diff options
| -rw-r--r-- | doc/html/nasmdo10.html | 165 | ||||
| -rw-r--r-- | doc/html/nasmdo11.html | 164 | ||||
| -rw-r--r-- | doc/html/nasmdo12.html | 178 | ||||
| -rw-r--r-- | doc/html/nasmdoc0.html | 378 | ||||
| -rw-r--r-- | doc/html/nasmdoc1.html | 159 | ||||
| -rw-r--r-- | doc/html/nasmdoc2.html | 650 | ||||
| -rw-r--r-- | doc/html/nasmdoc3.html | 753 | ||||
| -rw-r--r-- | doc/html/nasmdoc4.html | 1970 | ||||
| -rw-r--r-- | doc/html/nasmdoc5.html | 83 | ||||
| -rw-r--r-- | doc/html/nasmdoc6.html | 371 | ||||
| -rw-r--r-- | doc/html/nasmdoc7.html | 1267 | ||||
| -rw-r--r-- | doc/html/nasmdoc8.html | 808 | ||||
| -rw-r--r-- | doc/html/nasmdoc9.html | 482 | ||||
| -rw-r--r-- | doc/html/nasmdoca.html | 165 | ||||
| -rw-r--r-- | doc/html/nasmdocb.html | 3148 | ||||
| -rw-r--r-- | doc/html/nasmdocc.html | 1468 | ||||
| -rw-r--r-- | doc/html/nasmdoci.html | 2345 | ||||
| -rw-r--r-- | doc/info/nasm.info | 398 | ||||
| -rw-r--r-- | doc/info/nasm.info-1 | 7484 | ||||
| -rw-r--r-- | doc/info/nasm.info-2 | 6521 | ||||
| -rw-r--r-- | doc/info/nasm.info-3 | 1284 | ||||
| -rw-r--r-- | doc/inslist.src | 3130 | ||||
| -rw-r--r-- | doc/nasmdoc.pdf | bin | 689552 -> 0 bytes | |||
| -rw-r--r-- | doc/nasmdoc.ps | 12251 | ||||
| -rw-r--r-- | doc/nasmdoc.txt | 12394 | ||||
| -rw-r--r-- | packaging/nasm.manifest | 5 | ||||
| -rw-r--r-- | packaging/nasm.spec | 12 |
27 files changed, 2 insertions, 58031 deletions
diff --git a/doc/html/nasmdo10.html b/doc/html/nasmdo10.html deleted file mode 100644 index 151dee1..0000000 --- a/doc/html/nasmdo10.html +++ /dev/null @@ -1,165 +0,0 @@ -<html><head><title>NASM Manual</title></head> -<body><h1 align=center>The Netwide Assembler: NASM</h1> - -<p align=center><a href="nasmdo11.html">Next Chapter</a> | -<a href="nasmdoc9.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -<h2><a name="chapter-10">Chapter 10: Mixing 16 and 32 Bit Code</a></h2> -<p>This chapter tries to cover some of the issues, largely related to -unusual forms of addressing and jump instructions, encountered when writing -operating system code such as protected-mode initialisation routines, which -require code that operates in mixed segment sizes, such as code in a 16-bit -segment trying to modify data in a 32-bit one, or jumps between -different-size segments. -<h3><a name="section-10.1">10.1 Mixed-Size Jumps</a></h3> -<p>The most common form of mixed-size instruction is the one used when -writing a 32-bit OS: having done your setup in 16-bit mode, such as loading -the kernel, you then have to boot it by switching into protected mode and -jumping to the 32-bit kernel start address. In a fully 32-bit OS, this -tends to be the <em>only</em> mixed-size instruction you need, since -everything before it can be done in pure 16-bit code, and everything after -it can be pure 32-bit. -<p>This jump must specify a 48-bit far address, since the target segment is -a 32-bit one. However, it must be assembled in a 16-bit segment, so just -coding, for example, -<p><pre> - jmp 0x1234:0x56789ABC ; wrong! -</pre> -<p>will not work, since the offset part of the address will be truncated to -<code><nobr>0x9ABC</nobr></code> and the jump will be an ordinary 16-bit -far one. -<p>The Linux kernel setup code gets round the inability of -<code><nobr>as86</nobr></code> to generate the required instruction by -coding it manually, using <code><nobr>DB</nobr></code> instructions. NASM -can go one better than that, by actually generating the right instruction -itself. Here's how to do it right: -<p><pre> - jmp dword 0x1234:0x56789ABC ; right -</pre> -<p>The <code><nobr>DWORD</nobr></code> prefix (strictly speaking, it should -come <em>after</em> the colon, since it is declaring the <em>offset</em> -field to be a doubleword; but NASM will accept either form, since both are -unambiguous) forces the offset part to be treated as far, in the assumption -that you are deliberately writing a jump from a 16-bit segment to a 32-bit -one. -<p>You can do the reverse operation, jumping from a 32-bit segment to a -16-bit one, by means of the <code><nobr>WORD</nobr></code> prefix: -<p><pre> - jmp word 0x8765:0x4321 ; 32 to 16 bit -</pre> -<p>If the <code><nobr>WORD</nobr></code> prefix is specified in 16-bit -mode, or the <code><nobr>DWORD</nobr></code> prefix in 32-bit mode, they -will be ignored, since each is explicitly forcing NASM into a mode it was -in anyway. -<h3><a name="section-10.2">10.2 Addressing Between Different-Size Segments</a></h3> -<p>If your OS is mixed 16 and 32-bit, or if you are writing a DOS extender, -you are likely to have to deal with some 16-bit segments and some 32-bit -ones. At some point, you will probably end up writing code in a 16-bit -segment which has to access data in a 32-bit segment, or vice versa. -<p>If the data you are trying to access in a 32-bit segment lies within the -first 64K of the segment, you may be able to get away with using an -ordinary 16-bit addressing operation for the purpose; but sooner or later, -you will want to do 32-bit addressing from 16-bit mode. -<p>The easiest way to do this is to make sure you use a register for the -address, since any effective address containing a 32-bit register is forced -to be a 32-bit address. So you can do -<p><pre> - mov eax,offset_into_32_bit_segment_specified_by_fs - mov dword [fs:eax],0x11223344 -</pre> -<p>This is fine, but slightly cumbersome (since it wastes an instruction -and a register) if you already know the precise offset you are aiming at. -The x86 architecture does allow 32-bit effective addresses to specify -nothing but a 4-byte offset, so why shouldn't NASM be able to generate the -best instruction for the purpose? -<p>It can. As in <a href="#section-10.1">section 10.1</a>, you need only -prefix the address with the <code><nobr>DWORD</nobr></code> keyword, and it -will be forced to be a 32-bit address: -<p><pre> - mov dword [fs:dword my_offset],0x11223344 -</pre> -<p>Also as in <a href="#section-10.1">section 10.1</a>, NASM is not fussy -about whether the <code><nobr>DWORD</nobr></code> prefix comes before or -after the segment override, so arguably a nicer-looking way to code the -above instruction is -<p><pre> - mov dword [dword fs:my_offset],0x11223344 -</pre> -<p>Don't confuse the <code><nobr>DWORD</nobr></code> prefix -<em>outside</em> the square brackets, which controls the size of the data -stored at the address, with the one <code><nobr>inside</nobr></code> the -square brackets which controls the length of the address itself. The two -can quite easily be different: -<p><pre> - mov word [dword 0x12345678],0x9ABC -</pre> -<p>This moves 16 bits of data to an address specified by a 32-bit offset. -<p>You can also specify <code><nobr>WORD</nobr></code> or -<code><nobr>DWORD</nobr></code> prefixes along with the -<code><nobr>FAR</nobr></code> prefix to indirect far jumps or calls. For -example: -<p><pre> - call dword far [fs:word 0x4321] -</pre> -<p>This instruction contains an address specified by a 16-bit offset; it -loads a 48-bit far pointer from that (16-bit segment and 32-bit offset), -and calls that address. -<h3><a name="section-10.3">10.3 Other Mixed-Size Instructions</a></h3> -<p>The other way you might want to access data might be using the string -instructions (<code><nobr>LODSx</nobr></code>, -<code><nobr>STOSx</nobr></code> and so on) or the -<code><nobr>XLATB</nobr></code> instruction. These instructions, since they -take no parameters, might seem to have no easy way to make them perform -32-bit addressing when assembled in a 16-bit segment. -<p>This is the purpose of NASM's <code><nobr>a16</nobr></code>, -<code><nobr>a32</nobr></code> and <code><nobr>a64</nobr></code> prefixes. -If you are coding <code><nobr>LODSB</nobr></code> in a 16-bit segment but -it is supposed to be accessing a string in a 32-bit segment, you should -load the desired address into <code><nobr>ESI</nobr></code> and then code -<p><pre> - a32 lodsb -</pre> -<p>The prefix forces the addressing size to 32 bits, meaning that -<code><nobr>LODSB</nobr></code> loads from -<code><nobr>[DS:ESI]</nobr></code> instead of -<code><nobr>[DS:SI]</nobr></code>. To access a string in a 16-bit segment -when coding in a 32-bit one, the corresponding -<code><nobr>a16</nobr></code> prefix can be used. -<p>The <code><nobr>a16</nobr></code>, <code><nobr>a32</nobr></code> and -<code><nobr>a64</nobr></code> prefixes can be applied to any instruction in -NASM's instruction table, but most of them can generate all the useful -forms without them. The prefixes are necessary only for instructions with -implicit addressing: <code><nobr>CMPSx</nobr></code>, -<code><nobr>SCASx</nobr></code>, <code><nobr>LODSx</nobr></code>, -<code><nobr>STOSx</nobr></code>, <code><nobr>MOVSx</nobr></code>, -<code><nobr>INSx</nobr></code>, <code><nobr>OUTSx</nobr></code>, and -<code><nobr>XLATB</nobr></code>. Also, the various push and pop -instructions (<code><nobr>PUSHA</nobr></code> and -<code><nobr>POPF</nobr></code> as well as the more usual -<code><nobr>PUSH</nobr></code> and <code><nobr>POP</nobr></code>) can -accept <code><nobr>a16</nobr></code>, <code><nobr>a32</nobr></code> or -<code><nobr>a64</nobr></code> prefixes to force a particular one of -<code><nobr>SP</nobr></code>, <code><nobr>ESP</nobr></code> or -<code><nobr>RSP</nobr></code> to be used as a stack pointer, in case the -stack segment in use is a different size from the code segment. -<p><code><nobr>PUSH</nobr></code> and <code><nobr>POP</nobr></code>, when -applied to segment registers in 32-bit mode, also have the slightly odd -behaviour that they push and pop 4 bytes at a time, of which the top two -are ignored and the bottom two give the value of the segment register being -manipulated. To force the 16-bit behaviour of segment-register push and pop -instructions, you can use the operand-size prefix -<code><nobr>o16</nobr></code>: -<p><pre> - o16 push ss - o16 push ds -</pre> -<p>This code saves a doubleword of stack space by fitting two segment -registers into the space which would normally be consumed by pushing one. -<p>(You can also use the <code><nobr>o32</nobr></code> prefix to force the -32-bit behaviour when in 16-bit mode, but this seems less useful.) -<p align=center><a href="nasmdo11.html">Next Chapter</a> | -<a href="nasmdoc9.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -</body></html> diff --git a/doc/html/nasmdo11.html b/doc/html/nasmdo11.html deleted file mode 100644 index c40eede..0000000 --- a/doc/html/nasmdo11.html +++ /dev/null @@ -1,164 +0,0 @@ -<html><head><title>NASM Manual</title></head> -<body><h1 align=center>The Netwide Assembler: NASM</h1> - -<p align=center><a href="nasmdo12.html">Next Chapter</a> | -<a href="nasmdo10.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -<h2><a name="chapter-11">Chapter 11: Writing 64-bit Code (Unix, Win64)</a></h2> -<p>This chapter attempts to cover some of the common issues involved when -writing 64-bit code, to run under Win64 or Unix. It covers how to write -assembly code to interface with 64-bit C routines, and how to write -position-independent code for shared libraries. -<p>All 64-bit code uses a flat memory model, since segmentation is not -available in 64-bit mode. The one exception is the -<code><nobr>FS</nobr></code> and <code><nobr>GS</nobr></code> registers, -which still add their bases. -<p>Position independence in 64-bit mode is significantly simpler, since the -processor supports <code><nobr>RIP</nobr></code>-relative addressing -directly; see the <code><nobr>REL</nobr></code> keyword -(<a href="nasmdoc3.html#section-3.3">section 3.3</a>). On most 64-bit -platforms, it is probably desirable to make that the default, using the -directive <code><nobr>DEFAULT REL</nobr></code> -(<a href="nasmdoc6.html#section-6.2">section 6.2</a>). -<p>64-bit programming is relatively similar to 32-bit programming, but of -course pointers are 64 bits long; additionally, all existing platforms pass -arguments in registers rather than on the stack. Furthermore, 64-bit -platforms use SSE2 by default for floating point. Please see the ABI -documentation for your platform. -<p>64-bit platforms differ in the sizes of the fundamental datatypes, not -just from 32-bit platforms but from each other. If a specific size data -type is desired, it is probably best to use the types defined in the -Standard C header <code><nobr><inttypes.h></nobr></code>. -<p>In 64-bit mode, the default instruction size is still 32 bits. When -loading a value into a 32-bit register (but not an 8- or 16-bit register), -the upper 32 bits of the corresponding 64-bit register are set to zero. -<h3><a name="section-11.1">11.1 Register Names in 64-bit Mode</a></h3> -<p>NASM uses the following names for general-purpose registers in 64-bit -mode, for 8-, 16-, 32- and 64-bit references, respecitively: -<p><pre> - AL/AH, CL/CH, DL/DH, BL/BH, SPL, BPL, SIL, DIL, R8B-R15B - AX, CX, DX, BX, SP, BP, SI, DI, R8W-R15W - EAX, ECX, EDX, EBX, ESP, EBP, ESI, EDI, R8D-R15D - RAX, RCX, RDX, RBX, RSP, RBP, RSI, RDI, R8-R15 -</pre> -<p>This is consistent with the AMD documentation and most other assemblers. -The Intel documentation, however, uses the names -<code><nobr>R8L-R15L</nobr></code> for 8-bit references to the higher -registers. It is possible to use those names by definiting them as macros; -similarly, if one wants to use numeric names for the low 8 registers, -define them as macros. The standard macro package -<code><nobr>altreg</nobr></code> (see -<a href="nasmdoc5.html#section-5.1">section 5.1</a>) can be used for this -purpose. -<h3><a name="section-11.2">11.2 Immediates and Displacements in 64-bit Mode</a></h3> -<p>In 64-bit mode, immediates and displacements are generally only 32 bits -wide. NASM will therefore truncate most displacements and immediates to 32 -bits. -<p>The only instruction which takes a full 64-bit immediate is: -<p><pre> - MOV reg64,imm64 -</pre> -<p>NASM will produce this instruction whenever the programmer uses -<code><nobr>MOV</nobr></code> with an immediate into a 64-bit register. If -this is not desirable, simply specify the equivalent 32-bit register, which -will be automatically zero-extended by the processor, or specify the -immediate as <code><nobr>DWORD</nobr></code>: -<p><pre> - mov rax,foo ; 64-bit immediate - mov rax,qword foo ; (identical) - mov eax,foo ; 32-bit immediate, zero-extended - mov rax,dword foo ; 32-bit immediate, sign-extended -</pre> -<p>The length of these instructions are 10, 5 and 7 bytes, respectively. -<p>The only instructions which take a full 64-bit <em>displacement</em> is -loading or storing, using <code><nobr>MOV</nobr></code>, -<code><nobr>AL</nobr></code>, <code><nobr>AX</nobr></code>, -<code><nobr>EAX</nobr></code> or <code><nobr>RAX</nobr></code> (but no -other registers) to an absolute 64-bit address. Since this is a relatively -rarely used instruction (64-bit code generally uses relative addressing), -the programmer has to explicitly declare the displacement size as -<code><nobr>QWORD</nobr></code>: -<p><pre> - default abs - - mov eax,[foo] ; 32-bit absolute disp, sign-extended - mov eax,[a32 foo] ; 32-bit absolute disp, zero-extended - mov eax,[qword foo] ; 64-bit absolute disp - - default rel - - mov eax,[foo] ; 32-bit relative disp - mov eax,[a32 foo] ; d:o, address truncated to 32 bits(!) - mov eax,[qword foo] ; error - mov eax,[abs qword foo] ; 64-bit absolute disp -</pre> -<p>A sign-extended absolute displacement can access from -2 GB to +2 GB; a -zero-extended absolute displacement can access from 0 to 4 GB. -<h3><a name="section-11.3">11.3 Interfacing to 64-bit C Programs (Unix)</a></h3> -<p>On Unix, the 64-bit ABI is defined by the document: -<p> -<a href="http://www.x86-64.org/documentation/abi.pdf"><code><nobr>http://www.x86-64.org/documentation/abi.pdf</nobr></code></a> -<p>Although written for AT&T-syntax assembly, the concepts apply -equally well for NASM-style assembly. What follows is a simplified summary. -<p>The first six integer arguments (from the left) are passed in -<code><nobr>RDI</nobr></code>, <code><nobr>RSI</nobr></code>, -<code><nobr>RDX</nobr></code>, <code><nobr>RCX</nobr></code>, -<code><nobr>R8</nobr></code>, and <code><nobr>R9</nobr></code>, in that -order. Additional integer arguments are passed on the stack. These -registers, plus <code><nobr>RAX</nobr></code>, -<code><nobr>R10</nobr></code> and <code><nobr>R11</nobr></code> are -destroyed by function calls, and thus are available for use by the function -without saving. -<p>Integer return values are passed in <code><nobr>RAX</nobr></code> and -<code><nobr>RDX</nobr></code>, in that order. -<p>Floating point is done using SSE registers, except for -<code><nobr>long double</nobr></code>. Floating-point arguments are passed -in <code><nobr>XMM0</nobr></code> to <code><nobr>XMM7</nobr></code>; return -is <code><nobr>XMM0</nobr></code> and <code><nobr>XMM1</nobr></code>. -<code><nobr>long double</nobr></code> are passed on the stack, and returned -in <code><nobr>ST0</nobr></code> and <code><nobr>ST1</nobr></code>. -<p>All SSE and x87 registers are destroyed by function calls. -<p>On 64-bit Unix, <code><nobr>long</nobr></code> is 64 bits. -<p>Integer and SSE register arguments are counted separately, so for the -case of -<p><pre> - void foo(long a, double b, int c) -</pre> -<p><code><nobr>a</nobr></code> is passed in <code><nobr>RDI</nobr></code>, -<code><nobr>b</nobr></code> in <code><nobr>XMM0</nobr></code>, and -<code><nobr>c</nobr></code> in <code><nobr>ESI</nobr></code>. -<h3><a name="section-11.4">11.4 Interfacing to 64-bit C Programs (Win64)</a></h3> -<p>The Win64 ABI is described at: -<p> -<a href="http://msdn2.microsoft.com/en-gb/library/ms794533.aspx"><code><nobr>http://msdn2.microsoft.com/en-gb/library/ms794533.aspx</nobr></code></a> -<p>What follows is a simplified summary. -<p>The first four integer arguments are passed in -<code><nobr>RCX</nobr></code>, <code><nobr>RDX</nobr></code>, -<code><nobr>R8</nobr></code> and <code><nobr>R9</nobr></code>, in that -order. Additional integer arguments are passed on the stack. These -registers, plus <code><nobr>RAX</nobr></code>, -<code><nobr>R10</nobr></code> and <code><nobr>R11</nobr></code> are -destroyed by function calls, and thus are available for use by the function -without saving. -<p>Integer return values are passed in <code><nobr>RAX</nobr></code> only. -<p>Floating point is done using SSE registers, except for -<code><nobr>long double</nobr></code>. Floating-point arguments are passed -in <code><nobr>XMM0</nobr></code> to <code><nobr>XMM3</nobr></code>; return -is <code><nobr>XMM0</nobr></code> only. -<p>On Win64, <code><nobr>long</nobr></code> is 32 bits; -<code><nobr>long long</nobr></code> or <code><nobr>_int64</nobr></code> is -64 bits. -<p>Integer and SSE register arguments are counted together, so for the case -of -<p><pre> - void foo(long long a, double b, int c) -</pre> -<p><code><nobr>a</nobr></code> is passed in <code><nobr>RCX</nobr></code>, -<code><nobr>b</nobr></code> in <code><nobr>XMM1</nobr></code>, and -<code><nobr>c</nobr></code> in <code><nobr>R8D</nobr></code>. -<p align=center><a href="nasmdo12.html">Next Chapter</a> | -<a href="nasmdo10.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -</body></html> diff --git a/doc/html/nasmdo12.html b/doc/html/nasmdo12.html deleted file mode 100644 index 9c937f8..0000000 --- a/doc/html/nasmdo12.html +++ /dev/null @@ -1,178 +0,0 @@ -<html><head><title>NASM Manual</title></head> -<body><h1 align=center>The Netwide Assembler: NASM</h1> - -<p align=center><a href="nasmdoca.html">Next Chapter</a> | -<a href="nasmdo11.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -<h2><a name="chapter-12">Chapter 12: Troubleshooting</a></h2> -<p>This chapter describes some of the common problems that users have been -known to encounter with NASM, and answers them. It also gives instructions -for reporting bugs in NASM if you find a difficulty that isn't listed here. -<h3><a name="section-12.1">12.1 Common Problems</a></h3> -<h4><a name="section-12.1.1">12.1.1 NASM Generates Inefficient Code</a></h4> -<p>We sometimes get `bug' reports about NASM generating inefficient, or -even `wrong', code on instructions such as -<code><nobr>ADD ESP,8</nobr></code>. This is a deliberate design feature, -connected to predictability of output: NASM, on seeing -<code><nobr>ADD ESP,8</nobr></code>, will generate the form of the -instruction which leaves room for a 32-bit offset. You need to code -<code><nobr>ADD ESP,BYTE 8</nobr></code> if you want the space-efficient -form of the instruction. This isn't a bug, it's user error: if you prefer -to have NASM produce the more efficient code automatically enable -optimization with the <code><nobr>-O</nobr></code> option (see -<a href="nasmdoc2.html#section-2.1.22">section 2.1.22</a>). -<h4><a name="section-12.1.2">12.1.2 My Jumps are Out of Range</a></h4> -<p>Similarly, people complain that when they issue conditional jumps (which -are <code><nobr>SHORT</nobr></code> by default) that try to jump too far, -NASM reports `short jump out of range' instead of making the jumps longer. -<p>This, again, is partly a predictability issue, but in fact has a more -practical reason as well. NASM has no means of being told what type of -processor the code it is generating will be run on; so it cannot decide for -itself that it should generate <code><nobr>Jcc NEAR</nobr></code> type -instructions, because it doesn't know that it's working for a 386 or above. -Alternatively, it could replace the out-of-range short -<code><nobr>JNE</nobr></code> instruction with a very short -<code><nobr>JE</nobr></code> instruction that jumps over a -<code><nobr>JMP NEAR</nobr></code>; this is a sensible solution for -processors below a 386, but hardly efficient on processors which have good -branch prediction <em>and</em> could have used -<code><nobr>JNE NEAR</nobr></code> instead. So, once again, it's up to the -user, not the assembler, to decide what instructions should be generated. -See <a href="nasmdoc2.html#section-2.1.22">section 2.1.22</a>. -<h4><a name="section-12.1.3">12.1.3 <code><nobr>ORG</nobr></code> Doesn't Work</a></h4> -<p>People writing boot sector programs in the <code><nobr>bin</nobr></code> -format often complain that <code><nobr>ORG</nobr></code> doesn't work the -way they'd like: in order to place the <code><nobr>0xAA55</nobr></code> -signature word at the end of a 512-byte boot sector, people who are used to -MASM tend to code -<p><pre> - ORG 0 - - ; some boot sector code - - ORG 510 - DW 0xAA55 -</pre> -<p>This is not the intended use of the <code><nobr>ORG</nobr></code> -directive in NASM, and will not work. The correct way to solve this problem -in NASM is to use the <code><nobr>TIMES</nobr></code> directive, like this: -<p><pre> - ORG 0 - - ; some boot sector code - - TIMES 510-($-$$) DB 0 - DW 0xAA55 -</pre> -<p>The <code><nobr>TIMES</nobr></code> directive will insert exactly enough -zero bytes into the output to move the assembly point up to 510. This -method also has the advantage that if you accidentally fill your boot -sector too full, NASM will catch the problem at assembly time and report -it, so you won't end up with a boot sector that you have to disassemble to -find out what's wrong with it. -<h4><a name="section-12.1.4">12.1.4 <code><nobr>TIMES</nobr></code> Doesn't Work</a></h4> -<p>The other common problem with the above code is people who write the -<code><nobr>TIMES</nobr></code> line as -<p><pre> - TIMES 510-$ DB 0 -</pre> -<p>by reasoning that <code><nobr>$</nobr></code> should be a pure number, -just like 510, so the difference between them is also a pure number and can -happily be fed to <code><nobr>TIMES</nobr></code>. -<p>NASM is a <em>modular</em> assembler: the various component parts are -designed to be easily separable for re-use, so they don't exchange -information unnecessarily. In consequence, the -<code><nobr>bin</nobr></code> output format, even though it has been told -by the <code><nobr>ORG</nobr></code> directive that the -<code><nobr>.text</nobr></code> section should start at 0, does not pass -that information back to the expression evaluator. So from the evaluator's -point of view, <code><nobr>$</nobr></code> isn't a pure number: it's an -offset from a section base. Therefore the difference between -<code><nobr>$</nobr></code> and 510 is also not a pure number, but involves -a section base. Values involving section bases cannot be passed as -arguments to <code><nobr>TIMES</nobr></code>. -<p>The solution, as in the previous section, is to code the -<code><nobr>TIMES</nobr></code> line in the form -<p><pre> - TIMES 510-($-$$) DB 0 -</pre> -<p>in which <code><nobr>$</nobr></code> and <code><nobr>$$</nobr></code> -are offsets from the same section base, and so their difference is a pure -number. This will solve the problem and generate sensible code. -<h3><a name="section-12.2">12.2 Bugs</a></h3> -<p>We have never yet released a version of NASM with any <em>known</em> -bugs. That doesn't usually stop there being plenty we didn't know about, -though. Any that you find should be reported firstly via the -<code><nobr>bugtracker</nobr></code> at -<a href="https://sourceforge.net/projects/nasm/"><code><nobr>https://sourceforge.net/projects/nasm/</nobr></code></a> -(click on "Bugs"), or if that fails then through one of the contacts in -<a href="nasmdoc1.html#section-1.2">section 1.2</a>. -<p>Please read <a href="nasmdoc2.html#section-2.2">section 2.2</a> first, -and don't report the bug if it's listed in there as a deliberate feature. -(If you think the feature is badly thought out, feel free to send us -reasons why you think it should be changed, but don't just send us mail -saying `This is a bug' if the documentation says we did it on purpose.) -Then read <a href="#section-12.1">section 12.1</a>, and don't bother -reporting the bug if it's listed there. -<p>If you do report a bug, <em>please</em> give us all of the following -information: -<ul> -<li>What operating system you're running NASM under. DOS, Linux, NetBSD, -Win16, Win32, VMS (I'd be impressed), whatever. -<li>If you're running NASM under DOS or Win32, tell us whether you've -compiled your own executable from the DOS source archive, or whether you -were using the standard distribution binaries out of the archive. If you -were using a locally built executable, try to reproduce the problem using -one of the standard binaries, as this will make it easier for us to -reproduce your problem prior to fixing it. -<li>Which version of NASM you're using, and exactly how you invoked it. -Give us the precise command line, and the contents of the -<code><nobr>NASMENV</nobr></code> environment variable if any. -<li>Which versions of any supplementary programs you're using, and how you -invoked them. If the problem only becomes visible at link time, tell us -what linker you're using, what version of it you've got, and the exact -linker command line. If the problem involves linking against object files -generated by a compiler, tell us what compiler, what version, and what -command line or options you used. (If you're compiling in an IDE, please -try to reproduce the problem with the command-line version of the -compiler.) -<li>If at all possible, send us a NASM source file which exhibits the -problem. If this causes copyright problems (e.g. you can only reproduce the -bug in restricted-distribution code) then bear in mind the following two -points: firstly, we guarantee that any source code sent to us for the -purposes of debugging NASM will be used <em>only</em> for the purposes of -debugging NASM, and that we will delete all our copies of it as soon as we -have found and fixed the bug or bugs in question; and secondly, we would -prefer <em>not</em> to be mailed large chunks of code anyway. The smaller -the file, the better. A three-line sample file that does nothing useful -<em>except</em> demonstrate the problem is much easier to work with than a -fully fledged ten-thousand-line program. (Of course, some errors -<em>do</em> only crop up in large files, so this may not be possible.) -<li>A description of what the problem actually <em>is</em>. `It doesn't -work' is <em>not</em> a helpful description! Please describe exactly what -is happening that shouldn't be, or what isn't happening that should. -Examples might be: `NASM generates an error message saying Line 3 for an -error that's actually on Line 5'; `NASM generates an error message that I -believe it shouldn't be generating at all'; `NASM fails to generate an -error message that I believe it <em>should</em> be generating'; `the object -file produced from this source code crashes my linker'; `the ninth byte of -the output file is 66 and I think it should be 77 instead'. -<li>If you believe the output file from NASM to be faulty, send it to us. -That allows us to determine whether our own copy of NASM generates the same -file, or whether the problem is related to portability issues between our -development platforms and yours. We can handle binary files mailed to us as -MIME attachments, uuencoded, and even BinHex. Alternatively, we may be able -to provide an FTP site you can upload the suspect files to; but mailing -them is easier for us. -<li>Any other information or data files that might be helpful. If, for -example, the problem involves NASM failing to generate an object file while -TASM can generate an equivalent file without trouble, then send us -<em>both</em> object files, so we can see what TASM is doing differently -from us. -</ul> -<p align=center><a href="nasmdoca.html">Next Chapter</a> | -<a href="nasmdo11.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -</body></html> diff --git a/doc/html/nasmdoc0.html b/doc/html/nasmdoc0.html deleted file mode 100644 index 2b51d51..0000000 --- a/doc/html/nasmdoc0.html +++ /dev/null @@ -1,378 +0,0 @@ -<html><head><title>NASM Manual</title></head> -<body><h1 align=center>The Netwide Assembler: NASM</h1> - -<p>This manual documents NASM, the Netwide Assembler: an assembler -targetting the Intel x86 series of processors, with portable source. -<p><p><a href="nasmdoc1.html">Chapter 1: Introduction</a><br> -<a href="nasmdoc1.html#section-1.1">Section 1.1: What Is NASM?</a><br> -<a href="nasmdoc1.html#section-1.1.1">Section 1.1.1: Why Yet Another Assembler?</a><br> -<a href="nasmdoc1.html#section-1.1.2">Section 1.1.2: License Conditions</a><br> -<a href="nasmdoc1.html#section-1.2">Section 1.2: Contact Information</a><br> -<a href="nasmdoc1.html#section-1.3">Section 1.3: Installation</a><br> -<a href="nasmdoc1.html#section-1.3.1">Section 1.3.1: Installing NASM under MS-DOS or Windows</a><br> -<a href="nasmdoc1.html#section-1.3.2">Section 1.3.2: Installing NASM under Unix</a><br> -<p><a href="nasmdoc2.html">Chapter 2: Running NASM</a><br> -<a href="nasmdoc2.html#section-2.1">Section 2.1: NASM Command-Line Syntax</a><br> -<a href="nasmdoc2.html#section-2.1.1">Section 2.1.1: The <code><nobr>-o</nobr></code> Option: Specifying the Output File Name</a><br> -<a href="nasmdoc2.html#section-2.1.2">Section 2.1.2: The <code><nobr>-f</nobr></code> Option: Specifying the Output File Format</a><br> -<a href="nasmdoc2.html#section-2.1.3">Section 2.1.3: The <code><nobr>-l</nobr></code> Option: Generating a Listing File</a><br> -<a href="nasmdoc2.html#section-2.1.4">Section 2.1.4: The <code><nobr>-M</nobr></code> Option: Generate Makefile Dependencies</a><br> -<a href="nasmdoc2.html#section-2.1.5">Section 2.1.5: The <code><nobr>-MG</nobr></code> Option: Generate Makefile Dependencies</a><br> -<a href="nasmdoc2.html#section-2.1.6">Section 2.1.6: The <code><nobr>-MF</nobr></code> Option: Set Makefile Dependency File</a><br> -<a href="nasmdoc2.html#section-2.1.7">Section 2.1.7: The <code><nobr>-MD</nobr></code> Option: Assemble and Generate Dependencies</a><br> -<a href="nasmdoc2.html#section-2.1.8">Section 2.1.8: The <code><nobr>-MT</nobr></code> Option: Dependency Target Name</a><br> -<a href="nasmdoc2.html#section-2.1.9">Section 2.1.9: The <code><nobr>-MQ</nobr></code> Option: Dependency Target Name (Quoted)</a><br> -<a href="nasmdoc2.html#section-2.1.10">Section 2.1.10: The <code><nobr>-MP</nobr></code> Option: Emit phony targets</a><br> -<a href="nasmdoc2.html#section-2.1.11">Section 2.1.11: The <code><nobr>-F</nobr></code> Option: Selecting a Debug Information Format</a><br> -<a href="nasmdoc2.html#section-2.1.12">Section 2.1.12: The <code><nobr>-g</nobr></code> Option: Enabling Debug Information.</a><br> -<a href="nasmdoc2.html#section-2.1.13">Section 2.1.13: The <code><nobr>-X</nobr></code> Option: Selecting an Error Reporting Format</a><br> -<a href="nasmdoc2.html#section-2.1.14">Section 2.1.14: The <code><nobr>-Z</nobr></code> Option: Send Errors to a File</a><br> -<a href="nasmdoc2.html#section-2.1.15">Section 2.1.15: The <code><nobr>-s</nobr></code> Option: Send Errors to <code><nobr>stdout</nobr></code></a><br> -<a href="nasmdoc2.html#section-2.1.16">Section 2.1.16: The <code><nobr>-i</nobr></code> Option: Include File Search Directories</a><br> -<a href="nasmdoc2.html#section-2.1.17">Section 2.1.17: The <code><nobr>-p</nobr></code> Option: Pre-Include a File</a><br> -<a href="nasmdoc2.html#section-2.1.18">Section 2.1.18: The <code><nobr>-d</nobr></code> Option: Pre-Define a Macro</a><br> -<a href="nasmdoc2.html#section-2.1.19">Section 2.1.19: The <code><nobr>-u</nobr></code> Option: Undefine a Macro</a><br> -<a href="nasmdoc2.html#section-2.1.20">Section 2.1.20: The <code><nobr>-E</nobr></code> Option: Preprocess Only</a><br> -<a href="nasmdoc2.html#section-2.1.21">Section 2.1.21: The <code><nobr>-a</nobr></code> Option: Don't Preprocess At All</a><br> -<a href="nasmdoc2.html#section-2.1.22">Section 2.1.22: The <code><nobr>-O</nobr></code> Option: Specifying Multipass Optimization</a><br> -<a href="nasmdoc2.html#section-2.1.23">Section 2.1.23: The <code><nobr>-t</nobr></code> Option: Enable TASM Compatibility Mode</a><br> -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24: The <code><nobr>-w</nobr></code> and <code><nobr>-W</nobr></code> Options: Enable or Disable Assembly Warnings</a><br> -<a href="nasmdoc2.html#section-2.1.25">Section 2.1.25: The <code><nobr>-v</nobr></code> Option: Display Version Info</a><br> -<a href="nasmdoc2.html#section-2.1.26">Section 2.1.26: The <code><nobr>-y</nobr></code> Option: Display Available Debug Info Formats</a><br> -<a href="nasmdoc2.html#section-2.1.27">Section 2.1.27: The <code><nobr>--prefix</nobr></code> and <code><nobr>--postfix</nobr></code> Options.</a><br> -<a href="nasmdoc2.html#section-2.1.28">Section 2.1.28: The <code><nobr>NASMENV</nobr></code> Environment Variable</a><br> -<a href="nasmdoc2.html#section-2.2">Section 2.2: Quick Start for MASM Users</a><br> -<a href="nasmdoc2.html#section-2.2.1">Section 2.2.1: NASM Is Case-Sensitive</a><br> -<a href="nasmdoc2.html#section-2.2.2">Section 2.2.2: NASM Requires Square Brackets For Memory References</a><br> -<a href="nasmdoc2.html#section-2.2.3">Section 2.2.3: NASM Doesn't Store Variable Types</a><br> -<a href="nasmdoc2.html#section-2.2.4">Section 2.2.4: NASM Doesn't <code><nobr>ASSUME</nobr></code></a><br> -<a href="nasmdoc2.html#section-2.2.5">Section 2.2.5: NASM Doesn't Support Memory Models</a><br> -<a href="nasmdoc2.html#section-2.2.6">Section 2.2.6: Floating-Point Differences</a><br> -<a href="nasmdoc2.html#section-2.2.7">Section 2.2.7: Other Differences</a><br> -<p><a href="nasmdoc3.html">Chapter 3: The NASM Language</a><br> -<a href="nasmdoc3.html#section-3.1">Section 3.1: Layout of a NASM Source Line</a><br> -<a href="nasmdoc3.html#section-3.2">Section 3.2: Pseudo-Instructions</a><br> -<a href="nasmdoc3.html#section-3.2.1">Section 3.2.1: <code><nobr>DB</nobr></code> and Friends: Declaring Initialized Data</a><br> -<a href="nasmdoc3.html#section-3.2.2">Section 3.2.2: <code><nobr>RESB</nobr></code> and Friends: Declaring Uninitialized Data</a><br> -<a href="nasmdoc3.html#section-3.2.3">Section 3.2.3: <code><nobr>INCBIN</nobr></code>: Including External Binary Files</a><br> -<a href="nasmdoc3.html#section-3.2.4">Section 3.2.4: <code><nobr>EQU</nobr></code>: Defining Constants</a><br> -<a href="nasmdoc3.html#section-3.2.5">Section 3.2.5: <code><nobr>TIMES</nobr></code>: Repeating Instructions or Data</a><br> -<a href="nasmdoc3.html#section-3.3">Section 3.3: Effective Addresses</a><br> -<a href="nasmdoc3.html#section-3.4">Section 3.4: Constants</a><br> -<a href="nasmdoc3.html#section-3.4.1">Section 3.4.1: Numeric Constants</a><br> -<a href="nasmdoc3.html#section-3.4.2">Section 3.4.2: Character Strings</a><br> -<a href="nasmdoc3.html#section-3.4.3">Section 3.4.3: Character Constants</a><br> -<a href="nasmdoc3.html#section-3.4.4">Section 3.4.4: String Constants</a><br> -<a href="nasmdoc3.html#section-3.4.5">Section 3.4.5: Unicode Strings</a><br> -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6: Floating-Point Constants</a><br> -<a href="nasmdoc3.html#section-3.4.7">Section 3.4.7: Packed BCD Constants</a><br> -<a href="nasmdoc3.html#section-3.5">Section 3.5: Expressions</a><br> -<a href="nasmdoc3.html#section-3.5.1">Section 3.5.1: <code><nobr>|</nobr></code>: Bitwise OR Operator</a><br> -<a href="nasmdoc3.html#section-3.5.2">Section 3.5.2: <code><nobr>^</nobr></code>: Bitwise XOR Operator</a><br> -<a href="nasmdoc3.html#section-3.5.3">Section 3.5.3: <code><nobr>&</nobr></code>: Bitwise AND Operator</a><br> -<a href="nasmdoc3.html#section-3.5.4">Section 3.5.4: <code><nobr><<</nobr></code> and <code><nobr>>></nobr></code>: Bit Shift Operators</a><br> -<a href="nasmdoc3.html#section-3.5.5">Section 3.5.5: <code><nobr>+</nobr></code> and <code><nobr>-</nobr></code>: Addition and Subtraction Operators</a><br> -<a href="nasmdoc3.html#section-3.5.6">Section 3.5.6: <code><nobr>*</nobr></code>, <code><nobr>/</nobr></code>, <code><nobr>//</nobr></code>, <code><nobr>%</nobr></code> and <code><nobr>%%</nobr></code>: Multiplication and Division</a><br> -<a href="nasmdoc3.html#section-3.5.7">Section 3.5.7: Unary Operators: <code><nobr>+</nobr></code>, <code><nobr>-</nobr></code>, <code><nobr>~</nobr></code>, <code><nobr>!</nobr></code> and <code><nobr>SEG</nobr></code></a><br> -<a href="nasmdoc3.html#section-3.6">Section 3.6: <code><nobr>SEG</nobr></code> and <code><nobr>WRT</nobr></code></a><br> -<a href="nasmdoc3.html#section-3.7">Section 3.7: <code><nobr>STRICT</nobr></code>: Inhibiting Optimization</a><br> -<a href="nasmdoc3.html#section-3.8">Section 3.8: Critical Expressions</a><br> -<a href="nasmdoc3.html#section-3.9">Section 3.9: Local Labels</a><br> -<p><a href="nasmdoc4.html">Chapter 4: The NASM Preprocessor</a><br> -<a href="nasmdoc4.html#section-4.1">Section 4.1: Single-Line Macros</a><br> -<a href="nasmdoc4.html#section-4.1.1">Section 4.1.1: The Normal Way: <code><nobr>%define</nobr></code></a><br> -<a href="nasmdoc4.html#section-4.1.2">Section 4.1.2: Resolving <code><nobr>%define</nobr></code>: <code><nobr>%xdefine</nobr></code></a><br> -<a href="nasmdoc4.html#section-4.1.3">Section 4.1.3: Macro Indirection: <code><nobr>%[...]</nobr></code></a><br> -<a href="nasmdoc4.html#section-4.1.4">Section 4.1.4: Concatenating Single Line Macro Tokens: <code><nobr>%+</nobr></code></a><br> -<a href="nasmdoc4.html#section-4.1.5">Section 4.1.5: The Macro Name Itself: <code><nobr>%?</nobr></code> and <code><nobr>%??</nobr></code></a><br> -<a href="nasmdoc4.html#section-4.1.6">Section 4.1.6: Undefining Single-Line Macros: <code><nobr>%undef</nobr></code></a><br> -<a href="nasmdoc4.html#section-4.1.7">Section 4.1.7: Preprocessor Variables: <code><nobr>%assign</nobr></code></a><br> -<a href="nasmdoc4.html#section-4.1.8">Section 4.1.8: Defining Strings: <code><nobr>%defstr</nobr></code></a><br> -<a href="nasmdoc4.html#section-4.1.9">Section 4.1.9: Defining Tokens: <code><nobr>%deftok</nobr></code></a><br> -<a href="nasmdoc4.html#section-4.2">Section 4.2: String Manipulation in Macros</a><br> -<a href="nasmdoc4.html#section-4.2.1">Section 4.2.1: Concatenating Strings: <code><nobr>%strcat</nobr></code></a><br> -<a href="nasmdoc4.html#section-4.2.2">Section 4.2.2: String Length: <code><nobr>%strlen</nobr></code></a><br> -<a href="nasmdoc4.html#section-4.2.3">Section 4.2.3: Extracting Substrings: <code><nobr>%substr</nobr></code></a><br> -<a href="nasmdoc4.html#section-4.3">Section 4.3: Multi-Line Macros: <code><nobr>%macro</nobr></code></a><br> -<a href="nasmdoc4.html#section-4.3.1">Section 4.3.1: Recursive Multi-Line Macros: <code><nobr>%rmacro</nobr></code></a><br> -<a href="nasmdoc4.html#section-4.3.2">Section 4.3.2: Overloading Multi-Line Macros</a><br> -<a href="nasmdoc4.html#section-4.3.3">Section 4.3.3: Macro-Local Labels</a><br> -<a href="nasmdoc4.html#section-4.3.4">Section 4.3.4: Greedy Macro Parameters</a><br> -<a href="nasmdoc4.html#section-4.3.5">Section 4.3.5: Default Macro Parameters</a><br> -<a href="nasmdoc4.html#section-4.3.6">Section 4.3.6: <code><nobr>%0</nobr></code>: Macro Parameter Counter</a><br> -<a href="nasmdoc4.html#section-4.3.7">Section 4.3.7: <code><nobr>%rotate</nobr></code>: Rotating Macro Parameters</a><br> -<a href="nasmdoc4.html#section-4.3.8">Section 4.3.8: Concatenating Macro Parameters</a><br> -<a href="nasmdoc4.html#section-4.3.9">Section 4.3.9: Condition Codes as Macro Parameters</a><br> -<a href="nasmdoc4.html#section-4.3.10">Section 4.3.10: Disabling Listing Expansion</a><br> -<a href="nasmdoc4.html#section-4.3.11">Section 4.3.11: Undefining Multi-Line Macros: <code><nobr>%unmacro</nobr></code></a><br> -<a href="nasmdoc4.html#section-4.3.12">Section 4.3.12: Exiting Multi-Line Macros: <code><nobr>%exitmacro</nobr></code></a><br> -<a href="nasmdoc4.html#section-4.4">Section 4.4: Conditional Assembly</a><br> -<a href="nasmdoc4.html#section-4.4.1">Section 4.4.1: <code><nobr>%ifdef</nobr></code>: Testing Single-Line Macro Existence</a><br> -<a href="nasmdoc4.html#section-4.4.2">Section 4.4.2: <code><nobr>%ifmacro</nobr></code>: Testing Multi-Line Macro Existence</a><br> -<a href="nasmdoc4.html#section-4.4.3">Section 4.4.3: <code><nobr>%ifctx</nobr></code>: Testing the Context Stack</a><br> -<a href="nasmdoc4.html#section-4.4.4">Section 4.4.4: <code><nobr>%if</nobr></code>: Testing Arbitrary Numeric Expressions</a><br> -<a href="nasmdoc4.html#section-4.4.5">Section 4.4.5: <code><nobr>%ifidn</nobr></code> and <code><nobr>%ifidni</nobr></code>: Testing Exact Text Identity</a><br> -<a href="nasmdoc4.html#section-4.4.6">Section 4.4.6: <code><nobr>%ifid</nobr></code>, <code><nobr>%ifnum</nobr></code>, <code><nobr>%ifstr</nobr></code>: Testing Token Types</a><br> -<a href="nasmdoc4.html#section-4.4.7">Section 4.4.7: <code><nobr>%iftoken</nobr></code>: Test for a Single Token</a><br> -<a href="nasmdoc4.html#section-4.4.8">Section 4.4.8: <code><nobr>%ifempty</nobr></code>: Test for Empty Expansion</a><br> -<a href="nasmdoc4.html#section-4.5">Section 4.5: Preprocessor Loops: <code><nobr>%rep</nobr></code></a><br> -<a href="nasmdoc4.html#section-4.6">Section 4.6: Source Files and Dependencies</a><br> -<a href="nasmdoc4.html#section-4.6.1">Section 4.6.1: <code><nobr>%include</nobr></code>: Including Other Files</a><br> -<a href="nasmdoc4.html#section-4.6.2">Section 4.6.2: <code><nobr>%pathsearch</nobr></code>: Search the Include Path</a><br> -<a href="nasmdoc4.html#section-4.6.3">Section 4.6.3: <code><nobr>%depend</nobr></code>: Add Dependent Files</a><br> -<a href="nasmdoc4.html#section-4.6.4">Section 4.6.4: <code><nobr>%use</nobr></code>: Include Standard Macro Package</a><br> -<a href="nasmdoc4.html#section-4.7">Section 4.7: The Context Stack</a><br> -<a href="nasmdoc4.html#section-4.7.1">Section 4.7.1: <code><nobr>%push</nobr></code> and <code><nobr>%pop</nobr></code>: Creating and Removing Contexts</a><br> -<a href="nasmdoc4.html#section-4.7.2">Section 4.7.2: Context-Local Labels</a><br> -<a href="nasmdoc4.html#section-4.7.3">Section 4.7.3: Context-Local Single-Line Macros</a><br> -<a href="nasmdoc4.html#section-4.7.4">Section 4.7.4: <code><nobr>%repl</nobr></code>: Renaming a Context</a><br> -<a href="nasmdoc4.html#section-4.7.5">Section 4.7.5: Example Use of the Context Stack: Block IFs</a><br> -<a href="nasmdoc4.html#section-4.8">Section 4.8: Stack Relative Preprocessor Directives</a><br> -<a href="nasmdoc4.html#section-4.8.1">Section 4.8.1: <code><nobr>%arg</nobr></code> Directive</a><br> -<a href="nasmdoc4.html#section-4.8.2">Section 4.8.2: <code><nobr>%stacksize</nobr></code> Directive</a><br> -<a href="nasmdoc4.html#section-4.8.3">Section 4.8.3: <code><nobr>%local</nobr></code> Directive</a><br> -<a href="nasmdoc4.html#section-4.9">Section 4.9: Reporting User-Defined Errors: <code><nobr>%error</nobr></code>, <code><nobr>%warning</nobr></code>, <code><nobr>%fatal</nobr></code></a><br> -<a href="nasmdoc4.html#section-4.10">Section 4.10: Other Preprocessor Directives</a><br> -<a href="nasmdoc4.html#section-4.10.1">Section 4.10.1: <code><nobr>%line</nobr></code> Directive</a><br> -<a href="nasmdoc4.html#section-4.10.2">Section 4.10.2: <code><nobr>%!</nobr></code><code><nobr><env></nobr></code>: Read an environment variable.</a><br> -<a href="nasmdoc4.html#section-4.11">Section 4.11: Standard Macros</a><br> -<a href="nasmdoc4.html#section-4.11.1">Section 4.11.1: NASM Version Macros</a><br> -<a href="nasmdoc4.html#section-4.11.2">Section 4.11.2: <code><nobr>__NASM_VERSION_ID__</nobr></code>: NASM Version ID</a><br> -<a href="nasmdoc4.html#section-4.11.3">Section 4.11.3: <code><nobr>__NASM_VER__</nobr></code>: NASM Version string</a><br> -<a href="nasmdoc4.html#section-4.11.4">Section 4.11.4: <code><nobr>__FILE__</nobr></code> and <code><nobr>__LINE__</nobr></code>: File Name and Line Number</a><br> -<a href="nasmdoc4.html#section-4.11.5">Section 4.11.5: <code><nobr>__BITS__</nobr></code>: Current BITS Mode</a><br> -<a href="nasmdoc4.html#section-4.11.6">Section 4.11.6: <code><nobr>__OUTPUT_FORMAT__</nobr></code>: Current Output Format</a><br> -<a href="nasmdoc4.html#section-4.11.7">Section 4.11.7: Assembly Date and Time Macros</a><br> -<a href="nasmdoc4.html#section-4.11.8">Section 4.11.8: <code><nobr>__USE_</nobr></code><em>package</em><code><nobr>__</nobr></code>: Package Include Test</a><br> -<a href="nasmdoc4.html#section-4.11.9">Section 4.11.9: <code><nobr>__PASS__</nobr></code>: Assembly Pass</a><br> -<a href="nasmdoc4.html#section-4.11.10">Section 4.11.10: <code><nobr>STRUC</nobr></code> and <code><nobr>ENDSTRUC</nobr></code>: Declaring Structure Data Types</a><br> -<a href="nasmdoc4.html#section-4.11.11">Section 4.11.11: <code><nobr>ISTRUC</nobr></code>, <code><nobr>AT</nobr></code> and <code><nobr>IEND</nobr></code>: Declaring Instances of Structures</a><br> -<a href="nasmdoc4.html#section-4.11.12">Section 4.11.12: <code><nobr>ALIGN</nobr></code> and <code><nobr>ALIGNB</nobr></code>: Data Alignment</a><br> -<p><a href="nasmdoc5.html">Chapter 5: Standard Macro Packages</a><br> -<a href="nasmdoc5.html#section-5.1">Section 5.1: <code><nobr>altreg</nobr></code>: Alternate Register Names</a><br> -<a href="nasmdoc5.html#section-5.2">Section 5.2: <code><nobr>smartalign</nobr></code>: Smart <code><nobr>ALIGN</nobr></code> Macro</a><br> -<p><a href="nasmdoc6.html">Chapter 6: Assembler Directives</a><br> -<a href="nasmdoc6.html#section-6.1">Section 6.1: <code><nobr>BITS</nobr></code>: Specifying Target Processor Mode</a><br> -<a href="nasmdoc6.html#section-6.1.1">Section 6.1.1: <code><nobr>USE16</nobr></code> & <code><nobr>USE32</nobr></code>: Aliases for BITS</a><br> -<a href="nasmdoc6.html#section-6.2">Section 6.2: <code><nobr>DEFAULT</nobr></code>: Change the assembler defaults</a><br> -<a href="nasmdoc6.html#section-6.3">Section 6.3: <code><nobr>SECTION</nobr></code> or <code><nobr>SEGMENT</nobr></code>: Changing and Defining Sections</a><br> -<a href="nasmdoc6.html#section-6.3.1">Section 6.3.1: The <code><nobr>__SECT__</nobr></code> Macro</a><br> -<a href="nasmdoc6.html#section-6.4">Section 6.4: <code><nobr>ABSOLUTE</nobr></code>: Defining Absolute Labels</a><br> -<a href="nasmdoc6.html#section-6.5">Section 6.5: <code><nobr>EXTERN</nobr></code>: Importing Symbols from Other Modules</a><br> -<a href="nasmdoc6.html#section-6.6">Section 6.6: <code><nobr>GLOBAL</nobr></code>: Exporting Symbols to Other Modules</a><br> -<a href="nasmdoc6.html#section-6.7">Section 6.7: <code><nobr>COMMON</nobr></code>: Defining Common Data Areas</a><br> -<a href="nasmdoc6.html#section-6.8">Section 6.8: <code><nobr>CPU</nobr></code>: Defining CPU Dependencies</a><br> -<a href="nasmdoc6.html#section-6.9">Section 6.9: <code><nobr>FLOAT</nobr></code>: Handling of floating-point constants</a><br> -<p><a href="nasmdoc7.html">Chapter 7: Output Formats</a><br> -<a href="nasmdoc7.html#section-7.1">Section 7.1: <code><nobr>bin</nobr></code>: Flat-Form Binary Output</a><br> -<a href="nasmdoc7.html#section-7.1.1">Section 7.1.1: <code><nobr>ORG</nobr></code>: Binary File Program Origin</a><br> -<a href="nasmdoc7.html#section-7.1.2">Section 7.1.2: <code><nobr>bin</nobr></code> Extensions to the <code><nobr>SECTION</nobr></code> Directive</a><br> -<a href="nasmdoc7.html#section-7.1.3">Section 7.1.3: Multisection Support for the <code><nobr>bin</nobr></code> Format</a><br> -<a href="nasmdoc7.html#section-7.1.4">Section 7.1.4: Map Files</a><br> -<a href="nasmdoc7.html#section-7.2">Section 7.2: <code><nobr>ith</nobr></code>: Intel Hex Output</a><br> -<a href="nasmdoc7.html#section-7.3">Section 7.3: <code><nobr>srec</nobr></code>: Motorola S-Records Output</a><br> -<a href="nasmdoc7.html#section-7.4">Section 7.4: <code><nobr>obj</nobr></code>: Microsoft OMF Object Files</a><br> -<a href="nasmdoc7.html#section-7.4.1">Section 7.4.1: <code><nobr>obj</nobr></code> Extensions to the <code><nobr>SEGMENT</nobr></code> Directive</a><br> -<a href="nasmdoc7.html#section-7.4.2">Section 7.4.2: <code><nobr>GROUP</nobr></code>: Defining Groups of Segments</a><br> -<a href="nasmdoc7.html#section-7.4.3">Section 7.4.3: <code><nobr>UPPERCASE</nobr></code>: Disabling Case Sensitivity in Output</a><br> -<a href="nasmdoc7.html#section-7.4.4">Section 7.4.4: <code><nobr>IMPORT</nobr></code>: Importing DLL Symbols</a><br> -<a href="nasmdoc7.html#section-7.4.5">Section 7.4.5: <code><nobr>EXPORT</nobr></code>: Exporting DLL Symbols</a><br> -<a href="nasmdoc7.html#section-7.4.6">Section 7.4.6: <code><nobr>..start</nobr></code>: Defining the Program Entry Point</a><br> -<a href="nasmdoc7.html#section-7.4.7">Section 7.4.7: <code><nobr>obj</nobr></code> Extensions to the <code><nobr>EXTERN</nobr></code> Directive</a><br> -<a href="nasmdoc7.html#section-7.4.8">Section 7.4.8: <code><nobr>obj</nobr></code> Extensions to the <code><nobr>COMMON</nobr></code> Directive</a><br> -<a href="nasmdoc7.html#section-7.5">Section 7.5: <code><nobr>win32</nobr></code>: Microsoft Win32 Object Files</a><br> -<a href="nasmdoc7.html#section-7.5.1">Section 7.5.1: <code><nobr>win32</nobr></code> Extensions to the <code><nobr>SECTION</nobr></code> Directive</a><br> -<a href="nasmdoc7.html#section-7.5.2">Section 7.5.2: <code><nobr>win32</nobr></code>: Safe Structured Exception Handling</a><br> -<a href="nasmdoc7.html#section-7.6">Section 7.6: <code><nobr>win64</nobr></code>: Microsoft Win64 Object Files</a><br> -<a href="nasmdoc7.html#section-7.6.1">Section 7.6.1: <code><nobr>win64</nobr></code>: Writing Position-Independent Code</a><br> -<a href="nasmdoc7.html#section-7.6.2">Section 7.6.2: <code><nobr>win64</nobr></code>: Structured Exception Handling</a><br> -<a href="nasmdoc7.html#section-7.7">Section 7.7: <code><nobr>coff</nobr></code>: Common Object File Format</a><br> -<a href="nasmdoc7.html#section-7.8">Section 7.8: <code><nobr>macho32</nobr></code> and <code><nobr>macho64</nobr></code>: Mach Object File Format</a><br> -<a href="nasmdoc7.html#section-7.9">Section 7.9: <code><nobr>elf32</nobr></code> and <code><nobr>elf64</nobr></code>: Executable and Linkable Format Object Files</a><br> -<a href="nasmdoc7.html#section-7.9.1">Section 7.9.1: ELF specific directive <code><nobr>osabi</nobr></code></a><br> -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2: <code><nobr>elf</nobr></code> Extensions to the <code><nobr>SECTION</nobr></code> Directive</a><br> -<a href="nasmdoc7.html#section-7.9.3">Section 7.9.3: Position-Independent Code: <code><nobr>elf</nobr></code> Special Symbols and <code><nobr>WRT</nobr></code></a><br> -<a href="nasmdoc7.html#section-7.9.4">Section 7.9.4: Thread Local Storage: <code><nobr>elf</nobr></code> Special Symbols and <code><nobr>WRT</nobr></code></a><br> -<a href="nasmdoc7.html#section-7.9.5">Section 7.9.5: <code><nobr>elf</nobr></code> Extensions to the <code><nobr>GLOBAL</nobr></code> Directive</a><br> -<a href="nasmdoc7.html#section-7.9.6">Section 7.9.6: <code><nobr>elf</nobr></code> Extensions to the <code><nobr>COMMON</nobr></code> Directive </a><br> -<a href="nasmdoc7.html#section-7.9.7">Section 7.9.7: 16-bit code and ELF </a><br> -<a href="nasmdoc7.html#section-7.9.8">Section 7.9.8: Debug formats and ELF </a><br> -<a href="nasmdoc7.html#section-7.10">Section 7.10: <code><nobr>aout</nobr></code>: Linux <code><nobr>a.out</nobr></code> Object Files</a><br> -<a href="nasmdoc7.html#section-7.11">Section 7.11: <code><nobr>aoutb</nobr></code>: NetBSD/FreeBSD/OpenBSD <code><nobr>a.out</nobr></code> Object Files</a><br> -<a href="nasmdoc7.html#section-7.12">Section 7.12: <code><nobr>as86</nobr></code>: Minix/Linux <code><nobr>as86</nobr></code> Object Files</a><br> -<a href="nasmdoc7.html#section-7.13">Section 7.13: <code><nobr>rdf</nobr></code>: Relocatable Dynamic Object File Format</a><br> -<a href="nasmdoc7.html#section-7.13.1">Section 7.13.1: Requiring a Library: The <code><nobr>LIBRARY</nobr></code> Directive</a><br> -<a href="nasmdoc7.html#section-7.13.2">Section 7.13.2: Specifying a Module Name: The <code><nobr>MODULE</nobr></code> Directive</a><br> -<a href="nasmdoc7.html#section-7.13.3">Section 7.13.3: <code><nobr>rdf</nobr></code> Extensions to the <code><nobr>GLOBAL</nobr></code> Directive</a><br> -<a href="nasmdoc7.html#section-7.13.4">Section 7.13.4: <code><nobr>rdf</nobr></code> Extensions to the <code><nobr>EXTERN</nobr></code> Directive</a><br> -<a href="nasmdoc7.html#section-7.14">Section 7.14: <code><nobr>dbg</nobr></code>: Debugging Format</a><br> -<p><a href="nasmdoc8.html">Chapter 8: Writing 16-bit Code (DOS, Windows 3/3.1)</a><br> -<a href="nasmdoc8.html#section-8.1">Section 8.1: Producing <code><nobr>.EXE</nobr></code> Files</a><br> -<a href="nasmdoc8.html#section-8.1.1">Section 8.1.1: Using the <code><nobr>obj</nobr></code> Format To Generate <code><nobr>.EXE</nobr></code> Files</a><br> -<a href="nasmdoc8.html#section-8.1.2">Section 8.1.2: Using the <code><nobr>bin</nobr></code> Format To Generate <code><nobr>.EXE</nobr></code> Files</a><br> -<a href="nasmdoc8.html#section-8.2">Section 8.2: Producing <code><nobr>.COM</nobr></code> Files</a><br> -<a href="nasmdoc8.html#section-8.2.1">Section 8.2.1: Using the <code><nobr>bin</nobr></code> Format To Generate <code><nobr>.COM</nobr></code> Files</a><br> -<a href="nasmdoc8.html#section-8.2.2">Section 8.2.2: Using the <code><nobr>obj</nobr></code> Format To Generate <code><nobr>.COM</nobr></code> Files</a><br> -<a href="nasmdoc8.html#section-8.3">Section 8.3: Producing <code><nobr>.SYS</nobr></code> Files</a><br> -<a href="nasmdoc8.html#section-8.4">Section 8.4: Interfacing to 16-bit C Programs</a><br> -<a href="nasmdoc8.html#section-8.4.1">Section 8.4.1: External Symbol Names</a><br> -<a href="nasmdoc8.html#section-8.4.2">Section 8.4.2: Memory Models</a><br> -<a href="nasmdoc8.html#section-8.4.3">Section 8.4.3: Function Definitions and Function Calls</a><br> -<a href="nasmdoc8.html#section-8.4.4">Section 8.4.4: Accessing Data Items</a><br> -<a href="nasmdoc8.html#section-8.4.5">Section 8.4.5: <code><nobr>c16.mac</nobr></code>: Helper Macros for the 16-bit C Interface</a><br> -<a href="nasmdoc8.html#section-8.5">Section 8.5: Interfacing to Borland Pascal Programs</a><br> -<a href="nasmdoc8.html#section-8.5.1">Section 8.5.1: The Pascal Calling Convention</a><br> -<a href="nasmdoc8.html#section-8.5.2">Section 8.5.2: Borland Pascal Segment Name Restrictions</a><br> -<a href="nasmdoc8.html#section-8.5.3">Section 8.5.3: Using <code><nobr>c16.mac</nobr></code> With Pascal Programs</a><br> -<p><a href="nasmdoc9.html">Chapter 9: Writing 32-bit Code (Unix, Win32, DJGPP)</a><br> -<a href="nasmdoc9.html#section-9.1">Section 9.1: Interfacing to 32-bit C Programs</a><br> -<a href="nasmdoc9.html#section-9.1.1">Section 9.1.1: External Symbol Names</a><br> -<a href="nasmdoc9.html#section-9.1.2">Section 9.1.2: Function Definitions and Function Calls</a><br> -<a href="nasmdoc9.html#section-9.1.3">Section 9.1.3: Accessing Data Items</a><br> -<a href="nasmdoc9.html#section-9.1.4">Section 9.1.4: <code><nobr>c32.mac</nobr></code>: Helper Macros for the 32-bit C Interface</a><br> -<a href="nasmdoc9.html#section-9.2">Section 9.2: Writing NetBSD/FreeBSD/OpenBSD and Linux/ELF Shared Libraries</a><br> -<a href="nasmdoc9.html#section-9.2.1">Section 9.2.1: Obtaining the Address of the GOT</a><br> -<a href="nasmdoc9.html#section-9.2.2">Section 9.2.2: Finding Your Local Data Items</a><br> -<a href="nasmdoc9.html#section-9.2.3">Section 9.2.3: Finding External and Common Data Items</a><br> -<a href="nasmdoc9.html#section-9.2.4">Section 9.2.4: Exporting Symbols to the Library User</a><br> -<a href="nasmdoc9.html#section-9.2.5">Section 9.2.5: Calling Procedures Outside the Library</a><br> -<a href="nasmdoc9.html#section-9.2.6">Section 9.2.6: Generating the Library File</a><br> -<p><a href="nasmdo10.html">Chapter 10: Mixing 16 and 32 Bit Code</a><br> -<a href="nasmdo10.html#section-10.1">Section 10.1: Mixed-Size Jumps</a><br> -<a href="nasmdo10.html#section-10.2">Section 10.2: Addressing Between Different-Size Segments</a><br> -<a href="nasmdo10.html#section-10.3">Section 10.3: Other Mixed-Size Instructions</a><br> -<p><a href="nasmdo11.html">Chapter 11: Writing 64-bit Code (Unix, Win64)</a><br> -<a href="nasmdo11.html#section-11.1">Section 11.1: Register Names in 64-bit Mode</a><br> -<a href="nasmdo11.html#section-11.2">Section 11.2: Immediates and Displacements in 64-bit Mode</a><br> -<a href="nasmdo11.html#section-11.3">Section 11.3: Interfacing to 64-bit C Programs (Unix)</a><br> -<a href="nasmdo11.html#section-11.4">Section 11.4: Interfacing to 64-bit C Programs (Win64)</a><br> -<p><a href="nasmdo12.html">Chapter 12: Troubleshooting</a><br> -<a href="nasmdo12.html#section-12.1">Section 12.1: Common Problems</a><br> -<a href="nasmdo12.html#section-12.1.1">Section 12.1.1: NASM Generates Inefficient Code</a><br> -<a href="nasmdo12.html#section-12.1.2">Section 12.1.2: My Jumps are Out of Range</a><br> -<a href="nasmdo12.html#section-12.1.3">Section 12.1.3: <code><nobr>ORG</nobr></code> Doesn't Work</a><br> -<a href="nasmdo12.html#section-12.1.4">Section 12.1.4: <code><nobr>TIMES</nobr></code> Doesn't Work</a><br> -<a href="nasmdo12.html#section-12.2">Section 12.2: Bugs</a><br> -<p><a href="nasmdoca.html">Appendix A: Ndisasm</a><br> -<a href="nasmdoca.html#section-A.1">Section A.1: Introduction</a><br> -<a href="nasmdoca.html#section-A.2">Section A.2: Getting Started: Installation</a><br> -<a href="nasmdoca.html#section-A.3">Section A.3: Running NDISASM</a><br> -<a href="nasmdoca.html#section-A.3.1">Section A.3.1: COM Files: Specifying an Origin</a><br> -<a href="nasmdoca.html#section-A.3.2">Section A.3.2: Code Following Data: Synchronisation</a><br> -<a href="nasmdoca.html#section-A.3.3">Section A.3.3: Mixed Code and Data: Automatic (Intelligent) Synchronisation </a><br> -<a href="nasmdoca.html#section-A.3.4">Section A.3.4: Other Options</a><br> -<a href="nasmdoca.html#section-A.4">Section A.4: Bugs and Improvements</a><br> -<p><a href="nasmdocb.html">Appendix B: Instruction List</a><br> -<a href="nasmdocb.html#section-B.1">Section B.1: Introduction</a><br> -<a href="nasmdocb.html#section-B.1.1">Section B.1.1: Special instructions...</a><br> -<a href="nasmdocb.html#section-B.1.2">Section B.1.2: Conventional instructions</a><br> -<a href="nasmdocb.html#section-B.1.3">Section B.1.3: Katmai Streaming SIMD instructions (SSE -- a.k.a. KNI, XMM, MMX2)</a><br> -<a href="nasmdocb.html#section-B.1.4">Section B.1.4: Introduced in Deschutes but necessary for SSE support</a><br> -<a href="nasmdocb.html#section-B.1.5">Section B.1.5: XSAVE group (AVX and extended state)</a><br> -<a href="nasmdocb.html#section-B.1.6">Section B.1.6: Generic memory operations</a><br> -<a href="nasmdocb.html#section-B.1.7">Section B.1.7: New MMX instructions introduced in Katmai</a><br> -<a href="nasmdocb.html#section-B.1.8">Section B.1.8: AMD Enhanced 3DNow! (Athlon) instructions</a><br> -<a href="nasmdocb.html#section-B.1.9">Section B.1.9: Willamette SSE2 Cacheability Instructions</a><br> -<a href="nasmdocb.html#section-B.1.10">Section B.1.10: Willamette MMX instructions (SSE2 SIMD Integer Instructions)</a><br> -<a href="nasmdocb.html#section-B.1.11">Section B.1.11: Willamette Streaming SIMD instructions (SSE2)</a><br> -<a href="nasmdocb.html#section-B.1.12">Section B.1.12: Prescott New Instructions (SSE3)</a><br> -<a href="nasmdocb.html#section-B.1.13">Section B.1.13: VMX Instructions</a><br> -<a href="nasmdocb.html#section-B.1.14">Section B.1.14: Extended Page Tables VMX instructions</a><br> -<a href="nasmdocb.html#section-B.1.15">Section B.1.15: Tejas New Instructions (SSSE3)</a><br> -<a href="nasmdocb.html#section-B.1.16">Section B.1.16: AMD SSE4A</a><br> -<a href="nasmdocb.html#section-B.1.17">Section B.1.17: New instructions in Barcelona</a><br> -<a href="nasmdocb.html#section-B.1.18">Section B.1.18: Penryn New Instructions (SSE4.1)</a><br> -<a href="nasmdocb.html#section-B.1.19">Section B.1.19: Nehalem New Instructions (SSE4.2)</a><br> -<a href="nasmdocb.html#section-B.1.20">Section B.1.20: Intel SMX</a><br> -<a href="nasmdocb.html#section-B.1.21">Section B.1.21: Geode (Cyrix) 3DNow! additions</a><br> -<a href="nasmdocb.html#section-B.1.22">Section B.1.22: Intel new instructions in ???</a><br> -<a href="nasmdocb.html#section-B.1.23">Section B.1.23: Intel AES instructions</a><br> -<a href="nasmdocb.html#section-B.1.24">Section B.1.24: Intel AVX AES instructions</a><br> -<a href="nasmdocb.html#section-B.1.25">Section B.1.25: Intel AVX instructions</a><br> -<a href="nasmdocb.html#section-B.1.26">Section B.1.26: Intel Carry-Less Multiplication instructions (CLMUL)</a><br> -<a href="nasmdocb.html#section-B.1.27">Section B.1.27: Intel AVX Carry-Less Multiplication instructions (CLMUL)</a><br> -<a href="nasmdocb.html#section-B.1.28">Section B.1.28: Intel Fused Multiply-Add instructions (FMA)</a><br> -<a href="nasmdocb.html#section-B.1.29">Section B.1.29: VIA (Centaur) security instructions</a><br> -<a href="nasmdocb.html#section-B.1.30">Section B.1.30: AMD Lightweight Profiling (LWP) instructions</a><br> -<a href="nasmdocb.html#section-B.1.31">Section B.1.31: AMD XOP, FMA4 and CVT16 instructions (SSE5)</a><br> -<a href="nasmdocb.html#section-B.1.32">Section B.1.32: Systematic names for the hinting nop instructions</a><br> -<p><a href="nasmdocc.html">Appendix C: NASM Version History</a><br> -<a href="nasmdocc.html#section-C.1">Section C.1: NASM 2 Series</a><br> -<a href="nasmdocc.html#section-C.1.1">Section C.1.1: Version 2.08</a><br> -<a href="nasmdocc.html#section-C.1.2">Section C.1.2: Version 2.07</a><br> -<a href="nasmdocc.html#section-C.1.3">Section C.1.3: Version 2.06</a><br> -<a href="nasmdocc.html#section-C.1.4">Section C.1.4: Version 2.05.01</a><br> -<a href="nasmdocc.html#section-C.1.5">Section C.1.5: Version 2.05</a><br> -<a href="nasmdocc.html#section-C.1.6">Section C.1.6: Version 2.04</a><br> -<a href="nasmdocc.html#section-C.1.7">Section C.1.7: Version 2.03.01</a><br> -<a href="nasmdocc.html#section-C.1.8">Section C.1.8: Version 2.03</a><br> -<a href="nasmdocc.html#section-C.1.9">Section C.1.9: Version 2.02</a><br> -<a href="nasmdocc.html#section-C.1.10">Section C.1.10: Version 2.01</a><br> -<a href="nasmdocc.html#section-C.1.11">Section C.1.11: Version 2.00</a><br> -<a href="nasmdocc.html#section-C.2">Section C.2: NASM 0.98 Series</a><br> -<a href="nasmdocc.html#section-C.2.1">Section C.2.1: Version 0.98.39</a><br> -<a href="nasmdocc.html#section-C.2.2">Section C.2.2: Version 0.98.38</a><br> -<a href="nasmdocc.html#section-C.2.3">Section C.2.3: Version 0.98.37</a><br> -<a href="nasmdocc.html#section-C.2.4">Section C.2.4: Version 0.98.36</a><br> -<a href="nasmdocc.html#section-C.2.5">Section C.2.5: Version 0.98.35</a><br> -<a href="nasmdocc.html#section-C.2.6">Section C.2.6: Version 0.98.34</a><br> -<a href="nasmdocc.html#section-C.2.7">Section C.2.7: Version 0.98.33</a><br> -<a href="nasmdocc.html#section-C.2.8">Section C.2.8: Version 0.98.32</a><br> -<a href="nasmdocc.html#section-C.2.9">Section C.2.9: Version 0.98.31</a><br> -<a href="nasmdocc.html#section-C.2.10">Section C.2.10: Version 0.98.30</a><br> -<a href="nasmdocc.html#section-C.2.11">Section C.2.11: Version 0.98.28</a><br> -<a href="nasmdocc.html#section-C.2.12">Section C.2.12: Version 0.98.26</a><br> -<a href="nasmdocc.html#section-C.2.13">Section C.2.13: Version 0.98.25alt</a><br> -<a href="nasmdocc.html#section-C.2.14">Section C.2.14: Version 0.98.25</a><br> -<a href="nasmdocc.html#section-C.2.15">Section C.2.15: Version 0.98.24p1</a><br> -<a href="nasmdocc.html#section-C.2.16">Section C.2.16: Version 0.98.24</a><br> -<a href="nasmdocc.html#section-C.2.17">Section C.2.17: Version 0.98.23</a><br> -<a href="nasmdocc.html#section-C.2.18">Section C.2.18: Version 0.98.22</a><br> -<a href="nasmdocc.html#section-C.2.19">Section C.2.19: Version 0.98.21</a><br> -<a href="nasmdocc.html#section-C.2.20">Section C.2.20: Version 0.98.20</a><br> -<a href="nasmdocc.html#section-C.2.21">Section C.2.21: Version 0.98.19</a><br> -<a href="nasmdocc.html#section-C.2.22">Section C.2.22: Version 0.98.18</a><br> -<a href="nasmdocc.html#section-C.2.23">Section C.2.23: Version 0.98.17</a><br> -<a href="nasmdocc.html#section-C.2.24">Section C.2.24: Version 0.98.16</a><br> -<a href="nasmdocc.html#section-C.2.25">Section C.2.25: Version 0.98.15</a><br> -<a href="nasmdocc.html#section-C.2.26">Section C.2.26: Version 0.98.14</a><br> -<a href="nasmdocc.html#section-C.2.27">Section C.2.27: Version 0.98.13</a><br> -<a href="nasmdocc.html#section-C.2.28">Section C.2.28: Version 0.98.12</a><br> -<a href="nasmdocc.html#section-C.2.29">Section C.2.29: Version 0.98.11</a><br> -<a href="nasmdocc.html#section-C.2.30">Section C.2.30: Version 0.98.10</a><br> -<a href="nasmdocc.html#section-C.2.31">Section C.2.31: Version 0.98.09</a><br> -<a href="nasmdocc.html#section-C.2.32">Section C.2.32: Version 0.98.08</a><br> -<a href="nasmdocc.html#section-C.2.33">Section C.2.33: Version 0.98.09b with John Coffman patches released 28-Oct-2001</a><br> -<a href="nasmdocc.html#section-C.2.34">Section C.2.34: Version 0.98.07 released 01/28/01</a><br> -<a href="nasmdocc.html#section-C.2.35">Section C.2.35: Version 0.98.06f released 01/18/01</a><br> -<a href="nasmdocc.html#section-C.2.36">Section C.2.36: Version 0.98.06e released 01/09/01</a><br> -<a href="nasmdocc.html#section-C.2.37">Section C.2.37: Version 0.98p1</a><br> -<a href="nasmdocc.html#section-C.2.38">Section C.2.38: Version 0.98bf (bug-fixed)</a><br> -<a href="nasmdocc.html#section-C.2.39">Section C.2.39: Version 0.98.03 with John Coffman's changes released 27-Jul-2000</a><br> -<a href="nasmdocc.html#section-C.2.40">Section C.2.40: Version 0.98.03</a><br> -<a href="nasmdocc.html#section-C.2.41">Section C.2.41: Version 0.98</a><br> -<a href="nasmdocc.html#section-C.2.42">Section C.2.42: Version 0.98p9</a><br> -<a href="nasmdocc.html#section-C.2.43">Section C.2.43: Version 0.98p8</a><br> -<a href="nasmdocc.html#section-C.2.44">Section C.2.44: Version 0.98p7</a><br> -<a href="nasmdocc.html#section-C.2.45">Section C.2.45: Version 0.98p6</a><br> -<a href="nasmdocc.html#section-C.2.46">Section C.2.46: Version 0.98p3.7</a><br> -<a href="nasmdocc.html#section-C.2.47">Section C.2.47: Version 0.98p3.6</a><br> -<a href="nasmdocc.html#section-C.2.48">Section C.2.48: Version 0.98p3.5</a><br> -<a href="nasmdocc.html#section-C.2.49">Section C.2.49: Version 0.98p3.4</a><br> -<a href="nasmdocc.html#section-C.2.50">Section C.2.50: Version 0.98p3.3</a><br> -<a href="nasmdocc.html#section-C.2.51">Section C.2.51: Version 0.98p3.2</a><br> -<a href="nasmdocc.html#section-C.2.52">Section C.2.52: Version 0.98p3-hpa</a><br> -<a href="nasmdocc.html#section-C.2.53">Section C.2.53: Version 0.98 pre-release 3</a><br> -<a href="nasmdocc.html#section-C.2.54">Section C.2.54: Version 0.98 pre-release 2</a><br> -<a href="nasmdocc.html#section-C.2.55">Section C.2.55: Version 0.98 pre-release 1</a><br> -<a href="nasmdocc.html#section-C.3">Section C.3: NASM 0.9 Series</a><br> -<a href="nasmdocc.html#section-C.3.1">Section C.3.1: Version 0.97 released December 1997</a><br> -<a href="nasmdocc.html#section-C.3.2">Section C.3.2: Version 0.96 released November 1997</a><br> -<a href="nasmdocc.html#section-C.3.3">Section C.3.3: Version 0.95 released July 1997</a><br> -<a href="nasmdocc.html#section-C.3.4">Section C.3.4: Version 0.94 released April 1997</a><br> -<a href="nasmdocc.html#section-C.3.5">Section C.3.5: Version 0.93 released January 1997</a><br> -<a href="nasmdocc.html#section-C.3.6">Section C.3.6: Version 0.92 released January 1997</a><br> -<a href="nasmdocc.html#section-C.3.7">Section C.3.7: Version 0.91 released November 1996</a><br> -<a href="nasmdocc.html#section-C.3.8">Section C.3.8: Version 0.90 released October 1996</a><br> -<p><a href="nasmdoci.html">Index</a> -</body></html> diff --git a/doc/html/nasmdoc1.html b/doc/html/nasmdoc1.html deleted file mode 100644 index d370895..0000000 --- a/doc/html/nasmdoc1.html +++ /dev/null @@ -1,159 +0,0 @@ -<html><head><title>NASM Manual</title></head> -<body><h1 align=center>The Netwide Assembler: NASM</h1> - -<p align=center><a href="nasmdoc2.html">Next Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -<h2><a name="chapter-1">Chapter 1: Introduction</a></h2> -<h3><a name="section-1.1">1.1 What Is NASM?</a></h3> -<p>The Netwide Assembler, NASM, is an 80x86 and x86-64 assembler designed -for portability and modularity. It supports a range of object file formats, -including Linux and <code><nobr>*BSD</nobr></code> -<code><nobr>a.out</nobr></code>, <code><nobr>ELF</nobr></code>, -<code><nobr>COFF</nobr></code>, <code><nobr>Mach-O</nobr></code>, Microsoft -16-bit <code><nobr>OBJ</nobr></code>, <code><nobr>Win32</nobr></code> and -<code><nobr>Win64</nobr></code>. It will also output plain binary files. -Its syntax is designed to be simple and easy to understand, similar to -Intel's but less complex. It supports all currently known x86 architectural -extensions, and has strong support for macros. -<h4><a name="section-1.1.1">1.1.1 Why Yet Another Assembler?</a></h4> -<p>The Netwide Assembler grew out of an idea on -<code><nobr>comp.lang.asm.x86</nobr></code> (or possibly -<code><nobr>alt.lang.asm</nobr></code> - I forget which), which was -essentially that there didn't seem to be a good <em>free</em> x86-series -assembler around, and that maybe someone ought to write one. -<ul> -<li><code><nobr>a86</nobr></code> is good, but not free, and in particular -you don't get any 32-bit capability until you pay. It's DOS only, too. -<li><code><nobr>gas</nobr></code> is free, and ports over to DOS and Unix, -but it's not very good, since it's designed to be a back end to -<code><nobr>gcc</nobr></code>, which always feeds it correct code. So its -error checking is minimal. Also, its syntax is horrible, from the point of -view of anyone trying to actually <em>write</em> anything in it. Plus you -can't write 16-bit code in it (properly.) -<li><code><nobr>as86</nobr></code> is specific to Minix and Linux, and (my -version at least) doesn't seem to have much (or any) documentation. -<li><code><nobr>MASM</nobr></code> isn't very good, and it's (was) -expensive, and it runs only under DOS. -<li><code><nobr>TASM</nobr></code> is better, but still strives for MASM -compatibility, which means millions of directives and tons of red tape. And -its syntax is essentially MASM's, with the contradictions and quirks that -entails (although it sorts out some of those by means of Ideal mode.) It's -expensive too. And it's DOS-only. -</ul> -<p>So here, for your coding pleasure, is NASM. At present it's still in -prototype stage - we don't promise that it can outperform any of these -assemblers. But please, <em>please</em> send us bug reports, fixes, helpful -information, and anything else you can get your hands on (and thanks to the -many people who've done this already! You all know who you are), and we'll -improve it out of all recognition. Again. -<h4><a name="section-1.1.2">1.1.2 License Conditions</a></h4> -<p>Please see the file <code><nobr>LICENSE</nobr></code>, supplied as part -of any NASM distribution archive, for the license conditions under which -you may use NASM. NASM is now under the so-called 2-clause BSD license, -also known as the simplified BSD license. -<p>Copyright 1996-2009 the NASM Authors - All rights reserved. -<p>Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are met: -<ul> -<li>Redistributions of source code must retain the above copyright notice, -this list of conditions and the following disclaimer. -<li>Redistributions in binary form must reproduce the above copyright -notice, this list of conditions and the following disclaimer in the -documentation and/or other materials provided with the distribution. -</ul> -<p>THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS -IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, -THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR -PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR -CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, -EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, -PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; -OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, -WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR -OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF -ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. -<h3><a name="section-1.2">1.2 Contact Information</a></h3> -<p>The current version of NASM (since about 0.98.08) is maintained by a -team of developers, accessible through the -<code><nobr>nasm-devel</nobr></code> mailing list (see below for the link). -If you want to report a bug, please read -<a href="nasmdo12.html#section-12.2">section 12.2</a> first. -<p>NASM has a website at -<a href="http://www.nasm.us/"><code><nobr>http://www.nasm.us/</nobr></code></a>. -If it's not there, google for us! -<p>New releases, release candidates, and daily development snapshots of -NASM are available from the official web site. -<p>Announcements are posted to -<a href="news:comp.lang.asm.x86"><code><nobr>comp.lang.asm.x86</nobr></code></a>, -and to the web site -<a href="http://www.freshmeat.net/"><code><nobr>http://www.freshmeat.net/</nobr></code></a>. -<p>If you want information about the current development status, please -subscribe to the <code><nobr>nasm-devel</nobr></code> email list; see link -from the website. -<h3><a name="section-1.3">1.3 Installation</a></h3> -<h4><a name="section-1.3.1">1.3.1 Installing NASM under MS-DOS or Windows</a></h4> -<p>Once you've obtained the appropriate archive for NASM, -<code><nobr>nasm-XXX-dos.zip</nobr></code> or -<code><nobr>nasm-XXX-win32.zip</nobr></code> (where -<code><nobr>XXX</nobr></code> denotes the version number of NASM contained -in the archive), unpack it into its own directory (for example -<code><nobr>c:\nasm</nobr></code>). -<p>The archive will contain a set of executable files: the NASM executable -file <code><nobr>nasm.exe</nobr></code>, the NDISASM executable file -<code><nobr>ndisasm.exe</nobr></code>, and possibly additional utilities to -handle the RDOFF file format. -<p>The only file NASM needs to run is its own executable, so copy -<code><nobr>nasm.exe</nobr></code> to a directory on your PATH, or -alternatively edit <code><nobr>autoexec.bat</nobr></code> to add the -<code><nobr>nasm</nobr></code> directory to your -<code><nobr>PATH</nobr></code> (to do that under Windows XP, go to Start -> Control Panel > System > Advanced > Environment Variables; -these instructions may work under other versions of Windows as well.) -<p>That's it - NASM is installed. You don't need the nasm directory to be -present to run NASM (unless you've added it to your -<code><nobr>PATH</nobr></code>), so you can delete it if you need to save -space; however, you may want to keep the documentation or test programs. -<p>If you've downloaded the DOS source archive, -<code><nobr>nasm-XXX.zip</nobr></code>, the <code><nobr>nasm</nobr></code> -directory will also contain the full NASM source code, and a selection of -Makefiles you can (hopefully) use to rebuild your copy of NASM from -scratch. See the file <code><nobr>INSTALL</nobr></code> in the source -archive. -<p>Note that a number of files are generated from other files by Perl -scripts. Although the NASM source distribution includes these generated -files, you will need to rebuild them (and hence, will need a Perl -interpreter) if you change insns.dat, standard.mac or the documentation. It -is possible future source distributions may not include these files at all. -Ports of Perl for a variety of platforms, including DOS and Windows, are -available from <a href="http://www.cpan.org/ports/">www.cpan.org</a>. -<h4><a name="section-1.3.2">1.3.2 Installing NASM under Unix</a></h4> -<p>Once you've obtained the Unix source archive for NASM, -<code><nobr>nasm-XXX.tar.gz</nobr></code> (where -<code><nobr>XXX</nobr></code> denotes the version number of NASM contained -in the archive), unpack it into a directory such as -<code><nobr>/usr/local/src</nobr></code>. The archive, when unpacked, will -create its own subdirectory <code><nobr>nasm-XXX</nobr></code>. -<p>NASM is an auto-configuring package: once you've unpacked it, -<code><nobr>cd</nobr></code> to the directory it's been unpacked into and -type <code><nobr>./configure</nobr></code>. This shell script will find the -best C compiler to use for building NASM and set up Makefiles accordingly. -<p>Once NASM has auto-configured, you can type -<code><nobr>make</nobr></code> to build the <code><nobr>nasm</nobr></code> -and <code><nobr>ndisasm</nobr></code> binaries, and then -<code><nobr>make install</nobr></code> to install them in -<code><nobr>/usr/local/bin</nobr></code> and install the man pages -<code><nobr>nasm.1</nobr></code> and <code><nobr>ndisasm.1</nobr></code> in -<code><nobr>/usr/local/man/man1</nobr></code>. Alternatively, you can give -options such as <code><nobr>--prefix</nobr></code> to the configure script -(see the file <code><nobr>INSTALL</nobr></code> for more details), or -install the programs yourself. -<p>NASM also comes with a set of utilities for handling the -<code><nobr>RDOFF</nobr></code> custom object-file format, which are in the -<code><nobr>rdoff</nobr></code> subdirectory of the NASM archive. You can -build these with <code><nobr>make rdf</nobr></code> and install them with -<code><nobr>make rdf_install</nobr></code>, if you want them. -<p align=center><a href="nasmdoc2.html">Next Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -</body></html> diff --git a/doc/html/nasmdoc2.html b/doc/html/nasmdoc2.html deleted file mode 100644 index a639a4a..0000000 --- a/doc/html/nasmdoc2.html +++ /dev/null @@ -1,650 +0,0 @@ -<html><head><title>NASM Manual</title></head> -<body><h1 align=center>The Netwide Assembler: NASM</h1> - -<p align=center><a href="nasmdoc3.html">Next Chapter</a> | -<a href="nasmdoc1.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -<h2><a name="chapter-2">Chapter 2: Running NASM</a></h2> -<h3><a name="section-2.1">2.1 NASM Command-Line Syntax</a></h3> -<p>To assemble a file, you issue a command of the form -<p><pre> -nasm -f <format> <filename> [-o <output>] -</pre> -<p>For example, -<p><pre> -nasm -f elf myfile.asm -</pre> -<p>will assemble <code><nobr>myfile.asm</nobr></code> into an -<code><nobr>ELF</nobr></code> object file -<code><nobr>myfile.o</nobr></code>. And -<p><pre> -nasm -f bin myfile.asm -o myfile.com -</pre> -<p>will assemble <code><nobr>myfile.asm</nobr></code> into a raw binary -file <code><nobr>myfile.com</nobr></code>. -<p>To produce a listing file, with the hex codes output from NASM displayed -on the left of the original sources, use the <code><nobr>-l</nobr></code> -option to give a listing file name, for example: -<p><pre> -nasm -f coff myfile.asm -l myfile.lst -</pre> -<p>To get further usage instructions from NASM, try typing -<p><pre> -nasm -h -</pre> -<p>As <code><nobr>-hf</nobr></code>, this will also list the available -output file formats, and what they are. -<p>If you use Linux but aren't sure whether your system is -<code><nobr>a.out</nobr></code> or <code><nobr>ELF</nobr></code>, type -<p><pre> -file nasm -</pre> -<p>(in the directory in which you put the NASM binary when you installed -it). If it says something like -<p><pre> -nasm: ELF 32-bit LSB executable i386 (386 and up) Version 1 -</pre> -<p>then your system is <code><nobr>ELF</nobr></code>, and you should use -the option <code><nobr>-f elf</nobr></code> when you want NASM to produce -Linux object files. If it says -<p><pre> -nasm: Linux/i386 demand-paged executable (QMAGIC) -</pre> -<p>or something similar, your system is <code><nobr>a.out</nobr></code>, -and you should use <code><nobr>-f aout</nobr></code> instead (Linux -<code><nobr>a.out</nobr></code> systems have long been obsolete, and are -rare these days.) -<p>Like Unix compilers and assemblers, NASM is silent unless it goes wrong: -you won't see any output at all, unless it gives error messages. -<h4><a name="section-2.1.1">2.1.1 The <code><nobr>-o</nobr></code> Option: Specifying the Output File Name</a></h4> -<p>NASM will normally choose the name of your output file for you; -precisely how it does this is dependent on the object file format. For -Microsoft object file formats (<code><nobr>obj</nobr></code>, -<code><nobr>win32</nobr></code> and <code><nobr>win64</nobr></code>), it -will remove the <code><nobr>.asm</nobr></code> extension (or whatever -extension you like to use - NASM doesn't care) from your source file name -and substitute <code><nobr>.obj</nobr></code>. For Unix object file formats -(<code><nobr>aout</nobr></code>, <code><nobr>as86</nobr></code>, -<code><nobr>coff</nobr></code>, <code><nobr>elf32</nobr></code>, -<code><nobr>elf64</nobr></code>, <code><nobr>ieee</nobr></code>, -<code><nobr>macho32</nobr></code> and <code><nobr>macho64</nobr></code>) it -will substitute <code><nobr>.o</nobr></code>. For -<code><nobr>dbg</nobr></code>, <code><nobr>rdf</nobr></code>, -<code><nobr>ith</nobr></code> and <code><nobr>srec</nobr></code>, it will -use <code><nobr>.dbg</nobr></code>, <code><nobr>.rdf</nobr></code>, -<code><nobr>.ith</nobr></code> and <code><nobr>.srec</nobr></code>, -respectively, and for the <code><nobr>bin</nobr></code> format it will -simply remove the extension, so that <code><nobr>myfile.asm</nobr></code> -produces the output file <code><nobr>myfile</nobr></code>. -<p>If the output file already exists, NASM will overwrite it, unless it has -the same name as the input file, in which case it will give a warning and -use <code><nobr>nasm.out</nobr></code> as the output file name instead. -<p>For situations in which this behaviour is unacceptable, NASM provides -the <code><nobr>-o</nobr></code> command-line option, which allows you to -specify your desired output file name. You invoke -<code><nobr>-o</nobr></code> by following it with the name you wish for the -output file, either with or without an intervening space. For example: -<p><pre> -nasm -f bin program.asm -o program.com -nasm -f bin driver.asm -odriver.sys -</pre> -<p>Note that this is a small o, and is different from a capital O , which -is used to specify the number of optimisation passes required. See -<a href="#section-2.1.22">section 2.1.22</a>. -<h4><a name="section-2.1.2">2.1.2 The <code><nobr>-f</nobr></code> Option: Specifying the Output File Format</a></h4> -<p>If you do not supply the <code><nobr>-f</nobr></code> option to NASM, it -will choose an output file format for you itself. In the distribution -versions of NASM, the default is always <code><nobr>bin</nobr></code>; if -you've compiled your own copy of NASM, you can redefine -<code><nobr>OF_DEFAULT</nobr></code> at compile time and choose what you -want the default to be. -<p>Like <code><nobr>-o</nobr></code>, the intervening space between -<code><nobr>-f</nobr></code> and the output file format is optional; so -<code><nobr>-f elf</nobr></code> and <code><nobr>-felf</nobr></code> are -both valid. -<p>A complete list of the available output file formats can be given by -issuing the command <code><nobr>nasm -hf</nobr></code>. -<h4><a name="section-2.1.3">2.1.3 The <code><nobr>-l</nobr></code> Option: Generating a Listing File</a></h4> -<p>If you supply the <code><nobr>-l</nobr></code> option to NASM, followed -(with the usual optional space) by a file name, NASM will generate a -source-listing file for you, in which addresses and generated code are -listed on the left, and the actual source code, with expansions of -multi-line macros (except those which specifically request no expansion in -source listings: see <a href="nasmdoc4.html#section-4.3.10">section -4.3.10</a>) on the right. For example: -<p><pre> -nasm -f elf myfile.asm -l myfile.lst -</pre> -<p>If a list file is selected, you may turn off listing for a section of -your source with <code><nobr>[list -]</nobr></code>, and turn it back on -with <code><nobr>[list +]</nobr></code>, (the default, obviously). There is -no "user form" (without the brackets). This can be used to list only -sections of interest, avoiding excessively long listings. -<h4><a name="section-2.1.4">2.1.4 The <code><nobr>-M</nobr></code> Option: Generate Makefile Dependencies</a></h4> -<p>This option can be used to generate makefile dependencies on stdout. -This can be redirected to a file for further processing. For example: -<p><pre> -nasm -M myfile.asm > myfile.dep -</pre> -<h4><a name="section-2.1.5">2.1.5 The <code><nobr>-MG</nobr></code> Option: Generate Makefile Dependencies</a></h4> -<p>This option can be used to generate makefile dependencies on stdout. -This differs from the <code><nobr>-M</nobr></code> option in that if a -nonexisting file is encountered, it is assumed to be a generated file and -is added to the dependency list without a prefix. -<h4><a name="section-2.1.6">2.1.6 The <code><nobr>-MF</nobr></code> Option: Set Makefile Dependency File</a></h4> -<p>This option can be used with the <code><nobr>-M</nobr></code> or -<code><nobr>-MG</nobr></code> options to send the output to a file, rather -than to stdout. For example: -<p><pre> -nasm -M -MF myfile.dep myfile.asm -</pre> -<h4><a name="section-2.1.7">2.1.7 The <code><nobr>-MD</nobr></code> Option: Assemble and Generate Dependencies</a></h4> -<p>The <code><nobr>-MD</nobr></code> option acts as the combination of the -<code><nobr>-M</nobr></code> and <code><nobr>-MF</nobr></code> options -(i.e. a filename has to be specified.) However, unlike the -<code><nobr>-M</nobr></code> or <code><nobr>-MG</nobr></code> options, -<code><nobr>-MD</nobr></code> does <em>not</em> inhibit the normal -operation of the assembler. Use this to automatically generate updated -dependencies with every assembly session. For example: -<p><pre> -nasm -f elf -o myfile.o -MD myfile.dep myfile.asm -</pre> -<h4><a name="section-2.1.8">2.1.8 The <code><nobr>-MT</nobr></code> Option: Dependency Target Name</a></h4> -<p>The <code><nobr>-MT</nobr></code> option can be used to override the -default name of the dependency target. This is normally the same as the -output filename, specified by the <code><nobr>-o</nobr></code> option. -<h4><a name="section-2.1.9">2.1.9 The <code><nobr>-MQ</nobr></code> Option: Dependency Target Name (Quoted)</a></h4> -<p>The <code><nobr>-MQ</nobr></code> option acts as the -<code><nobr>-MT</nobr></code> option, except it tries to quote characters -that have special meaning in Makefile syntax. This is not foolproof, as not -all characters with special meaning are quotable in Make. -<h4><a name="section-2.1.10">2.1.10 The <code><nobr>-MP</nobr></code> Option: Emit phony targets</a></h4> -<p>When used with any of the dependency generation options, the -<code><nobr>-MP</nobr></code> option causes NASM to emit a phony target -without dependencies for each header file. This prevents Make from -complaining if a header file has been removed. -<h4><a name="section-2.1.11">2.1.11 The <code><nobr>-F</nobr></code> Option: Selecting a Debug Information Format</a></h4> -<p>This option is used to select the format of the debug information -emitted into the output file, to be used by a debugger (or <em>will</em> -be). Prior to version 2.03.01, the use of this switch did <em>not</em> -enable output of the selected debug info format. Use -<code><nobr>-g</nobr></code>, see <a href="#section-2.1.12">section -2.1.12</a>, to enable output. Versions 2.03.01 and later automatically -enable <code><nobr>-g</nobr></code> if <code><nobr>-F</nobr></code> is -specified. -<p>A complete list of the available debug file formats for an output format -can be seen by issuing the command -<code><nobr>nasm -f <format> -y</nobr></code>. Not all output formats -currently support debugging output. See <a href="#section-2.1.26">section -2.1.26</a>. -<p>This should not be confused with the <code><nobr>-f dbg</nobr></code> -output format option which is not built into NASM by default. For -information on how to enable it when building from the sources, see -<a href="nasmdoc7.html#section-7.14">section 7.14</a>. -<h4><a name="section-2.1.12">2.1.12 The <code><nobr>-g</nobr></code> Option: Enabling Debug Information.</a></h4> -<p>This option can be used to generate debugging information in the -specified format. See <a href="#section-2.1.11">section 2.1.11</a>. Using -<code><nobr>-g</nobr></code> without <code><nobr>-F</nobr></code> results -in emitting debug info in the default format, if any, for the selected -output format. If no debug information is currently implemented in the -selected output format, <code><nobr>-g</nobr></code> is <em>silently -ignored</em>. -<h4><a name="section-2.1.13">2.1.13 The <code><nobr>-X</nobr></code> Option: Selecting an Error Reporting Format</a></h4> -<p>This option can be used to select an error reporting format for any -error messages that might be produced by NASM. -<p>Currently, two error reporting formats may be selected. They are the -<code><nobr>-Xvc</nobr></code> option and the -<code><nobr>-Xgnu</nobr></code> option. The GNU format is the default and -looks like this: -<p><pre> -filename.asm:65: error: specific error message -</pre> -<p>where <code><nobr>filename.asm</nobr></code> is the name of the source -file in which the error was detected, <code><nobr>65</nobr></code> is the -source file line number on which the error was detected, -<code><nobr>error</nobr></code> is the severity of the error (this could be -<code><nobr>warning</nobr></code>), and -<code><nobr>specific error message</nobr></code> is a more detailed text -message which should help pinpoint the exact problem. -<p>The other format, specified by <code><nobr>-Xvc</nobr></code> is the -style used by Microsoft Visual C++ and some other programs. It looks like -this: -<p><pre> -filename.asm(65) : error: specific error message -</pre> -<p>where the only difference is that the line number is in parentheses -instead of being delimited by colons. -<p>See also the <code><nobr>Visual C++</nobr></code> output format, -<a href="nasmdoc7.html#section-7.5">section 7.5</a>. -<h4><a name="section-2.1.14">2.1.14 The <code><nobr>-Z</nobr></code> Option: Send Errors to a File</a></h4> -<p>Under <code><nobr>MS-DOS</nobr></code> it can be difficult (though there -are ways) to redirect the standard-error output of a program to a file. -Since NASM usually produces its warning and error messages on -<code><nobr>stderr</nobr></code>, this can make it hard to capture the -errors if (for example) you want to load them into an editor. -<p>NASM therefore provides the <code><nobr>-Z</nobr></code> option, taking -a filename argument which causes errors to be sent to the specified files -rather than standard error. Therefore you can redirect the errors into a -file by typing -<p><pre> -nasm -Z myfile.err -f obj myfile.asm -</pre> -<p>In earlier versions of NASM, this option was called -<code><nobr>-E</nobr></code>, but it was changed since -<code><nobr>-E</nobr></code> is an option conventionally used for -preprocessing only, with disastrous results. See -<a href="#section-2.1.20">section 2.1.20</a>. -<h4><a name="section-2.1.15">2.1.15 The <code><nobr>-s</nobr></code> Option: Send Errors to <code><nobr>stdout</nobr></code></a></h4> -<p>The <code><nobr>-s</nobr></code> option redirects error messages to -<code><nobr>stdout</nobr></code> rather than -<code><nobr>stderr</nobr></code>, so it can be redirected under -<code><nobr>MS-DOS</nobr></code>. To assemble the file -<code><nobr>myfile.asm</nobr></code> and pipe its output to the -<code><nobr>more</nobr></code> program, you can type: -<p><pre> -nasm -s -f obj myfile.asm | more -</pre> -<p>See also the <code><nobr>-Z</nobr></code> option, -<a href="#section-2.1.14">section 2.1.14</a>. -<h4><a name="section-2.1.16">2.1.16 The <code><nobr>-i</nobr></code> Option: Include File Search Directories</a></h4> -<p>When NASM sees the <code><nobr>%include</nobr></code> or -<code><nobr>%pathsearch</nobr></code> directive in a source file (see -<a href="nasmdoc4.html#section-4.6.1">section 4.6.1</a>, -<a href="nasmdoc4.html#section-4.6.2">section 4.6.2</a> or -<a href="nasmdoc3.html#section-3.2.3">section 3.2.3</a>), it will search -for the given file not only in the current directory, but also in any -directories specified on the command line by the use of the -<code><nobr>-i</nobr></code> option. Therefore you can include files from a -macro library, for example, by typing -<p><pre> -nasm -ic:\macrolib\ -f obj myfile.asm -</pre> -<p>(As usual, a space between <code><nobr>-i</nobr></code> and the path -name is allowed, and optional). -<p>NASM, in the interests of complete source-code portability, does not -understand the file naming conventions of the OS it is running on; the -string you provide as an argument to the <code><nobr>-i</nobr></code> -option will be prepended exactly as written to the name of the include -file. Therefore the trailing backslash in the above example is necessary. -Under Unix, a trailing forward slash is similarly necessary. -<p>(You can use this to your advantage, if you're really perverse, by -noting that the option <code><nobr>-ifoo</nobr></code> will cause -<code><nobr>%include "bar.i"</nobr></code> to search for the file -<code><nobr>foobar.i</nobr></code>...) -<p>If you want to define a <em>standard</em> include search path, similar -to <code><nobr>/usr/include</nobr></code> on Unix systems, you should place -one or more <code><nobr>-i</nobr></code> directives in the -<code><nobr>NASMENV</nobr></code> environment variable (see -<a href="#section-2.1.28">section 2.1.28</a>). -<p>For Makefile compatibility with many C compilers, this option can also -be specified as <code><nobr>-I</nobr></code>. -<h4><a name="section-2.1.17">2.1.17 The <code><nobr>-p</nobr></code> Option: Pre-Include a File</a></h4> -<p>NASM allows you to specify files to be <em>pre-included</em> into your -source file, by the use of the <code><nobr>-p</nobr></code> option. So -running -<p><pre> -nasm myfile.asm -p myinc.inc -</pre> -<p>is equivalent to running <code><nobr>nasm myfile.asm</nobr></code> and -placing the directive <code><nobr>%include "myinc.inc"</nobr></code> at the -start of the file. -<p>For consistency with the <code><nobr>-I</nobr></code>, -<code><nobr>-D</nobr></code> and <code><nobr>-U</nobr></code> options, this -option can also be specified as <code><nobr>-P</nobr></code>. -<h4><a name="section-2.1.18">2.1.18 The <code><nobr>-d</nobr></code> Option: Pre-Define a Macro</a></h4> -<p>Just as the <code><nobr>-p</nobr></code> option gives an alternative to -placing <code><nobr>%include</nobr></code> directives at the start of a -source file, the <code><nobr>-d</nobr></code> option gives an alternative -to placing a <code><nobr>%define</nobr></code> directive. You could code -<p><pre> -nasm myfile.asm -dFOO=100 -</pre> -<p>as an alternative to placing the directive -<p><pre> -%define FOO 100 -</pre> -<p>at the start of the file. You can miss off the macro value, as well: the -option <code><nobr>-dFOO</nobr></code> is equivalent to coding -<code><nobr>%define FOO</nobr></code>. This form of the directive may be -useful for selecting assembly-time options which are then tested using -<code><nobr>%ifdef</nobr></code>, for example -<code><nobr>-dDEBUG</nobr></code>. -<p>For Makefile compatibility with many C compilers, this option can also -be specified as <code><nobr>-D</nobr></code>. -<h4><a name="section-2.1.19">2.1.19 The <code><nobr>-u</nobr></code> Option: Undefine a Macro</a></h4> -<p>The <code><nobr>-u</nobr></code> option undefines a macro that would -otherwise have been pre-defined, either automatically or by a -<code><nobr>-p</nobr></code> or <code><nobr>-d</nobr></code> option -specified earlier on the command lines. -<p>For example, the following command line: -<p><pre> -nasm myfile.asm -dFOO=100 -uFOO -</pre> -<p>would result in <code><nobr>FOO</nobr></code> <em>not</em> being a -predefined macro in the program. This is useful to override options -specified at a different point in a Makefile. -<p>For Makefile compatibility with many C compilers, this option can also -be specified as <code><nobr>-U</nobr></code>. -<h4><a name="section-2.1.20">2.1.20 The <code><nobr>-E</nobr></code> Option: Preprocess Only</a></h4> -<p>NASM allows the preprocessor to be run on its own, up to a point. Using -the <code><nobr>-E</nobr></code> option (which requires no arguments) will -cause NASM to preprocess its input file, expand all the macro references, -remove all the comments and preprocessor directives, and print the -resulting file on standard output (or save it to a file, if the -<code><nobr>-o</nobr></code> option is also used). -<p>This option cannot be applied to programs which require the preprocessor -to evaluate expressions which depend on the values of symbols: so code such -as -<p><pre> -%assign tablesize ($-tablestart) -</pre> -<p>will cause an error in preprocess-only mode. -<p>For compatiblity with older version of NASM, this option can also be -written <code><nobr>-e</nobr></code>. <code><nobr>-E</nobr></code> in older -versions of NASM was the equivalent of the current -<code><nobr>-Z</nobr></code> option, <a href="#section-2.1.14">section -2.1.14</a>. -<h4><a name="section-2.1.21">2.1.21 The <code><nobr>-a</nobr></code> Option: Don't Preprocess At All</a></h4> -<p>If NASM is being used as the back end to a compiler, it might be -desirable to suppress preprocessing completely and assume the compiler has -already done it, to save time and increase compilation speeds. The -<code><nobr>-a</nobr></code> option, requiring no argument, instructs NASM -to replace its powerful preprocessor with a stub preprocessor which does -nothing. -<h4><a name="section-2.1.22">2.1.22 The <code><nobr>-O</nobr></code> Option: Specifying Multipass Optimization</a></h4> -<p>NASM defaults to not optimizing operands which can fit into a signed -byte. This means that if you want the shortest possible object code, you -have to enable optimization. -<p>Using the <code><nobr>-O</nobr></code> option, you can tell NASM to -carry out different levels of optimization. The syntax is: -<ul> -<li><code><nobr>-O0</nobr></code>: No optimization. All operands take their -long forms, if a short form is not specified, except conditional jumps. -This is intended to match NASM 0.98 behavior. -<li><code><nobr>-O1</nobr></code>: Minimal optimization. As above, but -immediate operands which will fit in a signed byte are optimized, unless -the long form is specified. Conditional jumps default to the long form -unless otherwise specified. -<li><code><nobr>-Ox</nobr></code> (where <code><nobr>x</nobr></code> is the -actual letter <code><nobr>x</nobr></code>): Multipass optimization. -Minimize branch offsets and signed immediate bytes, overriding size -specification unless the <code><nobr>strict</nobr></code> keyword has been -used (see <a href="nasmdoc3.html#section-3.7">section 3.7</a>). For -compatability with earlier releases, the letter <code><nobr>x</nobr></code> -may also be any number greater than one. This number has no effect on the -actual number of passes. -</ul> -<p>The <code><nobr>-Ox</nobr></code> mode is recommended for most uses. -<p>Note that this is a capital <code><nobr>O</nobr></code>, and is -different from a small <code><nobr>o</nobr></code>, which is used to -specify the output file name. See <a href="#section-2.1.1">section -2.1.1</a>. -<h4><a name="section-2.1.23">2.1.23 The <code><nobr>-t</nobr></code> Option: Enable TASM Compatibility Mode</a></h4> -<p>NASM includes a limited form of compatibility with Borland's -<code><nobr>TASM</nobr></code>. When NASM's <code><nobr>-t</nobr></code> -option is used, the following changes are made: -<ul> -<li>local labels may be prefixed with <code><nobr>@@</nobr></code> instead -of <code><nobr>.</nobr></code> -<li>size override is supported within brackets. In TASM compatible mode, a -size override inside square brackets changes the size of the operand, and -not the address type of the operand as it does in NASM syntax. E.g. -<code><nobr>mov eax,[DWORD val]</nobr></code> is valid syntax in TASM -compatibility mode. Note that you lose the ability to override the default -address type for the instruction. -<li>unprefixed forms of some directives supported -(<code><nobr>arg</nobr></code>, <code><nobr>elif</nobr></code>, -<code><nobr>else</nobr></code>, <code><nobr>endif</nobr></code>, -<code><nobr>if</nobr></code>, <code><nobr>ifdef</nobr></code>, -<code><nobr>ifdifi</nobr></code>, <code><nobr>ifndef</nobr></code>, -<code><nobr>include</nobr></code>, <code><nobr>local</nobr></code>) -</ul> -<h4><a name="section-2.1.24">2.1.24 The <code><nobr>-w</nobr></code> and <code><nobr>-W</nobr></code> Options: Enable or Disable Assembly Warnings</a></h4> -<p>NASM can observe many conditions during the course of assembly which are -worth mentioning to the user, but not a sufficiently severe error to -justify NASM refusing to generate an output file. These conditions are -reported like errors, but come up with the word `warning' before the -message. Warnings do not prevent NASM from generating an output file and -returning a success status to the operating system. -<p>Some conditions are even less severe than that: they are only sometimes -worth mentioning to the user. Therefore NASM supports the -<code><nobr>-w</nobr></code> command-line option, which enables or disables -certain classes of assembly warning. Such warning classes are described by -a name, for example <code><nobr>orphan-labels</nobr></code>; you can enable -warnings of this class by the command-line option -<code><nobr>-w+orphan-labels</nobr></code> and disable it by -<code><nobr>-w-orphan-labels</nobr></code>. -<p>The suppressible warning classes are: -<ul> -<li><code><nobr>macro-params</nobr></code> covers warnings about multi-line -macros being invoked with the wrong number of parameters. This warning -class is enabled by default; see -<a href="nasmdoc4.html#section-4.3.2">section 4.3.2</a> for an example of -why you might want to disable it. -<li><code><nobr>macro-selfref</nobr></code> warns if a macro references -itself. This warning class is disabled by default. -<li><code><nobr>macro-defaults</nobr></code> warns when a macro has more -default parameters than optional parameters. This warning class is enabled -by default; see <a href="nasmdoc4.html#section-4.3.5">section 4.3.5</a> for -why you might want to disable it. -<li><code><nobr>orphan-labels</nobr></code> covers warnings about source -lines which contain no instruction but define a label without a trailing -colon. NASM warns about this somewhat obscure condition by default; see -<a href="nasmdoc3.html#section-3.1">section 3.1</a> for more information. -<li><code><nobr>number-overflow</nobr></code> covers warnings about numeric -constants which don't fit in 64 bits. This warning class is enabled by -default. -<li><code><nobr>gnu-elf-extensions</nobr></code> warns if 8-bit or 16-bit -relocations are used in <code><nobr>-f elf</nobr></code> format. The GNU -extensions allow this. This warning class is disabled by default. -<li><code><nobr>float-overflow</nobr></code> warns about floating point -overflow. Enabled by default. -<li><code><nobr>float-denorm</nobr></code> warns about floating point -denormals. Disabled by default. -<li><code><nobr>float-underflow</nobr></code> warns about floating point -underflow. Disabled by default. -<li><code><nobr>float-toolong</nobr></code> warns about too many digits in -floating-point numbers. Enabled by default. -<li><code><nobr>user</nobr></code> controls -<code><nobr>%warning</nobr></code> directives (see -<a href="nasmdoc4.html#section-4.9">section 4.9</a>). Enabled by default. -<li><code><nobr>error</nobr></code> causes warnings to be treated as -errors. Disabled by default. -<li><code><nobr>all</nobr></code> is an alias for <em>all</em> suppressible -warning classes (not including <code><nobr>error</nobr></code>). Thus, -<code><nobr>-w+all</nobr></code> enables all available warnings. -</ul> -<p>In addition, you can set warning classes across sections. Warning -classes may be enabled with -<code><nobr>[warning +warning-name]</nobr></code>, disabled with -<code><nobr>[warning -warning-name]</nobr></code> or reset to their -original value with <code><nobr>[warning *warning-name]</nobr></code>. No -"user form" (without the brackets) exists. -<p>Since version 2.00, NASM has also supported the gcc-like syntax -<code><nobr>-Wwarning</nobr></code> and -<code><nobr>-Wno-warning</nobr></code> instead of -<code><nobr>-w+warning</nobr></code> and -<code><nobr>-w-warning</nobr></code>, respectively. -<h4><a name="section-2.1.25">2.1.25 The <code><nobr>-v</nobr></code> Option: Display Version Info</a></h4> -<p>Typing <code><nobr>NASM -v</nobr></code> will display the version of -NASM which you are using, and the date on which it was compiled. -<p>You will need the version number if you report a bug. -<h4><a name="section-2.1.26">2.1.26 The <code><nobr>-y</nobr></code> Option: Display Available Debug Info Formats</a></h4> -<p>Typing <code><nobr>nasm -f <option> -y</nobr></code> will display -a list of the available debug info formats for the given output format. The -default format is indicated by an asterisk. For example: -<p><pre> -nasm -f elf -y -</pre> -<p><pre> -valid debug formats for 'elf32' output format are - ('*' denotes default): - * stabs ELF32 (i386) stabs debug format for Linux - dwarf elf32 (i386) dwarf debug format for Linux -</pre> -<h4><a name="section-2.1.27">2.1.27 The <code><nobr>--prefix</nobr></code> and <code><nobr>--postfix</nobr></code> Options.</a></h4> -<p>The <code><nobr>--prefix</nobr></code> and -<code><nobr>--postfix</nobr></code> options prepend or append -(respectively) the given argument to all <code><nobr>global</nobr></code> -or <code><nobr>extern</nobr></code> variables. E.g. -<code><nobr>--prefix _</nobr></code> will prepend the underscore to all -global and external variables, as C sometimes (but not always) likes it. -<h4><a name="section-2.1.28">2.1.28 The <code><nobr>NASMENV</nobr></code> Environment Variable</a></h4> -<p>If you define an environment variable called -<code><nobr>NASMENV</nobr></code>, the program will interpret it as a list -of extra command-line options, which are processed before the real command -line. You can use this to define standard search directories for include -files, by putting <code><nobr>-i</nobr></code> options in the -<code><nobr>NASMENV</nobr></code> variable. -<p>The value of the variable is split up at white space, so that the value -<code><nobr>-s -ic:\nasmlib\</nobr></code> will be treated as two separate -options. However, that means that the value -<code><nobr>-dNAME="my name"</nobr></code> won't do what you might want, -because it will be split at the space and the NASM command-line processing -will get confused by the two nonsensical words -<code><nobr>-dNAME="my</nobr></code> and <code><nobr>name"</nobr></code>. -<p>To get round this, NASM provides a feature whereby, if you begin the -<code><nobr>NASMENV</nobr></code> environment variable with some character -that isn't a minus sign, then NASM will treat this character as the -separator character for options. So setting the -<code><nobr>NASMENV</nobr></code> variable to the value -<code><nobr>!-s!-ic:\nasmlib\</nobr></code> is equivalent to setting it to -<code><nobr>-s -ic:\nasmlib\</nobr></code>, but -<code><nobr>!-dNAME="my name"</nobr></code> will work. -<p>This environment variable was previously called -<code><nobr>NASM</nobr></code>. This was changed with version 0.98.31. -<h3><a name="section-2.2">2.2 Quick Start for MASM Users</a></h3> -<p>If you're used to writing programs with MASM, or with TASM in -MASM-compatible (non-Ideal) mode, or with <code><nobr>a86</nobr></code>, -this section attempts to outline the major differences between MASM's -syntax and NASM's. If you're not already used to MASM, it's probably worth -skipping this section. -<h4><a name="section-2.2.1">2.2.1 NASM Is Case-Sensitive</a></h4> -<p>One simple difference is that NASM is case-sensitive. It makes a -difference whether you call your label <code><nobr>foo</nobr></code>, -<code><nobr>Foo</nobr></code> or <code><nobr>FOO</nobr></code>. If you're -assembling to <code><nobr>DOS</nobr></code> or -<code><nobr>OS/2</nobr></code> <code><nobr>.OBJ</nobr></code> files, you -can invoke the <code><nobr>UPPERCASE</nobr></code> directive (documented in -<a href="nasmdoc7.html#section-7.4">section 7.4</a>) to ensure that all -symbols exported to other code modules are forced to be upper case; but -even then, <em>within</em> a single module, NASM will distinguish between -labels differing only in case. -<h4><a name="section-2.2.2">2.2.2 NASM Requires Square Brackets For Memory References</a></h4> -<p>NASM was designed with simplicity of syntax in mind. One of the design -goals of NASM is that it should be possible, as far as is practical, for -the user to look at a single line of NASM code and tell what opcode is -generated by it. You can't do this in MASM: if you declare, for example, -<p><pre> -foo equ 1 -bar dw 2 -</pre> -<p>then the two lines of code -<p><pre> - mov ax,foo - mov ax,bar -</pre> -<p>generate completely different opcodes, despite having identical-looking -syntaxes. -<p>NASM avoids this undesirable situation by having a much simpler syntax -for memory references. The rule is simply that any access to the -<em>contents</em> of a memory location requires square brackets around the -address, and any access to the <em>address</em> of a variable doesn't. So -an instruction of the form <code><nobr>mov ax,foo</nobr></code> will -<em>always</em> refer to a compile-time constant, whether it's an -<code><nobr>EQU</nobr></code> or the address of a variable; and to access -the <em>contents</em> of the variable <code><nobr>bar</nobr></code>, you -must code <code><nobr>mov ax,[bar]</nobr></code>. -<p>This also means that NASM has no need for MASM's -<code><nobr>OFFSET</nobr></code> keyword, since the MASM code -<code><nobr>mov ax,offset bar</nobr></code> means exactly the same thing as -NASM's <code><nobr>mov ax,bar</nobr></code>. If you're trying to get large -amounts of MASM code to assemble sensibly under NASM, you can always code -<code><nobr>%idefine offset</nobr></code> to make the preprocessor treat -the <code><nobr>OFFSET</nobr></code> keyword as a no-op. -<p>This issue is even more confusing in <code><nobr>a86</nobr></code>, -where declaring a label with a trailing colon defines it to be a `label' as -opposed to a `variable' and causes <code><nobr>a86</nobr></code> to adopt -NASM-style semantics; so in <code><nobr>a86</nobr></code>, -<code><nobr>mov ax,var</nobr></code> has different behaviour depending on -whether <code><nobr>var</nobr></code> was declared as -<code><nobr>var: dw 0</nobr></code> (a label) or -<code><nobr>var dw 0</nobr></code> (a word-size variable). NASM is very -simple by comparison: <em>everything</em> is a label. -<p>NASM, in the interests of simplicity, also does not support the hybrid -syntaxes supported by MASM and its clones, such as -<code><nobr>mov ax,table[bx]</nobr></code>, where a memory reference is -denoted by one portion outside square brackets and another portion inside. -The correct syntax for the above is -<code><nobr>mov ax,[table+bx]</nobr></code>. Likewise, -<code><nobr>mov ax,es:[di]</nobr></code> is wrong and -<code><nobr>mov ax,[es:di]</nobr></code> is right. -<h4><a name="section-2.2.3">2.2.3 NASM Doesn't Store Variable Types</a></h4> -<p>NASM, by design, chooses not to remember the types of variables you -declare. Whereas MASM will remember, on seeing -<code><nobr>var dw 0</nobr></code>, that you declared -<code><nobr>var</nobr></code> as a word-size variable, and will then be -able to fill in the ambiguity in the size of the instruction -<code><nobr>mov var,2</nobr></code>, NASM will deliberately remember -nothing about the symbol <code><nobr>var</nobr></code> except where it -begins, and so you must explicitly code -<code><nobr>mov word [var],2</nobr></code>. -<p>For this reason, NASM doesn't support the -<code><nobr>LODS</nobr></code>, <code><nobr>MOVS</nobr></code>, -<code><nobr>STOS</nobr></code>, <code><nobr>SCAS</nobr></code>, -<code><nobr>CMPS</nobr></code>, <code><nobr>INS</nobr></code>, or -<code><nobr>OUTS</nobr></code> instructions, but only supports the forms -such as <code><nobr>LODSB</nobr></code>, <code><nobr>MOVSW</nobr></code>, -and <code><nobr>SCASD</nobr></code>, which explicitly specify the size of -the components of the strings being manipulated. -<h4><a name="section-2.2.4">2.2.4 NASM Doesn't <code><nobr>ASSUME</nobr></code></a></h4> -<p>As part of NASM's drive for simplicity, it also does not support the -<code><nobr>ASSUME</nobr></code> directive. NASM will not keep track of -what values you choose to put in your segment registers, and will never -<em>automatically</em> generate a segment override prefix. -<h4><a name="section-2.2.5">2.2.5 NASM Doesn't Support Memory Models</a></h4> -<p>NASM also does not have any directives to support different 16-bit -memory models. The programmer has to keep track of which functions are -supposed to be called with a far call and which with a near call, and is -responsible for putting the correct form of <code><nobr>RET</nobr></code> -instruction (<code><nobr>RETN</nobr></code> or -<code><nobr>RETF</nobr></code>; NASM accepts <code><nobr>RET</nobr></code> -itself as an alternate form for <code><nobr>RETN</nobr></code>); in -addition, the programmer is responsible for coding CALL FAR instructions -where necessary when calling <em>external</em> functions, and must also -keep track of which external variable definitions are far and which are -near. -<h4><a name="section-2.2.6">2.2.6 Floating-Point Differences</a></h4> -<p>NASM uses different names to refer to floating-point registers from -MASM: where MASM would call them <code><nobr>ST(0)</nobr></code>, -<code><nobr>ST(1)</nobr></code> and so on, and -<code><nobr>a86</nobr></code> would call them simply -<code><nobr>0</nobr></code>, <code><nobr>1</nobr></code> and so on, NASM -chooses to call them <code><nobr>st0</nobr></code>, -<code><nobr>st1</nobr></code> etc. -<p>As of version 0.96, NASM now treats the instructions with `nowait' forms -in the same way as MASM-compatible assemblers. The idiosyncratic treatment -employed by 0.95 and earlier was based on a misunderstanding by the -authors. -<h4><a name="section-2.2.7">2.2.7 Other Differences</a></h4> -<p>For historical reasons, NASM uses the keyword -<code><nobr>TWORD</nobr></code> where MASM and compatible assemblers use -<code><nobr>TBYTE</nobr></code>. -<p>NASM does not declare uninitialized storage in the same way as MASM: -where a MASM programmer might use -<code><nobr>stack db 64 dup (?)</nobr></code>, NASM requires -<code><nobr>stack resb 64</nobr></code>, intended to be read as `reserve 64 -bytes'. For a limited amount of compatibility, since NASM treats -<code><nobr>?</nobr></code> as a valid character in symbol names, you can -code <code><nobr>? equ 0</nobr></code> and then writing -<code><nobr>dw ?</nobr></code> will at least do something vaguely useful. -<code><nobr>DUP</nobr></code> is still not a supported syntax, however. -<p>In addition to all of this, macros and directives work completely -differently to MASM. See <a href="nasmdoc4.html">chapter 4</a> and -<a href="nasmdoc6.html">chapter 6</a> for further details. -<p align=center><a href="nasmdoc3.html">Next Chapter</a> | -<a href="nasmdoc1.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -</body></html> diff --git a/doc/html/nasmdoc3.html b/doc/html/nasmdoc3.html deleted file mode 100644 index b70e438..0000000 --- a/doc/html/nasmdoc3.html +++ /dev/null @@ -1,753 +0,0 @@ -<html><head><title>NASM Manual</title></head> -<body><h1 align=center>The Netwide Assembler: NASM</h1> - -<p align=center><a href="nasmdoc4.html">Next Chapter</a> | -<a href="nasmdoc2.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -<h2><a name="chapter-3">Chapter 3: The NASM Language</a></h2> -<h3><a name="section-3.1">3.1 Layout of a NASM Source Line</a></h3> -<p>Like most assemblers, each NASM source line contains (unless it is a -macro, a preprocessor directive or an assembler directive: see -<a href="nasmdoc4.html">chapter 4</a> and <a href="nasmdoc6.html">chapter -6</a>) some combination of the four fields -<p><pre> -label: instruction operands ; comment -</pre> -<p>As usual, most of these fields are optional; the presence or absence of -any combination of a label, an instruction and a comment is allowed. Of -course, the operand field is either required or forbidden by the presence -and nature of the instruction field. -<p>NASM uses backslash (\) as the line continuation character; if a line -ends with backslash, the next line is considered to be a part of the -backslash-ended line. -<p>NASM places no restrictions on white space within a line: labels may -have white space before them, or instructions may have no space before -them, or anything. The colon after a label is also optional. (Note that -this means that if you intend to code <code><nobr>lodsb</nobr></code> alone -on a line, and type <code><nobr>lodab</nobr></code> by accident, then -that's still a valid source line which does nothing but define a label. -Running NASM with the command-line option -<code><nobr>-w+orphan-labels</nobr></code> will cause it to warn you if you -define a label alone on a line without a trailing colon.) -<p>Valid characters in labels are letters, numbers, -<code><nobr>_</nobr></code>, <code><nobr>$</nobr></code>, -<code><nobr>#</nobr></code>, <code><nobr>@</nobr></code>, -<code><nobr>~</nobr></code>, <code><nobr>.</nobr></code>, and -<code><nobr>?</nobr></code>. The only characters which may be used as the -<em>first</em> character of an identifier are letters, -<code><nobr>.</nobr></code> (with special meaning: see -<a href="#section-3.9">section 3.9</a>), <code><nobr>_</nobr></code> and -<code><nobr>?</nobr></code>. An identifier may also be prefixed with a -<code><nobr>$</nobr></code> to indicate that it is intended to be read as -an identifier and not a reserved word; thus, if some other module you are -linking with defines a symbol called <code><nobr>eax</nobr></code>, you can -refer to <code><nobr>$eax</nobr></code> in NASM code to distinguish the -symbol from the register. Maximum length of an identifier is 4095 -characters. -<p>The instruction field may contain any machine instruction: Pentium and -P6 instructions, FPU instructions, MMX instructions and even undocumented -instructions are all supported. The instruction may be prefixed by -<code><nobr>LOCK</nobr></code>, <code><nobr>REP</nobr></code>, -<code><nobr>REPE</nobr></code>/<code><nobr>REPZ</nobr></code> or -<code><nobr>REPNE</nobr></code>/<code><nobr>REPNZ</nobr></code>, in the -usual way. Explicit address-size and operand-size prefixes -<code><nobr>A16</nobr></code>, <code><nobr>A32</nobr></code>, -<code><nobr>A64</nobr></code>, <code><nobr>O16</nobr></code> and -<code><nobr>O32</nobr></code>, <code><nobr>O64</nobr></code> are provided - -one example of their use is given in <a href="nasmdo10.html">chapter -10</a>. You can also use the name of a segment register as an instruction -prefix: coding <code><nobr>es mov [bx],ax</nobr></code> is equivalent to -coding <code><nobr>mov [es:bx],ax</nobr></code>. We recommend the latter -syntax, since it is consistent with other syntactic features of the -language, but for instructions such as <code><nobr>LODSB</nobr></code>, -which has no operands and yet can require a segment override, there is no -clean syntactic way to proceed apart from -<code><nobr>es lodsb</nobr></code>. -<p>An instruction is not required to use a prefix: prefixes such as -<code><nobr>CS</nobr></code>, <code><nobr>A32</nobr></code>, -<code><nobr>LOCK</nobr></code> or <code><nobr>REPE</nobr></code> can appear -on a line by themselves, and NASM will just generate the prefix bytes. -<p>In addition to actual machine instructions, NASM also supports a number -of pseudo-instructions, described in <a href="#section-3.2">section -3.2</a>. -<p>Instruction operands may take a number of forms: they can be registers, -described simply by the register name (e.g. <code><nobr>ax</nobr></code>, -<code><nobr>bp</nobr></code>, <code><nobr>ebx</nobr></code>, -<code><nobr>cr0</nobr></code>: NASM does not use the -<code><nobr>gas</nobr></code>-style syntax in which register names must be -prefixed by a <code><nobr>%</nobr></code> sign), or they can be effective -addresses (see <a href="#section-3.3">section 3.3</a>), constants -(<a href="#section-3.4">section 3.4</a>) or expressions -(<a href="#section-3.5">section 3.5</a>). -<p>For x87 floating-point instructions, NASM accepts a wide range of -syntaxes: you can use two-operand forms like MASM supports, or you can use -NASM's native single-operand forms in most cases. For example, you can -code: -<p><pre> - fadd st1 ; this sets st0 := st0 + st1 - fadd st0,st1 ; so does this - - fadd st1,st0 ; this sets st1 := st1 + st0 - fadd to st1 ; so does this -</pre> -<p>Almost any x87 floating-point instruction that references memory must -use one of the prefixes <code><nobr>DWORD</nobr></code>, -<code><nobr>QWORD</nobr></code> or <code><nobr>TWORD</nobr></code> to -indicate what size of memory operand it refers to. -<h3><a name="section-3.2">3.2 Pseudo-Instructions</a></h3> -<p>Pseudo-instructions are things which, though not real x86 machine -instructions, are used in the instruction field anyway because that's the -most convenient place to put them. The current pseudo-instructions are -<code><nobr>DB</nobr></code>, <code><nobr>DW</nobr></code>, -<code><nobr>DD</nobr></code>, <code><nobr>DQ</nobr></code>, -<code><nobr>DT</nobr></code>, <code><nobr>DO</nobr></code> and -<code><nobr>DY</nobr></code>; their uninitialized counterparts -<code><nobr>RESB</nobr></code>, <code><nobr>RESW</nobr></code>, -<code><nobr>RESD</nobr></code>, <code><nobr>RESQ</nobr></code>, -<code><nobr>REST</nobr></code>, <code><nobr>RESO</nobr></code> and -<code><nobr>RESY</nobr></code>; the <code><nobr>INCBIN</nobr></code> -command, the <code><nobr>EQU</nobr></code> command, and the -<code><nobr>TIMES</nobr></code> prefix. -<h4><a name="section-3.2.1">3.2.1 <code><nobr>DB</nobr></code> and Friends: Declaring Initialized Data</a></h4> -<p><code><nobr>DB</nobr></code>, <code><nobr>DW</nobr></code>, -<code><nobr>DD</nobr></code>, <code><nobr>DQ</nobr></code>, -<code><nobr>DT</nobr></code>, <code><nobr>DO</nobr></code> and -<code><nobr>DY</nobr></code> are used, much as in MASM, to declare -initialized data in the output file. They can be invoked in a wide range of -ways: -<p><pre> - db 0x55 ; just the byte 0x55 - db 0x55,0x56,0x57 ; three bytes in succession - db 'a',0x55 ; character constants are OK - db 'hello',13,10,'$' ; so are string constants - dw 0x1234 ; 0x34 0x12 - dw 'a' ; 0x61 0x00 (it's just a number) - dw 'ab' ; 0x61 0x62 (character constant) - dw 'abc' ; 0x61 0x62 0x63 0x00 (string) - dd 0x12345678 ; 0x78 0x56 0x34 0x12 - dd 1.234567e20 ; floating-point constant - dq 0x123456789abcdef0 ; eight byte constant - dq 1.234567e20 ; double-precision float - dt 1.234567e20 ; extended-precision float -</pre> -<p><code><nobr>DT</nobr></code>, <code><nobr>DO</nobr></code> and -<code><nobr>DY</nobr></code> do not accept numeric constants as operands. -<h4><a name="section-3.2.2">3.2.2 <code><nobr>RESB</nobr></code> and Friends: Declaring Uninitialized Data</a></h4> -<p><code><nobr>RESB</nobr></code>, <code><nobr>RESW</nobr></code>, -<code><nobr>RESD</nobr></code>, <code><nobr>RESQ</nobr></code>, -<code><nobr>REST</nobr></code>, <code><nobr>RESO</nobr></code> and -<code><nobr>RESY</nobr></code> are designed to be used in the BSS section -of a module: they declare <em>uninitialized</em> storage space. Each takes -a single operand, which is the number of bytes, words, doublewords or -whatever to reserve. As stated in -<a href="nasmdoc2.html#section-2.2.7">section 2.2.7</a>, NASM does not -support the MASM/TASM syntax of reserving uninitialized space by writing -<code><nobr>DW ?</nobr></code> or similar things: this is what it does -instead. The operand to a <code><nobr>RESB</nobr></code>-type -pseudo-instruction is a <em>critical expression</em>: see -<a href="#section-3.8">section 3.8</a>. -<p>For example: -<p><pre> -buffer: resb 64 ; reserve 64 bytes -wordvar: resw 1 ; reserve a word -realarray resq 10 ; array of ten reals -ymmval: resy 1 ; one YMM register -</pre> -<h4><a name="section-3.2.3">3.2.3 <code><nobr>INCBIN</nobr></code>: Including External Binary Files</a></h4> -<p><code><nobr>INCBIN</nobr></code> is borrowed from the old Amiga -assembler DevPac: it includes a binary file verbatim into the output file. -This can be handy for (for example) including graphics and sound data -directly into a game executable file. It can be called in one of these -three ways: -<p><pre> - incbin "file.dat" ; include the whole file - incbin "file.dat",1024 ; skip the first 1024 bytes - incbin "file.dat",1024,512 ; skip the first 1024, and - ; actually include at most 512 -</pre> -<p><code><nobr>INCBIN</nobr></code> is both a directive and a standard -macro; the standard macro version searches for the file in the include file -search path and adds the file to the dependency lists. This macro can be -overridden if desired. -<h4><a name="section-3.2.4">3.2.4 <code><nobr>EQU</nobr></code>: Defining Constants</a></h4> -<p><code><nobr>EQU</nobr></code> defines a symbol to a given constant -value: when <code><nobr>EQU</nobr></code> is used, the source line must -contain a label. The action of <code><nobr>EQU</nobr></code> is to define -the given label name to the value of its (only) operand. This definition is -absolute, and cannot change later. So, for example, -<p><pre> -message db 'hello, world' -msglen equ $-message -</pre> -<p>defines <code><nobr>msglen</nobr></code> to be the constant 12. -<code><nobr>msglen</nobr></code> may not then be redefined later. This is -not a preprocessor definition either: the value of -<code><nobr>msglen</nobr></code> is evaluated <em>once</em>, using the -value of <code><nobr>$</nobr></code> (see <a href="#section-3.5">section -3.5</a> for an explanation of <code><nobr>$</nobr></code>) at the point of -definition, rather than being evaluated wherever it is referenced and using -the value of <code><nobr>$</nobr></code> at the point of reference. -<h4><a name="section-3.2.5">3.2.5 <code><nobr>TIMES</nobr></code>: Repeating Instructions or Data</a></h4> -<p>The <code><nobr>TIMES</nobr></code> prefix causes the instruction to be -assembled multiple times. This is partly present as NASM's equivalent of -the <code><nobr>DUP</nobr></code> syntax supported by MASM-compatible -assemblers, in that you can code -<p><pre> -zerobuf: times 64 db 0 -</pre> -<p>or similar things; but <code><nobr>TIMES</nobr></code> is more versatile -than that. The argument to <code><nobr>TIMES</nobr></code> is not just a -numeric constant, but a numeric <em>expression</em>, so you can do things -like -<p><pre> -buffer: db 'hello, world' - times 64-$+buffer db ' ' -</pre> -<p>which will store exactly enough spaces to make the total length of -<code><nobr>buffer</nobr></code> up to 64. Finally, -<code><nobr>TIMES</nobr></code> can be applied to ordinary instructions, so -you can code trivial unrolled loops in it: -<p><pre> - times 100 movsb -</pre> -<p>Note that there is no effective difference between -<code><nobr>times 100 resb 1</nobr></code> and -<code><nobr>resb 100</nobr></code>, except that the latter will be -assembled about 100 times faster due to the internal structure of the -assembler. -<p>The operand to <code><nobr>TIMES</nobr></code> is a critical expression -(<a href="#section-3.8">section 3.8</a>). -<p>Note also that <code><nobr>TIMES</nobr></code> can't be applied to -macros: the reason for this is that <code><nobr>TIMES</nobr></code> is -processed after the macro phase, which allows the argument to -<code><nobr>TIMES</nobr></code> to contain expressions such as -<code><nobr>64-$+buffer</nobr></code> as above. To repeat more than one -line of code, or a complex macro, use the preprocessor -<code><nobr>%rep</nobr></code> directive. -<h3><a name="section-3.3">3.3 Effective Addresses</a></h3> -<p>An effective address is any operand to an instruction which references -memory. Effective addresses, in NASM, have a very simple syntax: they -consist of an expression evaluating to the desired address, enclosed in -square brackets. For example: -<p><pre> -wordvar dw 123 - mov ax,[wordvar] - mov ax,[wordvar+1] - mov ax,[es:wordvar+bx] -</pre> -<p>Anything not conforming to this simple system is not a valid memory -reference in NASM, for example <code><nobr>es:wordvar[bx]</nobr></code>. -<p>More complicated effective addresses, such as those involving more than -one register, work in exactly the same way: -<p><pre> - mov eax,[ebx*2+ecx+offset] - mov ax,[bp+di+8] -</pre> -<p>NASM is capable of doing algebra on these effective addresses, so that -things which don't necessarily <em>look</em> legal are perfectly all right: -<p><pre> - mov eax,[ebx*5] ; assembles as [ebx*4+ebx] - mov eax,[label1*2-label2] ; ie [label1+(label1-label2)] -</pre> -<p>Some forms of effective address have more than one assembled form; in -most such cases NASM will generate the smallest form it can. For example, -there are distinct assembled forms for the 32-bit effective addresses -<code><nobr>[eax*2+0]</nobr></code> and -<code><nobr>[eax+eax]</nobr></code>, and NASM will generally generate the -latter on the grounds that the former requires four bytes to store a zero -offset. -<p>NASM has a hinting mechanism which will cause -<code><nobr>[eax+ebx]</nobr></code> and <code><nobr>[ebx+eax]</nobr></code> -to generate different opcodes; this is occasionally useful because -<code><nobr>[esi+ebp]</nobr></code> and <code><nobr>[ebp+esi]</nobr></code> -have different default segment registers. -<p>However, you can force NASM to generate an effective address in a -particular form by the use of the keywords <code><nobr>BYTE</nobr></code>, -<code><nobr>WORD</nobr></code>, <code><nobr>DWORD</nobr></code> and -<code><nobr>NOSPLIT</nobr></code>. If you need -<code><nobr>[eax+3]</nobr></code> to be assembled using a double-word -offset field instead of the one byte NASM will normally generate, you can -code <code><nobr>[dword eax+3]</nobr></code>. Similarly, you can force NASM -to use a byte offset for a small value which it hasn't seen on the first -pass (see <a href="#section-3.8">section 3.8</a> for an example of such a -code fragment) by using <code><nobr>[byte eax+offset]</nobr></code>. As -special cases, <code><nobr>[byte eax]</nobr></code> will code -<code><nobr>[eax+0]</nobr></code> with a byte offset of zero, and -<code><nobr>[dword eax]</nobr></code> will code it with a double-word -offset of zero. The normal form, <code><nobr>[eax]</nobr></code>, will be -coded with no offset field. -<p>The form described in the previous paragraph is also useful if you are -trying to access data in a 32-bit segment from within 16 bit code. For more -information on this see the section on mixed-size addressing -(<a href="nasmdo10.html#section-10.2">section 10.2</a>). In particular, if -you need to access data with a known offset that is larger than will fit in -a 16-bit value, if you don't specify that it is a dword offset, nasm will -cause the high word of the offset to be lost. -<p>Similarly, NASM will split <code><nobr>[eax*2]</nobr></code> into -<code><nobr>[eax+eax]</nobr></code> because that allows the offset field to -be absent and space to be saved; in fact, it will also split -<code><nobr>[eax*2+offset]</nobr></code> into -<code><nobr>[eax+eax+offset]</nobr></code>. You can combat this behaviour -by the use of the <code><nobr>NOSPLIT</nobr></code> keyword: -<code><nobr>[nosplit eax*2]</nobr></code> will force -<code><nobr>[eax*2+0]</nobr></code> to be generated literally. -<p>In 64-bit mode, NASM will by default generate absolute addresses. The -<code><nobr>REL</nobr></code> keyword makes it produce -<code><nobr>RIP</nobr></code>-relative addresses. Since this is frequently -the normally desired behaviour, see the <code><nobr>DEFAULT</nobr></code> -directive (<a href="nasmdoc6.html#section-6.2">section 6.2</a>). The -keyword <code><nobr>ABS</nobr></code> overrides -<code><nobr>REL</nobr></code>. -<h3><a name="section-3.4">3.4 Constants</a></h3> -<p>NASM understands four different types of constant: numeric, character, -string and floating-point. -<h4><a name="section-3.4.1">3.4.1 Numeric Constants</a></h4> -<p>A numeric constant is simply a number. NASM allows you to specify -numbers in a variety of number bases, in a variety of ways: you can suffix -<code><nobr>H</nobr></code> or <code><nobr>X</nobr></code>, -<code><nobr>Q</nobr></code> or <code><nobr>O</nobr></code>, and -<code><nobr>B</nobr></code> for hexadecimal, octal and binary respectively, -or you can prefix <code><nobr>0x</nobr></code> for hexadecimal in the style -of C, or you can prefix <code><nobr>$</nobr></code> for hexadecimal in the -style of Borland Pascal. Note, though, that the <code><nobr>$</nobr></code> -prefix does double duty as a prefix on identifiers (see -<a href="#section-3.1">section 3.1</a>), so a hex number prefixed with a -<code><nobr>$</nobr></code> sign must have a digit after the -<code><nobr>$</nobr></code> rather than a letter. In addition, current -versions of NASM accept the prefix <code><nobr>0h</nobr></code> for -hexadecimal, <code><nobr>0o</nobr></code> or <code><nobr>0q</nobr></code> -for octal, and <code><nobr>0b</nobr></code> for binary. Please note that -unlike C, a <code><nobr>0</nobr></code> prefix by itself does <em>not</em> -imply an octal constant! -<p>Numeric constants can have underscores (<code><nobr>_</nobr></code>) -interspersed to break up long strings. -<p>Some examples (all producing exactly the same code): -<p><pre> - mov ax,200 ; decimal - mov ax,0200 ; still decimal - mov ax,0200d ; explicitly decimal - mov ax,0d200 ; also decimal - mov ax,0c8h ; hex - mov ax,$0c8 ; hex again: the 0 is required - mov ax,0xc8 ; hex yet again - mov ax,0hc8 ; still hex - mov ax,310q ; octal - mov ax,310o ; octal again - mov ax,0o310 ; octal yet again - mov ax,0q310 ; hex yet again - mov ax,11001000b ; binary - mov ax,1100_1000b ; same binary constant - mov ax,0b1100_1000 ; same binary constant yet again -</pre> -<h4><a name="section-3.4.2">3.4.2 Character Strings</a></h4> -<p>A character string consists of up to eight characters enclosed in either -single quotes (<code><nobr>'...'</nobr></code>), double quotes -(<code><nobr>"..."</nobr></code>) or backquotes -(<code><nobr>`...`</nobr></code>). Single or double quotes are equivalent -to NASM (except of course that surrounding the constant with single quotes -allows double quotes to appear within it and vice versa); the contents of -those are represented verbatim. Strings enclosed in backquotes support -C-style <code><nobr>\</nobr></code>-escapes for special characters. -<p>The following escape sequences are recognized by backquoted strings: -<p><pre> - \' single quote (') - \" double quote (") - \` backquote (`) - \\ backslash (\) - \? question mark (?) - \a BEL (ASCII 7) - \b BS (ASCII 8) - \t TAB (ASCII 9) - \n LF (ASCII 10) - \v VT (ASCII 11) - \f FF (ASCII 12) - \r CR (ASCII 13) - \e ESC (ASCII 27) - \377 Up to 3 octal digits - literal byte - \xFF Up to 2 hexadecimal digits - literal byte - \u1234 4 hexadecimal digits - Unicode character - \U12345678 8 hexadecimal digits - Unicode character -</pre> -<p>All other escape sequences are reserved. Note that -<code><nobr>\0</nobr></code>, meaning a <code><nobr>NUL</nobr></code> -character (ASCII 0), is a special case of the octal escape sequence. -<p>Unicode characters specified with <code><nobr>\u</nobr></code> or -<code><nobr>\U</nobr></code> are converted to UTF-8. For example, the -following lines are all equivalent: -<p><pre> - db `\u263a` ; UTF-8 smiley face - db `\xe2\x98\xba` ; UTF-8 smiley face - db 0E2h, 098h, 0BAh ; UTF-8 smiley face -</pre> -<h4><a name="section-3.4.3">3.4.3 Character Constants</a></h4> -<p>A character constant consists of a string up to eight bytes long, used -in an expression context. It is treated as if it was an integer. -<p>A character constant with more than one byte will be arranged with -little-endian order in mind: if you code -<p><pre> - mov eax,'abcd' -</pre> -<p>then the constant generated is not <code><nobr>0x61626364</nobr></code>, -but <code><nobr>0x64636261</nobr></code>, so that if you were then to store -the value into memory, it would read <code><nobr>abcd</nobr></code> rather -than <code><nobr>dcba</nobr></code>. This is also the sense of character -constants understood by the Pentium's <code><nobr>CPUID</nobr></code> -instruction. -<h4><a name="section-3.4.4">3.4.4 String Constants</a></h4> -<p>String constants are character strings used in the context of some -pseudo-instructions, namely the <code><nobr>DB</nobr></code> family and -<code><nobr>INCBIN</nobr></code> (where it represents a filename.) They are -also used in certain preprocessor directives. -<p>A string constant looks like a character constant, only longer. It is -treated as a concatenation of maximum-size character constants for the -conditions. So the following are equivalent: -<p><pre> - db 'hello' ; string constant - db 'h','e','l','l','o' ; equivalent character constants -</pre> -<p>And the following are also equivalent: -<p><pre> - dd 'ninechars' ; doubleword string constant - dd 'nine','char','s' ; becomes three doublewords - db 'ninechars',0,0,0 ; and really looks like this -</pre> -<p>Note that when used in a string-supporting context, quoted strings are -treated as a string constants even if they are short enough to be a -character constant, because otherwise <code><nobr>db 'ab'</nobr></code> -would have the same effect as <code><nobr>db 'a'</nobr></code>, which would -be silly. Similarly, three-character or four-character constants are -treated as strings when they are operands to <code><nobr>DW</nobr></code>, -and so forth. -<h4><a name="section-3.4.5">3.4.5 Unicode Strings</a></h4> -<p>The special operators <code><nobr>__utf16__</nobr></code> and -<code><nobr>__utf32__</nobr></code> allows definition of Unicode strings. -They take a string in UTF-8 format and converts it to (littleendian) UTF-16 -or UTF-32, respectively. -<p>For example: -<p><pre> -%define u(x) __utf16__(x) -%define w(x) __utf32__(x) - - dw u('C:\WINDOWS'), 0 ; Pathname in UTF-16 - dd w(`A + B = \u206a`), 0 ; String in UTF-32 -</pre> -<p><code><nobr>__utf16__</nobr></code> and -<code><nobr>__utf32__</nobr></code> can be applied either to strings passed -to the <code><nobr>DB</nobr></code> family instructions, or to character -constants in an expression context. -<h4><a name="section-3.4.6">3.4.6 Floating-Point Constants</a></h4> -<p>Floating-point constants are acceptable only as arguments to -<code><nobr>DB</nobr></code>, <code><nobr>DW</nobr></code>, -<code><nobr>DD</nobr></code>, <code><nobr>DQ</nobr></code>, -<code><nobr>DT</nobr></code>, and <code><nobr>DO</nobr></code>, or as -arguments to the special operators <code><nobr>__float8__</nobr></code>, -<code><nobr>__float16__</nobr></code>, -<code><nobr>__float32__</nobr></code>, -<code><nobr>__float64__</nobr></code>, -<code><nobr>__float80m__</nobr></code>, -<code><nobr>__float80e__</nobr></code>, -<code><nobr>__float128l__</nobr></code>, and -<code><nobr>__float128h__</nobr></code>. -<p>Floating-point constants are expressed in the traditional form: digits, -then a period, then optionally more digits, then optionally an -<code><nobr>E</nobr></code> followed by an exponent. The period is -mandatory, so that NASM can distinguish between -<code><nobr>dd 1</nobr></code>, which declares an integer constant, and -<code><nobr>dd 1.0</nobr></code> which declares a floating-point constant. -NASM also support C99-style hexadecimal floating-point: -<code><nobr>0x</nobr></code>, hexadecimal digits, period, optionally more -hexadeximal digits, then optionally a <code><nobr>P</nobr></code> followed -by a <em>binary</em> (not hexadecimal) exponent in decimal notation. -<p>Underscores to break up groups of digits are permitted in floating-point -constants as well. -<p>Some examples: -<p><pre> - db -0.2 ; "Quarter precision" - dw -0.5 ; IEEE 754r/SSE5 half precision - dd 1.2 ; an easy one - dd 1.222_222_222 ; underscores are permitted - dd 0x1p+2 ; 1.0x2^2 = 4.0 - dq 0x1p+32 ; 1.0x2^32 = 4 294 967 296.0 - dq 1.e10 ; 10 000 000 000.0 - dq 1.e+10 ; synonymous with 1.e10 - dq 1.e-10 ; 0.000 000 000 1 - dt 3.141592653589793238462 ; pi - do 1.e+4000 ; IEEE 754r quad precision -</pre> -<p>The 8-bit "quarter-precision" floating-point format is -sign:exponent:mantissa = 1:4:3 with an exponent bias of 7. This appears to -be the most frequently used 8-bit floating-point format, although it is not -covered by any formal standard. This is sometimes called a "minifloat." -<p>The special operators are used to produce floating-point numbers in -other contexts. They produce the binary representation of a specific -floating-point number as an integer, and can use anywhere integer constants -are used in an expression. <code><nobr>__float80m__</nobr></code> and -<code><nobr>__float80e__</nobr></code> produce the 64-bit mantissa and -16-bit exponent of an 80-bit floating-point number, and -<code><nobr>__float128l__</nobr></code> and -<code><nobr>__float128h__</nobr></code> produce the lower and upper 64-bit -halves of a 128-bit floating-point number, respectively. -<p>For example: -<p><pre> - mov rax,__float64__(3.141592653589793238462) -</pre> -<p>... would assign the binary representation of pi as a 64-bit floating -point number into <code><nobr>RAX</nobr></code>. This is exactly equivalent -to: -<p><pre> - mov rax,0x400921fb54442d18 -</pre> -<p>NASM cannot do compile-time arithmetic on floating-point constants. This -is because NASM is designed to be portable - although it always generates -code to run on x86 processors, the assembler itself can run on any system -with an ANSI C compiler. Therefore, the assembler cannot guarantee the -presence of a floating-point unit capable of handling the Intel number -formats, and so for NASM to be able to do floating arithmetic it would have -to include its own complete set of floating-point routines, which would -significantly increase the size of the assembler for very little benefit. -<p>The special tokens <code><nobr>__Infinity__</nobr></code>, -<code><nobr>__QNaN__</nobr></code> (or <code><nobr>__NaN__</nobr></code>) -and <code><nobr>__SNaN__</nobr></code> can be used to generate infinities, -quiet NaNs, and signalling NaNs, respectively. These are normally used as -macros: -<p><pre> -%define Inf __Infinity__ -%define NaN __QNaN__ - - dq +1.5, -Inf, NaN ; Double-precision constants -</pre> -<h4><a name="section-3.4.7">3.4.7 Packed BCD Constants</a></h4> -<p>x87-style packed BCD constants can be used in the same contexts as -80-bit floating-point numbers. They are suffixed with -<code><nobr>p</nobr></code> or prefixed with <code><nobr>0p</nobr></code>, -and can include up to 18 decimal digits. -<p>As with other numeric constants, underscores can be used to separate -digits. -<p>For example: -<p><pre> - dt 12_345_678_901_245_678p - dt -12_345_678_901_245_678p - dt +0p33 - dt 33p -</pre> -<h3><a name="section-3.5">3.5 Expressions</a></h3> -<p>Expressions in NASM are similar in syntax to those in C. Expressions are -evaluated as 64-bit integers which are then adjusted to the appropriate -size. -<p>NASM supports two special tokens in expressions, allowing calculations -to involve the current assembly position: the <code><nobr>$</nobr></code> -and <code><nobr>$$</nobr></code> tokens. <code><nobr>$</nobr></code> -evaluates to the assembly position at the beginning of the line containing -the expression; so you can code an infinite loop using -<code><nobr>JMP $</nobr></code>. <code><nobr>$$</nobr></code> evaluates to -the beginning of the current section; so you can tell how far into the -section you are by using <code><nobr>($-$$)</nobr></code>. -<p>The arithmetic operators provided by NASM are listed here, in increasing -order of precedence. -<h4><a name="section-3.5.1">3.5.1 <code><nobr>|</nobr></code>: Bitwise OR Operator</a></h4> -<p>The <code><nobr>|</nobr></code> operator gives a bitwise OR, exactly as -performed by the <code><nobr>OR</nobr></code> machine instruction. Bitwise -OR is the lowest-priority arithmetic operator supported by NASM. -<h4><a name="section-3.5.2">3.5.2 <code><nobr>^</nobr></code>: Bitwise XOR Operator</a></h4> -<p><code><nobr>^</nobr></code> provides the bitwise XOR operation. -<h4><a name="section-3.5.3">3.5.3 <code><nobr>&</nobr></code>: Bitwise AND Operator</a></h4> -<p><code><nobr>&</nobr></code> provides the bitwise AND operation. -<h4><a name="section-3.5.4">3.5.4 <code><nobr><<</nobr></code> and <code><nobr>>></nobr></code>: Bit Shift Operators</a></h4> -<p><code><nobr><<</nobr></code> gives a bit-shift to the left, just -as it does in C. So <code><nobr>5<<3</nobr></code> evaluates to 5 -times 8, or 40. <code><nobr>>></nobr></code> gives a bit-shift to the -right; in NASM, such a shift is <em>always</em> unsigned, so that the bits -shifted in from the left-hand end are filled with zero rather than a -sign-extension of the previous highest bit. -<h4><a name="section-3.5.5">3.5.5 <code><nobr>+</nobr></code> and <code><nobr>-</nobr></code>: Addition and Subtraction Operators</a></h4> -<p>The <code><nobr>+</nobr></code> and <code><nobr>-</nobr></code> -operators do perfectly ordinary addition and subtraction. -<h4><a name="section-3.5.6">3.5.6 <code><nobr>*</nobr></code>, <code><nobr>/</nobr></code>, <code><nobr>//</nobr></code>, <code><nobr>%</nobr></code> and <code><nobr>%%</nobr></code>: Multiplication and Division</a></h4> -<p><code><nobr>*</nobr></code> is the multiplication operator. -<code><nobr>/</nobr></code> and <code><nobr>//</nobr></code> are both -division operators: <code><nobr>/</nobr></code> is unsigned division and -<code><nobr>//</nobr></code> is signed division. Similarly, -<code><nobr>%</nobr></code> and <code><nobr>%%</nobr></code> provide -unsigned and signed modulo operators respectively. -<p>NASM, like ANSI C, provides no guarantees about the sensible operation -of the signed modulo operator. -<p>Since the <code><nobr>%</nobr></code> character is used extensively by -the macro preprocessor, you should ensure that both the signed and unsigned -modulo operators are followed by white space wherever they appear. -<h4><a name="section-3.5.7">3.5.7 Unary Operators: <code><nobr>+</nobr></code>, <code><nobr>-</nobr></code>, <code><nobr>~</nobr></code>, <code><nobr>!</nobr></code> and <code><nobr>SEG</nobr></code></a></h4> -<p>The highest-priority operators in NASM's expression grammar are those -which only apply to one argument. <code><nobr>-</nobr></code> negates its -operand, <code><nobr>+</nobr></code> does nothing (it's provided for -symmetry with <code><nobr>-</nobr></code>), <code><nobr>~</nobr></code> -computes the one's complement of its operand, <code><nobr>!</nobr></code> -is the logical negation operator, and <code><nobr>SEG</nobr></code> -provides the segment address of its operand (explained in more detail in -<a href="#section-3.6">section 3.6</a>). -<h3><a name="section-3.6">3.6 <code><nobr>SEG</nobr></code> and <code><nobr>WRT</nobr></code></a></h3> -<p>When writing large 16-bit programs, which must be split into multiple -segments, it is often necessary to be able to refer to the segment part of -the address of a symbol. NASM supports the <code><nobr>SEG</nobr></code> -operator to perform this function. -<p>The <code><nobr>SEG</nobr></code> operator returns the -<em>preferred</em> segment base of a symbol, defined as the segment base -relative to which the offset of the symbol makes sense. So the code -<p><pre> - mov ax,seg symbol - mov es,ax - mov bx,symbol -</pre> -<p>will load <code><nobr>ES:BX</nobr></code> with a valid pointer to the -symbol <code><nobr>symbol</nobr></code>. -<p>Things can be more complex than this: since 16-bit segments and groups -may overlap, you might occasionally want to refer to some symbol using a -different segment base from the preferred one. NASM lets you do this, by -the use of the <code><nobr>WRT</nobr></code> (With Reference To) keyword. -So you can do things like -<p><pre> - mov ax,weird_seg ; weird_seg is a segment base - mov es,ax - mov bx,symbol wrt weird_seg -</pre> -<p>to load <code><nobr>ES:BX</nobr></code> with a different, but -functionally equivalent, pointer to the symbol -<code><nobr>symbol</nobr></code>. -<p>NASM supports far (inter-segment) calls and jumps by means of the syntax -<code><nobr>call segment:offset</nobr></code>, where -<code><nobr>segment</nobr></code> and <code><nobr>offset</nobr></code> both -represent immediate values. So to call a far procedure, you could code -either of -<p><pre> - call (seg procedure):procedure - call weird_seg:(procedure wrt weird_seg) -</pre> -<p>(The parentheses are included for clarity, to show the intended parsing -of the above instructions. They are not necessary in practice.) -<p>NASM supports the syntax <code><nobr>call far procedure</nobr></code> as -a synonym for the first of the above usages. <code><nobr>JMP</nobr></code> -works identically to <code><nobr>CALL</nobr></code> in these examples. -<p>To declare a far pointer to a data item in a data segment, you must code -<p><pre> - dw symbol, seg symbol -</pre> -<p>NASM supports no convenient synonym for this, though you can always -invent one using the macro processor. -<h3><a name="section-3.7">3.7 <code><nobr>STRICT</nobr></code>: Inhibiting Optimization</a></h3> -<p>When assembling with the optimizer set to level 2 or higher (see -<a href="nasmdoc2.html#section-2.1.22">section 2.1.22</a>), NASM will use -size specifiers (<code><nobr>BYTE</nobr></code>, -<code><nobr>WORD</nobr></code>, <code><nobr>DWORD</nobr></code>, -<code><nobr>QWORD</nobr></code>, <code><nobr>TWORD</nobr></code>, -<code><nobr>OWORD</nobr></code> or <code><nobr>YWORD</nobr></code>), but -will give them the smallest possible size. The keyword -<code><nobr>STRICT</nobr></code> can be used to inhibit optimization and -force a particular operand to be emitted in the specified size. For -example, with the optimizer on, and in <code><nobr>BITS 16</nobr></code> -mode, -<p><pre> - push dword 33 -</pre> -<p>is encoded in three bytes <code><nobr>66 6A 21</nobr></code>, whereas -<p><pre> - push strict dword 33 -</pre> -<p>is encoded in six bytes, with a full dword immediate operand -<code><nobr>66 68 21 00 00 00</nobr></code>. -<p>With the optimizer off, the same code (six bytes) is generated whether -the <code><nobr>STRICT</nobr></code> keyword was used or not. -<h3><a name="section-3.8">3.8 Critical Expressions</a></h3> -<p>Although NASM has an optional multi-pass optimizer, there are some -expressions which must be resolvable on the first pass. These are called -<em>Critical Expressions</em>. -<p>The first pass is used to determine the size of all the assembled code -and data, so that the second pass, when generating all the code, knows all -the symbol addresses the code refers to. So one thing NASM can't handle is -code whose size depends on the value of a symbol declared after the code in -question. For example, -<p><pre> - times (label-$) db 0 -label: db 'Where am I?' -</pre> -<p>The argument to <code><nobr>TIMES</nobr></code> in this case could -equally legally evaluate to anything at all; NASM will reject this example -because it cannot tell the size of the <code><nobr>TIMES</nobr></code> line -when it first sees it. It will just as firmly reject the slightly -paradoxical code -<p><pre> - times (label-$+1) db 0 -label: db 'NOW where am I?' -</pre> -<p>in which <em>any</em> value for the <code><nobr>TIMES</nobr></code> -argument is by definition wrong! -<p>NASM rejects these examples by means of a concept called a <em>critical -expression</em>, which is defined to be an expression whose value is -required to be computable in the first pass, and which must therefore -depend only on symbols defined before it. The argument to the -<code><nobr>TIMES</nobr></code> prefix is a critical expression. -<h3><a name="section-3.9">3.9 Local Labels</a></h3> -<p>NASM gives special treatment to symbols beginning with a period. A label -beginning with a single period is treated as a <em>local</em> label, which -means that it is associated with the previous non-local label. So, for -example: -<p><pre> -label1 ; some code - -.loop - ; some more code - - jne .loop - ret - -label2 ; some code - -.loop - ; some more code - - jne .loop - ret -</pre> -<p>In the above code fragment, each <code><nobr>JNE</nobr></code> -instruction jumps to the line immediately before it, because the two -definitions of <code><nobr>.loop</nobr></code> are kept separate by virtue -of each being associated with the previous non-local label. -<p>This form of local label handling is borrowed from the old Amiga -assembler DevPac; however, NASM goes one step further, in allowing access -to local labels from other parts of the code. This is achieved by means of -<em>defining</em> a local label in terms of the previous non-local label: -the first definition of <code><nobr>.loop</nobr></code> above is really -defining a symbol called <code><nobr>label1.loop</nobr></code>, and the -second defines a symbol called <code><nobr>label2.loop</nobr></code>. So, -if you really needed to, you could write -<p><pre> -label3 ; some more code - ; and some more - - jmp label1.loop -</pre> -<p>Sometimes it is useful - in a macro, for instance - to be able to define -a label which can be referenced from anywhere but which doesn't interfere -with the normal local-label mechanism. Such a label can't be non-local -because it would interfere with subsequent definitions of, and references -to, local labels; and it can't be local because the macro that defined it -wouldn't know the label's full name. NASM therefore introduces a third type -of label, which is probably only useful in macro definitions: if a label -begins with the special prefix <code><nobr>..@</nobr></code>, then it does -nothing to the local label mechanism. So you could code -<p><pre> -label1: ; a non-local label -.local: ; this is really label1.local -..@foo: ; this is a special symbol -label2: ; another non-local label -.local: ; this is really label2.local - - jmp ..@foo ; this will jump three lines up -</pre> -<p>NASM has the capacity to define other special symbols beginning with a -double period: for example, <code><nobr>..start</nobr></code> is used to -specify the entry point in the <code><nobr>obj</nobr></code> output format -(see <a href="nasmdoc7.html#section-7.4.6">section 7.4.6</a>). -<p align=center><a href="nasmdoc4.html">Next Chapter</a> | -<a href="nasmdoc2.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -</body></html> diff --git a/doc/html/nasmdoc4.html b/doc/html/nasmdoc4.html deleted file mode 100644 index 8c71881..0000000 --- a/doc/html/nasmdoc4.html +++ /dev/null @@ -1,1970 +0,0 @@ -<html><head><title>NASM Manual</title></head> -<body><h1 align=center>The Netwide Assembler: NASM</h1> - -<p align=center><a href="nasmdoc5.html">Next Chapter</a> | -<a href="nasmdoc3.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -<h2><a name="chapter-4">Chapter 4: The NASM Preprocessor</a></h2> -<p>NASM contains a powerful macro processor, which supports conditional -assembly, multi-level file inclusion, two forms of macro (single-line and -multi-line), and a `context stack' mechanism for extra macro power. -Preprocessor directives all begin with a <code><nobr>%</nobr></code> sign. -<p>The preprocessor collapses all lines which end with a backslash (\) -character into a single line. Thus: -<p><pre> -%define THIS_VERY_LONG_MACRO_NAME_IS_DEFINED_TO \ - THIS_VALUE -</pre> -<p>will work like a single-line macro without the backslash-newline -sequence. -<h3><a name="section-4.1">4.1 Single-Line Macros</a></h3> -<h4><a name="section-4.1.1">4.1.1 The Normal Way: <code><nobr>%define</nobr></code></a></h4> -<p>Single-line macros are defined using the -<code><nobr>%define</nobr></code> preprocessor directive. The definitions -work in a similar way to C; so you can do things like -<p><pre> -%define ctrl 0x1F & -%define param(a,b) ((a)+(a)*(b)) - - mov byte [param(2,ebx)], ctrl 'D' -</pre> -<p>which will expand to -<p><pre> - mov byte [(2)+(2)*(ebx)], 0x1F & 'D' -</pre> -<p>When the expansion of a single-line macro contains tokens which invoke -another macro, the expansion is performed at invocation time, not at -definition time. Thus the code -<p><pre> -%define a(x) 1+b(x) -%define b(x) 2*x - - mov ax,a(8) -</pre> -<p>will evaluate in the expected way to -<code><nobr>mov ax,1+2*8</nobr></code>, even though the macro -<code><nobr>b</nobr></code> wasn't defined at the time of definition of -<code><nobr>a</nobr></code>. -<p>Macros defined with <code><nobr>%define</nobr></code> are case -sensitive: after <code><nobr>%define foo bar</nobr></code>, only -<code><nobr>foo</nobr></code> will expand to <code><nobr>bar</nobr></code>: -<code><nobr>Foo</nobr></code> or <code><nobr>FOO</nobr></code> will not. By -using <code><nobr>%idefine</nobr></code> instead of -<code><nobr>%define</nobr></code> (the `i' stands for `insensitive') you -can define all the case variants of a macro at once, so that -<code><nobr>%idefine foo bar</nobr></code> would cause -<code><nobr>foo</nobr></code>, <code><nobr>Foo</nobr></code>, -<code><nobr>FOO</nobr></code>, <code><nobr>fOO</nobr></code> and so on all -to expand to <code><nobr>bar</nobr></code>. -<p>There is a mechanism which detects when a macro call has occurred as a -result of a previous expansion of the same macro, to guard against circular -references and infinite loops. If this happens, the preprocessor will only -expand the first occurrence of the macro. Hence, if you code -<p><pre> -%define a(x) 1+a(x) - - mov ax,a(3) -</pre> -<p>the macro <code><nobr>a(3)</nobr></code> will expand once, becoming -<code><nobr>1+a(3)</nobr></code>, and will then expand no further. This -behaviour can be useful: see <a href="nasmdoc9.html#section-9.1">section -9.1</a> for an example of its use. -<p>You can overload single-line macros: if you write -<p><pre> -%define foo(x) 1+x -%define foo(x,y) 1+x*y -</pre> -<p>the preprocessor will be able to handle both types of macro call, by -counting the parameters you pass; so <code><nobr>foo(3)</nobr></code> will -become <code><nobr>1+3</nobr></code> whereas -<code><nobr>foo(ebx,2)</nobr></code> will become -<code><nobr>1+ebx*2</nobr></code>. However, if you define -<p><pre> -%define foo bar -</pre> -<p>then no other definition of <code><nobr>foo</nobr></code> will be -accepted: a macro with no parameters prohibits the definition of the same -name as a macro <em>with</em> parameters, and vice versa. -<p>This doesn't prevent single-line macros being <em>redefined</em>: you -can perfectly well define a macro with -<p><pre> -%define foo bar -</pre> -<p>and then re-define it later in the same source file with -<p><pre> -%define foo baz -</pre> -<p>Then everywhere the macro <code><nobr>foo</nobr></code> is invoked, it -will be expanded according to the most recent definition. This is -particularly useful when defining single-line macros with -<code><nobr>%assign</nobr></code> (see <a href="#section-4.1.7">section -4.1.7</a>). -<p>You can pre-define single-line macros using the `-d' option on the NASM -command line: see <a href="nasmdoc2.html#section-2.1.18">section -2.1.18</a>. -<h4><a name="section-4.1.2">4.1.2 Resolving <code><nobr>%define</nobr></code>: <code><nobr>%xdefine</nobr></code></a></h4> -<p>To have a reference to an embedded single-line macro resolved at the -time that the embedding macro is <em>defined</em>, as opposed to when the -embedding macro is <em>expanded</em>, you need a different mechanism to the -one offered by <code><nobr>%define</nobr></code>. The solution is to use -<code><nobr>%xdefine</nobr></code>, or it's case-insensitive counterpart -<code><nobr>%ixdefine</nobr></code>. -<p>Suppose you have the following code: -<p><pre> -%define isTrue 1 -%define isFalse isTrue -%define isTrue 0 - -val1: db isFalse - -%define isTrue 1 - -val2: db isFalse -</pre> -<p>In this case, <code><nobr>val1</nobr></code> is equal to 0, and -<code><nobr>val2</nobr></code> is equal to 1. This is because, when a -single-line macro is defined using <code><nobr>%define</nobr></code>, it is -expanded only when it is called. As <code><nobr>isFalse</nobr></code> -expands to <code><nobr>isTrue</nobr></code>, the expansion will be the -current value of <code><nobr>isTrue</nobr></code>. The first time it is -called that is 0, and the second time it is 1. -<p>If you wanted <code><nobr>isFalse</nobr></code> to expand to the value -assigned to the embedded macro <code><nobr>isTrue</nobr></code> at the time -that <code><nobr>isFalse</nobr></code> was defined, you need to change the -above code to use <code><nobr>%xdefine</nobr></code>. -<p><pre> -%xdefine isTrue 1 -%xdefine isFalse isTrue -%xdefine isTrue 0 - -val1: db isFalse - -%xdefine isTrue 1 - -val2: db isFalse -</pre> -<p>Now, each time that <code><nobr>isFalse</nobr></code> is called, it -expands to 1, as that is what the embedded macro -<code><nobr>isTrue</nobr></code> expanded to at the time that -<code><nobr>isFalse</nobr></code> was defined. -<h4><a name="section-4.1.3">4.1.3 Macro Indirection: <code><nobr>%[...]</nobr></code></a></h4> -<p>The <code><nobr>%[...]</nobr></code> construct can be used to expand -macros in contexts where macro expansion would otherwise not occur, -including in the names other macros. For example, if you have a set of -macros named <code><nobr>Foo16</nobr></code>, -<code><nobr>Foo32</nobr></code> and <code><nobr>Foo64</nobr></code>, you -could write: -<p><pre> - mov ax,Foo%[__BITS__] ; The Foo value -</pre> -<p>to use the builtin macro <code><nobr>__BITS__</nobr></code> (see -<a href="#section-4.11.5">section 4.11.5</a>) to automatically select -between them. Similarly, the two statements: -<p><pre> -%xdefine Bar Quux ; Expands due to %xdefine -%define Bar %[Quux] ; Expands due to %[...] -</pre> -<p>have, in fact, exactly the same effect. -<p><code><nobr>%[...]</nobr></code> concatenates to adjacent tokens in the -same way that multi-line macro parameters do, see -<a href="#section-4.3.8">section 4.3.8</a> for details. -<h4><a name="section-4.1.4">4.1.4 Concatenating Single Line Macro Tokens: <code><nobr>%+</nobr></code></a></h4> -<p>Individual tokens in single line macros can be concatenated, to produce -longer tokens for later processing. This can be useful if there are several -similar macros that perform similar functions. -<p>Please note that a space is required after <code><nobr>%+</nobr></code>, -in order to disambiguate it from the syntax <code><nobr>%+1</nobr></code> -used in multiline macros. -<p>As an example, consider the following: -<p><pre> -%define BDASTART 400h ; Start of BIOS data area -</pre> -<p><pre> -struc tBIOSDA ; its structure - .COM1addr RESW 1 - .COM2addr RESW 1 - ; ..and so on -endstruc -</pre> -<p>Now, if we need to access the elements of tBIOSDA in different places, -we can end up with: -<p><pre> - mov ax,BDASTART + tBIOSDA.COM1addr - mov bx,BDASTART + tBIOSDA.COM2addr -</pre> -<p>This will become pretty ugly (and tedious) if used in many places, and -can be reduced in size significantly by using the following macro: -<p><pre> -; Macro to access BIOS variables by their names (from tBDA): -</pre> -<p><pre> -%define BDA(x) BDASTART + tBIOSDA. %+ x -</pre> -<p>Now the above code can be written as: -<p><pre> - mov ax,BDA(COM1addr) - mov bx,BDA(COM2addr) -</pre> -<p>Using this feature, we can simplify references to a lot of macros (and, -in turn, reduce typing errors). -<h4><a name="section-4.1.5">4.1.5 The Macro Name Itself: <code><nobr>%?</nobr></code> and <code><nobr>%??</nobr></code></a></h4> -<p>The special symbols <code><nobr>%?</nobr></code> and -<code><nobr>%??</nobr></code> can be used to reference the macro name -itself inside a macro expansion, this is supported for both single-and -multi-line macros. <code><nobr>%?</nobr></code> refers to the macro name as -<em>invoked</em>, whereas <code><nobr>%??</nobr></code> refers to the macro -name as <em>declared</em>. The two are always the same for case-sensitive -macros, but for case-insensitive macros, they can differ. -<p>For example: -<p><pre> -%idefine Foo mov %?,%?? - - foo - FOO -</pre> -<p>will expand to: -<p><pre> - mov foo,Foo - mov FOO,Foo -</pre> -<p>The sequence: -<p><pre> -%idefine keyword $%? -</pre> -<p>can be used to make a keyword "disappear", for example in case a new -instruction has been used as a label in older code. For example: -<p><pre> -%idefine pause $%? ; Hide the PAUSE instruction -</pre> -<h4><a name="section-4.1.6">4.1.6 Undefining Single-Line Macros: <code><nobr>%undef</nobr></code></a></h4> -<p>Single-line macros can be removed with the -<code><nobr>%undef</nobr></code> directive. For example, the following -sequence: -<p><pre> -%define foo bar -%undef foo - - mov eax, foo -</pre> -<p>will expand to the instruction <code><nobr>mov eax, foo</nobr></code>, -since after <code><nobr>%undef</nobr></code> the macro -<code><nobr>foo</nobr></code> is no longer defined. -<p>Macros that would otherwise be pre-defined can be undefined on the -command-line using the `-u' option on the NASM command line: see -<a href="nasmdoc2.html#section-2.1.19">section 2.1.19</a>. -<h4><a name="section-4.1.7">4.1.7 Preprocessor Variables: <code><nobr>%assign</nobr></code></a></h4> -<p>An alternative way to define single-line macros is by means of the -<code><nobr>%assign</nobr></code> command (and its case-insensitive -counterpart <code><nobr>%iassign</nobr></code>, which differs from -<code><nobr>%assign</nobr></code> in exactly the same way that -<code><nobr>%idefine</nobr></code> differs from -<code><nobr>%define</nobr></code>). -<p><code><nobr>%assign</nobr></code> is used to define single-line macros -which take no parameters and have a numeric value. This value can be -specified in the form of an expression, and it will be evaluated once, when -the <code><nobr>%assign</nobr></code> directive is processed. -<p>Like <code><nobr>%define</nobr></code>, macros defined using -<code><nobr>%assign</nobr></code> can be re-defined later, so you can do -things like -<p><pre> -%assign i i+1 -</pre> -<p>to increment the numeric value of a macro. -<p><code><nobr>%assign</nobr></code> is useful for controlling the -termination of <code><nobr>%rep</nobr></code> preprocessor loops: see -<a href="#section-4.5">section 4.5</a> for an example of this. Another use -for <code><nobr>%assign</nobr></code> is given in -<a href="nasmdoc8.html#section-8.4">section 8.4</a> and -<a href="nasmdoc9.html#section-9.1">section 9.1</a>. -<p>The expression passed to <code><nobr>%assign</nobr></code> is a critical -expression (see <a href="nasmdoc3.html#section-3.8">section 3.8</a>), and -must also evaluate to a pure number (rather than a relocatable reference -such as a code or data address, or anything involving a register). -<h4><a name="section-4.1.8">4.1.8 Defining Strings: <code><nobr>%defstr</nobr></code></a></h4> -<p><code><nobr>%defstr</nobr></code>, and its case-insensitive counterpart -<code><nobr>%idefstr</nobr></code>, define or redefine a single-line macro -without parameters but converts the entire right-hand side, after macro -expansion, to a quoted string before definition. -<p>For example: -<p><pre> -%defstr test TEST -</pre> -<p>is equivalent to -<p><pre> -%define test 'TEST' -</pre> -<p>This can be used, for example, with the <code><nobr>%!</nobr></code> -construct (see <a href="#section-4.10.2">section 4.10.2</a>): -<p><pre> -%defstr PATH %!PATH ; The operating system PATH variable -</pre> -<h4><a name="section-4.1.9">4.1.9 Defining Tokens: <code><nobr>%deftok</nobr></code></a></h4> -<p><code><nobr>%deftok</nobr></code>, and its case-insensitive counterpart -<code><nobr>%ideftok</nobr></code>, define or redefine a single-line macro -without parameters but converts the second parameter, after string -conversion, to a sequence of tokens. -<p>For example: -<p><pre> -%deftok test 'TEST' -</pre> -<p>is equivalent to -<p><pre> -%define test TEST -</pre> -<h3><a name="section-4.2">4.2 String Manipulation in Macros</a></h3> -<p>It's often useful to be able to handle strings in macros. NASM supports -a few simple string handling macro operators from which more complex -operations can be constructed. -<p>All the string operators define or redefine a value (either a string or -a numeric value) to a single-line macro. When producing a string value, it -may change the style of quoting of the input string or strings, and -possibly use <code><nobr>\</nobr></code>-escapes inside -<code><nobr>`</nobr></code>-quoted strings. -<h4><a name="section-4.2.1">4.2.1 Concatenating Strings: <code><nobr>%strcat</nobr></code></a></h4> -<p>The <code><nobr>%strcat</nobr></code> operator concatenates quoted -strings and assign them to a single-line macro. -<p>For example: -<p><pre> -%strcat alpha "Alpha: ", '12" screen' -</pre> -<p>... would assign the value <code><nobr>'Alpha: 12" screen'</nobr></code> -to <code><nobr>alpha</nobr></code>. Similarly: -<p><pre> -%strcat beta '"foo"\', "'bar'" -</pre> -<p>... would assign the value <code><nobr>`"foo"\\'bar'`</nobr></code> to -<code><nobr>beta</nobr></code>. -<p>The use of commas to separate strings is permitted but optional. -<h4><a name="section-4.2.2">4.2.2 String Length: <code><nobr>%strlen</nobr></code></a></h4> -<p>The <code><nobr>%strlen</nobr></code> operator assigns the length of a -string to a macro. For example: -<p><pre> -%strlen charcnt 'my string' -</pre> -<p>In this example, <code><nobr>charcnt</nobr></code> would receive the -value 9, just as if an <code><nobr>%assign</nobr></code> had been used. In -this example, <code><nobr>'my string'</nobr></code> was a literal string -but it could also have been a single-line macro that expands to a string, -as in the following example: -<p><pre> -%define sometext 'my string' -%strlen charcnt sometext -</pre> -<p>As in the first case, this would result in -<code><nobr>charcnt</nobr></code> being assigned the value of 9. -<h4><a name="section-4.2.3">4.2.3 Extracting Substrings: <code><nobr>%substr</nobr></code></a></h4> -<p>Individual letters or substrings in strings can be extracted using the -<code><nobr>%substr</nobr></code> operator. An example of its use is -probably more useful than the description: -<p><pre> -%substr mychar 'xyzw' 1 ; equivalent to %define mychar 'x' -%substr mychar 'xyzw' 2 ; equivalent to %define mychar 'y' -%substr mychar 'xyzw' 3 ; equivalent to %define mychar 'z' -%substr mychar 'xyzw' 2,2 ; equivalent to %define mychar 'yz' -%substr mychar 'xyzw' 2,-1 ; equivalent to %define mychar 'yzw' -%substr mychar 'xyzw' 2,-2 ; equivalent to %define mychar 'yz' -</pre> -<p>As with <code><nobr>%strlen</nobr></code> (see -<a href="#section-4.2.2">section 4.2.2</a>), the first parameter is the -single-line macro to be created and the second is the string. The third -parameter specifies the first character to be selected, and the optional -fourth parameter preceeded by comma) is the length. Note that the first -index is 1, not 0 and the last index is equal to the value that -<code><nobr>%strlen</nobr></code> would assign given the same string. Index -values out of range result in an empty string. A negative length means -"until N-1 characters before the end of string", i.e. -<code><nobr>-1</nobr></code> means until end of string, -<code><nobr>-2</nobr></code> until one character before, etc. -<h3><a name="section-4.3">4.3 Multi-Line Macros: <code><nobr>%macro</nobr></code></a></h3> -<p>Multi-line macros are much more like the type of macro seen in MASM and -TASM: a multi-line macro definition in NASM looks something like this. -<p><pre> -%macro prologue 1 - - push ebp - mov ebp,esp - sub esp,%1 - -%endmacro -</pre> -<p>This defines a C-like function prologue as a macro: so you would invoke -the macro with a call such as -<p><pre> -myfunc: prologue 12 -</pre> -<p>which would expand to the three lines of code -<p><pre> -myfunc: push ebp - mov ebp,esp - sub esp,12 -</pre> -<p>The number <code><nobr>1</nobr></code> after the macro name in the -<code><nobr>%macro</nobr></code> line defines the number of parameters the -macro <code><nobr>prologue</nobr></code> expects to receive. The use of -<code><nobr>%1</nobr></code> inside the macro definition refers to the -first parameter to the macro call. With a macro taking more than one -parameter, subsequent parameters would be referred to as -<code><nobr>%2</nobr></code>, <code><nobr>%3</nobr></code> and so on. -<p>Multi-line macros, like single-line macros, are case-sensitive, unless -you define them using the alternative directive -<code><nobr>%imacro</nobr></code>. -<p>If you need to pass a comma as <em>part</em> of a parameter to a -multi-line macro, you can do that by enclosing the entire parameter in -braces. So you could code things like -<p><pre> -%macro silly 2 - - %2: db %1 - -%endmacro - - silly 'a', letter_a ; letter_a: db 'a' - silly 'ab', string_ab ; string_ab: db 'ab' - silly {13,10}, crlf ; crlf: db 13,10 -</pre> -<h4><a name="section-4.3.1">4.3.1 Recursive Multi-Line Macros: <code><nobr>%rmacro</nobr></code></a></h4> -<p>A multi-line macro cannot be referenced within itself, in order to -prevent accidental infinite recursion. -<p>Recursive multi-line macros allow for self-referencing, with the caveat -that the user is aware of the existence, use and purpose of recursive -multi-line macros. There is also a generous, but sane, upper limit to the -number of recursions, in order to prevent run-away memory consumption in -case of accidental infinite recursion. -<p>As with non-recursive multi-line macros, recursive multi-line macros are -case-sensitive, unless you define them using the alternative directive -<code><nobr>%irmacro</nobr></code>. -<h4><a name="section-4.3.2">4.3.2 Overloading Multi-Line Macros</a></h4> -<p>As with single-line macros, multi-line macros can be overloaded by -defining the same macro name several times with different numbers of -parameters. This time, no exception is made for macros with no parameters -at all. So you could define -<p><pre> -%macro prologue 0 - - push ebp - mov ebp,esp - -%endmacro -</pre> -<p>to define an alternative form of the function prologue which allocates -no local stack space. -<p>Sometimes, however, you might want to `overload' a machine instruction; -for example, you might want to define -<p><pre> -%macro push 2 - - push %1 - push %2 - -%endmacro -</pre> -<p>so that you could code -<p><pre> - push ebx ; this line is not a macro call - push eax,ecx ; but this one is -</pre> -<p>Ordinarily, NASM will give a warning for the first of the above two -lines, since <code><nobr>push</nobr></code> is now defined to be a macro, -and is being invoked with a number of parameters for which no definition -has been given. The correct code will still be generated, but the assembler -will give a warning. This warning can be disabled by the use of the -<code><nobr>-w-macro-params</nobr></code> command-line option (see -<a href="nasmdoc2.html#section-2.1.24">section 2.1.24</a>). -<h4><a name="section-4.3.3">4.3.3 Macro-Local Labels</a></h4> -<p>NASM allows you to define labels within a multi-line macro definition in -such a way as to make them local to the macro call: so calling the same -macro multiple times will use a different label each time. You do this by -prefixing <code><nobr>%%</nobr></code> to the label name. So you can invent -an instruction which executes a <code><nobr>RET</nobr></code> if the -<code><nobr>Z</nobr></code> flag is set by doing this: -<p><pre> -%macro retz 0 - - jnz %%skip - ret - %%skip: - -%endmacro -</pre> -<p>You can call this macro as many times as you want, and every time you -call it NASM will make up a different `real' name to substitute for the -label <code><nobr>%%skip</nobr></code>. The names NASM invents are of the -form <code><nobr>..@2345.skip</nobr></code>, where the number 2345 changes -with every macro call. The <code><nobr>..@</nobr></code> prefix prevents -macro-local labels from interfering with the local label mechanism, as -described in <a href="nasmdoc3.html#section-3.9">section 3.9</a>. You -should avoid defining your own labels in this form (the -<code><nobr>..@</nobr></code> prefix, then a number, then another period) -in case they interfere with macro-local labels. -<h4><a name="section-4.3.4">4.3.4 Greedy Macro Parameters</a></h4> -<p>Occasionally it is useful to define a macro which lumps its entire -command line into one parameter definition, possibly after extracting one -or two smaller parameters from the front. An example might be a macro to -write a text string to a file in MS-DOS, where you might want to be able to -write -<p><pre> - writefile [filehandle],"hello, world",13,10 -</pre> -<p>NASM allows you to define the last parameter of a macro to be -<em>greedy</em>, meaning that if you invoke the macro with more parameters -than it expects, all the spare parameters get lumped into the last defined -one along with the separating commas. So if you code: -<p><pre> -%macro writefile 2+ - - jmp %%endstr - %%str: db %2 - %%endstr: - mov dx,%%str - mov cx,%%endstr-%%str - mov bx,%1 - mov ah,0x40 - int 0x21 - -%endmacro -</pre> -<p>then the example call to <code><nobr>writefile</nobr></code> above will -work as expected: the text before the first comma, -<code><nobr>[filehandle]</nobr></code>, is used as the first macro -parameter and expanded when <code><nobr>%1</nobr></code> is referred to, -and all the subsequent text is lumped into <code><nobr>%2</nobr></code> and -placed after the <code><nobr>db</nobr></code>. -<p>The greedy nature of the macro is indicated to NASM by the use of the -<code><nobr>+</nobr></code> sign after the parameter count on the -<code><nobr>%macro</nobr></code> line. -<p>If you define a greedy macro, you are effectively telling NASM how it -should expand the macro given <em>any</em> number of parameters from the -actual number specified up to infinity; in this case, for example, NASM now -knows what to do when it sees a call to <code><nobr>writefile</nobr></code> -with 2, 3, 4 or more parameters. NASM will take this into account when -overloading macros, and will not allow you to define another form of -<code><nobr>writefile</nobr></code> taking 4 parameters (for example). -<p>Of course, the above macro could have been implemented as a non-greedy -macro, in which case the call to it would have had to look like -<p><pre> - writefile [filehandle], {"hello, world",13,10} -</pre> -<p>NASM provides both mechanisms for putting commas in macro parameters, -and you choose which one you prefer for each macro definition. -<p>See <a href="nasmdoc6.html#section-6.3.1">section 6.3.1</a> for a better -way to write the above macro. -<h4><a name="section-4.3.5">4.3.5 Default Macro Parameters</a></h4> -<p>NASM also allows you to define a multi-line macro with a <em>range</em> -of allowable parameter counts. If you do this, you can specify defaults for -omitted parameters. So, for example: -<p><pre> -%macro die 0-1 "Painful program death has occurred." - - writefile 2,%1 - mov ax,0x4c01 - int 0x21 - -%endmacro -</pre> -<p>This macro (which makes use of the <code><nobr>writefile</nobr></code> -macro defined in <a href="#section-4.3.4">section 4.3.4</a>) can be called -with an explicit error message, which it will display on the error output -stream before exiting, or it can be called with no parameters, in which -case it will use the default error message supplied in the macro -definition. -<p>In general, you supply a minimum and maximum number of parameters for a -macro of this type; the minimum number of parameters are then required in -the macro call, and then you provide defaults for the optional ones. So if -a macro definition began with the line -<p><pre> -%macro foobar 1-3 eax,[ebx+2] -</pre> -<p>then it could be called with between one and three parameters, and -<code><nobr>%1</nobr></code> would always be taken from the macro call. -<code><nobr>%2</nobr></code>, if not specified by the macro call, would -default to <code><nobr>eax</nobr></code>, and <code><nobr>%3</nobr></code> -if not specified would default to <code><nobr>[ebx+2]</nobr></code>. -<p>You can provide extra information to a macro by providing too many -default parameters: -<p><pre> -%macro quux 1 something -</pre> -<p>This will trigger a warning by default; see -<a href="nasmdoc2.html#section-2.1.24">section 2.1.24</a> for more -information. When <code><nobr>quux</nobr></code> is invoked, it receives -not one but two parameters. <code><nobr>something</nobr></code> can be -referred to as <code><nobr>%2</nobr></code>. The difference between passing -<code><nobr>something</nobr></code> this way and writing -<code><nobr>something</nobr></code> in the macro body is that with this way -<code><nobr>something</nobr></code> is evaluated when the macro is defined, -not when it is expanded. -<p>You may omit parameter defaults from the macro definition, in which case -the parameter default is taken to be blank. This can be useful for macros -which can take a variable number of parameters, since the -<code><nobr>%0</nobr></code> token (see <a href="#section-4.3.6">section -4.3.6</a>) allows you to determine how many parameters were really passed -to the macro call. -<p>This defaulting mechanism can be combined with the greedy-parameter -mechanism; so the <code><nobr>die</nobr></code> macro above could be made -more powerful, and more useful, by changing the first line of the -definition to -<p><pre> -%macro die 0-1+ "Painful program death has occurred.",13,10 -</pre> -<p>The maximum parameter count can be infinite, denoted by -<code><nobr>*</nobr></code>. In this case, of course, it is impossible to -provide a <em>full</em> set of default parameters. Examples of this usage -are shown in <a href="#section-4.3.7">section 4.3.7</a>. -<h4><a name="section-4.3.6">4.3.6 <code><nobr>%0</nobr></code>: Macro Parameter Counter</a></h4> -<p>The parameter reference <code><nobr>%0</nobr></code> will return a -numeric constant giving the number of parameters received, that is, if -<code><nobr>%0</nobr></code> is n then <code><nobr>%</nobr></code>n is the -last parameter. <code><nobr>%0</nobr></code> is mostly useful for macros -that can take a variable number of parameters. It can be used as an -argument to <code><nobr>%rep</nobr></code> (see -<a href="#section-4.5">section 4.5</a>) in order to iterate through all the -parameters of a macro. Examples are given in -<a href="#section-4.3.7">section 4.3.7</a>. -<h4><a name="section-4.3.7">4.3.7 <code><nobr>%rotate</nobr></code>: Rotating Macro Parameters</a></h4> -<p>Unix shell programmers will be familiar with the -<code><nobr>shift</nobr></code> shell command, which allows the arguments -passed to a shell script (referenced as <code><nobr>$1</nobr></code>, -<code><nobr>$2</nobr></code> and so on) to be moved left by one place, so -that the argument previously referenced as <code><nobr>$2</nobr></code> -becomes available as <code><nobr>$1</nobr></code>, and the argument -previously referenced as <code><nobr>$1</nobr></code> is no longer -available at all. -<p>NASM provides a similar mechanism, in the form of -<code><nobr>%rotate</nobr></code>. As its name suggests, it differs from -the Unix <code><nobr>shift</nobr></code> in that no parameters are lost: -parameters rotated off the left end of the argument list reappear on the -right, and vice versa. -<p><code><nobr>%rotate</nobr></code> is invoked with a single numeric -argument (which may be an expression). The macro parameters are rotated to -the left by that many places. If the argument to -<code><nobr>%rotate</nobr></code> is negative, the macro parameters are -rotated to the right. -<p>So a pair of macros to save and restore a set of registers might work as -follows: -<p><pre> -%macro multipush 1-* - - %rep %0 - push %1 - %rotate 1 - %endrep - -%endmacro -</pre> -<p>This macro invokes the <code><nobr>PUSH</nobr></code> instruction on -each of its arguments in turn, from left to right. It begins by pushing its -first argument, <code><nobr>%1</nobr></code>, then invokes -<code><nobr>%rotate</nobr></code> to move all the arguments one place to -the left, so that the original second argument is now available as -<code><nobr>%1</nobr></code>. Repeating this procedure as many times as -there were arguments (achieved by supplying <code><nobr>%0</nobr></code> as -the argument to <code><nobr>%rep</nobr></code>) causes each argument in -turn to be pushed. -<p>Note also the use of <code><nobr>*</nobr></code> as the maximum -parameter count, indicating that there is no upper limit on the number of -parameters you may supply to the <code><nobr>multipush</nobr></code> macro. -<p>It would be convenient, when using this macro, to have a -<code><nobr>POP</nobr></code> equivalent, which <em>didn't</em> require the -arguments to be given in reverse order. Ideally, you would write the -<code><nobr>multipush</nobr></code> macro call, then cut-and-paste the line -to where the pop needed to be done, and change the name of the called macro -to <code><nobr>multipop</nobr></code>, and the macro would take care of -popping the registers in the opposite order from the one in which they were -pushed. -<p>This can be done by the following definition: -<p><pre> -%macro multipop 1-* - - %rep %0 - %rotate -1 - pop %1 - %endrep - -%endmacro -</pre> -<p>This macro begins by rotating its arguments one place to the -<em>right</em>, so that the original <em>last</em> argument appears as -<code><nobr>%1</nobr></code>. This is then popped, and the arguments are -rotated right again, so the second-to-last argument becomes -<code><nobr>%1</nobr></code>. Thus the arguments are iterated through in -reverse order. -<h4><a name="section-4.3.8">4.3.8 Concatenating Macro Parameters</a></h4> -<p>NASM can concatenate macro parameters and macro indirection constructs -on to other text surrounding them. This allows you to declare a family of -symbols, for example, in a macro definition. If, for example, you wanted to -generate a table of key codes along with offsets into the table, you could -code something like -<p><pre> -%macro keytab_entry 2 - - keypos%1 equ $-keytab - db %2 - -%endmacro - -keytab: - keytab_entry F1,128+1 - keytab_entry F2,128+2 - keytab_entry Return,13 -</pre> -<p>which would expand to -<p><pre> -keytab: -keyposF1 equ $-keytab - db 128+1 -keyposF2 equ $-keytab - db 128+2 -keyposReturn equ $-keytab - db 13 -</pre> -<p>You can just as easily concatenate text on to the other end of a macro -parameter, by writing <code><nobr>%1foo</nobr></code>. -<p>If you need to append a <em>digit</em> to a macro parameter, for example -defining labels <code><nobr>foo1</nobr></code> and -<code><nobr>foo2</nobr></code> when passed the parameter -<code><nobr>foo</nobr></code>, you can't code <code><nobr>%11</nobr></code> -because that would be taken as the eleventh macro parameter. Instead, you -must code <code><nobr>%{1}1</nobr></code>, which will separate the first -<code><nobr>1</nobr></code> (giving the number of the macro parameter) from -the second (literal text to be concatenated to the parameter). -<p>This concatenation can also be applied to other preprocessor in-line -objects, such as macro-local labels (<a href="#section-4.3.3">section -4.3.3</a>) and context-local labels (<a href="#section-4.7.2">section -4.7.2</a>). In all cases, ambiguities in syntax can be resolved by -enclosing everything after the <code><nobr>%</nobr></code> sign and before -the literal text in braces: so <code><nobr>%{%foo}bar</nobr></code> -concatenates the text <code><nobr>bar</nobr></code> to the end of the real -name of the macro-local label <code><nobr>%%foo</nobr></code>. (This is -unnecessary, since the form NASM uses for the real names of macro-local -labels means that the two usages <code><nobr>%{%foo}bar</nobr></code> and -<code><nobr>%%foobar</nobr></code> would both expand to the same thing -anyway; nevertheless, the capability is there.) -<p>The single-line macro indirection construct, -<code><nobr>%[...]</nobr></code> (<a href="#section-4.1.3">section -4.1.3</a>), behaves the same way as macro parameters for the purpose of -concatenation. -<p>See also the <code><nobr>%+</nobr></code> operator, -<a href="#section-4.1.4">section 4.1.4</a>. -<h4><a name="section-4.3.9">4.3.9 Condition Codes as Macro Parameters</a></h4> -<p>NASM can give special treatment to a macro parameter which contains a -condition code. For a start, you can refer to the macro parameter -<code><nobr>%1</nobr></code> by means of the alternative syntax -<code><nobr>%+1</nobr></code>, which informs NASM that this macro parameter -is supposed to contain a condition code, and will cause the preprocessor to -report an error message if the macro is called with a parameter which is -<em>not</em> a valid condition code. -<p>Far more usefully, though, you can refer to the macro parameter by means -of <code><nobr>%-1</nobr></code>, which NASM will expand as the -<em>inverse</em> condition code. So the <code><nobr>retz</nobr></code> -macro defined in <a href="#section-4.3.3">section 4.3.3</a> can be replaced -by a general conditional-return macro like this: -<p><pre> -%macro retc 1 - - j%-1 %%skip - ret - %%skip: - -%endmacro -</pre> -<p>This macro can now be invoked using calls like -<code><nobr>retc ne</nobr></code>, which will cause the conditional-jump -instruction in the macro expansion to come out as -<code><nobr>JE</nobr></code>, or <code><nobr>retc po</nobr></code> which -will make the jump a <code><nobr>JPE</nobr></code>. -<p>The <code><nobr>%+1</nobr></code> macro-parameter reference is quite -happy to interpret the arguments <code><nobr>CXZ</nobr></code> and -<code><nobr>ECXZ</nobr></code> as valid condition codes; however, -<code><nobr>%-1</nobr></code> will report an error if passed either of -these, because no inverse condition code exists. -<h4><a name="section-4.3.10">4.3.10 Disabling Listing Expansion</a></h4> -<p>When NASM is generating a listing file from your program, it will -generally expand multi-line macros by means of writing the macro call and -then listing each line of the expansion. This allows you to see which -instructions in the macro expansion are generating what code; however, for -some macros this clutters the listing up unnecessarily. -<p>NASM therefore provides the <code><nobr>.nolist</nobr></code> qualifier, -which you can include in a macro definition to inhibit the expansion of the -macro in the listing file. The <code><nobr>.nolist</nobr></code> qualifier -comes directly after the number of parameters, like this: -<p><pre> -%macro foo 1.nolist -</pre> -<p>Or like this: -<p><pre> -%macro bar 1-5+.nolist a,b,c,d,e,f,g,h -</pre> -<h4><a name="section-4.3.11">4.3.11 Undefining Multi-Line Macros: <code><nobr>%unmacro</nobr></code></a></h4> -<p>Multi-line macros can be removed with the -<code><nobr>%unmacro</nobr></code> directive. Unlike the -<code><nobr>%undef</nobr></code> directive, however, -<code><nobr>%unmacro</nobr></code> takes an argument specification, and -will only remove exact matches with that argument specification. -<p>For example: -<p><pre> -%macro foo 1-3 - ; Do something -%endmacro -%unmacro foo 1-3 -</pre> -<p>removes the previously defined macro <code><nobr>foo</nobr></code>, but -<p><pre> -%macro bar 1-3 - ; Do something -%endmacro -%unmacro bar 1 -</pre> -<p>does <em>not</em> remove the macro <code><nobr>bar</nobr></code>, since -the argument specification does not match exactly. -<h4><a name="section-4.3.12">4.3.12 Exiting Multi-Line Macros: <code><nobr>%exitmacro</nobr></code></a></h4> -<p>Multi-line macro expansions can be arbitrarily terminated with the -<code><nobr>%exitmacro</nobr></code> directive. -<p>For example: -<p><pre> -%macro foo 1-3 - ; Do something - %if<condition> - %exitmacro - %endif - ; Do something -%endmacro -</pre> -<h3><a name="section-4.4">4.4 Conditional Assembly</a></h3> -<p>Similarly to the C preprocessor, NASM allows sections of a source file -to be assembled only if certain conditions are met. The general syntax of -this feature looks like this: -<p><pre> -%if<condition> - ; some code which only appears if <condition> is met -%elif<condition2> - ; only appears if <condition> is not met but <condition2> is -%else - ; this appears if neither <condition> nor <condition2> was met -%endif -</pre> -<p>The inverse forms <code><nobr>%ifn</nobr></code> and -<code><nobr>%elifn</nobr></code> are also supported. -<p>The <code><nobr>%else</nobr></code> clause is optional, as is the -<code><nobr>%elif</nobr></code> clause. You can have more than one -<code><nobr>%elif</nobr></code> clause as well. -<p>There are a number of variants of the <code><nobr>%if</nobr></code> -directive. Each has its corresponding <code><nobr>%elif</nobr></code>, -<code><nobr>%ifn</nobr></code>, and <code><nobr>%elifn</nobr></code> -directives; for example, the equivalents to the -<code><nobr>%ifdef</nobr></code> directive are -<code><nobr>%elifdef</nobr></code>, <code><nobr>%ifndef</nobr></code>, and -<code><nobr>%elifndef</nobr></code>. -<h4><a name="section-4.4.1">4.4.1 <code><nobr>%ifdef</nobr></code>: Testing Single-Line Macro Existence</a></h4> -<p>Beginning a conditional-assembly block with the line -<code><nobr>%ifdef MACRO</nobr></code> will assemble the subsequent code -if, and only if, a single-line macro called <code><nobr>MACRO</nobr></code> -is defined. If not, then the <code><nobr>%elif</nobr></code> and -<code><nobr>%else</nobr></code> blocks (if any) will be processed instead. -<p>For example, when debugging a program, you might want to write code such -as -<p><pre> - ; perform some function -%ifdef DEBUG - writefile 2,"Function performed successfully",13,10 -%endif - ; go and do something else -</pre> -<p>Then you could use the command-line option -<code><nobr>-dDEBUG</nobr></code> to create a version of the program which -produced debugging messages, and remove the option to generate the final -release version of the program. -<p>You can test for a macro <em>not</em> being defined by using -<code><nobr>%ifndef</nobr></code> instead of -<code><nobr>%ifdef</nobr></code>. You can also test for macro definitions -in <code><nobr>%elif</nobr></code> blocks by using -<code><nobr>%elifdef</nobr></code> and <code><nobr>%elifndef</nobr></code>. -<h4><a name="section-4.4.2">4.4.2 <code><nobr>%ifmacro</nobr></code>: Testing Multi-Line Macro Existence</a></h4> -<p>The <code><nobr>%ifmacro</nobr></code> directive operates in the same -way as the <code><nobr>%ifdef</nobr></code> directive, except that it -checks for the existence of a multi-line macro. -<p>For example, you may be working with a large project and not have -control over the macros in a library. You may want to create a macro with -one name if it doesn't already exist, and another name if one with that -name does exist. -<p>The <code><nobr>%ifmacro</nobr></code> is considered true if defining a -macro with the given name and number of arguments would cause a definitions -conflict. For example: -<p><pre> -%ifmacro MyMacro 1-3 - - %error "MyMacro 1-3" causes a conflict with an existing macro. - -%else - - %macro MyMacro 1-3 - - ; insert code to define the macro - - %endmacro - -%endif -</pre> -<p>This will create the macro "MyMacro 1-3" if no macro already exists -which would conflict with it, and emits a warning if there would be a -definition conflict. -<p>You can test for the macro not existing by using the -<code><nobr>%ifnmacro</nobr></code> instead of -<code><nobr>%ifmacro</nobr></code>. Additional tests can be performed in -<code><nobr>%elif</nobr></code> blocks by using -<code><nobr>%elifmacro</nobr></code> and -<code><nobr>%elifnmacro</nobr></code>. -<h4><a name="section-4.4.3">4.4.3 <code><nobr>%ifctx</nobr></code>: Testing the Context Stack</a></h4> -<p>The conditional-assembly construct <code><nobr>%ifctx</nobr></code> will -cause the subsequent code to be assembled if and only if the top context on -the preprocessor's context stack has the same name as one of the arguments. -As with <code><nobr>%ifdef</nobr></code>, the inverse and -<code><nobr>%elif</nobr></code> forms <code><nobr>%ifnctx</nobr></code>, -<code><nobr>%elifctx</nobr></code> and <code><nobr>%elifnctx</nobr></code> -are also supported. -<p>For more details of the context stack, see -<a href="#section-4.7">section 4.7</a>. For a sample use of -<code><nobr>%ifctx</nobr></code>, see <a href="#section-4.7.5">section -4.7.5</a>. -<h4><a name="section-4.4.4">4.4.4 <code><nobr>%if</nobr></code>: Testing Arbitrary Numeric Expressions</a></h4> -<p>The conditional-assembly construct <code><nobr>%if expr</nobr></code> -will cause the subsequent code to be assembled if and only if the value of -the numeric expression <code><nobr>expr</nobr></code> is non-zero. An -example of the use of this feature is in deciding when to break out of a -<code><nobr>%rep</nobr></code> preprocessor loop: see -<a href="#section-4.5">section 4.5</a> for a detailed example. -<p>The expression given to <code><nobr>%if</nobr></code>, and its -counterpart <code><nobr>%elif</nobr></code>, is a critical expression (see -<a href="nasmdoc3.html#section-3.8">section 3.8</a>). -<p><code><nobr>%if</nobr></code> extends the normal NASM expression syntax, -by providing a set of relational operators which are not normally available -in expressions. The operators <code><nobr>=</nobr></code>, -<code><nobr><</nobr></code>, <code><nobr>></nobr></code>, -<code><nobr><=</nobr></code>, <code><nobr>>=</nobr></code> and -<code><nobr><></nobr></code> test equality, less-than, greater-than, -less-or-equal, greater-or-equal and not-equal respectively. The C-like -forms <code><nobr>==</nobr></code> and <code><nobr>!=</nobr></code> are -supported as alternative forms of <code><nobr>=</nobr></code> and -<code><nobr><></nobr></code>. In addition, low-priority logical -operators <code><nobr>&&</nobr></code>, -<code><nobr>^^</nobr></code> and <code><nobr>||</nobr></code> are provided, -supplying logical AND, logical XOR and logical OR. These work like the C -logical operators (although C has no logical XOR), in that they always -return either 0 or 1, and treat any non-zero input as 1 (so that -<code><nobr>^^</nobr></code>, for example, returns 1 if exactly one of its -inputs is zero, and 0 otherwise). The relational operators also return 1 -for true and 0 for false. -<p>Like other <code><nobr>%if</nobr></code> constructs, -<code><nobr>%if</nobr></code> has a counterpart -<code><nobr>%elif</nobr></code>, and negative forms -<code><nobr>%ifn</nobr></code> and <code><nobr>%elifn</nobr></code>. -<h4><a name="section-4.4.5">4.4.5 <code><nobr>%ifidn</nobr></code> and <code><nobr>%ifidni</nobr></code>: Testing Exact Text Identity</a></h4> -<p>The construct <code><nobr>%ifidn text1,text2</nobr></code> will cause -the subsequent code to be assembled if and only if -<code><nobr>text1</nobr></code> and <code><nobr>text2</nobr></code>, after -expanding single-line macros, are identical pieces of text. Differences in -white space are not counted. -<p><code><nobr>%ifidni</nobr></code> is similar to -<code><nobr>%ifidn</nobr></code>, but is case-insensitive. -<p>For example, the following macro pushes a register or number on the -stack, and allows you to treat <code><nobr>IP</nobr></code> as a real -register: -<p><pre> -%macro pushparam 1 - - %ifidni %1,ip - call %%label - %%label: - %else - push %1 - %endif - -%endmacro -</pre> -<p>Like other <code><nobr>%if</nobr></code> constructs, -<code><nobr>%ifidn</nobr></code> has a counterpart -<code><nobr>%elifidn</nobr></code>, and negative forms -<code><nobr>%ifnidn</nobr></code> and <code><nobr>%elifnidn</nobr></code>. -Similarly, <code><nobr>%ifidni</nobr></code> has counterparts -<code><nobr>%elifidni</nobr></code>, <code><nobr>%ifnidni</nobr></code> and -<code><nobr>%elifnidni</nobr></code>. -<h4><a name="section-4.4.6">4.4.6 <code><nobr>%ifid</nobr></code>, <code><nobr>%ifnum</nobr></code>, <code><nobr>%ifstr</nobr></code>: Testing Token Types</a></h4> -<p>Some macros will want to perform different tasks depending on whether -they are passed a number, a string, or an identifier. For example, a string -output macro might want to be able to cope with being passed either a -string constant or a pointer to an existing string. -<p>The conditional assembly construct <code><nobr>%ifid</nobr></code>, -taking one parameter (which may be blank), assembles the subsequent code if -and only if the first token in the parameter exists and is an identifier. -<code><nobr>%ifnum</nobr></code> works similarly, but tests for the token -being a numeric constant; <code><nobr>%ifstr</nobr></code> tests for it -being a string. -<p>For example, the <code><nobr>writefile</nobr></code> macro defined in -<a href="#section-4.3.4">section 4.3.4</a> can be extended to take -advantage of <code><nobr>%ifstr</nobr></code> in the following fashion: -<p><pre> -%macro writefile 2-3+ - - %ifstr %2 - jmp %%endstr - %if %0 = 3 - %%str: db %2,%3 - %else - %%str: db %2 - %endif - %%endstr: mov dx,%%str - mov cx,%%endstr-%%str - %else - mov dx,%2 - mov cx,%3 - %endif - mov bx,%1 - mov ah,0x40 - int 0x21 - -%endmacro -</pre> -<p>Then the <code><nobr>writefile</nobr></code> macro can cope with being -called in either of the following two ways: -<p><pre> - writefile [file], strpointer, length - writefile [file], "hello", 13, 10 -</pre> -<p>In the first, <code><nobr>strpointer</nobr></code> is used as the -address of an already-declared string, and <code><nobr>length</nobr></code> -is used as its length; in the second, a string is given to the macro, which -therefore declares it itself and works out the address and length for -itself. -<p>Note the use of <code><nobr>%if</nobr></code> inside the -<code><nobr>%ifstr</nobr></code>: this is to detect whether the macro was -passed two arguments (so the string would be a single string constant, and -<code><nobr>db %2</nobr></code> would be adequate) or more (in which case, -all but the first two would be lumped together into -<code><nobr>%3</nobr></code>, and <code><nobr>db %2,%3</nobr></code> would -be required). -<p>The usual <code><nobr>%elif</nobr></code>..., -<code><nobr>%ifn</nobr></code>..., and <code><nobr>%elifn</nobr></code>... -versions exist for each of <code><nobr>%ifid</nobr></code>, -<code><nobr>%ifnum</nobr></code> and <code><nobr>%ifstr</nobr></code>. -<h4><a name="section-4.4.7">4.4.7 <code><nobr>%iftoken</nobr></code>: Test for a Single Token</a></h4> -<p>Some macros will want to do different things depending on if it is -passed a single token (e.g. paste it to something else using -<code><nobr>%+</nobr></code>) versus a multi-token sequence. -<p>The conditional assembly construct <code><nobr>%iftoken</nobr></code> -assembles the subsequent code if and only if the expanded parameters -consist of exactly one token, possibly surrounded by whitespace. -<p>For example: -<p><pre> -%iftoken 1 -</pre> -<p>will assemble the subsequent code, but -<p><pre> -%iftoken -1 -</pre> -<p>will not, since <code><nobr>-1</nobr></code> contains two tokens: the -unary minus operator <code><nobr>-</nobr></code>, and the number -<code><nobr>1</nobr></code>. -<p>The usual <code><nobr>%eliftoken</nobr></code>, -<code><nobr>%ifntoken</nobr></code>, and -<code><nobr>%elifntoken</nobr></code> variants are also provided. -<h4><a name="section-4.4.8">4.4.8 <code><nobr>%ifempty</nobr></code>: Test for Empty Expansion</a></h4> -<p>The conditional assembly construct <code><nobr>%ifempty</nobr></code> -assembles the subsequent code if and only if the expanded parameters do not -contain any tokens at all, whitespace excepted. -<p>The usual <code><nobr>%elifempty</nobr></code>, -<code><nobr>%ifnempty</nobr></code>, and -<code><nobr>%elifnempty</nobr></code> variants are also provided. -<h3><a name="section-4.5">4.5 Preprocessor Loops: <code><nobr>%rep</nobr></code></a></h3> -<p>NASM's <code><nobr>TIMES</nobr></code> prefix, though useful, cannot be -used to invoke a multi-line macro multiple times, because it is processed -by NASM after macros have already been expanded. Therefore NASM provides -another form of loop, this time at the preprocessor level: -<code><nobr>%rep</nobr></code>. -<p>The directives <code><nobr>%rep</nobr></code> and -<code><nobr>%endrep</nobr></code> (<code><nobr>%rep</nobr></code> takes a -numeric argument, which can be an expression; -<code><nobr>%endrep</nobr></code> takes no arguments) can be used to -enclose a chunk of code, which is then replicated as many times as -specified by the preprocessor: -<p><pre> -%assign i 0 -%rep 64 - inc word [table+2*i] -%assign i i+1 -%endrep -</pre> -<p>This will generate a sequence of 64 <code><nobr>INC</nobr></code> -instructions, incrementing every word of memory from -<code><nobr>[table]</nobr></code> to <code><nobr>[table+126]</nobr></code>. -<p>For more complex termination conditions, or to break out of a repeat -loop part way along, you can use the <code><nobr>%exitrep</nobr></code> -directive to terminate the loop, like this: -<p><pre> -fibonacci: -%assign i 0 -%assign j 1 -%rep 100 -%if j > 65535 - %exitrep -%endif - dw j -%assign k j+i -%assign i j -%assign j k -%endrep - -fib_number equ ($-fibonacci)/2 -</pre> -<p>This produces a list of all the Fibonacci numbers that will fit in 16 -bits. Note that a maximum repeat count must still be given to -<code><nobr>%rep</nobr></code>. This is to prevent the possibility of NASM -getting into an infinite loop in the preprocessor, which (on multitasking -or multi-user systems) would typically cause all the system memory to be -gradually used up and other applications to start crashing. -<h3><a name="section-4.6">4.6 Source Files and Dependencies</a></h3> -<p>These commands allow you to split your sources into multiple files. -<h4><a name="section-4.6.1">4.6.1 <code><nobr>%include</nobr></code>: Including Other Files</a></h4> -<p>Using, once again, a very similar syntax to the C preprocessor, NASM's -preprocessor lets you include other source files into your code. This is -done by the use of the <code><nobr>%include</nobr></code> directive: -<p><pre> -%include "macros.mac" -</pre> -<p>will include the contents of the file -<code><nobr>macros.mac</nobr></code> into the source file containing the -<code><nobr>%include</nobr></code> directive. -<p>Include files are searched for in the current directory (the directory -you're in when you run NASM, as opposed to the location of the NASM -executable or the location of the source file), plus any directories -specified on the NASM command line using the <code><nobr>-i</nobr></code> -option. -<p>The standard C idiom for preventing a file being included more than once -is just as applicable in NASM: if the file -<code><nobr>macros.mac</nobr></code> has the form -<p><pre> -%ifndef MACROS_MAC - %define MACROS_MAC - ; now define some macros -%endif -</pre> -<p>then including the file more than once will not cause errors, because -the second time the file is included nothing will happen because the macro -<code><nobr>MACROS_MAC</nobr></code> will already be defined. -<p>You can force a file to be included even if there is no -<code><nobr>%include</nobr></code> directive that explicitly includes it, -by using the <code><nobr>-p</nobr></code> option on the NASM command line -(see <a href="nasmdoc2.html#section-2.1.17">section 2.1.17</a>). -<h4><a name="section-4.6.2">4.6.2 <code><nobr>%pathsearch</nobr></code>: Search the Include Path</a></h4> -<p>The <code><nobr>%pathsearch</nobr></code> directive takes a single-line -macro name and a filename, and declare or redefines the specified -single-line macro to be the include-path-resolved version of the filename, -if the file exists (otherwise, it is passed unchanged.) -<p>For example, -<p><pre> -%pathsearch MyFoo "foo.bin" -</pre> -<p>... with <code><nobr>-Ibins/</nobr></code> in the include path may end -up defining the macro <code><nobr>MyFoo</nobr></code> to be -<code><nobr>"bins/foo.bin"</nobr></code>. -<h4><a name="section-4.6.3">4.6.3 <code><nobr>%depend</nobr></code>: Add Dependent Files</a></h4> -<p>The <code><nobr>%depend</nobr></code> directive takes a filename and -adds it to the list of files to be emitted as dependency generation when -the <code><nobr>-M</nobr></code> options and its relatives (see -<a href="nasmdoc2.html#section-2.1.4">section 2.1.4</a>) are used. It -produces no output. -<p>This is generally used in conjunction with -<code><nobr>%pathsearch</nobr></code>. For example, a simplified version of -the standard macro wrapper for the <code><nobr>INCBIN</nobr></code> -directive looks like: -<p><pre> -%imacro incbin 1-2+ 0 -%pathsearch dep %1 -%depend dep - incbin dep,%2 -%endmacro -</pre> -<p>This first resolves the location of the file into the macro -<code><nobr>dep</nobr></code>, then adds it to the dependency lists, and -finally issues the assembler-level <code><nobr>INCBIN</nobr></code> -directive. -<h4><a name="section-4.6.4">4.6.4 <code><nobr>%use</nobr></code>: Include Standard Macro Package</a></h4> -<p>The <code><nobr>%use</nobr></code> directive is similar to -<code><nobr>%include</nobr></code>, but rather than including the contents -of a file, it includes a named standard macro package. The standard macro -packages are part of NASM, and are described in -<a href="nasmdoc5.html">chapter 5</a>. -<p>Unlike the <code><nobr>%include</nobr></code> directive, package names -for the <code><nobr>%use</nobr></code> directive do not require quotes, but -quotes are permitted. In NASM 2.04 and 2.05 the unquoted form would be -macro-expanded; this is no longer true. Thus, the following lines are -equivalent: -<p><pre> -%use altreg -%use 'altreg' -</pre> -<p>Standard macro packages are protected from multiple inclusion. When a -standard macro package is used, a testable single-line macro of the form -<code><nobr>__USE_</nobr></code><em>package</em><code><nobr>__</nobr></code> -is also defined, see <a href="#section-4.11.8">section 4.11.8</a>. -<h3><a name="section-4.7">4.7 The Context Stack</a></h3> -<p>Having labels that are local to a macro definition is sometimes not -quite powerful enough: sometimes you want to be able to share labels -between several macro calls. An example might be a -<code><nobr>REPEAT</nobr></code> ... <code><nobr>UNTIL</nobr></code> loop, -in which the expansion of the <code><nobr>REPEAT</nobr></code> macro would -need to be able to refer to a label which the -<code><nobr>UNTIL</nobr></code> macro had defined. However, for such a -macro you would also want to be able to nest these loops. -<p>NASM provides this level of power by means of a <em>context stack</em>. -The preprocessor maintains a stack of <em>contexts</em>, each of which is -characterized by a name. You add a new context to the stack using the -<code><nobr>%push</nobr></code> directive, and remove one using -<code><nobr>%pop</nobr></code>. You can define labels that are local to a -particular context on the stack. -<h4><a name="section-4.7.1">4.7.1 <code><nobr>%push</nobr></code> and <code><nobr>%pop</nobr></code>: Creating and Removing Contexts</a></h4> -<p>The <code><nobr>%push</nobr></code> directive is used to create a new -context and place it on the top of the context stack. -<code><nobr>%push</nobr></code> takes an optional argument, which is the -name of the context. For example: -<p><pre> -%push foobar -</pre> -<p>This pushes a new context called <code><nobr>foobar</nobr></code> on the -stack. You can have several contexts on the stack with the same name: they -can still be distinguished. If no name is given, the context is unnamed -(this is normally used when both the <code><nobr>%push</nobr></code> and -the <code><nobr>%pop</nobr></code> are inside a single macro definition.) -<p>The directive <code><nobr>%pop</nobr></code>, taking one optional -argument, removes the top context from the context stack and destroys it, -along with any labels associated with it. If an argument is given, it must -match the name of the current context, otherwise it will issue an error. -<h4><a name="section-4.7.2">4.7.2 Context-Local Labels</a></h4> -<p>Just as the usage <code><nobr>%%foo</nobr></code> defines a label which -is local to the particular macro call in which it is used, the usage -<code><nobr>%$foo</nobr></code> is used to define a label which is local to -the context on the top of the context stack. So the -<code><nobr>REPEAT</nobr></code> and <code><nobr>UNTIL</nobr></code> -example given above could be implemented by means of: -<p><pre> -%macro repeat 0 - - %push repeat - %$begin: - -%endmacro - -%macro until 1 - - j%-1 %$begin - %pop - -%endmacro -</pre> -<p>and invoked by means of, for example, -<p><pre> - mov cx,string - repeat - add cx,3 - scasb - until e -</pre> -<p>which would scan every fourth byte of a string in search of the byte in -<code><nobr>AL</nobr></code>. -<p>If you need to define, or access, labels local to the context -<em>below</em> the top one on the stack, you can use -<code><nobr>%$$foo</nobr></code>, or <code><nobr>%$$$foo</nobr></code> for -the context below that, and so on. -<h4><a name="section-4.7.3">4.7.3 Context-Local Single-Line Macros</a></h4> -<p>NASM also allows you to define single-line macros which are local to a -particular context, in just the same way: -<p><pre> -%define %$localmac 3 -</pre> -<p>will define the single-line macro <code><nobr>%$localmac</nobr></code> -to be local to the top context on the stack. Of course, after a subsequent -<code><nobr>%push</nobr></code>, it can then still be accessed by the name -<code><nobr>%$$localmac</nobr></code>. -<h4><a name="section-4.7.4">4.7.4 <code><nobr>%repl</nobr></code>: Renaming a Context</a></h4> -<p>If you need to change the name of the top context on the stack (in -order, for example, to have it respond differently to -<code><nobr>%ifctx</nobr></code>), you can execute a -<code><nobr>%pop</nobr></code> followed by a -<code><nobr>%push</nobr></code>; but this will have the side effect of -destroying all context-local labels and macros associated with the context -that was just popped. -<p>NASM provides the directive <code><nobr>%repl</nobr></code>, which -<em>replaces</em> a context with a different name, without touching the -associated macros and labels. So you could replace the destructive code -<p><pre> -%pop -%push newname -</pre> -<p>with the non-destructive version -<code><nobr>%repl newname</nobr></code>. -<h4><a name="section-4.7.5">4.7.5 Example Use of the Context Stack: Block IFs</a></h4> -<p>This example makes use of almost all the context-stack features, -including the conditional-assembly construct -<code><nobr>%ifctx</nobr></code>, to implement a block IF statement as a -set of macros. -<p><pre> -%macro if 1 - - %push if - j%-1 %$ifnot - -%endmacro - -%macro else 0 - - %ifctx if - %repl else - jmp %$ifend - %$ifnot: - %else - %error "expected `if' before `else'" - %endif - -%endmacro - -%macro endif 0 - - %ifctx if - %$ifnot: - %pop - %elifctx else - %$ifend: - %pop - %else - %error "expected `if' or `else' before `endif'" - %endif - -%endmacro -</pre> -<p>This code is more robust than the <code><nobr>REPEAT</nobr></code> and -<code><nobr>UNTIL</nobr></code> macros given in -<a href="#section-4.7.2">section 4.7.2</a>, because it uses conditional -assembly to check that the macros are issued in the right order (for -example, not calling <code><nobr>endif</nobr></code> before -<code><nobr>if</nobr></code>) and issues a <code><nobr>%error</nobr></code> -if they're not. -<p>In addition, the <code><nobr>endif</nobr></code> macro has to be able to -cope with the two distinct cases of either directly following an -<code><nobr>if</nobr></code>, or following an -<code><nobr>else</nobr></code>. It achieves this, again, by using -conditional assembly to do different things depending on whether the -context on top of the stack is <code><nobr>if</nobr></code> or -<code><nobr>else</nobr></code>. -<p>The <code><nobr>else</nobr></code> macro has to preserve the context on -the stack, in order to have the <code><nobr>%$ifnot</nobr></code> referred -to by the <code><nobr>if</nobr></code> macro be the same as the one defined -by the <code><nobr>endif</nobr></code> macro, but has to change the -context's name so that <code><nobr>endif</nobr></code> will know there was -an intervening <code><nobr>else</nobr></code>. It does this by the use of -<code><nobr>%repl</nobr></code>. -<p>A sample usage of these macros might look like: -<p><pre> - cmp ax,bx - - if ae - cmp bx,cx - - if ae - mov ax,cx - else - mov ax,bx - endif - - else - cmp ax,cx - - if ae - mov ax,cx - endif - - endif -</pre> -<p>The block-<code><nobr>IF</nobr></code> macros handle nesting quite -happily, by means of pushing another context, describing the inner -<code><nobr>if</nobr></code>, on top of the one describing the outer -<code><nobr>if</nobr></code>; thus <code><nobr>else</nobr></code> and -<code><nobr>endif</nobr></code> always refer to the last unmatched -<code><nobr>if</nobr></code> or <code><nobr>else</nobr></code>. -<h3><a name="section-4.8">4.8 Stack Relative Preprocessor Directives</a></h3> -<p>The following preprocessor directives provide a way to use labels to -refer to local variables allocated on the stack. -<ul> -<li><code><nobr>%arg</nobr></code> (see <a href="#section-4.8.1">section -4.8.1</a>) -<li><code><nobr>%stacksize</nobr></code> (see -<a href="#section-4.8.2">section 4.8.2</a>) -<li><code><nobr>%local</nobr></code> (see <a href="#section-4.8.3">section -4.8.3</a>) -</ul> -<h4><a name="section-4.8.1">4.8.1 <code><nobr>%arg</nobr></code> Directive</a></h4> -<p>The <code><nobr>%arg</nobr></code> directive is used to simplify the -handling of parameters passed on the stack. Stack based parameter passing -is used by many high level languages, including C, C++ and Pascal. -<p>While NASM has macros which attempt to duplicate this functionality (see -<a href="nasmdoc8.html#section-8.4.5">section 8.4.5</a>), the syntax is not -particularly convenient to use. and is not TASM compatible. Here is an -example which shows the use of <code><nobr>%arg</nobr></code> without any -external macros: -<p><pre> -some_function: - - %push mycontext ; save the current context - %stacksize large ; tell NASM to use bp - %arg i:word, j_ptr:word - - mov ax,[i] - mov bx,[j_ptr] - add ax,[bx] - ret - - %pop ; restore original context -</pre> -<p>This is similar to the procedure defined in -<a href="nasmdoc8.html#section-8.4.5">section 8.4.5</a> and adds the value -in i to the value pointed to by j_ptr and returns the sum in the ax -register. See <a href="#section-4.7.1">section 4.7.1</a> for an explanation -of <code><nobr>push</nobr></code> and <code><nobr>pop</nobr></code> and the -use of context stacks. -<h4><a name="section-4.8.2">4.8.2 <code><nobr>%stacksize</nobr></code> Directive</a></h4> -<p>The <code><nobr>%stacksize</nobr></code> directive is used in -conjunction with the <code><nobr>%arg</nobr></code> (see -<a href="#section-4.8.1">section 4.8.1</a>) and the -<code><nobr>%local</nobr></code> (see <a href="#section-4.8.3">section -4.8.3</a>) directives. It tells NASM the default size to use for subsequent -<code><nobr>%arg</nobr></code> and <code><nobr>%local</nobr></code> -directives. The <code><nobr>%stacksize</nobr></code> directive takes one -required argument which is one of <code><nobr>flat</nobr></code>, -<code><nobr>flat64</nobr></code>, <code><nobr>large</nobr></code> or -<code><nobr>small</nobr></code>. -<p><pre> -%stacksize flat -</pre> -<p>This form causes NASM to use stack-based parameter addressing relative -to <code><nobr>ebp</nobr></code> and it assumes that a near form of call -was used to get to this label (i.e. that <code><nobr>eip</nobr></code> is -on the stack). -<p><pre> -%stacksize flat64 -</pre> -<p>This form causes NASM to use stack-based parameter addressing relative -to <code><nobr>rbp</nobr></code> and it assumes that a near form of call -was used to get to this label (i.e. that <code><nobr>rip</nobr></code> is -on the stack). -<p><pre> -%stacksize large -</pre> -<p>This form uses <code><nobr>bp</nobr></code> to do stack-based parameter -addressing and assumes that a far form of call was used to get to this -address (i.e. that <code><nobr>ip</nobr></code> and -<code><nobr>cs</nobr></code> are on the stack). -<p><pre> -%stacksize small -</pre> -<p>This form also uses <code><nobr>bp</nobr></code> to address stack -parameters, but it is different from <code><nobr>large</nobr></code> -because it also assumes that the old value of bp is pushed onto the stack -(i.e. it expects an <code><nobr>ENTER</nobr></code> instruction). In other -words, it expects that <code><nobr>bp</nobr></code>, -<code><nobr>ip</nobr></code> and <code><nobr>cs</nobr></code> are on the -top of the stack, underneath any local space which may have been allocated -by <code><nobr>ENTER</nobr></code>. This form is probably most useful when -used in combination with the <code><nobr>%local</nobr></code> directive -(see <a href="#section-4.8.3">section 4.8.3</a>). -<h4><a name="section-4.8.3">4.8.3 <code><nobr>%local</nobr></code> Directive</a></h4> -<p>The <code><nobr>%local</nobr></code> directive is used to simplify the -use of local temporary stack variables allocated in a stack frame. -Automatic local variables in C are an example of this kind of variable. The -<code><nobr>%local</nobr></code> directive is most useful when used with -the <code><nobr>%stacksize</nobr></code> (see -<a href="#section-4.8.2">section 4.8.2</a> and is also compatible with the -<code><nobr>%arg</nobr></code> directive (see -<a href="#section-4.8.1">section 4.8.1</a>). It allows simplified reference -to variables on the stack which have been allocated typically by using the -<code><nobr>ENTER</nobr></code> instruction. An example of its use is the -following: -<p><pre> -silly_swap: - - %push mycontext ; save the current context - %stacksize small ; tell NASM to use bp - %assign %$localsize 0 ; see text for explanation - %local old_ax:word, old_dx:word - - enter %$localsize,0 ; see text for explanation - mov [old_ax],ax ; swap ax & bx - mov [old_dx],dx ; and swap dx & cx - mov ax,bx - mov dx,cx - mov bx,[old_ax] - mov cx,[old_dx] - leave ; restore old bp - ret ; - - %pop ; restore original context -</pre> -<p>The <code><nobr>%$localsize</nobr></code> variable is used internally by -the <code><nobr>%local</nobr></code> directive and <em>must</em> be defined -within the current context before the <code><nobr>%local</nobr></code> -directive may be used. Failure to do so will result in one expression -syntax error for each <code><nobr>%local</nobr></code> variable declared. -It then may be used in the construction of an appropriately sized ENTER -instruction as shown in the example. -<h3><a name="section-4.9">4.9 Reporting User-Defined Errors: <code><nobr>%error</nobr></code>, <code><nobr>%warning</nobr></code>, <code><nobr>%fatal</nobr></code></a></h3> -<p>The preprocessor directive <code><nobr>%error</nobr></code> will cause -NASM to report an error if it occurs in assembled code. So if other users -are going to try to assemble your source files, you can ensure that they -define the right macros by means of code like this: -<p><pre> -%ifdef F1 - ; do some setup -%elifdef F2 - ; do some different setup -%else - %error "Neither F1 nor F2 was defined." -%endif -</pre> -<p>Then any user who fails to understand the way your code is supposed to -be assembled will be quickly warned of their mistake, rather than having to -wait until the program crashes on being run and then not knowing what went -wrong. -<p>Similarly, <code><nobr>%warning</nobr></code> issues a warning, but -allows assembly to continue: -<p><pre> -%ifdef F1 - ; do some setup -%elifdef F2 - ; do some different setup -%else - %warning "Neither F1 nor F2 was defined, assuming F1." - %define F1 -%endif -</pre> -<p><code><nobr>%error</nobr></code> and <code><nobr>%warning</nobr></code> -are issued only on the final assembly pass. This makes them safe to use in -conjunction with tests that depend on symbol values. -<p><code><nobr>%fatal</nobr></code> terminates assembly immediately, -regardless of pass. This is useful when there is no point in continuing the -assembly further, and doing so is likely just going to cause a spew of -confusing error messages. -<p>It is optional for the message string after -<code><nobr>%error</nobr></code>, <code><nobr>%warning</nobr></code> or -<code><nobr>%fatal</nobr></code> to be quoted. If it is <em>not</em>, then -single-line macros are expanded in it, which can be used to display more -information to the user. For example: -<p><pre> -%if foo > 64 - %assign foo_over foo-64 - %error foo is foo_over bytes too large -%endif -</pre> -<h3><a name="section-4.10">4.10 Other Preprocessor Directives</a></h3> -<p>NASM also has preprocessor directives which allow access to information -from external sources. Currently they include: -<ul> -<li><code><nobr>%line</nobr></code> enables NASM to correctly handle the -output of another preprocessor (see <a href="#section-4.10.1">section -4.10.1</a>). -<li><code><nobr>%!</nobr></code> enables NASM to read in the value of an -environment variable, which can then be used in your program (see -<a href="#section-4.10.2">section 4.10.2</a>). -</ul> -<h4><a name="section-4.10.1">4.10.1 <code><nobr>%line</nobr></code> Directive</a></h4> -<p>The <code><nobr>%line</nobr></code> directive is used to notify NASM -that the input line corresponds to a specific line number in another file. -Typically this other file would be an original source file, with the -current NASM input being the output of a pre-processor. The -<code><nobr>%line</nobr></code> directive allows NASM to output messages -which indicate the line number of the original source file, instead of the -file that is being read by NASM. -<p>This preprocessor directive is not generally of use to programmers, by -may be of interest to preprocessor authors. The usage of the -<code><nobr>%line</nobr></code> preprocessor directive is as follows: -<p><pre> -%line nnn[+mmm] [filename] -</pre> -<p>In this directive, <code><nobr>nnn</nobr></code> identifies the line of -the original source file which this line corresponds to. -<code><nobr>mmm</nobr></code> is an optional parameter which specifies a -line increment value; each line of the input file read in is considered to -correspond to <code><nobr>mmm</nobr></code> lines of the original source -file. Finally, <code><nobr>filename</nobr></code> is an optional parameter -which specifies the file name of the original source file. -<p>After reading a <code><nobr>%line</nobr></code> preprocessor directive, -NASM will report all file name and line numbers relative to the values -specified therein. -<h4><a name="section-4.10.2">4.10.2 <code><nobr>%!</nobr></code><code><nobr><env></nobr></code>: Read an environment variable.</a></h4> -<p>The <code><nobr>%!<env></nobr></code> directive makes it possible -to read the value of an environment variable at assembly time. This could, -for example, be used to store the contents of an environment variable into -a string, which could be used at some other point in your code. -<p>For example, suppose that you have an environment variable -<code><nobr>FOO</nobr></code>, and you want the contents of -<code><nobr>FOO</nobr></code> to be embedded in your program. You could do -that as follows: -<p><pre> -%defstr FOO %!FOO -</pre> -<p>See <a href="#section-4.1.8">section 4.1.8</a> for notes on the -<code><nobr>%defstr</nobr></code> directive. -<h3><a name="section-4.11">4.11 Standard Macros</a></h3> -<p>NASM defines a set of standard macros, which are already defined when it -starts to process any source file. If you really need a program to be -assembled with no pre-defined macros, you can use the -<code><nobr>%clear</nobr></code> directive to empty the preprocessor of -everything but context-local preprocessor variables and single-line macros. -<p>Most user-level assembler directives (see -<a href="nasmdoc6.html">chapter 6</a>) are implemented as macros which -invoke primitive directives; these are described in -<a href="nasmdoc6.html">chapter 6</a>. The rest of the standard macro set -is described here. -<h4><a name="section-4.11.1">4.11.1 NASM Version Macros</a></h4> -<p>The single-line macros <code><nobr>__NASM_MAJOR__</nobr></code>, -<code><nobr>__NASM_MINOR__</nobr></code>, -<code><nobr>__NASM_SUBMINOR__</nobr></code> and -<code><nobr>___NASM_PATCHLEVEL__</nobr></code> expand to the major, minor, -subminor and patch level parts of the version number of NASM being used. -So, under NASM 0.98.32p1 for example, -<code><nobr>__NASM_MAJOR__</nobr></code> would be defined to be 0, -<code><nobr>__NASM_MINOR__</nobr></code> would be defined as 98, -<code><nobr>__NASM_SUBMINOR__</nobr></code> would be defined to 32, and -<code><nobr>___NASM_PATCHLEVEL__</nobr></code> would be defined as 1. -<p>Additionally, the macro <code><nobr>__NASM_SNAPSHOT__</nobr></code> is -defined for automatically generated snapshot releases <em>only</em>. -<h4><a name="section-4.11.2">4.11.2 <code><nobr>__NASM_VERSION_ID__</nobr></code>: NASM Version ID</a></h4> -<p>The single-line macro <code><nobr>__NASM_VERSION_ID__</nobr></code> -expands to a dword integer representing the full version number of the -version of nasm being used. The value is the equivalent to -<code><nobr>__NASM_MAJOR__</nobr></code>, -<code><nobr>__NASM_MINOR__</nobr></code>, -<code><nobr>__NASM_SUBMINOR__</nobr></code> and -<code><nobr>___NASM_PATCHLEVEL__</nobr></code> concatenated to produce a -single doubleword. Hence, for 0.98.32p1, the returned number would be -equivalent to: -<p><pre> - dd 0x00622001 -</pre> -<p>or -<p><pre> - db 1,32,98,0 -</pre> -<p>Note that the above lines are generate exactly the same code, the second -line is used just to give an indication of the order that the separate -values will be present in memory. -<h4><a name="section-4.11.3">4.11.3 <code><nobr>__NASM_VER__</nobr></code>: NASM Version string</a></h4> -<p>The single-line macro <code><nobr>__NASM_VER__</nobr></code> expands to -a string which defines the version number of nasm being used. So, under -NASM 0.98.32 for example, -<p><pre> - db __NASM_VER__ -</pre> -<p>would expand to -<p><pre> - db "0.98.32" -</pre> -<h4><a name="section-4.11.4">4.11.4 <code><nobr>__FILE__</nobr></code> and <code><nobr>__LINE__</nobr></code>: File Name and Line Number</a></h4> -<p>Like the C preprocessor, NASM allows the user to find out the file name -and line number containing the current instruction. The macro -<code><nobr>__FILE__</nobr></code> expands to a string constant giving the -name of the current input file (which may change through the course of -assembly if <code><nobr>%include</nobr></code> directives are used), and -<code><nobr>__LINE__</nobr></code> expands to a numeric constant giving the -current line number in the input file. -<p>These macros could be used, for example, to communicate debugging -information to a macro, since invoking <code><nobr>__LINE__</nobr></code> -inside a macro definition (either single-line or multi-line) will return -the line number of the macro <em>call</em>, rather than -<em>definition</em>. So to determine where in a piece of code a crash is -occurring, for example, one could write a routine -<code><nobr>stillhere</nobr></code>, which is passed a line number in -<code><nobr>EAX</nobr></code> and outputs something like `line 155: still -here'. You could then write a macro -<p><pre> -%macro notdeadyet 0 - - push eax - mov eax,__LINE__ - call stillhere - pop eax - -%endmacro -</pre> -<p>and then pepper your code with calls to -<code><nobr>notdeadyet</nobr></code> until you find the crash point. -<h4><a name="section-4.11.5">4.11.5 <code><nobr>__BITS__</nobr></code>: Current BITS Mode</a></h4> -<p>The <code><nobr>__BITS__</nobr></code> standard macro is updated every -time that the BITS mode is set using the <code><nobr>BITS XX</nobr></code> -or <code><nobr>[BITS XX]</nobr></code> directive, where XX is a valid mode -number of 16, 32 or 64. <code><nobr>__BITS__</nobr></code> receives the -specified mode number and makes it globally available. This can be very -useful for those who utilize mode-dependent macros. -<h4><a name="section-4.11.6">4.11.6 <code><nobr>__OUTPUT_FORMAT__</nobr></code>: Current Output Format</a></h4> -<p>The <code><nobr>__OUTPUT_FORMAT__</nobr></code> standard macro holds the -current Output Format, as given by the <code><nobr>-f</nobr></code> option -or NASM's default. Type <code><nobr>nasm -hf</nobr></code> for a list. -<p><pre> -%ifidn __OUTPUT_FORMAT__, win32 - %define NEWLINE 13, 10 -%elifidn __OUTPUT_FORMAT__, elf32 - %define NEWLINE 10 -%endif -</pre> -<h4><a name="section-4.11.7">4.11.7 Assembly Date and Time Macros</a></h4> -<p>NASM provides a variety of macros that represent the timestamp of the -assembly session. -<ul> -<li>The <code><nobr>__DATE__</nobr></code> and -<code><nobr>__TIME__</nobr></code> macros give the assembly date and time -as strings, in ISO 8601 format (<code><nobr>"YYYY-MM-DD"</nobr></code> and -<code><nobr>"HH:MM:SS"</nobr></code>, respectively.) -<li>The <code><nobr>__DATE_NUM__</nobr></code> and -<code><nobr>__TIME_NUM__</nobr></code> macros give the assembly date and -time in numeric form; in the format <code><nobr>YYYYMMDD</nobr></code> and -<code><nobr>HHMMSS</nobr></code> respectively. -<li>The <code><nobr>__UTC_DATE__</nobr></code> and -<code><nobr>__UTC_TIME__</nobr></code> macros give the assembly date and -time in universal time (UTC) as strings, in ISO 8601 format -(<code><nobr>"YYYY-MM-DD"</nobr></code> and -<code><nobr>"HH:MM:SS"</nobr></code>, respectively.) If the host platform -doesn't provide UTC time, these macros are undefined. -<li>The <code><nobr>__UTC_DATE_NUM__</nobr></code> and -<code><nobr>__UTC_TIME_NUM__</nobr></code> macros give the assembly date -and time universal time (UTC) in numeric form; in the format -<code><nobr>YYYYMMDD</nobr></code> and <code><nobr>HHMMSS</nobr></code> -respectively. If the host platform doesn't provide UTC time, these macros -are undefined. -<li>The <code><nobr>__POSIX_TIME__</nobr></code> macro is defined as a -number containing the number of seconds since the POSIX epoch, 1 January -1970 00:00:00 UTC; excluding any leap seconds. This is computed using UTC -time if available on the host platform, otherwise it is computed using the -local time as if it was UTC. -</ul> -<p>All instances of time and date macros in the same assembly session -produce consistent output. For example, in an assembly session started at -42 seconds after midnight on January 1, 2010 in Moscow (timezone UTC+3) -these macros would have the following values, assuming, of course, a -properly configured environment with a correct clock: -<p><pre> - __DATE__ "2010-01-01" - __TIME__ "00:00:42" - __DATE_NUM__ 20100101 - __TIME_NUM__ 000042 - __UTC_DATE__ "2009-12-31" - __UTC_TIME__ "21:00:42" - __UTC_DATE_NUM__ 20091231 - __UTC_TIME_NUM__ 210042 - __POSIX_TIME__ 1262293242 -</pre> -<h4><a name="section-4.11.8">4.11.8 <code><nobr>__USE_</nobr></code><em>package</em><code><nobr>__</nobr></code>: Package Include Test</a></h4> -<p>When a standard macro package (see <a href="nasmdoc5.html">chapter -5</a>) is included with the <code><nobr>%use</nobr></code> directive (see -<a href="#section-4.6.4">section 4.6.4</a>), a single-line macro of the -form -<code><nobr>__USE_</nobr></code><em>package</em><code><nobr>__</nobr></code> -is automatically defined. This allows testing if a particular package is -invoked or not. -<p>For example, if the <code><nobr>altreg</nobr></code> package is included -(see <a href="nasmdoc5.html#section-5.1">section 5.1</a>), then the macro -<code><nobr>__USE_ALTREG__</nobr></code> is defined. -<h4><a name="section-4.11.9">4.11.9 <code><nobr>__PASS__</nobr></code>: Assembly Pass</a></h4> -<p>The macro <code><nobr>__PASS__</nobr></code> is defined to be -<code><nobr>1</nobr></code> on preparatory passes, and -<code><nobr>2</nobr></code> on the final pass. In preprocess-only mode, it -is set to <code><nobr>3</nobr></code>, and when running only to generate -dependencies (due to the <code><nobr>-M</nobr></code> or -<code><nobr>-MG</nobr></code> option, see -<a href="nasmdoc2.html#section-2.1.4">section 2.1.4</a>) it is set to -<code><nobr>0</nobr></code>. -<p><em>Avoid using this macro if at all possible. It is tremendously easy -to generate very strange errors by misusing it, and the semantics may -change in future versions of NASM.</em> -<h4><a name="section-4.11.10">4.11.10 <code><nobr>STRUC</nobr></code> and <code><nobr>ENDSTRUC</nobr></code>: Declaring Structure Data Types</a></h4> -<p>The core of NASM contains no intrinsic means of defining data -structures; instead, the preprocessor is sufficiently powerful that data -structures can be implemented as a set of macros. The macros -<code><nobr>STRUC</nobr></code> and <code><nobr>ENDSTRUC</nobr></code> are -used to define a structure data type. -<p><code><nobr>STRUC</nobr></code> takes one or two parameters. The first -parameter is the name of the data type. The second, optional parameter is -the base offset of the structure. The name of the data type is defined as a -symbol with the value of the base offset, and the name of the data type -with the suffix <code><nobr>_size</nobr></code> appended to it is defined -as an <code><nobr>EQU</nobr></code> giving the size of the structure. Once -<code><nobr>STRUC</nobr></code> has been issued, you are defining the -structure, and should define fields using the -<code><nobr>RESB</nobr></code> family of pseudo-instructions, and then -invoke <code><nobr>ENDSTRUC</nobr></code> to finish the definition. -<p>For example, to define a structure called -<code><nobr>mytype</nobr></code> containing a longword, a word, a byte and -a string of bytes, you might code -<p><pre> -struc mytype - - mt_long: resd 1 - mt_word: resw 1 - mt_byte: resb 1 - mt_str: resb 32 - -endstruc -</pre> -<p>The above code defines six symbols: <code><nobr>mt_long</nobr></code> as -0 (the offset from the beginning of a <code><nobr>mytype</nobr></code> -structure to the longword field), <code><nobr>mt_word</nobr></code> as 4, -<code><nobr>mt_byte</nobr></code> as 6, <code><nobr>mt_str</nobr></code> as -7, <code><nobr>mytype_size</nobr></code> as 39, and -<code><nobr>mytype</nobr></code> itself as zero. -<p>The reason why the structure type name is defined at zero by default is -a side effect of allowing structures to work with the local label -mechanism: if your structure members tend to have the same names in more -than one structure, you can define the above structure like this: -<p><pre> -struc mytype - - .long: resd 1 - .word: resw 1 - .byte: resb 1 - .str: resb 32 - -endstruc -</pre> -<p>This defines the offsets to the structure fields as -<code><nobr>mytype.long</nobr></code>, -<code><nobr>mytype.word</nobr></code>, -<code><nobr>mytype.byte</nobr></code> and -<code><nobr>mytype.str</nobr></code>. -<p>NASM, since it has no <em>intrinsic</em> structure support, does not -support any form of period notation to refer to the elements of a structure -once you have one (except the above local-label notation), so code such as -<code><nobr>mov ax,[mystruc.mt_word]</nobr></code> is not valid. -<code><nobr>mt_word</nobr></code> is a constant just like any other -constant, so the correct syntax is -<code><nobr>mov ax,[mystruc+mt_word]</nobr></code> or -<code><nobr>mov ax,[mystruc+mytype.word]</nobr></code>. -<p>Sometimes you only have the address of the structure displaced by an -offset. For example, consider this standard stack frame setup: -<p><pre> -push ebp -mov ebp, esp -sub esp, 40 -</pre> -<p>In this case, you could access an element by subtracting the offset: -<p><pre> -mov [ebp - 40 + mytype.word], ax -</pre> -<p>However, if you do not want to repeat this offset, you can use -40 as a -base offset: -<p><pre> -struc mytype, -40 -</pre> -<p>And access an element this way: -<p><pre> -mov [ebp + mytype.word], ax -</pre> -<h4><a name="section-4.11.11">4.11.11 <code><nobr>ISTRUC</nobr></code>, <code><nobr>AT</nobr></code> and <code><nobr>IEND</nobr></code>: Declaring Instances of Structures</a></h4> -<p>Having defined a structure type, the next thing you typically want to do -is to declare instances of that structure in your data segment. NASM -provides an easy way to do this in the <code><nobr>ISTRUC</nobr></code> -mechanism. To declare a structure of type <code><nobr>mytype</nobr></code> -in a program, you code something like this: -<p><pre> -mystruc: - istruc mytype - - at mt_long, dd 123456 - at mt_word, dw 1024 - at mt_byte, db 'x' - at mt_str, db 'hello, world', 13, 10, 0 - - iend -</pre> -<p>The function of the <code><nobr>AT</nobr></code> macro is to make use of -the <code><nobr>TIMES</nobr></code> prefix to advance the assembly position -to the correct point for the specified structure field, and then to declare -the specified data. Therefore the structure fields must be declared in the -same order as they were specified in the structure definition. -<p>If the data to go in a structure field requires more than one source -line to specify, the remaining source lines can easily come after the -<code><nobr>AT</nobr></code> line. For example: -<p><pre> - at mt_str, db 123,134,145,156,167,178,189 - db 190,100,0 -</pre> -<p>Depending on personal taste, you can also omit the code part of the -<code><nobr>AT</nobr></code> line completely, and start the structure field -on the next line: -<p><pre> - at mt_str - db 'hello, world' - db 13,10,0 -</pre> -<h4><a name="section-4.11.12">4.11.12 <code><nobr>ALIGN</nobr></code> and <code><nobr>ALIGNB</nobr></code>: Data Alignment</a></h4> -<p>The <code><nobr>ALIGN</nobr></code> and <code><nobr>ALIGNB</nobr></code> -macros provides a convenient way to align code or data on a word, longword, -paragraph or other boundary. (Some assemblers call this directive -<code><nobr>EVEN</nobr></code>.) The syntax of the -<code><nobr>ALIGN</nobr></code> and <code><nobr>ALIGNB</nobr></code> macros -is -<p><pre> - align 4 ; align on 4-byte boundary - align 16 ; align on 16-byte boundary - align 8,db 0 ; pad with 0s rather than NOPs - align 4,resb 1 ; align to 4 in the BSS - alignb 4 ; equivalent to previous line -</pre> -<p>Both macros require their first argument to be a power of two; they both -compute the number of additional bytes required to bring the length of the -current section up to a multiple of that power of two, and then apply the -<code><nobr>TIMES</nobr></code> prefix to their second argument to perform -the alignment. -<p>If the second argument is not specified, the default for -<code><nobr>ALIGN</nobr></code> is <code><nobr>NOP</nobr></code>, and the -default for <code><nobr>ALIGNB</nobr></code> is -<code><nobr>RESB 1</nobr></code>. So if the second argument is specified, -the two macros are equivalent. Normally, you can just use -<code><nobr>ALIGN</nobr></code> in code and data sections and -<code><nobr>ALIGNB</nobr></code> in BSS sections, and never need the second -argument except for special purposes. -<p><code><nobr>ALIGN</nobr></code> and <code><nobr>ALIGNB</nobr></code>, -being simple macros, perform no error checking: they cannot warn you if -their first argument fails to be a power of two, or if their second -argument generates more than one byte of code. In each of these cases they -will silently do the wrong thing. -<p><code><nobr>ALIGNB</nobr></code> (or <code><nobr>ALIGN</nobr></code> -with a second argument of <code><nobr>RESB 1</nobr></code>) can be used -within structure definitions: -<p><pre> -struc mytype2 - - mt_byte: - resb 1 - alignb 2 - mt_word: - resw 1 - alignb 4 - mt_long: - resd 1 - mt_str: - resb 32 - -endstruc -</pre> -<p>This will ensure that the structure members are sensibly aligned -relative to the base of the structure. -<p>A final caveat: <code><nobr>ALIGN</nobr></code> and -<code><nobr>ALIGNB</nobr></code> work relative to the beginning of the -<em>section</em>, not the beginning of the address space in the final -executable. Aligning to a 16-byte boundary when the section you're in is -only guaranteed to be aligned to a 4-byte boundary, for example, is a waste -of effort. Again, NASM does not check that the section's alignment -characteristics are sensible for the use of <code><nobr>ALIGN</nobr></code> -or <code><nobr>ALIGNB</nobr></code>. -<p>See also the <code><nobr>smartalign</nobr></code> standard macro -package, <a href="nasmdoc5.html#section-5.2">section 5.2</a>. -<p align=center><a href="nasmdoc5.html">Next Chapter</a> | -<a href="nasmdoc3.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -</body></html> diff --git a/doc/html/nasmdoc5.html b/doc/html/nasmdoc5.html deleted file mode 100644 index 7effbbb..0000000 --- a/doc/html/nasmdoc5.html +++ /dev/null @@ -1,83 +0,0 @@ -<html><head><title>NASM Manual</title></head> -<body><h1 align=center>The Netwide Assembler: NASM</h1> - -<p align=center><a href="nasmdoc6.html">Next Chapter</a> | -<a href="nasmdoc4.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -<h2><a name="chapter-5">Chapter 5: Standard Macro Packages</a></h2> -<p>The <code><nobr>%use</nobr></code> directive (see -<a href="nasmdoc4.html#section-4.6.4">section 4.6.4</a>) includes one of -the standard macro packages included with the NASM distribution and -compiled into the NASM binary. It operates like the -<code><nobr>%include</nobr></code> directive (see -<a href="nasmdoc4.html#section-4.6.1">section 4.6.1</a>), but the included -contents is provided by NASM itself. -<p>The names of standard macro packages are case insensitive, and can be -quoted or not. -<h3><a name="section-5.1">5.1 <code><nobr>altreg</nobr></code>: Alternate Register Names</a></h3> -<p>The <code><nobr>altreg</nobr></code> standard macro package provides -alternate register names. It provides numeric register names for all -registers (not just -<code><nobr>R8</nobr></code>-<code><nobr>R15</nobr></code>), the -Intel-defined aliases -<code><nobr>R8L</nobr></code>-<code><nobr>R15L</nobr></code> for the low -bytes of register (as opposed to the NASM/AMD standard names -<code><nobr>R8B</nobr></code>-<code><nobr>R15B</nobr></code>), and the -names <code><nobr>R0H</nobr></code>-<code><nobr>R3H</nobr></code> (by -analogy with <code><nobr>R0L</nobr></code>-<code><nobr>R3L</nobr></code>) -for <code><nobr>AH</nobr></code>, <code><nobr>CH</nobr></code>, -<code><nobr>DH</nobr></code>, and <code><nobr>BH</nobr></code>. -<p>Example use: -<p><pre> -%use altreg - -proc: - mov r0l,r3h ; mov al,bh - ret -</pre> -<p>See also <a href="nasmdo11.html#section-11.1">section 11.1</a>. -<h3><a name="section-5.2">5.2 <code><nobr>smartalign</nobr></code>: Smart <code><nobr>ALIGN</nobr></code> Macro</a></h3> -<p>The <code><nobr>smartalign</nobr></code> standard macro package provides -for an <code><nobr>ALIGN</nobr></code> macro which is more powerful than -the default (and backwards-compatible) one (see -<a href="nasmdoc4.html#section-4.11.12">section 4.11.12</a>). When the -<code><nobr>smartalign</nobr></code> package is enabled, when -<code><nobr>ALIGN</nobr></code> is used without a second argument, NASM -will generate a sequence of instructions more efficient than a series of -<code><nobr>NOP</nobr></code>. Furthermore, if the padding exceeds a -specific threshold, then NASM will generate a jump over the entire padding -sequence. -<p>The specific instructions generated can be controlled with the new -<code><nobr>ALIGNMODE</nobr></code> macro. This macro takes two parameters: -one mode, and an optional jump threshold override. The modes are as -follows: -<ul> -<li><code><nobr>generic</nobr></code>: Works on all x86 CPUs and should -have reasonable performance. The default jump threshold is 8. This is the -default. -<li><code><nobr>nop</nobr></code>: Pad out with -<code><nobr>NOP</nobr></code> instructions. The only difference compared to -the standard <code><nobr>ALIGN</nobr></code> macro is that NASM can still -jump over a large padding area. The default jump threshold is 16. -<li><code><nobr>k7</nobr></code>: Optimize for the AMD K7 (Athlon/Althon -XP). These instructions should still work on all x86 CPUs. The default jump -threshold is 16. -<li><code><nobr>k8</nobr></code>: Optimize for the AMD K8 (Opteron/Althon -64). These instructions should still work on all x86 CPUs. The default jump -threshold is 16. -<li><code><nobr>p6</nobr></code>: Optimize for Intel CPUs. This uses the -long <code><nobr>NOP</nobr></code> instructions first introduced in Pentium -Pro. This is incompatible with all CPUs of family 5 or lower, as well as -some VIA CPUs and several virtualization solutions. The default jump -threshold is 16. -</ul> -<p>The macro <code><nobr>__ALIGNMODE__</nobr></code> is defined to contain -the current alignment mode. A number of other macros beginning with -<code><nobr>__ALIGN_</nobr></code> are used internally by this macro -package. -<p align=center><a href="nasmdoc6.html">Next Chapter</a> | -<a href="nasmdoc4.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -</body></html> diff --git a/doc/html/nasmdoc6.html b/doc/html/nasmdoc6.html deleted file mode 100644 index 144a661..0000000 --- a/doc/html/nasmdoc6.html +++ /dev/null @@ -1,371 +0,0 @@ -<html><head><title>NASM Manual</title></head> -<body><h1 align=center>The Netwide Assembler: NASM</h1> - -<p align=center><a href="nasmdoc7.html">Next Chapter</a> | -<a href="nasmdoc5.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -<h2><a name="chapter-6">Chapter 6: Assembler Directives</a></h2> -<p>NASM, though it attempts to avoid the bureaucracy of assemblers like -MASM and TASM, is nevertheless forced to support a <em>few</em> directives. -These are described in this chapter. -<p>NASM's directives come in two types: <em>user-level</em> directives and -<em>primitive</em> directives. Typically, each directive has a user-level -form and a primitive form. In almost all cases, we recommend that users use -the user-level forms of the directives, which are implemented as macros -which call the primitive forms. -<p>Primitive directives are enclosed in square brackets; user-level -directives are not. -<p>In addition to the universal directives described in this chapter, each -object file format can optionally supply extra directives in order to -control particular features of that file format. These -<em>format-specific</em> directives are documented along with the formats -that implement them, in <a href="nasmdoc7.html">chapter 7</a>. -<h3><a name="section-6.1">6.1 <code><nobr>BITS</nobr></code>: Specifying Target Processor Mode</a></h3> -<p>The <code><nobr>BITS</nobr></code> directive specifies whether NASM -should generate code designed to run on a processor operating in 16-bit -mode, 32-bit mode or 64-bit mode. The syntax is -<code><nobr>BITS XX</nobr></code>, where XX is 16, 32 or 64. -<p>In most cases, you should not need to use <code><nobr>BITS</nobr></code> -explicitly. The <code><nobr>aout</nobr></code>, -<code><nobr>coff</nobr></code>, <code><nobr>elf</nobr></code>, -<code><nobr>macho</nobr></code>, <code><nobr>win32</nobr></code> and -<code><nobr>win64</nobr></code> object formats, which are designed for use -in 32-bit or 64-bit operating systems, all cause NASM to select 32-bit or -64-bit mode, respectively, by default. The <code><nobr>obj</nobr></code> -object format allows you to specify each segment you define as either -<code><nobr>USE16</nobr></code> or <code><nobr>USE32</nobr></code>, and -NASM will set its operating mode accordingly, so the use of the -<code><nobr>BITS</nobr></code> directive is once again unnecessary. -<p>The most likely reason for using the <code><nobr>BITS</nobr></code> -directive is to write 32-bit or 64-bit code in a flat binary file; this is -because the <code><nobr>bin</nobr></code> output format defaults to 16-bit -mode in anticipation of it being used most frequently to write DOS -<code><nobr>.COM</nobr></code> programs, DOS <code><nobr>.SYS</nobr></code> -device drivers and boot loader software. -<p>You do <em>not</em> need to specify <code><nobr>BITS 32</nobr></code> -merely in order to use 32-bit instructions in a 16-bit DOS program; if you -do, the assembler will generate incorrect code because it will be writing -code targeted at a 32-bit platform, to be run on a 16-bit one. -<p>When NASM is in <code><nobr>BITS 16</nobr></code> mode, instructions -which use 32-bit data are prefixed with an 0x66 byte, and those referring -to 32-bit addresses have an 0x67 prefix. In -<code><nobr>BITS 32</nobr></code> mode, the reverse is true: 32-bit -instructions require no prefixes, whereas instructions using 16-bit data -need an 0x66 and those working on 16-bit addresses need an 0x67. -<p>When NASM is in <code><nobr>BITS 64</nobr></code> mode, most -instructions operate the same as they do for -<code><nobr>BITS 32</nobr></code> mode. However, there are 8 more general -and SSE registers, and 16-bit addressing is no longer supported. -<p>The default address size is 64 bits; 32-bit addressing can be selected -with the 0x67 prefix. The default operand size is still 32 bits, however, -and the 0x66 prefix selects 16-bit operand size. The -<code><nobr>REX</nobr></code> prefix is used both to select 64-bit operand -size, and to access the new registers. NASM automatically inserts REX -prefixes when necessary. -<p>When the <code><nobr>REX</nobr></code> prefix is used, the processor -does not know how to address the AH, BH, CH or DH (high 8-bit legacy) -registers. Instead, it is possible to access the the low 8-bits of the SP, -BP SI and DI registers as SPL, BPL, SIL and DIL, respectively; but only -when the REX prefix is used. -<p>The <code><nobr>BITS</nobr></code> directive has an exactly equivalent -primitive form, <code><nobr>[BITS 16]</nobr></code>, -<code><nobr>[BITS 32]</nobr></code> and -<code><nobr>[BITS 64]</nobr></code>. The user-level form is a macro which -has no function other than to call the primitive form. -<p>Note that the space is neccessary, e.g. <code><nobr>BITS32</nobr></code> -will <em>not</em> work! -<h4><a name="section-6.1.1">6.1.1 <code><nobr>USE16</nobr></code> & <code><nobr>USE32</nobr></code>: Aliases for BITS</a></h4> -<p>The `<code><nobr>USE16</nobr></code>' and -`<code><nobr>USE32</nobr></code>' directives can be used in place of -`<code><nobr>BITS 16</nobr></code>' and -`<code><nobr>BITS 32</nobr></code>', for compatibility with other -assemblers. -<h3><a name="section-6.2">6.2 <code><nobr>DEFAULT</nobr></code>: Change the assembler defaults</a></h3> -<p>The <code><nobr>DEFAULT</nobr></code> directive changes the assembler -defaults. Normally, NASM defaults to a mode where the programmer is -expected to explicitly specify most features directly. However, this is -occationally obnoxious, as the explicit form is pretty much the only one -one wishes to use. -<p>Currently, the only <code><nobr>DEFAULT</nobr></code> that is settable -is whether or not registerless instructions in 64-bit mode are -<code><nobr>RIP</nobr></code>-relative or not. By default, they are -absolute unless overridden with the <code><nobr>REL</nobr></code> specifier -(see <a href="nasmdoc3.html#section-3.3">section 3.3</a>). However, if -<code><nobr>DEFAULT REL</nobr></code> is specified, -<code><nobr>REL</nobr></code> is default, unless overridden with the -<code><nobr>ABS</nobr></code> specifier, <em>except when used with an FS or -GS segment override</em>. -<p>The special handling of <code><nobr>FS</nobr></code> and -<code><nobr>GS</nobr></code> overrides are due to the fact that these -registers are generally used as thread pointers or other special functions -in 64-bit mode, and generating <code><nobr>RIP</nobr></code>-relative -addresses would be extremely confusing. -<p><code><nobr>DEFAULT REL</nobr></code> is disabled with -<code><nobr>DEFAULT ABS</nobr></code>. -<h3><a name="section-6.3">6.3 <code><nobr>SECTION</nobr></code> or <code><nobr>SEGMENT</nobr></code>: Changing and Defining Sections</a></h3> -<p>The <code><nobr>SECTION</nobr></code> directive -(<code><nobr>SEGMENT</nobr></code> is an exactly equivalent synonym) -changes which section of the output file the code you write will be -assembled into. In some object file formats, the number and names of -sections are fixed; in others, the user may make up as many as they wish. -Hence <code><nobr>SECTION</nobr></code> may sometimes give an error -message, or may define a new section, if you try to switch to a section -that does not (yet) exist. -<p>The Unix object formats, and the <code><nobr>bin</nobr></code> object -format (but see <a href="nasmdoc7.html#section-7.1.3">section 7.1.3</a>, -all support the standardized section names <code><nobr>.text</nobr></code>, -<code><nobr>.data</nobr></code> and <code><nobr>.bss</nobr></code> for the -code, data and uninitialized-data sections. The -<code><nobr>obj</nobr></code> format, by contrast, does not recognize these -section names as being special, and indeed will strip off the leading -period of any section name that has one. -<h4><a name="section-6.3.1">6.3.1 The <code><nobr>__SECT__</nobr></code> Macro</a></h4> -<p>The <code><nobr>SECTION</nobr></code> directive is unusual in that its -user-level form functions differently from its primitive form. The -primitive form, <code><nobr>[SECTION xyz]</nobr></code>, simply switches -the current target section to the one given. The user-level form, -<code><nobr>SECTION xyz</nobr></code>, however, first defines the -single-line macro <code><nobr>__SECT__</nobr></code> to be the primitive -<code><nobr>[SECTION]</nobr></code> directive which it is about to issue, -and then issues it. So the user-level directive -<p><pre> - SECTION .text -</pre> -<p>expands to the two lines -<p><pre> -%define __SECT__ [SECTION .text] - [SECTION .text] -</pre> -<p>Users may find it useful to make use of this in their own macros. For -example, the <code><nobr>writefile</nobr></code> macro defined in -<a href="nasmdoc4.html#section-4.3.4">section 4.3.4</a> can be usefully -rewritten in the following more sophisticated form: -<p><pre> -%macro writefile 2+ - - [section .data] - - %%str: db %2 - %%endstr: - - __SECT__ - - mov dx,%%str - mov cx,%%endstr-%%str - mov bx,%1 - mov ah,0x40 - int 0x21 - -%endmacro -</pre> -<p>This form of the macro, once passed a string to output, first switches -temporarily to the data section of the file, using the primitive form of -the <code><nobr>SECTION</nobr></code> directive so as not to modify -<code><nobr>__SECT__</nobr></code>. It then declares its string in the data -section, and then invokes <code><nobr>__SECT__</nobr></code> to switch back -to <em>whichever</em> section the user was previously working in. It thus -avoids the need, in the previous version of the macro, to include a -<code><nobr>JMP</nobr></code> instruction to jump over the data, and also -does not fail if, in a complicated <code><nobr>OBJ</nobr></code> format -module, the user could potentially be assembling the code in any of several -separate code sections. -<h3><a name="section-6.4">6.4 <code><nobr>ABSOLUTE</nobr></code>: Defining Absolute Labels</a></h3> -<p>The <code><nobr>ABSOLUTE</nobr></code> directive can be thought of as an -alternative form of <code><nobr>SECTION</nobr></code>: it causes the -subsequent code to be directed at no physical section, but at the -hypothetical section starting at the given absolute address. The only -instructions you can use in this mode are the -<code><nobr>RESB</nobr></code> family. -<p><code><nobr>ABSOLUTE</nobr></code> is used as follows: -<p><pre> -absolute 0x1A - - kbuf_chr resw 1 - kbuf_free resw 1 - kbuf resw 16 -</pre> -<p>This example describes a section of the PC BIOS data area, at segment -address 0x40: the above code defines <code><nobr>kbuf_chr</nobr></code> to -be 0x1A, <code><nobr>kbuf_free</nobr></code> to be 0x1C, and -<code><nobr>kbuf</nobr></code> to be 0x1E. -<p>The user-level form of <code><nobr>ABSOLUTE</nobr></code>, like that of -<code><nobr>SECTION</nobr></code>, redefines the -<code><nobr>__SECT__</nobr></code> macro when it is invoked. -<p><code><nobr>STRUC</nobr></code> and <code><nobr>ENDSTRUC</nobr></code> -are defined as macros which use <code><nobr>ABSOLUTE</nobr></code> (and -also <code><nobr>__SECT__</nobr></code>). -<p><code><nobr>ABSOLUTE</nobr></code> doesn't have to take an absolute -constant as an argument: it can take an expression (actually, a critical -expression: see <a href="nasmdoc3.html#section-3.8">section 3.8</a>) and it -can be a value in a segment. For example, a TSR can re-use its setup code -as run-time BSS like this: -<p><pre> - org 100h ; it's a .COM program - - jmp setup ; setup code comes last - - ; the resident part of the TSR goes here -setup: - ; now write the code that installs the TSR here - -absolute setup - -runtimevar1 resw 1 -runtimevar2 resd 20 - -tsr_end: -</pre> -<p>This defines some variables `on top of' the setup code, so that after -the setup has finished running, the space it took up can be re-used as data -storage for the running TSR. The symbol `tsr_end' can be used to calculate -the total size of the part of the TSR that needs to be made resident. -<h3><a name="section-6.5">6.5 <code><nobr>EXTERN</nobr></code>: Importing Symbols from Other Modules</a></h3> -<p><code><nobr>EXTERN</nobr></code> is similar to the MASM directive -<code><nobr>EXTRN</nobr></code> and the C keyword -<code><nobr>extern</nobr></code>: it is used to declare a symbol which is -not defined anywhere in the module being assembled, but is assumed to be -defined in some other module and needs to be referred to by this one. Not -every object-file format can support external variables: the -<code><nobr>bin</nobr></code> format cannot. -<p>The <code><nobr>EXTERN</nobr></code> directive takes as many arguments -as you like. Each argument is the name of a symbol: -<p><pre> -extern _printf -extern _sscanf,_fscanf -</pre> -<p>Some object-file formats provide extra features to the -<code><nobr>EXTERN</nobr></code> directive. In all cases, the extra -features are used by suffixing a colon to the symbol name followed by -object-format specific text. For example, the <code><nobr>obj</nobr></code> -format allows you to declare that the default segment base of an external -should be the group <code><nobr>dgroup</nobr></code> by means of the -directive -<p><pre> -extern _variable:wrt dgroup -</pre> -<p>The primitive form of <code><nobr>EXTERN</nobr></code> differs from the -user-level form only in that it can take only one argument at a time: the -support for multiple arguments is implemented at the preprocessor level. -<p>You can declare the same variable as <code><nobr>EXTERN</nobr></code> -more than once: NASM will quietly ignore the second and later -redeclarations. You can't declare a variable as -<code><nobr>EXTERN</nobr></code> as well as something else, though. -<h3><a name="section-6.6">6.6 <code><nobr>GLOBAL</nobr></code>: Exporting Symbols to Other Modules</a></h3> -<p><code><nobr>GLOBAL</nobr></code> is the other end of -<code><nobr>EXTERN</nobr></code>: if one module declares a symbol as -<code><nobr>EXTERN</nobr></code> and refers to it, then in order to prevent -linker errors, some other module must actually <em>define</em> the symbol -and declare it as <code><nobr>GLOBAL</nobr></code>. Some assemblers use the -name <code><nobr>PUBLIC</nobr></code> for this purpose. -<p>The <code><nobr>GLOBAL</nobr></code> directive applying to a symbol must -appear <em>before</em> the definition of the symbol. -<p><code><nobr>GLOBAL</nobr></code> uses the same syntax as -<code><nobr>EXTERN</nobr></code>, except that it must refer to symbols -which <em>are</em> defined in the same module as the -<code><nobr>GLOBAL</nobr></code> directive. For example: -<p><pre> -global _main -_main: - ; some code -</pre> -<p><code><nobr>GLOBAL</nobr></code>, like <code><nobr>EXTERN</nobr></code>, -allows object formats to define private extensions by means of a colon. The -<code><nobr>elf</nobr></code> object format, for example, lets you specify -whether global data items are functions or data: -<p><pre> -global hashlookup:function, hashtable:data -</pre> -<p>Like <code><nobr>EXTERN</nobr></code>, the primitive form of -<code><nobr>GLOBAL</nobr></code> differs from the user-level form only in -that it can take only one argument at a time. -<h3><a name="section-6.7">6.7 <code><nobr>COMMON</nobr></code>: Defining Common Data Areas</a></h3> -<p>The <code><nobr>COMMON</nobr></code> directive is used to declare -<em>common variables</em>. A common variable is much like a global variable -declared in the uninitialized data section, so that -<p><pre> -common intvar 4 -</pre> -<p>is similar in function to -<p><pre> -global intvar -section .bss - -intvar resd 1 -</pre> -<p>The difference is that if more than one module defines the same common -variable, then at link time those variables will be <em>merged</em>, and -references to <code><nobr>intvar</nobr></code> in all modules will point at -the same piece of memory. -<p>Like <code><nobr>GLOBAL</nobr></code> and -<code><nobr>EXTERN</nobr></code>, <code><nobr>COMMON</nobr></code> supports -object-format specific extensions. For example, the -<code><nobr>obj</nobr></code> format allows common variables to be NEAR or -FAR, and the <code><nobr>elf</nobr></code> format allows you to specify the -alignment requirements of a common variable: -<p><pre> -common commvar 4:near ; works in OBJ -common intarray 100:4 ; works in ELF: 4 byte aligned -</pre> -<p>Once again, like <code><nobr>EXTERN</nobr></code> and -<code><nobr>GLOBAL</nobr></code>, the primitive form of -<code><nobr>COMMON</nobr></code> differs from the user-level form only in -that it can take only one argument at a time. -<h3><a name="section-6.8">6.8 <code><nobr>CPU</nobr></code>: Defining CPU Dependencies</a></h3> -<p>The <code><nobr>CPU</nobr></code> directive restricts assembly to those -instructions which are available on the specified CPU. -<p>Options are: -<ul> -<li><code><nobr>CPU 8086</nobr></code> Assemble only 8086 instruction set -<li><code><nobr>CPU 186</nobr></code> Assemble instructions up to the 80186 -instruction set -<li><code><nobr>CPU 286</nobr></code> Assemble instructions up to the 286 -instruction set -<li><code><nobr>CPU 386</nobr></code> Assemble instructions up to the 386 -instruction set -<li><code><nobr>CPU 486</nobr></code> 486 instruction set -<li><code><nobr>CPU 586</nobr></code> Pentium instruction set -<li><code><nobr>CPU PENTIUM</nobr></code> Same as 586 -<li><code><nobr>CPU 686</nobr></code> P6 instruction set -<li><code><nobr>CPU PPRO</nobr></code> Same as 686 -<li><code><nobr>CPU P2</nobr></code> Same as 686 -<li><code><nobr>CPU P3</nobr></code> Pentium III (Katmai) instruction sets -<li><code><nobr>CPU KATMAI</nobr></code> Same as P3 -<li><code><nobr>CPU P4</nobr></code> Pentium 4 (Willamette) instruction set -<li><code><nobr>CPU WILLAMETTE</nobr></code> Same as P4 -<li><code><nobr>CPU PRESCOTT</nobr></code> Prescott instruction set -<li><code><nobr>CPU X64</nobr></code> x86-64 (x64/AMD64/Intel 64) -instruction set -<li><code><nobr>CPU IA64</nobr></code> IA64 CPU (in x86 mode) instruction -set -</ul> -<p>All options are case insensitive. All instructions will be selected only -if they apply to the selected CPU or lower. By default, all instructions -are available. -<h3><a name="section-6.9">6.9 <code><nobr>FLOAT</nobr></code>: Handling of floating-point constants</a></h3> -<p>By default, floating-point constants are rounded to nearest, and IEEE -denormals are supported. The following options can be set to alter this -behaviour: -<ul> -<li><code><nobr>FLOAT DAZ</nobr></code> Flush denormals to zero -<li><code><nobr>FLOAT NODAZ</nobr></code> Do not flush denormals to zero -(default) -<li><code><nobr>FLOAT NEAR</nobr></code> Round to nearest (default) -<li><code><nobr>FLOAT UP</nobr></code> Round up (toward +Infinity) -<li><code><nobr>FLOAT DOWN</nobr></code> Round down (toward -Infinity) -<li><code><nobr>FLOAT ZERO</nobr></code> Round toward zero -<li><code><nobr>FLOAT DEFAULT</nobr></code> Restore default settings -</ul> -<p>The standard macros <code><nobr>__FLOAT_DAZ__</nobr></code>, -<code><nobr>__FLOAT_ROUND__</nobr></code>, and -<code><nobr>__FLOAT__</nobr></code> contain the current state, as long as -the programmer has avoided the use of the brackeded primitive form, -(<code><nobr>[FLOAT]</nobr></code>). -<p><code><nobr>__FLOAT__</nobr></code> contains the full set of -floating-point settings; this value can be saved away and invoked later to -restore the setting. -<p align=center><a href="nasmdoc7.html">Next Chapter</a> | -<a href="nasmdoc5.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -</body></html> diff --git a/doc/html/nasmdoc7.html b/doc/html/nasmdoc7.html deleted file mode 100644 index d1339a8..0000000 --- a/doc/html/nasmdoc7.html +++ /dev/null @@ -1,1267 +0,0 @@ -<html><head><title>NASM Manual</title></head> -<body><h1 align=center>The Netwide Assembler: NASM</h1> - -<p align=center><a href="nasmdoc8.html">Next Chapter</a> | -<a href="nasmdoc6.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -<h2><a name="chapter-7">Chapter 7: Output Formats</a></h2> -<p>NASM is a portable assembler, designed to be able to compile on any ANSI -C-supporting platform and produce output to run on a variety of Intel x86 -operating systems. For this reason, it has a large number of available -output formats, selected using the <code><nobr>-f</nobr></code> option on -the NASM command line. Each of these formats, along with its extensions to -the base NASM syntax, is detailed in this chapter. -<p>As stated in <a href="nasmdoc2.html#section-2.1.1">section 2.1.1</a>, -NASM chooses a default name for your output file based on the input file -name and the chosen output format. This will be generated by removing the -extension (<code><nobr>.asm</nobr></code>, <code><nobr>.s</nobr></code>, or -whatever you like to use) from the input file name, and substituting an -extension defined by the output format. The extensions are given with each -format below. -<h3><a name="section-7.1">7.1 <code><nobr>bin</nobr></code>: Flat-Form Binary Output</a></h3> -<p>The <code><nobr>bin</nobr></code> format does not produce object files: -it generates nothing in the output file except the code you wrote. Such -`pure binary' files are used by MS-DOS: <code><nobr>.COM</nobr></code> -executables and <code><nobr>.SYS</nobr></code> device drivers are pure -binary files. Pure binary output is also useful for operating system and -boot loader development. -<p>The <code><nobr>bin</nobr></code> format supports multiple section -names. For details of how NASM handles sections in the -<code><nobr>bin</nobr></code> format, see <a href="#section-7.1.3">section -7.1.3</a>. -<p>Using the <code><nobr>bin</nobr></code> format puts NASM by default into -16-bit mode (see <a href="nasmdoc6.html#section-6.1">section 6.1</a>). In -order to use <code><nobr>bin</nobr></code> to write 32-bit or 64-bit code, -such as an OS kernel, you need to explicitly issue the -<code><nobr>BITS 32</nobr></code> or <code><nobr>BITS 64</nobr></code> -directive. -<p><code><nobr>bin</nobr></code> has no default output file name extension: -instead, it leaves your file name as it is once the original extension has -been removed. Thus, the default is for NASM to assemble -<code><nobr>binprog.asm</nobr></code> into a binary file called -<code><nobr>binprog</nobr></code>. -<h4><a name="section-7.1.1">7.1.1 <code><nobr>ORG</nobr></code>: Binary File Program Origin</a></h4> -<p>The <code><nobr>bin</nobr></code> format provides an additional -directive to the list given in <a href="nasmdoc6.html">chapter 6</a>: -<code><nobr>ORG</nobr></code>. The function of the -<code><nobr>ORG</nobr></code> directive is to specify the origin address -which NASM will assume the program begins at when it is loaded into memory. -<p>For example, the following code will generate the longword -<code><nobr>0x00000104</nobr></code>: -<p><pre> - org 0x100 - dd label -label: -</pre> -<p>Unlike the <code><nobr>ORG</nobr></code> directive provided by -MASM-compatible assemblers, which allows you to jump around in the object -file and overwrite code you have already generated, NASM's -<code><nobr>ORG</nobr></code> does exactly what the directive says: -<em>origin</em>. Its sole function is to specify one offset which is added -to all internal address references within the section; it does not permit -any of the trickery that MASM's version does. See -<a href="nasmdo12.html#section-12.1.3">section 12.1.3</a> for further -comments. -<h4><a name="section-7.1.2">7.1.2 <code><nobr>bin</nobr></code> Extensions to the <code><nobr>SECTION</nobr></code> Directive</a></h4> -<p>The <code><nobr>bin</nobr></code> output format extends the -<code><nobr>SECTION</nobr></code> (or <code><nobr>SEGMENT</nobr></code>) -directive to allow you to specify the alignment requirements of segments. -This is done by appending the <code><nobr>ALIGN</nobr></code> qualifier to -the end of the section-definition line. For example, -<p><pre> -section .data align=16 -</pre> -<p>switches to the section <code><nobr>.data</nobr></code> and also -specifies that it must be aligned on a 16-byte boundary. -<p>The parameter to <code><nobr>ALIGN</nobr></code> specifies how many low -bits of the section start address must be forced to zero. The alignment -value given may be any power of two. -<h4><a name="section-7.1.3">7.1.3 Multisection Support for the <code><nobr>bin</nobr></code> Format</a></h4> -<p>The <code><nobr>bin</nobr></code> format allows the use of multiple -sections, of arbitrary names, besides the "known" -<code><nobr>.text</nobr></code>, <code><nobr>.data</nobr></code>, and -<code><nobr>.bss</nobr></code> names. -<ul> -<li>Sections may be designated <code><nobr>progbits</nobr></code> or -<code><nobr>nobits</nobr></code>. Default is -<code><nobr>progbits</nobr></code> (except <code><nobr>.bss</nobr></code>, -which defaults to <code><nobr>nobits</nobr></code>, of course). -<li>Sections can be aligned at a specified boundary following the previous -section with <code><nobr>align=</nobr></code>, or at an arbitrary -byte-granular position with <code><nobr>start=</nobr></code>. -<li>Sections can be given a virtual start address, which will be used for -the calculation of all memory references within that section with -<code><nobr>vstart=</nobr></code>. -<li>Sections can be ordered using -<code><nobr>follows=</nobr></code><code><nobr><section></nobr></code> -or -<code><nobr>vfollows=</nobr></code><code><nobr><section></nobr></code> -as an alternative to specifying an explicit start address. -<li>Arguments to <code><nobr>org</nobr></code>, -<code><nobr>start</nobr></code>, <code><nobr>vstart</nobr></code>, and -<code><nobr>align=</nobr></code> are critical expressions. See -<a href="nasmdoc3.html#section-3.8">section 3.8</a>. E.g. -<code><nobr>align=(1 << ALIGN_SHIFT)</nobr></code> - -<code><nobr>ALIGN_SHIFT</nobr></code> must be defined before it is used -here. -<li>Any code which comes before an explicit -<code><nobr>SECTION</nobr></code> directive is directed by default into the -<code><nobr>.text</nobr></code> section. -<li>If an <code><nobr>ORG</nobr></code> statement is not given, -<code><nobr>ORG 0</nobr></code> is used by default. -<li>The <code><nobr>.bss</nobr></code> section will be placed after the -last <code><nobr>progbits</nobr></code> section, unless -<code><nobr>start=</nobr></code>, <code><nobr>vstart=</nobr></code>, -<code><nobr>follows=</nobr></code>, or <code><nobr>vfollows=</nobr></code> -has been specified. -<li>All sections are aligned on dword boundaries, unless a different -alignment has been specified. -<li>Sections may not overlap. -<li>NASM creates the -<code><nobr>section.<secname>.start</nobr></code> for each section, -which may be used in your code. -</ul> -<h4><a name="section-7.1.4">7.1.4 Map Files</a></h4> -<p>Map files can be generated in <code><nobr>-f bin</nobr></code> format by -means of the <code><nobr>[map]</nobr></code> option. Map types of -<code><nobr>all</nobr></code> (default), <code><nobr>brief</nobr></code>, -<code><nobr>sections</nobr></code>, <code><nobr>segments</nobr></code>, or -<code><nobr>symbols</nobr></code> may be specified. Output may be directed -to <code><nobr>stdout</nobr></code> (default), -<code><nobr>stderr</nobr></code>, or a specified file. E.g. -<code><nobr>[map symbols myfile.map]</nobr></code>. No "user form" exists, -the square brackets must be used. -<h3><a name="section-7.2">7.2 <code><nobr>ith</nobr></code>: Intel Hex Output</a></h3> -<p>The <code><nobr>ith</nobr></code> file format produces Intel hex-format -files. Just as the <code><nobr>bin</nobr></code> format, this is a flat -memory image format with no support for relocation or linking. It is -usually used with ROM programmers and similar utilities. -<p>All extensions supported by the <code><nobr>bin</nobr></code> file -format is also supported by the <code><nobr>ith</nobr></code> file format. -<p><code><nobr>ith</nobr></code> provides a default output file-name -extension of <code><nobr>.ith</nobr></code>. -<h3><a name="section-7.3">7.3 <code><nobr>srec</nobr></code>: Motorola S-Records Output</a></h3> -<p>The <code><nobr>srec</nobr></code> file format produces Motorola -S-records files. Just as the <code><nobr>bin</nobr></code> format, this is -a flat memory image format with no support for relocation or linking. It is -usually used with ROM programmers and similar utilities. -<p>All extensions supported by the <code><nobr>bin</nobr></code> file -format is also supported by the <code><nobr>srec</nobr></code> file format. -<p><code><nobr>srec</nobr></code> provides a default output file-name -extension of <code><nobr>.srec</nobr></code>. -<h3><a name="section-7.4">7.4 <code><nobr>obj</nobr></code>: Microsoft OMF Object Files</a></h3> -<p>The <code><nobr>obj</nobr></code> file format (NASM calls it -<code><nobr>obj</nobr></code> rather than <code><nobr>omf</nobr></code> for -historical reasons) is the one produced by MASM and TASM, which is -typically fed to 16-bit DOS linkers to produce -<code><nobr>.EXE</nobr></code> files. It is also the format used by OS/2. -<p><code><nobr>obj</nobr></code> provides a default output file-name -extension of <code><nobr>.obj</nobr></code>. -<p><code><nobr>obj</nobr></code> is not exclusively a 16-bit format, -though: NASM has full support for the 32-bit extensions to the format. In -particular, 32-bit <code><nobr>obj</nobr></code> format files are used by -Borland's Win32 compilers, instead of using Microsoft's newer -<code><nobr>win32</nobr></code> object file format. -<p>The <code><nobr>obj</nobr></code> format does not define any special -segment names: you can call your segments anything you like. Typical names -for segments in <code><nobr>obj</nobr></code> format files are -<code><nobr>CODE</nobr></code>, <code><nobr>DATA</nobr></code> and -<code><nobr>BSS</nobr></code>. -<p>If your source file contains code before specifying an explicit -<code><nobr>SEGMENT</nobr></code> directive, then NASM will invent its own -segment called <code><nobr>__NASMDEFSEG</nobr></code> for you. -<p>When you define a segment in an <code><nobr>obj</nobr></code> file, NASM -defines the segment name as a symbol as well, so that you can access the -segment address of the segment. So, for example: -<p><pre> -segment data - -dvar: dw 1234 - -segment code - -function: - mov ax,data ; get segment address of data - mov ds,ax ; and move it into DS - inc word [dvar] ; now this reference will work - ret -</pre> -<p>The <code><nobr>obj</nobr></code> format also enables the use of the -<code><nobr>SEG</nobr></code> and <code><nobr>WRT</nobr></code> operators, -so that you can write code which does things like -<p><pre> -extern foo - - mov ax,seg foo ; get preferred segment of foo - mov ds,ax - mov ax,data ; a different segment - mov es,ax - mov ax,[ds:foo] ; this accesses `foo' - mov [es:foo wrt data],bx ; so does this -</pre> -<h4><a name="section-7.4.1">7.4.1 <code><nobr>obj</nobr></code> Extensions to the <code><nobr>SEGMENT</nobr></code> Directive</a></h4> -<p>The <code><nobr>obj</nobr></code> output format extends the -<code><nobr>SEGMENT</nobr></code> (or <code><nobr>SECTION</nobr></code>) -directive to allow you to specify various properties of the segment you are -defining. This is done by appending extra qualifiers to the end of the -segment-definition line. For example, -<p><pre> -segment code private align=16 -</pre> -<p>defines the segment <code><nobr>code</nobr></code>, but also declares it -to be a private segment, and requires that the portion of it described in -this code module must be aligned on a 16-byte boundary. -<p>The available qualifiers are: -<ul> -<li><code><nobr>PRIVATE</nobr></code>, <code><nobr>PUBLIC</nobr></code>, -<code><nobr>COMMON</nobr></code> and <code><nobr>STACK</nobr></code> -specify the combination characteristics of the segment. -<code><nobr>PRIVATE</nobr></code> segments do not get combined with any -others by the linker; <code><nobr>PUBLIC</nobr></code> and -<code><nobr>STACK</nobr></code> segments get concatenated together at link -time; and <code><nobr>COMMON</nobr></code> segments all get overlaid on top -of each other rather than stuck end-to-end. -<li><code><nobr>ALIGN</nobr></code> is used, as shown above, to specify how -many low bits of the segment start address must be forced to zero. The -alignment value given may be any power of two from 1 to 4096; in reality, -the only values supported are 1, 2, 4, 16, 256 and 4096, so if 8 is -specified it will be rounded up to 16, and 32, 64 and 128 will all be -rounded up to 256, and so on. Note that alignment to 4096-byte boundaries -is a PharLap extension to the format and may not be supported by all -linkers. -<li><code><nobr>CLASS</nobr></code> can be used to specify the segment -class; this feature indicates to the linker that segments of the same class -should be placed near each other in the output file. The class name can be -any word, e.g. <code><nobr>CLASS=CODE</nobr></code>. -<li><code><nobr>OVERLAY</nobr></code>, like -<code><nobr>CLASS</nobr></code>, is specified with an arbitrary word as an -argument, and provides overlay information to an overlay-capable linker. -<li>Segments can be declared as <code><nobr>USE16</nobr></code> or -<code><nobr>USE32</nobr></code>, which has the effect of recording the -choice in the object file and also ensuring that NASM's default assembly -mode when assembling in that segment is 16-bit or 32-bit respectively. -<li>When writing OS/2 object files, you should declare 32-bit segments as -<code><nobr>FLAT</nobr></code>, which causes the default segment base for -anything in the segment to be the special group -<code><nobr>FLAT</nobr></code>, and also defines the group if it is not -already defined. -<li>The <code><nobr>obj</nobr></code> file format also allows segments to -be declared as having a pre-defined absolute segment address, although no -linkers are currently known to make sensible use of this feature; -nevertheless, NASM allows you to declare a segment such as -<code><nobr>SEGMENT SCREEN ABSOLUTE=0xB800</nobr></code> if you need to. -The <code><nobr>ABSOLUTE</nobr></code> and <code><nobr>ALIGN</nobr></code> -keywords are mutually exclusive. -</ul> -<p>NASM's default segment attributes are <code><nobr>PUBLIC</nobr></code>, -<code><nobr>ALIGN=1</nobr></code>, no class, no overlay, and -<code><nobr>USE16</nobr></code>. -<h4><a name="section-7.4.2">7.4.2 <code><nobr>GROUP</nobr></code>: Defining Groups of Segments</a></h4> -<p>The <code><nobr>obj</nobr></code> format also allows segments to be -grouped, so that a single segment register can be used to refer to all the -segments in a group. NASM therefore supplies the -<code><nobr>GROUP</nobr></code> directive, whereby you can code -<p><pre> -segment data - - ; some data - -segment bss - - ; some uninitialized data - -group dgroup data bss -</pre> -<p>which will define a group called <code><nobr>dgroup</nobr></code> to -contain the segments <code><nobr>data</nobr></code> and -<code><nobr>bss</nobr></code>. Like <code><nobr>SEGMENT</nobr></code>, -<code><nobr>GROUP</nobr></code> causes the group name to be defined as a -symbol, so that you can refer to a variable <code><nobr>var</nobr></code> -in the <code><nobr>data</nobr></code> segment as -<code><nobr>var wrt data</nobr></code> or as -<code><nobr>var wrt dgroup</nobr></code>, depending on which segment value -is currently in your segment register. -<p>If you just refer to <code><nobr>var</nobr></code>, however, and -<code><nobr>var</nobr></code> is declared in a segment which is part of a -group, then NASM will default to giving you the offset of -<code><nobr>var</nobr></code> from the beginning of the <em>group</em>, not -the <em>segment</em>. Therefore <code><nobr>SEG var</nobr></code>, also, -will return the group base rather than the segment base. -<p>NASM will allow a segment to be part of more than one group, but will -generate a warning if you do this. Variables declared in a segment which is -part of more than one group will default to being relative to the first -group that was defined to contain the segment. -<p>A group does not have to contain any segments; you can still make -<code><nobr>WRT</nobr></code> references to a group which does not contain -the variable you are referring to. OS/2, for example, defines the special -group <code><nobr>FLAT</nobr></code> with no segments in it. -<h4><a name="section-7.4.3">7.4.3 <code><nobr>UPPERCASE</nobr></code>: Disabling Case Sensitivity in Output</a></h4> -<p>Although NASM itself is case sensitive, some OMF linkers are not; -therefore it can be useful for NASM to output single-case object files. The -<code><nobr>UPPERCASE</nobr></code> format-specific directive causes all -segment, group and symbol names that are written to the object file to be -forced to upper case just before being written. Within a source file, NASM -is still case-sensitive; but the object file can be written entirely in -upper case if desired. -<p><code><nobr>UPPERCASE</nobr></code> is used alone on a line; it requires -no parameters. -<h4><a name="section-7.4.4">7.4.4 <code><nobr>IMPORT</nobr></code>: Importing DLL Symbols</a></h4> -<p>The <code><nobr>IMPORT</nobr></code> format-specific directive defines a -symbol to be imported from a DLL, for use if you are writing a DLL's import -library in NASM. You still need to declare the symbol as -<code><nobr>EXTERN</nobr></code> as well as using the -<code><nobr>IMPORT</nobr></code> directive. -<p>The <code><nobr>IMPORT</nobr></code> directive takes two required -parameters, separated by white space, which are (respectively) the name of -the symbol you wish to import and the name of the library you wish to -import it from. For example: -<p><pre> - import WSAStartup wsock32.dll -</pre> -<p>A third optional parameter gives the name by which the symbol is known -in the library you are importing it from, in case this is not the same as -the name you wish the symbol to be known by to your code once you have -imported it. For example: -<p><pre> - import asyncsel wsock32.dll WSAAsyncSelect -</pre> -<h4><a name="section-7.4.5">7.4.5 <code><nobr>EXPORT</nobr></code>: Exporting DLL Symbols</a></h4> -<p>The <code><nobr>EXPORT</nobr></code> format-specific directive defines a -global symbol to be exported as a DLL symbol, for use if you are writing a -DLL in NASM. You still need to declare the symbol as -<code><nobr>GLOBAL</nobr></code> as well as using the -<code><nobr>EXPORT</nobr></code> directive. -<p><code><nobr>EXPORT</nobr></code> takes one required parameter, which is -the name of the symbol you wish to export, as it was defined in your source -file. An optional second parameter (separated by white space from the -first) gives the <em>external</em> name of the symbol: the name by which -you wish the symbol to be known to programs using the DLL. If this name is -the same as the internal name, you may leave the second parameter off. -<p>Further parameters can be given to define attributes of the exported -symbol. These parameters, like the second, are separated by white space. If -further parameters are given, the external name must also be specified, -even if it is the same as the internal name. The available attributes are: -<ul> -<li><code><nobr>resident</nobr></code> indicates that the exported name is -to be kept resident by the system loader. This is an optimisation for -frequently used symbols imported by name. -<li><code><nobr>nodata</nobr></code> indicates that the exported symbol is -a function which does not make use of any initialized data. -<li><code><nobr>parm=NNN</nobr></code>, where <code><nobr>NNN</nobr></code> -is an integer, sets the number of parameter words for the case in which the -symbol is a call gate between 32-bit and 16-bit segments. -<li>An attribute which is just a number indicates that the symbol should be -exported with an identifying number (ordinal), and gives the desired -number. -</ul> -<p>For example: -<p><pre> - export myfunc - export myfunc TheRealMoreFormalLookingFunctionName - export myfunc myfunc 1234 ; export by ordinal - export myfunc myfunc resident parm=23 nodata -</pre> -<h4><a name="section-7.4.6">7.4.6 <code><nobr>..start</nobr></code>: Defining the Program Entry Point</a></h4> -<p><code><nobr>OMF</nobr></code> linkers require exactly one of the object -files being linked to define the program entry point, where execution will -begin when the program is run. If the object file that defines the entry -point is assembled using NASM, you specify the entry point by declaring the -special symbol <code><nobr>..start</nobr></code> at the point where you -wish execution to begin. -<h4><a name="section-7.4.7">7.4.7 <code><nobr>obj</nobr></code> Extensions to the <code><nobr>EXTERN</nobr></code> Directive</a></h4> -<p>If you declare an external symbol with the directive -<p><pre> - extern foo -</pre> -<p>then references such as <code><nobr>mov ax,foo</nobr></code> will give -you the offset of <code><nobr>foo</nobr></code> from its preferred segment -base (as specified in whichever module <code><nobr>foo</nobr></code> is -actually defined in). So to access the contents of -<code><nobr>foo</nobr></code> you will usually need to do something like -<p><pre> - mov ax,seg foo ; get preferred segment base - mov es,ax ; move it into ES - mov ax,[es:foo] ; and use offset `foo' from it -</pre> -<p>This is a little unwieldy, particularly if you know that an external is -going to be accessible from a given segment or group, say -<code><nobr>dgroup</nobr></code>. So if <code><nobr>DS</nobr></code> -already contained <code><nobr>dgroup</nobr></code>, you could simply code -<p><pre> - mov ax,[foo wrt dgroup] -</pre> -<p>However, having to type this every time you want to access -<code><nobr>foo</nobr></code> can be a pain; so NASM allows you to declare -<code><nobr>foo</nobr></code> in the alternative form -<p><pre> - extern foo:wrt dgroup -</pre> -<p>This form causes NASM to pretend that the preferred segment base of -<code><nobr>foo</nobr></code> is in fact <code><nobr>dgroup</nobr></code>; -so the expression <code><nobr>seg foo</nobr></code> will now return -<code><nobr>dgroup</nobr></code>, and the expression -<code><nobr>foo</nobr></code> is equivalent to -<code><nobr>foo wrt dgroup</nobr></code>. -<p>This default-<code><nobr>WRT</nobr></code> mechanism can be used to make -externals appear to be relative to any group or segment in your program. It -can also be applied to common variables: see -<a href="#section-7.4.8">section 7.4.8</a>. -<h4><a name="section-7.4.8">7.4.8 <code><nobr>obj</nobr></code> Extensions to the <code><nobr>COMMON</nobr></code> Directive</a></h4> -<p>The <code><nobr>obj</nobr></code> format allows common variables to be -either near or far; NASM allows you to specify which your variables should -be by the use of the syntax -<p><pre> -common nearvar 2:near ; `nearvar' is a near common -common farvar 10:far ; and `farvar' is far -</pre> -<p>Far common variables may be greater in size than 64Kb, and so the OMF -specification says that they are declared as a number of <em>elements</em> -of a given size. So a 10-byte far common variable could be declared as ten -one-byte elements, five two-byte elements, two five-byte elements or one -ten-byte element. -<p>Some <code><nobr>OMF</nobr></code> linkers require the element size, as -well as the variable size, to match when resolving common variables -declared in more than one module. Therefore NASM must allow you to specify -the element size on your far common variables. This is done by the -following syntax: -<p><pre> -common c_5by2 10:far 5 ; two five-byte elements -common c_2by5 10:far 2 ; five two-byte elements -</pre> -<p>If no element size is specified, the default is 1. Also, the -<code><nobr>FAR</nobr></code> keyword is not required when an element size -is specified, since only far commons may have element sizes at all. So the -above declarations could equivalently be -<p><pre> -common c_5by2 10:5 ; two five-byte elements -common c_2by5 10:2 ; five two-byte elements -</pre> -<p>In addition to these extensions, the <code><nobr>COMMON</nobr></code> -directive in <code><nobr>obj</nobr></code> also supports -default-<code><nobr>WRT</nobr></code> specification like -<code><nobr>EXTERN</nobr></code> does (explained in -<a href="#section-7.4.7">section 7.4.7</a>). So you can also declare things -like -<p><pre> -common foo 10:wrt dgroup -common bar 16:far 2:wrt data -common baz 24:wrt data:6 -</pre> -<h3><a name="section-7.5">7.5 <code><nobr>win32</nobr></code>: Microsoft Win32 Object Files</a></h3> -<p>The <code><nobr>win32</nobr></code> output format generates Microsoft -Win32 object files, suitable for passing to Microsoft linkers such as -Visual C++. Note that Borland Win32 compilers do not use this format, but -use <code><nobr>obj</nobr></code> instead (see -<a href="#section-7.4">section 7.4</a>). -<p><code><nobr>win32</nobr></code> provides a default output file-name -extension of <code><nobr>.obj</nobr></code>. -<p>Note that although Microsoft say that Win32 object files follow the -<code><nobr>COFF</nobr></code> (Common Object File Format) standard, the -object files produced by Microsoft Win32 compilers are not compatible with -COFF linkers such as DJGPP's, and vice versa. This is due to a difference -of opinion over the precise semantics of PC-relative relocations. To -produce COFF files suitable for DJGPP, use NASM's -<code><nobr>coff</nobr></code> output format; conversely, the -<code><nobr>coff</nobr></code> format does not produce object files that -Win32 linkers can generate correct output from. -<h4><a name="section-7.5.1">7.5.1 <code><nobr>win32</nobr></code> Extensions to the <code><nobr>SECTION</nobr></code> Directive</a></h4> -<p>Like the <code><nobr>obj</nobr></code> format, -<code><nobr>win32</nobr></code> allows you to specify additional -information on the <code><nobr>SECTION</nobr></code> directive line, to -control the type and properties of sections you declare. Section types and -properties are generated automatically by NASM for the standard section -names <code><nobr>.text</nobr></code>, <code><nobr>.data</nobr></code> and -<code><nobr>.bss</nobr></code>, but may still be overridden by these -qualifiers. -<p>The available qualifiers are: -<ul> -<li><code><nobr>code</nobr></code>, or equivalently -<code><nobr>text</nobr></code>, defines the section to be a code section. -This marks the section as readable and executable, but not writable, and -also indicates to the linker that the type of the section is code. -<li><code><nobr>data</nobr></code> and <code><nobr>bss</nobr></code> define -the section to be a data section, analogously to -<code><nobr>code</nobr></code>. Data sections are marked as readable and -writable, but not executable. <code><nobr>data</nobr></code> declares an -initialized data section, whereas <code><nobr>bss</nobr></code> declares an -uninitialized data section. -<li><code><nobr>rdata</nobr></code> declares an initialized data section -that is readable but not writable. Microsoft compilers use this section to -place constants in it. -<li><code><nobr>info</nobr></code> defines the section to be an -informational section, which is not included in the executable file by the -linker, but may (for example) pass information <em>to</em> the linker. For -example, declaring an <code><nobr>info</nobr></code>-type section called -<code><nobr>.drectve</nobr></code> causes the linker to interpret the -contents of the section as command-line options. -<li><code><nobr>align=</nobr></code>, used with a trailing number as in -<code><nobr>obj</nobr></code>, gives the alignment requirements of the -section. The maximum you may specify is 64: the Win32 object file format -contains no means to request a greater section alignment than this. If -alignment is not explicitly specified, the defaults are 16-byte alignment -for code sections, 8-byte alignment for rdata sections and 4-byte alignment -for data (and BSS) sections. Informational sections get a default alignment -of 1 byte (no alignment), though the value does not matter. -</ul> -<p>The defaults assumed by NASM if you do not specify the above qualifiers -are: -<p><pre> -section .text code align=16 -section .data data align=4 -section .rdata rdata align=8 -section .bss bss align=4 -</pre> -<p>Any other section name is treated by default like -<code><nobr>.text</nobr></code>. -<h4><a name="section-7.5.2">7.5.2 <code><nobr>win32</nobr></code>: Safe Structured Exception Handling</a></h4> -<p>Among other improvements in Windows XP SP2 and Windows Server 2003 -Microsoft has introduced concept of "safe structured exception handling." -General idea is to collect handlers' entry points in designated read-only -table and have alleged entry point verified against this table prior -exception control is passed to the handler. In order for an executable -module to be equipped with such "safe exception handler table," all object -modules on linker command line has to comply with certain criteria. If one -single module among them does not, then the table in question is omitted -and above mentioned run-time checks will not be performed for application -in question. Table omission is by default silent and therefore can be -easily overlooked. One can instruct linker to refuse to produce binary -without such table by passing <code><nobr>/safeseh</nobr></code> command -line option. -<p>Without regard to this run-time check merits it's natural to expect NASM -to be capable of generating modules suitable for -<code><nobr>/safeseh</nobr></code> linking. From developer's viewpoint the -problem is two-fold: -<ul> -<li>how to adapt modules not deploying exception handlers of their own; -<li>how to adapt/develop modules utilizing custom exception handling; -</ul> -<p>Former can be easily achieved with any NASM version by adding following -line to source code: -<p><pre> -$@feat.00 equ 1 -</pre> -<p>As of version 2.03 NASM adds this absolute symbol automatically. If it's -not already present to be precise. I.e. if for whatever reason developer -would choose to assign another value in source file, it would still be -perfectly possible. -<p>Registering custom exception handler on the other hand requires certain -"magic." As of version 2.03 additional directive is implemented, -<code><nobr>safeseh</nobr></code>, which instructs the assembler to produce -appropriately formatted input data for above mentioned "safe exception -handler table." Its typical use would be: -<p><pre> -section .text -extern _MessageBoxA@16 -%if __NASM_VERSION_ID__ >= 0x02030000 -safeseh handler ; register handler as "safe handler" -%endif -handler: - push DWORD 1 ; MB_OKCANCEL - push DWORD caption - push DWORD text - push DWORD 0 - call _MessageBoxA@16 - sub eax,1 ; incidentally suits as return value - ; for exception handler - ret -global _main -_main: - push DWORD handler - push DWORD [fs:0] - mov DWORD [fs:0],esp ; engage exception handler - xor eax,eax - mov eax,DWORD[eax] ; cause exception - pop DWORD [fs:0] ; disengage exception handler - add esp,4 - ret -text: db 'OK to rethrow, CANCEL to generate core dump',0 -caption:db 'SEGV',0 - -section .drectve info - db '/defaultlib:user32.lib /defaultlib:msvcrt.lib ' -</pre> -<p>As you might imagine, it's perfectly possible to produce .exe binary -with "safe exception handler table" and yet engage unregistered exception -handler. Indeed, handler is engaged by simply manipulating -<code><nobr>[fs:0]</nobr></code> location at run-time, something linker has -no power over, run-time that is. It should be explicitly mentioned that -such failure to register handler's entry point with -<code><nobr>safeseh</nobr></code> directive has undesired side effect at -run-time. If exception is raised and unregistered handler is to be -executed, the application is abruptly terminated without any notification -whatsoever. One can argue that system could at least have logged some kind -"non-safe exception handler in x.exe at address n" message in event log, -but no, literally no notification is provided and user is left with no clue -on what caused application failure. -<p>Finally, all mentions of linker in this paragraph refer to Microsoft -linker version 7.x and later. Presence of -<code><nobr>@feat.00</nobr></code> symbol and input data for "safe -exception handler table" causes no backward incompatibilities and "safeseh" -modules generated by NASM 2.03 and later can still be linked by earlier -versions or non-Microsoft linkers. -<h3><a name="section-7.6">7.6 <code><nobr>win64</nobr></code>: Microsoft Win64 Object Files</a></h3> -<p>The <code><nobr>win64</nobr></code> output format generates Microsoft -Win64 object files, which is nearly 100% identical to the -<code><nobr>win32</nobr></code> object format -(<a href="#section-7.5">section 7.5</a>) with the exception that it is -meant to target 64-bit code and the x86-64 platform altogether. This object -file is used exactly the same as the <code><nobr>win32</nobr></code> object -format (<a href="#section-7.5">section 7.5</a>), in NASM, with regard to -this exception. -<h4><a name="section-7.6.1">7.6.1 <code><nobr>win64</nobr></code>: Writing Position-Independent Code</a></h4> -<p>While <code><nobr>REL</nobr></code> takes good care of RIP-relative -addressing, there is one aspect that is easy to overlook for a Win64 -programmer: indirect references. Consider a switch dispatch table: -<p><pre> - jmp QWORD[dsptch+rax*8] - ... -dsptch: dq case0 - dq case1 - ... -</pre> -<p>Even novice Win64 assembler programmer will soon realize that the code -is not 64-bit savvy. Most notably linker will refuse to link it with -"<code><nobr>'ADDR32' relocation to '.text' invalid without /LARGEADDRESSAWARE:NO</nobr></code>". -So [s]he will have to split jmp instruction as following: -<p><pre> - lea rbx,[rel dsptch] - jmp QWORD[rbx+rax*8] -</pre> -<p>What happens behind the scene is that effective address in -<code><nobr>lea</nobr></code> is encoded relative to instruction pointer, -or in perfectly position-independent manner. But this is only part of the -problem! Trouble is that in .dll context <code><nobr>caseN</nobr></code> -relocations will make their way to the final module and might have to be -adjusted at .dll load time. To be specific when it can't be loaded at -preferred address. And when this occurs, pages with such relocations will -be rendered private to current process, which kind of undermines the idea -of sharing .dll. But no worry, it's trivial to fix: -<p><pre> - lea rbx,[rel dsptch] - add rbx,QWORD[rbx+rax*8] - jmp rbx - ... -dsptch: dq case0-dsptch - dq case1-dsptch - ... -</pre> -<p>NASM version 2.03 and later provides another alternative, -<code><nobr>wrt ..imagebase</nobr></code> operator, which returns offset -from base address of the current image, be it .exe or .dll module, -therefore the name. For those acquainted with PE-COFF format base address -denotes start of <code><nobr>IMAGE_DOS_HEADER</nobr></code> structure. Here -is how to implement switch with these image-relative references: -<p><pre> - lea rbx,[rel dsptch] - mov eax,DWORD[rbx+rax*4] - sub rbx,dsptch wrt ..imagebase - add rbx,rax - jmp rbx - ... -dsptch: dd case0 wrt ..imagebase - dd case1 wrt ..imagebase -</pre> -<p>One can argue that the operator is redundant. Indeed, snippet before -last works just fine with any NASM version and is not even Windows -specific... The real reason for implementing -<code><nobr>wrt ..imagebase</nobr></code> will become apparent in next -paragraph. -<p>It should be noted that <code><nobr>wrt ..imagebase</nobr></code> is -defined as 32-bit operand only: -<p><pre> - dd label wrt ..imagebase ; ok - dq label wrt ..imagebase ; bad - mov eax,label wrt ..imagebase ; ok - mov rax,label wrt ..imagebase ; bad -</pre> -<h4><a name="section-7.6.2">7.6.2 <code><nobr>win64</nobr></code>: Structured Exception Handling</a></h4> -<p>Structured exception handing in Win64 is completely different matter -from Win32. Upon exception program counter value is noted, and -linker-generated table comprising start and end addresses of all the -functions [in given executable module] is traversed and compared to the -saved program counter. Thus so called <code><nobr>UNWIND_INFO</nobr></code> -structure is identified. If it's not found, then offending subroutine is -assumed to be "leaf" and just mentioned lookup procedure is attempted for -its caller. In Win64 leaf function is such function that does not call any -other function <em>nor</em> modifies any Win64 non-volatile registers, -including stack pointer. The latter ensures that it's possible to identify -leaf function's caller by simply pulling the value from the top of the -stack. -<p>While majority of subroutines written in assembler are not calling any -other function, requirement for non-volatile registers' immutability leaves -developer with not more than 7 registers and no stack frame, which is not -necessarily what [s]he counted with. Customarily one would meet the -requirement by saving non-volatile registers on stack and restoring them -upon return, so what can go wrong? If [and only if] an exception is raised -at run-time and no <code><nobr>UNWIND_INFO</nobr></code> structure is -associated with such "leaf" function, the stack unwind procedure will -expect to find caller's return address on the top of stack immediately -followed by its frame. Given that developer pushed caller's non-volatile -registers on stack, would the value on top point at some code segment or -even addressable space? Well, developer can attempt copying caller's return -address to the top of stack and this would actually work in some very -specific circumstances. But unless developer can guarantee that these -circumstances are always met, it's more appropriate to assume worst case -scenario, i.e. stack unwind procedure going berserk. Relevant question is -what happens then? Application is abruptly terminated without any -notification whatsoever. Just like in Win32 case, one can argue that system -could at least have logged "unwind procedure went berserk in x.exe at -address n" in event log, but no, no trace of failure is left. -<p>Now, when we understand significance of the -<code><nobr>UNWIND_INFO</nobr></code> structure, let's discuss what's in it -and/or how it's processed. First of all it is checked for presence of -reference to custom language-specific exception handler. If there is one, -then it's invoked. Depending on the return value, execution flow is resumed -(exception is said to be "handled"), <em>or</em> rest of -<code><nobr>UNWIND_INFO</nobr></code> structure is processed as following. -Beside optional reference to custom handler, it carries information about -current callee's stack frame and where non-volatile registers are saved. -Information is detailed enough to be able to reconstruct contents of -caller's non-volatile registers upon call to current callee. And so -caller's context is reconstructed, and then unwind procedure is repeated, -i.e. another <code><nobr>UNWIND_INFO</nobr></code> structure is associated, -this time, with caller's instruction pointer, which is then checked for -presence of reference to language-specific handler, etc. The procedure is -recursively repeated till exception is handled. As last resort system -"handles" it by generating memory core dump and terminating the -application. -<p>As for the moment of this writing NASM unfortunately does not facilitate -generation of above mentioned detailed information about stack frame -layout. But as of version 2.03 it implements building blocks for generating -structures involved in stack unwinding. As simplest example, here is how to -deploy custom exception handler for leaf function: -<p><pre> -default rel -section .text -extern MessageBoxA -handler: - sub rsp,40 - mov rcx,0 - lea rdx,[text] - lea r8,[caption] - mov r9,1 ; MB_OKCANCEL - call MessageBoxA - sub eax,1 ; incidentally suits as return value - ; for exception handler - add rsp,40 - ret -global main -main: - xor rax,rax - mov rax,QWORD[rax] ; cause exception - ret -main_end: -text: db 'OK to rethrow, CANCEL to generate core dump',0 -caption:db 'SEGV',0 - -section .pdata rdata align=4 - dd main wrt ..imagebase - dd main_end wrt ..imagebase - dd xmain wrt ..imagebase -section .xdata rdata align=8 -xmain: db 9,0,0,0 - dd handler wrt ..imagebase -section .drectve info - db '/defaultlib:user32.lib /defaultlib:msvcrt.lib ' -</pre> -<p>What you see in <code><nobr>.pdata</nobr></code> section is element of -the "table comprising start and end addresses of function" along with -reference to associated <code><nobr>UNWIND_INFO</nobr></code> structure. -And what you see in <code><nobr>.xdata</nobr></code> section is -<code><nobr>UNWIND_INFO</nobr></code> structure describing function with no -frame, but with designated exception handler. References are -<em>required</em> to be image-relative (which is the real reason for -implementing <code><nobr>wrt ..imagebase</nobr></code> operator). It should -be noted that <code><nobr>rdata align=n</nobr></code>, as well as -<code><nobr>wrt ..imagebase</nobr></code>, are optional in these two -segments' contexts, i.e. can be omitted. Latter means that <em>all</em> -32-bit references, not only above listed required ones, placed into these -two segments turn out image-relative. Why is it important to understand? -Developer is allowed to append handler-specific data to -<code><nobr>UNWIND_INFO</nobr></code> structure, and if [s]he adds a 32-bit -reference, then [s]he will have to remember to adjust its value to obtain -the real pointer. -<p>As already mentioned, in Win64 terms leaf function is one that does not -call any other function <em>nor</em> modifies any non-volatile register, -including stack pointer. But it's not uncommon that assembler programmer -plans to utilize every single register and sometimes even have variable -stack frame. Is there anything one can do with bare building blocks? I.e. -besides manually composing fully-fledged -<code><nobr>UNWIND_INFO</nobr></code> structure, which would surely be -considered error-prone? Yes, there is. Recall that exception handler is -called first, before stack layout is analyzed. As it turned out, it's -perfectly possible to manipulate current callee's context in custom handler -in manner that permits further stack unwinding. General idea is that -handler would not actually "handle" the exception, but instead restore -callee's context, as it was at its entry point and thus mimic leaf -function. In other words, handler would simply undertake part of unwinding -procedure. Consider following example: -<p><pre> -function: - mov rax,rsp ; copy rsp to volatile register - push r15 ; save non-volatile registers - push rbx - push rbp - mov r11,rsp ; prepare variable stack frame - sub r11,rcx - and r11,-64 - mov QWORD[r11],rax ; check for exceptions - mov rsp,r11 ; allocate stack frame - mov QWORD[rsp],rax ; save original rsp value -magic_point: - ... - mov r11,QWORD[rsp] ; pull original rsp value - mov rbp,QWORD[r11-24] - mov rbx,QWORD[r11-16] - mov r15,QWORD[r11-8] - mov rsp,r11 ; destroy frame - ret -</pre> -<p>The keyword is that up to <code><nobr>magic_point</nobr></code> original -<code><nobr>rsp</nobr></code> value remains in chosen volatile register and -no non-volatile register, except for <code><nobr>rsp</nobr></code>, is -modified. While past <code><nobr>magic_point</nobr></code> -<code><nobr>rsp</nobr></code> remains constant till the very end of the -<code><nobr>function</nobr></code>. In this case custom language-specific -exception handler would look like this: -<p><pre> -EXCEPTION_DISPOSITION handler (EXCEPTION_RECORD *rec,ULONG64 frame, - CONTEXT *context,DISPATCHER_CONTEXT *disp) -{ ULONG64 *rsp; - if (context->Rip<(ULONG64)magic_point) - rsp = (ULONG64 *)context->Rax; - else - { rsp = ((ULONG64 **)context->Rsp)[0]; - context->Rbp = rsp[-3]; - context->Rbx = rsp[-2]; - context->R15 = rsp[-1]; - } - context->Rsp = (ULONG64)rsp; - - memcpy (disp->ContextRecord,context,sizeof(CONTEXT)); - RtlVirtualUnwind(UNW_FLAG_NHANDLER,disp->ImageBase, - dips->ControlPc,disp->FunctionEntry,disp->ContextRecord, - &disp->HandlerData,&disp->EstablisherFrame,NULL); - return ExceptionContinueSearch; -} -</pre> -<p>As custom handler mimics leaf function, corresponding -<code><nobr>UNWIND_INFO</nobr></code> structure does not have to contain -any information about stack frame and its layout. -<h3><a name="section-7.7">7.7 <code><nobr>coff</nobr></code>: Common Object File Format</a></h3> -<p>The <code><nobr>coff</nobr></code> output type produces -<code><nobr>COFF</nobr></code> object files suitable for linking with the -DJGPP linker. -<p><code><nobr>coff</nobr></code> provides a default output file-name -extension of <code><nobr>.o</nobr></code>. -<p>The <code><nobr>coff</nobr></code> format supports the same extensions -to the <code><nobr>SECTION</nobr></code> directive as -<code><nobr>win32</nobr></code> does, except that the -<code><nobr>align</nobr></code> qualifier and the -<code><nobr>info</nobr></code> section type are not supported. -<h3><a name="section-7.8">7.8 <code><nobr>macho32</nobr></code> and <code><nobr>macho64</nobr></code>: Mach Object File Format</a></h3> -<p>The <code><nobr>macho32</nobr></code> and -<code><nobr>macho64</nobr></code> output formts produces -<code><nobr>Mach-O</nobr></code> object files suitable for linking with the -MacOS X linker. <code><nobr>macho</nobr></code> is a synonym for -<code><nobr>macho32</nobr></code>. -<p><code><nobr>macho</nobr></code> provides a default output file-name -extension of <code><nobr>.o</nobr></code>. -<h3><a name="section-7.9">7.9 <code><nobr>elf32</nobr></code> and <code><nobr>elf64</nobr></code>: Executable and Linkable Format Object Files</a></h3> -<p>The <code><nobr>elf32</nobr></code> and <code><nobr>elf64</nobr></code> -output formats generate <code><nobr>ELF32 and ELF64</nobr></code> -(Executable and Linkable Format) object files, as used by Linux as well as -Unix System V, including Solaris x86, UnixWare and SCO Unix. -<code><nobr>elf</nobr></code> provides a default output file-name extension -of <code><nobr>.o</nobr></code>. <code><nobr>elf</nobr></code> is a synonym -for <code><nobr>elf32</nobr></code>. -<h4><a name="section-7.9.1">7.9.1 ELF specific directive <code><nobr>osabi</nobr></code></a></h4> -<p>The ELF header specifies the application binary interface for the target -operating system (OSABI). This field can be set by using the -<code><nobr>osabi</nobr></code> directive with the numeric value (0-255) of -the target system. If this directive is not used, the default value will be -"UNIX System V ABI" (0) which will work on most systems which support ELF. -<h4><a name="section-7.9.2">7.9.2 <code><nobr>elf</nobr></code> Extensions to the <code><nobr>SECTION</nobr></code> Directive</a></h4> -<p>Like the <code><nobr>obj</nobr></code> format, -<code><nobr>elf</nobr></code> allows you to specify additional information -on the <code><nobr>SECTION</nobr></code> directive line, to control the -type and properties of sections you declare. Section types and properties -are generated automatically by NASM for the standard section names, but may -still be overridden by these qualifiers. -<p>The available qualifiers are: -<ul> -<li><code><nobr>alloc</nobr></code> defines the section to be one which is -loaded into memory when the program is run. -<code><nobr>noalloc</nobr></code> defines it to be one which is not, such -as an informational or comment section. -<li><code><nobr>exec</nobr></code> defines the section to be one which -should have execute permission when the program is run. -<code><nobr>noexec</nobr></code> defines it as one which should not. -<li><code><nobr>write</nobr></code> defines the section to be one which -should be writable when the program is run. -<code><nobr>nowrite</nobr></code> defines it as one which should not. -<li><code><nobr>progbits</nobr></code> defines the section to be one with -explicit contents stored in the object file: an ordinary code or data -section, for example, <code><nobr>nobits</nobr></code> defines the section -to be one with no explicit contents given, such as a BSS section. -<li><code><nobr>align=</nobr></code>, used with a trailing number as in -<code><nobr>obj</nobr></code>, gives the alignment requirements of the -section. -<li><code><nobr>tls</nobr></code> defines the section to be one which -contains thread local variables. -</ul> -<p>The defaults assumed by NASM if you do not specify the above qualifiers -are: -<p> -<p><pre> -section .text progbits alloc exec nowrite align=16 -section .rodata progbits alloc noexec nowrite align=4 -section .lrodata progbits alloc noexec nowrite align=4 -section .data progbits alloc noexec write align=4 -section .ldata progbits alloc noexec write align=4 -section .bss nobits alloc noexec write align=4 -section .lbss nobits alloc noexec write align=4 -section .tdata progbits alloc noexec write align=4 tls -section .tbss nobits alloc noexec write align=4 tls -section .comment progbits noalloc noexec nowrite align=1 -section other progbits alloc noexec nowrite align=1 -</pre> -<p>(Any section name other than those in the above table is treated by -default like <code><nobr>other</nobr></code> in the above table. Please -note that section names are case sensitive.) -<h4><a name="section-7.9.3">7.9.3 Position-Independent Code: <code><nobr>elf</nobr></code> Special Symbols and <code><nobr>WRT</nobr></code></a></h4> -<p>The <code><nobr>ELF</nobr></code> specification contains enough features -to allow position-independent code (PIC) to be written, which makes ELF -shared libraries very flexible. However, it also means NASM has to be able -to generate a variety of ELF specific relocation types in ELF object files, -if it is to be an assembler which can write PIC. -<p>Since <code><nobr>ELF</nobr></code> does not support segment-base -references, the <code><nobr>WRT</nobr></code> operator is not used for its -normal purpose; therefore NASM's <code><nobr>elf</nobr></code> output -format makes use of <code><nobr>WRT</nobr></code> for a different purpose, -namely the PIC-specific relocation types. -<p><code><nobr>elf</nobr></code> defines five special symbols which you can -use as the right-hand side of the <code><nobr>WRT</nobr></code> operator to -obtain PIC relocation types. They are <code><nobr>..gotpc</nobr></code>, -<code><nobr>..gotoff</nobr></code>, <code><nobr>..got</nobr></code>, -<code><nobr>..plt</nobr></code> and <code><nobr>..sym</nobr></code>. Their -functions are summarized here: -<ul> -<li>Referring to the symbol marking the global offset table base using -<code><nobr>wrt ..gotpc</nobr></code> will end up giving the distance from -the beginning of the current section to the global offset table. -(<code><nobr>_GLOBAL_OFFSET_TABLE_</nobr></code> is the standard symbol -name used to refer to the GOT.) So you would then need to add -<code><nobr>$$</nobr></code> to the result to get the real address of the -GOT. -<li>Referring to a location in one of your own sections using -<code><nobr>wrt ..gotoff</nobr></code> will give the distance from the -beginning of the GOT to the specified location, so that adding on the -address of the GOT would give the real address of the location you wanted. -<li>Referring to an external or global symbol using -<code><nobr>wrt ..got</nobr></code> causes the linker to build an entry -<em>in</em> the GOT containing the address of the symbol, and the reference -gives the distance from the beginning of the GOT to the entry; so you can -add on the address of the GOT, load from the resulting address, and end up -with the address of the symbol. -<li>Referring to a procedure name using <code><nobr>wrt ..plt</nobr></code> -causes the linker to build a procedure linkage table entry for the symbol, -and the reference gives the address of the PLT entry. You can only use this -in contexts which would generate a PC-relative relocation normally (i.e. as -the destination for <code><nobr>CALL</nobr></code> or -<code><nobr>JMP</nobr></code>), since ELF contains no relocation type to -refer to PLT entries absolutely. -<li>Referring to a symbol name using <code><nobr>wrt ..sym</nobr></code> -causes NASM to write an ordinary relocation, but instead of making the -relocation relative to the start of the section and then adding on the -offset to the symbol, it will write a relocation record aimed directly at -the symbol in question. The distinction is a necessary one due to a -peculiarity of the dynamic linker. -</ul> -<p>A fuller explanation of how to use these relocation types to write -shared libraries entirely in NASM is given in -<a href="nasmdoc9.html#section-9.2">section 9.2</a>. -<h4><a name="section-7.9.4">7.9.4 Thread Local Storage: <code><nobr>elf</nobr></code> Special Symbols and <code><nobr>WRT</nobr></code></a></h4> -<ul> -<li>In ELF32 mode, referring to an external or global symbol using -<code><nobr>wrt ..tlsie</nobr></code> causes the linker to build an entry -<em>in</em> the GOT containing the offset of the symbol within the TLS -block, so you can access the value of the symbol with code such as: -</ul> -<p><pre> - mov eax,[tid wrt ..tlsie] - mov [gs:eax],ebx -</pre> -<ul> -<li>In ELF64 mode, referring to an external or global symbol using -<code><nobr>wrt ..gottpoff</nobr></code> causes the linker to build an -entry <em>in</em> the GOT containing the offset of the symbol within the -TLS block, so you can access the value of the symbol with code such as: -</ul> -<p><pre> - mov rax,[rel tid wrt ..gottpoff] - mov rcx,[fs:rax] -</pre> -<h4><a name="section-7.9.5">7.9.5 <code><nobr>elf</nobr></code> Extensions to the <code><nobr>GLOBAL</nobr></code> Directive</a></h4> -<p><code><nobr>ELF</nobr></code> object files can contain more information -about a global symbol than just its address: they can contain the size of -the symbol and its type as well. These are not merely debugger -conveniences, but are actually necessary when the program being written is -a shared library. NASM therefore supports some extensions to the -<code><nobr>GLOBAL</nobr></code> directive, allowing you to specify these -features. -<p>You can specify whether a global variable is a function or a data object -by suffixing the name with a colon and the word -<code><nobr>function</nobr></code> or <code><nobr>data</nobr></code>. -(<code><nobr>object</nobr></code> is a synonym for -<code><nobr>data</nobr></code>.) For example: -<p><pre> -global hashlookup:function, hashtable:data -</pre> -<p>exports the global symbol <code><nobr>hashlookup</nobr></code> as a -function and <code><nobr>hashtable</nobr></code> as a data object. -<p>Optionally, you can control the ELF visibility of the symbol. Just add -one of the visibility keywords: <code><nobr>default</nobr></code>, -<code><nobr>internal</nobr></code>, <code><nobr>hidden</nobr></code>, or -<code><nobr>protected</nobr></code>. The default is -<code><nobr>default</nobr></code> of course. For example, to make -<code><nobr>hashlookup</nobr></code> hidden: -<p><pre> -global hashlookup:function hidden -</pre> -<p>You can also specify the size of the data associated with the symbol, as -a numeric expression (which may involve labels, and even forward -references) after the type specifier. Like this: -<p><pre> -global hashtable:data (hashtable.end - hashtable) - -hashtable: - db this,that,theother ; some data here -.end: -</pre> -<p>This makes NASM automatically calculate the length of the table and -place that information into the <code><nobr>ELF</nobr></code> symbol table. -<p>Declaring the type and size of global symbols is necessary when writing -shared library code. For more information, see -<a href="nasmdoc9.html#section-9.2.4">section 9.2.4</a>. -<h4><a name="section-7.9.6">7.9.6 <code><nobr>elf</nobr></code> Extensions to the <code><nobr>COMMON</nobr></code> Directive </a></h4> -<p><code><nobr>ELF</nobr></code> also allows you to specify alignment -requirements on common variables. This is done by putting a number (which -must be a power of two) after the name and size of the common variable, -separated (as usual) by a colon. For example, an array of doublewords would -benefit from 4-byte alignment: -<p><pre> -common dwordarray 128:4 -</pre> -<p>This declares the total size of the array to be 128 bytes, and requires -that it be aligned on a 4-byte boundary. -<h4><a name="section-7.9.7">7.9.7 16-bit code and ELF </a></h4> -<p>The <code><nobr>ELF32</nobr></code> specification doesn't provide -relocations for 8- and 16-bit values, but the GNU -<code><nobr>ld</nobr></code> linker adds these as an extension. NASM can -generate GNU-compatible relocations, to allow 16-bit code to be linked as -ELF using GNU <code><nobr>ld</nobr></code>. If NASM is used with the -<code><nobr>-w+gnu-elf-extensions</nobr></code> option, a warning is issued -when one of these relocations is generated. -<h4><a name="section-7.9.8">7.9.8 Debug formats and ELF </a></h4> -<p><code><nobr>ELF32</nobr></code> and <code><nobr>ELF64</nobr></code> -provide debug information in <code><nobr>STABS</nobr></code> and -<code><nobr>DWARF</nobr></code> formats. Line number information is -generated for all executable sections, but please note that only the -".text" section is executable by default. -<h3><a name="section-7.10">7.10 <code><nobr>aout</nobr></code>: Linux <code><nobr>a.out</nobr></code> Object Files</a></h3> -<p>The <code><nobr>aout</nobr></code> format generates -<code><nobr>a.out</nobr></code> object files, in the form used by early -Linux systems (current Linux systems use ELF, see -<a href="#section-7.9">section 7.9</a>.) These differ from other -<code><nobr>a.out</nobr></code> object files in that the magic number in -the first four bytes of the file is different; also, some implementations -of <code><nobr>a.out</nobr></code>, for example NetBSD's, support -position-independent code, which Linux's implementation does not. -<p><code><nobr>a.out</nobr></code> provides a default output file-name -extension of <code><nobr>.o</nobr></code>. -<p><code><nobr>a.out</nobr></code> is a very simple object format. It -supports no special directives, no special symbols, no use of -<code><nobr>SEG</nobr></code> or <code><nobr>WRT</nobr></code>, and no -extensions to any standard directives. It supports only the three standard -section names <code><nobr>.text</nobr></code>, -<code><nobr>.data</nobr></code> and <code><nobr>.bss</nobr></code>. -<h3><a name="section-7.11">7.11 <code><nobr>aoutb</nobr></code>: NetBSD/FreeBSD/OpenBSD <code><nobr>a.out</nobr></code> Object Files</a></h3> -<p>The <code><nobr>aoutb</nobr></code> format generates -<code><nobr>a.out</nobr></code> object files, in the form used by the -various free <code><nobr>BSD Unix</nobr></code> clones, -<code><nobr>NetBSD</nobr></code>, <code><nobr>FreeBSD</nobr></code> and -<code><nobr>OpenBSD</nobr></code>. For simple object files, this object -format is exactly the same as <code><nobr>aout</nobr></code> except for the -magic number in the first four bytes of the file. However, the -<code><nobr>aoutb</nobr></code> format supports position-independent code -in the same way as the <code><nobr>elf</nobr></code> format, so you can use -it to write <code><nobr>BSD</nobr></code> shared libraries. -<p><code><nobr>aoutb</nobr></code> provides a default output file-name -extension of <code><nobr>.o</nobr></code>. -<p><code><nobr>aoutb</nobr></code> supports no special directives, no -special symbols, and only the three standard section names -<code><nobr>.text</nobr></code>, <code><nobr>.data</nobr></code> and -<code><nobr>.bss</nobr></code>. However, it also supports the same use of -<code><nobr>WRT</nobr></code> as <code><nobr>elf</nobr></code> does, to -provide position-independent code relocation types. See -<a href="#section-7.9.3">section 7.9.3</a> for full documentation of this -feature. -<p><code><nobr>aoutb</nobr></code> also supports the same extensions to the -<code><nobr>GLOBAL</nobr></code> directive as <code><nobr>elf</nobr></code> -does: see <a href="#section-7.9.5">section 7.9.5</a> for documentation of -this. -<h3><a name="section-7.12">7.12 <code><nobr>as86</nobr></code>: Minix/Linux <code><nobr>as86</nobr></code> Object Files</a></h3> -<p>The Minix/Linux 16-bit assembler <code><nobr>as86</nobr></code> has its -own non-standard object file format. Although its companion linker -<code><nobr>ld86</nobr></code> produces something close to ordinary -<code><nobr>a.out</nobr></code> binaries as output, the object file format -used to communicate between <code><nobr>as86</nobr></code> and -<code><nobr>ld86</nobr></code> is not itself -<code><nobr>a.out</nobr></code>. -<p>NASM supports this format, just in case it is useful, as -<code><nobr>as86</nobr></code>. <code><nobr>as86</nobr></code> provides a -default output file-name extension of <code><nobr>.o</nobr></code>. -<p><code><nobr>as86</nobr></code> is a very simple object format (from the -NASM user's point of view). It supports no special directives, no use of -<code><nobr>SEG</nobr></code> or <code><nobr>WRT</nobr></code>, and no -extensions to any standard directives. It supports only the three standard -section names <code><nobr>.text</nobr></code>, -<code><nobr>.data</nobr></code> and <code><nobr>.bss</nobr></code>. The -only special symbol supported is <code><nobr>..start</nobr></code>. -<h3><a name="section-7.13">7.13 <code><nobr>rdf</nobr></code>: Relocatable Dynamic Object File Format</a></h3> -<p>The <code><nobr>rdf</nobr></code> output format produces -<code><nobr>RDOFF</nobr></code> object files. -<code><nobr>RDOFF</nobr></code> (Relocatable Dynamic Object File Format) is -a home-grown object-file format, designed alongside NASM itself and -reflecting in its file format the internal structure of the assembler. -<p><code><nobr>RDOFF</nobr></code> is not used by any well-known operating -systems. Those writing their own systems, however, may well wish to use -<code><nobr>RDOFF</nobr></code> as their object format, on the grounds that -it is designed primarily for simplicity and contains very little -file-header bureaucracy. -<p>The Unix NASM archive, and the DOS archive which includes sources, both -contain an <code><nobr>rdoff</nobr></code> subdirectory holding a set of -RDOFF utilities: an RDF linker, an <code><nobr>RDF</nobr></code> -static-library manager, an RDF file dump utility, and a program which will -load and execute an RDF executable under Linux. -<p><code><nobr>rdf</nobr></code> supports only the standard section names -<code><nobr>.text</nobr></code>, <code><nobr>.data</nobr></code> and -<code><nobr>.bss</nobr></code>. -<h4><a name="section-7.13.1">7.13.1 Requiring a Library: The <code><nobr>LIBRARY</nobr></code> Directive</a></h4> -<p><code><nobr>RDOFF</nobr></code> contains a mechanism for an object file -to demand a given library to be linked to the module, either at load time -or run time. This is done by the <code><nobr>LIBRARY</nobr></code> -directive, which takes one argument which is the name of the module: -<p><pre> - library mylib.rdl -</pre> -<h4><a name="section-7.13.2">7.13.2 Specifying a Module Name: The <code><nobr>MODULE</nobr></code> Directive</a></h4> -<p>Special <code><nobr>RDOFF</nobr></code> header record is used to store -the name of the module. It can be used, for example, by run-time loader to -perform dynamic linking. <code><nobr>MODULE</nobr></code> directive takes -one argument which is the name of current module: -<p><pre> - module mymodname -</pre> -<p>Note that when you statically link modules and tell linker to strip the -symbols from output file, all module names will be stripped too. To avoid -it, you should start module names with <code><nobr>$</nobr></code>, like: -<p><pre> - module $kernel.core -</pre> -<h4><a name="section-7.13.3">7.13.3 <code><nobr>rdf</nobr></code> Extensions to the <code><nobr>GLOBAL</nobr></code> Directive</a></h4> -<p><code><nobr>RDOFF</nobr></code> global symbols can contain additional -information needed by the static linker. You can mark a global symbol as -exported, thus telling the linker do not strip it from target executable or -library file. Like in <code><nobr>ELF</nobr></code>, you can also specify -whether an exported symbol is a procedure (function) or data object. -<p>Suffixing the name with a colon and the word -<code><nobr>export</nobr></code> you make the symbol exported: -<p><pre> - global sys_open:export -</pre> -<p>To specify that exported symbol is a procedure (function), you add the -word <code><nobr>proc</nobr></code> or <code><nobr>function</nobr></code> -after declaration: -<p><pre> - global sys_open:export proc -</pre> -<p>Similarly, to specify exported data object, add the word -<code><nobr>data</nobr></code> or <code><nobr>object</nobr></code> to the -directive: -<p><pre> - global kernel_ticks:export data -</pre> -<h4><a name="section-7.13.4">7.13.4 <code><nobr>rdf</nobr></code> Extensions to the <code><nobr>EXTERN</nobr></code> Directive</a></h4> -<p>By default the <code><nobr>EXTERN</nobr></code> directive in -<code><nobr>RDOFF</nobr></code> declares a "pure external" symbol (i.e. the -static linker will complain if such a symbol is not resolved). To declare -an "imported" symbol, which must be resolved later during a dynamic linking -phase, <code><nobr>RDOFF</nobr></code> offers an additional -<code><nobr>import</nobr></code> modifier. As in -<code><nobr>GLOBAL</nobr></code>, you can also specify whether an imported -symbol is a procedure (function) or data object. For example: -<p><pre> - library $libc - extern _open:import - extern _printf:import proc - extern _errno:import data -</pre> -<p>Here the directive <code><nobr>LIBRARY</nobr></code> is also included, -which gives the dynamic linker a hint as to where to find requested -symbols. -<h3><a name="section-7.14">7.14 <code><nobr>dbg</nobr></code>: Debugging Format</a></h3> -<p>The <code><nobr>dbg</nobr></code> output format is not built into NASM -in the default configuration. If you are building your own NASM executable -from the sources, you can define <code><nobr>OF_DBG</nobr></code> in -<code><nobr>output/outform.h</nobr></code> or on the compiler command line, -and obtain the <code><nobr>dbg</nobr></code> output format. -<p>The <code><nobr>dbg</nobr></code> format does not output an object file -as such; instead, it outputs a text file which contains a complete list of -all the transactions between the main body of NASM and the output-format -back end module. It is primarily intended to aid people who want to write -their own output drivers, so that they can get a clearer idea of the -various requests the main program makes of the output driver, and in what -order they happen. -<p>For simple files, one can easily use the <code><nobr>dbg</nobr></code> -format like this: -<p><pre> -nasm -f dbg filename.asm -</pre> -<p>which will generate a diagnostic file called -<code><nobr>filename.dbg</nobr></code>. However, this will not work well on -files which were designed for a different object format, because each -object format defines its own macros (usually user-level forms of -directives), and those macros will not be defined in the -<code><nobr>dbg</nobr></code> format. Therefore it can be useful to run -NASM twice, in order to do the preprocessing with the native object format -selected: -<p><pre> -nasm -e -f rdf -o rdfprog.i rdfprog.asm -nasm -a -f dbg rdfprog.i -</pre> -<p>This preprocesses <code><nobr>rdfprog.asm</nobr></code> into -<code><nobr>rdfprog.i</nobr></code>, keeping the -<code><nobr>rdf</nobr></code> object format selected in order to make sure -RDF special directives are converted into primitive form correctly. Then -the preprocessed source is fed through the <code><nobr>dbg</nobr></code> -format to generate the final diagnostic output. -<p>This workaround will still typically not work for programs intended for -<code><nobr>obj</nobr></code> format, because the -<code><nobr>obj</nobr></code> <code><nobr>SEGMENT</nobr></code> and -<code><nobr>GROUP</nobr></code> directives have side effects of defining -the segment and group names as symbols; <code><nobr>dbg</nobr></code> will -not do this, so the program will not assemble. You will have to work around -that by defining the symbols yourself (using -<code><nobr>EXTERN</nobr></code>, for example) if you really need to get a -<code><nobr>dbg</nobr></code> trace of an -<code><nobr>obj</nobr></code>-specific source file. -<p><code><nobr>dbg</nobr></code> accepts any section name and any -directives at all, and logs them all to its output file. -<p align=center><a href="nasmdoc8.html">Next Chapter</a> | -<a href="nasmdoc6.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -</body></html> diff --git a/doc/html/nasmdoc8.html b/doc/html/nasmdoc8.html deleted file mode 100644 index a7c2869..0000000 --- a/doc/html/nasmdoc8.html +++ /dev/null @@ -1,808 +0,0 @@ -<html><head><title>NASM Manual</title></head> -<body><h1 align=center>The Netwide Assembler: NASM</h1> - -<p align=center><a href="nasmdoc9.html">Next Chapter</a> | -<a href="nasmdoc7.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -<h2><a name="chapter-8">Chapter 8: Writing 16-bit Code (DOS, Windows 3/3.1)</a></h2> -<p>This chapter attempts to cover some of the common issues encountered -when writing 16-bit code to run under <code><nobr>MS-DOS</nobr></code> or -<code><nobr>Windows 3.x</nobr></code>. It covers how to link programs to -produce <code><nobr>.EXE</nobr></code> or <code><nobr>.COM</nobr></code> -files, how to write <code><nobr>.SYS</nobr></code> device drivers, and how -to interface assembly language code with 16-bit C compilers and with -Borland Pascal. -<h3><a name="section-8.1">8.1 Producing <code><nobr>.EXE</nobr></code> Files</a></h3> -<p>Any large program written under DOS needs to be built as a -<code><nobr>.EXE</nobr></code> file: only <code><nobr>.EXE</nobr></code> -files have the necessary internal structure required to span more than one -64K segment. Windows programs, also, have to be built as -<code><nobr>.EXE</nobr></code> files, since Windows does not support the -<code><nobr>.COM</nobr></code> format. -<p>In general, you generate <code><nobr>.EXE</nobr></code> files by using -the <code><nobr>obj</nobr></code> output format to produce one or more -<code><nobr>.OBJ</nobr></code> files, and then linking them together using -a linker. However, NASM also supports the direct generation of simple DOS -<code><nobr>.EXE</nobr></code> files using the -<code><nobr>bin</nobr></code> output format (by using -<code><nobr>DB</nobr></code> and <code><nobr>DW</nobr></code> to construct -the <code><nobr>.EXE</nobr></code> file header), and a macro package is -supplied to do this. Thanks to Yann Guidon for contributing the code for -this. -<p>NASM may also support <code><nobr>.EXE</nobr></code> natively as another -output format in future releases. -<h4><a name="section-8.1.1">8.1.1 Using the <code><nobr>obj</nobr></code> Format To Generate <code><nobr>.EXE</nobr></code> Files</a></h4> -<p>This section describes the usual method of generating -<code><nobr>.EXE</nobr></code> files by linking -<code><nobr>.OBJ</nobr></code> files together. -<p>Most 16-bit programming language packages come with a suitable linker; -if you have none of these, there is a free linker called VAL, available in -<code><nobr>LZH</nobr></code> archive format from -<a href="ftp://x2ftp.oulu.fi/pub/msdos/programming/lang/"><code><nobr>x2ftp.oulu.fi</nobr></code></a>. -An LZH archiver can be found at -<a href="ftp://ftp.simtel.net/pub/simtelnet/msdos/arcers"><code><nobr>ftp.simtel.net</nobr></code></a>. -There is another `free' linker (though this one doesn't come with sources) -called FREELINK, available from -<a href="http://www.pcorner.com/tpc/old/3-101.html"><code><nobr>www.pcorner.com</nobr></code></a>. -A third, <code><nobr>djlink</nobr></code>, written by DJ Delorie, is -available at -<a href="http://www.delorie.com/djgpp/16bit/djlink/"><code><nobr>www.delorie.com</nobr></code></a>. -A fourth linker, <code><nobr>ALINK</nobr></code>, written by Anthony A.J. -Williams, is available at -<a href="http://alink.sourceforge.net"><code><nobr>alink.sourceforge.net</nobr></code></a>. -<p>When linking several <code><nobr>.OBJ</nobr></code> files into a -<code><nobr>.EXE</nobr></code> file, you should ensure that exactly one of -them has a start point defined (using the <code><nobr>..start</nobr></code> -special symbol defined by the <code><nobr>obj</nobr></code> format: see -<a href="nasmdoc7.html#section-7.4.6">section 7.4.6</a>). If no module -defines a start point, the linker will not know what value to give the -entry-point field in the output file header; if more than one defines a -start point, the linker will not know <em>which</em> value to use. -<p>An example of a NASM source file which can be assembled to a -<code><nobr>.OBJ</nobr></code> file and linked on its own to a -<code><nobr>.EXE</nobr></code> is given here. It demonstrates the basic -principles of defining a stack, initialising the segment registers, and -declaring a start point. This file is also provided in the -<code><nobr>test</nobr></code> subdirectory of the NASM archives, under the -name <code><nobr>objexe.asm</nobr></code>. -<p><pre> -segment code - -..start: - mov ax,data - mov ds,ax - mov ax,stack - mov ss,ax - mov sp,stacktop -</pre> -<p>This initial piece of code sets up <code><nobr>DS</nobr></code> to point -to the data segment, and initializes <code><nobr>SS</nobr></code> and -<code><nobr>SP</nobr></code> to point to the top of the provided stack. -Notice that interrupts are implicitly disabled for one instruction after a -move into <code><nobr>SS</nobr></code>, precisely for this situation, so -that there's no chance of an interrupt occurring between the loads of -<code><nobr>SS</nobr></code> and <code><nobr>SP</nobr></code> and not -having a stack to execute on. -<p>Note also that the special symbol <code><nobr>..start</nobr></code> is -defined at the beginning of this code, which means that will be the entry -point into the resulting executable file. -<p><pre> - mov dx,hello - mov ah,9 - int 0x21 -</pre> -<p>The above is the main program: load <code><nobr>DS:DX</nobr></code> with -a pointer to the greeting message (<code><nobr>hello</nobr></code> is -implicitly relative to the segment <code><nobr>data</nobr></code>, which -was loaded into <code><nobr>DS</nobr></code> in the setup code, so the full -pointer is valid), and call the DOS print-string function. -<p><pre> - mov ax,0x4c00 - int 0x21 -</pre> -<p>This terminates the program using another DOS system call. -<p><pre> -segment data - -hello: db 'hello, world', 13, 10, '$' -</pre> -<p>The data segment contains the string we want to display. -<p><pre> -segment stack stack - resb 64 -stacktop: -</pre> -<p>The above code declares a stack segment containing 64 bytes of -uninitialized stack space, and points <code><nobr>stacktop</nobr></code> at -the top of it. The directive <code><nobr>segment stack stack</nobr></code> -defines a segment <em>called</em> <code><nobr>stack</nobr></code>, and also -of <em>type</em> <code><nobr>STACK</nobr></code>. The latter is not -necessary to the correct running of the program, but linkers are likely to -issue warnings or errors if your program has no segment of type -<code><nobr>STACK</nobr></code>. -<p>The above file, when assembled into a <code><nobr>.OBJ</nobr></code> -file, will link on its own to a valid <code><nobr>.EXE</nobr></code> file, -which when run will print `hello, world' and then exit. -<h4><a name="section-8.1.2">8.1.2 Using the <code><nobr>bin</nobr></code> Format To Generate <code><nobr>.EXE</nobr></code> Files</a></h4> -<p>The <code><nobr>.EXE</nobr></code> file format is simple enough that -it's possible to build a <code><nobr>.EXE</nobr></code> file by writing a -pure-binary program and sticking a 32-byte header on the front. This header -is simple enough that it can be generated using -<code><nobr>DB</nobr></code> and <code><nobr>DW</nobr></code> commands by -NASM itself, so that you can use the <code><nobr>bin</nobr></code> output -format to directly generate <code><nobr>.EXE</nobr></code> files. -<p>Included in the NASM archives, in the <code><nobr>misc</nobr></code> -subdirectory, is a file <code><nobr>exebin.mac</nobr></code> of macros. It -defines three macros: <code><nobr>EXE_begin</nobr></code>, -<code><nobr>EXE_stack</nobr></code> and <code><nobr>EXE_end</nobr></code>. -<p>To produce a <code><nobr>.EXE</nobr></code> file using this method, you -should start by using <code><nobr>%include</nobr></code> to load the -<code><nobr>exebin.mac</nobr></code> macro package into your source file. -You should then issue the <code><nobr>EXE_begin</nobr></code> macro call -(which takes no arguments) to generate the file header data. Then write -code as normal for the <code><nobr>bin</nobr></code> format - you can use -all three standard sections <code><nobr>.text</nobr></code>, -<code><nobr>.data</nobr></code> and <code><nobr>.bss</nobr></code>. At the -end of the file you should call the <code><nobr>EXE_end</nobr></code> macro -(again, no arguments), which defines some symbols to mark section sizes, -and these symbols are referred to in the header code generated by -<code><nobr>EXE_begin</nobr></code>. -<p>In this model, the code you end up writing starts at -<code><nobr>0x100</nobr></code>, just like a <code><nobr>.COM</nobr></code> -file - in fact, if you strip off the 32-byte header from the resulting -<code><nobr>.EXE</nobr></code> file, you will have a valid -<code><nobr>.COM</nobr></code> program. All the segment bases are the same, -so you are limited to a 64K program, again just like a -<code><nobr>.COM</nobr></code> file. Note that an -<code><nobr>ORG</nobr></code> directive is issued by the -<code><nobr>EXE_begin</nobr></code> macro, so you should not explicitly -issue one of your own. -<p>You can't directly refer to your segment base value, unfortunately, -since this would require a relocation in the header, and things would get a -lot more complicated. So you should get your segment base by copying it out -of <code><nobr>CS</nobr></code> instead. -<p>On entry to your <code><nobr>.EXE</nobr></code> file, -<code><nobr>SS:SP</nobr></code> are already set up to point to the top of a -2Kb stack. You can adjust the default stack size of 2Kb by calling the -<code><nobr>EXE_stack</nobr></code> macro. For example, to change the stack -size of your program to 64 bytes, you would call -<code><nobr>EXE_stack 64</nobr></code>. -<p>A sample program which generates a <code><nobr>.EXE</nobr></code> file -in this way is given in the <code><nobr>test</nobr></code> subdirectory of -the NASM archive, as <code><nobr>binexe.asm</nobr></code>. -<h3><a name="section-8.2">8.2 Producing <code><nobr>.COM</nobr></code> Files</a></h3> -<p>While large DOS programs must be written as -<code><nobr>.EXE</nobr></code> files, small ones are often better written -as <code><nobr>.COM</nobr></code> files. <code><nobr>.COM</nobr></code> -files are pure binary, and therefore most easily produced using the -<code><nobr>bin</nobr></code> output format. -<h4><a name="section-8.2.1">8.2.1 Using the <code><nobr>bin</nobr></code> Format To Generate <code><nobr>.COM</nobr></code> Files</a></h4> -<p><code><nobr>.COM</nobr></code> files expect to be loaded at offset -<code><nobr>100h</nobr></code> into their segment (though the segment may -change). Execution then begins at <code><nobr>100h</nobr></code>, i.e. -right at the start of the program. So to write a -<code><nobr>.COM</nobr></code> program, you would create a source file -looking like -<p><pre> - org 100h - -section .text - -start: - ; put your code here - -section .data - - ; put data items here - -section .bss - - ; put uninitialized data here -</pre> -<p>The <code><nobr>bin</nobr></code> format puts the -<code><nobr>.text</nobr></code> section first in the file, so you can -declare data or BSS items before beginning to write code if you want to and -the code will still end up at the front of the file where it belongs. -<p>The BSS (uninitialized data) section does not take up space in the -<code><nobr>.COM</nobr></code> file itself: instead, addresses of BSS items -are resolved to point at space beyond the end of the file, on the grounds -that this will be free memory when the program is run. Therefore you should -not rely on your BSS being initialized to all zeros when you run. -<p>To assemble the above program, you should use a command line like -<p><pre> -nasm myprog.asm -fbin -o myprog.com -</pre> -<p>The <code><nobr>bin</nobr></code> format would produce a file called -<code><nobr>myprog</nobr></code> if no explicit output file name were -specified, so you have to override it and give the desired file name. -<h4><a name="section-8.2.2">8.2.2 Using the <code><nobr>obj</nobr></code> Format To Generate <code><nobr>.COM</nobr></code> Files</a></h4> -<p>If you are writing a <code><nobr>.COM</nobr></code> program as more than -one module, you may wish to assemble several <code><nobr>.OBJ</nobr></code> -files and link them together into a <code><nobr>.COM</nobr></code> program. -You can do this, provided you have a linker capable of outputting -<code><nobr>.COM</nobr></code> files directly (TLINK does this), or -alternatively a converter program such as <code><nobr>EXE2BIN</nobr></code> -to transform the <code><nobr>.EXE</nobr></code> file output from the linker -into a <code><nobr>.COM</nobr></code> file. -<p>If you do this, you need to take care of several things: -<ul> -<li>The first object file containing code should start its code segment -with a line like <code><nobr>RESB 100h</nobr></code>. This is to ensure -that the code begins at offset <code><nobr>100h</nobr></code> relative to -the beginning of the code segment, so that the linker or converter program -does not have to adjust address references within the file when generating -the <code><nobr>.COM</nobr></code> file. Other assemblers use an -<code><nobr>ORG</nobr></code> directive for this purpose, but -<code><nobr>ORG</nobr></code> in NASM is a format-specific directive to the -<code><nobr>bin</nobr></code> output format, and does not mean the same -thing as it does in MASM-compatible assemblers. -<li>You don't need to define a stack segment. -<li>All your segments should be in the same group, so that every time your -code or data references a symbol offset, all offsets are relative to the -same segment base. This is because, when a <code><nobr>.COM</nobr></code> -file is loaded, all the segment registers contain the same value. -</ul> -<h3><a name="section-8.3">8.3 Producing <code><nobr>.SYS</nobr></code> Files</a></h3> -<p>MS-DOS device drivers - <code><nobr>.SYS</nobr></code> files - are pure -binary files, similar to <code><nobr>.COM</nobr></code> files, except that -they start at origin zero rather than <code><nobr>100h</nobr></code>. -Therefore, if you are writing a device driver using the -<code><nobr>bin</nobr></code> format, you do not need the -<code><nobr>ORG</nobr></code> directive, since the default origin for -<code><nobr>bin</nobr></code> is zero. Similarly, if you are using -<code><nobr>obj</nobr></code>, you do not need the -<code><nobr>RESB 100h</nobr></code> at the start of your code segment. -<p><code><nobr>.SYS</nobr></code> files start with a header structure, -containing pointers to the various routines inside the driver which do the -work. This structure should be defined at the start of the code segment, -even though it is not actually code. -<p>For more information on the format of <code><nobr>.SYS</nobr></code> -files, and the data which has to go in the header structure, a list of -books is given in the Frequently Asked Questions list for the newsgroup -<a href="news:comp.os.msdos.programmer"><code><nobr>comp.os.msdos.programmer</nobr></code></a>. -<h3><a name="section-8.4">8.4 Interfacing to 16-bit C Programs</a></h3> -<p>This section covers the basics of writing assembly routines that call, -or are called from, C programs. To do this, you would typically write an -assembly module as a <code><nobr>.OBJ</nobr></code> file, and link it with -your C modules to produce a mixed-language program. -<h4><a name="section-8.4.1">8.4.1 External Symbol Names</a></h4> -<p>C compilers have the convention that the names of all global symbols -(functions or data) they define are formed by prefixing an underscore to -the name as it appears in the C program. So, for example, the function a C -programmer thinks of as <code><nobr>printf</nobr></code> appears to an -assembly language programmer as <code><nobr>_printf</nobr></code>. This -means that in your assembly programs, you can define symbols without a -leading underscore, and not have to worry about name clashes with C -symbols. -<p>If you find the underscores inconvenient, you can define macros to -replace the <code><nobr>GLOBAL</nobr></code> and -<code><nobr>EXTERN</nobr></code> directives as follows: -<p><pre> -%macro cglobal 1 - - global _%1 - %define %1 _%1 - -%endmacro - -%macro cextern 1 - - extern _%1 - %define %1 _%1 - -%endmacro -</pre> -<p>(These forms of the macros only take one argument at a time; a -<code><nobr>%rep</nobr></code> construct could solve this.) -<p>If you then declare an external like this: -<p><pre> -cextern printf -</pre> -<p>then the macro will expand it as -<p><pre> -extern _printf -%define printf _printf -</pre> -<p>Thereafter, you can reference <code><nobr>printf</nobr></code> as if it -was a symbol, and the preprocessor will put the leading underscore on where -necessary. -<p>The <code><nobr>cglobal</nobr></code> macro works similarly. You must -use <code><nobr>cglobal</nobr></code> before defining the symbol in -question, but you would have had to do that anyway if you used -<code><nobr>GLOBAL</nobr></code>. -<p>Also see <a href="nasmdoc2.html#section-2.1.27">section 2.1.27</a>. -<h4><a name="section-8.4.2">8.4.2 Memory Models</a></h4> -<p>NASM contains no mechanism to support the various C memory models -directly; you have to keep track yourself of which one you are writing for. -This means you have to keep track of the following things: -<ul> -<li>In models using a single code segment (tiny, small and compact), -functions are near. This means that function pointers, when stored in data -segments or pushed on the stack as function arguments, are 16 bits long and -contain only an offset field (the <code><nobr>CS</nobr></code> register -never changes its value, and always gives the segment part of the full -function address), and that functions are called using ordinary near -<code><nobr>CALL</nobr></code> instructions and return using -<code><nobr>RETN</nobr></code> (which, in NASM, is synonymous with -<code><nobr>RET</nobr></code> anyway). This means both that you should -write your own routines to return with <code><nobr>RETN</nobr></code>, and -that you should call external C routines with near -<code><nobr>CALL</nobr></code> instructions. -<li>In models using more than one code segment (medium, large and huge), -functions are far. This means that function pointers are 32 bits long -(consisting of a 16-bit offset followed by a 16-bit segment), and that -functions are called using <code><nobr>CALL FAR</nobr></code> (or -<code><nobr>CALL seg:offset</nobr></code>) and return using -<code><nobr>RETF</nobr></code>. Again, you should therefore write your own -routines to return with <code><nobr>RETF</nobr></code> and use -<code><nobr>CALL FAR</nobr></code> to call external routines. -<li>In models using a single data segment (tiny, small and medium), data -pointers are 16 bits long, containing only an offset field (the -<code><nobr>DS</nobr></code> register doesn't change its value, and always -gives the segment part of the full data item address). -<li>In models using more than one data segment (compact, large and huge), -data pointers are 32 bits long, consisting of a 16-bit offset followed by a -16-bit segment. You should still be careful not to modify -<code><nobr>DS</nobr></code> in your routines without restoring it -afterwards, but <code><nobr>ES</nobr></code> is free for you to use to -access the contents of 32-bit data pointers you are passed. -<li>The huge memory model allows single data items to exceed 64K in size. -In all other memory models, you can access the whole of a data item just by -doing arithmetic on the offset field of the pointer you are given, whether -a segment field is present or not; in huge model, you have to be more -careful of your pointer arithmetic. -<li>In most memory models, there is a <em>default</em> data segment, whose -segment address is kept in <code><nobr>DS</nobr></code> throughout the -program. This data segment is typically the same segment as the stack, kept -in <code><nobr>SS</nobr></code>, so that functions' local variables (which -are stored on the stack) and global data items can both be accessed easily -without changing <code><nobr>DS</nobr></code>. Particularly large data -items are typically stored in other segments. However, some memory models -(though not the standard ones, usually) allow the assumption that -<code><nobr>SS</nobr></code> and <code><nobr>DS</nobr></code> hold the same -value to be removed. Be careful about functions' local variables in this -latter case. -</ul> -<p>In models with a single code segment, the segment is called -<code><nobr>_TEXT</nobr></code>, so your code segment must also go by this -name in order to be linked into the same place as the main code segment. In -models with a single data segment, or with a default data segment, it is -called <code><nobr>_DATA</nobr></code>. -<h4><a name="section-8.4.3">8.4.3 Function Definitions and Function Calls</a></h4> -<p>The C calling convention in 16-bit programs is as follows. In the -following description, the words <em>caller</em> and <em>callee</em> are -used to denote the function doing the calling and the function which gets -called. -<ul> -<li>The caller pushes the function's parameters on the stack, one after -another, in reverse order (right to left, so that the first argument -specified to the function is pushed last). -<li>The caller then executes a <code><nobr>CALL</nobr></code> instruction -to pass control to the callee. This <code><nobr>CALL</nobr></code> is -either near or far depending on the memory model. -<li>The callee receives control, and typically (although this is not -actually necessary, in functions which do not need to access their -parameters) starts by saving the value of <code><nobr>SP</nobr></code> in -<code><nobr>BP</nobr></code> so as to be able to use -<code><nobr>BP</nobr></code> as a base pointer to find its parameters on -the stack. However, the caller was probably doing this too, so part of the -calling convention states that <code><nobr>BP</nobr></code> must be -preserved by any C function. Hence the callee, if it is going to set up -<code><nobr>BP</nobr></code> as a <em>frame pointer</em>, must push the -previous value first. -<li>The callee may then access its parameters relative to -<code><nobr>BP</nobr></code>. The word at <code><nobr>[BP]</nobr></code> -holds the previous value of <code><nobr>BP</nobr></code> as it was pushed; -the next word, at <code><nobr>[BP+2]</nobr></code>, holds the offset part -of the return address, pushed implicitly by <code><nobr>CALL</nobr></code>. -In a small-model (near) function, the parameters start after that, at -<code><nobr>[BP+4]</nobr></code>; in a large-model (far) function, the -segment part of the return address lives at -<code><nobr>[BP+4]</nobr></code>, and the parameters begin at -<code><nobr>[BP+6]</nobr></code>. The leftmost parameter of the function, -since it was pushed last, is accessible at this offset from -<code><nobr>BP</nobr></code>; the others follow, at successively greater -offsets. Thus, in a function such as <code><nobr>printf</nobr></code> which -takes a variable number of parameters, the pushing of the parameters in -reverse order means that the function knows where to find its first -parameter, which tells it the number and type of the remaining ones. -<li>The callee may also wish to decrease <code><nobr>SP</nobr></code> -further, so as to allocate space on the stack for local variables, which -will then be accessible at negative offsets from -<code><nobr>BP</nobr></code>. -<li>The callee, if it wishes to return a value to the caller, should leave -the value in <code><nobr>AL</nobr></code>, <code><nobr>AX</nobr></code> or -<code><nobr>DX:AX</nobr></code> depending on the size of the value. -Floating-point results are sometimes (depending on the compiler) returned -in <code><nobr>ST0</nobr></code>. -<li>Once the callee has finished processing, it restores -<code><nobr>SP</nobr></code> from <code><nobr>BP</nobr></code> if it had -allocated local stack space, then pops the previous value of -<code><nobr>BP</nobr></code>, and returns via -<code><nobr>RETN</nobr></code> or <code><nobr>RETF</nobr></code> depending -on memory model. -<li>When the caller regains control from the callee, the function -parameters are still on the stack, so it typically adds an immediate -constant to <code><nobr>SP</nobr></code> to remove them (instead of -executing a number of slow <code><nobr>POP</nobr></code> instructions). -Thus, if a function is accidentally called with the wrong number of -parameters due to a prototype mismatch, the stack will still be returned to -a sensible state since the caller, which <em>knows</em> how many parameters -it pushed, does the removing. -</ul> -<p>It is instructive to compare this calling convention with that for -Pascal programs (described in <a href="#section-8.5.1">section 8.5.1</a>). -Pascal has a simpler convention, since no functions have variable numbers -of parameters. Therefore the callee knows how many parameters it should -have been passed, and is able to deallocate them from the stack itself by -passing an immediate argument to the <code><nobr>RET</nobr></code> or -<code><nobr>RETF</nobr></code> instruction, so the caller does not have to -do it. Also, the parameters are pushed in left-to-right order, not -right-to-left, which means that a compiler can give better guarantees about -sequence points without performance suffering. -<p>Thus, you would define a function in C style in the following way. The -following example is for small model: -<p><pre> -global _myfunc - -_myfunc: - push bp - mov bp,sp - sub sp,0x40 ; 64 bytes of local stack space - mov bx,[bp+4] ; first parameter to function - - ; some more code - - mov sp,bp ; undo "sub sp,0x40" above - pop bp - ret -</pre> -<p>For a large-model function, you would replace -<code><nobr>RET</nobr></code> by <code><nobr>RETF</nobr></code>, and look -for the first parameter at <code><nobr>[BP+6]</nobr></code> instead of -<code><nobr>[BP+4]</nobr></code>. Of course, if one of the parameters is a -pointer, then the offsets of <em>subsequent</em> parameters will change -depending on the memory model as well: far pointers take up four bytes on -the stack when passed as a parameter, whereas near pointers take up two. -<p>At the other end of the process, to call a C function from your assembly -code, you would do something like this: -<p><pre> -extern _printf - - ; and then, further down... - - push word [myint] ; one of my integer variables - push word mystring ; pointer into my data segment - call _printf - add sp,byte 4 ; `byte' saves space - - ; then those data items... - -segment _DATA - -myint dw 1234 -mystring db 'This number -> %d <- should be 1234',10,0 -</pre> -<p>This piece of code is the small-model assembly equivalent of the C code -<p><pre> - int myint = 1234; - printf("This number -> %d <- should be 1234\n", myint); -</pre> -<p>In large model, the function-call code might look more like this. In -this example, it is assumed that <code><nobr>DS</nobr></code> already holds -the segment base of the segment <code><nobr>_DATA</nobr></code>. If not, -you would have to initialize it first. -<p><pre> - push word [myint] - push word seg mystring ; Now push the segment, and... - push word mystring ; ... offset of "mystring" - call far _printf - add sp,byte 6 -</pre> -<p>The integer value still takes up one word on the stack, since large -model does not affect the size of the <code><nobr>int</nobr></code> data -type. The first argument (pushed last) to <code><nobr>printf</nobr></code>, -however, is a data pointer, and therefore has to contain a segment and -offset part. The segment should be stored second in memory, and therefore -must be pushed first. (Of course, <code><nobr>PUSH DS</nobr></code> would -have been a shorter instruction than -<code><nobr>PUSH WORD SEG mystring</nobr></code>, if -<code><nobr>DS</nobr></code> was set up as the above example assumed.) Then -the actual call becomes a far call, since functions expect far calls in -large model; and <code><nobr>SP</nobr></code> has to be increased by 6 -rather than 4 afterwards to make up for the extra word of parameters. -<h4><a name="section-8.4.4">8.4.4 Accessing Data Items</a></h4> -<p>To get at the contents of C variables, or to declare variables which C -can access, you need only declare the names as -<code><nobr>GLOBAL</nobr></code> or <code><nobr>EXTERN</nobr></code>. -(Again, the names require leading underscores, as stated in -<a href="#section-8.4.1">section 8.4.1</a>.) Thus, a C variable declared as -<code><nobr>int i</nobr></code> can be accessed from assembler as -<p><pre> -extern _i - - mov ax,[_i] -</pre> -<p>And to declare your own integer variable which C programs can access as -<code><nobr>extern int j</nobr></code>, you do this (making sure you are -assembling in the <code><nobr>_DATA</nobr></code> segment, if necessary): -<p><pre> -global _j - -_j dw 0 -</pre> -<p>To access a C array, you need to know the size of the components of the -array. For example, <code><nobr>int</nobr></code> variables are two bytes -long, so if a C program declares an array as -<code><nobr>int a[10]</nobr></code>, you can access -<code><nobr>a[3]</nobr></code> by coding -<code><nobr>mov ax,[_a+6]</nobr></code>. (The byte offset 6 is obtained by -multiplying the desired array index, 3, by the size of the array element, -2.) The sizes of the C base types in 16-bit compilers are: 1 for -<code><nobr>char</nobr></code>, 2 for <code><nobr>short</nobr></code> and -<code><nobr>int</nobr></code>, 4 for <code><nobr>long</nobr></code> and -<code><nobr>float</nobr></code>, and 8 for -<code><nobr>double</nobr></code>. -<p>To access a C data structure, you need to know the offset from the base -of the structure to the field you are interested in. You can either do this -by converting the C structure definition into a NASM structure definition -(using <code><nobr>STRUC</nobr></code>), or by calculating the one offset -and using just that. -<p>To do either of these, you should read your C compiler's manual to find -out how it organizes data structures. NASM gives no special alignment to -structure members in its own <code><nobr>STRUC</nobr></code> macro, so you -have to specify alignment yourself if the C compiler generates it. -Typically, you might find that a structure like -<p><pre> -struct { - char c; - int i; -} foo; -</pre> -<p>might be four bytes long rather than three, since the -<code><nobr>int</nobr></code> field would be aligned to a two-byte -boundary. However, this sort of feature tends to be a configurable option -in the C compiler, either using command-line options or -<code><nobr>#pragma</nobr></code> lines, so you have to find out how your -own compiler does it. -<h4><a name="section-8.4.5">8.4.5 <code><nobr>c16.mac</nobr></code>: Helper Macros for the 16-bit C Interface</a></h4> -<p>Included in the NASM archives, in the <code><nobr>misc</nobr></code> -directory, is a file <code><nobr>c16.mac</nobr></code> of macros. It -defines three macros: <code><nobr>proc</nobr></code>, -<code><nobr>arg</nobr></code> and <code><nobr>endproc</nobr></code>. These -are intended to be used for C-style procedure definitions, and they -automate a lot of the work involved in keeping track of the calling -convention. -<p>(An alternative, TASM compatible form of <code><nobr>arg</nobr></code> -is also now built into NASM's preprocessor. See -<a href="nasmdoc4.html#section-4.8">section 4.8</a> for details.) -<p>An example of an assembly function using the macro set is given here: -<p><pre> -proc _nearproc - -%$i arg -%$j arg - mov ax,[bp + %$i] - mov bx,[bp + %$j] - add ax,[bx] - -endproc -</pre> -<p>This defines <code><nobr>_nearproc</nobr></code> to be a procedure -taking two arguments, the first (<code><nobr>i</nobr></code>) an integer -and the second (<code><nobr>j</nobr></code>) a pointer to an integer. It -returns <code><nobr>i + *j</nobr></code>. -<p>Note that the <code><nobr>arg</nobr></code> macro has an -<code><nobr>EQU</nobr></code> as the first line of its expansion, and since -the label before the macro call gets prepended to the first line of the -expanded macro, the <code><nobr>EQU</nobr></code> works, defining -<code><nobr>%$i</nobr></code> to be an offset from -<code><nobr>BP</nobr></code>. A context-local variable is used, local to -the context pushed by the <code><nobr>proc</nobr></code> macro and popped -by the <code><nobr>endproc</nobr></code> macro, so that the same argument -name can be used in later procedures. Of course, you don't <em>have</em> to -do that. -<p>The macro set produces code for near functions (tiny, small and -compact-model code) by default. You can have it generate far functions -(medium, large and huge-model code) by means of coding -<code><nobr>%define FARCODE</nobr></code>. This changes the kind of return -instruction generated by <code><nobr>endproc</nobr></code>, and also -changes the starting point for the argument offsets. The macro set contains -no intrinsic dependency on whether data pointers are far or not. -<p><code><nobr>arg</nobr></code> can take an optional parameter, giving the -size of the argument. If no size is given, 2 is assumed, since it is likely -that many function parameters will be of type -<code><nobr>int</nobr></code>. -<p>The large-model equivalent of the above function would look like this: -<p><pre> -%define FARCODE - -proc _farproc - -%$i arg -%$j arg 4 - mov ax,[bp + %$i] - mov bx,[bp + %$j] - mov es,[bp + %$j + 2] - add ax,[bx] - -endproc -</pre> -<p>This makes use of the argument to the <code><nobr>arg</nobr></code> -macro to define a parameter of size 4, because <code><nobr>j</nobr></code> -is now a far pointer. When we load from <code><nobr>j</nobr></code>, we -must load a segment and an offset. -<h3><a name="section-8.5">8.5 Interfacing to Borland Pascal Programs</a></h3> -<p>Interfacing to Borland Pascal programs is similar in concept to -interfacing to 16-bit C programs. The differences are: -<ul> -<li>The leading underscore required for interfacing to C programs is not -required for Pascal. -<li>The memory model is always large: functions are far, data pointers are -far, and no data item can be more than 64K long. (Actually, some functions -are near, but only those functions that are local to a Pascal unit and -never called from outside it. All assembly functions that Pascal calls, and -all Pascal functions that assembly routines are able to call, are far.) -However, all static data declared in a Pascal program goes into the default -data segment, which is the one whose segment address will be in -<code><nobr>DS</nobr></code> when control is passed to your assembly code. -The only things that do not live in the default data segment are local -variables (they live in the stack segment) and dynamically allocated -variables. All data <em>pointers</em>, however, are far. -<li>The function calling convention is different - described below. -<li>Some data types, such as strings, are stored differently. -<li>There are restrictions on the segment names you are allowed to use - -Borland Pascal will ignore code or data declared in a segment it doesn't -like the name of. The restrictions are described below. -</ul> -<h4><a name="section-8.5.1">8.5.1 The Pascal Calling Convention</a></h4> -<p>The 16-bit Pascal calling convention is as follows. In the following -description, the words <em>caller</em> and <em>callee</em> are used to -denote the function doing the calling and the function which gets called. -<ul> -<li>The caller pushes the function's parameters on the stack, one after -another, in normal order (left to right, so that the first argument -specified to the function is pushed first). -<li>The caller then executes a far <code><nobr>CALL</nobr></code> -instruction to pass control to the callee. -<li>The callee receives control, and typically (although this is not -actually necessary, in functions which do not need to access their -parameters) starts by saving the value of <code><nobr>SP</nobr></code> in -<code><nobr>BP</nobr></code> so as to be able to use -<code><nobr>BP</nobr></code> as a base pointer to find its parameters on -the stack. However, the caller was probably doing this too, so part of the -calling convention states that <code><nobr>BP</nobr></code> must be -preserved by any function. Hence the callee, if it is going to set up -<code><nobr>BP</nobr></code> as a frame pointer, must push the previous -value first. -<li>The callee may then access its parameters relative to -<code><nobr>BP</nobr></code>. The word at <code><nobr>[BP]</nobr></code> -holds the previous value of <code><nobr>BP</nobr></code> as it was pushed. -The next word, at <code><nobr>[BP+2]</nobr></code>, holds the offset part -of the return address, and the next one at <code><nobr>[BP+4]</nobr></code> -the segment part. The parameters begin at <code><nobr>[BP+6]</nobr></code>. -The rightmost parameter of the function, since it was pushed last, is -accessible at this offset from <code><nobr>BP</nobr></code>; the others -follow, at successively greater offsets. -<li>The callee may also wish to decrease <code><nobr>SP</nobr></code> -further, so as to allocate space on the stack for local variables, which -will then be accessible at negative offsets from -<code><nobr>BP</nobr></code>. -<li>The callee, if it wishes to return a value to the caller, should leave -the value in <code><nobr>AL</nobr></code>, <code><nobr>AX</nobr></code> or -<code><nobr>DX:AX</nobr></code> depending on the size of the value. -Floating-point results are returned in <code><nobr>ST0</nobr></code>. -Results of type <code><nobr>Real</nobr></code> (Borland's own custom -floating-point data type, not handled directly by the FPU) are returned in -<code><nobr>DX:BX:AX</nobr></code>. To return a result of type -<code><nobr>String</nobr></code>, the caller pushes a pointer to a -temporary string before pushing the parameters, and the callee places the -returned string value at that location. The pointer is not a parameter, and -should not be removed from the stack by the <code><nobr>RETF</nobr></code> -instruction. -<li>Once the callee has finished processing, it restores -<code><nobr>SP</nobr></code> from <code><nobr>BP</nobr></code> if it had -allocated local stack space, then pops the previous value of -<code><nobr>BP</nobr></code>, and returns via -<code><nobr>RETF</nobr></code>. It uses the form of -<code><nobr>RETF</nobr></code> with an immediate parameter, giving the -number of bytes taken up by the parameters on the stack. This causes the -parameters to be removed from the stack as a side effect of the return -instruction. -<li>When the caller regains control from the callee, the function -parameters have already been removed from the stack, so it needs to do -nothing further. -</ul> -<p>Thus, you would define a function in Pascal style, taking two -<code><nobr>Integer</nobr></code>-type parameters, in the following way: -<p><pre> -global myfunc - -myfunc: push bp - mov bp,sp - sub sp,0x40 ; 64 bytes of local stack space - mov bx,[bp+8] ; first parameter to function - mov bx,[bp+6] ; second parameter to function - - ; some more code - - mov sp,bp ; undo "sub sp,0x40" above - pop bp - retf 4 ; total size of params is 4 -</pre> -<p>At the other end of the process, to call a Pascal function from your -assembly code, you would do something like this: -<p><pre> -extern SomeFunc - - ; and then, further down... - - push word seg mystring ; Now push the segment, and... - push word mystring ; ... offset of "mystring" - push word [myint] ; one of my variables - call far SomeFunc -</pre> -<p>This is equivalent to the Pascal code -<p><pre> -procedure SomeFunc(String: PChar; Int: Integer); - SomeFunc(@mystring, myint); -</pre> -<h4><a name="section-8.5.2">8.5.2 Borland Pascal Segment Name Restrictions</a></h4> -<p>Since Borland Pascal's internal unit file format is completely different -from <code><nobr>OBJ</nobr></code>, it only makes a very sketchy job of -actually reading and understanding the various information contained in a -real <code><nobr>OBJ</nobr></code> file when it links that in. Therefore an -object file intended to be linked to a Pascal program must obey a number of -restrictions: -<ul> -<li>Procedures and functions must be in a segment whose name is either -<code><nobr>CODE</nobr></code>, <code><nobr>CSEG</nobr></code>, or -something ending in <code><nobr>_TEXT</nobr></code>. -<li>initialized data must be in a segment whose name is either -<code><nobr>CONST</nobr></code> or something ending in -<code><nobr>_DATA</nobr></code>. -<li>Uninitialized data must be in a segment whose name is either -<code><nobr>DATA</nobr></code>, <code><nobr>DSEG</nobr></code>, or -something ending in <code><nobr>_BSS</nobr></code>. -<li>Any other segments in the object file are completely ignored. -<code><nobr>GROUP</nobr></code> directives and segment attributes are also -ignored. -</ul> -<h4><a name="section-8.5.3">8.5.3 Using <code><nobr>c16.mac</nobr></code> With Pascal Programs</a></h4> -<p>The <code><nobr>c16.mac</nobr></code> macro package, described in -<a href="#section-8.4.5">section 8.4.5</a>, can also be used to simplify -writing functions to be called from Pascal programs, if you code -<code><nobr>%define PASCAL</nobr></code>. This definition ensures that -functions are far (it implies <code><nobr>FARCODE</nobr></code>), and also -causes procedure return instructions to be generated with an operand. -<p>Defining <code><nobr>PASCAL</nobr></code> does not change the code which -calculates the argument offsets; you must declare your function's arguments -in reverse order. For example: -<p><pre> -%define PASCAL - -proc _pascalproc - -%$j arg 4 -%$i arg - mov ax,[bp + %$i] - mov bx,[bp + %$j] - mov es,[bp + %$j + 2] - add ax,[bx] - -endproc -</pre> -<p>This defines the same routine, conceptually, as the example in -<a href="#section-8.4.5">section 8.4.5</a>: it defines a function taking -two arguments, an integer and a pointer to an integer, which returns the -sum of the integer and the contents of the pointer. The only difference -between this code and the large-model C version is that -<code><nobr>PASCAL</nobr></code> is defined instead of -<code><nobr>FARCODE</nobr></code>, and that the arguments are declared in -reverse order. -<p align=center><a href="nasmdoc9.html">Next Chapter</a> | -<a href="nasmdoc7.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -</body></html> diff --git a/doc/html/nasmdoc9.html b/doc/html/nasmdoc9.html deleted file mode 100644 index 811af7c..0000000 --- a/doc/html/nasmdoc9.html +++ /dev/null @@ -1,482 +0,0 @@ -<html><head><title>NASM Manual</title></head> -<body><h1 align=center>The Netwide Assembler: NASM</h1> - -<p align=center><a href="nasmdo10.html">Next Chapter</a> | -<a href="nasmdoc8.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -<h2><a name="chapter-9">Chapter 9: Writing 32-bit Code (Unix, Win32, DJGPP)</a></h2> -<p>This chapter attempts to cover some of the common issues involved when -writing 32-bit code, to run under Win32 or Unix, or to be linked with C -code generated by a Unix-style C compiler such as DJGPP. It covers how to -write assembly code to interface with 32-bit C routines, and how to write -position-independent code for shared libraries. -<p>Almost all 32-bit code, and in particular all code running under -<code><nobr>Win32</nobr></code>, <code><nobr>DJGPP</nobr></code> or any of -the PC Unix variants, runs in <em>flat</em> memory model. This means that -the segment registers and paging have already been set up to give you the -same 32-bit 4Gb address space no matter what segment you work relative to, -and that you should ignore all segment registers completely. When writing -flat-model application code, you never need to use a segment override or -modify any segment register, and the code-section addresses you pass to -<code><nobr>CALL</nobr></code> and <code><nobr>JMP</nobr></code> live in -the same address space as the data-section addresses you access your -variables by and the stack-section addresses you access local variables and -procedure parameters by. Every address is 32 bits long and contains only an -offset part. -<h3><a name="section-9.1">9.1 Interfacing to 32-bit C Programs</a></h3> -<p>A lot of the discussion in <a href="nasmdoc8.html#section-8.4">section -8.4</a>, about interfacing to 16-bit C programs, still applies when working -in 32 bits. The absence of memory models or segmentation worries simplifies -things a lot. -<h4><a name="section-9.1.1">9.1.1 External Symbol Names</a></h4> -<p>Most 32-bit C compilers share the convention used by 16-bit compilers, -that the names of all global symbols (functions or data) they define are -formed by prefixing an underscore to the name as it appears in the C -program. However, not all of them do: the <code><nobr>ELF</nobr></code> -specification states that C symbols do <em>not</em> have a leading -underscore on their assembly-language names. -<p>The older Linux <code><nobr>a.out</nobr></code> C compiler, all -<code><nobr>Win32</nobr></code> compilers, <code><nobr>DJGPP</nobr></code>, -and <code><nobr>NetBSD</nobr></code> and <code><nobr>FreeBSD</nobr></code>, -all use the leading underscore; for these compilers, the macros -<code><nobr>cextern</nobr></code> and <code><nobr>cglobal</nobr></code>, as -given in <a href="nasmdoc8.html#section-8.4.1">section 8.4.1</a>, will -still work. For <code><nobr>ELF</nobr></code>, though, the leading -underscore should not be used. -<p>See also <a href="nasmdoc2.html#section-2.1.27">section 2.1.27</a>. -<h4><a name="section-9.1.2">9.1.2 Function Definitions and Function Calls</a></h4> -<p>The C calling convention in 32-bit programs is as follows. In the -following description, the words <em>caller</em> and <em>callee</em> are -used to denote the function doing the calling and the function which gets -called. -<ul> -<li>The caller pushes the function's parameters on the stack, one after -another, in reverse order (right to left, so that the first argument -specified to the function is pushed last). -<li>The caller then executes a near <code><nobr>CALL</nobr></code> -instruction to pass control to the callee. -<li>The callee receives control, and typically (although this is not -actually necessary, in functions which do not need to access their -parameters) starts by saving the value of <code><nobr>ESP</nobr></code> in -<code><nobr>EBP</nobr></code> so as to be able to use -<code><nobr>EBP</nobr></code> as a base pointer to find its parameters on -the stack. However, the caller was probably doing this too, so part of the -calling convention states that <code><nobr>EBP</nobr></code> must be -preserved by any C function. Hence the callee, if it is going to set up -<code><nobr>EBP</nobr></code> as a frame pointer, must push the previous -value first. -<li>The callee may then access its parameters relative to -<code><nobr>EBP</nobr></code>. The doubleword at -<code><nobr>[EBP]</nobr></code> holds the previous value of -<code><nobr>EBP</nobr></code> as it was pushed; the next doubleword, at -<code><nobr>[EBP+4]</nobr></code>, holds the return address, pushed -implicitly by <code><nobr>CALL</nobr></code>. The parameters start after -that, at <code><nobr>[EBP+8]</nobr></code>. The leftmost parameter of the -function, since it was pushed last, is accessible at this offset from -<code><nobr>EBP</nobr></code>; the others follow, at successively greater -offsets. Thus, in a function such as <code><nobr>printf</nobr></code> which -takes a variable number of parameters, the pushing of the parameters in -reverse order means that the function knows where to find its first -parameter, which tells it the number and type of the remaining ones. -<li>The callee may also wish to decrease <code><nobr>ESP</nobr></code> -further, so as to allocate space on the stack for local variables, which -will then be accessible at negative offsets from -<code><nobr>EBP</nobr></code>. -<li>The callee, if it wishes to return a value to the caller, should leave -the value in <code><nobr>AL</nobr></code>, <code><nobr>AX</nobr></code> or -<code><nobr>EAX</nobr></code> depending on the size of the value. -Floating-point results are typically returned in -<code><nobr>ST0</nobr></code>. -<li>Once the callee has finished processing, it restores -<code><nobr>ESP</nobr></code> from <code><nobr>EBP</nobr></code> if it had -allocated local stack space, then pops the previous value of -<code><nobr>EBP</nobr></code>, and returns via -<code><nobr>RET</nobr></code> (equivalently, -<code><nobr>RETN</nobr></code>). -<li>When the caller regains control from the callee, the function -parameters are still on the stack, so it typically adds an immediate -constant to <code><nobr>ESP</nobr></code> to remove them (instead of -executing a number of slow <code><nobr>POP</nobr></code> instructions). -Thus, if a function is accidentally called with the wrong number of -parameters due to a prototype mismatch, the stack will still be returned to -a sensible state since the caller, which <em>knows</em> how many parameters -it pushed, does the removing. -</ul> -<p>There is an alternative calling convention used by Win32 programs for -Windows API calls, and also for functions called <em>by</em> the Windows -API such as window procedures: they follow what Microsoft calls the -<code><nobr>__stdcall</nobr></code> convention. This is slightly closer to -the Pascal convention, in that the callee clears the stack by passing a -parameter to the <code><nobr>RET</nobr></code> instruction. However, the -parameters are still pushed in right-to-left order. -<p>Thus, you would define a function in C style in the following way: -<p><pre> -global _myfunc - -_myfunc: - push ebp - mov ebp,esp - sub esp,0x40 ; 64 bytes of local stack space - mov ebx,[ebp+8] ; first parameter to function - - ; some more code - - leave ; mov esp,ebp / pop ebp - ret -</pre> -<p>At the other end of the process, to call a C function from your assembly -code, you would do something like this: -<p><pre> -extern _printf - - ; and then, further down... - - push dword [myint] ; one of my integer variables - push dword mystring ; pointer into my data segment - call _printf - add esp,byte 8 ; `byte' saves space - - ; then those data items... - -segment _DATA - -myint dd 1234 -mystring db 'This number -> %d <- should be 1234',10,0 -</pre> -<p>This piece of code is the assembly equivalent of the C code -<p><pre> - int myint = 1234; - printf("This number -> %d <- should be 1234\n", myint); -</pre> -<h4><a name="section-9.1.3">9.1.3 Accessing Data Items</a></h4> -<p>To get at the contents of C variables, or to declare variables which C -can access, you need only declare the names as -<code><nobr>GLOBAL</nobr></code> or <code><nobr>EXTERN</nobr></code>. -(Again, the names require leading underscores, as stated in -<a href="#section-9.1.1">section 9.1.1</a>.) Thus, a C variable declared as -<code><nobr>int i</nobr></code> can be accessed from assembler as -<p><pre> - extern _i - mov eax,[_i] -</pre> -<p>And to declare your own integer variable which C programs can access as -<code><nobr>extern int j</nobr></code>, you do this (making sure you are -assembling in the <code><nobr>_DATA</nobr></code> segment, if necessary): -<p><pre> - global _j -_j dd 0 -</pre> -<p>To access a C array, you need to know the size of the components of the -array. For example, <code><nobr>int</nobr></code> variables are four bytes -long, so if a C program declares an array as -<code><nobr>int a[10]</nobr></code>, you can access -<code><nobr>a[3]</nobr></code> by coding -<code><nobr>mov ax,[_a+12]</nobr></code>. (The byte offset 12 is obtained -by multiplying the desired array index, 3, by the size of the array -element, 4.) The sizes of the C base types in 32-bit compilers are: 1 for -<code><nobr>char</nobr></code>, 2 for <code><nobr>short</nobr></code>, 4 -for <code><nobr>int</nobr></code>, <code><nobr>long</nobr></code> and -<code><nobr>float</nobr></code>, and 8 for -<code><nobr>double</nobr></code>. Pointers, being 32-bit addresses, are -also 4 bytes long. -<p>To access a C data structure, you need to know the offset from the base -of the structure to the field you are interested in. You can either do this -by converting the C structure definition into a NASM structure definition -(using <code><nobr>STRUC</nobr></code>), or by calculating the one offset -and using just that. -<p>To do either of these, you should read your C compiler's manual to find -out how it organizes data structures. NASM gives no special alignment to -structure members in its own <code><nobr>STRUC</nobr></code> macro, so you -have to specify alignment yourself if the C compiler generates it. -Typically, you might find that a structure like -<p><pre> -struct { - char c; - int i; -} foo; -</pre> -<p>might be eight bytes long rather than five, since the -<code><nobr>int</nobr></code> field would be aligned to a four-byte -boundary. However, this sort of feature is sometimes a configurable option -in the C compiler, either using command-line options or -<code><nobr>#pragma</nobr></code> lines, so you have to find out how your -own compiler does it. -<h4><a name="section-9.1.4">9.1.4 <code><nobr>c32.mac</nobr></code>: Helper Macros for the 32-bit C Interface</a></h4> -<p>Included in the NASM archives, in the <code><nobr>misc</nobr></code> -directory, is a file <code><nobr>c32.mac</nobr></code> of macros. It -defines three macros: <code><nobr>proc</nobr></code>, -<code><nobr>arg</nobr></code> and <code><nobr>endproc</nobr></code>. These -are intended to be used for C-style procedure definitions, and they -automate a lot of the work involved in keeping track of the calling -convention. -<p>An example of an assembly function using the macro set is given here: -<p><pre> -proc _proc32 - -%$i arg -%$j arg - mov eax,[ebp + %$i] - mov ebx,[ebp + %$j] - add eax,[ebx] - -endproc -</pre> -<p>This defines <code><nobr>_proc32</nobr></code> to be a procedure taking -two arguments, the first (<code><nobr>i</nobr></code>) an integer and the -second (<code><nobr>j</nobr></code>) a pointer to an integer. It returns -<code><nobr>i + *j</nobr></code>. -<p>Note that the <code><nobr>arg</nobr></code> macro has an -<code><nobr>EQU</nobr></code> as the first line of its expansion, and since -the label before the macro call gets prepended to the first line of the -expanded macro, the <code><nobr>EQU</nobr></code> works, defining -<code><nobr>%$i</nobr></code> to be an offset from -<code><nobr>BP</nobr></code>. A context-local variable is used, local to -the context pushed by the <code><nobr>proc</nobr></code> macro and popped -by the <code><nobr>endproc</nobr></code> macro, so that the same argument -name can be used in later procedures. Of course, you don't <em>have</em> to -do that. -<p><code><nobr>arg</nobr></code> can take an optional parameter, giving the -size of the argument. If no size is given, 4 is assumed, since it is likely -that many function parameters will be of type <code><nobr>int</nobr></code> -or pointers. -<h3><a name="section-9.2">9.2 Writing NetBSD/FreeBSD/OpenBSD and Linux/ELF Shared Libraries</a></h3> -<p><code><nobr>ELF</nobr></code> replaced the older -<code><nobr>a.out</nobr></code> object file format under Linux because it -contains support for position-independent code (PIC), which makes writing -shared libraries much easier. NASM supports the -<code><nobr>ELF</nobr></code> position-independent code features, so you -can write Linux <code><nobr>ELF</nobr></code> shared libraries in NASM. -<p>NetBSD, and its close cousins FreeBSD and OpenBSD, take a different -approach by hacking PIC support into the <code><nobr>a.out</nobr></code> -format. NASM supports this as the <code><nobr>aoutb</nobr></code> output -format, so you can write BSD shared libraries in NASM too. -<p>The operating system loads a PIC shared library by memory-mapping the -library file at an arbitrarily chosen point in the address space of the -running process. The contents of the library's code section must therefore -not depend on where it is loaded in memory. -<p>Therefore, you cannot get at your variables by writing code like this: -<p><pre> - mov eax,[myvar] ; WRONG -</pre> -<p>Instead, the linker provides an area of memory called the <em>global -offset table</em>, or GOT; the GOT is situated at a constant distance from -your library's code, so if you can find out where your library is loaded -(which is typically done using a <code><nobr>CALL</nobr></code> and -<code><nobr>POP</nobr></code> combination), you can obtain the address of -the GOT, and you can then load the addresses of your variables out of -linker-generated entries in the GOT. -<p>The <em>data</em> section of a PIC shared library does not have these -restrictions: since the data section is writable, it has to be copied into -memory anyway rather than just paged in from the library file, so as long -as it's being copied it can be relocated too. So you can put ordinary types -of relocation in the data section without too much worry (but see -<a href="#section-9.2.4">section 9.2.4</a> for a caveat). -<h4><a name="section-9.2.1">9.2.1 Obtaining the Address of the GOT</a></h4> -<p>Each code module in your shared library should define the GOT as an -external symbol: -<p><pre> -extern _GLOBAL_OFFSET_TABLE_ ; in ELF -extern __GLOBAL_OFFSET_TABLE_ ; in BSD a.out -</pre> -<p>At the beginning of any function in your shared library which plans to -access your data or BSS sections, you must first calculate the address of -the GOT. This is typically done by writing the function in this form: -<p><pre> -func: push ebp - mov ebp,esp - push ebx - call .get_GOT -.get_GOT: - pop ebx - add ebx,_GLOBAL_OFFSET_TABLE_+$$-.get_GOT wrt ..gotpc - - ; the function body comes here - - mov ebx,[ebp-4] - mov esp,ebp - pop ebp - ret -</pre> -<p>(For BSD, again, the symbol -<code><nobr>_GLOBAL_OFFSET_TABLE</nobr></code> requires a second leading -underscore.) -<p>The first two lines of this function are simply the standard C prologue -to set up a stack frame, and the last three lines are standard C function -epilogue. The third line, and the fourth to last line, save and restore the -<code><nobr>EBX</nobr></code> register, because PIC shared libraries use -this register to store the address of the GOT. -<p>The interesting bit is the <code><nobr>CALL</nobr></code> instruction -and the following two lines. The <code><nobr>CALL</nobr></code> and -<code><nobr>POP</nobr></code> combination obtains the address of the label -<code><nobr>.get_GOT</nobr></code>, without having to know in advance where -the program was loaded (since the <code><nobr>CALL</nobr></code> -instruction is encoded relative to the current position). The -<code><nobr>ADD</nobr></code> instruction makes use of one of the special -PIC relocation types: GOTPC relocation. With the -<code><nobr>WRT ..gotpc</nobr></code> qualifier specified, the symbol -referenced (here <code><nobr>_GLOBAL_OFFSET_TABLE_</nobr></code>, the -special symbol assigned to the GOT) is given as an offset from the -beginning of the section. (Actually, <code><nobr>ELF</nobr></code> encodes -it as the offset from the operand field of the -<code><nobr>ADD</nobr></code> instruction, but NASM simplifies this -deliberately, so you do things the same way for both -<code><nobr>ELF</nobr></code> and <code><nobr>BSD</nobr></code>.) So the -instruction then <em>adds</em> the beginning of the section, to get the -real address of the GOT, and subtracts the value of -<code><nobr>.get_GOT</nobr></code> which it knows is in -<code><nobr>EBX</nobr></code>. Therefore, by the time that instruction has -finished, <code><nobr>EBX</nobr></code> contains the address of the GOT. -<p>If you didn't follow that, don't worry: it's never necessary to obtain -the address of the GOT by any other means, so you can put those three -instructions into a macro and safely ignore them: -<p><pre> -%macro get_GOT 0 - - call %%getgot - %%getgot: - pop ebx - add ebx,_GLOBAL_OFFSET_TABLE_+$$-%%getgot wrt ..gotpc - -%endmacro -</pre> -<h4><a name="section-9.2.2">9.2.2 Finding Your Local Data Items</a></h4> -<p>Having got the GOT, you can then use it to obtain the addresses of your -data items. Most variables will reside in the sections you have declared; -they can be accessed using the <code><nobr>..gotoff</nobr></code> special -<code><nobr>WRT</nobr></code> type. The way this works is like this: -<p><pre> - lea eax,[ebx+myvar wrt ..gotoff] -</pre> -<p>The expression <code><nobr>myvar wrt ..gotoff</nobr></code> is -calculated, when the shared library is linked, to be the offset to the -local variable <code><nobr>myvar</nobr></code> from the beginning of the -GOT. Therefore, adding it to <code><nobr>EBX</nobr></code> as above will -place the real address of <code><nobr>myvar</nobr></code> in -<code><nobr>EAX</nobr></code>. -<p>If you declare variables as <code><nobr>GLOBAL</nobr></code> without -specifying a size for them, they are shared between code modules in the -library, but do not get exported from the library to the program that -loaded it. They will still be in your ordinary data and BSS sections, so -you can access them in the same way as local variables, using the above -<code><nobr>..gotoff</nobr></code> mechanism. -<p>Note that due to a peculiarity of the way BSD -<code><nobr>a.out</nobr></code> format handles this relocation type, there -must be at least one non-local symbol in the same section as the address -you're trying to access. -<h4><a name="section-9.2.3">9.2.3 Finding External and Common Data Items</a></h4> -<p>If your library needs to get at an external variable (external to the -<em>library</em>, not just to one of the modules within it), you must use -the <code><nobr>..got</nobr></code> type to get at it. The -<code><nobr>..got</nobr></code> type, instead of giving you the offset from -the GOT base to the variable, gives you the offset from the GOT base to a -GOT <em>entry</em> containing the address of the variable. The linker will -set up this GOT entry when it builds the library, and the dynamic linker -will place the correct address in it at load time. So to obtain the address -of an external variable <code><nobr>extvar</nobr></code> in -<code><nobr>EAX</nobr></code>, you would code -<p><pre> - mov eax,[ebx+extvar wrt ..got] -</pre> -<p>This loads the address of <code><nobr>extvar</nobr></code> out of an -entry in the GOT. The linker, when it builds the shared library, collects -together every relocation of type <code><nobr>..got</nobr></code>, and -builds the GOT so as to ensure it has every necessary entry present. -<p>Common variables must also be accessed in this way. -<h4><a name="section-9.2.4">9.2.4 Exporting Symbols to the Library User</a></h4> -<p>If you want to export symbols to the user of the library, you have to -declare whether they are functions or data, and if they are data, you have -to give the size of the data item. This is because the dynamic linker has -to build procedure linkage table entries for any exported functions, and -also moves exported data items away from the library's data section in -which they were declared. -<p>So to export a function to users of the library, you must use -<p><pre> -global func:function ; declare it as a function - -func: push ebp - - ; etc. -</pre> -<p>And to export a data item such as an array, you would have to code -<p><pre> -global array:data array.end-array ; give the size too - -array: resd 128 -.end: -</pre> -<p>Be careful: If you export a variable to the library user, by declaring -it as <code><nobr>GLOBAL</nobr></code> and supplying a size, the variable -will end up living in the data section of the main program, rather than in -your library's data section, where you declared it. So you will have to -access your own global variable with the <code><nobr>..got</nobr></code> -mechanism rather than <code><nobr>..gotoff</nobr></code>, as if it were -external (which, effectively, it has become). -<p>Equally, if you need to store the address of an exported global in one -of your data sections, you can't do it by means of the standard sort of -code: -<p><pre> -dataptr: dd global_data_item ; WRONG -</pre> -<p>NASM will interpret this code as an ordinary relocation, in which -<code><nobr>global_data_item</nobr></code> is merely an offset from the -beginning of the <code><nobr>.data</nobr></code> section (or whatever); so -this reference will end up pointing at your data section instead of at the -exported global which resides elsewhere. -<p>Instead of the above code, then, you must write -<p><pre> -dataptr: dd global_data_item wrt ..sym -</pre> -<p>which makes use of the special <code><nobr>WRT</nobr></code> type -<code><nobr>..sym</nobr></code> to instruct NASM to search the symbol table -for a particular symbol at that address, rather than just relocating by -section base. -<p>Either method will work for functions: referring to one of your -functions by means of -<p><pre> -funcptr: dd my_function -</pre> -<p>will give the user the address of the code you wrote, whereas -<p><pre> -funcptr: dd my_function wrt .sym -</pre> -<p>will give the address of the procedure linkage table for the function, -which is where the calling program will <em>believe</em> the function -lives. Either address is a valid way to call the function. -<h4><a name="section-9.2.5">9.2.5 Calling Procedures Outside the Library</a></h4> -<p>Calling procedures outside your shared library has to be done by means -of a <em>procedure linkage table</em>, or PLT. The PLT is placed at a known -offset from where the library is loaded, so the library code can make calls -to the PLT in a position-independent way. Within the PLT there is code to -jump to offsets contained in the GOT, so function calls to other shared -libraries or to routines in the main program can be transparently passed -off to their real destinations. -<p>To call an external routine, you must use another special PIC relocation -type, <code><nobr>WRT ..plt</nobr></code>. This is much easier than the -GOT-based ones: you simply replace calls such as -<code><nobr>CALL printf</nobr></code> with the PLT-relative version -<code><nobr>CALL printf WRT ..plt</nobr></code>. -<h4><a name="section-9.2.6">9.2.6 Generating the Library File</a></h4> -<p>Having written some code modules and assembled them to -<code><nobr>.o</nobr></code> files, you then generate your shared library -with a command such as -<p><pre> -ld -shared -o library.so module1.o module2.o # for ELF -ld -Bshareable -o library.so module1.o module2.o # for BSD -</pre> -<p>For ELF, if your shared library is going to reside in system directories -such as <code><nobr>/usr/lib</nobr></code> or -<code><nobr>/lib</nobr></code>, it is usually worth using the -<code><nobr>-soname</nobr></code> flag to the linker, to store the final -library file name, with a version number, into the library: -<p><pre> -ld -shared -soname library.so.1 -o library.so.1.2 *.o -</pre> -<p>You would then copy <code><nobr>library.so.1.2</nobr></code> into the -library directory, and create <code><nobr>library.so.1</nobr></code> as a -symbolic link to it. -<p align=center><a href="nasmdo10.html">Next Chapter</a> | -<a href="nasmdoc8.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -</body></html> diff --git a/doc/html/nasmdoca.html b/doc/html/nasmdoca.html deleted file mode 100644 index fda553d..0000000 --- a/doc/html/nasmdoca.html +++ /dev/null @@ -1,165 +0,0 @@ -<html><head><title>NASM Manual</title></head> -<body><h1 align=center>The Netwide Assembler: NASM</h1> - -<p align=center><a href="nasmdocb.html">Next Chapter</a> | -<a href="nasmdo12.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -<h2><a name="appendix-A">Appendix A: Ndisasm</a></h2> -<p>The Netwide Disassembler, NDISASM -<h3><a name="section-A.1">A.1 Introduction</a></h3> -<p>The Netwide Disassembler is a small companion program to the Netwide -Assembler, NASM. It seemed a shame to have an x86 assembler, complete with -a full instruction table, and not make as much use of it as possible, so -here's a disassembler which shares the instruction table (and some other -bits of code) with NASM. -<p>The Netwide Disassembler does nothing except to produce disassemblies of -<em>binary</em> source files. NDISASM does not have any understanding of -object file formats, like <code><nobr>objdump</nobr></code>, and it will -not understand <code><nobr>DOS .EXE</nobr></code> files like -<code><nobr>debug</nobr></code> will. It just disassembles. -<h3><a name="section-A.2">A.2 Getting Started: Installation</a></h3> -<p>See <a href="nasmdoc1.html#section-1.3">section 1.3</a> for installation -instructions. NDISASM, like NASM, has a <code><nobr>man page</nobr></code> -which you may want to put somewhere useful, if you are on a Unix system. -<h3><a name="section-A.3">A.3 Running NDISASM</a></h3> -<p>To disassemble a file, you will typically use a command of the form -<p><pre> - ndisasm -b {16|32|64} filename -</pre> -<p>NDISASM can disassemble 16-, 32- or 64-bit code equally easily, provided -of course that you remember to specify which it is to work with. If no -<code><nobr>-b</nobr></code> switch is present, NDISASM works in 16-bit -mode by default. The <code><nobr>-u</nobr></code> switch (for USE32) also -invokes 32-bit mode. -<p>Two more command line options are <code><nobr>-r</nobr></code> which -reports the version number of NDISASM you are running, and -<code><nobr>-h</nobr></code> which gives a short summary of command line -options. -<h4><a name="section-A.3.1">A.3.1 COM Files: Specifying an Origin</a></h4> -<p>To disassemble a <code><nobr>DOS .COM</nobr></code> file correctly, a -disassembler must assume that the first instruction in the file is loaded -at address <code><nobr>0x100</nobr></code>, rather than at zero. NDISASM, -which assumes by default that any file you give it is loaded at zero, will -therefore need to be informed of this. -<p>The <code><nobr>-o</nobr></code> option allows you to declare a -different origin for the file you are disassembling. Its argument may be -expressed in any of the NASM numeric formats: decimal by default, if it -begins with `<code><nobr>$</nobr></code>' or `<code><nobr>0x</nobr></code>' -or ends in `<code><nobr>H</nobr></code>' it's -<code><nobr>hex</nobr></code>, if it ends in `<code><nobr>Q</nobr></code>' -it's <code><nobr>octal</nobr></code>, and if it ends in -`<code><nobr>B</nobr></code>' it's <code><nobr>binary</nobr></code>. -<p>Hence, to disassemble a <code><nobr>.COM</nobr></code> file: -<p><pre> - ndisasm -o100h filename.com -</pre> -<p>will do the trick. -<h4><a name="section-A.3.2">A.3.2 Code Following Data: Synchronisation</a></h4> -<p>Suppose you are disassembling a file which contains some data which -isn't machine code, and <em>then</em> contains some machine code. NDISASM -will faithfully plough through the data section, producing machine -instructions wherever it can (although most of them will look bizarre, and -some may have unusual prefixes, e.g. -`<code><nobr>FS OR AX,0x240A</nobr></code>'), and generating `DB' -instructions ever so often if it's totally stumped. Then it will reach the -code section. -<p>Supposing NDISASM has just finished generating a strange machine -instruction from part of the data section, and its file position is now one -byte <em>before</em> the beginning of the code section. It's entirely -possible that another spurious instruction will get generated, starting -with the final byte of the data section, and then the correct first -instruction in the code section will not be seen because the starting point -skipped over it. This isn't really ideal. -<p>To avoid this, you can specify a -`<code><nobr>synchronisation</nobr></code>' point, or indeed as many -synchronisation points as you like (although NDISASM can only handle -2147483647 sync points internally). The definition of a sync point is this: -NDISASM guarantees to hit sync points exactly during disassembly. If it is -thinking about generating an instruction which would cause it to jump over -a sync point, it will discard that instruction and output a -`<code><nobr>db</nobr></code>' instead. So it <em>will</em> start -disassembly exactly from the sync point, and so you <em>will</em> see all -the instructions in your code section. -<p>Sync points are specified using the <code><nobr>-s</nobr></code> option: -they are measured in terms of the program origin, not the file position. So -if you want to synchronize after 32 bytes of a -<code><nobr>.COM</nobr></code> file, you would have to do -<p><pre> - ndisasm -o100h -s120h file.com -</pre> -<p>rather than -<p><pre> - ndisasm -o100h -s20h file.com -</pre> -<p>As stated above, you can specify multiple sync markers if you need to, -just by repeating the <code><nobr>-s</nobr></code> option. -<h4><a name="section-A.3.3">A.3.3 Mixed Code and Data: Automatic (Intelligent) Synchronisation </a></h4> -<p>Suppose you are disassembling the boot sector of a -<code><nobr>DOS</nobr></code> floppy (maybe it has a virus, and you need to -understand the virus so that you know what kinds of damage it might have -done you). Typically, this will contain a <code><nobr>JMP</nobr></code> -instruction, then some data, then the rest of the code. So there is a very -good chance of NDISASM being <em>misaligned</em> when the data ends and the -code begins. Hence a sync point is needed. -<p>On the other hand, why should you have to specify the sync point -manually? What you'd do in order to find where the sync point would be, -surely, would be to read the <code><nobr>JMP</nobr></code> instruction, and -then to use its target address as a sync point. So can NDISASM do that for -you? -<p>The answer, of course, is yes: using either of the synonymous switches -<code><nobr>-a</nobr></code> (for automatic sync) or -<code><nobr>-i</nobr></code> (for intelligent sync) will enable -<code><nobr>auto-sync</nobr></code> mode. Auto-sync mode automatically -generates a sync point for any forward-referring PC-relative jump or call -instruction that NDISASM encounters. (Since NDISASM is one-pass, if it -encounters a PC-relative jump whose target has already been processed, -there isn't much it can do about it...) -<p>Only PC-relative jumps are processed, since an absolute jump is either -through a register (in which case NDISASM doesn't know what the register -contains) or involves a segment address (in which case the target code -isn't in the same segment that NDISASM is working in, and so the sync point -can't be placed anywhere useful). -<p>For some kinds of file, this mechanism will automatically put sync -points in all the right places, and save you from having to place any sync -points manually. However, it should be stressed that auto-sync mode is -<em>not</em> guaranteed to catch all the sync points, and you may still -have to place some manually. -<p>Auto-sync mode doesn't prevent you from declaring manual sync points: it -just adds automatically generated ones to the ones you provide. It's -perfectly feasible to specify <code><nobr>-i</nobr></code> <em>and</em> -some <code><nobr>-s</nobr></code> options. -<p>Another caveat with auto-sync mode is that if, by some unpleasant fluke, -something in your data section should disassemble to a PC-relative call or -jump instruction, NDISASM may obediently place a sync point in a totally -random place, for example in the middle of one of the instructions in your -code section. So you may end up with a wrong disassembly even if you use -auto-sync. Again, there isn't much I can do about this. If you have -problems, you'll have to use manual sync points, or use the -<code><nobr>-k</nobr></code> option (documented below) to suppress -disassembly of the data area. -<h4><a name="section-A.3.4">A.3.4 Other Options</a></h4> -<p>The <code><nobr>-e</nobr></code> option skips a header on the file, by -ignoring the first N bytes. This means that the header is <em>not</em> -counted towards the disassembly offset: if you give -<code><nobr>-e10 -o10</nobr></code>, disassembly will start at byte 10 in -the file, and this will be given offset 10, not 20. -<p>The <code><nobr>-k</nobr></code> option is provided with two -comma-separated numeric arguments, the first of which is an assembly offset -and the second is a number of bytes to skip. This <em>will</em> count the -skipped bytes towards the assembly offset: its use is to suppress -disassembly of a data section which wouldn't contain anything you wanted to -see anyway. -<h3><a name="section-A.4">A.4 Bugs and Improvements</a></h3> -<p>There are no known bugs. However, any you find, with patches if -possible, should be sent to -<a href="mailto:nasm-bugs@lists.sourceforge.net"><code><nobr>nasm-bugs@lists.sourceforge.net</nobr></code></a>, -or to the developer's site at -<a href="https://sourceforge.net/projects/nasm/"><code><nobr>https://sourceforge.net/projects/nasm/</nobr></code></a> -and we'll try to fix them. Feel free to send contributions and new features -as well. -<p align=center><a href="nasmdocb.html">Next Chapter</a> | -<a href="nasmdo12.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -</body></html> diff --git a/doc/html/nasmdocb.html b/doc/html/nasmdocb.html deleted file mode 100644 index 15bca45..0000000 --- a/doc/html/nasmdocb.html +++ /dev/null @@ -1,3148 +0,0 @@ -<html><head><title>NASM Manual</title></head> -<body><h1 align=center>The Netwide Assembler: NASM</h1> - -<p align=center><a href="nasmdocc.html">Next Chapter</a> | -<a href="nasmdoca.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -<h2><a name="appendix-B">Appendix B: Instruction List</a></h2> -<h3><a name="section-B.1">B.1 Introduction</a></h3> -<p>The following sections show the instructions which NASM currently -supports. For each instruction, there is a separate entry for each -supported addressing mode. The third column shows the processor type in -which the instruction was introduced and, when appropriate, one or more -usage flags. -<h4><a name="section-B.1.1">B.1.1 Special instructions...</a></h4> -<p><pre> -DB -DW -DD -DQ -DT -DO -DY -RESB imm 8086 -RESW -RESD -RESQ -REST -RESO -RESY -</pre> -<h4><a name="section-B.1.2">B.1.2 Conventional instructions</a></h4> -<p><pre> -AAA 8086,NOLONG -AAD 8086,NOLONG -AAD imm 8086,NOLONG -AAM 8086,NOLONG -AAM imm 8086,NOLONG -AAS 8086,NOLONG -ADC mem,reg8 8086 -ADC reg8,reg8 8086 -ADC mem,reg16 8086 -ADC reg16,reg16 8086 -ADC mem,reg32 386 -ADC reg32,reg32 386 -ADC mem,reg64 X64 -ADC reg64,reg64 X64 -ADC reg8,mem 8086 -ADC reg8,reg8 8086 -ADC reg16,mem 8086 -ADC reg16,reg16 8086 -ADC reg32,mem 386 -ADC reg32,reg32 386 -ADC reg64,mem X64 -ADC reg64,reg64 X64 -ADC rm16,imm8 8086 -ADC rm32,imm8 386 -ADC rm64,imm8 X64 -ADC reg_al,imm 8086 -ADC reg_ax,sbyte16 8086 -ADC reg_ax,imm 8086 -ADC reg_eax,sbyte32 386 -ADC reg_eax,imm 386 -ADC reg_rax,sbyte64 X64 -ADC reg_rax,imm X64 -ADC rm8,imm 8086 -ADC rm16,imm 8086 -ADC rm32,imm 386 -ADC rm64,imm X64 -ADC mem,imm8 8086 -ADC mem,imm16 8086 -ADC mem,imm32 386 -ADD mem,reg8 8086 -ADD reg8,reg8 8086 -ADD mem,reg16 8086 -ADD reg16,reg16 8086 -ADD mem,reg32 386 -ADD reg32,reg32 386 -ADD mem,reg64 X64 -ADD reg64,reg64 X64 -ADD reg8,mem 8086 -ADD reg8,reg8 8086 -ADD reg16,mem 8086 -ADD reg16,reg16 8086 -ADD reg32,mem 386 -ADD reg32,reg32 386 -ADD reg64,mem X64 -ADD reg64,reg64 X64 -ADD rm16,imm8 8086 -ADD rm32,imm8 386 -ADD rm64,imm8 X64 -ADD reg_al,imm 8086 -ADD reg_ax,sbyte16 8086 -ADD reg_ax,imm 8086 -ADD reg_eax,sbyte32 386 -ADD reg_eax,imm 386 -ADD reg_rax,sbyte64 X64 -ADD reg_rax,imm X64 -ADD rm8,imm 8086 -ADD rm16,imm 8086 -ADD rm32,imm 386 -ADD rm64,imm X64 -ADD mem,imm8 8086 -ADD mem,imm16 8086 -ADD mem,imm32 386 -AND mem,reg8 8086 -AND reg8,reg8 8086 -AND mem,reg16 8086 -AND reg16,reg16 8086 -AND mem,reg32 386 -AND reg32,reg32 386 -AND mem,reg64 X64 -AND reg64,reg64 X64 -AND reg8,mem 8086 -AND reg8,reg8 8086 -AND reg16,mem 8086 -AND reg16,reg16 8086 -AND reg32,mem 386 -AND reg32,reg32 386 -AND reg64,mem X64 -AND reg64,reg64 X64 -AND rm16,imm8 8086 -AND rm32,imm8 386 -AND rm64,imm8 X64 -AND reg_al,imm 8086 -AND reg_ax,sbyte16 8086 -AND reg_ax,imm 8086 -AND reg_eax,sbyte32 386 -AND reg_eax,imm 386 -AND reg_rax,sbyte64 X64 -AND reg_rax,imm X64 -AND rm8,imm 8086 -AND rm16,imm 8086 -AND rm32,imm 386 -AND rm64,imm X64 -AND mem,imm8 8086 -AND mem,imm16 8086 -AND mem,imm32 386 -ARPL mem,reg16 286,PROT,NOLONG -ARPL reg16,reg16 286,PROT,NOLONG -BB0_RESET PENT,CYRIX,ND -BB1_RESET PENT,CYRIX,ND -BOUND reg16,mem 186,NOLONG -BOUND reg32,mem 386,NOLONG -BSF reg16,mem 386 -BSF reg16,reg16 386 -BSF reg32,mem 386 -BSF reg32,reg32 386 -BSF reg64,mem X64 -BSF reg64,reg64 X64 -BSR reg16,mem 386 -BSR reg16,reg16 386 -BSR reg32,mem 386 -BSR reg32,reg32 386 -BSR reg64,mem X64 -BSR reg64,reg64 X64 -BSWAP reg32 486 -BSWAP reg64 X64 -BT mem,reg16 386 -BT reg16,reg16 386 -BT mem,reg32 386 -BT reg32,reg32 386 -BT mem,reg64 X64 -BT reg64,reg64 X64 -BT rm16,imm 386 -BT rm32,imm 386 -BT rm64,imm X64 -BTC mem,reg16 386 -BTC reg16,reg16 386 -BTC mem,reg32 386 -BTC reg32,reg32 386 -BTC mem,reg64 X64 -BTC reg64,reg64 X64 -BTC rm16,imm 386 -BTC rm32,imm 386 -BTC rm64,imm X64 -BTR mem,reg16 386 -BTR reg16,reg16 386 -BTR mem,reg32 386 -BTR reg32,reg32 386 -BTR mem,reg64 X64 -BTR reg64,reg64 X64 -BTR rm16,imm 386 -BTR rm32,imm 386 -BTR rm64,imm X64 -BTS mem,reg16 386 -BTS reg16,reg16 386 -BTS mem,reg32 386 -BTS reg32,reg32 386 -BTS mem,reg64 X64 -BTS reg64,reg64 X64 -BTS rm16,imm 386 -BTS rm32,imm 386 -BTS rm64,imm X64 -CALL imm 8086 -CALL imm|near 8086 -CALL imm|far 8086,ND,NOLONG -CALL imm16 8086 -CALL imm16|near 8086 -CALL imm16|far 8086,ND,NOLONG -CALL imm32 386 -CALL imm32|near 386 -CALL imm32|far 386,ND,NOLONG -CALL imm:imm 8086,NOLONG -CALL imm16:imm 8086,NOLONG -CALL imm:imm16 8086,NOLONG -CALL imm32:imm 386,NOLONG -CALL imm:imm32 386,NOLONG -CALL mem|far 8086,NOLONG -CALL mem|far X64 -CALL mem16|far 8086 -CALL mem32|far 386 -CALL mem64|far X64 -CALL mem|near 8086 -CALL mem16|near 8086 -CALL mem32|near 386,NOLONG -CALL mem64|near X64 -CALL reg16 8086 -CALL reg32 386,NOLONG -CALL reg64 X64 -CALL mem 8086 -CALL mem16 8086 -CALL mem32 386,NOLONG -CALL mem64 X64 -CBW 8086 -CDQ 386 -CDQE X64 -CLC 8086 -CLD 8086 -CLGI X64,AMD -CLI 8086 -CLTS 286,PRIV -CMC 8086 -CMP mem,reg8 8086 -CMP reg8,reg8 8086 -CMP mem,reg16 8086 -CMP reg16,reg16 8086 -CMP mem,reg32 386 -CMP reg32,reg32 386 -CMP mem,reg64 X64 -CMP reg64,reg64 X64 -CMP reg8,mem 8086 -CMP reg8,reg8 8086 -CMP reg16,mem 8086 -CMP reg16,reg16 8086 -CMP reg32,mem 386 -CMP reg32,reg32 386 -CMP reg64,mem X64 -CMP reg64,reg64 X64 -CMP rm16,imm8 8086 -CMP rm32,imm8 386 -CMP rm64,imm8 X64 -CMP reg_al,imm 8086 -CMP reg_ax,sbyte16 8086 -CMP reg_ax,imm 8086 -CMP reg_eax,sbyte32 386 -CMP reg_eax,imm 386 -CMP reg_rax,sbyte64 X64 -CMP reg_rax,imm X64 -CMP rm8,imm 8086 -CMP rm16,imm 8086 -CMP rm32,imm 386 -CMP rm64,imm X64 -CMP mem,imm8 8086 -CMP mem,imm16 8086 -CMP mem,imm32 386 -CMPSB 8086 -CMPSD 386 -CMPSQ X64 -CMPSW 8086 -CMPXCHG mem,reg8 PENT -CMPXCHG reg8,reg8 PENT -CMPXCHG mem,reg16 PENT -CMPXCHG reg16,reg16 PENT -CMPXCHG mem,reg32 PENT -CMPXCHG reg32,reg32 PENT -CMPXCHG mem,reg64 X64 -CMPXCHG reg64,reg64 X64 -CMPXCHG486 mem,reg8 486,UNDOC,ND -CMPXCHG486 reg8,reg8 486,UNDOC,ND -CMPXCHG486 mem,reg16 486,UNDOC,ND -CMPXCHG486 reg16,reg16 486,UNDOC,ND -CMPXCHG486 mem,reg32 486,UNDOC,ND -CMPXCHG486 reg32,reg32 486,UNDOC,ND -CMPXCHG8B mem PENT -CMPXCHG16B mem X64 -CPUID PENT -CPU_READ PENT,CYRIX -CPU_WRITE PENT,CYRIX -CQO X64 -CWD 8086 -CWDE 386 -DAA 8086,NOLONG -DAS 8086,NOLONG -DEC reg16 8086,NOLONG -DEC reg32 386,NOLONG -DEC rm8 8086 -DEC rm16 8086 -DEC rm32 386 -DEC rm64 X64 -DIV rm8 8086 -DIV rm16 8086 -DIV rm32 386 -DIV rm64 X64 -DMINT P6,CYRIX -EMMS PENT,MMX -ENTER imm,imm 186 -EQU imm 8086 -EQU imm:imm 8086 -F2XM1 8086,FPU -FABS 8086,FPU -FADD mem32 8086,FPU -FADD mem64 8086,FPU -FADD fpureg|to 8086,FPU -FADD fpureg 8086,FPU -FADD fpureg,fpu0 8086,FPU -FADD fpu0,fpureg 8086,FPU -FADD 8086,FPU,ND -FADDP fpureg 8086,FPU -FADDP fpureg,fpu0 8086,FPU -FADDP 8086,FPU,ND -FBLD mem80 8086,FPU -FBLD mem 8086,FPU -FBSTP mem80 8086,FPU -FBSTP mem 8086,FPU -FCHS 8086,FPU -FCLEX 8086,FPU -FCMOVB fpureg P6,FPU -FCMOVB fpu0,fpureg P6,FPU -FCMOVB P6,FPU,ND -FCMOVBE fpureg P6,FPU -FCMOVBE fpu0,fpureg P6,FPU -FCMOVBE P6,FPU,ND -FCMOVE fpureg P6,FPU -FCMOVE fpu0,fpureg P6,FPU -FCMOVE P6,FPU,ND -FCMOVNB fpureg P6,FPU -FCMOVNB fpu0,fpureg P6,FPU -FCMOVNB P6,FPU,ND -FCMOVNBE fpureg P6,FPU -FCMOVNBE fpu0,fpureg P6,FPU -FCMOVNBE P6,FPU,ND -FCMOVNE fpureg P6,FPU -FCMOVNE fpu0,fpureg P6,FPU -FCMOVNE P6,FPU,ND -FCMOVNU fpureg P6,FPU -FCMOVNU fpu0,fpureg P6,FPU -FCMOVNU P6,FPU,ND -FCMOVU fpureg P6,FPU -FCMOVU fpu0,fpureg P6,FPU -FCMOVU P6,FPU,ND -FCOM mem32 8086,FPU -FCOM mem64 8086,FPU -FCOM fpureg 8086,FPU -FCOM fpu0,fpureg 8086,FPU -FCOM 8086,FPU,ND -FCOMI fpureg P6,FPU -FCOMI fpu0,fpureg P6,FPU -FCOMI P6,FPU,ND -FCOMIP fpureg P6,FPU -FCOMIP fpu0,fpureg P6,FPU -FCOMIP P6,FPU,ND -FCOMP mem32 8086,FPU -FCOMP mem64 8086,FPU -FCOMP fpureg 8086,FPU -FCOMP fpu0,fpureg 8086,FPU -FCOMP 8086,FPU,ND -FCOMPP 8086,FPU -FCOS 386,FPU -FDECSTP 8086,FPU -FDISI 8086,FPU -FDIV mem32 8086,FPU -FDIV mem64 8086,FPU -FDIV fpureg|to 8086,FPU -FDIV fpureg 8086,FPU -FDIV fpureg,fpu0 8086,FPU -FDIV fpu0,fpureg 8086,FPU -FDIV 8086,FPU,ND -FDIVP fpureg 8086,FPU -FDIVP fpureg,fpu0 8086,FPU -FDIVP 8086,FPU,ND -FDIVR mem32 8086,FPU -FDIVR mem64 8086,FPU -FDIVR fpureg|to 8086,FPU -FDIVR fpureg,fpu0 8086,FPU -FDIVR fpureg 8086,FPU -FDIVR fpu0,fpureg 8086,FPU -FDIVR 8086,FPU,ND -FDIVRP fpureg 8086,FPU -FDIVRP fpureg,fpu0 8086,FPU -FDIVRP 8086,FPU,ND -FEMMS PENT,3DNOW -FENI 8086,FPU -FFREE fpureg 8086,FPU -FFREE 8086,FPU -FFREEP fpureg 286,FPU,UNDOC -FFREEP 286,FPU,UNDOC -FIADD mem32 8086,FPU -FIADD mem16 8086,FPU -FICOM mem32 8086,FPU -FICOM mem16 8086,FPU -FICOMP mem32 8086,FPU -FICOMP mem16 8086,FPU -FIDIV mem32 8086,FPU -FIDIV mem16 8086,FPU -FIDIVR mem32 8086,FPU -FIDIVR mem16 8086,FPU -FILD mem32 8086,FPU -FILD mem16 8086,FPU -FILD mem64 8086,FPU -FIMUL mem32 8086,FPU -FIMUL mem16 8086,FPU -FINCSTP 8086,FPU -FINIT 8086,FPU -FIST mem32 8086,FPU -FIST mem16 8086,FPU -FISTP mem32 8086,FPU -FISTP mem16 8086,FPU -FISTP mem64 8086,FPU -FISTTP mem16 PRESCOTT,FPU -FISTTP mem32 PRESCOTT,FPU -FISTTP mem64 PRESCOTT,FPU -FISUB mem32 8086,FPU -FISUB mem16 8086,FPU -FISUBR mem32 8086,FPU -FISUBR mem16 8086,FPU -FLD mem32 8086,FPU -FLD mem64 8086,FPU -FLD mem80 8086,FPU -FLD fpureg 8086,FPU -FLD 8086,FPU,ND -FLD1 8086,FPU -FLDCW mem 8086,FPU,SW -FLDENV mem 8086,FPU -FLDL2E 8086,FPU -FLDL2T 8086,FPU -FLDLG2 8086,FPU -FLDLN2 8086,FPU -FLDPI 8086,FPU -FLDZ 8086,FPU -FMUL mem32 8086,FPU -FMUL mem64 8086,FPU -FMUL fpureg|to 8086,FPU -FMUL fpureg,fpu0 8086,FPU -FMUL fpureg 8086,FPU -FMUL fpu0,fpureg 8086,FPU -FMUL 8086,FPU,ND -FMULP fpureg 8086,FPU -FMULP fpureg,fpu0 8086,FPU -FMULP 8086,FPU,ND -FNCLEX 8086,FPU -FNDISI 8086,FPU -FNENI 8086,FPU -FNINIT 8086,FPU -FNOP 8086,FPU -FNSAVE mem 8086,FPU -FNSTCW mem 8086,FPU,SW -FNSTENV mem 8086,FPU -FNSTSW mem 8086,FPU,SW -FNSTSW reg_ax 286,FPU -FPATAN 8086,FPU -FPREM 8086,FPU -FPREM1 386,FPU -FPTAN 8086,FPU -FRNDINT 8086,FPU -FRSTOR mem 8086,FPU -FSAVE mem 8086,FPU -FSCALE 8086,FPU -FSETPM 286,FPU -FSIN 386,FPU -FSINCOS 386,FPU -FSQRT 8086,FPU -FST mem32 8086,FPU -FST mem64 8086,FPU -FST fpureg 8086,FPU -FST 8086,FPU,ND -FSTCW mem 8086,FPU,SW -FSTENV mem 8086,FPU -FSTP mem32 8086,FPU -FSTP mem64 8086,FPU -FSTP mem80 8086,FPU -FSTP fpureg 8086,FPU -FSTP 8086,FPU,ND -FSTSW mem 8086,FPU,SW -FSTSW reg_ax 286,FPU -FSUB mem32 8086,FPU -FSUB mem64 8086,FPU -FSUB fpureg|to 8086,FPU -FSUB fpureg,fpu0 8086,FPU -FSUB fpureg 8086,FPU -FSUB fpu0,fpureg 8086,FPU -FSUB 8086,FPU,ND -FSUBP fpureg 8086,FPU -FSUBP fpureg,fpu0 8086,FPU -FSUBP 8086,FPU,ND -FSUBR mem32 8086,FPU -FSUBR mem64 8086,FPU -FSUBR fpureg|to 8086,FPU -FSUBR fpureg,fpu0 8086,FPU -FSUBR fpureg 8086,FPU -FSUBR fpu0,fpureg 8086,FPU -FSUBR 8086,FPU,ND -FSUBRP fpureg 8086,FPU -FSUBRP fpureg,fpu0 8086,FPU -FSUBRP 8086,FPU,ND -FTST 8086,FPU -FUCOM fpureg 386,FPU -FUCOM fpu0,fpureg 386,FPU -FUCOM 386,FPU,ND -FUCOMI fpureg P6,FPU -FUCOMI fpu0,fpureg P6,FPU -FUCOMI P6,FPU,ND -FUCOMIP fpureg P6,FPU -FUCOMIP fpu0,fpureg P6,FPU -FUCOMIP P6,FPU,ND -FUCOMP fpureg 386,FPU -FUCOMP fpu0,fpureg 386,FPU -FUCOMP 386,FPU,ND -FUCOMPP 386,FPU -FXAM 8086,FPU -FXCH fpureg 8086,FPU -FXCH fpureg,fpu0 8086,FPU -FXCH fpu0,fpureg 8086,FPU -FXCH 8086,FPU,ND -FXTRACT 8086,FPU -FYL2X 8086,FPU -FYL2XP1 8086,FPU -HLT 8086,PRIV -IBTS mem,reg16 386,SW,UNDOC,ND -IBTS reg16,reg16 386,UNDOC,ND -IBTS mem,reg32 386,SD,UNDOC,ND -IBTS reg32,reg32 386,UNDOC,ND -ICEBP 386,ND -IDIV rm8 8086 -IDIV rm16 8086 -IDIV rm32 386 -IDIV rm64 X64 -IMUL rm8 8086 -IMUL rm16 8086 -IMUL rm32 386 -IMUL rm64 X64 -IMUL reg16,mem 386 -IMUL reg16,reg16 386 -IMUL reg32,mem 386 -IMUL reg32,reg32 386 -IMUL reg64,mem X64 -IMUL reg64,reg64 X64 -IMUL reg16,mem,imm8 186 -IMUL reg16,mem,sbyte16 186,ND -IMUL reg16,mem,imm16 186 -IMUL reg16,mem,imm 186,ND -IMUL reg16,reg16,imm8 186 -IMUL reg16,reg16,sbyte16 186,ND -IMUL reg16,reg16,imm16 186 -IMUL reg16,reg16,imm 186,ND -IMUL reg32,mem,imm8 386 -IMUL reg32,mem,sbyte32 386,ND -IMUL reg32,mem,imm32 386 -IMUL reg32,mem,imm 386,ND -IMUL reg32,reg32,imm8 386 -IMUL reg32,reg32,sbyte32 386,ND -IMUL reg32,reg32,imm32 386 -IMUL reg32,reg32,imm 386,ND -IMUL reg64,mem,imm8 X64 -IMUL reg64,mem,sbyte64 X64,ND -IMUL reg64,mem,imm32 X64 -IMUL reg64,mem,imm X64,ND -IMUL reg64,reg64,imm8 X64 -IMUL reg64,reg64,sbyte64 X64,ND -IMUL reg64,reg64,imm32 X64 -IMUL reg64,reg64,imm X64,ND -IMUL reg16,imm8 186 -IMUL reg16,sbyte16 186,ND -IMUL reg16,imm16 186 -IMUL reg16,imm 186,ND -IMUL reg32,imm8 386 -IMUL reg32,sbyte32 386,ND -IMUL reg32,imm32 386 -IMUL reg32,imm 386,ND -IMUL reg64,imm8 X64 -IMUL reg64,sbyte64 X64,ND -IMUL reg64,imm32 X64 -IMUL reg64,imm X64,ND -IN reg_al,imm 8086 -IN reg_ax,imm 8086 -IN reg_eax,imm 386 -IN reg_al,reg_dx 8086 -IN reg_ax,reg_dx 8086 -IN reg_eax,reg_dx 386 -INC reg16 8086,NOLONG -INC reg32 386,NOLONG -INC rm8 8086 -INC rm16 8086 -INC rm32 386 -INC rm64 X64 -INCBIN -INSB 186 -INSD 386 -INSW 186 -INT imm 8086 -INT01 386,ND -INT1 386 -INT03 8086,ND -INT3 8086 -INTO 8086,NOLONG -INVD 486,PRIV -INVLPG mem 486,PRIV -INVLPGA reg_ax,reg_ecx X86_64,AMD,NOLONG -INVLPGA reg_eax,reg_ecx X86_64,AMD -INVLPGA reg_rax,reg_ecx X64,AMD -INVLPGA X86_64,AMD -IRET 8086 -IRETD 386 -IRETQ X64 -IRETW 8086 -JCXZ imm 8086,NOLONG -JECXZ imm 386 -JRCXZ imm X64 -JMP imm|short 8086 -JMP imm 8086,ND -JMP imm 8086 -JMP imm|near 8086,ND -JMP imm|far 8086,ND,NOLONG -JMP imm16 8086 -JMP imm16|near 8086,ND -JMP imm16|far 8086,ND,NOLONG -JMP imm32 386 -JMP imm32|near 386,ND -JMP imm32|far 386,ND,NOLONG -JMP imm:imm 8086,NOLONG -JMP imm16:imm 8086,NOLONG -JMP imm:imm16 8086,NOLONG -JMP imm32:imm 386,NOLONG -JMP imm:imm32 386,NOLONG -JMP mem|far 8086,NOLONG -JMP mem|far X64 -JMP mem16|far 8086 -JMP mem32|far 386 -JMP mem64|far X64 -JMP mem|near 8086 -JMP mem16|near 8086 -JMP mem32|near 386,NOLONG -JMP mem64|near X64 -JMP reg16 8086 -JMP reg32 386,NOLONG -JMP reg64 X64 -JMP mem 8086 -JMP mem16 8086 -JMP mem32 386,NOLONG -JMP mem64 X64 -JMPE imm IA64 -JMPE imm16 IA64 -JMPE imm32 IA64 -JMPE rm16 IA64 -JMPE rm32 IA64 -LAHF 8086 -LAR reg16,mem 286,PROT,SW -LAR reg16,reg16 286,PROT -LAR reg16,reg32 386,PROT -LAR reg16,reg64 X64,PROT,ND -LAR reg32,mem 386,PROT,SW -LAR reg32,reg16 386,PROT -LAR reg32,reg32 386,PROT -LAR reg32,reg64 X64,PROT,ND -LAR reg64,mem X64,PROT,SW -LAR reg64,reg16 X64,PROT -LAR reg64,reg32 X64,PROT -LAR reg64,reg64 X64,PROT -LDS reg16,mem 8086,NOLONG -LDS reg32,mem 386,NOLONG -LEA reg16,mem 8086 -LEA reg32,mem 386 -LEA reg64,mem X64 -LEAVE 186 -LES reg16,mem 8086,NOLONG -LES reg32,mem 386,NOLONG -LFENCE X64,AMD -LFS reg16,mem 386 -LFS reg32,mem 386 -LGDT mem 286,PRIV -LGS reg16,mem 386 -LGS reg32,mem 386 -LIDT mem 286,PRIV -LLDT mem 286,PROT,PRIV -LLDT mem16 286,PROT,PRIV -LLDT reg16 286,PROT,PRIV -LMSW mem 286,PRIV -LMSW mem16 286,PRIV -LMSW reg16 286,PRIV -LOADALL 386,UNDOC -LOADALL286 286,UNDOC -LODSB 8086 -LODSD 386 -LODSQ X64 -LODSW 8086 -LOOP imm 8086 -LOOP imm,reg_cx 8086,NOLONG -LOOP imm,reg_ecx 386 -LOOP imm,reg_rcx X64 -LOOPE imm 8086 -LOOPE imm,reg_cx 8086,NOLONG -LOOPE imm,reg_ecx 386 -LOOPE imm,reg_rcx X64 -LOOPNE imm 8086 -LOOPNE imm,reg_cx 8086,NOLONG -LOOPNE imm,reg_ecx 386 -LOOPNE imm,reg_rcx X64 -LOOPNZ imm 8086 -LOOPNZ imm,reg_cx 8086,NOLONG -LOOPNZ imm,reg_ecx 386 -LOOPNZ imm,reg_rcx X64 -LOOPZ imm 8086 -LOOPZ imm,reg_cx 8086,NOLONG -LOOPZ imm,reg_ecx 386 -LOOPZ imm,reg_rcx X64 -LSL reg16,mem 286,PROT,SW -LSL reg16,reg16 286,PROT -LSL reg16,reg32 386,PROT -LSL reg16,reg64 X64,PROT,ND -LSL reg32,mem 386,PROT,SW -LSL reg32,reg16 386,PROT -LSL reg32,reg32 386,PROT -LSL reg32,reg64 X64,PROT,ND -LSL reg64,mem X64,PROT,SW -LSL reg64,reg16 X64,PROT -LSL reg64,reg32 X64,PROT -LSL reg64,reg64 X64,PROT -LSS reg16,mem 386 -LSS reg32,mem 386 -LTR mem 286,PROT,PRIV -LTR mem16 286,PROT,PRIV -LTR reg16 286,PROT,PRIV -MFENCE X64,AMD -MONITOR PRESCOTT -MONITOR reg_eax,reg_ecx,reg_edx PRESCOTT,ND -MONITOR reg_rax,reg_ecx,reg_edx X64,ND -MOV mem,reg_sreg 8086 -MOV reg16,reg_sreg 8086 -MOV reg32,reg_sreg 386 -MOV reg_sreg,mem 8086 -MOV reg_sreg,reg16 8086 -MOV reg_sreg,reg32 386 -MOV reg_al,mem_offs 8086 -MOV reg_ax,mem_offs 8086 -MOV reg_eax,mem_offs 386 -MOV reg_rax,mem_offs X64 -MOV mem_offs,reg_al 8086 -MOV mem_offs,reg_ax 8086 -MOV mem_offs,reg_eax 386 -MOV mem_offs,reg_rax X64 -MOV reg32,reg_creg 386,PRIV,NOLONG -MOV reg64,reg_creg X64,PRIV -MOV reg_creg,reg32 386,PRIV,NOLONG -MOV reg_creg,reg64 X64,PRIV -MOV reg32,reg_dreg 386,PRIV,NOLONG -MOV reg64,reg_dreg X64,PRIV -MOV reg_dreg,reg32 386,PRIV,NOLONG -MOV reg_dreg,reg64 X64,PRIV -MOV reg32,reg_treg 386,NOLONG,ND -MOV reg_treg,reg32 386,NOLONG,ND -MOV mem,reg8 8086 -MOV reg8,reg8 8086 -MOV mem,reg16 8086 -MOV reg16,reg16 8086 -MOV mem,reg32 386 -MOV reg32,reg32 386 -MOV mem,reg64 X64 -MOV reg64,reg64 X64 -MOV reg8,mem 8086 -MOV reg8,reg8 8086 -MOV reg16,mem 8086 -MOV reg16,reg16 8086 -MOV reg32,mem 386 -MOV reg32,reg32 386 -MOV reg64,mem X64 -MOV reg64,reg64 X64 -MOV reg8,imm 8086 -MOV reg16,imm 8086 -MOV reg32,imm 386 -MOV reg64,imm X64 -MOV reg64,imm32 X64 -MOV rm8,imm 8086 -MOV rm16,imm 8086 -MOV rm32,imm 386 -MOV rm64,imm X64 -MOV mem,imm8 8086 -MOV mem,imm16 8086 -MOV mem,imm32 386 -MOVD mmxreg,mem PENT,MMX,SD -MOVD mmxreg,reg32 PENT,MMX -MOVD mem,mmxreg PENT,MMX,SD -MOVD reg32,mmxreg PENT,MMX -MOVD xmmreg,mem X64,SD -MOVD xmmreg,reg32 X64 -MOVD mem,xmmreg X64,SD -MOVD reg32,xmmreg X64,SSE -MOVQ mmxreg,mmxrm PENT,MMX -MOVQ mmxrm,mmxreg PENT,MMX -MOVQ mmxreg,rm64 X64,MMX -MOVQ rm64,mmxreg X64,MMX -MOVSB 8086 -MOVSD 386 -MOVSQ X64 -MOVSW 8086 -MOVSX reg16,mem 386 -MOVSX reg16,reg8 386 -MOVSX reg32,rm8 386 -MOVSX reg32,rm16 386 -MOVSX reg64,rm8 X64 -MOVSX reg64,rm16 X64 -MOVSXD reg64,rm32 X64 -MOVSX reg64,rm32 X64,ND -MOVZX reg16,mem 386 -MOVZX reg16,reg8 386 -MOVZX reg32,rm8 386 -MOVZX reg32,rm16 386 -MOVZX reg64,rm8 X64 -MOVZX reg64,rm16 X64 -MUL rm8 8086 -MUL rm16 8086 -MUL rm32 386 -MUL rm64 X64 -MWAIT PRESCOTT -MWAIT reg_eax,reg_ecx PRESCOTT,ND -NEG rm8 8086 -NEG rm16 8086 -NEG rm32 386 -NEG rm64 X64 -NOP 8086 -NOP rm16 P6 -NOP rm32 P6 -NOP rm64 X64 -NOT rm8 8086 -NOT rm16 8086 -NOT rm32 386 -NOT rm64 X64 -OR mem,reg8 8086 -OR reg8,reg8 8086 -OR mem,reg16 8086 -OR reg16,reg16 8086 -OR mem,reg32 386 -OR reg32,reg32 386 -OR mem,reg64 X64 -OR reg64,reg64 X64 -OR reg8,mem 8086 -OR reg8,reg8 8086 -OR reg16,mem 8086 -OR reg16,reg16 8086 -OR reg32,mem 386 -OR reg32,reg32 386 -OR reg64,mem X64 -OR reg64,reg64 X64 -OR rm16,imm8 8086 -OR rm32,imm8 386 -OR rm64,imm8 X64 -OR reg_al,imm 8086 -OR reg_ax,sbyte16 8086 -OR reg_ax,imm 8086 -OR reg_eax,sbyte32 386 -OR reg_eax,imm 386 -OR reg_rax,sbyte64 X64 -OR reg_rax,imm X64 -OR rm8,imm 8086 -OR rm16,imm 8086 -OR rm32,imm 386 -OR rm64,imm X64 -OR mem,imm8 8086 -OR mem,imm16 8086 -OR mem,imm32 386 -OUT imm,reg_al 8086 -OUT imm,reg_ax 8086 -OUT imm,reg_eax 386 -OUT reg_dx,reg_al 8086 -OUT reg_dx,reg_ax 8086 -OUT reg_dx,reg_eax 386 -OUTSB 186 -OUTSD 386 -OUTSW 186 -PACKSSDW mmxreg,mmxrm PENT,MMX -PACKSSWB mmxreg,mmxrm PENT,MMX -PACKUSWB mmxreg,mmxrm PENT,MMX -PADDB mmxreg,mmxrm PENT,MMX -PADDD mmxreg,mmxrm PENT,MMX -PADDSB mmxreg,mmxrm PENT,MMX -PADDSIW mmxreg,mmxrm PENT,MMX,CYRIX -PADDSW mmxreg,mmxrm PENT,MMX -PADDUSB mmxreg,mmxrm PENT,MMX -PADDUSW mmxreg,mmxrm PENT,MMX -PADDW mmxreg,mmxrm PENT,MMX -PAND mmxreg,mmxrm PENT,MMX -PANDN mmxreg,mmxrm PENT,MMX -PAUSE 8086 -PAVEB mmxreg,mmxrm PENT,MMX,CYRIX -PAVGUSB mmxreg,mmxrm PENT,3DNOW -PCMPEQB mmxreg,mmxrm PENT,MMX -PCMPEQD mmxreg,mmxrm PENT,MMX -PCMPEQW mmxreg,mmxrm PENT,MMX -PCMPGTB mmxreg,mmxrm PENT,MMX -PCMPGTD mmxreg,mmxrm PENT,MMX -PCMPGTW mmxreg,mmxrm PENT,MMX -PDISTIB mmxreg,mem PENT,MMX,CYRIX -PF2ID mmxreg,mmxrm PENT,3DNOW -PFACC mmxreg,mmxrm PENT,3DNOW -PFADD mmxreg,mmxrm PENT,3DNOW -PFCMPEQ mmxreg,mmxrm PENT,3DNOW -PFCMPGE mmxreg,mmxrm PENT,3DNOW -PFCMPGT mmxreg,mmxrm PENT,3DNOW -PFMAX mmxreg,mmxrm PENT,3DNOW -PFMIN mmxreg,mmxrm PENT,3DNOW -PFMUL mmxreg,mmxrm PENT,3DNOW -PFRCP mmxreg,mmxrm PENT,3DNOW -PFRCPIT1 mmxreg,mmxrm PENT,3DNOW -PFRCPIT2 mmxreg,mmxrm PENT,3DNOW -PFRSQIT1 mmxreg,mmxrm PENT,3DNOW -PFRSQRT mmxreg,mmxrm PENT,3DNOW -PFSUB mmxreg,mmxrm PENT,3DNOW -PFSUBR mmxreg,mmxrm PENT,3DNOW -PI2FD mmxreg,mmxrm PENT,3DNOW -PMACHRIW mmxreg,mem PENT,MMX,CYRIX -PMADDWD mmxreg,mmxrm PENT,MMX -PMAGW mmxreg,mmxrm PENT,MMX,CYRIX -PMULHRIW mmxreg,mmxrm PENT,MMX,CYRIX -PMULHRWA mmxreg,mmxrm PENT,3DNOW -PMULHRWC mmxreg,mmxrm PENT,MMX,CYRIX -PMULHW mmxreg,mmxrm PENT,MMX -PMULLW mmxreg,mmxrm PENT,MMX -PMVGEZB mmxreg,mem PENT,MMX,CYRIX -PMVLZB mmxreg,mem PENT,MMX,CYRIX -PMVNZB mmxreg,mem PENT,MMX,CYRIX -PMVZB mmxreg,mem PENT,MMX,CYRIX -POP reg16 8086 -POP reg32 386,NOLONG -POP reg64 X64 -POP rm16 8086 -POP rm32 386,NOLONG -POP rm64 X64 -POP reg_cs 8086,UNDOC,ND -POP reg_dess 8086,NOLONG -POP reg_fsgs 386 -POPA 186,NOLONG -POPAD 386,NOLONG -POPAW 186,NOLONG -POPF 8086 -POPFD 386,NOLONG -POPFQ X64 -POPFW 8086 -POR mmxreg,mmxrm PENT,MMX -PREFETCH mem PENT,3DNOW -PREFETCHW mem PENT,3DNOW -PSLLD mmxreg,mmxrm PENT,MMX -PSLLD mmxreg,imm PENT,MMX -PSLLQ mmxreg,mmxrm PENT,MMX -PSLLQ mmxreg,imm PENT,MMX -PSLLW mmxreg,mmxrm PENT,MMX -PSLLW mmxreg,imm PENT,MMX -PSRAD mmxreg,mmxrm PENT,MMX -PSRAD mmxreg,imm PENT,MMX -PSRAW mmxreg,mmxrm PENT,MMX -PSRAW mmxreg,imm PENT,MMX -PSRLD mmxreg,mmxrm PENT,MMX -PSRLD mmxreg,imm PENT,MMX -PSRLQ mmxreg,mmxrm PENT,MMX -PSRLQ mmxreg,imm PENT,MMX -PSRLW mmxreg,mmxrm PENT,MMX -PSRLW mmxreg,imm PENT,MMX -PSUBB mmxreg,mmxrm PENT,MMX -PSUBD mmxreg,mmxrm PENT,MMX -PSUBSB mmxreg,mmxrm PENT,MMX -PSUBSIW mmxreg,mmxrm PENT,MMX,CYRIX -PSUBSW mmxreg,mmxrm PENT,MMX -PSUBUSB mmxreg,mmxrm PENT,MMX -PSUBUSW mmxreg,mmxrm PENT,MMX -PSUBW mmxreg,mmxrm PENT,MMX -PUNPCKHBW mmxreg,mmxrm PENT,MMX -PUNPCKHDQ mmxreg,mmxrm PENT,MMX -PUNPCKHWD mmxreg,mmxrm PENT,MMX -PUNPCKLBW mmxreg,mmxrm PENT,MMX -PUNPCKLDQ mmxreg,mmxrm PENT,MMX -PUNPCKLWD mmxreg,mmxrm PENT,MMX -PUSH reg16 8086 -PUSH reg32 386,NOLONG -PUSH reg64 X64 -PUSH rm16 8086 -PUSH rm32 386,NOLONG -PUSH rm64 X64 -PUSH reg_cs 8086,NOLONG -PUSH reg_dess 8086,NOLONG -PUSH reg_fsgs 386 -PUSH imm8 186 -PUSH imm16 186,AR0,SZ -PUSH imm32 386,NOLONG,AR0,SZ -PUSH imm32 386,NOLONG,SD -PUSH imm64 X64,AR0,SZ -PUSHA 186,NOLONG -PUSHAD 386,NOLONG -PUSHAW 186,NOLONG -PUSHF 8086 -PUSHFD 386,NOLONG -PUSHFQ X64 -PUSHFW 8086 -PXOR mmxreg,mmxrm PENT,MMX -RCL rm8,unity 8086 -RCL rm8,reg_cl 8086 -RCL rm8,imm 186 -RCL rm16,unity 8086 -RCL rm16,reg_cl 8086 -RCL rm16,imm 186 -RCL rm32,unity 386 -RCL rm32,reg_cl 386 -RCL rm32,imm 386 -RCL rm64,unity X64 -RCL rm64,reg_cl X64 -RCL rm64,imm X64 -RCR rm8,unity 8086 -RCR rm8,reg_cl 8086 -RCR rm8,imm 186 -RCR rm16,unity 8086 -RCR rm16,reg_cl 8086 -RCR rm16,imm 186 -RCR rm32,unity 386 -RCR rm32,reg_cl 386 -RCR rm32,imm 386 -RCR rm64,unity X64 -RCR rm64,reg_cl X64 -RCR rm64,imm X64 -RDSHR rm32 P6,CYRIXM -RDMSR PENT,PRIV -RDPMC P6 -RDTSC PENT -RDTSCP X86_64 -RET 8086 -RET imm 8086,SW -RETF 8086 -RETF imm 8086,SW -RETN 8086 -RETN imm 8086,SW -ROL rm8,unity 8086 -ROL rm8,reg_cl 8086 -ROL rm8,imm 186 -ROL rm16,unity 8086 -ROL rm16,reg_cl 8086 -ROL rm16,imm 186 -ROL rm32,unity 386 -ROL rm32,reg_cl 386 -ROL rm32,imm 386 -ROL rm64,unity X64 -ROL rm64,reg_cl X64 -ROL rm64,imm X64 -ROR rm8,unity 8086 -ROR rm8,reg_cl 8086 -ROR rm8,imm 186 -ROR rm16,unity 8086 -ROR rm16,reg_cl 8086 -ROR rm16,imm 186 -ROR rm32,unity 386 -ROR rm32,reg_cl 386 -ROR rm32,imm 386 -ROR rm64,unity X64 -ROR rm64,reg_cl X64 -ROR rm64,imm X64 -RDM P6,CYRIX,ND -RSDC reg_sreg,mem80 486,CYRIXM -RSLDT mem80 486,CYRIXM -RSM PENTM -RSTS mem80 486,CYRIXM -SAHF 8086 -SAL rm8,unity 8086,ND -SAL rm8,reg_cl 8086,ND -SAL rm8,imm 186,ND -SAL rm16,unity 8086,ND -SAL rm16,reg_cl 8086,ND -SAL rm16,imm 186,ND -SAL rm32,unity 386,ND -SAL rm32,reg_cl 386,ND -SAL rm32,imm 386,ND -SAL rm64,unity X64,ND -SAL rm64,reg_cl X64,ND -SAL rm64,imm X64,ND -SALC 8086,UNDOC -SAR rm8,unity 8086 -SAR rm8,reg_cl 8086 -SAR rm8,imm 186 -SAR rm16,unity 8086 -SAR rm16,reg_cl 8086 -SAR rm16,imm 186 -SAR rm32,unity 386 -SAR rm32,reg_cl 386 -SAR rm32,imm 386 -SAR rm64,unity X64 -SAR rm64,reg_cl X64 -SAR rm64,imm X64 -SBB mem,reg8 8086 -SBB reg8,reg8 8086 -SBB mem,reg16 8086 -SBB reg16,reg16 8086 -SBB mem,reg32 386 -SBB reg32,reg32 386 -SBB mem,reg64 X64 -SBB reg64,reg64 X64 -SBB reg8,mem 8086 -SBB reg8,reg8 8086 -SBB reg16,mem 8086 -SBB reg16,reg16 8086 -SBB reg32,mem 386 -SBB reg32,reg32 386 -SBB reg64,mem X64 -SBB reg64,reg64 X64 -SBB rm16,imm8 8086 -SBB rm32,imm8 386 -SBB rm64,imm8 X64 -SBB reg_al,imm 8086 -SBB reg_ax,sbyte16 8086 -SBB reg_ax,imm 8086 -SBB reg_eax,sbyte32 386 -SBB reg_eax,imm 386 -SBB reg_rax,sbyte64 X64 -SBB reg_rax,imm X64 -SBB rm8,imm 8086 -SBB rm16,imm 8086 -SBB rm32,imm 386 -SBB rm64,imm X64 -SBB mem,imm8 8086 -SBB mem,imm16 8086 -SBB mem,imm32 386 -SCASB 8086 -SCASD 386 -SCASQ X64 -SCASW 8086 -SFENCE X64,AMD -SGDT mem 286 -SHL rm8,unity 8086 -SHL rm8,reg_cl 8086 -SHL rm8,imm 186 -SHL rm16,unity 8086 -SHL rm16,reg_cl 8086 -SHL rm16,imm 186 -SHL rm32,unity 386 -SHL rm32,reg_cl 386 -SHL rm32,imm 386 -SHL rm64,unity X64 -SHL rm64,reg_cl X64 -SHL rm64,imm X64 -SHLD mem,reg16,imm 3862 -SHLD reg16,reg16,imm 3862 -SHLD mem,reg32,imm 3862 -SHLD reg32,reg32,imm 3862 -SHLD mem,reg64,imm X642 -SHLD reg64,reg64,imm X642 -SHLD mem,reg16,reg_cl 386 -SHLD reg16,reg16,reg_cl 386 -SHLD mem,reg32,reg_cl 386 -SHLD reg32,reg32,reg_cl 386 -SHLD mem,reg64,reg_cl X64 -SHLD reg64,reg64,reg_cl X64 -SHR rm8,unity 8086 -SHR rm8,reg_cl 8086 -SHR rm8,imm 186 -SHR rm16,unity 8086 -SHR rm16,reg_cl 8086 -SHR rm16,imm 186 -SHR rm32,unity 386 -SHR rm32,reg_cl 386 -SHR rm32,imm 386 -SHR rm64,unity X64 -SHR rm64,reg_cl X64 -SHR rm64,imm X64 -SHRD mem,reg16,imm 3862 -SHRD reg16,reg16,imm 3862 -SHRD mem,reg32,imm 3862 -SHRD reg32,reg32,imm 3862 -SHRD mem,reg64,imm X642 -SHRD reg64,reg64,imm X642 -SHRD mem,reg16,reg_cl 386 -SHRD reg16,reg16,reg_cl 386 -SHRD mem,reg32,reg_cl 386 -SHRD reg32,reg32,reg_cl 386 -SHRD mem,reg64,reg_cl X64 -SHRD reg64,reg64,reg_cl X64 -SIDT mem 286 -SLDT mem 286 -SLDT mem16 286 -SLDT reg16 286 -SLDT reg32 386 -SLDT reg64 X64,ND -SLDT reg64 X64 -SKINIT X64 -SMI 386,UNDOC -SMINT P6,CYRIX,ND -SMINTOLD 486,CYRIX,ND -SMSW mem 286 -SMSW mem16 286 -SMSW reg16 286 -SMSW reg32 386 -STC 8086 -STD 8086 -STGI X64 -STI 8086 -STOSB 8086 -STOSD 386 -STOSQ X64 -STOSW 8086 -STR mem 286,PROT -STR mem16 286,PROT -STR reg16 286,PROT -STR reg32 386,PROT -STR reg64 X64 -SUB mem,reg8 8086 -SUB reg8,reg8 8086 -SUB mem,reg16 8086 -SUB reg16,reg16 8086 -SUB mem,reg32 386 -SUB reg32,reg32 386 -SUB mem,reg64 X64 -SUB reg64,reg64 X64 -SUB reg8,mem 8086 -SUB reg8,reg8 8086 -SUB reg16,mem 8086 -SUB reg16,reg16 8086 -SUB reg32,mem 386 -SUB reg32,reg32 386 -SUB reg64,mem X64 -SUB reg64,reg64 X64 -SUB rm16,imm8 8086 -SUB rm32,imm8 386 -SUB rm64,imm8 X64 -SUB reg_al,imm 8086 -SUB reg_ax,sbyte16 8086 -SUB reg_ax,imm 8086 -SUB reg_eax,sbyte32 386 -SUB reg_eax,imm 386 -SUB reg_rax,sbyte64 X64 -SUB reg_rax,imm X64 -SUB rm8,imm 8086 -SUB rm16,imm 8086 -SUB rm32,imm 386 -SUB rm64,imm X64 -SUB mem,imm8 8086 -SUB mem,imm16 8086 -SUB mem,imm32 386 -SVDC mem80,reg_sreg 486,CYRIXM -SVLDT mem80 486,CYRIXM,ND -SVTS mem80 486,CYRIXM -SWAPGS X64 -SYSCALL P6,AMD -SYSENTER P6 -SYSEXIT P6,PRIV -SYSRET P6,PRIV,AMD -TEST mem,reg8 8086 -TEST reg8,reg8 8086 -TEST mem,reg16 8086 -TEST reg16,reg16 8086 -TEST mem,reg32 386 -TEST reg32,reg32 386 -TEST mem,reg64 X64 -TEST reg64,reg64 X64 -TEST reg8,mem 8086 -TEST reg16,mem 8086 -TEST reg32,mem 386 -TEST reg64,mem X64 -TEST reg_al,imm 8086 -TEST reg_ax,imm 8086 -TEST reg_eax,imm 386 -TEST reg_rax,imm X64 -TEST rm8,imm 8086 -TEST rm16,imm 8086 -TEST rm32,imm 386 -TEST rm64,imm X64 -TEST mem,imm8 8086 -TEST mem,imm16 8086 -TEST mem,imm32 386 -UD0 186,UNDOC -UD1 186,UNDOC -UD2B 186,UNDOC,ND -UD2 186 -UD2A 186,ND -UMOV mem,reg8 386,UNDOC,ND -UMOV reg8,reg8 386,UNDOC,ND -UMOV mem,reg16 386,UNDOC,ND -UMOV reg16,reg16 386,UNDOC,ND -UMOV mem,reg32 386,UNDOC,ND -UMOV reg32,reg32 386,UNDOC,ND -UMOV reg8,mem 386,UNDOC,ND -UMOV reg8,reg8 386,UNDOC,ND -UMOV reg16,mem 386,UNDOC,ND -UMOV reg16,reg16 386,UNDOC,ND -UMOV reg32,mem 386,UNDOC,ND -UMOV reg32,reg32 386,UNDOC,ND -VERR mem 286,PROT -VERR mem16 286,PROT -VERR reg16 286,PROT -VERW mem 286,PROT -VERW mem16 286,PROT -VERW reg16 286,PROT -FWAIT 8086 -WBINVD 486,PRIV -WRSHR rm32 P6,CYRIXM -WRMSR PENT,PRIV -XADD mem,reg8 486 -XADD reg8,reg8 486 -XADD mem,reg16 486 -XADD reg16,reg16 486 -XADD mem,reg32 486 -XADD reg32,reg32 486 -XADD mem,reg64 X64 -XADD reg64,reg64 X64 -XBTS reg16,mem 386,SW,UNDOC,ND -XBTS reg16,reg16 386,UNDOC,ND -XBTS reg32,mem 386,SD,UNDOC,ND -XBTS reg32,reg32 386,UNDOC,ND -XCHG reg_ax,reg16 8086 -XCHG reg_eax,reg32na 386 -XCHG reg_rax,reg64 X64 -XCHG reg16,reg_ax 8086 -XCHG reg32na,reg_eax 386 -XCHG reg64,reg_rax X64 -XCHG reg_eax,reg_eax 386,NOLONG -XCHG reg8,mem 8086 -XCHG reg8,reg8 8086 -XCHG reg16,mem 8086 -XCHG reg16,reg16 8086 -XCHG reg32,mem 386 -XCHG reg32,reg32 386 -XCHG reg64,mem X64 -XCHG reg64,reg64 X64 -XCHG mem,reg8 8086 -XCHG reg8,reg8 8086 -XCHG mem,reg16 8086 -XCHG reg16,reg16 8086 -XCHG mem,reg32 386 -XCHG reg32,reg32 386 -XCHG mem,reg64 X64 -XCHG reg64,reg64 X64 -XLATB 8086 -XLAT 8086 -XOR mem,reg8 8086 -XOR reg8,reg8 8086 -XOR mem,reg16 8086 -XOR reg16,reg16 8086 -XOR mem,reg32 386 -XOR reg32,reg32 386 -XOR mem,reg64 X64 -XOR reg64,reg64 X64 -XOR reg8,mem 8086 -XOR reg8,reg8 8086 -XOR reg16,mem 8086 -XOR reg16,reg16 8086 -XOR reg32,mem 386 -XOR reg32,reg32 386 -XOR reg64,mem X64 -XOR reg64,reg64 X64 -XOR rm16,imm8 8086 -XOR rm32,imm8 386 -XOR rm64,imm8 X64 -XOR reg_al,imm 8086 -XOR reg_ax,sbyte16 8086 -XOR reg_ax,imm 8086 -XOR reg_eax,sbyte32 386 -XOR reg_eax,imm 386 -XOR reg_rax,sbyte64 X64 -XOR reg_rax,imm X64 -XOR rm8,imm 8086 -XOR rm16,imm 8086 -XOR rm32,imm 386 -XOR rm64,imm X64 -XOR mem,imm8 8086 -XOR mem,imm16 8086 -XOR mem,imm32 386 -CMOVcc reg16,mem P6 -CMOVcc reg16,reg16 P6 -CMOVcc reg32,mem P6 -CMOVcc reg32,reg32 P6 -CMOVcc reg64,mem X64 -CMOVcc reg64,reg64 X64 -Jcc imm|near 386 -Jcc imm16|near 386 -Jcc imm32|near 386 -Jcc imm|short 8086,ND -Jcc imm 8086,ND -Jcc imm 386,ND -Jcc imm 8086,ND -Jcc imm 8086 -SETcc mem 386 -SETcc reg8 386 -</pre> -<h4><a name="section-B.1.3">B.1.3 Katmai Streaming SIMD instructions (SSE -- a.k.a. KNI, XMM, MMX2)</a></h4> -<p><pre> -ADDPS xmmreg,xmmrm KATMAI,SSE -ADDSS xmmreg,xmmrm KATMAI,SSE,SD -ANDNPS xmmreg,xmmrm KATMAI,SSE -ANDPS xmmreg,xmmrm KATMAI,SSE -CMPEQPS xmmreg,xmmrm KATMAI,SSE -CMPEQSS xmmreg,xmmrm KATMAI,SSE -CMPLEPS xmmreg,xmmrm KATMAI,SSE -CMPLESS xmmreg,xmmrm KATMAI,SSE -CMPLTPS xmmreg,xmmrm KATMAI,SSE -CMPLTSS xmmreg,xmmrm KATMAI,SSE -CMPNEQPS xmmreg,xmmrm KATMAI,SSE -CMPNEQSS xmmreg,xmmrm KATMAI,SSE -CMPNLEPS xmmreg,xmmrm KATMAI,SSE -CMPNLESS xmmreg,xmmrm KATMAI,SSE -CMPNLTPS xmmreg,xmmrm KATMAI,SSE -CMPNLTSS xmmreg,xmmrm KATMAI,SSE -CMPORDPS xmmreg,xmmrm KATMAI,SSE -CMPORDSS xmmreg,xmmrm KATMAI,SSE -CMPUNORDPS xmmreg,xmmrm KATMAI,SSE -CMPUNORDSS xmmreg,xmmrm KATMAI,SSE -CMPPS xmmreg,mem,imm KATMAI,SSE -CMPPS xmmreg,xmmreg,imm KATMAI,SSE -CMPSS xmmreg,mem,imm KATMAI,SSE -CMPSS xmmreg,xmmreg,imm KATMAI,SSE -COMISS xmmreg,xmmrm KATMAI,SSE -CVTPI2PS xmmreg,mmxrm KATMAI,SSE,MMX -CVTPS2PI mmxreg,xmmrm KATMAI,SSE,MMX -CVTSI2SS xmmreg,mem KATMAI,SSE,SD,AR1,ND -CVTSI2SS xmmreg,rm32 KATMAI,SSE,SD,AR1 -CVTSI2SS xmmreg,rm64 X64,SSE,AR1 -CVTSS2SI reg32,xmmreg KATMAI,SSE,SD,AR1 -CVTSS2SI reg32,mem KATMAI,SSE,SD,AR1 -CVTSS2SI reg64,xmmreg X64,SSE,SD,AR1 -CVTSS2SI reg64,mem X64,SSE,SD,AR1 -CVTTPS2PI mmxreg,xmmrm KATMAI,SSE,MMX -CVTTSS2SI reg32,xmmrm KATMAI,SSE,SD,AR1 -CVTTSS2SI reg64,xmmrm X64,SSE,SD,AR1 -DIVPS xmmreg,xmmrm KATMAI,SSE -DIVSS xmmreg,xmmrm KATMAI,SSE -LDMXCSR mem KATMAI,SSE,SD -MAXPS xmmreg,xmmrm KATMAI,SSE -MAXSS xmmreg,xmmrm KATMAI,SSE -MINPS xmmreg,xmmrm KATMAI,SSE -MINSS xmmreg,xmmrm KATMAI,SSE -MOVAPS xmmreg,mem KATMAI,SSE -MOVAPS mem,xmmreg KATMAI,SSE -MOVAPS xmmreg,xmmreg KATMAI,SSE -MOVAPS xmmreg,xmmreg KATMAI,SSE -MOVHPS xmmreg,mem KATMAI,SSE -MOVHPS mem,xmmreg KATMAI,SSE -MOVLHPS xmmreg,xmmreg KATMAI,SSE -MOVLPS xmmreg,mem KATMAI,SSE -MOVLPS mem,xmmreg KATMAI,SSE -MOVHLPS xmmreg,xmmreg KATMAI,SSE -MOVMSKPS reg32,xmmreg KATMAI,SSE -MOVMSKPS reg64,xmmreg X64,SSE -MOVNTPS mem,xmmreg KATMAI,SSE -MOVSS xmmreg,mem KATMAI,SSE -MOVSS mem,xmmreg KATMAI,SSE -MOVSS xmmreg,xmmreg KATMAI,SSE -MOVSS xmmreg,xmmreg KATMAI,SSE -MOVUPS xmmreg,mem KATMAI,SSE -MOVUPS mem,xmmreg KATMAI,SSE -MOVUPS xmmreg,xmmreg KATMAI,SSE -MOVUPS xmmreg,xmmreg KATMAI,SSE -MULPS xmmreg,xmmrm KATMAI,SSE -MULSS xmmreg,xmmrm KATMAI,SSE -ORPS xmmreg,xmmrm KATMAI,SSE -RCPPS xmmreg,xmmrm KATMAI,SSE -RCPSS xmmreg,xmmrm KATMAI,SSE -RSQRTPS xmmreg,xmmrm KATMAI,SSE -RSQRTSS xmmreg,xmmrm KATMAI,SSE -SHUFPS xmmreg,mem,imm KATMAI,SSE -SHUFPS xmmreg,xmmreg,imm KATMAI,SSE -SQRTPS xmmreg,xmmrm KATMAI,SSE -SQRTSS xmmreg,xmmrm KATMAI,SSE -STMXCSR mem KATMAI,SSE,SD -SUBPS xmmreg,xmmrm KATMAI,SSE -SUBSS xmmreg,xmmrm KATMAI,SSE -UCOMISS xmmreg,xmmrm KATMAI,SSE -UNPCKHPS xmmreg,xmmrm KATMAI,SSE -UNPCKLPS xmmreg,xmmrm KATMAI,SSE -XORPS xmmreg,xmmrm KATMAI,SSE -</pre> -<h4><a name="section-B.1.4">B.1.4 Introduced in Deschutes but necessary for SSE support</a></h4> -<p><pre> -FXRSTOR mem P6,SSE,FPU -FXSAVE mem P6,SSE,FPU -</pre> -<h4><a name="section-B.1.5">B.1.5 XSAVE group (AVX and extended state)</a></h4> -<p><pre> -XGETBV NEHALEM -XSETBV NEHALEM,PRIV -XSAVE mem NEHALEM -XRSTOR mem NEHALEM -</pre> -<h4><a name="section-B.1.6">B.1.6 Generic memory operations</a></h4> -<p><pre> -PREFETCHNTA mem KATMAI -PREFETCHT0 mem KATMAI -PREFETCHT1 mem KATMAI -PREFETCHT2 mem KATMAI -SFENCE KATMAI -</pre> -<h4><a name="section-B.1.7">B.1.7 New MMX instructions introduced in Katmai</a></h4> -<p><pre> -MASKMOVQ mmxreg,mmxreg KATMAI,MMX -MOVNTQ mem,mmxreg KATMAI,MMX -PAVGB mmxreg,mmxrm KATMAI,MMX -PAVGW mmxreg,mmxrm KATMAI,MMX -PEXTRW reg32,mmxreg,imm KATMAI,MMX -PINSRW mmxreg,mem,imm KATMAI,MMX -PINSRW mmxreg,rm16,imm KATMAI,MMX -PINSRW mmxreg,reg32,imm KATMAI,MMX -PMAXSW mmxreg,mmxrm KATMAI,MMX -PMAXUB mmxreg,mmxrm KATMAI,MMX -PMINSW mmxreg,mmxrm KATMAI,MMX -PMINUB mmxreg,mmxrm KATMAI,MMX -PMOVMSKB reg32,mmxreg KATMAI,MMX -PMULHUW mmxreg,mmxrm KATMAI,MMX -PSADBW mmxreg,mmxrm KATMAI,MMX -PSHUFW mmxreg,mmxrm,imm KATMAI,MMX2 -</pre> -<h4><a name="section-B.1.8">B.1.8 AMD Enhanced 3DNow! (Athlon) instructions</a></h4> -<p><pre> -PF2IW mmxreg,mmxrm PENT,3DNOW -PFNACC mmxreg,mmxrm PENT,3DNOW -PFPNACC mmxreg,mmxrm PENT,3DNOW -PI2FW mmxreg,mmxrm PENT,3DNOW -PSWAPD mmxreg,mmxrm PENT,3DNOW -</pre> -<h4><a name="section-B.1.9">B.1.9 Willamette SSE2 Cacheability Instructions</a></h4> -<p><pre> -MASKMOVDQU xmmreg,xmmreg WILLAMETTE,SSE2 -CLFLUSH mem WILLAMETTE,SSE2 -MOVNTDQ mem,xmmreg WILLAMETTE,SSE2,SO -MOVNTI mem,reg32 WILLAMETTE,SD -MOVNTI mem,reg64 X64 -MOVNTPD mem,xmmreg WILLAMETTE,SSE2,SO -LFENCE WILLAMETTE,SSE2 -MFENCE WILLAMETTE,SSE2 -</pre> -<h4><a name="section-B.1.10">B.1.10 Willamette MMX instructions (SSE2 SIMD Integer Instructions)</a></h4> -<p><pre> -MOVD mem,xmmreg WILLAMETTE,SSE2,SD -MOVD xmmreg,mem WILLAMETTE,SSE2,SD -MOVD xmmreg,rm32 WILLAMETTE,SSE2 -MOVD rm32,xmmreg WILLAMETTE,SSE2 -MOVDQA xmmreg,xmmreg WILLAMETTE,SSE2 -MOVDQA mem,xmmreg WILLAMETTE,SSE2,SO -MOVDQA xmmreg,mem WILLAMETTE,SSE2,SO -MOVDQA xmmreg,xmmreg WILLAMETTE,SSE2 -MOVDQU xmmreg,xmmreg WILLAMETTE,SSE2 -MOVDQU mem,xmmreg WILLAMETTE,SSE2,SO -MOVDQU xmmreg,mem WILLAMETTE,SSE2,SO -MOVDQU xmmreg,xmmreg WILLAMETTE,SSE2 -MOVDQ2Q mmxreg,xmmreg WILLAMETTE,SSE2 -MOVQ xmmreg,xmmreg WILLAMETTE,SSE2 -MOVQ xmmreg,xmmreg WILLAMETTE,SSE2 -MOVQ mem,xmmreg WILLAMETTE,SSE2 -MOVQ xmmreg,mem WILLAMETTE,SSE2 -MOVQ xmmreg,rm64 X64,SSE2 -MOVQ rm64,xmmreg X64,SSE2 -MOVQ2DQ xmmreg,mmxreg WILLAMETTE,SSE2 -PACKSSWB xmmreg,xmmrm WILLAMETTE,SSE2,SO -PACKSSDW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PACKUSWB xmmreg,xmmrm WILLAMETTE,SSE2,SO -PADDB xmmreg,xmmrm WILLAMETTE,SSE2,SO -PADDW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PADDD xmmreg,xmmrm WILLAMETTE,SSE2,SO -PADDQ mmxreg,mmxrm WILLAMETTE,MMX -PADDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -PADDSB xmmreg,xmmrm WILLAMETTE,SSE2,SO -PADDSW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PADDUSB xmmreg,xmmrm WILLAMETTE,SSE2,SO -PADDUSW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PAND xmmreg,xmmrm WILLAMETTE,SSE2,SO -PANDN xmmreg,xmmrm WILLAMETTE,SSE2,SO -PAVGB xmmreg,xmmrm WILLAMETTE,SSE2,SO -PAVGW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PCMPEQB xmmreg,xmmrm WILLAMETTE,SSE2,SO -PCMPEQW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PCMPEQD xmmreg,xmmrm WILLAMETTE,SSE2,SO -PCMPGTB xmmreg,xmmrm WILLAMETTE,SSE2,SO -PCMPGTW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PCMPGTD xmmreg,xmmrm WILLAMETTE,SSE2,SO -PEXTRW reg32,xmmreg,imm WILLAMETTE,SSE2 -PINSRW xmmreg,reg16,imm WILLAMETTE,SSE2 -PINSRW xmmreg,reg32,imm WILLAMETTE,SSE2,ND -PINSRW xmmreg,mem,imm WILLAMETTE,SSE2 -PINSRW xmmreg,mem16,imm WILLAMETTE,SSE2 -PMADDWD xmmreg,xmmrm WILLAMETTE,SSE2,SO -PMAXSW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PMAXUB xmmreg,xmmrm WILLAMETTE,SSE2,SO -PMINSW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PMINUB xmmreg,xmmrm WILLAMETTE,SSE2,SO -PMOVMSKB reg32,xmmreg WILLAMETTE,SSE2 -PMULHUW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PMULHW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PMULLW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PMULUDQ mmxreg,mmxrm WILLAMETTE,SSE2,SO -PMULUDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -POR xmmreg,xmmrm WILLAMETTE,SSE2,SO -PSADBW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PSHUFD xmmreg,xmmreg,imm WILLAMETTE,SSE2 -PSHUFD xmmreg,mem,imm WILLAMETTE,SSE22 -PSHUFHW xmmreg,xmmreg,imm WILLAMETTE,SSE2 -PSHUFHW xmmreg,mem,imm WILLAMETTE,SSE22 -PSHUFLW xmmreg,xmmreg,imm WILLAMETTE,SSE2 -PSHUFLW xmmreg,mem,imm WILLAMETTE,SSE22 -PSLLDQ xmmreg,imm WILLAMETTE,SSE2,AR1 -PSLLW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PSLLW xmmreg,imm WILLAMETTE,SSE2,AR1 -PSLLD xmmreg,xmmrm WILLAMETTE,SSE2,SO -PSLLD xmmreg,imm WILLAMETTE,SSE2,AR1 -PSLLQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -PSLLQ xmmreg,imm WILLAMETTE,SSE2,AR1 -PSRAW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PSRAW xmmreg,imm WILLAMETTE,SSE2,AR1 -PSRAD xmmreg,xmmrm WILLAMETTE,SSE2,SO -PSRAD xmmreg,imm WILLAMETTE,SSE2,AR1 -PSRLDQ xmmreg,imm WILLAMETTE,SSE2,AR1 -PSRLW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PSRLW xmmreg,imm WILLAMETTE,SSE2,AR1 -PSRLD xmmreg,xmmrm WILLAMETTE,SSE2,SO -PSRLD xmmreg,imm WILLAMETTE,SSE2,AR1 -PSRLQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -PSRLQ xmmreg,imm WILLAMETTE,SSE2,AR1 -PSUBB xmmreg,xmmrm WILLAMETTE,SSE2,SO -PSUBW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PSUBD xmmreg,xmmrm WILLAMETTE,SSE2,SO -PSUBQ mmxreg,mmxrm WILLAMETTE,SSE2,SO -PSUBQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -PSUBSB xmmreg,xmmrm WILLAMETTE,SSE2,SO -PSUBSW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PSUBUSB xmmreg,xmmrm WILLAMETTE,SSE2,SO -PSUBUSW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PUNPCKHBW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PUNPCKHWD xmmreg,xmmrm WILLAMETTE,SSE2,SO -PUNPCKHDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -PUNPCKHQDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -PUNPCKLBW xmmreg,xmmrm WILLAMETTE,SSE2,SO -PUNPCKLWD xmmreg,xmmrm WILLAMETTE,SSE2,SO -PUNPCKLDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -PUNPCKLQDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -PXOR xmmreg,xmmrm WILLAMETTE,SSE2,SO -</pre> -<h4><a name="section-B.1.11">B.1.11 Willamette Streaming SIMD instructions (SSE2)</a></h4> -<p><pre> -ADDPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -ADDSD xmmreg,xmmrm WILLAMETTE,SSE2 -ANDNPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -ANDPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -CMPEQPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -CMPEQSD xmmreg,xmmrm WILLAMETTE,SSE2 -CMPLEPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -CMPLESD xmmreg,xmmrm WILLAMETTE,SSE2 -CMPLTPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -CMPLTSD xmmreg,xmmrm WILLAMETTE,SSE2 -CMPNEQPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -CMPNEQSD xmmreg,xmmrm WILLAMETTE,SSE2 -CMPNLEPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -CMPNLESD xmmreg,xmmrm WILLAMETTE,SSE2 -CMPNLTPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -CMPNLTSD xmmreg,xmmrm WILLAMETTE,SSE2 -CMPORDPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -CMPORDSD xmmreg,xmmrm WILLAMETTE,SSE2 -CMPUNORDPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -CMPUNORDSD xmmreg,xmmrm WILLAMETTE,SSE2 -CMPPD xmmreg,xmmrm,imm WILLAMETTE,SSE22 -CMPSD xmmreg,xmmrm,imm WILLAMETTE,SSE2 -COMISD xmmreg,xmmrm WILLAMETTE,SSE2 -CVTDQ2PD xmmreg,xmmrm WILLAMETTE,SSE2 -CVTDQ2PS xmmreg,xmmrm WILLAMETTE,SSE2,SO -CVTPD2DQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -CVTPD2PI mmxreg,xmmrm WILLAMETTE,SSE2,SO -CVTPD2PS xmmreg,xmmrm WILLAMETTE,SSE2,SO -CVTPI2PD xmmreg,mmxrm WILLAMETTE,SSE2 -CVTPS2DQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -CVTPS2PD xmmreg,xmmrm WILLAMETTE,SSE2 -CVTSD2SI reg32,xmmreg WILLAMETTE,SSE2,AR1 -CVTSD2SI reg32,mem WILLAMETTE,SSE2,AR1 -CVTSD2SI reg64,xmmreg X64,SSE2,AR1 -CVTSD2SI reg64,mem X64,SSE2,AR1 -CVTSD2SS xmmreg,xmmrm WILLAMETTE,SSE2 -CVTSI2SD xmmreg,mem WILLAMETTE,SSE2,SD,AR1,ND -CVTSI2SD xmmreg,rm32 WILLAMETTE,SSE2,SD,AR1 -CVTSI2SD xmmreg,rm64 X64,SSE2,AR1 -CVTSS2SD xmmreg,xmmrm WILLAMETTE,SSE2,SD -CVTTPD2PI mmxreg,xmmrm WILLAMETTE,SSE2,SO -CVTTPD2DQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -CVTTPS2DQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -CVTTSD2SI reg32,xmmreg WILLAMETTE,SSE2,AR1 -CVTTSD2SI reg32,mem WILLAMETTE,SSE2,AR1 -CVTTSD2SI reg64,xmmreg X64,SSE2,AR1 -CVTTSD2SI reg64,mem X64,SSE2,AR1 -DIVPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -DIVSD xmmreg,xmmrm WILLAMETTE,SSE2 -MAXPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -MAXSD xmmreg,xmmrm WILLAMETTE,SSE2 -MINPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -MINSD xmmreg,xmmrm WILLAMETTE,SSE2 -MOVAPD xmmreg,xmmreg WILLAMETTE,SSE2 -MOVAPD xmmreg,xmmreg WILLAMETTE,SSE2 -MOVAPD mem,xmmreg WILLAMETTE,SSE2,SO -MOVAPD xmmreg,mem WILLAMETTE,SSE2,SO -MOVHPD mem,xmmreg WILLAMETTE,SSE2 -MOVHPD xmmreg,mem WILLAMETTE,SSE2 -MOVLPD mem,xmmreg WILLAMETTE,SSE2 -MOVLPD xmmreg,mem WILLAMETTE,SSE2 -MOVMSKPD reg32,xmmreg WILLAMETTE,SSE2 -MOVMSKPD reg64,xmmreg X64,SSE2 -MOVSD xmmreg,xmmreg WILLAMETTE,SSE2 -MOVSD xmmreg,xmmreg WILLAMETTE,SSE2 -MOVSD mem,xmmreg WILLAMETTE,SSE2 -MOVSD xmmreg,mem WILLAMETTE,SSE2 -MOVUPD xmmreg,xmmreg WILLAMETTE,SSE2 -MOVUPD xmmreg,xmmreg WILLAMETTE,SSE2 -MOVUPD mem,xmmreg WILLAMETTE,SSE2,SO -MOVUPD xmmreg,mem WILLAMETTE,SSE2,SO -MULPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -MULSD xmmreg,xmmrm WILLAMETTE,SSE2 -ORPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -SHUFPD xmmreg,xmmreg,imm WILLAMETTE,SSE2 -SHUFPD xmmreg,mem,imm WILLAMETTE,SSE2 -SQRTPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -SQRTSD xmmreg,xmmrm WILLAMETTE,SSE2 -SUBPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -SUBSD xmmreg,xmmrm WILLAMETTE,SSE2 -UCOMISD xmmreg,xmmrm WILLAMETTE,SSE2 -UNPCKHPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -UNPCKLPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -XORPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -</pre> -<h4><a name="section-B.1.12">B.1.12 Prescott New Instructions (SSE3)</a></h4> -<p><pre> -ADDSUBPD xmmreg,xmmrm PRESCOTT,SSE3,SO -ADDSUBPS xmmreg,xmmrm PRESCOTT,SSE3,SO -HADDPD xmmreg,xmmrm PRESCOTT,SSE3,SO -HADDPS xmmreg,xmmrm PRESCOTT,SSE3,SO -HSUBPD xmmreg,xmmrm PRESCOTT,SSE3,SO -HSUBPS xmmreg,xmmrm PRESCOTT,SSE3,SO -LDDQU xmmreg,mem PRESCOTT,SSE3,SO -MOVDDUP xmmreg,xmmrm PRESCOTT,SSE3 -MOVSHDUP xmmreg,xmmrm PRESCOTT,SSE3 -MOVSLDUP xmmreg,xmmrm PRESCOTT,SSE3 -</pre> -<h4><a name="section-B.1.13">B.1.13 VMX Instructions</a></h4> -<p><pre> -VMCALL VMX -VMCLEAR mem VMX -VMLAUNCH VMX -VMLOAD X64,VMX -VMMCALL X64,VMX -VMPTRLD mem VMX -VMPTRST mem VMX -VMREAD rm32,reg32 VMX,NOLONG,SD -VMREAD rm64,reg64 X64,VMX -VMRESUME VMX -VMRUN X64,VMX -VMSAVE X64,VMX -VMWRITE reg32,rm32 VMX,NOLONG,SD -VMWRITE reg64,rm64 X64,VMX -VMXOFF VMX -VMXON mem VMX -</pre> -<h4><a name="section-B.1.14">B.1.14 Extended Page Tables VMX instructions</a></h4> -<p><pre> -INVEPT reg32,mem VMX,SO,NOLONG -INVEPT reg64,mem VMX,SO,LONG -INVVPID reg32,mem VMX,SO,NOLONG -INVVPID reg64,mem VMX,SO,LONG -</pre> -<h4><a name="section-B.1.15">B.1.15 Tejas New Instructions (SSSE3)</a></h4> -<p><pre> -PABSB mmxreg,mmxrm SSSE3,MMX -PABSB xmmreg,xmmrm SSSE3 -PABSW mmxreg,mmxrm SSSE3,MMX -PABSW xmmreg,xmmrm SSSE3 -PABSD mmxreg,mmxrm SSSE3,MMX -PABSD xmmreg,xmmrm SSSE3 -PALIGNR mmxreg,mmxrm,imm SSSE3,MMX -PALIGNR xmmreg,xmmrm,imm SSSE3 -PHADDW mmxreg,mmxrm SSSE3,MMX -PHADDW xmmreg,xmmrm SSSE3 -PHADDD mmxreg,mmxrm SSSE3,MMX -PHADDD xmmreg,xmmrm SSSE3 -PHADDSW mmxreg,mmxrm SSSE3,MMX -PHADDSW xmmreg,xmmrm SSSE3 -PHSUBW mmxreg,mmxrm SSSE3,MMX -PHSUBW xmmreg,xmmrm SSSE3 -PHSUBD mmxreg,mmxrm SSSE3,MMX -PHSUBD xmmreg,xmmrm SSSE3 -PHSUBSW mmxreg,mmxrm SSSE3,MMX -PHSUBSW xmmreg,xmmrm SSSE3 -PMADDUBSW mmxreg,mmxrm SSSE3,MMX -PMADDUBSW xmmreg,xmmrm SSSE3 -PMULHRSW mmxreg,mmxrm SSSE3,MMX -PMULHRSW xmmreg,xmmrm SSSE3 -PSHUFB mmxreg,mmxrm SSSE3,MMX -PSHUFB xmmreg,xmmrm SSSE3 -PSIGNB mmxreg,mmxrm SSSE3,MMX -PSIGNB xmmreg,xmmrm SSSE3 -PSIGNW mmxreg,mmxrm SSSE3,MMX -PSIGNW xmmreg,xmmrm SSSE3 -PSIGND mmxreg,mmxrm SSSE3,MMX -PSIGND xmmreg,xmmrm SSSE3 -</pre> -<h4><a name="section-B.1.16">B.1.16 AMD SSE4A</a></h4> -<p><pre> -EXTRQ xmmreg,imm,imm SSE4A,AMD -EXTRQ xmmreg,xmmreg SSE4A,AMD -INSERTQ xmmreg,xmmreg,imm,imm SSE4A,AMD -INSERTQ xmmreg,xmmreg SSE4A,AMD -MOVNTSD mem,xmmreg SSE4A,AMD -MOVNTSS mem,xmmreg SSE4A,AMD,SD -</pre> -<h4><a name="section-B.1.17">B.1.17 New instructions in Barcelona</a></h4> -<p><pre> -LZCNT reg16,rm16 P6,AMD -LZCNT reg32,rm32 P6,AMD -LZCNT reg64,rm64 X64,AMD -</pre> -<h4><a name="section-B.1.18">B.1.18 Penryn New Instructions (SSE4.1)</a></h4> -<p><pre> -BLENDPD xmmreg,xmmrm,imm SSE41 -BLENDPS xmmreg,xmmrm,imm SSE41 -BLENDVPD xmmreg,xmmrm,xmm0 SSE41 -BLENDVPS xmmreg,xmmrm,xmm0 SSE41 -DPPD xmmreg,xmmrm,imm SSE41 -DPPS xmmreg,xmmrm,imm SSE41 -EXTRACTPS rm32,xmmreg,imm SSE41 -EXTRACTPS reg64,xmmreg,imm SSE41,X64 -INSERTPS xmmreg,xmmrm,imm SSE41,SD -MOVNTDQA xmmreg,mem SSE41 -MPSADBW xmmreg,xmmrm,imm SSE41 -PACKUSDW xmmreg,xmmrm SSE41 -PBLENDVB xmmreg,xmmrm,xmm0 SSE41 -PBLENDW xmmreg,xmmrm,imm SSE41 -PCMPEQQ xmmreg,xmmrm SSE41 -PEXTRB reg32,xmmreg,imm SSE41 -PEXTRB mem8,xmmreg,imm SSE41 -PEXTRB reg64,xmmreg,imm SSE41,X64 -PEXTRD rm32,xmmreg,imm SSE41 -PEXTRQ rm64,xmmreg,imm SSE41,X64 -PEXTRW reg32,xmmreg,imm SSE41 -PEXTRW mem16,xmmreg,imm SSE41 -PEXTRW reg64,xmmreg,imm SSE41,X64 -PHMINPOSUW xmmreg,xmmrm SSE41 -PINSRB xmmreg,mem,imm SSE41 -PINSRB xmmreg,rm8,imm SSE41 -PINSRB xmmreg,reg32,imm SSE41 -PINSRD xmmreg,mem,imm SSE41 -PINSRD xmmreg,rm32,imm SSE41 -PINSRQ xmmreg,mem,imm SSE41,X64 -PINSRQ xmmreg,rm64,imm SSE41,X64 -PMAXSB xmmreg,xmmrm SSE41 -PMAXSD xmmreg,xmmrm SSE41 -PMAXUD xmmreg,xmmrm SSE41 -PMAXUW xmmreg,xmmrm SSE41 -PMINSB xmmreg,xmmrm SSE41 -PMINSD xmmreg,xmmrm SSE41 -PMINUD xmmreg,xmmrm SSE41 -PMINUW xmmreg,xmmrm SSE41 -PMOVSXBW xmmreg,xmmrm SSE41 -PMOVSXBD xmmreg,xmmrm SSE41,SD -PMOVSXBQ xmmreg,xmmrm SSE41,SW -PMOVSXWD xmmreg,xmmrm SSE41 -PMOVSXWQ xmmreg,xmmrm SSE41,SD -PMOVSXDQ xmmreg,xmmrm SSE41 -PMOVZXBW xmmreg,xmmrm SSE41 -PMOVZXBD xmmreg,xmmrm SSE41,SD -PMOVZXBQ xmmreg,xmmrm SSE41,SW -PMOVZXWD xmmreg,xmmrm SSE41 -PMOVZXWQ xmmreg,xmmrm SSE41,SD -PMOVZXDQ xmmreg,xmmrm SSE41 -PMULDQ xmmreg,xmmrm SSE41 -PMULLD xmmreg,xmmrm SSE41 -PTEST xmmreg,xmmrm SSE41 -ROUNDPD xmmreg,xmmrm,imm SSE41 -ROUNDPS xmmreg,xmmrm,imm SSE41 -ROUNDSD xmmreg,xmmrm,imm SSE41 -ROUNDSS xmmreg,xmmrm,imm SSE41 -</pre> -<h4><a name="section-B.1.19">B.1.19 Nehalem New Instructions (SSE4.2)</a></h4> -<p><pre> -CRC32 reg32,rm8 SSE42 -CRC32 reg32,rm16 SSE42 -CRC32 reg32,rm32 SSE42 -CRC32 reg64,rm8 SSE42,X64 -CRC32 reg64,rm64 SSE42,X64 -PCMPESTRI xmmreg,xmmrm,imm SSE42 -PCMPESTRM xmmreg,xmmrm,imm SSE42 -PCMPISTRI xmmreg,xmmrm,imm SSE42 -PCMPISTRM xmmreg,xmmrm,imm SSE42 -PCMPGTQ xmmreg,xmmrm SSE42 -POPCNT reg16,rm16 NEHALEM,SW -POPCNT reg32,rm32 NEHALEM,SD -POPCNT reg64,rm64 NEHALEM,X64 -</pre> -<h4><a name="section-B.1.20">B.1.20 Intel SMX</a></h4> -<p><pre> -GETSEC KATMAI -</pre> -<h4><a name="section-B.1.21">B.1.21 Geode (Cyrix) 3DNow! additions</a></h4> -<p><pre> -PFRCPV mmxreg,mmxrm PENT,3DNOW,CYRIX -PFRSQRTV mmxreg,mmxrm PENT,3DNOW,CYRIX -</pre> -<h4><a name="section-B.1.22">B.1.22 Intel new instructions in ???</a></h4> -<p><pre> -MOVBE reg16,mem16 NEHALEM -MOVBE reg32,mem32 NEHALEM -MOVBE reg64,mem64 NEHALEM -MOVBE mem16,reg16 NEHALEM -MOVBE mem32,reg32 NEHALEM -MOVBE mem64,reg64 NEHALEM -</pre> -<h4><a name="section-B.1.23">B.1.23 Intel AES instructions</a></h4> -<p><pre> -AESENC xmmreg,xmmrm128 SSE,WESTMERE -AESENCLAST xmmreg,xmmrm128 SSE,WESTMERE -AESDEC xmmreg,xmmrm128 SSE,WESTMERE -AESDECLAST xmmreg,xmmrm128 SSE,WESTMERE -AESIMC xmmreg,xmmrm128 SSE,WESTMERE -AESKEYGENASSIST xmmreg,xmmrm128,imm8 SSE,WESTMERE -</pre> -<h4><a name="section-B.1.24">B.1.24 Intel AVX AES instructions</a></h4> -<p><pre> -VAESENC xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VAESENCLAST xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VAESDEC xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VAESDECLAST xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VAESIMC xmmreg,xmmrm128 AVX,SANDYBRIDGE -VAESKEYGENASSIST xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -</pre> -<h4><a name="section-B.1.25">B.1.25 Intel AVX instructions</a></h4> -<p><pre> -VADDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VADDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VADDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VADDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VADDSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VADDSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VADDSUBPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VADDSUBPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VADDSUBPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VADDSUBPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VANDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VANDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VANDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VANDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VANDNPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VANDNPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VANDNPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VANDNPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VBLENDPD xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -VBLENDPD ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE -VBLENDPS xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -VBLENDPS ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE -VBLENDVPD xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE -VBLENDVPD xmmreg,xmmrm128,xmm0 AVX,SANDYBRIDGE -VBLENDVPD ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE -VBLENDVPD ymmreg,ymmrm256,ymm0 AVX,SANDYBRIDGE -VBLENDVPS xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE -VBLENDVPS xmmreg,xmmrm128,xmm0 AVX,SANDYBRIDGE -VBLENDVPS ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE -VBLENDVPD ymmreg,ymmrm256,ymm0 AVX,SANDYBRIDGE -VBROADCASTSS xmmreg,mem32 AVX,SANDYBRIDGE -VBROADCASTSS ymmreg,mem32 AVX,SANDYBRIDGE -VBROADCASTSD ymmreg,mem64 AVX,SANDYBRIDGE -VBROADCASTF128 ymmreg,mem128 AVX,SANDYBRIDGE -VCMPEQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPEQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPLTPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPLTPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPLEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPLEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPUNORDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPUNORDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNEQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNEQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNLTPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNLTPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNLEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNLEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPORDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPORDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPEQ_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPEQ_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNGEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNGEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNGTPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNGTPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPFALSEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPFALSEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNEQ_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNEQ_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPGEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPGEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPGTPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPGTPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPTRUEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPTRUEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPEQ_OSPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPEQ_OSPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPLT_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPLT_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPLE_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPLE_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPUNORD_SPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPUNORD_SPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNEQ_USPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNEQ_USPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNLT_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNLT_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNLE_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNLE_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPORD_SPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPORD_SPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPEQ_USPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPEQ_USPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNGE_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNGE_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNGT_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNGT_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPFALSE_OSPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPFALSE_OSPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNEQ_OSPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNEQ_OSPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPGE_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPGE_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPGT_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPGT_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPTRUE_USPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPTRUE_USPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPPD xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -VCMPPD ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE -VCMPEQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPEQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPLTPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPLTPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPLEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPLEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPUNORDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPUNORDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNEQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNEQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNLTPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNLTPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNLEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNLEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPORDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPORDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPEQ_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPEQ_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNGEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNGEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNGTPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNGTPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPFALSEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPFALSEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNEQ_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNEQ_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPGEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPGEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPGTPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPGTPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPTRUEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPTRUEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPEQ_OSPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPEQ_OSPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPLT_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPLT_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPLE_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPLE_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPUNORD_SPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPUNORD_SPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNEQ_USPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNEQ_USPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNLT_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNLT_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNLE_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNLE_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPORD_SPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPORD_SPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPEQ_USPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPEQ_USPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNGE_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNGE_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNGT_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNGT_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPFALSE_OSPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPFALSE_OSPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPNEQ_OSPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPNEQ_OSPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPGE_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPGE_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPGT_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPGT_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPTRUE_USPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VCMPTRUE_USPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VCMPPS xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -VCMPPS ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE -VCMPEQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPLTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPLESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPUNORDSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPNEQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPNLTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPNLESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPORDSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPEQ_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPNGESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPNGTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPFALSESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPNEQ_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPGESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPGTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPTRUESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPEQ_OSSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPLT_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPLE_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPUNORD_SSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPNEQ_USSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPNLT_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPNLE_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPORD_SSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPEQ_USSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPNGE_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPNGT_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPFALSE_OSSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPNEQ_OSSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPGE_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPGT_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPTRUE_USSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCMPSD xmmreg,xmmreg*,xmmrm64,imm8 AVX,SANDYBRIDGE -VCMPEQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPLTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPLESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPUNORDSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPNEQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPNLTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPNLESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPORDSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPEQ_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPNGESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPNGTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPFALSESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPNEQ_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPGESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPGTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPTRUESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPEQ_OSSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPLT_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPLE_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPUNORD_SSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPNEQ_USSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPNLT_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPNLE_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPORD_SSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPEQ_USSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPNGE_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPNGT_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPFALSE_OSSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPNEQ_OSSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPGE_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPGT_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPTRUE_USSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCMPSS xmmreg,xmmreg*,xmmrm32,imm8 AVX,SANDYBRIDGE -VCOMISD xmmreg,xmmrm64 AVX,SANDYBRIDGE -VCOMISS xmmreg,xmmrm32 AVX,SANDYBRIDGE -VCVTDQ2PD xmmreg,xmmrm64 AVX,SANDYBRIDGE -VCVTDQ2PD ymmreg,xmmrm128 AVX,SANDYBRIDGE -VCVTDQ2PS xmmreg,xmmrm128 AVX,SANDYBRIDGE -VCVTDQ2PS ymmreg,ymmrm256 AVX,SANDYBRIDGE -VCVTPD2DQ xmmreg,xmmreg AVX,SANDYBRIDGE -VCVTPD2DQ xmmreg,mem128 AVX,SANDYBRIDGE,SO -VCVTPD2DQ xmmreg,ymmreg AVX,SANDYBRIDGE -VCVTPD2DQ xmmreg,mem256 AVX,SANDYBRIDGE,SY -VCVTPD2PS xmmreg,xmmreg AVX,SANDYBRIDGE -VCVTPD2PS xmmreg,mem128 AVX,SANDYBRIDGE,SO -VCVTPD2PS xmmreg,ymmreg AVX,SANDYBRIDGE -VCVTPD2PS xmmreg,mem256 AVX,SANDYBRIDGE,SY -VCVTPS2DQ xmmreg,xmmrm128 AVX,SANDYBRIDGE -VCVTPS2DQ ymmreg,ymmrm256 AVX,SANDYBRIDGE -VCVTPS2PD xmmreg,xmmrm64 AVX,SANDYBRIDGE -VCVTPS2PD ymmreg,xmmrm128 AVX,SANDYBRIDGE -VCVTSD2SI reg32,xmmrm64 AVX,SANDYBRIDGE -VCVTSD2SI reg64,xmmrm64 AVX,SANDYBRIDGE,LONG -VCVTSD2SS xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VCVTSI2SD xmmreg,xmmreg*,rm32 AVX,SANDYBRIDGE,SD -VCVTSI2SD xmmreg,xmmreg*,mem32 AVX,SANDYBRIDGE,ND,SD -VCVTSI2SD xmmreg,xmmreg*,rm64 AVX,SANDYBRIDGE,LONG -VCVTSI2SS xmmreg,xmmreg*,rm32 AVX,SANDYBRIDGE,SD -VCVTSI2SS xmmreg,xmmreg*,mem32 AVX,SANDYBRIDGE,ND,SD -VCVTSI2SS xmmreg,xmmreg*,rm64 AVX,SANDYBRIDGE,LONG -VCVTSS2SD xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VCVTSS2SI reg32,xmmrm32 AVX,SANDYBRIDGE -VCVTSS2SI reg64,xmmrm32 AVX,SANDYBRIDGE,LONG -VCVTTPD2DQ xmmreg,xmmreg AVX,SANDYBRIDGE -VCVTTPD2DQ xmmreg,mem128 AVX,SANDYBRIDGE,SO -VCVTTPD2DQ xmmreg,ymmreg AVX,SANDYBRIDGE -VCVTTPD2DQ xmmreg,mem256 AVX,SANDYBRIDGE,SY -VCVTTPS2DQ xmmreg,xmmrm128 AVX,SANDYBRIDGE -VCVTTPS2DQ ymmreg,ymmrm256 AVX,SANDYBRIDGE -VCVTTSD2SI reg32,xmmrm64 AVX,SANDYBRIDGE -VCVTTSD2SI reg64,xmmrm64 AVX,SANDYBRIDGE,LONG -VCVTTSS2SI reg32,xmmrm32 AVX,SANDYBRIDGE -VCVTTSS2SI reg64,xmmrm32 AVX,SANDYBRIDGE,LONG -VDIVPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VDIVPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VDIVPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VDIVPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VDIVSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VDIVSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VDPPD xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -VDPPS xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -VDPPS ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE -VEXTRACTF128 xmmrm128,xmmreg,imm8 AVX,SANDYBRIDGE -VEXTRACTPS rm32,xmmreg,imm8 AVX,SANDYBRIDGE -VHADDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VHADDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VHADDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VHADDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VHSUBPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VHSUBPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VHSUBPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VHSUBPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VINSERTF128 ymmreg,ymmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -VINSERTPS xmmreg,xmmreg*,xmmrm32,imm8 AVX,SANDYBRIDGE -VLDDQU xmmreg,mem128 AVX,SANDYBRIDGE -VLDQQU ymmreg,mem256 AVX,SANDYBRIDGE -VLDDQU ymmreg,mem256 AVX,SANDYBRIDGE -VLDMXCSR mem32 AVX,SANDYBRIDGE -VMASKMOVDQU xmmreg,xmmreg AVX,SANDYBRIDGE -VMASKMOVPS xmmreg,xmmreg,mem128 AVX,SANDYBRIDGE -VMASKMOVPS ymmreg,ymmreg,mem256 AVX,SANDYBRIDGE -VMASKMOVPS mem128,xmmreg,xmmreg AVX,SANDYBRIDGE,SO -VMASKMOVPS mem256,xmmreg,xmmreg AVX,SANDYBRIDGE,SY -VMASKMOVPD xmmreg,xmmreg,mem128 AVX,SANDYBRIDGE -VMASKMOVPD ymmreg,ymmreg,mem256 AVX,SANDYBRIDGE -VMASKMOVPD mem128,xmmreg,xmmreg AVX,SANDYBRIDGE -VMASKMOVPD mem256,ymmreg,ymmreg AVX,SANDYBRIDGE -VMAXPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VMAXPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VMAXPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VMAXPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VMAXSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VMAXSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VMINPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VMINPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VMINPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VMINPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VMINSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VMINSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VMOVAPD xmmreg,xmmrm128 AVX,SANDYBRIDGE -VMOVAPD xmmrm128,xmmreg AVX,SANDYBRIDGE -VMOVAPD ymmreg,ymmrm256 AVX,SANDYBRIDGE -VMOVAPD ymmrm256,ymmreg AVX,SANDYBRIDGE -VMOVAPS xmmreg,xmmrm128 AVX,SANDYBRIDGE -VMOVAPS xmmrm128,xmmreg AVX,SANDYBRIDGE -VMOVAPS ymmreg,ymmrm256 AVX,SANDYBRIDGE -VMOVAPS ymmrm256,ymmreg AVX,SANDYBRIDGE -VMOVQ xmmreg,xmmrm64 AVX,SANDYBRIDGE -VMOVQ xmmrm64,xmmreg AVX,SANDYBRIDGE -VMOVQ xmmreg,rm64 AVX,SANDYBRIDGE,LONG -VMOVQ rm64,xmmreg AVX,SANDYBRIDGE,LONG -VMOVD xmmreg,rm32 AVX,SANDYBRIDGE -VMOVD rm32,xmmreg AVX,SANDYBRIDGE -VMOVDDUP xmmreg,xmmrm64 AVX,SANDYBRIDGE -VMOVDDUP ymmreg,ymmrm256 AVX,SANDYBRIDGE -VMOVDQA xmmreg,xmmrm128 AVX,SANDYBRIDGE -VMOVDQA xmmrm128,xmmreg AVX,SANDYBRIDGE -VMOVQQA ymmreg,ymmrm256 AVX,SANDYBRIDGE -VMOVQQA ymmrm256,ymmreg AVX,SANDYBRIDGE -VMOVDQA ymmreg,ymmrm AVX,SANDYBRIDGE -VMOVDQA ymmrm256,ymmreg AVX,SANDYBRIDGE -VMOVDQU xmmreg,xmmrm128 AVX,SANDYBRIDGE -VMOVDQU xmmrm128,xmmreg AVX,SANDYBRIDGE -VMOVQQU ymmreg,ymmrm256 AVX,SANDYBRIDGE -VMOVQQU ymmrm256,ymmreg AVX,SANDYBRIDGE -VMOVDQU ymmreg,ymmrm256 AVX,SANDYBRIDGE -VMOVDQU ymmrm256,ymmreg AVX,SANDYBRIDGE -VMOVHLPS xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE -VMOVHPD xmmreg,xmmreg*,mem64 AVX,SANDYBRIDGE -VMOVHPD mem64,xmmreg AVX,SANDYBRIDGE -VMOVHPS xmmreg,xmmreg*,mem64 AVX,SANDYBRIDGE -VMOVHPS mem64,xmmreg AVX,SANDYBRIDGE -VMOVLHPS xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE -VMOVLPD xmmreg,xmmreg*,mem64 AVX,SANDYBRIDGE -VMOVLPD mem64,xmmreg AVX,SANDYBRIDGE -VMOVLPS xmmreg,xmmreg*,mem64 AVX,SANDYBRIDGE -VMOVLPS mem64,xmmreg AVX,SANDYBRIDGE -VMOVMSKPD reg64,xmmreg AVX,SANDYBRIDGE,LONG -VMOVMSKPD reg32,xmmreg AVX,SANDYBRIDGE -VMOVMSKPD reg64,ymmreg AVX,SANDYBRIDGE,LONG -VMOVMSKPD reg32,ymmreg AVX,SANDYBRIDGE -VMOVMSKPS reg64,xmmreg AVX,SANDYBRIDGE,LONG -VMOVMSKPS reg32,xmmreg AVX,SANDYBRIDGE -VMOVMSKPS reg64,ymmreg AVX,SANDYBRIDGE,LONG -VMOVMSKPS reg32,ymmreg AVX,SANDYBRIDGE -VMOVNTDQ mem128,xmmreg AVX,SANDYBRIDGE -VMOVNTQQ mem256,ymmreg AVX,SANDYBRIDGE -VMOVNTDQ mem256,ymmreg AVX,SANDYBRIDGE -VMOVNTDQA xmmreg,mem128 AVX,SANDYBRIDGE -VMOVNTPD mem128,xmmreg AVX,SANDYBRIDGE -VMOVNTPD mem256,ymmreg AVX,SANDYBRIDGE -VMOVNTPS mem128,xmmreg AVX,SANDYBRIDGE -VMOVNTPS mem128,ymmreg AVX,SANDYBRIDGE -VMOVSD xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE -VMOVSD xmmreg,mem64 AVX,SANDYBRIDGE -VMOVSD xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE -VMOVSD mem64,xmmreg AVX,SANDYBRIDGE -VMOVSHDUP xmmreg,xmmrm128 AVX,SANDYBRIDGE -VMOVSHDUP ymmreg,ymmrm256 AVX,SANDYBRIDGE -VMOVSLDUP xmmreg,xmmrm128 AVX,SANDYBRIDGE -VMOVSLDUP ymmreg,ymmrm256 AVX,SANDYBRIDGE -VMOVSS xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE -VMOVSS xmmreg,mem64 AVX,SANDYBRIDGE -VMOVSS xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE -VMOVSS mem64,xmmreg AVX,SANDYBRIDGE -VMOVUPD xmmreg,xmmrm128 AVX,SANDYBRIDGE -VMOVUPD xmmrm128,xmmreg AVX,SANDYBRIDGE -VMOVUPD ymmreg,ymmrm256 AVX,SANDYBRIDGE -VMOVUPD ymmrm256,ymmreg AVX,SANDYBRIDGE -VMOVUPS xmmreg,xmmrm128 AVX,SANDYBRIDGE -VMOVUPS xmmrm128,xmmreg AVX,SANDYBRIDGE -VMOVUPS ymmreg,ymmrm256 AVX,SANDYBRIDGE -VMOVUPS ymmrm256,ymmreg AVX,SANDYBRIDGE -VMPSADBW xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -VMULPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VMULPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VMULPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VMULPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VMULSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VMULSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VORPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VORPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VORPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VORPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VPABSB xmmreg,xmmrm128 AVX,SANDYBRIDGE -VPABSW xmmreg,xmmrm128 AVX,SANDYBRIDGE -VPABSD xmmreg,xmmrm128 AVX,SANDYBRIDGE -VPACKSSWB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPACKSSDW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPACKUSWB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPACKUSDW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPADDB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPADDW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPADDD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPADDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPADDSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPADDSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPADDUSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPADDUSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPALIGNR xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -VPAND xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPANDN xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPAVGB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPAVGW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPBLENDVB xmmreg,xmmreg*,xmmrm128,xmmreg AVX,SANDYBRIDGE -VPBLENDW xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -VPCMPESTRI xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -VPCMPESTRM xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -VPCMPISTRI xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -VPCMPISTRM xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -VPCMPEQB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPCMPEQW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPCMPEQD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPCMPEQQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPCMPGTB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPCMPGTW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPCMPGTD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPCMPGTQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPERMILPD xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE -VPERMILPD ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE -VPERMILPD xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -VPERMILPD ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE -VPERMILTD2PD xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE -VPERMILTD2PD xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE -VPERMILTD2PD ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE -VPERMILTD2PD ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE -VPERMILMO2PD xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE -VPERMILMO2PD xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE -VPERMILMO2PD ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE -VPERMILMO2PD ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE -VPERMILMZ2PD xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE -VPERMILMZ2PD xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE -VPERMILMZ2PD ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE -VPERMILMZ2PD ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE -VPERMIL2PD xmmreg,xmmreg,xmmrm128,xmmreg,imm8 AVX,SANDYBRIDGE -VPERMIL2PD xmmreg,xmmreg,xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -VPERMIL2PD ymmreg,ymmreg,ymmrm256,ymmreg,imm8 AVX,SANDYBRIDGE -VPERMIL2PD ymmreg,ymmreg,ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE -VPERMILPS xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE -VPERMILPS ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE -VPERMILPS xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -VPERMILPS ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE -VPERMILTD2PS xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE -VPERMILTD2PS xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE -VPERMILTD2PS ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE -VPERMILTD2PS ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE -VPERMILMO2PS xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE -VPERMILMO2PS xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE -VPERMILMO2PS ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE -VPERMILMO2PS ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE -VPERMILMZ2PS xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE -VPERMILMZ2PS xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE -VPERMILMZ2PS ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE -VPERMILMZ2PS ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE -VPERMIL2PS xmmreg,xmmreg,xmmrm128,xmmreg,imm8 AVX,SANDYBRIDGE -VPERMIL2PS xmmreg,xmmreg,xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -VPERMIL2PS ymmreg,ymmreg,ymmrm256,ymmreg,imm8 AVX,SANDYBRIDGE -VPERMIL2PS ymmreg,ymmreg,ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE -VPERM2F128 ymmreg,ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE -VPEXTRB reg64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG -VPEXTRB reg32,xmmreg,imm8 AVX,SANDYBRIDGE -VPEXTRB mem8,xmmreg,imm8 AVX,SANDYBRIDGE -VPEXTRW reg64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG -VPEXTRW reg32,xmmreg,imm8 AVX,SANDYBRIDGE -VPEXTRW mem16,xmmreg,imm8 AVX,SANDYBRIDGE -VPEXTRW reg64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG -VPEXTRW reg32,xmmreg,imm8 AVX,SANDYBRIDGE -VPEXTRW mem16,xmmreg,imm8 AVX,SANDYBRIDGE -VPEXTRD reg64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG -VPEXTRD rm32,xmmreg,imm8 AVX,SANDYBRIDGE -VPEXTRQ rm64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG -VPHADDW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPHADDD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPHADDSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPHMINPOSUW xmmreg,xmmrm128 AVX,SANDYBRIDGE -VPHSUBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPHSUBD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPHSUBSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPINSRB xmmreg,xmmreg*,mem8,imm8 AVX,SANDYBRIDGE -VPINSRB xmmreg,xmmreg*,rm8,imm8 AVX,SANDYBRIDGE -VPINSRB xmmreg,xmmreg*,reg32,imm8 AVX,SANDYBRIDGE -VPINSRW xmmreg,xmmreg*,mem16,imm8 AVX,SANDYBRIDGE -VPINSRW xmmreg,xmmreg*,rm16,imm8 AVX,SANDYBRIDGE -VPINSRW xmmreg,xmmreg*,reg32,imm8 AVX,SANDYBRIDGE -VPINSRD xmmreg,xmmreg*,mem32,imm8 AVX,SANDYBRIDGE -VPINSRD xmmreg,xmmreg*,rm32,imm8 AVX,SANDYBRIDGE -VPINSRQ xmmreg,xmmreg*,mem64,imm8 AVX,SANDYBRIDGE,LONG -VPINSRQ xmmreg,xmmreg*,rm64,imm8 AVX,SANDYBRIDGE,LONG -VPMADDWD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPMADDUBSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPMAXSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPMAXSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPMAXSD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPMAXUB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPMAXUW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPMAXUD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPMINSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPMINSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPMINSD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPMINUB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPMINUW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPMINUD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPMOVMSKB reg64,xmmreg AVX,SANDYBRIDGE,LONG -VPMOVMSKB reg32,xmmreg AVX,SANDYBRIDGE -VPMOVSXBW xmmreg,xmmrm64 AVX,SANDYBRIDGE -VPMOVSXBD xmmreg,xmmrm32 AVX,SANDYBRIDGE -VPMOVSXBQ xmmreg,xmmrm16 AVX,SANDYBRIDGE -VPMOVSXWD xmmreg,xmmrm64 AVX,SANDYBRIDGE -VPMOVSXWQ xmmreg,xmmrm32 AVX,SANDYBRIDGE -VPMOVSXDQ xmmreg,xmmrm64 AVX,SANDYBRIDGE -VPMOVZXBW xmmreg,xmmrm64 AVX,SANDYBRIDGE -VPMOVZXBD xmmreg,xmmrm32 AVX,SANDYBRIDGE -VPMOVZXBQ xmmreg,xmmrm16 AVX,SANDYBRIDGE -VPMOVZXWD xmmreg,xmmrm64 AVX,SANDYBRIDGE -VPMOVZXWQ xmmreg,xmmrm32 AVX,SANDYBRIDGE -VPMOVZXDQ xmmreg,xmmrm64 AVX,SANDYBRIDGE -VPMULHUW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPMULHRSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPMULHW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPMULLW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPMULLD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPMULUDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPMULDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPOR xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSADBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSHUFB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSHUFD xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -VPSHUFHW xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -VPSHUFLW xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -VPSIGNB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSIGNW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSIGND xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSLLDQ xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE -VPSRLDQ xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE -VPSLLW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSLLW xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE -VPSLLD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSLLD xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE -VPSLLQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSLLQ xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE -VPSRAW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSRAW xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE -VPSRAD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSRAD xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE -VPSRLW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSRLW xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE -VPSRLD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSRLD xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE -VPSRLQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSRLQ xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE -VPTEST xmmreg,xmmrm128 AVX,SANDYBRIDGE -VPTEST ymmreg,ymmrm256 AVX,SANDYBRIDGE -VPSUBB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSUBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSUBD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSUBQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSUBSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSUBSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSUBUSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPSUBUSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPUNPCKHBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPUNPCKHWD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPUNPCKHDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPUNPCKHQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPUNPCKLBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPUNPCKLWD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPUNPCKLDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPUNPCKLQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPXOR xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VRCPPS xmmreg,xmmrm128 AVX,SANDYBRIDGE -VRCPPS ymmreg,ymmrm256 AVX,SANDYBRIDGE -VRCPSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VRSQRTPS xmmreg,xmmrm128 AVX,SANDYBRIDGE -VRSQRTPS ymmreg,ymmrm256 AVX,SANDYBRIDGE -VRSQRTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VROUNDPD xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -VROUNDPD ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE -VROUNDPS xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -VROUNDPS ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE -VROUNDSD xmmreg,xmmreg*,xmmrm64,imm8 AVX,SANDYBRIDGE -VROUNDSS xmmreg,xmmreg*,xmmrm32,imm8 AVX,SANDYBRIDGE -VSHUFPD xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -VSHUFPD ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE -VSHUFPS xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -VSHUFPS ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE -VSQRTPD xmmreg,xmmrm128 AVX,SANDYBRIDGE -VSQRTPD ymmreg,ymmrm256 AVX,SANDYBRIDGE -VSQRTPS xmmreg,xmmrm128 AVX,SANDYBRIDGE -VSQRTPS ymmreg,ymmrm256 AVX,SANDYBRIDGE -VSQRTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VSQRTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VSTMXCSR mem32 AVX,SANDYBRIDGE -VSUBPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VSUBPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VSUBPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VSUBPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VSUBSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -VSUBSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -VTESTPS xmmreg,xmmrm128 AVX,SANDYBRIDGE -VTESTPS ymmreg,ymmrm256 AVX,SANDYBRIDGE -VTESTPD xmmreg,xmmrm128 AVX,SANDYBRIDGE -VTESTPD ymmreg,ymmrm256 AVX,SANDYBRIDGE -VUCOMISD xmmreg,xmmrm64 AVX,SANDYBRIDGE -VUCOMISS xmmreg,xmmrm32 AVX,SANDYBRIDGE -VUNPCKHPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VUNPCKHPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VUNPCKHPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VUNPCKHPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VUNPCKLPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VUNPCKLPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VUNPCKLPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VUNPCKLPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VXORPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VXORPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VXORPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VXORPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -VZEROALL AVX,SANDYBRIDGE -VZEROUPPER AVX,SANDYBRIDGE -</pre> -<h4><a name="section-B.1.26">B.1.26 Intel Carry-Less Multiplication instructions (CLMUL)</a></h4> -<p><pre> -PCLMULLQLQDQ xmmreg,xmmrm128 SSE,WESTMERE -PCLMULHQLQDQ xmmreg,xmmrm128 SSE,WESTMERE -PCLMULLQHQDQ xmmreg,xmmrm128 SSE,WESTMERE -PCLMULHQHQDQ xmmreg,xmmrm128 SSE,WESTMERE -PCLMULQDQ xmmreg,xmmrm128,imm8 SSE,WESTMERE -</pre> -<h4><a name="section-B.1.27">B.1.27 Intel AVX Carry-Less Multiplication instructions (CLMUL)</a></h4> -<p><pre> -VPCLMULLQLQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPCLMULHQLQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPCLMULLQHQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPCLMULHQHQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -VPCLMULQDQ xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -</pre> -<h4><a name="section-B.1.28">B.1.28 Intel Fused Multiply-Add instructions (FMA)</a></h4> -<p><pre> -VFMADD132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADD132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADD132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADD132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADD312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADD312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADD312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADD312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADD213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADD213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADD213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADD213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADD123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADD123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADD123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADD123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADD231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADD231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADD231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADD231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADD321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADD321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADD321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADD321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADDSUB132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADDSUB132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADDSUB132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADDSUB132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADDSUB312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADDSUB312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADDSUB312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADDSUB312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADDSUB213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADDSUB213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADDSUB213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADDSUB213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADDSUB123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADDSUB123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADDSUB123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADDSUB123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADDSUB231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADDSUB231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADDSUB231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADDSUB231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADDSUB321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADDSUB321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADDSUB321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMADDSUB321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUB132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUB132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUB132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUB132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUB312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUB312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUB312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUB312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUB213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUB213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUB213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUB213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUB123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUB123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUB123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUB123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUB231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUB231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUB231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUB231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUB321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUB321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUB321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUB321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUBADD132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUBADD132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUBADD132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUBADD132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUBADD312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUBADD312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUBADD312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUBADD312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUBADD213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUBADD213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUBADD213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUBADD213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUBADD123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUBADD123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUBADD123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUBADD123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUBADD231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUBADD231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUBADD231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUBADD231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUBADD321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUBADD321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMSUBADD321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFMSUBADD321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMADD132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMADD132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMADD132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMADD132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMADD312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMADD312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMADD312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMADD312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMADD213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMADD213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMADD213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMADD213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMADD123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMADD123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMADD123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMADD123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMADD231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMADD231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMADD231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMADD231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMADD321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMADD321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMADD321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMADD321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMSUB132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMSUB132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMSUB132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMSUB132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMSUB312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMSUB312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMSUB312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMSUB312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMSUB213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMSUB213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMSUB213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMSUB213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMSUB123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMSUB123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMSUB123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMSUB123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMSUB231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMSUB231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMSUB231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMSUB231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMSUB321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMSUB321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFNMSUB321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -VFNMSUB321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -VFMADD132SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFMADD132SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFMADD312SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFMADD312SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFMADD213SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFMADD213SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFMADD123SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFMADD123SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFMADD231SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFMADD231SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFMADD321SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFMADD321SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFMSUB132SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFMSUB132SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFMSUB312SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFMSUB312SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFMSUB213SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFMSUB213SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFMSUB123SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFMSUB123SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFMSUB231SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFMSUB231SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFMSUB321SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFMSUB321SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFNMADD132SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFNMADD132SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFNMADD312SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFNMADD312SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFNMADD213SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFNMADD213SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFNMADD123SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFNMADD123SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFNMADD231SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFNMADD231SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFNMADD321SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFNMADD321SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFNMSUB132SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFNMSUB132SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFNMSUB312SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFNMSUB312SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFNMSUB213SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFNMSUB213SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFNMSUB123SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFNMSUB123SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFNMSUB231SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFNMSUB231SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -VFNMSUB321SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -VFNMSUB321SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -</pre> -<h4><a name="section-B.1.29">B.1.29 VIA (Centaur) security instructions</a></h4> -<p><pre> -XSTORE PENT,CYRIX -XCRYPTECB PENT,CYRIX -XCRYPTCBC PENT,CYRIX -XCRYPTCTR PENT,CYRIX -XCRYPTCFB PENT,CYRIX -XCRYPTOFB PENT,CYRIX -MONTMUL PENT,CYRIX -XSHA1 PENT,CYRIX -XSHA256 PENT,CYRIX -</pre> -<h4><a name="section-B.1.30">B.1.30 AMD Lightweight Profiling (LWP) instructions</a></h4> -<p><pre> -LLWPCB reg16 AMD -LLWPCB reg32 AMD,386 -LLWPCB reg64 AMD,X64 -SLWPCB reg16 AMD -SLWPCB reg32 AMD,386 -SLWPCB reg64 AMD,X64 -LWPVAL reg16,rm32,imm16 AMD,386 -LWPVAL reg32,rm32,imm32 AMD,386 -LWPVAL reg64,rm32,imm32 AMD,X64 -LWPINS reg16,rm32,imm16 AMD,386 -LWPINS reg32,rm32,imm32 AMD,386 -LWPINS reg64,rm32,imm32 AMD,X64 -</pre> -<h4><a name="section-B.1.31">B.1.31 AMD XOP, FMA4 and CVT16 instructions (SSE5)</a></h4> -<p><pre> -VCVTPH2PS xmmreg,xmmrm64*,imm8 AMD,SSE5 -VCVTPH2PS ymmreg,xmmrm128,imm8 AMD,SSE5 -VCVTPH2PS ymmreg,ymmrm128*,imm8 AMD,SSE5 -VCVTPS2PH xmmrm64,xmmreg*,imm8 AMD,SSE5 -VCVTPS2PH xmmrm128,ymmreg,imm8 AMD,SSE5 -VCVTPS2PH ymmrm128,ymmreg*,imm8 AMD,SSE5 -VFMADDPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VFMADDPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -VFMADDPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -VFMADDPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -VFMADDPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VFMADDPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -VFMADDPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -VFMADDPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -VFMADDSD xmmreg,xmmreg*,xmmrm64,xmmreg AMD,SSE5 -VFMADDSD xmmreg,xmmreg*,xmmreg,xmmrm64 AMD,SSE5 -VFMADDSS xmmreg,xmmreg*,xmmrm32,xmmreg AMD,SSE5 -VFMADDSS xmmreg,xmmreg*,xmmreg,xmmrm32 AMD,SSE5 -VFMADDSUBPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VFMADDSUBPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -VFMADDSUBPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -VFMADDSUBPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -VFMADDSUBPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VFMADDSUBPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -VFMADDSUBPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -VFMADDSUBPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -VFMSUBADDPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VFMSUBADDPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -VFMSUBADDPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -VFMSUBADDPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -VFMSUBADDPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VFMSUBADDPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -VFMSUBADDPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -VFMSUBADDPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -VFMSUBPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VFMSUBPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -VFMSUBPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -VFMSUBPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -VFMSUBPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VFMSUBPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -VFMSUBPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -VFMSUBPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -VFMSUBSD xmmreg,xmmreg*,xmmrm64,xmmreg AMD,SSE5 -VFMSUBSD xmmreg,xmmreg*,xmmreg,xmmrm64 AMD,SSE5 -VFMSUBSS xmmreg,xmmreg*,xmmrm32,xmmreg AMD,SSE5 -VFMSUBSS xmmreg,xmmreg*,xmmreg,xmmrm32 AMD,SSE5 -VFNMADDPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VFNMADDPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -VFNMADDPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -VFNMADDPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -VFNMADDPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VFNMADDPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -VFNMADDPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -VFNMADDPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -VFNMADDSD xmmreg,xmmreg*,xmmrm64,xmmreg AMD,SSE5 -VFNMADDSD xmmreg,xmmreg*,xmmreg,xmmrm64 AMD,SSE5 -VFNMADDSS xmmreg,xmmreg*,xmmrm32,xmmreg AMD,SSE5 -VFNMADDSS xmmreg,xmmreg*,xmmreg,xmmrm32 AMD,SSE5 -VFNMSUBPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VFNMSUBPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -VFNMSUBPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -VFNMSUBPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -VFNMSUBPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VFNMSUBPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -VFNMSUBPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -VFNMSUBPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -VFNMSUBSD xmmreg,xmmreg*,xmmrm64,xmmreg AMD,SSE5 -VFNMSUBSD xmmreg,xmmreg*,xmmreg,xmmrm64 AMD,SSE5 -VFNMSUBSS xmmreg,xmmreg*,xmmrm32,xmmreg AMD,SSE5 -VFNMSUBSS xmmreg,xmmreg*,xmmreg,xmmrm32 AMD,SSE5 -VFRCZPD xmmreg,xmmrm128* AMD,SSE5 -VFRCZPD ymmreg,ymmrm256* AMD,SSE5 -VFRCZPS xmmreg,xmmrm128* AMD,SSE5 -VFRCZPS ymmreg,ymmrm256* AMD,SSE5 -VFRCZSD xmmreg,xmmrm64* AMD,SSE5 -VFRCZSS xmmreg,xmmrm32* AMD,SSE5 -VPCMOV xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VPCMOV ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -VPCMOV xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -VPCMOV ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -VPCOMB xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 -VPCOMD xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 -VPCOMQ xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 -VPCOMUB xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 -VPCOMUD xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 -VPCOMUQ xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 -VPCOMUW xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 -VPCOMW xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 -VPHADDBD xmmreg,xmmrm128* AMD,SSE5 -VPHADDBQ xmmreg,xmmrm128* AMD,SSE5 -VPHADDBW xmmreg,xmmrm128* AMD,SSE5 -VPHADDDQ xmmreg,xmmrm128* AMD,SSE5 -VPHADDUBD xmmreg,xmmrm128* AMD,SSE5 -VPHADDUBQ xmmreg,xmmrm128* AMD,SSE5 -VPHADDUBW xmmreg,xmmrm128* AMD,SSE5 -VPHADDUDQ xmmreg,xmmrm128* AMD,SSE5 -VPHADDUWD xmmreg,xmmrm128* AMD,SSE5 -VPHADDUWQ xmmreg,xmmrm128* AMD,SSE5 -VPHADDWD xmmreg,xmmrm128* AMD,SSE5 -VPHADDWQ xmmreg,xmmrm128* AMD,SSE5 -VPHSUBBW xmmreg,xmmrm128* AMD,SSE5 -VPHSUBDQ xmmreg,xmmrm128* AMD,SSE5 -VPHSUBWD xmmreg,xmmrm128* AMD,SSE5 -VPMACSDD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VPMACSDQH xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VPMACSDQL xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VPMACSSDD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VPMACSSDQH xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VPMACSSDQL xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VPMACSSWD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VPMACSSWW xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VPMACSWD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VPMACSWW xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VPMADCSSWD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VPMADCSWD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VPPERM xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -VPPERM xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -VPROTB xmmreg,xmmrm128*,xmmreg AMD,SSE5 -VPROTB xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -VPROTB xmmreg,xmmrm128*,imm8 AMD,SSE5 -VPROTD xmmreg,xmmrm128*,xmmreg AMD,SSE5 -VPROTD xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -VPROTD xmmreg,xmmrm128*,imm8 AMD,SSE5 -VPROTQ xmmreg,xmmrm128*,xmmreg AMD,SSE5 -VPROTQ xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -VPROTQ xmmreg,xmmrm128*,imm8 AMD,SSE5 -VPROTW xmmreg,xmmrm128*,xmmreg AMD,SSE5 -VPROTW xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -VPROTW xmmreg,xmmrm128*,imm8 AMD,SSE5 -VPSHAB xmmreg,xmmrm128*,xmmreg AMD,SSE5 -VPSHAB xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -VPSHAD xmmreg,xmmrm128*,xmmreg AMD,SSE5 -VPSHAD xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -VPSHAQ xmmreg,xmmrm128*,xmmreg AMD,SSE5 -VPSHAQ xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -VPSHAW xmmreg,xmmrm128*,xmmreg AMD,SSE5 -VPSHAW xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -VPSHLB xmmreg,xmmrm128*,xmmreg AMD,SSE5 -VPSHLB xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -VPSHLD xmmreg,xmmrm128*,xmmreg AMD,SSE5 -VPSHLD xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -VPSHLQ xmmreg,xmmrm128*,xmmreg AMD,SSE5 -VPSHLQ xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -VPSHLW xmmreg,xmmrm128*,xmmreg AMD,SSE5 -VPSHLW xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -</pre> -<h4><a name="section-B.1.32">B.1.32 Systematic names for the hinting nop instructions</a></h4> -<p><pre> -HINT_NOP0 rm16 P6,UNDOC -HINT_NOP0 rm32 P6,UNDOC -HINT_NOP0 rm64 X64,UNDOC -HINT_NOP1 rm16 P6,UNDOC -HINT_NOP1 rm32 P6,UNDOC -HINT_NOP1 rm64 X64,UNDOC -HINT_NOP2 rm16 P6,UNDOC -HINT_NOP2 rm32 P6,UNDOC -HINT_NOP2 rm64 X64,UNDOC -HINT_NOP3 rm16 P6,UNDOC -HINT_NOP3 rm32 P6,UNDOC -HINT_NOP3 rm64 X64,UNDOC -HINT_NOP4 rm16 P6,UNDOC -HINT_NOP4 rm32 P6,UNDOC -HINT_NOP4 rm64 X64,UNDOC -HINT_NOP5 rm16 P6,UNDOC -HINT_NOP5 rm32 P6,UNDOC -HINT_NOP5 rm64 X64,UNDOC -HINT_NOP6 rm16 P6,UNDOC -HINT_NOP6 rm32 P6,UNDOC -HINT_NOP6 rm64 X64,UNDOC -HINT_NOP7 rm16 P6,UNDOC -HINT_NOP7 rm32 P6,UNDOC -HINT_NOP7 rm64 X64,UNDOC -HINT_NOP8 rm16 P6,UNDOC -HINT_NOP8 rm32 P6,UNDOC -HINT_NOP8 rm64 X64,UNDOC -HINT_NOP9 rm16 P6,UNDOC -HINT_NOP9 rm32 P6,UNDOC -HINT_NOP9 rm64 X64,UNDOC -HINT_NOP10 rm16 P6,UNDOC -HINT_NOP10 rm32 P6,UNDOC -HINT_NOP10 rm64 X64,UNDOC -HINT_NOP11 rm16 P6,UNDOC -HINT_NOP11 rm32 P6,UNDOC -HINT_NOP11 rm64 X64,UNDOC -HINT_NOP12 rm16 P6,UNDOC -HINT_NOP12 rm32 P6,UNDOC -HINT_NOP12 rm64 X64,UNDOC -HINT_NOP13 rm16 P6,UNDOC -HINT_NOP13 rm32 P6,UNDOC -HINT_NOP13 rm64 X64,UNDOC -HINT_NOP14 rm16 P6,UNDOC -HINT_NOP14 rm32 P6,UNDOC -HINT_NOP14 rm64 X64,UNDOC -HINT_NOP15 rm16 P6,UNDOC -HINT_NOP15 rm32 P6,UNDOC -HINT_NOP15 rm64 X64,UNDOC -HINT_NOP16 rm16 P6,UNDOC -HINT_NOP16 rm32 P6,UNDOC -HINT_NOP16 rm64 X64,UNDOC -HINT_NOP17 rm16 P6,UNDOC -HINT_NOP17 rm32 P6,UNDOC -HINT_NOP17 rm64 X64,UNDOC -HINT_NOP18 rm16 P6,UNDOC -HINT_NOP18 rm32 P6,UNDOC -HINT_NOP18 rm64 X64,UNDOC -HINT_NOP19 rm16 P6,UNDOC -HINT_NOP19 rm32 P6,UNDOC -HINT_NOP19 rm64 X64,UNDOC -HINT_NOP20 rm16 P6,UNDOC -HINT_NOP20 rm32 P6,UNDOC -HINT_NOP20 rm64 X64,UNDOC -HINT_NOP21 rm16 P6,UNDOC -HINT_NOP21 rm32 P6,UNDOC -HINT_NOP21 rm64 X64,UNDOC -HINT_NOP22 rm16 P6,UNDOC -HINT_NOP22 rm32 P6,UNDOC -HINT_NOP22 rm64 X64,UNDOC -HINT_NOP23 rm16 P6,UNDOC -HINT_NOP23 rm32 P6,UNDOC -HINT_NOP23 rm64 X64,UNDOC -HINT_NOP24 rm16 P6,UNDOC -HINT_NOP24 rm32 P6,UNDOC -HINT_NOP24 rm64 X64,UNDOC -HINT_NOP25 rm16 P6,UNDOC -HINT_NOP25 rm32 P6,UNDOC -HINT_NOP25 rm64 X64,UNDOC -HINT_NOP26 rm16 P6,UNDOC -HINT_NOP26 rm32 P6,UNDOC -HINT_NOP26 rm64 X64,UNDOC -HINT_NOP27 rm16 P6,UNDOC -HINT_NOP27 rm32 P6,UNDOC -HINT_NOP27 rm64 X64,UNDOC -HINT_NOP28 rm16 P6,UNDOC -HINT_NOP28 rm32 P6,UNDOC -HINT_NOP28 rm64 X64,UNDOC -HINT_NOP29 rm16 P6,UNDOC -HINT_NOP29 rm32 P6,UNDOC -HINT_NOP29 rm64 X64,UNDOC -HINT_NOP30 rm16 P6,UNDOC -HINT_NOP30 rm32 P6,UNDOC -HINT_NOP30 rm64 X64,UNDOC -HINT_NOP31 rm16 P6,UNDOC -HINT_NOP31 rm32 P6,UNDOC -HINT_NOP31 rm64 X64,UNDOC -HINT_NOP32 rm16 P6,UNDOC -HINT_NOP32 rm32 P6,UNDOC -HINT_NOP32 rm64 X64,UNDOC -HINT_NOP33 rm16 P6,UNDOC -HINT_NOP33 rm32 P6,UNDOC -HINT_NOP33 rm64 X64,UNDOC -HINT_NOP34 rm16 P6,UNDOC -HINT_NOP34 rm32 P6,UNDOC -HINT_NOP34 rm64 X64,UNDOC -HINT_NOP35 rm16 P6,UNDOC -HINT_NOP35 rm32 P6,UNDOC -HINT_NOP35 rm64 X64,UNDOC -HINT_NOP36 rm16 P6,UNDOC -HINT_NOP36 rm32 P6,UNDOC -HINT_NOP36 rm64 X64,UNDOC -HINT_NOP37 rm16 P6,UNDOC -HINT_NOP37 rm32 P6,UNDOC -HINT_NOP37 rm64 X64,UNDOC -HINT_NOP38 rm16 P6,UNDOC -HINT_NOP38 rm32 P6,UNDOC -HINT_NOP38 rm64 X64,UNDOC -HINT_NOP39 rm16 P6,UNDOC -HINT_NOP39 rm32 P6,UNDOC -HINT_NOP39 rm64 X64,UNDOC -HINT_NOP40 rm16 P6,UNDOC -HINT_NOP40 rm32 P6,UNDOC -HINT_NOP40 rm64 X64,UNDOC -HINT_NOP41 rm16 P6,UNDOC -HINT_NOP41 rm32 P6,UNDOC -HINT_NOP41 rm64 X64,UNDOC -HINT_NOP42 rm16 P6,UNDOC -HINT_NOP42 rm32 P6,UNDOC -HINT_NOP42 rm64 X64,UNDOC -HINT_NOP43 rm16 P6,UNDOC -HINT_NOP43 rm32 P6,UNDOC -HINT_NOP43 rm64 X64,UNDOC -HINT_NOP44 rm16 P6,UNDOC -HINT_NOP44 rm32 P6,UNDOC -HINT_NOP44 rm64 X64,UNDOC -HINT_NOP45 rm16 P6,UNDOC -HINT_NOP45 rm32 P6,UNDOC -HINT_NOP45 rm64 X64,UNDOC -HINT_NOP46 rm16 P6,UNDOC -HINT_NOP46 rm32 P6,UNDOC -HINT_NOP46 rm64 X64,UNDOC -HINT_NOP47 rm16 P6,UNDOC -HINT_NOP47 rm32 P6,UNDOC -HINT_NOP47 rm64 X64,UNDOC -HINT_NOP48 rm16 P6,UNDOC -HINT_NOP48 rm32 P6,UNDOC -HINT_NOP48 rm64 X64,UNDOC -HINT_NOP49 rm16 P6,UNDOC -HINT_NOP49 rm32 P6,UNDOC -HINT_NOP49 rm64 X64,UNDOC -HINT_NOP50 rm16 P6,UNDOC -HINT_NOP50 rm32 P6,UNDOC -HINT_NOP50 rm64 X64,UNDOC -HINT_NOP51 rm16 P6,UNDOC -HINT_NOP51 rm32 P6,UNDOC -HINT_NOP51 rm64 X64,UNDOC -HINT_NOP52 rm16 P6,UNDOC -HINT_NOP52 rm32 P6,UNDOC -HINT_NOP52 rm64 X64,UNDOC -HINT_NOP53 rm16 P6,UNDOC -HINT_NOP53 rm32 P6,UNDOC -HINT_NOP53 rm64 X64,UNDOC -HINT_NOP54 rm16 P6,UNDOC -HINT_NOP54 rm32 P6,UNDOC -HINT_NOP54 rm64 X64,UNDOC -HINT_NOP55 rm16 P6,UNDOC -HINT_NOP55 rm32 P6,UNDOC -HINT_NOP55 rm64 X64,UNDOC -HINT_NOP56 rm16 P6,UNDOC -HINT_NOP56 rm32 P6,UNDOC -HINT_NOP56 rm64 X64,UNDOC -HINT_NOP57 rm16 P6,UNDOC -HINT_NOP57 rm32 P6,UNDOC -HINT_NOP57 rm64 X64,UNDOC -HINT_NOP58 rm16 P6,UNDOC -HINT_NOP58 rm32 P6,UNDOC -HINT_NOP58 rm64 X64,UNDOC -HINT_NOP59 rm16 P6,UNDOC -HINT_NOP59 rm32 P6,UNDOC -HINT_NOP59 rm64 X64,UNDOC -HINT_NOP60 rm16 P6,UNDOC -HINT_NOP60 rm32 P6,UNDOC -HINT_NOP60 rm64 X64,UNDOC -HINT_NOP61 rm16 P6,UNDOC -HINT_NOP61 rm32 P6,UNDOC -HINT_NOP61 rm64 X64,UNDOC -HINT_NOP62 rm16 P6,UNDOC -HINT_NOP62 rm32 P6,UNDOC -HINT_NOP62 rm64 X64,UNDOC -HINT_NOP63 rm16 P6,UNDOC -HINT_NOP63 rm32 P6,UNDOC -HINT_NOP63 rm64 X64,UNDOC -</pre> -<p align=center><a href="nasmdocc.html">Next Chapter</a> | -<a href="nasmdoca.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -</body></html> diff --git a/doc/html/nasmdocc.html b/doc/html/nasmdocc.html deleted file mode 100644 index b7ec3b2..0000000 --- a/doc/html/nasmdocc.html +++ /dev/null @@ -1,1468 +0,0 @@ -<html><head><title>NASM Manual</title></head> -<body><h1 align=center>The Netwide Assembler: NASM</h1> - -<p align=center><a href="nasmdocb.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -<h2><a name="appendix-C">Appendix C: NASM Version History</a></h2> -<h3><a name="section-C.1">C.1 NASM 2 Series</a></h3> -<p>The NASM 2 series support x86-64, and is the production version of NASM -since 2007. -<h4><a name="section-C.1.1">C.1.1 Version 2.08</a></h4> -<ul> -<li>A number of enhancements/fixes in macros area. -<li>Support for arbitrarily terminating macro expansions -<code><nobr>%exitmacro</nobr></code>. See -<a href="nasmdoc4.html#section-4.3.12">section 4.3.12</a>. -<li>Support for recursive macro expansion -<code><nobr>%rmacro/irmacro</nobr></code>. See -<a href="nasmdoc4.html#section-4.3.1">section 4.3.1</a>. -<li>Support for converting strings to tokens. See -<a href="nasmdoc4.html#section-4.1.9">section 4.1.9</a>. -<li>Fuzzy operand size logic introduced. -<li>Fix COFF stack overrun on too long export identifiers. -<li>Fix Macho-O alignment bug. -<li>Fix crashes with -fwin32 on file with many exports. -<li>Fix stack overrun for too long [DEBUG id]. -<li>Fix incorrect sbyte usage in IMUL (hit only if optimization flag -passed). -<li>Append ending token for <code><nobr>.stabs</nobr></code> records in the -ELF output format. -<li>New NSIS script which uses ModernUI and MultiUser approach. -<li>Visual Studio 2008 NASM integration (rules file). -<li>Warn a user if a constant is too long (and as result will be stripped). -<li>The obsoleted pre-XOP AMD SSE5 instruction set which was never -actualized was removed. -<li>Fix stack overrun on too long error file name passed from the command -line. -<li>Bind symbols to the .text section by default (ie in case if SECTION -directive was omitted) in the ELF output format. -<li>Fix sync points array index wrapping. -<li>A few fixes for FMA4 and XOP instruction templates. -<li>Add AMD Lightweight Profiling (LWP) instructions. -</ul> -<h4><a name="section-C.1.2">C.1.2 Version 2.07</a></h4> -<ul> -<li>NASM is now under the 2-clause BSD license. See -<a href="nasmdoc1.html#section-1.1.2">section 1.1.2</a>. -<li>Fix the section type for the <code><nobr>.strtab</nobr></code> section -in the <code><nobr>elf64</nobr></code> output format. -<li>Fix the handling of <code><nobr>COMMON</nobr></code> directives in the -<code><nobr>obj</nobr></code> output format. -<li>New <code><nobr>ith</nobr></code> and <code><nobr>srec</nobr></code> -output formats; these are variants of the <code><nobr>bin</nobr></code> -output format which output Intel hex and Motorola S-records, respectively. -See <a href="nasmdoc7.html#section-7.2">section 7.2</a> and -<a href="nasmdoc7.html#section-7.3">section 7.3</a>. -<li><code><nobr>rdf2ihx</nobr></code> replaced with an enhanced -<code><nobr>rdf2bin</nobr></code>, which can output binary, COM, Intel hex -or Motorola S-records. -<li>The Windows installer now puts the NASM directory first in the -<code><nobr>PATH</nobr></code> of the "NASM Shell". -<li>Revert the early expansion behavior of <code><nobr>%+</nobr></code> to -pre-2.06 behavior: <code><nobr>%+</nobr></code> is only expanded late. -<li>Yet another Mach-O alignment fix. -<li>Don't delete the list file on errors. Also, include error and warning -information in the list file. -<li>Support for 64-bit Mach-O output, see -<a href="nasmdoc7.html#section-7.8">section 7.8</a>. -<li>Fix assert failure on certain operations that involve strings with -high-bit bytes. -</ul> -<h4><a name="section-C.1.3">C.1.3 Version 2.06</a></h4> -<ul> -<li>This release is dedicated to the memory of Charles A. Crayne, long time -NASM developer as well as moderator of -<code><nobr>comp.lang.asm.x86</nobr></code> and author of the book -<em>Serious Assembler</em>. We miss you, Chuck. -<li>Support for indirect macro expansion -(<code><nobr>%[...]</nobr></code>). See -<a href="nasmdoc4.html#section-4.1.3">section 4.1.3</a>. -<li><code><nobr>%pop</nobr></code> can now take an argument, see -<a href="nasmdoc4.html#section-4.7.1">section 4.7.1</a>. -<li>The argument to <code><nobr>%use</nobr></code> is no longer -macro-expanded. Use <code><nobr>%[...]</nobr></code> if macro expansion is -desired. -<li>Support for thread-local storage in ELF32 and ELF64. See -<a href="nasmdoc7.html#section-7.9.4">section 7.9.4</a>. -<li>Fix crash on <code><nobr>%ifmacro</nobr></code> without an argument. -<li>Correct the arguments to the <code><nobr>POPCNT</nobr></code> -instruction. -<li>Fix section alignment in the Mach-O format. -<li>Update AVX support to version 5 of the Intel specification. -<li>Fix the handling of accesses to context-local macros from higher levels -in the context stack. -<li>Treat <code><nobr>WAIT</nobr></code> as a prefix rather than as an -instruction, thereby allowing constructs like -<code><nobr>O16 FSAVE</nobr></code> to work correctly. -<li>Support for structures with a non-zero base offset. See -<a href="nasmdoc4.html#section-4.11.10">section 4.11.10</a>. -<li>Correctly handle preprocessor token concatenation (see -<a href="nasmdoc4.html#section-4.3.8">section 4.3.8</a>) involving -floating-point numbers. -<li>The <code><nobr>PINSR</nobr></code> series of instructions have been -corrected and rationalized. -<li>Removed AMD SSE5, replaced with the new XOP/FMA4/CVT16 (rev 3.03) spec. -<li>The ELF backends no longer automatically generate a -<code><nobr>.comment</nobr></code> section. -<li>Add additional "well-known" ELF sections with default attributes. See -<a href="nasmdoc7.html#section-7.9.2">section 7.9.2</a>. -</ul> -<h4><a name="section-C.1.4">C.1.4 Version 2.05.01</a></h4> -<ul> -<li>Fix the <code><nobr>-w</nobr></code>/<code><nobr>-W</nobr></code> -option parsing, which was broken in NASM 2.05. -</ul> -<h4><a name="section-C.1.5">C.1.5 Version 2.05</a></h4> -<ul> -<li>Fix redundant REX.W prefix on <code><nobr>JMP reg64</nobr></code>. -<li>Make the behaviour of <code><nobr>-O0</nobr></code> match NASM 0.98 -legacy behavior. See <a href="nasmdoc2.html#section-2.1.22">section -2.1.22</a>. -<li><code><nobr>-w-user</nobr></code> can be used to suppress the output of -<code><nobr>%warning</nobr></code> directives. See -<a href="nasmdoc2.html#section-2.1.24">section 2.1.24</a>. -<li>Fix bug where <code><nobr>ALIGN</nobr></code> would issue a full -alignment datum instead of zero bytes. -<li>Fix offsets in list files. -<li>Fix <code><nobr>%include</nobr></code> inside multi-line macros or -loops. -<li>Fix error where NASM would generate a spurious warning on valid -optimizations of immediate values. -<li>Fix arguments to a number of the <code><nobr>CVT</nobr></code> SSE -instructions. -<li>Fix RIP-relative offsets when the instruction carries an immediate. -<li>Massive overhaul of the ELF64 backend for spec compliance. -<li>Fix the Geode <code><nobr>PFRCPV</nobr></code> and -<code><nobr>PFRSQRTV</nobr></code> instruction. -<li>Fix the SSE 4.2 <code><nobr>CRC32</nobr></code> instruction. -</ul> -<h4><a name="section-C.1.6">C.1.6 Version 2.04</a></h4> -<ul> -<li>Sanitize macro handing in the <code><nobr>%error</nobr></code> -directive. -<li>New <code><nobr>%warning</nobr></code> directive to issue -user-controlled warnings. -<li><code><nobr>%error</nobr></code> directives are now deferred to the -final assembly phase. -<li>New <code><nobr>%fatal</nobr></code> directive to immediately terminate -assembly. -<li>New <code><nobr>%strcat</nobr></code> directive to join quoted strings -together. -<li>New <code><nobr>%use</nobr></code> macro directive to support standard -macro directives. See <a href="nasmdoc4.html#section-4.6.4">section -4.6.4</a>. -<li>Excess default parameters to <code><nobr>%macro</nobr></code> now -issues a warning by default. See -<a href="nasmdoc4.html#section-4.3">section 4.3</a>. -<li>Fix <code><nobr>%ifn</nobr></code> and -<code><nobr>%elifn</nobr></code>. -<li>Fix nested <code><nobr>%else</nobr></code> clauses. -<li>Correct the handling of nested <code><nobr>%rep</nobr></code>s. -<li>New <code><nobr>%unmacro</nobr></code> directive to undeclare a -multi-line macro. See <a href="nasmdoc4.html#section-4.3.11">section -4.3.11</a>. -<li>Builtin macro <code><nobr>__PASS__</nobr></code> which expands to the -current assembly pass. See <a href="nasmdoc4.html#section-4.11.9">section -4.11.9</a>. -<li><code><nobr>__utf16__</nobr></code> and -<code><nobr>__utf32__</nobr></code> operators to generate UTF-16 and UTF-32 -strings. See <a href="nasmdoc3.html#section-3.4.5">section 3.4.5</a>. -<li>Fix bug in case-insensitive matching when compiled on platforms that -don't use the <code><nobr>configure</nobr></code> script. Of the official -release binaries, that only affected the OS/2 binary. -<li>Support for x87 packed BCD constants. See -<a href="nasmdoc3.html#section-3.4.7">section 3.4.7</a>. -<li>Correct the <code><nobr>LTR</nobr></code> and -<code><nobr>SLDT</nobr></code> instructions in 64-bit mode. -<li>Fix unnecessary REX.W prefix on indirect jumps in 64-bit mode. -<li>Add AVX versions of the AES instructions -(<code><nobr>VAES</nobr></code>...). -<li>Fix the 256-bit FMA instructions. -<li>Add 256-bit AVX stores per the latest AVX spec. -<li>VIA XCRYPT instructions can now be written either with or without -<code><nobr>REP</nobr></code>, apparently different versions of the VIA -spec wrote them differently. -<li>Add missing 64-bit <code><nobr>MOVNTI</nobr></code> instruction. -<li>Fix the operand size of <code><nobr>VMREAD</nobr></code> and -<code><nobr>VMWRITE</nobr></code>. -<li>Numerous bug fixes, especially to the AES, AVX and VTX instructions. -<li>The optimizer now always runs until it converges. It also runs even -when disabled, but doesn't optimize. This allows most forward references to -be resolved properly. -<li><code><nobr>%push</nobr></code> no longer needs a context identifier; -omitting the context identifier results in an anonymous context. -</ul> -<h4><a name="section-C.1.7">C.1.7 Version 2.03.01</a></h4> -<ul> -<li>Fix buffer overflow in the listing module. -<li>Fix the handling of hexadecimal escape codes in `...` strings. -<li>The Postscript/PDF documentation has been reformatted. -<li>The <code><nobr>-F</nobr></code> option now implies -<code><nobr>-g</nobr></code>. -</ul> -<h4><a name="section-C.1.8">C.1.8 Version 2.03</a></h4> -<ul> -<li>Add support for Intel AVX, CLMUL and FMA instructions, including YMM -registers. -<li><code><nobr>dy</nobr></code>, <code><nobr>resy</nobr></code> and -<code><nobr>yword</nobr></code> for 32-byte operands. -<li>Fix some SSE5 instructions. -<li>Intel <code><nobr>INVEPT</nobr></code>, -<code><nobr>INVVPID</nobr></code> and <code><nobr>MOVBE</nobr></code> -instructions. -<li>Fix checking for critical expressions when the optimizer is enabled. -<li>Support the DWARF debugging format for ELF targets. -<li>Fix optimizations of signed bytes. -<li>Fix operation on bigendian machines. -<li>Fix buffer overflow in the preprocessor. -<li><code><nobr>SAFESEH</nobr></code> support for Win32, -<code><nobr>IMAGEREL</nobr></code> for Win64 (SEH). -<li><code><nobr>%?</nobr></code> and <code><nobr>%??</nobr></code> to refer -to the name of a macro itself. In particular, -<code><nobr>%idefine keyword $%?</nobr></code> can be used to make a -keyword "disappear". -<li>New options for dependency generation: <code><nobr>-MD</nobr></code>, -<code><nobr>-MF</nobr></code>, <code><nobr>-MP</nobr></code>, -<code><nobr>-MT</nobr></code>, <code><nobr>-MQ</nobr></code>. -<li>New preprocessor directives <code><nobr>%pathsearch</nobr></code> and -<code><nobr>%depend</nobr></code>; INCBIN reimplemented as a macro. -<li><code><nobr>%include</nobr></code> now resolves macros in a sane -manner. -<li><code><nobr>%substr</nobr></code> can now be used to get other than -one-character substrings. -<li>New type of character/string constants, using backquotes -(<code><nobr>`...`</nobr></code>), which support C-style escape sequences. -<li><code><nobr>%defstr</nobr></code> and -<code><nobr>%idefstr</nobr></code> to stringize macro definitions before -creation. -<li>Fix forward references used in <code><nobr>EQU</nobr></code> -statements. -</ul> -<h4><a name="section-C.1.9">C.1.9 Version 2.02</a></h4> -<ul> -<li>Additional fixes for MMX operands with explicit -<code><nobr>qword</nobr></code>, as well as (hopefully) SSE operands with -<code><nobr>oword</nobr></code>. -<li>Fix handling of truncated strings with <code><nobr>DO</nobr></code>. -<li>Fix segfaults due to memory overwrites when floating-point constants -were used. -<li>Fix segfaults due to missing include files. -<li>Fix OpenWatcom Makefiles for DOS and OS/2. -<li>Add autogenerated instruction list back into the documentation. -<li>ELF: Fix segfault when generating stabs, and no symbols have been -defined. -<li>ELF: Experimental support for DWARF debugging information. -<li>New compile date and time standard macros. -<li><code><nobr>%ifnum</nobr></code> now returns true for negative numbers. -<li>New <code><nobr>%iftoken</nobr></code> test for a single token. -<li>New <code><nobr>%ifempty</nobr></code> test for empty expansion. -<li>Add support for the <code><nobr>XSAVE</nobr></code> instruction group. -<li>Makefile for Netware/gcc. -<li>Fix issue with some warnings getting emitted way too many times. -<li>Autogenerated instruction list added to the documentation. -</ul> -<h4><a name="section-C.1.10">C.1.10 Version 2.01</a></h4> -<ul> -<li>Fix the handling of MMX registers with explicit -<code><nobr>qword</nobr></code> tags on memory (broken in 2.00 due to -64-bit changes.) -<li>Fix the PREFETCH instructions. -<li>Fix the documentation. -<li>Fix debugging info when using <code><nobr>-f elf</nobr></code> -(backwards compatibility alias for <code><nobr>-f elf32</nobr></code>). -<li>Man pages for rdoff tools (from the Debian project.) -<li>ELF: handle large numbers of sections. -<li>Fix corrupt output when the optimizer runs out of passes. -</ul> -<h4><a name="section-C.1.11">C.1.11 Version 2.00</a></h4> -<ul> -<li>Added c99 data-type compliance. -<li>Added general x86-64 support. -<li>Added win64 (x86-64 COFF) output format. -<li>Added <code><nobr>__BITS__</nobr></code> standard macro. -<li>Renamed the <code><nobr>elf</nobr></code> output format to -<code><nobr>elf32</nobr></code> for clarity. -<li>Added <code><nobr>elf64</nobr></code> and -<code><nobr>macho</nobr></code> (MacOS X) output formats. -<li>Added Numeric constants in <code><nobr>dq</nobr></code> directive. -<li>Added <code><nobr>oword</nobr></code>, <code><nobr>do</nobr></code> and -<code><nobr>reso</nobr></code> pseudo operands. -<li>Allow underscores in numbers. -<li>Added 8-, 16- and 128-bit floating-point formats. -<li>Added binary, octal and hexadecimal floating-point. -<li>Correct the generation of floating-point constants. -<li>Added floating-point option control. -<li>Added Infinity and NaN floating point support. -<li>Added ELF Symbol Visibility support. -<li>Added setting OSABI value in ELF header directive. -<li>Added Generate Makefile Dependencies option. -<li>Added Unlimited Optimization Passes option. -<li>Added <code><nobr>%IFN</nobr></code> and -<code><nobr>%ELIFN</nobr></code> support. -<li>Added Logical Negation Operator. -<li>Enhanced Stack Relative Preprocessor Directives. -<li>Enhanced ELF Debug Formats. -<li>Enhanced Send Errors to a File option. -<li>Added SSSE3, SSE4.1, SSE4.2, SSE5 support. -<li>Added a large number of additional instructions. -<li>Significant performance improvements. -<li><code><nobr>-w+warning</nobr></code> and -<code><nobr>-w-warning</nobr></code> can now be written as -Wwarning and --Wno-warning, respectively. See -<a href="nasmdoc2.html#section-2.1.24">section 2.1.24</a>. -<li>Add <code><nobr>-w+error</nobr></code> to treat warnings as errors. See -<a href="nasmdoc2.html#section-2.1.24">section 2.1.24</a>. -<li>Add <code><nobr>-w+all</nobr></code> and -<code><nobr>-w-all</nobr></code> to enable or disable all suppressible -warnings. See <a href="nasmdoc2.html#section-2.1.24">section 2.1.24</a>. -</ul> -<h3><a name="section-C.2">C.2 NASM 0.98 Series</a></h3> -<p>The 0.98 series was the production versions of NASM from 1999 to 2007. -<h4><a name="section-C.2.1">C.2.1 Version 0.98.39</a></h4> -<ul> -<li>fix buffer overflow -<li>fix outas86's <code><nobr>.bss</nobr></code> handling -<li>"make spotless" no longer deletes config.h.in. -<li><code><nobr>%(el)if(n)idn</nobr></code> insensitivity to string quotes -difference (#809300). -<li>(nasm.c)<code><nobr>__OUTPUT_FORMAT__</nobr></code> changed to string -value instead of symbol. -</ul> -<h4><a name="section-C.2.2">C.2.2 Version 0.98.38</a></h4> -<ul> -<li>Add Makefile for 16-bit DOS binaries under OpenWatcom, and modify -<code><nobr>mkdep.pl</nobr></code> to be able to generate completely -pathless dependencies, as required by OpenWatcom wmake (it supports path -searches, but not explicit paths.) -<li>Fix the <code><nobr>STR</nobr></code> instruction. -<li>Fix the ELF output format, which was broken under certain circumstances -due to the addition of stabs support. -<li>Quick-fix Borland format debug-info for -<code><nobr>-f obj</nobr></code> -<li>Fix for <code><nobr>%rep</nobr></code> with no arguments (#560568) -<li>Fix concatenation of preprocessor function call (#794686) -<li>Fix long label causes coredump (#677841) -<li>Use autoheader as well as autoconf to keep configure from generating -ridiculously long command lines. -<li>Make sure that all of the formats which support debugging output -actually will suppress debugging output when <code><nobr>-g</nobr></code> -not specified. -</ul> -<h4><a name="section-C.2.3">C.2.3 Version 0.98.37</a></h4> -<ul> -<li>Paths given in <code><nobr>-I</nobr></code> switch searched for -<code><nobr>incbin</nobr></code>-ed as well as -<code><nobr>%include</nobr></code>-ed files. -<li>Added stabs debugging for the ELF output format, patch from Martin -Wawro. -<li>Fix <code><nobr>output/outbin.c</nobr></code> to allow origin > -80000000h. -<li>Make <code><nobr>-U</nobr></code> switch work. -<li>Fix the use of relative offsets with explicit prefixes, e.g. -<code><nobr>a32 loop foo</nobr></code>. -<li>Remove <code><nobr>backslash()</nobr></code>. -<li>Fix the <code><nobr>SMSW</nobr></code> and -<code><nobr>SLDT</nobr></code> instructions. -<li><code><nobr>-O2</nobr></code> and <code><nobr>-O3</nobr></code> are no -longer aliases for <code><nobr>-O10</nobr></code> and -<code><nobr>-O15</nobr></code>. If you mean the latter, please say so! :) -</ul> -<h4><a name="section-C.2.4">C.2.4 Version 0.98.36</a></h4> -<ul> -<li>Update rdoff - librarian/archiver - common rec - docs! -<li>Fix signed/unsigned problems. -<li>Fix <code><nobr>JMP FAR label</nobr></code> and -<code><nobr>CALL FAR label</nobr></code>. -<li>Add new multisection support - map files - fix align bug -<li>Fix sysexit, movhps/movlps reg,reg bugs in insns.dat -<li><code><nobr>Q</nobr></code> or <code><nobr>O</nobr></code> suffixes -indicate octal -<li>Support Prescott new instructions (PNI). -<li>Cyrix <code><nobr>XSTORE</nobr></code> instruction. -</ul> -<h4><a name="section-C.2.5">C.2.5 Version 0.98.35</a></h4> -<ul> -<li>Fix build failure on 16-bit DOS (Makefile.bc3 workaround for compiler -bug.) -<li>Fix dependencies and compiler warnings. -<li>Add "const" in a number of places. -<li>Add -X option to specify error reporting format (use -Xvc to integrate -with Microsoft Visual Studio.) -<li>Minor changes for code legibility. -<li>Drop use of tmpnam() in rdoff (security fix.) -</ul> -<h4><a name="section-C.2.6">C.2.6 Version 0.98.34</a></h4> -<ul> -<li>Correct additional address-size vs. operand-size confusions. -<li>Generate dependencies for all Makefiles automatically. -<li>Add support for unimplemented (but theoretically available) registers -such as tr0 and cr5. Segment registers 6 and 7 are called segr6 and segr7 -for the operations which they can be represented. -<li>Correct some disassembler bugs related to redundant address-size -prefixes. Some work still remains in this area. -<li>Correctly generate an error for things like "SEG eax". -<li>Add the JMPE instruction, enabled by "CPU IA64". -<li>Correct compilation on newer gcc/glibc platforms. -<li>Issue an error on things like "jmp far eax". -</ul> -<h4><a name="section-C.2.7">C.2.7 Version 0.98.33</a></h4> -<ul> -<li>New __NASM_PATCHLEVEL__ and __NASM_VERSION_ID__ standard macros to -round out the version-query macros. version.pl now understands X.YYplWW or -X.YY.ZZplWW as a version number, equivalent to X.YY.ZZ.WW (or X.YY.0.WW, as -appropriate). -<li>New keyword "strict" to disable the optimization of specific operands. -<li>Fix the handing of size overrides with JMP instructions (instructions -such as "jmp dword foo".) -<li>Fix the handling of "ABSOLUTE label", where "label" points into a -relocatable segment. -<li>Fix OBJ output format with lots of externs. -<li>More documentation updates. -<li>Add -Ov option to get verbose information about optimizations. -<li>Undo a braindead change which broke <code><nobr>%elif</nobr></code> -directives. -<li>Makefile updates. -</ul> -<h4><a name="section-C.2.8">C.2.8 Version 0.98.32</a></h4> -<ul> -<li>Fix NASM crashing when <code><nobr>%macro</nobr></code> directives were -left unterminated. -<li>Lots of documentation updates. -<li>Complete rewrite of the PostScript/PDF documentation generator. -<li>The MS Visual C++ Makefile was updated and corrected. -<li>Recognize .rodata as a standard section name in ELF. -<li>Fix some obsolete Perl4-isms in Perl scripts. -<li>Fix configure.in to work with autoconf 2.5x. -<li>Fix a couple of "make cleaner" misses. -<li>Make the normal "./configure && make" work with Cygwin. -</ul> -<h4><a name="section-C.2.9">C.2.9 Version 0.98.31</a></h4> -<ul> -<li>Correctly build in a separate object directory again. -<li>Derive all references to the version number from the version file. -<li>New standard macros __NASM_SUBMINOR__ and __NASM_VER__ macros. -<li>Lots of Makefile updates and bug fixes. -<li>New <code><nobr>%ifmacro</nobr></code> directive to test for multiline -macros. -<li>Documentation updates. -<li>Fixes for 16-bit OBJ format output. -<li>Changed the NASM environment variable to NASMENV. -</ul> -<h4><a name="section-C.2.10">C.2.10 Version 0.98.30</a></h4> -<ul> -<li>Changed doc files a lot: completely removed old READMExx and Wishlist -files, incorporating all information in CHANGES and TODO. -<li>I waited a long time to rename zoutieee.c to (original) outieee.c -<li>moved all output modules to output/ subdirectory. -<li>Added 'make strip' target to strip debug info from nasm & ndisasm. -<li>Added INSTALL file with installation instructions. -<li>Added -v option description to nasm man. -<li>Added dist makefile target to produce source distributions. -<li>16-bit support for ELF output format (GNU extension, but useful.) -</ul> -<h4><a name="section-C.2.11">C.2.11 Version 0.98.28</a></h4> -<ul> -<li>Fastcooked this for Debian's Woody release: Frank applied the INCBIN -bug patch to 0.98.25alt and called it 0.98.28 to not confuse poor little -apt-get. -</ul> -<h4><a name="section-C.2.12">C.2.12 Version 0.98.26</a></h4> -<ul> -<li>Reorganised files even better from 0.98.25alt -</ul> -<h4><a name="section-C.2.13">C.2.13 Version 0.98.25alt</a></h4> -<ul> -<li>Prettified the source tree. Moved files to more reasonable places. -<li>Added findleak.pl script to misc/ directory. -<li>Attempted to fix doc. -</ul> -<h4><a name="section-C.2.14">C.2.14 Version 0.98.25</a></h4> -<ul> -<li>Line continuation character <code><nobr>\</nobr></code>. -<li>Docs inadvertantly reverted - "dos packaging". -</ul> -<h4><a name="section-C.2.15">C.2.15 Version 0.98.24p1</a></h4> -<ul> -<li>FIXME: Someone, document this please. -</ul> -<h4><a name="section-C.2.16">C.2.16 Version 0.98.24</a></h4> -<ul> -<li>Documentation - Ndisasm doc added to Nasm.doc. -</ul> -<h4><a name="section-C.2.17">C.2.17 Version 0.98.23</a></h4> -<ul> -<li>Attempted to remove rdoff version1 -<li>Lino Mastrodomenico's patches to preproc.c (%$$ bug?). -</ul> -<h4><a name="section-C.2.18">C.2.18 Version 0.98.22</a></h4> -<ul> -<li>Update rdoff2 - attempt to remove v1. -</ul> -<h4><a name="section-C.2.19">C.2.19 Version 0.98.21</a></h4> -<ul> -<li>Optimization fixes. -</ul> -<h4><a name="section-C.2.20">C.2.20 Version 0.98.20</a></h4> -<ul> -<li>Optimization fixes. -</ul> -<h4><a name="section-C.2.21">C.2.21 Version 0.98.19</a></h4> -<ul> -<li>H. J. Lu's patch back out. -</ul> -<h4><a name="section-C.2.22">C.2.22 Version 0.98.18</a></h4> -<ul> -<li>Added ".rdata" to "-f win32". -</ul> -<h4><a name="section-C.2.23">C.2.23 Version 0.98.17</a></h4> -<ul> -<li>H. J. Lu's "bogus elf" patch. (Red Hat problem?) -</ul> -<h4><a name="section-C.2.24">C.2.24 Version 0.98.16</a></h4> -<ul> -<li>Fix whitespace before "[section ..." bug. -</ul> -<h4><a name="section-C.2.25">C.2.25 Version 0.98.15</a></h4> -<ul> -<li>Rdoff changes (?). -<li>Fix fixes to memory leaks. -</ul> -<h4><a name="section-C.2.26">C.2.26 Version 0.98.14</a></h4> -<ul> -<li>Fix memory leaks. -</ul> -<h4><a name="section-C.2.27">C.2.27 Version 0.98.13</a></h4> -<ul> -<li>There was no 0.98.13 -</ul> -<h4><a name="section-C.2.28">C.2.28 Version 0.98.12</a></h4> -<ul> -<li>Update optimization (new function of "-O1") -<li>Changes to test/bintest.asm (?). -</ul> -<h4><a name="section-C.2.29">C.2.29 Version 0.98.11</a></h4> -<ul> -<li>Optimization changes. -<li>Ndisasm fixed. -</ul> -<h4><a name="section-C.2.30">C.2.30 Version 0.98.10</a></h4> -<ul> -<li>There was no 0.98.10 -</ul> -<h4><a name="section-C.2.31">C.2.31 Version 0.98.09</a></h4> -<ul> -<li>Add multiple sections support to "-f bin". -<li>Changed GLOBAL_TEMP_BASE in outelf.c from 6 to 15. -<li>Add "-v" as an alias to the "-r" switch. -<li>Remove "#ifdef" from Tasm compatibility options. -<li>Remove redundant size-overrides on "mov ds, ex", etc. -<li>Fixes to SSE2, other insns.dat (?). -<li>Enable uppercase "I" and "P" switches. -<li>Case insinsitive "seg" and "wrt". -<li>Update install.sh (?). -<li>Allocate tokens in blocks. -<li>Improve "invalid effective address" messages. -</ul> -<h4><a name="section-C.2.32">C.2.32 Version 0.98.08</a></h4> -<ul> -<li>Add "<code><nobr>%strlen</nobr></code>" and -"<code><nobr>%substr</nobr></code>" macro operators -<li>Fixed broken c16.mac. -<li>Unterminated string error reported. -<li>Fixed bugs as per 0.98bf -</ul> -<h4><a name="section-C.2.33">C.2.33 Version 0.98.09b with John Coffman patches released 28-Oct-2001</a></h4> -<p>Changes from 0.98.07 release to 98.09b as of 28-Oct-2001 -<ul> -<li>More closely compatible with 0.98 when -O0 is implied or specified. Not -strictly identical, since backward branches in range of short offsets are -recognized, and signed byte values with no explicit size specification will -be assembled as a single byte. -<li>More forgiving with the PUSH instruction. 0.98 requires a size to be -specified always. 0.98.09b will imply the size from the current BITS -setting (16 or 32). -<li>Changed definition of the optimization flag: -</ul> -<p>-O0 strict two-pass assembly, JMP and Jcc are handled more like 0.98, -except that back- ward JMPs are short, if possible. -<p>-O1 strict two-pass assembly, but forward branches are assembled with -code guaranteed to reach; may produce larger code than -O0, but will -produce successful assembly more often if branch offset sizes are not -specified. -<p>-O2 multi-pass optimization, minimize branch offsets; also will minimize -signed immed- iate bytes, overriding size specification. -<p>-O3 like -O2, but more passes taken, if needed -<h4><a name="section-C.2.34">C.2.34 Version 0.98.07 released 01/28/01</a></h4> -<ul> -<li>Added Stepane Denis' SSE2 instructions to a *working* version of the -code - some earlier versions were based on broken code - sorry 'bout that. -version "0.98.07" -</ul> -<p>01/28/01 -<ul> -<li>Cosmetic modifications to nasm.c, nasm.h, AUTHORS, MODIFIED -</ul> -<h4><a name="section-C.2.35">C.2.35 Version 0.98.06f released 01/18/01</a></h4> -<ul> -<li>- Add "metalbrain"s jecxz bug fix in insns.dat - alter nasmdoc.src to -match - version "0.98.06f" -</ul> -<h4><a name="section-C.2.36">C.2.36 Version 0.98.06e released 01/09/01</a></h4> -<ul> -<li>Removed the "outforms.h" file - it appears to be someone's old backup -of "outform.h". version "0.98.06e" -</ul> -<p>01/09/01 -<ul> -<li>fbk - finally added the fix for the "multiple %includes bug", known -since 7/27/99 - reported originally (?) and sent to us by Austin Lunnen - -he reports that John Fine had a fix within the day. Here it is... -<li>Nelson Rush resigns from the group. Big thanks to Nelson for his -leadership and enthusiasm in getting these changes incorporated into Nasm! -<li>fbk - [list +], [list -] directives - ineptly implemented, should be -re-written or removed, perhaps. -<li>Brian Raiter / fbk - "elfso bug" fix - applied to aoutb format as well -- testing might be desirable... -</ul> -<p>08/07/00 -<ul> -<li>James Seter - -postfix, -prefix command line switches. -<li>Yuri Zaporogets - rdoff utility changes. -</ul> -<h4><a name="section-C.2.37">C.2.37 Version 0.98p1</a></h4> -<ul> -<li>GAS-like palign (Panos Minos) -<li>FIXME: Someone, fill this in with details -</ul> -<h4><a name="section-C.2.38">C.2.38 Version 0.98bf (bug-fixed)</a></h4> -<ul> -<li>Fixed - elf and aoutb bug - shared libraries - multiple "%include" bug -in "-f obj" - jcxz, jecxz bug - unrecognized option bug in ndisasm -</ul> -<h4><a name="section-C.2.39">C.2.39 Version 0.98.03 with John Coffman's changes released 27-Jul-2000</a></h4> -<ul> -<li>Added signed byte optimizations for the 0x81/0x83 class of -instructions: ADC, ADD, AND, CMP, OR, SBB, SUB, XOR: when used as 'ADD -reg16,imm' or 'ADD reg32,imm.' Also optimization of signed byte form of -'PUSH imm' and 'IMUL reg,imm'/'IMUL reg,reg,imm.' No size specification is -needed. -<li>Added multi-pass JMP and Jcc offset optimization. Offsets on forward -references will preferentially use the short form, without the need to code -a specific size (short or near) for the branch. Added instructions for 'Jcc -label' to use the form 'Jnotcc $+3/JMP label', in cases where a short -offset is out of bounds. If compiling for a 386 or higher CPU, then the 386 -form of Jcc will be used instead. -</ul> -<p>This feature is controlled by a new command-line switch: "O", (upper -case letter O). "-O0" reverts the assembler to no extra optimization -passes, "-O1" allows up to 5 extra passes, and "-O2"(default), allows up to -10 extra optimization passes. -<ul> -<li>Added a new directive: 'cpu XXX', where XXX is any of: 8086, 186, 286, -386, 486, 586, pentium, 686, PPro, P2, P3 or Katmai. All are case -insensitive. All instructions will be selected only if they apply to the -selected cpu or lower. Corrected a couple of bugs in cpu-dependence in -'insns.dat'. -<li>Added to 'standard.mac', the "use16" and "use32" forms of the "bits -16/32" directive. This is nothing new, just conforms to a lot of other -assemblers. (minor) -<li>Changed label allocation from 320/32 (10000 labels @ 200K+) to 32/37 -(1000 labels); makes running under DOS much easier. Since additional label -space is allocated dynamically, this should have no effect on large -programs with lots of labels. The 37 is a prime, believed to be better for -hashing. (minor) -</ul> -<h4><a name="section-C.2.40">C.2.40 Version 0.98.03</a></h4> -<p>"Integrated patchfile 0.98-0.98.01. I call this version 0.98.03 for -historical reasons: 0.98.02 was trashed." --John Coffman -<johninsd@san.rr.com>, 27-Jul-2000 -<ul> -<li>Kendall Bennett's SciTech MGL changes -<li>Note that you must define "TASM_COMPAT" at compile-time to get the Tasm -Ideal Mode compatibility. -<li>All changes can be compiled in and out using the TASM_COMPAT macros, -and when compiled without TASM_COMPAT defined we get the exact same binary -as the unmodified 0.98 sources. -<li>standard.mac, macros.c: Added macros to ignore TASM directives before -first include -<li>nasm.h: Added extern declaration for tasm_compatible_mode -<li>nasm.c: Added global variable tasm_compatible_mode -<li>Added command line switch for TASM compatible mode (-t) -<li>Changed version command line to reflect when compiled with TASM -additions -<li>Added response file processing to allow all arguments on a single line -(response file is @resp rather than -@resp for NASM format). -<li>labels.c: Changes islocal() macro to support TASM style @@local labels. -<li>Added islocalchar() macro to support TASM style @@local labels. -<li>parser.c: Added support for TASM style memory references (ie: mov -[DWORD eax],10 rather than the NASM style mov DWORD [eax],10). -<li>preproc.c: Added new directives, <code><nobr>%arg</nobr></code>, -<code><nobr>%local</nobr></code>, <code><nobr>%stacksize</nobr></code> to -directives table -<li>Added support for TASM style directives without a leading % symbol. -<li>Integrated a block of changes from Andrew Zabolotny -<bit@eltech.ru>: -<li>A new keyword <code><nobr>%xdefine</nobr></code> and its -case-insensitive counterpart <code><nobr>%ixdefine</nobr></code>. They work -almost the same way as <code><nobr>%define</nobr></code> and -<code><nobr>%idefine</nobr></code> but expand the definition immediately, -not on the invocation. Something like a cross between -<code><nobr>%define</nobr></code> and <code><nobr>%assign</nobr></code>. -The "x" suffix stands for "eXpand", so "xdefine" can be deciphered as -"expand-and-define". Thus you can do things like this: -</ul> -<p><pre> - %assign ofs 0 - - %macro arg 1 - %xdefine %1 dword [esp+ofs] - %assign ofs ofs+4 - %endmacro -</pre> -<ul> -<li>Changed the place where the expansion of %$name macros are expanded. -Now they are converted into ..@ctxnum.name form when detokenizing, so there -are no quirks as before when using %$name arguments to macros, in macros -etc. For example: -</ul> -<p><pre> - %macro abc 1 - %define %1 hello - %endm - - abc %$here - %$here -</pre> -<p>Now last line will be expanded into "hello" as expected. This also -allows for lots of goodies, a good example are extended "proc" macros -included in this archive. -<ul> -<li>Added a check for "cstk" in smacro_defined() before calling get_ctx() - -this allows for things like: -</ul> -<p><pre> - %ifdef %$abc - %endif -</pre> -<p>to work without warnings even in no context. -<ul> -<li>Added a check for "cstk" in %if*ctx and %elif*ctx directives - this -allows to use <code><nobr>%ifctx</nobr></code> without excessive warnings. -If there is no active context, <code><nobr>%ifctx</nobr></code> goes -through "false" branch. -<li>Removed "user error: " prefix with <code><nobr>%error</nobr></code> -directive: it just clobbers the output and has absolutely no functionality. -Besides, this allows to write macros that does not differ from built-in -functions in any way. -<li>Added expansion of string that is output by -<code><nobr>%error</nobr></code> directive. Now you can do things like: -</ul> -<p><pre> - %define hello(x) Hello, x! - - %define %$name andy - %error "hello(%$name)" -</pre> -<p>Same happened with <code><nobr>%include</nobr></code> directive. -<ul> -<li>Now all directives that expect an identifier will try to expand and -concatenate everything without whitespaces in between before usage. For -example, with "unfixed" nasm the commands -</ul> -<p><pre> - %define %$abc hello - %define __%$abc goodbye - __%$abc -</pre> -<p>would produce "incorrect" output: last line will expand to -<p><pre> - hello goodbyehello -</pre> -<p>Not quite what you expected, eh? :-) The answer is that preprocessor -treats the <code><nobr>%define</nobr></code> construct as if it would be -<p><pre> - %define __ %$abc goodbye -</pre> -<p>(note the white space between __ and %$abc). After my "fix" it will -"correctly" expand into -<p><pre> - goodbye -</pre> -<p>as expected. Note that I use quotes around words "correct", "incorrect" -etc because this is rather a feature not a bug; however current behaviour -is more logical (and allows more advanced macro usage :-). -<p>Same change was applied to: -<code><nobr>%push</nobr></code>,<code><nobr>%macro</nobr></code>,<code><nobr>%imacro</nobr></code>,<code><nobr>%define</nobr></code>,<code><nobr>%idefine</nobr></code>,<code><nobr>%xdefine</nobr></code>,<code><nobr>%ixdefine</nobr></code>, -<code><nobr>%assign</nobr></code>,<code><nobr>%iassign</nobr></code>,<code><nobr>%undef</nobr></code> -<ul> -<li>A new directive [WARNING {+|-}warning-id] have been added. It works -only if the assembly phase is enabled (i.e. it doesn't work with nasm -e). -<li>A new warning type: macro-selfref. By default this warning is disabled; -when enabled NASM warns when a macro self-references itself; for example -the following source: -</ul> -<p><pre> - [WARNING macro-selfref] - - %macro push 1-* - %rep %0 - push %1 - %rotate 1 - %endrep - %endmacro - - push eax,ebx,ecx -</pre> -<p>will produce a warning, but if we remove the first line we won't see it -anymore (which is The Right Thing To Do {tm} IMHO since C preprocessor eats -such constructs without warnings at all). -<ul> -<li>Added a "error" routine to preprocessor which always will set ERR_PASS1 -bit in severity_code. This removes annoying repeated errors on first and -second passes from preprocessor. -<li>Added the %+ operator in single-line macros for concatenating two -identifiers. Usage example: -</ul> -<p><pre> - %define _myfunc _otherfunc - %define cextern(x) _ %+ x - cextern (myfunc) -</pre> -<p>After first expansion, third line will become "_myfunc". After this -expansion is performed again so it becomes "_otherunc". -<ul> -<li>Now if preprocessor is in a non-emitting state, no warning or error -will be emitted. Example: -</ul> -<p><pre> - %if 1 - mov eax,ebx - %else - put anything you want between these two brackets, - even macro-parameter references %1 or local - labels %$zz or macro-local labels %%zz - no - warning will be emitted. - %endif -</pre> -<ul> -<li>Context-local variables on expansion as a last resort are looked up in -outer contexts. For example, the following piece: -</ul> -<p><pre> - %push outer - %define %$a [esp] - - %push inner - %$a - %pop - %pop -</pre> -<p>will expand correctly the fourth line to [esp]; if we'll define another -%$a inside the "inner" context, it will take precedence over outer -definition. However, this modification has been applied only to -expand_smacro and not to smacro_define: as a consequence expansion looks in -outer contexts, but <code><nobr>%ifdef</nobr></code> won't look in outer -contexts. -<p>This behaviour is needed because we don't want nested contexts to act on -already defined local macros. Example: -<p><pre> - %define %$arg1 [esp+4] - test eax,eax - if nz - mov eax,%$arg1 - endif -</pre> -<p>In this example the "if" mmacro enters into the "if" context, so %$arg1 -is not valid anymore inside "if". Of course it could be worked around by -using explicitely %$$arg1 but this is ugly IMHO. -<ul> -<li>Fixed memory leak in <code><nobr>%undef</nobr></code>. The origline -wasn't freed before exiting on success. -<li>Fixed trap in preprocessor when line expanded to empty set of tokens. -This happens, for example, in the following case: -</ul> -<p><pre> - #define SOMETHING - SOMETHING -</pre> -<h4><a name="section-C.2.41">C.2.41 Version 0.98</a></h4> -<p>All changes since NASM 0.98p3 have been produced by H. Peter Anvin -<hpa@zytor.com>. -<ul> -<li>The documentation comment delimiter is -<li>Allow EQU definitions to refer to external labels; reported by Pedro -Gimeno. -<li>Re-enable support for RDOFF v1; reported by Pedro Gimeno. -<li>Updated License file per OK from Simon and Julian. -</ul> -<h4><a name="section-C.2.42">C.2.42 Version 0.98p9</a></h4> -<ul> -<li>Update documentation (although the instruction set reference will have -to wait; I don't want to hold up the 0.98 release for it.) -<li>Verified that the NASM implementation of the PEXTRW and PMOVMSKB -instructions is correct. The encoding differs from what the Intel manuals -document, but the Pentium III behaviour matches NASM, not the Intel -manuals. -<li>Fix handling of implicit sizes in PSHUFW and PINSRW, reported by Stefan -Hoffmeister. -<li>Resurrect the -s option, which was removed when changing the diagnostic -output to stdout. -</ul> -<h4><a name="section-C.2.43">C.2.43 Version 0.98p8</a></h4> -<ul> -<li>Fix for "DB" when NASM is running on a bigendian machine. -<li>Invoke insns.pl once for each output script, making Makefile.in legal -for "make -j". -<li>Improve the Unix configure-based makefiles to make package creation -easier. -<li>Included an RPM .spec file for building RPM (RedHat Package Manager) -packages on Linux or Unix systems. -<li>Fix Makefile dependency problems. -<li>Change src/rdsrc.pl to include sectioning information in info output; -required for install-info to work. -<li>Updated the RDOFF distribution to version 2 from Jules; minor massaging -to make it compile in my environment. -<li>Split doc files that can be built by anyone with a Perl interpreter off -into a separate archive. -<li>"Dress rehearsal" release! -</ul> -<h4><a name="section-C.2.44">C.2.44 Version 0.98p7</a></h4> -<ul> -<li>Fixed opcodes with a third byte-sized immediate argument to not -complain if given "byte" on the immediate. -<li>Allow <code><nobr>%undef</nobr></code> to remove single-line macros -with arguments. This matches the behaviour of #undef in the C preprocessor. -<li>Allow -d, -u, -i and -p to be specified as -D, -U, -I and -P for -compatibility with most C compilers and preprocessors. This allows Makefile -options to be shared between cc and nasm, for example. -<li>Minor cleanups. -<li>Went through the list of Katmai instructions and hopefully fixed the -(rather few) mistakes in it. -<li>(Hopefully) fixed a number of disassembler bugs related to ambiguous -instructions (disambiguated by -p) and SSE instructions with REP. -<li>Fix for bug reported by Mark Junger: "call dword 0x12345678" should -work and may add an OSP (affected CALL, JMP, Jcc). -<li>Fix for environments when "stderr" isn't a compile-time constant. -</ul> -<h4><a name="section-C.2.45">C.2.45 Version 0.98p6</a></h4> -<ul> -<li>Took officially over coordination of the 0.98 release; so drop the p3.x -notation. Skipped p4 and p5 to avoid confusion with John Fine's J4 and J5 -releases. -<li>Update the documentation; however, it still doesn't include -documentation for the various new instructions. I somehow wonder if it -makes sense to have an instruction set reference in the assembler manual -when Intel et al have PDF versions of their manuals online. -<li>Recognize "idt" or "centaur" for the -p option to ndisasm. -<li>Changed error messages back to stderr where they belong, but add an -E -option to redirect them elsewhere (the DOS shell cannot redirect stderr.) -<li>-M option to generate Makefile dependencies (based on code from Alex -Verstak.) -<li><code><nobr>%undef</nobr></code> preprocessor directive, and -u option, -that undefines a single-line macro. -<li>OS/2 Makefile (Mkfiles/Makefile.os2) for Borland under OS/2; from Chuck -Crayne. -<li>Various minor bugfixes (reported by): - Dangling -<code><nobr>%s</nobr></code> in preproc.c (Martin Junker) -<li>THERE ARE KNOWN BUGS IN SSE AND THE OTHER KATMAI INSTRUCTIONS. I am on -a trip and didn't bring the Katmai instruction reference, so I can't work -on them right now. -<li>Updated the License file per agreement with Simon and Jules to include -a GPL distribution clause. -</ul> -<h4><a name="section-C.2.46">C.2.46 Version 0.98p3.7</a></h4> -<ul> -<li>(Hopefully) fixed the canned Makefiles to include the outrdf2 and -zoutieee modules. -<li>Renamed changes.asm to changed.asm. -</ul> -<h4><a name="section-C.2.47">C.2.47 Version 0.98p3.6</a></h4> -<ul> -<li>Fixed a bunch of instructions that were added in 0.98p3.5 which had -memory operands, and the address-size prefix was missing from the -instruction pattern. -</ul> -<h4><a name="section-C.2.48">C.2.48 Version 0.98p3.5</a></h4> -<ul> -<li>Merged in changes from John S. Fine's 0.98-J5 release. John's based -0.98-J5 on my 0.98p3.3 release; this merges the changes. -<li>Expanded the instructions flag field to a long so we can fit more -flags; mark SSE (KNI) and AMD or Katmai-specific instructions as such. -<li>Fix the "PRIV" flag on a bunch of instructions, and create new "PROT" -flag for protected-mode-only instructions (orthogonal to if the instruction -is privileged!) and new "SMM" flag for SMM-only instructions. -<li>Added AMD-only SYSCALL and SYSRET instructions. -<li>Make SSE actually work, and add new Katmai MMX instructions. -<li>Added a -p (preferred vendor) option to ndisasm so that it can -distinguish e.g. Cyrix opcodes also used in SSE. For example: -</ul> -<p><pre> - ndisasm -p cyrix aliased.bin - 00000000 670F514310 paddsiw mm0,[ebx+0x10] - 00000005 670F514320 paddsiw mm0,[ebx+0x20] - ndisasm -p intel aliased.bin - 00000000 670F514310 sqrtps xmm0,[ebx+0x10] - 00000005 670F514320 sqrtps xmm0,[ebx+0x20] -</pre> -<ul> -<li>Added a bunch of Cyrix-specific instructions. -</ul> -<h4><a name="section-C.2.49">C.2.49 Version 0.98p3.4</a></h4> -<ul> -<li>Made at least an attempt to modify all the additional Makefiles (in the -Mkfiles directory). I can't test it, but this was the best I could do. -<li>DOS DJGPP+"Opus Make" Makefile from John S. Fine. -<li>changes.asm changes from John S. Fine. -</ul> -<h4><a name="section-C.2.50">C.2.50 Version 0.98p3.3</a></h4> -<ul> -<li>Patch from Conan Brink to allow nesting of -<code><nobr>%rep</nobr></code> directives. -<li>If we're going to allow INT01 as an alias for INT1/ICEBP (one of Jules -0.98p3 changes), then we should allow INT03 as an alias for INT3 as well. -<li>Updated changes.asm to include the latest changes. -<li>Tried to clean up the <CR>s that had snuck in from a DOS/Windows -environment into my Unix environment, and try to make sure than DOS/Windows -users get them back. -<li>We would silently generate broken tools if insns.dat wasn't sorted -properly. Change insns.pl so that the order doesn't matter. -<li>Fix bug in insns.pl (introduced by me) which would cause conditional -instructions to have an extra "cc" in disassembly, e.g. "jnz" disassembled -as "jccnz". -</ul> -<h4><a name="section-C.2.51">C.2.51 Version 0.98p3.2</a></h4> -<ul> -<li>Merged in John S. Fine's changes from his 0.98-J4 prerelease; see -http://www.csoft.net/cz/johnfine/ -<li>Changed previous "spotless" Makefile target (appropriate for -distribution) to "distclean", and added "cleaner" target which is same as -"clean" except deletes files generated by Perl scripts; "spotless" is -union. -<li>Removed BASIC programs from distribution. Get a Perl interpreter -instead (see below.) -<li>Calling this "pre-release 3.2" rather than "p3-hpa2" because of John's -contributions. -<li>Actually link in the IEEE output format (zoutieee.c); fix a bunch of -compiler warnings in that file. Note I don't know what IEEE output is -supposed to look like, so these changes were made "blind". -</ul> -<h4><a name="section-C.2.52">C.2.52 Version 0.98p3-hpa</a></h4> -<ul> -<li>Merged nasm098p3.zip with nasm-0.97.tar.gz to create a fully buildable -version for Unix systems (Makefile.in updates, etc.) -<li>Changed insns.pl to create the instruction tables in nasm.h and -names.c, so that a new instruction can be added by adding it *only* to -insns.dat. -<li>Added the following new instructions: SYSENTER, SYSEXIT, FXSAVE, -FXRSTOR, UD1, UD2 (the latter two are two opcodes that Intel guarantee will -never be used; one of them is documented as UD2 in Intel documentation, the -other one just as "Undefined Opcode" -- calling it UD1 seemed to make -sense.) -<li>MAX_SYMBOL was defined to be 9, but LOADALL286 and LOADALL386 are 10 -characters long. Now MAX_SYMBOL is derived from insns.dat. -<li>A note on the BASIC programs included: forget them. insns.bas is -already out of date. Get yourself a Perl interpreter for your platform of -choice at -<a href="http://www.cpan.org/ports/index.html">http://www.cpan.org/ports/index.html</a>. -</ul> -<h4><a name="section-C.2.53">C.2.53 Version 0.98 pre-release 3</a></h4> -<ul> -<li>added response file support, improved command line handling, new layout -help screen -<li>fixed limit checking bug, 'OUT byte nn, reg' bug, and a couple of rdoff -related bugs, updated Wishlist; 0.98 Prerelease 3. -</ul> -<h4><a name="section-C.2.54">C.2.54 Version 0.98 pre-release 2</a></h4> -<ul> -<li>fixed bug in outcoff.c to do with truncating section names longer than -8 characters, referencing beyond end of string; 0.98 pre-release 2 -</ul> -<h4><a name="section-C.2.55">C.2.55 Version 0.98 pre-release 1</a></h4> -<ul> -<li>Fixed a bug whereby STRUC didn't work at all in RDF. -<li>Fixed a problem with group specification in PUBDEFs in OBJ. -<li>Improved ease of adding new output formats. Contribution due to Fox -Cutter. -<li>Fixed a bug in relocations in the `bin' format: was showing up when a -relocatable reference crossed an 8192-byte boundary in any output section. -<li>Fixed a bug in local labels: local-label lookups were inconsistent -between passes one and two if an EQU occurred between the definition of a -global label and the subsequent use of a local label local to that global. -<li>Fixed a seg-fault in the preprocessor (again) which happened when you -use a blank line as the first line of a multi-line macro definition and -then defined a label on the same line as a call to that macro. -<li>Fixed a stale-pointer bug in the handling of the NASM environment -variable. Thanks to Thomas McWilliams. -<li>ELF had a hard limit on the number of sections which caused segfaults -when transgressed. Fixed. -<li>Added ability for ndisasm to read from stdin by using `-' as the -filename. -<li>ndisasm wasn't outputting the TO keyword. Fixed. -<li>Fixed error cascade on bogus expression in -<code><nobr>%if</nobr></code> - an error in evaluation was causing the -entire <code><nobr>%if</nobr></code> to be discarded, thus creating trouble -later when the <code><nobr>%else</nobr></code> or -<code><nobr>%endif</nobr></code> was encountered. -<li>Forward reference tracking was instruction-granular not operand- -granular, which was causing 286-specific code to be generated needlessly on -code of the form `shr word [forwardref],1'. Thanks to Jim Hague for sending -a patch. -<li>All messages now appear on stdout, as sending them to stderr serves no -useful purpose other than to make redirection difficult. -<li>Fixed the problem with EQUs pointing to an external symbol - this now -generates an error message. -<li>Allowed multiple size prefixes to an operand, of which only the first -is taken into account. -<li>Incorporated John Fine's changes, including fixes of a large number of -preprocessor bugs, some small problems in OBJ, and a reworking of label -handling to define labels before their line is assembled, rather than -after. -<li>Reformatted a lot of the source code to be more readable. Included -'coding.txt' as a guideline for how to format code for contributors. -<li>Stopped nested <code><nobr>%reps</nobr></code> causing a panic - they -now cause a slightly more friendly error message instead. -<li>Fixed floating point constant problems (patch by Pedro Gimeno) -<li>Fixed the return value of insn_size() not being checked for -1, -indicating an error. -<li>Incorporated 3Dnow! instructions. -<li>Fixed the 'mov eax, eax + ebx' bug. -<li>Fixed the GLOBAL EQU bug in ELF. Released developers release 3. -<li>Incorporated John Fine's command line parsing changes -<li>Incorporated David Lindauer's OMF debug support -<li>Made changes for LCC 4.0 support -(<code><nobr>__NASM_CDecl__</nobr></code>, removed register size -specification warning when sizes agree). -</ul> -<h3><a name="section-C.3">C.3 NASM 0.9 Series</a></h3> -<p>Revisions before 0.98. -<h4><a name="section-C.3.1">C.3.1 Version 0.97 released December 1997</a></h4> -<ul> -<li>This was entirely a bug-fix release to 0.96, which seems to have got -cursed. Silly me. -<li>Fixed stupid mistake in OBJ which caused `MOV EAX,<constant>' to -fail. Caused by an error in the `MOV EAX,<segment>' support. -<li>ndisasm hung at EOF when compiled with lcc on Linux because lcc on -Linux somehow breaks feof(). ndisasm now does not rely on feof(). -<li>A heading in the documentation was missing due to a markup error in the -indexing. Fixed. -<li>Fixed failure to update all pointers on realloc() within extended- -operand code in parser.c. Was causing wrong behaviour and seg faults on -lines such as `dd 0.0,0.0,0.0,0.0,...' -<li>Fixed a subtle preprocessor bug whereby invoking one multi-line macro -on the first line of the expansion of another, when the second had been -invoked with a label defined before it, didn't expand the inner macro. -<li>Added internal.doc back in to the distribution archives - it was -missing in 0.96 *blush* -<li>Fixed bug causing 0.96 to be unable to assemble its own test files, -specifically objtest.asm. *blush again* -<li>Fixed seg-faults and bogus error messages caused by mismatching -<code><nobr>%rep</nobr></code> and <code><nobr>%endrep</nobr></code> within -multi-line macro definitions. -<li>Fixed a problem with buffer overrun in OBJ, which was causing -corruption at ends of long PUBDEF records. -<li>Separated DOS archives into main-program and documentation to reduce -download size. -</ul> -<h4><a name="section-C.3.2">C.3.2 Version 0.96 released November 1997</a></h4> -<ul> -<li>Fixed a bug whereby, if `nasm sourcefile' would cause a filename -collision warning and put output into `nasm.out', then `nasm sourcefile -o -outputfile' still gave the warning even though the `-o' was honoured. Fixed -name pollution under Digital UNIX: one of its header files defined R_SP, -which broke the enum in nasm.h. -<li>Fixed minor instruction table problems: FUCOM and FUCOMP didn't have -two-operand forms; NDISASM didn't recognise the longer register forms of -PUSH and POP (eg FF F3 for PUSH BX); TEST mem,imm32 was flagged as -undocumented; the 32-bit forms of CMOV had 16-bit operand size prefixes; -`AAD imm' and `AAM imm' are no longer flagged as undocumented because the -Intel Architecture reference documents them. -<li>Fixed a problem with the local-label mechanism, whereby strange types -of symbol (EQUs, auto-defined OBJ segment base symbols) interfered with the -`previous global label' value and screwed up local labels. -<li>Fixed a bug whereby the stub preprocessor didn't communicate with the -listing file generator, so that the -a and -l options in conjunction would -produce a useless listing file. -<li>Merged `os2' object file format back into `obj', after discovering that -`obj' _also_ shouldn't have a link pass separator in a module containing a -non-trivial MODEND. Flat segments are now declared using the FLAT -attribute. `os2' is no longer a valid object format name: use `obj'. -<li>Removed the fixed-size temporary storage in the evaluator. Very very -long expressions (like `mov ax,1+1+1+1+...' for two hundred 1s or so) -should now no longer crash NASM. -<li>Fixed a bug involving segfaults on disassembly of MMX instructions, by -changing the meaning of one of the operand-type flags in nasm.h. This may -cause other apparently unrelated MMX problems; it needs to be tested -thoroughly. -<li>Fixed some buffer overrun problems with large OBJ output files. Thanks -to DJ Delorie for the bug report and fix. -<li>Made preprocess-only mode actually listen to the -<code><nobr>%line</nobr></code> markers as it prints them, so that it can -report errors more sanely. -<li>Re-designed the evaluator to keep more sensible track of expressions -involving forward references: can now cope with previously-nightmare -situations such as: -</ul> -<p><pre> - mov ax,foo | bar - foo equ 1 - bar equ 2 -</pre> -<ul> -<li>Added the ALIGN and ALIGNB standard macros. -<li>Added PIC support in ELF: use of WRT to obtain the four extra -relocation types needed. -<li>Added the ability for output file formats to define their own -extensions to the GLOBAL, COMMON and EXTERN directives. -<li>Implemented common-variable alignment, and global-symbol type and size -declarations, in ELF. -<li>Implemented NEAR and FAR keywords for common variables, plus far-common -element size specification, in OBJ. -<li>Added a feature whereby EXTERNs and COMMONs in OBJ can be given a -default WRT specification (either a segment or a group). -<li>Transformed the Unix NASM archive into an auto-configuring package. -<li>Added a sanity-check for people applying SEG to things which are -already segment bases: this previously went unnoticed by the SEG processing -and caused OBJ-driver panics later. -<li>Added the ability, in OBJ format, to deal with `MOV -EAX,<segment>' type references: OBJ doesn't directly support -dword-size segment base fixups, but as long as the low two bytes of the -constant term are zero, a word-size fixup can be generated instead and it -will work. -<li>Added the ability to specify sections' alignment requirements in Win32 -object files and pure binary files. -<li>Added preprocess-time expression evaluation: the -<code><nobr>%assign</nobr></code> (and <code><nobr>%iassign</nobr></code>) -directive and the bare <code><nobr>%if</nobr></code> (and -<code><nobr>%elif</nobr></code>) conditional. Added relational operators to -the evaluator, for use only in <code><nobr>%if</nobr></code> constructs: -the standard relationals = < > <= >= <> (and C-like -synonyms == and !=) plus low-precedence logical operators &&, ^^ -and ||. -<li>Added a preprocessor repeat construct: <code><nobr>%rep</nobr></code> / -<code><nobr>%exitrep</nobr></code> / <code><nobr>%endrep</nobr></code>. -<li>Added the __FILE__ and __LINE__ standard macros. -<li>Added a sanity check for number constants being greater than -0xFFFFFFFF. The warning can be disabled. -<li>Added the %0 token whereby a variadic multi-line macro can tell how -many parameters it's been given in a specific invocation. -<li>Added <code><nobr>%rotate</nobr></code>, allowing multi-line macro -parameters to be cycled. -<li>Added the `*' option for the maximum parameter count on multi-line -macros, allowing them to take arbitrarily many parameters. -<li>Added the ability for the user-level forms of EXTERN, GLOBAL and COMMON -to take more than one argument. -<li>Added the IMPORT and EXPORT directives in OBJ format, to deal with -Windows DLLs. -<li>Added some more preprocessor <code><nobr>%if</nobr></code> constructs: -<code><nobr>%ifidn</nobr></code> / <code><nobr>%ifidni</nobr></code> (exact -textual identity), and <code><nobr>%ifid</nobr></code> / -<code><nobr>%ifnum</nobr></code> / <code><nobr>%ifstr</nobr></code> (token -type testing). -<li>Added the ability to distinguish SHL AX,1 (the 8086 version) from SHL -AX,BYTE 1 (the 286-and-upwards version whose constant happens to be 1). -<li>Added NetBSD/FreeBSD/OpenBSD's variant of a.out format, complete with -PIC shared library features. -<li>Changed NASM's idiosyncratic handling of FCLEX, FDISI, FENI, FINIT, -FSAVE, FSTCW, FSTENV, and FSTSW to bring it into line with the otherwise -accepted standard. The previous behaviour, though it was a deliberate -feature, was a deliberate feature based on a misunderstanding. Apologies -for the inconvenience. -<li>Improved the flexibility of ABSOLUTE: you can now give it an expression -rather than being restricted to a constant, and it can take relocatable -arguments as well. -<li>Added the ability for a variable to be declared as EXTERN multiple -times, and the subsequent definitions are just ignored. -<li>We now allow instruction prefixes (CS, DS, LOCK, REPZ etc) to be alone -on a line (without a following instruction). -<li>Improved sanity checks on whether the arguments to EXTERN, GLOBAL and -COMMON are valid identifiers. -<li>Added misc/exebin.mac to allow direct generation of .EXE files by -hacking up an EXE header using DB and DW; also added test/binexe.asm to -demonstrate the use of this. Thanks to Yann Guidon for contributing the EXE -header code. -<li>ndisasm forgot to check whether the input file had been successfully -opened. Now it does. Doh! -<li>Added the Cyrix extensions to the MMX instruction set. -<li>Added a hinting mechanism to allow [EAX+EBX] and [EBX+EAX] to be -assembled differently. This is important since [ESI+EBP] and [EBP+ESI] have -different default base segment registers. -<li>Added support for the PharLap OMF extension for 4096-byte segment -alignment. -</ul> -<h4><a name="section-C.3.3">C.3.3 Version 0.95 released July 1997</a></h4> -<ul> -<li>Fixed yet another ELF bug. This one manifested if the user relied on -the default segment, and attempted to define global symbols without first -explicitly declaring the target segment. -<li>Added makefiles (for NASM and the RDF tools) to build Win32 console -apps under Symantec C++. Donated by Mark Junker. -<li>Added `macros.bas' and `insns.bas', QBasic versions of the Perl scripts -that convert `standard.mac' to `macros.c' and convert `insns.dat' to -`insnsa.c' and `insnsd.c'. Also thanks to Mark Junker. -<li>Changed the diassembled forms of the conditional instructions so that -JB is now emitted as JC, and other similar changes. Suggested list by -Ulrich Doewich. -<li>Added `@' to the list of valid characters to begin an identifier with. -<li>Documentary changes, notably the addition of the `Common Problems' -section in nasm.doc. -<li>Fixed a bug relating to 32-bit PC-relative fixups in OBJ. -<li>Fixed a bug in perm_copy() in labels.c which was causing exceptions in -cleanup_labels() on some systems. -<li>Positivity sanity check in TIMES argument changed from a warning to an -error following a further complaint. -<li>Changed the acceptable limits on byte and word operands to allow things -like `~10111001b' to work. -<li>Fixed a major problem in the preprocessor which caused seg-faults if -macro definitions contained blank lines or comment-only lines. -<li>Fixed inadequate error checking on the commas separating the arguments -to `db', `dw' etc. -<li>Fixed a crippling bug in the handling of macros with operand counts -defined with a `+' modifier. -<li>Fixed a bug whereby object file formats which stored the input file -name in the output file (such as OBJ and COFF) weren't doing so correctly -when the output file name was specified on the command line. -<li>Removed [INC] and [INCLUDE] support for good, since they were obsolete -anyway. -<li>Fixed a bug in OBJ which caused all fixups to be output in 16-bit -(old-format) FIXUPP records, rather than putting the 32-bit ones in -FIXUPP32 (new-format) records. -<li>Added, tentatively, OS/2 object file support (as a minor variant on -OBJ). -<li>Updates to Fox Cutter's Borland C makefile, Makefile.bc2. -<li>Removed a spurious second fclose() on the output file. -<li>Added the `-s' command line option to redirect all messages which would -go to stderr (errors, help text) to stdout instead. -<li>Added the `-w' command line option to selectively suppress some classes -of assembly warning messages. -<li>Added the `-p' pre-include and `-d' pre-define command-line options. -<li>Added an include file search path: the `-i' command line option. -<li>Fixed a silly little preprocessor bug whereby starting a line with a -`%!' environment-variable reference caused an `unknown directive' error. -<li>Added the long-awaited listing file support: the `-l' command line -option. -<li>Fixed a problem with OBJ format whereby, in the absence of any explicit -segment definition, non-global symbols declared in the implicit default -segment generated spurious EXTDEF records in the output. -<li>Added the NASM environment variable. -<li>From this version forward, Win32 console-mode binaries will be included -in the DOS distribution in addition to the 16-bit binaries. Added -Makefile.vc for this purpose. -<li>Added `return 0;' to test/objlink.c to prevent compiler warnings. -<li>Added the __NASM_MAJOR__ and __NASM_MINOR__ standard defines. -<li>Added an alternative memory-reference syntax in which prefixing an -operand with `&' is equivalent to enclosing it in square brackets, at -the request of Fox Cutter. -<li>Errors in pass two now cause the program to return a non-zero error -code, which they didn't before. -<li>Fixed the single-line macro cycle detection, which didn't work at all -on macros with no parameters (caused an infinite loop). Also changed the -behaviour of single-line macro cycle detection to work like cpp, so that -macros like `extrn' as given in the documentation can be implemented. -<li>Fixed the implementation of WRT, which was too restrictive in that you -couldn't do `mov ax,[di+abc wrt dgroup]' because (di+abc) wasn't a -relocatable reference. -</ul> -<h4><a name="section-C.3.4">C.3.4 Version 0.94 released April 1997</a></h4> -<ul> -<li>Major item: added the macro processor. -<li>Added undocumented instructions SMI, IBTS, XBTS and LOADALL286. Also -reorganised CMPXCHG instruction into early-486 and Pentium forms. Thanks to -Thobias Jones for the information. -<li>Fixed two more stupid bugs in ELF, which were causing `ld' to continue -to seg-fault in a lot of non-trivial cases. -<li>Fixed a seg-fault in the label manager. -<li>Stopped FBLD and FBSTP from _requiring_ the TWORD keyword, which is the -only option for BCD loads/stores in any case. -<li>Ensured FLDCW, FSTCW and FSTSW can cope with the WORD keyword, if -anyone bothers to provide it. Previously they complained unless no keyword -at all was present. -<li>Some forms of FDIV/FDIVR and FSUB/FSUBR were still inverted: a vestige -of a bug that I thought had been fixed in 0.92. This was fixed, hopefully -for good this time... -<li>Another minor phase error (insofar as a phase error can _ever_ be -minor) fixed, this one occurring in code of the form -</ul> -<p><pre> - rol ax,forward_reference - forward_reference equ 1 -</pre> -<ul> -<li>The number supplied to TIMES is now sanity-checked for positivity, and -also may be greater than 64K (which previously didn't work on 16-bit -systems). -<li>Added Watcom C makefiles, and misc/pmw.bat, donated by Dominik Behr. -<li>Added the INCBIN pseudo-opcode. -<li>Due to the advent of the preprocessor, the [INCLUDE] and [INC] -directives have become obsolete. They are still supported in this version, -with a warning, but won't be in the next. -<li>Fixed a bug in OBJ format, which caused incorrect object records to be -output when absolute labels were made global. -<li>Updates to RDOFF subdirectory, and changes to outrdf.c. -</ul> -<h4><a name="section-C.3.5">C.3.5 Version 0.93 released January 1997</a></h4> -<p>This release went out in a great hurry after semi-crippling bugs were -found in 0.92. -<ul> -<li>Really <em>did</em> fix the stack overflows this time. *blush* -<li>Had problems with EA instruction sizes changing between passes, when an -offset contained a forward reference and so 4 bytes were allocated for the -offset in pass one; by pass two the symbol had been defined and happened to -be a small absolute value, so only 1 byte got allocated, causing -instruction size mismatch between passes and hence incorrect address -calculations. Fixed. -<li>Stupid bug in the revised ELF section generation fixed (associated -string-table section for .symtab was hard-coded as 7, even when this didn't -fit with the real section table). Was causing `ld' to seg-fault under -Linux. -<li>Included a new Borland C makefile, Makefile.bc2, donated by Fox Cutter -<lmb@comtch.iea.com>. -</ul> -<h4><a name="section-C.3.6">C.3.6 Version 0.92 released January 1997</a></h4> -<ul> -<li>The FDIVP/FDIVRP and FSUBP/FSUBRP pairs had been inverted: this was -fixed. This also affected the LCC driver. -<li>Fixed a bug regarding 32-bit effective addresses of the form -<code><nobr>[other_register+ESP]</nobr></code>. -<li>Documentary changes, notably documentation of the fact that Borland -Win32 compilers use `obj' rather than `win32' object format. -<li>Fixed the COMENT record in OBJ files, which was formatted incorrectly. -<li>Fixed a bug causing segfaults in large RDF files. -<li>OBJ format now strips initial periods from segment and group -definitions, in order to avoid complications with the local label syntax. -<li>Fixed a bug in disassembling far calls and jumps in NDISASM. -<li>Added support for user-defined sections in COFF and ELF files. -<li>Compiled the DOS binaries with a sensible amount of stack, to prevent -stack overflows on any arithmetic expression containing parentheses. -<li>Fixed a bug in handling of files that do not terminate in a newline. -</ul> -<h4><a name="section-C.3.7">C.3.7 Version 0.91 released November 1996</a></h4> -<ul> -<li>Loads of bug fixes. -<li>Support for RDF added. -<li>Support for DBG debugging format added. -<li>Support for 32-bit extensions to Microsoft OBJ format added. -<li>Revised for Borland C: some variable names changed, makefile added. -<li>LCC support revised to actually work. -<li>JMP/CALL NEAR/FAR notation added. -<li>`a16', `o16', `a32' and `o32' prefixes added. -<li>Range checking on short jumps implemented. -<li>MMX instruction support added. -<li>Negative floating point constant support added. -<li>Memory handling improved to bypass 64K barrier under DOS. -<li><code><nobr>$</nobr></code> prefix to force treatment of reserved words -as identifiers added. -<li>Default-size mechanism for object formats added. -<li>Compile-time configurability added. -<li><code><nobr>#</nobr></code>, <code><nobr>@</nobr></code>, -<code><nobr>~</nobr></code> and c{?} are now valid characters in labels. -<li><code><nobr>-e</nobr></code> and <code><nobr>-k</nobr></code> options -in NDISASM added. -</ul> -<h4><a name="section-C.3.8">C.3.8 Version 0.90 released October 1996</a></h4> -<p>First release version. First support for object file output. Other -changes from previous version (0.3x) too numerous to document. -<p align=center><a href="nasmdocb.html">Previous Chapter</a> | -<a href="nasmdoc0.html">Contents</a> | -<a href="nasmdoci.html">Index</a> -</body></html> diff --git a/doc/html/nasmdoci.html b/doc/html/nasmdoci.html deleted file mode 100644 index 54eb6a6..0000000 --- a/doc/html/nasmdoci.html +++ /dev/null @@ -1,2345 +0,0 @@ -<html><head><title>NASM Manual</title></head> -<body><h1 align=center>The Netwide Assembler: NASM</h1> - -<p align=center><a href="nasmdoc0.html">Contents</a> -<p><code><nobr>!</nobr></code> operator, unary: -<a href="nasmdoc3.html#section-3.5.7">Section 3.5.7</a> -<br> -<code><nobr>!=</nobr></code> operator: -<a href="nasmdoc4.html#section-4.4.4">Section 4.4.4</a> -<br> -<code><nobr>$$</nobr></code> token: -<a href="nasmdoc3.html#section-3.5">Section 3.5</a>, -<a href="nasmdoc7.html#section-7.9.3">Section 7.9.3</a> -<br> -<code><nobr>$</nobr></code>, Here token: -<a href="nasmdoc3.html#section-3.5">Section 3.5</a> -<br> -<code><nobr>$</nobr></code>, prefix: -<a href="nasmdoc3.html#section-3.1">Section 3.1</a>, -<a href="nasmdoc3.html#section-3.4.1">Section 3.4.1</a>, -<a href="nasmdoc7.html#section-7.13.2">Section 7.13.2</a> -<br> -<code><nobr>%</nobr></code> operator: -<a href="nasmdoc3.html#section-3.5.6">Section 3.5.6</a> -<br> -<code><nobr>%!</nobr></code>: -<a href="nasmdoc4.html#section-4.10.2">Section 4.10.2</a> -<br> -<code><nobr>%$</nobr></code> and <code><nobr>%$$</nobr></code> prefixes: -<a href="nasmdoc4.html#section-4.7.2">Section 4.7.2</a> -<br> -<code><nobr>%%</nobr></code> operator: -<a href="nasmdoc3.html#section-3.5.6">Section 3.5.6</a>, -<a href="nasmdoc4.html#section-4.3.3">Section 4.3.3</a> -<br> -<code><nobr>%+</nobr></code>: -<a href="nasmdoc4.html#section-4.1.4">Section 4.1.4</a> -<br> -<code><nobr>%?</nobr></code>: -<a href="nasmdoc4.html#section-4.1.5">Section 4.1.5</a> -<br> -<code><nobr>%??</nobr></code>: -<a href="nasmdoc4.html#section-4.1.5">Section 4.1.5</a> -<br> -<code><nobr>%[</nobr></code>: -<a href="nasmdoc4.html#section-4.1.3">Section 4.1.3</a> -<br> -<code><nobr>&</nobr></code> operator: -<a href="nasmdoc3.html#section-3.5.3">Section 3.5.3</a> -<br> -<code><nobr>&&</nobr></code> operator: -<a href="nasmdoc4.html#section-4.4.4">Section 4.4.4</a> -<br> -<code><nobr>*</nobr></code> operator: -<a href="nasmdoc3.html#section-3.5.6">Section 3.5.6</a> -<br> -<code><nobr>+</nobr></code> modifier: -<a href="nasmdoc4.html#section-4.3.4">Section 4.3.4</a> -<br> -<code><nobr>+</nobr></code> operator, binary: -<a href="nasmdoc3.html#section-3.5.5">Section 3.5.5</a> -<br> -<code><nobr>+</nobr></code> operator, unary: -<a href="nasmdoc3.html#section-3.5.7">Section 3.5.7</a> -<br> -<code><nobr>-</nobr></code> operator, binary: -<a href="nasmdoc3.html#section-3.5.5">Section 3.5.5</a> -<br> -<code><nobr>-</nobr></code> operator, unary: -<a href="nasmdoc3.html#section-3.5.7">Section 3.5.7</a> -<br> -<code><nobr>..@</nobr></code> symbol prefix: -<a href="nasmdoc3.html#section-3.9">Section 3.9</a>, -<a href="nasmdoc4.html#section-4.3.3">Section 4.3.3</a> -<br> -<code><nobr>/</nobr></code> operator: -<a href="nasmdoc3.html#section-3.5.6">Section 3.5.6</a> -<br> -<code><nobr>//</nobr></code> operator: -<a href="nasmdoc3.html#section-3.5.6">Section 3.5.6</a> -<br> -<code><nobr><</nobr></code> operator: -<a href="nasmdoc4.html#section-4.4.4">Section 4.4.4</a> -<br> -<code><nobr><<</nobr></code> operator: -<a href="nasmdoc3.html#section-3.5.4">Section 3.5.4</a> -<br> -<code><nobr><=</nobr></code> operator: -<a href="nasmdoc4.html#section-4.4.4">Section 4.4.4</a> -<br> -<code><nobr><></nobr></code> operator: -<a href="nasmdoc4.html#section-4.4.4">Section 4.4.4</a> -<br> -<code><nobr>=</nobr></code> operator: -<a href="nasmdoc4.html#section-4.4.4">Section 4.4.4</a> -<br> -<code><nobr>==</nobr></code> operator: -<a href="nasmdoc4.html#section-4.4.4">Section 4.4.4</a> -<br> -<code><nobr>></nobr></code> operator: -<a href="nasmdoc4.html#section-4.4.4">Section 4.4.4</a> -<br> -<code><nobr>>=</nobr></code> operator: -<a href="nasmdoc4.html#section-4.4.4">Section 4.4.4</a> -<br> -<code><nobr>>></nobr></code> operator: -<a href="nasmdoc3.html#section-3.5.4">Section 3.5.4</a> -<br> -<code><nobr>?</nobr></code> MASM syntax: -<a href="nasmdoc3.html#section-3.2.2">Section 3.2.2</a> -<br> -<code><nobr>^</nobr></code> operator: -<a href="nasmdoc3.html#section-3.5.2">Section 3.5.2</a> -<br> -<code><nobr>^^</nobr></code> operator: -<a href="nasmdoc4.html#section-4.4.4">Section 4.4.4</a> -<br> -<code><nobr>|</nobr></code> operator: -<a href="nasmdoc3.html#section-3.5.1">Section 3.5.1</a> -<br> -<code><nobr>||</nobr></code> operator: -<a href="nasmdoc4.html#section-4.4.4">Section 4.4.4</a> -<br> -<code><nobr>~</nobr></code> operator: -<a href="nasmdoc3.html#section-3.5.7">Section 3.5.7</a> -<br> -<code><nobr>%0</nobr></code> parameter count: -<a href="nasmdoc4.html#section-4.3.5">Section 4.3.5</a>, -<a href="nasmdoc4.html#section-4.3.6">Section 4.3.6</a> -<br> -<code><nobr>%+1</nobr></code> and <code><nobr>%-1</nobr></code> syntax: -<a href="nasmdoc4.html#section-4.3.9">Section 4.3.9</a> -<br> -16-bit mode, versus 32-bit mode: -<a href="nasmdoc6.html#section-6.1">Section 6.1</a> -<br> -64-bit displacement: <a href="nasmdo11.html#section-11.2">Section 11.2</a> -<br> -64-bit immediate: <a href="nasmdo11.html#section-11.2">Section 11.2</a> -<br> -<code><nobr>-a</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.21">Section 2.1.21</a>, -<a href="nasmdoca.html#section-A.3.3">Section A.3.3</a> -<br> -<code><nobr>A16</nobr></code>: -<a href="nasmdoc3.html#section-3.1">Section 3.1</a> -<br> -<code><nobr>a16</nobr></code>: -<a href="nasmdo10.html#section-10.3">Section 10.3</a> -<br> -<code><nobr>A32</nobr></code>: -<a href="nasmdoc3.html#section-3.1">Section 3.1</a> -<br> -<code><nobr>a32</nobr></code>: -<a href="nasmdo10.html#section-10.3">Section 10.3</a> -<br> -<code><nobr>A64</nobr></code>: -<a href="nasmdoc3.html#section-3.1">Section 3.1</a> -<br> -<code><nobr>a64</nobr></code>: -<a href="nasmdo10.html#section-10.3">Section 10.3</a> -<br> -<code><nobr>a86</nobr></code>: -<a href="nasmdoc1.html#section-1.1.1">Section 1.1.1</a>, -<a href="nasmdoc2.html#section-2.2">Section 2.2</a>, -<a href="nasmdoc2.html#section-2.2.2">Section 2.2.2</a>, -<a href="nasmdoc2.html#section-2.2.6">Section 2.2.6</a> -<br> -<code><nobr>ABS</nobr></code>: -<a href="nasmdoc3.html#section-3.3">Section 3.3</a> -<br> -<code><nobr>ABSOLUTE</nobr></code>: -<a href="nasmdoc6.html#section-6.4">Section 6.4</a>, -<a href="nasmdoc7.html#section-7.4.1">Section 7.4.1</a> -<br> -addition: <a href="nasmdoc3.html#section-3.5.5">Section 3.5.5</a> -<br> -addressing, mixed-size: -<a href="nasmdo10.html#section-10.2">Section 10.2</a> -<br> -address-size prefixes: <a href="nasmdoc3.html#section-3.1">Section 3.1</a> -<br> -algebra: <a href="nasmdoc3.html#section-3.3">Section 3.3</a> -<br> -<code><nobr>ALIGN</nobr></code>: -<a href="nasmdoc4.html#section-4.11.12">Section 4.11.12</a>, -<a href="nasmdoc5.html#section-5.2">Section 5.2</a>, -<a href="nasmdoc7.html#section-7.1.2">Section 7.1.2</a>, -<a href="nasmdoc7.html#section-7.4.1">Section 7.4.1</a> -<br> -<code><nobr>ALIGN</nobr></code>, smart: -<a href="nasmdoc5.html#section-5.2">Section 5.2</a> -<br> -<code><nobr>ALIGNB</nobr></code>: -<a href="nasmdoc4.html#section-4.11.12">Section 4.11.12</a> -<br> -alignment, in <code><nobr>bin</nobr></code> sections: -<a href="nasmdoc7.html#section-7.1.2">Section 7.1.2</a> -<br> -alignment, in <code><nobr>elf</nobr></code> sections: -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a> -<br> -alignment, in <code><nobr>obj</nobr></code> sections: -<a href="nasmdoc7.html#section-7.4.1">Section 7.4.1</a> -<br> -alignment, in <code><nobr>win32</nobr></code> sections: -<a href="nasmdoc7.html#section-7.5.1">Section 7.5.1</a> -<br> -alignment, of <code><nobr>elf</nobr></code> common variables: -<a href="nasmdoc7.html#section-7.9.6">Section 7.9.6</a> -<br> -<code><nobr>ALIGNMODE</nobr></code>: -<a href="nasmdoc5.html#section-5.2">Section 5.2</a> -<br> -<code><nobr>__ALIGNMODE__</nobr></code>: -<a href="nasmdoc5.html#section-5.2">Section 5.2</a> -<br> -<code><nobr>ALINK</nobr></code>: -<a href="nasmdoc8.html#section-8.1.1">Section 8.1.1</a> -<br> -<code><nobr>alink.sourceforge.net</nobr></code>: -<a href="nasmdoc8.html#section-8.1.1">Section 8.1.1</a> -<br> -<code><nobr>all</nobr></code>: -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a> -<br> -<code><nobr>alloc</nobr></code>: -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a> -<br> -alternate register names: -<a href="nasmdoc5.html#section-5.1">Section 5.1</a> -<br> -<code><nobr>alt.lang.asm</nobr></code>: -<a href="nasmdoc1.html#section-1.1.1">Section 1.1.1</a> -<br> -<code><nobr>altreg</nobr></code>: -<a href="nasmdoc5.html#section-5.1">Section 5.1</a> -<br> -ambiguity: <a href="nasmdoc2.html#section-2.2.3">Section 2.2.3</a> -<br> -<code><nobr>a.out</nobr></code>, BSD version: -<a href="nasmdoc7.html#section-7.11">Section 7.11</a> -<br> -<code><nobr>a.out</nobr></code>, Linux version: -<a href="nasmdoc7.html#section-7.10">Section 7.10</a> -<br> -<code><nobr>aout</nobr></code>: -<a href="nasmdoc7.html#section-7.10">Section 7.10</a> -<br> -<code><nobr>aoutb</nobr></code>: -<a href="nasmdoc7.html#section-7.11">Section 7.11</a>, -<a href="nasmdoc9.html#section-9.2">Section 9.2</a> -<br> -<code><nobr>%arg</nobr></code>: -<a href="nasmdoc4.html#section-4.8.1">Section 4.8.1</a> -<br> -<code><nobr>arg</nobr></code>: -<a href="nasmdoc8.html#section-8.4.5">Section 8.4.5</a>, -<a href="nasmdoc9.html#section-9.1.4">Section 9.1.4</a> -<br> -<code><nobr>as86</nobr></code>: -<a href="nasmdoc1.html#section-1.1.1">Section 1.1.1</a>, -<a href="nasmdoc7.html#section-7.12">Section 7.12</a> -<br> -assembler directives: <a href="nasmdoc6.html">Chapter 6</a> -<br> -assembly-time options: -<a href="nasmdoc2.html#section-2.1.18">Section 2.1.18</a> -<br> -<code><nobr>%assign</nobr></code>: -<a href="nasmdoc4.html#section-4.1.7">Section 4.1.7</a> -<br> -<code><nobr>ASSUME</nobr></code>: -<a href="nasmdoc2.html#section-2.2.4">Section 2.2.4</a> -<br> -<code><nobr>AT</nobr></code>: -<a href="nasmdoc4.html#section-4.11.11">Section 4.11.11</a> -<br> -Autoconf: <a href="nasmdoc1.html#section-1.3.2">Section 1.3.2</a> -<br> -<code><nobr>autoexec.bat</nobr></code>: -<a href="nasmdoc1.html#section-1.3.1">Section 1.3.1</a> -<br> -<code><nobr>auto-sync</nobr></code>: -<a href="nasmdoca.html#section-A.3.3">Section A.3.3</a> -<br> -<code><nobr>-b</nobr></code>: -<a href="nasmdoca.html#section-A.3">Section A.3</a> -<br> -bin: <a href="nasmdoc2.html#section-2.1.2">Section 2.1.2</a>, -<a href="nasmdoc7.html#section-7.1">Section 7.1</a> -<br> -bin, multisection: <a href="nasmdoc7.html#section-7.1.3">Section 7.1.3</a> -<br> -binary: <a href="nasmdoc3.html#section-3.4.1">Section 3.4.1</a> -<br> -binary files: <a href="nasmdoc3.html#section-3.2.3">Section 3.2.3</a> -<br> -bit shift: <a href="nasmdoc3.html#section-3.5.4">Section 3.5.4</a> -<br> -<code><nobr>BITS</nobr></code>: -<a href="nasmdoc6.html#section-6.1">Section 6.1</a>, -<a href="nasmdoc7.html#section-7.1">Section 7.1</a> -<br> -<code><nobr>__BITS__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.5">Section 4.11.5</a> -<br> -bitwise AND: <a href="nasmdoc3.html#section-3.5.3">Section 3.5.3</a> -<br> -bitwise OR: <a href="nasmdoc3.html#section-3.5.1">Section 3.5.1</a> -<br> -bitwise XOR: <a href="nasmdoc3.html#section-3.5.2">Section 3.5.2</a> -<br> -block IFs: <a href="nasmdoc4.html#section-4.7.5">Section 4.7.5</a> -<br> -boot loader: <a href="nasmdoc7.html#section-7.1">Section 7.1</a> -<br> -boot sector: <a href="nasmdo12.html#section-12.1.3">Section 12.1.3</a> -<br> -Borland, Pascal: <a href="nasmdoc8.html#section-8.5">Section 8.5</a> -<br> -Borland, Win32 compilers: -<a href="nasmdoc7.html#section-7.4">Section 7.4</a> -<br> -braces, after <code><nobr>%</nobr></code> sign: -<a href="nasmdoc4.html#section-4.3.8">Section 4.3.8</a> -<br> -braces, around macro parameters: -<a href="nasmdoc4.html#section-4.3">Section 4.3</a> -<br> -BSD: <a href="nasmdoc9.html#section-9.2">Section 9.2</a> -<br> -<code><nobr>.bss</nobr></code>: -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a>, -<a href="nasmdoc7.html#section-7.10">Section 7.10</a>, -<a href="nasmdoc7.html#section-7.11">Section 7.11</a>, -<a href="nasmdoc7.html#section-7.12">Section 7.12</a>, -<a href="nasmdoc7.html#section-7.13">Section 7.13</a> -<br> -bugs: <a href="nasmdo12.html#section-12.2">Section 12.2</a> -<br> -<code><nobr>bugtracker</nobr></code>: -<a href="nasmdo12.html#section-12.2">Section 12.2</a> -<br> -<code><nobr>BYTE</nobr></code>: -<a href="nasmdo12.html#section-12.1.1">Section 12.1.1</a> -<br> -C calling convention: -<a href="nasmdoc8.html#section-8.4.3">Section 8.4.3</a>, -<a href="nasmdoc9.html#section-9.1.2">Section 9.1.2</a> -<br> -C symbol names: <a href="nasmdoc8.html#section-8.4.1">Section 8.4.1</a> -<br> -<code><nobr>c16.mac</nobr></code>: -<a href="nasmdoc8.html#section-8.4.5">Section 8.4.5</a>, -<a href="nasmdoc8.html#section-8.5.3">Section 8.5.3</a> -<br> -<code><nobr>c32.mac</nobr></code>: -<a href="nasmdoc9.html#section-9.1.4">Section 9.1.4</a> -<br> -<code><nobr>CALL FAR</nobr></code>: -<a href="nasmdoc3.html#section-3.6">Section 3.6</a> -<br> -case sensitivity: <a href="nasmdoc2.html#section-2.2.1">Section 2.2.1</a>, -<a href="nasmdoc4.html#section-4.1.1">Section 4.1.1</a>, -<a href="nasmdoc4.html#section-4.1.2">Section 4.1.2</a>, -<a href="nasmdoc4.html#section-4.1.7">Section 4.1.7</a>, -<a href="nasmdoc4.html#section-4.3">Section 4.3</a>, -<a href="nasmdoc4.html#section-4.3.1">Section 4.3.1</a>, -<a href="nasmdoc4.html#section-4.4.5">Section 4.4.5</a>, -<a href="nasmdoc7.html#section-7.4.3">Section 7.4.3</a> -<br> -changing sections: <a href="nasmdoc6.html#section-6.3">Section 6.3</a> -<br> -character constant: -<a href="nasmdoc3.html#section-3.2.1">Section 3.2.1</a>, -<a href="nasmdoc3.html#section-3.4.3">Section 3.4.3</a> -<br> -character strings: <a href="nasmdoc3.html#section-3.4.2">Section 3.4.2</a> -<br> -circular references: -<a href="nasmdoc4.html#section-4.1.1">Section 4.1.1</a> -<br> -<code><nobr>CLASS</nobr></code>: -<a href="nasmdoc7.html#section-7.4.1">Section 7.4.1</a> -<br> -<code><nobr>%clear</nobr></code>: -<a href="nasmdoc4.html#section-4.11">Section 4.11</a> -<br> -<code><nobr>coff</nobr></code>: -<a href="nasmdoc7.html#section-7.7">Section 7.7</a> -<br> -colon: <a href="nasmdoc3.html#section-3.1">Section 3.1</a> -<br> -<code><nobr>.COM</nobr></code>: -<a href="nasmdoc7.html#section-7.1">Section 7.1</a>, -<a href="nasmdoc8.html#section-8.2">Section 8.2</a> -<br> -command-line: <a href="nasmdoc2.html#section-2.1">Section 2.1</a>, -<a href="nasmdoc7.html">Chapter 7</a> -<br> -commas in macro parameters: -<a href="nasmdoc4.html#section-4.3.4">Section 4.3.4</a> -<br> -<code><nobr>.comment</nobr></code>: -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a> -<br> -<code><nobr>COMMON</nobr></code>: -<a href="nasmdoc6.html#section-6.7">Section 6.7</a>, -<a href="nasmdoc7.html#section-7.4.1">Section 7.4.1</a> -<br> -<code><nobr>COMMON</nobr></code>, <code><nobr>elf</nobr></code> extensions -to: <a href="nasmdoc7.html#section-7.9.6">Section 7.9.6</a> -<br> -<code><nobr>COMMON</nobr></code>, <code><nobr>obj</nobr></code> extensions -to: <a href="nasmdoc7.html#section-7.4.8">Section 7.4.8</a> -<br> -Common Object File Format: -<a href="nasmdoc7.html#section-7.7">Section 7.7</a> -<br> -common variables: <a href="nasmdoc6.html#section-6.7">Section 6.7</a> -<br> -common variables, alignment in <code><nobr>elf</nobr></code>: -<a href="nasmdoc7.html#section-7.9.6">Section 7.9.6</a> -<br> -common variables, element size: -<a href="nasmdoc7.html#section-7.4.8">Section 7.4.8</a> -<br> -<code><nobr>comp.lang.asm.x86</nobr></code>: -<a href="nasmdoc1.html#section-1.1.1">Section 1.1.1</a>, -<a href="nasmdoc1.html#section-1.2">Section 1.2</a> -<br> -<code><nobr>comp.os.msdos.programmer</nobr></code>: -<a href="nasmdoc8.html#section-8.3">Section 8.3</a> -<br> -concatenating macro parameters: -<a href="nasmdoc4.html#section-4.3.8">Section 4.3.8</a> -<br> -concatenating strings: -<a href="nasmdoc4.html#section-4.2.1">Section 4.2.1</a> -<br> -condition codes as macro parameters: -<a href="nasmdoc4.html#section-4.3.9">Section 4.3.9</a> -<br> -conditional assembly: <a href="nasmdoc4.html#section-4.4">Section 4.4</a> -<br> -conditional jumps: -<a href="nasmdo12.html#section-12.1.2">Section 12.1.2</a> -<br> -conditional-return macro: -<a href="nasmdoc4.html#section-4.3.9">Section 4.3.9</a> -<br> -<code><nobr>configure</nobr></code>: -<a href="nasmdoc1.html#section-1.3.2">Section 1.3.2</a> -<br> -constants: <a href="nasmdoc3.html#section-3.4">Section 3.4</a> -<br> -context stack: <a href="nasmdoc4.html#section-4.7">Section 4.7</a>, -<a href="nasmdoc4.html#section-4.7.5">Section 4.7.5</a> -<br> -context-local labels: -<a href="nasmdoc4.html#section-4.7.2">Section 4.7.2</a> -<br> -context-local single-line macros: -<a href="nasmdoc4.html#section-4.7.3">Section 4.7.3</a> -<br> -counting macro parameters: -<a href="nasmdoc4.html#section-4.3.6">Section 4.3.6</a> -<br> -<code><nobr>CPU</nobr></code>: -<a href="nasmdoc6.html#section-6.8">Section 6.8</a> -<br> -<code><nobr>CPUID</nobr></code>: -<a href="nasmdoc3.html#section-3.4.3">Section 3.4.3</a> -<br> -creating contexts: <a href="nasmdoc4.html#section-4.7.1">Section 4.7.1</a> -<br> -critical expression: -<a href="nasmdoc3.html#section-3.2.2">Section 3.2.2</a>, -<a href="nasmdoc3.html#section-3.8">Section 3.8</a>, -<a href="nasmdoc4.html#section-4.1.7">Section 4.1.7</a>, -<a href="nasmdoc6.html#section-6.4">Section 6.4</a> -<br> -<code><nobr>-D</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.18">Section 2.1.18</a> -<br> -<code><nobr>-d</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.18">Section 2.1.18</a> -<br> -daily development snapshots: -<a href="nasmdoc1.html#section-1.2">Section 1.2</a> -<br> -<code><nobr>.data</nobr></code>: -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a>, -<a href="nasmdoc7.html#section-7.10">Section 7.10</a>, -<a href="nasmdoc7.html#section-7.11">Section 7.11</a>, -<a href="nasmdoc7.html#section-7.12">Section 7.12</a>, -<a href="nasmdoc7.html#section-7.13">Section 7.13</a> -<br> -<code><nobr>_DATA</nobr></code>: -<a href="nasmdoc8.html#section-8.4.2">Section 8.4.2</a> -<br> -<code><nobr>data</nobr></code>: -<a href="nasmdoc7.html#section-7.9.5">Section 7.9.5</a>, -<a href="nasmdoc7.html#section-7.13.3">Section 7.13.3</a> -<br> -data structure: <a href="nasmdoc8.html#section-8.4.4">Section 8.4.4</a>, -<a href="nasmdoc9.html#section-9.1.3">Section 9.1.3</a> -<br> -<code><nobr>__DATE__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.7">Section 4.11.7</a> -<br> -<code><nobr>__DATE_NUM__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.7">Section 4.11.7</a> -<br> -<code><nobr>DB</nobr></code>: -<a href="nasmdoc3.html#section-3.2">Section 3.2</a>, -<a href="nasmdoc3.html#section-3.2.1">Section 3.2.1</a>, -<a href="nasmdoc3.html#section-3.4.4">Section 3.4.4</a>, -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -<code><nobr>dbg</nobr></code>: -<a href="nasmdoc7.html#section-7.14">Section 7.14</a> -<br> -<code><nobr>DD</nobr></code>: -<a href="nasmdoc3.html#section-3.2">Section 3.2</a>, -<a href="nasmdoc3.html#section-3.2.1">Section 3.2.1</a>, -<a href="nasmdoc3.html#section-3.4.4">Section 3.4.4</a>, -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -debug information: -<a href="nasmdoc2.html#section-2.1.12">Section 2.1.12</a> -<br> -debug information format: -<a href="nasmdoc2.html#section-2.1.11">Section 2.1.11</a> -<br> -declaring structures: -<a href="nasmdoc4.html#section-4.11.10">Section 4.11.10</a> -<br> -<code><nobr>DEFAULT</nobr></code>: -<a href="nasmdoc6.html#section-6.2">Section 6.2</a> -<br> -<code><nobr>default</nobr></code>: -<a href="nasmdoc7.html#section-7.9.5">Section 7.9.5</a> -<br> -default macro parameters: -<a href="nasmdoc4.html#section-4.3.5">Section 4.3.5</a> -<br> -default name: <a href="nasmdoc7.html">Chapter 7</a> -<br> -default-<code><nobr>WRT</nobr></code> mechanism: -<a href="nasmdoc7.html#section-7.4.7">Section 7.4.7</a> -<br> -<code><nobr>%define</nobr></code>: -<a href="nasmdoc2.html#section-2.1.18">Section 2.1.18</a>, -<a href="nasmdoc4.html#section-4.1.1">Section 4.1.1</a> -<br> -defining sections: <a href="nasmdoc6.html#section-6.3">Section 6.3</a> -<br> -<code><nobr>%defstr</nobr></code>: -<a href="nasmdoc4.html#section-4.1.8">Section 4.1.8</a> -<br> -<code><nobr>%deftok</nobr></code>: -<a href="nasmdoc4.html#section-4.1.9">Section 4.1.9</a> -<br> -<code><nobr>%depend</nobr></code>: -<a href="nasmdoc4.html#section-4.6.3">Section 4.6.3</a> -<br> -design goals: <a href="nasmdoc2.html#section-2.2.2">Section 2.2.2</a> -<br> -DevPac: <a href="nasmdoc3.html#section-3.2.3">Section 3.2.3</a>, -<a href="nasmdoc3.html#section-3.9">Section 3.9</a> -<br> -disabling listing expansion: -<a href="nasmdoc4.html#section-4.3.10">Section 4.3.10</a> -<br> -division: <a href="nasmdoc3.html#section-3.5.6">Section 3.5.6</a> -<br> -DJGPP: <a href="nasmdoc7.html#section-7.7">Section 7.7</a>, -<a href="nasmdoc9.html">Chapter 9</a> -<br> -<code><nobr>djlink</nobr></code>: -<a href="nasmdoc8.html#section-8.1.1">Section 8.1.1</a> -<br> -DLL symbols, exporting: -<a href="nasmdoc7.html#section-7.4.5">Section 7.4.5</a> -<br> -DLL symbols, importing: -<a href="nasmdoc7.html#section-7.4.4">Section 7.4.4</a> -<br> -<code><nobr>DO</nobr></code>: -<a href="nasmdoc3.html#section-3.2">Section 3.2</a>, -<a href="nasmdoc3.html#section-3.2.1">Section 3.2.1</a>, -<a href="nasmdoc3.html#section-3.4.4">Section 3.4.4</a>, -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -DOS: <a href="nasmdoc1.html#section-1.3.1">Section 1.3.1</a>, -<a href="nasmdoc2.html#section-2.1.14">Section 2.1.14</a>, -<a href="nasmdoc2.html#section-2.1.15">Section 2.1.15</a> -<br> -DOS archive: -<br> -DOS source archive: <a href="nasmdoc1.html#section-1.3.1">Section 1.3.1</a> -<br> -<code><nobr>DQ</nobr></code>: -<a href="nasmdoc3.html#section-3.2">Section 3.2</a>, -<a href="nasmdoc3.html#section-3.2.1">Section 3.2.1</a>, -<a href="nasmdoc3.html#section-3.4.4">Section 3.4.4</a>, -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -<code><nobr>.drectve</nobr></code>: -<a href="nasmdoc7.html#section-7.5.1">Section 7.5.1</a> -<br> -<code><nobr>DT</nobr></code>: -<a href="nasmdoc3.html#section-3.2">Section 3.2</a>, -<a href="nasmdoc3.html#section-3.2.1">Section 3.2.1</a>, -<a href="nasmdoc3.html#section-3.4.4">Section 3.4.4</a>, -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -<code><nobr>DUP</nobr></code>: -<a href="nasmdoc2.html#section-2.2.7">Section 2.2.7</a>, -<a href="nasmdoc3.html#section-3.2.5">Section 3.2.5</a> -<br> -<code><nobr>DW</nobr></code>: -<a href="nasmdoc3.html#section-3.2">Section 3.2</a>, -<a href="nasmdoc3.html#section-3.2.1">Section 3.2.1</a>, -<a href="nasmdoc3.html#section-3.4.4">Section 3.4.4</a>, -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -<code><nobr>DWORD</nobr></code>: -<a href="nasmdoc3.html#section-3.1">Section 3.1</a> -<br> -<code><nobr>DY</nobr></code>: -<a href="nasmdoc3.html#section-3.2">Section 3.2</a>, -<a href="nasmdoc3.html#section-3.2.1">Section 3.2.1</a>, -<a href="nasmdoc3.html#section-3.4.4">Section 3.4.4</a> -<br> -<code><nobr>-E</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.20">Section 2.1.20</a> -<br> -<code><nobr>-e</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.20">Section 2.1.20</a>, -<a href="nasmdoca.html#section-A.3.4">Section A.3.4</a> -<br> -effective addresses: <a href="nasmdoc3.html#section-3.1">Section 3.1</a>, -<a href="nasmdoc3.html#section-3.3">Section 3.3</a> -<br> -element size, in common variables: -<a href="nasmdoc7.html#section-7.4.8">Section 7.4.8</a> -<br> -ELF: <a href="nasmdoc7.html#section-7.9">Section 7.9</a> -<br> -ELF, shared libraries: -<a href="nasmdoc7.html#section-7.9.3">Section 7.9.3</a> -<br> -ELF, 16-bit code and: -<a href="nasmdoc7.html#section-7.9.7">Section 7.9.7</a> -<br> -elf, debug formats and: -<a href="nasmdoc7.html#section-7.9.8">Section 7.9.8</a> -<br> -<code><nobr>elf32</nobr></code>: -<a href="nasmdoc7.html#section-7.9">Section 7.9</a> -<br> -<code><nobr>elf64</nobr></code>: -<a href="nasmdoc7.html#section-7.9">Section 7.9</a> -<br> -<code><nobr>%elif</nobr></code>: -<a href="nasmdoc4.html#section-4.4">Section 4.4</a>, -<a href="nasmdoc4.html#section-4.4.4">Section 4.4.4</a> -<br> -<code><nobr>%elifctx</nobr></code>: -<a href="nasmdoc4.html#section-4.4.3">Section 4.4.3</a> -<br> -<code><nobr>%elifdef</nobr></code>: -<a href="nasmdoc4.html#section-4.4.1">Section 4.4.1</a> -<br> -<code><nobr>%elifempty</nobr></code>: -<a href="nasmdoc4.html#section-4.4.8">Section 4.4.8</a> -<br> -<code><nobr>%elifid</nobr></code>: -<a href="nasmdoc4.html#section-4.4.6">Section 4.4.6</a> -<br> -<code><nobr>%elifidn</nobr></code>: -<a href="nasmdoc4.html#section-4.4.5">Section 4.4.5</a> -<br> -<code><nobr>%elifidni</nobr></code>: -<a href="nasmdoc4.html#section-4.4.5">Section 4.4.5</a> -<br> -<code><nobr>%elifmacro</nobr></code>: -<a href="nasmdoc4.html#section-4.4.2">Section 4.4.2</a> -<br> -<code><nobr>%elifn</nobr></code>: -<a href="nasmdoc4.html#section-4.4">Section 4.4</a>, -<a href="nasmdoc4.html#section-4.4.4">Section 4.4.4</a> -<br> -<code><nobr>%elifnctx</nobr></code>: -<a href="nasmdoc4.html#section-4.4.3">Section 4.4.3</a> -<br> -<code><nobr>%elifndef</nobr></code>: -<a href="nasmdoc4.html#section-4.4.1">Section 4.4.1</a> -<br> -<code><nobr>%elifnempty</nobr></code>: -<a href="nasmdoc4.html#section-4.4.8">Section 4.4.8</a> -<br> -<code><nobr>%elifnid</nobr></code>: -<a href="nasmdoc4.html#section-4.4.6">Section 4.4.6</a> -<br> -<code><nobr>%elifnidn</nobr></code>: -<a href="nasmdoc4.html#section-4.4.5">Section 4.4.5</a> -<br> -<code><nobr>%elifnidni</nobr></code>: -<a href="nasmdoc4.html#section-4.4.5">Section 4.4.5</a> -<br> -<code><nobr>%elifnmacro</nobr></code>: -<a href="nasmdoc4.html#section-4.4.2">Section 4.4.2</a> -<br> -<code><nobr>%elifnnum</nobr></code>: -<a href="nasmdoc4.html#section-4.4.6">Section 4.4.6</a> -<br> -<code><nobr>%elifnstr</nobr></code>: -<a href="nasmdoc4.html#section-4.4.6">Section 4.4.6</a> -<br> -<code><nobr>%elifntoken</nobr></code>: -<a href="nasmdoc4.html#section-4.4.7">Section 4.4.7</a> -<br> -<code><nobr>%elifnum</nobr></code>: -<a href="nasmdoc4.html#section-4.4.6">Section 4.4.6</a> -<br> -<code><nobr>%elifstr</nobr></code>: -<a href="nasmdoc4.html#section-4.4.6">Section 4.4.6</a> -<br> -<code><nobr>%eliftoken</nobr></code>: -<a href="nasmdoc4.html#section-4.4.7">Section 4.4.7</a> -<br> -<code><nobr>%else</nobr></code>: -<a href="nasmdoc4.html#section-4.4">Section 4.4</a> -<br> -<code><nobr>endproc</nobr></code>: -<a href="nasmdoc8.html#section-8.4.5">Section 8.4.5</a>, -<a href="nasmdoc9.html#section-9.1.4">Section 9.1.4</a> -<br> -<code><nobr>%endrep</nobr></code>: -<a href="nasmdoc4.html#section-4.5">Section 4.5</a> -<br> -<code><nobr>ENDSTRUC</nobr></code>: -<a href="nasmdoc4.html#section-4.11.10">Section 4.11.10</a>, -<a href="nasmdoc6.html#section-6.4">Section 6.4</a> -<br> -environment: <a href="nasmdoc2.html#section-2.1.28">Section 2.1.28</a> -<br> -<code><nobr>EQU</nobr></code>: -<a href="nasmdoc3.html#section-3.2">Section 3.2</a>, -<a href="nasmdoc3.html#section-3.2.4">Section 3.2.4</a> -<br> -<code><nobr>%error</nobr></code>: -<a href="nasmdoc4.html#section-4.9">Section 4.9</a> -<br> -<code><nobr>error</nobr></code>: -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a> -<br> -error messages: <a href="nasmdoc2.html#section-2.1.14">Section 2.1.14</a>, -<a href="nasmdoc2.html#section-2.1.15">Section 2.1.15</a> -<br> -error reporting format: -<a href="nasmdoc2.html#section-2.1.13">Section 2.1.13</a> -<br> -escape sequences: <a href="nasmdoc3.html#section-3.4.2">Section 3.4.2</a> -<br> -<code><nobr>EVEN</nobr></code>: -<a href="nasmdoc4.html#section-4.11.12">Section 4.11.12</a> -<br> -exact matches: <a href="nasmdoc4.html#section-4.3.11">Section 4.3.11</a> -<br> -<code><nobr>.EXE</nobr></code>: -<a href="nasmdoc7.html#section-7.4">Section 7.4</a>, -<a href="nasmdoc8.html#section-8.1">Section 8.1</a> -<br> -<code><nobr>EXE2BIN</nobr></code>: -<a href="nasmdoc8.html#section-8.2.2">Section 8.2.2</a> -<br> -<code><nobr>EXE_begin</nobr></code>: -<a href="nasmdoc8.html#section-8.1.2">Section 8.1.2</a> -<br> -<code><nobr>exebin.mac</nobr></code>: -<a href="nasmdoc8.html#section-8.1.2">Section 8.1.2</a> -<br> -<code><nobr>exec</nobr></code>: -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a> -<br> -Executable and Linkable Format: -<a href="nasmdoc7.html#section-7.9">Section 7.9</a> -<br> -<code><nobr>EXE_end</nobr></code>: -<a href="nasmdoc8.html#section-8.1.2">Section 8.1.2</a> -<br> -<code><nobr>EXE_stack</nobr></code>: -<a href="nasmdoc8.html#section-8.1.2">Section 8.1.2</a> -<br> -<code><nobr>%exitmacro</nobr></code>: -<a href="nasmdoc4.html#section-4.3.12">Section 4.3.12</a> -<br> -<code><nobr>%exitrep</nobr></code>: -<a href="nasmdoc4.html#section-4.5">Section 4.5</a> -<br> -<code><nobr>EXPORT</nobr></code>: -<a href="nasmdoc7.html#section-7.4.5">Section 7.4.5</a> -<br> -<code><nobr>export</nobr></code>: -<a href="nasmdoc7.html#section-7.13.3">Section 7.13.3</a> -<br> -exporting symbols: <a href="nasmdoc6.html#section-6.6">Section 6.6</a> -<br> -expressions: <a href="nasmdoc2.html#section-2.1.20">Section 2.1.20</a>, -<a href="nasmdoc3.html#section-3.5">Section 3.5</a> -<br> -extension: <a href="nasmdoc2.html#section-2.1.1">Section 2.1.1</a>, -<a href="nasmdoc7.html">Chapter 7</a> -<br> -<code><nobr>EXTERN</nobr></code>: -<a href="nasmdoc6.html#section-6.5">Section 6.5</a> -<br> -<code><nobr>EXTERN</nobr></code>, <code><nobr>obj</nobr></code> extensions -to: <a href="nasmdoc7.html#section-7.4.7">Section 7.4.7</a> -<br> -<code><nobr>EXTERN</nobr></code>, <code><nobr>rdf</nobr></code> extensions -to: <a href="nasmdoc7.html#section-7.13.4">Section 7.13.4</a> -<br> -extracting substrings: -<a href="nasmdoc4.html#section-4.2.3">Section 4.2.3</a> -<br> -<code><nobr>-F</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.11">Section 2.1.11</a> -<br> -<code><nobr>-f</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.2">Section 2.1.2</a>, -<a href="nasmdoc7.html">Chapter 7</a> -<br> -far call: <a href="nasmdoc2.html#section-2.2.5">Section 2.2.5</a> -<br> -far common variables: -<a href="nasmdoc7.html#section-7.4.8">Section 7.4.8</a> -<br> -far pointer: <a href="nasmdoc3.html#section-3.6">Section 3.6</a> -<br> -<code><nobr>FARCODE</nobr></code>: -<a href="nasmdoc8.html#section-8.4.5">Section 8.4.5</a>, -<a href="nasmdoc8.html#section-8.5.3">Section 8.5.3</a> -<br> -<code><nobr>%fatal</nobr></code>: -<a href="nasmdoc4.html#section-4.9">Section 4.9</a> -<br> -<code><nobr>__FILE__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.4">Section 4.11.4</a> -<br> -<code><nobr>FLAT</nobr></code>: -<a href="nasmdoc7.html#section-7.4.1">Section 7.4.1</a> -<br> -flat memory model: <a href="nasmdoc9.html">Chapter 9</a> -<br> -flat-form binary: <a href="nasmdoc7.html#section-7.1">Section 7.1</a> -<br> -<code><nobr>FLOAT</nobr></code>: -<a href="nasmdoc6.html#section-6.9">Section 6.9</a> -<br> -<code><nobr>__FLOAT__</nobr></code>: -<a href="nasmdoc6.html#section-6.9">Section 6.9</a> -<br> -<code><nobr>__float128h__</nobr></code>: -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -<code><nobr>__float128l__</nobr></code>: -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -<code><nobr>__float16__</nobr></code>: -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -<code><nobr>__float32__</nobr></code>: -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -<code><nobr>__float64__</nobr></code>: -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -<code><nobr>__float8__</nobr></code>: -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -<code><nobr>__float80e__</nobr></code>: -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -<code><nobr>__float80m__</nobr></code>: -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -<code><nobr>__FLOAT_DAZ__</nobr></code>: -<a href="nasmdoc6.html#section-6.9">Section 6.9</a> -<br> -<code><nobr>float-denorm</nobr></code>: -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a> -<br> -floating-point, constants: -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a>, -<a href="nasmdoc6.html#section-6.9">Section 6.9</a> -<br> -floating-point, packed BCD constants: -<a href="nasmdoc3.html#section-3.4.7">Section 3.4.7</a> -<br> -floating-point: <a href="nasmdoc2.html#section-2.2.6">Section 2.2.6</a>, -<a href="nasmdoc3.html#section-3.1">Section 3.1</a>, -<a href="nasmdoc3.html#section-3.2.1">Section 3.2.1</a>, -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -<code><nobr>float-overflow</nobr></code>: -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a> -<br> -<code><nobr>__FLOAT_ROUND__</nobr></code>: -<a href="nasmdoc6.html#section-6.9">Section 6.9</a> -<br> -<code><nobr>float-toolong</nobr></code>: -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a> -<br> -<code><nobr>float-underflow</nobr></code>: -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a> -<br> -<code><nobr>follows=</nobr></code>: -<a href="nasmdoc7.html#section-7.1.3">Section 7.1.3</a> -<br> -format-specific directives: <a href="nasmdoc6.html">Chapter 6</a> -<br> -frame pointer: <a href="nasmdoc8.html#section-8.4.3">Section 8.4.3</a>, -<a href="nasmdoc8.html#section-8.5.1">Section 8.5.1</a>, -<a href="nasmdoc9.html#section-9.1.2">Section 9.1.2</a> -<br> -FreeBSD: <a href="nasmdoc7.html#section-7.11">Section 7.11</a>, -<a href="nasmdoc9.html#section-9.2">Section 9.2</a> -<br> -FreeLink: <a href="nasmdoc8.html#section-8.1.1">Section 8.1.1</a> -<br> -<code><nobr>ftp.simtel.net</nobr></code>: -<a href="nasmdoc8.html#section-8.1.1">Section 8.1.1</a> -<br> -<code><nobr>function</nobr></code>: -<a href="nasmdoc7.html#section-7.9.5">Section 7.9.5</a>, -<a href="nasmdoc7.html#section-7.13.3">Section 7.13.3</a> -<br> -functions, C calling convention: -<a href="nasmdoc8.html#section-8.4.3">Section 8.4.3</a>, -<a href="nasmdoc9.html#section-9.1.2">Section 9.1.2</a> -<br> -functions, Pascal calling convention: -<a href="nasmdoc8.html#section-8.5.1">Section 8.5.1</a> -<br> -<code><nobr>-g</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.12">Section 2.1.12</a> -<br> -<code><nobr>gas</nobr></code>: -<a href="nasmdoc1.html#section-1.1.1">Section 1.1.1</a> -<br> -<code><nobr>gcc</nobr></code>: -<a href="nasmdoc1.html#section-1.1.1">Section 1.1.1</a> -<br> -<code><nobr>GLOBAL</nobr></code>: -<a href="nasmdoc6.html#section-6.6">Section 6.6</a> -<br> -<code><nobr>GLOBAL</nobr></code>, <code><nobr>aoutb</nobr></code> -extensions to: <a href="nasmdoc7.html#section-7.9.5">Section 7.9.5</a> -<br> -<code><nobr>GLOBAL</nobr></code>, <code><nobr>elf</nobr></code> extensions -to: <a href="nasmdoc7.html#section-7.9.5">Section 7.9.5</a> -<br> -<code><nobr>GLOBAL</nobr></code>, <code><nobr>rdf</nobr></code> extensions -to: <a href="nasmdoc7.html#section-7.13.3">Section 7.13.3</a> -<br> -global offset table: <a href="nasmdoc9.html#section-9.2">Section 9.2</a> -<br> -<code><nobr>_GLOBAL_OFFSET_TABLE_</nobr></code>: -<a href="nasmdoc7.html#section-7.9.3">Section 7.9.3</a> -<br> -<code><nobr>gnu-elf-extensions</nobr></code>: -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a> -<br> -<code><nobr>..got</nobr></code>: -<a href="nasmdoc7.html#section-7.9.3">Section 7.9.3</a> -<br> -<code><nobr>GOT</nobr></code> relocations: -<a href="nasmdoc9.html#section-9.2.3">Section 9.2.3</a> -<br> -GOT: <a href="nasmdoc7.html#section-7.9.3">Section 7.9.3</a>, -<a href="nasmdoc9.html#section-9.2">Section 9.2</a> -<br> -<code><nobr>..gotoff</nobr></code>: -<a href="nasmdoc7.html#section-7.9.3">Section 7.9.3</a> -<br> -<code><nobr>GOTOFF</nobr></code> relocations: -<a href="nasmdoc9.html#section-9.2.2">Section 9.2.2</a> -<br> -<code><nobr>..gotpc</nobr></code>: -<a href="nasmdoc7.html#section-7.9.3">Section 7.9.3</a> -<br> -<code><nobr>GOTPC</nobr></code> relocations: -<a href="nasmdoc9.html#section-9.2.1">Section 9.2.1</a> -<br> -<code><nobr>..gottpoff</nobr></code>: -<a href="nasmdoc7.html#section-7.9.4">Section 7.9.4</a> -<br> -graphics: <a href="nasmdoc3.html#section-3.2.3">Section 3.2.3</a> -<br> -greedy macro parameters: -<a href="nasmdoc4.html#section-4.3.4">Section 4.3.4</a> -<br> -<code><nobr>GROUP</nobr></code>: -<a href="nasmdoc7.html#section-7.4.2">Section 7.4.2</a> -<br> -groups: <a href="nasmdoc3.html#section-3.6">Section 3.6</a> -<br> -<code><nobr>-h</nobr></code>: -<a href="nasmdoca.html#section-A.3">Section A.3</a> -<br> -hexadecimal: <a href="nasmdoc3.html#section-3.4.1">Section 3.4.1</a> -<br> -<code><nobr>hidden</nobr></code>: -<a href="nasmdoc7.html#section-7.9.5">Section 7.9.5</a> -<br> -hybrid syntaxes: <a href="nasmdoc2.html#section-2.2.2">Section 2.2.2</a> -<br> -<code><nobr>-I</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.16">Section 2.1.16</a> -<br> -<code><nobr>-i</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.16">Section 2.1.16</a>, -<a href="nasmdoca.html#section-A.3.3">Section A.3.3</a> -<br> -<code><nobr>%iassign</nobr></code>: -<a href="nasmdoc4.html#section-4.1.7">Section 4.1.7</a> -<br> -<code><nobr>%idefine</nobr></code>: -<a href="nasmdoc4.html#section-4.1.1">Section 4.1.1</a> -<br> -<code><nobr>%idefstr</nobr></code>: -<a href="nasmdoc4.html#section-4.1.8">Section 4.1.8</a> -<br> -<code><nobr>%ideftok</nobr></code>: -<a href="nasmdoc4.html#section-4.1.9">Section 4.1.9</a> -<br> -<code><nobr>IEND</nobr></code>: -<a href="nasmdoc4.html#section-4.11.11">Section 4.11.11</a> -<br> -<code><nobr>%if</nobr></code>: -<a href="nasmdoc4.html#section-4.4">Section 4.4</a>, -<a href="nasmdoc4.html#section-4.4.4">Section 4.4.4</a> -<br> -<code><nobr>%ifctx</nobr></code>: -<a href="nasmdoc4.html#section-4.4.3">Section 4.4.3</a>, -<a href="nasmdoc4.html#section-4.7.5">Section 4.7.5</a> -<br> -<code><nobr>%ifdef</nobr></code>: -<a href="nasmdoc4.html#section-4.4.1">Section 4.4.1</a> -<br> -<code><nobr>%ifempty</nobr></code>: -<a href="nasmdoc4.html#section-4.4.8">Section 4.4.8</a> -<br> -<code><nobr>%ifid</nobr></code>: -<a href="nasmdoc4.html#section-4.4.6">Section 4.4.6</a> -<br> -<code><nobr>%ifidn</nobr></code>: -<a href="nasmdoc4.html#section-4.4.5">Section 4.4.5</a> -<br> -<code><nobr>%ifidni</nobr></code>: -<a href="nasmdoc4.html#section-4.4.5">Section 4.4.5</a> -<br> -<code><nobr>%ifmacro</nobr></code>: -<a href="nasmdoc4.html#section-4.4.2">Section 4.4.2</a> -<br> -<code><nobr>%ifn</nobr></code>: -<a href="nasmdoc4.html#section-4.4">Section 4.4</a>, -<a href="nasmdoc4.html#section-4.4.4">Section 4.4.4</a> -<br> -<code><nobr>%ifnctx</nobr></code>: -<a href="nasmdoc4.html#section-4.4.3">Section 4.4.3</a> -<br> -<code><nobr>%ifndef</nobr></code>: -<a href="nasmdoc4.html#section-4.4.1">Section 4.4.1</a> -<br> -<code><nobr>%ifnempty</nobr></code>: -<a href="nasmdoc4.html#section-4.4.8">Section 4.4.8</a> -<br> -<code><nobr>%ifnid</nobr></code>: -<a href="nasmdoc4.html#section-4.4.6">Section 4.4.6</a> -<br> -<code><nobr>%ifnidn</nobr></code>: -<a href="nasmdoc4.html#section-4.4.5">Section 4.4.5</a> -<br> -<code><nobr>%ifnidni</nobr></code>: -<a href="nasmdoc4.html#section-4.4.5">Section 4.4.5</a> -<br> -<code><nobr>%ifnmacro</nobr></code>: -<a href="nasmdoc4.html#section-4.4.2">Section 4.4.2</a> -<br> -<code><nobr>%ifnnum</nobr></code>: -<a href="nasmdoc4.html#section-4.4.6">Section 4.4.6</a> -<br> -<code><nobr>%ifnstr</nobr></code>: -<a href="nasmdoc4.html#section-4.4.6">Section 4.4.6</a> -<br> -<code><nobr>%ifntoken</nobr></code>: -<a href="nasmdoc4.html#section-4.4.7">Section 4.4.7</a> -<br> -<code><nobr>%ifnum</nobr></code>: -<a href="nasmdoc4.html#section-4.4.6">Section 4.4.6</a> -<br> -<code><nobr>%ifstr</nobr></code>: -<a href="nasmdoc4.html#section-4.4.6">Section 4.4.6</a> -<br> -<code><nobr>%iftoken</nobr></code>: -<a href="nasmdoc4.html#section-4.4.7">Section 4.4.7</a> -<br> -<code><nobr>%imacro</nobr></code>: -<a href="nasmdoc4.html#section-4.3">Section 4.3</a> -<br> -<code><nobr>IMPORT</nobr></code>: -<a href="nasmdoc7.html#section-7.4.4">Section 7.4.4</a> -<br> -import library: <a href="nasmdoc7.html#section-7.4.4">Section 7.4.4</a> -<br> -importing symbols: <a href="nasmdoc6.html#section-6.5">Section 6.5</a> -<br> -<code><nobr>INCBIN</nobr></code>: -<a href="nasmdoc3.html#section-3.2">Section 3.2</a>, -<a href="nasmdoc3.html#section-3.2.3">Section 3.2.3</a>, -<a href="nasmdoc3.html#section-3.4.4">Section 3.4.4</a> -<br> -<code><nobr>%include</nobr></code>: -<a href="nasmdoc2.html#section-2.1.16">Section 2.1.16</a>, -<a href="nasmdoc2.html#section-2.1.17">Section 2.1.17</a>, -<a href="nasmdoc4.html#section-4.6.1">Section 4.6.1</a> -<br> -include search path: -<a href="nasmdoc2.html#section-2.1.16">Section 2.1.16</a> -<br> -including other files: -<a href="nasmdoc4.html#section-4.6.1">Section 4.6.1</a> -<br> -inefficient code: <a href="nasmdo12.html#section-12.1.1">Section 12.1.1</a> -<br> -infinite loop: <a href="nasmdoc3.html#section-3.5">Section 3.5</a> -<br> -<code><nobr>__Infinity__</nobr></code>: -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -infinity: <a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -informational section: -<a href="nasmdoc7.html#section-7.5.1">Section 7.5.1</a> -<br> -<code><nobr>INSTALL</nobr></code>: -<a href="nasmdoc1.html#section-1.3.2">Section 1.3.2</a> -<br> -installing: <a href="nasmdoc1.html#section-1.3.1">Section 1.3.1</a> -<br> -instances of structures: -<a href="nasmdoc4.html#section-4.11.11">Section 4.11.11</a> -<br> -instruction list: <a href="nasmdocb.html">Appendix B</a> -<br> -intel hex: <a href="nasmdoc7.html#section-7.2">Section 7.2</a> -<br> -Intel number formats: -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -<code><nobr>internal</nobr></code>: -<a href="nasmdoc7.html#section-7.9.5">Section 7.9.5</a> -<br> -<code><nobr>%irmacro</nobr></code>: -<a href="nasmdoc4.html#section-4.3.1">Section 4.3.1</a> -<br> -<code><nobr>ISTRUC</nobr></code>: -<a href="nasmdoc4.html#section-4.11.11">Section 4.11.11</a> -<br> -iterating over macro parameters: -<a href="nasmdoc4.html#section-4.3.7">Section 4.3.7</a> -<br> -<code><nobr>ith</nobr></code>: -<a href="nasmdoc7.html#section-7.2">Section 7.2</a> -<br> -<code><nobr>%ixdefine</nobr></code>: -<a href="nasmdoc4.html#section-4.1.2">Section 4.1.2</a> -<br> -<code><nobr>Jcc NEAR</nobr></code>: -<a href="nasmdo12.html#section-12.1.2">Section 12.1.2</a> -<br> -<code><nobr>JMP DWORD</nobr></code>: -<a href="nasmdo10.html#section-10.1">Section 10.1</a> -<br> -jumps, mixed-size: <a href="nasmdo10.html#section-10.1">Section 10.1</a> -<br> -<code><nobr>-k</nobr></code>: -<a href="nasmdoca.html#section-A.3.4">Section A.3.4</a> -<br> -<code><nobr>-l</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.3">Section 2.1.3</a> -<br> -label prefix: <a href="nasmdoc3.html#section-3.9">Section 3.9</a> -<br> -<code><nobr>.lbss</nobr></code>: -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a> -<br> -<code><nobr>ld86</nobr></code>: -<a href="nasmdoc7.html#section-7.12">Section 7.12</a> -<br> -<code><nobr>.ldata</nobr></code>: -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a> -<br> -<code><nobr>LIBRARY</nobr></code>: -<a href="nasmdoc7.html#section-7.13.1">Section 7.13.1</a> -<br> -license: <a href="nasmdoc1.html#section-1.1.2">Section 1.1.2</a> -<br> -<code><nobr>%line</nobr></code>: -<a href="nasmdoc4.html#section-4.10.1">Section 4.10.1</a> -<br> -<code><nobr>__LINE__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.4">Section 4.11.4</a> -<br> -linker, free: <a href="nasmdoc8.html#section-8.1.1">Section 8.1.1</a> -<br> -Linux, <code><nobr>a.out</nobr></code>: -<a href="nasmdoc7.html#section-7.10">Section 7.10</a> -<br> -Linux, <code><nobr>as86</nobr></code>: -<a href="nasmdoc7.html#section-7.12">Section 7.12</a> -<br> -Linux, ELF: <a href="nasmdoc7.html#section-7.9">Section 7.9</a> -<br> -listing file: <a href="nasmdoc2.html#section-2.1.3">Section 2.1.3</a> -<br> -little-endian: <a href="nasmdoc3.html#section-3.4.3">Section 3.4.3</a> -<br> -<code><nobr>%local</nobr></code>: -<a href="nasmdoc4.html#section-4.8.3">Section 4.8.3</a> -<br> -local labels: <a href="nasmdoc3.html#section-3.9">Section 3.9</a> -<br> -logical AND: <a href="nasmdoc4.html#section-4.4.4">Section 4.4.4</a> -<br> -logical negation: <a href="nasmdoc3.html#section-3.5.7">Section 3.5.7</a> -<br> -logical OR: <a href="nasmdoc4.html#section-4.4.4">Section 4.4.4</a> -<br> -logical XOR: <a href="nasmdoc4.html#section-4.4.4">Section 4.4.4</a> -<br> -<code><nobr>.lrodata</nobr></code>: -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a> -<br> -<code><nobr>-M</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.4">Section 2.1.4</a> -<br> -Mach, object file format: -<a href="nasmdoc7.html#section-7.8">Section 7.8</a> -<br> -Mach-O: <a href="nasmdoc7.html#section-7.8">Section 7.8</a> -<br> -<code><nobr>macho</nobr></code>: -<a href="nasmdoc7.html#section-7.8">Section 7.8</a> -<br> -<code><nobr>macho32</nobr></code>: -<a href="nasmdoc7.html#section-7.8">Section 7.8</a> -<br> -<code><nobr>macho64</nobr></code>: -<a href="nasmdoc7.html#section-7.8">Section 7.8</a> -<br> -MacOS X: <a href="nasmdoc7.html#section-7.8">Section 7.8</a> -<br> -<code><nobr>%macro</nobr></code>: -<a href="nasmdoc4.html#section-4.3">Section 4.3</a> -<br> -macro indirection: <a href="nasmdoc4.html#section-4.1.3">Section 4.1.3</a> -<br> -macro library: <a href="nasmdoc2.html#section-2.1.16">Section 2.1.16</a> -<br> -macro processor: <a href="nasmdoc4.html">Chapter 4</a> -<br> -<code><nobr>macro-defaults</nobr></code>: -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a> -<br> -macro-local labels: <a href="nasmdoc4.html#section-4.3.3">Section 4.3.3</a> -<br> -<code><nobr>macro-params</nobr></code>: -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a> -<br> -macros: <a href="nasmdoc3.html#section-3.2.5">Section 3.2.5</a> -<br> -<code><nobr>macro-selfref</nobr></code>: -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a> -<br> -<code><nobr>make</nobr></code>: -<a href="nasmdoc1.html#section-1.3.2">Section 1.3.2</a> -<br> -makefile dependencies: -<a href="nasmdoc2.html#section-2.1.4">Section 2.1.4</a>, -<a href="nasmdoc2.html#section-2.1.5">Section 2.1.5</a> -<br> -makefiles: <a href="nasmdoc1.html#section-1.3.1">Section 1.3.1</a>, -<a href="nasmdoc1.html#section-1.3.2">Section 1.3.2</a> -<br> -man pages: <a href="nasmdoc1.html#section-1.3.2">Section 1.3.2</a> -<br> -map files: <a href="nasmdoc7.html#section-7.1.4">Section 7.1.4</a> -<br> -<code><nobr>MASM</nobr></code>: -<a href="nasmdoc1.html#section-1.1.1">Section 1.1.1</a> -<br> -MASM: <a href="nasmdoc2.html#section-2.2">Section 2.2</a>, -<a href="nasmdoc3.html#section-3.2.5">Section 3.2.5</a>, -<a href="nasmdoc7.html#section-7.4">Section 7.4</a> -<br> -<code><nobr>-MD</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.7">Section 2.1.7</a> -<br> -memory models: <a href="nasmdoc2.html#section-2.2.5">Section 2.2.5</a>, -<a href="nasmdoc8.html#section-8.4.2">Section 8.4.2</a> -<br> -memory operand: <a href="nasmdoc3.html#section-3.1">Section 3.1</a> -<br> -memory references: <a href="nasmdoc2.html#section-2.2.2">Section 2.2.2</a>, -<a href="nasmdoc3.html#section-3.3">Section 3.3</a> -<br> -<code><nobr>-MF</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.6">Section 2.1.6</a> -<br> -<code><nobr>-MG</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.5">Section 2.1.5</a> -<br> -Microsoft OMF: <a href="nasmdoc7.html#section-7.4">Section 7.4</a> -<br> -minifloat: <a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -Minix: <a href="nasmdoc7.html#section-7.12">Section 7.12</a> -<br> -<code><nobr>misc</nobr></code> subdirectory: -<a href="nasmdoc8.html#section-8.1.2">Section 8.1.2</a>, -<a href="nasmdoc8.html#section-8.4.5">Section 8.4.5</a>, -<a href="nasmdoc9.html#section-9.1.4">Section 9.1.4</a> -<br> -mixed-language program: <a href="nasmdoc8.html#section-8.4">Section 8.4</a> -<br> -mixed-size addressing: -<a href="nasmdo10.html#section-10.2">Section 10.2</a> -<br> -mixed-size instruction: -<a href="nasmdo10.html#section-10.1">Section 10.1</a> -<br> -MMX registers: -<br> -ModR/M byte: -<br> -<code><nobr>MODULE</nobr></code>: -<a href="nasmdoc7.html#section-7.13.2">Section 7.13.2</a> -<br> -modulo operators: <a href="nasmdoc3.html#section-3.5.6">Section 3.5.6</a> -<br> -motorola s-records: <a href="nasmdoc7.html#section-7.3">Section 7.3</a> -<br> -<code><nobr>-MP</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.10">Section 2.1.10</a> -<br> -<code><nobr>-MQ</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.9">Section 2.1.9</a> -<br> -MS-DOS: <a href="nasmdoc7.html#section-7.1">Section 7.1</a> -<br> -MS-DOS device drivers: <a href="nasmdoc8.html#section-8.3">Section 8.3</a> -<br> -<code><nobr>-MT</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.8">Section 2.1.8</a> -<br> -multi-line macros: -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a>, -<a href="nasmdoc4.html#section-4.3">Section 4.3</a> -<br> -multipass optimization: -<a href="nasmdoc2.html#section-2.1.22">Section 2.1.22</a> -<br> -multiple section names: <a href="nasmdoc7.html#section-7.1">Section 7.1</a> -<br> -multiplication: <a href="nasmdoc3.html#section-3.5.6">Section 3.5.6</a> -<br> -<code><nobr>multipush</nobr></code> macro: -<a href="nasmdoc4.html#section-4.3.7">Section 4.3.7</a> -<br> -multisection: <a href="nasmdoc7.html#section-7.1.3">Section 7.1.3</a> -<br> -<code><nobr>__NaN__</nobr></code>: -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -NaN: <a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -NASM version: <a href="nasmdoc4.html#section-4.11.1">Section 4.11.1</a> -<br> -nasm version history: <a href="nasmdocc.html">Appendix C</a> -<br> -nasm version id: <a href="nasmdoc4.html#section-4.11.2">Section 4.11.2</a> -<br> -nasm version string: -<a href="nasmdoc4.html#section-4.11.3">Section 4.11.3</a> -<br> -<code><nobr>nasm.1</nobr></code>: -<a href="nasmdoc1.html#section-1.3.2">Section 1.3.2</a> -<br> -<code><nobr>__NASMDEFSEG</nobr></code>: -<a href="nasmdoc7.html#section-7.4">Section 7.4</a> -<br> -<code><nobr>nasm-devel</nobr></code>: -<a href="nasmdoc1.html#section-1.2">Section 1.2</a> -<br> -<code><nobr>NASMENV</nobr></code>: -<a href="nasmdoc2.html#section-2.1.28">Section 2.1.28</a> -<br> -<code><nobr>nasm.exe</nobr></code>: -<a href="nasmdoc1.html#section-1.3.1">Section 1.3.1</a> -<br> -<code><nobr>nasm -hf</nobr></code>: -<a href="nasmdoc2.html#section-2.1.2">Section 2.1.2</a> -<br> -<code><nobr>__NASM_MAJOR__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.1">Section 4.11.1</a> -<br> -<code><nobr>__NASM_MINOR__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.1">Section 4.11.1</a> -<br> -<code><nobr>nasm.out</nobr></code>: -<a href="nasmdoc2.html#section-2.1.1">Section 2.1.1</a> -<br> -<code><nobr>___NASM_PATCHLEVEL__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.1">Section 4.11.1</a> -<br> -<code><nobr>__NASM_SNAPSHOT__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.1">Section 4.11.1</a> -<br> -<code><nobr>__NASM_SUBMINOR__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.1">Section 4.11.1</a> -<br> -<code><nobr>__NASM_VER__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.3">Section 4.11.3</a> -<br> -<code><nobr>__NASM_VERSION_ID__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.2">Section 4.11.2</a> -<br> -<code><nobr>nasm-XXX-dos.zip</nobr></code>: -<a href="nasmdoc1.html#section-1.3.1">Section 1.3.1</a> -<br> -<code><nobr>nasm-XXX.tar.gz</nobr></code>: -<a href="nasmdoc1.html#section-1.3.2">Section 1.3.2</a> -<br> -<code><nobr>nasm-XXX-win32.zip</nobr></code>: -<a href="nasmdoc1.html#section-1.3.1">Section 1.3.1</a> -<br> -<code><nobr>nasm-XXX.zip</nobr></code>: -<a href="nasmdoc1.html#section-1.3.1">Section 1.3.1</a> -<br> -ndisasm: <a href="nasmdoca.html">Appendix A</a> -<br> -<code><nobr>ndisasm.1</nobr></code>: -<a href="nasmdoc1.html#section-1.3.2">Section 1.3.2</a> -<br> -<code><nobr>ndisasm.exe</nobr></code>: -<a href="nasmdoc1.html#section-1.3.1">Section 1.3.1</a> -<br> -near call: <a href="nasmdoc2.html#section-2.2.5">Section 2.2.5</a> -<br> -near common variables: -<a href="nasmdoc7.html#section-7.4.8">Section 7.4.8</a> -<br> -NetBSD: <a href="nasmdoc7.html#section-7.11">Section 7.11</a>, -<a href="nasmdoc9.html#section-9.2">Section 9.2</a> -<br> -new releases: <a href="nasmdoc1.html#section-1.2">Section 1.2</a> -<br> -<code><nobr>noalloc</nobr></code>: -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a> -<br> -<code><nobr>nobits</nobr></code>: -<a href="nasmdoc7.html#section-7.1.3">Section 7.1.3</a>, -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a> -<br> -<code><nobr>noexec</nobr></code>: -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a> -<br> -<code><nobr>.nolist</nobr></code>: -<a href="nasmdoc4.html#section-4.3.10">Section 4.3.10</a> -<br> -`nowait': <a href="nasmdoc2.html#section-2.2.6">Section 2.2.6</a> -<br> -<code><nobr>nowrite</nobr></code>: -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a> -<br> -<code><nobr>number-overflow</nobr></code>: -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a> -<br> -numeric constants: <a href="nasmdoc3.html#section-3.2.1">Section 3.2.1</a>, -<a href="nasmdoc3.html#section-3.4.1">Section 3.4.1</a> -<br> -<code><nobr>-O</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.22">Section 2.1.22</a> -<br> -<code><nobr>-o</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.1">Section 2.1.1</a>, -<a href="nasmdoca.html#section-A.3.1">Section A.3.1</a> -<br> -<code><nobr>O16</nobr></code>: -<a href="nasmdoc3.html#section-3.1">Section 3.1</a> -<br> -<code><nobr>o16</nobr></code>: -<a href="nasmdo10.html#section-10.3">Section 10.3</a> -<br> -<code><nobr>O32</nobr></code>: -<a href="nasmdoc3.html#section-3.1">Section 3.1</a> -<br> -<code><nobr>o32</nobr></code>: -<a href="nasmdo10.html#section-10.3">Section 10.3</a> -<br> -<code><nobr>O64</nobr></code>: -<a href="nasmdoc3.html#section-3.1">Section 3.1</a> -<br> -<code><nobr>.OBJ</nobr></code>: -<a href="nasmdoc8.html#section-8.1">Section 8.1</a> -<br> -<code><nobr>obj</nobr></code>: -<a href="nasmdoc7.html#section-7.4">Section 7.4</a> -<br> -<code><nobr>object</nobr></code>: -<a href="nasmdoc7.html#section-7.9.5">Section 7.9.5</a>, -<a href="nasmdoc7.html#section-7.13.3">Section 7.13.3</a> -<br> -octal: <a href="nasmdoc3.html#section-3.4.1">Section 3.4.1</a> -<br> -<code><nobr>OF_DBG</nobr></code>: -<a href="nasmdoc7.html#section-7.14">Section 7.14</a> -<br> -<code><nobr>OF_DEFAULT</nobr></code>: -<a href="nasmdoc2.html#section-2.1.2">Section 2.1.2</a> -<br> -<code><nobr>OFFSET</nobr></code>: -<a href="nasmdoc2.html#section-2.2.2">Section 2.2.2</a> -<br> -OMF: <a href="nasmdoc7.html#section-7.4">Section 7.4</a> -<br> -omitted parameters: <a href="nasmdoc4.html#section-4.3.5">Section 4.3.5</a> -<br> -one's complement: <a href="nasmdoc3.html#section-3.5.7">Section 3.5.7</a> -<br> -OpenBSD: <a href="nasmdoc7.html#section-7.11">Section 7.11</a>, -<a href="nasmdoc9.html#section-9.2">Section 9.2</a> -<br> -operands: <a href="nasmdoc3.html#section-3.1">Section 3.1</a> -<br> -operand-size prefixes: <a href="nasmdoc3.html#section-3.1">Section 3.1</a> -<br> -operating system: <a href="nasmdoc7.html#section-7.1">Section 7.1</a> -<br> -operating system, writing: -<a href="nasmdo10.html#section-10.1">Section 10.1</a> -<br> -operators: <a href="nasmdoc3.html#section-3.5">Section 3.5</a> -<br> -<code><nobr>ORG</nobr></code>: -<a href="nasmdoc7.html#section-7.1.1">Section 7.1.1</a>, -<a href="nasmdoc8.html#section-8.2.1">Section 8.2.1</a>, -<a href="nasmdoc8.html#section-8.2.2">Section 8.2.2</a>, -<a href="nasmdo12.html#section-12.1.3">Section 12.1.3</a> -<br> -<code><nobr>orphan-labels</nobr></code>: -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a>, -<a href="nasmdoc3.html#section-3.1">Section 3.1</a> -<br> -OS/2: <a href="nasmdoc7.html#section-7.4">Section 7.4</a>, -<a href="nasmdoc7.html#section-7.4.1">Section 7.4.1</a> -<br> -<code><nobr>osabi</nobr></code>: -<a href="nasmdoc7.html#section-7.9.1">Section 7.9.1</a> -<br> -other preprocessor directives: -<a href="nasmdoc4.html#section-4.10">Section 4.10</a> -<br> -out of range, jumps: -<a href="nasmdo12.html#section-12.1.2">Section 12.1.2</a> -<br> -output file format: <a href="nasmdoc2.html#section-2.1.2">Section 2.1.2</a> -<br> -output formats: <a href="nasmdoc7.html">Chapter 7</a> -<br> -<code><nobr>__OUTPUT_FORMAT__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.6">Section 4.11.6</a> -<br> -overlapping segments: <a href="nasmdoc3.html#section-3.6">Section 3.6</a> -<br> -<code><nobr>OVERLAY</nobr></code>: -<a href="nasmdoc7.html#section-7.4.1">Section 7.4.1</a> -<br> -overloading, multi-line macros: -<a href="nasmdoc4.html#section-4.3.2">Section 4.3.2</a> -<br> -overloading, single-line macros: -<a href="nasmdoc4.html#section-4.1.1">Section 4.1.1</a> -<br> -<code><nobr>-P</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.17">Section 2.1.17</a> -<br> -<code><nobr>-p</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.17">Section 2.1.17</a>, -<a href="nasmdoc4.html#section-4.6.1">Section 4.6.1</a> -<br> -paradox: <a href="nasmdoc3.html#section-3.8">Section 3.8</a> -<br> -<code><nobr>PASCAL</nobr></code>: -<a href="nasmdoc8.html#section-8.5.3">Section 8.5.3</a> -<br> -Pascal calling convention: -<a href="nasmdoc8.html#section-8.5.1">Section 8.5.1</a> -<br> -<code><nobr>__PASS__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.9">Section 4.11.9</a> -<br> -passes, assembly: -<br> -<code><nobr>PATH</nobr></code>: -<a href="nasmdoc1.html#section-1.3.1">Section 1.3.1</a> -<br> -<code><nobr>%pathsearch</nobr></code>: -<a href="nasmdoc2.html#section-2.1.16">Section 2.1.16</a>, -<a href="nasmdoc4.html#section-4.6.2">Section 4.6.2</a> -<br> -period: <a href="nasmdoc3.html#section-3.9">Section 3.9</a> -<br> -Perl: <a href="nasmdoc1.html#section-1.3.1">Section 1.3.1</a> -<br> -perverse: <a href="nasmdoc2.html#section-2.1.16">Section 2.1.16</a> -<br> -PharLap: <a href="nasmdoc7.html#section-7.4.1">Section 7.4.1</a> -<br> -PIC: <a href="nasmdoc7.html#section-7.9.3">Section 7.9.3</a>, -<a href="nasmdoc7.html#section-7.11">Section 7.11</a>, -<a href="nasmdoc9.html#section-9.2">Section 9.2</a> -<br> -<code><nobr>..plt</nobr></code>: -<a href="nasmdoc7.html#section-7.9.3">Section 7.9.3</a> -<br> -<code><nobr>PLT</nobr></code> relocations: -<a href="nasmdoc7.html#section-7.9.3">Section 7.9.3</a>, -<a href="nasmdoc9.html#section-9.2.4">Section 9.2.4</a>, -<a href="nasmdoc9.html#section-9.2.5">Section 9.2.5</a> -<br> -plt relocations: <a href="nasmdoc9.html#section-9.2.5">Section 9.2.5</a> -<br> -<code><nobr>%pop</nobr></code>: -<a href="nasmdoc4.html#section-4.7">Section 4.7</a>, -<a href="nasmdoc4.html#section-4.7.1">Section 4.7.1</a> -<br> -position-independent code: -<a href="nasmdoc7.html#section-7.9.3">Section 7.9.3</a>, -<a href="nasmdoc7.html#section-7.11">Section 7.11</a>, -<a href="nasmdoc9.html#section-9.2">Section 9.2</a> -<br> -<code><nobr>--postfix</nobr></code>: -<a href="nasmdoc2.html#section-2.1.27">Section 2.1.27</a> -<br> -precedence: <a href="nasmdoc3.html#section-3.5">Section 3.5</a> -<br> -pre-defining macros: -<a href="nasmdoc2.html#section-2.1.18">Section 2.1.18</a>, -<a href="nasmdoc4.html#section-4.1.1">Section 4.1.1</a> -<br> -preferred: <a href="nasmdoc3.html#section-3.6">Section 3.6</a> -<br> -<code><nobr>--prefix</nobr></code>: -<a href="nasmdoc2.html#section-2.1.27">Section 2.1.27</a> -<br> -pre-including files: -<a href="nasmdoc2.html#section-2.1.17">Section 2.1.17</a> -<br> -preprocess-only mode: -<a href="nasmdoc2.html#section-2.1.20">Section 2.1.20</a> -<br> -preprocessor: <a href="nasmdoc2.html#section-2.1.20">Section 2.1.20</a>, -<a href="nasmdoc2.html#section-2.1.21">Section 2.1.21</a>, -<a href="nasmdoc3.html#section-3.2.4">Section 3.2.4</a>, -<a href="nasmdoc3.html#section-3.5.6">Section 3.5.6</a>, -<a href="nasmdoc4.html">Chapter 4</a> -<br> -preprocessor expressions: -<a href="nasmdoc2.html#section-2.1.20">Section 2.1.20</a> -<br> -preprocessor loops: <a href="nasmdoc4.html#section-4.5">Section 4.5</a> -<br> -preprocessor variables: -<a href="nasmdoc4.html#section-4.1.7">Section 4.1.7</a> -<br> -primitive directives: <a href="nasmdoc6.html">Chapter 6</a> -<br> -<code><nobr>PRIVATE</nobr></code>: -<a href="nasmdoc7.html#section-7.4.1">Section 7.4.1</a> -<br> -<code><nobr>proc</nobr></code>: -<a href="nasmdoc7.html#section-7.13.3">Section 7.13.3</a>, -<a href="nasmdoc8.html#section-8.4.5">Section 8.4.5</a>, -<a href="nasmdoc9.html#section-9.1.4">Section 9.1.4</a> -<br> -procedure linkage table: -<a href="nasmdoc7.html#section-7.9.3">Section 7.9.3</a>, -<a href="nasmdoc9.html#section-9.2.4">Section 9.2.4</a>, -<a href="nasmdoc9.html#section-9.2.5">Section 9.2.5</a> -<br> -processor mode: <a href="nasmdoc6.html#section-6.1">Section 6.1</a> -<br> -<code><nobr>progbits</nobr></code>: -<a href="nasmdoc7.html#section-7.1.3">Section 7.1.3</a>, -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a> -<br> -program entry point: -<a href="nasmdoc7.html#section-7.4.6">Section 7.4.6</a>, -<a href="nasmdoc8.html#section-8.1.1">Section 8.1.1</a> -<br> -program origin: <a href="nasmdoc7.html#section-7.1.1">Section 7.1.1</a> -<br> -<code><nobr>protected</nobr></code>: -<a href="nasmdoc7.html#section-7.9.5">Section 7.9.5</a> -<br> -pseudo-instructions: <a href="nasmdoc3.html#section-3.2">Section 3.2</a> -<br> -<code><nobr>PUBLIC</nobr></code>: -<a href="nasmdoc6.html#section-6.6">Section 6.6</a>, -<a href="nasmdoc7.html#section-7.4.1">Section 7.4.1</a> -<br> -pure binary: <a href="nasmdoc7.html#section-7.1">Section 7.1</a> -<br> -<code><nobr>%push</nobr></code>: -<a href="nasmdoc4.html#section-4.7">Section 4.7</a>, -<a href="nasmdoc4.html#section-4.7.1">Section 4.7.1</a> -<br> -<code><nobr>__QNaN__</nobr></code>: -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -quick start: <a href="nasmdoc2.html#section-2.2">Section 2.2</a> -<br> -<code><nobr>QWORD</nobr></code>: -<a href="nasmdoc3.html#section-3.1">Section 3.1</a> -<br> -<code><nobr>-r</nobr></code>: -<a href="nasmdoca.html#section-A.3">Section A.3</a> -<br> -<code><nobr>rdf</nobr></code>: -<a href="nasmdoc7.html#section-7.13">Section 7.13</a> -<br> -<code><nobr>rdoff</nobr></code> subdirectory: -<a href="nasmdoc1.html#section-1.3.2">Section 1.3.2</a>, -<a href="nasmdoc7.html#section-7.13">Section 7.13</a> -<br> -recursive multi-line macros: -<a href="nasmdoc4.html#section-4.3.1">Section 4.3.1</a> -<br> -redirecting errors: -<a href="nasmdoc2.html#section-2.1.14">Section 2.1.14</a> -<br> -<code><nobr>REL</nobr></code>: -<a href="nasmdoc3.html#section-3.3">Section 3.3</a>, -<a href="nasmdoc6.html#section-6.2">Section 6.2</a> -<br> -relational operators: -<a href="nasmdoc4.html#section-4.4.4">Section 4.4.4</a> -<br> -release candidates: <a href="nasmdoc1.html#section-1.2">Section 1.2</a> -<br> -Relocatable Dynamic Object File Format: -<a href="nasmdoc7.html#section-7.13">Section 7.13</a> -<br> -relocations, PIC-specific: -<a href="nasmdoc7.html#section-7.9.3">Section 7.9.3</a> -<br> -removing contexts: <a href="nasmdoc4.html#section-4.7.1">Section 4.7.1</a> -<br> -renaming contexts: <a href="nasmdoc4.html#section-4.7.4">Section 4.7.4</a> -<br> -<code><nobr>%rep</nobr></code>: -<a href="nasmdoc3.html#section-3.2.5">Section 3.2.5</a>, -<a href="nasmdoc4.html#section-4.5">Section 4.5</a> -<br> -repeating: <a href="nasmdoc3.html#section-3.2.5">Section 3.2.5</a>, -<a href="nasmdoc4.html#section-4.5">Section 4.5</a> -<br> -<code><nobr>%repl</nobr></code>: -<a href="nasmdoc4.html#section-4.7.4">Section 4.7.4</a> -<br> -reporting bugs: <a href="nasmdo12.html#section-12.2">Section 12.2</a> -<br> -<code><nobr>RESB</nobr></code>: -<a href="nasmdoc2.html#section-2.2.7">Section 2.2.7</a>, -<a href="nasmdoc3.html#section-3.2">Section 3.2</a>, -<a href="nasmdoc3.html#section-3.2.2">Section 3.2.2</a> -<br> -<code><nobr>RESD</nobr></code>: -<a href="nasmdoc3.html#section-3.2">Section 3.2</a>, -<a href="nasmdoc3.html#section-3.2.2">Section 3.2.2</a> -<br> -<code><nobr>RESO</nobr></code>: -<a href="nasmdoc3.html#section-3.2">Section 3.2</a>, -<a href="nasmdoc3.html#section-3.2.2">Section 3.2.2</a> -<br> -<code><nobr>RESQ</nobr></code>: -<a href="nasmdoc3.html#section-3.2">Section 3.2</a>, -<a href="nasmdoc3.html#section-3.2.2">Section 3.2.2</a> -<br> -<code><nobr>REST</nobr></code>: -<a href="nasmdoc3.html#section-3.2">Section 3.2</a>, -<a href="nasmdoc3.html#section-3.2.2">Section 3.2.2</a> -<br> -<code><nobr>RESW</nobr></code>: -<a href="nasmdoc3.html#section-3.2">Section 3.2</a>, -<a href="nasmdoc3.html#section-3.2.2">Section 3.2.2</a> -<br> -<code><nobr>RESY</nobr></code>: -<a href="nasmdoc3.html#section-3.2">Section 3.2</a>, -<a href="nasmdoc3.html#section-3.2.2">Section 3.2.2</a> -<br> -<code><nobr>%rmacro</nobr></code>: -<a href="nasmdoc4.html#section-4.3.1">Section 4.3.1</a> -<br> -<code><nobr>.rodata</nobr></code>: -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a> -<br> -<code><nobr>%rotate</nobr></code>: -<a href="nasmdoc4.html#section-4.3.7">Section 4.3.7</a> -<br> -rotating macro parameters: -<a href="nasmdoc4.html#section-4.3.7">Section 4.3.7</a> -<br> -<code><nobr>-s</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.15">Section 2.1.15</a>, -<a href="nasmdoca.html#section-A.3.2">Section A.3.2</a> -<br> -searching for include files: -<a href="nasmdoc4.html#section-4.6.1">Section 4.6.1</a> -<br> -<code><nobr>__SECT__</nobr></code>: -<a href="nasmdoc6.html#section-6.3.1">Section 6.3.1</a>, -<a href="nasmdoc6.html#section-6.4">Section 6.4</a> -<br> -<code><nobr>SECTION</nobr></code>: -<a href="nasmdoc6.html#section-6.3">Section 6.3</a> -<br> -<code><nobr>SECTION</nobr></code>, <code><nobr>elf</nobr></code> extensions -to: <a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a> -<br> -<code><nobr>SECTION</nobr></code>, <code><nobr>win32</nobr></code> -extensions to: <a href="nasmdoc7.html#section-7.5.1">Section 7.5.1</a> -<br> -section alignment, in <code><nobr>bin</nobr></code>: -<a href="nasmdoc7.html#section-7.1.2">Section 7.1.2</a> -<br> -section alignment, in <code><nobr>elf</nobr></code>: -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a> -<br> -section alignment, in <code><nobr>obj</nobr></code>: -<a href="nasmdoc7.html#section-7.4.1">Section 7.4.1</a> -<br> -section alignment, in <code><nobr>win32</nobr></code>: -<a href="nasmdoc7.html#section-7.5.1">Section 7.5.1</a> -<br> -section, bin extensions to: -<a href="nasmdoc7.html#section-7.1.2">Section 7.1.2</a> -<br> -<code><nobr>SEG</nobr></code>: -<a href="nasmdoc3.html#section-3.5.7">Section 3.5.7</a>, -<a href="nasmdoc3.html#section-3.6">Section 3.6</a>, -<a href="nasmdoc7.html#section-7.4">Section 7.4</a> -<br> -<code><nobr>SEGMENT</nobr></code>: -<a href="nasmdoc6.html#section-6.3">Section 6.3</a> -<br> -<code><nobr>SEGMENT</nobr></code>, <code><nobr>elf</nobr></code> extensions -to: <a href="nasmdoc7.html#section-7.4.1">Section 7.4.1</a> -<br> -segment address: <a href="nasmdoc3.html#section-3.5.7">Section 3.5.7</a>, -<a href="nasmdoc3.html#section-3.6">Section 3.6</a> -<br> -segment alignment, in <code><nobr>bin</nobr></code>: -<a href="nasmdoc7.html#section-7.1.2">Section 7.1.2</a> -<br> -segment alignment, in <code><nobr>obj</nobr></code>: -<a href="nasmdoc7.html#section-7.4.1">Section 7.4.1</a> -<br> -segment names, Borland Pascal: -<a href="nasmdoc8.html#section-8.5.2">Section 8.5.2</a> -<br> -segment override: <a href="nasmdoc2.html#section-2.2.4">Section 2.2.4</a>, -<a href="nasmdoc3.html#section-3.1">Section 3.1</a> -<br> -segments: <a href="nasmdoc3.html#section-3.6">Section 3.6</a> -<br> -segments, groups of: -<a href="nasmdoc7.html#section-7.4.2">Section 7.4.2</a> -<br> -separator character: -<a href="nasmdoc2.html#section-2.1.28">Section 2.1.28</a> -<br> -shared libraries: <a href="nasmdoc7.html#section-7.11">Section 7.11</a>, -<a href="nasmdoc9.html#section-9.2">Section 9.2</a> -<br> -shared library: <a href="nasmdoc7.html#section-7.9.5">Section 7.9.5</a> -<br> -<code><nobr>shift</nobr></code> command: -<a href="nasmdoc4.html#section-4.3.7">Section 4.3.7</a> -<br> -SIB byte: -<br> -signed division: <a href="nasmdoc3.html#section-3.5.6">Section 3.5.6</a> -<br> -signed modulo: <a href="nasmdoc3.html#section-3.5.6">Section 3.5.6</a> -<br> -single-line macros: <a href="nasmdoc4.html#section-4.1">Section 4.1</a> -<br> -size, of symbols: <a href="nasmdoc7.html#section-7.9.5">Section 7.9.5</a> -<br> -<code><nobr>smartalign</nobr></code>: -<a href="nasmdoc5.html#section-5.2">Section 5.2</a> -<br> -<code><nobr>__SNaN__</nobr></code>: -<a href="nasmdoc3.html#section-3.4.6">Section 3.4.6</a> -<br> -snapshots, daily development: -<a href="nasmdoc1.html#section-1.2">Section 1.2</a> -<br> -Solaris x86: <a href="nasmdoc7.html#section-7.9">Section 7.9</a> -<br> -<code><nobr>-soname</nobr></code>: -<a href="nasmdoc9.html#section-9.2.6">Section 9.2.6</a> -<br> -sound: <a href="nasmdoc3.html#section-3.2.3">Section 3.2.3</a> -<br> -source code: <a href="nasmdoc1.html#section-1.3.1">Section 1.3.1</a> -<br> -source-listing file: -<a href="nasmdoc2.html#section-2.1.3">Section 2.1.3</a> -<br> -square brackets: <a href="nasmdoc2.html#section-2.2.2">Section 2.2.2</a>, -<a href="nasmdoc3.html#section-3.3">Section 3.3</a> -<br> -<code><nobr>srec</nobr></code>: -<a href="nasmdoc7.html#section-7.3">Section 7.3</a> -<br> -<code><nobr>STACK</nobr></code>: -<a href="nasmdoc7.html#section-7.4.1">Section 7.4.1</a> -<br> -stack relative preprocessor directives: -<a href="nasmdoc4.html#section-4.8">Section 4.8</a> -<br> -<code><nobr>%stacksize</nobr></code>: -<a href="nasmdoc4.html#section-4.8.2">Section 4.8.2</a> -<br> -standard macro packages: <a href="nasmdoc5.html">Chapter 5</a> -<br> -standard macros: <a href="nasmdoc4.html#section-4.11">Section 4.11</a> -<br> -standardized section names: -<a href="nasmdoc6.html#section-6.3">Section 6.3</a>, -<a href="nasmdoc7.html#section-7.5.1">Section 7.5.1</a>, -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a>, -<a href="nasmdoc7.html#section-7.10">Section 7.10</a>, -<a href="nasmdoc7.html#section-7.11">Section 7.11</a>, -<a href="nasmdoc7.html#section-7.12">Section 7.12</a>, -<a href="nasmdoc7.html#section-7.13">Section 7.13</a> -<br> -<code><nobr>..start</nobr></code>: -<a href="nasmdoc7.html#section-7.4.6">Section 7.4.6</a>, -<a href="nasmdoc8.html#section-8.1.1">Section 8.1.1</a> -<br> -<code><nobr>start=</nobr></code>: -<a href="nasmdoc7.html#section-7.1.3">Section 7.1.3</a> -<br> -<code><nobr>stderr</nobr></code>: -<a href="nasmdoc2.html#section-2.1.14">Section 2.1.14</a> -<br> -<code><nobr>stdout</nobr></code>: -<a href="nasmdoc2.html#section-2.1.15">Section 2.1.15</a> -<br> -<code><nobr>%strcat</nobr></code>: -<a href="nasmdoc4.html#section-4.2.1">Section 4.2.1</a> -<br> -<code><nobr>STRICT</nobr></code>: -<a href="nasmdoc3.html#section-3.7">Section 3.7</a> -<br> -string constant: <a href="nasmdoc3.html#section-3.2.1">Section 3.2.1</a> -<br> -string constants: <a href="nasmdoc3.html#section-3.4.4">Section 3.4.4</a> -<br> -string length: <a href="nasmdoc4.html#section-4.2.2">Section 4.2.2</a> -<br> -string manipulation in macros: -<a href="nasmdoc4.html#section-4.2">Section 4.2</a> -<br> -strings: <a href="nasmdoc3.html#section-3.4.2">Section 3.4.2</a> -<br> -<code><nobr>%strlen</nobr></code>: -<a href="nasmdoc4.html#section-4.2.2">Section 4.2.2</a> -<br> -<code><nobr>STRUC</nobr></code>: -<a href="nasmdoc4.html#section-4.11.10">Section 4.11.10</a>, -<a href="nasmdoc6.html#section-6.4">Section 6.4</a>, -<a href="nasmdoc8.html#section-8.4.4">Section 8.4.4</a>, -<a href="nasmdoc9.html#section-9.1.3">Section 9.1.3</a> -<br> -stub preprocessor: -<a href="nasmdoc2.html#section-2.1.21">Section 2.1.21</a> -<br> -<code><nobr>%substr</nobr></code>: -<a href="nasmdoc4.html#section-4.2.3">Section 4.2.3</a> -<br> -subtraction: <a href="nasmdoc3.html#section-3.5.5">Section 3.5.5</a> -<br> -suppressible warning: -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a> -<br> -suppressing preprocessing: -<a href="nasmdoc2.html#section-2.1.21">Section 2.1.21</a> -<br> -switching between sections: -<a href="nasmdoc6.html#section-6.3">Section 6.3</a> -<br> -<code><nobr>..sym</nobr></code>: -<a href="nasmdoc7.html#section-7.9.3">Section 7.9.3</a> -<br> -symbol sizes, specifying: -<a href="nasmdoc7.html#section-7.9.5">Section 7.9.5</a> -<br> -symbol types, specifying: -<a href="nasmdoc7.html#section-7.9.5">Section 7.9.5</a> -<br> -symbols, exporting from DLLs: -<a href="nasmdoc7.html#section-7.4.5">Section 7.4.5</a> -<br> -symbols, importing from DLLs: -<a href="nasmdoc7.html#section-7.4.4">Section 7.4.4</a> -<br> -<code><nobr>synchronisation</nobr></code>: -<a href="nasmdoca.html#section-A.3.2">Section A.3.2</a> -<br> -<code><nobr>.SYS</nobr></code>: -<a href="nasmdoc7.html#section-7.1">Section 7.1</a>, -<a href="nasmdoc8.html#section-8.3">Section 8.3</a> -<br> -<code><nobr>-t</nobr></code>: -<a href="nasmdoc2.html#section-2.1.23">Section 2.1.23</a> -<br> -<code><nobr>TASM</nobr></code>: -<a href="nasmdoc1.html#section-1.1.1">Section 1.1.1</a>, -<a href="nasmdoc2.html#section-2.1.23">Section 2.1.23</a> -<br> -tasm: <a href="nasmdoc2.html#section-2.2">Section 2.2</a>, -<a href="nasmdoc7.html#section-7.4">Section 7.4</a> -<br> -<code><nobr>.tbss</nobr></code>: -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a> -<br> -<code><nobr>TBYTE</nobr></code>: -<a href="nasmdoc2.html#section-2.2.7">Section 2.2.7</a> -<br> -<code><nobr>.tdata</nobr></code>: -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a> -<br> -<code><nobr>test</nobr></code> subdirectory: -<a href="nasmdoc8.html#section-8.1.1">Section 8.1.1</a> -<br> -testing, arbitrary numeric expressions: -<a href="nasmdoc4.html#section-4.4.4">Section 4.4.4</a> -<br> -testing, context stack: -<a href="nasmdoc4.html#section-4.4.3">Section 4.4.3</a> -<br> -testing, exact text identity: -<a href="nasmdoc4.html#section-4.4.5">Section 4.4.5</a> -<br> -testing, multi-line macro existence: -<a href="nasmdoc4.html#section-4.4.2">Section 4.4.2</a> -<br> -testing, single-line macro existence: -<a href="nasmdoc4.html#section-4.4.1">Section 4.4.1</a> -<br> -testing, token types: -<a href="nasmdoc4.html#section-4.4.6">Section 4.4.6</a> -<br> -<code><nobr>.text</nobr></code>: -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a>, -<a href="nasmdoc7.html#section-7.10">Section 7.10</a>, -<a href="nasmdoc7.html#section-7.11">Section 7.11</a>, -<a href="nasmdoc7.html#section-7.12">Section 7.12</a>, -<a href="nasmdoc7.html#section-7.13">Section 7.13</a> -<br> -<code><nobr>_TEXT</nobr></code>: -<a href="nasmdoc8.html#section-8.4.2">Section 8.4.2</a> -<br> -thread local storage: -<a href="nasmdoc7.html#section-7.9.4">Section 7.9.4</a> -<br> -<code><nobr>__TIME__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.7">Section 4.11.7</a> -<br> -<code><nobr>__TIME_NUM__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.7">Section 4.11.7</a> -<br> -<code><nobr>TIMES</nobr></code>: -<a href="nasmdoc3.html#section-3.2">Section 3.2</a>, -<a href="nasmdoc3.html#section-3.2.5">Section 3.2.5</a>, -<a href="nasmdoc3.html#section-3.8">Section 3.8</a>, -<a href="nasmdo12.html#section-12.1.3">Section 12.1.3</a>, -<a href="nasmdo12.html#section-12.1.4">Section 12.1.4</a> -<br> -<code><nobr>TLINK</nobr></code>: -<a href="nasmdoc8.html#section-8.2.2">Section 8.2.2</a> -<br> -<code><nobr>tls</nobr></code>: -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a>, -<a href="nasmdoc7.html#section-7.9.4">Section 7.9.4</a> -<br> -<code><nobr>..tlsie</nobr></code>: -<a href="nasmdoc7.html#section-7.9.4">Section 7.9.4</a> -<br> -trailing colon: <a href="nasmdoc3.html#section-3.1">Section 3.1</a> -<br> -<code><nobr>TWORD</nobr></code>: -<a href="nasmdoc2.html#section-2.2.7">Section 2.2.7</a>, -<a href="nasmdoc3.html#section-3.1">Section 3.1</a> -<br> -type, of symbols: <a href="nasmdoc7.html#section-7.9.5">Section 7.9.5</a> -<br> -<code><nobr>-U</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.19">Section 2.1.19</a> -<br> -<code><nobr>-u</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.19">Section 2.1.19</a>, -<a href="nasmdoca.html#section-A.3">Section A.3</a> -<br> -unary operators: <a href="nasmdoc3.html#section-3.5.7">Section 3.5.7</a> -<br> -<code><nobr>%undef</nobr></code>: -<a href="nasmdoc2.html#section-2.1.19">Section 2.1.19</a>, -<a href="nasmdoc4.html#section-4.1.6">Section 4.1.6</a> -<br> -undefining macros: -<a href="nasmdoc2.html#section-2.1.19">Section 2.1.19</a> -<br> -underscore, in C symbols: -<a href="nasmdoc8.html#section-8.4.1">Section 8.4.1</a> -<br> -Unicode: <a href="nasmdoc3.html#section-3.4.2">Section 3.4.2</a>, -<a href="nasmdoc3.html#section-3.4.5">Section 3.4.5</a> -<br> -uninitialized: <a href="nasmdoc3.html#section-3.2">Section 3.2</a>, -<a href="nasmdoc3.html#section-3.2.2">Section 3.2.2</a> -<br> -uninitialized storage: -<a href="nasmdoc2.html#section-2.2.7">Section 2.2.7</a> -<br> -Unix: <a href="nasmdoc1.html#section-1.3.2">Section 1.3.2</a> -<br> -Unix, SCO: <a href="nasmdoc7.html#section-7.9">Section 7.9</a> -<br> -Unix, source archive: -<a href="nasmdoc1.html#section-1.3.2">Section 1.3.2</a> -<br> -Unix, System V: <a href="nasmdoc7.html#section-7.9">Section 7.9</a> -<br> -UnixWare: <a href="nasmdoc7.html#section-7.9">Section 7.9</a> -<br> -<code><nobr>%unmacro</nobr></code>: -<a href="nasmdoc4.html#section-4.3.11">Section 4.3.11</a> -<br> -unrolled loops: <a href="nasmdoc3.html#section-3.2.5">Section 3.2.5</a> -<br> -unsigned division: <a href="nasmdoc3.html#section-3.5.6">Section 3.5.6</a> -<br> -unsigned modulo: <a href="nasmdoc3.html#section-3.5.6">Section 3.5.6</a> -<br> -<code><nobr>UPPERCASE</nobr></code>: -<a href="nasmdoc2.html#section-2.2.1">Section 2.2.1</a>, -<a href="nasmdoc7.html#section-7.4.3">Section 7.4.3</a> -<br> -<code><nobr>%use</nobr></code>: -<a href="nasmdoc4.html#section-4.6.4">Section 4.6.4</a>, -<a href="nasmdoc5.html">Chapter 5</a> -<br> -<code><nobr>__USE_*__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.8">Section 4.11.8</a> -<br> -<code><nobr>USE16</nobr></code>: -<a href="nasmdoc6.html#section-6.1.1">Section 6.1.1</a>, -<a href="nasmdoc7.html#section-7.4.1">Section 7.4.1</a> -<br> -<code><nobr>USE32</nobr></code>: -<a href="nasmdoc6.html#section-6.1.1">Section 6.1.1</a>, -<a href="nasmdoc7.html#section-7.4.1">Section 7.4.1</a> -<br> -<code><nobr>user</nobr></code>: -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a> -<br> -user-defined errors: <a href="nasmdoc4.html#section-4.9">Section 4.9</a> -<br> -user-level assembler directives: -<a href="nasmdoc4.html#section-4.11">Section 4.11</a> -<br> -user-level directives: <a href="nasmdoc6.html">Chapter 6</a> -<br> -<code><nobr>__UTC_DATE__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.7">Section 4.11.7</a> -<br> -<code><nobr>__UTC_DATE_NUM__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.7">Section 4.11.7</a> -<br> -<code><nobr>__UTC_TIME__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.7">Section 4.11.7</a> -<br> -<code><nobr>__UTC_TIME_NUM__</nobr></code>: -<a href="nasmdoc4.html#section-4.11.7">Section 4.11.7</a> -<br> -UTF-16: <a href="nasmdoc3.html#section-3.4.5">Section 3.4.5</a> -<br> -UTF-32: <a href="nasmdoc3.html#section-3.4.5">Section 3.4.5</a> -<br> -UTF-8: <a href="nasmdoc3.html#section-3.4.2">Section 3.4.2</a> -<br> -<code><nobr>__utf16__</nobr></code>: -<a href="nasmdoc3.html#section-3.4.5">Section 3.4.5</a> -<br> -<code><nobr>__utf32__</nobr></code>: -<a href="nasmdoc3.html#section-3.4.5">Section 3.4.5</a> -<br> -<code><nobr>-v</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.25">Section 2.1.25</a> -<br> -VAL: <a href="nasmdoc8.html#section-8.1.1">Section 8.1.1</a> -<br> -valid characters: <a href="nasmdoc3.html#section-3.1">Section 3.1</a> -<br> -variable types: <a href="nasmdoc2.html#section-2.2.3">Section 2.2.3</a> -<br> -version: <a href="nasmdoc2.html#section-2.1.25">Section 2.1.25</a> -<br> -version number of NASM: -<a href="nasmdoc4.html#section-4.11.1">Section 4.11.1</a> -<br> -<code><nobr>vfollows=</nobr></code>: -<a href="nasmdoc7.html#section-7.1.3">Section 7.1.3</a> -<br> -Visual C++: <a href="nasmdoc7.html#section-7.5">Section 7.5</a> -<br> -<code><nobr>vstart=</nobr></code>: -<a href="nasmdoc7.html#section-7.1.3">Section 7.1.3</a> -<br> -<code><nobr>-W</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a> -<br> -<code><nobr>-w</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a> -<br> -<code><nobr>%warning</nobr></code>: -<a href="nasmdoc4.html#section-4.9">Section 4.9</a> -<br> -warnings: <a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a> -<br> -<code><nobr>[warning *warning-name]</nobr></code>: -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a> -<br> -<code><nobr>[warning +warning-name]</nobr></code>: -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a> -<br> -<code><nobr>[warning -warning-name]</nobr></code>: -<a href="nasmdoc2.html#section-2.1.24">Section 2.1.24</a> -<br> -website: <a href="nasmdoc1.html#section-1.2">Section 1.2</a> -<br> -<code><nobr>win64</nobr></code>: -<a href="nasmdoc7.html#section-7.6">Section 7.6</a>, -<a href="nasmdo11.html">Chapter 11</a> -<br> -Win64: <a href="nasmdoc7.html#section-7.4">Section 7.4</a>, -<a href="nasmdoc7.html#section-7.5">Section 7.5</a>, -<a href="nasmdoc9.html">Chapter 9</a> -<br> -Windows: <a href="nasmdoc8.html#section-8.1">Section 8.1</a> -<br> -Windows 95: -<br> -Windows NT: -<br> -<code><nobr>write</nobr></code>: -<a href="nasmdoc7.html#section-7.9.2">Section 7.9.2</a> -<br> -writing operating systems: -<a href="nasmdo10.html#section-10.1">Section 10.1</a> -<br> -<code><nobr>WRT</nobr></code>: -<a href="nasmdoc3.html#section-3.6">Section 3.6</a>, -<a href="nasmdoc7.html#section-7.4">Section 7.4</a>, -<a href="nasmdoc7.html#section-7.9.3">Section 7.9.3</a>, -<a href="nasmdoc7.html#section-7.9.4">Section 7.9.4</a>, -<a href="nasmdoc7.html#section-7.11">Section 7.11</a> -<br> -<code><nobr>WRT ..got</nobr></code>: -<a href="nasmdoc9.html#section-9.2.3">Section 9.2.3</a> -<br> -<code><nobr>WRT ..gotoff</nobr></code>: -<a href="nasmdoc9.html#section-9.2.2">Section 9.2.2</a> -<br> -<code><nobr>WRT ..gotpc</nobr></code>: -<a href="nasmdoc9.html#section-9.2.1">Section 9.2.1</a> -<br> -<code><nobr>WRT ..plt</nobr></code>: -<a href="nasmdoc9.html#section-9.2.5">Section 9.2.5</a> -<br> -<code><nobr>WRT ..sym</nobr></code>: -<a href="nasmdoc9.html#section-9.2.4">Section 9.2.4</a> -<br> -WWW page: -<br> -<code><nobr>www.cpan.org</nobr></code>: -<a href="nasmdoc1.html#section-1.3.1">Section 1.3.1</a> -<br> -<code><nobr>www.delorie.com</nobr></code>: -<a href="nasmdoc8.html#section-8.1.1">Section 8.1.1</a> -<br> -<code><nobr>www.pcorner.com</nobr></code>: -<a href="nasmdoc8.html#section-8.1.1">Section 8.1.1</a> -<br> -<code><nobr>-X</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.13">Section 2.1.13</a> -<br> -<code><nobr>x2ftp.oulu.fi</nobr></code>: -<a href="nasmdoc8.html#section-8.1.1">Section 8.1.1</a> -<br> -<code><nobr>%xdefine</nobr></code>: -<a href="nasmdoc4.html#section-4.1.2">Section 4.1.2</a> -<br> -<code><nobr>-y</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.26">Section 2.1.26</a> -<br> -<code><nobr>-Z</nobr></code> option: -<a href="nasmdoc2.html#section-2.1.14">Section 2.1.14</a> -<br> -<p align=center><a href="nasmdoc0.html">Contents</a> -</body></html> diff --git a/doc/info/nasm.info b/doc/info/nasm.info deleted file mode 100644 index 64ae230..0000000 --- a/doc/info/nasm.info +++ /dev/null @@ -1,398 +0,0 @@ -This is nasm.info, produced by makeinfo version 4.13 from nasmdoc.texi. - -INFO-DIR-SECTION Programming -START-INFO-DIR-ENTRY -* NASM: (nasm). The Netwide Assembler for x86. -END-INFO-DIR-ENTRY - - This file documents NASM, the Netwide Assembler: an assembler -targetting the Intel x86 series of processors, with portable source. - - Copyright 1996-2009 The NASM Development Team - - This document is redistributable under the license given in the file -"COPYING" distributed in the NASM archive. - - -Indirect: -nasm.info-1: 506 -nasm.info-2: 299030 -nasm.info-3: 597272 - -Tag Table: -(Indirect) -Node: Top506 -Node: Chapter 11340 -Node: Section 1.11573 -Node: Section 1.1.12288 -Node: Section 1.1.24165 -Node: Section 1.25885 -Node: Section 1.36714 -Node: Section 1.3.16961 -Node: Section 1.3.28939 -Node: Chapter 210269 -Node: Section 2.110501 -Node: Section 2.1.113922 -Node: Section 2.1.215572 -Node: Section 2.1.316326 -Node: Section 2.1.417278 -Node: Section 2.1.517670 -Node: Section 2.1.618127 -Node: Section 2.1.718501 -Node: Section 2.1.819097 -Node: Section 2.1.919462 -Node: Section 2.1.1019895 -Node: Section 2.1.1120315 -Node: Section 2.1.1221351 -Node: Section 2.1.1321902 -Node: Section 2.1.1423151 -Node: Section 2.1.1524104 -Node: Section 2.1.1624609 -Node: Section 2.1.1726221 -Node: Section 2.1.1826794 -Node: Section 2.1.1927652 -Node: Section 2.1.2028346 -Node: Section 2.1.2129324 -Node: Section 2.1.2229870 -Node: Section 2.1.2331394 -Node: Section 2.1.2432358 -Node: Section 2.1.2535726 -Node: Section 2.1.2636094 -Node: Section 2.1.2736730 -Node: Section 2.1.2837190 -Node: Section 2.238528 -Node: Section 2.2.139321 -Node: Section 2.2.239931 -Node: Section 2.2.342292 -Node: Section 2.2.443155 -Node: Section 2.2.543559 -Node: Section 2.2.644319 -Node: Section 2.2.744947 -Node: Chapter 345799 -Node: Section 3.146276 -Node: Section 3.250446 -Node: Section 3.2.151310 -Node: Section 3.2.252567 -Node: Section 3.2.353598 -Node: Section 3.2.454561 -Node: Section 3.2.555471 -Node: Section 3.356970 -Node: Section 3.460459 -Node: Section 3.4.160957 -Node: Section 3.4.262817 -Node: Section 3.4.364681 -Node: Section 3.4.465383 -Node: Section 3.4.566694 -Node: Section 3.4.667390 -Node: Section 3.4.771120 -Node: Section 3.571682 -Node: Section 3.5.172876 -Node: Section 3.5.273207 -Node: Section 3.5.373418 -Node: Section 3.5.473629 -Node: Section 3.5.574128 -Node: Section 3.5.674417 -Node: Section 3.5.775167 -Node: Section 3.675754 -Node: Section 3.777833 -Node: Section 3.878736 -Node: Section 3.980161 -Node: Chapter 482743 -Node: Section 4.183935 -Node: Section 4.1.184549 -Node: Section 4.1.287390 -Node: Section 4.1.388934 -Node: Section 4.1.489843 -Node: Section 4.1.591349 -Node: Section 4.1.692319 -Node: Section 4.1.792959 -Node: Section 4.1.894230 -Node: Section 4.1.994874 -Node: Section 4.295336 -Node: Section 4.2.196105 -Node: Section 4.2.296681 -Node: Section 4.2.397368 -Node: Section 4.398741 -Node: Section 4.3.1100959 -Node: Section 4.3.2101752 -Node: Section 4.3.3103123 -Node: Section 4.3.4104286 -Node: Section 4.3.5106612 -Node: Section 4.3.6109310 -Node: Section 4.3.7109896 -Node: Section 4.3.8112639 -Node: Section 4.3.9114996 -Node: Section 4.3.10116387 -Node: Section 4.3.11117234 -Node: Section 4.3.12117992 -Node: Section 4.4118457 -Node: Section 4.4.1119961 -Node: Section 4.4.2121062 -Node: Section 4.4.3122377 -Node: Section 4.4.4123000 -Node: Section 4.4.5124560 -Node: Section 4.4.6125561 -Node: Section 4.4.7127864 -Node: Section 4.4.8128682 -Node: Section 4.5129130 -Node: Section 4.6130826 -Node: Section 4.6.1131288 -Node: Section 4.6.2132695 -Node: Section 4.6.3133293 -Node: Section 4.6.4134133 -Node: Section 4.7135063 -Node: Section 4.7.1136233 -Node: Section 4.7.2137224 -Node: Section 4.7.3138261 -Node: Section 4.7.4138778 -Node: Section 4.7.5139538 -Node: Section 4.8142039 -Node: Section 4.8.1142600 -Node: Section 4.8.2143809 -Node: Section 4.8.3145438 -Node: Section 4.9147193 -Node: Section 4.10149062 -Node: Section 4.10.1149732 -Node: Section 4.10.2151046 -Node: Section 4.11151773 -Node: Section 4.11.1153179 -Node: Section 4.11.2153878 -Node: Section 4.11.3154703 -Node: Section 4.11.4155133 -Node: Section 4.11.5156498 -Node: Section 4.11.6157004 -Node: Section 4.11.7157505 -Node: Section 4.11.8159693 -Node: Section 4.11.9160294 -Node: Section 4.11.10160911 -Node: Section 4.11.11163932 -Node: Section 4.11.12165536 -Node: Chapter 5168133 -Node: Section 5.1168785 -Node: Section 5.2169475 -Node: Chapter 6171420 -Node: Section 6.1173000 -Node: Section 6.1.1175976 -Node: Section 6.2176296 -Node: Section 6.3177453 -Node: Section 6.3.1178524 -Node: Section 6.4180364 -Node: Section 6.5182293 -Node: Section 6.6183787 -Node: Section 6.7184971 -Node: Section 6.8186170 -Node: Section 6.9187418 -Node: Chapter 7188427 -Node: Section 7.1190123 -Node: Section 7.1.1191429 -Node: Section 7.1.2192457 -Node: Section 7.1.3193198 -Node: Section 7.1.4194959 -Node: Section 7.2195436 -Node: Section 7.3195980 -Node: Section 7.4196548 -Node: Section 7.4.1199011 -Node: Section 7.4.2201987 -Node: Section 7.4.3203656 -Node: Section 7.4.4204379 -Node: Section 7.4.5205355 -Node: Section 7.4.6207319 -Node: Section 7.4.7207871 -Node: Section 7.4.8209425 -Node: Section 7.5211274 -Node: Section 7.5.1212425 -Node: Section 7.5.2214883 -Node: Section 7.6219189 -Node: Section 7.6.1219880 -Node: Section 7.6.2222750 -Node: Section 7.7231628 -Node: Section 7.8232143 -Node: Section 7.9232570 -Node: Section 7.9.1233574 -Node: Section 7.9.2234090 -Node: Section 7.9.3236574 -Node: Section 7.9.4239580 -Node: Section 7.9.5240492 -Node: Section 7.9.6242295 -Node: Section 7.9.7242944 -Node: Section 7.9.8243460 -Node: Section 7.10243838 -Node: Section 7.11244741 -Node: Section 7.12245917 -Node: Section 7.13246833 -Node: Section 7.13.1248187 -Node: Section 7.13.2248663 -Node: Section 7.13.3249368 -Node: Section 7.13.4250315 -Node: Section 7.14251192 -Node: Chapter 8253473 -Node: Section 8.1254227 -Node: Section 8.1.1255319 -Node: Section 8.1.2258921 -Node: Section 8.2261299 -Node: Section 8.2.1261805 -Node: Section 8.2.2263314 -Node: Section 8.3264893 -Node: Section 8.4265899 -Node: Section 8.4.1266570 -Node: Section 8.4.2268084 -Node: Section 8.4.3271540 -Node: Section 8.4.4278288 -Node: Section 8.4.5280465 -Node: Section 8.5283090 -Node: Section 8.5.1284827 -Node: Section 8.5.2289081 -Node: Section 8.5.3290107 -Node: Chapter 9291435 -Node: Section 9.1292896 -Node: Section 9.1.1293491 -Node: Section 9.1.2294359 -Node: Section 9.1.3299030 -Node: Section 9.1.4301283 -Node: Section 9.2302862 -Node: Section 9.2.1305149 -Node: Section 9.2.2307928 -Node: Section 9.2.3309212 -Node: Section 9.2.4310327 -Node: Section 9.2.5312918 -Node: Section 9.2.6313825 -Node: Chapter 10314685 -Node: Section 10.1315390 -Node: Section 10.2317375 -Node: Section 10.3319959 -Node: Chapter 11322252 -Node: Section 11.1324207 -Node: Section 11.2325177 -Node: Section 11.3327200 -Node: Section 11.4328551 -Node: Chapter 12329634 -Node: Section 12.1330070 -Node: Section 12.1.1330398 -Node: Section 12.1.2331166 -Node: Section 12.1.3332323 -Node: Section 12.1.4333493 -Node: Section 12.2334851 -Node: Appendix A339510 -Node: Section A.1339830 -Node: Section A.2340543 -Node: Section A.3340876 -Node: Section A.3.1341789 -Node: Section A.3.2342678 -Node: Section A.3.3344784 -Node: Section A.3.4347578 -Node: Section A.4348326 -Node: Appendix B348762 -Node: Section B.1348948 -Node: Section B.1.1351150 -Node: Section B.1.2351483 -Node: Section B.1.3424168 -Node: Section B.1.4429304 -Node: Section B.1.5429647 -Node: Section B.1.6430065 -Node: Section B.1.7430506 -Node: Section B.1.8431638 -Node: Section B.1.9432131 -Node: Section B.1.10432831 -Node: Section B.1.11439728 -Node: Section B.1.12445350 -Node: Section B.1.13446171 -Node: Section B.1.14447189 -Node: Section B.1.15447629 -Node: Section B.1.16449575 -Node: Section B.1.17450064 -Node: Section B.1.18450411 -Node: Section B.1.19453720 -Node: Section B.1.20454625 -Node: Section B.1.21454823 -Node: Section B.1.22455137 -Node: Section B.1.23455651 -Node: Section B.1.24456181 -Node: Section B.1.25456737 -Node: Section B.1.26497504 -Node: Section B.1.27498034 -Node: Section B.1.28498591 -Node: Section B.1.29509939 -Node: Section B.1.30510657 -Node: Section B.1.31511523 -Node: Section B.1.32520363 -Node: Appendix C531399 -Node: Section C.1531662 -Node: Section C.1.1532242 -Node: Section C.1.2533796 -Node: Section C.1.3535031 -Node: Section C.1.4536766 -Node: Section C.1.5536987 -Node: Section C.1.6537982 -Node: Section C.1.7540282 -Node: Section C.1.8540651 -Node: Section C.1.9542095 -Node: Section C.1.10543237 -Node: Section C.1.11543839 -Node: Section C.2545564 -Node: Section C.2.1547911 -Node: Section C.2.2548336 -Node: Section C.2.3549349 -Node: Section C.2.4550042 -Node: Section C.2.5550581 -Node: Section C.2.6551111 -Node: Section C.2.7551952 -Node: Section C.2.8552881 -Node: Section C.2.9553540 -Node: Section C.2.10554127 -Node: Section C.2.11554863 -Node: Section C.2.12555186 -Node: Section C.2.13555393 -Node: Section C.2.14555706 -Node: Section C.2.15555953 -Node: Section C.2.16556156 -Node: Section C.2.17556364 -Node: Section C.2.18556621 -Node: Section C.2.19556820 -Node: Section C.2.20557001 -Node: Section C.2.21557182 -Node: Section C.2.22557370 -Node: Section C.2.23557561 -Node: Section C.2.24557771 -Node: Section C.2.25557974 -Node: Section C.2.26558187 -Node: Section C.2.27558366 -Node: Section C.2.28558548 -Node: Section C.2.29558792 -Node: Section C.2.30558996 -Node: Section C.2.31559178 -Node: Section C.2.32559842 -Node: Section C.2.33560152 -Node: Section C.2.34561495 -Node: Section C.2.35561944 -Node: Section C.2.36562245 -Node: Section C.2.37563300 -Node: Section C.2.38563537 -Node: Section C.2.39563868 -Node: Section C.2.40565946 -Node: Section C.2.41573822 -Node: Section C.2.42574310 -Node: Section C.2.43575038 -Node: Section C.2.44575971 -Node: Section C.2.45577069 -Node: Section C.2.46578569 -Node: Section C.2.47578863 -Node: Section C.2.48579191 -Node: Section C.2.49580498 -Node: Section C.2.50580915 -Node: Section C.2.51581846 -Node: Section C.2.52582760 -Node: Section C.2.53583882 -Node: Section C.2.54584288 -Node: Section C.2.55584619 -Node: Section C.3587873 -Node: Section C.3.1588473 -Node: Section C.3.2590132 -Node: Section C.3.3597272 -Node: Section C.3.4601832 -Node: Section C.3.5603782 -Node: Section C.3.6604873 -Node: Section C.3.7606084 -Node: Section C.3.8607145 -Node: Index607442 - -End Tag Table diff --git a/doc/info/nasm.info-1 b/doc/info/nasm.info-1 deleted file mode 100644 index d1896f8..0000000 --- a/doc/info/nasm.info-1 +++ /dev/null @@ -1,7484 +0,0 @@ -This is nasm.info, produced by makeinfo version 4.13 from nasmdoc.texi. - -INFO-DIR-SECTION Programming -START-INFO-DIR-ENTRY -* NASM: (nasm). The Netwide Assembler for x86. -END-INFO-DIR-ENTRY - - This file documents NASM, the Netwide Assembler: an assembler -targetting the Intel x86 series of processors, with portable source. - - Copyright 1996-2009 The NASM Development Team - - This document is redistributable under the license given in the file -"COPYING" distributed in the NASM archive. - - -File: nasm.info, Node: Top, Next: Chapter 1, Prev: (dir), Up: (dir) - -The Netwide Assembler for x86 -***************************** - -This file documents NASM, the Netwide Assembler: an assembler -targetting the Intel x86 series of processors, with portable source. - -* Menu: - -* Chapter 1:: Introduction -* Chapter 2:: Running NASM -* Chapter 3:: The NASM Language -* Chapter 4:: The NASM Preprocessor -* Chapter 5:: Standard Macro Packages -* Chapter 6:: Assembler Directives -* Chapter 7:: Output Formats -* Chapter 8:: Writing 16-bit Code (DOS, Windows 3/3.1) -* Chapter 9:: Writing 32-bit Code (Unix, Win32, DJGPP) -* Chapter 10:: Mixing 16 and 32 Bit Code -* Chapter 11:: Writing 64-bit Code (Unix, Win64) -* Chapter 12:: Troubleshooting -* Appendix A:: Ndisasm -* Appendix B:: Instruction List -* Appendix C:: NASM Version History -* Index:: - - -File: nasm.info, Node: Chapter 1, Next: Section 1.1, Prev: Top, Up: Top - -Chapter 1: Introduction -*********************** - -* Menu: - -* Section 1.1:: What Is NASM? -* Section 1.2:: Contact Information -* Section 1.3:: Installation - - -File: nasm.info, Node: Section 1.1, Next: Section 1.1.1, Prev: Chapter 1, Up: Chapter 1 - -1.1. What Is NASM? -================== - -The Netwide Assembler, NASM, is an 80x86 and x86-64 assembler designed -for portability and modularity. It supports a range of object file -formats, including Linux and `*BSD' `a.out', `ELF', `COFF', `Mach-O', -Microsoft 16-bit `OBJ', `Win32' and `Win64'. It will also output plain -binary files. Its syntax is designed to be simple and easy to -understand, similar to Intel's but less complex. It supports all -currently known x86 architectural extensions, and has strong support -for macros. - -* Menu: - -* Section 1.1.1:: Why Yet Another Assembler? -* Section 1.1.2:: License Conditions - - -File: nasm.info, Node: Section 1.1.1, Next: Section 1.1.2, Prev: Section 1.1, Up: Section 1.1 - -1.1.1. Why Yet Another Assembler? ---------------------------------- - -The Netwide Assembler grew out of an idea on `comp.lang.asm.x86' (or -possibly `alt.lang.asm' - I forget which), which was essentially that -there didn't seem to be a good _free_ x86-series assembler around, and -that maybe someone ought to write one. - - * `a86' is good, but not free, and in particular you don't get any - 32- bit capability until you pay. It's DOS only, too. - - * `gas' is free, and ports over to DOS and Unix, but it's not very - good, since it's designed to be a back end to `gcc', which always - feeds it correct code. So its error checking is minimal. Also, its - syntax is horrible, from the point of view of anyone trying to - actually _write_ anything in it. Plus you can't write 16-bit code - in it (properly.) - - * `as86' is specific to Minix and Linux, and (my version at least) - doesn't seem to have much (or any) documentation. - - * `MASM' isn't very good, and it's (was) expensive, and it runs only - under DOS. - - * `TASM' is better, but still strives for MASM compatibility, which - means millions of directives and tons of red tape. And its syntax - is essentially MASM's, with the contradictions and quirks that - entails (although it sorts out some of those by means of Ideal - mode.) It's expensive too. And it's DOS-only. - - So here, for your coding pleasure, is NASM. At present it's still in -prototype stage - we don't promise that it can outperform any of these -assemblers. But please, _please_ send us bug reports, fixes, helpful -information, and anything else you can get your hands on (and thanks to -the many people who've done this already! You all know who you are), -and we'll improve it out of all recognition. Again. - - -File: nasm.info, Node: Section 1.1.2, Next: Section 1.2, Prev: Section 1.1.1, Up: Section 1.1 - -1.1.2. License Conditions -------------------------- - -Please see the file `LICENSE', supplied as part of any NASM -distribution archive, for the license conditions under which you may use -NASM. NASM is now under the so-called 2-clause BSD license, also known -as the simplified BSD license. - - Copyright 1996-2009 the NASM Authors - All rights reserved. - - Redistribution and use in source and binary forms, with or without -modification, are permitted provided that the following conditions are -met: - - * Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in - the documentation and/or other materials provided with the - distribution. - - THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS -"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT -LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A -PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT -OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, -SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT -LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, -DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY -THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT -(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE -OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - - -File: nasm.info, Node: Section 1.2, Next: Section 1.3, Prev: Section 1.1.2, Up: Chapter 1 - -1.2. Contact Information -======================== - -The current version of NASM (since about 0.98.08) is maintained by a -team of developers, accessible through the `nasm-devel' mailing list -(see below for the link). If you want to report a bug, please read -*note Section 12.2:: first. - - NASM has a website at `http://www.nasm.us/'. If it's not there, -google for us! - - New releases, release candidates, and daily development snapshots of -NASM are available from the official web site. - - Announcements are posted to `comp.lang.asm.x86', and to the web site -`http://www.freshmeat.net/'. - - If you want information about the current development status, please -subscribe to the `nasm-devel' email list; see link from the website. - - -File: nasm.info, Node: Section 1.3, Next: Section 1.3.1, Prev: Section 1.2, Up: Chapter 1 - -1.3. Installation -================= - -* Menu: - -* Section 1.3.1:: Installing NASM under MS-DOS or Windows -* Section 1.3.2:: Installing NASM under Unix - - -File: nasm.info, Node: Section 1.3.1, Next: Section 1.3.2, Prev: Section 1.3, Up: Section 1.3 - -1.3.1. Installing NASM under MS-DOS or Windows ----------------------------------------------- - -Once you've obtained the appropriate archive for NASM, -`nasm-XXX-dos.zip' or `nasm-XXX-win32.zip' (where `XXX' denotes the -version number of NASM contained in the archive), unpack it into its -own directory (for example `c:\nasm'). - - The archive will contain a set of executable files: the NASM -executable file `nasm.exe', the NDISASM executable file `ndisasm.exe', -and possibly additional utilities to handle the RDOFF file format. - - The only file NASM needs to run is its own executable, so copy -`nasm.exe' to a directory on your PATH, or alternatively edit -`autoexec.bat' to add the `nasm' directory to your `PATH' (to do that -under Windows XP, go to Start > Control Panel > System > Advanced > -Environment Variables; these instructions may work under other versions -of Windows as well.) - - That's it - NASM is installed. You don't need the nasm directory to -be present to run NASM (unless you've added it to your `PATH'), so you -can delete it if you need to save space; however, you may want to keep -the documentation or test programs. - - If you've downloaded the DOS source archive, `nasm-XXX.zip', the -`nasm' directory will also contain the full NASM source code, and a -selection of Makefiles you can (hopefully) use to rebuild your copy of -NASM from scratch. See the file `INSTALL' in the source archive. - - Note that a number of files are generated from other files by Perl -scripts. Although the NASM source distribution includes these -generated files, you will need to rebuild them (and hence, will need a -Perl interpreter) if you change insns.dat, standard.mac or the -documentation. It is possible future source distributions may not -include these files at all. Ports of Perl for a variety of platforms, -including DOS and Windows, are available from www.cpan.org. - - -File: nasm.info, Node: Section 1.3.2, Next: Chapter 2, Prev: Section 1.3.1, Up: Section 1.3 - -1.3.2. Installing NASM under Unix ---------------------------------- - -Once you've obtained the Unix source archive for NASM, -`nasm-XXX.tar.gz' (where `XXX' denotes the version number of NASM -contained in the archive), unpack it into a directory such as -`/usr/local/src'. The archive, when unpacked, will create its own -subdirectory `nasm-XXX'. - - NASM is an auto-configuring package: once you've unpacked it, `cd' to -the directory it's been unpacked into and type `./configure'. This -shell script will find the best C compiler to use for building NASM and -set up Makefiles accordingly. - - Once NASM has auto-configured, you can type `make' to build the -`nasm' and `ndisasm' binaries, and then `make install' to install them -in `/usr/local/bin' and install the man pages `nasm.1' and `ndisasm.1' -in `/usr/local/man/man1'. Alternatively, you can give options such as -`--prefix' to the configure script (see the file `INSTALL' for more -details), or install the programs yourself. - - NASM also comes with a set of utilities for handling the `RDOFF' -custom object-file format, which are in the `rdoff' subdirectory of the -NASM archive. You can build these with `make rdf' and install them with -`make rdf_install', if you want them. - - -File: nasm.info, Node: Chapter 2, Next: Section 2.1, Prev: Section 1.3.2, Up: Top - -Chapter 2: Running NASM -*********************** - -* Menu: - -* Section 2.1:: NASM Command-Line Syntax -* Section 2.2:: Quick Start for MASM Users - - -File: nasm.info, Node: Section 2.1, Next: Section 2.1.1, Prev: Chapter 2, Up: Chapter 2 - -2.1. NASM Command-Line Syntax -============================= - -To assemble a file, you issue a command of the form - - nasm -f <format> <filename> [-o <output>] - - For example, - - nasm -f elf myfile.asm - - will assemble `myfile.asm' into an `ELF' object file `myfile.o'. And - - nasm -f bin myfile.asm -o myfile.com - - will assemble `myfile.asm' into a raw binary file `myfile.com'. - - To produce a listing file, with the hex codes output from NASM -displayed on the left of the original sources, use the `-l' option to -give a listing file name, for example: - - nasm -f coff myfile.asm -l myfile.lst - - To get further usage instructions from NASM, try typing - - nasm -h - - As `-hf', this will also list the available output file formats, and -what they are. - - If you use Linux but aren't sure whether your system is `a.out' or -`ELF', type - - file nasm - - (in the directory in which you put the NASM binary when you -installed it). If it says something like - - nasm: ELF 32-bit LSB executable i386 (386 and up) Version 1 - - then your system is `ELF', and you should use the option `-f elf' -when you want NASM to produce Linux object files. If it says - - nasm: Linux/i386 demand-paged executable (QMAGIC) - - or something similar, your system is `a.out', and you should use `-f -aout' instead (Linux `a.out' systems have long been obsolete, and are -rare these days.) - - Like Unix compilers and assemblers, NASM is silent unless it goes -wrong: you won't see any output at all, unless it gives error messages. - -* Menu: - -* Section 2.1.1:: The `-o' Option: Specifying the Output File Name -* Section 2.1.2:: The `-f' Option: Specifying the Output File Format -* Section 2.1.3:: The `-l' Option: Generating a Listing File -* Section 2.1.4:: The `-M' Option: Generate Makefile Dependencies -* Section 2.1.5:: The `-MG' Option: Generate Makefile Dependencies -* Section 2.1.6:: The `-MF' Option: Set Makefile Dependency File -* Section 2.1.7:: The `-MD' Option: Assemble and Generate Dependencies -* Section 2.1.8:: The `-MT' Option: Dependency Target Name -* Section 2.1.9:: The `-MQ' Option: Dependency Target Name (Quoted) -* Section 2.1.10:: The `-MP' Option: Emit phony targets -* Section 2.1.11:: The `-F' Option: Selecting a Debug Information Format -* Section 2.1.12:: The `-g' Option: Enabling Debug Information. -* Section 2.1.13:: The `-X' Option: Selecting an Error Reporting Format -* Section 2.1.14:: The `-Z' Option: Send Errors to a File -* Section 2.1.15:: The `-s' Option: Send Errors to `stdout' -* Section 2.1.16:: The `-i' Option: Include File Search Directories -* Section 2.1.17:: The `-p' Option: Pre-Include a File -* Section 2.1.18:: The `-d' Option: Pre-Define a Macro -* Section 2.1.19:: The `-u' Option: Undefine a Macro -* Section 2.1.20:: The `-E' Option: Preprocess Only -* Section 2.1.21:: The `-a' Option: Don't Preprocess At All -* Section 2.1.22:: The `-O' Option: Specifying Multipass Optimization -* Section 2.1.23:: The `-t' Option: Enable TASM Compatibility Mode -* Section 2.1.24:: The `-w' and `-W' Options: Enable or Disable Assembly Warnings -* Section 2.1.25:: The `-v' Option: Display Version Info -* Section 2.1.26:: The `-y' Option: Display Available Debug Info Formats -* Section 2.1.27:: The `--prefix' and `--postfix' Options. -* Section 2.1.28:: The `NASMENV' Environment Variable - - -File: nasm.info, Node: Section 2.1.1, Next: Section 2.1.2, Prev: Section 2.1, Up: Section 2.1 - -2.1.1. The `-o' Option: Specifying the Output File Name -------------------------------------------------------- - -NASM will normally choose the name of your output file for you; -precisely how it does this is dependent on the object file format. For -Microsoft object file formats (`obj', `win32' and `win64'), it will -remove the `.asm' extension (or whatever extension you like to use - -NASM doesn't care) from your source file name and substitute `.obj'. -For Unix object file formats (`aout', `as86', `coff', `elf32', `elf64', -`ieee', `macho32' and `macho64') it will substitute `.o'. For `dbg', -`rdf', `ith' and `srec', it will use `.dbg', `.rdf', `.ith' and -`.srec', respectively, and for the `bin' format it will simply remove -the extension, so that `myfile.asm' produces the output file `myfile'. - - If the output file already exists, NASM will overwrite it, unless it -has the same name as the input file, in which case it will give a -warning and use `nasm.out' as the output file name instead. - - For situations in which this behaviour is unacceptable, NASM -provides the `-o' command-line option, which allows you to specify your -desired output file name. You invoke `-o' by following it with the name -you wish for the output file, either with or without an intervening -space. For example: - - nasm -f bin program.asm -o program.com - nasm -f bin driver.asm -odriver.sys - - Note that this is a small o, and is different from a capital O , -which is used to specify the number of optimisation passes required. See -*note Section 2.1.22::. - - -File: nasm.info, Node: Section 2.1.2, Next: Section 2.1.3, Prev: Section 2.1.1, Up: Section 2.1 - -2.1.2. The `-f' Option: Specifying the Output File Format ---------------------------------------------------------- - -If you do not supply the `-f' option to NASM, it will choose an output -file format for you itself. In the distribution versions of NASM, the -default is always `bin'; if you've compiled your own copy of NASM, you -can redefine `OF_DEFAULT' at compile time and choose what you want the -default to be. - - Like `-o', the intervening space between `-f' and the output file -format is optional; so `-f elf' and `-felf' are both valid. - - A complete list of the available output file formats can be given by -issuing the command `nasm -hf'. - - -File: nasm.info, Node: Section 2.1.3, Next: Section 2.1.4, Prev: Section 2.1.2, Up: Section 2.1 - -2.1.3. The `-l' Option: Generating a Listing File -------------------------------------------------- - -If you supply the `-l' option to NASM, followed (with the usual -optional space) by a file name, NASM will generate a source-listing file -for you, in which addresses and generated code are listed on the left, -and the actual source code, with expansions of multi-line macros -(except those which specifically request no expansion in source -listings: see *note Section 4.3.10::) on the right. For example: - - nasm -f elf myfile.asm -l myfile.lst - - If a list file is selected, you may turn off listing for a section -of your source with `[list -]', and turn it back on with `[list +]', -(the default, obviously). There is no "user form" (without the -brackets). This can be used to list only sections of interest, avoiding -excessively long listings. - - -File: nasm.info, Node: Section 2.1.4, Next: Section 2.1.5, Prev: Section 2.1.3, Up: Section 2.1 - -2.1.4. The `-M' Option: Generate Makefile Dependencies ------------------------------------------------------- - -This option can be used to generate makefile dependencies on stdout. -This can be redirected to a file for further processing. For example: - - nasm -M myfile.asm > myfile.dep - - -File: nasm.info, Node: Section 2.1.5, Next: Section 2.1.6, Prev: Section 2.1.4, Up: Section 2.1 - -2.1.5. The `-MG' Option: Generate Makefile Dependencies -------------------------------------------------------- - -This option can be used to generate makefile dependencies on stdout. -This differs from the `-M' option in that if a nonexisting file is -encountered, it is assumed to be a generated file and is added to the -dependency list without a prefix. - - -File: nasm.info, Node: Section 2.1.6, Next: Section 2.1.7, Prev: Section 2.1.5, Up: Section 2.1 - -2.1.6. The `-MF' Option: Set Makefile Dependency File ------------------------------------------------------ - -This option can be used with the `-M' or `-MG' options to send the -output to a file, rather than to stdout. For example: - - nasm -M -MF myfile.dep myfile.asm - - -File: nasm.info, Node: Section 2.1.7, Next: Section 2.1.8, Prev: Section 2.1.6, Up: Section 2.1 - -2.1.7. The `-MD' Option: Assemble and Generate Dependencies ------------------------------------------------------------ - -The `-MD' option acts as the combination of the `-M' and `-MF' options -(i.e. a filename has to be specified.) However, unlike the `-M' or -`-MG' options, `-MD' does _not_ inhibit the normal operation of the -assembler. Use this to automatically generate updated dependencies with -every assembly session. For example: - - nasm -f elf -o myfile.o -MD myfile.dep myfile.asm - - -File: nasm.info, Node: Section 2.1.8, Next: Section 2.1.9, Prev: Section 2.1.7, Up: Section 2.1 - -2.1.8. The `-MT' Option: Dependency Target Name ------------------------------------------------ - -The `-MT' option can be used to override the default name of the -dependency target. This is normally the same as the output filename, -specified by the `-o' option. - - -File: nasm.info, Node: Section 2.1.9, Next: Section 2.1.10, Prev: Section 2.1.8, Up: Section 2.1 - -2.1.9. The `-MQ' Option: Dependency Target Name (Quoted) --------------------------------------------------------- - -The `-MQ' option acts as the `-MT' option, except it tries to quote -characters that have special meaning in Makefile syntax. This is not -foolproof, as not all characters with special meaning are quotable in -Make. - - -File: nasm.info, Node: Section 2.1.10, Next: Section 2.1.11, Prev: Section 2.1.9, Up: Section 2.1 - -2.1.10. The `-MP' Option: Emit phony targets --------------------------------------------- - -When used with any of the dependency generation options, the `-MP' -option causes NASM to emit a phony target without dependencies for each -header file. This prevents Make from complaining if a header file has -been removed. - - -File: nasm.info, Node: Section 2.1.11, Next: Section 2.1.12, Prev: Section 2.1.10, Up: Section 2.1 - -2.1.11. The `-F' Option: Selecting a Debug Information Format -------------------------------------------------------------- - -This option is used to select the format of the debug information -emitted into the output file, to be used by a debugger (or _will_ be). -Prior to version 2.03.01, the use of this switch did _not_ enable -output of the selected debug info format. Use `-g', see *note Section -2.1.12::, to enable output. Versions 2.03.01 and later automatically -enable `-g' if `-F' is specified. - - A complete list of the available debug file formats for an output -format can be seen by issuing the command `nasm -f <format> -y'. Not all -output formats currently support debugging output. See *note Section -2.1.26::. - - This should not be confused with the `-f dbg' output format option -which is not built into NASM by default. For information on how to -enable it when building from the sources, see *note Section 7.14::. - - -File: nasm.info, Node: Section 2.1.12, Next: Section 2.1.13, Prev: Section 2.1.11, Up: Section 2.1 - -2.1.12. The `-g' Option: Enabling Debug Information. ----------------------------------------------------- - -This option can be used to generate debugging information in the -specified format. See *note Section 2.1.11::. Using `-g' without `-F' -results in emitting debug info in the default format, if any, for the -selected output format. If no debug information is currently -implemented in the selected output format, `-g' is _silently ignored_. - - -File: nasm.info, Node: Section 2.1.13, Next: Section 2.1.14, Prev: Section 2.1.12, Up: Section 2.1 - -2.1.13. The `-X' Option: Selecting an Error Reporting Format ------------------------------------------------------------- - -This option can be used to select an error reporting format for any -error messages that might be produced by NASM. - - Currently, two error reporting formats may be selected. They are the -`-Xvc' option and the `-Xgnu' option. The GNU format is the default and -looks like this: - - filename.asm:65: error: specific error message - - where `filename.asm' is the name of the source file in which the -error was detected, `65' is the source file line number on which the -error was detected, `error' is the severity of the error (this could be -`warning'), and `specific error message' is a more detailed text -message which should help pinpoint the exact problem. - - The other format, specified by `-Xvc' is the style used by Microsoft -Visual C++ and some other programs. It looks like this: - - filename.asm(65) : error: specific error message - - where the only difference is that the line number is in parentheses -instead of being delimited by colons. - - See also the `Visual C++' output format, *note Section 7.5::. - - -File: nasm.info, Node: Section 2.1.14, Next: Section 2.1.15, Prev: Section 2.1.13, Up: Section 2.1 - -2.1.14. The `-Z' Option: Send Errors to a File ----------------------------------------------- - -Under `MS-DOS' it can be difficult (though there are ways) to redirect -the standard-error output of a program to a file. Since NASM usually -produces its warning and error messages on `stderr', this can make it -hard to capture the errors if (for example) you want to load them into -an editor. - - NASM therefore provides the `-Z' option, taking a filename argument -which causes errors to be sent to the specified files rather than -standard error. Therefore you can redirect the errors into a file by -typing - - nasm -Z myfile.err -f obj myfile.asm - - In earlier versions of NASM, this option was called `-E', but it was -changed since `-E' is an option conventionally used for preprocessing -only, with disastrous results. See *note Section 2.1.20::. - - -File: nasm.info, Node: Section 2.1.15, Next: Section 2.1.16, Prev: Section 2.1.14, Up: Section 2.1 - -2.1.15. The `-s' Option: Send Errors to `stdout' ------------------------------------------------- - -The `-s' option redirects error messages to `stdout' rather than -`stderr', so it can be redirected under `MS-DOS'. To assemble the file -`myfile.asm' and pipe its output to the `more' program, you can type: - - nasm -s -f obj myfile.asm | more - - See also the `-Z' option, *note Section 2.1.14::. - - -File: nasm.info, Node: Section 2.1.16, Next: Section 2.1.17, Prev: Section 2.1.15, Up: Section 2.1 - -2.1.16. The `-i' Option: Include File Search Directories --------------------------------------------------------- - -When NASM sees the `%include' or `%pathsearch' directive in a source -file (see *note Section 4.6.1::, *note Section 4.6.2:: or *note Section -3.2.3::), it will search for the given file not only in the current -directory, but also in any directories specified on the command line by -the use of the `-i' option. Therefore you can include files from a macro -library, for example, by typing - - nasm -ic:\macrolib\ -f obj myfile.asm - - (As usual, a space between `-i' and the path name is allowed, and -optional). - - NASM, in the interests of complete source-code portability, does not -understand the file naming conventions of the OS it is running on; the -string you provide as an argument to the `-i' option will be prepended -exactly as written to the name of the include file. Therefore the -trailing backslash in the above example is necessary. Under Unix, a -trailing forward slash is similarly necessary. - - (You can use this to your advantage, if you're really perverse, by -noting that the option `-ifoo' will cause `%include "bar.i"' to search -for the file `foobar.i'...) - - If you want to define a _standard_ include search path, similar to -`/usr/include' on Unix systems, you should place one or more `-i' -directives in the `NASMENV' environment variable (see *note Section -2.1.28::). - - For Makefile compatibility with many C compilers, this option can -also be specified as `-I'. - - -File: nasm.info, Node: Section 2.1.17, Next: Section 2.1.18, Prev: Section 2.1.16, Up: Section 2.1 - -2.1.17. The `-p' Option: Pre-Include a File -------------------------------------------- - -NASM allows you to specify files to be _pre-included_ into your source -file, by the use of the `-p' option. So running - - nasm myfile.asm -p myinc.inc - - is equivalent to running `nasm myfile.asm' and placing the directive -`%include "myinc.inc"' at the start of the file. - - For consistency with the `-I', `-D' and `-U' options, this option -can also be specified as `-P'. - - -File: nasm.info, Node: Section 2.1.18, Next: Section 2.1.19, Prev: Section 2.1.17, Up: Section 2.1 - -2.1.18. The `-d' Option: Pre-Define a Macro -------------------------------------------- - -Just as the `-p' option gives an alternative to placing `%include' -directives at the start of a source file, the `-d' option gives an -alternative to placing a `%define' directive. You could code - - nasm myfile.asm -dFOO=100 - - as an alternative to placing the directive - - %define FOO 100 - - at the start of the file. You can miss off the macro value, as well: -the option `-dFOO' is equivalent to coding `%define FOO'. This form of -the directive may be useful for selecting assembly-time options which -are then tested using `%ifdef', for example `-dDEBUG'. - - For Makefile compatibility with many C compilers, this option can -also be specified as `-D'. - - -File: nasm.info, Node: Section 2.1.19, Next: Section 2.1.20, Prev: Section 2.1.18, Up: Section 2.1 - -2.1.19. The `-u' Option: Undefine a Macro ------------------------------------------ - -The `-u' option undefines a macro that would otherwise have been pre- -defined, either automatically or by a `-p' or `-d' option specified -earlier on the command lines. - - For example, the following command line: - - nasm myfile.asm -dFOO=100 -uFOO - - would result in `FOO' _not_ being a predefined macro in the program. -This is useful to override options specified at a different point in a -Makefile. - - For Makefile compatibility with many C compilers, this option can -also be specified as `-U'. - - -File: nasm.info, Node: Section 2.1.20, Next: Section 2.1.21, Prev: Section 2.1.19, Up: Section 2.1 - -2.1.20. The `-E' Option: Preprocess Only ----------------------------------------- - -NASM allows the preprocessor to be run on its own, up to a point. Using -the `-E' option (which requires no arguments) will cause NASM to -preprocess its input file, expand all the macro references, remove all -the comments and preprocessor directives, and print the resulting file -on standard output (or save it to a file, if the `-o' option is also -used). - - This option cannot be applied to programs which require the -preprocessor to evaluate expressions which depend on the values of -symbols: so code such as - - %assign tablesize ($-tablestart) - - will cause an error in preprocess-only mode. - - For compatiblity with older version of NASM, this option can also be -written `-e'. `-E' in older versions of NASM was the equivalent of the -current `-Z' option, *note Section 2.1.14::. - - -File: nasm.info, Node: Section 2.1.21, Next: Section 2.1.22, Prev: Section 2.1.20, Up: Section 2.1 - -2.1.21. The `-a' Option: Don't Preprocess At All ------------------------------------------------- - -If NASM is being used as the back end to a compiler, it might be -desirable to suppress preprocessing completely and assume the compiler -has already done it, to save time and increase compilation speeds. The -`-a' option, requiring no argument, instructs NASM to replace its -powerful preprocessor with a stub preprocessor which does nothing. - - -File: nasm.info, Node: Section 2.1.22, Next: Section 2.1.23, Prev: Section 2.1.21, Up: Section 2.1 - -2.1.22. The `-O' Option: Specifying Multipass Optimization ----------------------------------------------------------- - -NASM defaults to not optimizing operands which can fit into a signed -byte. This means that if you want the shortest possible object code, -you have to enable optimization. - - Using the `-O' option, you can tell NASM to carry out different -levels of optimization. The syntax is: - - * `-O0': No optimization. All operands take their long forms, if a - short form is not specified, except conditional jumps. This is - intended to match NASM 0.98 behavior. - - * `-O1': Minimal optimization. As above, but immediate operands which - will fit in a signed byte are optimized, unless the long form is - specified. Conditional jumps default to the long form unless - otherwise specified. - - * `-Ox' (where `x' is the actual letter `x'): Multipass - optimization. Minimize branch offsets and signed immediate bytes, - overriding size specification unless the `strict' keyword has been - used (see *note Section 3.7::). For compatability with earlier - releases, the letter `x' may also be any number greater than one. - This number has no effect on the actual number of passes. - - The `-Ox' mode is recommended for most uses. - - Note that this is a capital `O', and is different from a small `o', -which is used to specify the output file name. See *note Section -2.1.1::. - - -File: nasm.info, Node: Section 2.1.23, Next: Section 2.1.24, Prev: Section 2.1.22, Up: Section 2.1 - -2.1.23. The `-t' Option: Enable TASM Compatibility Mode -------------------------------------------------------- - -NASM includes a limited form of compatibility with Borland's `TASM'. -When NASM's `-t' option is used, the following changes are made: - - * local labels may be prefixed with `@@' instead of `.' - - * size override is supported within brackets. In TASM compatible - mode, a size override inside square brackets changes the size of - the operand, and not the address type of the operand as it does in - NASM syntax. E.g. `mov eax,[DWORD val]' is valid syntax in TASM - compatibility mode. Note that you lose the ability to override the - default address type for the instruction. - - * unprefixed forms of some directives supported (`arg', `elif', - `else', `endif', `if', `ifdef', `ifdifi', `ifndef', `include', - `local') - - -File: nasm.info, Node: Section 2.1.24, Next: Section 2.1.25, Prev: Section 2.1.23, Up: Section 2.1 - -2.1.24. The `-w' and `-W' Options: Enable or Disable Assembly Warnings ----------------------------------------------------------------------- - -NASM can observe many conditions during the course of assembly which are -worth mentioning to the user, but not a sufficiently severe error to -justify NASM refusing to generate an output file. These conditions are -reported like errors, but come up with the word `warning' before the -message. Warnings do not prevent NASM from generating an output file and -returning a success status to the operating system. - - Some conditions are even less severe than that: they are only -sometimes worth mentioning to the user. Therefore NASM supports the `-w' -command-line option, which enables or disables certain classes of -assembly warning. Such warning classes are described by a name, for -example `orphan-labels'; you can enable warnings of this class by the -command- line option `-w+orphan-labels' and disable it by -`-w-orphan-labels'. - - The suppressible warning classes are: - - * `macro-params' covers warnings about multi-line macros being - invoked with the wrong number of parameters. This warning class is - enabled by default; see *note Section 4.3.2:: for an example of - why you might want to disable it. - - * `macro-selfref' warns if a macro references itself. This warning - class is disabled by default. - - * `macro-defaults' warns when a macro has more default parameters - than optional parameters. This warning class is enabled by - default; see *note Section 4.3.5:: for why you might want to - disable it. - - * `orphan-labels' covers warnings about source lines which contain no - instruction but define a label without a trailing colon. NASM - warns about this somewhat obscure condition by default; see *note - Section 3.1:: for more information. - - * `number-overflow' covers warnings about numeric constants which - don't fit in 64 bits. This warning class is enabled by default. - - * `gnu-elf-extensions' warns if 8-bit or 16-bit relocations are used - in `-f elf' format. The GNU extensions allow this. This warning - class is disabled by default. - - * `float-overflow' warns about floating point overflow. Enabled by - default. - - * `float-denorm' warns about floating point denormals. Disabled by - default. - - * `float-underflow' warns about floating point underflow. Disabled by - default. - - * `float-toolong' warns about too many digits in floating-point - numbers. Enabled by default. - - * `user' controls `%warning' directives (see *note Section 4.9::). - Enabled by default. - - * `error' causes warnings to be treated as errors. Disabled by - default. - - * `all' is an alias for _all_ suppressible warning classes (not - including `error'). Thus, `-w+all' enables all available warnings. - - In addition, you can set warning classes across sections. Warning -classes may be enabled with `[warning +warning-name]', disabled with -`[warning -warning-name]' or reset to their original value with -`[warning *warning-name]'. No "user form" (without the brackets) exists. - - Since version 2.00, NASM has also supported the gcc-like syntax -`-Wwarning' and `-Wno-warning' instead of `-w+warning' and -`-w-warning', respectively. - - -File: nasm.info, Node: Section 2.1.25, Next: Section 2.1.26, Prev: Section 2.1.24, Up: Section 2.1 - -2.1.25. The `-v' Option: Display Version Info ---------------------------------------------- - -Typing `NASM -v' will display the version of NASM which you are using, -and the date on which it was compiled. - - You will need the version number if you report a bug. - - -File: nasm.info, Node: Section 2.1.26, Next: Section 2.1.27, Prev: Section 2.1.25, Up: Section 2.1 - -2.1.26. The `-y' Option: Display Available Debug Info Formats -------------------------------------------------------------- - -Typing `nasm -f <option> -y' will display a list of the available debug -info formats for the given output format. The default format is -indicated by an asterisk. For example: - - nasm -f elf -y - - valid debug formats for 'elf32' output format are - ('*' denotes default): - * stabs ELF32 (i386) stabs debug format for Linux - dwarf elf32 (i386) dwarf debug format for Linux - - -File: nasm.info, Node: Section 2.1.27, Next: Section 2.1.28, Prev: Section 2.1.26, Up: Section 2.1 - -2.1.27. The `--prefix' and `--postfix' Options. ------------------------------------------------ - -The `--prefix' and `--postfix' options prepend or append (respectively) -the given argument to all `global' or `extern' variables. E.g. -`--prefix _' will prepend the underscore to all global and external -variables, as C sometimes (but not always) likes it. - - -File: nasm.info, Node: Section 2.1.28, Next: Section 2.2, Prev: Section 2.1.27, Up: Section 2.1 - -2.1.28. The `NASMENV' Environment Variable ------------------------------------------- - -If you define an environment variable called `NASMENV', the program -will interpret it as a list of extra command-line options, which are -processed before the real command line. You can use this to define -standard search directories for include files, by putting `-i' options -in the `NASMENV' variable. - - The value of the variable is split up at white space, so that the -value `-s -ic:\nasmlib\' will be treated as two separate options. -However, that means that the value `-dNAME="my name"' won't do what you -might want, because it will be split at the space and the NASM -command-line processing will get confused by the two nonsensical words -`-dNAME="my' and `name"'. - - To get round this, NASM provides a feature whereby, if you begin the -`NASMENV' environment variable with some character that isn't a minus -sign, then NASM will treat this character as the separator character for -options. So setting the `NASMENV' variable to the value -`!-s!-ic:\nasmlib\' is equivalent to setting it to `-s -ic:\nasmlib\', -but `!-dNAME="my name"' will work. - - This environment variable was previously called `NASM'. This was -changed with version 0.98.31. - - -File: nasm.info, Node: Section 2.2, Next: Section 2.2.1, Prev: Section 2.1.28, Up: Chapter 2 - -2.2. Quick Start for MASM Users -=============================== - -If you're used to writing programs with MASM, or with TASM in MASM- -compatible (non-Ideal) mode, or with `a86', this section attempts to -outline the major differences between MASM's syntax and NASM's. If -you're not already used to MASM, it's probably worth skipping this -section. - -* Menu: - -* Section 2.2.1:: NASM Is Case-Sensitive -* Section 2.2.2:: NASM Requires Square Brackets For Memory References -* Section 2.2.3:: NASM Doesn't Store Variable Types -* Section 2.2.4:: NASM Doesn't `ASSUME' -* Section 2.2.5:: NASM Doesn't Support Memory Models -* Section 2.2.6:: Floating-Point Differences -* Section 2.2.7:: Other Differences - - -File: nasm.info, Node: Section 2.2.1, Next: Section 2.2.2, Prev: Section 2.2, Up: Section 2.2 - -2.2.1. NASM Is Case-Sensitive ------------------------------ - -One simple difference is that NASM is case-sensitive. It makes a -difference whether you call your label `foo', `Foo' or `FOO'. If you're -assembling to `DOS' or `OS/2' `.OBJ' files, you can invoke the -`UPPERCASE' directive (documented in *note Section 7.4::) to ensure -that all symbols exported to other code modules are forced to be upper -case; but even then, _within_ a single module, NASM will distinguish -between labels differing only in case. - - -File: nasm.info, Node: Section 2.2.2, Next: Section 2.2.3, Prev: Section 2.2.1, Up: Section 2.2 - -2.2.2. NASM Requires Square Brackets For Memory References ----------------------------------------------------------- - -NASM was designed with simplicity of syntax in mind. One of the design -goals of NASM is that it should be possible, as far as is practical, for -the user to look at a single line of NASM code and tell what opcode is -generated by it. You can't do this in MASM: if you declare, for example, - - foo equ 1 - bar dw 2 - - then the two lines of code - - mov ax,foo - mov ax,bar - - generate completely different opcodes, despite having -identical-looking syntaxes. - - NASM avoids this undesirable situation by having a much simpler -syntax for memory references. The rule is simply that any access to the -_contents_ of a memory location requires square brackets around the -address, and any access to the _address_ of a variable doesn't. So an -instruction of the form `mov ax,foo' will _always_ refer to a -compile-time constant, whether it's an `EQU' or the address of a -variable; and to access the _contents_ of the variable `bar', you must -code `mov ax,[bar]'. - - This also means that NASM has no need for MASM's `OFFSET' keyword, -since the MASM code `mov ax,offset bar' means exactly the same thing as -NASM's `mov ax,bar'. If you're trying to get large amounts of MASM code -to assemble sensibly under NASM, you can always code `%idefine offset' -to make the preprocessor treat the `OFFSET' keyword as a no-op. - - This issue is even more confusing in `a86', where declaring a label -with a trailing colon defines it to be a `label' as opposed to a -`variable' and causes `a86' to adopt NASM-style semantics; so in `a86', -`mov ax,var' has different behaviour depending on whether `var' was -declared as `var: dw 0' (a label) or `var dw 0' (a word-size variable). -NASM is very simple by comparison: _everything_ is a label. - - NASM, in the interests of simplicity, also does not support the -hybrid syntaxes supported by MASM and its clones, such as `mov -ax,table[bx]', where a memory reference is denoted by one portion -outside square brackets and another portion inside. The correct syntax -for the above is `mov ax,[table+bx]'. Likewise, `mov ax,es:[di]' is -wrong and `mov ax,[es:di]' is right. - - -File: nasm.info, Node: Section 2.2.3, Next: Section 2.2.4, Prev: Section 2.2.2, Up: Section 2.2 - -2.2.3. NASM Doesn't Store Variable Types ----------------------------------------- - -NASM, by design, chooses not to remember the types of variables you -declare. Whereas MASM will remember, on seeing `var dw 0', that you -declared `var' as a word-size variable, and will then be able to fill -in the ambiguity in the size of the instruction `mov var,2', NASM will -deliberately remember nothing about the symbol `var' except where it -begins, and so you must explicitly code `mov word [var],2'. - - For this reason, NASM doesn't support the `LODS', `MOVS', `STOS', -`SCAS', `CMPS', `INS', or `OUTS' instructions, but only supports the -forms such as `LODSB', `MOVSW', and `SCASD', which explicitly specify -the size of the components of the strings being manipulated. - - -File: nasm.info, Node: Section 2.2.4, Next: Section 2.2.5, Prev: Section 2.2.3, Up: Section 2.2 - -2.2.4. NASM Doesn't `ASSUME' ----------------------------- - -As part of NASM's drive for simplicity, it also does not support the -`ASSUME' directive. NASM will not keep track of what values you choose -to put in your segment registers, and will never _automatically_ -generate a segment override prefix. - - -File: nasm.info, Node: Section 2.2.5, Next: Section 2.2.6, Prev: Section 2.2.4, Up: Section 2.2 - -2.2.5. NASM Doesn't Support Memory Models ------------------------------------------ - -NASM also does not have any directives to support different 16-bit -memory models. The programmer has to keep track of which functions are -supposed to be called with a far call and which with a near call, and -is responsible for putting the correct form of `RET' instruction -(`RETN' or `RETF'; NASM accepts `RET' itself as an alternate form for -`RETN'); in addition, the programmer is responsible for coding CALL FAR -instructions where necessary when calling _external_ functions, and -must also keep track of which external variable definitions are far and -which are near. - - -File: nasm.info, Node: Section 2.2.6, Next: Section 2.2.7, Prev: Section 2.2.5, Up: Section 2.2 - -2.2.6. Floating-Point Differences ---------------------------------- - -NASM uses different names to refer to floating-point registers from -MASM: where MASM would call them `ST(0)', `ST(1)' and so on, and `a86' -would call them simply `0', `1' and so on, NASM chooses to call them -`st0', `st1' etc. - - As of version 0.96, NASM now treats the instructions with `nowait' -forms in the same way as MASM-compatible assemblers. The idiosyncratic -treatment employed by 0.95 and earlier was based on a misunderstanding -by the authors. - - -File: nasm.info, Node: Section 2.2.7, Next: Chapter 3, Prev: Section 2.2.6, Up: Section 2.2 - -2.2.7. Other Differences ------------------------- - -For historical reasons, NASM uses the keyword `TWORD' where MASM and -compatible assemblers use `TBYTE'. - - NASM does not declare uninitialized storage in the same way as MASM: -where a MASM programmer might use `stack db 64 dup (?)', NASM requires -`stack resb 64', intended to be read as `reserve 64 bytes'. For a -limited amount of compatibility, since NASM treats `?' as a valid -character in symbol names, you can code `? equ 0' and then writing `dw -?' will at least do something vaguely useful. `DUP' is still not a -supported syntax, however. - - In addition to all of this, macros and directives work completely -differently to MASM. See *note Chapter 4:: and *note Chapter 6:: for -further details. - - -File: nasm.info, Node: Chapter 3, Next: Section 3.1, Prev: Section 2.2.7, Up: Top - -Chapter 3: The NASM Language -**************************** - -* Menu: - -* Section 3.1:: Layout of a NASM Source Line -* Section 3.2:: Pseudo-Instructions -* Section 3.3:: Effective Addresses -* Section 3.4:: Constants -* Section 3.5:: Expressions -* Section 3.6:: `SEG' and `WRT' -* Section 3.7:: `STRICT': Inhibiting Optimization -* Section 3.8:: Critical Expressions -* Section 3.9:: Local Labels - - -File: nasm.info, Node: Section 3.1, Next: Section 3.2, Prev: Chapter 3, Up: Chapter 3 - -3.1. Layout of a NASM Source Line -================================= - -Like most assemblers, each NASM source line contains (unless it is a -macro, a preprocessor directive or an assembler directive: see *note -Chapter 4:: and *note Chapter 6::) some combination of the four fields - - label: instruction operands ; comment - - As usual, most of these fields are optional; the presence or absence -of any combination of a label, an instruction and a comment is allowed. -Of course, the operand field is either required or forbidden by the -presence and nature of the instruction field. - - NASM uses backslash (\) as the line continuation character; if a -line ends with backslash, the next line is considered to be a part of -the backslash- ended line. - - NASM places no restrictions on white space within a line: labels may -have white space before them, or instructions may have no space before -them, or anything. The colon after a label is also optional. (Note that -this means that if you intend to code `lodsb' alone on a line, and type -`lodab' by accident, then that's still a valid source line which does -nothing but define a label. Running NASM with the command-line option -`-w+orphan-labels' will cause it to warn you if you define a label -alone on a line without a trailing colon.) - - Valid characters in labels are letters, numbers, `_', `$', `#', `@', -`~', `.', and `?'. The only characters which may be used as the _first_ -character of an identifier are letters, `.' (with special meaning: see -*note Section 3.9::), `_' and `?'. An identifier may also be prefixed -with a `$' to indicate that it is intended to be read as an identifier -and not a reserved word; thus, if some other module you are linking -with defines a symbol called `eax', you can refer to `$eax' in NASM -code to distinguish the symbol from the register. Maximum length of an -identifier is 4095 characters. - - The instruction field may contain any machine instruction: Pentium -and P6 instructions, FPU instructions, MMX instructions and even -undocumented instructions are all supported. The instruction may be -prefixed by `LOCK', `REP', `REPE'/`REPZ' or `REPNE'/`REPNZ', in the -usual way. Explicit address-size and operand-size prefixes `A16', -`A32', `A64', `O16' and `O32', `O64' are provided - one example of -their use is given in *note Chapter 10::. You can also use the name of -a segment register as an instruction prefix: coding `es mov [bx],ax' is -equivalent to coding `mov [es:bx],ax'. We recommend the latter syntax, -since it is consistent with other syntactic features of the language, -but for instructions such as `LODSB', which has no operands and yet can -require a segment override, there is no clean syntactic way to proceed -apart from `es lodsb'. - - An instruction is not required to use a prefix: prefixes such as -`CS', `A32', `LOCK' or `REPE' can appear on a line by themselves, and -NASM will just generate the prefix bytes. - - In addition to actual machine instructions, NASM also supports a -number of pseudo-instructions, described in *note Section 3.2::. - - Instruction operands may take a number of forms: they can be -registers, described simply by the register name (e.g. `ax', `bp', -`ebx', `cr0': NASM does not use the `gas'-style syntax in which -register names must be prefixed by a `%' sign), or they can be -effective addresses (see *note Section 3.3::), constants (*note Section -3.4::) or expressions (*note Section 3.5::). - - For x87 floating-point instructions, NASM accepts a wide range of -syntaxes: you can use two-operand forms like MASM supports, or you can -use NASM's native single-operand forms in most cases. For example, you -can code: - - fadd st1 ; this sets st0 := st0 + st1 - fadd st0,st1 ; so does this - - fadd st1,st0 ; this sets st1 := st1 + st0 - fadd to st1 ; so does this - - Almost any x87 floating-point instruction that references memory -must use one of the prefixes `DWORD', `QWORD' or `TWORD' to indicate -what size of memory operand it refers to. - - -File: nasm.info, Node: Section 3.2, Next: Section 3.2.1, Prev: Section 3.1, Up: Chapter 3 - -3.2. Pseudo-Instructions -======================== - -Pseudo-instructions are things which, though not real x86 machine -instructions, are used in the instruction field anyway because that's -the most convenient place to put them. The current pseudo-instructions -are `DB', `DW', `DD', `DQ', `DT', `DO' and `DY'; their uninitialized -counterparts `RESB', `RESW', `RESD', `RESQ', `REST', `RESO' and `RESY'; -the `INCBIN' command, the `EQU' command, and the `TIMES' prefix. - -* Menu: - -* Section 3.2.1:: `DB' and Friends: Declaring Initialized Data -* Section 3.2.2:: `RESB' and Friends: Declaring Uninitialized Data -* Section 3.2.3:: `INCBIN': Including External Binary Files -* Section 3.2.4:: `EQU': Defining Constants -* Section 3.2.5:: `TIMES': Repeating Instructions or Data - - -File: nasm.info, Node: Section 3.2.1, Next: Section 3.2.2, Prev: Section 3.2, Up: Section 3.2 - -3.2.1. `DB' and Friends: Declaring Initialized Data ---------------------------------------------------- - -`DB', `DW', `DD', `DQ', `DT', `DO' and `DY' are used, much as in MASM, -to declare initialized data in the output file. They can be invoked in -a wide range of ways: - - db 0x55 ; just the byte 0x55 - db 0x55,0x56,0x57 ; three bytes in succession - db 'a',0x55 ; character constants are OK - db 'hello',13,10,'$' ; so are string constants - dw 0x1234 ; 0x34 0x12 - dw 'a' ; 0x61 0x00 (it's just a number) - dw 'ab' ; 0x61 0x62 (character constant) - dw 'abc' ; 0x61 0x62 0x63 0x00 (string) - dd 0x12345678 ; 0x78 0x56 0x34 0x12 - dd 1.234567e20 ; floating-point constant - dq 0x123456789abcdef0 ; eight byte constant - dq 1.234567e20 ; double-precision float - dt 1.234567e20 ; extended-precision float - - `DT', `DO' and `DY' do not accept numeric constants as operands. - - -File: nasm.info, Node: Section 3.2.2, Next: Section 3.2.3, Prev: Section 3.2.1, Up: Section 3.2 - -3.2.2. `RESB' and Friends: Declaring Uninitialized Data -------------------------------------------------------- - -`RESB', `RESW', `RESD', `RESQ', `REST', `RESO' and `RESY' are designed -to be used in the BSS section of a module: they declare _uninitialized_ -storage space. Each takes a single operand, which is the number of -bytes, words, doublewords or whatever to reserve. As stated in *note -Section 2.2.7::, NASM does not support the MASM/TASM syntax of -reserving uninitialized space by writing `DW ?' or similar things: this -is what it does instead. The operand to a `RESB'-type -pseudo-instruction is a _critical expression_: see *note Section 3.8::. - - For example: - - buffer: resb 64 ; reserve 64 bytes - wordvar: resw 1 ; reserve a word - realarray resq 10 ; array of ten reals - ymmval: resy 1 ; one YMM register - - -File: nasm.info, Node: Section 3.2.3, Next: Section 3.2.4, Prev: Section 3.2.2, Up: Section 3.2 - -3.2.3. `INCBIN': Including External Binary Files ------------------------------------------------- - -`INCBIN' is borrowed from the old Amiga assembler DevPac: it includes a -binary file verbatim into the output file. This can be handy for (for -example) including graphics and sound data directly into a game -executable file. It can be called in one of these three ways: - - incbin "file.dat" ; include the whole file - incbin "file.dat",1024 ; skip the first 1024 bytes - incbin "file.dat",1024,512 ; skip the first 1024, and - ; actually include at most 512 - - `INCBIN' is both a directive and a standard macro; the standard macro -version searches for the file in the include file search path and adds -the file to the dependency lists. This macro can be overridden if -desired. - - -File: nasm.info, Node: Section 3.2.4, Next: Section 3.2.5, Prev: Section 3.2.3, Up: Section 3.2 - -3.2.4. `EQU': Defining Constants --------------------------------- - -`EQU' defines a symbol to a given constant value: when `EQU' is used, -the source line must contain a label. The action of `EQU' is to define -the given label name to the value of its (only) operand. This -definition is absolute, and cannot change later. So, for example, - - message db 'hello, world' - msglen equ $-message - - defines `msglen' to be the constant 12. `msglen' may not then be -redefined later. This is not a preprocessor definition either: the -value of `msglen' is evaluated _once_, using the value of `$' (see -*note Section 3.5:: for an explanation of `$') at the point of -definition, rather than being evaluated wherever it is referenced and -using the value of `$' at the point of reference. - - -File: nasm.info, Node: Section 3.2.5, Next: Section 3.3, Prev: Section 3.2.4, Up: Section 3.2 - -3.2.5. `TIMES': Repeating Instructions or Data ----------------------------------------------- - -The `TIMES' prefix causes the instruction to be assembled multiple -times. This is partly present as NASM's equivalent of the `DUP' syntax -supported by MASM-compatible assemblers, in that you can code - - zerobuf: times 64 db 0 - - or similar things; but `TIMES' is more versatile than that. The -argument to `TIMES' is not just a numeric constant, but a numeric -_expression_, so you can do things like - - buffer: db 'hello, world' - times 64-$+buffer db ' ' - - which will store exactly enough spaces to make the total length of -`buffer' up to 64. Finally, `TIMES' can be applied to ordinary -instructions, so you can code trivial unrolled loops in it: - - times 100 movsb - - Note that there is no effective difference between `times 100 resb 1' -and `resb 100', except that the latter will be assembled about 100 -times faster due to the internal structure of the assembler. - - The operand to `TIMES' is a critical expression (*note Section -3.8::). - - Note also that `TIMES' can't be applied to macros: the reason for -this is that `TIMES' is processed after the macro phase, which allows -the argument to `TIMES' to contain expressions such as `64-$+buffer' as -above. To repeat more than one line of code, or a complex macro, use the -preprocessor `%rep' directive. - - -File: nasm.info, Node: Section 3.3, Next: Section 3.4, Prev: Section 3.2.5, Up: Chapter 3 - -3.3. Effective Addresses -======================== - -An effective address is any operand to an instruction which references -memory. Effective addresses, in NASM, have a very simple syntax: they -consist of an expression evaluating to the desired address, enclosed in -square brackets. For example: - - wordvar dw 123 - mov ax,[wordvar] - mov ax,[wordvar+1] - mov ax,[es:wordvar+bx] - - Anything not conforming to this simple system is not a valid memory -reference in NASM, for example `es:wordvar[bx]'. - - More complicated effective addresses, such as those involving more -than one register, work in exactly the same way: - - mov eax,[ebx*2+ecx+offset] - mov ax,[bp+di+8] - - NASM is capable of doing algebra on these effective addresses, so -that things which don't necessarily _look_ legal are perfectly all -right: - - mov eax,[ebx*5] ; assembles as [ebx*4+ebx] - mov eax,[label1*2-label2] ; ie [label1+(label1-label2)] - - Some forms of effective address have more than one assembled form; -in most such cases NASM will generate the smallest form it can. For -example, there are distinct assembled forms for the 32-bit effective -addresses `[eax*2+0]' and `[eax+eax]', and NASM will generally generate -the latter on the grounds that the former requires four bytes to store -a zero offset. - - NASM has a hinting mechanism which will cause `[eax+ebx]' and -`[ebx+eax]' to generate different opcodes; this is occasionally useful -because `[esi+ebp]' and `[ebp+esi]' have different default segment -registers. - - However, you can force NASM to generate an effective address in a -particular form by the use of the keywords `BYTE', `WORD', `DWORD' and -`NOSPLIT'. If you need `[eax+3]' to be assembled using a double-word -offset field instead of the one byte NASM will normally generate, you -can code `[dword eax+3]'. Similarly, you can force NASM to use a byte -offset for a small value which it hasn't seen on the first pass (see -*note Section 3.8:: for an example of such a code fragment) by using -`[byte eax+offset]'. As special cases, `[byte eax]' will code `[eax+0]' -with a byte offset of zero, and `[dword eax]' will code it with a -double-word offset of zero. The normal form, `[eax]', will be coded -with no offset field. - - The form described in the previous paragraph is also useful if you -are trying to access data in a 32-bit segment from within 16 bit code. -For more information on this see the section on mixed-size addressing -(*note Section 10.2::). In particular, if you need to access data with -a known offset that is larger than will fit in a 16-bit value, if you -don't specify that it is a dword offset, nasm will cause the high word -of the offset to be lost. - - Similarly, NASM will split `[eax*2]' into `[eax+eax]' because that -allows the offset field to be absent and space to be saved; in fact, it -will also split `[eax*2+offset]' into `[eax+eax+offset]'. You can -combat this behaviour by the use of the `NOSPLIT' keyword: `[nosplit -eax*2]' will force `[eax*2+0]' to be generated literally. - - In 64-bit mode, NASM will by default generate absolute addresses. The -`REL' keyword makes it produce `RIP'-relative addresses. Since this is -frequently the normally desired behaviour, see the `DEFAULT' directive -(*note Section 6.2::). The keyword `ABS' overrides `REL'. - - -File: nasm.info, Node: Section 3.4, Next: Section 3.4.1, Prev: Section 3.3, Up: Chapter 3 - -3.4. Constants -============== - -NASM understands four different types of constant: numeric, character, -string and floating-point. - -* Menu: - -* Section 3.4.1:: Numeric Constants -* Section 3.4.2:: Character Strings -* Section 3.4.3:: Character Constants -* Section 3.4.4:: String Constants -* Section 3.4.5:: Unicode Strings -* Section 3.4.6:: Floating-Point Constants -* Section 3.4.7:: Packed BCD Constants - - -File: nasm.info, Node: Section 3.4.1, Next: Section 3.4.2, Prev: Section 3.4, Up: Section 3.4 - -3.4.1. Numeric Constants ------------------------- - -A numeric constant is simply a number. NASM allows you to specify -numbers in a variety of number bases, in a variety of ways: you can -suffix `H' or `X', `Q' or `O', and `B' for hexadecimal, octal and -binary respectively, or you can prefix `0x' for hexadecimal in the -style of C, or you can prefix `$' for hexadecimal in the style of -Borland Pascal. Note, though, that the `$' prefix does double duty as a -prefix on identifiers (see *note Section 3.1::), so a hex number -prefixed with a `$' sign must have a digit after the `$' rather than a -letter. In addition, current versions of NASM accept the prefix `0h' -for hexadecimal, `0o' or `0q' for octal, and `0b' for binary. Please -note that unlike C, a `0' prefix by itself does _not_ imply an octal -constant! - - Numeric constants can have underscores (`_') interspersed to break up -long strings. - - Some examples (all producing exactly the same code): - - mov ax,200 ; decimal - mov ax,0200 ; still decimal - mov ax,0200d ; explicitly decimal - mov ax,0d200 ; also decimal - mov ax,0c8h ; hex - mov ax,$0c8 ; hex again: the 0 is required - mov ax,0xc8 ; hex yet again - mov ax,0hc8 ; still hex - mov ax,310q ; octal - mov ax,310o ; octal again - mov ax,0o310 ; octal yet again - mov ax,0q310 ; hex yet again - mov ax,11001000b ; binary - mov ax,1100_1000b ; same binary constant - mov ax,0b1100_1000 ; same binary constant yet again - - -File: nasm.info, Node: Section 3.4.2, Next: Section 3.4.3, Prev: Section 3.4.1, Up: Section 3.4 - -3.4.2. Character Strings ------------------------- - -A character string consists of up to eight characters enclosed in either -single quotes (`'...''), double quotes (`"..."') or backquotes -(``...`'). Single or double quotes are equivalent to NASM (except of -course that surrounding the constant with single quotes allows double -quotes to appear within it and vice versa); the contents of those are -represented verbatim. Strings enclosed in backquotes support C-style -`\'-escapes for special characters. - - The following escape sequences are recognized by backquoted strings: - - \' single quote (') - \" double quote (") - \` backquote (`) - \\ backslash (\) - \? question mark (?) - \a BEL (ASCII 7) - \b BS (ASCII 8) - \t TAB (ASCII 9) - \n LF (ASCII 10) - \v VT (ASCII 11) - \f FF (ASCII 12) - \r CR (ASCII 13) - \e ESC (ASCII 27) - \377 Up to 3 octal digits - literal byte - \xFF Up to 2 hexadecimal digits - literal byte - \u1234 4 hexadecimal digits - Unicode character - \U12345678 8 hexadecimal digits - Unicode character - - All other escape sequences are reserved. Note that `\0', meaning a -`NUL' character (ASCII 0), is a special case of the octal escape -sequence. - - Unicode characters specified with `\u' or `\U' are converted to -UTF-8. For example, the following lines are all equivalent: - - db `\u263a` ; UTF-8 smiley face - db `\xe2\x98\xba` ; UTF-8 smiley face - db 0E2h, 098h, 0BAh ; UTF-8 smiley face - - -File: nasm.info, Node: Section 3.4.3, Next: Section 3.4.4, Prev: Section 3.4.2, Up: Section 3.4 - -3.4.3. Character Constants --------------------------- - -A character constant consists of a string up to eight bytes long, used -in an expression context. It is treated as if it was an integer. - - A character constant with more than one byte will be arranged with -little-endian order in mind: if you code - - mov eax,'abcd' - - then the constant generated is not `0x61626364', but `0x64636261', -so that if you were then to store the value into memory, it would read -`abcd' rather than `dcba'. This is also the sense of character -constants understood by the Pentium's `CPUID' instruction. - - -File: nasm.info, Node: Section 3.4.4, Next: Section 3.4.5, Prev: Section 3.4.3, Up: Section 3.4 - -3.4.4. String Constants ------------------------ - -String constants are character strings used in the context of some -pseudo- instructions, namely the `DB' family and `INCBIN' (where it -represents a filename.) They are also used in certain preprocessor -directives. - - A string constant looks like a character constant, only longer. It is -treated as a concatenation of maximum-size character constants for the -conditions. So the following are equivalent: - - db 'hello' ; string constant - db 'h','e','l','l','o' ; equivalent character constants - - And the following are also equivalent: - - dd 'ninechars' ; doubleword string constant - dd 'nine','char','s' ; becomes three doublewords - db 'ninechars',0,0,0 ; and really looks like this - - Note that when used in a string-supporting context, quoted strings -are treated as a string constants even if they are short enough to be a -character constant, because otherwise `db 'ab'' would have the same -effect as `db 'a'', which would be silly. Similarly, three-character or -four-character constants are treated as strings when they are operands -to `DW', and so forth. - - -File: nasm.info, Node: Section 3.4.5, Next: Section 3.4.6, Prev: Section 3.4.4, Up: Section 3.4 - -3.4.5. Unicode Strings ----------------------- - -The special operators `__utf16__' and `__utf32__' allows definition of -Unicode strings. They take a string in UTF-8 format and converts it to -(littleendian) UTF-16 or UTF-32, respectively. - - For example: - - %define u(x) __utf16__(x) - %define w(x) __utf32__(x) - - dw u('C:\WINDOWS'), 0 ; Pathname in UTF-16 - dd w(`A + B = \u206a`), 0 ; String in UTF-32 - - `__utf16__' and `__utf32__' can be applied either to strings passed -to the `DB' family instructions, or to character constants in an -expression context. - - -File: nasm.info, Node: Section 3.4.6, Next: Section 3.4.7, Prev: Section 3.4.5, Up: Section 3.4 - -3.4.6. Floating-Point Constants -------------------------------- - -Floating-point constants are acceptable only as arguments to `DB', -`DW', `DD', `DQ', `DT', and `DO', or as arguments to the special -operators `__float8__', `__float16__', `__float32__', `__float64__', -`__float80m__', `__float80e__', `__float128l__', and `__float128h__'. - - Floating-point constants are expressed in the traditional form: -digits, then a period, then optionally more digits, then optionally an -`E' followed by an exponent. The period is mandatory, so that NASM can -distinguish between `dd 1', which declares an integer constant, and `dd -1.0' which declares a floating-point constant. NASM also support -C99-style hexadecimal floating-point: `0x', hexadecimal digits, period, -optionally more hexadeximal digits, then optionally a `P' followed by a -_binary_ (not hexadecimal) exponent in decimal notation. - - Underscores to break up groups of digits are permitted in -floating-point constants as well. - - Some examples: - - db -0.2 ; "Quarter precision" - dw -0.5 ; IEEE 754r/SSE5 half precision - dd 1.2 ; an easy one - dd 1.222_222_222 ; underscores are permitted - dd 0x1p+2 ; 1.0x2^2 = 4.0 - dq 0x1p+32 ; 1.0x2^32 = 4 294 967 296.0 - dq 1.e10 ; 10 000 000 000.0 - dq 1.e+10 ; synonymous with 1.e10 - dq 1.e-10 ; 0.000 000 000 1 - dt 3.141592653589793238462 ; pi - do 1.e+4000 ; IEEE 754r quad precision - - The 8-bit "quarter-precision" floating-point format is -sign:exponent:mantissa = 1:4:3 with an exponent bias of 7. This appears -to be the most frequently used 8-bit floating-point format, although it -is not covered by any formal standard. This is sometimes called a -"minifloat." - - The special operators are used to produce floating-point numbers in -other contexts. They produce the binary representation of a specific -floating- point number as an integer, and can use anywhere integer -constants are used in an expression. `__float80m__' and `__float80e__' -produce the 64-bit mantissa and 16-bit exponent of an 80-bit -floating-point number, and `__float128l__' and `__float128h__' produce -the lower and upper 64-bit halves of a 128-bit floating-point number, -respectively. - - For example: - - mov rax,__float64__(3.141592653589793238462) - - ... would assign the binary representation of pi as a 64-bit -floating point number into `RAX'. This is exactly equivalent to: - - mov rax,0x400921fb54442d18 - - NASM cannot do compile-time arithmetic on floating-point constants. -This is because NASM is designed to be portable - although it always -generates code to run on x86 processors, the assembler itself can run -on any system with an ANSI C compiler. Therefore, the assembler cannot -guarantee the presence of a floating-point unit capable of handling the -Intel number formats, and so for NASM to be able to do floating -arithmetic it would have to include its own complete set of -floating-point routines, which would significantly increase the size of -the assembler for very little benefit. - - The special tokens `__Infinity__', `__QNaN__' (or `__NaN__') and -`__SNaN__' can be used to generate infinities, quiet NaNs, and -signalling NaNs, respectively. These are normally used as macros: - - %define Inf __Infinity__ - %define NaN __QNaN__ - - dq +1.5, -Inf, NaN ; Double-precision constants - - -File: nasm.info, Node: Section 3.4.7, Next: Section 3.5, Prev: Section 3.4.6, Up: Section 3.4 - -3.4.7. Packed BCD Constants ---------------------------- - -x87-style packed BCD constants can be used in the same contexts as -80-bit floating-point numbers. They are suffixed with `p' or prefixed -with `0p', and can include up to 18 decimal digits. - - As with other numeric constants, underscores can be used to separate -digits. - - For example: - - dt 12_345_678_901_245_678p - dt -12_345_678_901_245_678p - dt +0p33 - dt 33p - - -File: nasm.info, Node: Section 3.5, Next: Section 3.5.1, Prev: Section 3.4.7, Up: Chapter 3 - -3.5. Expressions -================ - -Expressions in NASM are similar in syntax to those in C. Expressions are -evaluated as 64-bit integers which are then adjusted to the appropriate -size. - - NASM supports two special tokens in expressions, allowing -calculations to involve the current assembly position: the `$' and `$$' -tokens. `$' evaluates to the assembly position at the beginning of the -line containing the expression; so you can code an infinite loop using -`JMP $'. `$$' evaluates to the beginning of the current section; so you -can tell how far into the section you are by using `($-$$)'. - - The arithmetic operators provided by NASM are listed here, in -increasing order of precedence. - -* Menu: - -* Section 3.5.1:: `|': Bitwise OR Operator -* Section 3.5.2:: `^': Bitwise XOR Operator -* Section 3.5.3:: `&': Bitwise AND Operator -* Section 3.5.4:: `<<' and `>>': Bit Shift Operators -* Section 3.5.5:: `+' and `-': Addition and Subtraction Operators -* Section 3.5.6:: `*', `/', `//', `%' and `%%': Multiplication and Division -* Section 3.5.7:: Unary Operators: `+', `-', `~', `!' and `SEG' - - -File: nasm.info, Node: Section 3.5.1, Next: Section 3.5.2, Prev: Section 3.5, Up: Section 3.5 - -3.5.1. `|': Bitwise OR Operator -------------------------------- - -The `|' operator gives a bitwise OR, exactly as performed by the `OR' -machine instruction. Bitwise OR is the lowest-priority arithmetic -operator supported by NASM. - - -File: nasm.info, Node: Section 3.5.2, Next: Section 3.5.3, Prev: Section 3.5.1, Up: Section 3.5 - -3.5.2. `^': Bitwise XOR Operator --------------------------------- - -`^' provides the bitwise XOR operation. - - -File: nasm.info, Node: Section 3.5.3, Next: Section 3.5.4, Prev: Section 3.5.2, Up: Section 3.5 - -3.5.3. `&': Bitwise AND Operator --------------------------------- - -`&' provides the bitwise AND operation. - - -File: nasm.info, Node: Section 3.5.4, Next: Section 3.5.5, Prev: Section 3.5.3, Up: Section 3.5 - -3.5.4. `<<' and `>>': Bit Shift Operators ------------------------------------------ - -`<<' gives a bit-shift to the left, just as it does in C. So `5<<3' -evaluates to 5 times 8, or 40. `>>' gives a bit-shift to the right; in -NASM, such a shift is _always_ unsigned, so that the bits shifted in -from the left-hand end are filled with zero rather than a -sign-extension of the previous highest bit. - - -File: nasm.info, Node: Section 3.5.5, Next: Section 3.5.6, Prev: Section 3.5.4, Up: Section 3.5 - -3.5.5. `+' and `-': Addition and Subtraction Operators ------------------------------------------------------- - -The `+' and `-' operators do perfectly ordinary addition and -subtraction. - - -File: nasm.info, Node: Section 3.5.6, Next: Section 3.5.7, Prev: Section 3.5.5, Up: Section 3.5 - -3.5.6. `*', `/', `//', `%' and `%%': Multiplication and Division ----------------------------------------------------------------- - -`*' is the multiplication operator. `/' and `//' are both division -operators: `/' is unsigned division and `//' is signed division. -Similarly, `%' and `%%' provide unsigned and signed modulo operators -respectively. - - NASM, like ANSI C, provides no guarantees about the sensible -operation of the signed modulo operator. - - Since the `%' character is used extensively by the macro -preprocessor, you should ensure that both the signed and unsigned -modulo operators are followed by white space wherever they appear. - - -File: nasm.info, Node: Section 3.5.7, Next: Section 3.6, Prev: Section 3.5.6, Up: Section 3.5 - -3.5.7. Unary Operators: `+', `-', `~', `!' and `SEG' ----------------------------------------------------- - -The highest-priority operators in NASM's expression grammar are those -which only apply to one argument. `-' negates its operand, `+' does -nothing (it's provided for symmetry with `-'), `~' computes the one's -complement of its operand, `!' is the logical negation operator, and -`SEG' provides the segment address of its operand (explained in more -detail in *note Section 3.6::). - - -File: nasm.info, Node: Section 3.6, Next: Section 3.7, Prev: Section 3.5.7, Up: Chapter 3 - -3.6. `SEG' and `WRT' -==================== - -When writing large 16-bit programs, which must be split into multiple -segments, it is often necessary to be able to refer to the segment part -of the address of a symbol. NASM supports the `SEG' operator to perform -this function. - - The `SEG' operator returns the _preferred_ segment base of a symbol, -defined as the segment base relative to which the offset of the symbol -makes sense. So the code - - mov ax,seg symbol - mov es,ax - mov bx,symbol - - will load `ES:BX' with a valid pointer to the symbol `symbol'. - - Things can be more complex than this: since 16-bit segments and -groups may overlap, you might occasionally want to refer to some symbol -using a different segment base from the preferred one. NASM lets you do -this, by the use of the `WRT' (With Reference To) keyword. So you can -do things like - - mov ax,weird_seg ; weird_seg is a segment base - mov es,ax - mov bx,symbol wrt weird_seg - - to load `ES:BX' with a different, but functionally equivalent, -pointer to the symbol `symbol'. - - NASM supports far (inter-segment) calls and jumps by means of the -syntax `call segment:offset', where `segment' and `offset' both -represent immediate values. So to call a far procedure, you could code -either of - - call (seg procedure):procedure - call weird_seg:(procedure wrt weird_seg) - - (The parentheses are included for clarity, to show the intended -parsing of the above instructions. They are not necessary in practice.) - - NASM supports the syntax `call far procedure' as a synonym for the -first of the above usages. `JMP' works identically to `CALL' in these -examples. - - To declare a far pointer to a data item in a data segment, you must -code - - dw symbol, seg symbol - - NASM supports no convenient synonym for this, though you can always -invent one using the macro processor. - - -File: nasm.info, Node: Section 3.7, Next: Section 3.8, Prev: Section 3.6, Up: Chapter 3 - -3.7. `STRICT': Inhibiting Optimization -====================================== - -When assembling with the optimizer set to level 2 or higher (see *note -Section 2.1.22::), NASM will use size specifiers (`BYTE', `WORD', -`DWORD', `QWORD', `TWORD', `OWORD' or `YWORD'), but will give them the -smallest possible size. The keyword `STRICT' can be used to inhibit -optimization and force a particular operand to be emitted in the -specified size. For example, with the optimizer on, and in `BITS 16' -mode, - - push dword 33 - - is encoded in three bytes `66 6A 21', whereas - - push strict dword 33 - - is encoded in six bytes, with a full dword immediate operand `66 68 -21 00 00 00'. - - With the optimizer off, the same code (six bytes) is generated -whether the `STRICT' keyword was used or not. - - -File: nasm.info, Node: Section 3.8, Next: Section 3.9, Prev: Section 3.7, Up: Chapter 3 - -3.8. Critical Expressions -========================= - -Although NASM has an optional multi-pass optimizer, there are some -expressions which must be resolvable on the first pass. These are called -_Critical Expressions_. - - The first pass is used to determine the size of all the assembled -code and data, so that the second pass, when generating all the code, -knows all the symbol addresses the code refers to. So one thing NASM -can't handle is code whose size depends on the value of a symbol -declared after the code in question. For example, - - times (label-$) db 0 - label: db 'Where am I?' - - The argument to `TIMES' in this case could equally legally evaluate -to anything at all; NASM will reject this example because it cannot -tell the size of the `TIMES' line when it first sees it. It will just -as firmly reject the slightly paradoxical code - - times (label-$+1) db 0 - label: db 'NOW where am I?' - - in which _any_ value for the `TIMES' argument is by definition wrong! - - NASM rejects these examples by means of a concept called a _critical -expression_, which is defined to be an expression whose value is -required to be computable in the first pass, and which must therefore -depend only on symbols defined before it. The argument to the `TIMES' -prefix is a critical expression. - - -File: nasm.info, Node: Section 3.9, Next: Chapter 4, Prev: Section 3.8, Up: Chapter 3 - -3.9. Local Labels -================= - -NASM gives special treatment to symbols beginning with a period. A label -beginning with a single period is treated as a _local_ label, which -means that it is associated with the previous non-local label. So, for -example: - - label1 ; some code - - .loop - ; some more code - - jne .loop - ret - - label2 ; some code - - .loop - ; some more code - - jne .loop - ret - - In the above code fragment, each `JNE' instruction jumps to the line -immediately before it, because the two definitions of `.loop' are kept -separate by virtue of each being associated with the previous non-local -label. - - This form of local label handling is borrowed from the old Amiga -assembler DevPac; however, NASM goes one step further, in allowing -access to local labels from other parts of the code. This is achieved -by means of _defining_ a local label in terms of the previous non-local -label: the first definition of `.loop' above is really defining a -symbol called `label1.loop', and the second defines a symbol called -`label2.loop'. So, if you really needed to, you could write - - label3 ; some more code - ; and some more - - jmp label1.loop - - Sometimes it is useful - in a macro, for instance - to be able to -define a label which can be referenced from anywhere but which doesn't -interfere with the normal local-label mechanism. Such a label can't be -non-local because it would interfere with subsequent definitions of, -and references to, local labels; and it can't be local because the -macro that defined it wouldn't know the label's full name. NASM -therefore introduces a third type of label, which is probably only -useful in macro definitions: if a label begins with the special prefix -`..@', then it does nothing to the local label mechanism. So you could -code - - label1: ; a non-local label - .local: ; this is really label1.local - ..@foo: ; this is a special symbol - label2: ; another non-local label - .local: ; this is really label2.local - - jmp ..@foo ; this will jump three lines up - - NASM has the capacity to define other special symbols beginning with -a double period: for example, `..start' is used to specify the entry -point in the `obj' output format (see *note Section 7.4.6::). - - -File: nasm.info, Node: Chapter 4, Next: Section 4.1, Prev: Section 3.9, Up: Top - -Chapter 4: The NASM Preprocessor -******************************** - -NASM contains a powerful macro processor, which supports conditional -assembly, multi-level file inclusion, two forms of macro (single-line -and multi-line), and a `context stack' mechanism for extra macro power. -Preprocessor directives all begin with a `%' sign. - - The preprocessor collapses all lines which end with a backslash (\) -character into a single line. Thus: - - %define THIS_VERY_LONG_MACRO_NAME_IS_DEFINED_TO \ - THIS_VALUE - - will work like a single-line macro without the backslash-newline -sequence. - -* Menu: - -* Section 4.1:: Single-Line Macros -* Section 4.2:: String Manipulation in Macros -* Section 4.3:: Multi-Line Macros: `%macro' -* Section 4.4:: Conditional Assembly -* Section 4.5:: Preprocessor Loops: `%rep' -* Section 4.6:: Source Files and Dependencies -* Section 4.7:: The Context Stack -* Section 4.8:: Stack Relative Preprocessor Directives -* Section 4.9:: Reporting User-Defined Errors: `%error', `%warning', `%fatal' -* Section 4.10:: Other Preprocessor Directives -* Section 4.11:: Standard Macros - - -File: nasm.info, Node: Section 4.1, Next: Section 4.1.1, Prev: Chapter 4, Up: Chapter 4 - -4.1. Single-Line Macros -======================= - -* Menu: - -* Section 4.1.1:: The Normal Way: `%define' -* Section 4.1.2:: Resolving `%define': `%xdefine' -* Section 4.1.3:: Macro Indirection: `%[...]' -* Section 4.1.4:: Concatenating Single Line Macro Tokens: `%+' -* Section 4.1.5:: The Macro Name Itself: `%?' and `%??' -* Section 4.1.6:: Undefining Single-Line Macros: `%undef' -* Section 4.1.7:: Preprocessor Variables: `%assign' -* Section 4.1.8:: Defining Strings: `%defstr' -* Section 4.1.9:: Defining Tokens: `%deftok' - - -File: nasm.info, Node: Section 4.1.1, Next: Section 4.1.2, Prev: Section 4.1, Up: Section 4.1 - -4.1.1. The Normal Way: `%define' --------------------------------- - -Single-line macros are defined using the `%define' preprocessor -directive. The definitions work in a similar way to C; so you can do -things like - - %define ctrl 0x1F & - %define param(a,b) ((a)+(a)*(b)) - - mov byte [param(2,ebx)], ctrl 'D' - - which will expand to - - mov byte [(2)+(2)*(ebx)], 0x1F & 'D' - - When the expansion of a single-line macro contains tokens which -invoke another macro, the expansion is performed at invocation time, -not at definition time. Thus the code - - %define a(x) 1+b(x) - %define b(x) 2*x - - mov ax,a(8) - - will evaluate in the expected way to `mov ax,1+2*8', even though the -macro `b' wasn't defined at the time of definition of `a'. - - Macros defined with `%define' are case sensitive: after `%define foo -bar', only `foo' will expand to `bar': `Foo' or `FOO' will not. By -using `%idefine' instead of `%define' (the `i' stands for -`insensitive') you can define all the case variants of a macro at once, -so that `%idefine foo bar' would cause `foo', `Foo', `FOO', `fOO' and -so on all to expand to `bar'. - - There is a mechanism which detects when a macro call has occurred as -a result of a previous expansion of the same macro, to guard against -circular references and infinite loops. If this happens, the -preprocessor will only expand the first occurrence of the macro. Hence, -if you code - - %define a(x) 1+a(x) - - mov ax,a(3) - - the macro `a(3)' will expand once, becoming `1+a(3)', and will then -expand no further. This behaviour can be useful: see *note Section 9.1:: -for an example of its use. - - You can overload single-line macros: if you write - - %define foo(x) 1+x - %define foo(x,y) 1+x*y - - the preprocessor will be able to handle both types of macro call, by -counting the parameters you pass; so `foo(3)' will become `1+3' whereas -`foo(ebx,2)' will become `1+ebx*2'. However, if you define - - %define foo bar - - then no other definition of `foo' will be accepted: a macro with no -parameters prohibits the definition of the same name as a macro _with_ -parameters, and vice versa. - - This doesn't prevent single-line macros being _redefined_: you can -perfectly well define a macro with - - %define foo bar - - and then re-define it later in the same source file with - - %define foo baz - - Then everywhere the macro `foo' is invoked, it will be expanded -according to the most recent definition. This is particularly useful -when defining single-line macros with `%assign' (see *note Section -4.1.7::). - - You can pre-define single-line macros using the `-d' option on the -NASM command line: see *note Section 2.1.18::. - - -File: nasm.info, Node: Section 4.1.2, Next: Section 4.1.3, Prev: Section 4.1.1, Up: Section 4.1 - -4.1.2. Resolving `%define': `%xdefine' --------------------------------------- - -To have a reference to an embedded single-line macro resolved at the -time that the embedding macro is _defined_, as opposed to when the -embedding macro is _expanded_, you need a different mechanism to the -one offered by `%define'. The solution is to use `%xdefine', or it's -case-insensitive counterpart `%ixdefine'. - - Suppose you have the following code: - - %define isTrue 1 - %define isFalse isTrue - %define isTrue 0 - - val1: db isFalse - - %define isTrue 1 - - val2: db isFalse - - In this case, `val1' is equal to 0, and `val2' is equal to 1. This -is because, when a single-line macro is defined using `%define', it is -expanded only when it is called. As `isFalse' expands to `isTrue', the -expansion will be the current value of `isTrue'. The first time it is -called that is 0, and the second time it is 1. - - If you wanted `isFalse' to expand to the value assigned to the -embedded macro `isTrue' at the time that `isFalse' was defined, you -need to change the above code to use `%xdefine'. - - %xdefine isTrue 1 - %xdefine isFalse isTrue - %xdefine isTrue 0 - - val1: db isFalse - - %xdefine isTrue 1 - - val2: db isFalse - - Now, each time that `isFalse' is called, it expands to 1, as that is -what the embedded macro `isTrue' expanded to at the time that `isFalse' -was defined. - - -File: nasm.info, Node: Section 4.1.3, Next: Section 4.1.4, Prev: Section 4.1.2, Up: Section 4.1 - -4.1.3. Macro Indirection: `%[...]' ----------------------------------- - -The `%[...]' construct can be used to expand macros in contexts where -macro expansion would otherwise not occur, including in the names other -macros. For example, if you have a set of macros named `Foo16', `Foo32' -and `Foo64', you could write: - - mov ax,Foo%[__BITS__] ; The Foo value - - to use the builtin macro `__BITS__' (see *note Section 4.11.5::) to -automatically select between them. Similarly, the two statements: - - %xdefine Bar Quux ; Expands due to %xdefine - %define Bar %[Quux] ; Expands due to %[...] - - have, in fact, exactly the same effect. - - `%[...]' concatenates to adjacent tokens in the same way that multi- -line macro parameters do, see *note Section 4.3.8:: for details. - - -File: nasm.info, Node: Section 4.1.4, Next: Section 4.1.5, Prev: Section 4.1.3, Up: Section 4.1 - -4.1.4. Concatenating Single Line Macro Tokens: `%+' ---------------------------------------------------- - -Individual tokens in single line macros can be concatenated, to produce -longer tokens for later processing. This can be useful if there are -several similar macros that perform similar functions. - - Please note that a space is required after `%+', in order to -disambiguate it from the syntax `%+1' used in multiline macros. - - As an example, consider the following: - - %define BDASTART 400h ; Start of BIOS data area - - struc tBIOSDA ; its structure - .COM1addr RESW 1 - .COM2addr RESW 1 - ; ..and so on - endstruc - - Now, if we need to access the elements of tBIOSDA in different -places, we can end up with: - - mov ax,BDASTART + tBIOSDA.COM1addr - mov bx,BDASTART + tBIOSDA.COM2addr - - This will become pretty ugly (and tedious) if used in many places, -and can be reduced in size significantly by using the following macro: - - ; Macro to access BIOS variables by their names (from tBDA): - - %define BDA(x) BDASTART + tBIOSDA. %+ x - - Now the above code can be written as: - - mov ax,BDA(COM1addr) - mov bx,BDA(COM2addr) - - Using this feature, we can simplify references to a lot of macros -(and, in turn, reduce typing errors). - - -File: nasm.info, Node: Section 4.1.5, Next: Section 4.1.6, Prev: Section 4.1.4, Up: Section 4.1 - -4.1.5. The Macro Name Itself: `%?' and `%??' --------------------------------------------- - -The special symbols `%?' and `%??' can be used to reference the macro -name itself inside a macro expansion, this is supported for both -single-and multi-line macros. `%?' refers to the macro name as -_invoked_, whereas `%??' refers to the macro name as _declared_. The -two are always the same for case-sensitive macros, but for -case-insensitive macros, they can differ. - - For example: - - %idefine Foo mov %?,%?? - - foo - FOO - - will expand to: - - mov foo,Foo - mov FOO,Foo - - The sequence: - - %idefine keyword $%? - - can be used to make a keyword "disappear", for example in case a new -instruction has been used as a label in older code. For example: - - %idefine pause $%? ; Hide the PAUSE instruction - - -File: nasm.info, Node: Section 4.1.6, Next: Section 4.1.7, Prev: Section 4.1.5, Up: Section 4.1 - -4.1.6. Undefining Single-Line Macros: `%undef' ----------------------------------------------- - -Single-line macros can be removed with the `%undef' directive. For -example, the following sequence: - - %define foo bar - %undef foo - - mov eax, foo - - will expand to the instruction `mov eax, foo', since after `%undef' -the macro `foo' is no longer defined. - - Macros that would otherwise be pre-defined can be undefined on the -command- line using the `-u' option on the NASM command line: see *note -Section 2.1.19::. - - -File: nasm.info, Node: Section 4.1.7, Next: Section 4.1.8, Prev: Section 4.1.6, Up: Section 4.1 - -4.1.7. Preprocessor Variables: `%assign' ----------------------------------------- - -An alternative way to define single-line macros is by means of the -`%assign' command (and its case-insensitive counterpart `%iassign', -which differs from `%assign' in exactly the same way that `%idefine' -differs from `%define'). - - `%assign' is used to define single-line macros which take no -parameters and have a numeric value. This value can be specified in the -form of an expression, and it will be evaluated once, when the -`%assign' directive is processed. - - Like `%define', macros defined using `%assign' can be re-defined -later, so you can do things like - - %assign i i+1 - - to increment the numeric value of a macro. - - `%assign' is useful for controlling the termination of `%rep' -preprocessor loops: see *note Section 4.5:: for an example of this. -Another use for `%assign' is given in *note Section 8.4:: and *note -Section 9.1::. - - The expression passed to `%assign' is a critical expression (see -*note Section 3.8::), and must also evaluate to a pure number (rather -than a relocatable reference such as a code or data address, or -anything involving a register). - - -File: nasm.info, Node: Section 4.1.8, Next: Section 4.1.9, Prev: Section 4.1.7, Up: Section 4.1 - -4.1.8. Defining Strings: `%defstr' ----------------------------------- - -`%defstr', and its case-insensitive counterpart `%idefstr', define or -redefine a single-line macro without parameters but converts the entire -right-hand side, after macro expansion, to a quoted string before -definition. - - For example: - - %defstr test TEST - - is equivalent to - - %define test 'TEST' - - This can be used, for example, with the `%!' construct (see *note -Section 4.10.2::): - - %defstr PATH %!PATH ; The operating system PATH variable - - -File: nasm.info, Node: Section 4.1.9, Next: Section 4.2, Prev: Section 4.1.8, Up: Section 4.1 - -4.1.9. Defining Tokens: `%deftok' ---------------------------------- - -`%deftok', and its case-insensitive counterpart `%ideftok', define or -redefine a single-line macro without parameters but converts the second -parameter, after string conversion, to a sequence of tokens. - - For example: - - %deftok test 'TEST' - - is equivalent to - - %define test TEST - - -File: nasm.info, Node: Section 4.2, Next: Section 4.2.1, Prev: Section 4.1.9, Up: Chapter 4 - -4.2. String Manipulation in Macros -================================== - -It's often useful to be able to handle strings in macros. NASM supports -a few simple string handling macro operators from which more complex -operations can be constructed. - - All the string operators define or redefine a value (either a string -or a numeric value) to a single-line macro. When producing a string -value, it may change the style of quoting of the input string or -strings, and possibly use `\'-escapes inside ``'-quoted strings. - -* Menu: - -* Section 4.2.1:: Concatenating Strings: `%strcat' -* Section 4.2.2:: String Length: `%strlen' -* Section 4.2.3:: Extracting Substrings: `%substr' - - -File: nasm.info, Node: Section 4.2.1, Next: Section 4.2.2, Prev: Section 4.2, Up: Section 4.2 - -4.2.1. Concatenating Strings: `%strcat' ---------------------------------------- - -The `%strcat' operator concatenates quoted strings and assign them to a -single-line macro. - - For example: - - %strcat alpha "Alpha: ", '12" screen' - - ... would assign the value `'Alpha: 12" screen'' to `alpha'. -Similarly: - - %strcat beta '"foo"\', "'bar'" - - ... would assign the value ``"foo"\\'bar'`' to `beta'. - - The use of commas to separate strings is permitted but optional. - - -File: nasm.info, Node: Section 4.2.2, Next: Section 4.2.3, Prev: Section 4.2.1, Up: Section 4.2 - -4.2.2. String Length: `%strlen' -------------------------------- - -The `%strlen' operator assigns the length of a string to a macro. For -example: - - %strlen charcnt 'my string' - - In this example, `charcnt' would receive the value 9, just as if an -`%assign' had been used. In this example, `'my string'' was a literal -string but it could also have been a single-line macro that expands to -a string, as in the following example: - - %define sometext 'my string' - %strlen charcnt sometext - - As in the first case, this would result in `charcnt' being assigned -the value of 9. - - -File: nasm.info, Node: Section 4.2.3, Next: Section 4.3, Prev: Section 4.2.2, Up: Section 4.2 - -4.2.3. Extracting Substrings: `%substr' ---------------------------------------- - -Individual letters or substrings in strings can be extracted using the -`%substr' operator. An example of its use is probably more useful than -the description: - - %substr mychar 'xyzw' 1 ; equivalent to %define mychar 'x' - %substr mychar 'xyzw' 2 ; equivalent to %define mychar 'y' - %substr mychar 'xyzw' 3 ; equivalent to %define mychar 'z' - %substr mychar 'xyzw' 2,2 ; equivalent to %define mychar 'yz' - %substr mychar 'xyzw' 2,-1 ; equivalent to %define mychar 'yzw' - %substr mychar 'xyzw' 2,-2 ; equivalent to %define mychar 'yz' - - As with `%strlen' (see *note Section 4.2.2::), the first parameter is -the single-line macro to be created and the second is the string. The -third parameter specifies the first character to be selected, and the -optional fourth parameter preceeded by comma) is the length. Note that -the first index is 1, not 0 and the last index is equal to the value -that `%strlen' would assign given the same string. Index values out of -range result in an empty string. A negative length means "until N-1 -characters before the end of string", i.e. `-1' means until end of -string, `-2' until one character before, etc. - - -File: nasm.info, Node: Section 4.3, Next: Section 4.3.1, Prev: Section 4.2.3, Up: Chapter 4 - -4.3. Multi-Line Macros: `%macro' -================================ - -Multi-line macros are much more like the type of macro seen in MASM and -TASM: a multi-line macro definition in NASM looks something like this. - - %macro prologue 1 - - push ebp - mov ebp,esp - sub esp,%1 - - %endmacro - - This defines a C-like function prologue as a macro: so you would -invoke the macro with a call such as - - myfunc: prologue 12 - - which would expand to the three lines of code - - myfunc: push ebp - mov ebp,esp - sub esp,12 - - The number `1' after the macro name in the `%macro' line defines the -number of parameters the macro `prologue' expects to receive. The use -of `%1' inside the macro definition refers to the first parameter to -the macro call. With a macro taking more than one parameter, subsequent -parameters would be referred to as `%2', `%3' and so on. - - Multi-line macros, like single-line macros, are case-sensitive, -unless you define them using the alternative directive `%imacro'. - - If you need to pass a comma as _part_ of a parameter to a multi-line -macro, you can do that by enclosing the entire parameter in braces. So -you could code things like - - %macro silly 2 - - %2: db %1 - - %endmacro - - silly 'a', letter_a ; letter_a: db 'a' - silly 'ab', string_ab ; string_ab: db 'ab' - silly {13,10}, crlf ; crlf: db 13,10 - -* Menu: - -* Section 4.3.1:: Recursive Multi-Line Macros: `%rmacro' -* Section 4.3.2:: Overloading Multi-Line Macros -* Section 4.3.3:: Macro-Local Labels -* Section 4.3.4:: Greedy Macro Parameters -* Section 4.3.5:: Default Macro Parameters -* Section 4.3.6:: `%0': Macro Parameter Counter -* Section 4.3.7:: `%rotate': Rotating Macro Parameters -* Section 4.3.8:: Concatenating Macro Parameters -* Section 4.3.9:: Condition Codes as Macro Parameters -* Section 4.3.10:: Disabling Listing Expansion -* Section 4.3.11:: Undefining Multi-Line Macros: `%unmacro' -* Section 4.3.12:: Exiting Multi-Line Macros: `%exitmacro' - - -File: nasm.info, Node: Section 4.3.1, Next: Section 4.3.2, Prev: Section 4.3, Up: Section 4.3 - -4.3.1. Recursive Multi-Line Macros: `%rmacro' ---------------------------------------------- - -A multi-line macro cannot be referenced within itself, in order to -prevent accidental infinite recursion. - - Recursive multi-line macros allow for self-referencing, with the -caveat that the user is aware of the existence, use and purpose of -recursive multi-line macros. There is also a generous, but sane, upper -limit to the number of recursions, in order to prevent run-away memory -consumption in case of accidental infinite recursion. - - As with non-recursive multi-line macros, recursive multi-line macros -are case-sensitive, unless you define them using the alternative -directive `%irmacro'. - - -File: nasm.info, Node: Section 4.3.2, Next: Section 4.3.3, Prev: Section 4.3.1, Up: Section 4.3 - -4.3.2. Overloading Multi-Line Macros ------------------------------------- - -As with single-line macros, multi-line macros can be overloaded by -defining the same macro name several times with different numbers of -parameters. This time, no exception is made for macros with no -parameters at all. So you could define - - %macro prologue 0 - - push ebp - mov ebp,esp - - %endmacro - - to define an alternative form of the function prologue which -allocates no local stack space. - - Sometimes, however, you might want to `overload' a machine -instruction; for example, you might want to define - - %macro push 2 - - push %1 - push %2 - - %endmacro - - so that you could code - - push ebx ; this line is not a macro call - push eax,ecx ; but this one is - - Ordinarily, NASM will give a warning for the first of the above two -lines, since `push' is now defined to be a macro, and is being invoked -with a number of parameters for which no definition has been given. The -correct code will still be generated, but the assembler will give a -warning. This warning can be disabled by the use of the -`-w-macro-params' command- line option (see *note Section 2.1.24::). - - -File: nasm.info, Node: Section 4.3.3, Next: Section 4.3.4, Prev: Section 4.3.2, Up: Section 4.3 - -4.3.3. Macro-Local Labels -------------------------- - -NASM allows you to define labels within a multi-line macro definition in -such a way as to make them local to the macro call: so calling the same -macro multiple times will use a different label each time. You do this -by prefixing `%%' to the label name. So you can invent an instruction -which executes a `RET' if the `Z' flag is set by doing this: - - %macro retz 0 - - jnz %%skip - ret - %%skip: - - %endmacro - - You can call this macro as many times as you want, and every time -you call it NASM will make up a different `real' name to substitute for -the label `%%skip'. The names NASM invents are of the form -`..@2345.skip', where the number 2345 changes with every macro call. -The `..@' prefix prevents macro-local labels from interfering with the -local label mechanism, as described in *note Section 3.9::. You should -avoid defining your own labels in this form (the `..@' prefix, then a -number, then another period) in case they interfere with macro-local -labels. - - -File: nasm.info, Node: Section 4.3.4, Next: Section 4.3.5, Prev: Section 4.3.3, Up: Section 4.3 - -4.3.4. Greedy Macro Parameters ------------------------------- - -Occasionally it is useful to define a macro which lumps its entire -command line into one parameter definition, possibly after extracting -one or two smaller parameters from the front. An example might be a -macro to write a text string to a file in MS-DOS, where you might want -to be able to write - - writefile [filehandle],"hello, world",13,10 - - NASM allows you to define the last parameter of a macro to be -_greedy_, meaning that if you invoke the macro with more parameters -than it expects, all the spare parameters get lumped into the last -defined one along with the separating commas. So if you code: - - %macro writefile 2+ - - jmp %%endstr - %%str: db %2 - %%endstr: - mov dx,%%str - mov cx,%%endstr-%%str - mov bx,%1 - mov ah,0x40 - int 0x21 - - %endmacro - - then the example call to `writefile' above will work as expected: the -text before the first comma, `[filehandle]', is used as the first macro -parameter and expanded when `%1' is referred to, and all the subsequent -text is lumped into `%2' and placed after the `db'. - - The greedy nature of the macro is indicated to NASM by the use of the -`+' sign after the parameter count on the `%macro' line. - - If you define a greedy macro, you are effectively telling NASM how it -should expand the macro given _any_ number of parameters from the -actual number specified up to infinity; in this case, for example, NASM -now knows what to do when it sees a call to `writefile' with 2, 3, 4 or -more parameters. NASM will take this into account when overloading -macros, and will not allow you to define another form of `writefile' -taking 4 parameters (for example). - - Of course, the above macro could have been implemented as a -non-greedy macro, in which case the call to it would have had to look -like - - writefile [filehandle], {"hello, world",13,10} - - NASM provides both mechanisms for putting commas in macro -parameters, and you choose which one you prefer for each macro -definition. - - See *note Section 6.3.1:: for a better way to write the above macro. - - -File: nasm.info, Node: Section 4.3.5, Next: Section 4.3.6, Prev: Section 4.3.4, Up: Section 4.3 - -4.3.5. Default Macro Parameters -------------------------------- - -NASM also allows you to define a multi-line macro with a _range_ of -allowable parameter counts. If you do this, you can specify defaults for -omitted parameters. So, for example: - - %macro die 0-1 "Painful program death has occurred." - - writefile 2,%1 - mov ax,0x4c01 - int 0x21 - - %endmacro - - This macro (which makes use of the `writefile' macro defined in -*note Section 4.3.4::) can be called with an explicit error message, -which it will display on the error output stream before exiting, or it -can be called with no parameters, in which case it will use the default -error message supplied in the macro definition. - - In general, you supply a minimum and maximum number of parameters -for a macro of this type; the minimum number of parameters are then -required in the macro call, and then you provide defaults for the -optional ones. So if a macro definition began with the line - - %macro foobar 1-3 eax,[ebx+2] - - then it could be called with between one and three parameters, and -`%1' would always be taken from the macro call. `%2', if not specified -by the macro call, would default to `eax', and `%3' if not specified -would default to `[ebx+2]'. - - You can provide extra information to a macro by providing too many -default parameters: - - %macro quux 1 something - - This will trigger a warning by default; see *note Section 2.1.24:: -for more information. When `quux' is invoked, it receives not one but -two parameters. `something' can be referred to as `%2'. The difference -between passing `something' this way and writing `something' in the -macro body is that with this way `something' is evaluated when the -macro is defined, not when it is expanded. - - You may omit parameter defaults from the macro definition, in which -case the parameter default is taken to be blank. This can be useful for -macros which can take a variable number of parameters, since the `%0' -token (see *note Section 4.3.6::) allows you to determine how many -parameters were really passed to the macro call. - - This defaulting mechanism can be combined with the greedy-parameter -mechanism; so the `die' macro above could be made more powerful, and -more useful, by changing the first line of the definition to - - %macro die 0-1+ "Painful program death has occurred.",13,10 - - The maximum parameter count can be infinite, denoted by `*'. In this -case, of course, it is impossible to provide a _full_ set of default -parameters. Examples of this usage are shown in *note Section 4.3.7::. - - -File: nasm.info, Node: Section 4.3.6, Next: Section 4.3.7, Prev: Section 4.3.5, Up: Section 4.3 - -4.3.6. `%0': Macro Parameter Counter ------------------------------------- - -The parameter reference `%0' will return a numeric constant giving the -number of parameters received, that is, if `%0' is n then `%'n is the -last parameter. `%0' is mostly useful for macros that can take a -variable number of parameters. It can be used as an argument to `%rep' -(see *note Section 4.5::) in order to iterate through all the -parameters of a macro. Examples are given in *note Section 4.3.7::. - - -File: nasm.info, Node: Section 4.3.7, Next: Section 4.3.8, Prev: Section 4.3.6, Up: Section 4.3 - -4.3.7. `%rotate': Rotating Macro Parameters -------------------------------------------- - -Unix shell programmers will be familiar with the `shift' shell command, -which allows the arguments passed to a shell script (referenced as -`$1', `$2' and so on) to be moved left by one place, so that the -argument previously referenced as `$2' becomes available as `$1', and -the argument previously referenced as `$1' is no longer available at -all. - - NASM provides a similar mechanism, in the form of `%rotate'. As its -name suggests, it differs from the Unix `shift' in that no parameters -are lost: parameters rotated off the left end of the argument list -reappear on the right, and vice versa. - - `%rotate' is invoked with a single numeric argument (which may be an -expression). The macro parameters are rotated to the left by that many -places. If the argument to `%rotate' is negative, the macro parameters -are rotated to the right. - - So a pair of macros to save and restore a set of registers might -work as follows: - - %macro multipush 1-* - - %rep %0 - push %1 - %rotate 1 - %endrep - - %endmacro - - This macro invokes the `PUSH' instruction on each of its arguments in -turn, from left to right. It begins by pushing its first argument, -`%1', then invokes `%rotate' to move all the arguments one place to the -left, so that the original second argument is now available as `%1'. -Repeating this procedure as many times as there were arguments -(achieved by supplying `%0' as the argument to `%rep') causes each -argument in turn to be pushed. - - Note also the use of `*' as the maximum parameter count, indicating -that there is no upper limit on the number of parameters you may supply -to the `multipush' macro. - - It would be convenient, when using this macro, to have a `POP' -equivalent, which _didn't_ require the arguments to be given in reverse -order. Ideally, you would write the `multipush' macro call, then -cut-and-paste the line to where the pop needed to be done, and change -the name of the called macro to `multipop', and the macro would take -care of popping the registers in the opposite order from the one in -which they were pushed. - - This can be done by the following definition: - - %macro multipop 1-* - - %rep %0 - %rotate -1 - pop %1 - %endrep - - %endmacro - - This macro begins by rotating its arguments one place to the _right_, -so that the original _last_ argument appears as `%1'. This is then -popped, and the arguments are rotated right again, so the second-to- -last argument becomes `%1'. Thus the arguments are iterated through in -reverse order. - - -File: nasm.info, Node: Section 4.3.8, Next: Section 4.3.9, Prev: Section 4.3.7, Up: Section 4.3 - -4.3.8. Concatenating Macro Parameters -------------------------------------- - -NASM can concatenate macro parameters and macro indirection constructs -on to other text surrounding them. This allows you to declare a family -of symbols, for example, in a macro definition. If, for example, you -wanted to generate a table of key codes along with offsets into the -table, you could code something like - - %macro keytab_entry 2 - - keypos%1 equ $-keytab - db %2 - - %endmacro - - keytab: - keytab_entry F1,128+1 - keytab_entry F2,128+2 - keytab_entry Return,13 - - which would expand to - - keytab: - keyposF1 equ $-keytab - db 128+1 - keyposF2 equ $-keytab - db 128+2 - keyposReturn equ $-keytab - db 13 - - You can just as easily concatenate text on to the other end of a -macro parameter, by writing `%1foo'. - - If you need to append a _digit_ to a macro parameter, for example -defining labels `foo1' and `foo2' when passed the parameter `foo', you -can't code `%11' because that would be taken as the eleventh macro -parameter. Instead, you must code `%{1}1', which will separate the -first `1' (giving the number of the macro parameter) from the second -(literal text to be concatenated to the parameter). - - This concatenation can also be applied to other preprocessor in-line -objects, such as macro-local labels (*note Section 4.3.3::) and -context-local labels (*note Section 4.7.2::). In all cases, ambiguities -in syntax can be resolved by enclosing everything after the `%' sign -and before the literal text in braces: so `%{%foo}bar' concatenates the -text `bar' to the end of the real name of the macro-local label -`%%foo'. (This is unnecessary, since the form NASM uses for the real -names of macro-local labels means that the two usages `%{%foo}bar' and -`%%foobar' would both expand to the same thing anyway; nevertheless, -the capability is there.) - - The single-line macro indirection construct, `%[...]' (*note Section -4.1.3::), behaves the same way as macro parameters for the purpose of -concatenation. - - See also the `%+' operator, *note Section 4.1.4::. - - -File: nasm.info, Node: Section 4.3.9, Next: Section 4.3.10, Prev: Section 4.3.8, Up: Section 4.3 - -4.3.9. Condition Codes as Macro Parameters ------------------------------------------- - -NASM can give special treatment to a macro parameter which contains a -condition code. For a start, you can refer to the macro parameter `%1' -by means of the alternative syntax `%+1', which informs NASM that this -macro parameter is supposed to contain a condition code, and will cause -the preprocessor to report an error message if the macro is called with -a parameter which is _not_ a valid condition code. - - Far more usefully, though, you can refer to the macro parameter by -means of `%-1', which NASM will expand as the _inverse_ condition code. -So the `retz' macro defined in *note Section 4.3.3:: can be replaced by -a general conditional-return macro like this: - - %macro retc 1 - - j%-1 %%skip - ret - %%skip: - - %endmacro - - This macro can now be invoked using calls like `retc ne', which will -cause the conditional-jump instruction in the macro expansion to come -out as `JE', or `retc po' which will make the jump a `JPE'. - - The `%+1' macro-parameter reference is quite happy to interpret the -arguments `CXZ' and `ECXZ' as valid condition codes; however, `%-1' -will report an error if passed either of these, because no inverse -condition code exists. - - -File: nasm.info, Node: Section 4.3.10, Next: Section 4.3.11, Prev: Section 4.3.9, Up: Section 4.3 - -4.3.10. Disabling Listing Expansion ------------------------------------ - -When NASM is generating a listing file from your program, it will -generally expand multi-line macros by means of writing the macro call -and then listing each line of the expansion. This allows you to see -which instructions in the macro expansion are generating what code; -however, for some macros this clutters the listing up unnecessarily. - - NASM therefore provides the `.nolist' qualifier, which you can -include in a macro definition to inhibit the expansion of the macro in -the listing file. The `.nolist' qualifier comes directly after the -number of parameters, like this: - - %macro foo 1.nolist - - Or like this: - - %macro bar 1-5+.nolist a,b,c,d,e,f,g,h - - -File: nasm.info, Node: Section 4.3.11, Next: Section 4.3.12, Prev: Section 4.3.10, Up: Section 4.3 - -4.3.11. Undefining Multi-Line Macros: `%unmacro' ------------------------------------------------- - -Multi-line macros can be removed with the `%unmacro' directive. Unlike -the `%undef' directive, however, `%unmacro' takes an argument -specification, and will only remove exact matches with that argument -specification. - - For example: - - %macro foo 1-3 - ; Do something - %endmacro - %unmacro foo 1-3 - - removes the previously defined macro `foo', but - - %macro bar 1-3 - ; Do something - %endmacro - %unmacro bar 1 - - does _not_ remove the macro `bar', since the argument specification -does not match exactly. - - -File: nasm.info, Node: Section 4.3.12, Next: Section 4.4, Prev: Section 4.3.11, Up: Section 4.3 - -4.3.12. Exiting Multi-Line Macros: `%exitmacro' ------------------------------------------------ - -Multi-line macro expansions can be arbitrarily terminated with the -`%exitmacro' directive. - - For example: - - %macro foo 1-3 - ; Do something - %if<condition> - %exitmacro - %endif - ; Do something - %endmacro - - -File: nasm.info, Node: Section 4.4, Next: Section 4.4.1, Prev: Section 4.3.12, Up: Chapter 4 - -4.4. Conditional Assembly -========================= - -Similarly to the C preprocessor, NASM allows sections of a source file -to be assembled only if certain conditions are met. The general syntax -of this feature looks like this: - - %if<condition> - ; some code which only appears if <condition> is met - %elif<condition2> - ; only appears if <condition> is not met but <condition2> is - %else - ; this appears if neither <condition> nor <condition2> was met - %endif - - The inverse forms `%ifn' and `%elifn' are also supported. - - The `%else' clause is optional, as is the `%elif' clause. You can -have more than one `%elif' clause as well. - - There are a number of variants of the `%if' directive. Each has its -corresponding `%elif', `%ifn', and `%elifn' directives; for example, -the equivalents to the `%ifdef' directive are `%elifdef', `%ifndef', -and `%elifndef'. - -* Menu: - -* Section 4.4.1:: `%ifdef': Testing Single-Line Macro Existence -* Section 4.4.2:: `%ifmacro': Testing Multi-Line Macro Existence -* Section 4.4.3:: `%ifctx': Testing the Context Stack -* Section 4.4.4:: `%if': Testing Arbitrary Numeric Expressions -* Section 4.4.5:: `%ifidn' and `%ifidni': Testing Exact Text Identity -* Section 4.4.6:: `%ifid', `%ifnum', `%ifstr': Testing Token Types -* Section 4.4.7:: `%iftoken': Test for a Single Token -* Section 4.4.8:: `%ifempty': Test for Empty Expansion - - -File: nasm.info, Node: Section 4.4.1, Next: Section 4.4.2, Prev: Section 4.4, Up: Section 4.4 - -4.4.1. `%ifdef': Testing Single-Line Macro Existence ----------------------------------------------------- - -Beginning a conditional-assembly block with the line `%ifdef MACRO' -will assemble the subsequent code if, and only if, a single-line macro -called `MACRO' is defined. If not, then the `%elif' and `%else' blocks -(if any) will be processed instead. - - For example, when debugging a program, you might want to write code -such as - - ; perform some function - %ifdef DEBUG - writefile 2,"Function performed successfully",13,10 - %endif - ; go and do something else - - Then you could use the command-line option `-dDEBUG' to create a -version of the program which produced debugging messages, and remove the -option to generate the final release version of the program. - - You can test for a macro _not_ being defined by using `%ifndef' -instead of `%ifdef'. You can also test for macro definitions in `%elif' -blocks by using `%elifdef' and `%elifndef'. - - -File: nasm.info, Node: Section 4.4.2, Next: Section 4.4.3, Prev: Section 4.4.1, Up: Section 4.4 - -4.4.2. `%ifmacro': Testing Multi-Line Macro Existence ------------------------------------------------------ - -The `%ifmacro' directive operates in the same way as the `%ifdef' -directive, except that it checks for the existence of a multi-line -macro. - - For example, you may be working with a large project and not have -control over the macros in a library. You may want to create a macro -with one name if it doesn't already exist, and another name if one with -that name does exist. - - The `%ifmacro' is considered true if defining a macro with the given -name and number of arguments would cause a definitions conflict. For -example: - - %ifmacro MyMacro 1-3 - - %error "MyMacro 1-3" causes a conflict with an existing macro. - - %else - - %macro MyMacro 1-3 - - ; insert code to define the macro - - %endmacro - - %endif - - This will create the macro "MyMacro 1-3" if no macro already exists -which would conflict with it, and emits a warning if there would be a -definition conflict. - - You can test for the macro not existing by using the `%ifnmacro' -instead of `%ifmacro'. Additional tests can be performed in `%elif' -blocks by using `%elifmacro' and `%elifnmacro'. - - -File: nasm.info, Node: Section 4.4.3, Next: Section 4.4.4, Prev: Section 4.4.2, Up: Section 4.4 - -4.4.3. `%ifctx': Testing the Context Stack ------------------------------------------- - -The conditional-assembly construct `%ifctx' will cause the subsequent -code to be assembled if and only if the top context on the -preprocessor's context stack has the same name as one of the arguments. -As with `%ifdef', the inverse and `%elif' forms `%ifnctx', `%elifctx' -and `%elifnctx' are also supported. - - For more details of the context stack, see *note Section 4.7::. For -a sample use of `%ifctx', see *note Section 4.7.5::. - - -File: nasm.info, Node: Section 4.4.4, Next: Section 4.4.5, Prev: Section 4.4.3, Up: Section 4.4 - -4.4.4. `%if': Testing Arbitrary Numeric Expressions ---------------------------------------------------- - -The conditional-assembly construct `%if expr' will cause the subsequent -code to be assembled if and only if the value of the numeric expression -`expr' is non-zero. An example of the use of this feature is in -deciding when to break out of a `%rep' preprocessor loop: see *note -Section 4.5:: for a detailed example. - - The expression given to `%if', and its counterpart `%elif', is a -critical expression (see *note Section 3.8::). - - `%if' extends the normal NASM expression syntax, by providing a set -of relational operators which are not normally available in -expressions. The operators `=', `<', `>', `<=', `>=' and `<>' test -equality, less-than, greater-than, less-or-equal, greater-or-equal and -not-equal respectively. The C-like forms `==' and `!=' are supported as -alternative forms of `=' and `<>'. In addition, low- priority logical -operators `&&', `^^' and `||' are provided, supplying logical AND, -logical XOR and logical OR. These work like the C logical operators -(although C has no logical XOR), in that they always return either 0 or -1, and treat any non-zero input as 1 (so that `^^', for example, -returns 1 if exactly one of its inputs is zero, and 0 otherwise). The -relational operators also return 1 for true and 0 for false. - - Like other `%if' constructs, `%if' has a counterpart `%elif', and -negative forms `%ifn' and `%elifn'. - - -File: nasm.info, Node: Section 4.4.5, Next: Section 4.4.6, Prev: Section 4.4.4, Up: Section 4.4 - -4.4.5. `%ifidn' and `%ifidni': Testing Exact Text Identity ----------------------------------------------------------- - -The construct `%ifidn text1,text2' will cause the subsequent code to be -assembled if and only if `text1' and `text2', after expanding -single-line macros, are identical pieces of text. Differences in white -space are not counted. - - `%ifidni' is similar to `%ifidn', but is case-insensitive. - - For example, the following macro pushes a register or number on the -stack, and allows you to treat `IP' as a real register: - - %macro pushparam 1 - - %ifidni %1,ip - call %%label - %%label: - %else - push %1 - %endif - - %endmacro - - Like other `%if' constructs, `%ifidn' has a counterpart `%elifidn', -and negative forms `%ifnidn' and `%elifnidn'. Similarly, `%ifidni' has -counterparts `%elifidni', `%ifnidni' and `%elifnidni'. - - -File: nasm.info, Node: Section 4.4.6, Next: Section 4.4.7, Prev: Section 4.4.5, Up: Section 4.4 - -4.4.6. `%ifid', `%ifnum', `%ifstr': Testing Token Types -------------------------------------------------------- - -Some macros will want to perform different tasks depending on whether -they are passed a number, a string, or an identifier. For example, a -string output macro might want to be able to cope with being passed -either a string constant or a pointer to an existing string. - - The conditional assembly construct `%ifid', taking one parameter -(which may be blank), assembles the subsequent code if and only if the -first token in the parameter exists and is an identifier. `%ifnum' -works similarly, but tests for the token being a numeric constant; -`%ifstr' tests for it being a string. - - For example, the `writefile' macro defined in *note Section 4.3.4:: -can be extended to take advantage of `%ifstr' in the following fashion: - - %macro writefile 2-3+ - - %ifstr %2 - jmp %%endstr - %if %0 = 3 - %%str: db %2,%3 - %else - %%str: db %2 - %endif - %%endstr: mov dx,%%str - mov cx,%%endstr-%%str - %else - mov dx,%2 - mov cx,%3 - %endif - mov bx,%1 - mov ah,0x40 - int 0x21 - - %endmacro - - Then the `writefile' macro can cope with being called in either of -the following two ways: - - writefile [file], strpointer, length - writefile [file], "hello", 13, 10 - - In the first, `strpointer' is used as the address of an already- -declared string, and `length' is used as its length; in the second, a -string is given to the macro, which therefore declares it itself and -works out the address and length for itself. - - Note the use of `%if' inside the `%ifstr': this is to detect whether -the macro was passed two arguments (so the string would be a single -string constant, and `db %2' would be adequate) or more (in which case, -all but the first two would be lumped together into `%3', and `db -%2,%3' would be required). - - The usual `%elif'..., `%ifn'..., and `%elifn'... versions exist for -each of `%ifid', `%ifnum' and `%ifstr'. - - -File: nasm.info, Node: Section 4.4.7, Next: Section 4.4.8, Prev: Section 4.4.6, Up: Section 4.4 - -4.4.7. `%iftoken': Test for a Single Token ------------------------------------------- - -Some macros will want to do different things depending on if it is -passed a single token (e.g. paste it to something else using `%+') -versus a multi-token sequence. - - The conditional assembly construct `%iftoken' assembles the -subsequent code if and only if the expanded parameters consist of -exactly one token, possibly surrounded by whitespace. - - For example: - - %iftoken 1 - - will assemble the subsequent code, but - - %iftoken -1 - - will not, since `-1' contains two tokens: the unary minus operator -`-', and the number `1'. - - The usual `%eliftoken', `%ifntoken', and `%elifntoken' variants are -also provided. - - -File: nasm.info, Node: Section 4.4.8, Next: Section 4.5, Prev: Section 4.4.7, Up: Section 4.4 - -4.4.8. `%ifempty': Test for Empty Expansion -------------------------------------------- - -The conditional assembly construct `%ifempty' assembles the subsequent -code if and only if the expanded parameters do not contain any tokens at -all, whitespace excepted. - - The usual `%elifempty', `%ifnempty', and `%elifnempty' variants are -also provided. - - -File: nasm.info, Node: Section 4.5, Next: Section 4.6, Prev: Section 4.4.8, Up: Chapter 4 - -4.5. Preprocessor Loops: `%rep' -=============================== - -NASM's `TIMES' prefix, though useful, cannot be used to invoke a -multi-line macro multiple times, because it is processed by NASM after -macros have already been expanded. Therefore NASM provides another form -of loop, this time at the preprocessor level: `%rep'. - - The directives `%rep' and `%endrep' (`%rep' takes a numeric -argument, which can be an expression; `%endrep' takes no arguments) can -be used to enclose a chunk of code, which is then replicated as many -times as specified by the preprocessor: - - %assign i 0 - %rep 64 - inc word [table+2*i] - %assign i i+1 - %endrep - - This will generate a sequence of 64 `INC' instructions, incrementing -every word of memory from `[table]' to `[table+126]'. - - For more complex termination conditions, or to break out of a repeat -loop part way along, you can use the `%exitrep' directive to terminate -the loop, like this: - - fibonacci: - %assign i 0 - %assign j 1 - %rep 100 - %if j > 65535 - %exitrep - %endif - dw j - %assign k j+i - %assign i j - %assign j k - %endrep - - fib_number equ ($-fibonacci)/2 - - This produces a list of all the Fibonacci numbers that will fit in -16 bits. Note that a maximum repeat count must still be given to -`%rep'. This is to prevent the possibility of NASM getting into an -infinite loop in the preprocessor, which (on multitasking or multi-user -systems) would typically cause all the system memory to be gradually -used up and other applications to start crashing. - - -File: nasm.info, Node: Section 4.6, Next: Section 4.6.1, Prev: Section 4.5, Up: Chapter 4 - -4.6. Source Files and Dependencies -================================== - -These commands allow you to split your sources into multiple files. - -* Menu: - -* Section 4.6.1:: `%include': Including Other Files -* Section 4.6.2:: `%pathsearch': Search the Include Path -* Section 4.6.3:: `%depend': Add Dependent Files -* Section 4.6.4:: `%use': Include Standard Macro Package - - -File: nasm.info, Node: Section 4.6.1, Next: Section 4.6.2, Prev: Section 4.6, Up: Section 4.6 - -4.6.1. `%include': Including Other Files ----------------------------------------- - -Using, once again, a very similar syntax to the C preprocessor, NASM's -preprocessor lets you include other source files into your code. This is -done by the use of the `%include' directive: - - %include "macros.mac" - - will include the contents of the file `macros.mac' into the source -file containing the `%include' directive. - - Include files are searched for in the current directory (the -directory you're in when you run NASM, as opposed to the location of -the NASM executable or the location of the source file), plus any -directories specified on the NASM command line using the `-i' option. - - The standard C idiom for preventing a file being included more than -once is just as applicable in NASM: if the file `macros.mac' has the -form - - %ifndef MACROS_MAC - %define MACROS_MAC - ; now define some macros - %endif - - then including the file more than once will not cause errors, -because the second time the file is included nothing will happen -because the macro `MACROS_MAC' will already be defined. - - You can force a file to be included even if there is no `%include' -directive that explicitly includes it, by using the `-p' option on the -NASM command line (see *note Section 2.1.17::). - - -File: nasm.info, Node: Section 4.6.2, Next: Section 4.6.3, Prev: Section 4.6.1, Up: Section 4.6 - -4.6.2. `%pathsearch': Search the Include Path ---------------------------------------------- - -The `%pathsearch' directive takes a single-line macro name and a -filename, and declare or redefines the specified single-line macro to be -the include-path-resolved version of the filename, if the file exists -(otherwise, it is passed unchanged.) - - For example, - - %pathsearch MyFoo "foo.bin" - - ... with `-Ibins/' in the include path may end up defining the macro -`MyFoo' to be `"bins/foo.bin"'. - - -File: nasm.info, Node: Section 4.6.3, Next: Section 4.6.4, Prev: Section 4.6.2, Up: Section 4.6 - -4.6.3. `%depend': Add Dependent Files -------------------------------------- - -The `%depend' directive takes a filename and adds it to the list of -files to be emitted as dependency generation when the `-M' options and -its relatives (see *note Section 2.1.4::) are used. It produces no -output. - - This is generally used in conjunction with `%pathsearch'. For -example, a simplified version of the standard macro wrapper for the -`INCBIN' directive looks like: - - %imacro incbin 1-2+ 0 - %pathsearch dep %1 - %depend dep - incbin dep,%2 - %endmacro - - This first resolves the location of the file into the macro `dep', -then adds it to the dependency lists, and finally issues the assembler- -level `INCBIN' directive. - - -File: nasm.info, Node: Section 4.6.4, Next: Section 4.7, Prev: Section 4.6.3, Up: Section 4.6 - -4.6.4. `%use': Include Standard Macro Package ---------------------------------------------- - -The `%use' directive is similar to `%include', but rather than -including the contents of a file, it includes a named standard macro -package. The standard macro packages are part of NASM, and are -described in *note Chapter 5::. - - Unlike the `%include' directive, package names for the `%use' -directive do not require quotes, but quotes are permitted. In NASM 2.04 -and 2.05 the unquoted form would be macro-expanded; this is no longer -true. Thus, the following lines are equivalent: - - %use altreg - %use 'altreg' - - Standard macro packages are protected from multiple inclusion. When a -standard macro package is used, a testable single-line macro of the form -`__USE_'_package_`__' is also defined, see *note Section 4.11.8::. - - -File: nasm.info, Node: Section 4.7, Next: Section 4.7.1, Prev: Section 4.6.4, Up: Chapter 4 - -4.7. The Context Stack -====================== - -Having labels that are local to a macro definition is sometimes not -quite powerful enough: sometimes you want to be able to share labels -between several macro calls. An example might be a `REPEAT' ... `UNTIL' -loop, in which the expansion of the `REPEAT' macro would need to be -able to refer to a label which the `UNTIL' macro had defined. However, -for such a macro you would also want to be able to nest these loops. - - NASM provides this level of power by means of a _context stack_. The -preprocessor maintains a stack of _contexts_, each of which is -characterized by a name. You add a new context to the stack using the -`%push' directive, and remove one using `%pop'. You can define labels -that are local to a particular context on the stack. - -* Menu: - -* Section 4.7.1:: `%push' and `%pop': Creating and Removing Contexts -* Section 4.7.2:: Context-Local Labels -* Section 4.7.3:: Context-Local Single-Line Macros -* Section 4.7.4:: `%repl': Renaming a Context -* Section 4.7.5:: Example Use of the Context Stack: Block IFs - - -File: nasm.info, Node: Section 4.7.1, Next: Section 4.7.2, Prev: Section 4.7, Up: Section 4.7 - -4.7.1. `%push' and `%pop': Creating and Removing Contexts ---------------------------------------------------------- - -The `%push' directive is used to create a new context and place it on -the top of the context stack. `%push' takes an optional argument, which -is the name of the context. For example: - - %push foobar - - This pushes a new context called `foobar' on the stack. You can have -several contexts on the stack with the same name: they can still be -distinguished. If no name is given, the context is unnamed (this is -normally used when both the `%push' and the `%pop' are inside a single -macro definition.) - - The directive `%pop', taking one optional argument, removes the top -context from the context stack and destroys it, along with any labels -associated with it. If an argument is given, it must match the name of -the current context, otherwise it will issue an error. - - -File: nasm.info, Node: Section 4.7.2, Next: Section 4.7.3, Prev: Section 4.7.1, Up: Section 4.7 - -4.7.2. Context-Local Labels ---------------------------- - -Just as the usage `%%foo' defines a label which is local to the -particular macro call in which it is used, the usage `%$foo' is used to -define a label which is local to the context on the top of the context -stack. So the `REPEAT' and `UNTIL' example given above could be -implemented by means of: - - %macro repeat 0 - - %push repeat - %$begin: - - %endmacro - - %macro until 1 - - j%-1 %$begin - %pop - - %endmacro - - and invoked by means of, for example, - - mov cx,string - repeat - add cx,3 - scasb - until e - - which would scan every fourth byte of a string in search of the byte -in `AL'. - - If you need to define, or access, labels local to the context _below_ -the top one on the stack, you can use `%$$foo', or `%$$$foo' for the -context below that, and so on. - - -File: nasm.info, Node: Section 4.7.3, Next: Section 4.7.4, Prev: Section 4.7.2, Up: Section 4.7 - -4.7.3. Context-Local Single-Line Macros ---------------------------------------- - -NASM also allows you to define single-line macros which are local to a -particular context, in just the same way: - - %define %$localmac 3 - - will define the single-line macro `%$localmac' to be local to the top -context on the stack. Of course, after a subsequent `%push', it can -then still be accessed by the name `%$$localmac'. - - -File: nasm.info, Node: Section 4.7.4, Next: Section 4.7.5, Prev: Section 4.7.3, Up: Section 4.7 - -4.7.4. `%repl': Renaming a Context ----------------------------------- - -If you need to change the name of the top context on the stack (in -order, for example, to have it respond differently to `%ifctx'), you can -execute a `%pop' followed by a `%push'; but this will have the side -effect of destroying all context-local labels and macros associated -with the context that was just popped. - - NASM provides the directive `%repl', which _replaces_ a context with -a different name, without touching the associated macros and labels. -So you could replace the destructive code - - %pop - %push newname - - with the non-destructive version `%repl newname'. - - -File: nasm.info, Node: Section 4.7.5, Next: Section 4.8, Prev: Section 4.7.4, Up: Section 4.7 - -4.7.5. Example Use of the Context Stack: Block IFs --------------------------------------------------- - -This example makes use of almost all the context-stack features, -including the conditional-assembly construct `%ifctx', to implement a -block IF statement as a set of macros. - - %macro if 1 - - %push if - j%-1 %$ifnot - - %endmacro - - %macro else 0 - - %ifctx if - %repl else - jmp %$ifend - %$ifnot: - %else - %error "expected `if' before `else'" - %endif - - %endmacro - - %macro endif 0 - - %ifctx if - %$ifnot: - %pop - %elifctx else - %$ifend: - %pop - %else - %error "expected `if' or `else' before `endif'" - %endif - - %endmacro - - This code is more robust than the `REPEAT' and `UNTIL' macros given -in *note Section 4.7.2::, because it uses conditional assembly to check -that the macros are issued in the right order (for example, not calling -`endif' before `if') and issues a `%error' if they're not. - - In addition, the `endif' macro has to be able to cope with the two -distinct cases of either directly following an `if', or following an -`else'. It achieves this, again, by using conditional assembly to do -different things depending on whether the context on top of the stack is -`if' or `else'. - - The `else' macro has to preserve the context on the stack, in order -to have the `%$ifnot' referred to by the `if' macro be the same as the -one defined by the `endif' macro, but has to change the context's name -so that `endif' will know there was an intervening `else'. It does -this by the use of `%repl'. - - A sample usage of these macros might look like: - - cmp ax,bx - - if ae - cmp bx,cx - - if ae - mov ax,cx - else - mov ax,bx - endif - - else - cmp ax,cx - - if ae - mov ax,cx - endif - - endif - - The block-`IF' macros handle nesting quite happily, by means of -pushing another context, describing the inner `if', on top of the one -describing the outer `if'; thus `else' and `endif' always refer to the -last unmatched `if' or `else'. - - -File: nasm.info, Node: Section 4.8, Next: Section 4.8.1, Prev: Section 4.7.5, Up: Chapter 4 - -4.8. Stack Relative Preprocessor Directives -=========================================== - -The following preprocessor directives provide a way to use labels to -refer to local variables allocated on the stack. - - * `%arg' (see *note Section 4.8.1::) - - * `%stacksize' (see *note Section 4.8.2::) - - * `%local' (see *note Section 4.8.3::) - -* Menu: - -* Section 4.8.1:: `%arg' Directive -* Section 4.8.2:: `%stacksize' Directive -* Section 4.8.3:: `%local' Directive - - -File: nasm.info, Node: Section 4.8.1, Next: Section 4.8.2, Prev: Section 4.8, Up: Section 4.8 - -4.8.1. `%arg' Directive ------------------------ - -The `%arg' directive is used to simplify the handling of parameters -passed on the stack. Stack based parameter passing is used by many high -level languages, including C, C++ and Pascal. - - While NASM has macros which attempt to duplicate this functionality -(see *note Section 8.4.5::), the syntax is not particularly convenient -to use. and is not TASM compatible. Here is an example which shows the -use of `%arg' without any external macros: - - some_function: - - %push mycontext ; save the current context - %stacksize large ; tell NASM to use bp - %arg i:word, j_ptr:word - - mov ax,[i] - mov bx,[j_ptr] - add ax,[bx] - ret - - %pop ; restore original context - - This is similar to the procedure defined in *note Section 8.4.5:: -and adds the value in i to the value pointed to by j_ptr and returns -the sum in the ax register. See *note Section 4.7.1:: for an -explanation of `push' and `pop' and the use of context stacks. - - -File: nasm.info, Node: Section 4.8.2, Next: Section 4.8.3, Prev: Section 4.8.1, Up: Section 4.8 - -4.8.2. `%stacksize' Directive ------------------------------ - -The `%stacksize' directive is used in conjunction with the `%arg' (see -*note Section 4.8.1::) and the `%local' (see *note Section 4.8.3::) -directives. It tells NASM the default size to use for subsequent `%arg' -and `%local' directives. The `%stacksize' directive takes one required -argument which is one of `flat', `flat64', `large' or `small'. - - %stacksize flat - - This form causes NASM to use stack-based parameter addressing -relative to `ebp' and it assumes that a near form of call was used to -get to this label (i.e. that `eip' is on the stack). - - %stacksize flat64 - - This form causes NASM to use stack-based parameter addressing -relative to `rbp' and it assumes that a near form of call was used to -get to this label (i.e. that `rip' is on the stack). - - %stacksize large - - This form uses `bp' to do stack-based parameter addressing and -assumes that a far form of call was used to get to this address (i.e. -that `ip' and `cs' are on the stack). - - %stacksize small - - This form also uses `bp' to address stack parameters, but it is -different from `large' because it also assumes that the old value of bp -is pushed onto the stack (i.e. it expects an `ENTER' instruction). In -other words, it expects that `bp', `ip' and `cs' are on the top of the -stack, underneath any local space which may have been allocated by -`ENTER'. This form is probably most useful when used in combination -with the `%local' directive (see *note Section 4.8.3::). - - -File: nasm.info, Node: Section 4.8.3, Next: Section 4.9, Prev: Section 4.8.2, Up: Section 4.8 - -4.8.3. `%local' Directive -------------------------- - -The `%local' directive is used to simplify the use of local temporary -stack variables allocated in a stack frame. Automatic local variables -in C are an example of this kind of variable. The `%local' directive is -most useful when used with the `%stacksize' (see *note Section 4.8.2:: -and is also compatible with the `%arg' directive (see *note Section -4.8.1::). It allows simplified reference to variables on the stack which -have been allocated typically by using the `ENTER' instruction. An -example of its use is the following: - - silly_swap: - - %push mycontext ; save the current context - %stacksize small ; tell NASM to use bp - %assign %$localsize 0 ; see text for explanation - %local old_ax:word, old_dx:word - - enter %$localsize,0 ; see text for explanation - mov [old_ax],ax ; swap ax & bx - mov [old_dx],dx ; and swap dx & cx - mov ax,bx - mov dx,cx - mov bx,[old_ax] - mov cx,[old_dx] - leave ; restore old bp - ret ; - - %pop ; restore original context - - The `%$localsize' variable is used internally by the `%local' -directive and _must_ be defined within the current context before the -`%local' directive may be used. Failure to do so will result in one -expression syntax error for each `%local' variable declared. It then -may be used in the construction of an appropriately sized ENTER -instruction as shown in the example. - - -File: nasm.info, Node: Section 4.9, Next: Section 4.10, Prev: Section 4.8.3, Up: Chapter 4 - -4.9. Reporting User-Defined Errors: `%error', `%warning', `%fatal' -================================================================== - -The preprocessor directive `%error' will cause NASM to report an error -if it occurs in assembled code. So if other users are going to try to -assemble your source files, you can ensure that they define the right -macros by means of code like this: - - %ifdef F1 - ; do some setup - %elifdef F2 - ; do some different setup - %else - %error "Neither F1 nor F2 was defined." - %endif - - Then any user who fails to understand the way your code is supposed -to be assembled will be quickly warned of their mistake, rather than -having to wait until the program crashes on being run and then not -knowing what went wrong. - - Similarly, `%warning' issues a warning, but allows assembly to -continue: - - %ifdef F1 - ; do some setup - %elifdef F2 - ; do some different setup - %else - %warning "Neither F1 nor F2 was defined, assuming F1." - %define F1 - %endif - - `%error' and `%warning' are issued only on the final assembly pass. -This makes them safe to use in conjunction with tests that depend on -symbol values. - - `%fatal' terminates assembly immediately, regardless of pass. This is -useful when there is no point in continuing the assembly further, and -doing so is likely just going to cause a spew of confusing error -messages. - - It is optional for the message string after `%error', `%warning' or -`%fatal' to be quoted. If it is _not_, then single-line macros are -expanded in it, which can be used to display more information to the -user. For example: - - %if foo > 64 - %assign foo_over foo-64 - %error foo is foo_over bytes too large - %endif - - -File: nasm.info, Node: Section 4.10, Next: Section 4.10.1, Prev: Section 4.9, Up: Chapter 4 - -4.10. Other Preprocessor Directives -=================================== - -NASM also has preprocessor directives which allow access to information -from external sources. Currently they include: - - * `%line' enables NASM to correctly handle the output of another - preprocessor (see *note Section 4.10.1::). - - * `%!' enables NASM to read in the value of an environment variable, - which can then be used in your program (see *note Section - 4.10.2::). - -* Menu: - -* Section 4.10.1:: `%line' Directive -* Section 4.10.2:: `%!'`<env>': Read an environment variable. - - -File: nasm.info, Node: Section 4.10.1, Next: Section 4.10.2, Prev: Section 4.10, Up: Section 4.10 - -4.10.1. `%line' Directive -------------------------- - -The `%line' directive is used to notify NASM that the input line -corresponds to a specific line number in another file. Typically this -other file would be an original source file, with the current NASM -input being the output of a pre-processor. The `%line' directive allows -NASM to output messages which indicate the line number of the original -source file, instead of the file that is being read by NASM. - - This preprocessor directive is not generally of use to programmers, -by may be of interest to preprocessor authors. The usage of the `%line' -preprocessor directive is as follows: - - %line nnn[+mmm] [filename] - - In this directive, `nnn' identifies the line of the original source -file which this line corresponds to. `mmm' is an optional parameter -which specifies a line increment value; each line of the input file -read in is considered to correspond to `mmm' lines of the original -source file. Finally, `filename' is an optional parameter which -specifies the file name of the original source file. - - After reading a `%line' preprocessor directive, NASM will report all -file name and line numbers relative to the values specified therein. - - -File: nasm.info, Node: Section 4.10.2, Next: Section 4.11, Prev: Section 4.10.1, Up: Section 4.10 - -4.10.2. `%!'`<env>': Read an environment variable. --------------------------------------------------- - -The `%!<env>' directive makes it possible to read the value of an -environment variable at assembly time. This could, for example, be used -to store the contents of an environment variable into a string, which -could be used at some other point in your code. - - For example, suppose that you have an environment variable `FOO', and -you want the contents of `FOO' to be embedded in your program. You -could do that as follows: - - %defstr FOO %!FOO - - See *note Section 4.1.8:: for notes on the `%defstr' directive. - - -File: nasm.info, Node: Section 4.11, Next: Section 4.11.1, Prev: Section 4.10.2, Up: Chapter 4 - -4.11. Standard Macros -===================== - -NASM defines a set of standard macros, which are already defined when it -starts to process any source file. If you really need a program to be -assembled with no pre-defined macros, you can use the `%clear' -directive to empty the preprocessor of everything but context-local -preprocessor variables and single-line macros. - - Most user-level assembler directives (see *note Chapter 6::) are -implemented as macros which invoke primitive directives; these are -described in *note Chapter 6::. The rest of the standard macro set is -described here. - -* Menu: - -* Section 4.11.1:: NASM Version Macros -* Section 4.11.2:: `__NASM_VERSION_ID__': NASM Version ID -* Section 4.11.3:: `__NASM_VER__': NASM Version string -* Section 4.11.4:: `__FILE__' and `__LINE__': File Name and Line Number -* Section 4.11.5:: `__BITS__': Current BITS Mode -* Section 4.11.6:: `__OUTPUT_FORMAT__': Current Output Format -* Section 4.11.7:: Assembly Date and Time Macros -* Section 4.11.8:: `__USE_'_package_`__': Package Include Test -* Section 4.11.9:: `__PASS__': Assembly Pass -* Section 4.11.10:: `STRUC' and `ENDSTRUC': Declaring Structure Data Types -* Section 4.11.11:: `ISTRUC', `AT' and `IEND': Declaring Instances of Structures -* Section 4.11.12:: `ALIGN' and `ALIGNB': Data Alignment - - -File: nasm.info, Node: Section 4.11.1, Next: Section 4.11.2, Prev: Section 4.11, Up: Section 4.11 - -4.11.1. NASM Version Macros ---------------------------- - -The single-line macros `__NASM_MAJOR__', `__NASM_MINOR__', -`__NASM_SUBMINOR__' and `___NASM_PATCHLEVEL__' expand to the major, -minor, subminor and patch level parts of the version number of NASM -being used. So, under NASM 0.98.32p1 for example, `__NASM_MAJOR__' -would be defined to be 0, `__NASM_MINOR__' would be defined as 98, -`__NASM_SUBMINOR__' would be defined to 32, and `___NASM_PATCHLEVEL__' -would be defined as 1. - - Additionally, the macro `__NASM_SNAPSHOT__' is defined for -automatically generated snapshot releases _only_. - - -File: nasm.info, Node: Section 4.11.2, Next: Section 4.11.3, Prev: Section 4.11.1, Up: Section 4.11 - -4.11.2. `__NASM_VERSION_ID__': NASM Version ID ----------------------------------------------- - -The single-line macro `__NASM_VERSION_ID__' expands to a dword integer -representing the full version number of the version of nasm being used. -The value is the equivalent to `__NASM_MAJOR__', `__NASM_MINOR__', -`__NASM_SUBMINOR__' and `___NASM_PATCHLEVEL__' concatenated to produce -a single doubleword. Hence, for 0.98.32p1, the returned number would be -equivalent to: - - dd 0x00622001 - - or - - db 1,32,98,0 - - Note that the above lines are generate exactly the same code, the -second line is used just to give an indication of the order that the -separate values will be present in memory. - - -File: nasm.info, Node: Section 4.11.3, Next: Section 4.11.4, Prev: Section 4.11.2, Up: Section 4.11 - -4.11.3. `__NASM_VER__': NASM Version string -------------------------------------------- - -The single-line macro `__NASM_VER__' expands to a string which defines -the version number of nasm being used. So, under NASM 0.98.32 for -example, - - db __NASM_VER__ - - would expand to - - db "0.98.32" - - -File: nasm.info, Node: Section 4.11.4, Next: Section 4.11.5, Prev: Section 4.11.3, Up: Section 4.11 - -4.11.4. `__FILE__' and `__LINE__': File Name and Line Number ------------------------------------------------------------- - -Like the C preprocessor, NASM allows the user to find out the file name -and line number containing the current instruction. The macro `__FILE__' -expands to a string constant giving the name of the current input file -(which may change through the course of assembly if `%include' -directives are used), and `__LINE__' expands to a numeric constant -giving the current line number in the input file. - - These macros could be used, for example, to communicate debugging -information to a macro, since invoking `__LINE__' inside a macro -definition (either single-line or multi-line) will return the line -number of the macro _call_, rather than _definition_. So to determine -where in a piece of code a crash is occurring, for example, one could -write a routine `stillhere', which is passed a line number in `EAX' and -outputs something like `line 155: still here'. You could then write a -macro - - %macro notdeadyet 0 - - push eax - mov eax,__LINE__ - call stillhere - pop eax - - %endmacro - - and then pepper your code with calls to `notdeadyet' until you find -the crash point. - - -File: nasm.info, Node: Section 4.11.5, Next: Section 4.11.6, Prev: Section 4.11.4, Up: Section 4.11 - -4.11.5. `__BITS__': Current BITS Mode -------------------------------------- - -The `__BITS__' standard macro is updated every time that the BITS mode -is set using the `BITS XX' or `[BITS XX]' directive, where XX is a -valid mode number of 16, 32 or 64. `__BITS__' receives the specified -mode number and makes it globally available. This can be very useful for -those who utilize mode-dependent macros. - - -File: nasm.info, Node: Section 4.11.6, Next: Section 4.11.7, Prev: Section 4.11.5, Up: Section 4.11 - -4.11.6. `__OUTPUT_FORMAT__': Current Output Format --------------------------------------------------- - -The `__OUTPUT_FORMAT__' standard macro holds the current Output Format, -as given by the `-f' option or NASM's default. Type `nasm -hf' for a -list. - - %ifidn __OUTPUT_FORMAT__, win32 - %define NEWLINE 13, 10 - %elifidn __OUTPUT_FORMAT__, elf32 - %define NEWLINE 10 - %endif - - -File: nasm.info, Node: Section 4.11.7, Next: Section 4.11.8, Prev: Section 4.11.6, Up: Section 4.11 - -4.11.7. Assembly Date and Time Macros -------------------------------------- - -NASM provides a variety of macros that represent the timestamp of the -assembly session. - - * The `__DATE__' and `__TIME__' macros give the assembly date and - time as strings, in ISO 8601 format (`"YYYY-MM-DD"' and - `"HH:MM:SS"', respectively.) - - * The `__DATE_NUM__' and `__TIME_NUM__' macros give the assembly - date and time in numeric form; in the format `YYYYMMDD' and - `HHMMSS' respectively. - - * The `__UTC_DATE__' and `__UTC_TIME__' macros give the assembly - date and time in universal time (UTC) as strings, in ISO 8601 - format (`"YYYY-MM-DD"' and `"HH:MM:SS"', respectively.) If the host - platform doesn't provide UTC time, these macros are undefined. - - * The `__UTC_DATE_NUM__' and `__UTC_TIME_NUM__' macros give the - assembly date and time universal time (UTC) in numeric form; in - the format `YYYYMMDD' and `HHMMSS' respectively. If the host - platform doesn't provide UTC time, these macros are undefined. - - * The `__POSIX_TIME__' macro is defined as a number containing the - number of seconds since the POSIX epoch, 1 January 1970 00:00:00 - UTC; excluding any leap seconds. This is computed using UTC time - if available on the host platform, otherwise it is computed using - the local time as if it was UTC. - - All instances of time and date macros in the same assembly session -produce consistent output. For example, in an assembly session started -at 42 seconds after midnight on January 1, 2010 in Moscow (timezone -UTC+3) these macros would have the following values, assuming, of -course, a properly configured environment with a correct clock: - - __DATE__ "2010-01-01" - __TIME__ "00:00:42" - __DATE_NUM__ 20100101 - __TIME_NUM__ 000042 - __UTC_DATE__ "2009-12-31" - __UTC_TIME__ "21:00:42" - __UTC_DATE_NUM__ 20091231 - __UTC_TIME_NUM__ 210042 - __POSIX_TIME__ 1262293242 - - -File: nasm.info, Node: Section 4.11.8, Next: Section 4.11.9, Prev: Section 4.11.7, Up: Section 4.11 - -4.11.8. `__USE_'_package_`__': Package Include Test ---------------------------------------------------- - -When a standard macro package (see *note Chapter 5::) is included with -the `%use' directive (see *note Section 4.6.4::), a single-line macro -of the form `__USE_'_package_`__' is automatically defined. This allows -testing if a particular package is invoked or not. - - For example, if the `altreg' package is included (see *note Section -5.1::), then the macro `__USE_ALTREG__' is defined. - - -File: nasm.info, Node: Section 4.11.9, Next: Section 4.11.10, Prev: Section 4.11.8, Up: Section 4.11 - -4.11.9. `__PASS__': Assembly Pass ---------------------------------- - -The macro `__PASS__' is defined to be `1' on preparatory passes, and -`2' on the final pass. In preprocess-only mode, it is set to `3', and -when running only to generate dependencies (due to the `-M' or `-MG' -option, see *note Section 2.1.4::) it is set to `0'. - - _Avoid using this macro if at all possible. It is tremendously easy -to generate very strange errors by misusing it, and the semantics may -change in future versions of NASM._ - - -File: nasm.info, Node: Section 4.11.10, Next: Section 4.11.11, Prev: Section 4.11.9, Up: Section 4.11 - -4.11.10. `STRUC' and `ENDSTRUC': Declaring Structure Data Types ---------------------------------------------------------------- - -The core of NASM contains no intrinsic means of defining data -structures; instead, the preprocessor is sufficiently powerful that -data structures can be implemented as a set of macros. The macros -`STRUC' and `ENDSTRUC' are used to define a structure data type. - - `STRUC' takes one or two parameters. The first parameter is the name -of the data type. The second, optional parameter is the base offset of -the structure. The name of the data type is defined as a symbol with -the value of the base offset, and the name of the data type with the -suffix `_size' appended to it is defined as an `EQU' giving the size of -the structure. Once `STRUC' has been issued, you are defining the -structure, and should define fields using the `RESB' family of pseudo- -instructions, and then invoke `ENDSTRUC' to finish the definition. - - For example, to define a structure called `mytype' containing a -longword, a word, a byte and a string of bytes, you might code - - struc mytype - - mt_long: resd 1 - mt_word: resw 1 - mt_byte: resb 1 - mt_str: resb 32 - - endstruc - - The above code defines six symbols: `mt_long' as 0 (the offset from -the beginning of a `mytype' structure to the longword field), `mt_word' -as 4, `mt_byte' as 6, `mt_str' as 7, `mytype_size' as 39, and `mytype' -itself as zero. - - The reason why the structure type name is defined at zero by default -is a side effect of allowing structures to work with the local label -mechanism: if your structure members tend to have the same names in -more than one structure, you can define the above structure like this: - - struc mytype - - .long: resd 1 - .word: resw 1 - .byte: resb 1 - .str: resb 32 - - endstruc - - This defines the offsets to the structure fields as `mytype.long', -`mytype.word', `mytype.byte' and `mytype.str'. - - NASM, since it has no _intrinsic_ structure support, does not support -any form of period notation to refer to the elements of a structure once -you have one (except the above local-label notation), so code such as -`mov ax,[mystruc.mt_word]' is not valid. `mt_word' is a constant just -like any other constant, so the correct syntax is `mov -ax,[mystruc+mt_word]' or `mov ax,[mystruc+mytype.word]'. - - Sometimes you only have the address of the structure displaced by an -offset. For example, consider this standard stack frame setup: - - push ebp - mov ebp, esp - sub esp, 40 - - In this case, you could access an element by subtracting the offset: - - mov [ebp - 40 + mytype.word], ax - - However, if you do not want to repeat this offset, you can use -40 -as a base offset: - - struc mytype, -40 - - And access an element this way: - - mov [ebp + mytype.word], ax - - -File: nasm.info, Node: Section 4.11.11, Next: Section 4.11.12, Prev: Section 4.11.10, Up: Section 4.11 - -4.11.11. `ISTRUC', `AT' and `IEND': Declaring Instances of Structures ---------------------------------------------------------------------- - -Having defined a structure type, the next thing you typically want to -do is to declare instances of that structure in your data segment. NASM -provides an easy way to do this in the `ISTRUC' mechanism. To declare a -structure of type `mytype' in a program, you code something like this: - - mystruc: - istruc mytype - - at mt_long, dd 123456 - at mt_word, dw 1024 - at mt_byte, db 'x' - at mt_str, db 'hello, world', 13, 10, 0 - - iend - - The function of the `AT' macro is to make use of the `TIMES' prefix -to advance the assembly position to the correct point for the specified -structure field, and then to declare the specified data. Therefore the -structure fields must be declared in the same order as they were -specified in the structure definition. - - If the data to go in a structure field requires more than one source -line to specify, the remaining source lines can easily come after the -`AT' line. For example: - - at mt_str, db 123,134,145,156,167,178,189 - db 190,100,0 - - Depending on personal taste, you can also omit the code part of the -`AT' line completely, and start the structure field on the next line: - - at mt_str - db 'hello, world' - db 13,10,0 - - -File: nasm.info, Node: Section 4.11.12, Next: Chapter 5, Prev: Section 4.11.11, Up: Section 4.11 - -4.11.12. `ALIGN' and `ALIGNB': Data Alignment ---------------------------------------------- - -The `ALIGN' and `ALIGNB' macros provides a convenient way to align code -or data on a word, longword, paragraph or other boundary. (Some -assemblers call this directive `EVEN'.) The syntax of the `ALIGN' and -`ALIGNB' macros is - - align 4 ; align on 4-byte boundary - align 16 ; align on 16-byte boundary - align 8,db 0 ; pad with 0s rather than NOPs - align 4,resb 1 ; align to 4 in the BSS - alignb 4 ; equivalent to previous line - - Both macros require their first argument to be a power of two; they -both compute the number of additional bytes required to bring the -length of the current section up to a multiple of that power of two, -and then apply the `TIMES' prefix to their second argument to perform -the alignment. - - If the second argument is not specified, the default for `ALIGN' is -`NOP', and the default for `ALIGNB' is `RESB 1'. So if the second -argument is specified, the two macros are equivalent. Normally, you can -just use `ALIGN' in code and data sections and `ALIGNB' in BSS -sections, and never need the second argument except for special -purposes. - - `ALIGN' and `ALIGNB', being simple macros, perform no error -checking: they cannot warn you if their first argument fails to be a -power of two, or if their second argument generates more than one byte -of code. In each of these cases they will silently do the wrong thing. - - `ALIGNB' (or `ALIGN' with a second argument of `RESB 1') can be used -within structure definitions: - - struc mytype2 - - mt_byte: - resb 1 - alignb 2 - mt_word: - resw 1 - alignb 4 - mt_long: - resd 1 - mt_str: - resb 32 - - endstruc - - This will ensure that the structure members are sensibly aligned -relative to the base of the structure. - - A final caveat: `ALIGN' and `ALIGNB' work relative to the beginning -of the _section_, not the beginning of the address space in the final -executable. Aligning to a 16-byte boundary when the section you're in -is only guaranteed to be aligned to a 4-byte boundary, for example, is -a waste of effort. Again, NASM does not check that the section's -alignment characteristics are sensible for the use of `ALIGN' or -`ALIGNB'. - - See also the `smartalign' standard macro package, *note Section -5.2::. - - -File: nasm.info, Node: Chapter 5, Next: Section 5.1, Prev: Section 4.11.12, Up: Top - -Chapter 5: Standard Macro Packages -********************************** - -The `%use' directive (see *note Section 4.6.4::) includes one of the -standard macro packages included with the NASM distribution and compiled -into the NASM binary. It operates like the `%include' directive (see -*note Section 4.6.1::), but the included contents is provided by NASM -itself. - - The names of standard macro packages are case insensitive, and can be -quoted or not. - -* Menu: - -* Section 5.1:: `altreg': Alternate Register Names -* Section 5.2:: `smartalign': Smart `ALIGN' Macro - - -File: nasm.info, Node: Section 5.1, Next: Section 5.2, Prev: Chapter 5, Up: Chapter 5 - -5.1. `altreg': Alternate Register Names -======================================= - -The `altreg' standard macro package provides alternate register names. -It provides numeric register names for all registers (not just `R8'- -`R15'), the Intel-defined aliases `R8L'-`R15L' for the low bytes of -register (as opposed to the NASM/AMD standard names `R8B'- `R15B'), and -the names `R0H'-`R3H' (by analogy with `R0L'-`R3L') for `AH', `CH', -`DH', and `BH'. - - Example use: - - %use altreg - - proc: - mov r0l,r3h ; mov al,bh - ret - - See also *note Section 11.1::. - - -File: nasm.info, Node: Section 5.2, Next: Chapter 6, Prev: Section 5.1, Up: Chapter 5 - -5.2. `smartalign': Smart `ALIGN' Macro -====================================== - -The `smartalign' standard macro package provides for an `ALIGN' macro -which is more powerful than the default (and backwards-compatible) one -(see *note Section 4.11.12::). When the `smartalign' package is -enabled, when `ALIGN' is used without a second argument, NASM will -generate a sequence of instructions more efficient than a series of -`NOP'. Furthermore, if the padding exceeds a specific threshold, then -NASM will generate a jump over the entire padding sequence. - - The specific instructions generated can be controlled with the new -`ALIGNMODE' macro. This macro takes two parameters: one mode, and an -optional jump threshold override. The modes are as follows: - - * `generic': Works on all x86 CPUs and should have reasonable - performance. The default jump threshold is 8. This is the default. - - * `nop': Pad out with `NOP' instructions. The only difference - compared to the standard `ALIGN' macro is that NASM can still jump - over a large padding area. The default jump threshold is 16. - - * `k7': Optimize for the AMD K7 (Athlon/Althon XP). These - instructions should still work on all x86 CPUs. The default jump - threshold is 16. - - * `k8': Optimize for the AMD K8 (Opteron/Althon 64). These - instructions should still work on all x86 CPUs. The default jump - threshold is 16. - - * `p6': Optimize for Intel CPUs. This uses the long `NOP' - instructions first introduced in Pentium Pro. This is incompatible - with all CPUs of family 5 or lower, as well as some VIA CPUs and - several virtualization solutions. The default jump threshold is 16. - - The macro `__ALIGNMODE__' is defined to contain the current alignment -mode. A number of other macros beginning with `__ALIGN_' are used -internally by this macro package. - - -File: nasm.info, Node: Chapter 6, Next: Section 6.1, Prev: Section 5.2, Up: Top - -Chapter 6: Assembler Directives -******************************* - -NASM, though it attempts to avoid the bureaucracy of assemblers like -MASM and TASM, is nevertheless forced to support a _few_ directives. -These are described in this chapter. - - NASM's directives come in two types: _user-level_ directives and -_primitive_ directives. Typically, each directive has a user-level form -and a primitive form. In almost all cases, we recommend that users use -the user-level forms of the directives, which are implemented as macros -which call the primitive forms. - - Primitive directives are enclosed in square brackets; user-level -directives are not. - - In addition to the universal directives described in this chapter, -each object file format can optionally supply extra directives in order -to control particular features of that file format. These -_format-specific_ directives are documented along with the formats that -implement them, in *note Chapter 7::. - -* Menu: - -* Section 6.1:: `BITS': Specifying Target Processor Mode -* Section 6.2:: `DEFAULT': Change the assembler defaults -* Section 6.3:: `SECTION' or `SEGMENT': Changing and Defining Sections -* Section 6.4:: `ABSOLUTE': Defining Absolute Labels -* Section 6.5:: `EXTERN': Importing Symbols from Other Modules -* Section 6.6:: `GLOBAL': Exporting Symbols to Other Modules -* Section 6.7:: `COMMON': Defining Common Data Areas -* Section 6.8:: `CPU': Defining CPU Dependencies -* Section 6.9:: `FLOAT': Handling of floating-point constants - - -File: nasm.info, Node: Section 6.1, Next: Section 6.1.1, Prev: Chapter 6, Up: Chapter 6 - -6.1. `BITS': Specifying Target Processor Mode -============================================= - -The `BITS' directive specifies whether NASM should generate code -designed to run on a processor operating in 16-bit mode, 32-bit mode or -64- bit mode. The syntax is `BITS XX', where XX is 16, 32 or 64. - - In most cases, you should not need to use `BITS' explicitly. The -`aout', `coff', `elf', `macho', `win32' and `win64' object formats, -which are designed for use in 32-bit or 64-bit operating systems, all -cause NASM to select 32-bit or 64-bit mode, respectively, by default. -The `obj' object format allows you to specify each segment you define -as either `USE16' or `USE32', and NASM will set its operating mode -accordingly, so the use of the `BITS' directive is once again -unnecessary. - - The most likely reason for using the `BITS' directive is to write 32- -bit or 64-bit code in a flat binary file; this is because the `bin' -output format defaults to 16-bit mode in anticipation of it being used -most frequently to write DOS `.COM' programs, DOS `.SYS' device drivers -and boot loader software. - - You do _not_ need to specify `BITS 32' merely in order to use 32- -bit instructions in a 16-bit DOS program; if you do, the assembler will -generate incorrect code because it will be writing code targeted at a -32- bit platform, to be run on a 16-bit one. - - When NASM is in `BITS 16' mode, instructions which use 32-bit data -are prefixed with an 0x66 byte, and those referring to 32-bit addresses -have an 0x67 prefix. In `BITS 32' mode, the reverse is true: 32-bit -instructions require no prefixes, whereas instructions using 16-bit data -need an 0x66 and those working on 16-bit addresses need an 0x67. - - When NASM is in `BITS 64' mode, most instructions operate the same as -they do for `BITS 32' mode. However, there are 8 more general and SSE -registers, and 16-bit addressing is no longer supported. - - The default address size is 64 bits; 32-bit addressing can be -selected with the 0x67 prefix. The default operand size is still 32 -bits, however, and the 0x66 prefix selects 16-bit operand size. The -`REX' prefix is used both to select 64-bit operand size, and to access -the new registers. NASM automatically inserts REX prefixes when -necessary. - - When the `REX' prefix is used, the processor does not know how to -address the AH, BH, CH or DH (high 8-bit legacy) registers. Instead, it -is possible to access the the low 8-bits of the SP, BP SI and DI -registers as SPL, BPL, SIL and DIL, respectively; but only when the REX -prefix is used. - - The `BITS' directive has an exactly equivalent primitive form, -`[BITS 16]', `[BITS 32]' and `[BITS 64]'. The user-level form is a -macro which has no function other than to call the primitive form. - - Note that the space is neccessary, e.g. `BITS32' will _not_ work! - -* Menu: - -* Section 6.1.1:: `USE16' & `USE32': Aliases for BITS - - -File: nasm.info, Node: Section 6.1.1, Next: Section 6.2, Prev: Section 6.1, Up: Section 6.1 - -6.1.1. `USE16' & `USE32': Aliases for BITS ------------------------------------------- - -The ``USE16'' and ``USE32'' directives can be used in place of ``BITS -16'' and ``BITS 32'', for compatibility with other assemblers. - - -File: nasm.info, Node: Section 6.2, Next: Section 6.3, Prev: Section 6.1.1, Up: Chapter 6 - -6.2. `DEFAULT': Change the assembler defaults -============================================= - -The `DEFAULT' directive changes the assembler defaults. Normally, NASM -defaults to a mode where the programmer is expected to explicitly -specify most features directly. However, this is occationally -obnoxious, as the explicit form is pretty much the only one one wishes -to use. - - Currently, the only `DEFAULT' that is settable is whether or not -registerless instructions in 64-bit mode are `RIP'-relative or not. By -default, they are absolute unless overridden with the `REL' specifier -(see *note Section 3.3::). However, if `DEFAULT REL' is specified, -`REL' is default, unless overridden with the `ABS' specifier, _except -when used with an FS or GS segment override_. - - The special handling of `FS' and `GS' overrides are due to the fact -that these registers are generally used as thread pointers or other -special functions in 64-bit mode, and generating `RIP'-relative -addresses would be extremely confusing. - - `DEFAULT REL' is disabled with `DEFAULT ABS'. - - -File: nasm.info, Node: Section 6.3, Next: Section 6.3.1, Prev: Section 6.2, Up: Chapter 6 - -6.3. `SECTION' or `SEGMENT': Changing and Defining Sections -=========================================================== - -The `SECTION' directive (`SEGMENT' is an exactly equivalent synonym) -changes which section of the output file the code you write will be -assembled into. In some object file formats, the number and names of -sections are fixed; in others, the user may make up as many as they -wish. Hence `SECTION' may sometimes give an error message, or may -define a new section, if you try to switch to a section that does not -(yet) exist. - - The Unix object formats, and the `bin' object format (but see *note -Section 7.1.3::, all support the standardized section names `.text', -`.data' and `.bss' for the code, data and uninitialized-data sections. -The `obj' format, by contrast, does not recognize these section names -as being special, and indeed will strip off the leading period of any -section name that has one. - -* Menu: - -* Section 6.3.1:: The `__SECT__' Macro - - -File: nasm.info, Node: Section 6.3.1, Next: Section 6.4, Prev: Section 6.3, Up: Section 6.3 - -6.3.1. The `__SECT__' Macro ---------------------------- - -The `SECTION' directive is unusual in that its user-level form -functions differently from its primitive form. The primitive form, -`[SECTION xyz]', simply switches the current target section to the one -given. The user-level form, `SECTION xyz', however, first defines the -single-line macro `__SECT__' to be the primitive `[SECTION]' directive -which it is about to issue, and then issues it. So the user-level -directive - - SECTION .text - - expands to the two lines - - %define __SECT__ [SECTION .text] - [SECTION .text] - - Users may find it useful to make use of this in their own macros. For -example, the `writefile' macro defined in *note Section 4.3.4:: can be -usefully rewritten in the following more sophisticated form: - - %macro writefile 2+ - - [section .data] - - %%str: db %2 - %%endstr: - - __SECT__ - - mov dx,%%str - mov cx,%%endstr-%%str - mov bx,%1 - mov ah,0x40 - int 0x21 - - %endmacro - - This form of the macro, once passed a string to output, first -switches temporarily to the data section of the file, using the -primitive form of the `SECTION' directive so as not to modify -`__SECT__'. It then declares its string in the data section, and then -invokes `__SECT__' to switch back to _whichever_ section the user was -previously working in. It thus avoids the need, in the previous version -of the macro, to include a `JMP' instruction to jump over the data, and -also does not fail if, in a complicated `OBJ' format module, the user -could potentially be assembling the code in any of several separate code -sections. - - -File: nasm.info, Node: Section 6.4, Next: Section 6.5, Prev: Section 6.3.1, Up: Chapter 6 - -6.4. `ABSOLUTE': Defining Absolute Labels -========================================= - -The `ABSOLUTE' directive can be thought of as an alternative form of -`SECTION': it causes the subsequent code to be directed at no physical -section, but at the hypothetical section starting at the given absolute -address. The only instructions you can use in this mode are the `RESB' -family. - - `ABSOLUTE' is used as follows: - - absolute 0x1A - - kbuf_chr resw 1 - kbuf_free resw 1 - kbuf resw 16 - - This example describes a section of the PC BIOS data area, at segment -address 0x40: the above code defines `kbuf_chr' to be 0x1A, `kbuf_free' -to be 0x1C, and `kbuf' to be 0x1E. - - The user-level form of `ABSOLUTE', like that of `SECTION', redefines -the `__SECT__' macro when it is invoked. - - `STRUC' and `ENDSTRUC' are defined as macros which use `ABSOLUTE' -(and also `__SECT__'). - - `ABSOLUTE' doesn't have to take an absolute constant as an argument: -it can take an expression (actually, a critical expression: see *note -Section 3.8::) and it can be a value in a segment. For example, a TSR -can re-use its setup code as run-time BSS like this: - - org 100h ; it's a .COM program - - jmp setup ; setup code comes last - - ; the resident part of the TSR goes here - setup: - ; now write the code that installs the TSR here - - absolute setup - - runtimevar1 resw 1 - runtimevar2 resd 20 - - tsr_end: - - This defines some variables `on top of' the setup code, so that -after the setup has finished running, the space it took up can be -re-used as data storage for the running TSR. The symbol `tsr_end' can -be used to calculate the total size of the part of the TSR that needs -to be made resident. - - -File: nasm.info, Node: Section 6.5, Next: Section 6.6, Prev: Section 6.4, Up: Chapter 6 - -6.5. `EXTERN': Importing Symbols from Other Modules -=================================================== - -`EXTERN' is similar to the MASM directive `EXTRN' and the C keyword -`extern': it is used to declare a symbol which is not defined anywhere -in the module being assembled, but is assumed to be defined in some -other module and needs to be referred to by this one. Not every -object-file format can support external variables: the `bin' format -cannot. - - The `EXTERN' directive takes as many arguments as you like. Each -argument is the name of a symbol: - - extern _printf - extern _sscanf,_fscanf - - Some object-file formats provide extra features to the `EXTERN' -directive. In all cases, the extra features are used by suffixing a -colon to the symbol name followed by object-format specific text. For -example, the `obj' format allows you to declare that the default -segment base of an external should be the group `dgroup' by means of -the directive - - extern _variable:wrt dgroup - - The primitive form of `EXTERN' differs from the user-level form only -in that it can take only one argument at a time: the support for -multiple arguments is implemented at the preprocessor level. - - You can declare the same variable as `EXTERN' more than once: NASM -will quietly ignore the second and later redeclarations. You can't -declare a variable as `EXTERN' as well as something else, though. - - -File: nasm.info, Node: Section 6.6, Next: Section 6.7, Prev: Section 6.5, Up: Chapter 6 - -6.6. `GLOBAL': Exporting Symbols to Other Modules -================================================= - -`GLOBAL' is the other end of `EXTERN': if one module declares a symbol -as `EXTERN' and refers to it, then in order to prevent linker errors, -some other module must actually _define_ the symbol and declare it as -`GLOBAL'. Some assemblers use the name `PUBLIC' for this purpose. - - The `GLOBAL' directive applying to a symbol must appear _before_ the -definition of the symbol. - - `GLOBAL' uses the same syntax as `EXTERN', except that it must refer -to symbols which _are_ defined in the same module as the `GLOBAL' -directive. For example: - - global _main - _main: - ; some code - - `GLOBAL', like `EXTERN', allows object formats to define private -extensions by means of a colon. The `elf' object format, for example, -lets you specify whether global data items are functions or data: - - global hashlookup:function, hashtable:data - - Like `EXTERN', the primitive form of `GLOBAL' differs from the -user-level form only in that it can take only one argument at a time. - - -File: nasm.info, Node: Section 6.7, Next: Section 6.8, Prev: Section 6.6, Up: Chapter 6 - -6.7. `COMMON': Defining Common Data Areas -========================================= - -The `COMMON' directive is used to declare _common variables_. A common -variable is much like a global variable declared in the uninitialized -data section, so that - - common intvar 4 - - is similar in function to - - global intvar - section .bss - - intvar resd 1 - - The difference is that if more than one module defines the same -common variable, then at link time those variables will be _merged_, and -references to `intvar' in all modules will point at the same piece of -memory. - - Like `GLOBAL' and `EXTERN', `COMMON' supports object-format specific -extensions. For example, the `obj' format allows common variables to be -NEAR or FAR, and the `elf' format allows you to specify the alignment -requirements of a common variable: - - common commvar 4:near ; works in OBJ - common intarray 100:4 ; works in ELF: 4 byte aligned - - Once again, like `EXTERN' and `GLOBAL', the primitive form of -`COMMON' differs from the user-level form only in that it can take only -one argument at a time. - - -File: nasm.info, Node: Section 6.8, Next: Section 6.9, Prev: Section 6.7, Up: Chapter 6 - -6.8. `CPU': Defining CPU Dependencies -===================================== - -The `CPU' directive restricts assembly to those instructions which are -available on the specified CPU. - - Options are: - - * `CPU 8086' Assemble only 8086 instruction set - - * `CPU 186' Assemble instructions up to the 80186 instruction set - - * `CPU 286' Assemble instructions up to the 286 instruction set - - * `CPU 386' Assemble instructions up to the 386 instruction set - - * `CPU 486' 486 instruction set - - * `CPU 586' Pentium instruction set - - * `CPU PENTIUM' Same as 586 - - * `CPU 686' P6 instruction set - - * `CPU PPRO' Same as 686 - - * `CPU P2' Same as 686 - - * `CPU P3' Pentium III (Katmai) instruction sets - - * `CPU KATMAI' Same as P3 - - * `CPU P4' Pentium 4 (Willamette) instruction set - - * `CPU WILLAMETTE' Same as P4 - - * `CPU PRESCOTT' Prescott instruction set - - * `CPU X64' x86-64 (x64/AMD64/Intel 64) instruction set - - * `CPU IA64' IA64 CPU (in x86 mode) instruction set - - All options are case insensitive. All instructions will be selected -only if they apply to the selected CPU or lower. By default, all -instructions are available. - - -File: nasm.info, Node: Section 6.9, Next: Chapter 7, Prev: Section 6.8, Up: Chapter 6 - -6.9. `FLOAT': Handling of floating-point constants -================================================== - -By default, floating-point constants are rounded to nearest, and IEEE -denormals are supported. The following options can be set to alter this -behaviour: - - * `FLOAT DAZ' Flush denormals to zero - - * `FLOAT NODAZ' Do not flush denormals to zero (default) - - * `FLOAT NEAR' Round to nearest (default) - - * `FLOAT UP' Round up (toward +Infinity) - - * `FLOAT DOWN' Round down (toward -Infinity) - - * `FLOAT ZERO' Round toward zero - - * `FLOAT DEFAULT' Restore default settings - - The standard macros `__FLOAT_DAZ__', `__FLOAT_ROUND__', and -`__FLOAT__' contain the current state, as long as the programmer has -avoided the use of the brackeded primitive form, (`[FLOAT]'). - - `__FLOAT__' contains the full set of floating-point settings; this -value can be saved away and invoked later to restore the setting. - - -File: nasm.info, Node: Chapter 7, Next: Section 7.1, Prev: Section 6.9, Up: Top - -Chapter 7: Output Formats -************************* - -NASM is a portable assembler, designed to be able to compile on any -ANSI C- supporting platform and produce output to run on a variety of -Intel x86 operating systems. For this reason, it has a large number of -available output formats, selected using the `-f' option on the NASM -command line. Each of these formats, along with its extensions to the -base NASM syntax, is detailed in this chapter. - - As stated in *note Section 2.1.1::, NASM chooses a default name for -your output file based on the input file name and the chosen output -format. This will be generated by removing the extension (`.asm', `.s', -or whatever you like to use) from the input file name, and substituting -an extension defined by the output format. The extensions are given -with each format below. - -* Menu: - -* Section 7.1:: `bin': Flat-Form Binary Output -* Section 7.2:: `ith': Intel Hex Output -* Section 7.3:: `srec': Motorola S-Records Output -* Section 7.4:: `obj': Microsoft OMF Object Files -* Section 7.5:: `win32': Microsoft Win32 Object Files -* Section 7.6:: `win64': Microsoft Win64 Object Files -* Section 7.7:: `coff': Common Object File Format -* Section 7.8:: `macho32' and `macho64': Mach Object File Format -* Section 7.9:: `elf32' and `elf64': Executable and Linkable Format Object Files -* Section 7.10:: `aout': Linux `a.out' Object Files -* Section 7.11:: `aoutb': NetBSD/FreeBSD/OpenBSD `a.out' Object Files -* Section 7.12:: `as86': Minix/Linux `as86' Object Files -* Section 7.13:: `rdf': Relocatable Dynamic Object File Format -* Section 7.14:: `dbg': Debugging Format - - -File: nasm.info, Node: Section 7.1, Next: Section 7.1.1, Prev: Chapter 7, Up: Chapter 7 - -7.1. `bin': Flat-Form Binary Output -=================================== - -The `bin' format does not produce object files: it generates nothing in -the output file except the code you wrote. Such `pure binary' files are -used by MS-DOS: `.COM' executables and `.SYS' device drivers are pure -binary files. Pure binary output is also useful for operating system -and boot loader development. - - The `bin' format supports multiple section names. For details of how -NASM handles sections in the `bin' format, see *note Section 7.1.3::. - - Using the `bin' format puts NASM by default into 16-bit mode (see -*note Section 6.1::). In order to use `bin' to write 32-bit or 64-bit -code, such as an OS kernel, you need to explicitly issue the `BITS 32' -or `BITS 64' directive. - - `bin' has no default output file name extension: instead, it leaves -your file name as it is once the original extension has been removed. -Thus, the default is for NASM to assemble `binprog.asm' into a binary -file called `binprog'. - -* Menu: - -* Section 7.1.1:: `ORG': Binary File Program Origin -* Section 7.1.2:: `bin' Extensions to the `SECTION' Directive -* Section 7.1.3:: Multisection Support for the `bin' Format -* Section 7.1.4:: Map Files - - -File: nasm.info, Node: Section 7.1.1, Next: Section 7.1.2, Prev: Section 7.1, Up: Section 7.1 - -7.1.1. `ORG': Binary File Program Origin ----------------------------------------- - -The `bin' format provides an additional directive to the list given in -*note Chapter 6::: `ORG'. The function of the `ORG' directive is to -specify the origin address which NASM will assume the program begins at -when it is loaded into memory. - - For example, the following code will generate the longword -`0x00000104': - - org 0x100 - dd label - label: - - Unlike the `ORG' directive provided by MASM-compatible assemblers, -which allows you to jump around in the object file and overwrite code -you have already generated, NASM's `ORG' does exactly what the directive -says: _origin_. Its sole function is to specify one offset which is -added to all internal address references within the section; it does not -permit any of the trickery that MASM's version does. See *note Section -12.1.3:: for further comments. - - -File: nasm.info, Node: Section 7.1.2, Next: Section 7.1.3, Prev: Section 7.1.1, Up: Section 7.1 - -7.1.2. `bin' Extensions to the `SECTION' Directive --------------------------------------------------- - -The `bin' output format extends the `SECTION' (or `SEGMENT') directive -to allow you to specify the alignment requirements of segments. This -is done by appending the `ALIGN' qualifier to the end of the -section-definition line. For example, - - section .data align=16 - - switches to the section `.data' and also specifies that it must be -aligned on a 16-byte boundary. - - The parameter to `ALIGN' specifies how many low bits of the section -start address must be forced to zero. The alignment value given may be -any power of two. - - -File: nasm.info, Node: Section 7.1.3, Next: Section 7.1.4, Prev: Section 7.1.2, Up: Section 7.1 - -7.1.3. Multisection Support for the `bin' Format ------------------------------------------------- - -The `bin' format allows the use of multiple sections, of arbitrary -names, besides the "known" `.text', `.data', and `.bss' names. - - * Sections may be designated `progbits' or `nobits'. Default is - `progbits' (except `.bss', which defaults to `nobits', of course). - - * Sections can be aligned at a specified boundary following the - previous section with `align=', or at an arbitrary byte-granular - position with `start='. - - * Sections can be given a virtual start address, which will be used - for the calculation of all memory references within that section - with `vstart='. - - * Sections can be ordered using `follows='`<section>' or - `vfollows='`<section>' as an alternative to specifying an explicit - start address. - - * Arguments to `org', `start', `vstart', and `align=' are critical - expressions. See *note Section 3.8::. E.g. `align=(1 << - ALIGN_SHIFT)' - `ALIGN_SHIFT' must be defined before it is used - here. - - * Any code which comes before an explicit `SECTION' directive is - directed by default into the `.text' section. - - * If an `ORG' statement is not given, `ORG 0' is used by default. - - * The `.bss' section will be placed after the last `progbits' - section, unless `start=', `vstart=', `follows=', or `vfollows=' - has been specified. - - * All sections are aligned on dword boundaries, unless a different - alignment has been specified. - - * Sections may not overlap. - - * NASM creates the `section.<secname>.start' for each section, which - may be used in your code. - - -File: nasm.info, Node: Section 7.1.4, Next: Section 7.2, Prev: Section 7.1.3, Up: Section 7.1 - -7.1.4. Map Files ----------------- - -Map files can be generated in `-f bin' format by means of the `[map]' -option. Map types of `all' (default), `brief', `sections', `segments', -or `symbols' may be specified. Output may be directed to `stdout' -(default), `stderr', or a specified file. E.g. `[map symbols -myfile.map]'. No "user form" exists, the square brackets must be used. - - -File: nasm.info, Node: Section 7.2, Next: Section 7.3, Prev: Section 7.1.4, Up: Chapter 7 - -7.2. `ith': Intel Hex Output -============================ - -The `ith' file format produces Intel hex-format files. Just as the -`bin' format, this is a flat memory image format with no support for -relocation or linking. It is usually used with ROM programmers and -similar utilities. - - All extensions supported by the `bin' file format is also supported -by the `ith' file format. - - `ith' provides a default output file-name extension of `.ith'. - - -File: nasm.info, Node: Section 7.3, Next: Section 7.4, Prev: Section 7.2, Up: Chapter 7 - -7.3. `srec': Motorola S-Records Output -====================================== - -The `srec' file format produces Motorola S-records files. Just as the -`bin' format, this is a flat memory image format with no support for -relocation or linking. It is usually used with ROM programmers and -similar utilities. - - All extensions supported by the `bin' file format is also supported -by the `srec' file format. - - `srec' provides a default output file-name extension of `.srec'. - - -File: nasm.info, Node: Section 7.4, Next: Section 7.4.1, Prev: Section 7.3, Up: Chapter 7 - -7.4. `obj': Microsoft OMF Object Files -====================================== - -The `obj' file format (NASM calls it `obj' rather than `omf' for -historical reasons) is the one produced by MASM and TASM, which is -typically fed to 16-bit DOS linkers to produce `.EXE' files. It is also -the format used by OS/2. - - `obj' provides a default output file-name extension of `.obj'. - - `obj' is not exclusively a 16-bit format, though: NASM has full -support for the 32-bit extensions to the format. In particular, 32-bit -`obj' format files are used by Borland's Win32 compilers, instead of -using Microsoft's newer `win32' object file format. - - The `obj' format does not define any special segment names: you can -call your segments anything you like. Typical names for segments in -`obj' format files are `CODE', `DATA' and `BSS'. - - If your source file contains code before specifying an explicit -`SEGMENT' directive, then NASM will invent its own segment called -`__NASMDEFSEG' for you. - - When you define a segment in an `obj' file, NASM defines the segment -name as a symbol as well, so that you can access the segment address of -the segment. So, for example: - - segment data - - dvar: dw 1234 - - segment code - - function: - mov ax,data ; get segment address of data - mov ds,ax ; and move it into DS - inc word [dvar] ; now this reference will work - ret - - The `obj' format also enables the use of the `SEG' and `WRT' -operators, so that you can write code which does things like - - extern foo - - mov ax,seg foo ; get preferred segment of foo - mov ds,ax - mov ax,data ; a different segment - mov es,ax - mov ax,[ds:foo] ; this accesses `foo' - mov [es:foo wrt data],bx ; so does this - -* Menu: - -* Section 7.4.1:: `obj' Extensions to the `SEGMENT' Directive -* Section 7.4.2:: `GROUP': Defining Groups of Segments -* Section 7.4.3:: `UPPERCASE': Disabling Case Sensitivity in Output -* Section 7.4.4:: `IMPORT': Importing DLL Symbols -* Section 7.4.5:: `EXPORT': Exporting DLL Symbols -* Section 7.4.6:: `..start': Defining the Program Entry Point -* Section 7.4.7:: `obj' Extensions to the `EXTERN' Directive -* Section 7.4.8:: `obj' Extensions to the `COMMON' Directive - - -File: nasm.info, Node: Section 7.4.1, Next: Section 7.4.2, Prev: Section 7.4, Up: Section 7.4 - -7.4.1. `obj' Extensions to the `SEGMENT' Directive --------------------------------------------------- - -The `obj' output format extends the `SEGMENT' (or `SECTION') directive -to allow you to specify various properties of the segment you are -defining. This is done by appending extra qualifiers to the end of the -segment-definition line. For example, - - segment code private align=16 - - defines the segment `code', but also declares it to be a private -segment, and requires that the portion of it described in this code -module must be aligned on a 16-byte boundary. - - The available qualifiers are: - - * `PRIVATE', `PUBLIC', `COMMON' and `STACK' specify the combination - characteristics of the segment. `PRIVATE' segments do not get - combined with any others by the linker; `PUBLIC' and `STACK' - segments get concatenated together at link time; and `COMMON' - segments all get overlaid on top of each other rather than stuck - end-to-end. - - * `ALIGN' is used, as shown above, to specify how many low bits of - the segment start address must be forced to zero. The alignment - value given may be any power of two from 1 to 4096; in reality, - the only values supported are 1, 2, 4, 16, 256 and 4096, so if 8 - is specified it will be rounded up to 16, and 32, 64 and 128 will - all be rounded up to 256, and so on. Note that alignment to - 4096-byte boundaries is a PharLap extension to the format and may - not be supported by all linkers. - - * `CLASS' can be used to specify the segment class; this feature - indicates to the linker that segments of the same class should be - placed near each other in the output file. The class name can be - any word, e.g. `CLASS=CODE'. - - * `OVERLAY', like `CLASS', is specified with an arbitrary word as an - argument, and provides overlay information to an overlay-capable - linker. - - * Segments can be declared as `USE16' or `USE32', which has the - effect of recording the choice in the object file and also - ensuring that NASM's default assembly mode when assembling in that - segment is 16-bit or 32-bit respectively. - - * When writing OS/2 object files, you should declare 32-bit segments - as `FLAT', which causes the default segment base for anything in - the segment to be the special group `FLAT', and also defines the - group if it is not already defined. - - * The `obj' file format also allows segments to be declared as - having a pre-defined absolute segment address, although no linkers - are currently known to make sensible use of this feature; - nevertheless, NASM allows you to declare a segment such as - `SEGMENT SCREEN ABSOLUTE=0xB800' if you need to. The `ABSOLUTE' - and `ALIGN' keywords are mutually exclusive. - - NASM's default segment attributes are `PUBLIC', `ALIGN=1', no class, -no overlay, and `USE16'. - - -File: nasm.info, Node: Section 7.4.2, Next: Section 7.4.3, Prev: Section 7.4.1, Up: Section 7.4 - -7.4.2. `GROUP': Defining Groups of Segments -------------------------------------------- - -The `obj' format also allows segments to be grouped, so that a single -segment register can be used to refer to all the segments in a group. -NASM therefore supplies the `GROUP' directive, whereby you can code - - segment data - - ; some data - - segment bss - - ; some uninitialized data - - group dgroup data bss - - which will define a group called `dgroup' to contain the segments -`data' and `bss'. Like `SEGMENT', `GROUP' causes the group name to be -defined as a symbol, so that you can refer to a variable `var' in the -`data' segment as `var wrt data' or as `var wrt dgroup', depending on -which segment value is currently in your segment register. - - If you just refer to `var', however, and `var' is declared in a -segment which is part of a group, then NASM will default to giving you -the offset of `var' from the beginning of the _group_, not the -_segment_. Therefore `SEG var', also, will return the group base rather -than the segment base. - - NASM will allow a segment to be part of more than one group, but will -generate a warning if you do this. Variables declared in a segment -which is part of more than one group will default to being relative to -the first group that was defined to contain the segment. - - A group does not have to contain any segments; you can still make -`WRT' references to a group which does not contain the variable you are -referring to. OS/2, for example, defines the special group `FLAT' with -no segments in it. - - -File: nasm.info, Node: Section 7.4.3, Next: Section 7.4.4, Prev: Section 7.4.2, Up: Section 7.4 - -7.4.3. `UPPERCASE': Disabling Case Sensitivity in Output --------------------------------------------------------- - -Although NASM itself is case sensitive, some OMF linkers are not; -therefore it can be useful for NASM to output single-case object files. -The `UPPERCASE' format-specific directive causes all segment, group and -symbol names that are written to the object file to be forced to upper -case just before being written. Within a source file, NASM is still -case- sensitive; but the object file can be written entirely in upper -case if desired. - - `UPPERCASE' is used alone on a line; it requires no parameters. - - -File: nasm.info, Node: Section 7.4.4, Next: Section 7.4.5, Prev: Section 7.4.3, Up: Section 7.4 - -7.4.4. `IMPORT': Importing DLL Symbols --------------------------------------- - -The `IMPORT' format-specific directive defines a symbol to be imported -from a DLL, for use if you are writing a DLL's import library in NASM. -You still need to declare the symbol as `EXTERN' as well as using the -`IMPORT' directive. - - The `IMPORT' directive takes two required parameters, separated by -white space, which are (respectively) the name of the symbol you wish to -import and the name of the library you wish to import it from. For -example: - - import WSAStartup wsock32.dll - - A third optional parameter gives the name by which the symbol is -known in the library you are importing it from, in case this is not the -same as the name you wish the symbol to be known by to your code once -you have imported it. For example: - - import asyncsel wsock32.dll WSAAsyncSelect - - -File: nasm.info, Node: Section 7.4.5, Next: Section 7.4.6, Prev: Section 7.4.4, Up: Section 7.4 - -7.4.5. `EXPORT': Exporting DLL Symbols --------------------------------------- - -The `EXPORT' format-specific directive defines a global symbol to be -exported as a DLL symbol, for use if you are writing a DLL in NASM. You -still need to declare the symbol as `GLOBAL' as well as using the -`EXPORT' directive. - - `EXPORT' takes one required parameter, which is the name of the -symbol you wish to export, as it was defined in your source file. An -optional second parameter (separated by white space from the first) -gives the _external_ name of the symbol: the name by which you wish the -symbol to be known to programs using the DLL. If this name is the same -as the internal name, you may leave the second parameter off. - - Further parameters can be given to define attributes of the exported -symbol. These parameters, like the second, are separated by white -space. If further parameters are given, the external name must also be -specified, even if it is the same as the internal name. The available -attributes are: - - * `resident' indicates that the exported name is to be kept resident - by the system loader. This is an optimisation for frequently used - symbols imported by name. - - * `nodata' indicates that the exported symbol is a function which - does not make use of any initialized data. - - * `parm=NNN', where `NNN' is an integer, sets the number of - parameter words for the case in which the symbol is a call gate - between 32- bit and 16-bit segments. - - * An attribute which is just a number indicates that the symbol - should be exported with an identifying number (ordinal), and gives - the desired number. - - For example: - - export myfunc - export myfunc TheRealMoreFormalLookingFunctionName - export myfunc myfunc 1234 ; export by ordinal - export myfunc myfunc resident parm=23 nodata - - -File: nasm.info, Node: Section 7.4.6, Next: Section 7.4.7, Prev: Section 7.4.5, Up: Section 7.4 - -7.4.6. `..start': Defining the Program Entry Point --------------------------------------------------- - -`OMF' linkers require exactly one of the object files being linked to -define the program entry point, where execution will begin when the -program is run. If the object file that defines the entry point is -assembled using NASM, you specify the entry point by declaring the -special symbol `..start' at the point where you wish execution to begin. - - -File: nasm.info, Node: Section 7.4.7, Next: Section 7.4.8, Prev: Section 7.4.6, Up: Section 7.4 - -7.4.7. `obj' Extensions to the `EXTERN' Directive -------------------------------------------------- - -If you declare an external symbol with the directive - - extern foo - - then references such as `mov ax,foo' will give you the offset of -`foo' from its preferred segment base (as specified in whichever module -`foo' is actually defined in). So to access the contents of `foo' you -will usually need to do something like - - mov ax,seg foo ; get preferred segment base - mov es,ax ; move it into ES - mov ax,[es:foo] ; and use offset `foo' from it - - This is a little unwieldy, particularly if you know that an external -is going to be accessible from a given segment or group, say `dgroup'. -So if `DS' already contained `dgroup', you could simply code - - mov ax,[foo wrt dgroup] - - However, having to type this every time you want to access `foo' can -be a pain; so NASM allows you to declare `foo' in the alternative form - - extern foo:wrt dgroup - - This form causes NASM to pretend that the preferred segment base of -`foo' is in fact `dgroup'; so the expression `seg foo' will now return -`dgroup', and the expression `foo' is equivalent to `foo wrt dgroup'. - - This default-`WRT' mechanism can be used to make externals appear to -be relative to any group or segment in your program. It can also be -applied to common variables: see *note Section 7.4.8::. - - -File: nasm.info, Node: Section 7.4.8, Next: Section 7.5, Prev: Section 7.4.7, Up: Section 7.4 - -7.4.8. `obj' Extensions to the `COMMON' Directive -------------------------------------------------- - -The `obj' format allows common variables to be either near or far; NASM -allows you to specify which your variables should be by the use of the -syntax - - common nearvar 2:near ; `nearvar' is a near common - common farvar 10:far ; and `farvar' is far - - Far common variables may be greater in size than 64Kb, and so the OMF -specification says that they are declared as a number of _elements_ of -a given size. So a 10-byte far common variable could be declared as ten -one-byte elements, five two-byte elements, two five-byte elements or one -ten-byte element. - - Some `OMF' linkers require the element size, as well as the variable -size, to match when resolving common variables declared in more than one -module. Therefore NASM must allow you to specify the element size on -your far common variables. This is done by the following syntax: - - common c_5by2 10:far 5 ; two five-byte elements - common c_2by5 10:far 2 ; five two-byte elements - - If no element size is specified, the default is 1. Also, the `FAR' -keyword is not required when an element size is specified, since only -far commons may have element sizes at all. So the above declarations -could equivalently be - - common c_5by2 10:5 ; two five-byte elements - common c_2by5 10:2 ; five two-byte elements - - In addition to these extensions, the `COMMON' directive in `obj' -also supports default-`WRT' specification like `EXTERN' does (explained -in *note Section 7.4.7::). So you can also declare things like - - common foo 10:wrt dgroup - common bar 16:far 2:wrt data - common baz 24:wrt data:6 - - -File: nasm.info, Node: Section 7.5, Next: Section 7.5.1, Prev: Section 7.4.8, Up: Chapter 7 - -7.5. `win32': Microsoft Win32 Object Files -========================================== - -The `win32' output format generates Microsoft Win32 object files, -suitable for passing to Microsoft linkers such as Visual C++. Note that -Borland Win32 compilers do not use this format, but use `obj' instead -(see *note Section 7.4::). - - `win32' provides a default output file-name extension of `.obj'. - - Note that although Microsoft say that Win32 object files follow the -`COFF' (Common Object File Format) standard, the object files produced -by Microsoft Win32 compilers are not compatible with COFF linkers such -as DJGPP's, and vice versa. This is due to a difference of opinion over -the precise semantics of PC-relative relocations. To produce COFF files -suitable for DJGPP, use NASM's `coff' output format; conversely, the -`coff' format does not produce object files that Win32 linkers can -generate correct output from. - -* Menu: - -* Section 7.5.1:: `win32' Extensions to the `SECTION' Directive -* Section 7.5.2:: `win32': Safe Structured Exception Handling - - -File: nasm.info, Node: Section 7.5.1, Next: Section 7.5.2, Prev: Section 7.5, Up: Section 7.5 - -7.5.1. `win32' Extensions to the `SECTION' Directive ----------------------------------------------------- - -Like the `obj' format, `win32' allows you to specify additional -information on the `SECTION' directive line, to control the type and -properties of sections you declare. Section types and properties are -generated automatically by NASM for the standard section names `.text', -`.data' and `.bss', but may still be overridden by these qualifiers. - - The available qualifiers are: - - * `code', or equivalently `text', defines the section to be a code - section. This marks the section as readable and executable, but not - writable, and also indicates to the linker that the type of the - section is code. - - * `data' and `bss' define the section to be a data section, - analogously to `code'. Data sections are marked as readable and - writable, but not executable. `data' declares an initialized data - section, whereas `bss' declares an uninitialized data section. - - * `rdata' declares an initialized data section that is readable but - not writable. Microsoft compilers use this section to place - constants in it. - - * `info' defines the section to be an informational section, which is - not included in the executable file by the linker, but may (for - example) pass information _to_ the linker. For example, declaring - an `info'-type section called `.drectve' causes the linker to - interpret the contents of the section as command-line options. - - * `align=', used with a trailing number as in `obj', gives the - alignment requirements of the section. The maximum you may specify - is 64: the Win32 object file format contains no means to request a - greater section alignment than this. If alignment is not - explicitly specified, the defaults are 16-byte alignment for code - sections, 8-byte alignment for rdata sections and 4-byte alignment - for data (and BSS) sections. Informational sections get a default - alignment of 1 byte (no alignment), though the value does not - matter. - - The defaults assumed by NASM if you do not specify the above -qualifiers are: - - section .text code align=16 - section .data data align=4 - section .rdata rdata align=8 - section .bss bss align=4 - - Any other section name is treated by default like `.text'. - - -File: nasm.info, Node: Section 7.5.2, Next: Section 7.6, Prev: Section 7.5.1, Up: Section 7.5 - -7.5.2. `win32': Safe Structured Exception Handling --------------------------------------------------- - -Among other improvements in Windows XP SP2 and Windows Server 2003 -Microsoft has introduced concept of "safe structured exception -handling." General idea is to collect handlers' entry points in -designated read-only table and have alleged entry point verified -against this table prior exception control is passed to the handler. In -order for an executable module to be equipped with such "safe exception -handler table," all object modules on linker command line has to comply -with certain criteria. If one single module among them does not, then -the table in question is omitted and above mentioned run-time checks -will not be performed for application in question. Table omission is by -default silent and therefore can be easily overlooked. One can instruct -linker to refuse to produce binary without such table by passing -`/safeseh' command line option. - - Without regard to this run-time check merits it's natural to expect -NASM to be capable of generating modules suitable for `/safeseh' -linking. From developer's viewpoint the problem is two-fold: - - * how to adapt modules not deploying exception handlers of their own; - - * how to adapt/develop modules utilizing custom exception handling; - - Former can be easily achieved with any NASM version by adding -following line to source code: - - $@feat.00 equ 1 - - As of version 2.03 NASM adds this absolute symbol automatically. If -it's not already present to be precise. I.e. if for whatever reason -developer would choose to assign another value in source file, it would -still be perfectly possible. - - Registering custom exception handler on the other hand requires -certain "magic." As of version 2.03 additional directive is implemented, -`safeseh', which instructs the assembler to produce appropriately -formatted input data for above mentioned "safe exception handler table." -Its typical use would be: - - section .text - extern _MessageBoxA@16 - %if __NASM_VERSION_ID__ >= 0x02030000 - safeseh handler ; register handler as "safe handler" - %endif - handler: - push DWORD 1 ; MB_OKCANCEL - push DWORD caption - push DWORD text - push DWORD 0 - call _MessageBoxA@16 - sub eax,1 ; incidentally suits as return value - ; for exception handler - ret - global _main - _main: - push DWORD handler - push DWORD [fs:0] - mov DWORD [fs:0],esp ; engage exception handler - xor eax,eax - mov eax,DWORD[eax] ; cause exception - pop DWORD [fs:0] ; disengage exception handler - add esp,4 - ret - text: db 'OK to rethrow, CANCEL to generate core dump',0 - caption:db 'SEGV',0 - - section .drectve info - db '/defaultlib:user32.lib /defaultlib:msvcrt.lib ' - - As you might imagine, it's perfectly possible to produce .exe binary -with "safe exception handler table" and yet engage unregistered -exception handler. Indeed, handler is engaged by simply manipulating -`[fs:0]' location at run-time, something linker has no power over, -run-time that is. It should be explicitly mentioned that such failure -to register handler's entry point with `safeseh' directive has -undesired side effect at run- time. If exception is raised and -unregistered handler is to be executed, the application is abruptly -terminated without any notification whatsoever. One can argue that -system could at least have logged some kind "non-safe exception handler -in x.exe at address n" message in event log, but no, literally no -notification is provided and user is left with no clue on what caused -application failure. - - Finally, all mentions of linker in this paragraph refer to Microsoft -linker version 7.x and later. Presence of `@feat.00' symbol and input -data for "safe exception handler table" causes no backward -incompatibilities and "safeseh" modules generated by NASM 2.03 and -later can still be linked by earlier versions or non-Microsoft linkers. - - -File: nasm.info, Node: Section 7.6, Next: Section 7.6.1, Prev: Section 7.5.2, Up: Chapter 7 - -7.6. `win64': Microsoft Win64 Object Files -========================================== - -The `win64' output format generates Microsoft Win64 object files, which -is nearly 100% identical to the `win32' object format (*note Section -7.5::) with the exception that it is meant to target 64-bit code and -the x86-64 platform altogether. This object file is used exactly the -same as the `win32' object format (*note Section 7.5::), in NASM, with -regard to this exception. - -* Menu: - -* Section 7.6.1:: `win64': Writing Position-Independent Code -* Section 7.6.2:: `win64': Structured Exception Handling - - -File: nasm.info, Node: Section 7.6.1, Next: Section 7.6.2, Prev: Section 7.6, Up: Section 7.6 - -7.6.1. `win64': Writing Position-Independent Code -------------------------------------------------- - -While `REL' takes good care of RIP-relative addressing, there is one -aspect that is easy to overlook for a Win64 programmer: indirect -references. Consider a switch dispatch table: - - jmp QWORD[dsptch+rax*8] - ... - dsptch: dq case0 - dq case1 - ... - - Even novice Win64 assembler programmer will soon realize that the -code is not 64-bit savvy. Most notably linker will refuse to link it -with "`'ADDR32' relocation to '.text' invalid without -/LARGEADDRESSAWARE:NO'". So [s]he will have to split jmp instruction -as following: - - lea rbx,[rel dsptch] - jmp QWORD[rbx+rax*8] - - What happens behind the scene is that effective address in `lea' is -encoded relative to instruction pointer, or in perfectly position- -independent manner. But this is only part of the problem! Trouble is -that in .dll context `caseN' relocations will make their way to the -final module and might have to be adjusted at .dll load time. To be -specific when it can't be loaded at preferred address. And when this -occurs, pages with such relocations will be rendered private to current -process, which kind of undermines the idea of sharing .dll. But no -worry, it's trivial to fix: - - lea rbx,[rel dsptch] - add rbx,QWORD[rbx+rax*8] - jmp rbx - ... - dsptch: dq case0-dsptch - dq case1-dsptch - ... - - NASM version 2.03 and later provides another alternative, `wrt -..imagebase' operator, which returns offset from base address of the -current image, be it .exe or .dll module, therefore the name. For those -acquainted with PE-COFF format base address denotes start of -`IMAGE_DOS_HEADER' structure. Here is how to implement switch with -these image-relative references: - - lea rbx,[rel dsptch] - mov eax,DWORD[rbx+rax*4] - sub rbx,dsptch wrt ..imagebase - add rbx,rax - jmp rbx - ... - dsptch: dd case0 wrt ..imagebase - dd case1 wrt ..imagebase - - One can argue that the operator is redundant. Indeed, snippet before -last works just fine with any NASM version and is not even Windows -specific... The real reason for implementing `wrt ..imagebase' will -become apparent in next paragraph. - - It should be noted that `wrt ..imagebase' is defined as 32-bit -operand only: - - dd label wrt ..imagebase ; ok - dq label wrt ..imagebase ; bad - mov eax,label wrt ..imagebase ; ok - mov rax,label wrt ..imagebase ; bad - - -File: nasm.info, Node: Section 7.6.2, Next: Section 7.7, Prev: Section 7.6.1, Up: Section 7.6 - -7.6.2. `win64': Structured Exception Handling ---------------------------------------------- - -Structured exception handing in Win64 is completely different matter -from Win32. Upon exception program counter value is noted, and -linker-generated table comprising start and end addresses of all the -functions [in given executable module] is traversed and compared to the -saved program counter. Thus so called `UNWIND_INFO' structure is -identified. If it's not found, then offending subroutine is assumed to -be "leaf" and just mentioned lookup procedure is attempted for its -caller. In Win64 leaf function is such function that does not call any -other function _nor_ modifies any Win64 non-volatile registers, -including stack pointer. The latter ensures that it's possible to -identify leaf function's caller by simply pulling the value from the -top of the stack. - - While majority of subroutines written in assembler are not calling -any other function, requirement for non-volatile registers' -immutability leaves developer with not more than 7 registers and no -stack frame, which is not necessarily what [s]he counted with. -Customarily one would meet the requirement by saving non-volatile -registers on stack and restoring them upon return, so what can go -wrong? If [and only if] an exception is raised at run-time and no -`UNWIND_INFO' structure is associated with such "leaf" function, the -stack unwind procedure will expect to find caller's return address on -the top of stack immediately followed by its frame. Given that -developer pushed caller's non-volatile registers on stack, would the -value on top point at some code segment or even addressable space? Well, -developer can attempt copying caller's return address to the top of -stack and this would actually work in some very specific circumstances. -But unless developer can guarantee that these circumstances are always -met, it's more appropriate to assume worst case scenario, i.e. stack -unwind procedure going berserk. Relevant question is what happens then? -Application is abruptly terminated without any notification whatsoever. -Just like in Win32 case, one can argue that system could at least have -logged "unwind procedure went berserk in x.exe at address n" in event -log, but no, no trace of failure is left. - - Now, when we understand significance of the `UNWIND_INFO' structure, -let's discuss what's in it and/or how it's processed. First of all it is -checked for presence of reference to custom language-specific exception -handler. If there is one, then it's invoked. Depending on the return -value, execution flow is resumed (exception is said to be "handled"), -_or_ rest of `UNWIND_INFO' structure is processed as following. Beside -optional reference to custom handler, it carries information about -current callee's stack frame and where non-volatile registers are saved. -Information is detailed enough to be able to reconstruct contents of -caller's non-volatile registers upon call to current callee. And so -caller's context is reconstructed, and then unwind procedure is -repeated, i.e. another `UNWIND_INFO' structure is associated, this -time, with caller's instruction pointer, which is then checked for -presence of reference to language-specific handler, etc. The procedure -is recursively repeated till exception is handled. As last resort -system "handles" it by generating memory core dump and terminating the -application. - - As for the moment of this writing NASM unfortunately does not -facilitate generation of above mentioned detailed information about -stack frame layout. But as of version 2.03 it implements building -blocks for generating structures involved in stack unwinding. As -simplest example, here is how to deploy custom exception handler for -leaf function: - - default rel - section .text - extern MessageBoxA - handler: - sub rsp,40 - mov rcx,0 - lea rdx,[text] - lea r8,[caption] - mov r9,1 ; MB_OKCANCEL - call MessageBoxA - sub eax,1 ; incidentally suits as return value - ; for exception handler - add rsp,40 - ret - global main - main: - xor rax,rax - mov rax,QWORD[rax] ; cause exception - ret - main_end: - text: db 'OK to rethrow, CANCEL to generate core dump',0 - caption:db 'SEGV',0 - - section .pdata rdata align=4 - dd main wrt ..imagebase - dd main_end wrt ..imagebase - dd xmain wrt ..imagebase - section .xdata rdata align=8 - xmain: db 9,0,0,0 - dd handler wrt ..imagebase - section .drectve info - db '/defaultlib:user32.lib /defaultlib:msvcrt.lib ' - - What you see in `.pdata' section is element of the "table comprising -start and end addresses of function" along with reference to associated -`UNWIND_INFO' structure. And what you see in `.xdata' section is -`UNWIND_INFO' structure describing function with no frame, but with -designated exception handler. References are _required_ to be image- -relative (which is the real reason for implementing `wrt ..imagebase' -operator). It should be noted that `rdata align=n', as well as `wrt -..imagebase', are optional in these two segments' contexts, i.e. can -be omitted. Latter means that _all_ 32-bit references, not only above -listed required ones, placed into these two segments turn out image- -relative. Why is it important to understand? Developer is allowed to -append handler-specific data to `UNWIND_INFO' structure, and if [s]he -adds a 32-bit reference, then [s]he will have to remember to adjust its -value to obtain the real pointer. - - As already mentioned, in Win64 terms leaf function is one that does -not call any other function _nor_ modifies any non-volatile register, -including stack pointer. But it's not uncommon that assembler programmer -plans to utilize every single register and sometimes even have variable -stack frame. Is there anything one can do with bare building blocks? -I.e. besides manually composing fully-fledged `UNWIND_INFO' structure, -which would surely be considered error-prone? Yes, there is. Recall that -exception handler is called first, before stack layout is analyzed. As -it turned out, it's perfectly possible to manipulate current callee's -context in custom handler in manner that permits further stack -unwinding. General idea is that handler would not actually "handle" the -exception, but instead restore callee's context, as it was at its entry -point and thus mimic leaf function. In other words, handler would -simply undertake part of unwinding procedure. Consider following -example: - - function: - mov rax,rsp ; copy rsp to volatile register - push r15 ; save non-volatile registers - push rbx - push rbp - mov r11,rsp ; prepare variable stack frame - sub r11,rcx - and r11,-64 - mov QWORD[r11],rax ; check for exceptions - mov rsp,r11 ; allocate stack frame - mov QWORD[rsp],rax ; save original rsp value - magic_point: - ... - mov r11,QWORD[rsp] ; pull original rsp value - mov rbp,QWORD[r11-24] - mov rbx,QWORD[r11-16] - mov r15,QWORD[r11-8] - mov rsp,r11 ; destroy frame - ret - - The keyword is that up to `magic_point' original `rsp' value remains -in chosen volatile register and no non-volatile register, except for -`rsp', is modified. While past `magic_point' `rsp' remains constant -till the very end of the `function'. In this case custom -language-specific exception handler would look like this: - - EXCEPTION_DISPOSITION handler (EXCEPTION_RECORD *rec,ULONG64 frame, - CONTEXT *context,DISPATCHER_CONTEXT *disp) - { ULONG64 *rsp; - if (context->Rip<(ULONG64)magic_point) - rsp = (ULONG64 *)context->Rax; - else - { rsp = ((ULONG64 **)context->Rsp)[0]; - context->Rbp = rsp[-3]; - context->Rbx = rsp[-2]; - context->R15 = rsp[-1]; - } - context->Rsp = (ULONG64)rsp; - - memcpy (disp->ContextRecord,context,sizeof(CONTEXT)); - RtlVirtualUnwind(UNW_FLAG_NHANDLER,disp->ImageBase, - dips->ControlPc,disp->FunctionEntry,disp->ContextRecord, - &disp->HandlerData,&disp->EstablisherFrame,NULL); - return ExceptionContinueSearch; - } - - As custom handler mimics leaf function, corresponding `UNWIND_INFO' -structure does not have to contain any information about stack frame and -its layout. - - -File: nasm.info, Node: Section 7.7, Next: Section 7.8, Prev: Section 7.6.2, Up: Chapter 7 - -7.7. `coff': Common Object File Format -====================================== - -The `coff' output type produces `COFF' object files suitable for -linking with the DJGPP linker. - - `coff' provides a default output file-name extension of `.o'. - - The `coff' format supports the same extensions to the `SECTION' -directive as `win32' does, except that the `align' qualifier and the -`info' section type are not supported. - - -File: nasm.info, Node: Section 7.8, Next: Section 7.9, Prev: Section 7.7, Up: Chapter 7 - -7.8. `macho32' and `macho64': Mach Object File Format -===================================================== - -The `macho32' and `macho64' output formts produces `Mach-O' object -files suitable for linking with the MacOS X linker. `macho' is a -synonym for `macho32'. - - `macho' provides a default output file-name extension of `.o'. - - -File: nasm.info, Node: Section 7.9, Next: Section 7.9.1, Prev: Section 7.8, Up: Chapter 7 - -7.9. `elf32' and `elf64': Executable and Linkable Format Object Files -===================================================================== - -The `elf32' and `elf64' output formats generate `ELF32 and ELF64' -(Executable and Linkable Format) object files, as used by Linux as well -as Unix System V, including Solaris x86, UnixWare and SCO Unix. `elf' -provides a default output file-name extension of `.o'. `elf' is a -synonym for `elf32'. - -* Menu: - -* Section 7.9.1:: ELF specific directive `osabi' -* Section 7.9.2:: `elf' Extensions to the `SECTION' Directive -* Section 7.9.3:: Position-Independent Code: `elf' Special Symbols and `WRT' -* Section 7.9.4:: Thread Local Storage: `elf' Special Symbols and `WRT' -* Section 7.9.5:: `elf' Extensions to the `GLOBAL' Directive -* Section 7.9.6:: `elf' Extensions to the `COMMON' Directive -* Section 7.9.7:: 16-bit code and ELF -* Section 7.9.8:: Debug formats and ELF - - -File: nasm.info, Node: Section 7.9.1, Next: Section 7.9.2, Prev: Section 7.9, Up: Section 7.9 - -7.9.1. ELF specific directive `osabi' -------------------------------------- - -The ELF header specifies the application binary interface for the target -operating system (OSABI). This field can be set by using the `osabi' -directive with the numeric value (0-255) of the target system. If this -directive is not used, the default value will be "UNIX System V ABI" (0) -which will work on most systems which support ELF. - - -File: nasm.info, Node: Section 7.9.2, Next: Section 7.9.3, Prev: Section 7.9.1, Up: Section 7.9 - -7.9.2. `elf' Extensions to the `SECTION' Directive --------------------------------------------------- - -Like the `obj' format, `elf' allows you to specify additional -information on the `SECTION' directive line, to control the type and -properties of sections you declare. Section types and properties are -generated automatically by NASM for the standard section names, but may -still be overridden by these qualifiers. - - The available qualifiers are: - - * `alloc' defines the section to be one which is loaded into memory - when the program is run. `noalloc' defines it to be one which is - not, such as an informational or comment section. - - * `exec' defines the section to be one which should have execute - permission when the program is run. `noexec' defines it as one - which should not. - - * `write' defines the section to be one which should be writable when - the program is run. `nowrite' defines it as one which should not. - - * `progbits' defines the section to be one with explicit contents - stored in the object file: an ordinary code or data section, for - example, `nobits' defines the section to be one with no explicit - contents given, such as a BSS section. - - * `align=', used with a trailing number as in `obj', gives the - alignment requirements of the section. - - * `tls' defines the section to be one which contains thread local - variables. - - The defaults assumed by NASM if you do not specify the above -qualifiers are: - - section .text progbits alloc exec nowrite align=16 - section .rodata progbits alloc noexec nowrite align=4 - section .lrodata progbits alloc noexec nowrite align=4 - section .data progbits alloc noexec write align=4 - section .ldata progbits alloc noexec write align=4 - section .bss nobits alloc noexec write align=4 - section .lbss nobits alloc noexec write align=4 - section .tdata progbits alloc noexec write align=4 tls - section .tbss nobits alloc noexec write align=4 tls - section .comment progbits noalloc noexec nowrite align=1 - section other progbits alloc noexec nowrite align=1 - - (Any section name other than those in the above table is treated by -default like `other' in the above table. Please note that section names -are case sensitive.) - - -File: nasm.info, Node: Section 7.9.3, Next: Section 7.9.4, Prev: Section 7.9.2, Up: Section 7.9 - -7.9.3. Position-Independent Code: `elf' Special Symbols and `WRT' ------------------------------------------------------------------ - -The `ELF' specification contains enough features to allow position- -independent code (PIC) to be written, which makes ELF shared libraries -very flexible. However, it also means NASM has to be able to generate a -variety of ELF specific relocation types in ELF object files, if it is -to be an assembler which can write PIC. - - Since `ELF' does not support segment-base references, the `WRT' -operator is not used for its normal purpose; therefore NASM's `elf' -output format makes use of `WRT' for a different purpose, namely the -PIC-specific relocation types. - - `elf' defines five special symbols which you can use as the -right-hand side of the `WRT' operator to obtain PIC relocation types. -They are `..gotpc', `..gotoff', `..got', `..plt' and `..sym'. Their -functions are summarized here: - - * Referring to the symbol marking the global offset table base using - `wrt ..gotpc' will end up giving the distance from the beginning of - the current section to the global offset table. - (`_GLOBAL_OFFSET_TABLE_' is the standard symbol name used to refer - to the GOT.) So you would then need to add `$$' to the result to - get the real address of the GOT. - - * Referring to a location in one of your own sections using `wrt - ..gotoff' will give the distance from the beginning of the GOT to - the specified location, so that adding on the address of the GOT - would give the real address of the location you wanted. - - * Referring to an external or global symbol using `wrt ..got' causes - the linker to build an entry _in_ the GOT containing the address - of the symbol, and the reference gives the distance from the - beginning of the GOT to the entry; so you can add on the address - of the GOT, load from the resulting address, and end up with the - address of the symbol. - - * Referring to a procedure name using `wrt ..plt' causes the linker - to build a procedure linkage table entry for the symbol, and the - reference gives the address of the PLT entry. You can only use - this in contexts which would generate a PC-relative relocation - normally (i.e. as the destination for `CALL' or `JMP'), since ELF - contains no relocation type to refer to PLT entries absolutely. - - * Referring to a symbol name using `wrt ..sym' causes NASM to write - an ordinary relocation, but instead of making the relocation - relative to the start of the section and then adding on the offset - to the symbol, it will write a relocation record aimed directly at - the symbol in question. The distinction is a necessary one due to - a peculiarity of the dynamic linker. - - A fuller explanation of how to use these relocation types to write -shared libraries entirely in NASM is given in *note Section 9.2::. - - -File: nasm.info, Node: Section 7.9.4, Next: Section 7.9.5, Prev: Section 7.9.3, Up: Section 7.9 - -7.9.4. Thread Local Storage: `elf' Special Symbols and `WRT' ------------------------------------------------------------- - - * In ELF32 mode, referring to an external or global symbol using - `wrt ..tlsie' causes the linker to build an entry _in_ the GOT - containing the offset of the symbol within the TLS block, so you - can access the value of the symbol with code such as: - - mov eax,[tid wrt ..tlsie] - mov [gs:eax],ebx - - * In ELF64 mode, referring to an external or global symbol using - `wrt ..gottpoff' causes the linker to build an entry _in_ the GOT - containing the offset of the symbol within the TLS block, so you - can access the value of the symbol with code such as: - - mov rax,[rel tid wrt ..gottpoff] - mov rcx,[fs:rax] - - -File: nasm.info, Node: Section 7.9.5, Next: Section 7.9.6, Prev: Section 7.9.4, Up: Section 7.9 - -7.9.5. `elf' Extensions to the `GLOBAL' Directive -------------------------------------------------- - -`ELF' object files can contain more information about a global symbol -than just its address: they can contain the size of the symbol and its -type as well. These are not merely debugger conveniences, but are -actually necessary when the program being written is a shared library. -NASM therefore supports some extensions to the `GLOBAL' directive, -allowing you to specify these features. - - You can specify whether a global variable is a function or a data -object by suffixing the name with a colon and the word `function' or -`data'. (`object' is a synonym for `data'.) For example: - - global hashlookup:function, hashtable:data - - exports the global symbol `hashlookup' as a function and `hashtable' -as a data object. - - Optionally, you can control the ELF visibility of the symbol. Just -add one of the visibility keywords: `default', `internal', `hidden', or -`protected'. The default is `default' of course. For example, to make -`hashlookup' hidden: - - global hashlookup:function hidden - - You can also specify the size of the data associated with the -symbol, as a numeric expression (which may involve labels, and even -forward references) after the type specifier. Like this: - - global hashtable:data (hashtable.end - hashtable) - - hashtable: - db this,that,theother ; some data here - .end: - - This makes NASM automatically calculate the length of the table and -place that information into the `ELF' symbol table. - - Declaring the type and size of global symbols is necessary when -writing shared library code. For more information, see *note Section -9.2.4::. - - -File: nasm.info, Node: Section 7.9.6, Next: Section 7.9.7, Prev: Section 7.9.5, Up: Section 7.9 - -7.9.6. `elf' Extensions to the `COMMON' Directive -------------------------------------------------- - -`ELF' also allows you to specify alignment requirements on common -variables. This is done by putting a number (which must be a power of -two) after the name and size of the common variable, separated (as -usual) by a colon. For example, an array of doublewords would benefit -from 4-byte alignment: - - common dwordarray 128:4 - - This declares the total size of the array to be 128 bytes, and -requires that it be aligned on a 4-byte boundary. - - -File: nasm.info, Node: Section 7.9.7, Next: Section 7.9.8, Prev: Section 7.9.6, Up: Section 7.9 - -7.9.7. 16-bit code and ELF --------------------------- - -The `ELF32' specification doesn't provide relocations for 8- and 16- -bit values, but the GNU `ld' linker adds these as an extension. NASM -can generate GNU-compatible relocations, to allow 16-bit code to be -linked as ELF using GNU `ld'. If NASM is used with the -`-w+gnu-elf-extensions' option, a warning is issued when one of these -relocations is generated. - - -File: nasm.info, Node: Section 7.9.8, Next: Section 7.10, Prev: Section 7.9.7, Up: Section 7.9 - -7.9.8. Debug formats and ELF ----------------------------- - -`ELF32' and `ELF64' provide debug information in `STABS' and `DWARF' -formats. Line number information is generated for all executable -sections, but please note that only the ".text" section is executable -by default. - - -File: nasm.info, Node: Section 7.10, Next: Section 7.11, Prev: Section 7.9.8, Up: Chapter 7 - -7.10. `aout': Linux `a.out' Object Files -======================================== - -The `aout' format generates `a.out' object files, in the form used by -early Linux systems (current Linux systems use ELF, see *note Section -7.9::.) These differ from other `a.out' object files in that the magic -number in the first four bytes of the file is different; also, some -implementations of `a.out', for example NetBSD's, support -position-independent code, which Linux's implementation does not. - - `a.out' provides a default output file-name extension of `.o'. - - `a.out' is a very simple object format. It supports no special -directives, no special symbols, no use of `SEG' or `WRT', and no -extensions to any standard directives. It supports only the three -standard section names `.text', `.data' and `.bss'. - - -File: nasm.info, Node: Section 7.11, Next: Section 7.12, Prev: Section 7.10, Up: Chapter 7 - -7.11. `aoutb': NetBSD/FreeBSD/OpenBSD `a.out' Object Files -========================================================== - -The `aoutb' format generates `a.out' object files, in the form used by -the various free `BSD Unix' clones, `NetBSD', `FreeBSD' and `OpenBSD'. -For simple object files, this object format is exactly the same as -`aout' except for the magic number in the first four bytes of the file. -However, the `aoutb' format supports position-independent code in the -same way as the `elf' format, so you can use it to write `BSD' shared -libraries. - - `aoutb' provides a default output file-name extension of `.o'. - - `aoutb' supports no special directives, no special symbols, and only -the three standard section names `.text', `.data' and `.bss'. However, -it also supports the same use of `WRT' as `elf' does, to provide -position-independent code relocation types. See *note Section 7.9.3:: -for full documentation of this feature. - - `aoutb' also supports the same extensions to the `GLOBAL' directive -as `elf' does: see *note Section 7.9.5:: for documentation of this. - - -File: nasm.info, Node: Section 7.12, Next: Section 7.13, Prev: Section 7.11, Up: Chapter 7 - -7.12. `as86': Minix/Linux `as86' Object Files -============================================= - -The Minix/Linux 16-bit assembler `as86' has its own non-standard object -file format. Although its companion linker `ld86' produces something -close to ordinary `a.out' binaries as output, the object file format -used to communicate between `as86' and `ld86' is not itself `a.out'. - - NASM supports this format, just in case it is useful, as `as86'. -`as86' provides a default output file-name extension of `.o'. - - `as86' is a very simple object format (from the NASM user's point of -view). It supports no special directives, no use of `SEG' or `WRT', and -no extensions to any standard directives. It supports only the three -standard section names `.text', `.data' and `.bss'. The only special -symbol supported is `..start'. - - -File: nasm.info, Node: Section 7.13, Next: Section 7.13.1, Prev: Section 7.12, Up: Chapter 7 - -7.13. `rdf': Relocatable Dynamic Object File Format -=================================================== - -The `rdf' output format produces `RDOFF' object files. `RDOFF' -(Relocatable Dynamic Object File Format) is a home-grown object-file -format, designed alongside NASM itself and reflecting in its file -format the internal structure of the assembler. - - `RDOFF' is not used by any well-known operating systems. Those -writing their own systems, however, may well wish to use `RDOFF' as -their object format, on the grounds that it is designed primarily for -simplicity and contains very little file-header bureaucracy. - - The Unix NASM archive, and the DOS archive which includes sources, -both contain an `rdoff' subdirectory holding a set of RDOFF utilities: -an RDF linker, an `RDF' static-library manager, an RDF file dump -utility, and a program which will load and execute an RDF executable -under Linux. - - `rdf' supports only the standard section names `.text', `.data' and -`.bss'. - -* Menu: - -* Section 7.13.1:: Requiring a Library: The `LIBRARY' Directive -* Section 7.13.2:: Specifying a Module Name: The `MODULE' Directive -* Section 7.13.3:: `rdf' Extensions to the `GLOBAL' Directive -* Section 7.13.4:: `rdf' Extensions to the `EXTERN' Directive - - -File: nasm.info, Node: Section 7.13.1, Next: Section 7.13.2, Prev: Section 7.13, Up: Section 7.13 - -7.13.1. Requiring a Library: The `LIBRARY' Directive ----------------------------------------------------- - -`RDOFF' contains a mechanism for an object file to demand a given -library to be linked to the module, either at load time or run time. -This is done by the `LIBRARY' directive, which takes one argument which -is the name of the module: - - library mylib.rdl - - -File: nasm.info, Node: Section 7.13.2, Next: Section 7.13.3, Prev: Section 7.13.1, Up: Section 7.13 - -7.13.2. Specifying a Module Name: The `MODULE' Directive --------------------------------------------------------- - -Special `RDOFF' header record is used to store the name of the module. -It can be used, for example, by run-time loader to perform dynamic -linking. `MODULE' directive takes one argument which is the name of -current module: - - module mymodname - - Note that when you statically link modules and tell linker to strip -the symbols from output file, all module names will be stripped too. To -avoid it, you should start module names with `$', like: - - module $kernel.core - - -File: nasm.info, Node: Section 7.13.3, Next: Section 7.13.4, Prev: Section 7.13.2, Up: Section 7.13 - -7.13.3. `rdf' Extensions to the `GLOBAL' Directive --------------------------------------------------- - -`RDOFF' global symbols can contain additional information needed by the -static linker. You can mark a global symbol as exported, thus telling -the linker do not strip it from target executable or library file. Like -in `ELF', you can also specify whether an exported symbol is a procedure -(function) or data object. - - Suffixing the name with a colon and the word `export' you make the -symbol exported: - - global sys_open:export - - To specify that exported symbol is a procedure (function), you add -the word `proc' or `function' after declaration: - - global sys_open:export proc - - Similarly, to specify exported data object, add the word `data' or -`object' to the directive: - - global kernel_ticks:export data - - -File: nasm.info, Node: Section 7.13.4, Next: Section 7.14, Prev: Section 7.13.3, Up: Section 7.13 - -7.13.4. `rdf' Extensions to the `EXTERN' Directive --------------------------------------------------- - -By default the `EXTERN' directive in `RDOFF' declares a "pure external" -symbol (i.e. the static linker will complain if such a symbol is not -resolved). To declare an "imported" symbol, which must be resolved -later during a dynamic linking phase, `RDOFF' offers an additional -`import' modifier. As in `GLOBAL', you can also specify whether an -imported symbol is a procedure (function) or data object. For example: - - library $libc - extern _open:import - extern _printf:import proc - extern _errno:import data - - Here the directive `LIBRARY' is also included, which gives the -dynamic linker a hint as to where to find requested symbols. - - -File: nasm.info, Node: Section 7.14, Next: Chapter 8, Prev: Section 7.13.4, Up: Chapter 7 - -7.14. `dbg': Debugging Format -============================= - -The `dbg' output format is not built into NASM in the default -configuration. If you are building your own NASM executable from the -sources, you can define `OF_DBG' in `output/outform.h' or on the -compiler command line, and obtain the `dbg' output format. - - The `dbg' format does not output an object file as such; instead, it -outputs a text file which contains a complete list of all the -transactions between the main body of NASM and the output-format back -end module. It is primarily intended to aid people who want to write -their own output drivers, so that they can get a clearer idea of the -various requests the main program makes of the output driver, and in -what order they happen. - - For simple files, one can easily use the `dbg' format like this: - - nasm -f dbg filename.asm - - which will generate a diagnostic file called `filename.dbg'. However, -this will not work well on files which were designed for a different -object format, because each object format defines its own macros -(usually user- level forms of directives), and those macros will not be -defined in the `dbg' format. Therefore it can be useful to run NASM -twice, in order to do the preprocessing with the native object format -selected: - - nasm -e -f rdf -o rdfprog.i rdfprog.asm - nasm -a -f dbg rdfprog.i - - This preprocesses `rdfprog.asm' into `rdfprog.i', keeping the `rdf' -object format selected in order to make sure RDF special directives are -converted into primitive form correctly. Then the preprocessed source -is fed through the `dbg' format to generate the final diagnostic output. - - This workaround will still typically not work for programs intended -for `obj' format, because the `obj' `SEGMENT' and `GROUP' directives -have side effects of defining the segment and group names as symbols; -`dbg' will not do this, so the program will not assemble. You will have -to work around that by defining the symbols yourself (using `EXTERN', -for example) if you really need to get a `dbg' trace of an -`obj'-specific source file. - - `dbg' accepts any section name and any directives at all, and logs -them all to its output file. - - -File: nasm.info, Node: Chapter 8, Next: Section 8.1, Prev: Section 7.14, Up: Top - -Chapter 8: Writing 16-bit Code (DOS, Windows 3/3.1) -*************************************************** - -This chapter attempts to cover some of the common issues encountered -when writing 16-bit code to run under `MS-DOS' or `Windows 3.x'. It -covers how to link programs to produce `.EXE' or `.COM' files, how to -write `.SYS' device drivers, and how to interface assembly language -code with 16-bit C compilers and with Borland Pascal. - -* Menu: - -* Section 8.1:: Producing `.EXE' Files -* Section 8.2:: Producing `.COM' Files -* Section 8.3:: Producing `.SYS' Files -* Section 8.4:: Interfacing to 16-bit C Programs -* Section 8.5:: Interfacing to Borland Pascal Programs - - -File: nasm.info, Node: Section 8.1, Next: Section 8.1.1, Prev: Chapter 8, Up: Chapter 8 - -8.1. Producing `.EXE' Files -=========================== - -Any large program written under DOS needs to be built as a `.EXE' file: -only `.EXE' files have the necessary internal structure required to -span more than one 64K segment. Windows programs, also, have to be built -as `.EXE' files, since Windows does not support the `.COM' format. - - In general, you generate `.EXE' files by using the `obj' output -format to produce one or more `.OBJ' files, and then linking them -together using a linker. However, NASM also supports the direct -generation of simple DOS `.EXE' files using the `bin' output format (by -using `DB' and `DW' to construct the `.EXE' file header), and a macro -package is supplied to do this. Thanks to Yann Guidon for contributing -the code for this. - - NASM may also support `.EXE' natively as another output format in -future releases. - -* Menu: - -* Section 8.1.1:: Using the `obj' Format To Generate `.EXE' Files -* Section 8.1.2:: Using the `bin' Format To Generate `.EXE' Files - - -File: nasm.info, Node: Section 8.1.1, Next: Section 8.1.2, Prev: Section 8.1, Up: Section 8.1 - -8.1.1. Using the `obj' Format To Generate `.EXE' Files ------------------------------------------------------- - -This section describes the usual method of generating `.EXE' files by -linking `.OBJ' files together. - - Most 16-bit programming language packages come with a suitable -linker; if you have none of these, there is a free linker called VAL, -available in `LZH' archive format from `x2ftp.oulu.fi'. An LZH archiver -can be found at `ftp.simtel.net'. There is another `free' linker -(though this one doesn't come with sources) called FREELINK, available -from `www.pcorner.com'. A third, `djlink', written by DJ Delorie, is -available at `www.delorie.com'. A fourth linker, `ALINK', written by -Anthony A.J. Williams, is available at `alink.sourceforge.net'. - - When linking several `.OBJ' files into a `.EXE' file, you should -ensure that exactly one of them has a start point defined (using the -`..start' special symbol defined by the `obj' format: see *note Section -7.4.6::). If no module defines a start point, the linker will not know -what value to give the entry-point field in the output file header; if -more than one defines a start point, the linker will not know _which_ -value to use. - - An example of a NASM source file which can be assembled to a `.OBJ' -file and linked on its own to a `.EXE' is given here. It demonstrates -the basic principles of defining a stack, initialising the segment -registers, and declaring a start point. This file is also provided in -the `test' subdirectory of the NASM archives, under the name -`objexe.asm'. - - segment code - - ..start: - mov ax,data - mov ds,ax - mov ax,stack - mov ss,ax - mov sp,stacktop - - This initial piece of code sets up `DS' to point to the data segment, -and initializes `SS' and `SP' to point to the top of the provided -stack. Notice that interrupts are implicitly disabled for one -instruction after a move into `SS', precisely for this situation, so -that there's no chance of an interrupt occurring between the loads of -`SS' and `SP' and not having a stack to execute on. - - Note also that the special symbol `..start' is defined at the -beginning of this code, which means that will be the entry point into -the resulting executable file. - - mov dx,hello - mov ah,9 - int 0x21 - - The above is the main program: load `DS:DX' with a pointer to the -greeting message (`hello' is implicitly relative to the segment `data', -which was loaded into `DS' in the setup code, so the full pointer is -valid), and call the DOS print-string function. - - mov ax,0x4c00 - int 0x21 - - This terminates the program using another DOS system call. - - segment data - - hello: db 'hello, world', 13, 10, '$' - - The data segment contains the string we want to display. - - segment stack stack - resb 64 - stacktop: - - The above code declares a stack segment containing 64 bytes of -uninitialized stack space, and points `stacktop' at the top of it. The -directive `segment stack stack' defines a segment _called_ `stack', and -also of _type_ `STACK'. The latter is not necessary to the correct -running of the program, but linkers are likely to issue warnings or -errors if your program has no segment of type `STACK'. - - The above file, when assembled into a `.OBJ' file, will link on its -own to a valid `.EXE' file, which when run will print `hello, world' -and then exit. - - -File: nasm.info, Node: Section 8.1.2, Next: Section 8.2, Prev: Section 8.1.1, Up: Section 8.1 - -8.1.2. Using the `bin' Format To Generate `.EXE' Files ------------------------------------------------------- - -The `.EXE' file format is simple enough that it's possible to build a -`.EXE' file by writing a pure-binary program and sticking a 32-byte -header on the front. This header is simple enough that it can be -generated using `DB' and `DW' commands by NASM itself, so that you can -use the `bin' output format to directly generate `.EXE' files. - - Included in the NASM archives, in the `misc' subdirectory, is a file -`exebin.mac' of macros. It defines three macros: `EXE_begin', -`EXE_stack' and `EXE_end'. - - To produce a `.EXE' file using this method, you should start by using -`%include' to load the `exebin.mac' macro package into your source -file. You should then issue the `EXE_begin' macro call (which takes no -arguments) to generate the file header data. Then write code as normal -for the `bin' format - you can use all three standard sections `.text', -`.data' and `.bss'. At the end of the file you should call the -`EXE_end' macro (again, no arguments), which defines some symbols to -mark section sizes, and these symbols are referred to in the header -code generated by `EXE_begin'. - - In this model, the code you end up writing starts at `0x100', just -like a `.COM' file - in fact, if you strip off the 32-byte header from -the resulting `.EXE' file, you will have a valid `.COM' program. All -the segment bases are the same, so you are limited to a 64K program, -again just like a `.COM' file. Note that an `ORG' directive is issued -by the `EXE_begin' macro, so you should not explicitly issue one of -your own. - - You can't directly refer to your segment base value, unfortunately, -since this would require a relocation in the header, and things would -get a lot more complicated. So you should get your segment base by -copying it out of `CS' instead. - - On entry to your `.EXE' file, `SS:SP' are already set up to point to -the top of a 2Kb stack. You can adjust the default stack size of 2Kb by -calling the `EXE_stack' macro. For example, to change the stack size of -your program to 64 bytes, you would call `EXE_stack 64'. - - A sample program which generates a `.EXE' file in this way is given -in the `test' subdirectory of the NASM archive, as `binexe.asm'. - - -File: nasm.info, Node: Section 8.2, Next: Section 8.2.1, Prev: Section 8.1.2, Up: Chapter 8 - -8.2. Producing `.COM' Files -=========================== - -While large DOS programs must be written as `.EXE' files, small ones -are often better written as `.COM' files. `.COM' files are pure binary, -and therefore most easily produced using the `bin' output format. - -* Menu: - -* Section 8.2.1:: Using the `bin' Format To Generate `.COM' Files -* Section 8.2.2:: Using the `obj' Format To Generate `.COM' Files - - -File: nasm.info, Node: Section 8.2.1, Next: Section 8.2.2, Prev: Section 8.2, Up: Section 8.2 - -8.2.1. Using the `bin' Format To Generate `.COM' Files ------------------------------------------------------- - -`.COM' files expect to be loaded at offset `100h' into their segment -(though the segment may change). Execution then begins at `100h', i.e. -right at the start of the program. So to write a `.COM' program, you -would create a source file looking like - - org 100h - - section .text - - start: - ; put your code here - - section .data - - ; put data items here - - section .bss - - ; put uninitialized data here - - The `bin' format puts the `.text' section first in the file, so you -can declare data or BSS items before beginning to write code if you -want to and the code will still end up at the front of the file where it -belongs. - - The BSS (uninitialized data) section does not take up space in the -`.COM' file itself: instead, addresses of BSS items are resolved to -point at space beyond the end of the file, on the grounds that this -will be free memory when the program is run. Therefore you should not -rely on your BSS being initialized to all zeros when you run. - - To assemble the above program, you should use a command line like - - nasm myprog.asm -fbin -o myprog.com - - The `bin' format would produce a file called `myprog' if no explicit -output file name were specified, so you have to override it and give -the desired file name. - - -File: nasm.info, Node: Section 8.2.2, Next: Section 8.3, Prev: Section 8.2.1, Up: Section 8.2 - -8.2.2. Using the `obj' Format To Generate `.COM' Files ------------------------------------------------------- - -If you are writing a `.COM' program as more than one module, you may -wish to assemble several `.OBJ' files and link them together into a -`.COM' program. You can do this, provided you have a linker capable of -outputting `.COM' files directly (TLINK does this), or alternatively a -converter program such as `EXE2BIN' to transform the `.EXE' file output -from the linker into a `.COM' file. - - If you do this, you need to take care of several things: - - * The first object file containing code should start its code - segment with a line like `RESB 100h'. This is to ensure that the - code begins at offset `100h' relative to the beginning of the code - segment, so that the linker or converter program does not have to - adjust address references within the file when generating the - `.COM' file. Other assemblers use an `ORG' directive for this - purpose, but `ORG' in NASM is a format-specific directive to the - `bin' output format, and does not mean the same thing as it does - in MASM-compatible assemblers. - - * You don't need to define a stack segment. - - * All your segments should be in the same group, so that every time - your code or data references a symbol offset, all offsets are - relative to the same segment base. This is because, when a `.COM' - file is loaded, all the segment registers contain the same value. - - -File: nasm.info, Node: Section 8.3, Next: Section 8.4, Prev: Section 8.2.2, Up: Chapter 8 - -8.3. Producing `.SYS' Files -=========================== - -MS-DOS device drivers - `.SYS' files - are pure binary files, similar -to `.COM' files, except that they start at origin zero rather than -`100h'. Therefore, if you are writing a device driver using the `bin' -format, you do not need the `ORG' directive, since the default origin -for `bin' is zero. Similarly, if you are using `obj', you do not need -the `RESB 100h' at the start of your code segment. - - `.SYS' files start with a header structure, containing pointers to -the various routines inside the driver which do the work. This -structure should be defined at the start of the code segment, even -though it is not actually code. - - For more information on the format of `.SYS' files, and the data -which has to go in the header structure, a list of books is given in the -Frequently Asked Questions list for the newsgroup -`comp.os.msdos.programmer'. - - -File: nasm.info, Node: Section 8.4, Next: Section 8.4.1, Prev: Section 8.3, Up: Chapter 8 - -8.4. Interfacing to 16-bit C Programs -===================================== - -This section covers the basics of writing assembly routines that call, -or are called from, C programs. To do this, you would typically write an -assembly module as a `.OBJ' file, and link it with your C modules to -produce a mixed-language program. - -* Menu: - -* Section 8.4.1:: External Symbol Names -* Section 8.4.2:: Memory Models -* Section 8.4.3:: Function Definitions and Function Calls -* Section 8.4.4:: Accessing Data Items -* Section 8.4.5:: `c16.mac': Helper Macros for the 16-bit C Interface - - -File: nasm.info, Node: Section 8.4.1, Next: Section 8.4.2, Prev: Section 8.4, Up: Section 8.4 - -8.4.1. External Symbol Names ----------------------------- - -C compilers have the convention that the names of all global symbols -(functions or data) they define are formed by prefixing an underscore to -the name as it appears in the C program. So, for example, the function -a C programmer thinks of as `printf' appears to an assembly language -programmer as `_printf'. This means that in your assembly programs, you -can define symbols without a leading underscore, and not have to worry -about name clashes with C symbols. - - If you find the underscores inconvenient, you can define macros to -replace the `GLOBAL' and `EXTERN' directives as follows: - - %macro cglobal 1 - - global _%1 - %define %1 _%1 - - %endmacro - - %macro cextern 1 - - extern _%1 - %define %1 _%1 - - %endmacro - - (These forms of the macros only take one argument at a time; a `%rep' -construct could solve this.) - - If you then declare an external like this: - - cextern printf - - then the macro will expand it as - - extern _printf - %define printf _printf - - Thereafter, you can reference `printf' as if it was a symbol, and the -preprocessor will put the leading underscore on where necessary. - - The `cglobal' macro works similarly. You must use `cglobal' before -defining the symbol in question, but you would have had to do that -anyway if you used `GLOBAL'. - - Also see *note Section 2.1.27::. - - -File: nasm.info, Node: Section 8.4.2, Next: Section 8.4.3, Prev: Section 8.4.1, Up: Section 8.4 - -8.4.2. Memory Models --------------------- - -NASM contains no mechanism to support the various C memory models -directly; you have to keep track yourself of which one you are writing -for. This means you have to keep track of the following things: - - * In models using a single code segment (tiny, small and compact), - functions are near. This means that function pointers, when stored - in data segments or pushed on the stack as function arguments, are - 16 bits long and contain only an offset field (the `CS' register - never changes its value, and always gives the segment part of the - full function address), and that functions are called using - ordinary near `CALL' instructions and return using `RETN' (which, - in NASM, is synonymous with `RET' anyway). This means both that - you should write your own routines to return with `RETN', and that - you should call external C routines with near `CALL' instructions. - - * In models using more than one code segment (medium, large and - huge), functions are far. This means that function pointers are 32 - bits long (consisting of a 16-bit offset followed by a 16-bit - segment), and that functions are called using `CALL FAR' (or `CALL - seg:offset') and return using `RETF'. Again, you should therefore - write your own routines to return with `RETF' and use `CALL FAR' - to call external routines. - - * In models using a single data segment (tiny, small and medium), - data pointers are 16 bits long, containing only an offset field - (the `DS' register doesn't change its value, and always gives the - segment part of the full data item address). - - * In models using more than one data segment (compact, large and - huge), data pointers are 32 bits long, consisting of a 16-bit - offset followed by a 16- bit segment. You should still be careful - not to modify `DS' in your routines without restoring it - afterwards, but `ES' is free for you to use to access the contents - of 32-bit data pointers you are passed. - - * The huge memory model allows single data items to exceed 64K in - size. In all other memory models, you can access the whole of a - data item just by doing arithmetic on the offset field of the - pointer you are given, whether a segment field is present or not; - in huge model, you have to be more careful of your pointer - arithmetic. - - * In most memory models, there is a _default_ data segment, whose - segment address is kept in `DS' throughout the program. This data - segment is typically the same segment as the stack, kept in `SS', - so that functions' local variables (which are stored on the stack) - and global data items can both be accessed easily without changing - `DS'. Particularly large data items are typically stored in other - segments. However, some memory models (though not the standard - ones, usually) allow the assumption that `SS' and `DS' hold the - same value to be removed. Be careful about functions' local - variables in this latter case. - - In models with a single code segment, the segment is called `_TEXT', -so your code segment must also go by this name in order to be linked -into the same place as the main code segment. In models with a single -data segment, or with a default data segment, it is called `_DATA'. - - -File: nasm.info, Node: Section 8.4.3, Next: Section 8.4.4, Prev: Section 8.4.2, Up: Section 8.4 - -8.4.3. Function Definitions and Function Calls ----------------------------------------------- - -The C calling convention in 16-bit programs is as follows. In the -following description, the words _caller_ and _callee_ are used to -denote the function doing the calling and the function which gets -called. - - * The caller pushes the function's parameters on the stack, one after - another, in reverse order (right to left, so that the first - argument specified to the function is pushed last). - - * The caller then executes a `CALL' instruction to pass control to - the callee. This `CALL' is either near or far depending on the - memory model. - - * The callee receives control, and typically (although this is not - actually necessary, in functions which do not need to access their - parameters) starts by saving the value of `SP' in `BP' so as to be - able to use `BP' as a base pointer to find its parameters on the - stack. However, the caller was probably doing this too, so part - of the calling convention states that `BP' must be preserved by - any C function. Hence the callee, if it is going to set up `BP' as - a _frame pointer_, must push the previous value first. - - * The callee may then access its parameters relative to `BP'. The - word at `[BP]' holds the previous value of `BP' as it was pushed; - the next word, at `[BP+2]', holds the offset part of the return - address, pushed implicitly by `CALL'. In a small-model (near) - function, the parameters start after that, at `[BP+4]'; in a - large-model (far) function, the segment part of the return address - lives at `[BP+4]', and the parameters begin at `[BP+6]'. The - leftmost parameter of the function, since it was pushed last, is - accessible at this offset from `BP'; the others follow, at - successively greater offsets. Thus, in a function such as `printf' - which takes a variable number of parameters, the pushing of the - parameters in reverse order means that the function knows where to - find its first parameter, which tells it the number and type of - the remaining ones. - - * The callee may also wish to decrease `SP' further, so as to - allocate space on the stack for local variables, which will then - be accessible at negative offsets from `BP'. - - * The callee, if it wishes to return a value to the caller, should - leave the value in `AL', `AX' or `DX:AX' depending on the size of - the value. Floating-point results are sometimes (depending on the - compiler) returned in `ST0'. - - * Once the callee has finished processing, it restores `SP' from - `BP' if it had allocated local stack space, then pops the previous - value of `BP', and returns via `RETN' or `RETF' depending on - memory model. - - * When the caller regains control from the callee, the function - parameters are still on the stack, so it typically adds an - immediate constant to `SP' to remove them (instead of executing a - number of slow `POP' instructions). Thus, if a function is - accidentally called with the wrong number of parameters due to a - prototype mismatch, the stack will still be returned to a sensible - state since the caller, which _knows_ how many parameters it - pushed, does the removing. - - It is instructive to compare this calling convention with that for -Pascal programs (described in *note Section 8.5.1::). Pascal has a -simpler convention, since no functions have variable numbers of -parameters. Therefore the callee knows how many parameters it should -have been passed, and is able to deallocate them from the stack itself -by passing an immediate argument to the `RET' or `RETF' instruction, so -the caller does not have to do it. Also, the parameters are pushed in -left-to- right order, not right-to-left, which means that a compiler -can give better guarantees about sequence points without performance -suffering. - - Thus, you would define a function in C style in the following way. -The following example is for small model: - - global _myfunc - - _myfunc: - push bp - mov bp,sp - sub sp,0x40 ; 64 bytes of local stack space - mov bx,[bp+4] ; first parameter to function - - ; some more code - - mov sp,bp ; undo "sub sp,0x40" above - pop bp - ret - - For a large-model function, you would replace `RET' by `RETF', and -look for the first parameter at `[BP+6]' instead of `[BP+4]'. Of -course, if one of the parameters is a pointer, then the offsets of -_subsequent_ parameters will change depending on the memory model as -well: far pointers take up four bytes on the stack when passed as a -parameter, whereas near pointers take up two. - - At the other end of the process, to call a C function from your -assembly code, you would do something like this: - - extern _printf - - ; and then, further down... - - push word [myint] ; one of my integer variables - push word mystring ; pointer into my data segment - call _printf - add sp,byte 4 ; `byte' saves space - - ; then those data items... - - segment _DATA - - myint dw 1234 - mystring db 'This number -> %d <- should be 1234',10,0 - - This piece of code is the small-model assembly equivalent of the C -code - - int myint = 1234; - printf("This number -> %d <- should be 1234\n", myint); - - In large model, the function-call code might look more like this. In -this example, it is assumed that `DS' already holds the segment base of -the segment `_DATA'. If not, you would have to initialize it first. - - push word [myint] - push word seg mystring ; Now push the segment, and... - push word mystring ; ... offset of "mystring" - call far _printf - add sp,byte 6 - - The integer value still takes up one word on the stack, since large -model does not affect the size of the `int' data type. The first -argument (pushed last) to `printf', however, is a data pointer, and -therefore has to contain a segment and offset part. The segment should -be stored second in memory, and therefore must be pushed first. (Of -course, `PUSH DS' would have been a shorter instruction than `PUSH WORD -SEG mystring', if `DS' was set up as the above example assumed.) Then -the actual call becomes a far call, since functions expect far calls in -large model; and `SP' has to be increased by 6 rather than 4 afterwards -to make up for the extra word of parameters. - - -File: nasm.info, Node: Section 8.4.4, Next: Section 8.4.5, Prev: Section 8.4.3, Up: Section 8.4 - -8.4.4. Accessing Data Items ---------------------------- - -To get at the contents of C variables, or to declare variables which C -can access, you need only declare the names as `GLOBAL' or `EXTERN'. -(Again, the names require leading underscores, as stated in *note -Section 8.4.1::.) Thus, a C variable declared as `int i' can be -accessed from assembler as - - extern _i - - mov ax,[_i] - - And to declare your own integer variable which C programs can access -as `extern int j', you do this (making sure you are assembling in the -`_DATA' segment, if necessary): - - global _j - - _j dw 0 - - To access a C array, you need to know the size of the components of -the array. For example, `int' variables are two bytes long, so if a C -program declares an array as `int a[10]', you can access `a[3]' by -coding `mov ax,[_a+6]'. (The byte offset 6 is obtained by multiplying -the desired array index, 3, by the size of the array element, 2.) The -sizes of the C base types in 16-bit compilers are: 1 for `char', 2 for -`short' and `int', 4 for `long' and `float', and 8 for `double'. - - To access a C data structure, you need to know the offset from the -base of the structure to the field you are interested in. You can -either do this by converting the C structure definition into a NASM -structure definition (using `STRUC'), or by calculating the one offset -and using just that. - - To do either of these, you should read your C compiler's manual to -find out how it organizes data structures. NASM gives no special -alignment to structure members in its own `STRUC' macro, so you have to -specify alignment yourself if the C compiler generates it. Typically, -you might find that a structure like - - struct { - char c; - int i; - } foo; - - might be four bytes long rather than three, since the `int' field -would be aligned to a two-byte boundary. However, this sort of feature -tends to be a configurable option in the C compiler, either using -command- line options or `#pragma' lines, so you have to find out how -your own compiler does it. - - -File: nasm.info, Node: Section 8.4.5, Next: Section 8.5, Prev: Section 8.4.4, Up: Section 8.4 - -8.4.5. `c16.mac': Helper Macros for the 16-bit C Interface ----------------------------------------------------------- - -Included in the NASM archives, in the `misc' directory, is a file -`c16.mac' of macros. It defines three macros: `proc', `arg' and -`endproc'. These are intended to be used for C-style procedure -definitions, and they automate a lot of the work involved in keeping -track of the calling convention. - - (An alternative, TASM compatible form of `arg' is also now built into -NASM's preprocessor. See *note Section 4.8:: for details.) - - An example of an assembly function using the macro set is given here: - - proc _nearproc - - %$i arg - %$j arg - mov ax,[bp + %$i] - mov bx,[bp + %$j] - add ax,[bx] - - endproc - - This defines `_nearproc' to be a procedure taking two arguments, the -first (`i') an integer and the second (`j') a pointer to an integer. It -returns `i + *j'. - - Note that the `arg' macro has an `EQU' as the first line of its -expansion, and since the label before the macro call gets prepended to -the first line of the expanded macro, the `EQU' works, defining `%$i' -to be an offset from `BP'. A context-local variable is used, local to -the context pushed by the `proc' macro and popped by the `endproc' -macro, so that the same argument name can be used in later procedures. -Of course, you don't _have_ to do that. - - The macro set produces code for near functions (tiny, small and -compact- model code) by default. You can have it generate far functions -(medium, large and huge-model code) by means of coding `%define -FARCODE'. This changes the kind of return instruction generated by -`endproc', and also changes the starting point for the argument -offsets. The macro set contains no intrinsic dependency on whether data -pointers are far or not. - - `arg' can take an optional parameter, giving the size of the -argument. If no size is given, 2 is assumed, since it is likely that -many function parameters will be of type `int'. - - The large-model equivalent of the above function would look like -this: - - %define FARCODE - - proc _farproc - - %$i arg - %$j arg 4 - mov ax,[bp + %$i] - mov bx,[bp + %$j] - mov es,[bp + %$j + 2] - add ax,[bx] - - endproc - - This makes use of the argument to the `arg' macro to define a -parameter of size 4, because `j' is now a far pointer. When we load -from `j', we must load a segment and an offset. - - -File: nasm.info, Node: Section 8.5, Next: Section 8.5.1, Prev: Section 8.4.5, Up: Chapter 8 - -8.5. Interfacing to Borland Pascal Programs -=========================================== - -Interfacing to Borland Pascal programs is similar in concept to -interfacing to 16-bit C programs. The differences are: - - * The leading underscore required for interfacing to C programs is - not required for Pascal. - - * The memory model is always large: functions are far, data pointers - are far, and no data item can be more than 64K long. (Actually, - some functions are near, but only those functions that are local - to a Pascal unit and never called from outside it. All assembly - functions that Pascal calls, and all Pascal functions that - assembly routines are able to call, are far.) However, all static - data declared in a Pascal program goes into the default data - segment, which is the one whose segment address will be in `DS' - when control is passed to your assembly code. The only things that - do not live in the default data segment are local variables (they - live in the stack segment) and dynamically allocated variables. - All data _pointers_, however, are far. - - * The function calling convention is different - described below. - - * Some data types, such as strings, are stored differently. - - * There are restrictions on the segment names you are allowed to use - - Borland Pascal will ignore code or data declared in a segment it - doesn't like the name of. The restrictions are described below. - -* Menu: - -* Section 8.5.1:: The Pascal Calling Convention -* Section 8.5.2:: Borland Pascal Segment Name Restrictions -* Section 8.5.3:: Using `c16.mac' With Pascal Programs - - -File: nasm.info, Node: Section 8.5.1, Next: Section 8.5.2, Prev: Section 8.5, Up: Section 8.5 - -8.5.1. The Pascal Calling Convention ------------------------------------- - -The 16-bit Pascal calling convention is as follows. In the following -description, the words _caller_ and _callee_ are used to denote the -function doing the calling and the function which gets called. - - * The caller pushes the function's parameters on the stack, one after - another, in normal order (left to right, so that the first argument - specified to the function is pushed first). - - * The caller then executes a far `CALL' instruction to pass control - to the callee. - - * The callee receives control, and typically (although this is not - actually necessary, in functions which do not need to access their - parameters) starts by saving the value of `SP' in `BP' so as to be - able to use `BP' as a base pointer to find its parameters on the - stack. However, the caller was probably doing this too, so part - of the calling convention states that `BP' must be preserved by - any function. Hence the callee, if it is going to set up `BP' as a - frame pointer, must push the previous value first. - - * The callee may then access its parameters relative to `BP'. The - word at `[BP]' holds the previous value of `BP' as it was pushed. - The next word, at `[BP+2]', holds the offset part of the return - address, and the next one at `[BP+4]' the segment part. The - parameters begin at `[BP+6]'. The rightmost parameter of the - function, since it was pushed last, is accessible at this offset - from `BP'; the others follow, at successively greater offsets. - - * The callee may also wish to decrease `SP' further, so as to - allocate space on the stack for local variables, which will then - be accessible at negative offsets from `BP'. - - * The callee, if it wishes to return a value to the caller, should - leave the value in `AL', `AX' or `DX:AX' depending on the size of - the value. Floating-point results are returned in `ST0'. Results - of type `Real' (Borland's own custom floating-point data type, not - handled directly by the FPU) are returned in `DX:BX:AX'. To return - a result of type `String', the caller pushes a pointer to a - temporary string before pushing the parameters, and the callee - places the returned string value at that location. The pointer is - not a parameter, and should not be removed from the stack by the - `RETF' instruction. - - * Once the callee has finished processing, it restores `SP' from - `BP' if it had allocated local stack space, then pops the previous - value of `BP', and returns via `RETF'. It uses the form of `RETF' - with an immediate parameter, giving the number of bytes taken up - by the parameters on the stack. This causes the parameters to be - removed from the stack as a side effect of the return instruction. - - * When the caller regains control from the callee, the function - parameters have already been removed from the stack, so it needs - to do nothing further. - - Thus, you would define a function in Pascal style, taking two -`Integer'-type parameters, in the following way: - - global myfunc - - myfunc: push bp - mov bp,sp - sub sp,0x40 ; 64 bytes of local stack space - mov bx,[bp+8] ; first parameter to function - mov bx,[bp+6] ; second parameter to function - - ; some more code - - mov sp,bp ; undo "sub sp,0x40" above - pop bp - retf 4 ; total size of params is 4 - - At the other end of the process, to call a Pascal function from your -assembly code, you would do something like this: - - extern SomeFunc - - ; and then, further down... - - push word seg mystring ; Now push the segment, and... - push word mystring ; ... offset of "mystring" - push word [myint] ; one of my variables - call far SomeFunc - - This is equivalent to the Pascal code - - procedure SomeFunc(String: PChar; Int: Integer); - SomeFunc(@mystring, myint); - - -File: nasm.info, Node: Section 8.5.2, Next: Section 8.5.3, Prev: Section 8.5.1, Up: Section 8.5 - -8.5.2. Borland Pascal Segment Name Restrictions ------------------------------------------------ - -Since Borland Pascal's internal unit file format is completely different -from `OBJ', it only makes a very sketchy job of actually reading and -understanding the various information contained in a real `OBJ' file -when it links that in. Therefore an object file intended to be linked -to a Pascal program must obey a number of restrictions: - - * Procedures and functions must be in a segment whose name is either - `CODE', `CSEG', or something ending in `_TEXT'. - - * initialized data must be in a segment whose name is either `CONST' - or something ending in `_DATA'. - - * Uninitialized data must be in a segment whose name is either - `DATA', `DSEG', or something ending in `_BSS'. - - * Any other segments in the object file are completely ignored. - `GROUP' directives and segment attributes are also ignored. - - -File: nasm.info, Node: Section 8.5.3, Next: Chapter 9, Prev: Section 8.5.2, Up: Section 8.5 - -8.5.3. Using `c16.mac' With Pascal Programs -------------------------------------------- - -The `c16.mac' macro package, described in *note Section 8.4.5::, can -also be used to simplify writing functions to be called from Pascal -programs, if you code `%define PASCAL'. This definition ensures that -functions are far (it implies `FARCODE'), and also causes procedure -return instructions to be generated with an operand. - - Defining `PASCAL' does not change the code which calculates the -argument offsets; you must declare your function's arguments in reverse -order. For example: - - %define PASCAL - - proc _pascalproc - - %$j arg 4 - %$i arg - mov ax,[bp + %$i] - mov bx,[bp + %$j] - mov es,[bp + %$j + 2] - add ax,[bx] - - endproc - - This defines the same routine, conceptually, as the example in *note -Section 8.4.5::: it defines a function taking two arguments, an integer -and a pointer to an integer, which returns the sum of the integer and -the contents of the pointer. The only difference between this code and -the large-model C version is that `PASCAL' is defined instead of -`FARCODE', and that the arguments are declared in reverse order. - - -File: nasm.info, Node: Chapter 9, Next: Section 9.1, Prev: Section 8.5.3, Up: Top - -Chapter 9: Writing 32-bit Code (Unix, Win32, DJGPP) -*************************************************** - -This chapter attempts to cover some of the common issues involved when -writing 32-bit code, to run under Win32 or Unix, or to be linked with C -code generated by a Unix-style C compiler such as DJGPP. It covers how -to write assembly code to interface with 32-bit C routines, and how to -write position-independent code for shared libraries. - - Almost all 32-bit code, and in particular all code running under -`Win32', `DJGPP' or any of the PC Unix variants, runs in _flat_ memory -model. This means that the segment registers and paging have already -been set up to give you the same 32-bit 4Gb address space no matter -what segment you work relative to, and that you should ignore all -segment registers completely. When writing flat-model application code, -you never need to use a segment override or modify any segment -register, and the code-section addresses you pass to `CALL' and `JMP' -live in the same address space as the data-section addresses you access -your variables by and the stack-section addresses you access local -variables and procedure parameters by. Every address is 32 bits long -and contains only an offset part. - -* Menu: - -* Section 9.1:: Interfacing to 32-bit C Programs -* Section 9.2:: Writing NetBSD/FreeBSD/OpenBSD and Linux/ELF Shared Libraries - - -File: nasm.info, Node: Section 9.1, Next: Section 9.1.1, Prev: Chapter 9, Up: Chapter 9 - -9.1. Interfacing to 32-bit C Programs -===================================== - -A lot of the discussion in *note Section 8.4::, about interfacing to -16-bit C programs, still applies when working in 32 bits. The absence -of memory models or segmentation worries simplifies things a lot. - -* Menu: - -* Section 9.1.1:: External Symbol Names -* Section 9.1.2:: Function Definitions and Function Calls -* Section 9.1.3:: Accessing Data Items -* Section 9.1.4:: `c32.mac': Helper Macros for the 32-bit C Interface - - -File: nasm.info, Node: Section 9.1.1, Next: Section 9.1.2, Prev: Section 9.1, Up: Section 9.1 - -9.1.1. External Symbol Names ----------------------------- - -Most 32-bit C compilers share the convention used by 16-bit compilers, -that the names of all global symbols (functions or data) they define -are formed by prefixing an underscore to the name as it appears in the -C program. However, not all of them do: the `ELF' specification states -that C symbols do _not_ have a leading underscore on their -assembly-language names. - - The older Linux `a.out' C compiler, all `Win32' compilers, `DJGPP', -and `NetBSD' and `FreeBSD', all use the leading underscore; for these -compilers, the macros `cextern' and `cglobal', as given in *note -Section 8.4.1::, will still work. For `ELF', though, the leading -underscore should not be used. - - See also *note Section 2.1.27::. - - -File: nasm.info, Node: Section 9.1.2, Next: Section 9.1.3, Prev: Section 9.1.1, Up: Section 9.1 - -9.1.2. Function Definitions and Function Calls ----------------------------------------------- - -The C calling convention in 32-bit programs is as follows. In the -following description, the words _caller_ and _callee_ are used to -denote the function doing the calling and the function which gets -called. - - * The caller pushes the function's parameters on the stack, one after - another, in reverse order (right to left, so that the first - argument specified to the function is pushed last). - - * The caller then executes a near `CALL' instruction to pass control - to the callee. - - * The callee receives control, and typically (although this is not - actually necessary, in functions which do not need to access their - parameters) starts by saving the value of `ESP' in `EBP' so as to - be able to use `EBP' as a base pointer to find its parameters on - the stack. However, the caller was probably doing this too, so - part of the calling convention states that `EBP' must be preserved - by any C function. Hence the callee, if it is going to set up - `EBP' as a frame pointer, must push the previous value first. - - * The callee may then access its parameters relative to `EBP'. The - doubleword at `[EBP]' holds the previous value of `EBP' as it was - pushed; the next doubleword, at `[EBP+4]', holds the return - address, pushed implicitly by `CALL'. The parameters start after - that, at `[EBP+8]'. The leftmost parameter of the function, since - it was pushed last, is accessible at this offset from `EBP'; the - others follow, at successively greater offsets. Thus, in a - function such as `printf' which takes a variable number of - parameters, the pushing of the parameters in reverse order means - that the function knows where to find its first parameter, which - tells it the number and type of the remaining ones. - - * The callee may also wish to decrease `ESP' further, so as to - allocate space on the stack for local variables, which will then - be accessible at negative offsets from `EBP'. - - * The callee, if it wishes to return a value to the caller, should - leave the value in `AL', `AX' or `EAX' depending on the size of the - value. Floating-point results are typically returned in `ST0'. - - * Once the callee has finished processing, it restores `ESP' from - `EBP' if it had allocated local stack space, then pops the previous - value of `EBP', and returns via `RET' (equivalently, `RETN'). - - * When the caller regains control from the callee, the function - parameters are still on the stack, so it typically adds an - immediate constant to `ESP' to remove them (instead of executing a - number of slow `POP' instructions). Thus, if a function is - accidentally called with the wrong number of parameters due to a - prototype mismatch, the stack will still be returned to a sensible - state since the caller, which _knows_ how many parameters it - pushed, does the removing. - - There is an alternative calling convention used by Win32 programs for -Windows API calls, and also for functions called _by_ the Windows API -such as window procedures: they follow what Microsoft calls the -`__stdcall' convention. This is slightly closer to the Pascal -convention, in that the callee clears the stack by passing a parameter -to the `RET' instruction. However, the parameters are still pushed in -right-to-left order. - - Thus, you would define a function in C style in the following way: - - global _myfunc - - _myfunc: - push ebp - mov ebp,esp - sub esp,0x40 ; 64 bytes of local stack space - mov ebx,[ebp+8] ; first parameter to function - - ; some more code - - leave ; mov esp,ebp / pop ebp - ret - - At the other end of the process, to call a C function from your -assembly code, you would do something like this: - - extern _printf - - ; and then, further down... - - push dword [myint] ; one of my integer variables - push dword mystring ; pointer into my data segment - call _printf - add esp,byte 8 ; `byte' saves space - - ; then those data items... - - segment _DATA - - myint dd 1234 - mystring db 'This number -> %d <- should be 1234',10,0 - - This piece of code is the assembly equivalent of the C code - - int myint = 1234; - printf("This number -> %d <- should be 1234\n", myint); - diff --git a/doc/info/nasm.info-2 b/doc/info/nasm.info-2 deleted file mode 100644 index dd66a29..0000000 --- a/doc/info/nasm.info-2 +++ /dev/null @@ -1,6521 +0,0 @@ -This is nasm.info, produced by makeinfo version 4.13 from nasmdoc.texi. - -INFO-DIR-SECTION Programming -START-INFO-DIR-ENTRY -* NASM: (nasm). The Netwide Assembler for x86. -END-INFO-DIR-ENTRY - - This file documents NASM, the Netwide Assembler: an assembler -targetting the Intel x86 series of processors, with portable source. - - Copyright 1996-2009 The NASM Development Team - - This document is redistributable under the license given in the file -"COPYING" distributed in the NASM archive. - - -File: nasm.info, Node: Section 9.1.3, Next: Section 9.1.4, Prev: Section 9.1.2, Up: Section 9.1 - -9.1.3. Accessing Data Items ---------------------------- - -To get at the contents of C variables, or to declare variables which C -can access, you need only declare the names as `GLOBAL' or `EXTERN'. -(Again, the names require leading underscores, as stated in *note -Section 9.1.1::.) Thus, a C variable declared as `int i' can be -accessed from assembler as - - extern _i - mov eax,[_i] - - And to declare your own integer variable which C programs can access -as `extern int j', you do this (making sure you are assembling in the -`_DATA' segment, if necessary): - - global _j - _j dd 0 - - To access a C array, you need to know the size of the components of -the array. For example, `int' variables are four bytes long, so if a C -program declares an array as `int a[10]', you can access `a[3]' by -coding `mov ax,[_a+12]'. (The byte offset 12 is obtained by multiplying -the desired array index, 3, by the size of the array element, 4.) The -sizes of the C base types in 32-bit compilers are: 1 for `char', 2 for -`short', 4 for `int', `long' and `float', and 8 for `double'. Pointers, -being 32-bit addresses, are also 4 bytes long. - - To access a C data structure, you need to know the offset from the -base of the structure to the field you are interested in. You can -either do this by converting the C structure definition into a NASM -structure definition (using `STRUC'), or by calculating the one offset -and using just that. - - To do either of these, you should read your C compiler's manual to -find out how it organizes data structures. NASM gives no special -alignment to structure members in its own `STRUC' macro, so you have to -specify alignment yourself if the C compiler generates it. Typically, -you might find that a structure like - - struct { - char c; - int i; - } foo; - - might be eight bytes long rather than five, since the `int' field -would be aligned to a four-byte boundary. However, this sort of feature -is sometimes a configurable option in the C compiler, either using -command- line options or `#pragma' lines, so you have to find out how -your own compiler does it. - - -File: nasm.info, Node: Section 9.1.4, Next: Section 9.2, Prev: Section 9.1.3, Up: Section 9.1 - -9.1.4. `c32.mac': Helper Macros for the 32-bit C Interface ----------------------------------------------------------- - -Included in the NASM archives, in the `misc' directory, is a file -`c32.mac' of macros. It defines three macros: `proc', `arg' and -`endproc'. These are intended to be used for C-style procedure -definitions, and they automate a lot of the work involved in keeping -track of the calling convention. - - An example of an assembly function using the macro set is given here: - - proc _proc32 - - %$i arg - %$j arg - mov eax,[ebp + %$i] - mov ebx,[ebp + %$j] - add eax,[ebx] - - endproc - - This defines `_proc32' to be a procedure taking two arguments, the -first (`i') an integer and the second (`j') a pointer to an integer. It -returns `i + *j'. - - Note that the `arg' macro has an `EQU' as the first line of its -expansion, and since the label before the macro call gets prepended to -the first line of the expanded macro, the `EQU' works, defining `%$i' -to be an offset from `BP'. A context-local variable is used, local to -the context pushed by the `proc' macro and popped by the `endproc' -macro, so that the same argument name can be used in later procedures. -Of course, you don't _have_ to do that. - - `arg' can take an optional parameter, giving the size of the -argument. If no size is given, 4 is assumed, since it is likely that -many function parameters will be of type `int' or pointers. - - -File: nasm.info, Node: Section 9.2, Next: Section 9.2.1, Prev: Section 9.1.4, Up: Chapter 9 - -9.2. Writing NetBSD/FreeBSD/OpenBSD and Linux/ELF Shared Libraries -================================================================== - -`ELF' replaced the older `a.out' object file format under Linux because -it contains support for position-independent code (PIC), which makes -writing shared libraries much easier. NASM supports the `ELF' -position-independent code features, so you can write Linux `ELF' shared -libraries in NASM. - - NetBSD, and its close cousins FreeBSD and OpenBSD, take a different -approach by hacking PIC support into the `a.out' format. NASM supports -this as the `aoutb' output format, so you can write BSD shared -libraries in NASM too. - - The operating system loads a PIC shared library by memory-mapping the -library file at an arbitrarily chosen point in the address space of the -running process. The contents of the library's code section must -therefore not depend on where it is loaded in memory. - - Therefore, you cannot get at your variables by writing code like -this: - - mov eax,[myvar] ; WRONG - - Instead, the linker provides an area of memory called the _global -offset table_, or GOT; the GOT is situated at a constant distance from -your library's code, so if you can find out where your library is -loaded (which is typically done using a `CALL' and `POP' combination), -you can obtain the address of the GOT, and you can then load the -addresses of your variables out of linker-generated entries in the GOT. - - The _data_ section of a PIC shared library does not have these -restrictions: since the data section is writable, it has to be copied -into memory anyway rather than just paged in from the library file, so -as long as it's being copied it can be relocated too. So you can put -ordinary types of relocation in the data section without too much worry -(but see *note Section 9.2.4:: for a caveat). - -* Menu: - -* Section 9.2.1:: Obtaining the Address of the GOT -* Section 9.2.2:: Finding Your Local Data Items -* Section 9.2.3:: Finding External and Common Data Items -* Section 9.2.4:: Exporting Symbols to the Library User -* Section 9.2.5:: Calling Procedures Outside the Library -* Section 9.2.6:: Generating the Library File - - -File: nasm.info, Node: Section 9.2.1, Next: Section 9.2.2, Prev: Section 9.2, Up: Section 9.2 - -9.2.1. Obtaining the Address of the GOT ---------------------------------------- - -Each code module in your shared library should define the GOT as an -external symbol: - - extern _GLOBAL_OFFSET_TABLE_ ; in ELF - extern __GLOBAL_OFFSET_TABLE_ ; in BSD a.out - - At the beginning of any function in your shared library which plans -to access your data or BSS sections, you must first calculate the -address of the GOT. This is typically done by writing the function in -this form: - - func: push ebp - mov ebp,esp - push ebx - call .get_GOT - .get_GOT: - pop ebx - add ebx,_GLOBAL_OFFSET_TABLE_+$$-.get_GOT wrt ..gotpc - - ; the function body comes here - - mov ebx,[ebp-4] - mov esp,ebp - pop ebp - ret - - (For BSD, again, the symbol `_GLOBAL_OFFSET_TABLE' requires a second -leading underscore.) - - The first two lines of this function are simply the standard C -prologue to set up a stack frame, and the last three lines are standard -C function epilogue. The third line, and the fourth to last line, save -and restore the `EBX' register, because PIC shared libraries use this -register to store the address of the GOT. - - The interesting bit is the `CALL' instruction and the following two -lines. The `CALL' and `POP' combination obtains the address of the -label `.get_GOT', without having to know in advance where the program -was loaded (since the `CALL' instruction is encoded relative to the -current position). The `ADD' instruction makes use of one of the -special PIC relocation types: GOTPC relocation. With the `WRT ..gotpc' -qualifier specified, the symbol referenced (here -`_GLOBAL_OFFSET_TABLE_', the special symbol assigned to the GOT) is -given as an offset from the beginning of the section. (Actually, `ELF' -encodes it as the offset from the operand field of the `ADD' -instruction, but NASM simplifies this deliberately, so you do things the -same way for both `ELF' and `BSD'.) So the instruction then _adds_ the -beginning of the section, to get the real address of the GOT, and -subtracts the value of `.get_GOT' which it knows is in `EBX'. -Therefore, by the time that instruction has finished, `EBX' contains -the address of the GOT. - - If you didn't follow that, don't worry: it's never necessary to -obtain the address of the GOT by any other means, so you can put those -three instructions into a macro and safely ignore them: - - %macro get_GOT 0 - - call %%getgot - %%getgot: - pop ebx - add ebx,_GLOBAL_OFFSET_TABLE_+$$-%%getgot wrt ..gotpc - - %endmacro - - -File: nasm.info, Node: Section 9.2.2, Next: Section 9.2.3, Prev: Section 9.2.1, Up: Section 9.2 - -9.2.2. Finding Your Local Data Items ------------------------------------- - -Having got the GOT, you can then use it to obtain the addresses of your -data items. Most variables will reside in the sections you have -declared; they can be accessed using the `..gotoff' special `WRT' type. -The way this works is like this: - - lea eax,[ebx+myvar wrt ..gotoff] - - The expression `myvar wrt ..gotoff' is calculated, when the shared -library is linked, to be the offset to the local variable `myvar' from -the beginning of the GOT. Therefore, adding it to `EBX' as above will -place the real address of `myvar' in `EAX'. - - If you declare variables as `GLOBAL' without specifying a size for -them, they are shared between code modules in the library, but do not -get exported from the library to the program that loaded it. They will -still be in your ordinary data and BSS sections, so you can access them -in the same way as local variables, using the above `..gotoff' -mechanism. - - Note that due to a peculiarity of the way BSD `a.out' format handles -this relocation type, there must be at least one non-local symbol in the -same section as the address you're trying to access. - - -File: nasm.info, Node: Section 9.2.3, Next: Section 9.2.4, Prev: Section 9.2.2, Up: Section 9.2 - -9.2.3. Finding External and Common Data Items ---------------------------------------------- - -If your library needs to get at an external variable (external to the -_library_, not just to one of the modules within it), you must use the -`..got' type to get at it. The `..got' type, instead of giving you the -offset from the GOT base to the variable, gives you the offset from the -GOT base to a GOT _entry_ containing the address of the variable. The -linker will set up this GOT entry when it builds the library, and the -dynamic linker will place the correct address in it at load time. So to -obtain the address of an external variable `extvar' in `EAX', you would -code - - mov eax,[ebx+extvar wrt ..got] - - This loads the address of `extvar' out of an entry in the GOT. The -linker, when it builds the shared library, collects together every -relocation of type `..got', and builds the GOT so as to ensure it has -every necessary entry present. - - Common variables must also be accessed in this way. - - -File: nasm.info, Node: Section 9.2.4, Next: Section 9.2.5, Prev: Section 9.2.3, Up: Section 9.2 - -9.2.4. Exporting Symbols to the Library User --------------------------------------------- - -If you want to export symbols to the user of the library, you have to -declare whether they are functions or data, and if they are data, you -have to give the size of the data item. This is because the dynamic -linker has to build procedure linkage table entries for any exported -functions, and also moves exported data items away from the library's -data section in which they were declared. - - So to export a function to users of the library, you must use - - global func:function ; declare it as a function - - func: push ebp - - ; etc. - - And to export a data item such as an array, you would have to code - - global array:data array.end-array ; give the size too - - array: resd 128 - .end: - - Be careful: If you export a variable to the library user, by -declaring it as `GLOBAL' and supplying a size, the variable will end up -living in the data section of the main program, rather than in your -library's data section, where you declared it. So you will have to -access your own global variable with the `..got' mechanism rather than -`..gotoff', as if it were external (which, effectively, it has become). - - Equally, if you need to store the address of an exported global in -one of your data sections, you can't do it by means of the standard -sort of code: - - dataptr: dd global_data_item ; WRONG - - NASM will interpret this code as an ordinary relocation, in which -`global_data_item' is merely an offset from the beginning of the -`.data' section (or whatever); so this reference will end up pointing -at your data section instead of at the exported global which resides -elsewhere. - - Instead of the above code, then, you must write - - dataptr: dd global_data_item wrt ..sym - - which makes use of the special `WRT' type `..sym' to instruct NASM -to search the symbol table for a particular symbol at that address, -rather than just relocating by section base. - - Either method will work for functions: referring to one of your -functions by means of - - funcptr: dd my_function - - will give the user the address of the code you wrote, whereas - - funcptr: dd my_function wrt .sym - - will give the address of the procedure linkage table for the -function, which is where the calling program will _believe_ the -function lives. Either address is a valid way to call the function. - - -File: nasm.info, Node: Section 9.2.5, Next: Section 9.2.6, Prev: Section 9.2.4, Up: Section 9.2 - -9.2.5. Calling Procedures Outside the Library ---------------------------------------------- - -Calling procedures outside your shared library has to be done by means -of a _procedure linkage table_, or PLT. The PLT is placed at a known -offset from where the library is loaded, so the library code can make -calls to the PLT in a position-independent way. Within the PLT there is -code to jump to offsets contained in the GOT, so function calls to -other shared libraries or to routines in the main program can be -transparently passed off to their real destinations. - - To call an external routine, you must use another special PIC -relocation type, `WRT ..plt'. This is much easier than the GOT-based -ones: you simply replace calls such as `CALL printf' with the -PLT-relative version `CALL printf WRT ..plt'. - - -File: nasm.info, Node: Section 9.2.6, Next: Chapter 10, Prev: Section 9.2.5, Up: Section 9.2 - -9.2.6. Generating the Library File ----------------------------------- - -Having written some code modules and assembled them to `.o' files, you -then generate your shared library with a command such as - - ld -shared -o library.so module1.o module2.o # for ELF - ld -Bshareable -o library.so module1.o module2.o # for BSD - - For ELF, if your shared library is going to reside in system -directories such as `/usr/lib' or `/lib', it is usually worth using the -`-soname' flag to the linker, to store the final library file name, -with a version number, into the library: - - ld -shared -soname library.so.1 -o library.so.1.2 *.o - - You would then copy `library.so.1.2' into the library directory, and -create `library.so.1' as a symbolic link to it. - - -File: nasm.info, Node: Chapter 10, Next: Section 10.1, Prev: Section 9.2.6, Up: Top - -Chapter 10: Mixing 16 and 32 Bit Code -************************************* - -This chapter tries to cover some of the issues, largely related to -unusual forms of addressing and jump instructions, encountered when -writing operating system code such as protected-mode initialisation -routines, which require code that operates in mixed segment sizes, such -as code in a 16-bit segment trying to modify data in a 32-bit one, or -jumps between different- size segments. - -* Menu: - -* Section 10.1:: Mixed-Size Jumps -* Section 10.2:: Addressing Between Different-Size Segments -* Section 10.3:: Other Mixed-Size Instructions - - -File: nasm.info, Node: Section 10.1, Next: Section 10.2, Prev: Chapter 10, Up: Chapter 10 - -10.1. Mixed-Size Jumps -====================== - -The most common form of mixed-size instruction is the one used when -writing a 32-bit OS: having done your setup in 16-bit mode, such as -loading the kernel, you then have to boot it by switching into -protected mode and jumping to the 32-bit kernel start address. In a -fully 32-bit OS, this tends to be the _only_ mixed-size instruction you -need, since everything before it can be done in pure 16-bit code, and -everything after it can be pure 32-bit. - - This jump must specify a 48-bit far address, since the target -segment is a 32-bit one. However, it must be assembled in a 16-bit -segment, so just coding, for example, - - jmp 0x1234:0x56789ABC ; wrong! - - will not work, since the offset part of the address will be -truncated to `0x9ABC' and the jump will be an ordinary 16-bit far one. - - The Linux kernel setup code gets round the inability of `as86' to -generate the required instruction by coding it manually, using `DB' -instructions. NASM can go one better than that, by actually generating -the right instruction itself. Here's how to do it right: - - jmp dword 0x1234:0x56789ABC ; right - - The `DWORD' prefix (strictly speaking, it should come _after_ the -colon, since it is declaring the _offset_ field to be a doubleword; but -NASM will accept either form, since both are unambiguous) forces the -offset part to be treated as far, in the assumption that you are -deliberately writing a jump from a 16-bit segment to a 32-bit one. - - You can do the reverse operation, jumping from a 32-bit segment to a -16-bit one, by means of the `WORD' prefix: - - jmp word 0x8765:0x4321 ; 32 to 16 bit - - If the `WORD' prefix is specified in 16-bit mode, or the `DWORD' -prefix in 32-bit mode, they will be ignored, since each is explicitly -forcing NASM into a mode it was in anyway. - - -File: nasm.info, Node: Section 10.2, Next: Section 10.3, Prev: Section 10.1, Up: Chapter 10 - -10.2. Addressing Between Different-Size Segments -================================================ - -If your OS is mixed 16 and 32-bit, or if you are writing a DOS extender, -you are likely to have to deal with some 16-bit segments and some 32-bit -ones. At some point, you will probably end up writing code in a 16-bit -segment which has to access data in a 32-bit segment, or vice versa. - - If the data you are trying to access in a 32-bit segment lies within -the first 64K of the segment, you may be able to get away with using an -ordinary 16-bit addressing operation for the purpose; but sooner or -later, you will want to do 32-bit addressing from 16-bit mode. - - The easiest way to do this is to make sure you use a register for the -address, since any effective address containing a 32-bit register is -forced to be a 32-bit address. So you can do - - mov eax,offset_into_32_bit_segment_specified_by_fs - mov dword [fs:eax],0x11223344 - - This is fine, but slightly cumbersome (since it wastes an -instruction and a register) if you already know the precise offset you -are aiming at. The x86 architecture does allow 32-bit effective -addresses to specify nothing but a 4-byte offset, so why shouldn't NASM -be able to generate the best instruction for the purpose? - - It can. As in *note Section 10.1::, you need only prefix the address -with the `DWORD' keyword, and it will be forced to be a 32-bit address: - - mov dword [fs:dword my_offset],0x11223344 - - Also as in *note Section 10.1::, NASM is not fussy about whether the -`DWORD' prefix comes before or after the segment override, so arguably -a nicer-looking way to code the above instruction is - - mov dword [dword fs:my_offset],0x11223344 - - Don't confuse the `DWORD' prefix _outside_ the square brackets, -which controls the size of the data stored at the address, with the one -`inside' the square brackets which controls the length of the address -itself. The two can quite easily be different: - - mov word [dword 0x12345678],0x9ABC - - This moves 16 bits of data to an address specified by a 32-bit -offset. - - You can also specify `WORD' or `DWORD' prefixes along with the `FAR' -prefix to indirect far jumps or calls. For example: - - call dword far [fs:word 0x4321] - - This instruction contains an address specified by a 16-bit offset; -it loads a 48-bit far pointer from that (16-bit segment and 32-bit -offset), and calls that address. - - -File: nasm.info, Node: Section 10.3, Next: Chapter 11, Prev: Section 10.2, Up: Chapter 10 - -10.3. Other Mixed-Size Instructions -=================================== - -The other way you might want to access data might be using the string -instructions (`LODSx', `STOSx' and so on) or the `XLATB' instruction. -These instructions, since they take no parameters, might seem to have -no easy way to make them perform 32-bit addressing when assembled in a -16-bit segment. - - This is the purpose of NASM's `a16', `a32' and `a64' prefixes. If -you are coding `LODSB' in a 16-bit segment but it is supposed to be -accessing a string in a 32-bit segment, you should load the desired -address into `ESI' and then code - - a32 lodsb - - The prefix forces the addressing size to 32 bits, meaning that -`LODSB' loads from `[DS:ESI]' instead of `[DS:SI]'. To access a string -in a 16-bit segment when coding in a 32-bit one, the corresponding `a16' -prefix can be used. - - The `a16', `a32' and `a64' prefixes can be applied to any -instruction in NASM's instruction table, but most of them can generate -all the useful forms without them. The prefixes are necessary only for -instructions with implicit addressing: `CMPSx', `SCASx', `LODSx', -`STOSx', `MOVSx', `INSx', `OUTSx', and `XLATB'. Also, the various push -and pop instructions (`PUSHA' and `POPF' as well as the more usual -`PUSH' and `POP') can accept `a16', `a32' or `a64' prefixes to force a -particular one of `SP', `ESP' or `RSP' to be used as a stack pointer, -in case the stack segment in use is a different size from the code -segment. - - `PUSH' and `POP', when applied to segment registers in 32-bit mode, -also have the slightly odd behaviour that they push and pop 4 bytes at -a time, of which the top two are ignored and the bottom two give the -value of the segment register being manipulated. To force the 16-bit -behaviour of segment-register push and pop instructions, you can use the -operand-size prefix `o16': - - o16 push ss - o16 push ds - - This code saves a doubleword of stack space by fitting two segment -registers into the space which would normally be consumed by pushing -one. - - (You can also use the `o32' prefix to force the 32-bit behaviour when -in 16-bit mode, but this seems less useful.) - - -File: nasm.info, Node: Chapter 11, Next: Section 11.1, Prev: Section 10.3, Up: Top - -Chapter 11: Writing 64-bit Code (Unix, Win64) -********************************************* - -This chapter attempts to cover some of the common issues involved when -writing 64-bit code, to run under Win64 or Unix. It covers how to write -assembly code to interface with 64-bit C routines, and how to write -position-independent code for shared libraries. - - All 64-bit code uses a flat memory model, since segmentation is not -available in 64-bit mode. The one exception is the `FS' and `GS' -registers, which still add their bases. - - Position independence in 64-bit mode is significantly simpler, since -the processor supports `RIP'-relative addressing directly; see the -`REL' keyword (*note Section 3.3::). On most 64-bit platforms, it is -probably desirable to make that the default, using the directive -`DEFAULT REL' (*note Section 6.2::). - - 64-bit programming is relatively similar to 32-bit programming, but -of course pointers are 64 bits long; additionally, all existing -platforms pass arguments in registers rather than on the stack. -Furthermore, 64-bit platforms use SSE2 by default for floating point. -Please see the ABI documentation for your platform. - - 64-bit platforms differ in the sizes of the fundamental datatypes, -not just from 32-bit platforms but from each other. If a specific size -data type is desired, it is probably best to use the types defined in -the Standard C header `<inttypes.h>'. - - In 64-bit mode, the default instruction size is still 32 bits. When -loading a value into a 32-bit register (but not an 8- or 16-bit -register), the upper 32 bits of the corresponding 64-bit register are -set to zero. - -* Menu: - -* Section 11.1:: Register Names in 64-bit Mode -* Section 11.2:: Immediates and Displacements in 64-bit Mode -* Section 11.3:: Interfacing to 64-bit C Programs (Unix) -* Section 11.4:: Interfacing to 64-bit C Programs (Win64) - - -File: nasm.info, Node: Section 11.1, Next: Section 11.2, Prev: Chapter 11, Up: Chapter 11 - -11.1. Register Names in 64-bit Mode -=================================== - -NASM uses the following names for general-purpose registers in 64-bit -mode, for 8-, 16-, 32- and 64-bit references, respecitively: - - AL/AH, CL/CH, DL/DH, BL/BH, SPL, BPL, SIL, DIL, R8B-R15B - AX, CX, DX, BX, SP, BP, SI, DI, R8W-R15W - EAX, ECX, EDX, EBX, ESP, EBP, ESI, EDI, R8D-R15D - RAX, RCX, RDX, RBX, RSP, RBP, RSI, RDI, R8-R15 - - This is consistent with the AMD documentation and most other -assemblers. The Intel documentation, however, uses the names -`R8L-R15L' for 8-bit references to the higher registers. It is possible -to use those names by definiting them as macros; similarly, if one -wants to use numeric names for the low 8 registers, define them as -macros. The standard macro package `altreg' (see *note Section 5.1::) -can be used for this purpose. - - -File: nasm.info, Node: Section 11.2, Next: Section 11.3, Prev: Section 11.1, Up: Chapter 11 - -11.2. Immediates and Displacements in 64-bit Mode -================================================= - -In 64-bit mode, immediates and displacements are generally only 32 bits -wide. NASM will therefore truncate most displacements and immediates to -32 bits. - - The only instruction which takes a full 64-bit immediate is: - - MOV reg64,imm64 - - NASM will produce this instruction whenever the programmer uses `MOV' -with an immediate into a 64-bit register. If this is not desirable, -simply specify the equivalent 32-bit register, which will be -automatically zero- extended by the processor, or specify the immediate -as `DWORD': - - mov rax,foo ; 64-bit immediate - mov rax,qword foo ; (identical) - mov eax,foo ; 32-bit immediate, zero-extended - mov rax,dword foo ; 32-bit immediate, sign-extended - - The length of these instructions are 10, 5 and 7 bytes, respectively. - - The only instructions which take a full 64-bit _displacement_ is -loading or storing, using `MOV', `AL', `AX', `EAX' or `RAX' (but no -other registers) to an absolute 64-bit address. Since this is a -relatively rarely used instruction (64-bit code generally uses relative -addressing), the programmer has to explicitly declare the displacement -size as `QWORD': - - default abs - - mov eax,[foo] ; 32-bit absolute disp, sign-extended - mov eax,[a32 foo] ; 32-bit absolute disp, zero-extended - mov eax,[qword foo] ; 64-bit absolute disp - - default rel - - mov eax,[foo] ; 32-bit relative disp - mov eax,[a32 foo] ; d:o, address truncated to 32 bits(!) - mov eax,[qword foo] ; error - mov eax,[abs qword foo] ; 64-bit absolute disp - - A sign-extended absolute displacement can access from -2 GB to +2 -GB; a zero-extended absolute displacement can access from 0 to 4 GB. - - -File: nasm.info, Node: Section 11.3, Next: Section 11.4, Prev: Section 11.2, Up: Chapter 11 - -11.3. Interfacing to 64-bit C Programs (Unix) -============================================= - -On Unix, the 64-bit ABI is defined by the document: - - `http://www.x86-64.org/documentation/abi.pdf' - - Although written for AT&T-syntax assembly, the concepts apply -equally well for NASM-style assembly. What follows is a simplified -summary. - - The first six integer arguments (from the left) are passed in `RDI', -`RSI', `RDX', `RCX', `R8', and `R9', in that order. Additional integer -arguments are passed on the stack. These registers, plus `RAX', `R10' -and `R11' are destroyed by function calls, and thus are available for -use by the function without saving. - - Integer return values are passed in `RAX' and `RDX', in that order. - - Floating point is done using SSE registers, except for `long double'. -Floating-point arguments are passed in `XMM0' to `XMM7'; return is -`XMM0' and `XMM1'. `long double' are passed on the stack, and returned -in `ST0' and `ST1'. - - All SSE and x87 registers are destroyed by function calls. - - On 64-bit Unix, `long' is 64 bits. - - Integer and SSE register arguments are counted separately, so for -the case of - - void foo(long a, double b, int c) - - `a' is passed in `RDI', `b' in `XMM0', and `c' in `ESI'. - - -File: nasm.info, Node: Section 11.4, Next: Chapter 12, Prev: Section 11.3, Up: Chapter 11 - -11.4. Interfacing to 64-bit C Programs (Win64) -============================================== - -The Win64 ABI is described at: - - `http://msdn2.microsoft.com/en-gb/library/ms794533.aspx' - - What follows is a simplified summary. - - The first four integer arguments are passed in `RCX', `RDX', `R8' -and `R9', in that order. Additional integer arguments are passed on the -stack. These registers, plus `RAX', `R10' and `R11' are destroyed by -function calls, and thus are available for use by the function without -saving. - - Integer return values are passed in `RAX' only. - - Floating point is done using SSE registers, except for `long double'. -Floating-point arguments are passed in `XMM0' to `XMM3'; return is -`XMM0' only. - - On Win64, `long' is 32 bits; `long long' or `_int64' is 64 bits. - - Integer and SSE register arguments are counted together, so for the -case of - - void foo(long long a, double b, int c) - - `a' is passed in `RCX', `b' in `XMM1', and `c' in `R8D'. - - -File: nasm.info, Node: Chapter 12, Next: Section 12.1, Prev: Section 11.4, Up: Top - -Chapter 12: Troubleshooting -*************************** - -This chapter describes some of the common problems that users have been -known to encounter with NASM, and answers them. It also gives -instructions for reporting bugs in NASM if you find a difficulty that -isn't listed here. - -* Menu: - -* Section 12.1:: Common Problems -* Section 12.2:: Bugs - - -File: nasm.info, Node: Section 12.1, Next: Section 12.1.1, Prev: Chapter 12, Up: Chapter 12 - -12.1. Common Problems -===================== - -* Menu: - -* Section 12.1.1:: NASM Generates Inefficient Code -* Section 12.1.2:: My Jumps are Out of Range -* Section 12.1.3:: `ORG' Doesn't Work -* Section 12.1.4:: `TIMES' Doesn't Work - - -File: nasm.info, Node: Section 12.1.1, Next: Section 12.1.2, Prev: Section 12.1, Up: Section 12.1 - -12.1.1. NASM Generates Inefficient Code ---------------------------------------- - -We sometimes get `bug' reports about NASM generating inefficient, or -even `wrong', code on instructions such as `ADD ESP,8'. This is a -deliberate design feature, connected to predictability of output: NASM, -on seeing `ADD ESP,8', will generate the form of the instruction which -leaves room for a 32-bit offset. You need to code `ADD ESP,BYTE 8' if -you want the space-efficient form of the instruction. This isn't a bug, -it's user error: if you prefer to have NASM produce the more efficient -code automatically enable optimization with the `-O' option (see *note -Section 2.1.22::). - - -File: nasm.info, Node: Section 12.1.2, Next: Section 12.1.3, Prev: Section 12.1.1, Up: Section 12.1 - -12.1.2. My Jumps are Out of Range ---------------------------------- - -Similarly, people complain that when they issue conditional jumps (which -are `SHORT' by default) that try to jump too far, NASM reports `short -jump out of range' instead of making the jumps longer. - - This, again, is partly a predictability issue, but in fact has a more -practical reason as well. NASM has no means of being told what type of -processor the code it is generating will be run on; so it cannot decide -for itself that it should generate `Jcc NEAR' type instructions, because -it doesn't know that it's working for a 386 or above. Alternatively, it -could replace the out-of-range short `JNE' instruction with a very -short `JE' instruction that jumps over a `JMP NEAR'; this is a sensible -solution for processors below a 386, but hardly efficient on processors -which have good branch prediction _and_ could have used `JNE NEAR' -instead. So, once again, it's up to the user, not the assembler, to -decide what instructions should be generated. See *note Section -2.1.22::. - - -File: nasm.info, Node: Section 12.1.3, Next: Section 12.1.4, Prev: Section 12.1.2, Up: Section 12.1 - -12.1.3. `ORG' Doesn't Work --------------------------- - -People writing boot sector programs in the `bin' format often complain -that `ORG' doesn't work the way they'd like: in order to place the -`0xAA55' signature word at the end of a 512-byte boot sector, people -who are used to MASM tend to code - - ORG 0 - - ; some boot sector code - - ORG 510 - DW 0xAA55 - - This is not the intended use of the `ORG' directive in NASM, and will -not work. The correct way to solve this problem in NASM is to use the -`TIMES' directive, like this: - - ORG 0 - - ; some boot sector code - - TIMES 510-($-$$) DB 0 - DW 0xAA55 - - The `TIMES' directive will insert exactly enough zero bytes into the -output to move the assembly point up to 510. This method also has the -advantage that if you accidentally fill your boot sector too full, NASM -will catch the problem at assembly time and report it, so you won't end -up with a boot sector that you have to disassemble to find out what's -wrong with it. - - -File: nasm.info, Node: Section 12.1.4, Next: Section 12.2, Prev: Section 12.1.3, Up: Section 12.1 - -12.1.4. `TIMES' Doesn't Work ----------------------------- - -The other common problem with the above code is people who write the -`TIMES' line as - - TIMES 510-$ DB 0 - - by reasoning that `$' should be a pure number, just like 510, so the -difference between them is also a pure number and can happily be fed to -`TIMES'. - - NASM is a _modular_ assembler: the various component parts are -designed to be easily separable for re-use, so they don't exchange -information unnecessarily. In consequence, the `bin' output format, -even though it has been told by the `ORG' directive that the `.text' -section should start at 0, does not pass that information back to the -expression evaluator. So from the evaluator's point of view, `$' isn't -a pure number: it's an offset from a section base. Therefore the -difference between `$' and 510 is also not a pure number, but involves -a section base. Values involving section bases cannot be passed as -arguments to `TIMES'. - - The solution, as in the previous section, is to code the `TIMES' line -in the form - - TIMES 510-($-$$) DB 0 - - in which `$' and `$$' are offsets from the same section base, and so -their difference is a pure number. This will solve the problem and -generate sensible code. - - -File: nasm.info, Node: Section 12.2, Next: Appendix A, Prev: Section 12.1.4, Up: Chapter 12 - -12.2. Bugs -========== - -We have never yet released a version of NASM with any _known_ bugs. -That doesn't usually stop there being plenty we didn't know about, -though. Any that you find should be reported firstly via the -`bugtracker' at `https://sourceforge.net/projects/nasm/' (click on -"Bugs"), or if that fails then through one of the contacts in *note -Section 1.2::. - - Please read *note Section 2.2:: first, and don't report the bug if -it's listed in there as a deliberate feature. (If you think the feature -is badly thought out, feel free to send us reasons why you think it -should be changed, but don't just send us mail saying `This is a bug' -if the documentation says we did it on purpose.) Then read *note -Section 12.1::, and don't bother reporting the bug if it's listed there. - - If you do report a bug, _please_ give us all of the following -information: - - * What operating system you're running NASM under. DOS, Linux, - NetBSD, Win16, Win32, VMS (I'd be impressed), whatever. - - * If you're running NASM under DOS or Win32, tell us whether you've - compiled your own executable from the DOS source archive, or - whether you were using the standard distribution binaries out of - the archive. If you were using a locally built executable, try to - reproduce the problem using one of the standard binaries, as this - will make it easier for us to reproduce your problem prior to - fixing it. - - * Which version of NASM you're using, and exactly how you invoked - it. Give us the precise command line, and the contents of the - `NASMENV' environment variable if any. - - * Which versions of any supplementary programs you're using, and how - you invoked them. If the problem only becomes visible at link - time, tell us what linker you're using, what version of it you've - got, and the exact linker command line. If the problem involves - linking against object files generated by a compiler, tell us what - compiler, what version, and what command line or options you used. - (If you're compiling in an IDE, please try to reproduce the - problem with the command-line version of the compiler.) - - * If at all possible, send us a NASM source file which exhibits the - problem. If this causes copyright problems (e.g. you can only - reproduce the bug in restricted-distribution code) then bear in - mind the following two points: firstly, we guarantee that any - source code sent to us for the purposes of debugging NASM will be - used _only_ for the purposes of debugging NASM, and that we will - delete all our copies of it as soon as we have found and fixed the - bug or bugs in question; and secondly, we would prefer _not_ to be - mailed large chunks of code anyway. The smaller the file, the - better. A three-line sample file that does nothing useful - _except_ demonstrate the problem is much easier to work with than - a fully fledged ten-thousand- line program. (Of course, some - errors _do_ only crop up in large files, so this may not be - possible.) - - * A description of what the problem actually _is_. `It doesn't work' - is _not_ a helpful description! Please describe exactly what is - happening that shouldn't be, or what isn't happening that should. - Examples might be: `NASM generates an error message saying Line 3 - for an error that's actually on Line 5'; `NASM generates an error - message that I believe it shouldn't be generating at all'; `NASM - fails to generate an error message that I believe it _should_ be - generating'; `the object file produced from this source code - crashes my linker'; `the ninth byte of the output file is 66 and I - think it should be 77 instead'. - - * If you believe the output file from NASM to be faulty, send it to - us. That allows us to determine whether our own copy of NASM - generates the same file, or whether the problem is related to - portability issues between our development platforms and yours. We - can handle binary files mailed to us as MIME attachments, - uuencoded, and even BinHex. Alternatively, we may be able to - provide an FTP site you can upload the suspect files to; but - mailing them is easier for us. - - * Any other information or data files that might be helpful. If, for - example, the problem involves NASM failing to generate an object - file while TASM can generate an equivalent file without trouble, - then send us _both_ object files, so we can see what TASM is doing - differently from us. - - -File: nasm.info, Node: Appendix A, Next: Section A.1, Prev: Section 12.2, Up: Top - -Appendix A: Ndisasm -******************* - -The Netwide Disassembler, NDISASM - -* Menu: - -* Section A.1:: Introduction -* Section A.2:: Getting Started: Installation -* Section A.3:: Running NDISASM -* Section A.4:: Bugs and Improvements - - -File: nasm.info, Node: Section A.1, Next: Section A.2, Prev: Appendix A, Up: Appendix A - -A.1. Introduction -================= - -The Netwide Disassembler is a small companion program to the Netwide -Assembler, NASM. It seemed a shame to have an x86 assembler, complete -with a full instruction table, and not make as much use of it as -possible, so here's a disassembler which shares the instruction table -(and some other bits of code) with NASM. - - The Netwide Disassembler does nothing except to produce -disassemblies of _binary_ source files. NDISASM does not have any -understanding of object file formats, like `objdump', and it will not -understand `DOS .EXE' files like `debug' will. It just disassembles. - - -File: nasm.info, Node: Section A.2, Next: Section A.3, Prev: Section A.1, Up: Appendix A - -A.2. Getting Started: Installation -================================== - -See *note Section 1.3:: for installation instructions. NDISASM, like -NASM, has a `man page' which you may want to put somewhere useful, if -you are on a Unix system. - - -File: nasm.info, Node: Section A.3, Next: Section A.3.1, Prev: Section A.2, Up: Appendix A - -A.3. Running NDISASM -==================== - -To disassemble a file, you will typically use a command of the form - - ndisasm -b {16|32|64} filename - - NDISASM can disassemble 16-, 32- or 64-bit code equally easily, -provided of course that you remember to specify which it is to work -with. If no `-b' switch is present, NDISASM works in 16-bit mode by -default. The `-u' switch (for USE32) also invokes 32-bit mode. - - Two more command line options are `-r' which reports the version -number of NDISASM you are running, and `-h' which gives a short summary -of command line options. - -* Menu: - -* Section A.3.1:: COM Files: Specifying an Origin -* Section A.3.2:: Code Following Data: Synchronisation -* Section A.3.3:: Mixed Code and Data: Automatic (Intelligent) Synchronisation -* Section A.3.4:: Other Options - - -File: nasm.info, Node: Section A.3.1, Next: Section A.3.2, Prev: Section A.3, Up: Section A.3 - -A.3.1. COM Files: Specifying an Origin --------------------------------------- - -To disassemble a `DOS .COM' file correctly, a disassembler must assume -that the first instruction in the file is loaded at address `0x100', -rather than at zero. NDISASM, which assumes by default that any file you -give it is loaded at zero, will therefore need to be informed of this. - - The `-o' option allows you to declare a different origin for the file -you are disassembling. Its argument may be expressed in any of the NASM -numeric formats: decimal by default, if it begins with ``$'' or ``0x'' -or ends in ``H'' it's `hex', if it ends in ``Q'' it's `octal', and if -it ends in ``B'' it's `binary'. - - Hence, to disassemble a `.COM' file: - - ndisasm -o100h filename.com - - will do the trick. - - -File: nasm.info, Node: Section A.3.2, Next: Section A.3.3, Prev: Section A.3.1, Up: Section A.3 - -A.3.2. Code Following Data: Synchronisation -------------------------------------------- - -Suppose you are disassembling a file which contains some data which -isn't machine code, and _then_ contains some machine code. NDISASM will -faithfully plough through the data section, producing machine -instructions wherever it can (although most of them will look bizarre, -and some may have unusual prefixes, e.g. ``FS OR AX,0x240A''), and -generating `DB' instructions ever so often if it's totally stumped. -Then it will reach the code section. - - Supposing NDISASM has just finished generating a strange machine -instruction from part of the data section, and its file position is now -one byte _before_ the beginning of the code section. It's entirely -possible that another spurious instruction will get generated, starting -with the final byte of the data section, and then the correct first -instruction in the code section will not be seen because the starting -point skipped over it. This isn't really ideal. - - To avoid this, you can specify a ``synchronisation'' point, or indeed -as many synchronisation points as you like (although NDISASM can only -handle 2147483647 sync points internally). The definition of a sync -point is this: NDISASM guarantees to hit sync points exactly during -disassembly. If it is thinking about generating an instruction which -would cause it to jump over a sync point, it will discard that -instruction and output a ``db'' instead. So it _will_ start disassembly -exactly from the sync point, and so you _will_ see all the instructions -in your code section. - - Sync points are specified using the `-s' option: they are measured in -terms of the program origin, not the file position. So if you want to -synchronize after 32 bytes of a `.COM' file, you would have to do - - ndisasm -o100h -s120h file.com - - rather than - - ndisasm -o100h -s20h file.com - - As stated above, you can specify multiple sync markers if you need -to, just by repeating the `-s' option. - - -File: nasm.info, Node: Section A.3.3, Next: Section A.3.4, Prev: Section A.3.2, Up: Section A.3 - -A.3.3. Mixed Code and Data: Automatic (Intelligent) Synchronisation -------------------------------------------------------------------- - -Suppose you are disassembling the boot sector of a `DOS' floppy (maybe -it has a virus, and you need to understand the virus so that you know -what kinds of damage it might have done you). Typically, this will -contain a `JMP' instruction, then some data, then the rest of the code. -So there is a very good chance of NDISASM being _misaligned_ when the -data ends and the code begins. Hence a sync point is needed. - - On the other hand, why should you have to specify the sync point -manually? What you'd do in order to find where the sync point would -be, surely, would be to read the `JMP' instruction, and then to use its -target address as a sync point. So can NDISASM do that for you? - - The answer, of course, is yes: using either of the synonymous -switches `-a' (for automatic sync) or `-i' (for intelligent sync) will -enable `auto-sync' mode. Auto-sync mode automatically generates a sync -point for any forward-referring PC-relative jump or call instruction -that NDISASM encounters. (Since NDISASM is one-pass, if it encounters a -PC- relative jump whose target has already been processed, there isn't -much it can do about it...) - - Only PC-relative jumps are processed, since an absolute jump is -either through a register (in which case NDISASM doesn't know what the -register contains) or involves a segment address (in which case the -target code isn't in the same segment that NDISASM is working in, and -so the sync point can't be placed anywhere useful). - - For some kinds of file, this mechanism will automatically put sync -points in all the right places, and save you from having to place any -sync points manually. However, it should be stressed that auto-sync -mode is _not_ guaranteed to catch all the sync points, and you may -still have to place some manually. - - Auto-sync mode doesn't prevent you from declaring manual sync -points: it just adds automatically generated ones to the ones you -provide. It's perfectly feasible to specify `-i' _and_ some `-s' -options. - - Another caveat with auto-sync mode is that if, by some unpleasant -fluke, something in your data section should disassemble to a -PC-relative call or jump instruction, NDISASM may obediently place a -sync point in a totally random place, for example in the middle of one -of the instructions in your code section. So you may end up with a -wrong disassembly even if you use auto-sync. Again, there isn't much I -can do about this. If you have problems, you'll have to use manual sync -points, or use the `-k' option (documented below) to suppress -disassembly of the data area. - - -File: nasm.info, Node: Section A.3.4, Next: Section A.4, Prev: Section A.3.3, Up: Section A.3 - -A.3.4. Other Options --------------------- - -The `-e' option skips a header on the file, by ignoring the first N -bytes. This means that the header is _not_ counted towards the -disassembly offset: if you give `-e10 -o10', disassembly will start at -byte 10 in the file, and this will be given offset 10, not 20. - - The `-k' option is provided with two comma-separated numeric -arguments, the first of which is an assembly offset and the second is a -number of bytes to skip. This _will_ count the skipped bytes towards -the assembly offset: its use is to suppress disassembly of a data -section which wouldn't contain anything you wanted to see anyway. - - -File: nasm.info, Node: Section A.4, Next: Appendix B, Prev: Section A.3.4, Up: Appendix A - -A.4. Bugs and Improvements -========================== - -There are no known bugs. However, any you find, with patches if -possible, should be sent to `nasm-bugs@lists.sourceforge.net', or to the -developer's site at `https://sourceforge.net/projects/nasm/' and we'll -try to fix them. Feel free to send contributions and new features as -well. - - -File: nasm.info, Node: Appendix B, Next: Section B.1, Prev: Section A.4, Up: Top - -Appendix B: Instruction List -**************************** - -* Menu: - -* Section B.1:: Introduction - - -File: nasm.info, Node: Section B.1, Next: Section B.1.1, Prev: Appendix B, Up: Appendix B - -B.1. Introduction -================= - -The following sections show the instructions which NASM currently -supports. For each instruction, there is a separate entry for each -supported addressing mode. The third column shows the processor type in -which the instruction was introduced and, when appropriate, one or more -usage flags. - -* Menu: - -* Section B.1.1:: Special instructions... -* Section B.1.2:: Conventional instructions -* Section B.1.3:: Katmai Streaming SIMD instructions (SSE ---- a.k.a. KNI, XMM, MMX2) -* Section B.1.4:: Introduced in Deschutes but necessary for SSE support -* Section B.1.5:: XSAVE group (AVX and extended state) -* Section B.1.6:: Generic memory operations -* Section B.1.7:: New MMX instructions introduced in Katmai -* Section B.1.8:: AMD Enhanced 3DNow! (Athlon) instructions -* Section B.1.9:: Willamette SSE2 Cacheability Instructions -* Section B.1.10:: Willamette MMX instructions (SSE2 SIMD Integer Instructions) -* Section B.1.11:: Willamette Streaming SIMD instructions (SSE2) -* Section B.1.12:: Prescott New Instructions (SSE3) -* Section B.1.13:: VMX Instructions -* Section B.1.14:: Extended Page Tables VMX instructions -* Section B.1.15:: Tejas New Instructions (SSSE3) -* Section B.1.16:: AMD SSE4A -* Section B.1.17:: New instructions in Barcelona -* Section B.1.18:: Penryn New Instructions (SSE4.1) -* Section B.1.19:: Nehalem New Instructions (SSE4.2) -* Section B.1.20:: Intel SMX -* Section B.1.21:: Geode (Cyrix) 3DNow! additions -* Section B.1.22:: Intel new instructions in ??? -* Section B.1.23:: Intel AES instructions -* Section B.1.24:: Intel AVX AES instructions -* Section B.1.25:: Intel AVX instructions -* Section B.1.26:: Intel Carry-Less Multiplication instructions (CLMUL) -* Section B.1.27:: Intel AVX Carry-Less Multiplication instructions (CLMUL) -* Section B.1.28:: Intel Fused Multiply-Add instructions (FMA) -* Section B.1.29:: VIA (Centaur) security instructions -* Section B.1.30:: AMD Lightweight Profiling (LWP) instructions -* Section B.1.31:: AMD XOP, FMA4 and CVT16 instructions (SSE5) -* Section B.1.32:: Systematic names for the hinting nop instructions - - -File: nasm.info, Node: Section B.1.1, Next: Section B.1.2, Prev: Section B.1, Up: Section B.1 - -B.1.1. Special instructions... ------------------------------- - - DB - DW - DD - DQ - DT - DO - DY - RESB imm 8086 - RESW - RESD - RESQ - REST - RESO - RESY - - -File: nasm.info, Node: Section B.1.2, Next: Section B.1.3, Prev: Section B.1.1, Up: Section B.1 - -B.1.2. Conventional instructions --------------------------------- - - AAA 8086,NOLONG - AAD 8086,NOLONG - AAD imm 8086,NOLONG - AAM 8086,NOLONG - AAM imm 8086,NOLONG - AAS 8086,NOLONG - ADC mem,reg8 8086 - ADC reg8,reg8 8086 - ADC mem,reg16 8086 - ADC reg16,reg16 8086 - ADC mem,reg32 386 - ADC reg32,reg32 386 - ADC mem,reg64 X64 - ADC reg64,reg64 X64 - ADC reg8,mem 8086 - ADC reg8,reg8 8086 - ADC reg16,mem 8086 - ADC reg16,reg16 8086 - ADC reg32,mem 386 - ADC reg32,reg32 386 - ADC reg64,mem X64 - ADC reg64,reg64 X64 - ADC rm16,imm8 8086 - ADC rm32,imm8 386 - ADC rm64,imm8 X64 - ADC reg_al,imm 8086 - ADC reg_ax,sbyte16 8086 - ADC reg_ax,imm 8086 - ADC reg_eax,sbyte32 386 - ADC reg_eax,imm 386 - ADC reg_rax,sbyte64 X64 - ADC reg_rax,imm X64 - ADC rm8,imm 8086 - ADC rm16,imm 8086 - ADC rm32,imm 386 - ADC rm64,imm X64 - ADC mem,imm8 8086 - ADC mem,imm16 8086 - ADC mem,imm32 386 - ADD mem,reg8 8086 - ADD reg8,reg8 8086 - ADD mem,reg16 8086 - ADD reg16,reg16 8086 - ADD mem,reg32 386 - ADD reg32,reg32 386 - ADD mem,reg64 X64 - ADD reg64,reg64 X64 - ADD reg8,mem 8086 - ADD reg8,reg8 8086 - ADD reg16,mem 8086 - ADD reg16,reg16 8086 - ADD reg32,mem 386 - ADD reg32,reg32 386 - ADD reg64,mem X64 - ADD reg64,reg64 X64 - ADD rm16,imm8 8086 - ADD rm32,imm8 386 - ADD rm64,imm8 X64 - ADD reg_al,imm 8086 - ADD reg_ax,sbyte16 8086 - ADD reg_ax,imm 8086 - ADD reg_eax,sbyte32 386 - ADD reg_eax,imm 386 - ADD reg_rax,sbyte64 X64 - ADD reg_rax,imm X64 - ADD rm8,imm 8086 - ADD rm16,imm 8086 - ADD rm32,imm 386 - ADD rm64,imm X64 - ADD mem,imm8 8086 - ADD mem,imm16 8086 - ADD mem,imm32 386 - AND mem,reg8 8086 - AND reg8,reg8 8086 - AND mem,reg16 8086 - AND reg16,reg16 8086 - AND mem,reg32 386 - AND reg32,reg32 386 - AND mem,reg64 X64 - AND reg64,reg64 X64 - AND reg8,mem 8086 - AND reg8,reg8 8086 - AND reg16,mem 8086 - AND reg16,reg16 8086 - AND reg32,mem 386 - AND reg32,reg32 386 - AND reg64,mem X64 - AND reg64,reg64 X64 - AND rm16,imm8 8086 - AND rm32,imm8 386 - AND rm64,imm8 X64 - AND reg_al,imm 8086 - AND reg_ax,sbyte16 8086 - AND reg_ax,imm 8086 - AND reg_eax,sbyte32 386 - AND reg_eax,imm 386 - AND reg_rax,sbyte64 X64 - AND reg_rax,imm X64 - AND rm8,imm 8086 - AND rm16,imm 8086 - AND rm32,imm 386 - AND rm64,imm X64 - AND mem,imm8 8086 - AND mem,imm16 8086 - AND mem,imm32 386 - ARPL mem,reg16 286,PROT,NOLONG - ARPL reg16,reg16 286,PROT,NOLONG - BB0_RESET PENT,CYRIX,ND - BB1_RESET PENT,CYRIX,ND - BOUND reg16,mem 186,NOLONG - BOUND reg32,mem 386,NOLONG - BSF reg16,mem 386 - BSF reg16,reg16 386 - BSF reg32,mem 386 - BSF reg32,reg32 386 - BSF reg64,mem X64 - BSF reg64,reg64 X64 - BSR reg16,mem 386 - BSR reg16,reg16 386 - BSR reg32,mem 386 - BSR reg32,reg32 386 - BSR reg64,mem X64 - BSR reg64,reg64 X64 - BSWAP reg32 486 - BSWAP reg64 X64 - BT mem,reg16 386 - BT reg16,reg16 386 - BT mem,reg32 386 - BT reg32,reg32 386 - BT mem,reg64 X64 - BT reg64,reg64 X64 - BT rm16,imm 386 - BT rm32,imm 386 - BT rm64,imm X64 - BTC mem,reg16 386 - BTC reg16,reg16 386 - BTC mem,reg32 386 - BTC reg32,reg32 386 - BTC mem,reg64 X64 - BTC reg64,reg64 X64 - BTC rm16,imm 386 - BTC rm32,imm 386 - BTC rm64,imm X64 - BTR mem,reg16 386 - BTR reg16,reg16 386 - BTR mem,reg32 386 - BTR reg32,reg32 386 - BTR mem,reg64 X64 - BTR reg64,reg64 X64 - BTR rm16,imm 386 - BTR rm32,imm 386 - BTR rm64,imm X64 - BTS mem,reg16 386 - BTS reg16,reg16 386 - BTS mem,reg32 386 - BTS reg32,reg32 386 - BTS mem,reg64 X64 - BTS reg64,reg64 X64 - BTS rm16,imm 386 - BTS rm32,imm 386 - BTS rm64,imm X64 - CALL imm 8086 - CALL imm|near 8086 - CALL imm|far 8086,ND,NOLONG - CALL imm16 8086 - CALL imm16|near 8086 - CALL imm16|far 8086,ND,NOLONG - CALL imm32 386 - CALL imm32|near 386 - CALL imm32|far 386,ND,NOLONG - CALL imm:imm 8086,NOLONG - CALL imm16:imm 8086,NOLONG - CALL imm:imm16 8086,NOLONG - CALL imm32:imm 386,NOLONG - CALL imm:imm32 386,NOLONG - CALL mem|far 8086,NOLONG - CALL mem|far X64 - CALL mem16|far 8086 - CALL mem32|far 386 - CALL mem64|far X64 - CALL mem|near 8086 - CALL mem16|near 8086 - CALL mem32|near 386,NOLONG - CALL mem64|near X64 - CALL reg16 8086 - CALL reg32 386,NOLONG - CALL reg64 X64 - CALL mem 8086 - CALL mem16 8086 - CALL mem32 386,NOLONG - CALL mem64 X64 - CBW 8086 - CDQ 386 - CDQE X64 - CLC 8086 - CLD 8086 - CLGI X64,AMD - CLI 8086 - CLTS 286,PRIV - CMC 8086 - CMP mem,reg8 8086 - CMP reg8,reg8 8086 - CMP mem,reg16 8086 - CMP reg16,reg16 8086 - CMP mem,reg32 386 - CMP reg32,reg32 386 - CMP mem,reg64 X64 - CMP reg64,reg64 X64 - CMP reg8,mem 8086 - CMP reg8,reg8 8086 - CMP reg16,mem 8086 - CMP reg16,reg16 8086 - CMP reg32,mem 386 - CMP reg32,reg32 386 - CMP reg64,mem X64 - CMP reg64,reg64 X64 - CMP rm16,imm8 8086 - CMP rm32,imm8 386 - CMP rm64,imm8 X64 - CMP reg_al,imm 8086 - CMP reg_ax,sbyte16 8086 - CMP reg_ax,imm 8086 - CMP reg_eax,sbyte32 386 - CMP reg_eax,imm 386 - CMP reg_rax,sbyte64 X64 - CMP reg_rax,imm X64 - CMP rm8,imm 8086 - CMP rm16,imm 8086 - CMP rm32,imm 386 - CMP rm64,imm X64 - CMP mem,imm8 8086 - CMP mem,imm16 8086 - CMP mem,imm32 386 - CMPSB 8086 - CMPSD 386 - CMPSQ X64 - CMPSW 8086 - CMPXCHG mem,reg8 PENT - CMPXCHG reg8,reg8 PENT - CMPXCHG mem,reg16 PENT - CMPXCHG reg16,reg16 PENT - CMPXCHG mem,reg32 PENT - CMPXCHG reg32,reg32 PENT - CMPXCHG mem,reg64 X64 - CMPXCHG reg64,reg64 X64 - CMPXCHG486 mem,reg8 486,UNDOC,ND - CMPXCHG486 reg8,reg8 486,UNDOC,ND - CMPXCHG486 mem,reg16 486,UNDOC,ND - CMPXCHG486 reg16,reg16 486,UNDOC,ND - CMPXCHG486 mem,reg32 486,UNDOC,ND - CMPXCHG486 reg32,reg32 486,UNDOC,ND - CMPXCHG8B mem PENT - CMPXCHG16B mem X64 - CPUID PENT - CPU_READ PENT,CYRIX - CPU_WRITE PENT,CYRIX - CQO X64 - CWD 8086 - CWDE 386 - DAA 8086,NOLONG - DAS 8086,NOLONG - DEC reg16 8086,NOLONG - DEC reg32 386,NOLONG - DEC rm8 8086 - DEC rm16 8086 - DEC rm32 386 - DEC rm64 X64 - DIV rm8 8086 - DIV rm16 8086 - DIV rm32 386 - DIV rm64 X64 - DMINT P6,CYRIX - EMMS PENT,MMX - ENTER imm,imm 186 - EQU imm 8086 - EQU imm:imm 8086 - F2XM1 8086,FPU - FABS 8086,FPU - FADD mem32 8086,FPU - FADD mem64 8086,FPU - FADD fpureg|to 8086,FPU - FADD fpureg 8086,FPU - FADD fpureg,fpu0 8086,FPU - FADD fpu0,fpureg 8086,FPU - FADD 8086,FPU,ND - FADDP fpureg 8086,FPU - FADDP fpureg,fpu0 8086,FPU - FADDP 8086,FPU,ND - FBLD mem80 8086,FPU - FBLD mem 8086,FPU - FBSTP mem80 8086,FPU - FBSTP mem 8086,FPU - FCHS 8086,FPU - FCLEX 8086,FPU - FCMOVB fpureg P6,FPU - FCMOVB fpu0,fpureg P6,FPU - FCMOVB P6,FPU,ND - FCMOVBE fpureg P6,FPU - FCMOVBE fpu0,fpureg P6,FPU - FCMOVBE P6,FPU,ND - FCMOVE fpureg P6,FPU - FCMOVE fpu0,fpureg P6,FPU - FCMOVE P6,FPU,ND - FCMOVNB fpureg P6,FPU - FCMOVNB fpu0,fpureg P6,FPU - FCMOVNB P6,FPU,ND - FCMOVNBE fpureg P6,FPU - FCMOVNBE fpu0,fpureg P6,FPU - FCMOVNBE P6,FPU,ND - FCMOVNE fpureg P6,FPU - FCMOVNE fpu0,fpureg P6,FPU - FCMOVNE P6,FPU,ND - FCMOVNU fpureg P6,FPU - FCMOVNU fpu0,fpureg P6,FPU - FCMOVNU P6,FPU,ND - FCMOVU fpureg P6,FPU - FCMOVU fpu0,fpureg P6,FPU - FCMOVU P6,FPU,ND - FCOM mem32 8086,FPU - FCOM mem64 8086,FPU - FCOM fpureg 8086,FPU - FCOM fpu0,fpureg 8086,FPU - FCOM 8086,FPU,ND - FCOMI fpureg P6,FPU - FCOMI fpu0,fpureg P6,FPU - FCOMI P6,FPU,ND - FCOMIP fpureg P6,FPU - FCOMIP fpu0,fpureg P6,FPU - FCOMIP P6,FPU,ND - FCOMP mem32 8086,FPU - FCOMP mem64 8086,FPU - FCOMP fpureg 8086,FPU - FCOMP fpu0,fpureg 8086,FPU - FCOMP 8086,FPU,ND - FCOMPP 8086,FPU - FCOS 386,FPU - FDECSTP 8086,FPU - FDISI 8086,FPU - FDIV mem32 8086,FPU - FDIV mem64 8086,FPU - FDIV fpureg|to 8086,FPU - FDIV fpureg 8086,FPU - FDIV fpureg,fpu0 8086,FPU - FDIV fpu0,fpureg 8086,FPU - FDIV 8086,FPU,ND - FDIVP fpureg 8086,FPU - FDIVP fpureg,fpu0 8086,FPU - FDIVP 8086,FPU,ND - FDIVR mem32 8086,FPU - FDIVR mem64 8086,FPU - FDIVR fpureg|to 8086,FPU - FDIVR fpureg,fpu0 8086,FPU - FDIVR fpureg 8086,FPU - FDIVR fpu0,fpureg 8086,FPU - FDIVR 8086,FPU,ND - FDIVRP fpureg 8086,FPU - FDIVRP fpureg,fpu0 8086,FPU - FDIVRP 8086,FPU,ND - FEMMS PENT,3DNOW - FENI 8086,FPU - FFREE fpureg 8086,FPU - FFREE 8086,FPU - FFREEP fpureg 286,FPU,UNDOC - FFREEP 286,FPU,UNDOC - FIADD mem32 8086,FPU - FIADD mem16 8086,FPU - FICOM mem32 8086,FPU - FICOM mem16 8086,FPU - FICOMP mem32 8086,FPU - FICOMP mem16 8086,FPU - FIDIV mem32 8086,FPU - FIDIV mem16 8086,FPU - FIDIVR mem32 8086,FPU - FIDIVR mem16 8086,FPU - FILD mem32 8086,FPU - FILD mem16 8086,FPU - FILD mem64 8086,FPU - FIMUL mem32 8086,FPU - FIMUL mem16 8086,FPU - FINCSTP 8086,FPU - FINIT 8086,FPU - FIST mem32 8086,FPU - FIST mem16 8086,FPU - FISTP mem32 8086,FPU - FISTP mem16 8086,FPU - FISTP mem64 8086,FPU - FISTTP mem16 PRESCOTT,FPU - FISTTP mem32 PRESCOTT,FPU - FISTTP mem64 PRESCOTT,FPU - FISUB mem32 8086,FPU - FISUB mem16 8086,FPU - FISUBR mem32 8086,FPU - FISUBR mem16 8086,FPU - FLD mem32 8086,FPU - FLD mem64 8086,FPU - FLD mem80 8086,FPU - FLD fpureg 8086,FPU - FLD 8086,FPU,ND - FLD1 8086,FPU - FLDCW mem 8086,FPU,SW - FLDENV mem 8086,FPU - FLDL2E 8086,FPU - FLDL2T 8086,FPU - FLDLG2 8086,FPU - FLDLN2 8086,FPU - FLDPI 8086,FPU - FLDZ 8086,FPU - FMUL mem32 8086,FPU - FMUL mem64 8086,FPU - FMUL fpureg|to 8086,FPU - FMUL fpureg,fpu0 8086,FPU - FMUL fpureg 8086,FPU - FMUL fpu0,fpureg 8086,FPU - FMUL 8086,FPU,ND - FMULP fpureg 8086,FPU - FMULP fpureg,fpu0 8086,FPU - FMULP 8086,FPU,ND - FNCLEX 8086,FPU - FNDISI 8086,FPU - FNENI 8086,FPU - FNINIT 8086,FPU - FNOP 8086,FPU - FNSAVE mem 8086,FPU - FNSTCW mem 8086,FPU,SW - FNSTENV mem 8086,FPU - FNSTSW mem 8086,FPU,SW - FNSTSW reg_ax 286,FPU - FPATAN 8086,FPU - FPREM 8086,FPU - FPREM1 386,FPU - FPTAN 8086,FPU - FRNDINT 8086,FPU - FRSTOR mem 8086,FPU - FSAVE mem 8086,FPU - FSCALE 8086,FPU - FSETPM 286,FPU - FSIN 386,FPU - FSINCOS 386,FPU - FSQRT 8086,FPU - FST mem32 8086,FPU - FST mem64 8086,FPU - FST fpureg 8086,FPU - FST 8086,FPU,ND - FSTCW mem 8086,FPU,SW - FSTENV mem 8086,FPU - FSTP mem32 8086,FPU - FSTP mem64 8086,FPU - FSTP mem80 8086,FPU - FSTP fpureg 8086,FPU - FSTP 8086,FPU,ND - FSTSW mem 8086,FPU,SW - FSTSW reg_ax 286,FPU - FSUB mem32 8086,FPU - FSUB mem64 8086,FPU - FSUB fpureg|to 8086,FPU - FSUB fpureg,fpu0 8086,FPU - FSUB fpureg 8086,FPU - FSUB fpu0,fpureg 8086,FPU - FSUB 8086,FPU,ND - FSUBP fpureg 8086,FPU - FSUBP fpureg,fpu0 8086,FPU - FSUBP 8086,FPU,ND - FSUBR mem32 8086,FPU - FSUBR mem64 8086,FPU - FSUBR fpureg|to 8086,FPU - FSUBR fpureg,fpu0 8086,FPU - FSUBR fpureg 8086,FPU - FSUBR fpu0,fpureg 8086,FPU - FSUBR 8086,FPU,ND - FSUBRP fpureg 8086,FPU - FSUBRP fpureg,fpu0 8086,FPU - FSUBRP 8086,FPU,ND - FTST 8086,FPU - FUCOM fpureg 386,FPU - FUCOM fpu0,fpureg 386,FPU - FUCOM 386,FPU,ND - FUCOMI fpureg P6,FPU - FUCOMI fpu0,fpureg P6,FPU - FUCOMI P6,FPU,ND - FUCOMIP fpureg P6,FPU - FUCOMIP fpu0,fpureg P6,FPU - FUCOMIP P6,FPU,ND - FUCOMP fpureg 386,FPU - FUCOMP fpu0,fpureg 386,FPU - FUCOMP 386,FPU,ND - FUCOMPP 386,FPU - FXAM 8086,FPU - FXCH fpureg 8086,FPU - FXCH fpureg,fpu0 8086,FPU - FXCH fpu0,fpureg 8086,FPU - FXCH 8086,FPU,ND - FXTRACT 8086,FPU - FYL2X 8086,FPU - FYL2XP1 8086,FPU - HLT 8086,PRIV - IBTS mem,reg16 386,SW,UNDOC,ND - IBTS reg16,reg16 386,UNDOC,ND - IBTS mem,reg32 386,SD,UNDOC,ND - IBTS reg32,reg32 386,UNDOC,ND - ICEBP 386,ND - IDIV rm8 8086 - IDIV rm16 8086 - IDIV rm32 386 - IDIV rm64 X64 - IMUL rm8 8086 - IMUL rm16 8086 - IMUL rm32 386 - IMUL rm64 X64 - IMUL reg16,mem 386 - IMUL reg16,reg16 386 - IMUL reg32,mem 386 - IMUL reg32,reg32 386 - IMUL reg64,mem X64 - IMUL reg64,reg64 X64 - IMUL reg16,mem,imm8 186 - IMUL reg16,mem,sbyte16 186,ND - IMUL reg16,mem,imm16 186 - IMUL reg16,mem,imm 186,ND - IMUL reg16,reg16,imm8 186 - IMUL reg16,reg16,sbyte16 186,ND - IMUL reg16,reg16,imm16 186 - IMUL reg16,reg16,imm 186,ND - IMUL reg32,mem,imm8 386 - IMUL reg32,mem,sbyte32 386,ND - IMUL reg32,mem,imm32 386 - IMUL reg32,mem,imm 386,ND - IMUL reg32,reg32,imm8 386 - IMUL reg32,reg32,sbyte32 386,ND - IMUL reg32,reg32,imm32 386 - IMUL reg32,reg32,imm 386,ND - IMUL reg64,mem,imm8 X64 - IMUL reg64,mem,sbyte64 X64,ND - IMUL reg64,mem,imm32 X64 - IMUL reg64,mem,imm X64,ND - IMUL reg64,reg64,imm8 X64 - IMUL reg64,reg64,sbyte64 X64,ND - IMUL reg64,reg64,imm32 X64 - IMUL reg64,reg64,imm X64,ND - IMUL reg16,imm8 186 - IMUL reg16,sbyte16 186,ND - IMUL reg16,imm16 186 - IMUL reg16,imm 186,ND - IMUL reg32,imm8 386 - IMUL reg32,sbyte32 386,ND - IMUL reg32,imm32 386 - IMUL reg32,imm 386,ND - IMUL reg64,imm8 X64 - IMUL reg64,sbyte64 X64,ND - IMUL reg64,imm32 X64 - IMUL reg64,imm X64,ND - IN reg_al,imm 8086 - IN reg_ax,imm 8086 - IN reg_eax,imm 386 - IN reg_al,reg_dx 8086 - IN reg_ax,reg_dx 8086 - IN reg_eax,reg_dx 386 - INC reg16 8086,NOLONG - INC reg32 386,NOLONG - INC rm8 8086 - INC rm16 8086 - INC rm32 386 - INC rm64 X64 - INCBIN - INSB 186 - INSD 386 - INSW 186 - INT imm 8086 - INT01 386,ND - INT1 386 - INT03 8086,ND - INT3 8086 - INTO 8086,NOLONG - INVD 486,PRIV - INVLPG mem 486,PRIV - INVLPGA reg_ax,reg_ecx X86_64,AMD,NOLONG - INVLPGA reg_eax,reg_ecx X86_64,AMD - INVLPGA reg_rax,reg_ecx X64,AMD - INVLPGA X86_64,AMD - IRET 8086 - IRETD 386 - IRETQ X64 - IRETW 8086 - JCXZ imm 8086,NOLONG - JECXZ imm 386 - JRCXZ imm X64 - JMP imm|short 8086 - JMP imm 8086,ND - JMP imm 8086 - JMP imm|near 8086,ND - JMP imm|far 8086,ND,NOLONG - JMP imm16 8086 - JMP imm16|near 8086,ND - JMP imm16|far 8086,ND,NOLONG - JMP imm32 386 - JMP imm32|near 386,ND - JMP imm32|far 386,ND,NOLONG - JMP imm:imm 8086,NOLONG - JMP imm16:imm 8086,NOLONG - JMP imm:imm16 8086,NOLONG - JMP imm32:imm 386,NOLONG - JMP imm:imm32 386,NOLONG - JMP mem|far 8086,NOLONG - JMP mem|far X64 - JMP mem16|far 8086 - JMP mem32|far 386 - JMP mem64|far X64 - JMP mem|near 8086 - JMP mem16|near 8086 - JMP mem32|near 386,NOLONG - JMP mem64|near X64 - JMP reg16 8086 - JMP reg32 386,NOLONG - JMP reg64 X64 - JMP mem 8086 - JMP mem16 8086 - JMP mem32 386,NOLONG - JMP mem64 X64 - JMPE imm IA64 - JMPE imm16 IA64 - JMPE imm32 IA64 - JMPE rm16 IA64 - JMPE rm32 IA64 - LAHF 8086 - LAR reg16,mem 286,PROT,SW - LAR reg16,reg16 286,PROT - LAR reg16,reg32 386,PROT - LAR reg16,reg64 X64,PROT,ND - LAR reg32,mem 386,PROT,SW - LAR reg32,reg16 386,PROT - LAR reg32,reg32 386,PROT - LAR reg32,reg64 X64,PROT,ND - LAR reg64,mem X64,PROT,SW - LAR reg64,reg16 X64,PROT - LAR reg64,reg32 X64,PROT - LAR reg64,reg64 X64,PROT - LDS reg16,mem 8086,NOLONG - LDS reg32,mem 386,NOLONG - LEA reg16,mem 8086 - LEA reg32,mem 386 - LEA reg64,mem X64 - LEAVE 186 - LES reg16,mem 8086,NOLONG - LES reg32,mem 386,NOLONG - LFENCE X64,AMD - LFS reg16,mem 386 - LFS reg32,mem 386 - LGDT mem 286,PRIV - LGS reg16,mem 386 - LGS reg32,mem 386 - LIDT mem 286,PRIV - LLDT mem 286,PROT,PRIV - LLDT mem16 286,PROT,PRIV - LLDT reg16 286,PROT,PRIV - LMSW mem 286,PRIV - LMSW mem16 286,PRIV - LMSW reg16 286,PRIV - LOADALL 386,UNDOC - LOADALL286 286,UNDOC - LODSB 8086 - LODSD 386 - LODSQ X64 - LODSW 8086 - LOOP imm 8086 - LOOP imm,reg_cx 8086,NOLONG - LOOP imm,reg_ecx 386 - LOOP imm,reg_rcx X64 - LOOPE imm 8086 - LOOPE imm,reg_cx 8086,NOLONG - LOOPE imm,reg_ecx 386 - LOOPE imm,reg_rcx X64 - LOOPNE imm 8086 - LOOPNE imm,reg_cx 8086,NOLONG - LOOPNE imm,reg_ecx 386 - LOOPNE imm,reg_rcx X64 - LOOPNZ imm 8086 - LOOPNZ imm,reg_cx 8086,NOLONG - LOOPNZ imm,reg_ecx 386 - LOOPNZ imm,reg_rcx X64 - LOOPZ imm 8086 - LOOPZ imm,reg_cx 8086,NOLONG - LOOPZ imm,reg_ecx 386 - LOOPZ imm,reg_rcx X64 - LSL reg16,mem 286,PROT,SW - LSL reg16,reg16 286,PROT - LSL reg16,reg32 386,PROT - LSL reg16,reg64 X64,PROT,ND - LSL reg32,mem 386,PROT,SW - LSL reg32,reg16 386,PROT - LSL reg32,reg32 386,PROT - LSL reg32,reg64 X64,PROT,ND - LSL reg64,mem X64,PROT,SW - LSL reg64,reg16 X64,PROT - LSL reg64,reg32 X64,PROT - LSL reg64,reg64 X64,PROT - LSS reg16,mem 386 - LSS reg32,mem 386 - LTR mem 286,PROT,PRIV - LTR mem16 286,PROT,PRIV - LTR reg16 286,PROT,PRIV - MFENCE X64,AMD - MONITOR PRESCOTT - MONITOR reg_eax,reg_ecx,reg_edx PRESCOTT,ND - MONITOR reg_rax,reg_ecx,reg_edx X64,ND - MOV mem,reg_sreg 8086 - MOV reg16,reg_sreg 8086 - MOV reg32,reg_sreg 386 - MOV reg_sreg,mem 8086 - MOV reg_sreg,reg16 8086 - MOV reg_sreg,reg32 386 - MOV reg_al,mem_offs 8086 - MOV reg_ax,mem_offs 8086 - MOV reg_eax,mem_offs 386 - MOV reg_rax,mem_offs X64 - MOV mem_offs,reg_al 8086 - MOV mem_offs,reg_ax 8086 - MOV mem_offs,reg_eax 386 - MOV mem_offs,reg_rax X64 - MOV reg32,reg_creg 386,PRIV,NOLONG - MOV reg64,reg_creg X64,PRIV - MOV reg_creg,reg32 386,PRIV,NOLONG - MOV reg_creg,reg64 X64,PRIV - MOV reg32,reg_dreg 386,PRIV,NOLONG - MOV reg64,reg_dreg X64,PRIV - MOV reg_dreg,reg32 386,PRIV,NOLONG - MOV reg_dreg,reg64 X64,PRIV - MOV reg32,reg_treg 386,NOLONG,ND - MOV reg_treg,reg32 386,NOLONG,ND - MOV mem,reg8 8086 - MOV reg8,reg8 8086 - MOV mem,reg16 8086 - MOV reg16,reg16 8086 - MOV mem,reg32 386 - MOV reg32,reg32 386 - MOV mem,reg64 X64 - MOV reg64,reg64 X64 - MOV reg8,mem 8086 - MOV reg8,reg8 8086 - MOV reg16,mem 8086 - MOV reg16,reg16 8086 - MOV reg32,mem 386 - MOV reg32,reg32 386 - MOV reg64,mem X64 - MOV reg64,reg64 X64 - MOV reg8,imm 8086 - MOV reg16,imm 8086 - MOV reg32,imm 386 - MOV reg64,imm X64 - MOV reg64,imm32 X64 - MOV rm8,imm 8086 - MOV rm16,imm 8086 - MOV rm32,imm 386 - MOV rm64,imm X64 - MOV mem,imm8 8086 - MOV mem,imm16 8086 - MOV mem,imm32 386 - MOVD mmxreg,mem PENT,MMX,SD - MOVD mmxreg,reg32 PENT,MMX - MOVD mem,mmxreg PENT,MMX,SD - MOVD reg32,mmxreg PENT,MMX - MOVD xmmreg,mem X64,SD - MOVD xmmreg,reg32 X64 - MOVD mem,xmmreg X64,SD - MOVD reg32,xmmreg X64,SSE - MOVQ mmxreg,mmxrm PENT,MMX - MOVQ mmxrm,mmxreg PENT,MMX - MOVQ mmxreg,rm64 X64,MMX - MOVQ rm64,mmxreg X64,MMX - MOVSB 8086 - MOVSD 386 - MOVSQ X64 - MOVSW 8086 - MOVSX reg16,mem 386 - MOVSX reg16,reg8 386 - MOVSX reg32,rm8 386 - MOVSX reg32,rm16 386 - MOVSX reg64,rm8 X64 - MOVSX reg64,rm16 X64 - MOVSXD reg64,rm32 X64 - MOVSX reg64,rm32 X64,ND - MOVZX reg16,mem 386 - MOVZX reg16,reg8 386 - MOVZX reg32,rm8 386 - MOVZX reg32,rm16 386 - MOVZX reg64,rm8 X64 - MOVZX reg64,rm16 X64 - MUL rm8 8086 - MUL rm16 8086 - MUL rm32 386 - MUL rm64 X64 - MWAIT PRESCOTT - MWAIT reg_eax,reg_ecx PRESCOTT,ND - NEG rm8 8086 - NEG rm16 8086 - NEG rm32 386 - NEG rm64 X64 - NOP 8086 - NOP rm16 P6 - NOP rm32 P6 - NOP rm64 X64 - NOT rm8 8086 - NOT rm16 8086 - NOT rm32 386 - NOT rm64 X64 - OR mem,reg8 8086 - OR reg8,reg8 8086 - OR mem,reg16 8086 - OR reg16,reg16 8086 - OR mem,reg32 386 - OR reg32,reg32 386 - OR mem,reg64 X64 - OR reg64,reg64 X64 - OR reg8,mem 8086 - OR reg8,reg8 8086 - OR reg16,mem 8086 - OR reg16,reg16 8086 - OR reg32,mem 386 - OR reg32,reg32 386 - OR reg64,mem X64 - OR reg64,reg64 X64 - OR rm16,imm8 8086 - OR rm32,imm8 386 - OR rm64,imm8 X64 - OR reg_al,imm 8086 - OR reg_ax,sbyte16 8086 - OR reg_ax,imm 8086 - OR reg_eax,sbyte32 386 - OR reg_eax,imm 386 - OR reg_rax,sbyte64 X64 - OR reg_rax,imm X64 - OR rm8,imm 8086 - OR rm16,imm 8086 - OR rm32,imm 386 - OR rm64,imm X64 - OR mem,imm8 8086 - OR mem,imm16 8086 - OR mem,imm32 386 - OUT imm,reg_al 8086 - OUT imm,reg_ax 8086 - OUT imm,reg_eax 386 - OUT reg_dx,reg_al 8086 - OUT reg_dx,reg_ax 8086 - OUT reg_dx,reg_eax 386 - OUTSB 186 - OUTSD 386 - OUTSW 186 - PACKSSDW mmxreg,mmxrm PENT,MMX - PACKSSWB mmxreg,mmxrm PENT,MMX - PACKUSWB mmxreg,mmxrm PENT,MMX - PADDB mmxreg,mmxrm PENT,MMX - PADDD mmxreg,mmxrm PENT,MMX - PADDSB mmxreg,mmxrm PENT,MMX - PADDSIW mmxreg,mmxrm PENT,MMX,CYRIX - PADDSW mmxreg,mmxrm PENT,MMX - PADDUSB mmxreg,mmxrm PENT,MMX - PADDUSW mmxreg,mmxrm PENT,MMX - PADDW mmxreg,mmxrm PENT,MMX - PAND mmxreg,mmxrm PENT,MMX - PANDN mmxreg,mmxrm PENT,MMX - PAUSE 8086 - PAVEB mmxreg,mmxrm PENT,MMX,CYRIX - PAVGUSB mmxreg,mmxrm PENT,3DNOW - PCMPEQB mmxreg,mmxrm PENT,MMX - PCMPEQD mmxreg,mmxrm PENT,MMX - PCMPEQW mmxreg,mmxrm PENT,MMX - PCMPGTB mmxreg,mmxrm PENT,MMX - PCMPGTD mmxreg,mmxrm PENT,MMX - PCMPGTW mmxreg,mmxrm PENT,MMX - PDISTIB mmxreg,mem PENT,MMX,CYRIX - PF2ID mmxreg,mmxrm PENT,3DNOW - PFACC mmxreg,mmxrm PENT,3DNOW - PFADD mmxreg,mmxrm PENT,3DNOW - PFCMPEQ mmxreg,mmxrm PENT,3DNOW - PFCMPGE mmxreg,mmxrm PENT,3DNOW - PFCMPGT mmxreg,mmxrm PENT,3DNOW - PFMAX mmxreg,mmxrm PENT,3DNOW - PFMIN mmxreg,mmxrm PENT,3DNOW - PFMUL mmxreg,mmxrm PENT,3DNOW - PFRCP mmxreg,mmxrm PENT,3DNOW - PFRCPIT1 mmxreg,mmxrm PENT,3DNOW - PFRCPIT2 mmxreg,mmxrm PENT,3DNOW - PFRSQIT1 mmxreg,mmxrm PENT,3DNOW - PFRSQRT mmxreg,mmxrm PENT,3DNOW - PFSUB mmxreg,mmxrm PENT,3DNOW - PFSUBR mmxreg,mmxrm PENT,3DNOW - PI2FD mmxreg,mmxrm PENT,3DNOW - PMACHRIW mmxreg,mem PENT,MMX,CYRIX - PMADDWD mmxreg,mmxrm PENT,MMX - PMAGW mmxreg,mmxrm PENT,MMX,CYRIX - PMULHRIW mmxreg,mmxrm PENT,MMX,CYRIX - PMULHRWA mmxreg,mmxrm PENT,3DNOW - PMULHRWC mmxreg,mmxrm PENT,MMX,CYRIX - PMULHW mmxreg,mmxrm PENT,MMX - PMULLW mmxreg,mmxrm PENT,MMX - PMVGEZB mmxreg,mem PENT,MMX,CYRIX - PMVLZB mmxreg,mem PENT,MMX,CYRIX - PMVNZB mmxreg,mem PENT,MMX,CYRIX - PMVZB mmxreg,mem PENT,MMX,CYRIX - POP reg16 8086 - POP reg32 386,NOLONG - POP reg64 X64 - POP rm16 8086 - POP rm32 386,NOLONG - POP rm64 X64 - POP reg_cs 8086,UNDOC,ND - POP reg_dess 8086,NOLONG - POP reg_fsgs 386 - POPA 186,NOLONG - POPAD 386,NOLONG - POPAW 186,NOLONG - POPF 8086 - POPFD 386,NOLONG - POPFQ X64 - POPFW 8086 - POR mmxreg,mmxrm PENT,MMX - PREFETCH mem PENT,3DNOW - PREFETCHW mem PENT,3DNOW - PSLLD mmxreg,mmxrm PENT,MMX - PSLLD mmxreg,imm PENT,MMX - PSLLQ mmxreg,mmxrm PENT,MMX - PSLLQ mmxreg,imm PENT,MMX - PSLLW mmxreg,mmxrm PENT,MMX - PSLLW mmxreg,imm PENT,MMX - PSRAD mmxreg,mmxrm PENT,MMX - PSRAD mmxreg,imm PENT,MMX - PSRAW mmxreg,mmxrm PENT,MMX - PSRAW mmxreg,imm PENT,MMX - PSRLD mmxreg,mmxrm PENT,MMX - PSRLD mmxreg,imm PENT,MMX - PSRLQ mmxreg,mmxrm PENT,MMX - PSRLQ mmxreg,imm PENT,MMX - PSRLW mmxreg,mmxrm PENT,MMX - PSRLW mmxreg,imm PENT,MMX - PSUBB mmxreg,mmxrm PENT,MMX - PSUBD mmxreg,mmxrm PENT,MMX - PSUBSB mmxreg,mmxrm PENT,MMX - PSUBSIW mmxreg,mmxrm PENT,MMX,CYRIX - PSUBSW mmxreg,mmxrm PENT,MMX - PSUBUSB mmxreg,mmxrm PENT,MMX - PSUBUSW mmxreg,mmxrm PENT,MMX - PSUBW mmxreg,mmxrm PENT,MMX - PUNPCKHBW mmxreg,mmxrm PENT,MMX - PUNPCKHDQ mmxreg,mmxrm PENT,MMX - PUNPCKHWD mmxreg,mmxrm PENT,MMX - PUNPCKLBW mmxreg,mmxrm PENT,MMX - PUNPCKLDQ mmxreg,mmxrm PENT,MMX - PUNPCKLWD mmxreg,mmxrm PENT,MMX - PUSH reg16 8086 - PUSH reg32 386,NOLONG - PUSH reg64 X64 - PUSH rm16 8086 - PUSH rm32 386,NOLONG - PUSH rm64 X64 - PUSH reg_cs 8086,NOLONG - PUSH reg_dess 8086,NOLONG - PUSH reg_fsgs 386 - PUSH imm8 186 - PUSH imm16 186,AR0,SZ - PUSH imm32 386,NOLONG,AR0,SZ - PUSH imm32 386,NOLONG,SD - PUSH imm64 X64,AR0,SZ - PUSHA 186,NOLONG - PUSHAD 386,NOLONG - PUSHAW 186,NOLONG - PUSHF 8086 - PUSHFD 386,NOLONG - PUSHFQ X64 - PUSHFW 8086 - PXOR mmxreg,mmxrm PENT,MMX - RCL rm8,unity 8086 - RCL rm8,reg_cl 8086 - RCL rm8,imm 186 - RCL rm16,unity 8086 - RCL rm16,reg_cl 8086 - RCL rm16,imm 186 - RCL rm32,unity 386 - RCL rm32,reg_cl 386 - RCL rm32,imm 386 - RCL rm64,unity X64 - RCL rm64,reg_cl X64 - RCL rm64,imm X64 - RCR rm8,unity 8086 - RCR rm8,reg_cl 8086 - RCR rm8,imm 186 - RCR rm16,unity 8086 - RCR rm16,reg_cl 8086 - RCR rm16,imm 186 - RCR rm32,unity 386 - RCR rm32,reg_cl 386 - RCR rm32,imm 386 - RCR rm64,unity X64 - RCR rm64,reg_cl X64 - RCR rm64,imm X64 - RDSHR rm32 P6,CYRIXM - RDMSR PENT,PRIV - RDPMC P6 - RDTSC PENT - RDTSCP X86_64 - RET 8086 - RET imm 8086,SW - RETF 8086 - RETF imm 8086,SW - RETN 8086 - RETN imm 8086,SW - ROL rm8,unity 8086 - ROL rm8,reg_cl 8086 - ROL rm8,imm 186 - ROL rm16,unity 8086 - ROL rm16,reg_cl 8086 - ROL rm16,imm 186 - ROL rm32,unity 386 - ROL rm32,reg_cl 386 - ROL rm32,imm 386 - ROL rm64,unity X64 - ROL rm64,reg_cl X64 - ROL rm64,imm X64 - ROR rm8,unity 8086 - ROR rm8,reg_cl 8086 - ROR rm8,imm 186 - ROR rm16,unity 8086 - ROR rm16,reg_cl 8086 - ROR rm16,imm 186 - ROR rm32,unity 386 - ROR rm32,reg_cl 386 - ROR rm32,imm 386 - ROR rm64,unity X64 - ROR rm64,reg_cl X64 - ROR rm64,imm X64 - RDM P6,CYRIX,ND - RSDC reg_sreg,mem80 486,CYRIXM - RSLDT mem80 486,CYRIXM - RSM PENTM - RSTS mem80 486,CYRIXM - SAHF 8086 - SAL rm8,unity 8086,ND - SAL rm8,reg_cl 8086,ND - SAL rm8,imm 186,ND - SAL rm16,unity 8086,ND - SAL rm16,reg_cl 8086,ND - SAL rm16,imm 186,ND - SAL rm32,unity 386,ND - SAL rm32,reg_cl 386,ND - SAL rm32,imm 386,ND - SAL rm64,unity X64,ND - SAL rm64,reg_cl X64,ND - SAL rm64,imm X64,ND - SALC 8086,UNDOC - SAR rm8,unity 8086 - SAR rm8,reg_cl 8086 - SAR rm8,imm 186 - SAR rm16,unity 8086 - SAR rm16,reg_cl 8086 - SAR rm16,imm 186 - SAR rm32,unity 386 - SAR rm32,reg_cl 386 - SAR rm32,imm 386 - SAR rm64,unity X64 - SAR rm64,reg_cl X64 - SAR rm64,imm X64 - SBB mem,reg8 8086 - SBB reg8,reg8 8086 - SBB mem,reg16 8086 - SBB reg16,reg16 8086 - SBB mem,reg32 386 - SBB reg32,reg32 386 - SBB mem,reg64 X64 - SBB reg64,reg64 X64 - SBB reg8,mem 8086 - SBB reg8,reg8 8086 - SBB reg16,mem 8086 - SBB reg16,reg16 8086 - SBB reg32,mem 386 - SBB reg32,reg32 386 - SBB reg64,mem X64 - SBB reg64,reg64 X64 - SBB rm16,imm8 8086 - SBB rm32,imm8 386 - SBB rm64,imm8 X64 - SBB reg_al,imm 8086 - SBB reg_ax,sbyte16 8086 - SBB reg_ax,imm 8086 - SBB reg_eax,sbyte32 386 - SBB reg_eax,imm 386 - SBB reg_rax,sbyte64 X64 - SBB reg_rax,imm X64 - SBB rm8,imm 8086 - SBB rm16,imm 8086 - SBB rm32,imm 386 - SBB rm64,imm X64 - SBB mem,imm8 8086 - SBB mem,imm16 8086 - SBB mem,imm32 386 - SCASB 8086 - SCASD 386 - SCASQ X64 - SCASW 8086 - SFENCE X64,AMD - SGDT mem 286 - SHL rm8,unity 8086 - SHL rm8,reg_cl 8086 - SHL rm8,imm 186 - SHL rm16,unity 8086 - SHL rm16,reg_cl 8086 - SHL rm16,imm 186 - SHL rm32,unity 386 - SHL rm32,reg_cl 386 - SHL rm32,imm 386 - SHL rm64,unity X64 - SHL rm64,reg_cl X64 - SHL rm64,imm X64 - SHLD mem,reg16,imm 3862 - SHLD reg16,reg16,imm 3862 - SHLD mem,reg32,imm 3862 - SHLD reg32,reg32,imm 3862 - SHLD mem,reg64,imm X642 - SHLD reg64,reg64,imm X642 - SHLD mem,reg16,reg_cl 386 - SHLD reg16,reg16,reg_cl 386 - SHLD mem,reg32,reg_cl 386 - SHLD reg32,reg32,reg_cl 386 - SHLD mem,reg64,reg_cl X64 - SHLD reg64,reg64,reg_cl X64 - SHR rm8,unity 8086 - SHR rm8,reg_cl 8086 - SHR rm8,imm 186 - SHR rm16,unity 8086 - SHR rm16,reg_cl 8086 - SHR rm16,imm 186 - SHR rm32,unity 386 - SHR rm32,reg_cl 386 - SHR rm32,imm 386 - SHR rm64,unity X64 - SHR rm64,reg_cl X64 - SHR rm64,imm X64 - SHRD mem,reg16,imm 3862 - SHRD reg16,reg16,imm 3862 - SHRD mem,reg32,imm 3862 - SHRD reg32,reg32,imm 3862 - SHRD mem,reg64,imm X642 - SHRD reg64,reg64,imm X642 - SHRD mem,reg16,reg_cl 386 - SHRD reg16,reg16,reg_cl 386 - SHRD mem,reg32,reg_cl 386 - SHRD reg32,reg32,reg_cl 386 - SHRD mem,reg64,reg_cl X64 - SHRD reg64,reg64,reg_cl X64 - SIDT mem 286 - SLDT mem 286 - SLDT mem16 286 - SLDT reg16 286 - SLDT reg32 386 - SLDT reg64 X64,ND - SLDT reg64 X64 - SKINIT X64 - SMI 386,UNDOC - SMINT P6,CYRIX,ND - SMINTOLD 486,CYRIX,ND - SMSW mem 286 - SMSW mem16 286 - SMSW reg16 286 - SMSW reg32 386 - STC 8086 - STD 8086 - STGI X64 - STI 8086 - STOSB 8086 - STOSD 386 - STOSQ X64 - STOSW 8086 - STR mem 286,PROT - STR mem16 286,PROT - STR reg16 286,PROT - STR reg32 386,PROT - STR reg64 X64 - SUB mem,reg8 8086 - SUB reg8,reg8 8086 - SUB mem,reg16 8086 - SUB reg16,reg16 8086 - SUB mem,reg32 386 - SUB reg32,reg32 386 - SUB mem,reg64 X64 - SUB reg64,reg64 X64 - SUB reg8,mem 8086 - SUB reg8,reg8 8086 - SUB reg16,mem 8086 - SUB reg16,reg16 8086 - SUB reg32,mem 386 - SUB reg32,reg32 386 - SUB reg64,mem X64 - SUB reg64,reg64 X64 - SUB rm16,imm8 8086 - SUB rm32,imm8 386 - SUB rm64,imm8 X64 - SUB reg_al,imm 8086 - SUB reg_ax,sbyte16 8086 - SUB reg_ax,imm 8086 - SUB reg_eax,sbyte32 386 - SUB reg_eax,imm 386 - SUB reg_rax,sbyte64 X64 - SUB reg_rax,imm X64 - SUB rm8,imm 8086 - SUB rm16,imm 8086 - SUB rm32,imm 386 - SUB rm64,imm X64 - SUB mem,imm8 8086 - SUB mem,imm16 8086 - SUB mem,imm32 386 - SVDC mem80,reg_sreg 486,CYRIXM - SVLDT mem80 486,CYRIXM,ND - SVTS mem80 486,CYRIXM - SWAPGS X64 - SYSCALL P6,AMD - SYSENTER P6 - SYSEXIT P6,PRIV - SYSRET P6,PRIV,AMD - TEST mem,reg8 8086 - TEST reg8,reg8 8086 - TEST mem,reg16 8086 - TEST reg16,reg16 8086 - TEST mem,reg32 386 - TEST reg32,reg32 386 - TEST mem,reg64 X64 - TEST reg64,reg64 X64 - TEST reg8,mem 8086 - TEST reg16,mem 8086 - TEST reg32,mem 386 - TEST reg64,mem X64 - TEST reg_al,imm 8086 - TEST reg_ax,imm 8086 - TEST reg_eax,imm 386 - TEST reg_rax,imm X64 - TEST rm8,imm 8086 - TEST rm16,imm 8086 - TEST rm32,imm 386 - TEST rm64,imm X64 - TEST mem,imm8 8086 - TEST mem,imm16 8086 - TEST mem,imm32 386 - UD0 186,UNDOC - UD1 186,UNDOC - UD2B 186,UNDOC,ND - UD2 186 - UD2A 186,ND - UMOV mem,reg8 386,UNDOC,ND - UMOV reg8,reg8 386,UNDOC,ND - UMOV mem,reg16 386,UNDOC,ND - UMOV reg16,reg16 386,UNDOC,ND - UMOV mem,reg32 386,UNDOC,ND - UMOV reg32,reg32 386,UNDOC,ND - UMOV reg8,mem 386,UNDOC,ND - UMOV reg8,reg8 386,UNDOC,ND - UMOV reg16,mem 386,UNDOC,ND - UMOV reg16,reg16 386,UNDOC,ND - UMOV reg32,mem 386,UNDOC,ND - UMOV reg32,reg32 386,UNDOC,ND - VERR mem 286,PROT - VERR mem16 286,PROT - VERR reg16 286,PROT - VERW mem 286,PROT - VERW mem16 286,PROT - VERW reg16 286,PROT - FWAIT 8086 - WBINVD 486,PRIV - WRSHR rm32 P6,CYRIXM - WRMSR PENT,PRIV - XADD mem,reg8 486 - XADD reg8,reg8 486 - XADD mem,reg16 486 - XADD reg16,reg16 486 - XADD mem,reg32 486 - XADD reg32,reg32 486 - XADD mem,reg64 X64 - XADD reg64,reg64 X64 - XBTS reg16,mem 386,SW,UNDOC,ND - XBTS reg16,reg16 386,UNDOC,ND - XBTS reg32,mem 386,SD,UNDOC,ND - XBTS reg32,reg32 386,UNDOC,ND - XCHG reg_ax,reg16 8086 - XCHG reg_eax,reg32na 386 - XCHG reg_rax,reg64 X64 - XCHG reg16,reg_ax 8086 - XCHG reg32na,reg_eax 386 - XCHG reg64,reg_rax X64 - XCHG reg_eax,reg_eax 386,NOLONG - XCHG reg8,mem 8086 - XCHG reg8,reg8 8086 - XCHG reg16,mem 8086 - XCHG reg16,reg16 8086 - XCHG reg32,mem 386 - XCHG reg32,reg32 386 - XCHG reg64,mem X64 - XCHG reg64,reg64 X64 - XCHG mem,reg8 8086 - XCHG reg8,reg8 8086 - XCHG mem,reg16 8086 - XCHG reg16,reg16 8086 - XCHG mem,reg32 386 - XCHG reg32,reg32 386 - XCHG mem,reg64 X64 - XCHG reg64,reg64 X64 - XLATB 8086 - XLAT 8086 - XOR mem,reg8 8086 - XOR reg8,reg8 8086 - XOR mem,reg16 8086 - XOR reg16,reg16 8086 - XOR mem,reg32 386 - XOR reg32,reg32 386 - XOR mem,reg64 X64 - XOR reg64,reg64 X64 - XOR reg8,mem 8086 - XOR reg8,reg8 8086 - XOR reg16,mem 8086 - XOR reg16,reg16 8086 - XOR reg32,mem 386 - XOR reg32,reg32 386 - XOR reg64,mem X64 - XOR reg64,reg64 X64 - XOR rm16,imm8 8086 - XOR rm32,imm8 386 - XOR rm64,imm8 X64 - XOR reg_al,imm 8086 - XOR reg_ax,sbyte16 8086 - XOR reg_ax,imm 8086 - XOR reg_eax,sbyte32 386 - XOR reg_eax,imm 386 - XOR reg_rax,sbyte64 X64 - XOR reg_rax,imm X64 - XOR rm8,imm 8086 - XOR rm16,imm 8086 - XOR rm32,imm 386 - XOR rm64,imm X64 - XOR mem,imm8 8086 - XOR mem,imm16 8086 - XOR mem,imm32 386 - CMOVcc reg16,mem P6 - CMOVcc reg16,reg16 P6 - CMOVcc reg32,mem P6 - CMOVcc reg32,reg32 P6 - CMOVcc reg64,mem X64 - CMOVcc reg64,reg64 X64 - Jcc imm|near 386 - Jcc imm16|near 386 - Jcc imm32|near 386 - Jcc imm|short 8086,ND - Jcc imm 8086,ND - Jcc imm 386,ND - Jcc imm 8086,ND - Jcc imm 8086 - SETcc mem 386 - SETcc reg8 386 - - -File: nasm.info, Node: Section B.1.3, Next: Section B.1.4, Prev: Section B.1.2, Up: Section B.1 - -B.1.3. Katmai Streaming SIMD instructions (SSE --- a.k.a. KNI, XMM, MMX2) -------------------------------------------------------------------------- - - ADDPS xmmreg,xmmrm KATMAI,SSE - ADDSS xmmreg,xmmrm KATMAI,SSE,SD - ANDNPS xmmreg,xmmrm KATMAI,SSE - ANDPS xmmreg,xmmrm KATMAI,SSE - CMPEQPS xmmreg,xmmrm KATMAI,SSE - CMPEQSS xmmreg,xmmrm KATMAI,SSE - CMPLEPS xmmreg,xmmrm KATMAI,SSE - CMPLESS xmmreg,xmmrm KATMAI,SSE - CMPLTPS xmmreg,xmmrm KATMAI,SSE - CMPLTSS xmmreg,xmmrm KATMAI,SSE - CMPNEQPS xmmreg,xmmrm KATMAI,SSE - CMPNEQSS xmmreg,xmmrm KATMAI,SSE - CMPNLEPS xmmreg,xmmrm KATMAI,SSE - CMPNLESS xmmreg,xmmrm KATMAI,SSE - CMPNLTPS xmmreg,xmmrm KATMAI,SSE - CMPNLTSS xmmreg,xmmrm KATMAI,SSE - CMPORDPS xmmreg,xmmrm KATMAI,SSE - CMPORDSS xmmreg,xmmrm KATMAI,SSE - CMPUNORDPS xmmreg,xmmrm KATMAI,SSE - CMPUNORDSS xmmreg,xmmrm KATMAI,SSE - CMPPS xmmreg,mem,imm KATMAI,SSE - CMPPS xmmreg,xmmreg,imm KATMAI,SSE - CMPSS xmmreg,mem,imm KATMAI,SSE - CMPSS xmmreg,xmmreg,imm KATMAI,SSE - COMISS xmmreg,xmmrm KATMAI,SSE - CVTPI2PS xmmreg,mmxrm KATMAI,SSE,MMX - CVTPS2PI mmxreg,xmmrm KATMAI,SSE,MMX - CVTSI2SS xmmreg,mem KATMAI,SSE,SD,AR1,ND - CVTSI2SS xmmreg,rm32 KATMAI,SSE,SD,AR1 - CVTSI2SS xmmreg,rm64 X64,SSE,AR1 - CVTSS2SI reg32,xmmreg KATMAI,SSE,SD,AR1 - CVTSS2SI reg32,mem KATMAI,SSE,SD,AR1 - CVTSS2SI reg64,xmmreg X64,SSE,SD,AR1 - CVTSS2SI reg64,mem X64,SSE,SD,AR1 - CVTTPS2PI mmxreg,xmmrm KATMAI,SSE,MMX - CVTTSS2SI reg32,xmmrm KATMAI,SSE,SD,AR1 - CVTTSS2SI reg64,xmmrm X64,SSE,SD,AR1 - DIVPS xmmreg,xmmrm KATMAI,SSE - DIVSS xmmreg,xmmrm KATMAI,SSE - LDMXCSR mem KATMAI,SSE,SD - MAXPS xmmreg,xmmrm KATMAI,SSE - MAXSS xmmreg,xmmrm KATMAI,SSE - MINPS xmmreg,xmmrm KATMAI,SSE - MINSS xmmreg,xmmrm KATMAI,SSE - MOVAPS xmmreg,mem KATMAI,SSE - MOVAPS mem,xmmreg KATMAI,SSE - MOVAPS xmmreg,xmmreg KATMAI,SSE - MOVAPS xmmreg,xmmreg KATMAI,SSE - MOVHPS xmmreg,mem KATMAI,SSE - MOVHPS mem,xmmreg KATMAI,SSE - MOVLHPS xmmreg,xmmreg KATMAI,SSE - MOVLPS xmmreg,mem KATMAI,SSE - MOVLPS mem,xmmreg KATMAI,SSE - MOVHLPS xmmreg,xmmreg KATMAI,SSE - MOVMSKPS reg32,xmmreg KATMAI,SSE - MOVMSKPS reg64,xmmreg X64,SSE - MOVNTPS mem,xmmreg KATMAI,SSE - MOVSS xmmreg,mem KATMAI,SSE - MOVSS mem,xmmreg KATMAI,SSE - MOVSS xmmreg,xmmreg KATMAI,SSE - MOVSS xmmreg,xmmreg KATMAI,SSE - MOVUPS xmmreg,mem KATMAI,SSE - MOVUPS mem,xmmreg KATMAI,SSE - MOVUPS xmmreg,xmmreg KATMAI,SSE - MOVUPS xmmreg,xmmreg KATMAI,SSE - MULPS xmmreg,xmmrm KATMAI,SSE - MULSS xmmreg,xmmrm KATMAI,SSE - ORPS xmmreg,xmmrm KATMAI,SSE - RCPPS xmmreg,xmmrm KATMAI,SSE - RCPSS xmmreg,xmmrm KATMAI,SSE - RSQRTPS xmmreg,xmmrm KATMAI,SSE - RSQRTSS xmmreg,xmmrm KATMAI,SSE - SHUFPS xmmreg,mem,imm KATMAI,SSE - SHUFPS xmmreg,xmmreg,imm KATMAI,SSE - SQRTPS xmmreg,xmmrm KATMAI,SSE - SQRTSS xmmreg,xmmrm KATMAI,SSE - STMXCSR mem KATMAI,SSE,SD - SUBPS xmmreg,xmmrm KATMAI,SSE - SUBSS xmmreg,xmmrm KATMAI,SSE - UCOMISS xmmreg,xmmrm KATMAI,SSE - UNPCKHPS xmmreg,xmmrm KATMAI,SSE - UNPCKLPS xmmreg,xmmrm KATMAI,SSE - XORPS xmmreg,xmmrm KATMAI,SSE - - -File: nasm.info, Node: Section B.1.4, Next: Section B.1.5, Prev: Section B.1.3, Up: Section B.1 - -B.1.4. Introduced in Deschutes but necessary for SSE support ------------------------------------------------------------- - - FXRSTOR mem P6,SSE,FPU - FXSAVE mem P6,SSE,FPU - - -File: nasm.info, Node: Section B.1.5, Next: Section B.1.6, Prev: Section B.1.4, Up: Section B.1 - -B.1.5. XSAVE group (AVX and extended state) -------------------------------------------- - - XGETBV NEHALEM - XSETBV NEHALEM,PRIV - XSAVE mem NEHALEM - XRSTOR mem NEHALEM - - -File: nasm.info, Node: Section B.1.6, Next: Section B.1.7, Prev: Section B.1.5, Up: Section B.1 - -B.1.6. Generic memory operations --------------------------------- - - PREFETCHNTA mem KATMAI - PREFETCHT0 mem KATMAI - PREFETCHT1 mem KATMAI - PREFETCHT2 mem KATMAI - SFENCE KATMAI - - -File: nasm.info, Node: Section B.1.7, Next: Section B.1.8, Prev: Section B.1.6, Up: Section B.1 - -B.1.7. New MMX instructions introduced in Katmai ------------------------------------------------- - - MASKMOVQ mmxreg,mmxreg KATMAI,MMX - MOVNTQ mem,mmxreg KATMAI,MMX - PAVGB mmxreg,mmxrm KATMAI,MMX - PAVGW mmxreg,mmxrm KATMAI,MMX - PEXTRW reg32,mmxreg,imm KATMAI,MMX - PINSRW mmxreg,mem,imm KATMAI,MMX - PINSRW mmxreg,rm16,imm KATMAI,MMX - PINSRW mmxreg,reg32,imm KATMAI,MMX - PMAXSW mmxreg,mmxrm KATMAI,MMX - PMAXUB mmxreg,mmxrm KATMAI,MMX - PMINSW mmxreg,mmxrm KATMAI,MMX - PMINUB mmxreg,mmxrm KATMAI,MMX - PMOVMSKB reg32,mmxreg KATMAI,MMX - PMULHUW mmxreg,mmxrm KATMAI,MMX - PSADBW mmxreg,mmxrm KATMAI,MMX - PSHUFW mmxreg,mmxrm,imm KATMAI,MMX2 - - -File: nasm.info, Node: Section B.1.8, Next: Section B.1.9, Prev: Section B.1.7, Up: Section B.1 - -B.1.8. AMD Enhanced 3DNow! (Athlon) instructions ------------------------------------------------- - - PF2IW mmxreg,mmxrm PENT,3DNOW - PFNACC mmxreg,mmxrm PENT,3DNOW - PFPNACC mmxreg,mmxrm PENT,3DNOW - PI2FW mmxreg,mmxrm PENT,3DNOW - PSWAPD mmxreg,mmxrm PENT,3DNOW - - -File: nasm.info, Node: Section B.1.9, Next: Section B.1.10, Prev: Section B.1.8, Up: Section B.1 - -B.1.9. Willamette SSE2 Cacheability Instructions ------------------------------------------------- - - MASKMOVDQU xmmreg,xmmreg WILLAMETTE,SSE2 - CLFLUSH mem WILLAMETTE,SSE2 - MOVNTDQ mem,xmmreg WILLAMETTE,SSE2,SO - MOVNTI mem,reg32 WILLAMETTE,SD - MOVNTI mem,reg64 X64 - MOVNTPD mem,xmmreg WILLAMETTE,SSE2,SO - LFENCE WILLAMETTE,SSE2 - MFENCE WILLAMETTE,SSE2 - - -File: nasm.info, Node: Section B.1.10, Next: Section B.1.11, Prev: Section B.1.9, Up: Section B.1 - -B.1.10. Willamette MMX instructions (SSE2 SIMD Integer Instructions) --------------------------------------------------------------------- - - MOVD mem,xmmreg WILLAMETTE,SSE2,SD - MOVD xmmreg,mem WILLAMETTE,SSE2,SD - MOVD xmmreg,rm32 WILLAMETTE,SSE2 - MOVD rm32,xmmreg WILLAMETTE,SSE2 - MOVDQA xmmreg,xmmreg WILLAMETTE,SSE2 - MOVDQA mem,xmmreg WILLAMETTE,SSE2,SO - MOVDQA xmmreg,mem WILLAMETTE,SSE2,SO - MOVDQA xmmreg,xmmreg WILLAMETTE,SSE2 - MOVDQU xmmreg,xmmreg WILLAMETTE,SSE2 - MOVDQU mem,xmmreg WILLAMETTE,SSE2,SO - MOVDQU xmmreg,mem WILLAMETTE,SSE2,SO - MOVDQU xmmreg,xmmreg WILLAMETTE,SSE2 - MOVDQ2Q mmxreg,xmmreg WILLAMETTE,SSE2 - MOVQ xmmreg,xmmreg WILLAMETTE,SSE2 - MOVQ xmmreg,xmmreg WILLAMETTE,SSE2 - MOVQ mem,xmmreg WILLAMETTE,SSE2 - MOVQ xmmreg,mem WILLAMETTE,SSE2 - MOVQ xmmreg,rm64 X64,SSE2 - MOVQ rm64,xmmreg X64,SSE2 - MOVQ2DQ xmmreg,mmxreg WILLAMETTE,SSE2 - PACKSSWB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PACKSSDW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PACKUSWB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PADDB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PADDW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PADDD xmmreg,xmmrm WILLAMETTE,SSE2,SO - PADDQ mmxreg,mmxrm WILLAMETTE,MMX - PADDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - PADDSB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PADDSW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PADDUSB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PADDUSW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PAND xmmreg,xmmrm WILLAMETTE,SSE2,SO - PANDN xmmreg,xmmrm WILLAMETTE,SSE2,SO - PAVGB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PAVGW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PCMPEQB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PCMPEQW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PCMPEQD xmmreg,xmmrm WILLAMETTE,SSE2,SO - PCMPGTB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PCMPGTW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PCMPGTD xmmreg,xmmrm WILLAMETTE,SSE2,SO - PEXTRW reg32,xmmreg,imm WILLAMETTE,SSE2 - PINSRW xmmreg,reg16,imm WILLAMETTE,SSE2 - PINSRW xmmreg,reg32,imm WILLAMETTE,SSE2,ND - PINSRW xmmreg,mem,imm WILLAMETTE,SSE2 - PINSRW xmmreg,mem16,imm WILLAMETTE,SSE2 - PMADDWD xmmreg,xmmrm WILLAMETTE,SSE2,SO - PMAXSW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PMAXUB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PMINSW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PMINUB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PMOVMSKB reg32,xmmreg WILLAMETTE,SSE2 - PMULHUW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PMULHW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PMULLW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PMULUDQ mmxreg,mmxrm WILLAMETTE,SSE2,SO - PMULUDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - POR xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSADBW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSHUFD xmmreg,xmmreg,imm WILLAMETTE,SSE2 - PSHUFD xmmreg,mem,imm WILLAMETTE,SSE22 - PSHUFHW xmmreg,xmmreg,imm WILLAMETTE,SSE2 - PSHUFHW xmmreg,mem,imm WILLAMETTE,SSE22 - PSHUFLW xmmreg,xmmreg,imm WILLAMETTE,SSE2 - PSHUFLW xmmreg,mem,imm WILLAMETTE,SSE22 - PSLLDQ xmmreg,imm WILLAMETTE,SSE2,AR1 - PSLLW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSLLW xmmreg,imm WILLAMETTE,SSE2,AR1 - PSLLD xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSLLD xmmreg,imm WILLAMETTE,SSE2,AR1 - PSLLQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSLLQ xmmreg,imm WILLAMETTE,SSE2,AR1 - PSRAW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSRAW xmmreg,imm WILLAMETTE,SSE2,AR1 - PSRAD xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSRAD xmmreg,imm WILLAMETTE,SSE2,AR1 - PSRLDQ xmmreg,imm WILLAMETTE,SSE2,AR1 - PSRLW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSRLW xmmreg,imm WILLAMETTE,SSE2,AR1 - PSRLD xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSRLD xmmreg,imm WILLAMETTE,SSE2,AR1 - PSRLQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSRLQ xmmreg,imm WILLAMETTE,SSE2,AR1 - PSUBB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSUBW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSUBD xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSUBQ mmxreg,mmxrm WILLAMETTE,SSE2,SO - PSUBQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSUBSB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSUBSW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSUBUSB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSUBUSW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PUNPCKHBW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PUNPCKHWD xmmreg,xmmrm WILLAMETTE,SSE2,SO - PUNPCKHDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - PUNPCKHQDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - PUNPCKLBW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PUNPCKLWD xmmreg,xmmrm WILLAMETTE,SSE2,SO - PUNPCKLDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - PUNPCKLQDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - PXOR xmmreg,xmmrm WILLAMETTE,SSE2,SO - - -File: nasm.info, Node: Section B.1.11, Next: Section B.1.12, Prev: Section B.1.10, Up: Section B.1 - -B.1.11. Willamette Streaming SIMD instructions (SSE2) ------------------------------------------------------ - - ADDPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - ADDSD xmmreg,xmmrm WILLAMETTE,SSE2 - ANDNPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - ANDPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - CMPEQPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - CMPEQSD xmmreg,xmmrm WILLAMETTE,SSE2 - CMPLEPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - CMPLESD xmmreg,xmmrm WILLAMETTE,SSE2 - CMPLTPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - CMPLTSD xmmreg,xmmrm WILLAMETTE,SSE2 - CMPNEQPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - CMPNEQSD xmmreg,xmmrm WILLAMETTE,SSE2 - CMPNLEPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - CMPNLESD xmmreg,xmmrm WILLAMETTE,SSE2 - CMPNLTPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - CMPNLTSD xmmreg,xmmrm WILLAMETTE,SSE2 - CMPORDPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - CMPORDSD xmmreg,xmmrm WILLAMETTE,SSE2 - CMPUNORDPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - CMPUNORDSD xmmreg,xmmrm WILLAMETTE,SSE2 - CMPPD xmmreg,xmmrm,imm WILLAMETTE,SSE22 - CMPSD xmmreg,xmmrm,imm WILLAMETTE,SSE2 - COMISD xmmreg,xmmrm WILLAMETTE,SSE2 - CVTDQ2PD xmmreg,xmmrm WILLAMETTE,SSE2 - CVTDQ2PS xmmreg,xmmrm WILLAMETTE,SSE2,SO - CVTPD2DQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - CVTPD2PI mmxreg,xmmrm WILLAMETTE,SSE2,SO - CVTPD2PS xmmreg,xmmrm WILLAMETTE,SSE2,SO - CVTPI2PD xmmreg,mmxrm WILLAMETTE,SSE2 - CVTPS2DQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - CVTPS2PD xmmreg,xmmrm WILLAMETTE,SSE2 - CVTSD2SI reg32,xmmreg WILLAMETTE,SSE2,AR1 - CVTSD2SI reg32,mem WILLAMETTE,SSE2,AR1 - CVTSD2SI reg64,xmmreg X64,SSE2,AR1 - CVTSD2SI reg64,mem X64,SSE2,AR1 - CVTSD2SS xmmreg,xmmrm WILLAMETTE,SSE2 - CVTSI2SD xmmreg,mem WILLAMETTE,SSE2,SD,AR1,ND - CVTSI2SD xmmreg,rm32 WILLAMETTE,SSE2,SD,AR1 - CVTSI2SD xmmreg,rm64 X64,SSE2,AR1 - CVTSS2SD xmmreg,xmmrm WILLAMETTE,SSE2,SD - CVTTPD2PI mmxreg,xmmrm WILLAMETTE,SSE2,SO - CVTTPD2DQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - CVTTPS2DQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - CVTTSD2SI reg32,xmmreg WILLAMETTE,SSE2,AR1 - CVTTSD2SI reg32,mem WILLAMETTE,SSE2,AR1 - CVTTSD2SI reg64,xmmreg X64,SSE2,AR1 - CVTTSD2SI reg64,mem X64,SSE2,AR1 - DIVPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - DIVSD xmmreg,xmmrm WILLAMETTE,SSE2 - MAXPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - MAXSD xmmreg,xmmrm WILLAMETTE,SSE2 - MINPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - MINSD xmmreg,xmmrm WILLAMETTE,SSE2 - MOVAPD xmmreg,xmmreg WILLAMETTE,SSE2 - MOVAPD xmmreg,xmmreg WILLAMETTE,SSE2 - MOVAPD mem,xmmreg WILLAMETTE,SSE2,SO - MOVAPD xmmreg,mem WILLAMETTE,SSE2,SO - MOVHPD mem,xmmreg WILLAMETTE,SSE2 - MOVHPD xmmreg,mem WILLAMETTE,SSE2 - MOVLPD mem,xmmreg WILLAMETTE,SSE2 - MOVLPD xmmreg,mem WILLAMETTE,SSE2 - MOVMSKPD reg32,xmmreg WILLAMETTE,SSE2 - MOVMSKPD reg64,xmmreg X64,SSE2 - MOVSD xmmreg,xmmreg WILLAMETTE,SSE2 - MOVSD xmmreg,xmmreg WILLAMETTE,SSE2 - MOVSD mem,xmmreg WILLAMETTE,SSE2 - MOVSD xmmreg,mem WILLAMETTE,SSE2 - MOVUPD xmmreg,xmmreg WILLAMETTE,SSE2 - MOVUPD xmmreg,xmmreg WILLAMETTE,SSE2 - MOVUPD mem,xmmreg WILLAMETTE,SSE2,SO - MOVUPD xmmreg,mem WILLAMETTE,SSE2,SO - MULPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - MULSD xmmreg,xmmrm WILLAMETTE,SSE2 - ORPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - SHUFPD xmmreg,xmmreg,imm WILLAMETTE,SSE2 - SHUFPD xmmreg,mem,imm WILLAMETTE,SSE2 - SQRTPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - SQRTSD xmmreg,xmmrm WILLAMETTE,SSE2 - SUBPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - SUBSD xmmreg,xmmrm WILLAMETTE,SSE2 - UCOMISD xmmreg,xmmrm WILLAMETTE,SSE2 - UNPCKHPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - UNPCKLPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - XORPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - - -File: nasm.info, Node: Section B.1.12, Next: Section B.1.13, Prev: Section B.1.11, Up: Section B.1 - -B.1.12. Prescott New Instructions (SSE3) ----------------------------------------- - - ADDSUBPD xmmreg,xmmrm PRESCOTT,SSE3,SO - ADDSUBPS xmmreg,xmmrm PRESCOTT,SSE3,SO - HADDPD xmmreg,xmmrm PRESCOTT,SSE3,SO - HADDPS xmmreg,xmmrm PRESCOTT,SSE3,SO - HSUBPD xmmreg,xmmrm PRESCOTT,SSE3,SO - HSUBPS xmmreg,xmmrm PRESCOTT,SSE3,SO - LDDQU xmmreg,mem PRESCOTT,SSE3,SO - MOVDDUP xmmreg,xmmrm PRESCOTT,SSE3 - MOVSHDUP xmmreg,xmmrm PRESCOTT,SSE3 - MOVSLDUP xmmreg,xmmrm PRESCOTT,SSE3 - - -File: nasm.info, Node: Section B.1.13, Next: Section B.1.14, Prev: Section B.1.12, Up: Section B.1 - -B.1.13. VMX Instructions ------------------------- - - VMCALL VMX - VMCLEAR mem VMX - VMLAUNCH VMX - VMLOAD X64,VMX - VMMCALL X64,VMX - VMPTRLD mem VMX - VMPTRST mem VMX - VMREAD rm32,reg32 VMX,NOLONG,SD - VMREAD rm64,reg64 X64,VMX - VMRESUME VMX - VMRUN X64,VMX - VMSAVE X64,VMX - VMWRITE reg32,rm32 VMX,NOLONG,SD - VMWRITE reg64,rm64 X64,VMX - VMXOFF VMX - VMXON mem VMX - - -File: nasm.info, Node: Section B.1.14, Next: Section B.1.15, Prev: Section B.1.13, Up: Section B.1 - -B.1.14. Extended Page Tables VMX instructions ---------------------------------------------- - - INVEPT reg32,mem VMX,SO,NOLONG - INVEPT reg64,mem VMX,SO,LONG - INVVPID reg32,mem VMX,SO,NOLONG - INVVPID reg64,mem VMX,SO,LONG - - -File: nasm.info, Node: Section B.1.15, Next: Section B.1.16, Prev: Section B.1.14, Up: Section B.1 - -B.1.15. Tejas New Instructions (SSSE3) --------------------------------------- - - PABSB mmxreg,mmxrm SSSE3,MMX - PABSB xmmreg,xmmrm SSSE3 - PABSW mmxreg,mmxrm SSSE3,MMX - PABSW xmmreg,xmmrm SSSE3 - PABSD mmxreg,mmxrm SSSE3,MMX - PABSD xmmreg,xmmrm SSSE3 - PALIGNR mmxreg,mmxrm,imm SSSE3,MMX - PALIGNR xmmreg,xmmrm,imm SSSE3 - PHADDW mmxreg,mmxrm SSSE3,MMX - PHADDW xmmreg,xmmrm SSSE3 - PHADDD mmxreg,mmxrm SSSE3,MMX - PHADDD xmmreg,xmmrm SSSE3 - PHADDSW mmxreg,mmxrm SSSE3,MMX - PHADDSW xmmreg,xmmrm SSSE3 - PHSUBW mmxreg,mmxrm SSSE3,MMX - PHSUBW xmmreg,xmmrm SSSE3 - PHSUBD mmxreg,mmxrm SSSE3,MMX - PHSUBD xmmreg,xmmrm SSSE3 - PHSUBSW mmxreg,mmxrm SSSE3,MMX - PHSUBSW xmmreg,xmmrm SSSE3 - PMADDUBSW mmxreg,mmxrm SSSE3,MMX - PMADDUBSW xmmreg,xmmrm SSSE3 - PMULHRSW mmxreg,mmxrm SSSE3,MMX - PMULHRSW xmmreg,xmmrm SSSE3 - PSHUFB mmxreg,mmxrm SSSE3,MMX - PSHUFB xmmreg,xmmrm SSSE3 - PSIGNB mmxreg,mmxrm SSSE3,MMX - PSIGNB xmmreg,xmmrm SSSE3 - PSIGNW mmxreg,mmxrm SSSE3,MMX - PSIGNW xmmreg,xmmrm SSSE3 - PSIGND mmxreg,mmxrm SSSE3,MMX - PSIGND xmmreg,xmmrm SSSE3 - - -File: nasm.info, Node: Section B.1.16, Next: Section B.1.17, Prev: Section B.1.15, Up: Section B.1 - -B.1.16. AMD SSE4A ------------------ - - EXTRQ xmmreg,imm,imm SSE4A,AMD - EXTRQ xmmreg,xmmreg SSE4A,AMD - INSERTQ xmmreg,xmmreg,imm,imm SSE4A,AMD - INSERTQ xmmreg,xmmreg SSE4A,AMD - MOVNTSD mem,xmmreg SSE4A,AMD - MOVNTSS mem,xmmreg SSE4A,AMD,SD - - -File: nasm.info, Node: Section B.1.17, Next: Section B.1.18, Prev: Section B.1.16, Up: Section B.1 - -B.1.17. New instructions in Barcelona -------------------------------------- - - LZCNT reg16,rm16 P6,AMD - LZCNT reg32,rm32 P6,AMD - LZCNT reg64,rm64 X64,AMD - - -File: nasm.info, Node: Section B.1.18, Next: Section B.1.19, Prev: Section B.1.17, Up: Section B.1 - -B.1.18. Penryn New Instructions (SSE4.1) ----------------------------------------- - - BLENDPD xmmreg,xmmrm,imm SSE41 - BLENDPS xmmreg,xmmrm,imm SSE41 - BLENDVPD xmmreg,xmmrm,xmm0 SSE41 - BLENDVPS xmmreg,xmmrm,xmm0 SSE41 - DPPD xmmreg,xmmrm,imm SSE41 - DPPS xmmreg,xmmrm,imm SSE41 - EXTRACTPS rm32,xmmreg,imm SSE41 - EXTRACTPS reg64,xmmreg,imm SSE41,X64 - INSERTPS xmmreg,xmmrm,imm SSE41,SD - MOVNTDQA xmmreg,mem SSE41 - MPSADBW xmmreg,xmmrm,imm SSE41 - PACKUSDW xmmreg,xmmrm SSE41 - PBLENDVB xmmreg,xmmrm,xmm0 SSE41 - PBLENDW xmmreg,xmmrm,imm SSE41 - PCMPEQQ xmmreg,xmmrm SSE41 - PEXTRB reg32,xmmreg,imm SSE41 - PEXTRB mem8,xmmreg,imm SSE41 - PEXTRB reg64,xmmreg,imm SSE41,X64 - PEXTRD rm32,xmmreg,imm SSE41 - PEXTRQ rm64,xmmreg,imm SSE41,X64 - PEXTRW reg32,xmmreg,imm SSE41 - PEXTRW mem16,xmmreg,imm SSE41 - PEXTRW reg64,xmmreg,imm SSE41,X64 - PHMINPOSUW xmmreg,xmmrm SSE41 - PINSRB xmmreg,mem,imm SSE41 - PINSRB xmmreg,rm8,imm SSE41 - PINSRB xmmreg,reg32,imm SSE41 - PINSRD xmmreg,mem,imm SSE41 - PINSRD xmmreg,rm32,imm SSE41 - PINSRQ xmmreg,mem,imm SSE41,X64 - PINSRQ xmmreg,rm64,imm SSE41,X64 - PMAXSB xmmreg,xmmrm SSE41 - PMAXSD xmmreg,xmmrm SSE41 - PMAXUD xmmreg,xmmrm SSE41 - PMAXUW xmmreg,xmmrm SSE41 - PMINSB xmmreg,xmmrm SSE41 - PMINSD xmmreg,xmmrm SSE41 - PMINUD xmmreg,xmmrm SSE41 - PMINUW xmmreg,xmmrm SSE41 - PMOVSXBW xmmreg,xmmrm SSE41 - PMOVSXBD xmmreg,xmmrm SSE41,SD - PMOVSXBQ xmmreg,xmmrm SSE41,SW - PMOVSXWD xmmreg,xmmrm SSE41 - PMOVSXWQ xmmreg,xmmrm SSE41,SD - PMOVSXDQ xmmreg,xmmrm SSE41 - PMOVZXBW xmmreg,xmmrm SSE41 - PMOVZXBD xmmreg,xmmrm SSE41,SD - PMOVZXBQ xmmreg,xmmrm SSE41,SW - PMOVZXWD xmmreg,xmmrm SSE41 - PMOVZXWQ xmmreg,xmmrm SSE41,SD - PMOVZXDQ xmmreg,xmmrm SSE41 - PMULDQ xmmreg,xmmrm SSE41 - PMULLD xmmreg,xmmrm SSE41 - PTEST xmmreg,xmmrm SSE41 - ROUNDPD xmmreg,xmmrm,imm SSE41 - ROUNDPS xmmreg,xmmrm,imm SSE41 - ROUNDSD xmmreg,xmmrm,imm SSE41 - ROUNDSS xmmreg,xmmrm,imm SSE41 - - -File: nasm.info, Node: Section B.1.19, Next: Section B.1.20, Prev: Section B.1.18, Up: Section B.1 - -B.1.19. Nehalem New Instructions (SSE4.2) ------------------------------------------ - - CRC32 reg32,rm8 SSE42 - CRC32 reg32,rm16 SSE42 - CRC32 reg32,rm32 SSE42 - CRC32 reg64,rm8 SSE42,X64 - CRC32 reg64,rm64 SSE42,X64 - PCMPESTRI xmmreg,xmmrm,imm SSE42 - PCMPESTRM xmmreg,xmmrm,imm SSE42 - PCMPISTRI xmmreg,xmmrm,imm SSE42 - PCMPISTRM xmmreg,xmmrm,imm SSE42 - PCMPGTQ xmmreg,xmmrm SSE42 - POPCNT reg16,rm16 NEHALEM,SW - POPCNT reg32,rm32 NEHALEM,SD - POPCNT reg64,rm64 NEHALEM,X64 - - -File: nasm.info, Node: Section B.1.20, Next: Section B.1.21, Prev: Section B.1.19, Up: Section B.1 - -B.1.20. Intel SMX ------------------ - - GETSEC KATMAI - - -File: nasm.info, Node: Section B.1.21, Next: Section B.1.22, Prev: Section B.1.20, Up: Section B.1 - -B.1.21. Geode (Cyrix) 3DNow! additions --------------------------------------- - - PFRCPV mmxreg,mmxrm PENT,3DNOW,CYRIX - PFRSQRTV mmxreg,mmxrm PENT,3DNOW,CYRIX - - -File: nasm.info, Node: Section B.1.22, Next: Section B.1.23, Prev: Section B.1.21, Up: Section B.1 - -B.1.22. Intel new instructions in ??? -------------------------------------- - - MOVBE reg16,mem16 NEHALEM - MOVBE reg32,mem32 NEHALEM - MOVBE reg64,mem64 NEHALEM - MOVBE mem16,reg16 NEHALEM - MOVBE mem32,reg32 NEHALEM - MOVBE mem64,reg64 NEHALEM - - -File: nasm.info, Node: Section B.1.23, Next: Section B.1.24, Prev: Section B.1.22, Up: Section B.1 - -B.1.23. Intel AES instructions ------------------------------- - - AESENC xmmreg,xmmrm128 SSE,WESTMERE - AESENCLAST xmmreg,xmmrm128 SSE,WESTMERE - AESDEC xmmreg,xmmrm128 SSE,WESTMERE - AESDECLAST xmmreg,xmmrm128 SSE,WESTMERE - AESIMC xmmreg,xmmrm128 SSE,WESTMERE - AESKEYGENASSIST xmmreg,xmmrm128,imm8 SSE,WESTMERE - - -File: nasm.info, Node: Section B.1.24, Next: Section B.1.25, Prev: Section B.1.23, Up: Section B.1 - -B.1.24. Intel AVX AES instructions ----------------------------------- - - VAESENC xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VAESENCLAST xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VAESDEC xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VAESDECLAST xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VAESIMC xmmreg,xmmrm128 AVX,SANDYBRIDGE - VAESKEYGENASSIST xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - - -File: nasm.info, Node: Section B.1.25, Next: Section B.1.26, Prev: Section B.1.24, Up: Section B.1 - -B.1.25. Intel AVX instructions ------------------------------- - - VADDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VADDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VADDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VADDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VADDSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VADDSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VADDSUBPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VADDSUBPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VADDSUBPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VADDSUBPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VANDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VANDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VANDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VANDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VANDNPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VANDNPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VANDNPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VANDNPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VBLENDPD xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VBLENDPD ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE - VBLENDPS xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VBLENDPS ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE - VBLENDVPD xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE - VBLENDVPD xmmreg,xmmrm128,xmm0 AVX,SANDYBRIDGE - VBLENDVPD ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE - VBLENDVPD ymmreg,ymmrm256,ymm0 AVX,SANDYBRIDGE - VBLENDVPS xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE - VBLENDVPS xmmreg,xmmrm128,xmm0 AVX,SANDYBRIDGE - VBLENDVPS ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE - VBLENDVPD ymmreg,ymmrm256,ymm0 AVX,SANDYBRIDGE - VBROADCASTSS xmmreg,mem32 AVX,SANDYBRIDGE - VBROADCASTSS ymmreg,mem32 AVX,SANDYBRIDGE - VBROADCASTSD ymmreg,mem64 AVX,SANDYBRIDGE - VBROADCASTF128 ymmreg,mem128 AVX,SANDYBRIDGE - VCMPEQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPEQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPLTPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPLTPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPLEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPLEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPUNORDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPUNORDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNEQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNEQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNLTPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNLTPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNLEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNLEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPORDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPORDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPEQ_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPEQ_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNGEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNGEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNGTPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNGTPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPFALSEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPFALSEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNEQ_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNEQ_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPGEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPGEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPGTPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPGTPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPTRUEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPTRUEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPEQ_OSPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPEQ_OSPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPLT_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPLT_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPLE_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPLE_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPUNORD_SPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPUNORD_SPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNEQ_USPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNEQ_USPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNLT_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNLT_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNLE_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNLE_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPORD_SPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPORD_SPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPEQ_USPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPEQ_USPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNGE_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNGE_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNGT_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNGT_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPFALSE_OSPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPFALSE_OSPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNEQ_OSPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNEQ_OSPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPGE_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPGE_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPGT_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPGT_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPTRUE_USPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPTRUE_USPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPPD xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VCMPPD ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE - VCMPEQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPEQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPLTPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPLTPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPLEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPLEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPUNORDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPUNORDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNEQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNEQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNLTPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNLTPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNLEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNLEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPORDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPORDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPEQ_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPEQ_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNGEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNGEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNGTPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNGTPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPFALSEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPFALSEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNEQ_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNEQ_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPGEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPGEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPGTPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPGTPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPTRUEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPTRUEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPEQ_OSPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPEQ_OSPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPLT_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPLT_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPLE_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPLE_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPUNORD_SPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPUNORD_SPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNEQ_USPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNEQ_USPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNLT_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNLT_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNLE_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNLE_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPORD_SPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPORD_SPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPEQ_USPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPEQ_USPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNGE_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNGE_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNGT_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNGT_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPFALSE_OSPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPFALSE_OSPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNEQ_OSPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNEQ_OSPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPGE_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPGE_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPGT_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPGT_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPTRUE_USPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPTRUE_USPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPPS xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VCMPPS ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE - VCMPEQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPLTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPLESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPUNORDSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNEQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNLTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNLESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPORDSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPEQ_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNGESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNGTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPFALSESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNEQ_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPGESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPGTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPTRUESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPEQ_OSSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPLT_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPLE_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPUNORD_SSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNEQ_USSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNLT_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNLE_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPORD_SSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPEQ_USSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNGE_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNGT_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPFALSE_OSSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNEQ_OSSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPGE_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPGT_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPTRUE_USSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPSD xmmreg,xmmreg*,xmmrm64,imm8 AVX,SANDYBRIDGE - VCMPEQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPLTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPLESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPUNORDSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNEQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNLTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNLESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPORDSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPEQ_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNGESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNGTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPFALSESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNEQ_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPGESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPGTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPTRUESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPEQ_OSSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPLT_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPLE_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPUNORD_SSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNEQ_USSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNLT_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNLE_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPORD_SSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPEQ_USSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNGE_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNGT_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPFALSE_OSSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNEQ_OSSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPGE_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPGT_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPTRUE_USSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPSS xmmreg,xmmreg*,xmmrm32,imm8 AVX,SANDYBRIDGE - VCOMISD xmmreg,xmmrm64 AVX,SANDYBRIDGE - VCOMISS xmmreg,xmmrm32 AVX,SANDYBRIDGE - VCVTDQ2PD xmmreg,xmmrm64 AVX,SANDYBRIDGE - VCVTDQ2PD ymmreg,xmmrm128 AVX,SANDYBRIDGE - VCVTDQ2PS xmmreg,xmmrm128 AVX,SANDYBRIDGE - VCVTDQ2PS ymmreg,ymmrm256 AVX,SANDYBRIDGE - VCVTPD2DQ xmmreg,xmmreg AVX,SANDYBRIDGE - VCVTPD2DQ xmmreg,mem128 AVX,SANDYBRIDGE,SO - VCVTPD2DQ xmmreg,ymmreg AVX,SANDYBRIDGE - VCVTPD2DQ xmmreg,mem256 AVX,SANDYBRIDGE,SY - VCVTPD2PS xmmreg,xmmreg AVX,SANDYBRIDGE - VCVTPD2PS xmmreg,mem128 AVX,SANDYBRIDGE,SO - VCVTPD2PS xmmreg,ymmreg AVX,SANDYBRIDGE - VCVTPD2PS xmmreg,mem256 AVX,SANDYBRIDGE,SY - VCVTPS2DQ xmmreg,xmmrm128 AVX,SANDYBRIDGE - VCVTPS2DQ ymmreg,ymmrm256 AVX,SANDYBRIDGE - VCVTPS2PD xmmreg,xmmrm64 AVX,SANDYBRIDGE - VCVTPS2PD ymmreg,xmmrm128 AVX,SANDYBRIDGE - VCVTSD2SI reg32,xmmrm64 AVX,SANDYBRIDGE - VCVTSD2SI reg64,xmmrm64 AVX,SANDYBRIDGE,LONG - VCVTSD2SS xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCVTSI2SD xmmreg,xmmreg*,rm32 AVX,SANDYBRIDGE,SD - VCVTSI2SD xmmreg,xmmreg*,mem32 AVX,SANDYBRIDGE,ND,SD - VCVTSI2SD xmmreg,xmmreg*,rm64 AVX,SANDYBRIDGE,LONG - VCVTSI2SS xmmreg,xmmreg*,rm32 AVX,SANDYBRIDGE,SD - VCVTSI2SS xmmreg,xmmreg*,mem32 AVX,SANDYBRIDGE,ND,SD - VCVTSI2SS xmmreg,xmmreg*,rm64 AVX,SANDYBRIDGE,LONG - VCVTSS2SD xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCVTSS2SI reg32,xmmrm32 AVX,SANDYBRIDGE - VCVTSS2SI reg64,xmmrm32 AVX,SANDYBRIDGE,LONG - VCVTTPD2DQ xmmreg,xmmreg AVX,SANDYBRIDGE - VCVTTPD2DQ xmmreg,mem128 AVX,SANDYBRIDGE,SO - VCVTTPD2DQ xmmreg,ymmreg AVX,SANDYBRIDGE - VCVTTPD2DQ xmmreg,mem256 AVX,SANDYBRIDGE,SY - VCVTTPS2DQ xmmreg,xmmrm128 AVX,SANDYBRIDGE - VCVTTPS2DQ ymmreg,ymmrm256 AVX,SANDYBRIDGE - VCVTTSD2SI reg32,xmmrm64 AVX,SANDYBRIDGE - VCVTTSD2SI reg64,xmmrm64 AVX,SANDYBRIDGE,LONG - VCVTTSS2SI reg32,xmmrm32 AVX,SANDYBRIDGE - VCVTTSS2SI reg64,xmmrm32 AVX,SANDYBRIDGE,LONG - VDIVPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VDIVPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VDIVPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VDIVPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VDIVSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VDIVSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VDPPD xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VDPPS xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VDPPS ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE - VEXTRACTF128 xmmrm128,xmmreg,imm8 AVX,SANDYBRIDGE - VEXTRACTPS rm32,xmmreg,imm8 AVX,SANDYBRIDGE - VHADDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VHADDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VHADDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VHADDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VHSUBPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VHSUBPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VHSUBPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VHSUBPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VINSERTF128 ymmreg,ymmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VINSERTPS xmmreg,xmmreg*,xmmrm32,imm8 AVX,SANDYBRIDGE - VLDDQU xmmreg,mem128 AVX,SANDYBRIDGE - VLDQQU ymmreg,mem256 AVX,SANDYBRIDGE - VLDDQU ymmreg,mem256 AVX,SANDYBRIDGE - VLDMXCSR mem32 AVX,SANDYBRIDGE - VMASKMOVDQU xmmreg,xmmreg AVX,SANDYBRIDGE - VMASKMOVPS xmmreg,xmmreg,mem128 AVX,SANDYBRIDGE - VMASKMOVPS ymmreg,ymmreg,mem256 AVX,SANDYBRIDGE - VMASKMOVPS mem128,xmmreg,xmmreg AVX,SANDYBRIDGE,SO - VMASKMOVPS mem256,xmmreg,xmmreg AVX,SANDYBRIDGE,SY - VMASKMOVPD xmmreg,xmmreg,mem128 AVX,SANDYBRIDGE - VMASKMOVPD ymmreg,ymmreg,mem256 AVX,SANDYBRIDGE - VMASKMOVPD mem128,xmmreg,xmmreg AVX,SANDYBRIDGE - VMASKMOVPD mem256,ymmreg,ymmreg AVX,SANDYBRIDGE - VMAXPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VMAXPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VMAXPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VMAXPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VMAXSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VMAXSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VMINPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VMINPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VMINPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VMINPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VMINSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VMINSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VMOVAPD xmmreg,xmmrm128 AVX,SANDYBRIDGE - VMOVAPD xmmrm128,xmmreg AVX,SANDYBRIDGE - VMOVAPD ymmreg,ymmrm256 AVX,SANDYBRIDGE - VMOVAPD ymmrm256,ymmreg AVX,SANDYBRIDGE - VMOVAPS xmmreg,xmmrm128 AVX,SANDYBRIDGE - VMOVAPS xmmrm128,xmmreg AVX,SANDYBRIDGE - VMOVAPS ymmreg,ymmrm256 AVX,SANDYBRIDGE - VMOVAPS ymmrm256,ymmreg AVX,SANDYBRIDGE - VMOVQ xmmreg,xmmrm64 AVX,SANDYBRIDGE - VMOVQ xmmrm64,xmmreg AVX,SANDYBRIDGE - VMOVQ xmmreg,rm64 AVX,SANDYBRIDGE,LONG - VMOVQ rm64,xmmreg AVX,SANDYBRIDGE,LONG - VMOVD xmmreg,rm32 AVX,SANDYBRIDGE - VMOVD rm32,xmmreg AVX,SANDYBRIDGE - VMOVDDUP xmmreg,xmmrm64 AVX,SANDYBRIDGE - VMOVDDUP ymmreg,ymmrm256 AVX,SANDYBRIDGE - VMOVDQA xmmreg,xmmrm128 AVX,SANDYBRIDGE - VMOVDQA xmmrm128,xmmreg AVX,SANDYBRIDGE - VMOVQQA ymmreg,ymmrm256 AVX,SANDYBRIDGE - VMOVQQA ymmrm256,ymmreg AVX,SANDYBRIDGE - VMOVDQA ymmreg,ymmrm AVX,SANDYBRIDGE - VMOVDQA ymmrm256,ymmreg AVX,SANDYBRIDGE - VMOVDQU xmmreg,xmmrm128 AVX,SANDYBRIDGE - VMOVDQU xmmrm128,xmmreg AVX,SANDYBRIDGE - VMOVQQU ymmreg,ymmrm256 AVX,SANDYBRIDGE - VMOVQQU ymmrm256,ymmreg AVX,SANDYBRIDGE - VMOVDQU ymmreg,ymmrm256 AVX,SANDYBRIDGE - VMOVDQU ymmrm256,ymmreg AVX,SANDYBRIDGE - VMOVHLPS xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE - VMOVHPD xmmreg,xmmreg*,mem64 AVX,SANDYBRIDGE - VMOVHPD mem64,xmmreg AVX,SANDYBRIDGE - VMOVHPS xmmreg,xmmreg*,mem64 AVX,SANDYBRIDGE - VMOVHPS mem64,xmmreg AVX,SANDYBRIDGE - VMOVLHPS xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE - VMOVLPD xmmreg,xmmreg*,mem64 AVX,SANDYBRIDGE - VMOVLPD mem64,xmmreg AVX,SANDYBRIDGE - VMOVLPS xmmreg,xmmreg*,mem64 AVX,SANDYBRIDGE - VMOVLPS mem64,xmmreg AVX,SANDYBRIDGE - VMOVMSKPD reg64,xmmreg AVX,SANDYBRIDGE,LONG - VMOVMSKPD reg32,xmmreg AVX,SANDYBRIDGE - VMOVMSKPD reg64,ymmreg AVX,SANDYBRIDGE,LONG - VMOVMSKPD reg32,ymmreg AVX,SANDYBRIDGE - VMOVMSKPS reg64,xmmreg AVX,SANDYBRIDGE,LONG - VMOVMSKPS reg32,xmmreg AVX,SANDYBRIDGE - VMOVMSKPS reg64,ymmreg AVX,SANDYBRIDGE,LONG - VMOVMSKPS reg32,ymmreg AVX,SANDYBRIDGE - VMOVNTDQ mem128,xmmreg AVX,SANDYBRIDGE - VMOVNTQQ mem256,ymmreg AVX,SANDYBRIDGE - VMOVNTDQ mem256,ymmreg AVX,SANDYBRIDGE - VMOVNTDQA xmmreg,mem128 AVX,SANDYBRIDGE - VMOVNTPD mem128,xmmreg AVX,SANDYBRIDGE - VMOVNTPD mem256,ymmreg AVX,SANDYBRIDGE - VMOVNTPS mem128,xmmreg AVX,SANDYBRIDGE - VMOVNTPS mem128,ymmreg AVX,SANDYBRIDGE - VMOVSD xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE - VMOVSD xmmreg,mem64 AVX,SANDYBRIDGE - VMOVSD xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE - VMOVSD mem64,xmmreg AVX,SANDYBRIDGE - VMOVSHDUP xmmreg,xmmrm128 AVX,SANDYBRIDGE - VMOVSHDUP ymmreg,ymmrm256 AVX,SANDYBRIDGE - VMOVSLDUP xmmreg,xmmrm128 AVX,SANDYBRIDGE - VMOVSLDUP ymmreg,ymmrm256 AVX,SANDYBRIDGE - VMOVSS xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE - VMOVSS xmmreg,mem64 AVX,SANDYBRIDGE - VMOVSS xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE - VMOVSS mem64,xmmreg AVX,SANDYBRIDGE - VMOVUPD xmmreg,xmmrm128 AVX,SANDYBRIDGE - VMOVUPD xmmrm128,xmmreg AVX,SANDYBRIDGE - VMOVUPD ymmreg,ymmrm256 AVX,SANDYBRIDGE - VMOVUPD ymmrm256,ymmreg AVX,SANDYBRIDGE - VMOVUPS xmmreg,xmmrm128 AVX,SANDYBRIDGE - VMOVUPS xmmrm128,xmmreg AVX,SANDYBRIDGE - VMOVUPS ymmreg,ymmrm256 AVX,SANDYBRIDGE - VMOVUPS ymmrm256,ymmreg AVX,SANDYBRIDGE - VMPSADBW xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VMULPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VMULPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VMULPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VMULPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VMULSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VMULSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VORPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VORPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VORPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VORPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VPABSB xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPABSW xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPABSD xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPACKSSWB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPACKSSDW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPACKUSWB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPACKUSDW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPADDB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPADDW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPADDD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPADDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPADDSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPADDSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPADDUSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPADDUSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPALIGNR xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VPAND xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPANDN xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPAVGB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPAVGW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPBLENDVB xmmreg,xmmreg*,xmmrm128,xmmreg AVX,SANDYBRIDGE - VPBLENDW xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VPCMPESTRI xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPCMPESTRM xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPCMPISTRI xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPCMPISTRM xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPCMPEQB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCMPEQW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCMPEQD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCMPEQQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCMPGTB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCMPGTW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCMPGTD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCMPGTQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPERMILPD xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPERMILPD ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE - VPERMILPD xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPERMILPD ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE - VPERMILTD2PD xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE - VPERMILTD2PD xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPERMILTD2PD ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE - VPERMILTD2PD ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE - VPERMILMO2PD xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE - VPERMILMO2PD xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPERMILMO2PD ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE - VPERMILMO2PD ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE - VPERMILMZ2PD xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE - VPERMILMZ2PD xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPERMILMZ2PD ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE - VPERMILMZ2PD ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE - VPERMIL2PD xmmreg,xmmreg,xmmrm128,xmmreg,imm8 AVX,SANDYBRIDGE - VPERMIL2PD xmmreg,xmmreg,xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPERMIL2PD ymmreg,ymmreg,ymmrm256,ymmreg,imm8 AVX,SANDYBRIDGE - VPERMIL2PD ymmreg,ymmreg,ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE - VPERMILPS xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPERMILPS ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE - VPERMILPS xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPERMILPS ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE - VPERMILTD2PS xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE - VPERMILTD2PS xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPERMILTD2PS ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE - VPERMILTD2PS ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE - VPERMILMO2PS xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE - VPERMILMO2PS xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPERMILMO2PS ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE - VPERMILMO2PS ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE - VPERMILMZ2PS xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE - VPERMILMZ2PS xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPERMILMZ2PS ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE - VPERMILMZ2PS ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE - VPERMIL2PS xmmreg,xmmreg,xmmrm128,xmmreg,imm8 AVX,SANDYBRIDGE - VPERMIL2PS xmmreg,xmmreg,xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPERMIL2PS ymmreg,ymmreg,ymmrm256,ymmreg,imm8 AVX,SANDYBRIDGE - VPERMIL2PS ymmreg,ymmreg,ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE - VPERM2F128 ymmreg,ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE - VPEXTRB reg64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG - VPEXTRB reg32,xmmreg,imm8 AVX,SANDYBRIDGE - VPEXTRB mem8,xmmreg,imm8 AVX,SANDYBRIDGE - VPEXTRW reg64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG - VPEXTRW reg32,xmmreg,imm8 AVX,SANDYBRIDGE - VPEXTRW mem16,xmmreg,imm8 AVX,SANDYBRIDGE - VPEXTRW reg64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG - VPEXTRW reg32,xmmreg,imm8 AVX,SANDYBRIDGE - VPEXTRW mem16,xmmreg,imm8 AVX,SANDYBRIDGE - VPEXTRD reg64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG - VPEXTRD rm32,xmmreg,imm8 AVX,SANDYBRIDGE - VPEXTRQ rm64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG - VPHADDW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPHADDD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPHADDSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPHMINPOSUW xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPHSUBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPHSUBD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPHSUBSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPINSRB xmmreg,xmmreg*,mem8,imm8 AVX,SANDYBRIDGE - VPINSRB xmmreg,xmmreg*,rm8,imm8 AVX,SANDYBRIDGE - VPINSRB xmmreg,xmmreg*,reg32,imm8 AVX,SANDYBRIDGE - VPINSRW xmmreg,xmmreg*,mem16,imm8 AVX,SANDYBRIDGE - VPINSRW xmmreg,xmmreg*,rm16,imm8 AVX,SANDYBRIDGE - VPINSRW xmmreg,xmmreg*,reg32,imm8 AVX,SANDYBRIDGE - VPINSRD xmmreg,xmmreg*,mem32,imm8 AVX,SANDYBRIDGE - VPINSRD xmmreg,xmmreg*,rm32,imm8 AVX,SANDYBRIDGE - VPINSRQ xmmreg,xmmreg*,mem64,imm8 AVX,SANDYBRIDGE,LONG - VPINSRQ xmmreg,xmmreg*,rm64,imm8 AVX,SANDYBRIDGE,LONG - VPMADDWD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMADDUBSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMAXSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMAXSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMAXSD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMAXUB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMAXUW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMAXUD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMINSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMINSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMINSD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMINUB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMINUW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMINUD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMOVMSKB reg64,xmmreg AVX,SANDYBRIDGE,LONG - VPMOVMSKB reg32,xmmreg AVX,SANDYBRIDGE - VPMOVSXBW xmmreg,xmmrm64 AVX,SANDYBRIDGE - VPMOVSXBD xmmreg,xmmrm32 AVX,SANDYBRIDGE - VPMOVSXBQ xmmreg,xmmrm16 AVX,SANDYBRIDGE - VPMOVSXWD xmmreg,xmmrm64 AVX,SANDYBRIDGE - VPMOVSXWQ xmmreg,xmmrm32 AVX,SANDYBRIDGE - VPMOVSXDQ xmmreg,xmmrm64 AVX,SANDYBRIDGE - VPMOVZXBW xmmreg,xmmrm64 AVX,SANDYBRIDGE - VPMOVZXBD xmmreg,xmmrm32 AVX,SANDYBRIDGE - VPMOVZXBQ xmmreg,xmmrm16 AVX,SANDYBRIDGE - VPMOVZXWD xmmreg,xmmrm64 AVX,SANDYBRIDGE - VPMOVZXWQ xmmreg,xmmrm32 AVX,SANDYBRIDGE - VPMOVZXDQ xmmreg,xmmrm64 AVX,SANDYBRIDGE - VPMULHUW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMULHRSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMULHW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMULLW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMULLD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMULUDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMULDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPOR xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSADBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSHUFB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSHUFD xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPSHUFHW xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPSHUFLW xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPSIGNB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSIGNW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSIGND xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSLLDQ xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE - VPSRLDQ xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE - VPSLLW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSLLW xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE - VPSLLD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSLLD xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE - VPSLLQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSLLQ xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE - VPSRAW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSRAW xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE - VPSRAD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSRAD xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE - VPSRLW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSRLW xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE - VPSRLD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSRLD xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE - VPSRLQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSRLQ xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE - VPTEST xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPTEST ymmreg,ymmrm256 AVX,SANDYBRIDGE - VPSUBB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSUBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSUBD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSUBQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSUBSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSUBSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSUBUSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSUBUSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPUNPCKHBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPUNPCKHWD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPUNPCKHDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPUNPCKHQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPUNPCKLBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPUNPCKLWD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPUNPCKLDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPUNPCKLQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPXOR xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VRCPPS xmmreg,xmmrm128 AVX,SANDYBRIDGE - VRCPPS ymmreg,ymmrm256 AVX,SANDYBRIDGE - VRCPSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VRSQRTPS xmmreg,xmmrm128 AVX,SANDYBRIDGE - VRSQRTPS ymmreg,ymmrm256 AVX,SANDYBRIDGE - VRSQRTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VROUNDPD xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VROUNDPD ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE - VROUNDPS xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VROUNDPS ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE - VROUNDSD xmmreg,xmmreg*,xmmrm64,imm8 AVX,SANDYBRIDGE - VROUNDSS xmmreg,xmmreg*,xmmrm32,imm8 AVX,SANDYBRIDGE - VSHUFPD xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VSHUFPD ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE - VSHUFPS xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VSHUFPS ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE - VSQRTPD xmmreg,xmmrm128 AVX,SANDYBRIDGE - VSQRTPD ymmreg,ymmrm256 AVX,SANDYBRIDGE - VSQRTPS xmmreg,xmmrm128 AVX,SANDYBRIDGE - VSQRTPS ymmreg,ymmrm256 AVX,SANDYBRIDGE - VSQRTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VSQRTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VSTMXCSR mem32 AVX,SANDYBRIDGE - VSUBPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VSUBPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VSUBPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VSUBPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VSUBSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VSUBSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VTESTPS xmmreg,xmmrm128 AVX,SANDYBRIDGE - VTESTPS ymmreg,ymmrm256 AVX,SANDYBRIDGE - VTESTPD xmmreg,xmmrm128 AVX,SANDYBRIDGE - VTESTPD ymmreg,ymmrm256 AVX,SANDYBRIDGE - VUCOMISD xmmreg,xmmrm64 AVX,SANDYBRIDGE - VUCOMISS xmmreg,xmmrm32 AVX,SANDYBRIDGE - VUNPCKHPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VUNPCKHPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VUNPCKHPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VUNPCKHPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VUNPCKLPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VUNPCKLPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VUNPCKLPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VUNPCKLPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VXORPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VXORPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VXORPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VXORPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VZEROALL AVX,SANDYBRIDGE - VZEROUPPER AVX,SANDYBRIDGE - - -File: nasm.info, Node: Section B.1.26, Next: Section B.1.27, Prev: Section B.1.25, Up: Section B.1 - -B.1.26. Intel Carry-Less Multiplication instructions (CLMUL) ------------------------------------------------------------- - - PCLMULLQLQDQ xmmreg,xmmrm128 SSE,WESTMERE - PCLMULHQLQDQ xmmreg,xmmrm128 SSE,WESTMERE - PCLMULLQHQDQ xmmreg,xmmrm128 SSE,WESTMERE - PCLMULHQHQDQ xmmreg,xmmrm128 SSE,WESTMERE - PCLMULQDQ xmmreg,xmmrm128,imm8 SSE,WESTMERE - - -File: nasm.info, Node: Section B.1.27, Next: Section B.1.28, Prev: Section B.1.26, Up: Section B.1 - -B.1.27. Intel AVX Carry-Less Multiplication instructions (CLMUL) ----------------------------------------------------------------- - - VPCLMULLQLQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCLMULHQLQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCLMULLQHQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCLMULHQHQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCLMULQDQ xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - - -File: nasm.info, Node: Section B.1.28, Next: Section B.1.29, Prev: Section B.1.27, Up: Section B.1 - -B.1.28. Intel Fused Multiply-Add instructions (FMA) ---------------------------------------------------- - - VFMADD132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD132SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMADD132SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMADD312SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMADD312SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMADD213SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMADD213SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMADD123SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMADD123SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMADD231SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMADD231SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMADD321SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMADD321SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMSUB132SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMSUB132SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMSUB312SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMSUB312SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMSUB213SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMSUB213SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMSUB123SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMSUB123SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMSUB231SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMSUB231SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMSUB321SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMSUB321SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMADD132SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMADD132SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMADD312SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMADD312SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMADD213SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMADD213SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMADD123SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMADD123SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMADD231SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMADD231SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMADD321SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMADD321SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMSUB132SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMSUB132SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMSUB312SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMSUB312SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMSUB213SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMSUB213SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMSUB123SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMSUB123SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMSUB231SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMSUB231SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMSUB321SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMSUB321SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - - -File: nasm.info, Node: Section B.1.29, Next: Section B.1.30, Prev: Section B.1.28, Up: Section B.1 - -B.1.29. VIA (Centaur) security instructions -------------------------------------------- - - XSTORE PENT,CYRIX - XCRYPTECB PENT,CYRIX - XCRYPTCBC PENT,CYRIX - XCRYPTCTR PENT,CYRIX - XCRYPTCFB PENT,CYRIX - XCRYPTOFB PENT,CYRIX - MONTMUL PENT,CYRIX - XSHA1 PENT,CYRIX - XSHA256 PENT,CYRIX - - -File: nasm.info, Node: Section B.1.30, Next: Section B.1.31, Prev: Section B.1.29, Up: Section B.1 - -B.1.30. AMD Lightweight Profiling (LWP) instructions ----------------------------------------------------- - - LLWPCB reg16 AMD - LLWPCB reg32 AMD,386 - LLWPCB reg64 AMD,X64 - SLWPCB reg16 AMD - SLWPCB reg32 AMD,386 - SLWPCB reg64 AMD,X64 - LWPVAL reg16,rm32,imm16 AMD,386 - LWPVAL reg32,rm32,imm32 AMD,386 - LWPVAL reg64,rm32,imm32 AMD,X64 - LWPINS reg16,rm32,imm16 AMD,386 - LWPINS reg32,rm32,imm32 AMD,386 - LWPINS reg64,rm32,imm32 AMD,X64 - - -File: nasm.info, Node: Section B.1.31, Next: Section B.1.32, Prev: Section B.1.30, Up: Section B.1 - -B.1.31. AMD XOP, FMA4 and CVT16 instructions (SSE5) ---------------------------------------------------- - - VCVTPH2PS xmmreg,xmmrm64*,imm8 AMD,SSE5 - VCVTPH2PS ymmreg,xmmrm128,imm8 AMD,SSE5 - VCVTPH2PS ymmreg,ymmrm128*,imm8 AMD,SSE5 - VCVTPS2PH xmmrm64,xmmreg*,imm8 AMD,SSE5 - VCVTPS2PH xmmrm128,ymmreg,imm8 AMD,SSE5 - VCVTPS2PH ymmrm128,ymmreg*,imm8 AMD,SSE5 - VFMADDPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFMADDPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFMADDPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFMADDPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFMADDPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFMADDPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFMADDPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFMADDPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFMADDSD xmmreg,xmmreg*,xmmrm64,xmmreg AMD,SSE5 - VFMADDSD xmmreg,xmmreg*,xmmreg,xmmrm64 AMD,SSE5 - VFMADDSS xmmreg,xmmreg*,xmmrm32,xmmreg AMD,SSE5 - VFMADDSS xmmreg,xmmreg*,xmmreg,xmmrm32 AMD,SSE5 - VFMADDSUBPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFMADDSUBPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFMADDSUBPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFMADDSUBPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFMADDSUBPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFMADDSUBPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFMADDSUBPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFMADDSUBPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFMSUBADDPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFMSUBADDPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFMSUBADDPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFMSUBADDPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFMSUBADDPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFMSUBADDPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFMSUBADDPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFMSUBADDPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFMSUBPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFMSUBPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFMSUBPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFMSUBPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFMSUBPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFMSUBPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFMSUBPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFMSUBPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFMSUBSD xmmreg,xmmreg*,xmmrm64,xmmreg AMD,SSE5 - VFMSUBSD xmmreg,xmmreg*,xmmreg,xmmrm64 AMD,SSE5 - VFMSUBSS xmmreg,xmmreg*,xmmrm32,xmmreg AMD,SSE5 - VFMSUBSS xmmreg,xmmreg*,xmmreg,xmmrm32 AMD,SSE5 - VFNMADDPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFNMADDPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFNMADDPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFNMADDPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFNMADDPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFNMADDPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFNMADDPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFNMADDPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFNMADDSD xmmreg,xmmreg*,xmmrm64,xmmreg AMD,SSE5 - VFNMADDSD xmmreg,xmmreg*,xmmreg,xmmrm64 AMD,SSE5 - VFNMADDSS xmmreg,xmmreg*,xmmrm32,xmmreg AMD,SSE5 - VFNMADDSS xmmreg,xmmreg*,xmmreg,xmmrm32 AMD,SSE5 - VFNMSUBPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFNMSUBPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFNMSUBPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFNMSUBPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFNMSUBPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFNMSUBPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFNMSUBPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFNMSUBPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFNMSUBSD xmmreg,xmmreg*,xmmrm64,xmmreg AMD,SSE5 - VFNMSUBSD xmmreg,xmmreg*,xmmreg,xmmrm64 AMD,SSE5 - VFNMSUBSS xmmreg,xmmreg*,xmmrm32,xmmreg AMD,SSE5 - VFNMSUBSS xmmreg,xmmreg*,xmmreg,xmmrm32 AMD,SSE5 - VFRCZPD xmmreg,xmmrm128* AMD,SSE5 - VFRCZPD ymmreg,ymmrm256* AMD,SSE5 - VFRCZPS xmmreg,xmmrm128* AMD,SSE5 - VFRCZPS ymmreg,ymmrm256* AMD,SSE5 - VFRCZSD xmmreg,xmmrm64* AMD,SSE5 - VFRCZSS xmmreg,xmmrm32* AMD,SSE5 - VPCMOV xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPCMOV ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VPCMOV xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VPCMOV ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VPCOMB xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 - VPCOMD xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 - VPCOMQ xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 - VPCOMUB xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 - VPCOMUD xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 - VPCOMUQ xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 - VPCOMUW xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 - VPCOMW xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 - VPHADDBD xmmreg,xmmrm128* AMD,SSE5 - VPHADDBQ xmmreg,xmmrm128* AMD,SSE5 - VPHADDBW xmmreg,xmmrm128* AMD,SSE5 - VPHADDDQ xmmreg,xmmrm128* AMD,SSE5 - VPHADDUBD xmmreg,xmmrm128* AMD,SSE5 - VPHADDUBQ xmmreg,xmmrm128* AMD,SSE5 - VPHADDUBW xmmreg,xmmrm128* AMD,SSE5 - VPHADDUDQ xmmreg,xmmrm128* AMD,SSE5 - VPHADDUWD xmmreg,xmmrm128* AMD,SSE5 - VPHADDUWQ xmmreg,xmmrm128* AMD,SSE5 - VPHADDWD xmmreg,xmmrm128* AMD,SSE5 - VPHADDWQ xmmreg,xmmrm128* AMD,SSE5 - VPHSUBBW xmmreg,xmmrm128* AMD,SSE5 - VPHSUBDQ xmmreg,xmmrm128* AMD,SSE5 - VPHSUBWD xmmreg,xmmrm128* AMD,SSE5 - VPMACSDD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMACSDQH xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMACSDQL xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMACSSDD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMACSSDQH xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMACSSDQL xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMACSSWD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMACSSWW xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMACSWD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMACSWW xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMADCSSWD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMADCSWD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPPERM xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VPPERM xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPROTB xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPROTB xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPROTB xmmreg,xmmrm128*,imm8 AMD,SSE5 - VPROTD xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPROTD xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPROTD xmmreg,xmmrm128*,imm8 AMD,SSE5 - VPROTQ xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPROTQ xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPROTQ xmmreg,xmmrm128*,imm8 AMD,SSE5 - VPROTW xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPROTW xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPROTW xmmreg,xmmrm128*,imm8 AMD,SSE5 - VPSHAB xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPSHAB xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPSHAD xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPSHAD xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPSHAQ xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPSHAQ xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPSHAW xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPSHAW xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPSHLB xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPSHLB xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPSHLD xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPSHLD xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPSHLQ xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPSHLQ xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPSHLW xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPSHLW xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - - -File: nasm.info, Node: Section B.1.32, Next: Appendix C, Prev: Section B.1.31, Up: Section B.1 - -B.1.32. Systematic names for the hinting nop instructions ---------------------------------------------------------- - - HINT_NOP0 rm16 P6,UNDOC - HINT_NOP0 rm32 P6,UNDOC - HINT_NOP0 rm64 X64,UNDOC - HINT_NOP1 rm16 P6,UNDOC - HINT_NOP1 rm32 P6,UNDOC - HINT_NOP1 rm64 X64,UNDOC - HINT_NOP2 rm16 P6,UNDOC - HINT_NOP2 rm32 P6,UNDOC - HINT_NOP2 rm64 X64,UNDOC - HINT_NOP3 rm16 P6,UNDOC - HINT_NOP3 rm32 P6,UNDOC - HINT_NOP3 rm64 X64,UNDOC - HINT_NOP4 rm16 P6,UNDOC - HINT_NOP4 rm32 P6,UNDOC - HINT_NOP4 rm64 X64,UNDOC - HINT_NOP5 rm16 P6,UNDOC - HINT_NOP5 rm32 P6,UNDOC - HINT_NOP5 rm64 X64,UNDOC - HINT_NOP6 rm16 P6,UNDOC - HINT_NOP6 rm32 P6,UNDOC - HINT_NOP6 rm64 X64,UNDOC - HINT_NOP7 rm16 P6,UNDOC - HINT_NOP7 rm32 P6,UNDOC - HINT_NOP7 rm64 X64,UNDOC - HINT_NOP8 rm16 P6,UNDOC - HINT_NOP8 rm32 P6,UNDOC - HINT_NOP8 rm64 X64,UNDOC - HINT_NOP9 rm16 P6,UNDOC - HINT_NOP9 rm32 P6,UNDOC - HINT_NOP9 rm64 X64,UNDOC - HINT_NOP10 rm16 P6,UNDOC - HINT_NOP10 rm32 P6,UNDOC - HINT_NOP10 rm64 X64,UNDOC - HINT_NOP11 rm16 P6,UNDOC - HINT_NOP11 rm32 P6,UNDOC - HINT_NOP11 rm64 X64,UNDOC - HINT_NOP12 rm16 P6,UNDOC - HINT_NOP12 rm32 P6,UNDOC - HINT_NOP12 rm64 X64,UNDOC - HINT_NOP13 rm16 P6,UNDOC - HINT_NOP13 rm32 P6,UNDOC - HINT_NOP13 rm64 X64,UNDOC - HINT_NOP14 rm16 P6,UNDOC - HINT_NOP14 rm32 P6,UNDOC - HINT_NOP14 rm64 X64,UNDOC - HINT_NOP15 rm16 P6,UNDOC - HINT_NOP15 rm32 P6,UNDOC - HINT_NOP15 rm64 X64,UNDOC - HINT_NOP16 rm16 P6,UNDOC - HINT_NOP16 rm32 P6,UNDOC - HINT_NOP16 rm64 X64,UNDOC - HINT_NOP17 rm16 P6,UNDOC - HINT_NOP17 rm32 P6,UNDOC - HINT_NOP17 rm64 X64,UNDOC - HINT_NOP18 rm16 P6,UNDOC - HINT_NOP18 rm32 P6,UNDOC - HINT_NOP18 rm64 X64,UNDOC - HINT_NOP19 rm16 P6,UNDOC - HINT_NOP19 rm32 P6,UNDOC - HINT_NOP19 rm64 X64,UNDOC - HINT_NOP20 rm16 P6,UNDOC - HINT_NOP20 rm32 P6,UNDOC - HINT_NOP20 rm64 X64,UNDOC - HINT_NOP21 rm16 P6,UNDOC - HINT_NOP21 rm32 P6,UNDOC - HINT_NOP21 rm64 X64,UNDOC - HINT_NOP22 rm16 P6,UNDOC - HINT_NOP22 rm32 P6,UNDOC - HINT_NOP22 rm64 X64,UNDOC - HINT_NOP23 rm16 P6,UNDOC - HINT_NOP23 rm32 P6,UNDOC - HINT_NOP23 rm64 X64,UNDOC - HINT_NOP24 rm16 P6,UNDOC - HINT_NOP24 rm32 P6,UNDOC - HINT_NOP24 rm64 X64,UNDOC - HINT_NOP25 rm16 P6,UNDOC - HINT_NOP25 rm32 P6,UNDOC - HINT_NOP25 rm64 X64,UNDOC - HINT_NOP26 rm16 P6,UNDOC - HINT_NOP26 rm32 P6,UNDOC - HINT_NOP26 rm64 X64,UNDOC - HINT_NOP27 rm16 P6,UNDOC - HINT_NOP27 rm32 P6,UNDOC - HINT_NOP27 rm64 X64,UNDOC - HINT_NOP28 rm16 P6,UNDOC - HINT_NOP28 rm32 P6,UNDOC - HINT_NOP28 rm64 X64,UNDOC - HINT_NOP29 rm16 P6,UNDOC - HINT_NOP29 rm32 P6,UNDOC - HINT_NOP29 rm64 X64,UNDOC - HINT_NOP30 rm16 P6,UNDOC - HINT_NOP30 rm32 P6,UNDOC - HINT_NOP30 rm64 X64,UNDOC - HINT_NOP31 rm16 P6,UNDOC - HINT_NOP31 rm32 P6,UNDOC - HINT_NOP31 rm64 X64,UNDOC - HINT_NOP32 rm16 P6,UNDOC - HINT_NOP32 rm32 P6,UNDOC - HINT_NOP32 rm64 X64,UNDOC - HINT_NOP33 rm16 P6,UNDOC - HINT_NOP33 rm32 P6,UNDOC - HINT_NOP33 rm64 X64,UNDOC - HINT_NOP34 rm16 P6,UNDOC - HINT_NOP34 rm32 P6,UNDOC - HINT_NOP34 rm64 X64,UNDOC - HINT_NOP35 rm16 P6,UNDOC - HINT_NOP35 rm32 P6,UNDOC - HINT_NOP35 rm64 X64,UNDOC - HINT_NOP36 rm16 P6,UNDOC - HINT_NOP36 rm32 P6,UNDOC - HINT_NOP36 rm64 X64,UNDOC - HINT_NOP37 rm16 P6,UNDOC - HINT_NOP37 rm32 P6,UNDOC - HINT_NOP37 rm64 X64,UNDOC - HINT_NOP38 rm16 P6,UNDOC - HINT_NOP38 rm32 P6,UNDOC - HINT_NOP38 rm64 X64,UNDOC - HINT_NOP39 rm16 P6,UNDOC - HINT_NOP39 rm32 P6,UNDOC - HINT_NOP39 rm64 X64,UNDOC - HINT_NOP40 rm16 P6,UNDOC - HINT_NOP40 rm32 P6,UNDOC - HINT_NOP40 rm64 X64,UNDOC - HINT_NOP41 rm16 P6,UNDOC - HINT_NOP41 rm32 P6,UNDOC - HINT_NOP41 rm64 X64,UNDOC - HINT_NOP42 rm16 P6,UNDOC - HINT_NOP42 rm32 P6,UNDOC - HINT_NOP42 rm64 X64,UNDOC - HINT_NOP43 rm16 P6,UNDOC - HINT_NOP43 rm32 P6,UNDOC - HINT_NOP43 rm64 X64,UNDOC - HINT_NOP44 rm16 P6,UNDOC - HINT_NOP44 rm32 P6,UNDOC - HINT_NOP44 rm64 X64,UNDOC - HINT_NOP45 rm16 P6,UNDOC - HINT_NOP45 rm32 P6,UNDOC - HINT_NOP45 rm64 X64,UNDOC - HINT_NOP46 rm16 P6,UNDOC - HINT_NOP46 rm32 P6,UNDOC - HINT_NOP46 rm64 X64,UNDOC - HINT_NOP47 rm16 P6,UNDOC - HINT_NOP47 rm32 P6,UNDOC - HINT_NOP47 rm64 X64,UNDOC - HINT_NOP48 rm16 P6,UNDOC - HINT_NOP48 rm32 P6,UNDOC - HINT_NOP48 rm64 X64,UNDOC - HINT_NOP49 rm16 P6,UNDOC - HINT_NOP49 rm32 P6,UNDOC - HINT_NOP49 rm64 X64,UNDOC - HINT_NOP50 rm16 P6,UNDOC - HINT_NOP50 rm32 P6,UNDOC - HINT_NOP50 rm64 X64,UNDOC - HINT_NOP51 rm16 P6,UNDOC - HINT_NOP51 rm32 P6,UNDOC - HINT_NOP51 rm64 X64,UNDOC - HINT_NOP52 rm16 P6,UNDOC - HINT_NOP52 rm32 P6,UNDOC - HINT_NOP52 rm64 X64,UNDOC - HINT_NOP53 rm16 P6,UNDOC - HINT_NOP53 rm32 P6,UNDOC - HINT_NOP53 rm64 X64,UNDOC - HINT_NOP54 rm16 P6,UNDOC - HINT_NOP54 rm32 P6,UNDOC - HINT_NOP54 rm64 X64,UNDOC - HINT_NOP55 rm16 P6,UNDOC - HINT_NOP55 rm32 P6,UNDOC - HINT_NOP55 rm64 X64,UNDOC - HINT_NOP56 rm16 P6,UNDOC - HINT_NOP56 rm32 P6,UNDOC - HINT_NOP56 rm64 X64,UNDOC - HINT_NOP57 rm16 P6,UNDOC - HINT_NOP57 rm32 P6,UNDOC - HINT_NOP57 rm64 X64,UNDOC - HINT_NOP58 rm16 P6,UNDOC - HINT_NOP58 rm32 P6,UNDOC - HINT_NOP58 rm64 X64,UNDOC - HINT_NOP59 rm16 P6,UNDOC - HINT_NOP59 rm32 P6,UNDOC - HINT_NOP59 rm64 X64,UNDOC - HINT_NOP60 rm16 P6,UNDOC - HINT_NOP60 rm32 P6,UNDOC - HINT_NOP60 rm64 X64,UNDOC - HINT_NOP61 rm16 P6,UNDOC - HINT_NOP61 rm32 P6,UNDOC - HINT_NOP61 rm64 X64,UNDOC - HINT_NOP62 rm16 P6,UNDOC - HINT_NOP62 rm32 P6,UNDOC - HINT_NOP62 rm64 X64,UNDOC - HINT_NOP63 rm16 P6,UNDOC - HINT_NOP63 rm32 P6,UNDOC - HINT_NOP63 rm64 X64,UNDOC - - -File: nasm.info, Node: Appendix C, Next: Section C.1, Prev: Section B.1.32, Up: Top - -Appendix C: NASM Version History -******************************** - -* Menu: - -* Section C.1:: NASM 2 Series -* Section C.2:: NASM 0.98 Series -* Section C.3:: NASM 0.9 Series - - -File: nasm.info, Node: Section C.1, Next: Section C.1.1, Prev: Appendix C, Up: Appendix C - -C.1. NASM 2 Series -================== - -The NASM 2 series support x86-64, and is the production version of NASM -since 2007. - -* Menu: - -* Section C.1.1:: Version 2.08 -* Section C.1.2:: Version 2.07 -* Section C.1.3:: Version 2.06 -* Section C.1.4:: Version 2.05.01 -* Section C.1.5:: Version 2.05 -* Section C.1.6:: Version 2.04 -* Section C.1.7:: Version 2.03.01 -* Section C.1.8:: Version 2.03 -* Section C.1.9:: Version 2.02 -* Section C.1.10:: Version 2.01 -* Section C.1.11:: Version 2.00 - - -File: nasm.info, Node: Section C.1.1, Next: Section C.1.2, Prev: Section C.1, Up: Section C.1 - -C.1.1. Version 2.08 -------------------- - - * A number of enhancements/fixes in macros area. - - * Support for arbitrarily terminating macro expansions `%exitmacro'. - See *note Section 4.3.12::. - - * Support for recursive macro expansion `%rmacro/irmacro'. See *note - Section 4.3.1::. - - * Support for converting strings to tokens. See *note Section - 4.1.9::. - - * Fuzzy operand size logic introduced. - - * Fix COFF stack overrun on too long export identifiers. - - * Fix Macho-O alignment bug. - - * Fix crashes with -fwin32 on file with many exports. - - * Fix stack overrun for too long [DEBUG id]. - - * Fix incorrect sbyte usage in IMUL (hit only if optimization flag - passed). - - * Append ending token for `.stabs' records in the ELF output format. - - * New NSIS script which uses ModernUI and MultiUser approach. - - * Visual Studio 2008 NASM integration (rules file). - - * Warn a user if a constant is too long (and as result will be - stripped). - - * The obsoleted pre-XOP AMD SSE5 instruction set which was never - actualized was removed. - - * Fix stack overrun on too long error file name passed from the - command line. - - * Bind symbols to the .text section by default (ie in case if SECTION - directive was omitted) in the ELF output format. - - * Fix sync points array index wrapping. - - * A few fixes for FMA4 and XOP instruction templates. - - * Add AMD Lightweight Profiling (LWP) instructions. - - -File: nasm.info, Node: Section C.1.2, Next: Section C.1.3, Prev: Section C.1.1, Up: Section C.1 - -C.1.2. Version 2.07 -------------------- - - * NASM is now under the 2-clause BSD license. See *note Section - 1.1.2::. - - * Fix the section type for the `.strtab' section in the `elf64' - output format. - - * Fix the handling of `COMMON' directives in the `obj' output format. - - * New `ith' and `srec' output formats; these are variants of the - `bin' output format which output Intel hex and Motorola S-records, - respectively. See *note Section 7.2:: and *note Section 7.3::. - - * `rdf2ihx' replaced with an enhanced `rdf2bin', which can output - binary, COM, Intel hex or Motorola S-records. - - * The Windows installer now puts the NASM directory first in the - `PATH' of the "NASM Shell". - - * Revert the early expansion behavior of `%+' to pre-2.06 behavior: - `%+' is only expanded late. - - * Yet another Mach-O alignment fix. - - * Don't delete the list file on errors. Also, include error and - warning information in the list file. - - * Support for 64-bit Mach-O output, see *note Section 7.8::. - - * Fix assert failure on certain operations that involve strings with - high-bit bytes. - - -File: nasm.info, Node: Section C.1.3, Next: Section C.1.4, Prev: Section C.1.2, Up: Section C.1 - -C.1.3. Version 2.06 -------------------- - - * This release is dedicated to the memory of Charles A. Crayne, long - time NASM developer as well as moderator of `comp.lang.asm.x86' - and author of the book _Serious Assembler_. We miss you, Chuck. - - * Support for indirect macro expansion (`%[...]'). See *note Section - 4.1.3::. - - * `%pop' can now take an argument, see *note Section 4.7.1::. - - * The argument to `%use' is no longer macro-expanded. Use `%[...]' - if macro expansion is desired. - - * Support for thread-local storage in ELF32 and ELF64. See *note - Section 7.9.4::. - - * Fix crash on `%ifmacro' without an argument. - - * Correct the arguments to the `POPCNT' instruction. - - * Fix section alignment in the Mach-O format. - - * Update AVX support to version 5 of the Intel specification. - - * Fix the handling of accesses to context-local macros from higher - levels in the context stack. - - * Treat `WAIT' as a prefix rather than as an instruction, thereby - allowing constructs like `O16 FSAVE' to work correctly. - - * Support for structures with a non-zero base offset. See *note - Section 4.11.10::. - - * Correctly handle preprocessor token concatenation (see *note - Section 4.3.8::) involving floating-point numbers. - - * The `PINSR' series of instructions have been corrected and - rationalized. - - * Removed AMD SSE5, replaced with the new XOP/FMA4/CVT16 (rev 3.03) - spec. - - * The ELF backends no longer automatically generate a `.comment' - section. - - * Add additional "well-known" ELF sections with default attributes. - See *note Section 7.9.2::. - - -File: nasm.info, Node: Section C.1.4, Next: Section C.1.5, Prev: Section C.1.3, Up: Section C.1 - -C.1.4. Version 2.05.01 ----------------------- - - * Fix the `-w'/`-W' option parsing, which was broken in NASM 2.05. - - -File: nasm.info, Node: Section C.1.5, Next: Section C.1.6, Prev: Section C.1.4, Up: Section C.1 - -C.1.5. Version 2.05 -------------------- - - * Fix redundant REX.W prefix on `JMP reg64'. - - * Make the behaviour of `-O0' match NASM 0.98 legacy behavior. See - *note Section 2.1.22::. - - * `-w-user' can be used to suppress the output of `%warning' - directives. See *note Section 2.1.24::. - - * Fix bug where `ALIGN' would issue a full alignment datum instead of - zero bytes. - - * Fix offsets in list files. - - * Fix `%include' inside multi-line macros or loops. - - * Fix error where NASM would generate a spurious warning on valid - optimizations of immediate values. - - * Fix arguments to a number of the `CVT' SSE instructions. - - * Fix RIP-relative offsets when the instruction carries an immediate. - - * Massive overhaul of the ELF64 backend for spec compliance. - - * Fix the Geode `PFRCPV' and `PFRSQRTV' instruction. - - * Fix the SSE 4.2 `CRC32' instruction. - - -File: nasm.info, Node: Section C.1.6, Next: Section C.1.7, Prev: Section C.1.5, Up: Section C.1 - -C.1.6. Version 2.04 -------------------- - - * Sanitize macro handing in the `%error' directive. - - * New `%warning' directive to issue user-controlled warnings. - - * `%error' directives are now deferred to the final assembly phase. - - * New `%fatal' directive to immediately terminate assembly. - - * New `%strcat' directive to join quoted strings together. - - * New `%use' macro directive to support standard macro directives. - See *note Section 4.6.4::. - - * Excess default parameters to `%macro' now issues a warning by - default. See *note Section 4.3::. - - * Fix `%ifn' and `%elifn'. - - * Fix nested `%else' clauses. - - * Correct the handling of nested `%rep's. - - * New `%unmacro' directive to undeclare a multi-line macro. See - *note Section 4.3.11::. - - * Builtin macro `__PASS__' which expands to the current assembly - pass. See *note Section 4.11.9::. - - * `__utf16__' and `__utf32__' operators to generate UTF-16 and UTF- - 32 strings. See *note Section 3.4.5::. - - * Fix bug in case-insensitive matching when compiled on platforms - that don't use the `configure' script. Of the official release - binaries, that only affected the OS/2 binary. - - * Support for x87 packed BCD constants. See *note Section 3.4.7::. - - * Correct the `LTR' and `SLDT' instructions in 64-bit mode. - - * Fix unnecessary REX.W prefix on indirect jumps in 64-bit mode. - - * Add AVX versions of the AES instructions (`VAES'...). - - * Fix the 256-bit FMA instructions. - - * Add 256-bit AVX stores per the latest AVX spec. - - * VIA XCRYPT instructions can now be written either with or without - `REP', apparently different versions of the VIA spec wrote them - differently. - - * Add missing 64-bit `MOVNTI' instruction. - - * Fix the operand size of `VMREAD' and `VMWRITE'. - - * Numerous bug fixes, especially to the AES, AVX and VTX - instructions. - - * The optimizer now always runs until it converges. It also runs - even when disabled, but doesn't optimize. This allows most forward - references to be resolved properly. - - * `%push' no longer needs a context identifier; omitting the context - identifier results in an anonymous context. - - -File: nasm.info, Node: Section C.1.7, Next: Section C.1.8, Prev: Section C.1.6, Up: Section C.1 - -C.1.7. Version 2.03.01 ----------------------- - - * Fix buffer overflow in the listing module. - - * Fix the handling of hexadecimal escape codes in `...` strings. - - * The Postscript/PDF documentation has been reformatted. - - * The `-F' option now implies `-g'. - - -File: nasm.info, Node: Section C.1.8, Next: Section C.1.9, Prev: Section C.1.7, Up: Section C.1 - -C.1.8. Version 2.03 -------------------- - - * Add support for Intel AVX, CLMUL and FMA instructions, including - YMM registers. - - * `dy', `resy' and `yword' for 32-byte operands. - - * Fix some SSE5 instructions. - - * Intel `INVEPT', `INVVPID' and `MOVBE' instructions. - - * Fix checking for critical expressions when the optimizer is - enabled. - - * Support the DWARF debugging format for ELF targets. - - * Fix optimizations of signed bytes. - - * Fix operation on bigendian machines. - - * Fix buffer overflow in the preprocessor. - - * `SAFESEH' support for Win32, `IMAGEREL' for Win64 (SEH). - - * `%?' and `%??' to refer to the name of a macro itself. In - particular, `%idefine keyword $%?' can be used to make a keyword - "disappear". - - * New options for dependency generation: `-MD', `-MF', `-MP', `-MT', - `-MQ'. - - * New preprocessor directives `%pathsearch' and `%depend'; INCBIN - reimplemented as a macro. - - * `%include' now resolves macros in a sane manner. - - * `%substr' can now be used to get other than one-character - substrings. - - * New type of character/string constants, using backquotes (``...`'), - which support C-style escape sequences. - - * `%defstr' and `%idefstr' to stringize macro definitions before - creation. - - * Fix forward references used in `EQU' statements. - - -File: nasm.info, Node: Section C.1.9, Next: Section C.1.10, Prev: Section C.1.8, Up: Section C.1 - -C.1.9. Version 2.02 -------------------- - - * Additional fixes for MMX operands with explicit `qword', as well as - (hopefully) SSE operands with `oword'. - - * Fix handling of truncated strings with `DO'. - - * Fix segfaults due to memory overwrites when floating-point - constants were used. - - * Fix segfaults due to missing include files. - - * Fix OpenWatcom Makefiles for DOS and OS/2. - - * Add autogenerated instruction list back into the documentation. - - * ELF: Fix segfault when generating stabs, and no symbols have been - defined. - - * ELF: Experimental support for DWARF debugging information. - - * New compile date and time standard macros. - - * `%ifnum' now returns true for negative numbers. - - * New `%iftoken' test for a single token. - - * New `%ifempty' test for empty expansion. - - * Add support for the `XSAVE' instruction group. - - * Makefile for Netware/gcc. - - * Fix issue with some warnings getting emitted way too many times. - - * Autogenerated instruction list added to the documentation. - - -File: nasm.info, Node: Section C.1.10, Next: Section C.1.11, Prev: Section C.1.9, Up: Section C.1 - -C.1.10. Version 2.01 --------------------- - - * Fix the handling of MMX registers with explicit `qword' tags on - memory (broken in 2.00 due to 64-bit changes.) - - * Fix the PREFETCH instructions. - - * Fix the documentation. - - * Fix debugging info when using `-f elf' (backwards compatibility - alias for `-f elf32'). - - * Man pages for rdoff tools (from the Debian project.) - - * ELF: handle large numbers of sections. - - * Fix corrupt output when the optimizer runs out of passes. - - -File: nasm.info, Node: Section C.1.11, Next: Section C.2, Prev: Section C.1.10, Up: Section C.1 - -C.1.11. Version 2.00 --------------------- - - * Added c99 data-type compliance. - - * Added general x86-64 support. - - * Added win64 (x86-64 COFF) output format. - - * Added `__BITS__' standard macro. - - * Renamed the `elf' output format to `elf32' for clarity. - - * Added `elf64' and `macho' (MacOS X) output formats. - - * Added Numeric constants in `dq' directive. - - * Added `oword', `do' and `reso' pseudo operands. - - * Allow underscores in numbers. - - * Added 8-, 16- and 128-bit floating-point formats. - - * Added binary, octal and hexadecimal floating-point. - - * Correct the generation of floating-point constants. - - * Added floating-point option control. - - * Added Infinity and NaN floating point support. - - * Added ELF Symbol Visibility support. - - * Added setting OSABI value in ELF header directive. - - * Added Generate Makefile Dependencies option. - - * Added Unlimited Optimization Passes option. - - * Added `%IFN' and `%ELIFN' support. - - * Added Logical Negation Operator. - - * Enhanced Stack Relative Preprocessor Directives. - - * Enhanced ELF Debug Formats. - - * Enhanced Send Errors to a File option. - - * Added SSSE3, SSE4.1, SSE4.2, SSE5 support. - - * Added a large number of additional instructions. - - * Significant performance improvements. - - * `-w+warning' and `-w-warning' can now be written as -Wwarning and - -Wno-warning, respectively. See *note Section 2.1.24::. - - * Add `-w+error' to treat warnings as errors. See *note Section - 2.1.24::. - - * Add `-w+all' and `-w-all' to enable or disable all suppressible - warnings. See *note Section 2.1.24::. - - -File: nasm.info, Node: Section C.2, Next: Section C.2.1, Prev: Section C.1.11, Up: Appendix C - -C.2. NASM 0.98 Series -===================== - -The 0.98 series was the production versions of NASM from 1999 to 2007. - -* Menu: - -* Section C.2.1:: Version 0.98.39 -* Section C.2.2:: Version 0.98.38 -* Section C.2.3:: Version 0.98.37 -* Section C.2.4:: Version 0.98.36 -* Section C.2.5:: Version 0.98.35 -* Section C.2.6:: Version 0.98.34 -* Section C.2.7:: Version 0.98.33 -* Section C.2.8:: Version 0.98.32 -* Section C.2.9:: Version 0.98.31 -* Section C.2.10:: Version 0.98.30 -* Section C.2.11:: Version 0.98.28 -* Section C.2.12:: Version 0.98.26 -* Section C.2.13:: Version 0.98.25alt -* Section C.2.14:: Version 0.98.25 -* Section C.2.15:: Version 0.98.24p1 -* Section C.2.16:: Version 0.98.24 -* Section C.2.17:: Version 0.98.23 -* Section C.2.18:: Version 0.98.22 -* Section C.2.19:: Version 0.98.21 -* Section C.2.20:: Version 0.98.20 -* Section C.2.21:: Version 0.98.19 -* Section C.2.22:: Version 0.98.18 -* Section C.2.23:: Version 0.98.17 -* Section C.2.24:: Version 0.98.16 -* Section C.2.25:: Version 0.98.15 -* Section C.2.26:: Version 0.98.14 -* Section C.2.27:: Version 0.98.13 -* Section C.2.28:: Version 0.98.12 -* Section C.2.29:: Version 0.98.11 -* Section C.2.30:: Version 0.98.10 -* Section C.2.31:: Version 0.98.09 -* Section C.2.32:: Version 0.98.08 -* Section C.2.33:: Version 0.98.09b with John Coffman patches released 28-Oct-2001 -* Section C.2.34:: Version 0.98.07 released 01/28/01 -* Section C.2.35:: Version 0.98.06f released 01/18/01 -* Section C.2.36:: Version 0.98.06e released 01/09/01 -* Section C.2.37:: Version 0.98p1 -* Section C.2.38:: Version 0.98bf (bug-fixed) -* Section C.2.39:: Version 0.98.03 with John Coffman's changes released 27-Jul-2000 -* Section C.2.40:: Version 0.98.03 -* Section C.2.41:: Version 0.98 -* Section C.2.42:: Version 0.98p9 -* Section C.2.43:: Version 0.98p8 -* Section C.2.44:: Version 0.98p7 -* Section C.2.45:: Version 0.98p6 -* Section C.2.46:: Version 0.98p3.7 -* Section C.2.47:: Version 0.98p3.6 -* Section C.2.48:: Version 0.98p3.5 -* Section C.2.49:: Version 0.98p3.4 -* Section C.2.50:: Version 0.98p3.3 -* Section C.2.51:: Version 0.98p3.2 -* Section C.2.52:: Version 0.98p3-hpa -* Section C.2.53:: Version 0.98 pre-release 3 -* Section C.2.54:: Version 0.98 pre-release 2 -* Section C.2.55:: Version 0.98 pre-release 1 - - -File: nasm.info, Node: Section C.2.1, Next: Section C.2.2, Prev: Section C.2, Up: Section C.2 - -C.2.1. Version 0.98.39 ----------------------- - - * fix buffer overflow - - * fix outas86's `.bss' handling - - * "make spotless" no longer deletes config.h.in. - - * `%(el)if(n)idn' insensitivity to string quotes difference - (#809300). - - * (nasm.c)`__OUTPUT_FORMAT__' changed to string value instead of - symbol. - - -File: nasm.info, Node: Section C.2.2, Next: Section C.2.3, Prev: Section C.2.1, Up: Section C.2 - -C.2.2. Version 0.98.38 ----------------------- - - * Add Makefile for 16-bit DOS binaries under OpenWatcom, and modify - `mkdep.pl' to be able to generate completely pathless - dependencies, as required by OpenWatcom wmake (it supports path - searches, but not explicit paths.) - - * Fix the `STR' instruction. - - * Fix the ELF output format, which was broken under certain - circumstances due to the addition of stabs support. - - * Quick-fix Borland format debug-info for `-f obj' - - * Fix for `%rep' with no arguments (#560568) - - * Fix concatenation of preprocessor function call (#794686) - - * Fix long label causes coredump (#677841) - - * Use autoheader as well as autoconf to keep configure from - generating ridiculously long command lines. - - * Make sure that all of the formats which support debugging output - actually will suppress debugging output when `-g' not specified. - - -File: nasm.info, Node: Section C.2.3, Next: Section C.2.4, Prev: Section C.2.2, Up: Section C.2 - -C.2.3. Version 0.98.37 ----------------------- - - * Paths given in `-I' switch searched for `incbin'-ed as well as - `%include'-ed files. - - * Added stabs debugging for the ELF output format, patch from Martin - Wawro. - - * Fix `output/outbin.c' to allow origin > 80000000h. - - * Make `-U' switch work. - - * Fix the use of relative offsets with explicit prefixes, e.g. `a32 - loop foo'. - - * Remove `backslash()'. - - * Fix the `SMSW' and `SLDT' instructions. - - * `-O2' and `-O3' are no longer aliases for `-O10' and `-O15'. If - you mean the latter, please say so! :) - - -File: nasm.info, Node: Section C.2.4, Next: Section C.2.5, Prev: Section C.2.3, Up: Section C.2 - -C.2.4. Version 0.98.36 ----------------------- - - * Update rdoff - librarian/archiver - common rec - docs! - - * Fix signed/unsigned problems. - - * Fix `JMP FAR label' and `CALL FAR label'. - - * Add new multisection support - map files - fix align bug - - * Fix sysexit, movhps/movlps reg,reg bugs in insns.dat - - * `Q' or `O' suffixes indicate octal - - * Support Prescott new instructions (PNI). - - * Cyrix `XSTORE' instruction. - - -File: nasm.info, Node: Section C.2.5, Next: Section C.2.6, Prev: Section C.2.4, Up: Section C.2 - -C.2.5. Version 0.98.35 ----------------------- - - * Fix build failure on 16-bit DOS (Makefile.bc3 workaround for - compiler bug.) - - * Fix dependencies and compiler warnings. - - * Add "const" in a number of places. - - * Add -X option to specify error reporting format (use -Xvc to - integrate with Microsoft Visual Studio.) - - * Minor changes for code legibility. - - * Drop use of tmpnam() in rdoff (security fix.) - - -File: nasm.info, Node: Section C.2.6, Next: Section C.2.7, Prev: Section C.2.5, Up: Section C.2 - -C.2.6. Version 0.98.34 ----------------------- - - * Correct additional address-size vs. operand-size confusions. - - * Generate dependencies for all Makefiles automatically. - - * Add support for unimplemented (but theoretically available) - registers such as tr0 and cr5. Segment registers 6 and 7 are - called segr6 and segr7 for the operations which they can be - represented. - - * Correct some disassembler bugs related to redundant address-size - prefixes. Some work still remains in this area. - - * Correctly generate an error for things like "SEG eax". - - * Add the JMPE instruction, enabled by "CPU IA64". - - * Correct compilation on newer gcc/glibc platforms. - - * Issue an error on things like "jmp far eax". - - -File: nasm.info, Node: Section C.2.7, Next: Section C.2.8, Prev: Section C.2.6, Up: Section C.2 - -C.2.7. Version 0.98.33 ----------------------- - - * New __NASM_PATCHLEVEL__ and __NASM_VERSION_ID__ standard macros to - round out the version-query macros. version.pl now understands - X.YYplWW or X.YY.ZZplWW as a version number, equivalent to - X.YY.ZZ.WW (or X.YY.0.WW, as appropriate). - - * New keyword "strict" to disable the optimization of specific - operands. - - * Fix the handing of size overrides with JMP instructions - (instructions such as "jmp dword foo".) - - * Fix the handling of "ABSOLUTE label", where "label" points into a - relocatable segment. - - * Fix OBJ output format with lots of externs. - - * More documentation updates. - - * Add -Ov option to get verbose information about optimizations. - - * Undo a braindead change which broke `%elif' directives. - - * Makefile updates. - - -File: nasm.info, Node: Section C.2.8, Next: Section C.2.9, Prev: Section C.2.7, Up: Section C.2 - -C.2.8. Version 0.98.32 ----------------------- - - * Fix NASM crashing when `%macro' directives were left unterminated. - - * Lots of documentation updates. - - * Complete rewrite of the PostScript/PDF documentation generator. - - * The MS Visual C++ Makefile was updated and corrected. - - * Recognize .rodata as a standard section name in ELF. - - * Fix some obsolete Perl4-isms in Perl scripts. - - * Fix configure.in to work with autoconf 2.5x. - - * Fix a couple of "make cleaner" misses. - - * Make the normal "./configure && make" work with Cygwin. - - -File: nasm.info, Node: Section C.2.9, Next: Section C.2.10, Prev: Section C.2.8, Up: Section C.2 - -C.2.9. Version 0.98.31 ----------------------- - - * Correctly build in a separate object directory again. - - * Derive all references to the version number from the version file. - - * New standard macros __NASM_SUBMINOR__ and __NASM_VER__ macros. - - * Lots of Makefile updates and bug fixes. - - * New `%ifmacro' directive to test for multiline macros. - - * Documentation updates. - - * Fixes for 16-bit OBJ format output. - - * Changed the NASM environment variable to NASMENV. - - -File: nasm.info, Node: Section C.2.10, Next: Section C.2.11, Prev: Section C.2.9, Up: Section C.2 - -C.2.10. Version 0.98.30 ------------------------ - - * Changed doc files a lot: completely removed old READMExx and - Wishlist files, incorporating all information in CHANGES and TODO. - - * I waited a long time to rename zoutieee.c to (original) outieee.c - - * moved all output modules to output/ subdirectory. - - * Added 'make strip' target to strip debug info from nasm & ndisasm. - - * Added INSTALL file with installation instructions. - - * Added -v option description to nasm man. - - * Added dist makefile target to produce source distributions. - - * 16-bit support for ELF output format (GNU extension, but useful.) - - -File: nasm.info, Node: Section C.2.11, Next: Section C.2.12, Prev: Section C.2.10, Up: Section C.2 - -C.2.11. Version 0.98.28 ------------------------ - - * Fastcooked this for Debian's Woody release: Frank applied the - INCBIN bug patch to 0.98.25alt and called it 0.98.28 to not - confuse poor little apt- get. - - -File: nasm.info, Node: Section C.2.12, Next: Section C.2.13, Prev: Section C.2.11, Up: Section C.2 - -C.2.12. Version 0.98.26 ------------------------ - - * Reorganised files even better from 0.98.25alt - - -File: nasm.info, Node: Section C.2.13, Next: Section C.2.14, Prev: Section C.2.12, Up: Section C.2 - -C.2.13. Version 0.98.25alt --------------------------- - - * Prettified the source tree. Moved files to more reasonable places. - - * Added findleak.pl script to misc/ directory. - - * Attempted to fix doc. - - -File: nasm.info, Node: Section C.2.14, Next: Section C.2.15, Prev: Section C.2.13, Up: Section C.2 - -C.2.14. Version 0.98.25 ------------------------ - - * Line continuation character `\'. - - * Docs inadvertantly reverted - "dos packaging". - - -File: nasm.info, Node: Section C.2.15, Next: Section C.2.16, Prev: Section C.2.14, Up: Section C.2 - -C.2.15. Version 0.98.24p1 -------------------------- - - * FIXME: Someone, document this please. - - -File: nasm.info, Node: Section C.2.16, Next: Section C.2.17, Prev: Section C.2.15, Up: Section C.2 - -C.2.16. Version 0.98.24 ------------------------ - - * Documentation - Ndisasm doc added to Nasm.doc. - - -File: nasm.info, Node: Section C.2.17, Next: Section C.2.18, Prev: Section C.2.16, Up: Section C.2 - -C.2.17. Version 0.98.23 ------------------------ - - * Attempted to remove rdoff version1 - - * Lino Mastrodomenico's patches to preproc.c (%$$ bug?). - - -File: nasm.info, Node: Section C.2.18, Next: Section C.2.19, Prev: Section C.2.17, Up: Section C.2 - -C.2.18. Version 0.98.22 ------------------------ - - * Update rdoff2 - attempt to remove v1. - - -File: nasm.info, Node: Section C.2.19, Next: Section C.2.20, Prev: Section C.2.18, Up: Section C.2 - -C.2.19. Version 0.98.21 ------------------------ - - * Optimization fixes. - - -File: nasm.info, Node: Section C.2.20, Next: Section C.2.21, Prev: Section C.2.19, Up: Section C.2 - -C.2.20. Version 0.98.20 ------------------------ - - * Optimization fixes. - - -File: nasm.info, Node: Section C.2.21, Next: Section C.2.22, Prev: Section C.2.20, Up: Section C.2 - -C.2.21. Version 0.98.19 ------------------------ - - * H. J. Lu's patch back out. - - -File: nasm.info, Node: Section C.2.22, Next: Section C.2.23, Prev: Section C.2.21, Up: Section C.2 - -C.2.22. Version 0.98.18 ------------------------ - - * Added ".rdata" to "-f win32". - - -File: nasm.info, Node: Section C.2.23, Next: Section C.2.24, Prev: Section C.2.22, Up: Section C.2 - -C.2.23. Version 0.98.17 ------------------------ - - * H. J. Lu's "bogus elf" patch. (Red Hat problem?) - - -File: nasm.info, Node: Section C.2.24, Next: Section C.2.25, Prev: Section C.2.23, Up: Section C.2 - -C.2.24. Version 0.98.16 ------------------------ - - * Fix whitespace before "[section ..." bug. - - -File: nasm.info, Node: Section C.2.25, Next: Section C.2.26, Prev: Section C.2.24, Up: Section C.2 - -C.2.25. Version 0.98.15 ------------------------ - - * Rdoff changes (?). - - * Fix fixes to memory leaks. - - -File: nasm.info, Node: Section C.2.26, Next: Section C.2.27, Prev: Section C.2.25, Up: Section C.2 - -C.2.26. Version 0.98.14 ------------------------ - - * Fix memory leaks. - - -File: nasm.info, Node: Section C.2.27, Next: Section C.2.28, Prev: Section C.2.26, Up: Section C.2 - -C.2.27. Version 0.98.13 ------------------------ - - * There was no 0.98.13 - - -File: nasm.info, Node: Section C.2.28, Next: Section C.2.29, Prev: Section C.2.27, Up: Section C.2 - -C.2.28. Version 0.98.12 ------------------------ - - * Update optimization (new function of "-O1") - - * Changes to test/bintest.asm (?). - - -File: nasm.info, Node: Section C.2.29, Next: Section C.2.30, Prev: Section C.2.28, Up: Section C.2 - -C.2.29. Version 0.98.11 ------------------------ - - * Optimization changes. - - * Ndisasm fixed. - - -File: nasm.info, Node: Section C.2.30, Next: Section C.2.31, Prev: Section C.2.29, Up: Section C.2 - -C.2.30. Version 0.98.10 ------------------------ - - * There was no 0.98.10 - - -File: nasm.info, Node: Section C.2.31, Next: Section C.2.32, Prev: Section C.2.30, Up: Section C.2 - -C.2.31. Version 0.98.09 ------------------------ - - * Add multiple sections support to "-f bin". - - * Changed GLOBAL_TEMP_BASE in outelf.c from 6 to 15. - - * Add "-v" as an alias to the "-r" switch. - - * Remove "#ifdef" from Tasm compatibility options. - - * Remove redundant size-overrides on "mov ds, ex", etc. - - * Fixes to SSE2, other insns.dat (?). - - * Enable uppercase "I" and "P" switches. - - * Case insinsitive "seg" and "wrt". - - * Update install.sh (?). - - * Allocate tokens in blocks. - - * Improve "invalid effective address" messages. - - -File: nasm.info, Node: Section C.2.32, Next: Section C.2.33, Prev: Section C.2.31, Up: Section C.2 - -C.2.32. Version 0.98.08 ------------------------ - - * Add "`%strlen'" and "`%substr'" macro operators - - * Fixed broken c16.mac. - - * Unterminated string error reported. - - * Fixed bugs as per 0.98bf - - -File: nasm.info, Node: Section C.2.33, Next: Section C.2.34, Prev: Section C.2.32, Up: Section C.2 - -C.2.33. Version 0.98.09b with John Coffman patches released 28-Oct-2001 ------------------------------------------------------------------------ - -Changes from 0.98.07 release to 98.09b as of 28-Oct-2001 - - * More closely compatible with 0.98 when -O0 is implied or - specified. Not strictly identical, since backward branches in - range of short offsets are recognized, and signed byte values with - no explicit size specification will be assembled as a single byte. - - * More forgiving with the PUSH instruction. 0.98 requires a size to - be specified always. 0.98.09b will imply the size from the current - BITS setting (16 or 32). - - * Changed definition of the optimization flag: - - -O0 strict two-pass assembly, JMP and Jcc are handled more like 0.98, -except that back- ward JMPs are short, if possible. - - -O1 strict two-pass assembly, but forward branches are assembled -with code guaranteed to reach; may produce larger code than -O0, but -will produce successful assembly more often if branch offset sizes are -not specified. - - -O2 multi-pass optimization, minimize branch offsets; also will -minimize signed immed- iate bytes, overriding size specification. - - -O3 like -O2, but more passes taken, if needed - - -File: nasm.info, Node: Section C.2.34, Next: Section C.2.35, Prev: Section C.2.33, Up: Section C.2 - -C.2.34. Version 0.98.07 released 01/28/01 ------------------------------------------ - - * Added Stepane Denis' SSE2 instructions to a *working* version of - the code - some earlier versions were based on broken code - sorry - 'bout that. version "0.98.07" - - 01/28/01 - - * Cosmetic modifications to nasm.c, nasm.h, AUTHORS, MODIFIED - - -File: nasm.info, Node: Section C.2.35, Next: Section C.2.36, Prev: Section C.2.34, Up: Section C.2 - -C.2.35. Version 0.98.06f released 01/18/01 ------------------------------------------- - - * - Add "metalbrain"s jecxz bug fix in insns.dat - alter nasmdoc.src - to match - version "0.98.06f" - - -File: nasm.info, Node: Section C.2.36, Next: Section C.2.37, Prev: Section C.2.35, Up: Section C.2 - -C.2.36. Version 0.98.06e released 01/09/01 ------------------------------------------- - - * Removed the "outforms.h" file - it appears to be someone's old - backup of "outform.h". version "0.98.06e" - - 01/09/01 - - * fbk - finally added the fix for the "multiple %includes bug", - known since 7/27/99 - reported originally (?) and sent to us by - Austin Lunnen - he reports that John Fine had a fix within the - day. Here it is... - - * Nelson Rush resigns from the group. Big thanks to Nelson for his - leadership and enthusiasm in getting these changes incorporated - into Nasm! - - * fbk - [list +], [list -] directives - ineptly implemented, should - be re- written or removed, perhaps. - - * Brian Raiter / fbk - "elfso bug" fix - applied to aoutb format as - well - testing might be desirable... - - 08/07/00 - - * James Seter - -postfix, -prefix command line switches. - - * Yuri Zaporogets - rdoff utility changes. - - -File: nasm.info, Node: Section C.2.37, Next: Section C.2.38, Prev: Section C.2.36, Up: Section C.2 - -C.2.37. Version 0.98p1 ----------------------- - - * GAS-like palign (Panos Minos) - - * FIXME: Someone, fill this in with details - - -File: nasm.info, Node: Section C.2.38, Next: Section C.2.39, Prev: Section C.2.37, Up: Section C.2 - -C.2.38. Version 0.98bf (bug-fixed) ----------------------------------- - - * Fixed - elf and aoutb bug - shared libraries - multiple "%include" - bug in "-f obj" - jcxz, jecxz bug - unrecognized option bug in - ndisasm - - -File: nasm.info, Node: Section C.2.39, Next: Section C.2.40, Prev: Section C.2.38, Up: Section C.2 - -C.2.39. Version 0.98.03 with John Coffman's changes released 27-Jul-2000 ------------------------------------------------------------------------- - - * Added signed byte optimizations for the 0x81/0x83 class of - instructions: ADC, ADD, AND, CMP, OR, SBB, SUB, XOR: when used as - 'ADD reg16,imm' or 'ADD reg32,imm.' Also optimization of signed - byte form of 'PUSH imm' and 'IMUL reg,imm'/'IMUL reg,reg,imm.' No - size specification is needed. - - * Added multi-pass JMP and Jcc offset optimization. Offsets on - forward references will preferentially use the short form, without - the need to code a specific size (short or near) for the branch. - Added instructions for 'Jcc label' to use the form 'Jnotcc $+3/JMP - label', in cases where a short offset is out of bounds. If - compiling for a 386 or higher CPU, then the 386 form of Jcc will - be used instead. - - This feature is controlled by a new command-line switch: "O", (upper -case letter O). "-O0" reverts the assembler to no extra optimization -passes, "- O1" allows up to 5 extra passes, and "-O2"(default), allows -up to 10 extra optimization passes. - - * Added a new directive: 'cpu XXX', where XXX is any of: 8086, 186, - 286, 386, 486, 586, pentium, 686, PPro, P2, P3 or Katmai. All are - case insensitive. All instructions will be selected only if they - apply to the selected cpu or lower. Corrected a couple of bugs in - cpu-dependence in 'insns.dat'. - - * Added to 'standard.mac', the "use16" and "use32" forms of the - "bits 16/32" directive. This is nothing new, just conforms to a - lot of other assemblers. (minor) - - * Changed label allocation from 320/32 (10000 labels @ 200K+) to - 32/37 (1000 labels); makes running under DOS much easier. Since - additional label space is allocated dynamically, this should have - no effect on large programs with lots of labels. The 37 is a - prime, believed to be better for hashing. (minor) - - -File: nasm.info, Node: Section C.2.40, Next: Section C.2.41, Prev: Section C.2.39, Up: Section C.2 - -C.2.40. Version 0.98.03 ------------------------ - -"Integrated patchfile 0.98-0.98.01. I call this version 0.98.03 for -historical reasons: 0.98.02 was trashed." ---John Coffman -<johninsd@san.rr.com>, 27-Jul-2000 - - * Kendall Bennett's SciTech MGL changes - - * Note that you must define "TASM_COMPAT" at compile-time to get the - Tasm Ideal Mode compatibility. - - * All changes can be compiled in and out using the TASM_COMPAT - macros, and when compiled without TASM_COMPAT defined we get the - exact same binary as the unmodified 0.98 sources. - - * standard.mac, macros.c: Added macros to ignore TASM directives - before first include - - * nasm.h: Added extern declaration for tasm_compatible_mode - - * nasm.c: Added global variable tasm_compatible_mode - - * Added command line switch for TASM compatible mode (-t) - - * Changed version command line to reflect when compiled with TASM - additions - - * Added response file processing to allow all arguments on a single - line (response file is @resp rather than -@resp for NASM format). - - * labels.c: Changes islocal() macro to support TASM style @@local - labels. - - * Added islocalchar() macro to support TASM style @@local labels. - - * parser.c: Added support for TASM style memory references (ie: mov - [DWORD eax],10 rather than the NASM style mov DWORD [eax],10). - - * preproc.c: Added new directives, `%arg', `%local', `%stacksize' to - directives table - - * Added support for TASM style directives without a leading % symbol. - - * Integrated a block of changes from Andrew Zabolotny - <bit@eltech.ru>: - - * A new keyword `%xdefine' and its case-insensitive counterpart - `%ixdefine'. They work almost the same way as `%define' and - `%idefine' but expand the definition immediately, not on the - invocation. Something like a cross between `%define' and - `%assign'. The "x" suffix stands for "eXpand", so "xdefine" can be - deciphered as "expand-and-define". Thus you can do things like - this: - - %assign ofs 0 - - %macro arg 1 - %xdefine %1 dword [esp+ofs] - %assign ofs ofs+4 - %endmacro - - * Changed the place where the expansion of %$name macros are - expanded. Now they are converted into ..@ctxnum.name form when - detokenizing, so there are no quirks as before when using %$name - arguments to macros, in macros etc. For example: - - %macro abc 1 - %define %1 hello - %endm - - abc %$here - %$here - - Now last line will be expanded into "hello" as expected. This also -allows for lots of goodies, a good example are extended "proc" macros -included in this archive. - - * Added a check for "cstk" in smacro_defined() before calling - get_ctx() - this allows for things like: - - %ifdef %$abc - %endif - - to work without warnings even in no context. - - * Added a check for "cstk" in %if*ctx and %elif*ctx directives - - this allows to use `%ifctx' without excessive warnings. If there - is no active context, `%ifctx' goes through "false" branch. - - * Removed "user error: " prefix with `%error' directive: it just - clobbers the output and has absolutely no functionality. Besides, - this allows to write macros that does not differ from built-in - functions in any way. - - * Added expansion of string that is output by `%error' directive. Now - you can do things like: - - %define hello(x) Hello, x! - - %define %$name andy - %error "hello(%$name)" - - Same happened with `%include' directive. - - * Now all directives that expect an identifier will try to expand and - concatenate everything without whitespaces in between before - usage. For example, with "unfixed" nasm the commands - - %define %$abc hello - %define __%$abc goodbye - __%$abc - - would produce "incorrect" output: last line will expand to - - hello goodbyehello - - Not quite what you expected, eh? :-) The answer is that preprocessor -treats the `%define' construct as if it would be - - %define __ %$abc goodbye - - (note the white space between __ and %$abc). After my "fix" it will -"correctly" expand into - - goodbye - - as expected. Note that I use quotes around words "correct", -"incorrect" etc because this is rather a feature not a bug; however -current behaviour is more logical (and allows more advanced macro usage -:-). - - Same change was applied to: -`%push',`%macro',`%imacro',`%define',`%idefine',`%xdefine',`%ixdefine', -`%assign',`%iassign',`%undef' - - * A new directive [WARNING {+|-}warning-id] have been added. It - works only if the assembly phase is enabled (i.e. it doesn't work - with nasm -e). - - * A new warning type: macro-selfref. By default this warning is - disabled; when enabled NASM warns when a macro self-references - itself; for example the following source: - - [WARNING macro-selfref] - - %macro push 1-* - %rep %0 - push %1 - %rotate 1 - %endrep - %endmacro - - push eax,ebx,ecx - - will produce a warning, but if we remove the first line we won't see -it anymore (which is The Right Thing To Do {tm} IMHO since C -preprocessor eats such constructs without warnings at all). - - * Added a "error" routine to preprocessor which always will set - ERR_PASS1 bit in severity_code. This removes annoying repeated - errors on first and second passes from preprocessor. - - * Added the %+ operator in single-line macros for concatenating two - identifiers. Usage example: - - %define _myfunc _otherfunc - %define cextern(x) _ %+ x - cextern (myfunc) - - After first expansion, third line will become "_myfunc". After this -expansion is performed again so it becomes "_otherunc". - - * Now if preprocessor is in a non-emitting state, no warning or - error will be emitted. Example: - - %if 1 - mov eax,ebx - %else - put anything you want between these two brackets, - even macro-parameter references %1 or local - labels %$zz or macro-local labels %%zz - no - warning will be emitted. - %endif - - * Context-local variables on expansion as a last resort are looked - up in outer contexts. For example, the following piece: - - %push outer - %define %$a [esp] - - %push inner - %$a - %pop - %pop - - will expand correctly the fourth line to [esp]; if we'll define -another %$a inside the "inner" context, it will take precedence over -outer definition. However, this modification has been applied only to -expand_smacro and not to smacro_define: as a consequence expansion -looks in outer contexts, but `%ifdef' won't look in outer contexts. - - This behaviour is needed because we don't want nested contexts to -act on already defined local macros. Example: - - %define %$arg1 [esp+4] - test eax,eax - if nz - mov eax,%$arg1 - endif - - In this example the "if" mmacro enters into the "if" context, so -%$arg1 is not valid anymore inside "if". Of course it could be worked -around by using explicitely %$$arg1 but this is ugly IMHO. - - * Fixed memory leak in `%undef'. The origline wasn't freed before - exiting on success. - - * Fixed trap in preprocessor when line expanded to empty set of - tokens. This happens, for example, in the following case: - - #define SOMETHING - SOMETHING - - -File: nasm.info, Node: Section C.2.41, Next: Section C.2.42, Prev: Section C.2.40, Up: Section C.2 - -C.2.41. Version 0.98 --------------------- - -All changes since NASM 0.98p3 have been produced by H. Peter Anvin -<hpa@zytor.com>. - - * The documentation comment delimiter is - - * Allow EQU definitions to refer to external labels; reported by - Pedro Gimeno. - - * Re-enable support for RDOFF v1; reported by Pedro Gimeno. - - * Updated License file per OK from Simon and Julian. - - -File: nasm.info, Node: Section C.2.42, Next: Section C.2.43, Prev: Section C.2.41, Up: Section C.2 - -C.2.42. Version 0.98p9 ----------------------- - - * Update documentation (although the instruction set reference will - have to wait; I don't want to hold up the 0.98 release for it.) - - * Verified that the NASM implementation of the PEXTRW and PMOVMSKB - instructions is correct. The encoding differs from what the Intel - manuals document, but the Pentium III behaviour matches NASM, not - the Intel manuals. - - * Fix handling of implicit sizes in PSHUFW and PINSRW, reported by - Stefan Hoffmeister. - - * Resurrect the -s option, which was removed when changing the - diagnostic output to stdout. - - -File: nasm.info, Node: Section C.2.43, Next: Section C.2.44, Prev: Section C.2.42, Up: Section C.2 - -C.2.43. Version 0.98p8 ----------------------- - - * Fix for "DB" when NASM is running on a bigendian machine. - - * Invoke insns.pl once for each output script, making Makefile.in - legal for "make -j". - - * Improve the Unix configure-based makefiles to make package - creation easier. - - * Included an RPM .spec file for building RPM (RedHat Package - Manager) packages on Linux or Unix systems. - - * Fix Makefile dependency problems. - - * Change src/rdsrc.pl to include sectioning information in info - output; required for install-info to work. - - * Updated the RDOFF distribution to version 2 from Jules; minor - massaging to make it compile in my environment. - - * Split doc files that can be built by anyone with a Perl - interpreter off into a separate archive. - - * "Dress rehearsal" release! - - -File: nasm.info, Node: Section C.2.44, Next: Section C.2.45, Prev: Section C.2.43, Up: Section C.2 - -C.2.44. Version 0.98p7 ----------------------- - - * Fixed opcodes with a third byte-sized immediate argument to not - complain if given "byte" on the immediate. - - * Allow `%undef' to remove single-line macros with arguments. This - matches the behaviour of #undef in the C preprocessor. - - * Allow -d, -u, -i and -p to be specified as -D, -U, -I and -P for - compatibility with most C compilers and preprocessors. This allows - Makefile options to be shared between cc and nasm, for example. - - * Minor cleanups. - - * Went through the list of Katmai instructions and hopefully fixed - the (rather few) mistakes in it. - - * (Hopefully) fixed a number of disassembler bugs related to - ambiguous instructions (disambiguated by -p) and SSE instructions - with REP. - - * Fix for bug reported by Mark Junger: "call dword 0x12345678" - should work and may add an OSP (affected CALL, JMP, Jcc). - - * Fix for environments when "stderr" isn't a compile-time constant. - - -File: nasm.info, Node: Section C.2.45, Next: Section C.2.46, Prev: Section C.2.44, Up: Section C.2 - -C.2.45. Version 0.98p6 ----------------------- - - * Took officially over coordination of the 0.98 release; so drop the - p3.x notation. Skipped p4 and p5 to avoid confusion with John - Fine's J4 and J5 releases. - - * Update the documentation; however, it still doesn't include - documentation for the various new instructions. I somehow wonder - if it makes sense to have an instruction set reference in the - assembler manual when Intel et al have PDF versions of their - manuals online. - - * Recognize "idt" or "centaur" for the -p option to ndisasm. - - * Changed error messages back to stderr where they belong, but add - an -E option to redirect them elsewhere (the DOS shell cannot - redirect stderr.) - - * -M option to generate Makefile dependencies (based on code from - Alex Verstak.) - - * `%undef' preprocessor directive, and -u option, that undefines a - single-line macro. - - * OS/2 Makefile (Mkfiles/Makefile.os2) for Borland under OS/2; from - Chuck Crayne. - - * Various minor bugfixes (reported by): - Dangling `%s' in preproc.c - (Martin Junker) - - * THERE ARE KNOWN BUGS IN SSE AND THE OTHER KATMAI INSTRUCTIONS. I - am on a trip and didn't bring the Katmai instruction reference, so - I can't work on them right now. - - * Updated the License file per agreement with Simon and Jules to - include a GPL distribution clause. - - -File: nasm.info, Node: Section C.2.46, Next: Section C.2.47, Prev: Section C.2.45, Up: Section C.2 - -C.2.46. Version 0.98p3.7 ------------------------- - - * (Hopefully) fixed the canned Makefiles to include the outrdf2 and - zoutieee modules. - - * Renamed changes.asm to changed.asm. - - -File: nasm.info, Node: Section C.2.47, Next: Section C.2.48, Prev: Section C.2.46, Up: Section C.2 - -C.2.47. Version 0.98p3.6 ------------------------- - - * Fixed a bunch of instructions that were added in 0.98p3.5 which - had memory operands, and the address-size prefix was missing from - the instruction pattern. - - -File: nasm.info, Node: Section C.2.48, Next: Section C.2.49, Prev: Section C.2.47, Up: Section C.2 - -C.2.48. Version 0.98p3.5 ------------------------- - - * Merged in changes from John S. Fine's 0.98-J5 release. John's - based 0.98-J5 on my 0.98p3.3 release; this merges the changes. - - * Expanded the instructions flag field to a long so we can fit more - flags; mark SSE (KNI) and AMD or Katmai-specific instructions as - such. - - * Fix the "PRIV" flag on a bunch of instructions, and create new - "PROT" flag for protected-mode-only instructions (orthogonal to if - the instruction is privileged!) and new "SMM" flag for SMM-only - instructions. - - * Added AMD-only SYSCALL and SYSRET instructions. - - * Make SSE actually work, and add new Katmai MMX instructions. - - * Added a -p (preferred vendor) option to ndisasm so that it can - distinguish e.g. Cyrix opcodes also used in SSE. For example: - - ndisasm -p cyrix aliased.bin - 00000000 670F514310 paddsiw mm0,[ebx+0x10] - 00000005 670F514320 paddsiw mm0,[ebx+0x20] - ndisasm -p intel aliased.bin - 00000000 670F514310 sqrtps xmm0,[ebx+0x10] - 00000005 670F514320 sqrtps xmm0,[ebx+0x20] - - * Added a bunch of Cyrix-specific instructions. - - -File: nasm.info, Node: Section C.2.49, Next: Section C.2.50, Prev: Section C.2.48, Up: Section C.2 - -C.2.49. Version 0.98p3.4 ------------------------- - - * Made at least an attempt to modify all the additional Makefiles - (in the Mkfiles directory). I can't test it, but this was the best - I could do. - - * DOS DJGPP+"Opus Make" Makefile from John S. Fine. - - * changes.asm changes from John S. Fine. - - -File: nasm.info, Node: Section C.2.50, Next: Section C.2.51, Prev: Section C.2.49, Up: Section C.2 - -C.2.50. Version 0.98p3.3 ------------------------- - - * Patch from Conan Brink to allow nesting of `%rep' directives. - - * If we're going to allow INT01 as an alias for INT1/ICEBP (one of - Jules 0.98p3 changes), then we should allow INT03 as an alias for - INT3 as well. - - * Updated changes.asm to include the latest changes. - - * Tried to clean up the <CR>s that had snuck in from a DOS/Windows - environment into my Unix environment, and try to make sure than - DOS/Windows users get them back. - - * We would silently generate broken tools if insns.dat wasn't sorted - properly. Change insns.pl so that the order doesn't matter. - - * Fix bug in insns.pl (introduced by me) which would cause - conditional instructions to have an extra "cc" in disassembly, - e.g. "jnz" disassembled as "jccnz". - - -File: nasm.info, Node: Section C.2.51, Next: Section C.2.52, Prev: Section C.2.50, Up: Section C.2 - -C.2.51. Version 0.98p3.2 ------------------------- - - * Merged in John S. Fine's changes from his 0.98-J4 prerelease; see - http://www.csoft.net/cz/johnfine/ - - * Changed previous "spotless" Makefile target (appropriate for - distribution) to "distclean", and added "cleaner" target which is - same as "clean" except deletes files generated by Perl scripts; - "spotless" is union. - - * Removed BASIC programs from distribution. Get a Perl interpreter - instead (see below.) - - * Calling this "pre-release 3.2" rather than "p3-hpa2" because of - John's contributions. - - * Actually link in the IEEE output format (zoutieee.c); fix a bunch - of compiler warnings in that file. Note I don't know what IEEE - output is supposed to look like, so these changes were made - "blind". - - -File: nasm.info, Node: Section C.2.52, Next: Section C.2.53, Prev: Section C.2.51, Up: Section C.2 - -C.2.52. Version 0.98p3-hpa --------------------------- - - * Merged nasm098p3.zip with nasm-0.97.tar.gz to create a fully - buildable version for Unix systems (Makefile.in updates, etc.) - - * Changed insns.pl to create the instruction tables in nasm.h and - names.c, so that a new instruction can be added by adding it - *only* to insns.dat. - - * Added the following new instructions: SYSENTER, SYSEXIT, FXSAVE, - FXRSTOR, UD1, UD2 (the latter two are two opcodes that Intel - guarantee will never be used; one of them is documented as UD2 in - Intel documentation, the other one just as "Undefined Opcode" --- - calling it UD1 seemed to make sense.) - - * MAX_SYMBOL was defined to be 9, but LOADALL286 and LOADALL386 are - 10 characters long. Now MAX_SYMBOL is derived from insns.dat. - - * A note on the BASIC programs included: forget them. insns.bas is - already out of date. Get yourself a Perl interpreter for your - platform of choice at http://www.cpan.org/ports/index.html. - - -File: nasm.info, Node: Section C.2.53, Next: Section C.2.54, Prev: Section C.2.52, Up: Section C.2 - -C.2.53. Version 0.98 pre-release 3 ----------------------------------- - - * added response file support, improved command line handling, new - layout help screen - - * fixed limit checking bug, 'OUT byte nn, reg' bug, and a couple of - rdoff related bugs, updated Wishlist; 0.98 Prerelease 3. - - -File: nasm.info, Node: Section C.2.54, Next: Section C.2.55, Prev: Section C.2.53, Up: Section C.2 - -C.2.54. Version 0.98 pre-release 2 ----------------------------------- - - * fixed bug in outcoff.c to do with truncating section names longer - than 8 characters, referencing beyond end of string; 0.98 - pre-release 2 - - -File: nasm.info, Node: Section C.2.55, Next: Section C.3, Prev: Section C.2.54, Up: Section C.2 - -C.2.55. Version 0.98 pre-release 1 ----------------------------------- - - * Fixed a bug whereby STRUC didn't work at all in RDF. - - * Fixed a problem with group specification in PUBDEFs in OBJ. - - * Improved ease of adding new output formats. Contribution due to - Fox Cutter. - - * Fixed a bug in relocations in the `bin' format: was showing up - when a relocatable reference crossed an 8192-byte boundary in any - output section. - - * Fixed a bug in local labels: local-label lookups were inconsistent - between passes one and two if an EQU occurred between the - definition of a global label and the subsequent use of a local - label local to that global. - - * Fixed a seg-fault in the preprocessor (again) which happened when - you use a blank line as the first line of a multi-line macro - definition and then defined a label on the same line as a call to - that macro. - - * Fixed a stale-pointer bug in the handling of the NASM environment - variable. Thanks to Thomas McWilliams. - - * ELF had a hard limit on the number of sections which caused - segfaults when transgressed. Fixed. - - * Added ability for ndisasm to read from stdin by using `-' as the - filename. - - * ndisasm wasn't outputting the TO keyword. Fixed. - - * Fixed error cascade on bogus expression in `%if' - an error in - evaluation was causing the entire `%if' to be discarded, thus - creating trouble later when the `%else' or `%endif' was - encountered. - - * Forward reference tracking was instruction-granular not operand- - granular, which was causing 286-specific code to be generated - needlessly on code of the form `shr word [forwardref],1'. Thanks - to Jim Hague for sending a patch. - - * All messages now appear on stdout, as sending them to stderr - serves no useful purpose other than to make redirection difficult. - - * Fixed the problem with EQUs pointing to an external symbol - this - now generates an error message. - - * Allowed multiple size prefixes to an operand, of which only the - first is taken into account. - - * Incorporated John Fine's changes, including fixes of a large - number of preprocessor bugs, some small problems in OBJ, and a - reworking of label handling to define labels before their line is - assembled, rather than after. - - * Reformatted a lot of the source code to be more readable. Included - 'coding.txt' as a guideline for how to format code for - contributors. - - * Stopped nested `%reps' causing a panic - they now cause a slightly - more friendly error message instead. - - * Fixed floating point constant problems (patch by Pedro Gimeno) - - * Fixed the return value of insn_size() not being checked for -1, - indicating an error. - - * Incorporated 3Dnow! instructions. - - * Fixed the 'mov eax, eax + ebx' bug. - - * Fixed the GLOBAL EQU bug in ELF. Released developers release 3. - - * Incorporated John Fine's command line parsing changes - - * Incorporated David Lindauer's OMF debug support - - * Made changes for LCC 4.0 support (`__NASM_CDecl__', removed - register size specification warning when sizes agree). - - -File: nasm.info, Node: Section C.3, Next: Section C.3.1, Prev: Section C.2.55, Up: Appendix C - -C.3. NASM 0.9 Series -==================== - -Revisions before 0.98. - -* Menu: - -* Section C.3.1:: Version 0.97 released December 1997 -* Section C.3.2:: Version 0.96 released November 1997 -* Section C.3.3:: Version 0.95 released July 1997 -* Section C.3.4:: Version 0.94 released April 1997 -* Section C.3.5:: Version 0.93 released January 1997 -* Section C.3.6:: Version 0.92 released January 1997 -* Section C.3.7:: Version 0.91 released November 1996 -* Section C.3.8:: Version 0.90 released October 1996 - - -File: nasm.info, Node: Section C.3.1, Next: Section C.3.2, Prev: Section C.3, Up: Section C.3 - -C.3.1. Version 0.97 released December 1997 ------------------------------------------- - - * This was entirely a bug-fix release to 0.96, which seems to have - got cursed. Silly me. - - * Fixed stupid mistake in OBJ which caused `MOV EAX,<constant>' to - fail. Caused by an error in the `MOV EAX,<segment>' support. - - * ndisasm hung at EOF when compiled with lcc on Linux because lcc on - Linux somehow breaks feof(). ndisasm now does not rely on feof(). - - * A heading in the documentation was missing due to a markup error - in the indexing. Fixed. - - * Fixed failure to update all pointers on realloc() within extended- - operand code in parser.c. Was causing wrong behaviour and seg - faults on lines such as `dd 0.0,0.0,0.0,0.0,...' - - * Fixed a subtle preprocessor bug whereby invoking one multi-line - macro on the first line of the expansion of another, when the - second had been invoked with a label defined before it, didn't - expand the inner macro. - - * Added internal.doc back in to the distribution archives - it was - missing in 0.96 *blush* - - * Fixed bug causing 0.96 to be unable to assemble its own test files, - specifically objtest.asm. *blush again* - - * Fixed seg-faults and bogus error messages caused by mismatching - `%rep' and `%endrep' within multi-line macro definitions. - - * Fixed a problem with buffer overrun in OBJ, which was causing - corruption at ends of long PUBDEF records. - - * Separated DOS archives into main-program and documentation to - reduce download size. - - -File: nasm.info, Node: Section C.3.2, Next: Section C.3.3, Prev: Section C.3.1, Up: Section C.3 - -C.3.2. Version 0.96 released November 1997 ------------------------------------------- - - * Fixed a bug whereby, if `nasm sourcefile' would cause a filename - collision warning and put output into `nasm.out', then `nasm - sourcefile -o outputfile' still gave the warning even though the - `-o' was honoured. Fixed name pollution under Digital UNIX: one of - its header files defined R_SP, which broke the enum in nasm.h. - - * Fixed minor instruction table problems: FUCOM and FUCOMP didn't - have two- operand forms; NDISASM didn't recognise the longer - register forms of PUSH and POP (eg FF F3 for PUSH BX); TEST - mem,imm32 was flagged as undocumented; the 32-bit forms of CMOV - had 16-bit operand size prefixes; `AAD imm' and `AAM imm' are no - longer flagged as undocumented because the Intel Architecture - reference documents them. - - * Fixed a problem with the local-label mechanism, whereby strange - types of symbol (EQUs, auto-defined OBJ segment base symbols) - interfered with the `previous global label' value and screwed up - local labels. - - * Fixed a bug whereby the stub preprocessor didn't communicate with - the listing file generator, so that the -a and -l options in - conjunction would produce a useless listing file. - - * Merged `os2' object file format back into `obj', after discovering - that `obj' _also_ shouldn't have a link pass separator in a module - containing a non-trivial MODEND. Flat segments are now declared - using the FLAT attribute. `os2' is no longer a valid object format - name: use `obj'. - - * Removed the fixed-size temporary storage in the evaluator. Very - very long expressions (like `mov ax,1+1+1+1+...' for two hundred - 1s or so) should now no longer crash NASM. - - * Fixed a bug involving segfaults on disassembly of MMX - instructions, by changing the meaning of one of the operand-type - flags in nasm.h. This may cause other apparently unrelated MMX - problems; it needs to be tested thoroughly. - - * Fixed some buffer overrun problems with large OBJ output files. - Thanks to DJ Delorie for the bug report and fix. - - * Made preprocess-only mode actually listen to the `%line' markers - as it prints them, so that it can report errors more sanely. - - * Re-designed the evaluator to keep more sensible track of - expressions involving forward references: can now cope with - previously-nightmare situations such as: - - mov ax,foo | bar - foo equ 1 - bar equ 2 - - * Added the ALIGN and ALIGNB standard macros. - - * Added PIC support in ELF: use of WRT to obtain the four extra - relocation types needed. - - * Added the ability for output file formats to define their own - extensions to the GLOBAL, COMMON and EXTERN directives. - - * Implemented common-variable alignment, and global-symbol type and - size declarations, in ELF. - - * Implemented NEAR and FAR keywords for common variables, plus - far-common element size specification, in OBJ. - - * Added a feature whereby EXTERNs and COMMONs in OBJ can be given a - default WRT specification (either a segment or a group). - - * Transformed the Unix NASM archive into an auto-configuring package. - - * Added a sanity-check for people applying SEG to things which are - already segment bases: this previously went unnoticed by the SEG - processing and caused OBJ-driver panics later. - - * Added the ability, in OBJ format, to deal with `MOV EAX,<segment>' - type references: OBJ doesn't directly support dword-size segment - base fixups, but as long as the low two bytes of the constant term - are zero, a word-size fixup can be generated instead and it will - work. - - * Added the ability to specify sections' alignment requirements in - Win32 object files and pure binary files. - - * Added preprocess-time expression evaluation: the `%assign' (and - `%iassign') directive and the bare `%if' (and `%elif') - conditional. Added relational operators to the evaluator, for use - only in `%if' constructs: the standard relationals = < > <= >= <> - (and C-like synonyms == and !=) plus low-precedence logical - operators &&, ^^ and ||. - - * Added a preprocessor repeat construct: `%rep' / `%exitrep' / - `%endrep'. - - * Added the __FILE__ and __LINE__ standard macros. - - * Added a sanity check for number constants being greater than - 0xFFFFFFFF. The warning can be disabled. - - * Added the %0 token whereby a variadic multi-line macro can tell - how many parameters it's been given in a specific invocation. - - * Added `%rotate', allowing multi-line macro parameters to be cycled. - - * Added the `*' option for the maximum parameter count on multi-line - macros, allowing them to take arbitrarily many parameters. - - * Added the ability for the user-level forms of EXTERN, GLOBAL and - COMMON to take more than one argument. - - * Added the IMPORT and EXPORT directives in OBJ format, to deal with - Windows DLLs. - - * Added some more preprocessor `%if' constructs: `%ifidn' / - `%ifidni' (exact textual identity), and `%ifid' / `%ifnum' / - `%ifstr' (token type testing). - - * Added the ability to distinguish SHL AX,1 (the 8086 version) from - SHL AX,BYTE 1 (the 286-and-upwards version whose constant happens - to be 1). - - * Added NetBSD/FreeBSD/OpenBSD's variant of a.out format, complete - with PIC shared library features. - - * Changed NASM's idiosyncratic handling of FCLEX, FDISI, FENI, - FINIT, FSAVE, FSTCW, FSTENV, and FSTSW to bring it into line with - the otherwise accepted standard. The previous behaviour, though it - was a deliberate feature, was a deliberate feature based on a - misunderstanding. Apologies for the inconvenience. - - * Improved the flexibility of ABSOLUTE: you can now give it an - expression rather than being restricted to a constant, and it can - take relocatable arguments as well. - - * Added the ability for a variable to be declared as EXTERN multiple - times, and the subsequent definitions are just ignored. - - * We now allow instruction prefixes (CS, DS, LOCK, REPZ etc) to be - alone on a line (without a following instruction). - - * Improved sanity checks on whether the arguments to EXTERN, GLOBAL - and COMMON are valid identifiers. - - * Added misc/exebin.mac to allow direct generation of .EXE files by - hacking up an EXE header using DB and DW; also added - test/binexe.asm to demonstrate the use of this. Thanks to Yann - Guidon for contributing the EXE header code. - - * ndisasm forgot to check whether the input file had been - successfully opened. Now it does. Doh! - - * Added the Cyrix extensions to the MMX instruction set. - - * Added a hinting mechanism to allow [EAX+EBX] and [EBX+EAX] to be - assembled differently. This is important since [ESI+EBP] and - [EBP+ESI] have different default base segment registers. - - * Added support for the PharLap OMF extension for 4096-byte segment - alignment. - diff --git a/doc/info/nasm.info-3 b/doc/info/nasm.info-3 deleted file mode 100644 index f70d131..0000000 --- a/doc/info/nasm.info-3 +++ /dev/null @@ -1,1284 +0,0 @@ -This is nasm.info, produced by makeinfo version 4.13 from nasmdoc.texi. - -INFO-DIR-SECTION Programming -START-INFO-DIR-ENTRY -* NASM: (nasm). The Netwide Assembler for x86. -END-INFO-DIR-ENTRY - - This file documents NASM, the Netwide Assembler: an assembler -targetting the Intel x86 series of processors, with portable source. - - Copyright 1996-2009 The NASM Development Team - - This document is redistributable under the license given in the file -"COPYING" distributed in the NASM archive. - - -File: nasm.info, Node: Section C.3.3, Next: Section C.3.4, Prev: Section C.3.2, Up: Section C.3 - -C.3.3. Version 0.95 released July 1997 --------------------------------------- - - * Fixed yet another ELF bug. This one manifested if the user relied - on the default segment, and attempted to define global symbols - without first explicitly declaring the target segment. - - * Added makefiles (for NASM and the RDF tools) to build Win32 - console apps under Symantec C++. Donated by Mark Junker. - - * Added `macros.bas' and `insns.bas', QBasic versions of the Perl - scripts that convert `standard.mac' to `macros.c' and convert - `insns.dat' to `insnsa.c' and `insnsd.c'. Also thanks to Mark - Junker. - - * Changed the diassembled forms of the conditional instructions so - that JB is now emitted as JC, and other similar changes. Suggested - list by Ulrich Doewich. - - * Added `@' to the list of valid characters to begin an identifier - with. - - * Documentary changes, notably the addition of the `Common Problems' - section in nasm.doc. - - * Fixed a bug relating to 32-bit PC-relative fixups in OBJ. - - * Fixed a bug in perm_copy() in labels.c which was causing - exceptions in cleanup_labels() on some systems. - - * Positivity sanity check in TIMES argument changed from a warning - to an error following a further complaint. - - * Changed the acceptable limits on byte and word operands to allow - things like `~10111001b' to work. - - * Fixed a major problem in the preprocessor which caused seg-faults - if macro definitions contained blank lines or comment-only lines. - - * Fixed inadequate error checking on the commas separating the - arguments to `db', `dw' etc. - - * Fixed a crippling bug in the handling of macros with operand - counts defined with a `+' modifier. - - * Fixed a bug whereby object file formats which stored the input - file name in the output file (such as OBJ and COFF) weren't doing - so correctly when the output file name was specified on the - command line. - - * Removed [INC] and [INCLUDE] support for good, since they were - obsolete anyway. - - * Fixed a bug in OBJ which caused all fixups to be output in 16-bit - (old- format) FIXUPP records, rather than putting the 32-bit ones - in FIXUPP32 (new-format) records. - - * Added, tentatively, OS/2 object file support (as a minor variant - on OBJ). - - * Updates to Fox Cutter's Borland C makefile, Makefile.bc2. - - * Removed a spurious second fclose() on the output file. - - * Added the `-s' command line option to redirect all messages which - would go to stderr (errors, help text) to stdout instead. - - * Added the `-w' command line option to selectively suppress some - classes of assembly warning messages. - - * Added the `-p' pre-include and `-d' pre-define command-line - options. - - * Added an include file search path: the `-i' command line option. - - * Fixed a silly little preprocessor bug whereby starting a line with - a `%!' environment-variable reference caused an `unknown - directive' error. - - * Added the long-awaited listing file support: the `-l' command line - option. - - * Fixed a problem with OBJ format whereby, in the absence of any - explicit segment definition, non-global symbols declared in the - implicit default segment generated spurious EXTDEF records in the - output. - - * Added the NASM environment variable. - - * From this version forward, Win32 console-mode binaries will be - included in the DOS distribution in addition to the 16-bit - binaries. Added Makefile.vc for this purpose. - - * Added `return 0;' to test/objlink.c to prevent compiler warnings. - - * Added the __NASM_MAJOR__ and __NASM_MINOR__ standard defines. - - * Added an alternative memory-reference syntax in which prefixing an - operand with `&' is equivalent to enclosing it in square brackets, - at the request of Fox Cutter. - - * Errors in pass two now cause the program to return a non-zero - error code, which they didn't before. - - * Fixed the single-line macro cycle detection, which didn't work at - all on macros with no parameters (caused an infinite loop). Also - changed the behaviour of single-line macro cycle detection to work - like cpp, so that macros like `extrn' as given in the - documentation can be implemented. - - * Fixed the implementation of WRT, which was too restrictive in that - you couldn't do `mov ax,[di+abc wrt dgroup]' because (di+abc) - wasn't a relocatable reference. - - -File: nasm.info, Node: Section C.3.4, Next: Section C.3.5, Prev: Section C.3.3, Up: Section C.3 - -C.3.4. Version 0.94 released April 1997 ---------------------------------------- - - * Major item: added the macro processor. - - * Added undocumented instructions SMI, IBTS, XBTS and LOADALL286. - Also reorganised CMPXCHG instruction into early-486 and Pentium - forms. Thanks to Thobias Jones for the information. - - * Fixed two more stupid bugs in ELF, which were causing `ld' to - continue to seg-fault in a lot of non-trivial cases. - - * Fixed a seg-fault in the label manager. - - * Stopped FBLD and FBSTP from _requiring_ the TWORD keyword, which - is the only option for BCD loads/stores in any case. - - * Ensured FLDCW, FSTCW and FSTSW can cope with the WORD keyword, if - anyone bothers to provide it. Previously they complained unless no - keyword at all was present. - - * Some forms of FDIV/FDIVR and FSUB/FSUBR were still inverted: a - vestige of a bug that I thought had been fixed in 0.92. This was - fixed, hopefully for good this time... - - * Another minor phase error (insofar as a phase error can _ever_ be - minor) fixed, this one occurring in code of the form - - rol ax,forward_reference - forward_reference equ 1 - - * The number supplied to TIMES is now sanity-checked for positivity, - and also may be greater than 64K (which previously didn't work on - 16-bit systems). - - * Added Watcom C makefiles, and misc/pmw.bat, donated by Dominik - Behr. - - * Added the INCBIN pseudo-opcode. - - * Due to the advent of the preprocessor, the [INCLUDE] and [INC] - directives have become obsolete. They are still supported in this - version, with a warning, but won't be in the next. - - * Fixed a bug in OBJ format, which caused incorrect object records - to be output when absolute labels were made global. - - * Updates to RDOFF subdirectory, and changes to outrdf.c. - - -File: nasm.info, Node: Section C.3.5, Next: Section C.3.6, Prev: Section C.3.4, Up: Section C.3 - -C.3.5. Version 0.93 released January 1997 ------------------------------------------ - -This release went out in a great hurry after semi-crippling bugs were -found in 0.92. - - * Really _did_ fix the stack overflows this time. *blush* - - * Had problems with EA instruction sizes changing between passes, - when an offset contained a forward reference and so 4 bytes were - allocated for the offset in pass one; by pass two the symbol had - been defined and happened to be a small absolute value, so only 1 - byte got allocated, causing instruction size mismatch between - passes and hence incorrect address calculations. Fixed. - - * Stupid bug in the revised ELF section generation fixed (associated - string- table section for .symtab was hard-coded as 7, even when - this didn't fit with the real section table). Was causing `ld' to - seg-fault under Linux. - - * Included a new Borland C makefile, Makefile.bc2, donated by Fox - Cutter <lmb@comtch.iea.com>. - - -File: nasm.info, Node: Section C.3.6, Next: Section C.3.7, Prev: Section C.3.5, Up: Section C.3 - -C.3.6. Version 0.92 released January 1997 ------------------------------------------ - - * The FDIVP/FDIVRP and FSUBP/FSUBRP pairs had been inverted: this - was fixed. This also affected the LCC driver. - - * Fixed a bug regarding 32-bit effective addresses of the form - `[other_register+ESP]'. - - * Documentary changes, notably documentation of the fact that - Borland Win32 compilers use `obj' rather than `win32' object - format. - - * Fixed the COMENT record in OBJ files, which was formatted - incorrectly. - - * Fixed a bug causing segfaults in large RDF files. - - * OBJ format now strips initial periods from segment and group - definitions, in order to avoid complications with the local label - syntax. - - * Fixed a bug in disassembling far calls and jumps in NDISASM. - - * Added support for user-defined sections in COFF and ELF files. - - * Compiled the DOS binaries with a sensible amount of stack, to - prevent stack overflows on any arithmetic expression containing - parentheses. - - * Fixed a bug in handling of files that do not terminate in a - newline. - - -File: nasm.info, Node: Section C.3.7, Next: Section C.3.8, Prev: Section C.3.6, Up: Section C.3 - -C.3.7. Version 0.91 released November 1996 ------------------------------------------- - - * Loads of bug fixes. - - * Support for RDF added. - - * Support for DBG debugging format added. - - * Support for 32-bit extensions to Microsoft OBJ format added. - - * Revised for Borland C: some variable names changed, makefile added. - - * LCC support revised to actually work. - - * JMP/CALL NEAR/FAR notation added. - - * `a16', `o16', `a32' and `o32' prefixes added. - - * Range checking on short jumps implemented. - - * MMX instruction support added. - - * Negative floating point constant support added. - - * Memory handling improved to bypass 64K barrier under DOS. - - * `$' prefix to force treatment of reserved words as identifiers - added. - - * Default-size mechanism for object formats added. - - * Compile-time configurability added. - - * `#', `@', `~' and c{?} are now valid characters in labels. - - * `-e' and `-k' options in NDISASM added. - - -File: nasm.info, Node: Section C.3.8, Prev: Section C.3.7, Up: Section C.3 - -C.3.8. Version 0.90 released October 1996 ------------------------------------------ - -First release version. First support for object file output. Other -changes from previous version (0.3x) too numerous to document. - - -File: nasm.info, Node: Index, Up: Top - -Index -***** - -* Menu: - -* `!' operator, unary (1): Section 3.5.7. -* `!=' operator (1): Section 4.4.4. -* `$$' token (1): Section 3.5. -* `$$' token (2): Section 7.9.3. -* `$', Here token (1): Section 3.5. -* `$', prefix (1): Section 3.1. -* `$', prefix (2): Section 3.4.1. -* `$', prefix (3): Section 7.13.2. -* `%' operator (1): Section 3.5.6. -* `%!' (1): Section 4.10.2. -* `%$' and `%$$' prefixes (1): Section 4.7.2. -* `%%' operator (1): Section 3.5.6. -* `%%' operator (2): Section 4.3.3. -* `%+' (1): Section 4.1.4. -* `%?' (1): Section 4.1.5. -* `%??' (1): Section 4.1.5. -* `%[' (1): Section 4.1.3. -* `&' operator (1): Section 3.5.3. -* `&&' operator (1): Section 4.4.4. -* `*' operator (1): Section 3.5.6. -* `+' modifier (1): Section 4.3.4. -* `+' operator, binary (1): Section 3.5.5. -* `+' operator, unary (1): Section 3.5.7. -* `-' operator, binary (1): Section 3.5.5. -* `-' operator, unary (1): Section 3.5.7. -* `..@' symbol prefix (1): Section 3.9. -* `..@' symbol prefix (2): Section 4.3.3. -* `/' operator (1): Section 3.5.6. -* `//' operator (1): Section 3.5.6. -* `<' operator (1): Section 4.4.4. -* `<<' operator (1): Section 3.5.4. -* `<=' operator (1): Section 4.4.4. -* `<>' operator (1): Section 4.4.4. -* `=' operator (1): Section 4.4.4. -* `==' operator (1): Section 4.4.4. -* `>' operator (1): Section 4.4.4. -* `>=' operator (1): Section 4.4.4. -* `>>' operator (1): Section 3.5.4. -* `?' MASM syntax (1): Section 3.2.2. -* `^' operator (1): Section 3.5.2. -* `^^' operator (1): Section 4.4.4. -* `|' operator (1): Section 3.5.1. -* `||' operator (1): Section 4.4.4. -* `~' operator (1): Section 3.5.7. -* `%0' parameter count (1): Section 4.3.5. -* `%0' parameter count (2): Section 4.3.6. -* `%+1' and `%-1' syntax (1): Section 4.3.9. -* 16-bit mode, versus 32-bit mode (1): Section 6.1. -* 64-bit displacement (1): Section 11.2. -* 64-bit immediate (1): Section 11.2. -* `-a' option (1): Section 2.1.21. -* `-a' option (2): Section A.3.3. -* `A16' (1): Section 3.1. -* `a16' (1): Section 10.3. -* `A32' (1): Section 3.1. -* `a32' (1): Section 10.3. -* `A64' (1): Section 3.1. -* `a64' (1): Section 10.3. -* `a86' (1): Section 1.1.1. -* `a86' (2): Section 2.2. -* `a86' (3): Section 2.2.2. -* `a86' (4): Section 2.2.6. -* `ABS' (1): Section 3.3. -* `ABSOLUTE' (1): Section 6.4. -* `ABSOLUTE' (2): Section 7.4.1. -* addition (1): Section 3.5.5. -* addressing, mixed-size (1): Section 10.2. -* address-size prefixes (1): Section 3.1. -* algebra (1): Section 3.3. -* `ALIGN' (1): Section 4.11.12. -* `ALIGN' (2): Section 5.2. -* `ALIGN' (3): Section 7.1.2. -* `ALIGN' (4): Section 7.4.1. -* `ALIGN', smart (1): Section 5.2. -* `ALIGNB' (1): Section 4.11.12. -* alignment, in `bin' sections (1): Section 7.1.2. -* alignment, in `elf' sections (1): Section 7.9.2. -* alignment, in `obj' sections (1): Section 7.4.1. -* alignment, in `win32' sections (1): Section 7.5.1. -* alignment, of `elf' common variables (1): Section 7.9.6. -* `ALIGNMODE' (1): Section 5.2. -* `__ALIGNMODE__' (1): Section 5.2. -* `ALINK' (1): Section 8.1.1. -* `alink.sourceforge.net' (1): Section 8.1.1. -* `all' (1): Section 2.1.24. -* `alloc' (1): Section 7.9.2. -* alternate register names (1): Section 5.1. -* `alt.lang.asm' (1): Section 1.1.1. -* `altreg' (1): Section 5.1. -* ambiguity (1): Section 2.2.3. -* `a.out', BSD version (1): Section 7.11. -* `a.out', Linux version (1): Section 7.10. -* `aout' (1): Section 7.10. -* `aoutb' (1): Section 7.11. -* `aoutb' (2): Section 9.2. -* `%arg' (1): Section 4.8.1. -* `arg' (1): Section 8.4.5. -* `arg' (2): Section 9.1.4. -* `as86' (1): Section 1.1.1. -* `as86' (2): Section 7.12. -* assembler directives (1): Chapter 6. -* assembly-time options (1): Section 2.1.18. -* `%assign' (1): Section 4.1.7. -* `ASSUME' (1): Section 2.2.4. -* `AT' (1): Section 4.11.11. -* Autoconf (1): Section 1.3.2. -* `autoexec.bat' (1): Section 1.3.1. -* `auto-sync' (1): Section A.3.3. -* `-b' (1): Section A.3. -* bin (1): Section 2.1.2. -* bin (2): Section 7.1. -* bin, multisection (1): Section 7.1.3. -* binary (1): Section 3.4.1. -* binary files (1): Section 3.2.3. -* bit shift (1): Section 3.5.4. -* `BITS' (1): Section 6.1. -* `BITS' (2): Section 7.1. -* `__BITS__' (1): Section 4.11.5. -* bitwise AND (1): Section 3.5.3. -* bitwise OR (1): Section 3.5.1. -* bitwise XOR (1): Section 3.5.2. -* block IFs (1): Section 4.7.5. -* boot loader (1): Section 7.1. -* boot sector (1): Section 12.1.3. -* Borland, Pascal (1): Section 8.5. -* Borland, Win32 compilers (1): Section 7.4. -* braces, after `%' sign (1): Section 4.3.8. -* braces, around macro parameters (1): Section 4.3. -* BSD (1): Section 9.2. -* `.bss' (1): Section 7.9.2. -* `.bss' (2): Section 7.10. -* `.bss' (3): Section 7.11. -* `.bss' (4): Section 7.12. -* `.bss' (5): Section 7.13. -* bugs (1): Section 12.2. -* `bugtracker' (1): Section 12.2. -* `BYTE' (1): Section 12.1.1. -* C calling convention (1): Section 8.4.3. -* C calling convention (2): Section 9.1.2. -* C symbol names (1): Section 8.4.1. -* `c16.mac' (1): Section 8.4.5. -* `c16.mac' (2): Section 8.5.3. -* `c32.mac' (1): Section 9.1.4. -* `CALL FAR' (1): Section 3.6. -* case sensitivity (1): Section 2.2.1. -* case sensitivity (2): Section 4.1.1. -* case sensitivity (3): Section 4.1.2. -* case sensitivity (4): Section 4.1.7. -* case sensitivity (5): Section 4.3. -* case sensitivity (6): Section 4.3.1. -* case sensitivity (7): Section 4.4.5. -* case sensitivity (8): Section 7.4.3. -* changing sections (1): Section 6.3. -* character constant (1): Section 3.2.1. -* character constant (2): Section 3.4.3. -* character strings (1): Section 3.4.2. -* circular references (1): Section 4.1.1. -* `CLASS' (1): Section 7.4.1. -* `%clear' (1): Section 4.11. -* `coff' (1): Section 7.7. -* colon (1): Section 3.1. -* `.COM' (1): Section 7.1. -* `.COM' (2): Section 8.2. -* command-line (1): Section 2.1. -* command-line (2): Chapter 7. -* commas in macro parameters (1): Section 4.3.4. -* `.comment' (1): Section 7.9.2. -* `COMMON' (1): Section 6.7. -* `COMMON' (2): Section 7.4.1. -* `COMMON', `elf' extensions to (1): Section 7.9.6. -* `COMMON', `obj' extensions to (1): Section 7.4.8. -* Common Object File Format (1): Section 7.7. -* common variables (1): Section 6.7. -* common variables, alignment in `elf' (1): Section 7.9.6. -* common variables, element size (1): Section 7.4.8. -* `comp.lang.asm.x86' (1): Section 1.1.1. -* `comp.lang.asm.x86' (2): Section 1.2. -* `comp.os.msdos.programmer' (1): Section 8.3. -* concatenating macro parameters (1): Section 4.3.8. -* concatenating strings (1): Section 4.2.1. -* condition codes as macro parameters (1): Section 4.3.9. -* conditional assembly (1): Section 4.4. -* conditional jumps (1): Section 12.1.2. -* conditional-return macro (1): Section 4.3.9. -* `configure' (1): Section 1.3.2. -* constants (1): Section 3.4. -* context stack (1): Section 4.7. -* context stack (2): Section 4.7.5. -* context-local labels (1): Section 4.7.2. -* context-local single-line macros (1): Section 4.7.3. -* counting macro parameters (1): Section 4.3.6. -* `CPU' (1): Section 6.8. -* `CPUID' (1): Section 3.4.3. -* creating contexts (1): Section 4.7.1. -* critical expression (1): Section 3.2.2. -* critical expression (2): Section 3.8. -* critical expression (3): Section 4.1.7. -* critical expression (4): Section 6.4. -* `-D' option (1): Section 2.1.18. -* `-d' option (1): Section 2.1.18. -* daily development snapshots (1): Section 1.2. -* `.data' (1): Section 7.9.2. -* `.data' (2): Section 7.10. -* `.data' (3): Section 7.11. -* `.data' (4): Section 7.12. -* `.data' (5): Section 7.13. -* `_DATA' (1): Section 8.4.2. -* `data' (1): Section 7.9.5. -* `data' (2): Section 7.13.3. -* data structure (1): Section 8.4.4. -* data structure (2): Section 9.1.3. -* `__DATE__' (1): Section 4.11.7. -* `__DATE_NUM__' (1): Section 4.11.7. -* `DB' (1): Section 3.2. -* `DB' (2): Section 3.2.1. -* `DB' (3): Section 3.4.4. -* `DB' (4): Section 3.4.6. -* `dbg' (1): Section 7.14. -* `DD' (1): Section 3.2. -* `DD' (2): Section 3.2.1. -* `DD' (3): Section 3.4.4. -* `DD' (4): Section 3.4.6. -* debug information (1): Section 2.1.12. -* debug information format (1): Section 2.1.11. -* declaring structures (1): Section 4.11.10. -* `DEFAULT' (1): Section 6.2. -* `default' (1): Section 7.9.5. -* default macro parameters (1): Section 4.3.5. -* default name (1): Chapter 7. -* default-`WRT' mechanism (1): Section 7.4.7. -* `%define' (1): Section 2.1.18. -* `%define' (2): Section 4.1.1. -* defining sections (1): Section 6.3. -* `%defstr' (1): Section 4.1.8. -* `%deftok' (1): Section 4.1.9. -* `%depend' (1): Section 4.6.3. -* design goals (1): Section 2.2.2. -* DevPac (1): Section 3.2.3. -* DevPac (2): Section 3.9. -* disabling listing expansion (1): Section 4.3.10. -* division (1): Section 3.5.6. -* DJGPP (1): Section 7.7. -* DJGPP (2): Chapter 9. -* `djlink' (1): Section 8.1.1. -* DLL symbols, exporting (1): Section 7.4.5. -* DLL symbols, importing (1): Section 7.4.4. -* `DO' (1): Section 3.2. -* `DO' (2): Section 3.2.1. -* `DO' (3): Section 3.4.4. -* `DO' (4): Section 3.4.6. -* DOS (1): Section 1.3.1. -* DOS (2): Section 2.1.14. -* DOS (3): Section 2.1.15. -* DOS source archive (1): Section 1.3.1. -* `DQ' (1): Section 3.2. -* `DQ' (2): Section 3.2.1. -* `DQ' (3): Section 3.4.4. -* `DQ' (4): Section 3.4.6. -* `.drectve' (1): Section 7.5.1. -* `DT' (1): Section 3.2. -* `DT' (2): Section 3.2.1. -* `DT' (3): Section 3.4.4. -* `DT' (4): Section 3.4.6. -* `DUP' (1): Section 2.2.7. -* `DUP' (2): Section 3.2.5. -* `DW' (1): Section 3.2. -* `DW' (2): Section 3.2.1. -* `DW' (3): Section 3.4.4. -* `DW' (4): Section 3.4.6. -* `DWORD' (1): Section 3.1. -* `DY' (1): Section 3.2. -* `DY' (2): Section 3.2.1. -* `DY' (3): Section 3.4.4. -* `-E' option (1): Section 2.1.20. -* `-e' option (1): Section 2.1.20. -* `-e' option (2): Section A.3.4. -* effective addresses (1): Section 3.1. -* effective addresses (2): Section 3.3. -* element size, in common variables (1): Section 7.4.8. -* ELF (1): Section 7.9. -* ELF, shared libraries (1): Section 7.9.3. -* ELF, 16-bit code and (1): Section 7.9.7. -* elf, debug formats and (1): Section 7.9.8. -* `elf32' (1): Section 7.9. -* `elf64' (1): Section 7.9. -* `%elif' (1): Section 4.4. -* `%elif' (2): Section 4.4.4. -* `%elifctx' (1): Section 4.4.3. -* `%elifdef' (1): Section 4.4.1. -* `%elifempty' (1): Section 4.4.8. -* `%elifid' (1): Section 4.4.6. -* `%elifidn' (1): Section 4.4.5. -* `%elifidni' (1): Section 4.4.5. -* `%elifmacro' (1): Section 4.4.2. -* `%elifn' (1): Section 4.4. -* `%elifn' (2): Section 4.4.4. -* `%elifnctx' (1): Section 4.4.3. -* `%elifndef' (1): Section 4.4.1. -* `%elifnempty' (1): Section 4.4.8. -* `%elifnid' (1): Section 4.4.6. -* `%elifnidn' (1): Section 4.4.5. -* `%elifnidni' (1): Section 4.4.5. -* `%elifnmacro' (1): Section 4.4.2. -* `%elifnnum' (1): Section 4.4.6. -* `%elifnstr' (1): Section 4.4.6. -* `%elifntoken' (1): Section 4.4.7. -* `%elifnum' (1): Section 4.4.6. -* `%elifstr' (1): Section 4.4.6. -* `%eliftoken' (1): Section 4.4.7. -* `%else' (1): Section 4.4. -* `endproc' (1): Section 8.4.5. -* `endproc' (2): Section 9.1.4. -* `%endrep' (1): Section 4.5. -* `ENDSTRUC' (1): Section 4.11.10. -* `ENDSTRUC' (2): Section 6.4. -* environment (1): Section 2.1.28. -* `EQU' (1): Section 3.2. -* `EQU' (2): Section 3.2.4. -* `%error' (1): Section 4.9. -* `error' (1): Section 2.1.24. -* error messages (1): Section 2.1.14. -* error messages (2): Section 2.1.15. -* error reporting format (1): Section 2.1.13. -* escape sequences (1): Section 3.4.2. -* `EVEN' (1): Section 4.11.12. -* exact matches (1): Section 4.3.11. -* `.EXE' (1): Section 7.4. -* `.EXE' (2): Section 8.1. -* `EXE2BIN' (1): Section 8.2.2. -* `EXE_begin' (1): Section 8.1.2. -* `exebin.mac' (1): Section 8.1.2. -* `exec' (1): Section 7.9.2. -* Executable and Linkable Format (1): Section 7.9. -* `EXE_end' (1): Section 8.1.2. -* `EXE_stack' (1): Section 8.1.2. -* `%exitmacro' (1): Section 4.3.12. -* `%exitrep' (1): Section 4.5. -* `EXPORT' (1): Section 7.4.5. -* `export' (1): Section 7.13.3. -* exporting symbols (1): Section 6.6. -* expressions (1): Section 2.1.20. -* expressions (2): Section 3.5. -* extension (1): Section 2.1.1. -* extension (2): Chapter 7. -* `EXTERN' (1): Section 6.5. -* `EXTERN', `obj' extensions to (1): Section 7.4.7. -* `EXTERN', `rdf' extensions to (1): Section 7.13.4. -* extracting substrings (1): Section 4.2.3. -* `-F' option (1): Section 2.1.11. -* `-f' option (1): Section 2.1.2. -* `-f' option (2): Chapter 7. -* far call (1): Section 2.2.5. -* far common variables (1): Section 7.4.8. -* far pointer (1): Section 3.6. -* `FARCODE' (1): Section 8.4.5. -* `FARCODE' (2): Section 8.5.3. -* `%fatal' (1): Section 4.9. -* `__FILE__' (1): Section 4.11.4. -* `FLAT' (1): Section 7.4.1. -* flat memory model (1): Chapter 9. -* flat-form binary (1): Section 7.1. -* `FLOAT' (1): Section 6.9. -* `__FLOAT__' (1): Section 6.9. -* `__float128h__' (1): Section 3.4.6. -* `__float128l__' (1): Section 3.4.6. -* `__float16__' (1): Section 3.4.6. -* `__float32__' (1): Section 3.4.6. -* `__float64__' (1): Section 3.4.6. -* `__float8__' (1): Section 3.4.6. -* `__float80e__' (1): Section 3.4.6. -* `__float80m__' (1): Section 3.4.6. -* `__FLOAT_DAZ__' (1): Section 6.9. -* `float-denorm' (1): Section 2.1.24. -* floating-point, constants (1): Section 3.4.6. -* floating-point, constants (2): Section 6.9. -* floating-point, packed BCD constants (1): Section 3.4.7. -* floating-point (1): Section 2.2.6. -* floating-point (2): Section 3.1. -* floating-point (3): Section 3.2.1. -* floating-point (4): Section 3.4.6. -* `float-overflow' (1): Section 2.1.24. -* `__FLOAT_ROUND__' (1): Section 6.9. -* `float-toolong' (1): Section 2.1.24. -* `float-underflow' (1): Section 2.1.24. -* `follows=' (1): Section 7.1.3. -* format-specific directives (1): Chapter 6. -* frame pointer (1): Section 8.4.3. -* frame pointer (2): Section 8.5.1. -* frame pointer (3): Section 9.1.2. -* FreeBSD (1): Section 7.11. -* FreeBSD (2): Section 9.2. -* FreeLink (1): Section 8.1.1. -* `ftp.simtel.net' (1): Section 8.1.1. -* `function' (1): Section 7.9.5. -* `function' (2): Section 7.13.3. -* functions, C calling convention (1): Section 8.4.3. -* functions, C calling convention (2): Section 9.1.2. -* functions, Pascal calling convention (1): Section 8.5.1. -* `-g' option (1): Section 2.1.12. -* `gas' (1): Section 1.1.1. -* `gcc' (1): Section 1.1.1. -* `GLOBAL' (1): Section 6.6. -* `GLOBAL', `aoutb' extensions to (1): Section 7.9.5. -* `GLOBAL', `elf' extensions to (1): Section 7.9.5. -* `GLOBAL', `rdf' extensions to (1): Section 7.13.3. -* global offset table (1): Section 9.2. -* `_GLOBAL_OFFSET_TABLE_' (1): Section 7.9.3. -* `gnu-elf-extensions' (1): Section 2.1.24. -* `..got' (1): Section 7.9.3. -* `GOT' relocations (1): Section 9.2.3. -* GOT (1): Section 7.9.3. -* GOT (2): Section 9.2. -* `..gotoff' (1): Section 7.9.3. -* `GOTOFF' relocations (1): Section 9.2.2. -* `..gotpc' (1): Section 7.9.3. -* `GOTPC' relocations (1): Section 9.2.1. -* `..gottpoff' (1): Section 7.9.4. -* graphics (1): Section 3.2.3. -* greedy macro parameters (1): Section 4.3.4. -* `GROUP' (1): Section 7.4.2. -* groups (1): Section 3.6. -* `-h' (1): Section A.3. -* hexadecimal (1): Section 3.4.1. -* `hidden' (1): Section 7.9.5. -* hybrid syntaxes (1): Section 2.2.2. -* `-I' option (1): Section 2.1.16. -* `-i' option (1): Section 2.1.16. -* `-i' option (2): Section A.3.3. -* `%iassign' (1): Section 4.1.7. -* `%idefine' (1): Section 4.1.1. -* `%idefstr' (1): Section 4.1.8. -* `%ideftok' (1): Section 4.1.9. -* `IEND' (1): Section 4.11.11. -* `%if' (1): Section 4.4. -* `%if' (2): Section 4.4.4. -* `%ifctx' (1): Section 4.4.3. -* `%ifctx' (2): Section 4.7.5. -* `%ifdef' (1): Section 4.4.1. -* `%ifempty' (1): Section 4.4.8. -* `%ifid' (1): Section 4.4.6. -* `%ifidn' (1): Section 4.4.5. -* `%ifidni' (1): Section 4.4.5. -* `%ifmacro' (1): Section 4.4.2. -* `%ifn' (1): Section 4.4. -* `%ifn' (2): Section 4.4.4. -* `%ifnctx' (1): Section 4.4.3. -* `%ifndef' (1): Section 4.4.1. -* `%ifnempty' (1): Section 4.4.8. -* `%ifnid' (1): Section 4.4.6. -* `%ifnidn' (1): Section 4.4.5. -* `%ifnidni' (1): Section 4.4.5. -* `%ifnmacro' (1): Section 4.4.2. -* `%ifnnum' (1): Section 4.4.6. -* `%ifnstr' (1): Section 4.4.6. -* `%ifntoken' (1): Section 4.4.7. -* `%ifnum' (1): Section 4.4.6. -* `%ifstr' (1): Section 4.4.6. -* `%iftoken' (1): Section 4.4.7. -* `%imacro' (1): Section 4.3. -* `IMPORT' (1): Section 7.4.4. -* import library (1): Section 7.4.4. -* importing symbols (1): Section 6.5. -* `INCBIN' (1): Section 3.2. -* `INCBIN' (2): Section 3.2.3. -* `INCBIN' (3): Section 3.4.4. -* `%include' (1): Section 2.1.16. -* `%include' (2): Section 2.1.17. -* `%include' (3): Section 4.6.1. -* include search path (1): Section 2.1.16. -* including other files (1): Section 4.6.1. -* inefficient code (1): Section 12.1.1. -* infinite loop (1): Section 3.5. -* `__Infinity__' (1): Section 3.4.6. -* infinity (1): Section 3.4.6. -* informational section (1): Section 7.5.1. -* `INSTALL' (1): Section 1.3.2. -* installing (1): Section 1.3.1. -* instances of structures (1): Section 4.11.11. -* instruction list (1): Appendix B. -* intel hex (1): Section 7.2. -* Intel number formats (1): Section 3.4.6. -* `internal' (1): Section 7.9.5. -* `%irmacro' (1): Section 4.3.1. -* `ISTRUC' (1): Section 4.11.11. -* iterating over macro parameters (1): Section 4.3.7. -* `ith' (1): Section 7.2. -* `%ixdefine' (1): Section 4.1.2. -* `Jcc NEAR' (1): Section 12.1.2. -* `JMP DWORD' (1): Section 10.1. -* jumps, mixed-size (1): Section 10.1. -* `-k' (1): Section A.3.4. -* `-l' option (1): Section 2.1.3. -* label prefix (1): Section 3.9. -* `.lbss' (1): Section 7.9.2. -* `ld86' (1): Section 7.12. -* `.ldata' (1): Section 7.9.2. -* `LIBRARY' (1): Section 7.13.1. -* license (1): Section 1.1.2. -* `%line' (1): Section 4.10.1. -* `__LINE__' (1): Section 4.11.4. -* linker, free (1): Section 8.1.1. -* Linux, `a.out' (1): Section 7.10. -* Linux, `as86' (1): Section 7.12. -* Linux, ELF (1): Section 7.9. -* listing file (1): Section 2.1.3. -* little-endian (1): Section 3.4.3. -* `%local' (1): Section 4.8.3. -* local labels (1): Section 3.9. -* logical AND (1): Section 4.4.4. -* logical negation (1): Section 3.5.7. -* logical OR (1): Section 4.4.4. -* logical XOR (1): Section 4.4.4. -* `.lrodata' (1): Section 7.9.2. -* `-M' option (1): Section 2.1.4. -* Mach, object file format (1): Section 7.8. -* Mach-O (1): Section 7.8. -* `macho' (1): Section 7.8. -* `macho32' (1): Section 7.8. -* `macho64' (1): Section 7.8. -* MacOS X (1): Section 7.8. -* `%macro' (1): Section 4.3. -* macro indirection (1): Section 4.1.3. -* macro library (1): Section 2.1.16. -* macro processor (1): Chapter 4. -* `macro-defaults' (1): Section 2.1.24. -* macro-local labels (1): Section 4.3.3. -* `macro-params' (1): Section 2.1.24. -* macros (1): Section 3.2.5. -* `macro-selfref' (1): Section 2.1.24. -* `make' (1): Section 1.3.2. -* makefile dependencies (1): Section 2.1.4. -* makefile dependencies (2): Section 2.1.5. -* makefiles (1): Section 1.3.1. -* makefiles (2): Section 1.3.2. -* man pages (1): Section 1.3.2. -* map files (1): Section 7.1.4. -* `MASM' (1): Section 1.1.1. -* MASM (1): Section 2.2. -* MASM (2): Section 3.2.5. -* MASM (3): Section 7.4. -* `-MD' option (1): Section 2.1.7. -* memory models (1): Section 2.2.5. -* memory models (2): Section 8.4.2. -* memory operand (1): Section 3.1. -* memory references (1): Section 2.2.2. -* memory references (2): Section 3.3. -* `-MF' option (1): Section 2.1.6. -* `-MG' option (1): Section 2.1.5. -* Microsoft OMF (1): Section 7.4. -* minifloat (1): Section 3.4.6. -* Minix (1): Section 7.12. -* `misc' subdirectory (1): Section 8.1.2. -* `misc' subdirectory (2): Section 8.4.5. -* `misc' subdirectory (3): Section 9.1.4. -* mixed-language program (1): Section 8.4. -* mixed-size addressing (1): Section 10.2. -* mixed-size instruction (1): Section 10.1. -* `MODULE' (1): Section 7.13.2. -* modulo operators (1): Section 3.5.6. -* motorola s-records (1): Section 7.3. -* `-MP' option (1): Section 2.1.10. -* `-MQ' option (1): Section 2.1.9. -* MS-DOS (1): Section 7.1. -* MS-DOS device drivers (1): Section 8.3. -* `-MT' option (1): Section 2.1.8. -* multi-line macros (1): Section 2.1.24. -* multi-line macros (2): Section 4.3. -* multipass optimization (1): Section 2.1.22. -* multiple section names (1): Section 7.1. -* multiplication (1): Section 3.5.6. -* `multipush' macro (1): Section 4.3.7. -* multisection (1): Section 7.1.3. -* `__NaN__' (1): Section 3.4.6. -* NaN (1): Section 3.4.6. -* NASM version (1): Section 4.11.1. -* nasm version history (1): Appendix C. -* nasm version id (1): Section 4.11.2. -* nasm version string (1): Section 4.11.3. -* `nasm.1' (1): Section 1.3.2. -* `__NASMDEFSEG' (1): Section 7.4. -* `nasm-devel' (1): Section 1.2. -* `NASMENV' (1): Section 2.1.28. -* `nasm.exe' (1): Section 1.3.1. -* `nasm -hf' (1): Section 2.1.2. -* `__NASM_MAJOR__' (1): Section 4.11.1. -* `__NASM_MINOR__' (1): Section 4.11.1. -* `nasm.out' (1): Section 2.1.1. -* `___NASM_PATCHLEVEL__' (1): Section 4.11.1. -* `__NASM_SNAPSHOT__' (1): Section 4.11.1. -* `__NASM_SUBMINOR__' (1): Section 4.11.1. -* `__NASM_VER__' (1): Section 4.11.3. -* `__NASM_VERSION_ID__' (1): Section 4.11.2. -* `nasm-XXX-dos.zip' (1): Section 1.3.1. -* `nasm-XXX.tar.gz' (1): Section 1.3.2. -* `nasm-XXX-win32.zip' (1): Section 1.3.1. -* `nasm-XXX.zip' (1): Section 1.3.1. -* ndisasm (1): Appendix A. -* `ndisasm.1' (1): Section 1.3.2. -* `ndisasm.exe' (1): Section 1.3.1. -* near call (1): Section 2.2.5. -* near common variables (1): Section 7.4.8. -* NetBSD (1): Section 7.11. -* NetBSD (2): Section 9.2. -* new releases (1): Section 1.2. -* `noalloc' (1): Section 7.9.2. -* `nobits' (1): Section 7.1.3. -* `nobits' (2): Section 7.9.2. -* `noexec' (1): Section 7.9.2. -* `.nolist' (1): Section 4.3.10. -* `nowait' (1): Section 2.2.6. -* `nowrite' (1): Section 7.9.2. -* `number-overflow' (1): Section 2.1.24. -* numeric constants (1): Section 3.2.1. -* numeric constants (2): Section 3.4.1. -* `-O' option (1): Section 2.1.22. -* `-o' option (1): Section 2.1.1. -* `-o' option (2): Section A.3.1. -* `O16' (1): Section 3.1. -* `o16' (1): Section 10.3. -* `O32' (1): Section 3.1. -* `o32' (1): Section 10.3. -* `O64' (1): Section 3.1. -* `.OBJ' (1): Section 8.1. -* `obj' (1): Section 7.4. -* `object' (1): Section 7.9.5. -* `object' (2): Section 7.13.3. -* octal (1): Section 3.4.1. -* `OF_DBG' (1): Section 7.14. -* `OF_DEFAULT' (1): Section 2.1.2. -* `OFFSET' (1): Section 2.2.2. -* OMF (1): Section 7.4. -* omitted parameters (1): Section 4.3.5. -* one's complement (1): Section 3.5.7. -* OpenBSD (1): Section 7.11. -* OpenBSD (2): Section 9.2. -* operands (1): Section 3.1. -* operand-size prefixes (1): Section 3.1. -* operating system (1): Section 7.1. -* operating system, writing (1): Section 10.1. -* operators (1): Section 3.5. -* `ORG' (1): Section 7.1.1. -* `ORG' (2): Section 8.2.1. -* `ORG' (3): Section 8.2.2. -* `ORG' (4): Section 12.1.3. -* `orphan-labels' (1): Section 2.1.24. -* `orphan-labels' (2): Section 3.1. -* OS/2 (1): Section 7.4. -* OS/2 (2): Section 7.4.1. -* `osabi' (1): Section 7.9.1. -* other preprocessor directives (1): Section 4.10. -* out of range, jumps (1): Section 12.1.2. -* output file format (1): Section 2.1.2. -* output formats (1): Chapter 7. -* `__OUTPUT_FORMAT__' (1): Section 4.11.6. -* overlapping segments (1): Section 3.6. -* `OVERLAY' (1): Section 7.4.1. -* overloading, multi-line macros (1): Section 4.3.2. -* overloading, single-line macros (1): Section 4.1.1. -* `-P' option (1): Section 2.1.17. -* `-p' option (1): Section 2.1.17. -* `-p' option (2): Section 4.6.1. -* paradox (1): Section 3.8. -* `PASCAL' (1): Section 8.5.3. -* Pascal calling convention (1): Section 8.5.1. -* `__PASS__' (1): Section 4.11.9. -* `PATH' (1): Section 1.3.1. -* `%pathsearch' (1): Section 2.1.16. -* `%pathsearch' (2): Section 4.6.2. -* period (1): Section 3.9. -* Perl (1): Section 1.3.1. -* perverse (1): Section 2.1.16. -* PharLap (1): Section 7.4.1. -* PIC (1): Section 7.9.3. -* PIC (2): Section 7.11. -* PIC (3): Section 9.2. -* `..plt' (1): Section 7.9.3. -* `PLT' relocations (1): Section 7.9.3. -* `PLT' relocations (2): Section 9.2.4. -* `PLT' relocations (3): Section 9.2.5. -* plt relocations (1): Section 9.2.5. -* `%pop' (1): Section 4.7. -* `%pop' (2): Section 4.7.1. -* position-independent code (1): Section 7.9.3. -* position-independent code (2): Section 7.11. -* position-independent code (3): Section 9.2. -* `--postfix' (1): Section 2.1.27. -* precedence (1): Section 3.5. -* pre-defining macros (1): Section 2.1.18. -* pre-defining macros (2): Section 4.1.1. -* preferred (1): Section 3.6. -* `--prefix' (1): Section 2.1.27. -* pre-including files (1): Section 2.1.17. -* preprocess-only mode (1): Section 2.1.20. -* preprocessor (1): Section 2.1.20. -* preprocessor (2): Section 2.1.21. -* preprocessor (3): Section 3.2.4. -* preprocessor (4): Section 3.5.6. -* preprocessor (5): Chapter 4. -* preprocessor expressions (1): Section 2.1.20. -* preprocessor loops (1): Section 4.5. -* preprocessor variables (1): Section 4.1.7. -* primitive directives (1): Chapter 6. -* `PRIVATE' (1): Section 7.4.1. -* `proc' (1): Section 7.13.3. -* `proc' (2): Section 8.4.5. -* `proc' (3): Section 9.1.4. -* procedure linkage table (1): Section 7.9.3. -* procedure linkage table (2): Section 9.2.4. -* procedure linkage table (3): Section 9.2.5. -* processor mode (1): Section 6.1. -* `progbits' (1): Section 7.1.3. -* `progbits' (2): Section 7.9.2. -* program entry point (1): Section 7.4.6. -* program entry point (2): Section 8.1.1. -* program origin (1): Section 7.1.1. -* `protected' (1): Section 7.9.5. -* pseudo-instructions (1): Section 3.2. -* `PUBLIC' (1): Section 6.6. -* `PUBLIC' (2): Section 7.4.1. -* pure binary (1): Section 7.1. -* `%push' (1): Section 4.7. -* `%push' (2): Section 4.7.1. -* `__QNaN__' (1): Section 3.4.6. -* quick start (1): Section 2.2. -* `QWORD' (1): Section 3.1. -* `-r' (1): Section A.3. -* `rdf' (1): Section 7.13. -* `rdoff' subdirectory (1): Section 1.3.2. -* `rdoff' subdirectory (2): Section 7.13. -* recursive multi-line macros (1): Section 4.3.1. -* redirecting errors (1): Section 2.1.14. -* `REL' (1): Section 3.3. -* `REL' (2): Section 6.2. -* relational operators (1): Section 4.4.4. -* release candidates (1): Section 1.2. -* Relocatable Dynamic Object File Format (1): Section 7.13. -* relocations, PIC-specific (1): Section 7.9.3. -* removing contexts (1): Section 4.7.1. -* renaming contexts (1): Section 4.7.4. -* `%rep' (1): Section 3.2.5. -* `%rep' (2): Section 4.5. -* repeating (1): Section 3.2.5. -* repeating (2): Section 4.5. -* `%repl' (1): Section 4.7.4. -* reporting bugs (1): Section 12.2. -* `RESB' (1): Section 2.2.7. -* `RESB' (2): Section 3.2. -* `RESB' (3): Section 3.2.2. -* `RESD' (1): Section 3.2. -* `RESD' (2): Section 3.2.2. -* `RESO' (1): Section 3.2. -* `RESO' (2): Section 3.2.2. -* `RESQ' (1): Section 3.2. -* `RESQ' (2): Section 3.2.2. -* `REST' (1): Section 3.2. -* `REST' (2): Section 3.2.2. -* `RESW' (1): Section 3.2. -* `RESW' (2): Section 3.2.2. -* `RESY' (1): Section 3.2. -* `RESY' (2): Section 3.2.2. -* `%rmacro' (1): Section 4.3.1. -* `.rodata' (1): Section 7.9.2. -* `%rotate' (1): Section 4.3.7. -* rotating macro parameters (1): Section 4.3.7. -* `-s' option (1): Section 2.1.15. -* `-s' option (2): Section A.3.2. -* searching for include files (1): Section 4.6.1. -* `__SECT__' (1): Section 6.3.1. -* `__SECT__' (2): Section 6.4. -* `SECTION' (1): Section 6.3. -* `SECTION', `elf' extensions to (1): Section 7.9.2. -* `SECTION', `win32' extensions to (1): Section 7.5.1. -* section alignment, in `bin' (1): Section 7.1.2. -* section alignment, in `elf' (1): Section 7.9.2. -* section alignment, in `obj' (1): Section 7.4.1. -* section alignment, in `win32' (1): Section 7.5.1. -* section, bin extensions to (1): Section 7.1.2. -* `SEG' (1): Section 3.5.7. -* `SEG' (2): Section 3.6. -* `SEG' (3): Section 7.4. -* `SEGMENT' (1): Section 6.3. -* `SEGMENT', `elf' extensions to (1): Section 7.4.1. -* segment address (1): Section 3.5.7. -* segment address (2): Section 3.6. -* segment alignment, in `bin' (1): Section 7.1.2. -* segment alignment, in `obj' (1): Section 7.4.1. -* segment names, Borland Pascal (1): Section 8.5.2. -* segment override (1): Section 2.2.4. -* segment override (2): Section 3.1. -* segments (1): Section 3.6. -* segments, groups of (1): Section 7.4.2. -* separator character (1): Section 2.1.28. -* shared libraries (1): Section 7.11. -* shared libraries (2): Section 9.2. -* shared library (1): Section 7.9.5. -* `shift' command (1): Section 4.3.7. -* signed division (1): Section 3.5.6. -* signed modulo (1): Section 3.5.6. -* single-line macros (1): Section 4.1. -* size, of symbols (1): Section 7.9.5. -* `smartalign' (1): Section 5.2. -* `__SNaN__' (1): Section 3.4.6. -* snapshots, daily development (1): Section 1.2. -* Solaris x86 (1): Section 7.9. -* `-soname' (1): Section 9.2.6. -* sound (1): Section 3.2.3. -* source code (1): Section 1.3.1. -* source-listing file (1): Section 2.1.3. -* square brackets (1): Section 2.2.2. -* square brackets (2): Section 3.3. -* `srec' (1): Section 7.3. -* `STACK' (1): Section 7.4.1. -* stack relative preprocessor directives (1): Section 4.8. -* `%stacksize' (1): Section 4.8.2. -* standard macro packages (1): Chapter 5. -* standard macros (1): Section 4.11. -* standardized section names (1): Section 6.3. -* standardized section names (2): Section 7.5.1. -* standardized section names (3): Section 7.9.2. -* standardized section names (4): Section 7.10. -* standardized section names (5): Section 7.11. -* standardized section names (6): Section 7.12. -* standardized section names (7): Section 7.13. -* `..start' (1): Section 7.4.6. -* `..start' (2): Section 8.1.1. -* `start=' (1): Section 7.1.3. -* `stderr' (1): Section 2.1.14. -* `stdout' (1): Section 2.1.15. -* `%strcat' (1): Section 4.2.1. -* `STRICT' (1): Section 3.7. -* string constant (1): Section 3.2.1. -* string constants (1): Section 3.4.4. -* string length (1): Section 4.2.2. -* string manipulation in macros (1): Section 4.2. -* strings (1): Section 3.4.2. -* `%strlen' (1): Section 4.2.2. -* `STRUC' (1): Section 4.11.10. -* `STRUC' (2): Section 6.4. -* `STRUC' (3): Section 8.4.4. -* `STRUC' (4): Section 9.1.3. -* stub preprocessor (1): Section 2.1.21. -* `%substr' (1): Section 4.2.3. -* subtraction (1): Section 3.5.5. -* suppressible warning (1): Section 2.1.24. -* suppressing preprocessing (1): Section 2.1.21. -* switching between sections (1): Section 6.3. -* `..sym' (1): Section 7.9.3. -* symbol sizes, specifying (1): Section 7.9.5. -* symbol types, specifying (1): Section 7.9.5. -* symbols, exporting from DLLs (1): Section 7.4.5. -* symbols, importing from DLLs (1): Section 7.4.4. -* `synchronisation' (1): Section A.3.2. -* `.SYS' (1): Section 7.1. -* `.SYS' (2): Section 8.3. -* `-t' (1): Section 2.1.23. -* `TASM' (1): Section 1.1.1. -* `TASM' (2): Section 2.1.23. -* tasm (1): Section 2.2. -* tasm (2): Section 7.4. -* `.tbss' (1): Section 7.9.2. -* `TBYTE' (1): Section 2.2.7. -* `.tdata' (1): Section 7.9.2. -* `test' subdirectory (1): Section 8.1.1. -* testing, arbitrary numeric expressions (1): Section 4.4.4. -* testing, context stack (1): Section 4.4.3. -* testing, exact text identity (1): Section 4.4.5. -* testing, multi-line macro existence (1): Section 4.4.2. -* testing, single-line macro existence (1): Section 4.4.1. -* testing, token types (1): Section 4.4.6. -* `.text' (1): Section 7.9.2. -* `.text' (2): Section 7.10. -* `.text' (3): Section 7.11. -* `.text' (4): Section 7.12. -* `.text' (5): Section 7.13. -* `_TEXT' (1): Section 8.4.2. -* thread local storage (1): Section 7.9.4. -* `__TIME__' (1): Section 4.11.7. -* `__TIME_NUM__' (1): Section 4.11.7. -* `TIMES' (1): Section 3.2. -* `TIMES' (2): Section 3.2.5. -* `TIMES' (3): Section 3.8. -* `TIMES' (4): Section 12.1.3. -* `TIMES' (5): Section 12.1.4. -* `TLINK' (1): Section 8.2.2. -* `tls' (1): Section 7.9.2. -* `tls' (2): Section 7.9.4. -* `..tlsie' (1): Section 7.9.4. -* trailing colon (1): Section 3.1. -* `TWORD' (1): Section 2.2.7. -* `TWORD' (2): Section 3.1. -* type, of symbols (1): Section 7.9.5. -* `-U' option (1): Section 2.1.19. -* `-u' option (1): Section 2.1.19. -* `-u' option (2): Section A.3. -* unary operators (1): Section 3.5.7. -* `%undef' (1): Section 2.1.19. -* `%undef' (2): Section 4.1.6. -* undefining macros (1): Section 2.1.19. -* underscore, in C symbols (1): Section 8.4.1. -* Unicode (1): Section 3.4.2. -* Unicode (2): Section 3.4.5. -* uninitialized (1): Section 3.2. -* uninitialized (2): Section 3.2.2. -* uninitialized storage (1): Section 2.2.7. -* Unix (1): Section 1.3.2. -* Unix, SCO (1): Section 7.9. -* Unix, source archive (1): Section 1.3.2. -* Unix, System V (1): Section 7.9. -* UnixWare (1): Section 7.9. -* `%unmacro' (1): Section 4.3.11. -* unrolled loops (1): Section 3.2.5. -* unsigned division (1): Section 3.5.6. -* unsigned modulo (1): Section 3.5.6. -* `UPPERCASE' (1): Section 2.2.1. -* `UPPERCASE' (2): Section 7.4.3. -* `%use' (1): Section 4.6.4. -* `%use' (2): Chapter 5. -* `__USE_*__' (1): Section 4.11.8. -* `USE16' (1): Section 6.1.1. -* `USE16' (2): Section 7.4.1. -* `USE32' (1): Section 6.1.1. -* `USE32' (2): Section 7.4.1. -* `user' (1): Section 2.1.24. -* user-defined errors (1): Section 4.9. -* user-level assembler directives (1): Section 4.11. -* user-level directives (1): Chapter 6. -* `__UTC_DATE__' (1): Section 4.11.7. -* `__UTC_DATE_NUM__' (1): Section 4.11.7. -* `__UTC_TIME__' (1): Section 4.11.7. -* `__UTC_TIME_NUM__' (1): Section 4.11.7. -* UTF-16 (1): Section 3.4.5. -* UTF-32 (1): Section 3.4.5. -* UTF-8 (1): Section 3.4.2. -* `__utf16__' (1): Section 3.4.5. -* `__utf32__' (1): Section 3.4.5. -* `-v' option (1): Section 2.1.25. -* VAL (1): Section 8.1.1. -* valid characters (1): Section 3.1. -* variable types (1): Section 2.2.3. -* version (1): Section 2.1.25. -* version number of NASM (1): Section 4.11.1. -* `vfollows=' (1): Section 7.1.3. -* Visual C++ (1): Section 7.5. -* `vstart=' (1): Section 7.1.3. -* `-W' option (1): Section 2.1.24. -* `-w' option (1): Section 2.1.24. -* `%warning' (1): Section 4.9. -* warnings (1): Section 2.1.24. -* `[warning *warning-name]' (1): Section 2.1.24. -* `[warning +warning-name]' (1): Section 2.1.24. -* `[warning -warning-name]' (1): Section 2.1.24. -* website (1): Section 1.2. -* `win64' (1): Section 7.6. -* `win64' (2): Chapter 11. -* Win64 (1): Section 7.4. -* Win64 (2): Section 7.5. -* Win64 (3): Chapter 9. -* Windows (1): Section 8.1. -* `write' (1): Section 7.9.2. -* writing operating systems (1): Section 10.1. -* `WRT' (1): Section 3.6. -* `WRT' (2): Section 7.4. -* `WRT' (3): Section 7.9.3. -* `WRT' (4): Section 7.9.4. -* `WRT' (5): Section 7.11. -* `WRT ..got' (1): Section 9.2.3. -* `WRT ..gotoff' (1): Section 9.2.2. -* `WRT ..gotpc' (1): Section 9.2.1. -* `WRT ..plt' (1): Section 9.2.5. -* `WRT ..sym' (1): Section 9.2.4. -* `www.cpan.org' (1): Section 1.3.1. -* `www.delorie.com' (1): Section 8.1.1. -* `www.pcorner.com' (1): Section 8.1.1. -* `-X' option (1): Section 2.1.13. -* `x2ftp.oulu.fi' (1): Section 8.1.1. -* `%xdefine' (1): Section 4.1.2. -* `-y' option (1): Section 2.1.26. -* `-Z' option (1): Section 2.1.14. - - diff --git a/doc/inslist.src b/doc/inslist.src deleted file mode 100644 index a5c7155..0000000 --- a/doc/inslist.src +++ /dev/null @@ -1,3130 +0,0 @@ - -\S{} Special instructions... - -\c DB -\c DW -\c DD -\c DQ -\c DT -\c DO -\c DY -\c RESB imm 8086 -\c RESW -\c RESD -\c RESQ -\c REST -\c RESO -\c RESY - -\S{} Conventional instructions - -\c AAA 8086,NOLONG -\c AAD 8086,NOLONG -\c AAD imm 8086,NOLONG -\c AAM 8086,NOLONG -\c AAM imm 8086,NOLONG -\c AAS 8086,NOLONG -\c ADC mem,reg8 8086 -\c ADC reg8,reg8 8086 -\c ADC mem,reg16 8086 -\c ADC reg16,reg16 8086 -\c ADC mem,reg32 386 -\c ADC reg32,reg32 386 -\c ADC mem,reg64 X64 -\c ADC reg64,reg64 X64 -\c ADC reg8,mem 8086 -\c ADC reg8,reg8 8086 -\c ADC reg16,mem 8086 -\c ADC reg16,reg16 8086 -\c ADC reg32,mem 386 -\c ADC reg32,reg32 386 -\c ADC reg64,mem X64 -\c ADC reg64,reg64 X64 -\c ADC rm16,imm8 8086 -\c ADC rm32,imm8 386 -\c ADC rm64,imm8 X64 -\c ADC reg_al,imm 8086 -\c ADC reg_ax,sbyte16 8086 -\c ADC reg_ax,imm 8086 -\c ADC reg_eax,sbyte32 386 -\c ADC reg_eax,imm 386 -\c ADC reg_rax,sbyte64 X64 -\c ADC reg_rax,imm X64 -\c ADC rm8,imm 8086 -\c ADC rm16,imm 8086 -\c ADC rm32,imm 386 -\c ADC rm64,imm X64 -\c ADC mem,imm8 8086 -\c ADC mem,imm16 8086 -\c ADC mem,imm32 386 -\c ADD mem,reg8 8086 -\c ADD reg8,reg8 8086 -\c ADD mem,reg16 8086 -\c ADD reg16,reg16 8086 -\c ADD mem,reg32 386 -\c ADD reg32,reg32 386 -\c ADD mem,reg64 X64 -\c ADD reg64,reg64 X64 -\c ADD reg8,mem 8086 -\c ADD reg8,reg8 8086 -\c ADD reg16,mem 8086 -\c ADD reg16,reg16 8086 -\c ADD reg32,mem 386 -\c ADD reg32,reg32 386 -\c ADD reg64,mem X64 -\c ADD reg64,reg64 X64 -\c ADD rm16,imm8 8086 -\c ADD rm32,imm8 386 -\c ADD rm64,imm8 X64 -\c ADD reg_al,imm 8086 -\c ADD reg_ax,sbyte16 8086 -\c ADD reg_ax,imm 8086 -\c ADD reg_eax,sbyte32 386 -\c ADD reg_eax,imm 386 -\c ADD reg_rax,sbyte64 X64 -\c ADD reg_rax,imm X64 -\c ADD rm8,imm 8086 -\c ADD rm16,imm 8086 -\c ADD rm32,imm 386 -\c ADD rm64,imm X64 -\c ADD mem,imm8 8086 -\c ADD mem,imm16 8086 -\c ADD mem,imm32 386 -\c AND mem,reg8 8086 -\c AND reg8,reg8 8086 -\c AND mem,reg16 8086 -\c AND reg16,reg16 8086 -\c AND mem,reg32 386 -\c AND reg32,reg32 386 -\c AND mem,reg64 X64 -\c AND reg64,reg64 X64 -\c AND reg8,mem 8086 -\c AND reg8,reg8 8086 -\c AND reg16,mem 8086 -\c AND reg16,reg16 8086 -\c AND reg32,mem 386 -\c AND reg32,reg32 386 -\c AND reg64,mem X64 -\c AND reg64,reg64 X64 -\c AND rm16,imm8 8086 -\c AND rm32,imm8 386 -\c AND rm64,imm8 X64 -\c AND reg_al,imm 8086 -\c AND reg_ax,sbyte16 8086 -\c AND reg_ax,imm 8086 -\c AND reg_eax,sbyte32 386 -\c AND reg_eax,imm 386 -\c AND reg_rax,sbyte64 X64 -\c AND reg_rax,imm X64 -\c AND rm8,imm 8086 -\c AND rm16,imm 8086 -\c AND rm32,imm 386 -\c AND rm64,imm X64 -\c AND mem,imm8 8086 -\c AND mem,imm16 8086 -\c AND mem,imm32 386 -\c ARPL mem,reg16 286,PROT,NOLONG -\c ARPL reg16,reg16 286,PROT,NOLONG -\c BB0_RESET PENT,CYRIX,ND -\c BB1_RESET PENT,CYRIX,ND -\c BOUND reg16,mem 186,NOLONG -\c BOUND reg32,mem 386,NOLONG -\c BSF reg16,mem 386 -\c BSF reg16,reg16 386 -\c BSF reg32,mem 386 -\c BSF reg32,reg32 386 -\c BSF reg64,mem X64 -\c BSF reg64,reg64 X64 -\c BSR reg16,mem 386 -\c BSR reg16,reg16 386 -\c BSR reg32,mem 386 -\c BSR reg32,reg32 386 -\c BSR reg64,mem X64 -\c BSR reg64,reg64 X64 -\c BSWAP reg32 486 -\c BSWAP reg64 X64 -\c BT mem,reg16 386 -\c BT reg16,reg16 386 -\c BT mem,reg32 386 -\c BT reg32,reg32 386 -\c BT mem,reg64 X64 -\c BT reg64,reg64 X64 -\c BT rm16,imm 386 -\c BT rm32,imm 386 -\c BT rm64,imm X64 -\c BTC mem,reg16 386 -\c BTC reg16,reg16 386 -\c BTC mem,reg32 386 -\c BTC reg32,reg32 386 -\c BTC mem,reg64 X64 -\c BTC reg64,reg64 X64 -\c BTC rm16,imm 386 -\c BTC rm32,imm 386 -\c BTC rm64,imm X64 -\c BTR mem,reg16 386 -\c BTR reg16,reg16 386 -\c BTR mem,reg32 386 -\c BTR reg32,reg32 386 -\c BTR mem,reg64 X64 -\c BTR reg64,reg64 X64 -\c BTR rm16,imm 386 -\c BTR rm32,imm 386 -\c BTR rm64,imm X64 -\c BTS mem,reg16 386 -\c BTS reg16,reg16 386 -\c BTS mem,reg32 386 -\c BTS reg32,reg32 386 -\c BTS mem,reg64 X64 -\c BTS reg64,reg64 X64 -\c BTS rm16,imm 386 -\c BTS rm32,imm 386 -\c BTS rm64,imm X64 -\c CALL imm 8086 -\c CALL imm|near 8086 -\c CALL imm|far 8086,ND,NOLONG -\c CALL imm16 8086 -\c CALL imm16|near 8086 -\c CALL imm16|far 8086,ND,NOLONG -\c CALL imm32 386 -\c CALL imm32|near 386 -\c CALL imm32|far 386,ND,NOLONG -\c CALL imm:imm 8086,NOLONG -\c CALL imm16:imm 8086,NOLONG -\c CALL imm:imm16 8086,NOLONG -\c CALL imm32:imm 386,NOLONG -\c CALL imm:imm32 386,NOLONG -\c CALL mem|far 8086,NOLONG -\c CALL mem|far X64 -\c CALL mem16|far 8086 -\c CALL mem32|far 386 -\c CALL mem64|far X64 -\c CALL mem|near 8086 -\c CALL mem16|near 8086 -\c CALL mem32|near 386,NOLONG -\c CALL mem64|near X64 -\c CALL reg16 8086 -\c CALL reg32 386,NOLONG -\c CALL reg64 X64 -\c CALL mem 8086 -\c CALL mem16 8086 -\c CALL mem32 386,NOLONG -\c CALL mem64 X64 -\c CBW 8086 -\c CDQ 386 -\c CDQE X64 -\c CLC 8086 -\c CLD 8086 -\c CLGI X64,AMD -\c CLI 8086 -\c CLTS 286,PRIV -\c CMC 8086 -\c CMP mem,reg8 8086 -\c CMP reg8,reg8 8086 -\c CMP mem,reg16 8086 -\c CMP reg16,reg16 8086 -\c CMP mem,reg32 386 -\c CMP reg32,reg32 386 -\c CMP mem,reg64 X64 -\c CMP reg64,reg64 X64 -\c CMP reg8,mem 8086 -\c CMP reg8,reg8 8086 -\c CMP reg16,mem 8086 -\c CMP reg16,reg16 8086 -\c CMP reg32,mem 386 -\c CMP reg32,reg32 386 -\c CMP reg64,mem X64 -\c CMP reg64,reg64 X64 -\c CMP rm16,imm8 8086 -\c CMP rm32,imm8 386 -\c CMP rm64,imm8 X64 -\c CMP reg_al,imm 8086 -\c CMP reg_ax,sbyte16 8086 -\c CMP reg_ax,imm 8086 -\c CMP reg_eax,sbyte32 386 -\c CMP reg_eax,imm 386 -\c CMP reg_rax,sbyte64 X64 -\c CMP reg_rax,imm X64 -\c CMP rm8,imm 8086 -\c CMP rm16,imm 8086 -\c CMP rm32,imm 386 -\c CMP rm64,imm X64 -\c CMP mem,imm8 8086 -\c CMP mem,imm16 8086 -\c CMP mem,imm32 386 -\c CMPSB 8086 -\c CMPSD 386 -\c CMPSQ X64 -\c CMPSW 8086 -\c CMPXCHG mem,reg8 PENT -\c CMPXCHG reg8,reg8 PENT -\c CMPXCHG mem,reg16 PENT -\c CMPXCHG reg16,reg16 PENT -\c CMPXCHG mem,reg32 PENT -\c CMPXCHG reg32,reg32 PENT -\c CMPXCHG mem,reg64 X64 -\c CMPXCHG reg64,reg64 X64 -\c CMPXCHG486 mem,reg8 486,UNDOC,ND -\c CMPXCHG486 reg8,reg8 486,UNDOC,ND -\c CMPXCHG486 mem,reg16 486,UNDOC,ND -\c CMPXCHG486 reg16,reg16 486,UNDOC,ND -\c CMPXCHG486 mem,reg32 486,UNDOC,ND -\c CMPXCHG486 reg32,reg32 486,UNDOC,ND -\c CMPXCHG8B mem PENT -\c CMPXCHG16B mem X64 -\c CPUID PENT -\c CPU_READ PENT,CYRIX -\c CPU_WRITE PENT,CYRIX -\c CQO X64 -\c CWD 8086 -\c CWDE 386 -\c DAA 8086,NOLONG -\c DAS 8086,NOLONG -\c DEC reg16 8086,NOLONG -\c DEC reg32 386,NOLONG -\c DEC rm8 8086 -\c DEC rm16 8086 -\c DEC rm32 386 -\c DEC rm64 X64 -\c DIV rm8 8086 -\c DIV rm16 8086 -\c DIV rm32 386 -\c DIV rm64 X64 -\c DMINT P6,CYRIX -\c EMMS PENT,MMX -\c ENTER imm,imm 186 -\c EQU imm 8086 -\c EQU imm:imm 8086 -\c F2XM1 8086,FPU -\c FABS 8086,FPU -\c FADD mem32 8086,FPU -\c FADD mem64 8086,FPU -\c FADD fpureg|to 8086,FPU -\c FADD fpureg 8086,FPU -\c FADD fpureg,fpu0 8086,FPU -\c FADD fpu0,fpureg 8086,FPU -\c FADD 8086,FPU,ND -\c FADDP fpureg 8086,FPU -\c FADDP fpureg,fpu0 8086,FPU -\c FADDP 8086,FPU,ND -\c FBLD mem80 8086,FPU -\c FBLD mem 8086,FPU -\c FBSTP mem80 8086,FPU -\c FBSTP mem 8086,FPU -\c FCHS 8086,FPU -\c FCLEX 8086,FPU -\c FCMOVB fpureg P6,FPU -\c FCMOVB fpu0,fpureg P6,FPU -\c FCMOVB P6,FPU,ND -\c FCMOVBE fpureg P6,FPU -\c FCMOVBE fpu0,fpureg P6,FPU -\c FCMOVBE P6,FPU,ND -\c FCMOVE fpureg P6,FPU -\c FCMOVE fpu0,fpureg P6,FPU -\c FCMOVE P6,FPU,ND -\c FCMOVNB fpureg P6,FPU -\c FCMOVNB fpu0,fpureg P6,FPU -\c FCMOVNB P6,FPU,ND -\c FCMOVNBE fpureg P6,FPU -\c FCMOVNBE fpu0,fpureg P6,FPU -\c FCMOVNBE P6,FPU,ND -\c FCMOVNE fpureg P6,FPU -\c FCMOVNE fpu0,fpureg P6,FPU -\c FCMOVNE P6,FPU,ND -\c FCMOVNU fpureg P6,FPU -\c FCMOVNU fpu0,fpureg P6,FPU -\c FCMOVNU P6,FPU,ND -\c FCMOVU fpureg P6,FPU -\c FCMOVU fpu0,fpureg P6,FPU -\c FCMOVU P6,FPU,ND -\c FCOM mem32 8086,FPU -\c FCOM mem64 8086,FPU -\c FCOM fpureg 8086,FPU -\c FCOM fpu0,fpureg 8086,FPU -\c FCOM 8086,FPU,ND -\c FCOMI fpureg P6,FPU -\c FCOMI fpu0,fpureg P6,FPU -\c FCOMI P6,FPU,ND -\c FCOMIP fpureg P6,FPU -\c FCOMIP fpu0,fpureg P6,FPU -\c FCOMIP P6,FPU,ND -\c FCOMP mem32 8086,FPU -\c FCOMP mem64 8086,FPU -\c FCOMP fpureg 8086,FPU -\c FCOMP fpu0,fpureg 8086,FPU -\c FCOMP 8086,FPU,ND -\c FCOMPP 8086,FPU -\c FCOS 386,FPU -\c FDECSTP 8086,FPU -\c FDISI 8086,FPU -\c FDIV mem32 8086,FPU -\c FDIV mem64 8086,FPU -\c FDIV fpureg|to 8086,FPU -\c FDIV fpureg 8086,FPU -\c FDIV fpureg,fpu0 8086,FPU -\c FDIV fpu0,fpureg 8086,FPU -\c FDIV 8086,FPU,ND -\c FDIVP fpureg 8086,FPU -\c FDIVP fpureg,fpu0 8086,FPU -\c FDIVP 8086,FPU,ND -\c FDIVR mem32 8086,FPU -\c FDIVR mem64 8086,FPU -\c FDIVR fpureg|to 8086,FPU -\c FDIVR fpureg,fpu0 8086,FPU -\c FDIVR fpureg 8086,FPU -\c FDIVR fpu0,fpureg 8086,FPU -\c FDIVR 8086,FPU,ND -\c FDIVRP fpureg 8086,FPU -\c FDIVRP fpureg,fpu0 8086,FPU -\c FDIVRP 8086,FPU,ND -\c FEMMS PENT,3DNOW -\c FENI 8086,FPU -\c FFREE fpureg 8086,FPU -\c FFREE 8086,FPU -\c FFREEP fpureg 286,FPU,UNDOC -\c FFREEP 286,FPU,UNDOC -\c FIADD mem32 8086,FPU -\c FIADD mem16 8086,FPU -\c FICOM mem32 8086,FPU -\c FICOM mem16 8086,FPU -\c FICOMP mem32 8086,FPU -\c FICOMP mem16 8086,FPU -\c FIDIV mem32 8086,FPU -\c FIDIV mem16 8086,FPU -\c FIDIVR mem32 8086,FPU -\c FIDIVR mem16 8086,FPU -\c FILD mem32 8086,FPU -\c FILD mem16 8086,FPU -\c FILD mem64 8086,FPU -\c FIMUL mem32 8086,FPU -\c FIMUL mem16 8086,FPU -\c FINCSTP 8086,FPU -\c FINIT 8086,FPU -\c FIST mem32 8086,FPU -\c FIST mem16 8086,FPU -\c FISTP mem32 8086,FPU -\c FISTP mem16 8086,FPU -\c FISTP mem64 8086,FPU -\c FISTTP mem16 PRESCOTT,FPU -\c FISTTP mem32 PRESCOTT,FPU -\c FISTTP mem64 PRESCOTT,FPU -\c FISUB mem32 8086,FPU -\c FISUB mem16 8086,FPU -\c FISUBR mem32 8086,FPU -\c FISUBR mem16 8086,FPU -\c FLD mem32 8086,FPU -\c FLD mem64 8086,FPU -\c FLD mem80 8086,FPU -\c FLD fpureg 8086,FPU -\c FLD 8086,FPU,ND -\c FLD1 8086,FPU -\c FLDCW mem 8086,FPU,SW -\c FLDENV mem 8086,FPU -\c FLDL2E 8086,FPU -\c FLDL2T 8086,FPU -\c FLDLG2 8086,FPU -\c FLDLN2 8086,FPU -\c FLDPI 8086,FPU -\c FLDZ 8086,FPU -\c FMUL mem32 8086,FPU -\c FMUL mem64 8086,FPU -\c FMUL fpureg|to 8086,FPU -\c FMUL fpureg,fpu0 8086,FPU -\c FMUL fpureg 8086,FPU -\c FMUL fpu0,fpureg 8086,FPU -\c FMUL 8086,FPU,ND -\c FMULP fpureg 8086,FPU -\c FMULP fpureg,fpu0 8086,FPU -\c FMULP 8086,FPU,ND -\c FNCLEX 8086,FPU -\c FNDISI 8086,FPU -\c FNENI 8086,FPU -\c FNINIT 8086,FPU -\c FNOP 8086,FPU -\c FNSAVE mem 8086,FPU -\c FNSTCW mem 8086,FPU,SW -\c FNSTENV mem 8086,FPU -\c FNSTSW mem 8086,FPU,SW -\c FNSTSW reg_ax 286,FPU -\c FPATAN 8086,FPU -\c FPREM 8086,FPU -\c FPREM1 386,FPU -\c FPTAN 8086,FPU -\c FRNDINT 8086,FPU -\c FRSTOR mem 8086,FPU -\c FSAVE mem 8086,FPU -\c FSCALE 8086,FPU -\c FSETPM 286,FPU -\c FSIN 386,FPU -\c FSINCOS 386,FPU -\c FSQRT 8086,FPU -\c FST mem32 8086,FPU -\c FST mem64 8086,FPU -\c FST fpureg 8086,FPU -\c FST 8086,FPU,ND -\c FSTCW mem 8086,FPU,SW -\c FSTENV mem 8086,FPU -\c FSTP mem32 8086,FPU -\c FSTP mem64 8086,FPU -\c FSTP mem80 8086,FPU -\c FSTP fpureg 8086,FPU -\c FSTP 8086,FPU,ND -\c FSTSW mem 8086,FPU,SW -\c FSTSW reg_ax 286,FPU -\c FSUB mem32 8086,FPU -\c FSUB mem64 8086,FPU -\c FSUB fpureg|to 8086,FPU -\c FSUB fpureg,fpu0 8086,FPU -\c FSUB fpureg 8086,FPU -\c FSUB fpu0,fpureg 8086,FPU -\c FSUB 8086,FPU,ND -\c FSUBP fpureg 8086,FPU -\c FSUBP fpureg,fpu0 8086,FPU -\c FSUBP 8086,FPU,ND -\c FSUBR mem32 8086,FPU -\c FSUBR mem64 8086,FPU -\c FSUBR fpureg|to 8086,FPU -\c FSUBR fpureg,fpu0 8086,FPU -\c FSUBR fpureg 8086,FPU -\c FSUBR fpu0,fpureg 8086,FPU -\c FSUBR 8086,FPU,ND -\c FSUBRP fpureg 8086,FPU -\c FSUBRP fpureg,fpu0 8086,FPU -\c FSUBRP 8086,FPU,ND -\c FTST 8086,FPU -\c FUCOM fpureg 386,FPU -\c FUCOM fpu0,fpureg 386,FPU -\c FUCOM 386,FPU,ND -\c FUCOMI fpureg P6,FPU -\c FUCOMI fpu0,fpureg P6,FPU -\c FUCOMI P6,FPU,ND -\c FUCOMIP fpureg P6,FPU -\c FUCOMIP fpu0,fpureg P6,FPU -\c FUCOMIP P6,FPU,ND -\c FUCOMP fpureg 386,FPU -\c FUCOMP fpu0,fpureg 386,FPU -\c FUCOMP 386,FPU,ND -\c FUCOMPP 386,FPU -\c FXAM 8086,FPU -\c FXCH fpureg 8086,FPU -\c FXCH fpureg,fpu0 8086,FPU -\c FXCH fpu0,fpureg 8086,FPU -\c FXCH 8086,FPU,ND -\c FXTRACT 8086,FPU -\c FYL2X 8086,FPU -\c FYL2XP1 8086,FPU -\c HLT 8086,PRIV -\c IBTS mem,reg16 386,SW,UNDOC,ND -\c IBTS reg16,reg16 386,UNDOC,ND -\c IBTS mem,reg32 386,SD,UNDOC,ND -\c IBTS reg32,reg32 386,UNDOC,ND -\c ICEBP 386,ND -\c IDIV rm8 8086 -\c IDIV rm16 8086 -\c IDIV rm32 386 -\c IDIV rm64 X64 -\c IMUL rm8 8086 -\c IMUL rm16 8086 -\c IMUL rm32 386 -\c IMUL rm64 X64 -\c IMUL reg16,mem 386 -\c IMUL reg16,reg16 386 -\c IMUL reg32,mem 386 -\c IMUL reg32,reg32 386 -\c IMUL reg64,mem X64 -\c IMUL reg64,reg64 X64 -\c IMUL reg16,mem,imm8 186 -\c IMUL reg16,mem,sbyte16 186,ND -\c IMUL reg16,mem,imm16 186 -\c IMUL reg16,mem,imm 186,ND -\c IMUL reg16,reg16,imm8 186 -\c IMUL reg16,reg16,sbyte16 186,ND -\c IMUL reg16,reg16,imm16 186 -\c IMUL reg16,reg16,imm 186,ND -\c IMUL reg32,mem,imm8 386 -\c IMUL reg32,mem,sbyte32 386,ND -\c IMUL reg32,mem,imm32 386 -\c IMUL reg32,mem,imm 386,ND -\c IMUL reg32,reg32,imm8 386 -\c IMUL reg32,reg32,sbyte32 386,ND -\c IMUL reg32,reg32,imm32 386 -\c IMUL reg32,reg32,imm 386,ND -\c IMUL reg64,mem,imm8 X64 -\c IMUL reg64,mem,sbyte64 X64,ND -\c IMUL reg64,mem,imm32 X64 -\c IMUL reg64,mem,imm X64,ND -\c IMUL reg64,reg64,imm8 X64 -\c IMUL reg64,reg64,sbyte64 X64,ND -\c IMUL reg64,reg64,imm32 X64 -\c IMUL reg64,reg64,imm X64,ND -\c IMUL reg16,imm8 186 -\c IMUL reg16,sbyte16 186,ND -\c IMUL reg16,imm16 186 -\c IMUL reg16,imm 186,ND -\c IMUL reg32,imm8 386 -\c IMUL reg32,sbyte32 386,ND -\c IMUL reg32,imm32 386 -\c IMUL reg32,imm 386,ND -\c IMUL reg64,imm8 X64 -\c IMUL reg64,sbyte64 X64,ND -\c IMUL reg64,imm32 X64 -\c IMUL reg64,imm X64,ND -\c IN reg_al,imm 8086 -\c IN reg_ax,imm 8086 -\c IN reg_eax,imm 386 -\c IN reg_al,reg_dx 8086 -\c IN reg_ax,reg_dx 8086 -\c IN reg_eax,reg_dx 386 -\c INC reg16 8086,NOLONG -\c INC reg32 386,NOLONG -\c INC rm8 8086 -\c INC rm16 8086 -\c INC rm32 386 -\c INC rm64 X64 -\c INCBIN -\c INSB 186 -\c INSD 386 -\c INSW 186 -\c INT imm 8086 -\c INT01 386,ND -\c INT1 386 -\c INT03 8086,ND -\c INT3 8086 -\c INTO 8086,NOLONG -\c INVD 486,PRIV -\c INVLPG mem 486,PRIV -\c INVLPGA reg_ax,reg_ecx X86_64,AMD,NOLONG -\c INVLPGA reg_eax,reg_ecx X86_64,AMD -\c INVLPGA reg_rax,reg_ecx X64,AMD -\c INVLPGA X86_64,AMD -\c IRET 8086 -\c IRETD 386 -\c IRETQ X64 -\c IRETW 8086 -\c JCXZ imm 8086,NOLONG -\c JECXZ imm 386 -\c JRCXZ imm X64 -\c JMP imm|short 8086 -\c JMP imm 8086,ND -\c JMP imm 8086 -\c JMP imm|near 8086,ND -\c JMP imm|far 8086,ND,NOLONG -\c JMP imm16 8086 -\c JMP imm16|near 8086,ND -\c JMP imm16|far 8086,ND,NOLONG -\c JMP imm32 386 -\c JMP imm32|near 386,ND -\c JMP imm32|far 386,ND,NOLONG -\c JMP imm:imm 8086,NOLONG -\c JMP imm16:imm 8086,NOLONG -\c JMP imm:imm16 8086,NOLONG -\c JMP imm32:imm 386,NOLONG -\c JMP imm:imm32 386,NOLONG -\c JMP mem|far 8086,NOLONG -\c JMP mem|far X64 -\c JMP mem16|far 8086 -\c JMP mem32|far 386 -\c JMP mem64|far X64 -\c JMP mem|near 8086 -\c JMP mem16|near 8086 -\c JMP mem32|near 386,NOLONG -\c JMP mem64|near X64 -\c JMP reg16 8086 -\c JMP reg32 386,NOLONG -\c JMP reg64 X64 -\c JMP mem 8086 -\c JMP mem16 8086 -\c JMP mem32 386,NOLONG -\c JMP mem64 X64 -\c JMPE imm IA64 -\c JMPE imm16 IA64 -\c JMPE imm32 IA64 -\c JMPE rm16 IA64 -\c JMPE rm32 IA64 -\c LAHF 8086 -\c LAR reg16,mem 286,PROT,SW -\c LAR reg16,reg16 286,PROT -\c LAR reg16,reg32 386,PROT -\c LAR reg16,reg64 X64,PROT,ND -\c LAR reg32,mem 386,PROT,SW -\c LAR reg32,reg16 386,PROT -\c LAR reg32,reg32 386,PROT -\c LAR reg32,reg64 X64,PROT,ND -\c LAR reg64,mem X64,PROT,SW -\c LAR reg64,reg16 X64,PROT -\c LAR reg64,reg32 X64,PROT -\c LAR reg64,reg64 X64,PROT -\c LDS reg16,mem 8086,NOLONG -\c LDS reg32,mem 386,NOLONG -\c LEA reg16,mem 8086 -\c LEA reg32,mem 386 -\c LEA reg64,mem X64 -\c LEAVE 186 -\c LES reg16,mem 8086,NOLONG -\c LES reg32,mem 386,NOLONG -\c LFENCE X64,AMD -\c LFS reg16,mem 386 -\c LFS reg32,mem 386 -\c LGDT mem 286,PRIV -\c LGS reg16,mem 386 -\c LGS reg32,mem 386 -\c LIDT mem 286,PRIV -\c LLDT mem 286,PROT,PRIV -\c LLDT mem16 286,PROT,PRIV -\c LLDT reg16 286,PROT,PRIV -\c LMSW mem 286,PRIV -\c LMSW mem16 286,PRIV -\c LMSW reg16 286,PRIV -\c LOADALL 386,UNDOC -\c LOADALL286 286,UNDOC -\c LODSB 8086 -\c LODSD 386 -\c LODSQ X64 -\c LODSW 8086 -\c LOOP imm 8086 -\c LOOP imm,reg_cx 8086,NOLONG -\c LOOP imm,reg_ecx 386 -\c LOOP imm,reg_rcx X64 -\c LOOPE imm 8086 -\c LOOPE imm,reg_cx 8086,NOLONG -\c LOOPE imm,reg_ecx 386 -\c LOOPE imm,reg_rcx X64 -\c LOOPNE imm 8086 -\c LOOPNE imm,reg_cx 8086,NOLONG -\c LOOPNE imm,reg_ecx 386 -\c LOOPNE imm,reg_rcx X64 -\c LOOPNZ imm 8086 -\c LOOPNZ imm,reg_cx 8086,NOLONG -\c LOOPNZ imm,reg_ecx 386 -\c LOOPNZ imm,reg_rcx X64 -\c LOOPZ imm 8086 -\c LOOPZ imm,reg_cx 8086,NOLONG -\c LOOPZ imm,reg_ecx 386 -\c LOOPZ imm,reg_rcx X64 -\c LSL reg16,mem 286,PROT,SW -\c LSL reg16,reg16 286,PROT -\c LSL reg16,reg32 386,PROT -\c LSL reg16,reg64 X64,PROT,ND -\c LSL reg32,mem 386,PROT,SW -\c LSL reg32,reg16 386,PROT -\c LSL reg32,reg32 386,PROT -\c LSL reg32,reg64 X64,PROT,ND -\c LSL reg64,mem X64,PROT,SW -\c LSL reg64,reg16 X64,PROT -\c LSL reg64,reg32 X64,PROT -\c LSL reg64,reg64 X64,PROT -\c LSS reg16,mem 386 -\c LSS reg32,mem 386 -\c LTR mem 286,PROT,PRIV -\c LTR mem16 286,PROT,PRIV -\c LTR reg16 286,PROT,PRIV -\c MFENCE X64,AMD -\c MONITOR PRESCOTT -\c MONITOR reg_eax,reg_ecx,reg_edx PRESCOTT,ND -\c MONITOR reg_rax,reg_ecx,reg_edx X64,ND -\c MOV mem,reg_sreg 8086 -\c MOV reg16,reg_sreg 8086 -\c MOV reg32,reg_sreg 386 -\c MOV reg_sreg,mem 8086 -\c MOV reg_sreg,reg16 8086 -\c MOV reg_sreg,reg32 386 -\c MOV reg_al,mem_offs 8086 -\c MOV reg_ax,mem_offs 8086 -\c MOV reg_eax,mem_offs 386 -\c MOV reg_rax,mem_offs X64 -\c MOV mem_offs,reg_al 8086 -\c MOV mem_offs,reg_ax 8086 -\c MOV mem_offs,reg_eax 386 -\c MOV mem_offs,reg_rax X64 -\c MOV reg32,reg_creg 386,PRIV,NOLONG -\c MOV reg64,reg_creg X64,PRIV -\c MOV reg_creg,reg32 386,PRIV,NOLONG -\c MOV reg_creg,reg64 X64,PRIV -\c MOV reg32,reg_dreg 386,PRIV,NOLONG -\c MOV reg64,reg_dreg X64,PRIV -\c MOV reg_dreg,reg32 386,PRIV,NOLONG -\c MOV reg_dreg,reg64 X64,PRIV -\c MOV reg32,reg_treg 386,NOLONG,ND -\c MOV reg_treg,reg32 386,NOLONG,ND -\c MOV mem,reg8 8086 -\c MOV reg8,reg8 8086 -\c MOV mem,reg16 8086 -\c MOV reg16,reg16 8086 -\c MOV mem,reg32 386 -\c MOV reg32,reg32 386 -\c MOV mem,reg64 X64 -\c MOV reg64,reg64 X64 -\c MOV reg8,mem 8086 -\c MOV reg8,reg8 8086 -\c MOV reg16,mem 8086 -\c MOV reg16,reg16 8086 -\c MOV reg32,mem 386 -\c MOV reg32,reg32 386 -\c MOV reg64,mem X64 -\c MOV reg64,reg64 X64 -\c MOV reg8,imm 8086 -\c MOV reg16,imm 8086 -\c MOV reg32,imm 386 -\c MOV reg64,imm X64 -\c MOV reg64,imm32 X64 -\c MOV rm8,imm 8086 -\c MOV rm16,imm 8086 -\c MOV rm32,imm 386 -\c MOV rm64,imm X64 -\c MOV mem,imm8 8086 -\c MOV mem,imm16 8086 -\c MOV mem,imm32 386 -\c MOVD mmxreg,mem PENT,MMX,SD -\c MOVD mmxreg,reg32 PENT,MMX -\c MOVD mem,mmxreg PENT,MMX,SD -\c MOVD reg32,mmxreg PENT,MMX -\c MOVD xmmreg,mem X64,SD -\c MOVD xmmreg,reg32 X64 -\c MOVD mem,xmmreg X64,SD -\c MOVD reg32,xmmreg X64,SSE -\c MOVQ mmxreg,mmxrm PENT,MMX -\c MOVQ mmxrm,mmxreg PENT,MMX -\c MOVQ mmxreg,rm64 X64,MMX -\c MOVQ rm64,mmxreg X64,MMX -\c MOVSB 8086 -\c MOVSD 386 -\c MOVSQ X64 -\c MOVSW 8086 -\c MOVSX reg16,mem 386 -\c MOVSX reg16,reg8 386 -\c MOVSX reg32,rm8 386 -\c MOVSX reg32,rm16 386 -\c MOVSX reg64,rm8 X64 -\c MOVSX reg64,rm16 X64 -\c MOVSXD reg64,rm32 X64 -\c MOVSX reg64,rm32 X64,ND -\c MOVZX reg16,mem 386 -\c MOVZX reg16,reg8 386 -\c MOVZX reg32,rm8 386 -\c MOVZX reg32,rm16 386 -\c MOVZX reg64,rm8 X64 -\c MOVZX reg64,rm16 X64 -\c MUL rm8 8086 -\c MUL rm16 8086 -\c MUL rm32 386 -\c MUL rm64 X64 -\c MWAIT PRESCOTT -\c MWAIT reg_eax,reg_ecx PRESCOTT,ND -\c NEG rm8 8086 -\c NEG rm16 8086 -\c NEG rm32 386 -\c NEG rm64 X64 -\c NOP 8086 -\c NOP rm16 P6 -\c NOP rm32 P6 -\c NOP rm64 X64 -\c NOT rm8 8086 -\c NOT rm16 8086 -\c NOT rm32 386 -\c NOT rm64 X64 -\c OR mem,reg8 8086 -\c OR reg8,reg8 8086 -\c OR mem,reg16 8086 -\c OR reg16,reg16 8086 -\c OR mem,reg32 386 -\c OR reg32,reg32 386 -\c OR mem,reg64 X64 -\c OR reg64,reg64 X64 -\c OR reg8,mem 8086 -\c OR reg8,reg8 8086 -\c OR reg16,mem 8086 -\c OR reg16,reg16 8086 -\c OR reg32,mem 386 -\c OR reg32,reg32 386 -\c OR reg64,mem X64 -\c OR reg64,reg64 X64 -\c OR rm16,imm8 8086 -\c OR rm32,imm8 386 -\c OR rm64,imm8 X64 -\c OR reg_al,imm 8086 -\c OR reg_ax,sbyte16 8086 -\c OR reg_ax,imm 8086 -\c OR reg_eax,sbyte32 386 -\c OR reg_eax,imm 386 -\c OR reg_rax,sbyte64 X64 -\c OR reg_rax,imm X64 -\c OR rm8,imm 8086 -\c OR rm16,imm 8086 -\c OR rm32,imm 386 -\c OR rm64,imm X64 -\c OR mem,imm8 8086 -\c OR mem,imm16 8086 -\c OR mem,imm32 386 -\c OUT imm,reg_al 8086 -\c OUT imm,reg_ax 8086 -\c OUT imm,reg_eax 386 -\c OUT reg_dx,reg_al 8086 -\c OUT reg_dx,reg_ax 8086 -\c OUT reg_dx,reg_eax 386 -\c OUTSB 186 -\c OUTSD 386 -\c OUTSW 186 -\c PACKSSDW mmxreg,mmxrm PENT,MMX -\c PACKSSWB mmxreg,mmxrm PENT,MMX -\c PACKUSWB mmxreg,mmxrm PENT,MMX -\c PADDB mmxreg,mmxrm PENT,MMX -\c PADDD mmxreg,mmxrm PENT,MMX -\c PADDSB mmxreg,mmxrm PENT,MMX -\c PADDSIW mmxreg,mmxrm PENT,MMX,CYRIX -\c PADDSW mmxreg,mmxrm PENT,MMX -\c PADDUSB mmxreg,mmxrm PENT,MMX -\c PADDUSW mmxreg,mmxrm PENT,MMX -\c PADDW mmxreg,mmxrm PENT,MMX -\c PAND mmxreg,mmxrm PENT,MMX -\c PANDN mmxreg,mmxrm PENT,MMX -\c PAUSE 8086 -\c PAVEB mmxreg,mmxrm PENT,MMX,CYRIX -\c PAVGUSB mmxreg,mmxrm PENT,3DNOW -\c PCMPEQB mmxreg,mmxrm PENT,MMX -\c PCMPEQD mmxreg,mmxrm PENT,MMX -\c PCMPEQW mmxreg,mmxrm PENT,MMX -\c PCMPGTB mmxreg,mmxrm PENT,MMX -\c PCMPGTD mmxreg,mmxrm PENT,MMX -\c PCMPGTW mmxreg,mmxrm PENT,MMX -\c PDISTIB mmxreg,mem PENT,MMX,CYRIX -\c PF2ID mmxreg,mmxrm PENT,3DNOW -\c PFACC mmxreg,mmxrm PENT,3DNOW -\c PFADD mmxreg,mmxrm PENT,3DNOW -\c PFCMPEQ mmxreg,mmxrm PENT,3DNOW -\c PFCMPGE mmxreg,mmxrm PENT,3DNOW -\c PFCMPGT mmxreg,mmxrm PENT,3DNOW -\c PFMAX mmxreg,mmxrm PENT,3DNOW -\c PFMIN mmxreg,mmxrm PENT,3DNOW -\c PFMUL mmxreg,mmxrm PENT,3DNOW -\c PFRCP mmxreg,mmxrm PENT,3DNOW -\c PFRCPIT1 mmxreg,mmxrm PENT,3DNOW -\c PFRCPIT2 mmxreg,mmxrm PENT,3DNOW -\c PFRSQIT1 mmxreg,mmxrm PENT,3DNOW -\c PFRSQRT mmxreg,mmxrm PENT,3DNOW -\c PFSUB mmxreg,mmxrm PENT,3DNOW -\c PFSUBR mmxreg,mmxrm PENT,3DNOW -\c PI2FD mmxreg,mmxrm PENT,3DNOW -\c PMACHRIW mmxreg,mem PENT,MMX,CYRIX -\c PMADDWD mmxreg,mmxrm PENT,MMX -\c PMAGW mmxreg,mmxrm PENT,MMX,CYRIX -\c PMULHRIW mmxreg,mmxrm PENT,MMX,CYRIX -\c PMULHRWA mmxreg,mmxrm PENT,3DNOW -\c PMULHRWC mmxreg,mmxrm PENT,MMX,CYRIX -\c PMULHW mmxreg,mmxrm PENT,MMX -\c PMULLW mmxreg,mmxrm PENT,MMX -\c PMVGEZB mmxreg,mem PENT,MMX,CYRIX -\c PMVLZB mmxreg,mem PENT,MMX,CYRIX -\c PMVNZB mmxreg,mem PENT,MMX,CYRIX -\c PMVZB mmxreg,mem PENT,MMX,CYRIX -\c POP reg16 8086 -\c POP reg32 386,NOLONG -\c POP reg64 X64 -\c POP rm16 8086 -\c POP rm32 386,NOLONG -\c POP rm64 X64 -\c POP reg_cs 8086,UNDOC,ND -\c POP reg_dess 8086,NOLONG -\c POP reg_fsgs 386 -\c POPA 186,NOLONG -\c POPAD 386,NOLONG -\c POPAW 186,NOLONG -\c POPF 8086 -\c POPFD 386,NOLONG -\c POPFQ X64 -\c POPFW 8086 -\c POR mmxreg,mmxrm PENT,MMX -\c PREFETCH mem PENT,3DNOW -\c PREFETCHW mem PENT,3DNOW -\c PSLLD mmxreg,mmxrm PENT,MMX -\c PSLLD mmxreg,imm PENT,MMX -\c PSLLQ mmxreg,mmxrm PENT,MMX -\c PSLLQ mmxreg,imm PENT,MMX -\c PSLLW mmxreg,mmxrm PENT,MMX -\c PSLLW mmxreg,imm PENT,MMX -\c PSRAD mmxreg,mmxrm PENT,MMX -\c PSRAD mmxreg,imm PENT,MMX -\c PSRAW mmxreg,mmxrm PENT,MMX -\c PSRAW mmxreg,imm PENT,MMX -\c PSRLD mmxreg,mmxrm PENT,MMX -\c PSRLD mmxreg,imm PENT,MMX -\c PSRLQ mmxreg,mmxrm PENT,MMX -\c PSRLQ mmxreg,imm PENT,MMX -\c PSRLW mmxreg,mmxrm PENT,MMX -\c PSRLW mmxreg,imm PENT,MMX -\c PSUBB mmxreg,mmxrm PENT,MMX -\c PSUBD mmxreg,mmxrm PENT,MMX -\c PSUBSB mmxreg,mmxrm PENT,MMX -\c PSUBSIW mmxreg,mmxrm PENT,MMX,CYRIX -\c PSUBSW mmxreg,mmxrm PENT,MMX -\c PSUBUSB mmxreg,mmxrm PENT,MMX -\c PSUBUSW mmxreg,mmxrm PENT,MMX -\c PSUBW mmxreg,mmxrm PENT,MMX -\c PUNPCKHBW mmxreg,mmxrm PENT,MMX -\c PUNPCKHDQ mmxreg,mmxrm PENT,MMX -\c PUNPCKHWD mmxreg,mmxrm PENT,MMX -\c PUNPCKLBW mmxreg,mmxrm PENT,MMX -\c PUNPCKLDQ mmxreg,mmxrm PENT,MMX -\c PUNPCKLWD mmxreg,mmxrm PENT,MMX -\c PUSH reg16 8086 -\c PUSH reg32 386,NOLONG -\c PUSH reg64 X64 -\c PUSH rm16 8086 -\c PUSH rm32 386,NOLONG -\c PUSH rm64 X64 -\c PUSH reg_cs 8086,NOLONG -\c PUSH reg_dess 8086,NOLONG -\c PUSH reg_fsgs 386 -\c PUSH imm8 186 -\c PUSH imm16 186,AR0,SZ -\c PUSH imm32 386,NOLONG,AR0,SZ -\c PUSH imm32 386,NOLONG,SD -\c PUSH imm64 X64,AR0,SZ -\c PUSHA 186,NOLONG -\c PUSHAD 386,NOLONG -\c PUSHAW 186,NOLONG -\c PUSHF 8086 -\c PUSHFD 386,NOLONG -\c PUSHFQ X64 -\c PUSHFW 8086 -\c PXOR mmxreg,mmxrm PENT,MMX -\c RCL rm8,unity 8086 -\c RCL rm8,reg_cl 8086 -\c RCL rm8,imm 186 -\c RCL rm16,unity 8086 -\c RCL rm16,reg_cl 8086 -\c RCL rm16,imm 186 -\c RCL rm32,unity 386 -\c RCL rm32,reg_cl 386 -\c RCL rm32,imm 386 -\c RCL rm64,unity X64 -\c RCL rm64,reg_cl X64 -\c RCL rm64,imm X64 -\c RCR rm8,unity 8086 -\c RCR rm8,reg_cl 8086 -\c RCR rm8,imm 186 -\c RCR rm16,unity 8086 -\c RCR rm16,reg_cl 8086 -\c RCR rm16,imm 186 -\c RCR rm32,unity 386 -\c RCR rm32,reg_cl 386 -\c RCR rm32,imm 386 -\c RCR rm64,unity X64 -\c RCR rm64,reg_cl X64 -\c RCR rm64,imm X64 -\c RDSHR rm32 P6,CYRIXM -\c RDMSR PENT,PRIV -\c RDPMC P6 -\c RDTSC PENT -\c RDTSCP X86_64 -\c RET 8086 -\c RET imm 8086,SW -\c RETF 8086 -\c RETF imm 8086,SW -\c RETN 8086 -\c RETN imm 8086,SW -\c ROL rm8,unity 8086 -\c ROL rm8,reg_cl 8086 -\c ROL rm8,imm 186 -\c ROL rm16,unity 8086 -\c ROL rm16,reg_cl 8086 -\c ROL rm16,imm 186 -\c ROL rm32,unity 386 -\c ROL rm32,reg_cl 386 -\c ROL rm32,imm 386 -\c ROL rm64,unity X64 -\c ROL rm64,reg_cl X64 -\c ROL rm64,imm X64 -\c ROR rm8,unity 8086 -\c ROR rm8,reg_cl 8086 -\c ROR rm8,imm 186 -\c ROR rm16,unity 8086 -\c ROR rm16,reg_cl 8086 -\c ROR rm16,imm 186 -\c ROR rm32,unity 386 -\c ROR rm32,reg_cl 386 -\c ROR rm32,imm 386 -\c ROR rm64,unity X64 -\c ROR rm64,reg_cl X64 -\c ROR rm64,imm X64 -\c RDM P6,CYRIX,ND -\c RSDC reg_sreg,mem80 486,CYRIXM -\c RSLDT mem80 486,CYRIXM -\c RSM PENTM -\c RSTS mem80 486,CYRIXM -\c SAHF 8086 -\c SAL rm8,unity 8086,ND -\c SAL rm8,reg_cl 8086,ND -\c SAL rm8,imm 186,ND -\c SAL rm16,unity 8086,ND -\c SAL rm16,reg_cl 8086,ND -\c SAL rm16,imm 186,ND -\c SAL rm32,unity 386,ND -\c SAL rm32,reg_cl 386,ND -\c SAL rm32,imm 386,ND -\c SAL rm64,unity X64,ND -\c SAL rm64,reg_cl X64,ND -\c SAL rm64,imm X64,ND -\c SALC 8086,UNDOC -\c SAR rm8,unity 8086 -\c SAR rm8,reg_cl 8086 -\c SAR rm8,imm 186 -\c SAR rm16,unity 8086 -\c SAR rm16,reg_cl 8086 -\c SAR rm16,imm 186 -\c SAR rm32,unity 386 -\c SAR rm32,reg_cl 386 -\c SAR rm32,imm 386 -\c SAR rm64,unity X64 -\c SAR rm64,reg_cl X64 -\c SAR rm64,imm X64 -\c SBB mem,reg8 8086 -\c SBB reg8,reg8 8086 -\c SBB mem,reg16 8086 -\c SBB reg16,reg16 8086 -\c SBB mem,reg32 386 -\c SBB reg32,reg32 386 -\c SBB mem,reg64 X64 -\c SBB reg64,reg64 X64 -\c SBB reg8,mem 8086 -\c SBB reg8,reg8 8086 -\c SBB reg16,mem 8086 -\c SBB reg16,reg16 8086 -\c SBB reg32,mem 386 -\c SBB reg32,reg32 386 -\c SBB reg64,mem X64 -\c SBB reg64,reg64 X64 -\c SBB rm16,imm8 8086 -\c SBB rm32,imm8 386 -\c SBB rm64,imm8 X64 -\c SBB reg_al,imm 8086 -\c SBB reg_ax,sbyte16 8086 -\c SBB reg_ax,imm 8086 -\c SBB reg_eax,sbyte32 386 -\c SBB reg_eax,imm 386 -\c SBB reg_rax,sbyte64 X64 -\c SBB reg_rax,imm X64 -\c SBB rm8,imm 8086 -\c SBB rm16,imm 8086 -\c SBB rm32,imm 386 -\c SBB rm64,imm X64 -\c SBB mem,imm8 8086 -\c SBB mem,imm16 8086 -\c SBB mem,imm32 386 -\c SCASB 8086 -\c SCASD 386 -\c SCASQ X64 -\c SCASW 8086 -\c SFENCE X64,AMD -\c SGDT mem 286 -\c SHL rm8,unity 8086 -\c SHL rm8,reg_cl 8086 -\c SHL rm8,imm 186 -\c SHL rm16,unity 8086 -\c SHL rm16,reg_cl 8086 -\c SHL rm16,imm 186 -\c SHL rm32,unity 386 -\c SHL rm32,reg_cl 386 -\c SHL rm32,imm 386 -\c SHL rm64,unity X64 -\c SHL rm64,reg_cl X64 -\c SHL rm64,imm X64 -\c SHLD mem,reg16,imm 3862 -\c SHLD reg16,reg16,imm 3862 -\c SHLD mem,reg32,imm 3862 -\c SHLD reg32,reg32,imm 3862 -\c SHLD mem,reg64,imm X642 -\c SHLD reg64,reg64,imm X642 -\c SHLD mem,reg16,reg_cl 386 -\c SHLD reg16,reg16,reg_cl 386 -\c SHLD mem,reg32,reg_cl 386 -\c SHLD reg32,reg32,reg_cl 386 -\c SHLD mem,reg64,reg_cl X64 -\c SHLD reg64,reg64,reg_cl X64 -\c SHR rm8,unity 8086 -\c SHR rm8,reg_cl 8086 -\c SHR rm8,imm 186 -\c SHR rm16,unity 8086 -\c SHR rm16,reg_cl 8086 -\c SHR rm16,imm 186 -\c SHR rm32,unity 386 -\c SHR rm32,reg_cl 386 -\c SHR rm32,imm 386 -\c SHR rm64,unity X64 -\c SHR rm64,reg_cl X64 -\c SHR rm64,imm X64 -\c SHRD mem,reg16,imm 3862 -\c SHRD reg16,reg16,imm 3862 -\c SHRD mem,reg32,imm 3862 -\c SHRD reg32,reg32,imm 3862 -\c SHRD mem,reg64,imm X642 -\c SHRD reg64,reg64,imm X642 -\c SHRD mem,reg16,reg_cl 386 -\c SHRD reg16,reg16,reg_cl 386 -\c SHRD mem,reg32,reg_cl 386 -\c SHRD reg32,reg32,reg_cl 386 -\c SHRD mem,reg64,reg_cl X64 -\c SHRD reg64,reg64,reg_cl X64 -\c SIDT mem 286 -\c SLDT mem 286 -\c SLDT mem16 286 -\c SLDT reg16 286 -\c SLDT reg32 386 -\c SLDT reg64 X64,ND -\c SLDT reg64 X64 -\c SKINIT X64 -\c SMI 386,UNDOC -\c SMINT P6,CYRIX,ND -\c SMINTOLD 486,CYRIX,ND -\c SMSW mem 286 -\c SMSW mem16 286 -\c SMSW reg16 286 -\c SMSW reg32 386 -\c STC 8086 -\c STD 8086 -\c STGI X64 -\c STI 8086 -\c STOSB 8086 -\c STOSD 386 -\c STOSQ X64 -\c STOSW 8086 -\c STR mem 286,PROT -\c STR mem16 286,PROT -\c STR reg16 286,PROT -\c STR reg32 386,PROT -\c STR reg64 X64 -\c SUB mem,reg8 8086 -\c SUB reg8,reg8 8086 -\c SUB mem,reg16 8086 -\c SUB reg16,reg16 8086 -\c SUB mem,reg32 386 -\c SUB reg32,reg32 386 -\c SUB mem,reg64 X64 -\c SUB reg64,reg64 X64 -\c SUB reg8,mem 8086 -\c SUB reg8,reg8 8086 -\c SUB reg16,mem 8086 -\c SUB reg16,reg16 8086 -\c SUB reg32,mem 386 -\c SUB reg32,reg32 386 -\c SUB reg64,mem X64 -\c SUB reg64,reg64 X64 -\c SUB rm16,imm8 8086 -\c SUB rm32,imm8 386 -\c SUB rm64,imm8 X64 -\c SUB reg_al,imm 8086 -\c SUB reg_ax,sbyte16 8086 -\c SUB reg_ax,imm 8086 -\c SUB reg_eax,sbyte32 386 -\c SUB reg_eax,imm 386 -\c SUB reg_rax,sbyte64 X64 -\c SUB reg_rax,imm X64 -\c SUB rm8,imm 8086 -\c SUB rm16,imm 8086 -\c SUB rm32,imm 386 -\c SUB rm64,imm X64 -\c SUB mem,imm8 8086 -\c SUB mem,imm16 8086 -\c SUB mem,imm32 386 -\c SVDC mem80,reg_sreg 486,CYRIXM -\c SVLDT mem80 486,CYRIXM,ND -\c SVTS mem80 486,CYRIXM -\c SWAPGS X64 -\c SYSCALL P6,AMD -\c SYSENTER P6 -\c SYSEXIT P6,PRIV -\c SYSRET P6,PRIV,AMD -\c TEST mem,reg8 8086 -\c TEST reg8,reg8 8086 -\c TEST mem,reg16 8086 -\c TEST reg16,reg16 8086 -\c TEST mem,reg32 386 -\c TEST reg32,reg32 386 -\c TEST mem,reg64 X64 -\c TEST reg64,reg64 X64 -\c TEST reg8,mem 8086 -\c TEST reg16,mem 8086 -\c TEST reg32,mem 386 -\c TEST reg64,mem X64 -\c TEST reg_al,imm 8086 -\c TEST reg_ax,imm 8086 -\c TEST reg_eax,imm 386 -\c TEST reg_rax,imm X64 -\c TEST rm8,imm 8086 -\c TEST rm16,imm 8086 -\c TEST rm32,imm 386 -\c TEST rm64,imm X64 -\c TEST mem,imm8 8086 -\c TEST mem,imm16 8086 -\c TEST mem,imm32 386 -\c UD0 186,UNDOC -\c UD1 186,UNDOC -\c UD2B 186,UNDOC,ND -\c UD2 186 -\c UD2A 186,ND -\c UMOV mem,reg8 386,UNDOC,ND -\c UMOV reg8,reg8 386,UNDOC,ND -\c UMOV mem,reg16 386,UNDOC,ND -\c UMOV reg16,reg16 386,UNDOC,ND -\c UMOV mem,reg32 386,UNDOC,ND -\c UMOV reg32,reg32 386,UNDOC,ND -\c UMOV reg8,mem 386,UNDOC,ND -\c UMOV reg8,reg8 386,UNDOC,ND -\c UMOV reg16,mem 386,UNDOC,ND -\c UMOV reg16,reg16 386,UNDOC,ND -\c UMOV reg32,mem 386,UNDOC,ND -\c UMOV reg32,reg32 386,UNDOC,ND -\c VERR mem 286,PROT -\c VERR mem16 286,PROT -\c VERR reg16 286,PROT -\c VERW mem 286,PROT -\c VERW mem16 286,PROT -\c VERW reg16 286,PROT -\c FWAIT 8086 -\c WBINVD 486,PRIV -\c WRSHR rm32 P6,CYRIXM -\c WRMSR PENT,PRIV -\c XADD mem,reg8 486 -\c XADD reg8,reg8 486 -\c XADD mem,reg16 486 -\c XADD reg16,reg16 486 -\c XADD mem,reg32 486 -\c XADD reg32,reg32 486 -\c XADD mem,reg64 X64 -\c XADD reg64,reg64 X64 -\c XBTS reg16,mem 386,SW,UNDOC,ND -\c XBTS reg16,reg16 386,UNDOC,ND -\c XBTS reg32,mem 386,SD,UNDOC,ND -\c XBTS reg32,reg32 386,UNDOC,ND -\c XCHG reg_ax,reg16 8086 -\c XCHG reg_eax,reg32na 386 -\c XCHG reg_rax,reg64 X64 -\c XCHG reg16,reg_ax 8086 -\c XCHG reg32na,reg_eax 386 -\c XCHG reg64,reg_rax X64 -\c XCHG reg_eax,reg_eax 386,NOLONG -\c XCHG reg8,mem 8086 -\c XCHG reg8,reg8 8086 -\c XCHG reg16,mem 8086 -\c XCHG reg16,reg16 8086 -\c XCHG reg32,mem 386 -\c XCHG reg32,reg32 386 -\c XCHG reg64,mem X64 -\c XCHG reg64,reg64 X64 -\c XCHG mem,reg8 8086 -\c XCHG reg8,reg8 8086 -\c XCHG mem,reg16 8086 -\c XCHG reg16,reg16 8086 -\c XCHG mem,reg32 386 -\c XCHG reg32,reg32 386 -\c XCHG mem,reg64 X64 -\c XCHG reg64,reg64 X64 -\c XLATB 8086 -\c XLAT 8086 -\c XOR mem,reg8 8086 -\c XOR reg8,reg8 8086 -\c XOR mem,reg16 8086 -\c XOR reg16,reg16 8086 -\c XOR mem,reg32 386 -\c XOR reg32,reg32 386 -\c XOR mem,reg64 X64 -\c XOR reg64,reg64 X64 -\c XOR reg8,mem 8086 -\c XOR reg8,reg8 8086 -\c XOR reg16,mem 8086 -\c XOR reg16,reg16 8086 -\c XOR reg32,mem 386 -\c XOR reg32,reg32 386 -\c XOR reg64,mem X64 -\c XOR reg64,reg64 X64 -\c XOR rm16,imm8 8086 -\c XOR rm32,imm8 386 -\c XOR rm64,imm8 X64 -\c XOR reg_al,imm 8086 -\c XOR reg_ax,sbyte16 8086 -\c XOR reg_ax,imm 8086 -\c XOR reg_eax,sbyte32 386 -\c XOR reg_eax,imm 386 -\c XOR reg_rax,sbyte64 X64 -\c XOR reg_rax,imm X64 -\c XOR rm8,imm 8086 -\c XOR rm16,imm 8086 -\c XOR rm32,imm 386 -\c XOR rm64,imm X64 -\c XOR mem,imm8 8086 -\c XOR mem,imm16 8086 -\c XOR mem,imm32 386 -\c CMOVcc reg16,mem P6 -\c CMOVcc reg16,reg16 P6 -\c CMOVcc reg32,mem P6 -\c CMOVcc reg32,reg32 P6 -\c CMOVcc reg64,mem X64 -\c CMOVcc reg64,reg64 X64 -\c Jcc imm|near 386 -\c Jcc imm16|near 386 -\c Jcc imm32|near 386 -\c Jcc imm|short 8086,ND -\c Jcc imm 8086,ND -\c Jcc imm 386,ND -\c Jcc imm 8086,ND -\c Jcc imm 8086 -\c SETcc mem 386 -\c SETcc reg8 386 - -\S{} Katmai Streaming SIMD instructions (SSE -- a.k.a. KNI, XMM, MMX2) - -\c ADDPS xmmreg,xmmrm KATMAI,SSE -\c ADDSS xmmreg,xmmrm KATMAI,SSE,SD -\c ANDNPS xmmreg,xmmrm KATMAI,SSE -\c ANDPS xmmreg,xmmrm KATMAI,SSE -\c CMPEQPS xmmreg,xmmrm KATMAI,SSE -\c CMPEQSS xmmreg,xmmrm KATMAI,SSE -\c CMPLEPS xmmreg,xmmrm KATMAI,SSE -\c CMPLESS xmmreg,xmmrm KATMAI,SSE -\c CMPLTPS xmmreg,xmmrm KATMAI,SSE -\c CMPLTSS xmmreg,xmmrm KATMAI,SSE -\c CMPNEQPS xmmreg,xmmrm KATMAI,SSE -\c CMPNEQSS xmmreg,xmmrm KATMAI,SSE -\c CMPNLEPS xmmreg,xmmrm KATMAI,SSE -\c CMPNLESS xmmreg,xmmrm KATMAI,SSE -\c CMPNLTPS xmmreg,xmmrm KATMAI,SSE -\c CMPNLTSS xmmreg,xmmrm KATMAI,SSE -\c CMPORDPS xmmreg,xmmrm KATMAI,SSE -\c CMPORDSS xmmreg,xmmrm KATMAI,SSE -\c CMPUNORDPS xmmreg,xmmrm KATMAI,SSE -\c CMPUNORDSS xmmreg,xmmrm KATMAI,SSE -\c CMPPS xmmreg,mem,imm KATMAI,SSE -\c CMPPS xmmreg,xmmreg,imm KATMAI,SSE -\c CMPSS xmmreg,mem,imm KATMAI,SSE -\c CMPSS xmmreg,xmmreg,imm KATMAI,SSE -\c COMISS xmmreg,xmmrm KATMAI,SSE -\c CVTPI2PS xmmreg,mmxrm KATMAI,SSE,MMX -\c CVTPS2PI mmxreg,xmmrm KATMAI,SSE,MMX -\c CVTSI2SS xmmreg,mem KATMAI,SSE,SD,AR1,ND -\c CVTSI2SS xmmreg,rm32 KATMAI,SSE,SD,AR1 -\c CVTSI2SS xmmreg,rm64 X64,SSE,AR1 -\c CVTSS2SI reg32,xmmreg KATMAI,SSE,SD,AR1 -\c CVTSS2SI reg32,mem KATMAI,SSE,SD,AR1 -\c CVTSS2SI reg64,xmmreg X64,SSE,SD,AR1 -\c CVTSS2SI reg64,mem X64,SSE,SD,AR1 -\c CVTTPS2PI mmxreg,xmmrm KATMAI,SSE,MMX -\c CVTTSS2SI reg32,xmmrm KATMAI,SSE,SD,AR1 -\c CVTTSS2SI reg64,xmmrm X64,SSE,SD,AR1 -\c DIVPS xmmreg,xmmrm KATMAI,SSE -\c DIVSS xmmreg,xmmrm KATMAI,SSE -\c LDMXCSR mem KATMAI,SSE,SD -\c MAXPS xmmreg,xmmrm KATMAI,SSE -\c MAXSS xmmreg,xmmrm KATMAI,SSE -\c MINPS xmmreg,xmmrm KATMAI,SSE -\c MINSS xmmreg,xmmrm KATMAI,SSE -\c MOVAPS xmmreg,mem KATMAI,SSE -\c MOVAPS mem,xmmreg KATMAI,SSE -\c MOVAPS xmmreg,xmmreg KATMAI,SSE -\c MOVAPS xmmreg,xmmreg KATMAI,SSE -\c MOVHPS xmmreg,mem KATMAI,SSE -\c MOVHPS mem,xmmreg KATMAI,SSE -\c MOVLHPS xmmreg,xmmreg KATMAI,SSE -\c MOVLPS xmmreg,mem KATMAI,SSE -\c MOVLPS mem,xmmreg KATMAI,SSE -\c MOVHLPS xmmreg,xmmreg KATMAI,SSE -\c MOVMSKPS reg32,xmmreg KATMAI,SSE -\c MOVMSKPS reg64,xmmreg X64,SSE -\c MOVNTPS mem,xmmreg KATMAI,SSE -\c MOVSS xmmreg,mem KATMAI,SSE -\c MOVSS mem,xmmreg KATMAI,SSE -\c MOVSS xmmreg,xmmreg KATMAI,SSE -\c MOVSS xmmreg,xmmreg KATMAI,SSE -\c MOVUPS xmmreg,mem KATMAI,SSE -\c MOVUPS mem,xmmreg KATMAI,SSE -\c MOVUPS xmmreg,xmmreg KATMAI,SSE -\c MOVUPS xmmreg,xmmreg KATMAI,SSE -\c MULPS xmmreg,xmmrm KATMAI,SSE -\c MULSS xmmreg,xmmrm KATMAI,SSE -\c ORPS xmmreg,xmmrm KATMAI,SSE -\c RCPPS xmmreg,xmmrm KATMAI,SSE -\c RCPSS xmmreg,xmmrm KATMAI,SSE -\c RSQRTPS xmmreg,xmmrm KATMAI,SSE -\c RSQRTSS xmmreg,xmmrm KATMAI,SSE -\c SHUFPS xmmreg,mem,imm KATMAI,SSE -\c SHUFPS xmmreg,xmmreg,imm KATMAI,SSE -\c SQRTPS xmmreg,xmmrm KATMAI,SSE -\c SQRTSS xmmreg,xmmrm KATMAI,SSE -\c STMXCSR mem KATMAI,SSE,SD -\c SUBPS xmmreg,xmmrm KATMAI,SSE -\c SUBSS xmmreg,xmmrm KATMAI,SSE -\c UCOMISS xmmreg,xmmrm KATMAI,SSE -\c UNPCKHPS xmmreg,xmmrm KATMAI,SSE -\c UNPCKLPS xmmreg,xmmrm KATMAI,SSE -\c XORPS xmmreg,xmmrm KATMAI,SSE - -\S{} Introduced in Deschutes but necessary for SSE support - -\c FXRSTOR mem P6,SSE,FPU -\c FXSAVE mem P6,SSE,FPU - -\S{} XSAVE group (AVX and extended state) - -\c XGETBV NEHALEM -\c XSETBV NEHALEM,PRIV -\c XSAVE mem NEHALEM -\c XRSTOR mem NEHALEM - -\S{} Generic memory operations - -\c PREFETCHNTA mem KATMAI -\c PREFETCHT0 mem KATMAI -\c PREFETCHT1 mem KATMAI -\c PREFETCHT2 mem KATMAI -\c SFENCE KATMAI - -\S{} New MMX instructions introduced in Katmai - -\c MASKMOVQ mmxreg,mmxreg KATMAI,MMX -\c MOVNTQ mem,mmxreg KATMAI,MMX -\c PAVGB mmxreg,mmxrm KATMAI,MMX -\c PAVGW mmxreg,mmxrm KATMAI,MMX -\c PEXTRW reg32,mmxreg,imm KATMAI,MMX -\c PINSRW mmxreg,mem,imm KATMAI,MMX -\c PINSRW mmxreg,rm16,imm KATMAI,MMX -\c PINSRW mmxreg,reg32,imm KATMAI,MMX -\c PMAXSW mmxreg,mmxrm KATMAI,MMX -\c PMAXUB mmxreg,mmxrm KATMAI,MMX -\c PMINSW mmxreg,mmxrm KATMAI,MMX -\c PMINUB mmxreg,mmxrm KATMAI,MMX -\c PMOVMSKB reg32,mmxreg KATMAI,MMX -\c PMULHUW mmxreg,mmxrm KATMAI,MMX -\c PSADBW mmxreg,mmxrm KATMAI,MMX -\c PSHUFW mmxreg,mmxrm,imm KATMAI,MMX2 - -\S{} AMD Enhanced 3DNow! (Athlon) instructions - -\c PF2IW mmxreg,mmxrm PENT,3DNOW -\c PFNACC mmxreg,mmxrm PENT,3DNOW -\c PFPNACC mmxreg,mmxrm PENT,3DNOW -\c PI2FW mmxreg,mmxrm PENT,3DNOW -\c PSWAPD mmxreg,mmxrm PENT,3DNOW - -\S{} Willamette SSE2 Cacheability Instructions - -\c MASKMOVDQU xmmreg,xmmreg WILLAMETTE,SSE2 -\c CLFLUSH mem WILLAMETTE,SSE2 -\c MOVNTDQ mem,xmmreg WILLAMETTE,SSE2,SO -\c MOVNTI mem,reg32 WILLAMETTE,SD -\c MOVNTI mem,reg64 X64 -\c MOVNTPD mem,xmmreg WILLAMETTE,SSE2,SO -\c LFENCE WILLAMETTE,SSE2 -\c MFENCE WILLAMETTE,SSE2 - -\S{} Willamette MMX instructions (SSE2 SIMD Integer Instructions) - -\c MOVD mem,xmmreg WILLAMETTE,SSE2,SD -\c MOVD xmmreg,mem WILLAMETTE,SSE2,SD -\c MOVD xmmreg,rm32 WILLAMETTE,SSE2 -\c MOVD rm32,xmmreg WILLAMETTE,SSE2 -\c MOVDQA xmmreg,xmmreg WILLAMETTE,SSE2 -\c MOVDQA mem,xmmreg WILLAMETTE,SSE2,SO -\c MOVDQA xmmreg,mem WILLAMETTE,SSE2,SO -\c MOVDQA xmmreg,xmmreg WILLAMETTE,SSE2 -\c MOVDQU xmmreg,xmmreg WILLAMETTE,SSE2 -\c MOVDQU mem,xmmreg WILLAMETTE,SSE2,SO -\c MOVDQU xmmreg,mem WILLAMETTE,SSE2,SO -\c MOVDQU xmmreg,xmmreg WILLAMETTE,SSE2 -\c MOVDQ2Q mmxreg,xmmreg WILLAMETTE,SSE2 -\c MOVQ xmmreg,xmmreg WILLAMETTE,SSE2 -\c MOVQ xmmreg,xmmreg WILLAMETTE,SSE2 -\c MOVQ mem,xmmreg WILLAMETTE,SSE2 -\c MOVQ xmmreg,mem WILLAMETTE,SSE2 -\c MOVQ xmmreg,rm64 X64,SSE2 -\c MOVQ rm64,xmmreg X64,SSE2 -\c MOVQ2DQ xmmreg,mmxreg WILLAMETTE,SSE2 -\c PACKSSWB xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PACKSSDW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PACKUSWB xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PADDB xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PADDW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PADDD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PADDQ mmxreg,mmxrm WILLAMETTE,MMX -\c PADDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PADDSB xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PADDSW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PADDUSB xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PADDUSW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PAND xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PANDN xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PAVGB xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PAVGW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PCMPEQB xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PCMPEQW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PCMPEQD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PCMPGTB xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PCMPGTW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PCMPGTD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PEXTRW reg32,xmmreg,imm WILLAMETTE,SSE2 -\c PINSRW xmmreg,reg16,imm WILLAMETTE,SSE2 -\c PINSRW xmmreg,reg32,imm WILLAMETTE,SSE2,ND -\c PINSRW xmmreg,mem,imm WILLAMETTE,SSE2 -\c PINSRW xmmreg,mem16,imm WILLAMETTE,SSE2 -\c PMADDWD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PMAXSW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PMAXUB xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PMINSW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PMINUB xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PMOVMSKB reg32,xmmreg WILLAMETTE,SSE2 -\c PMULHUW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PMULHW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PMULLW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PMULUDQ mmxreg,mmxrm WILLAMETTE,SSE2,SO -\c PMULUDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c POR xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PSADBW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PSHUFD xmmreg,xmmreg,imm WILLAMETTE,SSE2 -\c PSHUFD xmmreg,mem,imm WILLAMETTE,SSE22 -\c PSHUFHW xmmreg,xmmreg,imm WILLAMETTE,SSE2 -\c PSHUFHW xmmreg,mem,imm WILLAMETTE,SSE22 -\c PSHUFLW xmmreg,xmmreg,imm WILLAMETTE,SSE2 -\c PSHUFLW xmmreg,mem,imm WILLAMETTE,SSE22 -\c PSLLDQ xmmreg,imm WILLAMETTE,SSE2,AR1 -\c PSLLW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PSLLW xmmreg,imm WILLAMETTE,SSE2,AR1 -\c PSLLD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PSLLD xmmreg,imm WILLAMETTE,SSE2,AR1 -\c PSLLQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PSLLQ xmmreg,imm WILLAMETTE,SSE2,AR1 -\c PSRAW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PSRAW xmmreg,imm WILLAMETTE,SSE2,AR1 -\c PSRAD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PSRAD xmmreg,imm WILLAMETTE,SSE2,AR1 -\c PSRLDQ xmmreg,imm WILLAMETTE,SSE2,AR1 -\c PSRLW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PSRLW xmmreg,imm WILLAMETTE,SSE2,AR1 -\c PSRLD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PSRLD xmmreg,imm WILLAMETTE,SSE2,AR1 -\c PSRLQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PSRLQ xmmreg,imm WILLAMETTE,SSE2,AR1 -\c PSUBB xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PSUBW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PSUBD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PSUBQ mmxreg,mmxrm WILLAMETTE,SSE2,SO -\c PSUBQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PSUBSB xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PSUBSW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PSUBUSB xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PSUBUSW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PUNPCKHBW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PUNPCKHWD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PUNPCKHDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PUNPCKHQDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PUNPCKLBW xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PUNPCKLWD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PUNPCKLDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PUNPCKLQDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c PXOR xmmreg,xmmrm WILLAMETTE,SSE2,SO - -\S{} Willamette Streaming SIMD instructions (SSE2) - -\c ADDPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c ADDSD xmmreg,xmmrm WILLAMETTE,SSE2 -\c ANDNPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c ANDPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c CMPEQPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c CMPEQSD xmmreg,xmmrm WILLAMETTE,SSE2 -\c CMPLEPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c CMPLESD xmmreg,xmmrm WILLAMETTE,SSE2 -\c CMPLTPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c CMPLTSD xmmreg,xmmrm WILLAMETTE,SSE2 -\c CMPNEQPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c CMPNEQSD xmmreg,xmmrm WILLAMETTE,SSE2 -\c CMPNLEPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c CMPNLESD xmmreg,xmmrm WILLAMETTE,SSE2 -\c CMPNLTPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c CMPNLTSD xmmreg,xmmrm WILLAMETTE,SSE2 -\c CMPORDPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c CMPORDSD xmmreg,xmmrm WILLAMETTE,SSE2 -\c CMPUNORDPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c CMPUNORDSD xmmreg,xmmrm WILLAMETTE,SSE2 -\c CMPPD xmmreg,xmmrm,imm WILLAMETTE,SSE22 -\c CMPSD xmmreg,xmmrm,imm WILLAMETTE,SSE2 -\c COMISD xmmreg,xmmrm WILLAMETTE,SSE2 -\c CVTDQ2PD xmmreg,xmmrm WILLAMETTE,SSE2 -\c CVTDQ2PS xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c CVTPD2DQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c CVTPD2PI mmxreg,xmmrm WILLAMETTE,SSE2,SO -\c CVTPD2PS xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c CVTPI2PD xmmreg,mmxrm WILLAMETTE,SSE2 -\c CVTPS2DQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c CVTPS2PD xmmreg,xmmrm WILLAMETTE,SSE2 -\c CVTSD2SI reg32,xmmreg WILLAMETTE,SSE2,AR1 -\c CVTSD2SI reg32,mem WILLAMETTE,SSE2,AR1 -\c CVTSD2SI reg64,xmmreg X64,SSE2,AR1 -\c CVTSD2SI reg64,mem X64,SSE2,AR1 -\c CVTSD2SS xmmreg,xmmrm WILLAMETTE,SSE2 -\c CVTSI2SD xmmreg,mem WILLAMETTE,SSE2,SD,AR1,ND -\c CVTSI2SD xmmreg,rm32 WILLAMETTE,SSE2,SD,AR1 -\c CVTSI2SD xmmreg,rm64 X64,SSE2,AR1 -\c CVTSS2SD xmmreg,xmmrm WILLAMETTE,SSE2,SD -\c CVTTPD2PI mmxreg,xmmrm WILLAMETTE,SSE2,SO -\c CVTTPD2DQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c CVTTPS2DQ xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c CVTTSD2SI reg32,xmmreg WILLAMETTE,SSE2,AR1 -\c CVTTSD2SI reg32,mem WILLAMETTE,SSE2,AR1 -\c CVTTSD2SI reg64,xmmreg X64,SSE2,AR1 -\c CVTTSD2SI reg64,mem X64,SSE2,AR1 -\c DIVPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c DIVSD xmmreg,xmmrm WILLAMETTE,SSE2 -\c MAXPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c MAXSD xmmreg,xmmrm WILLAMETTE,SSE2 -\c MINPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c MINSD xmmreg,xmmrm WILLAMETTE,SSE2 -\c MOVAPD xmmreg,xmmreg WILLAMETTE,SSE2 -\c MOVAPD xmmreg,xmmreg WILLAMETTE,SSE2 -\c MOVAPD mem,xmmreg WILLAMETTE,SSE2,SO -\c MOVAPD xmmreg,mem WILLAMETTE,SSE2,SO -\c MOVHPD mem,xmmreg WILLAMETTE,SSE2 -\c MOVHPD xmmreg,mem WILLAMETTE,SSE2 -\c MOVLPD mem,xmmreg WILLAMETTE,SSE2 -\c MOVLPD xmmreg,mem WILLAMETTE,SSE2 -\c MOVMSKPD reg32,xmmreg WILLAMETTE,SSE2 -\c MOVMSKPD reg64,xmmreg X64,SSE2 -\c MOVSD xmmreg,xmmreg WILLAMETTE,SSE2 -\c MOVSD xmmreg,xmmreg WILLAMETTE,SSE2 -\c MOVSD mem,xmmreg WILLAMETTE,SSE2 -\c MOVSD xmmreg,mem WILLAMETTE,SSE2 -\c MOVUPD xmmreg,xmmreg WILLAMETTE,SSE2 -\c MOVUPD xmmreg,xmmreg WILLAMETTE,SSE2 -\c MOVUPD mem,xmmreg WILLAMETTE,SSE2,SO -\c MOVUPD xmmreg,mem WILLAMETTE,SSE2,SO -\c MULPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c MULSD xmmreg,xmmrm WILLAMETTE,SSE2 -\c ORPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c SHUFPD xmmreg,xmmreg,imm WILLAMETTE,SSE2 -\c SHUFPD xmmreg,mem,imm WILLAMETTE,SSE2 -\c SQRTPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c SQRTSD xmmreg,xmmrm WILLAMETTE,SSE2 -\c SUBPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c SUBSD xmmreg,xmmrm WILLAMETTE,SSE2 -\c UCOMISD xmmreg,xmmrm WILLAMETTE,SSE2 -\c UNPCKHPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c UNPCKLPD xmmreg,xmmrm WILLAMETTE,SSE2,SO -\c XORPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - -\S{} Prescott New Instructions (SSE3) - -\c ADDSUBPD xmmreg,xmmrm PRESCOTT,SSE3,SO -\c ADDSUBPS xmmreg,xmmrm PRESCOTT,SSE3,SO -\c HADDPD xmmreg,xmmrm PRESCOTT,SSE3,SO -\c HADDPS xmmreg,xmmrm PRESCOTT,SSE3,SO -\c HSUBPD xmmreg,xmmrm PRESCOTT,SSE3,SO -\c HSUBPS xmmreg,xmmrm PRESCOTT,SSE3,SO -\c LDDQU xmmreg,mem PRESCOTT,SSE3,SO -\c MOVDDUP xmmreg,xmmrm PRESCOTT,SSE3 -\c MOVSHDUP xmmreg,xmmrm PRESCOTT,SSE3 -\c MOVSLDUP xmmreg,xmmrm PRESCOTT,SSE3 - -\S{} VMX Instructions - -\c VMCALL VMX -\c VMCLEAR mem VMX -\c VMLAUNCH VMX -\c VMLOAD X64,VMX -\c VMMCALL X64,VMX -\c VMPTRLD mem VMX -\c VMPTRST mem VMX -\c VMREAD rm32,reg32 VMX,NOLONG,SD -\c VMREAD rm64,reg64 X64,VMX -\c VMRESUME VMX -\c VMRUN X64,VMX -\c VMSAVE X64,VMX -\c VMWRITE reg32,rm32 VMX,NOLONG,SD -\c VMWRITE reg64,rm64 X64,VMX -\c VMXOFF VMX -\c VMXON mem VMX - -\S{} Extended Page Tables VMX instructions - -\c INVEPT reg32,mem VMX,SO,NOLONG -\c INVEPT reg64,mem VMX,SO,LONG -\c INVVPID reg32,mem VMX,SO,NOLONG -\c INVVPID reg64,mem VMX,SO,LONG - -\S{} Tejas New Instructions (SSSE3) - -\c PABSB mmxreg,mmxrm SSSE3,MMX -\c PABSB xmmreg,xmmrm SSSE3 -\c PABSW mmxreg,mmxrm SSSE3,MMX -\c PABSW xmmreg,xmmrm SSSE3 -\c PABSD mmxreg,mmxrm SSSE3,MMX -\c PABSD xmmreg,xmmrm SSSE3 -\c PALIGNR mmxreg,mmxrm,imm SSSE3,MMX -\c PALIGNR xmmreg,xmmrm,imm SSSE3 -\c PHADDW mmxreg,mmxrm SSSE3,MMX -\c PHADDW xmmreg,xmmrm SSSE3 -\c PHADDD mmxreg,mmxrm SSSE3,MMX -\c PHADDD xmmreg,xmmrm SSSE3 -\c PHADDSW mmxreg,mmxrm SSSE3,MMX -\c PHADDSW xmmreg,xmmrm SSSE3 -\c PHSUBW mmxreg,mmxrm SSSE3,MMX -\c PHSUBW xmmreg,xmmrm SSSE3 -\c PHSUBD mmxreg,mmxrm SSSE3,MMX -\c PHSUBD xmmreg,xmmrm SSSE3 -\c PHSUBSW mmxreg,mmxrm SSSE3,MMX -\c PHSUBSW xmmreg,xmmrm SSSE3 -\c PMADDUBSW mmxreg,mmxrm SSSE3,MMX -\c PMADDUBSW xmmreg,xmmrm SSSE3 -\c PMULHRSW mmxreg,mmxrm SSSE3,MMX -\c PMULHRSW xmmreg,xmmrm SSSE3 -\c PSHUFB mmxreg,mmxrm SSSE3,MMX -\c PSHUFB xmmreg,xmmrm SSSE3 -\c PSIGNB mmxreg,mmxrm SSSE3,MMX -\c PSIGNB xmmreg,xmmrm SSSE3 -\c PSIGNW mmxreg,mmxrm SSSE3,MMX -\c PSIGNW xmmreg,xmmrm SSSE3 -\c PSIGND mmxreg,mmxrm SSSE3,MMX -\c PSIGND xmmreg,xmmrm SSSE3 - -\S{} AMD SSE4A - -\c EXTRQ xmmreg,imm,imm SSE4A,AMD -\c EXTRQ xmmreg,xmmreg SSE4A,AMD -\c INSERTQ xmmreg,xmmreg,imm,imm SSE4A,AMD -\c INSERTQ xmmreg,xmmreg SSE4A,AMD -\c MOVNTSD mem,xmmreg SSE4A,AMD -\c MOVNTSS mem,xmmreg SSE4A,AMD,SD - -\S{} New instructions in Barcelona - -\c LZCNT reg16,rm16 P6,AMD -\c LZCNT reg32,rm32 P6,AMD -\c LZCNT reg64,rm64 X64,AMD - -\S{} Penryn New Instructions (SSE4.1) - -\c BLENDPD xmmreg,xmmrm,imm SSE41 -\c BLENDPS xmmreg,xmmrm,imm SSE41 -\c BLENDVPD xmmreg,xmmrm,xmm0 SSE41 -\c BLENDVPS xmmreg,xmmrm,xmm0 SSE41 -\c DPPD xmmreg,xmmrm,imm SSE41 -\c DPPS xmmreg,xmmrm,imm SSE41 -\c EXTRACTPS rm32,xmmreg,imm SSE41 -\c EXTRACTPS reg64,xmmreg,imm SSE41,X64 -\c INSERTPS xmmreg,xmmrm,imm SSE41,SD -\c MOVNTDQA xmmreg,mem SSE41 -\c MPSADBW xmmreg,xmmrm,imm SSE41 -\c PACKUSDW xmmreg,xmmrm SSE41 -\c PBLENDVB xmmreg,xmmrm,xmm0 SSE41 -\c PBLENDW xmmreg,xmmrm,imm SSE41 -\c PCMPEQQ xmmreg,xmmrm SSE41 -\c PEXTRB reg32,xmmreg,imm SSE41 -\c PEXTRB mem8,xmmreg,imm SSE41 -\c PEXTRB reg64,xmmreg,imm SSE41,X64 -\c PEXTRD rm32,xmmreg,imm SSE41 -\c PEXTRQ rm64,xmmreg,imm SSE41,X64 -\c PEXTRW reg32,xmmreg,imm SSE41 -\c PEXTRW mem16,xmmreg,imm SSE41 -\c PEXTRW reg64,xmmreg,imm SSE41,X64 -\c PHMINPOSUW xmmreg,xmmrm SSE41 -\c PINSRB xmmreg,mem,imm SSE41 -\c PINSRB xmmreg,rm8,imm SSE41 -\c PINSRB xmmreg,reg32,imm SSE41 -\c PINSRD xmmreg,mem,imm SSE41 -\c PINSRD xmmreg,rm32,imm SSE41 -\c PINSRQ xmmreg,mem,imm SSE41,X64 -\c PINSRQ xmmreg,rm64,imm SSE41,X64 -\c PMAXSB xmmreg,xmmrm SSE41 -\c PMAXSD xmmreg,xmmrm SSE41 -\c PMAXUD xmmreg,xmmrm SSE41 -\c PMAXUW xmmreg,xmmrm SSE41 -\c PMINSB xmmreg,xmmrm SSE41 -\c PMINSD xmmreg,xmmrm SSE41 -\c PMINUD xmmreg,xmmrm SSE41 -\c PMINUW xmmreg,xmmrm SSE41 -\c PMOVSXBW xmmreg,xmmrm SSE41 -\c PMOVSXBD xmmreg,xmmrm SSE41,SD -\c PMOVSXBQ xmmreg,xmmrm SSE41,SW -\c PMOVSXWD xmmreg,xmmrm SSE41 -\c PMOVSXWQ xmmreg,xmmrm SSE41,SD -\c PMOVSXDQ xmmreg,xmmrm SSE41 -\c PMOVZXBW xmmreg,xmmrm SSE41 -\c PMOVZXBD xmmreg,xmmrm SSE41,SD -\c PMOVZXBQ xmmreg,xmmrm SSE41,SW -\c PMOVZXWD xmmreg,xmmrm SSE41 -\c PMOVZXWQ xmmreg,xmmrm SSE41,SD -\c PMOVZXDQ xmmreg,xmmrm SSE41 -\c PMULDQ xmmreg,xmmrm SSE41 -\c PMULLD xmmreg,xmmrm SSE41 -\c PTEST xmmreg,xmmrm SSE41 -\c ROUNDPD xmmreg,xmmrm,imm SSE41 -\c ROUNDPS xmmreg,xmmrm,imm SSE41 -\c ROUNDSD xmmreg,xmmrm,imm SSE41 -\c ROUNDSS xmmreg,xmmrm,imm SSE41 - -\S{} Nehalem New Instructions (SSE4.2) - -\c CRC32 reg32,rm8 SSE42 -\c CRC32 reg32,rm16 SSE42 -\c CRC32 reg32,rm32 SSE42 -\c CRC32 reg64,rm8 SSE42,X64 -\c CRC32 reg64,rm64 SSE42,X64 -\c PCMPESTRI xmmreg,xmmrm,imm SSE42 -\c PCMPESTRM xmmreg,xmmrm,imm SSE42 -\c PCMPISTRI xmmreg,xmmrm,imm SSE42 -\c PCMPISTRM xmmreg,xmmrm,imm SSE42 -\c PCMPGTQ xmmreg,xmmrm SSE42 -\c POPCNT reg16,rm16 NEHALEM,SW -\c POPCNT reg32,rm32 NEHALEM,SD -\c POPCNT reg64,rm64 NEHALEM,X64 - -\S{} Intel SMX - -\c GETSEC KATMAI - -\S{} Geode (Cyrix) 3DNow! additions - -\c PFRCPV mmxreg,mmxrm PENT,3DNOW,CYRIX -\c PFRSQRTV mmxreg,mmxrm PENT,3DNOW,CYRIX - -\S{} Intel new instructions in ??? - -\c MOVBE reg16,mem16 NEHALEM -\c MOVBE reg32,mem32 NEHALEM -\c MOVBE reg64,mem64 NEHALEM -\c MOVBE mem16,reg16 NEHALEM -\c MOVBE mem32,reg32 NEHALEM -\c MOVBE mem64,reg64 NEHALEM - -\S{} Intel AES instructions - -\c AESENC xmmreg,xmmrm128 SSE,WESTMERE -\c AESENCLAST xmmreg,xmmrm128 SSE,WESTMERE -\c AESDEC xmmreg,xmmrm128 SSE,WESTMERE -\c AESDECLAST xmmreg,xmmrm128 SSE,WESTMERE -\c AESIMC xmmreg,xmmrm128 SSE,WESTMERE -\c AESKEYGENASSIST xmmreg,xmmrm128,imm8 SSE,WESTMERE - -\S{} Intel AVX AES instructions - -\c VAESENC xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VAESENCLAST xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VAESDEC xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VAESDECLAST xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VAESIMC xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VAESKEYGENASSIST xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - -\S{} Intel AVX instructions - -\c VADDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VADDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VADDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VADDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VADDSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VADDSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VADDSUBPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VADDSUBPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VADDSUBPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VADDSUBPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VANDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VANDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VANDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VANDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VANDNPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VANDNPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VANDNPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VANDNPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VBLENDPD xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VBLENDPD ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE -\c VBLENDPS xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VBLENDPS ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE -\c VBLENDVPD xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE -\c VBLENDVPD xmmreg,xmmrm128,xmm0 AVX,SANDYBRIDGE -\c VBLENDVPD ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE -\c VBLENDVPD ymmreg,ymmrm256,ymm0 AVX,SANDYBRIDGE -\c VBLENDVPS xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE -\c VBLENDVPS xmmreg,xmmrm128,xmm0 AVX,SANDYBRIDGE -\c VBLENDVPS ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE -\c VBLENDVPD ymmreg,ymmrm256,ymm0 AVX,SANDYBRIDGE -\c VBROADCASTSS xmmreg,mem32 AVX,SANDYBRIDGE -\c VBROADCASTSS ymmreg,mem32 AVX,SANDYBRIDGE -\c VBROADCASTSD ymmreg,mem64 AVX,SANDYBRIDGE -\c VBROADCASTF128 ymmreg,mem128 AVX,SANDYBRIDGE -\c VCMPEQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPEQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPLTPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPLTPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPLEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPLEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPUNORDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPUNORDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNEQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNEQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNLTPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNLTPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNLEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNLEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPORDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPORDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPEQ_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPEQ_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNGEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNGEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNGTPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNGTPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPFALSEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPFALSEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNEQ_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNEQ_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPGEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPGEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPGTPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPGTPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPTRUEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPTRUEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPEQ_OSPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPEQ_OSPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPLT_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPLT_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPLE_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPLE_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPUNORD_SPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPUNORD_SPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNEQ_USPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNEQ_USPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNLT_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNLT_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNLE_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNLE_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPORD_SPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPORD_SPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPEQ_USPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPEQ_USPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNGE_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNGE_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNGT_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNGT_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPFALSE_OSPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPFALSE_OSPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNEQ_OSPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNEQ_OSPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPGE_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPGE_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPGT_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPGT_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPTRUE_USPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPTRUE_USPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPPD xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VCMPPD ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE -\c VCMPEQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPEQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPLTPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPLTPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPLEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPLEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPUNORDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPUNORDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNEQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNEQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNLTPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNLTPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNLEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNLEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPORDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPORDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPEQ_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPEQ_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNGEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNGEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNGTPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNGTPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPFALSEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPFALSEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNEQ_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNEQ_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPGEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPGEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPGTPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPGTPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPTRUEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPTRUEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPEQ_OSPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPEQ_OSPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPLT_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPLT_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPLE_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPLE_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPUNORD_SPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPUNORD_SPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNEQ_USPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNEQ_USPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNLT_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNLT_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNLE_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNLE_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPORD_SPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPORD_SPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPEQ_USPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPEQ_USPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNGE_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNGE_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNGT_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNGT_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPFALSE_OSPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPFALSE_OSPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPNEQ_OSPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPNEQ_OSPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPGE_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPGE_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPGT_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPGT_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPTRUE_USPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VCMPTRUE_USPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VCMPPS xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VCMPPS ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE -\c VCMPEQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPLTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPLESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPUNORDSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPNEQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPNLTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPNLESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPORDSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPEQ_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPNGESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPNGTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPFALSESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPNEQ_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPGESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPGTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPTRUESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPEQ_OSSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPLT_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPLE_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPUNORD_SSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPNEQ_USSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPNLT_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPNLE_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPORD_SSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPEQ_USSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPNGE_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPNGT_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPFALSE_OSSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPNEQ_OSSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPGE_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPGT_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPTRUE_USSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCMPSD xmmreg,xmmreg*,xmmrm64,imm8 AVX,SANDYBRIDGE -\c VCMPEQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPLTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPLESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPUNORDSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPNEQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPNLTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPNLESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPORDSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPEQ_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPNGESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPNGTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPFALSESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPNEQ_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPGESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPGTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPTRUESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPEQ_OSSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPLT_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPLE_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPUNORD_SSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPNEQ_USSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPNLT_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPNLE_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPORD_SSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPEQ_USSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPNGE_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPNGT_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPFALSE_OSSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPNEQ_OSSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPGE_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPGT_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPTRUE_USSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCMPSS xmmreg,xmmreg*,xmmrm32,imm8 AVX,SANDYBRIDGE -\c VCOMISD xmmreg,xmmrm64 AVX,SANDYBRIDGE -\c VCOMISS xmmreg,xmmrm32 AVX,SANDYBRIDGE -\c VCVTDQ2PD xmmreg,xmmrm64 AVX,SANDYBRIDGE -\c VCVTDQ2PD ymmreg,xmmrm128 AVX,SANDYBRIDGE -\c VCVTDQ2PS xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VCVTDQ2PS ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VCVTPD2DQ xmmreg,xmmreg AVX,SANDYBRIDGE -\c VCVTPD2DQ xmmreg,mem128 AVX,SANDYBRIDGE,SO -\c VCVTPD2DQ xmmreg,ymmreg AVX,SANDYBRIDGE -\c VCVTPD2DQ xmmreg,mem256 AVX,SANDYBRIDGE,SY -\c VCVTPD2PS xmmreg,xmmreg AVX,SANDYBRIDGE -\c VCVTPD2PS xmmreg,mem128 AVX,SANDYBRIDGE,SO -\c VCVTPD2PS xmmreg,ymmreg AVX,SANDYBRIDGE -\c VCVTPD2PS xmmreg,mem256 AVX,SANDYBRIDGE,SY -\c VCVTPS2DQ xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VCVTPS2DQ ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VCVTPS2PD xmmreg,xmmrm64 AVX,SANDYBRIDGE -\c VCVTPS2PD ymmreg,xmmrm128 AVX,SANDYBRIDGE -\c VCVTSD2SI reg32,xmmrm64 AVX,SANDYBRIDGE -\c VCVTSD2SI reg64,xmmrm64 AVX,SANDYBRIDGE,LONG -\c VCVTSD2SS xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VCVTSI2SD xmmreg,xmmreg*,rm32 AVX,SANDYBRIDGE,SD -\c VCVTSI2SD xmmreg,xmmreg*,mem32 AVX,SANDYBRIDGE,ND,SD -\c VCVTSI2SD xmmreg,xmmreg*,rm64 AVX,SANDYBRIDGE,LONG -\c VCVTSI2SS xmmreg,xmmreg*,rm32 AVX,SANDYBRIDGE,SD -\c VCVTSI2SS xmmreg,xmmreg*,mem32 AVX,SANDYBRIDGE,ND,SD -\c VCVTSI2SS xmmreg,xmmreg*,rm64 AVX,SANDYBRIDGE,LONG -\c VCVTSS2SD xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VCVTSS2SI reg32,xmmrm32 AVX,SANDYBRIDGE -\c VCVTSS2SI reg64,xmmrm32 AVX,SANDYBRIDGE,LONG -\c VCVTTPD2DQ xmmreg,xmmreg AVX,SANDYBRIDGE -\c VCVTTPD2DQ xmmreg,mem128 AVX,SANDYBRIDGE,SO -\c VCVTTPD2DQ xmmreg,ymmreg AVX,SANDYBRIDGE -\c VCVTTPD2DQ xmmreg,mem256 AVX,SANDYBRIDGE,SY -\c VCVTTPS2DQ xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VCVTTPS2DQ ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VCVTTSD2SI reg32,xmmrm64 AVX,SANDYBRIDGE -\c VCVTTSD2SI reg64,xmmrm64 AVX,SANDYBRIDGE,LONG -\c VCVTTSS2SI reg32,xmmrm32 AVX,SANDYBRIDGE -\c VCVTTSS2SI reg64,xmmrm32 AVX,SANDYBRIDGE,LONG -\c VDIVPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VDIVPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VDIVPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VDIVPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VDIVSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VDIVSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VDPPD xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VDPPS xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VDPPS ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE -\c VEXTRACTF128 xmmrm128,xmmreg,imm8 AVX,SANDYBRIDGE -\c VEXTRACTPS rm32,xmmreg,imm8 AVX,SANDYBRIDGE -\c VHADDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VHADDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VHADDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VHADDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VHSUBPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VHSUBPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VHSUBPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VHSUBPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VINSERTF128 ymmreg,ymmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VINSERTPS xmmreg,xmmreg*,xmmrm32,imm8 AVX,SANDYBRIDGE -\c VLDDQU xmmreg,mem128 AVX,SANDYBRIDGE -\c VLDQQU ymmreg,mem256 AVX,SANDYBRIDGE -\c VLDDQU ymmreg,mem256 AVX,SANDYBRIDGE -\c VLDMXCSR mem32 AVX,SANDYBRIDGE -\c VMASKMOVDQU xmmreg,xmmreg AVX,SANDYBRIDGE -\c VMASKMOVPS xmmreg,xmmreg,mem128 AVX,SANDYBRIDGE -\c VMASKMOVPS ymmreg,ymmreg,mem256 AVX,SANDYBRIDGE -\c VMASKMOVPS mem128,xmmreg,xmmreg AVX,SANDYBRIDGE,SO -\c VMASKMOVPS mem256,xmmreg,xmmreg AVX,SANDYBRIDGE,SY -\c VMASKMOVPD xmmreg,xmmreg,mem128 AVX,SANDYBRIDGE -\c VMASKMOVPD ymmreg,ymmreg,mem256 AVX,SANDYBRIDGE -\c VMASKMOVPD mem128,xmmreg,xmmreg AVX,SANDYBRIDGE -\c VMASKMOVPD mem256,ymmreg,ymmreg AVX,SANDYBRIDGE -\c VMAXPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VMAXPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VMAXPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VMAXPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VMAXSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VMAXSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VMINPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VMINPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VMINPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VMINPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VMINSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VMINSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VMOVAPD xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VMOVAPD xmmrm128,xmmreg AVX,SANDYBRIDGE -\c VMOVAPD ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VMOVAPD ymmrm256,ymmreg AVX,SANDYBRIDGE -\c VMOVAPS xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VMOVAPS xmmrm128,xmmreg AVX,SANDYBRIDGE -\c VMOVAPS ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VMOVAPS ymmrm256,ymmreg AVX,SANDYBRIDGE -\c VMOVQ xmmreg,xmmrm64 AVX,SANDYBRIDGE -\c VMOVQ xmmrm64,xmmreg AVX,SANDYBRIDGE -\c VMOVQ xmmreg,rm64 AVX,SANDYBRIDGE,LONG -\c VMOVQ rm64,xmmreg AVX,SANDYBRIDGE,LONG -\c VMOVD xmmreg,rm32 AVX,SANDYBRIDGE -\c VMOVD rm32,xmmreg AVX,SANDYBRIDGE -\c VMOVDDUP xmmreg,xmmrm64 AVX,SANDYBRIDGE -\c VMOVDDUP ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VMOVDQA xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VMOVDQA xmmrm128,xmmreg AVX,SANDYBRIDGE -\c VMOVQQA ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VMOVQQA ymmrm256,ymmreg AVX,SANDYBRIDGE -\c VMOVDQA ymmreg,ymmrm AVX,SANDYBRIDGE -\c VMOVDQA ymmrm256,ymmreg AVX,SANDYBRIDGE -\c VMOVDQU xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VMOVDQU xmmrm128,xmmreg AVX,SANDYBRIDGE -\c VMOVQQU ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VMOVQQU ymmrm256,ymmreg AVX,SANDYBRIDGE -\c VMOVDQU ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VMOVDQU ymmrm256,ymmreg AVX,SANDYBRIDGE -\c VMOVHLPS xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE -\c VMOVHPD xmmreg,xmmreg*,mem64 AVX,SANDYBRIDGE -\c VMOVHPD mem64,xmmreg AVX,SANDYBRIDGE -\c VMOVHPS xmmreg,xmmreg*,mem64 AVX,SANDYBRIDGE -\c VMOVHPS mem64,xmmreg AVX,SANDYBRIDGE -\c VMOVLHPS xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE -\c VMOVLPD xmmreg,xmmreg*,mem64 AVX,SANDYBRIDGE -\c VMOVLPD mem64,xmmreg AVX,SANDYBRIDGE -\c VMOVLPS xmmreg,xmmreg*,mem64 AVX,SANDYBRIDGE -\c VMOVLPS mem64,xmmreg AVX,SANDYBRIDGE -\c VMOVMSKPD reg64,xmmreg AVX,SANDYBRIDGE,LONG -\c VMOVMSKPD reg32,xmmreg AVX,SANDYBRIDGE -\c VMOVMSKPD reg64,ymmreg AVX,SANDYBRIDGE,LONG -\c VMOVMSKPD reg32,ymmreg AVX,SANDYBRIDGE -\c VMOVMSKPS reg64,xmmreg AVX,SANDYBRIDGE,LONG -\c VMOVMSKPS reg32,xmmreg AVX,SANDYBRIDGE -\c VMOVMSKPS reg64,ymmreg AVX,SANDYBRIDGE,LONG -\c VMOVMSKPS reg32,ymmreg AVX,SANDYBRIDGE -\c VMOVNTDQ mem128,xmmreg AVX,SANDYBRIDGE -\c VMOVNTQQ mem256,ymmreg AVX,SANDYBRIDGE -\c VMOVNTDQ mem256,ymmreg AVX,SANDYBRIDGE -\c VMOVNTDQA xmmreg,mem128 AVX,SANDYBRIDGE -\c VMOVNTPD mem128,xmmreg AVX,SANDYBRIDGE -\c VMOVNTPD mem256,ymmreg AVX,SANDYBRIDGE -\c VMOVNTPS mem128,xmmreg AVX,SANDYBRIDGE -\c VMOVNTPS mem128,ymmreg AVX,SANDYBRIDGE -\c VMOVSD xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE -\c VMOVSD xmmreg,mem64 AVX,SANDYBRIDGE -\c VMOVSD xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE -\c VMOVSD mem64,xmmreg AVX,SANDYBRIDGE -\c VMOVSHDUP xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VMOVSHDUP ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VMOVSLDUP xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VMOVSLDUP ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VMOVSS xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE -\c VMOVSS xmmreg,mem64 AVX,SANDYBRIDGE -\c VMOVSS xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE -\c VMOVSS mem64,xmmreg AVX,SANDYBRIDGE -\c VMOVUPD xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VMOVUPD xmmrm128,xmmreg AVX,SANDYBRIDGE -\c VMOVUPD ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VMOVUPD ymmrm256,ymmreg AVX,SANDYBRIDGE -\c VMOVUPS xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VMOVUPS xmmrm128,xmmreg AVX,SANDYBRIDGE -\c VMOVUPS ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VMOVUPS ymmrm256,ymmreg AVX,SANDYBRIDGE -\c VMPSADBW xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VMULPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VMULPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VMULPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VMULPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VMULSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VMULSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VORPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VORPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VORPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VORPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VPABSB xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VPABSW xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VPABSD xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VPACKSSWB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPACKSSDW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPACKUSWB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPACKUSDW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPADDB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPADDW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPADDD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPADDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPADDSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPADDSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPADDUSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPADDUSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPALIGNR xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VPAND xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPANDN xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPAVGB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPAVGW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPBLENDVB xmmreg,xmmreg*,xmmrm128,xmmreg AVX,SANDYBRIDGE -\c VPBLENDW xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VPCMPESTRI xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VPCMPESTRM xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VPCMPISTRI xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VPCMPISTRM xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VPCMPEQB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPCMPEQW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPCMPEQD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPCMPEQQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPCMPGTB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPCMPGTW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPCMPGTD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPCMPGTQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPERMILPD xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VPERMILPD ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VPERMILPD xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VPERMILPD ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE -\c VPERMILTD2PD xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE -\c VPERMILTD2PD xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VPERMILTD2PD ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE -\c VPERMILTD2PD ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VPERMILMO2PD xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE -\c VPERMILMO2PD xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VPERMILMO2PD ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE -\c VPERMILMO2PD ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VPERMILMZ2PD xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE -\c VPERMILMZ2PD xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VPERMILMZ2PD ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE -\c VPERMILMZ2PD ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VPERMIL2PD xmmreg,xmmreg,xmmrm128,xmmreg,imm8 AVX,SANDYBRIDGE -\c VPERMIL2PD xmmreg,xmmreg,xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VPERMIL2PD ymmreg,ymmreg,ymmrm256,ymmreg,imm8 AVX,SANDYBRIDGE -\c VPERMIL2PD ymmreg,ymmreg,ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE -\c VPERMILPS xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VPERMILPS ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VPERMILPS xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VPERMILPS ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE -\c VPERMILTD2PS xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE -\c VPERMILTD2PS xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VPERMILTD2PS ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE -\c VPERMILTD2PS ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VPERMILMO2PS xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE -\c VPERMILMO2PS xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VPERMILMO2PS ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE -\c VPERMILMO2PS ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VPERMILMZ2PS xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE -\c VPERMILMZ2PS xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VPERMILMZ2PS ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE -\c VPERMILMZ2PS ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VPERMIL2PS xmmreg,xmmreg,xmmrm128,xmmreg,imm8 AVX,SANDYBRIDGE -\c VPERMIL2PS xmmreg,xmmreg,xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VPERMIL2PS ymmreg,ymmreg,ymmrm256,ymmreg,imm8 AVX,SANDYBRIDGE -\c VPERMIL2PS ymmreg,ymmreg,ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE -\c VPERM2F128 ymmreg,ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE -\c VPEXTRB reg64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG -\c VPEXTRB reg32,xmmreg,imm8 AVX,SANDYBRIDGE -\c VPEXTRB mem8,xmmreg,imm8 AVX,SANDYBRIDGE -\c VPEXTRW reg64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG -\c VPEXTRW reg32,xmmreg,imm8 AVX,SANDYBRIDGE -\c VPEXTRW mem16,xmmreg,imm8 AVX,SANDYBRIDGE -\c VPEXTRW reg64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG -\c VPEXTRW reg32,xmmreg,imm8 AVX,SANDYBRIDGE -\c VPEXTRW mem16,xmmreg,imm8 AVX,SANDYBRIDGE -\c VPEXTRD reg64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG -\c VPEXTRD rm32,xmmreg,imm8 AVX,SANDYBRIDGE -\c VPEXTRQ rm64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG -\c VPHADDW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPHADDD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPHADDSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPHMINPOSUW xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VPHSUBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPHSUBD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPHSUBSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPINSRB xmmreg,xmmreg*,mem8,imm8 AVX,SANDYBRIDGE -\c VPINSRB xmmreg,xmmreg*,rm8,imm8 AVX,SANDYBRIDGE -\c VPINSRB xmmreg,xmmreg*,reg32,imm8 AVX,SANDYBRIDGE -\c VPINSRW xmmreg,xmmreg*,mem16,imm8 AVX,SANDYBRIDGE -\c VPINSRW xmmreg,xmmreg*,rm16,imm8 AVX,SANDYBRIDGE -\c VPINSRW xmmreg,xmmreg*,reg32,imm8 AVX,SANDYBRIDGE -\c VPINSRD xmmreg,xmmreg*,mem32,imm8 AVX,SANDYBRIDGE -\c VPINSRD xmmreg,xmmreg*,rm32,imm8 AVX,SANDYBRIDGE -\c VPINSRQ xmmreg,xmmreg*,mem64,imm8 AVX,SANDYBRIDGE,LONG -\c VPINSRQ xmmreg,xmmreg*,rm64,imm8 AVX,SANDYBRIDGE,LONG -\c VPMADDWD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPMADDUBSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPMAXSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPMAXSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPMAXSD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPMAXUB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPMAXUW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPMAXUD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPMINSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPMINSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPMINSD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPMINUB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPMINUW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPMINUD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPMOVMSKB reg64,xmmreg AVX,SANDYBRIDGE,LONG -\c VPMOVMSKB reg32,xmmreg AVX,SANDYBRIDGE -\c VPMOVSXBW xmmreg,xmmrm64 AVX,SANDYBRIDGE -\c VPMOVSXBD xmmreg,xmmrm32 AVX,SANDYBRIDGE -\c VPMOVSXBQ xmmreg,xmmrm16 AVX,SANDYBRIDGE -\c VPMOVSXWD xmmreg,xmmrm64 AVX,SANDYBRIDGE -\c VPMOVSXWQ xmmreg,xmmrm32 AVX,SANDYBRIDGE -\c VPMOVSXDQ xmmreg,xmmrm64 AVX,SANDYBRIDGE -\c VPMOVZXBW xmmreg,xmmrm64 AVX,SANDYBRIDGE -\c VPMOVZXBD xmmreg,xmmrm32 AVX,SANDYBRIDGE -\c VPMOVZXBQ xmmreg,xmmrm16 AVX,SANDYBRIDGE -\c VPMOVZXWD xmmreg,xmmrm64 AVX,SANDYBRIDGE -\c VPMOVZXWQ xmmreg,xmmrm32 AVX,SANDYBRIDGE -\c VPMOVZXDQ xmmreg,xmmrm64 AVX,SANDYBRIDGE -\c VPMULHUW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPMULHRSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPMULHW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPMULLW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPMULLD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPMULUDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPMULDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPOR xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSADBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSHUFB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSHUFD xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VPSHUFHW xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VPSHUFLW xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VPSIGNB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSIGNW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSIGND xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSLLDQ xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE -\c VPSRLDQ xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE -\c VPSLLW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSLLW xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE -\c VPSLLD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSLLD xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE -\c VPSLLQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSLLQ xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE -\c VPSRAW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSRAW xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE -\c VPSRAD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSRAD xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE -\c VPSRLW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSRLW xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE -\c VPSRLD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSRLD xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE -\c VPSRLQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSRLQ xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE -\c VPTEST xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VPTEST ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VPSUBB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSUBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSUBD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSUBQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSUBSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSUBSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSUBUSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPSUBUSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPUNPCKHBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPUNPCKHWD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPUNPCKHDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPUNPCKHQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPUNPCKLBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPUNPCKLWD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPUNPCKLDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPUNPCKLQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPXOR xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VRCPPS xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VRCPPS ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VRCPSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VRSQRTPS xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VRSQRTPS ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VRSQRTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VROUNDPD xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VROUNDPD ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE -\c VROUNDPS xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VROUNDPS ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE -\c VROUNDSD xmmreg,xmmreg*,xmmrm64,imm8 AVX,SANDYBRIDGE -\c VROUNDSS xmmreg,xmmreg*,xmmrm32,imm8 AVX,SANDYBRIDGE -\c VSHUFPD xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VSHUFPD ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE -\c VSHUFPS xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE -\c VSHUFPS ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE -\c VSQRTPD xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VSQRTPD ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VSQRTPS xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VSQRTPS ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VSQRTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VSQRTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VSTMXCSR mem32 AVX,SANDYBRIDGE -\c VSUBPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VSUBPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VSUBPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VSUBPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VSUBSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE -\c VSUBSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE -\c VTESTPS xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VTESTPS ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VTESTPD xmmreg,xmmrm128 AVX,SANDYBRIDGE -\c VTESTPD ymmreg,ymmrm256 AVX,SANDYBRIDGE -\c VUCOMISD xmmreg,xmmrm64 AVX,SANDYBRIDGE -\c VUCOMISS xmmreg,xmmrm32 AVX,SANDYBRIDGE -\c VUNPCKHPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VUNPCKHPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VUNPCKHPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VUNPCKHPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VUNPCKLPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VUNPCKLPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VUNPCKLPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VUNPCKLPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VXORPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VXORPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VXORPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VXORPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE -\c VZEROALL AVX,SANDYBRIDGE -\c VZEROUPPER AVX,SANDYBRIDGE - -\S{} Intel Carry-Less Multiplication instructions (CLMUL) - -\c PCLMULLQLQDQ xmmreg,xmmrm128 SSE,WESTMERE -\c PCLMULHQLQDQ xmmreg,xmmrm128 SSE,WESTMERE -\c PCLMULLQHQDQ xmmreg,xmmrm128 SSE,WESTMERE -\c PCLMULHQHQDQ xmmreg,xmmrm128 SSE,WESTMERE -\c PCLMULQDQ xmmreg,xmmrm128,imm8 SSE,WESTMERE - -\S{} Intel AVX Carry-Less Multiplication instructions (CLMUL) - -\c VPCLMULLQLQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPCLMULHQLQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPCLMULLQHQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPCLMULHQHQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE -\c VPCLMULQDQ xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - -\S{} Intel Fused Multiply-Add instructions (FMA) - -\c VFMADD132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADD132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADD132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADD132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADD312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADD312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADD312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADD312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADD213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADD213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADD213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADD213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADD123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADD123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADD123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADD123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADD231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADD231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADD231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADD231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADD321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADD321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADD321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADD321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADDSUB132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADDSUB132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADDSUB132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADDSUB132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADDSUB312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADDSUB312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADDSUB312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADDSUB312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADDSUB213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADDSUB213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADDSUB213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADDSUB213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADDSUB123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADDSUB123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADDSUB123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADDSUB123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADDSUB231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADDSUB231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADDSUB231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADDSUB231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADDSUB321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADDSUB321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADDSUB321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMADDSUB321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUB132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUB132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUB132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUB132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUB312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUB312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUB312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUB312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUB213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUB213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUB213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUB213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUB123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUB123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUB123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUB123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUB231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUB231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUB231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUB231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUB321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUB321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUB321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUB321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUBADD132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUBADD132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUBADD132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUBADD132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUBADD312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUBADD312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUBADD312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUBADD312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUBADD213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUBADD213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUBADD213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUBADD213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUBADD123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUBADD123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUBADD123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUBADD123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUBADD231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUBADD231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUBADD231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUBADD231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUBADD321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUBADD321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMSUBADD321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFMSUBADD321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMADD132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMADD132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMADD132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMADD132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMADD312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMADD312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMADD312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMADD312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMADD213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMADD213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMADD213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMADD213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMADD123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMADD123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMADD123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMADD123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMADD231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMADD231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMADD231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMADD231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMADD321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMADD321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMADD321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMADD321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMSUB132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMSUB132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMSUB132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMSUB132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMSUB312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMSUB312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMSUB312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMSUB312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMSUB213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMSUB213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMSUB213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMSUB213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMSUB123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMSUB123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMSUB123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMSUB123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMSUB231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMSUB231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMSUB231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMSUB231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMSUB321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMSUB321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFNMSUB321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE -\c VFNMSUB321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE -\c VFMADD132SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFMADD132SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFMADD312SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFMADD312SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFMADD213SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFMADD213SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFMADD123SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFMADD123SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFMADD231SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFMADD231SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFMADD321SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFMADD321SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFMSUB132SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFMSUB132SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFMSUB312SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFMSUB312SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFMSUB213SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFMSUB213SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFMSUB123SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFMSUB123SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFMSUB231SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFMSUB231SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFMSUB321SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFMSUB321SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFNMADD132SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFNMADD132SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFNMADD312SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFNMADD312SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFNMADD213SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFNMADD213SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFNMADD123SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFNMADD123SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFNMADD231SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFNMADD231SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFNMADD321SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFNMADD321SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFNMSUB132SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFNMSUB132SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFNMSUB312SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFNMSUB312SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFNMSUB213SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFNMSUB213SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFNMSUB123SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFNMSUB123SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFNMSUB231SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFNMSUB231SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE -\c VFNMSUB321SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE -\c VFNMSUB321SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - -\S{} VIA (Centaur) security instructions - -\c XSTORE PENT,CYRIX -\c XCRYPTECB PENT,CYRIX -\c XCRYPTCBC PENT,CYRIX -\c XCRYPTCTR PENT,CYRIX -\c XCRYPTCFB PENT,CYRIX -\c XCRYPTOFB PENT,CYRIX -\c MONTMUL PENT,CYRIX -\c XSHA1 PENT,CYRIX -\c XSHA256 PENT,CYRIX - -\S{} AMD Lightweight Profiling (LWP) instructions - -\c LLWPCB reg16 AMD -\c LLWPCB reg32 AMD,386 -\c LLWPCB reg64 AMD,X64 -\c SLWPCB reg16 AMD -\c SLWPCB reg32 AMD,386 -\c SLWPCB reg64 AMD,X64 -\c LWPVAL reg16,rm32,imm16 AMD,386 -\c LWPVAL reg32,rm32,imm32 AMD,386 -\c LWPVAL reg64,rm32,imm32 AMD,X64 -\c LWPINS reg16,rm32,imm16 AMD,386 -\c LWPINS reg32,rm32,imm32 AMD,386 -\c LWPINS reg64,rm32,imm32 AMD,X64 - -\S{} AMD XOP, FMA4 and CVT16 instructions (SSE5) - -\c VCVTPH2PS xmmreg,xmmrm64*,imm8 AMD,SSE5 -\c VCVTPH2PS ymmreg,xmmrm128,imm8 AMD,SSE5 -\c VCVTPH2PS ymmreg,ymmrm128*,imm8 AMD,SSE5 -\c VCVTPS2PH xmmrm64,xmmreg*,imm8 AMD,SSE5 -\c VCVTPS2PH xmmrm128,ymmreg,imm8 AMD,SSE5 -\c VCVTPS2PH ymmrm128,ymmreg*,imm8 AMD,SSE5 -\c VFMADDPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VFMADDPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -\c VFMADDPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -\c VFMADDPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -\c VFMADDPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VFMADDPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -\c VFMADDPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -\c VFMADDPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -\c VFMADDSD xmmreg,xmmreg*,xmmrm64,xmmreg AMD,SSE5 -\c VFMADDSD xmmreg,xmmreg*,xmmreg,xmmrm64 AMD,SSE5 -\c VFMADDSS xmmreg,xmmreg*,xmmrm32,xmmreg AMD,SSE5 -\c VFMADDSS xmmreg,xmmreg*,xmmreg,xmmrm32 AMD,SSE5 -\c VFMADDSUBPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VFMADDSUBPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -\c VFMADDSUBPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -\c VFMADDSUBPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -\c VFMADDSUBPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VFMADDSUBPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -\c VFMADDSUBPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -\c VFMADDSUBPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -\c VFMSUBADDPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VFMSUBADDPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -\c VFMSUBADDPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -\c VFMSUBADDPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -\c VFMSUBADDPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VFMSUBADDPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -\c VFMSUBADDPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -\c VFMSUBADDPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -\c VFMSUBPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VFMSUBPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -\c VFMSUBPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -\c VFMSUBPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -\c VFMSUBPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VFMSUBPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -\c VFMSUBPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -\c VFMSUBPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -\c VFMSUBSD xmmreg,xmmreg*,xmmrm64,xmmreg AMD,SSE5 -\c VFMSUBSD xmmreg,xmmreg*,xmmreg,xmmrm64 AMD,SSE5 -\c VFMSUBSS xmmreg,xmmreg*,xmmrm32,xmmreg AMD,SSE5 -\c VFMSUBSS xmmreg,xmmreg*,xmmreg,xmmrm32 AMD,SSE5 -\c VFNMADDPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VFNMADDPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -\c VFNMADDPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -\c VFNMADDPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -\c VFNMADDPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VFNMADDPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -\c VFNMADDPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -\c VFNMADDPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -\c VFNMADDSD xmmreg,xmmreg*,xmmrm64,xmmreg AMD,SSE5 -\c VFNMADDSD xmmreg,xmmreg*,xmmreg,xmmrm64 AMD,SSE5 -\c VFNMADDSS xmmreg,xmmreg*,xmmrm32,xmmreg AMD,SSE5 -\c VFNMADDSS xmmreg,xmmreg*,xmmreg,xmmrm32 AMD,SSE5 -\c VFNMSUBPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VFNMSUBPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -\c VFNMSUBPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -\c VFNMSUBPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -\c VFNMSUBPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VFNMSUBPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -\c VFNMSUBPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -\c VFNMSUBPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -\c VFNMSUBSD xmmreg,xmmreg*,xmmrm64,xmmreg AMD,SSE5 -\c VFNMSUBSD xmmreg,xmmreg*,xmmreg,xmmrm64 AMD,SSE5 -\c VFNMSUBSS xmmreg,xmmreg*,xmmrm32,xmmreg AMD,SSE5 -\c VFNMSUBSS xmmreg,xmmreg*,xmmreg,xmmrm32 AMD,SSE5 -\c VFRCZPD xmmreg,xmmrm128* AMD,SSE5 -\c VFRCZPD ymmreg,ymmrm256* AMD,SSE5 -\c VFRCZPS xmmreg,xmmrm128* AMD,SSE5 -\c VFRCZPS ymmreg,ymmrm256* AMD,SSE5 -\c VFRCZSD xmmreg,xmmrm64* AMD,SSE5 -\c VFRCZSS xmmreg,xmmrm32* AMD,SSE5 -\c VPCMOV xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VPCMOV ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 -\c VPCMOV xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -\c VPCMOV ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 -\c VPCOMB xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 -\c VPCOMD xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 -\c VPCOMQ xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 -\c VPCOMUB xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 -\c VPCOMUD xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 -\c VPCOMUQ xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 -\c VPCOMUW xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 -\c VPCOMW xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 -\c VPHADDBD xmmreg,xmmrm128* AMD,SSE5 -\c VPHADDBQ xmmreg,xmmrm128* AMD,SSE5 -\c VPHADDBW xmmreg,xmmrm128* AMD,SSE5 -\c VPHADDDQ xmmreg,xmmrm128* AMD,SSE5 -\c VPHADDUBD xmmreg,xmmrm128* AMD,SSE5 -\c VPHADDUBQ xmmreg,xmmrm128* AMD,SSE5 -\c VPHADDUBW xmmreg,xmmrm128* AMD,SSE5 -\c VPHADDUDQ xmmreg,xmmrm128* AMD,SSE5 -\c VPHADDUWD xmmreg,xmmrm128* AMD,SSE5 -\c VPHADDUWQ xmmreg,xmmrm128* AMD,SSE5 -\c VPHADDWD xmmreg,xmmrm128* AMD,SSE5 -\c VPHADDWQ xmmreg,xmmrm128* AMD,SSE5 -\c VPHSUBBW xmmreg,xmmrm128* AMD,SSE5 -\c VPHSUBDQ xmmreg,xmmrm128* AMD,SSE5 -\c VPHSUBWD xmmreg,xmmrm128* AMD,SSE5 -\c VPMACSDD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VPMACSDQH xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VPMACSDQL xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VPMACSSDD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VPMACSSDQH xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VPMACSSDQL xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VPMACSSWD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VPMACSSWW xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VPMACSWD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VPMACSWW xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VPMADCSSWD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VPMADCSWD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VPPERM xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 -\c VPPERM xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 -\c VPROTB xmmreg,xmmrm128*,xmmreg AMD,SSE5 -\c VPROTB xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -\c VPROTB xmmreg,xmmrm128*,imm8 AMD,SSE5 -\c VPROTD xmmreg,xmmrm128*,xmmreg AMD,SSE5 -\c VPROTD xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -\c VPROTD xmmreg,xmmrm128*,imm8 AMD,SSE5 -\c VPROTQ xmmreg,xmmrm128*,xmmreg AMD,SSE5 -\c VPROTQ xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -\c VPROTQ xmmreg,xmmrm128*,imm8 AMD,SSE5 -\c VPROTW xmmreg,xmmrm128*,xmmreg AMD,SSE5 -\c VPROTW xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -\c VPROTW xmmreg,xmmrm128*,imm8 AMD,SSE5 -\c VPSHAB xmmreg,xmmrm128*,xmmreg AMD,SSE5 -\c VPSHAB xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -\c VPSHAD xmmreg,xmmrm128*,xmmreg AMD,SSE5 -\c VPSHAD xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -\c VPSHAQ xmmreg,xmmrm128*,xmmreg AMD,SSE5 -\c VPSHAQ xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -\c VPSHAW xmmreg,xmmrm128*,xmmreg AMD,SSE5 -\c VPSHAW xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -\c VPSHLB xmmreg,xmmrm128*,xmmreg AMD,SSE5 -\c VPSHLB xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -\c VPSHLD xmmreg,xmmrm128*,xmmreg AMD,SSE5 -\c VPSHLD xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -\c VPSHLQ xmmreg,xmmrm128*,xmmreg AMD,SSE5 -\c VPSHLQ xmmreg,xmmreg*,xmmrm128 AMD,SSE5 -\c VPSHLW xmmreg,xmmrm128*,xmmreg AMD,SSE5 -\c VPSHLW xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - -\S{} Systematic names for the hinting nop instructions - -\c HINT_NOP0 rm16 P6,UNDOC -\c HINT_NOP0 rm32 P6,UNDOC -\c HINT_NOP0 rm64 X64,UNDOC -\c HINT_NOP1 rm16 P6,UNDOC -\c HINT_NOP1 rm32 P6,UNDOC -\c HINT_NOP1 rm64 X64,UNDOC -\c HINT_NOP2 rm16 P6,UNDOC -\c HINT_NOP2 rm32 P6,UNDOC -\c HINT_NOP2 rm64 X64,UNDOC -\c HINT_NOP3 rm16 P6,UNDOC -\c HINT_NOP3 rm32 P6,UNDOC -\c HINT_NOP3 rm64 X64,UNDOC -\c HINT_NOP4 rm16 P6,UNDOC -\c HINT_NOP4 rm32 P6,UNDOC -\c HINT_NOP4 rm64 X64,UNDOC -\c HINT_NOP5 rm16 P6,UNDOC -\c HINT_NOP5 rm32 P6,UNDOC -\c HINT_NOP5 rm64 X64,UNDOC -\c HINT_NOP6 rm16 P6,UNDOC -\c HINT_NOP6 rm32 P6,UNDOC -\c HINT_NOP6 rm64 X64,UNDOC -\c HINT_NOP7 rm16 P6,UNDOC -\c HINT_NOP7 rm32 P6,UNDOC -\c HINT_NOP7 rm64 X64,UNDOC -\c HINT_NOP8 rm16 P6,UNDOC -\c HINT_NOP8 rm32 P6,UNDOC -\c HINT_NOP8 rm64 X64,UNDOC -\c HINT_NOP9 rm16 P6,UNDOC -\c HINT_NOP9 rm32 P6,UNDOC -\c HINT_NOP9 rm64 X64,UNDOC -\c HINT_NOP10 rm16 P6,UNDOC -\c HINT_NOP10 rm32 P6,UNDOC -\c HINT_NOP10 rm64 X64,UNDOC -\c HINT_NOP11 rm16 P6,UNDOC -\c HINT_NOP11 rm32 P6,UNDOC -\c HINT_NOP11 rm64 X64,UNDOC -\c HINT_NOP12 rm16 P6,UNDOC -\c HINT_NOP12 rm32 P6,UNDOC -\c HINT_NOP12 rm64 X64,UNDOC -\c HINT_NOP13 rm16 P6,UNDOC -\c HINT_NOP13 rm32 P6,UNDOC -\c HINT_NOP13 rm64 X64,UNDOC -\c HINT_NOP14 rm16 P6,UNDOC -\c HINT_NOP14 rm32 P6,UNDOC -\c HINT_NOP14 rm64 X64,UNDOC -\c HINT_NOP15 rm16 P6,UNDOC -\c HINT_NOP15 rm32 P6,UNDOC -\c HINT_NOP15 rm64 X64,UNDOC -\c HINT_NOP16 rm16 P6,UNDOC -\c HINT_NOP16 rm32 P6,UNDOC -\c HINT_NOP16 rm64 X64,UNDOC -\c HINT_NOP17 rm16 P6,UNDOC -\c HINT_NOP17 rm32 P6,UNDOC -\c HINT_NOP17 rm64 X64,UNDOC -\c HINT_NOP18 rm16 P6,UNDOC -\c HINT_NOP18 rm32 P6,UNDOC -\c HINT_NOP18 rm64 X64,UNDOC -\c HINT_NOP19 rm16 P6,UNDOC -\c HINT_NOP19 rm32 P6,UNDOC -\c HINT_NOP19 rm64 X64,UNDOC -\c HINT_NOP20 rm16 P6,UNDOC -\c HINT_NOP20 rm32 P6,UNDOC -\c HINT_NOP20 rm64 X64,UNDOC -\c HINT_NOP21 rm16 P6,UNDOC -\c HINT_NOP21 rm32 P6,UNDOC -\c HINT_NOP21 rm64 X64,UNDOC -\c HINT_NOP22 rm16 P6,UNDOC -\c HINT_NOP22 rm32 P6,UNDOC -\c HINT_NOP22 rm64 X64,UNDOC -\c HINT_NOP23 rm16 P6,UNDOC -\c HINT_NOP23 rm32 P6,UNDOC -\c HINT_NOP23 rm64 X64,UNDOC -\c HINT_NOP24 rm16 P6,UNDOC -\c HINT_NOP24 rm32 P6,UNDOC -\c HINT_NOP24 rm64 X64,UNDOC -\c HINT_NOP25 rm16 P6,UNDOC -\c HINT_NOP25 rm32 P6,UNDOC -\c HINT_NOP25 rm64 X64,UNDOC -\c HINT_NOP26 rm16 P6,UNDOC -\c HINT_NOP26 rm32 P6,UNDOC -\c HINT_NOP26 rm64 X64,UNDOC -\c HINT_NOP27 rm16 P6,UNDOC -\c HINT_NOP27 rm32 P6,UNDOC -\c HINT_NOP27 rm64 X64,UNDOC -\c HINT_NOP28 rm16 P6,UNDOC -\c HINT_NOP28 rm32 P6,UNDOC -\c HINT_NOP28 rm64 X64,UNDOC -\c HINT_NOP29 rm16 P6,UNDOC -\c HINT_NOP29 rm32 P6,UNDOC -\c HINT_NOP29 rm64 X64,UNDOC -\c HINT_NOP30 rm16 P6,UNDOC -\c HINT_NOP30 rm32 P6,UNDOC -\c HINT_NOP30 rm64 X64,UNDOC -\c HINT_NOP31 rm16 P6,UNDOC -\c HINT_NOP31 rm32 P6,UNDOC -\c HINT_NOP31 rm64 X64,UNDOC -\c HINT_NOP32 rm16 P6,UNDOC -\c HINT_NOP32 rm32 P6,UNDOC -\c HINT_NOP32 rm64 X64,UNDOC -\c HINT_NOP33 rm16 P6,UNDOC -\c HINT_NOP33 rm32 P6,UNDOC -\c HINT_NOP33 rm64 X64,UNDOC -\c HINT_NOP34 rm16 P6,UNDOC -\c HINT_NOP34 rm32 P6,UNDOC -\c HINT_NOP34 rm64 X64,UNDOC -\c HINT_NOP35 rm16 P6,UNDOC -\c HINT_NOP35 rm32 P6,UNDOC -\c HINT_NOP35 rm64 X64,UNDOC -\c HINT_NOP36 rm16 P6,UNDOC -\c HINT_NOP36 rm32 P6,UNDOC -\c HINT_NOP36 rm64 X64,UNDOC -\c HINT_NOP37 rm16 P6,UNDOC -\c HINT_NOP37 rm32 P6,UNDOC -\c HINT_NOP37 rm64 X64,UNDOC -\c HINT_NOP38 rm16 P6,UNDOC -\c HINT_NOP38 rm32 P6,UNDOC -\c HINT_NOP38 rm64 X64,UNDOC -\c HINT_NOP39 rm16 P6,UNDOC -\c HINT_NOP39 rm32 P6,UNDOC -\c HINT_NOP39 rm64 X64,UNDOC -\c HINT_NOP40 rm16 P6,UNDOC -\c HINT_NOP40 rm32 P6,UNDOC -\c HINT_NOP40 rm64 X64,UNDOC -\c HINT_NOP41 rm16 P6,UNDOC -\c HINT_NOP41 rm32 P6,UNDOC -\c HINT_NOP41 rm64 X64,UNDOC -\c HINT_NOP42 rm16 P6,UNDOC -\c HINT_NOP42 rm32 P6,UNDOC -\c HINT_NOP42 rm64 X64,UNDOC -\c HINT_NOP43 rm16 P6,UNDOC -\c HINT_NOP43 rm32 P6,UNDOC -\c HINT_NOP43 rm64 X64,UNDOC -\c HINT_NOP44 rm16 P6,UNDOC -\c HINT_NOP44 rm32 P6,UNDOC -\c HINT_NOP44 rm64 X64,UNDOC -\c HINT_NOP45 rm16 P6,UNDOC -\c HINT_NOP45 rm32 P6,UNDOC -\c HINT_NOP45 rm64 X64,UNDOC -\c HINT_NOP46 rm16 P6,UNDOC -\c HINT_NOP46 rm32 P6,UNDOC -\c HINT_NOP46 rm64 X64,UNDOC -\c HINT_NOP47 rm16 P6,UNDOC -\c HINT_NOP47 rm32 P6,UNDOC -\c HINT_NOP47 rm64 X64,UNDOC -\c HINT_NOP48 rm16 P6,UNDOC -\c HINT_NOP48 rm32 P6,UNDOC -\c HINT_NOP48 rm64 X64,UNDOC -\c HINT_NOP49 rm16 P6,UNDOC -\c HINT_NOP49 rm32 P6,UNDOC -\c HINT_NOP49 rm64 X64,UNDOC -\c HINT_NOP50 rm16 P6,UNDOC -\c HINT_NOP50 rm32 P6,UNDOC -\c HINT_NOP50 rm64 X64,UNDOC -\c HINT_NOP51 rm16 P6,UNDOC -\c HINT_NOP51 rm32 P6,UNDOC -\c HINT_NOP51 rm64 X64,UNDOC -\c HINT_NOP52 rm16 P6,UNDOC -\c HINT_NOP52 rm32 P6,UNDOC -\c HINT_NOP52 rm64 X64,UNDOC -\c HINT_NOP53 rm16 P6,UNDOC -\c HINT_NOP53 rm32 P6,UNDOC -\c HINT_NOP53 rm64 X64,UNDOC -\c HINT_NOP54 rm16 P6,UNDOC -\c HINT_NOP54 rm32 P6,UNDOC -\c HINT_NOP54 rm64 X64,UNDOC -\c HINT_NOP55 rm16 P6,UNDOC -\c HINT_NOP55 rm32 P6,UNDOC -\c HINT_NOP55 rm64 X64,UNDOC -\c HINT_NOP56 rm16 P6,UNDOC -\c HINT_NOP56 rm32 P6,UNDOC -\c HINT_NOP56 rm64 X64,UNDOC -\c HINT_NOP57 rm16 P6,UNDOC -\c HINT_NOP57 rm32 P6,UNDOC -\c HINT_NOP57 rm64 X64,UNDOC -\c HINT_NOP58 rm16 P6,UNDOC -\c HINT_NOP58 rm32 P6,UNDOC -\c HINT_NOP58 rm64 X64,UNDOC -\c HINT_NOP59 rm16 P6,UNDOC -\c HINT_NOP59 rm32 P6,UNDOC -\c HINT_NOP59 rm64 X64,UNDOC -\c HINT_NOP60 rm16 P6,UNDOC -\c HINT_NOP60 rm32 P6,UNDOC -\c HINT_NOP60 rm64 X64,UNDOC -\c HINT_NOP61 rm16 P6,UNDOC -\c HINT_NOP61 rm32 P6,UNDOC -\c HINT_NOP61 rm64 X64,UNDOC -\c HINT_NOP62 rm16 P6,UNDOC -\c HINT_NOP62 rm32 P6,UNDOC -\c HINT_NOP62 rm64 X64,UNDOC -\c HINT_NOP63 rm16 P6,UNDOC -\c HINT_NOP63 rm32 P6,UNDOC -\c HINT_NOP63 rm64 X64,UNDOC - diff --git a/doc/nasmdoc.pdf b/doc/nasmdoc.pdf Binary files differdeleted file mode 100644 index 0dbf46a..0000000 --- a/doc/nasmdoc.pdf +++ /dev/null diff --git a/doc/nasmdoc.ps b/doc/nasmdoc.ps deleted file mode 100644 index e7d0040..0000000 --- a/doc/nasmdoc.ps +++ /dev/null @@ -1,12251 +0,0 @@ -%!PS-Adobe-3.0 -%%Pages: 213 -%%BoundingBox: 0 0 595 792 -%%Creator: (NASM psflow.pl) -%%DocumentData: Clean7Bit -%%DocumentFonts: Times-Italic Times-Bold Courier Times-Roman Courier-Bold Times-BoldItalic -%%DocumentNeededFonts: Times-Italic Times-Bold Courier Times-Roman Courier-Bold Times-BoldItalic -%%Orientation: Portrait -%%PageOrder: Ascend -%%EndComments -%%BeginProlog -/tocind 12 def -/pymarg 50 def -/idxindent 24 def -/pageheight 792 def -/pagewidth 595 def -/idxcolumns 2 def -/tocpnz 24 def -/bulladj 12 def -/plmarg 50 def -/idxgutter 24 def -/lmarg 100 def -/topmarg 100 def -/botmarg 100 def -/prmarg 0 def -/startcopyright 75 def -/tocdots 8 def -/rmarg 50 def -/idxspace 24 def -/colorlinks false def -/NASMEncoding [ /.notdef /.notdef /.notdef /.notdef /.notdef - /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef - /.notdef /.notdef /.notdef /.notdef /dotlessi /grave /acute - /circumflex /tilde /macron /breve /dotaccent /dieresis /.notdef /ring - /cedilla /.notdef /hungarumlaut /ogonek /caron /space /exclam - /quotedbl /numbersign /dollar /percent /ampersand /quoteright - /parenleft /parenright /asterisk /plus /comma /minus /period /slash - /zero /one /two /three /four /five /six /seven /eight /nine /colon - /semicolon /less /equal /greater /question /at /A /B /C /D /E /F /G /H - /I /J /K /L /M /N /O /P /Q /R /S /T /U /V /W /X /Y /Z /bracketleft - /backslash /bracketright /asciicircum /underscore /quoteleft /a /b /c - /d /e /f /g /h /i /j /k /l /m /n /o /p /q /r /s /t /u /v /w /x /y /z - /braceleft /bar /braceright /asciitilde /.notdef /.notdef /.notdef - /quotesinglbase /florin /quotedblbase /ellipsis /dagger /dbldagger - /circumflex /perthousand /Scaron /guilsinglleft /OE /.notdef /Zcaron - /.notdef /.notdef /grave /quotesingle /quotedblleft /quotedblright - /bullet /endash /emdash /tilde /trademark /scaron /guilsignlright /oe - /.notdef /zcaron /Ydieresis /space /exclamdown /cent /sterling - /currency /yen /brokenbar /section /dieresis /copyright /ordfeminine - /guillemotleft /logicalnot /hyphen /registered /macron /degree - /plusminus /twosuperior /threesuperior /acute /mu /paragraph - /periodcentered /cedilla /onesuperior /ordmasculine /guillemotright - /onequarter /onehalf /threequarters /questiondown /Agrave /Aacute - /Acircumflex /Atilde /Adieresis /Aring /AE /Ccedilla /Egrave /Eacute - /Ecircumflex /Edieresis /Igrave /Iacute /Icircumflex /Idieresis /Eth - /Ntilde /Ograve /Oacute /Ocircumflex /Otilde /Odieresis /multiply - /Oslash /Ugrave /Uacute /Ucircumflex /Udieresis /Yacute /Thorn - /germandbls /agrave /aacute /acircumflex /atilde /adieresis /aring /ae - /ccedilla /egrave /eacute /ecircumflex /edieresis /igrave /iacute - /icircumflex /idieresis /eth /ntilde /ograve /oacute /ocircumflex - /otilde /odieresis /divide /oslash /ugrave /uacute /ucircumflex - /udieresis /yacute /thorn /ydieresis ] def -/nasmenc { - findfont dup length dict begin - { 1 index /FID ne {def}{pop pop} ifelse } forall - /Encoding NASMEncoding def - currentdict - end - definefont pop -} def -/Times-Italic-NASM /Times-Italic nasmenc -/Times-Bold-NASM /Times-Bold nasmenc -/Courier-NASM /Courier nasmenc -/Times-Roman-NASM /Times-Roman nasmenc -/Courier-Bold-NASM /Courier-Bold nasmenc -/Times-BoldItalic-NASM /Times-BoldItalic nasmenc -/tfont0 /Times-Bold-NASM findfont 20 scalefont def -/tfont1 /Times-BoldItalic-NASM findfont 20 scalefont def -/tfont2 /Courier-Bold-NASM findfont 20 scalefont def -/tfont [tfont0 tfont1 tfont2] def -/cfont0 /Times-Bold-NASM findfont 18 scalefont def -/cfont1 /Times-BoldItalic-NASM findfont 18 scalefont def -/cfont2 /Courier-Bold-NASM findfont 18 scalefont def -/cfont [cfont0 cfont1 cfont2] def -/hfont0 /Times-Bold-NASM findfont 14 scalefont def -/hfont1 /Times-BoldItalic-NASM findfont 14 scalefont def -/hfont2 /Courier-Bold-NASM findfont 14 scalefont def -/hfont [hfont0 hfont1 hfont2] def -/sfont0 /Times-Bold-NASM findfont 12 scalefont def -/sfont1 /Times-BoldItalic-NASM findfont 12 scalefont def -/sfont2 /Courier-Bold-NASM findfont 12 scalefont def -/sfont [sfont0 sfont1 sfont2] def -/bfont0 /Times-Roman-NASM findfont 10 scalefont def -/bfont1 /Times-Italic-NASM findfont 10 scalefont def -/bfont2 /Courier-NASM findfont 10 scalefont def -/bfont [bfont0 bfont1 bfont2] def -/bullet [(\225)] def -% -% PostScript header for NASM documentation -% - -% Avoid barfing on old PS implementations -/pdfmark where -{pop} {userdict /pdfmark /cleartomark load put} ifelse -/setpagedevice where -{pop} {userdict /setpagedevice /pop load put} ifelse - -% Useful definition -/space 32 def - -% -% This asks the PostScript interpreter for the proper size paper -% -/setpagesize { - 1 dict dup /PageSize [pagewidth pageheight] put setpagedevice -} def - -% -% Code to handle links -% -/min { 2 copy gt { exch } if pop } def -/max { 2 copy lt { exch } if pop } def - -/lkbegun 0 def -/lktype null def -/lkury 0 def -/lkurx 0 def -/lklly 0 def -/lkllx 0 def -/lkxmarg 1 def % Extra space for link in x dir -/lkymarg 1 def % Extra space for link in y dir -/lktarget () def - -% target type -- -/linkbegin { - userdict begin - /lkbegun 1 def - /lktype exch def - /lktarget exch def - colorlinks { 0 0 0.4 setrgbcolor } if - end -} def - -% target -- -/linkbegindest { - /Dest linkbegin -} def - -% uristring -- -/linkbeginuri { - /URI linkbegin -} def - -% pageno -- -/linkbeginpage { - /Page linkbegin -} def - -% string spacepadding -- -/linkshow { - userdict begin - /lspad exch def /lss exch def - lkbegun 0 ne { - gsave lss true charpath flattenpath pathbbox grestore - lkbegun 1 eq { - /lkury exch def - lss spacecount lspad mul add /lkurx exch def - /lklly exch def - /lkllx exch def - /lkbegun 2 def - } { - lkury max /lkury exch def - lss spacecount lspad mul add lkurx max /lkurx exch def - lklly min /lklly exch def - lkllx min /lkllx exch def - } ifelse - } if - lspad 0 space lss widthshow - end -} def - -% -- -/linkend { - userdict begin - [ lktype /URI eq { - /Action 2 dict dup /Subtype /URI put dup /URI lktarget put - } { - /Dest lktarget - } ifelse - /Border [0 0 0] - /Rect [ lkllx lkxmarg sub - lklly lkymarg sub - lkurx lkxmarg add - lkury lkymarg add ] - /Subtype /Link - /ANN pdfmark - /lkbegun 0 def - colorlinks { 0 setgray } if - end -} def - -% targetname -- -/linkdest { - [ /Dest 3 -1 roll - /View [ /XYZ currentpoint null ] - /DEST pdfmark -} def - -% A "fontset" is an array of fonts; a "stream" is an array of strings -% and numbers or procedures: -% [ 0 (Foo) ( ) (mani) ( ) 1 (padme) 0 ( ) (hum.) ] -% A number choses a font from the current fontset. -% A procedure is invoked as-is when printing the stream. -% -% When printing justified, an equal amount of space is added in -% between each string. - -% string -- spacecount -% Count space characters in a string -/spacecount { - 0 exch { - space eq { 1 add } if - } forall -} def - -% stream fontset -- spacecount width -% Get the width of a stream in the given fontset, and the -% number of space characters in the stream -/streamwidth { - gsave - 6 dict begin - /f exch def - /w 0 def - /s 0 def - f 0 get setfont - /integertype { - f exch get setfont - } def - /stringtype { - dup stringwidth pop w add /w exch def - spacecount s add /s exch def - } def - /arraytype { pop } def - % The input stream is on the top of the stack now - { - dup type exec - } forall - s w - end - grestore -} def - -% stream fontset spacer -- -% Show the stream in the given fontset, but add a certain amount -% of space to each space character -/showstreamspc { - 5 dict begin - /spc exch def - /f exch def - f 0 get setfont - /integertype { - f exch get setfont - } def - /stringtype { - spc linkshow - } def - /arraytype { - exec - } def - % Now stream is on the top of the stack - { - dup type exec - } forall - end -} def - -% stream fontset -- -% Show the stream in the given fontset, with no extra spacing -/showstream { - 0 showstreamspc -} def - -% stream fontset totalspace -- -% Show the stream justified to fit into a certain number of pixels -/showstreamjust { - userdict begin - /ts exch def /fs exch def /st exch def - st fs - st fs streamwidth ts exch sub exch - dup 0 gt { div } { pop } ifelse - showstreamspc - end -} def - -/bullmarg lmarg bulladj add def -/lwidth pagewidth lmarg sub rmarg sub def -/bwidth lwidth bulladj sub def - -% -% The various paragraph types -% The number at the end indicates start (1) of para, end (2) of para -% -/chapline { - currentpoint exch pop 10 sub lmarg exch moveto - 0 setlinecap 3 setlinewidth - lwidth 0 rlineto stroke -} def - -/chap0 { lmarg exch moveto cfont lwidth showstreamjust } def -/chap1 { lmarg exch moveto cfont lwidth showstreamjust } def -/chap2 { lmarg exch moveto cfont showstream chapline } def -/chap3 { lmarg exch moveto cfont showstream chapline } def - -/appn0 {chap0} def -/appn1 {chap1} def -/appn2 {chap2} def -/appn3 {chap3} def - -% lbl ypos fontset -- ypos -/headlbl { - 3 -1 roll [exch ( )] exch % ypos strm fontset - 2 copy % ypos strm fontset strm fontset - streamwidth % ypos strm fontset spccount width - lmarg exch sub % ypos strm fontset spccount xpos - 4 index % ypos strm fontset spccount xpos ypos - moveto % ypos strm fontset spccount - pop % ypos strm fontset spccount - showstream % ypos -} def - -/head0 { lmarg exch moveto hfont lwidth showstreamjust } def -/head1 { hfont headlbl lmarg exch moveto hfont lwidth showstreamjust } def -/head2 { lmarg exch moveto hfont showstream } def -/head3 { hfont headlbl lmarg exch moveto hfont showstream } def - -/subh0 { lmarg exch moveto sfont lwidth showstreamjust } def -/subh1 { sfont headlbl lmarg exch moveto sfont lwidth showstreamjust } def -/subh2 { lmarg exch moveto sfont showstream } def -/subh3 { sfont headlbl lmarg exch moveto sfont showstream } def - -/norm0 { lmarg exch moveto bfont lwidth showstreamjust } def -/norm1 { lmarg exch moveto bfont lwidth showstreamjust } def -/norm2 { lmarg exch moveto bfont showstream } def -/norm3 { lmarg exch moveto bfont showstream } def - -/code0 { lmarg exch moveto bfont showstream } def -/code1 { lmarg exch moveto bfont showstream } def -/code2 { lmarg exch moveto bfont showstream } def -/code3 { lmarg exch moveto bfont showstream } def - -/bull0 { bullmarg exch moveto bfont bwidth showstreamjust } def -/bull1 { dup lmarg exch moveto bullet bfont showstream - bullmarg exch moveto bfont bwidth showstreamjust } def -/bull2 { bullmarg exch moveto bfont showstream } def -/bull3 { dup lmarg exch moveto bullet bfont showstream - bullmarg exch moveto bfont showstream } def - -/tocw0 lwidth tocpnz sub def -/tocw1 tocw0 tocind sub def -/tocw2 tocw1 tocind sub def - -/tocx0 lmarg def -/tocx1 tocx0 tocind add def -/tocx2 tocx1 tocind add def - -/tocpn { - bfont0 setfont - 3 dict begin - /s exch def - /x s stringwidth pop pagewidth rmarg sub exch sub def - currentpoint /y exch def - lmarg sub tocdots div ceiling tocdots mul lmarg add - tocdots x { - y moveto (.) 0 linkshow - } for - x y moveto s 0 linkshow - end - linkend -} def - -/toc00 { tocx0 exch moveto 0 rmoveto bfont showstream } def -/toc01 { tocx0 exch moveto - linkbegindest bfont0 setfont 0 linkshow bfont showstream } def -/toc02 { tocx0 exch moveto 3 1 roll - 0 rmoveto bfont showstream tocpn } def -/toc03 { tocx0 exch moveto 4 1 roll - linkbegindest bfont0 setfont 0 linkshow bfont showstream tocpn } def - -/toc10 { tocx1 exch moveto 0 rmoveto bfont showstream } def -/toc11 { tocx1 exch moveto - linkbegindest bfont0 setfont 0 linkshow bfont showstream } def -/toc12 { tocx1 exch moveto 3 1 roll - 0 rmoveto bfont showstream tocpn } def -/toc13 { tocx1 exch moveto 4 1 roll - linkbegindest bfont0 setfont 0 linkshow bfont showstream tocpn } def - -/toc20 { tocx2 exch moveto 0 rmoveto bfont showstream } def -/toc21 { tocx2 exch moveto - linkbegindest bfont0 setfont 0 linkshow bfont showstream } def -/toc22 { tocx2 exch moveto 3 1 roll - 0 rmoveto bfont showstream tocpn } def -/toc23 { tocx2 exch moveto 4 1 roll - linkbegindest bfont0 setfont 0 linkshow bfont showstream tocpn } def - -% Spacing between index columns -/indexcolumn pagewidth lmarg sub rmarg sub idxgutter add idxcolumns div def -% Width of an individual index column -/indexcolwid indexcolumn idxgutter sub def - -/idx03 { - 2 dict begin - indexcolumn mul lmarg add - /x exch def /y exch def x y moveto - exch bfont showstream - dup bfont streamwidth - x indexcolwid add exch sub exch pop y moveto - bfont showstream - end -} def -/idx00 {idx03} def -/idx01 {idx03} def -/idx02 {idx03} def - -/idx13 { - 2 dict begin - indexcolumn mul lmarg add idxindent add - /x exch def /y exch def x y moveto - exch bfont showstream - dup bfont streamwidth - x indexcolwid idxindent sub add exch sub exch pop y moveto - bfont showstream - end -} def -/idx10 {idx13} def -/idx11 {idx13} def -/idx12 {idx13} def - -% -% Page numbers -% -/pagey botmarg pymarg sub def -/pagel lmarg plmarg sub def -/pager pagewidth rmarg sub prmarg add def - -/pageeven { pagel pagey moveto bfont1 setfont show } def -/pageodd { bfont1 setfont dup stringwidth pop pager exch sub - pagey moveto show } def - -% -% Functions invoked during parsing -% -/xa { linkdest } def -/pa { 0 pageheight moveto linkdest } def -/xl { linkbegindest } def -/wl { linkbeginuri } def -/pl { linkbeginpage } def -/el { linkend } def - -% -% PDF viewer options -% -[/PageMode /UseOutlines /DOCVIEW pdfmark % Display bookmarks - -% -% Functions to include EPS -% -/BeginEPSF { - /Before_EPSF_State save def - /dict_count countdictstack def - /op_count count 1 sub def - userdict begin - /showpage {} def - 0 setgray 0 setlinecap - 1 setlinewidth 0 setlinejoin - 10 setmiterlimit [ ] 0 setdash newpath - /languagelevel where - { - pop languagelevel - 1 ne { - false setstrokeadjust false setoverprint - } if - } if -} bind def -/EndEPSF { - count op_count sub {pop} repeat - countdictstack dict_count sub {end} repeat - Before_EPSF_State restore -} bind def -%%EndProlog -%%BeginSetup -[/Title (Title) -/Dest /title /OUT pdfmark -[/Title (Contents) -/Dest /contents /OUT pdfmark -[/Title (Introduction) -/Count -3 /Dest /chapter-1 /OUT pdfmark -[/Title (What Is NASM?) -/Count -2 /Dest /section-1.1 /OUT pdfmark -[/Title (Why Yet Another Assembler?) -/Dest /section-1.1.1 /OUT pdfmark -[/Title (License Conditions) -/Dest /section-1.1.2 /OUT pdfmark -[/Title (Contact Information) -/Dest /section-1.2 /OUT pdfmark -[/Title (Installation) -/Count -2 /Dest /section-1.3 /OUT pdfmark -[/Title (Installing NASM under MS-DOS or Windows) -/Dest /section-1.3.1 /OUT pdfmark -[/Title (Installing NASM under Unix) -/Dest /section-1.3.2 /OUT pdfmark -[/Title (Running NASM) -/Count -2 /Dest /chapter-2 /OUT pdfmark -[/Title (NASM Command-Line Syntax) -/Count -28 /Dest /section-2.1 /OUT pdfmark -[/Title (The -o Option: Specifying the Output File Name) -/Dest /section-2.1.1 /OUT pdfmark -[/Title (The -f Option: Specifying the Output File Format) -/Dest /section-2.1.2 /OUT pdfmark -[/Title (The -l Option: Generating a Listing File) -/Dest /section-2.1.3 /OUT pdfmark -[/Title (The -M Option: Generate Makefile Dependencies) -/Dest /section-2.1.4 /OUT pdfmark -[/Title (The -MG Option: Generate Makefile Dependencies) -/Dest /section-2.1.5 /OUT pdfmark -[/Title (The -MF Option: Set Makefile Dependency File) -/Dest /section-2.1.6 /OUT pdfmark -[/Title (The -MD Option: Assemble and Generate Dependencies) -/Dest /section-2.1.7 /OUT pdfmark -[/Title (The -MT Option: Dependency Target Name) -/Dest /section-2.1.8 /OUT pdfmark -[/Title (The -MQ Option: Dependency Target Name \(Quoted\)) -/Dest /section-2.1.9 /OUT pdfmark -[/Title (The -MP Option: Emit phony targets) -/Dest /section-2.1.10 /OUT pdfmark -[/Title (The -F Option: Selecting a Debug Information Format) -/Dest /section-2.1.11 /OUT pdfmark -[/Title (The -g Option: Enabling Debug Information.) -/Dest /section-2.1.12 /OUT pdfmark -[/Title (The -X Option: Selecting an Error Reporting Format) -/Dest /section-2.1.13 /OUT pdfmark -[/Title (The -Z Option: Send Errors to a File) -/Dest /section-2.1.14 /OUT pdfmark -[/Title (The -s Option: Send Errors to stdout) -/Dest /section-2.1.15 /OUT pdfmark -[/Title (The -i Option: Include File Search Directories) -/Dest /section-2.1.16 /OUT pdfmark -[/Title (The -p Option: Pre-Include a File) -/Dest /section-2.1.17 /OUT pdfmark -[/Title (The -d Option: Pre-Define a Macro) -/Dest /section-2.1.18 /OUT pdfmark -[/Title (The -u Option: Undefine a Macro) -/Dest /section-2.1.19 /OUT pdfmark -[/Title (The -E Option: Preprocess Only) -/Dest /section-2.1.20 /OUT pdfmark -[/Title (The -a Option: Don't Preprocess At All) -/Dest /section-2.1.21 /OUT pdfmark -[/Title (The -O Option: Specifying Multipass Optimization) -/Dest /section-2.1.22 /OUT pdfmark -[/Title (The -t Option: Enable TASM Compatibility Mode) -/Dest /section-2.1.23 /OUT pdfmark -[/Title (The -w and -W Options: Enable or Disable Assembly Warnings) -/Dest /section-2.1.24 /OUT pdfmark -[/Title (The -v Option: Display Version Info) -/Dest /section-2.1.25 /OUT pdfmark -[/Title (The -y Option: Display Available Debug Info Formats) -/Dest /section-2.1.26 /OUT pdfmark -[/Title (The --prefix and --postfix Options.) -/Dest /section-2.1.27 /OUT pdfmark -[/Title (The NASMENV Environment Variable) -/Dest /section-2.1.28 /OUT pdfmark -[/Title (Quick Start for MASM Users) -/Count -7 /Dest /section-2.2 /OUT pdfmark -[/Title (NASM Is Case-Sensitive) -/Dest /section-2.2.1 /OUT pdfmark -[/Title (NASM Requires Square Brackets For Memory References) -/Dest /section-2.2.2 /OUT pdfmark -[/Title (NASM Doesn't Store Variable Types) -/Dest /section-2.2.3 /OUT pdfmark -[/Title (NASM Doesn't ASSUME) -/Dest /section-2.2.4 /OUT pdfmark -[/Title (NASM Doesn't Support Memory Models) -/Dest /section-2.2.5 /OUT pdfmark -[/Title (Floating-Point Differences) -/Dest /section-2.2.6 /OUT pdfmark -[/Title (Other Differences) -/Dest /section-2.2.7 /OUT pdfmark -[/Title (The NASM Language) -/Count -9 /Dest /chapter-3 /OUT pdfmark -[/Title (Layout of a NASM Source Line) -/Dest /section-3.1 /OUT pdfmark -[/Title (Pseudo-Instructions) -/Count -5 /Dest /section-3.2 /OUT pdfmark -[/Title (DB and Friends: Declaring Initialized Data) -/Dest /section-3.2.1 /OUT pdfmark -[/Title (RESB and Friends: Declaring Uninitialized Data) -/Dest /section-3.2.2 /OUT pdfmark -[/Title (INCBIN: Including External Binary Files) -/Dest /section-3.2.3 /OUT pdfmark -[/Title (EQU: Defining Constants) -/Dest /section-3.2.4 /OUT pdfmark -[/Title (TIMES: Repeating Instructions or Data) -/Dest /section-3.2.5 /OUT pdfmark -[/Title (Effective Addresses) -/Dest /section-3.3 /OUT pdfmark -[/Title (Constants) -/Count -7 /Dest /section-3.4 /OUT pdfmark -[/Title (Numeric Constants) -/Dest /section-3.4.1 /OUT pdfmark -[/Title (Character Strings) -/Dest /section-3.4.2 /OUT pdfmark -[/Title (Character Constants) -/Dest /section-3.4.3 /OUT pdfmark -[/Title (String Constants) -/Dest /section-3.4.4 /OUT pdfmark -[/Title (Unicode Strings) -/Dest /section-3.4.5 /OUT pdfmark -[/Title (Floating-Point Constants) -/Dest /section-3.4.6 /OUT pdfmark -[/Title (Packed BCD Constants) -/Dest /section-3.4.7 /OUT pdfmark -[/Title (Expressions) -/Count -7 /Dest /section-3.5 /OUT pdfmark -[/Title (|: Bitwise OR Operator) -/Dest /section-3.5.1 /OUT pdfmark -[/Title (^: Bitwise XOR Operator) -/Dest /section-3.5.2 /OUT pdfmark -[/Title (&: Bitwise AND Operator) -/Dest /section-3.5.3 /OUT pdfmark -[/Title (<< and >>: Bit Shift Operators) -/Dest /section-3.5.4 /OUT pdfmark -[/Title (+ and -: Addition and Subtraction Operators) -/Dest /section-3.5.5 /OUT pdfmark -[/Title (*, /, //, % and %%: Multiplication and Division) -/Dest /section-3.5.6 /OUT pdfmark -[/Title (Unary Operators: +, -, ~, ! and SEG) -/Dest /section-3.5.7 /OUT pdfmark -[/Title (SEG and WRT) -/Dest /section-3.6 /OUT pdfmark -[/Title (STRICT: Inhibiting Optimization) -/Dest /section-3.7 /OUT pdfmark -[/Title (Critical Expressions) -/Dest /section-3.8 /OUT pdfmark -[/Title (Local Labels) -/Dest /section-3.9 /OUT pdfmark -[/Title (The NASM Preprocessor) -/Count -11 /Dest /chapter-4 /OUT pdfmark -[/Title (Single-Line Macros) -/Count -9 /Dest /section-4.1 /OUT pdfmark -[/Title (The Normal Way: %define) -/Dest /section-4.1.1 /OUT pdfmark -[/Title (Resolving %define: %xdefine) -/Dest /section-4.1.2 /OUT pdfmark -[/Title (Macro Indirection: %[...]) -/Dest /section-4.1.3 /OUT pdfmark -[/Title (Concatenating Single Line Macro Tokens: %+) -/Dest /section-4.1.4 /OUT pdfmark -[/Title (The Macro Name Itself: %? and %??) -/Dest /section-4.1.5 /OUT pdfmark -[/Title (Undefining Single-Line Macros: %undef) -/Dest /section-4.1.6 /OUT pdfmark -[/Title (Preprocessor Variables: %assign) -/Dest /section-4.1.7 /OUT pdfmark -[/Title (Defining Strings: %defstr) -/Dest /section-4.1.8 /OUT pdfmark -[/Title (Defining Tokens: %deftok) -/Dest /section-4.1.9 /OUT pdfmark -[/Title (String Manipulation in Macros) -/Count -3 /Dest /section-4.2 /OUT pdfmark -[/Title (Concatenating Strings: %strcat) -/Dest /section-4.2.1 /OUT pdfmark -[/Title (String Length: %strlen) -/Dest /section-4.2.2 /OUT pdfmark -[/Title (Extracting Substrings: %substr) -/Dest /section-4.2.3 /OUT pdfmark -[/Title (Multi-Line Macros: %macro) -/Count -12 /Dest /section-4.3 /OUT pdfmark -[/Title (Recursive Multi-Line Macros: %rmacro) -/Dest /section-4.3.1 /OUT pdfmark -[/Title (Overloading Multi-Line Macros) -/Dest /section-4.3.2 /OUT pdfmark -[/Title (Macro-Local Labels) -/Dest /section-4.3.3 /OUT pdfmark -[/Title (Greedy Macro Parameters) -/Dest /section-4.3.4 /OUT pdfmark -[/Title (Default Macro Parameters) -/Dest /section-4.3.5 /OUT pdfmark -[/Title (%0: Macro Parameter Counter) -/Dest /section-4.3.6 /OUT pdfmark -[/Title (%rotate: Rotating Macro Parameters) -/Dest /section-4.3.7 /OUT pdfmark -[/Title (Concatenating Macro Parameters) -/Dest /section-4.3.8 /OUT pdfmark -[/Title (Condition Codes as Macro Parameters) -/Dest /section-4.3.9 /OUT pdfmark -[/Title (Disabling Listing Expansion) -/Dest /section-4.3.10 /OUT pdfmark -[/Title (Undefining Multi-Line Macros: %unmacro) -/Dest /section-4.3.11 /OUT pdfmark -[/Title (Exiting Multi-Line Macros: %exitmacro) -/Dest /section-4.3.12 /OUT pdfmark -[/Title (Conditional Assembly) -/Count -8 /Dest /section-4.4 /OUT pdfmark -[/Title (%ifdef: Testing Single-Line Macro Existence) -/Dest /section-4.4.1 /OUT pdfmark -[/Title (%ifmacro: Testing Multi-Line Macro Existence) -/Dest /section-4.4.2 /OUT pdfmark -[/Title (%ifctx: Testing the Context Stack) -/Dest /section-4.4.3 /OUT pdfmark -[/Title (%if: Testing Arbitrary Numeric Expressions) -/Dest /section-4.4.4 /OUT pdfmark -[/Title (%ifidn and %ifidni: Testing Exact Text Identity) -/Dest /section-4.4.5 /OUT pdfmark -[/Title (%ifid, %ifnum, %ifstr: Testing Token Types) -/Dest /section-4.4.6 /OUT pdfmark -[/Title (%iftoken: Test for a Single Token) -/Dest /section-4.4.7 /OUT pdfmark -[/Title (%ifempty: Test for Empty Expansion) -/Dest /section-4.4.8 /OUT pdfmark -[/Title (Preprocessor Loops: %rep) -/Dest /section-4.5 /OUT pdfmark -[/Title (Source Files and Dependencies) -/Count -4 /Dest /section-4.6 /OUT pdfmark -[/Title (%include: Including Other Files) -/Dest /section-4.6.1 /OUT pdfmark -[/Title (%pathsearch: Search the Include Path) -/Dest /section-4.6.2 /OUT pdfmark -[/Title (%depend: Add Dependent Files) -/Dest /section-4.6.3 /OUT pdfmark -[/Title (%use: Include Standard Macro Package) -/Dest /section-4.6.4 /OUT pdfmark -[/Title (The Context Stack) -/Count -5 /Dest /section-4.7 /OUT pdfmark -[/Title (%push and %pop: Creating and Removing Contexts) -/Dest /section-4.7.1 /OUT pdfmark -[/Title (Context-Local Labels) -/Dest /section-4.7.2 /OUT pdfmark -[/Title (Context-Local Single-Line Macros) -/Dest /section-4.7.3 /OUT pdfmark -[/Title (%repl: Renaming a Context) -/Dest /section-4.7.4 /OUT pdfmark -[/Title (Example Use of the Context Stack: Block IFs) -/Dest /section-4.7.5 /OUT pdfmark -[/Title (Stack Relative Preprocessor Directives) -/Count -3 /Dest /section-4.8 /OUT pdfmark -[/Title (%arg Directive) -/Dest /section-4.8.1 /OUT pdfmark -[/Title (%stacksize Directive) -/Dest /section-4.8.2 /OUT pdfmark -[/Title (%local Directive) -/Dest /section-4.8.3 /OUT pdfmark -[/Title (Reporting User-Defined Errors: %error, %warning, %fatal) -/Dest /section-4.9 /OUT pdfmark -[/Title (Other Preprocessor Directives) -/Count -2 /Dest /section-4.10 /OUT pdfmark -[/Title (%line Directive) -/Dest /section-4.10.1 /OUT pdfmark -[/Title (%!<env>: Read an environment variable.) -/Dest /section-4.10.2 /OUT pdfmark -[/Title (Standard Macros) -/Count -12 /Dest /section-4.11 /OUT pdfmark -[/Title (NASM Version Macros) -/Dest /section-4.11.1 /OUT pdfmark -[/Title (__NASM_VERSION_ID__: NASM Version ID) -/Dest /section-4.11.2 /OUT pdfmark -[/Title (__NASM_VER__: NASM Version string) -/Dest /section-4.11.3 /OUT pdfmark -[/Title (__FILE__ and __LINE__: File Name and Line Number) -/Dest /section-4.11.4 /OUT pdfmark -[/Title (__BITS__: Current BITS Mode) -/Dest /section-4.11.5 /OUT pdfmark -[/Title (__OUTPUT_FORMAT__: Current Output Format) -/Dest /section-4.11.6 /OUT pdfmark -[/Title (Assembly Date and Time Macros) -/Dest /section-4.11.7 /OUT pdfmark -[/Title (__USE_package__: Package Include Test) -/Dest /section-4.11.8 /OUT pdfmark -[/Title (__PASS__: Assembly Pass) -/Dest /section-4.11.9 /OUT pdfmark -[/Title (STRUC and ENDSTRUC: Declaring Structure Data Types) -/Dest /section-4.11.10 /OUT pdfmark -[/Title (ISTRUC, AT and IEND: Declaring Instances of Structures) -/Dest /section-4.11.11 /OUT pdfmark -[/Title (ALIGN and ALIGNB: Data Alignment) -/Dest /section-4.11.12 /OUT pdfmark -[/Title (Standard Macro Packages) -/Count -2 /Dest /chapter-5 /OUT pdfmark -[/Title (altreg: Alternate Register Names) -/Dest /section-5.1 /OUT pdfmark -[/Title (smartalign: Smart ALIGN Macro) -/Dest /section-5.2 /OUT pdfmark -[/Title (Assembler Directives) -/Count -9 /Dest /chapter-6 /OUT pdfmark -[/Title (BITS: Specifying Target Processor Mode) -/Count -1 /Dest /section-6.1 /OUT pdfmark -[/Title (USE16 & USE32: Aliases for BITS) -/Dest /section-6.1.1 /OUT pdfmark -[/Title (DEFAULT: Change the assembler defaults) -/Dest /section-6.2 /OUT pdfmark -[/Title (SECTION or SEGMENT: Changing and Defining Sections) -/Count -1 /Dest /section-6.3 /OUT pdfmark -[/Title (The __SECT__ Macro) -/Dest /section-6.3.1 /OUT pdfmark -[/Title (ABSOLUTE: Defining Absolute Labels) -/Dest /section-6.4 /OUT pdfmark -[/Title (EXTERN: Importing Symbols from Other Modules) -/Dest /section-6.5 /OUT pdfmark -[/Title (GLOBAL: Exporting Symbols to Other Modules) -/Dest /section-6.6 /OUT pdfmark -[/Title (COMMON: Defining Common Data Areas) -/Dest /section-6.7 /OUT pdfmark -[/Title (CPU: Defining CPU Dependencies) -/Dest /section-6.8 /OUT pdfmark -[/Title (FLOAT: Handling of floating-point constants) -/Dest /section-6.9 /OUT pdfmark -[/Title (Output Formats) -/Count -14 /Dest /chapter-7 /OUT pdfmark -[/Title (bin: Flat-Form Binary Output) -/Count -4 /Dest /section-7.1 /OUT pdfmark -[/Title (ORG: Binary File Program Origin) -/Dest /section-7.1.1 /OUT pdfmark -[/Title (bin Extensions to the SECTION Directive) -/Dest /section-7.1.2 /OUT pdfmark -[/Title (Multisection Support for the bin Format) -/Dest /section-7.1.3 /OUT pdfmark -[/Title (Map Files) -/Dest /section-7.1.4 /OUT pdfmark -[/Title (ith: Intel Hex Output) -/Dest /section-7.2 /OUT pdfmark -[/Title (srec: Motorola S-Records Output) -/Dest /section-7.3 /OUT pdfmark -[/Title (obj: Microsoft OMF Object Files) -/Count -8 /Dest /section-7.4 /OUT pdfmark -[/Title (obj Extensions to the SEGMENT Directive) -/Dest /section-7.4.1 /OUT pdfmark -[/Title (GROUP: Defining Groups of Segments) -/Dest /section-7.4.2 /OUT pdfmark -[/Title (UPPERCASE: Disabling Case Sensitivity in Output) -/Dest /section-7.4.3 /OUT pdfmark -[/Title (IMPORT: Importing DLL Symbols) -/Dest /section-7.4.4 /OUT pdfmark -[/Title (EXPORT: Exporting DLL Symbols) -/Dest /section-7.4.5 /OUT pdfmark -[/Title (..start: Defining the Program Entry Point) -/Dest /section-7.4.6 /OUT pdfmark -[/Title (obj Extensions to the EXTERN Directive) -/Dest /section-7.4.7 /OUT pdfmark -[/Title (obj Extensions to the COMMON Directive) -/Dest /section-7.4.8 /OUT pdfmark -[/Title (win32: Microsoft Win32 Object Files) -/Count -2 /Dest /section-7.5 /OUT pdfmark -[/Title (win32 Extensions to the SECTION Directive) -/Dest /section-7.5.1 /OUT pdfmark -[/Title (win32: Safe Structured Exception Handling) -/Dest /section-7.5.2 /OUT pdfmark -[/Title (win64: Microsoft Win64 Object Files) -/Count -2 /Dest /section-7.6 /OUT pdfmark -[/Title (win64: Writing Position-Independent Code) -/Dest /section-7.6.1 /OUT pdfmark -[/Title (win64: Structured Exception Handling) -/Dest /section-7.6.2 /OUT pdfmark -[/Title (coff: Common Object File Format) -/Dest /section-7.7 /OUT pdfmark -[/Title (macho32 and macho64: Mach Object File Format) -/Dest /section-7.8 /OUT pdfmark -[/Title (elf32 and elf64: Executable and Linkable Format Object Files) -/Count -8 /Dest /section-7.9 /OUT pdfmark -[/Title (ELF specific directive osabi) -/Dest /section-7.9.1 /OUT pdfmark -[/Title (elf Extensions to the SECTION Directive) -/Dest /section-7.9.2 /OUT pdfmark -[/Title (Position-Independent Code: elf Special Symbols and WRT) -/Dest /section-7.9.3 /OUT pdfmark -[/Title (Thread Local Storage: elf Special Symbols and WRT) -/Dest /section-7.9.4 /OUT pdfmark -[/Title (elf Extensions to the GLOBAL Directive) -/Dest /section-7.9.5 /OUT pdfmark -[/Title (elf Extensions to the COMMON Directive ) -/Dest /section-7.9.6 /OUT pdfmark -[/Title (16-bit code and ELF ) -/Dest /section-7.9.7 /OUT pdfmark -[/Title (Debug formats and ELF ) -/Dest /section-7.9.8 /OUT pdfmark -[/Title (aout: Linux a.out Object Files) -/Dest /section-7.10 /OUT pdfmark -[/Title (aoutb: NetBSD/FreeBSD/OpenBSD a.out Object Files) -/Dest /section-7.11 /OUT pdfmark -[/Title (as86: Minix/Linux as86 Object Files) -/Dest /section-7.12 /OUT pdfmark -[/Title (rdf: Relocatable Dynamic Object File Format) -/Count -4 /Dest /section-7.13 /OUT pdfmark -[/Title (Requiring a Library: The LIBRARY Directive) -/Dest /section-7.13.1 /OUT pdfmark -[/Title (Specifying a Module Name: The MODULE Directive) -/Dest /section-7.13.2 /OUT pdfmark -[/Title (rdf Extensions to the GLOBAL Directive) -/Dest /section-7.13.3 /OUT pdfmark -[/Title (rdf Extensions to the EXTERN Directive) -/Dest /section-7.13.4 /OUT pdfmark -[/Title (dbg: Debugging Format) -/Dest /section-7.14 /OUT pdfmark -[/Title (Writing 16-bit Code \(DOS, Windows 3/3.1\)) -/Count -5 /Dest /chapter-8 /OUT pdfmark -[/Title (Producing .EXE Files) -/Count -2 /Dest /section-8.1 /OUT pdfmark -[/Title (Using the obj Format To Generate .EXE Files) -/Dest /section-8.1.1 /OUT pdfmark -[/Title (Using the bin Format To Generate .EXE Files) -/Dest /section-8.1.2 /OUT pdfmark -[/Title (Producing .COM Files) -/Count -2 /Dest /section-8.2 /OUT pdfmark -[/Title (Using the bin Format To Generate .COM Files) -/Dest /section-8.2.1 /OUT pdfmark -[/Title (Using the obj Format To Generate .COM Files) -/Dest /section-8.2.2 /OUT pdfmark -[/Title (Producing .SYS Files) -/Dest /section-8.3 /OUT pdfmark -[/Title (Interfacing to 16-bit C Programs) -/Count -5 /Dest /section-8.4 /OUT pdfmark -[/Title (External Symbol Names) -/Dest /section-8.4.1 /OUT pdfmark -[/Title (Memory Models) -/Dest /section-8.4.2 /OUT pdfmark -[/Title (Function Definitions and Function Calls) -/Dest /section-8.4.3 /OUT pdfmark -[/Title (Accessing Data Items) -/Dest /section-8.4.4 /OUT pdfmark -[/Title (c16.mac: Helper Macros for the 16-bit C Interface) -/Dest /section-8.4.5 /OUT pdfmark -[/Title (Interfacing to Borland Pascal Programs) -/Count -3 /Dest /section-8.5 /OUT pdfmark -[/Title (The Pascal Calling Convention) -/Dest /section-8.5.1 /OUT pdfmark -[/Title (Borland Pascal Segment Name Restrictions) -/Dest /section-8.5.2 /OUT pdfmark -[/Title (Using c16.mac With Pascal Programs) -/Dest /section-8.5.3 /OUT pdfmark -[/Title (Writing 32-bit Code \(Unix, Win32, DJGPP\)) -/Count -2 /Dest /chapter-9 /OUT pdfmark -[/Title (Interfacing to 32-bit C Programs) -/Count -4 /Dest /section-9.1 /OUT pdfmark -[/Title (External Symbol Names) -/Dest /section-9.1.1 /OUT pdfmark -[/Title (Function Definitions and Function Calls) -/Dest /section-9.1.2 /OUT pdfmark -[/Title (Accessing Data Items) -/Dest /section-9.1.3 /OUT pdfmark -[/Title (c32.mac: Helper Macros for the 32-bit C Interface) -/Dest /section-9.1.4 /OUT pdfmark -[/Title (Writing NetBSD/FreeBSD/OpenBSD and Linux/ELF Shared Libraries) -/Count -6 /Dest /section-9.2 /OUT pdfmark -[/Title (Obtaining the Address of the GOT) -/Dest /section-9.2.1 /OUT pdfmark -[/Title (Finding Your Local Data Items) -/Dest /section-9.2.2 /OUT pdfmark -[/Title (Finding External and Common Data Items) -/Dest /section-9.2.3 /OUT pdfmark -[/Title (Exporting Symbols to the Library User) -/Dest /section-9.2.4 /OUT pdfmark -[/Title (Calling Procedures Outside the Library) -/Dest /section-9.2.5 /OUT pdfmark -[/Title (Generating the Library File) -/Dest /section-9.2.6 /OUT pdfmark -[/Title (Mixing 16 and 32 Bit Code) -/Count -3 /Dest /chapter-10 /OUT pdfmark -[/Title (Mixed-Size Jumps) -/Dest /section-10.1 /OUT pdfmark -[/Title (Addressing Between Different-Size Segments) -/Dest /section-10.2 /OUT pdfmark -[/Title (Other Mixed-Size Instructions) -/Dest /section-10.3 /OUT pdfmark -[/Title (Writing 64-bit Code \(Unix, Win64\)) -/Count -4 /Dest /chapter-11 /OUT pdfmark -[/Title (Register Names in 64-bit Mode) -/Dest /section-11.1 /OUT pdfmark -[/Title (Immediates and Displacements in 64-bit Mode) -/Dest /section-11.2 /OUT pdfmark -[/Title (Interfacing to 64-bit C Programs \(Unix\)) -/Dest /section-11.3 /OUT pdfmark -[/Title (Interfacing to 64-bit C Programs \(Win64\)) -/Dest /section-11.4 /OUT pdfmark -[/Title (Troubleshooting) -/Count -2 /Dest /chapter-12 /OUT pdfmark -[/Title (Common Problems) -/Count -4 /Dest /section-12.1 /OUT pdfmark -[/Title (NASM Generates Inefficient Code) -/Dest /section-12.1.1 /OUT pdfmark -[/Title (My Jumps are Out of Range) -/Dest /section-12.1.2 /OUT pdfmark -[/Title (ORG Doesn't Work) -/Dest /section-12.1.3 /OUT pdfmark -[/Title (TIMES Doesn't Work) -/Dest /section-12.1.4 /OUT pdfmark -[/Title (Bugs) -/Dest /section-12.2 /OUT pdfmark -[/Title (Ndisasm) -/Count -4 /Dest /appendix-A /OUT pdfmark -[/Title (Introduction) -/Dest /section-A.1 /OUT pdfmark -[/Title (Getting Started: Installation) -/Dest /section-A.2 /OUT pdfmark -[/Title (Running NDISASM) -/Count -4 /Dest /section-A.3 /OUT pdfmark -[/Title (COM Files: Specifying an Origin) -/Dest /section-A.3.1 /OUT pdfmark -[/Title (Code Following Data: Synchronisation) -/Dest /section-A.3.2 /OUT pdfmark -[/Title (Mixed Code and Data: Automatic \(Intelligent\) Synchronisation ) -/Dest /section-A.3.3 /OUT pdfmark -[/Title (Other Options) -/Dest /section-A.3.4 /OUT pdfmark -[/Title (Bugs and Improvements) -/Dest /section-A.4 /OUT pdfmark -[/Title (Instruction List) -/Count -1 /Dest /appendix-B /OUT pdfmark -[/Title (Introduction) -/Count -32 /Dest /section-B.1 /OUT pdfmark -[/Title (Special instructions...) -/Dest /section-B.1.1 /OUT pdfmark -[/Title (Conventional instructions) -/Dest /section-B.1.2 /OUT pdfmark -[/Title (Katmai Streaming SIMD instructions \(SSE \226\226 a.k.a. KNI, XMM, MMX2\)) -/Dest /section-B.1.3 /OUT pdfmark -[/Title (Introduced in Deschutes but necessary for SSE support) -/Dest /section-B.1.4 /OUT pdfmark -[/Title (XSAVE group \(AVX and extended state\)) -/Dest /section-B.1.5 /OUT pdfmark -[/Title (Generic memory operations) -/Dest /section-B.1.6 /OUT pdfmark -[/Title (New MMX instructions introduced in Katmai) -/Dest /section-B.1.7 /OUT pdfmark -[/Title (AMD Enhanced 3DNow! \(Athlon\) instructions) -/Dest /section-B.1.8 /OUT pdfmark -[/Title (Willamette SSE2 Cacheability Instructions) -/Dest /section-B.1.9 /OUT pdfmark -[/Title (Willamette MMX instructions \(SSE2 SIMD Integer Instructions\)) -/Dest /section-B.1.10 /OUT pdfmark -[/Title (Willamette Streaming SIMD instructions \(SSE2\)) -/Dest /section-B.1.11 /OUT pdfmark -[/Title (Prescott New Instructions \(SSE3\)) -/Dest /section-B.1.12 /OUT pdfmark -[/Title (VMX Instructions) -/Dest /section-B.1.13 /OUT pdfmark -[/Title (Extended Page Tables VMX instructions) -/Dest /section-B.1.14 /OUT pdfmark -[/Title (Tejas New Instructions \(SSSE3\)) -/Dest /section-B.1.15 /OUT pdfmark -[/Title (AMD SSE4A) -/Dest /section-B.1.16 /OUT pdfmark -[/Title (New instructions in Barcelona) -/Dest /section-B.1.17 /OUT pdfmark -[/Title (Penryn New Instructions \(SSE4.1\)) -/Dest /section-B.1.18 /OUT pdfmark -[/Title (Nehalem New Instructions \(SSE4.2\)) -/Dest /section-B.1.19 /OUT pdfmark -[/Title (Intel SMX) -/Dest /section-B.1.20 /OUT pdfmark -[/Title (Geode \(Cyrix\) 3DNow! additions) -/Dest /section-B.1.21 /OUT pdfmark -[/Title (Intel new instructions in ???) -/Dest /section-B.1.22 /OUT pdfmark -[/Title (Intel AES instructions) -/Dest /section-B.1.23 /OUT pdfmark -[/Title (Intel AVX AES instructions) -/Dest /section-B.1.24 /OUT pdfmark -[/Title (Intel AVX instructions) -/Dest /section-B.1.25 /OUT pdfmark -[/Title (Intel Carry-Less Multiplication instructions \(CLMUL\)) -/Dest /section-B.1.26 /OUT pdfmark -[/Title (Intel AVX Carry-Less Multiplication instructions \(CLMUL\)) -/Dest /section-B.1.27 /OUT pdfmark -[/Title (Intel Fused Multiply-Add instructions \(FMA\)) -/Dest /section-B.1.28 /OUT pdfmark -[/Title (VIA \(Centaur\) security instructions) -/Dest /section-B.1.29 /OUT pdfmark -[/Title (AMD Lightweight Profiling \(LWP\) instructions) -/Dest /section-B.1.30 /OUT pdfmark -[/Title (AMD XOP, FMA4 and CVT16 instructions \(SSE5\)) -/Dest /section-B.1.31 /OUT pdfmark -[/Title (Systematic names for the hinting nop instructions) -/Dest /section-B.1.32 /OUT pdfmark -[/Title (NASM Version History) -/Count -3 /Dest /appendix-C /OUT pdfmark -[/Title (NASM 2 Series) -/Count -11 /Dest /section-C.1 /OUT pdfmark -[/Title (Version 2.08) -/Dest /section-C.1.1 /OUT pdfmark -[/Title (Version 2.07) -/Dest /section-C.1.2 /OUT pdfmark -[/Title (Version 2.06) -/Dest /section-C.1.3 /OUT pdfmark -[/Title (Version 2.05.01) -/Dest /section-C.1.4 /OUT pdfmark -[/Title (Version 2.05) -/Dest /section-C.1.5 /OUT pdfmark -[/Title (Version 2.04) -/Dest /section-C.1.6 /OUT pdfmark -[/Title (Version 2.03.01) -/Dest /section-C.1.7 /OUT pdfmark -[/Title (Version 2.03) -/Dest /section-C.1.8 /OUT pdfmark -[/Title (Version 2.02) -/Dest /section-C.1.9 /OUT pdfmark -[/Title (Version 2.01) -/Dest /section-C.1.10 /OUT pdfmark -[/Title (Version 2.00) -/Dest /section-C.1.11 /OUT pdfmark -[/Title (NASM 0.98 Series) -/Count -55 /Dest /section-C.2 /OUT pdfmark -[/Title (Version 0.98.39) -/Dest /section-C.2.1 /OUT pdfmark -[/Title (Version 0.98.38) -/Dest /section-C.2.2 /OUT pdfmark -[/Title (Version 0.98.37) -/Dest /section-C.2.3 /OUT pdfmark -[/Title (Version 0.98.36) -/Dest /section-C.2.4 /OUT pdfmark -[/Title (Version 0.98.35) -/Dest /section-C.2.5 /OUT pdfmark -[/Title (Version 0.98.34) -/Dest /section-C.2.6 /OUT pdfmark -[/Title (Version 0.98.33) -/Dest /section-C.2.7 /OUT pdfmark -[/Title (Version 0.98.32) -/Dest /section-C.2.8 /OUT pdfmark -[/Title (Version 0.98.31) -/Dest /section-C.2.9 /OUT pdfmark -[/Title (Version 0.98.30) -/Dest /section-C.2.10 /OUT pdfmark -[/Title (Version 0.98.28) -/Dest /section-C.2.11 /OUT pdfmark -[/Title (Version 0.98.26) -/Dest /section-C.2.12 /OUT pdfmark -[/Title (Version 0.98.25alt) -/Dest /section-C.2.13 /OUT pdfmark -[/Title (Version 0.98.25) -/Dest /section-C.2.14 /OUT pdfmark -[/Title (Version 0.98.24p1) -/Dest /section-C.2.15 /OUT pdfmark -[/Title (Version 0.98.24) -/Dest /section-C.2.16 /OUT pdfmark -[/Title (Version 0.98.23) -/Dest /section-C.2.17 /OUT pdfmark -[/Title (Version 0.98.22) -/Dest /section-C.2.18 /OUT pdfmark -[/Title (Version 0.98.21) -/Dest /section-C.2.19 /OUT pdfmark -[/Title (Version 0.98.20) -/Dest /section-C.2.20 /OUT pdfmark -[/Title (Version 0.98.19) -/Dest /section-C.2.21 /OUT pdfmark -[/Title (Version 0.98.18) -/Dest /section-C.2.22 /OUT pdfmark -[/Title (Version 0.98.17) -/Dest /section-C.2.23 /OUT pdfmark -[/Title (Version 0.98.16) -/Dest /section-C.2.24 /OUT pdfmark -[/Title (Version 0.98.15) -/Dest /section-C.2.25 /OUT pdfmark -[/Title (Version 0.98.14) -/Dest /section-C.2.26 /OUT pdfmark -[/Title (Version 0.98.13) -/Dest /section-C.2.27 /OUT pdfmark -[/Title (Version 0.98.12) -/Dest /section-C.2.28 /OUT pdfmark -[/Title (Version 0.98.11) -/Dest /section-C.2.29 /OUT pdfmark -[/Title (Version 0.98.10) -/Dest /section-C.2.30 /OUT pdfmark -[/Title (Version 0.98.09) -/Dest /section-C.2.31 /OUT pdfmark -[/Title (Version 0.98.08) -/Dest /section-C.2.32 /OUT pdfmark -[/Title (Version 0.98.09b with John Coffman patches released 28-Oct-2001) -/Dest /section-C.2.33 /OUT pdfmark -[/Title (Version 0.98.07 released 01/28/01) -/Dest /section-C.2.34 /OUT pdfmark -[/Title (Version 0.98.06f released 01/18/01) -/Dest /section-C.2.35 /OUT pdfmark -[/Title (Version 0.98.06e released 01/09/01) -/Dest /section-C.2.36 /OUT pdfmark -[/Title (Version 0.98p1) -/Dest /section-C.2.37 /OUT pdfmark -[/Title (Version 0.98bf \(bug-fixed\)) -/Dest /section-C.2.38 /OUT pdfmark -[/Title (Version 0.98.03 with John Coffman's changes released 27-Jul-2000) -/Dest /section-C.2.39 /OUT pdfmark -[/Title (Version 0.98.03) -/Dest /section-C.2.40 /OUT pdfmark -[/Title (Version 0.98) -/Dest /section-C.2.41 /OUT pdfmark -[/Title (Version 0.98p9) -/Dest /section-C.2.42 /OUT pdfmark -[/Title (Version 0.98p8) -/Dest /section-C.2.43 /OUT pdfmark -[/Title (Version 0.98p7) -/Dest /section-C.2.44 /OUT pdfmark -[/Title (Version 0.98p6) -/Dest /section-C.2.45 /OUT pdfmark -[/Title (Version 0.98p3.7) -/Dest /section-C.2.46 /OUT pdfmark -[/Title (Version 0.98p3.6) -/Dest /section-C.2.47 /OUT pdfmark -[/Title (Version 0.98p3.5) -/Dest /section-C.2.48 /OUT pdfmark -[/Title (Version 0.98p3.4) -/Dest /section-C.2.49 /OUT pdfmark -[/Title (Version 0.98p3.3) -/Dest /section-C.2.50 /OUT pdfmark -[/Title (Version 0.98p3.2) -/Dest /section-C.2.51 /OUT pdfmark -[/Title (Version 0.98p3-hpa) -/Dest /section-C.2.52 /OUT pdfmark -[/Title (Version 0.98 pre-release 3) -/Dest /section-C.2.53 /OUT pdfmark -[/Title (Version 0.98 pre-release 2) -/Dest /section-C.2.54 /OUT pdfmark -[/Title (Version 0.98 pre-release 1) -/Dest /section-C.2.55 /OUT pdfmark -[/Title (NASM 0.9 Series) -/Count -8 /Dest /section-C.3 /OUT pdfmark -[/Title (Version 0.97 released December 1997) -/Dest /section-C.3.1 /OUT pdfmark -[/Title (Version 0.96 released November 1997) -/Dest /section-C.3.2 /OUT pdfmark -[/Title (Version 0.95 released July 1997) -/Dest /section-C.3.3 /OUT pdfmark -[/Title (Version 0.94 released April 1997) -/Dest /section-C.3.4 /OUT pdfmark -[/Title (Version 0.93 released January 1997) -/Dest /section-C.3.5 /OUT pdfmark -[/Title (Version 0.92 released January 1997) -/Dest /section-C.3.6 /OUT pdfmark -[/Title (Version 0.91 released November 1996) -/Dest /section-C.3.7 /OUT pdfmark -[/Title (Version 0.90 released October 1996) -/Dest /section-C.3.8 /OUT pdfmark -[/Title (Index) -/Dest /index /OUT pdfmark -setpagesize -%%EndSetup -%%Page: 1 1 -%%BeginPageSetup -save -%%EndPageSetup -/1 pa -/ti (NASM \227 The Netwide Assembler) def -/sti (version\2402.08rc7) def -lmarg pageheight 2 mul 3 div moveto -tfont0 setfont -/title linkdest ti show -lmarg pageheight 2 mul 3 div 10 sub moveto -0 setlinecap 3 setlinewidth -pagewidth lmarg sub rmarg sub 0 rlineto currentpoint stroke moveto -hfont1 setfont sti stringwidth pop neg -15.4 rmoveto -sti show -BeginEPSF -154.5 363 translate --99 -45 translate -99 45 moveto -385 45 lineto -385 111 lineto -99 111 lineto -99 45 lineto clip newpath -%%BeginDocument: (nasmlogo.eps) -%!PS-Adobe-3.0 EPSF-3.0 -%%BoundingBox: 99 45 385 111 -%%DocumentData: Clean7Bit -%%DocumentFonts: Courier-Bold -%%DocumentNeededFonts: Courier-Bold -%%Title: (NASM logo) -%%EndComments -%%Page 1 1 -%%BeginPageSetup -save -newpath -%%EndPageSetup -% x y pointsize -- -/nasmlogo { -gsave 1 dict begin -/sz exch def -/Courier-Bold findfont sz scalefont setfont -moveto -0.85 1.22 scale -[(-~~..~:#;L .-:#;L,.- .~:#:;.T -~~.~:;. .~:;. ) -( E8+U *T +U' *T# .97 *L E8+' *;T' *;, ) -( D97 `*L .97 '*L "T;E+:, D9 *L *L ) -( H7 I# T7 I# "*:. H7 I# I# ) -( U: :8 *#+ , :8 T, 79 U: :8 :8 ) -(,#B. .IE, "T;E* .IE, J *+;#:T*" ,#B. .IE, .IE,)] { -currentpoint 3 -1 roll -sz -0.10 mul 0 3 -1 roll ashow -sz 0.72 mul sub moveto -} forall -end grestore -} def -0.6 setgray 100 100 12 nasmlogo -%%PageTrailer -restore -%%EndDocument -EndEPSF -restore showpage -%%Page: 2 2 -%%BeginPageSetup -save -%%EndPageSetup -/2 pa -[(\251 1996-2009 The NASM Development Team \227 All Rights Reserved)]164 norm3 -[(This document is redistributable under the license given in the file "COPYING" distributed in the NASM)]147 norm1 -[(archive.)]136 norm2 -[(This release is dedicated to the memory of Charles A. Crayne. We miss you, Chuck.)]119 norm3 -restore showpage -%%Page: 3 3 -%%BeginPageSetup -save -%%EndPageSetup -/3 pa -[{/contents xa}(Contents)]642.8 chap3 -[(Introduction)](Chapter 1: )/chapter-1 (14)607.8 toc03 -[(What Is NASM?)](1.1 )/section-1.1 (14)590.8 toc13 -[(Why Yet Another Assembler?)](1.1.1 )/section-1.1.1 (14)573.8 toc23 -[(License Conditions)](1.1.2 )/section-1.1.2 (14)556.8 toc23 -[(Contact Information)](1.2 )/section-1.2 (15)539.8 toc13 -[(Installation)](1.3 )/section-1.3 (15)522.8 toc13 -[(Installing NASM under MS-DOS or Windows)](1.3.1 )/section-1.3.1 (15)505.8 toc23 -[(Installing NASM under Unix)](1.3.2 )/section-1.3.2 (16)488.8 toc23 -[(Running NASM)](Chapter 2: )/chapter-2 (17)471.8 toc03 -[(NASM Command-Line Syntax)](2.1 )/section-2.1 (17)454.8 toc13 -[(The )2(-o)0( Option: Specifying the Output File Name)](2.1.1 )/section-2.1.1 (17)437.8 toc23 -[(The )2(-f)0( Option: Specifying the Output File Format)](2.1.2 )/section-2.1.2 (18)420.8 toc23 -[(The )2(-l)0( Option: Generating a Listing File)](2.1.3 )/section-2.1.3 (18)403.8 toc23 -[(The )2(-M)0( Option: Generate Makefile Dependencies)](2.1.4 )/section-2.1.4 (18)386.8 toc23 -[(The )2(-MG)0( Option: Generate Makefile Dependencies)](2.1.5 )/section-2.1.5 (18)369.8 toc23 -[(The )2(-MF)0( Option: Set Makefile Dependency File)](2.1.6 )/section-2.1.6 (18)352.8 toc23 -[(The )2(-MD)0( Option: Assemble and Generate Dependencies)](2.1.7 )/section-2.1.7 (18)335.8 toc23 -[(The )2(-MT)0( Option: Dependency Target Name)](2.1.8 )/section-2.1.8 (19)318.8 toc23 -[(The )2(-MQ)0( Option: Dependency Target Name \(Quoted\))](2.1.9 )/section-2.1.9 (19)301.8 toc23 -[(The )2(-MP)0( Option: Emit phony targets)](2.1.10 )/section-2.1.10 (19)284.8 toc23 -[(The )2(-F)0( Option: Selecting a Debug Information Format)](2.1.11 )/section-2.1.11 (19)267.8 toc23 -[(The )2(-g)0( Option: Enabling Debug Information.)](2.1.12 )/section-2.1.12 (19)250.8 toc23 -[(The )2(-X)0( Option: Selecting an Error Reporting Format)](2.1.13 )/section-2.1.13 (19)233.8 toc23 -[(The )2(-Z)0( Option: Send Errors to a File)](2.1.14 )/section-2.1.14 (20)216.8 toc23 -[(The )2(-s)0( Option: Send Errors to )2(stdout)](2.1.15 )/section-2.1.15 (20)199.8 toc23 -[(The )2(-i)0( Option: Include File Search Directories)](2.1.16 )/section-2.1.16 (20)182.8 toc23 -[(The )2(-p)0( Option: Pre-Include a File)](2.1.17 )/section-2.1.17 (20)165.8 toc23 -[(The )2(-d)0( Option: Pre-Define a Macro)](2.1.18 )/section-2.1.18 (20)148.8 toc23 -[(The )2(-u)0( Option: Undefine a Macro)](2.1.19 )/section-2.1.19 (21)131.8 toc23 -[(The )2(-E)0( Option: Preprocess Only)](2.1.20 )/section-2.1.20 (21)114.8 toc23 -(3)pageodd -restore showpage -%%Page: 4 4 -%%BeginPageSetup -save -%%EndPageSetup -/4 pa -[(The )2(-a)0( Option: Don't Preprocess At All)](2.1.21 )/section-2.1.21 (21)681 toc23 -[(The )2(-O)0( Option: Specifying Multipass Optimization)](2.1.22 )/section-2.1.22 (21)664 toc23 -[(The )2(-t)0( Option: Enable TASM Compatibility Mode)](2.1.23 )/section-2.1.23 (22)647 toc23 -[(The )2(-w)0( and )2(-W)0( Options: Enable or Disable Assembly Warnings)](2.1.24 )/section-2.1.24 (22)630 toc23 -[(The )2(-v)0( Option: Display Version Info)](2.1.25 )/section-2.1.25 (23)613 toc23 -[(The )2(-y)0( Option: Display Available Debug Info Formats)](2.1.26 )/section-2.1.26 (23)596 toc23 -[(The )2(--prefix)0( and )2(--postfix)0( Options.)](2.1.27 )/section-2.1.27 (23)579 toc23 -[(The )2(NASMENV)0( Environment Variable)](2.1.28 )/section-2.1.28 (23)562 toc23 -[(Quick Start for MASM Users)](2.2 )/section-2.2 (24)545 toc13 -[(NASM Is Case-Sensitive)](2.2.1 )/section-2.2.1 (24)528 toc23 -[(NASM Requires Square Brackets For Memory References)](2.2.2 )/section-2.2.2 (24)511 toc23 -[(NASM Doesn't Store Variable Types)](2.2.3 )/section-2.2.3 (24)494 toc23 -[(NASM Doesn't )2(ASSUME)](2.2.4 )/section-2.2.4 (25)477 toc23 -[(NASM Doesn't Support Memory Models)](2.2.5 )/section-2.2.5 (25)460 toc23 -[(Floating-Point Differences)](2.2.6 )/section-2.2.6 (25)443 toc23 -[(Other Differences)](2.2.7 )/section-2.2.7 (25)426 toc23 -[(The NASM Language)](Chapter 3: )/chapter-3 (26)409 toc03 -[(Layout of a NASM Source Line)](3.1 )/section-3.1 (26)392 toc13 -[(Pseudo-Instructions)](3.2 )/section-3.2 (27)375 toc13 -[2(DB)0( and Friends: Declaring Initialized Data)](3.2.1 )/section-3.2.1 (27)358 toc23 -[2(RESB)0( and Friends: Declaring Uninitialized Data)](3.2.2 )/section-3.2.2 (27)341 toc23 -[2(INCBIN)0(: Including External Binary Files)](3.2.3 )/section-3.2.3 (27)324 toc23 -[2(EQU)0(: Defining Constants)](3.2.4 )/section-3.2.4 (28)307 toc23 -[2(TIMES)0(: Repeating Instructions or Data)](3.2.5 )/section-3.2.5 (28)290 toc23 -[(Effective Addresses)](3.3 )/section-3.3 (28)273 toc13 -[(Constants)](3.4 )/section-3.4 (29)256 toc13 -[(Numeric Constants)](3.4.1 )/section-3.4.1 (29)239 toc23 -[(Character Strings)](3.4.2 )/section-3.4.2 (30)222 toc23 -[(Character Constants)](3.4.3 )/section-3.4.3 (31)205 toc23 -[(String Constants)](3.4.4 )/section-3.4.4 (31)188 toc23 -[(Unicode Strings)](3.4.5 )/section-3.4.5 (31)171 toc23 -[(Floating-Point Constants)](3.4.6 )/section-3.4.6 (31)154 toc23 -[(Packed BCD Constants)](3.4.7 )/section-3.4.7 (33)137 toc23 -[(Expressions)](3.5 )/section-3.5 (33)120 toc13 -[2(|)0(: Bitwise OR Operator)](3.5.1 )/section-3.5.1 (33)103 toc23 -(4)pageeven -restore showpage -%%Page: 5 5 -%%BeginPageSetup -save -%%EndPageSetup -/5 pa -[2(^)0(: Bitwise XOR Operator)](3.5.2 )/section-3.5.2 (33)681 toc23 -[2(&)0(: Bitwise AND Operator)](3.5.3 )/section-3.5.3 (33)664 toc23 -[2(<<)0( and )2(>>)0(: Bit Shift Operators)](3.5.4 )/section-3.5.4 (33)647 toc23 -[2(+)0( and )2(-)0(: Addition and Subtraction Operators)](3.5.5 )/section-3.5.5 (33)630 toc23 -[2(*)0(, )2(/)0(, )2(//)0(, )2(%)0( and )2(%%)0(: Multiplication and Division)](3.5.6 )/section-3.5.6 (33)613 toc23 -[(Unary Operators: )2(+)0(, )2(-)0(, )2(~)0(, )2(!)0( and )2(SEG)](3.5.7 )/section-3.5.7 (34)596 toc23 -[2(SEG)0( and )2(WRT)](3.6 )/section-3.6 (34)579 toc13 -[2(STRICT)0(: Inhibiting Optimization)](3.7 )/section-3.7 (34)562 toc13 -[(Critical Expressions)](3.8 )/section-3.8 (35)545 toc13 -[(Local Labels)](3.9 )/section-3.9 (35)528 toc13 -[(The NASM Preprocessor)](Chapter 4: )/chapter-4 (37)511 toc03 -[(Single-Line Macros)](4.1 )/section-4.1 (37)494 toc13 -[(The Normal Way: )2(%define)](4.1.1 )/section-4.1.1 (37)477 toc23 -[(Resolving )2(%define)0(: )2(%xdefine)](4.1.2 )/section-4.1.2 (38)460 toc23 -[(Macro Indirection: )2(%[...])](4.1.3 )/section-4.1.3 (39)443 toc23 -[(Concatenating Single Line Macro Tokens: )2(%+)](4.1.4 )/section-4.1.4 (39)426 toc23 -[(The Macro Name Itself: )2(%?)0( and )2(%??)](4.1.5 )/section-4.1.5 (39)409 toc23 -[(Undefining Single-Line Macros: )2(%undef)](4.1.6 )/section-4.1.6 (40)392 toc23 -[(Preprocessor Variables: )2(%assign)](4.1.7 )/section-4.1.7 (40)375 toc23 -[(Defining Strings: )2(%defstr)](4.1.8 )/section-4.1.8 (41)358 toc23 -[(Defining Tokens: )2(%deftok)](4.1.9 )/section-4.1.9 (41)341 toc23 -[(String Manipulation in Macros)](4.2 )/section-4.2 (41)324 toc13 -[(Concatenating Strings: )2(%strcat)](4.2.1 )/section-4.2.1 (41)307 toc23 -[(String Length: )2(%strlen)](4.2.2 )/section-4.2.2 (41)290 toc23 -[(Extracting Substrings: )2(%substr)](4.2.3 )/section-4.2.3 (42)273 toc23 -[(Multi-Line Macros: )2(%macro)](4.3 )/section-4.3 (42)256 toc13 -[(Recursive Multi-Line Macros: )2(%rmacro)](4.3.1 )/section-4.3.1 (43)239 toc23 -[(Overloading Multi-Line Macros)](4.3.2 )/section-4.3.2 (43)222 toc23 -[(Macro-Local Labels)](4.3.3 )/section-4.3.3 (43)205 toc23 -[(Greedy Macro Parameters)](4.3.4 )/section-4.3.4 (44)188 toc23 -[(Default Macro Parameters)](4.3.5 )/section-4.3.5 (45)171 toc23 -[2(%0)0(: Macro Parameter Counter)](4.3.6 )/section-4.3.6 (45)154 toc23 -[2(%rotate)0(: Rotating Macro Parameters)](4.3.7 )/section-4.3.7 (46)137 toc23 -[(Concatenating Macro Parameters)](4.3.8 )/section-4.3.8 (46)120 toc23 -[(Condition Codes as Macro Parameters)](4.3.9 )/section-4.3.9 (47)103 toc23 -(5)pageodd -restore showpage -%%Page: 6 6 -%%BeginPageSetup -save -%%EndPageSetup -/6 pa -[(Disabling Listing Expansion)](4.3.10 )/section-4.3.10 (48)681 toc23 -[(Undefining Multi-Line Macros: )2(%unmacro)](4.3.11 )/section-4.3.11 (48)664 toc23 -[(Exiting Multi-Line Macros: )2(%exitmacro)](4.3.12 )/section-4.3.12 (48)647 toc23 -[(Conditional Assembly)](4.4 )/section-4.4 (49)630 toc13 -[2(%ifdef)0(: Testing Single-Line Macro Existence)](4.4.1 )/section-4.4.1 (49)613 toc23 -[2(%ifmacro)0(: Testing Multi-Line Macro Existence)](4.4.2 )/section-4.4.2 (49)596 toc23 -[2(%ifctx)0(: Testing the Context Stack)](4.4.3 )/section-4.4.3 (50)579 toc23 -[2(%if)0(: Testing Arbitrary Numeric Expressions)](4.4.4 )/section-4.4.4 (50)562 toc23 -[2(%ifidn)0( and )2(%ifidni)0(: Testing Exact Text Identity)](4.4.5 )/section-4.4.5 (50)545 toc23 -[2(%ifid)0(, )2(%ifnum)0(, )2(%ifstr)0(: Testing Token Types)](4.4.6 )/section-4.4.6 (51)528 toc23 -[2(%iftoken)0(: Test for a Single Token)](4.4.7 )/section-4.4.7 (52)511 toc23 -[2(%ifempty)0(: Test for Empty Expansion)](4.4.8 )/section-4.4.8 (52)494 toc23 -[(Preprocessor Loops: )2(%rep)](4.5 )/section-4.5 (52)477 toc13 -[(Source Files and Dependencies)](4.6 )/section-4.6 (53)460 toc13 -[2(%include)0(: Including Other Files)](4.6.1 )/section-4.6.1 (53)443 toc23 -[2(%pathsearch)0(: Search the Include Path)](4.6.2 )/section-4.6.2 (53)426 toc23 -[2(%depend)0(: Add Dependent Files)](4.6.3 )/section-4.6.3 (53)409 toc23 -[2(%use)0(: Include Standard Macro Package)](4.6.4 )/section-4.6.4 (54)392 toc23 -[(The Context Stack)](4.7 )/section-4.7 (54)375 toc13 -[2(%push)0( and )2(%pop)0(: Creating and Removing Contexts)](4.7.1 )/section-4.7.1 (54)358 toc23 -[(Context-Local Labels)](4.7.2 )/section-4.7.2 (54)341 toc23 -[(Context-Local Single-Line Macros)](4.7.3 )/section-4.7.3 (55)324 toc23 -[2(%repl)0(: Renaming a Context)](4.7.4 )/section-4.7.4 (55)307 toc23 -[(Example Use of the Context Stack: Block IFs)](4.7.5 )/section-4.7.5 (55)290 toc23 -[(Stack Relative Preprocessor Directives)](4.8 )/section-4.8 (57)273 toc13 -[2(%arg)0( Directive)](4.8.1 )/section-4.8.1 (57)256 toc23 -[2(%stacksize)0( Directive)](4.8.2 )/section-4.8.2 (57)239 toc23 -[2(%local)0( Directive)](4.8.3 )/section-4.8.3 (58)222 toc23 -[(Reporting User-Defined Errors: )2(%error)0(, )2(%warning)0(, )2(%fatal)](4.9 )/section-4.9 (58)205 toc13 -[(Other Preprocessor Directives)](4.10 )/section-4.10 (59)188 toc13 -[2(%line)0( Directive)](4.10.1 )/section-4.10.1 (59)171 toc23 -[2(%!<env>)0(: Read an environment variable.)](4.10.2 )/section-4.10.2 (60)154 toc23 -[(Standard Macros)](4.11 )/section-4.11 (60)137 toc13 -[(NASM Version Macros)](4.11.1 )/section-4.11.1 (60)120 toc23 -[2(__NASM_VERSION_ID__)0(: NASM Version ID)](4.11.2 )/section-4.11.2 (60)103 toc23 -(6)pageeven -restore showpage -%%Page: 7 7 -%%BeginPageSetup -save -%%EndPageSetup -/7 pa -[2(__NASM_VER__)0(: NASM Version string)](4.11.3 )/section-4.11.3 (60)681 toc23 -[2(__FILE__)0( and )2(__LINE__)0(: File Name and Line Number)](4.11.4 )/section-4.11.4 (61)664 toc23 -[2(__BITS__)0(: Current BITS Mode)](4.11.5 )/section-4.11.5 (61)647 toc23 -[2(__OUTPUT_FORMAT__)0(: Current Output Format)](4.11.6 )/section-4.11.6 (61)630 toc23 -[(Assembly Date and Time Macros)](4.11.7 )/section-4.11.7 (61)613 toc23 -[2(__USE_)1(package)2(__)0(: Package Include Test)](4.11.8 )/section-4.11.8 (62)596 toc23 -[2(__PASS__)0(: Assembly Pass)](4.11.9 )/section-4.11.9 (62)579 toc23 -[2(STRUC)0( and )2(ENDSTRUC)0(: Declaring Structure Data Types)](4.11.10 )/section-4.11.10 (62)562 toc23 -[2(ISTRUC)0(, )2(AT)0( and )2(IEND)0(: Declaring Instances of Structures)](4.11.11 )/section-4.11.11 (63)545 toc23 -[2(ALIGN)0( and )2(ALIGNB)0(: Data Alignment)](4.11.12 )/section-4.11.12 (64)528 toc23 -[(Standard Macro Packages)](Chapter 5: )/chapter-5 (66)511 toc03 -[2(altreg)0(: Alternate Register Names)](5.1 )/section-5.1 (66)494 toc13 -[2(smartalign)0(: Smart )2(ALIGN)0( Macro)](5.2 )/section-5.2 (66)477 toc13 -[(Assembler Directives)](Chapter 6: )/chapter-6 (67)460 toc03 -[2(BITS)0(: Specifying Target Processor Mode)](6.1 )/section-6.1 (67)443 toc13 -[2(USE16)0( & )2(USE32)0(: Aliases for BITS)](6.1.1 )/section-6.1.1 (68)426 toc23 -[2(DEFAULT)0(: Change the assembler defaults)](6.2 )/section-6.2 (68)409 toc13 -[2(SECTION)0( or )2(SEGMENT)0(: Changing and Defining Sections)](6.3 )/section-6.3 (68)392 toc13 -[(The )2(__SECT__)0( Macro)](6.3.1 )/section-6.3.1 (68)375 toc23 -[2(ABSOLUTE)0(: Defining Absolute Labels)](6.4 )/section-6.4 (69)358 toc13 -[2(EXTERN)0(: Importing Symbols from Other Modules)](6.5 )/section-6.5 (70)341 toc13 -[2(GLOBAL)0(: Exporting Symbols to Other Modules)](6.6 )/section-6.6 (70)324 toc13 -[2(COMMON)0(: Defining Common Data Areas)](6.7 )/section-6.7 (70)307 toc13 -[2(CPU)0(: Defining CPU Dependencies)](6.8 )/section-6.8 (71)290 toc13 -[2(FLOAT)0(: Handling of floating-point constants)](6.9 )/section-6.9 (71)273 toc13 -[(Output Formats)](Chapter 7: )/chapter-7 (73)256 toc03 -[2(bin)0(: Flat-Form Binary Output)](7.1 )/section-7.1 (73)239 toc13 -[2(ORG)0(: Binary File Program Origin)](7.1.1 )/section-7.1.1 (73)222 toc23 -[2(bin)0( Extensions to the )2(SECTION)0( Directive)](7.1.2 )/section-7.1.2 (73)205 toc23 -[(Multisection Support for the )2(bin)0( Format)](7.1.3 )/section-7.1.3 (74)188 toc23 -[(Map Files)](7.1.4 )/section-7.1.4 (74)171 toc23 -[2(ith)0(: Intel Hex Output)](7.2 )/section-7.2 (74)154 toc13 -[2(srec)0(: Motorola S-Records Output)](7.3 )/section-7.3 (74)137 toc13 -[2(obj)0(: Microsoft OMF Object Files)](7.4 )/section-7.4 (75)120 toc13 -[2(obj)0( Extensions to the )2(SEGMENT)0( Directive)](7.4.1 )/section-7.4.1 (75)103 toc23 -(7)pageodd -restore showpage -%%Page: 8 8 -%%BeginPageSetup -save -%%EndPageSetup -/8 pa -[2(GROUP)0(: Defining Groups of Segments)](7.4.2 )/section-7.4.2 (76)681 toc23 -[2(UPPERCASE)0(: Disabling Case Sensitivity in Output)](7.4.3 )/section-7.4.3 (77)664 toc23 -[2(IMPORT)0(: Importing DLL Symbols)](7.4.4 )/section-7.4.4 (77)647 toc23 -[2(EXPORT)0(: Exporting DLL Symbols)](7.4.5 )/section-7.4.5 (77)630 toc23 -[2(..start)0(: Defining the Program Entry Point)](7.4.6 )/section-7.4.6 (78)613 toc23 -[2(obj)0( Extensions to the )2(EXTERN)0( Directive)](7.4.7 )/section-7.4.7 (78)596 toc23 -[2(obj)0( Extensions to the )2(COMMON)0( Directive)](7.4.8 )/section-7.4.8 (78)579 toc23 -[2(win32)0(: Microsoft Win32 Object Files)](7.5 )/section-7.5 (79)562 toc13 -[2(win32)0( Extensions to the )2(SECTION)0( Directive)](7.5.1 )/section-7.5.1 (79)545 toc23 -[2(win32)0(: Safe Structured Exception Handling)](7.5.2 )/section-7.5.2 (80)528 toc23 -[2(win64)0(: Microsoft Win64 Object Files)](7.6 )/section-7.6 (81)511 toc13 -[2(win64)0(: Writing Position-Independent Code)](7.6.1 )/section-7.6.1 (81)494 toc23 -[2(win64)0(: Structured Exception Handling)](7.6.2 )/section-7.6.2 (82)477 toc23 -[2(coff)0(: Common Object File Format)](7.7 )/section-7.7 (85)460 toc13 -[2(macho32)0( and )2(macho64)0(: Mach Object File Format)](7.8 )/section-7.8 (85)443 toc13 -[2(elf32)0( and )2(elf64)0(: Executable and Linkable Format Object Files)](7.9 )/section-7.9 (85)426 toc13 -[(ELF specific directive )2(osabi)](7.9.1 )/section-7.9.1 (85)409 toc23 -[2(elf)0( Extensions to the )2(SECTION)0( Directive)](7.9.2 )/section-7.9.2 (85)392 toc23 -[(Position-Independent Code: )2(elf)0( Special Symbols and )2(WRT)](7.9.3 )/section-7.9.3 (86)375 toc23 -[(Thread Local Storage: )2(elf)0( Special Symbols and )2(WRT)](7.9.4 )/section-7.9.4 (87)358 toc23 -[2(elf)0( Extensions to the )2(GLOBAL)0( Directive)](7.9.5 )/section-7.9.5 (87)341 toc23 -[2(elf)0( Extensions to the )2(COMMON)0( Directive )](7.9.6 )/section-7.9.6 (87)324 toc23 -[(16-bit code and ELF )](7.9.7 )/section-7.9.7 (88)307 toc23 -[(Debug formats and ELF )](7.9.8 )/section-7.9.8 (88)290 toc23 -[2(aout)0(: Linux )2(a.out)0( Object Files)](7.10 )/section-7.10 (88)273 toc13 -[2(aoutb)0(: NetBSD/FreeBSD/OpenBSD )2(a.out)0( Object Files)](7.11 )/section-7.11 (88)256 toc13 -[2(as86)0(: Minix/Linux )2(as86)0( Object Files)](7.12 )/section-7.12 (88)239 toc13 -[2(rdf)0(: Relocatable Dynamic Object File Format)](7.13 )/section-7.13 (88)222 toc13 -[(Requiring a Library: The )2(LIBRARY)0( Directive)](7.13.1 )/section-7.13.1 (89)205 toc23 -[(Specifying a Module Name: The )2(MODULE)0( Directive)](7.13.2 )/section-7.13.2 (89)188 toc23 -[2(rdf)0( Extensions to the )2(GLOBAL)0( Directive)](7.13.3 )/section-7.13.3 (89)171 toc23 -[2(rdf)0( Extensions to the )2(EXTERN)0( Directive)](7.13.4 )/section-7.13.4 (89)154 toc23 -[2(dbg)0(: Debugging Format)](7.14 )/section-7.14 (90)137 toc13 -[(Writing 16-bit Code \(DOS, Windows 3/3.1\))](Chapter 8: )/chapter-8 (91)120 toc03 -[(Producing )2(.EXE)0( Files)](8.1 )/section-8.1 (91)103 toc13 -(8)pageeven -restore showpage -%%Page: 9 9 -%%BeginPageSetup -save -%%EndPageSetup -/9 pa -[(Using the )2(obj)0( Format To Generate )2(.EXE)0( Files)](8.1.1 )/section-8.1.1 (91)681 toc23 -[(Using the )2(bin)0( Format To Generate )2(.EXE)0( Files)](8.1.2 )/section-8.1.2 (92)664 toc23 -[(Producing )2(.COM)0( Files)](8.2 )/section-8.2 (93)647 toc13 -[(Using the )2(bin)0( Format To Generate )2(.COM)0( Files)](8.2.1 )/section-8.2.1 (93)630 toc23 -[(Using the )2(obj)0( Format To Generate )2(.COM)0( Files)](8.2.2 )/section-8.2.2 (93)613 toc23 -[(Producing )2(.SYS)0( Files)](8.3 )/section-8.3 (94)596 toc13 -[(Interfacing to 16-bit C Programs)](8.4 )/section-8.4 (94)579 toc13 -[(External Symbol Names)](8.4.1 )/section-8.4.1 (94)562 toc23 -[(Memory Models)](8.4.2 )/section-8.4.2 (95)545 toc23 -[(Function Definitions and Function Calls)](8.4.3 )/section-8.4.3 (96)528 toc23 -[(Accessing Data Items)](8.4.4 )/section-8.4.4 (98)511 toc23 -[2(c16.mac)0(: Helper Macros for the 16-bit C Interface)](8.4.5 )/section-8.4.5 (98)494 toc23 -[(Interfacing to Borland Pascal Programs)](8.5 )/section-8.5 (99)477 toc13 -[(The Pascal Calling Convention)](8.5.1 )/section-8.5.1 (100)460 toc23 -[(Borland Pascal Segment Name Restrictions)](8.5.2 )/section-8.5.2 (101)443 toc23 -[(Using )2(c16.mac)0( With Pascal Programs)](8.5.3 )/section-8.5.3 (101)426 toc23 -[(Writing 32-bit Code \(Unix, Win32, DJGPP\))](Chapter 9: )/chapter-9 (103)409 toc03 -[(Interfacing to 32-bit C Programs)](9.1 )/section-9.1 (103)392 toc13 -[(External Symbol Names)](9.1.1 )/section-9.1.1 (103)375 toc23 -[(Function Definitions and Function Calls)](9.1.2 )/section-9.1.2 (103)358 toc23 -[(Accessing Data Items)](9.1.3 )/section-9.1.3 (105)341 toc23 -[2(c32.mac)0(: Helper Macros for the 32-bit C Interface)](9.1.4 )/section-9.1.4 (105)324 toc23 -[(Writing NetBSD/FreeBSD/OpenBSD and Linux/ELF Shared Libraries)](9.2 )/section-9.2 (106)307 toc13 -[(Obtaining the Address of the GOT)](9.2.1 )/section-9.2.1 (106)290 toc23 -[(Finding Your Local Data Items)](9.2.2 )/section-9.2.2 (107)273 toc23 -[(Finding External and Common Data Items)](9.2.3 )/section-9.2.3 (107)256 toc23 -[(Exporting Symbols to the Library User)](9.2.4 )/section-9.2.4 (108)239 toc23 -[(Calling Procedures Outside the Library)](9.2.5 )/section-9.2.5 (109)222 toc23 -[(Generating the Library File)](9.2.6 )/section-9.2.6 (109)205 toc23 -[(Mixing 16 and 32 Bit Code)](Chapter 10: )/chapter-10 (110)188 toc03 -[(Mixed-Size Jumps)](10.1 )/section-10.1 (110)171 toc13 -[(Addressing Between Different-Size Segments)](10.2 )/section-10.2 (110)154 toc13 -[(Other Mixed-Size Instructions)](10.3 )/section-10.3 (111)137 toc13 -[(Writing 64-bit Code \(Unix, Win64\))](Chapter 11: )/chapter-11 (113)120 toc03 -[(Register Names in 64-bit Mode)](11.1 )/section-11.1 (113)103 toc13 -(9)pageodd -restore showpage -%%Page: 10 10 -%%BeginPageSetup -save -%%EndPageSetup -/10 pa -[(Immediates and Displacements in 64-bit Mode)](11.2 )/section-11.2 (113)681 toc13 -[(Interfacing to 64-bit C Programs \(Unix\))](11.3 )/section-11.3 (114)664 toc13 -[(Interfacing to 64-bit C Programs \(Win64\))](11.4 )/section-11.4 (114)647 toc13 -[(Troubleshooting)](Chapter 12: )/chapter-12 (116)630 toc03 -[(Common Problems)](12.1 )/section-12.1 (116)613 toc13 -[(NASM Generates Inefficient Code)](12.1.1 )/section-12.1.1 (116)596 toc23 -[(My Jumps are Out of Range)](12.1.2 )/section-12.1.2 (116)579 toc23 -[2(ORG)0( Doesn't Work)](12.1.3 )/section-12.1.3 (116)562 toc23 -[2(TIMES)0( Doesn't Work)](12.1.4 )/section-12.1.4 (117)545 toc23 -[(Bugs)](12.2 )/section-12.2 (117)528 toc13 -[(Ndisasm)](Appendix A: )/appendix-A (119)511 toc03 -[(Introduction)](A.1 )/section-A.1 (119)494 toc13 -[(Getting Started: Installation)](A.2 )/section-A.2 (119)477 toc13 -[(Running NDISASM)](A.3 )/section-A.3 (119)460 toc13 -[(COM Files: Specifying an Origin)](A.3.1 )/section-A.3.1 (119)443 toc23 -[(Code Following Data: Synchronisation)](A.3.2 )/section-A.3.2 (119)426 toc23 -[(Mixed Code and Data: Automatic \(Intelligent\) Synchronisation )](A.3.3 )/section-A.3.3 (120)409 toc23 -[(Other Options)](A.3.4 )/section-A.3.4 (121)392 toc23 -[(Bugs and Improvements)](A.4 )/section-A.4 (121)375 toc13 -[(Instruction List)](Appendix B: )/appendix-B (122)358 toc03 -[(Introduction)](B.1 )/section-B.1 (122)341 toc13 -[(Special instructions...)](B.1.1 )/section-B.1.1 (122)324 toc23 -[(Conventional instructions)](B.1.2 )/section-B.1.2 (122)307 toc23 -[(Katmai Streaming SIMD instructions \(SSE \226\226 a.k.a. KNI, XMM, MMX2\))](B.1.3 )/section-B.1.3 (148)290 toc23 -[(Introduced in Deschutes but necessary for SSE support)](B.1.4 )/section-B.1.4 (149)273 toc23 -[(XSAVE group \(AVX and extended state\))](B.1.5 )/section-B.1.5 (149)256 toc23 -[(Generic memory operations)](B.1.6 )/section-B.1.6 (149)239 toc23 -[(New MMX instructions introduced in Katmai)](B.1.7 )/section-B.1.7 (150)222 toc23 -[(AMD Enhanced 3DNow! \(Athlon\) instructions)](B.1.8 )/section-B.1.8 (150)205 toc23 -[(Willamette SSE2 Cacheability Instructions)](B.1.9 )/section-B.1.9 (150)188 toc23 -[(Willamette MMX instructions \(SSE2 SIMD Integer Instructions\))](B.1.10 )/section-B.1.10 (150)171 toc23 -[(Willamette Streaming SIMD instructions \(SSE2\))](B.1.11 )/section-B.1.11 (152)154 toc23 -[(Prescott New Instructions \(SSE3\))](B.1.12 )/section-B.1.12 (154)137 toc23 -[(VMX Instructions)](B.1.13 )/section-B.1.13 (154)120 toc23 -[(Extended Page Tables VMX instructions)](B.1.14 )/section-B.1.14 (154)103 toc23 -(10)pageeven -restore showpage -%%Page: 11 11 -%%BeginPageSetup -save -%%EndPageSetup -/11 pa -[(Tejas New Instructions \(SSSE3\))](B.1.15 )/section-B.1.15 (155)681 toc23 -[(AMD SSE4A)](B.1.16 )/section-B.1.16 (155)664 toc23 -[(New instructions in Barcelona)](B.1.17 )/section-B.1.17 (155)647 toc23 -[(Penryn New Instructions \(SSE4.1\))](B.1.18 )/section-B.1.18 (155)630 toc23 -[(Nehalem New Instructions \(SSE4.2\))](B.1.19 )/section-B.1.19 (157)613 toc23 -[(Intel SMX)](B.1.20 )/section-B.1.20 (157)596 toc23 -[(Geode \(Cyrix\) 3DNow! additions)](B.1.21 )/section-B.1.21 (157)579 toc23 -[(Intel new instructions in ???)](B.1.22 )/section-B.1.22 (157)562 toc23 -[(Intel AES instructions)](B.1.23 )/section-B.1.23 (157)545 toc23 -[(Intel AVX AES instructions)](B.1.24 )/section-B.1.24 (157)528 toc23 -[(Intel AVX instructions)](B.1.25 )/section-B.1.25 (158)511 toc23 -[(Intel Carry-Less Multiplication instructions \(CLMUL\))](B.1.26 )/section-B.1.26 (170)494 toc23 -[(Intel AVX Carry-Less Multiplication instructions \(CLMUL\))](B.1.27 )/section-B.1.27 (170)477 toc23 -[(Intel Fused Multiply-Add instructions \(FMA\))](B.1.28 )/section-B.1.28 (170)460 toc23 -[(VIA \(Centaur\) security instructions)](B.1.29 )/section-B.1.29 (174)443 toc23 -[(AMD Lightweight Profiling \(LWP\) instructions)](B.1.30 )/section-B.1.30 (174)426 toc23 -[(AMD XOP, FMA4 and CVT16 instructions \(SSE5\))](B.1.31 )/section-B.1.31 (174)409 toc23 -[(Systematic names for the hinting nop instructions)](B.1.32 )/section-B.1.32 (177)392 toc23 -[(NASM Version History)](Appendix C: )/appendix-C (181)375 toc03 -[(NASM 2 Series)](C.1 )/section-C.1 (181)358 toc13 -[(Version 2.08)](C.1.1 )/section-C.1.1 (181)341 toc23 -[(Version 2.07)](C.1.2 )/section-C.1.2 (181)324 toc23 -[(Version 2.06)](C.1.3 )/section-C.1.3 (182)307 toc23 -[(Version 2.05.01)](C.1.4 )/section-C.1.4 (182)290 toc23 -[(Version 2.05)](C.1.5 )/section-C.1.5 (182)273 toc23 -[(Version 2.04)](C.1.6 )/section-C.1.6 (183)256 toc23 -[(Version 2.03.01)](C.1.7 )/section-C.1.7 (184)239 toc23 -[(Version 2.03)](C.1.8 )/section-C.1.8 (184)222 toc23 -[(Version 2.02)](C.1.9 )/section-C.1.9 (184)205 toc23 -[(Version 2.01)](C.1.10 )/section-C.1.10 (185)188 toc23 -[(Version 2.00)](C.1.11 )/section-C.1.11 (185)171 toc23 -[(NASM 0.98 Series)](C.2 )/section-C.2 (186)154 toc13 -[(Version 0.98.39)](C.2.1 )/section-C.2.1 (186)137 toc23 -[(Version 0.98.38)](C.2.2 )/section-C.2.2 (186)120 toc23 -[(Version 0.98.37)](C.2.3 )/section-C.2.3 (187)103 toc23 -(11)pageodd -restore showpage -%%Page: 12 12 -%%BeginPageSetup -save -%%EndPageSetup -/12 pa -[(Version 0.98.36)](C.2.4 )/section-C.2.4 (187)681 toc23 -[(Version 0.98.35)](C.2.5 )/section-C.2.5 (187)664 toc23 -[(Version 0.98.34)](C.2.6 )/section-C.2.6 (187)647 toc23 -[(Version 0.98.33)](C.2.7 )/section-C.2.7 (188)630 toc23 -[(Version 0.98.32)](C.2.8 )/section-C.2.8 (188)613 toc23 -[(Version 0.98.31)](C.2.9 )/section-C.2.9 (188)596 toc23 -[(Version 0.98.30)](C.2.10 )/section-C.2.10 (189)579 toc23 -[(Version 0.98.28)](C.2.11 )/section-C.2.11 (189)562 toc23 -[(Version 0.98.26)](C.2.12 )/section-C.2.12 (189)545 toc23 -[(Version 0.98.25alt)](C.2.13 )/section-C.2.13 (189)528 toc23 -[(Version 0.98.25)](C.2.14 )/section-C.2.14 (189)511 toc23 -[(Version 0.98.24p1)](C.2.15 )/section-C.2.15 (189)494 toc23 -[(Version 0.98.24)](C.2.16 )/section-C.2.16 (189)477 toc23 -[(Version 0.98.23)](C.2.17 )/section-C.2.17 (189)460 toc23 -[(Version 0.98.22)](C.2.18 )/section-C.2.18 (189)443 toc23 -[(Version 0.98.21)](C.2.19 )/section-C.2.19 (189)426 toc23 -[(Version 0.98.20)](C.2.20 )/section-C.2.20 (190)409 toc23 -[(Version 0.98.19)](C.2.21 )/section-C.2.21 (190)392 toc23 -[(Version 0.98.18)](C.2.22 )/section-C.2.22 (190)375 toc23 -[(Version 0.98.17)](C.2.23 )/section-C.2.23 (190)358 toc23 -[(Version 0.98.16)](C.2.24 )/section-C.2.24 (190)341 toc23 -[(Version 0.98.15)](C.2.25 )/section-C.2.25 (190)324 toc23 -[(Version 0.98.14)](C.2.26 )/section-C.2.26 (190)307 toc23 -[(Version 0.98.13)](C.2.27 )/section-C.2.27 (190)290 toc23 -[(Version 0.98.12)](C.2.28 )/section-C.2.28 (190)273 toc23 -[(Version 0.98.11)](C.2.29 )/section-C.2.29 (190)256 toc23 -[(Version 0.98.10)](C.2.30 )/section-C.2.30 (190)239 toc23 -[(Version 0.98.09)](C.2.31 )/section-C.2.31 (190)222 toc23 -[(Version 0.98.08)](C.2.32 )/section-C.2.32 (191)205 toc23 -[(Version 0.98.09b with John Coffman patches released 28-Oct-2001)](C.2.33 )/section-C.2.33 (191)188 toc23 -[(Version 0.98.07 released 01/28/01)](C.2.34 )/section-C.2.34 (191)171 toc23 -[(Version 0.98.06f released 01/18/01)](C.2.35 )/section-C.2.35 (191)154 toc23 -[(Version 0.98.06e released 01/09/01)](C.2.36 )/section-C.2.36 (191)137 toc23 -[(Version 0.98p1)](C.2.37 )/section-C.2.37 (192)120 toc23 -[(Version 0.98bf \(bug-fixed\))](C.2.38 )/section-C.2.38 (192)103 toc23 -(12)pageeven -restore showpage -%%Page: 13 13 -%%BeginPageSetup -save -%%EndPageSetup -/13 pa -[(Version 0.98.03 with John Coffman's changes released 27-Jul-2000)](C.2.39 )/section-C.2.39 (192)681 toc23 -[(Version 0.98.03)](C.2.40 )/section-C.2.40 (192)664 toc23 -[(Version 0.98)](C.2.41 )/section-C.2.41 (195)647 toc23 -[(Version 0.98p9)](C.2.42 )/section-C.2.42 (196)630 toc23 -[(Version 0.98p8)](C.2.43 )/section-C.2.43 (196)613 toc23 -[(Version 0.98p7)](C.2.44 )/section-C.2.44 (196)596 toc23 -[(Version 0.98p6)](C.2.45 )/section-C.2.45 (197)579 toc23 -[(Version 0.98p3.7)](C.2.46 )/section-C.2.46 (197)562 toc23 -[(Version 0.98p3.6)](C.2.47 )/section-C.2.47 (197)545 toc23 -[(Version 0.98p3.5)](C.2.48 )/section-C.2.48 (197)528 toc23 -[(Version 0.98p3.4)](C.2.49 )/section-C.2.49 (198)511 toc23 -[(Version 0.98p3.3)](C.2.50 )/section-C.2.50 (198)494 toc23 -[(Version 0.98p3.2)](C.2.51 )/section-C.2.51 (198)477 toc23 -[(Version 0.98p3-hpa)](C.2.52 )/section-C.2.52 (198)460 toc23 -[(Version 0.98 pre-release 3)](C.2.53 )/section-C.2.53 (198)443 toc23 -[(Version 0.98 pre-release 2)](C.2.54 )/section-C.2.54 (199)426 toc23 -[(Version 0.98 pre-release 1)](C.2.55 )/section-C.2.55 (199)409 toc23 -[(NASM 0.9 Series)](C.3 )/section-C.3 (200)392 toc13 -[(Version 0.97 released December 1997)](C.3.1 )/section-C.3.1 (200)375 toc23 -[(Version 0.96 released November 1997)](C.3.2 )/section-C.3.2 (200)358 toc23 -[(Version 0.95 released July 1997)](C.3.3 )/section-C.3.3 (202)341 toc23 -[(Version 0.94 released April 1997)](C.3.4 )/section-C.3.4 (204)324 toc23 -[(Version 0.93 released January 1997)](C.3.5 )/section-C.3.5 (204)307 toc23 -[(Version 0.92 released January 1997)](C.3.6 )/section-C.3.6 (205)290 toc23 -[(Version 0.91 released November 1996)](C.3.7 )/section-C.3.7 (205)273 toc23 -[(Version 0.90 released October 1996)](C.3.8 )/section-C.3.8 (205)256 toc23 -(13)pageodd -restore showpage -%%Page: 14 14 -%%BeginPageSetup -save -%%EndPageSetup -/14 pa -[{/chapter-1 xa}(Chapter 1: Introduction)]642.8 chap3 -[{/section-1.1 xa}(What Is NASM?)](1.1)603.4 head3 -[(The Netwide Assembler, NASM, is an 80x86 and x86-64 assembler designed for portability and modularity.)]586.4 norm1 -[(It supports a range of object file formats, including Linux and )2(*BSD)0( )2(a.out)0(, )2(ELF)0(, )2(COFF)0(, )2(Mach-O)0(,)]575.4 norm0 -[(Microsoft 16-bit )2(OBJ)0(, )2(Win32)0( and )2(Win64)0(. It will also output plain binary files. Its syntax is designed to be)]564.4 norm0 -[(simple and easy to understand, similar to Intel's but less complex. It supports all currently known x86)]553.4 norm0 -[(architectural extensions, and has strong support for macros.)]542.4 norm2 -[{/section-1.1.1 xa}(Why Yet Another Assembler?)](1.1.1)523.2 subh3 -[(The Netwide Assembler grew out of an idea on )2(comp.lang.asm.x86)0( \(or possibly )2(alt.lang.asm)0( \226 I)]506.2 norm1 -[(forget which\), which was essentially that there didn't seem to be a good )1(free)0( x86-series assembler around,)]495.2 norm0 -[(and that maybe someone ought to write one.)]484.2 norm2 -[2(a86)0( is good, but not free, and in particular you don't get any 32-bit capability until you pay. It's DOS)]467.2 bull1 -[(only, too.)]456.2 bull2 -[2(gas)0( is free, and ports over to DOS and Unix, but it's not very good, since it's designed to be a back end to)]439.2 bull1 -[2(gcc)0(, which always feeds it correct code. So its error checking is minimal. Also, its syntax is horrible, from)]428.2 bull0 -[(the point of view of anyone trying to actually )1(write)0( anything in it. Plus you can't write 16-bit code in it)]417.2 bull0 -[(\(properly.\))]406.2 bull2 -[2(as86)0( is specific to Minix and Linux, and \(my version at least\) doesn't seem to have much \(or any\))]389.2 bull1 -[(documentation.)]378.2 bull2 -[2(MASM)0( isn't very good, and it's \(was\) expensive, and it runs only under DOS.)]361.2 bull3 -[2(TASM)0( is better, but still strives for MASM compatibility, which means millions of directives and tons of)]344.2 bull1 -[(red tape. And its syntax is essentially MASM's, with the contradictions and quirks that entails \(although it)]333.2 bull0 -[(sorts out some of those by means of Ideal mode.\) It's expensive too. And it's DOS-only.)]322.2 bull2 -[(So here, for your coding pleasure, is NASM. At present it's still in prototype stage \226 we don't promise that it)]305.2 norm1 -[(can outperform any of these assemblers. But please, )1(please)0( send us bug reports, fixes, helpful information,)]294.2 norm0 -[(and anything else you can get your hands on \(and thanks to the many people who've done this already! You)]283.2 norm0 -[(all know who you are\), and we'll improve it out of all recognition. Again.)]272.2 norm2 -[{/section-1.1.2 xa}(License Conditions)](1.1.2)253 subh3 -[(Please see the file )2(LICENSE)0(, supplied as part of any NASM distribution archive, for the license conditions)]236 norm1 -[(under which you may use NASM. NASM is now under the so-called 2-clause BSD license, also known as)]225 norm0 -[(the simplified BSD license.)]214 norm2 -[(Copyright 1996-2009 the NASM Authors \226 All rights reserved.)]197 norm3 -[(Redistribution and use in source and binary forms, with or without modification, are permitted provided that)]180 norm1 -[(the following conditions are met:)]169 norm2 -[(Redistributions of source code must retain the above copyright notice, this list of conditions and the)]152 bull1 -[(following disclaimer.)]141 bull2 -[(Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the)]124 bull1 -[(following disclaimer in the documentation and/or other materials provided with the distribution.)]113 bull2 -(14)pageeven -restore showpage -%%Page: 15 15 -%%BeginPageSetup -save -%%EndPageSetup -/15 pa -[(THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS")]681 norm1 -[(AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE)]670 norm0 -[(IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE)]659 norm0 -[(ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE)]648 norm0 -[(LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR)]637 norm0 -[(CONSEQUENTIAL DAMAGES \(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF)]626 norm0 -[(SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS)]615 norm0 -[(INTERRUPTION\) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN)]604 norm0 -[(CONTRACT, STRICT LIABILITY, OR TORT \(INCLUDING NEGLIGENCE OR OTHERWISE\))]593 norm0 -[(ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE)]582 norm0 -[(POSSIBILITY OF SUCH DAMAGE.)]571 norm2 -[{/section-1.2 xa}(Contact Information)](1.2)549.6 head3 -[(The current version of NASM \(since about 0.98.08\) is maintained by a team of developers, accessible through)]532.6 norm1 -[(the )2(nasm-devel)0( mailing list \(see below for the link\). If you want to report a bug, please read ){/section-12.2 xl}(section 12.2){el}]521.6 norm0 -[(first.)]510.6 norm2 -[(NASM has a )(website at ){(http://www.nasm.us/)wl}2(http://www.nasm.us/){el}0(. If it's not there, google for us!)]493.6 norm3 -[(New releases, )(release candidates, and )(daily development snapshots of NASM are available from the official)]476.6 norm1 -[(web site.)]465.6 norm2 -[(Announcements are posted to ){(news:comp.lang.asm.x86)wl}2(comp.lang.asm.x86){el}0(, and to the web site)]448.6 norm1 -[{(http://www.freshmeat.net/)wl}2(http://www.freshmeat.net/){el}0(.)]437.6 norm2 -[(If you want information about the current development status, please subscribe to the )2(nasm-devel)0( email)]420.6 norm1 -[(list; see link from the website.)]409.6 norm2 -[{/section-1.3 xa}(Installation)](1.3)388.2 head3 -[{/section-1.3.1 xa}(Installing NASM under MS-)(DOS or Windows)](1.3.1)369 subh3 -[(Once you've obtained the appropriate archive for NASM, )2(nasm-XXX-dos.zip)0( or)]352 norm1 -[2(nasm-XXX-win32.zip)0( \(where )2(XXX)0( denotes the version number of NASM contained in the archive\),)]341 norm0 -[(unpack it into its own directory \(for example )2(c:\\nasm)0(\).)]330 norm2 -[(The archive will contain a set of executable files: the NASM executable file )2(nasm.exe)0(, the NDISASM)]313 norm1 -[(executable file )2(ndisasm.exe)0(, and possibly additional utilities to handle the RDOFF file format.)]302 norm2 -[(The only file NASM needs to run is its own executable, so copy )2(nasm.exe)0( to a directory on your PATH, or)]285 norm1 -[(alternatively edit )2(autoexec.bat)0( to add the )2(nasm)0( directory to your )2(PATH)0( \(to do that under Windows XP,)]274 norm0 -[(go to Start > Control Panel > System > Advanced > Environment Variables; these instructions may work)]263 norm0 -[(under other versions of Windows as well.\))]252 norm2 -[(That's it \226 NASM is installed. You don't need the nasm directory to be present to run NASM \(unless you've)]235 norm1 -[(added it to your )2(PATH)0(\), so you can delete it if you need to save space; however, you may want to keep the)]224 norm0 -[(documentation or test programs.)]213 norm2 -[(If you've downloaded the )(DOS source archive, )2(nasm-XXX.zip)0(, the )2(nasm)0( directory will also contain the)]196 norm1 -[(full NASM )(source code, and a selection of )(Makefiles you can \(hopefully\) use to rebuild your copy of NASM)]185 norm0 -[(from scratch. See the file )2(INSTALL)0( in the source archive.)]174 norm2 -[(Note that a number of files are generated from other files by Perl scripts. Although the NASM source)]157 norm1 -[(distribution includes these generated files, you will need to rebuild them \(and hence, will need a Perl)]146 norm0 -[(interpreter\) if you change insns.dat, standard.mac or the documentation. It is possible future source)]135 norm0 -[(distributions may not include these files at all. Ports of )(Perl for a variety of platforms, including DOS and)]124 norm0 -[(Windows, are available from ){(http://www.cpan.org/ports/)wl}(www.cpan.org){el}(.)]113 norm2 -(15)pageodd -restore showpage -%%Page: 16 16 -%%BeginPageSetup -save -%%EndPageSetup -/16 pa -[{/section-1.3.2 xa}(Installing NASM under )(Unix)](1.3.2)678.8 subh3 -[(Once you've obtained the )(Unix source archive for NASM, )2(nasm-XXX.tar.gz)0( \(where )2(XXX)0( denotes the)]661.8 norm1 -[(version number of NASM contained in the archive\), unpack it into a directory such as )2(/usr/local/src)0(.)]650.8 norm0 -[(The archive, when unpacked, will create its own subdirectory )2(nasm-XXX)0(.)]639.8 norm2 -[(NASM is an )(auto-configuring package: once you've unpacked it, )2(cd)0( to the directory it's been unpacked into)]622.8 norm1 -[(and type )2(./configure)0(. This shell script will find the best C compiler to use for building NASM and set up)]611.8 norm0 -[(Makefiles accordingly.)]600.8 norm2 -[(Once NASM has auto-configured, you can type )2(make)0( to build the )2(nasm)0( and )2(ndisasm)0( binaries, and then)]583.8 norm1 -[2(make install)0( to install them in )2(/usr/local/bin)0( and install the )(man pages )2(nasm.1)0( and)]572.8 norm0 -[2(ndisasm.1)0( in )2(/usr/local/man/man1)0(. Alternatively, you can give options such as )2(--prefix)0( to the)]561.8 norm0 -[(configure script \(see the file )2(INSTALL)0( for more details\), or install the programs yourself.)]550.8 norm2 -[(NASM also comes with a set of utilities for handling the )2(RDOFF)0( custom object-file format, which are in the)]533.8 norm1 -[2(rdoff)0( subdirectory of the NASM archive. You can build these with )2(make rdf)0( and install them with)]522.8 norm0 -[2(make rdf_install)0(, if you want them.)]511.8 norm2 -(16)pageeven -restore showpage -%%Page: 17 17 -%%BeginPageSetup -save -%%EndPageSetup -/17 pa -[{/chapter-2 xa}(Chapter 2: Running NASM)]642.8 chap3 -[{/section-2.1 xa}(NASM )(Command-Line Syntax)](2.1)603.4 head3 -[(To assemble a file, you issue a command of the form)]586.4 norm3 -[2(nasm -f <format> <filename> [-o <output>])]569.4 code3 -[(For example,)]552.4 norm3 -[2(nasm -f elf myfile.asm)]535.4 code3 -[(will assemble )2(myfile.asm)0( into an )2(ELF)0( object file )2(myfile.o)0(. And)]518.4 norm3 -[2(nasm -f bin myfile.asm -o myfile.com)]501.4 code3 -[(will assemble )2(myfile.asm)0( into a raw binary file )2(myfile.com)0(.)]484.4 norm3 -[(To produce a listing file, with the hex codes output from NASM displayed on the left of the original sources,)]467.4 norm1 -[(use the )2(-l)0( option to give a listing file name, for example:)]456.4 norm2 -[2(nasm -f coff myfile.asm -l myfile.lst)]439.4 code3 -[(To get further usage instructions from NASM, try typing)]422.4 norm3 -[2(nasm -h)]405.4 code3 -[(As )2(-hf)0(, this will also list the available output file formats, and what they are.)]388.4 norm3 -[(If you use Linux but aren't sure whether your system is )2(a.out)0( or )2(ELF)0(, type)]371.4 norm3 -[2(file nasm)]354.4 code3 -[(\(in the directory in which you put the NASM binary when you installed it\). If it says something like)]337.4 norm3 -[2(nasm: ELF 32-bit LSB executable i386 \(386 and up\) Version 1)]320.4 code3 -[(then your system is )2(ELF)0(, and you should use the option )2(-f elf)0( when you want NASM to produce Linux)]303.4 norm1 -[(object files. If it says)]292.4 norm2 -[2(nasm: Linux/i386 demand-paged executable \(QMAGIC\))]275.4 code3 -[(or something similar, your system is )2(a.out)0(, and you should use )2(-f aout)0( instead \(Linux )2(a.out)0( systems)]258.4 norm1 -[(have long been obsolete, and are rare these days.\))]247.4 norm2 -[(Like Unix compilers and assemblers, NASM is silent unless it goes wrong: you won't see any output at all,)]230.4 norm1 -[(unless it gives error messages.)]219.4 norm2 -[{/section-2.1.1 xa}(The )2(-o)0( Option: Specifying the Output File Name)](2.1.1)200.2 subh3 -[(NASM will normally choose the name of your output file for you; precisely how it does this is dependent on)]183.2 norm1 -[(the object file format. For Microsoft object file formats \()2(obj)0(, )2(win32)0( and )2(win64)0(\), it will remove the )2(.asm)]172.2 norm0 -[(extension \(or whatever extension you like to use \226 NASM doesn't care\) from your source file name and)]161.2 norm0 -[(substitute )2(.obj)0(. For Unix object file formats \()2(aout)0(, )2(as86)0(, )2(coff)0(, )2(elf32)0(, )2(elf64)0(, )2(ieee)0(, )2(macho32)0( and)]150.2 norm0 -[2(macho64)0(\) it will substitute )2(.o)0(. For )2(dbg)0(, )2(rdf)0(, )2(ith)0( and )2(srec)0(, it will use )2(.dbg)0(, )2(.rdf)0(, )2(.ith)0( and )2(.srec)0(,)]139.2 norm0 -[(respectively, and for the )2(bin)0( format it will simply remove the extension, so that )2(myfile.asm)0( produces the)]128.2 norm0 -[(output file )2(myfile)0(.)]117.2 norm2 -(17)pageodd -restore showpage -%%Page: 18 18 -%%BeginPageSetup -save -%%EndPageSetup -/18 pa -[(If the output file already exists, NASM will overwrite it, unless it has the same name as the input file, in)]681 norm1 -[(which case it will give a warning and use )2(nasm.out)0( as the output file name instead.)]670 norm2 -[(For situations in which this behaviour is unacceptable, NASM provides the )2(-o)0( command-line option, which)]653 norm1 -[(allows you to specify your desired output file name. You invoke )2(-o)0( by following it with the name you wish)]642 norm0 -[(for the output file, either with or without an intervening space. For example:)]631 norm2 -[2(nasm -f bin program.asm -o program.com )]614 code1 -[2(nasm -f bin driver.asm -odriver.sys)]603 code2 -[(Note that this is a small o, and is different from a capital O , which is used to specify the number of)]586 norm1 -[(optimisation passes required. See ){/section-2.1.22 xl}(section 2.1.22){el}(.)]575 norm2 -[{/section-2.1.2 xa}(The )2(-f)0( Option: Specifying the )(Output File Format)](2.1.2)555.8 subh3 -[(If you do not supply the )2(-f)0( option to NASM, it will choose an output file format for you itself. In the)]538.8 norm1 -[(distribution versions of NASM, the default is always )2(bin)0(; if you've compiled your own copy of NASM, you)]527.8 norm0 -[(can redefine )2(OF_DEFAULT)0( at compile time and choose what you want the default to be.)]516.8 norm2 -[(Like )2(-o)0(, the intervening space between )2(-f)0( and the output file format is optional; so )2(-f elf)0( and )2(-felf)0( are)]499.8 norm1 -[(both valid.)]488.8 norm2 -[(A complete list of the available output file formats can be given by issuing the command )2(nasm -hf)0(.)]471.8 norm3 -[{/section-2.1.3 xa}(The )2(-l)0( Option: Generating a )(Listing File)](2.1.3)452.6 subh3 -[(If you supply the )2(-l)0( option to NASM, followed \(with the usual optional space\) by a file name, NASM will)]435.6 norm1 -[(generate a )(source-listing file for you, in which addresses and generated code are listed on the left, and the)]424.6 norm0 -[(actual source code, with expansions of multi-line macros \(except those which specifically request no)]413.6 norm0 -[(expansion in source listings: see ){/section-4.3.10 xl}(section 4.3.10){el}(\) on the right. For example:)]402.6 norm2 -[2(nasm -f elf myfile.asm -l myfile.lst)]385.6 code3 -[(If a list file is selected, you may turn off listing for a section of your source with )2([list -])0(, and turn it back)]368.6 norm1 -[(on with )2([list +])0(, \(the default, obviously\). There is no "user form" \(without the brackets\). This can be used)]357.6 norm0 -[(to list only sections of interest, avoiding excessively long listings.)]346.6 norm2 -[{/section-2.1.4 xa}(The )2(-M)0( Option: Generate )(Makefile Dependencies)](2.1.4)327.4 subh3 -[(This option can be used to generate makefile dependencies on stdout. This can be redirected to a file for)]310.4 norm1 -[(further processing. For example:)]299.4 norm2 -[2(nasm -M myfile.asm > myfile.dep)]282.4 code3 -[{/section-2.1.5 xa}(The )2(-MG)0( Option: Generate )(Makefile Dependencies)](2.1.5)263.2 subh3 -[(This option can be used to generate makefile dependencies on stdout. This differs from the )2(-M)0( option in that if)]246.2 norm1 -[(a nonexisting file is encountered, it is assumed to be a generated file and is added to the dependency list)]235.2 norm0 -[(without a prefix.)]224.2 norm2 -[{/section-2.1.6 xa}(The )2(-MF)0( Option: Set Makefile Dependency File)](2.1.6)205 subh3 -[(This option can be used with the )2(-M)0( or )2(-MG)0( options to send the output to a file, rather than to stdout. For)]188 norm1 -[(example:)]177 norm2 -[2(nasm -M -MF myfile.dep myfile.asm)]160 code3 -[{/section-2.1.7 xa}(The )2(-MD)0( Option: Assemble and Generate Dependencies)](2.1.7)140.8 subh3 -[(The )2(-MD)0( option acts as the combination of the )2(-M)0( and )2(-MF)0( options \(i.e. a filename has to be specified.\))]123.8 norm1 -[(However, unlike the )2(-M)0( or )2(-MG)0( options, )2(-MD)0( does )1(not)0( inhibit the normal operation of the assembler. Use this)]112.8 norm0 -[(to automatically generate updated dependencies with every assembly session. For example:)]101.8 norm2 -(18)pageeven -restore showpage -%%Page: 19 19 -%%BeginPageSetup -save -%%EndPageSetup -/19 pa -[2(nasm -f elf -o myfile.o -MD myfile.dep myfile.asm)]681 code3 -[{/section-2.1.8 xa}(The )2(-MT)0( Option: Dependency Target Name)](2.1.8)661.8 subh3 -[(The )2(-MT)0( option can be used to override the default name of the dependency target. This is normally the same)]644.8 norm1 -[(as the output filename, specified by the )2(-o)0( option.)]633.8 norm2 -[{/section-2.1.9 xa}(The )2(-MQ)0( Option: Dependency Target Name \(Quoted\))](2.1.9)614.6 subh3 -[(The )2(-MQ)0( option acts as the )2(-MT)0( option, except it tries to quote characters that have special meaning in)]597.6 norm1 -[(Makefile syntax. This is not foolproof, as not all characters with special meaning are quotable in Make.)]586.6 norm2 -[{/section-2.1.10 xa}(The )2(-MP)0( Option: Emit phony targets)](2.1.10)567.4 subh3 -[(When used with any of the dependency generation options, the )2(-MP)0( option causes NASM to emit a phony)]550.4 norm1 -[(target without dependencies for each header file. This prevents Make from complaining if a header file has)]539.4 norm0 -[(been removed.)]528.4 norm2 -[{/section-2.1.11 xa}(The )2(-F)0( Option: Selecting a )(Debug Information Format)](2.1.11)509.2 subh3 -[(This option is used to select the format of the debug information emitted into the output file, to be used by a)]492.2 norm1 -[(debugger \(or )1(will)0( be\). Prior to version 2.03.01, the use of this switch did )1(not)0( enable output of the selected)]481.2 norm0 -[(debug info format. Use )2(-g)0(, see ){/section-2.1.12 xl}(section 2.1.12){el}(, to enable output. Versions 2.03.01 and later automatically)]470.2 norm0 -[(enable )2(-g)0( if )2(-F)0( is specified.)]459.2 norm2 -[(A complete list of the available debug file formats for an output format can be seen by issuing the command)]442.2 norm1 -[2(nasm -f <format> -y)0(. Not all output formats currently support debugging output. See ){/section-2.1.26 xl}(section 2.1.26){el}(.)]431.2 norm2 -[(This should not be confused with the )2(-f dbg)0( output format option which is not built into NASM by default.)]414.2 norm1 -[(For information on how to enable it when building from the sources, see ){/section-7.14 xl}(section 7.14){el}(.)]403.2 norm2 -[{/section-2.1.12 xa}(The )2(-g)0( Option: Enabling )(Debug Information.)](2.1.12)384 subh3 -[(This option can be used to generate debugging information in the specified format. See ){/section-2.1.11 xl}(section 2.1.11){el}(. Using)]367 norm1 -[2(-g)0( without )2(-F)0( results in emitting debug info in the default format, if any, for the selected output format. If no)]356 norm0 -[(debug information is currently implemented in the selected output format, )2(-g)0( is )1(silently ignored)0(.)]345 norm2 -[{/section-2.1.13 xa}(The )2(-X)0( Option: Selecting an )(Error Reporting Format)](2.1.13)325.8 subh3 -[(This option can be used to select an error reporting format for any error messages that might be produced by)]308.8 norm1 -[(NASM.)]297.8 norm2 -[(Currently, two error reporting formats may be selected. They are the )2(-Xvc)0( option and the )2(-Xgnu)0( option. The)]280.8 norm1 -[(GNU format is the default and looks like this:)]269.8 norm2 -[2(filename.asm:65: error: specific error message)]252.8 code3 -[(where )2(filename.asm)0( is the name of the source file in which the error was detected, )2(65)0( is the source file)]235.8 norm1 -[(line number on which the error was detected, )2(error)0( is the severity of the error \(this could be )2(warning)0(\),)]224.8 norm0 -[(and )2(specific error message)0( is a more detailed text message which should help pinpoint the exact)]213.8 norm0 -[(problem.)]202.8 norm2 -[(The other format, specified by )2(-Xvc)0( is the style used by Microsoft Visual C++ and some other programs. It)]185.8 norm1 -[(looks like this:)]174.8 norm2 -[2(filename.asm\(65\) : error: specific error message)]157.8 code3 -[(where the only difference is that the line number is in parentheses instead of being delimited by colons.)]140.8 norm3 -[(See also the )2(Visual C++)0( output format, ){/section-7.5 xl}(section 7.5){el}(.)]123.8 norm3 -(19)pageodd -restore showpage -%%Page: 20 20 -%%BeginPageSetup -save -%%EndPageSetup -/20 pa -[{/section-2.1.14 xa}(The )2(-Z)0( Option: Send Errors to a File)](2.1.14)678.8 subh3 -[(Under )2(MS-DOS)0( it can be difficult \(though there are ways\) to redirect the standard-error output of a program)]661.8 norm1 -[(to a file. Since NASM usually produces its warning and )(error messages on )2(stderr)0(, this can make it hard to)]650.8 norm0 -[(capture the errors if \(for example\) you want to load them into an editor.)]639.8 norm2 -[(NASM therefore provides the )2(-Z)0( option, taking a filename argument which causes errors to be sent to the)]622.8 norm1 -[(specified files rather than standard error. Therefore you can )(redirect the errors into a file by typing)]611.8 norm2 -[2(nasm -Z myfile.err -f obj myfile.asm)]594.8 code3 -[(In earlier versions of NASM, this option was called )2(-E)0(, but it was changed since )2(-E)0( is an option)]577.8 norm1 -[(conventionally used for preprocessing only, with disastrous results. See ){/section-2.1.20 xl}(section 2.1.20){el}(.)]566.8 norm2 -[{/section-2.1.15 xa}(The )2(-s)0( Option: Send Errors to )2(stdout)](2.1.15)547.6 subh3 -[(The )2(-s)0( option redirects )(error messages to )2(stdout)0( rather than )2(stderr)0(, so it can be redirected under)]530.6 norm1 -[2(MS-DOS)0(. To assemble the file )2(myfile.asm)0( and pipe its output to the )2(more)0( program, you can type:)]519.6 norm2 -[2(nasm -s -f obj myfile.asm | more)]502.6 code3 -[(See also the )2(-Z)0( option, ){/section-2.1.14 xl}(section 2.1.14){el}(.)]485.6 norm3 -[{/section-2.1.16 xa}(The )2(-i)0( Option: Include File Search Directories)](2.1.16)466.4 subh3 -[(When NASM sees the )2(%include)0( or )2(%pathsearch)0( directive in a source file \(see ){/section-4.6.1 xl}(section 4.6.1){el}(, ){/section-4.6.2 xl}(section){el}]449.4 norm1 -[{/section-4.6.2 xl}(4.6.2){el}( or ){/section-3.2.3 xl}(section 3.2.3){el}(\), it will search for the given file not only in the current directory, but also in any)]438.4 norm0 -[(directories specified on the command line by the use of the )2(-i)0( option. Therefore you can include files from a)]427.4 norm0 -[(macro library, for example, by typing)]416.4 norm2 -[2(nasm -ic:\\macrolib\\ -f obj myfile.asm)]399.4 code3 -[(\(As usual, a space between )2(-i)0( and the path name is allowed, and optional\).)]382.4 norm3 -[(NASM, in the interests of complete source-code portability, does not understand the file naming conventions)]365.4 norm1 -[(of the OS it is running on; the string you provide as an argument to the )2(-i)0( option will be prepended exactly as)]354.4 norm0 -[(written to the name of the include file. Therefore the trailing backslash in the above example is necessary.)]343.4 norm0 -[(Under Unix, a trailing forward slash is similarly necessary.)]332.4 norm2 -[(\(You can use this to your advantage, if you're really )(perverse, by noting that the option )2(-ifoo)0( will cause)]315.4 norm1 -[2(%include "bar.i")0( to search for the file )2(foobar.i)0(...\))]304.4 norm2 -[(If you want to define a )1(standard)0( )(include search path, similar to )2(/usr/include)0( on Unix systems, you)]287.4 norm1 -[(should place one or more )2(-i)0( directives in the )2(NASMENV)0( environment variable \(see ){/section-2.1.28 xl}(section 2.1.28){el}(\).)]276.4 norm2 -[(For Makefile compatibility with many C compilers, this option can also be specified as )2(-I)0(.)]259.4 norm3 -[{/section-2.1.17 xa}(The )2(-p)0( Option: )(Pre-Include a File)](2.1.17)240.2 subh3 -[(NASM allows you to specify files to be )1(pre-included)0( into your source file, by the use of the )2(-p)0( option. So)]223.2 norm1 -[(running)]212.2 norm2 -[2(nasm myfile.asm -p myinc.inc)]195.2 code3 -[(is equivalent to running )2(nasm myfile.asm)0( and placing the directive )2(%include "myinc.inc")0( at the)]178.2 norm1 -[(start of the file.)]167.2 norm2 -[(For consistency with the )2(-I)0(, )2(-D)0( and )2(-U)0( options, this option can also be specified as )2(-P)0(.)]150.2 norm3 -[{/section-2.1.18 xa}(The )2(-d)0( Option: )(Pre-Define a Macro)](2.1.18)131 subh3 -[(Just as the )2(-p)0( option gives an alternative to placing )2(%include)0( directives at the start of a source file, the )2(-d)]114 norm1 -[(option gives an alternative to placing a )2(%define)0( directive. You could code)]103 norm2 -(20)pageeven -restore showpage -%%Page: 21 21 -%%BeginPageSetup -save -%%EndPageSetup -/21 pa -[2(nasm myfile.asm -dFOO=100)]681 code3 -[(as an alternative to placing the directive)]664 norm3 -[2(%define FOO 100)]647 code3 -[(at the start of the file. You can miss off the macro value, as well: the option )2(-dFOO)0( is equivalent to coding)]630 norm1 -[2(%define FOO)0(. This form of the directive may be useful for selecting )(assembly-time options which are then)]619 norm0 -[(tested using )2(%ifdef)0(, for example )2(-dDEBUG)0(.)]608 norm2 -[(For Makefile compatibility with many C compilers, this option can also be specified as )2(-D)0(.)]591 norm3 -[{/section-2.1.19 xa}(The )2(-u)0( Option: )(Undefine a Macro)](2.1.19)571.8 subh3 -[(The )2(-u)0( option undefines a macro that would otherwise have been pre-defined, either automatically or by a)]554.8 norm1 -[2(-p)0( or )2(-d)0( option specified earlier on the command lines.)]543.8 norm2 -[(For example, the following command line:)]526.8 norm3 -[2(nasm myfile.asm -dFOO=100 -uFOO)]509.8 code3 -[(would result in )2(FOO)0( )1(not)0( being a predefined macro in the program. This is useful to override options specified)]492.8 norm1 -[(at a different point in a Makefile.)]481.8 norm2 -[(For Makefile compatibility with many C compilers, this option can also be specified as )2(-U)0(.)]464.8 norm3 -[{/section-2.1.20 xa}(The )2(-E)0( Option: Preprocess Only)](2.1.20)445.6 subh3 -[(NASM allows the )(preprocessor to be run on its own, up to a point. Using the )2(-E)0( option \(which requires no)]428.6 norm1 -[(arguments\) will cause NASM to preprocess its input file, expand all the macro references, remove all the)]417.6 norm0 -[(comments and preprocessor directives, and print the resulting file on standard output \(or save it to a file, if the)]406.6 norm0 -[2(-o)0( option is also used\).)]395.6 norm2 -[(This option cannot be applied to programs which require the preprocessor to evaluate )(expressions which)]378.6 norm1 -[(depend on the values of symbols: so code such as)]367.6 norm2 -[2(%assign tablesize \($-tablestart\))]350.6 code3 -[(will cause an error in )(preprocess-only mode.)]333.6 norm3 -[(For compatiblity with older version of NASM, this option can also be written )2(-e)0(. )2(-E)0( in older versions of)]316.6 norm1 -[(NASM was the equivalent of the current )2(-Z)0( option, ){/section-2.1.14 xl}(section 2.1.14){el}(.)]305.6 norm2 -[{/section-2.1.21 xa}(The )2(-a)0( Option: Don't Preprocess At All)](2.1.21)286.4 subh3 -[(If NASM is being used as the back end to a compiler, it might be desirable to )(suppress preprocessing)]269.4 norm1 -[(completely and assume the compiler has already done it, to save time and increase compilation speeds. The)]258.4 norm0 -[2(-a)0( option, requiring no argument, instructs NASM to replace its powerful )(preprocessor with a )(stub)]247.4 norm0 -[(preprocessor which does nothing.)]236.4 norm2 -[{/section-2.1.22 xa}(The )2(-O)0( Option: Specifying )(Multipass Optimization)](2.1.22)217.2 subh3 -[(NASM defaults to not optimizing operands which can fit into a signed byte. This means that if you want the)]200.2 norm1 -[(shortest possible object code, you have to enable optimization.)]189.2 norm2 -[(Using the )2(-O)0( option, you can tell NASM to carry out different levels of optimization. The syntax is:)]172.2 norm3 -[2(-O0)0(: No optimization. All operands take their long forms, if a short form is not specified, except)]155.2 bull1 -[(conditional jumps. This is intended to match NASM 0.98 behavior.)]144.2 bull2 -[2(-O1)0(: Minimal optimization. As above, but immediate operands which will fit in a signed byte are)]127.2 bull1 -[(optimized, unless the long form is specified. Conditional jumps default to the long form unless otherwise)]116.2 bull0 -[(specified.)]105.2 bull2 -(21)pageodd -restore showpage -%%Page: 22 22 -%%BeginPageSetup -save -%%EndPageSetup -/22 pa -[2(-Ox)0( \(where )2(x)0( is the actual letter )2(x)0(\): Multipass optimization. Minimize branch offsets and signed)]681 bull1 -[(immediate bytes, overriding size specification unless the )2(strict)0( keyword has been used \(see ){/section-3.7 xl}(section){el}]670 bull0 -[{/section-3.7 xl}(3.7){el}(\). For compatability with earlier releases, the letter )2(x)0( may also be any number greater than one. This)]659 bull0 -[(number has no effect on the actual number of passes.)]648 bull2 -[(The )2(-Ox)0( mode is recommended for most uses.)]631 norm3 -[(Note that this is a capital )2(O)0(, and is different from a small )2(o)0(, which is used to specify the output file name. See)]614 norm1 -[{/section-2.1.1 xl}(section 2.1.1){el}(.)]603 norm2 -[{/section-2.1.23 xa}(The )2(-t)0( Option: Enable TASM Compatibility Mode)](2.1.23)583.8 subh3 -[(NASM includes a limited form of compatibility with Borland's )2(TASM)0(. When NASM's )2(-t)0( option is used, the)]566.8 norm1 -[(following changes are made:)]555.8 norm2 -[(local labels may be prefixed with )2(@@)0( instead of )2(.)]538.8 bull3 -[(size override is supported within brackets. In TASM compatible mode, a size override inside square)]521.8 bull1 -[(brackets changes the size of the operand, and not the address type of the operand as it does in NASM)]510.8 bull0 -[(syntax. E.g. )2(mov eax,[DWORD val])0( is valid syntax in TASM compatibility mode. Note that you lose)]499.8 bull0 -[(the ability to override the default address type for the instruction.)]488.8 bull2 -[(unprefixed forms of some directives supported \()2(arg)0(, )2(elif)0(, )2(else)0(, )2(endif)0(, )2(if)0(, )2(ifdef)0(, )2(ifdifi)0(,)]471.8 bull1 -[2(ifndef)0(, )2(include)0(, )2(local)0(\))]460.8 bull2 -[{/section-2.1.24 xa}(The )2(-w)0( and )2(-W)0( Options: Enable or Disable Assembly )(Warnings)](2.1.24)441.6 subh3 -[(NASM can observe many conditions during the course of assembly which are worth mentioning to the user,)]424.6 norm1 -[(but not a sufficiently severe error to justify NASM refusing to generate an output file. These conditions are)]413.6 norm0 -[(reported like errors, but come up with the word `warning' before the message. Warnings do not prevent)]402.6 norm0 -[(NASM from generating an output file and returning a success status to the operating system.)]391.6 norm2 -[(Some conditions are even less severe than that: they are only sometimes worth mentioning to the user.)]374.6 norm1 -[(Therefore NASM supports the )2(-w)0( command-line option, which enables or disables certain classes of)]363.6 norm0 -[(assembly warning. Such warning classes are described by a name, for example )2(orphan-labels)0(; you can)]352.6 norm0 -[(enable warnings of this class by the command-line option )2(-w+orphan-labels)0( and disable it by)]341.6 norm0 -[2(-w-orphan-labels)0(.)]330.6 norm2 -[(The )(suppressible warning classes are:)]313.6 norm3 -[2(macro-params)0( covers warnings about )(multi-line macros being invoked with the wrong number of)]296.6 bull1 -[(parameters. This warning class is enabled by default; see ){/section-4.3.2 xl}(section 4.3.2){el}( for an example of why you might)]285.6 bull0 -[(want to disable it.)]274.6 bull2 -[2(macro-selfref)0( warns if a macro references itself. This warning class is disabled by default.)]257.6 bull3 -[2(macro-defaults)0( warns when a macro has more default parameters than optional parameters. This)]240.6 bull1 -[(warning class is enabled by default; see ){/section-4.3.5 xl}(section 4.3.5){el}( for why you might want to disable it.)]229.6 bull2 -[2(orphan-labels)0( covers warnings about source lines which contain no instruction but define a label)]212.6 bull1 -[(without a trailing colon. NASM warns about this somewhat obscure condition by default; see ){/section-3.1 xl}(section 3.1){el}]201.6 bull0 -[(for more information.)]190.6 bull2 -[2(number-overflow)0( covers warnings about numeric constants which don't fit in 64 bits. This warning)]173.6 bull1 -[(class is enabled by default.)]162.6 bull2 -[2(gnu-elf-extensions)0( warns if 8-bit or 16-bit relocations are used in )2(-f elf)0( format. The GNU)]145.6 bull1 -[(extensions allow this. This warning class is disabled by default.)]134.6 bull2 -[2(float-overflow)0( warns about floating point overflow. Enabled by default.)]117.6 bull3 -[2(float-denorm)0( warns about floating point denormals. Disabled by default.)]100.6 bull3 -(22)pageeven -restore showpage -%%Page: 23 23 -%%BeginPageSetup -save -%%EndPageSetup -/23 pa -[2(float-underflow)0( warns about floating point underflow. Disabled by default.)]681 bull3 -[2(float-toolong)0( warns about too many digits in floating-point numbers. Enabled by default.)]664 bull3 -[2(user)0( controls )2(%warning)0( directives \(see ){/section-4.9 xl}(section 4.9){el}(\). Enabled by default.)]647 bull3 -[2(error)0( causes warnings to be treated as errors. Disabled by default.)]630 bull3 -[2(all)0( is an alias for )1(all)0( suppressible warning classes \(not including )2(error)0(\). Thus, )2(-w+all)0( enables all)]613 bull1 -[(available warnings.)]602 bull2 -[(In addition, you can set warning classes across sections. Warning classes may be enabled with)]585 norm1 -[2([warning +warning-name])0(, disabled with )2([warning -warning-name])0( or reset to their original)]574 norm0 -[(value with )2([warning *warning-name])0(. No "user form" \(without the brackets\) exists.)]563 norm2 -[(Since version 2.00, NASM has also supported the gcc-like syntax )2(-Wwarning)0( and )2(-Wno-warning)]546 norm1 -[(instead of )2(-w+warning)0( and )2(-w-warning)0(, respectively.)]535 norm2 -[{/section-2.1.25 xa}(The )2(-v)0( Option: Display )(Version Info)](2.1.25)515.8 subh3 -[(Typing )2(NASM -v)0( will display the version of NASM which you are using, and the date on which it was)]498.8 norm1 -[(compiled.)]487.8 norm2 -[(You will need the version number if you report a bug.)]470.8 norm3 -[{/section-2.1.26 xa}(The )2(-y)0( Option: Display Available Debug Info Formats)](2.1.26)451.6 subh3 -[(Typing )2(nasm -f <option> -y)0( will display a list of the available debug info formats for the given output)]434.6 norm1 -[(format. The default format is indicated by an asterisk. For example:)]423.6 norm2 -[2(nasm -f elf -y)]406.6 code3 -[2(valid debug formats for 'elf32' output format are )]389.6 code1 -[2( \('*' denotes default\): )]378.6 code0 -[2( * stabs ELF32 \(i386\) stabs debug format for Linux )]367.6 code0 -[2( dwarf elf32 \(i386\) dwarf debug format for Linux)]356.6 code2 -[{/section-2.1.27 xa}(The )2(--prefix)0( and )2(--postfix)0( Options.)](2.1.27)337.4 subh3 -[(The )2(--prefix)0( and )2(--postfix)0( options prepend or append \(respectively\) the given argument to all)]320.4 norm1 -[2(global)0( or )2(extern)0( variables. E.g. )2(--prefix _)0( will prepend the underscore to all global and external)]309.4 norm0 -[(variables, as C sometimes \(but not always\) likes it.)]298.4 norm2 -[{/section-2.1.28 xa}(The )2(NASMENV)0( )(Environment Variable)](2.1.28)279.2 subh3 -[(If you define an environment variable called )2(NASMENV)0(, the program will interpret it as a list of extra)]262.2 norm1 -[(command-line options, which are processed before the real command line. You can use this to define)]251.2 norm0 -[(standard search directories for include files, by putting )2(-i)0( options in the )2(NASMENV)0( variable.)]240.2 norm2 -[(The value of the variable is split up at white space, so that the value )2(-s -ic:\\nasmlib\\)0( will be treated as)]223.2 norm1 -[(two separate options. However, that means that the value )2(-dNAME="my name")0( won't do what you might)]212.2 norm0 -[(want, because it will be split at the space and the NASM command-line processing will get confused by the)]201.2 norm0 -[(two nonsensical words )2(-dNAME="my)0( and )2(name")0(.)]190.2 norm2 -[(To get round this, NASM provides a feature whereby, if you begin the )2(NASMENV)0( environment variable with)]173.2 norm1 -[(some character that isn't a minus sign, then NASM will treat this character as the )(separator character for)]162.2 norm0 -[(options. So setting the )2(NASMENV)0( variable to the value )2(!-s!-ic:\\nasmlib\\)0( is equivalent to setting it to)]151.2 norm0 -[2(-s -ic:\\nasmlib\\)0(, but )2(!-dNAME="my name")0( will work.)]140.2 norm2 -[(This environment variable was previously called )2(NASM)0(. This was changed with version 0.98.31.)]123.2 norm3 -(23)pageodd -restore showpage -%%Page: 24 24 -%%BeginPageSetup -save -%%EndPageSetup -/24 pa -[{/section-2.2 xa}(Quick Start for )(MASM Users)](2.2)676.6 head3 -[(If you're used to writing programs with MASM, or with )(TASM in MASM-compatible \(non-Ideal\) mode, or)]659.6 norm1 -[(with )2(a86)0(, this section attempts to outline the major differences between MASM's syntax and NASM's. If)]648.6 norm0 -[(you're not already used to MASM, it's probably worth skipping this section.)]637.6 norm2 -[{/section-2.2.1 xa}(NASM Is )(Case-Sensitive)](2.2.1)618.4 subh3 -[(One simple difference is that NASM is case-sensitive. It makes a difference whether you call your label )2(foo)0(,)]601.4 norm1 -[2(Foo)0( or )2(FOO)0(. If you're assembling to )2(DOS)0( or )2(OS/2)0( )2(.OBJ)0( files, you can invoke the )2(UPPERCASE)0( directive)]590.4 norm0 -[(\(documented in ){/section-7.4 xl}(section 7.4){el}(\) to ensure that all symbols exported to other code modules are forced to be upper)]579.4 norm0 -[(case; but even then, )1(within)0( a single module, NASM will distinguish between labels differing only in case.)]568.4 norm2 -[{/section-2.2.2 xa}(NASM Requires )(Square Brackets For )(Memory References)](2.2.2)549.2 subh3 -[(NASM was designed with simplicity of syntax in mind. One of the )(design goals of NASM is that it should be)]532.2 norm1 -[(possible, as far as is practical, for the user to look at a single line of NASM code and tell what opcode is)]521.2 norm0 -[(generated by it. You can't do this in MASM: if you declare, for example,)]510.2 norm2 -[2(foo equ 1 )]493.2 code1 -[2(bar dw 2)]482.2 code2 -[(then the two lines of code)]465.2 norm3 -[2( mov ax,foo )]448.2 code1 -[2( mov ax,bar)]437.2 code2 -[(generate completely different opcodes, despite having identical-looking syntaxes.)]420.2 norm3 -[(NASM avoids this undesirable situation by having a much simpler syntax for memory references. The rule is)]403.2 norm1 -[(simply that any access to the )1(contents)0( of a memory location requires square brackets around the address, and)]392.2 norm0 -[(any access to the )1(address)0( of a variable doesn't. So an instruction of the form )2(mov ax,foo)0( will )1(always)0( refer)]381.2 norm0 -[(to a compile-time constant, whether it's an )2(EQU)0( or the address of a variable; and to access the )1(contents)0( of the)]370.2 norm0 -[(variable )2(bar)0(, you must code )2(mov ax,[bar])0(.)]359.2 norm2 -[(This also means that NASM has no need for MASM's )2(OFFSET)0( keyword, since the MASM code)]342.2 norm1 -[2(mov ax,offset bar)0( means exactly the same thing as NASM's )2(mov ax,bar)0(. If you're trying to get)]331.2 norm0 -[(large amounts of MASM code to assemble sensibly under NASM, you can always code)]320.2 norm0 -[2(%idefine offset)0( to make the preprocessor treat the )2(OFFSET)0( keyword as a no-op.)]309.2 norm2 -[(This issue is even more confusing in )2(a86)0(, where declaring a label with a trailing colon defines it to be a)]292.2 norm1 -[(`label' as opposed to a `variable' and causes )2(a86)0( to adopt NASM-style semantics; so in )2(a86)0(, )2(mov ax,var)]281.2 norm0 -[(has different behaviour depending on whether )2(var)0( was declared as )2(var: dw 0)0( \(a label\) or )2(var dw 0)0( \(a)]270.2 norm0 -[(word-size variable\). NASM is very simple by comparison: )1(everything)0( is a label.)]259.2 norm2 -[(NASM, in the interests of simplicity, also does not support the )(hybrid syntaxes supported by MASM and its)]242.2 norm1 -[(clones, such as )2(mov ax,table[bx])0(, where a memory reference is denoted by one portion outside square)]231.2 norm0 -[(brackets and another portion inside. The correct syntax for the above is )2(mov ax,[table+bx])0(. Likewise,)]220.2 norm0 -[2(mov ax,es:[di])0( is wrong and )2(mov ax,[es:di])0( is right.)]209.2 norm2 -[{/section-2.2.3 xa}(NASM Doesn't Store )(Variable Types)](2.2.3)190 subh3 -[(NASM, by design, chooses not to remember the types of variables you declare. Whereas MASM will)]173 norm1 -[(remember, on seeing )2(var dw 0)0(, that you declared )2(var)0( as a word-size variable, and will then be able to fill)]162 norm0 -[(in the )(ambiguity in the size of the instruction )2(mov var,2)0(, NASM will deliberately remember nothing about)]151 norm0 -[(the symbol )2(var)0( except where it begins, and so you must explicitly code )2(mov word [var],2)0(.)]140 norm2 -[(For this reason, NASM doesn't support the )2(LODS)0(, )2(MOVS)0(, )2(STOS)0(, )2(SCAS)0(, )2(CMPS)0(, )2(INS)0(, or )2(OUTS)0( instructions,)]123 norm1 -[(but only supports the forms such as )2(LODSB)0(, )2(MOVSW)0(, and )2(SCASD)0(, which explicitly specify the size of the)]112 norm0 -[(components of the strings being manipulated.)]101 norm2 -(24)pageeven -restore showpage -%%Page: 25 25 -%%BeginPageSetup -save -%%EndPageSetup -/25 pa -[{/section-2.2.4 xa}(NASM Doesn't )2(ASSUME)](2.2.4)678.8 subh3 -[(As part of NASM's drive for simplicity, it also does not support the )2(ASSUME)0( directive. NASM will not keep)]661.8 norm1 -[(track of what values you choose to put in your segment registers, and will never )1(automatically)0( generate a)]650.8 norm0 -[(segment override prefix.)]639.8 norm2 -[{/section-2.2.5 xa}(NASM Doesn't Support )(Memory Models)](2.2.5)620.6 subh3 -[(NASM also does not have any directives to support different 16-bit memory models. The programmer has to)]603.6 norm1 -[(keep track of which functions are supposed to be called with a )(far call and which with a )(near call, and is)]592.6 norm0 -[(responsible for putting the correct form of )2(RET)0( instruction \()2(RETN)0( or )2(RETF)0(; NASM accepts )2(RET)0( itself as an)]581.6 norm0 -[(alternate form for )2(RETN)0(\); in addition, the programmer is responsible for coding CALL FAR instructions)]570.6 norm0 -[(where necessary when calling )1(external)0( functions, and must also keep track of which external variable)]559.6 norm0 -[(definitions are far and which are near.)]548.6 norm2 -[{/section-2.2.6 xa}(Floating-Point Differences)](2.2.6)529.4 subh3 -[(NASM uses different names to refer to floating-point registers from MASM: where MASM would call them)]512.4 norm1 -[2(ST\(0\))0(, )2(ST\(1\))0( and so on, and )2(a86)0( would call them simply )2(0)0(, )2(1)0( and so on, NASM chooses to call them)]501.4 norm0 -[2(st0)0(, )2(st1)0( etc.)]490.4 norm2 -[(As of version 0.96, NASM now treats the instructions with )(`nowait' forms in the same way as)]473.4 norm1 -[(MASM-compatible assemblers. The idiosyncratic treatment employed by 0.95 and earlier was based on a)]462.4 norm0 -[(misunderstanding by the authors.)]451.4 norm2 -[{/section-2.2.7 xa}(Other Differences)](2.2.7)432.2 subh3 -[(For historical reasons, NASM uses the keyword )2(TWORD)0( where MASM and compatible assemblers use)]415.2 norm1 -[2(TBYTE)0(.)]404.2 norm2 -[(NASM does not declare )(uninitialized storage in the same way as MASM: where a MASM programmer might)]387.2 norm1 -[(use )2(stack db 64 dup \(?\))0(, NASM requires )2(stack resb 64)0(, intended to be read as `reserve 64)]376.2 norm0 -[(bytes'. For a limited amount of compatibility, since NASM treats )2(?)0( as a valid character in symbol names, you)]365.2 norm0 -[(can code )2(? equ 0)0( and then writing )2(dw ?)0( will at least do something vaguely useful. )2(DUP)0( is still not a)]354.2 norm0 -[(supported syntax, however.)]343.2 norm2 -[(In addition to all of this, macros and directives work completely differently to MASM. See ){/chapter-4 xl}(chapter 4){el}( and)]326.2 norm1 -[{/chapter-6 xl}(chapter 6){el}( for further details.)]315.2 norm2 -(25)pageodd -restore showpage -%%Page: 26 26 -%%BeginPageSetup -save -%%EndPageSetup -/26 pa -[{/chapter-3 xa}(Chapter 3: The NASM Language)]642.8 chap3 -[{/section-3.1 xa}(Layout of a NASM Source Line)](3.1)603.4 head3 -[(Like most assemblers, each NASM source line contains \(unless it is a macro, a preprocessor directive or an)]586.4 norm1 -[(assembler directive: see ){/chapter-4 xl}(chapter 4){el}( and ){/chapter-6 xl}(chapter 6){el}(\) some combination of the four fields)]575.4 norm2 -[2(label: instruction operands ; comment)]558.4 code3 -[(As usual, most of these fields are optional; the presence or absence of any combination of a label, an)]541.4 norm1 -[(instruction and a comment is allowed. Of course, the operand field is either required or forbidden by the)]530.4 norm0 -[(presence and nature of the instruction field.)]519.4 norm2 -[(NASM uses backslash \(\\\) as the line continuation character; if a line ends with backslash, the next line is)]502.4 norm1 -[(considered to be a part of the backslash-ended line.)]491.4 norm2 -[(NASM places no restrictions on white space within a line: labels may have white space before them, or)]474.4 norm1 -[(instructions may have no space before them, or anything. The )(colon after a label is also optional. \(Note that)]463.4 norm0 -[(this means that if you intend to code )2(lodsb)0( alone on a line, and type )2(lodab)0( by accident, then that's still a)]452.4 norm0 -[(valid source line which does nothing but define a label. Running NASM with the command-line option)]441.4 norm0 -[2(-w+orphan-labels)0( will cause it to warn you if you define a label alone on a line without a )(trailing colon.\))]430.4 norm2 -[(Valid characters in labels are letters, numbers, )2(_)0(, )2($)0(, )2(#)0(, )2(@)0(, )2(~)0(, )2(.)0(, and )2(?)0(. The only characters which may be used)]413.4 norm1 -[(as the )1(first)0( character of an identifier are letters, )2(.)0( \(with special meaning: see ){/section-3.9 xl}(section 3.9){el}(\), )2(_)0( and )2(?)0(. An)]402.4 norm0 -[(identifier may also be prefixed with a )2($)0( to indicate that it is intended to be read as an identifier and not a)]391.4 norm0 -[(reserved word; thus, if some other module you are linking with defines a symbol called )2(eax)0(, you can refer to)]380.4 norm0 -[2($eax)0( in NASM code to distinguish the symbol from the register. Maximum length of an identifier is 4095)]369.4 norm0 -[(characters.)]358.4 norm2 -[(The instruction field may contain any machine instruction: Pentium and P6 instructions, FPU instructions,)]341.4 norm1 -[(MMX instructions and even undocumented instructions are all supported. The instruction may be prefixed by)]330.4 norm0 -[2(LOCK)0(, )2(REP)0(, )2(REPE)0(/)2(REPZ)0( or )2(REPNE)0(/)2(REPNZ)0(, in the usual way. Explicit )(address-size and )(operand-size)]319.4 norm0 -[(prefixes )2(A16)0(, )2(A32)0(, )2(A64)0(, )2(O16)0( and )2(O32)0(, )2(O64)0( are provided \226 one example of their use is given in ){/chapter-10 xl}(chapter 10){el}(.)]308.4 norm0 -[(You can also use the name of a )(segment register as an instruction prefix: coding )2(es mov [bx],ax)0( is)]297.4 norm0 -[(equivalent to coding )2(mov [es:bx],ax)0(. We recommend the latter syntax, since it is consistent with other)]286.4 norm0 -[(syntactic features of the language, but for instructions such as )2(LODSB)0(, which has no operands and yet can)]275.4 norm0 -[(require a segment override, there is no clean syntactic way to proceed apart from )2(es lodsb)0(.)]264.4 norm2 -[(An instruction is not required to use a prefix: prefixes such as )2(CS)0(, )2(A32)0(, )2(LOCK)0( or )2(REPE)0( can appear on a line)]247.4 norm1 -[(by themselves, and NASM will just generate the prefix bytes.)]236.4 norm2 -[(In addition to actual machine instructions, NASM also supports a number of pseudo-instructions, described in)]219.4 norm1 -[{/section-3.2 xl}(section 3.2){el}(.)]208.4 norm2 -[(Instruction )(operands may take a number of forms: they can be registers, described simply by the register name)]191.4 norm1 -[(\(e.g. )2(ax)0(, )2(bp)0(, )2(ebx)0(, )2(cr0)0(: NASM does not use the )2(gas)0(\226style syntax in which register names must be prefixed)]180.4 norm0 -[(by a )2(%)0( sign\), or they can be )(effective addresses \(see ){/section-3.3 xl}(section 3.3){el}(\), constants \(){/section-3.4 xl}(section 3.4){el}(\) or expressions)]169.4 norm0 -[(\(){/section-3.5 xl}(section 3.5){el}(\).)]158.4 norm2 -[(For x87 )(floating-point instructions, NASM accepts a wide range of syntaxes: you can use two-operand forms)]141.4 norm1 -[(like MASM supports, or you can use NASM's native single-operand forms in most cases. For example, you)]130.4 norm0 -[(can code:)]119.4 norm2 -(26)pageeven -restore showpage -%%Page: 27 27 -%%BeginPageSetup -save -%%EndPageSetup -/27 pa -[2( fadd st1 ; this sets st0 := st0 + st1 )]681 code1 -[2( fadd st0,st1 ; so does this )]670 code0 -[2()]659 code0 -[2( fadd st1,st0 ; this sets st1 := st1 + st0 )]648 code0 -[2( fadd to st1 ; so does this)]637 code2 -[(Almost any x87 floating-point instruction that references memory must use one of the prefixes )2(DWORD)0(,)]620 norm1 -[2(QWORD)0( or )2(TWORD)0( to indicate what size of )(memory operand it refers to.)]609 norm2 -[{/section-3.2 xa}(Pseudo-Instructions)](3.2)587.6 head3 -[(Pseudo-instructions are things which, though not real x86 machine instructions, are used in the instruction)]570.6 norm1 -[(field anyway because that's the most convenient place to put them. The current pseudo-instructions are )2(DB)0(,)]559.6 norm0 -[2(DW)0(, )2(DD)0(, )2(DQ)0(, )2(DT)0(, )2(DO)0( and )2(DY)0(; their )(uninitialized counterparts )2(RESB)0(, )2(RESW)0(, )2(RESD)0(, )2(RESQ)0(, )2(REST)0(, )2(RESO)0( and)]548.6 norm0 -[2(RESY)0(; the )2(INCBIN)0( command, the )2(EQU)0( command, and the )2(TIMES)0( prefix.)]537.6 norm2 -[{/section-3.2.1 xa}2(DB)0( and Friends: Declaring Initialized Data)](3.2.1)518.4 subh3 -[2(DB)0(, )2(DW)0(, )2(DD)0(, )2(DQ)0(, )2(DT)0(, )2(DO)0( and )2(DY)0( are used, much as in MASM, to declare initialized data in the output file.)]501.4 norm1 -[(They can be invoked in a wide range of ways: )]490.4 norm2 -[2( db 0x55 ; just the byte 0x55 )]473.4 code1 -[2( db 0x55,0x56,0x57 ; three bytes in succession )]462.4 code0 -[2( db 'a',0x55 ; character constants are OK )]451.4 code0 -[2( db 'hello',13,10,'$' ; so are string constants )]440.4 code0 -[2( dw 0x1234 ; 0x34 0x12 )]429.4 code0 -[2( dw 'a' ; 0x61 0x00 \(it's just a number\) )]418.4 code0 -[2( dw 'ab' ; 0x61 0x62 \(character constant\) )]407.4 code0 -[2( dw 'abc' ; 0x61 0x62 0x63 0x00 \(string\) )]396.4 code0 -[2( dd 0x12345678 ; 0x78 0x56 0x34 0x12 )]385.4 code0 -[2( dd 1.234567e20 ; floating-point constant )]374.4 code0 -[2( dq 0x123456789abcdef0 ; eight byte constant )]363.4 code0 -[2( dq 1.234567e20 ; double-precision float )]352.4 code0 -[2( dt 1.234567e20 ; extended-precision float)]341.4 code2 -[2(DT)0(, )2(DO)0( and )2(DY)0( do not accept )(numeric constants as operands.)]324.4 norm3 -[{/section-3.2.2 xa}2(RESB)0( and Friends: Declaring )(Uninitialized Data)](3.2.2)305.2 subh3 -[2(RESB)0(, )2(RESW)0(, )2(RESD)0(, )2(RESQ)0(, )2(REST)0(, )2(RESO)0( and )2(RESY)0( are designed to be used in the BSS section of a module:)]288.2 norm1 -[(they declare )1(uninitialized)0( storage space. Each takes a single operand, which is the number of bytes, words,)]277.2 norm0 -[(doublewords or whatever to reserve. As stated in ){/section-2.2.7 xl}(section 2.2.7){el}(, NASM does not support the MASM/TASM)]266.2 norm0 -[(syntax of reserving uninitialized space by writing )2(DW ?)0( or similar things: this is what it does instead. The)]255.2 norm0 -[(operand to a )2(RESB)0(\226type pseudo-instruction is a )1(critical expression)0(: see ){/section-3.8 xl}(section 3.8){el}(.)]244.2 norm2 -[(For example:)]227.2 norm3 -[2(buffer: resb 64 ; reserve 64 bytes )]210.2 code1 -[2(wordvar: resw 1 ; reserve a word )]199.2 code0 -[2(realarray resq 10 ; array of ten reals )]188.2 code0 -[2(ymmval: resy 1 ; one YMM register)]177.2 code2 -[{/section-3.2.3 xa}2(INCBIN)0(: Including External )(Binary Files)](3.2.3)158 subh3 -[2(INCBIN)0( is borrowed from the old Amiga assembler )(DevPac: it includes a binary file verbatim into the output)]141 norm1 -[(file. This can be handy for \(for example\) including )(graphics and )(sound data directly into a game executable)]130 norm0 -[(file. It can be called in one of these three ways:)]119 norm2 -(27)pageodd -restore showpage -%%Page: 28 28 -%%BeginPageSetup -save -%%EndPageSetup -/28 pa -[2( incbin "file.dat" ; include the whole file )]681 code1 -[2( incbin "file.dat",1024 ; skip the first 1024 bytes )]670 code0 -[2( incbin "file.dat",1024,512 ; skip the first 1024, and )]659 code0 -[2( ; actually include at most 512)]648 code2 -[2(INCBIN)0( is both a directive and a standard macro; the standard macro version searches for the file in the)]631 norm1 -[(include file search path and adds the file to the dependency lists. This macro can be overridden if desired.)]620 norm2 -[{/section-3.2.4 xa}2(EQU)0(: Defining Constants)](3.2.4)600.8 subh3 -[2(EQU)0( defines a symbol to a given constant value: when )2(EQU)0( is used, the source line must contain a label. The)]583.8 norm1 -[(action of )2(EQU)0( is to define the given label name to the value of its \(only\) operand. This definition is absolute,)]572.8 norm0 -[(and cannot change later. So, for example,)]561.8 norm2 -[2(message db 'hello, world' )]544.8 code1 -[2(msglen equ $-message)]533.8 code2 -[(defines )2(msglen)0( to be the constant 12. )2(msglen)0( may not then be redefined later. This is not a )(preprocessor)]516.8 norm1 -[(definition either: the value of )2(msglen)0( is evaluated )1(once)0(, using the value of )2($)0( \(see ){/section-3.5 xl}(section 3.5){el}( for an)]505.8 norm0 -[(explanation of )2($)0(\) at the point of definition, rather than being evaluated wherever it is referenced and using the)]494.8 norm0 -[(value of )2($)0( at the point of reference.)]483.8 norm2 -[{/section-3.2.5 xa}2(TIMES)0(: )(Repeating Instructions or Data)](3.2.5)464.6 subh3 -[(The )2(TIMES)0( prefix causes the instruction to be assembled multiple times. This is partly present as NASM's)]447.6 norm1 -[(equivalent of the )2(DUP)0( syntax supported by )(MASM\226compatible assemblers, in that you can code)]436.6 norm2 -[2(zerobuf: times 64 db 0)]419.6 code3 -[(or similar things; but )2(TIMES)0( is more versatile than that. The argument to )2(TIMES)0( is not just a numeric)]402.6 norm1 -[(constant, but a numeric )1(expression)0(, so you can do things like)]391.6 norm2 -[2(buffer: db 'hello, world' )]374.6 code1 -[2( times 64-$+buffer db ' ')]363.6 code2 -[(which will store exactly enough spaces to make the total length of )2(buffer)0( up to 64. Finally, )2(TIMES)0( can be)]346.6 norm1 -[(applied to ordinary instructions, so you can code trivial )(unrolled loops in it:)]335.6 norm2 -[2( times 100 movsb)]318.6 code3 -[(Note that there is no effective difference between )2(times 100 resb 1)0( and )2(resb 100)0(, except that the)]301.6 norm1 -[(latter will be assembled about 100 times faster due to the internal structure of the assembler.)]290.6 norm2 -[(The operand to )2(TIMES)0( is a critical expression \(){/section-3.8 xl}(section 3.8){el}(\).)]273.6 norm3 -[(Note also that )2(TIMES)0( can't be applied to )(macros: the reason for this is that )2(TIMES)0( is processed after the)]256.6 norm1 -[(macro phase, which allows the argument to )2(TIMES)0( to contain expressions such as )2(64-$+buffer)0( as above.)]245.6 norm0 -[(To repeat more than one line of code, or a complex macro, use the preprocessor )2(%rep)0( directive.)]234.6 norm2 -[{/section-3.3 xa}(Effective Addresses)](3.3)213.2 head3 -[(An )(effective address is any operand to an instruction which )(references memory. Effective addresses, in)]196.2 norm1 -[(NASM, have a very simple syntax: they consist of an expression evaluating to the desired address, enclosed in)]185.2 norm0 -[(square brackets. For example:)]174.2 norm2 -[2(wordvar dw 123 )]157.2 code1 -[2( mov ax,[wordvar] )]146.2 code0 -[2( mov ax,[wordvar+1] )]135.2 code0 -[2( mov ax,[es:wordvar+bx])]124.2 code2 -(28)pageeven -restore showpage -%%Page: 29 29 -%%BeginPageSetup -save -%%EndPageSetup -/29 pa -[(Anything not conforming to this simple system is not a valid memory reference in NASM, for example)]681 norm1 -[2(es:wordvar[bx])0(.)]670 norm2 -[(More complicated effective addresses, such as those involving more than one register, work in exactly the)]653 norm1 -[(same way:)]642 norm2 -[2( mov eax,[ebx*2+ecx+offset] )]625 code1 -[2( mov ax,[bp+di+8])]614 code2 -[(NASM is capable of doing )(algebra on these effective addresses, so that things which don't necessarily )1(look)]597 norm1 -[(legal are perfectly all right:)]586 norm2 -[2( mov eax,[ebx*5] ; assembles as [ebx*4+ebx] )]569 code1 -[2( mov eax,[label1*2-label2] ; ie [label1+\(label1-label2\)])]558 code2 -[(Some forms of effective address have more than one assembled form; in most such cases NASM will generate)]541 norm1 -[(the smallest form it can. For example, there are distinct assembled forms for the 32-bit effective addresses)]530 norm0 -[2([eax*2+0])0( and )2([eax+eax])0(, and NASM will generally generate the latter on the grounds that the former)]519 norm0 -[(requires four bytes to store a zero offset.)]508 norm2 -[(NASM has a hinting mechanism which will cause )2([eax+ebx])0( and )2([ebx+eax])0( to generate different)]491 norm1 -[(opcodes; this is occasionally useful because )2([esi+ebp])0( and )2([ebp+esi])0( have different default segment)]480 norm0 -[(registers.)]469 norm2 -[(However, you can force NASM to generate an effective address in a particular form by the use of the)]452 norm1 -[(keywords )2(BYTE)0(, )2(WORD)0(, )2(DWORD)0( and )2(NOSPLIT)0(. If you need )2([eax+3])0( to be assembled using a double-word)]441 norm0 -[(offset field instead of the one byte NASM will normally generate, you can code )2([dword eax+3])0(.)]430 norm0 -[(Similarly, you can force NASM to use a byte offset for a small value which it hasn't seen on the first pass \(see)]419 norm0 -[{/section-3.8 xl}(section 3.8){el}( for an example of such a code fragment\) by using )2([byte eax+offset])0(. As special cases,)]408 norm0 -[2([byte eax])0( will code )2([eax+0])0( with a byte offset of zero, and )2([dword eax])0( will code it with a)]397 norm0 -[(double-word offset of zero. The normal form, )2([eax])0(, will be coded with no offset field.)]386 norm2 -[(The form described in the previous paragraph is also useful if you are trying to access data in a 32-bit)]369 norm1 -[(segment from within 16 bit code. For more information on this see the section on mixed-size addressing)]358 norm0 -[(\(){/section-10.2 xl}(section 10.2){el}(\). In particular, if you need to access data with a known offset that is larger than will fit in a)]347 norm0 -[(16-bit value, if you don't specify that it is a dword offset, nasm will cause the high word of the offset to be)]336 norm0 -[(lost.)]325 norm2 -[(Similarly, NASM will split )2([eax*2])0( into )2([eax+eax])0( because that allows the offset field to be absent and)]308 norm1 -[(space to be saved; in fact, it will also split )2([eax*2+offset])0( into )2([eax+eax+offset])0(. You can combat)]297 norm0 -[(this behaviour by the use of the )2(NOSPLIT)0( keyword: )2([nosplit eax*2])0( will force )2([eax*2+0])0( to be)]286 norm0 -[(generated literally.)]275 norm2 -[(In 64-bit mode, NASM will by default generate absolute addresses. The )2(REL)0( keyword makes it produce)]258 norm1 -[2(RIP)0(\226relative addresses. Since this is frequently the normally desired behaviour, see the )2(DEFAULT)0( directive)]247 norm0 -[(\(){/section-6.2 xl}(section 6.2){el}(\). The keyword )2(ABS)0( overrides )2(REL)0(.)]236 norm2 -[{/section-3.4 xa}(Constants)](3.4)214.6 head3 -[(NASM understands four different types of constant: numeric, character, string and floating-point.)]197.6 norm3 -[{/section-3.4.1 xa}(Numeric Constants)](3.4.1)178.4 subh3 -[(A numeric constant is simply a number. NASM allows you to specify numbers in a variety of number bases,)]161.4 norm1 -[(in a variety of ways: you can suffix )2(H)0( or )2(X)0(, )2(Q)0( or )2(O)0(, and )2(B)0( for )(hexadecimal, )(octal and )(binary respectively, or)]150.4 norm0 -[(you can prefix )2(0x)0( for hexadecimal in the style of C, or you can prefix )2($)0( for hexadecimal in the style of)]139.4 norm0 -[(Borland Pascal. Note, though, that the )2($)0( prefix does double duty as a prefix on identifiers \(see ){/section-3.1 xl}(section 3.1){el}(\), so)]128.4 norm0 -[(a hex number prefixed with a )2($)0( sign must have a digit after the )2($)0( rather than a letter. In addition, current)]117.4 norm0 -(29)pageodd -restore showpage -%%Page: 30 30 -%%BeginPageSetup -save -%%EndPageSetup -/30 pa -[(versions of NASM accept the prefix )2(0h)0( for hexadecimal, )2(0o)0( or )2(0q)0( for octal, and )2(0b)0( for binary. Please note)]681 norm0 -[(that unlike C, a )2(0)0( prefix by itself does )1(not)0( imply an octal constant!)]670 norm2 -[(Numeric constants can have underscores \()2(_)0(\) interspersed to break up long strings.)]653 norm3 -[(Some examples \(all producing exactly the same code\):)]636 norm3 -[2( mov ax,200 ; decimal )]619 code1 -[2( mov ax,0200 ; still decimal )]608 code0 -[2( mov ax,0200d ; explicitly decimal )]597 code0 -[2( mov ax,0d200 ; also decimal )]586 code0 -[2( mov ax,0c8h ; hex )]575 code0 -[2( mov ax,$0c8 ; hex again: the 0 is required )]564 code0 -[2( mov ax,0xc8 ; hex yet again )]553 code0 -[2( mov ax,0hc8 ; still hex )]542 code0 -[2( mov ax,310q ; octal )]531 code0 -[2( mov ax,310o ; octal again )]520 code0 -[2( mov ax,0o310 ; octal yet again )]509 code0 -[2( mov ax,0q310 ; hex yet again )]498 code0 -[2( mov ax,11001000b ; binary )]487 code0 -[2( mov ax,1100_1000b ; same binary constant )]476 code0 -[2( mov ax,0b1100_1000 ; same binary constant yet again)]465 code2 -[{/section-3.4.2 xa}(Character Strings)](3.4.2)445.8 subh3 -[(A character string consists of up to eight characters enclosed in either single quotes \()2('...')0(\), double quotes)]428.8 norm1 -[(\()2("...")0(\) or backquotes \()2(`...`)0(\). Single or double quotes are equivalent to NASM \(except of course that)]417.8 norm0 -[(surrounding the constant with single quotes allows double quotes to appear within it and vice versa\); the)]406.8 norm0 -[(contents of those are represented verbatim. Strings enclosed in backquotes support C-style )2(\\)0(\226escapes for)]395.8 norm0 -[(special characters.)]384.8 norm2 -[(The following )(escape sequences are recognized by backquoted strings:)]367.8 norm3 -[2( \\' single quote \('\) )]350.8 code1 -[2( \\" double quote \("\) )]339.8 code0 -[2( \\` backquote \(`\) )]328.8 code0 -[2( \\\\ backslash \(\\\) )]317.8 code0 -[2( \\? question mark \(?\) )]306.8 code0 -[2( \\a BEL \(ASCII 7\) )]295.8 code0 -[2( \\b BS \(ASCII 8\) )]284.8 code0 -[2( \\t TAB \(ASCII 9\) )]273.8 code0 -[2( \\n LF \(ASCII 10\) )]262.8 code0 -[2( \\v VT \(ASCII 11\) )]251.8 code0 -[2( \\f FF \(ASCII 12\) )]240.8 code0 -[2( \\r CR \(ASCII 13\) )]229.8 code0 -[2( \\e ESC \(ASCII 27\) )]218.8 code0 -[2( \\377 Up to 3 octal digits - literal byte )]207.8 code0 -[2( \\xFF Up to 2 hexadecimal digits - literal byte )]196.8 code0 -[2( \\u1234 4 hexadecimal digits - Unicode character )]185.8 code0 -[2( \\U12345678 8 hexadecimal digits - Unicode character)]174.8 code2 -[(All other escape sequences are reserved. Note that )2(\\0)0(, meaning a )2(NUL)0( character \(ASCII 0\), is a special case)]157.8 norm1 -[(of the octal escape sequence.)]146.8 norm2 -[(Unicode characters specified with )2(\\u)0( or )2(\\U)0( are converted to )(UTF-8. For example, the following lines are all)]129.8 norm1 -[(equivalent:)]118.8 norm2 -(30)pageeven -restore showpage -%%Page: 31 31 -%%BeginPageSetup -save -%%EndPageSetup -/31 pa -[2( db `\\u263a` ; UTF-8 smiley face )]681 code1 -[2( db `\\xe2\\x98\\xba` ; UTF-8 smiley face )]670 code0 -[2( db 0E2h, 098h, 0BAh ; UTF-8 smiley face)]659 code2 -[{/section-3.4.3 xa}(Character Constants)](3.4.3)639.8 subh3 -[(A character constant consists of a string up to eight bytes long, used in an expression context. It is treated as if)]622.8 norm1 -[(it was an integer.)]611.8 norm2 -[(A character constant with more than one byte will be arranged with )(little-endian order in mind: if you code)]594.8 norm3 -[2( mov eax,'abcd')]577.8 code3 -[(then the constant generated is not )2(0x61626364)0(, but )2(0x64636261)0(, so that if you were then to store the)]560.8 norm1 -[(value into memory, it would read )2(abcd)0( rather than )2(dcba)0(. This is also the sense of character constants)]549.8 norm0 -[(understood by the Pentium's )2(CPUID)0( instruction.)]538.8 norm2 -[{/section-3.4.4 xa}(String Constants)](3.4.4)519.6 subh3 -[(String constants are character strings used in the context of some pseudo-instructions, namely the )2(DB)0( family)]502.6 norm1 -[(and )2(INCBIN)0( \(where it represents a filename.\) They are also used in certain preprocessor directives.)]491.6 norm2 -[(A string constant looks like a character constant, only longer. It is treated as a concatenation of)]474.6 norm1 -[(maximum-size character constants for the conditions. So the following are equivalent:)]463.6 norm2 -[2( db 'hello' ; string constant )]446.6 code1 -[2( db 'h','e','l','l','o' ; equivalent character constants)]435.6 code2 -[(And the following are also equivalent:)]418.6 norm3 -[2( dd 'ninechars' ; doubleword string constant )]401.6 code1 -[2( dd 'nine','char','s' ; becomes three doublewords )]390.6 code0 -[2( db 'ninechars',0,0,0 ; and really looks like this)]379.6 code2 -[(Note that when used in a string-supporting context, quoted strings are treated as a string constants even if)]362.6 norm1 -[(they are short enough to be a character constant, because otherwise )2(db 'ab')0( would have the same effect as)]351.6 norm0 -[2(db 'a')0(, which would be silly. Similarly, three-character or four-character constants are treated as strings)]340.6 norm0 -[(when they are operands to )2(DW)0(, and so forth.)]329.6 norm2 -[{/section-3.4.5 xa}(Unicode Strings)](3.4.5)310.4 subh3 -[(The special operators )2(__utf16__)0( and )2(__utf32__)0( allows definition of Unicode strings. They take a string)]293.4 norm1 -[(in UTF-8 format and converts it to \(littleendian\) UTF-16 or UTF-32, respectively.)]282.4 norm2 -[(For example:)]265.4 norm3 -[2(%define u\(x\) __utf16__\(x\) )]248.4 code1 -[2(%define w\(x\) __utf32__\(x\) )]237.4 code0 -[2()]226.4 code0 -[2( dw u\('C:\\WINDOWS'\), 0 ; Pathname in UTF-16 )]215.4 code0 -[2( dd w\(`A + B = \\u206a`\), 0 ; String in UTF-32)]204.4 code2 -[2(__utf16__)0( and )2(__utf32__)0( can be applied either to strings passed to the )2(DB)0( family instructions, or to)]187.4 norm1 -[(character constants in an expression context.)]176.4 norm2 -[{/section-3.4.6 xa}(Floating-Point Constants)](3.4.6)157.2 subh3 -[(Floating-point constants are acceptable only as arguments to )2(DB)0(, )2(DW)0(, )2(DD)0(, )2(DQ)0(, )2(DT)0(, and )2(DO)0(, or as arguments to)]140.2 norm1 -[(the special operators )2(__float8__)0(, )2(__float16__)0(, )2(__float32__)0(, )2(__float64__)0(, )2(__float80m__)0(,)]129.2 norm0 -[2(__float80e__)0(, )2(__float128l__)0(, and )2(__float128h__)0(.)]118.2 norm2 -(31)pageodd -restore showpage -%%Page: 32 32 -%%BeginPageSetup -save -%%EndPageSetup -/32 pa -[(Floating-point constants are expressed in the traditional form: digits, then a period, then optionally more)]681 norm1 -[(digits, then optionally an )2(E)0( followed by an exponent. The period is mandatory, so that NASM can distinguish)]670 norm0 -[(between )2(dd 1)0(, which declares an integer constant, and )2(dd 1.0)0( which declares a floating-point constant.)]659 norm0 -[(NASM also support C99-style hexadecimal floating-point: )2(0x)0(, hexadecimal digits, period, optionally more)]648 norm0 -[(hexadeximal digits, then optionally a )2(P)0( followed by a )1(binary)0( \(not hexadecimal\) exponent in decimal notation.)]637 norm2 -[(Underscores to break up groups of digits are permitted in floating-point constants as well.)]620 norm3 -[(Some examples:)]603 norm3 -[2( db -0.2 ; "Quarter precision" )]586 code1 -[2( dw -0.5 ; IEEE 754r/SSE5 half precision )]575 code0 -[2( dd 1.2 ; an easy one )]564 code0 -[2( dd 1.222_222_222 ; underscores are permitted )]553 code0 -[2( dd 0x1p+2 ; 1.0x2^2 = 4.0 )]542 code0 -[2( dq 0x1p+32 ; 1.0x2^32 = 4 294 967 296.0 )]531 code0 -[2( dq 1.e10 ; 10 000 000 000.0 )]520 code0 -[2( dq 1.e+10 ; synonymous with 1.e10 )]509 code0 -[2( dq 1.e-10 ; 0.000 000 000 1 )]498 code0 -[2( dt 3.141592653589793238462 ; pi )]487 code0 -[2( do 1.e+4000 ; IEEE 754r quad precision)]476 code2 -[(The 8-bit "quarter-precision" floating-point format is sign:exponent:mantissa = 1:4:3 with an exponent bias)]459 norm1 -[(of 7. This appears to be the most frequently used 8-bit floating-point format, although it is not covered by)]448 norm0 -[(any formal standard. This is sometimes called a ")(minifloat.")]437 norm2 -[(The special operators are used to produce floating-point numbers in other contexts. They produce the binary)]420 norm1 -[(representation of a specific floating-point number as an integer, and can use anywhere integer constants are)]409 norm0 -[(used in an expression. )2(__float80m__)0( and )2(__float80e__)0( produce the 64-bit mantissa and 16-bit)]398 norm0 -[(exponent of an 80-bit floating-point number, and )2(__float128l__)0( and )2(__float128h__)0( produce the)]387 norm0 -[(lower and upper 64-bit halves of a 128-bit floating-point number, respectively.)]376 norm2 -[(For example:)]359 norm3 -[2( mov rax,__float64__\(3.141592653589793238462\))]342 code3 -[(... would assign the binary representation of pi as a 64-bit floating point number into )2(RAX)0(. This is exactly)]325 norm1 -[(equivalent to:)]314 norm2 -[2( mov rax,0x400921fb54442d18)]297 code3 -[(NASM cannot do compile-time arithmetic on floating-point constants. This is because NASM is designed to)]280 norm1 -[(be portable \226 although it always generates code to run on x86 processors, the assembler itself can run on any)]269 norm0 -[(system with an ANSI C compiler. Therefore, the assembler cannot guarantee the presence of a floating-point)]258 norm0 -[(unit capable of handling the )(Intel number formats, and so for NASM to be able to do floating arithmetic it)]247 norm0 -[(would have to include its own complete set of floating-point routines, which would significantly increase the)]236 norm0 -[(size of the assembler for very little benefit.)]225 norm2 -[(The special tokens )2(__Infinity__)0(, )2(__QNaN__)0( \(or )2(__NaN__)0(\) and )2(__SNaN__)0( can be used to generate)]208 norm1 -[(infinities, quiet )(NaNs, and signalling NaNs, respectively. These are normally used as macros:)]197 norm2 -[2(%define Inf __Infinity__ )]180 code1 -[2(%define NaN __QNaN__ )]169 code0 -[2()]158 code0 -[2( dq +1.5, -Inf, NaN ; Double-precision constants)]147 code2 -(32)pageeven -restore showpage -%%Page: 33 33 -%%BeginPageSetup -save -%%EndPageSetup -/33 pa -[{/section-3.4.7 xa}(Packed BCD Constants)](3.4.7)678.8 subh3 -[(x87-style packed BCD constants can be used in the same contexts as 80-bit floating-point numbers. They)]661.8 norm1 -[(are suffixed with )2(p)0( or prefixed with )2(0p)0(, and can include up to 18 decimal digits.)]650.8 norm2 -[(As with other numeric constants, underscores can be used to separate digits.)]633.8 norm3 -[(For example:)]616.8 norm3 -[2( dt 12_345_678_901_245_678p )]599.8 code1 -[2( dt -12_345_678_901_245_678p )]588.8 code0 -[2( dt +0p33 )]577.8 code0 -[2( dt 33p)]566.8 code2 -[{/section-3.5 xa}(Expressions)](3.5)545.4 head3 -[(Expressions in NASM are similar in syntax to those in C. Expressions are evaluated as 64-bit integers which)]528.4 norm1 -[(are then adjusted to the appropriate size.)]517.4 norm2 -[(NASM supports two special tokens in expressions, allowing calculations to involve the current assembly)]500.4 norm1 -[(position: the )2($)0( and )2($$)0( tokens. )2($)0( evaluates to the assembly position at the beginning of the line containing the)]489.4 norm0 -[(expression; so you can code an )(infinite loop using )2(JMP $)0(. )2($$)0( evaluates to the beginning of the current)]478.4 norm0 -[(section; so you can tell how far into the section you are by using )2(\($-$$\))0(.)]467.4 norm2 -[(The arithmetic )(operators provided by NASM are listed here, in increasing order of )(precedence.)]450.4 norm3 -[{/section-3.5.1 xa}2(|)0(: )(Bitwise OR Operator)](3.5.1)431.2 subh3 -[(The )2(|)0( operator gives a bitwise OR, exactly as performed by the )2(OR)0( machine instruction. Bitwise OR is the)]414.2 norm1 -[(lowest-priority arithmetic operator supported by NASM.)]403.2 norm2 -[{/section-3.5.2 xa}2(^)0(: )(Bitwise XOR Operator)](3.5.2)384 subh3 -[2(^)0( provides the bitwise XOR operation.)]367 norm3 -[{/section-3.5.3 xa}2(&)0(: )(Bitwise AND Operator)](3.5.3)347.8 subh3 -[2(&)0( provides the bitwise AND operation.)]330.8 norm3 -[{/section-3.5.4 xa}2(<<)0( and )2(>>)0(: )(Bit Shift Operators)](3.5.4)311.6 subh3 -[2(<<)0( gives a bit-shift to the left, just as it does in C. So )2(5<<3)0( evaluates to 5 times 8, or 40. )2(>>)0( gives a bit-shift)]294.6 norm1 -[(to the right; in NASM, such a shift is )1(always)0( unsigned, so that the bits shifted in from the left-hand end are)]283.6 norm0 -[(filled with zero rather than a sign-extension of the previous highest bit.)]272.6 norm2 -[{/section-3.5.5 xa}2(+)0( and )2(-)0(: )(Addition and )(Subtraction Operators)](3.5.5)253.4 subh3 -[(The )2(+)0( and )2(-)0( operators do perfectly ordinary addition and subtraction.)]236.4 norm3 -[{/section-3.5.6 xa}2(*)0(, )2(/)0(, )2(//)0(, )2(%)0( and )2(%%)0(: )(Multiplication and )(Division)](3.5.6)217.2 subh3 -[2(*)0( is the multiplication operator. )2(/)0( and )2(//)0( are both division operators: )2(/)0( is )(unsigned division and )2(//)0( is )(signed)]200.2 norm1 -[(division. Similarly, )2(%)0( and )2(%%)0( provide )(unsigned and )(signed modulo operators respectively.)]189.2 norm2 -[(NASM, like ANSI C, provides no guarantees about the sensible operation of the signed modulo operator.)]172.2 norm3 -[(Since the )2(%)0( character is used extensively by the macro )(preprocessor, you should ensure that both the signed)]155.2 norm1 -[(and unsigned modulo operators are followed by white space wherever they appear.)]144.2 norm2 -(33)pageodd -restore showpage -%%Page: 34 34 -%%BeginPageSetup -save -%%EndPageSetup -/34 pa -[{/section-3.5.7 xa}(Unary Operators: )2(+)0(, )2(-)0(, )2(~)0(, )2(!)0( and )2(SEG)](3.5.7)678.8 subh3 -[(The highest-priority operators in NASM's expression grammar are those which only apply to one argument.)]661.8 norm1 -[2(-)0( negates its operand, )2(+)0( does nothing \(it's provided for symmetry with )2(-)0(\), )2(~)0( computes the )(one's complement)]650.8 norm0 -[(of its operand, )2(!)0( is the )(logical negation operator, and )2(SEG)0( provides the )(segment address of its operand)]639.8 norm0 -[(\(explained in more detail in ){/section-3.6 xl}(section 3.6){el}(\).)]628.8 norm2 -[{/section-3.6 xa}2(SEG)0( and )2(WRT)](3.6)607.4 head3 -[(When writing large 16-bit programs, which must be split into multiple )(segments, it is often necessary to be)]590.4 norm1 -[(able to refer to the )(segment part of the address of a symbol. NASM supports the )2(SEG)0( operator to perform this)]579.4 norm0 -[(function.)]568.4 norm2 -[(The )2(SEG)0( operator returns the )1(preferred)0( segment base of a symbol, defined as the segment base relative to)]551.4 norm1 -[(which the offset of the symbol makes sense. So the code)]540.4 norm2 -[2( mov ax,seg symbol )]523.4 code1 -[2( mov es,ax )]512.4 code0 -[2( mov bx,symbol)]501.4 code2 -[(will load )2(ES:BX)0( with a valid pointer to the symbol )2(symbol)0(.)]484.4 norm3 -[(Things can be more complex than this: since 16-bit segments and )(groups may )(overlap, you might)]467.4 norm1 -[(occasionally want to refer to some symbol using a different segment base from the preferred one. NASM lets)]456.4 norm0 -[(you do this, by the use of the )2(WRT)0( \(With Reference To\) keyword. So you can do things like)]445.4 norm2 -[2( mov ax,weird_seg ; weird_seg is a segment base )]428.4 code1 -[2( mov es,ax )]417.4 code0 -[2( mov bx,symbol wrt weird_seg)]406.4 code2 -[(to load )2(ES:BX)0( with a different, but functionally equivalent, pointer to the symbol )2(symbol)0(.)]389.4 norm3 -[(NASM supports far \(inter-segment\) calls and jumps by means of the syntax )2(call segment:offset)0(,)]372.4 norm1 -[(where )2(segment)0( and )2(offset)0( both represent immediate values. So to call a far procedure, you could code)]361.4 norm0 -[(either of)]350.4 norm2 -[2( call \(seg procedure\):procedure )]333.4 code1 -[2( call weird_seg:\(procedure wrt weird_seg\))]322.4 code2 -[(\(The parentheses are included for clarity, to show the intended parsing of the above instructions. They are not)]305.4 norm1 -[(necessary in practice.\))]294.4 norm2 -[(NASM supports the syntax )2(call far procedure)0( as a synonym for the first of the above usages. )2(JMP)]277.4 norm1 -[(works identically to )2(CALL)0( in these examples.)]266.4 norm2 -[(To declare a )(far pointer to a data item in a data segment, you must code)]249.4 norm3 -[2( dw symbol, seg symbol)]232.4 code3 -[(NASM supports no convenient synonym for this, though you can always invent one using the macro)]215.4 norm1 -[(processor.)]204.4 norm2 -[{/section-3.7 xa}2(STRICT)0(: Inhibiting Optimization)](3.7)183 head3 -[(When assembling with the optimizer set to level 2 or higher \(see ){/section-2.1.22 xl}(section 2.1.22){el}(\), NASM will use size)]166 norm1 -[(specifiers \()2(BYTE)0(, )2(WORD)0(, )2(DWORD)0(, )2(QWORD)0(, )2(TWORD)0(, )2(OWORD)0( or )2(YWORD)0(\), but will give them the smallest)]155 norm0 -[(possible size. The keyword )2(STRICT)0( can be used to inhibit optimization and force a particular operand to be)]144 norm0 -[(emitted in the specified size. For example, with the optimizer on, and in )2(BITS 16)0( mode,)]133 norm2 -[2( push dword 33)]116 code3 -(34)pageeven -restore showpage -%%Page: 35 35 -%%BeginPageSetup -save -%%EndPageSetup -/35 pa -[(is encoded in three bytes )2(66 6A 21)0(, whereas)]681 norm3 -[2( push strict dword 33)]664 code3 -[(is encoded in six bytes, with a full dword immediate operand )2(66 68 21 00 00 00)0(.)]647 norm3 -[(With the optimizer off, the same code \(six bytes\) is generated whether the )2(STRICT)0( keyword was used or not.)]630 norm3 -[{/section-3.8 xa}(Critical Expressions)](3.8)608.6 head3 -[(Although NASM has an optional multi-pass optimizer, there are some expressions which must be resolvable)]591.6 norm1 -[(on the first pass. These are called )1(Critical Expressions)0(.)]580.6 norm2 -[(The first pass is used to determine the size of all the assembled code and data, so that the second pass, when)]563.6 norm1 -[(generating all the code, knows all the symbol addresses the code refers to. So one thing NASM can't handle is)]552.6 norm0 -[(code whose size depends on the value of a symbol declared after the code in question. For example,)]541.6 norm2 -[2( times \(label-$\) db 0 )]524.6 code1 -[2(label: db 'Where am I?')]513.6 code2 -[(The argument to )2(TIMES)0( in this case could equally legally evaluate to anything at all; NASM will reject this)]496.6 norm1 -[(example because it cannot tell the size of the )2(TIMES)0( line when it first sees it. It will just as firmly reject the)]485.6 norm0 -[(slightly )(paradoxical code)]474.6 norm2 -[2( times \(label-$+1\) db 0 )]457.6 code1 -[2(label: db 'NOW where am I?')]446.6 code2 -[(in which )1(any)0( value for the )2(TIMES)0( argument is by definition wrong!)]429.6 norm3 -[(NASM rejects these examples by means of a concept called a )1(critical expression)0(, which is defined to be an)]412.6 norm1 -[(expression whose value is required to be computable in the first pass, and which must therefore depend only)]401.6 norm0 -[(on symbols defined before it. The argument to the )2(TIMES)0( prefix is a critical expression.)]390.6 norm2 -[{/section-3.9 xa}(Local Labels)](3.9)369.2 head3 -[(NASM gives special treatment to symbols beginning with a )(period. A label beginning with a single period is)]352.2 norm1 -[(treated as a )1(local)0( label, which means that it is associated with the previous non-local label. So, for example:)]341.2 norm2 -[2(label1 ; some code )]324.2 code1 -[2()]313.2 code0 -[2(.loop )]302.2 code0 -[2( ; some more code )]291.2 code0 -[2()]280.2 code0 -[2( jne .loop )]269.2 code0 -[2( ret )]258.2 code0 -[2()]247.2 code0 -[2(label2 ; some code )]236.2 code0 -[2()]225.2 code0 -[2(.loop )]214.2 code0 -[2( ; some more code )]203.2 code0 -[2()]192.2 code0 -[2( jne .loop )]181.2 code0 -[2( ret)]170.2 code2 -[(In the above code fragment, each )2(JNE)0( instruction jumps to the line immediately before it, because the two)]153.2 norm1 -[(definitions of )2(.loop)0( are kept separate by virtue of each being associated with the previous non-local label.)]142.2 norm2 -[(This form of local label handling is borrowed from the old Amiga assembler )(DevPac; however, NASM goes)]125.2 norm1 -[(one step further, in allowing access to local labels from other parts of the code. This is achieved by means of)]114.2 norm0 -[1(defining)0( a local label in terms of the previous non-local label: the first definition of )2(.loop)0( above is really)]103.2 norm0 -(35)pageodd -restore showpage -%%Page: 36 36 -%%BeginPageSetup -save -%%EndPageSetup -/36 pa -[(defining a symbol called )2(label1.loop)0(, and the second defines a symbol called )2(label2.loop)0(. So, if you)]681 norm0 -[(really needed to, you could write)]670 norm2 -[2(label3 ; some more code )]653 code1 -[2( ; and some more )]642 code0 -[2()]631 code0 -[2( jmp label1.loop)]620 code2 -[(Sometimes it is useful \226 in a macro, for instance \226 to be able to define a label which can be referenced from)]603 norm1 -[(anywhere but which doesn't interfere with the normal local-label mechanism. Such a label can't be non-local)]592 norm0 -[(because it would interfere with subsequent definitions of, and references to, local labels; and it can't be local)]581 norm0 -[(because the macro that defined it wouldn't know the label's full name. NASM therefore introduces a third)]570 norm0 -[(type of label, which is probably only useful in macro definitions: if a label begins with the )(special prefix )2(..@)0(,)]559 norm0 -[(then it does nothing to the local label mechanism. So you could code)]548 norm2 -[2(label1: ; a non-local label )]531 code1 -[2(.local: ; this is really label1.local )]520 code0 -[2(..@foo: ; this is a special symbol )]509 code0 -[2(label2: ; another non-local label )]498 code0 -[2(.local: ; this is really label2.local )]487 code0 -[2()]476 code0 -[2( jmp ..@foo ; this will jump three lines up)]465 code2 -[(NASM has the capacity to define other special symbols beginning with a double period: for example,)]448 norm1 -[2(..start)0( is used to specify the entry point in the )2(obj)0( output format \(see ){/section-7.4.6 xl}(section 7.4.6){el}(\).)]437 norm2 -(36)pageeven -restore showpage -%%Page: 37 37 -%%BeginPageSetup -save -%%EndPageSetup -/37 pa -[{/chapter-4 xa}(Chapter 4: The NASM )(Preprocessor)]642.8 chap3 -[(NASM contains a powerful )(macro processor, which supports conditional assembly, multi-level file inclusion,)]607.8 norm1 -[(two forms of macro \(single-line and multi-line\), and a `context stack' mechanism for extra macro power.)]596.8 norm0 -[(Preprocessor directives all begin with a )2(%)0( sign.)]585.8 norm2 -[(The preprocessor collapses all lines which end with a backslash \(\\\) character into a single line. Thus:)]568.8 norm3 -[2(%define THIS_VERY_LONG_MACRO_NAME_IS_DEFINED_TO \\ )]551.8 code1 -[2( THIS_VALUE)]540.8 code2 -[(will work like a single-line macro without the backslash-newline sequence.)]523.8 norm3 -[{/section-4.1 xa}(Single-Line Macros)](4.1)502.4 head3 -[{/section-4.1.1 xa}(The Normal Way: )2(%define)](4.1.1)483.2 subh3 -[(Single-line macros are defined using the )2(%define)0( preprocessor directive. The definitions work in a similar)]466.2 norm1 -[(way to C; so you can do things like)]455.2 norm2 -[2(%define ctrl 0x1F & )]438.2 code1 -[2(%define param\(a,b\) \(\(a\)+\(a\)*\(b\)\) )]427.2 code0 -[2()]416.2 code0 -[2( mov byte [param\(2,ebx\)], ctrl 'D')]405.2 code2 -[(which will expand to)]388.2 norm3 -[2( mov byte [\(2\)+\(2\)*\(ebx\)], 0x1F & 'D')]371.2 code3 -[(When the expansion of a single-line macro contains tokens which invoke another macro, the expansion is)]354.2 norm1 -[(performed at invocation time, not at definition time. Thus the code)]343.2 norm2 -[2(%define a\(x\) 1+b\(x\) )]326.2 code1 -[2(%define b\(x\) 2*x )]315.2 code0 -[2()]304.2 code0 -[2( mov ax,a\(8\))]293.2 code2 -[(will evaluate in the expected way to )2(mov ax,1+2*8)0(, even though the macro )2(b)0( wasn't defined at the time of)]276.2 norm1 -[(definition of )2(a)0(.)]265.2 norm2 -[(Macros defined with )2(%define)0( are )(case sensitive: after )2(%define foo bar)0(, only )2(foo)0( will expand to )2(bar)0(:)]248.2 norm1 -[2(Foo)0( or )2(FOO)0( will not. By using )2(%idefine)0( instead of )2(%define)0( \(the `i' stands for `insensitive'\) you can)]237.2 norm0 -[(define all the case variants of a macro at once, so that )2(%idefine foo bar)0( would cause )2(foo)0(, )2(Foo)0(, )2(FOO)0(,)]226.2 norm0 -[2(fOO)0( and so on all to expand to )2(bar)0(.)]215.2 norm2 -[(There is a mechanism which detects when a macro call has occurred as a result of a previous expansion of the)]198.2 norm1 -[(same macro, to guard against )(circular references and infinite loops. If this happens, the preprocessor will only)]187.2 norm0 -[(expand the first occurrence of the macro. Hence, if you code)]176.2 norm2 -[2(%define a\(x\) 1+a\(x\) )]159.2 code1 -[2()]148.2 code0 -[2( mov ax,a\(3\))]137.2 code2 -[(the macro )2(a\(3\))0( will expand once, becoming )2(1+a\(3\))0(, and will then expand no further. This behaviour can)]120.2 norm1 -[(be useful: see ){/section-9.1 xl}(section 9.1){el}( for an example of its use.)]109.2 norm2 -(37)pageodd -restore showpage -%%Page: 38 38 -%%BeginPageSetup -save -%%EndPageSetup -/38 pa -[(You can )(overload single-line macros: if you write)]681 norm3 -[2(%define foo\(x\) 1+x )]664 code1 -[2(%define foo\(x,y\) 1+x*y)]653 code2 -[(the preprocessor will be able to handle both types of macro call, by counting the parameters you pass; so)]636 norm1 -[2(foo\(3\))0( will become )2(1+3)0( whereas )2(foo\(ebx,2\))0( will become )2(1+ebx*2)0(. However, if you define)]625 norm2 -[2(%define foo bar)]608 code3 -[(then no other definition of )2(foo)0( will be accepted: a macro with no parameters prohibits the definition of the)]591 norm1 -[(same name as a macro )1(with)0( parameters, and vice versa.)]580 norm2 -[(This doesn't prevent single-line macros being )1(redefined)0(: you can perfectly well define a macro with)]563 norm3 -[2(%define foo bar)]546 code3 -[(and then re-define it later in the same source file with)]529 norm3 -[2(%define foo baz)]512 code3 -[(Then everywhere the macro )2(foo)0( is invoked, it will be expanded according to the most recent definition. This)]495 norm1 -[(is particularly useful when defining single-line macros with )2(%assign)0( \(see ){/section-4.1.7 xl}(section 4.1.7){el}(\).)]484 norm2 -[(You can )(pre-define single-line macros using the `-d' option on the NASM command line: see ){/section-2.1.18 xl}(section 2.1.18){el}(.)]467 norm3 -[{/section-4.1.2 xa}(Resolving )2(%define)0(: )2(%xdefine)](4.1.2)447.8 subh3 -[(To have a reference to an embedded single-line macro resolved at the time that the embedding macro is)]430.8 norm1 -[1(defined)0(, as opposed to when the embedding macro is )1(expanded)0(, you need a different mechanism to the one)]419.8 norm0 -[(offered by )2(%define)0(. The solution is to use )2(%xdefine)0(, or it's )(case-insensitive counterpart )2(%ixdefine)0(.)]408.8 norm2 -[(Suppose you have the following code:)]391.8 norm3 -[2(%define isTrue 1 )]374.8 code1 -[2(%define isFalse isTrue )]363.8 code0 -[2(%define isTrue 0 )]352.8 code0 -[2()]341.8 code0 -[2(val1: db isFalse )]330.8 code0 -[2()]319.8 code0 -[2(%define isTrue 1 )]308.8 code0 -[2()]297.8 code0 -[2(val2: db isFalse)]286.8 code2 -[(In this case, )2(val1)0( is equal to 0, and )2(val2)0( is equal to 1. This is because, when a single-line macro is defined)]269.8 norm1 -[(using )2(%define)0(, it is expanded only when it is called. As )2(isFalse)0( expands to )2(isTrue)0(, the expansion will)]258.8 norm0 -[(be the current value of )2(isTrue)0(. The first time it is called that is 0, and the second time it is 1.)]247.8 norm2 -[(If you wanted )2(isFalse)0( to expand to the value assigned to the embedded macro )2(isTrue)0( at the time that)]230.8 norm1 -[2(isFalse)0( was defined, you need to change the above code to use )2(%xdefine)0(.)]219.8 norm2 -[2(%xdefine isTrue 1 )]202.8 code1 -[2(%xdefine isFalse isTrue )]191.8 code0 -[2(%xdefine isTrue 0 )]180.8 code0 -[2()]169.8 code0 -[2(val1: db isFalse )]158.8 code0 -[2()]147.8 code0 -[2(%xdefine isTrue 1 )]136.8 code0 -[2()]125.8 code0 -[2(val2: db isFalse)]114.8 code2 -(38)pageeven -restore showpage -%%Page: 39 39 -%%BeginPageSetup -save -%%EndPageSetup -/39 pa -[(Now, each time that )2(isFalse)0( is called, it expands to 1, as that is what the embedded macro )2(isTrue)]681 norm1 -[(expanded to at the time that )2(isFalse)0( was defined.)]670 norm2 -[{/section-4.1.3 xa}(Macro Indirection: )2(%[...])](4.1.3)650.8 subh3 -[(The )2(%[...])0( construct can be used to expand macros in contexts where macro expansion would otherwise)]633.8 norm1 -[(not occur, including in the names other macros. For example, if you have a set of macros named )2(Foo16)0(,)]622.8 norm0 -[2(Foo32)0( and )2(Foo64)0(, you could write:)]611.8 norm2 -[2( mov ax,Foo%[__BITS__] ; The Foo value)]594.8 code3 -[(to use the builtin macro )2(__BITS__)0( \(see ){/section-4.11.5 xl}(section 4.11.5){el}(\) to automatically select between them. Similarly, the)]577.8 norm1 -[(two statements:)]566.8 norm2 -[2(%xdefine Bar Quux ; Expands due to %xdefine )]549.8 code1 -[2(%define Bar %[Quux] ; Expands due to %[...])]538.8 code2 -[(have, in fact, exactly the same effect.)]521.8 norm3 -[2(%[...])0( concatenates to adjacent tokens in the same way that multi-line macro parameters do, see ){/section-4.3.8 xl}(section){el}]504.8 norm1 -[{/section-4.3.8 xl}(4.3.8){el}( for details.)]493.8 norm2 -[{/section-4.1.4 xa}(Concatenating Single Line Macro Tokens: )2(%+)](4.1.4)474.6 subh3 -[(Individual tokens in single line macros can be concatenated, to produce longer tokens for later processing.)]457.6 norm1 -[(This can be useful if there are several similar macros that perform similar functions.)]446.6 norm2 -[(Please note that a space is required after )2(%+)0(, in order to disambiguate it from the syntax )2(%+1)0( used in multiline)]429.6 norm1 -[(macros.)]418.6 norm2 -[(As an example, consider the following:)]401.6 norm3 -[2(%define BDASTART 400h ; Start of BIOS data area)]384.6 code3 -[2(struc tBIOSDA ; its structure )]367.6 code1 -[2( .COM1addr RESW 1 )]356.6 code0 -[2( .COM2addr RESW 1 )]345.6 code0 -[2( ; ..and so on )]334.6 code0 -[2(endstruc)]323.6 code2 -[(Now, if we need to access the elements of tBIOSDA in different places, we can end up with:)]306.6 norm3 -[2( mov ax,BDASTART + tBIOSDA.COM1addr )]289.6 code1 -[2( mov bx,BDASTART + tBIOSDA.COM2addr)]278.6 code2 -[(This will become pretty ugly \(and tedious\) if used in many places, and can be reduced in size significantly by)]261.6 norm1 -[(using the following macro:)]250.6 norm2 -[2(; Macro to access BIOS variables by their names \(from tBDA\):)]233.6 code3 -[2(%define BDA\(x\) BDASTART + tBIOSDA. %+ x)]216.6 code3 -[(Now the above code can be written as:)]199.6 norm3 -[2( mov ax,BDA\(COM1addr\) )]182.6 code1 -[2( mov bx,BDA\(COM2addr\))]171.6 code2 -[(Using this feature, we can simplify references to a lot of macros \(and, in turn, reduce typing errors\).)]154.6 norm3 -[{/section-4.1.5 xa}(The Macro Name Itself: )2(%?)0( and )2(%??)](4.1.5)135.4 subh3 -[(The special symbols )2(%?)0( and )2(%??)0( can be used to reference the macro name itself inside a macro expansion,)]118.4 norm1 -[(this is supported for both single-and multi-line macros. )2(%?)0( refers to the macro name as )1(invoked)0(, whereas)]107.4 norm0 -(39)pageodd -restore showpage -%%Page: 40 40 -%%BeginPageSetup -save -%%EndPageSetup -/40 pa -[2(%??)0( refers to the macro name as )1(declared)0(. The two are always the same for case-sensitive macros, but for)]681 norm0 -[(case-insensitive macros, they can differ.)]670 norm2 -[(For example:)]653 norm3 -[2(%idefine Foo mov %?,%?? )]636 code1 -[2()]625 code0 -[2( foo )]614 code0 -[2( FOO)]603 code2 -[(will expand to:)]586 norm3 -[2( mov foo,Foo )]569 code1 -[2( mov FOO,Foo)]558 code2 -[(The sequence:)]541 norm3 -[2(%idefine keyword $%?)]524 code3 -[(can be used to make a keyword "disappear", for example in case a new instruction has been used as a label in)]507 norm1 -[(older code. For example:)]496 norm2 -[2(%idefine pause $%? ; Hide the PAUSE instruction)]479 code3 -[{/section-4.1.6 xa}(Undefining Single-Line Macros: )2(%undef)](4.1.6)459.8 subh3 -[(Single-line macros can be removed with the )2(%undef)0( directive. For example, the following sequence:)]442.8 norm3 -[2(%define foo bar )]425.8 code1 -[2(%undef foo )]414.8 code0 -[2()]403.8 code0 -[2( mov eax, foo)]392.8 code2 -[(will expand to the instruction )2(mov eax, foo)0(, since after )2(%undef)0( the macro )2(foo)0( is no longer defined.)]375.8 norm3 -[(Macros that would otherwise be pre-defined can be undefined on the command-line using the `-u' option on)]358.8 norm1 -[(the NASM command line: see ){/section-2.1.19 xl}(section 2.1.19){el}(.)]347.8 norm2 -[{/section-4.1.7 xa}(Preprocessor Variables: )2(%assign)](4.1.7)328.6 subh3 -[(An alternative way to define single-line macros is by means of the )2(%assign)0( command \(and its)]311.6 norm1 -[(case-insensitive counterpart )2(%iassign)0(, which differs from )2(%assign)0( in exactly the same way that)]300.6 norm0 -[2(%idefine)0( differs from )2(%define)0(\).)]289.6 norm2 -[2(%assign)0( is used to define single-line macros which take no parameters and have a numeric value. This)]272.6 norm1 -[(value can be specified in the form of an expression, and it will be evaluated once, when the )2(%assign)]261.6 norm0 -[(directive is processed.)]250.6 norm2 -[(Like )2(%define)0(, macros defined using )2(%assign)0( can be re-defined later, so you can do things like)]233.6 norm3 -[2(%assign i i+1)]216.6 code3 -[(to increment the numeric value of a macro.)]199.6 norm3 -[2(%assign)0( is useful for controlling the termination of )2(%rep)0( preprocessor loops: see ){/section-4.5 xl}(section 4.5){el}( for an)]182.6 norm1 -[(example of this. Another use for )2(%assign)0( is given in ){/section-8.4 xl}(section 8.4){el}( and ){/section-9.1 xl}(section 9.1){el}(.)]171.6 norm2 -[(The expression passed to )2(%assign)0( is a )(critical expression \(see ){/section-3.8 xl}(section 3.8){el}(\), and must also evaluate to a pure)]154.6 norm1 -[(number \(rather than a relocatable reference such as a code or data address, or anything involving a register\).)]143.6 norm2 -(40)pageeven -restore showpage -%%Page: 41 41 -%%BeginPageSetup -save -%%EndPageSetup -/41 pa -[{/section-4.1.8 xa}(Defining Strings: )2(%defstr)](4.1.8)678.8 subh3 -[2(%defstr)0(, and its case-insensitive counterpart )2(%idefstr)0(, define or redefine a single-line macro without)]661.8 norm1 -[(parameters but converts the entire right-hand side, after macro expansion, to a quoted string before definition.)]650.8 norm2 -[(For example:)]633.8 norm3 -[2(%defstr test TEST)]616.8 code3 -[(is equivalent to)]599.8 norm3 -[2(%define test 'TEST')]582.8 code3 -[(This can be used, for example, with the )2(%!)0( construct \(see ){/section-4.10.2 xl}(section 4.10.2){el}(\):)]565.8 norm3 -[2(%defstr PATH %!PATH ; The operating system PATH variable)]548.8 code3 -[{/section-4.1.9 xa}(Defining Tokens: )2(%deftok)](4.1.9)529.6 subh3 -[2(%deftok)0(, and its case-insensitive counterpart )2(%ideftok)0(, define or redefine a single-line macro without)]512.6 norm1 -[(parameters but converts the second parameter, after string conversion, to a sequence of tokens.)]501.6 norm2 -[(For example:)]484.6 norm3 -[2(%deftok test 'TEST')]467.6 code3 -[(is equivalent to)]450.6 norm3 -[2(%define test TEST)]433.6 code3 -[{/section-4.2 xa}(String Manipulation in Macros)](4.2)412.2 head3 -[(It's often useful to be able to handle strings in macros. NASM supports a few simple string handling macro)]395.2 norm1 -[(operators from which more complex operations can be constructed.)]384.2 norm2 -[(All the string operators define or redefine a value \(either a string or a numeric value\) to a single-line macro.)]367.2 norm1 -[(When producing a string value, it may change the style of quoting of the input string or strings, and possibly)]356.2 norm0 -[(use )2(\\)0(\226escapes inside )2(`)0(\226quoted strings.)]345.2 norm2 -[{/section-4.2.1 xa}(Concatenating Strings: )2(%strcat)](4.2.1)326 subh3 -[(The )2(%strcat)0( operator concatenates quoted strings and assign them to a single-line macro.)]309 norm3 -[(For example:)]292 norm3 -[2(%strcat alpha "Alpha: ", '12" screen')]275 code3 -[(... would assign the value )2('Alpha: 12" screen')0( to )2(alpha)0(. Similarly:)]258 norm3 -[2(%strcat beta '"foo"\\', "'bar'")]241 code3 -[(... would assign the value )2(`"foo"\\\\'bar'`)0( to )2(beta)0(.)]224 norm3 -[(The use of commas to separate strings is permitted but optional.)]207 norm3 -[{/section-4.2.2 xa}(String Length: )2(%strlen)](4.2.2)187.8 subh3 -[(The )2(%strlen)0( operator assigns the length of a string to a macro. For example:)]170.8 norm3 -[2(%strlen charcnt 'my string')]153.8 code3 -[(In this example, )2(charcnt)0( would receive the value 9, just as if an )2(%assign)0( had been used. In this example,)]136.8 norm1 -[2('my string')0( was a literal string but it could also have been a single-line macro that expands to a string, as)]125.8 norm0 -[(in the following example:)]114.8 norm2 -(41)pageodd -restore showpage -%%Page: 42 42 -%%BeginPageSetup -save -%%EndPageSetup -/42 pa -[2(%define sometext 'my string' )]681 code1 -[2(%strlen charcnt sometext)]670 code2 -[(As in the first case, this would result in )2(charcnt)0( being assigned the value of 9.)]653 norm3 -[{/section-4.2.3 xa}(Extracting Substrings: )2(%substr)](4.2.3)633.8 subh3 -[(Individual letters or substrings in strings can be extracted using the )2(%substr)0( operator. An example of its use)]616.8 norm1 -[(is probably more useful than the description:)]605.8 norm2 -[2(%substr mychar 'xyzw' 1 ; equivalent to %define mychar 'x' )]588.8 code1 -[2(%substr mychar 'xyzw' 2 ; equivalent to %define mychar 'y' )]577.8 code0 -[2(%substr mychar 'xyzw' 3 ; equivalent to %define mychar 'z' )]566.8 code0 -[2(%substr mychar 'xyzw' 2,2 ; equivalent to %define mychar 'yz' )]555.8 code0 -[2(%substr mychar 'xyzw' 2,-1 ; equivalent to %define mychar 'yzw' )]544.8 code0 -[2(%substr mychar 'xyzw' 2,-2 ; equivalent to %define mychar 'yz')]533.8 code2 -[(As with )2(%strlen)0( \(see ){/section-4.2.2 xl}(section 4.2.2){el}(\), the first parameter is the single-line macro to be created and the)]516.8 norm1 -[(second is the string. The third parameter specifies the first character to be selected, and the optional fourth)]505.8 norm0 -[(parameter preceeded by comma\) is the length. Note that the first index is 1, not 0 and the last index is equal to)]494.8 norm0 -[(the value that )2(%strlen)0( would assign given the same string. Index values out of range result in an empty)]483.8 norm0 -[(string. A negative length means "until N-1 characters before the end of string", i.e. )2(-1)0( means until end of)]472.8 norm0 -[(string, )2(-2)0( until one character before, etc.)]461.8 norm2 -[{/section-4.3 xa}(Multi-Line Macros: )2(%macro)](4.3)440.4 head3 -[(Multi-line macros are much more like the type of macro seen in MASM and TASM: a multi-line macro)]423.4 norm1 -[(definition in NASM looks something like this.)]412.4 norm2 -[2(%macro prologue 1 )]395.4 code1 -[2()]384.4 code0 -[2( push ebp )]373.4 code0 -[2( mov ebp,esp )]362.4 code0 -[2( sub esp,%1 )]351.4 code0 -[2()]340.4 code0 -[2(%endmacro)]329.4 code2 -[(This defines a C-like function prologue as a macro: so you would invoke the macro with a call such as)]312.4 norm3 -[2(myfunc: prologue 12)]295.4 code3 -[(which would expand to the three lines of code)]278.4 norm3 -[2(myfunc: push ebp )]261.4 code1 -[2( mov ebp,esp )]250.4 code0 -[2( sub esp,12)]239.4 code2 -[(The number )2(1)0( after the macro name in the )2(%macro)0( line defines the number of parameters the macro)]222.4 norm1 -[2(prologue)0( expects to receive. The use of )2(%1)0( inside the macro definition refers to the first parameter to the)]211.4 norm0 -[(macro call. With a macro taking more than one parameter, subsequent parameters would be referred to as )2(%2)0(,)]200.4 norm0 -[2(%3)0( and so on.)]189.4 norm2 -[(Multi-line macros, like single-line macros, are )(case-sensitive, unless you define them using the alternative)]172.4 norm1 -[(directive )2(%imacro)0(.)]161.4 norm2 -[(If you need to pass a comma as )1(part)0( of a parameter to a multi-line macro, you can do that by enclosing the)]144.4 norm1 -[(entire parameter in )(braces. So you could code things like)]133.4 norm2 -[2(%macro silly 2 )]116.4 code1 -[2()]105.4 code0 -(42)pageeven -restore showpage -%%Page: 43 43 -%%BeginPageSetup -save -%%EndPageSetup -/43 pa -[2( %2: db %1 )]681 code0 -[2()]670 code0 -[2(%endmacro )]659 code0 -[2()]648 code0 -[2( silly 'a', letter_a ; letter_a: db 'a' )]637 code0 -[2( silly 'ab', string_ab ; string_ab: db 'ab' )]626 code0 -[2( silly {13,10}, crlf ; crlf: db 13,10)]615 code2 -[{/section-4.3.1 xa}(Recursive Multi-Line Macros: )2(%rmacro)](4.3.1)595.8 subh3 -[(A multi-line macro cannot be referenced within itself, in order to prevent accidental infinite recursion.)]578.8 norm3 -[(Recursive multi-line macros allow for self-referencing, with the caveat that the user is aware of the)]561.8 norm1 -[(existence, use and purpose of recursive multi-line macros. There is also a generous, but sane, upper limit to)]550.8 norm0 -[(the number of recursions, in order to prevent run-away memory consumption in case of accidental infinite)]539.8 norm0 -[(recursion.)]528.8 norm2 -[(As with non-recursive multi-line macros, recursive multi-line macros are )(case-sensitive, unless you define)]511.8 norm1 -[(them using the alternative directive )2(%irmacro)0(.)]500.8 norm2 -[{/section-4.3.2 xa}(Overloading Multi-Line Macros)](4.3.2)481.6 subh3 -[(As with single-line macros, multi-line macros can be overloaded by defining the same macro name several)]464.6 norm1 -[(times with different numbers of parameters. This time, no exception is made for macros with no parameters at)]453.6 norm0 -[(all. So you could define)]442.6 norm2 -[2(%macro prologue 0 )]425.6 code1 -[2()]414.6 code0 -[2( push ebp )]403.6 code0 -[2( mov ebp,esp )]392.6 code0 -[2()]381.6 code0 -[2(%endmacro)]370.6 code2 -[(to define an alternative form of the function prologue which allocates no local stack space.)]353.6 norm3 -[(Sometimes, however, you might want to `overload' a machine instruction; for example, you might want to)]336.6 norm1 -[(define)]325.6 norm2 -[2(%macro push 2 )]308.6 code1 -[2()]297.6 code0 -[2( push %1 )]286.6 code0 -[2( push %2 )]275.6 code0 -[2()]264.6 code0 -[2(%endmacro)]253.6 code2 -[(so that you could code)]236.6 norm3 -[2( push ebx ; this line is not a macro call )]219.6 code1 -[2( push eax,ecx ; but this one is)]208.6 code2 -[(Ordinarily, NASM will give a warning for the first of the above two lines, since )2(push)0( is now defined to be a)]191.6 norm1 -[(macro, and is being invoked with a number of parameters for which no definition has been given. The correct)]180.6 norm0 -[(code will still be generated, but the assembler will give a warning. This warning can be disabled by the use of)]169.6 norm0 -[(the )2(-w-macro-params)0( command-line option \(see ){/section-2.1.24 xl}(section 2.1.24){el}(\).)]158.6 norm2 -[{/section-4.3.3 xa}(Macro-Local Labels)](4.3.3)139.4 subh3 -[(NASM allows you to define labels within a multi-line macro definition in such a way as to make them local)]122.4 norm1 -[(to the macro call: so calling the same macro multiple times will use a different label each time. You do this by)]111.4 norm0 -(43)pageodd -restore showpage -%%Page: 44 44 -%%BeginPageSetup -save -%%EndPageSetup -/44 pa -[(prefixing )2(%%)0( to the label name. So you can invent an instruction which executes a )2(RET)0( if the )2(Z)0( flag is set by)]681 norm0 -[(doing this:)]670 norm2 -[2(%macro retz 0 )]653 code1 -[2()]642 code0 -[2( jnz %%skip )]631 code0 -[2( ret )]620 code0 -[2( %%skip: )]609 code0 -[2()]598 code0 -[2(%endmacro)]587 code2 -[(You can call this macro as many times as you want, and every time you call it NASM will make up a different)]570 norm1 -[(`real' name to substitute for the label )2(%%skip)0(. The names NASM invents are of the form )2(..@2345.skip)0(,)]559 norm0 -[(where the number 2345 changes with every macro call. The )2(..@)0( prefix prevents macro-local labels from)]548 norm0 -[(interfering with the local label mechanism, as described in ){/section-3.9 xl}(section 3.9){el}(. You should avoid defining your own)]537 norm0 -[(labels in this form \(the )2(..@)0( prefix, then a number, then another period\) in case they interfere with)]526 norm0 -[(macro-local labels.)]515 norm2 -[{/section-4.3.4 xa}(Greedy Macro Parameters)](4.3.4)495.8 subh3 -[(Occasionally it is useful to define a macro which lumps its entire command line into one parameter definition,)]478.8 norm1 -[(possibly after extracting one or two smaller parameters from the front. An example might be a macro to write)]467.8 norm0 -[(a text string to a file in MS-DOS, where you might want to be able to write)]456.8 norm2 -[2( writefile [filehandle],"hello, world",13,10)]439.8 code3 -[(NASM allows you to define the last parameter of a macro to be )1(greedy)0(, meaning that if you invoke the macro)]422.8 norm1 -[(with more parameters than it expects, all the spare parameters get lumped into the last defined one along with)]411.8 norm0 -[(the separating commas. So if you code:)]400.8 norm2 -[2(%macro writefile 2+ )]383.8 code1 -[2()]372.8 code0 -[2( jmp %%endstr )]361.8 code0 -[2( %%str: db %2 )]350.8 code0 -[2( %%endstr: )]339.8 code0 -[2( mov dx,%%str )]328.8 code0 -[2( mov cx,%%endstr-%%str )]317.8 code0 -[2( mov bx,%1 )]306.8 code0 -[2( mov ah,0x40 )]295.8 code0 -[2( int 0x21 )]284.8 code0 -[2()]273.8 code0 -[2(%endmacro)]262.8 code2 -[(then the example call to )2(writefile)0( above will work as expected: the text before the first comma,)]245.8 norm1 -[2([filehandle])0(, is used as the first macro parameter and expanded when )2(%1)0( is referred to, and all the)]234.8 norm0 -[(subsequent text is lumped into )2(%2)0( and placed after the )2(db)0(.)]223.8 norm2 -[(The greedy nature of the macro is indicated to NASM by the use of the )2(+)0( sign after the parameter count on the)]206.8 norm1 -[2(%macro)0( line.)]195.8 norm2 -[(If you define a greedy macro, you are effectively telling NASM how it should expand the macro given )1(any)]178.8 norm1 -[(number of parameters from the actual number specified up to infinity; in this case, for example, NASM now)]167.8 norm0 -[(knows what to do when it sees a call to )2(writefile)0( with 2, 3, 4 or more parameters. NASM will take this)]156.8 norm0 -[(into account when overloading macros, and will not allow you to define another form of )2(writefile)0( taking)]145.8 norm0 -[(4 parameters \(for example\).)]134.8 norm2 -[(Of course, the above macro could have been implemented as a non-greedy macro, in which case the call to it)]117.8 norm1 -[(would have had to look like)]106.8 norm2 -(44)pageeven -restore showpage -%%Page: 45 45 -%%BeginPageSetup -save -%%EndPageSetup -/45 pa -[2( writefile [filehandle], {"hello, world",13,10})]681 code3 -[(NASM provides both mechanisms for putting )(commas in macro parameters, and you choose which one you)]664 norm1 -[(prefer for each macro definition.)]653 norm2 -[(See ){/section-6.3.1 xl}(section 6.3.1){el}( for a better way to write the above macro.)]636 norm3 -[{/section-4.3.5 xa}(Default Macro Parameters)](4.3.5)616.8 subh3 -[(NASM also allows you to define a multi-line macro with a )1(range)0( of allowable parameter counts. If you do)]599.8 norm1 -[(this, you can specify defaults for )(omitted parameters. So, for example:)]588.8 norm2 -[2(%macro die 0-1 "Painful program death has occurred." )]571.8 code1 -[2()]560.8 code0 -[2( writefile 2,%1 )]549.8 code0 -[2( mov ax,0x4c01 )]538.8 code0 -[2( int 0x21 )]527.8 code0 -[2()]516.8 code0 -[2(%endmacro)]505.8 code2 -[(This macro \(which makes use of the )2(writefile)0( macro defined in ){/section-4.3.4 xl}(section 4.3.4){el}(\) can be called with an)]488.8 norm1 -[(explicit error message, which it will display on the error output stream before exiting, or it can be called with)]477.8 norm0 -[(no parameters, in which case it will use the default error message supplied in the macro definition.)]466.8 norm2 -[(In general, you supply a minimum and maximum number of parameters for a macro of this type; the)]449.8 norm1 -[(minimum number of parameters are then required in the macro call, and then you provide defaults for the)]438.8 norm0 -[(optional ones. So if a macro definition began with the line)]427.8 norm2 -[2(%macro foobar 1-3 eax,[ebx+2])]410.8 code3 -[(then it could be called with between one and three parameters, and )2(%1)0( would always be taken from the macro)]393.8 norm1 -[(call. )2(%2)0(, if not specified by the macro call, would default to )2(eax)0(, and )2(%3)0( if not specified would default to)]382.8 norm0 -[2([ebx+2])0(.)]371.8 norm2 -[(You can provide extra information to a macro by providing too many default parameters:)]354.8 norm3 -[2(%macro quux 1 something)]337.8 code3 -[(This will trigger a warning by default; see ){/section-2.1.24 xl}(section 2.1.24){el}( for more information. When )2(quux)0( is invoked, it)]320.8 norm1 -[(receives not one but two parameters. )2(something)0( can be referred to as )2(%2)0(. The difference between passing)]309.8 norm0 -[2(something)0( this way and writing )2(something)0( in the macro body is that with this way )2(something)0( is)]298.8 norm0 -[(evaluated when the macro is defined, not when it is expanded.)]287.8 norm2 -[(You may omit parameter defaults from the macro definition, in which case the parameter default is taken to be)]270.8 norm1 -[(blank. This can be useful for macros which can take a variable number of parameters, since the )2(%0)0( token \(see)]259.8 norm0 -[{/section-4.3.6 xl}(section 4.3.6){el}(\) allows you to determine how many parameters were really passed to the macro call.)]248.8 norm2 -[(This defaulting mechanism can be combined with the greedy-parameter mechanism; so the )2(die)0( macro above)]231.8 norm1 -[(could be made more powerful, and more useful, by changing the first line of the definition to)]220.8 norm2 -[2(%macro die 0-1+ "Painful program death has occurred.",13,10)]203.8 code3 -[(The maximum parameter count can be infinite, denoted by )2(*)0(. In this case, of course, it is impossible to)]186.8 norm1 -[(provide a )1(full)0( set of default parameters. Examples of this usage are shown in ){/section-4.3.7 xl}(section 4.3.7){el}(.)]175.8 norm2 -[{/section-4.3.6 xa}2(%0)0(: )(Macro Parameter Counter)](4.3.6)156.6 subh3 -[(The parameter reference )2(%0)0( will return a numeric constant giving the number of parameters received, that is,)]139.6 norm1 -[(if )2(%0)0( is n then )2(%)0(n is the last parameter. )2(%0)0( is mostly useful for macros that can take a variable number of)]128.6 norm0 -[(parameters. It can be used as an argument to )2(%rep)0( \(see ){/section-4.5 xl}(section 4.5){el}(\) in order to iterate through all the)]117.6 norm0 -[(parameters of a macro. Examples are given in ){/section-4.3.7 xl}(section 4.3.7){el}(.)]106.6 norm2 -(45)pageodd -restore showpage -%%Page: 46 46 -%%BeginPageSetup -save -%%EndPageSetup -/46 pa -[{/section-4.3.7 xa}2(%rotate)0(: )(Rotating Macro Parameters)](4.3.7)678.8 subh3 -[(Unix shell programmers will be familiar with the )2(shift)0( shell command, which allows the arguments passed)]661.8 norm1 -[(to a shell script \(referenced as )2($1)0(, )2($2)0( and so on\) to be moved left by one place, so that the argument)]650.8 norm0 -[(previously referenced as )2($2)0( becomes available as )2($1)0(, and the argument previously referenced as )2($1)0( is no)]639.8 norm0 -[(longer available at all.)]628.8 norm2 -[(NASM provides a similar mechanism, in the form of )2(%rotate)0(. As its name suggests, it differs from the)]611.8 norm1 -[(Unix )2(shift)0( in that no parameters are lost: parameters rotated off the left end of the argument list reappear on)]600.8 norm0 -[(the right, and vice versa.)]589.8 norm2 -[2(%rotate)0( is invoked with a single numeric argument \(which may be an expression\). The macro parameters)]572.8 norm1 -[(are rotated to the left by that many places. If the argument to )2(%rotate)0( is negative, the macro parameters are)]561.8 norm0 -[(rotated to the right.)]550.8 norm2 -[(So a pair of macros to save and restore a set of registers might work as follows:)]533.8 norm3 -[2(%macro multipush 1-* )]516.8 code1 -[2()]505.8 code0 -[2( %rep %0 )]494.8 code0 -[2( push %1 )]483.8 code0 -[2( %rotate 1 )]472.8 code0 -[2( %endrep )]461.8 code0 -[2()]450.8 code0 -[2(%endmacro)]439.8 code2 -[(This macro invokes the )2(PUSH)0( instruction on each of its arguments in turn, from left to right. It begins by)]422.8 norm1 -[(pushing its first argument, )2(%1)0(, then invokes )2(%rotate)0( to move all the arguments one place to the left, so that)]411.8 norm0 -[(the original second argument is now available as )2(%1)0(. Repeating this procedure as many times as there were)]400.8 norm0 -[(arguments \(achieved by supplying )2(%0)0( as the argument to )2(%rep)0(\) causes each argument in turn to be pushed.)]389.8 norm2 -[(Note also the use of )2(*)0( as the maximum parameter count, indicating that there is no upper limit on the number)]372.8 norm1 -[(of parameters you may supply to the )2(multipush)0( macro.)]361.8 norm2 -[(It would be convenient, when using this macro, to have a )2(POP)0( equivalent, which )1(didn't)0( require the arguments)]344.8 norm1 -[(to be given in reverse order. Ideally, you would write the )2(multipush)0( macro call, then cut-and-paste the)]333.8 norm0 -[(line to where the pop needed to be done, and change the name of the called macro to )2(multipop)0(, and the)]322.8 norm0 -[(macro would take care of popping the registers in the opposite order from the one in which they were pushed.)]311.8 norm2 -[(This can be done by the following definition:)]294.8 norm3 -[2(%macro multipop 1-* )]277.8 code1 -[2()]266.8 code0 -[2( %rep %0 )]255.8 code0 -[2( %rotate -1 )]244.8 code0 -[2( pop %1 )]233.8 code0 -[2( %endrep )]222.8 code0 -[2()]211.8 code0 -[2(%endmacro)]200.8 code2 -[(This macro begins by rotating its arguments one place to the )1(right)0(, so that the original )1(last)0( argument appears)]183.8 norm1 -[(as )2(%1)0(. This is then popped, and the arguments are rotated right again, so the second-to-last argument)]172.8 norm0 -[(becomes )2(%1)0(. Thus the arguments are iterated through in reverse order.)]161.8 norm2 -[{/section-4.3.8 xa}(Concatenating Macro Parameters)](4.3.8)142.6 subh3 -[(NASM can concatenate macro parameters and macro indirection constructs on to other text surrounding them.)]125.6 norm1 -[(This allows you to declare a family of symbols, for example, in a macro definition. If, for example, you)]114.6 norm0 -[(wanted to generate a table of key codes along with offsets into the table, you could code something like)]103.6 norm2 -(46)pageeven -restore showpage -%%Page: 47 47 -%%BeginPageSetup -save -%%EndPageSetup -/47 pa -[2(%macro keytab_entry 2 )]681 code1 -[2()]670 code0 -[2( keypos%1 equ $-keytab )]659 code0 -[2( db %2 )]648 code0 -[2()]637 code0 -[2(%endmacro )]626 code0 -[2()]615 code0 -[2(keytab: )]604 code0 -[2( keytab_entry F1,128+1 )]593 code0 -[2( keytab_entry F2,128+2 )]582 code0 -[2( keytab_entry Return,13)]571 code2 -[(which would expand to)]554 norm3 -[2(keytab: )]537 code1 -[2(keyposF1 equ $-keytab )]526 code0 -[2( db 128+1 )]515 code0 -[2(keyposF2 equ $-keytab )]504 code0 -[2( db 128+2 )]493 code0 -[2(keyposReturn equ $-keytab )]482 code0 -[2( db 13)]471 code2 -[(You can just as easily concatenate text on to the other end of a macro parameter, by writing )2(%1foo)0(.)]454 norm3 -[(If you need to append a )1(digit)0( to a macro parameter, for example defining labels )2(foo1)0( and )2(foo2)0( when passed)]437 norm1 -[(the parameter )2(foo)0(, you can't code )2(%11)0( because that would be taken as the eleventh macro parameter.)]426 norm0 -[(Instead, you must code )2(%{1}1)0(, which will separate the first )2(1)0( \(giving the number of the macro parameter\))]415 norm0 -[(from the second \(literal text to be concatenated to the parameter\).)]404 norm2 -[(This concatenation can also be applied to other preprocessor in-line objects, such as macro-local labels)]387 norm1 -[(\(){/section-4.3.3 xl}(section 4.3.3){el}(\) and context-local labels \(){/section-4.7.2 xl}(section 4.7.2){el}(\). In all cases, ambiguities in syntax can be resolved by)]376 norm0 -[(enclosing everything after the )2(%)0( sign and before the literal text in braces: so )2(%{%foo}bar)0( concatenates the)]365 norm0 -[(text )2(bar)0( to the end of the real name of the macro-local label )2(%%foo)0(. \(This is unnecessary, since the form)]354 norm0 -[(NASM uses for the real names of macro-local labels means that the two usages )2(%{%foo}bar)0( and)]343 norm0 -[2(%%foobar)0( would both expand to the same thing anyway; nevertheless, the capability is there.\))]332 norm2 -[(The single-line macro indirection construct, )2(%[...])0( \(){/section-4.1.3 xl}(section 4.1.3){el}(\), behaves the same way as macro)]315 norm1 -[(parameters for the purpose of concatenation.)]304 norm2 -[(See also the )2(%+)0( operator, ){/section-4.1.4 xl}(section 4.1.4){el}(.)]287 norm3 -[{/section-4.3.9 xa}(Condition Codes as Macro Parameters)](4.3.9)267.8 subh3 -[(NASM can give special treatment to a macro parameter which contains a condition code. For a start, you can)]250.8 norm1 -[(refer to the macro parameter )2(%1)0( by means of the alternative syntax )2(%+1)0(, which informs NASM that this)]239.8 norm0 -[(macro parameter is supposed to contain a condition code, and will cause the preprocessor to report an error)]228.8 norm0 -[(message if the macro is called with a parameter which is )1(not)0( a valid condition code.)]217.8 norm2 -[(Far more usefully, though, you can refer to the macro parameter by means of )2(%-1)0(, which NASM will expand)]200.8 norm1 -[(as the )1(inverse)0( condition code. So the )2(retz)0( macro defined in ){/section-4.3.3 xl}(section 4.3.3){el}( can be replaced by a general)]189.8 norm0 -[(conditional-return macro like this:)]178.8 norm2 -[2(%macro retc 1 )]161.8 code1 -[2()]150.8 code0 -[2( j%-1 %%skip )]139.8 code0 -[2( ret )]128.8 code0 -[2( %%skip: )]117.8 code0 -(47)pageodd -restore showpage -%%Page: 48 48 -%%BeginPageSetup -save -%%EndPageSetup -/48 pa -[2()]681 code0 -[2(%endmacro)]670 code2 -[(This macro can now be invoked using calls like )2(retc ne)0(, which will cause the conditional-jump instruction)]653 norm1 -[(in the macro expansion to come out as )2(JE)0(, or )2(retc po)0( which will make the jump a )2(JPE)0(.)]642 norm2 -[(The )2(%+1)0( macro-parameter reference is quite happy to interpret the arguments )2(CXZ)0( and )2(ECXZ)0( as valid)]625 norm1 -[(condition codes; however, )2(%-1)0( will report an error if passed either of these, because no inverse condition)]614 norm0 -[(code exists.)]603 norm2 -[{/section-4.3.10 xa}(Disabling Listing Expansion)](4.3.10)583.8 subh3 -[(When NASM is generating a listing file from your program, it will generally expand multi-line macros by)]566.8 norm1 -[(means of writing the macro call and then listing each line of the expansion. This allows you to see which)]555.8 norm0 -[(instructions in the macro expansion are generating what code; however, for some macros this clutters the)]544.8 norm0 -[(listing up unnecessarily.)]533.8 norm2 -[(NASM therefore provides the )2(.nolist)0( qualifier, which you can include in a macro definition to inhibit the)]516.8 norm1 -[(expansion of the macro in the listing file. The )2(.nolist)0( qualifier comes directly after the number of)]505.8 norm0 -[(parameters, like this:)]494.8 norm2 -[2(%macro foo 1.nolist)]477.8 code3 -[(Or like this:)]460.8 norm3 -[2(%macro bar 1-5+.nolist a,b,c,d,e,f,g,h)]443.8 code3 -[{/section-4.3.11 xa}(Undefining Multi-Line Macros: )2(%unmacro)](4.3.11)424.6 subh3 -[(Multi-line macros can be removed with the )2(%unmacro)0( directive. Unlike the )2(%undef)0( directive, however,)]407.6 norm1 -[2(%unmacro)0( takes an argument specification, and will only remove )(exact matches with that argument)]396.6 norm0 -[(specification.)]385.6 norm2 -[(For example:)]368.6 norm3 -[2(%macro foo 1-3 )]351.6 code1 -[2( ; Do something )]340.6 code0 -[2(%endmacro )]329.6 code0 -[2(%unmacro foo 1-3)]318.6 code2 -[(removes the previously defined macro )2(foo)0(, but)]301.6 norm3 -[2(%macro bar 1-3 )]284.6 code1 -[2( ; Do something )]273.6 code0 -[2(%endmacro )]262.6 code0 -[2(%unmacro bar 1)]251.6 code2 -[(does )1(not)0( remove the macro )2(bar)0(, since the argument specification does not match exactly.)]234.6 norm3 -[{/section-4.3.12 xa}(Exiting Multi-Line Macros: )2(%exitmacro)](4.3.12)215.4 subh3 -[(Multi-line macro expansions can be arbitrarily terminated with the )2(%exitmacro)0( directive.)]198.4 norm3 -[(For example:)]181.4 norm3 -[2(%macro foo 1-3 )]164.4 code1 -[2( ; Do something )]153.4 code0 -[2( %if<condition> )]142.4 code0 -[2( %exitmacro )]131.4 code0 -[2( %endif )]120.4 code0 -(48)pageeven -restore showpage -%%Page: 49 49 -%%BeginPageSetup -save -%%EndPageSetup -/49 pa -[2( ; Do something )]681 code0 -[2(%endmacro)]670 code2 -[{/section-4.4 xa}(Conditional Assembly)](4.4)648.6 head3 -[(Similarly to the C preprocessor, NASM allows sections of a source file to be assembled only if certain)]631.6 norm1 -[(conditions are met. The general syntax of this feature looks like this:)]620.6 norm2 -[2(%if<condition> )]603.6 code1 -[2( ; some code which only appears if <condition> is met )]592.6 code0 -[2(%elif<condition2> )]581.6 code0 -[2( ; only appears if <condition> is not met but <condition2> is )]570.6 code0 -[2(%else )]559.6 code0 -[2( ; this appears if neither <condition> nor <condition2> was met )]548.6 code0 -[2(%endif)]537.6 code2 -[(The inverse forms )2(%ifn)0( and )2(%elifn)0( are also supported.)]520.6 norm3 -[(The )2(%else)0( clause is optional, as is the )2(%elif)0( clause. You can have more than one )2(%elif)0( clause as well.)]503.6 norm3 -[(There are a number of variants of the )2(%if)0( directive. Each has its corresponding )2(%elif)0(, )2(%ifn)0(, and )2(%elifn)]486.6 norm1 -[(directives; for example, the equivalents to the )2(%ifdef)0( directive are )2(%elifdef)0(, )2(%ifndef)0(, and)]475.6 norm0 -[2(%elifndef)0(.)]464.6 norm2 -[{/section-4.4.1 xa}2(%ifdef)0(: Testing Single-Line Macro Existence)](4.4.1)445.4 subh3 -[(Beginning a conditional-assembly block with the line )2(%ifdef MACRO)0( will assemble the subsequent code if,)]428.4 norm1 -[(and only if, a single-line macro called )2(MACRO)0( is defined. If not, then the )2(%elif)0( and )2(%else)0( blocks \(if any\))]417.4 norm0 -[(will be processed instead.)]406.4 norm2 -[(For example, when debugging a program, you might want to write code such as)]389.4 norm3 -[2( ; perform some function )]372.4 code1 -[2(%ifdef DEBUG )]361.4 code0 -[2( writefile 2,"Function performed successfully",13,10 )]350.4 code0 -[2(%endif )]339.4 code0 -[2( ; go and do something else)]328.4 code2 -[(Then you could use the command-line option )2(-dDEBUG)0( to create a version of the program which produced)]311.4 norm1 -[(debugging messages, and remove the option to generate the final release version of the program.)]300.4 norm2 -[(You can test for a macro )1(not)0( being defined by using )2(%ifndef)0( instead of )2(%ifdef)0(. You can also test for)]283.4 norm1 -[(macro definitions in )2(%elif)0( blocks by using )2(%elifdef)0( and )2(%elifndef)0(.)]272.4 norm2 -[{/section-4.4.2 xa}2(%ifmacro)0(: Testing Multi-Line Macro Existence)](4.4.2)253.2 subh3 -[(The )2(%ifmacro)0( directive operates in the same way as the )2(%ifdef)0( directive, except that it checks for the)]236.2 norm1 -[(existence of a multi-line macro.)]225.2 norm2 -[(For example, you may be working with a large project and not have control over the macros in a library. You)]208.2 norm1 -[(may want to create a macro with one name if it doesn't already exist, and another name if one with that name)]197.2 norm0 -[(does exist.)]186.2 norm2 -[(The )2(%ifmacro)0( is considered true if defining a macro with the given name and number of arguments would)]169.2 norm1 -[(cause a definitions conflict. For example:)]158.2 norm2 -[2(%ifmacro MyMacro 1-3 )]141.2 code1 -[2()]130.2 code0 -[2( %error "MyMacro 1-3" causes a conflict with an existing macro. )]119.2 code0 -[2()]108.2 code0 -(49)pageodd -restore showpage -%%Page: 50 50 -%%BeginPageSetup -save -%%EndPageSetup -/50 pa -[2(%else )]681 code0 -[2()]670 code0 -[2( %macro MyMacro 1-3 )]659 code0 -[2()]648 code0 -[2( ; insert code to define the macro )]637 code0 -[2()]626 code0 -[2( %endmacro )]615 code0 -[2()]604 code0 -[2(%endif)]593 code2 -[(This will create the macro "MyMacro 1-3" if no macro already exists which would conflict with it, and emits)]576 norm1 -[(a warning if there would be a definition conflict.)]565 norm2 -[(You can test for the macro not existing by using the )2(%ifnmacro)0( instead of )2(%ifmacro)0(. Additional tests can)]548 norm1 -[(be performed in )2(%elif)0( blocks by using )2(%elifmacro)0( and )2(%elifnmacro)0(.)]537 norm2 -[{/section-4.4.3 xa}2(%ifctx)0(: Testing the Context Stack)](4.4.3)517.8 subh3 -[(The conditional-assembly construct )2(%ifctx)0( will cause the subsequent code to be assembled if and only if)]500.8 norm1 -[(the top context on the preprocessor's context stack has the same name as one of the arguments. As with)]489.8 norm0 -[2(%ifdef)0(, the inverse and )2(%elif)0( forms )2(%ifnctx)0(, )2(%elifctx)0( and )2(%elifnctx)0( are also supported.)]478.8 norm2 -[(For more details of the context stack, see ){/section-4.7 xl}(section 4.7){el}(. For a sample use of )2(%ifctx)0(, see ){/section-4.7.5 xl}(section 4.7.5){el}(.)]461.8 norm3 -[{/section-4.4.4 xa}2(%if)0(: Testing Arbitrary Numeric Expressions)](4.4.4)442.6 subh3 -[(The conditional-assembly construct )2(%if expr)0( will cause the subsequent code to be assembled if and only if)]425.6 norm1 -[(the value of the numeric expression )2(expr)0( is non-zero. An example of the use of this feature is in deciding)]414.6 norm0 -[(when to break out of a )2(%rep)0( preprocessor loop: see ){/section-4.5 xl}(section 4.5){el}( for a detailed example.)]403.6 norm2 -[(The expression given to )2(%if)0(, and its counterpart )2(%elif)0(, is a critical expression \(see ){/section-3.8 xl}(section 3.8){el}(\).)]386.6 norm3 -[2(%if)0( extends the normal NASM expression syntax, by providing a set of )(relational operators which are not)]369.6 norm1 -[(normally available in expressions. The operators )2(=)0(, )2(<)0(, )2(>)0(, )2(<=)0(, )2(>=)0( and )2(<>)0( test equality, less-than, greater-than,)]358.6 norm0 -[(less-or-equal, greater-or-equal and not-equal respectively. The C-like forms )2(==)0( and )2(!=)0( are supported as)]347.6 norm0 -[(alternative forms of )2(=)0( and )2(<>)0(. In addition, low-priority logical operators )2(&&)0(, )2(^^)0( and )2(||)0( are provided,)]336.6 norm0 -[(supplying )(logical AND, )(logical XOR and )(logical OR. These work like the C logical operators \(although C has)]325.6 norm0 -[(no logical XOR\), in that they always return either 0 or 1, and treat any non-zero input as 1 \(so that )2(^^)0(, for)]314.6 norm0 -[(example, returns 1 if exactly one of its inputs is zero, and 0 otherwise\). The relational operators also return 1)]303.6 norm0 -[(for true and 0 for false.)]292.6 norm2 -[(Like other )2(%if)0( constructs, )2(%if)0( has a counterpart )2(%elif)0(, and negative forms )2(%ifn)0( and )2(%elifn)0(.)]275.6 norm3 -[{/section-4.4.5 xa}2(%ifidn)0( and )2(%ifidni)0(: Testing Exact Text Identity)](4.4.5)256.4 subh3 -[(The construct )2(%ifidn text1,text2)0( will cause the subsequent code to be assembled if and only if)]239.4 norm1 -[2(text1)0( and )2(text2)0(, after expanding single-line macros, are identical pieces of text. Differences in white)]228.4 norm0 -[(space are not counted.)]217.4 norm2 -[2(%ifidni)0( is similar to )2(%ifidn)0(, but is )(case-insensitive.)]200.4 norm3 -[(For example, the following macro pushes a register or number on the stack, and allows you to treat )2(IP)0( as a)]183.4 norm1 -[(real register:)]172.4 norm2 -[2(%macro pushparam 1 )]155.4 code1 -[2()]144.4 code0 -[2( %ifidni %1,ip )]133.4 code0 -[2( call %%label )]122.4 code0 -[2( %%label: )]111.4 code0 -[2( %else )]100.4 code0 -(50)pageeven -restore showpage -%%Page: 51 51 -%%BeginPageSetup -save -%%EndPageSetup -/51 pa -[2( push %1 )]681 code0 -[2( %endif )]670 code0 -[2()]659 code0 -[2(%endmacro)]648 code2 -[(Like other )2(%if)0( constructs, )2(%ifidn)0( has a counterpart )2(%elifidn)0(, and negative forms )2(%ifnidn)0( and)]631 norm1 -[2(%elifnidn)0(. Similarly, )2(%ifidni)0( has counterparts )2(%elifidni)0(, )2(%ifnidni)0( and )2(%elifnidni)0(.)]620 norm2 -[{/section-4.4.6 xa}2(%ifid)0(, )2(%ifnum)0(, )2(%ifstr)0(: Testing Token Types)](4.4.6)600.8 subh3 -[(Some macros will want to perform different tasks depending on whether they are passed a number, a string, or)]583.8 norm1 -[(an identifier. For example, a string output macro might want to be able to cope with being passed either a)]572.8 norm0 -[(string constant or a pointer to an existing string.)]561.8 norm2 -[(The conditional assembly construct )2(%ifid)0(, taking one parameter \(which may be blank\), assembles the)]544.8 norm1 -[(subsequent code if and only if the first token in the parameter exists and is an identifier. )2(%ifnum)0( works)]533.8 norm0 -[(similarly, but tests for the token being a numeric constant; )2(%ifstr)0( tests for it being a string.)]522.8 norm2 -[(For example, the )2(writefile)0( macro defined in ){/section-4.3.4 xl}(section 4.3.4){el}( can be extended to take advantage of )2(%ifstr)]505.8 norm1 -[(in the following fashion:)]494.8 norm2 -[2(%macro writefile 2-3+ )]477.8 code1 -[2()]466.8 code0 -[2( %ifstr %2 )]455.8 code0 -[2( jmp %%endstr )]444.8 code0 -[2( %if %0 = 3 )]433.8 code0 -[2( %%str: db %2,%3 )]422.8 code0 -[2( %else )]411.8 code0 -[2( %%str: db %2 )]400.8 code0 -[2( %endif )]389.8 code0 -[2( %%endstr: mov dx,%%str )]378.8 code0 -[2( mov cx,%%endstr-%%str )]367.8 code0 -[2( %else )]356.8 code0 -[2( mov dx,%2 )]345.8 code0 -[2( mov cx,%3 )]334.8 code0 -[2( %endif )]323.8 code0 -[2( mov bx,%1 )]312.8 code0 -[2( mov ah,0x40 )]301.8 code0 -[2( int 0x21 )]290.8 code0 -[2()]279.8 code0 -[2(%endmacro)]268.8 code2 -[(Then the )2(writefile)0( macro can cope with being called in either of the following two ways:)]251.8 norm3 -[2( writefile [file], strpointer, length )]234.8 code1 -[2( writefile [file], "hello", 13, 10)]223.8 code2 -[(In the first, )2(strpointer)0( is used as the address of an already-declared string, and )2(length)0( is used as its)]206.8 norm1 -[(length; in the second, a string is given to the macro, which therefore declares it itself and works out the)]195.8 norm0 -[(address and length for itself.)]184.8 norm2 -[(Note the use of )2(%if)0( inside the )2(%ifstr)0(: this is to detect whether the macro was passed two arguments \(so the)]167.8 norm1 -[(string would be a single string constant, and )2(db %2)0( would be adequate\) or more \(in which case, all but the)]156.8 norm0 -[(first two would be lumped together into )2(%3)0(, and )2(db %2,%3)0( would be required\).)]145.8 norm2 -[(The usual )2(%elif)0(..., )2(%ifn)0(..., and )2(%elifn)0(... versions exist for each of )2(%ifid)0(, )2(%ifnum)0( and )2(%ifstr)0(.)]128.8 norm3 -(51)pageodd -restore showpage -%%Page: 52 52 -%%BeginPageSetup -save -%%EndPageSetup -/52 pa -[{/section-4.4.7 xa}2(%iftoken)0(: Test for a Single Token)](4.4.7)678.8 subh3 -[(Some macros will want to do different things depending on if it is passed a single token \(e.g. paste it to)]661.8 norm1 -[(something else using )2(%+)0(\) versus a multi-token sequence.)]650.8 norm2 -[(The conditional assembly construct )2(%iftoken)0( assembles the subsequent code if and only if the expanded)]633.8 norm1 -[(parameters consist of exactly one token, possibly surrounded by whitespace.)]622.8 norm2 -[(For example:)]605.8 norm3 -[2(%iftoken 1)]588.8 code3 -[(will assemble the subsequent code, but)]571.8 norm3 -[2(%iftoken -1)]554.8 code3 -[(will not, since )2(-1)0( contains two tokens: the unary minus operator )2(-)0(, and the number )2(1)0(.)]537.8 norm3 -[(The usual )2(%eliftoken)0(, )2(%ifntoken)0(, and )2(%elifntoken)0( variants are also provided.)]520.8 norm3 -[{/section-4.4.8 xa}2(%ifempty)0(: Test for Empty Expansion)](4.4.8)501.6 subh3 -[(The conditional assembly construct )2(%ifempty)0( assembles the subsequent code if and only if the expanded)]484.6 norm1 -[(parameters do not contain any tokens at all, whitespace excepted.)]473.6 norm2 -[(The usual )2(%elifempty)0(, )2(%ifnempty)0(, and )2(%elifnempty)0( variants are also provided.)]456.6 norm3 -[{/section-4.5 xa}(Preprocessor Loops)(: )2(%rep)](4.5)435.2 head3 -[(NASM's )2(TIMES)0( prefix, though useful, cannot be used to invoke a multi-line macro multiple times, because)]418.2 norm1 -[(it is processed by NASM after macros have already been expanded. Therefore NASM provides another form)]407.2 norm0 -[(of loop, this time at the preprocessor level: )2(%rep)0(.)]396.2 norm2 -[(The directives )2(%rep)0( and )2(%endrep)0( \()2(%rep)0( takes a numeric argument, which can be an expression; )2(%endrep)]379.2 norm1 -[(takes no arguments\) can be used to enclose a chunk of code, which is then replicated as many times as)]368.2 norm0 -[(specified by the preprocessor:)]357.2 norm2 -[2(%assign i 0 )]340.2 code1 -[2(%rep 64 )]329.2 code0 -[2( inc word [table+2*i] )]318.2 code0 -[2(%assign i i+1 )]307.2 code0 -[2(%endrep)]296.2 code2 -[(This will generate a sequence of 64 )2(INC)0( instructions, incrementing every word of memory from )2([table])0( to)]279.2 norm1 -[2([table+126])0(.)]268.2 norm2 -[(For more complex termination conditions, or to break out of a repeat loop part way along, you can use the)]251.2 norm1 -[2(%exitrep)0( directive to terminate the loop, like this:)]240.2 norm2 -[2(fibonacci: )]223.2 code1 -[2(%assign i 0 )]212.2 code0 -[2(%assign j 1 )]201.2 code0 -[2(%rep 100 )]190.2 code0 -[2(%if j > 65535 )]179.2 code0 -[2( %exitrep )]168.2 code0 -[2(%endif )]157.2 code0 -[2( dw j )]146.2 code0 -[2(%assign k j+i )]135.2 code0 -[2(%assign i j )]124.2 code0 -[2(%assign j k )]113.2 code0 -[2(%endrep )]102.2 code0 -(52)pageeven -restore showpage -%%Page: 53 53 -%%BeginPageSetup -save -%%EndPageSetup -/53 pa -[2()]681 code0 -[2(fib_number equ \($-fibonacci\)/2)]670 code2 -[(This produces a list of all the Fibonacci numbers that will fit in 16 bits. Note that a maximum repeat count)]653 norm1 -[(must still be given to )2(%rep)0(. This is to prevent the possibility of NASM getting into an infinite loop in the)]642 norm0 -[(preprocessor, which \(on multitasking or multi-user systems\) would typically cause all the system memory to)]631 norm0 -[(be gradually used up and other applications to start crashing.)]620 norm2 -[{/section-4.6 xa}(Source Files and Dependencies)](4.6)598.6 head3 -[(These commands allow you to split your sources into multiple files.)]581.6 norm3 -[{/section-4.6.1 xa}2(%include)0(: )(Including Other Files)](4.6.1)562.4 subh3 -[(Using, once again, a very similar syntax to the C preprocessor, NASM's preprocessor lets you include other)]545.4 norm1 -[(source files into your code. This is done by the use of the )2(%include)0( directive:)]534.4 norm2 -[2(%include "macros.mac")]517.4 code3 -[(will include the contents of the file )2(macros.mac)0( into the source file containing the )2(%include)0( directive.)]500.4 norm3 -[(Include files are )(searched for in the current directory \(the directory you're in when you run NASM, as)]483.4 norm1 -[(opposed to the location of the NASM executable or the location of the source file\), plus any directories)]472.4 norm0 -[(specified on the NASM command line using the )2(-i)0( option.)]461.4 norm2 -[(The standard C idiom for preventing a file being included more than once is just as applicable in NASM: if)]444.4 norm1 -[(the file )2(macros.mac)0( has the form)]433.4 norm2 -[2(%ifndef MACROS_MAC )]416.4 code1 -[2( %define MACROS_MAC )]405.4 code0 -[2( ; now define some macros )]394.4 code0 -[2(%endif)]383.4 code2 -[(then including the file more than once will not cause errors, because the second time the file is included)]366.4 norm1 -[(nothing will happen because the macro )2(MACROS_MAC)0( will already be defined.)]355.4 norm2 -[(You can force a file to be included even if there is no )2(%include)0( directive that explicitly includes it, by using)]338.4 norm1 -[(the )2(-p)0( option on the NASM command line \(see ){/section-2.1.17 xl}(section 2.1.17){el}(\).)]327.4 norm2 -[{/section-4.6.2 xa}2(%pathsearch)0(: Search the Include Path)](4.6.2)308.2 subh3 -[(The )2(%pathsearch)0( directive takes a single-line macro name and a filename, and declare or redefines the)]291.2 norm1 -[(specified single-line macro to be the include-path-resolved version of the filename, if the file exists)]280.2 norm0 -[(\(otherwise, it is passed unchanged.\))]269.2 norm2 -[(For example,)]252.2 norm3 -[2(%pathsearch MyFoo "foo.bin")]235.2 code3 -[(... with )2(-Ibins/)0( in the include path may end up defining the macro )2(MyFoo)0( to be )2("bins/foo.bin")0(.)]218.2 norm3 -[{/section-4.6.3 xa}2(%depend)0(: Add Dependent Files)](4.6.3)199 subh3 -[(The )2(%depend)0( directive takes a filename and adds it to the list of files to be emitted as dependency generation)]182 norm1 -[(when the )2(-M)0( options and its relatives \(see ){/section-2.1.4 xl}(section 2.1.4){el}(\) are used. It produces no output.)]171 norm2 -[(This is generally used in conjunction with )2(%pathsearch)0(. For example, a simplified version of the standard)]154 norm1 -[(macro wrapper for the )2(INCBIN)0( directive looks like:)]143 norm2 -[2(%imacro incbin 1-2+ 0 )]126 code1 -[2(%pathsearch dep %1 )]115 code0 -[2(%depend dep )]104 code0 -(53)pageodd -restore showpage -%%Page: 54 54 -%%BeginPageSetup -save -%%EndPageSetup -/54 pa -[2( incbin dep,%2 )]681 code0 -[2(%endmacro)]670 code2 -[(This first resolves the location of the file into the macro )2(dep)0(, then adds it to the dependency lists, and finally)]653 norm1 -[(issues the assembler-level )2(INCBIN)0( directive.)]642 norm2 -[{/section-4.6.4 xa}2(%use)0(: Include Standard Macro Package)](4.6.4)622.8 subh3 -[(The )2(%use)0( directive is similar to )2(%include)0(, but rather than including the contents of a file, it includes a)]605.8 norm1 -[(named standard macro package. The standard macro packages are part of NASM, and are described in ){/chapter-5 xl}(chapter){el}]594.8 norm0 -[{/chapter-5 xl}(5){el}(.)]583.8 norm2 -[(Unlike the )2(%include)0( directive, package names for the )2(%use)0( directive do not require quotes, but quotes are)]566.8 norm1 -[(permitted. In NASM 2.04 and 2.05 the unquoted form would be macro-expanded; this is no longer true. Thus,)]555.8 norm0 -[(the following lines are equivalent:)]544.8 norm2 -[2(%use altreg )]527.8 code1 -[2(%use 'altreg')]516.8 code2 -[(Standard macro packages are protected from multiple inclusion. When a standard macro package is used, a)]499.8 norm1 -[(testable single-line macro of the form )2(__USE_)1(package)2(__)0( is also defined, see ){/section-4.11.8 xl}(section 4.11.8){el}(.)]488.8 norm2 -[{/section-4.7 xa}(The )(Context Stack)](4.7)467.4 head3 -[(Having labels that are local to a macro definition is sometimes not quite powerful enough: sometimes you)]450.4 norm1 -[(want to be able to share labels between several macro calls. An example might be a )2(REPEAT)0( ... )2(UNTIL)0( loop,)]439.4 norm0 -[(in which the expansion of the )2(REPEAT)0( macro would need to be able to refer to a label which the )2(UNTIL)]428.4 norm0 -[(macro had defined. However, for such a macro you would also want to be able to nest these loops.)]417.4 norm2 -[(NASM provides this level of power by means of a )1(context stack)0(. The preprocessor maintains a stack of)]400.4 norm1 -[1(contexts)0(, each of which is characterized by a name. You add a new context to the stack using the )2(%push)]389.4 norm0 -[(directive, and remove one using )2(%pop)0(. You can define labels that are local to a particular context on the stack.)]378.4 norm2 -[{/section-4.7.1 xa}2(%push)0( and )2(%pop)0(: )(Creating and Removing Contexts)](4.7.1)359.2 subh3 -[(The )2(%push)0( directive is used to create a new context and place it on the top of the context stack. )2(%push)0( takes)]342.2 norm1 -[(an optional argument, which is the name of the context. For example:)]331.2 norm2 -[2(%push foobar)]314.2 code3 -[(This pushes a new context called )2(foobar)0( on the stack. You can have several contexts on the stack with the)]297.2 norm1 -[(same name: they can still be distinguished. If no name is given, the context is unnamed \(this is normally used)]286.2 norm0 -[(when both the )2(%push)0( and the )2(%pop)0( are inside a single macro definition.\))]275.2 norm2 -[(The directive )2(%pop)0(, taking one optional argument, removes the top context from the context stack and)]258.2 norm1 -[(destroys it, along with any labels associated with it. If an argument is given, it must match the name of the)]247.2 norm0 -[(current context, otherwise it will issue an error.)]236.2 norm2 -[{/section-4.7.2 xa}(Context-Local Labels)](4.7.2)217 subh3 -[(Just as the usage )2(%%foo)0( defines a label which is local to the particular macro call in which it is used, the)]200 norm1 -[(usage )2(%$foo)0( is used to define a label which is local to the context on the top of the context stack. So the)]189 norm0 -[2(REPEAT)0( and )2(UNTIL)0( example given above could be implemented by means of:)]178 norm2 -[2(%macro repeat 0 )]161 code1 -[2()]150 code0 -[2( %push repeat )]139 code0 -[2( %$begin: )]128 code0 -[2()]117 code0 -[2(%endmacro )]106 code0 -(54)pageeven -restore showpage -%%Page: 55 55 -%%BeginPageSetup -save -%%EndPageSetup -/55 pa -[2()]681 code0 -[2(%macro until 1 )]670 code0 -[2()]659 code0 -[2( j%-1 %$begin )]648 code0 -[2( %pop )]637 code0 -[2()]626 code0 -[2(%endmacro)]615 code2 -[(and invoked by means of, for example,)]598 norm3 -[2( mov cx,string )]581 code1 -[2( repeat )]570 code0 -[2( add cx,3 )]559 code0 -[2( scasb )]548 code0 -[2( until e)]537 code2 -[(which would scan every fourth byte of a string in search of the byte in )2(AL)0(.)]520 norm3 -[(If you need to define, or access, labels local to the context )1(below)0( the top one on the stack, you can use)]503 norm1 -[2(%$$foo)0(, or )2(%$$$foo)0( for the context below that, and so on.)]492 norm2 -[{/section-4.7.3 xa}(Context-Local Single-Line Macros)](4.7.3)472.8 subh3 -[(NASM also allows you to define single-line macros which are local to a particular context, in just the same)]455.8 norm1 -[(way:)]444.8 norm2 -[2(%define %$localmac 3)]427.8 code3 -[(will define the single-line macro )2(%$localmac)0( to be local to the top context on the stack. Of course, after a)]410.8 norm1 -[(subsequent )2(%push)0(, it can then still be accessed by the name )2(%$$localmac)0(.)]399.8 norm2 -[{/section-4.7.4 xa}2(%repl)0(: )(Renaming a Context)](4.7.4)380.6 subh3 -[(If you need to change the name of the top context on the stack \(in order, for example, to have it respond)]363.6 norm1 -[(differently to )2(%ifctx)0(\), you can execute a )2(%pop)0( followed by a )2(%push)0(; but this will have the side effect of)]352.6 norm0 -[(destroying all context-local labels and macros associated with the context that was just popped.)]341.6 norm2 -[(NASM provides the directive )2(%repl)0(, which )1(replaces)0( a context with a different name, without touching the)]324.6 norm1 -[(associated macros and labels. So you could replace the destructive code)]313.6 norm2 -[2(%pop )]296.6 code1 -[2(%push newname)]285.6 code2 -[(with the non-destructive version )2(%repl newname)0(.)]268.6 norm3 -[{/section-4.7.5 xa}(Example Use of the )(Context Stack: )(Block IFs)](4.7.5)249.4 subh3 -[(This example makes use of almost all the context-stack features, including the conditional-assembly)]232.4 norm1 -[(construct )2(%ifctx)0(, to implement a block IF statement as a set of macros.)]221.4 norm2 -[2(%macro if 1 )]204.4 code1 -[2()]193.4 code0 -[2( %push if )]182.4 code0 -[2( j%-1 %$ifnot )]171.4 code0 -[2()]160.4 code0 -[2(%endmacro )]149.4 code0 -[2()]138.4 code0 -[2(%macro else 0 )]127.4 code0 -[2()]116.4 code0 -[2( %ifctx if )]105.4 code0 -(55)pageodd -restore showpage -%%Page: 56 56 -%%BeginPageSetup -save -%%EndPageSetup -/56 pa -[2( %repl else )]681 code0 -[2( jmp %$ifend )]670 code0 -[2( %$ifnot: )]659 code0 -[2( %else )]648 code0 -[2( %error "expected `if' before `else'" )]637 code0 -[2( %endif )]626 code0 -[2()]615 code0 -[2(%endmacro )]604 code0 -[2()]593 code0 -[2(%macro endif 0 )]582 code0 -[2()]571 code0 -[2( %ifctx if )]560 code0 -[2( %$ifnot: )]549 code0 -[2( %pop )]538 code0 -[2( %elifctx else )]527 code0 -[2( %$ifend: )]516 code0 -[2( %pop )]505 code0 -[2( %else )]494 code0 -[2( %error "expected `if' or `else' before `endif'" )]483 code0 -[2( %endif )]472 code0 -[2()]461 code0 -[2(%endmacro)]450 code2 -[(This code is more robust than the )2(REPEAT)0( and )2(UNTIL)0( macros given in ){/section-4.7.2 xl}(section 4.7.2){el}(, because it uses)]433 norm1 -[(conditional assembly to check that the macros are issued in the right order \(for example, not calling )2(endif)]422 norm0 -[(before )2(if)0(\) and issues a )2(%error)0( if they're not.)]411 norm2 -[(In addition, the )2(endif)0( macro has to be able to cope with the two distinct cases of either directly following an)]394 norm1 -[2(if)0(, or following an )2(else)0(. It achieves this, again, by using conditional assembly to do different things)]383 norm0 -[(depending on whether the context on top of the stack is )2(if)0( or )2(else)0(.)]372 norm2 -[(The )2(else)0( macro has to preserve the context on the stack, in order to have the )2(%$ifnot)0( referred to by the )2(if)]355 norm1 -[(macro be the same as the one defined by the )2(endif)0( macro, but has to change the context's name so that)]344 norm0 -[2(endif)0( will know there was an intervening )2(else)0(. It does this by the use of )2(%repl)0(.)]333 norm2 -[(A sample usage of these macros might look like:)]316 norm3 -[2( cmp ax,bx )]299 code1 -[2()]288 code0 -[2( if ae )]277 code0 -[2( cmp bx,cx )]266 code0 -[2()]255 code0 -[2( if ae )]244 code0 -[2( mov ax,cx )]233 code0 -[2( else )]222 code0 -[2( mov ax,bx )]211 code0 -[2( endif )]200 code0 -[2()]189 code0 -[2( else )]178 code0 -[2( cmp ax,cx )]167 code0 -[2()]156 code0 -[2( if ae )]145 code0 -[2( mov ax,cx )]134 code0 -[2( endif )]123 code0 -[2()]112 code0 -[2( endif)]101 code2 -(56)pageeven -restore showpage -%%Page: 57 57 -%%BeginPageSetup -save -%%EndPageSetup -/57 pa -[(The block-)2(IF)0( macros handle nesting quite happily, by means of pushing another context, describing the inner)]681 norm1 -[2(if)0(, on top of the one describing the outer )2(if)0(; thus )2(else)0( and )2(endif)0( always refer to the last unmatched )2(if)]670 norm0 -[(or )2(else)0(.)]659 norm2 -[{/section-4.8 xa}(Stack Relative Preprocessor Directives)](4.8)637.6 head3 -[(The following preprocessor directives provide a way to use labels to refer to local variables allocated on the)]620.6 norm1 -[(stack.)]609.6 norm2 -[2(%arg)0( \(see ){/section-4.8.1 xl}(section 4.8.1){el}(\))]592.6 bull3 -[2(%stacksize)0( \(see ){/section-4.8.2 xl}(section 4.8.2){el}(\))]575.6 bull3 -[2(%local)0( \(see ){/section-4.8.3 xl}(section 4.8.3){el}(\))]558.6 bull3 -[{/section-4.8.1 xa}2(%arg)0( Directive)](4.8.1)539.4 subh3 -[(The )2(%arg)0( directive is used to simplify the handling of parameters passed on the stack. Stack based parameter)]522.4 norm1 -[(passing is used by many high level languages, including C, C++ and Pascal.)]511.4 norm2 -[(While NASM has macros which attempt to duplicate this functionality \(see ){/section-8.4.5 xl}(section 8.4.5){el}(\), the syntax is not)]494.4 norm1 -[(particularly convenient to use. and is not TASM compatible. Here is an example which shows the use of)]483.4 norm0 -[2(%arg)0( without any external macros:)]472.4 norm2 -[2(some_function: )]455.4 code1 -[2()]444.4 code0 -[2( %push mycontext ; save the current context )]433.4 code0 -[2( %stacksize large ; tell NASM to use bp )]422.4 code0 -[2( %arg i:word, j_ptr:word )]411.4 code0 -[2()]400.4 code0 -[2( mov ax,[i] )]389.4 code0 -[2( mov bx,[j_ptr] )]378.4 code0 -[2( add ax,[bx] )]367.4 code0 -[2( ret )]356.4 code0 -[2()]345.4 code0 -[2( %pop ; restore original context)]334.4 code2 -[(This is similar to the procedure defined in ){/section-8.4.5 xl}(section 8.4.5){el}( and adds the value in i to the value pointed to by j_ptr)]317.4 norm1 -[(and returns the sum in the ax register. See ){/section-4.7.1 xl}(section 4.7.1){el}( for an explanation of )2(push)0( and )2(pop)0( and the use of)]306.4 norm0 -[(context stacks.)]295.4 norm2 -[{/section-4.8.2 xa}2(%stacksize)0( Directive)](4.8.2)276.2 subh3 -[(The )2(%stacksize)0( directive is used in conjunction with the )2(%arg)0( \(see ){/section-4.8.1 xl}(section 4.8.1){el}(\) and the )2(%local)0( \(see)]259.2 norm1 -[{/section-4.8.3 xl}(section 4.8.3){el}(\) directives. It tells NASM the default size to use for subsequent )2(%arg)0( and )2(%local)0( directives.)]248.2 norm0 -[(The )2(%stacksize)0( directive takes one required argument which is one of )2(flat)0(, )2(flat64)0(, )2(large)0( or )2(small)0(.)]237.2 norm2 -[2(%stacksize flat)]220.2 code3 -[(This form causes NASM to use stack-based parameter addressing relative to )2(ebp)0( and it assumes that a near)]203.2 norm1 -[(form of call was used to get to this label \(i.e. that )2(eip)0( is on the stack\).)]192.2 norm2 -[2(%stacksize flat64)]175.2 code3 -[(This form causes NASM to use stack-based parameter addressing relative to )2(rbp)0( and it assumes that a near)]158.2 norm1 -[(form of call was used to get to this label \(i.e. that )2(rip)0( is on the stack\).)]147.2 norm2 -[2(%stacksize large)]130.2 code3 -[(This form uses )2(bp)0( to do stack-based parameter addressing and assumes that a far form of call was used to get)]113.2 norm1 -[(to this address \(i.e. that )2(ip)0( and )2(cs)0( are on the stack\).)]102.2 norm2 -(57)pageodd -restore showpage -%%Page: 58 58 -%%BeginPageSetup -save -%%EndPageSetup -/58 pa -[2(%stacksize small)]681 code3 -[(This form also uses )2(bp)0( to address stack parameters, but it is different from )2(large)0( because it also assumes)]664 norm1 -[(that the old value of bp is pushed onto the stack \(i.e. it expects an )2(ENTER)0( instruction\). In other words, it)]653 norm0 -[(expects that )2(bp)0(, )2(ip)0( and )2(cs)0( are on the top of the stack, underneath any local space which may have been)]642 norm0 -[(allocated by )2(ENTER)0(. This form is probably most useful when used in combination with the )2(%local)0( directive)]631 norm0 -[(\(see ){/section-4.8.3 xl}(section 4.8.3){el}(\).)]620 norm2 -[{/section-4.8.3 xa}2(%local)0( Directive)](4.8.3)600.8 subh3 -[(The )2(%local)0( directive is used to simplify the use of local temporary stack variables allocated in a stack)]583.8 norm1 -[(frame. Automatic local variables in C are an example of this kind of variable. The )2(%local)0( directive is most)]572.8 norm0 -[(useful when used with the )2(%stacksize)0( \(see ){/section-4.8.2 xl}(section 4.8.2){el}( and is also compatible with the )2(%arg)0( directive)]561.8 norm0 -[(\(see ){/section-4.8.1 xl}(section 4.8.1){el}(\). It allows simplified reference to variables on the stack which have been allocated typically)]550.8 norm0 -[(by using the )2(ENTER)0( instruction. An example of its use is the following:)]539.8 norm2 -[2(silly_swap: )]522.8 code1 -[2()]511.8 code0 -[2( %push mycontext ; save the current context )]500.8 code0 -[2( %stacksize small ; tell NASM to use bp )]489.8 code0 -[2( %assign %$localsize 0 ; see text for explanation )]478.8 code0 -[2( %local old_ax:word, old_dx:word )]467.8 code0 -[2()]456.8 code0 -[2( enter %$localsize,0 ; see text for explanation )]445.8 code0 -[2( mov [old_ax],ax ; swap ax & bx )]434.8 code0 -[2( mov [old_dx],dx ; and swap dx & cx )]423.8 code0 -[2( mov ax,bx )]412.8 code0 -[2( mov dx,cx )]401.8 code0 -[2( mov bx,[old_ax] )]390.8 code0 -[2( mov cx,[old_dx] )]379.8 code0 -[2( leave ; restore old bp )]368.8 code0 -[2( ret ; )]357.8 code0 -[2()]346.8 code0 -[2( %pop ; restore original context)]335.8 code2 -[(The )2(%$localsize)0( variable is used internally by the )2(%local)0( directive and )1(must)0( be defined within the)]318.8 norm1 -[(current context before the )2(%local)0( directive may be used. Failure to do so will result in one expression)]307.8 norm0 -[(syntax error for each )2(%local)0( variable declared. It then may be used in the construction of an appropriately)]296.8 norm0 -[(sized ENTER instruction as shown in the example.)]285.8 norm2 -[{/section-4.9 xa}(Reporting )(User-Defined Errors: )2(%error)0(, )2(%warning)0(, )2(%fatal)](4.9)264.4 head3 -[(The preprocessor directive )2(%error)0( will cause NASM to report an error if it occurs in assembled code. So if)]247.4 norm1 -[(other users are going to try to assemble your source files, you can ensure that they define the right macros by)]236.4 norm0 -[(means of code like this:)]225.4 norm2 -[2(%ifdef F1 )]208.4 code1 -[2( ; do some setup )]197.4 code0 -[2(%elifdef F2 )]186.4 code0 -[2( ; do some different setup )]175.4 code0 -[2(%else )]164.4 code0 -[2( %error "Neither F1 nor F2 was defined." )]153.4 code0 -[2(%endif)]142.4 code2 -[(Then any user who fails to understand the way your code is supposed to be assembled will be quickly warned)]125.4 norm1 -[(of their mistake, rather than having to wait until the program crashes on being run and then not knowing what)]114.4 norm0 -[(went wrong.)]103.4 norm2 -(58)pageeven -restore showpage -%%Page: 59 59 -%%BeginPageSetup -save -%%EndPageSetup -/59 pa -[(Similarly, )2(%warning)0( issues a warning, but allows assembly to continue:)]681 norm3 -[2(%ifdef F1 )]664 code1 -[2( ; do some setup )]653 code0 -[2(%elifdef F2 )]642 code0 -[2( ; do some different setup )]631 code0 -[2(%else )]620 code0 -[2( %warning "Neither F1 nor F2 was defined, assuming F1." )]609 code0 -[2( %define F1 )]598 code0 -[2(%endif)]587 code2 -[2(%error)0( and )2(%warning)0( are issued only on the final assembly pass. This makes them safe to use in)]570 norm1 -[(conjunction with tests that depend on symbol values.)]559 norm2 -[2(%fatal)0( terminates assembly immediately, regardless of pass. This is useful when there is no point in)]542 norm1 -[(continuing the assembly further, and doing so is likely just going to cause a spew of confusing error messages.)]531 norm2 -[(It is optional for the message string after )2(%error)0(, )2(%warning)0( or )2(%fatal)0( to be quoted. If it is )1(not)0(, then)]514 norm1 -[(single-line macros are expanded in it, which can be used to display more information to the user. For)]503 norm0 -[(example:)]492 norm2 -[2(%if foo > 64 )]475 code1 -[2( %assign foo_over foo-64 )]464 code0 -[2( %error foo is foo_over bytes too large )]453 code0 -[2(%endif)]442 code2 -[{/section-4.10 xa}(Other Preprocessor Directives)](4.10)420.6 head3 -[(NASM also has preprocessor directives which allow access to information from external sources. Currently)]403.6 norm1 -[(they include:)]392.6 norm2 -[2(%line)0( enables NASM to correctly handle the output of another preprocessor \(see ){/section-4.10.1 xl}(section 4.10.1){el}(\).)]375.6 bull3 -[2(%!)0( enables NASM to read in the value of an environment variable, which can then be used in your)]358.6 bull1 -[(program \(see ){/section-4.10.2 xl}(section 4.10.2){el}(\).)]347.6 bull2 -[{/section-4.10.1 xa}2(%line)0( Directive)](4.10.1)328.4 subh3 -[(The )2(%line)0( directive is used to notify NASM that the input line corresponds to a specific line number in)]311.4 norm1 -[(another file. Typically this other file would be an original source file, with the current NASM input being the)]300.4 norm0 -[(output of a pre-processor. The )2(%line)0( directive allows NASM to output messages which indicate the line)]289.4 norm0 -[(number of the original source file, instead of the file that is being read by NASM.)]278.4 norm2 -[(This preprocessor directive is not generally of use to programmers, by may be of interest to preprocessor)]261.4 norm1 -[(authors. The usage of the )2(%line)0( preprocessor directive is as follows:)]250.4 norm2 -[2(%line nnn[+mmm] [filename])]233.4 code3 -[(In this directive, )2(nnn)0( identifies the line of the original source file which this line corresponds to. )2(mmm)0( is an)]216.4 norm1 -[(optional parameter which specifies a line increment value; each line of the input file read in is considered to)]205.4 norm0 -[(correspond to )2(mmm)0( lines of the original source file. Finally, )2(filename)0( is an optional parameter which)]194.4 norm0 -[(specifies the file name of the original source file.)]183.4 norm2 -[(After reading a )2(%line)0( preprocessor directive, NASM will report all file name and line numbers relative to)]166.4 norm1 -[(the values specified therein.)]155.4 norm2 -(59)pageodd -restore showpage -%%Page: 60 60 -%%BeginPageSetup -save -%%EndPageSetup -/60 pa -[{/section-4.10.2 xa}2(%!<env>)0(: Read an environment variable.)](4.10.2)678.8 subh3 -[(The )2(%!<env>)0( directive makes it possible to read the value of an environment variable at assembly time. This)]661.8 norm1 -[(could, for example, be used to store the contents of an environment variable into a string, which could be used)]650.8 norm0 -[(at some other point in your code.)]639.8 norm2 -[(For example, suppose that you have an environment variable )2(FOO)0(, and you want the contents of )2(FOO)0( to be)]622.8 norm1 -[(embedded in your program. You could do that as follows:)]611.8 norm2 -[2(%defstr FOO %!FOO)]594.8 code3 -[(See ){/section-4.1.8 xl}(section 4.1.8){el}( for notes on the )2(%defstr)0( directive.)]577.8 norm3 -[{/section-4.11 xa}(Standard Macros)](4.11)556.4 head3 -[(NASM defines a set of standard macros, which are already defined when it starts to process any source file. If)]539.4 norm1 -[(you really need a program to be assembled with no pre-defined macros, you can use the )2(%clear)0( directive to)]528.4 norm0 -[(empty the preprocessor of everything but context-local preprocessor variables and single-line macros.)]517.4 norm2 -[(Most )(user-level assembler directives \(see ){/chapter-6 xl}(chapter 6){el}(\) are implemented as macros which invoke primitive)]500.4 norm1 -[(directives; these are described in ){/chapter-6 xl}(chapter 6){el}(. The rest of the standard macro set is described here.)]489.4 norm2 -[{/section-4.11.1 xa}(NASM Version Macros)](4.11.1)470.2 subh3 -[(The single-line macros )2(__NASM_MAJOR__)0(, )2(__NASM_MINOR__)0(, )2(__NASM_SUBMINOR__)0( and)]453.2 norm1 -[2(___NASM_PATCHLEVEL__)0( expand to the major, minor, subminor and patch level parts of the )(version)]442.2 norm0 -[(number of NASM being used. So, under NASM 0.98.32p1 for example, )2(__NASM_MAJOR__)0( would be)]431.2 norm0 -[(defined to be 0, )2(__NASM_MINOR__)0( would be defined as 98, )2(__NASM_SUBMINOR__)0( would be defined to)]420.2 norm0 -[(32, and )2(___NASM_PATCHLEVEL__)0( would be defined as 1.)]409.2 norm2 -[(Additionally, the macro )2(__NASM_SNAPSHOT__)0( is defined for automatically generated snapshot releases)]392.2 norm1 -[1(only)0(.)]381.2 norm2 -[{/section-4.11.2 xa}2(__NASM_VERSION_ID__)0(: )(NASM Version ID)](4.11.2)362 subh3 -[(The single-line macro )2(__NASM_VERSION_ID__)0( expands to a dword integer representing the full version)]345 norm1 -[(number of the version of nasm being used. The value is the equivalent to )2(__NASM_MAJOR__)0(,)]334 norm0 -[2(__NASM_MINOR__)0(, )2(__NASM_SUBMINOR__)0( and )2(___NASM_PATCHLEVEL__)0( concatenated to produce a)]323 norm0 -[(single doubleword. Hence, for 0.98.32p1, the returned number would be equivalent to:)]312 norm2 -[2( dd 0x00622001)]295 code3 -[(or)]278 norm3 -[2( db 1,32,98,0)]261 code3 -[(Note that the above lines are generate exactly the same code, the second line is used just to give an indication)]244 norm1 -[(of the order that the separate values will be present in memory.)]233 norm2 -[{/section-4.11.3 xa}2(__NASM_VER__)0(: )(NASM Version string)](4.11.3)213.8 subh3 -[(The single-line macro )2(__NASM_VER__)0( expands to a string which defines the version number of nasm being)]196.8 norm1 -[(used. So, under NASM 0.98.32 for example,)]185.8 norm2 -[2( db __NASM_VER__)]168.8 code3 -[(would expand to)]151.8 norm3 -[2( db "0.98.32")]134.8 code3 -(60)pageeven -restore showpage -%%Page: 61 61 -%%BeginPageSetup -save -%%EndPageSetup -/61 pa -[{/section-4.11.4 xa}2(__FILE__)0( and )2(__LINE__)0(: File Name and Line Number)](4.11.4)678.8 subh3 -[(Like the C preprocessor, NASM allows the user to find out the file name and line number containing the)]661.8 norm1 -[(current instruction. The macro )2(__FILE__)0( expands to a string constant giving the name of the current input)]650.8 norm0 -[(file \(which may change through the course of assembly if )2(%include)0( directives are used\), and )2(__LINE__)]639.8 norm0 -[(expands to a numeric constant giving the current line number in the input file.)]628.8 norm2 -[(These macros could be used, for example, to communicate debugging information to a macro, since invoking)]611.8 norm1 -[2(__LINE__)0( inside a macro definition \(either single-line or multi-line\) will return the line number of the)]600.8 norm0 -[(macro )1(call)0(, rather than )1(definition)0(. So to determine where in a piece of code a crash is occurring, for example,)]589.8 norm0 -[(one could write a routine )2(stillhere)0(, which is passed a line number in )2(EAX)0( and outputs something like)]578.8 norm0 -[(`line 155: still here'. You could then write a macro)]567.8 norm2 -[2(%macro notdeadyet 0 )]550.8 code1 -[2()]539.8 code0 -[2( push eax )]528.8 code0 -[2( mov eax,__LINE__ )]517.8 code0 -[2( call stillhere )]506.8 code0 -[2( pop eax )]495.8 code0 -[2()]484.8 code0 -[2(%endmacro)]473.8 code2 -[(and then pepper your code with calls to )2(notdeadyet)0( until you find the crash point.)]456.8 norm3 -[{/section-4.11.5 xa}2(__BITS__)0(: Current BITS Mode)](4.11.5)437.6 subh3 -[(The )2(__BITS__)0( standard macro is updated every time that the BITS mode is set using the )2(BITS XX)0( or)]420.6 norm1 -[2([BITS XX])0( directive, where XX is a valid mode number of 16, 32 or 64. )2(__BITS__)0( receives the specified)]409.6 norm0 -[(mode number and makes it globally available. This can be very useful for those who utilize mode-dependent)]398.6 norm0 -[(macros.)]387.6 norm2 -[{/section-4.11.6 xa}2(__OUTPUT_FORMAT__)0(: Current Output Format)](4.11.6)368.4 subh3 -[(The )2(__OUTPUT_FORMAT__)0( standard macro holds the current Output Format, as given by the )2(-f)0( option or)]351.4 norm1 -[(NASM's default. Type )2(nasm -hf)0( for a list.)]340.4 norm2 -[2(%ifidn __OUTPUT_FORMAT__, win32 )]323.4 code1 -[2( %define NEWLINE 13, 10 )]312.4 code0 -[2(%elifidn __OUTPUT_FORMAT__, elf32 )]301.4 code0 -[2( %define NEWLINE 10 )]290.4 code0 -[2(%endif)]279.4 code2 -[{/section-4.11.7 xa}(Assembly Date and Time Macros)](4.11.7)260.2 subh3 -[(NASM provides a variety of macros that represent the timestamp of the assembly session.)]243.2 norm3 -[(The )2(__DATE__)0( and )2(__TIME__)0( macros give the assembly date and time as strings, in ISO 8601 format)]226.2 bull1 -[(\()2("YYYY-MM-DD")0( and )2("HH:MM:SS")0(, respectively.\))]215.2 bull2 -[(The )2(__DATE_NUM__)0( and )2(__TIME_NUM__)0( macros give the assembly date and time in numeric form; in)]198.2 bull1 -[(the format )2(YYYYMMDD)0( and )2(HHMMSS)0( respectively.)]187.2 bull2 -[(The )2(__UTC_DATE__)0( and )2(__UTC_TIME__)0( macros give the assembly date and time in universal time)]170.2 bull1 -[(\(UTC\) as strings, in ISO 8601 format \()2("YYYY-MM-DD")0( and )2("HH:MM:SS")0(, respectively.\) If the host)]159.2 bull0 -[(platform doesn't provide UTC time, these macros are undefined.)]148.2 bull2 -[(The )2(__UTC_DATE_NUM__)0( and )2(__UTC_TIME_NUM__)0( macros give the assembly date and time)]131.2 bull1 -[(universal time \(UTC\) in numeric form; in the format )2(YYYYMMDD)0( and )2(HHMMSS)0( respectively. If the host)]120.2 bull0 -[(platform doesn't provide UTC time, these macros are undefined.)]109.2 bull2 -(61)pageodd -restore showpage -%%Page: 62 62 -%%BeginPageSetup -save -%%EndPageSetup -/62 pa -[(The )2(__POSIX_TIME__)0( macro is defined as a number containing the number of seconds since the POSIX)]681 bull1 -[(epoch, 1 January 1970 00:00:00 UTC; excluding any leap seconds. This is computed using UTC time if)]670 bull0 -[(available on the host platform, otherwise it is computed using the local time as if it was UTC.)]659 bull2 -[(All instances of time and date macros in the same assembly session produce consistent output. For example,)]642 norm1 -[(in an assembly session started at 42 seconds after midnight on January 1, 2010 in Moscow \(timezone UTC+3\))]631 norm0 -[(these macros would have the following values, assuming, of course, a properly configured environment with a)]620 norm0 -[(correct clock:)]609 norm2 -[2( __DATE__ "2010-01-01" )]592 code1 -[2( __TIME__ "00:00:42" )]581 code0 -[2( __DATE_NUM__ 20100101 )]570 code0 -[2( __TIME_NUM__ 000042 )]559 code0 -[2( __UTC_DATE__ "2009-12-31" )]548 code0 -[2( __UTC_TIME__ "21:00:42" )]537 code0 -[2( __UTC_DATE_NUM__ 20091231 )]526 code0 -[2( __UTC_TIME_NUM__ 210042 )]515 code0 -[2( __POSIX_TIME__ 1262293242)]504 code2 -[{/section-4.11.8 xa}2(__USE_)1(package)2(__)0(: Package Include Test)](4.11.8)484.8 subh3 -[(When a standard macro package \(see ){/chapter-5 xl}(chapter 5){el}(\) is included with the )2(%use)0( directive \(see ){/section-4.6.4 xl}(section 4.6.4){el}(\), a)]467.8 norm1 -[(single-line macro of the form )2(__USE_)1(package)2(__)0( is automatically defined. This allows testing if a particular)]456.8 norm0 -[(package is invoked or not.)]445.8 norm2 -[(For example, if the )2(altreg)0( package is included \(see ){/section-5.1 xl}(section 5.1){el}(\), then the macro )2(__USE_ALTREG__)0( is)]428.8 norm1 -[(defined.)]417.8 norm2 -[{/section-4.11.9 xa}2(__PASS__)0(: Assembly Pass)](4.11.9)398.6 subh3 -[(The macro )2(__PASS__)0( is defined to be )2(1)0( on preparatory passes, and )2(2)0( on the final pass. In preprocess-only)]381.6 norm1 -[(mode, it is set to )2(3)0(, and when running only to generate dependencies \(due to the )2(-M)0( or )2(-MG)0( option, see)]370.6 norm0 -[{/section-2.1.4 xl}(section 2.1.4){el}(\) it is set to )2(0)0(.)]359.6 norm2 -[1(Avoid using this macro if at all possible. It is tremendously easy to generate very strange errors by misusing)]342.6 norm1 -[1(it, and the semantics may change in future versions of NASM.)]331.6 norm2 -[{/section-4.11.10 xa}2(STRUC)0( and )2(ENDSTRUC)0(: )(Declaring Structure Data Types)](4.11.10)312.4 subh3 -[(The core of NASM contains no intrinsic means of defining data structures; instead, the preprocessor is)]295.4 norm1 -[(sufficiently powerful that data structures can be implemented as a set of macros. The macros )2(STRUC)0( and)]284.4 norm0 -[2(ENDSTRUC)0( are used to define a structure data type.)]273.4 norm2 -[2(STRUC)0( takes one or two parameters. The first parameter is the name of the data type. The second, optional)]256.4 norm1 -[(parameter is the base offset of the structure. The name of the data type is defined as a symbol with the value)]245.4 norm0 -[(of the base offset, and the name of the data type with the suffix )2(_size)0( appended to it is defined as an )2(EQU)]234.4 norm0 -[(giving the size of the structure. Once )2(STRUC)0( has been issued, you are defining the structure, and should)]223.4 norm0 -[(define fields using the )2(RESB)0( family of pseudo-instructions, and then invoke )2(ENDSTRUC)0( to finish the)]212.4 norm0 -[(definition.)]201.4 norm2 -[(For example, to define a structure called )2(mytype)0( containing a longword, a word, a byte and a string of bytes,)]184.4 norm1 -[(you might code)]173.4 norm2 -[2(struc mytype )]156.4 code1 -[2()]145.4 code0 -[2( mt_long: resd 1 )]134.4 code0 -[2( mt_word: resw 1 )]123.4 code0 -[2( mt_byte: resb 1 )]112.4 code0 -[2( mt_str: resb 32 )]101.4 code0 -(62)pageeven -restore showpage -%%Page: 63 63 -%%BeginPageSetup -save -%%EndPageSetup -/63 pa -[2()]681 code0 -[2(endstruc)]670 code2 -[(The above code defines six symbols: )2(mt_long)0( as 0 \(the offset from the beginning of a )2(mytype)0( structure to)]653 norm1 -[(the longword field\), )2(mt_word)0( as 4, )2(mt_byte)0( as 6, )2(mt_str)0( as 7, )2(mytype_size)0( as 39, and )2(mytype)0( itself)]642 norm0 -[(as zero.)]631 norm2 -[(The reason why the structure type name is defined at zero by default is a side effect of allowing structures to)]614 norm1 -[(work with the local label mechanism: if your structure members tend to have the same names in more than)]603 norm0 -[(one structure, you can define the above structure like this:)]592 norm2 -[2(struc mytype )]575 code1 -[2()]564 code0 -[2( .long: resd 1 )]553 code0 -[2( .word: resw 1 )]542 code0 -[2( .byte: resb 1 )]531 code0 -[2( .str: resb 32 )]520 code0 -[2()]509 code0 -[2(endstruc)]498 code2 -[(This defines the offsets to the structure fields as )2(mytype.long)0(, )2(mytype.word)0(, )2(mytype.byte)0( and)]481 norm1 -[2(mytype.str)0(.)]470 norm2 -[(NASM, since it has no )1(intrinsic)0( structure support, does not support any form of period notation to refer to the)]453 norm1 -[(elements of a structure once you have one \(except the above local-label notation\), so code such as)]442 norm0 -[2(mov ax,[mystruc.mt_word])0( is not valid. )2(mt_word)0( is a constant just like any other constant, so the)]431 norm0 -[(correct syntax is )2(mov ax,[mystruc+mt_word])0( or )2(mov ax,[mystruc+mytype.word])0(.)]420 norm2 -[(Sometimes you only have the address of the structure displaced by an offset. For example, consider this)]403 norm1 -[(standard stack frame setup:)]392 norm2 -[2(push ebp )]375 code1 -[2(mov ebp, esp )]364 code0 -[2(sub esp, 40)]353 code2 -[(In this case, you could access an element by subtracting the offset:)]336 norm3 -[2(mov [ebp - 40 + mytype.word], ax)]319 code3 -[(However, if you do not want to repeat this offset, you can use \22640 as a base offset:)]302 norm3 -[2(struc mytype, -40)]285 code3 -[(And access an element this way:)]268 norm3 -[2(mov [ebp + mytype.word], ax)]251 code3 -[{/section-4.11.11 xa}2(ISTRUC)0(, )2(AT)0( and )2(IEND)0(: Declaring )(Instances of Structures)](4.11.11)231.8 subh3 -[(Having defined a structure type, the next thing you typically want to do is to declare instances of that structure)]214.8 norm1 -[(in your data segment. NASM provides an easy way to do this in the )2(ISTRUC)0( mechanism. To declare a)]203.8 norm0 -[(structure of type )2(mytype)0( in a program, you code something like this:)]192.8 norm2 -[2(mystruc: )]175.8 code1 -[2( istruc mytype )]164.8 code0 -[2()]153.8 code0 -[2( at mt_long, dd 123456 )]142.8 code0 -[2( at mt_word, dw 1024 )]131.8 code0 -[2( at mt_byte, db 'x' )]120.8 code0 -[2( at mt_str, db 'hello, world', 13, 10, 0 )]109.8 code0 -(63)pageodd -restore showpage -%%Page: 64 64 -%%BeginPageSetup -save -%%EndPageSetup -/64 pa -[2()]681 code0 -[2( iend)]670 code2 -[(The function of the )2(AT)0( macro is to make use of the )2(TIMES)0( prefix to advance the assembly position to the)]653 norm1 -[(correct point for the specified structure field, and then to declare the specified data. Therefore the structure)]642 norm0 -[(fields must be declared in the same order as they were specified in the structure definition.)]631 norm2 -[(If the data to go in a structure field requires more than one source line to specify, the remaining source lines)]614 norm1 -[(can easily come after the )2(AT)0( line. For example:)]603 norm2 -[2( at mt_str, db 123,134,145,156,167,178,189 )]586 code1 -[2( db 190,100,0)]575 code2 -[(Depending on personal taste, you can also omit the code part of the )2(AT)0( line completely, and start the structure)]558 norm1 -[(field on the next line:)]547 norm2 -[2( at mt_str )]530 code1 -[2( db 'hello, world' )]519 code0 -[2( db 13,10,0)]508 code2 -[{/section-4.11.12 xa}2(ALIGN)0( and )2(ALIGNB)0(: Data Alignment)](4.11.12)488.8 subh3 -[(The )2(ALIGN)0( and )2(ALIGNB)0( macros provides a convenient way to align code or data on a word, longword,)]471.8 norm1 -[(paragraph or other boundary. \(Some assemblers call this directive )2(EVEN)0(.\) The syntax of the )2(ALIGN)0( and)]460.8 norm0 -[2(ALIGNB)0( macros is)]449.8 norm2 -[2( align 4 ; align on 4-byte boundary )]432.8 code1 -[2( align 16 ; align on 16-byte boundary )]421.8 code0 -[2( align 8,db 0 ; pad with 0s rather than NOPs )]410.8 code0 -[2( align 4,resb 1 ; align to 4 in the BSS )]399.8 code0 -[2( alignb 4 ; equivalent to previous line)]388.8 code2 -[(Both macros require their first argument to be a power of two; they both compute the number of additional)]371.8 norm1 -[(bytes required to bring the length of the current section up to a multiple of that power of two, and then apply)]360.8 norm0 -[(the )2(TIMES)0( prefix to their second argument to perform the alignment.)]349.8 norm2 -[(If the second argument is not specified, the default for )2(ALIGN)0( is )2(NOP)0(, and the default for )2(ALIGNB)0( is)]332.8 norm1 -[2(RESB 1)0(. So if the second argument is specified, the two macros are equivalent. Normally, you can just use)]321.8 norm0 -[2(ALIGN)0( in code and data sections and )2(ALIGNB)0( in BSS sections, and never need the second argument except)]310.8 norm0 -[(for special purposes.)]299.8 norm2 -[2(ALIGN)0( and )2(ALIGNB)0(, being simple macros, perform no error checking: they cannot warn you if their first)]282.8 norm1 -[(argument fails to be a power of two, or if their second argument generates more than one byte of code. In each)]271.8 norm0 -[(of these cases they will silently do the wrong thing.)]260.8 norm2 -[2(ALIGNB)0( \(or )2(ALIGN)0( with a second argument of )2(RESB 1)0(\) can be used within structure definitions:)]243.8 norm3 -[2(struc mytype2 )]226.8 code1 -[2()]215.8 code0 -[2( mt_byte: )]204.8 code0 -[2( resb 1 )]193.8 code0 -[2( alignb 2 )]182.8 code0 -[2( mt_word: )]171.8 code0 -[2( resw 1 )]160.8 code0 -[2( alignb 4 )]149.8 code0 -[2( mt_long: )]138.8 code0 -[2( resd 1 )]127.8 code0 -[2( mt_str: )]116.8 code0 -[2( resb 32 )]105.8 code0 -(64)pageeven -restore showpage -%%Page: 65 65 -%%BeginPageSetup -save -%%EndPageSetup -/65 pa -[2()]681 code0 -[2(endstruc)]670 code2 -[(This will ensure that the structure members are sensibly aligned relative to the base of the structure.)]653 norm3 -[(A final caveat: )2(ALIGN)0( and )2(ALIGNB)0( work relative to the beginning of the )1(section)0(, not the beginning of the)]636 norm1 -[(address space in the final executable. Aligning to a 16-byte boundary when the section you're in is only)]625 norm0 -[(guaranteed to be aligned to a 4-byte boundary, for example, is a waste of effort. Again, NASM does not)]614 norm0 -[(check that the section's alignment characteristics are sensible for the use of )2(ALIGN)0( or )2(ALIGNB)0(.)]603 norm2 -[(See also the )2(smartalign)0( standard macro package, ){/section-5.2 xl}(section 5.2){el}(.)]586 norm3 -(65)pageodd -restore showpage -%%Page: 66 66 -%%BeginPageSetup -save -%%EndPageSetup -/66 pa -[{/chapter-5 xa}(Chapter 5: )(Standard Macro Packages)]642.8 chap3 -[(The )2(%use)0( directive \(see ){/section-4.6.4 xl}(section 4.6.4){el}(\) includes one of the standard macro packages included with the NASM)]607.8 norm1 -[(distribution and compiled into the NASM binary. It operates like the )2(%include)0( directive \(see ){/section-4.6.1 xl}(section 4.6.1){el}(\),)]596.8 norm0 -[(but the included contents is provided by NASM itself.)]585.8 norm2 -[(The names of standard macro packages are case insensitive, and can be quoted or not.)]568.8 norm3 -[{/section-5.1 xa}2(altreg)0(: )(Alternate Register Names)](5.1)547.4 head3 -[(The )2(altreg)0( standard macro package provides alternate register names. It provides numeric register names)]530.4 norm1 -[(for all registers \(not just )2(R8)0(\226)2(R15)0(\), the Intel-defined aliases )2(R8L)0(\226)2(R15L)0( for the low bytes of register \(as)]519.4 norm0 -[(opposed to the NASM/AMD standard names )2(R8B)0(\226)2(R15B)0(\), and the names )2(R0H)0(\226)2(R3H)0( \(by analogy with)]508.4 norm0 -[2(R0L)0(\226)2(R3L)0(\) for )2(AH)0(, )2(CH)0(, )2(DH)0(, and )2(BH)0(.)]497.4 norm2 -[(Example use:)]480.4 norm3 -[2(%use altreg )]463.4 code1 -[2()]452.4 code0 -[2(proc: )]441.4 code0 -[2( mov r0l,r3h ; mov al,bh )]430.4 code0 -[2( ret)]419.4 code2 -[(See also ){/section-11.1 xl}(section 11.1){el}(.)]402.4 norm3 -[{/section-5.2 xa}2(smartalign)0(: Smart )2(ALIGN)0( Macro)](5.2)381 head3 -[(The )2(smartalign)0( standard macro package provides for an )2(ALIGN)0( macro which is more powerful than the)]364 norm1 -[(default \(and backwards-compatible\) one \(see ){/section-4.11.12 xl}(section 4.11.12){el}(\). When the )2(smartalign)0( package is enabled,)]353 norm0 -[(when )2(ALIGN)0( is used without a second argument, NASM will generate a sequence of instructions more)]342 norm0 -[(efficient than a series of )2(NOP)0(. Furthermore, if the padding exceeds a specific threshold, then NASM will)]331 norm0 -[(generate a jump over the entire padding sequence.)]320 norm2 -[(The specific instructions generated can be controlled with the new )2(ALIGNMODE)0( macro. This macro takes two)]303 norm1 -[(parameters: one mode, and an optional jump threshold override. The modes are as follows:)]292 norm2 -[2(generic)0(: Works on all x86 CPUs and should have reasonable performance. The default jump threshold)]275 bull1 -[(is 8. This is the default.)]264 bull2 -[2(nop)0(: Pad out with )2(NOP)0( instructions. The only difference compared to the standard )2(ALIGN)0( macro is that)]247 bull1 -[(NASM can still jump over a large padding area. The default jump threshold is 16.)]236 bull2 -[2(k7)0(: Optimize for the AMD K7 \(Athlon/Althon XP\). These instructions should still work on all x86 CPUs.)]219 bull1 -[(The default jump threshold is 16.)]208 bull2 -[2(k8)0(: Optimize for the AMD K8 \(Opteron/Althon 64\). These instructions should still work on all x86 CPUs.)]191 bull1 -[(The default jump threshold is 16.)]180 bull2 -[2(p6)0(: Optimize for Intel CPUs. This uses the long )2(NOP)0( instructions first introduced in Pentium Pro. This is)]163 bull1 -[(incompatible with all CPUs of family 5 or lower, as well as some VIA CPUs and several virtualization)]152 bull0 -[(solutions. The default jump threshold is 16.)]141 bull2 -[(The macro )2(__ALIGNMODE__)0( is defined to contain the current alignment mode. A number of other macros)]124 norm1 -[(beginning with )2(__ALIGN_)0( are used internally by this macro package.)]113 norm2 -(66)pageeven -restore showpage -%%Page: 67 67 -%%BeginPageSetup -save -%%EndPageSetup -/67 pa -[{/chapter-6 xa}(Chapter 6: )(Assembler Directives)]642.8 chap3 -[(NASM, though it attempts to avoid the bureaucracy of assemblers like MASM and TASM, is nevertheless)]607.8 norm1 -[(forced to support a )1(few)0( directives. These are described in this chapter.)]596.8 norm2 -[(NASM's directives come in two types: )1(user-level)0( directives and )1(primitive)0( directives. Typically, each)]579.8 norm1 -[(directive has a user-level form and a primitive form. In almost all cases, we recommend that users use the)]568.8 norm0 -[(user-level forms of the directives, which are implemented as macros which call the primitive forms.)]557.8 norm2 -[(Primitive directives are enclosed in square brackets; user-level directives are not.)]540.8 norm3 -[(In addition to the universal directives described in this chapter, each object file format can optionally supply)]523.8 norm1 -[(extra directives in order to control particular features of that file format. These )1(format-specific)0( directives are)]512.8 norm0 -[(documented along with the formats that implement them, in ){/chapter-7 xl}(chapter 7){el}(.)]501.8 norm2 -[{/section-6.1 xa}2(BITS)0(: Specifying Target )(Processor Mode)](6.1)480.4 head3 -[(The )2(BITS)0( directive specifies whether NASM should generate code )(designed to run on a processor operating)]463.4 norm1 -[(in 16-bit mode, 32-bit mode or 64-bit mode. The syntax is )2(BITS XX)0(, where XX is 16, 32 or 64.)]452.4 norm2 -[(In most cases, you should not need to use )2(BITS)0( explicitly. The )2(aout)0(, )2(coff)0(, )2(elf)0(, )2(macho)0(, )2(win32)0( and)]435.4 norm1 -[2(win64)0( object formats, which are designed for use in 32-bit or 64-bit operating systems, all cause NASM to)]424.4 norm0 -[(select 32-bit or 64-bit mode, respectively, by default. The )2(obj)0( object format allows you to specify each)]413.4 norm0 -[(segment you define as either )2(USE16)0( or )2(USE32)0(, and NASM will set its operating mode accordingly, so the)]402.4 norm0 -[(use of the )2(BITS)0( directive is once again unnecessary.)]391.4 norm2 -[(The most likely reason for using the )2(BITS)0( directive is to write 32-bit or 64-bit code in a flat binary file; this)]374.4 norm1 -[(is because the )2(bin)0( output format defaults to 16-bit mode in anticipation of it being used most frequently to)]363.4 norm0 -[(write DOS )2(.COM)0( programs, DOS )2(.SYS)0( device drivers and boot loader software.)]352.4 norm2 -[(You do )1(not)0( need to specify )2(BITS 32)0( merely in order to use 32-bit instructions in a 16-bit DOS program; if)]335.4 norm1 -[(you do, the assembler will generate incorrect code because it will be writing code targeted at a 32-bit)]324.4 norm0 -[(platform, to be run on a 16-bit one.)]313.4 norm2 -[(When NASM is in )2(BITS 16)0( mode, instructions which use 32-bit data are prefixed with an 0x66 byte, and)]296.4 norm1 -[(those referring to 32-bit addresses have an 0x67 prefix. In )2(BITS 32)0( mode, the reverse is true: 32-bit)]285.4 norm0 -[(instructions require no prefixes, whereas instructions using 16-bit data need an 0x66 and those working on)]274.4 norm0 -[(16-bit addresses need an 0x67.)]263.4 norm2 -[(When NASM is in )2(BITS 64)0( mode, most instructions operate the same as they do for )2(BITS 32)0( mode.)]246.4 norm1 -[(However, there are 8 more general and SSE registers, and 16-bit addressing is no longer supported.)]235.4 norm2 -[(The default address size is 64 bits; 32-bit addressing can be selected with the 0x67 prefix. The default)]218.4 norm1 -[(operand size is still 32 bits, however, and the 0x66 prefix selects 16-bit operand size. The )2(REX)0( prefix is used)]207.4 norm0 -[(both to select 64-bit operand size, and to access the new registers. NASM automatically inserts REX prefixes)]196.4 norm0 -[(when necessary.)]185.4 norm2 -[(When the )2(REX)0( prefix is used, the processor does not know how to address the AH, BH, CH or DH \(high 8-bit)]168.4 norm1 -[(legacy\) registers. Instead, it is possible to access the the low 8-bits of the SP, BP SI and DI registers as SPL,)]157.4 norm0 -[(BPL, SIL and DIL, respectively; but only when the REX prefix is used.)]146.4 norm2 -[(The )2(BITS)0( directive has an exactly equivalent primitive form, )2([BITS 16])0(, )2([BITS 32])0( and )2([BITS 64])0(.)]129.4 norm1 -[(The user-level form is a macro which has no function other than to call the primitive form.)]118.4 norm2 -[(Note that the space is neccessary, e.g. )2(BITS32)0( will )1(not)0( work!)]101.4 norm3 -(67)pageodd -restore showpage -%%Page: 68 68 -%%BeginPageSetup -save -%%EndPageSetup -/68 pa -[{/section-6.1.1 xa}2(USE16)0( & )2(USE32)0(: Aliases for BITS)](6.1.1)678.8 subh3 -[(The `)2(USE16)0(' and `)2(USE32)0(' directives can be used in place of `)2(BITS 16)0(' and `)2(BITS 32)0(', for compatibility)]661.8 norm1 -[(with other assemblers.)]650.8 norm2 -[{/section-6.2 xa}2(DEFAULT)0(: Change the assembler defaults)](6.2)629.4 head3 -[(The )2(DEFAULT)0( directive changes the assembler defaults. Normally, NASM defaults to a mode where the)]612.4 norm1 -[(programmer is expected to explicitly specify most features directly. However, this is occationally obnoxious,)]601.4 norm0 -[(as the explicit form is pretty much the only one one wishes to use.)]590.4 norm2 -[(Currently, the only )2(DEFAULT)0( that is settable is whether or not registerless instructions in 64-bit mode are)]573.4 norm1 -[2(RIP)0(\226relative or not. By default, they are absolute unless overridden with the )2(REL)0( specifier \(see ){/section-3.3 xl}(section 3.3){el}(\).)]562.4 norm0 -[(However, if )2(DEFAULT REL)0( is specified, )2(REL)0( is default, unless overridden with the )2(ABS)0( specifier, )1(except)]551.4 norm0 -[1(when used with an FS or GS segment override)0(.)]540.4 norm2 -[(The special handling of )2(FS)0( and )2(GS)0( overrides are due to the fact that these registers are generally used as)]523.4 norm1 -[(thread pointers or other special functions in 64-bit mode, and generating )2(RIP)0(\226relative addresses would be)]512.4 norm0 -[(extremely confusing.)]501.4 norm2 -[2(DEFAULT REL)0( is disabled with )2(DEFAULT ABS)0(.)]484.4 norm3 -[{/section-6.3 xa}2(SECTION)0( or )2(SEGMENT)0(: Changing and )(Defining Sections)](6.3)463 head3 -[(The )2(SECTION)0( directive \()2(SEGMENT)0( is an exactly equivalent synonym\) changes which section of the output)]446 norm1 -[(file the code you write will be assembled into. In some object file formats, the number and names of sections)]435 norm0 -[(are fixed; in others, the user may make up as many as they wish. Hence )2(SECTION)0( may sometimes give an)]424 norm0 -[(error message, or may define a new section, if you try to switch to a section that does not \(yet\) exist.)]413 norm2 -[(The Unix object formats, and the )2(bin)0( object format \(but see ){/section-7.1.3 xl}(section 7.1.3){el}(, all support the )(standardized section)]396 norm1 -[(names )2(.text)0(, )2(.data)0( and )2(.bss)0( for the code, data and uninitialized-data sections. The )2(obj)0( format, by)]385 norm0 -[(contrast, does not recognize these section names as being special, and indeed will strip off the leading period)]374 norm0 -[(of any section name that has one.)]363 norm2 -[{/section-6.3.1 xa}(The )2(__SECT__)0( Macro)](6.3.1)343.8 subh3 -[(The )2(SECTION)0( directive is unusual in that its user-level form functions differently from its primitive form.)]326.8 norm1 -[(The primitive form, )2([SECTION xyz])0(, simply switches the current target section to the one given. The)]315.8 norm0 -[(user-level form, )2(SECTION xyz)0(, however, first defines the single-line macro )2(__SECT__)0( to be the primitive)]304.8 norm0 -[2([SECTION])0( directive which it is about to issue, and then issues it. So the user-level directive)]293.8 norm2 -[2( SECTION .text)]276.8 code3 -[(expands to the two lines)]259.8 norm3 -[2(%define __SECT__ [SECTION .text] )]242.8 code1 -[2( [SECTION .text])]231.8 code2 -[(Users may find it useful to make use of this in their own macros. For example, the )2(writefile)0( macro)]214.8 norm1 -[(defined in ){/section-4.3.4 xl}(section 4.3.4){el}( can be usefully rewritten in the following more sophisticated form:)]203.8 norm2 -[2(%macro writefile 2+ )]186.8 code1 -[2()]175.8 code0 -[2( [section .data] )]164.8 code0 -[2()]153.8 code0 -[2( %%str: db %2 )]142.8 code0 -[2( %%endstr: )]131.8 code0 -[2()]120.8 code0 -[2( __SECT__ )]109.8 code0 -(68)pageeven -restore showpage -%%Page: 69 69 -%%BeginPageSetup -save -%%EndPageSetup -/69 pa -[2()]681 code0 -[2( mov dx,%%str )]670 code0 -[2( mov cx,%%endstr-%%str )]659 code0 -[2( mov bx,%1 )]648 code0 -[2( mov ah,0x40 )]637 code0 -[2( int 0x21 )]626 code0 -[2()]615 code0 -[2(%endmacro)]604 code2 -[(This form of the macro, once passed a string to output, first switches temporarily to the data section of the file,)]587 norm1 -[(using the primitive form of the )2(SECTION)0( directive so as not to modify )2(__SECT__)0(. It then declares its string)]576 norm0 -[(in the data section, and then invokes )2(__SECT__)0( to switch back to )1(whichever)0( section the user was previously)]565 norm0 -[(working in. It thus avoids the need, in the previous version of the macro, to include a )2(JMP)0( instruction to jump)]554 norm0 -[(over the data, and also does not fail if, in a complicated )2(OBJ)0( format module, the user could potentially be)]543 norm0 -[(assembling the code in any of several separate code sections.)]532 norm2 -[{/section-6.4 xa}2(ABSOLUTE)0(: Defining Absolute Labels)](6.4)510.6 head3 -[(The )2(ABSOLUTE)0( directive can be thought of as an alternative form of )2(SECTION)0(: it causes the subsequent)]493.6 norm1 -[(code to be directed at no physical section, but at the hypothetical section starting at the given absolute address.)]482.6 norm0 -[(The only instructions you can use in this mode are the )2(RESB)0( family.)]471.6 norm2 -[2(ABSOLUTE)0( is used as follows:)]454.6 norm3 -[2(absolute 0x1A )]437.6 code1 -[2()]426.6 code0 -[2( kbuf_chr resw 1 )]415.6 code0 -[2( kbuf_free resw 1 )]404.6 code0 -[2( kbuf resw 16)]393.6 code2 -[(This example describes a section of the PC BIOS data area, at segment address 0x40: the above code defines)]376.6 norm1 -[2(kbuf_chr)0( to be 0x1A, )2(kbuf_free)0( to be 0x1C, and )2(kbuf)0( to be 0x1E.)]365.6 norm2 -[(The user-level form of )2(ABSOLUTE)0(, like that of )2(SECTION)0(, redefines the )2(__SECT__)0( macro when it is)]348.6 norm1 -[(invoked.)]337.6 norm2 -[2(STRUC)0( and )2(ENDSTRUC)0( are defined as macros which use )2(ABSOLUTE)0( \(and also )2(__SECT__)0(\).)]320.6 norm3 -[2(ABSOLUTE)0( doesn't have to take an absolute constant as an argument: it can take an expression \(actually, a)]303.6 norm1 -[(critical expression: see ){/section-3.8 xl}(section 3.8){el}(\) and it can be a value in a segment. For example, a TSR can re-use its)]292.6 norm0 -[(setup code as run-time BSS like this:)]281.6 norm2 -[2( org 100h ; it's a .COM program )]264.6 code1 -[2()]253.6 code0 -[2( jmp setup ; setup code comes last )]242.6 code0 -[2()]231.6 code0 -[2( ; the resident part of the TSR goes here )]220.6 code0 -[2(setup: )]209.6 code0 -[2( ; now write the code that installs the TSR here )]198.6 code0 -[2()]187.6 code0 -[2(absolute setup )]176.6 code0 -[2()]165.6 code0 -[2(runtimevar1 resw 1 )]154.6 code0 -[2(runtimevar2 resd 20 )]143.6 code0 -[2()]132.6 code0 -[2(tsr_end:)]121.6 code2 -(69)pageodd -restore showpage -%%Page: 70 70 -%%BeginPageSetup -save -%%EndPageSetup -/70 pa -[(This defines some variables `on top of' the setup code, so that after the setup has finished running, the space it)]681 norm1 -[(took up can be re-used as data storage for the running TSR. The symbol `tsr_end' can be used to calculate the)]670 norm0 -[(total size of the part of the TSR that needs to be made resident.)]659 norm2 -[{/section-6.5 xa}2(EXTERN)0(: )(Importing Symbols from Other Modules)](6.5)637.6 head3 -[2(EXTERN)0( is similar to the MASM directive )2(EXTRN)0( and the C keyword )2(extern)0(: it is used to declare a symbol)]620.6 norm1 -[(which is not defined anywhere in the module being assembled, but is assumed to be defined in some other)]609.6 norm0 -[(module and needs to be referred to by this one. Not every object-file format can support external variables:)]598.6 norm0 -[(the )2(bin)0( format cannot.)]587.6 norm2 -[(The )2(EXTERN)0( directive takes as many arguments as you like. Each argument is the name of a symbol:)]570.6 norm3 -[2(extern _printf )]553.6 code1 -[2(extern _sscanf,_fscanf)]542.6 code2 -[(Some object-file formats provide extra features to the )2(EXTERN)0( directive. In all cases, the extra features are)]525.6 norm1 -[(used by suffixing a colon to the symbol name followed by object-format specific text. For example, the )2(obj)]514.6 norm0 -[(format allows you to declare that the default segment base of an external should be the group )2(dgroup)0( by)]503.6 norm0 -[(means of the directive)]492.6 norm2 -[2(extern _variable:wrt dgroup)]475.6 code3 -[(The primitive form of )2(EXTERN)0( differs from the user-level form only in that it can take only one argument at)]458.6 norm1 -[(a time: the support for multiple arguments is implemented at the preprocessor level.)]447.6 norm2 -[(You can declare the same variable as )2(EXTERN)0( more than once: NASM will quietly ignore the second and)]430.6 norm1 -[(later redeclarations. You can't declare a variable as )2(EXTERN)0( as well as something else, though.)]419.6 norm2 -[{/section-6.6 xa}2(GLOBAL)0(: )(Exporting Symbols to Other Modules)](6.6)398.2 head3 -[2(GLOBAL)0( is the other end of )2(EXTERN)0(: if one module declares a symbol as )2(EXTERN)0( and refers to it, then in)]381.2 norm1 -[(order to prevent linker errors, some other module must actually )1(define)0( the symbol and declare it as )2(GLOBAL)0(.)]370.2 norm0 -[(Some assemblers use the name )2(PUBLIC)0( for this purpose.)]359.2 norm2 -[(The )2(GLOBAL)0( directive applying to a symbol must appear )1(before)0( the definition of the symbol.)]342.2 norm3 -[2(GLOBAL)0( uses the same syntax as )2(EXTERN)0(, except that it must refer to symbols which )1(are)0( defined in the same)]325.2 norm1 -[(module as the )2(GLOBAL)0( directive. For example:)]314.2 norm2 -[2(global _main )]297.2 code1 -[2(_main: )]286.2 code0 -[2( ; some code)]275.2 code2 -[2(GLOBAL)0(, like )2(EXTERN)0(, allows object formats to define private extensions by means of a colon. The )2(elf)]258.2 norm1 -[(object format, for example, lets you specify whether global data items are functions or data:)]247.2 norm2 -[2(global hashlookup:function, hashtable:data)]230.2 code3 -[(Like )2(EXTERN)0(, the primitive form of )2(GLOBAL)0( differs from the user-level form only in that it can take only)]213.2 norm1 -[(one argument at a time.)]202.2 norm2 -[{/section-6.7 xa}2(COMMON)0(: Defining Common Data Areas)](6.7)180.8 head3 -[(The )2(COMMON)0( directive is used to declare )1(common variables)0(. A common variable is much like a global)]163.8 norm1 -[(variable declared in the uninitialized data section, so that)]152.8 norm2 -[2(common intvar 4)]135.8 code3 -[(is similar in function to)]118.8 norm3 -(70)pageeven -restore showpage -%%Page: 71 71 -%%BeginPageSetup -save -%%EndPageSetup -/71 pa -[2(global intvar )]681 code1 -[2(section .bss )]670 code0 -[2()]659 code0 -[2(intvar resd 1)]648 code2 -[(The difference is that if more than one module defines the same common variable, then at link time those)]631 norm1 -[(variables will be )1(merged)0(, and references to )2(intvar)0( in all modules will point at the same piece of memory.)]620 norm2 -[(Like )2(GLOBAL)0( and )2(EXTERN)0(, )2(COMMON)0( supports object-format specific extensions. For example, the )2(obj)]603 norm1 -[(format allows common variables to be NEAR or FAR, and the )2(elf)0( format allows you to specify the)]592 norm0 -[(alignment requirements of a common variable:)]581 norm2 -[2(common commvar 4:near ; works in OBJ )]564 code1 -[2(common intarray 100:4 ; works in ELF: 4 byte aligned)]553 code2 -[(Once again, like )2(EXTERN)0( and )2(GLOBAL)0(, the primitive form of )2(COMMON)0( differs from the user-level form only)]536 norm1 -[(in that it can take only one argument at a time.)]525 norm2 -[{/section-6.8 xa}2(CPU)0(: Defining CPU Dependencies)](6.8)503.6 head3 -[(The )2(CPU)0( directive restricts assembly to those instructions which are available on the specified CPU.)]486.6 norm3 -[(Options are:)]469.6 norm3 -[2(CPU 8086)0( Assemble only 8086 instruction set)]452.6 bull3 -[2(CPU 186)0( Assemble instructions up to the 80186 instruction set)]435.6 bull3 -[2(CPU 286)0( Assemble instructions up to the 286 instruction set)]418.6 bull3 -[2(CPU 386)0( Assemble instructions up to the 386 instruction set)]401.6 bull3 -[2(CPU 486)0( 486 instruction set)]384.6 bull3 -[2(CPU 586)0( Pentium instruction set)]367.6 bull3 -[2(CPU PENTIUM)0( Same as 586)]350.6 bull3 -[2(CPU 686)0( P6 instruction set)]333.6 bull3 -[2(CPU PPRO)0( Same as 686)]316.6 bull3 -[2(CPU P2)0( Same as 686)]299.6 bull3 -[2(CPU P3)0( Pentium III \(Katmai\) instruction sets)]282.6 bull3 -[2(CPU KATMAI)0( Same as P3)]265.6 bull3 -[2(CPU P4)0( Pentium 4 \(Willamette\) instruction set)]248.6 bull3 -[2(CPU WILLAMETTE)0( Same as P4)]231.6 bull3 -[2(CPU PRESCOTT)0( Prescott instruction set)]214.6 bull3 -[2(CPU X64)0( x86-64 \(x64/AMD64/Intel 64\) instruction set)]197.6 bull3 -[2(CPU IA64)0( IA64 CPU \(in x86 mode\) instruction set)]180.6 bull3 -[(All options are case insensitive. All instructions will be selected only if they apply to the selected CPU or)]163.6 norm1 -[(lower. By default, all instructions are available.)]152.6 norm2 -[{/section-6.9 xa}2(FLOAT)0(: Handling of )(floating-point constants)](6.9)131.2 head3 -[(By default, floating-point constants are rounded to nearest, and IEEE denormals are supported. The following)]114.2 norm1 -[(options can be set to alter this behaviour:)]103.2 norm2 -(71)pageodd -restore showpage -%%Page: 72 72 -%%BeginPageSetup -save -%%EndPageSetup -/72 pa -[2(FLOAT DAZ)0( Flush denormals to zero)]681 bull3 -[2(FLOAT NODAZ)0( Do not flush denormals to zero \(default\))]664 bull3 -[2(FLOAT NEAR)0( Round to nearest \(default\))]647 bull3 -[2(FLOAT UP)0( Round up \(toward +Infinity\))]630 bull3 -[2(FLOAT DOWN)0( Round down \(toward \226Infinity\))]613 bull3 -[2(FLOAT ZERO)0( Round toward zero)]596 bull3 -[2(FLOAT DEFAULT)0( Restore default settings)]579 bull3 -[(The standard macros )2(__FLOAT_DAZ__)0(, )2(__FLOAT_ROUND__)0(, and )2(__FLOAT__)0( contain the current state,)]562 norm1 -[(as long as the programmer has avoided the use of the brackeded primitive form, \()2([FLOAT])0(\).)]551 norm2 -[2(__FLOAT__)0( contains the full set of floating-point settings; this value can be saved away and invoked later to)]534 norm1 -[(restore the setting.)]523 norm2 -(72)pageeven -restore showpage -%%Page: 73 73 -%%BeginPageSetup -save -%%EndPageSetup -/73 pa -[{/chapter-7 xa}(Chapter 7: )(Output Formats)]642.8 chap3 -[(NASM is a portable assembler, designed to be able to compile on any ANSI C-supporting platform and)]607.8 norm1 -[(produce output to run on a variety of Intel x86 operating systems. For this reason, it has a large number of)]596.8 norm0 -[(available output formats, selected using the )2(-f)0( option on the NASM )(command line. Each of these formats,)]585.8 norm0 -[(along with its extensions to the base NASM syntax, is detailed in this chapter.)]574.8 norm2 -[(As stated in ){/section-2.1.1 xl}(section 2.1.1){el}(, NASM chooses a )(default name for your output file based on the input file name and)]557.8 norm1 -[(the chosen output format. This will be generated by removing the )(extension \()2(.asm)0(, )2(.s)0(, or whatever you like)]546.8 norm0 -[(to use\) from the input file name, and substituting an extension defined by the output format. The extensions)]535.8 norm0 -[(are given with each format below.)]524.8 norm2 -[{/section-7.1 xa}2(bin)0(: )(Flat-Form Binary)( Output)](7.1)503.4 head3 -[(The )2(bin)0( format does not produce object files: it generates nothing in the output file except the code you)]486.4 norm1 -[(wrote. Such `pure binary' files are used by )(MS-DOS: )2(.COM)0( executables and )2(.SYS)0( device drivers are pure)]475.4 norm0 -[(binary files. Pure binary output is also useful for )(operating system and )(boot loader development.)]464.4 norm2 -[(The )2(bin)0( format supports )(multiple section names. For details of how NASM handles sections in the )2(bin)]447.4 norm1 -[(format, see ){/section-7.1.3 xl}(section 7.1.3){el}(.)]436.4 norm2 -[(Using the )2(bin)0( format puts NASM by default into 16-bit mode \(see ){/section-6.1 xl}(section 6.1){el}(\). In order to use )2(bin)0( to write)]419.4 norm1 -[(32-bit or 64-bit code, such as an OS kernel, you need to explicitly issue the )2(BITS 32)0( or )2(BITS 64)0( directive.)]408.4 norm2 -[2(bin)0( has no default output file name extension: instead, it leaves your file name as it is once the original)]391.4 norm1 -[(extension has been removed. Thus, the default is for NASM to assemble )2(binprog.asm)0( into a binary file)]380.4 norm0 -[(called )2(binprog)0(.)]369.4 norm2 -[{/section-7.1.1 xa}2(ORG)0(: Binary File )(Program Origin)](7.1.1)350.2 subh3 -[(The )2(bin)0( format provides an additional directive to the list given in ){/chapter-6 xl}(chapter 6){el}(: )2(ORG)0(. The function of the )2(ORG)]333.2 norm1 -[(directive is to specify the origin address which NASM will assume the program begins at when it is loaded)]322.2 norm0 -[(into memory.)]311.2 norm2 -[(For example, the following code will generate the longword )2(0x00000104)0(:)]294.2 norm3 -[2( org 0x100 )]277.2 code1 -[2( dd label )]266.2 code0 -[2(label:)]255.2 code2 -[(Unlike the )2(ORG)0( directive provided by MASM-compatible assemblers, which allows you to jump around in)]238.2 norm1 -[(the object file and overwrite code you have already generated, NASM's )2(ORG)0( does exactly what the directive)]227.2 norm0 -[(says: )1(origin)0(. Its sole function is to specify one offset which is added to all internal address references within)]216.2 norm0 -[(the section; it does not permit any of the trickery that MASM's version does. See ){/section-12.1.3 xl}(section 12.1.3){el}( for further)]205.2 norm0 -[(comments.)]194.2 norm2 -[{/section-7.1.2 xa}2(bin)0( Extensions to the )2(SECTION)0( Directive)](7.1.2)175 subh3 -[(The )2(bin)0( output format extends the )2(SECTION)0( \(or )2(SEGMENT)0(\) directive to allow you to specify the alignment)]158 norm1 -[(requirements of segments. This is done by appending the )2(ALIGN)0( qualifier to the end of the section-definition)]147 norm0 -[(line. For example,)]136 norm2 -[2(section .data align=16)]119 code3 -[(switches to the section )2(.data)0( and also specifies that it must be aligned on a 16-byte boundary.)]102 norm3 -(73)pageodd -restore showpage -%%Page: 74 74 -%%BeginPageSetup -save -%%EndPageSetup -/74 pa -[(The parameter to )2(ALIGN)0( specifies how many low bits of the section start address must be forced to zero. The)]681 norm1 -[(alignment value given may be any power of two.)]670 norm2 -[{/section-7.1.3 xa}(Multisection)( Support for the )2(bin)0( Format)](7.1.3)650.8 subh3 -[(The )2(bin)0( format allows the use of multiple sections, of arbitrary names, besides the "known" )2(.text)0(, )2(.data)0(,)]633.8 norm1 -[(and )2(.bss)0( names.)]622.8 norm2 -[(Sections may be designated )2(progbits)0( or )2(nobits)0(. Default is )2(progbits)0( \(except )2(.bss)0(, which defaults)]605.8 bull1 -[(to )2(nobits)0(, of course\).)]594.8 bull2 -[(Sections can be aligned at a specified boundary following the previous section with )2(align=)0(, or at an)]577.8 bull1 -[(arbitrary byte-granular position with )2(start=)0(.)]566.8 bull2 -[(Sections can be given a virtual start address, which will be used for the calculation of all memory)]549.8 bull1 -[(references within that section with )2(vstart=)0(.)]538.8 bull2 -[(Sections can be ordered using )2(follows=<section>)0( or )2(vfollows=<section>)0( as an alternative to)]521.8 bull1 -[(specifying an explicit start address.)]510.8 bull2 -[(Arguments to )2(org)0(, )2(start)0(, )2(vstart)0(, and )2(align=)0( are critical expressions. See ){/section-3.8 xl}(section 3.8){el}(. E.g.)]493.8 bull1 -[2(align=\(1 << ALIGN_SHIFT\))0( \226 )2(ALIGN_SHIFT)0( must be defined before it is used here.)]482.8 bull2 -[(Any code which comes before an explicit )2(SECTION)0( directive is directed by default into the )2(.text)]465.8 bull1 -[(section.)]454.8 bull2 -[(If an )2(ORG)0( statement is not given, )2(ORG 0)0( is used by default.)]437.8 bull3 -[(The )2(.bss)0( section will be placed after the last )2(progbits)0( section, unless )2(start=)0(, )2(vstart=)0(,)]420.8 bull1 -[2(follows=)0(, or )2(vfollows=)0( has been specified.)]409.8 bull2 -[(All sections are aligned on dword boundaries, unless a different alignment has been specified.)]392.8 bull3 -[(Sections may not overlap.)]375.8 bull3 -[(NASM creates the )2(section.<secname>.start)0( for each section, which may be used in your code.)]358.8 bull3 -[{/section-7.1.4 xa}(Map Files)](7.1.4)339.6 subh3 -[(Map files can be generated in )2(-f bin)0( format by means of the )2([map])0( option. Map types of )2(all)0( \(default\),)]322.6 norm1 -[2(brief)0(, )2(sections)0(, )2(segments)0(, or )2(symbols)0( may be specified. Output may be directed to )2(stdout)]311.6 norm0 -[(\(default\), )2(stderr)0(, or a specified file. E.g. )2([map symbols myfile.map])0(. No "user form" exists, the)]300.6 norm0 -[(square brackets must be used.)]289.6 norm2 -[{/section-7.2 xa}2(ith)0(: )(Intel Hex Output)](7.2)268.2 head3 -[(The )2(ith)0( file format produces Intel hex-format files. Just as the )2(bin)0( format, this is a flat memory image)]251.2 norm1 -[(format with no support for relocation or linking. It is usually used with ROM programmers and similar)]240.2 norm0 -[(utilities.)]229.2 norm2 -[(All extensions supported by the )2(bin)0( file format is also supported by the )2(ith)0( file format.)]212.2 norm3 -[2(ith)0( provides a default output file-name extension of )2(.ith)0(.)]195.2 norm3 -[{/section-7.3 xa}2(srec)0(: )(Motorola S-Records Output)](7.3)173.8 head3 -[(The )2(srec)0( file format produces Motorola S-records files. Just as the )2(bin)0( format, this is a flat memory image)]156.8 norm1 -[(format with no support for relocation or linking. It is usually used with ROM programmers and similar)]145.8 norm0 -[(utilities.)]134.8 norm2 -[(All extensions supported by the )2(bin)0( file format is also supported by the )2(srec)0( file format.)]117.8 norm3 -[2(srec)0( provides a default output file-name extension of )2(.srec)0(.)]100.8 norm3 -(74)pageeven -restore showpage -%%Page: 75 75 -%%BeginPageSetup -save -%%EndPageSetup -/75 pa -[{/section-7.4 xa}2(obj)0(: )(Microsoft OMF)( Object Files)](7.4)676.6 head3 -[(The )2(obj)0( file format \(NASM calls it )2(obj)0( rather than )2(omf)0( for historical reasons\) is the one produced by)]659.6 norm1 -[(MASM and )(TASM, which is typically fed to 16-bit DOS linkers to produce )2(.EXE)0( files. It is also the format)]648.6 norm0 -[(used by )(OS/2.)]637.6 norm2 -[2(obj)0( provides a default output file-name extension of )2(.obj)0(.)]620.6 norm3 -[2(obj)0( is not exclusively a 16-bit format, though: NASM has full support for the 32-bit extensions to the)]603.6 norm1 -[(format. In particular, 32-bit )2(obj)0( format files are used by )(Borland's Win32 compilers, instead of using)]592.6 norm0 -[(Microsoft's newer )2(win32)0( object file format.)]581.6 norm2 -[(The )2(obj)0( format does not define any special segment names: you can call your segments anything you like.)]564.6 norm1 -[(Typical names for segments in )2(obj)0( format files are )2(CODE)0(, )2(DATA)0( and )2(BSS)0(.)]553.6 norm2 -[(If your source file contains code before specifying an explicit )2(SEGMENT)0( directive, then NASM will invent its)]536.6 norm1 -[(own segment called )2(__NASMDEFSEG)0( for you.)]525.6 norm2 -[(When you define a segment in an )2(obj)0( file, NASM defines the segment name as a symbol as well, so that you)]508.6 norm1 -[(can access the segment address of the segment. So, for example:)]497.6 norm2 -[2(segment data )]480.6 code1 -[2()]469.6 code0 -[2(dvar: dw 1234 )]458.6 code0 -[2()]447.6 code0 -[2(segment code )]436.6 code0 -[2()]425.6 code0 -[2(function: )]414.6 code0 -[2( mov ax,data ; get segment address of data )]403.6 code0 -[2( mov ds,ax ; and move it into DS )]392.6 code0 -[2( inc word [dvar] ; now this reference will work )]381.6 code0 -[2( ret)]370.6 code2 -[(The )2(obj)0( format also enables the use of the )2(SEG)0( and )2(WRT)0( operators, so that you can write code which does)]353.6 norm1 -[(things like)]342.6 norm2 -[2(extern foo )]325.6 code1 -[2()]314.6 code0 -[2( mov ax,seg foo ; get preferred segment of foo )]303.6 code0 -[2( mov ds,ax )]292.6 code0 -[2( mov ax,data ; a different segment )]281.6 code0 -[2( mov es,ax )]270.6 code0 -[2( mov ax,[ds:foo] ; this accesses `foo' )]259.6 code0 -[2( mov [es:foo wrt data],bx ; so does this)]248.6 code2 -[{/section-7.4.1 xa}2(obj)0( Extensions to the )2(SEGMENT)0( Directive)](7.4.1)229.4 subh3 -[(The )2(obj)0( output format extends the )2(SEGMENT)0( \(or )2(SECTION)0(\) directive to allow you to specify various)]212.4 norm1 -[(properties of the segment you are defining. This is done by appending extra qualifiers to the end of the)]201.4 norm0 -[(segment-definition line. For example,)]190.4 norm2 -[2(segment code private align=16)]173.4 code3 -[(defines the segment )2(code)0(, but also declares it to be a private segment, and requires that the portion of it)]156.4 norm1 -[(described in this code module must be aligned on a 16-byte boundary.)]145.4 norm2 -[(The available qualifiers are:)]128.4 norm3 -[2(PRIVATE)0(, )2(PUBLIC)0(, )2(COMMON)0( and )2(STACK)0( specify the combination characteristics of the segment.)]111.4 bull1 -[2(PRIVATE)0( segments do not get combined with any others by the linker; )2(PUBLIC)0( and )2(STACK)0( segments get)]100.4 bull0 -(75)pageodd -restore showpage -%%Page: 76 76 -%%BeginPageSetup -save -%%EndPageSetup -/76 pa -[(concatenated together at link time; and )2(COMMON)0( segments all get overlaid on top of each other rather than)]681 bull0 -[(stuck end-to-end.)]670 bull2 -[2(ALIGN)0( is used, as shown above, to specify how many low bits of the segment start address must be forced)]653 bull1 -[(to zero. The alignment value given may be any power of two from 1 to 4096; in reality, the only values)]642 bull0 -[(supported are 1, 2, 4, 16, 256 and 4096, so if 8 is specified it will be rounded up to 16, and 32, 64 and 128)]631 bull0 -[(will all be rounded up to 256, and so on. Note that alignment to 4096-byte boundaries is a )(PharLap)]620 bull0 -[(extension to the format and may not be supported by all linkers.)]609 bull2 -[2(CLASS)0( can be used to specify the segment class; this feature indicates to the linker that segments of the)]592 bull1 -[(same class should be placed near each other in the output file. The class name can be any word, e.g.)]581 bull0 -[2(CLASS=CODE)0(.)]570 bull2 -[2(OVERLAY)0(, like )2(CLASS)0(, is specified with an arbitrary word as an argument, and provides overlay)]553 bull1 -[(information to an overlay-capable linker.)]542 bull2 -[(Segments can be declared as )2(USE16)0( or )2(USE32)0(, which has the effect of recording the choice in the object)]525 bull1 -[(file and also ensuring that NASM's default assembly mode when assembling in that segment is 16-bit or)]514 bull0 -[(32-bit respectively.)]503 bull2 -[(When writing )(OS/2 object files, you should declare 32-bit segments as )2(FLAT)0(, which causes the default)]486 bull1 -[(segment base for anything in the segment to be the special group )2(FLAT)0(, and also defines the group if it is)]475 bull0 -[(not already defined.)]464 bull2 -[(The )2(obj)0( file format also allows segments to be declared as having a pre-defined absolute segment)]447 bull1 -[(address, although no linkers are currently known to make sensible use of this feature; nevertheless, NASM)]436 bull0 -[(allows you to declare a segment such as )2(SEGMENT SCREEN ABSOLUTE=0xB800)0( if you need to. The)]425 bull0 -[2(ABSOLUTE)0( and )2(ALIGN)0( keywords are mutually exclusive.)]414 bull2 -[(NASM's default segment attributes are )2(PUBLIC)0(, )2(ALIGN=1)0(, no class, no overlay, and )2(USE16)0(.)]397 norm3 -[{/section-7.4.2 xa}2(GROUP)0(: Defining Groups of Segments)](7.4.2)377.8 subh3 -[(The )2(obj)0( format also allows segments to be grouped, so that a single segment register can be used to refer to)]360.8 norm1 -[(all the segments in a group. NASM therefore supplies the )2(GROUP)0( directive, whereby you can code)]349.8 norm2 -[2(segment data )]332.8 code1 -[2()]321.8 code0 -[2( ; some data )]310.8 code0 -[2()]299.8 code0 -[2(segment bss )]288.8 code0 -[2()]277.8 code0 -[2( ; some uninitialized data )]266.8 code0 -[2()]255.8 code0 -[2(group dgroup data bss)]244.8 code2 -[(which will define a group called )2(dgroup)0( to contain the segments )2(data)0( and )2(bss)0(. Like )2(SEGMENT)0(, )2(GROUP)]227.8 norm1 -[(causes the group name to be defined as a symbol, so that you can refer to a variable )2(var)0( in the )2(data)0( segment)]216.8 norm0 -[(as )2(var wrt data)0( or as )2(var wrt dgroup)0(, depending on which segment value is currently in your)]205.8 norm0 -[(segment register.)]194.8 norm2 -[(If you just refer to )2(var)0(, however, and )2(var)0( is declared in a segment which is part of a group, then NASM will)]177.8 norm1 -[(default to giving you the offset of )2(var)0( from the beginning of the )1(group)0(, not the )1(segment)0(. Therefore)]166.8 norm0 -[2(SEG var)0(, also, will return the group base rather than the segment base.)]155.8 norm2 -[(NASM will allow a segment to be part of more than one group, but will generate a warning if you do this.)]138.8 norm1 -[(Variables declared in a segment which is part of more than one group will default to being relative to the first)]127.8 norm0 -[(group that was defined to contain the segment.)]116.8 norm2 -(76)pageeven -restore showpage -%%Page: 77 77 -%%BeginPageSetup -save -%%EndPageSetup -/77 pa -[(A group does not have to contain any segments; you can still make )2(WRT)0( references to a group which does not)]681 norm1 -[(contain the variable you are referring to. OS/2, for example, defines the special group )2(FLAT)0( with no segments)]670 norm0 -[(in it.)]659 norm2 -[{/section-7.4.3 xa}2(UPPERCASE)0(: Disabling Case Sensitivity in Output)](7.4.3)639.8 subh3 -[(Although NASM itself is )(case sensitive, some OMF linkers are not; therefore it can be useful for NASM to)]622.8 norm1 -[(output single-case object files. The )2(UPPERCASE)0( format-specific directive causes all segment, group and)]611.8 norm0 -[(symbol names that are written to the object file to be forced to upper case just before being written. Within a)]600.8 norm0 -[(source file, NASM is still case-sensitive; but the object file can be written entirely in upper case if desired.)]589.8 norm2 -[2(UPPERCASE)0( is used alone on a line; it requires no parameters.)]572.8 norm3 -[{/section-7.4.4 xa}2(IMPORT)0(: Importing DLL Symbols)](7.4.4)553.6 subh3 -[(The )2(IMPORT)0( format-specific directive defines a symbol to be imported from a DLL, for use if you are)]536.6 norm1 -[(writing a DLL's )(import library in NASM. You still need to declare the symbol as )2(EXTERN)0( as well as using)]525.6 norm0 -[(the )2(IMPORT)0( directive.)]514.6 norm2 -[(The )2(IMPORT)0( directive takes two required parameters, separated by white space, which are \(respectively\) the)]497.6 norm1 -[(name of the symbol you wish to import and the name of the library you wish to import it from. For example:)]486.6 norm2 -[2( import WSAStartup wsock32.dll)]469.6 code3 -[(A third optional parameter gives the name by which the symbol is known in the library you are importing it)]452.6 norm1 -[(from, in case this is not the same as the name you wish the symbol to be known by to your code once you)]441.6 norm0 -[(have imported it. For example:)]430.6 norm2 -[2( import asyncsel wsock32.dll WSAAsyncSelect)]413.6 code3 -[{/section-7.4.5 xa}2(EXPORT)0(: Exporting DLL Symbols)](7.4.5)394.4 subh3 -[(The )2(EXPORT)0( format-specific directive defines a global symbol to be exported as a DLL symbol, for use if)]377.4 norm1 -[(you are writing a DLL in NASM. You still need to declare the symbol as )2(GLOBAL)0( as well as using the)]366.4 norm0 -[2(EXPORT)0( directive.)]355.4 norm2 -[2(EXPORT)0( takes one required parameter, which is the name of the symbol you wish to export, as it was defined)]338.4 norm1 -[(in your source file. An optional second parameter \(separated by white space from the first\) gives the )1(external)]327.4 norm0 -[(name of the symbol: the name by which you wish the symbol to be known to programs using the DLL. If this)]316.4 norm0 -[(name is the same as the internal name, you may leave the second parameter off.)]305.4 norm2 -[(Further parameters can be given to define attributes of the exported symbol. These parameters, like the)]288.4 norm1 -[(second, are separated by white space. If further parameters are given, the external name must also be)]277.4 norm0 -[(specified, even if it is the same as the internal name. The available attributes are:)]266.4 norm2 -[2(resident)0( indicates that the exported name is to be kept resident by the system loader. This is an)]249.4 bull1 -[(optimisation for frequently used symbols imported by name.)]238.4 bull2 -[2(nodata)0( indicates that the exported symbol is a function which does not make use of any initialized data.)]221.4 bull3 -[2(parm=NNN)0(, where )2(NNN)0( is an integer, sets the number of parameter words for the case in which the symbol)]204.4 bull1 -[(is a call gate between 32-bit and 16-bit segments.)]193.4 bull2 -[(An attribute which is just a number indicates that the symbol should be exported with an identifying)]176.4 bull1 -[(number \(ordinal\), and gives the desired number.)]165.4 bull2 -[(For example:)]148.4 norm3 -[2( export myfunc )]131.4 code1 -[2( export myfunc TheRealMoreFormalLookingFunctionName )]120.4 code0 -(77)pageodd -restore showpage -%%Page: 78 78 -%%BeginPageSetup -save -%%EndPageSetup -/78 pa -[2( export myfunc myfunc 1234 ; export by ordinal )]681 code0 -[2( export myfunc myfunc resident parm=23 nodata)]670 code2 -[{/section-7.4.6 xa}2(..start)0(: Defining the )(Program Entry Point)](7.4.6)650.8 subh3 -[2(OMF)0( linkers require exactly one of the object files being linked to define the program entry point, where)]633.8 norm1 -[(execution will begin when the program is run. If the object file that defines the entry point is assembled using)]622.8 norm0 -[(NASM, you specify the entry point by declaring the special symbol )2(..start)0( at the point where you wish)]611.8 norm0 -[(execution to begin.)]600.8 norm2 -[{/section-7.4.7 xa}2(obj)0( Extensions to the )2(EXTERN)0( Directive)](7.4.7)581.6 subh3 -[(If you declare an external symbol with the directive)]564.6 norm3 -[2( extern foo)]547.6 code3 -[(then references such as )2(mov ax,foo)0( will give you the offset of )2(foo)0( from its preferred segment base \(as)]530.6 norm1 -[(specified in whichever module )2(foo)0( is actually defined in\). So to access the contents of )2(foo)0( you will usually)]519.6 norm0 -[(need to do something like)]508.6 norm2 -[2( mov ax,seg foo ; get preferred segment base )]491.6 code1 -[2( mov es,ax ; move it into ES )]480.6 code0 -[2( mov ax,[es:foo] ; and use offset `foo' from it)]469.6 code2 -[(This is a little unwieldy, particularly if you know that an external is going to be accessible from a given)]452.6 norm1 -[(segment or group, say )2(dgroup)0(. So if )2(DS)0( already contained )2(dgroup)0(, you could simply code)]441.6 norm2 -[2( mov ax,[foo wrt dgroup])]424.6 code3 -[(However, having to type this every time you want to access )2(foo)0( can be a pain; so NASM allows you to)]407.6 norm1 -[(declare )2(foo)0( in the alternative form)]396.6 norm2 -[2( extern foo:wrt dgroup)]379.6 code3 -[(This form causes NASM to pretend that the preferred segment base of )2(foo)0( is in fact )2(dgroup)0(; so the)]362.6 norm1 -[(expression )2(seg foo)0( will now return )2(dgroup)0(, and the expression )2(foo)0( is equivalent to )2(foo wrt dgroup)0(.)]351.6 norm2 -[(This )(default-)2(WRT)0( mechanism can be used to make externals appear to be relative to any group or segment in)]334.6 norm1 -[(your program. It can also be applied to common variables: see ){/section-7.4.8 xl}(section 7.4.8){el}(.)]323.6 norm2 -[{/section-7.4.8 xa}2(obj)0( Extensions to the )2(COMMON)0( Directive)](7.4.8)304.4 subh3 -[(The )2(obj)0( format allows common variables to be either near)( or far)(; NASM allows you to specify which your)]287.4 norm1 -[(variables should be by the use of the syntax)]276.4 norm2 -[2(common nearvar 2:near ; `nearvar' is a near common )]259.4 code1 -[2(common farvar 10:far ; and `farvar' is far)]248.4 code2 -[(Far common variables may be greater in size than 64Kb, and so the OMF specification says that they are)]231.4 norm1 -[(declared as a number of )1(elements)0( of a given size. So a 10-byte far common variable could be declared as ten)]220.4 norm0 -[(one-byte elements, five two-byte elements, two five-byte elements or one ten-byte element.)]209.4 norm2 -[(Some )2(OMF)0( linkers require the )(element size, as well as the variable size, to match when resolving common)]192.4 norm1 -[(variables declared in more than one module. Therefore NASM must allow you to specify the element size on)]181.4 norm0 -[(your far common variables. This is done by the following syntax:)]170.4 norm2 -[2(common c_5by2 10:far 5 ; two five-byte elements )]153.4 code1 -[2(common c_2by5 10:far 2 ; five two-byte elements)]142.4 code2 -[(If no element size is specified, the default is 1. Also, the )2(FAR)0( keyword is not required when an element size is)]125.4 norm1 -[(specified, since only far commons may have element sizes at all. So the above declarations could equivalently)]114.4 norm0 -[(be)]103.4 norm2 -(78)pageeven -restore showpage -%%Page: 79 79 -%%BeginPageSetup -save -%%EndPageSetup -/79 pa -[2(common c_5by2 10:5 ; two five-byte elements )]681 code1 -[2(common c_2by5 10:2 ; five two-byte elements)]670 code2 -[(In addition to these extensions, the )2(COMMON)0( directive in )2(obj)0( also supports default-)2(WRT)0( specification like)]653 norm1 -[2(EXTERN)0( does \(explained in ){/section-7.4.7 xl}(section 7.4.7){el}(\). So you can also declare things like)]642 norm2 -[2(common foo 10:wrt dgroup )]625 code1 -[2(common bar 16:far 2:wrt data )]614 code0 -[2(common baz 24:wrt data:6)]603 code2 -[{/section-7.5 xa}2(win32)0(: Microsoft Win32 Object Files)](7.5)581.6 head3 -[(The )2(win32)0( output format generates Microsoft Win32 object files, suitable for passing to Microsoft linkers)]564.6 norm1 -[(such as )(Visual C++. Note that Borland Win32 compilers do not use this format, but use )2(obj)0( instead \(see)]553.6 norm0 -[{/section-7.4 xl}(section 7.4){el}(\).)]542.6 norm2 -[2(win32)0( provides a default output file-name extension of )2(.obj)0(.)]525.6 norm3 -[(Note that although Microsoft say that Win32 object files follow the )2(COFF)0( \(Common Object File Format\))]508.6 norm1 -[(standard, the object files produced by Microsoft Win32 compilers are not compatible with COFF linkers such)]497.6 norm0 -[(as DJGPP's, and vice versa. This is due to a difference of opinion over the precise semantics of PC-relative)]486.6 norm0 -[(relocations. To produce COFF files suitable for DJGPP, use NASM's )2(coff)0( output format; conversely, the)]475.6 norm0 -[2(coff)0( format does not produce object files that Win32 linkers can generate correct output from.)]464.6 norm2 -[{/section-7.5.1 xa}2(win32)0( Extensions to the )2(SECTION)0( Directive)](7.5.1)445.4 subh3 -[(Like the )2(obj)0( format, )2(win32)0( allows you to specify additional information on the )2(SECTION)0( directive line, to)]428.4 norm1 -[(control the type and properties of sections you declare. Section types and properties are generated)]417.4 norm0 -[(automatically by NASM for the )(standard section names )2(.text)0(, )2(.data)0( and )2(.bss)0(, but may still be)]406.4 norm0 -[(overridden by these qualifiers.)]395.4 norm2 -[(The available qualifiers are:)]378.4 norm3 -[2(code)0(, or equivalently )2(text)0(, defines the section to be a code section. This marks the section as readable)]361.4 bull1 -[(and executable, but not writable, and also indicates to the linker that the type of the section is code.)]350.4 bull2 -[2(data)0( and )2(bss)0( define the section to be a data section, analogously to )2(code)0(. Data sections are marked as)]333.4 bull1 -[(readable and writable, but not executable. )2(data)0( declares an initialized data section, whereas )2(bss)0( declares)]322.4 bull0 -[(an uninitialized data section.)]311.4 bull2 -[2(rdata)0( declares an initialized data section that is readable but not writable. Microsoft compilers use this)]294.4 bull1 -[(section to place constants in it.)]283.4 bull2 -[2(info)0( defines the section to be an )(informational section, which is not included in the executable file by the)]266.4 bull1 -[(linker, but may \(for example\) pass information )1(to)0( the linker. For example, declaring an )2(info)0(\226type section)]255.4 bull0 -[(called )2(.drectve)0( causes the linker to interpret the contents of the section as command-line options.)]244.4 bull2 -[2(align=)0(, used with a trailing number as in )2(obj)0(, gives the )(alignment requirements of the section. The)]227.4 bull1 -[(maximum you may specify is 64: the Win32 object file format contains no means to request a greater)]216.4 bull0 -[(section alignment than this. If alignment is not explicitly specified, the defaults are 16-byte alignment for)]205.4 bull0 -[(code sections, 8-byte alignment for rdata sections and 4-byte alignment for data \(and BSS\) sections.)]194.4 bull0 -[(Informational sections get a default alignment of 1 byte \(no alignment\), though the value does not matter.)]183.4 bull2 -[(The defaults assumed by NASM if you do not specify the above qualifiers are:)]166.4 norm3 -[2(section .text code align=16 )]149.4 code1 -[2(section .data data align=4 )]138.4 code0 -[2(section .rdata rdata align=8 )]127.4 code0 -[2(section .bss bss align=4)]116.4 code2 -(79)pageodd -restore showpage -%%Page: 80 80 -%%BeginPageSetup -save -%%EndPageSetup -/80 pa -[(Any other section name is treated by default like )2(.text)0(.)]681 norm3 -[{/section-7.5.2 xa}2(win32)0(: Safe Structured Exception Handling)](7.5.2)661.8 subh3 -[(Among other improvements in Windows XP SP2 and Windows Server 2003 Microsoft has introduced)]644.8 norm1 -[(concept of "safe structured exception handling." General idea is to collect handlers' entry points in designated)]633.8 norm0 -[(read-only table and have alleged entry point verified against this table prior exception control is passed to the)]622.8 norm0 -[(handler. In order for an executable module to be equipped with such "safe exception handler table," all object)]611.8 norm0 -[(modules on linker command line has to comply with certain criteria. If one single module among them does)]600.8 norm0 -[(not, then the table in question is omitted and above mentioned run-time checks will not be performed for)]589.8 norm0 -[(application in question. Table omission is by default silent and therefore can be easily overlooked. One can)]578.8 norm0 -[(instruct linker to refuse to produce binary without such table by passing )2(/safeseh)0( command line option.)]567.8 norm2 -[(Without regard to this run-time check merits it's natural to expect NASM to be capable of generating)]550.8 norm1 -[(modules suitable for )2(/safeseh)0( linking. From developer's viewpoint the problem is two-fold:)]539.8 norm2 -[(how to adapt modules not deploying exception handlers of their own;)]522.8 bull3 -[(how to adapt/develop modules utilizing custom exception handling;)]505.8 bull3 -[(Former can be easily achieved with any NASM version by adding following line to source code:)]488.8 norm3 -[2($@feat.00 equ 1)]471.8 code3 -[(As of version 2.03 NASM adds this absolute symbol automatically. If it's not already present to be precise.)]454.8 norm1 -[(I.e. if for whatever reason developer would choose to assign another value in source file, it would still be)]443.8 norm0 -[(perfectly possible.)]432.8 norm2 -[(Registering custom exception handler on the other hand requires certain "magic." As of version 2.03)]415.8 norm1 -[(additional directive is implemented, )2(safeseh)0(, which instructs the assembler to produce appropriately)]404.8 norm0 -[(formatted input data for above mentioned "safe exception handler table." Its typical use would be:)]393.8 norm2 -[2(section .text )]376.8 code1 -[2(extern _MessageBoxA@16 )]365.8 code0 -[2(%if __NASM_VERSION_ID__ >= 0x02030000 )]354.8 code0 -[2(safeseh handler ; register handler as "safe handler" )]343.8 code0 -[2(%endif )]332.8 code0 -[2(handler: )]321.8 code0 -[2( push DWORD 1 ; MB_OKCANCEL )]310.8 code0 -[2( push DWORD caption )]299.8 code0 -[2( push DWORD text )]288.8 code0 -[2( push DWORD 0 )]277.8 code0 -[2( call _MessageBoxA@16 )]266.8 code0 -[2( sub eax,1 ; incidentally suits as return value )]255.8 code0 -[2( ; for exception handler )]244.8 code0 -[2( ret )]233.8 code0 -[2(global _main )]222.8 code0 -[2(_main: )]211.8 code0 -[2( push DWORD handler )]200.8 code0 -[2( push DWORD [fs:0] )]189.8 code0 -[2( mov DWORD [fs:0],esp ; engage exception handler )]178.8 code0 -[2( xor eax,eax )]167.8 code0 -[2( mov eax,DWORD[eax] ; cause exception )]156.8 code0 -[2( pop DWORD [fs:0] ; disengage exception handler )]145.8 code0 -[2( add esp,4 )]134.8 code0 -[2( ret )]123.8 code0 -[2(text: db 'OK to rethrow, CANCEL to generate core dump',0 )]112.8 code0 -[2(caption:db 'SEGV',0 )]101.8 code0 -(80)pageeven -restore showpage -%%Page: 81 81 -%%BeginPageSetup -save -%%EndPageSetup -/81 pa -[2()]681 code0 -[2(section .drectve info )]670 code0 -[2( db '/defaultlib:user32.lib /defaultlib:msvcrt.lib ')]659 code2 -[(As you might imagine, it's perfectly possible to produce .exe binary with "safe exception handler table" and)]642 norm1 -[(yet engage unregistered exception handler. Indeed, handler is engaged by simply manipulating )2([fs:0])]631 norm0 -[(location at run-time, something linker has no power over, run-time that is. It should be explicitly mentioned)]620 norm0 -[(that such failure to register handler's entry point with )2(safeseh)0( directive has undesired side effect at)]609 norm0 -[(run-time. If exception is raised and unregistered handler is to be executed, the application is abruptly)]598 norm0 -[(terminated without any notification whatsoever. One can argue that system could at least have logged some)]587 norm0 -[(kind "non-safe exception handler in x.exe at address n" message in event log, but no, literally no notification)]576 norm0 -[(is provided and user is left with no clue on what caused application failure.)]565 norm2 -[(Finally, all mentions of linker in this paragraph refer to Microsoft linker version 7.x and later. Presence of)]548 norm1 -[2(@feat.00)0( symbol and input data for "safe exception handler table" causes no backward incompatibilities)]537 norm0 -[(and "safeseh" modules generated by NASM 2.03 and later can still be linked by earlier versions or)]526 norm0 -[(non-Microsoft linkers.)]515 norm2 -[{/section-7.6 xa}2(win64)0(: Microsoft Win64 Object Files)](7.6)493.6 head3 -[(The )2(win64)0( output format generates Microsoft Win64 object files, which is nearly 100% identical to the)]476.6 norm1 -[2(win32)0( object format \(){/section-7.5 xl}(section 7.5){el}(\) with the exception that it is meant to target 64-bit code and the x86-64)]465.6 norm0 -[(platform altogether. This object file is used exactly the same as the )2(win32)0( object format \(){/section-7.5 xl}(section 7.5){el}(\), in)]454.6 norm0 -[(NASM, with regard to this exception.)]443.6 norm2 -[{/section-7.6.1 xa}2(win64)0(: Writing Position-Independent Code)](7.6.1)424.4 subh3 -[(While )2(REL)0( takes good care of RIP-relative addressing, there is one aspect that is easy to overlook for a)]407.4 norm1 -[(Win64 programmer: indirect references. Consider a switch dispatch table:)]396.4 norm2 -[2( jmp QWORD[dsptch+rax*8] )]379.4 code1 -[2( ... )]368.4 code0 -[2(dsptch: dq case0 )]357.4 code0 -[2( dq case1 )]346.4 code0 -[2( ...)]335.4 code2 -[(Even novice Win64 assembler programmer will soon realize that the code is not 64-bit savvy. Most notably)]318.4 norm1 -[(linker will refuse to link it with)]307.4 norm0 -[(")2('ADDR32' relocation to '.text' invalid without /LARGEADDRESSAWARE:NO)0(". So)]296.4 norm0 -[([s]he will have to split jmp instruction as following:)]285.4 norm2 -[2( lea rbx,[rel dsptch] )]268.4 code1 -[2( jmp QWORD[rbx+rax*8])]257.4 code2 -[(What happens behind the scene is that effective address in )2(lea)0( is encoded relative to instruction pointer, or in)]240.4 norm1 -[(perfectly position-independent manner. But this is only part of the problem! Trouble is that in .dll context)]229.4 norm0 -[2(caseN)0( relocations will make their way to the final module and might have to be adjusted at .dll load time. To)]218.4 norm0 -[(be specific when it can't be loaded at preferred address. And when this occurs, pages with such relocations)]207.4 norm0 -[(will be rendered private to current process, which kind of undermines the idea of sharing .dll. But no worry,)]196.4 norm0 -[(it's trivial to fix:)]185.4 norm2 -[2( lea rbx,[rel dsptch] )]168.4 code1 -[2( add rbx,QWORD[rbx+rax*8] )]157.4 code0 -[2( jmp rbx )]146.4 code0 -[2( ... )]135.4 code0 -[2(dsptch: dq case0-dsptch )]124.4 code0 -[2( dq case1-dsptch )]113.4 code0 -[2( ...)]102.4 code2 -(81)pageodd -restore showpage -%%Page: 82 82 -%%BeginPageSetup -save -%%EndPageSetup -/82 pa -[(NASM version 2.03 and later provides another alternative, )2(wrt ..imagebase)0( operator, which returns)]681 norm1 -[(offset from base address of the current image, be it .exe or .dll module, therefore the name. For those)]670 norm0 -[(acquainted with PE-COFF format base address denotes start of )2(IMAGE_DOS_HEADER)0( structure. Here is)]659 norm0 -[(how to implement switch with these image-relative references:)]648 norm2 -[2( lea rbx,[rel dsptch] )]631 code1 -[2( mov eax,DWORD[rbx+rax*4] )]620 code0 -[2( sub rbx,dsptch wrt ..imagebase )]609 code0 -[2( add rbx,rax )]598 code0 -[2( jmp rbx )]587 code0 -[2( ... )]576 code0 -[2(dsptch: dd case0 wrt ..imagebase )]565 code0 -[2( dd case1 wrt ..imagebase)]554 code2 -[(One can argue that the operator is redundant. Indeed, snippet before last works just fine with any NASM)]537 norm1 -[(version and is not even Windows specific... The real reason for implementing )2(wrt ..imagebase)0( will)]526 norm0 -[(become apparent in next paragraph.)]515 norm2 -[(It should be noted that )2(wrt ..imagebase)0( is defined as 32-bit operand only:)]498 norm3 -[2( dd label wrt ..imagebase ; ok )]481 code1 -[2( dq label wrt ..imagebase ; bad )]470 code0 -[2( mov eax,label wrt ..imagebase ; ok )]459 code0 -[2( mov rax,label wrt ..imagebase ; bad)]448 code2 -[{/section-7.6.2 xa}2(win64)0(: Structured Exception Handling)](7.6.2)428.8 subh3 -[(Structured exception handing in Win64 is completely different matter from Win32. Upon exception program)]411.8 norm1 -[(counter value is noted, and linker-generated table comprising start and end addresses of all the functions [in)]400.8 norm0 -[(given executable module] is traversed and compared to the saved program counter. Thus so called)]389.8 norm0 -[2(UNWIND_INFO)0( structure is identified. If it's not found, then offending subroutine is assumed to be "leaf" and)]378.8 norm0 -[(just mentioned lookup procedure is attempted for its caller. In Win64 leaf function is such function that does)]367.8 norm0 -[(not call any other function )1(nor)0( modifies any Win64 non-volatile registers, including stack pointer. The latter)]356.8 norm0 -[(ensures that it's possible to identify leaf function's caller by simply pulling the value from the top of the stack.)]345.8 norm2 -[(While majority of subroutines written in assembler are not calling any other function, requirement for)]328.8 norm1 -[(non-volatile registers' immutability leaves developer with not more than 7 registers and no stack frame,)]317.8 norm0 -[(which is not necessarily what [s]he counted with. Customarily one would meet the requirement by saving)]306.8 norm0 -[(non-volatile registers on stack and restoring them upon return, so what can go wrong? If [and only if] an)]295.8 norm0 -[(exception is raised at run-time and no )2(UNWIND_INFO)0( structure is associated with such "leaf" function, the)]284.8 norm0 -[(stack unwind procedure will expect to find caller's return address on the top of stack immediately followed by)]273.8 norm0 -[(its frame. Given that developer pushed caller's non-volatile registers on stack, would the value on top point at)]262.8 norm0 -[(some code segment or even addressable space? Well, developer can attempt copying caller's return address to)]251.8 norm0 -[(the top of stack and this would actually work in some very specific circumstances. But unless developer can)]240.8 norm0 -[(guarantee that these circumstances are always met, it's more appropriate to assume worst case scenario, i.e.)]229.8 norm0 -[(stack unwind procedure going berserk. Relevant question is what happens then? Application is abruptly)]218.8 norm0 -[(terminated without any notification whatsoever. Just like in Win32 case, one can argue that system could at)]207.8 norm0 -[(least have logged "unwind procedure went berserk in x.exe at address n" in event log, but no, no trace of)]196.8 norm0 -[(failure is left.)]185.8 norm2 -[(Now, when we understand significance of the )2(UNWIND_INFO)0( structure, let's discuss what's in it and/or how)]168.8 norm1 -[(it's processed. First of all it is checked for presence of reference to custom language-specific exception)]157.8 norm0 -[(handler. If there is one, then it's invoked. Depending on the return value, execution flow is resumed)]146.8 norm0 -[(\(exception is said to be "handled"\), )1(or)0( rest of )2(UNWIND_INFO)0( structure is processed as following. Beside)]135.8 norm0 -[(optional reference to custom handler, it carries information about current callee's stack frame and where)]124.8 norm0 -[(non-volatile registers are saved. Information is detailed enough to be able to reconstruct contents of caller's)]113.8 norm0 -[(non-volatile registers upon call to current callee. And so caller's context is reconstructed, and then unwind)]102.8 norm0 -(82)pageeven -restore showpage -%%Page: 83 83 -%%BeginPageSetup -save -%%EndPageSetup -/83 pa -[(procedure is repeated, i.e. another )2(UNWIND_INFO)0( structure is associated, this time, with caller's instruction)]681 norm0 -[(pointer, which is then checked for presence of reference to language-specific handler, etc. The procedure is)]670 norm0 -[(recursively repeated till exception is handled. As last resort system "handles" it by generating memory core)]659 norm0 -[(dump and terminating the application.)]648 norm2 -[(As for the moment of this writing NASM unfortunately does not facilitate generation of above mentioned)]631 norm1 -[(detailed information about stack frame layout. But as of version 2.03 it implements building blocks for)]620 norm0 -[(generating structures involved in stack unwinding. As simplest example, here is how to deploy custom)]609 norm0 -[(exception handler for leaf function:)]598 norm2 -[2(default rel )]581 code1 -[2(section .text )]570 code0 -[2(extern MessageBoxA )]559 code0 -[2(handler: )]548 code0 -[2( sub rsp,40 )]537 code0 -[2( mov rcx,0 )]526 code0 -[2( lea rdx,[text] )]515 code0 -[2( lea r8,[caption] )]504 code0 -[2( mov r9,1 ; MB_OKCANCEL )]493 code0 -[2( call MessageBoxA )]482 code0 -[2( sub eax,1 ; incidentally suits as return value )]471 code0 -[2( ; for exception handler )]460 code0 -[2( add rsp,40 )]449 code0 -[2( ret )]438 code0 -[2(global main )]427 code0 -[2(main: )]416 code0 -[2( xor rax,rax )]405 code0 -[2( mov rax,QWORD[rax] ; cause exception )]394 code0 -[2( ret )]383 code0 -[2(main_end: )]372 code0 -[2(text: db 'OK to rethrow, CANCEL to generate core dump',0 )]361 code0 -[2(caption:db 'SEGV',0 )]350 code0 -[2()]339 code0 -[2(section .pdata rdata align=4 )]328 code0 -[2( dd main wrt ..imagebase )]317 code0 -[2( dd main_end wrt ..imagebase )]306 code0 -[2( dd xmain wrt ..imagebase )]295 code0 -[2(section .xdata rdata align=8 )]284 code0 -[2(xmain: db 9,0,0,0 )]273 code0 -[2( dd handler wrt ..imagebase )]262 code0 -[2(section .drectve info )]251 code0 -[2( db '/defaultlib:user32.lib /defaultlib:msvcrt.lib ')]240 code2 -[(What you see in )2(.pdata)0( section is element of the "table comprising start and end addresses of function")]223 norm1 -[(along with reference to associated )2(UNWIND_INFO)0( structure. And what you see in )2(.xdata)0( section is)]212 norm0 -[2(UNWIND_INFO)0( structure describing function with no frame, but with designated exception handler.)]201 norm0 -[(References are )1(required)0( to be image-relative \(which is the real reason for implementing)]190 norm0 -[2(wrt ..imagebase)0( operator\). It should be noted that )2(rdata align=n)0(, as well as)]179 norm0 -[2(wrt ..imagebase)0(, are optional in these two segments' contexts, i.e. can be omitted. Latter means that )1(all)]168 norm0 -[(32-bit references, not only above listed required ones, placed into these two segments turn out)]157 norm0 -[(image-relative. Why is it important to understand? Developer is allowed to append handler-specific data to)]146 norm0 -[2(UNWIND_INFO)0( structure, and if [s]he adds a 32-bit reference, then [s]he will have to remember to adjust its)]135 norm0 -[(value to obtain the real pointer.)]124 norm2 -(83)pageodd -restore showpage -%%Page: 84 84 -%%BeginPageSetup -save -%%EndPageSetup -/84 pa -[(As already mentioned, in Win64 terms leaf function is one that does not call any other function )1(nor)0( modifies)]681 norm1 -[(any non-volatile register, including stack pointer. But it's not uncommon that assembler programmer plans to)]670 norm0 -[(utilize every single register and sometimes even have variable stack frame. Is there anything one can do with)]659 norm0 -[(bare building blocks? I.e. besides manually composing fully-fledged )2(UNWIND_INFO)0( structure, which would)]648 norm0 -[(surely be considered error-prone? Yes, there is. Recall that exception handler is called first, before stack)]637 norm0 -[(layout is analyzed. As it turned out, it's perfectly possible to manipulate current callee's context in custom)]626 norm0 -[(handler in manner that permits further stack unwinding. General idea is that handler would not actually)]615 norm0 -[("handle" the exception, but instead restore callee's context, as it was at its entry point and thus mimic leaf)]604 norm0 -[(function. In other words, handler would simply undertake part of unwinding procedure. Consider following)]593 norm0 -[(example:)]582 norm2 -[2(function: )]565 code1 -[2( mov rax,rsp ; copy rsp to volatile register )]554 code0 -[2( push r15 ; save non-volatile registers )]543 code0 -[2( push rbx )]532 code0 -[2( push rbp )]521 code0 -[2( mov r11,rsp ; prepare variable stack frame )]510 code0 -[2( sub r11,rcx )]499 code0 -[2( and r11,-64 )]488 code0 -[2( mov QWORD[r11],rax ; check for exceptions )]477 code0 -[2( mov rsp,r11 ; allocate stack frame )]466 code0 -[2( mov QWORD[rsp],rax ; save original rsp value )]455 code0 -[2(magic_point: )]444 code0 -[2( ... )]433 code0 -[2( mov r11,QWORD[rsp] ; pull original rsp value )]422 code0 -[2( mov rbp,QWORD[r11-24] )]411 code0 -[2( mov rbx,QWORD[r11-16] )]400 code0 -[2( mov r15,QWORD[r11-8] )]389 code0 -[2( mov rsp,r11 ; destroy frame )]378 code0 -[2( ret)]367 code2 -[(The keyword is that up to )2(magic_point)0( original )2(rsp)0( value remains in chosen volatile register and no)]350 norm1 -[(non-volatile register, except for )2(rsp)0(, is modified. While past )2(magic_point)0( )2(rsp)0( remains constant till the)]339 norm0 -[(very end of the )2(function)0(. In this case custom language-specific exception handler would look like this:)]328 norm2 -[2(EXCEPTION_DISPOSITION handler \(EXCEPTION_RECORD *rec,ULONG64 frame, )]311 code1 -[2( CONTEXT *context,DISPATCHER_CONTEXT *disp\) )]300 code0 -[2({ ULONG64 *rsp; )]289 code0 -[2( if \(context->Rip<\(ULONG64\)magic_point\) )]278 code0 -[2( rsp = \(ULONG64 *\)context->Rax; )]267 code0 -[2( else )]256 code0 -[2( { rsp = \(\(ULONG64 **\)context->Rsp\)[0]; )]245 code0 -[2( context->Rbp = rsp[-3]; )]234 code0 -[2( context->Rbx = rsp[-2]; )]223 code0 -[2( context->R15 = rsp[-1]; )]212 code0 -[2( } )]201 code0 -[2( context->Rsp = \(ULONG64\)rsp; )]190 code0 -[2()]179 code0 -[2( memcpy \(disp->ContextRecord,context,sizeof\(CONTEXT\)\); )]168 code0 -[2( RtlVirtualUnwind\(UNW_FLAG_NHANDLER,disp->ImageBase, )]157 code0 -[2( dips->ControlPc,disp->FunctionEntry,disp->ContextRecord, )]146 code0 -[2( &disp->HandlerData,&disp->EstablisherFrame,NULL\); )]135 code0 -[2( return ExceptionContinueSearch; )]124 code0 -[2(})]113 code2 -(84)pageeven -restore showpage -%%Page: 85 85 -%%BeginPageSetup -save -%%EndPageSetup -/85 pa -[(As custom handler mimics leaf function, corresponding )2(UNWIND_INFO)0( structure does not have to contain)]681 norm1 -[(any information about stack frame and its layout.)]670 norm2 -[{/section-7.7 xa}2(coff)0(: )(Common Object File Format)](7.7)648.6 head3 -[(The )2(coff)0( output type produces )2(COFF)0( object files suitable for linking with the )(DJGPP linker.)]631.6 norm3 -[2(coff)0( provides a default output file-name extension of )2(.o)0(.)]614.6 norm3 -[(The )2(coff)0( format supports the same extensions to the )2(SECTION)0( directive as )2(win32)0( does, except that the)]597.6 norm1 -[2(align)0( qualifier and the )2(info)0( section type are not supported.)]586.6 norm2 -[{/section-7.8 xa}2(macho32)0( and )2(macho64)0(: )(Mach Object File Format)](7.8)565.2 head3 -[(The )2(macho32)0( and )2(macho64)0( output formts produces )2(Mach-O)0( object files suitable for linking with the)]548.2 norm1 -[(MacOS X linker. )2(macho)0( is a synonym for )2(macho32)0(.)]537.2 norm2 -[2(macho)0( provides a default output file-name extension of )2(.o)0(.)]520.2 norm3 -[{/section-7.9 xa}2(elf32)0( and )2(elf64)0(: )(Executable and Linkable Format Object Files)](7.9)498.8 head3 -[(The )2(elf32)0( and )2(elf64)0( output formats generate )2(ELF32 and ELF64)0( \(Executable and Linkable Format\))]481.8 norm1 -[(object files, as used by Linux as well as )(Unix System V, including )(Solaris x86, )(UnixWare and )(SCO Unix.)]470.8 norm0 -[2(elf)0( provides a default output file-name extension of )2(.o)0(. )2(elf)0( is a synonym for )2(elf32)0(.)]459.8 norm2 -[{/section-7.9.1 xa}(ELF specific directive )2(osabi)](7.9.1)440.6 subh3 -[(The ELF header specifies the application binary interface for the target operating system \(OSABI\). This field)]423.6 norm1 -[(can be set by using the )2(osabi)0( directive with the numeric value \(0-255\) of the target system. If this directive)]412.6 norm0 -[(is not used, the default value will be "UNIX System V ABI" \(0\) which will work on most systems which)]401.6 norm0 -[(support ELF.)]390.6 norm2 -[{/section-7.9.2 xa}2(elf)0( Extensions to the )2(SECTION)0( Directive)](7.9.2)371.4 subh3 -[(Like the )2(obj)0( format, )2(elf)0( allows you to specify additional information on the )2(SECTION)0( directive line, to)]354.4 norm1 -[(control the type and properties of sections you declare. Section types and properties are generated)]343.4 norm0 -[(automatically by NASM for the )(standard section names, but may still be overridden by these qualifiers.)]332.4 norm2 -[(The available qualifiers are:)]315.4 norm3 -[2(alloc)0( defines the section to be one which is loaded into memory when the program is run. )2(noalloc)]298.4 bull1 -[(defines it to be one which is not, such as an informational or comment section.)]287.4 bull2 -[2(exec)0( defines the section to be one which should have execute permission when the program is run.)]270.4 bull1 -[2(noexec)0( defines it as one which should not.)]259.4 bull2 -[2(write)0( defines the section to be one which should be writable when the program is run. )2(nowrite)0( defines)]242.4 bull1 -[(it as one which should not.)]231.4 bull2 -[2(progbits)0( defines the section to be one with explicit contents stored in the object file: an ordinary code)]214.4 bull1 -[(or data section, for example, )2(nobits)0( defines the section to be one with no explicit contents given, such as)]203.4 bull0 -[(a BSS section.)]192.4 bull2 -[2(align=)0(, used with a trailing number as in )2(obj)0(, gives the )(alignment requirements of the section.)]175.4 bull3 -[2(tls)0( defines the section to be one which contains thread local variables.)]158.4 bull3 -[(The defaults assumed by NASM if you do not specify the above qualifiers are:)]141.4 norm3 -[( )( )( )( )( )( )( )( )( )]124.4 norm3 -(85)pageodd -restore showpage -%%Page: 86 86 -%%BeginPageSetup -save -%%EndPageSetup -/86 pa -[2(section .text progbits alloc exec nowrite align=16 )]681 code1 -[2(section .rodata progbits alloc noexec nowrite align=4 )]670 code0 -[2(section .lrodata progbits alloc noexec nowrite align=4 )]659 code0 -[2(section .data progbits alloc noexec write align=4 )]648 code0 -[2(section .ldata progbits alloc noexec write align=4 )]637 code0 -[2(section .bss nobits alloc noexec write align=4 )]626 code0 -[2(section .lbss nobits alloc noexec write align=4 )]615 code0 -[2(section .tdata progbits alloc noexec write align=4 tls )]604 code0 -[2(section .tbss nobits alloc noexec write align=4 tls )]593 code0 -[2(section .comment progbits noalloc noexec nowrite align=1 )]582 code0 -[2(section other progbits alloc noexec nowrite align=1)]571 code2 -[(\(Any section name other than those in the above table is treated by default like )2(other)0( in the above table.)]554 norm1 -[(Please note that section names are case sensitive.\))]543 norm2 -[{/section-7.9.3 xa}(Position-Independent Code)(: )2(elf)0( Special Symbols and )2(WRT)](7.9.3)523.8 subh3 -[(The )2(ELF)0( specification contains enough features to allow position-independent code \(PIC\) to be written,)]506.8 norm1 -[(which makes )(ELF shared libraries very flexible. However, it also means NASM has to be able to generate a)]495.8 norm0 -[(variety of ELF specific relocation types in ELF object files, if it is to be an assembler which can write PIC.)]484.8 norm2 -[(Since )2(ELF)0( does not support segment-base references, the )2(WRT)0( operator is not used for its normal purpose;)]467.8 norm1 -[(therefore NASM's )2(elf)0( output format makes use of )2(WRT)0( for a different purpose, namely the PIC-specific)]456.8 norm0 -[(relocation types.)]445.8 norm2 -[2(elf)0( defines five special symbols which you can use as the right-hand side of the )2(WRT)0( operator to obtain PIC)]428.8 norm1 -[(relocation types. They are )2(..gotpc)0(, )2(..gotoff)0(, )2(..got)0(, )2(..plt)0( and )2(..sym)0(. Their functions are)]417.8 norm0 -[(summarized here:)]406.8 norm2 -[(Referring to the symbol marking the global offset table base using )2(wrt ..gotpc)0( will end up giving the)]389.8 bull1 -[(distance from the beginning of the current section to the global offset table.)]378.8 bull0 -[(\()2(_GLOBAL_OFFSET_TABLE_)0( is the standard symbol name used to refer to the )(GOT.\) So you would then)]367.8 bull0 -[(need to add )2($$)0( to the result to get the real address of the GOT.)]356.8 bull2 -[(Referring to a location in one of your own sections using )2(wrt ..gotoff)0( will give the distance from the)]339.8 bull1 -[(beginning of the GOT to the specified location, so that adding on the address of the GOT would give the)]328.8 bull0 -[(real address of the location you wanted.)]317.8 bull2 -[(Referring to an external or global symbol using )2(wrt ..got)0( causes the linker to build an entry )1(in)0( the)]300.8 bull1 -[(GOT containing the address of the symbol, and the reference gives the distance from the beginning of the)]289.8 bull0 -[(GOT to the entry; so you can add on the address of the GOT, load from the resulting address, and end up)]278.8 bull0 -[(with the address of the symbol.)]267.8 bull2 -[(Referring to a procedure name using )2(wrt ..plt)0( causes the linker to build a )(procedure linkage table)]250.8 bull1 -[(entry for the symbol, and the reference gives the address of the )(PLT entry. You can only use this in)]239.8 bull0 -[(contexts which would generate a PC-relative relocation normally \(i.e. as the destination for )2(CALL)0( or )2(JMP)0(\),)]228.8 bull0 -[(since ELF contains no relocation type to refer to PLT entries absolutely.)]217.8 bull2 -[(Referring to a symbol name using )2(wrt ..sym)0( causes NASM to write an ordinary relocation, but instead)]200.8 bull1 -[(of making the relocation relative to the start of the section and then adding on the offset to the symbol, it)]189.8 bull0 -[(will write a relocation record aimed directly at the symbol in question. The distinction is a necessary one)]178.8 bull0 -[(due to a peculiarity of the dynamic linker.)]167.8 bull2 -[(A fuller explanation of how to use these relocation types to write shared libraries entirely in NASM is given in)]150.8 norm1 -[{/section-9.2 xl}(section 9.2){el}(.)]139.8 norm2 -(86)pageeven -restore showpage -%%Page: 87 87 -%%BeginPageSetup -save -%%EndPageSetup -/87 pa -[{/section-7.9.4 xa}(Thread Local Storage)(: )2(elf)0( Special Symbols and )2(WRT)](7.9.4)678.8 subh3 -[(In ELF32 mode, referring to an external or global symbol using )2(wrt ..tlsie)0( )( causes the linker to build)]661.8 bull1 -[(an entry )1(in)0( the GOT containing the offset of the symbol within the TLS block, so you can access the value)]650.8 bull0 -[(of the symbol with code such as:)]639.8 bull2 -[2( mov eax,[tid wrt ..tlsie] )]622.8 code1 -[2( mov [gs:eax],ebx)]611.8 code2 -[(In ELF64 mode, referring to an external or global symbol using )2(wrt ..gottpoff)0( )( causes the linker to)]594.8 bull1 -[(build an entry )1(in)0( the GOT containing the offset of the symbol within the TLS block, so you can access the)]583.8 bull0 -[(value of the symbol with code such as:)]572.8 bull2 -[2( mov rax,[rel tid wrt ..gottpoff] )]555.8 code1 -[2( mov rcx,[fs:rax])]544.8 code2 -[{/section-7.9.5 xa}2(elf)0( Extensions to the )2(GLOBAL)0( Directive)](7.9.5)525.6 subh3 -[2(ELF)0( object files can contain more information about a global symbol than just its address: they can contain)]508.6 norm1 -[(the )(size of the symbol and its )(type as well. These are not merely debugger conveniences, but are actually)]497.6 norm0 -[(necessary when the program being written is a )(shared library. NASM therefore supports some extensions to)]486.6 norm0 -[(the )2(GLOBAL)0( directive, allowing you to specify these features.)]475.6 norm2 -[(You can specify whether a global variable is a function or a data object by suffixing the name with a colon)]458.6 norm1 -[(and the word )2(function)0( or )2(data)0(. \()2(object)0( is a synonym for )2(data)0(.\) For example:)]447.6 norm2 -[2(global hashlookup:function, hashtable:data)]430.6 code3 -[(exports the global symbol )2(hashlookup)0( as a function and )2(hashtable)0( as a data object.)]413.6 norm3 -[(Optionally, you can control the ELF visibility of the symbol. Just add one of the visibility keywords:)]396.6 norm1 -[2(default)0(, )2(internal)0(, )2(hidden)0(, or )2(protected)0(. The default is )2(default)0( of course. For example, to make)]385.6 norm0 -[2(hashlookup)0( hidden:)]374.6 norm2 -[2(global hashlookup:function hidden)]357.6 code3 -[(You can also specify the size of the data associated with the symbol, as a numeric expression \(which may)]340.6 norm1 -[(involve labels, and even forward references\) after the type specifier. Like this:)]329.6 norm2 -[2(global hashtable:data \(hashtable.end - hashtable\) )]312.6 code1 -[2()]301.6 code0 -[2(hashtable: )]290.6 code0 -[2( db this,that,theother ; some data here )]279.6 code0 -[2(.end:)]268.6 code2 -[(This makes NASM automatically calculate the length of the table and place that information into the )2(ELF)]251.6 norm1 -[(symbol table.)]240.6 norm2 -[(Declaring the type and size of global symbols is necessary when writing shared library code. For more)]223.6 norm1 -[(information, see ){/section-9.2.4 xl}(section 9.2.4){el}(.)]212.6 norm2 -[{/section-7.9.6 xa}2(elf)0( Extensions to the )2(COMMON)0( Directive )](7.9.6)193.4 subh3 -[2(ELF)0( also allows you to specify alignment requirements )(on common variables. This is done by putting a)]176.4 norm1 -[(number \(which must be a power of two\) after the name and size of the common variable, separated \(as usual\))]165.4 norm0 -[(by a colon. For example, an array of doublewords would benefit from 4-byte alignment:)]154.4 norm2 -[2(common dwordarray 128:4)]137.4 code3 -[(This declares the total size of the array to be 128 bytes, and requires that it be aligned on a 4-byte boundary.)]120.4 norm3 -(87)pageodd -restore showpage -%%Page: 88 88 -%%BeginPageSetup -save -%%EndPageSetup -/88 pa -[{/section-7.9.7 xa}(16-bit code and ELF )](7.9.7)678.8 subh3 -[(The )2(ELF32)0( specification doesn't provide relocations for 8- and 16-bit values, but the GNU )2(ld)0( linker adds)]661.8 norm1 -[(these as an extension. NASM can generate GNU-compatible relocations, to allow 16-bit code to be linked as)]650.8 norm0 -[(ELF using GNU )2(ld)0(. If NASM is used with the )2(-w+gnu-elf-extensions)0( option, a warning is issued)]639.8 norm0 -[(when one of these relocations is generated.)]628.8 norm2 -[{/section-7.9.8 xa}(Debug formats and ELF )](7.9.8)609.6 subh3 -[2(ELF32)0( and )2(ELF64)0( provide debug information in )2(STABS)0( and )2(DWARF)0( formats. Line number information is)]592.6 norm1 -[(generated for all executable sections, but please note that only the ".text" section is executable by default.)]581.6 norm2 -[{/section-7.10 xa}2(aout)0(: Linux )2(a.out)0( Object Files)](7.10)560.2 head3 -[(The )2(aout)0( format generates )2(a.out)0( object files, in the form used by early Linux systems \(current Linux)]543.2 norm1 -[(systems use ELF, see ){/section-7.9 xl}(section 7.9){el}(.\) These differ from other )2(a.out)0( object files in that the magic number in the)]532.2 norm0 -[(first four bytes of the file is different; also, some implementations of )2(a.out)0(, for example NetBSD's, support)]521.2 norm0 -[(position-independent code, which Linux's implementation does not.)]510.2 norm2 -[2(a.out)0( provides a default output file-name extension of )2(.o)0(.)]493.2 norm3 -[2(a.out)0( is a very simple object format. It supports no special directives, no special symbols, no use of )2(SEG)0( or)]476.2 norm1 -[2(WRT)0(, and no extensions to any standard directives. It supports only the three )(standard section names )2(.text)0(,)]465.2 norm0 -[2(.data)0( and )2(.bss)0(.)]454.2 norm2 -[{/section-7.11 xa}2(aoutb)0(: )(NetBSD/)(FreeBSD/)(OpenBSD )2(a.out)0( Object Files)](7.11)432.8 head3 -[(The )2(aoutb)0( format generates )2(a.out)0( object files, in the form used by the various free )2(BSD Unix)0( clones,)]415.8 norm1 -[2(NetBSD)0(, )2(FreeBSD)0( and )2(OpenBSD)0(. For simple object files, this object format is exactly the same as )2(aout)]404.8 norm0 -[(except for the magic number in the first four bytes of the file. However, the )2(aoutb)0( format supports)]393.8 norm0 -[(position-independent code in the same way as the )2(elf)0( format, so you can use it to write )2(BSD)0( )(shared libraries.)]382.8 norm2 -[2(aoutb)0( provides a default output file-name extension of )2(.o)0(.)]365.8 norm3 -[2(aoutb)0( supports no special directives, no special symbols, and only the three )(standard section names )2(.text)0(,)]348.8 norm1 -[2(.data)0( and )2(.bss)0(. However, it also supports the same use of )2(WRT)0( as )2(elf)0( does, to provide)]337.8 norm0 -[(position-independent code relocation types. See ){/section-7.9.3 xl}(section 7.9.3){el}( for full documentation of this feature.)]326.8 norm2 -[2(aoutb)0( also supports the same extensions to the )2(GLOBAL)0( directive as )2(elf)0( does: see ){/section-7.9.5 xl}(section 7.9.5){el}( for)]309.8 norm1 -[(documentation of this.)]298.8 norm2 -[{/section-7.12 xa}2(as86)0(: )(Minix/Linux)( )2(as86)0( Object Files)](7.12)277.4 head3 -[(The Minix/Linux 16-bit assembler )2(as86)0( has its own non-standard object file format. Although its)]260.4 norm1 -[(companion linker )2(ld86)0( produces something close to ordinary )2(a.out)0( binaries as output, the object file)]249.4 norm0 -[(format used to communicate between )2(as86)0( and )2(ld86)0( is not itself )2(a.out)0(.)]238.4 norm2 -[(NASM supports this format, just in case it is useful, as )2(as86)0(. )2(as86)0( provides a default output file-name)]221.4 norm1 -[(extension of )2(.o)0(.)]210.4 norm2 -[2(as86)0( is a very simple object format \(from the NASM user's point of view\). It supports no special directives,)]193.4 norm1 -[(no use of )2(SEG)0( or )2(WRT)0(, and no extensions to any standard directives. It supports only the three )(standard section)]182.4 norm0 -[(names )2(.text)0(, )2(.data)0( and )2(.bss)0(. The only special symbol supported is )2(..start)0(.)]171.4 norm2 -[{/section-7.13 xa}2(rdf)0(: )(Relocatable Dynamic Object File Format)](7.13)150 head3 -[(The )2(rdf)0( output format produces )2(RDOFF)0( object files. )2(RDOFF)0( \(Relocatable Dynamic Object File Format\) is a)]133 norm1 -[(home-grown object-file format, designed alongside NASM itself and reflecting in its file format the internal)]122 norm0 -[(structure of the assembler.)]111 norm2 -(88)pageeven -restore showpage -%%Page: 89 89 -%%BeginPageSetup -save -%%EndPageSetup -/89 pa -[2(RDOFF)0( is not used by any well-known operating systems. Those writing their own systems, however, may)]681 norm1 -[(well wish to use )2(RDOFF)0( as their object format, on the grounds that it is designed primarily for simplicity and)]670 norm0 -[(contains very little file-header bureaucracy.)]659 norm2 -[(The Unix NASM archive, and the DOS archive which includes sources, both contain an )2(rdoff)0( subdirectory)]642 norm1 -[(holding a set of RDOFF utilities: an RDF linker, an )2(RDF)0( static-library manager, an RDF file dump utility,)]631 norm0 -[(and a program which will load and execute an RDF executable under Linux.)]620 norm2 -[2(rdf)0( supports only the )(standard section names )2(.text)0(, )2(.data)0( and )2(.bss)0(.)]603 norm3 -[{/section-7.13.1 xa}(Requiring a Library: The )2(LIBRARY)0( Directive)](7.13.1)583.8 subh3 -[2(RDOFF)0( contains a mechanism for an object file to demand a given library to be linked to the module, either at)]566.8 norm1 -[(load time or run time. This is done by the )2(LIBRARY)0( directive, which takes one argument which is the name)]555.8 norm0 -[(of the module:)]544.8 norm2 -[2( library mylib.rdl)]527.8 code3 -[{/section-7.13.2 xa}(Specifying a Module Name: The )2(MODULE)0( Directive)](7.13.2)508.6 subh3 -[(Special )2(RDOFF)0( header record is used to store the name of the module. It can be used, for example, by)]491.6 norm1 -[(run-time loader to perform dynamic linking. )2(MODULE)0( directive takes one argument which is the name of)]480.6 norm0 -[(current module:)]469.6 norm2 -[2( module mymodname)]452.6 code3 -[(Note that when you statically link modules and tell linker to strip the symbols from output file, all module)]435.6 norm1 -[(names will be stripped too. To avoid it, you should start module names with )2($)0(, like:)]424.6 norm2 -[2( module $kernel.core)]407.6 code3 -[{/section-7.13.3 xa}2(rdf)0( Extensions to the )2(GLOBAL)0( Directive)](7.13.3)388.4 subh3 -[2(RDOFF)0( global symbols can contain additional information needed by the static linker. You can mark a global)]371.4 norm1 -[(symbol as exported, thus telling the linker do not strip it from target executable or library file. Like in )2(ELF)0(,)]360.4 norm0 -[(you can also specify whether an exported symbol is a procedure \(function\) or data object.)]349.4 norm2 -[(Suffixing the name with a colon and the word )2(export)0( you make the symbol exported:)]332.4 norm3 -[2( global sys_open:export)]315.4 code3 -[(To specify that exported symbol is a procedure \(function\), you add the word )2(proc)0( or )2(function)0( after)]298.4 norm1 -[(declaration:)]287.4 norm2 -[2( global sys_open:export proc)]270.4 code3 -[(Similarly, to specify exported data object, add the word )2(data)0( or )2(object)0( to the directive:)]253.4 norm3 -[2( global kernel_ticks:export data)]236.4 code3 -[{/section-7.13.4 xa}2(rdf)0( Extensions to the )2(EXTERN)0( Directive)](7.13.4)217.2 subh3 -[(By default the )2(EXTERN)0( directive in )2(RDOFF)0( declares a "pure external" symbol \(i.e. the static linker will)]200.2 norm1 -[(complain if such a symbol is not resolved\). To declare an "imported" symbol, which must be resolved later)]189.2 norm0 -[(during a dynamic linking phase, )2(RDOFF)0( offers an additional )2(import)0( modifier. As in )2(GLOBAL)0(, you can also)]178.2 norm0 -[(specify whether an imported symbol is a procedure \(function\) or data object. For example:)]167.2 norm2 -[2( library $libc )]150.2 code1 -[2( extern _open:import )]139.2 code0 -[2( extern _printf:import proc )]128.2 code0 -[2( extern _errno:import data)]117.2 code2 -(89)pageodd -restore showpage -%%Page: 90 90 -%%BeginPageSetup -save -%%EndPageSetup -/90 pa -[(Here the directive )2(LIBRARY)0( is also included, which gives the dynamic linker a hint as to where to find)]681 norm1 -[(requested symbols.)]670 norm2 -[{/section-7.14 xa}2(dbg)0(: Debugging Format)](7.14)648.6 head3 -[(The )2(dbg)0( output format is not built into NASM in the default configuration. If you are building your own)]631.6 norm1 -[(NASM executable from the sources, you can define )2(OF_DBG)0( in )2(output/outform.h)0( or on the compiler)]620.6 norm0 -[(command line, and obtain the )2(dbg)0( output format.)]609.6 norm2 -[(The )2(dbg)0( format does not output an object file as such; instead, it outputs a text file which contains a complete)]592.6 norm1 -[(list of all the transactions between the main body of NASM and the output-format back end module. It is)]581.6 norm0 -[(primarily intended to aid people who want to write their own output drivers, so that they can get a clearer idea)]570.6 norm0 -[(of the various requests the main program makes of the output driver, and in what order they happen.)]559.6 norm2 -[(For simple files, one can easily use the )2(dbg)0( format like this:)]542.6 norm3 -[2(nasm -f dbg filename.asm)]525.6 code3 -[(which will generate a diagnostic file called )2(filename.dbg)0(. However, this will not work well on files)]508.6 norm1 -[(which were designed for a different object format, because each object format defines its own macros \(usually)]497.6 norm0 -[(user-level forms of directives\), and those macros will not be defined in the )2(dbg)0( format. Therefore it can be)]486.6 norm0 -[(useful to run NASM twice, in order to do the preprocessing with the native object format selected:)]475.6 norm2 -[2(nasm -e -f rdf -o rdfprog.i rdfprog.asm )]458.6 code1 -[2(nasm -a -f dbg rdfprog.i)]447.6 code2 -[(This preprocesses )2(rdfprog.asm)0( into )2(rdfprog.i)0(, keeping the )2(rdf)0( object format selected in order to)]430.6 norm1 -[(make sure RDF special directives are converted into primitive form correctly. Then the preprocessed source is)]419.6 norm0 -[(fed through the )2(dbg)0( format to generate the final diagnostic output.)]408.6 norm2 -[(This workaround will still typically not work for programs intended for )2(obj)0( format, because the )2(obj)]391.6 norm1 -[2(SEGMENT)0( and )2(GROUP)0( directives have side effects of defining the segment and group names as symbols; )2(dbg)]380.6 norm0 -[(will not do this, so the program will not assemble. You will have to work around that by defining the symbols)]369.6 norm0 -[(yourself \(using )2(EXTERN)0(, for example\) if you really need to get a )2(dbg)0( trace of an )2(obj)0(\226specific source file.)]358.6 norm2 -[2(dbg)0( accepts any section name and any directives at all, and logs them all to its output file.)]341.6 norm3 -(90)pageeven -restore showpage -%%Page: 91 91 -%%BeginPageSetup -save -%%EndPageSetup -/91 pa -[{/chapter-8 xa}(Chapter 8: Writing 16-bit Code \(DOS, Windows 3/3.1\))]642.8 chap3 -[(This chapter attempts to cover some of the common issues encountered when writing 16-bit code to run)]607.8 norm1 -[(under )2(MS-DOS)0( or )2(Windows 3.x)0(. It covers how to link programs to produce )2(.EXE)0( or )2(.COM)0( files, how to)]596.8 norm0 -[(write )2(.SYS)0( device drivers, and how to interface assembly language code with 16-bit C compilers and with)]585.8 norm0 -[(Borland Pascal.)]574.8 norm2 -[{/section-8.1 xa}(Producing )2(.EXE)0( Files)](8.1)553.4 head3 -[(Any large program written under DOS needs to be built as a )2(.EXE)0( file: only )2(.EXE)0( files have the necessary)]536.4 norm1 -[(internal structure required to span more than one 64K segment. )(Windows programs, also, have to be built as)]525.4 norm0 -[2(.EXE)0( files, since Windows does not support the )2(.COM)0( format.)]514.4 norm2 -[(In general, you generate )2(.EXE)0( files by using the )2(obj)0( output format to produce one or more )2(.OBJ)0( files, and)]497.4 norm1 -[(then linking them together using a linker. However, NASM also supports the direct generation of simple DOS)]486.4 norm0 -[2(.EXE)0( files using the )2(bin)0( output format \(by using )2(DB)0( and )2(DW)0( to construct the )2(.EXE)0( file header\), and a macro)]475.4 norm0 -[(package is supplied to do this. Thanks to Yann Guidon for contributing the code for this.)]464.4 norm2 -[(NASM may also support )2(.EXE)0( natively as another output format in future releases.)]447.4 norm3 -[{/section-8.1.1 xa}(Using the )2(obj)0( Format To Generate )2(.EXE)0( Files)](8.1.1)428.2 subh3 -[(This section describes the usual method of generating )2(.EXE)0( files by linking )2(.OBJ)0( files together.)]411.2 norm3 -[(Most 16-bit programming language packages come with a suitable linker; if you have none of these, there is a)]394.2 norm1 -[(free linker called )(VAL)(, available in )2(LZH)0( archive format from ){(ftp://x2ftp.oulu.fi/pub/msdos/programming/lang/)wl}2(x2ftp.oulu.fi){el}0(. An LZH archiver can be)]383.2 norm0 -[(found at ){(ftp://ftp.simtel.net/pub/simtelnet/msdos/arcers)wl}2(ftp.simtel.net){el}0(. There is another `free' linker \(though this one doesn't come with sources\))]372.2 norm0 -[(called )(FREELINK, available from ){(http://www.pcorner.com/tpc/old/3-101.html)wl}2(www.pcorner.com){el}0(. A third, )2(djlink)0(, written by DJ Delorie, is)]361.2 norm0 -[(available at ){(http://www.delorie.com/djgpp/16bit/djlink/)wl}2(www.delorie.com){el}0(. A fourth linker, )2(ALINK)0(, written by Anthony A.J. Williams, is available at)]350.2 norm0 -[{(http://alink.sourceforge.net)wl}2(alink.sourceforge.net){el}0(.)]339.2 norm2 -[(When linking several )2(.OBJ)0( files into a )2(.EXE)0( file, you should ensure that exactly one of them has a start point)]322.2 norm1 -[(defined \(using the )2(..start)0( special symbol defined by the )2(obj)0( format: see ){/section-7.4.6 xl}(section 7.4.6){el}(\). If no module)]311.2 norm0 -[(defines a start point, the linker will not know what value to give the entry-point field in the output file header;)]300.2 norm0 -[(if more than one defines a start point, the linker will not know )1(which)0( value to use.)]289.2 norm2 -[(An example of a NASM source file which can be assembled to a )2(.OBJ)0( file and linked on its own to a )2(.EXE)]272.2 norm1 -[(is given here. It demonstrates the basic principles of defining a stack, initialising the segment registers, and)]261.2 norm0 -[(declaring a start point. This file is also provided in the )2(test)0( subdirectory of the NASM archives, under the)]250.2 norm0 -[(name )2(objexe.asm)0(.)]239.2 norm2 -[2(segment code )]222.2 code1 -[2()]211.2 code0 -[2(..start: )]200.2 code0 -[2( mov ax,data )]189.2 code0 -[2( mov ds,ax )]178.2 code0 -[2( mov ax,stack )]167.2 code0 -[2( mov ss,ax )]156.2 code0 -[2( mov sp,stacktop)]145.2 code2 -[(This initial piece of code sets up )2(DS)0( to point to the data segment, and initializes )2(SS)0( and )2(SP)0( to point to the top)]128.2 norm1 -[(of the provided stack. Notice that interrupts are implicitly disabled for one instruction after a move into )2(SS)0(,)]117.2 norm0 -(91)pageodd -restore showpage -%%Page: 92 92 -%%BeginPageSetup -save -%%EndPageSetup -/92 pa -[(precisely for this situation, so that there's no chance of an interrupt occurring between the loads of )2(SS)0( and )2(SP)]681 norm0 -[(and not having a stack to execute on.)]670 norm2 -[(Note also that the special symbol )2(..start)0( is defined at the beginning of this code, which means that will be)]653 norm1 -[(the entry point into the resulting executable file.)]642 norm2 -[2( mov dx,hello )]625 code1 -[2( mov ah,9 )]614 code0 -[2( int 0x21)]603 code2 -[(The above is the main program: load )2(DS:DX)0( with a pointer to the greeting message \()2(hello)0( is implicitly)]586 norm1 -[(relative to the segment )2(data)0(, which was loaded into )2(DS)0( in the setup code, so the full pointer is valid\), and)]575 norm0 -[(call the DOS print-string function.)]564 norm2 -[2( mov ax,0x4c00 )]547 code1 -[2( int 0x21)]536 code2 -[(This terminates the program using another DOS system call.)]519 norm3 -[2(segment data )]502 code1 -[2()]491 code0 -[2(hello: db 'hello, world', 13, 10, '$')]480 code2 -[(The data segment contains the string we want to display.)]463 norm3 -[2(segment stack stack )]446 code1 -[2( resb 64 )]435 code0 -[2(stacktop:)]424 code2 -[(The above code declares a stack segment containing 64 bytes of uninitialized stack space, and points)]407 norm1 -[2(stacktop)0( at the top of it. The directive )2(segment stack stack)0( defines a segment )1(called)0( )2(stack)0(, and)]396 norm0 -[(also of )1(type)0( )2(STACK)0(. The latter is not necessary to the correct running of the program, but linkers are likely to)]385 norm0 -[(issue warnings or errors if your program has no segment of type )2(STACK)0(.)]374 norm2 -[(The above file, when assembled into a )2(.OBJ)0( file, will link on its own to a valid )2(.EXE)0( file, which when run)]357 norm1 -[(will print `hello, world' and then exit.)]346 norm2 -[{/section-8.1.2 xa}(Using the )2(bin)0( Format To Generate )2(.EXE)0( Files)](8.1.2)326.8 subh3 -[(The )2(.EXE)0( file format is simple enough that it's possible to build a )2(.EXE)0( file by writing a pure-binary)]309.8 norm1 -[(program and sticking a 32-byte header on the front. This header is simple enough that it can be generated)]298.8 norm0 -[(using )2(DB)0( and )2(DW)0( commands by NASM itself, so that you can use the )2(bin)0( output format to directly generate)]287.8 norm0 -[2(.EXE)0( files.)]276.8 norm2 -[(Included in the NASM archives, in the )2(misc)0( subdirectory, is a file )2(exebin.mac)0( of macros. It defines three)]259.8 norm1 -[(macros: )2(EXE_begin)0(, )2(EXE_stack)0( and )2(EXE_end)0(.)]248.8 norm2 -[(To produce a )2(.EXE)0( file using this method, you should start by using )2(%include)0( to load the )2(exebin.mac)]231.8 norm1 -[(macro package into your source file. You should then issue the )2(EXE_begin)0( macro call \(which takes no)]220.8 norm0 -[(arguments\) to generate the file header data. Then write code as normal for the )2(bin)0( format \226 you can use all)]209.8 norm0 -[(three standard sections )2(.text)0(, )2(.data)0( and )2(.bss)0(. At the end of the file you should call the )2(EXE_end)0( macro)]198.8 norm0 -[(\(again, no arguments\), which defines some symbols to mark section sizes, and these symbols are referred to in)]187.8 norm0 -[(the header code generated by )2(EXE_begin)0(.)]176.8 norm2 -[(In this model, the code you end up writing starts at )2(0x100)0(, just like a )2(.COM)0( file \226 in fact, if you strip off the)]159.8 norm1 -[(32-byte header from the resulting )2(.EXE)0( file, you will have a valid )2(.COM)0( program. All the segment bases are)]148.8 norm0 -[(the same, so you are limited to a 64K program, again just like a )2(.COM)0( file. Note that an )2(ORG)0( directive is)]137.8 norm0 -[(issued by the )2(EXE_begin)0( macro, so you should not explicitly issue one of your own.)]126.8 norm2 -(92)pageeven -restore showpage -%%Page: 93 93 -%%BeginPageSetup -save -%%EndPageSetup -/93 pa -[(You can't directly refer to your segment base value, unfortunately, since this would require a relocation in the)]681 norm1 -[(header, and things would get a lot more complicated. So you should get your segment base by copying it out)]670 norm0 -[(of )2(CS)0( instead.)]659 norm2 -[(On entry to your )2(.EXE)0( file, )2(SS:SP)0( are already set up to point to the top of a 2Kb stack. You can adjust the)]642 norm1 -[(default stack size of 2Kb by calling the )2(EXE_stack)0( macro. For example, to change the stack size of your)]631 norm0 -[(program to 64 bytes, you would call )2(EXE_stack 64)0(.)]620 norm2 -[(A sample program which generates a )2(.EXE)0( file in this way is given in the )2(test)0( subdirectory of the NASM)]603 norm1 -[(archive, as )2(binexe.asm)0(.)]592 norm2 -[{/section-8.2 xa}(Producing )2(.COM)0( Files)](8.2)570.6 head3 -[(While large DOS programs must be written as )2(.EXE)0( files, small ones are often better written as )2(.COM)0( files.)]553.6 norm1 -[2(.COM)0( files are pure binary, and therefore most easily produced using the )2(bin)0( output format.)]542.6 norm2 -[{/section-8.2.1 xa}(Using the )2(bin)0( Format To Generate )2(.COM)0( Files)](8.2.1)523.4 subh3 -[2(.COM)0( files expect to be loaded at offset )2(100h)0( into their segment \(though the segment may change\).)]506.4 norm1 -[(Execution then begins at )2(100h)0(, i.e. right at the start of the program. So to write a )2(.COM)0( program, you would)]495.4 norm0 -[(create a source file looking like)]484.4 norm2 -[2( org 100h )]467.4 code1 -[2()]456.4 code0 -[2(section .text )]445.4 code0 -[2()]434.4 code0 -[2(start: )]423.4 code0 -[2( ; put your code here )]412.4 code0 -[2()]401.4 code0 -[2(section .data )]390.4 code0 -[2()]379.4 code0 -[2( ; put data items here )]368.4 code0 -[2()]357.4 code0 -[2(section .bss )]346.4 code0 -[2()]335.4 code0 -[2( ; put uninitialized data here)]324.4 code2 -[(The )2(bin)0( format puts the )2(.text)0( section first in the file, so you can declare data or BSS items before)]307.4 norm1 -[(beginning to write code if you want to and the code will still end up at the front of the file where it belongs.)]296.4 norm2 -[(The BSS \(uninitialized data\) section does not take up space in the )2(.COM)0( file itself: instead, addresses of BSS)]279.4 norm1 -[(items are resolved to point at space beyond the end of the file, on the grounds that this will be free memory)]268.4 norm0 -[(when the program is run. Therefore you should not rely on your BSS being initialized to all zeros when you)]257.4 norm0 -[(run.)]246.4 norm2 -[(To assemble the above program, you should use a command line like)]229.4 norm3 -[2(nasm myprog.asm -fbin -o myprog.com)]212.4 code3 -[(The )2(bin)0( format would produce a file called )2(myprog)0( if no explicit output file name were specified, so you)]195.4 norm1 -[(have to override it and give the desired file name.)]184.4 norm2 -[{/section-8.2.2 xa}(Using the )2(obj)0( Format To Generate )2(.COM)0( Files)](8.2.2)165.2 subh3 -[(If you are writing a )2(.COM)0( program as more than one module, you may wish to assemble several )2(.OBJ)0( files)]148.2 norm1 -[(and link them together into a )2(.COM)0( program. You can do this, provided you have a linker capable of)]137.2 norm0 -[(outputting )2(.COM)0( files directly \()(TLINK does this\), or alternatively a converter program such as )2(EXE2BIN)0( to)]126.2 norm0 -[(transform the )2(.EXE)0( file output from the linker into a )2(.COM)0( file.)]115.2 norm2 -(93)pageodd -restore showpage -%%Page: 94 94 -%%BeginPageSetup -save -%%EndPageSetup -/94 pa -[(If you do this, you need to take care of several things:)]681 norm3 -[(The first object file containing code should start its code segment with a line like )2(RESB 100h)0(. This is to)]664 bull1 -[(ensure that the code begins at offset )2(100h)0( relative to the beginning of the code segment, so that the linker)]653 bull0 -[(or converter program does not have to adjust address references within the file when generating the )2(.COM)]642 bull0 -[(file. Other assemblers use an )2(ORG)0( directive for this purpose, but )2(ORG)0( in NASM is a format-specific)]631 bull0 -[(directive to the )2(bin)0( output format, and does not mean the same thing as it does in MASM-compatible)]620 bull0 -[(assemblers.)]609 bull2 -[(You don't need to define a stack segment.)]592 bull3 -[(All your segments should be in the same group, so that every time your code or data references a symbol)]575 bull1 -[(offset, all offsets are relative to the same segment base. This is because, when a )2(.COM)0( file is loaded, all the)]564 bull0 -[(segment registers contain the same value.)]553 bull2 -[{/section-8.3 xa}(Producing )2(.SYS)0( Files)](8.3)531.6 head3 -[(MS-DOS device drivers \226 )2(.SYS)0( files \226 are pure binary files, similar to )2(.COM)0( files, except that they start at)]514.6 norm1 -[(origin zero rather than )2(100h)0(. Therefore, if you are writing a device driver using the )2(bin)0( format, you do not)]503.6 norm0 -[(need the )2(ORG)0( directive, since the default origin for )2(bin)0( is zero. Similarly, if you are using )2(obj)0(, you do not)]492.6 norm0 -[(need the )2(RESB 100h)0( at the start of your code segment.)]481.6 norm2 -[2(.SYS)0( files start with a header structure, containing pointers to the various routines inside the driver which do)]464.6 norm1 -[(the work. This structure should be defined at the start of the code segment, even though it is not actually code.)]453.6 norm2 -[(For more information on the format of )2(.SYS)0( files, and the data which has to go in the header structure, a list)]436.6 norm1 -[(of books is given in the Frequently Asked Questions list for the newsgroup)]425.6 norm0 -[{(news:comp.os.msdos.programmer)wl}2(comp.os.msdos.programmer){el}0(.)]414.6 norm2 -[{/section-8.4 xa}(Interfacing to 16-bit C Programs)](8.4)393.2 head3 -[(This section covers the basics of writing assembly routines that call, or are called from, C programs. To do)]376.2 norm1 -[(this, you would typically write an assembly module as a )2(.OBJ)0( file, and link it with your C modules to)]365.2 norm0 -[(produce a )(mixed-language program.)]354.2 norm2 -[{/section-8.4.1 xa}(External Symbol Names)](8.4.1)335 subh3 -[(C compilers have the convention that the names of all global symbols \(functions or data\) they define are)]318 norm1 -[(formed by prefixing an underscore to the name as it appears in the C program. So, for example, the function a)]307 norm0 -[(C programmer thinks of as )2(printf)0( appears to an assembly language programmer as )2(_printf)0(. This means)]296 norm0 -[(that in your assembly programs, you can define symbols without a leading underscore, and not have to worry)]285 norm0 -[(about name clashes with C symbols.)]274 norm2 -[(If you find the underscores inconvenient, you can define macros to replace the )2(GLOBAL)0( and )2(EXTERN)]257 norm1 -[(directives as follows:)]246 norm2 -[2(%macro cglobal 1 )]229 code1 -[2()]218 code0 -[2( global _%1 )]207 code0 -[2( %define %1 _%1 )]196 code0 -[2()]185 code0 -[2(%endmacro )]174 code0 -[2()]163 code0 -[2(%macro cextern 1 )]152 code0 -[2()]141 code0 -[2( extern _%1 )]130 code0 -[2( %define %1 _%1 )]119 code0 -(94)pageeven -restore showpage -%%Page: 95 95 -%%BeginPageSetup -save -%%EndPageSetup -/95 pa -[2()]681 code0 -[2(%endmacro)]670 code2 -[(\(These forms of the macros only take one argument at a time; a )2(%rep)0( construct could solve this.\))]653 norm3 -[(If you then declare an external like this:)]636 norm3 -[2(cextern printf)]619 code3 -[(then the macro will expand it as)]602 norm3 -[2(extern _printf )]585 code1 -[2(%define printf _printf)]574 code2 -[(Thereafter, you can reference )2(printf)0( as if it was a symbol, and the preprocessor will put the leading)]557 norm1 -[(underscore on where necessary.)]546 norm2 -[(The )2(cglobal)0( macro works similarly. You must use )2(cglobal)0( before defining the symbol in question, but)]529 norm1 -[(you would have had to do that anyway if you used )2(GLOBAL)0(.)]518 norm2 -[(Also see ){/section-2.1.27 xl}(section 2.1.27){el}(.)]501 norm3 -[{/section-8.4.2 xa}(Memory Models)](8.4.2)481.8 subh3 -[(NASM contains no mechanism to support the various C memory models directly; you have to keep track)]464.8 norm1 -[(yourself of which one you are writing for. This means you have to keep track of the following things:)]453.8 norm2 -[(In models using a single code segment \(tiny, small and compact\), functions are near. This means that)]436.8 bull1 -[(function pointers, when stored in data segments or pushed on the stack as function arguments, are 16 bits)]425.8 bull0 -[(long and contain only an offset field \(the )2(CS)0( register never changes its value, and always gives the segment)]414.8 bull0 -[(part of the full function address\), and that functions are called using ordinary near )2(CALL)0( instructions and)]403.8 bull0 -[(return using )2(RETN)0( \(which, in NASM, is synonymous with )2(RET)0( anyway\). This means both that you should)]392.8 bull0 -[(write your own routines to return with )2(RETN)0(, and that you should call external C routines with near )2(CALL)]381.8 bull0 -[(instructions.)]370.8 bull2 -[(In models using more than one code segment \(medium, large and huge\), functions are far. This means that)]353.8 bull1 -[(function pointers are 32 bits long \(consisting of a 16-bit offset followed by a 16-bit segment\), and that)]342.8 bull0 -[(functions are called using )2(CALL FAR)0( \(or )2(CALL seg:offset)0(\) and return using )2(RETF)0(. Again, you)]331.8 bull0 -[(should therefore write your own routines to return with )2(RETF)0( and use )2(CALL FAR)0( to call external routines.)]320.8 bull2 -[(In models using a single data segment \(tiny, small and medium\), data pointers are 16 bits long, containing)]303.8 bull1 -[(only an offset field \(the )2(DS)0( register doesn't change its value, and always gives the segment part of the full)]292.8 bull0 -[(data item address\).)]281.8 bull2 -[(In models using more than one data segment \(compact, large and huge\), data pointers are 32 bits long,)]264.8 bull1 -[(consisting of a 16-bit offset followed by a 16-bit segment. You should still be careful not to modify )2(DS)0( in)]253.8 bull0 -[(your routines without restoring it afterwards, but )2(ES)0( is free for you to use to access the contents of 32-bit)]242.8 bull0 -[(data pointers you are passed.)]231.8 bull2 -[(The huge memory model allows single data items to exceed 64K in size. In all other memory models, you)]214.8 bull1 -[(can access the whole of a data item just by doing arithmetic on the offset field of the pointer you are given,)]203.8 bull0 -[(whether a segment field is present or not; in huge model, you have to be more careful of your pointer)]192.8 bull0 -[(arithmetic.)]181.8 bull2 -[(In most memory models, there is a )1(default)0( data segment, whose segment address is kept in )2(DS)0( throughout)]164.8 bull1 -[(the program. This data segment is typically the same segment as the stack, kept in )2(SS)0(, so that functions')]153.8 bull0 -[(local variables \(which are stored on the stack\) and global data items can both be accessed easily without)]142.8 bull0 -[(changing )2(DS)0(. Particularly large data items are typically stored in other segments. However, some memory)]131.8 bull0 -[(models \(though not the standard ones, usually\) allow the assumption that )2(SS)0( and )2(DS)0( hold the same value to)]120.8 bull0 -[(be removed. Be careful about functions' local variables in this latter case.)]109.8 bull2 -(95)pageodd -restore showpage -%%Page: 96 96 -%%BeginPageSetup -save -%%EndPageSetup -/96 pa -[(In models with a single code segment, the segment is called )2(_TEXT)0(, so your code segment must also go by)]681 norm1 -[(this name in order to be linked into the same place as the main code segment. In models with a single data)]670 norm0 -[(segment, or with a default data segment, it is called )2(_DATA)0(.)]659 norm2 -[{/section-8.4.3 xa}(Function Definitions and Function Calls)](8.4.3)639.8 subh3 -[(The )(C calling convention in 16-bit programs is as follows. In the following description, the words )1(caller)0( and)]622.8 norm1 -[1(callee)0( are used to denote the function doing the calling and the function which gets called.)]611.8 norm2 -[(The caller pushes the function's parameters on the stack, one after another, in reverse order \(right to left, so)]594.8 bull1 -[(that the first argument specified to the function is pushed last\).)]583.8 bull2 -[(The caller then executes a )2(CALL)0( instruction to pass control to the callee. This )2(CALL)0( is either near or far)]566.8 bull1 -[(depending on the memory model.)]555.8 bull2 -[(The callee receives control, and typically \(although this is not actually necessary, in functions which do not)]538.8 bull1 -[(need to access their parameters\) starts by saving the value of )2(SP)0( in )2(BP)0( so as to be able to use )2(BP)0( as a base)]527.8 bull0 -[(pointer to find its parameters on the stack. However, the caller was probably doing this too, so part of the)]516.8 bull0 -[(calling convention states that )2(BP)0( must be preserved by any C function. Hence the callee, if it is going to set)]505.8 bull0 -[(up )2(BP)0( as a )1(frame pointer)0(, must push the previous value first.)]494.8 bull2 -[(The callee may then access its parameters relative to )2(BP)0(. The word at )2([BP])0( holds the previous value of )2(BP)]477.8 bull1 -[(as it was pushed; the next word, at )2([BP+2])0(, holds the offset part of the return address, pushed implicitly)]466.8 bull0 -[(by )2(CALL)0(. In a small-model \(near\) function, the parameters start after that, at )2([BP+4])0(; in a large-model)]455.8 bull0 -[(\(far\) function, the segment part of the return address lives at )2([BP+4])0(, and the parameters begin at)]444.8 bull0 -[2([BP+6])0(. The leftmost parameter of the function, since it was pushed last, is accessible at this offset from)]433.8 bull0 -[2(BP)0(; the others follow, at successively greater offsets. Thus, in a function such as )2(printf)0( which takes a)]422.8 bull0 -[(variable number of parameters, the pushing of the parameters in reverse order means that the function)]411.8 bull0 -[(knows where to find its first parameter, which tells it the number and type of the remaining ones.)]400.8 bull2 -[(The callee may also wish to decrease )2(SP)0( further, so as to allocate space on the stack for local variables,)]383.8 bull1 -[(which will then be accessible at negative offsets from )2(BP)0(.)]372.8 bull2 -[(The callee, if it wishes to return a value to the caller, should leave the value in )2(AL)0(, )2(AX)0( or )2(DX:AX)]355.8 bull1 -[(depending on the size of the value. Floating-point results are sometimes \(depending on the compiler\))]344.8 bull0 -[(returned in )2(ST0)0(.)]333.8 bull2 -[(Once the callee has finished processing, it restores )2(SP)0( from )2(BP)0( if it had allocated local stack space, then)]316.8 bull1 -[(pops the previous value of )2(BP)0(, and returns via )2(RETN)0( or )2(RETF)0( depending on memory model.)]305.8 bull2 -[(When the caller regains control from the callee, the function parameters are still on the stack, so it typically)]288.8 bull1 -[(adds an immediate constant to )2(SP)0( to remove them \(instead of executing a number of slow )2(POP)]277.8 bull0 -[(instructions\). Thus, if a function is accidentally called with the wrong number of parameters due to a)]266.8 bull0 -[(prototype mismatch, the stack will still be returned to a sensible state since the caller, which )1(knows)0( how)]255.8 bull0 -[(many parameters it pushed, does the removing.)]244.8 bull2 -[(It is instructive to compare this calling convention with that for Pascal programs \(described in ){/section-8.5.1 xl}(section 8.5.1){el}(\).)]227.8 norm1 -[(Pascal has a simpler convention, since no functions have variable numbers of parameters. Therefore the callee)]216.8 norm0 -[(knows how many parameters it should have been passed, and is able to deallocate them from the stack itself)]205.8 norm0 -[(by passing an immediate argument to the )2(RET)0( or )2(RETF)0( instruction, so the caller does not have to do it. Also,)]194.8 norm0 -[(the parameters are pushed in left-to-right order, not right-to-left, which means that a compiler can give)]183.8 norm0 -[(better guarantees about sequence points without performance suffering.)]172.8 norm2 -[(Thus, you would define a function in C style in the following way. The following example is for small model:)]155.8 norm3 -[2(global _myfunc )]138.8 code1 -[2()]127.8 code0 -[2(_myfunc: )]116.8 code0 -[2( push bp )]105.8 code0 -(96)pageeven -restore showpage -%%Page: 97 97 -%%BeginPageSetup -save -%%EndPageSetup -/97 pa -[2( mov bp,sp )]681 code0 -[2( sub sp,0x40 ; 64 bytes of local stack space )]670 code0 -[2( mov bx,[bp+4] ; first parameter to function )]659 code0 -[2()]648 code0 -[2( ; some more code )]637 code0 -[2()]626 code0 -[2( mov sp,bp ; undo "sub sp,0x40" above )]615 code0 -[2( pop bp )]604 code0 -[2( ret)]593 code2 -[(For a large-model function, you would replace )2(RET)0( by )2(RETF)0(, and look for the first parameter at )2([BP+6])]576 norm1 -[(instead of )2([BP+4])0(. Of course, if one of the parameters is a pointer, then the offsets of )1(subsequent)0( parameters)]565 norm0 -[(will change depending on the memory model as well: far pointers take up four bytes on the stack when passed)]554 norm0 -[(as a parameter, whereas near pointers take up two.)]543 norm2 -[(At the other end of the process, to call a C function from your assembly code, you would do something like)]526 norm1 -[(this:)]515 norm2 -[2(extern _printf )]498 code1 -[2()]487 code0 -[2( ; and then, further down... )]476 code0 -[2()]465 code0 -[2( push word [myint] ; one of my integer variables )]454 code0 -[2( push word mystring ; pointer into my data segment )]443 code0 -[2( call _printf )]432 code0 -[2( add sp,byte 4 ; `byte' saves space )]421 code0 -[2()]410 code0 -[2( ; then those data items... )]399 code0 -[2()]388 code0 -[2(segment _DATA )]377 code0 -[2()]366 code0 -[2(myint dw 1234 )]355 code0 -[2(mystring db 'This number -> %d <- should be 1234',10,0)]344 code2 -[(This piece of code is the small-model assembly equivalent of the C code)]327 norm3 -[2( int myint = 1234; )]310 code1 -[2( printf\("This number -> %d <- should be 1234\\n", myint\);)]299 code2 -[(In large model, the function-call code might look more like this. In this example, it is assumed that )2(DS)]282 norm1 -[(already holds the segment base of the segment )2(_DATA)0(. If not, you would have to initialize it first.)]271 norm2 -[2( push word [myint] )]254 code1 -[2( push word seg mystring ; Now push the segment, and... )]243 code0 -[2( push word mystring ; ... offset of "mystring" )]232 code0 -[2( call far _printf )]221 code0 -[2( add sp,byte 6)]210 code2 -[(The integer value still takes up one word on the stack, since large model does not affect the size of the )2(int)]193 norm1 -[(data type. The first argument \(pushed last\) to )2(printf)0(, however, is a data pointer, and therefore has to)]182 norm0 -[(contain a segment and offset part. The segment should be stored second in memory, and therefore must be)]171 norm0 -[(pushed first. \(Of course, )2(PUSH DS)0( would have been a shorter instruction than)]160 norm0 -[2(PUSH WORD SEG mystring)0(, if )2(DS)0( was set up as the above example assumed.\) Then the actual call)]149 norm0 -[(becomes a far call, since functions expect far calls in large model; and )2(SP)0( has to be increased by 6 rather than)]138 norm0 -[(4 afterwards to make up for the extra word of parameters.)]127 norm2 -(97)pageodd -restore showpage -%%Page: 98 98 -%%BeginPageSetup -save -%%EndPageSetup -/98 pa -[{/section-8.4.4 xa}(Accessing Data Items)](8.4.4)678.8 subh3 -[(To get at the contents of C variables, or to declare variables which C can access, you need only declare the)]661.8 norm1 -[(names as )2(GLOBAL)0( or )2(EXTERN)0(. \(Again, the names require leading underscores, as stated in ){/section-8.4.1 xl}(section 8.4.1){el}(.\))]650.8 norm0 -[(Thus, a C variable declared as )2(int i)0( can be accessed from assembler as)]639.8 norm2 -[2(extern _i )]622.8 code1 -[2()]611.8 code0 -[2( mov ax,[_i])]600.8 code2 -[(And to declare your own integer variable which C programs can access as )2(extern int j)0(, you do this)]583.8 norm1 -[(\(making sure you are assembling in the )2(_DATA)0( segment, if necessary\):)]572.8 norm2 -[2(global _j )]555.8 code1 -[2()]544.8 code0 -[2(_j dw 0)]533.8 code2 -[(To access a C array, you need to know the size of the components of the array. For example, )2(int)0( variables)]516.8 norm1 -[(are two bytes long, so if a C program declares an array as )2(int a[10])0(, you can access )2(a[3])0( by coding)]505.8 norm0 -[2(mov ax,[_a+6])0(. \(The byte offset 6 is obtained by multiplying the desired array index, 3, by the size of the)]494.8 norm0 -[(array element, 2.\) The sizes of the C base types in 16-bit compilers are: 1 for )2(char)0(, 2 for )2(short)0( and )2(int)0(, 4)]483.8 norm0 -[(for )2(long)0( and )2(float)0(, and 8 for )2(double)0(.)]472.8 norm2 -[(To access a C )(data structure, you need to know the offset from the base of the structure to the field you are)]455.8 norm1 -[(interested in. You can either do this by converting the C structure definition into a NASM structure definition)]444.8 norm0 -[(\(using )2(STRUC)0(\), or by calculating the one offset and using just that.)]433.8 norm2 -[(To do either of these, you should read your C compiler's manual to find out how it organizes data structures.)]416.8 norm1 -[(NASM gives no special alignment to structure members in its own )2(STRUC)0( macro, so you have to specify)]405.8 norm0 -[(alignment yourself if the C compiler generates it. Typically, you might find that a structure like)]394.8 norm2 -[2(struct { )]377.8 code1 -[2( char c; )]366.8 code0 -[2( int i; )]355.8 code0 -[2(} foo;)]344.8 code2 -[(might be four bytes long rather than three, since the )2(int)0( field would be aligned to a two-byte boundary.)]327.8 norm1 -[(However, this sort of feature tends to be a configurable option in the C compiler, either using command-line)]316.8 norm0 -[(options or )2(#pragma)0( lines, so you have to find out how your own compiler does it.)]305.8 norm2 -[{/section-8.4.5 xa}2(c16.mac)0(: Helper Macros for the 16-bit C Interface)](8.4.5)286.6 subh3 -[(Included in the NASM archives, in the )2(misc)0( directory, is a file )2(c16.mac)0( of macros. It defines three macros:)]269.6 norm1 -[2(proc)0(, )2(arg)0( and )2(endproc)0(. These are intended to be used for C-style procedure definitions, and they)]258.6 norm0 -[(automate a lot of the work involved in keeping track of the calling convention.)]247.6 norm2 -[(\(An alternative, TASM compatible form of )2(arg)0( is also now built into NASM's preprocessor. See ){/section-4.8 xl}(section 4.8){el}]230.6 norm1 -[(for details.\))]219.6 norm2 -[(An example of an assembly function using the macro set is given here:)]202.6 norm3 -[2(proc _nearproc )]185.6 code1 -[2()]174.6 code0 -[2(%$i arg )]163.6 code0 -[2(%$j arg )]152.6 code0 -[2( mov ax,[bp + %$i] )]141.6 code0 -[2( mov bx,[bp + %$j] )]130.6 code0 -[2( add ax,[bx] )]119.6 code0 -(98)pageeven -restore showpage -%%Page: 99 99 -%%BeginPageSetup -save -%%EndPageSetup -/99 pa -[2()]681 code0 -[2(endproc)]670 code2 -[(This defines )2(_nearproc)0( to be a procedure taking two arguments, the first \()2(i)0(\) an integer and the second \()2(j)0(\))]653 norm1 -[(a pointer to an integer. It returns )2(i + *j)0(.)]642 norm2 -[(Note that the )2(arg)0( macro has an )2(EQU)0( as the first line of its expansion, and since the label before the macro call)]625 norm1 -[(gets prepended to the first line of the expanded macro, the )2(EQU)0( works, defining )2(%$i)0( to be an offset from )2(BP)0(.)]614 norm0 -[(A context-local variable is used, local to the context pushed by the )2(proc)0( macro and popped by the)]603 norm0 -[2(endproc)0( macro, so that the same argument name can be used in later procedures. Of course, you don't )1(have)]592 norm0 -[(to do that.)]581 norm2 -[(The macro set produces code for near functions \(tiny, small and compact-model code\) by default. You can)]564 norm1 -[(have it generate far functions \(medium, large and huge-model code\) by means of coding)]553 norm0 -[2(%define FARCODE)0(. This changes the kind of return instruction generated by )2(endproc)0(, and also changes)]542 norm0 -[(the starting point for the argument offsets. The macro set contains no intrinsic dependency on whether data)]531 norm0 -[(pointers are far or not.)]520 norm2 -[2(arg)0( can take an optional parameter, giving the size of the argument. If no size is given, 2 is assumed, since it)]503 norm1 -[(is likely that many function parameters will be of type )2(int)0(.)]492 norm2 -[(The large-model equivalent of the above function would look like this:)]475 norm3 -[2(%define FARCODE )]458 code1 -[2()]447 code0 -[2(proc _farproc )]436 code0 -[2()]425 code0 -[2(%$i arg )]414 code0 -[2(%$j arg 4 )]403 code0 -[2( mov ax,[bp + %$i] )]392 code0 -[2( mov bx,[bp + %$j] )]381 code0 -[2( mov es,[bp + %$j + 2] )]370 code0 -[2( add ax,[bx] )]359 code0 -[2()]348 code0 -[2(endproc)]337 code2 -[(This makes use of the argument to the )2(arg)0( macro to define a parameter of size 4, because )2(j)0( is now a far)]320 norm1 -[(pointer. When we load from )2(j)0(, we must load a segment and an offset.)]309 norm2 -[{/section-8.5 xa}(Interfacing to )(Borland Pascal Programs)](8.5)287.6 head3 -[(Interfacing to Borland Pascal programs is similar in concept to interfacing to 16-bit C programs. The)]270.6 norm1 -[(differences are:)]259.6 norm2 -[(The leading underscore required for interfacing to C programs is not required for Pascal.)]242.6 bull3 -[(The memory model is always large: functions are far, data pointers are far, and no data item can be more)]225.6 bull1 -[(than 64K long. \(Actually, some functions are near, but only those functions that are local to a Pascal unit)]214.6 bull0 -[(and never called from outside it. All assembly functions that Pascal calls, and all Pascal functions that)]203.6 bull0 -[(assembly routines are able to call, are far.\) However, all static data declared in a Pascal program goes into)]192.6 bull0 -[(the default data segment, which is the one whose segment address will be in )2(DS)0( when control is passed to)]181.6 bull0 -[(your assembly code. The only things that do not live in the default data segment are local variables \(they)]170.6 bull0 -[(live in the stack segment\) and dynamically allocated variables. All data )1(pointers)0(, however, are far.)]159.6 bull2 -[(The function calling convention is different \226 described below.)]142.6 bull3 -[(Some data types, such as strings, are stored differently.)]125.6 bull3 -(99)pageodd -restore showpage -%%Page: 100 100 -%%BeginPageSetup -save -%%EndPageSetup -/100 pa -[(There are restrictions on the segment names you are allowed to use \226 Borland Pascal will ignore code or)]681 bull1 -[(data declared in a segment it doesn't like the name of. The restrictions are described below.)]670 bull2 -[{/section-8.5.1 xa}(The Pascal Calling Convention)](8.5.1)650.8 subh3 -[(The 16-bit Pascal calling convention is as follows. In the following description, the words )1(caller)0( and )1(callee)]633.8 norm1 -[(are used to denote the function doing the calling and the function which gets called.)]622.8 norm2 -[(The caller pushes the function's parameters on the stack, one after another, in normal order \(left to right, so)]605.8 bull1 -[(that the first argument specified to the function is pushed first\).)]594.8 bull2 -[(The caller then executes a far )2(CALL)0( instruction to pass control to the callee.)]577.8 bull3 -[(The callee receives control, and typically \(although this is not actually necessary, in functions which do not)]560.8 bull1 -[(need to access their parameters\) starts by saving the value of )2(SP)0( in )2(BP)0( so as to be able to use )2(BP)0( as a base)]549.8 bull0 -[(pointer to find its parameters on the stack. However, the caller was probably doing this too, so part of the)]538.8 bull0 -[(calling convention states that )2(BP)0( must be preserved by any function. Hence the callee, if it is going to set)]527.8 bull0 -[(up )2(BP)0( as a )(frame pointer, must push the previous value first.)]516.8 bull2 -[(The callee may then access its parameters relative to )2(BP)0(. The word at )2([BP])0( holds the previous value of )2(BP)]499.8 bull1 -[(as it was pushed. The next word, at )2([BP+2])0(, holds the offset part of the return address, and the next one at)]488.8 bull0 -[2([BP+4])0( the segment part. The parameters begin at )2([BP+6])0(. The rightmost parameter of the function,)]477.8 bull0 -[(since it was pushed last, is accessible at this offset from )2(BP)0(; the others follow, at successively greater)]466.8 bull0 -[(offsets.)]455.8 bull2 -[(The callee may also wish to decrease )2(SP)0( further, so as to allocate space on the stack for local variables,)]438.8 bull1 -[(which will then be accessible at negative offsets from )2(BP)0(.)]427.8 bull2 -[(The callee, if it wishes to return a value to the caller, should leave the value in )2(AL)0(, )2(AX)0( or )2(DX:AX)]410.8 bull1 -[(depending on the size of the value. Floating-point results are returned in )2(ST0)0(. Results of type )2(Real)]399.8 bull0 -[(\(Borland's own custom floating-point data type, not handled directly by the FPU\) are returned in)]388.8 bull0 -[2(DX:BX:AX)0(. To return a result of type )2(String)0(, the caller pushes a pointer to a temporary string before)]377.8 bull0 -[(pushing the parameters, and the callee places the returned string value at that location. The pointer is not a)]366.8 bull0 -[(parameter, and should not be removed from the stack by the )2(RETF)0( instruction.)]355.8 bull2 -[(Once the callee has finished processing, it restores )2(SP)0( from )2(BP)0( if it had allocated local stack space, then)]338.8 bull1 -[(pops the previous value of )2(BP)0(, and returns via )2(RETF)0(. It uses the form of )2(RETF)0( with an immediate)]327.8 bull0 -[(parameter, giving the number of bytes taken up by the parameters on the stack. This causes the parameters)]316.8 bull0 -[(to be removed from the stack as a side effect of the return instruction.)]305.8 bull2 -[(When the caller regains control from the callee, the function parameters have already been removed from)]288.8 bull1 -[(the stack, so it needs to do nothing further.)]277.8 bull2 -[(Thus, you would define a function in Pascal style, taking two )2(Integer)0(\226type parameters, in the following)]260.8 norm1 -[(way:)]249.8 norm2 -[2(global myfunc )]232.8 code1 -[2()]221.8 code0 -[2(myfunc: push bp )]210.8 code0 -[2( mov bp,sp )]199.8 code0 -[2( sub sp,0x40 ; 64 bytes of local stack space )]188.8 code0 -[2( mov bx,[bp+8] ; first parameter to function )]177.8 code0 -[2( mov bx,[bp+6] ; second parameter to function )]166.8 code0 -[2()]155.8 code0 -[2( ; some more code )]144.8 code0 -[2()]133.8 code0 -[2( mov sp,bp ; undo "sub sp,0x40" above )]122.8 code0 -[2( pop bp )]111.8 code0 -[2( retf 4 ; total size of params is 4)]100.8 code2 -(100)pageeven -restore showpage -%%Page: 101 101 -%%BeginPageSetup -save -%%EndPageSetup -/101 pa -[(At the other end of the process, to call a Pascal function from your assembly code, you would do something)]681 norm1 -[(like this:)]670 norm2 -[2(extern SomeFunc )]653 code1 -[2()]642 code0 -[2( ; and then, further down... )]631 code0 -[2()]620 code0 -[2( push word seg mystring ; Now push the segment, and... )]609 code0 -[2( push word mystring ; ... offset of "mystring" )]598 code0 -[2( push word [myint] ; one of my variables )]587 code0 -[2( call far SomeFunc)]576 code2 -[(This is equivalent to the Pascal code)]559 norm3 -[2(procedure SomeFunc\(String: PChar; Int: Integer\); )]542 code1 -[2( SomeFunc\(@mystring, myint\);)]531 code2 -[{/section-8.5.2 xa}(Borland Pascal )(Segment Name Restrictions)](8.5.2)511.8 subh3 -[(Since Borland Pascal's internal unit file format is completely different from )2(OBJ)0(, it only makes a very)]494.8 norm1 -[(sketchy job of actually reading and understanding the various information contained in a real )2(OBJ)0( file when it)]483.8 norm0 -[(links that in. Therefore an object file intended to be linked to a Pascal program must obey a number of)]472.8 norm0 -[(restrictions:)]461.8 norm2 -[(Procedures and functions must be in a segment whose name is either )2(CODE)0(, )2(CSEG)0(, or something ending in)]444.8 bull1 -[2(_TEXT)0(.)]433.8 bull2 -[(initialized data must be in a segment whose name is either )2(CONST)0( or something ending in )2(_DATA)0(.)]416.8 bull3 -[(Uninitialized data must be in a segment whose name is either )2(DATA)0(, )2(DSEG)0(, or something ending in )2(_BSS)0(.)]399.8 bull3 -[(Any other segments in the object file are completely ignored. )2(GROUP)0( directives and segment attributes are)]382.8 bull1 -[(also ignored.)]371.8 bull2 -[{/section-8.5.3 xa}(Using )2(c16.mac)0( With Pascal Programs)](8.5.3)352.6 subh3 -[(The )2(c16.mac)0( macro package, described in ){/section-8.4.5 xl}(section 8.4.5){el}(, can also be used to simplify writing functions to be)]335.6 norm1 -[(called from Pascal programs, if you code )2(%define PASCAL)0(. This definition ensures that functions are far)]324.6 norm0 -[(\(it implies )2(FARCODE)0(\), and also causes procedure return instructions to be generated with an operand.)]313.6 norm2 -[(Defining )2(PASCAL)0( does not change the code which calculates the argument offsets; you must declare your)]296.6 norm1 -[(function's arguments in reverse order. For example:)]285.6 norm2 -[2(%define PASCAL )]268.6 code1 -[2()]257.6 code0 -[2(proc _pascalproc )]246.6 code0 -[2()]235.6 code0 -[2(%$j arg 4 )]224.6 code0 -[2(%$i arg )]213.6 code0 -[2( mov ax,[bp + %$i] )]202.6 code0 -[2( mov bx,[bp + %$j] )]191.6 code0 -[2( mov es,[bp + %$j + 2] )]180.6 code0 -[2( add ax,[bx] )]169.6 code0 -[2()]158.6 code0 -[2(endproc)]147.6 code2 -[(This defines the same routine, conceptually, as the example in ){/section-8.4.5 xl}(section 8.4.5){el}(: it defines a function taking two)]130.6 norm1 -[(arguments, an integer and a pointer to an integer, which returns the sum of the integer and the contents of the)]119.6 norm0 -(101)pageodd -restore showpage -%%Page: 102 102 -%%BeginPageSetup -save -%%EndPageSetup -/102 pa -[(pointer. The only difference between this code and the large-model C version is that )2(PASCAL)0( is defined)]681 norm0 -[(instead of )2(FARCODE)0(, and that the arguments are declared in reverse order.)]670 norm2 -(102)pageeven -restore showpage -%%Page: 103 103 -%%BeginPageSetup -save -%%EndPageSetup -/103 pa -[{/chapter-9 xa}(Chapter 9: Writing 32-bit Code \(Unix, Win32, DJGPP\))]642.8 chap3 -[(This chapter attempts to cover some of the common issues involved when writing 32-bit code, to run under)]607.8 norm1 -[(Win32 or Unix, or to be linked with C code generated by a Unix-style C compiler such as )(DJGPP. It covers)]596.8 norm0 -[(how to write assembly code to interface with 32-bit C routines, and how to write position-independent code)]585.8 norm0 -[(for shared libraries.)]574.8 norm2 -[(Almost all 32-bit code, and in particular all code running under )2(Win32)0(, )2(DJGPP)0( or any of the PC Unix)]557.8 norm1 -[(variants, runs in )1(flat)0( memory model. This means that the segment registers and paging have already been set)]546.8 norm0 -[(up to give you the same 32-bit 4Gb address space no matter what segment you work relative to, and that you)]535.8 norm0 -[(should ignore all segment registers completely. When writing flat-model application code, you never need to)]524.8 norm0 -[(use a segment override or modify any segment register, and the code-section addresses you pass to )2(CALL)0( and)]513.8 norm0 -[2(JMP)0( live in the same address space as the data-section addresses you access your variables by and the)]502.8 norm0 -[(stack-section addresses you access local variables and procedure parameters by. Every address is 32 bits long)]491.8 norm0 -[(and contains only an offset part.)]480.8 norm2 -[{/section-9.1 xa}(Interfacing to 32-bit C Programs)](9.1)459.4 head3 -[(A lot of the discussion in ){/section-8.4 xl}(section 8.4){el}(, about interfacing to 16-bit C programs, still applies when working in 32)]442.4 norm1 -[(bits. The absence of memory models or segmentation worries simplifies things a lot.)]431.4 norm2 -[{/section-9.1.1 xa}(External Symbol Names)](9.1.1)412.2 subh3 -[(Most 32-bit C compilers share the convention used by 16-bit compilers, that the names of all global symbols)]395.2 norm1 -[(\(functions or data\) they define are formed by prefixing an underscore to the name as it appears in the C)]384.2 norm0 -[(program. However, not all of them do: the )2(ELF)0( specification states that C symbols do )1(not)0( have a leading)]373.2 norm0 -[(underscore on their assembly-language names.)]362.2 norm2 -[(The older Linux )2(a.out)0( C compiler, all )2(Win32)0( compilers, )2(DJGPP)0(, and )2(NetBSD)0( and )2(FreeBSD)0(, all use the)]345.2 norm1 -[(leading underscore; for these compilers, the macros )2(cextern)0( and )2(cglobal)0(, as given in ){/section-8.4.1 xl}(section 8.4.1){el}(, will)]334.2 norm0 -[(still work. For )2(ELF)0(, though, the leading underscore should not be used.)]323.2 norm2 -[(See also ){/section-2.1.27 xl}(section 2.1.27){el}(.)]306.2 norm3 -[{/section-9.1.2 xa}(Function Definitions and Function Calls)](9.1.2)287 subh3 -[(The )(C calling convention in 32-bit programs is as follows. In the following description, the words )1(caller)0( and)]270 norm1 -[1(callee)0( are used to denote the function doing the calling and the function which gets called.)]259 norm2 -[(The caller pushes the function's parameters on the stack, one after another, in reverse order \(right to left, so)]242 bull1 -[(that the first argument specified to the function is pushed last\).)]231 bull2 -[(The caller then executes a near )2(CALL)0( instruction to pass control to the callee.)]214 bull3 -[(The callee receives control, and typically \(although this is not actually necessary, in functions which do not)]197 bull1 -[(need to access their parameters\) starts by saving the value of )2(ESP)0( in )2(EBP)0( so as to be able to use )2(EBP)0( as a)]186 bull0 -[(base pointer to find its parameters on the stack. However, the caller was probably doing this too, so part of)]175 bull0 -[(the calling convention states that )2(EBP)0( must be preserved by any C function. Hence the callee, if it is going)]164 bull0 -[(to set up )2(EBP)0( as a )(frame pointer, must push the previous value first.)]153 bull2 -[(The callee may then access its parameters relative to )2(EBP)0(. The doubleword at )2([EBP])0( holds the previous)]136 bull1 -[(value of )2(EBP)0( as it was pushed; the next doubleword, at )2([EBP+4])0(, holds the return address, pushed)]125 bull0 -[(implicitly by )2(CALL)0(. The parameters start after that, at )2([EBP+8])0(. The leftmost parameter of the function,)]114 bull0 -[(since it was pushed last, is accessible at this offset from )2(EBP)0(; the others follow, at successively greater)]103 bull0 -(103)pageodd -restore showpage -%%Page: 104 104 -%%BeginPageSetup -save -%%EndPageSetup -/104 pa -[(offsets. Thus, in a function such as )2(printf)0( which takes a variable number of parameters, the pushing of)]681 bull0 -[(the parameters in reverse order means that the function knows where to find its first parameter, which tells)]670 bull0 -[(it the number and type of the remaining ones.)]659 bull2 -[(The callee may also wish to decrease )2(ESP)0( further, so as to allocate space on the stack for local variables,)]642 bull1 -[(which will then be accessible at negative offsets from )2(EBP)0(.)]631 bull2 -[(The callee, if it wishes to return a value to the caller, should leave the value in )2(AL)0(, )2(AX)0( or )2(EAX)0( depending)]614 bull1 -[(on the size of the value. Floating-point results are typically returned in )2(ST0)0(.)]603 bull2 -[(Once the callee has finished processing, it restores )2(ESP)0( from )2(EBP)0( if it had allocated local stack space, then)]586 bull1 -[(pops the previous value of )2(EBP)0(, and returns via )2(RET)0( \(equivalently, )2(RETN)0(\).)]575 bull2 -[(When the caller regains control from the callee, the function parameters are still on the stack, so it typically)]558 bull1 -[(adds an immediate constant to )2(ESP)0( to remove them \(instead of executing a number of slow )2(POP)]547 bull0 -[(instructions\). Thus, if a function is accidentally called with the wrong number of parameters due to a)]536 bull0 -[(prototype mismatch, the stack will still be returned to a sensible state since the caller, which )1(knows)0( how)]525 bull0 -[(many parameters it pushed, does the removing.)]514 bull2 -[(There is an alternative calling convention used by Win32 programs for Windows API calls, and also for)]497 norm1 -[(functions called )1(by)0( the Windows API such as window procedures: they follow what Microsoft calls the)]486 norm0 -[2(__stdcall)0( convention. This is slightly closer to the Pascal convention, in that the callee clears the stack by)]475 norm0 -[(passing a parameter to the )2(RET)0( instruction. However, the parameters are still pushed in right-to-left order.)]464 norm2 -[(Thus, you would define a function in C style in the following way:)]447 norm3 -[2(global _myfunc )]430 code1 -[2()]419 code0 -[2(_myfunc: )]408 code0 -[2( push ebp )]397 code0 -[2( mov ebp,esp )]386 code0 -[2( sub esp,0x40 ; 64 bytes of local stack space )]375 code0 -[2( mov ebx,[ebp+8] ; first parameter to function )]364 code0 -[2()]353 code0 -[2( ; some more code )]342 code0 -[2()]331 code0 -[2( leave ; mov esp,ebp / pop ebp )]320 code0 -[2( ret)]309 code2 -[(At the other end of the process, to call a C function from your assembly code, you would do something like)]292 norm1 -[(this:)]281 norm2 -[2(extern _printf )]264 code1 -[2()]253 code0 -[2( ; and then, further down... )]242 code0 -[2()]231 code0 -[2( push dword [myint] ; one of my integer variables )]220 code0 -[2( push dword mystring ; pointer into my data segment )]209 code0 -[2( call _printf )]198 code0 -[2( add esp,byte 8 ; `byte' saves space )]187 code0 -[2()]176 code0 -[2( ; then those data items... )]165 code0 -[2()]154 code0 -[2(segment _DATA )]143 code0 -[2()]132 code0 -[2(myint dd 1234 )]121 code0 -[2(mystring db 'This number -> %d <- should be 1234',10,0)]110 code2 -(104)pageeven -restore showpage -%%Page: 105 105 -%%BeginPageSetup -save -%%EndPageSetup -/105 pa -[(This piece of code is the assembly equivalent of the C code)]681 norm3 -[2( int myint = 1234; )]664 code1 -[2( printf\("This number -> %d <- should be 1234\\n", myint\);)]653 code2 -[{/section-9.1.3 xa}(Accessing Data Items)](9.1.3)633.8 subh3 -[(To get at the contents of C variables, or to declare variables which C can access, you need only declare the)]616.8 norm1 -[(names as )2(GLOBAL)0( or )2(EXTERN)0(. \(Again, the names require leading underscores, as stated in ){/section-9.1.1 xl}(section 9.1.1){el}(.\))]605.8 norm0 -[(Thus, a C variable declared as )2(int i)0( can be accessed from assembler as)]594.8 norm2 -[2( extern _i )]577.8 code1 -[2( mov eax,[_i])]566.8 code2 -[(And to declare your own integer variable which C programs can access as )2(extern int j)0(, you do this)]549.8 norm1 -[(\(making sure you are assembling in the )2(_DATA)0( segment, if necessary\):)]538.8 norm2 -[2( global _j )]521.8 code1 -[2(_j dd 0)]510.8 code2 -[(To access a C array, you need to know the size of the components of the array. For example, )2(int)0( variables)]493.8 norm1 -[(are four bytes long, so if a C program declares an array as )2(int a[10])0(, you can access )2(a[3])0( by coding)]482.8 norm0 -[2(mov ax,[_a+12])0(. \(The byte offset 12 is obtained by multiplying the desired array index, 3, by the size of)]471.8 norm0 -[(the array element, 4.\) The sizes of the C base types in 32-bit compilers are: 1 for )2(char)0(, 2 for )2(short)0(, 4 for)]460.8 norm0 -[2(int)0(, )2(long)0( and )2(float)0(, and 8 for )2(double)0(. Pointers, being 32-bit addresses, are also 4 bytes long.)]449.8 norm2 -[(To access a C )(data structure, you need to know the offset from the base of the structure to the field you are)]432.8 norm1 -[(interested in. You can either do this by converting the C structure definition into a NASM structure definition)]421.8 norm0 -[(\(using )2(STRUC)0(\), or by calculating the one offset and using just that.)]410.8 norm2 -[(To do either of these, you should read your C compiler's manual to find out how it organizes data structures.)]393.8 norm1 -[(NASM gives no special alignment to structure members in its own )2(STRUC)0( macro, so you have to specify)]382.8 norm0 -[(alignment yourself if the C compiler generates it. Typically, you might find that a structure like)]371.8 norm2 -[2(struct { )]354.8 code1 -[2( char c; )]343.8 code0 -[2( int i; )]332.8 code0 -[2(} foo;)]321.8 code2 -[(might be eight bytes long rather than five, since the )2(int)0( field would be aligned to a four-byte boundary.)]304.8 norm1 -[(However, this sort of feature is sometimes a configurable option in the C compiler, either using)]293.8 norm0 -[(command-line options or )2(#pragma)0( lines, so you have to find out how your own compiler does it.)]282.8 norm2 -[{/section-9.1.4 xa}2(c32.mac)0(: Helper Macros for the 32-bit C Interface)](9.1.4)263.6 subh3 -[(Included in the NASM archives, in the )2(misc)0( directory, is a file )2(c32.mac)0( of macros. It defines three macros:)]246.6 norm1 -[2(proc)0(, )2(arg)0( and )2(endproc)0(. These are intended to be used for C-style procedure definitions, and they)]235.6 norm0 -[(automate a lot of the work involved in keeping track of the calling convention.)]224.6 norm2 -[(An example of an assembly function using the macro set is given here:)]207.6 norm3 -[2(proc _proc32 )]190.6 code1 -[2()]179.6 code0 -[2(%$i arg )]168.6 code0 -[2(%$j arg )]157.6 code0 -[2( mov eax,[ebp + %$i] )]146.6 code0 -[2( mov ebx,[ebp + %$j] )]135.6 code0 -[2( add eax,[ebx] )]124.6 code0 -[2()]113.6 code0 -[2(endproc)]102.6 code2 -(105)pageodd -restore showpage -%%Page: 106 106 -%%BeginPageSetup -save -%%EndPageSetup -/106 pa -[(This defines )2(_proc32)0( to be a procedure taking two arguments, the first \()2(i)0(\) an integer and the second \()2(j)0(\) a)]681 norm1 -[(pointer to an integer. It returns )2(i + *j)0(.)]670 norm2 -[(Note that the )2(arg)0( macro has an )2(EQU)0( as the first line of its expansion, and since the label before the macro call)]653 norm1 -[(gets prepended to the first line of the expanded macro, the )2(EQU)0( works, defining )2(%$i)0( to be an offset from )2(BP)0(.)]642 norm0 -[(A context-local variable is used, local to the context pushed by the )2(proc)0( macro and popped by the)]631 norm0 -[2(endproc)0( macro, so that the same argument name can be used in later procedures. Of course, you don't )1(have)]620 norm0 -[(to do that.)]609 norm2 -[2(arg)0( can take an optional parameter, giving the size of the argument. If no size is given, 4 is assumed, since it)]592 norm1 -[(is likely that many function parameters will be of type )2(int)0( or pointers.)]581 norm2 -[{/section-9.2 xa}(Writing NetBSD/FreeBSD/OpenBSD and Linux/ELF )(Shared Libraries)](9.2)559.6 head3 -[2(ELF)0( replaced the older )2(a.out)0( object file format under Linux because it contains support for)]542.6 norm1 -[(position-independent code \()(PIC\), which makes writing shared libraries much easier. NASM supports the )2(ELF)]531.6 norm0 -[(position-independent code features, so you can write Linux )2(ELF)0( shared libraries in NASM.)]520.6 norm2 -[(NetBSD, and its close cousins )(FreeBSD and )(OpenBSD, take a different approach by hacking PIC support into)]503.6 norm1 -[(the )2(a.out)0( format. NASM supports this as the )2(aoutb)0( output format, so you can write )(BSD shared libraries)]492.6 norm0 -[(in NASM too.)]481.6 norm2 -[(The operating system loads a PIC shared library by memory-mapping the library file at an arbitrarily chosen)]464.6 norm1 -[(point in the address space of the running process. The contents of the library's code section must therefore not)]453.6 norm0 -[(depend on where it is loaded in memory.)]442.6 norm2 -[(Therefore, you cannot get at your variables by writing code like this:)]425.6 norm3 -[2( mov eax,[myvar] ; WRONG)]408.6 code3 -[(Instead, the linker provides an area of memory called the )1(global offset table)0(, or )(GOT; the GOT is situated at a)]391.6 norm1 -[(constant distance from your library's code, so if you can find out where your library is loaded \(which is)]380.6 norm0 -[(typically done using a )2(CALL)0( and )2(POP)0( combination\), you can obtain the address of the GOT, and you can then)]369.6 norm0 -[(load the addresses of your variables out of linker-generated entries in the GOT.)]358.6 norm2 -[(The )1(data)0( section of a PIC shared library does not have these restrictions: since the data section is writable, it)]341.6 norm1 -[(has to be copied into memory anyway rather than just paged in from the library file, so as long as it's being)]330.6 norm0 -[(copied it can be relocated too. So you can put ordinary types of relocation in the data section without too)]319.6 norm0 -[(much worry \(but see ){/section-9.2.4 xl}(section 9.2.4){el}( for a caveat\).)]308.6 norm2 -[{/section-9.2.1 xa}(Obtaining the Address of the GOT)](9.2.1)289.4 subh3 -[(Each code module in your shared library should define the GOT as an external symbol:)]272.4 norm3 -[2(extern _GLOBAL_OFFSET_TABLE_ ; in ELF )]255.4 code1 -[2(extern __GLOBAL_OFFSET_TABLE_ ; in BSD a.out)]244.4 code2 -[(At the beginning of any function in your shared library which plans to access your data or BSS sections, you)]227.4 norm1 -[(must first calculate the address of the GOT. This is typically done by writing the function in this form:)]216.4 norm2 -[2(func: push ebp )]199.4 code1 -[2( mov ebp,esp )]188.4 code0 -[2( push ebx )]177.4 code0 -[2( call .get_GOT )]166.4 code0 -[2(.get_GOT: )]155.4 code0 -[2( pop ebx )]144.4 code0 -[2( add ebx,_GLOBAL_OFFSET_TABLE_+$$-.get_GOT wrt ..gotpc )]133.4 code0 -[2()]122.4 code0 -[2( ; the function body comes here )]111.4 code0 -[2()]100.4 code0 -(106)pageeven -restore showpage -%%Page: 107 107 -%%BeginPageSetup -save -%%EndPageSetup -/107 pa -[2( mov ebx,[ebp-4] )]681 code0 -[2( mov esp,ebp )]670 code0 -[2( pop ebp )]659 code0 -[2( ret)]648 code2 -[(\(For BSD, again, the symbol )2(_GLOBAL_OFFSET_TABLE)0( requires a second leading underscore.\))]631 norm3 -[(The first two lines of this function are simply the standard C prologue to set up a stack frame, and the last)]614 norm1 -[(three lines are standard C function epilogue. The third line, and the fourth to last line, save and restore the)]603 norm0 -[2(EBX)0( register, because PIC shared libraries use this register to store the address of the GOT.)]592 norm2 -[(The interesting bit is the )2(CALL)0( instruction and the following two lines. The )2(CALL)0( and )2(POP)0( combination)]575 norm1 -[(obtains the address of the label )2(.get_GOT)0(, without having to know in advance where the program was)]564 norm0 -[(loaded \(since the )2(CALL)0( instruction is encoded relative to the current position\). The )2(ADD)0( instruction makes use)]553 norm0 -[(of one of the special PIC relocation types: )(GOTPC relocation. With the )2(WRT ..gotpc)0( qualifier specified,)]542 norm0 -[(the symbol referenced \(here )2(_GLOBAL_OFFSET_TABLE_)0(, the special symbol assigned to the GOT\) is given)]531 norm0 -[(as an offset from the beginning of the section. \(Actually, )2(ELF)0( encodes it as the offset from the operand field)]520 norm0 -[(of the )2(ADD)0( instruction, but NASM simplifies this deliberately, so you do things the same way for both )2(ELF)]509 norm0 -[(and )2(BSD)0(.\) So the instruction then )1(adds)0( the beginning of the section, to get the real address of the GOT, and)]498 norm0 -[(subtracts the value of )2(.get_GOT)0( which it knows is in )2(EBX)0(. Therefore, by the time that instruction has)]487 norm0 -[(finished, )2(EBX)0( contains the address of the GOT.)]476 norm2 -[(If you didn't follow that, don't worry: it's never necessary to obtain the address of the GOT by any other)]459 norm1 -[(means, so you can put those three instructions into a macro and safely ignore them:)]448 norm2 -[2(%macro get_GOT 0 )]431 code1 -[2()]420 code0 -[2( call %%getgot )]409 code0 -[2( %%getgot: )]398 code0 -[2( pop ebx )]387 code0 -[2( add ebx,_GLOBAL_OFFSET_TABLE_+$$-%%getgot wrt ..gotpc )]376 code0 -[2()]365 code0 -[2(%endmacro)]354 code2 -[{/section-9.2.2 xa}(Finding Your Local Data Items)](9.2.2)334.8 subh3 -[(Having got the GOT, you can then use it to obtain the addresses of your data items. Most variables will reside)]317.8 norm1 -[(in the sections you have declared; they can be accessed using the )2(..gotoff)0( special )2(WRT)0( type. The way this)]306.8 norm0 -[(works is like this:)]295.8 norm2 -[2( lea eax,[ebx+myvar wrt ..gotoff])]278.8 code3 -[(The expression )2(myvar wrt ..gotoff)0( is calculated, when the shared library is linked, to be the offset to)]261.8 norm1 -[(the local variable )2(myvar)0( from the beginning of the GOT. Therefore, adding it to )2(EBX)0( as above will place the)]250.8 norm0 -[(real address of )2(myvar)0( in )2(EAX)0(.)]239.8 norm2 -[(If you declare variables as )2(GLOBAL)0( without specifying a size for them, they are shared between code modules)]222.8 norm1 -[(in the library, but do not get exported from the library to the program that loaded it. They will still be in your)]211.8 norm0 -[(ordinary data and BSS sections, so you can access them in the same way as local variables, using the above)]200.8 norm0 -[2(..gotoff)0( mechanism.)]189.8 norm2 -[(Note that due to a peculiarity of the way BSD )2(a.out)0( format handles this relocation type, there must be at)]172.8 norm1 -[(least one non-local symbol in the same section as the address you're trying to access.)]161.8 norm2 -[{/section-9.2.3 xa}(Finding External and Common Data Items)](9.2.3)142.6 subh3 -[(If your library needs to get at an external variable \(external to the )1(library)0(, not just to one of the modules)]125.6 norm1 -[(within it\), you must use the )2(..got)0( type to get at it. The )2(..got)0( type, instead of giving you the offset from)]114.6 norm0 -[(the GOT base to the variable, gives you the offset from the GOT base to a GOT )1(entry)0( containing the address)]103.6 norm0 -(107)pageodd -restore showpage -%%Page: 108 108 -%%BeginPageSetup -save -%%EndPageSetup -/108 pa -[(of the variable. The linker will set up this GOT entry when it builds the library, and the dynamic linker will)]681 norm0 -[(place the correct address in it at load time. So to obtain the address of an external variable )2(extvar)0( in )2(EAX)0(,)]670 norm0 -[(you would code)]659 norm2 -[2( mov eax,[ebx+extvar wrt ..got])]642 code3 -[(This loads the address of )2(extvar)0( out of an entry in the GOT. The linker, when it builds the shared library,)]625 norm1 -[(collects together every relocation of type )2(..got)0(, and builds the GOT so as to ensure it has every necessary)]614 norm0 -[(entry present.)]603 norm2 -[(Common variables must also be accessed in this way.)]586 norm3 -[{/section-9.2.4 xa}(Exporting Symbols to the Library User)](9.2.4)566.8 subh3 -[(If you want to export symbols to the user of the library, you have to declare whether they are functions or)]549.8 norm1 -[(data, and if they are data, you have to give the size of the data item. This is because the dynamic linker has to)]538.8 norm0 -[(build )(procedure linkage table entries for any exported functions, and also moves exported data items away)]527.8 norm0 -[(from the library's data section in which they were declared.)]516.8 norm2 -[(So to export a function to users of the library, you must use)]499.8 norm3 -[2(global func:function ; declare it as a function )]482.8 code1 -[2()]471.8 code0 -[2(func: push ebp )]460.8 code0 -[2()]449.8 code0 -[2( ; etc.)]438.8 code2 -[(And to export a data item such as an array, you would have to code)]421.8 norm3 -[2(global array:data array.end-array ; give the size too )]404.8 code1 -[2()]393.8 code0 -[2(array: resd 128 )]382.8 code0 -[2(.end:)]371.8 code2 -[(Be careful: If you export a variable to the library user, by declaring it as )2(GLOBAL)0( and supplying a size, the)]354.8 norm1 -[(variable will end up living in the data section of the main program, rather than in your library's data section,)]343.8 norm0 -[(where you declared it. So you will have to access your own global variable with the )2(..got)0( mechanism rather)]332.8 norm0 -[(than )2(..gotoff)0(, as if it were external \(which, effectively, it has become\).)]321.8 norm2 -[(Equally, if you need to store the address of an exported global in one of your data sections, you can't do it by)]304.8 norm1 -[(means of the standard sort of code:)]293.8 norm2 -[2(dataptr: dd global_data_item ; WRONG)]276.8 code3 -[(NASM will interpret this code as an ordinary relocation, in which )2(global_data_item)0( is merely an offset)]259.8 norm1 -[(from the beginning of the )2(.data)0( section \(or whatever\); so this reference will end up pointing at your data)]248.8 norm0 -[(section instead of at the exported global which resides elsewhere.)]237.8 norm2 -[(Instead of the above code, then, you must write)]220.8 norm3 -[2(dataptr: dd global_data_item wrt ..sym)]203.8 code3 -[(which makes use of the special )2(WRT)0( type )2(..sym)0( to instruct NASM to search the symbol table for a particular)]186.8 norm1 -[(symbol at that address, rather than just relocating by section base.)]175.8 norm2 -[(Either method will work for functions: referring to one of your functions by means of)]158.8 norm3 -[2(funcptr: dd my_function)]141.8 code3 -[(will give the user the address of the code you wrote, whereas)]124.8 norm3 -[2(funcptr: dd my_function wrt .sym)]107.8 code3 -(108)pageeven -restore showpage -%%Page: 109 109 -%%BeginPageSetup -save -%%EndPageSetup -/109 pa -[(will give the address of the procedure linkage table for the function, which is where the calling program will)]681 norm1 -[1(believe)0( the function lives. Either address is a valid way to call the function.)]670 norm2 -[{/section-9.2.5 xa}(Calling Procedures Outside the Library)](9.2.5)650.8 subh3 -[(Calling procedures outside your shared library has to be done by means of a )1(procedure linkage table)0(, or )(PLT.)]633.8 norm1 -[(The PLT is placed at a known offset from where the library is loaded, so the library code can make calls to the)]622.8 norm0 -[(PLT in a position-independent way. Within the PLT there is code to jump to offsets contained in the GOT, so)]611.8 norm0 -[(function calls to other shared libraries or to routines in the main program can be transparently passed off to)]600.8 norm0 -[(their real destinations.)]589.8 norm2 -[(To call an external routine, you must use another special PIC relocation type, )2(WRT ..plt)0(. This is much)]572.8 norm1 -[(easier than the GOT-based ones: you simply replace calls such as )2(CALL printf)0( with the PLT-relative)]561.8 norm0 -[(version )2(CALL printf WRT ..plt)0(.)]550.8 norm2 -[{/section-9.2.6 xa}(Generating the Library File)](9.2.6)531.6 subh3 -[(Having written some code modules and assembled them to )2(.o)0( files, you then generate your shared library)]514.6 norm1 -[(with a command such as)]503.6 norm2 -[2(ld -shared -o library.so module1.o module2.o # for ELF )]486.6 code1 -[2(ld -Bshareable -o library.so module1.o module2.o # for BSD)]475.6 code2 -[(For ELF, if your shared library is going to reside in system directories such as )2(/usr/lib)0( or )2(/lib)0(, it is)]458.6 norm1 -[(usually worth using the )2(-soname)0( flag to the linker, to store the final library file name, with a version)]447.6 norm0 -[(number, into the library:)]436.6 norm2 -[2(ld -shared -soname library.so.1 -o library.so.1.2 *.o)]419.6 code3 -[(You would then copy )2(library.so.1.2)0( into the library directory, and create )2(library.so.1)0( as a)]402.6 norm1 -[(symbolic link to it.)]391.6 norm2 -(109)pageodd -restore showpage -%%Page: 110 110 -%%BeginPageSetup -save -%%EndPageSetup -/110 pa -[{/chapter-10 xa}(Chapter 10: Mixing 16 and 32 Bit Code)]642.8 chap3 -[(This chapter tries to cover some of the issues, largely related to unusual forms of addressing and jump)]607.8 norm1 -[(instructions, encountered when writing operating system code such as protected-mode initialisation routines,)]596.8 norm0 -[(which require code that operates in mixed segment sizes, such as code in a 16-bit segment trying to modify)]585.8 norm0 -[(data in a 32-bit one, or jumps between different-size segments.)]574.8 norm2 -[{/section-10.1 xa}(Mixed-Size Jumps)](10.1)553.4 head3 -[(The most common form of )(mixed-size instruction is the one used when writing a 32-bit OS: having done)]536.4 norm1 -[(your setup in 16-bit mode, such as loading the kernel, you then have to boot it by switching into protected)]525.4 norm0 -[(mode and jumping to the 32-bit kernel start address. In a fully 32-bit OS, this tends to be the )1(only)0( mixed-size)]514.4 norm0 -[(instruction you need, since everything before it can be done in pure 16-bit code, and everything after it can be)]503.4 norm0 -[(pure 32-bit.)]492.4 norm2 -[(This jump must specify a 48-bit far address, since the target segment is a 32-bit one. However, it must be)]475.4 norm1 -[(assembled in a 16-bit segment, so just coding, for example,)]464.4 norm2 -[2( jmp 0x1234:0x56789ABC ; wrong!)]447.4 code3 -[(will not work, since the offset part of the address will be truncated to )2(0x9ABC)0( and the jump will be an)]430.4 norm1 -[(ordinary 16-bit far one.)]419.4 norm2 -[(The Linux kernel setup code gets round the inability of )2(as86)0( to generate the required instruction by coding it)]402.4 norm1 -[(manually, using )2(DB)0( instructions. NASM can go one better than that, by actually generating the right)]391.4 norm0 -[(instruction itself. Here's how to do it right:)]380.4 norm2 -[2( jmp dword 0x1234:0x56789ABC ; right)]363.4 code3 -[(The )2(DWORD)0( prefix \(strictly speaking, it should come )1(after)0( the colon, since it is declaring the )1(offset)0( field to be a)]346.4 norm1 -[(doubleword; but NASM will accept either form, since both are unambiguous\) forces the offset part to be)]335.4 norm0 -[(treated as far, in the assumption that you are deliberately writing a jump from a 16-bit segment to a 32-bit)]324.4 norm0 -[(one.)]313.4 norm2 -[(You can do the reverse operation, jumping from a 32-bit segment to a 16-bit one, by means of the )2(WORD)]296.4 norm1 -[(prefix:)]285.4 norm2 -[2( jmp word 0x8765:0x4321 ; 32 to 16 bit)]268.4 code3 -[(If the )2(WORD)0( prefix is specified in 16-bit mode, or the )2(DWORD)0( prefix in 32-bit mode, they will be ignored,)]251.4 norm1 -[(since each is explicitly forcing NASM into a mode it was in anyway.)]240.4 norm2 -[{/section-10.2 xa}(Addressing Between Different-Size Segments)](10.2)219 head3 -[(If your OS is mixed 16 and 32-bit, or if you are writing a DOS extender, you are likely to have to deal with)]202 norm1 -[(some 16-bit segments and some 32-bit ones. At some point, you will probably end up writing code in a)]191 norm0 -[(16-bit segment which has to access data in a 32-bit segment, or vice versa.)]180 norm2 -[(If the data you are trying to access in a 32-bit segment lies within the first 64K of the segment, you may be)]163 norm1 -[(able to get away with using an ordinary 16-bit addressing operation for the purpose; but sooner or later, you)]152 norm0 -[(will want to do 32-bit addressing from 16-bit mode.)]141 norm2 -[(The easiest way to do this is to make sure you use a register for the address, since any effective address)]124 norm1 -[(containing a 32-bit register is forced to be a 32-bit address. So you can do)]113 norm2 -(110)pageeven -restore showpage -%%Page: 111 111 -%%BeginPageSetup -save -%%EndPageSetup -/111 pa -[2( mov eax,offset_into_32_bit_segment_specified_by_fs )]681 code1 -[2( mov dword [fs:eax],0x11223344)]670 code2 -[(This is fine, but slightly cumbersome \(since it wastes an instruction and a register\) if you already know the)]653 norm1 -[(precise offset you are aiming at. The x86 architecture does allow 32-bit effective addresses to specify nothing)]642 norm0 -[(but a 4-byte offset, so why shouldn't NASM be able to generate the best instruction for the purpose?)]631 norm2 -[(It can. As in ){/section-10.1 xl}(section 10.1){el}(, you need only prefix the address with the )2(DWORD)0( keyword, and it will be forced to)]614 norm1 -[(be a 32-bit address:)]603 norm2 -[2( mov dword [fs:dword my_offset],0x11223344)]586 code3 -[(Also as in ){/section-10.1 xl}(section 10.1){el}(, NASM is not fussy about whether the )2(DWORD)0( prefix comes before or after the)]569 norm1 -[(segment override, so arguably a nicer-looking way to code the above instruction is)]558 norm2 -[2( mov dword [dword fs:my_offset],0x11223344)]541 code3 -[(Don't confuse the )2(DWORD)0( prefix )1(outside)0( the square brackets, which controls the size of the data stored at the)]524 norm1 -[(address, with the one )2(inside)0( the square brackets which controls the length of the address itself. The two can)]513 norm0 -[(quite easily be different:)]502 norm2 -[2( mov word [dword 0x12345678],0x9ABC)]485 code3 -[(This moves 16 bits of data to an address specified by a 32-bit offset.)]468 norm3 -[(You can also specify )2(WORD)0( or )2(DWORD)0( prefixes along with the )2(FAR)0( prefix to indirect far jumps or calls. For)]451 norm1 -[(example:)]440 norm2 -[2( call dword far [fs:word 0x4321])]423 code3 -[(This instruction contains an address specified by a 16-bit offset; it loads a 48-bit far pointer from that \(16-bit)]406 norm1 -[(segment and 32-bit offset\), and calls that address.)]395 norm2 -[{/section-10.3 xa}(Other Mixed-Size Instructions)](10.3)373.6 head3 -[(The other way you might want to access data might be using the string instructions \()2(LODSx)0(, )2(STOSx)0( and so)]356.6 norm1 -[(on\) or the )2(XLATB)0( instruction. These instructions, since they take no parameters, might seem to have no easy)]345.6 norm0 -[(way to make them perform 32-bit addressing when assembled in a 16-bit segment.)]334.6 norm2 -[(This is the purpose of NASM's )2(a16)0(, )2(a32)0( and )2(a64)0( prefixes. If you are coding )2(LODSB)0( in a 16-bit segment)]317.6 norm1 -[(but it is supposed to be accessing a string in a 32-bit segment, you should load the desired address into )2(ESI)]306.6 norm0 -[(and then code)]295.6 norm2 -[2( a32 lodsb)]278.6 code3 -[(The prefix forces the addressing size to 32 bits, meaning that )2(LODSB)0( loads from )2([DS:ESI])0( instead of)]261.6 norm1 -[2([DS:SI])0(. To access a string in a 16-bit segment when coding in a 32-bit one, the corresponding )2(a16)0( prefix)]250.6 norm0 -[(can be used.)]239.6 norm2 -[(The )2(a16)0(, )2(a32)0( and )2(a64)0( prefixes can be applied to any instruction in NASM's instruction table, but most of)]222.6 norm1 -[(them can generate all the useful forms without them. The prefixes are necessary only for instructions with)]211.6 norm0 -[(implicit addressing: )2(CMPSx)0(, )2(SCASx)0(, )2(LODSx)0(, )2(STOSx)0(, )2(MOVSx)0(, )2(INSx)0(, )2(OUTSx)0(, and )2(XLATB)0(. Also, the various)]200.6 norm0 -[(push and pop instructions \()2(PUSHA)0( and )2(POPF)0( as well as the more usual )2(PUSH)0( and )2(POP)0(\) can accept )2(a16)0(, )2(a32)]189.6 norm0 -[(or )2(a64)0( prefixes to force a particular one of )2(SP)0(, )2(ESP)0( or )2(RSP)0( to be used as a stack pointer, in case the stack)]178.6 norm0 -[(segment in use is a different size from the code segment.)]167.6 norm2 -[2(PUSH)0( and )2(POP)0(, when applied to segment registers in 32-bit mode, also have the slightly odd behaviour that)]150.6 norm1 -[(they push and pop 4 bytes at a time, of which the top two are ignored and the bottom two give the value of the)]139.6 norm0 -[(segment register being manipulated. To force the 16-bit behaviour of segment-register push and pop)]128.6 norm0 -[(instructions, you can use the operand-size prefix )2(o16)0(:)]117.6 norm2 -(111)pageodd -restore showpage -%%Page: 112 112 -%%BeginPageSetup -save -%%EndPageSetup -/112 pa -[2( o16 push ss )]681 code1 -[2( o16 push ds)]670 code2 -[(This code saves a doubleword of stack space by fitting two segment registers into the space which would)]653 norm1 -[(normally be consumed by pushing one.)]642 norm2 -[(\(You can also use the )2(o32)0( prefix to force the 32-bit behaviour when in 16-bit mode, but this seems less)]625 norm1 -[(useful.\))]614 norm2 -(112)pageeven -restore showpage -%%Page: 113 113 -%%BeginPageSetup -save -%%EndPageSetup -/113 pa -[{/chapter-11 xa}(Chapter 11: Writing 64-bit Code \(Unix, Win64\))]642.8 chap3 -[(This chapter attempts to cover some of the common issues involved when writing 64-bit code, to run under)]607.8 norm1 -[(Win64 or Unix. It covers how to write assembly code to interface with 64-bit C routines, and how to write)]596.8 norm0 -[(position-independent code for shared libraries.)]585.8 norm2 -[(All 64-bit code uses a flat memory model, since segmentation is not available in 64-bit mode. The one)]568.8 norm1 -[(exception is the )2(FS)0( and )2(GS)0( registers, which still add their bases.)]557.8 norm2 -[(Position independence in 64-bit mode is significantly simpler, since the processor supports )2(RIP)0(\226relative)]540.8 norm1 -[(addressing directly; see the )2(REL)0( keyword \(){/section-3.3 xl}(section 3.3){el}(\). On most 64-bit platforms, it is probably desirable to)]529.8 norm0 -[(make that the default, using the directive )2(DEFAULT REL)0( \(){/section-6.2 xl}(section 6.2){el}(\).)]518.8 norm2 -[(64-bit programming is relatively similar to 32-bit programming, but of course pointers are 64 bits long;)]501.8 norm1 -[(additionally, all existing platforms pass arguments in registers rather than on the stack. Furthermore, 64-bit)]490.8 norm0 -[(platforms use SSE2 by default for floating point. Please see the ABI documentation for your platform.)]479.8 norm2 -[(64-bit platforms differ in the sizes of the fundamental datatypes, not just from 32-bit platforms but from each)]462.8 norm1 -[(other. If a specific size data type is desired, it is probably best to use the types defined in the Standard C)]451.8 norm0 -[(header )2(<inttypes.h>)0(.)]440.8 norm2 -[(In 64-bit mode, the default instruction size is still 32 bits. When loading a value into a 32-bit register \(but not)]423.8 norm1 -[(an 8- or 16-bit register\), the upper 32 bits of the corresponding 64-bit register are set to zero.)]412.8 norm2 -[{/section-11.1 xa}(Register Names in 64-bit Mode)](11.1)391.4 head3 -[(NASM uses the following names for general-purpose registers in 64-bit mode, for 8-, 16-, 32- and 64-bit)]374.4 norm1 -[(references, respecitively:)]363.4 norm2 -[2( AL/AH, CL/CH, DL/DH, BL/BH, SPL, BPL, SIL, DIL, R8B-R15B )]346.4 code1 -[2( AX, CX, DX, BX, SP, BP, SI, DI, R8W-R15W )]335.4 code0 -[2( EAX, ECX, EDX, EBX, ESP, EBP, ESI, EDI, R8D-R15D )]324.4 code0 -[2( RAX, RCX, RDX, RBX, RSP, RBP, RSI, RDI, R8-R15)]313.4 code2 -[(This is consistent with the AMD documentation and most other assemblers. The Intel documentation,)]296.4 norm1 -[(however, uses the names )2(R8L-R15L)0( for 8-bit references to the higher registers. It is possible to use those)]285.4 norm0 -[(names by definiting them as macros; similarly, if one wants to use numeric names for the low 8 registers,)]274.4 norm0 -[(define them as macros. The standard macro package )2(altreg)0( \(see ){/section-5.1 xl}(section 5.1){el}(\) can be used for this purpose.)]263.4 norm2 -[{/section-11.2 xa}(Immediates and Displacements in 64-bit Mode)](11.2)242 head3 -[(In 64-bit mode, immediates and displacements are generally only 32 bits wide. NASM will therefore truncate)]225 norm1 -[(most displacements and immediates to 32 bits.)]214 norm2 -[(The only instruction which takes a full )(64-bit immediate is:)]197 norm3 -[2( MOV reg64,imm64)]180 code3 -[(NASM will produce this instruction whenever the programmer uses )2(MOV)0( with an immediate into a 64-bit)]163 norm1 -[(register. If this is not desirable, simply specify the equivalent 32-bit register, which will be automatically)]152 norm0 -[(zero-extended by the processor, or specify the immediate as )2(DWORD)0(:)]141 norm2 -[2( mov rax,foo ; 64-bit immediate )]124 code1 -[2( mov rax,qword foo ; \(identical\) )]113 code0 -(113)pageodd -restore showpage -%%Page: 114 114 -%%BeginPageSetup -save -%%EndPageSetup -/114 pa -[2( mov eax,foo ; 32-bit immediate, zero-extended )]681 code0 -[2( mov rax,dword foo ; 32-bit immediate, sign-extended)]670 code2 -[(The length of these instructions are 10, 5 and 7 bytes, respectively.)]653 norm3 -[(The only instructions which take a full )(64-bit )1(displacement)0( is loading or storing, using )2(MOV)0(, )2(AL)0(, )2(AX)0(, )2(EAX)0( or)]636 norm1 -[2(RAX)0( \(but no other registers\) to an absolute 64-bit address. Since this is a relatively rarely used instruction)]625 norm0 -[(\(64-bit code generally uses relative addressing\), the programmer has to explicitly declare the displacement)]614 norm0 -[(size as )2(QWORD)0(:)]603 norm2 -[2( default abs )]586 code1 -[2()]575 code0 -[2( mov eax,[foo] ; 32-bit absolute disp, sign-extended )]564 code0 -[2( mov eax,[a32 foo] ; 32-bit absolute disp, zero-extended )]553 code0 -[2( mov eax,[qword foo] ; 64-bit absolute disp )]542 code0 -[2()]531 code0 -[2( default rel )]520 code0 -[2()]509 code0 -[2( mov eax,[foo] ; 32-bit relative disp )]498 code0 -[2( mov eax,[a32 foo] ; d:o, address truncated to 32 bits\(!\) )]487 code0 -[2( mov eax,[qword foo] ; error )]476 code0 -[2( mov eax,[abs qword foo] ; 64-bit absolute disp)]465 code2 -[(A sign-extended absolute displacement can access from \2262 GB to +2 GB; a zero-extended absolute)]448 norm1 -[(displacement can access from 0 to 4 GB.)]437 norm2 -[{/section-11.3 xa}(Interfacing to 64-bit C Programs \(Unix\))](11.3)415.6 head3 -[(On Unix, the 64-bit ABI is defined by the document:)]398.6 norm3 -[{(http://www.x86-64.org/documentation/abi.pdf)wl}2(http://www.x86-64.org/documentation/abi.pdf){el}]381.6 norm3 -[(Although written for AT&T-syntax assembly, the concepts apply equally well for NASM-style assembly.)]364.6 norm1 -[(What follows is a simplified summary.)]353.6 norm2 -[(The first six integer arguments \(from the left\) are passed in )2(RDI)0(, )2(RSI)0(, )2(RDX)0(, )2(RCX)0(, )2(R8)0(, and )2(R9)0(, in that order.)]336.6 norm1 -[(Additional integer arguments are passed on the stack. These registers, plus )2(RAX)0(, )2(R10)0( and )2(R11)0( are destroyed)]325.6 norm0 -[(by function calls, and thus are available for use by the function without saving.)]314.6 norm2 -[(Integer return values are passed in )2(RAX)0( and )2(RDX)0(, in that order.)]297.6 norm3 -[(Floating point is done using SSE registers, except for )2(long double)0(. Floating-point arguments are passed)]280.6 norm1 -[(in )2(XMM0)0( to )2(XMM7)0(; return is )2(XMM0)0( and )2(XMM1)0(. )2(long double)0( are passed on the stack, and returned in )2(ST0)]269.6 norm0 -[(and )2(ST1)0(.)]258.6 norm2 -[(All SSE and x87 registers are destroyed by function calls.)]241.6 norm3 -[(On 64-bit Unix, )2(long)0( is 64 bits.)]224.6 norm3 -[(Integer and SSE register arguments are counted separately, so for the case of)]207.6 norm3 -[2( void foo\(long a, double b, int c\))]190.6 code3 -[2(a)0( is passed in )2(RDI)0(, )2(b)0( in )2(XMM0)0(, and )2(c)0( in )2(ESI)0(.)]173.6 norm3 -[{/section-11.4 xa}(Interfacing to 64-bit C Programs \(Win64\))](11.4)152.2 head3 -[(The Win64 ABI is described at:)]135.2 norm3 -[{(http://msdn2.microsoft.com/en-gb/library/ms794533.aspx)wl}2(http://msdn2.microsoft.com/en-gb/library/ms794533.aspx){el}]118.2 norm3 -[(What follows is a simplified summary.)]101.2 norm3 -(114)pageeven -restore showpage -%%Page: 115 115 -%%BeginPageSetup -save -%%EndPageSetup -/115 pa -[(The first four integer arguments are passed in )2(RCX)0(, )2(RDX)0(, )2(R8)0( and )2(R9)0(, in that order. Additional integer)]681 norm1 -[(arguments are passed on the stack. These registers, plus )2(RAX)0(, )2(R10)0( and )2(R11)0( are destroyed by function calls,)]670 norm0 -[(and thus are available for use by the function without saving.)]659 norm2 -[(Integer return values are passed in )2(RAX)0( only.)]642 norm3 -[(Floating point is done using SSE registers, except for )2(long double)0(. Floating-point arguments are passed)]625 norm1 -[(in )2(XMM0)0( to )2(XMM3)0(; return is )2(XMM0)0( only.)]614 norm2 -[(On Win64, )2(long)0( is 32 bits; )2(long long)0( or )2(_int64)0( is 64 bits.)]597 norm3 -[(Integer and SSE register arguments are counted together, so for the case of)]580 norm3 -[2( void foo\(long long a, double b, int c\))]563 code3 -[2(a)0( is passed in )2(RCX)0(, )2(b)0( in )2(XMM1)0(, and )2(c)0( in )2(R8D)0(.)]546 norm3 -(115)pageodd -restore showpage -%%Page: 116 116 -%%BeginPageSetup -save -%%EndPageSetup -/116 pa -[{/chapter-12 xa}(Chapter 12: Troubleshooting)]642.8 chap3 -[(This chapter describes some of the common problems that users have been known to encounter with NASM,)]607.8 norm1 -[(and answers them. It also gives instructions for reporting bugs in NASM if you find a difficulty that isn't)]596.8 norm0 -[(listed here.)]585.8 norm2 -[{/section-12.1 xa}(Common Problems)](12.1)564.4 head3 -[{/section-12.1.1 xa}(NASM Generates )(Inefficient Code)](12.1.1)545.2 subh3 -[(We sometimes get `bug' reports about NASM generating inefficient, or even `wrong', code on instructions)]528.2 norm1 -[(such as )2(ADD ESP,8)0(. This is a deliberate design feature, connected to predictability of output: NASM, on)]517.2 norm0 -[(seeing )2(ADD ESP,8)0(, will generate the form of the instruction which leaves room for a 32-bit offset. You)]506.2 norm0 -[(need to code )2(ADD ESP,BYTE 8)0( if you want the space-efficient form of the instruction. This isn't a bug, it's)]495.2 norm0 -[(user error: if you prefer to have NASM produce the more efficient code automatically enable optimization)]484.2 norm0 -[(with the )2(-O)0( option \(see ){/section-2.1.22 xl}(section 2.1.22){el}(\).)]473.2 norm2 -[{/section-12.1.2 xa}(My Jumps are Out of Range)](12.1.2)454 subh3 -[(Similarly, people complain that when they issue )(conditional jumps \(which are )2(SHORT)0( by default\) that try to)]437 norm1 -[(jump too far, NASM reports `short jump out of range' instead of making the jumps longer.)]426 norm2 -[(This, again, is partly a predictability issue, but in fact has a more practical reason as well. NASM has no)]409 norm1 -[(means of being told what type of processor the code it is generating will be run on; so it cannot decide for)]398 norm0 -[(itself that it should generate )2(Jcc NEAR)0( type instructions, because it doesn't know that it's working for a 386)]387 norm0 -[(or above. Alternatively, it could replace the out-of-range short )2(JNE)0( instruction with a very short )2(JE)]376 norm0 -[(instruction that jumps over a )2(JMP NEAR)0(; this is a sensible solution for processors below a 386, but hardly)]365 norm0 -[(efficient on processors which have good branch prediction )1(and)0( could have used )2(JNE NEAR)0( instead. So, once)]354 norm0 -[(again, it's up to the user, not the assembler, to decide what instructions should be generated. See ){/section-2.1.22 xl}(section){el}]343 norm0 -[{/section-2.1.22 xl}(2.1.22){el}(.)]332 norm2 -[{/section-12.1.3 xa}2(ORG)0( Doesn't Work)](12.1.3)312.8 subh3 -[(People writing )(boot sector programs in the )2(bin)0( format often complain that )2(ORG)0( doesn't work the way they'd)]295.8 norm1 -[(like: in order to place the )2(0xAA55)0( signature word at the end of a 512-byte boot sector, people who are used)]284.8 norm0 -[(to MASM tend to code)]273.8 norm2 -[2( ORG 0 )]256.8 code1 -[2()]245.8 code0 -[2( ; some boot sector code )]234.8 code0 -[2()]223.8 code0 -[2( ORG 510 )]212.8 code0 -[2( DW 0xAA55)]201.8 code2 -[(This is not the intended use of the )2(ORG)0( directive in NASM, and will not work. The correct way to solve this)]184.8 norm1 -[(problem in NASM is to use the )2(TIMES)0( directive, like this:)]173.8 norm2 -[2( ORG 0 )]156.8 code1 -[2()]145.8 code0 -[2( ; some boot sector code )]134.8 code0 -[2()]123.8 code0 -[2( TIMES 510-\($-$$\) DB 0 )]112.8 code0 -[2( DW 0xAA55)]101.8 code2 -(116)pageeven -restore showpage -%%Page: 117 117 -%%BeginPageSetup -save -%%EndPageSetup -/117 pa -[(The )2(TIMES)0( directive will insert exactly enough zero bytes into the output to move the assembly point up to)]681 norm1 -[(510. This method also has the advantage that if you accidentally fill your boot sector too full, NASM will)]670 norm0 -[(catch the problem at assembly time and report it, so you won't end up with a boot sector that you have to)]659 norm0 -[(disassemble to find out what's wrong with it.)]648 norm2 -[{/section-12.1.4 xa}2(TIMES)0( Doesn't Work)](12.1.4)628.8 subh3 -[(The other common problem with the above code is people who write the )2(TIMES)0( line as)]611.8 norm3 -[2( TIMES 510-$ DB 0)]594.8 code3 -[(by reasoning that )2($)0( should be a pure number, just like 510, so the difference between them is also a pure)]577.8 norm1 -[(number and can happily be fed to )2(TIMES)0(.)]566.8 norm2 -[(NASM is a )1(modular)0( assembler: the various component parts are designed to be easily separable for re-use, so)]549.8 norm1 -[(they don't exchange information unnecessarily. In consequence, the )2(bin)0( output format, even though it has)]538.8 norm0 -[(been told by the )2(ORG)0( directive that the )2(.text)0( section should start at 0, does not pass that information back to)]527.8 norm0 -[(the expression evaluator. So from the evaluator's point of view, )2($)0( isn't a pure number: it's an offset from a)]516.8 norm0 -[(section base. Therefore the difference between )2($)0( and 510 is also not a pure number, but involves a section)]505.8 norm0 -[(base. Values involving section bases cannot be passed as arguments to )2(TIMES)0(.)]494.8 norm2 -[(The solution, as in the previous section, is to code the )2(TIMES)0( line in the form)]477.8 norm3 -[2( TIMES 510-\($-$$\) DB 0)]460.8 code3 -[(in which )2($)0( and )2($$)0( are offsets from the same section base, and so their difference is a pure number. This will)]443.8 norm1 -[(solve the problem and generate sensible code.)]432.8 norm2 -[{/section-12.2 xa}(Bugs)](12.2)411.4 head3 -[(We have never yet released a version of NASM with any )1(known)0( bugs. That doesn't usually stop there being)]394.4 norm1 -[(plenty we didn't know about, though. Any that you find should be reported firstly via the )2(bugtracker)0( at)]383.4 norm0 -[{(https://sourceforge.net/projects/nasm/)wl}2(https://sourceforge.net/projects/nasm/){el}0( \(click on "Bugs"\), or if that fails then through one of)]372.4 norm0 -[(the contacts in ){/section-1.2 xl}(section 1.2){el}(.)]361.4 norm2 -[(Please read ){/section-2.2 xl}(section 2.2){el}( first, and don't report the bug if it's listed in there as a deliberate feature. \(If you think)]344.4 norm1 -[(the feature is badly thought out, feel free to send us reasons why you think it should be changed, but don't just)]333.4 norm0 -[(send us mail saying `This is a bug' if the documentation says we did it on purpose.\) Then read ){/section-12.1 xl}(section 12.1){el}(,)]322.4 norm0 -[(and don't bother reporting the bug if it's listed there.)]311.4 norm2 -[(If you do report a bug, )1(please)0( give us all of the following information:)]294.4 norm3 -[(What operating system you're running NASM under. DOS, Linux, NetBSD, Win16, Win32, VMS \(I'd be)]277.4 bull1 -[(impressed\), whatever.)]266.4 bull2 -[(If you're running NASM under DOS or Win32, tell us whether you've compiled your own executable from)]249.4 bull1 -[(the DOS source archive, or whether you were using the standard distribution binaries out of the archive. If)]238.4 bull0 -[(you were using a locally built executable, try to reproduce the problem using one of the standard binaries,)]227.4 bull0 -[(as this will make it easier for us to reproduce your problem prior to fixing it.)]216.4 bull2 -[(Which version of NASM you're using, and exactly how you invoked it. Give us the precise command line,)]199.4 bull1 -[(and the contents of the )2(NASMENV)0( environment variable if any.)]188.4 bull2 -[(Which versions of any supplementary programs you're using, and how you invoked them. If the problem)]171.4 bull1 -[(only becomes visible at link time, tell us what linker you're using, what version of it you've got, and the)]160.4 bull0 -[(exact linker command line. If the problem involves linking against object files generated by a compiler, tell)]149.4 bull0 -[(us what compiler, what version, and what command line or options you used. \(If you're compiling in an)]138.4 bull0 -[(IDE, please try to reproduce the problem with the command-line version of the compiler.\))]127.4 bull2 -(117)pageodd -restore showpage -%%Page: 118 118 -%%BeginPageSetup -save -%%EndPageSetup -/118 pa -[(If at all possible, send us a NASM source file which exhibits the problem. If this causes copyright)]681 bull1 -[(problems \(e.g. you can only reproduce the bug in restricted-distribution code\) then bear in mind the)]670 bull0 -[(following two points: firstly, we guarantee that any source code sent to us for the purposes of debugging)]659 bull0 -[(NASM will be used )1(only)0( for the purposes of debugging NASM, and that we will delete all our copies of it)]648 bull0 -[(as soon as we have found and fixed the bug or bugs in question; and secondly, we would prefer )1(not)0( to be)]637 bull0 -[(mailed large chunks of code anyway. The smaller the file, the better. A three-line sample file that does)]626 bull0 -[(nothing useful )1(except)0( demonstrate the problem is much easier to work with than a fully fledged)]615 bull0 -[(ten-thousand-line program. \(Of course, some errors )1(do)0( only crop up in large files, so this may not be)]604 bull0 -[(possible.\))]593 bull2 -[(A description of what the problem actually )1(is)0(. `It doesn't work' is )1(not)0( a helpful description! Please describe)]576 bull1 -[(exactly what is happening that shouldn't be, or what isn't happening that should. Examples might be:)]565 bull0 -[(`NASM generates an error message saying Line 3 for an error that's actually on Line 5'; `NASM generates)]554 bull0 -[(an error message that I believe it shouldn't be generating at all'; `NASM fails to generate an error message)]543 bull0 -[(that I believe it )1(should)0( be generating'; `the object file produced from this source code crashes my linker';)]532 bull0 -[(`the ninth byte of the output file is 66 and I think it should be 77 instead'.)]521 bull2 -[(If you believe the output file from NASM to be faulty, send it to us. That allows us to determine whether)]504 bull1 -[(our own copy of NASM generates the same file, or whether the problem is related to portability issues)]493 bull0 -[(between our development platforms and yours. We can handle binary files mailed to us as MIME)]482 bull0 -[(attachments, uuencoded, and even BinHex. Alternatively, we may be able to provide an FTP site you can)]471 bull0 -[(upload the suspect files to; but mailing them is easier for us.)]460 bull2 -[(Any other information or data files that might be helpful. If, for example, the problem involves NASM)]443 bull1 -[(failing to generate an object file while TASM can generate an equivalent file without trouble, then send us)]432 bull0 -[1(both)0( object files, so we can see what TASM is doing differently from us.)]421 bull2 -(118)pageeven -restore showpage -%%Page: 119 119 -%%BeginPageSetup -save -%%EndPageSetup -/119 pa -[{/appendix-A xa}(Appendix A: )(Ndisasm)]642.8 appn3 -[(The Netwide Disassembler, NDISASM)]601.8 norm3 -[{/section-A.1 xa}(Introduction)](A.1)580.4 head3 -[(The Netwide Disassembler is a small companion program to the Netwide Assembler, NASM. It seemed a)]563.4 norm1 -[(shame to have an x86 assembler, complete with a full instruction table, and not make as much use of it as)]552.4 norm0 -[(possible, so here's a disassembler which shares the instruction table \(and some other bits of code\) with NASM.)]541.4 norm2 -[(The Netwide Disassembler does nothing except to produce disassemblies of )1(binary)0( source files. NDISASM)]524.4 norm1 -[(does not have any understanding of object file formats, like )2(objdump)0(, and it will not understand )2(DOS .EXE)]513.4 norm0 -[(files like )2(debug)0( will. It just disassembles.)]502.4 norm2 -[{/section-A.2 xa}(Getting Started: Installation)](A.2)481 head3 -[(See ){/section-1.3 xl}(section 1.3){el}( for installation instructions. NDISASM, like NASM, has a )2(man page)0( which you may want)]464 norm1 -[(to put somewhere useful, if you are on a Unix system.)]453 norm2 -[{/section-A.3 xa}(Running NDISASM)](A.3)431.6 head3 -[(To disassemble a file, you will typically use a command of the form)]414.6 norm3 -[2( ndisasm -b {16|32|64} filename)]397.6 code3 -[(NDISASM can disassemble 16-, 32- or 64-bit code equally easily, provided of course that you remember to)]380.6 norm1 -[(specify which it is to work with. If no )2(-b)0( switch is present, NDISASM works in 16-bit mode by default. The)]369.6 norm0 -[2(-u)0( switch \(for USE32\) also invokes 32-bit mode.)]358.6 norm2 -[(Two more command line options are )2(-r)0( which reports the version number of NDISASM you are running, and)]341.6 norm1 -[2(-h)0( which gives a short summary of command line options.)]330.6 norm2 -[{/section-A.3.1 xa}(COM Files: Specifying an Origin)](A.3.1)311.4 subh3 -[(To disassemble a )2(DOS .COM)0( file correctly, a disassembler must assume that the first instruction in the file is)]294.4 norm1 -[(loaded at address )2(0x100)0(, rather than at zero. NDISASM, which assumes by default that any file you give it is)]283.4 norm0 -[(loaded at zero, will therefore need to be informed of this.)]272.4 norm2 -[(The )2(-o)0( option allows you to declare a different origin for the file you are disassembling. Its argument may be)]255.4 norm1 -[(expressed in any of the NASM numeric formats: decimal by default, if it begins with `)2($)0(' or `)2(0x)0(' or ends in)]244.4 norm0 -[(`)2(H)0(' it's )2(hex)0(, if it ends in `)2(Q)0(' it's )2(octal)0(, and if it ends in `)2(B)0(' it's )2(binary)0(.)]233.4 norm2 -[(Hence, to disassemble a )2(.COM)0( file:)]216.4 norm3 -[2( ndisasm -o100h filename.com)]199.4 code3 -[(will do the trick.)]182.4 norm3 -[{/section-A.3.2 xa}(Code Following Data: Synchronisation)](A.3.2)163.2 subh3 -[(Suppose you are disassembling a file which contains some data which isn't machine code, and )1(then)0( contains)]146.2 norm1 -[(some machine code. NDISASM will faithfully plough through the data section, producing machine)]135.2 norm0 -[(instructions wherever it can \(although most of them will look bizarre, and some may have unusual prefixes,)]124.2 norm0 -[(e.g. `)2(FS OR AX,0x240A)0('\), and generating `DB' instructions ever so often if it's totally stumped. Then it)]113.2 norm0 -[(will reach the code section.)]102.2 norm2 -(119)pageodd -restore showpage -%%Page: 120 120 -%%BeginPageSetup -save -%%EndPageSetup -/120 pa -[(Supposing NDISASM has just finished generating a strange machine instruction from part of the data section,)]681 norm1 -[(and its file position is now one byte )1(before)0( the beginning of the code section. It's entirely possible that another)]670 norm0 -[(spurious instruction will get generated, starting with the final byte of the data section, and then the correct first)]659 norm0 -[(instruction in the code section will not be seen because the starting point skipped over it. This isn't really ideal.)]648 norm2 -[(To avoid this, you can specify a `)2(synchronisation)0(' point, or indeed as many synchronisation points as)]631 norm1 -[(you like \(although NDISASM can only handle 2147483647 sync points internally\). The definition of a sync)]620 norm0 -[(point is this: NDISASM guarantees to hit sync points exactly during disassembly. If it is thinking about)]609 norm0 -[(generating an instruction which would cause it to jump over a sync point, it will discard that instruction and)]598 norm0 -[(output a `)2(db)0(' instead. So it )1(will)0( start disassembly exactly from the sync point, and so you )1(will)0( see all the)]587 norm0 -[(instructions in your code section.)]576 norm2 -[(Sync points are specified using the )2(-s)0( option: they are measured in terms of the program origin, not the file)]559 norm1 -[(position. So if you want to synchronize after 32 bytes of a )2(.COM)0( file, you would have to do)]548 norm2 -[2( ndisasm -o100h -s120h file.com)]531 code3 -[(rather than)]514 norm3 -[2( ndisasm -o100h -s20h file.com)]497 code3 -[(As stated above, you can specify multiple sync markers if you need to, just by repeating the )2(-s)0( option.)]480 norm3 -[{/section-A.3.3 xa}(Mixed Code and Data: Automatic \(Intelligent\) Synchronisation )](A.3.3)460.8 subh3 -[(Suppose you are disassembling the boot sector of a )2(DOS)0( floppy \(maybe it has a virus, and you need to)]443.8 norm1 -[(understand the virus so that you know what kinds of damage it might have done you\). Typically, this will)]432.8 norm0 -[(contain a )2(JMP)0( instruction, then some data, then the rest of the code. So there is a very good chance of)]421.8 norm0 -[(NDISASM being )1(misaligned)0( when the data ends and the code begins. Hence a sync point is needed.)]410.8 norm2 -[(On the other hand, why should you have to specify the sync point manually? What you'd do in order to find)]393.8 norm1 -[(where the sync point would be, surely, would be to read the )2(JMP)0( instruction, and then to use its target address)]382.8 norm0 -[(as a sync point. So can NDISASM do that for you?)]371.8 norm2 -[(The answer, of course, is yes: using either of the synonymous switches )2(-a)0( \(for automatic sync\) or )2(-i)0( \(for)]354.8 norm1 -[(intelligent sync\) will enable )2(auto-sync)0( mode. Auto-sync mode automatically generates a sync point for)]343.8 norm0 -[(any forward-referring PC-relative jump or call instruction that NDISASM encounters. \(Since NDISASM is)]332.8 norm0 -[(one-pass, if it encounters a PC-relative jump whose target has already been processed, there isn't much it can)]321.8 norm0 -[(do about it...\))]310.8 norm2 -[(Only PC-relative jumps are processed, since an absolute jump is either through a register \(in which case)]293.8 norm1 -[(NDISASM doesn't know what the register contains\) or involves a segment address \(in which case the target)]282.8 norm0 -[(code isn't in the same segment that NDISASM is working in, and so the sync point can't be placed anywhere)]271.8 norm0 -[(useful\).)]260.8 norm2 -[(For some kinds of file, this mechanism will automatically put sync points in all the right places, and save you)]243.8 norm1 -[(from having to place any sync points manually. However, it should be stressed that auto-sync mode is )1(not)]232.8 norm0 -[(guaranteed to catch all the sync points, and you may still have to place some manually.)]221.8 norm2 -[(Auto-sync mode doesn't prevent you from declaring manual sync points: it just adds automatically generated)]204.8 norm1 -[(ones to the ones you provide. It's perfectly feasible to specify )2(-i)0( )1(and)0( some )2(-s)0( options.)]193.8 norm2 -[(Another caveat with auto-sync mode is that if, by some unpleasant fluke, something in your data section)]176.8 norm1 -[(should disassemble to a PC-relative call or jump instruction, NDISASM may obediently place a sync point in)]165.8 norm0 -[(a totally random place, for example in the middle of one of the instructions in your code section. So you may)]154.8 norm0 -[(end up with a wrong disassembly even if you use auto-sync. Again, there isn't much I can do about this. If)]143.8 norm0 -[(you have problems, you'll have to use manual sync points, or use the )2(-k)0( option \(documented below\) to)]132.8 norm0 -[(suppress disassembly of the data area.)]121.8 norm2 -(120)pageeven -restore showpage -%%Page: 121 121 -%%BeginPageSetup -save -%%EndPageSetup -/121 pa -[{/section-A.3.4 xa}(Other Options)](A.3.4)678.8 subh3 -[(The )2(-e)0( option skips a header on the file, by ignoring the first N bytes. This means that the header is )1(not)]661.8 norm1 -[(counted towards the disassembly offset: if you give )2(-e10 -o10)0(, disassembly will start at byte 10 in the file,)]650.8 norm0 -[(and this will be given offset 10, not 20.)]639.8 norm2 -[(The )2(-k)0( option is provided with two comma-separated numeric arguments, the first of which is an assembly)]622.8 norm1 -[(offset and the second is a number of bytes to skip. This )1(will)0( count the skipped bytes towards the assembly)]611.8 norm0 -[(offset: its use is to suppress disassembly of a data section which wouldn't contain anything you wanted to see)]600.8 norm0 -[(anyway.)]589.8 norm2 -[{/section-A.4 xa}(Bugs and Improvements)](A.4)568.4 head3 -[(There are no known bugs. However, any you find, with patches if possible, should be sent to)]551.4 norm1 -[{(mailto:nasm-bugs@lists.sourceforge.net)wl}2(nasm-bugs@lists.sourceforge.net){el}0(, or to the developer's site at)]540.4 norm0 -[{(https://sourceforge.net/projects/nasm/)wl}2(https://sourceforge.net/projects/nasm/){el}0( and we'll try to fix them. Feel free to send)]529.4 norm0 -[(contributions and new features as well.)]518.4 norm2 -(121)pageodd -restore showpage -%%Page: 122 122 -%%BeginPageSetup -save -%%EndPageSetup -/122 pa -[{/appendix-B xa}(Appendix B: )(Instruction List)]642.8 appn3 -[{/section-B.1 xa}(Introduction)](B.1)597.4 head3 -[(The following sections show the instructions which NASM currently supports. For each instruction, there is a)]580.4 norm1 -[(separate entry for each supported addressing mode. The third column shows the processor type in which the)]569.4 norm0 -[(instruction was introduced and, when appropriate, one or more usage flags.)]558.4 norm2 -[{/section-B.1.1 xa}(Special instructions...)](B.1.1)539.2 subh3 -[2(DB )]522.2 code1 -[2(DW )]511.2 code0 -[2(DD )]500.2 code0 -[2(DQ )]489.2 code0 -[2(DT )]478.2 code0 -[2(DO )]467.2 code0 -[2(DY )]456.2 code0 -[2(RESB imm 8086 )]445.2 code0 -[2(RESW )]434.2 code0 -[2(RESD )]423.2 code0 -[2(RESQ )]412.2 code0 -[2(REST )]401.2 code0 -[2(RESO )]390.2 code0 -[2(RESY )]379.2 code2 -[{/section-B.1.2 xa}(Conventional instructions)](B.1.2)360 subh3 -[2(AAA 8086,NOLONG )]343 code1 -[2(AAD 8086,NOLONG )]332 code0 -[2(AAD imm 8086,NOLONG )]321 code0 -[2(AAM 8086,NOLONG )]310 code0 -[2(AAM imm 8086,NOLONG )]299 code0 -[2(AAS 8086,NOLONG )]288 code0 -[2(ADC mem,reg8 8086 )]277 code0 -[2(ADC reg8,reg8 8086 )]266 code0 -[2(ADC mem,reg16 8086 )]255 code0 -[2(ADC reg16,reg16 8086 )]244 code0 -[2(ADC mem,reg32 386 )]233 code0 -[2(ADC reg32,reg32 386 )]222 code0 -[2(ADC mem,reg64 X64 )]211 code0 -[2(ADC reg64,reg64 X64 )]200 code0 -[2(ADC reg8,mem 8086 )]189 code0 -[2(ADC reg8,reg8 8086 )]178 code0 -[2(ADC reg16,mem 8086 )]167 code0 -[2(ADC reg16,reg16 8086 )]156 code0 -[2(ADC reg32,mem 386 )]145 code0 -[2(ADC reg32,reg32 386 )]134 code0 -[2(ADC reg64,mem X64 )]123 code0 -[2(ADC reg64,reg64 X64 )]112 code0 -[2(ADC rm16,imm8 8086 )]101 code0 -(122)pageeven -restore showpage -%%Page: 123 123 -%%BeginPageSetup -save -%%EndPageSetup -/123 pa -[2(ADC rm32,imm8 386 )]681 code0 -[2(ADC rm64,imm8 X64 )]670 code0 -[2(ADC reg_al,imm 8086 )]659 code0 -[2(ADC reg_ax,sbyte16 8086 )]648 code0 -[2(ADC reg_ax,imm 8086 )]637 code0 -[2(ADC reg_eax,sbyte32 386 )]626 code0 -[2(ADC reg_eax,imm 386 )]615 code0 -[2(ADC reg_rax,sbyte64 X64 )]604 code0 -[2(ADC reg_rax,imm X64 )]593 code0 -[2(ADC rm8,imm 8086 )]582 code0 -[2(ADC rm16,imm 8086 )]571 code0 -[2(ADC rm32,imm 386 )]560 code0 -[2(ADC rm64,imm X64 )]549 code0 -[2(ADC mem,imm8 8086 )]538 code0 -[2(ADC mem,imm16 8086 )]527 code0 -[2(ADC mem,imm32 386 )]516 code0 -[2(ADD mem,reg8 8086 )]505 code0 -[2(ADD reg8,reg8 8086 )]494 code0 -[2(ADD mem,reg16 8086 )]483 code0 -[2(ADD reg16,reg16 8086 )]472 code0 -[2(ADD mem,reg32 386 )]461 code0 -[2(ADD reg32,reg32 386 )]450 code0 -[2(ADD mem,reg64 X64 )]439 code0 -[2(ADD reg64,reg64 X64 )]428 code0 -[2(ADD reg8,mem 8086 )]417 code0 -[2(ADD reg8,reg8 8086 )]406 code0 -[2(ADD reg16,mem 8086 )]395 code0 -[2(ADD reg16,reg16 8086 )]384 code0 -[2(ADD reg32,mem 386 )]373 code0 -[2(ADD reg32,reg32 386 )]362 code0 -[2(ADD reg64,mem X64 )]351 code0 -[2(ADD reg64,reg64 X64 )]340 code0 -[2(ADD rm16,imm8 8086 )]329 code0 -[2(ADD rm32,imm8 386 )]318 code0 -[2(ADD rm64,imm8 X64 )]307 code0 -[2(ADD reg_al,imm 8086 )]296 code0 -[2(ADD reg_ax,sbyte16 8086 )]285 code0 -[2(ADD reg_ax,imm 8086 )]274 code0 -[2(ADD reg_eax,sbyte32 386 )]263 code0 -[2(ADD reg_eax,imm 386 )]252 code0 -[2(ADD reg_rax,sbyte64 X64 )]241 code0 -[2(ADD reg_rax,imm X64 )]230 code0 -[2(ADD rm8,imm 8086 )]219 code0 -[2(ADD rm16,imm 8086 )]208 code0 -[2(ADD rm32,imm 386 )]197 code0 -[2(ADD rm64,imm X64 )]186 code0 -[2(ADD mem,imm8 8086 )]175 code0 -[2(ADD mem,imm16 8086 )]164 code0 -[2(ADD mem,imm32 386 )]153 code0 -[2(AND mem,reg8 8086 )]142 code0 -[2(AND reg8,reg8 8086 )]131 code0 -[2(AND mem,reg16 8086 )]120 code0 -[2(AND reg16,reg16 8086 )]109 code0 -(123)pageodd -restore showpage -%%Page: 124 124 -%%BeginPageSetup -save -%%EndPageSetup -/124 pa -[2(AND mem,reg32 386 )]681 code0 -[2(AND reg32,reg32 386 )]670 code0 -[2(AND mem,reg64 X64 )]659 code0 -[2(AND reg64,reg64 X64 )]648 code0 -[2(AND reg8,mem 8086 )]637 code0 -[2(AND reg8,reg8 8086 )]626 code0 -[2(AND reg16,mem 8086 )]615 code0 -[2(AND reg16,reg16 8086 )]604 code0 -[2(AND reg32,mem 386 )]593 code0 -[2(AND reg32,reg32 386 )]582 code0 -[2(AND reg64,mem X64 )]571 code0 -[2(AND reg64,reg64 X64 )]560 code0 -[2(AND rm16,imm8 8086 )]549 code0 -[2(AND rm32,imm8 386 )]538 code0 -[2(AND rm64,imm8 X64 )]527 code0 -[2(AND reg_al,imm 8086 )]516 code0 -[2(AND reg_ax,sbyte16 8086 )]505 code0 -[2(AND reg_ax,imm 8086 )]494 code0 -[2(AND reg_eax,sbyte32 386 )]483 code0 -[2(AND reg_eax,imm 386 )]472 code0 -[2(AND reg_rax,sbyte64 X64 )]461 code0 -[2(AND reg_rax,imm X64 )]450 code0 -[2(AND rm8,imm 8086 )]439 code0 -[2(AND rm16,imm 8086 )]428 code0 -[2(AND rm32,imm 386 )]417 code0 -[2(AND rm64,imm X64 )]406 code0 -[2(AND mem,imm8 8086 )]395 code0 -[2(AND mem,imm16 8086 )]384 code0 -[2(AND mem,imm32 386 )]373 code0 -[2(ARPL mem,reg16 286,PROT,NOLONG )]362 code0 -[2(ARPL reg16,reg16 286,PROT,NOLONG )]351 code0 -[2(BB0_RESET PENT,CYRIX,ND )]340 code0 -[2(BB1_RESET PENT,CYRIX,ND )]329 code0 -[2(BOUND reg16,mem 186,NOLONG )]318 code0 -[2(BOUND reg32,mem 386,NOLONG )]307 code0 -[2(BSF reg16,mem 386 )]296 code0 -[2(BSF reg16,reg16 386 )]285 code0 -[2(BSF reg32,mem 386 )]274 code0 -[2(BSF reg32,reg32 386 )]263 code0 -[2(BSF reg64,mem X64 )]252 code0 -[2(BSF reg64,reg64 X64 )]241 code0 -[2(BSR reg16,mem 386 )]230 code0 -[2(BSR reg16,reg16 386 )]219 code0 -[2(BSR reg32,mem 386 )]208 code0 -[2(BSR reg32,reg32 386 )]197 code0 -[2(BSR reg64,mem X64 )]186 code0 -[2(BSR reg64,reg64 X64 )]175 code0 -[2(BSWAP reg32 486 )]164 code0 -[2(BSWAP reg64 X64 )]153 code0 -[2(BT mem,reg16 386 )]142 code0 -[2(BT reg16,reg16 386 )]131 code0 -[2(BT mem,reg32 386 )]120 code0 -[2(BT reg32,reg32 386 )]109 code0 -(124)pageeven -restore showpage -%%Page: 125 125 -%%BeginPageSetup -save -%%EndPageSetup -/125 pa -[2(BT mem,reg64 X64 )]681 code0 -[2(BT reg64,reg64 X64 )]670 code0 -[2(BT rm16,imm 386 )]659 code0 -[2(BT rm32,imm 386 )]648 code0 -[2(BT rm64,imm X64 )]637 code0 -[2(BTC mem,reg16 386 )]626 code0 -[2(BTC reg16,reg16 386 )]615 code0 -[2(BTC mem,reg32 386 )]604 code0 -[2(BTC reg32,reg32 386 )]593 code0 -[2(BTC mem,reg64 X64 )]582 code0 -[2(BTC reg64,reg64 X64 )]571 code0 -[2(BTC rm16,imm 386 )]560 code0 -[2(BTC rm32,imm 386 )]549 code0 -[2(BTC rm64,imm X64 )]538 code0 -[2(BTR mem,reg16 386 )]527 code0 -[2(BTR reg16,reg16 386 )]516 code0 -[2(BTR mem,reg32 386 )]505 code0 -[2(BTR reg32,reg32 386 )]494 code0 -[2(BTR mem,reg64 X64 )]483 code0 -[2(BTR reg64,reg64 X64 )]472 code0 -[2(BTR rm16,imm 386 )]461 code0 -[2(BTR rm32,imm 386 )]450 code0 -[2(BTR rm64,imm X64 )]439 code0 -[2(BTS mem,reg16 386 )]428 code0 -[2(BTS reg16,reg16 386 )]417 code0 -[2(BTS mem,reg32 386 )]406 code0 -[2(BTS reg32,reg32 386 )]395 code0 -[2(BTS mem,reg64 X64 )]384 code0 -[2(BTS reg64,reg64 X64 )]373 code0 -[2(BTS rm16,imm 386 )]362 code0 -[2(BTS rm32,imm 386 )]351 code0 -[2(BTS rm64,imm X64 )]340 code0 -[2(CALL imm 8086 )]329 code0 -[2(CALL imm|near 8086 )]318 code0 -[2(CALL imm|far 8086,ND,NOLONG )]307 code0 -[2(CALL imm16 8086 )]296 code0 -[2(CALL imm16|near 8086 )]285 code0 -[2(CALL imm16|far 8086,ND,NOLONG )]274 code0 -[2(CALL imm32 386 )]263 code0 -[2(CALL imm32|near 386 )]252 code0 -[2(CALL imm32|far 386,ND,NOLONG )]241 code0 -[2(CALL imm:imm 8086,NOLONG )]230 code0 -[2(CALL imm16:imm 8086,NOLONG )]219 code0 -[2(CALL imm:imm16 8086,NOLONG )]208 code0 -[2(CALL imm32:imm 386,NOLONG )]197 code0 -[2(CALL imm:imm32 386,NOLONG )]186 code0 -[2(CALL mem|far 8086,NOLONG )]175 code0 -[2(CALL mem|far X64 )]164 code0 -[2(CALL mem16|far 8086 )]153 code0 -[2(CALL mem32|far 386 )]142 code0 -[2(CALL mem64|far X64 )]131 code0 -[2(CALL mem|near 8086 )]120 code0 -[2(CALL mem16|near 8086 )]109 code0 -(125)pageodd -restore showpage -%%Page: 126 126 -%%BeginPageSetup -save -%%EndPageSetup -/126 pa -[2(CALL mem32|near 386,NOLONG )]681 code0 -[2(CALL mem64|near X64 )]670 code0 -[2(CALL reg16 8086 )]659 code0 -[2(CALL reg32 386,NOLONG )]648 code0 -[2(CALL reg64 X64 )]637 code0 -[2(CALL mem 8086 )]626 code0 -[2(CALL mem16 8086 )]615 code0 -[2(CALL mem32 386,NOLONG )]604 code0 -[2(CALL mem64 X64 )]593 code0 -[2(CBW 8086 )]582 code0 -[2(CDQ 386 )]571 code0 -[2(CDQE X64 )]560 code0 -[2(CLC 8086 )]549 code0 -[2(CLD 8086 )]538 code0 -[2(CLGI X64,AMD )]527 code0 -[2(CLI 8086 )]516 code0 -[2(CLTS 286,PRIV )]505 code0 -[2(CMC 8086 )]494 code0 -[2(CMP mem,reg8 8086 )]483 code0 -[2(CMP reg8,reg8 8086 )]472 code0 -[2(CMP mem,reg16 8086 )]461 code0 -[2(CMP reg16,reg16 8086 )]450 code0 -[2(CMP mem,reg32 386 )]439 code0 -[2(CMP reg32,reg32 386 )]428 code0 -[2(CMP mem,reg64 X64 )]417 code0 -[2(CMP reg64,reg64 X64 )]406 code0 -[2(CMP reg8,mem 8086 )]395 code0 -[2(CMP reg8,reg8 8086 )]384 code0 -[2(CMP reg16,mem 8086 )]373 code0 -[2(CMP reg16,reg16 8086 )]362 code0 -[2(CMP reg32,mem 386 )]351 code0 -[2(CMP reg32,reg32 386 )]340 code0 -[2(CMP reg64,mem X64 )]329 code0 -[2(CMP reg64,reg64 X64 )]318 code0 -[2(CMP rm16,imm8 8086 )]307 code0 -[2(CMP rm32,imm8 386 )]296 code0 -[2(CMP rm64,imm8 X64 )]285 code0 -[2(CMP reg_al,imm 8086 )]274 code0 -[2(CMP reg_ax,sbyte16 8086 )]263 code0 -[2(CMP reg_ax,imm 8086 )]252 code0 -[2(CMP reg_eax,sbyte32 386 )]241 code0 -[2(CMP reg_eax,imm 386 )]230 code0 -[2(CMP reg_rax,sbyte64 X64 )]219 code0 -[2(CMP reg_rax,imm X64 )]208 code0 -[2(CMP rm8,imm 8086 )]197 code0 -[2(CMP rm16,imm 8086 )]186 code0 -[2(CMP rm32,imm 386 )]175 code0 -[2(CMP rm64,imm X64 )]164 code0 -[2(CMP mem,imm8 8086 )]153 code0 -[2(CMP mem,imm16 8086 )]142 code0 -[2(CMP mem,imm32 386 )]131 code0 -[2(CMPSB 8086 )]120 code0 -[2(CMPSD 386 )]109 code0 -(126)pageeven -restore showpage -%%Page: 127 127 -%%BeginPageSetup -save -%%EndPageSetup -/127 pa -[2(CMPSQ X64 )]681 code0 -[2(CMPSW 8086 )]670 code0 -[2(CMPXCHG mem,reg8 PENT )]659 code0 -[2(CMPXCHG reg8,reg8 PENT )]648 code0 -[2(CMPXCHG mem,reg16 PENT )]637 code0 -[2(CMPXCHG reg16,reg16 PENT )]626 code0 -[2(CMPXCHG mem,reg32 PENT )]615 code0 -[2(CMPXCHG reg32,reg32 PENT )]604 code0 -[2(CMPXCHG mem,reg64 X64 )]593 code0 -[2(CMPXCHG reg64,reg64 X64 )]582 code0 -[2(CMPXCHG486 mem,reg8 486,UNDOC,ND )]571 code0 -[2(CMPXCHG486 reg8,reg8 486,UNDOC,ND )]560 code0 -[2(CMPXCHG486 mem,reg16 486,UNDOC,ND )]549 code0 -[2(CMPXCHG486 reg16,reg16 486,UNDOC,ND )]538 code0 -[2(CMPXCHG486 mem,reg32 486,UNDOC,ND )]527 code0 -[2(CMPXCHG486 reg32,reg32 486,UNDOC,ND )]516 code0 -[2(CMPXCHG8B mem PENT )]505 code0 -[2(CMPXCHG16B mem X64 )]494 code0 -[2(CPUID PENT )]483 code0 -[2(CPU_READ PENT,CYRIX )]472 code0 -[2(CPU_WRITE PENT,CYRIX )]461 code0 -[2(CQO X64 )]450 code0 -[2(CWD 8086 )]439 code0 -[2(CWDE 386 )]428 code0 -[2(DAA 8086,NOLONG )]417 code0 -[2(DAS 8086,NOLONG )]406 code0 -[2(DEC reg16 8086,NOLONG )]395 code0 -[2(DEC reg32 386,NOLONG )]384 code0 -[2(DEC rm8 8086 )]373 code0 -[2(DEC rm16 8086 )]362 code0 -[2(DEC rm32 386 )]351 code0 -[2(DEC rm64 X64 )]340 code0 -[2(DIV rm8 8086 )]329 code0 -[2(DIV rm16 8086 )]318 code0 -[2(DIV rm32 386 )]307 code0 -[2(DIV rm64 X64 )]296 code0 -[2(DMINT P6,CYRIX )]285 code0 -[2(EMMS PENT,MMX )]274 code0 -[2(ENTER imm,imm 186 )]263 code0 -[2(EQU imm 8086 )]252 code0 -[2(EQU imm:imm 8086 )]241 code0 -[2(F2XM1 8086,FPU )]230 code0 -[2(FABS 8086,FPU )]219 code0 -[2(FADD mem32 8086,FPU )]208 code0 -[2(FADD mem64 8086,FPU )]197 code0 -[2(FADD fpureg|to 8086,FPU )]186 code0 -[2(FADD fpureg 8086,FPU )]175 code0 -[2(FADD fpureg,fpu0 8086,FPU )]164 code0 -[2(FADD fpu0,fpureg 8086,FPU )]153 code0 -[2(FADD 8086,FPU,ND )]142 code0 -[2(FADDP fpureg 8086,FPU )]131 code0 -[2(FADDP fpureg,fpu0 8086,FPU )]120 code0 -[2(FADDP 8086,FPU,ND )]109 code0 -(127)pageodd -restore showpage -%%Page: 128 128 -%%BeginPageSetup -save -%%EndPageSetup -/128 pa -[2(FBLD mem80 8086,FPU )]681 code0 -[2(FBLD mem 8086,FPU )]670 code0 -[2(FBSTP mem80 8086,FPU )]659 code0 -[2(FBSTP mem 8086,FPU )]648 code0 -[2(FCHS 8086,FPU )]637 code0 -[2(FCLEX 8086,FPU )]626 code0 -[2(FCMOVB fpureg P6,FPU )]615 code0 -[2(FCMOVB fpu0,fpureg P6,FPU )]604 code0 -[2(FCMOVB P6,FPU,ND )]593 code0 -[2(FCMOVBE fpureg P6,FPU )]582 code0 -[2(FCMOVBE fpu0,fpureg P6,FPU )]571 code0 -[2(FCMOVBE P6,FPU,ND )]560 code0 -[2(FCMOVE fpureg P6,FPU )]549 code0 -[2(FCMOVE fpu0,fpureg P6,FPU )]538 code0 -[2(FCMOVE P6,FPU,ND )]527 code0 -[2(FCMOVNB fpureg P6,FPU )]516 code0 -[2(FCMOVNB fpu0,fpureg P6,FPU )]505 code0 -[2(FCMOVNB P6,FPU,ND )]494 code0 -[2(FCMOVNBE fpureg P6,FPU )]483 code0 -[2(FCMOVNBE fpu0,fpureg P6,FPU )]472 code0 -[2(FCMOVNBE P6,FPU,ND )]461 code0 -[2(FCMOVNE fpureg P6,FPU )]450 code0 -[2(FCMOVNE fpu0,fpureg P6,FPU )]439 code0 -[2(FCMOVNE P6,FPU,ND )]428 code0 -[2(FCMOVNU fpureg P6,FPU )]417 code0 -[2(FCMOVNU fpu0,fpureg P6,FPU )]406 code0 -[2(FCMOVNU P6,FPU,ND )]395 code0 -[2(FCMOVU fpureg P6,FPU )]384 code0 -[2(FCMOVU fpu0,fpureg P6,FPU )]373 code0 -[2(FCMOVU P6,FPU,ND )]362 code0 -[2(FCOM mem32 8086,FPU )]351 code0 -[2(FCOM mem64 8086,FPU )]340 code0 -[2(FCOM fpureg 8086,FPU )]329 code0 -[2(FCOM fpu0,fpureg 8086,FPU )]318 code0 -[2(FCOM 8086,FPU,ND )]307 code0 -[2(FCOMI fpureg P6,FPU )]296 code0 -[2(FCOMI fpu0,fpureg P6,FPU )]285 code0 -[2(FCOMI P6,FPU,ND )]274 code0 -[2(FCOMIP fpureg P6,FPU )]263 code0 -[2(FCOMIP fpu0,fpureg P6,FPU )]252 code0 -[2(FCOMIP P6,FPU,ND )]241 code0 -[2(FCOMP mem32 8086,FPU )]230 code0 -[2(FCOMP mem64 8086,FPU )]219 code0 -[2(FCOMP fpureg 8086,FPU )]208 code0 -[2(FCOMP fpu0,fpureg 8086,FPU )]197 code0 -[2(FCOMP 8086,FPU,ND )]186 code0 -[2(FCOMPP 8086,FPU )]175 code0 -[2(FCOS 386,FPU )]164 code0 -[2(FDECSTP 8086,FPU )]153 code0 -[2(FDISI 8086,FPU )]142 code0 -[2(FDIV mem32 8086,FPU )]131 code0 -[2(FDIV mem64 8086,FPU )]120 code0 -[2(FDIV fpureg|to 8086,FPU )]109 code0 -(128)pageeven -restore showpage -%%Page: 129 129 -%%BeginPageSetup -save -%%EndPageSetup -/129 pa -[2(FDIV fpureg 8086,FPU )]681 code0 -[2(FDIV fpureg,fpu0 8086,FPU )]670 code0 -[2(FDIV fpu0,fpureg 8086,FPU )]659 code0 -[2(FDIV 8086,FPU,ND )]648 code0 -[2(FDIVP fpureg 8086,FPU )]637 code0 -[2(FDIVP fpureg,fpu0 8086,FPU )]626 code0 -[2(FDIVP 8086,FPU,ND )]615 code0 -[2(FDIVR mem32 8086,FPU )]604 code0 -[2(FDIVR mem64 8086,FPU )]593 code0 -[2(FDIVR fpureg|to 8086,FPU )]582 code0 -[2(FDIVR fpureg,fpu0 8086,FPU )]571 code0 -[2(FDIVR fpureg 8086,FPU )]560 code0 -[2(FDIVR fpu0,fpureg 8086,FPU )]549 code0 -[2(FDIVR 8086,FPU,ND )]538 code0 -[2(FDIVRP fpureg 8086,FPU )]527 code0 -[2(FDIVRP fpureg,fpu0 8086,FPU )]516 code0 -[2(FDIVRP 8086,FPU,ND )]505 code0 -[2(FEMMS PENT,3DNOW )]494 code0 -[2(FENI 8086,FPU )]483 code0 -[2(FFREE fpureg 8086,FPU )]472 code0 -[2(FFREE 8086,FPU )]461 code0 -[2(FFREEP fpureg 286,FPU,UNDOC )]450 code0 -[2(FFREEP 286,FPU,UNDOC )]439 code0 -[2(FIADD mem32 8086,FPU )]428 code0 -[2(FIADD mem16 8086,FPU )]417 code0 -[2(FICOM mem32 8086,FPU )]406 code0 -[2(FICOM mem16 8086,FPU )]395 code0 -[2(FICOMP mem32 8086,FPU )]384 code0 -[2(FICOMP mem16 8086,FPU )]373 code0 -[2(FIDIV mem32 8086,FPU )]362 code0 -[2(FIDIV mem16 8086,FPU )]351 code0 -[2(FIDIVR mem32 8086,FPU )]340 code0 -[2(FIDIVR mem16 8086,FPU )]329 code0 -[2(FILD mem32 8086,FPU )]318 code0 -[2(FILD mem16 8086,FPU )]307 code0 -[2(FILD mem64 8086,FPU )]296 code0 -[2(FIMUL mem32 8086,FPU )]285 code0 -[2(FIMUL mem16 8086,FPU )]274 code0 -[2(FINCSTP 8086,FPU )]263 code0 -[2(FINIT 8086,FPU )]252 code0 -[2(FIST mem32 8086,FPU )]241 code0 -[2(FIST mem16 8086,FPU )]230 code0 -[2(FISTP mem32 8086,FPU )]219 code0 -[2(FISTP mem16 8086,FPU )]208 code0 -[2(FISTP mem64 8086,FPU )]197 code0 -[2(FISTTP mem16 PRESCOTT,FPU )]186 code0 -[2(FISTTP mem32 PRESCOTT,FPU )]175 code0 -[2(FISTTP mem64 PRESCOTT,FPU )]164 code0 -[2(FISUB mem32 8086,FPU )]153 code0 -[2(FISUB mem16 8086,FPU )]142 code0 -[2(FISUBR mem32 8086,FPU )]131 code0 -[2(FISUBR mem16 8086,FPU )]120 code0 -[2(FLD mem32 8086,FPU )]109 code0 -(129)pageodd -restore showpage -%%Page: 130 130 -%%BeginPageSetup -save -%%EndPageSetup -/130 pa -[2(FLD mem64 8086,FPU )]681 code0 -[2(FLD mem80 8086,FPU )]670 code0 -[2(FLD fpureg 8086,FPU )]659 code0 -[2(FLD 8086,FPU,ND )]648 code0 -[2(FLD1 8086,FPU )]637 code0 -[2(FLDCW mem 8086,FPU,SW )]626 code0 -[2(FLDENV mem 8086,FPU )]615 code0 -[2(FLDL2E 8086,FPU )]604 code0 -[2(FLDL2T 8086,FPU )]593 code0 -[2(FLDLG2 8086,FPU )]582 code0 -[2(FLDLN2 8086,FPU )]571 code0 -[2(FLDPI 8086,FPU )]560 code0 -[2(FLDZ 8086,FPU )]549 code0 -[2(FMUL mem32 8086,FPU )]538 code0 -[2(FMUL mem64 8086,FPU )]527 code0 -[2(FMUL fpureg|to 8086,FPU )]516 code0 -[2(FMUL fpureg,fpu0 8086,FPU )]505 code0 -[2(FMUL fpureg 8086,FPU )]494 code0 -[2(FMUL fpu0,fpureg 8086,FPU )]483 code0 -[2(FMUL 8086,FPU,ND )]472 code0 -[2(FMULP fpureg 8086,FPU )]461 code0 -[2(FMULP fpureg,fpu0 8086,FPU )]450 code0 -[2(FMULP 8086,FPU,ND )]439 code0 -[2(FNCLEX 8086,FPU )]428 code0 -[2(FNDISI 8086,FPU )]417 code0 -[2(FNENI 8086,FPU )]406 code0 -[2(FNINIT 8086,FPU )]395 code0 -[2(FNOP 8086,FPU )]384 code0 -[2(FNSAVE mem 8086,FPU )]373 code0 -[2(FNSTCW mem 8086,FPU,SW )]362 code0 -[2(FNSTENV mem 8086,FPU )]351 code0 -[2(FNSTSW mem 8086,FPU,SW )]340 code0 -[2(FNSTSW reg_ax 286,FPU )]329 code0 -[2(FPATAN 8086,FPU )]318 code0 -[2(FPREM 8086,FPU )]307 code0 -[2(FPREM1 386,FPU )]296 code0 -[2(FPTAN 8086,FPU )]285 code0 -[2(FRNDINT 8086,FPU )]274 code0 -[2(FRSTOR mem 8086,FPU )]263 code0 -[2(FSAVE mem 8086,FPU )]252 code0 -[2(FSCALE 8086,FPU )]241 code0 -[2(FSETPM 286,FPU )]230 code0 -[2(FSIN 386,FPU )]219 code0 -[2(FSINCOS 386,FPU )]208 code0 -[2(FSQRT 8086,FPU )]197 code0 -[2(FST mem32 8086,FPU )]186 code0 -[2(FST mem64 8086,FPU )]175 code0 -[2(FST fpureg 8086,FPU )]164 code0 -[2(FST 8086,FPU,ND )]153 code0 -[2(FSTCW mem 8086,FPU,SW )]142 code0 -[2(FSTENV mem 8086,FPU )]131 code0 -[2(FSTP mem32 8086,FPU )]120 code0 -[2(FSTP mem64 8086,FPU )]109 code0 -(130)pageeven -restore showpage -%%Page: 131 131 -%%BeginPageSetup -save -%%EndPageSetup -/131 pa -[2(FSTP mem80 8086,FPU )]681 code0 -[2(FSTP fpureg 8086,FPU )]670 code0 -[2(FSTP 8086,FPU,ND )]659 code0 -[2(FSTSW mem 8086,FPU,SW )]648 code0 -[2(FSTSW reg_ax 286,FPU )]637 code0 -[2(FSUB mem32 8086,FPU )]626 code0 -[2(FSUB mem64 8086,FPU )]615 code0 -[2(FSUB fpureg|to 8086,FPU )]604 code0 -[2(FSUB fpureg,fpu0 8086,FPU )]593 code0 -[2(FSUB fpureg 8086,FPU )]582 code0 -[2(FSUB fpu0,fpureg 8086,FPU )]571 code0 -[2(FSUB 8086,FPU,ND )]560 code0 -[2(FSUBP fpureg 8086,FPU )]549 code0 -[2(FSUBP fpureg,fpu0 8086,FPU )]538 code0 -[2(FSUBP 8086,FPU,ND )]527 code0 -[2(FSUBR mem32 8086,FPU )]516 code0 -[2(FSUBR mem64 8086,FPU )]505 code0 -[2(FSUBR fpureg|to 8086,FPU )]494 code0 -[2(FSUBR fpureg,fpu0 8086,FPU )]483 code0 -[2(FSUBR fpureg 8086,FPU )]472 code0 -[2(FSUBR fpu0,fpureg 8086,FPU )]461 code0 -[2(FSUBR 8086,FPU,ND )]450 code0 -[2(FSUBRP fpureg 8086,FPU )]439 code0 -[2(FSUBRP fpureg,fpu0 8086,FPU )]428 code0 -[2(FSUBRP 8086,FPU,ND )]417 code0 -[2(FTST 8086,FPU )]406 code0 -[2(FUCOM fpureg 386,FPU )]395 code0 -[2(FUCOM fpu0,fpureg 386,FPU )]384 code0 -[2(FUCOM 386,FPU,ND )]373 code0 -[2(FUCOMI fpureg P6,FPU )]362 code0 -[2(FUCOMI fpu0,fpureg P6,FPU )]351 code0 -[2(FUCOMI P6,FPU,ND )]340 code0 -[2(FUCOMIP fpureg P6,FPU )]329 code0 -[2(FUCOMIP fpu0,fpureg P6,FPU )]318 code0 -[2(FUCOMIP P6,FPU,ND )]307 code0 -[2(FUCOMP fpureg 386,FPU )]296 code0 -[2(FUCOMP fpu0,fpureg 386,FPU )]285 code0 -[2(FUCOMP 386,FPU,ND )]274 code0 -[2(FUCOMPP 386,FPU )]263 code0 -[2(FXAM 8086,FPU )]252 code0 -[2(FXCH fpureg 8086,FPU )]241 code0 -[2(FXCH fpureg,fpu0 8086,FPU )]230 code0 -[2(FXCH fpu0,fpureg 8086,FPU )]219 code0 -[2(FXCH 8086,FPU,ND )]208 code0 -[2(FXTRACT 8086,FPU )]197 code0 -[2(FYL2X 8086,FPU )]186 code0 -[2(FYL2XP1 8086,FPU )]175 code0 -[2(HLT 8086,PRIV )]164 code0 -[2(IBTS mem,reg16 386,SW,UNDOC,ND )]153 code0 -[2(IBTS reg16,reg16 386,UNDOC,ND )]142 code0 -[2(IBTS mem,reg32 386,SD,UNDOC,ND )]131 code0 -[2(IBTS reg32,reg32 386,UNDOC,ND )]120 code0 -[2(ICEBP 386,ND )]109 code0 -(131)pageodd -restore showpage -%%Page: 132 132 -%%BeginPageSetup -save -%%EndPageSetup -/132 pa -[2(IDIV rm8 8086 )]681 code0 -[2(IDIV rm16 8086 )]670 code0 -[2(IDIV rm32 386 )]659 code0 -[2(IDIV rm64 X64 )]648 code0 -[2(IMUL rm8 8086 )]637 code0 -[2(IMUL rm16 8086 )]626 code0 -[2(IMUL rm32 386 )]615 code0 -[2(IMUL rm64 X64 )]604 code0 -[2(IMUL reg16,mem 386 )]593 code0 -[2(IMUL reg16,reg16 386 )]582 code0 -[2(IMUL reg32,mem 386 )]571 code0 -[2(IMUL reg32,reg32 386 )]560 code0 -[2(IMUL reg64,mem X64 )]549 code0 -[2(IMUL reg64,reg64 X64 )]538 code0 -[2(IMUL reg16,mem,imm8 186 )]527 code0 -[2(IMUL reg16,mem,sbyte16 186,ND )]516 code0 -[2(IMUL reg16,mem,imm16 186 )]505 code0 -[2(IMUL reg16,mem,imm 186,ND )]494 code0 -[2(IMUL reg16,reg16,imm8 186 )]483 code0 -[2(IMUL reg16,reg16,sbyte16 186,ND )]472 code0 -[2(IMUL reg16,reg16,imm16 186 )]461 code0 -[2(IMUL reg16,reg16,imm 186,ND )]450 code0 -[2(IMUL reg32,mem,imm8 386 )]439 code0 -[2(IMUL reg32,mem,sbyte32 386,ND )]428 code0 -[2(IMUL reg32,mem,imm32 386 )]417 code0 -[2(IMUL reg32,mem,imm 386,ND )]406 code0 -[2(IMUL reg32,reg32,imm8 386 )]395 code0 -[2(IMUL reg32,reg32,sbyte32 386,ND )]384 code0 -[2(IMUL reg32,reg32,imm32 386 )]373 code0 -[2(IMUL reg32,reg32,imm 386,ND )]362 code0 -[2(IMUL reg64,mem,imm8 X64 )]351 code0 -[2(IMUL reg64,mem,sbyte64 X64,ND )]340 code0 -[2(IMUL reg64,mem,imm32 X64 )]329 code0 -[2(IMUL reg64,mem,imm X64,ND )]318 code0 -[2(IMUL reg64,reg64,imm8 X64 )]307 code0 -[2(IMUL reg64,reg64,sbyte64 X64,ND )]296 code0 -[2(IMUL reg64,reg64,imm32 X64 )]285 code0 -[2(IMUL reg64,reg64,imm X64,ND )]274 code0 -[2(IMUL reg16,imm8 186 )]263 code0 -[2(IMUL reg16,sbyte16 186,ND )]252 code0 -[2(IMUL reg16,imm16 186 )]241 code0 -[2(IMUL reg16,imm 186,ND )]230 code0 -[2(IMUL reg32,imm8 386 )]219 code0 -[2(IMUL reg32,sbyte32 386,ND )]208 code0 -[2(IMUL reg32,imm32 386 )]197 code0 -[2(IMUL reg32,imm 386,ND )]186 code0 -[2(IMUL reg64,imm8 X64 )]175 code0 -[2(IMUL reg64,sbyte64 X64,ND )]164 code0 -[2(IMUL reg64,imm32 X64 )]153 code0 -[2(IMUL reg64,imm X64,ND )]142 code0 -[2(IN reg_al,imm 8086 )]131 code0 -[2(IN reg_ax,imm 8086 )]120 code0 -[2(IN reg_eax,imm 386 )]109 code0 -(132)pageeven -restore showpage -%%Page: 133 133 -%%BeginPageSetup -save -%%EndPageSetup -/133 pa -[2(IN reg_al,reg_dx 8086 )]681 code0 -[2(IN reg_ax,reg_dx 8086 )]670 code0 -[2(IN reg_eax,reg_dx 386 )]659 code0 -[2(INC reg16 8086,NOLONG )]648 code0 -[2(INC reg32 386,NOLONG )]637 code0 -[2(INC rm8 8086 )]626 code0 -[2(INC rm16 8086 )]615 code0 -[2(INC rm32 386 )]604 code0 -[2(INC rm64 X64 )]593 code0 -[2(INCBIN )]582 code0 -[2(INSB 186 )]571 code0 -[2(INSD 386 )]560 code0 -[2(INSW 186 )]549 code0 -[2(INT imm 8086 )]538 code0 -[2(INT01 386,ND )]527 code0 -[2(INT1 386 )]516 code0 -[2(INT03 8086,ND )]505 code0 -[2(INT3 8086 )]494 code0 -[2(INTO 8086,NOLONG )]483 code0 -[2(INVD 486,PRIV )]472 code0 -[2(INVLPG mem 486,PRIV )]461 code0 -[2(INVLPGA reg_ax,reg_ecx X86_64,AMD,NOLONG )]450 code0 -[2(INVLPGA reg_eax,reg_ecx X86_64,AMD )]439 code0 -[2(INVLPGA reg_rax,reg_ecx X64,AMD )]428 code0 -[2(INVLPGA X86_64,AMD )]417 code0 -[2(IRET 8086 )]406 code0 -[2(IRETD 386 )]395 code0 -[2(IRETQ X64 )]384 code0 -[2(IRETW 8086 )]373 code0 -[2(JCXZ imm 8086,NOLONG )]362 code0 -[2(JECXZ imm 386 )]351 code0 -[2(JRCXZ imm X64 )]340 code0 -[2(JMP imm|short 8086 )]329 code0 -[2(JMP imm 8086,ND )]318 code0 -[2(JMP imm 8086 )]307 code0 -[2(JMP imm|near 8086,ND )]296 code0 -[2(JMP imm|far 8086,ND,NOLONG )]285 code0 -[2(JMP imm16 8086 )]274 code0 -[2(JMP imm16|near 8086,ND )]263 code0 -[2(JMP imm16|far 8086,ND,NOLONG )]252 code0 -[2(JMP imm32 386 )]241 code0 -[2(JMP imm32|near 386,ND )]230 code0 -[2(JMP imm32|far 386,ND,NOLONG )]219 code0 -[2(JMP imm:imm 8086,NOLONG )]208 code0 -[2(JMP imm16:imm 8086,NOLONG )]197 code0 -[2(JMP imm:imm16 8086,NOLONG )]186 code0 -[2(JMP imm32:imm 386,NOLONG )]175 code0 -[2(JMP imm:imm32 386,NOLONG )]164 code0 -[2(JMP mem|far 8086,NOLONG )]153 code0 -[2(JMP mem|far X64 )]142 code0 -[2(JMP mem16|far 8086 )]131 code0 -[2(JMP mem32|far 386 )]120 code0 -[2(JMP mem64|far X64 )]109 code0 -(133)pageodd -restore showpage -%%Page: 134 134 -%%BeginPageSetup -save -%%EndPageSetup -/134 pa -[2(JMP mem|near 8086 )]681 code0 -[2(JMP mem16|near 8086 )]670 code0 -[2(JMP mem32|near 386,NOLONG )]659 code0 -[2(JMP mem64|near X64 )]648 code0 -[2(JMP reg16 8086 )]637 code0 -[2(JMP reg32 386,NOLONG )]626 code0 -[2(JMP reg64 X64 )]615 code0 -[2(JMP mem 8086 )]604 code0 -[2(JMP mem16 8086 )]593 code0 -[2(JMP mem32 386,NOLONG )]582 code0 -[2(JMP mem64 X64 )]571 code0 -[2(JMPE imm IA64 )]560 code0 -[2(JMPE imm16 IA64 )]549 code0 -[2(JMPE imm32 IA64 )]538 code0 -[2(JMPE rm16 IA64 )]527 code0 -[2(JMPE rm32 IA64 )]516 code0 -[2(LAHF 8086 )]505 code0 -[2(LAR reg16,mem 286,PROT,SW )]494 code0 -[2(LAR reg16,reg16 286,PROT )]483 code0 -[2(LAR reg16,reg32 386,PROT )]472 code0 -[2(LAR reg16,reg64 X64,PROT,ND )]461 code0 -[2(LAR reg32,mem 386,PROT,SW )]450 code0 -[2(LAR reg32,reg16 386,PROT )]439 code0 -[2(LAR reg32,reg32 386,PROT )]428 code0 -[2(LAR reg32,reg64 X64,PROT,ND )]417 code0 -[2(LAR reg64,mem X64,PROT,SW )]406 code0 -[2(LAR reg64,reg16 X64,PROT )]395 code0 -[2(LAR reg64,reg32 X64,PROT )]384 code0 -[2(LAR reg64,reg64 X64,PROT )]373 code0 -[2(LDS reg16,mem 8086,NOLONG )]362 code0 -[2(LDS reg32,mem 386,NOLONG )]351 code0 -[2(LEA reg16,mem 8086 )]340 code0 -[2(LEA reg32,mem 386 )]329 code0 -[2(LEA reg64,mem X64 )]318 code0 -[2(LEAVE 186 )]307 code0 -[2(LES reg16,mem 8086,NOLONG )]296 code0 -[2(LES reg32,mem 386,NOLONG )]285 code0 -[2(LFENCE X64,AMD )]274 code0 -[2(LFS reg16,mem 386 )]263 code0 -[2(LFS reg32,mem 386 )]252 code0 -[2(LGDT mem 286,PRIV )]241 code0 -[2(LGS reg16,mem 386 )]230 code0 -[2(LGS reg32,mem 386 )]219 code0 -[2(LIDT mem 286,PRIV )]208 code0 -[2(LLDT mem 286,PROT,PRIV )]197 code0 -[2(LLDT mem16 286,PROT,PRIV )]186 code0 -[2(LLDT reg16 286,PROT,PRIV )]175 code0 -[2(LMSW mem 286,PRIV )]164 code0 -[2(LMSW mem16 286,PRIV )]153 code0 -[2(LMSW reg16 286,PRIV )]142 code0 -[2(LOADALL 386,UNDOC )]131 code0 -[2(LOADALL286 286,UNDOC )]120 code0 -[2(LODSB 8086 )]109 code0 -(134)pageeven -restore showpage -%%Page: 135 135 -%%BeginPageSetup -save -%%EndPageSetup -/135 pa -[2(LODSD 386 )]681 code0 -[2(LODSQ X64 )]670 code0 -[2(LODSW 8086 )]659 code0 -[2(LOOP imm 8086 )]648 code0 -[2(LOOP imm,reg_cx 8086,NOLONG )]637 code0 -[2(LOOP imm,reg_ecx 386 )]626 code0 -[2(LOOP imm,reg_rcx X64 )]615 code0 -[2(LOOPE imm 8086 )]604 code0 -[2(LOOPE imm,reg_cx 8086,NOLONG )]593 code0 -[2(LOOPE imm,reg_ecx 386 )]582 code0 -[2(LOOPE imm,reg_rcx X64 )]571 code0 -[2(LOOPNE imm 8086 )]560 code0 -[2(LOOPNE imm,reg_cx 8086,NOLONG )]549 code0 -[2(LOOPNE imm,reg_ecx 386 )]538 code0 -[2(LOOPNE imm,reg_rcx X64 )]527 code0 -[2(LOOPNZ imm 8086 )]516 code0 -[2(LOOPNZ imm,reg_cx 8086,NOLONG )]505 code0 -[2(LOOPNZ imm,reg_ecx 386 )]494 code0 -[2(LOOPNZ imm,reg_rcx X64 )]483 code0 -[2(LOOPZ imm 8086 )]472 code0 -[2(LOOPZ imm,reg_cx 8086,NOLONG )]461 code0 -[2(LOOPZ imm,reg_ecx 386 )]450 code0 -[2(LOOPZ imm,reg_rcx X64 )]439 code0 -[2(LSL reg16,mem 286,PROT,SW )]428 code0 -[2(LSL reg16,reg16 286,PROT )]417 code0 -[2(LSL reg16,reg32 386,PROT )]406 code0 -[2(LSL reg16,reg64 X64,PROT,ND )]395 code0 -[2(LSL reg32,mem 386,PROT,SW )]384 code0 -[2(LSL reg32,reg16 386,PROT )]373 code0 -[2(LSL reg32,reg32 386,PROT )]362 code0 -[2(LSL reg32,reg64 X64,PROT,ND )]351 code0 -[2(LSL reg64,mem X64,PROT,SW )]340 code0 -[2(LSL reg64,reg16 X64,PROT )]329 code0 -[2(LSL reg64,reg32 X64,PROT )]318 code0 -[2(LSL reg64,reg64 X64,PROT )]307 code0 -[2(LSS reg16,mem 386 )]296 code0 -[2(LSS reg32,mem 386 )]285 code0 -[2(LTR mem 286,PROT,PRIV )]274 code0 -[2(LTR mem16 286,PROT,PRIV )]263 code0 -[2(LTR reg16 286,PROT,PRIV )]252 code0 -[2(MFENCE X64,AMD )]241 code0 -[2(MONITOR PRESCOTT )]230 code0 -[2(MONITOR reg_eax,reg_ecx,reg_edx PRESCOTT,ND )]219 code0 -[2(MONITOR reg_rax,reg_ecx,reg_edx X64,ND )]208 code0 -[2(MOV mem,reg_sreg 8086 )]197 code0 -[2(MOV reg16,reg_sreg 8086 )]186 code0 -[2(MOV reg32,reg_sreg 386 )]175 code0 -[2(MOV reg_sreg,mem 8086 )]164 code0 -[2(MOV reg_sreg,reg16 8086 )]153 code0 -[2(MOV reg_sreg,reg32 386 )]142 code0 -[2(MOV reg_al,mem_offs 8086 )]131 code0 -[2(MOV reg_ax,mem_offs 8086 )]120 code0 -[2(MOV reg_eax,mem_offs 386 )]109 code0 -(135)pageodd -restore showpage -%%Page: 136 136 -%%BeginPageSetup -save -%%EndPageSetup -/136 pa -[2(MOV reg_rax,mem_offs X64 )]681 code0 -[2(MOV mem_offs,reg_al 8086 )]670 code0 -[2(MOV mem_offs,reg_ax 8086 )]659 code0 -[2(MOV mem_offs,reg_eax 386 )]648 code0 -[2(MOV mem_offs,reg_rax X64 )]637 code0 -[2(MOV reg32,reg_creg 386,PRIV,NOLONG )]626 code0 -[2(MOV reg64,reg_creg X64,PRIV )]615 code0 -[2(MOV reg_creg,reg32 386,PRIV,NOLONG )]604 code0 -[2(MOV reg_creg,reg64 X64,PRIV )]593 code0 -[2(MOV reg32,reg_dreg 386,PRIV,NOLONG )]582 code0 -[2(MOV reg64,reg_dreg X64,PRIV )]571 code0 -[2(MOV reg_dreg,reg32 386,PRIV,NOLONG )]560 code0 -[2(MOV reg_dreg,reg64 X64,PRIV )]549 code0 -[2(MOV reg32,reg_treg 386,NOLONG,ND )]538 code0 -[2(MOV reg_treg,reg32 386,NOLONG,ND )]527 code0 -[2(MOV mem,reg8 8086 )]516 code0 -[2(MOV reg8,reg8 8086 )]505 code0 -[2(MOV mem,reg16 8086 )]494 code0 -[2(MOV reg16,reg16 8086 )]483 code0 -[2(MOV mem,reg32 386 )]472 code0 -[2(MOV reg32,reg32 386 )]461 code0 -[2(MOV mem,reg64 X64 )]450 code0 -[2(MOV reg64,reg64 X64 )]439 code0 -[2(MOV reg8,mem 8086 )]428 code0 -[2(MOV reg8,reg8 8086 )]417 code0 -[2(MOV reg16,mem 8086 )]406 code0 -[2(MOV reg16,reg16 8086 )]395 code0 -[2(MOV reg32,mem 386 )]384 code0 -[2(MOV reg32,reg32 386 )]373 code0 -[2(MOV reg64,mem X64 )]362 code0 -[2(MOV reg64,reg64 X64 )]351 code0 -[2(MOV reg8,imm 8086 )]340 code0 -[2(MOV reg16,imm 8086 )]329 code0 -[2(MOV reg32,imm 386 )]318 code0 -[2(MOV reg64,imm X64 )]307 code0 -[2(MOV reg64,imm32 X64 )]296 code0 -[2(MOV rm8,imm 8086 )]285 code0 -[2(MOV rm16,imm 8086 )]274 code0 -[2(MOV rm32,imm 386 )]263 code0 -[2(MOV rm64,imm X64 )]252 code0 -[2(MOV mem,imm8 8086 )]241 code0 -[2(MOV mem,imm16 8086 )]230 code0 -[2(MOV mem,imm32 386 )]219 code0 -[2(MOVD mmxreg,mem PENT,MMX,SD )]208 code0 -[2(MOVD mmxreg,reg32 PENT,MMX )]197 code0 -[2(MOVD mem,mmxreg PENT,MMX,SD )]186 code0 -[2(MOVD reg32,mmxreg PENT,MMX )]175 code0 -[2(MOVD xmmreg,mem X64,SD )]164 code0 -[2(MOVD xmmreg,reg32 X64 )]153 code0 -[2(MOVD mem,xmmreg X64,SD )]142 code0 -[2(MOVD reg32,xmmreg X64,SSE )]131 code0 -[2(MOVQ mmxreg,mmxrm PENT,MMX )]120 code0 -[2(MOVQ mmxrm,mmxreg PENT,MMX )]109 code0 -(136)pageeven -restore showpage -%%Page: 137 137 -%%BeginPageSetup -save -%%EndPageSetup -/137 pa -[2(MOVQ mmxreg,rm64 X64,MMX )]681 code0 -[2(MOVQ rm64,mmxreg X64,MMX )]670 code0 -[2(MOVSB 8086 )]659 code0 -[2(MOVSD 386 )]648 code0 -[2(MOVSQ X64 )]637 code0 -[2(MOVSW 8086 )]626 code0 -[2(MOVSX reg16,mem 386 )]615 code0 -[2(MOVSX reg16,reg8 386 )]604 code0 -[2(MOVSX reg32,rm8 386 )]593 code0 -[2(MOVSX reg32,rm16 386 )]582 code0 -[2(MOVSX reg64,rm8 X64 )]571 code0 -[2(MOVSX reg64,rm16 X64 )]560 code0 -[2(MOVSXD reg64,rm32 X64 )]549 code0 -[2(MOVSX reg64,rm32 X64,ND )]538 code0 -[2(MOVZX reg16,mem 386 )]527 code0 -[2(MOVZX reg16,reg8 386 )]516 code0 -[2(MOVZX reg32,rm8 386 )]505 code0 -[2(MOVZX reg32,rm16 386 )]494 code0 -[2(MOVZX reg64,rm8 X64 )]483 code0 -[2(MOVZX reg64,rm16 X64 )]472 code0 -[2(MUL rm8 8086 )]461 code0 -[2(MUL rm16 8086 )]450 code0 -[2(MUL rm32 386 )]439 code0 -[2(MUL rm64 X64 )]428 code0 -[2(MWAIT PRESCOTT )]417 code0 -[2(MWAIT reg_eax,reg_ecx PRESCOTT,ND )]406 code0 -[2(NEG rm8 8086 )]395 code0 -[2(NEG rm16 8086 )]384 code0 -[2(NEG rm32 386 )]373 code0 -[2(NEG rm64 X64 )]362 code0 -[2(NOP 8086 )]351 code0 -[2(NOP rm16 P6 )]340 code0 -[2(NOP rm32 P6 )]329 code0 -[2(NOP rm64 X64 )]318 code0 -[2(NOT rm8 8086 )]307 code0 -[2(NOT rm16 8086 )]296 code0 -[2(NOT rm32 386 )]285 code0 -[2(NOT rm64 X64 )]274 code0 -[2(OR mem,reg8 8086 )]263 code0 -[2(OR reg8,reg8 8086 )]252 code0 -[2(OR mem,reg16 8086 )]241 code0 -[2(OR reg16,reg16 8086 )]230 code0 -[2(OR mem,reg32 386 )]219 code0 -[2(OR reg32,reg32 386 )]208 code0 -[2(OR mem,reg64 X64 )]197 code0 -[2(OR reg64,reg64 X64 )]186 code0 -[2(OR reg8,mem 8086 )]175 code0 -[2(OR reg8,reg8 8086 )]164 code0 -[2(OR reg16,mem 8086 )]153 code0 -[2(OR reg16,reg16 8086 )]142 code0 -[2(OR reg32,mem 386 )]131 code0 -[2(OR reg32,reg32 386 )]120 code0 -[2(OR reg64,mem X64 )]109 code0 -(137)pageodd -restore showpage -%%Page: 138 138 -%%BeginPageSetup -save -%%EndPageSetup -/138 pa -[2(OR reg64,reg64 X64 )]681 code0 -[2(OR rm16,imm8 8086 )]670 code0 -[2(OR rm32,imm8 386 )]659 code0 -[2(OR rm64,imm8 X64 )]648 code0 -[2(OR reg_al,imm 8086 )]637 code0 -[2(OR reg_ax,sbyte16 8086 )]626 code0 -[2(OR reg_ax,imm 8086 )]615 code0 -[2(OR reg_eax,sbyte32 386 )]604 code0 -[2(OR reg_eax,imm 386 )]593 code0 -[2(OR reg_rax,sbyte64 X64 )]582 code0 -[2(OR reg_rax,imm X64 )]571 code0 -[2(OR rm8,imm 8086 )]560 code0 -[2(OR rm16,imm 8086 )]549 code0 -[2(OR rm32,imm 386 )]538 code0 -[2(OR rm64,imm X64 )]527 code0 -[2(OR mem,imm8 8086 )]516 code0 -[2(OR mem,imm16 8086 )]505 code0 -[2(OR mem,imm32 386 )]494 code0 -[2(OUT imm,reg_al 8086 )]483 code0 -[2(OUT imm,reg_ax 8086 )]472 code0 -[2(OUT imm,reg_eax 386 )]461 code0 -[2(OUT reg_dx,reg_al 8086 )]450 code0 -[2(OUT reg_dx,reg_ax 8086 )]439 code0 -[2(OUT reg_dx,reg_eax 386 )]428 code0 -[2(OUTSB 186 )]417 code0 -[2(OUTSD 386 )]406 code0 -[2(OUTSW 186 )]395 code0 -[2(PACKSSDW mmxreg,mmxrm PENT,MMX )]384 code0 -[2(PACKSSWB mmxreg,mmxrm PENT,MMX )]373 code0 -[2(PACKUSWB mmxreg,mmxrm PENT,MMX )]362 code0 -[2(PADDB mmxreg,mmxrm PENT,MMX )]351 code0 -[2(PADDD mmxreg,mmxrm PENT,MMX )]340 code0 -[2(PADDSB mmxreg,mmxrm PENT,MMX )]329 code0 -[2(PADDSIW mmxreg,mmxrm PENT,MMX,CYRIX )]318 code0 -[2(PADDSW mmxreg,mmxrm PENT,MMX )]307 code0 -[2(PADDUSB mmxreg,mmxrm PENT,MMX )]296 code0 -[2(PADDUSW mmxreg,mmxrm PENT,MMX )]285 code0 -[2(PADDW mmxreg,mmxrm PENT,MMX )]274 code0 -[2(PAND mmxreg,mmxrm PENT,MMX )]263 code0 -[2(PANDN mmxreg,mmxrm PENT,MMX )]252 code0 -[2(PAUSE 8086 )]241 code0 -[2(PAVEB mmxreg,mmxrm PENT,MMX,CYRIX )]230 code0 -[2(PAVGUSB mmxreg,mmxrm PENT,3DNOW )]219 code0 -[2(PCMPEQB mmxreg,mmxrm PENT,MMX )]208 code0 -[2(PCMPEQD mmxreg,mmxrm PENT,MMX )]197 code0 -[2(PCMPEQW mmxreg,mmxrm PENT,MMX )]186 code0 -[2(PCMPGTB mmxreg,mmxrm PENT,MMX )]175 code0 -[2(PCMPGTD mmxreg,mmxrm PENT,MMX )]164 code0 -[2(PCMPGTW mmxreg,mmxrm PENT,MMX )]153 code0 -[2(PDISTIB mmxreg,mem PENT,MMX,CYRIX )]142 code0 -[2(PF2ID mmxreg,mmxrm PENT,3DNOW )]131 code0 -[2(PFACC mmxreg,mmxrm PENT,3DNOW )]120 code0 -[2(PFADD mmxreg,mmxrm PENT,3DNOW )]109 code0 -(138)pageeven -restore showpage -%%Page: 139 139 -%%BeginPageSetup -save -%%EndPageSetup -/139 pa -[2(PFCMPEQ mmxreg,mmxrm PENT,3DNOW )]681 code0 -[2(PFCMPGE mmxreg,mmxrm PENT,3DNOW )]670 code0 -[2(PFCMPGT mmxreg,mmxrm PENT,3DNOW )]659 code0 -[2(PFMAX mmxreg,mmxrm PENT,3DNOW )]648 code0 -[2(PFMIN mmxreg,mmxrm PENT,3DNOW )]637 code0 -[2(PFMUL mmxreg,mmxrm PENT,3DNOW )]626 code0 -[2(PFRCP mmxreg,mmxrm PENT,3DNOW )]615 code0 -[2(PFRCPIT1 mmxreg,mmxrm PENT,3DNOW )]604 code0 -[2(PFRCPIT2 mmxreg,mmxrm PENT,3DNOW )]593 code0 -[2(PFRSQIT1 mmxreg,mmxrm PENT,3DNOW )]582 code0 -[2(PFRSQRT mmxreg,mmxrm PENT,3DNOW )]571 code0 -[2(PFSUB mmxreg,mmxrm PENT,3DNOW )]560 code0 -[2(PFSUBR mmxreg,mmxrm PENT,3DNOW )]549 code0 -[2(PI2FD mmxreg,mmxrm PENT,3DNOW )]538 code0 -[2(PMACHRIW mmxreg,mem PENT,MMX,CYRIX )]527 code0 -[2(PMADDWD mmxreg,mmxrm PENT,MMX )]516 code0 -[2(PMAGW mmxreg,mmxrm PENT,MMX,CYRIX )]505 code0 -[2(PMULHRIW mmxreg,mmxrm PENT,MMX,CYRIX )]494 code0 -[2(PMULHRWA mmxreg,mmxrm PENT,3DNOW )]483 code0 -[2(PMULHRWC mmxreg,mmxrm PENT,MMX,CYRIX )]472 code0 -[2(PMULHW mmxreg,mmxrm PENT,MMX )]461 code0 -[2(PMULLW mmxreg,mmxrm PENT,MMX )]450 code0 -[2(PMVGEZB mmxreg,mem PENT,MMX,CYRIX )]439 code0 -[2(PMVLZB mmxreg,mem PENT,MMX,CYRIX )]428 code0 -[2(PMVNZB mmxreg,mem PENT,MMX,CYRIX )]417 code0 -[2(PMVZB mmxreg,mem PENT,MMX,CYRIX )]406 code0 -[2(POP reg16 8086 )]395 code0 -[2(POP reg32 386,NOLONG )]384 code0 -[2(POP reg64 X64 )]373 code0 -[2(POP rm16 8086 )]362 code0 -[2(POP rm32 386,NOLONG )]351 code0 -[2(POP rm64 X64 )]340 code0 -[2(POP reg_cs 8086,UNDOC,ND )]329 code0 -[2(POP reg_dess 8086,NOLONG )]318 code0 -[2(POP reg_fsgs 386 )]307 code0 -[2(POPA 186,NOLONG )]296 code0 -[2(POPAD 386,NOLONG )]285 code0 -[2(POPAW 186,NOLONG )]274 code0 -[2(POPF 8086 )]263 code0 -[2(POPFD 386,NOLONG )]252 code0 -[2(POPFQ X64 )]241 code0 -[2(POPFW 8086 )]230 code0 -[2(POR mmxreg,mmxrm PENT,MMX )]219 code0 -[2(PREFETCH mem PENT,3DNOW )]208 code0 -[2(PREFETCHW mem PENT,3DNOW )]197 code0 -[2(PSLLD mmxreg,mmxrm PENT,MMX )]186 code0 -[2(PSLLD mmxreg,imm PENT,MMX )]175 code0 -[2(PSLLQ mmxreg,mmxrm PENT,MMX )]164 code0 -[2(PSLLQ mmxreg,imm PENT,MMX )]153 code0 -[2(PSLLW mmxreg,mmxrm PENT,MMX )]142 code0 -[2(PSLLW mmxreg,imm PENT,MMX )]131 code0 -[2(PSRAD mmxreg,mmxrm PENT,MMX )]120 code0 -[2(PSRAD mmxreg,imm PENT,MMX )]109 code0 -(139)pageodd -restore showpage -%%Page: 140 140 -%%BeginPageSetup -save -%%EndPageSetup -/140 pa -[2(PSRAW mmxreg,mmxrm PENT,MMX )]681 code0 -[2(PSRAW mmxreg,imm PENT,MMX )]670 code0 -[2(PSRLD mmxreg,mmxrm PENT,MMX )]659 code0 -[2(PSRLD mmxreg,imm PENT,MMX )]648 code0 -[2(PSRLQ mmxreg,mmxrm PENT,MMX )]637 code0 -[2(PSRLQ mmxreg,imm PENT,MMX )]626 code0 -[2(PSRLW mmxreg,mmxrm PENT,MMX )]615 code0 -[2(PSRLW mmxreg,imm PENT,MMX )]604 code0 -[2(PSUBB mmxreg,mmxrm PENT,MMX )]593 code0 -[2(PSUBD mmxreg,mmxrm PENT,MMX )]582 code0 -[2(PSUBSB mmxreg,mmxrm PENT,MMX )]571 code0 -[2(PSUBSIW mmxreg,mmxrm PENT,MMX,CYRIX )]560 code0 -[2(PSUBSW mmxreg,mmxrm PENT,MMX )]549 code0 -[2(PSUBUSB mmxreg,mmxrm PENT,MMX )]538 code0 -[2(PSUBUSW mmxreg,mmxrm PENT,MMX )]527 code0 -[2(PSUBW mmxreg,mmxrm PENT,MMX )]516 code0 -[2(PUNPCKHBW mmxreg,mmxrm PENT,MMX )]505 code0 -[2(PUNPCKHDQ mmxreg,mmxrm PENT,MMX )]494 code0 -[2(PUNPCKHWD mmxreg,mmxrm PENT,MMX )]483 code0 -[2(PUNPCKLBW mmxreg,mmxrm PENT,MMX )]472 code0 -[2(PUNPCKLDQ mmxreg,mmxrm PENT,MMX )]461 code0 -[2(PUNPCKLWD mmxreg,mmxrm PENT,MMX )]450 code0 -[2(PUSH reg16 8086 )]439 code0 -[2(PUSH reg32 386,NOLONG )]428 code0 -[2(PUSH reg64 X64 )]417 code0 -[2(PUSH rm16 8086 )]406 code0 -[2(PUSH rm32 386,NOLONG )]395 code0 -[2(PUSH rm64 X64 )]384 code0 -[2(PUSH reg_cs 8086,NOLONG )]373 code0 -[2(PUSH reg_dess 8086,NOLONG )]362 code0 -[2(PUSH reg_fsgs 386 )]351 code0 -[2(PUSH imm8 186 )]340 code0 -[2(PUSH imm16 186,AR0,SZ )]329 code0 -[2(PUSH imm32 386,NOLONG,AR0,SZ )]318 code0 -[2(PUSH imm32 386,NOLONG,SD )]307 code0 -[2(PUSH imm64 X64,AR0,SZ )]296 code0 -[2(PUSHA 186,NOLONG )]285 code0 -[2(PUSHAD 386,NOLONG )]274 code0 -[2(PUSHAW 186,NOLONG )]263 code0 -[2(PUSHF 8086 )]252 code0 -[2(PUSHFD 386,NOLONG )]241 code0 -[2(PUSHFQ X64 )]230 code0 -[2(PUSHFW 8086 )]219 code0 -[2(PXOR mmxreg,mmxrm PENT,MMX )]208 code0 -[2(RCL rm8,unity 8086 )]197 code0 -[2(RCL rm8,reg_cl 8086 )]186 code0 -[2(RCL rm8,imm 186 )]175 code0 -[2(RCL rm16,unity 8086 )]164 code0 -[2(RCL rm16,reg_cl 8086 )]153 code0 -[2(RCL rm16,imm 186 )]142 code0 -[2(RCL rm32,unity 386 )]131 code0 -[2(RCL rm32,reg_cl 386 )]120 code0 -[2(RCL rm32,imm 386 )]109 code0 -(140)pageeven -restore showpage -%%Page: 141 141 -%%BeginPageSetup -save -%%EndPageSetup -/141 pa -[2(RCL rm64,unity X64 )]681 code0 -[2(RCL rm64,reg_cl X64 )]670 code0 -[2(RCL rm64,imm X64 )]659 code0 -[2(RCR rm8,unity 8086 )]648 code0 -[2(RCR rm8,reg_cl 8086 )]637 code0 -[2(RCR rm8,imm 186 )]626 code0 -[2(RCR rm16,unity 8086 )]615 code0 -[2(RCR rm16,reg_cl 8086 )]604 code0 -[2(RCR rm16,imm 186 )]593 code0 -[2(RCR rm32,unity 386 )]582 code0 -[2(RCR rm32,reg_cl 386 )]571 code0 -[2(RCR rm32,imm 386 )]560 code0 -[2(RCR rm64,unity X64 )]549 code0 -[2(RCR rm64,reg_cl X64 )]538 code0 -[2(RCR rm64,imm X64 )]527 code0 -[2(RDSHR rm32 P6,CYRIXM )]516 code0 -[2(RDMSR PENT,PRIV )]505 code0 -[2(RDPMC P6 )]494 code0 -[2(RDTSC PENT )]483 code0 -[2(RDTSCP X86_64 )]472 code0 -[2(RET 8086 )]461 code0 -[2(RET imm 8086,SW )]450 code0 -[2(RETF 8086 )]439 code0 -[2(RETF imm 8086,SW )]428 code0 -[2(RETN 8086 )]417 code0 -[2(RETN imm 8086,SW )]406 code0 -[2(ROL rm8,unity 8086 )]395 code0 -[2(ROL rm8,reg_cl 8086 )]384 code0 -[2(ROL rm8,imm 186 )]373 code0 -[2(ROL rm16,unity 8086 )]362 code0 -[2(ROL rm16,reg_cl 8086 )]351 code0 -[2(ROL rm16,imm 186 )]340 code0 -[2(ROL rm32,unity 386 )]329 code0 -[2(ROL rm32,reg_cl 386 )]318 code0 -[2(ROL rm32,imm 386 )]307 code0 -[2(ROL rm64,unity X64 )]296 code0 -[2(ROL rm64,reg_cl X64 )]285 code0 -[2(ROL rm64,imm X64 )]274 code0 -[2(ROR rm8,unity 8086 )]263 code0 -[2(ROR rm8,reg_cl 8086 )]252 code0 -[2(ROR rm8,imm 186 )]241 code0 -[2(ROR rm16,unity 8086 )]230 code0 -[2(ROR rm16,reg_cl 8086 )]219 code0 -[2(ROR rm16,imm 186 )]208 code0 -[2(ROR rm32,unity 386 )]197 code0 -[2(ROR rm32,reg_cl 386 )]186 code0 -[2(ROR rm32,imm 386 )]175 code0 -[2(ROR rm64,unity X64 )]164 code0 -[2(ROR rm64,reg_cl X64 )]153 code0 -[2(ROR rm64,imm X64 )]142 code0 -[2(RDM P6,CYRIX,ND )]131 code0 -[2(RSDC reg_sreg,mem80 486,CYRIXM )]120 code0 -[2(RSLDT mem80 486,CYRIXM )]109 code0 -(141)pageodd -restore showpage -%%Page: 142 142 -%%BeginPageSetup -save -%%EndPageSetup -/142 pa -[2(RSM PENTM )]681 code0 -[2(RSTS mem80 486,CYRIXM )]670 code0 -[2(SAHF 8086 )]659 code0 -[2(SAL rm8,unity 8086,ND )]648 code0 -[2(SAL rm8,reg_cl 8086,ND )]637 code0 -[2(SAL rm8,imm 186,ND )]626 code0 -[2(SAL rm16,unity 8086,ND )]615 code0 -[2(SAL rm16,reg_cl 8086,ND )]604 code0 -[2(SAL rm16,imm 186,ND )]593 code0 -[2(SAL rm32,unity 386,ND )]582 code0 -[2(SAL rm32,reg_cl 386,ND )]571 code0 -[2(SAL rm32,imm 386,ND )]560 code0 -[2(SAL rm64,unity X64,ND )]549 code0 -[2(SAL rm64,reg_cl X64,ND )]538 code0 -[2(SAL rm64,imm X64,ND )]527 code0 -[2(SALC 8086,UNDOC )]516 code0 -[2(SAR rm8,unity 8086 )]505 code0 -[2(SAR rm8,reg_cl 8086 )]494 code0 -[2(SAR rm8,imm 186 )]483 code0 -[2(SAR rm16,unity 8086 )]472 code0 -[2(SAR rm16,reg_cl 8086 )]461 code0 -[2(SAR rm16,imm 186 )]450 code0 -[2(SAR rm32,unity 386 )]439 code0 -[2(SAR rm32,reg_cl 386 )]428 code0 -[2(SAR rm32,imm 386 )]417 code0 -[2(SAR rm64,unity X64 )]406 code0 -[2(SAR rm64,reg_cl X64 )]395 code0 -[2(SAR rm64,imm X64 )]384 code0 -[2(SBB mem,reg8 8086 )]373 code0 -[2(SBB reg8,reg8 8086 )]362 code0 -[2(SBB mem,reg16 8086 )]351 code0 -[2(SBB reg16,reg16 8086 )]340 code0 -[2(SBB mem,reg32 386 )]329 code0 -[2(SBB reg32,reg32 386 )]318 code0 -[2(SBB mem,reg64 X64 )]307 code0 -[2(SBB reg64,reg64 X64 )]296 code0 -[2(SBB reg8,mem 8086 )]285 code0 -[2(SBB reg8,reg8 8086 )]274 code0 -[2(SBB reg16,mem 8086 )]263 code0 -[2(SBB reg16,reg16 8086 )]252 code0 -[2(SBB reg32,mem 386 )]241 code0 -[2(SBB reg32,reg32 386 )]230 code0 -[2(SBB reg64,mem X64 )]219 code0 -[2(SBB reg64,reg64 X64 )]208 code0 -[2(SBB rm16,imm8 8086 )]197 code0 -[2(SBB rm32,imm8 386 )]186 code0 -[2(SBB rm64,imm8 X64 )]175 code0 -[2(SBB reg_al,imm 8086 )]164 code0 -[2(SBB reg_ax,sbyte16 8086 )]153 code0 -[2(SBB reg_ax,imm 8086 )]142 code0 -[2(SBB reg_eax,sbyte32 386 )]131 code0 -[2(SBB reg_eax,imm 386 )]120 code0 -[2(SBB reg_rax,sbyte64 X64 )]109 code0 -(142)pageeven -restore showpage -%%Page: 143 143 -%%BeginPageSetup -save -%%EndPageSetup -/143 pa -[2(SBB reg_rax,imm X64 )]681 code0 -[2(SBB rm8,imm 8086 )]670 code0 -[2(SBB rm16,imm 8086 )]659 code0 -[2(SBB rm32,imm 386 )]648 code0 -[2(SBB rm64,imm X64 )]637 code0 -[2(SBB mem,imm8 8086 )]626 code0 -[2(SBB mem,imm16 8086 )]615 code0 -[2(SBB mem,imm32 386 )]604 code0 -[2(SCASB 8086 )]593 code0 -[2(SCASD 386 )]582 code0 -[2(SCASQ X64 )]571 code0 -[2(SCASW 8086 )]560 code0 -[2(SFENCE X64,AMD )]549 code0 -[2(SGDT mem 286 )]538 code0 -[2(SHL rm8,unity 8086 )]527 code0 -[2(SHL rm8,reg_cl 8086 )]516 code0 -[2(SHL rm8,imm 186 )]505 code0 -[2(SHL rm16,unity 8086 )]494 code0 -[2(SHL rm16,reg_cl 8086 )]483 code0 -[2(SHL rm16,imm 186 )]472 code0 -[2(SHL rm32,unity 386 )]461 code0 -[2(SHL rm32,reg_cl 386 )]450 code0 -[2(SHL rm32,imm 386 )]439 code0 -[2(SHL rm64,unity X64 )]428 code0 -[2(SHL rm64,reg_cl X64 )]417 code0 -[2(SHL rm64,imm X64 )]406 code0 -[2(SHLD mem,reg16,imm 3862 )]395 code0 -[2(SHLD reg16,reg16,imm 3862 )]384 code0 -[2(SHLD mem,reg32,imm 3862 )]373 code0 -[2(SHLD reg32,reg32,imm 3862 )]362 code0 -[2(SHLD mem,reg64,imm X642 )]351 code0 -[2(SHLD reg64,reg64,imm X642 )]340 code0 -[2(SHLD mem,reg16,reg_cl 386 )]329 code0 -[2(SHLD reg16,reg16,reg_cl 386 )]318 code0 -[2(SHLD mem,reg32,reg_cl 386 )]307 code0 -[2(SHLD reg32,reg32,reg_cl 386 )]296 code0 -[2(SHLD mem,reg64,reg_cl X64 )]285 code0 -[2(SHLD reg64,reg64,reg_cl X64 )]274 code0 -[2(SHR rm8,unity 8086 )]263 code0 -[2(SHR rm8,reg_cl 8086 )]252 code0 -[2(SHR rm8,imm 186 )]241 code0 -[2(SHR rm16,unity 8086 )]230 code0 -[2(SHR rm16,reg_cl 8086 )]219 code0 -[2(SHR rm16,imm 186 )]208 code0 -[2(SHR rm32,unity 386 )]197 code0 -[2(SHR rm32,reg_cl 386 )]186 code0 -[2(SHR rm32,imm 386 )]175 code0 -[2(SHR rm64,unity X64 )]164 code0 -[2(SHR rm64,reg_cl X64 )]153 code0 -[2(SHR rm64,imm X64 )]142 code0 -[2(SHRD mem,reg16,imm 3862 )]131 code0 -[2(SHRD reg16,reg16,imm 3862 )]120 code0 -[2(SHRD mem,reg32,imm 3862 )]109 code0 -(143)pageodd -restore showpage -%%Page: 144 144 -%%BeginPageSetup -save -%%EndPageSetup -/144 pa -[2(SHRD reg32,reg32,imm 3862 )]681 code0 -[2(SHRD mem,reg64,imm X642 )]670 code0 -[2(SHRD reg64,reg64,imm X642 )]659 code0 -[2(SHRD mem,reg16,reg_cl 386 )]648 code0 -[2(SHRD reg16,reg16,reg_cl 386 )]637 code0 -[2(SHRD mem,reg32,reg_cl 386 )]626 code0 -[2(SHRD reg32,reg32,reg_cl 386 )]615 code0 -[2(SHRD mem,reg64,reg_cl X64 )]604 code0 -[2(SHRD reg64,reg64,reg_cl X64 )]593 code0 -[2(SIDT mem 286 )]582 code0 -[2(SLDT mem 286 )]571 code0 -[2(SLDT mem16 286 )]560 code0 -[2(SLDT reg16 286 )]549 code0 -[2(SLDT reg32 386 )]538 code0 -[2(SLDT reg64 X64,ND )]527 code0 -[2(SLDT reg64 X64 )]516 code0 -[2(SKINIT X64 )]505 code0 -[2(SMI 386,UNDOC )]494 code0 -[2(SMINT P6,CYRIX,ND )]483 code0 -[2(SMINTOLD 486,CYRIX,ND )]472 code0 -[2(SMSW mem 286 )]461 code0 -[2(SMSW mem16 286 )]450 code0 -[2(SMSW reg16 286 )]439 code0 -[2(SMSW reg32 386 )]428 code0 -[2(STC 8086 )]417 code0 -[2(STD 8086 )]406 code0 -[2(STGI X64 )]395 code0 -[2(STI 8086 )]384 code0 -[2(STOSB 8086 )]373 code0 -[2(STOSD 386 )]362 code0 -[2(STOSQ X64 )]351 code0 -[2(STOSW 8086 )]340 code0 -[2(STR mem 286,PROT )]329 code0 -[2(STR mem16 286,PROT )]318 code0 -[2(STR reg16 286,PROT )]307 code0 -[2(STR reg32 386,PROT )]296 code0 -[2(STR reg64 X64 )]285 code0 -[2(SUB mem,reg8 8086 )]274 code0 -[2(SUB reg8,reg8 8086 )]263 code0 -[2(SUB mem,reg16 8086 )]252 code0 -[2(SUB reg16,reg16 8086 )]241 code0 -[2(SUB mem,reg32 386 )]230 code0 -[2(SUB reg32,reg32 386 )]219 code0 -[2(SUB mem,reg64 X64 )]208 code0 -[2(SUB reg64,reg64 X64 )]197 code0 -[2(SUB reg8,mem 8086 )]186 code0 -[2(SUB reg8,reg8 8086 )]175 code0 -[2(SUB reg16,mem 8086 )]164 code0 -[2(SUB reg16,reg16 8086 )]153 code0 -[2(SUB reg32,mem 386 )]142 code0 -[2(SUB reg32,reg32 386 )]131 code0 -[2(SUB reg64,mem X64 )]120 code0 -[2(SUB reg64,reg64 X64 )]109 code0 -(144)pageeven -restore showpage -%%Page: 145 145 -%%BeginPageSetup -save -%%EndPageSetup -/145 pa -[2(SUB rm16,imm8 8086 )]681 code0 -[2(SUB rm32,imm8 386 )]670 code0 -[2(SUB rm64,imm8 X64 )]659 code0 -[2(SUB reg_al,imm 8086 )]648 code0 -[2(SUB reg_ax,sbyte16 8086 )]637 code0 -[2(SUB reg_ax,imm 8086 )]626 code0 -[2(SUB reg_eax,sbyte32 386 )]615 code0 -[2(SUB reg_eax,imm 386 )]604 code0 -[2(SUB reg_rax,sbyte64 X64 )]593 code0 -[2(SUB reg_rax,imm X64 )]582 code0 -[2(SUB rm8,imm 8086 )]571 code0 -[2(SUB rm16,imm 8086 )]560 code0 -[2(SUB rm32,imm 386 )]549 code0 -[2(SUB rm64,imm X64 )]538 code0 -[2(SUB mem,imm8 8086 )]527 code0 -[2(SUB mem,imm16 8086 )]516 code0 -[2(SUB mem,imm32 386 )]505 code0 -[2(SVDC mem80,reg_sreg 486,CYRIXM )]494 code0 -[2(SVLDT mem80 486,CYRIXM,ND )]483 code0 -[2(SVTS mem80 486,CYRIXM )]472 code0 -[2(SWAPGS X64 )]461 code0 -[2(SYSCALL P6,AMD )]450 code0 -[2(SYSENTER P6 )]439 code0 -[2(SYSEXIT P6,PRIV )]428 code0 -[2(SYSRET P6,PRIV,AMD )]417 code0 -[2(TEST mem,reg8 8086 )]406 code0 -[2(TEST reg8,reg8 8086 )]395 code0 -[2(TEST mem,reg16 8086 )]384 code0 -[2(TEST reg16,reg16 8086 )]373 code0 -[2(TEST mem,reg32 386 )]362 code0 -[2(TEST reg32,reg32 386 )]351 code0 -[2(TEST mem,reg64 X64 )]340 code0 -[2(TEST reg64,reg64 X64 )]329 code0 -[2(TEST reg8,mem 8086 )]318 code0 -[2(TEST reg16,mem 8086 )]307 code0 -[2(TEST reg32,mem 386 )]296 code0 -[2(TEST reg64,mem X64 )]285 code0 -[2(TEST reg_al,imm 8086 )]274 code0 -[2(TEST reg_ax,imm 8086 )]263 code0 -[2(TEST reg_eax,imm 386 )]252 code0 -[2(TEST reg_rax,imm X64 )]241 code0 -[2(TEST rm8,imm 8086 )]230 code0 -[2(TEST rm16,imm 8086 )]219 code0 -[2(TEST rm32,imm 386 )]208 code0 -[2(TEST rm64,imm X64 )]197 code0 -[2(TEST mem,imm8 8086 )]186 code0 -[2(TEST mem,imm16 8086 )]175 code0 -[2(TEST mem,imm32 386 )]164 code0 -[2(UD0 186,UNDOC )]153 code0 -[2(UD1 186,UNDOC )]142 code0 -[2(UD2B 186,UNDOC,ND )]131 code0 -[2(UD2 186 )]120 code0 -[2(UD2A 186,ND )]109 code0 -(145)pageodd -restore showpage -%%Page: 146 146 -%%BeginPageSetup -save -%%EndPageSetup -/146 pa -[2(UMOV mem,reg8 386,UNDOC,ND )]681 code0 -[2(UMOV reg8,reg8 386,UNDOC,ND )]670 code0 -[2(UMOV mem,reg16 386,UNDOC,ND )]659 code0 -[2(UMOV reg16,reg16 386,UNDOC,ND )]648 code0 -[2(UMOV mem,reg32 386,UNDOC,ND )]637 code0 -[2(UMOV reg32,reg32 386,UNDOC,ND )]626 code0 -[2(UMOV reg8,mem 386,UNDOC,ND )]615 code0 -[2(UMOV reg8,reg8 386,UNDOC,ND )]604 code0 -[2(UMOV reg16,mem 386,UNDOC,ND )]593 code0 -[2(UMOV reg16,reg16 386,UNDOC,ND )]582 code0 -[2(UMOV reg32,mem 386,UNDOC,ND )]571 code0 -[2(UMOV reg32,reg32 386,UNDOC,ND )]560 code0 -[2(VERR mem 286,PROT )]549 code0 -[2(VERR mem16 286,PROT )]538 code0 -[2(VERR reg16 286,PROT )]527 code0 -[2(VERW mem 286,PROT )]516 code0 -[2(VERW mem16 286,PROT )]505 code0 -[2(VERW reg16 286,PROT )]494 code0 -[2(FWAIT 8086 )]483 code0 -[2(WBINVD 486,PRIV )]472 code0 -[2(WRSHR rm32 P6,CYRIXM )]461 code0 -[2(WRMSR PENT,PRIV )]450 code0 -[2(XADD mem,reg8 486 )]439 code0 -[2(XADD reg8,reg8 486 )]428 code0 -[2(XADD mem,reg16 486 )]417 code0 -[2(XADD reg16,reg16 486 )]406 code0 -[2(XADD mem,reg32 486 )]395 code0 -[2(XADD reg32,reg32 486 )]384 code0 -[2(XADD mem,reg64 X64 )]373 code0 -[2(XADD reg64,reg64 X64 )]362 code0 -[2(XBTS reg16,mem 386,SW,UNDOC,ND )]351 code0 -[2(XBTS reg16,reg16 386,UNDOC,ND )]340 code0 -[2(XBTS reg32,mem 386,SD,UNDOC,ND )]329 code0 -[2(XBTS reg32,reg32 386,UNDOC,ND )]318 code0 -[2(XCHG reg_ax,reg16 8086 )]307 code0 -[2(XCHG reg_eax,reg32na 386 )]296 code0 -[2(XCHG reg_rax,reg64 X64 )]285 code0 -[2(XCHG reg16,reg_ax 8086 )]274 code0 -[2(XCHG reg32na,reg_eax 386 )]263 code0 -[2(XCHG reg64,reg_rax X64 )]252 code0 -[2(XCHG reg_eax,reg_eax 386,NOLONG )]241 code0 -[2(XCHG reg8,mem 8086 )]230 code0 -[2(XCHG reg8,reg8 8086 )]219 code0 -[2(XCHG reg16,mem 8086 )]208 code0 -[2(XCHG reg16,reg16 8086 )]197 code0 -[2(XCHG reg32,mem 386 )]186 code0 -[2(XCHG reg32,reg32 386 )]175 code0 -[2(XCHG reg64,mem X64 )]164 code0 -[2(XCHG reg64,reg64 X64 )]153 code0 -[2(XCHG mem,reg8 8086 )]142 code0 -[2(XCHG reg8,reg8 8086 )]131 code0 -[2(XCHG mem,reg16 8086 )]120 code0 -[2(XCHG reg16,reg16 8086 )]109 code0 -(146)pageeven -restore showpage -%%Page: 147 147 -%%BeginPageSetup -save -%%EndPageSetup -/147 pa -[2(XCHG mem,reg32 386 )]681 code0 -[2(XCHG reg32,reg32 386 )]670 code0 -[2(XCHG mem,reg64 X64 )]659 code0 -[2(XCHG reg64,reg64 X64 )]648 code0 -[2(XLATB 8086 )]637 code0 -[2(XLAT 8086 )]626 code0 -[2(XOR mem,reg8 8086 )]615 code0 -[2(XOR reg8,reg8 8086 )]604 code0 -[2(XOR mem,reg16 8086 )]593 code0 -[2(XOR reg16,reg16 8086 )]582 code0 -[2(XOR mem,reg32 386 )]571 code0 -[2(XOR reg32,reg32 386 )]560 code0 -[2(XOR mem,reg64 X64 )]549 code0 -[2(XOR reg64,reg64 X64 )]538 code0 -[2(XOR reg8,mem 8086 )]527 code0 -[2(XOR reg8,reg8 8086 )]516 code0 -[2(XOR reg16,mem 8086 )]505 code0 -[2(XOR reg16,reg16 8086 )]494 code0 -[2(XOR reg32,mem 386 )]483 code0 -[2(XOR reg32,reg32 386 )]472 code0 -[2(XOR reg64,mem X64 )]461 code0 -[2(XOR reg64,reg64 X64 )]450 code0 -[2(XOR rm16,imm8 8086 )]439 code0 -[2(XOR rm32,imm8 386 )]428 code0 -[2(XOR rm64,imm8 X64 )]417 code0 -[2(XOR reg_al,imm 8086 )]406 code0 -[2(XOR reg_ax,sbyte16 8086 )]395 code0 -[2(XOR reg_ax,imm 8086 )]384 code0 -[2(XOR reg_eax,sbyte32 386 )]373 code0 -[2(XOR reg_eax,imm 386 )]362 code0 -[2(XOR reg_rax,sbyte64 X64 )]351 code0 -[2(XOR reg_rax,imm X64 )]340 code0 -[2(XOR rm8,imm 8086 )]329 code0 -[2(XOR rm16,imm 8086 )]318 code0 -[2(XOR rm32,imm 386 )]307 code0 -[2(XOR rm64,imm X64 )]296 code0 -[2(XOR mem,imm8 8086 )]285 code0 -[2(XOR mem,imm16 8086 )]274 code0 -[2(XOR mem,imm32 386 )]263 code0 -[2(CMOVcc reg16,mem P6 )]252 code0 -[2(CMOVcc reg16,reg16 P6 )]241 code0 -[2(CMOVcc reg32,mem P6 )]230 code0 -[2(CMOVcc reg32,reg32 P6 )]219 code0 -[2(CMOVcc reg64,mem X64 )]208 code0 -[2(CMOVcc reg64,reg64 X64 )]197 code0 -[2(Jcc imm|near 386 )]186 code0 -[2(Jcc imm16|near 386 )]175 code0 -[2(Jcc imm32|near 386 )]164 code0 -[2(Jcc imm|short 8086,ND )]153 code0 -[2(Jcc imm 8086,ND )]142 code0 -[2(Jcc imm 386,ND )]131 code0 -[2(Jcc imm 8086,ND )]120 code0 -[2(Jcc imm 8086 )]109 code0 -(147)pageodd -restore showpage -%%Page: 148 148 -%%BeginPageSetup -save -%%EndPageSetup -/148 pa -[2(SETcc mem 386 )]681 code0 -[2(SETcc reg8 386)]670 code2 -[{/section-B.1.3 xa}(Katmai Streaming SIMD instructions \(SSE \226\226 a.k.a. KNI, XMM, MMX2\))](B.1.3)650.8 subh3 -[2(ADDPS xmmreg,xmmrm KATMAI,SSE )]633.8 code1 -[2(ADDSS xmmreg,xmmrm KATMAI,SSE,SD )]622.8 code0 -[2(ANDNPS xmmreg,xmmrm KATMAI,SSE )]611.8 code0 -[2(ANDPS xmmreg,xmmrm KATMAI,SSE )]600.8 code0 -[2(CMPEQPS xmmreg,xmmrm KATMAI,SSE )]589.8 code0 -[2(CMPEQSS xmmreg,xmmrm KATMAI,SSE )]578.8 code0 -[2(CMPLEPS xmmreg,xmmrm KATMAI,SSE )]567.8 code0 -[2(CMPLESS xmmreg,xmmrm KATMAI,SSE )]556.8 code0 -[2(CMPLTPS xmmreg,xmmrm KATMAI,SSE )]545.8 code0 -[2(CMPLTSS xmmreg,xmmrm KATMAI,SSE )]534.8 code0 -[2(CMPNEQPS xmmreg,xmmrm KATMAI,SSE )]523.8 code0 -[2(CMPNEQSS xmmreg,xmmrm KATMAI,SSE )]512.8 code0 -[2(CMPNLEPS xmmreg,xmmrm KATMAI,SSE )]501.8 code0 -[2(CMPNLESS xmmreg,xmmrm KATMAI,SSE )]490.8 code0 -[2(CMPNLTPS xmmreg,xmmrm KATMAI,SSE )]479.8 code0 -[2(CMPNLTSS xmmreg,xmmrm KATMAI,SSE )]468.8 code0 -[2(CMPORDPS xmmreg,xmmrm KATMAI,SSE )]457.8 code0 -[2(CMPORDSS xmmreg,xmmrm KATMAI,SSE )]446.8 code0 -[2(CMPUNORDPS xmmreg,xmmrm KATMAI,SSE )]435.8 code0 -[2(CMPUNORDSS xmmreg,xmmrm KATMAI,SSE )]424.8 code0 -[2(CMPPS xmmreg,mem,imm KATMAI,SSE )]413.8 code0 -[2(CMPPS xmmreg,xmmreg,imm KATMAI,SSE )]402.8 code0 -[2(CMPSS xmmreg,mem,imm KATMAI,SSE )]391.8 code0 -[2(CMPSS xmmreg,xmmreg,imm KATMAI,SSE )]380.8 code0 -[2(COMISS xmmreg,xmmrm KATMAI,SSE )]369.8 code0 -[2(CVTPI2PS xmmreg,mmxrm KATMAI,SSE,MMX )]358.8 code0 -[2(CVTPS2PI mmxreg,xmmrm KATMAI,SSE,MMX )]347.8 code0 -[2(CVTSI2SS xmmreg,mem KATMAI,SSE,SD,AR1,ND )]336.8 code0 -[2(CVTSI2SS xmmreg,rm32 KATMAI,SSE,SD,AR1 )]325.8 code0 -[2(CVTSI2SS xmmreg,rm64 X64,SSE,AR1 )]314.8 code0 -[2(CVTSS2SI reg32,xmmreg KATMAI,SSE,SD,AR1 )]303.8 code0 -[2(CVTSS2SI reg32,mem KATMAI,SSE,SD,AR1 )]292.8 code0 -[2(CVTSS2SI reg64,xmmreg X64,SSE,SD,AR1 )]281.8 code0 -[2(CVTSS2SI reg64,mem X64,SSE,SD,AR1 )]270.8 code0 -[2(CVTTPS2PI mmxreg,xmmrm KATMAI,SSE,MMX )]259.8 code0 -[2(CVTTSS2SI reg32,xmmrm KATMAI,SSE,SD,AR1 )]248.8 code0 -[2(CVTTSS2SI reg64,xmmrm X64,SSE,SD,AR1 )]237.8 code0 -[2(DIVPS xmmreg,xmmrm KATMAI,SSE )]226.8 code0 -[2(DIVSS xmmreg,xmmrm KATMAI,SSE )]215.8 code0 -[2(LDMXCSR mem KATMAI,SSE,SD )]204.8 code0 -[2(MAXPS xmmreg,xmmrm KATMAI,SSE )]193.8 code0 -[2(MAXSS xmmreg,xmmrm KATMAI,SSE )]182.8 code0 -[2(MINPS xmmreg,xmmrm KATMAI,SSE )]171.8 code0 -[2(MINSS xmmreg,xmmrm KATMAI,SSE )]160.8 code0 -[2(MOVAPS xmmreg,mem KATMAI,SSE )]149.8 code0 -[2(MOVAPS mem,xmmreg KATMAI,SSE )]138.8 code0 -[2(MOVAPS xmmreg,xmmreg KATMAI,SSE )]127.8 code0 -[2(MOVAPS xmmreg,xmmreg KATMAI,SSE )]116.8 code0 -[2(MOVHPS xmmreg,mem KATMAI,SSE )]105.8 code0 -(148)pageeven -restore showpage -%%Page: 149 149 -%%BeginPageSetup -save -%%EndPageSetup -/149 pa -[2(MOVHPS mem,xmmreg KATMAI,SSE )]681 code0 -[2(MOVLHPS xmmreg,xmmreg KATMAI,SSE )]670 code0 -[2(MOVLPS xmmreg,mem KATMAI,SSE )]659 code0 -[2(MOVLPS mem,xmmreg KATMAI,SSE )]648 code0 -[2(MOVHLPS xmmreg,xmmreg KATMAI,SSE )]637 code0 -[2(MOVMSKPS reg32,xmmreg KATMAI,SSE )]626 code0 -[2(MOVMSKPS reg64,xmmreg X64,SSE )]615 code0 -[2(MOVNTPS mem,xmmreg KATMAI,SSE )]604 code0 -[2(MOVSS xmmreg,mem KATMAI,SSE )]593 code0 -[2(MOVSS mem,xmmreg KATMAI,SSE )]582 code0 -[2(MOVSS xmmreg,xmmreg KATMAI,SSE )]571 code0 -[2(MOVSS xmmreg,xmmreg KATMAI,SSE )]560 code0 -[2(MOVUPS xmmreg,mem KATMAI,SSE )]549 code0 -[2(MOVUPS mem,xmmreg KATMAI,SSE )]538 code0 -[2(MOVUPS xmmreg,xmmreg KATMAI,SSE )]527 code0 -[2(MOVUPS xmmreg,xmmreg KATMAI,SSE )]516 code0 -[2(MULPS xmmreg,xmmrm KATMAI,SSE )]505 code0 -[2(MULSS xmmreg,xmmrm KATMAI,SSE )]494 code0 -[2(ORPS xmmreg,xmmrm KATMAI,SSE )]483 code0 -[2(RCPPS xmmreg,xmmrm KATMAI,SSE )]472 code0 -[2(RCPSS xmmreg,xmmrm KATMAI,SSE )]461 code0 -[2(RSQRTPS xmmreg,xmmrm KATMAI,SSE )]450 code0 -[2(RSQRTSS xmmreg,xmmrm KATMAI,SSE )]439 code0 -[2(SHUFPS xmmreg,mem,imm KATMAI,SSE )]428 code0 -[2(SHUFPS xmmreg,xmmreg,imm KATMAI,SSE )]417 code0 -[2(SQRTPS xmmreg,xmmrm KATMAI,SSE )]406 code0 -[2(SQRTSS xmmreg,xmmrm KATMAI,SSE )]395 code0 -[2(STMXCSR mem KATMAI,SSE,SD )]384 code0 -[2(SUBPS xmmreg,xmmrm KATMAI,SSE )]373 code0 -[2(SUBSS xmmreg,xmmrm KATMAI,SSE )]362 code0 -[2(UCOMISS xmmreg,xmmrm KATMAI,SSE )]351 code0 -[2(UNPCKHPS xmmreg,xmmrm KATMAI,SSE )]340 code0 -[2(UNPCKLPS xmmreg,xmmrm KATMAI,SSE )]329 code0 -[2(XORPS xmmreg,xmmrm KATMAI,SSE)]318 code2 -[{/section-B.1.4 xa}(Introduced in Deschutes but necessary for SSE support)](B.1.4)298.8 subh3 -[2(FXRSTOR mem P6,SSE,FPU )]281.8 code1 -[2(FXSAVE mem P6,SSE,FPU)]270.8 code2 -[{/section-B.1.5 xa}(XSAVE group \(AVX and extended state\))](B.1.5)251.6 subh3 -[2(XGETBV NEHALEM )]234.6 code1 -[2(XSETBV NEHALEM,PRIV )]223.6 code0 -[2(XSAVE mem NEHALEM )]212.6 code0 -[2(XRSTOR mem NEHALEM)]201.6 code2 -[{/section-B.1.6 xa}(Generic memory operations)](B.1.6)182.4 subh3 -[2(PREFETCHNTA mem KATMAI )]165.4 code1 -[2(PREFETCHT0 mem KATMAI )]154.4 code0 -[2(PREFETCHT1 mem KATMAI )]143.4 code0 -[2(PREFETCHT2 mem KATMAI )]132.4 code0 -[2(SFENCE KATMAI)]121.4 code2 -(149)pageodd -restore showpage -%%Page: 150 150 -%%BeginPageSetup -save -%%EndPageSetup -/150 pa -[{/section-B.1.7 xa}(New MMX instructions introduced in Katmai)](B.1.7)678.8 subh3 -[2(MASKMOVQ mmxreg,mmxreg KATMAI,MMX )]661.8 code1 -[2(MOVNTQ mem,mmxreg KATMAI,MMX )]650.8 code0 -[2(PAVGB mmxreg,mmxrm KATMAI,MMX )]639.8 code0 -[2(PAVGW mmxreg,mmxrm KATMAI,MMX )]628.8 code0 -[2(PEXTRW reg32,mmxreg,imm KATMAI,MMX )]617.8 code0 -[2(PINSRW mmxreg,mem,imm KATMAI,MMX )]606.8 code0 -[2(PINSRW mmxreg,rm16,imm KATMAI,MMX )]595.8 code0 -[2(PINSRW mmxreg,reg32,imm KATMAI,MMX )]584.8 code0 -[2(PMAXSW mmxreg,mmxrm KATMAI,MMX )]573.8 code0 -[2(PMAXUB mmxreg,mmxrm KATMAI,MMX )]562.8 code0 -[2(PMINSW mmxreg,mmxrm KATMAI,MMX )]551.8 code0 -[2(PMINUB mmxreg,mmxrm KATMAI,MMX )]540.8 code0 -[2(PMOVMSKB reg32,mmxreg KATMAI,MMX )]529.8 code0 -[2(PMULHUW mmxreg,mmxrm KATMAI,MMX )]518.8 code0 -[2(PSADBW mmxreg,mmxrm KATMAI,MMX )]507.8 code0 -[2(PSHUFW mmxreg,mmxrm,imm KATMAI,MMX2)]496.8 code2 -[{/section-B.1.8 xa}(AMD Enhanced 3DNow! \(Athlon\) instructions)](B.1.8)477.6 subh3 -[2(PF2IW mmxreg,mmxrm PENT,3DNOW )]460.6 code1 -[2(PFNACC mmxreg,mmxrm PENT,3DNOW )]449.6 code0 -[2(PFPNACC mmxreg,mmxrm PENT,3DNOW )]438.6 code0 -[2(PI2FW mmxreg,mmxrm PENT,3DNOW )]427.6 code0 -[2(PSWAPD mmxreg,mmxrm PENT,3DNOW)]416.6 code2 -[{/section-B.1.9 xa}(Willamette SSE2 Cacheability Instructions)](B.1.9)397.4 subh3 -[2(MASKMOVDQU xmmreg,xmmreg WILLAMETTE,SSE2 )]380.4 code1 -[2(CLFLUSH mem WILLAMETTE,SSE2 )]369.4 code0 -[2(MOVNTDQ mem,xmmreg WILLAMETTE,SSE2,SO )]358.4 code0 -[2(MOVNTI mem,reg32 WILLAMETTE,SD )]347.4 code0 -[2(MOVNTI mem,reg64 X64 )]336.4 code0 -[2(MOVNTPD mem,xmmreg WILLAMETTE,SSE2,SO )]325.4 code0 -[2(LFENCE WILLAMETTE,SSE2 )]314.4 code0 -[2(MFENCE WILLAMETTE,SSE2)]303.4 code2 -[{/section-B.1.10 xa}(Willamette MMX instructions \(SSE2 SIMD Integer Instructions\))](B.1.10)284.2 subh3 -[2(MOVD mem,xmmreg WILLAMETTE,SSE2,SD )]267.2 code1 -[2(MOVD xmmreg,mem WILLAMETTE,SSE2,SD )]256.2 code0 -[2(MOVD xmmreg,rm32 WILLAMETTE,SSE2 )]245.2 code0 -[2(MOVD rm32,xmmreg WILLAMETTE,SSE2 )]234.2 code0 -[2(MOVDQA xmmreg,xmmreg WILLAMETTE,SSE2 )]223.2 code0 -[2(MOVDQA mem,xmmreg WILLAMETTE,SSE2,SO )]212.2 code0 -[2(MOVDQA xmmreg,mem WILLAMETTE,SSE2,SO )]201.2 code0 -[2(MOVDQA xmmreg,xmmreg WILLAMETTE,SSE2 )]190.2 code0 -[2(MOVDQU xmmreg,xmmreg WILLAMETTE,SSE2 )]179.2 code0 -[2(MOVDQU mem,xmmreg WILLAMETTE,SSE2,SO )]168.2 code0 -[2(MOVDQU xmmreg,mem WILLAMETTE,SSE2,SO )]157.2 code0 -[2(MOVDQU xmmreg,xmmreg WILLAMETTE,SSE2 )]146.2 code0 -[2(MOVDQ2Q mmxreg,xmmreg WILLAMETTE,SSE2 )]135.2 code0 -[2(MOVQ xmmreg,xmmreg WILLAMETTE,SSE2 )]124.2 code0 -[2(MOVQ xmmreg,xmmreg WILLAMETTE,SSE2 )]113.2 code0 -[2(MOVQ mem,xmmreg WILLAMETTE,SSE2 )]102.2 code0 -(150)pageeven -restore showpage -%%Page: 151 151 -%%BeginPageSetup -save -%%EndPageSetup -/151 pa -[2(MOVQ xmmreg,mem WILLAMETTE,SSE2 )]681 code0 -[2(MOVQ xmmreg,rm64 X64,SSE2 )]670 code0 -[2(MOVQ rm64,xmmreg X64,SSE2 )]659 code0 -[2(MOVQ2DQ xmmreg,mmxreg WILLAMETTE,SSE2 )]648 code0 -[2(PACKSSWB xmmreg,xmmrm WILLAMETTE,SSE2,SO )]637 code0 -[2(PACKSSDW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]626 code0 -[2(PACKUSWB xmmreg,xmmrm WILLAMETTE,SSE2,SO )]615 code0 -[2(PADDB xmmreg,xmmrm WILLAMETTE,SSE2,SO )]604 code0 -[2(PADDW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]593 code0 -[2(PADDD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]582 code0 -[2(PADDQ mmxreg,mmxrm WILLAMETTE,MMX )]571 code0 -[2(PADDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO )]560 code0 -[2(PADDSB xmmreg,xmmrm WILLAMETTE,SSE2,SO )]549 code0 -[2(PADDSW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]538 code0 -[2(PADDUSB xmmreg,xmmrm WILLAMETTE,SSE2,SO )]527 code0 -[2(PADDUSW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]516 code0 -[2(PAND xmmreg,xmmrm WILLAMETTE,SSE2,SO )]505 code0 -[2(PANDN xmmreg,xmmrm WILLAMETTE,SSE2,SO )]494 code0 -[2(PAVGB xmmreg,xmmrm WILLAMETTE,SSE2,SO )]483 code0 -[2(PAVGW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]472 code0 -[2(PCMPEQB xmmreg,xmmrm WILLAMETTE,SSE2,SO )]461 code0 -[2(PCMPEQW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]450 code0 -[2(PCMPEQD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]439 code0 -[2(PCMPGTB xmmreg,xmmrm WILLAMETTE,SSE2,SO )]428 code0 -[2(PCMPGTW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]417 code0 -[2(PCMPGTD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]406 code0 -[2(PEXTRW reg32,xmmreg,imm WILLAMETTE,SSE2 )]395 code0 -[2(PINSRW xmmreg,reg16,imm WILLAMETTE,SSE2 )]384 code0 -[2(PINSRW xmmreg,reg32,imm WILLAMETTE,SSE2,ND )]373 code0 -[2(PINSRW xmmreg,mem,imm WILLAMETTE,SSE2 )]362 code0 -[2(PINSRW xmmreg,mem16,imm WILLAMETTE,SSE2 )]351 code0 -[2(PMADDWD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]340 code0 -[2(PMAXSW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]329 code0 -[2(PMAXUB xmmreg,xmmrm WILLAMETTE,SSE2,SO )]318 code0 -[2(PMINSW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]307 code0 -[2(PMINUB xmmreg,xmmrm WILLAMETTE,SSE2,SO )]296 code0 -[2(PMOVMSKB reg32,xmmreg WILLAMETTE,SSE2 )]285 code0 -[2(PMULHUW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]274 code0 -[2(PMULHW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]263 code0 -[2(PMULLW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]252 code0 -[2(PMULUDQ mmxreg,mmxrm WILLAMETTE,SSE2,SO )]241 code0 -[2(PMULUDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO )]230 code0 -[2(POR xmmreg,xmmrm WILLAMETTE,SSE2,SO )]219 code0 -[2(PSADBW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]208 code0 -[2(PSHUFD xmmreg,xmmreg,imm WILLAMETTE,SSE2 )]197 code0 -[2(PSHUFD xmmreg,mem,imm WILLAMETTE,SSE22 )]186 code0 -[2(PSHUFHW xmmreg,xmmreg,imm WILLAMETTE,SSE2 )]175 code0 -[2(PSHUFHW xmmreg,mem,imm WILLAMETTE,SSE22 )]164 code0 -[2(PSHUFLW xmmreg,xmmreg,imm WILLAMETTE,SSE2 )]153 code0 -[2(PSHUFLW xmmreg,mem,imm WILLAMETTE,SSE22 )]142 code0 -[2(PSLLDQ xmmreg,imm WILLAMETTE,SSE2,AR1 )]131 code0 -[2(PSLLW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]120 code0 -[2(PSLLW xmmreg,imm WILLAMETTE,SSE2,AR1 )]109 code0 -(151)pageodd -restore showpage -%%Page: 152 152 -%%BeginPageSetup -save -%%EndPageSetup -/152 pa -[2(PSLLD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]681 code0 -[2(PSLLD xmmreg,imm WILLAMETTE,SSE2,AR1 )]670 code0 -[2(PSLLQ xmmreg,xmmrm WILLAMETTE,SSE2,SO )]659 code0 -[2(PSLLQ xmmreg,imm WILLAMETTE,SSE2,AR1 )]648 code0 -[2(PSRAW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]637 code0 -[2(PSRAW xmmreg,imm WILLAMETTE,SSE2,AR1 )]626 code0 -[2(PSRAD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]615 code0 -[2(PSRAD xmmreg,imm WILLAMETTE,SSE2,AR1 )]604 code0 -[2(PSRLDQ xmmreg,imm WILLAMETTE,SSE2,AR1 )]593 code0 -[2(PSRLW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]582 code0 -[2(PSRLW xmmreg,imm WILLAMETTE,SSE2,AR1 )]571 code0 -[2(PSRLD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]560 code0 -[2(PSRLD xmmreg,imm WILLAMETTE,SSE2,AR1 )]549 code0 -[2(PSRLQ xmmreg,xmmrm WILLAMETTE,SSE2,SO )]538 code0 -[2(PSRLQ xmmreg,imm WILLAMETTE,SSE2,AR1 )]527 code0 -[2(PSUBB xmmreg,xmmrm WILLAMETTE,SSE2,SO )]516 code0 -[2(PSUBW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]505 code0 -[2(PSUBD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]494 code0 -[2(PSUBQ mmxreg,mmxrm WILLAMETTE,SSE2,SO )]483 code0 -[2(PSUBQ xmmreg,xmmrm WILLAMETTE,SSE2,SO )]472 code0 -[2(PSUBSB xmmreg,xmmrm WILLAMETTE,SSE2,SO )]461 code0 -[2(PSUBSW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]450 code0 -[2(PSUBUSB xmmreg,xmmrm WILLAMETTE,SSE2,SO )]439 code0 -[2(PSUBUSW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]428 code0 -[2(PUNPCKHBW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]417 code0 -[2(PUNPCKHWD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]406 code0 -[2(PUNPCKHDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO )]395 code0 -[2(PUNPCKHQDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO )]384 code0 -[2(PUNPCKLBW xmmreg,xmmrm WILLAMETTE,SSE2,SO )]373 code0 -[2(PUNPCKLWD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]362 code0 -[2(PUNPCKLDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO )]351 code0 -[2(PUNPCKLQDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO )]340 code0 -[2(PXOR xmmreg,xmmrm WILLAMETTE,SSE2,SO)]329 code2 -[{/section-B.1.11 xa}(Willamette Streaming SIMD instructions \(SSE2\))](B.1.11)309.8 subh3 -[2(ADDPD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]292.8 code1 -[2(ADDSD xmmreg,xmmrm WILLAMETTE,SSE2 )]281.8 code0 -[2(ANDNPD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]270.8 code0 -[2(ANDPD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]259.8 code0 -[2(CMPEQPD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]248.8 code0 -[2(CMPEQSD xmmreg,xmmrm WILLAMETTE,SSE2 )]237.8 code0 -[2(CMPLEPD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]226.8 code0 -[2(CMPLESD xmmreg,xmmrm WILLAMETTE,SSE2 )]215.8 code0 -[2(CMPLTPD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]204.8 code0 -[2(CMPLTSD xmmreg,xmmrm WILLAMETTE,SSE2 )]193.8 code0 -[2(CMPNEQPD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]182.8 code0 -[2(CMPNEQSD xmmreg,xmmrm WILLAMETTE,SSE2 )]171.8 code0 -[2(CMPNLEPD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]160.8 code0 -[2(CMPNLESD xmmreg,xmmrm WILLAMETTE,SSE2 )]149.8 code0 -[2(CMPNLTPD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]138.8 code0 -[2(CMPNLTSD xmmreg,xmmrm WILLAMETTE,SSE2 )]127.8 code0 -[2(CMPORDPD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]116.8 code0 -[2(CMPORDSD xmmreg,xmmrm WILLAMETTE,SSE2 )]105.8 code0 -(152)pageeven -restore showpage -%%Page: 153 153 -%%BeginPageSetup -save -%%EndPageSetup -/153 pa -[2(CMPUNORDPD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]681 code0 -[2(CMPUNORDSD xmmreg,xmmrm WILLAMETTE,SSE2 )]670 code0 -[2(CMPPD xmmreg,xmmrm,imm WILLAMETTE,SSE22 )]659 code0 -[2(CMPSD xmmreg,xmmrm,imm WILLAMETTE,SSE2 )]648 code0 -[2(COMISD xmmreg,xmmrm WILLAMETTE,SSE2 )]637 code0 -[2(CVTDQ2PD xmmreg,xmmrm WILLAMETTE,SSE2 )]626 code0 -[2(CVTDQ2PS xmmreg,xmmrm WILLAMETTE,SSE2,SO )]615 code0 -[2(CVTPD2DQ xmmreg,xmmrm WILLAMETTE,SSE2,SO )]604 code0 -[2(CVTPD2PI mmxreg,xmmrm WILLAMETTE,SSE2,SO )]593 code0 -[2(CVTPD2PS xmmreg,xmmrm WILLAMETTE,SSE2,SO )]582 code0 -[2(CVTPI2PD xmmreg,mmxrm WILLAMETTE,SSE2 )]571 code0 -[2(CVTPS2DQ xmmreg,xmmrm WILLAMETTE,SSE2,SO )]560 code0 -[2(CVTPS2PD xmmreg,xmmrm WILLAMETTE,SSE2 )]549 code0 -[2(CVTSD2SI reg32,xmmreg WILLAMETTE,SSE2,AR1 )]538 code0 -[2(CVTSD2SI reg32,mem WILLAMETTE,SSE2,AR1 )]527 code0 -[2(CVTSD2SI reg64,xmmreg X64,SSE2,AR1 )]516 code0 -[2(CVTSD2SI reg64,mem X64,SSE2,AR1 )]505 code0 -[2(CVTSD2SS xmmreg,xmmrm WILLAMETTE,SSE2 )]494 code0 -[2(CVTSI2SD xmmreg,mem WILLAMETTE,SSE2,SD,AR1,ND )]483 code0 -[2(CVTSI2SD xmmreg,rm32 WILLAMETTE,SSE2,SD,AR1 )]472 code0 -[2(CVTSI2SD xmmreg,rm64 X64,SSE2,AR1 )]461 code0 -[2(CVTSS2SD xmmreg,xmmrm WILLAMETTE,SSE2,SD )]450 code0 -[2(CVTTPD2PI mmxreg,xmmrm WILLAMETTE,SSE2,SO )]439 code0 -[2(CVTTPD2DQ xmmreg,xmmrm WILLAMETTE,SSE2,SO )]428 code0 -[2(CVTTPS2DQ xmmreg,xmmrm WILLAMETTE,SSE2,SO )]417 code0 -[2(CVTTSD2SI reg32,xmmreg WILLAMETTE,SSE2,AR1 )]406 code0 -[2(CVTTSD2SI reg32,mem WILLAMETTE,SSE2,AR1 )]395 code0 -[2(CVTTSD2SI reg64,xmmreg X64,SSE2,AR1 )]384 code0 -[2(CVTTSD2SI reg64,mem X64,SSE2,AR1 )]373 code0 -[2(DIVPD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]362 code0 -[2(DIVSD xmmreg,xmmrm WILLAMETTE,SSE2 )]351 code0 -[2(MAXPD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]340 code0 -[2(MAXSD xmmreg,xmmrm WILLAMETTE,SSE2 )]329 code0 -[2(MINPD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]318 code0 -[2(MINSD xmmreg,xmmrm WILLAMETTE,SSE2 )]307 code0 -[2(MOVAPD xmmreg,xmmreg WILLAMETTE,SSE2 )]296 code0 -[2(MOVAPD xmmreg,xmmreg WILLAMETTE,SSE2 )]285 code0 -[2(MOVAPD mem,xmmreg WILLAMETTE,SSE2,SO )]274 code0 -[2(MOVAPD xmmreg,mem WILLAMETTE,SSE2,SO )]263 code0 -[2(MOVHPD mem,xmmreg WILLAMETTE,SSE2 )]252 code0 -[2(MOVHPD xmmreg,mem WILLAMETTE,SSE2 )]241 code0 -[2(MOVLPD mem,xmmreg WILLAMETTE,SSE2 )]230 code0 -[2(MOVLPD xmmreg,mem WILLAMETTE,SSE2 )]219 code0 -[2(MOVMSKPD reg32,xmmreg WILLAMETTE,SSE2 )]208 code0 -[2(MOVMSKPD reg64,xmmreg X64,SSE2 )]197 code0 -[2(MOVSD xmmreg,xmmreg WILLAMETTE,SSE2 )]186 code0 -[2(MOVSD xmmreg,xmmreg WILLAMETTE,SSE2 )]175 code0 -[2(MOVSD mem,xmmreg WILLAMETTE,SSE2 )]164 code0 -[2(MOVSD xmmreg,mem WILLAMETTE,SSE2 )]153 code0 -[2(MOVUPD xmmreg,xmmreg WILLAMETTE,SSE2 )]142 code0 -[2(MOVUPD xmmreg,xmmreg WILLAMETTE,SSE2 )]131 code0 -[2(MOVUPD mem,xmmreg WILLAMETTE,SSE2,SO )]120 code0 -[2(MOVUPD xmmreg,mem WILLAMETTE,SSE2,SO )]109 code0 -(153)pageodd -restore showpage -%%Page: 154 154 -%%BeginPageSetup -save -%%EndPageSetup -/154 pa -[2(MULPD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]681 code0 -[2(MULSD xmmreg,xmmrm WILLAMETTE,SSE2 )]670 code0 -[2(ORPD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]659 code0 -[2(SHUFPD xmmreg,xmmreg,imm WILLAMETTE,SSE2 )]648 code0 -[2(SHUFPD xmmreg,mem,imm WILLAMETTE,SSE2 )]637 code0 -[2(SQRTPD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]626 code0 -[2(SQRTSD xmmreg,xmmrm WILLAMETTE,SSE2 )]615 code0 -[2(SUBPD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]604 code0 -[2(SUBSD xmmreg,xmmrm WILLAMETTE,SSE2 )]593 code0 -[2(UCOMISD xmmreg,xmmrm WILLAMETTE,SSE2 )]582 code0 -[2(UNPCKHPD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]571 code0 -[2(UNPCKLPD xmmreg,xmmrm WILLAMETTE,SSE2,SO )]560 code0 -[2(XORPD xmmreg,xmmrm WILLAMETTE,SSE2,SO)]549 code2 -[{/section-B.1.12 xa}(Prescott New Instructions \(SSE3\))](B.1.12)529.8 subh3 -[2(ADDSUBPD xmmreg,xmmrm PRESCOTT,SSE3,SO )]512.8 code1 -[2(ADDSUBPS xmmreg,xmmrm PRESCOTT,SSE3,SO )]501.8 code0 -[2(HADDPD xmmreg,xmmrm PRESCOTT,SSE3,SO )]490.8 code0 -[2(HADDPS xmmreg,xmmrm PRESCOTT,SSE3,SO )]479.8 code0 -[2(HSUBPD xmmreg,xmmrm PRESCOTT,SSE3,SO )]468.8 code0 -[2(HSUBPS xmmreg,xmmrm PRESCOTT,SSE3,SO )]457.8 code0 -[2(LDDQU xmmreg,mem PRESCOTT,SSE3,SO )]446.8 code0 -[2(MOVDDUP xmmreg,xmmrm PRESCOTT,SSE3 )]435.8 code0 -[2(MOVSHDUP xmmreg,xmmrm PRESCOTT,SSE3 )]424.8 code0 -[2(MOVSLDUP xmmreg,xmmrm PRESCOTT,SSE3)]413.8 code2 -[{/section-B.1.13 xa}(VMX Instructions)](B.1.13)394.6 subh3 -[2(VMCALL VMX )]377.6 code1 -[2(VMCLEAR mem VMX )]366.6 code0 -[2(VMLAUNCH VMX )]355.6 code0 -[2(VMLOAD X64,VMX )]344.6 code0 -[2(VMMCALL X64,VMX )]333.6 code0 -[2(VMPTRLD mem VMX )]322.6 code0 -[2(VMPTRST mem VMX )]311.6 code0 -[2(VMREAD rm32,reg32 VMX,NOLONG,SD )]300.6 code0 -[2(VMREAD rm64,reg64 X64,VMX )]289.6 code0 -[2(VMRESUME VMX )]278.6 code0 -[2(VMRUN X64,VMX )]267.6 code0 -[2(VMSAVE X64,VMX )]256.6 code0 -[2(VMWRITE reg32,rm32 VMX,NOLONG,SD )]245.6 code0 -[2(VMWRITE reg64,rm64 X64,VMX )]234.6 code0 -[2(VMXOFF VMX )]223.6 code0 -[2(VMXON mem VMX)]212.6 code2 -[{/section-B.1.14 xa}(Extended Page Tables VMX instructions)](B.1.14)193.4 subh3 -[2(INVEPT reg32,mem VMX,SO,NOLONG )]176.4 code1 -[2(INVEPT reg64,mem VMX,SO,LONG )]165.4 code0 -[2(INVVPID reg32,mem VMX,SO,NOLONG )]154.4 code0 -[2(INVVPID reg64,mem VMX,SO,LONG)]143.4 code2 -(154)pageeven -restore showpage -%%Page: 155 155 -%%BeginPageSetup -save -%%EndPageSetup -/155 pa -[{/section-B.1.15 xa}(Tejas New Instructions \(SSSE3\))](B.1.15)678.8 subh3 -[2(PABSB mmxreg,mmxrm SSSE3,MMX )]661.8 code1 -[2(PABSB xmmreg,xmmrm SSSE3 )]650.8 code0 -[2(PABSW mmxreg,mmxrm SSSE3,MMX )]639.8 code0 -[2(PABSW xmmreg,xmmrm SSSE3 )]628.8 code0 -[2(PABSD mmxreg,mmxrm SSSE3,MMX )]617.8 code0 -[2(PABSD xmmreg,xmmrm SSSE3 )]606.8 code0 -[2(PALIGNR mmxreg,mmxrm,imm SSSE3,MMX )]595.8 code0 -[2(PALIGNR xmmreg,xmmrm,imm SSSE3 )]584.8 code0 -[2(PHADDW mmxreg,mmxrm SSSE3,MMX )]573.8 code0 -[2(PHADDW xmmreg,xmmrm SSSE3 )]562.8 code0 -[2(PHADDD mmxreg,mmxrm SSSE3,MMX )]551.8 code0 -[2(PHADDD xmmreg,xmmrm SSSE3 )]540.8 code0 -[2(PHADDSW mmxreg,mmxrm SSSE3,MMX )]529.8 code0 -[2(PHADDSW xmmreg,xmmrm SSSE3 )]518.8 code0 -[2(PHSUBW mmxreg,mmxrm SSSE3,MMX )]507.8 code0 -[2(PHSUBW xmmreg,xmmrm SSSE3 )]496.8 code0 -[2(PHSUBD mmxreg,mmxrm SSSE3,MMX )]485.8 code0 -[2(PHSUBD xmmreg,xmmrm SSSE3 )]474.8 code0 -[2(PHSUBSW mmxreg,mmxrm SSSE3,MMX )]463.8 code0 -[2(PHSUBSW xmmreg,xmmrm SSSE3 )]452.8 code0 -[2(PMADDUBSW mmxreg,mmxrm SSSE3,MMX )]441.8 code0 -[2(PMADDUBSW xmmreg,xmmrm SSSE3 )]430.8 code0 -[2(PMULHRSW mmxreg,mmxrm SSSE3,MMX )]419.8 code0 -[2(PMULHRSW xmmreg,xmmrm SSSE3 )]408.8 code0 -[2(PSHUFB mmxreg,mmxrm SSSE3,MMX )]397.8 code0 -[2(PSHUFB xmmreg,xmmrm SSSE3 )]386.8 code0 -[2(PSIGNB mmxreg,mmxrm SSSE3,MMX )]375.8 code0 -[2(PSIGNB xmmreg,xmmrm SSSE3 )]364.8 code0 -[2(PSIGNW mmxreg,mmxrm SSSE3,MMX )]353.8 code0 -[2(PSIGNW xmmreg,xmmrm SSSE3 )]342.8 code0 -[2(PSIGND mmxreg,mmxrm SSSE3,MMX )]331.8 code0 -[2(PSIGND xmmreg,xmmrm SSSE3)]320.8 code2 -[{/section-B.1.16 xa}(AMD SSE4A)](B.1.16)301.6 subh3 -[2(EXTRQ xmmreg,imm,imm SSE4A,AMD )]284.6 code1 -[2(EXTRQ xmmreg,xmmreg SSE4A,AMD )]273.6 code0 -[2(INSERTQ xmmreg,xmmreg,imm,imm SSE4A,AMD )]262.6 code0 -[2(INSERTQ xmmreg,xmmreg SSE4A,AMD )]251.6 code0 -[2(MOVNTSD mem,xmmreg SSE4A,AMD )]240.6 code0 -[2(MOVNTSS mem,xmmreg SSE4A,AMD,SD)]229.6 code2 -[{/section-B.1.17 xa}(New instructions in Barcelona)](B.1.17)210.4 subh3 -[2(LZCNT reg16,rm16 P6,AMD )]193.4 code1 -[2(LZCNT reg32,rm32 P6,AMD )]182.4 code0 -[2(LZCNT reg64,rm64 X64,AMD)]171.4 code2 -[{/section-B.1.18 xa}(Penryn New Instructions \(SSE4.1\))](B.1.18)152.2 subh3 -[2(BLENDPD xmmreg,xmmrm,imm SSE41 )]135.2 code1 -[2(BLENDPS xmmreg,xmmrm,imm SSE41 )]124.2 code0 -[2(BLENDVPD xmmreg,xmmrm,xmm0 SSE41 )]113.2 code0 -[2(BLENDVPS xmmreg,xmmrm,xmm0 SSE41 )]102.2 code0 -(155)pageodd -restore showpage -%%Page: 156 156 -%%BeginPageSetup -save -%%EndPageSetup -/156 pa -[2(DPPD xmmreg,xmmrm,imm SSE41 )]681 code0 -[2(DPPS xmmreg,xmmrm,imm SSE41 )]670 code0 -[2(EXTRACTPS rm32,xmmreg,imm SSE41 )]659 code0 -[2(EXTRACTPS reg64,xmmreg,imm SSE41,X64 )]648 code0 -[2(INSERTPS xmmreg,xmmrm,imm SSE41,SD )]637 code0 -[2(MOVNTDQA xmmreg,mem SSE41 )]626 code0 -[2(MPSADBW xmmreg,xmmrm,imm SSE41 )]615 code0 -[2(PACKUSDW xmmreg,xmmrm SSE41 )]604 code0 -[2(PBLENDVB xmmreg,xmmrm,xmm0 SSE41 )]593 code0 -[2(PBLENDW xmmreg,xmmrm,imm SSE41 )]582 code0 -[2(PCMPEQQ xmmreg,xmmrm SSE41 )]571 code0 -[2(PEXTRB reg32,xmmreg,imm SSE41 )]560 code0 -[2(PEXTRB mem8,xmmreg,imm SSE41 )]549 code0 -[2(PEXTRB reg64,xmmreg,imm SSE41,X64 )]538 code0 -[2(PEXTRD rm32,xmmreg,imm SSE41 )]527 code0 -[2(PEXTRQ rm64,xmmreg,imm SSE41,X64 )]516 code0 -[2(PEXTRW reg32,xmmreg,imm SSE41 )]505 code0 -[2(PEXTRW mem16,xmmreg,imm SSE41 )]494 code0 -[2(PEXTRW reg64,xmmreg,imm SSE41,X64 )]483 code0 -[2(PHMINPOSUW xmmreg,xmmrm SSE41 )]472 code0 -[2(PINSRB xmmreg,mem,imm SSE41 )]461 code0 -[2(PINSRB xmmreg,rm8,imm SSE41 )]450 code0 -[2(PINSRB xmmreg,reg32,imm SSE41 )]439 code0 -[2(PINSRD xmmreg,mem,imm SSE41 )]428 code0 -[2(PINSRD xmmreg,rm32,imm SSE41 )]417 code0 -[2(PINSRQ xmmreg,mem,imm SSE41,X64 )]406 code0 -[2(PINSRQ xmmreg,rm64,imm SSE41,X64 )]395 code0 -[2(PMAXSB xmmreg,xmmrm SSE41 )]384 code0 -[2(PMAXSD xmmreg,xmmrm SSE41 )]373 code0 -[2(PMAXUD xmmreg,xmmrm SSE41 )]362 code0 -[2(PMAXUW xmmreg,xmmrm SSE41 )]351 code0 -[2(PMINSB xmmreg,xmmrm SSE41 )]340 code0 -[2(PMINSD xmmreg,xmmrm SSE41 )]329 code0 -[2(PMINUD xmmreg,xmmrm SSE41 )]318 code0 -[2(PMINUW xmmreg,xmmrm SSE41 )]307 code0 -[2(PMOVSXBW xmmreg,xmmrm SSE41 )]296 code0 -[2(PMOVSXBD xmmreg,xmmrm SSE41,SD )]285 code0 -[2(PMOVSXBQ xmmreg,xmmrm SSE41,SW )]274 code0 -[2(PMOVSXWD xmmreg,xmmrm SSE41 )]263 code0 -[2(PMOVSXWQ xmmreg,xmmrm SSE41,SD )]252 code0 -[2(PMOVSXDQ xmmreg,xmmrm SSE41 )]241 code0 -[2(PMOVZXBW xmmreg,xmmrm SSE41 )]230 code0 -[2(PMOVZXBD xmmreg,xmmrm SSE41,SD )]219 code0 -[2(PMOVZXBQ xmmreg,xmmrm SSE41,SW )]208 code0 -[2(PMOVZXWD xmmreg,xmmrm SSE41 )]197 code0 -[2(PMOVZXWQ xmmreg,xmmrm SSE41,SD )]186 code0 -[2(PMOVZXDQ xmmreg,xmmrm SSE41 )]175 code0 -[2(PMULDQ xmmreg,xmmrm SSE41 )]164 code0 -[2(PMULLD xmmreg,xmmrm SSE41 )]153 code0 -[2(PTEST xmmreg,xmmrm SSE41 )]142 code0 -[2(ROUNDPD xmmreg,xmmrm,imm SSE41 )]131 code0 -[2(ROUNDPS xmmreg,xmmrm,imm SSE41 )]120 code0 -(156)pageeven -restore showpage -%%Page: 157 157 -%%BeginPageSetup -save -%%EndPageSetup -/157 pa -[2(ROUNDSD xmmreg,xmmrm,imm SSE41 )]681 code0 -[2(ROUNDSS xmmreg,xmmrm,imm SSE41)]670 code2 -[{/section-B.1.19 xa}(Nehalem New Instructions \(SSE4.2\))](B.1.19)650.8 subh3 -[2(CRC32 reg32,rm8 SSE42 )]633.8 code1 -[2(CRC32 reg32,rm16 SSE42 )]622.8 code0 -[2(CRC32 reg32,rm32 SSE42 )]611.8 code0 -[2(CRC32 reg64,rm8 SSE42,X64 )]600.8 code0 -[2(CRC32 reg64,rm64 SSE42,X64 )]589.8 code0 -[2(PCMPESTRI xmmreg,xmmrm,imm SSE42 )]578.8 code0 -[2(PCMPESTRM xmmreg,xmmrm,imm SSE42 )]567.8 code0 -[2(PCMPISTRI xmmreg,xmmrm,imm SSE42 )]556.8 code0 -[2(PCMPISTRM xmmreg,xmmrm,imm SSE42 )]545.8 code0 -[2(PCMPGTQ xmmreg,xmmrm SSE42 )]534.8 code0 -[2(POPCNT reg16,rm16 NEHALEM,SW )]523.8 code0 -[2(POPCNT reg32,rm32 NEHALEM,SD )]512.8 code0 -[2(POPCNT reg64,rm64 NEHALEM,X64)]501.8 code2 -[{/section-B.1.20 xa}(Intel SMX)](B.1.20)482.6 subh3 -[2(GETSEC KATMAI)]465.6 code3 -[{/section-B.1.21 xa}(Geode \(Cyrix\) 3DNow! additions)](B.1.21)446.4 subh3 -[2(PFRCPV mmxreg,mmxrm PENT,3DNOW,CYRIX )]429.4 code1 -[2(PFRSQRTV mmxreg,mmxrm PENT,3DNOW,CYRIX)]418.4 code2 -[{/section-B.1.22 xa}(Intel new instructions in ???)](B.1.22)399.2 subh3 -[2(MOVBE reg16,mem16 NEHALEM )]382.2 code1 -[2(MOVBE reg32,mem32 NEHALEM )]371.2 code0 -[2(MOVBE reg64,mem64 NEHALEM )]360.2 code0 -[2(MOVBE mem16,reg16 NEHALEM )]349.2 code0 -[2(MOVBE mem32,reg32 NEHALEM )]338.2 code0 -[2(MOVBE mem64,reg64 NEHALEM)]327.2 code2 -[{/section-B.1.23 xa}(Intel AES instructions)](B.1.23)308 subh3 -[2(AESENC xmmreg,xmmrm128 SSE,WESTMERE )]291 code1 -[2(AESENCLAST xmmreg,xmmrm128 SSE,WESTMERE )]280 code0 -[2(AESDEC xmmreg,xmmrm128 SSE,WESTMERE )]269 code0 -[2(AESDECLAST xmmreg,xmmrm128 SSE,WESTMERE )]258 code0 -[2(AESIMC xmmreg,xmmrm128 SSE,WESTMERE )]247 code0 -[2(AESKEYGENASSIST xmmreg,xmmrm128,imm8 SSE,WESTMERE)]236 code2 -[{/section-B.1.24 xa}(Intel AVX AES instructions)](B.1.24)216.8 subh3 -[2(VAESENC xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]199.8 code1 -[2(VAESENCLAST xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]188.8 code0 -[2(VAESDEC xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]177.8 code0 -[2(VAESDECLAST xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]166.8 code0 -[2(VAESIMC xmmreg,xmmrm128 AVX,SANDYBRIDGE )]155.8 code0 -[2(VAESKEYGENASSIST xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE)]144.8 code2 -(157)pageodd -restore showpage -%%Page: 158 158 -%%BeginPageSetup -save -%%EndPageSetup -/158 pa -[{/section-B.1.25 xa}(Intel AVX instructions)](B.1.25)678.8 subh3 -[2(VADDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]661.8 code1 -[2(VADDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]650.8 code0 -[2(VADDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]639.8 code0 -[2(VADDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]628.8 code0 -[2(VADDSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]617.8 code0 -[2(VADDSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]606.8 code0 -[2(VADDSUBPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]595.8 code0 -[2(VADDSUBPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]584.8 code0 -[2(VADDSUBPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]573.8 code0 -[2(VADDSUBPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]562.8 code0 -[2(VANDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]551.8 code0 -[2(VANDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]540.8 code0 -[2(VANDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]529.8 code0 -[2(VANDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]518.8 code0 -[2(VANDNPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]507.8 code0 -[2(VANDNPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]496.8 code0 -[2(VANDNPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]485.8 code0 -[2(VANDNPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]474.8 code0 -[2(VBLENDPD xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE )]463.8 code0 -[2(VBLENDPD ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE )]452.8 code0 -[2(VBLENDPS xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE )]441.8 code0 -[2(VBLENDPS ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE )]430.8 code0 -[2(VBLENDVPD xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE )]419.8 code0 -[2(VBLENDVPD xmmreg,xmmrm128,xmm0 AVX,SANDYBRIDGE )]408.8 code0 -[2(VBLENDVPD ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE )]397.8 code0 -[2(VBLENDVPD ymmreg,ymmrm256,ymm0 AVX,SANDYBRIDGE )]386.8 code0 -[2(VBLENDVPS xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE )]375.8 code0 -[2(VBLENDVPS xmmreg,xmmrm128,xmm0 AVX,SANDYBRIDGE )]364.8 code0 -[2(VBLENDVPS ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE )]353.8 code0 -[2(VBLENDVPD ymmreg,ymmrm256,ymm0 AVX,SANDYBRIDGE )]342.8 code0 -[2(VBROADCASTSS xmmreg,mem32 AVX,SANDYBRIDGE )]331.8 code0 -[2(VBROADCASTSS ymmreg,mem32 AVX,SANDYBRIDGE )]320.8 code0 -[2(VBROADCASTSD ymmreg,mem64 AVX,SANDYBRIDGE )]309.8 code0 -[2(VBROADCASTF128 ymmreg,mem128 AVX,SANDYBRIDGE )]298.8 code0 -[2(VCMPEQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]287.8 code0 -[2(VCMPEQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]276.8 code0 -[2(VCMPLTPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]265.8 code0 -[2(VCMPLTPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]254.8 code0 -[2(VCMPLEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]243.8 code0 -[2(VCMPLEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]232.8 code0 -[2(VCMPUNORDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]221.8 code0 -[2(VCMPUNORDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]210.8 code0 -[2(VCMPNEQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]199.8 code0 -[2(VCMPNEQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]188.8 code0 -[2(VCMPNLTPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]177.8 code0 -[2(VCMPNLTPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]166.8 code0 -[2(VCMPNLEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]155.8 code0 -[2(VCMPNLEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]144.8 code0 -[2(VCMPORDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]133.8 code0 -[2(VCMPORDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]122.8 code0 -[2(VCMPEQ_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]111.8 code0 -[2(VCMPEQ_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]100.8 code0 -(158)pageeven -restore showpage -%%Page: 159 159 -%%BeginPageSetup -save -%%EndPageSetup -/159 pa -[2(VCMPNGEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]681 code0 -[2(VCMPNGEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]670 code0 -[2(VCMPNGTPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]659 code0 -[2(VCMPNGTPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]648 code0 -[2(VCMPFALSEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]637 code0 -[2(VCMPFALSEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]626 code0 -[2(VCMPNEQ_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]615 code0 -[2(VCMPNEQ_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]604 code0 -[2(VCMPGEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]593 code0 -[2(VCMPGEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]582 code0 -[2(VCMPGTPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]571 code0 -[2(VCMPGTPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]560 code0 -[2(VCMPTRUEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]549 code0 -[2(VCMPTRUEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]538 code0 -[2(VCMPEQ_OSPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]527 code0 -[2(VCMPEQ_OSPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]516 code0 -[2(VCMPLT_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]505 code0 -[2(VCMPLT_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]494 code0 -[2(VCMPLE_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]483 code0 -[2(VCMPLE_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]472 code0 -[2(VCMPUNORD_SPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]461 code0 -[2(VCMPUNORD_SPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]450 code0 -[2(VCMPNEQ_USPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]439 code0 -[2(VCMPNEQ_USPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]428 code0 -[2(VCMPNLT_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]417 code0 -[2(VCMPNLT_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]406 code0 -[2(VCMPNLE_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]395 code0 -[2(VCMPNLE_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]384 code0 -[2(VCMPORD_SPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]373 code0 -[2(VCMPORD_SPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]362 code0 -[2(VCMPEQ_USPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]351 code0 -[2(VCMPEQ_USPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]340 code0 -[2(VCMPNGE_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]329 code0 -[2(VCMPNGE_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]318 code0 -[2(VCMPNGT_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]307 code0 -[2(VCMPNGT_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]296 code0 -[2(VCMPFALSE_OSPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]285 code0 -[2(VCMPFALSE_OSPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]274 code0 -[2(VCMPNEQ_OSPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]263 code0 -[2(VCMPNEQ_OSPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]252 code0 -[2(VCMPGE_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]241 code0 -[2(VCMPGE_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]230 code0 -[2(VCMPGT_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]219 code0 -[2(VCMPGT_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]208 code0 -[2(VCMPTRUE_USPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]197 code0 -[2(VCMPTRUE_USPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]186 code0 -[2(VCMPPD xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE )]175 code0 -[2(VCMPPD ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE )]164 code0 -[2(VCMPEQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]153 code0 -[2(VCMPEQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]142 code0 -[2(VCMPLTPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]131 code0 -[2(VCMPLTPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]120 code0 -[2(VCMPLEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]109 code0 -(159)pageodd -restore showpage -%%Page: 160 160 -%%BeginPageSetup -save -%%EndPageSetup -/160 pa -[2(VCMPLEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]681 code0 -[2(VCMPUNORDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]670 code0 -[2(VCMPUNORDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]659 code0 -[2(VCMPNEQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]648 code0 -[2(VCMPNEQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]637 code0 -[2(VCMPNLTPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]626 code0 -[2(VCMPNLTPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]615 code0 -[2(VCMPNLEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]604 code0 -[2(VCMPNLEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]593 code0 -[2(VCMPORDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]582 code0 -[2(VCMPORDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]571 code0 -[2(VCMPEQ_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]560 code0 -[2(VCMPEQ_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]549 code0 -[2(VCMPNGEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]538 code0 -[2(VCMPNGEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]527 code0 -[2(VCMPNGTPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]516 code0 -[2(VCMPNGTPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]505 code0 -[2(VCMPFALSEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]494 code0 -[2(VCMPFALSEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]483 code0 -[2(VCMPNEQ_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]472 code0 -[2(VCMPNEQ_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]461 code0 -[2(VCMPGEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]450 code0 -[2(VCMPGEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]439 code0 -[2(VCMPGTPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]428 code0 -[2(VCMPGTPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]417 code0 -[2(VCMPTRUEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]406 code0 -[2(VCMPTRUEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]395 code0 -[2(VCMPEQ_OSPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]384 code0 -[2(VCMPEQ_OSPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]373 code0 -[2(VCMPLT_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]362 code0 -[2(VCMPLT_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]351 code0 -[2(VCMPLE_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]340 code0 -[2(VCMPLE_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]329 code0 -[2(VCMPUNORD_SPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]318 code0 -[2(VCMPUNORD_SPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]307 code0 -[2(VCMPNEQ_USPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]296 code0 -[2(VCMPNEQ_USPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]285 code0 -[2(VCMPNLT_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]274 code0 -[2(VCMPNLT_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]263 code0 -[2(VCMPNLE_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]252 code0 -[2(VCMPNLE_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]241 code0 -[2(VCMPORD_SPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]230 code0 -[2(VCMPORD_SPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]219 code0 -[2(VCMPEQ_USPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]208 code0 -[2(VCMPEQ_USPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]197 code0 -[2(VCMPNGE_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]186 code0 -[2(VCMPNGE_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]175 code0 -[2(VCMPNGT_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]164 code0 -[2(VCMPNGT_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]153 code0 -[2(VCMPFALSE_OSPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]142 code0 -[2(VCMPFALSE_OSPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]131 code0 -[2(VCMPNEQ_OSPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]120 code0 -[2(VCMPNEQ_OSPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]109 code0 -(160)pageeven -restore showpage -%%Page: 161 161 -%%BeginPageSetup -save -%%EndPageSetup -/161 pa -[2(VCMPGE_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]681 code0 -[2(VCMPGE_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]670 code0 -[2(VCMPGT_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]659 code0 -[2(VCMPGT_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]648 code0 -[2(VCMPTRUE_USPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]637 code0 -[2(VCMPTRUE_USPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]626 code0 -[2(VCMPPS xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE )]615 code0 -[2(VCMPPS ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE )]604 code0 -[2(VCMPEQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]593 code0 -[2(VCMPLTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]582 code0 -[2(VCMPLESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]571 code0 -[2(VCMPUNORDSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]560 code0 -[2(VCMPNEQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]549 code0 -[2(VCMPNLTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]538 code0 -[2(VCMPNLESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]527 code0 -[2(VCMPORDSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]516 code0 -[2(VCMPEQ_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]505 code0 -[2(VCMPNGESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]494 code0 -[2(VCMPNGTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]483 code0 -[2(VCMPFALSESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]472 code0 -[2(VCMPNEQ_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]461 code0 -[2(VCMPGESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]450 code0 -[2(VCMPGTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]439 code0 -[2(VCMPTRUESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]428 code0 -[2(VCMPEQ_OSSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]417 code0 -[2(VCMPLT_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]406 code0 -[2(VCMPLE_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]395 code0 -[2(VCMPUNORD_SSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]384 code0 -[2(VCMPNEQ_USSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]373 code0 -[2(VCMPNLT_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]362 code0 -[2(VCMPNLE_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]351 code0 -[2(VCMPORD_SSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]340 code0 -[2(VCMPEQ_USSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]329 code0 -[2(VCMPNGE_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]318 code0 -[2(VCMPNGT_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]307 code0 -[2(VCMPFALSE_OSSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]296 code0 -[2(VCMPNEQ_OSSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]285 code0 -[2(VCMPGE_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]274 code0 -[2(VCMPGT_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]263 code0 -[2(VCMPTRUE_USSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]252 code0 -[2(VCMPSD xmmreg,xmmreg*,xmmrm64,imm8 AVX,SANDYBRIDGE )]241 code0 -[2(VCMPEQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]230 code0 -[2(VCMPLTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]219 code0 -[2(VCMPLESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]208 code0 -[2(VCMPUNORDSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]197 code0 -[2(VCMPNEQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]186 code0 -[2(VCMPNLTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]175 code0 -[2(VCMPNLESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]164 code0 -[2(VCMPORDSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]153 code0 -[2(VCMPEQ_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]142 code0 -[2(VCMPNGESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]131 code0 -[2(VCMPNGTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]120 code0 -[2(VCMPFALSESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]109 code0 -(161)pageodd -restore showpage -%%Page: 162 162 -%%BeginPageSetup -save -%%EndPageSetup -/162 pa -[2(VCMPNEQ_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]681 code0 -[2(VCMPGESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]670 code0 -[2(VCMPGTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]659 code0 -[2(VCMPTRUESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]648 code0 -[2(VCMPEQ_OSSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]637 code0 -[2(VCMPLT_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]626 code0 -[2(VCMPLE_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]615 code0 -[2(VCMPUNORD_SSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]604 code0 -[2(VCMPNEQ_USSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]593 code0 -[2(VCMPNLT_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]582 code0 -[2(VCMPNLE_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]571 code0 -[2(VCMPORD_SSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]560 code0 -[2(VCMPEQ_USSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]549 code0 -[2(VCMPNGE_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]538 code0 -[2(VCMPNGT_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]527 code0 -[2(VCMPFALSE_OSSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]516 code0 -[2(VCMPNEQ_OSSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]505 code0 -[2(VCMPGE_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]494 code0 -[2(VCMPGT_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]483 code0 -[2(VCMPTRUE_USSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]472 code0 -[2(VCMPSS xmmreg,xmmreg*,xmmrm32,imm8 AVX,SANDYBRIDGE )]461 code0 -[2(VCOMISD xmmreg,xmmrm64 AVX,SANDYBRIDGE )]450 code0 -[2(VCOMISS xmmreg,xmmrm32 AVX,SANDYBRIDGE )]439 code0 -[2(VCVTDQ2PD xmmreg,xmmrm64 AVX,SANDYBRIDGE )]428 code0 -[2(VCVTDQ2PD ymmreg,xmmrm128 AVX,SANDYBRIDGE )]417 code0 -[2(VCVTDQ2PS xmmreg,xmmrm128 AVX,SANDYBRIDGE )]406 code0 -[2(VCVTDQ2PS ymmreg,ymmrm256 AVX,SANDYBRIDGE )]395 code0 -[2(VCVTPD2DQ xmmreg,xmmreg AVX,SANDYBRIDGE )]384 code0 -[2(VCVTPD2DQ xmmreg,mem128 AVX,SANDYBRIDGE,SO )]373 code0 -[2(VCVTPD2DQ xmmreg,ymmreg AVX,SANDYBRIDGE )]362 code0 -[2(VCVTPD2DQ xmmreg,mem256 AVX,SANDYBRIDGE,SY )]351 code0 -[2(VCVTPD2PS xmmreg,xmmreg AVX,SANDYBRIDGE )]340 code0 -[2(VCVTPD2PS xmmreg,mem128 AVX,SANDYBRIDGE,SO )]329 code0 -[2(VCVTPD2PS xmmreg,ymmreg AVX,SANDYBRIDGE )]318 code0 -[2(VCVTPD2PS xmmreg,mem256 AVX,SANDYBRIDGE,SY )]307 code0 -[2(VCVTPS2DQ xmmreg,xmmrm128 AVX,SANDYBRIDGE )]296 code0 -[2(VCVTPS2DQ ymmreg,ymmrm256 AVX,SANDYBRIDGE )]285 code0 -[2(VCVTPS2PD xmmreg,xmmrm64 AVX,SANDYBRIDGE )]274 code0 -[2(VCVTPS2PD ymmreg,xmmrm128 AVX,SANDYBRIDGE )]263 code0 -[2(VCVTSD2SI reg32,xmmrm64 AVX,SANDYBRIDGE )]252 code0 -[2(VCVTSD2SI reg64,xmmrm64 AVX,SANDYBRIDGE,LONG )]241 code0 -[2(VCVTSD2SS xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]230 code0 -[2(VCVTSI2SD xmmreg,xmmreg*,rm32 AVX,SANDYBRIDGE,SD )]219 code0 -[2(VCVTSI2SD xmmreg,xmmreg*,mem32 AVX,SANDYBRIDGE,ND,SD )]208 code0 -[2(VCVTSI2SD xmmreg,xmmreg*,rm64 AVX,SANDYBRIDGE,LONG )]197 code0 -[2(VCVTSI2SS xmmreg,xmmreg*,rm32 AVX,SANDYBRIDGE,SD )]186 code0 -[2(VCVTSI2SS xmmreg,xmmreg*,mem32 AVX,SANDYBRIDGE,ND,SD )]175 code0 -[2(VCVTSI2SS xmmreg,xmmreg*,rm64 AVX,SANDYBRIDGE,LONG )]164 code0 -[2(VCVTSS2SD xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]153 code0 -[2(VCVTSS2SI reg32,xmmrm32 AVX,SANDYBRIDGE )]142 code0 -[2(VCVTSS2SI reg64,xmmrm32 AVX,SANDYBRIDGE,LONG )]131 code0 -[2(VCVTTPD2DQ xmmreg,xmmreg AVX,SANDYBRIDGE )]120 code0 -[2(VCVTTPD2DQ xmmreg,mem128 AVX,SANDYBRIDGE,SO )]109 code0 -(162)pageeven -restore showpage -%%Page: 163 163 -%%BeginPageSetup -save -%%EndPageSetup -/163 pa -[2(VCVTTPD2DQ xmmreg,ymmreg AVX,SANDYBRIDGE )]681 code0 -[2(VCVTTPD2DQ xmmreg,mem256 AVX,SANDYBRIDGE,SY )]670 code0 -[2(VCVTTPS2DQ xmmreg,xmmrm128 AVX,SANDYBRIDGE )]659 code0 -[2(VCVTTPS2DQ ymmreg,ymmrm256 AVX,SANDYBRIDGE )]648 code0 -[2(VCVTTSD2SI reg32,xmmrm64 AVX,SANDYBRIDGE )]637 code0 -[2(VCVTTSD2SI reg64,xmmrm64 AVX,SANDYBRIDGE,LONG )]626 code0 -[2(VCVTTSS2SI reg32,xmmrm32 AVX,SANDYBRIDGE )]615 code0 -[2(VCVTTSS2SI reg64,xmmrm32 AVX,SANDYBRIDGE,LONG )]604 code0 -[2(VDIVPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]593 code0 -[2(VDIVPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]582 code0 -[2(VDIVPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]571 code0 -[2(VDIVPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]560 code0 -[2(VDIVSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]549 code0 -[2(VDIVSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]538 code0 -[2(VDPPD xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE )]527 code0 -[2(VDPPS xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE )]516 code0 -[2(VDPPS ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE )]505 code0 -[2(VEXTRACTF128 xmmrm128,xmmreg,imm8 AVX,SANDYBRIDGE )]494 code0 -[2(VEXTRACTPS rm32,xmmreg,imm8 AVX,SANDYBRIDGE )]483 code0 -[2(VHADDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]472 code0 -[2(VHADDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]461 code0 -[2(VHADDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]450 code0 -[2(VHADDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]439 code0 -[2(VHSUBPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]428 code0 -[2(VHSUBPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]417 code0 -[2(VHSUBPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]406 code0 -[2(VHSUBPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]395 code0 -[2(VINSERTF128 ymmreg,ymmreg,xmmrm128,imm8 AVX,SANDYBRIDGE )]384 code0 -[2(VINSERTPS xmmreg,xmmreg*,xmmrm32,imm8 AVX,SANDYBRIDGE )]373 code0 -[2(VLDDQU xmmreg,mem128 AVX,SANDYBRIDGE )]362 code0 -[2(VLDQQU ymmreg,mem256 AVX,SANDYBRIDGE )]351 code0 -[2(VLDDQU ymmreg,mem256 AVX,SANDYBRIDGE )]340 code0 -[2(VLDMXCSR mem32 AVX,SANDYBRIDGE )]329 code0 -[2(VMASKMOVDQU xmmreg,xmmreg AVX,SANDYBRIDGE )]318 code0 -[2(VMASKMOVPS xmmreg,xmmreg,mem128 AVX,SANDYBRIDGE )]307 code0 -[2(VMASKMOVPS ymmreg,ymmreg,mem256 AVX,SANDYBRIDGE )]296 code0 -[2(VMASKMOVPS mem128,xmmreg,xmmreg AVX,SANDYBRIDGE,SO )]285 code0 -[2(VMASKMOVPS mem256,xmmreg,xmmreg AVX,SANDYBRIDGE,SY )]274 code0 -[2(VMASKMOVPD xmmreg,xmmreg,mem128 AVX,SANDYBRIDGE )]263 code0 -[2(VMASKMOVPD ymmreg,ymmreg,mem256 AVX,SANDYBRIDGE )]252 code0 -[2(VMASKMOVPD mem128,xmmreg,xmmreg AVX,SANDYBRIDGE )]241 code0 -[2(VMASKMOVPD mem256,ymmreg,ymmreg AVX,SANDYBRIDGE )]230 code0 -[2(VMAXPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]219 code0 -[2(VMAXPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]208 code0 -[2(VMAXPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]197 code0 -[2(VMAXPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]186 code0 -[2(VMAXSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]175 code0 -[2(VMAXSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]164 code0 -[2(VMINPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]153 code0 -[2(VMINPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]142 code0 -[2(VMINPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]131 code0 -[2(VMINPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]120 code0 -[2(VMINSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]109 code0 -(163)pageodd -restore showpage -%%Page: 164 164 -%%BeginPageSetup -save -%%EndPageSetup -/164 pa -[2(VMINSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]681 code0 -[2(VMOVAPD xmmreg,xmmrm128 AVX,SANDYBRIDGE )]670 code0 -[2(VMOVAPD xmmrm128,xmmreg AVX,SANDYBRIDGE )]659 code0 -[2(VMOVAPD ymmreg,ymmrm256 AVX,SANDYBRIDGE )]648 code0 -[2(VMOVAPD ymmrm256,ymmreg AVX,SANDYBRIDGE )]637 code0 -[2(VMOVAPS xmmreg,xmmrm128 AVX,SANDYBRIDGE )]626 code0 -[2(VMOVAPS xmmrm128,xmmreg AVX,SANDYBRIDGE )]615 code0 -[2(VMOVAPS ymmreg,ymmrm256 AVX,SANDYBRIDGE )]604 code0 -[2(VMOVAPS ymmrm256,ymmreg AVX,SANDYBRIDGE )]593 code0 -[2(VMOVQ xmmreg,xmmrm64 AVX,SANDYBRIDGE )]582 code0 -[2(VMOVQ xmmrm64,xmmreg AVX,SANDYBRIDGE )]571 code0 -[2(VMOVQ xmmreg,rm64 AVX,SANDYBRIDGE,LONG )]560 code0 -[2(VMOVQ rm64,xmmreg AVX,SANDYBRIDGE,LONG )]549 code0 -[2(VMOVD xmmreg,rm32 AVX,SANDYBRIDGE )]538 code0 -[2(VMOVD rm32,xmmreg AVX,SANDYBRIDGE )]527 code0 -[2(VMOVDDUP xmmreg,xmmrm64 AVX,SANDYBRIDGE )]516 code0 -[2(VMOVDDUP ymmreg,ymmrm256 AVX,SANDYBRIDGE )]505 code0 -[2(VMOVDQA xmmreg,xmmrm128 AVX,SANDYBRIDGE )]494 code0 -[2(VMOVDQA xmmrm128,xmmreg AVX,SANDYBRIDGE )]483 code0 -[2(VMOVQQA ymmreg,ymmrm256 AVX,SANDYBRIDGE )]472 code0 -[2(VMOVQQA ymmrm256,ymmreg AVX,SANDYBRIDGE )]461 code0 -[2(VMOVDQA ymmreg,ymmrm AVX,SANDYBRIDGE )]450 code0 -[2(VMOVDQA ymmrm256,ymmreg AVX,SANDYBRIDGE )]439 code0 -[2(VMOVDQU xmmreg,xmmrm128 AVX,SANDYBRIDGE )]428 code0 -[2(VMOVDQU xmmrm128,xmmreg AVX,SANDYBRIDGE )]417 code0 -[2(VMOVQQU ymmreg,ymmrm256 AVX,SANDYBRIDGE )]406 code0 -[2(VMOVQQU ymmrm256,ymmreg AVX,SANDYBRIDGE )]395 code0 -[2(VMOVDQU ymmreg,ymmrm256 AVX,SANDYBRIDGE )]384 code0 -[2(VMOVDQU ymmrm256,ymmreg AVX,SANDYBRIDGE )]373 code0 -[2(VMOVHLPS xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE )]362 code0 -[2(VMOVHPD xmmreg,xmmreg*,mem64 AVX,SANDYBRIDGE )]351 code0 -[2(VMOVHPD mem64,xmmreg AVX,SANDYBRIDGE )]340 code0 -[2(VMOVHPS xmmreg,xmmreg*,mem64 AVX,SANDYBRIDGE )]329 code0 -[2(VMOVHPS mem64,xmmreg AVX,SANDYBRIDGE )]318 code0 -[2(VMOVLHPS xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE )]307 code0 -[2(VMOVLPD xmmreg,xmmreg*,mem64 AVX,SANDYBRIDGE )]296 code0 -[2(VMOVLPD mem64,xmmreg AVX,SANDYBRIDGE )]285 code0 -[2(VMOVLPS xmmreg,xmmreg*,mem64 AVX,SANDYBRIDGE )]274 code0 -[2(VMOVLPS mem64,xmmreg AVX,SANDYBRIDGE )]263 code0 -[2(VMOVMSKPD reg64,xmmreg AVX,SANDYBRIDGE,LONG )]252 code0 -[2(VMOVMSKPD reg32,xmmreg AVX,SANDYBRIDGE )]241 code0 -[2(VMOVMSKPD reg64,ymmreg AVX,SANDYBRIDGE,LONG )]230 code0 -[2(VMOVMSKPD reg32,ymmreg AVX,SANDYBRIDGE )]219 code0 -[2(VMOVMSKPS reg64,xmmreg AVX,SANDYBRIDGE,LONG )]208 code0 -[2(VMOVMSKPS reg32,xmmreg AVX,SANDYBRIDGE )]197 code0 -[2(VMOVMSKPS reg64,ymmreg AVX,SANDYBRIDGE,LONG )]186 code0 -[2(VMOVMSKPS reg32,ymmreg AVX,SANDYBRIDGE )]175 code0 -[2(VMOVNTDQ mem128,xmmreg AVX,SANDYBRIDGE )]164 code0 -[2(VMOVNTQQ mem256,ymmreg AVX,SANDYBRIDGE )]153 code0 -[2(VMOVNTDQ mem256,ymmreg AVX,SANDYBRIDGE )]142 code0 -[2(VMOVNTDQA xmmreg,mem128 AVX,SANDYBRIDGE )]131 code0 -[2(VMOVNTPD mem128,xmmreg AVX,SANDYBRIDGE )]120 code0 -[2(VMOVNTPD mem256,ymmreg AVX,SANDYBRIDGE )]109 code0 -(164)pageeven -restore showpage -%%Page: 165 165 -%%BeginPageSetup -save -%%EndPageSetup -/165 pa -[2(VMOVNTPS mem128,xmmreg AVX,SANDYBRIDGE )]681 code0 -[2(VMOVNTPS mem128,ymmreg AVX,SANDYBRIDGE )]670 code0 -[2(VMOVSD xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE )]659 code0 -[2(VMOVSD xmmreg,mem64 AVX,SANDYBRIDGE )]648 code0 -[2(VMOVSD xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE )]637 code0 -[2(VMOVSD mem64,xmmreg AVX,SANDYBRIDGE )]626 code0 -[2(VMOVSHDUP xmmreg,xmmrm128 AVX,SANDYBRIDGE )]615 code0 -[2(VMOVSHDUP ymmreg,ymmrm256 AVX,SANDYBRIDGE )]604 code0 -[2(VMOVSLDUP xmmreg,xmmrm128 AVX,SANDYBRIDGE )]593 code0 -[2(VMOVSLDUP ymmreg,ymmrm256 AVX,SANDYBRIDGE )]582 code0 -[2(VMOVSS xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE )]571 code0 -[2(VMOVSS xmmreg,mem64 AVX,SANDYBRIDGE )]560 code0 -[2(VMOVSS xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE )]549 code0 -[2(VMOVSS mem64,xmmreg AVX,SANDYBRIDGE )]538 code0 -[2(VMOVUPD xmmreg,xmmrm128 AVX,SANDYBRIDGE )]527 code0 -[2(VMOVUPD xmmrm128,xmmreg AVX,SANDYBRIDGE )]516 code0 -[2(VMOVUPD ymmreg,ymmrm256 AVX,SANDYBRIDGE )]505 code0 -[2(VMOVUPD ymmrm256,ymmreg AVX,SANDYBRIDGE )]494 code0 -[2(VMOVUPS xmmreg,xmmrm128 AVX,SANDYBRIDGE )]483 code0 -[2(VMOVUPS xmmrm128,xmmreg AVX,SANDYBRIDGE )]472 code0 -[2(VMOVUPS ymmreg,ymmrm256 AVX,SANDYBRIDGE )]461 code0 -[2(VMOVUPS ymmrm256,ymmreg AVX,SANDYBRIDGE )]450 code0 -[2(VMPSADBW xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE )]439 code0 -[2(VMULPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]428 code0 -[2(VMULPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]417 code0 -[2(VMULPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]406 code0 -[2(VMULPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]395 code0 -[2(VMULSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]384 code0 -[2(VMULSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]373 code0 -[2(VORPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]362 code0 -[2(VORPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]351 code0 -[2(VORPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]340 code0 -[2(VORPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]329 code0 -[2(VPABSB xmmreg,xmmrm128 AVX,SANDYBRIDGE )]318 code0 -[2(VPABSW xmmreg,xmmrm128 AVX,SANDYBRIDGE )]307 code0 -[2(VPABSD xmmreg,xmmrm128 AVX,SANDYBRIDGE )]296 code0 -[2(VPACKSSWB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]285 code0 -[2(VPACKSSDW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]274 code0 -[2(VPACKUSWB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]263 code0 -[2(VPACKUSDW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]252 code0 -[2(VPADDB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]241 code0 -[2(VPADDW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]230 code0 -[2(VPADDD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]219 code0 -[2(VPADDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]208 code0 -[2(VPADDSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]197 code0 -[2(VPADDSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]186 code0 -[2(VPADDUSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]175 code0 -[2(VPADDUSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]164 code0 -[2(VPALIGNR xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE )]153 code0 -[2(VPAND xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]142 code0 -[2(VPANDN xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]131 code0 -[2(VPAVGB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]120 code0 -[2(VPAVGW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]109 code0 -(165)pageodd -restore showpage -%%Page: 166 166 -%%BeginPageSetup -save -%%EndPageSetup -/166 pa -[2(VPBLENDVB xmmreg,xmmreg*,xmmrm128,xmmreg AVX,SANDYBRIDGE )]681 code0 -[2(VPBLENDW xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE )]670 code0 -[2(VPCMPESTRI xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE )]659 code0 -[2(VPCMPESTRM xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE )]648 code0 -[2(VPCMPISTRI xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE )]637 code0 -[2(VPCMPISTRM xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE )]626 code0 -[2(VPCMPEQB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]615 code0 -[2(VPCMPEQW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]604 code0 -[2(VPCMPEQD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]593 code0 -[2(VPCMPEQQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]582 code0 -[2(VPCMPGTB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]571 code0 -[2(VPCMPGTW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]560 code0 -[2(VPCMPGTD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]549 code0 -[2(VPCMPGTQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]538 code0 -[2(VPERMILPD xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE )]527 code0 -[2(VPERMILPD ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE )]516 code0 -[2(VPERMILPD xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE )]505 code0 -[2(VPERMILPD ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE )]494 code0 -[2(VPERMILTD2PD xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE )]483 code0 -[2(VPERMILTD2PD xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE )]472 code0 -[2(VPERMILTD2PD ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE )]461 code0 -[2(VPERMILTD2PD ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE )]450 code0 -[2(VPERMILMO2PD xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE )]439 code0 -[2(VPERMILMO2PD xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE )]428 code0 -[2(VPERMILMO2PD ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE )]417 code0 -[2(VPERMILMO2PD ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE )]406 code0 -[2(VPERMILMZ2PD xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE )]395 code0 -[2(VPERMILMZ2PD xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE )]384 code0 -[2(VPERMILMZ2PD ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE )]373 code0 -[2(VPERMILMZ2PD ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE )]362 code0 -[2(VPERMIL2PD xmmreg,xmmreg,xmmrm128,xmmreg,imm8 AVX,SANDYBRIDGE )]351 code0 -[2(VPERMIL2PD xmmreg,xmmreg,xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE )]340 code0 -[2(VPERMIL2PD ymmreg,ymmreg,ymmrm256,ymmreg,imm8 AVX,SANDYBRIDGE )]329 code0 -[2(VPERMIL2PD ymmreg,ymmreg,ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE )]318 code0 -[2(VPERMILPS xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE )]307 code0 -[2(VPERMILPS ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE )]296 code0 -[2(VPERMILPS xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE )]285 code0 -[2(VPERMILPS ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE )]274 code0 -[2(VPERMILTD2PS xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE )]263 code0 -[2(VPERMILTD2PS xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE )]252 code0 -[2(VPERMILTD2PS ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE )]241 code0 -[2(VPERMILTD2PS ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE )]230 code0 -[2(VPERMILMO2PS xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE )]219 code0 -[2(VPERMILMO2PS xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE )]208 code0 -[2(VPERMILMO2PS ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE )]197 code0 -[2(VPERMILMO2PS ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE )]186 code0 -[2(VPERMILMZ2PS xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE )]175 code0 -[2(VPERMILMZ2PS xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE )]164 code0 -[2(VPERMILMZ2PS ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE )]153 code0 -[2(VPERMILMZ2PS ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE )]142 code0 -[2(VPERMIL2PS xmmreg,xmmreg,xmmrm128,xmmreg,imm8 AVX,SANDYBRIDGE )]131 code0 -[2(VPERMIL2PS xmmreg,xmmreg,xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE )]120 code0 -[2(VPERMIL2PS ymmreg,ymmreg,ymmrm256,ymmreg,imm8 AVX,SANDYBRIDGE )]109 code0 -(166)pageeven -restore showpage -%%Page: 167 167 -%%BeginPageSetup -save -%%EndPageSetup -/167 pa -[2(VPERMIL2PS ymmreg,ymmreg,ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE )]681 code0 -[2(VPERM2F128 ymmreg,ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE )]670 code0 -[2(VPEXTRB reg64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG )]659 code0 -[2(VPEXTRB reg32,xmmreg,imm8 AVX,SANDYBRIDGE )]648 code0 -[2(VPEXTRB mem8,xmmreg,imm8 AVX,SANDYBRIDGE )]637 code0 -[2(VPEXTRW reg64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG )]626 code0 -[2(VPEXTRW reg32,xmmreg,imm8 AVX,SANDYBRIDGE )]615 code0 -[2(VPEXTRW mem16,xmmreg,imm8 AVX,SANDYBRIDGE )]604 code0 -[2(VPEXTRW reg64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG )]593 code0 -[2(VPEXTRW reg32,xmmreg,imm8 AVX,SANDYBRIDGE )]582 code0 -[2(VPEXTRW mem16,xmmreg,imm8 AVX,SANDYBRIDGE )]571 code0 -[2(VPEXTRD reg64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG )]560 code0 -[2(VPEXTRD rm32,xmmreg,imm8 AVX,SANDYBRIDGE )]549 code0 -[2(VPEXTRQ rm64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG )]538 code0 -[2(VPHADDW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]527 code0 -[2(VPHADDD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]516 code0 -[2(VPHADDSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]505 code0 -[2(VPHMINPOSUW xmmreg,xmmrm128 AVX,SANDYBRIDGE )]494 code0 -[2(VPHSUBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]483 code0 -[2(VPHSUBD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]472 code0 -[2(VPHSUBSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]461 code0 -[2(VPINSRB xmmreg,xmmreg*,mem8,imm8 AVX,SANDYBRIDGE )]450 code0 -[2(VPINSRB xmmreg,xmmreg*,rm8,imm8 AVX,SANDYBRIDGE )]439 code0 -[2(VPINSRB xmmreg,xmmreg*,reg32,imm8 AVX,SANDYBRIDGE )]428 code0 -[2(VPINSRW xmmreg,xmmreg*,mem16,imm8 AVX,SANDYBRIDGE )]417 code0 -[2(VPINSRW xmmreg,xmmreg*,rm16,imm8 AVX,SANDYBRIDGE )]406 code0 -[2(VPINSRW xmmreg,xmmreg*,reg32,imm8 AVX,SANDYBRIDGE )]395 code0 -[2(VPINSRD xmmreg,xmmreg*,mem32,imm8 AVX,SANDYBRIDGE )]384 code0 -[2(VPINSRD xmmreg,xmmreg*,rm32,imm8 AVX,SANDYBRIDGE )]373 code0 -[2(VPINSRQ xmmreg,xmmreg*,mem64,imm8 AVX,SANDYBRIDGE,LONG )]362 code0 -[2(VPINSRQ xmmreg,xmmreg*,rm64,imm8 AVX,SANDYBRIDGE,LONG )]351 code0 -[2(VPMADDWD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]340 code0 -[2(VPMADDUBSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]329 code0 -[2(VPMAXSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]318 code0 -[2(VPMAXSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]307 code0 -[2(VPMAXSD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]296 code0 -[2(VPMAXUB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]285 code0 -[2(VPMAXUW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]274 code0 -[2(VPMAXUD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]263 code0 -[2(VPMINSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]252 code0 -[2(VPMINSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]241 code0 -[2(VPMINSD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]230 code0 -[2(VPMINUB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]219 code0 -[2(VPMINUW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]208 code0 -[2(VPMINUD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]197 code0 -[2(VPMOVMSKB reg64,xmmreg AVX,SANDYBRIDGE,LONG )]186 code0 -[2(VPMOVMSKB reg32,xmmreg AVX,SANDYBRIDGE )]175 code0 -[2(VPMOVSXBW xmmreg,xmmrm64 AVX,SANDYBRIDGE )]164 code0 -[2(VPMOVSXBD xmmreg,xmmrm32 AVX,SANDYBRIDGE )]153 code0 -[2(VPMOVSXBQ xmmreg,xmmrm16 AVX,SANDYBRIDGE )]142 code0 -[2(VPMOVSXWD xmmreg,xmmrm64 AVX,SANDYBRIDGE )]131 code0 -[2(VPMOVSXWQ xmmreg,xmmrm32 AVX,SANDYBRIDGE )]120 code0 -[2(VPMOVSXDQ xmmreg,xmmrm64 AVX,SANDYBRIDGE )]109 code0 -(167)pageodd -restore showpage -%%Page: 168 168 -%%BeginPageSetup -save -%%EndPageSetup -/168 pa -[2(VPMOVZXBW xmmreg,xmmrm64 AVX,SANDYBRIDGE )]681 code0 -[2(VPMOVZXBD xmmreg,xmmrm32 AVX,SANDYBRIDGE )]670 code0 -[2(VPMOVZXBQ xmmreg,xmmrm16 AVX,SANDYBRIDGE )]659 code0 -[2(VPMOVZXWD xmmreg,xmmrm64 AVX,SANDYBRIDGE )]648 code0 -[2(VPMOVZXWQ xmmreg,xmmrm32 AVX,SANDYBRIDGE )]637 code0 -[2(VPMOVZXDQ xmmreg,xmmrm64 AVX,SANDYBRIDGE )]626 code0 -[2(VPMULHUW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]615 code0 -[2(VPMULHRSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]604 code0 -[2(VPMULHW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]593 code0 -[2(VPMULLW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]582 code0 -[2(VPMULLD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]571 code0 -[2(VPMULUDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]560 code0 -[2(VPMULDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]549 code0 -[2(VPOR xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]538 code0 -[2(VPSADBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]527 code0 -[2(VPSHUFB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]516 code0 -[2(VPSHUFD xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE )]505 code0 -[2(VPSHUFHW xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE )]494 code0 -[2(VPSHUFLW xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE )]483 code0 -[2(VPSIGNB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]472 code0 -[2(VPSIGNW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]461 code0 -[2(VPSIGND xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]450 code0 -[2(VPSLLDQ xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE )]439 code0 -[2(VPSRLDQ xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE )]428 code0 -[2(VPSLLW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]417 code0 -[2(VPSLLW xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE )]406 code0 -[2(VPSLLD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]395 code0 -[2(VPSLLD xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE )]384 code0 -[2(VPSLLQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]373 code0 -[2(VPSLLQ xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE )]362 code0 -[2(VPSRAW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]351 code0 -[2(VPSRAW xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE )]340 code0 -[2(VPSRAD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]329 code0 -[2(VPSRAD xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE )]318 code0 -[2(VPSRLW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]307 code0 -[2(VPSRLW xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE )]296 code0 -[2(VPSRLD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]285 code0 -[2(VPSRLD xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE )]274 code0 -[2(VPSRLQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]263 code0 -[2(VPSRLQ xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE )]252 code0 -[2(VPTEST xmmreg,xmmrm128 AVX,SANDYBRIDGE )]241 code0 -[2(VPTEST ymmreg,ymmrm256 AVX,SANDYBRIDGE )]230 code0 -[2(VPSUBB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]219 code0 -[2(VPSUBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]208 code0 -[2(VPSUBD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]197 code0 -[2(VPSUBQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]186 code0 -[2(VPSUBSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]175 code0 -[2(VPSUBSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]164 code0 -[2(VPSUBUSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]153 code0 -[2(VPSUBUSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]142 code0 -[2(VPUNPCKHBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]131 code0 -[2(VPUNPCKHWD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]120 code0 -[2(VPUNPCKHDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]109 code0 -(168)pageeven -restore showpage -%%Page: 169 169 -%%BeginPageSetup -save -%%EndPageSetup -/169 pa -[2(VPUNPCKHQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]681 code0 -[2(VPUNPCKLBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]670 code0 -[2(VPUNPCKLWD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]659 code0 -[2(VPUNPCKLDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]648 code0 -[2(VPUNPCKLQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]637 code0 -[2(VPXOR xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]626 code0 -[2(VRCPPS xmmreg,xmmrm128 AVX,SANDYBRIDGE )]615 code0 -[2(VRCPPS ymmreg,ymmrm256 AVX,SANDYBRIDGE )]604 code0 -[2(VRCPSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]593 code0 -[2(VRSQRTPS xmmreg,xmmrm128 AVX,SANDYBRIDGE )]582 code0 -[2(VRSQRTPS ymmreg,ymmrm256 AVX,SANDYBRIDGE )]571 code0 -[2(VRSQRTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]560 code0 -[2(VROUNDPD xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE )]549 code0 -[2(VROUNDPD ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE )]538 code0 -[2(VROUNDPS xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE )]527 code0 -[2(VROUNDPS ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE )]516 code0 -[2(VROUNDSD xmmreg,xmmreg*,xmmrm64,imm8 AVX,SANDYBRIDGE )]505 code0 -[2(VROUNDSS xmmreg,xmmreg*,xmmrm32,imm8 AVX,SANDYBRIDGE )]494 code0 -[2(VSHUFPD xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE )]483 code0 -[2(VSHUFPD ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE )]472 code0 -[2(VSHUFPS xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE )]461 code0 -[2(VSHUFPS ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE )]450 code0 -[2(VSQRTPD xmmreg,xmmrm128 AVX,SANDYBRIDGE )]439 code0 -[2(VSQRTPD ymmreg,ymmrm256 AVX,SANDYBRIDGE )]428 code0 -[2(VSQRTPS xmmreg,xmmrm128 AVX,SANDYBRIDGE )]417 code0 -[2(VSQRTPS ymmreg,ymmrm256 AVX,SANDYBRIDGE )]406 code0 -[2(VSQRTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]395 code0 -[2(VSQRTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]384 code0 -[2(VSTMXCSR mem32 AVX,SANDYBRIDGE )]373 code0 -[2(VSUBPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]362 code0 -[2(VSUBPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]351 code0 -[2(VSUBPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]340 code0 -[2(VSUBPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]329 code0 -[2(VSUBSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE )]318 code0 -[2(VSUBSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE )]307 code0 -[2(VTESTPS xmmreg,xmmrm128 AVX,SANDYBRIDGE )]296 code0 -[2(VTESTPS ymmreg,ymmrm256 AVX,SANDYBRIDGE )]285 code0 -[2(VTESTPD xmmreg,xmmrm128 AVX,SANDYBRIDGE )]274 code0 -[2(VTESTPD ymmreg,ymmrm256 AVX,SANDYBRIDGE )]263 code0 -[2(VUCOMISD xmmreg,xmmrm64 AVX,SANDYBRIDGE )]252 code0 -[2(VUCOMISS xmmreg,xmmrm32 AVX,SANDYBRIDGE )]241 code0 -[2(VUNPCKHPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]230 code0 -[2(VUNPCKHPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]219 code0 -[2(VUNPCKHPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]208 code0 -[2(VUNPCKHPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]197 code0 -[2(VUNPCKLPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]186 code0 -[2(VUNPCKLPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]175 code0 -[2(VUNPCKLPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]164 code0 -[2(VUNPCKLPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]153 code0 -[2(VXORPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]142 code0 -[2(VXORPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]131 code0 -[2(VXORPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]120 code0 -[2(VXORPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE )]109 code0 -(169)pageodd -restore showpage -%%Page: 170 170 -%%BeginPageSetup -save -%%EndPageSetup -/170 pa -[2(VZEROALL AVX,SANDYBRIDGE )]681 code0 -[2(VZEROUPPER AVX,SANDYBRIDGE)]670 code2 -[{/section-B.1.26 xa}(Intel Carry-Less Multiplication instructions \(CLMUL\))](B.1.26)650.8 subh3 -[2(PCLMULLQLQDQ xmmreg,xmmrm128 SSE,WESTMERE )]633.8 code1 -[2(PCLMULHQLQDQ xmmreg,xmmrm128 SSE,WESTMERE )]622.8 code0 -[2(PCLMULLQHQDQ xmmreg,xmmrm128 SSE,WESTMERE )]611.8 code0 -[2(PCLMULHQHQDQ xmmreg,xmmrm128 SSE,WESTMERE )]600.8 code0 -[2(PCLMULQDQ xmmreg,xmmrm128,imm8 SSE,WESTMERE)]589.8 code2 -[{/section-B.1.27 xa}(Intel AVX Carry-Less Multiplication instructions \(CLMUL\))](B.1.27)570.6 subh3 -[2(VPCLMULLQLQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]553.6 code1 -[2(VPCLMULHQLQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]542.6 code0 -[2(VPCLMULLQHQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]531.6 code0 -[2(VPCLMULHQHQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE )]520.6 code0 -[2(VPCLMULQDQ xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE)]509.6 code2 -[{/section-B.1.28 xa}(Intel Fused Multiply-Add instructions \(FMA\))](B.1.28)490.4 subh3 -[2(VFMADD132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]473.4 code1 -[2(VFMADD132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]462.4 code0 -[2(VFMADD132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]451.4 code0 -[2(VFMADD132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]440.4 code0 -[2(VFMADD312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]429.4 code0 -[2(VFMADD312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]418.4 code0 -[2(VFMADD312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]407.4 code0 -[2(VFMADD312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]396.4 code0 -[2(VFMADD213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]385.4 code0 -[2(VFMADD213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]374.4 code0 -[2(VFMADD213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]363.4 code0 -[2(VFMADD213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]352.4 code0 -[2(VFMADD123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]341.4 code0 -[2(VFMADD123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]330.4 code0 -[2(VFMADD123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]319.4 code0 -[2(VFMADD123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]308.4 code0 -[2(VFMADD231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]297.4 code0 -[2(VFMADD231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]286.4 code0 -[2(VFMADD231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]275.4 code0 -[2(VFMADD231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]264.4 code0 -[2(VFMADD321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]253.4 code0 -[2(VFMADD321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]242.4 code0 -[2(VFMADD321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]231.4 code0 -[2(VFMADD321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]220.4 code0 -[2(VFMADDSUB132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]209.4 code0 -[2(VFMADDSUB132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]198.4 code0 -[2(VFMADDSUB132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]187.4 code0 -[2(VFMADDSUB132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]176.4 code0 -[2(VFMADDSUB312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]165.4 code0 -[2(VFMADDSUB312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]154.4 code0 -[2(VFMADDSUB312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]143.4 code0 -[2(VFMADDSUB312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]132.4 code0 -[2(VFMADDSUB213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]121.4 code0 -[2(VFMADDSUB213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]110.4 code0 -(170)pageeven -restore showpage -%%Page: 171 171 -%%BeginPageSetup -save -%%EndPageSetup -/171 pa -[2(VFMADDSUB213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]681 code0 -[2(VFMADDSUB213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]670 code0 -[2(VFMADDSUB123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]659 code0 -[2(VFMADDSUB123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]648 code0 -[2(VFMADDSUB123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]637 code0 -[2(VFMADDSUB123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]626 code0 -[2(VFMADDSUB231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]615 code0 -[2(VFMADDSUB231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]604 code0 -[2(VFMADDSUB231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]593 code0 -[2(VFMADDSUB231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]582 code0 -[2(VFMADDSUB321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]571 code0 -[2(VFMADDSUB321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]560 code0 -[2(VFMADDSUB321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]549 code0 -[2(VFMADDSUB321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]538 code0 -[2(VFMSUB132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]527 code0 -[2(VFMSUB132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]516 code0 -[2(VFMSUB132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]505 code0 -[2(VFMSUB132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]494 code0 -[2(VFMSUB312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]483 code0 -[2(VFMSUB312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]472 code0 -[2(VFMSUB312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]461 code0 -[2(VFMSUB312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]450 code0 -[2(VFMSUB213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]439 code0 -[2(VFMSUB213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]428 code0 -[2(VFMSUB213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]417 code0 -[2(VFMSUB213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]406 code0 -[2(VFMSUB123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]395 code0 -[2(VFMSUB123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]384 code0 -[2(VFMSUB123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]373 code0 -[2(VFMSUB123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]362 code0 -[2(VFMSUB231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]351 code0 -[2(VFMSUB231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]340 code0 -[2(VFMSUB231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]329 code0 -[2(VFMSUB231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]318 code0 -[2(VFMSUB321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]307 code0 -[2(VFMSUB321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]296 code0 -[2(VFMSUB321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]285 code0 -[2(VFMSUB321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]274 code0 -[2(VFMSUBADD132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]263 code0 -[2(VFMSUBADD132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]252 code0 -[2(VFMSUBADD132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]241 code0 -[2(VFMSUBADD132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]230 code0 -[2(VFMSUBADD312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]219 code0 -[2(VFMSUBADD312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]208 code0 -[2(VFMSUBADD312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]197 code0 -[2(VFMSUBADD312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]186 code0 -[2(VFMSUBADD213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]175 code0 -[2(VFMSUBADD213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]164 code0 -[2(VFMSUBADD213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]153 code0 -[2(VFMSUBADD213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]142 code0 -[2(VFMSUBADD123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]131 code0 -[2(VFMSUBADD123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]120 code0 -[2(VFMSUBADD123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]109 code0 -(171)pageodd -restore showpage -%%Page: 172 172 -%%BeginPageSetup -save -%%EndPageSetup -/172 pa -[2(VFMSUBADD123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]681 code0 -[2(VFMSUBADD231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]670 code0 -[2(VFMSUBADD231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]659 code0 -[2(VFMSUBADD231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]648 code0 -[2(VFMSUBADD231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]637 code0 -[2(VFMSUBADD321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]626 code0 -[2(VFMSUBADD321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]615 code0 -[2(VFMSUBADD321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]604 code0 -[2(VFMSUBADD321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]593 code0 -[2(VFNMADD132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]582 code0 -[2(VFNMADD132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]571 code0 -[2(VFNMADD132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]560 code0 -[2(VFNMADD132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]549 code0 -[2(VFNMADD312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]538 code0 -[2(VFNMADD312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]527 code0 -[2(VFNMADD312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]516 code0 -[2(VFNMADD312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]505 code0 -[2(VFNMADD213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]494 code0 -[2(VFNMADD213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]483 code0 -[2(VFNMADD213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]472 code0 -[2(VFNMADD213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]461 code0 -[2(VFNMADD123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]450 code0 -[2(VFNMADD123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]439 code0 -[2(VFNMADD123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]428 code0 -[2(VFNMADD123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]417 code0 -[2(VFNMADD231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]406 code0 -[2(VFNMADD231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]395 code0 -[2(VFNMADD231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]384 code0 -[2(VFNMADD231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]373 code0 -[2(VFNMADD321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]362 code0 -[2(VFNMADD321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]351 code0 -[2(VFNMADD321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]340 code0 -[2(VFNMADD321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]329 code0 -[2(VFNMSUB132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]318 code0 -[2(VFNMSUB132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]307 code0 -[2(VFNMSUB132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]296 code0 -[2(VFNMSUB132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]285 code0 -[2(VFNMSUB312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]274 code0 -[2(VFNMSUB312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]263 code0 -[2(VFNMSUB312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]252 code0 -[2(VFNMSUB312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]241 code0 -[2(VFNMSUB213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]230 code0 -[2(VFNMSUB213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]219 code0 -[2(VFNMSUB213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]208 code0 -[2(VFNMSUB213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]197 code0 -[2(VFNMSUB123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]186 code0 -[2(VFNMSUB123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]175 code0 -[2(VFNMSUB123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]164 code0 -[2(VFNMSUB123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]153 code0 -[2(VFNMSUB231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]142 code0 -[2(VFNMSUB231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]131 code0 -[2(VFNMSUB231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]120 code0 -[2(VFNMSUB231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]109 code0 -(172)pageeven -restore showpage -%%Page: 173 173 -%%BeginPageSetup -save -%%EndPageSetup -/173 pa -[2(VFNMSUB321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]681 code0 -[2(VFNMSUB321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]670 code0 -[2(VFNMSUB321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE )]659 code0 -[2(VFNMSUB321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE )]648 code0 -[2(VFMADD132SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]637 code0 -[2(VFMADD132SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]626 code0 -[2(VFMADD312SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]615 code0 -[2(VFMADD312SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]604 code0 -[2(VFMADD213SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]593 code0 -[2(VFMADD213SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]582 code0 -[2(VFMADD123SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]571 code0 -[2(VFMADD123SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]560 code0 -[2(VFMADD231SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]549 code0 -[2(VFMADD231SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]538 code0 -[2(VFMADD321SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]527 code0 -[2(VFMADD321SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]516 code0 -[2(VFMSUB132SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]505 code0 -[2(VFMSUB132SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]494 code0 -[2(VFMSUB312SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]483 code0 -[2(VFMSUB312SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]472 code0 -[2(VFMSUB213SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]461 code0 -[2(VFMSUB213SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]450 code0 -[2(VFMSUB123SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]439 code0 -[2(VFMSUB123SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]428 code0 -[2(VFMSUB231SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]417 code0 -[2(VFMSUB231SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]406 code0 -[2(VFMSUB321SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]395 code0 -[2(VFMSUB321SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]384 code0 -[2(VFNMADD132SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]373 code0 -[2(VFNMADD132SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]362 code0 -[2(VFNMADD312SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]351 code0 -[2(VFNMADD312SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]340 code0 -[2(VFNMADD213SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]329 code0 -[2(VFNMADD213SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]318 code0 -[2(VFNMADD123SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]307 code0 -[2(VFNMADD123SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]296 code0 -[2(VFNMADD231SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]285 code0 -[2(VFNMADD231SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]274 code0 -[2(VFNMADD321SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]263 code0 -[2(VFNMADD321SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]252 code0 -[2(VFNMSUB132SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]241 code0 -[2(VFNMSUB132SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]230 code0 -[2(VFNMSUB312SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]219 code0 -[2(VFNMSUB312SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]208 code0 -[2(VFNMSUB213SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]197 code0 -[2(VFNMSUB213SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]186 code0 -[2(VFNMSUB123SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]175 code0 -[2(VFNMSUB123SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]164 code0 -[2(VFNMSUB231SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]153 code0 -[2(VFNMSUB231SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE )]142 code0 -[2(VFNMSUB321SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE )]131 code0 -[2(VFNMSUB321SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE)]120 code2 -(173)pageodd -restore showpage -%%Page: 174 174 -%%BeginPageSetup -save -%%EndPageSetup -/174 pa -[{/section-B.1.29 xa}(VIA \(Centaur\) security instructions)](B.1.29)678.8 subh3 -[2(XSTORE PENT,CYRIX )]661.8 code1 -[2(XCRYPTECB PENT,CYRIX )]650.8 code0 -[2(XCRYPTCBC PENT,CYRIX )]639.8 code0 -[2(XCRYPTCTR PENT,CYRIX )]628.8 code0 -[2(XCRYPTCFB PENT,CYRIX )]617.8 code0 -[2(XCRYPTOFB PENT,CYRIX )]606.8 code0 -[2(MONTMUL PENT,CYRIX )]595.8 code0 -[2(XSHA1 PENT,CYRIX )]584.8 code0 -[2(XSHA256 PENT,CYRIX)]573.8 code2 -[{/section-B.1.30 xa}(AMD Lightweight Profiling \(LWP\) instructions)](B.1.30)554.6 subh3 -[2(LLWPCB reg16 AMD )]537.6 code1 -[2(LLWPCB reg32 AMD,386 )]526.6 code0 -[2(LLWPCB reg64 AMD,X64 )]515.6 code0 -[2(SLWPCB reg16 AMD )]504.6 code0 -[2(SLWPCB reg32 AMD,386 )]493.6 code0 -[2(SLWPCB reg64 AMD,X64 )]482.6 code0 -[2(LWPVAL reg16,rm32,imm16 AMD,386 )]471.6 code0 -[2(LWPVAL reg32,rm32,imm32 AMD,386 )]460.6 code0 -[2(LWPVAL reg64,rm32,imm32 AMD,X64 )]449.6 code0 -[2(LWPINS reg16,rm32,imm16 AMD,386 )]438.6 code0 -[2(LWPINS reg32,rm32,imm32 AMD,386 )]427.6 code0 -[2(LWPINS reg64,rm32,imm32 AMD,X64)]416.6 code2 -[{/section-B.1.31 xa}(AMD XOP, FMA4 and CVT16 instructions \(SSE5\))](B.1.31)397.4 subh3 -[2(VCVTPH2PS xmmreg,xmmrm64*,imm8 AMD,SSE5 )]380.4 code1 -[2(VCVTPH2PS ymmreg,xmmrm128,imm8 AMD,SSE5 )]369.4 code0 -[2(VCVTPH2PS ymmreg,ymmrm128*,imm8 AMD,SSE5 )]358.4 code0 -[2(VCVTPS2PH xmmrm64,xmmreg*,imm8 AMD,SSE5 )]347.4 code0 -[2(VCVTPS2PH xmmrm128,ymmreg,imm8 AMD,SSE5 )]336.4 code0 -[2(VCVTPS2PH ymmrm128,ymmreg*,imm8 AMD,SSE5 )]325.4 code0 -[2(VFMADDPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]314.4 code0 -[2(VFMADDPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 )]303.4 code0 -[2(VFMADDPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 )]292.4 code0 -[2(VFMADDPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 )]281.4 code0 -[2(VFMADDPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]270.4 code0 -[2(VFMADDPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 )]259.4 code0 -[2(VFMADDPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 )]248.4 code0 -[2(VFMADDPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 )]237.4 code0 -[2(VFMADDSD xmmreg,xmmreg*,xmmrm64,xmmreg AMD,SSE5 )]226.4 code0 -[2(VFMADDSD xmmreg,xmmreg*,xmmreg,xmmrm64 AMD,SSE5 )]215.4 code0 -[2(VFMADDSS xmmreg,xmmreg*,xmmrm32,xmmreg AMD,SSE5 )]204.4 code0 -[2(VFMADDSS xmmreg,xmmreg*,xmmreg,xmmrm32 AMD,SSE5 )]193.4 code0 -[2(VFMADDSUBPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]182.4 code0 -[2(VFMADDSUBPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 )]171.4 code0 -[2(VFMADDSUBPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 )]160.4 code0 -[2(VFMADDSUBPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 )]149.4 code0 -[2(VFMADDSUBPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]138.4 code0 -[2(VFMADDSUBPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 )]127.4 code0 -[2(VFMADDSUBPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 )]116.4 code0 -[2(VFMADDSUBPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 )]105.4 code0 -(174)pageeven -restore showpage -%%Page: 175 175 -%%BeginPageSetup -save -%%EndPageSetup -/175 pa -[2(VFMSUBADDPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]681 code0 -[2(VFMSUBADDPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 )]670 code0 -[2(VFMSUBADDPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 )]659 code0 -[2(VFMSUBADDPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 )]648 code0 -[2(VFMSUBADDPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]637 code0 -[2(VFMSUBADDPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 )]626 code0 -[2(VFMSUBADDPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 )]615 code0 -[2(VFMSUBADDPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 )]604 code0 -[2(VFMSUBPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]593 code0 -[2(VFMSUBPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 )]582 code0 -[2(VFMSUBPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 )]571 code0 -[2(VFMSUBPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 )]560 code0 -[2(VFMSUBPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]549 code0 -[2(VFMSUBPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 )]538 code0 -[2(VFMSUBPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 )]527 code0 -[2(VFMSUBPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 )]516 code0 -[2(VFMSUBSD xmmreg,xmmreg*,xmmrm64,xmmreg AMD,SSE5 )]505 code0 -[2(VFMSUBSD xmmreg,xmmreg*,xmmreg,xmmrm64 AMD,SSE5 )]494 code0 -[2(VFMSUBSS xmmreg,xmmreg*,xmmrm32,xmmreg AMD,SSE5 )]483 code0 -[2(VFMSUBSS xmmreg,xmmreg*,xmmreg,xmmrm32 AMD,SSE5 )]472 code0 -[2(VFNMADDPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]461 code0 -[2(VFNMADDPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 )]450 code0 -[2(VFNMADDPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 )]439 code0 -[2(VFNMADDPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 )]428 code0 -[2(VFNMADDPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]417 code0 -[2(VFNMADDPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 )]406 code0 -[2(VFNMADDPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 )]395 code0 -[2(VFNMADDPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 )]384 code0 -[2(VFNMADDSD xmmreg,xmmreg*,xmmrm64,xmmreg AMD,SSE5 )]373 code0 -[2(VFNMADDSD xmmreg,xmmreg*,xmmreg,xmmrm64 AMD,SSE5 )]362 code0 -[2(VFNMADDSS xmmreg,xmmreg*,xmmrm32,xmmreg AMD,SSE5 )]351 code0 -[2(VFNMADDSS xmmreg,xmmreg*,xmmreg,xmmrm32 AMD,SSE5 )]340 code0 -[2(VFNMSUBPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]329 code0 -[2(VFNMSUBPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 )]318 code0 -[2(VFNMSUBPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 )]307 code0 -[2(VFNMSUBPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 )]296 code0 -[2(VFNMSUBPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]285 code0 -[2(VFNMSUBPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 )]274 code0 -[2(VFNMSUBPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 )]263 code0 -[2(VFNMSUBPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 )]252 code0 -[2(VFNMSUBSD xmmreg,xmmreg*,xmmrm64,xmmreg AMD,SSE5 )]241 code0 -[2(VFNMSUBSD xmmreg,xmmreg*,xmmreg,xmmrm64 AMD,SSE5 )]230 code0 -[2(VFNMSUBSS xmmreg,xmmreg*,xmmrm32,xmmreg AMD,SSE5 )]219 code0 -[2(VFNMSUBSS xmmreg,xmmreg*,xmmreg,xmmrm32 AMD,SSE5 )]208 code0 -[2(VFRCZPD xmmreg,xmmrm128* AMD,SSE5 )]197 code0 -[2(VFRCZPD ymmreg,ymmrm256* AMD,SSE5 )]186 code0 -[2(VFRCZPS xmmreg,xmmrm128* AMD,SSE5 )]175 code0 -[2(VFRCZPS ymmreg,ymmrm256* AMD,SSE5 )]164 code0 -[2(VFRCZSD xmmreg,xmmrm64* AMD,SSE5 )]153 code0 -[2(VFRCZSS xmmreg,xmmrm32* AMD,SSE5 )]142 code0 -[2(VPCMOV xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]131 code0 -[2(VPCMOV ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 )]120 code0 -[2(VPCMOV xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 )]109 code0 -(175)pageodd -restore showpage -%%Page: 176 176 -%%BeginPageSetup -save -%%EndPageSetup -/176 pa -[2(VPCMOV ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 )]681 code0 -[2(VPCOMB xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 )]670 code0 -[2(VPCOMD xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 )]659 code0 -[2(VPCOMQ xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 )]648 code0 -[2(VPCOMUB xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 )]637 code0 -[2(VPCOMUD xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 )]626 code0 -[2(VPCOMUQ xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 )]615 code0 -[2(VPCOMUW xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 )]604 code0 -[2(VPCOMW xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 )]593 code0 -[2(VPHADDBD xmmreg,xmmrm128* AMD,SSE5 )]582 code0 -[2(VPHADDBQ xmmreg,xmmrm128* AMD,SSE5 )]571 code0 -[2(VPHADDBW xmmreg,xmmrm128* AMD,SSE5 )]560 code0 -[2(VPHADDDQ xmmreg,xmmrm128* AMD,SSE5 )]549 code0 -[2(VPHADDUBD xmmreg,xmmrm128* AMD,SSE5 )]538 code0 -[2(VPHADDUBQ xmmreg,xmmrm128* AMD,SSE5 )]527 code0 -[2(VPHADDUBW xmmreg,xmmrm128* AMD,SSE5 )]516 code0 -[2(VPHADDUDQ xmmreg,xmmrm128* AMD,SSE5 )]505 code0 -[2(VPHADDUWD xmmreg,xmmrm128* AMD,SSE5 )]494 code0 -[2(VPHADDUWQ xmmreg,xmmrm128* AMD,SSE5 )]483 code0 -[2(VPHADDWD xmmreg,xmmrm128* AMD,SSE5 )]472 code0 -[2(VPHADDWQ xmmreg,xmmrm128* AMD,SSE5 )]461 code0 -[2(VPHSUBBW xmmreg,xmmrm128* AMD,SSE5 )]450 code0 -[2(VPHSUBDQ xmmreg,xmmrm128* AMD,SSE5 )]439 code0 -[2(VPHSUBWD xmmreg,xmmrm128* AMD,SSE5 )]428 code0 -[2(VPMACSDD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]417 code0 -[2(VPMACSDQH xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]406 code0 -[2(VPMACSDQL xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]395 code0 -[2(VPMACSSDD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]384 code0 -[2(VPMACSSDQH xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]373 code0 -[2(VPMACSSDQL xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]362 code0 -[2(VPMACSSWD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]351 code0 -[2(VPMACSSWW xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]340 code0 -[2(VPMACSWD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]329 code0 -[2(VPMACSWW xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]318 code0 -[2(VPMADCSSWD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]307 code0 -[2(VPMADCSWD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]296 code0 -[2(VPPERM xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 )]285 code0 -[2(VPPERM xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 )]274 code0 -[2(VPROTB xmmreg,xmmrm128*,xmmreg AMD,SSE5 )]263 code0 -[2(VPROTB xmmreg,xmmreg*,xmmrm128 AMD,SSE5 )]252 code0 -[2(VPROTB xmmreg,xmmrm128*,imm8 AMD,SSE5 )]241 code0 -[2(VPROTD xmmreg,xmmrm128*,xmmreg AMD,SSE5 )]230 code0 -[2(VPROTD xmmreg,xmmreg*,xmmrm128 AMD,SSE5 )]219 code0 -[2(VPROTD xmmreg,xmmrm128*,imm8 AMD,SSE5 )]208 code0 -[2(VPROTQ xmmreg,xmmrm128*,xmmreg AMD,SSE5 )]197 code0 -[2(VPROTQ xmmreg,xmmreg*,xmmrm128 AMD,SSE5 )]186 code0 -[2(VPROTQ xmmreg,xmmrm128*,imm8 AMD,SSE5 )]175 code0 -[2(VPROTW xmmreg,xmmrm128*,xmmreg AMD,SSE5 )]164 code0 -[2(VPROTW xmmreg,xmmreg*,xmmrm128 AMD,SSE5 )]153 code0 -[2(VPROTW xmmreg,xmmrm128*,imm8 AMD,SSE5 )]142 code0 -[2(VPSHAB xmmreg,xmmrm128*,xmmreg AMD,SSE5 )]131 code0 -[2(VPSHAB xmmreg,xmmreg*,xmmrm128 AMD,SSE5 )]120 code0 -[2(VPSHAD xmmreg,xmmrm128*,xmmreg AMD,SSE5 )]109 code0 -(176)pageeven -restore showpage -%%Page: 177 177 -%%BeginPageSetup -save -%%EndPageSetup -/177 pa -[2(VPSHAD xmmreg,xmmreg*,xmmrm128 AMD,SSE5 )]681 code0 -[2(VPSHAQ xmmreg,xmmrm128*,xmmreg AMD,SSE5 )]670 code0 -[2(VPSHAQ xmmreg,xmmreg*,xmmrm128 AMD,SSE5 )]659 code0 -[2(VPSHAW xmmreg,xmmrm128*,xmmreg AMD,SSE5 )]648 code0 -[2(VPSHAW xmmreg,xmmreg*,xmmrm128 AMD,SSE5 )]637 code0 -[2(VPSHLB xmmreg,xmmrm128*,xmmreg AMD,SSE5 )]626 code0 -[2(VPSHLB xmmreg,xmmreg*,xmmrm128 AMD,SSE5 )]615 code0 -[2(VPSHLD xmmreg,xmmrm128*,xmmreg AMD,SSE5 )]604 code0 -[2(VPSHLD xmmreg,xmmreg*,xmmrm128 AMD,SSE5 )]593 code0 -[2(VPSHLQ xmmreg,xmmrm128*,xmmreg AMD,SSE5 )]582 code0 -[2(VPSHLQ xmmreg,xmmreg*,xmmrm128 AMD,SSE5 )]571 code0 -[2(VPSHLW xmmreg,xmmrm128*,xmmreg AMD,SSE5 )]560 code0 -[2(VPSHLW xmmreg,xmmreg*,xmmrm128 AMD,SSE5)]549 code2 -[{/section-B.1.32 xa}(Systematic names for the hinting nop instructions)](B.1.32)529.8 subh3 -[2(HINT_NOP0 rm16 P6,UNDOC )]512.8 code1 -[2(HINT_NOP0 rm32 P6,UNDOC )]501.8 code0 -[2(HINT_NOP0 rm64 X64,UNDOC )]490.8 code0 -[2(HINT_NOP1 rm16 P6,UNDOC )]479.8 code0 -[2(HINT_NOP1 rm32 P6,UNDOC )]468.8 code0 -[2(HINT_NOP1 rm64 X64,UNDOC )]457.8 code0 -[2(HINT_NOP2 rm16 P6,UNDOC )]446.8 code0 -[2(HINT_NOP2 rm32 P6,UNDOC )]435.8 code0 -[2(HINT_NOP2 rm64 X64,UNDOC )]424.8 code0 -[2(HINT_NOP3 rm16 P6,UNDOC )]413.8 code0 -[2(HINT_NOP3 rm32 P6,UNDOC )]402.8 code0 -[2(HINT_NOP3 rm64 X64,UNDOC )]391.8 code0 -[2(HINT_NOP4 rm16 P6,UNDOC )]380.8 code0 -[2(HINT_NOP4 rm32 P6,UNDOC )]369.8 code0 -[2(HINT_NOP4 rm64 X64,UNDOC )]358.8 code0 -[2(HINT_NOP5 rm16 P6,UNDOC )]347.8 code0 -[2(HINT_NOP5 rm32 P6,UNDOC )]336.8 code0 -[2(HINT_NOP5 rm64 X64,UNDOC )]325.8 code0 -[2(HINT_NOP6 rm16 P6,UNDOC )]314.8 code0 -[2(HINT_NOP6 rm32 P6,UNDOC )]303.8 code0 -[2(HINT_NOP6 rm64 X64,UNDOC )]292.8 code0 -[2(HINT_NOP7 rm16 P6,UNDOC )]281.8 code0 -[2(HINT_NOP7 rm32 P6,UNDOC )]270.8 code0 -[2(HINT_NOP7 rm64 X64,UNDOC )]259.8 code0 -[2(HINT_NOP8 rm16 P6,UNDOC )]248.8 code0 -[2(HINT_NOP8 rm32 P6,UNDOC )]237.8 code0 -[2(HINT_NOP8 rm64 X64,UNDOC )]226.8 code0 -[2(HINT_NOP9 rm16 P6,UNDOC )]215.8 code0 -[2(HINT_NOP9 rm32 P6,UNDOC )]204.8 code0 -[2(HINT_NOP9 rm64 X64,UNDOC )]193.8 code0 -[2(HINT_NOP10 rm16 P6,UNDOC )]182.8 code0 -[2(HINT_NOP10 rm32 P6,UNDOC )]171.8 code0 -[2(HINT_NOP10 rm64 X64,UNDOC )]160.8 code0 -[2(HINT_NOP11 rm16 P6,UNDOC )]149.8 code0 -[2(HINT_NOP11 rm32 P6,UNDOC )]138.8 code0 -[2(HINT_NOP11 rm64 X64,UNDOC )]127.8 code0 -[2(HINT_NOP12 rm16 P6,UNDOC )]116.8 code0 -[2(HINT_NOP12 rm32 P6,UNDOC )]105.8 code0 -(177)pageodd -restore showpage -%%Page: 178 178 -%%BeginPageSetup -save -%%EndPageSetup -/178 pa -[2(HINT_NOP12 rm64 X64,UNDOC )]681 code0 -[2(HINT_NOP13 rm16 P6,UNDOC )]670 code0 -[2(HINT_NOP13 rm32 P6,UNDOC )]659 code0 -[2(HINT_NOP13 rm64 X64,UNDOC )]648 code0 -[2(HINT_NOP14 rm16 P6,UNDOC )]637 code0 -[2(HINT_NOP14 rm32 P6,UNDOC )]626 code0 -[2(HINT_NOP14 rm64 X64,UNDOC )]615 code0 -[2(HINT_NOP15 rm16 P6,UNDOC )]604 code0 -[2(HINT_NOP15 rm32 P6,UNDOC )]593 code0 -[2(HINT_NOP15 rm64 X64,UNDOC )]582 code0 -[2(HINT_NOP16 rm16 P6,UNDOC )]571 code0 -[2(HINT_NOP16 rm32 P6,UNDOC )]560 code0 -[2(HINT_NOP16 rm64 X64,UNDOC )]549 code0 -[2(HINT_NOP17 rm16 P6,UNDOC )]538 code0 -[2(HINT_NOP17 rm32 P6,UNDOC )]527 code0 -[2(HINT_NOP17 rm64 X64,UNDOC )]516 code0 -[2(HINT_NOP18 rm16 P6,UNDOC )]505 code0 -[2(HINT_NOP18 rm32 P6,UNDOC )]494 code0 -[2(HINT_NOP18 rm64 X64,UNDOC )]483 code0 -[2(HINT_NOP19 rm16 P6,UNDOC )]472 code0 -[2(HINT_NOP19 rm32 P6,UNDOC )]461 code0 -[2(HINT_NOP19 rm64 X64,UNDOC )]450 code0 -[2(HINT_NOP20 rm16 P6,UNDOC )]439 code0 -[2(HINT_NOP20 rm32 P6,UNDOC )]428 code0 -[2(HINT_NOP20 rm64 X64,UNDOC )]417 code0 -[2(HINT_NOP21 rm16 P6,UNDOC )]406 code0 -[2(HINT_NOP21 rm32 P6,UNDOC )]395 code0 -[2(HINT_NOP21 rm64 X64,UNDOC )]384 code0 -[2(HINT_NOP22 rm16 P6,UNDOC )]373 code0 -[2(HINT_NOP22 rm32 P6,UNDOC )]362 code0 -[2(HINT_NOP22 rm64 X64,UNDOC )]351 code0 -[2(HINT_NOP23 rm16 P6,UNDOC )]340 code0 -[2(HINT_NOP23 rm32 P6,UNDOC )]329 code0 -[2(HINT_NOP23 rm64 X64,UNDOC )]318 code0 -[2(HINT_NOP24 rm16 P6,UNDOC )]307 code0 -[2(HINT_NOP24 rm32 P6,UNDOC )]296 code0 -[2(HINT_NOP24 rm64 X64,UNDOC )]285 code0 -[2(HINT_NOP25 rm16 P6,UNDOC )]274 code0 -[2(HINT_NOP25 rm32 P6,UNDOC )]263 code0 -[2(HINT_NOP25 rm64 X64,UNDOC )]252 code0 -[2(HINT_NOP26 rm16 P6,UNDOC )]241 code0 -[2(HINT_NOP26 rm32 P6,UNDOC )]230 code0 -[2(HINT_NOP26 rm64 X64,UNDOC )]219 code0 -[2(HINT_NOP27 rm16 P6,UNDOC )]208 code0 -[2(HINT_NOP27 rm32 P6,UNDOC )]197 code0 -[2(HINT_NOP27 rm64 X64,UNDOC )]186 code0 -[2(HINT_NOP28 rm16 P6,UNDOC )]175 code0 -[2(HINT_NOP28 rm32 P6,UNDOC )]164 code0 -[2(HINT_NOP28 rm64 X64,UNDOC )]153 code0 -[2(HINT_NOP29 rm16 P6,UNDOC )]142 code0 -[2(HINT_NOP29 rm32 P6,UNDOC )]131 code0 -[2(HINT_NOP29 rm64 X64,UNDOC )]120 code0 -[2(HINT_NOP30 rm16 P6,UNDOC )]109 code0 -(178)pageeven -restore showpage -%%Page: 179 179 -%%BeginPageSetup -save -%%EndPageSetup -/179 pa -[2(HINT_NOP30 rm32 P6,UNDOC )]681 code0 -[2(HINT_NOP30 rm64 X64,UNDOC )]670 code0 -[2(HINT_NOP31 rm16 P6,UNDOC )]659 code0 -[2(HINT_NOP31 rm32 P6,UNDOC )]648 code0 -[2(HINT_NOP31 rm64 X64,UNDOC )]637 code0 -[2(HINT_NOP32 rm16 P6,UNDOC )]626 code0 -[2(HINT_NOP32 rm32 P6,UNDOC )]615 code0 -[2(HINT_NOP32 rm64 X64,UNDOC )]604 code0 -[2(HINT_NOP33 rm16 P6,UNDOC )]593 code0 -[2(HINT_NOP33 rm32 P6,UNDOC )]582 code0 -[2(HINT_NOP33 rm64 X64,UNDOC )]571 code0 -[2(HINT_NOP34 rm16 P6,UNDOC )]560 code0 -[2(HINT_NOP34 rm32 P6,UNDOC )]549 code0 -[2(HINT_NOP34 rm64 X64,UNDOC )]538 code0 -[2(HINT_NOP35 rm16 P6,UNDOC )]527 code0 -[2(HINT_NOP35 rm32 P6,UNDOC )]516 code0 -[2(HINT_NOP35 rm64 X64,UNDOC )]505 code0 -[2(HINT_NOP36 rm16 P6,UNDOC )]494 code0 -[2(HINT_NOP36 rm32 P6,UNDOC )]483 code0 -[2(HINT_NOP36 rm64 X64,UNDOC )]472 code0 -[2(HINT_NOP37 rm16 P6,UNDOC )]461 code0 -[2(HINT_NOP37 rm32 P6,UNDOC )]450 code0 -[2(HINT_NOP37 rm64 X64,UNDOC )]439 code0 -[2(HINT_NOP38 rm16 P6,UNDOC )]428 code0 -[2(HINT_NOP38 rm32 P6,UNDOC )]417 code0 -[2(HINT_NOP38 rm64 X64,UNDOC )]406 code0 -[2(HINT_NOP39 rm16 P6,UNDOC )]395 code0 -[2(HINT_NOP39 rm32 P6,UNDOC )]384 code0 -[2(HINT_NOP39 rm64 X64,UNDOC )]373 code0 -[2(HINT_NOP40 rm16 P6,UNDOC )]362 code0 -[2(HINT_NOP40 rm32 P6,UNDOC )]351 code0 -[2(HINT_NOP40 rm64 X64,UNDOC )]340 code0 -[2(HINT_NOP41 rm16 P6,UNDOC )]329 code0 -[2(HINT_NOP41 rm32 P6,UNDOC )]318 code0 -[2(HINT_NOP41 rm64 X64,UNDOC )]307 code0 -[2(HINT_NOP42 rm16 P6,UNDOC )]296 code0 -[2(HINT_NOP42 rm32 P6,UNDOC )]285 code0 -[2(HINT_NOP42 rm64 X64,UNDOC )]274 code0 -[2(HINT_NOP43 rm16 P6,UNDOC )]263 code0 -[2(HINT_NOP43 rm32 P6,UNDOC )]252 code0 -[2(HINT_NOP43 rm64 X64,UNDOC )]241 code0 -[2(HINT_NOP44 rm16 P6,UNDOC )]230 code0 -[2(HINT_NOP44 rm32 P6,UNDOC )]219 code0 -[2(HINT_NOP44 rm64 X64,UNDOC )]208 code0 -[2(HINT_NOP45 rm16 P6,UNDOC )]197 code0 -[2(HINT_NOP45 rm32 P6,UNDOC )]186 code0 -[2(HINT_NOP45 rm64 X64,UNDOC )]175 code0 -[2(HINT_NOP46 rm16 P6,UNDOC )]164 code0 -[2(HINT_NOP46 rm32 P6,UNDOC )]153 code0 -[2(HINT_NOP46 rm64 X64,UNDOC )]142 code0 -[2(HINT_NOP47 rm16 P6,UNDOC )]131 code0 -[2(HINT_NOP47 rm32 P6,UNDOC )]120 code0 -[2(HINT_NOP47 rm64 X64,UNDOC )]109 code0 -(179)pageodd -restore showpage -%%Page: 180 180 -%%BeginPageSetup -save -%%EndPageSetup -/180 pa -[2(HINT_NOP48 rm16 P6,UNDOC )]681 code0 -[2(HINT_NOP48 rm32 P6,UNDOC )]670 code0 -[2(HINT_NOP48 rm64 X64,UNDOC )]659 code0 -[2(HINT_NOP49 rm16 P6,UNDOC )]648 code0 -[2(HINT_NOP49 rm32 P6,UNDOC )]637 code0 -[2(HINT_NOP49 rm64 X64,UNDOC )]626 code0 -[2(HINT_NOP50 rm16 P6,UNDOC )]615 code0 -[2(HINT_NOP50 rm32 P6,UNDOC )]604 code0 -[2(HINT_NOP50 rm64 X64,UNDOC )]593 code0 -[2(HINT_NOP51 rm16 P6,UNDOC )]582 code0 -[2(HINT_NOP51 rm32 P6,UNDOC )]571 code0 -[2(HINT_NOP51 rm64 X64,UNDOC )]560 code0 -[2(HINT_NOP52 rm16 P6,UNDOC )]549 code0 -[2(HINT_NOP52 rm32 P6,UNDOC )]538 code0 -[2(HINT_NOP52 rm64 X64,UNDOC )]527 code0 -[2(HINT_NOP53 rm16 P6,UNDOC )]516 code0 -[2(HINT_NOP53 rm32 P6,UNDOC )]505 code0 -[2(HINT_NOP53 rm64 X64,UNDOC )]494 code0 -[2(HINT_NOP54 rm16 P6,UNDOC )]483 code0 -[2(HINT_NOP54 rm32 P6,UNDOC )]472 code0 -[2(HINT_NOP54 rm64 X64,UNDOC )]461 code0 -[2(HINT_NOP55 rm16 P6,UNDOC )]450 code0 -[2(HINT_NOP55 rm32 P6,UNDOC )]439 code0 -[2(HINT_NOP55 rm64 X64,UNDOC )]428 code0 -[2(HINT_NOP56 rm16 P6,UNDOC )]417 code0 -[2(HINT_NOP56 rm32 P6,UNDOC )]406 code0 -[2(HINT_NOP56 rm64 X64,UNDOC )]395 code0 -[2(HINT_NOP57 rm16 P6,UNDOC )]384 code0 -[2(HINT_NOP57 rm32 P6,UNDOC )]373 code0 -[2(HINT_NOP57 rm64 X64,UNDOC )]362 code0 -[2(HINT_NOP58 rm16 P6,UNDOC )]351 code0 -[2(HINT_NOP58 rm32 P6,UNDOC )]340 code0 -[2(HINT_NOP58 rm64 X64,UNDOC )]329 code0 -[2(HINT_NOP59 rm16 P6,UNDOC )]318 code0 -[2(HINT_NOP59 rm32 P6,UNDOC )]307 code0 -[2(HINT_NOP59 rm64 X64,UNDOC )]296 code0 -[2(HINT_NOP60 rm16 P6,UNDOC )]285 code0 -[2(HINT_NOP60 rm32 P6,UNDOC )]274 code0 -[2(HINT_NOP60 rm64 X64,UNDOC )]263 code0 -[2(HINT_NOP61 rm16 P6,UNDOC )]252 code0 -[2(HINT_NOP61 rm32 P6,UNDOC )]241 code0 -[2(HINT_NOP61 rm64 X64,UNDOC )]230 code0 -[2(HINT_NOP62 rm16 P6,UNDOC )]219 code0 -[2(HINT_NOP62 rm32 P6,UNDOC )]208 code0 -[2(HINT_NOP62 rm64 X64,UNDOC )]197 code0 -[2(HINT_NOP63 rm16 P6,UNDOC )]186 code0 -[2(HINT_NOP63 rm32 P6,UNDOC )]175 code0 -[2(HINT_NOP63 rm64 X64,UNDOC)]164 code2 -(180)pageeven -restore showpage -%%Page: 181 181 -%%BeginPageSetup -save -%%EndPageSetup -/181 pa -[{/appendix-C xa}(Appendix C: )(NASM Version History)]642.8 appn3 -[{/section-C.1 xa}(NASM 2 Series)](C.1)597.4 head3 -[(The NASM 2 series support x86-64, and is the production version of NASM since 2007.)]580.4 norm3 -[{/section-C.1.1 xa}(Version 2.08)](C.1.1)561.2 subh3 -[(A number of enhancements/fixes in macros area.)]544.2 bull3 -[(Support for arbitrarily terminating macro expansions )2(%exitmacro)0(. See ){/section-4.3.12 xl}(section 4.3.12){el}(.)]527.2 bull3 -[(Support for recursive macro expansion )2(%rmacro/irmacro)0(. See ){/section-4.3.1 xl}(section 4.3.1){el}(.)]510.2 bull3 -[(Support for converting strings to tokens. See ){/section-4.1.9 xl}(section 4.1.9){el}(.)]493.2 bull3 -[(Fuzzy operand size logic introduced.)]476.2 bull3 -[(Fix COFF stack overrun on too long export identifiers.)]459.2 bull3 -[(Fix Macho-O alignment bug.)]442.2 bull3 -[(Fix crashes with \226fwin32 on file with many exports.)]425.2 bull3 -[(Fix stack overrun for too long [DEBUG id].)]408.2 bull3 -[(Fix incorrect sbyte usage in IMUL \(hit only if optimization flag passed\).)]391.2 bull3 -[(Append ending token for )2(.stabs)0( records in the ELF output format.)]374.2 bull3 -[(New NSIS script which uses ModernUI and MultiUser approach.)]357.2 bull3 -[(Visual Studio 2008 NASM integration \(rules file\).)]340.2 bull3 -[(Warn a user if a constant is too long \(and as result will be stripped\).)]323.2 bull3 -[(The obsoleted pre-XOP AMD SSE5 instruction set which was never actualized was removed.)]306.2 bull3 -[(Fix stack overrun on too long error file name passed from the command line.)]289.2 bull3 -[(Bind symbols to the .text section by default \(ie in case if SECTION directive was omitted\) in the ELF)]272.2 bull1 -[(output format.)]261.2 bull2 -[(Fix sync points array index wrapping.)]244.2 bull3 -[(A few fixes for FMA4 and XOP instruction templates.)]227.2 bull3 -[(Add AMD Lightweight Profiling \(LWP\) instructions.)]210.2 bull3 -[{/section-C.1.2 xa}(Version 2.07)](C.1.2)191 subh3 -[(NASM is now under the 2-clause BSD license. See ){/section-1.1.2 xl}(section 1.1.2){el}(.)]174 bull3 -[(Fix the section type for the )2(.strtab)0( section in the )2(elf64)0( output format.)]157 bull3 -[(Fix the handling of )2(COMMON)0( directives in the )2(obj)0( output format.)]140 bull3 -[(New )2(ith)0( and )2(srec)0( output formats; these are variants of the )2(bin)0( output format which output Intel hex)]123 bull1 -[(and Motorola S-records, respectively. See ){/section-7.2 xl}(section 7.2){el}( and ){/section-7.3 xl}(section 7.3){el}(.)]112 bull2 -(181)pageodd -restore showpage -%%Page: 182 182 -%%BeginPageSetup -save -%%EndPageSetup -/182 pa -[2(rdf2ihx)0( replaced with an enhanced )2(rdf2bin)0(, which can output binary, COM, Intel hex or Motorola)]681 bull1 -[(S-records.)]670 bull2 -[(The Windows installer now puts the NASM directory first in the )2(PATH)0( of the "NASM Shell".)]653 bull3 -[(Revert the early expansion behavior of )2(%+)0( to pre-2.06 behavior: )2(%+)0( is only expanded late.)]636 bull3 -[(Yet another Mach-O alignment fix.)]619 bull3 -[(Don't delete the list file on errors. Also, include error and warning information in the list file.)]602 bull3 -[(Support for 64-bit Mach-O output, see ){/section-7.8 xl}(section 7.8){el}(.)]585 bull3 -[(Fix assert failure on certain operations that involve strings with high-bit bytes.)]568 bull3 -[{/section-C.1.3 xa}(Version 2.06)](C.1.3)548.8 subh3 -[(This release is dedicated to the memory of Charles A. Crayne, long time NASM developer as well as)]531.8 bull1 -[(moderator of )2(comp.lang.asm.x86)0( and author of the book )1(Serious Assembler)0(. We miss you, Chuck.)]520.8 bull2 -[(Support for indirect macro expansion \()2(%[...])0(\). See ){/section-4.1.3 xl}(section 4.1.3){el}(.)]503.8 bull3 -[2(%pop)0( can now take an argument, see ){/section-4.7.1 xl}(section 4.7.1){el}(.)]486.8 bull3 -[(The argument to )2(%use)0( is no longer macro-expanded. Use )2(%[...])0( if macro expansion is desired.)]469.8 bull3 -[(Support for thread-local storage in ELF32 and ELF64. See ){/section-7.9.4 xl}(section 7.9.4){el}(.)]452.8 bull3 -[(Fix crash on )2(%ifmacro)0( without an argument.)]435.8 bull3 -[(Correct the arguments to the )2(POPCNT)0( instruction.)]418.8 bull3 -[(Fix section alignment in the Mach-O format.)]401.8 bull3 -[(Update AVX support to version 5 of the Intel specification.)]384.8 bull3 -[(Fix the handling of accesses to context-local macros from higher levels in the context stack.)]367.8 bull3 -[(Treat )2(WAIT)0( as a prefix rather than as an instruction, thereby allowing constructs like )2(O16 FSAVE)0( to work)]350.8 bull1 -[(correctly.)]339.8 bull2 -[(Support for structures with a non-zero base offset. See ){/section-4.11.10 xl}(section 4.11.10){el}(.)]322.8 bull3 -[(Correctly handle preprocessor token concatenation \(see ){/section-4.3.8 xl}(section 4.3.8){el}(\) involving floating-point numbers.)]305.8 bull3 -[(The )2(PINSR)0( series of instructions have been corrected and rationalized.)]288.8 bull3 -[(Removed AMD SSE5, replaced with the new XOP/FMA4/CVT16 \(rev 3.03\) spec.)]271.8 bull3 -[(The ELF backends no longer automatically generate a )2(.comment)0( section.)]254.8 bull3 -[(Add additional "well-known" ELF sections with default attributes. See ){/section-7.9.2 xl}(section 7.9.2){el}(.)]237.8 bull3 -[{/section-C.1.4 xa}(Version 2.05.01)](C.1.4)218.6 subh3 -[(Fix the )2(-w)0(/)2(-W)0( option parsing, which was broken in NASM 2.05.)]201.6 bull3 -[{/section-C.1.5 xa}(Version 2.05)](C.1.5)182.4 subh3 -[(Fix redundant REX.W prefix on )2(JMP reg64)0(.)]165.4 bull3 -[(Make the behaviour of )2(-O0)0( match NASM 0.98 legacy behavior. See ){/section-2.1.22 xl}(section 2.1.22){el}(.)]148.4 bull3 -[2(-w-user)0( can be used to suppress the output of )2(%warning)0( directives. See ){/section-2.1.24 xl}(section 2.1.24){el}(.)]131.4 bull3 -[(Fix bug where )2(ALIGN)0( would issue a full alignment datum instead of zero bytes.)]114.4 bull3 -(182)pageeven -restore showpage -%%Page: 183 183 -%%BeginPageSetup -save -%%EndPageSetup -/183 pa -[(Fix offsets in list files.)]681 bull3 -[(Fix )2(%include)0( inside multi-line macros or loops.)]664 bull3 -[(Fix error where NASM would generate a spurious warning on valid optimizations of immediate values.)]647 bull3 -[(Fix arguments to a number of the )2(CVT)0( SSE instructions.)]630 bull3 -[(Fix RIP-relative offsets when the instruction carries an immediate.)]613 bull3 -[(Massive overhaul of the ELF64 backend for spec compliance.)]596 bull3 -[(Fix the Geode )2(PFRCPV)0( and )2(PFRSQRTV)0( instruction.)]579 bull3 -[(Fix the SSE 4.2 )2(CRC32)0( instruction.)]562 bull3 -[{/section-C.1.6 xa}(Version 2.04)](C.1.6)542.8 subh3 -[(Sanitize macro handing in the )2(%error)0( directive.)]525.8 bull3 -[(New )2(%warning)0( directive to issue user-controlled warnings.)]508.8 bull3 -[2(%error)0( directives are now deferred to the final assembly phase.)]491.8 bull3 -[(New )2(%fatal)0( directive to immediately terminate assembly.)]474.8 bull3 -[(New )2(%strcat)0( directive to join quoted strings together.)]457.8 bull3 -[(New )2(%use)0( macro directive to support standard macro directives. See ){/section-4.6.4 xl}(section 4.6.4){el}(.)]440.8 bull3 -[(Excess default parameters to )2(%macro)0( now issues a warning by default. See ){/section-4.3 xl}(section 4.3){el}(.)]423.8 bull3 -[(Fix )2(%ifn)0( and )2(%elifn)0(.)]406.8 bull3 -[(Fix nested )2(%else)0( clauses.)]389.8 bull3 -[(Correct the handling of nested )2(%rep)0(s.)]372.8 bull3 -[(New )2(%unmacro)0( directive to undeclare a multi-line macro. See ){/section-4.3.11 xl}(section 4.3.11){el}(.)]355.8 bull3 -[(Builtin macro )2(__PASS__)0( which expands to the current assembly pass. See ){/section-4.11.9 xl}(section 4.11.9){el}(.)]338.8 bull3 -[2(__utf16__)0( and )2(__utf32__)0( operators to generate UTF-16 and UTF-32 strings. See ){/section-3.4.5 xl}(section 3.4.5){el}(.)]321.8 bull3 -[(Fix bug in case-insensitive matching when compiled on platforms that don't use the )2(configure)0( script.)]304.8 bull1 -[(Of the official release binaries, that only affected the OS/2 binary.)]293.8 bull2 -[(Support for x87 packed BCD constants. See ){/section-3.4.7 xl}(section 3.4.7){el}(.)]276.8 bull3 -[(Correct the )2(LTR)0( and )2(SLDT)0( instructions in 64-bit mode.)]259.8 bull3 -[(Fix unnecessary REX.W prefix on indirect jumps in 64-bit mode.)]242.8 bull3 -[(Add AVX versions of the AES instructions \()2(VAES)0(...\).)]225.8 bull3 -[(Fix the 256-bit FMA instructions.)]208.8 bull3 -[(Add 256-bit AVX stores per the latest AVX spec.)]191.8 bull3 -[(VIA XCRYPT instructions can now be written either with or without )2(REP)0(, apparently different versions of)]174.8 bull1 -[(the VIA spec wrote them differently.)]163.8 bull2 -[(Add missing 64-bit )2(MOVNTI)0( instruction.)]146.8 bull3 -[(Fix the operand size of )2(VMREAD)0( and )2(VMWRITE)0(.)]129.8 bull3 -[(Numerous bug fixes, especially to the AES, AVX and VTX instructions.)]112.8 bull3 -(183)pageodd -restore showpage -%%Page: 184 184 -%%BeginPageSetup -save -%%EndPageSetup -/184 pa -[(The optimizer now always runs until it converges. It also runs even when disabled, but doesn't optimize.)]681 bull1 -[(This allows most forward references to be resolved properly.)]670 bull2 -[2(%push)0( no longer needs a context identifier; omitting the context identifier results in an anonymous context.)]653 bull3 -[{/section-C.1.7 xa}(Version 2.03.01)](C.1.7)633.8 subh3 -[(Fix buffer overflow in the listing module.)]616.8 bull3 -[(Fix the handling of hexadecimal escape codes in `...` strings.)]599.8 bull3 -[(The Postscript/PDF documentation has been reformatted.)]582.8 bull3 -[(The )2(-F)0( option now implies )2(-g)0(.)]565.8 bull3 -[{/section-C.1.8 xa}(Version 2.03)](C.1.8)546.6 subh3 -[(Add support for Intel AVX, CLMUL and FMA instructions, including YMM registers.)]529.6 bull3 -[2(dy)0(, )2(resy)0( and )2(yword)0( for 32-byte operands.)]512.6 bull3 -[(Fix some SSE5 instructions.)]495.6 bull3 -[(Intel )2(INVEPT)0(, )2(INVVPID)0( and )2(MOVBE)0( instructions.)]478.6 bull3 -[(Fix checking for critical expressions when the optimizer is enabled.)]461.6 bull3 -[(Support the DWARF debugging format for ELF targets.)]444.6 bull3 -[(Fix optimizations of signed bytes.)]427.6 bull3 -[(Fix operation on bigendian machines.)]410.6 bull3 -[(Fix buffer overflow in the preprocessor.)]393.6 bull3 -[2(SAFESEH)0( support for Win32, )2(IMAGEREL)0( for Win64 \(SEH\).)]376.6 bull3 -[2(%?)0( and )2(%??)0( to refer to the name of a macro itself. In particular, )2(%idefine keyword $%?)0( can be used)]359.6 bull1 -[(to make a keyword "disappear".)]348.6 bull2 -[(New options for dependency generation: )2(-MD)0(, )2(-MF)0(, )2(-MP)0(, )2(-MT)0(, )2(-MQ)0(.)]331.6 bull3 -[(New preprocessor directives )2(%pathsearch)0( and )2(%depend)0(; INCBIN reimplemented as a macro.)]314.6 bull3 -[2(%include)0( now resolves macros in a sane manner.)]297.6 bull3 -[2(%substr)0( can now be used to get other than one-character substrings.)]280.6 bull3 -[(New type of character/string constants, using backquotes \()2(`...`)0(\), which support C-style escape)]263.6 bull1 -[(sequences.)]252.6 bull2 -[2(%defstr)0( and )2(%idefstr)0( to stringize macro definitions before creation.)]235.6 bull3 -[(Fix forward references used in )2(EQU)0( statements.)]218.6 bull3 -[{/section-C.1.9 xa}(Version 2.02)](C.1.9)199.4 subh3 -[(Additional fixes for MMX operands with explicit )2(qword)0(, as well as \(hopefully\) SSE operands with)]182.4 bull1 -[2(oword)0(.)]171.4 bull2 -[(Fix handling of truncated strings with )2(DO)0(.)]154.4 bull3 -[(Fix segfaults due to memory overwrites when floating-point constants were used.)]137.4 bull3 -[(Fix segfaults due to missing include files.)]120.4 bull3 -[(Fix OpenWatcom Makefiles for DOS and OS/2.)]103.4 bull3 -(184)pageeven -restore showpage -%%Page: 185 185 -%%BeginPageSetup -save -%%EndPageSetup -/185 pa -[(Add autogenerated instruction list back into the documentation.)]681 bull3 -[(ELF: Fix segfault when generating stabs, and no symbols have been defined.)]664 bull3 -[(ELF: Experimental support for DWARF debugging information.)]647 bull3 -[(New compile date and time standard macros.)]630 bull3 -[2(%ifnum)0( now returns true for negative numbers.)]613 bull3 -[(New )2(%iftoken)0( test for a single token.)]596 bull3 -[(New )2(%ifempty)0( test for empty expansion.)]579 bull3 -[(Add support for the )2(XSAVE)0( instruction group.)]562 bull3 -[(Makefile for Netware/gcc.)]545 bull3 -[(Fix issue with some warnings getting emitted way too many times.)]528 bull3 -[(Autogenerated instruction list added to the documentation.)]511 bull3 -[{/section-C.1.10 xa}(Version 2.01)](C.1.10)491.8 subh3 -[(Fix the handling of MMX registers with explicit )2(qword)0( tags on memory \(broken in 2.00 due to 64-bit)]474.8 bull1 -[(changes.\))]463.8 bull2 -[(Fix the PREFETCH instructions.)]446.8 bull3 -[(Fix the documentation.)]429.8 bull3 -[(Fix debugging info when using )2(-f elf)0( \(backwards compatibility alias for )2(-f elf32)0(\).)]412.8 bull3 -[(Man pages for rdoff tools \(from the Debian project.\))]395.8 bull3 -[(ELF: handle large numbers of sections.)]378.8 bull3 -[(Fix corrupt output when the optimizer runs out of passes.)]361.8 bull3 -[{/section-C.1.11 xa}(Version 2.00)](C.1.11)342.6 subh3 -[(Added c99 data-type compliance.)]325.6 bull3 -[(Added general x86-64 support.)]308.6 bull3 -[(Added win64 \(x86-64 COFF\) output format.)]291.6 bull3 -[(Added )2(__BITS__)0( standard macro.)]274.6 bull3 -[(Renamed the )2(elf)0( output format to )2(elf32)0( for clarity.)]257.6 bull3 -[(Added )2(elf64)0( and )2(macho)0( \(MacOS X\) output formats.)]240.6 bull3 -[(Added Numeric constants in )2(dq)0( directive.)]223.6 bull3 -[(Added )2(oword)0(, )2(do)0( and )2(reso)0( pseudo operands.)]206.6 bull3 -[(Allow underscores in numbers.)]189.6 bull3 -[(Added 8-, 16- and 128-bit floating-point formats.)]172.6 bull3 -[(Added binary, octal and hexadecimal floating-point.)]155.6 bull3 -[(Correct the generation of floating-point constants.)]138.6 bull3 -[(Added floating-point option control.)]121.6 bull3 -[(Added Infinity and NaN floating point support.)]104.6 bull3 -(185)pageodd -restore showpage -%%Page: 186 186 -%%BeginPageSetup -save -%%EndPageSetup -/186 pa -[(Added ELF Symbol Visibility support.)]681 bull3 -[(Added setting OSABI value in ELF header directive.)]664 bull3 -[(Added Generate Makefile Dependencies option.)]647 bull3 -[(Added Unlimited Optimization Passes option.)]630 bull3 -[(Added )2(%IFN)0( and )2(%ELIFN)0( support.)]613 bull3 -[(Added Logical Negation Operator.)]596 bull3 -[(Enhanced Stack Relative Preprocessor Directives.)]579 bull3 -[(Enhanced ELF Debug Formats.)]562 bull3 -[(Enhanced Send Errors to a File option.)]545 bull3 -[(Added SSSE3, SSE4.1, SSE4.2, SSE5 support.)]528 bull3 -[(Added a large number of additional instructions.)]511 bull3 -[(Significant performance improvements.)]494 bull3 -[2(-w+warning)0( and )2(-w-warning)0( can now be written as \226Wwarning and \226Wno-warning, respectively.)]477 bull1 -[(See ){/section-2.1.24 xl}(section 2.1.24){el}(.)]466 bull2 -[(Add )2(-w+error)0( to treat warnings as errors. See ){/section-2.1.24 xl}(section 2.1.24){el}(.)]449 bull3 -[(Add )2(-w+all)0( and )2(-w-all)0( to enable or disable all suppressible warnings. See ){/section-2.1.24 xl}(section 2.1.24){el}(.)]432 bull3 -[{/section-C.2 xa}(NASM 0.98 Series)](C.2)410.6 head3 -[(The 0.98 series was the production versions of NASM from 1999 to 2007.)]393.6 norm3 -[{/section-C.2.1 xa}(Version 0.98.39)](C.2.1)374.4 subh3 -[(fix buffer overflow)]357.4 bull3 -[(fix outas86's )2(.bss)0( handling)]340.4 bull3 -[("make spotless" no longer deletes config.h.in.)]323.4 bull3 -[2(%\(el\)if\(n\)idn)0( insensitivity to string quotes difference \(#809300\).)]306.4 bull3 -[(\(nasm.c\))2(__OUTPUT_FORMAT__)0( changed to string value instead of symbol.)]289.4 bull3 -[{/section-C.2.2 xa}(Version 0.98.38)](C.2.2)270.2 subh3 -[(Add Makefile for 16-bit DOS binaries under OpenWatcom, and modify )2(mkdep.pl)0( to be able to generate)]253.2 bull1 -[(completely pathless dependencies, as required by OpenWatcom wmake \(it supports path searches, but not)]242.2 bull0 -[(explicit paths.\))]231.2 bull2 -[(Fix the )2(STR)0( instruction.)]214.2 bull3 -[(Fix the ELF output format, which was broken under certain circumstances due to the addition of stabs)]197.2 bull1 -[(support.)]186.2 bull2 -[(Quick-fix Borland format debug-info for )2(-f obj)]169.2 bull3 -[(Fix for )2(%rep)0( with no arguments \(#560568\))]152.2 bull3 -[(Fix concatenation of preprocessor function call \(#794686\))]135.2 bull3 -[(Fix long label causes coredump \(#677841\))]118.2 bull3 -[(Use autoheader as well as autoconf to keep configure from generating ridiculously long command lines.)]101.2 bull3 -(186)pageeven -restore showpage -%%Page: 187 187 -%%BeginPageSetup -save -%%EndPageSetup -/187 pa -[(Make sure that all of the formats which support debugging output actually will suppress debugging output)]681 bull1 -[(when )2(-g)0( not specified.)]670 bull2 -[{/section-C.2.3 xa}(Version 0.98.37)](C.2.3)650.8 subh3 -[(Paths given in )2(-I)0( switch searched for )2(incbin)0(\226ed as well as )2(%include)0(\226ed files.)]633.8 bull3 -[(Added stabs debugging for the ELF output format, patch from Martin Wawro.)]616.8 bull3 -[(Fix )2(output/outbin.c)0( to allow origin > 80000000h.)]599.8 bull3 -[(Make )2(-U)0( switch work.)]582.8 bull3 -[(Fix the use of relative offsets with explicit prefixes, e.g. )2(a32 loop foo)0(.)]565.8 bull3 -[(Remove )2(backslash\(\))0(.)]548.8 bull3 -[(Fix the )2(SMSW)0( and )2(SLDT)0( instructions.)]531.8 bull3 -[2(-O2)0( and )2(-O3)0( are no longer aliases for )2(-O10)0( and )2(-O15)0(. If you mean the latter, please say so! :\))]514.8 bull3 -[{/section-C.2.4 xa}(Version 0.98.36)](C.2.4)495.6 subh3 -[(Update rdoff \226 librarian/archiver \226 common rec \226 docs!)]478.6 bull3 -[(Fix signed/unsigned problems.)]461.6 bull3 -[(Fix )2(JMP FAR label)0( and )2(CALL FAR label)0(.)]444.6 bull3 -[(Add new multisection support \226 map files \226 fix align bug)]427.6 bull3 -[(Fix sysexit, movhps/movlps reg,reg bugs in insns.dat)]410.6 bull3 -[2(Q)0( or )2(O)0( suffixes indicate octal)]393.6 bull3 -[(Support Prescott new instructions \(PNI\).)]376.6 bull3 -[(Cyrix )2(XSTORE)0( instruction.)]359.6 bull3 -[{/section-C.2.5 xa}(Version 0.98.35)](C.2.5)340.4 subh3 -[(Fix build failure on 16-bit DOS \(Makefile.bc3 workaround for compiler bug.\))]323.4 bull3 -[(Fix dependencies and compiler warnings.)]306.4 bull3 -[(Add "const" in a number of places.)]289.4 bull3 -[(Add \226X option to specify error reporting format \(use \226Xvc to integrate with Microsoft Visual Studio.\))]272.4 bull3 -[(Minor changes for code legibility.)]255.4 bull3 -[(Drop use of tmpnam\(\) in rdoff \(security fix.\))]238.4 bull3 -[{/section-C.2.6 xa}(Version 0.98.34)](C.2.6)219.2 subh3 -[(Correct additional address-size vs. operand-size confusions.)]202.2 bull3 -[(Generate dependencies for all Makefiles automatically.)]185.2 bull3 -[(Add support for unimplemented \(but theoretically available\) registers such as tr0 and cr5. Segment)]168.2 bull1 -[(registers 6 and 7 are called segr6 and segr7 for the operations which they can be represented.)]157.2 bull2 -[(Correct some disassembler bugs related to redundant address-size prefixes. Some work still remains in this)]140.2 bull1 -[(area.)]129.2 bull2 -[(Correctly generate an error for things like "SEG eax".)]112.2 bull3 -(187)pageodd -restore showpage -%%Page: 188 188 -%%BeginPageSetup -save -%%EndPageSetup -/188 pa -[(Add the JMPE instruction, enabled by "CPU IA64".)]681 bull3 -[(Correct compilation on newer gcc/glibc platforms.)]664 bull3 -[(Issue an error on things like "jmp far eax".)]647 bull3 -[{/section-C.2.7 xa}(Version 0.98.33)](C.2.7)627.8 subh3 -[(New __NASM_PATCHLEVEL__ and __NASM_VERSION_ID__ standard macros to round out the)]610.8 bull1 -[(version-query macros. version.pl now understands X.YYplWW or X.YY.ZZplWW as a version number,)]599.8 bull0 -[(equivalent to X.YY.ZZ.WW \(or X.YY.0.WW, as appropriate\).)]588.8 bull2 -[(New keyword "strict" to disable the optimization of specific operands.)]571.8 bull3 -[(Fix the handing of size overrides with JMP instructions \(instructions such as "jmp dword foo".\))]554.8 bull3 -[(Fix the handling of "ABSOLUTE label", where "label" points into a relocatable segment.)]537.8 bull3 -[(Fix OBJ output format with lots of externs.)]520.8 bull3 -[(More documentation updates.)]503.8 bull3 -[(Add \226Ov option to get verbose information about optimizations.)]486.8 bull3 -[(Undo a braindead change which broke )2(%elif)0( directives.)]469.8 bull3 -[(Makefile updates.)]452.8 bull3 -[{/section-C.2.8 xa}(Version 0.98.32)](C.2.8)433.6 subh3 -[(Fix NASM crashing when )2(%macro)0( directives were left unterminated.)]416.6 bull3 -[(Lots of documentation updates.)]399.6 bull3 -[(Complete rewrite of the PostScript/PDF documentation generator.)]382.6 bull3 -[(The MS Visual C++ Makefile was updated and corrected.)]365.6 bull3 -[(Recognize .rodata as a standard section name in ELF.)]348.6 bull3 -[(Fix some obsolete Perl4-isms in Perl scripts.)]331.6 bull3 -[(Fix configure.in to work with autoconf 2.5x.)]314.6 bull3 -[(Fix a couple of "make cleaner" misses.)]297.6 bull3 -[(Make the normal "./configure && make" work with Cygwin.)]280.6 bull3 -[{/section-C.2.9 xa}(Version 0.98.31)](C.2.9)261.4 subh3 -[(Correctly build in a separate object directory again.)]244.4 bull3 -[(Derive all references to the version number from the version file.)]227.4 bull3 -[(New standard macros __NASM_SUBMINOR__ and __NASM_VER__ macros.)]210.4 bull3 -[(Lots of Makefile updates and bug fixes.)]193.4 bull3 -[(New )2(%ifmacro)0( directive to test for multiline macros.)]176.4 bull3 -[(Documentation updates.)]159.4 bull3 -[(Fixes for 16-bit OBJ format output.)]142.4 bull3 -[(Changed the NASM environment variable to NASMENV.)]125.4 bull3 -(188)pageeven -restore showpage -%%Page: 189 189 -%%BeginPageSetup -save -%%EndPageSetup -/189 pa -[{/section-C.2.10 xa}(Version 0.98.30)](C.2.10)678.8 subh3 -[(Changed doc files a lot: completely removed old READMExx and Wishlist files, incorporating all)]661.8 bull1 -[(information in CHANGES and TODO.)]650.8 bull2 -[(I waited a long time to rename zoutieee.c to \(original\) outieee.c)]633.8 bull3 -[(moved all output modules to output/ subdirectory.)]616.8 bull3 -[(Added 'make strip' target to strip debug info from nasm & ndisasm.)]599.8 bull3 -[(Added INSTALL file with installation instructions.)]582.8 bull3 -[(Added \226v option description to nasm man.)]565.8 bull3 -[(Added dist makefile target to produce source distributions.)]548.8 bull3 -[(16-bit support for ELF output format \(GNU extension, but useful.\))]531.8 bull3 -[{/section-C.2.11 xa}(Version 0.98.28)](C.2.11)512.6 subh3 -[(Fastcooked this for Debian's Woody release: Frank applied the INCBIN bug patch to 0.98.25alt and called)]495.6 bull1 -[(it 0.98.28 to not confuse poor little apt-get.)]484.6 bull2 -[{/section-C.2.12 xa}(Version 0.98.26)](C.2.12)465.4 subh3 -[(Reorganised files even better from 0.98.25alt)]448.4 bull3 -[{/section-C.2.13 xa}(Version 0.98.25alt)](C.2.13)429.2 subh3 -[(Prettified the source tree. Moved files to more reasonable places.)]412.2 bull3 -[(Added findleak.pl script to misc/ directory.)]395.2 bull3 -[(Attempted to fix doc.)]378.2 bull3 -[{/section-C.2.14 xa}(Version 0.98.25)](C.2.14)359 subh3 -[(Line continuation character )2(\\)0(.)]342 bull3 -[(Docs inadvertantly reverted \226 "dos packaging".)]325 bull3 -[{/section-C.2.15 xa}(Version 0.98.24p1)](C.2.15)305.8 subh3 -[(FIXME: Someone, document this please.)]288.8 bull3 -[{/section-C.2.16 xa}(Version 0.98.24)](C.2.16)269.6 subh3 -[(Documentation \226 Ndisasm doc added to Nasm.doc.)]252.6 bull3 -[{/section-C.2.17 xa}(Version 0.98.23)](C.2.17)233.4 subh3 -[(Attempted to remove rdoff version1)]216.4 bull3 -[(Lino Mastrodomenico's patches to preproc.c \(%$$ bug?\).)]199.4 bull3 -[{/section-C.2.18 xa}(Version 0.98.22)](C.2.18)180.2 subh3 -[(Update rdoff2 \226 attempt to remove v1.)]163.2 bull3 -[{/section-C.2.19 xa}(Version 0.98.21)](C.2.19)144 subh3 -[(Optimization fixes.)]127 bull3 -(189)pageodd -restore showpage -%%Page: 190 190 -%%BeginPageSetup -save -%%EndPageSetup -/190 pa -[{/section-C.2.20 xa}(Version 0.98.20)](C.2.20)678.8 subh3 -[(Optimization fixes.)]661.8 bull3 -[{/section-C.2.21 xa}(Version 0.98.19)](C.2.21)642.6 subh3 -[(H. J. Lu's patch back out.)]625.6 bull3 -[{/section-C.2.22 xa}(Version 0.98.18)](C.2.22)606.4 subh3 -[(Added ".rdata" to "-f win32".)]589.4 bull3 -[{/section-C.2.23 xa}(Version 0.98.17)](C.2.23)570.2 subh3 -[(H. J. Lu's "bogus elf" patch. \(Red Hat problem?\))]553.2 bull3 -[{/section-C.2.24 xa}(Version 0.98.16)](C.2.24)534 subh3 -[(Fix whitespace before "[section ..." bug.)]517 bull3 -[{/section-C.2.25 xa}(Version 0.98.15)](C.2.25)497.8 subh3 -[(Rdoff changes \(?\).)]480.8 bull3 -[(Fix fixes to memory leaks.)]463.8 bull3 -[{/section-C.2.26 xa}(Version 0.98.14)](C.2.26)444.6 subh3 -[(Fix memory leaks.)]427.6 bull3 -[{/section-C.2.27 xa}(Version 0.98.13)](C.2.27)408.4 subh3 -[(There was no 0.98.13)]391.4 bull3 -[{/section-C.2.28 xa}(Version 0.98.12)](C.2.28)372.2 subh3 -[(Update optimization \(new function of "-O1"\))]355.2 bull3 -[(Changes to test/bintest.asm \(?\).)]338.2 bull3 -[{/section-C.2.29 xa}(Version 0.98.11)](C.2.29)319 subh3 -[(Optimization changes.)]302 bull3 -[(Ndisasm fixed.)]285 bull3 -[{/section-C.2.30 xa}(Version 0.98.10)](C.2.30)265.8 subh3 -[(There was no 0.98.10)]248.8 bull3 -[{/section-C.2.31 xa}(Version 0.98.09)](C.2.31)229.6 subh3 -[(Add multiple sections support to "-f bin".)]212.6 bull3 -[(Changed GLOBAL_TEMP_BASE in outelf.c from 6 to 15.)]195.6 bull3 -[(Add "-v" as an alias to the "-r" switch.)]178.6 bull3 -[(Remove "#ifdef" from Tasm compatibility options.)]161.6 bull3 -[(Remove redundant size-overrides on "mov ds, ex", etc.)]144.6 bull3 -[(Fixes to SSE2, other insns.dat \(?\).)]127.6 bull3 -[(Enable uppercase "I" and "P" switches.)]110.6 bull3 -(190)pageeven -restore showpage -%%Page: 191 191 -%%BeginPageSetup -save -%%EndPageSetup -/191 pa -[(Case insinsitive "seg" and "wrt".)]681 bull3 -[(Update install.sh \(?\).)]664 bull3 -[(Allocate tokens in blocks.)]647 bull3 -[(Improve "invalid effective address" messages.)]630 bull3 -[{/section-C.2.32 xa}(Version 0.98.08)](C.2.32)610.8 subh3 -[(Add ")2(%strlen)0(" and ")2(%substr)0(" macro operators)]593.8 bull3 -[(Fixed broken c16.mac.)]576.8 bull3 -[(Unterminated string error reported.)]559.8 bull3 -[(Fixed bugs as per 0.98bf)]542.8 bull3 -[{/section-C.2.33 xa}(Version 0.98.09b with John Coffman patches released 28-Oct-2001)](C.2.33)523.6 subh3 -[(Changes from 0.98.07 release to 98.09b as of 28-Oct-2001)]506.6 norm3 -[(More closely compatible with 0.98 when \226O0 is implied or specified. Not strictly identical, since backward)]489.6 bull1 -[(branches in range of short offsets are recognized, and signed byte values with no explicit size specification)]478.6 bull0 -[(will be assembled as a single byte.)]467.6 bull2 -[(More forgiving with the PUSH instruction. 0.98 requires a size to be specified always. 0.98.09b will imply)]450.6 bull1 -[(the size from the current BITS setting \(16 or 32\).)]439.6 bull2 -[(Changed definition of the optimization flag:)]422.6 bull3 -[(\226O0 strict two-pass assembly, JMP and Jcc are handled more like 0.98, except that back- ward JMPs are)]405.6 norm1 -[(short, if possible.)]394.6 norm2 -[(\226O1 strict two-pass assembly, but forward branches are assembled with code guaranteed to reach; may)]377.6 norm1 -[(produce larger code than \226O0, but will produce successful assembly more often if branch offset sizes are not)]366.6 norm0 -[(specified.)]355.6 norm2 -[(\226O2 multi-pass optimization, minimize branch offsets; also will minimize signed immed- iate bytes,)]338.6 norm1 -[(overriding size specification.)]327.6 norm2 -[(\226O3 like \226O2, but more passes taken, if needed)]310.6 norm3 -[{/section-C.2.34 xa}(Version 0.98.07 released 01/28/01)](C.2.34)291.4 subh3 -[(Added Stepane Denis' SSE2 instructions to a *working* version of the code \226 some earlier versions were)]274.4 bull1 -[(based on broken code \226 sorry 'bout that. version "0.98.07")]263.4 bull2 -[(01/28/01)]246.4 norm3 -[(Cosmetic modifications to nasm.c, nasm.h, AUTHORS, MODIFIED)]229.4 bull3 -[{/section-C.2.35 xa}(Version 0.98.06f released 01/18/01)](C.2.35)210.2 subh3 -[(\226 Add "metalbrain"s jecxz bug fix in insns.dat \226 alter nasmdoc.src to match \226 version "0.98.06f")]193.2 bull3 -[{/section-C.2.36 xa}(Version 0.98.06e released 01/09/01)](C.2.36)174 subh3 -[(Removed the "outforms.h" file \226 it appears to be someone's old backup of "outform.h". version "0.98.06e")]157 bull3 -[(01/09/01)]140 norm3 -[(fbk \226 finally added the fix for the "multiple %includes bug", known since 7/27/99 \226 reported originally \(?\))]123 bull1 -[(and sent to us by Austin Lunnen \226 he reports that John Fine had a fix within the day. Here it is...)]112 bull2 -(191)pageodd -restore showpage -%%Page: 192 192 -%%BeginPageSetup -save -%%EndPageSetup -/192 pa -[(Nelson Rush resigns from the group. Big thanks to Nelson for his leadership and enthusiasm in getting)]681 bull1 -[(these changes incorporated into Nasm!)]670 bull2 -[(fbk \226 [list +], [list \226] directives \226 ineptly implemented, should be re-written or removed, perhaps.)]653 bull3 -[(Brian Raiter / fbk \226 "elfso bug" fix \226 applied to aoutb format as well \226 testing might be desirable...)]636 bull3 -[(08/07/00)]619 norm3 -[(James Seter \226 \226postfix, \226prefix command line switches.)]602 bull3 -[(Yuri Zaporogets \226 rdoff utility changes.)]585 bull3 -[{/section-C.2.37 xa}(Version 0.98p1)](C.2.37)565.8 subh3 -[(GAS-like palign \(Panos Minos\))]548.8 bull3 -[(FIXME: Someone, fill this in with details)]531.8 bull3 -[{/section-C.2.38 xa}(Version 0.98bf \(bug-fixed\))](C.2.38)512.6 subh3 -[(Fixed \226 elf and aoutb bug \226 shared libraries \226 multiple "%include" bug in "-f obj" \226 jcxz, jecxz bug \226)]495.6 bull1 -[(unrecognized option bug in ndisasm)]484.6 bull2 -[{/section-C.2.39 xa}(Version 0.98.03 with John Coffman's changes released 27-Jul-2000)](C.2.39)465.4 subh3 -[(Added signed byte optimizations for the 0x81/0x83 class of instructions: ADC, ADD, AND, CMP, OR,)]448.4 bull1 -[(SBB, SUB, XOR: when used as 'ADD reg16,imm' or 'ADD reg32,imm.' Also optimization of signed byte)]437.4 bull0 -[(form of 'PUSH imm' and 'IMUL reg,imm'/'IMUL reg,reg,imm.' No size specification is needed.)]426.4 bull2 -[(Added multi-pass JMP and Jcc offset optimization. Offsets on forward references will preferentially use)]409.4 bull1 -[(the short form, without the need to code a specific size \(short or near\) for the branch. Added instructions)]398.4 bull0 -[(for 'Jcc label' to use the form 'Jnotcc $+3/JMP label', in cases where a short offset is out of bounds. If)]387.4 bull0 -[(compiling for a 386 or higher CPU, then the 386 form of Jcc will be used instead.)]376.4 bull2 -[(This feature is controlled by a new command-line switch: "O", \(upper case letter O\). "-O0" reverts the)]359.4 norm1 -[(assembler to no extra optimization passes, "-O1" allows up to 5 extra passes, and "-O2"\(default\), allows up)]348.4 norm0 -[(to 10 extra optimization passes.)]337.4 norm2 -[(Added a new directive: 'cpu XXX', where XXX is any of: 8086, 186, 286, 386, 486, 586, pentium, 686,)]320.4 bull1 -[(PPro, P2, P3 or Katmai. All are case insensitive. All instructions will be selected only if they apply to the)]309.4 bull0 -[(selected cpu or lower. Corrected a couple of bugs in cpu-dependence in 'insns.dat'.)]298.4 bull2 -[(Added to 'standard.mac', the "use16" and "use32" forms of the "bits 16/32" directive. This is nothing new,)]281.4 bull1 -[(just conforms to a lot of other assemblers. \(minor\))]270.4 bull2 -[(Changed label allocation from 320/32 \(10000 labels @ 200K+\) to 32/37 \(1000 labels\); makes running)]253.4 bull1 -[(under DOS much easier. Since additional label space is allocated dynamically, this should have no effect)]242.4 bull0 -[(on large programs with lots of labels. The 37 is a prime, believed to be better for hashing. \(minor\))]231.4 bull2 -[{/section-C.2.40 xa}(Version 0.98.03)](C.2.40)212.2 subh3 -[("Integrated patchfile 0.98-0.98.01. I call this version 0.98.03 for historical reasons: 0.98.02 was trashed.")]195.2 norm1 -[(\226\226John Coffman <johninsd@san.rr.com>, 27-Jul-2000)]184.2 norm2 -[(Kendall Bennett's SciTech MGL changes)]167.2 bull3 -[(Note that you must define "TASM_COMPAT" at compile-time to get the Tasm Ideal Mode compatibility.)]150.2 bull3 -[(All changes can be compiled in and out using the TASM_COMPAT macros, and when compiled without)]133.2 bull1 -[(TASM_COMPAT defined we get the exact same binary as the unmodified 0.98 sources.)]122.2 bull2 -[(standard.mac, macros.c: Added macros to ignore TASM directives before first include)]105.2 bull3 -(192)pageeven -restore showpage -%%Page: 193 193 -%%BeginPageSetup -save -%%EndPageSetup -/193 pa -[(nasm.h: Added extern declaration for tasm_compatible_mode)]681 bull3 -[(nasm.c: Added global variable tasm_compatible_mode)]664 bull3 -[(Added command line switch for TASM compatible mode \(-t\))]647 bull3 -[(Changed version command line to reflect when compiled with TASM additions)]630 bull3 -[(Added response file processing to allow all arguments on a single line \(response file is @resp rather than)]613 bull1 -[(\226@resp for NASM format\).)]602 bull2 -[(labels.c: Changes islocal\(\) macro to support TASM style @@local labels.)]585 bull3 -[(Added islocalchar\(\) macro to support TASM style @@local labels.)]568 bull3 -[(parser.c: Added support for TASM style memory references \(ie: mov [DWORD eax],10 rather than the)]551 bull1 -[(NASM style mov DWORD [eax],10\).)]540 bull2 -[(preproc.c: Added new directives, )2(%arg)0(, )2(%local)0(, )2(%stacksize)0( to directives table)]523 bull3 -[(Added support for TASM style directives without a leading % symbol.)]506 bull3 -[(Integrated a block of changes from Andrew Zabolotny <bit@eltech.ru>:)]489 bull3 -[(A new keyword )2(%xdefine)0( and its case-insensitive counterpart )2(%ixdefine)0(. They work almost the)]472 bull1 -[(same way as )2(%define)0( and )2(%idefine)0( but expand the definition immediately, not on the invocation.)]461 bull0 -[(Something like a cross between )2(%define)0( and )2(%assign)0(. The "x" suffix stands for "eXpand", so)]450 bull0 -[("xdefine" can be deciphered as "expand-and-define". Thus you can do things like this:)]439 bull2 -[2( %assign ofs 0 )]422 code1 -[2()]411 code0 -[2( %macro arg 1 )]400 code0 -[2( %xdefine %1 dword [esp+ofs] )]389 code0 -[2( %assign ofs ofs+4 )]378 code0 -[2( %endmacro)]367 code2 -[(Changed the place where the expansion of %$name macros are expanded. Now they are converted into)]350 bull1 -[(..@ctxnum.name form when detokenizing, so there are no quirks as before when using %$name arguments)]339 bull0 -[(to macros, in macros etc. For example:)]328 bull2 -[2( %macro abc 1 )]311 code1 -[2( %define %1 hello )]300 code0 -[2( %endm )]289 code0 -[2()]278 code0 -[2( abc %$here )]267 code0 -[2( %$here)]256 code2 -[(Now last line will be expanded into "hello" as expected. This also allows for lots of goodies, a good example)]239 norm1 -[(are extended "proc" macros included in this archive.)]228 norm2 -[(Added a check for "cstk" in smacro_defined\(\) before calling get_ctx\(\) \226 this allows for things like:)]211 bull3 -[2( %ifdef %$abc )]194 code1 -[2( %endif)]183 code2 -[(to work without warnings even in no context.)]166 norm3 -[(Added a check for "cstk" in %if*ctx and %elif*ctx directives \226 this allows to use )2(%ifctx)0( without)]149 bull1 -[(excessive warnings. If there is no active context, )2(%ifctx)0( goes through "false" branch.)]138 bull2 -[(Removed "user error: " prefix with )2(%error)0( directive: it just clobbers the output and has absolutely no)]121 bull1 -[(functionality. Besides, this allows to write macros that does not differ from built-in functions in any way.)]110 bull2 -(193)pageodd -restore showpage -%%Page: 194 194 -%%BeginPageSetup -save -%%EndPageSetup -/194 pa -[(Added expansion of string that is output by )2(%error)0( directive. Now you can do things like:)]681 bull3 -[2( %define hello\(x\) Hello, x! )]664 code1 -[2()]653 code0 -[2( %define %$name andy )]642 code0 -[2( %error "hello\(%$name\)")]631 code2 -[(Same happened with )2(%include)0( directive.)]614 norm3 -[(Now all directives that expect an identifier will try to expand and concatenate everything without)]597 bull1 -[(whitespaces in between before usage. For example, with "unfixed" nasm the commands)]586 bull2 -[2( %define %$abc hello )]569 code1 -[2( %define __%$abc goodbye )]558 code0 -[2( __%$abc)]547 code2 -[(would produce "incorrect" output: last line will expand to)]530 norm3 -[2( hello goodbyehello)]513 code3 -[(Not quite what you expected, eh? :-\) The answer is that preprocessor treats the )2(%define)0( construct as if it)]496 norm1 -[(would be)]485 norm2 -[2( %define __ %$abc goodbye)]468 code3 -[(\(note the white space between __ and %$abc\). After my "fix" it will "correctly" expand into)]451 norm3 -[2( goodbye)]434 code3 -[(as expected. Note that I use quotes around words "correct", "incorrect" etc because this is rather a feature not)]417 norm1 -[(a bug; however current behaviour is more logical \(and allows more advanced macro usage :-\).)]406 norm2 -[(Same change was applied to: )2(%push)0(,)2(%macro)0(,)2(%imacro)0(,)2(%define)0(,)2(%idefine)0(,)2(%xdefine)0(,)2(%ixdefine)0(,)]389 norm1 -[2(%assign)0(,)2(%iassign)0(,)2(%undef)]378 norm2 -[(A new directive [WARNING {+|-}warning-id] have been added. It works only if the assembly phase is)]361 bull1 -[(enabled \(i.e. it doesn't work with nasm \226e\).)]350 bull2 -[(A new warning type: macro-selfref. By default this warning is disabled; when enabled NASM warns when)]333 bull1 -[(a macro self-references itself; for example the following source:)]322 bull2 -[2( [WARNING macro-selfref] )]305 code1 -[2()]294 code0 -[2( %macro push 1-* )]283 code0 -[2( %rep %0 )]272 code0 -[2( push %1 )]261 code0 -[2( %rotate 1 )]250 code0 -[2( %endrep )]239 code0 -[2( %endmacro )]228 code0 -[2()]217 code0 -[2( push eax,ebx,ecx)]206 code2 -[(will produce a warning, but if we remove the first line we won't see it anymore \(which is The Right Thing To)]189 norm1 -[(Do {tm} IMHO since C preprocessor eats such constructs without warnings at all\).)]178 norm2 -[(Added a "error" routine to preprocessor which always will set ERR_PASS1 bit in severity_code. This)]161 bull1 -[(removes annoying repeated errors on first and second passes from preprocessor.)]150 bull2 -[(Added the %+ operator in single-line macros for concatenating two identifiers. Usage example:)]133 bull3 -(194)pageeven -restore showpage -%%Page: 195 195 -%%BeginPageSetup -save -%%EndPageSetup -/195 pa -[2( %define _myfunc _otherfunc )]681 code1 -[2( %define cextern\(x\) _ %+ x )]670 code0 -[2( cextern \(myfunc\))]659 code2 -[(After first expansion, third line will become "_myfunc". After this expansion is performed again so it)]642 norm1 -[(becomes "_otherunc".)]631 norm2 -[(Now if preprocessor is in a non-emitting state, no warning or error will be emitted. Example:)]614 bull3 -[2( %if 1 )]597 code1 -[2( mov eax,ebx )]586 code0 -[2( %else )]575 code0 -[2( put anything you want between these two brackets, )]564 code0 -[2( even macro-parameter references %1 or local )]553 code0 -[2( labels %$zz or macro-local labels %%zz - no )]542 code0 -[2( warning will be emitted. )]531 code0 -[2( %endif)]520 code2 -[(Context-local variables on expansion as a last resort are looked up in outer contexts. For example, the)]503 bull1 -[(following piece:)]492 bull2 -[2( %push outer )]475 code1 -[2( %define %$a [esp] )]464 code0 -[2()]453 code0 -[2( %push inner )]442 code0 -[2( %$a )]431 code0 -[2( %pop )]420 code0 -[2( %pop)]409 code2 -[(will expand correctly the fourth line to [esp]; if we'll define another %$a inside the "inner" context, it will)]392 norm1 -[(take precedence over outer definition. However, this modification has been applied only to expand_smacro)]381 norm0 -[(and not to smacro_define: as a consequence expansion looks in outer contexts, but )2(%ifdef)0( won't look in)]370 norm0 -[(outer contexts.)]359 norm2 -[(This behaviour is needed because we don't want nested contexts to act on already defined local macros.)]342 norm1 -[(Example:)]331 norm2 -[2( %define %$arg1 [esp+4] )]314 code1 -[2( test eax,eax )]303 code0 -[2( if nz )]292 code0 -[2( mov eax,%$arg1 )]281 code0 -[2( endif)]270 code2 -[(In this example the "if" mmacro enters into the "if" context, so %$arg1 is not valid anymore inside "if". Of)]253 norm1 -[(course it could be worked around by using explicitely %$$arg1 but this is ugly IMHO.)]242 norm2 -[(Fixed memory leak in )2(%undef)0(. The origline wasn't freed before exiting on success.)]225 bull3 -[(Fixed trap in preprocessor when line expanded to empty set of tokens. This happens, for example, in the)]208 bull1 -[(following case:)]197 bull2 -[2( #define SOMETHING )]180 code1 -[2( SOMETHING)]169 code2 -[{/section-C.2.41 xa}(Version 0.98)](C.2.41)149.8 subh3 -[(All changes since NASM 0.98p3 have been produced by H. Peter Anvin <hpa@zytor.com>.)]132.8 norm3 -[(The documentation comment delimiter is)]115.8 bull3 -(195)pageodd -restore showpage -%%Page: 196 196 -%%BeginPageSetup -save -%%EndPageSetup -/196 pa -[(Allow EQU definitions to refer to external labels; reported by Pedro Gimeno.)]681 bull3 -[(Re-enable support for RDOFF v1; reported by Pedro Gimeno.)]664 bull3 -[(Updated License file per OK from Simon and Julian.)]647 bull3 -[{/section-C.2.42 xa}(Version 0.98p9)](C.2.42)627.8 subh3 -[(Update documentation \(although the instruction set reference will have to wait; I don't want to hold up the)]610.8 bull1 -[(0.98 release for it.\))]599.8 bull2 -[(Verified that the NASM implementation of the PEXTRW and PMOVMSKB instructions is correct. The)]582.8 bull1 -[(encoding differs from what the Intel manuals document, but the Pentium III behaviour matches NASM, not)]571.8 bull0 -[(the Intel manuals.)]560.8 bull2 -[(Fix handling of implicit sizes in PSHUFW and PINSRW, reported by Stefan Hoffmeister.)]543.8 bull3 -[(Resurrect the \226s option, which was removed when changing the diagnostic output to stdout.)]526.8 bull3 -[{/section-C.2.43 xa}(Version 0.98p8)](C.2.43)507.6 subh3 -[(Fix for "DB" when NASM is running on a bigendian machine.)]490.6 bull3 -[(Invoke insns.pl once for each output script, making Makefile.in legal for "make \226j".)]473.6 bull3 -[(Improve the Unix configure-based makefiles to make package creation easier.)]456.6 bull3 -[(Included an RPM .spec file for building RPM \(RedHat Package Manager\) packages on Linux or Unix)]439.6 bull1 -[(systems.)]428.6 bull2 -[(Fix Makefile dependency problems.)]411.6 bull3 -[(Change src/rdsrc.pl to include sectioning information in info output; required for install-info to work.)]394.6 bull3 -[(Updated the RDOFF distribution to version 2 from Jules; minor massaging to make it compile in my)]377.6 bull1 -[(environment.)]366.6 bull2 -[(Split doc files that can be built by anyone with a Perl interpreter off into a separate archive.)]349.6 bull3 -[("Dress rehearsal" release!)]332.6 bull3 -[{/section-C.2.44 xa}(Version 0.98p7)](C.2.44)313.4 subh3 -[(Fixed opcodes with a third byte-sized immediate argument to not complain if given "byte" on the)]296.4 bull1 -[(immediate.)]285.4 bull2 -[(Allow )2(%undef)0( to remove single-line macros with arguments. This matches the behaviour of #undef in the)]268.4 bull1 -[(C preprocessor.)]257.4 bull2 -[(Allow \226d, \226u, \226i and \226p to be specified as \226D, \226U, \226I and \226P for compatibility with most C compilers and)]240.4 bull1 -[(preprocessors. This allows Makefile options to be shared between cc and nasm, for example.)]229.4 bull2 -[(Minor cleanups.)]212.4 bull3 -[(Went through the list of Katmai instructions and hopefully fixed the \(rather few\) mistakes in it.)]195.4 bull3 -[(\(Hopefully\) fixed a number of disassembler bugs related to ambiguous instructions \(disambiguated by \226p\))]178.4 bull1 -[(and SSE instructions with REP.)]167.4 bull2 -[(Fix for bug reported by Mark Junger: "call dword 0x12345678" should work and may add an OSP)]150.4 bull1 -[(\(affected CALL, JMP, Jcc\).)]139.4 bull2 -[(Fix for environments when "stderr" isn't a compile-time constant.)]122.4 bull3 -(196)pageeven -restore showpage -%%Page: 197 197 -%%BeginPageSetup -save -%%EndPageSetup -/197 pa -[{/section-C.2.45 xa}(Version 0.98p6)](C.2.45)678.8 subh3 -[(Took officially over coordination of the 0.98 release; so drop the p3.x notation. Skipped p4 and p5 to avoid)]661.8 bull1 -[(confusion with John Fine's J4 and J5 releases.)]650.8 bull2 -[(Update the documentation; however, it still doesn't include documentation for the various new)]633.8 bull1 -[(instructions. I somehow wonder if it makes sense to have an instruction set reference in the assembler)]622.8 bull0 -[(manual when Intel et al have PDF versions of their manuals online.)]611.8 bull2 -[(Recognize "idt" or "centaur" for the \226p option to ndisasm.)]594.8 bull3 -[(Changed error messages back to stderr where they belong, but add an \226E option to redirect them elsewhere)]577.8 bull1 -[(\(the DOS shell cannot redirect stderr.\))]566.8 bull2 -[(\226M option to generate Makefile dependencies \(based on code from Alex Verstak.\))]549.8 bull3 -[2(%undef)0( preprocessor directive, and \226u option, that undefines a single-line macro.)]532.8 bull3 -[(OS/2 Makefile \(Mkfiles/Makefile.os2\) for Borland under OS/2; from Chuck Crayne.)]515.8 bull3 -[(Various minor bugfixes \(reported by\): \226 Dangling )2(%s)0( in preproc.c \(Martin Junker\))]498.8 bull3 -[(THERE ARE KNOWN BUGS IN SSE AND THE OTHER KATMAI INSTRUCTIONS. I am on a trip)]481.8 bull1 -[(and didn't bring the Katmai instruction reference, so I can't work on them right now.)]470.8 bull2 -[(Updated the License file per agreement with Simon and Jules to include a GPL distribution clause.)]453.8 bull3 -[{/section-C.2.46 xa}(Version 0.98p3.7)](C.2.46)434.6 subh3 -[(\(Hopefully\) fixed the canned Makefiles to include the outrdf2 and zoutieee modules.)]417.6 bull3 -[(Renamed changes.asm to changed.asm.)]400.6 bull3 -[{/section-C.2.47 xa}(Version 0.98p3.6)](C.2.47)381.4 subh3 -[(Fixed a bunch of instructions that were added in 0.98p3.5 which had memory operands, and the)]364.4 bull1 -[(address-size prefix was missing from the instruction pattern.)]353.4 bull2 -[{/section-C.2.48 xa}(Version 0.98p3.5)](C.2.48)334.2 subh3 -[(Merged in changes from John S. Fine's 0.98-J5 release. John's based 0.98-J5 on my 0.98p3.3 release; this)]317.2 bull1 -[(merges the changes.)]306.2 bull2 -[(Expanded the instructions flag field to a long so we can fit more flags; mark SSE \(KNI\) and AMD or)]289.2 bull1 -[(Katmai-specific instructions as such.)]278.2 bull2 -[(Fix the "PRIV" flag on a bunch of instructions, and create new "PROT" flag for protected-mode-only)]261.2 bull1 -[(instructions \(orthogonal to if the instruction is privileged!\) and new "SMM" flag for SMM-only)]250.2 bull0 -[(instructions.)]239.2 bull2 -[(Added AMD-only SYSCALL and SYSRET instructions.)]222.2 bull3 -[(Make SSE actually work, and add new Katmai MMX instructions.)]205.2 bull3 -[(Added a \226p \(preferred vendor\) option to ndisasm so that it can distinguish e.g. Cyrix opcodes also used in)]188.2 bull1 -[(SSE. For example:)]177.2 bull2 -[2( ndisasm -p cyrix aliased.bin )]160.2 code1 -[2( 00000000 670F514310 paddsiw mm0,[ebx+0x10] )]149.2 code0 -[2( 00000005 670F514320 paddsiw mm0,[ebx+0x20] )]138.2 code0 -[2( ndisasm -p intel aliased.bin )]127.2 code0 -[2( 00000000 670F514310 sqrtps xmm0,[ebx+0x10] )]116.2 code0 -[2( 00000005 670F514320 sqrtps xmm0,[ebx+0x20])]105.2 code2 -(197)pageodd -restore showpage -%%Page: 198 198 -%%BeginPageSetup -save -%%EndPageSetup -/198 pa -[(Added a bunch of Cyrix-specific instructions.)]681 bull3 -[{/section-C.2.49 xa}(Version 0.98p3.4)](C.2.49)661.8 subh3 -[(Made at least an attempt to modify all the additional Makefiles \(in the Mkfiles directory\). I can't test it, but)]644.8 bull1 -[(this was the best I could do.)]633.8 bull2 -[(DOS DJGPP+"Opus Make" Makefile from John S. Fine.)]616.8 bull3 -[(changes.asm changes from John S. Fine.)]599.8 bull3 -[{/section-C.2.50 xa}(Version 0.98p3.3)](C.2.50)580.6 subh3 -[(Patch from Conan Brink to allow nesting of )2(%rep)0( directives.)]563.6 bull3 -[(If we're going to allow INT01 as an alias for INT1/ICEBP \(one of Jules 0.98p3 changes\), then we should)]546.6 bull1 -[(allow INT03 as an alias for INT3 as well.)]535.6 bull2 -[(Updated changes.asm to include the latest changes.)]518.6 bull3 -[(Tried to clean up the <CR>s that had snuck in from a DOS/Windows environment into my Unix)]501.6 bull1 -[(environment, and try to make sure than DOS/Windows users get them back.)]490.6 bull2 -[(We would silently generate broken tools if insns.dat wasn't sorted properly. Change insns.pl so that the)]473.6 bull1 -[(order doesn't matter.)]462.6 bull2 -[(Fix bug in insns.pl \(introduced by me\) which would cause conditional instructions to have an extra "cc" in)]445.6 bull1 -[(disassembly, e.g. "jnz" disassembled as "jccnz".)]434.6 bull2 -[{/section-C.2.51 xa}(Version 0.98p3.2)](C.2.51)415.4 subh3 -[(Merged in John S. Fine's changes from his 0.98-J4 prerelease; see http://www.csoft.net/cz/johnfine/)]398.4 bull3 -[(Changed previous "spotless" Makefile target \(appropriate for distribution\) to "distclean", and added)]381.4 bull1 -[("cleaner" target which is same as "clean" except deletes files generated by Perl scripts; "spotless" is union.)]370.4 bull2 -[(Removed BASIC programs from distribution. Get a Perl interpreter instead \(see below.\))]353.4 bull3 -[(Calling this "pre-release 3.2" rather than "p3-hpa2" because of John's contributions.)]336.4 bull3 -[(Actually link in the IEEE output format \(zoutieee.c\); fix a bunch of compiler warnings in that file. Note I)]319.4 bull1 -[(don't know what IEEE output is supposed to look like, so these changes were made "blind".)]308.4 bull2 -[{/section-C.2.52 xa}(Version 0.98p3-hpa)](C.2.52)289.2 subh3 -[(Merged nasm098p3.zip with nasm-0.97.tar.gz to create a fully buildable version for Unix systems)]272.2 bull1 -[(\(Makefile.in updates, etc.\))]261.2 bull2 -[(Changed insns.pl to create the instruction tables in nasm.h and names.c, so that a new instruction can be)]244.2 bull1 -[(added by adding it *only* to insns.dat.)]233.2 bull2 -[(Added the following new instructions: SYSENTER, SYSEXIT, FXSAVE, FXRSTOR, UD1, UD2 \(the)]216.2 bull1 -[(latter two are two opcodes that Intel guarantee will never be used; one of them is documented as UD2 in)]205.2 bull0 -[(Intel documentation, the other one just as "Undefined Opcode" \226\226 calling it UD1 seemed to make sense.\))]194.2 bull2 -[(MAX_SYMBOL was defined to be 9, but LOADALL286 and LOADALL386 are 10 characters long. Now)]177.2 bull1 -[(MAX_SYMBOL is derived from insns.dat.)]166.2 bull2 -[(A note on the BASIC programs included: forget them. insns.bas is already out of date. Get yourself a Perl)]149.2 bull1 -[(interpreter for your platform of choice at ){(http://www.cpan.org/ports/index.html)wl}(http://www.cpan.org/ports/index.html){el}(.)]138.2 bull2 -[{/section-C.2.53 xa}(Version 0.98 pre-release 3)](C.2.53)119 subh3 -[(added response file support, improved command line handling, new layout help screen)]102 bull3 -(198)pageeven -restore showpage -%%Page: 199 199 -%%BeginPageSetup -save -%%EndPageSetup -/199 pa -[(fixed limit checking bug, 'OUT byte nn, reg' bug, and a couple of rdoff related bugs, updated Wishlist;)]681 bull1 -[(0.98 Prerelease 3.)]670 bull2 -[{/section-C.2.54 xa}(Version 0.98 pre-release 2)](C.2.54)650.8 subh3 -[(fixed bug in outcoff.c to do with truncating section names longer than 8 characters, referencing beyond end)]633.8 bull1 -[(of string; 0.98 pre-release 2)]622.8 bull2 -[{/section-C.2.55 xa}(Version 0.98 pre-release 1)](C.2.55)603.6 subh3 -[(Fixed a bug whereby STRUC didn't work at all in RDF.)]586.6 bull3 -[(Fixed a problem with group specification in PUBDEFs in OBJ.)]569.6 bull3 -[(Improved ease of adding new output formats. Contribution due to Fox Cutter.)]552.6 bull3 -[(Fixed a bug in relocations in the `bin' format: was showing up when a relocatable reference crossed an)]535.6 bull1 -[(8192-byte boundary in any output section.)]524.6 bull2 -[(Fixed a bug in local labels: local-label lookups were inconsistent between passes one and two if an EQU)]507.6 bull1 -[(occurred between the definition of a global label and the subsequent use of a local label local to that global.)]496.6 bull2 -[(Fixed a seg-fault in the preprocessor \(again\) which happened when you use a blank line as the first line of)]479.6 bull1 -[(a multi-line macro definition and then defined a label on the same line as a call to that macro.)]468.6 bull2 -[(Fixed a stale-pointer bug in the handling of the NASM environment variable. Thanks to Thomas)]451.6 bull1 -[(McWilliams.)]440.6 bull2 -[(ELF had a hard limit on the number of sections which caused segfaults when transgressed. Fixed.)]423.6 bull3 -[(Added ability for ndisasm to read from stdin by using `-' as the filename.)]406.6 bull3 -[(ndisasm wasn't outputting the TO keyword. Fixed.)]389.6 bull3 -[(Fixed error cascade on bogus expression in )2(%if)0( \226 an error in evaluation was causing the entire )2(%if)0( to be)]372.6 bull1 -[(discarded, thus creating trouble later when the )2(%else)0( or )2(%endif)0( was encountered.)]361.6 bull2 -[(Forward reference tracking was instruction-granular not operand- granular, which was causing)]344.6 bull1 -[(286-specific code to be generated needlessly on code of the form `shr word [forwardref],1'. Thanks to Jim)]333.6 bull0 -[(Hague for sending a patch.)]322.6 bull2 -[(All messages now appear on stdout, as sending them to stderr serves no useful purpose other than to make)]305.6 bull1 -[(redirection difficult.)]294.6 bull2 -[(Fixed the problem with EQUs pointing to an external symbol \226 this now generates an error message.)]277.6 bull3 -[(Allowed multiple size prefixes to an operand, of which only the first is taken into account.)]260.6 bull3 -[(Incorporated John Fine's changes, including fixes of a large number of preprocessor bugs, some small)]243.6 bull1 -[(problems in OBJ, and a reworking of label handling to define labels before their line is assembled, rather)]232.6 bull0 -[(than after.)]221.6 bull2 -[(Reformatted a lot of the source code to be more readable. Included 'coding.txt' as a guideline for how to)]204.6 bull1 -[(format code for contributors.)]193.6 bull2 -[(Stopped nested )2(%reps)0( causing a panic \226 they now cause a slightly more friendly error message instead.)]176.6 bull3 -[(Fixed floating point constant problems \(patch by Pedro Gimeno\))]159.6 bull3 -[(Fixed the return value of insn_size\(\) not being checked for \2261, indicating an error.)]142.6 bull3 -[(Incorporated 3Dnow! instructions.)]125.6 bull3 -[(Fixed the 'mov eax, eax + ebx' bug.)]108.6 bull3 -(199)pageodd -restore showpage -%%Page: 200 200 -%%BeginPageSetup -save -%%EndPageSetup -/200 pa -[(Fixed the GLOBAL EQU bug in ELF. Released developers release 3.)]681 bull3 -[(Incorporated John Fine's command line parsing changes)]664 bull3 -[(Incorporated David Lindauer's OMF debug support)]647 bull3 -[(Made changes for LCC 4.0 support \()2(__NASM_CDecl__)0(, removed register size specification warning)]630 bull1 -[(when sizes agree\).)]619 bull2 -[{/section-C.3 xa}(NASM 0.9 Series)](C.3)597.6 head3 -[(Revisions before 0.98.)]580.6 norm3 -[{/section-C.3.1 xa}(Version 0.97 released December 1997)](C.3.1)561.4 subh3 -[(This was entirely a bug-fix release to 0.96, which seems to have got cursed. Silly me.)]544.4 bull3 -[(Fixed stupid mistake in OBJ which caused `MOV EAX,<constant>' to fail. Caused by an error in the)]527.4 bull1 -[(`MOV EAX,<segment>' support.)]516.4 bull2 -[(ndisasm hung at EOF when compiled with lcc on Linux because lcc on Linux somehow breaks feof\(\).)]499.4 bull1 -[(ndisasm now does not rely on feof\(\).)]488.4 bull2 -[(A heading in the documentation was missing due to a markup error in the indexing. Fixed.)]471.4 bull3 -[(Fixed failure to update all pointers on realloc\(\) within extended- operand code in parser.c. Was causing)]454.4 bull1 -[(wrong behaviour and seg faults on lines such as `dd 0.0,0.0,0.0,0.0,...')]443.4 bull2 -[(Fixed a subtle preprocessor bug whereby invoking one multi-line macro on the first line of the expansion)]426.4 bull1 -[(of another, when the second had been invoked with a label defined before it, didn't expand the inner macro.)]415.4 bull2 -[(Added internal.doc back in to the distribution archives \226 it was missing in 0.96 *blush*)]398.4 bull3 -[(Fixed bug causing 0.96 to be unable to assemble its own test files, specifically objtest.asm. *blush again*)]381.4 bull3 -[(Fixed seg-faults and bogus error messages caused by mismatching )2(%rep)0( and )2(%endrep)0( within multi-line)]364.4 bull1 -[(macro definitions.)]353.4 bull2 -[(Fixed a problem with buffer overrun in OBJ, which was causing corruption at ends of long PUBDEF)]336.4 bull1 -[(records.)]325.4 bull2 -[(Separated DOS archives into main-program and documentation to reduce download size.)]308.4 bull3 -[{/section-C.3.2 xa}(Version 0.96 released November 1997)](C.3.2)289.2 subh3 -[(Fixed a bug whereby, if `nasm sourcefile' would cause a filename collision warning and put output into)]272.2 bull1 -[(`nasm.out', then `nasm sourcefile \226o outputfile' still gave the warning even though the `-o' was honoured.)]261.2 bull0 -[(Fixed name pollution under Digital UNIX: one of its header files defined R_SP, which broke the enum in)]250.2 bull0 -[(nasm.h.)]239.2 bull2 -[(Fixed minor instruction table problems: FUCOM and FUCOMP didn't have two-operand forms;)]222.2 bull1 -[(NDISASM didn't recognise the longer register forms of PUSH and POP \(eg FF F3 for PUSH BX\); TEST)]211.2 bull0 -[(mem,imm32 was flagged as undocumented; the 32-bit forms of CMOV had 16-bit operand size prefixes;)]200.2 bull0 -[(`AAD imm' and `AAM imm' are no longer flagged as undocumented because the Intel Architecture)]189.2 bull0 -[(reference documents them.)]178.2 bull2 -[(Fixed a problem with the local-label mechanism, whereby strange types of symbol \(EQUs, auto-defined)]161.2 bull1 -[(OBJ segment base symbols\) interfered with the `previous global label' value and screwed up local labels.)]150.2 bull2 -[(Fixed a bug whereby the stub preprocessor didn't communicate with the listing file generator, so that the)]133.2 bull1 -[(\226a and \226l options in conjunction would produce a useless listing file.)]122.2 bull2 -(200)pageeven -restore showpage -%%Page: 201 201 -%%BeginPageSetup -save -%%EndPageSetup -/201 pa -[(Merged `os2' object file format back into `obj', after discovering that `obj' _also_ shouldn't have a link)]681 bull1 -[(pass separator in a module containing a non-trivial MODEND. Flat segments are now declared using the)]670 bull0 -[(FLAT attribute. `os2' is no longer a valid object format name: use `obj'.)]659 bull2 -[(Removed the fixed-size temporary storage in the evaluator. Very very long expressions \(like `mov)]642 bull1 -[(ax,1+1+1+1+...' for two hundred 1s or so\) should now no longer crash NASM.)]631 bull2 -[(Fixed a bug involving segfaults on disassembly of MMX instructions, by changing the meaning of one of)]614 bull1 -[(the operand-type flags in nasm.h. This may cause other apparently unrelated MMX problems; it needs to)]603 bull0 -[(be tested thoroughly.)]592 bull2 -[(Fixed some buffer overrun problems with large OBJ output files. Thanks to DJ Delorie for the bug report)]575 bull1 -[(and fix.)]564 bull2 -[(Made preprocess-only mode actually listen to the )2(%line)0( markers as it prints them, so that it can report)]547 bull1 -[(errors more sanely.)]536 bull2 -[(Re-designed the evaluator to keep more sensible track of expressions involving forward references: can)]519 bull1 -[(now cope with previously-nightmare situations such as:)]508 bull2 -[2( mov ax,foo | bar )]491 code1 -[2( foo equ 1 )]480 code0 -[2( bar equ 2)]469 code2 -[(Added the ALIGN and ALIGNB standard macros.)]452 bull3 -[(Added PIC support in ELF: use of WRT to obtain the four extra relocation types needed.)]435 bull3 -[(Added the ability for output file formats to define their own extensions to the GLOBAL, COMMON and)]418 bull1 -[(EXTERN directives.)]407 bull2 -[(Implemented common-variable alignment, and global-symbol type and size declarations, in ELF.)]390 bull3 -[(Implemented NEAR and FAR keywords for common variables, plus far-common element size)]373 bull1 -[(specification, in OBJ.)]362 bull2 -[(Added a feature whereby EXTERNs and COMMONs in OBJ can be given a default WRT specification)]345 bull1 -[(\(either a segment or a group\).)]334 bull2 -[(Transformed the Unix NASM archive into an auto-configuring package.)]317 bull3 -[(Added a sanity-check for people applying SEG to things which are already segment bases: this previously)]300 bull1 -[(went unnoticed by the SEG processing and caused OBJ-driver panics later.)]289 bull2 -[(Added the ability, in OBJ format, to deal with `MOV EAX,<segment>' type references: OBJ doesn't)]272 bull1 -[(directly support dword-size segment base fixups, but as long as the low two bytes of the constant term are)]261 bull0 -[(zero, a word-size fixup can be generated instead and it will work.)]250 bull2 -[(Added the ability to specify sections' alignment requirements in Win32 object files and pure binary files.)]233 bull3 -[(Added preprocess-time expression evaluation: the )2(%assign)0( \(and )2(%iassign)0(\) directive and the bare )2(%if)]216 bull1 -[(\(and )2(%elif)0(\) conditional. Added relational operators to the evaluator, for use only in )2(%if)0( constructs: the)]205 bull0 -[(standard relationals = < > <= >= <> \(and C-like synonyms == and !=\) plus low-precedence logical)]194 bull0 -[(operators &&, ^^ and ||.)]183 bull2 -[(Added a preprocessor repeat construct: )2(%rep)0( / )2(%exitrep)0( / )2(%endrep)0(.)]166 bull3 -[(Added the __FILE__ and __LINE__ standard macros.)]149 bull3 -[(Added a sanity check for number constants being greater than 0xFFFFFFFF. The warning can be disabled.)]132 bull3 -[(Added the %0 token whereby a variadic multi-line macro can tell how many parameters it's been given in)]115 bull1 -[(a specific invocation.)]104 bull2 -(201)pageodd -restore showpage -%%Page: 202 202 -%%BeginPageSetup -save -%%EndPageSetup -/202 pa -[(Added )2(%rotate)0(, allowing multi-line macro parameters to be cycled.)]681 bull3 -[(Added the `*' option for the maximum parameter count on multi-line macros, allowing them to take)]664 bull1 -[(arbitrarily many parameters.)]653 bull2 -[(Added the ability for the user-level forms of EXTERN, GLOBAL and COMMON to take more than one)]636 bull1 -[(argument.)]625 bull2 -[(Added the IMPORT and EXPORT directives in OBJ format, to deal with Windows DLLs.)]608 bull3 -[(Added some more preprocessor )2(%if)0( constructs: )2(%ifidn)0( / )2(%ifidni)0( \(exact textual identity\), and )2(%ifid)]591 bull1 -[(/ )2(%ifnum)0( / )2(%ifstr)0( \(token type testing\).)]580 bull2 -[(Added the ability to distinguish SHL AX,1 \(the 8086 version\) from SHL AX,BYTE 1 \(the)]563 bull1 -[(286-and-upwards version whose constant happens to be 1\).)]552 bull2 -[(Added NetBSD/FreeBSD/OpenBSD's variant of a.out format, complete with PIC shared library features.)]535 bull3 -[(Changed NASM's idiosyncratic handling of FCLEX, FDISI, FENI, FINIT, FSAVE, FSTCW, FSTENV,)]518 bull1 -[(and FSTSW to bring it into line with the otherwise accepted standard. The previous behaviour, though it)]507 bull0 -[(was a deliberate feature, was a deliberate feature based on a misunderstanding. Apologies for the)]496 bull0 -[(inconvenience.)]485 bull2 -[(Improved the flexibility of ABSOLUTE: you can now give it an expression rather than being restricted to a)]468 bull1 -[(constant, and it can take relocatable arguments as well.)]457 bull2 -[(Added the ability for a variable to be declared as EXTERN multiple times, and the subsequent definitions)]440 bull1 -[(are just ignored.)]429 bull2 -[(We now allow instruction prefixes \(CS, DS, LOCK, REPZ etc\) to be alone on a line \(without a following)]412 bull1 -[(instruction\).)]401 bull2 -[(Improved sanity checks on whether the arguments to EXTERN, GLOBAL and COMMON are valid)]384 bull1 -[(identifiers.)]373 bull2 -[(Added misc/exebin.mac to allow direct generation of .EXE files by hacking up an EXE header using DB)]356 bull1 -[(and DW; also added test/binexe.asm to demonstrate the use of this. Thanks to Yann Guidon for)]345 bull0 -[(contributing the EXE header code.)]334 bull2 -[(ndisasm forgot to check whether the input file had been successfully opened. Now it does. Doh!)]317 bull3 -[(Added the Cyrix extensions to the MMX instruction set.)]300 bull3 -[(Added a hinting mechanism to allow [EAX+EBX] and [EBX+EAX] to be assembled differently. This is)]283 bull1 -[(important since [ESI+EBP] and [EBP+ESI] have different default base segment registers.)]272 bull2 -[(Added support for the PharLap OMF extension for 4096-byte segment alignment.)]255 bull3 -[{/section-C.3.3 xa}(Version 0.95 released July 1997)](C.3.3)235.8 subh3 -[(Fixed yet another ELF bug. This one manifested if the user relied on the default segment, and attempted to)]218.8 bull1 -[(define global symbols without first explicitly declaring the target segment.)]207.8 bull2 -[(Added makefiles \(for NASM and the RDF tools\) to build Win32 console apps under Symantec C++.)]190.8 bull1 -[(Donated by Mark Junker.)]179.8 bull2 -[(Added `macros.bas' and `insns.bas', QBasic versions of the Perl scripts that convert `standard.mac' to)]162.8 bull1 -[(`macros.c' and convert `insns.dat' to `insnsa.c' and `insnsd.c'. Also thanks to Mark Junker.)]151.8 bull2 -[(Changed the diassembled forms of the conditional instructions so that JB is now emitted as JC, and other)]134.8 bull1 -[(similar changes. Suggested list by Ulrich Doewich.)]123.8 bull2 -[(Added `@' to the list of valid characters to begin an identifier with.)]106.8 bull3 -(202)pageeven -restore showpage -%%Page: 203 203 -%%BeginPageSetup -save -%%EndPageSetup -/203 pa -[(Documentary changes, notably the addition of the `Common Problems' section in nasm.doc.)]681 bull3 -[(Fixed a bug relating to 32-bit PC-relative fixups in OBJ.)]664 bull3 -[(Fixed a bug in perm_copy\(\) in labels.c which was causing exceptions in cleanup_labels\(\) on some systems.)]647 bull3 -[(Positivity sanity check in TIMES argument changed from a warning to an error following a further)]630 bull1 -[(complaint.)]619 bull2 -[(Changed the acceptable limits on byte and word operands to allow things like `~10111001b' to work.)]602 bull3 -[(Fixed a major problem in the preprocessor which caused seg-faults if macro definitions contained blank)]585 bull1 -[(lines or comment-only lines.)]574 bull2 -[(Fixed inadequate error checking on the commas separating the arguments to `db', `dw' etc.)]557 bull3 -[(Fixed a crippling bug in the handling of macros with operand counts defined with a `+' modifier.)]540 bull3 -[(Fixed a bug whereby object file formats which stored the input file name in the output file \(such as OBJ)]523 bull1 -[(and COFF\) weren't doing so correctly when the output file name was specified on the command line.)]512 bull2 -[(Removed [INC] and [INCLUDE] support for good, since they were obsolete anyway.)]495 bull3 -[(Fixed a bug in OBJ which caused all fixups to be output in 16-bit \(old-format\) FIXUPP records, rather)]478 bull1 -[(than putting the 32-bit ones in FIXUPP32 \(new-format\) records.)]467 bull2 -[(Added, tentatively, OS/2 object file support \(as a minor variant on OBJ\).)]450 bull3 -[(Updates to Fox Cutter's Borland C makefile, Makefile.bc2.)]433 bull3 -[(Removed a spurious second fclose\(\) on the output file.)]416 bull3 -[(Added the `-s' command line option to redirect all messages which would go to stderr \(errors, help text\) to)]399 bull1 -[(stdout instead.)]388 bull2 -[(Added the `-w' command line option to selectively suppress some classes of assembly warning messages.)]371 bull3 -[(Added the `-p' pre-include and `-d' pre-define command-line options.)]354 bull3 -[(Added an include file search path: the `-i' command line option.)]337 bull3 -[(Fixed a silly little preprocessor bug whereby starting a line with a `%!' environment-variable reference)]320 bull1 -[(caused an `unknown directive' error.)]309 bull2 -[(Added the long-awaited listing file support: the `-l' command line option.)]292 bull3 -[(Fixed a problem with OBJ format whereby, in the absence of any explicit segment definition, non-global)]275 bull1 -[(symbols declared in the implicit default segment generated spurious EXTDEF records in the output.)]264 bull2 -[(Added the NASM environment variable.)]247 bull3 -[(From this version forward, Win32 console-mode binaries will be included in the DOS distribution in)]230 bull1 -[(addition to the 16-bit binaries. Added Makefile.vc for this purpose.)]219 bull2 -[(Added `return 0;' to test/objlink.c to prevent compiler warnings.)]202 bull3 -[(Added the __NASM_MAJOR__ and __NASM_MINOR__ standard defines.)]185 bull3 -[(Added an alternative memory-reference syntax in which prefixing an operand with `&' is equivalent to)]168 bull1 -[(enclosing it in square brackets, at the request of Fox Cutter.)]157 bull2 -[(Errors in pass two now cause the program to return a non-zero error code, which they didn't before.)]140 bull3 -[(Fixed the single-line macro cycle detection, which didn't work at all on macros with no parameters)]123 bull1 -[(\(caused an infinite loop\). Also changed the behaviour of single-line macro cycle detection to work like)]112 bull0 -[(cpp, so that macros like `extrn' as given in the documentation can be implemented.)]101 bull2 -(203)pageodd -restore showpage -%%Page: 204 204 -%%BeginPageSetup -save -%%EndPageSetup -/204 pa -[(Fixed the implementation of WRT, which was too restrictive in that you couldn't do `mov ax,[di+abc wrt)]681 bull1 -[(dgroup]' because \(di+abc\) wasn't a relocatable reference.)]670 bull2 -[{/section-C.3.4 xa}(Version 0.94 released April 1997)](C.3.4)650.8 subh3 -[(Major item: added the macro processor.)]633.8 bull3 -[(Added undocumented instructions SMI, IBTS, XBTS and LOADALL286. Also reorganised CMPXCHG)]616.8 bull1 -[(instruction into early-486 and Pentium forms. Thanks to Thobias Jones for the information.)]605.8 bull2 -[(Fixed two more stupid bugs in ELF, which were causing `ld' to continue to seg-fault in a lot of non-trivial)]588.8 bull1 -[(cases.)]577.8 bull2 -[(Fixed a seg-fault in the label manager.)]560.8 bull3 -[(Stopped FBLD and FBSTP from _requiring_ the TWORD keyword, which is the only option for BCD)]543.8 bull1 -[(loads/stores in any case.)]532.8 bull2 -[(Ensured FLDCW, FSTCW and FSTSW can cope with the WORD keyword, if anyone bothers to provide)]515.8 bull1 -[(it. Previously they complained unless no keyword at all was present.)]504.8 bull2 -[(Some forms of FDIV/FDIVR and FSUB/FSUBR were still inverted: a vestige of a bug that I thought had)]487.8 bull1 -[(been fixed in 0.92. This was fixed, hopefully for good this time...)]476.8 bull2 -[(Another minor phase error \(insofar as a phase error can _ever_ be minor\) fixed, this one occurring in code)]459.8 bull1 -[(of the form)]448.8 bull2 -[2( rol ax,forward_reference )]431.8 code1 -[2( forward_reference equ 1)]420.8 code2 -[(The number supplied to TIMES is now sanity-checked for positivity, and also may be greater than 64K)]403.8 bull1 -[(\(which previously didn't work on 16-bit systems\).)]392.8 bull2 -[(Added Watcom C makefiles, and misc/pmw.bat, donated by Dominik Behr.)]375.8 bull3 -[(Added the INCBIN pseudo-opcode.)]358.8 bull3 -[(Due to the advent of the preprocessor, the [INCLUDE] and [INC] directives have become obsolete. They)]341.8 bull1 -[(are still supported in this version, with a warning, but won't be in the next.)]330.8 bull2 -[(Fixed a bug in OBJ format, which caused incorrect object records to be output when absolute labels were)]313.8 bull1 -[(made global.)]302.8 bull2 -[(Updates to RDOFF subdirectory, and changes to outrdf.c.)]285.8 bull3 -[{/section-C.3.5 xa}(Version 0.93 released January 1997)](C.3.5)266.6 subh3 -[(This release went out in a great hurry after semi-crippling bugs were found in 0.92.)]249.6 norm3 -[(Really )1(did)0( fix the stack overflows this time. *blush*)]232.6 bull3 -[(Had problems with EA instruction sizes changing between passes, when an offset contained a forward)]215.6 bull1 -[(reference and so 4 bytes were allocated for the offset in pass one; by pass two the symbol had been defined)]204.6 bull0 -[(and happened to be a small absolute value, so only 1 byte got allocated, causing instruction size mismatch)]193.6 bull0 -[(between passes and hence incorrect address calculations. Fixed.)]182.6 bull2 -[(Stupid bug in the revised ELF section generation fixed \(associated string-table section for .symtab was)]165.6 bull1 -[(hard-coded as 7, even when this didn't fit with the real section table\). Was causing `ld' to seg-fault under)]154.6 bull0 -[(Linux.)]143.6 bull2 -[(Included a new Borland C makefile, Makefile.bc2, donated by Fox Cutter <lmb@comtch.iea.com>.)]126.6 bull3 -(204)pageeven -restore showpage -%%Page: 205 205 -%%BeginPageSetup -save -%%EndPageSetup -/205 pa -[{/section-C.3.6 xa}(Version 0.92 released January 1997)](C.3.6)678.8 subh3 -[(The FDIVP/FDIVRP and FSUBP/FSUBRP pairs had been inverted: this was fixed. This also affected the)]661.8 bull1 -[(LCC driver.)]650.8 bull2 -[(Fixed a bug regarding 32-bit effective addresses of the form )2([other_register+ESP])0(.)]633.8 bull3 -[(Documentary changes, notably documentation of the fact that Borland Win32 compilers use `obj' rather)]616.8 bull1 -[(than `win32' object format.)]605.8 bull2 -[(Fixed the COMENT record in OBJ files, which was formatted incorrectly.)]588.8 bull3 -[(Fixed a bug causing segfaults in large RDF files.)]571.8 bull3 -[(OBJ format now strips initial periods from segment and group definitions, in order to avoid complications)]554.8 bull1 -[(with the local label syntax.)]543.8 bull2 -[(Fixed a bug in disassembling far calls and jumps in NDISASM.)]526.8 bull3 -[(Added support for user-defined sections in COFF and ELF files.)]509.8 bull3 -[(Compiled the DOS binaries with a sensible amount of stack, to prevent stack overflows on any arithmetic)]492.8 bull1 -[(expression containing parentheses.)]481.8 bull2 -[(Fixed a bug in handling of files that do not terminate in a newline.)]464.8 bull3 -[{/section-C.3.7 xa}(Version 0.91 released November 1996)](C.3.7)445.6 subh3 -[(Loads of bug fixes.)]428.6 bull3 -[(Support for RDF added.)]411.6 bull3 -[(Support for DBG debugging format added.)]394.6 bull3 -[(Support for 32-bit extensions to Microsoft OBJ format added.)]377.6 bull3 -[(Revised for Borland C: some variable names changed, makefile added.)]360.6 bull3 -[(LCC support revised to actually work.)]343.6 bull3 -[(JMP/CALL NEAR/FAR notation added.)]326.6 bull3 -[(`a16', `o16', `a32' and `o32' prefixes added.)]309.6 bull3 -[(Range checking on short jumps implemented.)]292.6 bull3 -[(MMX instruction support added.)]275.6 bull3 -[(Negative floating point constant support added.)]258.6 bull3 -[(Memory handling improved to bypass 64K barrier under DOS.)]241.6 bull3 -[2($)0( prefix to force treatment of reserved words as identifiers added.)]224.6 bull3 -[(Default-size mechanism for object formats added.)]207.6 bull3 -[(Compile-time configurability added.)]190.6 bull3 -[2(#)0(, )2(@)0(, )2(~)0( and c{?} are now valid characters in labels.)]173.6 bull3 -[2(-e)0( and )2(-k)0( options in NDISASM added.)]156.6 bull3 -[{/section-C.3.8 xa}(Version 0.90 released October 1996)](C.3.8)137.4 subh3 -[(First release version. First support for object file output. Other changes from previous version \(0.3x\) too)]120.4 norm1 -[(numerous to document.)]109.4 norm2 -(205)pageodd -restore showpage -%%Page: 206 206 -%%BeginPageSetup -save -%%EndPageSetup -/206 pa -[{/index xa}(Index)]642.8 chap3 -[2(!)0( operator, unary)][{/34 pl}(34){el}]607.8 0 idx03 -[2(!=)0( operator)][{/50 pl}(50){el}]596.8 0 idx03 -[2($$)0( token)][{/33 pl}(33,){el}( ){/86 pl}(86){el}]585.8 0 idx03 -[2($)][]574.8 0 idx03 -[(Here token)][{/33 pl}(33){el}]563.8 0 idx13 -[(prefix)][{/26 pl}(26,){el}( ){/29 pl}(29,){el}( ){/89 pl}(89){el}]552.8 0 idx13 -[2(%)0( operator)][{/33 pl}(33){el}]541.8 0 idx03 -[2(%!)][{/60 pl}(60){el}]530.8 0 idx03 -[2(%$)0( and )2(%$$)0( prefixes)][{/54 pl}(54,){el}( ){/55 pl}(55){el}]519.8 0 idx03 -[2(%%)0( operator)][{/33 pl}(33,){el}( ){/44 pl}(44){el}]508.8 0 idx03 -[2(%+)][{/39 pl}(39){el}]497.8 0 idx03 -[2(%?)][{/39 pl}(39){el}]486.8 0 idx03 -[2(%??)][{/39 pl}(39){el}]475.8 0 idx03 -[2(%[)][{/39 pl}(39){el}]464.8 0 idx03 -[2(&)0( operator)][{/33 pl}(33){el}]453.8 0 idx03 -[2(&&)0( operator)][{/50 pl}(50){el}]442.8 0 idx03 -[2(*)0( operator)][{/33 pl}(33){el}]431.8 0 idx03 -[2(+)0( modifier)][{/44 pl}(44){el}]420.8 0 idx03 -[2(+)0( operator)][]409.8 0 idx03 -[(binary)][{/33 pl}(33){el}]398.8 0 idx13 -[(unary)][{/34 pl}(34){el}]387.8 0 idx13 -[2(-)0( operator)][]376.8 0 idx03 -[(binary)][{/33 pl}(33){el}]365.8 0 idx13 -[(unary)][{/34 pl}(34){el}]354.8 0 idx13 -[2(..@)0( symbol prefix)][{/36 pl}(36,){el}( ){/44 pl}(44){el}]343.8 0 idx03 -[2(/)0( operator)][{/33 pl}(33){el}]332.8 0 idx03 -[2(//)0( operator)][{/33 pl}(33){el}]321.8 0 idx03 -[2(<)0( operator)][{/50 pl}(50){el}]310.8 0 idx03 -[2(<<)0( operator)][{/33 pl}(33){el}]299.8 0 idx03 -[2(<=)0( operator)][{/50 pl}(50){el}]288.8 0 idx03 -[2(<>)0( operator)][{/50 pl}(50){el}]277.8 0 idx03 -[2(=)0( operator)][{/50 pl}(50){el}]266.8 0 idx03 -[2(==)0( operator)][{/50 pl}(50){el}]255.8 0 idx03 -[2(>)0( operator)][{/50 pl}(50){el}]244.8 0 idx03 -[2(>=)0( operator)][{/50 pl}(50){el}]233.8 0 idx03 -[2(>>)0( operator)][{/33 pl}(33){el}]222.8 0 idx03 -[2(?)0( MASM syntax)][{/27 pl}(27){el}]211.8 0 idx03 -[2(^)0( operator)][{/33 pl}(33){el}]200.8 0 idx03 -[2(^^)0( operator)][{/50 pl}(50){el}]189.8 0 idx03 -[2(|)0( operator)][{/33 pl}(33){el}]178.8 0 idx03 -[2(||)0( operator)][{/50 pl}(50){el}]167.8 0 idx03 -[2(~)0( operator)][{/34 pl}(34){el}]156.8 0 idx03 -[2(%0)0( parameter count)][{/45 pl}(45){el}]145.8 0 idx03 -[2(%+1)0( and )2(%-1)0( syntax)][{/47 pl}(47){el}]134.8 0 idx03 -[(16-bit mode, versus 32-bit mode)][{/67 pl}(67){el}]123.8 0 idx03 -[(64-bit displacement)][{/114 pl}(114){el}]112.8 0 idx03 -[(64-bit immediate)][{/113 pl}(113){el}]101.8 0 idx03 -[2(-a)0( option)][{/21 pl}(21,){el}( ){/120 pl}(120){el}]607.8 1 idx03 -[2(A16)][{/26 pl}(26){el}]596.8 1 idx03 -[2(a16)][{/111 pl}(111){el}]585.8 1 idx03 -[2(A32)][{/26 pl}(26){el}]574.8 1 idx03 -[2(a32)][{/111 pl}(111){el}]563.8 1 idx03 -[2(A64)][{/26 pl}(26){el}]552.8 1 idx03 -[2(a64)][{/111 pl}(111){el}]541.8 1 idx03 -[2(a86)][{/14 pl}(14,){el}( ){/24 pl}(24,){el}( ){/25 pl}(25){el}]530.8 1 idx03 -[2(ABS)][{/29 pl}(29){el}]519.8 1 idx03 -[2(ABSOLUTE)][{/69 pl}(69,){el}( ){/76 pl}(76){el}]508.8 1 idx03 -[(addition)][{/33 pl}(33){el}]497.8 1 idx03 -[(addressing, mixed-size)][{/110 pl}(110){el}]486.8 1 idx03 -[(address-size prefixes)][{/26 pl}(26){el}]475.8 1 idx03 -[(algebra)][{/29 pl}(29){el}]464.8 1 idx03 -[2(ALIGN)][{/64 pl}(64,){el}( ){/66 pl}(66,){el}( ){/73 pl}(73,){el}( ){/76 pl}(76){el}]453.8 1 idx03 -[(smart)][{/66 pl}(66){el}]442.8 1 idx13 -[2(ALIGNB)][{/64 pl}(64){el}]431.8 1 idx03 -[(alignment)][]420.8 1 idx03 -[(in )2(bin)0( sections)][{/74 pl}(74){el}]409.8 1 idx13 -[(in )2(elf)0( sections)][{/85 pl}(85){el}]398.8 1 idx13 -[(in )2(obj)0( sections)][{/76 pl}(76){el}]387.8 1 idx13 -[(in )2(win32)0( sections)][{/79 pl}(79){el}]376.8 1 idx13 -[(of )2(elf)0( common variables)][{/87 pl}(87){el}]365.8 1 idx13 -[2(ALIGNMODE)][{/66 pl}(66){el}]354.8 1 idx03 -[2(__ALIGNMODE__)][{/66 pl}(66){el}]343.8 1 idx03 -[2(ALINK)][{/91 pl}(91){el}]332.8 1 idx03 -[2(alink.sourceforge.net)][{/91 pl}(91){el}]321.8 1 idx03 -[2(all)][{/23 pl}(23){el}]310.8 1 idx03 -[2(alloc)][{/85 pl}(85){el}]299.8 1 idx03 -[(alternate register names)][{/66 pl}(66){el}]288.8 1 idx03 -[2(alt.lang.asm)][{/14 pl}(14){el}]277.8 1 idx03 -[2(altreg)][{/66 pl}(66){el}]266.8 1 idx03 -[(ambiguity)][{/24 pl}(24){el}]255.8 1 idx03 -[2(a.out)][]244.8 1 idx03 -[(BSD version)][{/88 pl}(88){el}]233.8 1 idx13 -[(Linux version)][{/88 pl}(88){el}]222.8 1 idx13 -[2(aout)][{/88 pl}(88){el}]211.8 1 idx03 -[2(aoutb)][{/88 pl}(88,){el}( ){/106 pl}(106){el}]200.8 1 idx03 -[2(%arg)][{/57 pl}(57){el}]189.8 1 idx03 -[2(arg)][{/98 pl}(98,){el}( ){/105 pl}(105){el}]178.8 1 idx03 -[2(as86)][{/14 pl}(14,){el}( ){/88 pl}(88){el}]167.8 1 idx03 -[(assembler directives)][{/67 pl}(67){el}]156.8 1 idx03 -[(assembly-time options)][{/21 pl}(21){el}]145.8 1 idx03 -[2(%assign)][{/40 pl}(40){el}]134.8 1 idx03 -[2(ASSUME)][{/25 pl}(25){el}]123.8 1 idx03 -[2(AT)][{/63 pl}(63){el}]112.8 1 idx03 -[(Autoconf)][{/16 pl}(16){el}]101.8 1 idx03 -(206)pageeven -restore showpage -%%Page: 207 207 -%%BeginPageSetup -save -%%EndPageSetup -/207 pa -[2(autoexec.bat)][{/15 pl}(15){el}]681 0 idx03 -[2(auto-sync)][{/120 pl}(120){el}]670 0 idx03 -[2(-b)][{/119 pl}(119){el}]659 0 idx03 -[(bin)][{/18 pl}(18,){el}( ){/73 pl}(73){el}]648 0 idx03 -[(multisection)][{/74 pl}(74){el}]637 0 idx13 -[(binary)][{/29 pl}(29){el}]626 0 idx03 -[(binary files)][{/27 pl}(27){el}]615 0 idx03 -[(bit shift)][{/33 pl}(33){el}]604 0 idx03 -[2(BITS)][{/67 pl}(67,){el}( ){/73 pl}(73){el}]593 0 idx03 -[2(__BITS__)][{/61 pl}(61){el}]582 0 idx03 -[(bitwise AND)][{/33 pl}(33){el}]571 0 idx03 -[(bitwise OR)][{/33 pl}(33){el}]560 0 idx03 -[(bitwise XOR)][{/33 pl}(33){el}]549 0 idx03 -[(block IFs)][{/55 pl}(55){el}]538 0 idx03 -[(boot loader)][{/73 pl}(73){el}]527 0 idx03 -[(boot sector)][{/116 pl}(116){el}]516 0 idx03 -[(Borland)][]505 0 idx03 -[(Pascal)][{/99 pl}(99){el}]494 0 idx13 -[(Win32 compilers)][{/75 pl}(75){el}]483 0 idx13 -[(braces)][]472 0 idx03 -[(after )2(%)0( sign)][{/47 pl}(47){el}]461 0 idx13 -[(around macro parameters)][{/42 pl}(42){el}]450 0 idx13 -[(BSD)][{/106 pl}(106){el}]439 0 idx03 -[2(.bss)][{/85 pl}(85,){el}( ){/88 pl}(88,){el}( ){/89 pl}(89){el}]428 0 idx03 -[(bugs)][{/117 pl}(117){el}]417 0 idx03 -[2(bugtracker)][{/117 pl}(117){el}]406 0 idx03 -[2(BYTE)][{/116 pl}(116){el}]395 0 idx03 -[(C calling convention)][{/96 pl}(96,){el}( ){/103 pl}(103){el}]384 0 idx03 -[(C symbol names)][{/94 pl}(94){el}]373 0 idx03 -[2(c16.mac)][{/98 pl}(98,){el}( ){/101 pl}(101){el}]362 0 idx03 -[2(c32.mac)][{/105 pl}(105){el}]351 0 idx03 -[2(CALL FAR)][{/34 pl}(34){el}]340 0 idx03 -[(case sensitivity)][{/24 pl}(24,){el}( ){/37 pl}(37,){el}( ){/38 pl}(38,){el}( ){/40 pl}(40,){el}( ){/42 pl}(42,){el}( ){/43 pl}(43,){el}( ){/50 pl}(50,){el}( ){/77 pl}(77){el}]329 0 idx03 -[(changing sections)][{/68 pl}(68){el}]318 0 idx03 -[(character constant)][{/27 pl}(27,){el}( ){/31 pl}(31){el}]307 0 idx03 -[(character strings)][{/30 pl}(30){el}]296 0 idx03 -[(circular references)][{/37 pl}(37){el}]285 0 idx03 -[2(CLASS)][{/76 pl}(76){el}]274 0 idx03 -[2(%clear)][{/60 pl}(60){el}]263 0 idx03 -[2(coff)][{/85 pl}(85){el}]252 0 idx03 -[(colon)][{/26 pl}(26){el}]241 0 idx03 -[2(.COM)][{/73 pl}(73,){el}( ){/93 pl}(93){el}]230 0 idx03 -[(command-line)][{/17 pl}(17,){el}( ){/73 pl}(73){el}]219 0 idx03 -[(commas in macro parameters)][{/45 pl}(45){el}]208 0 idx03 -[2(.comment)][{/85 pl}(85){el}]197 0 idx03 -[2(COMMON)][{/70 pl}(70,){el}( ){/75 pl}(75){el}]186 0 idx03 -[2(elf)0( extensions to)][{/87 pl}(87){el}]175 0 idx13 -[2(obj)0( extensions to)][{/78 pl}(78){el}]164 0 idx13 -[(Common Object File Format)][{/85 pl}(85){el}]153 0 idx03 -[(common variables)][{/70 pl}(70){el}]142 0 idx03 -[(alignment in )2(elf)][{/87 pl}(87){el}]131 0 idx13 -[(element size)][{/78 pl}(78){el}]120 0 idx13 -[2(comp.lang.asm.x86)][{/14 pl}(14,){el}( ){/15 pl}(15){el}]109 0 idx03 -[2(comp.os.msdos.programmer)][{/94 pl}(94){el}]681 1 idx03 -[(concatenating macro parameters)][{/46 pl}(46){el}]670 1 idx03 -[(concatenating strings)][{/41 pl}(41){el}]659 1 idx03 -[(condition codes as macro parameters)][{/47 pl}(47){el}]648 1 idx03 -[(conditional assembly)][{/49 pl}(49){el}]637 1 idx03 -[(conditional jumps)][{/116 pl}(116){el}]626 1 idx03 -[(conditional-return macro)][{/47 pl}(47){el}]615 1 idx03 -[2(configure)][{/16 pl}(16){el}]604 1 idx03 -[(constants)][{/29 pl}(29){el}]593 1 idx03 -[(context stack)][{/54 pl}(54,){el}( ){/55 pl}(55){el}]582 1 idx03 -[(context-local labels)][{/54 pl}(54){el}]571 1 idx03 -[(context-local single-line macros)][{/55 pl}(55){el}]560 1 idx03 -[(counting macro parameters)][{/45 pl}(45){el}]549 1 idx03 -[2(CPU)][{/71 pl}(71){el}]538 1 idx03 -[2(CPUID)][{/31 pl}(31){el}]527 1 idx03 -[(creating contexts)][{/54 pl}(54){el}]516 1 idx03 -[(critical expression)][{/27 pl}(27,){el}( ){/35 pl}(35,){el}( ){/40 pl}(40,){el}( ){/69 pl}(69){el}]505 1 idx03 -[2(-D)0( option)][{/20 pl}(20){el}]494 1 idx03 -[2(-d)0( option)][{/20 pl}(20){el}]483 1 idx03 -[(daily development snapshots)][{/15 pl}(15){el}]472 1 idx03 -[2(.data)][{/85 pl}(85,){el}( ){/88 pl}(88,){el}( ){/89 pl}(89){el}]461 1 idx03 -[2(_DATA)][{/96 pl}(96){el}]450 1 idx03 -[2(data)][{/87 pl}(87,){el}( ){/89 pl}(89){el}]439 1 idx03 -[(data structure)][{/98 pl}(98,){el}( ){/105 pl}(105){el}]428 1 idx03 -[2(__DATE__)][{/61 pl}(61){el}]417 1 idx03 -[2(__DATE_NUM__)][{/61 pl}(61){el}]406 1 idx03 -[2(DB)][{/27 pl}(27,){el}( ){/31 pl}(31){el}]395 1 idx03 -[2(dbg)][{/90 pl}(90){el}]384 1 idx03 -[2(DD)][{/27 pl}(27,){el}( ){/31 pl}(31){el}]373 1 idx03 -[(debug information)][{/19 pl}(19){el}]362 1 idx03 -[(debug information format)][{/19 pl}(19){el}]351 1 idx03 -[(declaring structures)][{/62 pl}(62){el}]340 1 idx03 -[2(DEFAULT)][{/68 pl}(68){el}]329 1 idx03 -[2(default)][{/87 pl}(87){el}]318 1 idx03 -[(default macro parameters)][{/45 pl}(45){el}]307 1 idx03 -[(default name)][{/73 pl}(73){el}]296 1 idx03 -[(default-)2(WRT)0( mechanism)][{/78 pl}(78){el}]285 1 idx03 -[2(%define)][{/20 pl}(20,){el}( ){/37 pl}(37){el}]274 1 idx03 -[(defining sections)][{/68 pl}(68){el}]263 1 idx03 -[2(%defstr)][{/41 pl}(41){el}]252 1 idx03 -[2(%deftok)][{/41 pl}(41){el}]241 1 idx03 -[2(%depend)][{/53 pl}(53){el}]230 1 idx03 -[(design goals)][{/24 pl}(24){el}]219 1 idx03 -[(DevPac)][{/27 pl}(27,){el}( ){/35 pl}(35){el}]208 1 idx03 -[(disabling listing expansion)][{/48 pl}(48){el}]197 1 idx03 -[(division)][{/33 pl}(33){el}]186 1 idx03 -[(DJGPP)][{/85 pl}(85,){el}( ){/103 pl}(103){el}]175 1 idx03 -[2(djlink)][{/91 pl}(91){el}]164 1 idx03 -[(DLL symbols)][]153 1 idx03 -[(exporting)][{/77 pl}(77){el}]142 1 idx13 -[(importing)][{/77 pl}(77){el}]131 1 idx13 -[2(DO)][{/27 pl}(27,){el}( ){/31 pl}(31){el}]120 1 idx03 -[(DOS)][{/15 pl}(15,){el}( ){/20 pl}(20){el}]109 1 idx03 -(207)pageodd -restore showpage -%%Page: 208 208 -%%BeginPageSetup -save -%%EndPageSetup -/208 pa -[(DOS archive)][]681 0 idx03 -[(DOS source archive)][{/15 pl}(15){el}]670 0 idx03 -[2(DQ)][{/27 pl}(27,){el}( ){/31 pl}(31){el}]659 0 idx03 -[2(.drectve)][{/79 pl}(79){el}]648 0 idx03 -[2(DT)][{/27 pl}(27,){el}( ){/31 pl}(31){el}]637 0 idx03 -[2(DUP)][{/25 pl}(25,){el}( ){/28 pl}(28){el}]626 0 idx03 -[2(DW)][{/27 pl}(27,){el}( ){/31 pl}(31){el}]615 0 idx03 -[2(DWORD)][{/27 pl}(27){el}]604 0 idx03 -[2(DY)][{/27 pl}(27,){el}( ){/31 pl}(31){el}]593 0 idx03 -[2(-E)0( option)][{/21 pl}(21){el}]582 0 idx03 -[2(-e)0( option)][{/21 pl}(21,){el}( ){/121 pl}(121){el}]571 0 idx03 -[(effective addresses)][{/26 pl}(26,){el}( ){/28 pl}(28){el}]560 0 idx03 -[(element size, in common variables)][{/78 pl}(78){el}]549 0 idx03 -[(ELF)][{/85 pl}(85){el}]538 0 idx03 -[(shared libraries)][{/86 pl}(86){el}]527 0 idx13 -[(16-bit code and)][{/88 pl}(88){el}]516 0 idx13 -[(elf, debug formats and)][{/88 pl}(88){el}]505 0 idx03 -[2(elf32)][{/85 pl}(85){el}]494 0 idx03 -[2(elf64)][{/85 pl}(85){el}]483 0 idx03 -[2(%elif)][{/49 pl}(49,){el}( ){/50 pl}(50){el}]472 0 idx03 -[2(%elifctx)][{/50 pl}(50){el}]461 0 idx03 -[2(%elifdef)][{/49 pl}(49){el}]450 0 idx03 -[2(%elifempty)][{/52 pl}(52){el}]439 0 idx03 -[2(%elifid)][{/51 pl}(51){el}]428 0 idx03 -[2(%elifidn)][{/51 pl}(51){el}]417 0 idx03 -[2(%elifidni)][{/51 pl}(51){el}]406 0 idx03 -[2(%elifmacro)][{/50 pl}(50){el}]395 0 idx03 -[2(%elifn)][{/49 pl}(49,){el}( ){/50 pl}(50){el}]384 0 idx03 -[2(%elifnctx)][{/50 pl}(50){el}]373 0 idx03 -[2(%elifndef)][{/49 pl}(49){el}]362 0 idx03 -[2(%elifnempty)][{/52 pl}(52){el}]351 0 idx03 -[2(%elifnid)][{/51 pl}(51){el}]340 0 idx03 -[2(%elifnidn)][{/51 pl}(51){el}]329 0 idx03 -[2(%elifnidni)][{/51 pl}(51){el}]318 0 idx03 -[2(%elifnmacro)][{/50 pl}(50){el}]307 0 idx03 -[2(%elifnnum)][{/51 pl}(51){el}]296 0 idx03 -[2(%elifnstr)][{/51 pl}(51){el}]285 0 idx03 -[2(%elifntoken)][{/52 pl}(52){el}]274 0 idx03 -[2(%elifnum)][{/51 pl}(51){el}]263 0 idx03 -[2(%elifstr)][{/51 pl}(51){el}]252 0 idx03 -[2(%eliftoken)][{/52 pl}(52){el}]241 0 idx03 -[2(%else)][{/49 pl}(49){el}]230 0 idx03 -[2(endproc)][{/98 pl}(98,){el}( ){/105 pl}(105){el}]219 0 idx03 -[2(%endrep)][{/52 pl}(52){el}]208 0 idx03 -[2(ENDSTRUC)][{/62 pl}(62,){el}( ){/69 pl}(69){el}]197 0 idx03 -[(environment)][{/23 pl}(23){el}]186 0 idx03 -[2(EQU)][{/27 pl}(27,){el}( ){/28 pl}(28){el}]175 0 idx03 -[2(%error)][{/58 pl}(58){el}]164 0 idx03 -[2(error)][{/23 pl}(23){el}]153 0 idx03 -[(error messages)][{/20 pl}(20){el}]142 0 idx03 -[(error reporting format)][{/19 pl}(19){el}]131 0 idx03 -[(escape sequences)][{/30 pl}(30){el}]120 0 idx03 -[2(EVEN)][{/64 pl}(64){el}]109 0 idx03 -[(exact matches)][{/48 pl}(48){el}]681 1 idx03 -[2(.EXE)][{/75 pl}(75,){el}( ){/91 pl}(91){el}]670 1 idx03 -[2(EXE2BIN)][{/93 pl}(93){el}]659 1 idx03 -[2(EXE_begin)][{/92 pl}(92){el}]648 1 idx03 -[2(exebin.mac)][{/92 pl}(92){el}]637 1 idx03 -[2(exec)][{/85 pl}(85){el}]626 1 idx03 -[(Executable and Linkable Format)][{/85 pl}(85){el}]615 1 idx03 -[2(EXE_end)][{/92 pl}(92){el}]604 1 idx03 -[2(EXE_stack)][{/92 pl}(92){el}]593 1 idx03 -[2(%exitmacro)][{/48 pl}(48){el}]582 1 idx03 -[2(%exitrep)][{/52 pl}(52){el}]571 1 idx03 -[2(EXPORT)][{/77 pl}(77){el}]560 1 idx03 -[2(export)][{/89 pl}(89){el}]549 1 idx03 -[(exporting symbols)][{/70 pl}(70){el}]538 1 idx03 -[(expressions)][{/21 pl}(21,){el}( ){/33 pl}(33){el}]527 1 idx03 -[(extension)][{/17 pl}(17,){el}( ){/73 pl}(73){el}]516 1 idx03 -[2(EXTERN)][{/70 pl}(70){el}]505 1 idx03 -[2(obj)0( extensions to)][{/78 pl}(78){el}]494 1 idx13 -[2(rdf)0( extensions to)][{/89 pl}(89){el}]483 1 idx13 -[(extracting substrings)][{/42 pl}(42){el}]472 1 idx03 -[2(-F)0( option)][{/19 pl}(19){el}]461 1 idx03 -[2(-f)0( option)][{/18 pl}(18,){el}( ){/73 pl}(73){el}]450 1 idx03 -[(far call)][{/25 pl}(25){el}]439 1 idx03 -[(far common variables)][{/78 pl}(78){el}]428 1 idx03 -[(far pointer)][{/34 pl}(34){el}]417 1 idx03 -[2(FARCODE)][{/99 pl}(99,){el}( ){/101 pl}(101){el}]406 1 idx03 -[2(%fatal)][{/58 pl}(58){el}]395 1 idx03 -[2(__FILE__)][{/61 pl}(61){el}]384 1 idx03 -[2(FLAT)][{/76 pl}(76){el}]373 1 idx03 -[(flat memory model)][{/103 pl}(103){el}]362 1 idx03 -[(flat-form binary)][{/73 pl}(73){el}]351 1 idx03 -[2(FLOAT)][{/71 pl}(71){el}]340 1 idx03 -[2(__FLOAT__)][{/72 pl}(72){el}]329 1 idx03 -[2(__float128h__)][{/31 pl}(31){el}]318 1 idx03 -[2(__float128l__)][{/31 pl}(31){el}]307 1 idx03 -[2(__float16__)][{/31 pl}(31){el}]296 1 idx03 -[2(__float32__)][{/31 pl}(31){el}]285 1 idx03 -[2(__float64__)][{/31 pl}(31){el}]274 1 idx03 -[2(__float8__)][{/31 pl}(31){el}]263 1 idx03 -[2(__float80e__)][{/31 pl}(31){el}]252 1 idx03 -[2(__float80m__)][{/31 pl}(31){el}]241 1 idx03 -[2(__FLOAT_DAZ__)][{/72 pl}(72){el}]230 1 idx03 -[2(float-denorm)][{/22 pl}(22){el}]219 1 idx03 -[(floating-point)][]208 1 idx03 -[(constants)][{/31 pl}(31,){el}( ){/71 pl}(71){el}]197 1 idx13 -[(packed BCD constants)][{/33 pl}(33){el}]186 1 idx13 -[(floating-point)][{/25 pl}(25,){el}( ){/26 pl}(26,){el}( ){/27 pl}(27,){el}( ){/31 pl}(31){el}]175 1 idx03 -[2(float-overflow)][{/22 pl}(22){el}]164 1 idx03 -[2(__FLOAT_ROUND__)][{/72 pl}(72){el}]153 1 idx03 -[2(float-toolong)][{/23 pl}(23){el}]142 1 idx03 -[2(float-underflow)][{/23 pl}(23){el}]131 1 idx03 -[2(follows=)][{/74 pl}(74){el}]120 1 idx03 -[(format-specific directives)][{/67 pl}(67){el}]109 1 idx03 -(208)pageeven -restore showpage -%%Page: 209 209 -%%BeginPageSetup -save -%%EndPageSetup -/209 pa -[(frame pointer)][{/96 pl}(96,){el}( ){/100 pl}(100,){el}( ){/103 pl}(103){el}]681 0 idx03 -[(FreeBSD)][{/88 pl}(88,){el}( ){/106 pl}(106){el}]670 0 idx03 -[(FreeLink)][{/91 pl}(91){el}]659 0 idx03 -[2(ftp.simtel.net)][{/91 pl}(91){el}]648 0 idx03 -[2(function)][{/87 pl}(87,){el}( ){/89 pl}(89){el}]637 0 idx03 -[(functions)][]626 0 idx03 -[(C calling convention)][{/96 pl}(96,){el}( ){/103 pl}(103){el}]615 0 idx13 -[(Pascal calling convention)][{/100 pl}(100){el}]604 0 idx13 -[2(-g)0( option)][{/19 pl}(19){el}]593 0 idx03 -[2(gas)][{/14 pl}(14){el}]582 0 idx03 -[2(gcc)][{/14 pl}(14){el}]571 0 idx03 -[2(GLOBAL)][{/70 pl}(70){el}]560 0 idx03 -[2(aoutb)0( extensions to)][{/87 pl}(87){el}]549 0 idx13 -[2(elf)0( extensions to)][{/87 pl}(87){el}]538 0 idx13 -[2(rdf)0( extensions to)][{/89 pl}(89){el}]527 0 idx13 -[(global offset table)][{/106 pl}(106){el}]516 0 idx03 -[2(_GLOBAL_OFFSET_TABLE_)][{/86 pl}(86){el}]505 0 idx03 -[2(gnu-elf-extensions)][{/22 pl}(22){el}]494 0 idx03 -[2(..got)][{/86 pl}(86){el}]483 0 idx03 -[2(GOT)0( relocations)][{/107 pl}(107){el}]472 0 idx03 -[(GOT)][{/86 pl}(86,){el}( ){/106 pl}(106){el}]461 0 idx03 -[2(..gotoff)][{/86 pl}(86){el}]450 0 idx03 -[2(GOTOFF)0( relocations)][{/107 pl}(107){el}]439 0 idx03 -[2(..gotpc)][{/86 pl}(86){el}]428 0 idx03 -[2(GOTPC)0( relocations)][{/107 pl}(107){el}]417 0 idx03 -[2(..gottpoff)][{/87 pl}(87){el}]406 0 idx03 -[(graphics)][{/27 pl}(27){el}]395 0 idx03 -[(greedy macro parameters)][{/44 pl}(44){el}]384 0 idx03 -[2(GROUP)][{/76 pl}(76){el}]373 0 idx03 -[(groups)][{/34 pl}(34){el}]362 0 idx03 -[2(-h)][{/119 pl}(119){el}]351 0 idx03 -[(hexadecimal)][{/29 pl}(29){el}]340 0 idx03 -[2(hidden)][{/87 pl}(87){el}]329 0 idx03 -[(hybrid syntaxes)][{/24 pl}(24){el}]318 0 idx03 -[2(-I)0( option)][{/20 pl}(20){el}]307 0 idx03 -[2(-i)0( option)][{/20 pl}(20,){el}( ){/120 pl}(120){el}]296 0 idx03 -[2(%iassign)][{/40 pl}(40){el}]285 0 idx03 -[2(%idefine)][{/37 pl}(37){el}]274 0 idx03 -[2(%idefstr)][{/41 pl}(41){el}]263 0 idx03 -[2(%ideftok)][{/41 pl}(41){el}]252 0 idx03 -[2(IEND)][{/63 pl}(63){el}]241 0 idx03 -[2(%if)][{/49 pl}(49,){el}( ){/50 pl}(50){el}]230 0 idx03 -[2(%ifctx)][{/50 pl}(50,){el}( ){/55 pl}(55){el}]219 0 idx03 -[2(%ifdef)][{/49 pl}(49){el}]208 0 idx03 -[2(%ifempty)][{/52 pl}(52){el}]197 0 idx03 -[2(%ifid)][{/51 pl}(51){el}]186 0 idx03 -[2(%ifidn)][{/50 pl}(50){el}]175 0 idx03 -[2(%ifidni)][{/50 pl}(50){el}]164 0 idx03 -[2(%ifmacro)][{/49 pl}(49){el}]153 0 idx03 -[2(%ifn)][{/49 pl}(49,){el}( ){/50 pl}(50){el}]142 0 idx03 -[2(%ifnctx)][{/50 pl}(50){el}]131 0 idx03 -[2(%ifndef)][{/49 pl}(49){el}]120 0 idx03 -[2(%ifnempty)][{/52 pl}(52){el}]109 0 idx03 -[2(%ifnid)][{/51 pl}(51){el}]681 1 idx03 -[2(%ifnidn)][{/51 pl}(51){el}]670 1 idx03 -[2(%ifnidni)][{/51 pl}(51){el}]659 1 idx03 -[2(%ifnmacro)][{/50 pl}(50){el}]648 1 idx03 -[2(%ifnnum)][{/51 pl}(51){el}]637 1 idx03 -[2(%ifnstr)][{/51 pl}(51){el}]626 1 idx03 -[2(%ifntoken)][{/52 pl}(52){el}]615 1 idx03 -[2(%ifnum)][{/51 pl}(51){el}]604 1 idx03 -[2(%ifstr)][{/51 pl}(51){el}]593 1 idx03 -[2(%iftoken)][{/52 pl}(52){el}]582 1 idx03 -[2(%imacro)][{/42 pl}(42){el}]571 1 idx03 -[2(IMPORT)][{/77 pl}(77){el}]560 1 idx03 -[(import library)][{/77 pl}(77){el}]549 1 idx03 -[(importing symbols)][{/70 pl}(70){el}]538 1 idx03 -[2(INCBIN)][{/27 pl}(27,){el}( ){/31 pl}(31){el}]527 1 idx03 -[2(%include)][{/20 pl}(20,){el}( ){/53 pl}(53){el}]516 1 idx03 -[(include search path)][{/20 pl}(20){el}]505 1 idx03 -[(including other files)][{/53 pl}(53){el}]494 1 idx03 -[(inefficient code)][{/116 pl}(116){el}]483 1 idx03 -[(infinite loop)][{/33 pl}(33){el}]472 1 idx03 -[2(__Infinity__)][{/32 pl}(32){el}]461 1 idx03 -[(infinity)][{/32 pl}(32){el}]450 1 idx03 -[(informational section)][{/79 pl}(79){el}]439 1 idx03 -[2(INSTALL)][{/16 pl}(16){el}]428 1 idx03 -[(installing)][{/15 pl}(15){el}]417 1 idx03 -[(instances of structures)][{/63 pl}(63){el}]406 1 idx03 -[(instruction list)][{/122 pl}(122){el}]395 1 idx03 -[(intel hex)][{/74 pl}(74){el}]384 1 idx03 -[(Intel number formats)][{/32 pl}(32){el}]373 1 idx03 -[2(internal)][{/87 pl}(87){el}]362 1 idx03 -[2(%irmacro)][{/43 pl}(43){el}]351 1 idx03 -[2(ISTRUC)][{/63 pl}(63){el}]340 1 idx03 -[(iterating over macro parameters)][{/46 pl}(46){el}]329 1 idx03 -[2(ith)][{/74 pl}(74){el}]318 1 idx03 -[2(%ixdefine)][{/38 pl}(38){el}]307 1 idx03 -[2(Jcc NEAR)][{/116 pl}(116){el}]296 1 idx03 -[2(JMP DWORD)][{/110 pl}(110){el}]285 1 idx03 -[(jumps, mixed-size)][{/110 pl}(110){el}]274 1 idx03 -[2(-k)][{/121 pl}(121){el}]263 1 idx03 -[2(-l)0( option)][{/18 pl}(18){el}]252 1 idx03 -[(label prefix)][{/36 pl}(36){el}]241 1 idx03 -[2(.lbss)][{/85 pl}(85){el}]230 1 idx03 -[2(ld86)][{/88 pl}(88){el}]219 1 idx03 -[2(.ldata)][{/85 pl}(85){el}]208 1 idx03 -[2(LIBRARY)][{/89 pl}(89){el}]197 1 idx03 -[(license)][{/14 pl}(14){el}]186 1 idx03 -[2(%line)][{/59 pl}(59){el}]175 1 idx03 -[2(__LINE__)][{/61 pl}(61){el}]164 1 idx03 -[(linker, free)][{/91 pl}(91){el}]153 1 idx03 -[(Linux)][]142 1 idx03 -[2(a.out)][{/88 pl}(88){el}]131 1 idx13 -[2(as86)][{/88 pl}(88){el}]120 1 idx13 -[(ELF)][{/85 pl}(85){el}]109 1 idx13 -(209)pageodd -restore showpage -%%Page: 210 210 -%%BeginPageSetup -save -%%EndPageSetup -/210 pa -[(listing file)][{/18 pl}(18){el}]681 0 idx03 -[(little-endian)][{/31 pl}(31){el}]670 0 idx03 -[2(%local)][{/58 pl}(58){el}]659 0 idx03 -[(local labels)][{/35 pl}(35){el}]648 0 idx03 -[(logical AND)][{/50 pl}(50){el}]637 0 idx03 -[(logical negation)][{/34 pl}(34){el}]626 0 idx03 -[(logical OR)][{/50 pl}(50){el}]615 0 idx03 -[(logical XOR)][{/50 pl}(50){el}]604 0 idx03 -[2(.lrodata)][{/85 pl}(85){el}]593 0 idx03 -[2(-M)0( option)][{/18 pl}(18){el}]582 0 idx03 -[(Mach, object file format)][{/85 pl}(85){el}]571 0 idx03 -[(Mach-O)][{/85 pl}(85){el}]560 0 idx03 -[2(macho)][{/85 pl}(85){el}]549 0 idx03 -[2(macho32)][{/85 pl}(85){el}]538 0 idx03 -[2(macho64)][{/85 pl}(85){el}]527 0 idx03 -[(MacOS X)][{/85 pl}(85){el}]516 0 idx03 -[2(%macro)][{/42 pl}(42){el}]505 0 idx03 -[(macro indirection)][{/39 pl}(39){el}]494 0 idx03 -[(macro library)][{/20 pl}(20){el}]483 0 idx03 -[(macro processor)][{/37 pl}(37){el}]472 0 idx03 -[2(macro-defaults)][{/22 pl}(22){el}]461 0 idx03 -[(macro-local labels)][{/43 pl}(43){el}]450 0 idx03 -[2(macro-params)][{/22 pl}(22){el}]439 0 idx03 -[(macros)][{/28 pl}(28){el}]428 0 idx03 -[2(macro-selfref)][{/22 pl}(22){el}]417 0 idx03 -[2(make)][{/16 pl}(16){el}]406 0 idx03 -[(makefile dependencies)][{/18 pl}(18){el}]395 0 idx03 -[(makefiles)][{/15 pl}(15,){el}( ){/16 pl}(16){el}]384 0 idx03 -[(man pages)][{/16 pl}(16){el}]373 0 idx03 -[(map files)][{/74 pl}(74){el}]362 0 idx03 -[2(MASM)][{/14 pl}(14){el}]351 0 idx03 -[(MASM)][{/24 pl}(24,){el}( ){/28 pl}(28,){el}( ){/75 pl}(75){el}]340 0 idx03 -[2(-MD)0( option)][{/18 pl}(18){el}]329 0 idx03 -[(memory models)][{/25 pl}(25,){el}( ){/95 pl}(95){el}]318 0 idx03 -[(memory operand)][{/27 pl}(27){el}]307 0 idx03 -[(memory references)][{/24 pl}(24,){el}( ){/28 pl}(28){el}]296 0 idx03 -[2(-MF)0( option)][{/18 pl}(18){el}]285 0 idx03 -[2(-MG)0( option)][{/18 pl}(18){el}]274 0 idx03 -[(Microsoft OMF)][{/75 pl}(75){el}]263 0 idx03 -[(minifloat)][{/32 pl}(32){el}]252 0 idx03 -[(Minix)][{/88 pl}(88){el}]241 0 idx03 -[2(misc)0( subdirectory)][{/92 pl}(92,){el}( ){/98 pl}(98,){el}( ){/105 pl}(105){el}]230 0 idx03 -[(mixed-language program)][{/94 pl}(94){el}]219 0 idx03 -[(mixed-size addressing)][{/110 pl}(110){el}]208 0 idx03 -[(mixed-size instruction)][{/110 pl}(110){el}]197 0 idx03 -[(MMX registers)][]186 0 idx03 -[(ModR/M byte)][]175 0 idx03 -[2(MODULE)][{/89 pl}(89){el}]164 0 idx03 -[(modulo operators)][{/33 pl}(33){el}]153 0 idx03 -[(motorola s-records)][{/74 pl}(74){el}]142 0 idx03 -[2(-MP)0( option)][{/19 pl}(19){el}]131 0 idx03 -[2(-MQ)0( option)][{/19 pl}(19){el}]120 0 idx03 -[(MS-DOS)][{/73 pl}(73){el}]109 0 idx03 -[(MS-DOS device drivers)][{/94 pl}(94){el}]681 1 idx03 -[2(-MT)0( option)][{/19 pl}(19){el}]670 1 idx03 -[(multi-line macros)][{/22 pl}(22,){el}( ){/42 pl}(42){el}]659 1 idx03 -[(multipass optimization)][{/21 pl}(21){el}]648 1 idx03 -[(multiple section names)][{/73 pl}(73){el}]637 1 idx03 -[(multiplication)][{/33 pl}(33){el}]626 1 idx03 -[2(multipush)0( macro)][{/46 pl}(46){el}]615 1 idx03 -[(multisection)][{/74 pl}(74){el}]604 1 idx03 -[2(__NaN__)][{/32 pl}(32){el}]593 1 idx03 -[(NaN)][{/32 pl}(32){el}]582 1 idx03 -[(NASM version)][{/60 pl}(60){el}]571 1 idx03 -[(nasm version history)][{/181 pl}(181){el}]560 1 idx03 -[(nasm version id)][{/60 pl}(60){el}]549 1 idx03 -[(nasm version string)][{/60 pl}(60){el}]538 1 idx03 -[2(nasm.1)][{/16 pl}(16){el}]527 1 idx03 -[2(__NASMDEFSEG)][{/75 pl}(75){el}]516 1 idx03 -[2(nasm-devel)][{/15 pl}(15){el}]505 1 idx03 -[2(NASMENV)][{/23 pl}(23){el}]494 1 idx03 -[2(nasm.exe)][{/15 pl}(15){el}]483 1 idx03 -[2(nasm -hf)][{/18 pl}(18){el}]472 1 idx03 -[2(__NASM_MAJOR__)][{/60 pl}(60){el}]461 1 idx03 -[2(__NASM_MINOR__)][{/60 pl}(60){el}]450 1 idx03 -[2(nasm.out)][{/18 pl}(18){el}]439 1 idx03 -[2(___NASM_PATCHLEVEL__)][{/60 pl}(60){el}]428 1 idx03 -[2(__NASM_SNAPSHOT__)][{/60 pl}(60){el}]417 1 idx03 -[2(__NASM_SUBMINOR__)][{/60 pl}(60){el}]406 1 idx03 -[2(__NASM_VER__)][{/60 pl}(60){el}]395 1 idx03 -[2(__NASM_VERSION_ID__)][{/60 pl}(60){el}]384 1 idx03 -[2(nasm-XXX-dos.zip)][{/15 pl}(15){el}]373 1 idx03 -[2(nasm-XXX.tar.gz)][{/16 pl}(16){el}]362 1 idx03 -[2(nasm-XXX-win32.zip)][{/15 pl}(15){el}]351 1 idx03 -[2(nasm-XXX.zip)][{/15 pl}(15){el}]340 1 idx03 -[(ndisasm)][{/119 pl}(119){el}]329 1 idx03 -[2(ndisasm.1)][{/16 pl}(16){el}]318 1 idx03 -[2(ndisasm.exe)][{/15 pl}(15){el}]307 1 idx03 -[(near call)][{/25 pl}(25){el}]296 1 idx03 -[(near common variables)][{/78 pl}(78){el}]285 1 idx03 -[(NetBSD)][{/88 pl}(88,){el}( ){/106 pl}(106){el}]274 1 idx03 -[(new releases)][{/15 pl}(15){el}]263 1 idx03 -[2(noalloc)][{/85 pl}(85){el}]252 1 idx03 -[2(nobits)][{/74 pl}(74,){el}( ){/85 pl}(85){el}]241 1 idx03 -[2(noexec)][{/85 pl}(85){el}]230 1 idx03 -[2(.nolist)][{/48 pl}(48){el}]219 1 idx03 -[(`nowait')][{/25 pl}(25){el}]208 1 idx03 -[2(nowrite)][{/85 pl}(85){el}]197 1 idx03 -[2(number-overflow)][{/22 pl}(22){el}]186 1 idx03 -[(numeric constants)][{/27 pl}(27,){el}( ){/29 pl}(29){el}]175 1 idx03 -[2(-O)0( option)][{/21 pl}(21){el}]164 1 idx03 -[2(-o)0( option)][{/17 pl}(17,){el}( ){/119 pl}(119){el}]153 1 idx03 -[2(O16)][{/26 pl}(26){el}]142 1 idx03 -[2(o16)][{/111 pl}(111){el}]131 1 idx03 -[2(O32)][{/26 pl}(26){el}]120 1 idx03 -[2(o32)][{/112 pl}(112){el}]109 1 idx03 -(210)pageeven -restore showpage -%%Page: 211 211 -%%BeginPageSetup -save -%%EndPageSetup -/211 pa -[2(O64)][{/26 pl}(26){el}]681 0 idx03 -[2(.OBJ)][{/91 pl}(91){el}]670 0 idx03 -[2(obj)][{/75 pl}(75){el}]659 0 idx03 -[2(object)][{/87 pl}(87,){el}( ){/89 pl}(89){el}]648 0 idx03 -[(octal)][{/29 pl}(29){el}]637 0 idx03 -[2(OF_DBG)][{/90 pl}(90){el}]626 0 idx03 -[2(OF_DEFAULT)][{/18 pl}(18){el}]615 0 idx03 -[2(OFFSET)][{/24 pl}(24){el}]604 0 idx03 -[(OMF)][{/75 pl}(75){el}]593 0 idx03 -[(omitted parameters)][{/45 pl}(45){el}]582 0 idx03 -[(one's complement)][{/34 pl}(34){el}]571 0 idx03 -[(OpenBSD)][{/88 pl}(88,){el}( ){/106 pl}(106){el}]560 0 idx03 -[(operands)][{/26 pl}(26){el}]549 0 idx03 -[(operand-size prefixes)][{/26 pl}(26){el}]538 0 idx03 -[(operating system)][{/73 pl}(73){el}]527 0 idx03 -[(writing)][{/110 pl}(110){el}]516 0 idx13 -[(operators)][{/33 pl}(33){el}]505 0 idx03 -[2(ORG)][{/73 pl}(73,){el}( ){/93 pl}(93,){el}( ){/94 pl}(94,){el}( ){/116 pl}(116){el}]494 0 idx03 -[2(orphan-labels)][{/22 pl}(22,){el}( ){/26 pl}(26){el}]483 0 idx03 -[(OS/2)][{/75 pl}(75,){el}( ){/76 pl}(76){el}]472 0 idx03 -[2(osabi)][{/85 pl}(85){el}]461 0 idx03 -[(other preprocessor directives)][{/59 pl}(59){el}]450 0 idx03 -[(out of range, jumps)][{/116 pl}(116){el}]439 0 idx03 -[(output file format)][{/18 pl}(18){el}]428 0 idx03 -[(output formats)][{/73 pl}(73){el}]417 0 idx03 -[2(__OUTPUT_FORMAT__)][{/61 pl}(61){el}]406 0 idx03 -[(overlapping segments)][{/34 pl}(34){el}]395 0 idx03 -[2(OVERLAY)][{/76 pl}(76){el}]384 0 idx03 -[(overloading)][]373 0 idx03 -[(multi-line macros)][{/43 pl}(43){el}]362 0 idx13 -[(single-line macros)][{/38 pl}(38){el}]351 0 idx13 -[2(-P)0( option)][{/20 pl}(20){el}]340 0 idx03 -[2(-p)0( option)][{/20 pl}(20,){el}( ){/53 pl}(53){el}]329 0 idx03 -[(paradox)][{/35 pl}(35){el}]318 0 idx03 -[2(PASCAL)][{/101 pl}(101){el}]307 0 idx03 -[(Pascal calling convention)][{/100 pl}(100){el}]296 0 idx03 -[2(__PASS__)][{/62 pl}(62){el}]285 0 idx03 -[(passes, assembly)][]274 0 idx03 -[2(PATH)][{/15 pl}(15){el}]263 0 idx03 -[2(%pathsearch)][{/20 pl}(20,){el}( ){/53 pl}(53){el}]252 0 idx03 -[(period)][{/35 pl}(35){el}]241 0 idx03 -[(Perl)][{/15 pl}(15){el}]230 0 idx03 -[(perverse)][{/20 pl}(20){el}]219 0 idx03 -[(PharLap)][{/76 pl}(76){el}]208 0 idx03 -[(PIC)][{/86 pl}(86,){el}( ){/88 pl}(88,){el}( ){/106 pl}(106){el}]197 0 idx03 -[2(..plt)][{/86 pl}(86){el}]186 0 idx03 -[2(PLT)0( relocations)][{/86 pl}(86,){el}( ){/108 pl}(108,){el}( ){/109 pl}(109){el}]175 0 idx03 -[(plt relocations)][{/109 pl}(109){el}]164 0 idx03 -[2(%pop)][{/54 pl}(54){el}]153 0 idx03 -[(position-independent code)][{/86 pl}(86,){el}( ){/88 pl}(88,){el}( ){/106 pl}(106){el}]142 0 idx03 -[2(--postfix)][{/23 pl}(23){el}]131 0 idx03 -[(precedence)][{/33 pl}(33){el}]120 0 idx03 -[(pre-defining macros)][{/20 pl}(20,){el}( ){/38 pl}(38){el}]109 0 idx03 -[(preferred)][{/34 pl}(34){el}]681 1 idx03 -[2(--prefix)][{/23 pl}(23){el}]670 1 idx03 -[(pre-including files)][{/20 pl}(20){el}]659 1 idx03 -[(preprocess-only mode)][{/21 pl}(21){el}]648 1 idx03 -[(preprocessor)][{/21 pl}(21,){el}( ){/28 pl}(28,){el}( ){/33 pl}(33,){el}( ){/37 pl}(37){el}]637 1 idx03 -[(preprocessor expressions)][{/21 pl}(21){el}]626 1 idx03 -[(preprocessor loops)][{/52 pl}(52){el}]615 1 idx03 -[(preprocessor variables)][{/40 pl}(40){el}]604 1 idx03 -[(primitive directives)][{/67 pl}(67){el}]593 1 idx03 -[2(PRIVATE)][{/75 pl}(75){el}]582 1 idx03 -[2(proc)][{/89 pl}(89,){el}( ){/98 pl}(98,){el}( ){/105 pl}(105){el}]571 1 idx03 -[(procedure linkage table)][{/86 pl}(86,){el}( ){/108 pl}(108,){el}( ){/109 pl}(109){el}]560 1 idx03 -[(processor mode)][{/67 pl}(67){el}]549 1 idx03 -[2(progbits)][{/74 pl}(74,){el}( ){/85 pl}(85){el}]538 1 idx03 -[(program entry point)][{/78 pl}(78,){el}( ){/91 pl}(91){el}]527 1 idx03 -[(program origin)][{/73 pl}(73){el}]516 1 idx03 -[2(protected)][{/87 pl}(87){el}]505 1 idx03 -[(pseudo-instructions)][{/27 pl}(27){el}]494 1 idx03 -[2(PUBLIC)][{/70 pl}(70,){el}( ){/75 pl}(75){el}]483 1 idx03 -[(pure binary)][{/73 pl}(73){el}]472 1 idx03 -[2(%push)][{/54 pl}(54){el}]461 1 idx03 -[2(__QNaN__)][{/32 pl}(32){el}]450 1 idx03 -[(quick start)][{/24 pl}(24){el}]439 1 idx03 -[2(QWORD)][{/27 pl}(27){el}]428 1 idx03 -[2(-r)][{/119 pl}(119){el}]417 1 idx03 -[2(rdf)][{/88 pl}(88){el}]406 1 idx03 -[2(rdoff)0( subdirectory)][{/16 pl}(16,){el}( ){/88 pl}(88,){el}( ){/89 pl}(89){el}]395 1 idx03 -[(recursive multi-line macros)][{/43 pl}(43){el}]384 1 idx03 -[(redirecting errors)][{/20 pl}(20){el}]373 1 idx03 -[2(REL)][{/29 pl}(29,){el}( ){/68 pl}(68){el}]362 1 idx03 -[(relational operators)][{/50 pl}(50){el}]351 1 idx03 -[(release candidates)][{/15 pl}(15){el}]340 1 idx03 -[(Relocatable Dynamic Object File Format)][{/88 pl}(88){el}]329 1 idx03 -[(relocations, PIC-specific)][{/86 pl}(86){el}]318 1 idx03 -[(removing contexts)][{/54 pl}(54){el}]307 1 idx03 -[(renaming contexts)][{/55 pl}(55){el}]296 1 idx03 -[2(%rep)][{/28 pl}(28,){el}( ){/52 pl}(52){el}]285 1 idx03 -[(repeating)][{/28 pl}(28,){el}( ){/52 pl}(52){el}]274 1 idx03 -[2(%repl)][{/55 pl}(55){el}]263 1 idx03 -[(reporting bugs)][{/117 pl}(117){el}]252 1 idx03 -[2(RESB)][{/25 pl}(25,){el}( ){/27 pl}(27){el}]241 1 idx03 -[2(RESD)][{/27 pl}(27){el}]230 1 idx03 -[2(RESO)][{/27 pl}(27){el}]219 1 idx03 -[2(RESQ)][{/27 pl}(27){el}]208 1 idx03 -[2(REST)][{/27 pl}(27){el}]197 1 idx03 -[2(RESW)][{/27 pl}(27){el}]186 1 idx03 -[2(RESY)][{/27 pl}(27){el}]175 1 idx03 -[2(%rmacro)][{/43 pl}(43){el}]164 1 idx03 -[2(.rodata)][{/85 pl}(85){el}]153 1 idx03 -[2(%rotate)][{/46 pl}(46){el}]142 1 idx03 -[(rotating macro parameters)][{/46 pl}(46){el}]131 1 idx03 -[2(-s)0( option)][{/20 pl}(20,){el}( ){/120 pl}(120){el}]120 1 idx03 -[(searching for include files)][{/53 pl}(53){el}]109 1 idx03 -(211)pageodd -restore showpage -%%Page: 212 212 -%%BeginPageSetup -save -%%EndPageSetup -/212 pa -[2(__SECT__)][{/68 pl}(68,){el}( ){/69 pl}(69){el}]681 0 idx03 -[2(SECTION)][{/68 pl}(68){el}]670 0 idx03 -[2(elf)0( extensions to)][{/85 pl}(85){el}]659 0 idx13 -[2(win32)0( extensions to)][{/79 pl}(79){el}]648 0 idx13 -[(section alignment)][]637 0 idx03 -[(in )2(bin)][{/74 pl}(74){el}]626 0 idx13 -[(in )2(elf)][{/85 pl}(85){el}]615 0 idx13 -[(in )2(obj)][{/76 pl}(76){el}]604 0 idx13 -[(in )2(win32)][{/79 pl}(79){el}]593 0 idx13 -[(section, bin extensions to)][{/73 pl}(73){el}]582 0 idx03 -[2(SEG)][{/34 pl}(34,){el}( ){/75 pl}(75){el}]571 0 idx03 -[2(SEGMENT)][{/68 pl}(68){el}]560 0 idx03 -[2(elf)0( extensions to)][{/75 pl}(75){el}]549 0 idx13 -[(segment address)][{/34 pl}(34){el}]538 0 idx03 -[(segment alignment)][]527 0 idx03 -[(in )2(bin)][{/74 pl}(74){el}]516 0 idx13 -[(in )2(obj)][{/76 pl}(76){el}]505 0 idx13 -[(segment names, Borland Pascal)][{/101 pl}(101){el}]494 0 idx03 -[(segment override)][{/25 pl}(25,){el}( ){/26 pl}(26){el}]483 0 idx03 -[(segments)][{/34 pl}(34){el}]472 0 idx03 -[(groups of)][{/76 pl}(76){el}]461 0 idx13 -[(separator character)][{/23 pl}(23){el}]450 0 idx03 -[(shared libraries)][{/88 pl}(88,){el}( ){/106 pl}(106){el}]439 0 idx03 -[(shared library)][{/87 pl}(87){el}]428 0 idx03 -[2(shift)0( command)][{/46 pl}(46){el}]417 0 idx03 -[(SIB byte)][]406 0 idx03 -[(signed division)][{/33 pl}(33){el}]395 0 idx03 -[(signed modulo)][{/33 pl}(33){el}]384 0 idx03 -[(single-line macros)][{/37 pl}(37){el}]373 0 idx03 -[(size, of symbols)][{/87 pl}(87){el}]362 0 idx03 -[2(smartalign)][{/66 pl}(66){el}]351 0 idx03 -[2(__SNaN__)][{/32 pl}(32){el}]340 0 idx03 -[(snapshots, daily development)][{/15 pl}(15){el}]329 0 idx03 -[(Solaris x86)][{/85 pl}(85){el}]318 0 idx03 -[2(-soname)][{/109 pl}(109){el}]307 0 idx03 -[(sound)][{/27 pl}(27){el}]296 0 idx03 -[(source code)][{/15 pl}(15){el}]285 0 idx03 -[(source-listing file)][{/18 pl}(18){el}]274 0 idx03 -[(square brackets)][{/24 pl}(24,){el}( ){/28 pl}(28){el}]263 0 idx03 -[2(srec)][{/74 pl}(74){el}]252 0 idx03 -[2(STACK)][{/75 pl}(75){el}]241 0 idx03 -[(stack relative preprocessor directives)][{/57 pl}(57){el}]230 0 idx03 -[2(%stacksize)][{/57 pl}(57){el}]219 0 idx03 -[(standard macro packages)][{/66 pl}(66){el}]208 0 idx03 -[(standard macros)][{/60 pl}(60){el}]197 0 idx03 -[(standardized section names)][{/68 pl}(68,){el}( ){/79 pl}(79,){el}( ){/85 pl}(85,){el}( ){/88 pl}(88,){el}( ){/89 pl}(89){el}]186 0 idx03 -[2(..start)][{/78 pl}(78,){el}( ){/91 pl}(91){el}]175 0 idx03 -[2(start=)][{/74 pl}(74){el}]164 0 idx03 -[2(stderr)][{/20 pl}(20){el}]153 0 idx03 -[2(stdout)][{/20 pl}(20){el}]142 0 idx03 -[2(%strcat)][{/41 pl}(41){el}]131 0 idx03 -[2(STRICT)][{/34 pl}(34){el}]120 0 idx03 -[(string constant)][{/27 pl}(27){el}]109 0 idx03 -[(string constants)][{/31 pl}(31){el}]681 1 idx03 -[(string length)][{/41 pl}(41){el}]670 1 idx03 -[(string manipulation in macros)][{/41 pl}(41){el}]659 1 idx03 -[(strings)][{/30 pl}(30){el}]648 1 idx03 -[2(%strlen)][{/41 pl}(41){el}]637 1 idx03 -[2(STRUC)][{/62 pl}(62,){el}( ){/69 pl}(69,){el}( ){/98 pl}(98,){el}( ){/105 pl}(105){el}]626 1 idx03 -[(stub preprocessor)][{/21 pl}(21){el}]615 1 idx03 -[2(%substr)][{/42 pl}(42){el}]604 1 idx03 -[(subtraction)][{/33 pl}(33){el}]593 1 idx03 -[(suppressible warning)][{/22 pl}(22){el}]582 1 idx03 -[(suppressing preprocessing)][{/21 pl}(21){el}]571 1 idx03 -[(switching between sections)][{/68 pl}(68){el}]560 1 idx03 -[2(..sym)][{/86 pl}(86){el}]549 1 idx03 -[(symbol sizes, specifying)][{/87 pl}(87){el}]538 1 idx03 -[(symbol types, specifying)][{/87 pl}(87){el}]527 1 idx03 -[(symbols)][]516 1 idx03 -[(exporting from DLLs)][{/77 pl}(77){el}]505 1 idx13 -[(importing from DLLs)][{/77 pl}(77){el}]494 1 idx13 -[2(synchronisation)][{/120 pl}(120){el}]483 1 idx03 -[2(.SYS)][{/73 pl}(73,){el}( ){/94 pl}(94){el}]472 1 idx03 -[2(-t)][{/22 pl}(22){el}]461 1 idx03 -[2(TASM)][{/14 pl}(14,){el}( ){/22 pl}(22){el}]450 1 idx03 -[(tasm)][{/24 pl}(24,){el}( ){/75 pl}(75){el}]439 1 idx03 -[2(.tbss)][{/85 pl}(85){el}]428 1 idx03 -[2(TBYTE)][{/25 pl}(25){el}]417 1 idx03 -[2(.tdata)][{/85 pl}(85){el}]406 1 idx03 -[2(test)0( subdirectory)][{/91 pl}(91){el}]395 1 idx03 -[(testing)][]384 1 idx03 -[(arbitrary numeric expressions)][{/50 pl}(50){el}]373 1 idx13 -[(context stack)][{/50 pl}(50){el}]362 1 idx13 -[(exact text identity)][{/50 pl}(50){el}]351 1 idx13 -[(multi-line macro existence)][{/49 pl}(49){el}]340 1 idx13 -[(single-line macro existence)][{/49 pl}(49){el}]329 1 idx13 -[(token types)][{/51 pl}(51){el}]318 1 idx13 -[2(.text)][{/85 pl}(85,){el}( ){/88 pl}(88,){el}( ){/89 pl}(89){el}]307 1 idx03 -[2(_TEXT)][{/96 pl}(96){el}]296 1 idx03 -[(thread local storage)][{/87 pl}(87){el}]285 1 idx03 -[2(__TIME__)][{/61 pl}(61){el}]274 1 idx03 -[2(__TIME_NUM__)][{/61 pl}(61){el}]263 1 idx03 -[2(TIMES)][{/27 pl}(27,){el}( ){/28 pl}(28,){el}( ){/35 pl}(35,){el}( ){/116 pl}(116,){el}( ){/117 pl}(117){el}]252 1 idx03 -[2(TLINK)][{/93 pl}(93){el}]241 1 idx03 -[2(tls)][{/85 pl}(85,){el}( ){/87 pl}(87){el}]230 1 idx03 -[2(..tlsie)][{/87 pl}(87){el}]219 1 idx03 -[(trailing colon)][{/26 pl}(26){el}]208 1 idx03 -[2(TWORD)][{/25 pl}(25,){el}( ){/27 pl}(27){el}]197 1 idx03 -[(type, of symbols)][{/87 pl}(87){el}]186 1 idx03 -[2(-U)0( option)][{/21 pl}(21){el}]175 1 idx03 -[2(-u)0( option)][{/21 pl}(21,){el}( ){/119 pl}(119){el}]164 1 idx03 -[(unary operators)][{/34 pl}(34){el}]153 1 idx03 -[2(%undef)][{/21 pl}(21,){el}( ){/40 pl}(40){el}]142 1 idx03 -[(undefining macros)][{/21 pl}(21){el}]131 1 idx03 -[(underscore, in C symbols)][{/94 pl}(94){el}]120 1 idx03 -[(Unicode)][{/30 pl}(30,){el}( ){/31 pl}(31){el}]109 1 idx03 -(212)pageeven -restore showpage -%%Page: 213 213 -%%BeginPageSetup -save -%%EndPageSetup -/213 pa -[(uninitialized)][{/27 pl}(27){el}]681 0 idx03 -[(uninitialized storage)][{/25 pl}(25){el}]670 0 idx03 -[(Unix)][{/16 pl}(16){el}]659 0 idx03 -[(SCO)][{/85 pl}(85){el}]648 0 idx13 -[(source archive)][{/16 pl}(16){el}]637 0 idx13 -[(System V)][{/85 pl}(85){el}]626 0 idx13 -[(UnixWare)][{/85 pl}(85){el}]615 0 idx03 -[2(%unmacro)][{/48 pl}(48){el}]604 0 idx03 -[(unrolled loops)][{/28 pl}(28){el}]593 0 idx03 -[(unsigned division)][{/33 pl}(33){el}]582 0 idx03 -[(unsigned modulo)][{/33 pl}(33){el}]571 0 idx03 -[2(UPPERCASE)][{/24 pl}(24,){el}( ){/77 pl}(77){el}]560 0 idx03 -[2(%use)][{/54 pl}(54,){el}( ){/66 pl}(66){el}]549 0 idx03 -[2(__USE_*__)][{/62 pl}(62){el}]538 0 idx03 -[2(USE16)][{/68 pl}(68,){el}( ){/76 pl}(76){el}]527 0 idx03 -[2(USE32)][{/68 pl}(68,){el}( ){/76 pl}(76){el}]516 0 idx03 -[2(user)][{/23 pl}(23){el}]505 0 idx03 -[(user-defined errors)][{/58 pl}(58){el}]494 0 idx03 -[(user-level assembler directives)][{/60 pl}(60){el}]483 0 idx03 -[(user-level directives)][{/67 pl}(67){el}]472 0 idx03 -[2(__UTC_DATE__)][{/61 pl}(61){el}]461 0 idx03 -[2(__UTC_DATE_NUM__)][{/61 pl}(61){el}]450 0 idx03 -[2(__UTC_TIME__)][{/61 pl}(61){el}]439 0 idx03 -[2(__UTC_TIME_NUM__)][{/61 pl}(61){el}]428 0 idx03 -[(UTF-16)][{/31 pl}(31){el}]417 0 idx03 -[(UTF-32)][{/31 pl}(31){el}]406 0 idx03 -[(UTF-8)][{/30 pl}(30){el}]395 0 idx03 -[2(__utf16__)][{/31 pl}(31){el}]384 0 idx03 -[2(__utf32__)][{/31 pl}(31){el}]373 0 idx03 -[2(-v)0( option)][{/23 pl}(23){el}]362 0 idx03 -[(VAL)][{/91 pl}(91){el}]351 0 idx03 -[(valid characters)][{/26 pl}(26){el}]340 0 idx03 -[(variable types)][{/24 pl}(24){el}]329 0 idx03 -[(version)][{/23 pl}(23){el}]318 0 idx03 -[(version number of NASM)][{/60 pl}(60){el}]307 0 idx03 -[2(vfollows=)][{/74 pl}(74){el}]296 0 idx03 -[(Visual C++)][{/79 pl}(79){el}]285 0 idx03 -[2(vstart=)][{/74 pl}(74){el}]274 0 idx03 -[2(-W)0( option)][{/22 pl}(22){el}]263 0 idx03 -[2(-w)0( option)][{/22 pl}(22){el}]252 0 idx03 -[2(%warning)][{/58 pl}(58){el}]241 0 idx03 -[(warnings)][{/22 pl}(22){el}]230 0 idx03 -[2([warning *warning-name])][{/23 pl}(23){el}]219 0 idx03 -[2([warning +warning-name])][{/23 pl}(23){el}]208 0 idx03 -[2([warning -warning-name])][{/23 pl}(23){el}]197 0 idx03 -[(website)][{/15 pl}(15){el}]186 0 idx03 -[2(win64)][{/81 pl}(81,){el}( ){/113 pl}(113){el}]175 0 idx03 -[(Win64)][{/75 pl}(75,){el}( ){/79 pl}(79,){el}( ){/103 pl}(103){el}]164 0 idx03 -[(Windows)][{/91 pl}(91){el}]153 0 idx03 -[(Windows 95)][]142 0 idx03 -[(Windows NT)][]131 0 idx03 -[2(write)][{/85 pl}(85){el}]120 0 idx03 -[(writing operating systems)][{/110 pl}(110){el}]109 0 idx03 -[2(WRT)][{/34 pl}(34,){el}( ){/75 pl}(75,){el}( ){/86 pl}(86,){el}( ){/87 pl}(87,){el}( ){/88 pl}(88){el}]681 1 idx03 -[2(WRT ..got)][{/107 pl}(107){el}]670 1 idx03 -[2(WRT ..gotoff)][{/107 pl}(107){el}]659 1 idx03 -[2(WRT ..gotpc)][{/107 pl}(107){el}]648 1 idx03 -[2(WRT ..plt)][{/109 pl}(109){el}]637 1 idx03 -[2(WRT ..sym)][{/108 pl}(108){el}]626 1 idx03 -[(WWW page)][]615 1 idx03 -[2(www.cpan.org)][{/15 pl}(15){el}]604 1 idx03 -[2(www.delorie.com)][{/91 pl}(91){el}]593 1 idx03 -[2(www.pcorner.com)][{/91 pl}(91){el}]582 1 idx03 -[2(-X)0( option)][{/19 pl}(19){el}]571 1 idx03 -[2(x2ftp.oulu.fi)][{/91 pl}(91){el}]560 1 idx03 -[2(%xdefine)][{/38 pl}(38){el}]549 1 idx03 -[2(-y)0( option)][{/23 pl}(23){el}]538 1 idx03 -[2(-Z)0( option)][{/20 pl}(20){el}]527 1 idx03 -(213)pageodd -restore showpage -%%EOF diff --git a/doc/nasmdoc.txt b/doc/nasmdoc.txt deleted file mode 100644 index 5a7286b..0000000 --- a/doc/nasmdoc.txt +++ /dev/null @@ -1,12394 +0,0 @@ - The Netwide Assembler: NASM - =========================== - -Chapter 1: Introduction ------------------------ - - 1.1 What Is NASM? - - The Netwide Assembler, NASM, is an 80x86 and x86-64 assembler - designed for portability and modularity. It supports a range of - object file formats, including Linux and `*BSD' `a.out', `ELF', - `COFF', `Mach-O', Microsoft 16-bit `OBJ', `Win32' and `Win64'. It - will also output plain binary files. Its syntax is designed to be - simple and easy to understand, similar to Intel's but less complex. - It supports all currently known x86 architectural extensions, and - has strong support for macros. - - 1.1.1 Why Yet Another Assembler? - - The Netwide Assembler grew out of an idea on `comp.lang.asm.x86' (or - possibly `alt.lang.asm' - I forget which), which was essentially - that there didn't seem to be a good _free_ x86-series assembler - around, and that maybe someone ought to write one. - - (*) `a86' is good, but not free, and in particular you don't get any - 32-bit capability until you pay. It's DOS only, too. - - (*) `gas' is free, and ports over to DOS and Unix, but it's not very - good, since it's designed to be a back end to `gcc', which - always feeds it correct code. So its error checking is minimal. - Also, its syntax is horrible, from the point of view of anyone - trying to actually _write_ anything in it. Plus you can't write - 16-bit code in it (properly.) - - (*) `as86' is specific to Minix and Linux, and (my version at least) - doesn't seem to have much (or any) documentation. - - (*) `MASM' isn't very good, and it's (was) expensive, and it runs - only under DOS. - - (*) `TASM' is better, but still strives for MASM compatibility, - which means millions of directives and tons of red tape. And its - syntax is essentially MASM's, with the contradictions and quirks - that entails (although it sorts out some of those by means of - Ideal mode.) It's expensive too. And it's DOS-only. - - So here, for your coding pleasure, is NASM. At present it's still in - prototype stage - we don't promise that it can outperform any of - these assemblers. But please, _please_ send us bug reports, fixes, - helpful information, and anything else you can get your hands on - (and thanks to the many people who've done this already! You all - know who you are), and we'll improve it out of all recognition. - Again. - - 1.1.2 License Conditions - - Please see the file `LICENSE', supplied as part of any NASM - distribution archive, for the license conditions under which you may - use NASM. NASM is now under the so-called 2-clause BSD license, also - known as the simplified BSD license. - - Copyright 1996-2009 the NASM Authors - All rights reserved. - - Redistribution and use in source and binary forms, with or without - modification, are permitted provided that the following conditions - are met: - - (*) Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. - - (*) Redistributions in binary form must reproduce the above - copyright notice, this list of conditions and the following - disclaimer in the documentation and/or other materials provided - with the distribution. - - THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS - FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE - COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, - INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, - BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; - LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER - CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT - LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN - ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE - POSSIBILITY OF SUCH DAMAGE. - - 1.2 Contact Information - - The current version of NASM (since about 0.98.08) is maintained by a - team of developers, accessible through the `nasm-devel' mailing list - (see below for the link). If you want to report a bug, please read - section 12.2 first. - - NASM has a website at `http://www.nasm.us/'. If it's not there, - google for us! - - New releases, release candidates, and daily development snapshots of - NASM are available from the official web site. - - Announcements are posted to `comp.lang.asm.x86', and to the web site - `http://www.freshmeat.net/'. - - If you want information about the current development status, please - subscribe to the `nasm-devel' email list; see link from the website. - - 1.3 Installation - - 1.3.1 Installing NASM under MS-DOS or Windows - - Once you've obtained the appropriate archive for NASM, - `nasm-XXX-dos.zip' or `nasm-XXX-win32.zip' (where `XXX' denotes the - version number of NASM contained in the archive), unpack it into its - own directory (for example `c:\nasm'). - - The archive will contain a set of executable files: the NASM - executable file `nasm.exe', the NDISASM executable file - `ndisasm.exe', and possibly additional utilities to handle the RDOFF - file format. - - The only file NASM needs to run is its own executable, so copy - `nasm.exe' to a directory on your PATH, or alternatively edit - `autoexec.bat' to add the `nasm' directory to your `PATH' (to do - that under Windows XP, go to Start > Control Panel > System > - Advanced > Environment Variables; these instructions may work under - other versions of Windows as well.) - - That's it - NASM is installed. You don't need the nasm directory to - be present to run NASM (unless you've added it to your `PATH'), so - you can delete it if you need to save space; however, you may want - to keep the documentation or test programs. - - If you've downloaded the DOS source archive, `nasm-XXX.zip', the - `nasm' directory will also contain the full NASM source code, and a - selection of Makefiles you can (hopefully) use to rebuild your copy - of NASM from scratch. See the file `INSTALL' in the source archive. - - Note that a number of files are generated from other files by Perl - scripts. Although the NASM source distribution includes these - generated files, you will need to rebuild them (and hence, will need - a Perl interpreter) if you change insns.dat, standard.mac or the - documentation. It is possible future source distributions may not - include these files at all. Ports of Perl for a variety of - platforms, including DOS and Windows, are available from - www.cpan.org. - - 1.3.2 Installing NASM under Unix - - Once you've obtained the Unix source archive for NASM, - `nasm-XXX.tar.gz' (where `XXX' denotes the version number of NASM - contained in the archive), unpack it into a directory such as - `/usr/local/src'. The archive, when unpacked, will create its own - subdirectory `nasm-XXX'. - - NASM is an auto-configuring package: once you've unpacked it, `cd' - to the directory it's been unpacked into and type `./configure'. - This shell script will find the best C compiler to use for building - NASM and set up Makefiles accordingly. - - Once NASM has auto-configured, you can type `make' to build the - `nasm' and `ndisasm' binaries, and then `make install' to install - them in `/usr/local/bin' and install the man pages `nasm.1' and - `ndisasm.1' in `/usr/local/man/man1'. Alternatively, you can give - options such as `--prefix' to the configure script (see the file - `INSTALL' for more details), or install the programs yourself. - - NASM also comes with a set of utilities for handling the `RDOFF' - custom object-file format, which are in the `rdoff' subdirectory of - the NASM archive. You can build these with `make rdf' and install - them with `make rdf_install', if you want them. - -Chapter 2: Running NASM ------------------------ - - 2.1 NASM Command-Line Syntax - - To assemble a file, you issue a command of the form - - nasm -f <format> <filename> [-o <output>] - - For example, - - nasm -f elf myfile.asm - - will assemble `myfile.asm' into an `ELF' object file `myfile.o'. And - - nasm -f bin myfile.asm -o myfile.com - - will assemble `myfile.asm' into a raw binary file `myfile.com'. - - To produce a listing file, with the hex codes output from NASM - displayed on the left of the original sources, use the `-l' option - to give a listing file name, for example: - - nasm -f coff myfile.asm -l myfile.lst - - To get further usage instructions from NASM, try typing - - nasm -h - - As `-hf', this will also list the available output file formats, and - what they are. - - If you use Linux but aren't sure whether your system is `a.out' or - `ELF', type - - file nasm - - (in the directory in which you put the NASM binary when you - installed it). If it says something like - - nasm: ELF 32-bit LSB executable i386 (386 and up) Version 1 - - then your system is `ELF', and you should use the option `-f elf' - when you want NASM to produce Linux object files. If it says - - nasm: Linux/i386 demand-paged executable (QMAGIC) - - or something similar, your system is `a.out', and you should use - `-f aout' instead (Linux `a.out' systems have long been obsolete, - and are rare these days.) - - Like Unix compilers and assemblers, NASM is silent unless it goes - wrong: you won't see any output at all, unless it gives error - messages. - - 2.1.1 The `-o' Option: Specifying the Output File Name - - NASM will normally choose the name of your output file for you; - precisely how it does this is dependent on the object file format. - For Microsoft object file formats (`obj', `win32' and `win64'), it - will remove the `.asm' extension (or whatever extension you like to - use - NASM doesn't care) from your source file name and substitute - `.obj'. For Unix object file formats (`aout', `as86', `coff', - `elf32', `elf64', `ieee', `macho32' and `macho64') it will - substitute `.o'. For `dbg', `rdf', `ith' and `srec', it will use - `.dbg', `.rdf', `.ith' and `.srec', respectively, and for the `bin' - format it will simply remove the extension, so that `myfile.asm' - produces the output file `myfile'. - - If the output file already exists, NASM will overwrite it, unless it - has the same name as the input file, in which case it will give a - warning and use `nasm.out' as the output file name instead. - - For situations in which this behaviour is unacceptable, NASM - provides the `-o' command-line option, which allows you to specify - your desired output file name. You invoke `-o' by following it with - the name you wish for the output file, either with or without an - intervening space. For example: - - nasm -f bin program.asm -o program.com - nasm -f bin driver.asm -odriver.sys - - Note that this is a small o, and is different from a capital O , - which is used to specify the number of optimisation passes required. - See section 2.1.22. - - 2.1.2 The `-f' Option: Specifying the Output File Format - - If you do not supply the `-f' option to NASM, it will choose an - output file format for you itself. In the distribution versions of - NASM, the default is always `bin'; if you've compiled your own copy - of NASM, you can redefine `OF_DEFAULT' at compile time and choose - what you want the default to be. - - Like `-o', the intervening space between `-f' and the output file - format is optional; so `-f elf' and `-felf' are both valid. - - A complete list of the available output file formats can be given by - issuing the command `nasm -hf'. - - 2.1.3 The `-l' Option: Generating a Listing File - - If you supply the `-l' option to NASM, followed (with the usual - optional space) by a file name, NASM will generate a source-listing - file for you, in which addresses and generated code are listed on - the left, and the actual source code, with expansions of multi-line - macros (except those which specifically request no expansion in - source listings: see section 4.3.10) on the right. For example: - - nasm -f elf myfile.asm -l myfile.lst - - If a list file is selected, you may turn off listing for a section - of your source with `[list -]', and turn it back on with `[list +]', - (the default, obviously). There is no "user form" (without the - brackets). This can be used to list only sections of interest, - avoiding excessively long listings. - - 2.1.4 The `-M' Option: Generate Makefile Dependencies - - This option can be used to generate makefile dependencies on stdout. - This can be redirected to a file for further processing. For - example: - - nasm -M myfile.asm > myfile.dep - - 2.1.5 The `-MG' Option: Generate Makefile Dependencies - - This option can be used to generate makefile dependencies on stdout. - This differs from the `-M' option in that if a nonexisting file is - encountered, it is assumed to be a generated file and is added to - the dependency list without a prefix. - - 2.1.6 The `-MF' Option: Set Makefile Dependency File - - This option can be used with the `-M' or `-MG' options to send the - output to a file, rather than to stdout. For example: - - nasm -M -MF myfile.dep myfile.asm - - 2.1.7 The `-MD' Option: Assemble and Generate Dependencies - - The `-MD' option acts as the combination of the `-M' and `-MF' - options (i.e. a filename has to be specified.) However, unlike the - `-M' or `-MG' options, `-MD' does _not_ inhibit the normal operation - of the assembler. Use this to automatically generate updated - dependencies with every assembly session. For example: - - nasm -f elf -o myfile.o -MD myfile.dep myfile.asm - - 2.1.8 The `-MT' Option: Dependency Target Name - - The `-MT' option can be used to override the default name of the - dependency target. This is normally the same as the output filename, - specified by the `-o' option. - - 2.1.9 The `-MQ' Option: Dependency Target Name (Quoted) - - The `-MQ' option acts as the `-MT' option, except it tries to quote - characters that have special meaning in Makefile syntax. This is not - foolproof, as not all characters with special meaning are quotable - in Make. - -2.1.10 The `-MP' Option: Emit phony targets - - When used with any of the dependency generation options, the `-MP' - option causes NASM to emit a phony target without dependencies for - each header file. This prevents Make from complaining if a header - file has been removed. - -2.1.11 The `-F' Option: Selecting a Debug Information Format - - This option is used to select the format of the debug information - emitted into the output file, to be used by a debugger (or _will_ - be). Prior to version 2.03.01, the use of this switch did _not_ - enable output of the selected debug info format. Use `-g', see - section 2.1.12, to enable output. Versions 2.03.01 and later - automatically enable `-g' if `-F' is specified. - - A complete list of the available debug file formats for an output - format can be seen by issuing the command `nasm -f <format> -y'. Not - all output formats currently support debugging output. See section - 2.1.26. - - This should not be confused with the `-f dbg' output format option - which is not built into NASM by default. For information on how to - enable it when building from the sources, see section 7.14. - -2.1.12 The `-g' Option: Enabling Debug Information. - - This option can be used to generate debugging information in the - specified format. See section 2.1.11. Using `-g' without `-F' - results in emitting debug info in the default format, if any, for - the selected output format. If no debug information is currently - implemented in the selected output format, `-g' is _silently - ignored_. - -2.1.13 The `-X' Option: Selecting an Error Reporting Format - - This option can be used to select an error reporting format for any - error messages that might be produced by NASM. - - Currently, two error reporting formats may be selected. They are the - `-Xvc' option and the `-Xgnu' option. The GNU format is the default - and looks like this: - - filename.asm:65: error: specific error message - - where `filename.asm' is the name of the source file in which the - error was detected, `65' is the source file line number on which the - error was detected, `error' is the severity of the error (this could - be `warning'), and `specific error message' is a more detailed text - message which should help pinpoint the exact problem. - - The other format, specified by `-Xvc' is the style used by Microsoft - Visual C++ and some other programs. It looks like this: - - filename.asm(65) : error: specific error message - - where the only difference is that the line number is in parentheses - instead of being delimited by colons. - - See also the `Visual C++' output format, section 7.5. - -2.1.14 The `-Z' Option: Send Errors to a File - - Under `MS-DOS' it can be difficult (though there are ways) to - redirect the standard-error output of a program to a file. Since - NASM usually produces its warning and error messages on `stderr', - this can make it hard to capture the errors if (for example) you - want to load them into an editor. - - NASM therefore provides the `-Z' option, taking a filename argument - which causes errors to be sent to the specified files rather than - standard error. Therefore you can redirect the errors into a file by - typing - - nasm -Z myfile.err -f obj myfile.asm - - In earlier versions of NASM, this option was called `-E', but it was - changed since `-E' is an option conventionally used for - preprocessing only, with disastrous results. See section 2.1.20. - -2.1.15 The `-s' Option: Send Errors to `stdout' - - The `-s' option redirects error messages to `stdout' rather than - `stderr', so it can be redirected under `MS-DOS'. To assemble the - file `myfile.asm' and pipe its output to the `more' program, you can - type: - - nasm -s -f obj myfile.asm | more - - See also the `-Z' option, section 2.1.14. - -2.1.16 The `-i' Option: Include File Search Directories - - When NASM sees the `%include' or `%pathsearch' directive in a source - file (see section 4.6.1, section 4.6.2 or section 3.2.3), it will - search for the given file not only in the current directory, but - also in any directories specified on the command line by the use of - the `-i' option. Therefore you can include files from a macro - library, for example, by typing - - nasm -ic:\macrolib\ -f obj myfile.asm - - (As usual, a space between `-i' and the path name is allowed, and - optional). - - NASM, in the interests of complete source-code portability, does not - understand the file naming conventions of the OS it is running on; - the string you provide as an argument to the `-i' option will be - prepended exactly as written to the name of the include file. - Therefore the trailing backslash in the above example is necessary. - Under Unix, a trailing forward slash is similarly necessary. - - (You can use this to your advantage, if you're really perverse, by - noting that the option `-ifoo' will cause `%include "bar.i"' to - search for the file `foobar.i'...) - - If you want to define a _standard_ include search path, similar to - `/usr/include' on Unix systems, you should place one or more `-i' - directives in the `NASMENV' environment variable (see section - 2.1.28). - - For Makefile compatibility with many C compilers, this option can - also be specified as `-I'. - -2.1.17 The `-p' Option: Pre-Include a File - - NASM allows you to specify files to be _pre-included_ into your - source file, by the use of the `-p' option. So running - - nasm myfile.asm -p myinc.inc - - is equivalent to running `nasm myfile.asm' and placing the directive - `%include "myinc.inc"' at the start of the file. - - For consistency with the `-I', `-D' and `-U' options, this option - can also be specified as `-P'. - -2.1.18 The `-d' Option: Pre-Define a Macro - - Just as the `-p' option gives an alternative to placing `%include' - directives at the start of a source file, the `-d' option gives an - alternative to placing a `%define' directive. You could code - - nasm myfile.asm -dFOO=100 - - as an alternative to placing the directive - - %define FOO 100 - - at the start of the file. You can miss off the macro value, as well: - the option `-dFOO' is equivalent to coding `%define FOO'. This form - of the directive may be useful for selecting assembly-time options - which are then tested using `%ifdef', for example `-dDEBUG'. - - For Makefile compatibility with many C compilers, this option can - also be specified as `-D'. - -2.1.19 The `-u' Option: Undefine a Macro - - The `-u' option undefines a macro that would otherwise have been - pre-defined, either automatically or by a `-p' or `-d' option - specified earlier on the command lines. - - For example, the following command line: - - nasm myfile.asm -dFOO=100 -uFOO - - would result in `FOO' _not_ being a predefined macro in the program. - This is useful to override options specified at a different point in - a Makefile. - - For Makefile compatibility with many C compilers, this option can - also be specified as `-U'. - -2.1.20 The `-E' Option: Preprocess Only - - NASM allows the preprocessor to be run on its own, up to a point. - Using the `-E' option (which requires no arguments) will cause NASM - to preprocess its input file, expand all the macro references, - remove all the comments and preprocessor directives, and print the - resulting file on standard output (or save it to a file, if the `-o' - option is also used). - - This option cannot be applied to programs which require the - preprocessor to evaluate expressions which depend on the values of - symbols: so code such as - - %assign tablesize ($-tablestart) - - will cause an error in preprocess-only mode. - - For compatiblity with older version of NASM, this option can also be - written `-e'. `-E' in older versions of NASM was the equivalent of - the current `-Z' option, section 2.1.14. - -2.1.21 The `-a' Option: Don't Preprocess At All - - If NASM is being used as the back end to a compiler, it might be - desirable to suppress preprocessing completely and assume the - compiler has already done it, to save time and increase compilation - speeds. The `-a' option, requiring no argument, instructs NASM to - replace its powerful preprocessor with a stub preprocessor which - does nothing. - -2.1.22 The `-O' Option: Specifying Multipass Optimization - - NASM defaults to not optimizing operands which can fit into a signed - byte. This means that if you want the shortest possible object code, - you have to enable optimization. - - Using the `-O' option, you can tell NASM to carry out different - levels of optimization. The syntax is: - - (*) `-O0': No optimization. All operands take their long forms, if a - short form is not specified, except conditional jumps. This is - intended to match NASM 0.98 behavior. - - (*) `-O1': Minimal optimization. As above, but immediate operands - which will fit in a signed byte are optimized, unless the long - form is specified. Conditional jumps default to the long form - unless otherwise specified. - - (*) `-Ox' (where `x' is the actual letter `x'): Multipass - optimization. Minimize branch offsets and signed immediate - bytes, overriding size specification unless the `strict' keyword - has been used (see section 3.7). For compatability with earlier - releases, the letter `x' may also be any number greater than - one. This number has no effect on the actual number of passes. - - The `-Ox' mode is recommended for most uses. - - Note that this is a capital `O', and is different from a small `o', - which is used to specify the output file name. See section 2.1.1. - -2.1.23 The `-t' Option: Enable TASM Compatibility Mode - - NASM includes a limited form of compatibility with Borland's `TASM'. - When NASM's `-t' option is used, the following changes are made: - - (*) local labels may be prefixed with `@@' instead of `.' - - (*) size override is supported within brackets. In TASM compatible - mode, a size override inside square brackets changes the size of - the operand, and not the address type of the operand as it does - in NASM syntax. E.g. `mov eax,[DWORD val]' is valid syntax in - TASM compatibility mode. Note that you lose the ability to - override the default address type for the instruction. - - (*) unprefixed forms of some directives supported (`arg', `elif', - `else', `endif', `if', `ifdef', `ifdifi', `ifndef', `include', - `local') - -2.1.24 The `-w' and `-W' Options: Enable or Disable Assembly Warnings - - NASM can observe many conditions during the course of assembly which - are worth mentioning to the user, but not a sufficiently severe - error to justify NASM refusing to generate an output file. These - conditions are reported like errors, but come up with the word - `warning' before the message. Warnings do not prevent NASM from - generating an output file and returning a success status to the - operating system. - - Some conditions are even less severe than that: they are only - sometimes worth mentioning to the user. Therefore NASM supports the - `-w' command-line option, which enables or disables certain classes - of assembly warning. Such warning classes are described by a name, - for example `orphan-labels'; you can enable warnings of this class - by the command-line option `-w+orphan-labels' and disable it by - `-w-orphan-labels'. - - The suppressible warning classes are: - - (*) `macro-params' covers warnings about multi-line macros being - invoked with the wrong number of parameters. This warning class - is enabled by default; see section 4.3.2 for an example of why - you might want to disable it. - - (*) `macro-selfref' warns if a macro references itself. This warning - class is disabled by default. - - (*) `macro-defaults' warns when a macro has more default parameters - than optional parameters. This warning class is enabled by - default; see section 4.3.5 for why you might want to disable it. - - (*) `orphan-labels' covers warnings about source lines which contain - no instruction but define a label without a trailing colon. NASM - warns about this somewhat obscure condition by default; see - section 3.1 for more information. - - (*) `number-overflow' covers warnings about numeric constants which - don't fit in 64 bits. This warning class is enabled by default. - - (*) `gnu-elf-extensions' warns if 8-bit or 16-bit relocations are - used in `-f elf' format. The GNU extensions allow this. This - warning class is disabled by default. - - (*) `float-overflow' warns about floating point overflow. Enabled by - default. - - (*) `float-denorm' warns about floating point denormals. Disabled by - default. - - (*) `float-underflow' warns about floating point underflow. Disabled - by default. - - (*) `float-toolong' warns about too many digits in floating-point - numbers. Enabled by default. - - (*) `user' controls `%warning' directives (see section 4.9). Enabled - by default. - - (*) `error' causes warnings to be treated as errors. Disabled by - default. - - (*) `all' is an alias for _all_ suppressible warning classes (not - including `error'). Thus, `-w+all' enables all available - warnings. - - In addition, you can set warning classes across sections. Warning - classes may be enabled with `[warning +warning-name]', disabled with - `[warning -warning-name]' or reset to their original value with - `[warning *warning-name]'. No "user form" (without the brackets) - exists. - - Since version 2.00, NASM has also supported the gcc-like syntax - `-Wwarning' and `-Wno-warning' instead of `-w+warning' and - `-w-warning', respectively. - -2.1.25 The `-v' Option: Display Version Info - - Typing `NASM -v' will display the version of NASM which you are - using, and the date on which it was compiled. - - You will need the version number if you report a bug. - -2.1.26 The `-y' Option: Display Available Debug Info Formats - - Typing `nasm -f <option> -y' will display a list of the available - debug info formats for the given output format. The default format - is indicated by an asterisk. For example: - - nasm -f elf -y - - valid debug formats for 'elf32' output format are - ('*' denotes default): - * stabs ELF32 (i386) stabs debug format for Linux - dwarf elf32 (i386) dwarf debug format for Linux - -2.1.27 The `--prefix' and `--postfix' Options. - - The `--prefix' and `--postfix' options prepend or append - (respectively) the given argument to all `global' or `extern' - variables. E.g. `--prefix _' will prepend the underscore to all - global and external variables, as C sometimes (but not always) likes - it. - -2.1.28 The `NASMENV' Environment Variable - - If you define an environment variable called `NASMENV', the program - will interpret it as a list of extra command-line options, which are - processed before the real command line. You can use this to define - standard search directories for include files, by putting `-i' - options in the `NASMENV' variable. - - The value of the variable is split up at white space, so that the - value `-s -ic:\nasmlib\' will be treated as two separate options. - However, that means that the value `-dNAME="my name"' won't do what - you might want, because it will be split at the space and the NASM - command-line processing will get confused by the two nonsensical - words `-dNAME="my' and `name"'. - - To get round this, NASM provides a feature whereby, if you begin the - `NASMENV' environment variable with some character that isn't a - minus sign, then NASM will treat this character as the separator - character for options. So setting the `NASMENV' variable to the - value `!-s!-ic:\nasmlib\' is equivalent to setting it to - `-s -ic:\nasmlib\', but `!-dNAME="my name"' will work. - - This environment variable was previously called `NASM'. This was - changed with version 0.98.31. - - 2.2 Quick Start for MASM Users - - If you're used to writing programs with MASM, or with TASM in MASM- - compatible (non-Ideal) mode, or with `a86', this section attempts to - outline the major differences between MASM's syntax and NASM's. If - you're not already used to MASM, it's probably worth skipping this - section. - - 2.2.1 NASM Is Case-Sensitive - - One simple difference is that NASM is case-sensitive. It makes a - difference whether you call your label `foo', `Foo' or `FOO'. If - you're assembling to `DOS' or `OS/2' `.OBJ' files, you can invoke - the `UPPERCASE' directive (documented in section 7.4) to ensure that - all symbols exported to other code modules are forced to be upper - case; but even then, _within_ a single module, NASM will distinguish - between labels differing only in case. - - 2.2.2 NASM Requires Square Brackets For Memory References - - NASM was designed with simplicity of syntax in mind. One of the - design goals of NASM is that it should be possible, as far as is - practical, for the user to look at a single line of NASM code and - tell what opcode is generated by it. You can't do this in MASM: if - you declare, for example, - - foo equ 1 - bar dw 2 - - then the two lines of code - - mov ax,foo - mov ax,bar - - generate completely different opcodes, despite having identical- - looking syntaxes. - - NASM avoids this undesirable situation by having a much simpler - syntax for memory references. The rule is simply that any access to - the _contents_ of a memory location requires square brackets around - the address, and any access to the _address_ of a variable doesn't. - So an instruction of the form `mov ax,foo' will _always_ refer to a - compile-time constant, whether it's an `EQU' or the address of a - variable; and to access the _contents_ of the variable `bar', you - must code `mov ax,[bar]'. - - This also means that NASM has no need for MASM's `OFFSET' keyword, - since the MASM code `mov ax,offset bar' means exactly the same thing - as NASM's `mov ax,bar'. If you're trying to get large amounts of - MASM code to assemble sensibly under NASM, you can always code - `%idefine offset' to make the preprocessor treat the `OFFSET' - keyword as a no-op. - - This issue is even more confusing in `a86', where declaring a label - with a trailing colon defines it to be a `label' as opposed to a - `variable' and causes `a86' to adopt NASM-style semantics; so in - `a86', `mov ax,var' has different behaviour depending on whether - `var' was declared as `var: dw 0' (a label) or `var dw 0' (a word- - size variable). NASM is very simple by comparison: _everything_ is a - label. - - NASM, in the interests of simplicity, also does not support the - hybrid syntaxes supported by MASM and its clones, such as - `mov ax,table[bx]', where a memory reference is denoted by one - portion outside square brackets and another portion inside. The - correct syntax for the above is `mov ax,[table+bx]'. Likewise, - `mov ax,es:[di]' is wrong and `mov ax,[es:di]' is right. - - 2.2.3 NASM Doesn't Store Variable Types - - NASM, by design, chooses not to remember the types of variables you - declare. Whereas MASM will remember, on seeing `var dw 0', that you - declared `var' as a word-size variable, and will then be able to - fill in the ambiguity in the size of the instruction `mov var,2', - NASM will deliberately remember nothing about the symbol `var' - except where it begins, and so you must explicitly code - `mov word [var],2'. - - For this reason, NASM doesn't support the `LODS', `MOVS', `STOS', - `SCAS', `CMPS', `INS', or `OUTS' instructions, but only supports the - forms such as `LODSB', `MOVSW', and `SCASD', which explicitly - specify the size of the components of the strings being manipulated. - - 2.2.4 NASM Doesn't `ASSUME' - - As part of NASM's drive for simplicity, it also does not support the - `ASSUME' directive. NASM will not keep track of what values you - choose to put in your segment registers, and will never - _automatically_ generate a segment override prefix. - - 2.2.5 NASM Doesn't Support Memory Models - - NASM also does not have any directives to support different 16-bit - memory models. The programmer has to keep track of which functions - are supposed to be called with a far call and which with a near - call, and is responsible for putting the correct form of `RET' - instruction (`RETN' or `RETF'; NASM accepts `RET' itself as an - alternate form for `RETN'); in addition, the programmer is - responsible for coding CALL FAR instructions where necessary when - calling _external_ functions, and must also keep track of which - external variable definitions are far and which are near. - - 2.2.6 Floating-Point Differences - - NASM uses different names to refer to floating-point registers from - MASM: where MASM would call them `ST(0)', `ST(1)' and so on, and - `a86' would call them simply `0', `1' and so on, NASM chooses to - call them `st0', `st1' etc. - - As of version 0.96, NASM now treats the instructions with `nowait' - forms in the same way as MASM-compatible assemblers. The - idiosyncratic treatment employed by 0.95 and earlier was based on a - misunderstanding by the authors. - - 2.2.7 Other Differences - - For historical reasons, NASM uses the keyword `TWORD' where MASM and - compatible assemblers use `TBYTE'. - - NASM does not declare uninitialized storage in the same way as MASM: - where a MASM programmer might use `stack db 64 dup (?)', NASM - requires `stack resb 64', intended to be read as `reserve 64 bytes'. - For a limited amount of compatibility, since NASM treats `?' as a - valid character in symbol names, you can code `? equ 0' and then - writing `dw ?' will at least do something vaguely useful. `DUP' is - still not a supported syntax, however. - - In addition to all of this, macros and directives work completely - differently to MASM. See chapter 4 and chapter 6 for further - details. - -Chapter 3: The NASM Language ----------------------------- - - 3.1 Layout of a NASM Source Line - - Like most assemblers, each NASM source line contains (unless it is a - macro, a preprocessor directive or an assembler directive: see - chapter 4 and chapter 6) some combination of the four fields - - label: instruction operands ; comment - - As usual, most of these fields are optional; the presence or absence - of any combination of a label, an instruction and a comment is - allowed. Of course, the operand field is either required or - forbidden by the presence and nature of the instruction field. - - NASM uses backslash (\) as the line continuation character; if a - line ends with backslash, the next line is considered to be a part - of the backslash-ended line. - - NASM places no restrictions on white space within a line: labels may - have white space before them, or instructions may have no space - before them, or anything. The colon after a label is also optional. - (Note that this means that if you intend to code `lodsb' alone on a - line, and type `lodab' by accident, then that's still a valid source - line which does nothing but define a label. Running NASM with the - command-line option `-w+orphan-labels' will cause it to warn you if - you define a label alone on a line without a trailing colon.) - - Valid characters in labels are letters, numbers, `_', `$', `#', `@', - `~', `.', and `?'. The only characters which may be used as the - _first_ character of an identifier are letters, `.' (with special - meaning: see section 3.9), `_' and `?'. An identifier may also be - prefixed with a `$' to indicate that it is intended to be read as an - identifier and not a reserved word; thus, if some other module you - are linking with defines a symbol called `eax', you can refer to - `$eax' in NASM code to distinguish the symbol from the register. - Maximum length of an identifier is 4095 characters. - - The instruction field may contain any machine instruction: Pentium - and P6 instructions, FPU instructions, MMX instructions and even - undocumented instructions are all supported. The instruction may be - prefixed by `LOCK', `REP', `REPE'/`REPZ' or `REPNE'/`REPNZ', in the - usual way. Explicit address-size and operand-size prefixes `A16', - `A32', `A64', `O16' and `O32', `O64' are provided - one example of - their use is given in chapter 10. You can also use the name of a - segment register as an instruction prefix: coding `es mov [bx],ax' - is equivalent to coding `mov [es:bx],ax'. We recommend the latter - syntax, since it is consistent with other syntactic features of the - language, but for instructions such as `LODSB', which has no - operands and yet can require a segment override, there is no clean - syntactic way to proceed apart from `es lodsb'. - - An instruction is not required to use a prefix: prefixes such as - `CS', `A32', `LOCK' or `REPE' can appear on a line by themselves, - and NASM will just generate the prefix bytes. - - In addition to actual machine instructions, NASM also supports a - number of pseudo-instructions, described in section 3.2. - - Instruction operands may take a number of forms: they can be - registers, described simply by the register name (e.g. `ax', `bp', - `ebx', `cr0': NASM does not use the `gas'-style syntax in which - register names must be prefixed by a `%' sign), or they can be - effective addresses (see section 3.3), constants (section 3.4) or - expressions (section 3.5). - - For x87 floating-point instructions, NASM accepts a wide range of - syntaxes: you can use two-operand forms like MASM supports, or you - can use NASM's native single-operand forms in most cases. For - example, you can code: - - fadd st1 ; this sets st0 := st0 + st1 - fadd st0,st1 ; so does this - - fadd st1,st0 ; this sets st1 := st1 + st0 - fadd to st1 ; so does this - - Almost any x87 floating-point instruction that references memory - must use one of the prefixes `DWORD', `QWORD' or `TWORD' to indicate - what size of memory operand it refers to. - - 3.2 Pseudo-Instructions - - Pseudo-instructions are things which, though not real x86 machine - instructions, are used in the instruction field anyway because - that's the most convenient place to put them. The current pseudo- - instructions are `DB', `DW', `DD', `DQ', `DT', `DO' and `DY'; their - uninitialized counterparts `RESB', `RESW', `RESD', `RESQ', `REST', - `RESO' and `RESY'; the `INCBIN' command, the `EQU' command, and the - `TIMES' prefix. - - 3.2.1 `DB' and Friends: Declaring Initialized Data - - `DB', `DW', `DD', `DQ', `DT', `DO' and `DY' are used, much as in - MASM, to declare initialized data in the output file. They can be - invoked in a wide range of ways: - - db 0x55 ; just the byte 0x55 - db 0x55,0x56,0x57 ; three bytes in succession - db 'a',0x55 ; character constants are OK - db 'hello',13,10,'$' ; so are string constants - dw 0x1234 ; 0x34 0x12 - dw 'a' ; 0x61 0x00 (it's just a number) - dw 'ab' ; 0x61 0x62 (character constant) - dw 'abc' ; 0x61 0x62 0x63 0x00 (string) - dd 0x12345678 ; 0x78 0x56 0x34 0x12 - dd 1.234567e20 ; floating-point constant - dq 0x123456789abcdef0 ; eight byte constant - dq 1.234567e20 ; double-precision float - dt 1.234567e20 ; extended-precision float - - `DT', `DO' and `DY' do not accept numeric constants as operands. - - 3.2.2 `RESB' and Friends: Declaring Uninitialized Data - - `RESB', `RESW', `RESD', `RESQ', `REST', `RESO' and `RESY' are - designed to be used in the BSS section of a module: they declare - _uninitialized_ storage space. Each takes a single operand, which is - the number of bytes, words, doublewords or whatever to reserve. As - stated in section 2.2.7, NASM does not support the MASM/TASM syntax - of reserving uninitialized space by writing `DW ?' or similar - things: this is what it does instead. The operand to a `RESB'-type - pseudo-instruction is a _critical expression_: see section 3.8. - - For example: - - buffer: resb 64 ; reserve 64 bytes - wordvar: resw 1 ; reserve a word - realarray resq 10 ; array of ten reals - ymmval: resy 1 ; one YMM register - - 3.2.3 `INCBIN': Including External Binary Files - - `INCBIN' is borrowed from the old Amiga assembler DevPac: it - includes a binary file verbatim into the output file. This can be - handy for (for example) including graphics and sound data directly - into a game executable file. It can be called in one of these three - ways: - - incbin "file.dat" ; include the whole file - incbin "file.dat",1024 ; skip the first 1024 bytes - incbin "file.dat",1024,512 ; skip the first 1024, and - ; actually include at most 512 - - `INCBIN' is both a directive and a standard macro; the standard - macro version searches for the file in the include file search path - and adds the file to the dependency lists. This macro can be - overridden if desired. - - 3.2.4 `EQU': Defining Constants - - `EQU' defines a symbol to a given constant value: when `EQU' is - used, the source line must contain a label. The action of `EQU' is - to define the given label name to the value of its (only) operand. - This definition is absolute, and cannot change later. So, for - example, - - message db 'hello, world' - msglen equ $-message - - defines `msglen' to be the constant 12. `msglen' may not then be - redefined later. This is not a preprocessor definition either: the - value of `msglen' is evaluated _once_, using the value of `$' (see - section 3.5 for an explanation of `$') at the point of definition, - rather than being evaluated wherever it is referenced and using the - value of `$' at the point of reference. - - 3.2.5 `TIMES': Repeating Instructions or Data - - The `TIMES' prefix causes the instruction to be assembled multiple - times. This is partly present as NASM's equivalent of the `DUP' - syntax supported by MASM-compatible assemblers, in that you can code - - zerobuf: times 64 db 0 - - or similar things; but `TIMES' is more versatile than that. The - argument to `TIMES' is not just a numeric constant, but a numeric - _expression_, so you can do things like - - buffer: db 'hello, world' - times 64-$+buffer db ' ' - - which will store exactly enough spaces to make the total length of - `buffer' up to 64. Finally, `TIMES' can be applied to ordinary - instructions, so you can code trivial unrolled loops in it: - - times 100 movsb - - Note that there is no effective difference between - `times 100 resb 1' and `resb 100', except that the latter will be - assembled about 100 times faster due to the internal structure of - the assembler. - - The operand to `TIMES' is a critical expression (section 3.8). - - Note also that `TIMES' can't be applied to macros: the reason for - this is that `TIMES' is processed after the macro phase, which - allows the argument to `TIMES' to contain expressions such as - `64-$+buffer' as above. To repeat more than one line of code, or a - complex macro, use the preprocessor `%rep' directive. - - 3.3 Effective Addresses - - An effective address is any operand to an instruction which - references memory. Effective addresses, in NASM, have a very simple - syntax: they consist of an expression evaluating to the desired - address, enclosed in square brackets. For example: - - wordvar dw 123 - mov ax,[wordvar] - mov ax,[wordvar+1] - mov ax,[es:wordvar+bx] - - Anything not conforming to this simple system is not a valid memory - reference in NASM, for example `es:wordvar[bx]'. - - More complicated effective addresses, such as those involving more - than one register, work in exactly the same way: - - mov eax,[ebx*2+ecx+offset] - mov ax,[bp+di+8] - - NASM is capable of doing algebra on these effective addresses, so - that things which don't necessarily _look_ legal are perfectly all - right: - - mov eax,[ebx*5] ; assembles as [ebx*4+ebx] - mov eax,[label1*2-label2] ; ie [label1+(label1-label2)] - - Some forms of effective address have more than one assembled form; - in most such cases NASM will generate the smallest form it can. For - example, there are distinct assembled forms for the 32-bit effective - addresses `[eax*2+0]' and `[eax+eax]', and NASM will generally - generate the latter on the grounds that the former requires four - bytes to store a zero offset. - - NASM has a hinting mechanism which will cause `[eax+ebx]' and - `[ebx+eax]' to generate different opcodes; this is occasionally - useful because `[esi+ebp]' and `[ebp+esi]' have different default - segment registers. - - However, you can force NASM to generate an effective address in a - particular form by the use of the keywords `BYTE', `WORD', `DWORD' - and `NOSPLIT'. If you need `[eax+3]' to be assembled using a double- - word offset field instead of the one byte NASM will normally - generate, you can code `[dword eax+3]'. Similarly, you can force - NASM to use a byte offset for a small value which it hasn't seen on - the first pass (see section 3.8 for an example of such a code - fragment) by using `[byte eax+offset]'. As special cases, - `[byte eax]' will code `[eax+0]' with a byte offset of zero, and - `[dword eax]' will code it with a double-word offset of zero. The - normal form, `[eax]', will be coded with no offset field. - - The form described in the previous paragraph is also useful if you - are trying to access data in a 32-bit segment from within 16 bit - code. For more information on this see the section on mixed-size - addressing (section 10.2). In particular, if you need to access data - with a known offset that is larger than will fit in a 16-bit value, - if you don't specify that it is a dword offset, nasm will cause the - high word of the offset to be lost. - - Similarly, NASM will split `[eax*2]' into `[eax+eax]' because that - allows the offset field to be absent and space to be saved; in fact, - it will also split `[eax*2+offset]' into `[eax+eax+offset]'. You can - combat this behaviour by the use of the `NOSPLIT' keyword: - `[nosplit eax*2]' will force `[eax*2+0]' to be generated literally. - - In 64-bit mode, NASM will by default generate absolute addresses. - The `REL' keyword makes it produce `RIP'-relative addresses. Since - this is frequently the normally desired behaviour, see the `DEFAULT' - directive (section 6.2). The keyword `ABS' overrides `REL'. - - 3.4 Constants - - NASM understands four different types of constant: numeric, - character, string and floating-point. - - 3.4.1 Numeric Constants - - A numeric constant is simply a number. NASM allows you to specify - numbers in a variety of number bases, in a variety of ways: you can - suffix `H' or `X', `Q' or `O', and `B' for hexadecimal, octal and - binary respectively, or you can prefix `0x' for hexadecimal in the - style of C, or you can prefix `$' for hexadecimal in the style of - Borland Pascal. Note, though, that the `$' prefix does double duty - as a prefix on identifiers (see section 3.1), so a hex number - prefixed with a `$' sign must have a digit after the `$' rather than - a letter. In addition, current versions of NASM accept the prefix - `0h' for hexadecimal, `0o' or `0q' for octal, and `0b' for binary. - Please note that unlike C, a `0' prefix by itself does _not_ imply - an octal constant! - - Numeric constants can have underscores (`_') interspersed to break - up long strings. - - Some examples (all producing exactly the same code): - - mov ax,200 ; decimal - mov ax,0200 ; still decimal - mov ax,0200d ; explicitly decimal - mov ax,0d200 ; also decimal - mov ax,0c8h ; hex - mov ax,$0c8 ; hex again: the 0 is required - mov ax,0xc8 ; hex yet again - mov ax,0hc8 ; still hex - mov ax,310q ; octal - mov ax,310o ; octal again - mov ax,0o310 ; octal yet again - mov ax,0q310 ; hex yet again - mov ax,11001000b ; binary - mov ax,1100_1000b ; same binary constant - mov ax,0b1100_1000 ; same binary constant yet again - - 3.4.2 Character Strings - - A character string consists of up to eight characters enclosed in - either single quotes (`'...''), double quotes (`"..."') or - backquotes (``...`'). Single or double quotes are equivalent to NASM - (except of course that surrounding the constant with single quotes - allows double quotes to appear within it and vice versa); the - contents of those are represented verbatim. Strings enclosed in - backquotes support C-style `\'-escapes for special characters. - - The following escape sequences are recognized by backquoted strings: - - \' single quote (') - \" double quote (") - \` backquote (`) - \\ backslash (\) - \? question mark (?) - \a BEL (ASCII 7) - \b BS (ASCII 8) - \t TAB (ASCII 9) - \n LF (ASCII 10) - \v VT (ASCII 11) - \f FF (ASCII 12) - \r CR (ASCII 13) - \e ESC (ASCII 27) - \377 Up to 3 octal digits - literal byte - \xFF Up to 2 hexadecimal digits - literal byte - \u1234 4 hexadecimal digits - Unicode character - \U12345678 8 hexadecimal digits - Unicode character - - All other escape sequences are reserved. Note that `\0', meaning a - `NUL' character (ASCII 0), is a special case of the octal escape - sequence. - - Unicode characters specified with `\u' or `\U' are converted to - UTF-8. For example, the following lines are all equivalent: - - db `\u263a` ; UTF-8 smiley face - db `\xe2\x98\xba` ; UTF-8 smiley face - db 0E2h, 098h, 0BAh ; UTF-8 smiley face - - 3.4.3 Character Constants - - A character constant consists of a string up to eight bytes long, - used in an expression context. It is treated as if it was an - integer. - - A character constant with more than one byte will be arranged with - little-endian order in mind: if you code - - mov eax,'abcd' - - then the constant generated is not `0x61626364', but `0x64636261', - so that if you were then to store the value into memory, it would - read `abcd' rather than `dcba'. This is also the sense of character - constants understood by the Pentium's `CPUID' instruction. - - 3.4.4 String Constants - - String constants are character strings used in the context of some - pseudo-instructions, namely the `DB' family and `INCBIN' (where it - represents a filename.) They are also used in certain preprocessor - directives. - - A string constant looks like a character constant, only longer. It - is treated as a concatenation of maximum-size character constants - for the conditions. So the following are equivalent: - - db 'hello' ; string constant - db 'h','e','l','l','o' ; equivalent character constants - - And the following are also equivalent: - - dd 'ninechars' ; doubleword string constant - dd 'nine','char','s' ; becomes three doublewords - db 'ninechars',0,0,0 ; and really looks like this - - Note that when used in a string-supporting context, quoted strings - are treated as a string constants even if they are short enough to - be a character constant, because otherwise `db 'ab'' would have the - same effect as `db 'a'', which would be silly. Similarly, three- - character or four-character constants are treated as strings when - they are operands to `DW', and so forth. - - 3.4.5 Unicode Strings - - The special operators `__utf16__' and `__utf32__' allows definition - of Unicode strings. They take a string in UTF-8 format and converts - it to (littleendian) UTF-16 or UTF-32, respectively. - - For example: - - %define u(x) __utf16__(x) - %define w(x) __utf32__(x) - - dw u('C:\WINDOWS'), 0 ; Pathname in UTF-16 - dd w(`A + B = \u206a`), 0 ; String in UTF-32 - - `__utf16__' and `__utf32__' can be applied either to strings passed - to the `DB' family instructions, or to character constants in an - expression context. - - 3.4.6 Floating-Point Constants - - Floating-point constants are acceptable only as arguments to `DB', - `DW', `DD', `DQ', `DT', and `DO', or as arguments to the special - operators `__float8__', `__float16__', `__float32__', `__float64__', - `__float80m__', `__float80e__', `__float128l__', and - `__float128h__'. - - Floating-point constants are expressed in the traditional form: - digits, then a period, then optionally more digits, then optionally - an `E' followed by an exponent. The period is mandatory, so that - NASM can distinguish between `dd 1', which declares an integer - constant, and `dd 1.0' which declares a floating-point constant. - NASM also support C99-style hexadecimal floating-point: `0x', - hexadecimal digits, period, optionally more hexadeximal digits, then - optionally a `P' followed by a _binary_ (not hexadecimal) exponent - in decimal notation. - - Underscores to break up groups of digits are permitted in floating- - point constants as well. - - Some examples: - - db -0.2 ; "Quarter precision" - dw -0.5 ; IEEE 754r/SSE5 half precision - dd 1.2 ; an easy one - dd 1.222_222_222 ; underscores are permitted - dd 0x1p+2 ; 1.0x2^2 = 4.0 - dq 0x1p+32 ; 1.0x2^32 = 4 294 967 296.0 - dq 1.e10 ; 10 000 000 000.0 - dq 1.e+10 ; synonymous with 1.e10 - dq 1.e-10 ; 0.000 000 000 1 - dt 3.141592653589793238462 ; pi - do 1.e+4000 ; IEEE 754r quad precision - - The 8-bit "quarter-precision" floating-point format is - sign:exponent:mantissa = 1:4:3 with an exponent bias of 7. This - appears to be the most frequently used 8-bit floating-point format, - although it is not covered by any formal standard. This is sometimes - called a "minifloat." - - The special operators are used to produce floating-point numbers in - other contexts. They produce the binary representation of a specific - floating-point number as an integer, and can use anywhere integer - constants are used in an expression. `__float80m__' and - `__float80e__' produce the 64-bit mantissa and 16-bit exponent of an - 80-bit floating-point number, and `__float128l__' and - `__float128h__' produce the lower and upper 64-bit halves of a 128- - bit floating-point number, respectively. - - For example: - - mov rax,__float64__(3.141592653589793238462) - - ... would assign the binary representation of pi as a 64-bit - floating point number into `RAX'. This is exactly equivalent to: - - mov rax,0x400921fb54442d18 - - NASM cannot do compile-time arithmetic on floating-point constants. - This is because NASM is designed to be portable - although it always - generates code to run on x86 processors, the assembler itself can - run on any system with an ANSI C compiler. Therefore, the assembler - cannot guarantee the presence of a floating-point unit capable of - handling the Intel number formats, and so for NASM to be able to do - floating arithmetic it would have to include its own complete set of - floating-point routines, which would significantly increase the size - of the assembler for very little benefit. - - The special tokens `__Infinity__', `__QNaN__' (or `__NaN__') and - `__SNaN__' can be used to generate infinities, quiet NaNs, and - signalling NaNs, respectively. These are normally used as macros: - - %define Inf __Infinity__ - %define NaN __QNaN__ - - dq +1.5, -Inf, NaN ; Double-precision constants - - 3.4.7 Packed BCD Constants - - x87-style packed BCD constants can be used in the same contexts as - 80-bit floating-point numbers. They are suffixed with `p' or - prefixed with `0p', and can include up to 18 decimal digits. - - As with other numeric constants, underscores can be used to separate - digits. - - For example: - - dt 12_345_678_901_245_678p - dt -12_345_678_901_245_678p - dt +0p33 - dt 33p - - 3.5 Expressions - - Expressions in NASM are similar in syntax to those in C. Expressions - are evaluated as 64-bit integers which are then adjusted to the - appropriate size. - - NASM supports two special tokens in expressions, allowing - calculations to involve the current assembly position: the `$' and - `$$' tokens. `$' evaluates to the assembly position at the beginning - of the line containing the expression; so you can code an infinite - loop using `JMP $'. `$$' evaluates to the beginning of the current - section; so you can tell how far into the section you are by using - `($-$$)'. - - The arithmetic operators provided by NASM are listed here, in - increasing order of precedence. - - 3.5.1 `|': Bitwise OR Operator - - The `|' operator gives a bitwise OR, exactly as performed by the - `OR' machine instruction. Bitwise OR is the lowest-priority - arithmetic operator supported by NASM. - - 3.5.2 `^': Bitwise XOR Operator - - `^' provides the bitwise XOR operation. - - 3.5.3 `&': Bitwise AND Operator - - `&' provides the bitwise AND operation. - - 3.5.4 `<<' and `>>': Bit Shift Operators - - `<<' gives a bit-shift to the left, just as it does in C. So `5<<3' - evaluates to 5 times 8, or 40. `>>' gives a bit-shift to the right; - in NASM, such a shift is _always_ unsigned, so that the bits shifted - in from the left-hand end are filled with zero rather than a sign- - extension of the previous highest bit. - - 3.5.5 `+' and `-': Addition and Subtraction Operators - - The `+' and `-' operators do perfectly ordinary addition and - subtraction. - - 3.5.6 `*', `/', `//', `%' and `%%': Multiplication and Division - - `*' is the multiplication operator. `/' and `//' are both division - operators: `/' is unsigned division and `//' is signed division. - Similarly, `%' and `%%' provide unsigned and signed modulo operators - respectively. - - NASM, like ANSI C, provides no guarantees about the sensible - operation of the signed modulo operator. - - Since the `%' character is used extensively by the macro - preprocessor, you should ensure that both the signed and unsigned - modulo operators are followed by white space wherever they appear. - - 3.5.7 Unary Operators: `+', `-', `~', `!' and `SEG' - - The highest-priority operators in NASM's expression grammar are - those which only apply to one argument. `-' negates its operand, `+' - does nothing (it's provided for symmetry with `-'), `~' computes the - one's complement of its operand, `!' is the logical negation - operator, and `SEG' provides the segment address of its operand - (explained in more detail in section 3.6). - - 3.6 `SEG' and `WRT' - - When writing large 16-bit programs, which must be split into - multiple segments, it is often necessary to be able to refer to the - segment part of the address of a symbol. NASM supports the `SEG' - operator to perform this function. - - The `SEG' operator returns the _preferred_ segment base of a symbol, - defined as the segment base relative to which the offset of the - symbol makes sense. So the code - - mov ax,seg symbol - mov es,ax - mov bx,symbol - - will load `ES:BX' with a valid pointer to the symbol `symbol'. - - Things can be more complex than this: since 16-bit segments and - groups may overlap, you might occasionally want to refer to some - symbol using a different segment base from the preferred one. NASM - lets you do this, by the use of the `WRT' (With Reference To) - keyword. So you can do things like - - mov ax,weird_seg ; weird_seg is a segment base - mov es,ax - mov bx,symbol wrt weird_seg - - to load `ES:BX' with a different, but functionally equivalent, - pointer to the symbol `symbol'. - - NASM supports far (inter-segment) calls and jumps by means of the - syntax `call segment:offset', where `segment' and `offset' both - represent immediate values. So to call a far procedure, you could - code either of - - call (seg procedure):procedure - call weird_seg:(procedure wrt weird_seg) - - (The parentheses are included for clarity, to show the intended - parsing of the above instructions. They are not necessary in - practice.) - - NASM supports the syntax `call far procedure' as a synonym for the - first of the above usages. `JMP' works identically to `CALL' in - these examples. - - To declare a far pointer to a data item in a data segment, you must - code - - dw symbol, seg symbol - - NASM supports no convenient synonym for this, though you can always - invent one using the macro processor. - - 3.7 `STRICT': Inhibiting Optimization - - When assembling with the optimizer set to level 2 or higher (see - section 2.1.22), NASM will use size specifiers (`BYTE', `WORD', - `DWORD', `QWORD', `TWORD', `OWORD' or `YWORD'), but will give them - the smallest possible size. The keyword `STRICT' can be used to - inhibit optimization and force a particular operand to be emitted in - the specified size. For example, with the optimizer on, and in - `BITS 16' mode, - - push dword 33 - - is encoded in three bytes `66 6A 21', whereas - - push strict dword 33 - - is encoded in six bytes, with a full dword immediate operand - `66 68 21 00 00 00'. - - With the optimizer off, the same code (six bytes) is generated - whether the `STRICT' keyword was used or not. - - 3.8 Critical Expressions - - Although NASM has an optional multi-pass optimizer, there are some - expressions which must be resolvable on the first pass. These are - called _Critical Expressions_. - - The first pass is used to determine the size of all the assembled - code and data, so that the second pass, when generating all the - code, knows all the symbol addresses the code refers to. So one - thing NASM can't handle is code whose size depends on the value of a - symbol declared after the code in question. For example, - - times (label-$) db 0 - label: db 'Where am I?' - - The argument to `TIMES' in this case could equally legally evaluate - to anything at all; NASM will reject this example because it cannot - tell the size of the `TIMES' line when it first sees it. It will - just as firmly reject the slightly paradoxical code - - times (label-$+1) db 0 - label: db 'NOW where am I?' - - in which _any_ value for the `TIMES' argument is by definition - wrong! - - NASM rejects these examples by means of a concept called a _critical - expression_, which is defined to be an expression whose value is - required to be computable in the first pass, and which must - therefore depend only on symbols defined before it. The argument to - the `TIMES' prefix is a critical expression. - - 3.9 Local Labels - - NASM gives special treatment to symbols beginning with a period. A - label beginning with a single period is treated as a _local_ label, - which means that it is associated with the previous non-local label. - So, for example: - - label1 ; some code - - .loop - ; some more code - - jne .loop - ret - - label2 ; some code - - .loop - ; some more code - - jne .loop - ret - - In the above code fragment, each `JNE' instruction jumps to the line - immediately before it, because the two definitions of `.loop' are - kept separate by virtue of each being associated with the previous - non-local label. - - This form of local label handling is borrowed from the old Amiga - assembler DevPac; however, NASM goes one step further, in allowing - access to local labels from other parts of the code. This is - achieved by means of _defining_ a local label in terms of the - previous non-local label: the first definition of `.loop' above is - really defining a symbol called `label1.loop', and the second - defines a symbol called `label2.loop'. So, if you really needed to, - you could write - - label3 ; some more code - ; and some more - - jmp label1.loop - - Sometimes it is useful - in a macro, for instance - to be able to - define a label which can be referenced from anywhere but which - doesn't interfere with the normal local-label mechanism. Such a - label can't be non-local because it would interfere with subsequent - definitions of, and references to, local labels; and it can't be - local because the macro that defined it wouldn't know the label's - full name. NASM therefore introduces a third type of label, which is - probably only useful in macro definitions: if a label begins with - the special prefix `..@', then it does nothing to the local label - mechanism. So you could code - - label1: ; a non-local label - .local: ; this is really label1.local - ..@foo: ; this is a special symbol - label2: ; another non-local label - .local: ; this is really label2.local - - jmp ..@foo ; this will jump three lines up - - NASM has the capacity to define other special symbols beginning with - a double period: for example, `..start' is used to specify the entry - point in the `obj' output format (see section 7.4.6). - -Chapter 4: The NASM Preprocessor --------------------------------- - - NASM contains a powerful macro processor, which supports conditional - assembly, multi-level file inclusion, two forms of macro (single- - line and multi-line), and a `context stack' mechanism for extra - macro power. Preprocessor directives all begin with a `%' sign. - - The preprocessor collapses all lines which end with a backslash (\) - character into a single line. Thus: - - %define THIS_VERY_LONG_MACRO_NAME_IS_DEFINED_TO \ - THIS_VALUE - - will work like a single-line macro without the backslash-newline - sequence. - - 4.1 Single-Line Macros - - 4.1.1 The Normal Way: `%define' - - Single-line macros are defined using the `%define' preprocessor - directive. The definitions work in a similar way to C; so you can do - things like - - %define ctrl 0x1F & - %define param(a,b) ((a)+(a)*(b)) - - mov byte [param(2,ebx)], ctrl 'D' - - which will expand to - - mov byte [(2)+(2)*(ebx)], 0x1F & 'D' - - When the expansion of a single-line macro contains tokens which - invoke another macro, the expansion is performed at invocation time, - not at definition time. Thus the code - - %define a(x) 1+b(x) - %define b(x) 2*x - - mov ax,a(8) - - will evaluate in the expected way to `mov ax,1+2*8', even though the - macro `b' wasn't defined at the time of definition of `a'. - - Macros defined with `%define' are case sensitive: after - `%define foo bar', only `foo' will expand to `bar': `Foo' or `FOO' - will not. By using `%idefine' instead of `%define' (the `i' stands - for `insensitive') you can define all the case variants of a macro - at once, so that `%idefine foo bar' would cause `foo', `Foo', `FOO', - `fOO' and so on all to expand to `bar'. - - There is a mechanism which detects when a macro call has occurred as - a result of a previous expansion of the same macro, to guard against - circular references and infinite loops. If this happens, the - preprocessor will only expand the first occurrence of the macro. - Hence, if you code - - %define a(x) 1+a(x) - - mov ax,a(3) - - the macro `a(3)' will expand once, becoming `1+a(3)', and will then - expand no further. This behaviour can be useful: see section 9.1 for - an example of its use. - - You can overload single-line macros: if you write - - %define foo(x) 1+x - %define foo(x,y) 1+x*y - - the preprocessor will be able to handle both types of macro call, by - counting the parameters you pass; so `foo(3)' will become `1+3' - whereas `foo(ebx,2)' will become `1+ebx*2'. However, if you define - - %define foo bar - - then no other definition of `foo' will be accepted: a macro with no - parameters prohibits the definition of the same name as a macro - _with_ parameters, and vice versa. - - This doesn't prevent single-line macros being _redefined_: you can - perfectly well define a macro with - - %define foo bar - - and then re-define it later in the same source file with - - %define foo baz - - Then everywhere the macro `foo' is invoked, it will be expanded - according to the most recent definition. This is particularly useful - when defining single-line macros with `%assign' (see section 4.1.7). - - You can pre-define single-line macros using the `-d' option on the - NASM command line: see section 2.1.18. - - 4.1.2 Resolving `%define': `%xdefine' - - To have a reference to an embedded single-line macro resolved at the - time that the embedding macro is _defined_, as opposed to when the - embedding macro is _expanded_, you need a different mechanism to the - one offered by `%define'. The solution is to use `%xdefine', or it's - case-insensitive counterpart `%ixdefine'. - - Suppose you have the following code: - - %define isTrue 1 - %define isFalse isTrue - %define isTrue 0 - - val1: db isFalse - - %define isTrue 1 - - val2: db isFalse - - In this case, `val1' is equal to 0, and `val2' is equal to 1. This - is because, when a single-line macro is defined using `%define', it - is expanded only when it is called. As `isFalse' expands to - `isTrue', the expansion will be the current value of `isTrue'. The - first time it is called that is 0, and the second time it is 1. - - If you wanted `isFalse' to expand to the value assigned to the - embedded macro `isTrue' at the time that `isFalse' was defined, you - need to change the above code to use `%xdefine'. - - %xdefine isTrue 1 - %xdefine isFalse isTrue - %xdefine isTrue 0 - - val1: db isFalse - - %xdefine isTrue 1 - - val2: db isFalse - - Now, each time that `isFalse' is called, it expands to 1, as that is - what the embedded macro `isTrue' expanded to at the time that - `isFalse' was defined. - - 4.1.3 Macro Indirection: `%[...]' - - The `%[...]' construct can be used to expand macros in contexts - where macro expansion would otherwise not occur, including in the - names other macros. For example, if you have a set of macros named - `Foo16', `Foo32' and `Foo64', you could write: - - mov ax,Foo%[__BITS__] ; The Foo value - - to use the builtin macro `__BITS__' (see section 4.11.5) to - automatically select between them. Similarly, the two statements: - - %xdefine Bar Quux ; Expands due to %xdefine - %define Bar %[Quux] ; Expands due to %[...] - - have, in fact, exactly the same effect. - - `%[...]' concatenates to adjacent tokens in the same way that multi- - line macro parameters do, see section 4.3.8 for details. - - 4.1.4 Concatenating Single Line Macro Tokens: `%+' - - Individual tokens in single line macros can be concatenated, to - produce longer tokens for later processing. This can be useful if - there are several similar macros that perform similar functions. - - Please note that a space is required after `%+', in order to - disambiguate it from the syntax `%+1' used in multiline macros. - - As an example, consider the following: - - %define BDASTART 400h ; Start of BIOS data area - - struc tBIOSDA ; its structure - .COM1addr RESW 1 - .COM2addr RESW 1 - ; ..and so on - endstruc - - Now, if we need to access the elements of tBIOSDA in different - places, we can end up with: - - mov ax,BDASTART + tBIOSDA.COM1addr - mov bx,BDASTART + tBIOSDA.COM2addr - - This will become pretty ugly (and tedious) if used in many places, - and can be reduced in size significantly by using the following - macro: - - ; Macro to access BIOS variables by their names (from tBDA): - - %define BDA(x) BDASTART + tBIOSDA. %+ x - - Now the above code can be written as: - - mov ax,BDA(COM1addr) - mov bx,BDA(COM2addr) - - Using this feature, we can simplify references to a lot of macros - (and, in turn, reduce typing errors). - - 4.1.5 The Macro Name Itself: `%?' and `%??' - - The special symbols `%?' and `%??' can be used to reference the - macro name itself inside a macro expansion, this is supported for - both single-and multi-line macros. `%?' refers to the macro name as - _invoked_, whereas `%??' refers to the macro name as _declared_. The - two are always the same for case-sensitive macros, but for case- - insensitive macros, they can differ. - - For example: - - %idefine Foo mov %?,%?? - - foo - FOO - - will expand to: - - mov foo,Foo - mov FOO,Foo - - The sequence: - - %idefine keyword $%? - - can be used to make a keyword "disappear", for example in case a new - instruction has been used as a label in older code. For example: - - %idefine pause $%? ; Hide the PAUSE instruction - - 4.1.6 Undefining Single-Line Macros: `%undef' - - Single-line macros can be removed with the `%undef' directive. For - example, the following sequence: - - %define foo bar - %undef foo - - mov eax, foo - - will expand to the instruction `mov eax, foo', since after `%undef' - the macro `foo' is no longer defined. - - Macros that would otherwise be pre-defined can be undefined on the - command-line using the `-u' option on the NASM command line: see - section 2.1.19. - - 4.1.7 Preprocessor Variables: `%assign' - - An alternative way to define single-line macros is by means of the - `%assign' command (and its case-insensitive counterpart `%iassign', - which differs from `%assign' in exactly the same way that `%idefine' - differs from `%define'). - - `%assign' is used to define single-line macros which take no - parameters and have a numeric value. This value can be specified in - the form of an expression, and it will be evaluated once, when the - `%assign' directive is processed. - - Like `%define', macros defined using `%assign' can be re-defined - later, so you can do things like - - %assign i i+1 - - to increment the numeric value of a macro. - - `%assign' is useful for controlling the termination of `%rep' - preprocessor loops: see section 4.5 for an example of this. Another - use for `%assign' is given in section 8.4 and section 9.1. - - The expression passed to `%assign' is a critical expression (see - section 3.8), and must also evaluate to a pure number (rather than a - relocatable reference such as a code or data address, or anything - involving a register). - - 4.1.8 Defining Strings: `%defstr' - - `%defstr', and its case-insensitive counterpart `%idefstr', define - or redefine a single-line macro without parameters but converts the - entire right-hand side, after macro expansion, to a quoted string - before definition. - - For example: - - %defstr test TEST - - is equivalent to - - %define test 'TEST' - - This can be used, for example, with the `%!' construct (see section - 4.10.2): - - %defstr PATH %!PATH ; The operating system PATH variable - - 4.1.9 Defining Tokens: `%deftok' - - `%deftok', and its case-insensitive counterpart `%ideftok', define - or redefine a single-line macro without parameters but converts the - second parameter, after string conversion, to a sequence of tokens. - - For example: - - %deftok test 'TEST' - - is equivalent to - - %define test TEST - - 4.2 String Manipulation in Macros - - It's often useful to be able to handle strings in macros. NASM - supports a few simple string handling macro operators from which - more complex operations can be constructed. - - All the string operators define or redefine a value (either a string - or a numeric value) to a single-line macro. When producing a string - value, it may change the style of quoting of the input string or - strings, and possibly use `\'-escapes inside ``'-quoted strings. - - 4.2.1 Concatenating Strings: `%strcat' - - The `%strcat' operator concatenates quoted strings and assign them - to a single-line macro. - - For example: - - %strcat alpha "Alpha: ", '12" screen' - - ... would assign the value `'Alpha: 12" screen'' to `alpha'. - Similarly: - - %strcat beta '"foo"\', "'bar'" - - ... would assign the value ``"foo"\\'bar'`' to `beta'. - - The use of commas to separate strings is permitted but optional. - - 4.2.2 String Length: `%strlen' - - The `%strlen' operator assigns the length of a string to a macro. - For example: - - %strlen charcnt 'my string' - - In this example, `charcnt' would receive the value 9, just as if an - `%assign' had been used. In this example, `'my string'' was a - literal string but it could also have been a single-line macro that - expands to a string, as in the following example: - - %define sometext 'my string' - %strlen charcnt sometext - - As in the first case, this would result in `charcnt' being assigned - the value of 9. - - 4.2.3 Extracting Substrings: `%substr' - - Individual letters or substrings in strings can be extracted using - the `%substr' operator. An example of its use is probably more - useful than the description: - - %substr mychar 'xyzw' 1 ; equivalent to %define mychar 'x' - %substr mychar 'xyzw' 2 ; equivalent to %define mychar 'y' - %substr mychar 'xyzw' 3 ; equivalent to %define mychar 'z' - %substr mychar 'xyzw' 2,2 ; equivalent to %define mychar 'yz' - %substr mychar 'xyzw' 2,-1 ; equivalent to %define mychar 'yzw' - %substr mychar 'xyzw' 2,-2 ; equivalent to %define mychar 'yz' - - As with `%strlen' (see section 4.2.2), the first parameter is the - single-line macro to be created and the second is the string. The - third parameter specifies the first character to be selected, and - the optional fourth parameter preceeded by comma) is the length. - Note that the first index is 1, not 0 and the last index is equal to - the value that `%strlen' would assign given the same string. Index - values out of range result in an empty string. A negative length - means "until N-1 characters before the end of string", i.e. `-1' - means until end of string, `-2' until one character before, etc. - - 4.3 Multi-Line Macros: `%macro' - - Multi-line macros are much more like the type of macro seen in MASM - and TASM: a multi-line macro definition in NASM looks something like - this. - - %macro prologue 1 - - push ebp - mov ebp,esp - sub esp,%1 - - %endmacro - - This defines a C-like function prologue as a macro: so you would - invoke the macro with a call such as - - myfunc: prologue 12 - - which would expand to the three lines of code - - myfunc: push ebp - mov ebp,esp - sub esp,12 - - The number `1' after the macro name in the `%macro' line defines the - number of parameters the macro `prologue' expects to receive. The - use of `%1' inside the macro definition refers to the first - parameter to the macro call. With a macro taking more than one - parameter, subsequent parameters would be referred to as `%2', `%3' - and so on. - - Multi-line macros, like single-line macros, are case-sensitive, - unless you define them using the alternative directive `%imacro'. - - If you need to pass a comma as _part_ of a parameter to a multi-line - macro, you can do that by enclosing the entire parameter in braces. - So you could code things like - - %macro silly 2 - - %2: db %1 - - %endmacro - - silly 'a', letter_a ; letter_a: db 'a' - silly 'ab', string_ab ; string_ab: db 'ab' - silly {13,10}, crlf ; crlf: db 13,10 - - 4.3.1 Recursive Multi-Line Macros: `%rmacro' - - A multi-line macro cannot be referenced within itself, in order to - prevent accidental infinite recursion. - - Recursive multi-line macros allow for self-referencing, with the - caveat that the user is aware of the existence, use and purpose of - recursive multi-line macros. There is also a generous, but sane, - upper limit to the number of recursions, in order to prevent run- - away memory consumption in case of accidental infinite recursion. - - As with non-recursive multi-line macros, recursive multi-line macros - are case-sensitive, unless you define them using the alternative - directive `%irmacro'. - - 4.3.2 Overloading Multi-Line Macros - - As with single-line macros, multi-line macros can be overloaded by - defining the same macro name several times with different numbers of - parameters. This time, no exception is made for macros with no - parameters at all. So you could define - - %macro prologue 0 - - push ebp - mov ebp,esp - - %endmacro - - to define an alternative form of the function prologue which - allocates no local stack space. - - Sometimes, however, you might want to `overload' a machine - instruction; for example, you might want to define - - %macro push 2 - - push %1 - push %2 - - %endmacro - - so that you could code - - push ebx ; this line is not a macro call - push eax,ecx ; but this one is - - Ordinarily, NASM will give a warning for the first of the above two - lines, since `push' is now defined to be a macro, and is being - invoked with a number of parameters for which no definition has been - given. The correct code will still be generated, but the assembler - will give a warning. This warning can be disabled by the use of the - `-w-macro-params' command-line option (see section 2.1.24). - - 4.3.3 Macro-Local Labels - - NASM allows you to define labels within a multi-line macro - definition in such a way as to make them local to the macro call: so - calling the same macro multiple times will use a different label - each time. You do this by prefixing `%%' to the label name. So you - can invent an instruction which executes a `RET' if the `Z' flag is - set by doing this: - - %macro retz 0 - - jnz %%skip - ret - %%skip: - - %endmacro - - You can call this macro as many times as you want, and every time - you call it NASM will make up a different `real' name to substitute - for the label `%%skip'. The names NASM invents are of the form - `..@2345.skip', where the number 2345 changes with every macro call. - The `..@' prefix prevents macro-local labels from interfering with - the local label mechanism, as described in section 3.9. You should - avoid defining your own labels in this form (the `..@' prefix, then - a number, then another period) in case they interfere with macro- - local labels. - - 4.3.4 Greedy Macro Parameters - - Occasionally it is useful to define a macro which lumps its entire - command line into one parameter definition, possibly after - extracting one or two smaller parameters from the front. An example - might be a macro to write a text string to a file in MS-DOS, where - you might want to be able to write - - writefile [filehandle],"hello, world",13,10 - - NASM allows you to define the last parameter of a macro to be - _greedy_, meaning that if you invoke the macro with more parameters - than it expects, all the spare parameters get lumped into the last - defined one along with the separating commas. So if you code: - - %macro writefile 2+ - - jmp %%endstr - %%str: db %2 - %%endstr: - mov dx,%%str - mov cx,%%endstr-%%str - mov bx,%1 - mov ah,0x40 - int 0x21 - - %endmacro - - then the example call to `writefile' above will work as expected: - the text before the first comma, `[filehandle]', is used as the - first macro parameter and expanded when `%1' is referred to, and all - the subsequent text is lumped into `%2' and placed after the `db'. - - The greedy nature of the macro is indicated to NASM by the use of - the `+' sign after the parameter count on the `%macro' line. - - If you define a greedy macro, you are effectively telling NASM how - it should expand the macro given _any_ number of parameters from the - actual number specified up to infinity; in this case, for example, - NASM now knows what to do when it sees a call to `writefile' with 2, - 3, 4 or more parameters. NASM will take this into account when - overloading macros, and will not allow you to define another form of - `writefile' taking 4 parameters (for example). - - Of course, the above macro could have been implemented as a non- - greedy macro, in which case the call to it would have had to look - like - - writefile [filehandle], {"hello, world",13,10} - - NASM provides both mechanisms for putting commas in macro - parameters, and you choose which one you prefer for each macro - definition. - - See section 6.3.1 for a better way to write the above macro. - - 4.3.5 Default Macro Parameters - - NASM also allows you to define a multi-line macro with a _range_ of - allowable parameter counts. If you do this, you can specify defaults - for omitted parameters. So, for example: - - %macro die 0-1 "Painful program death has occurred." - - writefile 2,%1 - mov ax,0x4c01 - int 0x21 - - %endmacro - - This macro (which makes use of the `writefile' macro defined in - section 4.3.4) can be called with an explicit error message, which - it will display on the error output stream before exiting, or it can - be called with no parameters, in which case it will use the default - error message supplied in the macro definition. - - In general, you supply a minimum and maximum number of parameters - for a macro of this type; the minimum number of parameters are then - required in the macro call, and then you provide defaults for the - optional ones. So if a macro definition began with the line - - %macro foobar 1-3 eax,[ebx+2] - - then it could be called with between one and three parameters, and - `%1' would always be taken from the macro call. `%2', if not - specified by the macro call, would default to `eax', and `%3' if not - specified would default to `[ebx+2]'. - - You can provide extra information to a macro by providing too many - default parameters: - - %macro quux 1 something - - This will trigger a warning by default; see section 2.1.24 for more - information. When `quux' is invoked, it receives not one but two - parameters. `something' can be referred to as `%2'. The difference - between passing `something' this way and writing `something' in the - macro body is that with this way `something' is evaluated when the - macro is defined, not when it is expanded. - - You may omit parameter defaults from the macro definition, in which - case the parameter default is taken to be blank. This can be useful - for macros which can take a variable number of parameters, since the - `%0' token (see section 4.3.6) allows you to determine how many - parameters were really passed to the macro call. - - This defaulting mechanism can be combined with the greedy-parameter - mechanism; so the `die' macro above could be made more powerful, and - more useful, by changing the first line of the definition to - - %macro die 0-1+ "Painful program death has occurred.",13,10 - - The maximum parameter count can be infinite, denoted by `*'. In this - case, of course, it is impossible to provide a _full_ set of default - parameters. Examples of this usage are shown in section 4.3.7. - - 4.3.6 `%0': Macro Parameter Counter - - The parameter reference `%0' will return a numeric constant giving - the number of parameters received, that is, if `%0' is n then `%'n - is the last parameter. `%0' is mostly useful for macros that can - take a variable number of parameters. It can be used as an argument - to `%rep' (see section 4.5) in order to iterate through all the - parameters of a macro. Examples are given in section 4.3.7. - - 4.3.7 `%rotate': Rotating Macro Parameters - - Unix shell programmers will be familiar with the `shift' shell - command, which allows the arguments passed to a shell script - (referenced as `$1', `$2' and so on) to be moved left by one place, - so that the argument previously referenced as `$2' becomes available - as `$1', and the argument previously referenced as `$1' is no longer - available at all. - - NASM provides a similar mechanism, in the form of `%rotate'. As its - name suggests, it differs from the Unix `shift' in that no - parameters are lost: parameters rotated off the left end of the - argument list reappear on the right, and vice versa. - - `%rotate' is invoked with a single numeric argument (which may be an - expression). The macro parameters are rotated to the left by that - many places. If the argument to `%rotate' is negative, the macro - parameters are rotated to the right. - - So a pair of macros to save and restore a set of registers might - work as follows: - - %macro multipush 1-* - - %rep %0 - push %1 - %rotate 1 - %endrep - - %endmacro - - This macro invokes the `PUSH' instruction on each of its arguments - in turn, from left to right. It begins by pushing its first - argument, `%1', then invokes `%rotate' to move all the arguments one - place to the left, so that the original second argument is now - available as `%1'. Repeating this procedure as many times as there - were arguments (achieved by supplying `%0' as the argument to - `%rep') causes each argument in turn to be pushed. - - Note also the use of `*' as the maximum parameter count, indicating - that there is no upper limit on the number of parameters you may - supply to the `multipush' macro. - - It would be convenient, when using this macro, to have a `POP' - equivalent, which _didn't_ require the arguments to be given in - reverse order. Ideally, you would write the `multipush' macro call, - then cut-and-paste the line to where the pop needed to be done, and - change the name of the called macro to `multipop', and the macro - would take care of popping the registers in the opposite order from - the one in which they were pushed. - - This can be done by the following definition: - - %macro multipop 1-* - - %rep %0 - %rotate -1 - pop %1 - %endrep - - %endmacro - - This macro begins by rotating its arguments one place to the - _right_, so that the original _last_ argument appears as `%1'. This - is then popped, and the arguments are rotated right again, so the - second-to-last argument becomes `%1'. Thus the arguments are - iterated through in reverse order. - - 4.3.8 Concatenating Macro Parameters - - NASM can concatenate macro parameters and macro indirection - constructs on to other text surrounding them. This allows you to - declare a family of symbols, for example, in a macro definition. If, - for example, you wanted to generate a table of key codes along with - offsets into the table, you could code something like - - %macro keytab_entry 2 - - keypos%1 equ $-keytab - db %2 - - %endmacro - - keytab: - keytab_entry F1,128+1 - keytab_entry F2,128+2 - keytab_entry Return,13 - - which would expand to - - keytab: - keyposF1 equ $-keytab - db 128+1 - keyposF2 equ $-keytab - db 128+2 - keyposReturn equ $-keytab - db 13 - - You can just as easily concatenate text on to the other end of a - macro parameter, by writing `%1foo'. - - If you need to append a _digit_ to a macro parameter, for example - defining labels `foo1' and `foo2' when passed the parameter `foo', - you can't code `%11' because that would be taken as the eleventh - macro parameter. Instead, you must code `%{1}1', which will separate - the first `1' (giving the number of the macro parameter) from the - second (literal text to be concatenated to the parameter). - - This concatenation can also be applied to other preprocessor in-line - objects, such as macro-local labels (section 4.3.3) and context- - local labels (section 4.7.2). In all cases, ambiguities in syntax - can be resolved by enclosing everything after the `%' sign and - before the literal text in braces: so `%{%foo}bar' concatenates the - text `bar' to the end of the real name of the macro-local label - `%%foo'. (This is unnecessary, since the form NASM uses for the real - names of macro-local labels means that the two usages `%{%foo}bar' - and `%%foobar' would both expand to the same thing anyway; - nevertheless, the capability is there.) - - The single-line macro indirection construct, `%[...]' (section - 4.1.3), behaves the same way as macro parameters for the purpose of - concatenation. - - See also the `%+' operator, section 4.1.4. - - 4.3.9 Condition Codes as Macro Parameters - - NASM can give special treatment to a macro parameter which contains - a condition code. For a start, you can refer to the macro parameter - `%1' by means of the alternative syntax `%+1', which informs NASM - that this macro parameter is supposed to contain a condition code, - and will cause the preprocessor to report an error message if the - macro is called with a parameter which is _not_ a valid condition - code. - - Far more usefully, though, you can refer to the macro parameter by - means of `%-1', which NASM will expand as the _inverse_ condition - code. So the `retz' macro defined in section 4.3.3 can be replaced - by a general conditional-return macro like this: - - %macro retc 1 - - j%-1 %%skip - ret - %%skip: - - %endmacro - - This macro can now be invoked using calls like `retc ne', which will - cause the conditional-jump instruction in the macro expansion to - come out as `JE', or `retc po' which will make the jump a `JPE'. - - The `%+1' macro-parameter reference is quite happy to interpret the - arguments `CXZ' and `ECXZ' as valid condition codes; however, `%-1' - will report an error if passed either of these, because no inverse - condition code exists. - -4.3.10 Disabling Listing Expansion - - When NASM is generating a listing file from your program, it will - generally expand multi-line macros by means of writing the macro - call and then listing each line of the expansion. This allows you to - see which instructions in the macro expansion are generating what - code; however, for some macros this clutters the listing up - unnecessarily. - - NASM therefore provides the `.nolist' qualifier, which you can - include in a macro definition to inhibit the expansion of the macro - in the listing file. The `.nolist' qualifier comes directly after - the number of parameters, like this: - - %macro foo 1.nolist - - Or like this: - - %macro bar 1-5+.nolist a,b,c,d,e,f,g,h - -4.3.11 Undefining Multi-Line Macros: `%unmacro' - - Multi-line macros can be removed with the `%unmacro' directive. - Unlike the `%undef' directive, however, `%unmacro' takes an argument - specification, and will only remove exact matches with that argument - specification. - - For example: - - %macro foo 1-3 - ; Do something - %endmacro - %unmacro foo 1-3 - - removes the previously defined macro `foo', but - - %macro bar 1-3 - ; Do something - %endmacro - %unmacro bar 1 - - does _not_ remove the macro `bar', since the argument specification - does not match exactly. - -4.3.12 Exiting Multi-Line Macros: `%exitmacro' - - Multi-line macro expansions can be arbitrarily terminated with the - `%exitmacro' directive. - - For example: - - %macro foo 1-3 - ; Do something - %if<condition> - %exitmacro - %endif - ; Do something - %endmacro - - 4.4 Conditional Assembly - - Similarly to the C preprocessor, NASM allows sections of a source - file to be assembled only if certain conditions are met. The general - syntax of this feature looks like this: - - %if<condition> - ; some code which only appears if <condition> is met - %elif<condition2> - ; only appears if <condition> is not met but <condition2> is - %else - ; this appears if neither <condition> nor <condition2> was met - %endif - - The inverse forms `%ifn' and `%elifn' are also supported. - - The `%else' clause is optional, as is the `%elif' clause. You can - have more than one `%elif' clause as well. - - There are a number of variants of the `%if' directive. Each has its - corresponding `%elif', `%ifn', and `%elifn' directives; for example, - the equivalents to the `%ifdef' directive are `%elifdef', `%ifndef', - and `%elifndef'. - - 4.4.1 `%ifdef': Testing Single-Line Macro Existence - - Beginning a conditional-assembly block with the line `%ifdef MACRO' - will assemble the subsequent code if, and only if, a single-line - macro called `MACRO' is defined. If not, then the `%elif' and - `%else' blocks (if any) will be processed instead. - - For example, when debugging a program, you might want to write code - such as - - ; perform some function - %ifdef DEBUG - writefile 2,"Function performed successfully",13,10 - %endif - ; go and do something else - - Then you could use the command-line option `-dDEBUG' to create a - version of the program which produced debugging messages, and remove - the option to generate the final release version of the program. - - You can test for a macro _not_ being defined by using `%ifndef' - instead of `%ifdef'. You can also test for macro definitions in - `%elif' blocks by using `%elifdef' and `%elifndef'. - - 4.4.2 `%ifmacro': Testing Multi-Line Macro Existence - - The `%ifmacro' directive operates in the same way as the `%ifdef' - directive, except that it checks for the existence of a multi-line - macro. - - For example, you may be working with a large project and not have - control over the macros in a library. You may want to create a macro - with one name if it doesn't already exist, and another name if one - with that name does exist. - - The `%ifmacro' is considered true if defining a macro with the given - name and number of arguments would cause a definitions conflict. For - example: - - %ifmacro MyMacro 1-3 - - %error "MyMacro 1-3" causes a conflict with an existing macro. - - %else - - %macro MyMacro 1-3 - - ; insert code to define the macro - - %endmacro - - %endif - - This will create the macro "MyMacro 1-3" if no macro already exists - which would conflict with it, and emits a warning if there would be - a definition conflict. - - You can test for the macro not existing by using the `%ifnmacro' - instead of `%ifmacro'. Additional tests can be performed in `%elif' - blocks by using `%elifmacro' and `%elifnmacro'. - - 4.4.3 `%ifctx': Testing the Context Stack - - The conditional-assembly construct `%ifctx' will cause the - subsequent code to be assembled if and only if the top context on - the preprocessor's context stack has the same name as one of the - arguments. As with `%ifdef', the inverse and `%elif' forms - `%ifnctx', `%elifctx' and `%elifnctx' are also supported. - - For more details of the context stack, see section 4.7. For a sample - use of `%ifctx', see section 4.7.5. - - 4.4.4 `%if': Testing Arbitrary Numeric Expressions - - The conditional-assembly construct `%if expr' will cause the - subsequent code to be assembled if and only if the value of the - numeric expression `expr' is non-zero. An example of the use of this - feature is in deciding when to break out of a `%rep' preprocessor - loop: see section 4.5 for a detailed example. - - The expression given to `%if', and its counterpart `%elif', is a - critical expression (see section 3.8). - - `%if' extends the normal NASM expression syntax, by providing a set - of relational operators which are not normally available in - expressions. The operators `=', `<', `>', `<=', `>=' and `<>' test - equality, less-than, greater-than, less-or-equal, greater-or-equal - and not-equal respectively. The C-like forms `==' and `!=' are - supported as alternative forms of `=' and `<>'. In addition, low- - priority logical operators `&&', `^^' and `||' are provided, - supplying logical AND, logical XOR and logical OR. These work like - the C logical operators (although C has no logical XOR), in that - they always return either 0 or 1, and treat any non-zero input as 1 - (so that `^^', for example, returns 1 if exactly one of its inputs - is zero, and 0 otherwise). The relational operators also return 1 - for true and 0 for false. - - Like other `%if' constructs, `%if' has a counterpart `%elif', and - negative forms `%ifn' and `%elifn'. - - 4.4.5 `%ifidn' and `%ifidni': Testing Exact Text Identity - - The construct `%ifidn text1,text2' will cause the subsequent code to - be assembled if and only if `text1' and `text2', after expanding - single-line macros, are identical pieces of text. Differences in - white space are not counted. - - `%ifidni' is similar to `%ifidn', but is case-insensitive. - - For example, the following macro pushes a register or number on the - stack, and allows you to treat `IP' as a real register: - - %macro pushparam 1 - - %ifidni %1,ip - call %%label - %%label: - %else - push %1 - %endif - - %endmacro - - Like other `%if' constructs, `%ifidn' has a counterpart `%elifidn', - and negative forms `%ifnidn' and `%elifnidn'. Similarly, `%ifidni' - has counterparts `%elifidni', `%ifnidni' and `%elifnidni'. - - 4.4.6 `%ifid', `%ifnum', `%ifstr': Testing Token Types - - Some macros will want to perform different tasks depending on - whether they are passed a number, a string, or an identifier. For - example, a string output macro might want to be able to cope with - being passed either a string constant or a pointer to an existing - string. - - The conditional assembly construct `%ifid', taking one parameter - (which may be blank), assembles the subsequent code if and only if - the first token in the parameter exists and is an identifier. - `%ifnum' works similarly, but tests for the token being a numeric - constant; `%ifstr' tests for it being a string. - - For example, the `writefile' macro defined in section 4.3.4 can be - extended to take advantage of `%ifstr' in the following fashion: - - %macro writefile 2-3+ - - %ifstr %2 - jmp %%endstr - %if %0 = 3 - %%str: db %2,%3 - %else - %%str: db %2 - %endif - %%endstr: mov dx,%%str - mov cx,%%endstr-%%str - %else - mov dx,%2 - mov cx,%3 - %endif - mov bx,%1 - mov ah,0x40 - int 0x21 - - %endmacro - - Then the `writefile' macro can cope with being called in either of - the following two ways: - - writefile [file], strpointer, length - writefile [file], "hello", 13, 10 - - In the first, `strpointer' is used as the address of an already- - declared string, and `length' is used as its length; in the second, - a string is given to the macro, which therefore declares it itself - and works out the address and length for itself. - - Note the use of `%if' inside the `%ifstr': this is to detect whether - the macro was passed two arguments (so the string would be a single - string constant, and `db %2' would be adequate) or more (in which - case, all but the first two would be lumped together into `%3', and - `db %2,%3' would be required). - - The usual `%elif'..., `%ifn'..., and `%elifn'... versions exist for - each of `%ifid', `%ifnum' and `%ifstr'. - - 4.4.7 `%iftoken': Test for a Single Token - - Some macros will want to do different things depending on if it is - passed a single token (e.g. paste it to something else using `%+') - versus a multi-token sequence. - - The conditional assembly construct `%iftoken' assembles the - subsequent code if and only if the expanded parameters consist of - exactly one token, possibly surrounded by whitespace. - - For example: - - %iftoken 1 - - will assemble the subsequent code, but - - %iftoken -1 - - will not, since `-1' contains two tokens: the unary minus operator - `-', and the number `1'. - - The usual `%eliftoken', `%ifntoken', and `%elifntoken' variants are - also provided. - - 4.4.8 `%ifempty': Test for Empty Expansion - - The conditional assembly construct `%ifempty' assembles the - subsequent code if and only if the expanded parameters do not - contain any tokens at all, whitespace excepted. - - The usual `%elifempty', `%ifnempty', and `%elifnempty' variants are - also provided. - - 4.5 Preprocessor Loops: `%rep' - - NASM's `TIMES' prefix, though useful, cannot be used to invoke a - multi-line macro multiple times, because it is processed by NASM - after macros have already been expanded. Therefore NASM provides - another form of loop, this time at the preprocessor level: `%rep'. - - The directives `%rep' and `%endrep' (`%rep' takes a numeric - argument, which can be an expression; `%endrep' takes no arguments) - can be used to enclose a chunk of code, which is then replicated as - many times as specified by the preprocessor: - - %assign i 0 - %rep 64 - inc word [table+2*i] - %assign i i+1 - %endrep - - This will generate a sequence of 64 `INC' instructions, incrementing - every word of memory from `[table]' to `[table+126]'. - - For more complex termination conditions, or to break out of a repeat - loop part way along, you can use the `%exitrep' directive to - terminate the loop, like this: - - fibonacci: - %assign i 0 - %assign j 1 - %rep 100 - %if j > 65535 - %exitrep - %endif - dw j - %assign k j+i - %assign i j - %assign j k - %endrep - - fib_number equ ($-fibonacci)/2 - - This produces a list of all the Fibonacci numbers that will fit in - 16 bits. Note that a maximum repeat count must still be given to - `%rep'. This is to prevent the possibility of NASM getting into an - infinite loop in the preprocessor, which (on multitasking or multi- - user systems) would typically cause all the system memory to be - gradually used up and other applications to start crashing. - - 4.6 Source Files and Dependencies - - These commands allow you to split your sources into multiple files. - - 4.6.1 `%include': Including Other Files - - Using, once again, a very similar syntax to the C preprocessor, - NASM's preprocessor lets you include other source files into your - code. This is done by the use of the `%include' directive: - - %include "macros.mac" - - will include the contents of the file `macros.mac' into the source - file containing the `%include' directive. - - Include files are searched for in the current directory (the - directory you're in when you run NASM, as opposed to the location of - the NASM executable or the location of the source file), plus any - directories specified on the NASM command line using the `-i' - option. - - The standard C idiom for preventing a file being included more than - once is just as applicable in NASM: if the file `macros.mac' has the - form - - %ifndef MACROS_MAC - %define MACROS_MAC - ; now define some macros - %endif - - then including the file more than once will not cause errors, - because the second time the file is included nothing will happen - because the macro `MACROS_MAC' will already be defined. - - You can force a file to be included even if there is no `%include' - directive that explicitly includes it, by using the `-p' option on - the NASM command line (see section 2.1.17). - - 4.6.2 `%pathsearch': Search the Include Path - - The `%pathsearch' directive takes a single-line macro name and a - filename, and declare or redefines the specified single-line macro - to be the include-path-resolved version of the filename, if the file - exists (otherwise, it is passed unchanged.) - - For example, - - %pathsearch MyFoo "foo.bin" - - ... with `-Ibins/' in the include path may end up defining the macro - `MyFoo' to be `"bins/foo.bin"'. - - 4.6.3 `%depend': Add Dependent Files - - The `%depend' directive takes a filename and adds it to the list of - files to be emitted as dependency generation when the `-M' options - and its relatives (see section 2.1.4) are used. It produces no - output. - - This is generally used in conjunction with `%pathsearch'. For - example, a simplified version of the standard macro wrapper for the - `INCBIN' directive looks like: - - %imacro incbin 1-2+ 0 - %pathsearch dep %1 - %depend dep - incbin dep,%2 - %endmacro - - This first resolves the location of the file into the macro `dep', - then adds it to the dependency lists, and finally issues the - assembler-level `INCBIN' directive. - - 4.6.4 `%use': Include Standard Macro Package - - The `%use' directive is similar to `%include', but rather than - including the contents of a file, it includes a named standard macro - package. The standard macro packages are part of NASM, and are - described in chapter 5. - - Unlike the `%include' directive, package names for the `%use' - directive do not require quotes, but quotes are permitted. In NASM - 2.04 and 2.05 the unquoted form would be macro-expanded; this is no - longer true. Thus, the following lines are equivalent: - - %use altreg - %use 'altreg' - - Standard macro packages are protected from multiple inclusion. When - a standard macro package is used, a testable single-line macro of - the form `__USE_'_package_`__' is also defined, see section 4.11.8. - - 4.7 The Context Stack - - Having labels that are local to a macro definition is sometimes not - quite powerful enough: sometimes you want to be able to share labels - between several macro calls. An example might be a `REPEAT' ... - `UNTIL' loop, in which the expansion of the `REPEAT' macro would - need to be able to refer to a label which the `UNTIL' macro had - defined. However, for such a macro you would also want to be able to - nest these loops. - - NASM provides this level of power by means of a _context stack_. The - preprocessor maintains a stack of _contexts_, each of which is - characterized by a name. You add a new context to the stack using - the `%push' directive, and remove one using `%pop'. You can define - labels that are local to a particular context on the stack. - - 4.7.1 `%push' and `%pop': Creating and Removing Contexts - - The `%push' directive is used to create a new context and place it - on the top of the context stack. `%push' takes an optional argument, - which is the name of the context. For example: - - %push foobar - - This pushes a new context called `foobar' on the stack. You can have - several contexts on the stack with the same name: they can still be - distinguished. If no name is given, the context is unnamed (this is - normally used when both the `%push' and the `%pop' are inside a - single macro definition.) - - The directive `%pop', taking one optional argument, removes the top - context from the context stack and destroys it, along with any - labels associated with it. If an argument is given, it must match - the name of the current context, otherwise it will issue an error. - - 4.7.2 Context-Local Labels - - Just as the usage `%%foo' defines a label which is local to the - particular macro call in which it is used, the usage `%$foo' is used - to define a label which is local to the context on the top of the - context stack. So the `REPEAT' and `UNTIL' example given above could - be implemented by means of: - - %macro repeat 0 - - %push repeat - %$begin: - - %endmacro - - %macro until 1 - - j%-1 %$begin - %pop - - %endmacro - - and invoked by means of, for example, - - mov cx,string - repeat - add cx,3 - scasb - until e - - which would scan every fourth byte of a string in search of the byte - in `AL'. - - If you need to define, or access, labels local to the context - _below_ the top one on the stack, you can use `%$$foo', or `%$$$foo' - for the context below that, and so on. - - 4.7.3 Context-Local Single-Line Macros - - NASM also allows you to define single-line macros which are local to - a particular context, in just the same way: - - %define %$localmac 3 - - will define the single-line macro `%$localmac' to be local to the - top context on the stack. Of course, after a subsequent `%push', it - can then still be accessed by the name `%$$localmac'. - - 4.7.4 `%repl': Renaming a Context - - If you need to change the name of the top context on the stack (in - order, for example, to have it respond differently to `%ifctx'), you - can execute a `%pop' followed by a `%push'; but this will have the - side effect of destroying all context-local labels and macros - associated with the context that was just popped. - - NASM provides the directive `%repl', which _replaces_ a context with - a different name, without touching the associated macros and labels. - So you could replace the destructive code - - %pop - %push newname - - with the non-destructive version `%repl newname'. - - 4.7.5 Example Use of the Context Stack: Block IFs - - This example makes use of almost all the context-stack features, - including the conditional-assembly construct `%ifctx', to implement - a block IF statement as a set of macros. - - %macro if 1 - - %push if - j%-1 %$ifnot - - %endmacro - - %macro else 0 - - %ifctx if - %repl else - jmp %$ifend - %$ifnot: - %else - %error "expected `if' before `else'" - %endif - - %endmacro - - %macro endif 0 - - %ifctx if - %$ifnot: - %pop - %elifctx else - %$ifend: - %pop - %else - %error "expected `if' or `else' before `endif'" - %endif - - %endmacro - - This code is more robust than the `REPEAT' and `UNTIL' macros given - in section 4.7.2, because it uses conditional assembly to check that - the macros are issued in the right order (for example, not calling - `endif' before `if') and issues a `%error' if they're not. - - In addition, the `endif' macro has to be able to cope with the two - distinct cases of either directly following an `if', or following an - `else'. It achieves this, again, by using conditional assembly to do - different things depending on whether the context on top of the - stack is `if' or `else'. - - The `else' macro has to preserve the context on the stack, in order - to have the `%$ifnot' referred to by the `if' macro be the same as - the one defined by the `endif' macro, but has to change the - context's name so that `endif' will know there was an intervening - `else'. It does this by the use of `%repl'. - - A sample usage of these macros might look like: - - cmp ax,bx - - if ae - cmp bx,cx - - if ae - mov ax,cx - else - mov ax,bx - endif - - else - cmp ax,cx - - if ae - mov ax,cx - endif - - endif - - The block-`IF' macros handle nesting quite happily, by means of - pushing another context, describing the inner `if', on top of the - one describing the outer `if'; thus `else' and `endif' always refer - to the last unmatched `if' or `else'. - - 4.8 Stack Relative Preprocessor Directives - - The following preprocessor directives provide a way to use labels to - refer to local variables allocated on the stack. - - (*) `%arg' (see section 4.8.1) - - (*) `%stacksize' (see section 4.8.2) - - (*) `%local' (see section 4.8.3) - - 4.8.1 `%arg' Directive - - The `%arg' directive is used to simplify the handling of parameters - passed on the stack. Stack based parameter passing is used by many - high level languages, including C, C++ and Pascal. - - While NASM has macros which attempt to duplicate this functionality - (see section 8.4.5), the syntax is not particularly convenient to - use. and is not TASM compatible. Here is an example which shows the - use of `%arg' without any external macros: - - some_function: - - %push mycontext ; save the current context - %stacksize large ; tell NASM to use bp - %arg i:word, j_ptr:word - - mov ax,[i] - mov bx,[j_ptr] - add ax,[bx] - ret - - %pop ; restore original context - - This is similar to the procedure defined in section 8.4.5 and adds - the value in i to the value pointed to by j_ptr and returns the sum - in the ax register. See section 4.7.1 for an explanation of `push' - and `pop' and the use of context stacks. - - 4.8.2 `%stacksize' Directive - - The `%stacksize' directive is used in conjunction with the `%arg' - (see section 4.8.1) and the `%local' (see section 4.8.3) directives. - It tells NASM the default size to use for subsequent `%arg' and - `%local' directives. The `%stacksize' directive takes one required - argument which is one of `flat', `flat64', `large' or `small'. - - %stacksize flat - - This form causes NASM to use stack-based parameter addressing - relative to `ebp' and it assumes that a near form of call was used - to get to this label (i.e. that `eip' is on the stack). - - %stacksize flat64 - - This form causes NASM to use stack-based parameter addressing - relative to `rbp' and it assumes that a near form of call was used - to get to this label (i.e. that `rip' is on the stack). - - %stacksize large - - This form uses `bp' to do stack-based parameter addressing and - assumes that a far form of call was used to get to this address - (i.e. that `ip' and `cs' are on the stack). - - %stacksize small - - This form also uses `bp' to address stack parameters, but it is - different from `large' because it also assumes that the old value of - bp is pushed onto the stack (i.e. it expects an `ENTER' - instruction). In other words, it expects that `bp', `ip' and `cs' - are on the top of the stack, underneath any local space which may - have been allocated by `ENTER'. This form is probably most useful - when used in combination with the `%local' directive (see section - 4.8.3). - - 4.8.3 `%local' Directive - - The `%local' directive is used to simplify the use of local - temporary stack variables allocated in a stack frame. Automatic - local variables in C are an example of this kind of variable. The - `%local' directive is most useful when used with the `%stacksize' - (see section 4.8.2 and is also compatible with the `%arg' directive - (see section 4.8.1). It allows simplified reference to variables on - the stack which have been allocated typically by using the `ENTER' - instruction. An example of its use is the following: - - silly_swap: - - %push mycontext ; save the current context - %stacksize small ; tell NASM to use bp - %assign %$localsize 0 ; see text for explanation - %local old_ax:word, old_dx:word - - enter %$localsize,0 ; see text for explanation - mov [old_ax],ax ; swap ax & bx - mov [old_dx],dx ; and swap dx & cx - mov ax,bx - mov dx,cx - mov bx,[old_ax] - mov cx,[old_dx] - leave ; restore old bp - ret ; - - %pop ; restore original context - - The `%$localsize' variable is used internally by the `%local' - directive and _must_ be defined within the current context before - the `%local' directive may be used. Failure to do so will result in - one expression syntax error for each `%local' variable declared. It - then may be used in the construction of an appropriately sized ENTER - instruction as shown in the example. - - 4.9 Reporting User-Defined Errors: `%error', `%warning', `%fatal' - - The preprocessor directive `%error' will cause NASM to report an - error if it occurs in assembled code. So if other users are going to - try to assemble your source files, you can ensure that they define - the right macros by means of code like this: - - %ifdef F1 - ; do some setup - %elifdef F2 - ; do some different setup - %else - %error "Neither F1 nor F2 was defined." - %endif - - Then any user who fails to understand the way your code is supposed - to be assembled will be quickly warned of their mistake, rather than - having to wait until the program crashes on being run and then not - knowing what went wrong. - - Similarly, `%warning' issues a warning, but allows assembly to - continue: - - %ifdef F1 - ; do some setup - %elifdef F2 - ; do some different setup - %else - %warning "Neither F1 nor F2 was defined, assuming F1." - %define F1 - %endif - - `%error' and `%warning' are issued only on the final assembly pass. - This makes them safe to use in conjunction with tests that depend on - symbol values. - - `%fatal' terminates assembly immediately, regardless of pass. This - is useful when there is no point in continuing the assembly further, - and doing so is likely just going to cause a spew of confusing error - messages. - - It is optional for the message string after `%error', `%warning' or - `%fatal' to be quoted. If it is _not_, then single-line macros are - expanded in it, which can be used to display more information to the - user. For example: - - %if foo > 64 - %assign foo_over foo-64 - %error foo is foo_over bytes too large - %endif - - 4.10 Other Preprocessor Directives - - NASM also has preprocessor directives which allow access to - information from external sources. Currently they include: - - (*) `%line' enables NASM to correctly handle the output of another - preprocessor (see section 4.10.1). - - (*) `%!' enables NASM to read in the value of an environment - variable, which can then be used in your program (see section - 4.10.2). - -4.10.1 `%line' Directive - - The `%line' directive is used to notify NASM that the input line - corresponds to a specific line number in another file. Typically - this other file would be an original source file, with the current - NASM input being the output of a pre-processor. The `%line' - directive allows NASM to output messages which indicate the line - number of the original source file, instead of the file that is - being read by NASM. - - This preprocessor directive is not generally of use to programmers, - by may be of interest to preprocessor authors. The usage of the - `%line' preprocessor directive is as follows: - - %line nnn[+mmm] [filename] - - In this directive, `nnn' identifies the line of the original source - file which this line corresponds to. `mmm' is an optional parameter - which specifies a line increment value; each line of the input file - read in is considered to correspond to `mmm' lines of the original - source file. Finally, `filename' is an optional parameter which - specifies the file name of the original source file. - - After reading a `%line' preprocessor directive, NASM will report all - file name and line numbers relative to the values specified therein. - -4.10.2 `%!'`<env>': Read an environment variable. - - The `%!<env>' directive makes it possible to read the value of an - environment variable at assembly time. This could, for example, be - used to store the contents of an environment variable into a string, - which could be used at some other point in your code. - - For example, suppose that you have an environment variable `FOO', - and you want the contents of `FOO' to be embedded in your program. - You could do that as follows: - - %defstr FOO %!FOO - - See section 4.1.8 for notes on the `%defstr' directive. - - 4.11 Standard Macros - - NASM defines a set of standard macros, which are already defined - when it starts to process any source file. If you really need a - program to be assembled with no pre-defined macros, you can use the - `%clear' directive to empty the preprocessor of everything but - context-local preprocessor variables and single-line macros. - - Most user-level assembler directives (see chapter 6) are implemented - as macros which invoke primitive directives; these are described in - chapter 6. The rest of the standard macro set is described here. - -4.11.1 NASM Version Macros - - The single-line macros `__NASM_MAJOR__', `__NASM_MINOR__', - `__NASM_SUBMINOR__' and `___NASM_PATCHLEVEL__' expand to the major, - minor, subminor and patch level parts of the version number of NASM - being used. So, under NASM 0.98.32p1 for example, `__NASM_MAJOR__' - would be defined to be 0, `__NASM_MINOR__' would be defined as 98, - `__NASM_SUBMINOR__' would be defined to 32, and - `___NASM_PATCHLEVEL__' would be defined as 1. - - Additionally, the macro `__NASM_SNAPSHOT__' is defined for - automatically generated snapshot releases _only_. - -4.11.2 `__NASM_VERSION_ID__': NASM Version ID - - The single-line macro `__NASM_VERSION_ID__' expands to a dword - integer representing the full version number of the version of nasm - being used. The value is the equivalent to `__NASM_MAJOR__', - `__NASM_MINOR__', `__NASM_SUBMINOR__' and `___NASM_PATCHLEVEL__' - concatenated to produce a single doubleword. Hence, for 0.98.32p1, - the returned number would be equivalent to: - - dd 0x00622001 - - or - - db 1,32,98,0 - - Note that the above lines are generate exactly the same code, the - second line is used just to give an indication of the order that the - separate values will be present in memory. - -4.11.3 `__NASM_VER__': NASM Version string - - The single-line macro `__NASM_VER__' expands to a string which - defines the version number of nasm being used. So, under NASM - 0.98.32 for example, - - db __NASM_VER__ - - would expand to - - db "0.98.32" - -4.11.4 `__FILE__' and `__LINE__': File Name and Line Number - - Like the C preprocessor, NASM allows the user to find out the file - name and line number containing the current instruction. The macro - `__FILE__' expands to a string constant giving the name of the - current input file (which may change through the course of assembly - if `%include' directives are used), and `__LINE__' expands to a - numeric constant giving the current line number in the input file. - - These macros could be used, for example, to communicate debugging - information to a macro, since invoking `__LINE__' inside a macro - definition (either single-line or multi-line) will return the line - number of the macro _call_, rather than _definition_. So to - determine where in a piece of code a crash is occurring, for - example, one could write a routine `stillhere', which is passed a - line number in `EAX' and outputs something like `line 155: still - here'. You could then write a macro - - %macro notdeadyet 0 - - push eax - mov eax,__LINE__ - call stillhere - pop eax - - %endmacro - - and then pepper your code with calls to `notdeadyet' until you find - the crash point. - -4.11.5 `__BITS__': Current BITS Mode - - The `__BITS__' standard macro is updated every time that the BITS - mode is set using the `BITS XX' or `[BITS XX]' directive, where XX - is a valid mode number of 16, 32 or 64. `__BITS__' receives the - specified mode number and makes it globally available. This can be - very useful for those who utilize mode-dependent macros. - -4.11.6 `__OUTPUT_FORMAT__': Current Output Format - - The `__OUTPUT_FORMAT__' standard macro holds the current Output - Format, as given by the `-f' option or NASM's default. Type - `nasm -hf' for a list. - - %ifidn __OUTPUT_FORMAT__, win32 - %define NEWLINE 13, 10 - %elifidn __OUTPUT_FORMAT__, elf32 - %define NEWLINE 10 - %endif - -4.11.7 Assembly Date and Time Macros - - NASM provides a variety of macros that represent the timestamp of - the assembly session. - - (*) The `__DATE__' and `__TIME__' macros give the assembly date and - time as strings, in ISO 8601 format (`"YYYY-MM-DD"' and - `"HH:MM:SS"', respectively.) - - (*) The `__DATE_NUM__' and `__TIME_NUM__' macros give the assembly - date and time in numeric form; in the format `YYYYMMDD' and - `HHMMSS' respectively. - - (*) The `__UTC_DATE__' and `__UTC_TIME__' macros give the assembly - date and time in universal time (UTC) as strings, in ISO 8601 - format (`"YYYY-MM-DD"' and `"HH:MM:SS"', respectively.) If the - host platform doesn't provide UTC time, these macros are - undefined. - - (*) The `__UTC_DATE_NUM__' and `__UTC_TIME_NUM__' macros give the - assembly date and time universal time (UTC) in numeric form; in - the format `YYYYMMDD' and `HHMMSS' respectively. If the host - platform doesn't provide UTC time, these macros are undefined. - - (*) The `__POSIX_TIME__' macro is defined as a number containing the - number of seconds since the POSIX epoch, 1 January 1970 00:00:00 - UTC; excluding any leap seconds. This is computed using UTC time - if available on the host platform, otherwise it is computed - using the local time as if it was UTC. - - All instances of time and date macros in the same assembly session - produce consistent output. For example, in an assembly session - started at 42 seconds after midnight on January 1, 2010 in Moscow - (timezone UTC+3) these macros would have the following values, - assuming, of course, a properly configured environment with a - correct clock: - - __DATE__ "2010-01-01" - __TIME__ "00:00:42" - __DATE_NUM__ 20100101 - __TIME_NUM__ 000042 - __UTC_DATE__ "2009-12-31" - __UTC_TIME__ "21:00:42" - __UTC_DATE_NUM__ 20091231 - __UTC_TIME_NUM__ 210042 - __POSIX_TIME__ 1262293242 - -4.11.8 `__USE_'_package_`__': Package Include Test - - When a standard macro package (see chapter 5) is included with the - `%use' directive (see section 4.6.4), a single-line macro of the - form `__USE_'_package_`__' is automatically defined. This allows - testing if a particular package is invoked or not. - - For example, if the `altreg' package is included (see section 5.1), - then the macro `__USE_ALTREG__' is defined. - -4.11.9 `__PASS__': Assembly Pass - - The macro `__PASS__' is defined to be `1' on preparatory passes, and - `2' on the final pass. In preprocess-only mode, it is set to `3', - and when running only to generate dependencies (due to the `-M' or - `-MG' option, see section 2.1.4) it is set to `0'. - - _Avoid using this macro if at all possible. It is tremendously easy - to generate very strange errors by misusing it, and the semantics - may change in future versions of NASM._ - -4.11.10 `STRUC' and `ENDSTRUC': Declaring Structure Data Types - - The core of NASM contains no intrinsic means of defining data - structures; instead, the preprocessor is sufficiently powerful that - data structures can be implemented as a set of macros. The macros - `STRUC' and `ENDSTRUC' are used to define a structure data type. - - `STRUC' takes one or two parameters. The first parameter is the name - of the data type. The second, optional parameter is the base offset - of the structure. The name of the data type is defined as a symbol - with the value of the base offset, and the name of the data type - with the suffix `_size' appended to it is defined as an `EQU' giving - the size of the structure. Once `STRUC' has been issued, you are - defining the structure, and should define fields using the `RESB' - family of pseudo-instructions, and then invoke `ENDSTRUC' to finish - the definition. - - For example, to define a structure called `mytype' containing a - longword, a word, a byte and a string of bytes, you might code - - struc mytype - - mt_long: resd 1 - mt_word: resw 1 - mt_byte: resb 1 - mt_str: resb 32 - - endstruc - - The above code defines six symbols: `mt_long' as 0 (the offset from - the beginning of a `mytype' structure to the longword field), - `mt_word' as 4, `mt_byte' as 6, `mt_str' as 7, `mytype_size' as 39, - and `mytype' itself as zero. - - The reason why the structure type name is defined at zero by default - is a side effect of allowing structures to work with the local label - mechanism: if your structure members tend to have the same names in - more than one structure, you can define the above structure like - this: - - struc mytype - - .long: resd 1 - .word: resw 1 - .byte: resb 1 - .str: resb 32 - - endstruc - - This defines the offsets to the structure fields as `mytype.long', - `mytype.word', `mytype.byte' and `mytype.str'. - - NASM, since it has no _intrinsic_ structure support, does not - support any form of period notation to refer to the elements of a - structure once you have one (except the above local-label notation), - so code such as `mov ax,[mystruc.mt_word]' is not valid. `mt_word' - is a constant just like any other constant, so the correct syntax is - `mov ax,[mystruc+mt_word]' or `mov ax,[mystruc+mytype.word]'. - - Sometimes you only have the address of the structure displaced by an - offset. For example, consider this standard stack frame setup: - - push ebp - mov ebp, esp - sub esp, 40 - - In this case, you could access an element by subtracting the offset: - - mov [ebp - 40 + mytype.word], ax - - However, if you do not want to repeat this offset, you can use -40 - as a base offset: - - struc mytype, -40 - - And access an element this way: - - mov [ebp + mytype.word], ax - -4.11.11 `ISTRUC', `AT' and `IEND': Declaring Instances of Structures - - Having defined a structure type, the next thing you typically want - to do is to declare instances of that structure in your data - segment. NASM provides an easy way to do this in the `ISTRUC' - mechanism. To declare a structure of type `mytype' in a program, you - code something like this: - - mystruc: - istruc mytype - - at mt_long, dd 123456 - at mt_word, dw 1024 - at mt_byte, db 'x' - at mt_str, db 'hello, world', 13, 10, 0 - - iend - - The function of the `AT' macro is to make use of the `TIMES' prefix - to advance the assembly position to the correct point for the - specified structure field, and then to declare the specified data. - Therefore the structure fields must be declared in the same order as - they were specified in the structure definition. - - If the data to go in a structure field requires more than one source - line to specify, the remaining source lines can easily come after - the `AT' line. For example: - - at mt_str, db 123,134,145,156,167,178,189 - db 190,100,0 - - Depending on personal taste, you can also omit the code part of the - `AT' line completely, and start the structure field on the next - line: - - at mt_str - db 'hello, world' - db 13,10,0 - -4.11.12 `ALIGN' and `ALIGNB': Data Alignment - - The `ALIGN' and `ALIGNB' macros provides a convenient way to align - code or data on a word, longword, paragraph or other boundary. (Some - assemblers call this directive `EVEN'.) The syntax of the `ALIGN' - and `ALIGNB' macros is - - align 4 ; align on 4-byte boundary - align 16 ; align on 16-byte boundary - align 8,db 0 ; pad with 0s rather than NOPs - align 4,resb 1 ; align to 4 in the BSS - alignb 4 ; equivalent to previous line - - Both macros require their first argument to be a power of two; they - both compute the number of additional bytes required to bring the - length of the current section up to a multiple of that power of two, - and then apply the `TIMES' prefix to their second argument to - perform the alignment. - - If the second argument is not specified, the default for `ALIGN' is - `NOP', and the default for `ALIGNB' is `RESB 1'. So if the second - argument is specified, the two macros are equivalent. Normally, you - can just use `ALIGN' in code and data sections and `ALIGNB' in BSS - sections, and never need the second argument except for special - purposes. - - `ALIGN' and `ALIGNB', being simple macros, perform no error - checking: they cannot warn you if their first argument fails to be a - power of two, or if their second argument generates more than one - byte of code. In each of these cases they will silently do the wrong - thing. - - `ALIGNB' (or `ALIGN' with a second argument of `RESB 1') can be used - within structure definitions: - - struc mytype2 - - mt_byte: - resb 1 - alignb 2 - mt_word: - resw 1 - alignb 4 - mt_long: - resd 1 - mt_str: - resb 32 - - endstruc - - This will ensure that the structure members are sensibly aligned - relative to the base of the structure. - - A final caveat: `ALIGN' and `ALIGNB' work relative to the beginning - of the _section_, not the beginning of the address space in the - final executable. Aligning to a 16-byte boundary when the section - you're in is only guaranteed to be aligned to a 4-byte boundary, for - example, is a waste of effort. Again, NASM does not check that the - section's alignment characteristics are sensible for the use of - `ALIGN' or `ALIGNB'. - - See also the `smartalign' standard macro package, section 5.2. - -Chapter 5: Standard Macro Packages ----------------------------------- - - The `%use' directive (see section 4.6.4) includes one of the - standard macro packages included with the NASM distribution and - compiled into the NASM binary. It operates like the `%include' - directive (see section 4.6.1), but the included contents is provided - by NASM itself. - - The names of standard macro packages are case insensitive, and can - be quoted or not. - - 5.1 `altreg': Alternate Register Names - - The `altreg' standard macro package provides alternate register - names. It provides numeric register names for all registers (not - just `R8'-`R15'), the Intel-defined aliases `R8L'-`R15L' for the low - bytes of register (as opposed to the NASM/AMD standard names `R8B'- - `R15B'), and the names `R0H'-`R3H' (by analogy with `R0L'-`R3L') for - `AH', `CH', `DH', and `BH'. - - Example use: - - %use altreg - - proc: - mov r0l,r3h ; mov al,bh - ret - - See also section 11.1. - - 5.2 `smartalign': Smart `ALIGN' Macro - - The `smartalign' standard macro package provides for an `ALIGN' - macro which is more powerful than the default (and backwards- - compatible) one (see section 4.11.12). When the `smartalign' package - is enabled, when `ALIGN' is used without a second argument, NASM - will generate a sequence of instructions more efficient than a - series of `NOP'. Furthermore, if the padding exceeds a specific - threshold, then NASM will generate a jump over the entire padding - sequence. - - The specific instructions generated can be controlled with the new - `ALIGNMODE' macro. This macro takes two parameters: one mode, and an - optional jump threshold override. The modes are as follows: - - (*) `generic': Works on all x86 CPUs and should have reasonable - performance. The default jump threshold is 8. This is the - default. - - (*) `nop': Pad out with `NOP' instructions. The only difference - compared to the standard `ALIGN' macro is that NASM can still - jump over a large padding area. The default jump threshold is - 16. - - (*) `k7': Optimize for the AMD K7 (Athlon/Althon XP). These - instructions should still work on all x86 CPUs. The default jump - threshold is 16. - - (*) `k8': Optimize for the AMD K8 (Opteron/Althon 64). These - instructions should still work on all x86 CPUs. The default jump - threshold is 16. - - (*) `p6': Optimize for Intel CPUs. This uses the long `NOP' - instructions first introduced in Pentium Pro. This is - incompatible with all CPUs of family 5 or lower, as well as some - VIA CPUs and several virtualization solutions. The default jump - threshold is 16. - - The macro `__ALIGNMODE__' is defined to contain the current - alignment mode. A number of other macros beginning with `__ALIGN_' - are used internally by this macro package. - -Chapter 6: Assembler Directives -------------------------------- - - NASM, though it attempts to avoid the bureaucracy of assemblers like - MASM and TASM, is nevertheless forced to support a _few_ directives. - These are described in this chapter. - - NASM's directives come in two types: _user-level_ directives and - _primitive_ directives. Typically, each directive has a user-level - form and a primitive form. In almost all cases, we recommend that - users use the user-level forms of the directives, which are - implemented as macros which call the primitive forms. - - Primitive directives are enclosed in square brackets; user-level - directives are not. - - In addition to the universal directives described in this chapter, - each object file format can optionally supply extra directives in - order to control particular features of that file format. These - _format-specific_ directives are documented along with the formats - that implement them, in chapter 7. - - 6.1 `BITS': Specifying Target Processor Mode - - The `BITS' directive specifies whether NASM should generate code - designed to run on a processor operating in 16-bit mode, 32-bit mode - or 64-bit mode. The syntax is `BITS XX', where XX is 16, 32 or 64. - - In most cases, you should not need to use `BITS' explicitly. The - `aout', `coff', `elf', `macho', `win32' and `win64' object formats, - which are designed for use in 32-bit or 64-bit operating systems, - all cause NASM to select 32-bit or 64-bit mode, respectively, by - default. The `obj' object format allows you to specify each segment - you define as either `USE16' or `USE32', and NASM will set its - operating mode accordingly, so the use of the `BITS' directive is - once again unnecessary. - - The most likely reason for using the `BITS' directive is to write - 32-bit or 64-bit code in a flat binary file; this is because the - `bin' output format defaults to 16-bit mode in anticipation of it - being used most frequently to write DOS `.COM' programs, DOS `.SYS' - device drivers and boot loader software. - - You do _not_ need to specify `BITS 32' merely in order to use 32-bit - instructions in a 16-bit DOS program; if you do, the assembler will - generate incorrect code because it will be writing code targeted at - a 32-bit platform, to be run on a 16-bit one. - - When NASM is in `BITS 16' mode, instructions which use 32-bit data - are prefixed with an 0x66 byte, and those referring to 32-bit - addresses have an 0x67 prefix. In `BITS 32' mode, the reverse is - true: 32-bit instructions require no prefixes, whereas instructions - using 16-bit data need an 0x66 and those working on 16-bit addresses - need an 0x67. - - When NASM is in `BITS 64' mode, most instructions operate the same - as they do for `BITS 32' mode. However, there are 8 more general and - SSE registers, and 16-bit addressing is no longer supported. - - The default address size is 64 bits; 32-bit addressing can be - selected with the 0x67 prefix. The default operand size is still 32 - bits, however, and the 0x66 prefix selects 16-bit operand size. The - `REX' prefix is used both to select 64-bit operand size, and to - access the new registers. NASM automatically inserts REX prefixes - when necessary. - - When the `REX' prefix is used, the processor does not know how to - address the AH, BH, CH or DH (high 8-bit legacy) registers. Instead, - it is possible to access the the low 8-bits of the SP, BP SI and DI - registers as SPL, BPL, SIL and DIL, respectively; but only when the - REX prefix is used. - - The `BITS' directive has an exactly equivalent primitive form, - `[BITS 16]', `[BITS 32]' and `[BITS 64]'. The user-level form is a - macro which has no function other than to call the primitive form. - - Note that the space is neccessary, e.g. `BITS32' will _not_ work! - - 6.1.1 `USE16' & `USE32': Aliases for BITS - - The ``USE16'' and ``USE32'' directives can be used in place of - ``BITS 16'' and ``BITS 32'', for compatibility with other - assemblers. - - 6.2 `DEFAULT': Change the assembler defaults - - The `DEFAULT' directive changes the assembler defaults. Normally, - NASM defaults to a mode where the programmer is expected to - explicitly specify most features directly. However, this is - occationally obnoxious, as the explicit form is pretty much the only - one one wishes to use. - - Currently, the only `DEFAULT' that is settable is whether or not - registerless instructions in 64-bit mode are `RIP'-relative or not. - By default, they are absolute unless overridden with the `REL' - specifier (see section 3.3). However, if `DEFAULT REL' is specified, - `REL' is default, unless overridden with the `ABS' specifier, - _except when used with an FS or GS segment override_. - - The special handling of `FS' and `GS' overrides are due to the fact - that these registers are generally used as thread pointers or other - special functions in 64-bit mode, and generating `RIP'-relative - addresses would be extremely confusing. - - `DEFAULT REL' is disabled with `DEFAULT ABS'. - - 6.3 `SECTION' or `SEGMENT': Changing and Defining Sections - - The `SECTION' directive (`SEGMENT' is an exactly equivalent synonym) - changes which section of the output file the code you write will be - assembled into. In some object file formats, the number and names of - sections are fixed; in others, the user may make up as many as they - wish. Hence `SECTION' may sometimes give an error message, or may - define a new section, if you try to switch to a section that does - not (yet) exist. - - The Unix object formats, and the `bin' object format (but see - section 7.1.3, all support the standardized section names `.text', - `.data' and `.bss' for the code, data and uninitialized-data - sections. The `obj' format, by contrast, does not recognize these - section names as being special, and indeed will strip off the - leading period of any section name that has one. - - 6.3.1 The `__SECT__' Macro - - The `SECTION' directive is unusual in that its user-level form - functions differently from its primitive form. The primitive form, - `[SECTION xyz]', simply switches the current target section to the - one given. The user-level form, `SECTION xyz', however, first - defines the single-line macro `__SECT__' to be the primitive - `[SECTION]' directive which it is about to issue, and then issues - it. So the user-level directive - - SECTION .text - - expands to the two lines - - %define __SECT__ [SECTION .text] - [SECTION .text] - - Users may find it useful to make use of this in their own macros. - For example, the `writefile' macro defined in section 4.3.4 can be - usefully rewritten in the following more sophisticated form: - - %macro writefile 2+ - - [section .data] - - %%str: db %2 - %%endstr: - - __SECT__ - - mov dx,%%str - mov cx,%%endstr-%%str - mov bx,%1 - mov ah,0x40 - int 0x21 - - %endmacro - - This form of the macro, once passed a string to output, first - switches temporarily to the data section of the file, using the - primitive form of the `SECTION' directive so as not to modify - `__SECT__'. It then declares its string in the data section, and - then invokes `__SECT__' to switch back to _whichever_ section the - user was previously working in. It thus avoids the need, in the - previous version of the macro, to include a `JMP' instruction to - jump over the data, and also does not fail if, in a complicated - `OBJ' format module, the user could potentially be assembling the - code in any of several separate code sections. - - 6.4 `ABSOLUTE': Defining Absolute Labels - - The `ABSOLUTE' directive can be thought of as an alternative form of - `SECTION': it causes the subsequent code to be directed at no - physical section, but at the hypothetical section starting at the - given absolute address. The only instructions you can use in this - mode are the `RESB' family. - - `ABSOLUTE' is used as follows: - - absolute 0x1A - - kbuf_chr resw 1 - kbuf_free resw 1 - kbuf resw 16 - - This example describes a section of the PC BIOS data area, at - segment address 0x40: the above code defines `kbuf_chr' to be 0x1A, - `kbuf_free' to be 0x1C, and `kbuf' to be 0x1E. - - The user-level form of `ABSOLUTE', like that of `SECTION', redefines - the `__SECT__' macro when it is invoked. - - `STRUC' and `ENDSTRUC' are defined as macros which use `ABSOLUTE' - (and also `__SECT__'). - - `ABSOLUTE' doesn't have to take an absolute constant as an argument: - it can take an expression (actually, a critical expression: see - section 3.8) and it can be a value in a segment. For example, a TSR - can re-use its setup code as run-time BSS like this: - - org 100h ; it's a .COM program - - jmp setup ; setup code comes last - - ; the resident part of the TSR goes here - setup: - ; now write the code that installs the TSR here - - absolute setup - - runtimevar1 resw 1 - runtimevar2 resd 20 - - tsr_end: - - This defines some variables `on top of' the setup code, so that - after the setup has finished running, the space it took up can be - re-used as data storage for the running TSR. The symbol `tsr_end' - can be used to calculate the total size of the part of the TSR that - needs to be made resident. - - 6.5 `EXTERN': Importing Symbols from Other Modules - - `EXTERN' is similar to the MASM directive `EXTRN' and the C keyword - `extern': it is used to declare a symbol which is not defined - anywhere in the module being assembled, but is assumed to be defined - in some other module and needs to be referred to by this one. Not - every object-file format can support external variables: the `bin' - format cannot. - - The `EXTERN' directive takes as many arguments as you like. Each - argument is the name of a symbol: - - extern _printf - extern _sscanf,_fscanf - - Some object-file formats provide extra features to the `EXTERN' - directive. In all cases, the extra features are used by suffixing a - colon to the symbol name followed by object-format specific text. - For example, the `obj' format allows you to declare that the default - segment base of an external should be the group `dgroup' by means of - the directive - - extern _variable:wrt dgroup - - The primitive form of `EXTERN' differs from the user-level form only - in that it can take only one argument at a time: the support for - multiple arguments is implemented at the preprocessor level. - - You can declare the same variable as `EXTERN' more than once: NASM - will quietly ignore the second and later redeclarations. You can't - declare a variable as `EXTERN' as well as something else, though. - - 6.6 `GLOBAL': Exporting Symbols to Other Modules - - `GLOBAL' is the other end of `EXTERN': if one module declares a - symbol as `EXTERN' and refers to it, then in order to prevent linker - errors, some other module must actually _define_ the symbol and - declare it as `GLOBAL'. Some assemblers use the name `PUBLIC' for - this purpose. - - The `GLOBAL' directive applying to a symbol must appear _before_ the - definition of the symbol. - - `GLOBAL' uses the same syntax as `EXTERN', except that it must refer - to symbols which _are_ defined in the same module as the `GLOBAL' - directive. For example: - - global _main - _main: - ; some code - - `GLOBAL', like `EXTERN', allows object formats to define private - extensions by means of a colon. The `elf' object format, for - example, lets you specify whether global data items are functions or - data: - - global hashlookup:function, hashtable:data - - Like `EXTERN', the primitive form of `GLOBAL' differs from the user- - level form only in that it can take only one argument at a time. - - 6.7 `COMMON': Defining Common Data Areas - - The `COMMON' directive is used to declare _common variables_. A - common variable is much like a global variable declared in the - uninitialized data section, so that - - common intvar 4 - - is similar in function to - - global intvar - section .bss - - intvar resd 1 - - The difference is that if more than one module defines the same - common variable, then at link time those variables will be _merged_, - and references to `intvar' in all modules will point at the same - piece of memory. - - Like `GLOBAL' and `EXTERN', `COMMON' supports object-format specific - extensions. For example, the `obj' format allows common variables to - be NEAR or FAR, and the `elf' format allows you to specify the - alignment requirements of a common variable: - - common commvar 4:near ; works in OBJ - common intarray 100:4 ; works in ELF: 4 byte aligned - - Once again, like `EXTERN' and `GLOBAL', the primitive form of - `COMMON' differs from the user-level form only in that it can take - only one argument at a time. - - 6.8 `CPU': Defining CPU Dependencies - - The `CPU' directive restricts assembly to those instructions which - are available on the specified CPU. - - Options are: - - (*) `CPU 8086' Assemble only 8086 instruction set - - (*) `CPU 186' Assemble instructions up to the 80186 instruction set - - (*) `CPU 286' Assemble instructions up to the 286 instruction set - - (*) `CPU 386' Assemble instructions up to the 386 instruction set - - (*) `CPU 486' 486 instruction set - - (*) `CPU 586' Pentium instruction set - - (*) `CPU PENTIUM' Same as 586 - - (*) `CPU 686' P6 instruction set - - (*) `CPU PPRO' Same as 686 - - (*) `CPU P2' Same as 686 - - (*) `CPU P3' Pentium III (Katmai) instruction sets - - (*) `CPU KATMAI' Same as P3 - - (*) `CPU P4' Pentium 4 (Willamette) instruction set - - (*) `CPU WILLAMETTE' Same as P4 - - (*) `CPU PRESCOTT' Prescott instruction set - - (*) `CPU X64' x86-64 (x64/AMD64/Intel 64) instruction set - - (*) `CPU IA64' IA64 CPU (in x86 mode) instruction set - - All options are case insensitive. All instructions will be selected - only if they apply to the selected CPU or lower. By default, all - instructions are available. - - 6.9 `FLOAT': Handling of floating-point constants - - By default, floating-point constants are rounded to nearest, and - IEEE denormals are supported. The following options can be set to - alter this behaviour: - - (*) `FLOAT DAZ' Flush denormals to zero - - (*) `FLOAT NODAZ' Do not flush denormals to zero (default) - - (*) `FLOAT NEAR' Round to nearest (default) - - (*) `FLOAT UP' Round up (toward +Infinity) - - (*) `FLOAT DOWN' Round down (toward -Infinity) - - (*) `FLOAT ZERO' Round toward zero - - (*) `FLOAT DEFAULT' Restore default settings - - The standard macros `__FLOAT_DAZ__', `__FLOAT_ROUND__', and - `__FLOAT__' contain the current state, as long as the programmer has - avoided the use of the brackeded primitive form, (`[FLOAT]'). - - `__FLOAT__' contains the full set of floating-point settings; this - value can be saved away and invoked later to restore the setting. - -Chapter 7: Output Formats -------------------------- - - NASM is a portable assembler, designed to be able to compile on any - ANSI C-supporting platform and produce output to run on a variety of - Intel x86 operating systems. For this reason, it has a large number - of available output formats, selected using the `-f' option on the - NASM command line. Each of these formats, along with its extensions - to the base NASM syntax, is detailed in this chapter. - - As stated in section 2.1.1, NASM chooses a default name for your - output file based on the input file name and the chosen output - format. This will be generated by removing the extension (`.asm', - `.s', or whatever you like to use) from the input file name, and - substituting an extension defined by the output format. The - extensions are given with each format below. - - 7.1 `bin': Flat-Form Binary Output - - The `bin' format does not produce object files: it generates nothing - in the output file except the code you wrote. Such `pure binary' - files are used by MS-DOS: `.COM' executables and `.SYS' device - drivers are pure binary files. Pure binary output is also useful for - operating system and boot loader development. - - The `bin' format supports multiple section names. For details of how - NASM handles sections in the `bin' format, see section 7.1.3. - - Using the `bin' format puts NASM by default into 16-bit mode (see - section 6.1). In order to use `bin' to write 32-bit or 64-bit code, - such as an OS kernel, you need to explicitly issue the `BITS 32' or - `BITS 64' directive. - - `bin' has no default output file name extension: instead, it leaves - your file name as it is once the original extension has been - removed. Thus, the default is for NASM to assemble `binprog.asm' - into a binary file called `binprog'. - - 7.1.1 `ORG': Binary File Program Origin - - The `bin' format provides an additional directive to the list given - in chapter 6: `ORG'. The function of the `ORG' directive is to - specify the origin address which NASM will assume the program begins - at when it is loaded into memory. - - For example, the following code will generate the longword - `0x00000104': - - org 0x100 - dd label - label: - - Unlike the `ORG' directive provided by MASM-compatible assemblers, - which allows you to jump around in the object file and overwrite - code you have already generated, NASM's `ORG' does exactly what the - directive says: _origin_. Its sole function is to specify one offset - which is added to all internal address references within the - section; it does not permit any of the trickery that MASM's version - does. See section 12.1.3 for further comments. - - 7.1.2 `bin' Extensions to the `SECTION' Directive - - The `bin' output format extends the `SECTION' (or `SEGMENT') - directive to allow you to specify the alignment requirements of - segments. This is done by appending the `ALIGN' qualifier to the end - of the section-definition line. For example, - - section .data align=16 - - switches to the section `.data' and also specifies that it must be - aligned on a 16-byte boundary. - - The parameter to `ALIGN' specifies how many low bits of the section - start address must be forced to zero. The alignment value given may - be any power of two. - - 7.1.3 Multisection Support for the `bin' Format - - The `bin' format allows the use of multiple sections, of arbitrary - names, besides the "known" `.text', `.data', and `.bss' names. - - (*) Sections may be designated `progbits' or `nobits'. Default is - `progbits' (except `.bss', which defaults to `nobits', of - course). - - (*) Sections can be aligned at a specified boundary following the - previous section with `align=', or at an arbitrary byte-granular - position with `start='. - - (*) Sections can be given a virtual start address, which will be - used for the calculation of all memory references within that - section with `vstart='. - - (*) Sections can be ordered using `follows='`<section>' or - `vfollows='`<section>' as an alternative to specifying an - explicit start address. - - (*) Arguments to `org', `start', `vstart', and `align=' are critical - expressions. See section 3.8. E.g. `align=(1 << ALIGN_SHIFT)' - - `ALIGN_SHIFT' must be defined before it is used here. - - (*) Any code which comes before an explicit `SECTION' directive is - directed by default into the `.text' section. - - (*) If an `ORG' statement is not given, `ORG 0' is used by default. - - (*) The `.bss' section will be placed after the last `progbits' - section, unless `start=', `vstart=', `follows=', or `vfollows=' - has been specified. - - (*) All sections are aligned on dword boundaries, unless a different - alignment has been specified. - - (*) Sections may not overlap. - - (*) NASM creates the `section.<secname>.start' for each section, - which may be used in your code. - - 7.1.4 Map Files - - Map files can be generated in `-f bin' format by means of the - `[map]' option. Map types of `all' (default), `brief', `sections', - `segments', or `symbols' may be specified. Output may be directed to - `stdout' (default), `stderr', or a specified file. E.g. - `[map symbols myfile.map]'. No "user form" exists, the square - brackets must be used. - - 7.2 `ith': Intel Hex Output - - The `ith' file format produces Intel hex-format files. Just as the - `bin' format, this is a flat memory image format with no support for - relocation or linking. It is usually used with ROM programmers and - similar utilities. - - All extensions supported by the `bin' file format is also supported - by the `ith' file format. - - `ith' provides a default output file-name extension of `.ith'. - - 7.3 `srec': Motorola S-Records Output - - The `srec' file format produces Motorola S-records files. Just as - the `bin' format, this is a flat memory image format with no support - for relocation or linking. It is usually used with ROM programmers - and similar utilities. - - All extensions supported by the `bin' file format is also supported - by the `srec' file format. - - `srec' provides a default output file-name extension of `.srec'. - - 7.4 `obj': Microsoft OMF Object Files - - The `obj' file format (NASM calls it `obj' rather than `omf' for - historical reasons) is the one produced by MASM and TASM, which is - typically fed to 16-bit DOS linkers to produce `.EXE' files. It is - also the format used by OS/2. - - `obj' provides a default output file-name extension of `.obj'. - - `obj' is not exclusively a 16-bit format, though: NASM has full - support for the 32-bit extensions to the format. In particular, 32- - bit `obj' format files are used by Borland's Win32 compilers, - instead of using Microsoft's newer `win32' object file format. - - The `obj' format does not define any special segment names: you can - call your segments anything you like. Typical names for segments in - `obj' format files are `CODE', `DATA' and `BSS'. - - If your source file contains code before specifying an explicit - `SEGMENT' directive, then NASM will invent its own segment called - `__NASMDEFSEG' for you. - - When you define a segment in an `obj' file, NASM defines the segment - name as a symbol as well, so that you can access the segment address - of the segment. So, for example: - - segment data - - dvar: dw 1234 - - segment code - - function: - mov ax,data ; get segment address of data - mov ds,ax ; and move it into DS - inc word [dvar] ; now this reference will work - ret - - The `obj' format also enables the use of the `SEG' and `WRT' - operators, so that you can write code which does things like - - extern foo - - mov ax,seg foo ; get preferred segment of foo - mov ds,ax - mov ax,data ; a different segment - mov es,ax - mov ax,[ds:foo] ; this accesses `foo' - mov [es:foo wrt data],bx ; so does this - - 7.4.1 `obj' Extensions to the `SEGMENT' Directive - - The `obj' output format extends the `SEGMENT' (or `SECTION') - directive to allow you to specify various properties of the segment - you are defining. This is done by appending extra qualifiers to the - end of the segment-definition line. For example, - - segment code private align=16 - - defines the segment `code', but also declares it to be a private - segment, and requires that the portion of it described in this code - module must be aligned on a 16-byte boundary. - - The available qualifiers are: - - (*) `PRIVATE', `PUBLIC', `COMMON' and `STACK' specify the - combination characteristics of the segment. `PRIVATE' segments - do not get combined with any others by the linker; `PUBLIC' and - `STACK' segments get concatenated together at link time; and - `COMMON' segments all get overlaid on top of each other rather - than stuck end-to-end. - - (*) `ALIGN' is used, as shown above, to specify how many low bits of - the segment start address must be forced to zero. The alignment - value given may be any power of two from 1 to 4096; in reality, - the only values supported are 1, 2, 4, 16, 256 and 4096, so if 8 - is specified it will be rounded up to 16, and 32, 64 and 128 - will all be rounded up to 256, and so on. Note that alignment to - 4096-byte boundaries is a PharLap extension to the format and - may not be supported by all linkers. - - (*) `CLASS' can be used to specify the segment class; this feature - indicates to the linker that segments of the same class should - be placed near each other in the output file. The class name can - be any word, e.g. `CLASS=CODE'. - - (*) `OVERLAY', like `CLASS', is specified with an arbitrary word as - an argument, and provides overlay information to an overlay- - capable linker. - - (*) Segments can be declared as `USE16' or `USE32', which has the - effect of recording the choice in the object file and also - ensuring that NASM's default assembly mode when assembling in - that segment is 16-bit or 32-bit respectively. - - (*) When writing OS/2 object files, you should declare 32-bit - segments as `FLAT', which causes the default segment base for - anything in the segment to be the special group `FLAT', and also - defines the group if it is not already defined. - - (*) The `obj' file format also allows segments to be declared as - having a pre-defined absolute segment address, although no - linkers are currently known to make sensible use of this - feature; nevertheless, NASM allows you to declare a segment such - as `SEGMENT SCREEN ABSOLUTE=0xB800' if you need to. The - `ABSOLUTE' and `ALIGN' keywords are mutually exclusive. - - NASM's default segment attributes are `PUBLIC', `ALIGN=1', no class, - no overlay, and `USE16'. - - 7.4.2 `GROUP': Defining Groups of Segments - - The `obj' format also allows segments to be grouped, so that a - single segment register can be used to refer to all the segments in - a group. NASM therefore supplies the `GROUP' directive, whereby you - can code - - segment data - - ; some data - - segment bss - - ; some uninitialized data - - group dgroup data bss - - which will define a group called `dgroup' to contain the segments - `data' and `bss'. Like `SEGMENT', `GROUP' causes the group name to - be defined as a symbol, so that you can refer to a variable `var' in - the `data' segment as `var wrt data' or as `var wrt dgroup', - depending on which segment value is currently in your segment - register. - - If you just refer to `var', however, and `var' is declared in a - segment which is part of a group, then NASM will default to giving - you the offset of `var' from the beginning of the _group_, not the - _segment_. Therefore `SEG var', also, will return the group base - rather than the segment base. - - NASM will allow a segment to be part of more than one group, but - will generate a warning if you do this. Variables declared in a - segment which is part of more than one group will default to being - relative to the first group that was defined to contain the segment. - - A group does not have to contain any segments; you can still make - `WRT' references to a group which does not contain the variable you - are referring to. OS/2, for example, defines the special group - `FLAT' with no segments in it. - - 7.4.3 `UPPERCASE': Disabling Case Sensitivity in Output - - Although NASM itself is case sensitive, some OMF linkers are not; - therefore it can be useful for NASM to output single-case object - files. The `UPPERCASE' format-specific directive causes all segment, - group and symbol names that are written to the object file to be - forced to upper case just before being written. Within a source - file, NASM is still case-sensitive; but the object file can be - written entirely in upper case if desired. - - `UPPERCASE' is used alone on a line; it requires no parameters. - - 7.4.4 `IMPORT': Importing DLL Symbols - - The `IMPORT' format-specific directive defines a symbol to be - imported from a DLL, for use if you are writing a DLL's import - library in NASM. You still need to declare the symbol as `EXTERN' as - well as using the `IMPORT' directive. - - The `IMPORT' directive takes two required parameters, separated by - white space, which are (respectively) the name of the symbol you - wish to import and the name of the library you wish to import it - from. For example: - - import WSAStartup wsock32.dll - - A third optional parameter gives the name by which the symbol is - known in the library you are importing it from, in case this is not - the same as the name you wish the symbol to be known by to your code - once you have imported it. For example: - - import asyncsel wsock32.dll WSAAsyncSelect - - 7.4.5 `EXPORT': Exporting DLL Symbols - - The `EXPORT' format-specific directive defines a global symbol to be - exported as a DLL symbol, for use if you are writing a DLL in NASM. - You still need to declare the symbol as `GLOBAL' as well as using - the `EXPORT' directive. - - `EXPORT' takes one required parameter, which is the name of the - symbol you wish to export, as it was defined in your source file. An - optional second parameter (separated by white space from the first) - gives the _external_ name of the symbol: the name by which you wish - the symbol to be known to programs using the DLL. If this name is - the same as the internal name, you may leave the second parameter - off. - - Further parameters can be given to define attributes of the exported - symbol. These parameters, like the second, are separated by white - space. If further parameters are given, the external name must also - be specified, even if it is the same as the internal name. The - available attributes are: - - (*) `resident' indicates that the exported name is to be kept - resident by the system loader. This is an optimisation for - frequently used symbols imported by name. - - (*) `nodata' indicates that the exported symbol is a function which - does not make use of any initialized data. - - (*) `parm=NNN', where `NNN' is an integer, sets the number of - parameter words for the case in which the symbol is a call gate - between 32-bit and 16-bit segments. - - (*) An attribute which is just a number indicates that the symbol - should be exported with an identifying number (ordinal), and - gives the desired number. - - For example: - - export myfunc - export myfunc TheRealMoreFormalLookingFunctionName - export myfunc myfunc 1234 ; export by ordinal - export myfunc myfunc resident parm=23 nodata - - 7.4.6 `..start': Defining the Program Entry Point - - `OMF' linkers require exactly one of the object files being linked - to define the program entry point, where execution will begin when - the program is run. If the object file that defines the entry point - is assembled using NASM, you specify the entry point by declaring - the special symbol `..start' at the point where you wish execution - to begin. - - 7.4.7 `obj' Extensions to the `EXTERN' Directive - - If you declare an external symbol with the directive - - extern foo - - then references such as `mov ax,foo' will give you the offset of - `foo' from its preferred segment base (as specified in whichever - module `foo' is actually defined in). So to access the contents of - `foo' you will usually need to do something like - - mov ax,seg foo ; get preferred segment base - mov es,ax ; move it into ES - mov ax,[es:foo] ; and use offset `foo' from it - - This is a little unwieldy, particularly if you know that an external - is going to be accessible from a given segment or group, say - `dgroup'. So if `DS' already contained `dgroup', you could simply - code - - mov ax,[foo wrt dgroup] - - However, having to type this every time you want to access `foo' can - be a pain; so NASM allows you to declare `foo' in the alternative - form - - extern foo:wrt dgroup - - This form causes NASM to pretend that the preferred segment base of - `foo' is in fact `dgroup'; so the expression `seg foo' will now - return `dgroup', and the expression `foo' is equivalent to - `foo wrt dgroup'. - - This default-`WRT' mechanism can be used to make externals appear to - be relative to any group or segment in your program. It can also be - applied to common variables: see section 7.4.8. - - 7.4.8 `obj' Extensions to the `COMMON' Directive - - The `obj' format allows common variables to be either near or far; - NASM allows you to specify which your variables should be by the use - of the syntax - - common nearvar 2:near ; `nearvar' is a near common - common farvar 10:far ; and `farvar' is far - - Far common variables may be greater in size than 64Kb, and so the - OMF specification says that they are declared as a number of - _elements_ of a given size. So a 10-byte far common variable could - be declared as ten one-byte elements, five two-byte elements, two - five-byte elements or one ten-byte element. - - Some `OMF' linkers require the element size, as well as the variable - size, to match when resolving common variables declared in more than - one module. Therefore NASM must allow you to specify the element - size on your far common variables. This is done by the following - syntax: - - common c_5by2 10:far 5 ; two five-byte elements - common c_2by5 10:far 2 ; five two-byte elements - - If no element size is specified, the default is 1. Also, the `FAR' - keyword is not required when an element size is specified, since - only far commons may have element sizes at all. So the above - declarations could equivalently be - - common c_5by2 10:5 ; two five-byte elements - common c_2by5 10:2 ; five two-byte elements - - In addition to these extensions, the `COMMON' directive in `obj' - also supports default-`WRT' specification like `EXTERN' does - (explained in section 7.4.7). So you can also declare things like - - common foo 10:wrt dgroup - common bar 16:far 2:wrt data - common baz 24:wrt data:6 - - 7.5 `win32': Microsoft Win32 Object Files - - The `win32' output format generates Microsoft Win32 object files, - suitable for passing to Microsoft linkers such as Visual C++. Note - that Borland Win32 compilers do not use this format, but use `obj' - instead (see section 7.4). - - `win32' provides a default output file-name extension of `.obj'. - - Note that although Microsoft say that Win32 object files follow the - `COFF' (Common Object File Format) standard, the object files - produced by Microsoft Win32 compilers are not compatible with COFF - linkers such as DJGPP's, and vice versa. This is due to a difference - of opinion over the precise semantics of PC-relative relocations. To - produce COFF files suitable for DJGPP, use NASM's `coff' output - format; conversely, the `coff' format does not produce object files - that Win32 linkers can generate correct output from. - - 7.5.1 `win32' Extensions to the `SECTION' Directive - - Like the `obj' format, `win32' allows you to specify additional - information on the `SECTION' directive line, to control the type and - properties of sections you declare. Section types and properties are - generated automatically by NASM for the standard section names - `.text', `.data' and `.bss', but may still be overridden by these - qualifiers. - - The available qualifiers are: - - (*) `code', or equivalently `text', defines the section to be a code - section. This marks the section as readable and executable, but - not writable, and also indicates to the linker that the type of - the section is code. - - (*) `data' and `bss' define the section to be a data section, - analogously to `code'. Data sections are marked as readable and - writable, but not executable. `data' declares an initialized - data section, whereas `bss' declares an uninitialized data - section. - - (*) `rdata' declares an initialized data section that is readable - but not writable. Microsoft compilers use this section to place - constants in it. - - (*) `info' defines the section to be an informational section, which - is not included in the executable file by the linker, but may - (for example) pass information _to_ the linker. For example, - declaring an `info'-type section called `.drectve' causes the - linker to interpret the contents of the section as command-line - options. - - (*) `align=', used with a trailing number as in `obj', gives the - alignment requirements of the section. The maximum you may - specify is 64: the Win32 object file format contains no means to - request a greater section alignment than this. If alignment is - not explicitly specified, the defaults are 16-byte alignment for - code sections, 8-byte alignment for rdata sections and 4-byte - alignment for data (and BSS) sections. Informational sections - get a default alignment of 1 byte (no alignment), though the - value does not matter. - - The defaults assumed by NASM if you do not specify the above - qualifiers are: - - section .text code align=16 - section .data data align=4 - section .rdata rdata align=8 - section .bss bss align=4 - - Any other section name is treated by default like `.text'. - - 7.5.2 `win32': Safe Structured Exception Handling - - Among other improvements in Windows XP SP2 and Windows Server 2003 - Microsoft has introduced concept of "safe structured exception - handling." General idea is to collect handlers' entry points in - designated read-only table and have alleged entry point verified - against this table prior exception control is passed to the handler. - In order for an executable module to be equipped with such "safe - exception handler table," all object modules on linker command line - has to comply with certain criteria. If one single module among them - does not, then the table in question is omitted and above mentioned - run-time checks will not be performed for application in question. - Table omission is by default silent and therefore can be easily - overlooked. One can instruct linker to refuse to produce binary - without such table by passing `/safeseh' command line option. - - Without regard to this run-time check merits it's natural to expect - NASM to be capable of generating modules suitable for `/safeseh' - linking. From developer's viewpoint the problem is two-fold: - - (*) how to adapt modules not deploying exception handlers of their - own; - - (*) how to adapt/develop modules utilizing custom exception - handling; - - Former can be easily achieved with any NASM version by adding - following line to source code: - - $@feat.00 equ 1 - - As of version 2.03 NASM adds this absolute symbol automatically. If - it's not already present to be precise. I.e. if for whatever reason - developer would choose to assign another value in source file, it - would still be perfectly possible. - - Registering custom exception handler on the other hand requires - certain "magic." As of version 2.03 additional directive is - implemented, `safeseh', which instructs the assembler to produce - appropriately formatted input data for above mentioned "safe - exception handler table." Its typical use would be: - - section .text - extern _MessageBoxA@16 - %if __NASM_VERSION_ID__ >= 0x02030000 - safeseh handler ; register handler as "safe handler" - %endif - handler: - push DWORD 1 ; MB_OKCANCEL - push DWORD caption - push DWORD text - push DWORD 0 - call _MessageBoxA@16 - sub eax,1 ; incidentally suits as return value - ; for exception handler - ret - global _main - _main: - push DWORD handler - push DWORD [fs:0] - mov DWORD [fs:0],esp ; engage exception handler - xor eax,eax - mov eax,DWORD[eax] ; cause exception - pop DWORD [fs:0] ; disengage exception handler - add esp,4 - ret - text: db 'OK to rethrow, CANCEL to generate core dump',0 - caption:db 'SEGV',0 - - section .drectve info - db '/defaultlib:user32.lib /defaultlib:msvcrt.lib ' - - As you might imagine, it's perfectly possible to produce .exe binary - with "safe exception handler table" and yet engage unregistered - exception handler. Indeed, handler is engaged by simply manipulating - `[fs:0]' location at run-time, something linker has no power over, - run-time that is. It should be explicitly mentioned that such - failure to register handler's entry point with `safeseh' directive - has undesired side effect at run-time. If exception is raised and - unregistered handler is to be executed, the application is abruptly - terminated without any notification whatsoever. One can argue that - system could at least have logged some kind "non-safe exception - handler in x.exe at address n" message in event log, but no, - literally no notification is provided and user is left with no clue - on what caused application failure. - - Finally, all mentions of linker in this paragraph refer to Microsoft - linker version 7.x and later. Presence of `@feat.00' symbol and - input data for "safe exception handler table" causes no backward - incompatibilities and "safeseh" modules generated by NASM 2.03 and - later can still be linked by earlier versions or non-Microsoft - linkers. - - 7.6 `win64': Microsoft Win64 Object Files - - The `win64' output format generates Microsoft Win64 object files, - which is nearly 100% identical to the `win32' object format (section - 7.5) with the exception that it is meant to target 64-bit code and - the x86-64 platform altogether. This object file is used exactly the - same as the `win32' object format (section 7.5), in NASM, with - regard to this exception. - - 7.6.1 `win64': Writing Position-Independent Code - - While `REL' takes good care of RIP-relative addressing, there is one - aspect that is easy to overlook for a Win64 programmer: indirect - references. Consider a switch dispatch table: - - jmp QWORD[dsptch+rax*8] - ... - dsptch: dq case0 - dq case1 - ... - - Even novice Win64 assembler programmer will soon realize that the - code is not 64-bit savvy. Most notably linker will refuse to link it - with - "`'ADDR32' relocation to '.text' invalid without /LARGEADDRESSAWARE:NO'". - So [s]he will have to split jmp instruction as following: - - lea rbx,[rel dsptch] - jmp QWORD[rbx+rax*8] - - What happens behind the scene is that effective address in `lea' is - encoded relative to instruction pointer, or in perfectly position- - independent manner. But this is only part of the problem! Trouble is - that in .dll context `caseN' relocations will make their way to the - final module and might have to be adjusted at .dll load time. To be - specific when it can't be loaded at preferred address. And when this - occurs, pages with such relocations will be rendered private to - current process, which kind of undermines the idea of sharing .dll. - But no worry, it's trivial to fix: - - lea rbx,[rel dsptch] - add rbx,QWORD[rbx+rax*8] - jmp rbx - ... - dsptch: dq case0-dsptch - dq case1-dsptch - ... - - NASM version 2.03 and later provides another alternative, - `wrt ..imagebase' operator, which returns offset from base address - of the current image, be it .exe or .dll module, therefore the name. - For those acquainted with PE-COFF format base address denotes start - of `IMAGE_DOS_HEADER' structure. Here is how to implement switch - with these image-relative references: - - lea rbx,[rel dsptch] - mov eax,DWORD[rbx+rax*4] - sub rbx,dsptch wrt ..imagebase - add rbx,rax - jmp rbx - ... - dsptch: dd case0 wrt ..imagebase - dd case1 wrt ..imagebase - - One can argue that the operator is redundant. Indeed, snippet before - last works just fine with any NASM version and is not even Windows - specific... The real reason for implementing `wrt ..imagebase' will - become apparent in next paragraph. - - It should be noted that `wrt ..imagebase' is defined as 32-bit - operand only: - - dd label wrt ..imagebase ; ok - dq label wrt ..imagebase ; bad - mov eax,label wrt ..imagebase ; ok - mov rax,label wrt ..imagebase ; bad - - 7.6.2 `win64': Structured Exception Handling - - Structured exception handing in Win64 is completely different matter - from Win32. Upon exception program counter value is noted, and - linker-generated table comprising start and end addresses of all the - functions [in given executable module] is traversed and compared to - the saved program counter. Thus so called `UNWIND_INFO' structure is - identified. If it's not found, then offending subroutine is assumed - to be "leaf" and just mentioned lookup procedure is attempted for - its caller. In Win64 leaf function is such function that does not - call any other function _nor_ modifies any Win64 non-volatile - registers, including stack pointer. The latter ensures that it's - possible to identify leaf function's caller by simply pulling the - value from the top of the stack. - - While majority of subroutines written in assembler are not calling - any other function, requirement for non-volatile registers' - immutability leaves developer with not more than 7 registers and no - stack frame, which is not necessarily what [s]he counted with. - Customarily one would meet the requirement by saving non-volatile - registers on stack and restoring them upon return, so what can go - wrong? If [and only if] an exception is raised at run-time and no - `UNWIND_INFO' structure is associated with such "leaf" function, the - stack unwind procedure will expect to find caller's return address - on the top of stack immediately followed by its frame. Given that - developer pushed caller's non-volatile registers on stack, would the - value on top point at some code segment or even addressable space? - Well, developer can attempt copying caller's return address to the - top of stack and this would actually work in some very specific - circumstances. But unless developer can guarantee that these - circumstances are always met, it's more appropriate to assume worst - case scenario, i.e. stack unwind procedure going berserk. Relevant - question is what happens then? Application is abruptly terminated - without any notification whatsoever. Just like in Win32 case, one - can argue that system could at least have logged "unwind procedure - went berserk in x.exe at address n" in event log, but no, no trace - of failure is left. - - Now, when we understand significance of the `UNWIND_INFO' structure, - let's discuss what's in it and/or how it's processed. First of all - it is checked for presence of reference to custom language-specific - exception handler. If there is one, then it's invoked. Depending on - the return value, execution flow is resumed (exception is said to be - "handled"), _or_ rest of `UNWIND_INFO' structure is processed as - following. Beside optional reference to custom handler, it carries - information about current callee's stack frame and where non- - volatile registers are saved. Information is detailed enough to be - able to reconstruct contents of caller's non-volatile registers upon - call to current callee. And so caller's context is reconstructed, - and then unwind procedure is repeated, i.e. another `UNWIND_INFO' - structure is associated, this time, with caller's instruction - pointer, which is then checked for presence of reference to - language-specific handler, etc. The procedure is recursively - repeated till exception is handled. As last resort system "handles" - it by generating memory core dump and terminating the application. - - As for the moment of this writing NASM unfortunately does not - facilitate generation of above mentioned detailed information about - stack frame layout. But as of version 2.03 it implements building - blocks for generating structures involved in stack unwinding. As - simplest example, here is how to deploy custom exception handler for - leaf function: - - default rel - section .text - extern MessageBoxA - handler: - sub rsp,40 - mov rcx,0 - lea rdx,[text] - lea r8,[caption] - mov r9,1 ; MB_OKCANCEL - call MessageBoxA - sub eax,1 ; incidentally suits as return value - ; for exception handler - add rsp,40 - ret - global main - main: - xor rax,rax - mov rax,QWORD[rax] ; cause exception - ret - main_end: - text: db 'OK to rethrow, CANCEL to generate core dump',0 - caption:db 'SEGV',0 - - section .pdata rdata align=4 - dd main wrt ..imagebase - dd main_end wrt ..imagebase - dd xmain wrt ..imagebase - section .xdata rdata align=8 - xmain: db 9,0,0,0 - dd handler wrt ..imagebase - section .drectve info - db '/defaultlib:user32.lib /defaultlib:msvcrt.lib ' - - What you see in `.pdata' section is element of the "table comprising - start and end addresses of function" along with reference to - associated `UNWIND_INFO' structure. And what you see in `.xdata' - section is `UNWIND_INFO' structure describing function with no - frame, but with designated exception handler. References are - _required_ to be image-relative (which is the real reason for - implementing `wrt ..imagebase' operator). It should be noted that - `rdata align=n', as well as `wrt ..imagebase', are optional in these - two segments' contexts, i.e. can be omitted. Latter means that _all_ - 32-bit references, not only above listed required ones, placed into - these two segments turn out image-relative. Why is it important to - understand? Developer is allowed to append handler-specific data to - `UNWIND_INFO' structure, and if [s]he adds a 32-bit reference, then - [s]he will have to remember to adjust its value to obtain the real - pointer. - - As already mentioned, in Win64 terms leaf function is one that does - not call any other function _nor_ modifies any non-volatile - register, including stack pointer. But it's not uncommon that - assembler programmer plans to utilize every single register and - sometimes even have variable stack frame. Is there anything one can - do with bare building blocks? I.e. besides manually composing fully- - fledged `UNWIND_INFO' structure, which would surely be considered - error-prone? Yes, there is. Recall that exception handler is called - first, before stack layout is analyzed. As it turned out, it's - perfectly possible to manipulate current callee's context in custom - handler in manner that permits further stack unwinding. General idea - is that handler would not actually "handle" the exception, but - instead restore callee's context, as it was at its entry point and - thus mimic leaf function. In other words, handler would simply - undertake part of unwinding procedure. Consider following example: - - function: - mov rax,rsp ; copy rsp to volatile register - push r15 ; save non-volatile registers - push rbx - push rbp - mov r11,rsp ; prepare variable stack frame - sub r11,rcx - and r11,-64 - mov QWORD[r11],rax ; check for exceptions - mov rsp,r11 ; allocate stack frame - mov QWORD[rsp],rax ; save original rsp value - magic_point: - ... - mov r11,QWORD[rsp] ; pull original rsp value - mov rbp,QWORD[r11-24] - mov rbx,QWORD[r11-16] - mov r15,QWORD[r11-8] - mov rsp,r11 ; destroy frame - ret - - The keyword is that up to `magic_point' original `rsp' value remains - in chosen volatile register and no non-volatile register, except for - `rsp', is modified. While past `magic_point' `rsp' remains constant - till the very end of the `function'. In this case custom language- - specific exception handler would look like this: - - EXCEPTION_DISPOSITION handler (EXCEPTION_RECORD *rec,ULONG64 frame, - CONTEXT *context,DISPATCHER_CONTEXT *disp) - { ULONG64 *rsp; - if (context->Rip<(ULONG64)magic_point) - rsp = (ULONG64 *)context->Rax; - else - { rsp = ((ULONG64 **)context->Rsp)[0]; - context->Rbp = rsp[-3]; - context->Rbx = rsp[-2]; - context->R15 = rsp[-1]; - } - context->Rsp = (ULONG64)rsp; - - memcpy (disp->ContextRecord,context,sizeof(CONTEXT)); - RtlVirtualUnwind(UNW_FLAG_NHANDLER,disp->ImageBase, - dips->ControlPc,disp->FunctionEntry,disp->ContextRecord, - &disp->HandlerData,&disp->EstablisherFrame,NULL); - return ExceptionContinueSearch; - } - - As custom handler mimics leaf function, corresponding `UNWIND_INFO' - structure does not have to contain any information about stack frame - and its layout. - - 7.7 `coff': Common Object File Format - - The `coff' output type produces `COFF' object files suitable for - linking with the DJGPP linker. - - `coff' provides a default output file-name extension of `.o'. - - The `coff' format supports the same extensions to the `SECTION' - directive as `win32' does, except that the `align' qualifier and the - `info' section type are not supported. - - 7.8 `macho32' and `macho64': Mach Object File Format - - The `macho32' and `macho64' output formts produces `Mach-O' object - files suitable for linking with the MacOS X linker. `macho' is a - synonym for `macho32'. - - `macho' provides a default output file-name extension of `.o'. - - 7.9 `elf32' and `elf64': Executable and Linkable Format Object Files - - The `elf32' and `elf64' output formats generate `ELF32 and ELF64' - (Executable and Linkable Format) object files, as used by Linux as - well as Unix System V, including Solaris x86, UnixWare and SCO Unix. - `elf' provides a default output file-name extension of `.o'. `elf' - is a synonym for `elf32'. - - 7.9.1 ELF specific directive `osabi' - - The ELF header specifies the application binary interface for the - target operating system (OSABI). This field can be set by using the - `osabi' directive with the numeric value (0-255) of the target - system. If this directive is not used, the default value will be - "UNIX System V ABI" (0) which will work on most systems which - support ELF. - - 7.9.2 `elf' Extensions to the `SECTION' Directive - - Like the `obj' format, `elf' allows you to specify additional - information on the `SECTION' directive line, to control the type and - properties of sections you declare. Section types and properties are - generated automatically by NASM for the standard section names, but - may still be overridden by these qualifiers. - - The available qualifiers are: - - (*) `alloc' defines the section to be one which is loaded into - memory when the program is run. `noalloc' defines it to be one - which is not, such as an informational or comment section. - - (*) `exec' defines the section to be one which should have execute - permission when the program is run. `noexec' defines it as one - which should not. - - (*) `write' defines the section to be one which should be writable - when the program is run. `nowrite' defines it as one which - should not. - - (*) `progbits' defines the section to be one with explicit contents - stored in the object file: an ordinary code or data section, for - example, `nobits' defines the section to be one with no explicit - contents given, such as a BSS section. - - (*) `align=', used with a trailing number as in `obj', gives the - alignment requirements of the section. - - (*) `tls' defines the section to be one which contains thread local - variables. - - The defaults assumed by NASM if you do not specify the above - qualifiers are: - - - section .text progbits alloc exec nowrite align=16 - section .rodata progbits alloc noexec nowrite align=4 - section .lrodata progbits alloc noexec nowrite align=4 - section .data progbits alloc noexec write align=4 - section .ldata progbits alloc noexec write align=4 - section .bss nobits alloc noexec write align=4 - section .lbss nobits alloc noexec write align=4 - section .tdata progbits alloc noexec write align=4 tls - section .tbss nobits alloc noexec write align=4 tls - section .comment progbits noalloc noexec nowrite align=1 - section other progbits alloc noexec nowrite align=1 - - (Any section name other than those in the above table is treated by - default like `other' in the above table. Please note that section - names are case sensitive.) - - 7.9.3 Position-Independent Code: `elf' Special Symbols and `WRT' - - The `ELF' specification contains enough features to allow position- - independent code (PIC) to be written, which makes ELF shared - libraries very flexible. However, it also means NASM has to be able - to generate a variety of ELF specific relocation types in ELF object - files, if it is to be an assembler which can write PIC. - - Since `ELF' does not support segment-base references, the `WRT' - operator is not used for its normal purpose; therefore NASM's `elf' - output format makes use of `WRT' for a different purpose, namely the - PIC-specific relocation types. - - `elf' defines five special symbols which you can use as the right- - hand side of the `WRT' operator to obtain PIC relocation types. They - are `..gotpc', `..gotoff', `..got', `..plt' and `..sym'. Their - functions are summarized here: - - (*) Referring to the symbol marking the global offset table base - using `wrt ..gotpc' will end up giving the distance from the - beginning of the current section to the global offset table. - (`_GLOBAL_OFFSET_TABLE_' is the standard symbol name used to - refer to the GOT.) So you would then need to add `$$' to the - result to get the real address of the GOT. - - (*) Referring to a location in one of your own sections using - `wrt ..gotoff' will give the distance from the beginning of the - GOT to the specified location, so that adding on the address of - the GOT would give the real address of the location you wanted. - - (*) Referring to an external or global symbol using `wrt ..got' - causes the linker to build an entry _in_ the GOT containing the - address of the symbol, and the reference gives the distance from - the beginning of the GOT to the entry; so you can add on the - address of the GOT, load from the resulting address, and end up - with the address of the symbol. - - (*) Referring to a procedure name using `wrt ..plt' causes the - linker to build a procedure linkage table entry for the symbol, - and the reference gives the address of the PLT entry. You can - only use this in contexts which would generate a PC-relative - relocation normally (i.e. as the destination for `CALL' or - `JMP'), since ELF contains no relocation type to refer to PLT - entries absolutely. - - (*) Referring to a symbol name using `wrt ..sym' causes NASM to - write an ordinary relocation, but instead of making the - relocation relative to the start of the section and then adding - on the offset to the symbol, it will write a relocation record - aimed directly at the symbol in question. The distinction is a - necessary one due to a peculiarity of the dynamic linker. - - A fuller explanation of how to use these relocation types to write - shared libraries entirely in NASM is given in section 9.2. - - 7.9.4 Thread Local Storage: `elf' Special Symbols and `WRT' - - (*) In ELF32 mode, referring to an external or global symbol using - `wrt ..tlsie' causes the linker to build an entry _in_ the GOT - containing the offset of the symbol within the TLS block, so you - can access the value of the symbol with code such as: - - mov eax,[tid wrt ..tlsie] - mov [gs:eax],ebx - - (*) In ELF64 mode, referring to an external or global symbol using - `wrt ..gottpoff' causes the linker to build an entry _in_ the - GOT containing the offset of the symbol within the TLS block, so - you can access the value of the symbol with code such as: - - mov rax,[rel tid wrt ..gottpoff] - mov rcx,[fs:rax] - - 7.9.5 `elf' Extensions to the `GLOBAL' Directive - - `ELF' object files can contain more information about a global - symbol than just its address: they can contain the size of the - symbol and its type as well. These are not merely debugger - conveniences, but are actually necessary when the program being - written is a shared library. NASM therefore supports some extensions - to the `GLOBAL' directive, allowing you to specify these features. - - You can specify whether a global variable is a function or a data - object by suffixing the name with a colon and the word `function' or - `data'. (`object' is a synonym for `data'.) For example: - - global hashlookup:function, hashtable:data - - exports the global symbol `hashlookup' as a function and `hashtable' - as a data object. - - Optionally, you can control the ELF visibility of the symbol. Just - add one of the visibility keywords: `default', `internal', `hidden', - or `protected'. The default is `default' of course. For example, to - make `hashlookup' hidden: - - global hashlookup:function hidden - - You can also specify the size of the data associated with the - symbol, as a numeric expression (which may involve labels, and even - forward references) after the type specifier. Like this: - - global hashtable:data (hashtable.end - hashtable) - - hashtable: - db this,that,theother ; some data here - .end: - - This makes NASM automatically calculate the length of the table and - place that information into the `ELF' symbol table. - - Declaring the type and size of global symbols is necessary when - writing shared library code. For more information, see section - 9.2.4. - - 7.9.6 `elf' Extensions to the `COMMON' Directive - - `ELF' also allows you to specify alignment requirements on common - variables. This is done by putting a number (which must be a power - of two) after the name and size of the common variable, separated - (as usual) by a colon. For example, an array of doublewords would - benefit from 4-byte alignment: - - common dwordarray 128:4 - - This declares the total size of the array to be 128 bytes, and - requires that it be aligned on a 4-byte boundary. - - 7.9.7 16-bit code and ELF - - The `ELF32' specification doesn't provide relocations for 8- and 16- - bit values, but the GNU `ld' linker adds these as an extension. NASM - can generate GNU-compatible relocations, to allow 16-bit code to be - linked as ELF using GNU `ld'. If NASM is used with the - `-w+gnu-elf-extensions' option, a warning is issued when one of - these relocations is generated. - - 7.9.8 Debug formats and ELF - - `ELF32' and `ELF64' provide debug information in `STABS' and `DWARF' - formats. Line number information is generated for all executable - sections, but please note that only the ".text" section is - executable by default. - - 7.10 `aout': Linux `a.out' Object Files - - The `aout' format generates `a.out' object files, in the form used - by early Linux systems (current Linux systems use ELF, see section - 7.9.) These differ from other `a.out' object files in that the magic - number in the first four bytes of the file is different; also, some - implementations of `a.out', for example NetBSD's, support position- - independent code, which Linux's implementation does not. - - `a.out' provides a default output file-name extension of `.o'. - - `a.out' is a very simple object format. It supports no special - directives, no special symbols, no use of `SEG' or `WRT', and no - extensions to any standard directives. It supports only the three - standard section names `.text', `.data' and `.bss'. - - 7.11 `aoutb': NetBSD/FreeBSD/OpenBSD `a.out' Object Files - - The `aoutb' format generates `a.out' object files, in the form used - by the various free `BSD Unix' clones, `NetBSD', `FreeBSD' and - `OpenBSD'. For simple object files, this object format is exactly - the same as `aout' except for the magic number in the first four - bytes of the file. However, the `aoutb' format supports - position-independent code in the same way as the `elf' format, so - you can use it to write `BSD' shared libraries. - - `aoutb' provides a default output file-name extension of `.o'. - - `aoutb' supports no special directives, no special symbols, and only - the three standard section names `.text', `.data' and `.bss'. - However, it also supports the same use of `WRT' as `elf' does, to - provide position-independent code relocation types. See section - 7.9.3 for full documentation of this feature. - - `aoutb' also supports the same extensions to the `GLOBAL' directive - as `elf' does: see section 7.9.5 for documentation of this. - - 7.12 `as86': Minix/Linux `as86' Object Files - - The Minix/Linux 16-bit assembler `as86' has its own non-standard - object file format. Although its companion linker `ld86' produces - something close to ordinary `a.out' binaries as output, the object - file format used to communicate between `as86' and `ld86' is not - itself `a.out'. - - NASM supports this format, just in case it is useful, as `as86'. - `as86' provides a default output file-name extension of `.o'. - - `as86' is a very simple object format (from the NASM user's point of - view). It supports no special directives, no use of `SEG' or `WRT', - and no extensions to any standard directives. It supports only the - three standard section names `.text', `.data' and `.bss'. The only - special symbol supported is `..start'. - - 7.13 `rdf': Relocatable Dynamic Object File Format - - The `rdf' output format produces `RDOFF' object files. `RDOFF' - (Relocatable Dynamic Object File Format) is a home-grown object-file - format, designed alongside NASM itself and reflecting in its file - format the internal structure of the assembler. - - `RDOFF' is not used by any well-known operating systems. Those - writing their own systems, however, may well wish to use `RDOFF' as - their object format, on the grounds that it is designed primarily - for simplicity and contains very little file-header bureaucracy. - - The Unix NASM archive, and the DOS archive which includes sources, - both contain an `rdoff' subdirectory holding a set of RDOFF - utilities: an RDF linker, an `RDF' static-library manager, an RDF - file dump utility, and a program which will load and execute an RDF - executable under Linux. - - `rdf' supports only the standard section names `.text', `.data' and - `.bss'. - -7.13.1 Requiring a Library: The `LIBRARY' Directive - - `RDOFF' contains a mechanism for an object file to demand a given - library to be linked to the module, either at load time or run time. - This is done by the `LIBRARY' directive, which takes one argument - which is the name of the module: - - library mylib.rdl - -7.13.2 Specifying a Module Name: The `MODULE' Directive - - Special `RDOFF' header record is used to store the name of the - module. It can be used, for example, by run-time loader to perform - dynamic linking. `MODULE' directive takes one argument which is the - name of current module: - - module mymodname - - Note that when you statically link modules and tell linker to strip - the symbols from output file, all module names will be stripped too. - To avoid it, you should start module names with `$', like: - - module $kernel.core - -7.13.3 `rdf' Extensions to the `GLOBAL' Directive - - `RDOFF' global symbols can contain additional information needed by - the static linker. You can mark a global symbol as exported, thus - telling the linker do not strip it from target executable or library - file. Like in `ELF', you can also specify whether an exported symbol - is a procedure (function) or data object. - - Suffixing the name with a colon and the word `export' you make the - symbol exported: - - global sys_open:export - - To specify that exported symbol is a procedure (function), you add - the word `proc' or `function' after declaration: - - global sys_open:export proc - - Similarly, to specify exported data object, add the word `data' or - `object' to the directive: - - global kernel_ticks:export data - -7.13.4 `rdf' Extensions to the `EXTERN' Directive - - By default the `EXTERN' directive in `RDOFF' declares a "pure - external" symbol (i.e. the static linker will complain if such a - symbol is not resolved). To declare an "imported" symbol, which must - be resolved later during a dynamic linking phase, `RDOFF' offers an - additional `import' modifier. As in `GLOBAL', you can also specify - whether an imported symbol is a procedure (function) or data object. - For example: - - library $libc - extern _open:import - extern _printf:import proc - extern _errno:import data - - Here the directive `LIBRARY' is also included, which gives the - dynamic linker a hint as to where to find requested symbols. - - 7.14 `dbg': Debugging Format - - The `dbg' output format is not built into NASM in the default - configuration. If you are building your own NASM executable from the - sources, you can define `OF_DBG' in `output/outform.h' or on the - compiler command line, and obtain the `dbg' output format. - - The `dbg' format does not output an object file as such; instead, it - outputs a text file which contains a complete list of all the - transactions between the main body of NASM and the output-format - back end module. It is primarily intended to aid people who want to - write their own output drivers, so that they can get a clearer idea - of the various requests the main program makes of the output driver, - and in what order they happen. - - For simple files, one can easily use the `dbg' format like this: - - nasm -f dbg filename.asm - - which will generate a diagnostic file called `filename.dbg'. - However, this will not work well on files which were designed for a - different object format, because each object format defines its own - macros (usually user-level forms of directives), and those macros - will not be defined in the `dbg' format. Therefore it can be useful - to run NASM twice, in order to do the preprocessing with the native - object format selected: - - nasm -e -f rdf -o rdfprog.i rdfprog.asm - nasm -a -f dbg rdfprog.i - - This preprocesses `rdfprog.asm' into `rdfprog.i', keeping the `rdf' - object format selected in order to make sure RDF special directives - are converted into primitive form correctly. Then the preprocessed - source is fed through the `dbg' format to generate the final - diagnostic output. - - This workaround will still typically not work for programs intended - for `obj' format, because the `obj' `SEGMENT' and `GROUP' directives - have side effects of defining the segment and group names as - symbols; `dbg' will not do this, so the program will not assemble. - You will have to work around that by defining the symbols yourself - (using `EXTERN', for example) if you really need to get a `dbg' - trace of an `obj'-specific source file. - - `dbg' accepts any section name and any directives at all, and logs - them all to its output file. - -Chapter 8: Writing 16-bit Code (DOS, Windows 3/3.1) ---------------------------------------------------- - - This chapter attempts to cover some of the common issues encountered - when writing 16-bit code to run under `MS-DOS' or `Windows 3.x'. It - covers how to link programs to produce `.EXE' or `.COM' files, how - to write `.SYS' device drivers, and how to interface assembly - language code with 16-bit C compilers and with Borland Pascal. - - 8.1 Producing `.EXE' Files - - Any large program written under DOS needs to be built as a `.EXE' - file: only `.EXE' files have the necessary internal structure - required to span more than one 64K segment. Windows programs, also, - have to be built as `.EXE' files, since Windows does not support the - `.COM' format. - - In general, you generate `.EXE' files by using the `obj' output - format to produce one or more `.OBJ' files, and then linking them - together using a linker. However, NASM also supports the direct - generation of simple DOS `.EXE' files using the `bin' output format - (by using `DB' and `DW' to construct the `.EXE' file header), and a - macro package is supplied to do this. Thanks to Yann Guidon for - contributing the code for this. - - NASM may also support `.EXE' natively as another output format in - future releases. - - 8.1.1 Using the `obj' Format To Generate `.EXE' Files - - This section describes the usual method of generating `.EXE' files - by linking `.OBJ' files together. - - Most 16-bit programming language packages come with a suitable - linker; if you have none of these, there is a free linker called - VAL, available in `LZH' archive format from `x2ftp.oulu.fi'. An LZH - archiver can be found at `ftp.simtel.net'. There is another `free' - linker (though this one doesn't come with sources) called FREELINK, - available from `www.pcorner.com'. A third, `djlink', written by DJ - Delorie, is available at `www.delorie.com'. A fourth linker, - `ALINK', written by Anthony A.J. Williams, is available at - `alink.sourceforge.net'. - - When linking several `.OBJ' files into a `.EXE' file, you should - ensure that exactly one of them has a start point defined (using the - `..start' special symbol defined by the `obj' format: see section - 7.4.6). If no module defines a start point, the linker will not know - what value to give the entry-point field in the output file header; - if more than one defines a start point, the linker will not know - _which_ value to use. - - An example of a NASM source file which can be assembled to a `.OBJ' - file and linked on its own to a `.EXE' is given here. It - demonstrates the basic principles of defining a stack, initialising - the segment registers, and declaring a start point. This file is - also provided in the `test' subdirectory of the NASM archives, under - the name `objexe.asm'. - - segment code - - ..start: - mov ax,data - mov ds,ax - mov ax,stack - mov ss,ax - mov sp,stacktop - - This initial piece of code sets up `DS' to point to the data - segment, and initializes `SS' and `SP' to point to the top of the - provided stack. Notice that interrupts are implicitly disabled for - one instruction after a move into `SS', precisely for this - situation, so that there's no chance of an interrupt occurring - between the loads of `SS' and `SP' and not having a stack to execute - on. - - Note also that the special symbol `..start' is defined at the - beginning of this code, which means that will be the entry point - into the resulting executable file. - - mov dx,hello - mov ah,9 - int 0x21 - - The above is the main program: load `DS:DX' with a pointer to the - greeting message (`hello' is implicitly relative to the segment - `data', which was loaded into `DS' in the setup code, so the full - pointer is valid), and call the DOS print-string function. - - mov ax,0x4c00 - int 0x21 - - This terminates the program using another DOS system call. - - segment data - - hello: db 'hello, world', 13, 10, '$' - - The data segment contains the string we want to display. - - segment stack stack - resb 64 - stacktop: - - The above code declares a stack segment containing 64 bytes of - uninitialized stack space, and points `stacktop' at the top of it. - The directive `segment stack stack' defines a segment _called_ - `stack', and also of _type_ `STACK'. The latter is not necessary to - the correct running of the program, but linkers are likely to issue - warnings or errors if your program has no segment of type `STACK'. - - The above file, when assembled into a `.OBJ' file, will link on its - own to a valid `.EXE' file, which when run will print `hello, world' - and then exit. - - 8.1.2 Using the `bin' Format To Generate `.EXE' Files - - The `.EXE' file format is simple enough that it's possible to build - a `.EXE' file by writing a pure-binary program and sticking a 32- - byte header on the front. This header is simple enough that it can - be generated using `DB' and `DW' commands by NASM itself, so that - you can use the `bin' output format to directly generate `.EXE' - files. - - Included in the NASM archives, in the `misc' subdirectory, is a file - `exebin.mac' of macros. It defines three macros: `EXE_begin', - `EXE_stack' and `EXE_end'. - - To produce a `.EXE' file using this method, you should start by - using `%include' to load the `exebin.mac' macro package into your - source file. You should then issue the `EXE_begin' macro call (which - takes no arguments) to generate the file header data. Then write - code as normal for the `bin' format - you can use all three standard - sections `.text', `.data' and `.bss'. At the end of the file you - should call the `EXE_end' macro (again, no arguments), which defines - some symbols to mark section sizes, and these symbols are referred - to in the header code generated by `EXE_begin'. - - In this model, the code you end up writing starts at `0x100', just - like a `.COM' file - in fact, if you strip off the 32-byte header - from the resulting `.EXE' file, you will have a valid `.COM' - program. All the segment bases are the same, so you are limited to a - 64K program, again just like a `.COM' file. Note that an `ORG' - directive is issued by the `EXE_begin' macro, so you should not - explicitly issue one of your own. - - You can't directly refer to your segment base value, unfortunately, - since this would require a relocation in the header, and things - would get a lot more complicated. So you should get your segment - base by copying it out of `CS' instead. - - On entry to your `.EXE' file, `SS:SP' are already set up to point to - the top of a 2Kb stack. You can adjust the default stack size of 2Kb - by calling the `EXE_stack' macro. For example, to change the stack - size of your program to 64 bytes, you would call `EXE_stack 64'. - - A sample program which generates a `.EXE' file in this way is given - in the `test' subdirectory of the NASM archive, as `binexe.asm'. - - 8.2 Producing `.COM' Files - - While large DOS programs must be written as `.EXE' files, small ones - are often better written as `.COM' files. `.COM' files are pure - binary, and therefore most easily produced using the `bin' output - format. - - 8.2.1 Using the `bin' Format To Generate `.COM' Files - - `.COM' files expect to be loaded at offset `100h' into their segment - (though the segment may change). Execution then begins at `100h', - i.e. right at the start of the program. So to write a `.COM' - program, you would create a source file looking like - - org 100h - - section .text - - start: - ; put your code here - - section .data - - ; put data items here - - section .bss - - ; put uninitialized data here - - The `bin' format puts the `.text' section first in the file, so you - can declare data or BSS items before beginning to write code if you - want to and the code will still end up at the front of the file - where it belongs. - - The BSS (uninitialized data) section does not take up space in the - `.COM' file itself: instead, addresses of BSS items are resolved to - point at space beyond the end of the file, on the grounds that this - will be free memory when the program is run. Therefore you should - not rely on your BSS being initialized to all zeros when you run. - - To assemble the above program, you should use a command line like - - nasm myprog.asm -fbin -o myprog.com - - The `bin' format would produce a file called `myprog' if no explicit - output file name were specified, so you have to override it and give - the desired file name. - - 8.2.2 Using the `obj' Format To Generate `.COM' Files - - If you are writing a `.COM' program as more than one module, you may - wish to assemble several `.OBJ' files and link them together into a - `.COM' program. You can do this, provided you have a linker capable - of outputting `.COM' files directly (TLINK does this), or - alternatively a converter program such as `EXE2BIN' to transform the - `.EXE' file output from the linker into a `.COM' file. - - If you do this, you need to take care of several things: - - (*) The first object file containing code should start its code - segment with a line like `RESB 100h'. This is to ensure that the - code begins at offset `100h' relative to the beginning of the - code segment, so that the linker or converter program does not - have to adjust address references within the file when - generating the `.COM' file. Other assemblers use an `ORG' - directive for this purpose, but `ORG' in NASM is a format- - specific directive to the `bin' output format, and does not mean - the same thing as it does in MASM-compatible assemblers. - - (*) You don't need to define a stack segment. - - (*) All your segments should be in the same group, so that every - time your code or data references a symbol offset, all offsets - are relative to the same segment base. This is because, when a - `.COM' file is loaded, all the segment registers contain the - same value. - - 8.3 Producing `.SYS' Files - - MS-DOS device drivers - `.SYS' files - are pure binary files, - similar to `.COM' files, except that they start at origin zero - rather than `100h'. Therefore, if you are writing a device driver - using the `bin' format, you do not need the `ORG' directive, since - the default origin for `bin' is zero. Similarly, if you are using - `obj', you do not need the `RESB 100h' at the start of your code - segment. - - `.SYS' files start with a header structure, containing pointers to - the various routines inside the driver which do the work. This - structure should be defined at the start of the code segment, even - though it is not actually code. - - For more information on the format of `.SYS' files, and the data - which has to go in the header structure, a list of books is given in - the Frequently Asked Questions list for the newsgroup - `comp.os.msdos.programmer'. - - 8.4 Interfacing to 16-bit C Programs - - This section covers the basics of writing assembly routines that - call, or are called from, C programs. To do this, you would - typically write an assembly module as a `.OBJ' file, and link it - with your C modules to produce a mixed-language program. - - 8.4.1 External Symbol Names - - C compilers have the convention that the names of all global symbols - (functions or data) they define are formed by prefixing an - underscore to the name as it appears in the C program. So, for - example, the function a C programmer thinks of as `printf' appears - to an assembly language programmer as `_printf'. This means that in - your assembly programs, you can define symbols without a leading - underscore, and not have to worry about name clashes with C symbols. - - If you find the underscores inconvenient, you can define macros to - replace the `GLOBAL' and `EXTERN' directives as follows: - - %macro cglobal 1 - - global _%1 - %define %1 _%1 - - %endmacro - - %macro cextern 1 - - extern _%1 - %define %1 _%1 - - %endmacro - - (These forms of the macros only take one argument at a time; a - `%rep' construct could solve this.) - - If you then declare an external like this: - - cextern printf - - then the macro will expand it as - - extern _printf - %define printf _printf - - Thereafter, you can reference `printf' as if it was a symbol, and - the preprocessor will put the leading underscore on where necessary. - - The `cglobal' macro works similarly. You must use `cglobal' before - defining the symbol in question, but you would have had to do that - anyway if you used `GLOBAL'. - - Also see section 2.1.27. - - 8.4.2 Memory Models - - NASM contains no mechanism to support the various C memory models - directly; you have to keep track yourself of which one you are - writing for. This means you have to keep track of the following - things: - - (*) In models using a single code segment (tiny, small and compact), - functions are near. This means that function pointers, when - stored in data segments or pushed on the stack as function - arguments, are 16 bits long and contain only an offset field - (the `CS' register never changes its value, and always gives the - segment part of the full function address), and that functions - are called using ordinary near `CALL' instructions and return - using `RETN' (which, in NASM, is synonymous with `RET' anyway). - This means both that you should write your own routines to - return with `RETN', and that you should call external C routines - with near `CALL' instructions. - - (*) In models using more than one code segment (medium, large and - huge), functions are far. This means that function pointers are - 32 bits long (consisting of a 16-bit offset followed by a 16-bit - segment), and that functions are called using `CALL FAR' (or - `CALL seg:offset') and return using `RETF'. Again, you should - therefore write your own routines to return with `RETF' and use - `CALL FAR' to call external routines. - - (*) In models using a single data segment (tiny, small and medium), - data pointers are 16 bits long, containing only an offset field - (the `DS' register doesn't change its value, and always gives - the segment part of the full data item address). - - (*) In models using more than one data segment (compact, large and - huge), data pointers are 32 bits long, consisting of a 16-bit - offset followed by a 16-bit segment. You should still be careful - not to modify `DS' in your routines without restoring it - afterwards, but `ES' is free for you to use to access the - contents of 32-bit data pointers you are passed. - - (*) The huge memory model allows single data items to exceed 64K in - size. In all other memory models, you can access the whole of a - data item just by doing arithmetic on the offset field of the - pointer you are given, whether a segment field is present or - not; in huge model, you have to be more careful of your pointer - arithmetic. - - (*) In most memory models, there is a _default_ data segment, whose - segment address is kept in `DS' throughout the program. This - data segment is typically the same segment as the stack, kept in - `SS', so that functions' local variables (which are stored on - the stack) and global data items can both be accessed easily - without changing `DS'. Particularly large data items are - typically stored in other segments. However, some memory models - (though not the standard ones, usually) allow the assumption - that `SS' and `DS' hold the same value to be removed. Be careful - about functions' local variables in this latter case. - - In models with a single code segment, the segment is called `_TEXT', - so your code segment must also go by this name in order to be linked - into the same place as the main code segment. In models with a - single data segment, or with a default data segment, it is called - `_DATA'. - - 8.4.3 Function Definitions and Function Calls - - The C calling convention in 16-bit programs is as follows. In the - following description, the words _caller_ and _callee_ are used to - denote the function doing the calling and the function which gets - called. - - (*) The caller pushes the function's parameters on the stack, one - after another, in reverse order (right to left, so that the - first argument specified to the function is pushed last). - - (*) The caller then executes a `CALL' instruction to pass control to - the callee. This `CALL' is either near or far depending on the - memory model. - - (*) The callee receives control, and typically (although this is not - actually necessary, in functions which do not need to access - their parameters) starts by saving the value of `SP' in `BP' so - as to be able to use `BP' as a base pointer to find its - parameters on the stack. However, the caller was probably doing - this too, so part of the calling convention states that `BP' - must be preserved by any C function. Hence the callee, if it is - going to set up `BP' as a _frame pointer_, must push the - previous value first. - - (*) The callee may then access its parameters relative to `BP'. The - word at `[BP]' holds the previous value of `BP' as it was - pushed; the next word, at `[BP+2]', holds the offset part of the - return address, pushed implicitly by `CALL'. In a small-model - (near) function, the parameters start after that, at `[BP+4]'; - in a large-model (far) function, the segment part of the return - address lives at `[BP+4]', and the parameters begin at `[BP+6]'. - The leftmost parameter of the function, since it was pushed - last, is accessible at this offset from `BP'; the others follow, - at successively greater offsets. Thus, in a function such as - `printf' which takes a variable number of parameters, the - pushing of the parameters in reverse order means that the - function knows where to find its first parameter, which tells it - the number and type of the remaining ones. - - (*) The callee may also wish to decrease `SP' further, so as to - allocate space on the stack for local variables, which will then - be accessible at negative offsets from `BP'. - - (*) The callee, if it wishes to return a value to the caller, should - leave the value in `AL', `AX' or `DX:AX' depending on the size - of the value. Floating-point results are sometimes (depending on - the compiler) returned in `ST0'. - - (*) Once the callee has finished processing, it restores `SP' from - `BP' if it had allocated local stack space, then pops the - previous value of `BP', and returns via `RETN' or `RETF' - depending on memory model. - - (*) When the caller regains control from the callee, the function - parameters are still on the stack, so it typically adds an - immediate constant to `SP' to remove them (instead of executing - a number of slow `POP' instructions). Thus, if a function is - accidentally called with the wrong number of parameters due to a - prototype mismatch, the stack will still be returned to a - sensible state since the caller, which _knows_ how many - parameters it pushed, does the removing. - - It is instructive to compare this calling convention with that for - Pascal programs (described in section 8.5.1). Pascal has a simpler - convention, since no functions have variable numbers of parameters. - Therefore the callee knows how many parameters it should have been - passed, and is able to deallocate them from the stack itself by - passing an immediate argument to the `RET' or `RETF' instruction, so - the caller does not have to do it. Also, the parameters are pushed - in left-to-right order, not right-to-left, which means that a - compiler can give better guarantees about sequence points without - performance suffering. - - Thus, you would define a function in C style in the following way. - The following example is for small model: - - global _myfunc - - _myfunc: - push bp - mov bp,sp - sub sp,0x40 ; 64 bytes of local stack space - mov bx,[bp+4] ; first parameter to function - - ; some more code - - mov sp,bp ; undo "sub sp,0x40" above - pop bp - ret - - For a large-model function, you would replace `RET' by `RETF', and - look for the first parameter at `[BP+6]' instead of `[BP+4]'. Of - course, if one of the parameters is a pointer, then the offsets of - _subsequent_ parameters will change depending on the memory model as - well: far pointers take up four bytes on the stack when passed as a - parameter, whereas near pointers take up two. - - At the other end of the process, to call a C function from your - assembly code, you would do something like this: - - extern _printf - - ; and then, further down... - - push word [myint] ; one of my integer variables - push word mystring ; pointer into my data segment - call _printf - add sp,byte 4 ; `byte' saves space - - ; then those data items... - - segment _DATA - - myint dw 1234 - mystring db 'This number -> %d <- should be 1234',10,0 - - This piece of code is the small-model assembly equivalent of the C - code - - int myint = 1234; - printf("This number -> %d <- should be 1234\n", myint); - - In large model, the function-call code might look more like this. In - this example, it is assumed that `DS' already holds the segment base - of the segment `_DATA'. If not, you would have to initialize it - first. - - push word [myint] - push word seg mystring ; Now push the segment, and... - push word mystring ; ... offset of "mystring" - call far _printf - add sp,byte 6 - - The integer value still takes up one word on the stack, since large - model does not affect the size of the `int' data type. The first - argument (pushed last) to `printf', however, is a data pointer, and - therefore has to contain a segment and offset part. The segment - should be stored second in memory, and therefore must be pushed - first. (Of course, `PUSH DS' would have been a shorter instruction - than `PUSH WORD SEG mystring', if `DS' was set up as the above - example assumed.) Then the actual call becomes a far call, since - functions expect far calls in large model; and `SP' has to be - increased by 6 rather than 4 afterwards to make up for the extra - word of parameters. - - 8.4.4 Accessing Data Items - - To get at the contents of C variables, or to declare variables which - C can access, you need only declare the names as `GLOBAL' or - `EXTERN'. (Again, the names require leading underscores, as stated - in section 8.4.1.) Thus, a C variable declared as `int i' can be - accessed from assembler as - - extern _i - - mov ax,[_i] - - And to declare your own integer variable which C programs can access - as `extern int j', you do this (making sure you are assembling in - the `_DATA' segment, if necessary): - - global _j - - _j dw 0 - - To access a C array, you need to know the size of the components of - the array. For example, `int' variables are two bytes long, so if a - C program declares an array as `int a[10]', you can access `a[3]' by - coding `mov ax,[_a+6]'. (The byte offset 6 is obtained by - multiplying the desired array index, 3, by the size of the array - element, 2.) The sizes of the C base types in 16-bit compilers are: - 1 for `char', 2 for `short' and `int', 4 for `long' and `float', and - 8 for `double'. - - To access a C data structure, you need to know the offset from the - base of the structure to the field you are interested in. You can - either do this by converting the C structure definition into a NASM - structure definition (using `STRUC'), or by calculating the one - offset and using just that. - - To do either of these, you should read your C compiler's manual to - find out how it organizes data structures. NASM gives no special - alignment to structure members in its own `STRUC' macro, so you have - to specify alignment yourself if the C compiler generates it. - Typically, you might find that a structure like - - struct { - char c; - int i; - } foo; - - might be four bytes long rather than three, since the `int' field - would be aligned to a two-byte boundary. However, this sort of - feature tends to be a configurable option in the C compiler, either - using command-line options or `#pragma' lines, so you have to find - out how your own compiler does it. - - 8.4.5 `c16.mac': Helper Macros for the 16-bit C Interface - - Included in the NASM archives, in the `misc' directory, is a file - `c16.mac' of macros. It defines three macros: `proc', `arg' and - `endproc'. These are intended to be used for C-style procedure - definitions, and they automate a lot of the work involved in keeping - track of the calling convention. - - (An alternative, TASM compatible form of `arg' is also now built - into NASM's preprocessor. See section 4.8 for details.) - - An example of an assembly function using the macro set is given - here: - - proc _nearproc - - %$i arg - %$j arg - mov ax,[bp + %$i] - mov bx,[bp + %$j] - add ax,[bx] - - endproc - - This defines `_nearproc' to be a procedure taking two arguments, the - first (`i') an integer and the second (`j') a pointer to an integer. - It returns `i + *j'. - - Note that the `arg' macro has an `EQU' as the first line of its - expansion, and since the label before the macro call gets prepended - to the first line of the expanded macro, the `EQU' works, defining - `%$i' to be an offset from `BP'. A context-local variable is used, - local to the context pushed by the `proc' macro and popped by the - `endproc' macro, so that the same argument name can be used in later - procedures. Of course, you don't _have_ to do that. - - The macro set produces code for near functions (tiny, small and - compact-model code) by default. You can have it generate far - functions (medium, large and huge-model code) by means of coding - `%define FARCODE'. This changes the kind of return instruction - generated by `endproc', and also changes the starting point for the - argument offsets. The macro set contains no intrinsic dependency on - whether data pointers are far or not. - - `arg' can take an optional parameter, giving the size of the - argument. If no size is given, 2 is assumed, since it is likely that - many function parameters will be of type `int'. - - The large-model equivalent of the above function would look like - this: - - %define FARCODE - - proc _farproc - - %$i arg - %$j arg 4 - mov ax,[bp + %$i] - mov bx,[bp + %$j] - mov es,[bp + %$j + 2] - add ax,[bx] - - endproc - - This makes use of the argument to the `arg' macro to define a - parameter of size 4, because `j' is now a far pointer. When we load - from `j', we must load a segment and an offset. - - 8.5 Interfacing to Borland Pascal Programs - - Interfacing to Borland Pascal programs is similar in concept to - interfacing to 16-bit C programs. The differences are: - - (*) The leading underscore required for interfacing to C programs is - not required for Pascal. - - (*) The memory model is always large: functions are far, data - pointers are far, and no data item can be more than 64K long. - (Actually, some functions are near, but only those functions - that are local to a Pascal unit and never called from outside - it. All assembly functions that Pascal calls, and all Pascal - functions that assembly routines are able to call, are far.) - However, all static data declared in a Pascal program goes into - the default data segment, which is the one whose segment address - will be in `DS' when control is passed to your assembly code. - The only things that do not live in the default data segment are - local variables (they live in the stack segment) and dynamically - allocated variables. All data _pointers_, however, are far. - - (*) The function calling convention is different - described below. - - (*) Some data types, such as strings, are stored differently. - - (*) There are restrictions on the segment names you are allowed to - use - Borland Pascal will ignore code or data declared in a - segment it doesn't like the name of. The restrictions are - described below. - - 8.5.1 The Pascal Calling Convention - - The 16-bit Pascal calling convention is as follows. In the following - description, the words _caller_ and _callee_ are used to denote the - function doing the calling and the function which gets called. - - (*) The caller pushes the function's parameters on the stack, one - after another, in normal order (left to right, so that the first - argument specified to the function is pushed first). - - (*) The caller then executes a far `CALL' instruction to pass - control to the callee. - - (*) The callee receives control, and typically (although this is not - actually necessary, in functions which do not need to access - their parameters) starts by saving the value of `SP' in `BP' so - as to be able to use `BP' as a base pointer to find its - parameters on the stack. However, the caller was probably doing - this too, so part of the calling convention states that `BP' - must be preserved by any function. Hence the callee, if it is - going to set up `BP' as a frame pointer, must push the previous - value first. - - (*) The callee may then access its parameters relative to `BP'. The - word at `[BP]' holds the previous value of `BP' as it was - pushed. The next word, at `[BP+2]', holds the offset part of the - return address, and the next one at `[BP+4]' the segment part. - The parameters begin at `[BP+6]'. The rightmost parameter of the - function, since it was pushed last, is accessible at this offset - from `BP'; the others follow, at successively greater offsets. - - (*) The callee may also wish to decrease `SP' further, so as to - allocate space on the stack for local variables, which will then - be accessible at negative offsets from `BP'. - - (*) The callee, if it wishes to return a value to the caller, should - leave the value in `AL', `AX' or `DX:AX' depending on the size - of the value. Floating-point results are returned in `ST0'. - Results of type `Real' (Borland's own custom floating-point data - type, not handled directly by the FPU) are returned in - `DX:BX:AX'. To return a result of type `String', the caller - pushes a pointer to a temporary string before pushing the - parameters, and the callee places the returned string value at - that location. The pointer is not a parameter, and should not be - removed from the stack by the `RETF' instruction. - - (*) Once the callee has finished processing, it restores `SP' from - `BP' if it had allocated local stack space, then pops the - previous value of `BP', and returns via `RETF'. It uses the form - of `RETF' with an immediate parameter, giving the number of - bytes taken up by the parameters on the stack. This causes the - parameters to be removed from the stack as a side effect of the - return instruction. - - (*) When the caller regains control from the callee, the function - parameters have already been removed from the stack, so it needs - to do nothing further. - - Thus, you would define a function in Pascal style, taking two - `Integer'-type parameters, in the following way: - - global myfunc - - myfunc: push bp - mov bp,sp - sub sp,0x40 ; 64 bytes of local stack space - mov bx,[bp+8] ; first parameter to function - mov bx,[bp+6] ; second parameter to function - - ; some more code - - mov sp,bp ; undo "sub sp,0x40" above - pop bp - retf 4 ; total size of params is 4 - - At the other end of the process, to call a Pascal function from your - assembly code, you would do something like this: - - extern SomeFunc - - ; and then, further down... - - push word seg mystring ; Now push the segment, and... - push word mystring ; ... offset of "mystring" - push word [myint] ; one of my variables - call far SomeFunc - - This is equivalent to the Pascal code - - procedure SomeFunc(String: PChar; Int: Integer); - SomeFunc(@mystring, myint); - - 8.5.2 Borland Pascal Segment Name Restrictions - - Since Borland Pascal's internal unit file format is completely - different from `OBJ', it only makes a very sketchy job of actually - reading and understanding the various information contained in a - real `OBJ' file when it links that in. Therefore an object file - intended to be linked to a Pascal program must obey a number of - restrictions: - - (*) Procedures and functions must be in a segment whose name is - either `CODE', `CSEG', or something ending in `_TEXT'. - - (*) initialized data must be in a segment whose name is either - `CONST' or something ending in `_DATA'. - - (*) Uninitialized data must be in a segment whose name is either - `DATA', `DSEG', or something ending in `_BSS'. - - (*) Any other segments in the object file are completely ignored. - `GROUP' directives and segment attributes are also ignored. - - 8.5.3 Using `c16.mac' With Pascal Programs - - The `c16.mac' macro package, described in section 8.4.5, can also be - used to simplify writing functions to be called from Pascal - programs, if you code `%define PASCAL'. This definition ensures that - functions are far (it implies `FARCODE'), and also causes procedure - return instructions to be generated with an operand. - - Defining `PASCAL' does not change the code which calculates the - argument offsets; you must declare your function's arguments in - reverse order. For example: - - %define PASCAL - - proc _pascalproc - - %$j arg 4 - %$i arg - mov ax,[bp + %$i] - mov bx,[bp + %$j] - mov es,[bp + %$j + 2] - add ax,[bx] - - endproc - - This defines the same routine, conceptually, as the example in - section 8.4.5: it defines a function taking two arguments, an - integer and a pointer to an integer, which returns the sum of the - integer and the contents of the pointer. The only difference between - this code and the large-model C version is that `PASCAL' is defined - instead of `FARCODE', and that the arguments are declared in reverse - order. - -Chapter 9: Writing 32-bit Code (Unix, Win32, DJGPP) ---------------------------------------------------- - - This chapter attempts to cover some of the common issues involved - when writing 32-bit code, to run under Win32 or Unix, or to be - linked with C code generated by a Unix-style C compiler such as - DJGPP. It covers how to write assembly code to interface with 32-bit - C routines, and how to write position-independent code for shared - libraries. - - Almost all 32-bit code, and in particular all code running under - `Win32', `DJGPP' or any of the PC Unix variants, runs in _flat_ - memory model. This means that the segment registers and paging have - already been set up to give you the same 32-bit 4Gb address space no - matter what segment you work relative to, and that you should ignore - all segment registers completely. When writing flat-model - application code, you never need to use a segment override or modify - any segment register, and the code-section addresses you pass to - `CALL' and `JMP' live in the same address space as the data-section - addresses you access your variables by and the stack-section - addresses you access local variables and procedure parameters by. - Every address is 32 bits long and contains only an offset part. - - 9.1 Interfacing to 32-bit C Programs - - A lot of the discussion in section 8.4, about interfacing to 16-bit - C programs, still applies when working in 32 bits. The absence of - memory models or segmentation worries simplifies things a lot. - - 9.1.1 External Symbol Names - - Most 32-bit C compilers share the convention used by 16-bit - compilers, that the names of all global symbols (functions or data) - they define are formed by prefixing an underscore to the name as it - appears in the C program. However, not all of them do: the `ELF' - specification states that C symbols do _not_ have a leading - underscore on their assembly-language names. - - The older Linux `a.out' C compiler, all `Win32' compilers, `DJGPP', - and `NetBSD' and `FreeBSD', all use the leading underscore; for - these compilers, the macros `cextern' and `cglobal', as given in - section 8.4.1, will still work. For `ELF', though, the leading - underscore should not be used. - - See also section 2.1.27. - - 9.1.2 Function Definitions and Function Calls - - The C calling convention in 32-bit programs is as follows. In the - following description, the words _caller_ and _callee_ are used to - denote the function doing the calling and the function which gets - called. - - (*) The caller pushes the function's parameters on the stack, one - after another, in reverse order (right to left, so that the - first argument specified to the function is pushed last). - - (*) The caller then executes a near `CALL' instruction to pass - control to the callee. - - (*) The callee receives control, and typically (although this is not - actually necessary, in functions which do not need to access - their parameters) starts by saving the value of `ESP' in `EBP' - so as to be able to use `EBP' as a base pointer to find its - parameters on the stack. However, the caller was probably doing - this too, so part of the calling convention states that `EBP' - must be preserved by any C function. Hence the callee, if it is - going to set up `EBP' as a frame pointer, must push the previous - value first. - - (*) The callee may then access its parameters relative to `EBP'. The - doubleword at `[EBP]' holds the previous value of `EBP' as it - was pushed; the next doubleword, at `[EBP+4]', holds the return - address, pushed implicitly by `CALL'. The parameters start after - that, at `[EBP+8]'. The leftmost parameter of the function, - since it was pushed last, is accessible at this offset from - `EBP'; the others follow, at successively greater offsets. Thus, - in a function such as `printf' which takes a variable number of - parameters, the pushing of the parameters in reverse order means - that the function knows where to find its first parameter, which - tells it the number and type of the remaining ones. - - (*) The callee may also wish to decrease `ESP' further, so as to - allocate space on the stack for local variables, which will then - be accessible at negative offsets from `EBP'. - - (*) The callee, if it wishes to return a value to the caller, should - leave the value in `AL', `AX' or `EAX' depending on the size of - the value. Floating-point results are typically returned in - `ST0'. - - (*) Once the callee has finished processing, it restores `ESP' from - `EBP' if it had allocated local stack space, then pops the - previous value of `EBP', and returns via `RET' (equivalently, - `RETN'). - - (*) When the caller regains control from the callee, the function - parameters are still on the stack, so it typically adds an - immediate constant to `ESP' to remove them (instead of executing - a number of slow `POP' instructions). Thus, if a function is - accidentally called with the wrong number of parameters due to a - prototype mismatch, the stack will still be returned to a - sensible state since the caller, which _knows_ how many - parameters it pushed, does the removing. - - There is an alternative calling convention used by Win32 programs - for Windows API calls, and also for functions called _by_ the - Windows API such as window procedures: they follow what Microsoft - calls the `__stdcall' convention. This is slightly closer to the - Pascal convention, in that the callee clears the stack by passing a - parameter to the `RET' instruction. However, the parameters are - still pushed in right-to-left order. - - Thus, you would define a function in C style in the following way: - - global _myfunc - - _myfunc: - push ebp - mov ebp,esp - sub esp,0x40 ; 64 bytes of local stack space - mov ebx,[ebp+8] ; first parameter to function - - ; some more code - - leave ; mov esp,ebp / pop ebp - ret - - At the other end of the process, to call a C function from your - assembly code, you would do something like this: - - extern _printf - - ; and then, further down... - - push dword [myint] ; one of my integer variables - push dword mystring ; pointer into my data segment - call _printf - add esp,byte 8 ; `byte' saves space - - ; then those data items... - - segment _DATA - - myint dd 1234 - mystring db 'This number -> %d <- should be 1234',10,0 - - This piece of code is the assembly equivalent of the C code - - int myint = 1234; - printf("This number -> %d <- should be 1234\n", myint); - - 9.1.3 Accessing Data Items - - To get at the contents of C variables, or to declare variables which - C can access, you need only declare the names as `GLOBAL' or - `EXTERN'. (Again, the names require leading underscores, as stated - in section 9.1.1.) Thus, a C variable declared as `int i' can be - accessed from assembler as - - extern _i - mov eax,[_i] - - And to declare your own integer variable which C programs can access - as `extern int j', you do this (making sure you are assembling in - the `_DATA' segment, if necessary): - - global _j - _j dd 0 - - To access a C array, you need to know the size of the components of - the array. For example, `int' variables are four bytes long, so if a - C program declares an array as `int a[10]', you can access `a[3]' by - coding `mov ax,[_a+12]'. (The byte offset 12 is obtained by - multiplying the desired array index, 3, by the size of the array - element, 4.) The sizes of the C base types in 32-bit compilers are: - 1 for `char', 2 for `short', 4 for `int', `long' and `float', and 8 - for `double'. Pointers, being 32-bit addresses, are also 4 bytes - long. - - To access a C data structure, you need to know the offset from the - base of the structure to the field you are interested in. You can - either do this by converting the C structure definition into a NASM - structure definition (using `STRUC'), or by calculating the one - offset and using just that. - - To do either of these, you should read your C compiler's manual to - find out how it organizes data structures. NASM gives no special - alignment to structure members in its own `STRUC' macro, so you have - to specify alignment yourself if the C compiler generates it. - Typically, you might find that a structure like - - struct { - char c; - int i; - } foo; - - might be eight bytes long rather than five, since the `int' field - would be aligned to a four-byte boundary. However, this sort of - feature is sometimes a configurable option in the C compiler, either - using command-line options or `#pragma' lines, so you have to find - out how your own compiler does it. - - 9.1.4 `c32.mac': Helper Macros for the 32-bit C Interface - - Included in the NASM archives, in the `misc' directory, is a file - `c32.mac' of macros. It defines three macros: `proc', `arg' and - `endproc'. These are intended to be used for C-style procedure - definitions, and they automate a lot of the work involved in keeping - track of the calling convention. - - An example of an assembly function using the macro set is given - here: - - proc _proc32 - - %$i arg - %$j arg - mov eax,[ebp + %$i] - mov ebx,[ebp + %$j] - add eax,[ebx] - - endproc - - This defines `_proc32' to be a procedure taking two arguments, the - first (`i') an integer and the second (`j') a pointer to an integer. - It returns `i + *j'. - - Note that the `arg' macro has an `EQU' as the first line of its - expansion, and since the label before the macro call gets prepended - to the first line of the expanded macro, the `EQU' works, defining - `%$i' to be an offset from `BP'. A context-local variable is used, - local to the context pushed by the `proc' macro and popped by the - `endproc' macro, so that the same argument name can be used in later - procedures. Of course, you don't _have_ to do that. - - `arg' can take an optional parameter, giving the size of the - argument. If no size is given, 4 is assumed, since it is likely that - many function parameters will be of type `int' or pointers. - - 9.2 Writing NetBSD/FreeBSD/OpenBSD and Linux/ELF Shared Libraries - - `ELF' replaced the older `a.out' object file format under Linux - because it contains support for position-independent code (PIC), - which makes writing shared libraries much easier. NASM supports the - `ELF' position-independent code features, so you can write Linux - `ELF' shared libraries in NASM. - - NetBSD, and its close cousins FreeBSD and OpenBSD, take a different - approach by hacking PIC support into the `a.out' format. NASM - supports this as the `aoutb' output format, so you can write BSD - shared libraries in NASM too. - - The operating system loads a PIC shared library by memory-mapping - the library file at an arbitrarily chosen point in the address space - of the running process. The contents of the library's code section - must therefore not depend on where it is loaded in memory. - - Therefore, you cannot get at your variables by writing code like - this: - - mov eax,[myvar] ; WRONG - - Instead, the linker provides an area of memory called the _global - offset table_, or GOT; the GOT is situated at a constant distance - from your library's code, so if you can find out where your library - is loaded (which is typically done using a `CALL' and `POP' - combination), you can obtain the address of the GOT, and you can - then load the addresses of your variables out of linker-generated - entries in the GOT. - - The _data_ section of a PIC shared library does not have these - restrictions: since the data section is writable, it has to be - copied into memory anyway rather than just paged in from the library - file, so as long as it's being copied it can be relocated too. So - you can put ordinary types of relocation in the data section without - too much worry (but see section 9.2.4 for a caveat). - - 9.2.1 Obtaining the Address of the GOT - - Each code module in your shared library should define the GOT as an - external symbol: - - extern _GLOBAL_OFFSET_TABLE_ ; in ELF - extern __GLOBAL_OFFSET_TABLE_ ; in BSD a.out - - At the beginning of any function in your shared library which plans - to access your data or BSS sections, you must first calculate the - address of the GOT. This is typically done by writing the function - in this form: - - func: push ebp - mov ebp,esp - push ebx - call .get_GOT - .get_GOT: - pop ebx - add ebx,_GLOBAL_OFFSET_TABLE_+$$-.get_GOT wrt ..gotpc - - ; the function body comes here - - mov ebx,[ebp-4] - mov esp,ebp - pop ebp - ret - - (For BSD, again, the symbol `_GLOBAL_OFFSET_TABLE' requires a second - leading underscore.) - - The first two lines of this function are simply the standard C - prologue to set up a stack frame, and the last three lines are - standard C function epilogue. The third line, and the fourth to last - line, save and restore the `EBX' register, because PIC shared - libraries use this register to store the address of the GOT. - - The interesting bit is the `CALL' instruction and the following two - lines. The `CALL' and `POP' combination obtains the address of the - label `.get_GOT', without having to know in advance where the - program was loaded (since the `CALL' instruction is encoded relative - to the current position). The `ADD' instruction makes use of one of - the special PIC relocation types: GOTPC relocation. With the - `WRT ..gotpc' qualifier specified, the symbol referenced (here - `_GLOBAL_OFFSET_TABLE_', the special symbol assigned to the GOT) is - given as an offset from the beginning of the section. (Actually, - `ELF' encodes it as the offset from the operand field of the `ADD' - instruction, but NASM simplifies this deliberately, so you do things - the same way for both `ELF' and `BSD'.) So the instruction then - _adds_ the beginning of the section, to get the real address of the - GOT, and subtracts the value of `.get_GOT' which it knows is in - `EBX'. Therefore, by the time that instruction has finished, `EBX' - contains the address of the GOT. - - If you didn't follow that, don't worry: it's never necessary to - obtain the address of the GOT by any other means, so you can put - those three instructions into a macro and safely ignore them: - - %macro get_GOT 0 - - call %%getgot - %%getgot: - pop ebx - add ebx,_GLOBAL_OFFSET_TABLE_+$$-%%getgot wrt ..gotpc - - %endmacro - - 9.2.2 Finding Your Local Data Items - - Having got the GOT, you can then use it to obtain the addresses of - your data items. Most variables will reside in the sections you have - declared; they can be accessed using the `..gotoff' special `WRT' - type. The way this works is like this: - - lea eax,[ebx+myvar wrt ..gotoff] - - The expression `myvar wrt ..gotoff' is calculated, when the shared - library is linked, to be the offset to the local variable `myvar' - from the beginning of the GOT. Therefore, adding it to `EBX' as - above will place the real address of `myvar' in `EAX'. - - If you declare variables as `GLOBAL' without specifying a size for - them, they are shared between code modules in the library, but do - not get exported from the library to the program that loaded it. - They will still be in your ordinary data and BSS sections, so you - can access them in the same way as local variables, using the above - `..gotoff' mechanism. - - Note that due to a peculiarity of the way BSD `a.out' format handles - this relocation type, there must be at least one non-local symbol in - the same section as the address you're trying to access. - - 9.2.3 Finding External and Common Data Items - - If your library needs to get at an external variable (external to - the _library_, not just to one of the modules within it), you must - use the `..got' type to get at it. The `..got' type, instead of - giving you the offset from the GOT base to the variable, gives you - the offset from the GOT base to a GOT _entry_ containing the address - of the variable. The linker will set up this GOT entry when it - builds the library, and the dynamic linker will place the correct - address in it at load time. So to obtain the address of an external - variable `extvar' in `EAX', you would code - - mov eax,[ebx+extvar wrt ..got] - - This loads the address of `extvar' out of an entry in the GOT. The - linker, when it builds the shared library, collects together every - relocation of type `..got', and builds the GOT so as to ensure it - has every necessary entry present. - - Common variables must also be accessed in this way. - - 9.2.4 Exporting Symbols to the Library User - - If you want to export symbols to the user of the library, you have - to declare whether they are functions or data, and if they are data, - you have to give the size of the data item. This is because the - dynamic linker has to build procedure linkage table entries for any - exported functions, and also moves exported data items away from the - library's data section in which they were declared. - - So to export a function to users of the library, you must use - - global func:function ; declare it as a function - - func: push ebp - - ; etc. - - And to export a data item such as an array, you would have to code - - global array:data array.end-array ; give the size too - - array: resd 128 - .end: - - Be careful: If you export a variable to the library user, by - declaring it as `GLOBAL' and supplying a size, the variable will end - up living in the data section of the main program, rather than in - your library's data section, where you declared it. So you will have - to access your own global variable with the `..got' mechanism rather - than `..gotoff', as if it were external (which, effectively, it has - become). - - Equally, if you need to store the address of an exported global in - one of your data sections, you can't do it by means of the standard - sort of code: - - dataptr: dd global_data_item ; WRONG - - NASM will interpret this code as an ordinary relocation, in which - `global_data_item' is merely an offset from the beginning of the - `.data' section (or whatever); so this reference will end up - pointing at your data section instead of at the exported global - which resides elsewhere. - - Instead of the above code, then, you must write - - dataptr: dd global_data_item wrt ..sym - - which makes use of the special `WRT' type `..sym' to instruct NASM - to search the symbol table for a particular symbol at that address, - rather than just relocating by section base. - - Either method will work for functions: referring to one of your - functions by means of - - funcptr: dd my_function - - will give the user the address of the code you wrote, whereas - - funcptr: dd my_function wrt .sym - - will give the address of the procedure linkage table for the - function, which is where the calling program will _believe_ the - function lives. Either address is a valid way to call the function. - - 9.2.5 Calling Procedures Outside the Library - - Calling procedures outside your shared library has to be done by - means of a _procedure linkage table_, or PLT. The PLT is placed at a - known offset from where the library is loaded, so the library code - can make calls to the PLT in a position-independent way. Within the - PLT there is code to jump to offsets contained in the GOT, so - function calls to other shared libraries or to routines in the main - program can be transparently passed off to their real destinations. - - To call an external routine, you must use another special PIC - relocation type, `WRT ..plt'. This is much easier than the GOT-based - ones: you simply replace calls such as `CALL printf' with the PLT- - relative version `CALL printf WRT ..plt'. - - 9.2.6 Generating the Library File - - Having written some code modules and assembled them to `.o' files, - you then generate your shared library with a command such as - - ld -shared -o library.so module1.o module2.o # for ELF - ld -Bshareable -o library.so module1.o module2.o # for BSD - - For ELF, if your shared library is going to reside in system - directories such as `/usr/lib' or `/lib', it is usually worth using - the `-soname' flag to the linker, to store the final library file - name, with a version number, into the library: - - ld -shared -soname library.so.1 -o library.so.1.2 *.o - - You would then copy `library.so.1.2' into the library directory, and - create `library.so.1' as a symbolic link to it. - -Chapter 10: Mixing 16 and 32 Bit Code -------------------------------------- - - This chapter tries to cover some of the issues, largely related to - unusual forms of addressing and jump instructions, encountered when - writing operating system code such as protected-mode initialisation - routines, which require code that operates in mixed segment sizes, - such as code in a 16-bit segment trying to modify data in a 32-bit - one, or jumps between different-size segments. - - 10.1 Mixed-Size Jumps - - The most common form of mixed-size instruction is the one used when - writing a 32-bit OS: having done your setup in 16-bit mode, such as - loading the kernel, you then have to boot it by switching into - protected mode and jumping to the 32-bit kernel start address. In a - fully 32-bit OS, this tends to be the _only_ mixed-size instruction - you need, since everything before it can be done in pure 16-bit - code, and everything after it can be pure 32-bit. - - This jump must specify a 48-bit far address, since the target - segment is a 32-bit one. However, it must be assembled in a 16-bit - segment, so just coding, for example, - - jmp 0x1234:0x56789ABC ; wrong! - - will not work, since the offset part of the address will be - truncated to `0x9ABC' and the jump will be an ordinary 16-bit far - one. - - The Linux kernel setup code gets round the inability of `as86' to - generate the required instruction by coding it manually, using `DB' - instructions. NASM can go one better than that, by actually - generating the right instruction itself. Here's how to do it right: - - jmp dword 0x1234:0x56789ABC ; right - - The `DWORD' prefix (strictly speaking, it should come _after_ the - colon, since it is declaring the _offset_ field to be a doubleword; - but NASM will accept either form, since both are unambiguous) forces - the offset part to be treated as far, in the assumption that you are - deliberately writing a jump from a 16-bit segment to a 32-bit one. - - You can do the reverse operation, jumping from a 32-bit segment to a - 16-bit one, by means of the `WORD' prefix: - - jmp word 0x8765:0x4321 ; 32 to 16 bit - - If the `WORD' prefix is specified in 16-bit mode, or the `DWORD' - prefix in 32-bit mode, they will be ignored, since each is - explicitly forcing NASM into a mode it was in anyway. - - 10.2 Addressing Between Different-Size Segments - - If your OS is mixed 16 and 32-bit, or if you are writing a DOS - extender, you are likely to have to deal with some 16-bit segments - and some 32-bit ones. At some point, you will probably end up - writing code in a 16-bit segment which has to access data in a 32- - bit segment, or vice versa. - - If the data you are trying to access in a 32-bit segment lies within - the first 64K of the segment, you may be able to get away with using - an ordinary 16-bit addressing operation for the purpose; but sooner - or later, you will want to do 32-bit addressing from 16-bit mode. - - The easiest way to do this is to make sure you use a register for - the address, since any effective address containing a 32-bit - register is forced to be a 32-bit address. So you can do - - mov eax,offset_into_32_bit_segment_specified_by_fs - mov dword [fs:eax],0x11223344 - - This is fine, but slightly cumbersome (since it wastes an - instruction and a register) if you already know the precise offset - you are aiming at. The x86 architecture does allow 32-bit effective - addresses to specify nothing but a 4-byte offset, so why shouldn't - NASM be able to generate the best instruction for the purpose? - - It can. As in section 10.1, you need only prefix the address with - the `DWORD' keyword, and it will be forced to be a 32-bit address: - - mov dword [fs:dword my_offset],0x11223344 - - Also as in section 10.1, NASM is not fussy about whether the `DWORD' - prefix comes before or after the segment override, so arguably a - nicer-looking way to code the above instruction is - - mov dword [dword fs:my_offset],0x11223344 - - Don't confuse the `DWORD' prefix _outside_ the square brackets, - which controls the size of the data stored at the address, with the - one `inside' the square brackets which controls the length of the - address itself. The two can quite easily be different: - - mov word [dword 0x12345678],0x9ABC - - This moves 16 bits of data to an address specified by a 32-bit - offset. - - You can also specify `WORD' or `DWORD' prefixes along with the `FAR' - prefix to indirect far jumps or calls. For example: - - call dword far [fs:word 0x4321] - - This instruction contains an address specified by a 16-bit offset; - it loads a 48-bit far pointer from that (16-bit segment and 32-bit - offset), and calls that address. - - 10.3 Other Mixed-Size Instructions - - The other way you might want to access data might be using the - string instructions (`LODSx', `STOSx' and so on) or the `XLATB' - instruction. These instructions, since they take no parameters, - might seem to have no easy way to make them perform 32-bit - addressing when assembled in a 16-bit segment. - - This is the purpose of NASM's `a16', `a32' and `a64' prefixes. If - you are coding `LODSB' in a 16-bit segment but it is supposed to be - accessing a string in a 32-bit segment, you should load the desired - address into `ESI' and then code - - a32 lodsb - - The prefix forces the addressing size to 32 bits, meaning that - `LODSB' loads from `[DS:ESI]' instead of `[DS:SI]'. To access a - string in a 16-bit segment when coding in a 32-bit one, the - corresponding `a16' prefix can be used. - - The `a16', `a32' and `a64' prefixes can be applied to any - instruction in NASM's instruction table, but most of them can - generate all the useful forms without them. The prefixes are - necessary only for instructions with implicit addressing: `CMPSx', - `SCASx', `LODSx', `STOSx', `MOVSx', `INSx', `OUTSx', and `XLATB'. - Also, the various push and pop instructions (`PUSHA' and `POPF' as - well as the more usual `PUSH' and `POP') can accept `a16', `a32' or - `a64' prefixes to force a particular one of `SP', `ESP' or `RSP' to - be used as a stack pointer, in case the stack segment in use is a - different size from the code segment. - - `PUSH' and `POP', when applied to segment registers in 32-bit mode, - also have the slightly odd behaviour that they push and pop 4 bytes - at a time, of which the top two are ignored and the bottom two give - the value of the segment register being manipulated. To force the - 16-bit behaviour of segment-register push and pop instructions, you - can use the operand-size prefix `o16': - - o16 push ss - o16 push ds - - This code saves a doubleword of stack space by fitting two segment - registers into the space which would normally be consumed by pushing - one. - - (You can also use the `o32' prefix to force the 32-bit behaviour - when in 16-bit mode, but this seems less useful.) - -Chapter 11: Writing 64-bit Code (Unix, Win64) ---------------------------------------------- - - This chapter attempts to cover some of the common issues involved - when writing 64-bit code, to run under Win64 or Unix. It covers how - to write assembly code to interface with 64-bit C routines, and how - to write position-independent code for shared libraries. - - All 64-bit code uses a flat memory model, since segmentation is not - available in 64-bit mode. The one exception is the `FS' and `GS' - registers, which still add their bases. - - Position independence in 64-bit mode is significantly simpler, since - the processor supports `RIP'-relative addressing directly; see the - `REL' keyword (section 3.3). On most 64-bit platforms, it is - probably desirable to make that the default, using the directive - `DEFAULT REL' (section 6.2). - - 64-bit programming is relatively similar to 32-bit programming, but - of course pointers are 64 bits long; additionally, all existing - platforms pass arguments in registers rather than on the stack. - Furthermore, 64-bit platforms use SSE2 by default for floating - point. Please see the ABI documentation for your platform. - - 64-bit platforms differ in the sizes of the fundamental datatypes, - not just from 32-bit platforms but from each other. If a specific - size data type is desired, it is probably best to use the types - defined in the Standard C header `<inttypes.h>'. - - In 64-bit mode, the default instruction size is still 32 bits. When - loading a value into a 32-bit register (but not an 8- or 16-bit - register), the upper 32 bits of the corresponding 64-bit register - are set to zero. - - 11.1 Register Names in 64-bit Mode - - NASM uses the following names for general-purpose registers in 64- - bit mode, for 8-, 16-, 32- and 64-bit references, respecitively: - - AL/AH, CL/CH, DL/DH, BL/BH, SPL, BPL, SIL, DIL, R8B-R15B - AX, CX, DX, BX, SP, BP, SI, DI, R8W-R15W - EAX, ECX, EDX, EBX, ESP, EBP, ESI, EDI, R8D-R15D - RAX, RCX, RDX, RBX, RSP, RBP, RSI, RDI, R8-R15 - - This is consistent with the AMD documentation and most other - assemblers. The Intel documentation, however, uses the names - `R8L-R15L' for 8-bit references to the higher registers. It is - possible to use those names by definiting them as macros; similarly, - if one wants to use numeric names for the low 8 registers, define - them as macros. The standard macro package `altreg' (see section - 5.1) can be used for this purpose. - - 11.2 Immediates and Displacements in 64-bit Mode - - In 64-bit mode, immediates and displacements are generally only 32 - bits wide. NASM will therefore truncate most displacements and - immediates to 32 bits. - - The only instruction which takes a full 64-bit immediate is: - - MOV reg64,imm64 - - NASM will produce this instruction whenever the programmer uses - `MOV' with an immediate into a 64-bit register. If this is not - desirable, simply specify the equivalent 32-bit register, which will - be automatically zero-extended by the processor, or specify the - immediate as `DWORD': - - mov rax,foo ; 64-bit immediate - mov rax,qword foo ; (identical) - mov eax,foo ; 32-bit immediate, zero-extended - mov rax,dword foo ; 32-bit immediate, sign-extended - - The length of these instructions are 10, 5 and 7 bytes, - respectively. - - The only instructions which take a full 64-bit _displacement_ is - loading or storing, using `MOV', `AL', `AX', `EAX' or `RAX' (but no - other registers) to an absolute 64-bit address. Since this is a - relatively rarely used instruction (64-bit code generally uses - relative addressing), the programmer has to explicitly declare the - displacement size as `QWORD': - - default abs - - mov eax,[foo] ; 32-bit absolute disp, sign-extended - mov eax,[a32 foo] ; 32-bit absolute disp, zero-extended - mov eax,[qword foo] ; 64-bit absolute disp - - default rel - - mov eax,[foo] ; 32-bit relative disp - mov eax,[a32 foo] ; d:o, address truncated to 32 bits(!) - mov eax,[qword foo] ; error - mov eax,[abs qword foo] ; 64-bit absolute disp - - A sign-extended absolute displacement can access from -2 GB to +2 - GB; a zero-extended absolute displacement can access from 0 to 4 GB. - - 11.3 Interfacing to 64-bit C Programs (Unix) - - On Unix, the 64-bit ABI is defined by the document: - - `http://www.x86-64.org/documentation/abi.pdf' - - Although written for AT&T-syntax assembly, the concepts apply - equally well for NASM-style assembly. What follows is a simplified - summary. - - The first six integer arguments (from the left) are passed in `RDI', - `RSI', `RDX', `RCX', `R8', and `R9', in that order. Additional - integer arguments are passed on the stack. These registers, plus - `RAX', `R10' and `R11' are destroyed by function calls, and thus are - available for use by the function without saving. - - Integer return values are passed in `RAX' and `RDX', in that order. - - Floating point is done using SSE registers, except for - `long double'. Floating-point arguments are passed in `XMM0' to - `XMM7'; return is `XMM0' and `XMM1'. `long double' are passed on the - stack, and returned in `ST0' and `ST1'. - - All SSE and x87 registers are destroyed by function calls. - - On 64-bit Unix, `long' is 64 bits. - - Integer and SSE register arguments are counted separately, so for - the case of - - void foo(long a, double b, int c) - - `a' is passed in `RDI', `b' in `XMM0', and `c' in `ESI'. - - 11.4 Interfacing to 64-bit C Programs (Win64) - - The Win64 ABI is described at: - - `http://msdn2.microsoft.com/en-gb/library/ms794533.aspx' - - What follows is a simplified summary. - - The first four integer arguments are passed in `RCX', `RDX', `R8' - and `R9', in that order. Additional integer arguments are passed on - the stack. These registers, plus `RAX', `R10' and `R11' are - destroyed by function calls, and thus are available for use by the - function without saving. - - Integer return values are passed in `RAX' only. - - Floating point is done using SSE registers, except for - `long double'. Floating-point arguments are passed in `XMM0' to - `XMM3'; return is `XMM0' only. - - On Win64, `long' is 32 bits; `long long' or `_int64' is 64 bits. - - Integer and SSE register arguments are counted together, so for the - case of - - void foo(long long a, double b, int c) - - `a' is passed in `RCX', `b' in `XMM1', and `c' in `R8D'. - -Chapter 12: Troubleshooting ---------------------------- - - This chapter describes some of the common problems that users have - been known to encounter with NASM, and answers them. It also gives - instructions for reporting bugs in NASM if you find a difficulty - that isn't listed here. - - 12.1 Common Problems - -12.1.1 NASM Generates Inefficient Code - - We sometimes get `bug' reports about NASM generating inefficient, or - even `wrong', code on instructions such as `ADD ESP,8'. This is a - deliberate design feature, connected to predictability of output: - NASM, on seeing `ADD ESP,8', will generate the form of the - instruction which leaves room for a 32-bit offset. You need to code - `ADD ESP,BYTE 8' if you want the space-efficient form of the - instruction. This isn't a bug, it's user error: if you prefer to - have NASM produce the more efficient code automatically enable - optimization with the `-O' option (see section 2.1.22). - -12.1.2 My Jumps are Out of Range - - Similarly, people complain that when they issue conditional jumps - (which are `SHORT' by default) that try to jump too far, NASM - reports `short jump out of range' instead of making the jumps - longer. - - This, again, is partly a predictability issue, but in fact has a - more practical reason as well. NASM has no means of being told what - type of processor the code it is generating will be run on; so it - cannot decide for itself that it should generate `Jcc NEAR' type - instructions, because it doesn't know that it's working for a 386 or - above. Alternatively, it could replace the out-of-range short `JNE' - instruction with a very short `JE' instruction that jumps over a - `JMP NEAR'; this is a sensible solution for processors below a 386, - but hardly efficient on processors which have good branch prediction - _and_ could have used `JNE NEAR' instead. So, once again, it's up to - the user, not the assembler, to decide what instructions should be - generated. See section 2.1.22. - -12.1.3 `ORG' Doesn't Work - - People writing boot sector programs in the `bin' format often - complain that `ORG' doesn't work the way they'd like: in order to - place the `0xAA55' signature word at the end of a 512-byte boot - sector, people who are used to MASM tend to code - - ORG 0 - - ; some boot sector code - - ORG 510 - DW 0xAA55 - - This is not the intended use of the `ORG' directive in NASM, and - will not work. The correct way to solve this problem in NASM is to - use the `TIMES' directive, like this: - - ORG 0 - - ; some boot sector code - - TIMES 510-($-$$) DB 0 - DW 0xAA55 - - The `TIMES' directive will insert exactly enough zero bytes into the - output to move the assembly point up to 510. This method also has - the advantage that if you accidentally fill your boot sector too - full, NASM will catch the problem at assembly time and report it, so - you won't end up with a boot sector that you have to disassemble to - find out what's wrong with it. - -12.1.4 `TIMES' Doesn't Work - - The other common problem with the above code is people who write the - `TIMES' line as - - TIMES 510-$ DB 0 - - by reasoning that `$' should be a pure number, just like 510, so the - difference between them is also a pure number and can happily be fed - to `TIMES'. - - NASM is a _modular_ assembler: the various component parts are - designed to be easily separable for re-use, so they don't exchange - information unnecessarily. In consequence, the `bin' output format, - even though it has been told by the `ORG' directive that the `.text' - section should start at 0, does not pass that information back to - the expression evaluator. So from the evaluator's point of view, `$' - isn't a pure number: it's an offset from a section base. Therefore - the difference between `$' and 510 is also not a pure number, but - involves a section base. Values involving section bases cannot be - passed as arguments to `TIMES'. - - The solution, as in the previous section, is to code the `TIMES' - line in the form - - TIMES 510-($-$$) DB 0 - - in which `$' and `$$' are offsets from the same section base, and so - their difference is a pure number. This will solve the problem and - generate sensible code. - - 12.2 Bugs - - We have never yet released a version of NASM with any _known_ bugs. - That doesn't usually stop there being plenty we didn't know about, - though. Any that you find should be reported firstly via the - `bugtracker' at `https://sourceforge.net/projects/nasm/' (click on - "Bugs"), or if that fails then through one of the contacts in - section 1.2. - - Please read section 2.2 first, and don't report the bug if it's - listed in there as a deliberate feature. (If you think the feature - is badly thought out, feel free to send us reasons why you think it - should be changed, but don't just send us mail saying `This is a - bug' if the documentation says we did it on purpose.) Then read - section 12.1, and don't bother reporting the bug if it's listed - there. - - If you do report a bug, _please_ give us all of the following - information: - - (*) What operating system you're running NASM under. DOS, Linux, - NetBSD, Win16, Win32, VMS (I'd be impressed), whatever. - - (*) If you're running NASM under DOS or Win32, tell us whether - you've compiled your own executable from the DOS source archive, - or whether you were using the standard distribution binaries out - of the archive. If you were using a locally built executable, - try to reproduce the problem using one of the standard binaries, - as this will make it easier for us to reproduce your problem - prior to fixing it. - - (*) Which version of NASM you're using, and exactly how you invoked - it. Give us the precise command line, and the contents of the - `NASMENV' environment variable if any. - - (*) Which versions of any supplementary programs you're using, and - how you invoked them. If the problem only becomes visible at - link time, tell us what linker you're using, what version of it - you've got, and the exact linker command line. If the problem - involves linking against object files generated by a compiler, - tell us what compiler, what version, and what command line or - options you used. (If you're compiling in an IDE, please try to - reproduce the problem with the command-line version of the - compiler.) - - (*) If at all possible, send us a NASM source file which exhibits - the problem. If this causes copyright problems (e.g. you can - only reproduce the bug in restricted-distribution code) then - bear in mind the following two points: firstly, we guarantee - that any source code sent to us for the purposes of debugging - NASM will be used _only_ for the purposes of debugging NASM, and - that we will delete all our copies of it as soon as we have - found and fixed the bug or bugs in question; and secondly, we - would prefer _not_ to be mailed large chunks of code anyway. The - smaller the file, the better. A three-line sample file that does - nothing useful _except_ demonstrate the problem is much easier - to work with than a fully fledged ten-thousand-line program. (Of - course, some errors _do_ only crop up in large files, so this - may not be possible.) - - (*) A description of what the problem actually _is_. `It doesn't - work' is _not_ a helpful description! Please describe exactly - what is happening that shouldn't be, or what isn't happening - that should. Examples might be: `NASM generates an error message - saying Line 3 for an error that's actually on Line 5'; `NASM - generates an error message that I believe it shouldn't be - generating at all'; `NASM fails to generate an error message - that I believe it _should_ be generating'; `the object file - produced from this source code crashes my linker'; `the ninth - byte of the output file is 66 and I think it should be 77 - instead'. - - (*) If you believe the output file from NASM to be faulty, send it - to us. That allows us to determine whether our own copy of NASM - generates the same file, or whether the problem is related to - portability issues between our development platforms and yours. - We can handle binary files mailed to us as MIME attachments, - uuencoded, and even BinHex. Alternatively, we may be able to - provide an FTP site you can upload the suspect files to; but - mailing them is easier for us. - - (*) Any other information or data files that might be helpful. If, - for example, the problem involves NASM failing to generate an - object file while TASM can generate an equivalent file without - trouble, then send us _both_ object files, so we can see what - TASM is doing differently from us. - -Appendix A: Ndisasm -------------------- - - The Netwide Disassembler, NDISASM - - A.1 Introduction - - The Netwide Disassembler is a small companion program to the Netwide - Assembler, NASM. It seemed a shame to have an x86 assembler, - complete with a full instruction table, and not make as much use of - it as possible, so here's a disassembler which shares the - instruction table (and some other bits of code) with NASM. - - The Netwide Disassembler does nothing except to produce - disassemblies of _binary_ source files. NDISASM does not have any - understanding of object file formats, like `objdump', and it will - not understand `DOS .EXE' files like `debug' will. It just - disassembles. - - A.2 Getting Started: Installation - - See section 1.3 for installation instructions. NDISASM, like NASM, - has a `man page' which you may want to put somewhere useful, if you - are on a Unix system. - - A.3 Running NDISASM - - To disassemble a file, you will typically use a command of the form - - ndisasm -b {16|32|64} filename - - NDISASM can disassemble 16-, 32- or 64-bit code equally easily, - provided of course that you remember to specify which it is to work - with. If no `-b' switch is present, NDISASM works in 16-bit mode by - default. The `-u' switch (for USE32) also invokes 32-bit mode. - - Two more command line options are `-r' which reports the version - number of NDISASM you are running, and `-h' which gives a short - summary of command line options. - - A.3.1 COM Files: Specifying an Origin - - To disassemble a `DOS .COM' file correctly, a disassembler must - assume that the first instruction in the file is loaded at address - `0x100', rather than at zero. NDISASM, which assumes by default that - any file you give it is loaded at zero, will therefore need to be - informed of this. - - The `-o' option allows you to declare a different origin for the - file you are disassembling. Its argument may be expressed in any of - the NASM numeric formats: decimal by default, if it begins with - ``$'' or ``0x'' or ends in ``H'' it's `hex', if it ends in ``Q'' - it's `octal', and if it ends in ``B'' it's `binary'. - - Hence, to disassemble a `.COM' file: - - ndisasm -o100h filename.com - - will do the trick. - - A.3.2 Code Following Data: Synchronisation - - Suppose you are disassembling a file which contains some data which - isn't machine code, and _then_ contains some machine code. NDISASM - will faithfully plough through the data section, producing machine - instructions wherever it can (although most of them will look - bizarre, and some may have unusual prefixes, e.g. - ``FS OR AX,0x240A''), and generating `DB' instructions ever so often - if it's totally stumped. Then it will reach the code section. - - Supposing NDISASM has just finished generating a strange machine - instruction from part of the data section, and its file position is - now one byte _before_ the beginning of the code section. It's - entirely possible that another spurious instruction will get - generated, starting with the final byte of the data section, and - then the correct first instruction in the code section will not be - seen because the starting point skipped over it. This isn't really - ideal. - - To avoid this, you can specify a ``synchronisation'' point, or - indeed as many synchronisation points as you like (although NDISASM - can only handle 2147483647 sync points internally). The definition - of a sync point is this: NDISASM guarantees to hit sync points - exactly during disassembly. If it is thinking about generating an - instruction which would cause it to jump over a sync point, it will - discard that instruction and output a ``db'' instead. So it _will_ - start disassembly exactly from the sync point, and so you _will_ see - all the instructions in your code section. - - Sync points are specified using the `-s' option: they are measured - in terms of the program origin, not the file position. So if you - want to synchronize after 32 bytes of a `.COM' file, you would have - to do - - ndisasm -o100h -s120h file.com - - rather than - - ndisasm -o100h -s20h file.com - - As stated above, you can specify multiple sync markers if you need - to, just by repeating the `-s' option. - - A.3.3 Mixed Code and Data: Automatic (Intelligent) Synchronisation - - Suppose you are disassembling the boot sector of a `DOS' floppy - (maybe it has a virus, and you need to understand the virus so that - you know what kinds of damage it might have done you). Typically, - this will contain a `JMP' instruction, then some data, then the rest - of the code. So there is a very good chance of NDISASM being - _misaligned_ when the data ends and the code begins. Hence a sync - point is needed. - - On the other hand, why should you have to specify the sync point - manually? What you'd do in order to find where the sync point would - be, surely, would be to read the `JMP' instruction, and then to use - its target address as a sync point. So can NDISASM do that for you? - - The answer, of course, is yes: using either of the synonymous - switches `-a' (for automatic sync) or `-i' (for intelligent sync) - will enable `auto-sync' mode. Auto-sync mode automatically generates - a sync point for any forward-referring PC-relative jump or call - instruction that NDISASM encounters. (Since NDISASM is one-pass, if - it encounters a PC-relative jump whose target has already been - processed, there isn't much it can do about it...) - - Only PC-relative jumps are processed, since an absolute jump is - either through a register (in which case NDISASM doesn't know what - the register contains) or involves a segment address (in which case - the target code isn't in the same segment that NDISASM is working - in, and so the sync point can't be placed anywhere useful). - - For some kinds of file, this mechanism will automatically put sync - points in all the right places, and save you from having to place - any sync points manually. However, it should be stressed that auto- - sync mode is _not_ guaranteed to catch all the sync points, and you - may still have to place some manually. - - Auto-sync mode doesn't prevent you from declaring manual sync - points: it just adds automatically generated ones to the ones you - provide. It's perfectly feasible to specify `-i' _and_ some `-s' - options. - - Another caveat with auto-sync mode is that if, by some unpleasant - fluke, something in your data section should disassemble to a PC- - relative call or jump instruction, NDISASM may obediently place a - sync point in a totally random place, for example in the middle of - one of the instructions in your code section. So you may end up with - a wrong disassembly even if you use auto-sync. Again, there isn't - much I can do about this. If you have problems, you'll have to use - manual sync points, or use the `-k' option (documented below) to - suppress disassembly of the data area. - - A.3.4 Other Options - - The `-e' option skips a header on the file, by ignoring the first N - bytes. This means that the header is _not_ counted towards the - disassembly offset: if you give `-e10 -o10', disassembly will start - at byte 10 in the file, and this will be given offset 10, not 20. - - The `-k' option is provided with two comma-separated numeric - arguments, the first of which is an assembly offset and the second - is a number of bytes to skip. This _will_ count the skipped bytes - towards the assembly offset: its use is to suppress disassembly of a - data section which wouldn't contain anything you wanted to see - anyway. - - A.4 Bugs and Improvements - - There are no known bugs. However, any you find, with patches if - possible, should be sent to `nasm-bugs@lists.sourceforge.net', or to - the developer's site at `https://sourceforge.net/projects/nasm/' and - we'll try to fix them. Feel free to send contributions and new - features as well. - -Appendix B: Instruction List ----------------------------- - - B.1 Introduction - - The following sections show the instructions which NASM currently - supports. For each instruction, there is a separate entry for each - supported addressing mode. The third column shows the processor type - in which the instruction was introduced and, when appropriate, one - or more usage flags. - - B.1.1 Special instructions... - - DB - DW - DD - DQ - DT - DO - DY - RESB imm 8086 - RESW - RESD - RESQ - REST - RESO - RESY - - B.1.2 Conventional instructions - - AAA 8086,NOLONG - AAD 8086,NOLONG - AAD imm 8086,NOLONG - AAM 8086,NOLONG - AAM imm 8086,NOLONG - AAS 8086,NOLONG - ADC mem,reg8 8086 - ADC reg8,reg8 8086 - ADC mem,reg16 8086 - ADC reg16,reg16 8086 - ADC mem,reg32 386 - ADC reg32,reg32 386 - ADC mem,reg64 X64 - ADC reg64,reg64 X64 - ADC reg8,mem 8086 - ADC reg8,reg8 8086 - ADC reg16,mem 8086 - ADC reg16,reg16 8086 - ADC reg32,mem 386 - ADC reg32,reg32 386 - ADC reg64,mem X64 - ADC reg64,reg64 X64 - ADC rm16,imm8 8086 - ADC rm32,imm8 386 - ADC rm64,imm8 X64 - ADC reg_al,imm 8086 - ADC reg_ax,sbyte16 8086 - ADC reg_ax,imm 8086 - ADC reg_eax,sbyte32 386 - ADC reg_eax,imm 386 - ADC reg_rax,sbyte64 X64 - ADC reg_rax,imm X64 - ADC rm8,imm 8086 - ADC rm16,imm 8086 - ADC rm32,imm 386 - ADC rm64,imm X64 - ADC mem,imm8 8086 - ADC mem,imm16 8086 - ADC mem,imm32 386 - ADD mem,reg8 8086 - ADD reg8,reg8 8086 - ADD mem,reg16 8086 - ADD reg16,reg16 8086 - ADD mem,reg32 386 - ADD reg32,reg32 386 - ADD mem,reg64 X64 - ADD reg64,reg64 X64 - ADD reg8,mem 8086 - ADD reg8,reg8 8086 - ADD reg16,mem 8086 - ADD reg16,reg16 8086 - ADD reg32,mem 386 - ADD reg32,reg32 386 - ADD reg64,mem X64 - ADD reg64,reg64 X64 - ADD rm16,imm8 8086 - ADD rm32,imm8 386 - ADD rm64,imm8 X64 - ADD reg_al,imm 8086 - ADD reg_ax,sbyte16 8086 - ADD reg_ax,imm 8086 - ADD reg_eax,sbyte32 386 - ADD reg_eax,imm 386 - ADD reg_rax,sbyte64 X64 - ADD reg_rax,imm X64 - ADD rm8,imm 8086 - ADD rm16,imm 8086 - ADD rm32,imm 386 - ADD rm64,imm X64 - ADD mem,imm8 8086 - ADD mem,imm16 8086 - ADD mem,imm32 386 - AND mem,reg8 8086 - AND reg8,reg8 8086 - AND mem,reg16 8086 - AND reg16,reg16 8086 - AND mem,reg32 386 - AND reg32,reg32 386 - AND mem,reg64 X64 - AND reg64,reg64 X64 - AND reg8,mem 8086 - AND reg8,reg8 8086 - AND reg16,mem 8086 - AND reg16,reg16 8086 - AND reg32,mem 386 - AND reg32,reg32 386 - AND reg64,mem X64 - AND reg64,reg64 X64 - AND rm16,imm8 8086 - AND rm32,imm8 386 - AND rm64,imm8 X64 - AND reg_al,imm 8086 - AND reg_ax,sbyte16 8086 - AND reg_ax,imm 8086 - AND reg_eax,sbyte32 386 - AND reg_eax,imm 386 - AND reg_rax,sbyte64 X64 - AND reg_rax,imm X64 - AND rm8,imm 8086 - AND rm16,imm 8086 - AND rm32,imm 386 - AND rm64,imm X64 - AND mem,imm8 8086 - AND mem,imm16 8086 - AND mem,imm32 386 - ARPL mem,reg16 286,PROT,NOLONG - ARPL reg16,reg16 286,PROT,NOLONG - BB0_RESET PENT,CYRIX,ND - BB1_RESET PENT,CYRIX,ND - BOUND reg16,mem 186,NOLONG - BOUND reg32,mem 386,NOLONG - BSF reg16,mem 386 - BSF reg16,reg16 386 - BSF reg32,mem 386 - BSF reg32,reg32 386 - BSF reg64,mem X64 - BSF reg64,reg64 X64 - BSR reg16,mem 386 - BSR reg16,reg16 386 - BSR reg32,mem 386 - BSR reg32,reg32 386 - BSR reg64,mem X64 - BSR reg64,reg64 X64 - BSWAP reg32 486 - BSWAP reg64 X64 - BT mem,reg16 386 - BT reg16,reg16 386 - BT mem,reg32 386 - BT reg32,reg32 386 - BT mem,reg64 X64 - BT reg64,reg64 X64 - BT rm16,imm 386 - BT rm32,imm 386 - BT rm64,imm X64 - BTC mem,reg16 386 - BTC reg16,reg16 386 - BTC mem,reg32 386 - BTC reg32,reg32 386 - BTC mem,reg64 X64 - BTC reg64,reg64 X64 - BTC rm16,imm 386 - BTC rm32,imm 386 - BTC rm64,imm X64 - BTR mem,reg16 386 - BTR reg16,reg16 386 - BTR mem,reg32 386 - BTR reg32,reg32 386 - BTR mem,reg64 X64 - BTR reg64,reg64 X64 - BTR rm16,imm 386 - BTR rm32,imm 386 - BTR rm64,imm X64 - BTS mem,reg16 386 - BTS reg16,reg16 386 - BTS mem,reg32 386 - BTS reg32,reg32 386 - BTS mem,reg64 X64 - BTS reg64,reg64 X64 - BTS rm16,imm 386 - BTS rm32,imm 386 - BTS rm64,imm X64 - CALL imm 8086 - CALL imm|near 8086 - CALL imm|far 8086,ND,NOLONG - CALL imm16 8086 - CALL imm16|near 8086 - CALL imm16|far 8086,ND,NOLONG - CALL imm32 386 - CALL imm32|near 386 - CALL imm32|far 386,ND,NOLONG - CALL imm:imm 8086,NOLONG - CALL imm16:imm 8086,NOLONG - CALL imm:imm16 8086,NOLONG - CALL imm32:imm 386,NOLONG - CALL imm:imm32 386,NOLONG - CALL mem|far 8086,NOLONG - CALL mem|far X64 - CALL mem16|far 8086 - CALL mem32|far 386 - CALL mem64|far X64 - CALL mem|near 8086 - CALL mem16|near 8086 - CALL mem32|near 386,NOLONG - CALL mem64|near X64 - CALL reg16 8086 - CALL reg32 386,NOLONG - CALL reg64 X64 - CALL mem 8086 - CALL mem16 8086 - CALL mem32 386,NOLONG - CALL mem64 X64 - CBW 8086 - CDQ 386 - CDQE X64 - CLC 8086 - CLD 8086 - CLGI X64,AMD - CLI 8086 - CLTS 286,PRIV - CMC 8086 - CMP mem,reg8 8086 - CMP reg8,reg8 8086 - CMP mem,reg16 8086 - CMP reg16,reg16 8086 - CMP mem,reg32 386 - CMP reg32,reg32 386 - CMP mem,reg64 X64 - CMP reg64,reg64 X64 - CMP reg8,mem 8086 - CMP reg8,reg8 8086 - CMP reg16,mem 8086 - CMP reg16,reg16 8086 - CMP reg32,mem 386 - CMP reg32,reg32 386 - CMP reg64,mem X64 - CMP reg64,reg64 X64 - CMP rm16,imm8 8086 - CMP rm32,imm8 386 - CMP rm64,imm8 X64 - CMP reg_al,imm 8086 - CMP reg_ax,sbyte16 8086 - CMP reg_ax,imm 8086 - CMP reg_eax,sbyte32 386 - CMP reg_eax,imm 386 - CMP reg_rax,sbyte64 X64 - CMP reg_rax,imm X64 - CMP rm8,imm 8086 - CMP rm16,imm 8086 - CMP rm32,imm 386 - CMP rm64,imm X64 - CMP mem,imm8 8086 - CMP mem,imm16 8086 - CMP mem,imm32 386 - CMPSB 8086 - CMPSD 386 - CMPSQ X64 - CMPSW 8086 - CMPXCHG mem,reg8 PENT - CMPXCHG reg8,reg8 PENT - CMPXCHG mem,reg16 PENT - CMPXCHG reg16,reg16 PENT - CMPXCHG mem,reg32 PENT - CMPXCHG reg32,reg32 PENT - CMPXCHG mem,reg64 X64 - CMPXCHG reg64,reg64 X64 - CMPXCHG486 mem,reg8 486,UNDOC,ND - CMPXCHG486 reg8,reg8 486,UNDOC,ND - CMPXCHG486 mem,reg16 486,UNDOC,ND - CMPXCHG486 reg16,reg16 486,UNDOC,ND - CMPXCHG486 mem,reg32 486,UNDOC,ND - CMPXCHG486 reg32,reg32 486,UNDOC,ND - CMPXCHG8B mem PENT - CMPXCHG16B mem X64 - CPUID PENT - CPU_READ PENT,CYRIX - CPU_WRITE PENT,CYRIX - CQO X64 - CWD 8086 - CWDE 386 - DAA 8086,NOLONG - DAS 8086,NOLONG - DEC reg16 8086,NOLONG - DEC reg32 386,NOLONG - DEC rm8 8086 - DEC rm16 8086 - DEC rm32 386 - DEC rm64 X64 - DIV rm8 8086 - DIV rm16 8086 - DIV rm32 386 - DIV rm64 X64 - DMINT P6,CYRIX - EMMS PENT,MMX - ENTER imm,imm 186 - EQU imm 8086 - EQU imm:imm 8086 - F2XM1 8086,FPU - FABS 8086,FPU - FADD mem32 8086,FPU - FADD mem64 8086,FPU - FADD fpureg|to 8086,FPU - FADD fpureg 8086,FPU - FADD fpureg,fpu0 8086,FPU - FADD fpu0,fpureg 8086,FPU - FADD 8086,FPU,ND - FADDP fpureg 8086,FPU - FADDP fpureg,fpu0 8086,FPU - FADDP 8086,FPU,ND - FBLD mem80 8086,FPU - FBLD mem 8086,FPU - FBSTP mem80 8086,FPU - FBSTP mem 8086,FPU - FCHS 8086,FPU - FCLEX 8086,FPU - FCMOVB fpureg P6,FPU - FCMOVB fpu0,fpureg P6,FPU - FCMOVB P6,FPU,ND - FCMOVBE fpureg P6,FPU - FCMOVBE fpu0,fpureg P6,FPU - FCMOVBE P6,FPU,ND - FCMOVE fpureg P6,FPU - FCMOVE fpu0,fpureg P6,FPU - FCMOVE P6,FPU,ND - FCMOVNB fpureg P6,FPU - FCMOVNB fpu0,fpureg P6,FPU - FCMOVNB P6,FPU,ND - FCMOVNBE fpureg P6,FPU - FCMOVNBE fpu0,fpureg P6,FPU - FCMOVNBE P6,FPU,ND - FCMOVNE fpureg P6,FPU - FCMOVNE fpu0,fpureg P6,FPU - FCMOVNE P6,FPU,ND - FCMOVNU fpureg P6,FPU - FCMOVNU fpu0,fpureg P6,FPU - FCMOVNU P6,FPU,ND - FCMOVU fpureg P6,FPU - FCMOVU fpu0,fpureg P6,FPU - FCMOVU P6,FPU,ND - FCOM mem32 8086,FPU - FCOM mem64 8086,FPU - FCOM fpureg 8086,FPU - FCOM fpu0,fpureg 8086,FPU - FCOM 8086,FPU,ND - FCOMI fpureg P6,FPU - FCOMI fpu0,fpureg P6,FPU - FCOMI P6,FPU,ND - FCOMIP fpureg P6,FPU - FCOMIP fpu0,fpureg P6,FPU - FCOMIP P6,FPU,ND - FCOMP mem32 8086,FPU - FCOMP mem64 8086,FPU - FCOMP fpureg 8086,FPU - FCOMP fpu0,fpureg 8086,FPU - FCOMP 8086,FPU,ND - FCOMPP 8086,FPU - FCOS 386,FPU - FDECSTP 8086,FPU - FDISI 8086,FPU - FDIV mem32 8086,FPU - FDIV mem64 8086,FPU - FDIV fpureg|to 8086,FPU - FDIV fpureg 8086,FPU - FDIV fpureg,fpu0 8086,FPU - FDIV fpu0,fpureg 8086,FPU - FDIV 8086,FPU,ND - FDIVP fpureg 8086,FPU - FDIVP fpureg,fpu0 8086,FPU - FDIVP 8086,FPU,ND - FDIVR mem32 8086,FPU - FDIVR mem64 8086,FPU - FDIVR fpureg|to 8086,FPU - FDIVR fpureg,fpu0 8086,FPU - FDIVR fpureg 8086,FPU - FDIVR fpu0,fpureg 8086,FPU - FDIVR 8086,FPU,ND - FDIVRP fpureg 8086,FPU - FDIVRP fpureg,fpu0 8086,FPU - FDIVRP 8086,FPU,ND - FEMMS PENT,3DNOW - FENI 8086,FPU - FFREE fpureg 8086,FPU - FFREE 8086,FPU - FFREEP fpureg 286,FPU,UNDOC - FFREEP 286,FPU,UNDOC - FIADD mem32 8086,FPU - FIADD mem16 8086,FPU - FICOM mem32 8086,FPU - FICOM mem16 8086,FPU - FICOMP mem32 8086,FPU - FICOMP mem16 8086,FPU - FIDIV mem32 8086,FPU - FIDIV mem16 8086,FPU - FIDIVR mem32 8086,FPU - FIDIVR mem16 8086,FPU - FILD mem32 8086,FPU - FILD mem16 8086,FPU - FILD mem64 8086,FPU - FIMUL mem32 8086,FPU - FIMUL mem16 8086,FPU - FINCSTP 8086,FPU - FINIT 8086,FPU - FIST mem32 8086,FPU - FIST mem16 8086,FPU - FISTP mem32 8086,FPU - FISTP mem16 8086,FPU - FISTP mem64 8086,FPU - FISTTP mem16 PRESCOTT,FPU - FISTTP mem32 PRESCOTT,FPU - FISTTP mem64 PRESCOTT,FPU - FISUB mem32 8086,FPU - FISUB mem16 8086,FPU - FISUBR mem32 8086,FPU - FISUBR mem16 8086,FPU - FLD mem32 8086,FPU - FLD mem64 8086,FPU - FLD mem80 8086,FPU - FLD fpureg 8086,FPU - FLD 8086,FPU,ND - FLD1 8086,FPU - FLDCW mem 8086,FPU,SW - FLDENV mem 8086,FPU - FLDL2E 8086,FPU - FLDL2T 8086,FPU - FLDLG2 8086,FPU - FLDLN2 8086,FPU - FLDPI 8086,FPU - FLDZ 8086,FPU - FMUL mem32 8086,FPU - FMUL mem64 8086,FPU - FMUL fpureg|to 8086,FPU - FMUL fpureg,fpu0 8086,FPU - FMUL fpureg 8086,FPU - FMUL fpu0,fpureg 8086,FPU - FMUL 8086,FPU,ND - FMULP fpureg 8086,FPU - FMULP fpureg,fpu0 8086,FPU - FMULP 8086,FPU,ND - FNCLEX 8086,FPU - FNDISI 8086,FPU - FNENI 8086,FPU - FNINIT 8086,FPU - FNOP 8086,FPU - FNSAVE mem 8086,FPU - FNSTCW mem 8086,FPU,SW - FNSTENV mem 8086,FPU - FNSTSW mem 8086,FPU,SW - FNSTSW reg_ax 286,FPU - FPATAN 8086,FPU - FPREM 8086,FPU - FPREM1 386,FPU - FPTAN 8086,FPU - FRNDINT 8086,FPU - FRSTOR mem 8086,FPU - FSAVE mem 8086,FPU - FSCALE 8086,FPU - FSETPM 286,FPU - FSIN 386,FPU - FSINCOS 386,FPU - FSQRT 8086,FPU - FST mem32 8086,FPU - FST mem64 8086,FPU - FST fpureg 8086,FPU - FST 8086,FPU,ND - FSTCW mem 8086,FPU,SW - FSTENV mem 8086,FPU - FSTP mem32 8086,FPU - FSTP mem64 8086,FPU - FSTP mem80 8086,FPU - FSTP fpureg 8086,FPU - FSTP 8086,FPU,ND - FSTSW mem 8086,FPU,SW - FSTSW reg_ax 286,FPU - FSUB mem32 8086,FPU - FSUB mem64 8086,FPU - FSUB fpureg|to 8086,FPU - FSUB fpureg,fpu0 8086,FPU - FSUB fpureg 8086,FPU - FSUB fpu0,fpureg 8086,FPU - FSUB 8086,FPU,ND - FSUBP fpureg 8086,FPU - FSUBP fpureg,fpu0 8086,FPU - FSUBP 8086,FPU,ND - FSUBR mem32 8086,FPU - FSUBR mem64 8086,FPU - FSUBR fpureg|to 8086,FPU - FSUBR fpureg,fpu0 8086,FPU - FSUBR fpureg 8086,FPU - FSUBR fpu0,fpureg 8086,FPU - FSUBR 8086,FPU,ND - FSUBRP fpureg 8086,FPU - FSUBRP fpureg,fpu0 8086,FPU - FSUBRP 8086,FPU,ND - FTST 8086,FPU - FUCOM fpureg 386,FPU - FUCOM fpu0,fpureg 386,FPU - FUCOM 386,FPU,ND - FUCOMI fpureg P6,FPU - FUCOMI fpu0,fpureg P6,FPU - FUCOMI P6,FPU,ND - FUCOMIP fpureg P6,FPU - FUCOMIP fpu0,fpureg P6,FPU - FUCOMIP P6,FPU,ND - FUCOMP fpureg 386,FPU - FUCOMP fpu0,fpureg 386,FPU - FUCOMP 386,FPU,ND - FUCOMPP 386,FPU - FXAM 8086,FPU - FXCH fpureg 8086,FPU - FXCH fpureg,fpu0 8086,FPU - FXCH fpu0,fpureg 8086,FPU - FXCH 8086,FPU,ND - FXTRACT 8086,FPU - FYL2X 8086,FPU - FYL2XP1 8086,FPU - HLT 8086,PRIV - IBTS mem,reg16 386,SW,UNDOC,ND - IBTS reg16,reg16 386,UNDOC,ND - IBTS mem,reg32 386,SD,UNDOC,ND - IBTS reg32,reg32 386,UNDOC,ND - ICEBP 386,ND - IDIV rm8 8086 - IDIV rm16 8086 - IDIV rm32 386 - IDIV rm64 X64 - IMUL rm8 8086 - IMUL rm16 8086 - IMUL rm32 386 - IMUL rm64 X64 - IMUL reg16,mem 386 - IMUL reg16,reg16 386 - IMUL reg32,mem 386 - IMUL reg32,reg32 386 - IMUL reg64,mem X64 - IMUL reg64,reg64 X64 - IMUL reg16,mem,imm8 186 - IMUL reg16,mem,sbyte16 186,ND - IMUL reg16,mem,imm16 186 - IMUL reg16,mem,imm 186,ND - IMUL reg16,reg16,imm8 186 - IMUL reg16,reg16,sbyte16 186,ND - IMUL reg16,reg16,imm16 186 - IMUL reg16,reg16,imm 186,ND - IMUL reg32,mem,imm8 386 - IMUL reg32,mem,sbyte32 386,ND - IMUL reg32,mem,imm32 386 - IMUL reg32,mem,imm 386,ND - IMUL reg32,reg32,imm8 386 - IMUL reg32,reg32,sbyte32 386,ND - IMUL reg32,reg32,imm32 386 - IMUL reg32,reg32,imm 386,ND - IMUL reg64,mem,imm8 X64 - IMUL reg64,mem,sbyte64 X64,ND - IMUL reg64,mem,imm32 X64 - IMUL reg64,mem,imm X64,ND - IMUL reg64,reg64,imm8 X64 - IMUL reg64,reg64,sbyte64 X64,ND - IMUL reg64,reg64,imm32 X64 - IMUL reg64,reg64,imm X64,ND - IMUL reg16,imm8 186 - IMUL reg16,sbyte16 186,ND - IMUL reg16,imm16 186 - IMUL reg16,imm 186,ND - IMUL reg32,imm8 386 - IMUL reg32,sbyte32 386,ND - IMUL reg32,imm32 386 - IMUL reg32,imm 386,ND - IMUL reg64,imm8 X64 - IMUL reg64,sbyte64 X64,ND - IMUL reg64,imm32 X64 - IMUL reg64,imm X64,ND - IN reg_al,imm 8086 - IN reg_ax,imm 8086 - IN reg_eax,imm 386 - IN reg_al,reg_dx 8086 - IN reg_ax,reg_dx 8086 - IN reg_eax,reg_dx 386 - INC reg16 8086,NOLONG - INC reg32 386,NOLONG - INC rm8 8086 - INC rm16 8086 - INC rm32 386 - INC rm64 X64 - INCBIN - INSB 186 - INSD 386 - INSW 186 - INT imm 8086 - INT01 386,ND - INT1 386 - INT03 8086,ND - INT3 8086 - INTO 8086,NOLONG - INVD 486,PRIV - INVLPG mem 486,PRIV - INVLPGA reg_ax,reg_ecx X86_64,AMD,NOLONG - INVLPGA reg_eax,reg_ecx X86_64,AMD - INVLPGA reg_rax,reg_ecx X64,AMD - INVLPGA X86_64,AMD - IRET 8086 - IRETD 386 - IRETQ X64 - IRETW 8086 - JCXZ imm 8086,NOLONG - JECXZ imm 386 - JRCXZ imm X64 - JMP imm|short 8086 - JMP imm 8086,ND - JMP imm 8086 - JMP imm|near 8086,ND - JMP imm|far 8086,ND,NOLONG - JMP imm16 8086 - JMP imm16|near 8086,ND - JMP imm16|far 8086,ND,NOLONG - JMP imm32 386 - JMP imm32|near 386,ND - JMP imm32|far 386,ND,NOLONG - JMP imm:imm 8086,NOLONG - JMP imm16:imm 8086,NOLONG - JMP imm:imm16 8086,NOLONG - JMP imm32:imm 386,NOLONG - JMP imm:imm32 386,NOLONG - JMP mem|far 8086,NOLONG - JMP mem|far X64 - JMP mem16|far 8086 - JMP mem32|far 386 - JMP mem64|far X64 - JMP mem|near 8086 - JMP mem16|near 8086 - JMP mem32|near 386,NOLONG - JMP mem64|near X64 - JMP reg16 8086 - JMP reg32 386,NOLONG - JMP reg64 X64 - JMP mem 8086 - JMP mem16 8086 - JMP mem32 386,NOLONG - JMP mem64 X64 - JMPE imm IA64 - JMPE imm16 IA64 - JMPE imm32 IA64 - JMPE rm16 IA64 - JMPE rm32 IA64 - LAHF 8086 - LAR reg16,mem 286,PROT,SW - LAR reg16,reg16 286,PROT - LAR reg16,reg32 386,PROT - LAR reg16,reg64 X64,PROT,ND - LAR reg32,mem 386,PROT,SW - LAR reg32,reg16 386,PROT - LAR reg32,reg32 386,PROT - LAR reg32,reg64 X64,PROT,ND - LAR reg64,mem X64,PROT,SW - LAR reg64,reg16 X64,PROT - LAR reg64,reg32 X64,PROT - LAR reg64,reg64 X64,PROT - LDS reg16,mem 8086,NOLONG - LDS reg32,mem 386,NOLONG - LEA reg16,mem 8086 - LEA reg32,mem 386 - LEA reg64,mem X64 - LEAVE 186 - LES reg16,mem 8086,NOLONG - LES reg32,mem 386,NOLONG - LFENCE X64,AMD - LFS reg16,mem 386 - LFS reg32,mem 386 - LGDT mem 286,PRIV - LGS reg16,mem 386 - LGS reg32,mem 386 - LIDT mem 286,PRIV - LLDT mem 286,PROT,PRIV - LLDT mem16 286,PROT,PRIV - LLDT reg16 286,PROT,PRIV - LMSW mem 286,PRIV - LMSW mem16 286,PRIV - LMSW reg16 286,PRIV - LOADALL 386,UNDOC - LOADALL286 286,UNDOC - LODSB 8086 - LODSD 386 - LODSQ X64 - LODSW 8086 - LOOP imm 8086 - LOOP imm,reg_cx 8086,NOLONG - LOOP imm,reg_ecx 386 - LOOP imm,reg_rcx X64 - LOOPE imm 8086 - LOOPE imm,reg_cx 8086,NOLONG - LOOPE imm,reg_ecx 386 - LOOPE imm,reg_rcx X64 - LOOPNE imm 8086 - LOOPNE imm,reg_cx 8086,NOLONG - LOOPNE imm,reg_ecx 386 - LOOPNE imm,reg_rcx X64 - LOOPNZ imm 8086 - LOOPNZ imm,reg_cx 8086,NOLONG - LOOPNZ imm,reg_ecx 386 - LOOPNZ imm,reg_rcx X64 - LOOPZ imm 8086 - LOOPZ imm,reg_cx 8086,NOLONG - LOOPZ imm,reg_ecx 386 - LOOPZ imm,reg_rcx X64 - LSL reg16,mem 286,PROT,SW - LSL reg16,reg16 286,PROT - LSL reg16,reg32 386,PROT - LSL reg16,reg64 X64,PROT,ND - LSL reg32,mem 386,PROT,SW - LSL reg32,reg16 386,PROT - LSL reg32,reg32 386,PROT - LSL reg32,reg64 X64,PROT,ND - LSL reg64,mem X64,PROT,SW - LSL reg64,reg16 X64,PROT - LSL reg64,reg32 X64,PROT - LSL reg64,reg64 X64,PROT - LSS reg16,mem 386 - LSS reg32,mem 386 - LTR mem 286,PROT,PRIV - LTR mem16 286,PROT,PRIV - LTR reg16 286,PROT,PRIV - MFENCE X64,AMD - MONITOR PRESCOTT - MONITOR reg_eax,reg_ecx,reg_edx PRESCOTT,ND - MONITOR reg_rax,reg_ecx,reg_edx X64,ND - MOV mem,reg_sreg 8086 - MOV reg16,reg_sreg 8086 - MOV reg32,reg_sreg 386 - MOV reg_sreg,mem 8086 - MOV reg_sreg,reg16 8086 - MOV reg_sreg,reg32 386 - MOV reg_al,mem_offs 8086 - MOV reg_ax,mem_offs 8086 - MOV reg_eax,mem_offs 386 - MOV reg_rax,mem_offs X64 - MOV mem_offs,reg_al 8086 - MOV mem_offs,reg_ax 8086 - MOV mem_offs,reg_eax 386 - MOV mem_offs,reg_rax X64 - MOV reg32,reg_creg 386,PRIV,NOLONG - MOV reg64,reg_creg X64,PRIV - MOV reg_creg,reg32 386,PRIV,NOLONG - MOV reg_creg,reg64 X64,PRIV - MOV reg32,reg_dreg 386,PRIV,NOLONG - MOV reg64,reg_dreg X64,PRIV - MOV reg_dreg,reg32 386,PRIV,NOLONG - MOV reg_dreg,reg64 X64,PRIV - MOV reg32,reg_treg 386,NOLONG,ND - MOV reg_treg,reg32 386,NOLONG,ND - MOV mem,reg8 8086 - MOV reg8,reg8 8086 - MOV mem,reg16 8086 - MOV reg16,reg16 8086 - MOV mem,reg32 386 - MOV reg32,reg32 386 - MOV mem,reg64 X64 - MOV reg64,reg64 X64 - MOV reg8,mem 8086 - MOV reg8,reg8 8086 - MOV reg16,mem 8086 - MOV reg16,reg16 8086 - MOV reg32,mem 386 - MOV reg32,reg32 386 - MOV reg64,mem X64 - MOV reg64,reg64 X64 - MOV reg8,imm 8086 - MOV reg16,imm 8086 - MOV reg32,imm 386 - MOV reg64,imm X64 - MOV reg64,imm32 X64 - MOV rm8,imm 8086 - MOV rm16,imm 8086 - MOV rm32,imm 386 - MOV rm64,imm X64 - MOV mem,imm8 8086 - MOV mem,imm16 8086 - MOV mem,imm32 386 - MOVD mmxreg,mem PENT,MMX,SD - MOVD mmxreg,reg32 PENT,MMX - MOVD mem,mmxreg PENT,MMX,SD - MOVD reg32,mmxreg PENT,MMX - MOVD xmmreg,mem X64,SD - MOVD xmmreg,reg32 X64 - MOVD mem,xmmreg X64,SD - MOVD reg32,xmmreg X64,SSE - MOVQ mmxreg,mmxrm PENT,MMX - MOVQ mmxrm,mmxreg PENT,MMX - MOVQ mmxreg,rm64 X64,MMX - MOVQ rm64,mmxreg X64,MMX - MOVSB 8086 - MOVSD 386 - MOVSQ X64 - MOVSW 8086 - MOVSX reg16,mem 386 - MOVSX reg16,reg8 386 - MOVSX reg32,rm8 386 - MOVSX reg32,rm16 386 - MOVSX reg64,rm8 X64 - MOVSX reg64,rm16 X64 - MOVSXD reg64,rm32 X64 - MOVSX reg64,rm32 X64,ND - MOVZX reg16,mem 386 - MOVZX reg16,reg8 386 - MOVZX reg32,rm8 386 - MOVZX reg32,rm16 386 - MOVZX reg64,rm8 X64 - MOVZX reg64,rm16 X64 - MUL rm8 8086 - MUL rm16 8086 - MUL rm32 386 - MUL rm64 X64 - MWAIT PRESCOTT - MWAIT reg_eax,reg_ecx PRESCOTT,ND - NEG rm8 8086 - NEG rm16 8086 - NEG rm32 386 - NEG rm64 X64 - NOP 8086 - NOP rm16 P6 - NOP rm32 P6 - NOP rm64 X64 - NOT rm8 8086 - NOT rm16 8086 - NOT rm32 386 - NOT rm64 X64 - OR mem,reg8 8086 - OR reg8,reg8 8086 - OR mem,reg16 8086 - OR reg16,reg16 8086 - OR mem,reg32 386 - OR reg32,reg32 386 - OR mem,reg64 X64 - OR reg64,reg64 X64 - OR reg8,mem 8086 - OR reg8,reg8 8086 - OR reg16,mem 8086 - OR reg16,reg16 8086 - OR reg32,mem 386 - OR reg32,reg32 386 - OR reg64,mem X64 - OR reg64,reg64 X64 - OR rm16,imm8 8086 - OR rm32,imm8 386 - OR rm64,imm8 X64 - OR reg_al,imm 8086 - OR reg_ax,sbyte16 8086 - OR reg_ax,imm 8086 - OR reg_eax,sbyte32 386 - OR reg_eax,imm 386 - OR reg_rax,sbyte64 X64 - OR reg_rax,imm X64 - OR rm8,imm 8086 - OR rm16,imm 8086 - OR rm32,imm 386 - OR rm64,imm X64 - OR mem,imm8 8086 - OR mem,imm16 8086 - OR mem,imm32 386 - OUT imm,reg_al 8086 - OUT imm,reg_ax 8086 - OUT imm,reg_eax 386 - OUT reg_dx,reg_al 8086 - OUT reg_dx,reg_ax 8086 - OUT reg_dx,reg_eax 386 - OUTSB 186 - OUTSD 386 - OUTSW 186 - PACKSSDW mmxreg,mmxrm PENT,MMX - PACKSSWB mmxreg,mmxrm PENT,MMX - PACKUSWB mmxreg,mmxrm PENT,MMX - PADDB mmxreg,mmxrm PENT,MMX - PADDD mmxreg,mmxrm PENT,MMX - PADDSB mmxreg,mmxrm PENT,MMX - PADDSIW mmxreg,mmxrm PENT,MMX,CYRIX - PADDSW mmxreg,mmxrm PENT,MMX - PADDUSB mmxreg,mmxrm PENT,MMX - PADDUSW mmxreg,mmxrm PENT,MMX - PADDW mmxreg,mmxrm PENT,MMX - PAND mmxreg,mmxrm PENT,MMX - PANDN mmxreg,mmxrm PENT,MMX - PAUSE 8086 - PAVEB mmxreg,mmxrm PENT,MMX,CYRIX - PAVGUSB mmxreg,mmxrm PENT,3DNOW - PCMPEQB mmxreg,mmxrm PENT,MMX - PCMPEQD mmxreg,mmxrm PENT,MMX - PCMPEQW mmxreg,mmxrm PENT,MMX - PCMPGTB mmxreg,mmxrm PENT,MMX - PCMPGTD mmxreg,mmxrm PENT,MMX - PCMPGTW mmxreg,mmxrm PENT,MMX - PDISTIB mmxreg,mem PENT,MMX,CYRIX - PF2ID mmxreg,mmxrm PENT,3DNOW - PFACC mmxreg,mmxrm PENT,3DNOW - PFADD mmxreg,mmxrm PENT,3DNOW - PFCMPEQ mmxreg,mmxrm PENT,3DNOW - PFCMPGE mmxreg,mmxrm PENT,3DNOW - PFCMPGT mmxreg,mmxrm PENT,3DNOW - PFMAX mmxreg,mmxrm PENT,3DNOW - PFMIN mmxreg,mmxrm PENT,3DNOW - PFMUL mmxreg,mmxrm PENT,3DNOW - PFRCP mmxreg,mmxrm PENT,3DNOW - PFRCPIT1 mmxreg,mmxrm PENT,3DNOW - PFRCPIT2 mmxreg,mmxrm PENT,3DNOW - PFRSQIT1 mmxreg,mmxrm PENT,3DNOW - PFRSQRT mmxreg,mmxrm PENT,3DNOW - PFSUB mmxreg,mmxrm PENT,3DNOW - PFSUBR mmxreg,mmxrm PENT,3DNOW - PI2FD mmxreg,mmxrm PENT,3DNOW - PMACHRIW mmxreg,mem PENT,MMX,CYRIX - PMADDWD mmxreg,mmxrm PENT,MMX - PMAGW mmxreg,mmxrm PENT,MMX,CYRIX - PMULHRIW mmxreg,mmxrm PENT,MMX,CYRIX - PMULHRWA mmxreg,mmxrm PENT,3DNOW - PMULHRWC mmxreg,mmxrm PENT,MMX,CYRIX - PMULHW mmxreg,mmxrm PENT,MMX - PMULLW mmxreg,mmxrm PENT,MMX - PMVGEZB mmxreg,mem PENT,MMX,CYRIX - PMVLZB mmxreg,mem PENT,MMX,CYRIX - PMVNZB mmxreg,mem PENT,MMX,CYRIX - PMVZB mmxreg,mem PENT,MMX,CYRIX - POP reg16 8086 - POP reg32 386,NOLONG - POP reg64 X64 - POP rm16 8086 - POP rm32 386,NOLONG - POP rm64 X64 - POP reg_cs 8086,UNDOC,ND - POP reg_dess 8086,NOLONG - POP reg_fsgs 386 - POPA 186,NOLONG - POPAD 386,NOLONG - POPAW 186,NOLONG - POPF 8086 - POPFD 386,NOLONG - POPFQ X64 - POPFW 8086 - POR mmxreg,mmxrm PENT,MMX - PREFETCH mem PENT,3DNOW - PREFETCHW mem PENT,3DNOW - PSLLD mmxreg,mmxrm PENT,MMX - PSLLD mmxreg,imm PENT,MMX - PSLLQ mmxreg,mmxrm PENT,MMX - PSLLQ mmxreg,imm PENT,MMX - PSLLW mmxreg,mmxrm PENT,MMX - PSLLW mmxreg,imm PENT,MMX - PSRAD mmxreg,mmxrm PENT,MMX - PSRAD mmxreg,imm PENT,MMX - PSRAW mmxreg,mmxrm PENT,MMX - PSRAW mmxreg,imm PENT,MMX - PSRLD mmxreg,mmxrm PENT,MMX - PSRLD mmxreg,imm PENT,MMX - PSRLQ mmxreg,mmxrm PENT,MMX - PSRLQ mmxreg,imm PENT,MMX - PSRLW mmxreg,mmxrm PENT,MMX - PSRLW mmxreg,imm PENT,MMX - PSUBB mmxreg,mmxrm PENT,MMX - PSUBD mmxreg,mmxrm PENT,MMX - PSUBSB mmxreg,mmxrm PENT,MMX - PSUBSIW mmxreg,mmxrm PENT,MMX,CYRIX - PSUBSW mmxreg,mmxrm PENT,MMX - PSUBUSB mmxreg,mmxrm PENT,MMX - PSUBUSW mmxreg,mmxrm PENT,MMX - PSUBW mmxreg,mmxrm PENT,MMX - PUNPCKHBW mmxreg,mmxrm PENT,MMX - PUNPCKHDQ mmxreg,mmxrm PENT,MMX - PUNPCKHWD mmxreg,mmxrm PENT,MMX - PUNPCKLBW mmxreg,mmxrm PENT,MMX - PUNPCKLDQ mmxreg,mmxrm PENT,MMX - PUNPCKLWD mmxreg,mmxrm PENT,MMX - PUSH reg16 8086 - PUSH reg32 386,NOLONG - PUSH reg64 X64 - PUSH rm16 8086 - PUSH rm32 386,NOLONG - PUSH rm64 X64 - PUSH reg_cs 8086,NOLONG - PUSH reg_dess 8086,NOLONG - PUSH reg_fsgs 386 - PUSH imm8 186 - PUSH imm16 186,AR0,SZ - PUSH imm32 386,NOLONG,AR0,SZ - PUSH imm32 386,NOLONG,SD - PUSH imm64 X64,AR0,SZ - PUSHA 186,NOLONG - PUSHAD 386,NOLONG - PUSHAW 186,NOLONG - PUSHF 8086 - PUSHFD 386,NOLONG - PUSHFQ X64 - PUSHFW 8086 - PXOR mmxreg,mmxrm PENT,MMX - RCL rm8,unity 8086 - RCL rm8,reg_cl 8086 - RCL rm8,imm 186 - RCL rm16,unity 8086 - RCL rm16,reg_cl 8086 - RCL rm16,imm 186 - RCL rm32,unity 386 - RCL rm32,reg_cl 386 - RCL rm32,imm 386 - RCL rm64,unity X64 - RCL rm64,reg_cl X64 - RCL rm64,imm X64 - RCR rm8,unity 8086 - RCR rm8,reg_cl 8086 - RCR rm8,imm 186 - RCR rm16,unity 8086 - RCR rm16,reg_cl 8086 - RCR rm16,imm 186 - RCR rm32,unity 386 - RCR rm32,reg_cl 386 - RCR rm32,imm 386 - RCR rm64,unity X64 - RCR rm64,reg_cl X64 - RCR rm64,imm X64 - RDSHR rm32 P6,CYRIXM - RDMSR PENT,PRIV - RDPMC P6 - RDTSC PENT - RDTSCP X86_64 - RET 8086 - RET imm 8086,SW - RETF 8086 - RETF imm 8086,SW - RETN 8086 - RETN imm 8086,SW - ROL rm8,unity 8086 - ROL rm8,reg_cl 8086 - ROL rm8,imm 186 - ROL rm16,unity 8086 - ROL rm16,reg_cl 8086 - ROL rm16,imm 186 - ROL rm32,unity 386 - ROL rm32,reg_cl 386 - ROL rm32,imm 386 - ROL rm64,unity X64 - ROL rm64,reg_cl X64 - ROL rm64,imm X64 - ROR rm8,unity 8086 - ROR rm8,reg_cl 8086 - ROR rm8,imm 186 - ROR rm16,unity 8086 - ROR rm16,reg_cl 8086 - ROR rm16,imm 186 - ROR rm32,unity 386 - ROR rm32,reg_cl 386 - ROR rm32,imm 386 - ROR rm64,unity X64 - ROR rm64,reg_cl X64 - ROR rm64,imm X64 - RDM P6,CYRIX,ND - RSDC reg_sreg,mem80 486,CYRIXM - RSLDT mem80 486,CYRIXM - RSM PENTM - RSTS mem80 486,CYRIXM - SAHF 8086 - SAL rm8,unity 8086,ND - SAL rm8,reg_cl 8086,ND - SAL rm8,imm 186,ND - SAL rm16,unity 8086,ND - SAL rm16,reg_cl 8086,ND - SAL rm16,imm 186,ND - SAL rm32,unity 386,ND - SAL rm32,reg_cl 386,ND - SAL rm32,imm 386,ND - SAL rm64,unity X64,ND - SAL rm64,reg_cl X64,ND - SAL rm64,imm X64,ND - SALC 8086,UNDOC - SAR rm8,unity 8086 - SAR rm8,reg_cl 8086 - SAR rm8,imm 186 - SAR rm16,unity 8086 - SAR rm16,reg_cl 8086 - SAR rm16,imm 186 - SAR rm32,unity 386 - SAR rm32,reg_cl 386 - SAR rm32,imm 386 - SAR rm64,unity X64 - SAR rm64,reg_cl X64 - SAR rm64,imm X64 - SBB mem,reg8 8086 - SBB reg8,reg8 8086 - SBB mem,reg16 8086 - SBB reg16,reg16 8086 - SBB mem,reg32 386 - SBB reg32,reg32 386 - SBB mem,reg64 X64 - SBB reg64,reg64 X64 - SBB reg8,mem 8086 - SBB reg8,reg8 8086 - SBB reg16,mem 8086 - SBB reg16,reg16 8086 - SBB reg32,mem 386 - SBB reg32,reg32 386 - SBB reg64,mem X64 - SBB reg64,reg64 X64 - SBB rm16,imm8 8086 - SBB rm32,imm8 386 - SBB rm64,imm8 X64 - SBB reg_al,imm 8086 - SBB reg_ax,sbyte16 8086 - SBB reg_ax,imm 8086 - SBB reg_eax,sbyte32 386 - SBB reg_eax,imm 386 - SBB reg_rax,sbyte64 X64 - SBB reg_rax,imm X64 - SBB rm8,imm 8086 - SBB rm16,imm 8086 - SBB rm32,imm 386 - SBB rm64,imm X64 - SBB mem,imm8 8086 - SBB mem,imm16 8086 - SBB mem,imm32 386 - SCASB 8086 - SCASD 386 - SCASQ X64 - SCASW 8086 - SFENCE X64,AMD - SGDT mem 286 - SHL rm8,unity 8086 - SHL rm8,reg_cl 8086 - SHL rm8,imm 186 - SHL rm16,unity 8086 - SHL rm16,reg_cl 8086 - SHL rm16,imm 186 - SHL rm32,unity 386 - SHL rm32,reg_cl 386 - SHL rm32,imm 386 - SHL rm64,unity X64 - SHL rm64,reg_cl X64 - SHL rm64,imm X64 - SHLD mem,reg16,imm 3862 - SHLD reg16,reg16,imm 3862 - SHLD mem,reg32,imm 3862 - SHLD reg32,reg32,imm 3862 - SHLD mem,reg64,imm X642 - SHLD reg64,reg64,imm X642 - SHLD mem,reg16,reg_cl 386 - SHLD reg16,reg16,reg_cl 386 - SHLD mem,reg32,reg_cl 386 - SHLD reg32,reg32,reg_cl 386 - SHLD mem,reg64,reg_cl X64 - SHLD reg64,reg64,reg_cl X64 - SHR rm8,unity 8086 - SHR rm8,reg_cl 8086 - SHR rm8,imm 186 - SHR rm16,unity 8086 - SHR rm16,reg_cl 8086 - SHR rm16,imm 186 - SHR rm32,unity 386 - SHR rm32,reg_cl 386 - SHR rm32,imm 386 - SHR rm64,unity X64 - SHR rm64,reg_cl X64 - SHR rm64,imm X64 - SHRD mem,reg16,imm 3862 - SHRD reg16,reg16,imm 3862 - SHRD mem,reg32,imm 3862 - SHRD reg32,reg32,imm 3862 - SHRD mem,reg64,imm X642 - SHRD reg64,reg64,imm X642 - SHRD mem,reg16,reg_cl 386 - SHRD reg16,reg16,reg_cl 386 - SHRD mem,reg32,reg_cl 386 - SHRD reg32,reg32,reg_cl 386 - SHRD mem,reg64,reg_cl X64 - SHRD reg64,reg64,reg_cl X64 - SIDT mem 286 - SLDT mem 286 - SLDT mem16 286 - SLDT reg16 286 - SLDT reg32 386 - SLDT reg64 X64,ND - SLDT reg64 X64 - SKINIT X64 - SMI 386,UNDOC - SMINT P6,CYRIX,ND - SMINTOLD 486,CYRIX,ND - SMSW mem 286 - SMSW mem16 286 - SMSW reg16 286 - SMSW reg32 386 - STC 8086 - STD 8086 - STGI X64 - STI 8086 - STOSB 8086 - STOSD 386 - STOSQ X64 - STOSW 8086 - STR mem 286,PROT - STR mem16 286,PROT - STR reg16 286,PROT - STR reg32 386,PROT - STR reg64 X64 - SUB mem,reg8 8086 - SUB reg8,reg8 8086 - SUB mem,reg16 8086 - SUB reg16,reg16 8086 - SUB mem,reg32 386 - SUB reg32,reg32 386 - SUB mem,reg64 X64 - SUB reg64,reg64 X64 - SUB reg8,mem 8086 - SUB reg8,reg8 8086 - SUB reg16,mem 8086 - SUB reg16,reg16 8086 - SUB reg32,mem 386 - SUB reg32,reg32 386 - SUB reg64,mem X64 - SUB reg64,reg64 X64 - SUB rm16,imm8 8086 - SUB rm32,imm8 386 - SUB rm64,imm8 X64 - SUB reg_al,imm 8086 - SUB reg_ax,sbyte16 8086 - SUB reg_ax,imm 8086 - SUB reg_eax,sbyte32 386 - SUB reg_eax,imm 386 - SUB reg_rax,sbyte64 X64 - SUB reg_rax,imm X64 - SUB rm8,imm 8086 - SUB rm16,imm 8086 - SUB rm32,imm 386 - SUB rm64,imm X64 - SUB mem,imm8 8086 - SUB mem,imm16 8086 - SUB mem,imm32 386 - SVDC mem80,reg_sreg 486,CYRIXM - SVLDT mem80 486,CYRIXM,ND - SVTS mem80 486,CYRIXM - SWAPGS X64 - SYSCALL P6,AMD - SYSENTER P6 - SYSEXIT P6,PRIV - SYSRET P6,PRIV,AMD - TEST mem,reg8 8086 - TEST reg8,reg8 8086 - TEST mem,reg16 8086 - TEST reg16,reg16 8086 - TEST mem,reg32 386 - TEST reg32,reg32 386 - TEST mem,reg64 X64 - TEST reg64,reg64 X64 - TEST reg8,mem 8086 - TEST reg16,mem 8086 - TEST reg32,mem 386 - TEST reg64,mem X64 - TEST reg_al,imm 8086 - TEST reg_ax,imm 8086 - TEST reg_eax,imm 386 - TEST reg_rax,imm X64 - TEST rm8,imm 8086 - TEST rm16,imm 8086 - TEST rm32,imm 386 - TEST rm64,imm X64 - TEST mem,imm8 8086 - TEST mem,imm16 8086 - TEST mem,imm32 386 - UD0 186,UNDOC - UD1 186,UNDOC - UD2B 186,UNDOC,ND - UD2 186 - UD2A 186,ND - UMOV mem,reg8 386,UNDOC,ND - UMOV reg8,reg8 386,UNDOC,ND - UMOV mem,reg16 386,UNDOC,ND - UMOV reg16,reg16 386,UNDOC,ND - UMOV mem,reg32 386,UNDOC,ND - UMOV reg32,reg32 386,UNDOC,ND - UMOV reg8,mem 386,UNDOC,ND - UMOV reg8,reg8 386,UNDOC,ND - UMOV reg16,mem 386,UNDOC,ND - UMOV reg16,reg16 386,UNDOC,ND - UMOV reg32,mem 386,UNDOC,ND - UMOV reg32,reg32 386,UNDOC,ND - VERR mem 286,PROT - VERR mem16 286,PROT - VERR reg16 286,PROT - VERW mem 286,PROT - VERW mem16 286,PROT - VERW reg16 286,PROT - FWAIT 8086 - WBINVD 486,PRIV - WRSHR rm32 P6,CYRIXM - WRMSR PENT,PRIV - XADD mem,reg8 486 - XADD reg8,reg8 486 - XADD mem,reg16 486 - XADD reg16,reg16 486 - XADD mem,reg32 486 - XADD reg32,reg32 486 - XADD mem,reg64 X64 - XADD reg64,reg64 X64 - XBTS reg16,mem 386,SW,UNDOC,ND - XBTS reg16,reg16 386,UNDOC,ND - XBTS reg32,mem 386,SD,UNDOC,ND - XBTS reg32,reg32 386,UNDOC,ND - XCHG reg_ax,reg16 8086 - XCHG reg_eax,reg32na 386 - XCHG reg_rax,reg64 X64 - XCHG reg16,reg_ax 8086 - XCHG reg32na,reg_eax 386 - XCHG reg64,reg_rax X64 - XCHG reg_eax,reg_eax 386,NOLONG - XCHG reg8,mem 8086 - XCHG reg8,reg8 8086 - XCHG reg16,mem 8086 - XCHG reg16,reg16 8086 - XCHG reg32,mem 386 - XCHG reg32,reg32 386 - XCHG reg64,mem X64 - XCHG reg64,reg64 X64 - XCHG mem,reg8 8086 - XCHG reg8,reg8 8086 - XCHG mem,reg16 8086 - XCHG reg16,reg16 8086 - XCHG mem,reg32 386 - XCHG reg32,reg32 386 - XCHG mem,reg64 X64 - XCHG reg64,reg64 X64 - XLATB 8086 - XLAT 8086 - XOR mem,reg8 8086 - XOR reg8,reg8 8086 - XOR mem,reg16 8086 - XOR reg16,reg16 8086 - XOR mem,reg32 386 - XOR reg32,reg32 386 - XOR mem,reg64 X64 - XOR reg64,reg64 X64 - XOR reg8,mem 8086 - XOR reg8,reg8 8086 - XOR reg16,mem 8086 - XOR reg16,reg16 8086 - XOR reg32,mem 386 - XOR reg32,reg32 386 - XOR reg64,mem X64 - XOR reg64,reg64 X64 - XOR rm16,imm8 8086 - XOR rm32,imm8 386 - XOR rm64,imm8 X64 - XOR reg_al,imm 8086 - XOR reg_ax,sbyte16 8086 - XOR reg_ax,imm 8086 - XOR reg_eax,sbyte32 386 - XOR reg_eax,imm 386 - XOR reg_rax,sbyte64 X64 - XOR reg_rax,imm X64 - XOR rm8,imm 8086 - XOR rm16,imm 8086 - XOR rm32,imm 386 - XOR rm64,imm X64 - XOR mem,imm8 8086 - XOR mem,imm16 8086 - XOR mem,imm32 386 - CMOVcc reg16,mem P6 - CMOVcc reg16,reg16 P6 - CMOVcc reg32,mem P6 - CMOVcc reg32,reg32 P6 - CMOVcc reg64,mem X64 - CMOVcc reg64,reg64 X64 - Jcc imm|near 386 - Jcc imm16|near 386 - Jcc imm32|near 386 - Jcc imm|short 8086,ND - Jcc imm 8086,ND - Jcc imm 386,ND - Jcc imm 8086,ND - Jcc imm 8086 - SETcc mem 386 - SETcc reg8 386 - - B.1.3 Katmai Streaming SIMD instructions (SSE -- a.k.a. KNI, XMM, MMX2) - - ADDPS xmmreg,xmmrm KATMAI,SSE - ADDSS xmmreg,xmmrm KATMAI,SSE,SD - ANDNPS xmmreg,xmmrm KATMAI,SSE - ANDPS xmmreg,xmmrm KATMAI,SSE - CMPEQPS xmmreg,xmmrm KATMAI,SSE - CMPEQSS xmmreg,xmmrm KATMAI,SSE - CMPLEPS xmmreg,xmmrm KATMAI,SSE - CMPLESS xmmreg,xmmrm KATMAI,SSE - CMPLTPS xmmreg,xmmrm KATMAI,SSE - CMPLTSS xmmreg,xmmrm KATMAI,SSE - CMPNEQPS xmmreg,xmmrm KATMAI,SSE - CMPNEQSS xmmreg,xmmrm KATMAI,SSE - CMPNLEPS xmmreg,xmmrm KATMAI,SSE - CMPNLESS xmmreg,xmmrm KATMAI,SSE - CMPNLTPS xmmreg,xmmrm KATMAI,SSE - CMPNLTSS xmmreg,xmmrm KATMAI,SSE - CMPORDPS xmmreg,xmmrm KATMAI,SSE - CMPORDSS xmmreg,xmmrm KATMAI,SSE - CMPUNORDPS xmmreg,xmmrm KATMAI,SSE - CMPUNORDSS xmmreg,xmmrm KATMAI,SSE - CMPPS xmmreg,mem,imm KATMAI,SSE - CMPPS xmmreg,xmmreg,imm KATMAI,SSE - CMPSS xmmreg,mem,imm KATMAI,SSE - CMPSS xmmreg,xmmreg,imm KATMAI,SSE - COMISS xmmreg,xmmrm KATMAI,SSE - CVTPI2PS xmmreg,mmxrm KATMAI,SSE,MMX - CVTPS2PI mmxreg,xmmrm KATMAI,SSE,MMX - CVTSI2SS xmmreg,mem KATMAI,SSE,SD,AR1,ND - CVTSI2SS xmmreg,rm32 KATMAI,SSE,SD,AR1 - CVTSI2SS xmmreg,rm64 X64,SSE,AR1 - CVTSS2SI reg32,xmmreg KATMAI,SSE,SD,AR1 - CVTSS2SI reg32,mem KATMAI,SSE,SD,AR1 - CVTSS2SI reg64,xmmreg X64,SSE,SD,AR1 - CVTSS2SI reg64,mem X64,SSE,SD,AR1 - CVTTPS2PI mmxreg,xmmrm KATMAI,SSE,MMX - CVTTSS2SI reg32,xmmrm KATMAI,SSE,SD,AR1 - CVTTSS2SI reg64,xmmrm X64,SSE,SD,AR1 - DIVPS xmmreg,xmmrm KATMAI,SSE - DIVSS xmmreg,xmmrm KATMAI,SSE - LDMXCSR mem KATMAI,SSE,SD - MAXPS xmmreg,xmmrm KATMAI,SSE - MAXSS xmmreg,xmmrm KATMAI,SSE - MINPS xmmreg,xmmrm KATMAI,SSE - MINSS xmmreg,xmmrm KATMAI,SSE - MOVAPS xmmreg,mem KATMAI,SSE - MOVAPS mem,xmmreg KATMAI,SSE - MOVAPS xmmreg,xmmreg KATMAI,SSE - MOVAPS xmmreg,xmmreg KATMAI,SSE - MOVHPS xmmreg,mem KATMAI,SSE - MOVHPS mem,xmmreg KATMAI,SSE - MOVLHPS xmmreg,xmmreg KATMAI,SSE - MOVLPS xmmreg,mem KATMAI,SSE - MOVLPS mem,xmmreg KATMAI,SSE - MOVHLPS xmmreg,xmmreg KATMAI,SSE - MOVMSKPS reg32,xmmreg KATMAI,SSE - MOVMSKPS reg64,xmmreg X64,SSE - MOVNTPS mem,xmmreg KATMAI,SSE - MOVSS xmmreg,mem KATMAI,SSE - MOVSS mem,xmmreg KATMAI,SSE - MOVSS xmmreg,xmmreg KATMAI,SSE - MOVSS xmmreg,xmmreg KATMAI,SSE - MOVUPS xmmreg,mem KATMAI,SSE - MOVUPS mem,xmmreg KATMAI,SSE - MOVUPS xmmreg,xmmreg KATMAI,SSE - MOVUPS xmmreg,xmmreg KATMAI,SSE - MULPS xmmreg,xmmrm KATMAI,SSE - MULSS xmmreg,xmmrm KATMAI,SSE - ORPS xmmreg,xmmrm KATMAI,SSE - RCPPS xmmreg,xmmrm KATMAI,SSE - RCPSS xmmreg,xmmrm KATMAI,SSE - RSQRTPS xmmreg,xmmrm KATMAI,SSE - RSQRTSS xmmreg,xmmrm KATMAI,SSE - SHUFPS xmmreg,mem,imm KATMAI,SSE - SHUFPS xmmreg,xmmreg,imm KATMAI,SSE - SQRTPS xmmreg,xmmrm KATMAI,SSE - SQRTSS xmmreg,xmmrm KATMAI,SSE - STMXCSR mem KATMAI,SSE,SD - SUBPS xmmreg,xmmrm KATMAI,SSE - SUBSS xmmreg,xmmrm KATMAI,SSE - UCOMISS xmmreg,xmmrm KATMAI,SSE - UNPCKHPS xmmreg,xmmrm KATMAI,SSE - UNPCKLPS xmmreg,xmmrm KATMAI,SSE - XORPS xmmreg,xmmrm KATMAI,SSE - - B.1.4 Introduced in Deschutes but necessary for SSE support - - FXRSTOR mem P6,SSE,FPU - FXSAVE mem P6,SSE,FPU - - B.1.5 XSAVE group (AVX and extended state) - - XGETBV NEHALEM - XSETBV NEHALEM,PRIV - XSAVE mem NEHALEM - XRSTOR mem NEHALEM - - B.1.6 Generic memory operations - - PREFETCHNTA mem KATMAI - PREFETCHT0 mem KATMAI - PREFETCHT1 mem KATMAI - PREFETCHT2 mem KATMAI - SFENCE KATMAI - - B.1.7 New MMX instructions introduced in Katmai - - MASKMOVQ mmxreg,mmxreg KATMAI,MMX - MOVNTQ mem,mmxreg KATMAI,MMX - PAVGB mmxreg,mmxrm KATMAI,MMX - PAVGW mmxreg,mmxrm KATMAI,MMX - PEXTRW reg32,mmxreg,imm KATMAI,MMX - PINSRW mmxreg,mem,imm KATMAI,MMX - PINSRW mmxreg,rm16,imm KATMAI,MMX - PINSRW mmxreg,reg32,imm KATMAI,MMX - PMAXSW mmxreg,mmxrm KATMAI,MMX - PMAXUB mmxreg,mmxrm KATMAI,MMX - PMINSW mmxreg,mmxrm KATMAI,MMX - PMINUB mmxreg,mmxrm KATMAI,MMX - PMOVMSKB reg32,mmxreg KATMAI,MMX - PMULHUW mmxreg,mmxrm KATMAI,MMX - PSADBW mmxreg,mmxrm KATMAI,MMX - PSHUFW mmxreg,mmxrm,imm KATMAI,MMX2 - - B.1.8 AMD Enhanced 3DNow! (Athlon) instructions - - PF2IW mmxreg,mmxrm PENT,3DNOW - PFNACC mmxreg,mmxrm PENT,3DNOW - PFPNACC mmxreg,mmxrm PENT,3DNOW - PI2FW mmxreg,mmxrm PENT,3DNOW - PSWAPD mmxreg,mmxrm PENT,3DNOW - - B.1.9 Willamette SSE2 Cacheability Instructions - - MASKMOVDQU xmmreg,xmmreg WILLAMETTE,SSE2 - CLFLUSH mem WILLAMETTE,SSE2 - MOVNTDQ mem,xmmreg WILLAMETTE,SSE2,SO - MOVNTI mem,reg32 WILLAMETTE,SD - MOVNTI mem,reg64 X64 - MOVNTPD mem,xmmreg WILLAMETTE,SSE2,SO - LFENCE WILLAMETTE,SSE2 - MFENCE WILLAMETTE,SSE2 - -B.1.10 Willamette MMX instructions (SSE2 SIMD Integer Instructions) - - MOVD mem,xmmreg WILLAMETTE,SSE2,SD - MOVD xmmreg,mem WILLAMETTE,SSE2,SD - MOVD xmmreg,rm32 WILLAMETTE,SSE2 - MOVD rm32,xmmreg WILLAMETTE,SSE2 - MOVDQA xmmreg,xmmreg WILLAMETTE,SSE2 - MOVDQA mem,xmmreg WILLAMETTE,SSE2,SO - MOVDQA xmmreg,mem WILLAMETTE,SSE2,SO - MOVDQA xmmreg,xmmreg WILLAMETTE,SSE2 - MOVDQU xmmreg,xmmreg WILLAMETTE,SSE2 - MOVDQU mem,xmmreg WILLAMETTE,SSE2,SO - MOVDQU xmmreg,mem WILLAMETTE,SSE2,SO - MOVDQU xmmreg,xmmreg WILLAMETTE,SSE2 - MOVDQ2Q mmxreg,xmmreg WILLAMETTE,SSE2 - MOVQ xmmreg,xmmreg WILLAMETTE,SSE2 - MOVQ xmmreg,xmmreg WILLAMETTE,SSE2 - MOVQ mem,xmmreg WILLAMETTE,SSE2 - MOVQ xmmreg,mem WILLAMETTE,SSE2 - MOVQ xmmreg,rm64 X64,SSE2 - MOVQ rm64,xmmreg X64,SSE2 - MOVQ2DQ xmmreg,mmxreg WILLAMETTE,SSE2 - PACKSSWB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PACKSSDW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PACKUSWB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PADDB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PADDW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PADDD xmmreg,xmmrm WILLAMETTE,SSE2,SO - PADDQ mmxreg,mmxrm WILLAMETTE,MMX - PADDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - PADDSB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PADDSW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PADDUSB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PADDUSW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PAND xmmreg,xmmrm WILLAMETTE,SSE2,SO - PANDN xmmreg,xmmrm WILLAMETTE,SSE2,SO - PAVGB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PAVGW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PCMPEQB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PCMPEQW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PCMPEQD xmmreg,xmmrm WILLAMETTE,SSE2,SO - PCMPGTB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PCMPGTW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PCMPGTD xmmreg,xmmrm WILLAMETTE,SSE2,SO - PEXTRW reg32,xmmreg,imm WILLAMETTE,SSE2 - PINSRW xmmreg,reg16,imm WILLAMETTE,SSE2 - PINSRW xmmreg,reg32,imm WILLAMETTE,SSE2,ND - PINSRW xmmreg,mem,imm WILLAMETTE,SSE2 - PINSRW xmmreg,mem16,imm WILLAMETTE,SSE2 - PMADDWD xmmreg,xmmrm WILLAMETTE,SSE2,SO - PMAXSW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PMAXUB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PMINSW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PMINUB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PMOVMSKB reg32,xmmreg WILLAMETTE,SSE2 - PMULHUW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PMULHW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PMULLW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PMULUDQ mmxreg,mmxrm WILLAMETTE,SSE2,SO - PMULUDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - POR xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSADBW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSHUFD xmmreg,xmmreg,imm WILLAMETTE,SSE2 - PSHUFD xmmreg,mem,imm WILLAMETTE,SSE22 - PSHUFHW xmmreg,xmmreg,imm WILLAMETTE,SSE2 - PSHUFHW xmmreg,mem,imm WILLAMETTE,SSE22 - PSHUFLW xmmreg,xmmreg,imm WILLAMETTE,SSE2 - PSHUFLW xmmreg,mem,imm WILLAMETTE,SSE22 - PSLLDQ xmmreg,imm WILLAMETTE,SSE2,AR1 - PSLLW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSLLW xmmreg,imm WILLAMETTE,SSE2,AR1 - PSLLD xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSLLD xmmreg,imm WILLAMETTE,SSE2,AR1 - PSLLQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSLLQ xmmreg,imm WILLAMETTE,SSE2,AR1 - PSRAW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSRAW xmmreg,imm WILLAMETTE,SSE2,AR1 - PSRAD xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSRAD xmmreg,imm WILLAMETTE,SSE2,AR1 - PSRLDQ xmmreg,imm WILLAMETTE,SSE2,AR1 - PSRLW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSRLW xmmreg,imm WILLAMETTE,SSE2,AR1 - PSRLD xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSRLD xmmreg,imm WILLAMETTE,SSE2,AR1 - PSRLQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSRLQ xmmreg,imm WILLAMETTE,SSE2,AR1 - PSUBB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSUBW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSUBD xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSUBQ mmxreg,mmxrm WILLAMETTE,SSE2,SO - PSUBQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSUBSB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSUBSW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSUBUSB xmmreg,xmmrm WILLAMETTE,SSE2,SO - PSUBUSW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PUNPCKHBW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PUNPCKHWD xmmreg,xmmrm WILLAMETTE,SSE2,SO - PUNPCKHDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - PUNPCKHQDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - PUNPCKLBW xmmreg,xmmrm WILLAMETTE,SSE2,SO - PUNPCKLWD xmmreg,xmmrm WILLAMETTE,SSE2,SO - PUNPCKLDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - PUNPCKLQDQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - PXOR xmmreg,xmmrm WILLAMETTE,SSE2,SO - -B.1.11 Willamette Streaming SIMD instructions (SSE2) - - ADDPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - ADDSD xmmreg,xmmrm WILLAMETTE,SSE2 - ANDNPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - ANDPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - CMPEQPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - CMPEQSD xmmreg,xmmrm WILLAMETTE,SSE2 - CMPLEPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - CMPLESD xmmreg,xmmrm WILLAMETTE,SSE2 - CMPLTPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - CMPLTSD xmmreg,xmmrm WILLAMETTE,SSE2 - CMPNEQPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - CMPNEQSD xmmreg,xmmrm WILLAMETTE,SSE2 - CMPNLEPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - CMPNLESD xmmreg,xmmrm WILLAMETTE,SSE2 - CMPNLTPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - CMPNLTSD xmmreg,xmmrm WILLAMETTE,SSE2 - CMPORDPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - CMPORDSD xmmreg,xmmrm WILLAMETTE,SSE2 - CMPUNORDPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - CMPUNORDSD xmmreg,xmmrm WILLAMETTE,SSE2 - CMPPD xmmreg,xmmrm,imm WILLAMETTE,SSE22 - CMPSD xmmreg,xmmrm,imm WILLAMETTE,SSE2 - COMISD xmmreg,xmmrm WILLAMETTE,SSE2 - CVTDQ2PD xmmreg,xmmrm WILLAMETTE,SSE2 - CVTDQ2PS xmmreg,xmmrm WILLAMETTE,SSE2,SO - CVTPD2DQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - CVTPD2PI mmxreg,xmmrm WILLAMETTE,SSE2,SO - CVTPD2PS xmmreg,xmmrm WILLAMETTE,SSE2,SO - CVTPI2PD xmmreg,mmxrm WILLAMETTE,SSE2 - CVTPS2DQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - CVTPS2PD xmmreg,xmmrm WILLAMETTE,SSE2 - CVTSD2SI reg32,xmmreg WILLAMETTE,SSE2,AR1 - CVTSD2SI reg32,mem WILLAMETTE,SSE2,AR1 - CVTSD2SI reg64,xmmreg X64,SSE2,AR1 - CVTSD2SI reg64,mem X64,SSE2,AR1 - CVTSD2SS xmmreg,xmmrm WILLAMETTE,SSE2 - CVTSI2SD xmmreg,mem WILLAMETTE,SSE2,SD,AR1,ND - CVTSI2SD xmmreg,rm32 WILLAMETTE,SSE2,SD,AR1 - CVTSI2SD xmmreg,rm64 X64,SSE2,AR1 - CVTSS2SD xmmreg,xmmrm WILLAMETTE,SSE2,SD - CVTTPD2PI mmxreg,xmmrm WILLAMETTE,SSE2,SO - CVTTPD2DQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - CVTTPS2DQ xmmreg,xmmrm WILLAMETTE,SSE2,SO - CVTTSD2SI reg32,xmmreg WILLAMETTE,SSE2,AR1 - CVTTSD2SI reg32,mem WILLAMETTE,SSE2,AR1 - CVTTSD2SI reg64,xmmreg X64,SSE2,AR1 - CVTTSD2SI reg64,mem X64,SSE2,AR1 - DIVPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - DIVSD xmmreg,xmmrm WILLAMETTE,SSE2 - MAXPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - MAXSD xmmreg,xmmrm WILLAMETTE,SSE2 - MINPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - MINSD xmmreg,xmmrm WILLAMETTE,SSE2 - MOVAPD xmmreg,xmmreg WILLAMETTE,SSE2 - MOVAPD xmmreg,xmmreg WILLAMETTE,SSE2 - MOVAPD mem,xmmreg WILLAMETTE,SSE2,SO - MOVAPD xmmreg,mem WILLAMETTE,SSE2,SO - MOVHPD mem,xmmreg WILLAMETTE,SSE2 - MOVHPD xmmreg,mem WILLAMETTE,SSE2 - MOVLPD mem,xmmreg WILLAMETTE,SSE2 - MOVLPD xmmreg,mem WILLAMETTE,SSE2 - MOVMSKPD reg32,xmmreg WILLAMETTE,SSE2 - MOVMSKPD reg64,xmmreg X64,SSE2 - MOVSD xmmreg,xmmreg WILLAMETTE,SSE2 - MOVSD xmmreg,xmmreg WILLAMETTE,SSE2 - MOVSD mem,xmmreg WILLAMETTE,SSE2 - MOVSD xmmreg,mem WILLAMETTE,SSE2 - MOVUPD xmmreg,xmmreg WILLAMETTE,SSE2 - MOVUPD xmmreg,xmmreg WILLAMETTE,SSE2 - MOVUPD mem,xmmreg WILLAMETTE,SSE2,SO - MOVUPD xmmreg,mem WILLAMETTE,SSE2,SO - MULPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - MULSD xmmreg,xmmrm WILLAMETTE,SSE2 - ORPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - SHUFPD xmmreg,xmmreg,imm WILLAMETTE,SSE2 - SHUFPD xmmreg,mem,imm WILLAMETTE,SSE2 - SQRTPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - SQRTSD xmmreg,xmmrm WILLAMETTE,SSE2 - SUBPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - SUBSD xmmreg,xmmrm WILLAMETTE,SSE2 - UCOMISD xmmreg,xmmrm WILLAMETTE,SSE2 - UNPCKHPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - UNPCKLPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - XORPD xmmreg,xmmrm WILLAMETTE,SSE2,SO - -B.1.12 Prescott New Instructions (SSE3) - - ADDSUBPD xmmreg,xmmrm PRESCOTT,SSE3,SO - ADDSUBPS xmmreg,xmmrm PRESCOTT,SSE3,SO - HADDPD xmmreg,xmmrm PRESCOTT,SSE3,SO - HADDPS xmmreg,xmmrm PRESCOTT,SSE3,SO - HSUBPD xmmreg,xmmrm PRESCOTT,SSE3,SO - HSUBPS xmmreg,xmmrm PRESCOTT,SSE3,SO - LDDQU xmmreg,mem PRESCOTT,SSE3,SO - MOVDDUP xmmreg,xmmrm PRESCOTT,SSE3 - MOVSHDUP xmmreg,xmmrm PRESCOTT,SSE3 - MOVSLDUP xmmreg,xmmrm PRESCOTT,SSE3 - -B.1.13 VMX Instructions - - VMCALL VMX - VMCLEAR mem VMX - VMLAUNCH VMX - VMLOAD X64,VMX - VMMCALL X64,VMX - VMPTRLD mem VMX - VMPTRST mem VMX - VMREAD rm32,reg32 VMX,NOLONG,SD - VMREAD rm64,reg64 X64,VMX - VMRESUME VMX - VMRUN X64,VMX - VMSAVE X64,VMX - VMWRITE reg32,rm32 VMX,NOLONG,SD - VMWRITE reg64,rm64 X64,VMX - VMXOFF VMX - VMXON mem VMX - -B.1.14 Extended Page Tables VMX instructions - - INVEPT reg32,mem VMX,SO,NOLONG - INVEPT reg64,mem VMX,SO,LONG - INVVPID reg32,mem VMX,SO,NOLONG - INVVPID reg64,mem VMX,SO,LONG - -B.1.15 Tejas New Instructions (SSSE3) - - PABSB mmxreg,mmxrm SSSE3,MMX - PABSB xmmreg,xmmrm SSSE3 - PABSW mmxreg,mmxrm SSSE3,MMX - PABSW xmmreg,xmmrm SSSE3 - PABSD mmxreg,mmxrm SSSE3,MMX - PABSD xmmreg,xmmrm SSSE3 - PALIGNR mmxreg,mmxrm,imm SSSE3,MMX - PALIGNR xmmreg,xmmrm,imm SSSE3 - PHADDW mmxreg,mmxrm SSSE3,MMX - PHADDW xmmreg,xmmrm SSSE3 - PHADDD mmxreg,mmxrm SSSE3,MMX - PHADDD xmmreg,xmmrm SSSE3 - PHADDSW mmxreg,mmxrm SSSE3,MMX - PHADDSW xmmreg,xmmrm SSSE3 - PHSUBW mmxreg,mmxrm SSSE3,MMX - PHSUBW xmmreg,xmmrm SSSE3 - PHSUBD mmxreg,mmxrm SSSE3,MMX - PHSUBD xmmreg,xmmrm SSSE3 - PHSUBSW mmxreg,mmxrm SSSE3,MMX - PHSUBSW xmmreg,xmmrm SSSE3 - PMADDUBSW mmxreg,mmxrm SSSE3,MMX - PMADDUBSW xmmreg,xmmrm SSSE3 - PMULHRSW mmxreg,mmxrm SSSE3,MMX - PMULHRSW xmmreg,xmmrm SSSE3 - PSHUFB mmxreg,mmxrm SSSE3,MMX - PSHUFB xmmreg,xmmrm SSSE3 - PSIGNB mmxreg,mmxrm SSSE3,MMX - PSIGNB xmmreg,xmmrm SSSE3 - PSIGNW mmxreg,mmxrm SSSE3,MMX - PSIGNW xmmreg,xmmrm SSSE3 - PSIGND mmxreg,mmxrm SSSE3,MMX - PSIGND xmmreg,xmmrm SSSE3 - -B.1.16 AMD SSE4A - - EXTRQ xmmreg,imm,imm SSE4A,AMD - EXTRQ xmmreg,xmmreg SSE4A,AMD - INSERTQ xmmreg,xmmreg,imm,imm SSE4A,AMD - INSERTQ xmmreg,xmmreg SSE4A,AMD - MOVNTSD mem,xmmreg SSE4A,AMD - MOVNTSS mem,xmmreg SSE4A,AMD,SD - -B.1.17 New instructions in Barcelona - - LZCNT reg16,rm16 P6,AMD - LZCNT reg32,rm32 P6,AMD - LZCNT reg64,rm64 X64,AMD - -B.1.18 Penryn New Instructions (SSE4.1) - - BLENDPD xmmreg,xmmrm,imm SSE41 - BLENDPS xmmreg,xmmrm,imm SSE41 - BLENDVPD xmmreg,xmmrm,xmm0 SSE41 - BLENDVPS xmmreg,xmmrm,xmm0 SSE41 - DPPD xmmreg,xmmrm,imm SSE41 - DPPS xmmreg,xmmrm,imm SSE41 - EXTRACTPS rm32,xmmreg,imm SSE41 - EXTRACTPS reg64,xmmreg,imm SSE41,X64 - INSERTPS xmmreg,xmmrm,imm SSE41,SD - MOVNTDQA xmmreg,mem SSE41 - MPSADBW xmmreg,xmmrm,imm SSE41 - PACKUSDW xmmreg,xmmrm SSE41 - PBLENDVB xmmreg,xmmrm,xmm0 SSE41 - PBLENDW xmmreg,xmmrm,imm SSE41 - PCMPEQQ xmmreg,xmmrm SSE41 - PEXTRB reg32,xmmreg,imm SSE41 - PEXTRB mem8,xmmreg,imm SSE41 - PEXTRB reg64,xmmreg,imm SSE41,X64 - PEXTRD rm32,xmmreg,imm SSE41 - PEXTRQ rm64,xmmreg,imm SSE41,X64 - PEXTRW reg32,xmmreg,imm SSE41 - PEXTRW mem16,xmmreg,imm SSE41 - PEXTRW reg64,xmmreg,imm SSE41,X64 - PHMINPOSUW xmmreg,xmmrm SSE41 - PINSRB xmmreg,mem,imm SSE41 - PINSRB xmmreg,rm8,imm SSE41 - PINSRB xmmreg,reg32,imm SSE41 - PINSRD xmmreg,mem,imm SSE41 - PINSRD xmmreg,rm32,imm SSE41 - PINSRQ xmmreg,mem,imm SSE41,X64 - PINSRQ xmmreg,rm64,imm SSE41,X64 - PMAXSB xmmreg,xmmrm SSE41 - PMAXSD xmmreg,xmmrm SSE41 - PMAXUD xmmreg,xmmrm SSE41 - PMAXUW xmmreg,xmmrm SSE41 - PMINSB xmmreg,xmmrm SSE41 - PMINSD xmmreg,xmmrm SSE41 - PMINUD xmmreg,xmmrm SSE41 - PMINUW xmmreg,xmmrm SSE41 - PMOVSXBW xmmreg,xmmrm SSE41 - PMOVSXBD xmmreg,xmmrm SSE41,SD - PMOVSXBQ xmmreg,xmmrm SSE41,SW - PMOVSXWD xmmreg,xmmrm SSE41 - PMOVSXWQ xmmreg,xmmrm SSE41,SD - PMOVSXDQ xmmreg,xmmrm SSE41 - PMOVZXBW xmmreg,xmmrm SSE41 - PMOVZXBD xmmreg,xmmrm SSE41,SD - PMOVZXBQ xmmreg,xmmrm SSE41,SW - PMOVZXWD xmmreg,xmmrm SSE41 - PMOVZXWQ xmmreg,xmmrm SSE41,SD - PMOVZXDQ xmmreg,xmmrm SSE41 - PMULDQ xmmreg,xmmrm SSE41 - PMULLD xmmreg,xmmrm SSE41 - PTEST xmmreg,xmmrm SSE41 - ROUNDPD xmmreg,xmmrm,imm SSE41 - ROUNDPS xmmreg,xmmrm,imm SSE41 - ROUNDSD xmmreg,xmmrm,imm SSE41 - ROUNDSS xmmreg,xmmrm,imm SSE41 - -B.1.19 Nehalem New Instructions (SSE4.2) - - CRC32 reg32,rm8 SSE42 - CRC32 reg32,rm16 SSE42 - CRC32 reg32,rm32 SSE42 - CRC32 reg64,rm8 SSE42,X64 - CRC32 reg64,rm64 SSE42,X64 - PCMPESTRI xmmreg,xmmrm,imm SSE42 - PCMPESTRM xmmreg,xmmrm,imm SSE42 - PCMPISTRI xmmreg,xmmrm,imm SSE42 - PCMPISTRM xmmreg,xmmrm,imm SSE42 - PCMPGTQ xmmreg,xmmrm SSE42 - POPCNT reg16,rm16 NEHALEM,SW - POPCNT reg32,rm32 NEHALEM,SD - POPCNT reg64,rm64 NEHALEM,X64 - -B.1.20 Intel SMX - - GETSEC KATMAI - -B.1.21 Geode (Cyrix) 3DNow! additions - - PFRCPV mmxreg,mmxrm PENT,3DNOW,CYRIX - PFRSQRTV mmxreg,mmxrm PENT,3DNOW,CYRIX - -B.1.22 Intel new instructions in ??? - - MOVBE reg16,mem16 NEHALEM - MOVBE reg32,mem32 NEHALEM - MOVBE reg64,mem64 NEHALEM - MOVBE mem16,reg16 NEHALEM - MOVBE mem32,reg32 NEHALEM - MOVBE mem64,reg64 NEHALEM - -B.1.23 Intel AES instructions - - AESENC xmmreg,xmmrm128 SSE,WESTMERE - AESENCLAST xmmreg,xmmrm128 SSE,WESTMERE - AESDEC xmmreg,xmmrm128 SSE,WESTMERE - AESDECLAST xmmreg,xmmrm128 SSE,WESTMERE - AESIMC xmmreg,xmmrm128 SSE,WESTMERE - AESKEYGENASSIST xmmreg,xmmrm128,imm8 SSE,WESTMERE - -B.1.24 Intel AVX AES instructions - - VAESENC xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VAESENCLAST xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VAESDEC xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VAESDECLAST xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VAESIMC xmmreg,xmmrm128 AVX,SANDYBRIDGE - VAESKEYGENASSIST xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - -B.1.25 Intel AVX instructions - - VADDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VADDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VADDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VADDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VADDSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VADDSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VADDSUBPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VADDSUBPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VADDSUBPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VADDSUBPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VANDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VANDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VANDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VANDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VANDNPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VANDNPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VANDNPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VANDNPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VBLENDPD xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VBLENDPD ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE - VBLENDPS xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VBLENDPS ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE - VBLENDVPD xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE - VBLENDVPD xmmreg,xmmrm128,xmm0 AVX,SANDYBRIDGE - VBLENDVPD ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE - VBLENDVPD ymmreg,ymmrm256,ymm0 AVX,SANDYBRIDGE - VBLENDVPS xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE - VBLENDVPS xmmreg,xmmrm128,xmm0 AVX,SANDYBRIDGE - VBLENDVPS ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE - VBLENDVPD ymmreg,ymmrm256,ymm0 AVX,SANDYBRIDGE - VBROADCASTSS xmmreg,mem32 AVX,SANDYBRIDGE - VBROADCASTSS ymmreg,mem32 AVX,SANDYBRIDGE - VBROADCASTSD ymmreg,mem64 AVX,SANDYBRIDGE - VBROADCASTF128 ymmreg,mem128 AVX,SANDYBRIDGE - VCMPEQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPEQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPLTPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPLTPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPLEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPLEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPUNORDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPUNORDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNEQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNEQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNLTPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNLTPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNLEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNLEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPORDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPORDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPEQ_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPEQ_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNGEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNGEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNGTPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNGTPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPFALSEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPFALSEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNEQ_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNEQ_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPGEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPGEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPGTPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPGTPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPTRUEPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPTRUEPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPEQ_OSPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPEQ_OSPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPLT_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPLT_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPLE_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPLE_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPUNORD_SPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPUNORD_SPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNEQ_USPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNEQ_USPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNLT_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNLT_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNLE_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNLE_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPORD_SPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPORD_SPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPEQ_USPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPEQ_USPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNGE_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNGE_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNGT_UQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNGT_UQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPFALSE_OSPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPFALSE_OSPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNEQ_OSPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNEQ_OSPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPGE_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPGE_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPGT_OQPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPGT_OQPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPTRUE_USPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPTRUE_USPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPPD xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VCMPPD ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE - VCMPEQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPEQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPLTPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPLTPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPLEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPLEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPUNORDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPUNORDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNEQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNEQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNLTPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNLTPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNLEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNLEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPORDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPORDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPEQ_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPEQ_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNGEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNGEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNGTPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNGTPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPFALSEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPFALSEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNEQ_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNEQ_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPGEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPGEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPGTPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPGTPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPTRUEPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPTRUEPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPEQ_OSPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPEQ_OSPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPLT_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPLT_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPLE_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPLE_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPUNORD_SPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPUNORD_SPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNEQ_USPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNEQ_USPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNLT_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNLT_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNLE_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNLE_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPORD_SPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPORD_SPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPEQ_USPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPEQ_USPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNGE_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNGE_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNGT_UQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNGT_UQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPFALSE_OSPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPFALSE_OSPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPNEQ_OSPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPNEQ_OSPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPGE_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPGE_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPGT_OQPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPGT_OQPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPTRUE_USPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VCMPTRUE_USPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VCMPPS xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VCMPPS ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE - VCMPEQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPLTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPLESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPUNORDSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNEQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNLTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNLESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPORDSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPEQ_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNGESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNGTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPFALSESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNEQ_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPGESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPGTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPTRUESD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPEQ_OSSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPLT_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPLE_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPUNORD_SSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNEQ_USSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNLT_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNLE_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPORD_SSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPEQ_USSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNGE_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNGT_UQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPFALSE_OSSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPNEQ_OSSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPGE_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPGT_OQSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPTRUE_USSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCMPSD xmmreg,xmmreg*,xmmrm64,imm8 AVX,SANDYBRIDGE - VCMPEQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPLTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPLESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPUNORDSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNEQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNLTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNLESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPORDSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPEQ_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNGESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNGTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPFALSESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNEQ_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPGESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPGTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPTRUESS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPEQ_OSSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPLT_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPLE_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPUNORD_SSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNEQ_USSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNLT_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNLE_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPORD_SSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPEQ_USSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNGE_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNGT_UQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPFALSE_OSSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPNEQ_OSSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPGE_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPGT_OQSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPTRUE_USSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCMPSS xmmreg,xmmreg*,xmmrm32,imm8 AVX,SANDYBRIDGE - VCOMISD xmmreg,xmmrm64 AVX,SANDYBRIDGE - VCOMISS xmmreg,xmmrm32 AVX,SANDYBRIDGE - VCVTDQ2PD xmmreg,xmmrm64 AVX,SANDYBRIDGE - VCVTDQ2PD ymmreg,xmmrm128 AVX,SANDYBRIDGE - VCVTDQ2PS xmmreg,xmmrm128 AVX,SANDYBRIDGE - VCVTDQ2PS ymmreg,ymmrm256 AVX,SANDYBRIDGE - VCVTPD2DQ xmmreg,xmmreg AVX,SANDYBRIDGE - VCVTPD2DQ xmmreg,mem128 AVX,SANDYBRIDGE,SO - VCVTPD2DQ xmmreg,ymmreg AVX,SANDYBRIDGE - VCVTPD2DQ xmmreg,mem256 AVX,SANDYBRIDGE,SY - VCVTPD2PS xmmreg,xmmreg AVX,SANDYBRIDGE - VCVTPD2PS xmmreg,mem128 AVX,SANDYBRIDGE,SO - VCVTPD2PS xmmreg,ymmreg AVX,SANDYBRIDGE - VCVTPD2PS xmmreg,mem256 AVX,SANDYBRIDGE,SY - VCVTPS2DQ xmmreg,xmmrm128 AVX,SANDYBRIDGE - VCVTPS2DQ ymmreg,ymmrm256 AVX,SANDYBRIDGE - VCVTPS2PD xmmreg,xmmrm64 AVX,SANDYBRIDGE - VCVTPS2PD ymmreg,xmmrm128 AVX,SANDYBRIDGE - VCVTSD2SI reg32,xmmrm64 AVX,SANDYBRIDGE - VCVTSD2SI reg64,xmmrm64 AVX,SANDYBRIDGE,LONG - VCVTSD2SS xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VCVTSI2SD xmmreg,xmmreg*,rm32 AVX,SANDYBRIDGE,SD - VCVTSI2SD xmmreg,xmmreg*,mem32 AVX,SANDYBRIDGE,ND,SD - VCVTSI2SD xmmreg,xmmreg*,rm64 AVX,SANDYBRIDGE,LONG - VCVTSI2SS xmmreg,xmmreg*,rm32 AVX,SANDYBRIDGE,SD - VCVTSI2SS xmmreg,xmmreg*,mem32 AVX,SANDYBRIDGE,ND,SD - VCVTSI2SS xmmreg,xmmreg*,rm64 AVX,SANDYBRIDGE,LONG - VCVTSS2SD xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VCVTSS2SI reg32,xmmrm32 AVX,SANDYBRIDGE - VCVTSS2SI reg64,xmmrm32 AVX,SANDYBRIDGE,LONG - VCVTTPD2DQ xmmreg,xmmreg AVX,SANDYBRIDGE - VCVTTPD2DQ xmmreg,mem128 AVX,SANDYBRIDGE,SO - VCVTTPD2DQ xmmreg,ymmreg AVX,SANDYBRIDGE - VCVTTPD2DQ xmmreg,mem256 AVX,SANDYBRIDGE,SY - VCVTTPS2DQ xmmreg,xmmrm128 AVX,SANDYBRIDGE - VCVTTPS2DQ ymmreg,ymmrm256 AVX,SANDYBRIDGE - VCVTTSD2SI reg32,xmmrm64 AVX,SANDYBRIDGE - VCVTTSD2SI reg64,xmmrm64 AVX,SANDYBRIDGE,LONG - VCVTTSS2SI reg32,xmmrm32 AVX,SANDYBRIDGE - VCVTTSS2SI reg64,xmmrm32 AVX,SANDYBRIDGE,LONG - VDIVPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VDIVPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VDIVPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VDIVPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VDIVSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VDIVSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VDPPD xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VDPPS xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VDPPS ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE - VEXTRACTF128 xmmrm128,xmmreg,imm8 AVX,SANDYBRIDGE - VEXTRACTPS rm32,xmmreg,imm8 AVX,SANDYBRIDGE - VHADDPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VHADDPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VHADDPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VHADDPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VHSUBPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VHSUBPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VHSUBPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VHSUBPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VINSERTF128 ymmreg,ymmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VINSERTPS xmmreg,xmmreg*,xmmrm32,imm8 AVX,SANDYBRIDGE - VLDDQU xmmreg,mem128 AVX,SANDYBRIDGE - VLDQQU ymmreg,mem256 AVX,SANDYBRIDGE - VLDDQU ymmreg,mem256 AVX,SANDYBRIDGE - VLDMXCSR mem32 AVX,SANDYBRIDGE - VMASKMOVDQU xmmreg,xmmreg AVX,SANDYBRIDGE - VMASKMOVPS xmmreg,xmmreg,mem128 AVX,SANDYBRIDGE - VMASKMOVPS ymmreg,ymmreg,mem256 AVX,SANDYBRIDGE - VMASKMOVPS mem128,xmmreg,xmmreg AVX,SANDYBRIDGE,SO - VMASKMOVPS mem256,xmmreg,xmmreg AVX,SANDYBRIDGE,SY - VMASKMOVPD xmmreg,xmmreg,mem128 AVX,SANDYBRIDGE - VMASKMOVPD ymmreg,ymmreg,mem256 AVX,SANDYBRIDGE - VMASKMOVPD mem128,xmmreg,xmmreg AVX,SANDYBRIDGE - VMASKMOVPD mem256,ymmreg,ymmreg AVX,SANDYBRIDGE - VMAXPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VMAXPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VMAXPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VMAXPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VMAXSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VMAXSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VMINPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VMINPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VMINPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VMINPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VMINSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VMINSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VMOVAPD xmmreg,xmmrm128 AVX,SANDYBRIDGE - VMOVAPD xmmrm128,xmmreg AVX,SANDYBRIDGE - VMOVAPD ymmreg,ymmrm256 AVX,SANDYBRIDGE - VMOVAPD ymmrm256,ymmreg AVX,SANDYBRIDGE - VMOVAPS xmmreg,xmmrm128 AVX,SANDYBRIDGE - VMOVAPS xmmrm128,xmmreg AVX,SANDYBRIDGE - VMOVAPS ymmreg,ymmrm256 AVX,SANDYBRIDGE - VMOVAPS ymmrm256,ymmreg AVX,SANDYBRIDGE - VMOVQ xmmreg,xmmrm64 AVX,SANDYBRIDGE - VMOVQ xmmrm64,xmmreg AVX,SANDYBRIDGE - VMOVQ xmmreg,rm64 AVX,SANDYBRIDGE,LONG - VMOVQ rm64,xmmreg AVX,SANDYBRIDGE,LONG - VMOVD xmmreg,rm32 AVX,SANDYBRIDGE - VMOVD rm32,xmmreg AVX,SANDYBRIDGE - VMOVDDUP xmmreg,xmmrm64 AVX,SANDYBRIDGE - VMOVDDUP ymmreg,ymmrm256 AVX,SANDYBRIDGE - VMOVDQA xmmreg,xmmrm128 AVX,SANDYBRIDGE - VMOVDQA xmmrm128,xmmreg AVX,SANDYBRIDGE - VMOVQQA ymmreg,ymmrm256 AVX,SANDYBRIDGE - VMOVQQA ymmrm256,ymmreg AVX,SANDYBRIDGE - VMOVDQA ymmreg,ymmrm AVX,SANDYBRIDGE - VMOVDQA ymmrm256,ymmreg AVX,SANDYBRIDGE - VMOVDQU xmmreg,xmmrm128 AVX,SANDYBRIDGE - VMOVDQU xmmrm128,xmmreg AVX,SANDYBRIDGE - VMOVQQU ymmreg,ymmrm256 AVX,SANDYBRIDGE - VMOVQQU ymmrm256,ymmreg AVX,SANDYBRIDGE - VMOVDQU ymmreg,ymmrm256 AVX,SANDYBRIDGE - VMOVDQU ymmrm256,ymmreg AVX,SANDYBRIDGE - VMOVHLPS xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE - VMOVHPD xmmreg,xmmreg*,mem64 AVX,SANDYBRIDGE - VMOVHPD mem64,xmmreg AVX,SANDYBRIDGE - VMOVHPS xmmreg,xmmreg*,mem64 AVX,SANDYBRIDGE - VMOVHPS mem64,xmmreg AVX,SANDYBRIDGE - VMOVLHPS xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE - VMOVLPD xmmreg,xmmreg*,mem64 AVX,SANDYBRIDGE - VMOVLPD mem64,xmmreg AVX,SANDYBRIDGE - VMOVLPS xmmreg,xmmreg*,mem64 AVX,SANDYBRIDGE - VMOVLPS mem64,xmmreg AVX,SANDYBRIDGE - VMOVMSKPD reg64,xmmreg AVX,SANDYBRIDGE,LONG - VMOVMSKPD reg32,xmmreg AVX,SANDYBRIDGE - VMOVMSKPD reg64,ymmreg AVX,SANDYBRIDGE,LONG - VMOVMSKPD reg32,ymmreg AVX,SANDYBRIDGE - VMOVMSKPS reg64,xmmreg AVX,SANDYBRIDGE,LONG - VMOVMSKPS reg32,xmmreg AVX,SANDYBRIDGE - VMOVMSKPS reg64,ymmreg AVX,SANDYBRIDGE,LONG - VMOVMSKPS reg32,ymmreg AVX,SANDYBRIDGE - VMOVNTDQ mem128,xmmreg AVX,SANDYBRIDGE - VMOVNTQQ mem256,ymmreg AVX,SANDYBRIDGE - VMOVNTDQ mem256,ymmreg AVX,SANDYBRIDGE - VMOVNTDQA xmmreg,mem128 AVX,SANDYBRIDGE - VMOVNTPD mem128,xmmreg AVX,SANDYBRIDGE - VMOVNTPD mem256,ymmreg AVX,SANDYBRIDGE - VMOVNTPS mem128,xmmreg AVX,SANDYBRIDGE - VMOVNTPS mem128,ymmreg AVX,SANDYBRIDGE - VMOVSD xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE - VMOVSD xmmreg,mem64 AVX,SANDYBRIDGE - VMOVSD xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE - VMOVSD mem64,xmmreg AVX,SANDYBRIDGE - VMOVSHDUP xmmreg,xmmrm128 AVX,SANDYBRIDGE - VMOVSHDUP ymmreg,ymmrm256 AVX,SANDYBRIDGE - VMOVSLDUP xmmreg,xmmrm128 AVX,SANDYBRIDGE - VMOVSLDUP ymmreg,ymmrm256 AVX,SANDYBRIDGE - VMOVSS xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE - VMOVSS xmmreg,mem64 AVX,SANDYBRIDGE - VMOVSS xmmreg,xmmreg*,xmmreg AVX,SANDYBRIDGE - VMOVSS mem64,xmmreg AVX,SANDYBRIDGE - VMOVUPD xmmreg,xmmrm128 AVX,SANDYBRIDGE - VMOVUPD xmmrm128,xmmreg AVX,SANDYBRIDGE - VMOVUPD ymmreg,ymmrm256 AVX,SANDYBRIDGE - VMOVUPD ymmrm256,ymmreg AVX,SANDYBRIDGE - VMOVUPS xmmreg,xmmrm128 AVX,SANDYBRIDGE - VMOVUPS xmmrm128,xmmreg AVX,SANDYBRIDGE - VMOVUPS ymmreg,ymmrm256 AVX,SANDYBRIDGE - VMOVUPS ymmrm256,ymmreg AVX,SANDYBRIDGE - VMPSADBW xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VMULPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VMULPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VMULPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VMULPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VMULSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VMULSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VORPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VORPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VORPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VORPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VPABSB xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPABSW xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPABSD xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPACKSSWB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPACKSSDW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPACKUSWB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPACKUSDW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPADDB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPADDW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPADDD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPADDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPADDSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPADDSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPADDUSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPADDUSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPALIGNR xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VPAND xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPANDN xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPAVGB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPAVGW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPBLENDVB xmmreg,xmmreg*,xmmrm128,xmmreg AVX,SANDYBRIDGE - VPBLENDW xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VPCMPESTRI xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPCMPESTRM xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPCMPISTRI xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPCMPISTRM xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPCMPEQB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCMPEQW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCMPEQD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCMPEQQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCMPGTB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCMPGTW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCMPGTD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCMPGTQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPERMILPD xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPERMILPD ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE - VPERMILPD xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPERMILPD ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE - VPERMILTD2PD xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE - VPERMILTD2PD xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPERMILTD2PD ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE - VPERMILTD2PD ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE - VPERMILMO2PD xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE - VPERMILMO2PD xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPERMILMO2PD ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE - VPERMILMO2PD ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE - VPERMILMZ2PD xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE - VPERMILMZ2PD xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPERMILMZ2PD ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE - VPERMILMZ2PD ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE - VPERMIL2PD xmmreg,xmmreg,xmmrm128,xmmreg,imm8 AVX,SANDYBRIDGE - VPERMIL2PD xmmreg,xmmreg,xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPERMIL2PD ymmreg,ymmreg,ymmrm256,ymmreg,imm8 AVX,SANDYBRIDGE - VPERMIL2PD ymmreg,ymmreg,ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE - VPERMILPS xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPERMILPS ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE - VPERMILPS xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPERMILPS ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE - VPERMILTD2PS xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE - VPERMILTD2PS xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPERMILTD2PS ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE - VPERMILTD2PS ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE - VPERMILMO2PS xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE - VPERMILMO2PS xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPERMILMO2PS ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE - VPERMILMO2PS ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE - VPERMILMZ2PS xmmreg,xmmreg,xmmrm128,xmmreg AVX,SANDYBRIDGE - VPERMILMZ2PS xmmreg,xmmreg,xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPERMILMZ2PS ymmreg,ymmreg,ymmrm256,ymmreg AVX,SANDYBRIDGE - VPERMILMZ2PS ymmreg,ymmreg,ymmreg,ymmrm256 AVX,SANDYBRIDGE - VPERMIL2PS xmmreg,xmmreg,xmmrm128,xmmreg,imm8 AVX,SANDYBRIDGE - VPERMIL2PS xmmreg,xmmreg,xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPERMIL2PS ymmreg,ymmreg,ymmrm256,ymmreg,imm8 AVX,SANDYBRIDGE - VPERMIL2PS ymmreg,ymmreg,ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE - VPERM2F128 ymmreg,ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE - VPEXTRB reg64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG - VPEXTRB reg32,xmmreg,imm8 AVX,SANDYBRIDGE - VPEXTRB mem8,xmmreg,imm8 AVX,SANDYBRIDGE - VPEXTRW reg64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG - VPEXTRW reg32,xmmreg,imm8 AVX,SANDYBRIDGE - VPEXTRW mem16,xmmreg,imm8 AVX,SANDYBRIDGE - VPEXTRW reg64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG - VPEXTRW reg32,xmmreg,imm8 AVX,SANDYBRIDGE - VPEXTRW mem16,xmmreg,imm8 AVX,SANDYBRIDGE - VPEXTRD reg64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG - VPEXTRD rm32,xmmreg,imm8 AVX,SANDYBRIDGE - VPEXTRQ rm64,xmmreg,imm8 AVX,SANDYBRIDGE,LONG - VPHADDW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPHADDD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPHADDSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPHMINPOSUW xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPHSUBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPHSUBD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPHSUBSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPINSRB xmmreg,xmmreg*,mem8,imm8 AVX,SANDYBRIDGE - VPINSRB xmmreg,xmmreg*,rm8,imm8 AVX,SANDYBRIDGE - VPINSRB xmmreg,xmmreg*,reg32,imm8 AVX,SANDYBRIDGE - VPINSRW xmmreg,xmmreg*,mem16,imm8 AVX,SANDYBRIDGE - VPINSRW xmmreg,xmmreg*,rm16,imm8 AVX,SANDYBRIDGE - VPINSRW xmmreg,xmmreg*,reg32,imm8 AVX,SANDYBRIDGE - VPINSRD xmmreg,xmmreg*,mem32,imm8 AVX,SANDYBRIDGE - VPINSRD xmmreg,xmmreg*,rm32,imm8 AVX,SANDYBRIDGE - VPINSRQ xmmreg,xmmreg*,mem64,imm8 AVX,SANDYBRIDGE,LONG - VPINSRQ xmmreg,xmmreg*,rm64,imm8 AVX,SANDYBRIDGE,LONG - VPMADDWD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMADDUBSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMAXSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMAXSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMAXSD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMAXUB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMAXUW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMAXUD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMINSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMINSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMINSD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMINUB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMINUW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMINUD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMOVMSKB reg64,xmmreg AVX,SANDYBRIDGE,LONG - VPMOVMSKB reg32,xmmreg AVX,SANDYBRIDGE - VPMOVSXBW xmmreg,xmmrm64 AVX,SANDYBRIDGE - VPMOVSXBD xmmreg,xmmrm32 AVX,SANDYBRIDGE - VPMOVSXBQ xmmreg,xmmrm16 AVX,SANDYBRIDGE - VPMOVSXWD xmmreg,xmmrm64 AVX,SANDYBRIDGE - VPMOVSXWQ xmmreg,xmmrm32 AVX,SANDYBRIDGE - VPMOVSXDQ xmmreg,xmmrm64 AVX,SANDYBRIDGE - VPMOVZXBW xmmreg,xmmrm64 AVX,SANDYBRIDGE - VPMOVZXBD xmmreg,xmmrm32 AVX,SANDYBRIDGE - VPMOVZXBQ xmmreg,xmmrm16 AVX,SANDYBRIDGE - VPMOVZXWD xmmreg,xmmrm64 AVX,SANDYBRIDGE - VPMOVZXWQ xmmreg,xmmrm32 AVX,SANDYBRIDGE - VPMOVZXDQ xmmreg,xmmrm64 AVX,SANDYBRIDGE - VPMULHUW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMULHRSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMULHW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMULLW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMULLD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMULUDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPMULDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPOR xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSADBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSHUFB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSHUFD xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPSHUFHW xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPSHUFLW xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VPSIGNB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSIGNW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSIGND xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSLLDQ xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE - VPSRLDQ xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE - VPSLLW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSLLW xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE - VPSLLD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSLLD xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE - VPSLLQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSLLQ xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE - VPSRAW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSRAW xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE - VPSRAD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSRAD xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE - VPSRLW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSRLW xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE - VPSRLD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSRLD xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE - VPSRLQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSRLQ xmmreg,xmmreg*,imm8 AVX,SANDYBRIDGE - VPTEST xmmreg,xmmrm128 AVX,SANDYBRIDGE - VPTEST ymmreg,ymmrm256 AVX,SANDYBRIDGE - VPSUBB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSUBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSUBD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSUBQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSUBSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSUBSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSUBUSB xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPSUBUSW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPUNPCKHBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPUNPCKHWD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPUNPCKHDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPUNPCKHQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPUNPCKLBW xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPUNPCKLWD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPUNPCKLDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPUNPCKLQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPXOR xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VRCPPS xmmreg,xmmrm128 AVX,SANDYBRIDGE - VRCPPS ymmreg,ymmrm256 AVX,SANDYBRIDGE - VRCPSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VRSQRTPS xmmreg,xmmrm128 AVX,SANDYBRIDGE - VRSQRTPS ymmreg,ymmrm256 AVX,SANDYBRIDGE - VRSQRTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VROUNDPD xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VROUNDPD ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE - VROUNDPS xmmreg,xmmrm128,imm8 AVX,SANDYBRIDGE - VROUNDPS ymmreg,ymmrm256,imm8 AVX,SANDYBRIDGE - VROUNDSD xmmreg,xmmreg*,xmmrm64,imm8 AVX,SANDYBRIDGE - VROUNDSS xmmreg,xmmreg*,xmmrm32,imm8 AVX,SANDYBRIDGE - VSHUFPD xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VSHUFPD ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE - VSHUFPS xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - VSHUFPS ymmreg,ymmreg*,ymmrm256,imm8 AVX,SANDYBRIDGE - VSQRTPD xmmreg,xmmrm128 AVX,SANDYBRIDGE - VSQRTPD ymmreg,ymmrm256 AVX,SANDYBRIDGE - VSQRTPS xmmreg,xmmrm128 AVX,SANDYBRIDGE - VSQRTPS ymmreg,ymmrm256 AVX,SANDYBRIDGE - VSQRTSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VSQRTSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VSTMXCSR mem32 AVX,SANDYBRIDGE - VSUBPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VSUBPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VSUBPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VSUBPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VSUBSD xmmreg,xmmreg*,xmmrm64 AVX,SANDYBRIDGE - VSUBSS xmmreg,xmmreg*,xmmrm32 AVX,SANDYBRIDGE - VTESTPS xmmreg,xmmrm128 AVX,SANDYBRIDGE - VTESTPS ymmreg,ymmrm256 AVX,SANDYBRIDGE - VTESTPD xmmreg,xmmrm128 AVX,SANDYBRIDGE - VTESTPD ymmreg,ymmrm256 AVX,SANDYBRIDGE - VUCOMISD xmmreg,xmmrm64 AVX,SANDYBRIDGE - VUCOMISS xmmreg,xmmrm32 AVX,SANDYBRIDGE - VUNPCKHPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VUNPCKHPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VUNPCKHPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VUNPCKHPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VUNPCKLPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VUNPCKLPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VUNPCKLPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VUNPCKLPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VXORPD xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VXORPD ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VXORPS xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VXORPS ymmreg,ymmreg*,ymmrm256 AVX,SANDYBRIDGE - VZEROALL AVX,SANDYBRIDGE - VZEROUPPER AVX,SANDYBRIDGE - -B.1.26 Intel Carry-Less Multiplication instructions (CLMUL) - - PCLMULLQLQDQ xmmreg,xmmrm128 SSE,WESTMERE - PCLMULHQLQDQ xmmreg,xmmrm128 SSE,WESTMERE - PCLMULLQHQDQ xmmreg,xmmrm128 SSE,WESTMERE - PCLMULHQHQDQ xmmreg,xmmrm128 SSE,WESTMERE - PCLMULQDQ xmmreg,xmmrm128,imm8 SSE,WESTMERE - -B.1.27 Intel AVX Carry-Less Multiplication instructions (CLMUL) - - VPCLMULLQLQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCLMULHQLQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCLMULLQHQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCLMULHQHQDQ xmmreg,xmmreg*,xmmrm128 AVX,SANDYBRIDGE - VPCLMULQDQ xmmreg,xmmreg*,xmmrm128,imm8 AVX,SANDYBRIDGE - -B.1.28 Intel Fused Multiply-Add instructions (FMA) - - VFMADD132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADD321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADDSUB321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMADDSUB321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUB321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUB321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMSUBADD321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFMSUBADD321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMADD321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMADD321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB132PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB132PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB132PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB132PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB312PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB312PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB312PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB312PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB213PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB213PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB213PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB213PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB123PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB123PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB123PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB123PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB231PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB231PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB231PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB231PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB321PS xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB321PS ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFNMSUB321PD xmmreg,xmmreg,xmmrm128 FMA,FUTURE - VFNMSUB321PD ymmreg,ymmreg,ymmrm256 FMA,FUTURE - VFMADD132SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMADD132SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMADD312SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMADD312SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMADD213SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMADD213SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMADD123SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMADD123SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMADD231SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMADD231SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMADD321SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMADD321SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMSUB132SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMSUB132SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMSUB312SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMSUB312SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMSUB213SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMSUB213SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMSUB123SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMSUB123SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMSUB231SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMSUB231SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFMSUB321SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFMSUB321SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMADD132SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMADD132SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMADD312SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMADD312SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMADD213SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMADD213SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMADD123SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMADD123SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMADD231SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMADD231SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMADD321SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMADD321SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMSUB132SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMSUB132SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMSUB312SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMSUB312SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMSUB213SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMSUB213SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMSUB123SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMSUB123SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMSUB231SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMSUB231SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - VFNMSUB321SS xmmreg,xmmreg,xmmrm32 FMA,FUTURE - VFNMSUB321SD xmmreg,xmmreg,xmmrm64 FMA,FUTURE - -B.1.29 VIA (Centaur) security instructions - - XSTORE PENT,CYRIX - XCRYPTECB PENT,CYRIX - XCRYPTCBC PENT,CYRIX - XCRYPTCTR PENT,CYRIX - XCRYPTCFB PENT,CYRIX - XCRYPTOFB PENT,CYRIX - MONTMUL PENT,CYRIX - XSHA1 PENT,CYRIX - XSHA256 PENT,CYRIX - -B.1.30 AMD Lightweight Profiling (LWP) instructions - - LLWPCB reg16 AMD - LLWPCB reg32 AMD,386 - LLWPCB reg64 AMD,X64 - SLWPCB reg16 AMD - SLWPCB reg32 AMD,386 - SLWPCB reg64 AMD,X64 - LWPVAL reg16,rm32,imm16 AMD,386 - LWPVAL reg32,rm32,imm32 AMD,386 - LWPVAL reg64,rm32,imm32 AMD,X64 - LWPINS reg16,rm32,imm16 AMD,386 - LWPINS reg32,rm32,imm32 AMD,386 - LWPINS reg64,rm32,imm32 AMD,X64 - -B.1.31 AMD XOP, FMA4 and CVT16 instructions (SSE5) - - VCVTPH2PS xmmreg,xmmrm64*,imm8 AMD,SSE5 - VCVTPH2PS ymmreg,xmmrm128,imm8 AMD,SSE5 - VCVTPH2PS ymmreg,ymmrm128*,imm8 AMD,SSE5 - VCVTPS2PH xmmrm64,xmmreg*,imm8 AMD,SSE5 - VCVTPS2PH xmmrm128,ymmreg,imm8 AMD,SSE5 - VCVTPS2PH ymmrm128,ymmreg*,imm8 AMD,SSE5 - VFMADDPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFMADDPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFMADDPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFMADDPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFMADDPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFMADDPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFMADDPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFMADDPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFMADDSD xmmreg,xmmreg*,xmmrm64,xmmreg AMD,SSE5 - VFMADDSD xmmreg,xmmreg*,xmmreg,xmmrm64 AMD,SSE5 - VFMADDSS xmmreg,xmmreg*,xmmrm32,xmmreg AMD,SSE5 - VFMADDSS xmmreg,xmmreg*,xmmreg,xmmrm32 AMD,SSE5 - VFMADDSUBPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFMADDSUBPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFMADDSUBPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFMADDSUBPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFMADDSUBPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFMADDSUBPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFMADDSUBPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFMADDSUBPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFMSUBADDPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFMSUBADDPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFMSUBADDPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFMSUBADDPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFMSUBADDPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFMSUBADDPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFMSUBADDPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFMSUBADDPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFMSUBPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFMSUBPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFMSUBPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFMSUBPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFMSUBPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFMSUBPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFMSUBPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFMSUBPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFMSUBSD xmmreg,xmmreg*,xmmrm64,xmmreg AMD,SSE5 - VFMSUBSD xmmreg,xmmreg*,xmmreg,xmmrm64 AMD,SSE5 - VFMSUBSS xmmreg,xmmreg*,xmmrm32,xmmreg AMD,SSE5 - VFMSUBSS xmmreg,xmmreg*,xmmreg,xmmrm32 AMD,SSE5 - VFNMADDPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFNMADDPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFNMADDPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFNMADDPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFNMADDPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFNMADDPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFNMADDPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFNMADDPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFNMADDSD xmmreg,xmmreg*,xmmrm64,xmmreg AMD,SSE5 - VFNMADDSD xmmreg,xmmreg*,xmmreg,xmmrm64 AMD,SSE5 - VFNMADDSS xmmreg,xmmreg*,xmmrm32,xmmreg AMD,SSE5 - VFNMADDSS xmmreg,xmmreg*,xmmreg,xmmrm32 AMD,SSE5 - VFNMSUBPD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFNMSUBPD ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFNMSUBPD xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFNMSUBPD ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFNMSUBPS xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VFNMSUBPS ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VFNMSUBPS xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VFNMSUBPS ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VFNMSUBSD xmmreg,xmmreg*,xmmrm64,xmmreg AMD,SSE5 - VFNMSUBSD xmmreg,xmmreg*,xmmreg,xmmrm64 AMD,SSE5 - VFNMSUBSS xmmreg,xmmreg*,xmmrm32,xmmreg AMD,SSE5 - VFNMSUBSS xmmreg,xmmreg*,xmmreg,xmmrm32 AMD,SSE5 - VFRCZPD xmmreg,xmmrm128* AMD,SSE5 - VFRCZPD ymmreg,ymmrm256* AMD,SSE5 - VFRCZPS xmmreg,xmmrm128* AMD,SSE5 - VFRCZPS ymmreg,ymmrm256* AMD,SSE5 - VFRCZSD xmmreg,xmmrm64* AMD,SSE5 - VFRCZSS xmmreg,xmmrm32* AMD,SSE5 - VPCMOV xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPCMOV ymmreg,ymmreg*,ymmrm256,ymmreg AMD,SSE5 - VPCMOV xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VPCMOV ymmreg,ymmreg*,ymmreg,ymmrm256 AMD,SSE5 - VPCOMB xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 - VPCOMD xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 - VPCOMQ xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 - VPCOMUB xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 - VPCOMUD xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 - VPCOMUQ xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 - VPCOMUW xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 - VPCOMW xmmreg,xmmreg*,xmmrm128,imm8 AMD,SSE5 - VPHADDBD xmmreg,xmmrm128* AMD,SSE5 - VPHADDBQ xmmreg,xmmrm128* AMD,SSE5 - VPHADDBW xmmreg,xmmrm128* AMD,SSE5 - VPHADDDQ xmmreg,xmmrm128* AMD,SSE5 - VPHADDUBD xmmreg,xmmrm128* AMD,SSE5 - VPHADDUBQ xmmreg,xmmrm128* AMD,SSE5 - VPHADDUBW xmmreg,xmmrm128* AMD,SSE5 - VPHADDUDQ xmmreg,xmmrm128* AMD,SSE5 - VPHADDUWD xmmreg,xmmrm128* AMD,SSE5 - VPHADDUWQ xmmreg,xmmrm128* AMD,SSE5 - VPHADDWD xmmreg,xmmrm128* AMD,SSE5 - VPHADDWQ xmmreg,xmmrm128* AMD,SSE5 - VPHSUBBW xmmreg,xmmrm128* AMD,SSE5 - VPHSUBDQ xmmreg,xmmrm128* AMD,SSE5 - VPHSUBWD xmmreg,xmmrm128* AMD,SSE5 - VPMACSDD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMACSDQH xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMACSDQL xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMACSSDD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMACSSDQH xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMACSSDQL xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMACSSWD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMACSSWW xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMACSWD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMACSWW xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMADCSSWD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPMADCSWD xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPPERM xmmreg,xmmreg*,xmmreg,xmmrm128 AMD,SSE5 - VPPERM xmmreg,xmmreg*,xmmrm128,xmmreg AMD,SSE5 - VPROTB xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPROTB xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPROTB xmmreg,xmmrm128*,imm8 AMD,SSE5 - VPROTD xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPROTD xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPROTD xmmreg,xmmrm128*,imm8 AMD,SSE5 - VPROTQ xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPROTQ xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPROTQ xmmreg,xmmrm128*,imm8 AMD,SSE5 - VPROTW xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPROTW xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPROTW xmmreg,xmmrm128*,imm8 AMD,SSE5 - VPSHAB xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPSHAB xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPSHAD xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPSHAD xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPSHAQ xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPSHAQ xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPSHAW xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPSHAW xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPSHLB xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPSHLB xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPSHLD xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPSHLD xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPSHLQ xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPSHLQ xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - VPSHLW xmmreg,xmmrm128*,xmmreg AMD,SSE5 - VPSHLW xmmreg,xmmreg*,xmmrm128 AMD,SSE5 - -B.1.32 Systematic names for the hinting nop instructions - - HINT_NOP0 rm16 P6,UNDOC - HINT_NOP0 rm32 P6,UNDOC - HINT_NOP0 rm64 X64,UNDOC - HINT_NOP1 rm16 P6,UNDOC - HINT_NOP1 rm32 P6,UNDOC - HINT_NOP1 rm64 X64,UNDOC - HINT_NOP2 rm16 P6,UNDOC - HINT_NOP2 rm32 P6,UNDOC - HINT_NOP2 rm64 X64,UNDOC - HINT_NOP3 rm16 P6,UNDOC - HINT_NOP3 rm32 P6,UNDOC - HINT_NOP3 rm64 X64,UNDOC - HINT_NOP4 rm16 P6,UNDOC - HINT_NOP4 rm32 P6,UNDOC - HINT_NOP4 rm64 X64,UNDOC - HINT_NOP5 rm16 P6,UNDOC - HINT_NOP5 rm32 P6,UNDOC - HINT_NOP5 rm64 X64,UNDOC - HINT_NOP6 rm16 P6,UNDOC - HINT_NOP6 rm32 P6,UNDOC - HINT_NOP6 rm64 X64,UNDOC - HINT_NOP7 rm16 P6,UNDOC - HINT_NOP7 rm32 P6,UNDOC - HINT_NOP7 rm64 X64,UNDOC - HINT_NOP8 rm16 P6,UNDOC - HINT_NOP8 rm32 P6,UNDOC - HINT_NOP8 rm64 X64,UNDOC - HINT_NOP9 rm16 P6,UNDOC - HINT_NOP9 rm32 P6,UNDOC - HINT_NOP9 rm64 X64,UNDOC - HINT_NOP10 rm16 P6,UNDOC - HINT_NOP10 rm32 P6,UNDOC - HINT_NOP10 rm64 X64,UNDOC - HINT_NOP11 rm16 P6,UNDOC - HINT_NOP11 rm32 P6,UNDOC - HINT_NOP11 rm64 X64,UNDOC - HINT_NOP12 rm16 P6,UNDOC - HINT_NOP12 rm32 P6,UNDOC - HINT_NOP12 rm64 X64,UNDOC - HINT_NOP13 rm16 P6,UNDOC - HINT_NOP13 rm32 P6,UNDOC - HINT_NOP13 rm64 X64,UNDOC - HINT_NOP14 rm16 P6,UNDOC - HINT_NOP14 rm32 P6,UNDOC - HINT_NOP14 rm64 X64,UNDOC - HINT_NOP15 rm16 P6,UNDOC - HINT_NOP15 rm32 P6,UNDOC - HINT_NOP15 rm64 X64,UNDOC - HINT_NOP16 rm16 P6,UNDOC - HINT_NOP16 rm32 P6,UNDOC - HINT_NOP16 rm64 X64,UNDOC - HINT_NOP17 rm16 P6,UNDOC - HINT_NOP17 rm32 P6,UNDOC - HINT_NOP17 rm64 X64,UNDOC - HINT_NOP18 rm16 P6,UNDOC - HINT_NOP18 rm32 P6,UNDOC - HINT_NOP18 rm64 X64,UNDOC - HINT_NOP19 rm16 P6,UNDOC - HINT_NOP19 rm32 P6,UNDOC - HINT_NOP19 rm64 X64,UNDOC - HINT_NOP20 rm16 P6,UNDOC - HINT_NOP20 rm32 P6,UNDOC - HINT_NOP20 rm64 X64,UNDOC - HINT_NOP21 rm16 P6,UNDOC - HINT_NOP21 rm32 P6,UNDOC - HINT_NOP21 rm64 X64,UNDOC - HINT_NOP22 rm16 P6,UNDOC - HINT_NOP22 rm32 P6,UNDOC - HINT_NOP22 rm64 X64,UNDOC - HINT_NOP23 rm16 P6,UNDOC - HINT_NOP23 rm32 P6,UNDOC - HINT_NOP23 rm64 X64,UNDOC - HINT_NOP24 rm16 P6,UNDOC - HINT_NOP24 rm32 P6,UNDOC - HINT_NOP24 rm64 X64,UNDOC - HINT_NOP25 rm16 P6,UNDOC - HINT_NOP25 rm32 P6,UNDOC - HINT_NOP25 rm64 X64,UNDOC - HINT_NOP26 rm16 P6,UNDOC - HINT_NOP26 rm32 P6,UNDOC - HINT_NOP26 rm64 X64,UNDOC - HINT_NOP27 rm16 P6,UNDOC - HINT_NOP27 rm32 P6,UNDOC - HINT_NOP27 rm64 X64,UNDOC - HINT_NOP28 rm16 P6,UNDOC - HINT_NOP28 rm32 P6,UNDOC - HINT_NOP28 rm64 X64,UNDOC - HINT_NOP29 rm16 P6,UNDOC - HINT_NOP29 rm32 P6,UNDOC - HINT_NOP29 rm64 X64,UNDOC - HINT_NOP30 rm16 P6,UNDOC - HINT_NOP30 rm32 P6,UNDOC - HINT_NOP30 rm64 X64,UNDOC - HINT_NOP31 rm16 P6,UNDOC - HINT_NOP31 rm32 P6,UNDOC - HINT_NOP31 rm64 X64,UNDOC - HINT_NOP32 rm16 P6,UNDOC - HINT_NOP32 rm32 P6,UNDOC - HINT_NOP32 rm64 X64,UNDOC - HINT_NOP33 rm16 P6,UNDOC - HINT_NOP33 rm32 P6,UNDOC - HINT_NOP33 rm64 X64,UNDOC - HINT_NOP34 rm16 P6,UNDOC - HINT_NOP34 rm32 P6,UNDOC - HINT_NOP34 rm64 X64,UNDOC - HINT_NOP35 rm16 P6,UNDOC - HINT_NOP35 rm32 P6,UNDOC - HINT_NOP35 rm64 X64,UNDOC - HINT_NOP36 rm16 P6,UNDOC - HINT_NOP36 rm32 P6,UNDOC - HINT_NOP36 rm64 X64,UNDOC - HINT_NOP37 rm16 P6,UNDOC - HINT_NOP37 rm32 P6,UNDOC - HINT_NOP37 rm64 X64,UNDOC - HINT_NOP38 rm16 P6,UNDOC - HINT_NOP38 rm32 P6,UNDOC - HINT_NOP38 rm64 X64,UNDOC - HINT_NOP39 rm16 P6,UNDOC - HINT_NOP39 rm32 P6,UNDOC - HINT_NOP39 rm64 X64,UNDOC - HINT_NOP40 rm16 P6,UNDOC - HINT_NOP40 rm32 P6,UNDOC - HINT_NOP40 rm64 X64,UNDOC - HINT_NOP41 rm16 P6,UNDOC - HINT_NOP41 rm32 P6,UNDOC - HINT_NOP41 rm64 X64,UNDOC - HINT_NOP42 rm16 P6,UNDOC - HINT_NOP42 rm32 P6,UNDOC - HINT_NOP42 rm64 X64,UNDOC - HINT_NOP43 rm16 P6,UNDOC - HINT_NOP43 rm32 P6,UNDOC - HINT_NOP43 rm64 X64,UNDOC - HINT_NOP44 rm16 P6,UNDOC - HINT_NOP44 rm32 P6,UNDOC - HINT_NOP44 rm64 X64,UNDOC - HINT_NOP45 rm16 P6,UNDOC - HINT_NOP45 rm32 P6,UNDOC - HINT_NOP45 rm64 X64,UNDOC - HINT_NOP46 rm16 P6,UNDOC - HINT_NOP46 rm32 P6,UNDOC - HINT_NOP46 rm64 X64,UNDOC - HINT_NOP47 rm16 P6,UNDOC - HINT_NOP47 rm32 P6,UNDOC - HINT_NOP47 rm64 X64,UNDOC - HINT_NOP48 rm16 P6,UNDOC - HINT_NOP48 rm32 P6,UNDOC - HINT_NOP48 rm64 X64,UNDOC - HINT_NOP49 rm16 P6,UNDOC - HINT_NOP49 rm32 P6,UNDOC - HINT_NOP49 rm64 X64,UNDOC - HINT_NOP50 rm16 P6,UNDOC - HINT_NOP50 rm32 P6,UNDOC - HINT_NOP50 rm64 X64,UNDOC - HINT_NOP51 rm16 P6,UNDOC - HINT_NOP51 rm32 P6,UNDOC - HINT_NOP51 rm64 X64,UNDOC - HINT_NOP52 rm16 P6,UNDOC - HINT_NOP52 rm32 P6,UNDOC - HINT_NOP52 rm64 X64,UNDOC - HINT_NOP53 rm16 P6,UNDOC - HINT_NOP53 rm32 P6,UNDOC - HINT_NOP53 rm64 X64,UNDOC - HINT_NOP54 rm16 P6,UNDOC - HINT_NOP54 rm32 P6,UNDOC - HINT_NOP54 rm64 X64,UNDOC - HINT_NOP55 rm16 P6,UNDOC - HINT_NOP55 rm32 P6,UNDOC - HINT_NOP55 rm64 X64,UNDOC - HINT_NOP56 rm16 P6,UNDOC - HINT_NOP56 rm32 P6,UNDOC - HINT_NOP56 rm64 X64,UNDOC - HINT_NOP57 rm16 P6,UNDOC - HINT_NOP57 rm32 P6,UNDOC - HINT_NOP57 rm64 X64,UNDOC - HINT_NOP58 rm16 P6,UNDOC - HINT_NOP58 rm32 P6,UNDOC - HINT_NOP58 rm64 X64,UNDOC - HINT_NOP59 rm16 P6,UNDOC - HINT_NOP59 rm32 P6,UNDOC - HINT_NOP59 rm64 X64,UNDOC - HINT_NOP60 rm16 P6,UNDOC - HINT_NOP60 rm32 P6,UNDOC - HINT_NOP60 rm64 X64,UNDOC - HINT_NOP61 rm16 P6,UNDOC - HINT_NOP61 rm32 P6,UNDOC - HINT_NOP61 rm64 X64,UNDOC - HINT_NOP62 rm16 P6,UNDOC - HINT_NOP62 rm32 P6,UNDOC - HINT_NOP62 rm64 X64,UNDOC - HINT_NOP63 rm16 P6,UNDOC - HINT_NOP63 rm32 P6,UNDOC - HINT_NOP63 rm64 X64,UNDOC - -Appendix C: NASM Version History --------------------------------- - - C.1 NASM 2 Series - - The NASM 2 series support x86-64, and is the production version of - NASM since 2007. - - C.1.1 Version 2.08 - - (*) A number of enhancements/fixes in macros area. - - (*) Support for arbitrarily terminating macro expansions - `%exitmacro'. See section 4.3.12. - - (*) Support for recursive macro expansion `%rmacro/irmacro'. See - section 4.3.1. - - (*) Support for converting strings to tokens. See section 4.1.9. - - (*) Fuzzy operand size logic introduced. - - (*) Fix COFF stack overrun on too long export identifiers. - - (*) Fix Macho-O alignment bug. - - (*) Fix crashes with -fwin32 on file with many exports. - - (*) Fix stack overrun for too long [DEBUG id]. - - (*) Fix incorrect sbyte usage in IMUL (hit only if optimization flag - passed). - - (*) Append ending token for `.stabs' records in the ELF output - format. - - (*) New NSIS script which uses ModernUI and MultiUser approach. - - (*) Visual Studio 2008 NASM integration (rules file). - - (*) Warn a user if a constant is too long (and as result will be - stripped). - - (*) The obsoleted pre-XOP AMD SSE5 instruction set which was never - actualized was removed. - - (*) Fix stack overrun on too long error file name passed from the - command line. - - (*) Bind symbols to the .text section by default (ie in case if - SECTION directive was omitted) in the ELF output format. - - (*) Fix sync points array index wrapping. - - (*) A few fixes for FMA4 and XOP instruction templates. - - (*) Add AMD Lightweight Profiling (LWP) instructions. - - C.1.2 Version 2.07 - - (*) NASM is now under the 2-clause BSD license. See section 1.1.2. - - (*) Fix the section type for the `.strtab' section in the `elf64' - output format. - - (*) Fix the handling of `COMMON' directives in the `obj' output - format. - - (*) New `ith' and `srec' output formats; these are variants of the - `bin' output format which output Intel hex and Motorola S- - records, respectively. See section 7.2 and section 7.3. - - (*) `rdf2ihx' replaced with an enhanced `rdf2bin', which can output - binary, COM, Intel hex or Motorola S-records. - - (*) The Windows installer now puts the NASM directory first in the - `PATH' of the "NASM Shell". - - (*) Revert the early expansion behavior of `%+' to pre-2.06 - behavior: `%+' is only expanded late. - - (*) Yet another Mach-O alignment fix. - - (*) Don't delete the list file on errors. Also, include error and - warning information in the list file. - - (*) Support for 64-bit Mach-O output, see section 7.8. - - (*) Fix assert failure on certain operations that involve strings - with high-bit bytes. - - C.1.3 Version 2.06 - - (*) This release is dedicated to the memory of Charles A. Crayne, - long time NASM developer as well as moderator of - `comp.lang.asm.x86' and author of the book _Serious Assembler_. - We miss you, Chuck. - - (*) Support for indirect macro expansion (`%[...]'). See section - 4.1.3. - - (*) `%pop' can now take an argument, see section 4.7.1. - - (*) The argument to `%use' is no longer macro-expanded. Use `%[...]' - if macro expansion is desired. - - (*) Support for thread-local storage in ELF32 and ELF64. See section - 7.9.4. - - (*) Fix crash on `%ifmacro' without an argument. - - (*) Correct the arguments to the `POPCNT' instruction. - - (*) Fix section alignment in the Mach-O format. - - (*) Update AVX support to version 5 of the Intel specification. - - (*) Fix the handling of accesses to context-local macros from higher - levels in the context stack. - - (*) Treat `WAIT' as a prefix rather than as an instruction, thereby - allowing constructs like `O16 FSAVE' to work correctly. - - (*) Support for structures with a non-zero base offset. See section - 4.11.10. - - (*) Correctly handle preprocessor token concatenation (see section - 4.3.8) involving floating-point numbers. - - (*) The `PINSR' series of instructions have been corrected and - rationalized. - - (*) Removed AMD SSE5, replaced with the new XOP/FMA4/CVT16 (rev - 3.03) spec. - - (*) The ELF backends no longer automatically generate a `.comment' - section. - - (*) Add additional "well-known" ELF sections with default - attributes. See section 7.9.2. - - C.1.4 Version 2.05.01 - - (*) Fix the `-w'/`-W' option parsing, which was broken in NASM 2.05. - - C.1.5 Version 2.05 - - (*) Fix redundant REX.W prefix on `JMP reg64'. - - (*) Make the behaviour of `-O0' match NASM 0.98 legacy behavior. See - section 2.1.22. - - (*) `-w-user' can be used to suppress the output of `%warning' - directives. See section 2.1.24. - - (*) Fix bug where `ALIGN' would issue a full alignment datum instead - of zero bytes. - - (*) Fix offsets in list files. - - (*) Fix `%include' inside multi-line macros or loops. - - (*) Fix error where NASM would generate a spurious warning on valid - optimizations of immediate values. - - (*) Fix arguments to a number of the `CVT' SSE instructions. - - (*) Fix RIP-relative offsets when the instruction carries an - immediate. - - (*) Massive overhaul of the ELF64 backend for spec compliance. - - (*) Fix the Geode `PFRCPV' and `PFRSQRTV' instruction. - - (*) Fix the SSE 4.2 `CRC32' instruction. - - C.1.6 Version 2.04 - - (*) Sanitize macro handing in the `%error' directive. - - (*) New `%warning' directive to issue user-controlled warnings. - - (*) `%error' directives are now deferred to the final assembly - phase. - - (*) New `%fatal' directive to immediately terminate assembly. - - (*) New `%strcat' directive to join quoted strings together. - - (*) New `%use' macro directive to support standard macro directives. - See section 4.6.4. - - (*) Excess default parameters to `%macro' now issues a warning by - default. See section 4.3. - - (*) Fix `%ifn' and `%elifn'. - - (*) Fix nested `%else' clauses. - - (*) Correct the handling of nested `%rep's. - - (*) New `%unmacro' directive to undeclare a multi-line macro. See - section 4.3.11. - - (*) Builtin macro `__PASS__' which expands to the current assembly - pass. See section 4.11.9. - - (*) `__utf16__' and `__utf32__' operators to generate UTF-16 and - UTF-32 strings. See section 3.4.5. - - (*) Fix bug in case-insensitive matching when compiled on platforms - that don't use the `configure' script. Of the official release - binaries, that only affected the OS/2 binary. - - (*) Support for x87 packed BCD constants. See section 3.4.7. - - (*) Correct the `LTR' and `SLDT' instructions in 64-bit mode. - - (*) Fix unnecessary REX.W prefix on indirect jumps in 64-bit mode. - - (*) Add AVX versions of the AES instructions (`VAES'...). - - (*) Fix the 256-bit FMA instructions. - - (*) Add 256-bit AVX stores per the latest AVX spec. - - (*) VIA XCRYPT instructions can now be written either with or - without `REP', apparently different versions of the VIA spec - wrote them differently. - - (*) Add missing 64-bit `MOVNTI' instruction. - - (*) Fix the operand size of `VMREAD' and `VMWRITE'. - - (*) Numerous bug fixes, especially to the AES, AVX and VTX - instructions. - - (*) The optimizer now always runs until it converges. It also runs - even when disabled, but doesn't optimize. This allows most - forward references to be resolved properly. - - (*) `%push' no longer needs a context identifier; omitting the - context identifier results in an anonymous context. - - C.1.7 Version 2.03.01 - - (*) Fix buffer overflow in the listing module. - - (*) Fix the handling of hexadecimal escape codes in `...` strings. - - (*) The Postscript/PDF documentation has been reformatted. - - (*) The `-F' option now implies `-g'. - - C.1.8 Version 2.03 - - (*) Add support for Intel AVX, CLMUL and FMA instructions, including - YMM registers. - - (*) `dy', `resy' and `yword' for 32-byte operands. - - (*) Fix some SSE5 instructions. - - (*) Intel `INVEPT', `INVVPID' and `MOVBE' instructions. - - (*) Fix checking for critical expressions when the optimizer is - enabled. - - (*) Support the DWARF debugging format for ELF targets. - - (*) Fix optimizations of signed bytes. - - (*) Fix operation on bigendian machines. - - (*) Fix buffer overflow in the preprocessor. - - (*) `SAFESEH' support for Win32, `IMAGEREL' for Win64 (SEH). - - (*) `%?' and `%??' to refer to the name of a macro itself. In - particular, `%idefine keyword $%?' can be used to make a keyword - "disappear". - - (*) New options for dependency generation: `-MD', `-MF', `-MP', - `-MT', `-MQ'. - - (*) New preprocessor directives `%pathsearch' and `%depend'; INCBIN - reimplemented as a macro. - - (*) `%include' now resolves macros in a sane manner. - - (*) `%substr' can now be used to get other than one-character - substrings. - - (*) New type of character/string constants, using backquotes - (``...`'), which support C-style escape sequences. - - (*) `%defstr' and `%idefstr' to stringize macro definitions before - creation. - - (*) Fix forward references used in `EQU' statements. - - C.1.9 Version 2.02 - - (*) Additional fixes for MMX operands with explicit `qword', as well - as (hopefully) SSE operands with `oword'. - - (*) Fix handling of truncated strings with `DO'. - - (*) Fix segfaults due to memory overwrites when floating-point - constants were used. - - (*) Fix segfaults due to missing include files. - - (*) Fix OpenWatcom Makefiles for DOS and OS/2. - - (*) Add autogenerated instruction list back into the documentation. - - (*) ELF: Fix segfault when generating stabs, and no symbols have - been defined. - - (*) ELF: Experimental support for DWARF debugging information. - - (*) New compile date and time standard macros. - - (*) `%ifnum' now returns true for negative numbers. - - (*) New `%iftoken' test for a single token. - - (*) New `%ifempty' test for empty expansion. - - (*) Add support for the `XSAVE' instruction group. - - (*) Makefile for Netware/gcc. - - (*) Fix issue with some warnings getting emitted way too many times. - - (*) Autogenerated instruction list added to the documentation. - -C.1.10 Version 2.01 - - (*) Fix the handling of MMX registers with explicit `qword' tags on - memory (broken in 2.00 due to 64-bit changes.) - - (*) Fix the PREFETCH instructions. - - (*) Fix the documentation. - - (*) Fix debugging info when using `-f elf' (backwards compatibility - alias for `-f elf32'). - - (*) Man pages for rdoff tools (from the Debian project.) - - (*) ELF: handle large numbers of sections. - - (*) Fix corrupt output when the optimizer runs out of passes. - -C.1.11 Version 2.00 - - (*) Added c99 data-type compliance. - - (*) Added general x86-64 support. - - (*) Added win64 (x86-64 COFF) output format. - - (*) Added `__BITS__' standard macro. - - (*) Renamed the `elf' output format to `elf32' for clarity. - - (*) Added `elf64' and `macho' (MacOS X) output formats. - - (*) Added Numeric constants in `dq' directive. - - (*) Added `oword', `do' and `reso' pseudo operands. - - (*) Allow underscores in numbers. - - (*) Added 8-, 16- and 128-bit floating-point formats. - - (*) Added binary, octal and hexadecimal floating-point. - - (*) Correct the generation of floating-point constants. - - (*) Added floating-point option control. - - (*) Added Infinity and NaN floating point support. - - (*) Added ELF Symbol Visibility support. - - (*) Added setting OSABI value in ELF header directive. - - (*) Added Generate Makefile Dependencies option. - - (*) Added Unlimited Optimization Passes option. - - (*) Added `%IFN' and `%ELIFN' support. - - (*) Added Logical Negation Operator. - - (*) Enhanced Stack Relative Preprocessor Directives. - - (*) Enhanced ELF Debug Formats. - - (*) Enhanced Send Errors to a File option. - - (*) Added SSSE3, SSE4.1, SSE4.2, SSE5 support. - - (*) Added a large number of additional instructions. - - (*) Significant performance improvements. - - (*) `-w+warning' and `-w-warning' can now be written as -Wwarning - and -Wno-warning, respectively. See section 2.1.24. - - (*) Add `-w+error' to treat warnings as errors. See section 2.1.24. - - (*) Add `-w+all' and `-w-all' to enable or disable all suppressible - warnings. See section 2.1.24. - - C.2 NASM 0.98 Series - - The 0.98 series was the production versions of NASM from 1999 to - 2007. - - C.2.1 Version 0.98.39 - - (*) fix buffer overflow - - (*) fix outas86's `.bss' handling - - (*) "make spotless" no longer deletes config.h.in. - - (*) `%(el)if(n)idn' insensitivity to string quotes difference - (#809300). - - (*) (nasm.c)`__OUTPUT_FORMAT__' changed to string value instead of - symbol. - - C.2.2 Version 0.98.38 - - (*) Add Makefile for 16-bit DOS binaries under OpenWatcom, and - modify `mkdep.pl' to be able to generate completely pathless - dependencies, as required by OpenWatcom wmake (it supports path - searches, but not explicit paths.) - - (*) Fix the `STR' instruction. - - (*) Fix the ELF output format, which was broken under certain - circumstances due to the addition of stabs support. - - (*) Quick-fix Borland format debug-info for `-f obj' - - (*) Fix for `%rep' with no arguments (#560568) - - (*) Fix concatenation of preprocessor function call (#794686) - - (*) Fix long label causes coredump (#677841) - - (*) Use autoheader as well as autoconf to keep configure from - generating ridiculously long command lines. - - (*) Make sure that all of the formats which support debugging output - actually will suppress debugging output when `-g' not specified. - - C.2.3 Version 0.98.37 - - (*) Paths given in `-I' switch searched for `incbin'-ed as well as - `%include'-ed files. - - (*) Added stabs debugging for the ELF output format, patch from - Martin Wawro. - - (*) Fix `output/outbin.c' to allow origin > 80000000h. - - (*) Make `-U' switch work. - - (*) Fix the use of relative offsets with explicit prefixes, e.g. - `a32 loop foo'. - - (*) Remove `backslash()'. - - (*) Fix the `SMSW' and `SLDT' instructions. - - (*) `-O2' and `-O3' are no longer aliases for `-O10' and `-O15'. If - you mean the latter, please say so! :) - - C.2.4 Version 0.98.36 - - (*) Update rdoff - librarian/archiver - common rec - docs! - - (*) Fix signed/unsigned problems. - - (*) Fix `JMP FAR label' and `CALL FAR label'. - - (*) Add new multisection support - map files - fix align bug - - (*) Fix sysexit, movhps/movlps reg,reg bugs in insns.dat - - (*) `Q' or `O' suffixes indicate octal - - (*) Support Prescott new instructions (PNI). - - (*) Cyrix `XSTORE' instruction. - - C.2.5 Version 0.98.35 - - (*) Fix build failure on 16-bit DOS (Makefile.bc3 workaround for - compiler bug.) - - (*) Fix dependencies and compiler warnings. - - (*) Add "const" in a number of places. - - (*) Add -X option to specify error reporting format (use -Xvc to - integrate with Microsoft Visual Studio.) - - (*) Minor changes for code legibility. - - (*) Drop use of tmpnam() in rdoff (security fix.) - - C.2.6 Version 0.98.34 - - (*) Correct additional address-size vs. operand-size confusions. - - (*) Generate dependencies for all Makefiles automatically. - - (*) Add support for unimplemented (but theoretically available) - registers such as tr0 and cr5. Segment registers 6 and 7 are - called segr6 and segr7 for the operations which they can be - represented. - - (*) Correct some disassembler bugs related to redundant address-size - prefixes. Some work still remains in this area. - - (*) Correctly generate an error for things like "SEG eax". - - (*) Add the JMPE instruction, enabled by "CPU IA64". - - (*) Correct compilation on newer gcc/glibc platforms. - - (*) Issue an error on things like "jmp far eax". - - C.2.7 Version 0.98.33 - - (*) New __NASM_PATCHLEVEL__ and __NASM_VERSION_ID__ standard macros - to round out the version-query macros. version.pl now - understands X.YYplWW or X.YY.ZZplWW as a version number, - equivalent to X.YY.ZZ.WW (or X.YY.0.WW, as appropriate). - - (*) New keyword "strict" to disable the optimization of specific - operands. - - (*) Fix the handing of size overrides with JMP instructions - (instructions such as "jmp dword foo".) - - (*) Fix the handling of "ABSOLUTE label", where "label" points into - a relocatable segment. - - (*) Fix OBJ output format with lots of externs. - - (*) More documentation updates. - - (*) Add -Ov option to get verbose information about optimizations. - - (*) Undo a braindead change which broke `%elif' directives. - - (*) Makefile updates. - - C.2.8 Version 0.98.32 - - (*) Fix NASM crashing when `%macro' directives were left - unterminated. - - (*) Lots of documentation updates. - - (*) Complete rewrite of the PostScript/PDF documentation generator. - - (*) The MS Visual C++ Makefile was updated and corrected. - - (*) Recognize .rodata as a standard section name in ELF. - - (*) Fix some obsolete Perl4-isms in Perl scripts. - - (*) Fix configure.in to work with autoconf 2.5x. - - (*) Fix a couple of "make cleaner" misses. - - (*) Make the normal "./configure && make" work with Cygwin. - - C.2.9 Version 0.98.31 - - (*) Correctly build in a separate object directory again. - - (*) Derive all references to the version number from the version - file. - - (*) New standard macros __NASM_SUBMINOR__ and __NASM_VER__ macros. - - (*) Lots of Makefile updates and bug fixes. - - (*) New `%ifmacro' directive to test for multiline macros. - - (*) Documentation updates. - - (*) Fixes for 16-bit OBJ format output. - - (*) Changed the NASM environment variable to NASMENV. - -C.2.10 Version 0.98.30 - - (*) Changed doc files a lot: completely removed old READMExx and - Wishlist files, incorporating all information in CHANGES and - TODO. - - (*) I waited a long time to rename zoutieee.c to (original) - outieee.c - - (*) moved all output modules to output/ subdirectory. - - (*) Added 'make strip' target to strip debug info from nasm & - ndisasm. - - (*) Added INSTALL file with installation instructions. - - (*) Added -v option description to nasm man. - - (*) Added dist makefile target to produce source distributions. - - (*) 16-bit support for ELF output format (GNU extension, but - useful.) - -C.2.11 Version 0.98.28 - - (*) Fastcooked this for Debian's Woody release: Frank applied the - INCBIN bug patch to 0.98.25alt and called it 0.98.28 to not - confuse poor little apt-get. - -C.2.12 Version 0.98.26 - - (*) Reorganised files even better from 0.98.25alt - -C.2.13 Version 0.98.25alt - - (*) Prettified the source tree. Moved files to more reasonable - places. - - (*) Added findleak.pl script to misc/ directory. - - (*) Attempted to fix doc. - -C.2.14 Version 0.98.25 - - (*) Line continuation character `\'. - - (*) Docs inadvertantly reverted - "dos packaging". - -C.2.15 Version 0.98.24p1 - - (*) FIXME: Someone, document this please. - -C.2.16 Version 0.98.24 - - (*) Documentation - Ndisasm doc added to Nasm.doc. - -C.2.17 Version 0.98.23 - - (*) Attempted to remove rdoff version1 - - (*) Lino Mastrodomenico's patches to preproc.c (%$$ bug?). - -C.2.18 Version 0.98.22 - - (*) Update rdoff2 - attempt to remove v1. - -C.2.19 Version 0.98.21 - - (*) Optimization fixes. - -C.2.20 Version 0.98.20 - - (*) Optimization fixes. - -C.2.21 Version 0.98.19 - - (*) H. J. Lu's patch back out. - -C.2.22 Version 0.98.18 - - (*) Added ".rdata" to "-f win32". - -C.2.23 Version 0.98.17 - - (*) H. J. Lu's "bogus elf" patch. (Red Hat problem?) - -C.2.24 Version 0.98.16 - - (*) Fix whitespace before "[section ..." bug. - -C.2.25 Version 0.98.15 - - (*) Rdoff changes (?). - - (*) Fix fixes to memory leaks. - -C.2.26 Version 0.98.14 - - (*) Fix memory leaks. - -C.2.27 Version 0.98.13 - - (*) There was no 0.98.13 - -C.2.28 Version 0.98.12 - - (*) Update optimization (new function of "-O1") - - (*) Changes to test/bintest.asm (?). - -C.2.29 Version 0.98.11 - - (*) Optimization changes. - - (*) Ndisasm fixed. - -C.2.30 Version 0.98.10 - - (*) There was no 0.98.10 - -C.2.31 Version 0.98.09 - - (*) Add multiple sections support to "-f bin". - - (*) Changed GLOBAL_TEMP_BASE in outelf.c from 6 to 15. - - (*) Add "-v" as an alias to the "-r" switch. - - (*) Remove "#ifdef" from Tasm compatibility options. - - (*) Remove redundant size-overrides on "mov ds, ex", etc. - - (*) Fixes to SSE2, other insns.dat (?). - - (*) Enable uppercase "I" and "P" switches. - - (*) Case insinsitive "seg" and "wrt". - - (*) Update install.sh (?). - - (*) Allocate tokens in blocks. - - (*) Improve "invalid effective address" messages. - -C.2.32 Version 0.98.08 - - (*) Add "`%strlen'" and "`%substr'" macro operators - - (*) Fixed broken c16.mac. - - (*) Unterminated string error reported. - - (*) Fixed bugs as per 0.98bf - -C.2.33 Version 0.98.09b with John Coffman patches released 28-Oct-2001 - - Changes from 0.98.07 release to 98.09b as of 28-Oct-2001 - - (*) More closely compatible with 0.98 when -O0 is implied or - specified. Not strictly identical, since backward branches in - range of short offsets are recognized, and signed byte values - with no explicit size specification will be assembled as a - single byte. - - (*) More forgiving with the PUSH instruction. 0.98 requires a size - to be specified always. 0.98.09b will imply the size from the - current BITS setting (16 or 32). - - (*) Changed definition of the optimization flag: - - -O0 strict two-pass assembly, JMP and Jcc are handled more like - 0.98, except that back- ward JMPs are short, if possible. - - -O1 strict two-pass assembly, but forward branches are assembled - with code guaranteed to reach; may produce larger code than -O0, but - will produce successful assembly more often if branch offset sizes - are not specified. - - -O2 multi-pass optimization, minimize branch offsets; also will - minimize signed immed- iate bytes, overriding size specification. - - -O3 like -O2, but more passes taken, if needed - -C.2.34 Version 0.98.07 released 01/28/01 - - (*) Added Stepane Denis' SSE2 instructions to a *working* version of - the code - some earlier versions were based on broken code - - sorry 'bout that. version "0.98.07" - - 01/28/01 - - (*) Cosmetic modifications to nasm.c, nasm.h, AUTHORS, MODIFIED - -C.2.35 Version 0.98.06f released 01/18/01 - - (*) - Add "metalbrain"s jecxz bug fix in insns.dat - alter - nasmdoc.src to match - version "0.98.06f" - -C.2.36 Version 0.98.06e released 01/09/01 - - (*) Removed the "outforms.h" file - it appears to be someone's old - backup of "outform.h". version "0.98.06e" - - 01/09/01 - - (*) fbk - finally added the fix for the "multiple %includes bug", - known since 7/27/99 - reported originally (?) and sent to us by - Austin Lunnen - he reports that John Fine had a fix within the - day. Here it is... - - (*) Nelson Rush resigns from the group. Big thanks to Nelson for his - leadership and enthusiasm in getting these changes incorporated - into Nasm! - - (*) fbk - [list +], [list -] directives - ineptly implemented, - should be re-written or removed, perhaps. - - (*) Brian Raiter / fbk - "elfso bug" fix - applied to aoutb format - as well - testing might be desirable... - - 08/07/00 - - (*) James Seter - -postfix, -prefix command line switches. - - (*) Yuri Zaporogets - rdoff utility changes. - -C.2.37 Version 0.98p1 - - (*) GAS-like palign (Panos Minos) - - (*) FIXME: Someone, fill this in with details - -C.2.38 Version 0.98bf (bug-fixed) - - (*) Fixed - elf and aoutb bug - shared libraries - multiple - "%include" bug in "-f obj" - jcxz, jecxz bug - unrecognized - option bug in ndisasm - -C.2.39 Version 0.98.03 with John Coffman's changes released 27-Jul-2000 - - (*) Added signed byte optimizations for the 0x81/0x83 class of - instructions: ADC, ADD, AND, CMP, OR, SBB, SUB, XOR: when used - as 'ADD reg16,imm' or 'ADD reg32,imm.' Also optimization of - signed byte form of 'PUSH imm' and 'IMUL reg,imm'/'IMUL - reg,reg,imm.' No size specification is needed. - - (*) Added multi-pass JMP and Jcc offset optimization. Offsets on - forward references will preferentially use the short form, - without the need to code a specific size (short or near) for the - branch. Added instructions for 'Jcc label' to use the form - 'Jnotcc $+3/JMP label', in cases where a short offset is out of - bounds. If compiling for a 386 or higher CPU, then the 386 form - of Jcc will be used instead. - - This feature is controlled by a new command-line switch: "O", (upper - case letter O). "-O0" reverts the assembler to no extra optimization - passes, "-O1" allows up to 5 extra passes, and "-O2"(default), - allows up to 10 extra optimization passes. - - (*) Added a new directive: 'cpu XXX', where XXX is any of: 8086, - 186, 286, 386, 486, 586, pentium, 686, PPro, P2, P3 or Katmai. - All are case insensitive. All instructions will be selected only - if they apply to the selected cpu or lower. Corrected a couple - of bugs in cpu-dependence in 'insns.dat'. - - (*) Added to 'standard.mac', the "use16" and "use32" forms of the - "bits 16/32" directive. This is nothing new, just conforms to a - lot of other assemblers. (minor) - - (*) Changed label allocation from 320/32 (10000 labels @ 200K+) to - 32/37 (1000 labels); makes running under DOS much easier. Since - additional label space is allocated dynamically, this should - have no effect on large programs with lots of labels. The 37 is - a prime, believed to be better for hashing. (minor) - -C.2.40 Version 0.98.03 - - "Integrated patchfile 0.98-0.98.01. I call this version 0.98.03 for - historical reasons: 0.98.02 was trashed." --John Coffman - <johninsd@san.rr.com>, 27-Jul-2000 - - (*) Kendall Bennett's SciTech MGL changes - - (*) Note that you must define "TASM_COMPAT" at compile-time to get - the Tasm Ideal Mode compatibility. - - (*) All changes can be compiled in and out using the TASM_COMPAT - macros, and when compiled without TASM_COMPAT defined we get the - exact same binary as the unmodified 0.98 sources. - - (*) standard.mac, macros.c: Added macros to ignore TASM directives - before first include - - (*) nasm.h: Added extern declaration for tasm_compatible_mode - - (*) nasm.c: Added global variable tasm_compatible_mode - - (*) Added command line switch for TASM compatible mode (-t) - - (*) Changed version command line to reflect when compiled with TASM - additions - - (*) Added response file processing to allow all arguments on a - single line (response file is @resp rather than -@resp for NASM - format). - - (*) labels.c: Changes islocal() macro to support TASM style @@local - labels. - - (*) Added islocalchar() macro to support TASM style @@local labels. - - (*) parser.c: Added support for TASM style memory references (ie: - mov [DWORD eax],10 rather than the NASM style mov DWORD - [eax],10). - - (*) preproc.c: Added new directives, `%arg', `%local', `%stacksize' - to directives table - - (*) Added support for TASM style directives without a leading % - symbol. - - (*) Integrated a block of changes from Andrew Zabolotny - <bit@eltech.ru>: - - (*) A new keyword `%xdefine' and its case-insensitive counterpart - `%ixdefine'. They work almost the same way as `%define' and - `%idefine' but expand the definition immediately, not on the - invocation. Something like a cross between `%define' and - `%assign'. The "x" suffix stands for "eXpand", so "xdefine" can - be deciphered as "expand-and-define". Thus you can do things - like this: - - %assign ofs 0 - - %macro arg 1 - %xdefine %1 dword [esp+ofs] - %assign ofs ofs+4 - %endmacro - - (*) Changed the place where the expansion of %$name macros are - expanded. Now they are converted into ..@ctxnum.name form when - detokenizing, so there are no quirks as before when using %$name - arguments to macros, in macros etc. For example: - - %macro abc 1 - %define %1 hello - %endm - - abc %$here - %$here - - Now last line will be expanded into "hello" as expected. This also - allows for lots of goodies, a good example are extended "proc" - macros included in this archive. - - (*) Added a check for "cstk" in smacro_defined() before calling - get_ctx() - this allows for things like: - - %ifdef %$abc - %endif - - to work without warnings even in no context. - - (*) Added a check for "cstk" in %if*ctx and %elif*ctx directives - - this allows to use `%ifctx' without excessive warnings. If there - is no active context, `%ifctx' goes through "false" branch. - - (*) Removed "user error: " prefix with `%error' directive: it just - clobbers the output and has absolutely no functionality. - Besides, this allows to write macros that does not differ from - built-in functions in any way. - - (*) Added expansion of string that is output by `%error' directive. - Now you can do things like: - - %define hello(x) Hello, x! - - %define %$name andy - %error "hello(%$name)" - - Same happened with `%include' directive. - - (*) Now all directives that expect an identifier will try to expand - and concatenate everything without whitespaces in between before - usage. For example, with "unfixed" nasm the commands - - %define %$abc hello - %define __%$abc goodbye - __%$abc - - would produce "incorrect" output: last line will expand to - - hello goodbyehello - - Not quite what you expected, eh? :-) The answer is that preprocessor - treats the `%define' construct as if it would be - - %define __ %$abc goodbye - - (note the white space between __ and %$abc). After my "fix" it will - "correctly" expand into - - goodbye - - as expected. Note that I use quotes around words "correct", - "incorrect" etc because this is rather a feature not a bug; however - current behaviour is more logical (and allows more advanced macro - usage :-). - - Same change was applied to: - `%push',`%macro',`%imacro',`%define',`%idefine',`%xdefine',`%ixdefine', - `%assign',`%iassign',`%undef' - - (*) A new directive [WARNING {+|-}warning-id] have been added. It - works only if the assembly phase is enabled (i.e. it doesn't - work with nasm -e). - - (*) A new warning type: macro-selfref. By default this warning is - disabled; when enabled NASM warns when a macro self-references - itself; for example the following source: - - [WARNING macro-selfref] - - %macro push 1-* - %rep %0 - push %1 - %rotate 1 - %endrep - %endmacro - - push eax,ebx,ecx - - will produce a warning, but if we remove the first line we won't see - it anymore (which is The Right Thing To Do {tm} IMHO since C - preprocessor eats such constructs without warnings at all). - - (*) Added a "error" routine to preprocessor which always will set - ERR_PASS1 bit in severity_code. This removes annoying repeated - errors on first and second passes from preprocessor. - - (*) Added the %+ operator in single-line macros for concatenating - two identifiers. Usage example: - - %define _myfunc _otherfunc - %define cextern(x) _ %+ x - cextern (myfunc) - - After first expansion, third line will become "_myfunc". After this - expansion is performed again so it becomes "_otherunc". - - (*) Now if preprocessor is in a non-emitting state, no warning or - error will be emitted. Example: - - %if 1 - mov eax,ebx - %else - put anything you want between these two brackets, - even macro-parameter references %1 or local - labels %$zz or macro-local labels %%zz - no - warning will be emitted. - %endif - - (*) Context-local variables on expansion as a last resort are looked - up in outer contexts. For example, the following piece: - - %push outer - %define %$a [esp] - - %push inner - %$a - %pop - %pop - - will expand correctly the fourth line to [esp]; if we'll define - another %$a inside the "inner" context, it will take precedence over - outer definition. However, this modification has been applied only - to expand_smacro and not to smacro_define: as a consequence - expansion looks in outer contexts, but `%ifdef' won't look in outer - contexts. - - This behaviour is needed because we don't want nested contexts to - act on already defined local macros. Example: - - %define %$arg1 [esp+4] - test eax,eax - if nz - mov eax,%$arg1 - endif - - In this example the "if" mmacro enters into the "if" context, so - %$arg1 is not valid anymore inside "if". Of course it could be - worked around by using explicitely %$$arg1 but this is ugly IMHO. - - (*) Fixed memory leak in `%undef'. The origline wasn't freed before - exiting on success. - - (*) Fixed trap in preprocessor when line expanded to empty set of - tokens. This happens, for example, in the following case: - - #define SOMETHING - SOMETHING - -C.2.41 Version 0.98 - - All changes since NASM 0.98p3 have been produced by H. Peter Anvin - <hpa@zytor.com>. - - (*) The documentation comment delimiter is - - (*) Allow EQU definitions to refer to external labels; reported by - Pedro Gimeno. - - (*) Re-enable support for RDOFF v1; reported by Pedro Gimeno. - - (*) Updated License file per OK from Simon and Julian. - -C.2.42 Version 0.98p9 - - (*) Update documentation (although the instruction set reference - will have to wait; I don't want to hold up the 0.98 release for - it.) - - (*) Verified that the NASM implementation of the PEXTRW and PMOVMSKB - instructions is correct. The encoding differs from what the - Intel manuals document, but the Pentium III behaviour matches - NASM, not the Intel manuals. - - (*) Fix handling of implicit sizes in PSHUFW and PINSRW, reported by - Stefan Hoffmeister. - - (*) Resurrect the -s option, which was removed when changing the - diagnostic output to stdout. - -C.2.43 Version 0.98p8 - - (*) Fix for "DB" when NASM is running on a bigendian machine. - - (*) Invoke insns.pl once for each output script, making Makefile.in - legal for "make -j". - - (*) Improve the Unix configure-based makefiles to make package - creation easier. - - (*) Included an RPM .spec file for building RPM (RedHat Package - Manager) packages on Linux or Unix systems. - - (*) Fix Makefile dependency problems. - - (*) Change src/rdsrc.pl to include sectioning information in info - output; required for install-info to work. - - (*) Updated the RDOFF distribution to version 2 from Jules; minor - massaging to make it compile in my environment. - - (*) Split doc files that can be built by anyone with a Perl - interpreter off into a separate archive. - - (*) "Dress rehearsal" release! - -C.2.44 Version 0.98p7 - - (*) Fixed opcodes with a third byte-sized immediate argument to not - complain if given "byte" on the immediate. - - (*) Allow `%undef' to remove single-line macros with arguments. This - matches the behaviour of #undef in the C preprocessor. - - (*) Allow -d, -u, -i and -p to be specified as -D, -U, -I and -P for - compatibility with most C compilers and preprocessors. This - allows Makefile options to be shared between cc and nasm, for - example. - - (*) Minor cleanups. - - (*) Went through the list of Katmai instructions and hopefully fixed - the (rather few) mistakes in it. - - (*) (Hopefully) fixed a number of disassembler bugs related to - ambiguous instructions (disambiguated by -p) and SSE - instructions with REP. - - (*) Fix for bug reported by Mark Junger: "call dword 0x12345678" - should work and may add an OSP (affected CALL, JMP, Jcc). - - (*) Fix for environments when "stderr" isn't a compile-time - constant. - -C.2.45 Version 0.98p6 - - (*) Took officially over coordination of the 0.98 release; so drop - the p3.x notation. Skipped p4 and p5 to avoid confusion with - John Fine's J4 and J5 releases. - - (*) Update the documentation; however, it still doesn't include - documentation for the various new instructions. I somehow wonder - if it makes sense to have an instruction set reference in the - assembler manual when Intel et al have PDF versions of their - manuals online. - - (*) Recognize "idt" or "centaur" for the -p option to ndisasm. - - (*) Changed error messages back to stderr where they belong, but add - an -E option to redirect them elsewhere (the DOS shell cannot - redirect stderr.) - - (*) -M option to generate Makefile dependencies (based on code from - Alex Verstak.) - - (*) `%undef' preprocessor directive, and -u option, that undefines a - single-line macro. - - (*) OS/2 Makefile (Mkfiles/Makefile.os2) for Borland under OS/2; - from Chuck Crayne. - - (*) Various minor bugfixes (reported by): - Dangling `%s' in - preproc.c (Martin Junker) - - (*) THERE ARE KNOWN BUGS IN SSE AND THE OTHER KATMAI INSTRUCTIONS. I - am on a trip and didn't bring the Katmai instruction reference, - so I can't work on them right now. - - (*) Updated the License file per agreement with Simon and Jules to - include a GPL distribution clause. - -C.2.46 Version 0.98p3.7 - - (*) (Hopefully) fixed the canned Makefiles to include the outrdf2 - and zoutieee modules. - - (*) Renamed changes.asm to changed.asm. - -C.2.47 Version 0.98p3.6 - - (*) Fixed a bunch of instructions that were added in 0.98p3.5 which - had memory operands, and the address-size prefix was missing - from the instruction pattern. - -C.2.48 Version 0.98p3.5 - - (*) Merged in changes from John S. Fine's 0.98-J5 release. John's - based 0.98-J5 on my 0.98p3.3 release; this merges the changes. - - (*) Expanded the instructions flag field to a long so we can fit - more flags; mark SSE (KNI) and AMD or Katmai-specific - instructions as such. - - (*) Fix the "PRIV" flag on a bunch of instructions, and create new - "PROT" flag for protected-mode-only instructions (orthogonal to - if the instruction is privileged!) and new "SMM" flag for SMM- - only instructions. - - (*) Added AMD-only SYSCALL and SYSRET instructions. - - (*) Make SSE actually work, and add new Katmai MMX instructions. - - (*) Added a -p (preferred vendor) option to ndisasm so that it can - distinguish e.g. Cyrix opcodes also used in SSE. For example: - - ndisasm -p cyrix aliased.bin - 00000000 670F514310 paddsiw mm0,[ebx+0x10] - 00000005 670F514320 paddsiw mm0,[ebx+0x20] - ndisasm -p intel aliased.bin - 00000000 670F514310 sqrtps xmm0,[ebx+0x10] - 00000005 670F514320 sqrtps xmm0,[ebx+0x20] - - (*) Added a bunch of Cyrix-specific instructions. - -C.2.49 Version 0.98p3.4 - - (*) Made at least an attempt to modify all the additional Makefiles - (in the Mkfiles directory). I can't test it, but this was the - best I could do. - - (*) DOS DJGPP+"Opus Make" Makefile from John S. Fine. - - (*) changes.asm changes from John S. Fine. - -C.2.50 Version 0.98p3.3 - - (*) Patch from Conan Brink to allow nesting of `%rep' directives. - - (*) If we're going to allow INT01 as an alias for INT1/ICEBP (one of - Jules 0.98p3 changes), then we should allow INT03 as an alias - for INT3 as well. - - (*) Updated changes.asm to include the latest changes. - - (*) Tried to clean up the <CR>s that had snuck in from a DOS/Windows - environment into my Unix environment, and try to make sure than - DOS/Windows users get them back. - - (*) We would silently generate broken tools if insns.dat wasn't - sorted properly. Change insns.pl so that the order doesn't - matter. - - (*) Fix bug in insns.pl (introduced by me) which would cause - conditional instructions to have an extra "cc" in disassembly, - e.g. "jnz" disassembled as "jccnz". - -C.2.51 Version 0.98p3.2 - - (*) Merged in John S. Fine's changes from his 0.98-J4 prerelease; - see http://www.csoft.net/cz/johnfine/ - - (*) Changed previous "spotless" Makefile target (appropriate for - distribution) to "distclean", and added "cleaner" target which - is same as "clean" except deletes files generated by Perl - scripts; "spotless" is union. - - (*) Removed BASIC programs from distribution. Get a Perl interpreter - instead (see below.) - - (*) Calling this "pre-release 3.2" rather than "p3-hpa2" because of - John's contributions. - - (*) Actually link in the IEEE output format (zoutieee.c); fix a - bunch of compiler warnings in that file. Note I don't know what - IEEE output is supposed to look like, so these changes were made - "blind". - -C.2.52 Version 0.98p3-hpa - - (*) Merged nasm098p3.zip with nasm-0.97.tar.gz to create a fully - buildable version for Unix systems (Makefile.in updates, etc.) - - (*) Changed insns.pl to create the instruction tables in nasm.h and - names.c, so that a new instruction can be added by adding it - *only* to insns.dat. - - (*) Added the following new instructions: SYSENTER, SYSEXIT, FXSAVE, - FXRSTOR, UD1, UD2 (the latter two are two opcodes that Intel - guarantee will never be used; one of them is documented as UD2 - in Intel documentation, the other one just as "Undefined Opcode" - -- calling it UD1 seemed to make sense.) - - (*) MAX_SYMBOL was defined to be 9, but LOADALL286 and LOADALL386 - are 10 characters long. Now MAX_SYMBOL is derived from - insns.dat. - - (*) A note on the BASIC programs included: forget them. insns.bas is - already out of date. Get yourself a Perl interpreter for your - platform of choice at http://www.cpan.org/ports/index.html. - -C.2.53 Version 0.98 pre-release 3 - - (*) added response file support, improved command line handling, new - layout help screen - - (*) fixed limit checking bug, 'OUT byte nn, reg' bug, and a couple - of rdoff related bugs, updated Wishlist; 0.98 Prerelease 3. - -C.2.54 Version 0.98 pre-release 2 - - (*) fixed bug in outcoff.c to do with truncating section names - longer than 8 characters, referencing beyond end of string; 0.98 - pre-release 2 - -C.2.55 Version 0.98 pre-release 1 - - (*) Fixed a bug whereby STRUC didn't work at all in RDF. - - (*) Fixed a problem with group specification in PUBDEFs in OBJ. - - (*) Improved ease of adding new output formats. Contribution due to - Fox Cutter. - - (*) Fixed a bug in relocations in the `bin' format: was showing up - when a relocatable reference crossed an 8192-byte boundary in - any output section. - - (*) Fixed a bug in local labels: local-label lookups were - inconsistent between passes one and two if an EQU occurred - between the definition of a global label and the subsequent use - of a local label local to that global. - - (*) Fixed a seg-fault in the preprocessor (again) which happened - when you use a blank line as the first line of a multi-line - macro definition and then defined a label on the same line as a - call to that macro. - - (*) Fixed a stale-pointer bug in the handling of the NASM - environment variable. Thanks to Thomas McWilliams. - - (*) ELF had a hard limit on the number of sections which caused - segfaults when transgressed. Fixed. - - (*) Added ability for ndisasm to read from stdin by using `-' as the - filename. - - (*) ndisasm wasn't outputting the TO keyword. Fixed. - - (*) Fixed error cascade on bogus expression in `%if' - an error in - evaluation was causing the entire `%if' to be discarded, thus - creating trouble later when the `%else' or `%endif' was - encountered. - - (*) Forward reference tracking was instruction-granular not operand- - granular, which was causing 286-specific code to be generated - needlessly on code of the form `shr word [forwardref],1'. Thanks - to Jim Hague for sending a patch. - - (*) All messages now appear on stdout, as sending them to stderr - serves no useful purpose other than to make redirection - difficult. - - (*) Fixed the problem with EQUs pointing to an external symbol - - this now generates an error message. - - (*) Allowed multiple size prefixes to an operand, of which only the - first is taken into account. - - (*) Incorporated John Fine's changes, including fixes of a large - number of preprocessor bugs, some small problems in OBJ, and a - reworking of label handling to define labels before their line - is assembled, rather than after. - - (*) Reformatted a lot of the source code to be more readable. - Included 'coding.txt' as a guideline for how to format code for - contributors. - - (*) Stopped nested `%reps' causing a panic - they now cause a - slightly more friendly error message instead. - - (*) Fixed floating point constant problems (patch by Pedro Gimeno) - - (*) Fixed the return value of insn_size() not being checked for -1, - indicating an error. - - (*) Incorporated 3Dnow! instructions. - - (*) Fixed the 'mov eax, eax + ebx' bug. - - (*) Fixed the GLOBAL EQU bug in ELF. Released developers release 3. - - (*) Incorporated John Fine's command line parsing changes - - (*) Incorporated David Lindauer's OMF debug support - - (*) Made changes for LCC 4.0 support (`__NASM_CDecl__', removed - register size specification warning when sizes agree). - - C.3 NASM 0.9 Series - - Revisions before 0.98. - - C.3.1 Version 0.97 released December 1997 - - (*) This was entirely a bug-fix release to 0.96, which seems to have - got cursed. Silly me. - - (*) Fixed stupid mistake in OBJ which caused `MOV EAX,<constant>' to - fail. Caused by an error in the `MOV EAX,<segment>' support. - - (*) ndisasm hung at EOF when compiled with lcc on Linux because lcc - on Linux somehow breaks feof(). ndisasm now does not rely on - feof(). - - (*) A heading in the documentation was missing due to a markup error - in the indexing. Fixed. - - (*) Fixed failure to update all pointers on realloc() within - extended- operand code in parser.c. Was causing wrong behaviour - and seg faults on lines such as `dd 0.0,0.0,0.0,0.0,...' - - (*) Fixed a subtle preprocessor bug whereby invoking one multi-line - macro on the first line of the expansion of another, when the - second had been invoked with a label defined before it, didn't - expand the inner macro. - - (*) Added internal.doc back in to the distribution archives - it was - missing in 0.96 *blush* - - (*) Fixed bug causing 0.96 to be unable to assemble its own test - files, specifically objtest.asm. *blush again* - - (*) Fixed seg-faults and bogus error messages caused by mismatching - `%rep' and `%endrep' within multi-line macro definitions. - - (*) Fixed a problem with buffer overrun in OBJ, which was causing - corruption at ends of long PUBDEF records. - - (*) Separated DOS archives into main-program and documentation to - reduce download size. - - C.3.2 Version 0.96 released November 1997 - - (*) Fixed a bug whereby, if `nasm sourcefile' would cause a filename - collision warning and put output into `nasm.out', then `nasm - sourcefile -o outputfile' still gave the warning even though the - `-o' was honoured. Fixed name pollution under Digital UNIX: one - of its header files defined R_SP, which broke the enum in - nasm.h. - - (*) Fixed minor instruction table problems: FUCOM and FUCOMP didn't - have two-operand forms; NDISASM didn't recognise the longer - register forms of PUSH and POP (eg FF F3 for PUSH BX); TEST - mem,imm32 was flagged as undocumented; the 32-bit forms of CMOV - had 16-bit operand size prefixes; `AAD imm' and `AAM imm' are no - longer flagged as undocumented because the Intel Architecture - reference documents them. - - (*) Fixed a problem with the local-label mechanism, whereby strange - types of symbol (EQUs, auto-defined OBJ segment base symbols) - interfered with the `previous global label' value and screwed up - local labels. - - (*) Fixed a bug whereby the stub preprocessor didn't communicate - with the listing file generator, so that the -a and -l options - in conjunction would produce a useless listing file. - - (*) Merged `os2' object file format back into `obj', after - discovering that `obj' _also_ shouldn't have a link pass - separator in a module containing a non-trivial MODEND. Flat - segments are now declared using the FLAT attribute. `os2' is no - longer a valid object format name: use `obj'. - - (*) Removed the fixed-size temporary storage in the evaluator. Very - very long expressions (like `mov ax,1+1+1+1+...' for two hundred - 1s or so) should now no longer crash NASM. - - (*) Fixed a bug involving segfaults on disassembly of MMX - instructions, by changing the meaning of one of the operand-type - flags in nasm.h. This may cause other apparently unrelated MMX - problems; it needs to be tested thoroughly. - - (*) Fixed some buffer overrun problems with large OBJ output files. - Thanks to DJ Delorie for the bug report and fix. - - (*) Made preprocess-only mode actually listen to the `%line' markers - as it prints them, so that it can report errors more sanely. - - (*) Re-designed the evaluator to keep more sensible track of - expressions involving forward references: can now cope with - previously-nightmare situations such as: - - mov ax,foo | bar - foo equ 1 - bar equ 2 - - (*) Added the ALIGN and ALIGNB standard macros. - - (*) Added PIC support in ELF: use of WRT to obtain the four extra - relocation types needed. - - (*) Added the ability for output file formats to define their own - extensions to the GLOBAL, COMMON and EXTERN directives. - - (*) Implemented common-variable alignment, and global-symbol type - and size declarations, in ELF. - - (*) Implemented NEAR and FAR keywords for common variables, plus - far-common element size specification, in OBJ. - - (*) Added a feature whereby EXTERNs and COMMONs in OBJ can be given - a default WRT specification (either a segment or a group). - - (*) Transformed the Unix NASM archive into an auto-configuring - package. - - (*) Added a sanity-check for people applying SEG to things which are - already segment bases: this previously went unnoticed by the SEG - processing and caused OBJ-driver panics later. - - (*) Added the ability, in OBJ format, to deal with `MOV - EAX,<segment>' type references: OBJ doesn't directly support - dword-size segment base fixups, but as long as the low two bytes - of the constant term are zero, a word-size fixup can be - generated instead and it will work. - - (*) Added the ability to specify sections' alignment requirements in - Win32 object files and pure binary files. - - (*) Added preprocess-time expression evaluation: the `%assign' (and - `%iassign') directive and the bare `%if' (and `%elif') - conditional. Added relational operators to the evaluator, for - use only in `%if' constructs: the standard relationals = < > <= - >= <> (and C-like synonyms == and !=) plus low-precedence - logical operators &&, ^^ and ||. - - (*) Added a preprocessor repeat construct: `%rep' / `%exitrep' / - `%endrep'. - - (*) Added the __FILE__ and __LINE__ standard macros. - - (*) Added a sanity check for number constants being greater than - 0xFFFFFFFF. The warning can be disabled. - - (*) Added the %0 token whereby a variadic multi-line macro can tell - how many parameters it's been given in a specific invocation. - - (*) Added `%rotate', allowing multi-line macro parameters to be - cycled. - - (*) Added the `*' option for the maximum parameter count on multi- - line macros, allowing them to take arbitrarily many parameters. - - (*) Added the ability for the user-level forms of EXTERN, GLOBAL and - COMMON to take more than one argument. - - (*) Added the IMPORT and EXPORT directives in OBJ format, to deal - with Windows DLLs. - - (*) Added some more preprocessor `%if' constructs: `%ifidn' / - `%ifidni' (exact textual identity), and `%ifid' / `%ifnum' / - `%ifstr' (token type testing). - - (*) Added the ability to distinguish SHL AX,1 (the 8086 version) - from SHL AX,BYTE 1 (the 286-and-upwards version whose constant - happens to be 1). - - (*) Added NetBSD/FreeBSD/OpenBSD's variant of a.out format, complete - with PIC shared library features. - - (*) Changed NASM's idiosyncratic handling of FCLEX, FDISI, FENI, - FINIT, FSAVE, FSTCW, FSTENV, and FSTSW to bring it into line - with the otherwise accepted standard. The previous behaviour, - though it was a deliberate feature, was a deliberate feature - based on a misunderstanding. Apologies for the inconvenience. - - (*) Improved the flexibility of ABSOLUTE: you can now give it an - expression rather than being restricted to a constant, and it - can take relocatable arguments as well. - - (*) Added the ability for a variable to be declared as EXTERN - multiple times, and the subsequent definitions are just ignored. - - (*) We now allow instruction prefixes (CS, DS, LOCK, REPZ etc) to be - alone on a line (without a following instruction). - - (*) Improved sanity checks on whether the arguments to EXTERN, - GLOBAL and COMMON are valid identifiers. - - (*) Added misc/exebin.mac to allow direct generation of .EXE files - by hacking up an EXE header using DB and DW; also added - test/binexe.asm to demonstrate the use of this. Thanks to Yann - Guidon for contributing the EXE header code. - - (*) ndisasm forgot to check whether the input file had been - successfully opened. Now it does. Doh! - - (*) Added the Cyrix extensions to the MMX instruction set. - - (*) Added a hinting mechanism to allow [EAX+EBX] and [EBX+EAX] to be - assembled differently. This is important since [ESI+EBP] and - [EBP+ESI] have different default base segment registers. - - (*) Added support for the PharLap OMF extension for 4096-byte - segment alignment. - - C.3.3 Version 0.95 released July 1997 - - (*) Fixed yet another ELF bug. This one manifested if the user - relied on the default segment, and attempted to define global - symbols without first explicitly declaring the target segment. - - (*) Added makefiles (for NASM and the RDF tools) to build Win32 - console apps under Symantec C++. Donated by Mark Junker. - - (*) Added `macros.bas' and `insns.bas', QBasic versions of the Perl - scripts that convert `standard.mac' to `macros.c' and convert - `insns.dat' to `insnsa.c' and `insnsd.c'. Also thanks to Mark - Junker. - - (*) Changed the diassembled forms of the conditional instructions so - that JB is now emitted as JC, and other similar changes. - Suggested list by Ulrich Doewich. - - (*) Added `@' to the list of valid characters to begin an identifier - with. - - (*) Documentary changes, notably the addition of the `Common - Problems' section in nasm.doc. - - (*) Fixed a bug relating to 32-bit PC-relative fixups in OBJ. - - (*) Fixed a bug in perm_copy() in labels.c which was causing - exceptions in cleanup_labels() on some systems. - - (*) Positivity sanity check in TIMES argument changed from a warning - to an error following a further complaint. - - (*) Changed the acceptable limits on byte and word operands to allow - things like `~10111001b' to work. - - (*) Fixed a major problem in the preprocessor which caused seg- - faults if macro definitions contained blank lines or comment- - only lines. - - (*) Fixed inadequate error checking on the commas separating the - arguments to `db', `dw' etc. - - (*) Fixed a crippling bug in the handling of macros with operand - counts defined with a `+' modifier. - - (*) Fixed a bug whereby object file formats which stored the input - file name in the output file (such as OBJ and COFF) weren't - doing so correctly when the output file name was specified on - the command line. - - (*) Removed [INC] and [INCLUDE] support for good, since they were - obsolete anyway. - - (*) Fixed a bug in OBJ which caused all fixups to be output in 16- - bit (old-format) FIXUPP records, rather than putting the 32-bit - ones in FIXUPP32 (new-format) records. - - (*) Added, tentatively, OS/2 object file support (as a minor variant - on OBJ). - - (*) Updates to Fox Cutter's Borland C makefile, Makefile.bc2. - - (*) Removed a spurious second fclose() on the output file. - - (*) Added the `-s' command line option to redirect all messages - which would go to stderr (errors, help text) to stdout instead. - - (*) Added the `-w' command line option to selectively suppress some - classes of assembly warning messages. - - (*) Added the `-p' pre-include and `-d' pre-define command-line - options. - - (*) Added an include file search path: the `-i' command line option. - - (*) Fixed a silly little preprocessor bug whereby starting a line - with a `%!' environment-variable reference caused an `unknown - directive' error. - - (*) Added the long-awaited listing file support: the `-l' command - line option. - - (*) Fixed a problem with OBJ format whereby, in the absence of any - explicit segment definition, non-global symbols declared in the - implicit default segment generated spurious EXTDEF records in - the output. - - (*) Added the NASM environment variable. - - (*) From this version forward, Win32 console-mode binaries will be - included in the DOS distribution in addition to the 16-bit - binaries. Added Makefile.vc for this purpose. - - (*) Added `return 0;' to test/objlink.c to prevent compiler - warnings. - - (*) Added the __NASM_MAJOR__ and __NASM_MINOR__ standard defines. - - (*) Added an alternative memory-reference syntax in which prefixing - an operand with `&' is equivalent to enclosing it in square - brackets, at the request of Fox Cutter. - - (*) Errors in pass two now cause the program to return a non-zero - error code, which they didn't before. - - (*) Fixed the single-line macro cycle detection, which didn't work - at all on macros with no parameters (caused an infinite loop). - Also changed the behaviour of single-line macro cycle detection - to work like cpp, so that macros like `extrn' as given in the - documentation can be implemented. - - (*) Fixed the implementation of WRT, which was too restrictive in - that you couldn't do `mov ax,[di+abc wrt dgroup]' because - (di+abc) wasn't a relocatable reference. - - C.3.4 Version 0.94 released April 1997 - - (*) Major item: added the macro processor. - - (*) Added undocumented instructions SMI, IBTS, XBTS and LOADALL286. - Also reorganised CMPXCHG instruction into early-486 and Pentium - forms. Thanks to Thobias Jones for the information. - - (*) Fixed two more stupid bugs in ELF, which were causing `ld' to - continue to seg-fault in a lot of non-trivial cases. - - (*) Fixed a seg-fault in the label manager. - - (*) Stopped FBLD and FBSTP from _requiring_ the TWORD keyword, which - is the only option for BCD loads/stores in any case. - - (*) Ensured FLDCW, FSTCW and FSTSW can cope with the WORD keyword, - if anyone bothers to provide it. Previously they complained - unless no keyword at all was present. - - (*) Some forms of FDIV/FDIVR and FSUB/FSUBR were still inverted: a - vestige of a bug that I thought had been fixed in 0.92. This was - fixed, hopefully for good this time... - - (*) Another minor phase error (insofar as a phase error can _ever_ - be minor) fixed, this one occurring in code of the form - - rol ax,forward_reference - forward_reference equ 1 - - (*) The number supplied to TIMES is now sanity-checked for - positivity, and also may be greater than 64K (which previously - didn't work on 16-bit systems). - - (*) Added Watcom C makefiles, and misc/pmw.bat, donated by Dominik - Behr. - - (*) Added the INCBIN pseudo-opcode. - - (*) Due to the advent of the preprocessor, the [INCLUDE] and [INC] - directives have become obsolete. They are still supported in - this version, with a warning, but won't be in the next. - - (*) Fixed a bug in OBJ format, which caused incorrect object records - to be output when absolute labels were made global. - - (*) Updates to RDOFF subdirectory, and changes to outrdf.c. - - C.3.5 Version 0.93 released January 1997 - - This release went out in a great hurry after semi-crippling bugs - were found in 0.92. - - (*) Really _did_ fix the stack overflows this time. *blush* - - (*) Had problems with EA instruction sizes changing between passes, - when an offset contained a forward reference and so 4 bytes were - allocated for the offset in pass one; by pass two the symbol had - been defined and happened to be a small absolute value, so only - 1 byte got allocated, causing instruction size mismatch between - passes and hence incorrect address calculations. Fixed. - - (*) Stupid bug in the revised ELF section generation fixed - (associated string-table section for .symtab was hard-coded as - 7, even when this didn't fit with the real section table). Was - causing `ld' to seg-fault under Linux. - - (*) Included a new Borland C makefile, Makefile.bc2, donated by Fox - Cutter <lmb@comtch.iea.com>. - - C.3.6 Version 0.92 released January 1997 - - (*) The FDIVP/FDIVRP and FSUBP/FSUBRP pairs had been inverted: this - was fixed. This also affected the LCC driver. - - (*) Fixed a bug regarding 32-bit effective addresses of the form - `[other_register+ESP]'. - - (*) Documentary changes, notably documentation of the fact that - Borland Win32 compilers use `obj' rather than `win32' object - format. - - (*) Fixed the COMENT record in OBJ files, which was formatted - incorrectly. - - (*) Fixed a bug causing segfaults in large RDF files. - - (*) OBJ format now strips initial periods from segment and group - definitions, in order to avoid complications with the local - label syntax. - - (*) Fixed a bug in disassembling far calls and jumps in NDISASM. - - (*) Added support for user-defined sections in COFF and ELF files. - - (*) Compiled the DOS binaries with a sensible amount of stack, to - prevent stack overflows on any arithmetic expression containing - parentheses. - - (*) Fixed a bug in handling of files that do not terminate in a - newline. - - C.3.7 Version 0.91 released November 1996 - - (*) Loads of bug fixes. - - (*) Support for RDF added. - - (*) Support for DBG debugging format added. - - (*) Support for 32-bit extensions to Microsoft OBJ format added. - - (*) Revised for Borland C: some variable names changed, makefile - added. - - (*) LCC support revised to actually work. - - (*) JMP/CALL NEAR/FAR notation added. - - (*) `a16', `o16', `a32' and `o32' prefixes added. - - (*) Range checking on short jumps implemented. - - (*) MMX instruction support added. - - (*) Negative floating point constant support added. - - (*) Memory handling improved to bypass 64K barrier under DOS. - - (*) `$' prefix to force treatment of reserved words as identifiers - added. - - (*) Default-size mechanism for object formats added. - - (*) Compile-time configurability added. - - (*) `#', `@', `~' and c{?} are now valid characters in labels. - - (*) `-e' and `-k' options in NDISASM added. - - C.3.8 Version 0.90 released October 1996 - - First release version. First support for object file output. Other - changes from previous version (0.3x) too numerous to document. diff --git a/packaging/nasm.manifest b/packaging/nasm.manifest deleted file mode 100644 index 017d22d..0000000 --- a/packaging/nasm.manifest +++ /dev/null @@ -1,5 +0,0 @@ -<manifest> - <request> - <domain name="_"/> - </request> -</manifest> diff --git a/packaging/nasm.spec b/packaging/nasm.spec index dcfda70..17fa4d7 100644 --- a/packaging/nasm.spec +++ b/packaging/nasm.spec @@ -2,16 +2,13 @@ Summary: A portable x86 assembler which uses Intel-like syntax Name: nasm Version: 2.08rc7 -Release: 2.16 +Release: 2.1 License: simplified BSD license Group: Development/Languages URL: http://nasm.sourceforge.net/ Source0: nasm-%{version}.tar.bz2 Source1: nasm-%{version}-xdoc.tar.bz2 Source2: nasm.sh -Source1001: packaging/nasm.manifest -# >> gbp-patch-tags # auto-added by gbp -# << gbp-patch-tags # auto-added by gbp BuildRequires: perl %package rdoff @@ -30,14 +27,11 @@ include linker, library manager, loader, and information dump. %prep %setup -q -# >> gbp-apply-patches # auto-added by gbp -# << gbp-apply-patches # auto-added by gbp tar xjf %{SOURCE1} --strip-components 1 %build -cp %{SOURCE1001} . %configure -make all +make all %{?_smp_mflags} %install rm -rf $RPM_BUILD_ROOT @@ -51,14 +45,12 @@ cp %{SOURCE2} $RPM_BUILD_ROOT/etc/profile.d/ %remove_docs %files -%manifest nasm.manifest %{_bindir}/nasm %{_bindir}/ndisasm /etc/profile.d/nasm.sh %files rdoff -%manifest nasm.manifest %defattr(-,root,root) %{_bindir}/ldrdf %{_bindir}/rdf2bin |