Age | Commit message (Collapse) | Author | Files | Lines |
|
|
|
|
|
|
|
(#24918)
- Microsoft.DotNet.Arcade.Sdk - 1.0.0-beta.19302.2
- Microsoft.DotNet.Build.Tasks.Feed - 2.2.0-beta.19302.2
- Microsoft.DotNet.Build.Tasks.Packaging - 1.0.0-beta.19302.2
- Microsoft.DotNet.Helix.Sdk - 2.0.0-beta.19302.2
|
|
* JIT HW Intrinsic implementation doc
|
|
|
|
|
|
available to trace
|
|
* Update dependencies from https://github.com/dotnet/arcade build 20190601.2
- Microsoft.DotNet.Arcade.Sdk - 1.0.0-beta.19301.2
- Microsoft.DotNet.Build.Tasks.Feed - 2.2.0-beta.19301.2
- Microsoft.DotNet.Build.Tasks.Packaging - 1.0.0-beta.19301.2
- Microsoft.DotNet.Helix.Sdk - 2.0.0-beta.19301.2
|
|
* Improve fatal err msg
* Match SO format
* a
* Remove dead define
* More adjustments to ex msg
* And PAL
* Remove special case for SOE
* Remove excess Debug.Assert newline
* Remove excess newline
* typo
* New format
* Remove DebugProvider redundancy
* Adjustments
* Remove preceding newline
* Make other SOE and OOM consistent
* Tidy up assertion msg
* Fix missing newline after inner exception divider
* CR when no inner exception
* ToString never CR terminated
* disable corefx tests temporarily
|
|
* Move EventPipeEventSource deletion to a better place
* typo
|
|
Flush() behaves like Flush(false) and writes out any buffered data but doesn't P/Invoke to FlushFileBuffers/FSync to flush the OS buffers.
But whereas FlushAsync() is supposed to just be an async equivalent of Flush(), it's actually behaving like Flush(true). This makes FlushAsync() inconsistent and much more expensive. (This is separate from FlushAsync not actually being async, which is an impactful problem to be solved separately.)
This changes FlushAsync to behave like Flush()/Flush(false) rather than Flush(true). If someone wants the FlushFileBuffers/FSync behavior, they can call Flush(true).
|
|
|
|
20190530.2 (#24882)" (#24898)
This reverts commit aa26688ee5e2b03945a22cda324887341e10595f.
|
|
|
|
20190530.01 (#24884)
- Microsoft.NETCore.App - 3.0.0-preview6-27730-01
|
|
(#24882)
- Microsoft.DotNet.Arcade.Sdk - 1.0.0-beta.19280.2
- Microsoft.DotNet.Build.Tasks.Feed - 2.2.0-beta.19280.2
- Microsoft.DotNet.Build.Tasks.Packaging - 1.0.0-beta.19280.2
- Microsoft.DotNet.Helix.Sdk - 2.0.0-beta.19280.2
|
|
Fix #24846
|
|
When adding the signature copying some time ago, I have overlooked one
place where the moduleIndex was set to -1 instead of MODULE_INDEX_NONE.
I was using the -1 during the development and replaced it by creating
MODULE_INDEX_NONE during final cleanup at all but one places.
This caused issues during crossgen even without large version bubble
enabled. Methods requiring generic constraint were not crossgen-ed due
to this.
|
|
* Makes GetTotalAllocatedBytes monotonic in nonprecise case.
Fixes:#24615
* read `dead_threads_non_alloc_bytes` atomically on 32bit platforms
* Update src/vm/comutilnative.cpp
Co-Authored-By: Jan Kotas <jkotas@microsoft.com>
|
|
|
|
the amount of work we do when we're going to return early anyway. (#24872)
|
|
Pre SDK 3.0, a project type that is not .csproj would not include Directory.Build.targets. Change empty.proj to empty.csproj to allow source build, which uses a 2.1 sdk to build, correctly import Directory.Build.targets.
|
|
* Remove unused code in vm/util.*
|
|
(#24865)
|
|
Fixes for building the Desktop JIT in CodegenMirror
|
|
|
|
|
|
* Remove BuildTools imports from product build
* Split scripts for installing dotnet vs BuildTools
|
|
|
|
* Add build-tools.cmd
This script builds selected tools in the src/tools directory (for the
time being, it builds just the runincontext tool). This is a preparation
step to enable nightly runs with unloadability tests. The
build-tools.cmd should build helper tools needed during test runs.
* Remove the build-tools.cmd
* Build the tools as part of the managed runtime build.
* Update the R2RDump project to put binaries into the bin/Product/...
* Update runincontext.cmd/sh scripts
Reflect the new location of the runincontext binary
|
|
|
|
* Update dependencies from https://github.com/dotnet/arcade build 20190529.5
- Microsoft.DotNet.Arcade.Sdk - 1.0.0-beta.19279.5
- Microsoft.DotNet.Build.Tasks.Feed - 2.2.0-beta.19279.5
- Microsoft.DotNet.Build.Tasks.Packaging - 1.0.0-beta.19279.5
- Microsoft.DotNet.Helix.Sdk - 2.0.0-beta.19279.5
* Work around dotnet/arcade#2673
|
|
|
|
|
|
Synchronized methods show up when `Console` is used and the missing helper mapping would cause us not to AOT compile them.
|
|
(#24855)
- Microsoft.Bcl.AsyncInterfaces - 1.0.0-preview6.19279.8
- Microsoft.NETCore.Platforms - 3.0.0-preview6.19279.8
- Microsoft.Private.CoreFx.NETCoreApp - 4.6.0-preview6.19279.8
|
|
20190529.07 (#24856)
- Microsoft.NETCore.App - 3.0.0-preview6-27729-07
|
|
* fix compilation with gcc7.x
In file included from src/inc/sstring.h:1029:0,
from src/inc/ex.h:19,
from src/inc/stgpool.h:28,
from src/inc/../md/inc/metamodel.h:18,
from src/inc/../md/inc/metamodelro.h:19,
from src/inc/metadata.h:17,
from src/inc/corpriv.h:21,
from src/md/compiler/stdafx.h:20,
from src/md/compiler/mdutil.cpp:12:
src/inc/sstring.inl: In member function ‘const WCHAR* SString::GetUnicode() const’:
src/inc/sstring.inl:650:5: error: nonnull argument ‘this’ compared to NULL [-Werror=nonnull-compare]
if (this == NULL)
^~
add version check
* Use the abstracted version of __nodebug__
* cc1: warning: command line option ‘-fpermissive’ is valid for C++/ObjC++ but not for C
|
|
We currently don't precompile methods that use hardware intrinsics because we don't know the CPU that the generated code will run on. Jitting these methods slows down startup and accounts for 3% of startup time in PowerShell.
With this change, we're going to lift this restriction for CoreLib (the thing that matters for startup) and support generating HW intrinsics for our minimum supported target ISA (SSE/SSE2).
|
|
* Remove all uses of CanRunManagedCode()
|
|
test case run at (#24844)
least 4 to 5 times faster than before.
Fallback to old transport ReadMemory if /proc/<pid>/mem can't be opened. This happens
on attach because of permissions/access, but works fine on the launch (the most
important case).
|
|
* default value is 1, and when set to 0 will disable loading LTTng.
Co-Authored-By: Jan Kotas <jkotas@microsoft.com>
|
|
* Update dependencies from https://github.com/dotnet/arcade build 20190524.6
- Microsoft.DotNet.Arcade.Sdk - 1.0.0-beta.19274.6
- Microsoft.DotNet.Build.Tasks.Feed - 2.2.0-beta.19274.6
- Microsoft.DotNet.Build.Tasks.Packaging - 1.0.0-beta.19274.6
- Microsoft.DotNet.Helix.Sdk - 2.0.0-beta.19274.6
* Update dependencies from https://github.com/dotnet/arcade build 20190528.1
- Microsoft.DotNet.Arcade.Sdk - 1.0.0-beta.19278.1
- Microsoft.DotNet.Build.Tasks.Feed - 2.2.0-beta.19278.1
- Microsoft.DotNet.Build.Tasks.Packaging - 1.0.0-beta.19278.1
- Microsoft.DotNet.Helix.Sdk - 2.0.0-beta.19278.1
* Work around native initialization
|
|
* Avoid using ins_Load/ins_Store with constant type
* Use ldr to load array lengths/lower bounds
genCodeForArrIndex and genCodeForArrOffset emit a ldrsw on ARM64 but that's not necessary. Array lengths are always positive. Lower bounds are signed but then the subsequent subtract is anyway EA_4BYTE so sign extension isn't actually needed. Just use ldr on both ARM32 and ARM64.
* Use ldr to load array length (part 2)
genCodeForIndexAddr's range check generation is messed up:
- It uses ldrsw to load array length on ARM64. Not needed, the length is positive.
- It uses sxtw to sign etxend the array length if the index is native int. But it was already extended by the load.
- It creates IND and LEA nodes out of thin air. Just call the appropiate emitter functions.
- It always generates a TYP_I_IMPL cmp, even when the index is TYP_INT. Well, that's just bogus.
* Stop the using the instruction size for immediate scaling on ARM32
The scaling of immediate operands is a property of the instruction and its encoding. It doesnt' make sense to throw the instruction size (emitAttr) into the mix, that's a codegen/emitter specific concept. On ARM32 it's also almost meaningless, at least in the case of integer types - all instructions are really EA_4BYTE, even ldrb, ldrh etc.
The ARM64 emitter already extracts the scaling factor from the instruction. It can't use the instruction size as on ARM64 the size is used to select between 32 bit and 64 bit instructions so it's never EA_1BYTE/EA_2BYTE.
* Stop using ldrsw for TYP_INT loads
ARM64's ins_Load returns INS_ldrsw for TYP_INT but there's nothing in the JIT type system that requires sign extending TYP_INT values on load. The fact that an indir node has TYP_INT doesn't even imply that the value to load is signed, it may be unsigned and indir nodes will never have type TYP_UINT nor have the GTF_UNSIGNED flag set.
XARCH uses a mov (rather than movsxd, the equivalent of ldrsw) so it zero extends. There's no reason for ARM64 to behave differently and doing so makes it more difficult to share codegen logic between XARCH and ARM64
Other ARM64 compilers also use ldr rather than ldrsw.
This requires patching up emitInsAdjustLoadStoreAttr so EA_4BYTE loads don't end up using EA_8BYTE, which ldrsw requires.
* Cleanup genCodeForIndir/genCodeForStoreInd
In particular, cleanup selection of acquire/release instructions. The existing code starts by selecting a "normal" instruction, only to throw it away and redo the type/size logic in the volatile case. And get it wrong in the process, it required that "targetType" be TYP_UINT or TYP_LONG to use ldar. But there are no TYP_UINT indirs.
Also rename "targetType" to "type", using "target" is misleading. The real target type would be genActualType(tree->TypeGet()).
* Remove ARM32/64 load/store instruction size inconsistencies
- Require EA_4BYTE for byte/halfword instructions on ARM32.
- Remove emitInsAdjustLoadStoreAttr on ARM64. Getting the correct instruction size simply requires using emitActualTypeSize, that will provide the correct size on both ARM32 and ARM64.
- Replace emitTypeSize with emitActualTypeSize as needed.
* Remove unnecessary insUnscaleImm parameter
|
|
|
|
|
|
* On systems that have both cmake 2 and cmake 3, use cmake 3.
Move CMAKE_EXPORT_COMPILE_COMMANDS to top level instead of in gen-buildsys-*. Define the CMake install prefix in gen-buildsys-* instead of pulling from an environment variable.
Define C++ standard as CMake property instead of as flag.
Move CLR_DEFINES_*_INIT out of overrides and into configurecompiler.cmake
Move flags that generate debug info into configurecompiler.cmake
Remove the CMAKE_USER_RULES_OVERRIDE files.
Add cmake version output for determining what cmake versions each CI system has.
Fix syntax in gen-buildsys-clang.
Change add_compile_definitions back to add_definitions
Add -D prefix for adding definitions via add_definitions
Remove extraneous double-quote
Change default config definition adding to the syntax in master
Switch back to old CMAKE_<LANG>_FLAGS way of setting the language standards and try to go back to 2.8.12 minimum
Switch back setting compile definitions for non-Windows branch too.
Use SET with CMAKE_<LANG>_FLAGS. Convert some usages of appending to CMAKE_<LANG>_FLAGS to add_compile_options where possible.
Set CMAKE_<LANG>_FLAGS_INIT instead of CMAKE_<LANG>_FLAGS
Make sure configureopimitzation.cmake is included correctly in test build.
Try to add brackets to get the Linux ARM compilation working correctly.
Define standard language version in configurecompiler.cmake instead of root CMakeLists (so tests get it)
Try to move langauge standard check to configure.cmake
define language standard in each root CMakeLists.txt
Fix off-Windows test build.
Set CMAKE_EXPORT_COMPILE_COMMANDS after the project() call
* Set CMAKE_USER_MAKE_RULES_OVERRIDE to "" to not break incremental builds after building on a branch that had it set to a path.
* Remove CMake version output.
* Move comment outside of multiline command.
* Retry setting CMAKE_USER_MAKE_RULES_OVERRIDE
* Remove unnecessary variable wrappers.
* Simplify cmake 3 resolution.
* Explicitly use CMAKE_CXX_FLAGS for C++-only flags that GCC complains about on C files.
* Set -Wall via CMAKE_<LANG>_FLAGS until we can move all flag settings to add_compile_options
* Fix typos
* Another temporary precedence issue.
* include configureoptimization.cmake in configurecompiler.cmake
* Move setting CMAKE_EXPORT_COMPILE_COMMANDS to configurecompiler.cmake.
* Rename configure.cmake -> verify_lto.cmake.
* Fix path to verify-lto
* Try using CMAKE_<LANG>_FLAGS instead of CMAKE_<LANG>_FLAGS_INIT.
* Revert name change to configure.camek
|
|
|
|
|