diff options
564 files changed, 29058 insertions, 11882 deletions
diff --git a/Documentation/i2c/busses/i2c-viapro b/Documentation/i2c/busses/i2c-viapro index 9363b8bd610..16775663b9f 100644 --- a/Documentation/i2c/busses/i2c-viapro +++ b/Documentation/i2c/busses/i2c-viapro @@ -7,12 +7,10 @@ Supported adapters: * VIA Technologies, Inc. VT82C686A/B Datasheet: Sometimes available at the VIA website - * VIA Technologies, Inc. VT8231, VT8233, VT8233A, VT8235, VT8237 - Datasheet: available on request from Via + * VIA Technologies, Inc. VT8231, VT8233, VT8233A, VT8235, VT8237R + Datasheet: available on request from VIA Authors: - Frodo Looijaard <frodol@dds.nl>, - Philip Edelbrock <phil@netroedge.com>, Kyösti Mälkki <kmalkki@cc.hut.fi>, Mark D. Studebaker <mdsxyz123@yahoo.com>, Jean Delvare <khali@linux-fr.org> diff --git a/Documentation/i2c/writing-clients b/Documentation/i2c/writing-clients index cff7b652588..d19993cc060 100644 --- a/Documentation/i2c/writing-clients +++ b/Documentation/i2c/writing-clients @@ -412,7 +412,7 @@ For now, you can ignore the `flags' parameter. It is there for future use. release_region(address,FOO_EXTENT); /* SENSORS ONLY END */ ERROR1: - kfree(new_client); + kfree(data); ERROR0: return err; } @@ -443,7 +443,7 @@ much simpler than the attachment code, fortunately! release_region(client->addr,LM78_EXTENT); /* HYBRID SENSORS CHIP ONLY END */ - kfree(data); + kfree(i2c_get_clientdata(client)); return 0; } diff --git a/Documentation/networking/dccp.txt b/Documentation/networking/dccp.txt new file mode 100644 index 00000000000..c45daabd3bf --- /dev/null +++ b/Documentation/networking/dccp.txt @@ -0,0 +1,56 @@ +DCCP protocol +============ + +Last updated: 10 November 2005 + +Contents +======== + +- Introduction +- Missing features +- Socket options +- Notes + +Introduction +============ + +Datagram Congestion Control Protocol (DCCP) is an unreliable, connection +based protocol designed to solve issues present in UDP and TCP particularly +for real time and multimedia traffic. + +It has a base protocol and pluggable congestion control IDs (CCIDs). + +It is at draft RFC status and the homepage for DCCP as a protocol is at: + http://www.icir.org/kohler/dcp/ + +Missing features +================ + +The DCCP implementation does not currently have all the features that are in +the draft RFC. + +In particular the following are missing: +- CCID2 support +- feature negotiation + +When testing against other implementations it appears that elapsed time +options are not coded compliant to the specification. + +Socket options +============== + +DCCP_SOCKOPT_PACKET_SIZE is used for CCID3 to set default packet size for +calculations. + +DCCP_SOCKOPT_SERVICE sets the service. This is compulsory as per the +specification. If you don't set it you will get EPROTO. + +Notes +===== + +SELinux does not yet have support for DCCP. You will need to turn it off or +else you will get EACCES. + +DCCP does not travel through NAT successfully at present. This is because +the checksum covers the psuedo-header as per TCP and UDP. It should be +relatively trivial to add Linux NAT support for DCCP. diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index 65895bb5141..ebc09a159f6 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -78,6 +78,11 @@ inet_peer_gc_maxtime - INTEGER TCP variables: +tcp_abc - INTEGER + Controls Appropriate Byte Count defined in RFC3465. If set to + 0 then does congestion avoid once per ack. 1 is conservative + value, and 2 is more agressive. + tcp_syn_retries - INTEGER Number of times initial SYNs for an active TCP connection attempt will be retransmitted. Should not be higher than 255. Default value diff --git a/MAINTAINERS b/MAINTAINERS index 5541f9970b8..0b03a88e88b 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -707,7 +707,7 @@ DCCP PROTOCOL P: Arnaldo Carvalho de Melo M: acme@mandriva.com L: dccp@vger.kernel.org -W: http://www.wlug.org.nz/DCCP +W: http://linux-net.osdl.org/index.php/DCCP S: Maintained DECnet NETWORK LAYER @@ -347,7 +347,7 @@ AFLAGS_KERNEL = # Needed to be compatible with the O= option LINUXINCLUDE := -Iinclude \ $(if $(KBUILD_SRC),-Iinclude2 -I$(srctree)/include) \ - -imacros include/linux/autoconf.h + -include include/linux/autoconf.h CPPFLAGS := -D__KERNEL__ $(LINUXINCLUDE) @@ -407,7 +407,7 @@ outputmakefile: # of make so .config is not included in this case either (for *config). no-dot-config-targets := clean mrproper distclean \ - cscope TAGS tags help %docs check% + cscope TAGS tags help %docs check% kernelrelease config-targets := 0 mixed-targets := 0 diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index ec77721507c..3df7cbd924a 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -239,6 +239,8 @@ source "arch/arm/plat-omap/Kconfig" source "arch/arm/mach-omap1/Kconfig" +source "arch/arm/mach-omap2/Kconfig" + source "arch/arm/mach-s3c2410/Kconfig" source "arch/arm/mach-lh7a40x/Kconfig" diff --git a/arch/arm/Makefile b/arch/arm/Makefile index 114cda7f1b7..81bd2193fe6 100644 --- a/arch/arm/Makefile +++ b/arch/arm/Makefile @@ -93,6 +93,7 @@ textaddr-$(CONFIG_ARCH_FORTUNET) := 0xc0008000 machine-$(CONFIG_ARCH_IXP4XX) := ixp4xx machine-$(CONFIG_ARCH_IXP2000) := ixp2000 machine-$(CONFIG_ARCH_OMAP1) := omap1 + machine-$(CONFIG_ARCH_OMAP2) := omap2 incdir-$(CONFIG_ARCH_OMAP) := omap machine-$(CONFIG_ARCH_S3C2410) := s3c2410 machine-$(CONFIG_ARCH_LH7A40X) := lh7a40x diff --git a/arch/arm/configs/omap_h2_1610_defconfig b/arch/arm/configs/omap_h2_1610_defconfig index 4198677cd39..529f0f72e1e 100644 --- a/arch/arm/configs/omap_h2_1610_defconfig +++ b/arch/arm/configs/omap_h2_1610_defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.13 -# Mon Sep 5 18:07:12 2005 +# Linux kernel version: 2.6.14 +# Wed Nov 9 18:53:40 2005 # CONFIG_ARM=y CONFIG_MMU=y @@ -22,6 +22,7 @@ CONFIG_INIT_ENV_ARG_LIMIT=32 # General setup # CONFIG_LOCALVERSION="" +CONFIG_LOCALVERSION_AUTO=y CONFIG_SWAP=y CONFIG_SYSVIPC=y # CONFIG_POSIX_MQUEUE is not set @@ -31,6 +32,7 @@ CONFIG_SYSCTL=y # CONFIG_HOTPLUG is not set CONFIG_KOBJECT_UEVENT=y # CONFIG_IKCONFIG is not set +CONFIG_INITRAMFS_SOURCE="" # CONFIG_EMBEDDED is not set CONFIG_KALLSYMS=y # CONFIG_KALLSYMS_EXTRA_PASS is not set @@ -60,6 +62,23 @@ CONFIG_OBSOLETE_MODPARM=y # CONFIG_KMOD is not set # +# Block layer +# + +# +# IO Schedulers +# +CONFIG_IOSCHED_NOOP=y +CONFIG_IOSCHED_AS=y +CONFIG_IOSCHED_DEADLINE=y +CONFIG_IOSCHED_CFQ=y +CONFIG_DEFAULT_AS=y +# CONFIG_DEFAULT_DEADLINE is not set +# CONFIG_DEFAULT_CFQ is not set +# CONFIG_DEFAULT_NOOP is not set +CONFIG_DEFAULT_IOSCHED="anticipatory" + +# # System Type # # CONFIG_ARCH_CLPS7500 is not set @@ -81,6 +100,7 @@ CONFIG_OBSOLETE_MODPARM=y # CONFIG_ARCH_LH7A40X is not set CONFIG_ARCH_OMAP=y # CONFIG_ARCH_VERSATILE is not set +# CONFIG_ARCH_REALVIEW is not set # CONFIG_ARCH_IMX is not set # CONFIG_ARCH_H720X is not set # CONFIG_ARCH_AAEC2000 is not set @@ -112,7 +132,7 @@ CONFIG_OMAP_SERIAL_WAKE=y # OMAP Core Type # # CONFIG_ARCH_OMAP730 is not set -# CONFIG_ARCH_OMAP1510 is not set +# CONFIG_ARCH_OMAP15XX is not set CONFIG_ARCH_OMAP16XX=y # @@ -177,6 +197,8 @@ CONFIG_FLATMEM_MANUAL=y # CONFIG_SPARSEMEM_MANUAL is not set CONFIG_FLATMEM=y CONFIG_FLAT_NODE_MEM_MAP=y +# CONFIG_SPARSEMEM_STATIC is not set +CONFIG_SPLIT_PTLOCK_CPUS=4096 # CONFIG_LEDS is not set CONFIG_ALIGNMENT_TRAP=y @@ -258,14 +280,19 @@ CONFIG_IP_PNP_BOOTP=y # CONFIG_INET_ESP is not set # CONFIG_INET_IPCOMP is not set # CONFIG_INET_TUNNEL is not set -CONFIG_IP_TCPDIAG=y -# CONFIG_IP_TCPDIAG_IPV6 is not set +CONFIG_INET_DIAG=y +CONFIG_INET_TCP_DIAG=y # CONFIG_TCP_CONG_ADVANCED is not set CONFIG_TCP_CONG_BIC=y # CONFIG_IPV6 is not set # CONFIG_NETFILTER is not set # +# DCCP Configuration (EXPERIMENTAL) +# +# CONFIG_IP_DCCP is not set + +# # SCTP Configuration (EXPERIMENTAL) # # CONFIG_IP_SCTP is not set @@ -281,6 +308,10 @@ CONFIG_TCP_CONG_BIC=y # CONFIG_NET_DIVERT is not set # CONFIG_ECONET is not set # CONFIG_WAN_ROUTER is not set + +# +# QoS and/or fair queueing +# # CONFIG_NET_SCHED is not set # CONFIG_NET_CLS_ROUTE is not set @@ -291,6 +322,7 @@ CONFIG_TCP_CONG_BIC=y # CONFIG_HAMRADIO is not set # CONFIG_IRDA is not set # CONFIG_BT is not set +# CONFIG_IEEE80211 is not set # # Device Drivers @@ -328,21 +360,13 @@ CONFIG_BLK_DEV_RAM=y CONFIG_BLK_DEV_RAM_COUNT=16 CONFIG_BLK_DEV_RAM_SIZE=8192 CONFIG_BLK_DEV_INITRD=y -CONFIG_INITRAMFS_SOURCE="" # CONFIG_CDROM_PKTCDVD is not set - -# -# IO Schedulers -# -CONFIG_IOSCHED_NOOP=y -CONFIG_IOSCHED_AS=y -CONFIG_IOSCHED_DEADLINE=y -CONFIG_IOSCHED_CFQ=y CONFIG_ATA_OVER_ETH=m # # SCSI device support # +# CONFIG_RAID_ATTRS is not set CONFIG_SCSI=y CONFIG_SCSI_PROC_FS=y @@ -369,10 +393,12 @@ CONFIG_SCSI_PROC_FS=y # CONFIG_SCSI_SPI_ATTRS is not set # CONFIG_SCSI_FC_ATTRS is not set # CONFIG_SCSI_ISCSI_ATTRS is not set +# CONFIG_SCSI_SAS_ATTRS is not set # # SCSI low-level drivers # +# CONFIG_ISCSI_TCP is not set # CONFIG_SCSI_SATA is not set # CONFIG_SCSI_DEBUG is not set @@ -404,6 +430,11 @@ CONFIG_NETDEVICES=y # CONFIG_TUN is not set # +# PHY device support +# +# CONFIG_PHYLIB is not set + +# # Ethernet (10 or 100Mbit) # CONFIG_NET_ETHERNET=y @@ -439,6 +470,7 @@ CONFIG_PPP=y # CONFIG_PPP_SYNC_TTY is not set # CONFIG_PPP_DEFLATE is not set # CONFIG_PPP_BSDCOMP is not set +# CONFIG_PPP_MPPE is not set # CONFIG_PPPOE is not set CONFIG_SLIP=y CONFIG_SLIP_COMPRESSED=y @@ -541,18 +573,18 @@ CONFIG_WATCHDOG_NOWAYOUT=y # # TPM devices # +# CONFIG_TELCLOCK is not set # # I2C support # # CONFIG_I2C is not set -# CONFIG_I2C_SENSOR is not set -CONFIG_ISP1301_OMAP=y # # Hardware Monitoring support # CONFIG_HWMON=y +# CONFIG_HWMON_VID is not set # CONFIG_HWMON_DEBUG_CHIP is not set # @@ -560,6 +592,10 @@ CONFIG_HWMON=y # # +# Multimedia Capabilities Port drivers +# + +# # Multimedia devices # # CONFIG_VIDEO_DEV is not set @@ -576,7 +612,6 @@ CONFIG_FB=y # CONFIG_FB_CFB_FILLRECT is not set # CONFIG_FB_CFB_COPYAREA is not set # CONFIG_FB_CFB_IMAGEBLIT is not set -# CONFIG_FB_SOFT_CURSOR is not set # CONFIG_FB_MACMODES is not set CONFIG_FB_MODE_HELPERS=y # CONFIG_FB_TILEBLITTING is not set @@ -589,6 +624,7 @@ CONFIG_FB_MODE_HELPERS=y # CONFIG_VGA_CONSOLE is not set CONFIG_DUMMY_CONSOLE=y CONFIG_FRAMEBUFFER_CONSOLE=y +# CONFIG_FRAMEBUFFER_CONSOLE_ROTATION is not set CONFIG_FONTS=y CONFIG_FONT_8x8=y CONFIG_FONT_8x16=y @@ -600,6 +636,7 @@ CONFIG_FONT_8x16=y # CONFIG_FONT_SUN8x16 is not set # CONFIG_FONT_SUN12x22 is not set # CONFIG_FONT_10x18 is not set +# CONFIG_FONT_RL is not set # # Logo configuration @@ -624,10 +661,10 @@ CONFIG_SOUND=y # Open Sound System # CONFIG_SOUND_PRIME=y +# CONFIG_OBSOLETE_OSS_DRIVER is not set # CONFIG_SOUND_MSNDCLAS is not set # CONFIG_SOUND_MSNDPIN is not set # CONFIG_SOUND_OSS is not set -# CONFIG_SOUND_AD1980 is not set # # USB support @@ -637,22 +674,21 @@ CONFIG_USB_ARCH_HAS_OHCI=y # CONFIG_USB is not set # +# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support' +# + +# # USB Gadget Support # -CONFIG_USB_GADGET=y -# CONFIG_USB_GADGET_DEBUG_FILES is not set -CONFIG_USB_GADGET_SELECTED=y +# CONFIG_USB_GADGET is not set # CONFIG_USB_GADGET_NET2280 is not set # CONFIG_USB_GADGET_PXA2XX is not set # CONFIG_USB_GADGET_GOKU is not set # CONFIG_USB_GADGET_LH7A40X is not set -CONFIG_USB_GADGET_OMAP=y -CONFIG_USB_OMAP=y +# CONFIG_USB_GADGET_OMAP is not set # CONFIG_USB_GADGET_DUMMY_HCD is not set -# CONFIG_USB_GADGET_DUALSPEED is not set # CONFIG_USB_ZERO is not set -CONFIG_USB_ETH=y -CONFIG_USB_ETH_RNDIS=y +# CONFIG_USB_ETH is not set # CONFIG_USB_GADGETFS is not set # CONFIG_USB_FILE_STORAGE is not set # CONFIG_USB_G_SERIAL is not set @@ -673,10 +709,6 @@ CONFIG_EXT2_FS=y # CONFIG_REISERFS_FS is not set # CONFIG_JFS_FS is not set # CONFIG_FS_POSIX_ACL is not set - -# -# XFS support -# # CONFIG_XFS_FS is not set # CONFIG_MINIX_FS is not set CONFIG_ROMFS_FS=y @@ -685,6 +717,7 @@ CONFIG_INOTIFY=y CONFIG_DNOTIFY=y # CONFIG_AUTOFS_FS is not set # CONFIG_AUTOFS4_FS is not set +# CONFIG_FUSE_FS is not set # # CD-ROM/DVD Filesystems @@ -706,10 +739,10 @@ CONFIG_FAT_DEFAULT_CODEPAGE=437 # CONFIG_PROC_FS=y CONFIG_SYSFS=y -# CONFIG_DEVPTS_FS_XATTR is not set # CONFIG_TMPFS is not set # CONFIG_HUGETLB_PAGE is not set CONFIG_RAMFS=y +# CONFIG_RELAYFS_FS is not set # # Miscellaneous filesystems @@ -750,6 +783,7 @@ CONFIG_RPCSEC_GSS_KRB5=y # CONFIG_NCP_FS is not set # CONFIG_CODA_FS is not set # CONFIG_AFS_FS is not set +# CONFIG_9P_FS is not set # # Partition Types @@ -859,6 +893,7 @@ CONFIG_CRYPTO_DES=y # Library routines # # CONFIG_CRC_CCITT is not set +# CONFIG_CRC16 is not set CONFIG_CRC32=y # CONFIG_LIBCRC32C is not set CONFIG_ZLIB_INFLATE=y diff --git a/arch/arm/lib/csumpartial.S b/arch/arm/lib/csumpartial.S index cb5e3708f11..fe797cf320b 100644 --- a/arch/arm/lib/csumpartial.S +++ b/arch/arm/lib/csumpartial.S @@ -39,6 +39,7 @@ td3 .req lr /* we must have at least one byte. */ tst buf, #1 @ odd address? + movne sum, sum, ror #8 ldrneb td0, [buf], #1 subne len, len, #1 adcnes sum, sum, td0, put_byte_1 @@ -103,6 +104,9 @@ ENTRY(csum_partial) cmp len, #8 @ Ensure that we have at least blo .less8 @ 8 bytes to copy. + tst buf, #1 + movne sum, sum, ror #8 + adds sum, sum, #0 @ C = 0 tst buf, #3 @ Test destination alignment blne .not_aligned @ aligh destination, return here diff --git a/arch/arm/mach-ixp4xx/Kconfig b/arch/arm/mach-ixp4xx/Kconfig index 89762a26495..385285851cb 100644 --- a/arch/arm/mach-ixp4xx/Kconfig +++ b/arch/arm/mach-ixp4xx/Kconfig @@ -8,6 +8,16 @@ menu "Intel IXP4xx Implementation Options" comment "IXP4xx Platforms" +# This entry is placed on top because otherwise it would have +# been shown as a submenu. +config MACH_NSLU2 + bool + prompt "NSLU2" if !(MACH_IXDP465 || MACH_IXDPG425 || ARCH_IXDP425 || ARCH_ADI_COYOTE || ARCH_AVILA || ARCH_IXCDP1100 || ARCH_PRPMC1100 || MACH_GTWX5715) + help + Say 'Y' here if you want your kernel to support Linksys's + NSLU2 NAS device. For more information on this platform, + see http://www.nslu2-linux.org + config ARCH_AVILA bool "Avila" help diff --git a/arch/arm/mach-ixp4xx/Makefile b/arch/arm/mach-ixp4xx/Makefile index ddecbda4a63..7a15629c18d 100644 --- a/arch/arm/mach-ixp4xx/Makefile +++ b/arch/arm/mach-ixp4xx/Makefile @@ -8,4 +8,5 @@ obj-$(CONFIG_ARCH_IXDP4XX) += ixdp425-pci.o ixdp425-setup.o obj-$(CONFIG_MACH_IXDPG425) += ixdpg425-pci.o coyote-setup.o obj-$(CONFIG_ARCH_ADI_COYOTE) += coyote-pci.o coyote-setup.o obj-$(CONFIG_MACH_GTWX5715) += gtwx5715-pci.o gtwx5715-setup.o +obj-$(CONFIG_MACH_NSLU2) += nslu2-pci.o nslu2-setup.o nslu2-power.o diff --git a/arch/arm/mach-ixp4xx/nslu2-pci.c b/arch/arm/mach-ixp4xx/nslu2-pci.c new file mode 100644 index 00000000000..a575f2e0b2c --- /dev/null +++ b/arch/arm/mach-ixp4xx/nslu2-pci.c @@ -0,0 +1,77 @@ +/* + * arch/arm/mach-ixp4xx/nslu2-pci.c + * + * NSLU2 board-level PCI initialization + * + * based on ixdp425-pci.c: + * Copyright (C) 2002 Intel Corporation. + * Copyright (C) 2003-2004 MontaVista Software, Inc. + * + * Maintainer: http://www.nslu2-linux.org/ + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ + +#include <linux/config.h> +#include <linux/pci.h> +#include <linux/init.h> + +#include <asm/mach/pci.h> +#include <asm/mach-types.h> + +void __init nslu2_pci_preinit(void) +{ + set_irq_type(IRQ_NSLU2_PCI_INTA, IRQT_LOW); + set_irq_type(IRQ_NSLU2_PCI_INTB, IRQT_LOW); + set_irq_type(IRQ_NSLU2_PCI_INTC, IRQT_LOW); + + gpio_line_isr_clear(NSLU2_PCI_INTA_PIN); + gpio_line_isr_clear(NSLU2_PCI_INTB_PIN); + gpio_line_isr_clear(NSLU2_PCI_INTC_PIN); + + /* INTD is not configured as GPIO is used + * for the power input button. + */ + + ixp4xx_pci_preinit(); +} + +static int __init nslu2_map_irq(struct pci_dev *dev, u8 slot, u8 pin) +{ + static int pci_irq_table[NSLU2_PCI_IRQ_LINES] = { + IRQ_NSLU2_PCI_INTA, + IRQ_NSLU2_PCI_INTB, + IRQ_NSLU2_PCI_INTC, + }; + + int irq = -1; + + if (slot >= 1 && slot <= NSLU2_PCI_MAX_DEV && + pin >= 1 && pin <= NSLU2_PCI_IRQ_LINES) { + irq = pci_irq_table[(slot + pin - 2) % NSLU2_PCI_IRQ_LINES]; + } + + return irq; +} + +struct hw_pci __initdata nslu2_pci = { + .nr_controllers = 1, + .preinit = nslu2_pci_preinit, + .swizzle = pci_std_swizzle, + .setup = ixp4xx_setup, + .scan = ixp4xx_scan_bus, + .map_irq = nslu2_map_irq, +}; + +int __init nslu2_pci_init(void) /* monkey see, monkey do */ +{ + if (machine_is_nslu2()) + pci_common_init(&nslu2_pci); + + return 0; +} + +subsys_initcall(nslu2_pci_init); diff --git a/arch/arm/mach-ixp4xx/nslu2-power.c b/arch/arm/mach-ixp4xx/nslu2-power.c new file mode 100644 index 00000000000..18fbc8c0fb3 --- /dev/null +++ b/arch/arm/mach-ixp4xx/nslu2-power.c @@ -0,0 +1,92 @@ +/* + * arch/arm/mach-ixp4xx/nslu2-power.c + * + * NSLU2 Power/Reset driver + * + * Copyright (C) 2005 Tower Technologies + * + * based on nslu2-io.c + * Copyright (C) 2004 Karen Spearel + * + * Author: Alessandro Zummo <a.zummo@towertech.it> + * Maintainers: http://www.nslu2-linux.org/ + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ + +#include <linux/module.h> +#include <linux/reboot.h> +#include <linux/interrupt.h> + +#include <asm/mach-types.h> + +extern void ctrl_alt_del(void); + +static irqreturn_t nslu2_power_handler(int irq, void *dev_id, struct pt_regs *regs) +{ + /* Signal init to do the ctrlaltdel action, this will bypass init if + * it hasn't started and do a kernel_restart. + */ + ctrl_alt_del(); + + return IRQ_HANDLED; +} + +static irqreturn_t nslu2_reset_handler(int irq, void *dev_id, struct pt_regs *regs) +{ + /* This is the paper-clip reset, it shuts the machine down directly. + */ + machine_power_off(); + + return IRQ_HANDLED; +} + +static int __init nslu2_power_init(void) +{ + if (!(machine_is_nslu2())) + return 0; + + *IXP4XX_GPIO_GPISR = 0x20400000; /* read the 2 irqs to clr */ + + set_irq_type(NSLU2_RB_IRQ, IRQT_LOW); + set_irq_type(NSLU2_PB_IRQ, IRQT_HIGH); + + gpio_line_isr_clear(NSLU2_RB_GPIO); + gpio_line_isr_clear(NSLU2_PB_GPIO); + + if (request_irq(NSLU2_RB_IRQ, &nslu2_reset_handler, + SA_INTERRUPT, "NSLU2 reset button", NULL) < 0) { + + printk(KERN_DEBUG "Reset Button IRQ %d not available\n", + NSLU2_RB_IRQ); + + return -EIO; + } + + if (request_irq(NSLU2_PB_IRQ, &nslu2_power_handler, + SA_INTERRUPT, "NSLU2 power button", NULL) < 0) { + + printk(KERN_DEBUG "Power Button IRQ %d not available\n", + NSLU2_PB_IRQ); + + return -EIO; + } + + return 0; +} + +static void __exit nslu2_power_exit(void) +{ + free_irq(NSLU2_RB_IRQ, NULL); + free_irq(NSLU2_PB_IRQ, NULL); +} + +module_init(nslu2_power_init); +module_exit(nslu2_power_exit); + +MODULE_AUTHOR("Alessandro Zummo <a.zummo@towertech.it>"); +MODULE_DESCRIPTION("NSLU2 Power/Reset driver"); +MODULE_LICENSE("GPL"); diff --git a/arch/arm/mach-ixp4xx/nslu2-setup.c b/arch/arm/mach-ixp4xx/nslu2-setup.c new file mode 100644 index 00000000000..289e94cb65c --- /dev/null +++ b/arch/arm/mach-ixp4xx/nslu2-setup.c @@ -0,0 +1,134 @@ +/* + * arch/arm/mach-ixp4xx/nslu2-setup.c + * + * NSLU2 board-setup + * + * based ixdp425-setup.c: + * Copyright (C) 2003-2004 MontaVista Software, Inc. + * + * Author: Mark Rakes <mrakes at mac.com> + * Maintainers: http://www.nslu2-linux.org/ + * + * Fixed missing init_time in MACHINE_START kas11 10/22/04 + * Changed to conform to new style __init ixdp425 kas11 10/22/04 + */ + +#include <linux/kernel.h> +#include <linux/serial.h> +#include <linux/serial_8250.h> + +#include <asm/mach-types.h> +#include <asm/mach/arch.h> +#include <asm/mach/flash.h> + +static struct flash_platform_data nslu2_flash_data = { + .map_name = "cfi_probe", + .width = 2, +}; + +static struct resource nslu2_flash_resource = { + .start = NSLU2_FLASH_BASE, + .end = NSLU2_FLASH_BASE + NSLU2_FLASH_SIZE, + .flags = IORESOURCE_MEM, +}; + +static struct platform_device nslu2_flash = { + .name = "IXP4XX-Flash", + .id = 0, + .dev.platform_data = &nslu2_flash_data, + .num_resources = 1, + .resource = &nslu2_flash_resource, +}; + +static struct ixp4xx_i2c_pins nslu2_i2c_gpio_pins = { + .sda_pin = NSLU2_SDA_PIN, + .scl_pin = NSLU2_SCL_PIN, +}; + +static struct platform_device nslu2_i2c_controller = { + .name = "IXP4XX-I2C", + .id = 0, + .dev.platform_data = &nslu2_i2c_gpio_pins, + .num_resources = 0, +}; + +static struct resource nslu2_uart_resources[] = { + { + .start = IXP4XX_UART1_BASE_PHYS, + .end = IXP4XX_UART1_BASE_PHYS + 0x0fff, + .flags = IORESOURCE_MEM, + }, + { + .start = IXP4XX_UART2_BASE_PHYS, + .end = IXP4XX_UART2_BASE_PHYS + 0x0fff, + .flags = IORESOURCE_MEM, + } +}; + +static struct plat_serial8250_port nslu2_uart_data[] = { + { + .mapbase = IXP4XX_UART1_BASE_PHYS, + .membase = (char *)IXP4XX_UART1_BASE_VIRT + REG_OFFSET, + .irq = IRQ_IXP4XX_UART1, + .flags = UPF_BOOT_AUTOCONF, + .iotype = UPIO_MEM, + .regshift = 2, + .uartclk = IXP4XX_UART_XTAL, + }, + { + .mapbase = IXP4XX_UART2_BASE_PHYS, + .membase = (char *)IXP4XX_UART2_BASE_VIRT + REG_OFFSET, + .irq = IRQ_IXP4XX_UART2, + .flags = UPF_BOOT_AUTOCONF, + .iotype = UPIO_MEM, + .regshift = 2, + .uartclk = IXP4XX_UART_XTAL, + }, + { } +}; + +static struct platform_device nslu2_uart = { + .name = "serial8250", + .id = PLAT8250_DEV_PLATFORM, + .dev.platform_data = nslu2_uart_data, + .num_resources = 2, + .resource = nslu2_uart_resources, +}; + +static struct platform_device *nslu2_devices[] __initdata = { + &nslu2_i2c_controller, + &nslu2_flash, + &nslu2_uart, +}; + +static void nslu2_power_off(void) +{ + /* This causes the box to drop the power and go dead. */ + + /* enable the pwr cntl gpio */ + gpio_line_config(NSLU2_PO_GPIO, IXP4XX_GPIO_OUT); + + /* do the deed */ + gpio_line_set(NSLU2_PO_GPIO, IXP4XX_GPIO_HIGH); +} + +static void __init nslu2_init(void) +{ + ixp4xx_sys_init(); + + pm_power_off = nslu2_power_off; + + platform_add_devices(nslu2_devices, ARRAY_SIZE(nslu2_devices)); +} + +MACHINE_START(NSLU2, "Linksys NSLU2") + /* Maintainer: www.nslu2-linux.org */ + .phys_ram = PHYS_OFFSET, + .phys_io = IXP4XX_PERIPHERAL_BASE_PHYS, + .io_pg_offst = ((IXP4XX_PERIPHERAL_BASE_VIRT) >> 18) & 0xFFFC, + .boot_params = 0x00000100, + .map_io = ixp4xx_map_io, + .init_irq = ixp4xx_init_irq, + .timer = &ixp4xx_timer, + .init_machine = nslu2_init, +MACHINE_END diff --git a/arch/arm/mach-omap1/Kconfig b/arch/arm/mach-omap1/Kconfig index 27fc2e8e5fc..86a0f0d1434 100644 --- a/arch/arm/mach-omap1/Kconfig +++ b/arch/arm/mach-omap1/Kconfig @@ -6,10 +6,10 @@ config ARCH_OMAP730 bool "OMAP730 Based System" select ARCH_OMAP_OTG -config ARCH_OMAP1510 +config ARCH_OMAP15XX depends on ARCH_OMAP1 default y - bool "OMAP1510 Based System" + bool "OMAP15xx Based System" config ARCH_OMAP16XX depends on ARCH_OMAP1 @@ -21,7 +21,7 @@ comment "OMAP Board Type" config MACH_OMAP_INNOVATOR bool "TI Innovator" - depends on ARCH_OMAP1 && (ARCH_OMAP1510 || ARCH_OMAP16XX) + depends on ARCH_OMAP1 && (ARCH_OMAP15XX || ARCH_OMAP16XX) help TI OMAP 1510 or 1610 Innovator board support. Say Y here if you have such a board. @@ -64,20 +64,30 @@ config MACH_OMAP_PERSEUS2 config MACH_VOICEBLUE bool "Voiceblue" - depends on ARCH_OMAP1 && ARCH_OMAP1510 + depends on ARCH_OMAP1 && ARCH_OMAP15XX help Support for Voiceblue GSM/VoIP gateway. Say Y here if you have such a board. config MACH_NETSTAR bool "NetStar" - depends on ARCH_OMAP1 && ARCH_OMAP1510 + depends on ARCH_OMAP1 && ARCH_OMAP15XX help Support for NetStar PBX. Say Y here if you have such a board. +config MACH_OMAP_PALMTE + bool "Palm Tungsten E" + depends on ARCH_OMAP1 && ARCH_OMAP15XX + help + Support for the Palm Tungsten E PDA. Currently only the LCD panel + is supported. To boot the kernel, you'll need a PalmOS compatible + bootloader; check out http://palmtelinux.sourceforge.net for more + informations. + Say Y here if you have such a PDA, say NO otherwise. + config MACH_OMAP_GENERIC bool "Generic OMAP board" - depends on ARCH_OMAP1 && (ARCH_OMAP1510 || ARCH_OMAP16XX) + depends on ARCH_OMAP1 && (ARCH_OMAP15XX || ARCH_OMAP16XX) help Support for generic OMAP-1510, 1610 or 1710 board with no FPGA. Can be used as template for porting Linux to @@ -121,32 +131,32 @@ config OMAP_ARM_182MHZ config OMAP_ARM_168MHZ bool "OMAP ARM 168 MHz CPU" - depends on ARCH_OMAP1 && (ARCH_OMAP1510 || ARCH_OMAP16XX || ARCH_OMAP730) + depends on ARCH_OMAP1 && (ARCH_OMAP15XX || ARCH_OMAP16XX || ARCH_OMAP730) help Enable 168MHz clock for OMAP CPU. If unsure, say N. config OMAP_ARM_150MHZ bool "OMAP ARM 150 MHz CPU" - depends on ARCH_OMAP1 && ARCH_OMAP1510 + depends on ARCH_OMAP1 && ARCH_OMAP15XX help Enable 150MHz clock for OMAP CPU. If unsure, say N. config OMAP_ARM_120MHZ bool "OMAP ARM 120 MHz CPU" - depends on ARCH_OMAP1 && (ARCH_OMAP1510 || ARCH_OMAP16XX || ARCH_OMAP730) + depends on ARCH_OMAP1 && (ARCH_OMAP15XX || ARCH_OMAP16XX || ARCH_OMAP730) help Enable 120MHz clock for OMAP CPU. If unsure, say N. config OMAP_ARM_60MHZ bool "OMAP ARM 60 MHz CPU" - depends on ARCH_OMAP1 && (ARCH_OMAP1510 || ARCH_OMAP16XX || ARCH_OMAP730) + depends on ARCH_OMAP1 && (ARCH_OMAP15XX || ARCH_OMAP16XX || ARCH_OMAP730) default y help Enable 60MHz clock for OMAP CPU. If unsure, say Y. config OMAP_ARM_30MHZ bool "OMAP ARM 30 MHz CPU" - depends on ARCH_OMAP1 && (ARCH_OMAP1510 || ARCH_OMAP16XX || ARCH_OMAP730) + depends on ARCH_OMAP1 && (ARCH_OMAP15XX || ARCH_OMAP16XX || ARCH_OMAP730) help Enable 30MHz clock for OMAP CPU. If unsure, say N. diff --git a/arch/arm/mach-omap1/Makefile b/arch/arm/mach-omap1/Makefile index 181a93deaae..b0b00156faa 100644 --- a/arch/arm/mach-omap1/Makefile +++ b/arch/arm/mach-omap1/Makefile @@ -3,7 +3,7 @@ # # Common support -obj-y := io.o id.o irq.o time.o serial.o devices.o +obj-y := io.o id.o clock.o irq.o time.o mux.o serial.o devices.o led-y := leds.o # Specific board support @@ -15,8 +15,9 @@ obj-$(CONFIG_MACH_OMAP_OSK) += board-osk.o obj-$(CONFIG_MACH_OMAP_H3) += board-h3.o obj-$(CONFIG_MACH_VOICEBLUE) += board-voiceblue.o obj-$(CONFIG_MACH_NETSTAR) += board-netstar.o +obj-$(CONFIG_MACH_OMAP_PALMTE) += board-palmte.o -ifeq ($(CONFIG_ARCH_OMAP1510),y) +ifeq ($(CONFIG_ARCH_OMAP15XX),y) # Innovator-1510 FPGA obj-$(CONFIG_MACH_OMAP_INNOVATOR) += fpga.o endif diff --git a/arch/arm/mach-omap1/board-generic.c b/arch/arm/mach-omap1/board-generic.c index c209c7172a9..4b292e93fbe 100644 --- a/arch/arm/mach-omap1/board-generic.c +++ b/arch/arm/mach-omap1/board-generic.c @@ -15,7 +15,7 @@ #include <linux/kernel.h> #include <linux/init.h> -#include <linux/device.h> +#include <linux/platform_device.h> #include <asm/hardware.h> #include <asm/mach-types.h> @@ -28,8 +28,6 @@ #include <asm/arch/board.h> #include <asm/arch/common.h> -static int __initdata generic_serial_ports[OMAP_MAX_NR_PORTS] = {1, 1, 1}; - static void __init omap_generic_init_irq(void) { omap_init_irq(); @@ -37,7 +35,7 @@ static void __init omap_generic_init_irq(void) /* assume no Mini-AB port */ -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX static struct omap_usb_config generic1510_usb_config __initdata = { .register_host = 1, .register_dev = 1, @@ -76,21 +74,19 @@ static struct omap_mmc_config generic_mmc_config __initdata = { #endif +static struct omap_uart_config generic_uart_config __initdata = { + .enabled_uarts = ((1 << 0) | (1 << 1) | (1 << 2)), +}; + static struct omap_board_config_kernel generic_config[] = { { OMAP_TAG_USB, NULL }, { OMAP_TAG_MMC, &generic_mmc_config }, + { OMAP_TAG_UART, &generic_uart_config }, }; static void __init omap_generic_init(void) { - const struct omap_uart_config *uart_conf; - - /* - * Make sure the serial ports are muxed on at this point. - * You have to mux them off in device drivers later on - * if not needed. - */ -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX if (cpu_is_omap1510()) { generic_config[0].data = &generic1510_usb_config; } @@ -101,20 +97,9 @@ static void __init omap_generic_init(void) } #endif - uart_conf = omap_get_config(OMAP_TAG_UART, struct omap_uart_config); - if (uart_conf != NULL) { - unsigned int enabled_ports, i; - - enabled_ports = uart_conf->enabled_uarts; - for (i = 0; i < 3; i++) { - if (!(enabled_ports & (1 << i))) - generic_serial_ports[i] = 0; - } - } - omap_board_config = generic_config; omap_board_config_size = ARRAY_SIZE(generic_config); - omap_serial_init(generic_serial_ports); + omap_serial_init(); } static void __init omap_generic_map_io(void) diff --git a/arch/arm/mach-omap1/board-h2.c b/arch/arm/mach-omap1/board-h2.c index 4ee6bd8a50b..a07e2c9307f 100644 --- a/arch/arm/mach-omap1/board-h2.c +++ b/arch/arm/mach-omap1/board-h2.c @@ -40,8 +40,6 @@ extern int omap_gpio_init(void); -static int __initdata h2_serial_ports[OMAP_MAX_NR_PORTS] = {1, 1, 1}; - static struct mtd_partition h2_partitions[] = { /* bootloader (U-Boot, etc) in first sector */ { @@ -160,9 +158,20 @@ static struct omap_mmc_config h2_mmc_config __initdata = { }, }; +static struct omap_uart_config h2_uart_config __initdata = { + .enabled_uarts = ((1 << 0) | (1 << 1) | (1 << 2)), +}; + +static struct omap_lcd_config h2_lcd_config __initdata = { + .panel_name = "h2", + .ctrl_name = "internal", +}; + static struct omap_board_config_kernel h2_config[] = { { OMAP_TAG_USB, &h2_usb_config }, { OMAP_TAG_MMC, &h2_mmc_config }, + { OMAP_TAG_UART, &h2_uart_config }, + { OMAP_TAG_LCD, &h2_lcd_config }, }; static void __init h2_init(void) @@ -180,12 +189,12 @@ static void __init h2_init(void) platform_add_devices(h2_devices, ARRAY_SIZE(h2_devices)); omap_board_config = h2_config; omap_board_config_size = ARRAY_SIZE(h2_config); + omap_serial_init(); } static void __init h2_map_io(void) { omap_map_common_io(); - omap_serial_init(h2_serial_ports); } MACHINE_START(OMAP_H2, "TI-H2") diff --git a/arch/arm/mach-omap1/board-h3.c b/arch/arm/mach-omap1/board-h3.c index fc824361430..668e278433c 100644 --- a/arch/arm/mach-omap1/board-h3.c +++ b/arch/arm/mach-omap1/board-h3.c @@ -41,8 +41,6 @@ extern int omap_gpio_init(void); -static int __initdata h3_serial_ports[OMAP_MAX_NR_PORTS] = {1, 1, 1}; - static struct mtd_partition h3_partitions[] = { /* bootloader (U-Boot, etc) in first sector */ { @@ -168,9 +166,20 @@ static struct omap_mmc_config h3_mmc_config __initdata = { }, }; +static struct omap_uart_config h3_uart_config __initdata = { + .enabled_uarts = ((1 << 0) | (1 << 1) | (1 << 2)), +}; + +static struct omap_lcd_config h3_lcd_config __initdata = { + .panel_name = "h3", + .ctrl_name = "internal", +}; + static struct omap_board_config_kernel h3_config[] = { - { OMAP_TAG_USB, &h3_usb_config }, - { OMAP_TAG_MMC, &h3_mmc_config }, + { OMAP_TAG_USB, &h3_usb_config }, + { OMAP_TAG_MMC, &h3_mmc_config }, + { OMAP_TAG_UART, &h3_uart_config }, + { OMAP_TAG_LCD, &h3_lcd_config }, }; static void __init h3_init(void) @@ -180,6 +189,7 @@ static void __init h3_init(void) (void) platform_add_devices(devices, ARRAY_SIZE(devices)); omap_board_config = h3_config; omap_board_config_size = ARRAY_SIZE(h3_config); + omap_serial_init(); } static void __init h3_init_smc91x(void) @@ -201,7 +211,6 @@ void h3_init_irq(void) static void __init h3_map_io(void) { omap_map_common_io(); - omap_serial_init(h3_serial_ports); } MACHINE_START(OMAP_H3, "TI OMAP1710 H3 board") diff --git a/arch/arm/mach-omap1/board-innovator.c b/arch/arm/mach-omap1/board-innovator.c index a2eac853b2d..95f1ff36cdc 100644 --- a/arch/arm/mach-omap1/board-innovator.c +++ b/arch/arm/mach-omap1/board-innovator.c @@ -36,8 +36,6 @@ #include <asm/arch/usb.h> #include <asm/arch/common.h> -static int __initdata innovator_serial_ports[OMAP_MAX_NR_PORTS] = {1, 1, 1}; - static struct mtd_partition innovator_partitions[] = { /* bootloader (U-Boot, etc) in first sector */ { @@ -99,7 +97,7 @@ static struct platform_device innovator_flash_device = { .resource = &innovator_flash_resource, }; -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX /* Only FPGA needs to be mapped here. All others are done with ioremap */ static struct map_desc innovator1510_io_desc[] __initdata = { @@ -136,7 +134,7 @@ static struct platform_device *innovator1510_devices[] __initdata = { &innovator1510_smc91x_device, }; -#endif /* CONFIG_ARCH_OMAP1510 */ +#endif /* CONFIG_ARCH_OMAP15XX */ #ifdef CONFIG_ARCH_OMAP16XX @@ -185,7 +183,7 @@ void innovator_init_irq(void) { omap_init_irq(); omap_gpio_init(); -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX if (cpu_is_omap1510()) { omap1510_fpga_init_irq(); } @@ -193,7 +191,7 @@ void innovator_init_irq(void) innovator_init_smc91x(); } -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX static struct omap_usb_config innovator1510_usb_config __initdata = { /* for bundled non-standard host and peripheral cables */ .hmc_mode = 4, @@ -205,6 +203,11 @@ static struct omap_usb_config innovator1510_usb_config __initdata = { .register_dev = 1, .pins[0] = 2, }; + +static struct omap_lcd_config innovator1510_lcd_config __initdata = { + .panel_name = "inn1510", + .ctrl_name = "internal", +}; #endif #ifdef CONFIG_ARCH_OMAP16XX @@ -222,6 +225,11 @@ static struct omap_usb_config h2_usb_config __initdata = { .pins[1] = 3, }; + +static struct omap_lcd_config innovator1610_lcd_config __initdata = { + .panel_name = "inn1610", + .ctrl_name = "internal", +}; #endif static struct omap_mmc_config innovator_mmc_config __initdata = { @@ -234,14 +242,20 @@ static struct omap_mmc_config innovator_mmc_config __initdata = { }, }; +static struct omap_uart_config innovator_uart_config __initdata = { + .enabled_uarts = ((1 << 0) | (1 << 1) | (1 << 2)), +}; + static struct omap_board_config_kernel innovator_config[] = { { OMAP_TAG_USB, NULL }, + { OMAP_TAG_LCD, NULL }, { OMAP_TAG_MMC, &innovator_mmc_config }, + { OMAP_TAG_UART, &innovator_uart_config }, }; static void __init innovator_init(void) { -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX if (cpu_is_omap1510()) { platform_add_devices(innovator1510_devices, ARRAY_SIZE(innovator1510_devices)); } @@ -252,23 +266,28 @@ static void __init innovator_init(void) } #endif -#ifdef CONFIG_ARCH_OMAP1510 - if (cpu_is_omap1510()) +#ifdef CONFIG_ARCH_OMAP15XX + if (cpu_is_omap1510()) { innovator_config[0].data = &innovator1510_usb_config; + innovator_config[1].data = &innovator1510_lcd_config; + } #endif #ifdef CONFIG_ARCH_OMAP16XX - if (cpu_is_omap1610()) + if (cpu_is_omap1610()) { innovator_config[0].data = &h2_usb_config; + innovator_config[1].data = &innovator1610_lcd_config; + } #endif omap_board_config = innovator_config; omap_board_config_size = ARRAY_SIZE(innovator_config); + omap_serial_init(); } static void __init innovator_map_io(void) { omap_map_common_io(); -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX if (cpu_is_omap1510()) { iotable_init(innovator1510_io_desc, ARRAY_SIZE(innovator1510_io_desc)); udelay(10); /* Delay needed for FPGA */ @@ -280,7 +299,6 @@ static void __init innovator_map_io(void) fpga_read(OMAP1510_FPGA_BOARD_REV)); } #endif - omap_serial_init(innovator_serial_ports); } MACHINE_START(OMAP_INNOVATOR, "TI-Innovator") diff --git a/arch/arm/mach-omap1/board-netstar.c b/arch/arm/mach-omap1/board-netstar.c index c851c2e4dfc..0448fa7de8a 100644 --- a/arch/arm/mach-omap1/board-netstar.c +++ b/arch/arm/mach-omap1/board-netstar.c @@ -55,6 +55,14 @@ static struct platform_device *netstar_devices[] __initdata = { &netstar_smc91x_device, }; +static struct omap_uart_config netstar_uart_config __initdata = { + .enabled_uarts = ((1 << 0) | (1 << 1) | (1 << 2)), +}; + +static struct omap_board_config_kernel netstar_config[] = { + { OMAP_TAG_UART, &netstar_uart_config }, +}; + static void __init netstar_init_irq(void) { omap_init_irq(); @@ -92,14 +100,15 @@ static void __init netstar_init(void) /* Switch off red LED */ omap_writeb(0x00, OMAP_LPG1_PMR); /* Disable clock */ omap_writeb(0x80, OMAP_LPG1_LCR); -} -static int __initdata omap_serial_ports[OMAP_MAX_NR_PORTS] = {1, 1, 1}; + omap_board_config = netstar_config; + omap_board_config_size = ARRAY_SIZE(netstar_config); + omap_serial_init(); +} static void __init netstar_map_io(void) { omap_map_common_io(); - omap_serial_init(omap_serial_ports); } #define MACHINE_PANICED 1 diff --git a/arch/arm/mach-omap1/board-osk.c b/arch/arm/mach-omap1/board-osk.c index a88524e7c31..e990e1bc166 100644 --- a/arch/arm/mach-omap1/board-osk.c +++ b/arch/arm/mach-omap1/board-osk.c @@ -46,8 +46,6 @@ #include <asm/arch/tc.h> #include <asm/arch/common.h> -static int __initdata osk_serial_ports[OMAP_MAX_NR_PORTS] = {1, 0, 0}; - static struct mtd_partition osk_partitions[] = { /* bootloader (U-Boot, etc) in first sector */ { @@ -155,7 +153,7 @@ static void __init osk_init_smc91x(void) } /* Check EMIFS wait states to fix errors with SMC_GET_PKT_HDR */ - EMIFS_CCS(1) |= 0x2; + EMIFS_CCS(1) |= 0x3; } static void __init osk_init_cf(void) @@ -193,8 +191,19 @@ static struct omap_usb_config osk_usb_config __initdata = { .pins[0] = 2, }; +static struct omap_uart_config osk_uart_config __initdata = { + .enabled_uarts = (1 << 0), +}; + +static struct omap_lcd_config osk_lcd_config __initdata = { + .panel_name = "osk", + .ctrl_name = "internal", +}; + static struct omap_board_config_kernel osk_config[] = { { OMAP_TAG_USB, &osk_usb_config }, + { OMAP_TAG_UART, &osk_uart_config }, + { OMAP_TAG_LCD, &osk_lcd_config }, }; #ifdef CONFIG_OMAP_OSK_MISTRAL @@ -254,13 +263,13 @@ static void __init osk_init(void) omap_board_config_size = ARRAY_SIZE(osk_config); USB_TRANSCEIVER_CTRL_REG |= (3 << 1); + omap_serial_init(); osk_mistral_init(); } static void __init osk_map_io(void) { omap_map_common_io(); - omap_serial_init(osk_serial_ports); } MACHINE_START(OMAP_OSK, "TI-OSK") diff --git a/arch/arm/mach-omap1/board-palmte.c b/arch/arm/mach-omap1/board-palmte.c new file mode 100644 index 00000000000..540b20d78cc --- /dev/null +++ b/arch/arm/mach-omap1/board-palmte.c @@ -0,0 +1,87 @@ +/* + * linux/arch/arm/mach-omap1/board-palmte.c + * + * Modified from board-generic.c + * + * Support for the Palm Tungsten E PDA. + * + * Original version : Laurent Gonzalez + * + * Maintainters : http://palmtelinux.sf.net + * palmtelinux-developpers@lists.sf.net + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/platform_device.h> +#include <linux/notifier.h> + +#include <asm/hardware.h> +#include <asm/mach-types.h> +#include <asm/mach/arch.h> +#include <asm/mach/map.h> + +#include <asm/arch/gpio.h> +#include <asm/arch/mux.h> +#include <asm/arch/usb.h> +#include <asm/arch/board.h> +#include <asm/arch/common.h> +#include <asm/hardware/clock.h> + +static void __init omap_generic_init_irq(void) +{ + omap_init_irq(); +} + +static struct omap_usb_config palmte_usb_config __initdata = { + .register_dev = 1, + .hmc_mode = 0, + .pins[0] = 3, +}; + +static struct omap_mmc_config palmte_mmc_config __initdata = { + .mmc [0] = { + .enabled = 1, + .wire4 = 1, + .wp_pin = OMAP_MPUIO(3), + .power_pin = -1, + .switch_pin = -1, + }, +}; + +static struct omap_lcd_config palmte_lcd_config __initdata = { + .panel_name = "palmte", + .ctrl_name = "internal", +}; + +static struct omap_board_config_kernel palmte_config[] = { + { OMAP_TAG_USB, &palmte_usb_config }, + { OMAP_TAG_MMC, &palmte_mmc_config }, + { OMAP_TAG_LCD, &palmte_lcd_config }, +}; + +static void __init omap_generic_init(void) +{ + omap_board_config = palmte_config; + omap_board_config_size = ARRAY_SIZE(palmte_config); +} + +static void __init omap_generic_map_io(void) +{ + omap_map_common_io(); +} + +MACHINE_START(OMAP_PALMTE, "OMAP310 based Palm Tungsten E") + .phys_ram = 0x10000000, + .phys_io = 0xfff00000, + .io_pg_offst = ((0xfef00000) >> 18) & 0xfffc, + .boot_params = 0x10000100, + .map_io = omap_generic_map_io, + .init_irq = omap_generic_init_irq, + .init_machine = omap_generic_init, + .timer = &omap_timer, +MACHINE_END diff --git a/arch/arm/mach-omap1/board-perseus2.c b/arch/arm/mach-omap1/board-perseus2.c index 354b157acb3..bd900b7ab33 100644 --- a/arch/arm/mach-omap1/board-perseus2.c +++ b/arch/arm/mach-omap1/board-perseus2.c @@ -29,6 +29,7 @@ #include <asm/arch/mux.h> #include <asm/arch/fpga.h> #include <asm/arch/common.h> +#include <asm/arch/board.h> static struct resource smc91x_resources[] = { [0] = { @@ -43,8 +44,6 @@ static struct resource smc91x_resources[] = { }, }; -static int __initdata p2_serial_ports[OMAP_MAX_NR_PORTS] = {1, 1, 0}; - static struct mtd_partition p2_partitions[] = { /* bootloader (U-Boot, etc) in first sector */ { @@ -111,9 +110,27 @@ static struct platform_device *devices[] __initdata = { &smc91x_device, }; +static struct omap_uart_config perseus2_uart_config __initdata = { + .enabled_uarts = ((1 << 0) | (1 << 1)), +}; + +static struct omap_lcd_config perseus2_lcd_config __initdata = { + .panel_name = "p2", + .ctrl_name = "internal", +}; + +static struct omap_board_config_kernel perseus2_config[] = { + { OMAP_TAG_UART, &perseus2_uart_config }, + { OMAP_TAG_LCD, &perseus2_lcd_config }, +}; + static void __init omap_perseus2_init(void) { (void) platform_add_devices(devices, ARRAY_SIZE(devices)); + + omap_board_config = perseus2_config; + omap_board_config_size = ARRAY_SIZE(perseus2_config); + omap_serial_init(); } static void __init perseus2_init_smc91x(void) @@ -131,7 +148,6 @@ void omap_perseus2_init_irq(void) omap_gpio_init(); perseus2_init_smc91x(); } - /* Only FPGA needs to be mapped here. All others are done with ioremap */ static struct map_desc omap_perseus2_io_desc[] __initdata = { { @@ -179,7 +195,6 @@ static void __init omap_perseus2_map_io(void) * It is used as the Ethernet controller interrupt */ omap_writel(omap_readl(OMAP730_IO_CONF_9) & 0x1FFFFFFF, OMAP730_IO_CONF_9); - omap_serial_init(p2_serial_ports); } MACHINE_START(OMAP_PERSEUS2, "OMAP730 Perseus2") diff --git a/arch/arm/mach-omap1/board-voiceblue.c b/arch/arm/mach-omap1/board-voiceblue.c index 3f018b29686..6f9a6220e78 100644 --- a/arch/arm/mach-omap1/board-voiceblue.c +++ b/arch/arm/mach-omap1/board-voiceblue.c @@ -150,9 +150,14 @@ static struct omap_mmc_config voiceblue_mmc_config __initdata = { }, }; +static struct omap_uart_config voiceblue_uart_config __initdata = { + .enabled_uarts = ((1 << 0) | (1 << 1) | (1 << 2)), +}; + static struct omap_board_config_kernel voiceblue_config[] = { { OMAP_TAG_USB, &voiceblue_usb_config }, { OMAP_TAG_MMC, &voiceblue_mmc_config }, + { OMAP_TAG_UART, &voiceblue_uart_config }, }; static void __init voiceblue_init_irq(void) @@ -191,6 +196,7 @@ static void __init voiceblue_init(void) platform_add_devices(voiceblue_devices, ARRAY_SIZE(voiceblue_devices)); omap_board_config = voiceblue_config; omap_board_config_size = ARRAY_SIZE(voiceblue_config); + omap_serial_init(); /* There is a good chance board is going up, so enable power LED * (it is connected through invertor) */ @@ -198,12 +204,9 @@ static void __init voiceblue_init(void) omap_writeb(0x00, OMAP_LPG1_PMR); /* Disable clock */ } -static int __initdata omap_serial_ports[OMAP_MAX_NR_PORTS] = {1, 1, 1}; - static void __init voiceblue_map_io(void) { omap_map_common_io(); - omap_serial_init(omap_serial_ports); } #define MACHINE_PANICED 1 diff --git a/arch/arm/mach-omap1/clock.c b/arch/arm/mach-omap1/clock.c new file mode 100644 index 00000000000..4277eee44ed --- /dev/null +++ b/arch/arm/mach-omap1/clock.c @@ -0,0 +1,792 @@ +/* + * linux/arch/arm/mach-omap1/clock.c + * + * Copyright (C) 2004 - 2005 Nokia corporation + * Written by Tuukka Tikkanen <tuukka.tikkanen@elektrobit.com> + * + * Modified to use omap shared clock framework by + * Tony Lindgren <tony@atomide.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/list.h> +#include <linux/errno.h> +#include <linux/err.h> + +#include <asm/io.h> +#include <asm/hardware/clock.h> + +#include <asm/arch/usb.h> +#include <asm/arch/clock.h> +#include <asm/arch/sram.h> + +#include "clock.h" + +__u32 arm_idlect1_mask; + +/*------------------------------------------------------------------------- + * Omap1 specific clock functions + *-------------------------------------------------------------------------*/ + +static void omap1_watchdog_recalc(struct clk * clk) +{ + clk->rate = clk->parent->rate / 14; +} + +static void omap1_uart_recalc(struct clk * clk) +{ + unsigned int val = omap_readl(clk->enable_reg); + if (val & clk->enable_bit) + clk->rate = 48000000; + else + clk->rate = 12000000; +} + +static int omap1_clk_enable_dsp_domain(struct clk *clk) +{ + int retval; + + retval = omap1_clk_use(&api_ck.clk); + if (!retval) { + retval = omap1_clk_enable(clk); + omap1_clk_unuse(&api_ck.clk); + } + + return retval; +} + +static void omap1_clk_disable_dsp_domain(struct clk *clk) +{ + if (omap1_clk_use(&api_ck.clk) == 0) { + omap1_clk_disable(clk); + omap1_clk_unuse(&api_ck.clk); + } +} + +static int omap1_clk_enable_uart_functional(struct clk *clk) +{ + int ret; + struct uart_clk *uclk; + + ret = omap1_clk_enable(clk); + if (ret == 0) { + /* Set smart idle acknowledgement mode */ + uclk = (struct uart_clk *)clk; + omap_writeb((omap_readb(uclk->sysc_addr) & ~0x10) | 8, + uclk->sysc_addr); + } + + return ret; +} + +static void omap1_clk_disable_uart_functional(struct clk *clk) +{ + struct uart_clk *uclk; + + /* Set force idle acknowledgement mode */ + uclk = (struct uart_clk *)clk; + omap_writeb((omap_readb(uclk->sysc_addr) & ~0x18), uclk->sysc_addr); + + omap1_clk_disable(clk); +} + +static void omap1_clk_allow_idle(struct clk *clk) +{ + struct arm_idlect1_clk * iclk = (struct arm_idlect1_clk *)clk; + + if (!(clk->flags & CLOCK_IDLE_CONTROL)) + return; + + if (iclk->no_idle_count > 0 && !(--iclk->no_idle_count)) + arm_idlect1_mask |= 1 << iclk->idlect_shift; +} + +static void omap1_clk_deny_idle(struct clk *clk) +{ + struct arm_idlect1_clk * iclk = (struct arm_idlect1_clk *)clk; + + if (!(clk->flags & CLOCK_IDLE_CONTROL)) + return; + + if (iclk->no_idle_count++ == 0) + arm_idlect1_mask &= ~(1 << iclk->idlect_shift); +} + +static __u16 verify_ckctl_value(__u16 newval) +{ + /* This function checks for following limitations set + * by the hardware (all conditions must be true): + * DSPMMU_CK == DSP_CK or DSPMMU_CK == DSP_CK/2 + * ARM_CK >= TC_CK + * DSP_CK >= TC_CK + * DSPMMU_CK >= TC_CK + * + * In addition following rules are enforced: + * LCD_CK <= TC_CK + * ARMPER_CK <= TC_CK + * + * However, maximum frequencies are not checked for! + */ + __u8 per_exp; + __u8 lcd_exp; + __u8 arm_exp; + __u8 dsp_exp; + __u8 tc_exp; + __u8 dspmmu_exp; + + per_exp = (newval >> CKCTL_PERDIV_OFFSET) & 3; + lcd_exp = (newval >> CKCTL_LCDDIV_OFFSET) & 3; + arm_exp = (newval >> CKCTL_ARMDIV_OFFSET) & 3; + dsp_exp = (newval >> CKCTL_DSPDIV_OFFSET) & 3; + tc_exp = (newval >> CKCTL_TCDIV_OFFSET) & 3; + dspmmu_exp = (newval >> CKCTL_DSPMMUDIV_OFFSET) & 3; + + if (dspmmu_exp < dsp_exp) + dspmmu_exp = dsp_exp; + if (dspmmu_exp > dsp_exp+1) + dspmmu_exp = dsp_exp+1; + if (tc_exp < arm_exp) + tc_exp = arm_exp; + if (tc_exp < dspmmu_exp) + tc_exp = dspmmu_exp; + if (tc_exp > lcd_exp) + lcd_exp = tc_exp; + if (tc_exp > per_exp) + per_exp = tc_exp; + + newval &= 0xf000; + newval |= per_exp << CKCTL_PERDIV_OFFSET; + newval |= lcd_exp << CKCTL_LCDDIV_OFFSET; + newval |= arm_exp << CKCTL_ARMDIV_OFFSET; + newval |= dsp_exp << CKCTL_DSPDIV_OFFSET; + newval |= tc_exp << CKCTL_TCDIV_OFFSET; + newval |= dspmmu_exp << CKCTL_DSPMMUDIV_OFFSET; + + return newval; +} + +static int calc_dsor_exp(struct clk *clk, unsigned long rate) +{ + /* Note: If target frequency is too low, this function will return 4, + * which is invalid value. Caller must check for this value and act + * accordingly. + * + * Note: This function does not check for following limitations set + * by the hardware (all conditions must be true): + * DSPMMU_CK == DSP_CK or DSPMMU_CK == DSP_CK/2 + * ARM_CK >= TC_CK + * DSP_CK >= TC_CK + * DSPMMU_CK >= TC_CK + */ + unsigned long realrate; + struct clk * parent; + unsigned dsor_exp; + + if (unlikely(!(clk->flags & RATE_CKCTL))) + return -EINVAL; + + parent = clk->parent; + if (unlikely(parent == 0)) + return -EIO; + + realrate = parent->rate; + for (dsor_exp=0; dsor_exp<4; dsor_exp++) { + if (realrate <= rate) + break; + + realrate /= 2; + } + + return dsor_exp; +} + +static void omap1_ckctl_recalc(struct clk * clk) +{ + int dsor; + + /* Calculate divisor encoded as 2-bit exponent */ + dsor = 1 << (3 & (omap_readw(ARM_CKCTL) >> clk->rate_offset)); + + if (unlikely(clk->rate == clk->parent->rate / dsor)) + return; /* No change, quick exit */ + clk->rate = clk->parent->rate / dsor; + + if (unlikely(clk->flags & RATE_PROPAGATES)) + propagate_rate(clk); +} + +static void omap1_ckctl_recalc_dsp_domain(struct clk * clk) +{ + int dsor; + + /* Calculate divisor encoded as 2-bit exponent + * + * The clock control bits are in DSP domain, + * so api_ck is needed for access. + * Note that DSP_CKCTL virt addr = phys addr, so + * we must use __raw_readw() instead of omap_readw(). + */ + omap1_clk_use(&api_ck.clk); + dsor = 1 << (3 & (__raw_readw(DSP_CKCTL) >> clk->rate_offset)); + omap1_clk_unuse(&api_ck.clk); + + if (unlikely(clk->rate == clk->parent->rate / dsor)) + return; /* No change, quick exit */ + clk->rate = clk->parent->rate / dsor; + + if (unlikely(clk->flags & RATE_PROPAGATES)) + propagate_rate(clk); +} + +/* MPU virtual clock functions */ +static int omap1_select_table_rate(struct clk * clk, unsigned long rate) +{ + /* Find the highest supported frequency <= rate and switch to it */ + struct mpu_rate * ptr; + + if (clk != &virtual_ck_mpu) + return -EINVAL; + + for (ptr = rate_table; ptr->rate; ptr++) { + if (ptr->xtal != ck_ref.rate) + continue; + + /* DPLL1 cannot be reprogrammed without risking system crash */ + if (likely(ck_dpll1.rate!=0) && ptr->pll_rate != ck_dpll1.rate) + continue; + + /* Can check only after xtal frequency check */ + if (ptr->rate <= rate) + break; + } + + if (!ptr->rate) + return -EINVAL; + + /* + * In most cases we should not need to reprogram DPLL. + * Reprogramming the DPLL is tricky, it must be done from SRAM. + */ + omap_sram_reprogram_clock(ptr->dpllctl_val, ptr->ckctl_val); + + ck_dpll1.rate = ptr->pll_rate; + propagate_rate(&ck_dpll1); + return 0; +} + +static int omap1_clk_set_rate_dsp_domain(struct clk *clk, unsigned long rate) +{ + int ret = -EINVAL; + int dsor_exp; + __u16 regval; + + if (clk->flags & RATE_CKCTL) { + dsor_exp = calc_dsor_exp(clk, rate); + if (dsor_exp > 3) + dsor_exp = -EINVAL; + if (dsor_exp < 0) + return dsor_exp; + + regval = __raw_readw(DSP_CKCTL); + regval &= ~(3 << clk->rate_offset); + regval |= dsor_exp << clk->rate_offset; + __raw_writew(regval, DSP_CKCTL); + clk->rate = clk->parent->rate / (1 << dsor_exp); + ret = 0; + } + + if (unlikely(ret == 0 && (clk->flags & RATE_PROPAGATES))) + propagate_rate(clk); + + return ret; +} + +static long omap1_round_to_table_rate(struct clk * clk, unsigned long rate) +{ + /* Find the highest supported frequency <= rate */ + struct mpu_rate * ptr; + long highest_rate; + + if (clk != &virtual_ck_mpu) + return -EINVAL; + + highest_rate = -EINVAL; + + for (ptr = rate_table; ptr->rate; ptr++) { + if (ptr->xtal != ck_ref.rate) + continue; + + highest_rate = ptr->rate; + + /* Can check only after xtal frequency check */ + if (ptr->rate <= rate) + break; + } + + return highest_rate; +} + +static unsigned calc_ext_dsor(unsigned long rate) +{ + unsigned dsor; + + /* MCLK and BCLK divisor selection is not linear: + * freq = 96MHz / dsor + * + * RATIO_SEL range: dsor <-> RATIO_SEL + * 0..6: (RATIO_SEL+2) <-> (dsor-2) + * 6..48: (8+(RATIO_SEL-6)*2) <-> ((dsor-8)/2+6) + * Minimum dsor is 2 and maximum is 96. Odd divisors starting from 9 + * can not be used. + */ + for (dsor = 2; dsor < 96; ++dsor) { + if ((dsor & 1) && dsor > 8) + continue; + if (rate >= 96000000 / dsor) + break; + } + return dsor; +} + +/* Only needed on 1510 */ +static int omap1_set_uart_rate(struct clk * clk, unsigned long rate) +{ + unsigned int val; + + val = omap_readl(clk->enable_reg); + if (rate == 12000000) + val &= ~(1 << clk->enable_bit); + else if (rate == 48000000) + val |= (1 << clk->enable_bit); + else + return -EINVAL; + omap_writel(val, clk->enable_reg); + clk->rate = rate; + + return 0; +} + +/* External clock (MCLK & BCLK) functions */ +static int omap1_set_ext_clk_rate(struct clk * clk, unsigned long rate) +{ + unsigned dsor; + __u16 ratio_bits; + + dsor = calc_ext_dsor(rate); + clk->rate = 96000000 / dsor; + if (dsor > 8) + ratio_bits = ((dsor - 8) / 2 + 6) << 2; + else + ratio_bits = (dsor - 2) << 2; + + ratio_bits |= omap_readw(clk->enable_reg) & ~0xfd; + omap_writew(ratio_bits, clk->enable_reg); + + return 0; +} + +static long omap1_round_ext_clk_rate(struct clk * clk, unsigned long rate) +{ + return 96000000 / calc_ext_dsor(rate); +} + +static void omap1_init_ext_clk(struct clk * clk) +{ + unsigned dsor; + __u16 ratio_bits; + + /* Determine current rate and ensure clock is based on 96MHz APLL */ + ratio_bits = omap_readw(clk->enable_reg) & ~1; + omap_writew(ratio_bits, clk->enable_reg); + + ratio_bits = (ratio_bits & 0xfc) >> 2; + if (ratio_bits > 6) + dsor = (ratio_bits - 6) * 2 + 8; + else + dsor = ratio_bits + 2; + + clk-> rate = 96000000 / dsor; +} + +static int omap1_clk_use(struct clk *clk) +{ + int ret = 0; + if (clk->usecount++ == 0) { + if (likely(clk->parent)) { + ret = omap1_clk_use(clk->parent); + + if (unlikely(ret != 0)) { + clk->usecount--; + return ret; + } + + if (clk->flags & CLOCK_NO_IDLE_PARENT) + if (!cpu_is_omap24xx()) + omap1_clk_deny_idle(clk->parent); + } + + ret = clk->enable(clk); + + if (unlikely(ret != 0) && clk->parent) { + omap1_clk_unuse(clk->parent); + clk->usecount--; + } + } + + return ret; +} + +static void omap1_clk_unuse(struct clk *clk) +{ + if (clk->usecount > 0 && !(--clk->usecount)) { + clk->disable(clk); + if (likely(clk->parent)) { + omap1_clk_unuse(clk->parent); + if (clk->flags & CLOCK_NO_IDLE_PARENT) + if (!cpu_is_omap24xx()) + omap1_clk_allow_idle(clk->parent); + } + } +} + +static int omap1_clk_enable(struct clk *clk) +{ + __u16 regval16; + __u32 regval32; + + if (clk->flags & ALWAYS_ENABLED) + return 0; + + if (unlikely(clk->enable_reg == 0)) { + printk(KERN_ERR "clock.c: Enable for %s without enable code\n", + clk->name); + return 0; + } + + if (clk->flags & ENABLE_REG_32BIT) { + if (clk->flags & VIRTUAL_IO_ADDRESS) { + regval32 = __raw_readl(clk->enable_reg); + regval32 |= (1 << clk->enable_bit); + __raw_writel(regval32, clk->enable_reg); + } else { + regval32 = omap_readl(clk->enable_reg); + regval32 |= (1 << clk->enable_bit); + omap_writel(regval32, clk->enable_reg); + } + } else { + if (clk->flags & VIRTUAL_IO_ADDRESS) { + regval16 = __raw_readw(clk->enable_reg); + regval16 |= (1 << clk->enable_bit); + __raw_writew(regval16, clk->enable_reg); + } else { + regval16 = omap_readw(clk->enable_reg); + regval16 |= (1 << clk->enable_bit); + omap_writew(regval16, clk->enable_reg); + } + } + + return 0; +} + +static void omap1_clk_disable(struct clk *clk) +{ + __u16 regval16; + __u32 regval32; + + if (clk->enable_reg == 0) + return; + + if (clk->flags & ENABLE_REG_32BIT) { + if (clk->flags & VIRTUAL_IO_ADDRESS) { + regval32 = __raw_readl(clk->enable_reg); + regval32 &= ~(1 << clk->enable_bit); + __raw_writel(regval32, clk->enable_reg); + } else { + regval32 = omap_readl(clk->enable_reg); + regval32 &= ~(1 << clk->enable_bit); + omap_writel(regval32, clk->enable_reg); + } + } else { + if (clk->flags & VIRTUAL_IO_ADDRESS) { + regval16 = __raw_readw(clk->enable_reg); + regval16 &= ~(1 << clk->enable_bit); + __raw_writew(regval16, clk->enable_reg); + } else { + regval16 = omap_readw(clk->enable_reg); + regval16 &= ~(1 << clk->enable_bit); + omap_writew(regval16, clk->enable_reg); + } + } +} + +static long omap1_clk_round_rate(struct clk *clk, unsigned long rate) +{ + int dsor_exp; + + if (clk->flags & RATE_FIXED) + return clk->rate; + + if (clk->flags & RATE_CKCTL) { + dsor_exp = calc_dsor_exp(clk, rate); + if (dsor_exp < 0) + return dsor_exp; + if (dsor_exp > 3) + dsor_exp = 3; + return clk->parent->rate / (1 << dsor_exp); + } + + if(clk->round_rate != 0) + return clk->round_rate(clk, rate); + + return clk->rate; +} + +static int omap1_clk_set_rate(struct clk *clk, unsigned long rate) +{ + int ret = -EINVAL; + int dsor_exp; + __u16 regval; + + if (clk->set_rate) + ret = clk->set_rate(clk, rate); + else if (clk->flags & RATE_CKCTL) { + dsor_exp = calc_dsor_exp(clk, rate); + if (dsor_exp > 3) + dsor_exp = -EINVAL; + if (dsor_exp < 0) + return dsor_exp; + + regval = omap_readw(ARM_CKCTL); + regval &= ~(3 << clk->rate_offset); + regval |= dsor_exp << clk->rate_offset; + regval = verify_ckctl_value(regval); + omap_writew(regval, ARM_CKCTL); + clk->rate = clk->parent->rate / (1 << dsor_exp); + ret = 0; + } + + if (unlikely(ret == 0 && (clk->flags & RATE_PROPAGATES))) + propagate_rate(clk); + + return ret; +} + +/*------------------------------------------------------------------------- + * Omap1 clock reset and init functions + *-------------------------------------------------------------------------*/ + +#ifdef CONFIG_OMAP_RESET_CLOCKS +/* + * Resets some clocks that may be left on from bootloader, + * but leaves serial clocks on. See also omap_late_clk_reset(). + */ +static inline void omap1_early_clk_reset(void) +{ + //omap_writel(0x3 << 29, MOD_CONF_CTRL_0); +} + +static int __init omap1_late_clk_reset(void) +{ + /* Turn off all unused clocks */ + struct clk *p; + __u32 regval32; + + /* USB_REQ_EN will be disabled later if necessary (usb_dc_ck) */ + regval32 = omap_readw(SOFT_REQ_REG) & (1 << 4); + omap_writew(regval32, SOFT_REQ_REG); + omap_writew(0, SOFT_REQ_REG2); + + list_for_each_entry(p, &clocks, node) { + if (p->usecount > 0 || (p->flags & ALWAYS_ENABLED) || + p->enable_reg == 0) + continue; + + /* Clocks in the DSP domain need api_ck. Just assume bootloader + * has not enabled any DSP clocks */ + if ((u32)p->enable_reg == DSP_IDLECT2) { + printk(KERN_INFO "Skipping reset check for DSP domain " + "clock \"%s\"\n", p->name); + continue; + } + + /* Is the clock already disabled? */ + if (p->flags & ENABLE_REG_32BIT) { + if (p->flags & VIRTUAL_IO_ADDRESS) + regval32 = __raw_readl(p->enable_reg); + else + regval32 = omap_readl(p->enable_reg); + } else { + if (p->flags & VIRTUAL_IO_ADDRESS) + regval32 = __raw_readw(p->enable_reg); + else + regval32 = omap_readw(p->enable_reg); + } + + if ((regval32 & (1 << p->enable_bit)) == 0) + continue; + + /* FIXME: This clock seems to be necessary but no-one + * has asked for its activation. */ + if (p == &tc2_ck // FIX: pm.c (SRAM), CCP, Camera + || p == &ck_dpll1out.clk // FIX: SoSSI, SSR + || p == &arm_gpio_ck // FIX: GPIO code for 1510 + ) { + printk(KERN_INFO "FIXME: Clock \"%s\" seems unused\n", + p->name); + continue; + } + + printk(KERN_INFO "Disabling unused clock \"%s\"... ", p->name); + p->disable(p); + printk(" done\n"); + } + + return 0; +} +late_initcall(omap1_late_clk_reset); + +#else +#define omap1_early_clk_reset() {} +#endif + +static struct clk_functions omap1_clk_functions = { + .clk_use = omap1_clk_use, + .clk_unuse = omap1_clk_unuse, + .clk_round_rate = omap1_clk_round_rate, + .clk_set_rate = omap1_clk_set_rate, +}; + +int __init omap1_clk_init(void) +{ + struct clk ** clkp; + const struct omap_clock_config *info; + int crystal_type = 0; /* Default 12 MHz */ + + omap1_early_clk_reset(); + clk_init(&omap1_clk_functions); + + /* By default all idlect1 clocks are allowed to idle */ + arm_idlect1_mask = ~0; + + for (clkp = onchip_clks; clkp < onchip_clks+ARRAY_SIZE(onchip_clks); clkp++) { + if (((*clkp)->flags &CLOCK_IN_OMAP1510) && cpu_is_omap1510()) { + clk_register(*clkp); + continue; + } + + if (((*clkp)->flags &CLOCK_IN_OMAP16XX) && cpu_is_omap16xx()) { + clk_register(*clkp); + continue; + } + + if (((*clkp)->flags &CLOCK_IN_OMAP730) && cpu_is_omap730()) { + clk_register(*clkp); + continue; + } + } + + info = omap_get_config(OMAP_TAG_CLOCK, struct omap_clock_config); + if (info != NULL) { + if (!cpu_is_omap1510()) + crystal_type = info->system_clock_type; + } + +#if defined(CONFIG_ARCH_OMAP730) + ck_ref.rate = 13000000; +#elif defined(CONFIG_ARCH_OMAP16XX) + if (crystal_type == 2) + ck_ref.rate = 19200000; +#endif + + printk("Clocks: ARM_SYSST: 0x%04x DPLL_CTL: 0x%04x ARM_CKCTL: 0x%04x\n", + omap_readw(ARM_SYSST), omap_readw(DPLL_CTL), + omap_readw(ARM_CKCTL)); + + /* We want to be in syncronous scalable mode */ + omap_writew(0x1000, ARM_SYSST); + +#ifdef CONFIG_OMAP_CLOCKS_SET_BY_BOOTLOADER + /* Use values set by bootloader. Determine PLL rate and recalculate + * dependent clocks as if kernel had changed PLL or divisors. + */ + { + unsigned pll_ctl_val = omap_readw(DPLL_CTL); + + ck_dpll1.rate = ck_ref.rate; /* Base xtal rate */ + if (pll_ctl_val & 0x10) { + /* PLL enabled, apply multiplier and divisor */ + if (pll_ctl_val & 0xf80) + ck_dpll1.rate *= (pll_ctl_val & 0xf80) >> 7; + ck_dpll1.rate /= ((pll_ctl_val & 0x60) >> 5) + 1; + } else { + /* PLL disabled, apply bypass divisor */ + switch (pll_ctl_val & 0xc) { + case 0: + break; + case 0x4: + ck_dpll1.rate /= 2; + break; + default: + ck_dpll1.rate /= 4; + break; + } + } + } + propagate_rate(&ck_dpll1); +#else + /* Find the highest supported frequency and enable it */ + if (omap1_select_table_rate(&virtual_ck_mpu, ~0)) { + printk(KERN_ERR "System frequencies not set. Check your config.\n"); + /* Guess sane values (60MHz) */ + omap_writew(0x2290, DPLL_CTL); + omap_writew(0x1005, ARM_CKCTL); + ck_dpll1.rate = 60000000; + propagate_rate(&ck_dpll1); + } +#endif + /* Cache rates for clocks connected to ck_ref (not dpll1) */ + propagate_rate(&ck_ref); + printk(KERN_INFO "Clocking rate (xtal/DPLL1/MPU): " + "%ld.%01ld/%ld.%01ld/%ld.%01ld MHz\n", + ck_ref.rate / 1000000, (ck_ref.rate / 100000) % 10, + ck_dpll1.rate / 1000000, (ck_dpll1.rate / 100000) % 10, + arm_ck.rate / 1000000, (arm_ck.rate / 100000) % 10); + +#ifdef CONFIG_MACH_OMAP_PERSEUS2 + /* Select slicer output as OMAP input clock */ + omap_writew(omap_readw(OMAP730_PCC_UPLD_CTRL) & ~0x1, OMAP730_PCC_UPLD_CTRL); +#endif + + /* Turn off DSP and ARM_TIMXO. Make sure ARM_INTHCK is not divided */ + omap_writew(omap_readw(ARM_CKCTL) & 0x0fff, ARM_CKCTL); + + /* Put DSP/MPUI into reset until needed */ + omap_writew(0, ARM_RSTCT1); + omap_writew(1, ARM_RSTCT2); + omap_writew(0x400, ARM_IDLECT1); + + /* + * According to OMAP5910 Erratum SYS_DMA_1, bit DMACK_REQ (bit 8) + * of the ARM_IDLECT2 register must be set to zero. The power-on + * default value of this bit is one. + */ + omap_writew(0x0000, ARM_IDLECT2); /* Turn LCD clock off also */ + + /* + * Only enable those clocks we will need, let the drivers + * enable other clocks as necessary + */ + clk_use(&armper_ck.clk); + clk_use(&armxor_ck.clk); + clk_use(&armtim_ck.clk); /* This should be done by timer code */ + + if (cpu_is_omap1510()) + clk_enable(&arm_gpio_ck); + + return 0; +} + diff --git a/arch/arm/mach-omap1/clock.h b/arch/arm/mach-omap1/clock.h new file mode 100644 index 00000000000..f3bdfb50e01 --- /dev/null +++ b/arch/arm/mach-omap1/clock.h @@ -0,0 +1,768 @@ +/* + * linux/arch/arm/mach-omap1/clock.h + * + * Copyright (C) 2004 - 2005 Nokia corporation + * Written by Tuukka Tikkanen <tuukka.tikkanen@elektrobit.com> + * Based on clocks.h by Tony Lindgren, Gordon McNutt and RidgeRun, Inc + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#ifndef __ARCH_ARM_MACH_OMAP1_CLOCK_H +#define __ARCH_ARM_MACH_OMAP1_CLOCK_H + +static int omap1_clk_enable(struct clk * clk); +static void omap1_clk_disable(struct clk * clk); +static void omap1_ckctl_recalc(struct clk * clk); +static void omap1_watchdog_recalc(struct clk * clk); +static void omap1_ckctl_recalc_dsp_domain(struct clk * clk); +static int omap1_clk_enable_dsp_domain(struct clk * clk); +static int omap1_clk_set_rate_dsp_domain(struct clk * clk, unsigned long rate); +static void omap1_clk_disable_dsp_domain(struct clk * clk); +static int omap1_set_uart_rate(struct clk * clk, unsigned long rate); +static void omap1_uart_recalc(struct clk * clk); +static int omap1_clk_enable_uart_functional(struct clk * clk); +static void omap1_clk_disable_uart_functional(struct clk * clk); +static int omap1_set_ext_clk_rate(struct clk * clk, unsigned long rate); +static long omap1_round_ext_clk_rate(struct clk * clk, unsigned long rate); +static void omap1_init_ext_clk(struct clk * clk); +static int omap1_select_table_rate(struct clk * clk, unsigned long rate); +static long omap1_round_to_table_rate(struct clk * clk, unsigned long rate); +static int omap1_clk_use(struct clk *clk); +static void omap1_clk_unuse(struct clk *clk); + +struct mpu_rate { + unsigned long rate; + unsigned long xtal; + unsigned long pll_rate; + __u16 ckctl_val; + __u16 dpllctl_val; +}; + +struct uart_clk { + struct clk clk; + unsigned long sysc_addr; +}; + +/* Provide a method for preventing idling some ARM IDLECT clocks */ +struct arm_idlect1_clk { + struct clk clk; + unsigned long no_idle_count; + __u8 idlect_shift; +}; + +/* ARM_CKCTL bit shifts */ +#define CKCTL_PERDIV_OFFSET 0 +#define CKCTL_LCDDIV_OFFSET 2 +#define CKCTL_ARMDIV_OFFSET 4 +#define CKCTL_DSPDIV_OFFSET 6 +#define CKCTL_TCDIV_OFFSET 8 +#define CKCTL_DSPMMUDIV_OFFSET 10 +/*#define ARM_TIMXO 12*/ +#define EN_DSPCK 13 +/*#define ARM_INTHCK_SEL 14*/ /* Divide-by-2 for mpu inth_ck */ +/* DSP_CKCTL bit shifts */ +#define CKCTL_DSPPERDIV_OFFSET 0 + +/* ARM_IDLECT2 bit shifts */ +#define EN_WDTCK 0 +#define EN_XORPCK 1 +#define EN_PERCK 2 +#define EN_LCDCK 3 +#define EN_LBCK 4 /* Not on 1610/1710 */ +/*#define EN_HSABCK 5*/ +#define EN_APICK 6 +#define EN_TIMCK 7 +#define DMACK_REQ 8 +#define EN_GPIOCK 9 /* Not on 1610/1710 */ +/*#define EN_LBFREECK 10*/ +#define EN_CKOUT_ARM 11 + +/* ARM_IDLECT3 bit shifts */ +#define EN_OCPI_CK 0 +#define EN_TC1_CK 2 +#define EN_TC2_CK 4 + +/* DSP_IDLECT2 bit shifts (0,1,2 are same as for ARM_IDLECT2) */ +#define EN_DSPTIMCK 5 + +/* Various register defines for clock controls scattered around OMAP chip */ +#define USB_MCLK_EN_BIT 4 /* In ULPD_CLKC_CTRL */ +#define USB_HOST_HHC_UHOST_EN 9 /* In MOD_CONF_CTRL_0 */ +#define SWD_ULPD_PLL_CLK_REQ 1 /* In SWD_CLK_DIV_CTRL_SEL */ +#define COM_ULPD_PLL_CLK_REQ 1 /* In COM_CLK_DIV_CTRL_SEL */ +#define SWD_CLK_DIV_CTRL_SEL 0xfffe0874 +#define COM_CLK_DIV_CTRL_SEL 0xfffe0878 +#define SOFT_REQ_REG 0xfffe0834 +#define SOFT_REQ_REG2 0xfffe0880 + +/*------------------------------------------------------------------------- + * Omap1 MPU rate table + *-------------------------------------------------------------------------*/ +static struct mpu_rate rate_table[] = { + /* MPU MHz, xtal MHz, dpll1 MHz, CKCTL, DPLL_CTL + * NOTE: Comment order here is different from bits in CKCTL value: + * armdiv, dspdiv, dspmmu, tcdiv, perdiv, lcddiv + */ +#if defined(CONFIG_OMAP_ARM_216MHZ) + { 216000000, 12000000, 216000000, 0x050d, 0x2910 }, /* 1/1/2/2/2/8 */ +#endif +#if defined(CONFIG_OMAP_ARM_195MHZ) + { 195000000, 13000000, 195000000, 0x050e, 0x2790 }, /* 1/1/2/2/4/8 */ +#endif +#if defined(CONFIG_OMAP_ARM_192MHZ) + { 192000000, 19200000, 192000000, 0x050f, 0x2510 }, /* 1/1/2/2/8/8 */ + { 192000000, 12000000, 192000000, 0x050f, 0x2810 }, /* 1/1/2/2/8/8 */ + { 96000000, 12000000, 192000000, 0x055f, 0x2810 }, /* 2/2/2/2/8/8 */ + { 48000000, 12000000, 192000000, 0x0baf, 0x2810 }, /* 4/4/4/8/8/8 */ + { 24000000, 12000000, 192000000, 0x0fff, 0x2810 }, /* 8/8/8/8/8/8 */ +#endif +#if defined(CONFIG_OMAP_ARM_182MHZ) + { 182000000, 13000000, 182000000, 0x050e, 0x2710 }, /* 1/1/2/2/4/8 */ +#endif +#if defined(CONFIG_OMAP_ARM_168MHZ) + { 168000000, 12000000, 168000000, 0x010f, 0x2710 }, /* 1/1/1/2/8/8 */ +#endif +#if defined(CONFIG_OMAP_ARM_150MHZ) + { 150000000, 12000000, 150000000, 0x010a, 0x2cb0 }, /* 1/1/1/2/4/4 */ +#endif +#if defined(CONFIG_OMAP_ARM_120MHZ) + { 120000000, 12000000, 120000000, 0x010a, 0x2510 }, /* 1/1/1/2/4/4 */ +#endif +#if defined(CONFIG_OMAP_ARM_96MHZ) + { 96000000, 12000000, 96000000, 0x0005, 0x2410 }, /* 1/1/1/1/2/2 */ +#endif +#if defined(CONFIG_OMAP_ARM_60MHZ) + { 60000000, 12000000, 60000000, 0x0005, 0x2290 }, /* 1/1/1/1/2/2 */ +#endif +#if defined(CONFIG_OMAP_ARM_30MHZ) + { 30000000, 12000000, 60000000, 0x0555, 0x2290 }, /* 2/2/2/2/2/2 */ +#endif + { 0, 0, 0, 0, 0 }, +}; + +/*------------------------------------------------------------------------- + * Omap1 clocks + *-------------------------------------------------------------------------*/ + +static struct clk ck_ref = { + .name = "ck_ref", + .rate = 12000000, + .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | + ALWAYS_ENABLED, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk ck_dpll1 = { + .name = "ck_dpll1", + .parent = &ck_ref, + .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | + RATE_PROPAGATES | ALWAYS_ENABLED, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct arm_idlect1_clk ck_dpll1out = { + .clk = { + .name = "ck_dpll1out", + .parent = &ck_dpll1, + .flags = CLOCK_IN_OMAP16XX | CLOCK_IDLE_CONTROL, + .enable_reg = (void __iomem *)ARM_IDLECT2, + .enable_bit = EN_CKOUT_ARM, + .recalc = &followparent_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, + }, + .idlect_shift = 12, +}; + +static struct clk arm_ck = { + .name = "arm_ck", + .parent = &ck_dpll1, + .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | + RATE_CKCTL | RATE_PROPAGATES | ALWAYS_ENABLED, + .rate_offset = CKCTL_ARMDIV_OFFSET, + .recalc = &omap1_ckctl_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct arm_idlect1_clk armper_ck = { + .clk = { + .name = "armper_ck", + .parent = &ck_dpll1, + .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | + RATE_CKCTL | CLOCK_IDLE_CONTROL, + .enable_reg = (void __iomem *)ARM_IDLECT2, + .enable_bit = EN_PERCK, + .rate_offset = CKCTL_PERDIV_OFFSET, + .recalc = &omap1_ckctl_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, + }, + .idlect_shift = 2, +}; + +static struct clk arm_gpio_ck = { + .name = "arm_gpio_ck", + .parent = &ck_dpll1, + .flags = CLOCK_IN_OMAP1510, + .enable_reg = (void __iomem *)ARM_IDLECT2, + .enable_bit = EN_GPIOCK, + .recalc = &followparent_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct arm_idlect1_clk armxor_ck = { + .clk = { + .name = "armxor_ck", + .parent = &ck_ref, + .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | + CLOCK_IDLE_CONTROL, + .enable_reg = (void __iomem *)ARM_IDLECT2, + .enable_bit = EN_XORPCK, + .recalc = &followparent_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, + }, + .idlect_shift = 1, +}; + +static struct arm_idlect1_clk armtim_ck = { + .clk = { + .name = "armtim_ck", + .parent = &ck_ref, + .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | + CLOCK_IDLE_CONTROL, + .enable_reg = (void __iomem *)ARM_IDLECT2, + .enable_bit = EN_TIMCK, + .recalc = &followparent_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, + }, + .idlect_shift = 9, +}; + +static struct arm_idlect1_clk armwdt_ck = { + .clk = { + .name = "armwdt_ck", + .parent = &ck_ref, + .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | + CLOCK_IDLE_CONTROL, + .enable_reg = (void __iomem *)ARM_IDLECT2, + .enable_bit = EN_WDTCK, + .recalc = &omap1_watchdog_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, + }, + .idlect_shift = 0, +}; + +static struct clk arminth_ck16xx = { + .name = "arminth_ck", + .parent = &arm_ck, + .flags = CLOCK_IN_OMAP16XX | ALWAYS_ENABLED, + .recalc = &followparent_recalc, + /* Note: On 16xx the frequency can be divided by 2 by programming + * ARM_CKCTL:ARM_INTHCK_SEL(14) to 1 + * + * 1510 version is in TC clocks. + */ + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk dsp_ck = { + .name = "dsp_ck", + .parent = &ck_dpll1, + .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | + RATE_CKCTL, + .enable_reg = (void __iomem *)ARM_CKCTL, + .enable_bit = EN_DSPCK, + .rate_offset = CKCTL_DSPDIV_OFFSET, + .recalc = &omap1_ckctl_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk dspmmu_ck = { + .name = "dspmmu_ck", + .parent = &ck_dpll1, + .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | + RATE_CKCTL | ALWAYS_ENABLED, + .rate_offset = CKCTL_DSPMMUDIV_OFFSET, + .recalc = &omap1_ckctl_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk dspper_ck = { + .name = "dspper_ck", + .parent = &ck_dpll1, + .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | + RATE_CKCTL | VIRTUAL_IO_ADDRESS, + .enable_reg = (void __iomem *)DSP_IDLECT2, + .enable_bit = EN_PERCK, + .rate_offset = CKCTL_PERDIV_OFFSET, + .recalc = &omap1_ckctl_recalc_dsp_domain, + .set_rate = &omap1_clk_set_rate_dsp_domain, + .enable = &omap1_clk_enable_dsp_domain, + .disable = &omap1_clk_disable_dsp_domain, +}; + +static struct clk dspxor_ck = { + .name = "dspxor_ck", + .parent = &ck_ref, + .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | + VIRTUAL_IO_ADDRESS, + .enable_reg = (void __iomem *)DSP_IDLECT2, + .enable_bit = EN_XORPCK, + .recalc = &followparent_recalc, + .enable = &omap1_clk_enable_dsp_domain, + .disable = &omap1_clk_disable_dsp_domain, +}; + +static struct clk dsptim_ck = { + .name = "dsptim_ck", + .parent = &ck_ref, + .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | + VIRTUAL_IO_ADDRESS, + .enable_reg = (void __iomem *)DSP_IDLECT2, + .enable_bit = EN_DSPTIMCK, + .recalc = &followparent_recalc, + .enable = &omap1_clk_enable_dsp_domain, + .disable = &omap1_clk_disable_dsp_domain, +}; + +/* Tie ARM_IDLECT1:IDLIF_ARM to this logical clock structure */ +static struct arm_idlect1_clk tc_ck = { + .clk = { + .name = "tc_ck", + .parent = &ck_dpll1, + .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | + CLOCK_IN_OMAP730 | RATE_CKCTL | + RATE_PROPAGATES | ALWAYS_ENABLED | + CLOCK_IDLE_CONTROL, + .rate_offset = CKCTL_TCDIV_OFFSET, + .recalc = &omap1_ckctl_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, + }, + .idlect_shift = 6, +}; + +static struct clk arminth_ck1510 = { + .name = "arminth_ck", + .parent = &tc_ck.clk, + .flags = CLOCK_IN_OMAP1510 | ALWAYS_ENABLED, + .recalc = &followparent_recalc, + /* Note: On 1510 the frequency follows TC_CK + * + * 16xx version is in MPU clocks. + */ + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk tipb_ck = { + /* No-idle controlled by "tc_ck" */ + .name = "tibp_ck", + .parent = &tc_ck.clk, + .flags = CLOCK_IN_OMAP1510 | ALWAYS_ENABLED, + .recalc = &followparent_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk l3_ocpi_ck = { + /* No-idle controlled by "tc_ck" */ + .name = "l3_ocpi_ck", + .parent = &tc_ck.clk, + .flags = CLOCK_IN_OMAP16XX, + .enable_reg = (void __iomem *)ARM_IDLECT3, + .enable_bit = EN_OCPI_CK, + .recalc = &followparent_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk tc1_ck = { + .name = "tc1_ck", + .parent = &tc_ck.clk, + .flags = CLOCK_IN_OMAP16XX, + .enable_reg = (void __iomem *)ARM_IDLECT3, + .enable_bit = EN_TC1_CK, + .recalc = &followparent_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk tc2_ck = { + .name = "tc2_ck", + .parent = &tc_ck.clk, + .flags = CLOCK_IN_OMAP16XX, + .enable_reg = (void __iomem *)ARM_IDLECT3, + .enable_bit = EN_TC2_CK, + .recalc = &followparent_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk dma_ck = { + /* No-idle controlled by "tc_ck" */ + .name = "dma_ck", + .parent = &tc_ck.clk, + .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | + ALWAYS_ENABLED, + .recalc = &followparent_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk dma_lcdfree_ck = { + .name = "dma_lcdfree_ck", + .parent = &tc_ck.clk, + .flags = CLOCK_IN_OMAP16XX | ALWAYS_ENABLED, + .recalc = &followparent_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct arm_idlect1_clk api_ck = { + .clk = { + .name = "api_ck", + .parent = &tc_ck.clk, + .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | + CLOCK_IDLE_CONTROL, + .enable_reg = (void __iomem *)ARM_IDLECT2, + .enable_bit = EN_APICK, + .recalc = &followparent_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, + }, + .idlect_shift = 8, +}; + +static struct arm_idlect1_clk lb_ck = { + .clk = { + .name = "lb_ck", + .parent = &tc_ck.clk, + .flags = CLOCK_IN_OMAP1510 | CLOCK_IDLE_CONTROL, + .enable_reg = (void __iomem *)ARM_IDLECT2, + .enable_bit = EN_LBCK, + .recalc = &followparent_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, + }, + .idlect_shift = 4, +}; + +static struct clk rhea1_ck = { + .name = "rhea1_ck", + .parent = &tc_ck.clk, + .flags = CLOCK_IN_OMAP16XX | ALWAYS_ENABLED, + .recalc = &followparent_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk rhea2_ck = { + .name = "rhea2_ck", + .parent = &tc_ck.clk, + .flags = CLOCK_IN_OMAP16XX | ALWAYS_ENABLED, + .recalc = &followparent_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk lcd_ck_16xx = { + .name = "lcd_ck", + .parent = &ck_dpll1, + .flags = CLOCK_IN_OMAP16XX | CLOCK_IN_OMAP730 | RATE_CKCTL, + .enable_reg = (void __iomem *)ARM_IDLECT2, + .enable_bit = EN_LCDCK, + .rate_offset = CKCTL_LCDDIV_OFFSET, + .recalc = &omap1_ckctl_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct arm_idlect1_clk lcd_ck_1510 = { + .clk = { + .name = "lcd_ck", + .parent = &ck_dpll1, + .flags = CLOCK_IN_OMAP1510 | RATE_CKCTL | + CLOCK_IDLE_CONTROL, + .enable_reg = (void __iomem *)ARM_IDLECT2, + .enable_bit = EN_LCDCK, + .rate_offset = CKCTL_LCDDIV_OFFSET, + .recalc = &omap1_ckctl_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, + }, + .idlect_shift = 3, +}; + +static struct clk uart1_1510 = { + .name = "uart1_ck", + /* Direct from ULPD, no real parent */ + .parent = &armper_ck.clk, + .rate = 12000000, + .flags = CLOCK_IN_OMAP1510 | ENABLE_REG_32BIT | + ALWAYS_ENABLED | CLOCK_NO_IDLE_PARENT, + .enable_reg = (void __iomem *)MOD_CONF_CTRL_0, + .enable_bit = 29, /* Chooses between 12MHz and 48MHz */ + .set_rate = &omap1_set_uart_rate, + .recalc = &omap1_uart_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct uart_clk uart1_16xx = { + .clk = { + .name = "uart1_ck", + /* Direct from ULPD, no real parent */ + .parent = &armper_ck.clk, + .rate = 48000000, + .flags = CLOCK_IN_OMAP16XX | RATE_FIXED | + ENABLE_REG_32BIT | CLOCK_NO_IDLE_PARENT, + .enable_reg = (void __iomem *)MOD_CONF_CTRL_0, + .enable_bit = 29, + .enable = &omap1_clk_enable_uart_functional, + .disable = &omap1_clk_disable_uart_functional, + }, + .sysc_addr = 0xfffb0054, +}; + +static struct clk uart2_ck = { + .name = "uart2_ck", + /* Direct from ULPD, no real parent */ + .parent = &armper_ck.clk, + .rate = 12000000, + .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | + ENABLE_REG_32BIT | ALWAYS_ENABLED | + CLOCK_NO_IDLE_PARENT, + .enable_reg = (void __iomem *)MOD_CONF_CTRL_0, + .enable_bit = 30, /* Chooses between 12MHz and 48MHz */ + .set_rate = &omap1_set_uart_rate, + .recalc = &omap1_uart_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk uart3_1510 = { + .name = "uart3_ck", + /* Direct from ULPD, no real parent */ + .parent = &armper_ck.clk, + .rate = 12000000, + .flags = CLOCK_IN_OMAP1510 | ENABLE_REG_32BIT | + ALWAYS_ENABLED | CLOCK_NO_IDLE_PARENT, + .enable_reg = (void __iomem *)MOD_CONF_CTRL_0, + .enable_bit = 31, /* Chooses between 12MHz and 48MHz */ + .set_rate = &omap1_set_uart_rate, + .recalc = &omap1_uart_recalc, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct uart_clk uart3_16xx = { + .clk = { + .name = "uart3_ck", + /* Direct from ULPD, no real parent */ + .parent = &armper_ck.clk, + .rate = 48000000, + .flags = CLOCK_IN_OMAP16XX | RATE_FIXED | + ENABLE_REG_32BIT | CLOCK_NO_IDLE_PARENT, + .enable_reg = (void __iomem *)MOD_CONF_CTRL_0, + .enable_bit = 31, + .enable = &omap1_clk_enable_uart_functional, + .disable = &omap1_clk_disable_uart_functional, + }, + .sysc_addr = 0xfffb9854, +}; + +static struct clk usb_clko = { /* 6 MHz output on W4_USB_CLKO */ + .name = "usb_clko", + /* Direct from ULPD, no parent */ + .rate = 6000000, + .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | + RATE_FIXED | ENABLE_REG_32BIT, + .enable_reg = (void __iomem *)ULPD_CLOCK_CTRL, + .enable_bit = USB_MCLK_EN_BIT, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk usb_hhc_ck1510 = { + .name = "usb_hhc_ck", + /* Direct from ULPD, no parent */ + .rate = 48000000, /* Actually 2 clocks, 12MHz and 48MHz */ + .flags = CLOCK_IN_OMAP1510 | + RATE_FIXED | ENABLE_REG_32BIT, + .enable_reg = (void __iomem *)MOD_CONF_CTRL_0, + .enable_bit = USB_HOST_HHC_UHOST_EN, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk usb_hhc_ck16xx = { + .name = "usb_hhc_ck", + /* Direct from ULPD, no parent */ + .rate = 48000000, + /* OTG_SYSCON_2.OTG_PADEN == 0 (not 1510-compatible) */ + .flags = CLOCK_IN_OMAP16XX | + RATE_FIXED | ENABLE_REG_32BIT, + .enable_reg = (void __iomem *)OTG_BASE + 0x08 /* OTG_SYSCON_2 */, + .enable_bit = 8 /* UHOST_EN */, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk usb_dc_ck = { + .name = "usb_dc_ck", + /* Direct from ULPD, no parent */ + .rate = 48000000, + .flags = CLOCK_IN_OMAP16XX | RATE_FIXED, + .enable_reg = (void __iomem *)SOFT_REQ_REG, + .enable_bit = 4, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk mclk_1510 = { + .name = "mclk", + /* Direct from ULPD, no parent. May be enabled by ext hardware. */ + .rate = 12000000, + .flags = CLOCK_IN_OMAP1510 | RATE_FIXED, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk mclk_16xx = { + .name = "mclk", + /* Direct from ULPD, no parent. May be enabled by ext hardware. */ + .flags = CLOCK_IN_OMAP16XX, + .enable_reg = (void __iomem *)COM_CLK_DIV_CTRL_SEL, + .enable_bit = COM_ULPD_PLL_CLK_REQ, + .set_rate = &omap1_set_ext_clk_rate, + .round_rate = &omap1_round_ext_clk_rate, + .init = &omap1_init_ext_clk, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk bclk_1510 = { + .name = "bclk", + /* Direct from ULPD, no parent. May be enabled by ext hardware. */ + .rate = 12000000, + .flags = CLOCK_IN_OMAP1510 | RATE_FIXED, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk bclk_16xx = { + .name = "bclk", + /* Direct from ULPD, no parent. May be enabled by ext hardware. */ + .flags = CLOCK_IN_OMAP16XX, + .enable_reg = (void __iomem *)SWD_CLK_DIV_CTRL_SEL, + .enable_bit = SWD_ULPD_PLL_CLK_REQ, + .set_rate = &omap1_set_ext_clk_rate, + .round_rate = &omap1_round_ext_clk_rate, + .init = &omap1_init_ext_clk, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk mmc1_ck = { + .name = "mmc1_ck", + /* Functional clock is direct from ULPD, interface clock is ARMPER */ + .parent = &armper_ck.clk, + .rate = 48000000, + .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | + RATE_FIXED | ENABLE_REG_32BIT | CLOCK_NO_IDLE_PARENT, + .enable_reg = (void __iomem *)MOD_CONF_CTRL_0, + .enable_bit = 23, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk mmc2_ck = { + .name = "mmc2_ck", + /* Functional clock is direct from ULPD, interface clock is ARMPER */ + .parent = &armper_ck.clk, + .rate = 48000000, + .flags = CLOCK_IN_OMAP16XX | + RATE_FIXED | ENABLE_REG_32BIT | CLOCK_NO_IDLE_PARENT, + .enable_reg = (void __iomem *)MOD_CONF_CTRL_0, + .enable_bit = 20, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk virtual_ck_mpu = { + .name = "mpu", + .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | + VIRTUAL_CLOCK | ALWAYS_ENABLED, + .parent = &arm_ck, /* Is smarter alias for */ + .recalc = &followparent_recalc, + .set_rate = &omap1_select_table_rate, + .round_rate = &omap1_round_to_table_rate, + .enable = &omap1_clk_enable, + .disable = &omap1_clk_disable, +}; + +static struct clk * onchip_clks[] = { + /* non-ULPD clocks */ + &ck_ref, + &ck_dpll1, + /* CK_GEN1 clocks */ + &ck_dpll1out.clk, + &arm_ck, + &armper_ck.clk, + &arm_gpio_ck, + &armxor_ck.clk, + &armtim_ck.clk, + &armwdt_ck.clk, + &arminth_ck1510, &arminth_ck16xx, + /* CK_GEN2 clocks */ + &dsp_ck, + &dspmmu_ck, + &dspper_ck, + &dspxor_ck, + &dsptim_ck, + /* CK_GEN3 clocks */ + &tc_ck.clk, + &tipb_ck, + &l3_ocpi_ck, + &tc1_ck, + &tc2_ck, + &dma_ck, + &dma_lcdfree_ck, + &api_ck.clk, + &lb_ck.clk, + &rhea1_ck, + &rhea2_ck, + &lcd_ck_16xx, + &lcd_ck_1510.clk, + /* ULPD clocks */ + &uart1_1510, + &uart1_16xx.clk, + &uart2_ck, + &uart3_1510, + &uart3_16xx.clk, + &usb_clko, + &usb_hhc_ck1510, &usb_hhc_ck16xx, + &usb_dc_ck, + &mclk_1510, &mclk_16xx, + &bclk_1510, &bclk_16xx, + &mmc1_ck, + &mmc2_ck, + /* Virtual clocks */ + &virtual_ck_mpu, +}; + +#endif diff --git a/arch/arm/mach-omap1/devices.c b/arch/arm/mach-omap1/devices.c index 3c5d901efea..ecbc47514ad 100644 --- a/arch/arm/mach-omap1/devices.c +++ b/arch/arm/mach-omap1/devices.c @@ -25,56 +25,7 @@ #include <asm/arch/mux.h> #include <asm/arch/gpio.h> - -static void omap_nop_release(struct device *dev) -{ - /* Nothing */ -} - -/*-------------------------------------------------------------------------*/ - -#if defined(CONFIG_I2C_OMAP) || defined(CONFIG_I2C_OMAP_MODULE) - -#define OMAP_I2C_BASE 0xfffb3800 - -static struct resource i2c_resources[] = { - { - .start = OMAP_I2C_BASE, - .end = OMAP_I2C_BASE + 0x3f, - .flags = IORESOURCE_MEM, - }, - { - .start = INT_I2C, - .flags = IORESOURCE_IRQ, - }, -}; - -/* DMA not used; works around erratum writing to non-empty i2c fifo */ - -static struct platform_device omap_i2c_device = { - .name = "i2c_omap", - .id = -1, - .dev = { - .release = omap_nop_release, - }, - .num_resources = ARRAY_SIZE(i2c_resources), - .resource = i2c_resources, -}; - -static void omap_init_i2c(void) -{ - /* FIXME define and use a boot tag, in case of boards that - * either don't wire up I2C, or chips that mux it differently... - * it can include clocking and address info, maybe more. - */ - omap_cfg_reg(I2C_SCL); - omap_cfg_reg(I2C_SDA); - - (void) platform_device_register(&omap_i2c_device); -} -#else -static inline void omap_init_i2c(void) {} -#endif +extern void omap_nop_release(struct device *dev); /*-------------------------------------------------------------------------*/ @@ -110,137 +61,6 @@ static inline void omap_init_irda(void) {} /*-------------------------------------------------------------------------*/ -#if defined(CONFIG_MMC_OMAP) || defined(CONFIG_MMC_OMAP_MODULE) - -#define OMAP_MMC1_BASE 0xfffb7800 -#define OMAP_MMC2_BASE 0xfffb7c00 /* omap16xx only */ - -static struct omap_mmc_conf mmc1_conf; - -static u64 mmc1_dmamask = 0xffffffff; - -static struct resource mmc1_resources[] = { - { - .start = IO_ADDRESS(OMAP_MMC1_BASE), - .end = IO_ADDRESS(OMAP_MMC1_BASE) + 0x7f, - .flags = IORESOURCE_MEM, - }, - { - .start = INT_MMC, - .flags = IORESOURCE_IRQ, - }, -}; - -static struct platform_device mmc_omap_device1 = { - .name = "mmci-omap", - .id = 1, - .dev = { - .release = omap_nop_release, - .dma_mask = &mmc1_dmamask, - .platform_data = &mmc1_conf, - }, - .num_resources = ARRAY_SIZE(mmc1_resources), - .resource = mmc1_resources, -}; - -#ifdef CONFIG_ARCH_OMAP16XX - -static struct omap_mmc_conf mmc2_conf; - -static u64 mmc2_dmamask = 0xffffffff; - -static struct resource mmc2_resources[] = { - { - .start = IO_ADDRESS(OMAP_MMC2_BASE), - .end = IO_ADDRESS(OMAP_MMC2_BASE) + 0x7f, - .flags = IORESOURCE_MEM, - }, - { - .start = INT_1610_MMC2, - .flags = IORESOURCE_IRQ, - }, -}; - -static struct platform_device mmc_omap_device2 = { - .name = "mmci-omap", - .id = 2, - .dev = { - .release = omap_nop_release, - .dma_mask = &mmc2_dmamask, - .platform_data = &mmc2_conf, - }, - .num_resources = ARRAY_SIZE(mmc2_resources), - .resource = mmc2_resources, -}; -#endif - -static void __init omap_init_mmc(void) -{ - const struct omap_mmc_config *mmc_conf; - const struct omap_mmc_conf *mmc; - - /* NOTE: assumes MMC was never (wrongly) enabled */ - mmc_conf = omap_get_config(OMAP_TAG_MMC, struct omap_mmc_config); - if (!mmc_conf) - return; - - /* block 1 is always available and has just one pinout option */ - mmc = &mmc_conf->mmc[0]; - if (mmc->enabled) { - omap_cfg_reg(MMC_CMD); - omap_cfg_reg(MMC_CLK); - omap_cfg_reg(MMC_DAT0); - if (cpu_is_omap1710()) { - omap_cfg_reg(M15_1710_MMC_CLKI); - omap_cfg_reg(P19_1710_MMC_CMDDIR); - omap_cfg_reg(P20_1710_MMC_DATDIR0); - } - if (mmc->wire4) { - omap_cfg_reg(MMC_DAT1); - /* NOTE: DAT2 can be on W10 (here) or M15 */ - if (!mmc->nomux) - omap_cfg_reg(MMC_DAT2); - omap_cfg_reg(MMC_DAT3); - } - mmc1_conf = *mmc; - (void) platform_device_register(&mmc_omap_device1); - } - -#ifdef CONFIG_ARCH_OMAP16XX - /* block 2 is on newer chips, and has many pinout options */ - mmc = &mmc_conf->mmc[1]; - if (mmc->enabled) { - if (!mmc->nomux) { - omap_cfg_reg(Y8_1610_MMC2_CMD); - omap_cfg_reg(Y10_1610_MMC2_CLK); - omap_cfg_reg(R18_1610_MMC2_CLKIN); - omap_cfg_reg(W8_1610_MMC2_DAT0); - if (mmc->wire4) { - omap_cfg_reg(V8_1610_MMC2_DAT1); - omap_cfg_reg(W15_1610_MMC2_DAT2); - omap_cfg_reg(R10_1610_MMC2_DAT3); - } - - /* These are needed for the level shifter */ - omap_cfg_reg(V9_1610_MMC2_CMDDIR); - omap_cfg_reg(V5_1610_MMC2_DATDIR0); - omap_cfg_reg(W19_1610_MMC2_DATDIR1); - } - - /* Feedback clock must be set on OMAP-1710 MMC2 */ - if (cpu_is_omap1710()) - omap_writel(omap_readl(MOD_CONF_CTRL_1) | (1 << 24), - MOD_CONF_CTRL_1); - mmc2_conf = *mmc; - (void) platform_device_register(&mmc_omap_device2); - } -#endif - return; -} -#else -static inline void omap_init_mmc(void) {} -#endif - #if defined(CONFIG_OMAP_RTC) || defined(CONFIG_OMAP_RTC) #define OMAP_RTC_BASE 0xfffb4800 @@ -279,38 +99,6 @@ static void omap_init_rtc(void) static inline void omap_init_rtc(void) {} #endif -/*-------------------------------------------------------------------------*/ - -#if defined(CONFIG_OMAP16XX_WATCHDOG) || defined(CONFIG_OMAP16XX_WATCHDOG_MODULE) - -#define OMAP_WDT_BASE 0xfffeb000 - -static struct resource wdt_resources[] = { - { - .start = OMAP_WDT_BASE, - .end = OMAP_WDT_BASE + 0x4f, - .flags = IORESOURCE_MEM, - }, -}; - -static struct platform_device omap_wdt_device = { - .name = "omap1610_wdt", - .id = -1, - .dev = { - .release = omap_nop_release, - }, - .num_resources = ARRAY_SIZE(wdt_resources), - .resource = wdt_resources, -}; - -static void omap_init_wdt(void) -{ - (void) platform_device_register(&omap_wdt_device); -} -#else -static inline void omap_init_wdt(void) {} -#endif - /*-------------------------------------------------------------------------*/ @@ -334,18 +122,15 @@ static inline void omap_init_wdt(void) {} * may be handled by the boot loader, and drivers should expect it will * normally have been done by the time they're probed. */ -static int __init omap_init_devices(void) +static int __init omap1_init_devices(void) { /* please keep these calls, and their implementations above, * in alphabetical order so they're easier to sort through. */ - omap_init_i2c(); omap_init_irda(); - omap_init_mmc(); omap_init_rtc(); - omap_init_wdt(); return 0; } -arch_initcall(omap_init_devices); +arch_initcall(omap1_init_devices); diff --git a/arch/arm/mach-omap1/id.c b/arch/arm/mach-omap1/id.c index 986c3b7e09b..5c637c04836 100644 --- a/arch/arm/mach-omap1/id.c +++ b/arch/arm/mach-omap1/id.c @@ -18,6 +18,13 @@ #include <asm/io.h> +#define OMAP_DIE_ID_0 0xfffe1800 +#define OMAP_DIE_ID_1 0xfffe1804 +#define OMAP_PRODUCTION_ID_0 0xfffe2000 +#define OMAP_PRODUCTION_ID_1 0xfffe2004 +#define OMAP32_ID_0 0xfffed400 +#define OMAP32_ID_1 0xfffed404 + struct omap_id { u16 jtag_id; /* Used to determine OMAP type */ u8 die_rev; /* Processor revision */ @@ -27,6 +34,7 @@ struct omap_id { /* Register values to detect the OMAP version */ static struct omap_id omap_ids[] __initdata = { + { .jtag_id = 0xb574, .die_rev = 0x2, .omap_id = 0x03310315, .type = 0x03100000}, { .jtag_id = 0x355f, .die_rev = 0x0, .omap_id = 0x03320000, .type = 0x07300100}, { .jtag_id = 0xb55f, .die_rev = 0x0, .omap_id = 0x03320000, .type = 0x07300300}, { .jtag_id = 0xb470, .die_rev = 0x0, .omap_id = 0x03310100, .type = 0x15100000}, @@ -164,6 +172,7 @@ void __init omap_check_revision(void) case 0x07: system_rev |= 0x07; break; + case 0x03: case 0x15: system_rev |= 0x15; break; diff --git a/arch/arm/mach-omap1/io.c b/arch/arm/mach-omap1/io.c index 79fb86535eb..a7a19f75b9e 100644 --- a/arch/arm/mach-omap1/io.c +++ b/arch/arm/mach-omap1/io.c @@ -15,9 +15,10 @@ #include <asm/mach/map.h> #include <asm/io.h> +#include <asm/arch/mux.h> #include <asm/arch/tc.h> -extern int clk_init(void); +extern int omap1_clk_init(void); extern void omap_check_revision(void); extern void omap_sram_init(void); @@ -50,7 +51,7 @@ static struct map_desc omap730_io_desc[] __initdata = { }; #endif -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX static struct map_desc omap1510_io_desc[] __initdata = { { .virtual = OMAP1510_DSP_BASE, @@ -98,7 +99,7 @@ static void __init _omap_map_io(void) iotable_init(omap730_io_desc, ARRAY_SIZE(omap730_io_desc)); } #endif -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX if (cpu_is_omap1510()) { iotable_init(omap1510_io_desc, ARRAY_SIZE(omap1510_io_desc)); } @@ -119,7 +120,7 @@ static void __init _omap_map_io(void) /* Must init clocks early to assure that timer interrupt works */ - clk_init(); + omap1_clk_init(); } /* @@ -127,7 +128,9 @@ static void __init _omap_map_io(void) */ void __init omap_map_common_io(void) { - if (!initialized) + if (!initialized) { _omap_map_io(); + omap1_mux_init(); + } } diff --git a/arch/arm/mach-omap1/irq.c b/arch/arm/mach-omap1/irq.c index 192ce6055fa..ed65a7d2e94 100644 --- a/arch/arm/mach-omap1/irq.c +++ b/arch/arm/mach-omap1/irq.c @@ -47,6 +47,7 @@ #include <asm/irq.h> #include <asm/mach/irq.h> #include <asm/arch/gpio.h> +#include <asm/arch/cpu.h> #include <asm/io.h> @@ -147,11 +148,15 @@ static struct omap_irq_bank omap730_irq_banks[] = { }; #endif -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX static struct omap_irq_bank omap1510_irq_banks[] = { { .base_reg = OMAP_IH1_BASE, .trigger_map = 0xb3febfff }, { .base_reg = OMAP_IH2_BASE, .trigger_map = 0xffbfffed }, }; +static struct omap_irq_bank omap310_irq_banks[] = { + { .base_reg = OMAP_IH1_BASE, .trigger_map = 0xb3faefc3 }, + { .base_reg = OMAP_IH2_BASE, .trigger_map = 0x65b3c061 }, +}; #endif #if defined(CONFIG_ARCH_OMAP16XX) @@ -181,11 +186,15 @@ void __init omap_init_irq(void) irq_bank_count = ARRAY_SIZE(omap730_irq_banks); } #endif -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX if (cpu_is_omap1510()) { irq_banks = omap1510_irq_banks; irq_bank_count = ARRAY_SIZE(omap1510_irq_banks); } + if (cpu_is_omap310()) { + irq_banks = omap310_irq_banks; + irq_bank_count = ARRAY_SIZE(omap310_irq_banks); + } #endif #if defined(CONFIG_ARCH_OMAP16XX) if (cpu_is_omap16xx()) { @@ -226,9 +235,11 @@ void __init omap_init_irq(void) } /* Unmask level 2 handler */ - if (cpu_is_omap730()) { + + if (cpu_is_omap730()) omap_unmask_irq(INT_730_IH2_IRQ); - } else { - omap_unmask_irq(INT_IH2_IRQ); - } + else if (cpu_is_omap1510()) + omap_unmask_irq(INT_1510_IH2_IRQ); + else if (cpu_is_omap16xx()) + omap_unmask_irq(INT_1610_IH2_IRQ); } diff --git a/arch/arm/mach-omap1/leds-h2p2-debug.c b/arch/arm/mach-omap1/leds-h2p2-debug.c index 399010c1403..65065081591 100644 --- a/arch/arm/mach-omap1/leds-h2p2-debug.c +++ b/arch/arm/mach-omap1/leds-h2p2-debug.c @@ -18,6 +18,7 @@ #include <asm/hardware.h> #include <asm/leds.h> #include <asm/system.h> +#include <asm/mach-types.h> #include <asm/arch/fpga.h> #include <asm/arch/gpio.h> @@ -63,14 +64,19 @@ void h2p2_dbg_leds_event(led_event_t evt) case led_stop: case led_halted: /* all leds off during suspend or shutdown */ - omap_set_gpio_dataout(GPIO_TIMER, 0); - omap_set_gpio_dataout(GPIO_IDLE, 0); + + if (! machine_is_omap_perseus2()) { + omap_set_gpio_dataout(GPIO_TIMER, 0); + omap_set_gpio_dataout(GPIO_IDLE, 0); + } + __raw_writew(~0, &fpga->leds); led_state &= ~LED_STATE_ENABLED; if (evt == led_halted) { iounmap(fpga); fpga = NULL; } + goto done; case led_claim: @@ -85,18 +91,37 @@ void h2p2_dbg_leds_event(led_event_t evt) #ifdef CONFIG_LEDS_TIMER case led_timer: led_state ^= LED_TIMER_ON; - omap_set_gpio_dataout(GPIO_TIMER, led_state & LED_TIMER_ON); - goto done; + + if (machine_is_omap_perseus2()) + hw_led_state ^= H2P2_DBG_FPGA_P2_LED_TIMER; + else { + omap_set_gpio_dataout(GPIO_TIMER, led_state & LED_TIMER_ON); + goto done; + } + + break; #endif #ifdef CONFIG_LEDS_CPU case led_idle_start: - omap_set_gpio_dataout(GPIO_IDLE, 1); - goto done; + if (machine_is_omap_perseus2()) + hw_led_state |= H2P2_DBG_FPGA_P2_LED_IDLE; + else { + omap_set_gpio_dataout(GPIO_IDLE, 1); + goto done; + } + + break; case led_idle_end: - omap_set_gpio_dataout(GPIO_IDLE, 0); - goto done; + if (machine_is_omap_perseus2()) + hw_led_state &= ~H2P2_DBG_FPGA_P2_LED_IDLE; + else { + omap_set_gpio_dataout(GPIO_IDLE, 0); + goto done; + } + + break; #endif case led_green_on: @@ -135,7 +160,7 @@ void h2p2_dbg_leds_event(led_event_t evt) /* * Actually burn the LEDs */ - if (led_state & LED_STATE_CLAIMED) + if (led_state & LED_STATE_ENABLED) __raw_writew(~hw_led_state, &fpga->leds); done: diff --git a/arch/arm/mach-omap1/leds.c b/arch/arm/mach-omap1/leds.c index 5c6b1bb6e72..3f9dcac4fd4 100644 --- a/arch/arm/mach-omap1/leds.c +++ b/arch/arm/mach-omap1/leds.c @@ -33,7 +33,6 @@ omap_leds_init(void) if (machine_is_omap_h2() || machine_is_omap_h3() - || machine_is_omap_perseus2() #ifdef CONFIG_OMAP_OSK_MISTRAL || machine_is_omap_osk() #endif diff --git a/arch/arm/mach-omap1/mux.c b/arch/arm/mach-omap1/mux.c new file mode 100644 index 00000000000..d4b8d624e74 --- /dev/null +++ b/arch/arm/mach-omap1/mux.c @@ -0,0 +1,289 @@ +/* + * linux/arch/arm/mach-omap1/mux.c + * + * OMAP1 pin multiplexing configurations + * + * Copyright (C) 2003 - 2005 Nokia Corporation + * + * Written by Tony Lindgren <tony.lindgren@nokia.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ +#include <linux/config.h> +#include <linux/module.h> +#include <linux/init.h> +#include <asm/system.h> +#include <asm/io.h> +#include <linux/spinlock.h> + +#include <asm/arch/mux.h> + +#ifdef CONFIG_OMAP_MUX + +#ifdef CONFIG_ARCH_OMAP730 +struct pin_config __initdata_or_module omap730_pins[] = { +MUX_CFG_730("E2_730_KBR0", 12, 21, 0, 0, 20, 1, NA, 0, 0) +MUX_CFG_730("J7_730_KBR1", 12, 25, 0, 0, 24, 1, NA, 0, 0) +MUX_CFG_730("E1_730_KBR2", 12, 29, 0, 0, 28, 1, NA, 0, 0) +MUX_CFG_730("F3_730_KBR3", 13, 1, 0, 0, 0, 1, NA, 0, 0) +MUX_CFG_730("D2_730_KBR4", 13, 5, 0, 0, 4, 1, NA, 0, 0) +MUX_CFG_730("C2_730_KBC0", 13, 9, 0, 0, 8, 1, NA, 0, 0) +MUX_CFG_730("D3_730_KBC1", 13, 13, 0, 0, 12, 1, NA, 0, 0) +MUX_CFG_730("E4_730_KBC2", 13, 17, 0, 0, 16, 1, NA, 0, 0) +MUX_CFG_730("F4_730_KBC3", 13, 21, 0, 0, 20, 1, NA, 0, 0) +MUX_CFG_730("E3_730_KBC4", 13, 25, 0, 0, 24, 1, NA, 0, 0) +}; +#endif + +#if defined(CONFIG_ARCH_OMAP15XX) || defined(CONFIG_ARCH_OMAP16XX) +struct pin_config __initdata_or_module omap1xxx_pins[] = { +/* + * description mux mode mux pull pull pull pu_pd pu dbg + * reg offset mode reg bit ena reg + */ +MUX_CFG("UART1_TX", 9, 21, 1, 2, 3, 0, NA, 0, 0) +MUX_CFG("UART1_RTS", 9, 12, 1, 2, 0, 0, NA, 0, 0) + +/* UART2 (COM_UART_GATING), conflicts with USB2 */ +MUX_CFG("UART2_TX", C, 27, 1, 3, 3, 0, NA, 0, 0) +MUX_CFG("UART2_RX", C, 18, 0, 3, 1, 1, NA, 0, 0) +MUX_CFG("UART2_CTS", C, 21, 0, 3, 1, 1, NA, 0, 0) +MUX_CFG("UART2_RTS", C, 24, 1, 3, 2, 0, NA, 0, 0) + +/* UART3 (GIGA_UART_GATING) */ +MUX_CFG("UART3_TX", 6, 0, 1, 0, 30, 0, NA, 0, 0) +MUX_CFG("UART3_RX", 6, 3, 0, 0, 31, 1, NA, 0, 0) +MUX_CFG("UART3_CTS", 5, 12, 2, 0, 24, 0, NA, 0, 0) +MUX_CFG("UART3_RTS", 5, 15, 2, 0, 25, 0, NA, 0, 0) +MUX_CFG("UART3_CLKREQ", 9, 27, 0, 2, 5, 0, NA, 0, 0) +MUX_CFG("UART3_BCLK", A, 0, 0, 2, 6, 0, NA, 0, 0) +MUX_CFG("Y15_1610_UART3_RTS", A, 0, 1, 2, 6, 0, NA, 0, 0) + +/* PWT & PWL, conflicts with UART3 */ +MUX_CFG("PWT", 6, 0, 2, 0, 30, 0, NA, 0, 0) +MUX_CFG("PWL", 6, 3, 1, 0, 31, 1, NA, 0, 0) + +/* USB internal master generic */ +MUX_CFG("R18_USB_VBUS", 7, 9, 2, 1, 11, 0, NA, 0, 1) +MUX_CFG("R18_1510_USB_GPIO0", 7, 9, 0, 1, 11, 1, NA, 0, 1) +/* works around erratum: W4_USB_PUEN and W4_USB_PUDIS are switched! */ +MUX_CFG("W4_USB_PUEN", D, 3, 3, 3, 5, 1, NA, 0, 1) +MUX_CFG("W4_USB_CLKO", D, 3, 1, 3, 5, 0, NA, 0, 1) +MUX_CFG("W4_USB_HIGHZ", D, 3, 4, 3, 5, 0, 3, 0, 1) +MUX_CFG("W4_GPIO58", D, 3, 7, 3, 5, 0, 3, 0, 1) + +/* USB1 master */ +MUX_CFG("USB1_SUSP", 8, 27, 2, 1, 27, 0, NA, 0, 1) +MUX_CFG("USB1_SE0", 9, 0, 2, 1, 28, 0, NA, 0, 1) +MUX_CFG("W13_1610_USB1_SE0", 9, 0, 4, 1, 28, 0, NA, 0, 1) +MUX_CFG("USB1_TXEN", 9, 3, 2, 1, 29, 0, NA, 0, 1) +MUX_CFG("USB1_TXD", 9, 24, 1, 2, 4, 0, NA, 0, 1) +MUX_CFG("USB1_VP", A, 3, 1, 2, 7, 0, NA, 0, 1) +MUX_CFG("USB1_VM", A, 6, 1, 2, 8, 0, NA, 0, 1) +MUX_CFG("USB1_RCV", A, 9, 1, 2, 9, 0, NA, 0, 1) +MUX_CFG("USB1_SPEED", A, 12, 2, 2, 10, 0, NA, 0, 1) +MUX_CFG("R13_1610_USB1_SPEED", A, 12, 5, 2, 10, 0, NA, 0, 1) +MUX_CFG("R13_1710_USB1_SEO", A, 12, 5, 2, 10, 0, NA, 0, 1) + +/* USB2 master */ +MUX_CFG("USB2_SUSP", B, 3, 1, 2, 17, 0, NA, 0, 1) +MUX_CFG("USB2_VP", B, 6, 1, 2, 18, 0, NA, 0, 1) +MUX_CFG("USB2_TXEN", B, 9, 1, 2, 19, 0, NA, 0, 1) +MUX_CFG("USB2_VM", C, 18, 1, 3, 0, 0, NA, 0, 1) +MUX_CFG("USB2_RCV", C, 21, 1, 3, 1, 0, NA, 0, 1) +MUX_CFG("USB2_SE0", C, 24, 2, 3, 2, 0, NA, 0, 1) +MUX_CFG("USB2_TXD", C, 27, 2, 3, 3, 0, NA, 0, 1) + +/* OMAP-1510 GPIO */ +MUX_CFG("R18_1510_GPIO0", 7, 9, 0, 1, 11, 1, 0, 0, 1) +MUX_CFG("R19_1510_GPIO1", 7, 6, 0, 1, 10, 1, 0, 0, 1) +MUX_CFG("M14_1510_GPIO2", 7, 3, 0, 1, 9, 1, 0, 0, 1) + +/* OMAP1610 GPIO */ +MUX_CFG("P18_1610_GPIO3", 7, 0, 0, 1, 8, 0, NA, 0, 1) +MUX_CFG("Y15_1610_GPIO17", A, 0, 7, 2, 6, 0, NA, 0, 1) + +/* OMAP-1710 GPIO */ +MUX_CFG("R18_1710_GPIO0", 7, 9, 0, 1, 11, 1, 1, 1, 1) +MUX_CFG("V2_1710_GPIO10", F, 27, 1, 4, 3, 1, 4, 1, 1) +MUX_CFG("N21_1710_GPIO14", 6, 9, 0, 1, 1, 1, 1, 1, 1) +MUX_CFG("W15_1710_GPIO40", 9, 27, 7, 2, 5, 1, 2, 1, 1) + +/* MPUIO */ +MUX_CFG("MPUIO2", 7, 18, 0, 1, 14, 1, NA, 0, 1) +MUX_CFG("N15_1610_MPUIO2", 7, 18, 0, 1, 14, 1, 1, 0, 1) +MUX_CFG("MPUIO4", 7, 15, 0, 1, 13, 1, NA, 0, 1) +MUX_CFG("MPUIO5", 7, 12, 0, 1, 12, 1, NA, 0, 1) + +MUX_CFG("T20_1610_MPUIO5", 7, 12, 0, 1, 12, 0, 3, 0, 1) +MUX_CFG("W11_1610_MPUIO6", 10, 15, 2, 3, 8, 0, 3, 0, 1) +MUX_CFG("V10_1610_MPUIO7", A, 24, 2, 2, 14, 0, 2, 0, 1) +MUX_CFG("W11_1610_MPUIO9", 10, 15, 1, 3, 8, 0, 3, 0, 1) +MUX_CFG("V10_1610_MPUIO10", A, 24, 1, 2, 14, 0, 2, 0, 1) +MUX_CFG("W10_1610_MPUIO11", A, 18, 2, 2, 11, 0, 2, 0, 1) +MUX_CFG("E20_1610_MPUIO13", 3, 21, 1, 0, 7, 0, 0, 0, 1) +MUX_CFG("U20_1610_MPUIO14", 9, 6, 6, 0, 30, 0, 0, 0, 1) +MUX_CFG("E19_1610_MPUIO15", 3, 18, 1, 0, 6, 0, 0, 0, 1) + +/* MCBSP2 */ +MUX_CFG("MCBSP2_CLKR", C, 6, 0, 2, 27, 1, NA, 0, 1) +MUX_CFG("MCBSP2_CLKX", C, 9, 0, 2, 29, 1, NA, 0, 1) +MUX_CFG("MCBSP2_DR", C, 0, 0, 2, 26, 1, NA, 0, 1) +MUX_CFG("MCBSP2_DX", C, 15, 0, 2, 31, 1, NA, 0, 1) +MUX_CFG("MCBSP2_FSR", C, 12, 0, 2, 30, 1, NA, 0, 1) +MUX_CFG("MCBSP2_FSX", C, 3, 0, 2, 27, 1, NA, 0, 1) + +/* MCBSP3 NOTE: Mode must 1 for clock */ +MUX_CFG("MCBSP3_CLKX", 9, 3, 1, 1, 29, 0, NA, 0, 1) + +/* Misc ballouts */ +MUX_CFG("BALLOUT_V8_ARMIO3", B, 18, 0, 2, 25, 1, NA, 0, 1) +MUX_CFG("N20_HDQ", 6, 18, 1, 1, 4, 0, 1, 4, 0) + +/* OMAP-1610 MMC2 */ +MUX_CFG("W8_1610_MMC2_DAT0", B, 21, 6, 2, 23, 1, 2, 1, 1) +MUX_CFG("V8_1610_MMC2_DAT1", B, 27, 6, 2, 25, 1, 2, 1, 1) +MUX_CFG("W15_1610_MMC2_DAT2", 9, 12, 6, 2, 5, 1, 2, 1, 1) +MUX_CFG("R10_1610_MMC2_DAT3", B, 18, 6, 2, 22, 1, 2, 1, 1) +MUX_CFG("Y10_1610_MMC2_CLK", B, 3, 6, 2, 17, 0, 2, 0, 1) +MUX_CFG("Y8_1610_MMC2_CMD", B, 24, 6, 2, 24, 1, 2, 1, 1) +MUX_CFG("V9_1610_MMC2_CMDDIR", B, 12, 6, 2, 20, 0, 2, 1, 1) +MUX_CFG("V5_1610_MMC2_DATDIR0", B, 15, 6, 2, 21, 0, 2, 1, 1) +MUX_CFG("W19_1610_MMC2_DATDIR1", 8, 15, 6, 1, 23, 0, 1, 1, 1) +MUX_CFG("R18_1610_MMC2_CLKIN", 7, 9, 6, 1, 11, 0, 1, 11, 1) + +/* OMAP-1610 External Trace Interface */ +MUX_CFG("M19_1610_ETM_PSTAT0", 5, 27, 1, 0, 29, 0, 0, 0, 1) +MUX_CFG("L15_1610_ETM_PSTAT1", 5, 24, 1, 0, 28, 0, 0, 0, 1) +MUX_CFG("L18_1610_ETM_PSTAT2", 5, 21, 1, 0, 27, 0, 0, 0, 1) +MUX_CFG("L19_1610_ETM_D0", 5, 18, 1, 0, 26, 0, 0, 0, 1) +MUX_CFG("J19_1610_ETM_D6", 5, 0, 1, 0, 20, 0, 0, 0, 1) +MUX_CFG("J18_1610_ETM_D7", 5, 27, 1, 0, 19, 0, 0, 0, 1) + +/* OMAP16XX GPIO */ +MUX_CFG("P20_1610_GPIO4", 6, 27, 0, 1, 7, 0, 1, 1, 1) +MUX_CFG("V9_1610_GPIO7", B, 12, 1, 2, 20, 0, 2, 1, 1) +MUX_CFG("W8_1610_GPIO9", B, 21, 0, 2, 23, 0, 2, 1, 1) +MUX_CFG("N20_1610_GPIO11", 6, 18, 0, 1, 4, 0, 1, 1, 1) +MUX_CFG("N19_1610_GPIO13", 6, 12, 0, 1, 2, 0, 1, 1, 1) +MUX_CFG("P10_1610_GPIO22", C, 0, 7, 2, 26, 0, 2, 1, 1) +MUX_CFG("V5_1610_GPIO24", B, 15, 7, 2, 21, 0, 2, 1, 1) +MUX_CFG("AA20_1610_GPIO_41", 9, 9, 7, 1, 31, 0, 1, 1, 1) +MUX_CFG("W19_1610_GPIO48", 8, 15, 7, 1, 23, 1, 1, 0, 1) +MUX_CFG("M7_1610_GPIO62", 10, 0, 0, 4, 24, 0, 4, 0, 1) +MUX_CFG("V14_16XX_GPIO37", 9, 18, 7, 2, 2, 0, 2, 2, 0) +MUX_CFG("R9_16XX_GPIO18", C, 18, 7, 3, 0, 0, 3, 0, 0) +MUX_CFG("L14_16XX_GPIO49", 6, 3, 7, 0, 31, 0, 0, 31, 0) + +/* OMAP-1610 uWire */ +MUX_CFG("V19_1610_UWIRE_SCLK", 8, 6, 0, 1, 20, 0, 1, 1, 1) +MUX_CFG("U18_1610_UWIRE_SDI", 8, 0, 0, 1, 18, 0, 1, 1, 1) +MUX_CFG("W21_1610_UWIRE_SDO", 8, 3, 0, 1, 19, 0, 1, 1, 1) +MUX_CFG("N14_1610_UWIRE_CS0", 8, 9, 1, 1, 21, 0, 1, 1, 1) +MUX_CFG("P15_1610_UWIRE_CS3", 8, 12, 1, 1, 22, 0, 1, 1, 1) +MUX_CFG("N15_1610_UWIRE_CS1", 7, 18, 2, 1, 14, 0, NA, 0, 1) + +/* OMAP-1610 Flash */ +MUX_CFG("L3_1610_FLASH_CS2B_OE",10, 6, 1, NA, 0, 0, NA, 0, 1) +MUX_CFG("M8_1610_FLASH_CS2B_WE",10, 3, 1, NA, 0, 0, NA, 0, 1) + +/* First MMC interface, same on 1510, 1610 and 1710 */ +MUX_CFG("MMC_CMD", A, 27, 0, 2, 15, 1, 2, 1, 1) +MUX_CFG("MMC_DAT1", A, 24, 0, 2, 14, 1, 2, 1, 1) +MUX_CFG("MMC_DAT2", A, 18, 0, 2, 12, 1, 2, 1, 1) +MUX_CFG("MMC_DAT0", B, 0, 0, 2, 16, 1, 2, 1, 1) +MUX_CFG("MMC_CLK", A, 21, 0, NA, 0, 0, NA, 0, 1) +MUX_CFG("MMC_DAT3", 10, 15, 0, 3, 8, 1, 3, 1, 1) +MUX_CFG("M15_1710_MMC_CLKI", 6, 21, 2, 0, 0, 0, NA, 0, 1) +MUX_CFG("P19_1710_MMC_CMDDIR", 6, 24, 6, 0, 0, 0, NA, 0, 1) +MUX_CFG("P20_1710_MMC_DATDIR0", 6, 27, 5, 0, 0, 0, NA, 0, 1) + +/* OMAP-1610 USB0 alternate configuration */ +MUX_CFG("W9_USB0_TXEN", B, 9, 5, 2, 19, 0, 2, 0, 1) +MUX_CFG("AA9_USB0_VP", B, 6, 5, 2, 18, 0, 2, 0, 1) +MUX_CFG("Y5_USB0_RCV", C, 21, 5, 3, 1, 0, 1, 0, 1) +MUX_CFG("R9_USB0_VM", C, 18, 5, 3, 0, 0, 3, 0, 1) +MUX_CFG("V6_USB0_TXD", C, 27, 5, 3, 3, 0, 3, 0, 1) +MUX_CFG("W5_USB0_SE0", C, 24, 5, 3, 2, 0, 3, 0, 1) +MUX_CFG("V9_USB0_SPEED", B, 12, 5, 2, 20, 0, 2, 0, 1) +MUX_CFG("Y10_USB0_SUSP", B, 3, 5, 2, 17, 0, 2, 0, 1) + +/* USB2 interface */ +MUX_CFG("W9_USB2_TXEN", B, 9, 1, NA, 0, 0, NA, 0, 1) +MUX_CFG("AA9_USB2_VP", B, 6, 1, NA, 0, 0, NA, 0, 1) +MUX_CFG("Y5_USB2_RCV", C, 21, 1, NA, 0, 0, NA, 0, 1) +MUX_CFG("R9_USB2_VM", C, 18, 1, NA, 0, 0, NA, 0, 1) +MUX_CFG("V6_USB2_TXD", C, 27, 2, NA, 0, 0, NA, 0, 1) +MUX_CFG("W5_USB2_SE0", C, 24, 2, NA, 0, 0, NA, 0, 1) + +/* 16XX UART */ +MUX_CFG("R13_1610_UART1_TX", A, 12, 6, 2, 10, 0, 2, 10, 1) +MUX_CFG("V14_16XX_UART1_RX", 9, 18, 0, 2, 2, 0, 2, 2, 1) +MUX_CFG("R14_1610_UART1_CTS", 9, 15, 0, 2, 1, 0, 2, 1, 1) +MUX_CFG("AA15_1610_UART1_RTS", 9, 12, 1, 2, 0, 0, 2, 0, 1) +MUX_CFG("R9_16XX_UART2_RX", C, 18, 0, 3, 0, 0, 3, 0, 1) +MUX_CFG("L14_16XX_UART3_RX", 6, 3, 0, 0, 31, 0, 0, 31, 1) + +/* I2C interface */ +MUX_CFG("I2C_SCL", 7, 24, 0, NA, 0, 0, NA, 0, 0) +MUX_CFG("I2C_SDA", 7, 27, 0, NA, 0, 0, NA, 0, 0) + +/* Keypad */ +MUX_CFG("F18_1610_KBC0", 3, 15, 0, 0, 5, 1, 0, 0, 0) +MUX_CFG("D20_1610_KBC1", 3, 12, 0, 0, 4, 1, 0, 0, 0) +MUX_CFG("D19_1610_KBC2", 3, 9, 0, 0, 3, 1, 0, 0, 0) +MUX_CFG("E18_1610_KBC3", 3, 6, 0, 0, 2, 1, 0, 0, 0) +MUX_CFG("C21_1610_KBC4", 3, 3, 0, 0, 1, 1, 0, 0, 0) +MUX_CFG("G18_1610_KBR0", 4, 0, 0, 0, 10, 1, 0, 1, 0) +MUX_CFG("F19_1610_KBR1", 3, 27, 0, 0, 9, 1, 0, 1, 0) +MUX_CFG("H14_1610_KBR2", 3, 24, 0, 0, 8, 1, 0, 1, 0) +MUX_CFG("E20_1610_KBR3", 3, 21, 0, 0, 7, 1, 0, 1, 0) +MUX_CFG("E19_1610_KBR4", 3, 18, 0, 0, 6, 1, 0, 1, 0) +MUX_CFG("N19_1610_KBR5", 6, 12, 1, 1, 2, 1, 1, 1, 0) + +/* Power management */ +MUX_CFG("T20_1610_LOW_PWR", 7, 12, 1, NA, 0, 0, NA, 0, 0) + +/* MCLK Settings */ +MUX_CFG("V5_1710_MCLK_ON", B, 15, 0, NA, 0, 0, NA, 0, 0) +MUX_CFG("V5_1710_MCLK_OFF", B, 15, 6, NA, 0, 0, NA, 0, 0) +MUX_CFG("R10_1610_MCLK_ON", B, 18, 0, NA, 22, 0, NA, 1, 0) +MUX_CFG("R10_1610_MCLK_OFF", B, 18, 6, 2, 22, 1, 2, 1, 1) + +/* CompactFlash controller, conflicts with MMC1 */ +MUX_CFG("P11_1610_CF_CD2", A, 27, 3, 2, 15, 1, 2, 1, 1) +MUX_CFG("R11_1610_CF_IOIS16", B, 0, 3, 2, 16, 1, 2, 1, 1) +MUX_CFG("V10_1610_CF_IREQ", A, 24, 3, 2, 14, 0, 2, 0, 1) +MUX_CFG("W10_1610_CF_RESET", A, 18, 3, 2, 12, 1, 2, 1, 1) +MUX_CFG("W11_1610_CF_CD1", 10, 15, 3, 3, 8, 1, 3, 1, 1) +}; +#endif /* CONFIG_ARCH_OMAP15XX || CONFIG_ARCH_OMAP16XX */ + +int __init omap1_mux_init(void) +{ + +#ifdef CONFIG_ARCH_OMAP730 + omap_mux_register(omap730_pins, ARRAY_SIZE(omap730_pins)); +#endif + +#if defined(CONFIG_ARCH_OMAP15XX) || defined(CONFIG_ARCH_OMAP16XX) + omap_mux_register(omap1xxx_pins, ARRAY_SIZE(omap1xxx_pins)); +#endif + + return 0; +} + +#endif diff --git a/arch/arm/mach-omap1/serial.c b/arch/arm/mach-omap1/serial.c index 40c4f7c40e7..6810cfb8446 100644 --- a/arch/arm/mach-omap1/serial.c +++ b/arch/arm/mach-omap1/serial.c @@ -109,9 +109,10 @@ static struct platform_device serial_device = { * By default UART2 does not work on Innovator-1510 if you have * USB OHCI enabled. To use UART2, you must disable USB2 first. */ -void __init omap_serial_init(int ports[OMAP_MAX_NR_PORTS]) +void __init omap_serial_init(void) { int i; + const struct omap_uart_config *info; if (cpu_is_omap730()) { serial_platform_data[0].regshift = 0; @@ -126,10 +127,14 @@ void __init omap_serial_init(int ports[OMAP_MAX_NR_PORTS]) serial_platform_data[2].uartclk = OMAP1510_BASE_BAUD * 16; } + info = omap_get_config(OMAP_TAG_UART, struct omap_uart_config); + if (info == NULL) + return; + for (i = 0; i < OMAP_MAX_NR_PORTS; i++) { unsigned char reg; - if (ports[i] == 0) { + if (!((1 << i) & info->enabled_uarts)) { serial_platform_data[i].membase = NULL; serial_platform_data[i].mapbase = 0; continue; diff --git a/arch/arm/mach-omap1/time.c b/arch/arm/mach-omap1/time.c index 191a9b1ee9b..cdbf4d7620c 100644 --- a/arch/arm/mach-omap1/time.c +++ b/arch/arm/mach-omap1/time.c @@ -226,8 +226,8 @@ unsigned long long sched_clock(void) #ifdef CONFIG_OMAP_32K_TIMER -#ifdef CONFIG_ARCH_OMAP1510 -#error OMAP 32KHz timer does not currently work on 1510! +#ifdef CONFIG_ARCH_OMAP15XX +#error OMAP 32KHz timer does not currently work on 15XX! #endif /* diff --git a/arch/arm/mach-omap2/Kconfig b/arch/arm/mach-omap2/Kconfig new file mode 100644 index 00000000000..578880943cf --- /dev/null +++ b/arch/arm/mach-omap2/Kconfig @@ -0,0 +1,22 @@ +comment "OMAP Core Type" + depends on ARCH_OMAP2 + +config ARCH_OMAP24XX + bool "OMAP24xx Based System" + depends on ARCH_OMAP2 + +config ARCH_OMAP2420 + bool "OMAP2420 support" + depends on ARCH_OMAP24XX + +comment "OMAP Board Type" + depends on ARCH_OMAP2 + +config MACH_OMAP_GENERIC + bool "Generic OMAP board" + depends on ARCH_OMAP2 && ARCH_OMAP24XX + +config MACH_OMAP_H4 + bool "OMAP 2420 H4 board" + depends on ARCH_OMAP2 && ARCH_OMAP24XX + diff --git a/arch/arm/mach-omap2/Makefile b/arch/arm/mach-omap2/Makefile new file mode 100644 index 00000000000..42041166435 --- /dev/null +++ b/arch/arm/mach-omap2/Makefile @@ -0,0 +1,13 @@ +# +# Makefile for the linux kernel. +# + +# Common support +obj-y := irq.o id.o io.o sram-fn.o clock.o mux.o devices.o serial.o + +obj-$(CONFIG_OMAP_MPU_TIMER) += timer-gp.o + +# Specific board support +obj-$(CONFIG_MACH_OMAP_GENERIC) += board-generic.o +obj-$(CONFIG_MACH_OMAP_H4) += board-h4.o + diff --git a/arch/arm/mach-omap2/Makefile.boot b/arch/arm/mach-omap2/Makefile.boot new file mode 100644 index 00000000000..565aff7f37a --- /dev/null +++ b/arch/arm/mach-omap2/Makefile.boot @@ -0,0 +1,3 @@ + zreladdr-y := 0x80008000 +params_phys-y := 0x80000100 +initrd_phys-y := 0x80800000 diff --git a/arch/arm/mach-omap2/board-generic.c b/arch/arm/mach-omap2/board-generic.c new file mode 100644 index 00000000000..c602e7a3d93 --- /dev/null +++ b/arch/arm/mach-omap2/board-generic.c @@ -0,0 +1,80 @@ +/* + * linux/arch/arm/mach-omap/omap2/board-generic.c + * + * Copyright (C) 2005 Nokia Corporation + * Author: Paul Mundt <paul.mundt@nokia.com> + * + * Modified from mach-omap/omap1/board-generic.c + * + * Code for generic OMAP2 board. Should work on many OMAP2 systems where + * the bootloader passes the board-specific data to the kernel. + * Do not put any board specific code to this file; create a new machine + * type if you need custom low-level initializations. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/device.h> + +#include <asm/hardware.h> +#include <asm/mach-types.h> +#include <asm/mach/arch.h> +#include <asm/mach/map.h> + +#include <asm/arch/gpio.h> +#include <asm/arch/mux.h> +#include <asm/arch/usb.h> +#include <asm/arch/board.h> +#include <asm/arch/common.h> + +static void __init omap_generic_init_irq(void) +{ + omap_init_irq(); +} + +static struct omap_uart_config generic_uart_config __initdata = { + .enabled_uarts = ((1 << 0) | (1 << 1) | (1 << 2)), +}; + +static struct omap_mmc_config generic_mmc_config __initdata = { + .mmc [0] = { + .enabled = 0, + .wire4 = 0, + .wp_pin = -1, + .power_pin = -1, + .switch_pin = -1, + }, +}; + +static struct omap_board_config_kernel generic_config[] = { + { OMAP_TAG_UART, &generic_uart_config }, + { OMAP_TAG_MMC, &generic_mmc_config }, +}; + +static void __init omap_generic_init(void) +{ + omap_board_config = generic_config; + omap_board_config_size = ARRAY_SIZE(generic_config); + omap_serial_init(); +} + +static void __init omap_generic_map_io(void) +{ + omap_map_common_io(); +} + +MACHINE_START(OMAP_GENERIC, "Generic OMAP24xx") + /* Maintainer: Paul Mundt <paul.mundt@nokia.com> */ + .phys_ram = 0x80000000, + .phys_io = 0x48000000, + .io_pg_offst = ((0xd8000000) >> 18) & 0xfffc, + .boot_params = 0x80000100, + .map_io = omap_generic_map_io, + .init_irq = omap_generic_init_irq, + .init_machine = omap_generic_init, + .timer = &omap_timer, +MACHINE_END diff --git a/arch/arm/mach-omap2/board-h4.c b/arch/arm/mach-omap2/board-h4.c new file mode 100644 index 00000000000..f2554469a76 --- /dev/null +++ b/arch/arm/mach-omap2/board-h4.c @@ -0,0 +1,197 @@ +/* + * linux/arch/arm/mach-omap/omap2/board-h4.c + * + * Copyright (C) 2005 Nokia Corporation + * Author: Paul Mundt <paul.mundt@nokia.com> + * + * Modified from mach-omap/omap1/board-generic.c + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/platform_device.h> +#include <linux/mtd/mtd.h> +#include <linux/mtd/partitions.h> +#include <linux/delay.h> + +#include <asm/hardware.h> +#include <asm/mach-types.h> +#include <asm/mach/arch.h> +#include <asm/mach/map.h> +#include <asm/mach/flash.h> + +#include <asm/arch/gpio.h> +#include <asm/arch/mux.h> +#include <asm/arch/usb.h> +#include <asm/arch/board.h> +#include <asm/arch/common.h> +#include <asm/arch/prcm.h> + +#include <asm/io.h> +#include <asm/delay.h> + +static struct mtd_partition h4_partitions[] = { + /* bootloader (U-Boot, etc) in first sector */ + { + .name = "bootloader", + .offset = 0, + .size = SZ_128K, + .mask_flags = MTD_WRITEABLE, /* force read-only */ + }, + /* bootloader params in the next sector */ + { + .name = "params", + .offset = MTDPART_OFS_APPEND, + .size = SZ_128K, + .mask_flags = 0, + }, + /* kernel */ + { + .name = "kernel", + .offset = MTDPART_OFS_APPEND, + .size = SZ_2M, + .mask_flags = 0 + }, + /* file system */ + { + .name = "filesystem", + .offset = MTDPART_OFS_APPEND, + .size = MTDPART_SIZ_FULL, + .mask_flags = 0 + } +}; + +static struct flash_platform_data h4_flash_data = { + .map_name = "cfi_probe", + .width = 2, + .parts = h4_partitions, + .nr_parts = ARRAY_SIZE(h4_partitions), +}; + +static struct resource h4_flash_resource = { + .start = H4_CS0_BASE, + .end = H4_CS0_BASE + SZ_64M - 1, + .flags = IORESOURCE_MEM, +}; + +static struct platform_device h4_flash_device = { + .name = "omapflash", + .id = 0, + .dev = { + .platform_data = &h4_flash_data, + }, + .num_resources = 1, + .resource = &h4_flash_resource, +}; + +static struct resource h4_smc91x_resources[] = { + [0] = { + .start = OMAP24XX_ETHR_START, /* Physical */ + .end = OMAP24XX_ETHR_START + 0xf, + .flags = IORESOURCE_MEM, + }, + [1] = { + .start = OMAP_GPIO_IRQ(OMAP24XX_ETHR_GPIO_IRQ), + .end = OMAP_GPIO_IRQ(OMAP24XX_ETHR_GPIO_IRQ), + .flags = IORESOURCE_IRQ, + }, +}; + +static struct platform_device h4_smc91x_device = { + .name = "smc91x", + .id = -1, + .num_resources = ARRAY_SIZE(h4_smc91x_resources), + .resource = h4_smc91x_resources, +}; + +static struct platform_device *h4_devices[] __initdata = { + &h4_smc91x_device, + &h4_flash_device, +}; + +static inline void __init h4_init_smc91x(void) +{ + /* Make sure CS1 timings are correct */ + GPMC_CONFIG1_1 = 0x00011200; + GPMC_CONFIG2_1 = 0x001f1f01; + GPMC_CONFIG3_1 = 0x00080803; + GPMC_CONFIG4_1 = 0x1c091c09; + GPMC_CONFIG5_1 = 0x041f1f1f; + GPMC_CONFIG6_1 = 0x000004c4; + GPMC_CONFIG7_1 = 0x00000f40 | (0x08000000 >> 24); + udelay(100); + + omap_cfg_reg(M15_24XX_GPIO92); + if (omap_request_gpio(OMAP24XX_ETHR_GPIO_IRQ) < 0) { + printk(KERN_ERR "Failed to request GPIO%d for smc91x IRQ\n", + OMAP24XX_ETHR_GPIO_IRQ); + return; + } + omap_set_gpio_direction(OMAP24XX_ETHR_GPIO_IRQ, 1); +} + +static void __init omap_h4_init_irq(void) +{ + omap_init_irq(); + omap_gpio_init(); + h4_init_smc91x(); +} + +static struct omap_uart_config h4_uart_config __initdata = { + .enabled_uarts = ((1 << 0) | (1 << 1) | (1 << 2)), +}; + +static struct omap_mmc_config h4_mmc_config __initdata = { + .mmc [0] = { + .enabled = 1, + .wire4 = 1, + .wp_pin = -1, + .power_pin = -1, + .switch_pin = -1, + }, +}; + +static struct omap_lcd_config h4_lcd_config __initdata = { + .panel_name = "h4", + .ctrl_name = "internal", +}; + +static struct omap_board_config_kernel h4_config[] = { + { OMAP_TAG_UART, &h4_uart_config }, + { OMAP_TAG_MMC, &h4_mmc_config }, + { OMAP_TAG_LCD, &h4_lcd_config }, +}; + +static void __init omap_h4_init(void) +{ + /* + * Make sure the serial ports are muxed on at this point. + * You have to mux them off in device drivers later on + * if not needed. + */ + platform_add_devices(h4_devices, ARRAY_SIZE(h4_devices)); + omap_board_config = h4_config; + omap_board_config_size = ARRAY_SIZE(h4_config); + omap_serial_init(); +} + +static void __init omap_h4_map_io(void) +{ + omap_map_common_io(); +} + +MACHINE_START(OMAP_H4, "OMAP2420 H4 board") + /* Maintainer: Paul Mundt <paul.mundt@nokia.com> */ + .phys_ram = 0x80000000, + .phys_io = 0x48000000, + .io_pg_offst = ((0xd8000000) >> 18) & 0xfffc, + .boot_params = 0x80000100, + .map_io = omap_h4_map_io, + .init_irq = omap_h4_init_irq, + .init_machine = omap_h4_init, + .timer = &omap_timer, +MACHINE_END diff --git a/arch/arm/mach-omap2/clock.c b/arch/arm/mach-omap2/clock.c new file mode 100644 index 00000000000..85818d9f263 --- /dev/null +++ b/arch/arm/mach-omap2/clock.c @@ -0,0 +1,1129 @@ +/* + * linux/arch/arm/mach-omap2/clock.c + * + * Copyright (C) 2005 Texas Instruments Inc. + * Richard Woodruff <r-woodruff2@ti.com> + * Created for OMAP2. + * + * Cleaned up and modified to use omap shared clock framework by + * Tony Lindgren <tony@atomide.com> + * + * Based on omap1 clock.c, Copyright (C) 2004 - 2005 Nokia corporation + * Written by Tuukka Tikkanen <tuukka.tikkanen@elektrobit.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include <linux/config.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/device.h> +#include <linux/list.h> +#include <linux/errno.h> +#include <linux/delay.h> + +#include <asm/io.h> + +#include <asm/hardware/clock.h> +#include <asm/arch/clock.h> +#include <asm/arch/sram.h> +#include <asm/arch/prcm.h> + +#include "clock.h" + +//#define DOWN_VARIABLE_DPLL 1 /* Experimental */ + +static struct prcm_config *curr_prcm_set; +static struct memory_timings mem_timings; +static u32 curr_perf_level = PRCM_FULL_SPEED; + +/*------------------------------------------------------------------------- + * Omap2 specific clock functions + *-------------------------------------------------------------------------*/ + +/* Recalculate SYST_CLK */ +static void omap2_sys_clk_recalc(struct clk * clk) +{ + u32 div = PRCM_CLKSRC_CTRL; + div &= (1 << 7) | (1 << 6); /* Test if ext clk divided by 1 or 2 */ + div >>= clk->rate_offset; + clk->rate = (clk->parent->rate / div); + propagate_rate(clk); +} + +static u32 omap2_get_dpll_rate(struct clk * tclk) +{ + int dpll_clk, dpll_mult, dpll_div, amult; + + dpll_mult = (CM_CLKSEL1_PLL >> 12) & 0x03ff; /* 10 bits */ + dpll_div = (CM_CLKSEL1_PLL >> 8) & 0x0f; /* 4 bits */ + dpll_clk = (tclk->parent->rate * dpll_mult) / (dpll_div + 1); + amult = CM_CLKSEL2_PLL & 0x3; + dpll_clk *= amult; + + return dpll_clk; +} + +static void omap2_followparent_recalc(struct clk *clk) +{ + followparent_recalc(clk); +} + +static void omap2_propagate_rate(struct clk * clk) +{ + if (!(clk->flags & RATE_FIXED)) + clk->rate = clk->parent->rate; + + propagate_rate(clk); +} + +/* Enable an APLL if off */ +static void omap2_clk_fixed_enable(struct clk *clk) +{ + u32 cval, i=0; + + if (clk->enable_bit == 0xff) /* Parent will do it */ + return; + + cval = CM_CLKEN_PLL; + + if ((cval & (0x3 << clk->enable_bit)) == (0x3 << clk->enable_bit)) + return; + + cval &= ~(0x3 << clk->enable_bit); + cval |= (0x3 << clk->enable_bit); + CM_CLKEN_PLL = cval; + + if (clk == &apll96_ck) + cval = (1 << 8); + else if (clk == &apll54_ck) + cval = (1 << 6); + + while (!CM_IDLEST_CKGEN & cval) { /* Wait for lock */ + ++i; + udelay(1); + if (i == 100000) + break; + } +} + +/* Enables clock without considering parent dependencies or use count + * REVISIT: Maybe change this to use clk->enable like on omap1? + */ +static int omap2_clk_enable(struct clk * clk) +{ + u32 regval32; + + if (clk->flags & ALWAYS_ENABLED) + return 0; + + if (unlikely(clk->enable_reg == 0)) { + printk(KERN_ERR "clock.c: Enable for %s without enable code\n", + clk->name); + return 0; + } + + if (clk->enable_reg == (void __iomem *)&CM_CLKEN_PLL) { + omap2_clk_fixed_enable(clk); + return 0; + } + + regval32 = __raw_readl(clk->enable_reg); + regval32 |= (1 << clk->enable_bit); + __raw_writel(regval32, clk->enable_reg); + + return 0; +} + +/* Stop APLL */ +static void omap2_clk_fixed_disable(struct clk *clk) +{ + u32 cval; + + if(clk->enable_bit == 0xff) /* let parent off do it */ + return; + + cval = CM_CLKEN_PLL; + cval &= ~(0x3 << clk->enable_bit); + CM_CLKEN_PLL = cval; +} + +/* Disables clock without considering parent dependencies or use count */ +static void omap2_clk_disable(struct clk *clk) +{ + u32 regval32; + + if (clk->enable_reg == 0) + return; + + if (clk->enable_reg == (void __iomem *)&CM_CLKEN_PLL) { + omap2_clk_fixed_disable(clk); + return; + } + + regval32 = __raw_readl(clk->enable_reg); + regval32 &= ~(1 << clk->enable_bit); + __raw_writel(regval32, clk->enable_reg); +} + +static int omap2_clk_use(struct clk *clk) +{ + int ret = 0; + + if (clk->usecount++ == 0) { + if (likely((u32)clk->parent)) + ret = omap2_clk_use(clk->parent); + + if (unlikely(ret != 0)) { + clk->usecount--; + return ret; + } + + ret = omap2_clk_enable(clk); + + if (unlikely(ret != 0) && clk->parent) { + omap2_clk_unuse(clk->parent); + clk->usecount--; + } + } + + return ret; +} + +static void omap2_clk_unuse(struct clk *clk) +{ + if (clk->usecount > 0 && !(--clk->usecount)) { + omap2_clk_disable(clk); + if (likely((u32)clk->parent)) + omap2_clk_unuse(clk->parent); + } +} + +/* + * Uses the current prcm set to tell if a rate is valid. + * You can go slower, but not faster within a given rate set. + */ +static u32 omap2_dpll_round_rate(unsigned long target_rate) +{ + u32 high, low; + + if ((CM_CLKSEL2_PLL & 0x3) == 1) { /* DPLL clockout */ + high = curr_prcm_set->dpll_speed * 2; + low = curr_prcm_set->dpll_speed; + } else { /* DPLL clockout x 2 */ + high = curr_prcm_set->dpll_speed; + low = curr_prcm_set->dpll_speed / 2; + } + +#ifdef DOWN_VARIABLE_DPLL + if (target_rate > high) + return high; + else + return target_rate; +#else + if (target_rate > low) + return high; + else + return low; +#endif + +} + +/* + * Used for clocks that are part of CLKSEL_xyz governed clocks. + * REVISIT: Maybe change to use clk->enable() functions like on omap1? + */ +static void omap2_clksel_recalc(struct clk * clk) +{ + u32 fixed = 0, div = 0; + + if (clk == &dpll_ck) { + clk->rate = omap2_get_dpll_rate(clk); + fixed = 1; + div = 0; + } + + if (clk == &iva1_mpu_int_ifck) { + div = 2; + fixed = 1; + } + + if ((clk == &dss1_fck) && ((CM_CLKSEL1_CORE & (0x1f << 8)) == 0)) { + clk->rate = sys_ck.rate; + return; + } + + if (!fixed) { + div = omap2_clksel_get_divisor(clk); + if (div == 0) + return; + } + + if (div != 0) { + if (unlikely(clk->rate == clk->parent->rate / div)) + return; + clk->rate = clk->parent->rate / div; + } + + if (unlikely(clk->flags & RATE_PROPAGATES)) + propagate_rate(clk); +} + +/* + * Finds best divider value in an array based on the source and target + * rates. The divider array must be sorted with smallest divider first. + */ +static inline u32 omap2_divider_from_table(u32 size, u32 *div_array, + u32 src_rate, u32 tgt_rate) +{ + int i, test_rate; + + if (div_array == NULL) + return ~1; + + for (i=0; i < size; i++) { + test_rate = src_rate / *div_array; + if (test_rate <= tgt_rate) + return *div_array; + ++div_array; + } + + return ~0; /* No acceptable divider */ +} + +/* + * Find divisor for the given clock and target rate. + * + * Note that this will not work for clocks which are part of CONFIG_PARTICIPANT, + * they are only settable as part of virtual_prcm set. + */ +static u32 omap2_clksel_round_rate(struct clk *tclk, u32 target_rate, + u32 *new_div) +{ + u32 gfx_div[] = {2, 3, 4}; + u32 sysclkout_div[] = {1, 2, 4, 8, 16}; + u32 dss1_div[] = {1, 2, 3, 4, 5, 6, 8, 9, 12, 16}; + u32 vylnq_div[] = {1, 2, 3, 4, 6, 8, 9, 12, 16, 18}; + u32 best_div = ~0, asize = 0; + u32 *div_array = NULL; + + switch (tclk->flags & SRC_RATE_SEL_MASK) { + case CM_GFX_SEL1: + asize = 3; + div_array = gfx_div; + break; + case CM_PLL_SEL1: + return omap2_dpll_round_rate(target_rate); + case CM_SYSCLKOUT_SEL1: + asize = 5; + div_array = sysclkout_div; + break; + case CM_CORE_SEL1: + if(tclk == &dss1_fck){ + if(tclk->parent == &core_ck){ + asize = 10; + div_array = dss1_div; + } else { + *new_div = 0; /* fixed clk */ + return(tclk->parent->rate); + } + } else if((tclk == &vlynq_fck) && cpu_is_omap2420()){ + if(tclk->parent == &core_ck){ + asize = 10; + div_array = vylnq_div; + } else { + *new_div = 0; /* fixed clk */ + return(tclk->parent->rate); + } + } + break; + } + + best_div = omap2_divider_from_table(asize, div_array, + tclk->parent->rate, target_rate); + if (best_div == ~0){ + *new_div = 1; + return best_div; /* signal error */ + } + + *new_div = best_div; + return (tclk->parent->rate / best_div); +} + +/* Given a clock and a rate apply a clock specific rounding function */ +static long omap2_clk_round_rate(struct clk *clk, unsigned long rate) +{ + u32 new_div = 0; + int valid_rate; + + if (clk->flags & RATE_FIXED) + return clk->rate; + + if (clk->flags & RATE_CKCTL) { + valid_rate = omap2_clksel_round_rate(clk, rate, &new_div); + return valid_rate; + } + + if (clk->round_rate != 0) + return clk->round_rate(clk, rate); + + return clk->rate; +} + +/* + * Check the DLL lock state, and return tue if running in unlock mode. + * This is needed to compenste for the shifted DLL value in unlock mode. + */ +static u32 omap2_dll_force_needed(void) +{ + u32 dll_state = SDRC_DLLA_CTRL; /* dlla and dllb are a set */ + + if ((dll_state & (1 << 2)) == (1 << 2)) + return 1; + else + return 0; +} + +static void omap2_init_memory_params(u32 force_lock_to_unlock_mode) +{ + unsigned long dll_cnt; + u32 fast_dll = 0; + + mem_timings.m_type = !((SDRC_MR_0 & 0x3) == 0x1); /* DDR = 1, SDR = 0 */ + + /* 2422 es2.05 and beyond has a single SIP DDR instead of 2 like others. + * In the case of 2422, its ok to use CS1 instead of CS0. + */ + +#if 0 /* FIXME: Enable after 24xx cpu detection works */ + ctype = get_cpu_type(); + if (cpu_is_omap2422()) + mem_timings.base_cs = 1; + else +#endif + mem_timings.base_cs = 0; + + if (mem_timings.m_type != M_DDR) + return; + + /* With DDR we need to determine the low frequency DLL value */ + if (((mem_timings.fast_dll_ctrl & (1 << 2)) == M_LOCK_CTRL)) + mem_timings.dll_mode = M_UNLOCK; + else + mem_timings.dll_mode = M_LOCK; + + if (mem_timings.base_cs == 0) { + fast_dll = SDRC_DLLA_CTRL; + dll_cnt = SDRC_DLLA_STATUS & 0xff00; + } else { + fast_dll = SDRC_DLLB_CTRL; + dll_cnt = SDRC_DLLB_STATUS & 0xff00; + } + if (force_lock_to_unlock_mode) { + fast_dll &= ~0xff00; + fast_dll |= dll_cnt; /* Current lock mode */ + } + mem_timings.fast_dll_ctrl = fast_dll; + + /* No disruptions, DDR will be offline & C-ABI not followed */ + omap2_sram_ddr_init(&mem_timings.slow_dll_ctrl, + mem_timings.fast_dll_ctrl, + mem_timings.base_cs, + force_lock_to_unlock_mode); + mem_timings.slow_dll_ctrl &= 0xff00; /* Keep lock value */ + + /* Turn status into unlock ctrl */ + mem_timings.slow_dll_ctrl |= + ((mem_timings.fast_dll_ctrl & 0xF) | (1 << 2)); + + /* 90 degree phase for anything below 133Mhz */ + mem_timings.slow_dll_ctrl |= (1 << 1); +} + +static u32 omap2_reprogram_sdrc(u32 level, u32 force) +{ + u32 prev = curr_perf_level, flags; + + if ((curr_perf_level == level) && !force) + return prev; + + if (level == PRCM_HALF_SPEED) { + local_irq_save(flags); + PRCM_VOLTSETUP = 0xffff; + omap2_sram_reprogram_sdrc(PRCM_HALF_SPEED, + mem_timings.slow_dll_ctrl, + mem_timings.m_type); + curr_perf_level = PRCM_HALF_SPEED; + local_irq_restore(flags); + } + if (level == PRCM_FULL_SPEED) { + local_irq_save(flags); + PRCM_VOLTSETUP = 0xffff; + omap2_sram_reprogram_sdrc(PRCM_FULL_SPEED, + mem_timings.fast_dll_ctrl, + mem_timings.m_type); + curr_perf_level = PRCM_FULL_SPEED; + local_irq_restore(flags); + } + + return prev; +} + +static int omap2_reprogram_dpll(struct clk * clk, unsigned long rate) +{ + u32 flags, cur_rate, low, mult, div, valid_rate, done_rate; + u32 bypass = 0; + struct prcm_config tmpset; + int ret = -EINVAL; + + local_irq_save(flags); + cur_rate = omap2_get_dpll_rate(&dpll_ck); + mult = CM_CLKSEL2_PLL & 0x3; + + if ((rate == (cur_rate / 2)) && (mult == 2)) { + omap2_reprogram_sdrc(PRCM_HALF_SPEED, 1); + } else if ((rate == (cur_rate * 2)) && (mult == 1)) { + omap2_reprogram_sdrc(PRCM_FULL_SPEED, 1); + } else if (rate != cur_rate) { + valid_rate = omap2_dpll_round_rate(rate); + if (valid_rate != rate) + goto dpll_exit; + + if ((CM_CLKSEL2_PLL & 0x3) == 1) + low = curr_prcm_set->dpll_speed; + else + low = curr_prcm_set->dpll_speed / 2; + + tmpset.cm_clksel1_pll = CM_CLKSEL1_PLL; + tmpset.cm_clksel1_pll &= ~(0x3FFF << 8); + div = ((curr_prcm_set->xtal_speed / 1000000) - 1); + tmpset.cm_clksel2_pll = CM_CLKSEL2_PLL; + tmpset.cm_clksel2_pll &= ~0x3; + if (rate > low) { + tmpset.cm_clksel2_pll |= 0x2; + mult = ((rate / 2) / 1000000); + done_rate = PRCM_FULL_SPEED; + } else { + tmpset.cm_clksel2_pll |= 0x1; + mult = (rate / 1000000); + done_rate = PRCM_HALF_SPEED; + } + tmpset.cm_clksel1_pll |= ((div << 8) | (mult << 12)); + + /* Worst case */ + tmpset.base_sdrc_rfr = V24XX_SDRC_RFR_CTRL_BYPASS; + + if (rate == curr_prcm_set->xtal_speed) /* If asking for 1-1 */ + bypass = 1; + + omap2_reprogram_sdrc(PRCM_FULL_SPEED, 1); /* For init_mem */ + + /* Force dll lock mode */ + omap2_set_prcm(tmpset.cm_clksel1_pll, tmpset.base_sdrc_rfr, + bypass); + + /* Errata: ret dll entry state */ + omap2_init_memory_params(omap2_dll_force_needed()); + omap2_reprogram_sdrc(done_rate, 0); + } + omap2_clksel_recalc(&dpll_ck); + ret = 0; + +dpll_exit: + local_irq_restore(flags); + return(ret); +} + +/* Just return the MPU speed */ +static void omap2_mpu_recalc(struct clk * clk) +{ + clk->rate = curr_prcm_set->mpu_speed; +} + +/* + * Look for a rate equal or less than the target rate given a configuration set. + * + * What's not entirely clear is "which" field represents the key field. + * Some might argue L3-DDR, others ARM, others IVA. This code is simple and + * just uses the ARM rates. + */ +static long omap2_round_to_table_rate(struct clk * clk, unsigned long rate) +{ + struct prcm_config * ptr; + long highest_rate; + + if (clk != &virt_prcm_set) + return -EINVAL; + + highest_rate = -EINVAL; + + for (ptr = rate_table; ptr->mpu_speed; ptr++) { + if (ptr->xtal_speed != sys_ck.rate) + continue; + + highest_rate = ptr->mpu_speed; + + /* Can check only after xtal frequency check */ + if (ptr->mpu_speed <= rate) + break; + } + return highest_rate; +} + +/* + * omap2_convert_field_to_div() - turn field value into integer divider + */ +static u32 omap2_clksel_to_divisor(u32 div_sel, u32 field_val) +{ + u32 i; + u32 clkout_array[] = {1, 2, 4, 8, 16}; + + if ((div_sel & SRC_RATE_SEL_MASK) == CM_SYSCLKOUT_SEL1) { + for (i = 0; i < 5; i++) { + if (field_val == i) + return clkout_array[i]; + } + return ~0; + } else + return field_val; +} + +/* + * Returns the CLKSEL divider register value + * REVISIT: This should be cleaned up to work nicely with void __iomem * + */ +static u32 omap2_get_clksel(u32 *div_sel, u32 *field_mask, + struct clk *clk) +{ + int ret = ~0; + u32 reg_val, div_off; + u32 div_addr = 0; + u32 mask = ~0; + + div_off = clk->rate_offset; + + switch ((*div_sel & SRC_RATE_SEL_MASK)) { + case CM_MPU_SEL1: + div_addr = (u32)&CM_CLKSEL_MPU; + mask = 0x1f; + break; + case CM_DSP_SEL1: + div_addr = (u32)&CM_CLKSEL_DSP; + if (cpu_is_omap2420()) { + if ((div_off == 0) || (div_off == 8)) + mask = 0x1f; + else if (div_off == 5) + mask = 0x3; + } else if (cpu_is_omap2430()) { + if (div_off == 0) + mask = 0x1f; + else if (div_off == 5) + mask = 0x3; + } + break; + case CM_GFX_SEL1: + div_addr = (u32)&CM_CLKSEL_GFX; + if (div_off == 0) + mask = 0x7; + break; + case CM_MODEM_SEL1: + div_addr = (u32)&CM_CLKSEL_MDM; + if (div_off == 0) + mask = 0xf; + break; + case CM_SYSCLKOUT_SEL1: + div_addr = (u32)&PRCM_CLKOUT_CTRL; + if ((div_off == 3) || (div_off = 11)) + mask= 0x3; + break; + case CM_CORE_SEL1: + div_addr = (u32)&CM_CLKSEL1_CORE; + switch (div_off) { + case 0: /* l3 */ + case 8: /* dss1 */ + case 15: /* vylnc-2420 */ + case 20: /* ssi */ + mask = 0x1f; break; + case 5: /* l4 */ + mask = 0x3; break; + case 13: /* dss2 */ + mask = 0x1; break; + case 25: /* usb */ + mask = 0xf; break; + } + } + + *field_mask = mask; + + if (unlikely(mask == ~0)) + div_addr = 0; + + *div_sel = div_addr; + + if (unlikely(div_addr == 0)) + return ret; + + /* Isolate field */ + reg_val = __raw_readl((void __iomem *)div_addr) & (mask << div_off); + + /* Normalize back to divider value */ + reg_val >>= div_off; + + return reg_val; +} + +/* + * Return divider to be applied to parent clock. + * Return 0 on error. + */ +static u32 omap2_clksel_get_divisor(struct clk *clk) +{ + int ret = 0; + u32 div, div_sel, div_off, field_mask, field_val; + + /* isolate control register */ + div_sel = (SRC_RATE_SEL_MASK & clk->flags); + + div_off = clk->rate_offset; + field_val = omap2_get_clksel(&div_sel, &field_mask, clk); + if (div_sel == 0) + return ret; + + div_sel = (SRC_RATE_SEL_MASK & clk->flags); + div = omap2_clksel_to_divisor(div_sel, field_val); + + return div; +} + +/* Set the clock rate for a clock source */ +static int omap2_clk_set_rate(struct clk *clk, unsigned long rate) + +{ + int ret = -EINVAL; + void __iomem * reg; + u32 div_sel, div_off, field_mask, field_val, reg_val, validrate; + u32 new_div = 0; + + if (!(clk->flags & CONFIG_PARTICIPANT) && (clk->flags & RATE_CKCTL)) { + if (clk == &dpll_ck) + return omap2_reprogram_dpll(clk, rate); + + /* Isolate control register */ + div_sel = (SRC_RATE_SEL_MASK & clk->flags); + div_off = clk->src_offset; + + validrate = omap2_clksel_round_rate(clk, rate, &new_div); + if(validrate != rate) + return(ret); + + field_val = omap2_get_clksel(&div_sel, &field_mask, clk); + if (div_sel == 0) + return ret; + + if(clk->flags & CM_SYSCLKOUT_SEL1){ + switch(new_div){ + case 16: field_val = 4; break; + case 8: field_val = 3; break; + case 4: field_val = 2; break; + case 2: field_val = 1; break; + case 1: field_val = 0; break; + } + } + else + field_val = new_div; + + reg = (void __iomem *)div_sel; + + reg_val = __raw_readl(reg); + reg_val &= ~(field_mask << div_off); + reg_val |= (field_val << div_off); + + __raw_writel(reg_val, reg); + clk->rate = clk->parent->rate / field_val; + + if (clk->flags & DELAYED_APP) + __raw_writel(0x1, (void __iomem *)&PRCM_CLKCFG_CTRL); + ret = 0; + } else if (clk->set_rate != 0) + ret = clk->set_rate(clk, rate); + + if (unlikely(ret == 0 && (clk->flags & RATE_PROPAGATES))) + propagate_rate(clk); + + return ret; +} + +/* Converts encoded control register address into a full address */ +static u32 omap2_get_src_field(u32 *type_to_addr, u32 reg_offset, + struct clk *src_clk, u32 *field_mask) +{ + u32 val = ~0, src_reg_addr = 0, mask = 0; + + /* Find target control register.*/ + switch ((*type_to_addr & SRC_RATE_SEL_MASK)) { + case CM_CORE_SEL1: + src_reg_addr = (u32)&CM_CLKSEL1_CORE; + if (reg_offset == 13) { /* DSS2_fclk */ + mask = 0x1; + if (src_clk == &sys_ck) + val = 0; + if (src_clk == &func_48m_ck) + val = 1; + } else if (reg_offset == 8) { /* DSS1_fclk */ + mask = 0x1f; + if (src_clk == &sys_ck) + val = 0; + else if (src_clk == &core_ck) /* divided clock */ + val = 0x10; /* rate needs fixing */ + } else if ((reg_offset == 15) && cpu_is_omap2420()){ /*vlnyq*/ + mask = 0x1F; + if(src_clk == &func_96m_ck) + val = 0; + else if (src_clk == &core_ck) + val = 0x10; + } + break; + case CM_CORE_SEL2: + src_reg_addr = (u32)&CM_CLKSEL2_CORE; + mask = 0x3; + if (src_clk == &func_32k_ck) + val = 0x0; + if (src_clk == &sys_ck) + val = 0x1; + if (src_clk == &alt_ck) + val = 0x2; + break; + case CM_WKUP_SEL1: + src_reg_addr = (u32)&CM_CLKSEL2_CORE; + mask = 0x3; + if (src_clk == &func_32k_ck) + val = 0x0; + if (src_clk == &sys_ck) + val = 0x1; + if (src_clk == &alt_ck) + val = 0x2; + break; + case CM_PLL_SEL1: + src_reg_addr = (u32)&CM_CLKSEL1_PLL; + mask = 0x1; + if (reg_offset == 0x3) { + if (src_clk == &apll96_ck) + val = 0; + if (src_clk == &alt_ck) + val = 1; + } + else if (reg_offset == 0x5) { + if (src_clk == &apll54_ck) + val = 0; + if (src_clk == &alt_ck) + val = 1; + } + break; + case CM_PLL_SEL2: + src_reg_addr = (u32)&CM_CLKSEL2_PLL; + mask = 0x3; + if (src_clk == &func_32k_ck) + val = 0x0; + if (src_clk == &dpll_ck) + val = 0x2; + break; + case CM_SYSCLKOUT_SEL1: + src_reg_addr = (u32)&PRCM_CLKOUT_CTRL; + mask = 0x3; + if (src_clk == &dpll_ck) + val = 0; + if (src_clk == &sys_ck) + val = 1; + if (src_clk == &func_54m_ck) + val = 2; + if (src_clk == &func_96m_ck) + val = 3; + break; + } + + if (val == ~0) /* Catch errors in offset */ + *type_to_addr = 0; + else + *type_to_addr = src_reg_addr; + *field_mask = mask; + + return val; +} + +static int omap2_clk_set_parent(struct clk *clk, struct clk *new_parent) +{ + void __iomem * reg; + u32 src_sel, src_off, field_val, field_mask, reg_val, rate; + int ret = -EINVAL; + + if (unlikely(clk->flags & CONFIG_PARTICIPANT)) + return ret; + + if (clk->flags & SRC_SEL_MASK) { /* On-chip SEL collection */ + src_sel = (SRC_RATE_SEL_MASK & clk->flags); + src_off = clk->src_offset; + + if (src_sel == 0) + goto set_parent_error; + + field_val = omap2_get_src_field(&src_sel, src_off, new_parent, + &field_mask); + + reg = (void __iomem *)src_sel; + + if (clk->usecount > 0) + omap2_clk_disable(clk); + + /* Set new source value (previous dividers if any in effect) */ + reg_val = __raw_readl(reg) & ~(field_mask << src_off); + reg_val |= (field_val << src_off); + __raw_writel(reg_val, reg); + + if (clk->flags & DELAYED_APP) + __raw_writel(0x1, (void __iomem *)&PRCM_CLKCFG_CTRL); + + if (clk->usecount > 0) + omap2_clk_enable(clk); + + clk->parent = new_parent; + + /* SRC_RATE_SEL_MASK clocks follow their parents rates.*/ + if ((new_parent == &core_ck) && (clk == &dss1_fck)) + clk->rate = new_parent->rate / 0x10; + else + clk->rate = new_parent->rate; + + if (unlikely(clk->flags & RATE_PROPAGATES)) + propagate_rate(clk); + + return 0; + } else { + clk->parent = new_parent; + rate = new_parent->rate; + omap2_clk_set_rate(clk, rate); + ret = 0; + } + + set_parent_error: + return ret; +} + +/* Sets basic clocks based on the specified rate */ +static int omap2_select_table_rate(struct clk * clk, unsigned long rate) +{ + u32 flags, cur_rate, done_rate, bypass = 0; + u8 cpu_mask = 0; + struct prcm_config *prcm; + unsigned long found_speed = 0; + + if (clk != &virt_prcm_set) + return -EINVAL; + + /* FIXME: Change cpu_is_omap2420() to cpu_is_omap242x() */ + if (cpu_is_omap2420()) + cpu_mask = RATE_IN_242X; + else if (cpu_is_omap2430()) + cpu_mask = RATE_IN_243X; + + for (prcm = rate_table; prcm->mpu_speed; prcm++) { + if (!(prcm->flags & cpu_mask)) + continue; + + if (prcm->xtal_speed != sys_ck.rate) + continue; + + if (prcm->mpu_speed <= rate) { + found_speed = prcm->mpu_speed; + break; + } + } + + if (!found_speed) { + printk(KERN_INFO "Could not set MPU rate to %luMHz\n", + rate / 1000000); + return -EINVAL; + } + + curr_prcm_set = prcm; + cur_rate = omap2_get_dpll_rate(&dpll_ck); + + if (prcm->dpll_speed == cur_rate / 2) { + omap2_reprogram_sdrc(PRCM_HALF_SPEED, 1); + } else if (prcm->dpll_speed == cur_rate * 2) { + omap2_reprogram_sdrc(PRCM_FULL_SPEED, 1); + } else if (prcm->dpll_speed != cur_rate) { + local_irq_save(flags); + + if (prcm->dpll_speed == prcm->xtal_speed) + bypass = 1; + + if ((prcm->cm_clksel2_pll & 0x3) == 2) + done_rate = PRCM_FULL_SPEED; + else + done_rate = PRCM_HALF_SPEED; + + /* MPU divider */ + CM_CLKSEL_MPU = prcm->cm_clksel_mpu; + + /* dsp + iva1 div(2420), iva2.1(2430) */ + CM_CLKSEL_DSP = prcm->cm_clksel_dsp; + + CM_CLKSEL_GFX = prcm->cm_clksel_gfx; + + /* Major subsystem dividers */ + CM_CLKSEL1_CORE = prcm->cm_clksel1_core; + if (cpu_is_omap2430()) + CM_CLKSEL_MDM = prcm->cm_clksel_mdm; + + /* x2 to enter init_mem */ + omap2_reprogram_sdrc(PRCM_FULL_SPEED, 1); + + omap2_set_prcm(prcm->cm_clksel1_pll, prcm->base_sdrc_rfr, + bypass); + + omap2_init_memory_params(omap2_dll_force_needed()); + omap2_reprogram_sdrc(done_rate, 0); + + local_irq_restore(flags); + } + omap2_clksel_recalc(&dpll_ck); + + return 0; +} + +/*------------------------------------------------------------------------- + * Omap2 clock reset and init functions + *-------------------------------------------------------------------------*/ + +static struct clk_functions omap2_clk_functions = { + .clk_enable = omap2_clk_enable, + .clk_disable = omap2_clk_disable, + .clk_use = omap2_clk_use, + .clk_unuse = omap2_clk_unuse, + .clk_round_rate = omap2_clk_round_rate, + .clk_set_rate = omap2_clk_set_rate, + .clk_set_parent = omap2_clk_set_parent, +}; + +static void __init omap2_get_crystal_rate(struct clk *osc, struct clk *sys) +{ + u32 div, aplls, sclk = 13000000; + + aplls = CM_CLKSEL1_PLL; + aplls &= ((1 << 23) | (1 << 24) | (1 << 25)); + aplls >>= 23; /* Isolate field, 0,2,3 */ + + if (aplls == 0) + sclk = 19200000; + else if (aplls == 2) + sclk = 13000000; + else if (aplls == 3) + sclk = 12000000; + + div = PRCM_CLKSRC_CTRL; + div &= ((1 << 7) | (1 << 6)); + div >>= sys->rate_offset; + + osc->rate = sclk * div; + sys->rate = sclk; +} + +#ifdef CONFIG_OMAP_RESET_CLOCKS +static void __init omap2_disable_unused_clocks(void) +{ + struct clk *ck; + u32 regval32; + + list_for_each_entry(ck, &clocks, node) { + if (ck->usecount > 0 || (ck->flags & ALWAYS_ENABLED) || + ck->enable_reg == 0) + continue; + + regval32 = __raw_readl(ck->enable_reg); + if ((regval32 & (1 << ck->enable_bit)) == 0) + continue; + + printk(KERN_INFO "Disabling unused clock \"%s\"\n", ck->name); + omap2_clk_disable(ck); + } +} +late_initcall(omap2_disable_unused_clocks); +#endif + +/* + * Switch the MPU rate if specified on cmdline. + * We cannot do this early until cmdline is parsed. + */ +static int __init omap2_clk_arch_init(void) +{ + if (!mpurate) + return -EINVAL; + + if (omap2_select_table_rate(&virt_prcm_set, mpurate)) + printk(KERN_ERR "Could not find matching MPU rate\n"); + + propagate_rate(&osc_ck); /* update main root fast */ + propagate_rate(&func_32k_ck); /* update main root slow */ + + printk(KERN_INFO "Switched to new clocking rate (Crystal/DPLL/MPU): " + "%ld.%01ld/%ld/%ld MHz\n", + (sys_ck.rate / 1000000), (sys_ck.rate / 100000) % 10, + (dpll_ck.rate / 1000000), (mpu_ck.rate / 1000000)) ; + + return 0; +} +arch_initcall(omap2_clk_arch_init); + +int __init omap2_clk_init(void) +{ + struct prcm_config *prcm; + struct clk ** clkp; + u32 clkrate; + + clk_init(&omap2_clk_functions); + omap2_get_crystal_rate(&osc_ck, &sys_ck); + + for (clkp = onchip_clks; clkp < onchip_clks + ARRAY_SIZE(onchip_clks); + clkp++) { + + if ((*clkp)->flags & CLOCK_IN_OMAP242X && cpu_is_omap2420()) { + clk_register(*clkp); + continue; + } + + if ((*clkp)->flags & CLOCK_IN_OMAP243X && cpu_is_omap2430()) { + clk_register(*clkp); + continue; + } + } + + /* Check the MPU rate set by bootloader */ + clkrate = omap2_get_dpll_rate(&dpll_ck); + for (prcm = rate_table; prcm->mpu_speed; prcm++) { + if (prcm->xtal_speed != sys_ck.rate) + continue; + if (prcm->dpll_speed <= clkrate) + break; + } + curr_prcm_set = prcm; + + propagate_rate(&osc_ck); /* update main root fast */ + propagate_rate(&func_32k_ck); /* update main root slow */ + + printk(KERN_INFO "Clocking rate (Crystal/DPLL/MPU): " + "%ld.%01ld/%ld/%ld MHz\n", + (sys_ck.rate / 1000000), (sys_ck.rate / 100000) % 10, + (dpll_ck.rate / 1000000), (mpu_ck.rate / 1000000)) ; + + /* + * Only enable those clocks we will need, let the drivers + * enable other clocks as necessary + */ + clk_use(&sync_32k_ick); + clk_use(&omapctrl_ick); + if (cpu_is_omap2430()) + clk_use(&sdrc_ick); + + return 0; +} diff --git a/arch/arm/mach-omap2/clock.h b/arch/arm/mach-omap2/clock.h new file mode 100644 index 00000000000..4aeab5591bd --- /dev/null +++ b/arch/arm/mach-omap2/clock.h @@ -0,0 +1,2103 @@ +/* + * linux/arch/arm/mach-omap24xx/clock.h + * + * Copyright (C) 2005 Texas Instruments Inc. + * Richard Woodruff <r-woodruff2@ti.com> + * Created for OMAP2. + * + * Copyright (C) 2004 Nokia corporation + * Written by Tuukka Tikkanen <tuukka.tikkanen@elektrobit.com> + * Based on clocks.h by Tony Lindgren, Gordon McNutt and RidgeRun, Inc + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#ifndef __ARCH_ARM_MACH_OMAP2_CLOCK_H +#define __ARCH_ARM_MACH_OMAP2_CLOCK_H + +static void omap2_sys_clk_recalc(struct clk * clk); +static void omap2_clksel_recalc(struct clk * clk); +static void omap2_followparent_recalc(struct clk * clk); +static void omap2_propagate_rate(struct clk * clk); +static void omap2_mpu_recalc(struct clk * clk); +static int omap2_select_table_rate(struct clk * clk, unsigned long rate); +static long omap2_round_to_table_rate(struct clk * clk, unsigned long rate); +static void omap2_clk_unuse(struct clk *clk); +static void omap2_sys_clk_recalc(struct clk * clk); +static u32 omap2_clksel_to_divisor(u32 div_sel, u32 field_val); +static u32 omap2_clksel_get_divisor(struct clk *clk); + + +#define RATE_IN_242X (1 << 0) +#define RATE_IN_243X (1 << 1) + +/* Memory timings */ +#define M_DDR 1 +#define M_LOCK_CTRL (1 << 2) +#define M_UNLOCK 0 +#define M_LOCK 1 + +struct memory_timings { + u32 m_type; /* ddr = 1, sdr = 0 */ + u32 dll_mode; /* use lock mode = 1, unlock mode = 0 */ + u32 slow_dll_ctrl; /* unlock mode, dll value for slow speed */ + u32 fast_dll_ctrl; /* unlock mode, dll value for fast speed */ + u32 base_cs; /* base chip select to use for calculations */ +}; + +/* Key dividers which make up a PRCM set. Ratio's for a PRCM are mandated. + * xtal_speed, dpll_speed, mpu_speed, CM_CLKSEL_MPU,CM_CLKSEL_DSP + * CM_CLKSEL_GFX, CM_CLKSEL1_CORE, CM_CLKSEL1_PLL CM_CLKSEL2_PLL, CM_CLKSEL_MDM + */ +struct prcm_config { + unsigned long xtal_speed; /* crystal rate */ + unsigned long dpll_speed; /* dpll: out*xtal*M/(N-1)table_recalc */ + unsigned long mpu_speed; /* speed of MPU */ + unsigned long cm_clksel_mpu; /* mpu divider */ + unsigned long cm_clksel_dsp; /* dsp+iva1 div(2420), iva2.1(2430) */ + unsigned long cm_clksel_gfx; /* gfx dividers */ + unsigned long cm_clksel1_core; /* major subsystem dividers */ + unsigned long cm_clksel1_pll; /* m,n */ + unsigned long cm_clksel2_pll; /* dpllx1 or x2 out */ + unsigned long cm_clksel_mdm; /* modem dividers 2430 only */ + unsigned long base_sdrc_rfr; /* base refresh timing for a set */ + unsigned char flags; +}; + +/* Mask for clksel which support parent settign in set_rate */ +#define SRC_SEL_MASK (CM_CORE_SEL1 | CM_CORE_SEL2 | CM_WKUP_SEL1 | \ + CM_PLL_SEL1 | CM_PLL_SEL2 | CM_SYSCLKOUT_SEL1) + +/* Mask for clksel regs which support rate operations */ +#define SRC_RATE_SEL_MASK (CM_MPU_SEL1 | CM_DSP_SEL1 | CM_GFX_SEL1 | \ + CM_MODEM_SEL1 | CM_CORE_SEL1 | CM_CORE_SEL2 | \ + CM_WKUP_SEL1 | CM_PLL_SEL1 | CM_PLL_SEL2 | \ + CM_SYSCLKOUT_SEL1) + +/* + * The OMAP2 processor can be run at several discrete 'PRCM configurations'. + * These configurations are characterized by voltage and speed for clocks. + * The device is only validated for certain combinations. One way to express + * these combinations is via the 'ratio's' which the clocks operate with + * respect to each other. These ratio sets are for a given voltage/DPLL + * setting. All configurations can be described by a DPLL setting and a ratio + * There are 3 ratio sets for the 2430 and X ratio sets for 2420. + * + * 2430 differs from 2420 in that there are no more phase synchronizers used. + * They both have a slightly different clock domain setup. 2420(iva1,dsp) vs + * 2430 (iva2.1, NOdsp, mdm) + */ + +/* Core fields for cm_clksel, not ratio governed */ +#define RX_CLKSEL_DSS1 (0x10 << 8) +#define RX_CLKSEL_DSS2 (0x0 << 13) +#define RX_CLKSEL_SSI (0x5 << 20) + +/*------------------------------------------------------------------------- + * Voltage/DPLL ratios + *-------------------------------------------------------------------------*/ + +/* 2430 Ratio's, 2430-Ratio Config 1 */ +#define R1_CLKSEL_L3 (4 << 0) +#define R1_CLKSEL_L4 (2 << 5) +#define R1_CLKSEL_USB (4 << 25) +#define R1_CM_CLKSEL1_CORE_VAL R1_CLKSEL_USB | RX_CLKSEL_SSI | \ + RX_CLKSEL_DSS2 | RX_CLKSEL_DSS1 | \ + R1_CLKSEL_L4 | R1_CLKSEL_L3 +#define R1_CLKSEL_MPU (2 << 0) +#define R1_CM_CLKSEL_MPU_VAL R1_CLKSEL_MPU +#define R1_CLKSEL_DSP (2 << 0) +#define R1_CLKSEL_DSP_IF (2 << 5) +#define R1_CM_CLKSEL_DSP_VAL R1_CLKSEL_DSP | R1_CLKSEL_DSP_IF +#define R1_CLKSEL_GFX (2 << 0) +#define R1_CM_CLKSEL_GFX_VAL R1_CLKSEL_GFX +#define R1_CLKSEL_MDM (4 << 0) +#define R1_CM_CLKSEL_MDM_VAL R1_CLKSEL_MDM + +/* 2430-Ratio Config 2 */ +#define R2_CLKSEL_L3 (6 << 0) +#define R2_CLKSEL_L4 (2 << 5) +#define R2_CLKSEL_USB (2 << 25) +#define R2_CM_CLKSEL1_CORE_VAL R2_CLKSEL_USB | RX_CLKSEL_SSI | \ + RX_CLKSEL_DSS2 | RX_CLKSEL_DSS1 | \ + R2_CLKSEL_L4 | R2_CLKSEL_L3 +#define R2_CLKSEL_MPU (2 << 0) +#define R2_CM_CLKSEL_MPU_VAL R2_CLKSEL_MPU +#define R2_CLKSEL_DSP (2 << 0) +#define R2_CLKSEL_DSP_IF (3 << 5) +#define R2_CM_CLKSEL_DSP_VAL R2_CLKSEL_DSP | R2_CLKSEL_DSP_IF +#define R2_CLKSEL_GFX (2 << 0) +#define R2_CM_CLKSEL_GFX_VAL R2_CLKSEL_GFX +#define R2_CLKSEL_MDM (6 << 0) +#define R2_CM_CLKSEL_MDM_VAL R2_CLKSEL_MDM + +/* 2430-Ratio Bootm (BYPASS) */ +#define RB_CLKSEL_L3 (1 << 0) +#define RB_CLKSEL_L4 (1 << 5) +#define RB_CLKSEL_USB (1 << 25) +#define RB_CM_CLKSEL1_CORE_VAL RB_CLKSEL_USB | RX_CLKSEL_SSI | \ + RX_CLKSEL_DSS2 | RX_CLKSEL_DSS1 | \ + RB_CLKSEL_L4 | RB_CLKSEL_L3 +#define RB_CLKSEL_MPU (1 << 0) +#define RB_CM_CLKSEL_MPU_VAL RB_CLKSEL_MPU +#define RB_CLKSEL_DSP (1 << 0) +#define RB_CLKSEL_DSP_IF (1 << 5) +#define RB_CM_CLKSEL_DSP_VAL RB_CLKSEL_DSP | RB_CLKSEL_DSP_IF +#define RB_CLKSEL_GFX (1 << 0) +#define RB_CM_CLKSEL_GFX_VAL RB_CLKSEL_GFX +#define RB_CLKSEL_MDM (1 << 0) +#define RB_CM_CLKSEL_MDM_VAL RB_CLKSEL_MDM + +/* 2420 Ratio Equivalents */ +#define RXX_CLKSEL_VLYNQ (0x12 << 15) +#define RXX_CLKSEL_SSI (0x8 << 20) + +/* 2420-PRCM III 532MHz core */ +#define RIII_CLKSEL_L3 (4 << 0) /* 133MHz */ +#define RIII_CLKSEL_L4 (2 << 5) /* 66.5MHz */ +#define RIII_CLKSEL_USB (4 << 25) /* 33.25MHz */ +#define RIII_CM_CLKSEL1_CORE_VAL RIII_CLKSEL_USB | RXX_CLKSEL_SSI | \ + RXX_CLKSEL_VLYNQ | RX_CLKSEL_DSS2 | \ + RX_CLKSEL_DSS1 | RIII_CLKSEL_L4 | \ + RIII_CLKSEL_L3 +#define RIII_CLKSEL_MPU (2 << 0) /* 266MHz */ +#define RIII_CM_CLKSEL_MPU_VAL RIII_CLKSEL_MPU +#define RIII_CLKSEL_DSP (3 << 0) /* c5x - 177.3MHz */ +#define RIII_CLKSEL_DSP_IF (2 << 5) /* c5x - 88.67MHz */ +#define RIII_SYNC_DSP (1 << 7) /* Enable sync */ +#define RIII_CLKSEL_IVA (6 << 8) /* iva1 - 88.67MHz */ +#define RIII_SYNC_IVA (1 << 13) /* Enable sync */ +#define RIII_CM_CLKSEL_DSP_VAL RIII_SYNC_IVA | RIII_CLKSEL_IVA | \ + RIII_SYNC_DSP | RIII_CLKSEL_DSP_IF | \ + RIII_CLKSEL_DSP +#define RIII_CLKSEL_GFX (2 << 0) /* 66.5MHz */ +#define RIII_CM_CLKSEL_GFX_VAL RIII_CLKSEL_GFX + +/* 2420-PRCM II 600MHz core */ +#define RII_CLKSEL_L3 (6 << 0) /* 100MHz */ +#define RII_CLKSEL_L4 (2 << 5) /* 50MHz */ +#define RII_CLKSEL_USB (2 << 25) /* 50MHz */ +#define RII_CM_CLKSEL1_CORE_VAL RII_CLKSEL_USB | \ + RXX_CLKSEL_SSI | RXX_CLKSEL_VLYNQ | \ + RX_CLKSEL_DSS2 | RX_CLKSEL_DSS1 | \ + RII_CLKSEL_L4 | RII_CLKSEL_L3 +#define RII_CLKSEL_MPU (2 << 0) /* 300MHz */ +#define RII_CM_CLKSEL_MPU_VAL RII_CLKSEL_MPU +#define RII_CLKSEL_DSP (3 << 0) /* c5x - 200MHz */ +#define RII_CLKSEL_DSP_IF (2 << 5) /* c5x - 100MHz */ +#define RII_SYNC_DSP (0 << 7) /* Bypass sync */ +#define RII_CLKSEL_IVA (6 << 8) /* iva1 - 200MHz */ +#define RII_SYNC_IVA (0 << 13) /* Bypass sync */ +#define RII_CM_CLKSEL_DSP_VAL RII_SYNC_IVA | RII_CLKSEL_IVA | \ + RII_SYNC_DSP | RII_CLKSEL_DSP_IF | \ + RII_CLKSEL_DSP +#define RII_CLKSEL_GFX (2 << 0) /* 50MHz */ +#define RII_CM_CLKSEL_GFX_VAL RII_CLKSEL_GFX + +/* 2420-PRCM VII (boot) */ +#define RVII_CLKSEL_L3 (1 << 0) +#define RVII_CLKSEL_L4 (1 << 5) +#define RVII_CLKSEL_DSS1 (1 << 8) +#define RVII_CLKSEL_DSS2 (0 << 13) +#define RVII_CLKSEL_VLYNQ (1 << 15) +#define RVII_CLKSEL_SSI (1 << 20) +#define RVII_CLKSEL_USB (1 << 25) + +#define RVII_CM_CLKSEL1_CORE_VAL RVII_CLKSEL_USB | RVII_CLKSEL_SSI | \ + RVII_CLKSEL_VLYNQ | RVII_CLKSEL_DSS2 | \ + RVII_CLKSEL_DSS1 | RVII_CLKSEL_L4 | RVII_CLKSEL_L3 + +#define RVII_CLKSEL_MPU (1 << 0) /* all divide by 1 */ +#define RVII_CM_CLKSEL_MPU_VAL RVII_CLKSEL_MPU + +#define RVII_CLKSEL_DSP (1 << 0) +#define RVII_CLKSEL_DSP_IF (1 << 5) +#define RVII_SYNC_DSP (0 << 7) +#define RVII_CLKSEL_IVA (1 << 8) +#define RVII_SYNC_IVA (0 << 13) +#define RVII_CM_CLKSEL_DSP_VAL RVII_SYNC_IVA | RVII_CLKSEL_IVA | RVII_SYNC_DSP | \ + RVII_CLKSEL_DSP_IF | RVII_CLKSEL_DSP + +#define RVII_CLKSEL_GFX (1 << 0) +#define RVII_CM_CLKSEL_GFX_VAL RVII_CLKSEL_GFX + +/*------------------------------------------------------------------------- + * 2430 Target modes: Along with each configuration the CPU has several + * modes which goes along with them. Modes mainly are the addition of + * describe DPLL combinations to go along with a ratio. + *-------------------------------------------------------------------------*/ + +/* Hardware governed */ +#define MX_48M_SRC (0 << 3) +#define MX_54M_SRC (0 << 5) +#define MX_APLLS_CLIKIN_12 (3 << 23) +#define MX_APLLS_CLIKIN_13 (2 << 23) +#define MX_APLLS_CLIKIN_19_2 (0 << 23) + +/* + * 2430 - standalone, 2*ref*M/(n+1), M/N is for exactness not relock speed + * #2 (ratio1) baseport-target + * #5a (ratio1) baseport-target, target DPLL = 266*2 = 532MHz + */ +#define M5A_DPLL_MULT_12 (133 << 12) +#define M5A_DPLL_DIV_12 (5 << 8) +#define M5A_CM_CLKSEL1_PLL_12_VAL MX_48M_SRC | MX_54M_SRC | \ + M5A_DPLL_DIV_12 | M5A_DPLL_MULT_12 | \ + MX_APLLS_CLIKIN_12 +#define M5A_DPLL_MULT_13 (266 << 12) +#define M5A_DPLL_DIV_13 (12 << 8) +#define M5A_CM_CLKSEL1_PLL_13_VAL MX_48M_SRC | MX_54M_SRC | \ + M5A_DPLL_DIV_13 | M5A_DPLL_MULT_13 | \ + MX_APLLS_CLIKIN_13 +#define M5A_DPLL_MULT_19 (180 << 12) +#define M5A_DPLL_DIV_19 (12 << 8) +#define M5A_CM_CLKSEL1_PLL_19_VAL MX_48M_SRC | MX_54M_SRC | \ + M5A_DPLL_DIV_19 | M5A_DPLL_MULT_19 | \ + MX_APLLS_CLIKIN_19_2 +/* #5b (ratio1) target DPLL = 200*2 = 400MHz */ +#define M5B_DPLL_MULT_12 (50 << 12) +#define M5B_DPLL_DIV_12 (2 << 8) +#define M5B_CM_CLKSEL1_PLL_12_VAL MX_48M_SRC | MX_54M_SRC | \ + M5B_DPLL_DIV_12 | M5B_DPLL_MULT_12 | \ + MX_APLLS_CLIKIN_12 +#define M5B_DPLL_MULT_13 (200 << 12) +#define M5B_DPLL_DIV_13 (12 << 8) + +#define M5B_CM_CLKSEL1_PLL_13_VAL MX_48M_SRC | MX_54M_SRC | \ + M5B_DPLL_DIV_13 | M5B_DPLL_MULT_13 | \ + MX_APLLS_CLIKIN_13 +#define M5B_DPLL_MULT_19 (125 << 12) +#define M5B_DPLL_DIV_19 (31 << 8) +#define M5B_CM_CLKSEL1_PLL_19_VAL MX_48M_SRC | MX_54M_SRC | \ + M5B_DPLL_DIV_19 | M5B_DPLL_MULT_19 | \ + MX_APLLS_CLIKIN_19_2 +/* + * #4 (ratio2) + * #3 (ratio2) baseport-target, target DPLL = 330*2 = 660MHz + */ +#define M3_DPLL_MULT_12 (55 << 12) +#define M3_DPLL_DIV_12 (1 << 8) +#define M3_CM_CLKSEL1_PLL_12_VAL MX_48M_SRC | MX_54M_SRC | \ + M3_DPLL_DIV_12 | M3_DPLL_MULT_12 | \ + MX_APLLS_CLIKIN_12 +#define M3_DPLL_MULT_13 (330 << 12) +#define M3_DPLL_DIV_13 (12 << 8) +#define M3_CM_CLKSEL1_PLL_13_VAL MX_48M_SRC | MX_54M_SRC | \ + M3_DPLL_DIV_13 | M3_DPLL_MULT_13 | \ + MX_APLLS_CLIKIN_13 +#define M3_DPLL_MULT_19 (275 << 12) +#define M3_DPLL_DIV_19 (15 << 8) +#define M3_CM_CLKSEL1_PLL_19_VAL MX_48M_SRC | MX_54M_SRC | \ + M3_DPLL_DIV_19 | M3_DPLL_MULT_19 | \ + MX_APLLS_CLIKIN_19_2 +/* boot (boot) */ +#define MB_DPLL_MULT (1 << 12) +#define MB_DPLL_DIV (0 << 8) +#define MB_CM_CLKSEL1_PLL_12_VAL MX_48M_SRC | MX_54M_SRC | MB_DPLL_DIV |\ + MB_DPLL_MULT | MX_APLLS_CLIKIN_12 + +#define MB_CM_CLKSEL1_PLL_13_VAL MX_48M_SRC | MX_54M_SRC | MB_DPLL_DIV |\ + MB_DPLL_MULT | MX_APLLS_CLIKIN_13 + +#define MB_CM_CLKSEL1_PLL_19_VAL MX_48M_SRC | MX_54M_SRC | MB_DPLL_DIV |\ + MB_DPLL_MULT | MX_APLLS_CLIKIN_19 + +/* + * 2430 - chassis (sedna) + * 165 (ratio1) same as above #2 + * 150 (ratio1) + * 133 (ratio2) same as above #4 + * 110 (ratio2) same as above #3 + * 104 (ratio2) + * boot (boot) + */ + +/* + * 2420 Equivalent - mode registers + * PRCM II , target DPLL = 2*300MHz = 600MHz + */ +#define MII_DPLL_MULT_12 (50 << 12) +#define MII_DPLL_DIV_12 (1 << 8) +#define MII_CM_CLKSEL1_PLL_12_VAL MX_48M_SRC | MX_54M_SRC | \ + MII_DPLL_DIV_12 | MII_DPLL_MULT_12 | \ + MX_APLLS_CLIKIN_12 +#define MII_DPLL_MULT_13 (300 << 12) +#define MII_DPLL_DIV_13 (12 << 8) +#define MII_CM_CLKSEL1_PLL_13_VAL MX_48M_SRC | MX_54M_SRC | \ + MII_DPLL_DIV_13 | MII_DPLL_MULT_13 | \ + MX_APLLS_CLIKIN_13 + +/* PRCM III target DPLL = 2*266 = 532MHz*/ +#define MIII_DPLL_MULT_12 (133 << 12) +#define MIII_DPLL_DIV_12 (5 << 8) +#define MIII_CM_CLKSEL1_PLL_12_VAL MX_48M_SRC | MX_54M_SRC | \ + MIII_DPLL_DIV_12 | MIII_DPLL_MULT_12 | \ + MX_APLLS_CLIKIN_12 +#define MIII_DPLL_MULT_13 (266 << 12) +#define MIII_DPLL_DIV_13 (12 << 8) +#define MIII_CM_CLKSEL1_PLL_13_VAL MX_48M_SRC | MX_54M_SRC | \ + MIII_DPLL_DIV_13 | MIII_DPLL_MULT_13 | \ + MX_APLLS_CLIKIN_13 + +/* PRCM VII (boot bypass) */ +#define MVII_CM_CLKSEL1_PLL_12_VAL MB_CM_CLKSEL1_PLL_12_VAL +#define MVII_CM_CLKSEL1_PLL_13_VAL MB_CM_CLKSEL1_PLL_13_VAL + +/* High and low operation value */ +#define MX_CLKSEL2_PLL_2x_VAL (2 << 0) +#define MX_CLKSEL2_PLL_1x_VAL (1 << 0) + +/* + * These represent optimal values for common parts, it won't work for all. + * As long as you scale down, most parameters are still work, they just + * become sub-optimal. The RFR value goes in the oppisite direction. If you + * don't adjust it down as your clock period increases the refresh interval + * will not be met. Setting all parameters for complete worst case may work, + * but may cut memory performance by 2x. Due to errata the DLLs need to be + * unlocked and their value needs run time calibration. A dynamic call is + * need for that as no single right value exists acorss production samples. + * + * Only the FULL speed values are given. Current code is such that rate + * changes must be made at DPLLoutx2. The actual value adjustment for low + * frequency operation will be handled by omap_set_performance() + * + * By having the boot loader boot up in the fastest L4 speed available likely + * will result in something which you can switch between. + */ +#define V24XX_SDRC_RFR_CTRL_133MHz (0x0003de00 | 1) +#define V24XX_SDRC_RFR_CTRL_100MHz (0x0002da01 | 1) +#define V24XX_SDRC_RFR_CTRL_110MHz (0x0002da01 | 1) /* Need to calc */ +#define V24XX_SDRC_RFR_CTRL_BYPASS (0x00005000 | 1) /* Need to calc */ + +/* MPU speed defines */ +#define S12M 12000000 +#define S13M 13000000 +#define S19M 19200000 +#define S26M 26000000 +#define S100M 100000000 +#define S133M 133000000 +#define S150M 150000000 +#define S165M 165000000 +#define S200M 200000000 +#define S266M 266000000 +#define S300M 300000000 +#define S330M 330000000 +#define S400M 400000000 +#define S532M 532000000 +#define S600M 600000000 +#define S660M 660000000 + +/*------------------------------------------------------------------------- + * Key dividers which make up a PRCM set. Ratio's for a PRCM are mandated. + * xtal_speed, dpll_speed, mpu_speed, CM_CLKSEL_MPU, + * CM_CLKSEL_DSP, CM_CLKSEL_GFX, CM_CLKSEL1_CORE, CM_CLKSEL1_PLL, + * CM_CLKSEL2_PLL, CM_CLKSEL_MDM + * + * Filling in table based on H4 boards and 2430-SDPs variants available. + * There are quite a few more rates combinations which could be defined. + * + * When multiple values are defiend the start up will try and choose the + * fastest one. If a 'fast' value is defined, then automatically, the /2 + * one should be included as it can be used. Generally having more that + * one fast set does not make sense, as static timings need to be changed + * to change the set. The exception is the bypass setting which is + * availble for low power bypass. + * + * Note: This table needs to be sorted, fastest to slowest. + *-------------------------------------------------------------------------*/ +static struct prcm_config rate_table[] = { + /* PRCM II - FAST */ + {S12M, S600M, S300M, RII_CM_CLKSEL_MPU_VAL, /* 300MHz ARM */ + RII_CM_CLKSEL_DSP_VAL, RII_CM_CLKSEL_GFX_VAL, + RII_CM_CLKSEL1_CORE_VAL, MII_CM_CLKSEL1_PLL_12_VAL, + MX_CLKSEL2_PLL_2x_VAL, 0, V24XX_SDRC_RFR_CTRL_100MHz, + RATE_IN_242X}, + + {S13M, S600M, S300M, RII_CM_CLKSEL_MPU_VAL, /* 300MHz ARM */ + RII_CM_CLKSEL_DSP_VAL, RII_CM_CLKSEL_GFX_VAL, + RII_CM_CLKSEL1_CORE_VAL, MII_CM_CLKSEL1_PLL_13_VAL, + MX_CLKSEL2_PLL_2x_VAL, 0, V24XX_SDRC_RFR_CTRL_100MHz, + RATE_IN_242X}, + + /* PRCM III - FAST */ + {S12M, S532M, S266M, RIII_CM_CLKSEL_MPU_VAL, /* 266MHz ARM */ + RIII_CM_CLKSEL_DSP_VAL, RIII_CM_CLKSEL_GFX_VAL, + RIII_CM_CLKSEL1_CORE_VAL, MIII_CM_CLKSEL1_PLL_12_VAL, + MX_CLKSEL2_PLL_2x_VAL, 0, V24XX_SDRC_RFR_CTRL_133MHz, + RATE_IN_242X}, + + {S13M, S532M, S266M, RIII_CM_CLKSEL_MPU_VAL, /* 266MHz ARM */ + RIII_CM_CLKSEL_DSP_VAL, RIII_CM_CLKSEL_GFX_VAL, + RIII_CM_CLKSEL1_CORE_VAL, MIII_CM_CLKSEL1_PLL_13_VAL, + MX_CLKSEL2_PLL_2x_VAL, 0, V24XX_SDRC_RFR_CTRL_133MHz, + RATE_IN_242X}, + + /* PRCM II - SLOW */ + {S12M, S300M, S150M, RII_CM_CLKSEL_MPU_VAL, /* 150MHz ARM */ + RII_CM_CLKSEL_DSP_VAL, RII_CM_CLKSEL_GFX_VAL, + RII_CM_CLKSEL1_CORE_VAL, MII_CM_CLKSEL1_PLL_12_VAL, + MX_CLKSEL2_PLL_2x_VAL, 0, V24XX_SDRC_RFR_CTRL_100MHz, + RATE_IN_242X}, + + {S13M, S300M, S150M, RII_CM_CLKSEL_MPU_VAL, /* 150MHz ARM */ + RII_CM_CLKSEL_DSP_VAL, RII_CM_CLKSEL_GFX_VAL, + RII_CM_CLKSEL1_CORE_VAL, MII_CM_CLKSEL1_PLL_13_VAL, + MX_CLKSEL2_PLL_2x_VAL, 0, V24XX_SDRC_RFR_CTRL_100MHz, + RATE_IN_242X}, + + /* PRCM III - SLOW */ + {S12M, S266M, S133M, RIII_CM_CLKSEL_MPU_VAL, /* 133MHz ARM */ + RIII_CM_CLKSEL_DSP_VAL, RIII_CM_CLKSEL_GFX_VAL, + RIII_CM_CLKSEL1_CORE_VAL, MIII_CM_CLKSEL1_PLL_12_VAL, + MX_CLKSEL2_PLL_2x_VAL, 0, V24XX_SDRC_RFR_CTRL_133MHz, + RATE_IN_242X}, + + {S13M, S266M, S133M, RIII_CM_CLKSEL_MPU_VAL, /* 133MHz ARM */ + RIII_CM_CLKSEL_DSP_VAL, RIII_CM_CLKSEL_GFX_VAL, + RIII_CM_CLKSEL1_CORE_VAL, MIII_CM_CLKSEL1_PLL_13_VAL, + MX_CLKSEL2_PLL_2x_VAL, 0, V24XX_SDRC_RFR_CTRL_133MHz, + RATE_IN_242X}, + + /* PRCM-VII (boot-bypass) */ + {S12M, S12M, S12M, RVII_CM_CLKSEL_MPU_VAL, /* 12MHz ARM*/ + RVII_CM_CLKSEL_DSP_VAL, RVII_CM_CLKSEL_GFX_VAL, + RVII_CM_CLKSEL1_CORE_VAL, MVII_CM_CLKSEL1_PLL_12_VAL, + MX_CLKSEL2_PLL_2x_VAL, 0, V24XX_SDRC_RFR_CTRL_BYPASS, + RATE_IN_242X}, + + /* PRCM-VII (boot-bypass) */ + {S13M, S13M, S13M, RVII_CM_CLKSEL_MPU_VAL, /* 13MHz ARM */ + RVII_CM_CLKSEL_DSP_VAL, RVII_CM_CLKSEL_GFX_VAL, + RVII_CM_CLKSEL1_CORE_VAL, MVII_CM_CLKSEL1_PLL_13_VAL, + MX_CLKSEL2_PLL_2x_VAL, 0, V24XX_SDRC_RFR_CTRL_BYPASS, + RATE_IN_242X}, + + /* PRCM #3 - ratio2 (ES2) - FAST */ + {S13M, S660M, S330M, R2_CM_CLKSEL_MPU_VAL, /* 330MHz ARM */ + R2_CM_CLKSEL_DSP_VAL, R2_CM_CLKSEL_GFX_VAL, + R2_CM_CLKSEL1_CORE_VAL, M3_CM_CLKSEL1_PLL_13_VAL, + MX_CLKSEL2_PLL_2x_VAL, R2_CM_CLKSEL_MDM_VAL, + V24XX_SDRC_RFR_CTRL_110MHz, + RATE_IN_243X}, + + /* PRCM #5a - ratio1 - FAST */ + {S13M, S532M, S266M, R1_CM_CLKSEL_MPU_VAL, /* 266MHz ARM */ + R1_CM_CLKSEL_DSP_VAL, R1_CM_CLKSEL_GFX_VAL, + R1_CM_CLKSEL1_CORE_VAL, M5A_CM_CLKSEL1_PLL_13_VAL, + MX_CLKSEL2_PLL_2x_VAL, R1_CM_CLKSEL_MDM_VAL, + V24XX_SDRC_RFR_CTRL_133MHz, + RATE_IN_243X}, + + /* PRCM #5b - ratio1 - FAST */ + {S13M, S400M, S200M, R1_CM_CLKSEL_MPU_VAL, /* 200MHz ARM */ + R1_CM_CLKSEL_DSP_VAL, R1_CM_CLKSEL_GFX_VAL, + R1_CM_CLKSEL1_CORE_VAL, M5B_CM_CLKSEL1_PLL_13_VAL, + MX_CLKSEL2_PLL_2x_VAL, R1_CM_CLKSEL_MDM_VAL, + V24XX_SDRC_RFR_CTRL_100MHz, + RATE_IN_243X}, + + /* PRCM #3 - ratio2 (ES2) - SLOW */ + {S13M, S330M, S165M, R2_CM_CLKSEL_MPU_VAL, /* 165MHz ARM */ + R2_CM_CLKSEL_DSP_VAL, R2_CM_CLKSEL_GFX_VAL, + R2_CM_CLKSEL1_CORE_VAL, M3_CM_CLKSEL1_PLL_13_VAL, + MX_CLKSEL2_PLL_1x_VAL, R2_CM_CLKSEL_MDM_VAL, + V24XX_SDRC_RFR_CTRL_110MHz, + RATE_IN_243X}, + + /* PRCM #5a - ratio1 - SLOW */ + {S13M, S266M, S133M, R1_CM_CLKSEL_MPU_VAL, /* 133MHz ARM */ + R1_CM_CLKSEL_DSP_VAL, R1_CM_CLKSEL_GFX_VAL, + R1_CM_CLKSEL1_CORE_VAL, M5A_CM_CLKSEL1_PLL_13_VAL, + MX_CLKSEL2_PLL_1x_VAL, R1_CM_CLKSEL_MDM_VAL, + V24XX_SDRC_RFR_CTRL_133MHz, + RATE_IN_243X}, + + /* PRCM #5b - ratio1 - SLOW*/ + {S13M, S200M, S100M, R1_CM_CLKSEL_MPU_VAL, /* 100MHz ARM */ + R1_CM_CLKSEL_DSP_VAL, R1_CM_CLKSEL_GFX_VAL, + R1_CM_CLKSEL1_CORE_VAL, M5B_CM_CLKSEL1_PLL_13_VAL, + MX_CLKSEL2_PLL_1x_VAL, R1_CM_CLKSEL_MDM_VAL, + V24XX_SDRC_RFR_CTRL_100MHz, + RATE_IN_243X}, + + /* PRCM-boot/bypass */ + {S13M, S13M, S13M, RB_CM_CLKSEL_MPU_VAL, /* 13Mhz */ + RB_CM_CLKSEL_DSP_VAL, RB_CM_CLKSEL_GFX_VAL, + RB_CM_CLKSEL1_CORE_VAL, MB_CM_CLKSEL1_PLL_13_VAL, + MX_CLKSEL2_PLL_2x_VAL, RB_CM_CLKSEL_MDM_VAL, + V24XX_SDRC_RFR_CTRL_BYPASS, + RATE_IN_243X}, + + /* PRCM-boot/bypass */ + {S12M, S12M, S12M, RB_CM_CLKSEL_MPU_VAL, /* 12Mhz */ + RB_CM_CLKSEL_DSP_VAL, RB_CM_CLKSEL_GFX_VAL, + RB_CM_CLKSEL1_CORE_VAL, MB_CM_CLKSEL1_PLL_12_VAL, + MX_CLKSEL2_PLL_2x_VAL, RB_CM_CLKSEL_MDM_VAL, + V24XX_SDRC_RFR_CTRL_BYPASS, + RATE_IN_243X}, + + { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0}, +}; + +/*------------------------------------------------------------------------- + * 24xx clock tree. + * + * NOTE:In many cases here we are assigning a 'default' parent. In many + * cases the parent is selectable. The get/set parent calls will also + * switch sources. + * + * Many some clocks say always_enabled, but they can be auto idled for + * power savings. They will always be available upon clock request. + * + * Several sources are given initial rates which may be wrong, this will + * be fixed up in the init func. + * + * Things are broadly separated below by clock domains. It is + * noteworthy that most periferals have dependencies on multiple clock + * domains. Many get their interface clocks from the L4 domain, but get + * functional clocks from fixed sources or other core domain derived + * clocks. + *-------------------------------------------------------------------------*/ + +/* Base external input clocks */ +static struct clk func_32k_ck = { + .name = "func_32k_ck", + .rate = 32000, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + RATE_FIXED | ALWAYS_ENABLED, +}; + +/* Typical 12/13MHz in standalone mode, will be 26Mhz in chassis mode */ +static struct clk osc_ck = { /* (*12, *13, 19.2, *26, 38.4)MHz */ + .name = "osc_ck", + .rate = 26000000, /* fixed up in clock init */ + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + RATE_FIXED | ALWAYS_ENABLED | RATE_PROPAGATES, +}; + +/* With out modem likely 12MHz, with modem likely 13MHz */ +static struct clk sys_ck = { /* (*12, *13, 19.2, 26, 38.4)MHz */ + .name = "sys_ck", /* ~ ref_clk also */ + .parent = &osc_ck, + .rate = 13000000, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + RATE_FIXED | ALWAYS_ENABLED | RATE_PROPAGATES, + .rate_offset = 6, /* sysclkdiv 1 or 2, already handled or no boot */ + .recalc = &omap2_sys_clk_recalc, +}; + +static struct clk alt_ck = { /* Typical 54M or 48M, may not exist */ + .name = "alt_ck", + .rate = 54000000, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + RATE_FIXED | ALWAYS_ENABLED | RATE_PROPAGATES, + .recalc = &omap2_propagate_rate, +}; + +/* + * Analog domain root source clocks + */ + +/* dpll_ck, is broken out in to special cases through clksel */ +static struct clk dpll_ck = { + .name = "dpll_ck", + .parent = &sys_ck, /* Can be func_32k also */ + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + RATE_PROPAGATES | RATE_CKCTL | CM_PLL_SEL1, + .recalc = &omap2_clksel_recalc, +}; + +static struct clk apll96_ck = { + .name = "apll96_ck", + .parent = &sys_ck, + .rate = 96000000, + .flags = CLOCK_IN_OMAP242X |CLOCK_IN_OMAP243X | + RATE_FIXED | RATE_PROPAGATES, + .enable_reg = (void __iomem *)&CM_CLKEN_PLL, + .enable_bit = 0x2, + .recalc = &omap2_propagate_rate, +}; + +static struct clk apll54_ck = { + .name = "apll54_ck", + .parent = &sys_ck, + .rate = 54000000, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + RATE_FIXED | RATE_PROPAGATES, + .enable_reg = (void __iomem *)&CM_CLKEN_PLL, + .enable_bit = 0x6, + .recalc = &omap2_propagate_rate, +}; + +/* + * PRCM digital base sources + */ +static struct clk func_54m_ck = { + .name = "func_54m_ck", + .parent = &apll54_ck, /* can also be alt_clk */ + .rate = 54000000, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + RATE_FIXED | CM_PLL_SEL1 | RATE_PROPAGATES, + .src_offset = 5, + .enable_reg = (void __iomem *)&CM_CLKEN_PLL, + .enable_bit = 0xff, + .recalc = &omap2_propagate_rate, +}; + +static struct clk core_ck = { + .name = "core_ck", + .parent = &dpll_ck, /* can also be 32k */ + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + ALWAYS_ENABLED | RATE_PROPAGATES, + .recalc = &omap2_propagate_rate, +}; + +static struct clk sleep_ck = { /* sys_clk or 32k */ + .name = "sleep_ck", + .parent = &func_32k_ck, + .rate = 32000, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .recalc = &omap2_propagate_rate, +}; + +static struct clk func_96m_ck = { + .name = "func_96m_ck", + .parent = &apll96_ck, + .rate = 96000000, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + RATE_FIXED | RATE_PROPAGATES, + .enable_reg = (void __iomem *)&CM_CLKEN_PLL, + .enable_bit = 0xff, + .recalc = &omap2_propagate_rate, +}; + +static struct clk func_48m_ck = { + .name = "func_48m_ck", + .parent = &apll96_ck, /* 96M or Alt */ + .rate = 48000000, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + RATE_FIXED | CM_PLL_SEL1 | RATE_PROPAGATES, + .src_offset = 3, + .enable_reg = (void __iomem *)&CM_CLKEN_PLL, + .enable_bit = 0xff, + .recalc = &omap2_propagate_rate, +}; + +static struct clk func_12m_ck = { + .name = "func_12m_ck", + .parent = &func_48m_ck, + .rate = 12000000, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + RATE_FIXED | RATE_PROPAGATES, + .recalc = &omap2_propagate_rate, + .enable_reg = (void __iomem *)&CM_CLKEN_PLL, + .enable_bit = 0xff, +}; + +/* Secure timer, only available in secure mode */ +static struct clk wdt1_osc_ck = { + .name = "ck_wdt1_osc", + .parent = &osc_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk sys_clkout = { + .name = "sys_clkout", + .parent = &func_54m_ck, + .rate = 54000000, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + CM_SYSCLKOUT_SEL1 | RATE_CKCTL, + .src_offset = 0, + .enable_reg = (void __iomem *)&PRCM_CLKOUT_CTRL, + .enable_bit = 7, + .rate_offset = 3, + .recalc = &omap2_clksel_recalc, +}; + +/* In 2430, new in 2420 ES2 */ +static struct clk sys_clkout2 = { + .name = "sys_clkout2", + .parent = &func_54m_ck, + .rate = 54000000, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + CM_SYSCLKOUT_SEL1 | RATE_CKCTL, + .src_offset = 8, + .enable_reg = (void __iomem *)&PRCM_CLKOUT_CTRL, + .enable_bit = 15, + .rate_offset = 11, + .recalc = &omap2_clksel_recalc, +}; + +/* + * MPU clock domain + * Clocks: + * MPU_FCLK, MPU_ICLK + * INT_M_FCLK, INT_M_I_CLK + * + * - Individual clocks are hardware managed. + * - Base divider comes from: CM_CLKSEL_MPU + * + */ +static struct clk mpu_ck = { /* Control cpu */ + .name = "mpu_ck", + .parent = &core_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | RATE_CKCTL | + ALWAYS_ENABLED | CM_MPU_SEL1 | DELAYED_APP | + CONFIG_PARTICIPANT | RATE_PROPAGATES, + .rate_offset = 0, /* bits 0-4 */ + .recalc = &omap2_clksel_recalc, +}; + +/* + * DSP (2430-IVA2.1) (2420-UMA+IVA1) clock domain + * Clocks: + * 2430: IVA2.1_FCLK, IVA2.1_ICLK + * 2420: UMA_FCLK, UMA_ICLK, IVA_MPU, IVA_COP + */ +static struct clk iva2_1_fck = { + .name = "iva2_1_fck", + .parent = &core_ck, + .flags = CLOCK_IN_OMAP243X | RATE_CKCTL | CM_DSP_SEL1 | + DELAYED_APP | RATE_PROPAGATES | + CONFIG_PARTICIPANT, + .rate_offset = 0, + .enable_reg = (void __iomem *)&CM_FCLKEN_DSP, + .enable_bit = 0, + .recalc = &omap2_clksel_recalc, +}; + +static struct clk iva2_1_ick = { + .name = "iva2_1_ick", + .parent = &iva2_1_fck, + .flags = CLOCK_IN_OMAP243X | RATE_CKCTL | CM_DSP_SEL1 | + DELAYED_APP | CONFIG_PARTICIPANT, + .rate_offset = 5, + .recalc = &omap2_clksel_recalc, +}; + +/* + * Won't be too specific here. The core clock comes into this block + * it is divided then tee'ed. One branch goes directly to xyz enable + * controls. The other branch gets further divided by 2 then possibly + * routed into a synchronizer and out of clocks abc. + */ +static struct clk dsp_fck = { + .name = "dsp_fck", + .parent = &core_ck, + .flags = CLOCK_IN_OMAP242X | RATE_CKCTL | CM_DSP_SEL1 | + DELAYED_APP | CONFIG_PARTICIPANT | RATE_PROPAGATES, + .rate_offset = 0, + .enable_reg = (void __iomem *)&CM_FCLKEN_DSP, + .enable_bit = 0, + .recalc = &omap2_clksel_recalc, +}; + +static struct clk dsp_ick = { + .name = "dsp_ick", /* apparently ipi and isp */ + .parent = &dsp_fck, + .flags = CLOCK_IN_OMAP242X | RATE_CKCTL | CM_DSP_SEL1 | + DELAYED_APP | CONFIG_PARTICIPANT, + .rate_offset = 5, + .enable_reg = (void __iomem *)&CM_ICLKEN_DSP, + .enable_bit = 1, /* for ipi */ + .recalc = &omap2_clksel_recalc, +}; + +static struct clk iva1_ifck = { + .name = "iva1_ifck", + .parent = &core_ck, + .flags = CLOCK_IN_OMAP242X | CM_DSP_SEL1 | RATE_CKCTL | + CONFIG_PARTICIPANT | RATE_PROPAGATES | DELAYED_APP, + .rate_offset= 8, + .enable_reg = (void __iomem *)&CM_FCLKEN_DSP, + .enable_bit = 10, + .recalc = &omap2_clksel_recalc, +}; + +/* IVA1 mpu/int/i/f clocks are /2 of parent */ +static struct clk iva1_mpu_int_ifck = { + .name = "iva1_mpu_int_ifck", + .parent = &iva1_ifck, + .flags = CLOCK_IN_OMAP242X | RATE_CKCTL | CM_DSP_SEL1, + .enable_reg = (void __iomem *)&CM_FCLKEN_DSP, + .enable_bit = 8, + .recalc = &omap2_clksel_recalc, +}; + +/* + * L3 clock domain + * L3 clocks are used for both interface and functional clocks to + * multiple entities. Some of these clocks are completely managed + * by hardware, and some others allow software control. Hardware + * managed ones general are based on directly CLK_REQ signals and + * various auto idle settings. The functional spec sets many of these + * as 'tie-high' for their enables. + * + * I-CLOCKS: + * L3-Interconnect, SMS, GPMC, SDRC, OCM_RAM, OCM_ROM, SDMA + * CAM, HS-USB. + * F-CLOCK + * SSI. + * + * GPMC memories and SDRC have timing and clock sensitive registers which + * may very well need notification when the clock changes. Currently for low + * operating points, these are taken care of in sleep.S. + */ +static struct clk core_l3_ck = { /* Used for ick and fck, interconnect */ + .name = "core_l3_ck", + .parent = &core_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + RATE_CKCTL | ALWAYS_ENABLED | CM_CORE_SEL1 | + DELAYED_APP | CONFIG_PARTICIPANT | + RATE_PROPAGATES, + .rate_offset = 0, + .recalc = &omap2_clksel_recalc, +}; + +static struct clk usb_l4_ick = { /* FS-USB interface clock */ + .name = "usb_l4_ick", + .parent = &core_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + RATE_CKCTL | CM_CORE_SEL1 | DELAYED_APP | + CONFIG_PARTICIPANT, + .enable_reg = (void __iomem *)&CM_ICLKEN2_CORE, + .enable_bit = 0, + .rate_offset = 25, + .recalc = &omap2_clksel_recalc, +}; + +/* + * SSI is in L3 management domain, its direct parent is core not l3, + * many core power domain entities are grouped into the L3 clock + * domain. + * SSI_SSR_FCLK, SSI_SST_FCLK, SSI_L4_CLIK + * + * ssr = core/1/2/3/4/5, sst = 1/2 ssr. + */ +static struct clk ssi_ssr_sst_fck = { + .name = "ssi_fck", + .parent = &core_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + RATE_CKCTL | CM_CORE_SEL1 | DELAYED_APP, + .enable_reg = (void __iomem *)&CM_FCLKEN2_CORE, /* bit 1 */ + .enable_bit = 1, + .rate_offset = 20, + .recalc = &omap2_clksel_recalc, +}; + +/* + * GFX clock domain + * Clocks: + * GFX_FCLK, GFX_ICLK + * GFX_CG1(2d), GFX_CG2(3d) + * + * GFX_FCLK runs from L3, and is divided by (1,2,3,4) + * The 2d and 3d clocks run at a hardware determined + * divided value of fclk. + * + */ +static struct clk gfx_3d_fck = { + .name = "gfx_3d_fck", + .parent = &core_l3_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + RATE_CKCTL | CM_GFX_SEL1, + .enable_reg = (void __iomem *)&CM_FCLKEN_GFX, + .enable_bit = 2, + .rate_offset= 0, + .recalc = &omap2_clksel_recalc, +}; + +static struct clk gfx_2d_fck = { + .name = "gfx_2d_fck", + .parent = &core_l3_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + RATE_CKCTL | CM_GFX_SEL1, + .enable_reg = (void __iomem *)&CM_FCLKEN_GFX, + .enable_bit = 1, + .rate_offset= 0, + .recalc = &omap2_clksel_recalc, +}; + +static struct clk gfx_ick = { + .name = "gfx_ick", /* From l3 */ + .parent = &core_l3_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + RATE_CKCTL, + .enable_reg = (void __iomem *)&CM_ICLKEN_GFX, /* bit 0 */ + .enable_bit = 0, + .recalc = &omap2_followparent_recalc, +}; + +/* + * Modem clock domain (2430) + * CLOCKS: + * MDM_OSC_CLK + * MDM_ICLK + */ +static struct clk mdm_ick = { /* used both as a ick and fck */ + .name = "mdm_ick", + .parent = &core_ck, + .flags = CLOCK_IN_OMAP243X | RATE_CKCTL | CM_MODEM_SEL1 | + DELAYED_APP | CONFIG_PARTICIPANT, + .rate_offset = 0, + .enable_reg = (void __iomem *)&CM_ICLKEN_MDM, + .enable_bit = 0, + .recalc = &omap2_clksel_recalc, +}; + +static struct clk mdm_osc_ck = { + .name = "mdm_osc_ck", + .rate = 26000000, + .parent = &osc_ck, + .flags = CLOCK_IN_OMAP243X | RATE_FIXED, + .enable_reg = (void __iomem *)&CM_FCLKEN_MDM, + .enable_bit = 1, + .recalc = &omap2_followparent_recalc, +}; + +/* + * L4 clock management domain + * + * This domain contains lots of interface clocks from the L4 interface, some + * functional clocks. Fixed APLL functional source clocks are managed in + * this domain. + */ +static struct clk l4_ck = { /* used both as an ick and fck */ + .name = "l4_ck", + .parent = &core_l3_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + RATE_CKCTL | ALWAYS_ENABLED | CM_CORE_SEL1 | + DELAYED_APP | RATE_PROPAGATES, + .rate_offset = 5, + .recalc = &omap2_clksel_recalc, +}; + +static struct clk ssi_l4_ick = { + .name = "ssi_l4_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | RATE_CKCTL, + .enable_reg = (void __iomem *)&CM_ICLKEN2_CORE, /* bit 1 */ + .enable_bit = 1, + .recalc = &omap2_followparent_recalc, +}; + +/* + * DSS clock domain + * CLOCKs: + * DSS_L4_ICLK, DSS_L3_ICLK, + * DSS_CLK1, DSS_CLK2, DSS_54MHz_CLK + * + * DSS is both initiator and target. + */ +static struct clk dss_ick = { /* Enables both L3,L4 ICLK's */ + .name = "dss_ick", + .parent = &l4_ck, /* really both l3 and l4 */ + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | RATE_CKCTL, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .enable_bit = 0, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk dss1_fck = { + .name = "dss1_fck", + .parent = &core_ck, /* Core or sys */ + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + RATE_CKCTL | CM_CORE_SEL1 | DELAYED_APP, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 0, + .rate_offset = 8, + .src_offset = 8, + .recalc = &omap2_clksel_recalc, +}; + +static struct clk dss2_fck = { /* Alt clk used in power management */ + .name = "dss2_fck", + .parent = &sys_ck, /* fixed at sys_ck or 48MHz */ + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + RATE_CKCTL | CM_CORE_SEL1 | RATE_FIXED, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 1, + .src_offset = 13, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk dss_54m_fck = { /* Alt clk used in power management */ + .name = "dss_54m_fck", /* 54m tv clk */ + .parent = &func_54m_ck, + .rate = 54000000, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + RATE_FIXED | RATE_PROPAGATES, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 2, + .recalc = &omap2_propagate_rate, +}; + +/* + * CORE power domain ICLK & FCLK defines. + * Many of the these can have more than one possible parent. Entries + * here will likely have an L4 interface parent, and may have multiple + * functional clock parents. + */ +static struct clk gpt1_ick = { + .name = "gpt1_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN_WKUP, /* Bit4 */ + .enable_bit = 0, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt1_fck = { + .name = "gpt1_fck", + .parent = &func_32k_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + CM_WKUP_SEL1, + .enable_reg = (void __iomem *)&CM_FCLKEN_WKUP, + .enable_bit = 0, + .src_offset = 0, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt2_ick = { + .name = "gpt2_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, /* bit4 */ + .enable_bit = 0, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt2_fck = { + .name = "gpt2_fck", + .parent = &func_32k_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + CM_CORE_SEL2, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 4, + .src_offset = 2, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt3_ick = { + .name = "gpt3_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, /* Bit5 */ + .enable_bit = 5, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt3_fck = { + .name = "gpt3_fck", + .parent = &func_32k_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + CM_CORE_SEL2, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 5, + .src_offset = 4, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt4_ick = { + .name = "gpt4_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, /* Bit6 */ + .enable_bit = 6, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt4_fck = { + .name = "gpt4_fck", + .parent = &func_32k_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + CM_CORE_SEL2, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 6, + .src_offset = 6, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt5_ick = { + .name = "gpt5_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, /* Bit7 */ + .enable_bit = 7, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt5_fck = { + .name = "gpt5_fck", + .parent = &func_32k_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + CM_CORE_SEL2, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 7, + .src_offset = 8, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt6_ick = { + .name = "gpt6_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_bit = 8, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, /* bit8 */ + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt6_fck = { + .name = "gpt6_fck", + .parent = &func_32k_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + CM_CORE_SEL2, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 8, + .src_offset = 10, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt7_ick = { + .name = "gpt7_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, /* bit9 */ + .enable_bit = 9, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt7_fck = { + .name = "gpt7_fck", + .parent = &func_32k_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + CM_CORE_SEL2, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 9, + .src_offset = 12, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt8_ick = { + .name = "gpt8_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, /* bit10 */ + .enable_bit = 10, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt8_fck = { + .name = "gpt8_fck", + .parent = &func_32k_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + CM_CORE_SEL2, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 10, + .src_offset = 14, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt9_ick = { + .name = "gpt9_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .enable_bit = 11, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt9_fck = { + .name = "gpt9_fck", + .parent = &func_32k_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + CM_CORE_SEL2, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 11, + .src_offset = 16, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt10_ick = { + .name = "gpt10_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .enable_bit = 12, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt10_fck = { + .name = "gpt10_fck", + .parent = &func_32k_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + CM_CORE_SEL2, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 12, + .src_offset = 18, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt11_ick = { + .name = "gpt11_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .enable_bit = 13, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt11_fck = { + .name = "gpt11_fck", + .parent = &func_32k_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + CM_CORE_SEL2, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 13, + .src_offset = 20, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt12_ick = { + .name = "gpt12_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, /* bit14 */ + .enable_bit = 14, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpt12_fck = { + .name = "gpt12_fck", + .parent = &func_32k_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + CM_CORE_SEL2, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 14, + .src_offset = 22, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mcbsp1_ick = { + .name = "mcbsp1_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_bit = 15, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, /* bit16 */ + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mcbsp1_fck = { + .name = "mcbsp1_fck", + .parent = &func_96m_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_bit = 15, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mcbsp2_ick = { + .name = "mcbsp2_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_bit = 16, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mcbsp2_fck = { + .name = "mcbsp2_fck", + .parent = &func_96m_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_bit = 16, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mcbsp3_ick = { + .name = "mcbsp3_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN2_CORE, + .enable_bit = 3, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mcbsp3_fck = { + .name = "mcbsp3_fck", + .parent = &func_96m_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN2_CORE, + .enable_bit = 3, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mcbsp4_ick = { + .name = "mcbsp4_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN2_CORE, + .enable_bit = 4, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mcbsp4_fck = { + .name = "mcbsp4_fck", + .parent = &func_96m_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN2_CORE, + .enable_bit = 4, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mcbsp5_ick = { + .name = "mcbsp5_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN2_CORE, + .enable_bit = 5, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mcbsp5_fck = { + .name = "mcbsp5_fck", + .parent = &func_96m_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN2_CORE, + .enable_bit = 5, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mcspi1_ick = { + .name = "mcspi1_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .enable_bit = 17, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mcspi1_fck = { + .name = "mcspi1_fck", + .parent = &func_48m_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 17, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mcspi2_ick = { + .name = "mcspi2_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .enable_bit = 18, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mcspi2_fck = { + .name = "mcspi2_fck", + .parent = &func_48m_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 18, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mcspi3_ick = { + .name = "mcspi3_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN2_CORE, + .enable_bit = 9, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mcspi3_fck = { + .name = "mcspi3_fck", + .parent = &func_48m_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN2_CORE, + .enable_bit = 9, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk uart1_ick = { + .name = "uart1_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .enable_bit = 21, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk uart1_fck = { + .name = "uart1_fck", + .parent = &func_48m_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 21, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk uart2_ick = { + .name = "uart2_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .enable_bit = 22, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk uart2_fck = { + .name = "uart2_fck", + .parent = &func_48m_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 22, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk uart3_ick = { + .name = "uart3_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN2_CORE, + .enable_bit = 2, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk uart3_fck = { + .name = "uart3_fck", + .parent = &func_48m_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN2_CORE, + .enable_bit = 2, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpios_ick = { + .name = "gpios_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN_WKUP, + .enable_bit = 2, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpios_fck = { + .name = "gpios_fck", + .parent = &func_32k_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN_WKUP, + .enable_bit = 2, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mpu_wdt_ick = { + .name = "mpu_wdt_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN_WKUP, + .enable_bit = 3, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mpu_wdt_fck = { + .name = "mpu_wdt_fck", + .parent = &func_32k_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN_WKUP, + .enable_bit = 3, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk sync_32k_ick = { + .name = "sync_32k_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN_WKUP, + .enable_bit = 1, + .recalc = &omap2_followparent_recalc, +}; +static struct clk wdt1_ick = { + .name = "wdt1_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN_WKUP, + .enable_bit = 4, + .recalc = &omap2_followparent_recalc, +}; +static struct clk omapctrl_ick = { + .name = "omapctrl_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN_WKUP, + .enable_bit = 5, + .recalc = &omap2_followparent_recalc, +}; +static struct clk icr_ick = { + .name = "icr_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN_WKUP, + .enable_bit = 6, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk cam_ick = { + .name = "cam_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .enable_bit = 31, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk cam_fck = { + .name = "cam_fck", + .parent = &func_96m_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 31, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mailboxes_ick = { + .name = "mailboxes_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .enable_bit = 30, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk wdt4_ick = { + .name = "wdt4_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .enable_bit = 29, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk wdt4_fck = { + .name = "wdt4_fck", + .parent = &func_32k_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 29, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk wdt3_ick = { + .name = "wdt3_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .enable_bit = 28, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk wdt3_fck = { + .name = "wdt3_fck", + .parent = &func_32k_ck, + .flags = CLOCK_IN_OMAP242X, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 28, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mspro_ick = { + .name = "mspro_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .enable_bit = 27, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mspro_fck = { + .name = "mspro_fck", + .parent = &func_96m_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 27, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mmc_ick = { + .name = "mmc_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .enable_bit = 26, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mmc_fck = { + .name = "mmc_fck", + .parent = &func_96m_ck, + .flags = CLOCK_IN_OMAP242X, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 26, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk fac_ick = { + .name = "fac_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .enable_bit = 25, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk fac_fck = { + .name = "fac_fck", + .parent = &func_12m_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 25, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk eac_ick = { + .name = "eac_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .enable_bit = 24, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk eac_fck = { + .name = "eac_fck", + .parent = &func_96m_ck, + .flags = CLOCK_IN_OMAP242X, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 24, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk hdq_ick = { + .name = "hdq_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .enable_bit = 23, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk hdq_fck = { + .name = "hdq_fck", + .parent = &func_12m_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 23, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk i2c2_ick = { + .name = "i2c2_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .enable_bit = 20, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk i2c2_fck = { + .name = "i2c2_fck", + .parent = &func_12m_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 20, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk i2chs2_fck = { + .name = "i2chs2_fck", + .parent = &func_96m_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN2_CORE, + .enable_bit = 20, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk i2c1_ick = { + .name = "i2c1_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .enable_bit = 19, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk i2c1_fck = { + .name = "i2c1_fck", + .parent = &func_12m_ck, + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 19, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk i2chs1_fck = { + .name = "i2chs1_fck", + .parent = &func_96m_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN2_CORE, + .enable_bit = 19, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk vlynq_ick = { + .name = "vlynq_ick", + .parent = &core_l3_ck, + .flags = CLOCK_IN_OMAP242X, + .enable_reg = (void __iomem *)&CM_ICLKEN1_CORE, + .enable_bit = 3, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk vlynq_fck = { + .name = "vlynq_fck", + .parent = &func_96m_ck, + .flags = CLOCK_IN_OMAP242X | RATE_CKCTL | CM_CORE_SEL1 | DELAYED_APP, + .enable_reg = (void __iomem *)&CM_FCLKEN1_CORE, + .enable_bit = 3, + .src_offset = 15, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk sdrc_ick = { + .name = "sdrc_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN3_CORE, + .enable_bit = 2, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk des_ick = { + .name = "des_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP243X | CLOCK_IN_OMAP242X, + .enable_reg = (void __iomem *)&CM_ICLKEN4_CORE, + .enable_bit = 0, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk sha_ick = { + .name = "sha_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP243X | CLOCK_IN_OMAP242X, + .enable_reg = (void __iomem *)&CM_ICLKEN4_CORE, + .enable_bit = 1, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk rng_ick = { + .name = "rng_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP243X | CLOCK_IN_OMAP242X, + .enable_reg = (void __iomem *)&CM_ICLKEN4_CORE, + .enable_bit = 2, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk aes_ick = { + .name = "aes_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP243X | CLOCK_IN_OMAP242X, + .enable_reg = (void __iomem *)&CM_ICLKEN4_CORE, + .enable_bit = 3, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk pka_ick = { + .name = "pka_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP243X | CLOCK_IN_OMAP242X, + .enable_reg = (void __iomem *)&CM_ICLKEN4_CORE, + .enable_bit = 4, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk usb_fck = { + .name = "usb_fck", + .parent = &func_48m_ck, + .flags = CLOCK_IN_OMAP243X | CLOCK_IN_OMAP242X, + .enable_reg = (void __iomem *)&CM_FCLKEN2_CORE, + .enable_bit = 0, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk usbhs_ick = { + .name = "usbhs_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN2_CORE, + .enable_bit = 6, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mmchs1_ick = { + .name = "mmchs1_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN2_CORE, + .enable_bit = 7, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mmchs1_fck = { + .name = "mmchs1_fck", + .parent = &func_96m_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN2_CORE, + .enable_bit = 7, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mmchs2_ick = { + .name = "mmchs2_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN2_CORE, + .enable_bit = 8, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mmchs2_fck = { + .name = "mmchs2_fck", + .parent = &func_96m_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN2_CORE, + .enable_bit = 8, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpio5_ick = { + .name = "gpio5_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN2_CORE, + .enable_bit = 10, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk gpio5_fck = { + .name = "gpio5_fck", + .parent = &func_32k_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN2_CORE, + .enable_bit = 10, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mdm_intc_ick = { + .name = "mdm_intc_ick", + .parent = &l4_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_ICLKEN2_CORE, + .enable_bit = 11, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mmchsdb1_fck = { + .name = "mmchsdb1_fck", + .parent = &func_32k_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN2_CORE, + .enable_bit = 16, + .recalc = &omap2_followparent_recalc, +}; + +static struct clk mmchsdb2_fck = { + .name = "mmchsdb2_fck", + .parent = &func_32k_ck, + .flags = CLOCK_IN_OMAP243X, + .enable_reg = (void __iomem *)&CM_FCLKEN2_CORE, + .enable_bit = 17, + .recalc = &omap2_followparent_recalc, +}; + +/* + * This clock is a composite clock which does entire set changes then + * forces a rebalance. It keys on the MPU speed, but it really could + * be any key speed part of a set in the rate table. + * + * to really change a set, you need memory table sets which get changed + * in sram, pre-notifiers & post notifiers, changing the top set, without + * having low level display recalc's won't work... this is why dpm notifiers + * work, isr's off, walk a list of clocks already _off_ and not messing with + * the bus. + * + * This clock should have no parent. It embodies the entire upper level + * active set. A parent will mess up some of the init also. + */ +static struct clk virt_prcm_set = { + .name = "virt_prcm_set", + .flags = CLOCK_IN_OMAP242X | CLOCK_IN_OMAP243X | + VIRTUAL_CLOCK | ALWAYS_ENABLED | DELAYED_APP, + .parent = &mpu_ck, /* Indexed by mpu speed, no parent */ + .recalc = &omap2_mpu_recalc, /* sets are keyed on mpu rate */ + .set_rate = &omap2_select_table_rate, + .round_rate = &omap2_round_to_table_rate, +}; + +static struct clk *onchip_clks[] = { + /* external root sources */ + &func_32k_ck, + &osc_ck, + &sys_ck, + &alt_ck, + /* internal analog sources */ + &dpll_ck, + &apll96_ck, + &apll54_ck, + /* internal prcm root sources */ + &func_54m_ck, + &core_ck, + &sleep_ck, + &func_96m_ck, + &func_48m_ck, + &func_12m_ck, + &wdt1_osc_ck, + &sys_clkout, + &sys_clkout2, + /* mpu domain clocks */ + &mpu_ck, + /* dsp domain clocks */ + &iva2_1_fck, /* 2430 */ + &iva2_1_ick, + &dsp_ick, /* 2420 */ + &dsp_fck, + &iva1_ifck, + &iva1_mpu_int_ifck, + /* GFX domain clocks */ + &gfx_3d_fck, + &gfx_2d_fck, + &gfx_ick, + /* Modem domain clocks */ + &mdm_ick, + &mdm_osc_ck, + /* DSS domain clocks */ + &dss_ick, + &dss1_fck, + &dss2_fck, + &dss_54m_fck, + /* L3 domain clocks */ + &core_l3_ck, + &ssi_ssr_sst_fck, + &usb_l4_ick, + /* L4 domain clocks */ + &l4_ck, /* used as both core_l4 and wu_l4 */ + &ssi_l4_ick, + /* virtual meta-group clock */ + &virt_prcm_set, + /* general l4 interface ck, multi-parent functional clk */ + &gpt1_ick, + &gpt1_fck, + &gpt2_ick, + &gpt2_fck, + &gpt3_ick, + &gpt3_fck, + &gpt4_ick, + &gpt4_fck, + &gpt5_ick, + &gpt5_fck, + &gpt6_ick, + &gpt6_fck, + &gpt7_ick, + &gpt7_fck, + &gpt8_ick, + &gpt8_fck, + &gpt9_ick, + &gpt9_fck, + &gpt10_ick, + &gpt10_fck, + &gpt11_ick, + &gpt11_fck, + &gpt12_ick, + &gpt12_fck, + &mcbsp1_ick, + &mcbsp1_fck, + &mcbsp2_ick, + &mcbsp2_fck, + &mcbsp3_ick, + &mcbsp3_fck, + &mcbsp4_ick, + &mcbsp4_fck, + &mcbsp5_ick, + &mcbsp5_fck, + &mcspi1_ick, + &mcspi1_fck, + &mcspi2_ick, + &mcspi2_fck, + &mcspi3_ick, + &mcspi3_fck, + &uart1_ick, + &uart1_fck, + &uart2_ick, + &uart2_fck, + &uart3_ick, + &uart3_fck, + &gpios_ick, + &gpios_fck, + &mpu_wdt_ick, + &mpu_wdt_fck, + &sync_32k_ick, + &wdt1_ick, + &omapctrl_ick, + &icr_ick, + &cam_fck, + &cam_ick, + &mailboxes_ick, + &wdt4_ick, + &wdt4_fck, + &wdt3_ick, + &wdt3_fck, + &mspro_ick, + &mspro_fck, + &mmc_ick, + &mmc_fck, + &fac_ick, + &fac_fck, + &eac_ick, + &eac_fck, + &hdq_ick, + &hdq_fck, + &i2c1_ick, + &i2c1_fck, + &i2chs1_fck, + &i2c2_ick, + &i2c2_fck, + &i2chs2_fck, + &vlynq_ick, + &vlynq_fck, + &sdrc_ick, + &des_ick, + &sha_ick, + &rng_ick, + &aes_ick, + &pka_ick, + &usb_fck, + &usbhs_ick, + &mmchs1_ick, + &mmchs1_fck, + &mmchs2_ick, + &mmchs2_fck, + &gpio5_ick, + &gpio5_fck, + &mdm_intc_ick, + &mmchsdb1_fck, + &mmchsdb2_fck, +}; + +#endif diff --git a/arch/arm/mach-omap2/devices.c b/arch/arm/mach-omap2/devices.c new file mode 100644 index 00000000000..7181edb8935 --- /dev/null +++ b/arch/arm/mach-omap2/devices.c @@ -0,0 +1,89 @@ +/* + * linux/arch/arm/mach-omap2/devices.c + * + * OMAP2 platform device setup/initialization + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include <linux/config.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/platform_device.h> + +#include <asm/hardware.h> +#include <asm/io.h> +#include <asm/mach-types.h> +#include <asm/mach/map.h> + +#include <asm/arch/tc.h> +#include <asm/arch/board.h> +#include <asm/arch/mux.h> +#include <asm/arch/gpio.h> + +extern void omap_nop_release(struct device *dev); + +/*-------------------------------------------------------------------------*/ + +#if defined(CONFIG_I2C_OMAP) || defined(CONFIG_I2C_OMAP_MODULE) + +#define OMAP2_I2C_BASE2 0x48072000 +#define OMAP2_I2C_INT2 57 + +static struct resource i2c_resources2[] = { + { + .start = OMAP2_I2C_BASE2, + .end = OMAP2_I2C_BASE2 + 0x3f, + .flags = IORESOURCE_MEM, + }, + { + .start = OMAP2_I2C_INT2, + .flags = IORESOURCE_IRQ, + }, +}; + +static struct platform_device omap_i2c_device2 = { + .name = "i2c_omap", + .id = 2, + .dev = { + .release = omap_nop_release, + }, + .num_resources = ARRAY_SIZE(i2c_resources2), + .resource = i2c_resources2, +}; + +/* See also arch/arm/plat-omap/devices.c for first I2C on 24xx */ +static void omap_init_i2c(void) +{ + /* REVISIT: Second I2C not in use on H4? */ + if (machine_is_omap_h4()) + return; + + omap_cfg_reg(J15_24XX_I2C2_SCL); + omap_cfg_reg(H19_24XX_I2C2_SDA); + (void) platform_device_register(&omap_i2c_device2); +} + +#else + +static void omap_init_i2c(void) {} + +#endif + +/*-------------------------------------------------------------------------*/ + +static int __init omap2_init_devices(void) +{ + /* please keep these calls, and their implementations above, + * in alphabetical order so they're easier to sort through. + */ + omap_init_i2c(); + + return 0; +} +arch_initcall(omap2_init_devices); + diff --git a/arch/arm/mach-omap2/id.c b/arch/arm/mach-omap2/id.c new file mode 100644 index 00000000000..76187300f2b --- /dev/null +++ b/arch/arm/mach-omap2/id.c @@ -0,0 +1,124 @@ +/* + * linux/arch/arm/mach-omap2/id.c + * + * OMAP2 CPU identification code + * + * Copyright (C) 2005 Nokia Corporation + * Written by Tony Lindgren <tony@atomide.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include <linux/config.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/init.h> + +#include <asm/io.h> + +#define OMAP24XX_TAP_BASE io_p2v(0x48014000) + +#define OMAP_TAP_IDCODE 0x0204 +#define OMAP_TAP_PROD_ID 0x0208 + +#define OMAP_TAP_DIE_ID_0 0x0218 +#define OMAP_TAP_DIE_ID_1 0x021C +#define OMAP_TAP_DIE_ID_2 0x0220 +#define OMAP_TAP_DIE_ID_3 0x0224 + +/* system_rev fields for OMAP2 processors: + * CPU id bits [31:16], + * CPU device type [15:12], (unprg,normal,POP) + * CPU revision [11:08] + * CPU class bits [07:00] + */ + +struct omap_id { + u16 hawkeye; /* Silicon type (Hawkeye id) */ + u8 dev; /* Device type from production_id reg */ + u32 type; /* combined type id copied to system_rev */ +}; + +/* Register values to detect the OMAP version */ +static struct omap_id omap_ids[] __initdata = { + { .hawkeye = 0xb5d9, .dev = 0x0, .type = 0x24200000 }, + { .hawkeye = 0xb5d9, .dev = 0x1, .type = 0x24201000 }, + { .hawkeye = 0xb5d9, .dev = 0x2, .type = 0x24202000 }, + { .hawkeye = 0xb5d9, .dev = 0x4, .type = 0x24220000 }, + { .hawkeye = 0xb5d9, .dev = 0x8, .type = 0x24230000 }, + { .hawkeye = 0xb68a, .dev = 0x0, .type = 0x24300000 }, +}; + +static u32 __init read_tap_reg(int reg) +{ + return __raw_readl(OMAP24XX_TAP_BASE + reg); +} + +void __init omap2_check_revision(void) +{ + int i, j; + u32 idcode; + u32 prod_id; + u16 hawkeye; + u8 dev_type; + u8 rev; + + idcode = read_tap_reg(OMAP_TAP_IDCODE); + prod_id = read_tap_reg(OMAP_TAP_PROD_ID); + hawkeye = (idcode >> 12) & 0xffff; + rev = (idcode >> 28) & 0x0f; + dev_type = (prod_id >> 16) & 0x0f; + +#ifdef DEBUG + printk(KERN_DEBUG "OMAP_TAP_IDCODE 0x%08x REV %i HAWKEYE 0x%04x MANF %03x\n", + idcode, rev, hawkeye, (idcode >> 1) & 0x7ff); + printk(KERN_DEBUG "OMAP_TAP_DIE_ID_0: 0x%08x\n", + read_tap_reg(OMAP_TAP_DIE_ID_0)); + printk(KERN_DEBUG "OMAP_TAP_DIE_ID_1: 0x%08x DEV_REV: %i\n", + read_tap_reg(OMAP_TAP_DIE_ID_1), + (read_tap_reg(OMAP_TAP_DIE_ID_1) >> 28) & 0xf); + printk(KERN_DEBUG "OMAP_TAP_DIE_ID_2: 0x%08x\n", + read_tap_reg(OMAP_TAP_DIE_ID_2)); + printk(KERN_DEBUG "OMAP_TAP_DIE_ID_3: 0x%08x\n", + read_tap_reg(OMAP_TAP_DIE_ID_3)); + printk(KERN_DEBUG "OMAP_TAP_PROD_ID_0: 0x%08x DEV_TYPE: %i\n", + prod_id, dev_type); +#endif + + /* Check hawkeye ids */ + for (i = 0; i < ARRAY_SIZE(omap_ids); i++) { + if (hawkeye == omap_ids[i].hawkeye) + break; + } + + if (i == ARRAY_SIZE(omap_ids)) { + printk(KERN_ERR "Unknown OMAP CPU id\n"); + return; + } + + for (j = i; j < ARRAY_SIZE(omap_ids); j++) { + if (dev_type == omap_ids[j].dev) + break; + } + + if (j == ARRAY_SIZE(omap_ids)) { + printk(KERN_ERR "Unknown OMAP device type. " + "Handling it as OMAP%04x\n", + omap_ids[i].type >> 16); + j = i; + } + system_rev = omap_ids[j].type; + + system_rev |= rev << 8; + + /* Add the cpu class info (24xx) */ + system_rev |= 0x24; + + pr_info("OMAP%04x", system_rev >> 16); + if ((system_rev >> 8) & 0x0f) + printk("%x", (system_rev >> 8) & 0x0f); + printk("\n"); +} + diff --git a/arch/arm/mach-omap2/io.c b/arch/arm/mach-omap2/io.c new file mode 100644 index 00000000000..8ea67bf196a --- /dev/null +++ b/arch/arm/mach-omap2/io.c @@ -0,0 +1,53 @@ +/* + * linux/arch/arm/mach-omap2/io.c + * + * OMAP2 I/O mapping code + * + * Copyright (C) 2005 Nokia Corporation + * Author: Juha Yrjölä <juha.yrjola@nokia.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include <linux/config.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/init.h> + +#include <asm/mach/map.h> +#include <asm/io.h> +#include <asm/arch/mux.h> + +extern void omap_sram_init(void); +extern int omap2_clk_init(void); +extern void omap2_check_revision(void); + +/* + * The machine specific code may provide the extra mapping besides the + * default mapping provided here. + */ +static struct map_desc omap2_io_desc[] __initdata = { + { + .virtual = L3_24XX_VIRT, + .pfn = __phys_to_pfn(L3_24XX_PHYS), + .length = L3_24XX_SIZE, + .type = MT_DEVICE + }, + { + .virtual = L4_24XX_VIRT, + .pfn = __phys_to_pfn(L4_24XX_PHYS), + .length = L4_24XX_SIZE, + .type = MT_DEVICE + } +}; + +void __init omap_map_common_io(void) +{ + iotable_init(omap2_io_desc, ARRAY_SIZE(omap2_io_desc)); + omap2_check_revision(); + omap_sram_init(); + omap2_mux_init(); + omap2_clk_init(); +} diff --git a/arch/arm/mach-omap2/irq.c b/arch/arm/mach-omap2/irq.c new file mode 100644 index 00000000000..d7baff675cf --- /dev/null +++ b/arch/arm/mach-omap2/irq.c @@ -0,0 +1,149 @@ +/* + * linux/arch/arm/mach-omap/omap2/irq.c + * + * Interrupt handler for OMAP2 boards. + * + * Copyright (C) 2005 Nokia Corporation + * Author: Paul Mundt <paul.mundt@nokia.com> + * + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + */ +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/config.h> +#include <linux/interrupt.h> +#include <asm/hardware.h> +#include <asm/mach/irq.h> +#include <asm/irq.h> +#include <asm/io.h> + +#define INTC_REVISION 0x0000 +#define INTC_SYSCONFIG 0x0010 +#define INTC_SYSSTATUS 0x0014 +#define INTC_CONTROL 0x0048 +#define INTC_MIR_CLEAR0 0x0088 +#define INTC_MIR_SET0 0x008c + +/* + * OMAP2 has a number of different interrupt controllers, each interrupt + * controller is identified as its own "bank". Register definitions are + * fairly consistent for each bank, but not all registers are implemented + * for each bank.. when in doubt, consult the TRM. + */ +static struct omap_irq_bank { + unsigned long base_reg; + unsigned int nr_irqs; +} __attribute__ ((aligned(4))) irq_banks[] = { + { + /* MPU INTC */ + .base_reg = OMAP24XX_IC_BASE, + .nr_irqs = 96, + }, { + /* XXX: DSP INTC */ + +#if 0 + /* + * Commented out for now until we fix the IVA clocking + */ +#ifdef CONFIG_ARCH_OMAP2420 + }, { + /* IVA INTC (2420 only) */ + .base_reg = OMAP24XX_IVA_INTC_BASE, + .nr_irqs = 16, /* Actually 32, but only 16 are used */ +#endif +#endif + } +}; + +/* XXX: FIQ and additional INTC support (only MPU at the moment) */ +static void omap_ack_irq(unsigned int irq) +{ + omap_writel(0x1, irq_banks[0].base_reg + INTC_CONTROL); +} + +static void omap_mask_irq(unsigned int irq) +{ + int offset = (irq >> 5) << 5; + + if (irq >= 64) { + irq %= 64; + } else if (irq >= 32) { + irq %= 32; + } + + omap_writel(1 << irq, irq_banks[0].base_reg + INTC_MIR_SET0 + offset); +} + +static void omap_unmask_irq(unsigned int irq) +{ + int offset = (irq >> 5) << 5; + + if (irq >= 64) { + irq %= 64; + } else if (irq >= 32) { + irq %= 32; + } + + omap_writel(1 << irq, irq_banks[0].base_reg + INTC_MIR_CLEAR0 + offset); +} + +static void omap_mask_ack_irq(unsigned int irq) +{ + omap_mask_irq(irq); + omap_ack_irq(irq); +} + +static struct irqchip omap_irq_chip = { + .ack = omap_mask_ack_irq, + .mask = omap_mask_irq, + .unmask = omap_unmask_irq, +}; + +static void __init omap_irq_bank_init_one(struct omap_irq_bank *bank) +{ + unsigned long tmp; + + tmp = omap_readl(bank->base_reg + INTC_REVISION) & 0xff; + printk(KERN_INFO "IRQ: Found an INTC at 0x%08lx " + "(revision %ld.%ld) with %d interrupts\n", + bank->base_reg, tmp >> 4, tmp & 0xf, bank->nr_irqs); + + tmp = omap_readl(bank->base_reg + INTC_SYSCONFIG); + tmp |= 1 << 1; /* soft reset */ + omap_writel(tmp, bank->base_reg + INTC_SYSCONFIG); + + while (!(omap_readl(bank->base_reg + INTC_SYSSTATUS) & 0x1)) + /* Wait for reset to complete */; +} + +void __init omap_init_irq(void) +{ + unsigned long nr_irqs = 0; + unsigned int nr_banks = 0; + int i; + + for (i = 0; i < ARRAY_SIZE(irq_banks); i++) { + struct omap_irq_bank *bank = irq_banks + i; + + /* XXX */ + if (!bank->base_reg) + continue; + + omap_irq_bank_init_one(bank); + + nr_irqs += bank->nr_irqs; + nr_banks++; + } + + printk(KERN_INFO "Total of %ld interrupts on %d active controller%s\n", + nr_irqs, nr_banks, nr_banks > 1 ? "s" : ""); + + for (i = 0; i < nr_irqs; i++) { + set_irq_chip(i, &omap_irq_chip); + set_irq_handler(i, do_level_IRQ); + set_irq_flags(i, IRQF_VALID); + } +} + diff --git a/arch/arm/mach-omap2/mux.c b/arch/arm/mach-omap2/mux.c new file mode 100644 index 00000000000..ea4654815dd --- /dev/null +++ b/arch/arm/mach-omap2/mux.c @@ -0,0 +1,65 @@ +/* + * linux/arch/arm/mach-omap2/mux.c + * + * OMAP1 pin multiplexing configurations + * + * Copyright (C) 2003 - 2005 Nokia Corporation + * + * Written by Tony Lindgren <tony.lindgren@nokia.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ +#include <linux/config.h> +#include <linux/module.h> +#include <linux/init.h> +#include <asm/system.h> +#include <asm/io.h> +#include <linux/spinlock.h> + +#include <asm/arch/mux.h> + +#ifdef CONFIG_OMAP_MUX + +/* NOTE: See mux.h for the enumeration */ + +struct pin_config __initdata_or_module omap24xx_pins[] = { +/* + * description mux mux pull pull debug + * offset mode ena type + */ + +/* 24xx I2C */ +MUX_CFG_24XX("M19_24XX_I2C1_SCL", 0x111, 0, 0, 0, 1) +MUX_CFG_24XX("L15_24XX_I2C1_SDA", 0x112, 0, 0, 0, 1) +MUX_CFG_24XX("J15_24XX_I2C2_SCL", 0x113, 0, 0, 0, 1) +MUX_CFG_24XX("H19_24XX_I2C2_SDA", 0x114, 0, 0, 0, 1) + +/* Menelaus interrupt */ +MUX_CFG_24XX("W19_24XX_SYS_NIRQ", 0x12c, 0, 1, 1, 1) + +/* 24xx GPIO */ +MUX_CFG_24XX("Y20_24XX_GPIO60", 0x12c, 3, 0, 0, 1) +MUX_CFG_24XX("M15_24XX_GPIO92", 0x10a, 3, 0, 0, 1) + +}; + +int __init omap2_mux_init(void) +{ + omap_mux_register(omap24xx_pins, ARRAY_SIZE(omap24xx_pins)); + return 0; +} + +#endif diff --git a/arch/arm/mach-omap2/prcm.h b/arch/arm/mach-omap2/prcm.h new file mode 100644 index 00000000000..2eb89b936c8 --- /dev/null +++ b/arch/arm/mach-omap2/prcm.h @@ -0,0 +1,419 @@ +/* + * prcm.h - Access definations for use in OMAP24XX clock and power management + * + * Copyright (C) 2005 Texas Instruments, Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#ifndef __ASM_ARM_ARCH_DPM_PRCM_H +#define __ASM_ARM_ARCH_DPM_PRCM_H + +/* SET_PERFORMANCE_LEVEL PARAMETERS */ +#define PRCM_HALF_SPEED 1 +#define PRCM_FULL_SPEED 2 + +#ifndef __ASSEMBLER__ + +#define PRCM_REG32(offset) __REG32(OMAP24XX_PRCM_BASE + (offset)) + +#define PRCM_REVISION PRCM_REG32(0x000) +#define PRCM_SYSCONFIG PRCM_REG32(0x010) +#define PRCM_IRQSTATUS_MPU PRCM_REG32(0x018) +#define PRCM_IRQENABLE_MPU PRCM_REG32(0x01C) +#define PRCM_VOLTCTRL PRCM_REG32(0x050) +#define PRCM_VOLTST PRCM_REG32(0x054) +#define PRCM_CLKSRC_CTRL PRCM_REG32(0x060) +#define PRCM_CLKOUT_CTRL PRCM_REG32(0x070) +#define PRCM_CLKEMUL_CTRL PRCM_REG32(0x078) +#define PRCM_CLKCFG_CTRL PRCM_REG32(0x080) +#define PRCM_CLKCFG_STATUS PRCM_REG32(0x084) +#define PRCM_VOLTSETUP PRCM_REG32(0x090) +#define PRCM_CLKSSETUP PRCM_REG32(0x094) +#define PRCM_POLCTRL PRCM_REG32(0x098) + +/* GENERAL PURPOSE */ +#define GENERAL_PURPOSE1 PRCM_REG32(0x0B0) +#define GENERAL_PURPOSE2 PRCM_REG32(0x0B4) +#define GENERAL_PURPOSE3 PRCM_REG32(0x0B8) +#define GENERAL_PURPOSE4 PRCM_REG32(0x0BC) +#define GENERAL_PURPOSE5 PRCM_REG32(0x0C0) +#define GENERAL_PURPOSE6 PRCM_REG32(0x0C4) +#define GENERAL_PURPOSE7 PRCM_REG32(0x0C8) +#define GENERAL_PURPOSE8 PRCM_REG32(0x0CC) +#define GENERAL_PURPOSE9 PRCM_REG32(0x0D0) +#define GENERAL_PURPOSE10 PRCM_REG32(0x0D4) +#define GENERAL_PURPOSE11 PRCM_REG32(0x0D8) +#define GENERAL_PURPOSE12 PRCM_REG32(0x0DC) +#define GENERAL_PURPOSE13 PRCM_REG32(0x0E0) +#define GENERAL_PURPOSE14 PRCM_REG32(0x0E4) +#define GENERAL_PURPOSE15 PRCM_REG32(0x0E8) +#define GENERAL_PURPOSE16 PRCM_REG32(0x0EC) +#define GENERAL_PURPOSE17 PRCM_REG32(0x0F0) +#define GENERAL_PURPOSE18 PRCM_REG32(0x0F4) +#define GENERAL_PURPOSE19 PRCM_REG32(0x0F8) +#define GENERAL_PURPOSE20 PRCM_REG32(0x0FC) + +/* MPU */ +#define CM_CLKSEL_MPU PRCM_REG32(0x140) +#define CM_CLKSTCTRL_MPU PRCM_REG32(0x148) +#define RM_RSTST_MPU PRCM_REG32(0x158) +#define PM_WKDEP_MPU PRCM_REG32(0x1C8) +#define PM_EVGENCTRL_MPU PRCM_REG32(0x1D4) +#define PM_EVEGENONTIM_MPU PRCM_REG32(0x1D8) +#define PM_EVEGENOFFTIM_MPU PRCM_REG32(0x1DC) +#define PM_PWSTCTRL_MPU PRCM_REG32(0x1E0) +#define PM_PWSTST_MPU PRCM_REG32(0x1E4) + +/* CORE */ +#define CM_FCLKEN1_CORE PRCM_REG32(0x200) +#define CM_FCLKEN2_CORE PRCM_REG32(0x204) +#define CM_FCLKEN3_CORE PRCM_REG32(0x208) +#define CM_ICLKEN1_CORE PRCM_REG32(0x210) +#define CM_ICLKEN2_CORE PRCM_REG32(0x214) +#define CM_ICLKEN3_CORE PRCM_REG32(0x218) +#define CM_ICLKEN4_CORE PRCM_REG32(0x21C) +#define CM_IDLEST1_CORE PRCM_REG32(0x220) +#define CM_IDLEST2_CORE PRCM_REG32(0x224) +#define CM_IDLEST3_CORE PRCM_REG32(0x228) +#define CM_IDLEST4_CORE PRCM_REG32(0x22C) +#define CM_AUTOIDLE1_CORE PRCM_REG32(0x230) +#define CM_AUTOIDLE2_CORE PRCM_REG32(0x234) +#define CM_AUTOIDLE3_CORE PRCM_REG32(0x238) +#define CM_AUTOIDLE4_CORE PRCM_REG32(0x23C) +#define CM_CLKSEL1_CORE PRCM_REG32(0x240) +#define CM_CLKSEL2_CORE PRCM_REG32(0x244) +#define CM_CLKSTCTRL_CORE PRCM_REG32(0x248) +#define PM_WKEN1_CORE PRCM_REG32(0x2A0) +#define PM_WKEN2_CORE PRCM_REG32(0x2A4) +#define PM_WKST1_CORE PRCM_REG32(0x2B0) +#define PM_WKST2_CORE PRCM_REG32(0x2B4) +#define PM_WKDEP_CORE PRCM_REG32(0x2C8) +#define PM_PWSTCTRL_CORE PRCM_REG32(0x2E0) +#define PM_PWSTST_CORE PRCM_REG32(0x2E4) + +/* GFX */ +#define CM_FCLKEN_GFX PRCM_REG32(0x300) +#define CM_ICLKEN_GFX PRCM_REG32(0x310) +#define CM_IDLEST_GFX PRCM_REG32(0x320) +#define CM_CLKSEL_GFX PRCM_REG32(0x340) +#define CM_CLKSTCTRL_GFX PRCM_REG32(0x348) +#define RM_RSTCTRL_GFX PRCM_REG32(0x350) +#define RM_RSTST_GFX PRCM_REG32(0x358) +#define PM_WKDEP_GFX PRCM_REG32(0x3C8) +#define PM_PWSTCTRL_GFX PRCM_REG32(0x3E0) +#define PM_PWSTST_GFX PRCM_REG32(0x3E4) + +/* WAKE-UP */ +#define CM_FCLKEN_WKUP PRCM_REG32(0x400) +#define CM_ICLKEN_WKUP PRCM_REG32(0x410) +#define CM_IDLEST_WKUP PRCM_REG32(0x420) +#define CM_AUTOIDLE_WKUP PRCM_REG32(0x430) +#define CM_CLKSEL_WKUP PRCM_REG32(0x440) +#define RM_RSTCTRL_WKUP PRCM_REG32(0x450) +#define RM_RSTTIME_WKUP PRCM_REG32(0x454) +#define RM_RSTST_WKUP PRCM_REG32(0x458) +#define PM_WKEN_WKUP PRCM_REG32(0x4A0) +#define PM_WKST_WKUP PRCM_REG32(0x4B0) + +/* CLOCKS */ +#define CM_CLKEN_PLL PRCM_REG32(0x500) +#define CM_IDLEST_CKGEN PRCM_REG32(0x520) +#define CM_AUTOIDLE_PLL PRCM_REG32(0x530) +#define CM_CLKSEL1_PLL PRCM_REG32(0x540) +#define CM_CLKSEL2_PLL PRCM_REG32(0x544) + +/* DSP */ +#define CM_FCLKEN_DSP PRCM_REG32(0x800) +#define CM_ICLKEN_DSP PRCM_REG32(0x810) +#define CM_IDLEST_DSP PRCM_REG32(0x820) +#define CM_AUTOIDLE_DSP PRCM_REG32(0x830) +#define CM_CLKSEL_DSP PRCM_REG32(0x840) +#define CM_CLKSTCTRL_DSP PRCM_REG32(0x848) +#define RM_RSTCTRL_DSP PRCM_REG32(0x850) +#define RM_RSTST_DSP PRCM_REG32(0x858) +#define PM_WKEN_DSP PRCM_REG32(0x8A0) +#define PM_WKDEP_DSP PRCM_REG32(0x8C8) +#define PM_PWSTCTRL_DSP PRCM_REG32(0x8E0) +#define PM_PWSTST_DSP PRCM_REG32(0x8E4) +#define PRCM_IRQSTATUS_DSP PRCM_REG32(0x8F0) +#define PRCM_IRQENABLE_DSP PRCM_REG32(0x8F4) + +/* IVA */ +#define PRCM_IRQSTATUS_IVA PRCM_REG32(0x8F8) +#define PRCM_IRQENABLE_IVA PRCM_REG32(0x8FC) + +/* Modem on 2430 */ +#define CM_FCLKEN_MDM PRCM_REG32(0xC00) +#define CM_ICLKEN_MDM PRCM_REG32(0xC10) +#define CM_IDLEST_MDM PRCM_REG32(0xC20) +#define CM_CLKSEL_MDM PRCM_REG32(0xC40) + +/* FIXME: Move to header for 2430 */ +#define DISP_BASE (OMAP24XX_L4_IO_BASE+0x50000) +#define DISP_REG32(offset) __REG32(DISP_BASE + (offset)) + +#define GPMC_BASE (OMAP24XX_GPMC_BASE) +#define GPMC_REG32(offset) __REG32(GPMC_BASE + (offset)) + +#define GPT1_BASE (OMAP24XX_GPT1) +#define GPT1_REG32(offset) __REG32(GPT1_BASE + (offset)) + +/* Misc sysconfig */ +#define DISPC_SYSCONFIG DISP_REG32(0x410) +#define SPI_BASE (OMAP24XX_L4_IO_BASE+0x98000) +#define MCSPI1_SYSCONFIG __REG32(SPI_BASE + 0x10) +#define MCSPI2_SYSCONFIG __REG32(SPI_BASE+0x2000 + 0x10) + +//#define DSP_MMU_SYSCONFIG 0x5A000010 +#define CAMERA_MMU_SYSCONFIG __REG32(DISP_BASE+0x2C10) +//#define IVA_MMU_SYSCONFIG 0x5D000010 +//#define DSP_DMA_SYSCONFIG 0x00FCC02C +#define CAMERA_DMA_SYSCONFIG __REG32(DISP_BASE+0x282C) +#define SYSTEM_DMA_SYSCONFIG __REG32(DISP_BASE+0x602C) +#define GPMC_SYSCONFIG GPMC_REG32(0x010) +#define MAILBOXES_SYSCONFIG __REG32(OMAP24XX_L4_IO_BASE+0x94010) +#define UART1_SYSCONFIG __REG32(OMAP24XX_L4_IO_BASE+0x6A054) +#define UART2_SYSCONFIG __REG32(OMAP24XX_L4_IO_BASE+0x6C054) +#define UART3_SYSCONFIG __REG32(OMAP24XX_L4_IO_BASE+0x6E054) +//#define IVA_SYSCONFIG 0x5C060010 +#define SDRC_SYSCONFIG __REG32(OMAP24XX_SDRC_BASE+0x10) +#define SMS_SYSCONFIG __REG32(OMAP24XX_SMS_BASE+0x10) +#define SSI_SYSCONFIG __REG32(DISP_BASE+0x8010) +//#define VLYNQ_SYSCONFIG 0x67FFFE10 + +/* rkw - good cannidates for PM_ to start what nm was trying */ +#define OMAP24XX_GPT2 (OMAP24XX_L4_IO_BASE+0x2A000) +#define OMAP24XX_GPT3 (OMAP24XX_L4_IO_BASE+0x78000) +#define OMAP24XX_GPT4 (OMAP24XX_L4_IO_BASE+0x7A000) +#define OMAP24XX_GPT5 (OMAP24XX_L4_IO_BASE+0x7C000) +#define OMAP24XX_GPT6 (OMAP24XX_L4_IO_BASE+0x7E000) +#define OMAP24XX_GPT7 (OMAP24XX_L4_IO_BASE+0x80000) +#define OMAP24XX_GPT8 (OMAP24XX_L4_IO_BASE+0x82000) +#define OMAP24XX_GPT9 (OMAP24XX_L4_IO_BASE+0x84000) +#define OMAP24XX_GPT10 (OMAP24XX_L4_IO_BASE+0x86000) +#define OMAP24XX_GPT11 (OMAP24XX_L4_IO_BASE+0x88000) +#define OMAP24XX_GPT12 (OMAP24XX_L4_IO_BASE+0x8A000) + +#define GPTIMER1_SYSCONFIG GPT1_REG32(0x010) +#define GPTIMER2_SYSCONFIG __REG32(OMAP24XX_GPT2 + 0x10) +#define GPTIMER3_SYSCONFIG __REG32(OMAP24XX_GPT3 + 0x10) +#define GPTIMER4_SYSCONFIG __REG32(OMAP24XX_GPT4 + 0x10) +#define GPTIMER5_SYSCONFIG __REG32(OMAP24XX_GPT5 + 0x10) +#define GPTIMER6_SYSCONFIG __REG32(OMAP24XX_GPT6 + 0x10) +#define GPTIMER7_SYSCONFIG __REG32(OMAP24XX_GPT7 + 0x10) +#define GPTIMER8_SYSCONFIG __REG32(OMAP24XX_GPT8 + 0x10) +#define GPTIMER9_SYSCONFIG __REG32(OMAP24XX_GPT9 + 0x10) +#define GPTIMER10_SYSCONFIG __REG32(OMAP24XX_GPT10 + 0x10) +#define GPTIMER11_SYSCONFIG __REG32(OMAP24XX_GPT11 + 0x10) +#define GPTIMER12_SYSCONFIG __REG32(OMAP24XX_GPT12 + 0x10) + +#define GPIOX_BASE(X) (OMAP24XX_GPIO_BASE+(0x2000*((X)-1))) + +#define GPIO1_SYSCONFIG __REG32((GPIOX_BASE(1)+0x10)) +#define GPIO2_SYSCONFIG __REG32((GPIOX_BASE(2)+0x10)) +#define GPIO3_SYSCONFIG __REG32((GPIOX_BASE(3)+0x10)) +#define GPIO4_SYSCONFIG __REG32((GPIOX_BASE(4)+0x10)) + +/* GP TIMER 1 */ +#define GPTIMER1_TISTAT GPT1_REG32(0x014) +#define GPTIMER1_TISR GPT1_REG32(0x018) +#define GPTIMER1_TIER GPT1_REG32(0x01C) +#define GPTIMER1_TWER GPT1_REG32(0x020) +#define GPTIMER1_TCLR GPT1_REG32(0x024) +#define GPTIMER1_TCRR GPT1_REG32(0x028) +#define GPTIMER1_TLDR GPT1_REG32(0x02C) +#define GPTIMER1_TTGR GPT1_REG32(0x030) +#define GPTIMER1_TWPS GPT1_REG32(0x034) +#define GPTIMER1_TMAR GPT1_REG32(0x038) +#define GPTIMER1_TCAR1 GPT1_REG32(0x03C) +#define GPTIMER1_TSICR GPT1_REG32(0x040) +#define GPTIMER1_TCAR2 GPT1_REG32(0x044) + +/* rkw -- base fix up please... */ +#define GPTIMER3_TISR __REG32(OMAP24XX_L4_IO_BASE+0x78018) + +/* SDRC */ +#define SDRC_DLLA_CTRL __REG32(OMAP24XX_SDRC_BASE+0x060) +#define SDRC_DLLA_STATUS __REG32(OMAP24XX_SDRC_BASE+0x064) +#define SDRC_DLLB_CTRL __REG32(OMAP24XX_SDRC_BASE+0x068) +#define SDRC_DLLB_STATUS __REG32(OMAP24XX_SDRC_BASE+0x06C) +#define SDRC_POWER __REG32(OMAP24XX_SDRC_BASE+0x070) +#define SDRC_MR_0 __REG32(OMAP24XX_SDRC_BASE+0x084) + +/* GPIO 1 */ +#define GPIO1_BASE GPIOX_BASE(1) +#define GPIO1_REG32(offset) __REG32(GPIO1_BASE + (offset)) +#define GPIO1_IRQENABLE1 GPIO1_REG32(0x01C) +#define GPIO1_IRQSTATUS1 GPIO1_REG32(0x018) +#define GPIO1_IRQENABLE2 GPIO1_REG32(0x02C) +#define GPIO1_IRQSTATUS2 GPIO1_REG32(0x028) +#define GPIO1_WAKEUPENABLE GPIO1_REG32(0x020) +#define GPIO1_RISINGDETECT GPIO1_REG32(0x048) +#define GPIO1_DATAIN GPIO1_REG32(0x038) +#define GPIO1_OE GPIO1_REG32(0x034) +#define GPIO1_DATAOUT GPIO1_REG32(0x03C) + +/* GPIO2 */ +#define GPIO2_BASE GPIOX_BASE(2) +#define GPIO2_REG32(offset) __REG32(GPIO2_BASE + (offset)) +#define GPIO2_IRQENABLE1 GPIO2_REG32(0x01C) +#define GPIO2_IRQSTATUS1 GPIO2_REG32(0x018) +#define GPIO2_IRQENABLE2 GPIO2_REG32(0x02C) +#define GPIO2_IRQSTATUS2 GPIO2_REG32(0x028) +#define GPIO2_WAKEUPENABLE GPIO2_REG32(0x020) +#define GPIO2_RISINGDETECT GPIO2_REG32(0x048) +#define GPIO2_DATAIN GPIO2_REG32(0x038) +#define GPIO2_OE GPIO2_REG32(0x034) +#define GPIO2_DATAOUT GPIO2_REG32(0x03C) + +/* GPIO 3 */ +#define GPIO3_BASE GPIOX_BASE(3) +#define GPIO3_REG32(offset) __REG32(GPIO3_BASE + (offset)) +#define GPIO3_IRQENABLE1 GPIO3_REG32(0x01C) +#define GPIO3_IRQSTATUS1 GPIO3_REG32(0x018) +#define GPIO3_IRQENABLE2 GPIO3_REG32(0x02C) +#define GPIO3_IRQSTATUS2 GPIO3_REG32(0x028) +#define GPIO3_WAKEUPENABLE GPIO3_REG32(0x020) +#define GPIO3_RISINGDETECT GPIO3_REG32(0x048) +#define GPIO3_FALLINGDETECT GPIO3_REG32(0x04C) +#define GPIO3_DATAIN GPIO3_REG32(0x038) +#define GPIO3_OE GPIO3_REG32(0x034) +#define GPIO3_DATAOUT GPIO3_REG32(0x03C) +#define GPIO3_DEBOUNCENABLE GPIO3_REG32(0x050) +#define GPIO3_DEBOUNCINGTIME GPIO3_REG32(0x054) + +/* GPIO 4 */ +#define GPIO4_BASE GPIOX_BASE(4) +#define GPIO4_REG32(offset) __REG32(GPIO4_BASE + (offset)) +#define GPIO4_IRQENABLE1 GPIO4_REG32(0x01C) +#define GPIO4_IRQSTATUS1 GPIO4_REG32(0x018) +#define GPIO4_IRQENABLE2 GPIO4_REG32(0x02C) +#define GPIO4_IRQSTATUS2 GPIO4_REG32(0x028) +#define GPIO4_WAKEUPENABLE GPIO4_REG32(0x020) +#define GPIO4_RISINGDETECT GPIO4_REG32(0x048) +#define GPIO4_FALLINGDETECT GPIO4_REG32(0x04C) +#define GPIO4_DATAIN GPIO4_REG32(0x038) +#define GPIO4_OE GPIO4_REG32(0x034) +#define GPIO4_DATAOUT GPIO4_REG32(0x03C) +#define GPIO4_DEBOUNCENABLE GPIO4_REG32(0x050) +#define GPIO4_DEBOUNCINGTIME GPIO4_REG32(0x054) + + +/* IO CONFIG */ +#define CONTROL_BASE (OMAP24XX_CTRL_BASE) +#define CONTROL_REG32(offset) __REG32(CONTROL_BASE + (offset)) + +#define CONTROL_PADCONF_SPI1_NCS2 CONTROL_REG32(0x104) +#define CONTROL_PADCONF_SYS_XTALOUT CONTROL_REG32(0x134) +#define CONTROL_PADCONF_UART1_RX CONTROL_REG32(0x0C8) +#define CONTROL_PADCONF_MCBSP1_DX CONTROL_REG32(0x10C) +#define CONTROL_PADCONF_GPMC_NCS4 CONTROL_REG32(0x090) +#define CONTROL_PADCONF_DSS_D5 CONTROL_REG32(0x0B8) +#define CONTROL_PADCONF_DSS_D9 CONTROL_REG32(0x0BC) +#define CONTROL_PADCONF_DSS_D13 CONTROL_REG32(0x0C0) +#define CONTROL_PADCONF_DSS_VSYNC CONTROL_REG32(0x0CC) + +/* CONTROL */ +#define CONTROL_DEVCONF CONTROL_REG32(0x274) + +/* INTERRUPT CONTROLLER */ +#define INTC_BASE (OMAP24XX_L4_IO_BASE+0xfe000) +#define INTC_REG32(offset) __REG32(INTC_BASE + (offset)) + +#define INTC1_U_BASE INTC_REG32(0x000) +#define INTC_MIR0 INTC_REG32(0x084) +#define INTC_MIR_SET0 INTC_REG32(0x08C) +#define INTC_MIR_CLEAR0 INTC_REG32(0x088) +#define INTC_ISR_CLEAR0 INTC_REG32(0x094) +#define INTC_MIR1 INTC_REG32(0x0A4) +#define INTC_MIR_SET1 INTC_REG32(0x0AC) +#define INTC_MIR_CLEAR1 INTC_REG32(0x0A8) +#define INTC_ISR_CLEAR1 INTC_REG32(0x0B4) +#define INTC_MIR2 INTC_REG32(0x0C4) +#define INTC_MIR_SET2 INTC_REG32(0x0CC) +#define INTC_MIR_CLEAR2 INTC_REG32(0x0C8) +#define INTC_ISR_CLEAR2 INTC_REG32(0x0D4) +#define INTC_SIR_IRQ INTC_REG32(0x040) +#define INTC_CONTROL INTC_REG32(0x048) +#define INTC_ILR11 INTC_REG32(0x12C) +#define INTC_ILR32 INTC_REG32(0x180) +#define INTC_ILR37 INTC_REG32(0x194) +#define INTC_SYSCONFIG INTC_REG32(0x010) + +/* RAM FIREWALL */ +#define RAMFW_BASE (0x68005000) +#define RAMFW_REG32(offset) __REG32(RAMFW_BASE + (offset)) + +#define RAMFW_REQINFOPERM0 RAMFW_REG32(0x048) +#define RAMFW_READPERM0 RAMFW_REG32(0x050) +#define RAMFW_WRITEPERM0 RAMFW_REG32(0x058) + +/* GPMC CS1 FPGA ON USER INTERFACE MODULE */ +//#define DEBUG_BOARD_LED_REGISTER 0x04000014 + +/* GPMC CS0 */ +#define GPMC_CONFIG1_0 GPMC_REG32(0x060) +#define GPMC_CONFIG2_0 GPMC_REG32(0x064) +#define GPMC_CONFIG3_0 GPMC_REG32(0x068) +#define GPMC_CONFIG4_0 GPMC_REG32(0x06C) +#define GPMC_CONFIG5_0 GPMC_REG32(0x070) +#define GPMC_CONFIG6_0 GPMC_REG32(0x074) +#define GPMC_CONFIG7_0 GPMC_REG32(0x078) + +/* DSS */ +#define DSS_CONTROL DISP_REG32(0x040) +#define DISPC_CONTROL DISP_REG32(0x440) +#define DISPC_SYSSTATUS DISP_REG32(0x414) +#define DISPC_IRQSTATUS DISP_REG32(0x418) +#define DISPC_IRQENABLE DISP_REG32(0x41C) +#define DISPC_CONFIG DISP_REG32(0x444) +#define DISPC_DEFAULT_COLOR0 DISP_REG32(0x44C) +#define DISPC_DEFAULT_COLOR1 DISP_REG32(0x450) +#define DISPC_TRANS_COLOR0 DISP_REG32(0x454) +#define DISPC_TRANS_COLOR1 DISP_REG32(0x458) +#define DISPC_LINE_NUMBER DISP_REG32(0x460) +#define DISPC_TIMING_H DISP_REG32(0x464) +#define DISPC_TIMING_V DISP_REG32(0x468) +#define DISPC_POL_FREQ DISP_REG32(0x46C) +#define DISPC_DIVISOR DISP_REG32(0x470) +#define DISPC_SIZE_DIG DISP_REG32(0x478) +#define DISPC_SIZE_LCD DISP_REG32(0x47C) +#define DISPC_GFX_BA0 DISP_REG32(0x480) +#define DISPC_GFX_BA1 DISP_REG32(0x484) +#define DISPC_GFX_POSITION DISP_REG32(0x488) +#define DISPC_GFX_SIZE DISP_REG32(0x48C) +#define DISPC_GFX_ATTRIBUTES DISP_REG32(0x4A0) +#define DISPC_GFX_FIFO_THRESHOLD DISP_REG32(0x4A4) +#define DISPC_GFX_ROW_INC DISP_REG32(0x4AC) +#define DISPC_GFX_PIXEL_INC DISP_REG32(0x4B0) +#define DISPC_GFX_WINDOW_SKIP DISP_REG32(0x4B4) +#define DISPC_GFX_TABLE_BA DISP_REG32(0x4B8) +#define DISPC_DATA_CYCLE1 DISP_REG32(0x5D4) +#define DISPC_DATA_CYCLE2 DISP_REG32(0x5D8) +#define DISPC_DATA_CYCLE3 DISP_REG32(0x5DC) + +/* Wake up define for board */ +#define GPIO97 (1 << 1) +#define GPIO88 (1 << 24) + +#endif /* __ASSEMBLER__ */ + +#endif + + + + + diff --git a/arch/arm/mach-omap2/serial.c b/arch/arm/mach-omap2/serial.c new file mode 100644 index 00000000000..f4df04fe1dd --- /dev/null +++ b/arch/arm/mach-omap2/serial.c @@ -0,0 +1,180 @@ +/* + * arch/arm/mach-omap/omap2/serial.c + * + * OMAP2 serial support. + * + * Copyright (C) 2005 Nokia Corporation + * Author: Paul Mundt <paul.mundt@nokia.com> + * + * Based off of arch/arm/mach-omap/omap1/serial.c + * + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + */ +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/serial_8250.h> +#include <linux/serial_reg.h> + +#include <asm/io.h> +#include <asm/hardware/clock.h> + +#include <asm/arch/common.h> +#include <asm/arch/board.h> + +static struct clk * uart1_ick = NULL; +static struct clk * uart1_fck = NULL; +static struct clk * uart2_ick = NULL; +static struct clk * uart2_fck = NULL; +static struct clk * uart3_ick = NULL; +static struct clk * uart3_fck = NULL; + +static struct plat_serial8250_port serial_platform_data[] = { + { + .membase = (char *)IO_ADDRESS(OMAP_UART1_BASE), + .mapbase = (unsigned long)OMAP_UART1_BASE, + .irq = 72, + .flags = UPF_BOOT_AUTOCONF, + .iotype = UPIO_MEM, + .regshift = 2, + .uartclk = OMAP16XX_BASE_BAUD * 16, + }, { + .membase = (char *)IO_ADDRESS(OMAP_UART2_BASE), + .mapbase = (unsigned long)OMAP_UART2_BASE, + .irq = 73, + .flags = UPF_BOOT_AUTOCONF, + .iotype = UPIO_MEM, + .regshift = 2, + .uartclk = OMAP16XX_BASE_BAUD * 16, + }, { + .membase = (char *)IO_ADDRESS(OMAP_UART3_BASE), + .mapbase = (unsigned long)OMAP_UART3_BASE, + .irq = 74, + .flags = UPF_BOOT_AUTOCONF, + .iotype = UPIO_MEM, + .regshift = 2, + .uartclk = OMAP16XX_BASE_BAUD * 16, + }, { + .flags = 0 + } +}; + +static inline unsigned int serial_read_reg(struct plat_serial8250_port *up, + int offset) +{ + offset <<= up->regshift; + return (unsigned int)__raw_readb(up->membase + offset); +} + +static inline void serial_write_reg(struct plat_serial8250_port *p, int offset, + int value) +{ + offset <<= p->regshift; + __raw_writeb(value, (unsigned long)(p->membase + offset)); +} + +/* + * Internal UARTs need to be initialized for the 8250 autoconfig to work + * properly. Note that the TX watermark initialization may not be needed + * once the 8250.c watermark handling code is merged. + */ +static inline void __init omap_serial_reset(struct plat_serial8250_port *p) +{ + serial_write_reg(p, UART_OMAP_MDR1, 0x07); + serial_write_reg(p, UART_OMAP_SCR, 0x08); + serial_write_reg(p, UART_OMAP_MDR1, 0x00); + serial_write_reg(p, UART_OMAP_SYSC, 0x01); +} + +void __init omap_serial_init() +{ + int i; + const struct omap_uart_config *info; + + /* + * Make sure the serial ports are muxed on at this point. + * You have to mux them off in device drivers later on + * if not needed. + */ + + info = omap_get_config(OMAP_TAG_UART, + struct omap_uart_config); + + if (info == NULL) + return; + + for (i = 0; i < OMAP_MAX_NR_PORTS; i++) { + struct plat_serial8250_port *p = serial_platform_data + i; + + if (!(info->enabled_uarts & (1 << i))) { + p->membase = 0; + p->mapbase = 0; + continue; + } + + switch (i) { + case 0: + uart1_ick = clk_get(NULL, "uart1_ick"); + if (IS_ERR(uart1_ick)) + printk("Could not get uart1_ick\n"); + else { + clk_use(uart1_ick); + } + + uart1_fck = clk_get(NULL, "uart1_fck"); + if (IS_ERR(uart1_fck)) + printk("Could not get uart1_fck\n"); + else { + clk_use(uart1_fck); + } + break; + case 1: + uart2_ick = clk_get(NULL, "uart2_ick"); + if (IS_ERR(uart2_ick)) + printk("Could not get uart2_ick\n"); + else { + clk_use(uart2_ick); + } + + uart2_fck = clk_get(NULL, "uart2_fck"); + if (IS_ERR(uart2_fck)) + printk("Could not get uart2_fck\n"); + else { + clk_use(uart2_fck); + } + break; + case 2: + uart3_ick = clk_get(NULL, "uart3_ick"); + if (IS_ERR(uart3_ick)) + printk("Could not get uart3_ick\n"); + else { + clk_use(uart3_ick); + } + + uart3_fck = clk_get(NULL, "uart3_fck"); + if (IS_ERR(uart3_fck)) + printk("Could not get uart3_fck\n"); + else { + clk_use(uart3_fck); + } + break; + } + + omap_serial_reset(p); + } +} + +static struct platform_device serial_device = { + .name = "serial8250", + .id = 0, + .dev = { + .platform_data = serial_platform_data, + }, +}; + +static int __init omap_init(void) +{ + return platform_device_register(&serial_device); +} +arch_initcall(omap_init); diff --git a/arch/arm/mach-omap2/sram-fn.S b/arch/arm/mach-omap2/sram-fn.S new file mode 100644 index 00000000000..2a869e20334 --- /dev/null +++ b/arch/arm/mach-omap2/sram-fn.S @@ -0,0 +1,333 @@ +/* + * linux/arch/arm/mach-omap1/sram.S + * + * Omap2 specific functions that need to be run in internal SRAM + * + * (C) Copyright 2004 + * Texas Instruments, <www.ti.com> + * Richard Woodruff <r-woodruff2@ti.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation; either version 2 of + * the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR /PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, + * MA 02111-1307 USA + */ +#include <linux/config.h> +#include <linux/linkage.h> +#include <asm/assembler.h> +#include <asm/arch/io.h> +#include <asm/hardware.h> + +#include <asm/arch/prcm.h> + +#define TIMER_32KSYNCT_CR_V IO_ADDRESS(OMAP24XX_32KSYNCT_BASE + 0x010) + +#define CM_CLKSEL2_PLL_V IO_ADDRESS(OMAP24XX_PRCM_BASE + 0x544) +#define PRCM_VOLTCTRL_V IO_ADDRESS(OMAP24XX_PRCM_BASE + 0x050) +#define PRCM_CLKCFG_CTRL_V IO_ADDRESS(OMAP24XX_PRCM_BASE + 0x080) +#define CM_CLKEN_PLL_V IO_ADDRESS(OMAP24XX_PRCM_BASE + 0x500) +#define CM_IDLEST_CKGEN_V IO_ADDRESS(OMAP24XX_PRCM_BASE + 0x520) +#define CM_CLKSEL1_PLL_V IO_ADDRESS(OMAP24XX_PRCM_BASE + 0x540) + +#define SDRC_DLLA_CTRL_V IO_ADDRESS(OMAP24XX_SDRC_BASE + 0x060) +#define SDRC_RFR_CTRL_V IO_ADDRESS(OMAP24XX_SDRC_BASE + 0x0a4) + + .text + +ENTRY(sram_ddr_init) + stmfd sp!, {r0 - r12, lr} @ save registers on stack + + mov r12, r2 @ capture CS1 vs CS0 + mov r8, r3 @ capture force parameter + + /* frequency shift down */ + ldr r2, cm_clksel2_pll @ get address of dpllout reg + mov r3, #0x1 @ value for 1x operation + str r3, [r2] @ go to L1-freq operation + + /* voltage shift down */ + mov r9, #0x1 @ set up for L1 voltage call + bl voltage_shift @ go drop voltage + + /* dll lock mode */ + ldr r11, sdrc_dlla_ctrl @ addr of dlla ctrl + ldr r10, [r11] @ get current val + cmp r12, #0x1 @ cs1 base (2422 es2.05/1) + addeq r11, r11, #0x8 @ if cs1 base, move to DLLB + mvn r9, #0x4 @ mask to get clear bit2 + and r10, r10, r9 @ clear bit2 for lock mode. + orr r10, r10, #0x8 @ make sure DLL on (es2 bit pos) + orr r10, r10, #0x2 @ 90 degree phase for all below 133Mhz + str r10, [r11] @ commit to DLLA_CTRL + bl i_dll_wait @ wait for dll to lock + + /* get dll value */ + add r11, r11, #0x4 @ get addr of status reg + ldr r10, [r11] @ get locked value + + /* voltage shift up */ + mov r9, #0x0 @ shift back to L0-voltage + bl voltage_shift @ go raise voltage + + /* frequency shift up */ + mov r3, #0x2 @ value for 2x operation + str r3, [r2] @ go to L0-freq operation + + /* reset entry mode for dllctrl */ + sub r11, r11, #0x4 @ move from status to ctrl + cmp r12, #0x1 @ normalize if cs1 based + subeq r11, r11, #0x8 @ possibly back to DLLA + cmp r8, #0x1 @ if forced unlock exit + orreq r1, r1, #0x4 @ make sure exit with unlocked value + str r1, [r11] @ restore DLLA_CTRL high value + add r11, r11, #0x8 @ move to DLLB_CTRL addr + str r1, [r11] @ set value DLLB_CTRL + bl i_dll_wait @ wait for possible lock + + /* set up for return, DDR should be good */ + str r10, [r0] @ write dll_status and return counter + ldmfd sp!, {r0 - r12, pc} @ restore regs and return + + /* ensure the DLL has relocked */ +i_dll_wait: + mov r4, #0x800 @ delay DLL relock, min 0x400 L3 clocks +i_dll_delay: + subs r4, r4, #0x1 + bne i_dll_delay + mov pc, lr + + /* + * shift up or down voltage, use R9 as input to tell level. + * wait for it to finish, use 32k sync counter, 1tick=31uS. + */ +voltage_shift: + ldr r4, prcm_voltctrl @ get addr of volt ctrl. + ldr r5, [r4] @ get value. + ldr r6, prcm_mask_val @ get value of mask + and r5, r5, r6 @ apply mask to clear bits + orr r5, r5, r9 @ bulld value for L0/L1-volt operation. + str r5, [r4] @ set up for change. + mov r3, #0x4000 @ get val for force + orr r5, r5, r3 @ build value for force + str r5, [r4] @ Force transition to L1 + + ldr r3, timer_32ksynct_cr @ get addr of counter + ldr r5, [r3] @ get value + add r5, r5, #0x3 @ give it at most 93uS +volt_delay: + ldr r7, [r3] @ get timer value + cmp r5, r7 @ time up? + bhi volt_delay @ not yet->branch + mov pc, lr @ back to caller. + +/* relative load constants */ +cm_clksel2_pll: + .word CM_CLKSEL2_PLL_V +sdrc_dlla_ctrl: + .word SDRC_DLLA_CTRL_V +prcm_voltctrl: + .word PRCM_VOLTCTRL_V +prcm_mask_val: + .word 0xFFFF3FFC +timer_32ksynct_cr: + .word TIMER_32KSYNCT_CR_V +ENTRY(sram_ddr_init_sz) + .word . - sram_ddr_init + +/* + * Reprograms memory timings. + * r0 = [PRCM_FULL | PRCM_HALF] r1 = SDRC_DLLA_CTRL value r2 = [DDR | SDR] + * PRCM_FULL = 2, PRCM_HALF = 1, DDR = 1, SDR = 0 + */ +ENTRY(sram_reprogram_sdrc) + stmfd sp!, {r0 - r10, lr} @ save registers on stack + mov r3, #0x0 @ clear for mrc call + mcr p15, 0, r3, c7, c10, 4 @ memory barrier, finish ARM SDR/DDR + nop + nop + ldr r6, ddr_sdrc_rfr_ctrl @ get addr of refresh reg + ldr r5, [r6] @ get value + mov r5, r5, lsr #8 @ isolate rfr field and drop burst + + cmp r0, #0x1 @ going to half speed? + movne r9, #0x0 @ if up set flag up for pre up, hi volt + + blne voltage_shift_c @ adjust voltage + + cmp r0, #0x1 @ going to half speed (post branch link) + moveq r5, r5, lsr #1 @ divide by 2 if to half + movne r5, r5, lsl #1 @ mult by 2 if to full + mov r5, r5, lsl #8 @ put rfr field back into place + add r5, r5, #0x1 @ turn on burst of 1 + ldr r4, ddr_cm_clksel2_pll @ get address of out reg + ldr r3, [r4] @ get curr value + orr r3, r3, #0x3 + bic r3, r3, #0x3 @ clear lower bits + orr r3, r3, r0 @ new state value + str r3, [r4] @ set new state (pll/x, x=1 or 2) + nop + nop + + moveq r9, #0x1 @ if speed down, post down, drop volt + bleq voltage_shift_c + + mcr p15, 0, r3, c7, c10, 4 @ memory barrier + str r5, [r6] @ set new RFR_1 value + add r6, r6, #0x30 @ get RFR_2 addr + str r5, [r6] @ set RFR_2 + nop + cmp r2, #0x1 @ (SDR or DDR) do we need to adjust DLL + bne freq_out @ leave if SDR, no DLL function + + /* With DDR, we need to take care of the DLL for the frequency change */ + ldr r2, ddr_sdrc_dlla_ctrl @ addr of dlla ctrl + str r1, [r2] @ write out new SDRC_DLLA_CTRL + add r2, r2, #0x8 @ addr to SDRC_DLLB_CTRL + str r1, [r2] @ commit to SDRC_DLLB_CTRL + mov r1, #0x2000 @ wait DLL relock, min 0x400 L3 clocks +dll_wait: + subs r1, r1, #0x1 + bne dll_wait +freq_out: + ldmfd sp!, {r0 - r10, pc} @ restore regs and return + + /* + * shift up or down voltage, use R9 as input to tell level. + * wait for it to finish, use 32k sync counter, 1tick=31uS. + */ +voltage_shift_c: + ldr r10, ddr_prcm_voltctrl @ get addr of volt ctrl + ldr r8, [r10] @ get value + ldr r7, ddr_prcm_mask_val @ get value of mask + and r8, r8, r7 @ apply mask to clear bits + orr r8, r8, r9 @ bulld value for L0/L1-volt operation. + str r8, [r10] @ set up for change. + mov r7, #0x4000 @ get val for force + orr r8, r8, r7 @ build value for force + str r8, [r10] @ Force transition to L1 + + ldr r10, ddr_timer_32ksynct @ get addr of counter + ldr r8, [r10] @ get value + add r8, r8, #0x2 @ give it at most 62uS (min 31+) +volt_delay_c: + ldr r7, [r10] @ get timer value + cmp r8, r7 @ time up? + bhi volt_delay_c @ not yet->branch + mov pc, lr @ back to caller + +ddr_cm_clksel2_pll: + .word CM_CLKSEL2_PLL_V +ddr_sdrc_dlla_ctrl: + .word SDRC_DLLA_CTRL_V +ddr_sdrc_rfr_ctrl: + .word SDRC_RFR_CTRL_V +ddr_prcm_voltctrl: + .word PRCM_VOLTCTRL_V +ddr_prcm_mask_val: + .word 0xFFFF3FFC +ddr_timer_32ksynct: + .word TIMER_32KSYNCT_CR_V + +ENTRY(sram_reprogram_sdrc_sz) + .word . - sram_reprogram_sdrc + +/* + * Set dividers and pll. Also recalculate DLL value for DDR and unlock mode. + */ +ENTRY(sram_set_prcm) + stmfd sp!, {r0-r12, lr} @ regs to stack + adr r4, pbegin @ addr of preload start + adr r8, pend @ addr of preload end + mcrr p15, 1, r8, r4, c12 @ preload into icache +pbegin: + /* move into fast relock bypass */ + ldr r8, pll_ctl @ get addr + ldr r5, [r8] @ get val + mvn r6, #0x3 @ clear mask + and r5, r5, r6 @ clear field + orr r7, r5, #0x2 @ fast relock val + str r7, [r8] @ go to fast relock + ldr r4, pll_stat @ addr of stat +block: + /* wait for bypass */ + ldr r8, [r4] @ stat value + and r8, r8, #0x3 @ mask for stat + cmp r8, #0x1 @ there yet + bne block @ loop if not + + /* set new dpll dividers _after_ in bypass */ + ldr r4, pll_div @ get addr + str r0, [r4] @ set dpll ctrl val + + ldr r4, set_config @ get addr + mov r8, #1 @ valid cfg msk + str r8, [r4] @ make dividers take + + mov r4, #100 @ dead spin a bit +wait_a_bit: + subs r4, r4, #1 @ dec loop + bne wait_a_bit @ delay done? + + /* check if staying in bypass */ + cmp r2, #0x1 @ stay in bypass? + beq pend @ jump over dpll relock + + /* relock DPLL with new vals */ + ldr r5, pll_stat @ get addr + ldr r4, pll_ctl @ get addr + orr r8, r7, #0x3 @ val for lock dpll + str r8, [r4] @ set val + mov r0, #1000 @ dead spin a bit +wait_more: + subs r0, r0, #1 @ dec loop + bne wait_more @ delay done? +wait_lock: + ldr r8, [r5] @ get lock val + and r8, r8, #3 @ isolate field + cmp r8, #2 @ locked? + bne wait_lock @ wait if not +pend: + /* update memory timings & briefly lock dll */ + ldr r4, sdrc_rfr @ get addr + str r1, [r4] @ update refresh timing + ldr r11, dlla_ctrl @ get addr of DLLA ctrl + ldr r10, [r11] @ get current val + mvn r9, #0x4 @ mask to get clear bit2 + and r10, r10, r9 @ clear bit2 for lock mode + orr r10, r10, #0x8 @ make sure DLL on (es2 bit pos) + str r10, [r11] @ commit to DLLA_CTRL + add r11, r11, #0x8 @ move to dllb + str r10, [r11] @ hit DLLB also + + mov r4, #0x800 @ relock time (min 0x400 L3 clocks) +wait_dll_lock: + subs r4, r4, #0x1 + bne wait_dll_lock + nop + ldmfd sp!, {r0-r12, pc} @ restore regs and return + +set_config: + .word PRCM_CLKCFG_CTRL_V +pll_ctl: + .word CM_CLKEN_PLL_V +pll_stat: + .word CM_IDLEST_CKGEN_V +pll_div: + .word CM_CLKSEL1_PLL_V +sdrc_rfr: + .word SDRC_RFR_CTRL_V +dlla_ctrl: + .word SDRC_DLLA_CTRL_V + +ENTRY(sram_set_prcm_sz) + .word . - sram_set_prcm diff --git a/arch/arm/mach-omap2/timer-gp.c b/arch/arm/mach-omap2/timer-gp.c new file mode 100644 index 00000000000..9ec11443200 --- /dev/null +++ b/arch/arm/mach-omap2/timer-gp.c @@ -0,0 +1,126 @@ +/* + * linux/arch/arm/mach-omap2/timer-gp.c + * + * OMAP2 GP timer support. + * + * Copyright (C) 2005 Nokia Corporation + * Author: Paul Mundt <paul.mundt@nokia.com> + * Juha Yrjölä <juha.yrjola@nokia.com> + * + * Some parts based off of TI's 24xx code: + * + * Copyright (C) 2004 Texas Instruments, Inc. + * + * Roughly modelled after the OMAP1 MPU timer code. + * + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + */ +#include <linux/init.h> +#include <linux/time.h> +#include <linux/interrupt.h> +#include <linux/err.h> +#include <asm/mach/time.h> +#include <asm/delay.h> +#include <asm/io.h> +#include <asm/hardware/clock.h> + +#define OMAP2_GP_TIMER1_BASE 0x48028000 +#define OMAP2_GP_TIMER2_BASE 0x4802a000 +#define OMAP2_GP_TIMER3_BASE 0x48078000 +#define OMAP2_GP_TIMER4_BASE 0x4807a000 + +#define GP_TIMER_TIDR 0x00 +#define GP_TIMER_TISR 0x18 +#define GP_TIMER_TIER 0x1c +#define GP_TIMER_TCLR 0x24 +#define GP_TIMER_TCRR 0x28 +#define GP_TIMER_TLDR 0x2c +#define GP_TIMER_TSICR 0x40 + +#define OS_TIMER_NR 1 /* GP timer 2 */ + +static unsigned long timer_base[] = { + IO_ADDRESS(OMAP2_GP_TIMER1_BASE), + IO_ADDRESS(OMAP2_GP_TIMER2_BASE), + IO_ADDRESS(OMAP2_GP_TIMER3_BASE), + IO_ADDRESS(OMAP2_GP_TIMER4_BASE), +}; + +static inline unsigned int timer_read_reg(int nr, unsigned int reg) +{ + return __raw_readl(timer_base[nr] + reg); +} + +static inline void timer_write_reg(int nr, unsigned int reg, unsigned int val) +{ + __raw_writel(val, timer_base[nr] + reg); +} + +/* Note that we always enable the clock prescale divider bit */ +static inline void omap2_gp_timer_start(int nr, unsigned long load_val) +{ + unsigned int tmp; + + tmp = 0xffffffff - load_val; + + timer_write_reg(nr, GP_TIMER_TLDR, tmp); + timer_write_reg(nr, GP_TIMER_TCRR, tmp); + timer_write_reg(nr, GP_TIMER_TIER, 1 << 1); + timer_write_reg(nr, GP_TIMER_TCLR, (1 << 5) | (1 << 1) | 1); +} + +static irqreturn_t omap2_gp_timer_interrupt(int irq, void *dev_id, + struct pt_regs *regs) +{ + write_seqlock(&xtime_lock); + + timer_write_reg(OS_TIMER_NR, GP_TIMER_TISR, 1 << 1); + timer_tick(regs); + + write_sequnlock(&xtime_lock); + + return IRQ_HANDLED; +} + +static struct irqaction omap2_gp_timer_irq = { + .name = "gp timer", + .flags = SA_INTERRUPT, + .handler = omap2_gp_timer_interrupt, +}; + +static void __init omap2_gp_timer_init(void) +{ + struct clk * sys_ck; + u32 tick_period = 120000; + u32 l; + + /* Reset clock and prescale value */ + timer_write_reg(OS_TIMER_NR, GP_TIMER_TCLR, 0); + + sys_ck = clk_get(NULL, "sys_ck"); + if (IS_ERR(sys_ck)) + printk(KERN_ERR "Could not get sys_ck\n"); + else { + clk_use(sys_ck); + tick_period = clk_get_rate(sys_ck) / 100; + clk_put(sys_ck); + } + + tick_period /= 2; /* Minimum prescale divider is 2 */ + tick_period -= 1; + + l = timer_read_reg(OS_TIMER_NR, GP_TIMER_TIDR); + printk(KERN_INFO "OMAP2 GP timer (HW version %d.%d)\n", + (l >> 4) & 0x0f, l & 0x0f); + + setup_irq(38, &omap2_gp_timer_irq); + + omap2_gp_timer_start(OS_TIMER_NR, tick_period); +} + +struct sys_timer omap_timer = { + .init = omap2_gp_timer_init, +}; + diff --git a/arch/arm/mach-pxa/Kconfig b/arch/arm/mach-pxa/Kconfig index b380a438e68..e201aa9765b 100644 --- a/arch/arm/mach-pxa/Kconfig +++ b/arch/arm/mach-pxa/Kconfig @@ -60,6 +60,7 @@ config MACH_CORGI bool "Enable Sharp SL-C700 (Corgi) Support" depends PXA_SHARPSL_25x select PXA_SHARP_C7xx + select PXA_SSP config MACH_SHEPHERD bool "Enable Sharp SL-C750 (Shepherd) Support" @@ -102,12 +103,18 @@ config IWMMXT config PXA_SHARP_C7xx bool + select PXA_SSP help Enable support for all Sharp C7xx models config PXA_SHARP_Cxx00 bool + select PXA_SSP help Enable common support for Sharp Cxx00 models +config PXA_SSP + tristate + help + Enable support for PXA2xx SSP ports endif diff --git a/arch/arm/mach-pxa/Makefile b/arch/arm/mach-pxa/Makefile index 8bc72d07cea..d210bd5032c 100644 --- a/arch/arm/mach-pxa/Makefile +++ b/arch/arm/mach-pxa/Makefile @@ -11,8 +11,8 @@ obj-$(CONFIG_PXA27x) += pxa27x.o obj-$(CONFIG_ARCH_LUBBOCK) += lubbock.o obj-$(CONFIG_MACH_MAINSTONE) += mainstone.o obj-$(CONFIG_ARCH_PXA_IDP) += idp.o -obj-$(CONFIG_PXA_SHARP_C7xx) += corgi.o corgi_ssp.o corgi_lcd.o ssp.o -obj-$(CONFIG_PXA_SHARP_Cxx00) += spitz.o corgi_ssp.o corgi_lcd.o ssp.o +obj-$(CONFIG_PXA_SHARP_C7xx) += corgi.o corgi_ssp.o corgi_lcd.o +obj-$(CONFIG_PXA_SHARP_Cxx00) += spitz.o corgi_ssp.o corgi_lcd.o obj-$(CONFIG_MACH_POODLE) += poodle.o obj-$(CONFIG_MACH_TOSA) += tosa.o @@ -26,6 +26,7 @@ obj-$(CONFIG_LEDS) += $(led-y) # Misc features obj-$(CONFIG_PM) += pm.o sleep.o +obj-$(CONFIG_PXA_SSP) += ssp.o ifeq ($(CONFIG_PXA27x),y) obj-$(CONFIG_PM) += standby.o diff --git a/arch/arm/mach-pxa/corgi_ssp.c b/arch/arm/mach-pxa/corgi_ssp.c index 591e5f32dbe..bdf10cfa944 100644 --- a/arch/arm/mach-pxa/corgi_ssp.c +++ b/arch/arm/mach-pxa/corgi_ssp.c @@ -203,7 +203,7 @@ static int __init corgi_ssp_probe(struct device *dev) GPDR(ssp_machinfo->cs_ads7846) |= GPIO_bit(ssp_machinfo->cs_ads7846); /* output */ GPSR(ssp_machinfo->cs_ads7846) = GPIO_bit(ssp_machinfo->cs_ads7846); /* High - Disable ADS7846*/ - ret = ssp_init(&corgi_ssp_dev,ssp_machinfo->port); + ret = ssp_init(&corgi_ssp_dev, ssp_machinfo->port, 0); if (ret) printk(KERN_ERR "Unable to register SSP handler!\n"); diff --git a/arch/arm/mach-pxa/sharpsl.h b/arch/arm/mach-pxa/sharpsl.h index 3977a77aacd..4879c0f7da7 100644 --- a/arch/arm/mach-pxa/sharpsl.h +++ b/arch/arm/mach-pxa/sharpsl.h @@ -32,3 +32,90 @@ void corgi_put_hsync(void); void spitz_put_hsync(void); void corgi_wait_hsync(void); void spitz_wait_hsync(void); + +/* + * SharpSL Battery/PM Driver + */ + +struct sharpsl_charger_machinfo { + void (*init)(void); + int gpio_acin; + int gpio_batfull; + int gpio_batlock; + int gpio_fatal; + int (*status_acin)(void); + void (*discharge)(int); + void (*discharge1)(int); + void (*charge)(int); + void (*chargeled)(int); + void (*measure_temp)(int); + void (*presuspend)(void); + void (*postsuspend)(void); + unsigned long (*charger_wakeup)(void); + int (*should_wakeup)(unsigned int resume_on_alarm); + int bat_levels; + struct battery_thresh *bat_levels_noac; + struct battery_thresh *bat_levels_acin; + int status_high_acin; + int status_low_acin; + int status_high_noac; + int status_low_noac; +}; + +struct battery_thresh { + int voltage; + int percentage; +}; + +struct battery_stat { + int ac_status; /* APM AC Present/Not Present */ + int mainbat_status; /* APM Main Battery Status */ + int mainbat_percent; /* Main Battery Percentage Charge */ + int mainbat_voltage; /* Main Battery Voltage */ +}; + +struct sharpsl_pm_status { + struct device *dev; + struct timer_list ac_timer; + struct timer_list chrg_full_timer; + + int charge_mode; +#define CHRG_ERROR (-1) +#define CHRG_OFF (0) +#define CHRG_ON (1) +#define CHRG_DONE (2) + + unsigned int flags; +#define SHARPSL_SUSPENDED (1 << 0) /* Device is Suspended */ +#define SHARPSL_ALARM_ACTIVE (1 << 1) /* Alarm is for charging event (not user) */ +#define SHARPSL_BL_LIMIT (1 << 2) /* Backlight Intensity Limited */ +#define SHARPSL_APM_QUEUED (1 << 3) /* APM Event Queued */ +#define SHARPSL_DO_OFFLINE_CHRG (1 << 4) /* Trigger the offline charger */ + + int full_count; + unsigned long charge_start_time; + struct sharpsl_charger_machinfo *machinfo; + struct battery_stat battstat; +}; + +extern struct sharpsl_pm_status sharpsl_pm; +extern struct battery_thresh spitz_battery_levels_acin[]; +extern struct battery_thresh spitz_battery_levels_noac[]; + +#define READ_GPIO_BIT(x) (GPLR(x) & GPIO_bit(x)) + +#define SHARPSL_LED_ERROR 2 +#define SHARPSL_LED_ON 1 +#define SHARPSL_LED_OFF 0 + +#define CHARGE_ON() sharpsl_pm.machinfo->charge(1) +#define CHARGE_OFF() sharpsl_pm.machinfo->charge(0) +#define CHARGE_LED_ON() sharpsl_pm.machinfo->chargeled(SHARPSL_LED_ON) +#define CHARGE_LED_OFF() sharpsl_pm.machinfo->chargeled(SHARPSL_LED_OFF) +#define CHARGE_LED_ERR() sharpsl_pm.machinfo->chargeled(SHARPSL_LED_ERROR) +#define DISCHARGE_ON() sharpsl_pm.machinfo->discharge(1) +#define DISCHARGE_OFF() sharpsl_pm.machinfo->discharge(0) +#define STATUS_AC_IN sharpsl_pm.machinfo->status_acin() +#define STATUS_BATT_LOCKED READ_GPIO_BIT(sharpsl_pm.machinfo->gpio_batlock) +#define STATUS_CHRG_FULL READ_GPIO_BIT(sharpsl_pm.machinfo->gpio_batfull) +#define STATUS_FATAL READ_GPIO_BIT(sharpsl_pm.machinfo->gpio_fatal) diff --git a/arch/arm/mach-pxa/sharpsl_pm.c b/arch/arm/mach-pxa/sharpsl_pm.c new file mode 100644 index 00000000000..6c9e871c53d --- /dev/null +++ b/arch/arm/mach-pxa/sharpsl_pm.c @@ -0,0 +1,992 @@ +/* + * Battery and Power Management code for the Sharp SL-C7xx and SL-Cxx00 + * series of PDAs + * + * Copyright (c) 2004-2005 Richard Purdie + * + * Based on code written by Sharp for 2.4 kernels + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ + +#undef DEBUG + +#include <linux/module.h> +#include <linux/timer.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/apm_bios.h> +#include <linux/delay.h> +#include <linux/interrupt.h> +#include <linux/device.h> + +#include <asm/hardware.h> +#include <asm/hardware/scoop.h> +#include <asm/mach-types.h> +#include <asm/irq.h> +#include <asm/apm.h> + +#include <asm/arch/pm.h> +#include <asm/arch/pxa-regs.h> +#include <asm/arch/sharpsl.h> +#include "sharpsl.h" + +/* + * Constants + */ +#define SHARPSL_CHARGE_ON_TIME_INTERVAL (msecs_to_jiffies(1*60*1000)) /* 1 min */ +#define SHARPSL_CHARGE_FINISH_TIME (msecs_to_jiffies(10*60*1000)) /* 10 min */ +#define SHARPSL_BATCHK_TIME (msecs_to_jiffies(15*1000)) /* 15 sec */ +#define SHARPSL_BATCHK_TIME_SUSPEND (60*10) /* 10 min */ +#define SHARPSL_WAIT_CO_TIME 15 /* 15 sec */ +#define SHARPSL_WAIT_DISCHARGE_ON 100 /* 100 msec */ +#define SHARPSL_CHECK_BATTERY_WAIT_TIME_TEMP 10 /* 10 msec */ +#define SHARPSL_CHECK_BATTERY_WAIT_TIME_VOLT 10 /* 10 msec */ +#define SHARPSL_CHECK_BATTERY_WAIT_TIME_JKVAD 10 /* 10 msec */ +#define SHARPSL_CHARGE_WAIT_TIME 15 /* 15 msec */ +#define SHARPSL_CHARGE_CO_CHECK_TIME 5 /* 5 msec */ +#define SHARPSL_CHARGE_RETRY_CNT 1 /* eqv. 10 min */ + +#define SHARPSL_CHARGE_ON_VOLT 0x99 /* 2.9V */ +#define SHARPSL_CHARGE_ON_TEMP 0xe0 /* 2.9V */ +#define SHARPSL_CHARGE_ON_JKVAD_HIGH 0x9b /* 6V */ +#define SHARPSL_CHARGE_ON_JKVAD_LOW 0x34 /* 2V */ +#define SHARPSL_FATAL_ACIN_VOLT 182 /* 3.45V */ +#define SHARPSL_FATAL_NOACIN_VOLT 170 /* 3.40V */ + +struct battery_thresh spitz_battery_levels_acin[] = { + { 213, 100}, + { 212, 98}, + { 211, 95}, + { 210, 93}, + { 209, 90}, + { 208, 88}, + { 207, 85}, + { 206, 83}, + { 205, 80}, + { 204, 78}, + { 203, 75}, + { 202, 73}, + { 201, 70}, + { 200, 68}, + { 199, 65}, + { 198, 63}, + { 197, 60}, + { 196, 58}, + { 195, 55}, + { 194, 53}, + { 193, 50}, + { 192, 48}, + { 192, 45}, + { 191, 43}, + { 191, 40}, + { 190, 38}, + { 190, 35}, + { 189, 33}, + { 188, 30}, + { 187, 28}, + { 186, 25}, + { 185, 23}, + { 184, 20}, + { 183, 18}, + { 182, 15}, + { 181, 13}, + { 180, 10}, + { 179, 8}, + { 178, 5}, + { 0, 0}, +}; + +struct battery_thresh spitz_battery_levels_noac[] = { + { 213, 100}, + { 212, 98}, + { 211, 95}, + { 210, 93}, + { 209, 90}, + { 208, 88}, + { 207, 85}, + { 206, 83}, + { 205, 80}, + { 204, 78}, + { 203, 75}, + { 202, 73}, + { 201, 70}, + { 200, 68}, + { 199, 65}, + { 198, 63}, + { 197, 60}, + { 196, 58}, + { 195, 55}, + { 194, 53}, + { 193, 50}, + { 192, 48}, + { 191, 45}, + { 190, 43}, + { 189, 40}, + { 188, 38}, + { 187, 35}, + { 186, 33}, + { 185, 30}, + { 184, 28}, + { 183, 25}, + { 182, 23}, + { 181, 20}, + { 180, 18}, + { 179, 15}, + { 178, 13}, + { 177, 10}, + { 176, 8}, + { 175, 5}, + { 0, 0}, +}; + +/* MAX1111 Commands */ +#define MAXCTRL_PD0 1u << 0 +#define MAXCTRL_PD1 1u << 1 +#define MAXCTRL_SGL 1u << 2 +#define MAXCTRL_UNI 1u << 3 +#define MAXCTRL_SEL_SH 4 +#define MAXCTRL_STR 1u << 7 + +/* MAX1111 Channel Definitions */ +#define BATT_AD 4u +#define BATT_THM 2u +#define JK_VAD 6u + + +/* + * Prototypes + */ +static int sharpsl_read_MainBattery(void); +static int sharpsl_off_charge_battery(void); +static int sharpsl_check_battery(int mode); +static int sharpsl_ac_check(void); +static int sharpsl_fatal_check(void); +static int sharpsl_average_value(int ad); +static void sharpsl_average_clear(void); +static void sharpsl_charge_toggle(void *private_); +static void sharpsl_battery_thread(void *private_); + + +/* + * Variables + */ +struct sharpsl_pm_status sharpsl_pm; +DECLARE_WORK(toggle_charger, sharpsl_charge_toggle, NULL); +DECLARE_WORK(sharpsl_bat, sharpsl_battery_thread, NULL); + + +static int get_percentage(int voltage) +{ + int i = sharpsl_pm.machinfo->bat_levels - 1; + struct battery_thresh *thresh; + + if (sharpsl_pm.charge_mode == CHRG_ON) + thresh=sharpsl_pm.machinfo->bat_levels_acin; + else + thresh=sharpsl_pm.machinfo->bat_levels_noac; + + while (i > 0 && (voltage > thresh[i].voltage)) + i--; + + return thresh[i].percentage; +} + +static int get_apm_status(int voltage) +{ + int low_thresh, high_thresh; + + if (sharpsl_pm.charge_mode == CHRG_ON) { + high_thresh = sharpsl_pm.machinfo->status_high_acin; + low_thresh = sharpsl_pm.machinfo->status_low_acin; + } else { + high_thresh = sharpsl_pm.machinfo->status_high_noac; + low_thresh = sharpsl_pm.machinfo->status_low_noac; + } + + if (voltage >= high_thresh) + return APM_BATTERY_STATUS_HIGH; + if (voltage >= low_thresh) + return APM_BATTERY_STATUS_LOW; + return APM_BATTERY_STATUS_CRITICAL; +} + +void sharpsl_battery_kick(void) +{ + schedule_delayed_work(&sharpsl_bat, msecs_to_jiffies(125)); +} +EXPORT_SYMBOL(sharpsl_battery_kick); + + +static void sharpsl_battery_thread(void *private_) +{ + int voltage, percent, apm_status, i = 0; + + if (!sharpsl_pm.machinfo) + return; + + sharpsl_pm.battstat.ac_status = (!(STATUS_AC_IN) ? APM_AC_OFFLINE : APM_AC_ONLINE); + + /* Corgi cannot confirm when battery fully charged so periodically kick! */ + if (machine_is_corgi() && (sharpsl_pm.charge_mode == CHRG_ON) + && time_after(jiffies, sharpsl_pm.charge_start_time + SHARPSL_CHARGE_ON_TIME_INTERVAL)) + schedule_work(&toggle_charger); + + while(1) { + voltage = sharpsl_read_MainBattery(); + if (voltage > 0) break; + if (i++ > 5) { + voltage = sharpsl_pm.machinfo->bat_levels_noac[0].voltage; + dev_warn(sharpsl_pm.dev, "Warning: Cannot read main battery!\n"); + break; + } + } + + voltage = sharpsl_average_value(voltage); + apm_status = get_apm_status(voltage); + percent = get_percentage(voltage); + + /* At low battery voltages, the voltage has a tendency to start + creeping back up so we try to avoid this here */ + if ((sharpsl_pm.battstat.ac_status == APM_AC_ONLINE) || (apm_status == APM_BATTERY_STATUS_HIGH) || percent <= sharpsl_pm.battstat.mainbat_percent) { + sharpsl_pm.battstat.mainbat_voltage = voltage; + sharpsl_pm.battstat.mainbat_status = apm_status; + sharpsl_pm.battstat.mainbat_percent = percent; + } + + dev_dbg(sharpsl_pm.dev, "Battery: voltage: %d, status: %d, percentage: %d, time: %d\n", voltage, + sharpsl_pm.battstat.mainbat_status, sharpsl_pm.battstat.mainbat_percent, jiffies); + + /* If battery is low. limit backlight intensity to save power. */ + if ((sharpsl_pm.battstat.ac_status != APM_AC_ONLINE) + && ((sharpsl_pm.battstat.mainbat_status == APM_BATTERY_STATUS_LOW) || + (sharpsl_pm.battstat.mainbat_status == APM_BATTERY_STATUS_CRITICAL))) { + if (!(sharpsl_pm.flags & SHARPSL_BL_LIMIT)) { + corgibl_limit_intensity(1); + sharpsl_pm.flags |= SHARPSL_BL_LIMIT; + } + } else if (sharpsl_pm.flags & SHARPSL_BL_LIMIT) { + corgibl_limit_intensity(0); + sharpsl_pm.flags &= ~SHARPSL_BL_LIMIT; + } + + /* Suspend if critical battery level */ + if ((sharpsl_pm.battstat.ac_status != APM_AC_ONLINE) + && (sharpsl_pm.battstat.mainbat_status == APM_BATTERY_STATUS_CRITICAL) + && !(sharpsl_pm.flags & SHARPSL_APM_QUEUED)) { + sharpsl_pm.flags |= SHARPSL_APM_QUEUED; + dev_err(sharpsl_pm.dev, "Fatal Off\n"); + apm_queue_event(APM_CRITICAL_SUSPEND); + } + + schedule_delayed_work(&sharpsl_bat, SHARPSL_BATCHK_TIME); +} + +static void sharpsl_charge_on(void) +{ + dev_dbg(sharpsl_pm.dev, "Turning Charger On\n"); + + sharpsl_pm.full_count = 0; + sharpsl_pm.charge_mode = CHRG_ON; + schedule_delayed_work(&toggle_charger, msecs_to_jiffies(250)); + schedule_delayed_work(&sharpsl_bat, msecs_to_jiffies(500)); +} + +static void sharpsl_charge_off(void) +{ + dev_dbg(sharpsl_pm.dev, "Turning Charger Off\n"); + + CHARGE_OFF(); + CHARGE_LED_OFF(); + sharpsl_pm.charge_mode = CHRG_OFF; + + schedule_work(&sharpsl_bat); +} + +static void sharpsl_charge_error(void) +{ + CHARGE_LED_ERR(); + CHARGE_OFF(); + sharpsl_pm.charge_mode = CHRG_ERROR; +} + +static void sharpsl_charge_toggle(void *private_) +{ + dev_dbg(sharpsl_pm.dev, "Toogling Charger at time: %lx\n", jiffies); + + if (STATUS_AC_IN == 0) { + sharpsl_charge_off(); + return; + } else if ((sharpsl_check_battery(1) < 0) || (sharpsl_ac_check() < 0)) { + sharpsl_charge_error(); + return; + } + + CHARGE_LED_ON(); + CHARGE_OFF(); + mdelay(SHARPSL_CHARGE_WAIT_TIME); + CHARGE_ON(); + + sharpsl_pm.charge_start_time = jiffies; +} + +static void sharpsl_ac_timer(unsigned long data) +{ + int acin = STATUS_AC_IN; + + dev_dbg(sharpsl_pm.dev, "AC Status: %d\n",acin); + + sharpsl_average_clear(); + if (acin && (sharpsl_pm.charge_mode != CHRG_ON)) + sharpsl_charge_on(); + else if (sharpsl_pm.charge_mode == CHRG_ON) + sharpsl_charge_off(); + + schedule_work(&sharpsl_bat); +} + + +static irqreturn_t sharpsl_ac_isr(int irq, void *dev_id, struct pt_regs *fp) +{ + /* Delay the event slightly to debounce */ + /* Must be a smaller delay than the chrg_full_isr below */ + mod_timer(&sharpsl_pm.ac_timer, jiffies + msecs_to_jiffies(250)); + + return IRQ_HANDLED; +} + +static void sharpsl_chrg_full_timer(unsigned long data) +{ + dev_dbg(sharpsl_pm.dev, "Charge Full at time: %lx\n", jiffies); + + sharpsl_pm.full_count++; + + if (STATUS_AC_IN == 0) { + dev_dbg(sharpsl_pm.dev, "Charge Full: AC removed - stop charging!\n"); + if (sharpsl_pm.charge_mode == CHRG_ON) + sharpsl_charge_off(); + } else if (sharpsl_pm.full_count < 2) { + dev_dbg(sharpsl_pm.dev, "Charge Full: Count too low\n"); + schedule_work(&toggle_charger); + } else if (time_after(jiffies, sharpsl_pm.charge_start_time + SHARPSL_CHARGE_FINISH_TIME)) { + dev_dbg(sharpsl_pm.dev, "Charge Full: Interrupt generated too slowly - retry.\n"); + schedule_work(&toggle_charger); + } else { + sharpsl_charge_off(); + sharpsl_pm.charge_mode = CHRG_DONE; + dev_dbg(sharpsl_pm.dev, "Charge Full: Charging Finished\n"); + } +} + +/* Charging Finished Interrupt (Not present on Corgi) */ +/* Can trigger at the same time as an AC staus change so + delay until after that has been processed */ +static irqreturn_t sharpsl_chrg_full_isr(int irq, void *dev_id, struct pt_regs *fp) +{ + if (sharpsl_pm.flags & SHARPSL_SUSPENDED) + return IRQ_HANDLED; + + /* delay until after any ac interrupt */ + mod_timer(&sharpsl_pm.chrg_full_timer, jiffies + msecs_to_jiffies(500)); + + return IRQ_HANDLED; +} + +static irqreturn_t sharpsl_fatal_isr(int irq, void *dev_id, struct pt_regs *fp) +{ + int is_fatal = 0; + + if (STATUS_BATT_LOCKED == 0) { + dev_err(sharpsl_pm.dev, "Battery now Unlocked! Suspending.\n"); + is_fatal = 1; + } + + if (sharpsl_pm.machinfo->gpio_fatal && (STATUS_FATAL == 0)) { + dev_err(sharpsl_pm.dev, "Fatal Batt Error! Suspending.\n"); + is_fatal = 1; + } + + if (!(sharpsl_pm.flags & SHARPSL_APM_QUEUED) && is_fatal) { + sharpsl_pm.flags |= SHARPSL_APM_QUEUED; + apm_queue_event(APM_CRITICAL_SUSPEND); + } + + return IRQ_HANDLED; +} + +/* + * Maintain an average of the last 10 readings + */ +#define SHARPSL_CNV_VALUE_NUM 10 +static int sharpsl_ad_index; + +static void sharpsl_average_clear(void) +{ + sharpsl_ad_index = 0; +} + +static int sharpsl_average_value(int ad) +{ + int i, ad_val = 0; + static int sharpsl_ad[SHARPSL_CNV_VALUE_NUM+1]; + + if (sharpsl_pm.battstat.mainbat_status != APM_BATTERY_STATUS_HIGH) { + sharpsl_ad_index = 0; + return ad; + } + + sharpsl_ad[sharpsl_ad_index] = ad; + sharpsl_ad_index++; + if (sharpsl_ad_index >= SHARPSL_CNV_VALUE_NUM) { + for (i=0; i < (SHARPSL_CNV_VALUE_NUM-1); i++) + sharpsl_ad[i] = sharpsl_ad[i+1]; + sharpsl_ad_index = SHARPSL_CNV_VALUE_NUM - 1; + } + for (i=0; i < sharpsl_ad_index; i++) + ad_val += sharpsl_ad[i]; + + return (ad_val / sharpsl_ad_index); +} + + +/* + * Read MAX1111 ADC + */ +static int read_max1111(int channel) +{ + return corgi_ssp_max1111_get((channel << MAXCTRL_SEL_SH) | MAXCTRL_PD0 | MAXCTRL_PD1 + | MAXCTRL_SGL | MAXCTRL_UNI | MAXCTRL_STR); +} + +static int sharpsl_read_MainBattery(void) +{ + return read_max1111(BATT_AD); +} + +static int sharpsl_read_Temp(void) +{ + int temp; + + sharpsl_pm.machinfo->measure_temp(1); + + mdelay(SHARPSL_CHECK_BATTERY_WAIT_TIME_TEMP); + temp = read_max1111(BATT_THM); + + sharpsl_pm.machinfo->measure_temp(0); + + return temp; +} + +static int sharpsl_read_jkvad(void) +{ + return read_max1111(JK_VAD); +} + +/* + * Take an array of 5 integers, remove the maximum and minimum values + * and return the average. + */ +static int get_select_val(int *val) +{ + int i, j, k, temp, sum = 0; + + /* Find MAX val */ + temp = val[0]; + j=0; + for (i=1; i<5; i++) { + if (temp < val[i]) { + temp = val[i]; + j = i; + } + } + + /* Find MIN val */ + temp = val[4]; + k=4; + for (i=3; i>=0; i--) { + if (temp > val[i]) { + temp = val[i]; + k = i; + } + } + + for (i=0; i<5; i++) + if (i != j && i != k ) + sum += val[i]; + + dev_dbg(sharpsl_pm.dev, "Average: %d from values: %d, %d, %d, %d, %d\n", sum/3, val[0], val[1], val[2], val[3], val[4]); + + return (sum/3); +} + +/* mode 0 - Check temperature and voltage + * 1 - Check temperature only */ +static int sharpsl_check_battery(int mode) +{ + int val, i, buff[5]; + + /* Check battery temperature */ + for (i=0; i<5; i++) { + mdelay(SHARPSL_CHECK_BATTERY_WAIT_TIME_TEMP); + buff[i] = sharpsl_read_Temp(); + } + + val = get_select_val(buff); + + dev_dbg(sharpsl_pm.dev, "Temperature: %d\n", val); + if (val > SHARPSL_CHARGE_ON_TEMP) + return -1; + if (mode == 1) + return 0; + + /* disable charge, enable discharge */ + CHARGE_OFF(); + DISCHARGE_ON(); + mdelay(SHARPSL_WAIT_DISCHARGE_ON); + + if (sharpsl_pm.machinfo->discharge1) + sharpsl_pm.machinfo->discharge1(1); + + /* Check battery voltage */ + for (i=0; i<5; i++) { + buff[i] = sharpsl_read_MainBattery(); + mdelay(SHARPSL_CHECK_BATTERY_WAIT_TIME_VOLT); + } + + if (sharpsl_pm.machinfo->discharge1) + sharpsl_pm.machinfo->discharge1(0); + + DISCHARGE_OFF(); + + val = get_select_val(buff); + dev_dbg(sharpsl_pm.dev, "Battery Voltage: %d\n", val); + + if (val < SHARPSL_CHARGE_ON_VOLT) + return -1; + + return 0; +} + +static int sharpsl_ac_check(void) +{ + int temp, i, buff[5]; + + for (i=0; i<5; i++) { + buff[i] = sharpsl_read_jkvad(); + mdelay(SHARPSL_CHECK_BATTERY_WAIT_TIME_JKVAD); + } + + temp = get_select_val(buff); + dev_dbg(sharpsl_pm.dev, "AC Voltage: %d\n",temp); + + if ((temp > SHARPSL_CHARGE_ON_JKVAD_HIGH) || (temp < SHARPSL_CHARGE_ON_JKVAD_LOW)) { + dev_err(sharpsl_pm.dev, "Error: AC check failed.\n"); + return -1; + } + + return 0; +} + +#ifdef CONFIG_PM +static int sharpsl_pm_suspend(struct device *dev, pm_message_t state) +{ + sharpsl_pm.flags |= SHARPSL_SUSPENDED; + flush_scheduled_work(); + + if (sharpsl_pm.charge_mode == CHRG_ON) + sharpsl_pm.flags |= SHARPSL_DO_OFFLINE_CHRG; + else + sharpsl_pm.flags &= ~SHARPSL_DO_OFFLINE_CHRG; + + return 0; +} + +static int sharpsl_pm_resume(struct device *dev) +{ + /* Clear the reset source indicators as they break the bootloader upon reboot */ + RCSR = 0x0f; + sharpsl_average_clear(); + sharpsl_pm.flags &= ~SHARPSL_APM_QUEUED; + sharpsl_pm.flags &= ~SHARPSL_SUSPENDED; + + return 0; +} + +static void corgi_goto_sleep(unsigned long alarm_time, unsigned int alarm_enable, suspend_state_t state) +{ + dev_dbg(sharpsl_pm.dev, "Time is: %08x\n",RCNR); + + dev_dbg(sharpsl_pm.dev, "Offline Charge Activate = %d\n",sharpsl_pm.flags & SHARPSL_DO_OFFLINE_CHRG); + /* not charging and AC-IN! */ + + if ((sharpsl_pm.flags & SHARPSL_DO_OFFLINE_CHRG) && (STATUS_AC_IN != 0)) { + dev_dbg(sharpsl_pm.dev, "Activating Offline Charger...\n"); + sharpsl_pm.charge_mode = CHRG_OFF; + sharpsl_pm.flags &= ~SHARPSL_DO_OFFLINE_CHRG; + sharpsl_off_charge_battery(); + } + + sharpsl_pm.machinfo->presuspend(); + + PEDR = 0xffffffff; /* clear it */ + + sharpsl_pm.flags &= ~SHARPSL_ALARM_ACTIVE; + if ((sharpsl_pm.charge_mode == CHRG_ON) && ((alarm_enable && ((alarm_time - RCNR) > (SHARPSL_BATCHK_TIME_SUSPEND + 30))) || !alarm_enable)) { + RTSR &= RTSR_ALE; + RTAR = RCNR + SHARPSL_BATCHK_TIME_SUSPEND; + dev_dbg(sharpsl_pm.dev, "Charging alarm at: %08x\n",RTAR); + sharpsl_pm.flags |= SHARPSL_ALARM_ACTIVE; + } else if (alarm_enable) { + RTSR &= RTSR_ALE; + RTAR = alarm_time; + dev_dbg(sharpsl_pm.dev, "User alarm at: %08x\n",RTAR); + } else { + dev_dbg(sharpsl_pm.dev, "No alarms set.\n"); + } + + pxa_pm_enter(state); + + sharpsl_pm.machinfo->postsuspend(); + + dev_dbg(sharpsl_pm.dev, "Corgi woken up from suspend: %08x\n",PEDR); +} + +static int corgi_enter_suspend(unsigned long alarm_time, unsigned int alarm_enable, suspend_state_t state) +{ + if (!sharpsl_pm.machinfo->should_wakeup(!(sharpsl_pm.flags & SHARPSL_ALARM_ACTIVE) && alarm_enable) ) + { + if (!(sharpsl_pm.flags & SHARPSL_ALARM_ACTIVE)) { + dev_dbg(sharpsl_pm.dev, "No user triggered wakeup events and not charging. Strange. Suspend.\n"); + corgi_goto_sleep(alarm_time, alarm_enable, state); + return 1; + } + if(sharpsl_off_charge_battery()) { + dev_dbg(sharpsl_pm.dev, "Charging. Suspend...\n"); + corgi_goto_sleep(alarm_time, alarm_enable, state); + return 1; + } + dev_dbg(sharpsl_pm.dev, "User triggered wakeup in offline charger.\n"); + } + + if ((STATUS_BATT_LOCKED == 0) || (sharpsl_fatal_check() < 0) ) + { + dev_err(sharpsl_pm.dev, "Fatal condition. Suspend.\n"); + corgi_goto_sleep(alarm_time, alarm_enable, state); + return 1; + } + + return 0; +} + +static int corgi_pxa_pm_enter(suspend_state_t state) +{ + unsigned long alarm_time = RTAR; + unsigned int alarm_status = ((RTSR & RTSR_ALE) != 0); + + dev_dbg(sharpsl_pm.dev, "SharpSL suspending for first time.\n"); + + corgi_goto_sleep(alarm_time, alarm_status, state); + + while (corgi_enter_suspend(alarm_time,alarm_status,state)) + {} + + dev_dbg(sharpsl_pm.dev, "SharpSL resuming...\n"); + + return 0; +} +#endif + + +/* + * Check for fatal battery errors + * Fatal returns -1 + */ +static int sharpsl_fatal_check(void) +{ + int buff[5], temp, i, acin; + + dev_dbg(sharpsl_pm.dev, "sharpsl_fatal_check entered\n"); + + /* Check AC-Adapter */ + acin = STATUS_AC_IN; + + if (acin && (sharpsl_pm.charge_mode == CHRG_ON)) { + CHARGE_OFF(); + udelay(100); + DISCHARGE_ON(); /* enable discharge */ + mdelay(SHARPSL_WAIT_DISCHARGE_ON); + } + + if (sharpsl_pm.machinfo->discharge1) + sharpsl_pm.machinfo->discharge1(1); + + /* Check battery : check inserting battery ? */ + for (i=0; i<5; i++) { + buff[i] = sharpsl_read_MainBattery(); + mdelay(SHARPSL_CHECK_BATTERY_WAIT_TIME_VOLT); + } + + if (sharpsl_pm.machinfo->discharge1) + sharpsl_pm.machinfo->discharge1(0); + + if (acin && (sharpsl_pm.charge_mode == CHRG_ON)) { + udelay(100); + CHARGE_ON(); + DISCHARGE_OFF(); + } + + temp = get_select_val(buff); + dev_dbg(sharpsl_pm.dev, "sharpsl_fatal_check: acin: %d, discharge voltage: %d, no discharge: %d\n", acin, temp, sharpsl_read_MainBattery()); + + if ((acin && (temp < SHARPSL_FATAL_ACIN_VOLT)) || + (!acin && (temp < SHARPSL_FATAL_NOACIN_VOLT))) + return -1; + return 0; +} + +static int sharpsl_off_charge_error(void) +{ + dev_err(sharpsl_pm.dev, "Offline Charger: Error occured.\n"); + CHARGE_OFF(); + CHARGE_LED_ERR(); + sharpsl_pm.charge_mode = CHRG_ERROR; + return 1; +} + +/* + * Charging Control while suspended + * Return 1 - go straight to sleep + * Return 0 - sleep or wakeup depending on other factors + */ +static int sharpsl_off_charge_battery(void) +{ + int time; + + dev_dbg(sharpsl_pm.dev, "Charge Mode: %d\n", sharpsl_pm.charge_mode); + + if (sharpsl_pm.charge_mode == CHRG_OFF) { + dev_dbg(sharpsl_pm.dev, "Offline Charger: Step 1\n"); + + /* AC Check */ + if ((sharpsl_ac_check() < 0) || (sharpsl_check_battery(1) < 0)) + return sharpsl_off_charge_error(); + + /* Start Charging */ + CHARGE_LED_ON(); + CHARGE_OFF(); + mdelay(SHARPSL_CHARGE_WAIT_TIME); + CHARGE_ON(); + + sharpsl_pm.charge_mode = CHRG_ON; + sharpsl_pm.full_count = 0; + + return 1; + } else if (sharpsl_pm.charge_mode != CHRG_ON) { + return 1; + } + + if (sharpsl_pm.full_count == 0) { + int time; + + dev_dbg(sharpsl_pm.dev, "Offline Charger: Step 2\n"); + + if (sharpsl_check_battery(0) < 0) + return sharpsl_off_charge_error(); + + CHARGE_OFF(); + mdelay(SHARPSL_CHARGE_WAIT_TIME); + CHARGE_ON(); + sharpsl_pm.charge_mode = CHRG_ON; + + mdelay(SHARPSL_CHARGE_CO_CHECK_TIME); + + time = RCNR; + while(1) { + /* Check if any wakeup event had occured */ + if (sharpsl_pm.machinfo->charger_wakeup() != 0) + return 0; + /* Check for timeout */ + if ((RCNR - time) > SHARPSL_WAIT_CO_TIME) + return 1; + if (STATUS_CHRG_FULL) { + dev_dbg(sharpsl_pm.dev, "Offline Charger: Charge full occured. Retrying to check\n"); + sharpsl_pm.full_count++; + CHARGE_OFF(); + mdelay(SHARPSL_CHARGE_WAIT_TIME); + CHARGE_ON(); + return 1; + } + } + } + + dev_dbg(sharpsl_pm.dev, "Offline Charger: Step 3\n"); + + mdelay(SHARPSL_CHARGE_CO_CHECK_TIME); + + time = RCNR; + while(1) { + /* Check if any wakeup event had occured */ + if (sharpsl_pm.machinfo->charger_wakeup() != 0) + return 0; + /* Check for timeout */ + if ((RCNR-time) > SHARPSL_WAIT_CO_TIME) { + if (sharpsl_pm.full_count > SHARPSL_CHARGE_RETRY_CNT) { + dev_dbg(sharpsl_pm.dev, "Offline Charger: Not charged sufficiently. Retrying.\n"); + sharpsl_pm.full_count = 0; + } + sharpsl_pm.full_count++; + return 1; + } + if (STATUS_CHRG_FULL) { + dev_dbg(sharpsl_pm.dev, "Offline Charger: Charging complete.\n"); + CHARGE_LED_OFF(); + CHARGE_OFF(); + sharpsl_pm.charge_mode = CHRG_DONE; + return 1; + } + } +} + + +static ssize_t battery_percentage_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + return sprintf(buf, "%d\n",sharpsl_pm.battstat.mainbat_percent); +} + +static ssize_t battery_voltage_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + return sprintf(buf, "%d\n",sharpsl_pm.battstat.mainbat_voltage); +} + +static DEVICE_ATTR(battery_percentage, 0444, battery_percentage_show, NULL); +static DEVICE_ATTR(battery_voltage, 0444, battery_voltage_show, NULL); + +extern void (*apm_get_power_status)(struct apm_power_info *); + +static void sharpsl_apm_get_power_status(struct apm_power_info *info) +{ + info->ac_line_status = sharpsl_pm.battstat.ac_status; + + if (sharpsl_pm.charge_mode == CHRG_ON) + info->battery_status = APM_BATTERY_STATUS_CHARGING; + else + info->battery_status = sharpsl_pm.battstat.mainbat_status; + + info->battery_flag = (1 << info->battery_status); + info->battery_life = sharpsl_pm.battstat.mainbat_percent; +} + +static struct pm_ops sharpsl_pm_ops = { + .pm_disk_mode = PM_DISK_FIRMWARE, + .prepare = pxa_pm_prepare, + .enter = corgi_pxa_pm_enter, + .finish = pxa_pm_finish, +}; + +static int __init sharpsl_pm_probe(struct device *dev) +{ + if (!dev->platform_data) + return -EINVAL; + + sharpsl_pm.dev = dev; + sharpsl_pm.machinfo = dev->platform_data; + sharpsl_pm.charge_mode = CHRG_OFF; + sharpsl_pm.flags = 0; + + sharpsl_pm.machinfo->init(); + + init_timer(&sharpsl_pm.ac_timer); + sharpsl_pm.ac_timer.function = sharpsl_ac_timer; + + init_timer(&sharpsl_pm.chrg_full_timer); + sharpsl_pm.chrg_full_timer.function = sharpsl_chrg_full_timer; + + pxa_gpio_mode(sharpsl_pm.machinfo->gpio_acin | GPIO_IN); + pxa_gpio_mode(sharpsl_pm.machinfo->gpio_batfull | GPIO_IN); + pxa_gpio_mode(sharpsl_pm.machinfo->gpio_batlock | GPIO_IN); + + /* Register interrupt handlers */ + if (request_irq(IRQ_GPIO(sharpsl_pm.machinfo->gpio_acin), sharpsl_ac_isr, SA_INTERRUPT, "AC Input Detect", sharpsl_ac_isr)) { + dev_err(sharpsl_pm.dev, "Could not get irq %d.\n", IRQ_GPIO(sharpsl_pm.machinfo->gpio_acin)); + } + else set_irq_type(IRQ_GPIO(sharpsl_pm.machinfo->gpio_acin),IRQT_BOTHEDGE); + + if (request_irq(IRQ_GPIO(sharpsl_pm.machinfo->gpio_batlock), sharpsl_fatal_isr, SA_INTERRUPT, "Battery Cover", sharpsl_fatal_isr)) { + dev_err(sharpsl_pm.dev, "Could not get irq %d.\n", IRQ_GPIO(sharpsl_pm.machinfo->gpio_batlock)); + } + else set_irq_type(IRQ_GPIO(sharpsl_pm.machinfo->gpio_batlock),IRQT_FALLING); + + if (sharpsl_pm.machinfo->gpio_fatal) { + if (request_irq(IRQ_GPIO(sharpsl_pm.machinfo->gpio_fatal), sharpsl_fatal_isr, SA_INTERRUPT, "Fatal Battery", sharpsl_fatal_isr)) { + dev_err(sharpsl_pm.dev, "Could not get irq %d.\n", IRQ_GPIO(sharpsl_pm.machinfo->gpio_fatal)); + } + else set_irq_type(IRQ_GPIO(sharpsl_pm.machinfo->gpio_fatal),IRQT_FALLING); + } + + if (!machine_is_corgi()) + { + /* Register interrupt handler. */ + if (request_irq(IRQ_GPIO(sharpsl_pm.machinfo->gpio_batfull), sharpsl_chrg_full_isr, SA_INTERRUPT, "CO", sharpsl_chrg_full_isr)) { + dev_err(sharpsl_pm.dev, "Could not get irq %d.\n", IRQ_GPIO(sharpsl_pm.machinfo->gpio_batfull)); + } + else set_irq_type(IRQ_GPIO(sharpsl_pm.machinfo->gpio_batfull),IRQT_RISING); + } + + device_create_file(dev, &dev_attr_battery_percentage); + device_create_file(dev, &dev_attr_battery_voltage); + + apm_get_power_status = sharpsl_apm_get_power_status; + + pm_set_ops(&sharpsl_pm_ops); + + mod_timer(&sharpsl_pm.ac_timer, jiffies + msecs_to_jiffies(250)); + + return 0; +} + +static int sharpsl_pm_remove(struct device *dev) +{ + pm_set_ops(NULL); + + device_remove_file(dev, &dev_attr_battery_percentage); + device_remove_file(dev, &dev_attr_battery_voltage); + + free_irq(IRQ_GPIO(sharpsl_pm.machinfo->gpio_acin), sharpsl_ac_isr); + free_irq(IRQ_GPIO(sharpsl_pm.machinfo->gpio_batlock), sharpsl_fatal_isr); + + if (sharpsl_pm.machinfo->gpio_fatal) + free_irq(IRQ_GPIO(sharpsl_pm.machinfo->gpio_fatal), sharpsl_fatal_isr); + + if (!machine_is_corgi()) + free_irq(IRQ_GPIO(sharpsl_pm.machinfo->gpio_batfull), sharpsl_chrg_full_isr); + + del_timer_sync(&sharpsl_pm.chrg_full_timer); + del_timer_sync(&sharpsl_pm.ac_timer); + + return 0; +} + +static struct device_driver sharpsl_pm_driver = { + .name = "sharpsl-pm", + .bus = &platform_bus_type, + .probe = sharpsl_pm_probe, + .remove = sharpsl_pm_remove, + .suspend = sharpsl_pm_suspend, + .resume = sharpsl_pm_resume, +}; + +static int __devinit sharpsl_pm_init(void) +{ + return driver_register(&sharpsl_pm_driver); +} + +static void sharpsl_pm_exit(void) +{ + driver_unregister(&sharpsl_pm_driver); +} + +late_initcall(sharpsl_pm_init); +module_exit(sharpsl_pm_exit); diff --git a/arch/arm/mach-pxa/ssp.c b/arch/arm/mach-pxa/ssp.c index 4d826c02131..a68b30eff4d 100644 --- a/arch/arm/mach-pxa/ssp.c +++ b/arch/arm/mach-pxa/ssp.c @@ -19,6 +19,8 @@ * 22nd Aug 2003 Initial version. * 20th Dec 2004 Added ssp_config for changing port config without * closing the port. + * 4th Aug 2005 Added option to disable irq handler registration and + * cleaned up irq and clock detection. */ #include <linux/module.h> @@ -37,6 +39,26 @@ #define PXA_SSP_PORTS 3 +struct ssp_info_ { + int irq; + u32 clock; +}; + +/* + * SSP port clock and IRQ settings + */ +static const struct ssp_info_ ssp_info[PXA_SSP_PORTS] = { +#if defined (CONFIG_PXA27x) + {IRQ_SSP, CKEN23_SSP1}, + {IRQ_SSP2, CKEN3_SSP2}, + {IRQ_SSP3, CKEN4_SSP3}, +#else + {IRQ_SSP, CKEN3_SSP}, + {IRQ_NSSP, CKEN9_NSSP}, + {IRQ_ASSP, CKEN10_ASSP}, +#endif +}; + static DECLARE_MUTEX(sem); static int use_count[PXA_SSP_PORTS] = {0, 0, 0}; @@ -210,9 +232,9 @@ int ssp_config(struct ssp_dev *dev, u32 mode, u32 flags, u32 psp_flags, u32 spee * %-EBUSY if the resources are already in use * %0 on success */ -int ssp_init(struct ssp_dev *dev, u32 port) +int ssp_init(struct ssp_dev *dev, u32 port, u32 init_flags) { - int ret, irq; + int ret; if (port > PXA_SSP_PORTS || port == 0) return -ENODEV; @@ -229,61 +251,20 @@ int ssp_init(struct ssp_dev *dev, u32 port) up(&sem); return -EBUSY; } - - switch (port) { - case 1: - irq = IRQ_SSP; - break; -#if defined (CONFIG_PXA27x) - case 2: - irq = IRQ_SSP2; - break; - case 3: - irq = IRQ_SSP3; - break; -#else - case 2: - irq = IRQ_NSSP; - break; - case 3: - irq = IRQ_ASSP; - break; -#endif - default: - return -ENODEV; - } - dev->port = port; - ret = request_irq(irq, ssp_interrupt, 0, "SSP", dev); - if (ret) - goto out_region; + /* do we need to get irq */ + if (!(init_flags & SSP_NO_IRQ)) { + ret = request_irq(ssp_info[port-1].irq, ssp_interrupt, + 0, "SSP", dev); + if (ret) + goto out_region; + dev->irq = ssp_info[port-1].irq; + } else + dev->irq = 0; /* turn on SSP port clock */ - switch (dev->port) { -#if defined (CONFIG_PXA27x) - case 1: - pxa_set_cken(CKEN23_SSP1, 1); - break; - case 2: - pxa_set_cken(CKEN3_SSP2, 1); - break; - case 3: - pxa_set_cken(CKEN4_SSP3, 1); - break; -#else - case 1: - pxa_set_cken(CKEN3_SSP, 1); - break; - case 2: - pxa_set_cken(CKEN9_NSSP, 1); - break; - case 3: - pxa_set_cken(CKEN10_ASSP, 1); - break; -#endif - } - + pxa_set_cken(ssp_info[port-1].clock, 1); up(&sem); return 0; @@ -301,46 +282,17 @@ out_region: */ void ssp_exit(struct ssp_dev *dev) { - int irq; - down(&sem); SSCR0_P(dev->port) &= ~SSCR0_SSE; - /* find irq, save power and turn off SSP port clock */ - switch (dev->port) { -#if defined (CONFIG_PXA27x) - case 1: - irq = IRQ_SSP; - pxa_set_cken(CKEN23_SSP1, 0); - break; - case 2: - irq = IRQ_SSP2; - pxa_set_cken(CKEN3_SSP2, 0); - break; - case 3: - irq = IRQ_SSP3; - pxa_set_cken(CKEN4_SSP3, 0); - break; -#else - case 1: - irq = IRQ_SSP; - pxa_set_cken(CKEN3_SSP, 0); - break; - case 2: - irq = IRQ_NSSP; - pxa_set_cken(CKEN9_NSSP, 0); - break; - case 3: - irq = IRQ_ASSP; - pxa_set_cken(CKEN10_ASSP, 0); - break; -#endif - default: - printk(KERN_WARNING "SSP: tried to close invalid port\n"); - return; + if (dev->port > PXA_SSP_PORTS || dev->port == 0) { + printk(KERN_WARNING "SSP: tried to close invalid port\n"); + return; } - free_irq(irq, dev); + pxa_set_cken(ssp_info[dev->port-1].clock, 0); + if (dev->irq) + free_irq(dev->irq, dev); release_mem_region(__PREG(SSCR0_P(dev->port)), 0x2c); use_count[dev->port - 1]--; up(&sem); diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig index e3c14d6b432..e84fdde6edf 100644 --- a/arch/arm/mm/Kconfig +++ b/arch/arm/mm/Kconfig @@ -102,8 +102,8 @@ config CPU_ARM922T # ARM925T config CPU_ARM925T bool "Support ARM925T processor" if ARCH_OMAP1 - depends on ARCH_OMAP1510 - default y if ARCH_OMAP1510 + depends on ARCH_OMAP15XX + default y if ARCH_OMAP15XX select CPU_32v4 select CPU_ABRT_EV4T select CPU_CACHE_V4WT @@ -242,7 +242,7 @@ config CPU_XSCALE # ARMv6 config CPU_V6 bool "Support ARM V6 processor" - depends on ARCH_INTEGRATOR || MACH_REALVIEW_EB + depends on ARCH_INTEGRATOR || MACH_REALVIEW_EB || ARCH_OMAP2 select CPU_32v6 select CPU_ABRT_EV6 select CPU_CACHE_V6 diff --git a/arch/arm/plat-omap/Makefile b/arch/arm/plat-omap/Makefile index 7e144f9cad1..9ccf1943fc9 100644 --- a/arch/arm/plat-omap/Makefile +++ b/arch/arm/plat-omap/Makefile @@ -3,7 +3,7 @@ # # Common support -obj-y := common.o sram.o sram-fn.o clock.o dma.o mux.o gpio.o mcbsp.o usb.o +obj-y := common.o sram.o sram-fn.o clock.o devices.o dma.o mux.o gpio.o mcbsp.o usb.o obj-m := obj-n := obj- := diff --git a/arch/arm/plat-omap/clock.c b/arch/arm/plat-omap/clock.c index a020fe16428..7ce39b986e2 100644 --- a/arch/arm/plat-omap/clock.c +++ b/arch/arm/plat-omap/clock.c @@ -1,15 +1,20 @@ /* * linux/arch/arm/plat-omap/clock.c * - * Copyright (C) 2004 Nokia corporation + * Copyright (C) 2004 - 2005 Nokia corporation * Written by Tuukka Tikkanen <tuukka.tikkanen@elektrobit.com> * + * Modified for omap shared clock framework by Tony Lindgren <tony@atomide.com> + * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as * published by the Free Software Foundation. */ -#include <linux/module.h> +#include <linux/version.h> +#include <linux/config.h> #include <linux/kernel.h> +#include <linux/init.h> +#include <linux/module.h> #include <linux/list.h> #include <linux/errno.h> #include <linux/err.h> @@ -18,562 +23,20 @@ #include <asm/io.h> #include <asm/semaphore.h> #include <asm/hardware/clock.h> -#include <asm/arch/board.h> -#include <asm/arch/usb.h> -#include "clock.h" -#include "sram.h" +#include <asm/arch/clock.h> -static LIST_HEAD(clocks); +LIST_HEAD(clocks); static DECLARE_MUTEX(clocks_sem); -static DEFINE_SPINLOCK(clockfw_lock); -static void propagate_rate(struct clk * clk); -/* UART clock function */ -static int set_uart_rate(struct clk * clk, unsigned long rate); -/* External clock (MCLK & BCLK) functions */ -static int set_ext_clk_rate(struct clk * clk, unsigned long rate); -static long round_ext_clk_rate(struct clk * clk, unsigned long rate); -static void init_ext_clk(struct clk * clk); -/* MPU virtual clock functions */ -static int select_table_rate(struct clk * clk, unsigned long rate); -static long round_to_table_rate(struct clk * clk, unsigned long rate); -void clk_setdpll(__u16, __u16); - -static struct mpu_rate rate_table[] = { - /* MPU MHz, xtal MHz, dpll1 MHz, CKCTL, DPLL_CTL - * armdiv, dspdiv, dspmmu, tcdiv, perdiv, lcddiv - */ -#if defined(CONFIG_OMAP_ARM_216MHZ) - { 216000000, 12000000, 216000000, 0x050d, 0x2910 }, /* 1/1/2/2/2/8 */ -#endif -#if defined(CONFIG_OMAP_ARM_195MHZ) - { 195000000, 13000000, 195000000, 0x050e, 0x2790 }, /* 1/1/2/2/4/8 */ -#endif -#if defined(CONFIG_OMAP_ARM_192MHZ) - { 192000000, 19200000, 192000000, 0x050f, 0x2510 }, /* 1/1/2/2/8/8 */ - { 192000000, 12000000, 192000000, 0x050f, 0x2810 }, /* 1/1/2/2/8/8 */ - { 96000000, 12000000, 192000000, 0x055f, 0x2810 }, /* 2/2/2/2/8/8 */ - { 48000000, 12000000, 192000000, 0x0baf, 0x2810 }, /* 4/8/4/4/8/8 */ - { 24000000, 12000000, 192000000, 0x0fff, 0x2810 }, /* 8/8/8/8/8/8 */ -#endif -#if defined(CONFIG_OMAP_ARM_182MHZ) - { 182000000, 13000000, 182000000, 0x050e, 0x2710 }, /* 1/1/2/2/4/8 */ -#endif -#if defined(CONFIG_OMAP_ARM_168MHZ) - { 168000000, 12000000, 168000000, 0x010f, 0x2710 }, /* 1/1/1/2/8/8 */ -#endif -#if defined(CONFIG_OMAP_ARM_150MHZ) - { 150000000, 12000000, 150000000, 0x010a, 0x2cb0 }, /* 1/1/1/2/4/4 */ -#endif -#if defined(CONFIG_OMAP_ARM_120MHZ) - { 120000000, 12000000, 120000000, 0x010a, 0x2510 }, /* 1/1/1/2/4/4 */ -#endif -#if defined(CONFIG_OMAP_ARM_96MHZ) - { 96000000, 12000000, 96000000, 0x0005, 0x2410 }, /* 1/1/1/1/2/2 */ -#endif -#if defined(CONFIG_OMAP_ARM_60MHZ) - { 60000000, 12000000, 60000000, 0x0005, 0x2290 }, /* 1/1/1/1/2/2 */ -#endif -#if defined(CONFIG_OMAP_ARM_30MHZ) - { 30000000, 12000000, 60000000, 0x0555, 0x2290 }, /* 2/2/2/2/2/2 */ -#endif - { 0, 0, 0, 0, 0 }, -}; - - -static void ckctl_recalc(struct clk * clk); -int __clk_enable(struct clk *clk); -void __clk_disable(struct clk *clk); -void __clk_unuse(struct clk *clk); -int __clk_use(struct clk *clk); - - -static void followparent_recalc(struct clk * clk) -{ - clk->rate = clk->parent->rate; -} - - -static void watchdog_recalc(struct clk * clk) -{ - clk->rate = clk->parent->rate / 14; -} - -static void uart_recalc(struct clk * clk) -{ - unsigned int val = omap_readl(clk->enable_reg); - if (val & clk->enable_bit) - clk->rate = 48000000; - else - clk->rate = 12000000; -} - -static struct clk ck_ref = { - .name = "ck_ref", - .rate = 12000000, - .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | - ALWAYS_ENABLED, -}; - -static struct clk ck_dpll1 = { - .name = "ck_dpll1", - .parent = &ck_ref, - .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | - RATE_PROPAGATES | ALWAYS_ENABLED, -}; - -static struct clk ck_dpll1out = { - .name = "ck_dpll1out", - .parent = &ck_dpll1, - .flags = CLOCK_IN_OMAP16XX, - .enable_reg = ARM_IDLECT2, - .enable_bit = EN_CKOUT_ARM, - .recalc = &followparent_recalc, -}; - -static struct clk arm_ck = { - .name = "arm_ck", - .parent = &ck_dpll1, - .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | - RATE_CKCTL | RATE_PROPAGATES | ALWAYS_ENABLED, - .rate_offset = CKCTL_ARMDIV_OFFSET, - .recalc = &ckctl_recalc, -}; - -static struct clk armper_ck = { - .name = "armper_ck", - .parent = &ck_dpll1, - .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | - RATE_CKCTL, - .enable_reg = ARM_IDLECT2, - .enable_bit = EN_PERCK, - .rate_offset = CKCTL_PERDIV_OFFSET, - .recalc = &ckctl_recalc, -}; - -static struct clk arm_gpio_ck = { - .name = "arm_gpio_ck", - .parent = &ck_dpll1, - .flags = CLOCK_IN_OMAP1510, - .enable_reg = ARM_IDLECT2, - .enable_bit = EN_GPIOCK, - .recalc = &followparent_recalc, -}; - -static struct clk armxor_ck = { - .name = "armxor_ck", - .parent = &ck_ref, - .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX, - .enable_reg = ARM_IDLECT2, - .enable_bit = EN_XORPCK, - .recalc = &followparent_recalc, -}; - -static struct clk armtim_ck = { - .name = "armtim_ck", - .parent = &ck_ref, - .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX, - .enable_reg = ARM_IDLECT2, - .enable_bit = EN_TIMCK, - .recalc = &followparent_recalc, -}; - -static struct clk armwdt_ck = { - .name = "armwdt_ck", - .parent = &ck_ref, - .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX, - .enable_reg = ARM_IDLECT2, - .enable_bit = EN_WDTCK, - .recalc = &watchdog_recalc, -}; - -static struct clk arminth_ck16xx = { - .name = "arminth_ck", - .parent = &arm_ck, - .flags = CLOCK_IN_OMAP16XX | ALWAYS_ENABLED, - .recalc = &followparent_recalc, - /* Note: On 16xx the frequency can be divided by 2 by programming - * ARM_CKCTL:ARM_INTHCK_SEL(14) to 1 - * - * 1510 version is in TC clocks. - */ -}; - -static struct clk dsp_ck = { - .name = "dsp_ck", - .parent = &ck_dpll1, - .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | - RATE_CKCTL, - .enable_reg = ARM_CKCTL, - .enable_bit = EN_DSPCK, - .rate_offset = CKCTL_DSPDIV_OFFSET, - .recalc = &ckctl_recalc, -}; - -static struct clk dspmmu_ck = { - .name = "dspmmu_ck", - .parent = &ck_dpll1, - .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | - RATE_CKCTL | ALWAYS_ENABLED, - .rate_offset = CKCTL_DSPMMUDIV_OFFSET, - .recalc = &ckctl_recalc, -}; - -static struct clk dspper_ck = { - .name = "dspper_ck", - .parent = &ck_dpll1, - .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | - RATE_CKCTL | DSP_DOMAIN_CLOCK | VIRTUAL_IO_ADDRESS, - .enable_reg = DSP_IDLECT2, - .enable_bit = EN_PERCK, - .rate_offset = CKCTL_PERDIV_OFFSET, - .recalc = &followparent_recalc, - //.recalc = &ckctl_recalc, -}; - -static struct clk dspxor_ck = { - .name = "dspxor_ck", - .parent = &ck_ref, - .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | - DSP_DOMAIN_CLOCK | VIRTUAL_IO_ADDRESS, - .enable_reg = DSP_IDLECT2, - .enable_bit = EN_XORPCK, - .recalc = &followparent_recalc, -}; - -static struct clk dsptim_ck = { - .name = "dsptim_ck", - .parent = &ck_ref, - .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | - DSP_DOMAIN_CLOCK | VIRTUAL_IO_ADDRESS, - .enable_reg = DSP_IDLECT2, - .enable_bit = EN_DSPTIMCK, - .recalc = &followparent_recalc, -}; - -static struct clk tc_ck = { - .name = "tc_ck", - .parent = &ck_dpll1, - .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | CLOCK_IN_OMAP730 | - RATE_CKCTL | RATE_PROPAGATES | ALWAYS_ENABLED, - .rate_offset = CKCTL_TCDIV_OFFSET, - .recalc = &ckctl_recalc, -}; - -static struct clk arminth_ck1510 = { - .name = "arminth_ck", - .parent = &tc_ck, - .flags = CLOCK_IN_OMAP1510 | ALWAYS_ENABLED, - .recalc = &followparent_recalc, - /* Note: On 1510 the frequency follows TC_CK - * - * 16xx version is in MPU clocks. - */ -}; - -static struct clk tipb_ck = { - .name = "tibp_ck", - .parent = &tc_ck, - .flags = CLOCK_IN_OMAP1510 | ALWAYS_ENABLED, - .recalc = &followparent_recalc, -}; - -static struct clk l3_ocpi_ck = { - .name = "l3_ocpi_ck", - .parent = &tc_ck, - .flags = CLOCK_IN_OMAP16XX, - .enable_reg = ARM_IDLECT3, - .enable_bit = EN_OCPI_CK, - .recalc = &followparent_recalc, -}; +DEFINE_SPINLOCK(clockfw_lock); -static struct clk tc1_ck = { - .name = "tc1_ck", - .parent = &tc_ck, - .flags = CLOCK_IN_OMAP16XX, - .enable_reg = ARM_IDLECT3, - .enable_bit = EN_TC1_CK, - .recalc = &followparent_recalc, -}; +static struct clk_functions *arch_clock; -static struct clk tc2_ck = { - .name = "tc2_ck", - .parent = &tc_ck, - .flags = CLOCK_IN_OMAP16XX, - .enable_reg = ARM_IDLECT3, - .enable_bit = EN_TC2_CK, - .recalc = &followparent_recalc, -}; +/*------------------------------------------------------------------------- + * Standard clock functions defined in asm/hardware/clock.h + *-------------------------------------------------------------------------*/ -static struct clk dma_ck = { - .name = "dma_ck", - .parent = &tc_ck, - .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | - ALWAYS_ENABLED, - .recalc = &followparent_recalc, -}; - -static struct clk dma_lcdfree_ck = { - .name = "dma_lcdfree_ck", - .parent = &tc_ck, - .flags = CLOCK_IN_OMAP16XX | ALWAYS_ENABLED, - .recalc = &followparent_recalc, -}; - -static struct clk api_ck = { - .name = "api_ck", - .parent = &tc_ck, - .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX, - .enable_reg = ARM_IDLECT2, - .enable_bit = EN_APICK, - .recalc = &followparent_recalc, -}; - -static struct clk lb_ck = { - .name = "lb_ck", - .parent = &tc_ck, - .flags = CLOCK_IN_OMAP1510, - .enable_reg = ARM_IDLECT2, - .enable_bit = EN_LBCK, - .recalc = &followparent_recalc, -}; - -static struct clk rhea1_ck = { - .name = "rhea1_ck", - .parent = &tc_ck, - .flags = CLOCK_IN_OMAP16XX | ALWAYS_ENABLED, - .recalc = &followparent_recalc, -}; - -static struct clk rhea2_ck = { - .name = "rhea2_ck", - .parent = &tc_ck, - .flags = CLOCK_IN_OMAP16XX | ALWAYS_ENABLED, - .recalc = &followparent_recalc, -}; - -static struct clk lcd_ck = { - .name = "lcd_ck", - .parent = &ck_dpll1, - .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | CLOCK_IN_OMAP730 | - RATE_CKCTL, - .enable_reg = ARM_IDLECT2, - .enable_bit = EN_LCDCK, - .rate_offset = CKCTL_LCDDIV_OFFSET, - .recalc = &ckctl_recalc, -}; - -static struct clk uart1_1510 = { - .name = "uart1_ck", - /* Direct from ULPD, no parent */ - .rate = 12000000, - .flags = CLOCK_IN_OMAP1510 | ENABLE_REG_32BIT | ALWAYS_ENABLED, - .enable_reg = MOD_CONF_CTRL_0, - .enable_bit = 29, /* Chooses between 12MHz and 48MHz */ - .set_rate = &set_uart_rate, - .recalc = &uart_recalc, -}; - -static struct clk uart1_16xx = { - .name = "uart1_ck", - /* Direct from ULPD, no parent */ - .rate = 48000000, - .flags = CLOCK_IN_OMAP16XX | RATE_FIXED | ENABLE_REG_32BIT, - .enable_reg = MOD_CONF_CTRL_0, - .enable_bit = 29, -}; - -static struct clk uart2_ck = { - .name = "uart2_ck", - /* Direct from ULPD, no parent */ - .rate = 12000000, - .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | ENABLE_REG_32BIT | - ALWAYS_ENABLED, - .enable_reg = MOD_CONF_CTRL_0, - .enable_bit = 30, /* Chooses between 12MHz and 48MHz */ - .set_rate = &set_uart_rate, - .recalc = &uart_recalc, -}; - -static struct clk uart3_1510 = { - .name = "uart3_ck", - /* Direct from ULPD, no parent */ - .rate = 12000000, - .flags = CLOCK_IN_OMAP1510 | ENABLE_REG_32BIT | ALWAYS_ENABLED, - .enable_reg = MOD_CONF_CTRL_0, - .enable_bit = 31, /* Chooses between 12MHz and 48MHz */ - .set_rate = &set_uart_rate, - .recalc = &uart_recalc, -}; - -static struct clk uart3_16xx = { - .name = "uart3_ck", - /* Direct from ULPD, no parent */ - .rate = 48000000, - .flags = CLOCK_IN_OMAP16XX | RATE_FIXED | ENABLE_REG_32BIT, - .enable_reg = MOD_CONF_CTRL_0, - .enable_bit = 31, -}; - -static struct clk usb_clko = { /* 6 MHz output on W4_USB_CLKO */ - .name = "usb_clko", - /* Direct from ULPD, no parent */ - .rate = 6000000, - .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | - RATE_FIXED | ENABLE_REG_32BIT, - .enable_reg = ULPD_CLOCK_CTRL, - .enable_bit = USB_MCLK_EN_BIT, -}; - -static struct clk usb_hhc_ck1510 = { - .name = "usb_hhc_ck", - /* Direct from ULPD, no parent */ - .rate = 48000000, /* Actually 2 clocks, 12MHz and 48MHz */ - .flags = CLOCK_IN_OMAP1510 | - RATE_FIXED | ENABLE_REG_32BIT, - .enable_reg = MOD_CONF_CTRL_0, - .enable_bit = USB_HOST_HHC_UHOST_EN, -}; - -static struct clk usb_hhc_ck16xx = { - .name = "usb_hhc_ck", - /* Direct from ULPD, no parent */ - .rate = 48000000, - /* OTG_SYSCON_2.OTG_PADEN == 0 (not 1510-compatible) */ - .flags = CLOCK_IN_OMAP16XX | - RATE_FIXED | ENABLE_REG_32BIT, - .enable_reg = OTG_BASE + 0x08 /* OTG_SYSCON_2 */, - .enable_bit = 8 /* UHOST_EN */, -}; - -static struct clk usb_dc_ck = { - .name = "usb_dc_ck", - /* Direct from ULPD, no parent */ - .rate = 48000000, - .flags = CLOCK_IN_OMAP16XX | RATE_FIXED, - .enable_reg = SOFT_REQ_REG, - .enable_bit = 4, -}; - -static struct clk mclk_1510 = { - .name = "mclk", - /* Direct from ULPD, no parent. May be enabled by ext hardware. */ - .rate = 12000000, - .flags = CLOCK_IN_OMAP1510 | RATE_FIXED, -}; - -static struct clk mclk_16xx = { - .name = "mclk", - /* Direct from ULPD, no parent. May be enabled by ext hardware. */ - .flags = CLOCK_IN_OMAP16XX, - .enable_reg = COM_CLK_DIV_CTRL_SEL, - .enable_bit = COM_ULPD_PLL_CLK_REQ, - .set_rate = &set_ext_clk_rate, - .round_rate = &round_ext_clk_rate, - .init = &init_ext_clk, -}; - -static struct clk bclk_1510 = { - .name = "bclk", - /* Direct from ULPD, no parent. May be enabled by ext hardware. */ - .rate = 12000000, - .flags = CLOCK_IN_OMAP1510 | RATE_FIXED, -}; - -static struct clk bclk_16xx = { - .name = "bclk", - /* Direct from ULPD, no parent. May be enabled by ext hardware. */ - .flags = CLOCK_IN_OMAP16XX, - .enable_reg = SWD_CLK_DIV_CTRL_SEL, - .enable_bit = SWD_ULPD_PLL_CLK_REQ, - .set_rate = &set_ext_clk_rate, - .round_rate = &round_ext_clk_rate, - .init = &init_ext_clk, -}; - -static struct clk mmc1_ck = { - .name = "mmc1_ck", - /* Functional clock is direct from ULPD, interface clock is ARMPER */ - .parent = &armper_ck, - .rate = 48000000, - .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | - RATE_FIXED | ENABLE_REG_32BIT, - .enable_reg = MOD_CONF_CTRL_0, - .enable_bit = 23, -}; - -static struct clk mmc2_ck = { - .name = "mmc2_ck", - /* Functional clock is direct from ULPD, interface clock is ARMPER */ - .parent = &armper_ck, - .rate = 48000000, - .flags = CLOCK_IN_OMAP16XX | - RATE_FIXED | ENABLE_REG_32BIT, - .enable_reg = MOD_CONF_CTRL_0, - .enable_bit = 20, -}; - -static struct clk virtual_ck_mpu = { - .name = "mpu", - .flags = CLOCK_IN_OMAP1510 | CLOCK_IN_OMAP16XX | - VIRTUAL_CLOCK | ALWAYS_ENABLED, - .parent = &arm_ck, /* Is smarter alias for */ - .recalc = &followparent_recalc, - .set_rate = &select_table_rate, - .round_rate = &round_to_table_rate, -}; - - -static struct clk * onchip_clks[] = { - /* non-ULPD clocks */ - &ck_ref, - &ck_dpll1, - /* CK_GEN1 clocks */ - &ck_dpll1out, - &arm_ck, - &armper_ck, - &arm_gpio_ck, - &armxor_ck, - &armtim_ck, - &armwdt_ck, - &arminth_ck1510, &arminth_ck16xx, - /* CK_GEN2 clocks */ - &dsp_ck, - &dspmmu_ck, - &dspper_ck, - &dspxor_ck, - &dsptim_ck, - /* CK_GEN3 clocks */ - &tc_ck, - &tipb_ck, - &l3_ocpi_ck, - &tc1_ck, - &tc2_ck, - &dma_ck, - &dma_lcdfree_ck, - &api_ck, - &lb_ck, - &rhea1_ck, - &rhea2_ck, - &lcd_ck, - /* ULPD clocks */ - &uart1_1510, - &uart1_16xx, - &uart2_ck, - &uart3_1510, - &uart3_16xx, - &usb_clko, - &usb_hhc_ck1510, &usb_hhc_ck16xx, - &usb_dc_ck, - &mclk_1510, &mclk_16xx, - &bclk_1510, &bclk_16xx, - &mmc1_ck, - &mmc2_ck, - /* Virtual clocks */ - &virtual_ck_mpu, -}; - -struct clk *clk_get(struct device *dev, const char *id) +struct clk * clk_get(struct device *dev, const char *id) { struct clk *p, *clk = ERR_PTR(-ENOENT); @@ -590,534 +53,200 @@ struct clk *clk_get(struct device *dev, const char *id) } EXPORT_SYMBOL(clk_get); - -void clk_put(struct clk *clk) -{ - if (clk && !IS_ERR(clk)) - module_put(clk->owner); -} -EXPORT_SYMBOL(clk_put); - - -int __clk_enable(struct clk *clk) -{ - __u16 regval16; - __u32 regval32; - - if (clk->flags & ALWAYS_ENABLED) - return 0; - - if (unlikely(clk->enable_reg == 0)) { - printk(KERN_ERR "clock.c: Enable for %s without enable code\n", - clk->name); - return 0; - } - - if (clk->flags & DSP_DOMAIN_CLOCK) { - __clk_use(&api_ck); - } - - if (clk->flags & ENABLE_REG_32BIT) { - if (clk->flags & VIRTUAL_IO_ADDRESS) { - regval32 = __raw_readl(clk->enable_reg); - regval32 |= (1 << clk->enable_bit); - __raw_writel(regval32, clk->enable_reg); - } else { - regval32 = omap_readl(clk->enable_reg); - regval32 |= (1 << clk->enable_bit); - omap_writel(regval32, clk->enable_reg); - } - } else { - if (clk->flags & VIRTUAL_IO_ADDRESS) { - regval16 = __raw_readw(clk->enable_reg); - regval16 |= (1 << clk->enable_bit); - __raw_writew(regval16, clk->enable_reg); - } else { - regval16 = omap_readw(clk->enable_reg); - regval16 |= (1 << clk->enable_bit); - omap_writew(regval16, clk->enable_reg); - } - } - - if (clk->flags & DSP_DOMAIN_CLOCK) { - __clk_unuse(&api_ck); - } - - return 0; -} - - -void __clk_disable(struct clk *clk) -{ - __u16 regval16; - __u32 regval32; - - if (clk->enable_reg == 0) - return; - - if (clk->flags & DSP_DOMAIN_CLOCK) { - __clk_use(&api_ck); - } - - if (clk->flags & ENABLE_REG_32BIT) { - if (clk->flags & VIRTUAL_IO_ADDRESS) { - regval32 = __raw_readl(clk->enable_reg); - regval32 &= ~(1 << clk->enable_bit); - __raw_writel(regval32, clk->enable_reg); - } else { - regval32 = omap_readl(clk->enable_reg); - regval32 &= ~(1 << clk->enable_bit); - omap_writel(regval32, clk->enable_reg); - } - } else { - if (clk->flags & VIRTUAL_IO_ADDRESS) { - regval16 = __raw_readw(clk->enable_reg); - regval16 &= ~(1 << clk->enable_bit); - __raw_writew(regval16, clk->enable_reg); - } else { - regval16 = omap_readw(clk->enable_reg); - regval16 &= ~(1 << clk->enable_bit); - omap_writew(regval16, clk->enable_reg); - } - } - - if (clk->flags & DSP_DOMAIN_CLOCK) { - __clk_unuse(&api_ck); - } -} - - -void __clk_unuse(struct clk *clk) -{ - if (clk->usecount > 0 && !(--clk->usecount)) { - __clk_disable(clk); - if (likely(clk->parent)) - __clk_unuse(clk->parent); - } -} - - -int __clk_use(struct clk *clk) -{ - int ret = 0; - if (clk->usecount++ == 0) { - if (likely(clk->parent)) - ret = __clk_use(clk->parent); - - if (unlikely(ret != 0)) { - clk->usecount--; - return ret; - } - - ret = __clk_enable(clk); - - if (unlikely(ret != 0) && clk->parent) { - __clk_unuse(clk->parent); - clk->usecount--; - } - } - - return ret; -} - - int clk_enable(struct clk *clk) { unsigned long flags; - int ret; + int ret = 0; spin_lock_irqsave(&clockfw_lock, flags); - ret = __clk_enable(clk); + if (clk->enable) + ret = clk->enable(clk); + else if (arch_clock->clk_enable) + ret = arch_clock->clk_enable(clk); + else + printk(KERN_ERR "Could not enable clock %s\n", clk->name); spin_unlock_irqrestore(&clockfw_lock, flags); + return ret; } EXPORT_SYMBOL(clk_enable); - void clk_disable(struct clk *clk) { unsigned long flags; spin_lock_irqsave(&clockfw_lock, flags); - __clk_disable(clk); + if (clk->disable) + clk->disable(clk); + else if (arch_clock->clk_disable) + arch_clock->clk_disable(clk); + else + printk(KERN_ERR "Could not disable clock %s\n", clk->name); spin_unlock_irqrestore(&clockfw_lock, flags); } EXPORT_SYMBOL(clk_disable); - int clk_use(struct clk *clk) { unsigned long flags; int ret = 0; spin_lock_irqsave(&clockfw_lock, flags); - ret = __clk_use(clk); + if (arch_clock->clk_use) + ret = arch_clock->clk_use(clk); spin_unlock_irqrestore(&clockfw_lock, flags); + return ret; } EXPORT_SYMBOL(clk_use); - void clk_unuse(struct clk *clk) { unsigned long flags; spin_lock_irqsave(&clockfw_lock, flags); - __clk_unuse(clk); + if (arch_clock->clk_unuse) + arch_clock->clk_unuse(clk); spin_unlock_irqrestore(&clockfw_lock, flags); } EXPORT_SYMBOL(clk_unuse); - int clk_get_usecount(struct clk *clk) { - return clk->usecount; -} -EXPORT_SYMBOL(clk_get_usecount); - - -unsigned long clk_get_rate(struct clk *clk) -{ - return clk->rate; -} -EXPORT_SYMBOL(clk_get_rate); - - -static __u16 verify_ckctl_value(__u16 newval) -{ - /* This function checks for following limitations set - * by the hardware (all conditions must be true): - * DSPMMU_CK == DSP_CK or DSPMMU_CK == DSP_CK/2 - * ARM_CK >= TC_CK - * DSP_CK >= TC_CK - * DSPMMU_CK >= TC_CK - * - * In addition following rules are enforced: - * LCD_CK <= TC_CK - * ARMPER_CK <= TC_CK - * - * However, maximum frequencies are not checked for! - */ - __u8 per_exp; - __u8 lcd_exp; - __u8 arm_exp; - __u8 dsp_exp; - __u8 tc_exp; - __u8 dspmmu_exp; - - per_exp = (newval >> CKCTL_PERDIV_OFFSET) & 3; - lcd_exp = (newval >> CKCTL_LCDDIV_OFFSET) & 3; - arm_exp = (newval >> CKCTL_ARMDIV_OFFSET) & 3; - dsp_exp = (newval >> CKCTL_DSPDIV_OFFSET) & 3; - tc_exp = (newval >> CKCTL_TCDIV_OFFSET) & 3; - dspmmu_exp = (newval >> CKCTL_DSPMMUDIV_OFFSET) & 3; - - if (dspmmu_exp < dsp_exp) - dspmmu_exp = dsp_exp; - if (dspmmu_exp > dsp_exp+1) - dspmmu_exp = dsp_exp+1; - if (tc_exp < arm_exp) - tc_exp = arm_exp; - if (tc_exp < dspmmu_exp) - tc_exp = dspmmu_exp; - if (tc_exp > lcd_exp) - lcd_exp = tc_exp; - if (tc_exp > per_exp) - per_exp = tc_exp; + unsigned long flags; + int ret = 0; - newval &= 0xf000; - newval |= per_exp << CKCTL_PERDIV_OFFSET; - newval |= lcd_exp << CKCTL_LCDDIV_OFFSET; - newval |= arm_exp << CKCTL_ARMDIV_OFFSET; - newval |= dsp_exp << CKCTL_DSPDIV_OFFSET; - newval |= tc_exp << CKCTL_TCDIV_OFFSET; - newval |= dspmmu_exp << CKCTL_DSPMMUDIV_OFFSET; + spin_lock_irqsave(&clockfw_lock, flags); + ret = clk->usecount; + spin_unlock_irqrestore(&clockfw_lock, flags); - return newval; + return ret; } +EXPORT_SYMBOL(clk_get_usecount); - -static int calc_dsor_exp(struct clk *clk, unsigned long rate) +unsigned long clk_get_rate(struct clk *clk) { - /* Note: If target frequency is too low, this function will return 4, - * which is invalid value. Caller must check for this value and act - * accordingly. - * - * Note: This function does not check for following limitations set - * by the hardware (all conditions must be true): - * DSPMMU_CK == DSP_CK or DSPMMU_CK == DSP_CK/2 - * ARM_CK >= TC_CK - * DSP_CK >= TC_CK - * DSPMMU_CK >= TC_CK - */ - unsigned long realrate; - struct clk * parent; - unsigned dsor_exp; - - if (unlikely(!(clk->flags & RATE_CKCTL))) - return -EINVAL; - - parent = clk->parent; - if (unlikely(parent == 0)) - return -EIO; - - realrate = parent->rate; - for (dsor_exp=0; dsor_exp<4; dsor_exp++) { - if (realrate <= rate) - break; + unsigned long flags; + unsigned long ret = 0; - realrate /= 2; - } + spin_lock_irqsave(&clockfw_lock, flags); + ret = clk->rate; + spin_unlock_irqrestore(&clockfw_lock, flags); - return dsor_exp; + return ret; } +EXPORT_SYMBOL(clk_get_rate); - -static void ckctl_recalc(struct clk * clk) +void clk_put(struct clk *clk) { - int dsor; - - /* Calculate divisor encoded as 2-bit exponent */ - if (clk->flags & DSP_DOMAIN_CLOCK) { - /* The clock control bits are in DSP domain, - * so api_ck is needed for access. - * Note that DSP_CKCTL virt addr = phys addr, so - * we must use __raw_readw() instead of omap_readw(). - */ - __clk_use(&api_ck); - dsor = 1 << (3 & (__raw_readw(DSP_CKCTL) >> clk->rate_offset)); - __clk_unuse(&api_ck); - } else { - dsor = 1 << (3 & (omap_readw(ARM_CKCTL) >> clk->rate_offset)); - } - if (unlikely(clk->rate == clk->parent->rate / dsor)) - return; /* No change, quick exit */ - clk->rate = clk->parent->rate / dsor; - - if (unlikely(clk->flags & RATE_PROPAGATES)) - propagate_rate(clk); + if (clk && !IS_ERR(clk)) + module_put(clk->owner); } +EXPORT_SYMBOL(clk_put); +/*------------------------------------------------------------------------- + * Optional clock functions defined in asm/hardware/clock.h + *-------------------------------------------------------------------------*/ long clk_round_rate(struct clk *clk, unsigned long rate) { - int dsor_exp; - - if (clk->flags & RATE_FIXED) - return clk->rate; - - if (clk->flags & RATE_CKCTL) { - dsor_exp = calc_dsor_exp(clk, rate); - if (dsor_exp < 0) - return dsor_exp; - if (dsor_exp > 3) - dsor_exp = 3; - return clk->parent->rate / (1 << dsor_exp); - } + unsigned long flags; + long ret = 0; - if(clk->round_rate != 0) - return clk->round_rate(clk, rate); + spin_lock_irqsave(&clockfw_lock, flags); + if (arch_clock->clk_round_rate) + ret = arch_clock->clk_round_rate(clk, rate); + spin_unlock_irqrestore(&clockfw_lock, flags); - return clk->rate; + return ret; } EXPORT_SYMBOL(clk_round_rate); - -static void propagate_rate(struct clk * clk) -{ - struct clk ** clkp; - - for (clkp = onchip_clks; clkp < onchip_clks+ARRAY_SIZE(onchip_clks); clkp++) { - if (likely((*clkp)->parent != clk)) continue; - if (likely((*clkp)->recalc)) - (*clkp)->recalc(*clkp); - } -} - - -static int select_table_rate(struct clk * clk, unsigned long rate) +int clk_set_rate(struct clk *clk, unsigned long rate) { - /* Find the highest supported frequency <= rate and switch to it */ - struct mpu_rate * ptr; - - if (clk != &virtual_ck_mpu) - return -EINVAL; - - for (ptr = rate_table; ptr->rate; ptr++) { - if (ptr->xtal != ck_ref.rate) - continue; - - /* DPLL1 cannot be reprogrammed without risking system crash */ - if (likely(ck_dpll1.rate!=0) && ptr->pll_rate != ck_dpll1.rate) - continue; - - /* Can check only after xtal frequency check */ - if (ptr->rate <= rate) - break; - } - - if (!ptr->rate) - return -EINVAL; + unsigned long flags; + int ret = 0; - /* - * In most cases we should not need to reprogram DPLL. - * Reprogramming the DPLL is tricky, it must be done from SRAM. - */ - omap_sram_reprogram_clock(ptr->dpllctl_val, ptr->ckctl_val); + spin_lock_irqsave(&clockfw_lock, flags); + if (arch_clock->clk_set_rate) + ret = arch_clock->clk_set_rate(clk, rate); + spin_unlock_irqrestore(&clockfw_lock, flags); - ck_dpll1.rate = ptr->pll_rate; - propagate_rate(&ck_dpll1); - return 0; + return ret; } +EXPORT_SYMBOL(clk_set_rate); - -static long round_to_table_rate(struct clk * clk, unsigned long rate) +int clk_set_parent(struct clk *clk, struct clk *parent) { - /* Find the highest supported frequency <= rate */ - struct mpu_rate * ptr; - long highest_rate; - - if (clk != &virtual_ck_mpu) - return -EINVAL; - - highest_rate = -EINVAL; - - for (ptr = rate_table; ptr->rate; ptr++) { - if (ptr->xtal != ck_ref.rate) - continue; - - highest_rate = ptr->rate; + unsigned long flags; + int ret = 0; - /* Can check only after xtal frequency check */ - if (ptr->rate <= rate) - break; - } + spin_lock_irqsave(&clockfw_lock, flags); + if (arch_clock->clk_set_parent) + ret = arch_clock->clk_set_parent(clk, parent); + spin_unlock_irqrestore(&clockfw_lock, flags); - return highest_rate; + return ret; } +EXPORT_SYMBOL(clk_set_parent); - -int clk_set_rate(struct clk *clk, unsigned long rate) +struct clk *clk_get_parent(struct clk *clk) { - int ret = -EINVAL; - int dsor_exp; - __u16 regval; - unsigned long flags; - - if (clk->flags & RATE_CKCTL) { - dsor_exp = calc_dsor_exp(clk, rate); - if (dsor_exp > 3) - dsor_exp = -EINVAL; - if (dsor_exp < 0) - return dsor_exp; - - spin_lock_irqsave(&clockfw_lock, flags); - regval = omap_readw(ARM_CKCTL); - regval &= ~(3 << clk->rate_offset); - regval |= dsor_exp << clk->rate_offset; - regval = verify_ckctl_value(regval); - omap_writew(regval, ARM_CKCTL); - clk->rate = clk->parent->rate / (1 << dsor_exp); - spin_unlock_irqrestore(&clockfw_lock, flags); - ret = 0; - } else if(clk->set_rate != 0) { - spin_lock_irqsave(&clockfw_lock, flags); - ret = clk->set_rate(clk, rate); - spin_unlock_irqrestore(&clockfw_lock, flags); - } + unsigned long flags; + struct clk * ret = NULL; - if (unlikely(ret == 0 && (clk->flags & RATE_PROPAGATES))) - propagate_rate(clk); + spin_lock_irqsave(&clockfw_lock, flags); + if (arch_clock->clk_get_parent) + ret = arch_clock->clk_get_parent(clk); + spin_unlock_irqrestore(&clockfw_lock, flags); return ret; } -EXPORT_SYMBOL(clk_set_rate); +EXPORT_SYMBOL(clk_get_parent); +/*------------------------------------------------------------------------- + * OMAP specific clock functions shared between omap1 and omap2 + *-------------------------------------------------------------------------*/ -static unsigned calc_ext_dsor(unsigned long rate) -{ - unsigned dsor; +unsigned int __initdata mpurate; - /* MCLK and BCLK divisor selection is not linear: - * freq = 96MHz / dsor - * - * RATIO_SEL range: dsor <-> RATIO_SEL - * 0..6: (RATIO_SEL+2) <-> (dsor-2) - * 6..48: (8+(RATIO_SEL-6)*2) <-> ((dsor-8)/2+6) - * Minimum dsor is 2 and maximum is 96. Odd divisors starting from 9 - * can not be used. - */ - for (dsor = 2; dsor < 96; ++dsor) { - if ((dsor & 1) && dsor > 8) - continue; - if (rate >= 96000000 / dsor) - break; - } - return dsor; -} - -/* Only needed on 1510 */ -static int set_uart_rate(struct clk * clk, unsigned long rate) -{ - unsigned int val; - - val = omap_readl(clk->enable_reg); - if (rate == 12000000) - val &= ~(1 << clk->enable_bit); - else if (rate == 48000000) - val |= (1 << clk->enable_bit); - else - return -EINVAL; - omap_writel(val, clk->enable_reg); - clk->rate = rate; - - return 0; -} - -static int set_ext_clk_rate(struct clk * clk, unsigned long rate) +/* + * By default we use the rate set by the bootloader. + * You can override this with mpurate= cmdline option. + */ +static int __init omap_clk_setup(char *str) { - unsigned dsor; - __u16 ratio_bits; + get_option(&str, &mpurate); - dsor = calc_ext_dsor(rate); - clk->rate = 96000000 / dsor; - if (dsor > 8) - ratio_bits = ((dsor - 8) / 2 + 6) << 2; - else - ratio_bits = (dsor - 2) << 2; + if (!mpurate) + return 1; - ratio_bits |= omap_readw(clk->enable_reg) & ~0xfd; - omap_writew(ratio_bits, clk->enable_reg); + if (mpurate < 1000) + mpurate *= 1000000; - return 0; + return 1; } +__setup("mpurate=", omap_clk_setup); - -static long round_ext_clk_rate(struct clk * clk, unsigned long rate) +/* Used for clocks that always have same value as the parent clock */ +void followparent_recalc(struct clk *clk) { - return 96000000 / calc_ext_dsor(rate); + clk->rate = clk->parent->rate; } - -static void init_ext_clk(struct clk * clk) +/* Propagate rate to children */ +void propagate_rate(struct clk * tclk) { - unsigned dsor; - __u16 ratio_bits; + struct clk *clkp; - /* Determine current rate and ensure clock is based on 96MHz APLL */ - ratio_bits = omap_readw(clk->enable_reg) & ~1; - omap_writew(ratio_bits, clk->enable_reg); - - ratio_bits = (ratio_bits & 0xfc) >> 2; - if (ratio_bits > 6) - dsor = (ratio_bits - 6) * 2 + 8; - else - dsor = ratio_bits + 2; - - clk-> rate = 96000000 / dsor; + list_for_each_entry(clkp, &clocks, node) { + if (likely(clkp->parent != tclk)) + continue; + if (likely((u32)clkp->recalc)) + clkp->recalc(clkp); + } } - int clk_register(struct clk *clk) { down(&clocks_sem); @@ -1125,6 +254,7 @@ int clk_register(struct clk *clk) if (clk->init) clk->init(clk); up(&clocks_sem); + return 0; } EXPORT_SYMBOL(clk_register); @@ -1137,203 +267,38 @@ void clk_unregister(struct clk *clk) } EXPORT_SYMBOL(clk_unregister); -#ifdef CONFIG_OMAP_RESET_CLOCKS -/* - * Resets some clocks that may be left on from bootloader, - * but leaves serial clocks on. See also omap_late_clk_reset(). - */ -static inline void omap_early_clk_reset(void) +void clk_deny_idle(struct clk *clk) { - //omap_writel(0x3 << 29, MOD_CONF_CTRL_0); + unsigned long flags; + + spin_lock_irqsave(&clockfw_lock, flags); + if (arch_clock->clk_deny_idle) + arch_clock->clk_deny_idle(clk); + spin_unlock_irqrestore(&clockfw_lock, flags); } -#else -#define omap_early_clk_reset() {} -#endif +EXPORT_SYMBOL(clk_deny_idle); -int __init clk_init(void) +void clk_allow_idle(struct clk *clk) { - struct clk ** clkp; - const struct omap_clock_config *info; - int crystal_type = 0; /* Default 12 MHz */ - - omap_early_clk_reset(); - - for (clkp = onchip_clks; clkp < onchip_clks+ARRAY_SIZE(onchip_clks); clkp++) { - if (((*clkp)->flags &CLOCK_IN_OMAP1510) && cpu_is_omap1510()) { - clk_register(*clkp); - continue; - } - - if (((*clkp)->flags &CLOCK_IN_OMAP16XX) && cpu_is_omap16xx()) { - clk_register(*clkp); - continue; - } - - if (((*clkp)->flags &CLOCK_IN_OMAP730) && cpu_is_omap730()) { - clk_register(*clkp); - continue; - } - } - - info = omap_get_config(OMAP_TAG_CLOCK, struct omap_clock_config); - if (info != NULL) { - if (!cpu_is_omap1510()) - crystal_type = info->system_clock_type; - } - -#if defined(CONFIG_ARCH_OMAP730) - ck_ref.rate = 13000000; -#elif defined(CONFIG_ARCH_OMAP16XX) - if (crystal_type == 2) - ck_ref.rate = 19200000; -#endif - - printk("Clocks: ARM_SYSST: 0x%04x DPLL_CTL: 0x%04x ARM_CKCTL: 0x%04x\n", - omap_readw(ARM_SYSST), omap_readw(DPLL_CTL), - omap_readw(ARM_CKCTL)); - - /* We want to be in syncronous scalable mode */ - omap_writew(0x1000, ARM_SYSST); - -#ifdef CONFIG_OMAP_CLOCKS_SET_BY_BOOTLOADER - /* Use values set by bootloader. Determine PLL rate and recalculate - * dependent clocks as if kernel had changed PLL or divisors. - */ - { - unsigned pll_ctl_val = omap_readw(DPLL_CTL); - - ck_dpll1.rate = ck_ref.rate; /* Base xtal rate */ - if (pll_ctl_val & 0x10) { - /* PLL enabled, apply multiplier and divisor */ - if (pll_ctl_val & 0xf80) - ck_dpll1.rate *= (pll_ctl_val & 0xf80) >> 7; - ck_dpll1.rate /= ((pll_ctl_val & 0x60) >> 5) + 1; - } else { - /* PLL disabled, apply bypass divisor */ - switch (pll_ctl_val & 0xc) { - case 0: - break; - case 0x4: - ck_dpll1.rate /= 2; - break; - default: - ck_dpll1.rate /= 4; - break; - } - } - } - propagate_rate(&ck_dpll1); -#else - /* Find the highest supported frequency and enable it */ - if (select_table_rate(&virtual_ck_mpu, ~0)) { - printk(KERN_ERR "System frequencies not set. Check your config.\n"); - /* Guess sane values (60MHz) */ - omap_writew(0x2290, DPLL_CTL); - omap_writew(0x1005, ARM_CKCTL); - ck_dpll1.rate = 60000000; - propagate_rate(&ck_dpll1); - } -#endif - /* Cache rates for clocks connected to ck_ref (not dpll1) */ - propagate_rate(&ck_ref); - printk(KERN_INFO "Clocking rate (xtal/DPLL1/MPU): " - "%ld.%01ld/%ld.%01ld/%ld.%01ld MHz\n", - ck_ref.rate / 1000000, (ck_ref.rate / 100000) % 10, - ck_dpll1.rate / 1000000, (ck_dpll1.rate / 100000) % 10, - arm_ck.rate / 1000000, (arm_ck.rate / 100000) % 10); - -#ifdef CONFIG_MACH_OMAP_PERSEUS2 - /* Select slicer output as OMAP input clock */ - omap_writew(omap_readw(OMAP730_PCC_UPLD_CTRL) & ~0x1, OMAP730_PCC_UPLD_CTRL); -#endif - - /* Turn off DSP and ARM_TIMXO. Make sure ARM_INTHCK is not divided */ - omap_writew(omap_readw(ARM_CKCTL) & 0x0fff, ARM_CKCTL); - - /* Put DSP/MPUI into reset until needed */ - omap_writew(0, ARM_RSTCT1); - omap_writew(1, ARM_RSTCT2); - omap_writew(0x400, ARM_IDLECT1); - - /* - * According to OMAP5910 Erratum SYS_DMA_1, bit DMACK_REQ (bit 8) - * of the ARM_IDLECT2 register must be set to zero. The power-on - * default value of this bit is one. - */ - omap_writew(0x0000, ARM_IDLECT2); /* Turn LCD clock off also */ - - /* - * Only enable those clocks we will need, let the drivers - * enable other clocks as necessary - */ - clk_use(&armper_ck); - clk_use(&armxor_ck); - clk_use(&armtim_ck); - - if (cpu_is_omap1510()) - clk_enable(&arm_gpio_ck); + unsigned long flags; - return 0; + spin_lock_irqsave(&clockfw_lock, flags); + if (arch_clock->clk_allow_idle) + arch_clock->clk_allow_idle(clk); + spin_unlock_irqrestore(&clockfw_lock, flags); } +EXPORT_SYMBOL(clk_allow_idle); +/*-------------------------------------------------------------------------*/ -#ifdef CONFIG_OMAP_RESET_CLOCKS - -static int __init omap_late_clk_reset(void) +int __init clk_init(struct clk_functions * custom_clocks) { - /* Turn off all unused clocks */ - struct clk *p; - __u32 regval32; - - /* USB_REQ_EN will be disabled later if necessary (usb_dc_ck) */ - regval32 = omap_readw(SOFT_REQ_REG) & (1 << 4); - omap_writew(regval32, SOFT_REQ_REG); - omap_writew(0, SOFT_REQ_REG2); - - list_for_each_entry(p, &clocks, node) { - if (p->usecount > 0 || (p->flags & ALWAYS_ENABLED) || - p->enable_reg == 0) - continue; - - /* Assume no DSP clocks have been activated by bootloader */ - if (p->flags & DSP_DOMAIN_CLOCK) - continue; - - /* Is the clock already disabled? */ - if (p->flags & ENABLE_REG_32BIT) { - if (p->flags & VIRTUAL_IO_ADDRESS) - regval32 = __raw_readl(p->enable_reg); - else - regval32 = omap_readl(p->enable_reg); - } else { - if (p->flags & VIRTUAL_IO_ADDRESS) - regval32 = __raw_readw(p->enable_reg); - else - regval32 = omap_readw(p->enable_reg); - } - - if ((regval32 & (1 << p->enable_bit)) == 0) - continue; - - /* FIXME: This clock seems to be necessary but no-one - * has asked for its activation. */ - if (p == &tc2_ck // FIX: pm.c (SRAM), CCP, Camera - || p == &ck_dpll1out // FIX: SoSSI, SSR - || p == &arm_gpio_ck // FIX: GPIO code for 1510 - ) { - printk(KERN_INFO "FIXME: Clock \"%s\" seems unused\n", - p->name); - continue; - } - - printk(KERN_INFO "Disabling unused clock \"%s\"... ", p->name); - __clk_disable(p); - printk(" done\n"); + if (!custom_clocks) { + printk(KERN_ERR "No custom clock functions registered\n"); + BUG(); } + arch_clock = custom_clocks; + return 0; } - -late_initcall(omap_late_clk_reset); - -#endif diff --git a/arch/arm/plat-omap/clock.h b/arch/arm/plat-omap/clock.h deleted file mode 100644 index a89e1e8c251..00000000000 --- a/arch/arm/plat-omap/clock.h +++ /dev/null @@ -1,120 +0,0 @@ -/* - * linux/arch/arm/plat-omap/clock.h - * - * Copyright (C) 2004 Nokia corporation - * Written by Tuukka Tikkanen <tuukka.tikkanen@elektrobit.com> - * Based on clocks.h by Tony Lindgren, Gordon McNutt and RidgeRun, Inc - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - */ - -#ifndef __ARCH_ARM_OMAP_CLOCK_H -#define __ARCH_ARM_OMAP_CLOCK_H - -struct module; - -struct clk { - struct list_head node; - struct module *owner; - const char *name; - struct clk *parent; - unsigned long rate; - __s8 usecount; - __u16 flags; - __u32 enable_reg; - __u8 enable_bit; - __u8 rate_offset; - void (*recalc)(struct clk *); - int (*set_rate)(struct clk *, unsigned long); - long (*round_rate)(struct clk *, unsigned long); - void (*init)(struct clk *); -}; - - -struct mpu_rate { - unsigned long rate; - unsigned long xtal; - unsigned long pll_rate; - __u16 ckctl_val; - __u16 dpllctl_val; -}; - - -/* Clock flags */ -#define RATE_CKCTL 1 -#define RATE_FIXED 2 -#define RATE_PROPAGATES 4 -#define VIRTUAL_CLOCK 8 -#define ALWAYS_ENABLED 16 -#define ENABLE_REG_32BIT 32 -#define CLOCK_IN_OMAP16XX 64 -#define CLOCK_IN_OMAP1510 128 -#define CLOCK_IN_OMAP730 256 -#define DSP_DOMAIN_CLOCK 512 -#define VIRTUAL_IO_ADDRESS 1024 - -/* ARM_CKCTL bit shifts */ -#define CKCTL_PERDIV_OFFSET 0 -#define CKCTL_LCDDIV_OFFSET 2 -#define CKCTL_ARMDIV_OFFSET 4 -#define CKCTL_DSPDIV_OFFSET 6 -#define CKCTL_TCDIV_OFFSET 8 -#define CKCTL_DSPMMUDIV_OFFSET 10 -/*#define ARM_TIMXO 12*/ -#define EN_DSPCK 13 -/*#define ARM_INTHCK_SEL 14*/ /* Divide-by-2 for mpu inth_ck */ -/* DSP_CKCTL bit shifts */ -#define CKCTL_DSPPERDIV_OFFSET 0 - -/* ARM_IDLECT1 bit shifts */ -/*#define IDLWDT_ARM 0*/ -/*#define IDLXORP_ARM 1*/ -/*#define IDLPER_ARM 2*/ -/*#define IDLLCD_ARM 3*/ -/*#define IDLLB_ARM 4*/ -/*#define IDLHSAB_ARM 5*/ -/*#define IDLIF_ARM 6*/ -/*#define IDLDPLL_ARM 7*/ -/*#define IDLAPI_ARM 8*/ -/*#define IDLTIM_ARM 9*/ -/*#define SETARM_IDLE 11*/ - -/* ARM_IDLECT2 bit shifts */ -#define EN_WDTCK 0 -#define EN_XORPCK 1 -#define EN_PERCK 2 -#define EN_LCDCK 3 -#define EN_LBCK 4 /* Not on 1610/1710 */ -/*#define EN_HSABCK 5*/ -#define EN_APICK 6 -#define EN_TIMCK 7 -#define DMACK_REQ 8 -#define EN_GPIOCK 9 /* Not on 1610/1710 */ -/*#define EN_LBFREECK 10*/ -#define EN_CKOUT_ARM 11 - -/* ARM_IDLECT3 bit shifts */ -#define EN_OCPI_CK 0 -#define EN_TC1_CK 2 -#define EN_TC2_CK 4 - -/* DSP_IDLECT2 bit shifts (0,1,2 are same as for ARM_IDLECT2) */ -#define EN_DSPTIMCK 5 - -/* Various register defines for clock controls scattered around OMAP chip */ -#define USB_MCLK_EN_BIT 4 /* In ULPD_CLKC_CTRL */ -#define USB_HOST_HHC_UHOST_EN 9 /* In MOD_CONF_CTRL_0 */ -#define SWD_ULPD_PLL_CLK_REQ 1 /* In SWD_CLK_DIV_CTRL_SEL */ -#define COM_ULPD_PLL_CLK_REQ 1 /* In COM_CLK_DIV_CTRL_SEL */ -#define SWD_CLK_DIV_CTRL_SEL 0xfffe0874 -#define COM_CLK_DIV_CTRL_SEL 0xfffe0878 -#define SOFT_REQ_REG 0xfffe0834 -#define SOFT_REQ_REG2 0xfffe0880 - -int clk_register(struct clk *clk); -void clk_unregister(struct clk *clk); -int clk_init(void); - -#endif diff --git a/arch/arm/plat-omap/common.c b/arch/arm/plat-omap/common.c index 02bcc6c1cd1..ccdb452630c 100644 --- a/arch/arm/plat-omap/common.c +++ b/arch/arm/plat-omap/common.c @@ -31,7 +31,7 @@ #include <asm/arch/mux.h> #include <asm/arch/fpga.h> -#include "clock.h" +#include <asm/arch/clock.h> #define NO_LENGTH_CHECK 0xffffffff @@ -117,19 +117,43 @@ EXPORT_SYMBOL(omap_get_var_config); static int __init omap_add_serial_console(void) { - const struct omap_serial_console_config *info; - - info = omap_get_config(OMAP_TAG_SERIAL_CONSOLE, - struct omap_serial_console_config); - if (info != NULL && info->console_uart) { - static char speed[11], *opt = NULL; + const struct omap_serial_console_config *con_info; + const struct omap_uart_config *uart_info; + static char speed[11], *opt = NULL; + int line, i, uart_idx; + + uart_info = omap_get_config(OMAP_TAG_UART, struct omap_uart_config); + con_info = omap_get_config(OMAP_TAG_SERIAL_CONSOLE, + struct omap_serial_console_config); + if (uart_info == NULL || con_info == NULL) + return 0; + + if (con_info->console_uart == 0) + return 0; + + if (con_info->console_speed) { + snprintf(speed, sizeof(speed), "%u", con_info->console_speed); + opt = speed; + } - if (info->console_speed) { - snprintf(speed, sizeof(speed), "%u", info->console_speed); - opt = speed; - } - return add_preferred_console("ttyS", info->console_uart - 1, opt); + uart_idx = con_info->console_uart - 1; + if (uart_idx >= OMAP_MAX_NR_PORTS) { + printk(KERN_INFO "Console: external UART#%d. " + "Not adding it as console this time.\n", + uart_idx + 1); + return 0; + } + if (!(uart_info->enabled_uarts & (1 << uart_idx))) { + printk(KERN_ERR "Console: Selected UART#%d is " + "not enabled for this platform\n", + uart_idx + 1); + return -1; + } + line = 0; + for (i = 0; i < uart_idx; i++) { + if (uart_info->enabled_uarts & (1 << i)) + line++; } - return 0; + return add_preferred_console("ttyS", line, opt); } console_initcall(omap_add_serial_console); diff --git a/arch/arm/plat-omap/devices.c b/arch/arm/plat-omap/devices.c new file mode 100644 index 00000000000..9dcce904b60 --- /dev/null +++ b/arch/arm/plat-omap/devices.c @@ -0,0 +1,381 @@ +/* + * linux/arch/arm/plat-omap/devices.c + * + * Common platform device setup/initialization for OMAP1 and OMAP2 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include <linux/config.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/platform_device.h> + +#include <asm/hardware.h> +#include <asm/io.h> +#include <asm/mach-types.h> +#include <asm/mach/map.h> + +#include <asm/arch/tc.h> +#include <asm/arch/board.h> +#include <asm/arch/mux.h> +#include <asm/arch/gpio.h> + + +void omap_nop_release(struct device *dev) +{ + /* Nothing */ +} + +/*-------------------------------------------------------------------------*/ + +#if defined(CONFIG_I2C_OMAP) || defined(CONFIG_I2C_OMAP_MODULE) + +#define OMAP1_I2C_BASE 0xfffb3800 +#define OMAP2_I2C_BASE1 0x48070000 +#define OMAP_I2C_SIZE 0x3f +#define OMAP1_I2C_INT INT_I2C +#define OMAP2_I2C_INT1 56 + +static struct resource i2c_resources1[] = { + { + .start = 0, + .end = 0, + .flags = IORESOURCE_MEM, + }, + { + .start = 0, + .flags = IORESOURCE_IRQ, + }, +}; + +/* DMA not used; works around erratum writing to non-empty i2c fifo */ + +static struct platform_device omap_i2c_device1 = { + .name = "i2c_omap", + .id = 1, + .dev = { + .release = omap_nop_release, + }, + .num_resources = ARRAY_SIZE(i2c_resources1), + .resource = i2c_resources1, +}; + +/* See also arch/arm/mach-omap2/devices.c for second I2C on 24xx */ +static void omap_init_i2c(void) +{ + if (cpu_is_omap24xx()) { + i2c_resources1[0].start = OMAP2_I2C_BASE1; + i2c_resources1[0].end = OMAP2_I2C_BASE1 + OMAP_I2C_SIZE; + i2c_resources1[1].start = OMAP2_I2C_INT1; + } else { + i2c_resources1[0].start = OMAP1_I2C_BASE; + i2c_resources1[0].end = OMAP1_I2C_BASE + OMAP_I2C_SIZE; + i2c_resources1[1].start = OMAP1_I2C_INT; + } + + /* FIXME define and use a boot tag, in case of boards that + * either don't wire up I2C, or chips that mux it differently... + * it can include clocking and address info, maybe more. + */ + if (cpu_is_omap24xx()) { + omap_cfg_reg(M19_24XX_I2C1_SCL); + omap_cfg_reg(L15_24XX_I2C1_SDA); + } else { + omap_cfg_reg(I2C_SCL); + omap_cfg_reg(I2C_SDA); + } + + (void) platform_device_register(&omap_i2c_device1); +} + +#else +static inline void omap_init_i2c(void) {} +#endif + +/*-------------------------------------------------------------------------*/ + +#if defined(CONFIG_MMC_OMAP) || defined(CONFIG_MMC_OMAP_MODULE) + +#ifdef CONFIG_ARCH_OMAP24XX +#define OMAP_MMC1_BASE 0x4809c000 +#define OMAP_MMC1_INT 83 +#else +#define OMAP_MMC1_BASE 0xfffb7800 +#define OMAP_MMC1_INT INT_MMC +#endif +#define OMAP_MMC2_BASE 0xfffb7c00 /* omap16xx only */ + +static struct omap_mmc_conf mmc1_conf; + +static u64 mmc1_dmamask = 0xffffffff; + +static struct resource mmc1_resources[] = { + { + .start = IO_ADDRESS(OMAP_MMC1_BASE), + .end = IO_ADDRESS(OMAP_MMC1_BASE) + 0x7f, + .flags = IORESOURCE_MEM, + }, + { + .start = OMAP_MMC1_INT, + .flags = IORESOURCE_IRQ, + }, +}; + +static struct platform_device mmc_omap_device1 = { + .name = "mmci-omap", + .id = 1, + .dev = { + .release = omap_nop_release, + .dma_mask = &mmc1_dmamask, + .platform_data = &mmc1_conf, + }, + .num_resources = ARRAY_SIZE(mmc1_resources), + .resource = mmc1_resources, +}; + +#ifdef CONFIG_ARCH_OMAP16XX + +static struct omap_mmc_conf mmc2_conf; + +static u64 mmc2_dmamask = 0xffffffff; + +static struct resource mmc2_resources[] = { + { + .start = IO_ADDRESS(OMAP_MMC2_BASE), + .end = IO_ADDRESS(OMAP_MMC2_BASE) + 0x7f, + .flags = IORESOURCE_MEM, + }, + { + .start = INT_1610_MMC2, + .flags = IORESOURCE_IRQ, + }, +}; + +static struct platform_device mmc_omap_device2 = { + .name = "mmci-omap", + .id = 2, + .dev = { + .release = omap_nop_release, + .dma_mask = &mmc2_dmamask, + .platform_data = &mmc2_conf, + }, + .num_resources = ARRAY_SIZE(mmc2_resources), + .resource = mmc2_resources, +}; +#endif + +static void __init omap_init_mmc(void) +{ + const struct omap_mmc_config *mmc_conf; + const struct omap_mmc_conf *mmc; + + /* NOTE: assumes MMC was never (wrongly) enabled */ + mmc_conf = omap_get_config(OMAP_TAG_MMC, struct omap_mmc_config); + if (!mmc_conf) + return; + + /* block 1 is always available and has just one pinout option */ + mmc = &mmc_conf->mmc[0]; + if (mmc->enabled) { + if (!cpu_is_omap24xx()) { + omap_cfg_reg(MMC_CMD); + omap_cfg_reg(MMC_CLK); + omap_cfg_reg(MMC_DAT0); + if (cpu_is_omap1710()) { + omap_cfg_reg(M15_1710_MMC_CLKI); + omap_cfg_reg(P19_1710_MMC_CMDDIR); + omap_cfg_reg(P20_1710_MMC_DATDIR0); + } + } + if (mmc->wire4) { + if (!cpu_is_omap24xx()) { + omap_cfg_reg(MMC_DAT1); + /* NOTE: DAT2 can be on W10 (here) or M15 */ + if (!mmc->nomux) + omap_cfg_reg(MMC_DAT2); + omap_cfg_reg(MMC_DAT3); + } + } + mmc1_conf = *mmc; + (void) platform_device_register(&mmc_omap_device1); + } + +#ifdef CONFIG_ARCH_OMAP16XX + /* block 2 is on newer chips, and has many pinout options */ + mmc = &mmc_conf->mmc[1]; + if (mmc->enabled) { + if (!mmc->nomux) { + omap_cfg_reg(Y8_1610_MMC2_CMD); + omap_cfg_reg(Y10_1610_MMC2_CLK); + omap_cfg_reg(R18_1610_MMC2_CLKIN); + omap_cfg_reg(W8_1610_MMC2_DAT0); + if (mmc->wire4) { + omap_cfg_reg(V8_1610_MMC2_DAT1); + omap_cfg_reg(W15_1610_MMC2_DAT2); + omap_cfg_reg(R10_1610_MMC2_DAT3); + } + + /* These are needed for the level shifter */ + omap_cfg_reg(V9_1610_MMC2_CMDDIR); + omap_cfg_reg(V5_1610_MMC2_DATDIR0); + omap_cfg_reg(W19_1610_MMC2_DATDIR1); + } + + /* Feedback clock must be set on OMAP-1710 MMC2 */ + if (cpu_is_omap1710()) + omap_writel(omap_readl(MOD_CONF_CTRL_1) | (1 << 24), + MOD_CONF_CTRL_1); + mmc2_conf = *mmc; + (void) platform_device_register(&mmc_omap_device2); + } +#endif + return; +} +#else +static inline void omap_init_mmc(void) {} +#endif + +#if defined(CONFIG_OMAP_WATCHDOG) || defined(CONFIG_OMAP_WATCHDOG_MODULE) + +#ifdef CONFIG_ARCH_OMAP24XX +#define OMAP_WDT_BASE 0x48022000 +#else +#define OMAP_WDT_BASE 0xfffeb000 +#endif + +static struct resource wdt_resources[] = { + { + .start = OMAP_WDT_BASE, + .end = OMAP_WDT_BASE + 0x4f, + .flags = IORESOURCE_MEM, + }, +}; + +static struct platform_device omap_wdt_device = { + .name = "omap_wdt", + .id = -1, + .dev = { + .release = omap_nop_release, + }, + .num_resources = ARRAY_SIZE(wdt_resources), + .resource = wdt_resources, +}; + +static void omap_init_wdt(void) +{ + (void) platform_device_register(&omap_wdt_device); +} +#else +static inline void omap_init_wdt(void) {} +#endif + +/*-------------------------------------------------------------------------*/ + +#if defined(CONFIG_OMAP_RNG) || defined(CONFIG_OMAP_RNG_MODULE) + +#ifdef CONFIG_ARCH_OMAP24XX +#define OMAP_RNG_BASE 0x480A0000 +#else +#define OMAP_RNG_BASE 0xfffe5000 +#endif + +static struct resource rng_resources[] = { + { + .start = OMAP_RNG_BASE, + .end = OMAP_RNG_BASE + 0x4f, + .flags = IORESOURCE_MEM, + }, +}; + +static struct platform_device omap_rng_device = { + .name = "omap_rng", + .id = -1, + .dev = { + .release = omap_nop_release, + }, + .num_resources = ARRAY_SIZE(rng_resources), + .resource = rng_resources, +}; + +static void omap_init_rng(void) +{ + (void) platform_device_register(&omap_rng_device); +} +#else +static inline void omap_init_rng(void) {} +#endif + +#if defined(CONFIG_FB_OMAP) || defined(CONFIG_FB_OMAP_MODULE) + +static struct omap_lcd_config omap_fb_conf; + +static u64 omap_fb_dma_mask = ~(u32)0; + +static struct platform_device omap_fb_device = { + .name = "omapfb", + .id = -1, + .dev = { + .release = omap_nop_release, + .dma_mask = &omap_fb_dma_mask, + .coherent_dma_mask = ~(u32)0, + .platform_data = &omap_fb_conf, + }, + .num_resources = 0, +}; + +static inline void omap_init_fb(void) +{ + const struct omap_lcd_config *conf; + + conf = omap_get_config(OMAP_TAG_LCD, struct omap_lcd_config); + if (conf != NULL) + omap_fb_conf = *conf; + platform_device_register(&omap_fb_device); +} + +#else + +static inline void omap_init_fb(void) {} + +#endif + +/* + * This gets called after board-specific INIT_MACHINE, and initializes most + * on-chip peripherals accessible on this board (except for few like USB): + * + * (a) Does any "standard config" pin muxing needed. Board-specific + * code will have muxed GPIO pins and done "nonstandard" setup; + * that code could live in the boot loader. + * (b) Populating board-specific platform_data with the data drivers + * rely on to handle wiring variations. + * (c) Creating platform devices as meaningful on this board and + * with this kernel configuration. + * + * Claiming GPIOs, and setting their direction and initial values, is the + * responsibility of the device drivers. So is responding to probe(). + * + * Board-specific knowlege like creating devices or pin setup is to be + * kept out of drivers as much as possible. In particular, pin setup + * may be handled by the boot loader, and drivers should expect it will + * normally have been done by the time they're probed. + */ +static int __init omap_init_devices(void) +{ + /* please keep these calls, and their implementations above, + * in alphabetical order so they're easier to sort through. + */ + omap_init_fb(); + omap_init_i2c(); + omap_init_mmc(); + omap_init_wdt(); + omap_init_rng(); + + return 0; +} +arch_initcall(omap_init_devices); + diff --git a/arch/arm/plat-omap/dma.c b/arch/arm/plat-omap/dma.c index da7b6514565..f5cc21ad095 100644 --- a/arch/arm/plat-omap/dma.c +++ b/arch/arm/plat-omap/dma.c @@ -6,6 +6,8 @@ * DMA channel linking for 1610 by Samuel Ortiz <samuel.ortiz@nokia.com> * Graphics DMA and LCD DMA graphics tranformations * by Imre Deak <imre.deak@nokia.com> + * OMAP2 support Copyright (C) 2004-2005 Texas Instruments, Inc. + * Merged to support both OMAP1 and OMAP2 by Tony Lindgren <tony@atomide.com> * Some functions based on earlier dma-omap.c Copyright (C) 2001 RidgeRun, Inc. * * Support functions for the OMAP internal DMA channels. @@ -31,8 +33,15 @@ #include <asm/arch/tc.h> -#define OMAP_DMA_ACTIVE 0x01 +#define DEBUG_PRINTS +#undef DEBUG_PRINTS +#ifdef DEBUG_PRINTS +#define debug_printk(x) printk x +#else +#define debug_printk(x) +#endif +#define OMAP_DMA_ACTIVE 0x01 #define OMAP_DMA_CCR_EN (1 << 7) #define OMAP_FUNC_MUX_ARM_BASE (0xfffe1000 + 0xec) @@ -55,7 +64,7 @@ static int dma_chan_count; static spinlock_t dma_chan_lock; static struct omap_dma_lch dma_chan[OMAP_LOGICAL_DMA_CH_COUNT]; -const static u8 dma_irq[OMAP_LOGICAL_DMA_CH_COUNT] = { +const static u8 omap1_dma_irq[OMAP_LOGICAL_DMA_CH_COUNT] = { INT_DMA_CH0_6, INT_DMA_CH1_7, INT_DMA_CH2_8, INT_DMA_CH3, INT_DMA_CH4, INT_DMA_CH5, INT_1610_DMA_CH6, INT_1610_DMA_CH7, INT_1610_DMA_CH8, INT_1610_DMA_CH9, INT_1610_DMA_CH10, @@ -63,6 +72,20 @@ const static u8 dma_irq[OMAP_LOGICAL_DMA_CH_COUNT] = { INT_1610_DMA_CH14, INT_1610_DMA_CH15, INT_DMA_LCD }; +#define REVISIT_24XX() printk(KERN_ERR "FIXME: no %s on 24xx\n", \ + __FUNCTION__); + +#ifdef CONFIG_ARCH_OMAP15XX +/* Returns 1 if the DMA module is in OMAP1510-compatible mode, 0 otherwise */ +int omap_dma_in_1510_mode(void) +{ + return enable_1510_mode; +} +#else +#define omap_dma_in_1510_mode() 0 +#endif + +#ifdef CONFIG_ARCH_OMAP1 static inline int get_gdma_dev(int req) { u32 reg = OMAP_FUNC_MUX_ARM_BASE + ((req - 1) / 5) * 4; @@ -82,6 +105,9 @@ static inline void set_gdma_dev(int req, int dev) l |= (dev - 1) << shift; omap_writel(l, reg); } +#else +#define set_gdma_dev(req, dev) do {} while (0) +#endif static void clear_lch_regs(int lch) { @@ -121,38 +147,62 @@ void omap_set_dma_priority(int dst_port, int priority) } void omap_set_dma_transfer_params(int lch, int data_type, int elem_count, - int frame_count, int sync_mode) + int frame_count, int sync_mode, + int dma_trigger, int src_or_dst_synch) { - u16 w; + OMAP_DMA_CSDP_REG(lch) &= ~0x03; + OMAP_DMA_CSDP_REG(lch) |= data_type; - w = omap_readw(OMAP_DMA_CSDP(lch)); - w &= ~0x03; - w |= data_type; - omap_writew(w, OMAP_DMA_CSDP(lch)); + if (cpu_class_is_omap1()) { + OMAP_DMA_CCR_REG(lch) &= ~(1 << 5); + if (sync_mode == OMAP_DMA_SYNC_FRAME) + OMAP_DMA_CCR_REG(lch) |= 1 << 5; + + OMAP1_DMA_CCR2_REG(lch) &= ~(1 << 2); + if (sync_mode == OMAP_DMA_SYNC_BLOCK) + OMAP1_DMA_CCR2_REG(lch) |= 1 << 2; + } + + if (cpu_is_omap24xx() && dma_trigger) { + u32 val = OMAP_DMA_CCR_REG(lch); + + if (dma_trigger > 63) + val |= 1 << 20; + if (dma_trigger > 31) + val |= 1 << 19; - w = omap_readw(OMAP_DMA_CCR(lch)); - w &= ~(1 << 5); - if (sync_mode == OMAP_DMA_SYNC_FRAME) - w |= 1 << 5; - omap_writew(w, OMAP_DMA_CCR(lch)); + val |= (dma_trigger & 0x1f); - w = omap_readw(OMAP_DMA_CCR2(lch)); - w &= ~(1 << 2); - if (sync_mode == OMAP_DMA_SYNC_BLOCK) - w |= 1 << 2; - omap_writew(w, OMAP_DMA_CCR2(lch)); + if (sync_mode & OMAP_DMA_SYNC_FRAME) + val |= 1 << 5; - omap_writew(elem_count, OMAP_DMA_CEN(lch)); - omap_writew(frame_count, OMAP_DMA_CFN(lch)); + if (sync_mode & OMAP_DMA_SYNC_BLOCK) + val |= 1 << 18; + if (src_or_dst_synch) + val |= 1 << 24; /* source synch */ + else + val &= ~(1 << 24); /* dest synch */ + + OMAP_DMA_CCR_REG(lch) = val; + } + + OMAP_DMA_CEN_REG(lch) = elem_count; + OMAP_DMA_CFN_REG(lch) = frame_count; } + void omap_set_dma_color_mode(int lch, enum omap_dma_color_mode mode, u32 color) { u16 w; BUG_ON(omap_dma_in_1510_mode()); - w = omap_readw(OMAP_DMA_CCR2(lch)) & ~0x03; + if (cpu_is_omap24xx()) { + REVISIT_24XX(); + return; + } + + w = OMAP1_DMA_CCR2_REG(lch) & ~0x03; switch (mode) { case OMAP_DMA_CONSTANT_FILL: w |= 0x01; @@ -165,63 +215,84 @@ void omap_set_dma_color_mode(int lch, enum omap_dma_color_mode mode, u32 color) default: BUG(); } - omap_writew(w, OMAP_DMA_CCR2(lch)); + OMAP1_DMA_CCR2_REG(lch) = w; - w = omap_readw(OMAP_DMA_LCH_CTRL(lch)) & ~0x0f; + w = OMAP1_DMA_LCH_CTRL_REG(lch) & ~0x0f; /* Default is channel type 2D */ if (mode) { - omap_writew((u16)color, OMAP_DMA_COLOR_L(lch)); - omap_writew((u16)(color >> 16), OMAP_DMA_COLOR_U(lch)); + OMAP1_DMA_COLOR_L_REG(lch) = (u16)color; + OMAP1_DMA_COLOR_U_REG(lch) = (u16)(color >> 16); w |= 1; /* Channel type G */ } - omap_writew(w, OMAP_DMA_LCH_CTRL(lch)); + OMAP1_DMA_LCH_CTRL_REG(lch) = w; } - +/* Note that src_port is only for omap1 */ void omap_set_dma_src_params(int lch, int src_port, int src_amode, - unsigned long src_start) + unsigned long src_start, + int src_ei, int src_fi) { - u16 w; + if (cpu_class_is_omap1()) { + OMAP_DMA_CSDP_REG(lch) &= ~(0x1f << 2); + OMAP_DMA_CSDP_REG(lch) |= src_port << 2; + } + + OMAP_DMA_CCR_REG(lch) &= ~(0x03 << 12); + OMAP_DMA_CCR_REG(lch) |= src_amode << 12; + + if (cpu_class_is_omap1()) { + OMAP1_DMA_CSSA_U_REG(lch) = src_start >> 16; + OMAP1_DMA_CSSA_L_REG(lch) = src_start; + } - w = omap_readw(OMAP_DMA_CSDP(lch)); - w &= ~(0x1f << 2); - w |= src_port << 2; - omap_writew(w, OMAP_DMA_CSDP(lch)); + if (cpu_is_omap24xx()) + OMAP2_DMA_CSSA_REG(lch) = src_start; - w = omap_readw(OMAP_DMA_CCR(lch)); - w &= ~(0x03 << 12); - w |= src_amode << 12; - omap_writew(w, OMAP_DMA_CCR(lch)); + OMAP_DMA_CSEI_REG(lch) = src_ei; + OMAP_DMA_CSFI_REG(lch) = src_fi; +} - omap_writew(src_start >> 16, OMAP_DMA_CSSA_U(lch)); - omap_writew(src_start, OMAP_DMA_CSSA_L(lch)); +void omap_set_dma_params(int lch, struct omap_dma_channel_params * params) +{ + omap_set_dma_transfer_params(lch, params->data_type, + params->elem_count, params->frame_count, + params->sync_mode, params->trigger, + params->src_or_dst_synch); + omap_set_dma_src_params(lch, params->src_port, + params->src_amode, params->src_start, + params->src_ei, params->src_fi); + + omap_set_dma_dest_params(lch, params->dst_port, + params->dst_amode, params->dst_start, + params->dst_ei, params->dst_fi); } void omap_set_dma_src_index(int lch, int eidx, int fidx) { - omap_writew(eidx, OMAP_DMA_CSEI(lch)); - omap_writew(fidx, OMAP_DMA_CSFI(lch)); + if (cpu_is_omap24xx()) { + REVISIT_24XX(); + return; + } + OMAP_DMA_CSEI_REG(lch) = eidx; + OMAP_DMA_CSFI_REG(lch) = fidx; } void omap_set_dma_src_data_pack(int lch, int enable) { - u16 w; - - w = omap_readw(OMAP_DMA_CSDP(lch)) & ~(1 << 6); - w |= enable ? (1 << 6) : 0; - omap_writew(w, OMAP_DMA_CSDP(lch)); + OMAP_DMA_CSDP_REG(lch) &= ~(1 << 6); + if (enable) + OMAP_DMA_CSDP_REG(lch) |= (1 << 6); } void omap_set_dma_src_burst_mode(int lch, enum omap_dma_burst_mode burst_mode) { - u16 w; + OMAP_DMA_CSDP_REG(lch) &= ~(0x03 << 7); - w = omap_readw(OMAP_DMA_CSDP(lch)) & ~(0x03 << 7); switch (burst_mode) { case OMAP_DMA_DATA_BURST_DIS: break; case OMAP_DMA_DATA_BURST_4: - w |= (0x01 << 7); + OMAP_DMA_CSDP_REG(lch) |= (0x02 << 7); break; case OMAP_DMA_DATA_BURST_8: /* not supported by current hardware @@ -231,110 +302,283 @@ void omap_set_dma_src_burst_mode(int lch, enum omap_dma_burst_mode burst_mode) default: BUG(); } - omap_writew(w, OMAP_DMA_CSDP(lch)); } +/* Note that dest_port is only for OMAP1 */ void omap_set_dma_dest_params(int lch, int dest_port, int dest_amode, - unsigned long dest_start) + unsigned long dest_start, + int dst_ei, int dst_fi) { - u16 w; + if (cpu_class_is_omap1()) { + OMAP_DMA_CSDP_REG(lch) &= ~(0x1f << 9); + OMAP_DMA_CSDP_REG(lch) |= dest_port << 9; + } - w = omap_readw(OMAP_DMA_CSDP(lch)); - w &= ~(0x1f << 9); - w |= dest_port << 9; - omap_writew(w, OMAP_DMA_CSDP(lch)); + OMAP_DMA_CCR_REG(lch) &= ~(0x03 << 14); + OMAP_DMA_CCR_REG(lch) |= dest_amode << 14; + + if (cpu_class_is_omap1()) { + OMAP1_DMA_CDSA_U_REG(lch) = dest_start >> 16; + OMAP1_DMA_CDSA_L_REG(lch) = dest_start; + } - w = omap_readw(OMAP_DMA_CCR(lch)); - w &= ~(0x03 << 14); - w |= dest_amode << 14; - omap_writew(w, OMAP_DMA_CCR(lch)); + if (cpu_is_omap24xx()) + OMAP2_DMA_CDSA_REG(lch) = dest_start; - omap_writew(dest_start >> 16, OMAP_DMA_CDSA_U(lch)); - omap_writew(dest_start, OMAP_DMA_CDSA_L(lch)); + OMAP_DMA_CDEI_REG(lch) = dst_ei; + OMAP_DMA_CDFI_REG(lch) = dst_fi; } void omap_set_dma_dest_index(int lch, int eidx, int fidx) { - omap_writew(eidx, OMAP_DMA_CDEI(lch)); - omap_writew(fidx, OMAP_DMA_CDFI(lch)); + if (cpu_is_omap24xx()) { + REVISIT_24XX(); + return; + } + OMAP_DMA_CDEI_REG(lch) = eidx; + OMAP_DMA_CDFI_REG(lch) = fidx; } void omap_set_dma_dest_data_pack(int lch, int enable) { - u16 w; - - w = omap_readw(OMAP_DMA_CSDP(lch)) & ~(1 << 13); - w |= enable ? (1 << 13) : 0; - omap_writew(w, OMAP_DMA_CSDP(lch)); + OMAP_DMA_CSDP_REG(lch) &= ~(1 << 13); + if (enable) + OMAP_DMA_CSDP_REG(lch) |= 1 << 13; } void omap_set_dma_dest_burst_mode(int lch, enum omap_dma_burst_mode burst_mode) { - u16 w; + OMAP_DMA_CSDP_REG(lch) &= ~(0x03 << 14); - w = omap_readw(OMAP_DMA_CSDP(lch)) & ~(0x03 << 14); switch (burst_mode) { case OMAP_DMA_DATA_BURST_DIS: break; case OMAP_DMA_DATA_BURST_4: - w |= (0x01 << 14); + OMAP_DMA_CSDP_REG(lch) |= (0x02 << 14); break; case OMAP_DMA_DATA_BURST_8: - w |= (0x03 << 14); + OMAP_DMA_CSDP_REG(lch) |= (0x03 << 14); break; default: printk(KERN_ERR "Invalid DMA burst mode\n"); BUG(); return; } - omap_writew(w, OMAP_DMA_CSDP(lch)); } -static inline void init_intr(int lch) +static inline void omap_enable_channel_irq(int lch) { - u16 w; + u32 status; /* Read CSR to make sure it's cleared. */ - w = omap_readw(OMAP_DMA_CSR(lch)); + status = OMAP_DMA_CSR_REG(lch); + /* Enable some nice interrupts. */ - omap_writew(dma_chan[lch].enabled_irqs, OMAP_DMA_CICR(lch)); + OMAP_DMA_CICR_REG(lch) = dma_chan[lch].enabled_irqs; + dma_chan[lch].flags |= OMAP_DMA_ACTIVE; } -static inline void enable_lnk(int lch) +static void omap_disable_channel_irq(int lch) { - u16 w; + if (cpu_is_omap24xx()) + OMAP_DMA_CICR_REG(lch) = 0; +} + +void omap_enable_dma_irq(int lch, u16 bits) +{ + dma_chan[lch].enabled_irqs |= bits; +} - /* Clear the STOP_LNK bits */ - w = omap_readw(OMAP_DMA_CLNK_CTRL(lch)); - w &= ~(1 << 14); - omap_writew(w, OMAP_DMA_CLNK_CTRL(lch)); +void omap_disable_dma_irq(int lch, u16 bits) +{ + dma_chan[lch].enabled_irqs &= ~bits; +} + +static inline void enable_lnk(int lch) +{ + if (cpu_class_is_omap1()) + OMAP_DMA_CLNK_CTRL_REG(lch) &= ~(1 << 14); - /* And set the ENABLE_LNK bits */ + /* Set the ENABLE_LNK bits */ if (dma_chan[lch].next_lch != -1) - omap_writew(dma_chan[lch].next_lch | (1 << 15), - OMAP_DMA_CLNK_CTRL(lch)); + OMAP_DMA_CLNK_CTRL_REG(lch) = + dma_chan[lch].next_lch | (1 << 15); } static inline void disable_lnk(int lch) { - u16 w; - /* Disable interrupts */ - omap_writew(0, OMAP_DMA_CICR(lch)); + if (cpu_class_is_omap1()) { + OMAP_DMA_CICR_REG(lch) = 0; + /* Set the STOP_LNK bit */ + OMAP_DMA_CLNK_CTRL_REG(lch) |= 1 << 14; + } - /* Set the STOP_LNK bit */ - w = omap_readw(OMAP_DMA_CLNK_CTRL(lch)); - w |= (1 << 14); - w = omap_writew(w, OMAP_DMA_CLNK_CTRL(lch)); + if (cpu_is_omap24xx()) { + omap_disable_channel_irq(lch); + /* Clear the ENABLE_LNK bit */ + OMAP_DMA_CLNK_CTRL_REG(lch) &= ~(1 << 15); + } dma_chan[lch].flags &= ~OMAP_DMA_ACTIVE; } -void omap_start_dma(int lch) +static inline void omap2_enable_irq_lch(int lch) { - u16 w; + u32 val; + + if (!cpu_is_omap24xx()) + return; + + val = omap_readl(OMAP_DMA4_IRQENABLE_L0); + val |= 1 << lch; + omap_writel(val, OMAP_DMA4_IRQENABLE_L0); +} + +int omap_request_dma(int dev_id, const char *dev_name, + void (* callback)(int lch, u16 ch_status, void *data), + void *data, int *dma_ch_out) +{ + int ch, free_ch = -1; + unsigned long flags; + struct omap_dma_lch *chan; + + spin_lock_irqsave(&dma_chan_lock, flags); + for (ch = 0; ch < dma_chan_count; ch++) { + if (free_ch == -1 && dma_chan[ch].dev_id == -1) { + free_ch = ch; + if (dev_id == 0) + break; + } + } + if (free_ch == -1) { + spin_unlock_irqrestore(&dma_chan_lock, flags); + return -EBUSY; + } + chan = dma_chan + free_ch; + chan->dev_id = dev_id; + + if (cpu_class_is_omap1()) + clear_lch_regs(free_ch); + if (cpu_is_omap24xx()) + omap_clear_dma(free_ch); + + spin_unlock_irqrestore(&dma_chan_lock, flags); + + chan->dev_name = dev_name; + chan->callback = callback; + chan->data = data; + chan->enabled_irqs = OMAP_DMA_TOUT_IRQ | OMAP_DMA_DROP_IRQ | + OMAP_DMA_BLOCK_IRQ; + + if (cpu_is_omap24xx()) + chan->enabled_irqs |= OMAP2_DMA_TRANS_ERR_IRQ; + + if (cpu_is_omap16xx()) { + /* If the sync device is set, configure it dynamically. */ + if (dev_id != 0) { + set_gdma_dev(free_ch + 1, dev_id); + dev_id = free_ch + 1; + } + /* Disable the 1510 compatibility mode and set the sync device + * id. */ + OMAP_DMA_CCR_REG(free_ch) = dev_id | (1 << 10); + } else if (cpu_is_omap730() || cpu_is_omap15xx()) { + OMAP_DMA_CCR_REG(free_ch) = dev_id; + } + + if (cpu_is_omap24xx()) { + omap2_enable_irq_lch(free_ch); + + omap_enable_channel_irq(free_ch); + /* Clear the CSR register and IRQ status register */ + OMAP_DMA_CSR_REG(free_ch) = 0x0; + omap_writel(~0x0, OMAP_DMA4_IRQSTATUS_L0); + } + + *dma_ch_out = free_ch; + + return 0; +} + +void omap_free_dma(int lch) +{ + unsigned long flags; + + spin_lock_irqsave(&dma_chan_lock, flags); + if (dma_chan[lch].dev_id == -1) { + printk("omap_dma: trying to free nonallocated DMA channel %d\n", + lch); + spin_unlock_irqrestore(&dma_chan_lock, flags); + return; + } + dma_chan[lch].dev_id = -1; + dma_chan[lch].next_lch = -1; + dma_chan[lch].callback = NULL; + spin_unlock_irqrestore(&dma_chan_lock, flags); + + if (cpu_class_is_omap1()) { + /* Disable all DMA interrupts for the channel. */ + OMAP_DMA_CICR_REG(lch) = 0; + /* Make sure the DMA transfer is stopped. */ + OMAP_DMA_CCR_REG(lch) = 0; + } + + if (cpu_is_omap24xx()) { + u32 val; + /* Disable interrupts */ + val = omap_readl(OMAP_DMA4_IRQENABLE_L0); + val &= ~(1 << lch); + omap_writel(val, OMAP_DMA4_IRQENABLE_L0); + + /* Clear the CSR register and IRQ status register */ + OMAP_DMA_CSR_REG(lch) = 0x0; + + val = omap_readl(OMAP_DMA4_IRQSTATUS_L0); + val |= 1 << lch; + omap_writel(val, OMAP_DMA4_IRQSTATUS_L0); + + /* Disable all DMA interrupts for the channel. */ + OMAP_DMA_CICR_REG(lch) = 0; + + /* Make sure the DMA transfer is stopped. */ + OMAP_DMA_CCR_REG(lch) = 0; + omap_clear_dma(lch); + } +} + +/* + * Clears any DMA state so the DMA engine is ready to restart with new buffers + * through omap_start_dma(). Any buffers in flight are discarded. + */ +void omap_clear_dma(int lch) +{ + unsigned long flags; + + local_irq_save(flags); + + if (cpu_class_is_omap1()) { + int status; + OMAP_DMA_CCR_REG(lch) &= ~OMAP_DMA_CCR_EN; + + /* Clear pending interrupts */ + status = OMAP_DMA_CSR_REG(lch); + } + + if (cpu_is_omap24xx()) { + int i; + u32 lch_base = OMAP24XX_DMA_BASE + lch * 0x60 + 0x80; + for (i = 0; i < 0x44; i += 4) + omap_writel(0, lch_base + i); + } + + local_irq_restore(flags); +} + +void omap_start_dma(int lch) +{ if (!omap_dma_in_1510_mode() && dma_chan[lch].next_lch != -1) { int next_lch, cur_lch; char dma_chan_link_map[OMAP_LOGICAL_DMA_CH_COUNT]; @@ -348,31 +592,37 @@ void omap_start_dma(int lch) do { next_lch = dma_chan[cur_lch].next_lch; - /* The loop case: we've been here already */ + /* The loop case: we've been here already */ if (dma_chan_link_map[cur_lch]) break; /* Mark the current channel */ dma_chan_link_map[cur_lch] = 1; enable_lnk(cur_lch); - init_intr(cur_lch); + omap_enable_channel_irq(cur_lch); cur_lch = next_lch; } while (next_lch != -1); + } else if (cpu_is_omap24xx()) { + /* Errata: Need to write lch even if not using chaining */ + OMAP_DMA_CLNK_CTRL_REG(lch) = lch; } - init_intr(lch); + omap_enable_channel_irq(lch); + + /* Errata: On ES2.0 BUFFERING disable must be set. + * This will always fail on ES1.0 */ + if (cpu_is_omap24xx()) { + OMAP_DMA_CCR_REG(lch) |= OMAP_DMA_CCR_EN; + } + + OMAP_DMA_CCR_REG(lch) |= OMAP_DMA_CCR_EN; - w = omap_readw(OMAP_DMA_CCR(lch)); - w |= OMAP_DMA_CCR_EN; - omap_writew(w, OMAP_DMA_CCR(lch)); dma_chan[lch].flags |= OMAP_DMA_ACTIVE; } void omap_stop_dma(int lch) { - u16 w; - if (!omap_dma_in_1510_mode() && dma_chan[lch].next_lch != -1) { int next_lch, cur_lch = lch; char dma_chan_link_map[OMAP_LOGICAL_DMA_CH_COUNT]; @@ -393,146 +643,83 @@ void omap_stop_dma(int lch) return; } + /* Disable all interrupts on the channel */ - omap_writew(0, OMAP_DMA_CICR(lch)); + if (cpu_class_is_omap1()) + OMAP_DMA_CICR_REG(lch) = 0; - w = omap_readw(OMAP_DMA_CCR(lch)); - w &= ~OMAP_DMA_CCR_EN; - omap_writew(w, OMAP_DMA_CCR(lch)); + OMAP_DMA_CCR_REG(lch) &= ~OMAP_DMA_CCR_EN; dma_chan[lch].flags &= ~OMAP_DMA_ACTIVE; } -void omap_enable_dma_irq(int lch, u16 bits) +/* + * Returns current physical source address for the given DMA channel. + * If the channel is running the caller must disable interrupts prior calling + * this function and process the returned value before re-enabling interrupt to + * prevent races with the interrupt handler. Note that in continuous mode there + * is a chance for CSSA_L register overflow inbetween the two reads resulting + * in incorrect return value. + */ +dma_addr_t omap_get_dma_src_pos(int lch) { - dma_chan[lch].enabled_irqs |= bits; -} + dma_addr_t offset; -void omap_disable_dma_irq(int lch, u16 bits) -{ - dma_chan[lch].enabled_irqs &= ~bits; -} + if (cpu_class_is_omap1()) + offset = (dma_addr_t) (OMAP1_DMA_CSSA_L_REG(lch) | + (OMAP1_DMA_CSSA_U_REG(lch) << 16)); -static int dma_handle_ch(int ch) -{ - u16 csr; + if (cpu_is_omap24xx()) + offset = OMAP_DMA_CSAC_REG(lch); - if (enable_1510_mode && ch >= 6) { - csr = dma_chan[ch].saved_csr; - dma_chan[ch].saved_csr = 0; - } else - csr = omap_readw(OMAP_DMA_CSR(ch)); - if (enable_1510_mode && ch <= 2 && (csr >> 7) != 0) { - dma_chan[ch + 6].saved_csr = csr >> 7; - csr &= 0x7f; - } - if ((csr & 0x3f) == 0) - return 0; - if (unlikely(dma_chan[ch].dev_id == -1)) { - printk(KERN_WARNING "Spurious interrupt from DMA channel %d (CSR %04x)\n", - ch, csr); - return 0; - } - if (unlikely(csr & OMAP_DMA_TOUT_IRQ)) - printk(KERN_WARNING "DMA timeout with device %d\n", dma_chan[ch].dev_id); - if (unlikely(csr & OMAP_DMA_DROP_IRQ)) - printk(KERN_WARNING "DMA synchronization event drop occurred with device %d\n", - dma_chan[ch].dev_id); - if (likely(csr & OMAP_DMA_BLOCK_IRQ)) - dma_chan[ch].flags &= ~OMAP_DMA_ACTIVE; - if (likely(dma_chan[ch].callback != NULL)) - dma_chan[ch].callback(ch, csr, dma_chan[ch].data); - return 1; + return offset; } -static irqreturn_t dma_irq_handler(int irq, void *dev_id, struct pt_regs *regs) +/* + * Returns current physical destination address for the given DMA channel. + * If the channel is running the caller must disable interrupts prior calling + * this function and process the returned value before re-enabling interrupt to + * prevent races with the interrupt handler. Note that in continuous mode there + * is a chance for CDSA_L register overflow inbetween the two reads resulting + * in incorrect return value. + */ +dma_addr_t omap_get_dma_dst_pos(int lch) { - int ch = ((int) dev_id) - 1; - int handled = 0; + dma_addr_t offset; - for (;;) { - int handled_now = 0; + if (cpu_class_is_omap1()) + offset = (dma_addr_t) (OMAP1_DMA_CDSA_L_REG(lch) | + (OMAP1_DMA_CDSA_U_REG(lch) << 16)); - handled_now += dma_handle_ch(ch); - if (enable_1510_mode && dma_chan[ch + 6].saved_csr) - handled_now += dma_handle_ch(ch + 6); - if (!handled_now) - break; - handled += handled_now; - } + if (cpu_is_omap24xx()) + offset = OMAP2_DMA_CDSA_REG(lch); - return handled ? IRQ_HANDLED : IRQ_NONE; + return offset; } -int omap_request_dma(int dev_id, const char *dev_name, - void (* callback)(int lch, u16 ch_status, void *data), - void *data, int *dma_ch_out) +/* + * Returns current source transfer counting for the given DMA channel. + * Can be used to monitor the progress of a transfer inside a block. + * It must be called with disabled interrupts. + */ +int omap_get_dma_src_addr_counter(int lch) { - int ch, free_ch = -1; - unsigned long flags; - struct omap_dma_lch *chan; - - spin_lock_irqsave(&dma_chan_lock, flags); - for (ch = 0; ch < dma_chan_count; ch++) { - if (free_ch == -1 && dma_chan[ch].dev_id == -1) { - free_ch = ch; - if (dev_id == 0) - break; - } - } - if (free_ch == -1) { - spin_unlock_irqrestore(&dma_chan_lock, flags); - return -EBUSY; - } - chan = dma_chan + free_ch; - chan->dev_id = dev_id; - clear_lch_regs(free_ch); - spin_unlock_irqrestore(&dma_chan_lock, flags); - - chan->dev_id = dev_id; - chan->dev_name = dev_name; - chan->callback = callback; - chan->data = data; - chan->enabled_irqs = OMAP_DMA_TOUT_IRQ | OMAP_DMA_DROP_IRQ | OMAP_DMA_BLOCK_IRQ; - - if (cpu_is_omap16xx()) { - /* If the sync device is set, configure it dynamically. */ - if (dev_id != 0) { - set_gdma_dev(free_ch + 1, dev_id); - dev_id = free_ch + 1; - } - /* Disable the 1510 compatibility mode and set the sync device - * id. */ - omap_writew(dev_id | (1 << 10), OMAP_DMA_CCR(free_ch)); - } else { - omap_writew(dev_id, OMAP_DMA_CCR(free_ch)); - } - *dma_ch_out = free_ch; - - return 0; + return (dma_addr_t) OMAP_DMA_CSAC_REG(lch); } -void omap_free_dma(int ch) +int omap_dma_running(void) { - unsigned long flags; + int lch; - spin_lock_irqsave(&dma_chan_lock, flags); - if (dma_chan[ch].dev_id == -1) { - printk("omap_dma: trying to free nonallocated DMA channel %d\n", ch); - spin_unlock_irqrestore(&dma_chan_lock, flags); - return; - } - dma_chan[ch].dev_id = -1; - spin_unlock_irqrestore(&dma_chan_lock, flags); + /* Check if LCD DMA is running */ + if (cpu_is_omap16xx()) + if (omap_readw(OMAP1610_DMA_LCD_CCR) & OMAP_DMA_CCR_EN) + return 1; - /* Disable all DMA interrupts for the channel. */ - omap_writew(0, OMAP_DMA_CICR(ch)); - /* Make sure the DMA transfer is stopped. */ - omap_writew(0, OMAP_DMA_CCR(ch)); -} + for (lch = 0; lch < dma_chan_count; lch++) + if (OMAP_DMA_CCR_REG(lch) & OMAP_DMA_CCR_EN) + return 1; -int omap_dma_in_1510_mode(void) -{ - return enable_1510_mode; + return 0; } /* @@ -550,7 +737,8 @@ void omap_dma_link_lch (int lch_head, int lch_queue) if ((dma_chan[lch_head].dev_id == -1) || (dma_chan[lch_queue].dev_id == -1)) { - printk(KERN_ERR "omap_dma: trying to link non requested channels\n"); + printk(KERN_ERR "omap_dma: trying to link " + "non requested channels\n"); dump_stack(); } @@ -570,20 +758,149 @@ void omap_dma_unlink_lch (int lch_head, int lch_queue) if (dma_chan[lch_head].next_lch != lch_queue || dma_chan[lch_head].next_lch == -1) { - printk(KERN_ERR "omap_dma: trying to unlink non linked channels\n"); + printk(KERN_ERR "omap_dma: trying to unlink " + "non linked channels\n"); dump_stack(); } if ((dma_chan[lch_head].flags & OMAP_DMA_ACTIVE) || (dma_chan[lch_head].flags & OMAP_DMA_ACTIVE)) { - printk(KERN_ERR "omap_dma: You need to stop the DMA channels before unlinking\n"); + printk(KERN_ERR "omap_dma: You need to stop the DMA channels " + "before unlinking\n"); dump_stack(); } dma_chan[lch_head].next_lch = -1; } +/*----------------------------------------------------------------------------*/ + +#ifdef CONFIG_ARCH_OMAP1 + +static int omap1_dma_handle_ch(int ch) +{ + u16 csr; + + if (enable_1510_mode && ch >= 6) { + csr = dma_chan[ch].saved_csr; + dma_chan[ch].saved_csr = 0; + } else + csr = OMAP_DMA_CSR_REG(ch); + if (enable_1510_mode && ch <= 2 && (csr >> 7) != 0) { + dma_chan[ch + 6].saved_csr = csr >> 7; + csr &= 0x7f; + } + if ((csr & 0x3f) == 0) + return 0; + if (unlikely(dma_chan[ch].dev_id == -1)) { + printk(KERN_WARNING "Spurious interrupt from DMA channel " + "%d (CSR %04x)\n", ch, csr); + return 0; + } + if (unlikely(csr & OMAP_DMA_TOUT_IRQ)) + printk(KERN_WARNING "DMA timeout with device %d\n", + dma_chan[ch].dev_id); + if (unlikely(csr & OMAP_DMA_DROP_IRQ)) + printk(KERN_WARNING "DMA synchronization event drop occurred " + "with device %d\n", dma_chan[ch].dev_id); + if (likely(csr & OMAP_DMA_BLOCK_IRQ)) + dma_chan[ch].flags &= ~OMAP_DMA_ACTIVE; + if (likely(dma_chan[ch].callback != NULL)) + dma_chan[ch].callback(ch, csr, dma_chan[ch].data); + return 1; +} + +static irqreturn_t omap1_dma_irq_handler(int irq, void *dev_id, + struct pt_regs *regs) +{ + int ch = ((int) dev_id) - 1; + int handled = 0; + + for (;;) { + int handled_now = 0; + + handled_now += omap1_dma_handle_ch(ch); + if (enable_1510_mode && dma_chan[ch + 6].saved_csr) + handled_now += omap1_dma_handle_ch(ch + 6); + if (!handled_now) + break; + handled += handled_now; + } + + return handled ? IRQ_HANDLED : IRQ_NONE; +} + +#else +#define omap1_dma_irq_handler NULL +#endif + +#ifdef CONFIG_ARCH_OMAP2 + +static int omap2_dma_handle_ch(int ch) +{ + u32 status = OMAP_DMA_CSR_REG(ch); + u32 val; + + if (!status) + return 0; + if (unlikely(dma_chan[ch].dev_id == -1)) + return 0; + /* REVISIT: According to 24xx TRM, there's no TOUT_IE */ + if (unlikely(status & OMAP_DMA_TOUT_IRQ)) + printk(KERN_INFO "DMA timeout with device %d\n", + dma_chan[ch].dev_id); + if (unlikely(status & OMAP_DMA_DROP_IRQ)) + printk(KERN_INFO + "DMA synchronization event drop occurred with device " + "%d\n", dma_chan[ch].dev_id); + + if (unlikely(status & OMAP2_DMA_TRANS_ERR_IRQ)) + printk(KERN_INFO "DMA transaction error with device %d\n", + dma_chan[ch].dev_id); + + OMAP_DMA_CSR_REG(ch) = 0x20; + + val = omap_readl(OMAP_DMA4_IRQSTATUS_L0); + /* ch in this function is from 0-31 while in register it is 1-32 */ + val = 1 << (ch); + omap_writel(val, OMAP_DMA4_IRQSTATUS_L0); + + if (likely(dma_chan[ch].callback != NULL)) + dma_chan[ch].callback(ch, status, dma_chan[ch].data); + + return 0; +} + +/* STATUS register count is from 1-32 while our is 0-31 */ +static irqreturn_t omap2_dma_irq_handler(int irq, void *dev_id, + struct pt_regs *regs) +{ + u32 val; + int i; + + val = omap_readl(OMAP_DMA4_IRQSTATUS_L0); + + for (i = 1; i <= OMAP_LOGICAL_DMA_CH_COUNT; i++) { + int active = val & (1 << (i - 1)); + if (active) + omap2_dma_handle_ch(i - 1); + } + + return IRQ_HANDLED; +} + +static struct irqaction omap24xx_dma_irq = { + .name = "DMA", + .handler = omap2_dma_irq_handler, + .flags = SA_INTERRUPT +}; + +#else +static struct irqaction omap24xx_dma_irq; +#endif + +/*----------------------------------------------------------------------------*/ static struct lcd_dma_info { spinlock_t lock; @@ -795,7 +1112,7 @@ static void set_b1_regs(void) /* Always set the source port as SDRAM for now*/ w &= ~(0x03 << 6); if (lcd_dma.callback != NULL) - w |= 1 << 1; /* Block interrupt enable */ + w |= 1 << 1; /* Block interrupt enable */ else w &= ~(1 << 1); omap_writew(w, OMAP1610_DMA_LCD_CTRL); @@ -814,7 +1131,8 @@ static void set_b1_regs(void) omap_writew(fi, OMAP1610_DMA_LCD_SRC_FI_B1_L); } -static irqreturn_t lcd_dma_irq_handler(int irq, void *dev_id, struct pt_regs *regs) +static irqreturn_t lcd_dma_irq_handler(int irq, void *dev_id, + struct pt_regs *regs) { u16 w; @@ -870,7 +1188,8 @@ void omap_free_lcd_dma(void) return; } if (!enable_1510_mode) - omap_writew(omap_readw(OMAP1610_DMA_LCD_CCR) & ~1, OMAP1610_DMA_LCD_CCR); + omap_writew(omap_readw(OMAP1610_DMA_LCD_CCR) & ~1, + OMAP1610_DMA_LCD_CCR); lcd_dma.reserved = 0; spin_unlock(&lcd_dma.lock); } @@ -939,93 +1258,24 @@ void omap_stop_lcd_dma(void) omap_writew(w, OMAP1610_DMA_LCD_CTRL); } -/* - * Clears any DMA state so the DMA engine is ready to restart with new buffers - * through omap_start_dma(). Any buffers in flight are discarded. - */ -void omap_clear_dma(int lch) -{ - unsigned long flags; - int status; - - local_irq_save(flags); - omap_writew(omap_readw(OMAP_DMA_CCR(lch)) & ~OMAP_DMA_CCR_EN, - OMAP_DMA_CCR(lch)); - status = OMAP_DMA_CSR(lch); /* clear pending interrupts */ - local_irq_restore(flags); -} - -/* - * Returns current physical source address for the given DMA channel. - * If the channel is running the caller must disable interrupts prior calling - * this function and process the returned value before re-enabling interrupt to - * prevent races with the interrupt handler. Note that in continuous mode there - * is a chance for CSSA_L register overflow inbetween the two reads resulting - * in incorrect return value. - */ -dma_addr_t omap_get_dma_src_pos(int lch) -{ - return (dma_addr_t) (omap_readw(OMAP_DMA_CSSA_L(lch)) | - (omap_readw(OMAP_DMA_CSSA_U(lch)) << 16)); -} - -/* - * Returns current physical destination address for the given DMA channel. - * If the channel is running the caller must disable interrupts prior calling - * this function and process the returned value before re-enabling interrupt to - * prevent races with the interrupt handler. Note that in continuous mode there - * is a chance for CDSA_L register overflow inbetween the two reads resulting - * in incorrect return value. - */ -dma_addr_t omap_get_dma_dst_pos(int lch) -{ - return (dma_addr_t) (omap_readw(OMAP_DMA_CDSA_L(lch)) | - (omap_readw(OMAP_DMA_CDSA_U(lch)) << 16)); -} - -/* - * Returns current source transfer counting for the given DMA channel. - * Can be used to monitor the progress of a transfer inside a block. - * It must be called with disabled interrupts. - */ -int omap_get_dma_src_addr_counter(int lch) -{ - return (dma_addr_t) omap_readw(OMAP_DMA_CSAC(lch)); -} - -int omap_dma_running(void) -{ - int lch; - - /* Check if LCD DMA is running */ - if (cpu_is_omap16xx()) - if (omap_readw(OMAP1610_DMA_LCD_CCR) & OMAP_DMA_CCR_EN) - return 1; - - for (lch = 0; lch < dma_chan_count; lch++) { - u16 w; - - w = omap_readw(OMAP_DMA_CCR(lch)); - if (w & OMAP_DMA_CCR_EN) - return 1; - } - return 0; -} +/*----------------------------------------------------------------------------*/ static int __init omap_init_dma(void) { int ch, r; - if (cpu_is_omap1510()) { - printk(KERN_INFO "DMA support for OMAP1510 initialized\n"); + if (cpu_is_omap15xx()) { + printk(KERN_INFO "DMA support for OMAP15xx initialized\n"); dma_chan_count = 9; enable_1510_mode = 1; } else if (cpu_is_omap16xx() || cpu_is_omap730()) { printk(KERN_INFO "OMAP DMA hardware version %d\n", omap_readw(OMAP_DMA_HW_ID)); printk(KERN_INFO "DMA capabilities: %08x:%08x:%04x:%04x:%04x\n", - (omap_readw(OMAP_DMA_CAPS_0_U) << 16) | omap_readw(OMAP_DMA_CAPS_0_L), - (omap_readw(OMAP_DMA_CAPS_1_U) << 16) | omap_readw(OMAP_DMA_CAPS_1_L), + (omap_readw(OMAP_DMA_CAPS_0_U) << 16) | + omap_readw(OMAP_DMA_CAPS_0_L), + (omap_readw(OMAP_DMA_CAPS_1_U) << 16) | + omap_readw(OMAP_DMA_CAPS_1_L), omap_readw(OMAP_DMA_CAPS_2), omap_readw(OMAP_DMA_CAPS_3), omap_readw(OMAP_DMA_CAPS_4)); if (!enable_1510_mode) { @@ -1038,6 +1288,11 @@ static int __init omap_init_dma(void) dma_chan_count = 16; } else dma_chan_count = 9; + } else if (cpu_is_omap24xx()) { + u8 revision = omap_readb(OMAP_DMA4_REVISION); + printk(KERN_INFO "OMAP DMA hardware revision %d.%d\n", + revision >> 4, revision & 0xf); + dma_chan_count = OMAP_LOGICAL_DMA_CH_COUNT; } else { dma_chan_count = 0; return 0; @@ -1049,41 +1304,56 @@ static int __init omap_init_dma(void) memset(&dma_chan, 0, sizeof(dma_chan)); for (ch = 0; ch < dma_chan_count; ch++) { + omap_clear_dma(ch); dma_chan[ch].dev_id = -1; dma_chan[ch].next_lch = -1; if (ch >= 6 && enable_1510_mode) continue; - /* request_irq() doesn't like dev_id (ie. ch) being zero, - * so we have to kludge around this. */ - r = request_irq(dma_irq[ch], dma_irq_handler, 0, "DMA", - (void *) (ch + 1)); + if (cpu_class_is_omap1()) { + /* request_irq() doesn't like dev_id (ie. ch) being + * zero, so we have to kludge around this. */ + r = request_irq(omap1_dma_irq[ch], + omap1_dma_irq_handler, 0, "DMA", + (void *) (ch + 1)); + if (r != 0) { + int i; + + printk(KERN_ERR "unable to request IRQ %d " + "for DMA (error %d)\n", + omap1_dma_irq[ch], r); + for (i = 0; i < ch; i++) + free_irq(omap1_dma_irq[i], + (void *) (i + 1)); + return r; + } + } + } + + if (cpu_is_omap24xx()) + setup_irq(INT_24XX_SDMA_IRQ0, &omap24xx_dma_irq); + + /* FIXME: Update LCD DMA to work on 24xx */ + if (cpu_class_is_omap1()) { + r = request_irq(INT_DMA_LCD, lcd_dma_irq_handler, 0, + "LCD DMA", NULL); if (r != 0) { int i; - printk(KERN_ERR "unable to request IRQ %d for DMA (error %d)\n", - dma_irq[ch], r); - for (i = 0; i < ch; i++) - free_irq(dma_irq[i], (void *) (i + 1)); + printk(KERN_ERR "unable to request IRQ for LCD DMA " + "(error %d)\n", r); + for (i = 0; i < dma_chan_count; i++) + free_irq(omap1_dma_irq[i], (void *) (i + 1)); return r; } } - r = request_irq(INT_DMA_LCD, lcd_dma_irq_handler, 0, "LCD DMA", NULL); - if (r != 0) { - int i; - printk(KERN_ERR "unable to request IRQ for LCD DMA (error %d)\n", r); - for (i = 0; i < dma_chan_count; i++) - free_irq(dma_irq[i], (void *) (i + 1)); - return r; - } return 0; } arch_initcall(omap_init_dma); - EXPORT_SYMBOL(omap_get_dma_src_pos); EXPORT_SYMBOL(omap_get_dma_dst_pos); EXPORT_SYMBOL(omap_get_dma_src_addr_counter); @@ -1109,6 +1379,8 @@ EXPORT_SYMBOL(omap_set_dma_dest_index); EXPORT_SYMBOL(omap_set_dma_dest_data_pack); EXPORT_SYMBOL(omap_set_dma_dest_burst_mode); +EXPORT_SYMBOL(omap_set_dma_params); + EXPORT_SYMBOL(omap_dma_link_lch); EXPORT_SYMBOL(omap_dma_unlink_lch); diff --git a/arch/arm/plat-omap/gpio.c b/arch/arm/plat-omap/gpio.c index 55059a24ad4..76f721d8513 100644 --- a/arch/arm/plat-omap/gpio.c +++ b/arch/arm/plat-omap/gpio.c @@ -140,7 +140,7 @@ static struct gpio_bank gpio_bank_1610[5] = { }; #endif -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX static struct gpio_bank gpio_bank_1510[2] = { { OMAP_MPUIO_BASE, INT_MPUIO, IH_MPUIO_BASE, METHOD_MPUIO }, { OMAP1510_GPIO_BASE, INT_GPIO_BANK1, IH_GPIO_BASE, METHOD_GPIO_1510 } @@ -173,7 +173,7 @@ static int gpio_bank_count; static inline struct gpio_bank *get_gpio_bank(int gpio) { -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX if (cpu_is_omap1510()) { if (OMAP_GPIO_IS_MPUIO(gpio)) return &gpio_bank[0]; @@ -222,7 +222,7 @@ static inline int gpio_valid(int gpio) return -1; return 0; } -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX if (cpu_is_omap1510() && gpio < 16) return 0; #endif @@ -654,7 +654,7 @@ int omap_request_gpio(int gpio) /* Set trigger to none. You need to enable the trigger after request_irq */ _set_gpio_triggering(bank, get_gpio_index(gpio), IRQT_NOEDGE); -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX if (bank->method == METHOD_GPIO_1510) { void __iomem *reg; @@ -739,7 +739,7 @@ static void gpio_irq_handler(unsigned int irq, struct irqdesc *desc, bank = (struct gpio_bank *) desc->data; if (bank->method == METHOD_MPUIO) isr_reg = bank->base + OMAP_MPUIO_GPIO_INT; -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX if (bank->method == METHOD_GPIO_1510) isr_reg = bank->base + OMAP1510_GPIO_INT_STATUS; #endif @@ -774,7 +774,7 @@ static void gpio_irq_handler(unsigned int irq, struct irqdesc *desc, d = irq_desc + gpio_irq; desc_handle_irq(gpio_irq, d, regs); } - } + } } static void gpio_ack_irq(unsigned int irq) @@ -837,8 +837,9 @@ static struct irqchip mpuio_irq_chip = { .unmask = mpuio_unmask_irq }; -static int initialized = 0; -static struct clk * gpio_ck = NULL; +static int initialized; +static struct clk * gpio_ick; +static struct clk * gpio_fck; static int __init _omap_gpio_init(void) { @@ -848,14 +849,26 @@ static int __init _omap_gpio_init(void) initialized = 1; if (cpu_is_omap1510()) { - gpio_ck = clk_get(NULL, "arm_gpio_ck"); - if (IS_ERR(gpio_ck)) + gpio_ick = clk_get(NULL, "arm_gpio_ck"); + if (IS_ERR(gpio_ick)) printk("Could not get arm_gpio_ck\n"); else - clk_use(gpio_ck); + clk_use(gpio_ick); + } + if (cpu_is_omap24xx()) { + gpio_ick = clk_get(NULL, "gpios_ick"); + if (IS_ERR(gpio_ick)) + printk("Could not get gpios_ick\n"); + else + clk_use(gpio_ick); + gpio_fck = clk_get(NULL, "gpios_fck"); + if (IS_ERR(gpio_ick)) + printk("Could not get gpios_fck\n"); + else + clk_use(gpio_fck); } -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX if (cpu_is_omap1510()) { printk(KERN_INFO "OMAP1510 GPIO hardware\n"); gpio_bank_count = 2; @@ -901,7 +914,7 @@ static int __init _omap_gpio_init(void) if (bank->method == METHOD_MPUIO) { omap_writew(0xFFFF, OMAP_MPUIO_BASE + OMAP_MPUIO_GPIO_MASKIT); } -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX if (bank->method == METHOD_GPIO_1510) { __raw_writew(0xffff, bank->base + OMAP1510_GPIO_INT_MASK); __raw_writew(0x0000, bank->base + OMAP1510_GPIO_INT_STATUS); @@ -1038,6 +1051,7 @@ static struct sys_device omap_gpio_device = { /* * This may get called early from board specific init + * for boards that have interrupts routed via FPGA. */ int omap_gpio_init(void) { diff --git a/arch/arm/plat-omap/mcbsp.c b/arch/arm/plat-omap/mcbsp.c index 9c9b7df3faf..ea9475c8665 100644 --- a/arch/arm/plat-omap/mcbsp.c +++ b/arch/arm/plat-omap/mcbsp.c @@ -491,17 +491,20 @@ int omap_mcbsp_xmit_buffer(unsigned int id, dma_addr_t buffer, unsigned int leng omap_set_dma_transfer_params(mcbsp[id].dma_tx_lch, OMAP_DMA_DATA_TYPE_S16, length >> 1, 1, - OMAP_DMA_SYNC_ELEMENT); + OMAP_DMA_SYNC_ELEMENT, + 0, 0); omap_set_dma_dest_params(mcbsp[id].dma_tx_lch, OMAP_DMA_PORT_TIPB, OMAP_DMA_AMODE_CONSTANT, - mcbsp[id].io_base + OMAP_MCBSP_REG_DXR1); + mcbsp[id].io_base + OMAP_MCBSP_REG_DXR1, + 0, 0); omap_set_dma_src_params(mcbsp[id].dma_tx_lch, OMAP_DMA_PORT_EMIFF, OMAP_DMA_AMODE_POST_INC, - buffer); + buffer, + 0, 0); omap_start_dma(mcbsp[id].dma_tx_lch); wait_for_completion(&(mcbsp[id].tx_dma_completion)); @@ -531,17 +534,20 @@ int omap_mcbsp_recv_buffer(unsigned int id, dma_addr_t buffer, unsigned int leng omap_set_dma_transfer_params(mcbsp[id].dma_rx_lch, OMAP_DMA_DATA_TYPE_S16, length >> 1, 1, - OMAP_DMA_SYNC_ELEMENT); + OMAP_DMA_SYNC_ELEMENT, + 0, 0); omap_set_dma_src_params(mcbsp[id].dma_rx_lch, OMAP_DMA_PORT_TIPB, OMAP_DMA_AMODE_CONSTANT, - mcbsp[id].io_base + OMAP_MCBSP_REG_DRR1); + mcbsp[id].io_base + OMAP_MCBSP_REG_DRR1, + 0, 0); omap_set_dma_dest_params(mcbsp[id].dma_rx_lch, OMAP_DMA_PORT_EMIFF, OMAP_DMA_AMODE_POST_INC, - buffer); + buffer, + 0, 0); omap_start_dma(mcbsp[id].dma_rx_lch); wait_for_completion(&(mcbsp[id].rx_dma_completion)); @@ -643,7 +649,7 @@ static const struct omap_mcbsp_info mcbsp_730[] = { }; #endif -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX static const struct omap_mcbsp_info mcbsp_1510[] = { [0] = { .virt_base = OMAP1510_MCBSP1_BASE, .dma_rx_sync = OMAP_DMA_MCBSP1_RX, @@ -712,7 +718,7 @@ static int __init omap_mcbsp_init(void) mcbsp_count = ARRAY_SIZE(mcbsp_730); } #endif -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX if (cpu_is_omap1510()) { mcbsp_info = mcbsp_1510; mcbsp_count = ARRAY_SIZE(mcbsp_1510); diff --git a/arch/arm/plat-omap/mux.c b/arch/arm/plat-omap/mux.c index 64482040f89..8c1c016aa68 100644 --- a/arch/arm/plat-omap/mux.c +++ b/arch/arm/plat-omap/mux.c @@ -3,7 +3,7 @@ * * Utility to set the Omap MUX and PULL_DWN registers from a table in mux.h * - * Copyright (C) 2003 Nokia Corporation + * Copyright (C) 2003 - 2005 Nokia Corporation * * Written by Tony Lindgren <tony.lindgren@nokia.com> * @@ -25,38 +25,74 @@ #include <linux/config.h> #include <linux/module.h> #include <linux/init.h> +#include <linux/kernel.h> #include <asm/system.h> #include <asm/io.h> #include <linux/spinlock.h> - -#define __MUX_C__ #include <asm/arch/mux.h> #ifdef CONFIG_OMAP_MUX +#define OMAP24XX_L4_BASE 0x48000000 +#define OMAP24XX_PULL_ENA (1 << 3) +#define OMAP24XX_PULL_UP (1 << 4) + +static struct pin_config * pin_table; +static unsigned long pin_table_sz; + +extern struct pin_config * omap730_pins; +extern struct pin_config * omap1xxx_pins; +extern struct pin_config * omap24xx_pins; + +int __init omap_mux_register(struct pin_config * pins, unsigned long size) +{ + pin_table = pins; + pin_table_sz = size; + + return 0; +} + /* * Sets the Omap MUX and PULL_DWN registers based on the table */ -int __init_or_module -omap_cfg_reg(const reg_cfg_t reg_cfg) +int __init_or_module omap_cfg_reg(const unsigned long index) { static DEFINE_SPINLOCK(mux_spin_lock); unsigned long flags; - reg_cfg_set *cfg; + struct pin_config *cfg; unsigned int reg_orig = 0, reg = 0, pu_pd_orig = 0, pu_pd = 0, pull_orig = 0, pull = 0; unsigned int mask, warn = 0; - if (cpu_is_omap7xx()) - return 0; + if (!pin_table) + BUG(); - if (reg_cfg > ARRAY_SIZE(reg_cfg_table)) { - printk(KERN_ERR "MUX: reg_cfg %d\n", reg_cfg); - return -EINVAL; + if (index >= pin_table_sz) { + printk(KERN_ERR "Invalid pin mux index: %lu (%lu)\n", + index, pin_table_sz); + dump_stack(); + return -ENODEV; } - cfg = (reg_cfg_set *)®_cfg_table[reg_cfg]; + cfg = (struct pin_config *)&pin_table[index]; + if (cpu_is_omap24xx()) { + u8 reg = 0; + + reg |= cfg->mask & 0x7; + if (cfg->pull_val) + reg |= OMAP24XX_PULL_ENA; + if(cfg->pu_pd_val) + reg |= OMAP24XX_PULL_UP; +#ifdef CONFIG_OMAP_MUX_DEBUG + printk("Muxing %s (0x%08x): 0x%02x -> 0x%02x\n", + cfg->name, OMAP24XX_L4_BASE + cfg->mux_reg, + omap_readb(OMAP24XX_L4_BASE + cfg->mux_reg), reg); +#endif + omap_writeb(reg, OMAP24XX_L4_BASE + cfg->mux_reg); + + return 0; + } /* Check the mux register in question */ if (cfg->mux_reg) { @@ -157,7 +193,8 @@ omap_cfg_reg(const reg_cfg_t reg_cfg) return 0; #endif } - EXPORT_SYMBOL(omap_cfg_reg); - +#else +#define omap_mux_init() do {} while(0) +#define omap_cfg_reg(x) do {} while(0) #endif /* CONFIG_OMAP_MUX */ diff --git a/arch/arm/plat-omap/pm.c b/arch/arm/plat-omap/pm.c index e15c6c1ddec..966cca031ca 100644 --- a/arch/arm/plat-omap/pm.c +++ b/arch/arm/plat-omap/pm.c @@ -54,11 +54,12 @@ #include <asm/arch/tps65010.h> #include <asm/arch/dsp_common.h> -#include "clock.h" -#include "sram.h" +#include <asm/arch/clock.h> +#include <asm/arch/sram.h> static unsigned int arm_sleep_save[ARM_SLEEP_SAVE_SIZE]; static unsigned short ulpd_sleep_save[ULPD_SLEEP_SAVE_SIZE]; +static unsigned int mpui730_sleep_save[MPUI730_SLEEP_SAVE_SIZE]; static unsigned int mpui1510_sleep_save[MPUI1510_SLEEP_SAVE_SIZE]; static unsigned int mpui1610_sleep_save[MPUI1610_SLEEP_SAVE_SIZE]; @@ -120,8 +121,8 @@ void omap_pm_idle(void) */ static void omap_pm_wakeup_setup(void) { - u32 level1_wake = OMAP_IRQ_BIT(INT_IH2_IRQ); - u32 level2_wake = OMAP_IRQ_BIT(INT_UART2) | OMAP_IRQ_BIT(INT_KEYBOARD); + u32 level1_wake = 0; + u32 level2_wake = OMAP_IRQ_BIT(INT_UART2); /* * Turn off all interrupts except GPIO bank 1, L1-2nd level cascade, @@ -129,19 +130,29 @@ static void omap_pm_wakeup_setup(void) * drivers must still separately call omap_set_gpio_wakeup() to * wake up to a GPIO interrupt. */ - if (cpu_is_omap1510() || cpu_is_omap16xx()) - level1_wake |= OMAP_IRQ_BIT(INT_GPIO_BANK1); - else if (cpu_is_omap730()) - level1_wake |= OMAP_IRQ_BIT(INT_730_GPIO_BANK1); + if (cpu_is_omap730()) + level1_wake = OMAP_IRQ_BIT(INT_730_GPIO_BANK1) | + OMAP_IRQ_BIT(INT_730_IH2_IRQ); + else if (cpu_is_omap1510()) + level1_wake = OMAP_IRQ_BIT(INT_GPIO_BANK1) | + OMAP_IRQ_BIT(INT_1510_IH2_IRQ); + else if (cpu_is_omap16xx()) + level1_wake = OMAP_IRQ_BIT(INT_GPIO_BANK1) | + OMAP_IRQ_BIT(INT_1610_IH2_IRQ); omap_writel(~level1_wake, OMAP_IH1_MIR); - if (cpu_is_omap1510()) + if (cpu_is_omap730()) { + omap_writel(~level2_wake, OMAP_IH2_0_MIR); + omap_writel(~(OMAP_IRQ_BIT(INT_730_WAKE_UP_REQ) | OMAP_IRQ_BIT(INT_730_MPUIO_KEYPAD)), OMAP_IH2_1_MIR); + } else if (cpu_is_omap1510()) { + level2_wake |= OMAP_IRQ_BIT(INT_KEYBOARD); omap_writel(~level2_wake, OMAP_IH2_MIR); - - /* INT_1610_WAKE_UP_REQ is needed for GPIO wakeup... */ - if (cpu_is_omap16xx()) { + } else if (cpu_is_omap16xx()) { + level2_wake |= OMAP_IRQ_BIT(INT_KEYBOARD); omap_writel(~level2_wake, OMAP_IH2_0_MIR); + + /* INT_1610_WAKE_UP_REQ is needed for GPIO wakeup... */ omap_writel(~OMAP_IRQ_BIT(INT_1610_WAKE_UP_REQ), OMAP_IH2_1_MIR); omap_writel(~0x0, OMAP_IH2_2_MIR); omap_writel(~0x0, OMAP_IH2_3_MIR); @@ -185,7 +196,17 @@ void omap_pm_suspend(void) * Save interrupt, MPUI, ARM and UPLD control registers. */ - if (cpu_is_omap1510()) { + if (cpu_is_omap730()) { + MPUI730_SAVE(OMAP_IH1_MIR); + MPUI730_SAVE(OMAP_IH2_0_MIR); + MPUI730_SAVE(OMAP_IH2_1_MIR); + MPUI730_SAVE(MPUI_CTRL); + MPUI730_SAVE(MPUI_DSP_BOOT_CONFIG); + MPUI730_SAVE(MPUI_DSP_API_CONFIG); + MPUI730_SAVE(EMIFS_CONFIG); + MPUI730_SAVE(EMIFF_SDRAM_CONFIG); + + } else if (cpu_is_omap1510()) { MPUI1510_SAVE(OMAP_IH1_MIR); MPUI1510_SAVE(OMAP_IH2_MIR); MPUI1510_SAVE(MPUI_CTRL); @@ -280,7 +301,13 @@ void omap_pm_suspend(void) ULPD_RESTORE(ULPD_CLOCK_CTRL); ULPD_RESTORE(ULPD_STATUS_REQ); - if (cpu_is_omap1510()) { + if (cpu_is_omap730()) { + MPUI730_RESTORE(EMIFS_CONFIG); + MPUI730_RESTORE(EMIFF_SDRAM_CONFIG); + MPUI730_RESTORE(OMAP_IH1_MIR); + MPUI730_RESTORE(OMAP_IH2_0_MIR); + MPUI730_RESTORE(OMAP_IH2_1_MIR); + } else if (cpu_is_omap1510()) { MPUI1510_RESTORE(MPUI_CTRL); MPUI1510_RESTORE(MPUI_DSP_BOOT_CONFIG); MPUI1510_RESTORE(MPUI_DSP_API_CONFIG); @@ -355,7 +382,14 @@ static int omap_pm_read_proc( ULPD_SAVE(ULPD_DPLL_CTRL); ULPD_SAVE(ULPD_POWER_CTRL); - if (cpu_is_omap1510()) { + if (cpu_is_omap730()) { + MPUI730_SAVE(MPUI_CTRL); + MPUI730_SAVE(MPUI_DSP_STATUS); + MPUI730_SAVE(MPUI_DSP_BOOT_CONFIG); + MPUI730_SAVE(MPUI_DSP_API_CONFIG); + MPUI730_SAVE(EMIFF_SDRAM_CONFIG); + MPUI730_SAVE(EMIFS_CONFIG); + } else if (cpu_is_omap1510()) { MPUI1510_SAVE(MPUI_CTRL); MPUI1510_SAVE(MPUI_DSP_STATUS); MPUI1510_SAVE(MPUI_DSP_BOOT_CONFIG); @@ -404,7 +438,21 @@ static int omap_pm_read_proc( ULPD_SHOW(ULPD_STATUS_REQ), ULPD_SHOW(ULPD_POWER_CTRL)); - if (cpu_is_omap1510()) { + if (cpu_is_omap730()) { + my_buffer_offset += sprintf(my_base + my_buffer_offset, + "MPUI730_CTRL_REG 0x%-8x \n" + "MPUI730_DSP_STATUS_REG: 0x%-8x \n" + "MPUI730_DSP_BOOT_CONFIG_REG: 0x%-8x \n" + "MPUI730_DSP_API_CONFIG_REG: 0x%-8x \n" + "MPUI730_SDRAM_CONFIG_REG: 0x%-8x \n" + "MPUI730_EMIFS_CONFIG_REG: 0x%-8x \n", + MPUI730_SHOW(MPUI_CTRL), + MPUI730_SHOW(MPUI_DSP_STATUS), + MPUI730_SHOW(MPUI_DSP_BOOT_CONFIG), + MPUI730_SHOW(MPUI_DSP_API_CONFIG), + MPUI730_SHOW(EMIFF_SDRAM_CONFIG), + MPUI730_SHOW(EMIFS_CONFIG)); + } else if (cpu_is_omap1510()) { my_buffer_offset += sprintf(my_base + my_buffer_offset, "MPUI1510_CTRL_REG 0x%-8x \n" "MPUI1510_DSP_STATUS_REG: 0x%-8x \n" @@ -553,7 +601,12 @@ static int __init omap_pm_init(void) * These routines need to be in SRAM as that's the only * memory the MPU can see when it wakes up. */ - if (cpu_is_omap1510()) { + if (cpu_is_omap730()) { + omap_sram_idle = omap_sram_push(omap730_idle_loop_suspend, + omap730_idle_loop_suspend_sz); + omap_sram_suspend = omap_sram_push(omap730_cpu_suspend, + omap730_cpu_suspend_sz); + } else if (cpu_is_omap1510()) { omap_sram_idle = omap_sram_push(omap1510_idle_loop_suspend, omap1510_idle_loop_suspend_sz); omap_sram_suspend = omap_sram_push(omap1510_cpu_suspend, @@ -572,7 +625,11 @@ static int __init omap_pm_init(void) pm_idle = omap_pm_idle; - setup_irq(INT_1610_WAKE_UP_REQ, &omap_wakeup_irq); + if (cpu_is_omap730()) + setup_irq(INT_730_WAKE_UP_REQ, &omap_wakeup_irq); + else if (cpu_is_omap16xx()) + setup_irq(INT_1610_WAKE_UP_REQ, &omap_wakeup_irq); + #if 0 /* --- BEGIN BOARD-DEPENDENT CODE --- */ /* Sleepx mask direction */ @@ -591,7 +648,9 @@ static int __init omap_pm_init(void) omap_writew(ULPD_POWER_CTRL_REG_VAL, ULPD_POWER_CTRL); /* Configure IDLECT3 */ - if (cpu_is_omap16xx()) + if (cpu_is_omap730()) + omap_writel(OMAP730_IDLECT3_VAL, OMAP730_IDLECT3); + else if (cpu_is_omap16xx()) omap_writel(OMAP1610_IDLECT3_VAL, OMAP1610_IDLECT3); pm_set_ops(&omap_pm_ops); @@ -600,8 +659,10 @@ static int __init omap_pm_init(void) omap_pm_init_proc(); #endif - /* configure LOW_PWR pin */ - omap_cfg_reg(T20_1610_LOW_PWR); + if (cpu_is_omap16xx()) { + /* configure LOW_PWR pin */ + omap_cfg_reg(T20_1610_LOW_PWR); + } return 0; } diff --git a/arch/arm/plat-omap/sleep.S b/arch/arm/plat-omap/sleep.S index 9f745836f6a..4cd7d292f85 100644 --- a/arch/arm/plat-omap/sleep.S +++ b/arch/arm/plat-omap/sleep.S @@ -1,7 +1,7 @@ /* * linux/arch/arm/plat-omap/sleep.S * - * Low-level OMAP1510/1610 sleep/wakeUp support + * Low-level OMAP730/1510/1610 sleep/wakeUp support * * Initial SA1110 code: * Copyright (c) 2001 Cliff Brake <cbrake@accelent.com> @@ -52,7 +52,57 @@ * processor specific functions here. */ -#ifdef CONFIG_ARCH_OMAP1510 +#if defined(CONFIG_ARCH_OMAP730) +ENTRY(omap730_idle_loop_suspend) + + stmfd sp!, {r0 - r12, lr} @ save registers on stack + + @ load base address of ARM_IDLECT1 and ARM_IDLECT2 + mov r4, #CLKGEN_REG_ASM_BASE & 0xff000000 + orr r4, r4, #CLKGEN_REG_ASM_BASE & 0x00ff0000 + orr r4, r4, #CLKGEN_REG_ASM_BASE & 0x0000ff00 + + @ turn off clock domains + @ get ARM_IDLECT2 into r2 + ldrh r2, [r4, #ARM_IDLECT2_ASM_OFFSET & 0xff] + mov r5, #OMAP730_IDLECT2_SLEEP_VAL & 0xff + orr r5, r5, #OMAP730_IDLECT2_SLEEP_VAL & 0xff00 + strh r5, [r4, #ARM_IDLECT2_ASM_OFFSET & 0xff] + + @ request ARM idle + @ get ARM_IDLECT1 into r1 + ldrh r1, [r4, #ARM_IDLECT1_ASM_OFFSET & 0xff] + orr r3, r1, #OMAP730_IDLE_LOOP_REQUEST & 0xffff + strh r3, [r4, #ARM_IDLECT1_ASM_OFFSET & 0xff] + + mov r5, #IDLE_WAIT_CYCLES & 0xff + orr r5, r5, #IDLE_WAIT_CYCLES & 0xff00 +l_730: subs r5, r5, #1 + bne l_730 +/* + * Let's wait for the next clock tick to wake us up. + */ + mov r0, #0 + mcr p15, 0, r0, c7, c0, 4 @ wait for interrupt +/* + * omap730_idle_loop_suspend()'s resume point. + * + * It will just start executing here, so we'll restore stuff from the + * stack, reset the ARM_IDLECT1 and ARM_IDLECT2. + */ + + @ restore ARM_IDLECT1 and ARM_IDLECT2 and return + @ r1 has ARM_IDLECT1 and r2 still has ARM_IDLECT2 + strh r2, [r4, #ARM_IDLECT2_ASM_OFFSET & 0xff] + strh r1, [r4, #ARM_IDLECT1_ASM_OFFSET & 0xff] + + ldmfd sp!, {r0 - r12, pc} @ restore regs and return + +ENTRY(omap730_idle_loop_suspend_sz) + .word . - omap730_idle_loop_suspend +#endif /* CONFIG_ARCH_OMAP730 */ + +#ifdef CONFIG_ARCH_OMAP15XX ENTRY(omap1510_idle_loop_suspend) stmfd sp!, {r0 - r12, lr} @ save registers on stack @@ -100,7 +150,7 @@ l_1510: subs r5, r5, #1 ENTRY(omap1510_idle_loop_suspend_sz) .word . - omap1510_idle_loop_suspend -#endif /* CONFIG_ARCH_OMAP1510 */ +#endif /* CONFIG_ARCH_OMAP15XX */ #if defined(CONFIG_ARCH_OMAP16XX) ENTRY(omap1610_idle_loop_suspend) @@ -169,7 +219,86 @@ ENTRY(omap1610_idle_loop_suspend_sz) * */ -#ifdef CONFIG_ARCH_OMAP1510 +#if defined(CONFIG_ARCH_OMAP730) +ENTRY(omap730_cpu_suspend) + + @ save registers on stack + stmfd sp!, {r0 - r12, lr} + + @ Drain write cache + mov r4, #0 + mcr p15, 0, r0, c7, c10, 4 + nop + + @ load base address of Traffic Controller + mov r6, #TCMIF_ASM_BASE & 0xff000000 + orr r6, r6, #TCMIF_ASM_BASE & 0x00ff0000 + orr r6, r6, #TCMIF_ASM_BASE & 0x0000ff00 + + @ prepare to put SDRAM into self-refresh manually + ldr r7, [r6, #EMIFF_SDRAM_CONFIG_ASM_OFFSET & 0xff] + orr r9, r7, #SELF_REFRESH_MODE & 0xff000000 + orr r9, r9, #SELF_REFRESH_MODE & 0x000000ff + str r9, [r6, #EMIFF_SDRAM_CONFIG_ASM_OFFSET & 0xff] + + @ prepare to put EMIFS to Sleep + ldr r8, [r6, #EMIFS_CONFIG_ASM_OFFSET & 0xff] + orr r9, r8, #IDLE_EMIFS_REQUEST & 0xff + str r9, [r6, #EMIFS_CONFIG_ASM_OFFSET & 0xff] + + @ load base address of ARM_IDLECT1 and ARM_IDLECT2 + mov r4, #CLKGEN_REG_ASM_BASE & 0xff000000 + orr r4, r4, #CLKGEN_REG_ASM_BASE & 0x00ff0000 + orr r4, r4, #CLKGEN_REG_ASM_BASE & 0x0000ff00 + + @ turn off clock domains + @ do not disable PERCK (0x04) + mov r5, #OMAP730_IDLECT2_SLEEP_VAL & 0xff + orr r5, r5, #OMAP730_IDLECT2_SLEEP_VAL & 0xff00 + strh r5, [r4, #ARM_IDLECT2_ASM_OFFSET & 0xff] + + @ request ARM idle + mov r3, #OMAP730_IDLECT1_SLEEP_VAL & 0xff + orr r3, r3, #OMAP730_IDLECT1_SLEEP_VAL & 0xff00 + strh r3, [r4, #ARM_IDLECT1_ASM_OFFSET & 0xff] + + @ disable instruction cache + mrc p15, 0, r9, c1, c0, 0 + bic r2, r9, #0x1000 + mcr p15, 0, r2, c1, c0, 0 + nop + +/* + * Let's wait for the next wake up event to wake us up. r0 can't be + * used here because r0 holds ARM_IDLECT1 + */ + mov r2, #0 + mcr p15, 0, r2, c7, c0, 4 @ wait for interrupt +/* + * omap730_cpu_suspend()'s resume point. + * + * It will just start executing here, so we'll restore stuff from the + * stack. + */ + @ re-enable Icache + mcr p15, 0, r9, c1, c0, 0 + + @ reset the ARM_IDLECT1 and ARM_IDLECT2. + strh r1, [r4, #ARM_IDLECT2_ASM_OFFSET & 0xff] + strh r0, [r4, #ARM_IDLECT1_ASM_OFFSET & 0xff] + + @ Restore EMIFF controls + str r7, [r6, #EMIFF_SDRAM_CONFIG_ASM_OFFSET & 0xff] + str r8, [r6, #EMIFS_CONFIG_ASM_OFFSET & 0xff] + + @ restore regs and return + ldmfd sp!, {r0 - r12, pc} + +ENTRY(omap730_cpu_suspend_sz) + .word . - omap730_cpu_suspend +#endif /* CONFIG_ARCH_OMAP730 */ + +#ifdef CONFIG_ARCH_OMAP15XX ENTRY(omap1510_cpu_suspend) @ save registers on stack @@ -241,7 +370,7 @@ l_1510_2: ENTRY(omap1510_cpu_suspend_sz) .word . - omap1510_cpu_suspend -#endif /* CONFIG_ARCH_OMAP1510 */ +#endif /* CONFIG_ARCH_OMAP15XX */ #if defined(CONFIG_ARCH_OMAP16XX) ENTRY(omap1610_cpu_suspend) diff --git a/arch/arm/plat-omap/sram.c b/arch/arm/plat-omap/sram.c index 7ad69f14a3e..792f6637583 100644 --- a/arch/arm/plat-omap/sram.c +++ b/arch/arm/plat-omap/sram.c @@ -20,10 +20,13 @@ #include <asm/io.h> #include <asm/cacheflush.h> -#include "sram.h" +#include <asm/arch/sram.h> + +#define OMAP1_SRAM_PA 0x20000000 +#define OMAP1_SRAM_VA 0xd0000000 +#define OMAP2_SRAM_PA 0x40200000 +#define OMAP2_SRAM_VA 0xd0000000 -#define OMAP1_SRAM_BASE 0xd0000000 -#define OMAP1_SRAM_START 0x20000000 #define SRAM_BOOTLOADER_SZ 0x80 static unsigned long omap_sram_base; @@ -31,37 +34,40 @@ static unsigned long omap_sram_size; static unsigned long omap_sram_ceil; /* - * The amount of SRAM depends on the core type: - * 730 = 200K, 1510 = 512K, 5912 = 256K, 1610 = 16K, 1710 = 16K + * The amount of SRAM depends on the core type. * Note that we cannot try to test for SRAM here because writes * to secure SRAM will hang the system. Also the SRAM is not * yet mapped at this point. */ void __init omap_detect_sram(void) { - omap_sram_base = OMAP1_SRAM_BASE; + if (!cpu_is_omap24xx()) + omap_sram_base = OMAP1_SRAM_VA; + else + omap_sram_base = OMAP2_SRAM_VA; if (cpu_is_omap730()) - omap_sram_size = 0x32000; - else if (cpu_is_omap1510()) - omap_sram_size = 0x80000; + omap_sram_size = 0x32000; /* 200K */ + else if (cpu_is_omap15xx()) + omap_sram_size = 0x30000; /* 192K */ else if (cpu_is_omap1610() || cpu_is_omap1621() || cpu_is_omap1710()) - omap_sram_size = 0x4000; + omap_sram_size = 0x4000; /* 16K */ else if (cpu_is_omap1611()) - omap_sram_size = 0x3e800; + omap_sram_size = 0x3e800; /* 250K */ + else if (cpu_is_omap2420()) + omap_sram_size = 0xa0014; /* 640K */ else { printk(KERN_ERR "Could not detect SRAM size\n"); omap_sram_size = 0x4000; } - printk(KERN_INFO "SRAM size: 0x%lx\n", omap_sram_size); omap_sram_ceil = omap_sram_base + omap_sram_size; } static struct map_desc omap_sram_io_desc[] __initdata = { { /* .length gets filled in at runtime */ - .virtual = OMAP1_SRAM_BASE, - .pfn = __phys_to_pfn(OMAP1_SRAM_START), + .virtual = OMAP1_SRAM_VA, + .pfn = __phys_to_pfn(OMAP1_SRAM_PA), .type = MT_DEVICE } }; @@ -76,10 +82,19 @@ void __init omap_map_sram(void) if (omap_sram_size == 0) return; + if (cpu_is_omap24xx()) { + omap_sram_io_desc[0].virtual = OMAP2_SRAM_VA; + omap_sram_io_desc[0].pfn = __phys_to_pfn(OMAP2_SRAM_PA); + } + omap_sram_io_desc[0].length = (omap_sram_size + PAGE_SIZE-1)/PAGE_SIZE; omap_sram_io_desc[0].length *= PAGE_SIZE; iotable_init(omap_sram_io_desc, ARRAY_SIZE(omap_sram_io_desc)); + printk(KERN_INFO "SRAM: Mapped pa 0x%08lx to va 0x%08lx size: 0x%lx\n", + omap_sram_io_desc[0].pfn, omap_sram_io_desc[0].virtual, + omap_sram_io_desc[0].length); + /* * Looks like we need to preserve some bootloader code at the * beginning of SRAM for jumping to flash for reboot to work... @@ -88,16 +103,6 @@ void __init omap_map_sram(void) omap_sram_size - SRAM_BOOTLOADER_SZ); } -static void (*_omap_sram_reprogram_clock)(u32 dpllctl, u32 ckctl) = NULL; - -void omap_sram_reprogram_clock(u32 dpllctl, u32 ckctl) -{ - if (_omap_sram_reprogram_clock == NULL) - panic("Cannot use SRAM"); - - return _omap_sram_reprogram_clock(dpllctl, ckctl); -} - void * omap_sram_push(void * start, unsigned long size) { if (size > (omap_sram_ceil - (omap_sram_base + SRAM_BOOTLOADER_SZ))) { @@ -111,10 +116,94 @@ void * omap_sram_push(void * start, unsigned long size) return (void *)omap_sram_ceil; } -void __init omap_sram_init(void) +static void omap_sram_error(void) +{ + panic("Uninitialized SRAM function\n"); +} + +#ifdef CONFIG_ARCH_OMAP1 + +static void (*_omap_sram_reprogram_clock)(u32 dpllctl, u32 ckctl); + +void omap_sram_reprogram_clock(u32 dpllctl, u32 ckctl) +{ + if (!_omap_sram_reprogram_clock) + omap_sram_error(); + + return _omap_sram_reprogram_clock(dpllctl, ckctl); +} + +int __init omap1_sram_init(void) { - omap_detect_sram(); - omap_map_sram(); _omap_sram_reprogram_clock = omap_sram_push(sram_reprogram_clock, sram_reprogram_clock_sz); + + return 0; +} + +#else +#define omap1_sram_init() do {} while (0) +#endif + +#ifdef CONFIG_ARCH_OMAP2 + +static void (*_omap2_sram_ddr_init)(u32 *slow_dll_ctrl, u32 fast_dll_ctrl, + u32 base_cs, u32 force_unlock); + +void omap2_sram_ddr_init(u32 *slow_dll_ctrl, u32 fast_dll_ctrl, + u32 base_cs, u32 force_unlock) +{ + if (!_omap2_sram_ddr_init) + omap_sram_error(); + + return _omap2_sram_ddr_init(slow_dll_ctrl, fast_dll_ctrl, + base_cs, force_unlock); +} + +static void (*_omap2_sram_reprogram_sdrc)(u32 perf_level, u32 dll_val, + u32 mem_type); + +void omap2_sram_reprogram_sdrc(u32 perf_level, u32 dll_val, u32 mem_type) +{ + if (!_omap2_sram_reprogram_sdrc) + omap_sram_error(); + + return _omap2_sram_reprogram_sdrc(perf_level, dll_val, mem_type); +} + +static u32 (*_omap2_set_prcm)(u32 dpll_ctrl_val, u32 sdrc_rfr_val, int bypass); + +u32 omap2_set_prcm(u32 dpll_ctrl_val, u32 sdrc_rfr_val, int bypass) +{ + if (!_omap2_set_prcm) + omap_sram_error(); + + return _omap2_set_prcm(dpll_ctrl_val, sdrc_rfr_val, bypass); +} + +int __init omap2_sram_init(void) +{ + _omap2_sram_ddr_init = omap_sram_push(sram_ddr_init, sram_ddr_init_sz); + + _omap2_sram_reprogram_sdrc = omap_sram_push(sram_reprogram_sdrc, + sram_reprogram_sdrc_sz); + _omap2_set_prcm = omap_sram_push(sram_set_prcm, sram_set_prcm_sz); + + return 0; +} +#else +#define omap2_sram_init() do {} while (0) +#endif + +int __init omap_sram_init(void) +{ + omap_detect_sram(); + omap_map_sram(); + + if (!cpu_is_omap24xx()) + omap1_sram_init(); + else + omap2_sram_init(); + + return 0; } diff --git a/arch/arm/plat-omap/sram.h b/arch/arm/plat-omap/sram.h deleted file mode 100644 index 71984efa6ae..00000000000 --- a/arch/arm/plat-omap/sram.h +++ /dev/null @@ -1,21 +0,0 @@ -/* - * linux/arch/arm/plat-omap/sram.h - * - * Interface for functions that need to be run in internal SRAM - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - */ - -#ifndef __ARCH_ARM_OMAP_SRAM_H -#define __ARCH_ARM_OMAP_SRAM_H - -extern void * omap_sram_push(void * start, unsigned long size); -extern void omap_sram_reprogram_clock(u32 dpllctl, u32 ckctl); - -/* Do not use these */ -extern void sram_reprogram_clock(u32 ckctl, u32 dpllctl); -extern unsigned long sram_reprogram_clock_sz; - -#endif diff --git a/arch/arm/plat-omap/usb.c b/arch/arm/plat-omap/usb.c index 205e2d0b826..00afc7a8c2a 100644 --- a/arch/arm/plat-omap/usb.c +++ b/arch/arm/plat-omap/usb.c @@ -91,6 +91,8 @@ EXPORT_SYMBOL(otg_set_transceiver); /*-------------------------------------------------------------------------*/ +#if defined(CONFIG_ARCH_OMAP_OTG) || defined(CONFIG_ARCH_OMAP15XX) + static u32 __init omap_usb0_init(unsigned nwires, unsigned is_device) { u32 syscon1 = 0; @@ -271,6 +273,8 @@ static u32 __init omap_usb2_init(unsigned nwires, unsigned alt_pingroup) return syscon1 << 24; } +#endif + /*-------------------------------------------------------------------------*/ #if defined(CONFIG_USB_GADGET_OMAP) || \ @@ -494,7 +498,7 @@ static inline void omap_otg_init(struct omap_usb_config *config) {} /*-------------------------------------------------------------------------*/ -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX #define ULPD_DPLL_CTRL_REG __REG16(ULPD_DPLL_CTRL) #define DPLL_IOB (1 << 13) @@ -507,7 +511,6 @@ static inline void omap_otg_init(struct omap_usb_config *config) {} static void __init omap_1510_usb_init(struct omap_usb_config *config) { - int status; unsigned int val; omap_usb0_init(config->pins[0], is_usb0_device(config)); @@ -539,6 +542,8 @@ static void __init omap_1510_usb_init(struct omap_usb_config *config) #ifdef CONFIG_USB_GADGET_OMAP if (config->register_dev) { + int status; + udc_device.dev.platform_data = config; status = platform_device_register(&udc_device); if (status) @@ -549,6 +554,8 @@ static void __init omap_1510_usb_init(struct omap_usb_config *config) #if defined(CONFIG_USB_OHCI_HCD) || defined(CONFIG_USB_OHCI_HCD_MODULE) if (config->register_host) { + int status; + ohci_device.dev.platform_data = config; status = platform_device_register(&ohci_device); if (status) diff --git a/arch/i386/pci/fixup.c b/arch/i386/pci/fixup.c index 3984226a8b9..eeb1b1f2d54 100644 --- a/arch/i386/pci/fixup.c +++ b/arch/i386/pci/fixup.c @@ -433,9 +433,8 @@ static void __devinit pci_post_fixup_toshiba_ohci1394(struct pci_dev *dev) return; /* only applies to certain Toshibas (so far) */ /* Restore config space on Toshiba laptops */ - mdelay(10); pci_write_config_word(dev, PCI_CACHE_LINE_SIZE, toshiba_line_size); - pci_write_config_word(dev, PCI_INTERRUPT_LINE, dev->irq); + pci_read_config_byte(dev, PCI_INTERRUPT_LINE, (u8 *)&dev->irq); pci_write_config_dword(dev, PCI_BASE_ADDRESS_0, pci_resource_start(dev, 0)); pci_write_config_dword(dev, PCI_BASE_ADDRESS_1, diff --git a/arch/ia64/kernel/efi.c b/arch/ia64/kernel/efi.c index f72ea6aebcb..a3aa45cbcfa 100644 --- a/arch/ia64/kernel/efi.c +++ b/arch/ia64/kernel/efi.c @@ -987,7 +987,7 @@ efi_initialize_iomem_resources(struct resource *code_resource, break; } - if ((res = kcalloc(1, sizeof(struct resource), GFP_KERNEL)) == NULL) { + if ((res = kzalloc(sizeof(struct resource), GFP_KERNEL)) == NULL) { printk(KERN_ERR "failed to alocate resource for iomem\n"); return; } diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c index 96736a119c9..801eeaeaf3d 100644 --- a/arch/ia64/kernel/kprobes.c +++ b/arch/ia64/kernel/kprobes.c @@ -347,7 +347,7 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs) ((struct fnptr *)kretprobe_trampoline)->ip; spin_lock_irqsave(&kretprobe_lock, flags); - head = kretprobe_inst_table_head(current); + head = kretprobe_inst_table_head(current); /* * It is possible to have multiple instances associated with a given @@ -363,9 +363,9 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs) * kretprobe_trampoline */ hlist_for_each_entry_safe(ri, node, tmp, head, hlist) { - if (ri->task != current) + if (ri->task != current) /* another task is sharing our hash bucket */ - continue; + continue; if (ri->rp && ri->rp->handler) ri->rp->handler(ri, regs); @@ -394,7 +394,7 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs) * kprobe_handler() that we don't want the post_handler * to run (and have re-enabled preemption) */ - return 1; + return 1; } /* Called with kretprobe_lock held */ @@ -739,12 +739,16 @@ int __kprobes kprobe_exceptions_notify(struct notifier_block *self, switch(val) { case DIE_BREAK: - if (pre_kprobes_handler(args)) - ret = NOTIFY_STOP; + /* err is break number from ia64_bad_break() */ + if (args->err == 0x80200 || args->err == 0x80300) + if (pre_kprobes_handler(args)) + ret = NOTIFY_STOP; break; - case DIE_SS: - if (post_kprobes_handler(args->regs)) - ret = NOTIFY_STOP; + case DIE_FAULT: + /* err is vector number from ia64_fault() */ + if (args->err == 36) + if (post_kprobes_handler(args->regs)) + ret = NOTIFY_STOP; break; case DIE_PAGE_FAULT: /* kprobe_running() needs smp_processor_id() */ diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c index 52c47da1724..355af15287c 100644 --- a/arch/ia64/kernel/mca.c +++ b/arch/ia64/kernel/mca.c @@ -51,6 +51,9 @@ * * 2005-08-12 Keith Owens <kaos@sgi.com> * Convert MCA/INIT handlers to use per event stacks and SAL/OS state. + * + * 2005-10-07 Keith Owens <kaos@sgi.com> + * Add notify_die() hooks. */ #include <linux/config.h> #include <linux/types.h> @@ -58,7 +61,6 @@ #include <linux/sched.h> #include <linux/interrupt.h> #include <linux/irq.h> -#include <linux/kallsyms.h> #include <linux/smp_lock.h> #include <linux/bootmem.h> #include <linux/acpi.h> @@ -69,6 +71,7 @@ #include <linux/workqueue.h> #include <asm/delay.h> +#include <asm/kdebug.h> #include <asm/machvec.h> #include <asm/meminit.h> #include <asm/page.h> @@ -132,6 +135,14 @@ extern void salinfo_log_wakeup(int type, u8 *buffer, u64 size, int irqsafe); static int mca_init; + +static void inline +ia64_mca_spin(const char *func) +{ + printk(KERN_EMERG "%s: spinning here, not returning to SAL\n", func); + while (1) + cpu_relax(); +} /* * IA64_MCA log support */ @@ -526,13 +537,16 @@ ia64_mca_wakeup_all(void) * Outputs : None */ static irqreturn_t -ia64_mca_rendez_int_handler(int rendez_irq, void *arg, struct pt_regs *ptregs) +ia64_mca_rendez_int_handler(int rendez_irq, void *arg, struct pt_regs *regs) { unsigned long flags; int cpu = smp_processor_id(); /* Mask all interrupts */ local_irq_save(flags); + if (notify_die(DIE_MCA_RENDZVOUS_ENTER, "MCA", regs, 0, 0, 0) + == NOTIFY_STOP) + ia64_mca_spin(__FUNCTION__); ia64_mc_info.imi_rendez_checkin[cpu] = IA64_MCA_RENDEZ_CHECKIN_DONE; /* Register with the SAL monarch that the slave has @@ -540,10 +554,18 @@ ia64_mca_rendez_int_handler(int rendez_irq, void *arg, struct pt_regs *ptregs) */ ia64_sal_mc_rendez(); + if (notify_die(DIE_MCA_RENDZVOUS_PROCESS, "MCA", regs, 0, 0, 0) + == NOTIFY_STOP) + ia64_mca_spin(__FUNCTION__); + /* Wait for the monarch cpu to exit. */ while (monarch_cpu != -1) cpu_relax(); /* spin until monarch leaves */ + if (notify_die(DIE_MCA_RENDZVOUS_LEAVE, "MCA", regs, 0, 0, 0) + == NOTIFY_STOP) + ia64_mca_spin(__FUNCTION__); + /* Enable all interrupts */ local_irq_restore(flags); return IRQ_HANDLED; @@ -933,6 +955,9 @@ ia64_mca_handler(struct pt_regs *regs, struct switch_stack *sw, oops_in_progress = 1; /* FIXME: make printk NMI/MCA/INIT safe */ previous_current = ia64_mca_modify_original_stack(regs, sw, sos, "MCA"); monarch_cpu = cpu; + if (notify_die(DIE_MCA_MONARCH_ENTER, "MCA", regs, 0, 0, 0) + == NOTIFY_STOP) + ia64_mca_spin(__FUNCTION__); ia64_wait_for_slaves(cpu); /* Wakeup all the processors which are spinning in the rendezvous loop. @@ -942,6 +967,9 @@ ia64_mca_handler(struct pt_regs *regs, struct switch_stack *sw, * spinning in SAL does not work. */ ia64_mca_wakeup_all(); + if (notify_die(DIE_MCA_MONARCH_PROCESS, "MCA", regs, 0, 0, 0) + == NOTIFY_STOP) + ia64_mca_spin(__FUNCTION__); /* Get the MCA error record and log it */ ia64_mca_log_sal_error_record(SAL_INFO_TYPE_MCA); @@ -960,6 +988,9 @@ ia64_mca_handler(struct pt_regs *regs, struct switch_stack *sw, ia64_sal_clear_state_info(SAL_INFO_TYPE_MCA); sos->os_status = IA64_MCA_CORRECTED; } + if (notify_die(DIE_MCA_MONARCH_LEAVE, "MCA", regs, 0, 0, recover) + == NOTIFY_STOP) + ia64_mca_spin(__FUNCTION__); set_curr_task(cpu, previous_current); monarch_cpu = -1; @@ -1188,6 +1219,37 @@ ia64_mca_cpe_poll (unsigned long dummy) #endif /* CONFIG_ACPI */ +static int +default_monarch_init_process(struct notifier_block *self, unsigned long val, void *data) +{ + int c; + struct task_struct *g, *t; + if (val != DIE_INIT_MONARCH_PROCESS) + return NOTIFY_DONE; + printk(KERN_ERR "Processes interrupted by INIT -"); + for_each_online_cpu(c) { + struct ia64_sal_os_state *s; + t = __va(__per_cpu_mca[c] + IA64_MCA_CPU_INIT_STACK_OFFSET); + s = (struct ia64_sal_os_state *)((char *)t + MCA_SOS_OFFSET); + g = s->prev_task; + if (g) { + if (g->pid) + printk(" %d", g->pid); + else + printk(" %d (cpu %d task 0x%p)", g->pid, task_cpu(g), g); + } + } + printk("\n\n"); + if (read_trylock(&tasklist_lock)) { + do_each_thread (g, t) { + printk("\nBacktrace of pid %d (%s)\n", t->pid, t->comm); + show_stack(t, NULL); + } while_each_thread (g, t); + read_unlock(&tasklist_lock); + } + return NOTIFY_DONE; +} + /* * C portion of the OS INIT handler * @@ -1212,8 +1274,7 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw, static atomic_t slaves; static atomic_t monarchs; task_t *previous_current; - int cpu = smp_processor_id(), c; - struct task_struct *g, *t; + int cpu = smp_processor_id(); oops_in_progress = 1; /* FIXME: make printk NMI/MCA/INIT safe */ console_loglevel = 15; /* make sure printks make it to console */ @@ -1253,8 +1314,17 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw, ia64_mc_info.imi_rendez_checkin[cpu] = IA64_MCA_RENDEZ_CHECKIN_INIT; while (monarch_cpu == -1) cpu_relax(); /* spin until monarch enters */ + if (notify_die(DIE_INIT_SLAVE_ENTER, "INIT", regs, 0, 0, 0) + == NOTIFY_STOP) + ia64_mca_spin(__FUNCTION__); + if (notify_die(DIE_INIT_SLAVE_PROCESS, "INIT", regs, 0, 0, 0) + == NOTIFY_STOP) + ia64_mca_spin(__FUNCTION__); while (monarch_cpu != -1) cpu_relax(); /* spin until monarch leaves */ + if (notify_die(DIE_INIT_SLAVE_LEAVE, "INIT", regs, 0, 0, 0) + == NOTIFY_STOP) + ia64_mca_spin(__FUNCTION__); printk("Slave on cpu %d returning to normal service.\n", cpu); set_curr_task(cpu, previous_current); ia64_mc_info.imi_rendez_checkin[cpu] = IA64_MCA_RENDEZ_CHECKIN_NOTDONE; @@ -1263,6 +1333,9 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw, } monarch_cpu = cpu; + if (notify_die(DIE_INIT_MONARCH_ENTER, "INIT", regs, 0, 0, 0) + == NOTIFY_STOP) + ia64_mca_spin(__FUNCTION__); /* * Wait for a bit. On some machines (e.g., HP's zx2000 and zx6000, INIT can be @@ -1273,27 +1346,16 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw, printk("Delaying for 5 seconds...\n"); udelay(5*1000000); ia64_wait_for_slaves(cpu); - printk(KERN_ERR "Processes interrupted by INIT -"); - for_each_online_cpu(c) { - struct ia64_sal_os_state *s; - t = __va(__per_cpu_mca[c] + IA64_MCA_CPU_INIT_STACK_OFFSET); - s = (struct ia64_sal_os_state *)((char *)t + MCA_SOS_OFFSET); - g = s->prev_task; - if (g) { - if (g->pid) - printk(" %d", g->pid); - else - printk(" %d (cpu %d task 0x%p)", g->pid, task_cpu(g), g); - } - } - printk("\n\n"); - if (read_trylock(&tasklist_lock)) { - do_each_thread (g, t) { - printk("\nBacktrace of pid %d (%s)\n", t->pid, t->comm); - show_stack(t, NULL); - } while_each_thread (g, t); - read_unlock(&tasklist_lock); - } + /* If nobody intercepts DIE_INIT_MONARCH_PROCESS then we drop through + * to default_monarch_init_process() above and just print all the + * tasks. + */ + if (notify_die(DIE_INIT_MONARCH_PROCESS, "INIT", regs, 0, 0, 0) + == NOTIFY_STOP) + ia64_mca_spin(__FUNCTION__); + if (notify_die(DIE_INIT_MONARCH_LEAVE, "INIT", regs, 0, 0, 0) + == NOTIFY_STOP) + ia64_mca_spin(__FUNCTION__); printk("\nINIT dump complete. Monarch on cpu %d returning to normal service.\n", cpu); atomic_dec(&monarchs); set_curr_task(cpu, previous_current); @@ -1462,6 +1524,10 @@ ia64_mca_init(void) s64 rc; struct ia64_sal_retval isrv; u64 timeout = IA64_MCA_RENDEZ_TIMEOUT; /* platform specific */ + static struct notifier_block default_init_monarch_nb = { + .notifier_call = default_monarch_init_process, + .priority = 0/* we need to notified last */ + }; IA64_MCA_DEBUG("%s: begin\n", __FUNCTION__); @@ -1555,6 +1621,10 @@ ia64_mca_init(void) "(status %ld)\n", rc); return; } + if (register_die_notifier(&default_init_monarch_nb)) { + printk(KERN_ERR "Failed to register default monarch INIT process\n"); + return; + } IA64_MCA_DEBUG("%s: registered OS INIT handler with SAL\n", __FUNCTION__); diff --git a/arch/ia64/kernel/mca_drv.c b/arch/ia64/kernel/mca_drv.c index eb39bc9c133..3492e3211a4 100644 --- a/arch/ia64/kernel/mca_drv.c +++ b/arch/ia64/kernel/mca_drv.c @@ -547,9 +547,20 @@ recover_from_processor_error(int platform, slidx_table_t *slidx, (pal_processor_state_info_t*)peidx_psp(peidx); /* - * We cannot recover errors with other than bus_check. + * Processor recovery status must key off of the PAL recovery + * status in the Processor State Parameter. */ - if (psp->cc || psp->rc || psp->uc) + + /* + * The machine check is corrected. + */ + if (psp->cm == 1) + return 1; + + /* + * The error was not contained. Software must be reset. + */ + if (psp->us || psp->ci == 0) return 0; /* @@ -570,8 +581,6 @@ recover_from_processor_error(int platform, slidx_table_t *slidx, return 0; if (pbci->eb && pbci->bsi > 0) return 0; - if (psp->ci == 0) - return 0; /* * This is a local MCA and estimated as recoverble external bus error. diff --git a/arch/ia64/kernel/process.c b/arch/ia64/kernel/process.c index 640d6908f8e..e92ea64d804 100644 --- a/arch/ia64/kernel/process.c +++ b/arch/ia64/kernel/process.c @@ -4,6 +4,9 @@ * Copyright (C) 1998-2003 Hewlett-Packard Co * David Mosberger-Tang <davidm@hpl.hp.com> * 04/11/17 Ashok Raj <ashok.raj@intel.com> Added CPU Hotplug Support + * + * 2005-10-07 Keith Owens <kaos@sgi.com> + * Add notify_die() hooks. */ #define __KERNEL_SYSCALLS__ /* see <asm/unistd.h> */ #include <linux/config.h> @@ -34,6 +37,7 @@ #include <asm/elf.h> #include <asm/ia32.h> #include <asm/irq.h> +#include <asm/kdebug.h> #include <asm/pgalloc.h> #include <asm/processor.h> #include <asm/sal.h> @@ -808,12 +812,14 @@ cpu_halt (void) void machine_restart (char *restart_cmd) { + (void) notify_die(DIE_MACHINE_RESTART, restart_cmd, NULL, 0, 0, 0); (*efi.reset_system)(EFI_RESET_WARM, 0, 0, NULL); } void machine_halt (void) { + (void) notify_die(DIE_MACHINE_HALT, "", NULL, 0, 0, 0); cpu_halt(); } diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c index 3af6de36a48..5add0bcf87a 100644 --- a/arch/ia64/kernel/setup.c +++ b/arch/ia64/kernel/setup.c @@ -461,6 +461,7 @@ setup_arch (char **cmdline_p) #endif cpu_init(); /* initialize the bootstrap CPU */ + mmu_context_init(); /* initialize context_id bitmap */ #ifdef CONFIG_ACPI acpi_boot_init(); diff --git a/arch/ia64/kernel/signal.c b/arch/ia64/kernel/signal.c index 774f34b675c..58ce07efc56 100644 --- a/arch/ia64/kernel/signal.c +++ b/arch/ia64/kernel/signal.c @@ -387,15 +387,14 @@ setup_frame (int sig, struct k_sigaction *ka, siginfo_t *info, sigset_t *set, struct sigscratch *scr) { extern char __kernel_sigtramp[]; - unsigned long tramp_addr, new_rbs = 0; + unsigned long tramp_addr, new_rbs = 0, new_sp; struct sigframe __user *frame; long err; - frame = (void __user *) scr->pt.r12; + new_sp = scr->pt.r12; tramp_addr = (unsigned long) __kernel_sigtramp; - if ((ka->sa.sa_flags & SA_ONSTACK) && sas_ss_flags((unsigned long) frame) == 0) { - frame = (void __user *) ((current->sas_ss_sp + current->sas_ss_size) - & ~(STACK_ALIGN - 1)); + if ((ka->sa.sa_flags & SA_ONSTACK) && sas_ss_flags(new_sp) == 0) { + new_sp = current->sas_ss_sp + current->sas_ss_size; /* * We need to check for the register stack being on the signal stack * separately, because it's switched separately (memory stack is switched @@ -404,7 +403,7 @@ setup_frame (int sig, struct k_sigaction *ka, siginfo_t *info, sigset_t *set, if (!rbs_on_sig_stack(scr->pt.ar_bspstore)) new_rbs = (current->sas_ss_sp + sizeof(long) - 1) & ~(sizeof(long) - 1); } - frame = (void __user *) frame - ((sizeof(*frame) + STACK_ALIGN - 1) & ~(STACK_ALIGN - 1)); + frame = (void __user *) ((new_sp - sizeof(*frame)) & -STACK_ALIGN); if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame))) return force_sigsegv_info(sig, frame); diff --git a/arch/ia64/kernel/traps.c b/arch/ia64/kernel/traps.c index f970359e7ed..fba5fdd1f96 100644 --- a/arch/ia64/kernel/traps.c +++ b/arch/ia64/kernel/traps.c @@ -30,17 +30,20 @@ fpswa_interface_t *fpswa_interface; EXPORT_SYMBOL(fpswa_interface); struct notifier_block *ia64die_chain; -static DEFINE_SPINLOCK(die_notifier_lock); -int register_die_notifier(struct notifier_block *nb) +int +register_die_notifier(struct notifier_block *nb) { - int err = 0; - unsigned long flags; - spin_lock_irqsave(&die_notifier_lock, flags); - err = notifier_chain_register(&ia64die_chain, nb); - spin_unlock_irqrestore(&die_notifier_lock, flags); - return err; + return notifier_chain_register(&ia64die_chain, nb); } +EXPORT_SYMBOL_GPL(register_die_notifier); + +int +unregister_die_notifier(struct notifier_block *nb) +{ + return notifier_chain_unregister(&ia64die_chain, nb); +} +EXPORT_SYMBOL_GPL(unregister_die_notifier); void __init trap_init (void) @@ -105,6 +108,7 @@ die (const char *str, struct pt_regs *regs, long err) if (++die.lock_owner_depth < 3) { printk("%s[%d]: %s %ld [%d]\n", current->comm, current->pid, str, err, ++die_counter); + (void) notify_die(DIE_OOPS, (char *)str, regs, err, 255, SIGSEGV); show_regs(regs); } else printk(KERN_ERR "Recursive die() failure, output suppressed\n"); @@ -155,9 +159,8 @@ __kprobes ia64_bad_break (unsigned long break_num, struct pt_regs *regs) switch (break_num) { case 0: /* unknown error (used by GCC for __builtin_abort()) */ if (notify_die(DIE_BREAK, "break 0", regs, break_num, TRAP_BRKPT, SIGTRAP) - == NOTIFY_STOP) { + == NOTIFY_STOP) return; - } die_if_kernel("bugcheck!", regs, break_num); sig = SIGILL; code = ILL_ILLOPC; break; @@ -210,15 +213,6 @@ __kprobes ia64_bad_break (unsigned long break_num, struct pt_regs *regs) sig = SIGILL; code = __ILL_BNDMOD; break; - case 0x80200: - case 0x80300: - if (notify_die(DIE_BREAK, "kprobe", regs, break_num, TRAP_BRKPT, SIGTRAP) - == NOTIFY_STOP) { - return; - } - sig = SIGTRAP; code = TRAP_BRKPT; - break; - default: if (break_num < 0x40000 || break_num > 0x100000) die_if_kernel("Bad break", regs, break_num); @@ -226,6 +220,9 @@ __kprobes ia64_bad_break (unsigned long break_num, struct pt_regs *regs) if (break_num < 0x80000) { sig = SIGILL; code = __ILL_BREAK; } else { + if (notify_die(DIE_BREAK, "bad break", regs, break_num, TRAP_BRKPT, SIGTRAP) + == NOTIFY_STOP) + return; sig = SIGTRAP; code = TRAP_BRKPT; } } @@ -578,12 +575,11 @@ ia64_fault (unsigned long vector, unsigned long isr, unsigned long ifa, #endif break; case 35: siginfo.si_code = TRAP_BRANCH; ifa = 0; break; - case 36: - if (notify_die(DIE_SS, "ss", ®s, vector, - vector, SIGTRAP) == NOTIFY_STOP) - return; - siginfo.si_code = TRAP_TRACE; ifa = 0; break; + case 36: siginfo.si_code = TRAP_TRACE; ifa = 0; break; } + if (notify_die(DIE_FAULT, "ia64_fault", ®s, vector, siginfo.si_code, SIGTRAP) + == NOTIFY_STOP) + return; siginfo.si_signo = SIGTRAP; siginfo.si_errno = 0; siginfo.si_addr = (void __user *) ifa; diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c index a88cdb7232f..0f776b032d3 100644 --- a/arch/ia64/mm/discontig.c +++ b/arch/ia64/mm/discontig.c @@ -350,14 +350,12 @@ static void __init initialize_pernode_data(void) * for best. * @nid: node id * @pernodesize: size of this node's pernode data - * @align: alignment to use for this node's pernode data */ -static void __init *memory_less_node_alloc(int nid, unsigned long pernodesize, - unsigned long align) +static void __init *memory_less_node_alloc(int nid, unsigned long pernodesize) { void *ptr = NULL; u8 best = 0xff; - int bestnode = -1, node; + int bestnode = -1, node, anynode = 0; for_each_online_node(node) { if (node_isset(node, memory_less_mask)) @@ -366,13 +364,15 @@ static void __init *memory_less_node_alloc(int nid, unsigned long pernodesize, best = node_distance(nid, node); bestnode = node; } + anynode = node; } - ptr = __alloc_bootmem_node(mem_data[bestnode].pgdat, - pernodesize, align, __pa(MAX_DMA_ADDRESS)); + if (bestnode == -1) + bestnode = anynode; + + ptr = __alloc_bootmem_node(mem_data[bestnode].pgdat, pernodesize, + PERCPU_PAGE_SIZE, __pa(MAX_DMA_ADDRESS)); - if (!ptr) - panic("NO memory for memory less node\n"); return ptr; } @@ -413,8 +413,7 @@ static void __init memory_less_nodes(void) for_each_node_mask(node, memory_less_mask) { pernodesize = compute_pernodesize(node); - pernode = memory_less_node_alloc(node, pernodesize, - (node) ? (node * PERCPU_PAGE_SIZE) : (1024*1024)); + pernode = memory_less_node_alloc(node, pernodesize); fill_pernode(node, __pa(pernode), pernodesize); } diff --git a/arch/ia64/mm/tlb.c b/arch/ia64/mm/tlb.c index c79a9b96d02..41105d45442 100644 --- a/arch/ia64/mm/tlb.c +++ b/arch/ia64/mm/tlb.c @@ -8,6 +8,8 @@ * Modified RID allocation for SMP * Goutham Rao <goutham.rao@intel.com> * IPI based ptc implementation and A-step IPI implementation. + * Rohit Seth <rohit.seth@intel.com> + * Ken Chen <kenneth.w.chen@intel.com> */ #include <linux/config.h> #include <linux/module.h> @@ -16,78 +18,75 @@ #include <linux/sched.h> #include <linux/smp.h> #include <linux/mm.h> +#include <linux/bootmem.h> #include <asm/delay.h> #include <asm/mmu_context.h> #include <asm/pgalloc.h> #include <asm/pal.h> #include <asm/tlbflush.h> +#include <asm/dma.h> static struct { unsigned long mask; /* mask of supported purge page-sizes */ - unsigned long max_bits; /* log2() of largest supported purge page-size */ + unsigned long max_bits; /* log2 of largest supported purge page-size */ } purge; struct ia64_ctx ia64_ctx = { .lock = SPIN_LOCK_UNLOCKED, .next = 1, - .limit = (1 << 15) - 1, /* start out with the safe (architected) limit */ .max_ctx = ~0U }; DEFINE_PER_CPU(u8, ia64_need_tlb_flush); /* + * Initializes the ia64_ctx.bitmap array based on max_ctx+1. + * Called after cpu_init() has setup ia64_ctx.max_ctx based on + * maximum RID that is supported by boot CPU. + */ +void __init +mmu_context_init (void) +{ + ia64_ctx.bitmap = alloc_bootmem((ia64_ctx.max_ctx+1)>>3); + ia64_ctx.flushmap = alloc_bootmem((ia64_ctx.max_ctx+1)>>3); +} + +/* * Acquire the ia64_ctx.lock before calling this function! */ void wrap_mmu_context (struct mm_struct *mm) { - unsigned long tsk_context, max_ctx = ia64_ctx.max_ctx; - struct task_struct *tsk; - int i; + int i, cpu; + unsigned long flush_bit; - if (ia64_ctx.next > max_ctx) - ia64_ctx.next = 300; /* skip daemons */ - ia64_ctx.limit = max_ctx + 1; + for (i=0; i <= ia64_ctx.max_ctx / BITS_PER_LONG; i++) { + flush_bit = xchg(&ia64_ctx.flushmap[i], 0); + ia64_ctx.bitmap[i] ^= flush_bit; + } + + /* use offset at 300 to skip daemons */ + ia64_ctx.next = find_next_zero_bit(ia64_ctx.bitmap, + ia64_ctx.max_ctx, 300); + ia64_ctx.limit = find_next_bit(ia64_ctx.bitmap, + ia64_ctx.max_ctx, ia64_ctx.next); /* - * Scan all the task's mm->context and set proper safe range + * can't call flush_tlb_all() here because of race condition + * with O(1) scheduler [EF] */ - - read_lock(&tasklist_lock); - repeat: - for_each_process(tsk) { - if (!tsk->mm) - continue; - tsk_context = tsk->mm->context; - if (tsk_context == ia64_ctx.next) { - if (++ia64_ctx.next >= ia64_ctx.limit) { - /* empty range: reset the range limit and start over */ - if (ia64_ctx.next > max_ctx) - ia64_ctx.next = 300; - ia64_ctx.limit = max_ctx + 1; - goto repeat; - } - } - if ((tsk_context > ia64_ctx.next) && (tsk_context < ia64_ctx.limit)) - ia64_ctx.limit = tsk_context; - } - read_unlock(&tasklist_lock); - /* can't call flush_tlb_all() here because of race condition with O(1) scheduler [EF] */ - { - int cpu = get_cpu(); /* prevent preemption/migration */ - for_each_online_cpu(i) { - if (i != cpu) - per_cpu(ia64_need_tlb_flush, i) = 1; - } - put_cpu(); - } + cpu = get_cpu(); /* prevent preemption/migration */ + for_each_online_cpu(i) + if (i != cpu) + per_cpu(ia64_need_tlb_flush, i) = 1; + put_cpu(); local_flush_tlb_all(); } void -ia64_global_tlb_purge (struct mm_struct *mm, unsigned long start, unsigned long end, unsigned long nbits) +ia64_global_tlb_purge (struct mm_struct *mm, unsigned long start, + unsigned long end, unsigned long nbits) { static DEFINE_SPINLOCK(ptcg_lock); @@ -135,7 +134,8 @@ local_flush_tlb_all (void) } void -flush_tlb_range (struct vm_area_struct *vma, unsigned long start, unsigned long end) +flush_tlb_range (struct vm_area_struct *vma, unsigned long start, + unsigned long end) { struct mm_struct *mm = vma->vm_mm; unsigned long size = end - start; @@ -149,7 +149,8 @@ flush_tlb_range (struct vm_area_struct *vma, unsigned long start, unsigned long #endif nbits = ia64_fls(size + 0xfff); - while (unlikely (((1UL << nbits) & purge.mask) == 0) && (nbits < purge.max_bits)) + while (unlikely (((1UL << nbits) & purge.mask) == 0) && + (nbits < purge.max_bits)) ++nbits; if (nbits > purge.max_bits) nbits = purge.max_bits; @@ -191,5 +192,5 @@ ia64_tlb_init (void) local_cpu_data->ptce_stride[0] = ptce_info.stride[0]; local_cpu_data->ptce_stride[1] = ptce_info.stride[1]; - local_flush_tlb_all(); /* nuke left overs from bootstrapping... */ + local_flush_tlb_all(); /* nuke left overs from bootstrapping... */ } diff --git a/arch/ia64/pci/pci.c b/arch/ia64/pci/pci.c index 017cfc3f478..20d76fae24e 100644 --- a/arch/ia64/pci/pci.c +++ b/arch/ia64/pci/pci.c @@ -95,7 +95,7 @@ pci_sal_write (unsigned int seg, unsigned int bus, unsigned int devfn, } static struct pci_raw_ops pci_sal_ops = { - .read = pci_sal_read, + .read = pci_sal_read, .write = pci_sal_write }; @@ -137,35 +137,98 @@ alloc_pci_controller (int seg) return controller; } -static u64 __devinit -add_io_space (struct acpi_resource_address64 *addr) +struct pci_root_info { + struct pci_controller *controller; + char *name; +}; + +static unsigned int +new_space (u64 phys_base, int sparse) { - u64 offset; - int sparse = 0; + u64 mmio_base; int i; - if (addr->address_translation_offset == 0) - return IO_SPACE_BASE(0); /* part of legacy IO space */ - - if (addr->attribute.io.translation_attribute == ACPI_SPARSE_TRANSLATION) - sparse = 1; + if (phys_base == 0) + return 0; /* legacy I/O port space */ - offset = (u64) ioremap(addr->address_translation_offset, 0); + mmio_base = (u64) ioremap(phys_base, 0); for (i = 0; i < num_io_spaces; i++) - if (io_space[i].mmio_base == offset && + if (io_space[i].mmio_base == mmio_base && io_space[i].sparse == sparse) - return IO_SPACE_BASE(i); + return i; if (num_io_spaces == MAX_IO_SPACES) { - printk("Too many IO port spaces\n"); + printk(KERN_ERR "PCI: Too many IO port spaces " + "(MAX_IO_SPACES=%lu)\n", MAX_IO_SPACES); return ~0; } i = num_io_spaces++; - io_space[i].mmio_base = offset; + io_space[i].mmio_base = mmio_base; io_space[i].sparse = sparse; - return IO_SPACE_BASE(i); + return i; +} + +static u64 __devinit +add_io_space (struct pci_root_info *info, struct acpi_resource_address64 *addr) +{ + struct resource *resource; + char *name; + u64 base, min, max, base_port; + unsigned int sparse = 0, space_nr, len; + + resource = kzalloc(sizeof(*resource), GFP_KERNEL); + if (!resource) { + printk(KERN_ERR "PCI: No memory for %s I/O port space\n", + info->name); + goto out; + } + + len = strlen(info->name) + 32; + name = kzalloc(len, GFP_KERNEL); + if (!name) { + printk(KERN_ERR "PCI: No memory for %s I/O port space name\n", + info->name); + goto free_resource; + } + + min = addr->min_address_range; + max = min + addr->address_length - 1; + if (addr->attribute.io.translation_attribute == ACPI_SPARSE_TRANSLATION) + sparse = 1; + + space_nr = new_space(addr->address_translation_offset, sparse); + if (space_nr == ~0) + goto free_name; + + base = __pa(io_space[space_nr].mmio_base); + base_port = IO_SPACE_BASE(space_nr); + snprintf(name, len, "%s I/O Ports %08lx-%08lx", info->name, + base_port + min, base_port + max); + + /* + * The SDM guarantees the legacy 0-64K space is sparse, but if the + * mapping is done by the processor (not the bridge), ACPI may not + * mark it as sparse. + */ + if (space_nr == 0) + sparse = 1; + + resource->name = name; + resource->flags = IORESOURCE_MEM; + resource->start = base + (sparse ? IO_SPACE_SPARSE_ENCODING(min) : min); + resource->end = base + (sparse ? IO_SPACE_SPARSE_ENCODING(max) : max); + insert_resource(&iomem_resource, resource); + + return base_port; + +free_name: + kfree(name); +free_resource: + kfree(resource); +out: + return ~0; } static acpi_status __devinit resource_to_window(struct acpi_resource *resource, @@ -205,11 +268,6 @@ count_window (struct acpi_resource *resource, void *data) return AE_OK; } -struct pci_root_info { - struct pci_controller *controller; - char *name; -}; - static __devinit acpi_status add_window(struct acpi_resource *res, void *data) { struct pci_root_info *info = data; @@ -231,7 +289,7 @@ static __devinit acpi_status add_window(struct acpi_resource *res, void *data) } else if (addr.resource_type == ACPI_IO_RANGE) { flags = IORESOURCE_IO; root = &ioport_resource; - offset = add_io_space(&addr); + offset = add_io_space(info, &addr); if (offset == ~0) return AE_OK; } else @@ -241,7 +299,7 @@ static __devinit acpi_status add_window(struct acpi_resource *res, void *data) window->resource.name = info->name; window->resource.flags = flags; window->resource.start = addr.min_address_range + offset; - window->resource.end = addr.max_address_range + offset; + window->resource.end = window->resource.start + addr.address_length - 1; window->resource.child = NULL; window->offset = offset; @@ -739,7 +797,7 @@ int pci_vector_resources(int last, int nr_released) { int count = nr_released; - count += (IA64_LAST_DEVICE_VECTOR - last); + count += (IA64_LAST_DEVICE_VECTOR - last); return count; } diff --git a/arch/ia64/sn/kernel/io_init.c b/arch/ia64/sn/kernel/io_init.c index b4f5053f5e1..05e4ea88998 100644 --- a/arch/ia64/sn/kernel/io_init.c +++ b/arch/ia64/sn/kernel/io_init.c @@ -349,7 +349,7 @@ void sn_pci_controller_fixup(int segment, int busnum, struct pci_bus *bus) return; /*bus # does not exist */ prom_bussoft_ptr = __va(prom_bussoft_ptr); - controller = kcalloc(1,sizeof(struct pci_controller), GFP_KERNEL); + controller = kzalloc(sizeof(struct pci_controller), GFP_KERNEL); controller->segment = segment; if (!controller) BUG(); diff --git a/arch/ia64/sn/kernel/xpc.h b/arch/ia64/sn/kernel/xpc.h index fbcedc7c27f..5483a9f227d 100644 --- a/arch/ia64/sn/kernel/xpc.h +++ b/arch/ia64/sn/kernel/xpc.h @@ -163,7 +163,7 @@ struct xpc_vars { u8 version; u64 heartbeat; u64 heartbeating_to_mask; - u64 kdb_status; /* 0 = machine running */ + u64 heartbeat_offline; /* if 0, heartbeat should be changing */ int act_nasid; int act_phys_cpuid; u64 vars_part_pa; diff --git a/arch/ia64/sn/kernel/xpc_main.c b/arch/ia64/sn/kernel/xpc_main.c index cece3c7c69b..b617236524c 100644 --- a/arch/ia64/sn/kernel/xpc_main.c +++ b/arch/ia64/sn/kernel/xpc_main.c @@ -57,6 +57,7 @@ #include <linux/reboot.h> #include <asm/sn/intr.h> #include <asm/sn/sn_sal.h> +#include <asm/kdebug.h> #include <asm/uaccess.h> #include "xpc.h" @@ -188,6 +189,11 @@ static struct notifier_block xpc_reboot_notifier = { .notifier_call = xpc_system_reboot, }; +static int xpc_system_die(struct notifier_block *, unsigned long, void *); +static struct notifier_block xpc_die_notifier = { + .notifier_call = xpc_system_die, +}; + /* * Timer function to enforce the timelimit on the partition disengage request. @@ -997,6 +1003,9 @@ xpc_do_exit(enum xpc_retval reason) /* take ourselves off of the reboot_notifier_list */ (void) unregister_reboot_notifier(&xpc_reboot_notifier); + /* take ourselves off of the die_notifier list */ + (void) unregister_die_notifier(&xpc_die_notifier); + /* close down protections for IPI operations */ xpc_restrict_IPI_ops(); @@ -1011,6 +1020,63 @@ xpc_do_exit(enum xpc_retval reason) /* + * Called when the system is about to be either restarted or halted. + */ +static void +xpc_die_disengage(void) +{ + struct xpc_partition *part; + partid_t partid; + unsigned long engaged; + long time, print_time, disengage_request_timeout; + + + /* keep xpc_hb_checker thread from doing anything (just in case) */ + xpc_exiting = 1; + + xpc_vars->heartbeating_to_mask = 0; /* indicate we're deactivated */ + + for (partid = 1; partid < XP_MAX_PARTITIONS; partid++) { + part = &xpc_partitions[partid]; + + if (!XPC_SUPPORTS_DISENGAGE_REQUEST(part-> + remote_vars_version)) { + + /* just in case it was left set by an earlier XPC */ + xpc_clear_partition_engaged(1UL << partid); + continue; + } + + if (xpc_partition_engaged(1UL << partid) || + part->act_state != XPC_P_INACTIVE) { + xpc_request_partition_disengage(part); + xpc_mark_partition_disengaged(part); + xpc_IPI_send_disengage(part); + } + } + + print_time = rtc_time(); + disengage_request_timeout = print_time + + (xpc_disengage_request_timelimit * sn_rtc_cycles_per_second); + + /* wait for all other partitions to disengage from us */ + + while ((engaged = xpc_partition_engaged(-1UL)) && + (time = rtc_time()) < disengage_request_timeout) { + + if (time >= print_time) { + dev_info(xpc_part, "waiting for remote partitions to " + "disengage, engaged=0x%lx\n", engaged); + print_time = time + (XPC_DISENGAGE_PRINTMSG_INTERVAL * + sn_rtc_cycles_per_second); + } + } + dev_info(xpc_part, "finished waiting for remote partitions to " + "disengage, engaged=0x%lx\n", engaged); +} + + +/* * This function is called when the system is being rebooted. */ static int @@ -1038,6 +1104,33 @@ xpc_system_reboot(struct notifier_block *nb, unsigned long event, void *unused) } +/* + * This function is called when the system is being rebooted. + */ +static int +xpc_system_die(struct notifier_block *nb, unsigned long event, void *unused) +{ + switch (event) { + case DIE_MACHINE_RESTART: + case DIE_MACHINE_HALT: + xpc_die_disengage(); + break; + case DIE_MCA_MONARCH_ENTER: + case DIE_INIT_MONARCH_ENTER: + xpc_vars->heartbeat++; + xpc_vars->heartbeat_offline = 1; + break; + case DIE_MCA_MONARCH_LEAVE: + case DIE_INIT_MONARCH_LEAVE: + xpc_vars->heartbeat++; + xpc_vars->heartbeat_offline = 0; + break; + } + + return NOTIFY_DONE; +} + + int __init xpc_init(void) { @@ -1154,6 +1247,12 @@ xpc_init(void) dev_warn(xpc_part, "can't register reboot notifier\n"); } + /* add ourselves to the die_notifier list (i.e., ia64die_chain) */ + ret = register_die_notifier(&xpc_die_notifier); + if (ret != 0) { + dev_warn(xpc_part, "can't register die notifier\n"); + } + /* * Set the beating to other partitions into motion. This is @@ -1179,6 +1278,9 @@ xpc_init(void) /* take ourselves off of the reboot_notifier_list */ (void) unregister_reboot_notifier(&xpc_reboot_notifier); + /* take ourselves off of the die_notifier list */ + (void) unregister_die_notifier(&xpc_die_notifier); + del_timer_sync(&xpc_hb_timer); free_irq(SGI_XPC_ACTIVATE, NULL); xpc_restrict_IPI_ops(); diff --git a/arch/ia64/sn/kernel/xpc_partition.c b/arch/ia64/sn/kernel/xpc_partition.c index 581e113d2d3..cdd6431853a 100644 --- a/arch/ia64/sn/kernel/xpc_partition.c +++ b/arch/ia64/sn/kernel/xpc_partition.c @@ -436,13 +436,13 @@ xpc_check_remote_hb(void) } dev_dbg(xpc_part, "partid = %d, heartbeat = %ld, last_heartbeat" - " = %ld, kdb_status = %ld, HB_mask = 0x%lx\n", partid, - remote_vars->heartbeat, part->last_heartbeat, - remote_vars->kdb_status, + " = %ld, heartbeat_offline = %ld, HB_mask = 0x%lx\n", + partid, remote_vars->heartbeat, part->last_heartbeat, + remote_vars->heartbeat_offline, remote_vars->heartbeating_to_mask); if (((remote_vars->heartbeat == part->last_heartbeat) && - (remote_vars->kdb_status == 0)) || + (remote_vars->heartbeat_offline == 0)) || !xpc_hb_allowed(sn_partition_id, remote_vars)) { XPC_DEACTIVATE_PARTITION(part, xpcNoHeartbeat); diff --git a/arch/ia64/sn/pci/tioce_provider.c b/arch/ia64/sn/pci/tioce_provider.c index 9f03d4e5121..dda196c9e32 100644 --- a/arch/ia64/sn/pci/tioce_provider.c +++ b/arch/ia64/sn/pci/tioce_provider.c @@ -218,7 +218,7 @@ tioce_alloc_map(struct tioce_kernel *ce_kern, int type, int port, if (i > last) return 0; - map = kcalloc(1, sizeof(struct tioce_dmamap), GFP_ATOMIC); + map = kzalloc(sizeof(struct tioce_dmamap), GFP_ATOMIC); if (!map) return 0; @@ -555,7 +555,7 @@ tioce_kern_init(struct tioce_common *tioce_common) struct tioce *tioce_mmr; struct tioce_kernel *tioce_kern; - tioce_kern = kcalloc(1, sizeof(struct tioce_kernel), GFP_KERNEL); + tioce_kern = kzalloc(sizeof(struct tioce_kernel), GFP_KERNEL); if (!tioce_kern) { return NULL; } @@ -727,7 +727,7 @@ tioce_bus_fixup(struct pcibus_bussoft *prom_bussoft, struct pci_controller *cont * Allocate kernel bus soft and copy from prom. */ - tioce_common = kcalloc(1, sizeof(struct tioce_common), GFP_KERNEL); + tioce_common = kzalloc(sizeof(struct tioce_common), GFP_KERNEL); if (!tioce_common) return NULL; diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 1493c7896fe..ed31062029f 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -599,6 +599,10 @@ config HAVE_ARCH_EARLY_PFN_TO_NID def_bool y depends on NEED_MULTIPLE_NODES +config ARCH_MEMORY_PROBE + def_bool y + depends on MEMORY_HOTPLUG + # Some NUMA nodes have memory ranges that span # other nodes. Even though a pfn is valid and # between a node's start and end pfns, it may not diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile index b3ae2993efb..c04bbd32059 100644 --- a/arch/powerpc/kernel/Makefile +++ b/arch/powerpc/kernel/Makefile @@ -4,6 +4,7 @@ ifeq ($(CONFIG_PPC64),y) EXTRA_CFLAGS += -mno-minimal-toc +CFLAGS_ioctl32.o += -Ifs/ endif ifeq ($(CONFIG_PPC32),y) CFLAGS_prom_init.o += -fPIC @@ -11,15 +12,21 @@ CFLAGS_btext.o += -fPIC endif obj-y := semaphore.o cputable.o ptrace.o syscalls.o \ - signal_32.o pmc.o + irq.o signal_32.o pmc.o obj-$(CONFIG_PPC64) += setup_64.o binfmt_elf32.o sys_ppc32.o \ - signal_64.o ptrace32.o systbl.o + signal_64.o ptrace32.o systbl.o \ + paca.o ioctl32.o cpu_setup_power4.o \ + firmware.o sysfs.o obj-$(CONFIG_ALTIVEC) += vecemu.o vector.o obj-$(CONFIG_POWER4) += idle_power4.o obj-$(CONFIG_PPC_OF) += of_device.o -obj-$(CONFIG_PPC_RTAS) += rtas.o +procfs-$(CONFIG_PPC64) := proc_ppc64.o +obj-$(CONFIG_PROC_FS) += $(procfs-y) +rtaspci-$(CONFIG_PPC64) := rtas_pci.o +obj-$(CONFIG_PPC_RTAS) += rtas.o $(rtaspci-y) obj-$(CONFIG_RTAS_FLASH) += rtas_flash.o obj-$(CONFIG_RTAS_PROC) += rtas-proc.o +obj-$(CONFIG_LPARCFG) += lparcfg.o obj-$(CONFIG_IBMVIO) += vio.o obj-$(CONFIG_GENERIC_TBSYNC) += smp-tbsync.o diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c index b7575725199..8793102711a 100644 --- a/arch/powerpc/kernel/asm-offsets.c +++ b/arch/powerpc/kernel/asm-offsets.c @@ -106,7 +106,6 @@ int main(void) DEFINE(ICACHEL1LINESIZE, offsetof(struct ppc64_caches, iline_size)); DEFINE(ICACHEL1LOGLINESIZE, offsetof(struct ppc64_caches, log_iline_size)); DEFINE(ICACHEL1LINESPERPAGE, offsetof(struct ppc64_caches, ilines_per_page)); - DEFINE(PLATFORM, offsetof(struct systemcfg, platform)); DEFINE(PLATFORM_LPAR, PLATFORM_LPAR); /* paca */ diff --git a/arch/ppc64/kernel/cpu_setup_power4.S b/arch/powerpc/kernel/cpu_setup_power4.S index 1fb673c511f..cca942fe611 100644 --- a/arch/ppc64/kernel/cpu_setup_power4.S +++ b/arch/powerpc/kernel/cpu_setup_power4.S @@ -114,11 +114,11 @@ _GLOBAL(__setup_cpu_ppc970) .data .balign L1_CACHE_BYTES,0 -cpu_state_storage: +cpu_state_storage: .space CS_SIZE .balign L1_CACHE_BYTES,0 .text - + /* Called in normal context to backup CPU 0 state. This * does not include cache settings. This function is also * called for machine sleep. This does not include the MMU @@ -151,7 +151,7 @@ _GLOBAL(__save_cpu_setup) std r3,CS_HID4(r5) mfspr r3,SPRN_HID5 std r3,CS_HID5(r5) - + 2: mtcr r7 blr @@ -213,7 +213,7 @@ _GLOBAL(__restore_cpu_setup) mtspr SPRN_HID1,r3 sync isync - + /* Restore HID4 */ ld r3,CS_HID4(r5) sync diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c index cc4e9eb1c13..1d85cedbbb7 100644 --- a/arch/powerpc/kernel/cputable.c +++ b/arch/powerpc/kernel/cputable.c @@ -52,6 +52,9 @@ extern void __setup_cpu_ppc970(unsigned long offset, struct cpu_spec* spec); #define COMMON_USER (PPC_FEATURE_32 | PPC_FEATURE_HAS_FPU | \ PPC_FEATURE_HAS_MMU) #define COMMON_USER_PPC64 (COMMON_USER | PPC_FEATURE_64) +#define COMMON_USER_POWER4 (COMMON_USER_PPC64 | PPC_FEATURE_POWER4) +#define COMMON_USER_POWER5 (COMMON_USER_PPC64 | PPC_FEATURE_POWER5) +#define COMMON_USER_POWER5_PLUS (COMMON_USER_PPC64 | PPC_FEATURE_POWER5_PLUS) /* We only set the spe features if the kernel was compiled with @@ -160,7 +163,7 @@ struct cpu_spec cpu_specs[] = { .pvr_value = 0x00350000, .cpu_name = "POWER4 (gp)", .cpu_features = CPU_FTRS_POWER4, - .cpu_user_features = COMMON_USER_PPC64, + .cpu_user_features = COMMON_USER_POWER4, .icache_bsize = 128, .dcache_bsize = 128, .num_pmcs = 8, @@ -175,7 +178,7 @@ struct cpu_spec cpu_specs[] = { .pvr_value = 0x00380000, .cpu_name = "POWER4+ (gq)", .cpu_features = CPU_FTRS_POWER4, - .cpu_user_features = COMMON_USER_PPC64, + .cpu_user_features = COMMON_USER_POWER4, .icache_bsize = 128, .dcache_bsize = 128, .num_pmcs = 8, @@ -190,7 +193,7 @@ struct cpu_spec cpu_specs[] = { .pvr_value = 0x00390000, .cpu_name = "PPC970", .cpu_features = CPU_FTRS_PPC970, - .cpu_user_features = COMMON_USER_PPC64 | + .cpu_user_features = COMMON_USER_POWER4 | PPC_FEATURE_HAS_ALTIVEC_COMP, .icache_bsize = 128, .dcache_bsize = 128, @@ -212,7 +215,7 @@ struct cpu_spec cpu_specs[] = { #else .cpu_features = CPU_FTRS_PPC970, #endif - .cpu_user_features = COMMON_USER_PPC64 | + .cpu_user_features = COMMON_USER_POWER4 | PPC_FEATURE_HAS_ALTIVEC_COMP, .icache_bsize = 128, .dcache_bsize = 128, @@ -230,7 +233,7 @@ struct cpu_spec cpu_specs[] = { .pvr_value = 0x00440000, .cpu_name = "PPC970MP", .cpu_features = CPU_FTRS_PPC970, - .cpu_user_features = COMMON_USER_PPC64 | + .cpu_user_features = COMMON_USER_POWER4 | PPC_FEATURE_HAS_ALTIVEC_COMP, .icache_bsize = 128, .dcache_bsize = 128, @@ -245,7 +248,7 @@ struct cpu_spec cpu_specs[] = { .pvr_value = 0x003a0000, .cpu_name = "POWER5 (gr)", .cpu_features = CPU_FTRS_POWER5, - .cpu_user_features = COMMON_USER_PPC64, + .cpu_user_features = COMMON_USER_POWER5, .icache_bsize = 128, .dcache_bsize = 128, .num_pmcs = 6, @@ -260,7 +263,7 @@ struct cpu_spec cpu_specs[] = { .pvr_value = 0x003b0000, .cpu_name = "POWER5 (gs)", .cpu_features = CPU_FTRS_POWER5, - .cpu_user_features = COMMON_USER_PPC64, + .cpu_user_features = COMMON_USER_POWER5_PLUS, .icache_bsize = 128, .dcache_bsize = 128, .num_pmcs = 6, @@ -276,7 +279,7 @@ struct cpu_spec cpu_specs[] = { .cpu_name = "Cell Broadband Engine", .cpu_features = CPU_FTRS_CELL, .cpu_user_features = COMMON_USER_PPC64 | - PPC_FEATURE_HAS_ALTIVEC_COMP, + PPC_FEATURE_CELL | PPC_FEATURE_HAS_ALTIVEC_COMP, .icache_bsize = 128, .dcache_bsize = 128, .cpu_setup = __setup_cpu_be, diff --git a/arch/ppc64/kernel/firmware.c b/arch/powerpc/kernel/firmware.c index d8432c0fb27..65eae752a52 100644 --- a/arch/ppc64/kernel/firmware.c +++ b/arch/powerpc/kernel/firmware.c @@ -1,6 +1,4 @@ /* - * arch/ppc64/kernel/firmware.c - * * Extracted from cputable.c * * Copyright (C) 2001 Ben. Herrenschmidt (benh@kernel.crashing.org) diff --git a/arch/powerpc/kernel/fpu.S b/arch/powerpc/kernel/fpu.S index 4d6001fa1cf..b780b42c95f 100644 --- a/arch/powerpc/kernel/fpu.S +++ b/arch/powerpc/kernel/fpu.S @@ -41,20 +41,20 @@ _GLOBAL(load_up_fpu) #ifndef CONFIG_SMP LOADBASE(r3, last_task_used_math) toreal(r3) - LDL r4,OFF(last_task_used_math)(r3) - CMPI 0,r4,0 + PPC_LL r4,OFF(last_task_used_math)(r3) + PPC_LCMPI 0,r4,0 beq 1f toreal(r4) addi r4,r4,THREAD /* want last_task_used_math->thread */ SAVE_32FPRS(0, r4) mffs fr0 stfd fr0,THREAD_FPSCR(r4) - LDL r5,PT_REGS(r4) + PPC_LL r5,PT_REGS(r4) toreal(r5) - LDL r4,_MSR-STACK_FRAME_OVERHEAD(r5) + PPC_LL r4,_MSR-STACK_FRAME_OVERHEAD(r5) li r10,MSR_FP|MSR_FE0|MSR_FE1 andc r4,r4,r10 /* disable FP for previous task */ - STL r4,_MSR-STACK_FRAME_OVERHEAD(r5) + PPC_STL r4,_MSR-STACK_FRAME_OVERHEAD(r5) 1: #endif /* CONFIG_SMP */ /* enable use of FP after return */ @@ -77,7 +77,7 @@ _GLOBAL(load_up_fpu) #ifndef CONFIG_SMP subi r4,r5,THREAD fromreal(r4) - STL r4,OFF(last_task_used_math)(r3) + PPC_STL r4,OFF(last_task_used_math)(r3) #endif /* CONFIG_SMP */ /* restore registers and return */ /* we haven't used ctr or xer or lr */ @@ -97,24 +97,24 @@ _GLOBAL(giveup_fpu) MTMSRD(r5) /* enable use of fpu now */ SYNC_601 isync - CMPI 0,r3,0 + PPC_LCMPI 0,r3,0 beqlr- /* if no previous owner, done */ addi r3,r3,THREAD /* want THREAD of task */ - LDL r5,PT_REGS(r3) - CMPI 0,r5,0 + PPC_LL r5,PT_REGS(r3) + PPC_LCMPI 0,r5,0 SAVE_32FPRS(0, r3) mffs fr0 stfd fr0,THREAD_FPSCR(r3) beq 1f - LDL r4,_MSR-STACK_FRAME_OVERHEAD(r5) + PPC_LL r4,_MSR-STACK_FRAME_OVERHEAD(r5) li r3,MSR_FP|MSR_FE0|MSR_FE1 andc r4,r4,r3 /* disable FP for previous task */ - STL r4,_MSR-STACK_FRAME_OVERHEAD(r5) + PPC_STL r4,_MSR-STACK_FRAME_OVERHEAD(r5) 1: #ifndef CONFIG_SMP li r5,0 LOADBASE(r4,last_task_used_math) - STL r5,OFF(last_task_used_math)(r4) + PPC_STL r5,OFF(last_task_used_math)(r4) #endif /* CONFIG_SMP */ blr diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S index 16ab40daa73..8a8bf79ef04 100644 --- a/arch/powerpc/kernel/head_64.S +++ b/arch/powerpc/kernel/head_64.S @@ -28,7 +28,6 @@ #include <asm/reg.h> #include <asm/page.h> #include <asm/mmu.h> -#include <asm/systemcfg.h> #include <asm/ppc_asm.h> #include <asm/asm-offsets.h> #include <asm/bug.h> @@ -1697,25 +1696,14 @@ _GLOBAL(pmac_secondary_start) * SPRG3 = paca virtual address */ _GLOBAL(__secondary_start) + /* Set thread priority to MEDIUM */ + HMT_MEDIUM - HMT_MEDIUM /* Set thread priority to MEDIUM */ - + /* Load TOC */ ld r2,PACATOC(r13) - li r6,0 - stb r6,PACAPROCENABLED(r13) - -#ifndef CONFIG_PPC_ISERIES - /* Initialize the page table pointer register. */ - LOADADDR(r6,_SDR1) - ld r6,0(r6) /* get the value of _SDR1 */ - mtspr SPRN_SDR1,r6 /* set the htab location */ -#endif - /* Initialize the first segment table (or SLB) entry */ - ld r3,PACASTABVIRT(r13) /* get addr of segment table */ -BEGIN_FTR_SECTION - bl .stab_initialize -END_FTR_SECTION_IFCLR(CPU_FTR_SLB) - bl .slb_initialize + + /* Do early setup for that CPU (stab, slb, hash table pointer) */ + bl .early_setup_secondary /* Initialize the kernel stack. Just a repeat for iSeries. */ LOADADDR(r3,current_set) @@ -1724,37 +1712,7 @@ END_FTR_SECTION_IFCLR(CPU_FTR_SLB) addi r1,r1,THREAD_SIZE-STACK_FRAME_OVERHEAD std r1,PACAKSAVE(r13) - ld r3,PACASTABREAL(r13) /* get raddr of segment table */ - ori r4,r3,1 /* turn on valid bit */ - -#ifdef CONFIG_PPC_ISERIES - li r0,-1 /* hypervisor call */ - li r3,1 - sldi r3,r3,63 /* 0x8000000000000000 */ - ori r3,r3,4 /* 0x8000000000000004 */ - sc /* HvCall_setASR */ -#else - /* set the ASR */ - ld r3,systemcfg@got(r2) /* r3 = ptr to systemcfg */ - ld r3,0(r3) - lwz r3,PLATFORM(r3) /* r3 = platform flags */ - andi. r3,r3,PLATFORM_LPAR /* Test if bit 0 is set (LPAR bit) */ - beq 98f /* branch if result is 0 */ - mfspr r3,SPRN_PVR - srwi r3,r3,16 - cmpwi r3,0x37 /* SStar */ - beq 97f - cmpwi r3,0x36 /* IStar */ - beq 97f - cmpwi r3,0x34 /* Pulsar */ - bne 98f -97: li r3,H_SET_ASR /* hcall = H_SET_ASR */ - HVSC /* Invoking hcall */ - b 99f -98: /* !(rpa hypervisor) || !(star) */ - mtasr r4 /* set the stab location */ -99: -#endif + /* Clear backchain so we get nice backtraces */ li r7,0 mtlr r7 @@ -1777,6 +1735,7 @@ _GLOBAL(start_secondary_prolog) li r3,0 std r3,0(r1) /* Zero the stack frame pointer */ bl .start_secondary + b . #endif /* @@ -1896,40 +1855,6 @@ _STATIC(start_here_multiplatform) mr r3,r31 bl .early_setup - /* set the ASR */ - ld r3,PACASTABREAL(r13) - ori r4,r3,1 /* turn on valid bit */ - ld r3,systemcfg@got(r2) /* r3 = ptr to systemcfg */ - ld r3,0(r3) - lwz r3,PLATFORM(r3) /* r3 = platform flags */ - andi. r3,r3,PLATFORM_LPAR /* Test if bit 0 is set (LPAR bit) */ - beq 98f /* branch if result is 0 */ - mfspr r3,SPRN_PVR - srwi r3,r3,16 - cmpwi r3,0x37 /* SStar */ - beq 97f - cmpwi r3,0x36 /* IStar */ - beq 97f - cmpwi r3,0x34 /* Pulsar */ - bne 98f -97: li r3,H_SET_ASR /* hcall = H_SET_ASR */ - HVSC /* Invoking hcall */ - b 99f -98: /* !(rpa hypervisor) || !(star) */ - mtasr r4 /* set the stab location */ -99: - /* Set SDR1 (hash table pointer) */ - ld r3,systemcfg@got(r2) /* r3 = ptr to systemcfg */ - ld r3,0(r3) - lwz r3,PLATFORM(r3) /* r3 = platform flags */ - /* Test if bit 0 is set (LPAR bit) */ - andi. r3,r3,PLATFORM_LPAR - bne 98f /* branch if result is !0 */ - LOADADDR(r6,_SDR1) /* Only if NOT LPAR */ - add r6,r6,r26 - ld r6,0(r6) /* get the value of _SDR1 */ - mtspr SPRN_SDR1,r6 /* set the htab location */ -98: LOADADDR(r3,.start_here_common) SET_REG_TO_CONST(r4, MSR_KERNEL) mtspr SPRN_SRR0,r3 diff --git a/arch/ppc64/kernel/ioctl32.c b/arch/powerpc/kernel/ioctl32.c index ba4a899045c..3fa6a93adbd 100644 --- a/arch/ppc64/kernel/ioctl32.c +++ b/arch/powerpc/kernel/ioctl32.c @@ -1,6 +1,6 @@ -/* +/* * ioctl32.c: Conversion between 32bit and 64bit native ioctls. - * + * * Based on sparc64 ioctl32.c by: * * Copyright (C) 1997-2000 Jakub Jelinek (jakub@redhat.com) diff --git a/arch/ppc64/kernel/irq.c b/arch/powerpc/kernel/irq.c index 87474584033..4b7940693f3 100644 --- a/arch/ppc64/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -5,12 +5,12 @@ * Copyright (C) 1992 Linus Torvalds * Adapted from arch/i386 by Gary Thomas * Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org) - * Updated and modified by Cort Dougan (cort@cs.nmt.edu) - * Copyright (C) 1996 Cort Dougan + * Updated and modified by Cort Dougan <cort@fsmlabs.com> + * Copyright (C) 1996-2001 Cort Dougan * Adapted for Power Macintosh by Paul Mackerras * Copyright (C) 1996 Paul Mackerras (paulus@cs.anu.edu.au) * Amiga/APUS changes by Jesper Skov (jskov@cygnus.co.uk). - * + * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License * as published by the Free Software Foundation; either version @@ -21,6 +21,14 @@ * instead of just grabbing them. Thus setups with different IRQ numbers * shouldn't result in any weird surprises, and installing new handlers * should be easier. + * + * The MPC8xx has an interrupt mask in the SIU. If a bit is set, the + * interrupt is _enabled_. As expected, IRQ0 is bit 0 in the 32-bit + * mask register (of which only 16 are defined), hence the weird shifting + * and complement of the cached_irq_mask. I want to be able to stuff + * this right into the SIU SMASK register. + * Many of the prep/chrp functions are conditional compiled on CONFIG_8xx + * to reduce code space and undefined function references. */ #include <linux/errno.h> @@ -29,6 +37,7 @@ #include <linux/kernel_stat.h> #include <linux/signal.h> #include <linux/sched.h> +#include <linux/ptrace.h> #include <linux/ioport.h> #include <linux/interrupt.h> #include <linux/timex.h> @@ -40,9 +49,13 @@ #include <linux/irq.h> #include <linux/proc_fs.h> #include <linux/random.h> -#include <linux/kallsyms.h> +#include <linux/seq_file.h> +#include <linux/cpumask.h> #include <linux/profile.h> #include <linux/bitops.h> +#ifdef CONFIG_PPC64 +#include <linux/kallsyms.h> +#endif #include <asm/uaccess.h> #include <asm/system.h> @@ -52,35 +65,54 @@ #include <asm/cache.h> #include <asm/prom.h> #include <asm/ptrace.h> -#include <asm/iseries/it_lp_queue.h> #include <asm/machdep.h> +#ifdef CONFIG_PPC64 +#include <asm/iseries/it_lp_queue.h> #include <asm/paca.h> +#endif -#ifdef CONFIG_SMP -extern void iSeries_smp_message_recv( struct pt_regs * ); +static int ppc_spurious_interrupts; + +#if defined(CONFIG_PPC_ISERIES) && defined(CONFIG_SMP) +extern void iSeries_smp_message_recv(struct pt_regs *); +#endif + +#ifdef CONFIG_PPC32 +#define NR_MASK_WORDS ((NR_IRQS + 31) / 32) + +unsigned long ppc_cached_irq_mask[NR_MASK_WORDS]; +atomic_t ppc_n_lost_interrupts; + +#ifdef CONFIG_TAU_INT +extern int tau_initialized; +extern int tau_interrupts(int); +#endif + +#if defined(CONFIG_SMP) && !defined(CONFIG_PPC_MERGE) +extern atomic_t ipi_recv; +extern atomic_t ipi_sent; #endif +#endif /* CONFIG_PPC32 */ -extern irq_desc_t irq_desc[NR_IRQS]; +#ifdef CONFIG_PPC64 EXPORT_SYMBOL(irq_desc); int distribute_irqs = 1; int __irq_offset_value; -int ppc_spurious_interrupts; u64 ppc64_interrupt_controller; +#endif /* CONFIG_PPC64 */ int show_interrupts(struct seq_file *p, void *v) { - int i = *(loff_t *) v, j; - struct irqaction * action; + int i = *(loff_t *)v, j; + struct irqaction *action; irq_desc_t *desc; unsigned long flags; if (i == 0) { - seq_printf(p, " "); - for (j=0; j<NR_CPUS; j++) { - if (cpu_online(j)) - seq_printf(p, "CPU%d ",j); - } + seq_puts(p, " "); + for_each_online_cpu(j) + seq_printf(p, "CPU%d ", j); seq_putc(p, '\n'); } @@ -92,26 +124,41 @@ int show_interrupts(struct seq_file *p, void *v) goto skip; seq_printf(p, "%3d: ", i); #ifdef CONFIG_SMP - for (j = 0; j < NR_CPUS; j++) { - if (cpu_online(j)) - seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]); - } + for_each_online_cpu(j) + seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]); #else seq_printf(p, "%10u ", kstat_irqs(i)); #endif /* CONFIG_SMP */ if (desc->handler) - seq_printf(p, " %s ", desc->handler->typename ); + seq_printf(p, " %s ", desc->handler->typename); else - seq_printf(p, " None "); + seq_puts(p, " None "); seq_printf(p, "%s", (desc->status & IRQ_LEVEL) ? "Level " : "Edge "); - seq_printf(p, " %s",action->name); - for (action=action->next; action; action = action->next) + seq_printf(p, " %s", action->name); + for (action = action->next; action; action = action->next) seq_printf(p, ", %s", action->name); seq_putc(p, '\n'); skip: spin_unlock_irqrestore(&desc->lock, flags); - } else if (i == NR_IRQS) + } else if (i == NR_IRQS) { +#ifdef CONFIG_PPC32 +#ifdef CONFIG_TAU_INT + if (tau_initialized){ + seq_puts(p, "TAU: "); + for (j = 0; j < NR_CPUS; j++) + if (cpu_online(j)) + seq_printf(p, "%10u ", tau_interrupts(j)); + seq_puts(p, " PowerPC Thermal Assist (cpu temp)\n"); + } +#endif +#if defined(CONFIG_SMP) && !defined(CONFIG_PPC_MERGE) + /* should this be per processor send/receive? */ + seq_printf(p, "IPI (recv/sent): %10u/%u\n", + atomic_read(&ipi_recv), atomic_read(&ipi_sent)); +#endif +#endif /* CONFIG_PPC32 */ seq_printf(p, "BAD: %10u\n", ppc_spurious_interrupts); + } return 0; } @@ -144,126 +191,6 @@ void fixup_irqs(cpumask_t map) } #endif -extern int noirqdebug; - -/* - * Eventually, this should take an array of interrupts and an array size - * so it can dispatch multiple interrupts. - */ -void ppc_irq_dispatch_handler(struct pt_regs *regs, int irq) -{ - int status; - struct irqaction *action; - int cpu = smp_processor_id(); - irq_desc_t *desc = get_irq_desc(irq); - irqreturn_t action_ret; -#ifdef CONFIG_IRQSTACKS - struct thread_info *curtp, *irqtp; -#endif - - kstat_cpu(cpu).irqs[irq]++; - - if (desc->status & IRQ_PER_CPU) { - /* no locking required for CPU-local interrupts: */ - ack_irq(irq); - action_ret = handle_IRQ_event(irq, regs, desc->action); - desc->handler->end(irq); - return; - } - - spin_lock(&desc->lock); - ack_irq(irq); - /* - REPLAY is when Linux resends an IRQ that was dropped earlier - WAITING is used by probe to mark irqs that are being tested - */ - status = desc->status & ~(IRQ_REPLAY | IRQ_WAITING); - status |= IRQ_PENDING; /* we _want_ to handle it */ - - /* - * If the IRQ is disabled for whatever reason, we cannot - * use the action we have. - */ - action = NULL; - if (likely(!(status & (IRQ_DISABLED | IRQ_INPROGRESS)))) { - action = desc->action; - if (!action || !action->handler) { - ppc_spurious_interrupts++; - printk(KERN_DEBUG "Unhandled interrupt %x, disabled\n", irq); - /* We can't call disable_irq here, it would deadlock */ - if (!desc->depth) - desc->depth = 1; - desc->status |= IRQ_DISABLED; - /* This is not a real spurrious interrupt, we - * have to eoi it, so we jump to out - */ - mask_irq(irq); - goto out; - } - status &= ~IRQ_PENDING; /* we commit to handling */ - status |= IRQ_INPROGRESS; /* we are handling it */ - } - desc->status = status; - - /* - * If there is no IRQ handler or it was disabled, exit early. - Since we set PENDING, if another processor is handling - a different instance of this same irq, the other processor - will take care of it. - */ - if (unlikely(!action)) - goto out; - - /* - * Edge triggered interrupts need to remember - * pending events. - * This applies to any hw interrupts that allow a second - * instance of the same irq to arrive while we are in do_IRQ - * or in the handler. But the code here only handles the _second_ - * instance of the irq, not the third or fourth. So it is mostly - * useful for irq hardware that does not mask cleanly in an - * SMP environment. - */ - for (;;) { - spin_unlock(&desc->lock); - -#ifdef CONFIG_IRQSTACKS - /* Switch to the irq stack to handle this */ - curtp = current_thread_info(); - irqtp = hardirq_ctx[smp_processor_id()]; - if (curtp != irqtp) { - irqtp->task = curtp->task; - irqtp->flags = 0; - action_ret = call_handle_IRQ_event(irq, regs, action, irqtp); - irqtp->task = NULL; - if (irqtp->flags) - set_bits(irqtp->flags, &curtp->flags); - } else -#endif - action_ret = handle_IRQ_event(irq, regs, action); - - spin_lock(&desc->lock); - if (!noirqdebug) - note_interrupt(irq, desc, action_ret, regs); - if (likely(!(desc->status & IRQ_PENDING))) - break; - desc->status &= ~IRQ_PENDING; - } -out: - desc->status &= ~IRQ_INPROGRESS; - /* - * The ->end() handler has to deal with interrupts which got - * disabled while the handler was running. - */ - if (desc->handler) { - if (desc->handler->end) - desc->handler->end(irq); - else if (desc->handler->enable) - desc->handler->enable(irq); - } - spin_unlock(&desc->lock); -} - #ifdef CONFIG_PPC_ISERIES void do_IRQ(struct pt_regs *regs) { @@ -310,8 +237,11 @@ void do_IRQ(struct pt_regs *regs) void do_IRQ(struct pt_regs *regs) { int irq; +#ifdef CONFIG_IRQSTACKS + struct thread_info *curtp, *irqtp; +#endif - irq_enter(); + irq_enter(); #ifdef CONFIG_DEBUG_STACKOVERFLOW /* Debugging check for stack overflow: is there less than 2KB free? */ @@ -328,20 +258,44 @@ void do_IRQ(struct pt_regs *regs) } #endif + /* + * Every platform is required to implement ppc_md.get_irq. + * This function will either return an irq number or -1 to + * indicate there are no more pending. + * The value -2 is for buggy hardware and means that this IRQ + * has already been handled. -- Tom + */ irq = ppc_md.get_irq(regs); - if (irq >= 0) - ppc_irq_dispatch_handler(regs, irq); - else - /* That's not SMP safe ... but who cares ? */ - ppc_spurious_interrupts++; - - irq_exit(); + if (irq >= 0) { +#ifdef CONFIG_IRQSTACKS + /* Switch to the irq stack to handle this */ + curtp = current_thread_info(); + irqtp = hardirq_ctx[smp_processor_id()]; + if (curtp != irqtp) { + irqtp->task = curtp->task; + irqtp->flags = 0; + call___do_IRQ(irq, regs, irqtp); + irqtp->task = NULL; + if (irqtp->flags) + set_bits(irqtp->flags, &curtp->flags); + } else +#endif + __do_IRQ(irq, regs); + } else +#ifdef CONFIG_PPC32 + if (irq != -2) +#endif + /* That's not SMP safe ... but who cares ? */ + ppc_spurious_interrupts++; + irq_exit(); } + #endif /* CONFIG_PPC_ISERIES */ void __init init_IRQ(void) { +#ifdef CONFIG_PPC64 static int once = 0; if (once) @@ -349,10 +303,14 @@ void __init init_IRQ(void) once++; +#endif ppc_md.init_IRQ(); +#ifdef CONFIG_PPC64 irq_ctx_init(); +#endif } +#ifdef CONFIG_PPC64 #ifndef CONFIG_PPC_ISERIES /* * Virtual IRQ mapping code, used on systems with XICS interrupt controllers. @@ -517,3 +475,4 @@ static int __init setup_noirqdistrib(char *str) } __setup("noirqdistrib", setup_noirqdistrib); +#endif /* CONFIG_PPC64 */ diff --git a/arch/ppc64/kernel/lparcfg.c b/arch/powerpc/kernel/lparcfg.c index 3e7b2f28ec8..5e954fae031 100644 --- a/arch/ppc64/kernel/lparcfg.c +++ b/arch/powerpc/kernel/lparcfg.c @@ -35,6 +35,7 @@ #include <asm/time.h> #include <asm/iseries/it_exp_vpd_panel.h> #include <asm/prom.h> +#include <asm/systemcfg.h> #define MODULE_VERS "1.6" #define MODULE_NAME "lparcfg" @@ -96,7 +97,7 @@ static unsigned long get_purr(void) #define lparcfg_write NULL -/* +/* * Methods used to fetch LPAR data when running on an iSeries platform. */ static int lparcfg_data(struct seq_file *m, void *v) @@ -168,7 +169,7 @@ static int lparcfg_data(struct seq_file *m, void *v) #endif /* CONFIG_PPC_ISERIES */ #ifdef CONFIG_PPC_PSERIES -/* +/* * Methods used to fetch LPAR data when running on a pSeries platform. */ @@ -177,7 +178,7 @@ static int lparcfg_data(struct seq_file *m, void *v) * entitled_capacity,unallocated_capacity, * aggregation, resource_capability). * - * R4 = Entitled Processor Capacity Percentage. + * R4 = Entitled Processor Capacity Percentage. * R5 = Unallocated Processor Capacity Percentage. * R6 (AABBCCDDEEFFGGHH). * XXXX - reserved (0) @@ -190,7 +191,7 @@ static int lparcfg_data(struct seq_file *m, void *v) * XX - variable processor Capacity Weight * XX - Unallocated Variable Processor Capacity Weight. * XXXX - Active processors in Physical Processor Pool. - * XXXX - Processors active on platform. + * XXXX - Processors active on platform. */ static unsigned int h_get_ppp(unsigned long *entitled, unsigned long *unallocated, @@ -273,7 +274,7 @@ static void parse_system_parameter_string(struct seq_file *m) if (!workbuffer) { printk(KERN_ERR "%s %s kmalloc failure at line %d \n", __FILE__, __FUNCTION__, __LINE__); - kfree(local_buffer); + kfree(local_buffer); return; } #ifdef LPARCFG_DEBUG @@ -371,7 +372,7 @@ static int lparcfg_data(struct seq_file *m, void *v) lrdrp = (int *)get_property(rtas_node, "ibm,lrdr-capacity", NULL); if (lrdrp == NULL) { - partition_potential_processors = systemcfg->processorCount; + partition_potential_processors = _systemcfg->processorCount; } else { partition_potential_processors = *(lrdrp + 4); } diff --git a/arch/powerpc/kernel/misc_32.S b/arch/powerpc/kernel/misc_32.S index 3bedb532aed..f6d84a75ed2 100644 --- a/arch/powerpc/kernel/misc_32.S +++ b/arch/powerpc/kernel/misc_32.S @@ -519,7 +519,7 @@ END_FTR_SECTION_IFCLR(CPU_FTR_SPLIT_ID_CACHE) * * flush_icache_range(unsigned long start, unsigned long stop) */ -_GLOBAL(flush_icache_range) +_GLOBAL(__flush_icache_range) BEGIN_FTR_SECTION blr /* for 601, do nothing */ END_FTR_SECTION_IFCLR(CPU_FTR_SPLIT_ID_CACHE) @@ -607,27 +607,6 @@ _GLOBAL(invalidate_dcache_range) sync /* wait for dcbi's to get to ram */ blr -#ifdef CONFIG_NOT_COHERENT_CACHE -/* - * 40x cores have 8K or 16K dcache and 32 byte line size. - * 44x has a 32K dcache and 32 byte line size. - * 8xx has 1, 2, 4, 8K variants. - * For now, cover the worst case of the 44x. - * Must be called with external interrupts disabled. - */ -#define CACHE_NWAYS 64 -#define CACHE_NLINES 16 - -_GLOBAL(flush_dcache_all) - li r4, (2 * CACHE_NWAYS * CACHE_NLINES) - mtctr r4 - lis r5, KERNELBASE@h -1: lwz r3, 0(r5) /* Load one word from every line */ - addi r5, r5, L1_CACHE_BYTES - bdnz 1b - blr -#endif /* CONFIG_NOT_COHERENT_CACHE */ - /* * Flush a particular page from the data cache to RAM. * Note: this is necessary because the instruction cache does *not* diff --git a/arch/powerpc/kernel/misc_64.S b/arch/powerpc/kernel/misc_64.S index ae1433da09b..ae48a002f81 100644 --- a/arch/powerpc/kernel/misc_64.S +++ b/arch/powerpc/kernel/misc_64.S @@ -89,12 +89,12 @@ _GLOBAL(call_do_softirq) mtlr r0 blr -_GLOBAL(call_handle_IRQ_event) +_GLOBAL(call___do_IRQ) mflr r0 std r0,16(r1) - stdu r1,THREAD_SIZE-112(r6) - mr r1,r6 - bl .handle_IRQ_event + stdu r1,THREAD_SIZE-112(r5) + mr r1,r5 + bl .__do_IRQ ld r1,0(r1) ld r0,16(r1) mtlr r0 diff --git a/arch/ppc64/kernel/pacaData.c b/arch/powerpc/kernel/paca.c index 3133c72b28e..3cf2517c5f9 100644 --- a/arch/ppc64/kernel/pacaData.c +++ b/arch/powerpc/kernel/paca.c @@ -15,7 +15,7 @@ #include <asm/processor.h> #include <asm/ptrace.h> #include <asm/page.h> - +#include <asm/systemcfg.h> #include <asm/lppaca.h> #include <asm/iseries/it_lp_queue.h> #include <asm/paca.h> @@ -24,15 +24,14 @@ static union { struct systemcfg data; u8 page[PAGE_SIZE]; } systemcfg_store __attribute__((__section__(".data.page.aligned"))); -struct systemcfg *systemcfg = &systemcfg_store.data; -EXPORT_SYMBOL(systemcfg); +struct systemcfg *_systemcfg = &systemcfg_store.data; /* This symbol is provided by the linker - let it fill in the paca * field correctly */ extern unsigned long __toc_start; -/* The Paca is an array with one entry per processor. Each contains an +/* The Paca is an array with one entry per processor. Each contains an * lppaca, which contains the information shared between the * hypervisor and Linux. Each also contains an ItLpRegSave area which * is used by the hypervisor to save registers. diff --git a/arch/powerpc/kernel/ppc_ksyms.c b/arch/powerpc/kernel/ppc_ksyms.c index 47d6f7e2ea9..5dcf4ba05ee 100644 --- a/arch/powerpc/kernel/ppc_ksyms.c +++ b/arch/powerpc/kernel/ppc_ksyms.c @@ -44,6 +44,7 @@ #include <asm/cputable.h> #include <asm/btext.h> #include <asm/div64.h> +#include <asm/signal.h> #ifdef CONFIG_8xx #include <asm/commproc.h> @@ -56,7 +57,6 @@ extern void machine_check_exception(struct pt_regs *regs); extern void alignment_exception(struct pt_regs *regs); extern void program_check_exception(struct pt_regs *regs); extern void single_step_exception(struct pt_regs *regs); -extern int do_signal(sigset_t *, struct pt_regs *); extern int pmac_newworld; extern int sys_sigreturn(struct pt_regs *regs); @@ -188,9 +188,6 @@ EXPORT_SYMBOL(adb_try_handler_change); EXPORT_SYMBOL(cuda_request); EXPORT_SYMBOL(cuda_poll); #endif /* CONFIG_ADB_CUDA */ -#if defined(CONFIG_PPC_MULTIPLATFORM) && defined(CONFIG_PPC32) -EXPORT_SYMBOL(_machine); -#endif #ifdef CONFIG_PPC_PMAC EXPORT_SYMBOL(sys_ctrler); #endif diff --git a/arch/ppc64/kernel/proc_ppc64.c b/arch/powerpc/kernel/proc_ppc64.c index 24e955ee948..a1c19502fe8 100644 --- a/arch/ppc64/kernel/proc_ppc64.c +++ b/arch/powerpc/kernel/proc_ppc64.c @@ -1,18 +1,16 @@ /* - * arch/ppc64/kernel/proc_ppc64.c - * * Copyright (C) 2001 Mike Corrigan & Dave Engebretsen IBM Corporation - * + * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by * the Free Software Foundation; either version 2 of the License, or * (at your option) any later version. - * + * * This program is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * GNU General Public License for more details. - * + * * You should have received a copy of the GNU General Public License * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA @@ -53,7 +51,7 @@ static int __init proc_ppc64_create(void) if (!root) return 1; - if (!(systemcfg->platform & (PLATFORM_PSERIES | PLATFORM_CELL))) + if (!(platform_is_pseries() || _machine == PLATFORM_CELL)) return 0; if (!proc_mkdir("rtas", root)) @@ -74,7 +72,7 @@ static int __init proc_ppc64_init(void) if (!pde) return 1; pde->nlink = 1; - pde->data = systemcfg; + pde->data = _systemcfg; pde->size = PAGE_SIZE; pde->proc_fops = &page_map_fops; diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c index f645adb5753..6a5b468edb4 100644 --- a/arch/powerpc/kernel/prom.c +++ b/arch/powerpc/kernel/prom.c @@ -48,9 +48,6 @@ #include <asm/machdep.h> #include <asm/pSeries_reconfig.h> #include <asm/pci-bridge.h> -#ifdef CONFIG_PPC64 -#include <asm/systemcfg.h> -#endif #ifdef DEBUG #define DBG(fmt...) printk(KERN_ERR fmt) @@ -74,10 +71,6 @@ struct isa_reg_property { typedef int interpret_func(struct device_node *, unsigned long *, int, int, int); -extern struct rtas_t rtas; -extern struct lmb lmb; -extern unsigned long klimit; - static int __initdata dt_root_addr_cells; static int __initdata dt_root_size_cells; @@ -391,7 +384,7 @@ static int __devinit finish_node_interrupts(struct device_node *np, #ifdef CONFIG_PPC64 /* We offset irq numbers for the u3 MPIC by 128 in PowerMac */ - if (systemcfg->platform == PLATFORM_POWERMAC && ic && ic->parent) { + if (_machine == PLATFORM_POWERMAC && ic && ic->parent) { char *name = get_property(ic->parent, "name", NULL); if (name && !strcmp(name, "u3")) np->intrs[intrcount].line += 128; @@ -1087,9 +1080,9 @@ void __init unflatten_device_tree(void) static int __init early_init_dt_scan_cpus(unsigned long node, const char *uname, int depth, void *data) { - char *type = of_get_flat_dt_prop(node, "device_type", NULL); u32 *prop; - unsigned long size = 0; + unsigned long size; + char *type = of_get_flat_dt_prop(node, "device_type", &size); /* We are scanning "cpu" nodes only */ if (type == NULL || strcmp(type, "cpu") != 0) @@ -1115,7 +1108,7 @@ static int __init early_init_dt_scan_cpus(unsigned long node, #ifdef CONFIG_ALTIVEC /* Check if we have a VMX and eventually update CPU features */ - prop = (u32 *)of_get_flat_dt_prop(node, "ibm,vmx", &size); + prop = (u32 *)of_get_flat_dt_prop(node, "ibm,vmx", NULL); if (prop && (*prop) > 0) { cur_cpu_spec->cpu_features |= CPU_FTR_ALTIVEC; cur_cpu_spec->cpu_user_features |= PPC_FEATURE_HAS_ALTIVEC; @@ -1161,13 +1154,9 @@ static int __init early_init_dt_scan_chosen(unsigned long node, prop = (u32 *)of_get_flat_dt_prop(node, "linux,platform", NULL); if (prop == NULL) return 0; -#ifdef CONFIG_PPC64 - systemcfg->platform = *prop; -#else #ifdef CONFIG_PPC_MULTIPLATFORM _machine = *prop; #endif -#endif #ifdef CONFIG_PPC64 /* check if iommu is forced on or off */ @@ -1264,7 +1253,14 @@ static int __init early_init_dt_scan_memory(unsigned long node, unsigned long l; /* We are scanning "memory" nodes only */ - if (type == NULL || strcmp(type, "memory") != 0) + if (type == NULL) { + /* + * The longtrail doesn't have a device_type on the + * /memory node, so look for the node called /memory@0. + */ + if (depth != 1 || strcmp(uname, "memory@0") != 0) + return 0; + } else if (strcmp(type, "memory") != 0) return 0; reg = (cell_t *)of_get_flat_dt_prop(node, "reg", &l); @@ -1339,9 +1335,6 @@ void __init early_init_devtree(void *params) of_scan_flat_dt(early_init_dt_scan_memory, NULL); lmb_enforce_memory_limit(memory_limit); lmb_analyze(); -#ifdef CONFIG_PPC64 - systemcfg->physicalMemorySize = lmb_phys_mem_size(); -#endif lmb_reserve(0, __pa(klimit)); DBG("Phys. mem: %lx\n", lmb_phys_mem_size()); @@ -1908,7 +1901,7 @@ static int of_finish_dynamic_node(struct device_node *node, /* We don't support that function on PowerMac, at least * not yet */ - if (systemcfg->platform == PLATFORM_POWERMAC) + if (_machine == PLATFORM_POWERMAC) return -ENODEV; /* fix up new node's linux_phandle field */ @@ -1992,9 +1985,11 @@ int prom_add_property(struct device_node* np, struct property* prop) *next = prop; write_unlock(&devtree_lock); +#ifdef CONFIG_PROC_DEVICETREE /* try to add to proc as well if it was initialized */ if (np->pde) proc_device_tree_add_prop(np->pde, prop); +#endif /* CONFIG_PROC_DEVICETREE */ return 0; } diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c index 6dc33d19fc2..4ce0105c308 100644 --- a/arch/powerpc/kernel/prom_init.c +++ b/arch/powerpc/kernel/prom_init.c @@ -94,11 +94,17 @@ extern const struct linux_logo logo_linux_clut224; #ifdef CONFIG_PPC64 #define RELOC(x) (*PTRRELOC(&(x))) #define ADDR(x) (u32) add_reloc_offset((unsigned long)(x)) +#define OF_WORKAROUNDS 0 #else #define RELOC(x) (x) #define ADDR(x) (u32) (x) +#define OF_WORKAROUNDS of_workarounds +int of_workarounds; #endif +#define OF_WA_CLAIM 1 /* do phys/virt claim separately, then map */ +#define OF_WA_LONGTRAIL 2 /* work around longtrail bugs */ + #define PROM_BUG() do { \ prom_printf("kernel BUG at %s line 0x%x!\n", \ RELOC(__FILE__), __LINE__); \ @@ -111,11 +117,6 @@ extern const struct linux_logo logo_linux_clut224; #define prom_debug(x...) #endif -#ifdef CONFIG_PPC32 -#define PLATFORM_POWERMAC _MACH_Pmac -#define PLATFORM_CHRP _MACH_chrp -#endif - typedef u32 prom_arg_t; @@ -128,10 +129,11 @@ struct prom_args { struct prom_t { ihandle root; - ihandle chosen; + phandle chosen; int cpu; ihandle stdout; ihandle mmumap; + ihandle memory; }; struct mem_map_entry { @@ -360,16 +362,36 @@ static void __init prom_printf(const char *format, ...) static unsigned int __init prom_claim(unsigned long virt, unsigned long size, unsigned long align) { - int ret; struct prom_t *_prom = &RELOC(prom); - ret = call_prom("claim", 3, 1, (prom_arg_t)virt, (prom_arg_t)size, - (prom_arg_t)align); - if (ret != -1 && _prom->mmumap != 0) - /* old pmacs need us to map as well */ + if (align == 0 && (OF_WORKAROUNDS & OF_WA_CLAIM)) { + /* + * Old OF requires we claim physical and virtual separately + * and then map explicitly (assuming virtual mode) + */ + int ret; + prom_arg_t result; + + ret = call_prom_ret("call-method", 5, 2, &result, + ADDR("claim"), _prom->memory, + align, size, virt); + if (ret != 0 || result == -1) + return -1; + ret = call_prom_ret("call-method", 5, 2, &result, + ADDR("claim"), _prom->mmumap, + align, size, virt); + if (ret != 0) { + call_prom("call-method", 4, 1, ADDR("release"), + _prom->memory, size, virt); + return -1; + } + /* the 0x12 is M (coherence) + PP == read/write */ call_prom("call-method", 6, 1, - ADDR("map"), _prom->mmumap, 0, size, virt, virt); - return ret; + ADDR("map"), _prom->mmumap, 0x12, size, virt, virt); + return virt; + } + return call_prom("claim", 3, 1, (prom_arg_t)virt, (prom_arg_t)size, + (prom_arg_t)align); } static void __init __attribute__((noreturn)) prom_panic(const char *reason) @@ -415,11 +437,52 @@ static int inline prom_getproplen(phandle node, const char *pname) return call_prom("getproplen", 2, 1, node, ADDR(pname)); } -static int inline prom_setprop(phandle node, const char *pname, - void *value, size_t valuelen) +static void add_string(char **str, const char *q) { - return call_prom("setprop", 4, 1, node, ADDR(pname), - (u32)(unsigned long) value, (u32) valuelen); + char *p = *str; + + while (*q) + *p++ = *q++; + *p++ = ' '; + *str = p; +} + +static char *tohex(unsigned int x) +{ + static char digits[] = "0123456789abcdef"; + static char result[9]; + int i; + + result[8] = 0; + i = 8; + do { + --i; + result[i] = digits[x & 0xf]; + x >>= 4; + } while (x != 0 && i > 0); + return &result[i]; +} + +static int __init prom_setprop(phandle node, const char *nodename, + const char *pname, void *value, size_t valuelen) +{ + char cmd[256], *p; + + if (!(OF_WORKAROUNDS & OF_WA_LONGTRAIL)) + return call_prom("setprop", 4, 1, node, ADDR(pname), + (u32)(unsigned long) value, (u32) valuelen); + + /* gah... setprop doesn't work on longtrail, have to use interpret */ + p = cmd; + add_string(&p, "dev"); + add_string(&p, nodename); + add_string(&p, tohex((u32)(unsigned long) value)); + add_string(&p, tohex(valuelen)); + add_string(&p, tohex(ADDR(pname))); + add_string(&p, tohex(strlen(RELOC(pname)))); + add_string(&p, "property"); + *p = 0; + return call_prom("interpret", 1, 1, (u32)(unsigned long) cmd); } /* We can't use the standard versions because of RELOC headaches. */ @@ -980,7 +1043,7 @@ static void __init prom_instantiate_rtas(void) rtas_inst = call_prom("open", 1, 1, ADDR("/rtas")); if (!IHANDLE_VALID(rtas_inst)) { - prom_printf("opening rtas package failed"); + prom_printf("opening rtas package failed (%x)\n", rtas_inst); return; } @@ -988,7 +1051,7 @@ static void __init prom_instantiate_rtas(void) if (call_prom_ret("call-method", 3, 2, &entry, ADDR("instantiate-rtas"), - rtas_inst, base) == PROM_ERROR + rtas_inst, base) != 0 || entry == 0) { prom_printf(" failed\n"); return; @@ -997,8 +1060,10 @@ static void __init prom_instantiate_rtas(void) reserve_mem(base, size); - prom_setprop(rtas_node, "linux,rtas-base", &base, sizeof(base)); - prom_setprop(rtas_node, "linux,rtas-entry", &entry, sizeof(entry)); + prom_setprop(rtas_node, "/rtas", "linux,rtas-base", + &base, sizeof(base)); + prom_setprop(rtas_node, "/rtas", "linux,rtas-entry", + &entry, sizeof(entry)); prom_debug("rtas base = 0x%x\n", base); prom_debug("rtas entry = 0x%x\n", entry); @@ -1089,10 +1154,6 @@ static void __init prom_initialize_tce_table(void) if (base < local_alloc_bottom) local_alloc_bottom = base; - /* Save away the TCE table attributes for later use. */ - prom_setprop(node, "linux,tce-base", &base, sizeof(base)); - prom_setprop(node, "linux,tce-size", &minsize, sizeof(minsize)); - /* It seems OF doesn't null-terminate the path :-( */ memset(path, 0, sizeof(path)); /* Call OF to setup the TCE hardware */ @@ -1101,6 +1162,10 @@ static void __init prom_initialize_tce_table(void) prom_printf("package-to-path failed\n"); } + /* Save away the TCE table attributes for later use. */ + prom_setprop(node, path, "linux,tce-base", &base, sizeof(base)); + prom_setprop(node, path, "linux,tce-size", &minsize, sizeof(minsize)); + prom_debug("TCE table: %s\n", path); prom_debug("\tnode = 0x%x\n", node); prom_debug("\tbase = 0x%x\n", base); @@ -1342,6 +1407,7 @@ static void __init prom_init_client_services(unsigned long pp) /* * For really old powermacs, we need to map things we claim. * For that, we need the ihandle of the mmu. + * Also, on the longtrail, we need to work around other bugs. */ static void __init prom_find_mmu(void) { @@ -1355,12 +1421,19 @@ static void __init prom_find_mmu(void) if (prom_getprop(oprom, "model", version, sizeof(version)) <= 0) return; version[sizeof(version) - 1] = 0; - prom_printf("OF version is '%s'\n", version); /* XXX might need to add other versions here */ - if (strcmp(version, "Open Firmware, 1.0.5") != 0) + if (strcmp(version, "Open Firmware, 1.0.5") == 0) + of_workarounds = OF_WA_CLAIM; + else if (strncmp(version, "FirmWorks,3.", 12) == 0) { + of_workarounds = OF_WA_CLAIM | OF_WA_LONGTRAIL; + call_prom("interpret", 1, 1, "dev /memory 0 to allow-reclaim"); + } else return; + _prom->memory = call_prom("open", 1, 1, ADDR("/memory")); prom_getprop(_prom->chosen, "mmu", &_prom->mmumap, sizeof(_prom->mmumap)); + if (!IHANDLE_VALID(_prom->memory) || !IHANDLE_VALID(_prom->mmumap)) + of_workarounds &= ~OF_WA_CLAIM; /* hmmm */ } #else #define prom_find_mmu() @@ -1382,16 +1455,17 @@ static void __init prom_init_stdout(void) memset(path, 0, 256); call_prom("instance-to-path", 3, 1, _prom->stdout, path, 255); val = call_prom("instance-to-package", 1, 1, _prom->stdout); - prom_setprop(_prom->chosen, "linux,stdout-package", &val, sizeof(val)); + prom_setprop(_prom->chosen, "/chosen", "linux,stdout-package", + &val, sizeof(val)); prom_printf("OF stdout device is: %s\n", RELOC(of_stdout_device)); - prom_setprop(_prom->chosen, "linux,stdout-path", - RELOC(of_stdout_device), strlen(RELOC(of_stdout_device))+1); + prom_setprop(_prom->chosen, "/chosen", "linux,stdout-path", + path, strlen(path) + 1); /* If it's a display, note it */ memset(type, 0, sizeof(type)); prom_getprop(val, "device_type", type, sizeof(type)); if (strcmp(type, RELOC("display")) == 0) - prom_setprop(val, "linux,boot-display", NULL, 0); + prom_setprop(val, path, "linux,boot-display", NULL, 0); } static void __init prom_close_stdin(void) @@ -1514,7 +1588,7 @@ static void __init prom_check_displays(void) /* Success */ prom_printf("done\n"); - prom_setprop(node, "linux,opened", NULL, 0); + prom_setprop(node, path, "linux,opened", NULL, 0); /* Setup a usable color table when the appropriate * method is available. Should update this to set-colors */ @@ -1884,9 +1958,11 @@ static void __init fixup_device_tree(void) /* interrupt on this revision of u3 is number 0 and level */ interrupts[0] = 0; interrupts[1] = 1; - prom_setprop(i2c, "interrupts", &interrupts, sizeof(interrupts)); + prom_setprop(i2c, "/u3@0,f8000000/i2c@f8001000", "interrupts", + &interrupts, sizeof(interrupts)); parent = (u32)mpic; - prom_setprop(i2c, "interrupt-parent", &parent, sizeof(parent)); + prom_setprop(i2c, "/u3@0,f8000000/i2c@f8001000", "interrupt-parent", + &parent, sizeof(parent)); #endif } @@ -1922,11 +1998,11 @@ static void __init prom_check_initrd(unsigned long r3, unsigned long r4) RELOC(prom_initrd_end) = RELOC(prom_initrd_start) + r4; val = RELOC(prom_initrd_start); - prom_setprop(_prom->chosen, "linux,initrd-start", &val, - sizeof(val)); + prom_setprop(_prom->chosen, "/chosen", "linux,initrd-start", + &val, sizeof(val)); val = RELOC(prom_initrd_end); - prom_setprop(_prom->chosen, "linux,initrd-end", &val, - sizeof(val)); + prom_setprop(_prom->chosen, "/chosen", "linux,initrd-end", + &val, sizeof(val)); reserve_mem(RELOC(prom_initrd_start), RELOC(prom_initrd_end) - RELOC(prom_initrd_start)); @@ -1969,14 +2045,15 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4, prom_init_client_services(pp); /* - * Init prom stdout device + * See if this OF is old enough that we need to do explicit maps + * and other workarounds */ - prom_init_stdout(); + prom_find_mmu(); /* - * See if this OF is old enough that we need to do explicit maps + * Init prom stdout device */ - prom_find_mmu(); + prom_init_stdout(); /* * Check for an initrd @@ -1989,14 +2066,15 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4, */ RELOC(of_platform) = prom_find_machine_type(); getprop_rval = RELOC(of_platform); - prom_setprop(_prom->chosen, "linux,platform", + prom_setprop(_prom->chosen, "/chosen", "linux,platform", &getprop_rval, sizeof(getprop_rval)); #ifdef CONFIG_PPC_PSERIES /* * On pSeries, inform the firmware about our capabilities */ - if (RELOC(of_platform) & PLATFORM_PSERIES) + if (RELOC(of_platform) == PLATFORM_PSERIES || + RELOC(of_platform) == PLATFORM_PSERIES_LPAR) prom_send_capabilities(); #endif @@ -2050,21 +2128,23 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4, * Fill in some infos for use by the kernel later on */ if (RELOC(prom_memory_limit)) - prom_setprop(_prom->chosen, "linux,memory-limit", + prom_setprop(_prom->chosen, "/chosen", "linux,memory-limit", &RELOC(prom_memory_limit), sizeof(prom_memory_limit)); #ifdef CONFIG_PPC64 if (RELOC(ppc64_iommu_off)) - prom_setprop(_prom->chosen, "linux,iommu-off", NULL, 0); + prom_setprop(_prom->chosen, "/chosen", "linux,iommu-off", + NULL, 0); if (RELOC(iommu_force_on)) - prom_setprop(_prom->chosen, "linux,iommu-force-on", NULL, 0); + prom_setprop(_prom->chosen, "/chosen", "linux,iommu-force-on", + NULL, 0); if (RELOC(prom_tce_alloc_start)) { - prom_setprop(_prom->chosen, "linux,tce-alloc-start", + prom_setprop(_prom->chosen, "/chosen", "linux,tce-alloc-start", &RELOC(prom_tce_alloc_start), sizeof(prom_tce_alloc_start)); - prom_setprop(_prom->chosen, "linux,tce-alloc-end", + prom_setprop(_prom->chosen, "/chosen", "linux,tce-alloc-end", &RELOC(prom_tce_alloc_end), sizeof(prom_tce_alloc_end)); } @@ -2081,8 +2161,13 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4, prom_printf("copying OF device tree ...\n"); flatten_device_tree(); - /* in case stdin is USB and still active on IBM machines... */ - prom_close_stdin(); + /* + * in case stdin is USB and still active on IBM machines... + * Unfortunately quiesce crashes on some powermacs if we have + * closed stdin already (in particular the powerbook 101). + */ + if (RELOC(of_platform) != PLATFORM_POWERMAC) + prom_close_stdin(); /* * Call OF "quiesce" method to shut down pending DMA's from diff --git a/arch/powerpc/kernel/rtas-proc.c b/arch/powerpc/kernel/rtas-proc.c index 5bdd5b079d9..ae1a36449cc 100644 --- a/arch/powerpc/kernel/rtas-proc.c +++ b/arch/powerpc/kernel/rtas-proc.c @@ -259,7 +259,7 @@ static int __init proc_rtas_init(void) { struct proc_dir_entry *entry; - if (!(systemcfg->platform & PLATFORM_PSERIES)) + if (_machine != PLATFORM_PSERIES && _machine != PLATFORM_PSERIES_LPAR) return 1; rtas_node = of_find_node_by_name(NULL, "rtas"); diff --git a/arch/powerpc/kernel/rtas.c b/arch/powerpc/kernel/rtas.c index 9d4e07f6f1e..4283fa33f78 100644 --- a/arch/powerpc/kernel/rtas.c +++ b/arch/powerpc/kernel/rtas.c @@ -29,9 +29,6 @@ #include <asm/delay.h> #include <asm/uaccess.h> #include <asm/lmb.h> -#ifdef CONFIG_PPC64 -#include <asm/systemcfg.h> -#endif struct rtas_t rtas = { .lock = SPIN_LOCK_UNLOCKED @@ -671,7 +668,7 @@ void __init rtas_initialize(void) * the stop-self token if any */ #ifdef CONFIG_PPC64 - if (systemcfg->platform == PLATFORM_PSERIES_LPAR) + if (_machine == PLATFORM_PSERIES_LPAR) rtas_region = min(lmb.rmo_size, RTAS_INSTANTIATE_MAX); #endif rtas_rmo_buf = lmb_alloc_base(RTAS_RMOBUF_MAX, PAGE_SIZE, rtas_region); diff --git a/arch/ppc64/kernel/rtas_pci.c b/arch/powerpc/kernel/rtas_pci.c index 3c3f19192fc..0e5a8e11665 100644 --- a/arch/ppc64/kernel/rtas_pci.c +++ b/arch/powerpc/kernel/rtas_pci.c @@ -5,19 +5,19 @@ * Copyright (C) 2003 Anton Blanchard <anton@au.ibm.com>, IBM * * RTAS specific routines for PCI. - * + * * Based on code from pci.c, chrp_pci.c and pSeries_pci.c * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by * the Free Software Foundation; either version 2 of the License, or * (at your option) any later version. - * + * * This program is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * GNU General Public License for more details. - * + * * You should have received a copy of the GNU General Public License * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA @@ -47,7 +47,7 @@ static int write_pci_config; static int ibm_read_pci_config; static int ibm_write_pci_config; -static int config_access_valid(struct pci_dn *dn, int where) +static inline int config_access_valid(struct pci_dn *dn, int where) { if (where < 256) return 1; @@ -72,16 +72,14 @@ static int of_device_available(struct device_node * dn) return 0; } -static int rtas_read_config(struct device_node *dn, int where, int size, u32 *val) +static int rtas_read_config(struct pci_dn *pdn, int where, int size, u32 *val) { int returnval = -1; unsigned long buid, addr; int ret; - struct pci_dn *pdn; - if (!dn || !dn->data) + if (!pdn) return PCIBIOS_DEVICE_NOT_FOUND; - pdn = dn->data; if (!config_access_valid(pdn, where)) return PCIBIOS_BAD_REGISTER_NUMBER; @@ -90,7 +88,7 @@ static int rtas_read_config(struct device_node *dn, int where, int size, u32 *va buid = pdn->phb->buid; if (buid) { ret = rtas_call(ibm_read_pci_config, 4, 2, &returnval, - addr, buid >> 32, buid & 0xffffffff, size); + addr, BUID_HI(buid), BUID_LO(buid), size); } else { ret = rtas_call(read_pci_config, 2, 2, &returnval, addr, size); } @@ -100,7 +98,7 @@ static int rtas_read_config(struct device_node *dn, int where, int size, u32 *va return PCIBIOS_DEVICE_NOT_FOUND; if (returnval == EEH_IO_ERROR_VALUE(size) && - eeh_dn_check_failure (dn, NULL)) + eeh_dn_check_failure (pdn->node, NULL)) return PCIBIOS_DEVICE_NOT_FOUND; return PCIBIOS_SUCCESSFUL; @@ -118,23 +116,23 @@ static int rtas_pci_read_config(struct pci_bus *bus, busdn = bus->sysdata; /* must be a phb */ /* Search only direct children of the bus */ - for (dn = busdn->child; dn; dn = dn->sibling) - if (dn->data && PCI_DN(dn)->devfn == devfn + for (dn = busdn->child; dn; dn = dn->sibling) { + struct pci_dn *pdn = PCI_DN(dn); + if (pdn && pdn->devfn == devfn && of_device_available(dn)) - return rtas_read_config(dn, where, size, val); + return rtas_read_config(pdn, where, size, val); + } return PCIBIOS_DEVICE_NOT_FOUND; } -int rtas_write_config(struct device_node *dn, int where, int size, u32 val) +int rtas_write_config(struct pci_dn *pdn, int where, int size, u32 val) { unsigned long buid, addr; int ret; - struct pci_dn *pdn; - if (!dn || !dn->data) + if (!pdn) return PCIBIOS_DEVICE_NOT_FOUND; - pdn = dn->data; if (!config_access_valid(pdn, where)) return PCIBIOS_BAD_REGISTER_NUMBER; @@ -142,7 +140,8 @@ int rtas_write_config(struct device_node *dn, int where, int size, u32 val) (pdn->devfn << 8) | (where & 0xff); buid = pdn->phb->buid; if (buid) { - ret = rtas_call(ibm_write_pci_config, 5, 1, NULL, addr, buid >> 32, buid & 0xffffffff, size, (ulong) val); + ret = rtas_call(ibm_write_pci_config, 5, 1, NULL, addr, + BUID_HI(buid), BUID_LO(buid), size, (ulong) val); } else { ret = rtas_call(write_pci_config, 3, 1, NULL, addr, size, (ulong)val); } @@ -165,10 +164,12 @@ static int rtas_pci_write_config(struct pci_bus *bus, busdn = bus->sysdata; /* must be a phb */ /* Search only direct children of the bus */ - for (dn = busdn->child; dn; dn = dn->sibling) - if (dn->data && PCI_DN(dn)->devfn == devfn + for (dn = busdn->child; dn; dn = dn->sibling) { + struct pci_dn *pdn = PCI_DN(dn); + if (pdn && pdn->devfn == devfn && of_device_available(dn)) - return rtas_write_config(dn, where, size, val); + return rtas_write_config(pdn, where, size, val); + } return PCIBIOS_DEVICE_NOT_FOUND; } @@ -221,7 +222,7 @@ static void python_countermeasures(struct device_node *dev, /* Python's register file is 1 MB in size. */ chip_regs = ioremap(reg_struct.address & ~(0xfffffUL), 0x100000); - /* + /* * Firmware doesn't always clear this bit which is critical * for good performance - Anton */ @@ -292,7 +293,7 @@ static int phb_set_bus_ranges(struct device_node *dev, if (bus_range == NULL || len < 2 * sizeof(int)) { return 1; } - + phb->first_busno = bus_range[0]; phb->last_busno = bus_range[1]; diff --git a/arch/powerpc/kernel/setup-common.c b/arch/powerpc/kernel/setup-common.c index e22856ecb5a..bae4bff138f 100644 --- a/arch/powerpc/kernel/setup-common.c +++ b/arch/powerpc/kernel/setup-common.c @@ -33,6 +33,7 @@ #include <asm/io.h> #include <asm/prom.h> #include <asm/processor.h> +#include <asm/systemcfg.h> #include <asm/pgtable.h> #include <asm/smp.h> #include <asm/elf.h> @@ -51,6 +52,9 @@ #include <asm/page.h> #include <asm/mmu.h> #include <asm/lmb.h> +#include <asm/xmon.h> + +#include "setup.h" #undef DEBUG @@ -60,6 +64,13 @@ #define DBG(fmt...) #endif +#ifdef CONFIG_PPC_MULTIPLATFORM +int _machine = 0; +EXPORT_SYMBOL(_machine); +#endif + +unsigned long klimit = (unsigned long) _end; + /* * This still seems to be needed... -- paulus */ @@ -510,8 +521,8 @@ void __init smp_setup_cpu_maps(void) * On pSeries LPAR, we need to know how many cpus * could possibly be added to this partition. */ - if (systemcfg->platform == PLATFORM_PSERIES_LPAR && - (dn = of_find_node_by_path("/rtas"))) { + if (_machine == PLATFORM_PSERIES_LPAR && + (dn = of_find_node_by_path("/rtas"))) { int num_addr_cell, num_size_cell, maxcpus; unsigned int *ireg; @@ -555,7 +566,27 @@ void __init smp_setup_cpu_maps(void) cpu_set(cpu ^ 0x1, cpu_sibling_map[cpu]); } - systemcfg->processorCount = num_present_cpus(); + _systemcfg->processorCount = num_present_cpus(); #endif /* CONFIG_PPC64 */ } #endif /* CONFIG_SMP */ + +#ifdef CONFIG_XMON +static int __init early_xmon(char *p) +{ + /* ensure xmon is enabled */ + if (p) { + if (strncmp(p, "on", 2) == 0) + xmon_init(1); + if (strncmp(p, "off", 3) == 0) + xmon_init(0); + if (strncmp(p, "early", 5) != 0) + return 0; + } + xmon_init(1); + debugger(NULL); + + return 0; +} +early_param("xmon", early_xmon); +#endif diff --git a/arch/powerpc/kernel/setup.h b/arch/powerpc/kernel/setup.h new file mode 100644 index 00000000000..2ebba755272 --- /dev/null +++ b/arch/powerpc/kernel/setup.h @@ -0,0 +1,6 @@ +#ifndef _POWERPC_KERNEL_SETUP_H +#define _POWERPC_KERNEL_SETUP_H + +void check_for_initrd(void); + +#endif /* _POWERPC_KERNEL_SETUP_H */ diff --git a/arch/powerpc/kernel/setup_32.c b/arch/powerpc/kernel/setup_32.c index 3af2631e3fa..c98cfcc9cd9 100644 --- a/arch/powerpc/kernel/setup_32.c +++ b/arch/powerpc/kernel/setup_32.c @@ -40,6 +40,8 @@ #include <asm/xmon.h> #include <asm/time.h> +#include "setup.h" + #define DBG(fmt...) #if defined CONFIG_KGDB @@ -70,8 +72,6 @@ unsigned int DMA_MODE_WRITE; int have_of = 1; #ifdef CONFIG_PPC_MULTIPLATFORM -int _machine = 0; - extern void prep_init(void); extern void pmac_init(void); extern void chrp_init(void); @@ -279,7 +279,6 @@ arch_initcall(ppc_init); /* Warning, IO base is not yet inited */ void __init setup_arch(char **cmdline_p) { - extern char *klimit; extern void do_init_bootmem(void); /* so udelay does something sensible, assume <= 1000 bogomips */ @@ -303,14 +302,9 @@ void __init setup_arch(char **cmdline_p) pmac_feature_init(); /* New cool way */ #endif -#ifdef CONFIG_XMON - xmon_map_scc(); - if (strstr(cmd_line, "xmon")) { - xmon_init(1); - debugger(NULL); - } -#endif /* CONFIG_XMON */ - if ( ppc_md.progress ) ppc_md.progress("setup_arch: enter", 0x3eab); +#ifdef CONFIG_XMON_DEFAULT + xmon_init(1); +#endif #if defined(CONFIG_KGDB) if (ppc_md.kgdb_map_scc) @@ -343,7 +337,7 @@ void __init setup_arch(char **cmdline_p) init_mm.start_code = PAGE_OFFSET; init_mm.end_code = (unsigned long) _etext; init_mm.end_data = (unsigned long) _edata; - init_mm.brk = (unsigned long) klimit; + init_mm.brk = klimit; /* Save unparsed command line copy for /proc/cmdline */ strlcpy(saved_command_line, cmd_line, COMMAND_LINE_SIZE); diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c index 0471e843b6c..6791668213e 100644 --- a/arch/powerpc/kernel/setup_64.c +++ b/arch/powerpc/kernel/setup_64.c @@ -61,6 +61,8 @@ #include <asm/xmon.h> #include <asm/udbg.h> +#include "setup.h" + #ifdef DEBUG #define DBG(fmt...) udbg_printf(fmt) #else @@ -94,15 +96,6 @@ extern void udbg_init_maple_realmode(void); do { udbg_putc = call_rtas_display_status_delay; } while(0) #endif -/* extern void *stab; */ -extern unsigned long klimit; - -extern void mm_init_ppc64(void); -extern void stab_initialize(unsigned long stab); -extern void htab_initialize(void); -extern void early_init_devtree(void *flat_dt); -extern void unflatten_device_tree(void); - int have_of = 1; int boot_cpuid = 0; int boot_cpuid_phys = 0; @@ -254,11 +247,10 @@ void __init early_setup(unsigned long dt_ptr) * Iterate all ppc_md structures until we find the proper * one for the current machine type */ - DBG("Probing machine type for platform %x...\n", - systemcfg->platform); + DBG("Probing machine type for platform %x...\n", _machine); for (mach = machines; *mach; mach++) { - if ((*mach)->probe(systemcfg->platform)) + if ((*mach)->probe(_machine)) break; } /* What can we do if we didn't find ? */ @@ -290,6 +282,28 @@ void __init early_setup(unsigned long dt_ptr) DBG(" <- early_setup()\n"); } +#ifdef CONFIG_SMP +void early_setup_secondary(void) +{ + struct paca_struct *lpaca = get_paca(); + + /* Mark enabled in PACA */ + lpaca->proc_enabled = 0; + + /* Initialize hash table for that CPU */ + htab_initialize_secondary(); + + /* Initialize STAB/SLB. We use a virtual address as it works + * in real mode on pSeries and we want a virutal address on + * iSeries anyway + */ + if (cpu_has_feature(CPU_FTR_SLB)) + slb_initialize(); + else + stab_initialize(lpaca->stab_addr); +} + +#endif /* CONFIG_SMP */ #if defined(CONFIG_SMP) || defined(CONFIG_KEXEC) void smp_release_cpus(void) @@ -315,7 +329,8 @@ void smp_release_cpus(void) #endif /* CONFIG_SMP || CONFIG_KEXEC */ /* - * Initialize some remaining members of the ppc64_caches and systemcfg structures + * Initialize some remaining members of the ppc64_caches and systemcfg + * structures * (at least until we get rid of them completely). This is mostly some * cache informations about the CPU that will be used by cache flush * routines and/or provided to userland @@ -340,7 +355,7 @@ static void __init initialize_cache_info(void) const char *dc, *ic; /* Then read cache informations */ - if (systemcfg->platform == PLATFORM_POWERMAC) { + if (_machine == PLATFORM_POWERMAC) { dc = "d-cache-block-size"; ic = "i-cache-block-size"; } else { @@ -360,8 +375,8 @@ static void __init initialize_cache_info(void) DBG("Argh, can't find dcache properties ! " "sizep: %p, lsizep: %p\n", sizep, lsizep); - systemcfg->dcache_size = ppc64_caches.dsize = size; - systemcfg->dcache_line_size = + _systemcfg->dcache_size = ppc64_caches.dsize = size; + _systemcfg->dcache_line_size = ppc64_caches.dline_size = lsize; ppc64_caches.log_dline_size = __ilog2(lsize); ppc64_caches.dlines_per_page = PAGE_SIZE / lsize; @@ -378,8 +393,8 @@ static void __init initialize_cache_info(void) DBG("Argh, can't find icache properties ! " "sizep: %p, lsizep: %p\n", sizep, lsizep); - systemcfg->icache_size = ppc64_caches.isize = size; - systemcfg->icache_line_size = + _systemcfg->icache_size = ppc64_caches.isize = size; + _systemcfg->icache_line_size = ppc64_caches.iline_size = lsize; ppc64_caches.log_iline_size = __ilog2(lsize); ppc64_caches.ilines_per_page = PAGE_SIZE / lsize; @@ -387,10 +402,12 @@ static void __init initialize_cache_info(void) } /* Add an eye catcher and the systemcfg layout version number */ - strcpy(systemcfg->eye_catcher, "SYSTEMCFG:PPC64"); - systemcfg->version.major = SYSTEMCFG_MAJOR; - systemcfg->version.minor = SYSTEMCFG_MINOR; - systemcfg->processor = mfspr(SPRN_PVR); + strcpy(_systemcfg->eye_catcher, "SYSTEMCFG:PPC64"); + _systemcfg->version.major = SYSTEMCFG_MAJOR; + _systemcfg->version.minor = SYSTEMCFG_MINOR; + _systemcfg->processor = mfspr(SPRN_PVR); + _systemcfg->platform = _machine; + _systemcfg->physicalMemorySize = lmb_phys_mem_size(); DBG(" <- initialize_cache_info()\n"); } @@ -479,10 +496,10 @@ void __init setup_system(void) printk("-----------------------------------------------------\n"); printk("ppc64_pft_size = 0x%lx\n", ppc64_pft_size); printk("ppc64_interrupt_controller = 0x%ld\n", ppc64_interrupt_controller); - printk("systemcfg = 0x%p\n", systemcfg); - printk("systemcfg->platform = 0x%x\n", systemcfg->platform); - printk("systemcfg->processorCount = 0x%lx\n", systemcfg->processorCount); - printk("systemcfg->physicalMemorySize = 0x%lx\n", systemcfg->physicalMemorySize); + printk("systemcfg = 0x%p\n", _systemcfg); + printk("systemcfg->platform = 0x%x\n", _systemcfg->platform); + printk("systemcfg->processorCount = 0x%lx\n", _systemcfg->processorCount); + printk("systemcfg->physicalMemorySize = 0x%lx\n", _systemcfg->physicalMemorySize); printk("ppc64_caches.dcache_line_size = 0x%x\n", ppc64_caches.dline_size); printk("ppc64_caches.icache_line_size = 0x%x\n", @@ -564,12 +581,12 @@ void __init setup_syscall_map(void) for (i = 0; i < __NR_syscalls; i++) { if (sys_call_table[i*2] != sys_ni_syscall) { count64++; - systemcfg->syscall_map_64[i >> 5] |= + _systemcfg->syscall_map_64[i >> 5] |= 0x80000000UL >> (i & 0x1f); } if (sys_call_table[i*2+1] != sys_ni_syscall) { count32++; - systemcfg->syscall_map_32[i >> 5] |= + _systemcfg->syscall_map_32[i >> 5] |= 0x80000000UL >> (i & 0x1f); } } @@ -858,26 +875,6 @@ int check_legacy_ioport(unsigned long base_port) } EXPORT_SYMBOL(check_legacy_ioport); -#ifdef CONFIG_XMON -static int __init early_xmon(char *p) -{ - /* ensure xmon is enabled */ - if (p) { - if (strncmp(p, "on", 2) == 0) - xmon_init(1); - if (strncmp(p, "off", 3) == 0) - xmon_init(0); - if (strncmp(p, "early", 5) != 0) - return 0; - } - xmon_init(1); - debugger(NULL); - - return 0; -} -early_param("xmon", early_xmon); -#endif - void cpu_die(void) { if (ppc_md.cpu_die) diff --git a/arch/powerpc/kernel/signal_32.c b/arch/powerpc/kernel/signal_32.c index 081d931eae4..a7c4515f320 100644 --- a/arch/powerpc/kernel/signal_32.c +++ b/arch/powerpc/kernel/signal_32.c @@ -42,6 +42,7 @@ #include <asm/uaccess.h> #include <asm/cacheflush.h> +#include <asm/sigcontext.h> #ifdef CONFIG_PPC64 #include "ppc32.h" #include <asm/unistd.h> diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index 5c330c3366e..e28a139c29d 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -44,6 +44,7 @@ #include <asm/cputable.h> #include <asm/system.h> #include <asm/mpic.h> +#include <asm/systemcfg.h> #ifdef CONFIG_PPC64 #include <asm/paca.h> #endif @@ -368,9 +369,11 @@ int generic_cpu_disable(void) if (cpu == boot_cpuid) return -EBUSY; - systemcfg->processorCount--; cpu_clear(cpu, cpu_online_map); +#ifdef CONFIG_PPC64 + _systemcfg->processorCount--; fixup_irqs(cpu_online_map); +#endif return 0; } @@ -388,9 +391,11 @@ int generic_cpu_enable(unsigned int cpu) while (!cpu_online(cpu)) cpu_relax(); +#ifdef CONFIG_PPC64 fixup_irqs(cpu_online_map); /* counter the irq disable in fixup_irqs */ local_irq_enable(); +#endif return 0; } @@ -419,7 +424,9 @@ void generic_mach_cpu_die(void) while (__get_cpu_var(cpu_state) != CPU_UP_PREPARE) cpu_relax(); +#ifdef CONFIG_PPC64 flush_tlb_pending(); +#endif cpu_set(cpu, cpu_online_map); local_irq_enable(); } @@ -510,6 +517,7 @@ int __devinit start_secondary(void *unused) smp_store_cpu_info(cpu); set_dec(tb_ticks_per_jiffy); + preempt_disable(); cpu_callin_map[cpu] = 1; smp_ops->setup_cpu(cpu); diff --git a/arch/powerpc/kernel/sys_ppc32.c b/arch/powerpc/kernel/sys_ppc32.c index a8210ed5c68..9c921d1c408 100644 --- a/arch/powerpc/kernel/sys_ppc32.c +++ b/arch/powerpc/kernel/sys_ppc32.c @@ -52,7 +52,6 @@ #include <asm/semaphore.h> #include <asm/time.h> #include <asm/mmu_context.h> -#include <asm/systemcfg.h> #include <asm/ppc-pci.h> /* readdir & getdents */ diff --git a/arch/ppc64/kernel/sysfs.c b/arch/powerpc/kernel/sysfs.c index e99ec62c2c5..850af198fb5 100644 --- a/arch/ppc64/kernel/sysfs.c +++ b/arch/powerpc/kernel/sysfs.c @@ -232,7 +232,7 @@ static void register_cpu_online(unsigned int cpu) sysdev_create_file(s, &attr_pmc7); if (cur_cpu_spec->num_pmcs >= 8) sysdev_create_file(s, &attr_pmc8); - + if (cpu_has_feature(CPU_FTR_SMT)) sysdev_create_file(s, &attr_purr); } diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index a6282b625b4..260b6ecd26a 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -271,13 +271,13 @@ static inline void update_gtod(u64 new_tb_stamp, u64 new_stamp_xsec, * tb_to_xs and stamp_xsec values are consistent. If not, then it * loops back and reads them again until this criteria is met. */ - ++(systemcfg->tb_update_count); + ++(_systemcfg->tb_update_count); smp_wmb(); - systemcfg->tb_orig_stamp = new_tb_stamp; - systemcfg->stamp_xsec = new_stamp_xsec; - systemcfg->tb_to_xs = new_tb_to_xs; + _systemcfg->tb_orig_stamp = new_tb_stamp; + _systemcfg->stamp_xsec = new_stamp_xsec; + _systemcfg->tb_to_xs = new_tb_to_xs; smp_wmb(); - ++(systemcfg->tb_update_count); + ++(_systemcfg->tb_update_count); #endif } @@ -357,8 +357,9 @@ static void iSeries_tb_recal(void) do_gtod.tb_ticks_per_sec = tb_ticks_per_sec; tb_to_xs = divres.result_low; do_gtod.varp->tb_to_xs = tb_to_xs; - systemcfg->tb_ticks_per_sec = tb_ticks_per_sec; - systemcfg->tb_to_xs = tb_to_xs; + _systemcfg->tb_ticks_per_sec = + tb_ticks_per_sec; + _systemcfg->tb_to_xs = tb_to_xs; } else { printk( "Titan recalibrate: FAILED (difference > 4 percent)\n" @@ -483,6 +484,8 @@ void __init smp_space_timers(unsigned int max_cpus) unsigned long offset = tb_ticks_per_jiffy / max_cpus; unsigned long previous_tb = per_cpu(last_jiffy, boot_cpuid); + /* make sure tb > per_cpu(last_jiffy, cpu) for all cpus always */ + previous_tb -= tb_ticks_per_jiffy; for_each_cpu(i) { if (i != boot_cpuid) { previous_tb += offset; @@ -559,8 +562,8 @@ int do_settimeofday(struct timespec *tv) update_gtod(tb_last_jiffy, new_xsec, do_gtod.varp->tb_to_xs); #ifdef CONFIG_PPC64 - systemcfg->tz_minuteswest = sys_tz.tz_minuteswest; - systemcfg->tz_dsttime = sys_tz.tz_dsttime; + _systemcfg->tz_minuteswest = sys_tz.tz_minuteswest; + _systemcfg->tz_dsttime = sys_tz.tz_dsttime; #endif write_sequnlock_irqrestore(&xtime_lock, flags); @@ -711,11 +714,11 @@ void __init time_init(void) do_gtod.varp->tb_to_xs = tb_to_xs; do_gtod.tb_to_us = tb_to_us; #ifdef CONFIG_PPC64 - systemcfg->tb_orig_stamp = tb_last_jiffy; - systemcfg->tb_update_count = 0; - systemcfg->tb_ticks_per_sec = tb_ticks_per_sec; - systemcfg->stamp_xsec = xtime.tv_sec * XSEC_PER_SEC; - systemcfg->tb_to_xs = tb_to_xs; + _systemcfg->tb_orig_stamp = tb_last_jiffy; + _systemcfg->tb_update_count = 0; + _systemcfg->tb_ticks_per_sec = tb_ticks_per_sec; + _systemcfg->stamp_xsec = xtime.tv_sec * XSEC_PER_SEC; + _systemcfg->tb_to_xs = tb_to_xs; #endif time_freq = 0; diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c index 0578f838760..2020bb7648f 100644 --- a/arch/powerpc/kernel/traps.c +++ b/arch/powerpc/kernel/traps.c @@ -129,7 +129,7 @@ int die(const char *str, struct pt_regs *regs, long err) nl = 1; #endif #ifdef CONFIG_PPC64 - switch (systemcfg->platform) { + switch (_machine) { case PLATFORM_PSERIES: printk("PSERIES "); nl = 1; diff --git a/arch/powerpc/lib/bitops.c b/arch/powerpc/lib/bitops.c index b67ce3004eb..f68ad71a018 100644 --- a/arch/powerpc/lib/bitops.c +++ b/arch/powerpc/lib/bitops.c @@ -41,7 +41,7 @@ unsigned long find_next_bit(const unsigned long *addr, unsigned long size, tmp = *p; found_first: - tmp &= (~0UL >> (64 - size)); + tmp &= (~0UL >> (BITS_PER_LONG - size)); if (tmp == 0UL) /* Are any bits set? */ return result + size; /* Nope. */ found_middle: diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c index 22e47487613..706e8a63ced 100644 --- a/arch/powerpc/mm/hash_utils_64.c +++ b/arch/powerpc/mm/hash_utils_64.c @@ -84,10 +84,11 @@ extern unsigned long dart_tablebase; #endif /* CONFIG_U3_DART */ +static unsigned long _SDR1; +struct mmu_psize_def mmu_psize_defs[MMU_PAGE_COUNT]; + hpte_t *htab_address; unsigned long htab_hash_mask; -unsigned long _SDR1; -struct mmu_psize_def mmu_psize_defs[MMU_PAGE_COUNT]; int mmu_linear_psize = MMU_PAGE_4K; int mmu_virtual_psize = MMU_PAGE_4K; #ifdef CONFIG_HUGETLB_PAGE @@ -165,7 +166,7 @@ int htab_bolt_mapping(unsigned long vstart, unsigned long vend, * normal insert callback here. */ #ifdef CONFIG_PPC_ISERIES - if (systemcfg->platform == PLATFORM_ISERIES_LPAR) + if (_machine == PLATFORM_ISERIES_LPAR) ret = iSeries_hpte_insert(hpteg, va, virt_to_abs(paddr), tmp_mode, @@ -174,7 +175,7 @@ int htab_bolt_mapping(unsigned long vstart, unsigned long vend, else #endif #ifdef CONFIG_PPC_PSERIES - if (systemcfg->platform & PLATFORM_LPAR) + if (_machine & PLATFORM_LPAR) ret = pSeries_lpar_hpte_insert(hpteg, va, virt_to_abs(paddr), tmp_mode, @@ -293,7 +294,7 @@ static void __init htab_init_page_sizes(void) * Not in the device-tree, let's fallback on known size * list for 16M capable GP & GR */ - if ((systemcfg->platform != PLATFORM_ISERIES_LPAR) && + if ((_machine != PLATFORM_ISERIES_LPAR) && cpu_has_feature(CPU_FTR_16M_PAGE)) memcpy(mmu_psize_defs, mmu_psize_defaults_gp, sizeof(mmu_psize_defaults_gp)); @@ -364,7 +365,7 @@ static int __init htab_dt_scan_pftsize(unsigned long node, static unsigned long __init htab_get_table_size(void) { - unsigned long rnd_mem_size, pteg_count; + unsigned long mem_size, rnd_mem_size, pteg_count; /* If hash size isn't already provided by the platform, we try to * retreive it from the device-tree. If it's not there neither, we @@ -376,8 +377,9 @@ static unsigned long __init htab_get_table_size(void) return 1UL << ppc64_pft_size; /* round mem_size up to next power of 2 */ - rnd_mem_size = 1UL << __ilog2(systemcfg->physicalMemorySize); - if (rnd_mem_size < systemcfg->physicalMemorySize) + mem_size = lmb_phys_mem_size(); + rnd_mem_size = 1UL << __ilog2(mem_size); + if (rnd_mem_size < mem_size) rnd_mem_size <<= 1; /* # pages / 2 */ @@ -386,6 +388,15 @@ static unsigned long __init htab_get_table_size(void) return pteg_count << 7; } +#ifdef CONFIG_MEMORY_HOTPLUG +void create_section_mapping(unsigned long start, unsigned long end) +{ + BUG_ON(htab_bolt_mapping(start, end, start, + _PAGE_ACCESSED | _PAGE_DIRTY | _PAGE_COHERENT | PP_RWXX, + mmu_linear_psize)); +} +#endif /* CONFIG_MEMORY_HOTPLUG */ + void __init htab_initialize(void) { unsigned long table, htab_size_bytes; @@ -410,7 +421,7 @@ void __init htab_initialize(void) htab_hash_mask = pteg_count - 1; - if (systemcfg->platform & PLATFORM_LPAR) { + if (platform_is_lpar()) { /* Using a hypervisor which owns the htab */ htab_address = NULL; _SDR1 = 0; @@ -431,6 +442,9 @@ void __init htab_initialize(void) /* Initialize the HPT with no entries */ memset((void *)table, 0, htab_size_bytes); + + /* Set SDR1 */ + mtspr(SPRN_SDR1, _SDR1); } mode_rw = _PAGE_ACCESSED | _PAGE_DIRTY | _PAGE_COHERENT | PP_RWXX; @@ -500,6 +514,12 @@ void __init htab_initialize(void) #undef KB #undef MB +void __init htab_initialize_secondary(void) +{ + if (!platform_is_lpar()) + mtspr(SPRN_SDR1, _SDR1); +} + /* * Called by asm hashtable.S for doing lazy icache flush */ diff --git a/arch/powerpc/mm/init_32.c b/arch/powerpc/mm/init_32.c index 4612a79dfb6..7d4b8b5f060 100644 --- a/arch/powerpc/mm/init_32.c +++ b/arch/powerpc/mm/init_32.c @@ -84,9 +84,6 @@ void MMU_init(void); /* XXX should be in current.h -- paulus */ extern struct task_struct *current_set[NR_CPUS]; -char *klimit = _end; -struct device_node *memory_node; - extern int init_bootmem_done; /* diff --git a/arch/powerpc/mm/init_64.c b/arch/powerpc/mm/init_64.c index ce974c83d88..1134f70f231 100644 --- a/arch/powerpc/mm/init_64.c +++ b/arch/powerpc/mm/init_64.c @@ -20,6 +20,8 @@ * */ +#undef DEBUG + #include <linux/config.h> #include <linux/signal.h> #include <linux/sched.h> @@ -64,6 +66,12 @@ #include <asm/vdso.h> #include <asm/imalloc.h> +#ifdef DEBUG +#define DBG(fmt...) printk(fmt) +#else +#define DBG(fmt...) +#endif + #if PGTABLE_RANGE > USER_VSID_RANGE #warning Limited user VSID range means pagetable space is wasted #endif @@ -72,8 +80,6 @@ #warning TASK_SIZE is smaller than it needs to be. #endif -unsigned long klimit = (unsigned long)_end; - /* max amount of RAM to use */ unsigned long __max_memory; @@ -188,14 +194,14 @@ static void zero_ctor(void *addr, kmem_cache_t *cache, unsigned long flags) } #ifdef CONFIG_PPC_64K_PAGES -static const int pgtable_cache_size[2] = { - PTE_TABLE_SIZE, PGD_TABLE_SIZE +static const unsigned int pgtable_cache_size[3] = { + PTE_TABLE_SIZE, PMD_TABLE_SIZE, PGD_TABLE_SIZE }; static const char *pgtable_cache_name[ARRAY_SIZE(pgtable_cache_size)] = { - "pte_pmd_cache", "pgd_cache", + "pte_pmd_cache", "pmd_cache", "pgd_cache", }; #else -static const int pgtable_cache_size[2] = { +static const unsigned int pgtable_cache_size[2] = { PTE_TABLE_SIZE, PMD_TABLE_SIZE }; static const char *pgtable_cache_name[ARRAY_SIZE(pgtable_cache_size)] = { @@ -213,6 +219,8 @@ void pgtable_cache_init(void) int size = pgtable_cache_size[i]; const char *name = pgtable_cache_name[i]; + DBG("Allocating page table cache %s (#%d) " + "for size: %08x...\n", name, i, size); pgtable_cache[i] = kmem_cache_create(name, size, size, SLAB_HWCACHE_ALIGN | diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c index 6f55efd9be9..1dd3cc69a49 100644 --- a/arch/powerpc/mm/mem.c +++ b/arch/powerpc/mm/mem.c @@ -110,6 +110,7 @@ EXPORT_SYMBOL(phys_mem_access_prot); void online_page(struct page *page) { ClearPageReserved(page); + set_page_count(page, 0); free_cold_page(page); totalram_pages++; num_physpages++; @@ -127,6 +128,9 @@ int __devinit add_memory(u64 start, u64 size) unsigned long start_pfn = start >> PAGE_SHIFT; unsigned long nr_pages = size >> PAGE_SHIFT; + start += KERNELBASE; + create_section_mapping(start, start + size); + /* this should work for most non-highmem platforms */ zone = pgdata->node_zones; diff --git a/arch/powerpc/mm/pgtable_64.c b/arch/powerpc/mm/pgtable_64.c index 900842451bd..c7f7bb6f30b 100644 --- a/arch/powerpc/mm/pgtable_64.c +++ b/arch/powerpc/mm/pgtable_64.c @@ -122,8 +122,11 @@ static int map_io_page(unsigned long ea, unsigned long pa, int flags) * */ if (htab_bolt_mapping(ea, ea + PAGE_SIZE, pa, flags, - mmu_virtual_psize)) - panic("Can't map bolted IO mapping"); + mmu_virtual_psize)) { + printk(KERN_ERR "Failed to do bolted mapping IO " + "memory at %016lx !\n", pa); + return -ENOMEM; + } } return 0; } diff --git a/arch/powerpc/mm/stab.c b/arch/powerpc/mm/stab.c index fa325dbf98f..cfbb4e1f966 100644 --- a/arch/powerpc/mm/stab.c +++ b/arch/powerpc/mm/stab.c @@ -20,6 +20,7 @@ #include <asm/cputable.h> #include <asm/lmb.h> #include <asm/abs_addr.h> +#include <asm/firmware.h> struct stab_entry { unsigned long esid_data; @@ -256,7 +257,7 @@ void stabs_alloc(void) paca[cpu].stab_addr = newstab; paca[cpu].stab_real = virt_to_abs(newstab); - printk(KERN_DEBUG "Segment table for CPU %d at 0x%lx " + printk(KERN_INFO "Segment table for CPU %d at 0x%lx " "virtual, 0x%lx absolute\n", cpu, paca[cpu].stab_addr, paca[cpu].stab_real); } @@ -270,10 +271,28 @@ void stabs_alloc(void) void stab_initialize(unsigned long stab) { unsigned long vsid = get_kernel_vsid(KERNELBASE); + unsigned long stabreal; asm volatile("isync; slbia; isync":::"memory"); make_ste(stab, GET_ESID(KERNELBASE), vsid); /* Order update */ asm volatile("sync":::"memory"); + + /* Set ASR */ + stabreal = get_paca()->stab_real | 0x1ul; + +#ifdef CONFIG_PPC_ISERIES + if (firmware_has_feature(FW_FEATURE_ISERIES)) { + HvCall1(HvCallBaseSetASR, stabreal); + return; + } +#endif /* CONFIG_PPC_ISERIES */ +#ifdef CONFIG_PPC_PSERIES + if (platform_is_lpar()) { + plpar_hcall_norets(H_SET_ASR, stabreal); + return; + } +#endif + mtspr(SPRN_ASR, stabreal); } diff --git a/arch/powerpc/oprofile/op_model_power4.c b/arch/powerpc/oprofile/op_model_power4.c index c4ee5478427..e3a024e324b 100644 --- a/arch/powerpc/oprofile/op_model_power4.c +++ b/arch/powerpc/oprofile/op_model_power4.c @@ -233,8 +233,7 @@ static unsigned long get_pc(struct pt_regs *regs) mmcra = mfspr(SPRN_MMCRA); /* Were we in the hypervisor? */ - if ((systemcfg->platform == PLATFORM_PSERIES_LPAR) && - (mmcra & MMCRA_SIHV)) + if (platform_is_lpar() && (mmcra & MMCRA_SIHV)) /* function descriptor madness */ return *((unsigned long *)hypervisor_bucket); diff --git a/arch/powerpc/platforms/chrp/setup.c b/arch/powerpc/platforms/chrp/setup.c index ecd32d5d85f..4099ddab920 100644 --- a/arch/powerpc/platforms/chrp/setup.c +++ b/arch/powerpc/platforms/chrp/setup.c @@ -361,7 +361,9 @@ static void __init chrp_find_openpic(void) printk(KERN_INFO "OpenPIC at %lx\n", opaddr); irq_count = NR_IRQS - NUM_ISA_INTERRUPTS - 4; /* leave room for IPIs */ - prom_get_irq_senses(init_senses, NUM_8259_INTERRUPTS, NR_IRQS - 4); + prom_get_irq_senses(init_senses, NUM_ISA_INTERRUPTS, NR_IRQS - 4); + /* i8259 cascade is always positive level */ + init_senses[0] = IRQ_SENSE_LEVEL | IRQ_POLARITY_POSITIVE; iranges = (unsigned int *) get_property(np, "interrupt-ranges", &len); if (iranges == NULL) diff --git a/arch/powerpc/platforms/iseries/irq.c b/arch/powerpc/platforms/iseries/irq.c index a06603d84a4..01090e9ce0c 100644 --- a/arch/powerpc/platforms/iseries/irq.c +++ b/arch/powerpc/platforms/iseries/irq.c @@ -103,6 +103,9 @@ static void intReceived(struct XmPciLpEvent *eventParm, struct pt_regs *regsParm) { int irq; +#ifdef CONFIG_IRQSTACKS + struct thread_info *curtp, *irqtp; +#endif ++Pci_Interrupt_Count; @@ -110,7 +113,20 @@ static void intReceived(struct XmPciLpEvent *eventParm, case XmPciLpEvent_SlotInterrupt: irq = eventParm->hvLpEvent.xCorrelationToken; /* Dispatch the interrupt handlers for this irq */ - ppc_irq_dispatch_handler(regsParm, irq); +#ifdef CONFIG_IRQSTACKS + /* Switch to the irq stack to handle this */ + curtp = current_thread_info(); + irqtp = hardirq_ctx[smp_processor_id()]; + if (curtp != irqtp) { + irqtp->task = curtp->task; + irqtp->flags = 0; + call___do_IRQ(irq, regsParm, irqtp); + irqtp->task = NULL; + if (irqtp->flags) + set_bits(irqtp->flags, &curtp->flags); + } else +#endif + __do_IRQ(irq, regsParm); HvCallPci_eoi(eventParm->eventData.slotInterrupt.busNumber, eventParm->eventData.slotInterrupt.subBusNumber, eventParm->eventData.slotInterrupt.deviceId); @@ -310,10 +326,8 @@ static void iSeries_disable_IRQ(unsigned int irq) } /* - * Need to define this so ppc_irq_dispatch_handler will NOT call - * enable_IRQ at the end of interrupt handling. However, this does - * nothing because there is not enough information provided to do - * the EOI HvCall. This is done by XmPciLpEvent.c + * This does nothing because there is not enough information + * provided to do the EOI HvCall. This is done by XmPciLpEvent.c */ static void iSeries_end_IRQ(unsigned int irq) { diff --git a/arch/powerpc/platforms/iseries/misc.S b/arch/powerpc/platforms/iseries/misc.S index 09f14522e17..dfe7aa1ba09 100644 --- a/arch/powerpc/platforms/iseries/misc.S +++ b/arch/powerpc/platforms/iseries/misc.S @@ -15,6 +15,7 @@ #include <asm/processor.h> #include <asm/asm-offsets.h> +#include <asm/ppc_asm.h> .text diff --git a/arch/powerpc/platforms/iseries/setup.c b/arch/powerpc/platforms/iseries/setup.c index 7f8f0cda6a7..6a29f301436 100644 --- a/arch/powerpc/platforms/iseries/setup.c +++ b/arch/powerpc/platforms/iseries/setup.c @@ -39,7 +39,8 @@ #include <asm/sections.h> #include <asm/iommu.h> #include <asm/firmware.h> - +#include <asm/systemcfg.h> +#include <asm/system.h> #include <asm/time.h> #include <asm/paca.h> #include <asm/cache.h> @@ -71,7 +72,7 @@ extern void hvlog(char *fmt, ...); #endif /* Function Prototypes */ -static void build_iSeries_Memory_Map(void); +static unsigned long build_iSeries_Memory_Map(void); static void iseries_shared_idle(void); static void iseries_dedicated_idle(void); #ifdef CONFIG_PCI @@ -84,7 +85,6 @@ static void iSeries_pci_final_fixup(void) { } int piranha_simulator; extern int rd_size; /* Defined in drivers/block/rd.c */ -extern unsigned long klimit; extern unsigned long embedded_sysmap_start; extern unsigned long embedded_sysmap_end; @@ -403,9 +403,11 @@ void mschunks_alloc(unsigned long num_chunks) * a table used to translate Linux's physical addresses to these * absolute addresses. Absolute addresses are needed when * communicating with the hypervisor (e.g. to build HPT entries) + * + * Returns the physical memory size */ -static void __init build_iSeries_Memory_Map(void) +static unsigned long __init build_iSeries_Memory_Map(void) { u32 loadAreaFirstChunk, loadAreaLastChunk, loadAreaSize; u32 nextPhysChunk; @@ -538,7 +540,7 @@ static void __init build_iSeries_Memory_Map(void) * which should be equal to * nextPhysChunk */ - systemcfg->physicalMemorySize = chunk_to_addr(nextPhysChunk); + return chunk_to_addr(nextPhysChunk); } /* @@ -564,8 +566,8 @@ static void __init iSeries_setup_arch(void) printk("Max physical processors = %d\n", itVpdAreas.xSlicMaxPhysicalProcs); - systemcfg->processor = xIoHriProcessorVpd[procIx].xPVR; - printk("Processor version = %x\n", systemcfg->processor); + _systemcfg->processor = xIoHriProcessorVpd[procIx].xPVR; + printk("Processor version = %x\n", _systemcfg->processor); } static void iSeries_show_cpuinfo(struct seq_file *m) @@ -702,7 +704,6 @@ static void iseries_shared_idle(void) static void iseries_dedicated_idle(void) { - long oldval; set_thread_flag(TIF_POLLING_NRFLAG); while (1) { @@ -929,7 +930,7 @@ void dt_cpus(struct iseries_flat_dt *dt) dt_end_node(dt); } -void build_flat_dt(struct iseries_flat_dt *dt) +void build_flat_dt(struct iseries_flat_dt *dt, unsigned long phys_mem_size) { u64 tmp[2]; @@ -945,7 +946,7 @@ void build_flat_dt(struct iseries_flat_dt *dt) dt_prop_str(dt, "name", "memory"); dt_prop_str(dt, "device_type", "memory"); tmp[0] = 0; - tmp[1] = systemcfg->physicalMemorySize; + tmp[1] = phys_mem_size; dt_prop_u64_list(dt, "reg", tmp, 2); dt_end_node(dt); @@ -965,13 +966,15 @@ void build_flat_dt(struct iseries_flat_dt *dt) void * __init iSeries_early_setup(void) { + unsigned long phys_mem_size; + iSeries_fixup_klimit(); /* * Initialize the table which translate Linux physical addresses to * AS/400 absolute addresses */ - build_iSeries_Memory_Map(); + phys_mem_size = build_iSeries_Memory_Map(); iSeries_get_cmdline(); @@ -981,7 +984,7 @@ void * __init iSeries_early_setup(void) /* Parse early parameters, in particular mem=x */ parse_early_param(); - build_flat_dt(&iseries_dt); + build_flat_dt(&iseries_dt, phys_mem_size); return (void *) __pa(&iseries_dt); } diff --git a/arch/powerpc/platforms/maple/pci.c b/arch/powerpc/platforms/maple/pci.c index 340c21caeae..895aeb3f75d 100644 --- a/arch/powerpc/platforms/maple/pci.c +++ b/arch/powerpc/platforms/maple/pci.c @@ -380,9 +380,6 @@ void __init maple_pcibios_fixup(void) for_each_pci_dev(dev) pci_read_irq_line(dev); - /* Do the mapping of the IO space */ - phbs_remap_io(); - DBG(" <- maple_pcibios_fixup\n"); } diff --git a/arch/powerpc/platforms/powermac/pci.c b/arch/powerpc/platforms/powermac/pci.c index 8f818d092e2..dfd41b9781a 100644 --- a/arch/powerpc/platforms/powermac/pci.c +++ b/arch/powerpc/platforms/powermac/pci.c @@ -918,9 +918,6 @@ void __init pmac_pci_init(void) PCI_DN(np)->busno = 0xf0; } - /* map in PCI I/O space */ - phbs_remap_io(); - /* pmac_check_ht_link(); */ /* Tell pci.c to not use the common resource allocation mechanism */ diff --git a/arch/powerpc/platforms/powermac/pic.c b/arch/powerpc/platforms/powermac/pic.c index 83a49e80ac2..90040c49494 100644 --- a/arch/powerpc/platforms/powermac/pic.c +++ b/arch/powerpc/platforms/powermac/pic.c @@ -74,6 +74,9 @@ static DEFINE_SPINLOCK(pmac_pic_lock); #define GATWICK_IRQ_POOL_SIZE 10 static struct interrupt_info gatwick_int_pool[GATWICK_IRQ_POOL_SIZE]; +#define NR_MASK_WORDS ((NR_IRQS + 31) / 32) +static unsigned long ppc_lost_interrupts[NR_MASK_WORDS]; + /* * Mark an irq as "lost". This is only used on the pmac * since it can lose interrupts (see pmac_set_irq_mask). diff --git a/arch/powerpc/platforms/powermac/smp.c b/arch/powerpc/platforms/powermac/smp.c index e1f9443cc87..957b0910342 100644 --- a/arch/powerpc/platforms/powermac/smp.c +++ b/arch/powerpc/platforms/powermac/smp.c @@ -305,9 +305,19 @@ static int __init smp_psurge_probe(void) psurge_start = ioremap(PSURGE_START, 4); psurge_pri_intr = ioremap(PSURGE_PRI_INTR, 4); - /* this is not actually strictly necessary -- paulus. */ - for (i = 1; i < ncpus; ++i) - smp_hw_index[i] = i; + /* + * This is necessary because OF doesn't know about the + * secondary cpu(s), and thus there aren't nodes in the + * device tree for them, and smp_setup_cpu_maps hasn't + * set their bits in cpu_possible_map and cpu_present_map. + */ + if (ncpus > NR_CPUS) + ncpus = NR_CPUS; + for (i = 1; i < ncpus ; ++i) { + cpu_set(i, cpu_present_map); + cpu_set(i, cpu_possible_map); + set_hard_smp_processor_id(i, i); + } if (ppc_md.progress) ppc_md.progress("smp_psurge_probe - done", 0x352); @@ -348,6 +358,7 @@ static void __init psurge_dual_sync_tb(int cpu_nr) int t; set_dec(tb_ticks_per_jiffy); + /* XXX fixme */ set_tb(0, 0); last_jiffy_stamp(cpu_nr) = 0; @@ -363,8 +374,6 @@ static void __init psurge_dual_sync_tb(int cpu_nr) /* now interrupt the secondary, starting both TBs */ psurge_set_ipi(1); - - smp_tb_synchronized = 1; } static struct irqaction psurge_irqaction = { @@ -625,9 +634,8 @@ void smp_core99_give_timebase(void) for (t = 100000; t > 0 && sec_tb_reset; --t) udelay(10); if (sec_tb_reset) + /* XXX BUG_ON here? */ printk(KERN_WARNING "Timeout waiting sync(2) on second CPU\n"); - else - smp_tb_synchronized = 1; /* Now, restart the timebase by leaving the GPIO to an open collector */ pmac_call_feature(PMAC_FTR_WRITE_GPIO, NULL, core99_tb_gpio, 0); @@ -810,19 +818,9 @@ static void __devinit smp_core99_setup_cpu(int cpu_nr) } -/* Core99 Macs (dual G4s and G5s) */ -struct smp_ops_t core99_smp_ops = { - .message_pass = smp_mpic_message_pass, - .probe = smp_core99_probe, - .kick_cpu = smp_core99_kick_cpu, - .setup_cpu = smp_core99_setup_cpu, - .give_timebase = smp_core99_give_timebase, - .take_timebase = smp_core99_take_timebase, -}; - #if defined(CONFIG_HOTPLUG_CPU) && defined(CONFIG_PPC32) -int __cpu_disable(void) +int smp_core99_cpu_disable(void) { cpu_clear(smp_processor_id(), cpu_online_map); @@ -846,7 +844,7 @@ void cpu_die(void) low_cpu_die(); } -void __cpu_die(unsigned int cpu) +void smp_core99_cpu_die(unsigned int cpu) { int timeout; @@ -858,8 +856,21 @@ void __cpu_die(unsigned int cpu) } msleep(1); } - cpu_callin_map[cpu] = 0; cpu_dead[cpu] = 0; } #endif + +/* Core99 Macs (dual G4s and G5s) */ +struct smp_ops_t core99_smp_ops = { + .message_pass = smp_mpic_message_pass, + .probe = smp_core99_probe, + .kick_cpu = smp_core99_kick_cpu, + .setup_cpu = smp_core99_setup_cpu, + .give_timebase = smp_core99_give_timebase, + .take_timebase = smp_core99_take_timebase, +#if defined(CONFIG_HOTPLUG_CPU) && defined(CONFIG_PPC32) + .cpu_disable = smp_core99_cpu_disable, + .cpu_die = smp_core99_cpu_die, +#endif +}; diff --git a/arch/powerpc/platforms/pseries/Makefile b/arch/powerpc/platforms/pseries/Makefile index b9938fece78..e7ca5b1f591 100644 --- a/arch/powerpc/platforms/pseries/Makefile +++ b/arch/powerpc/platforms/pseries/Makefile @@ -3,3 +3,5 @@ obj-y := pci.o lpar.o hvCall.o nvram.o reconfig.o \ obj-$(CONFIG_SMP) += smp.o obj-$(CONFIG_IBMVIO) += vio.o obj-$(CONFIG_XICS) += xics.o +obj-$(CONFIG_SCANLOG) += scanlog.o +obj-$(CONFIG_EEH) += eeh.o eeh_event.o diff --git a/arch/ppc64/kernel/eeh.c b/arch/powerpc/platforms/pseries/eeh.c index 035d1b14a20..79de2310e70 100644 --- a/arch/ppc64/kernel/eeh.c +++ b/arch/powerpc/platforms/pseries/eeh.c @@ -1,39 +1,37 @@ /* * eeh.c * Copyright (C) 2001 Dave Engebretsen & Todd Inglett IBM Corporation - * + * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by * the Free Software Foundation; either version 2 of the License, or * (at your option) any later version. - * + * * This program is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * GNU General Public License for more details. - * + * * You should have received a copy of the GNU General Public License * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ -#include <linux/bootmem.h> +#include <linux/delay.h> #include <linux/init.h> #include <linux/list.h> -#include <linux/mm.h> -#include <linux/notifier.h> #include <linux/pci.h> #include <linux/proc_fs.h> #include <linux/rbtree.h> #include <linux/seq_file.h> #include <linux/spinlock.h> +#include <asm/atomic.h> #include <asm/eeh.h> +#include <asm/eeh_event.h> #include <asm/io.h> #include <asm/machdep.h> -#include <asm/rtas.h> -#include <asm/atomic.h> -#include <asm/systemcfg.h> #include <asm/ppc-pci.h> +#include <asm/rtas.h> #undef DEBUG @@ -49,8 +47,8 @@ * were "empty": all reads return 0xff's and all writes are silently * ignored. EEH slot isolation events can be triggered by parity * errors on the address or data busses (e.g. during posted writes), - * which in turn might be caused by dust, vibration, humidity, - * radioactivity or plain-old failed hardware. + * which in turn might be caused by low voltage on the bus, dust, + * vibration, humidity, radioactivity or plain-old failed hardware. * * Note, however, that one of the leading causes of EEH slot * freeze events are buggy device drivers, buggy device microcode, @@ -71,26 +69,15 @@ * and sent out for processing. */ -/** Bus Unit ID macros; get low and hi 32-bits of the 64-bit BUID */ -#define BUID_HI(buid) ((buid) >> 32) -#define BUID_LO(buid) ((buid) & 0xffffffff) - -/* EEH event workqueue setup. */ -static DEFINE_SPINLOCK(eeh_eventlist_lock); -LIST_HEAD(eeh_eventlist); -static void eeh_event_handler(void *); -DECLARE_WORK(eeh_event_wq, eeh_event_handler, NULL); - -static struct notifier_block *eeh_notifier_chain; - -/* - * If a device driver keeps reading an MMIO register in an interrupt +/* If a device driver keeps reading an MMIO register in an interrupt * handler after a slot isolation event has occurred, we assume it * is broken and panic. This sets the threshold for how many read * attempts we allow before panicking. */ -#define EEH_MAX_FAILS 1000 -static atomic_t eeh_fail_count; +#define EEH_MAX_FAILS 100000 + +/* Misc forward declaraions */ +static void eeh_save_bars(struct pci_dev * pdev, struct pci_dn *pdn); /* RTAS tokens */ static int ibm_set_eeh_option; @@ -101,12 +88,19 @@ static int ibm_slot_error_detail; static int eeh_subsystem_enabled; +/* Lock to avoid races due to multiple reports of an error */ +static DEFINE_SPINLOCK(confirm_error_lock); + /* Buffer for reporting slot-error-detail rtas calls */ static unsigned char slot_errbuf[RTAS_ERROR_LOG_MAX]; static DEFINE_SPINLOCK(slot_errbuf_lock); static int eeh_error_buf_size; /* System monitoring statistics */ +static DEFINE_PER_CPU(unsigned long, no_device); +static DEFINE_PER_CPU(unsigned long, no_dn); +static DEFINE_PER_CPU(unsigned long, no_cfg_addr); +static DEFINE_PER_CPU(unsigned long, ignored_check); static DEFINE_PER_CPU(unsigned long, total_mmio_ffs); static DEFINE_PER_CPU(unsigned long, false_positives); static DEFINE_PER_CPU(unsigned long, ignored_failures); @@ -224,9 +218,9 @@ pci_addr_cache_insert(struct pci_dev *dev, unsigned long alo, while (*p) { parent = *p; piar = rb_entry(parent, struct pci_io_addr_range, rb_node); - if (alo < piar->addr_lo) { + if (ahi < piar->addr_lo) { p = &parent->rb_left; - } else if (ahi > piar->addr_hi) { + } else if (alo > piar->addr_hi) { p = &parent->rb_right; } else { if (dev != piar->pcidev || @@ -245,6 +239,11 @@ pci_addr_cache_insert(struct pci_dev *dev, unsigned long alo, piar->pcidev = dev; piar->flags = flags; +#ifdef DEBUG + printk(KERN_DEBUG "PIAR: insert range=[%lx:%lx] dev=%s\n", + alo, ahi, pci_name (dev)); +#endif + rb_link_node(&piar->rb_node, parent, p); rb_insert_color(&piar->rb_node, &pci_io_addr_cache_root.rb_root); @@ -260,18 +259,17 @@ static void __pci_addr_cache_insert_device(struct pci_dev *dev) dn = pci_device_to_OF_node(dev); if (!dn) { - printk(KERN_WARNING "PCI: no pci dn found for dev=%s\n", - pci_name(dev)); + printk(KERN_WARNING "PCI: no pci dn found for dev=%s\n", pci_name(dev)); return; } /* Skip any devices for which EEH is not enabled. */ - pdn = dn->data; + pdn = PCI_DN(dn); if (!(pdn->eeh_mode & EEH_MODE_SUPPORTED) || pdn->eeh_mode & EEH_MODE_NOCHECK) { #ifdef DEBUG - printk(KERN_INFO "PCI: skip building address cache for=%s\n", - pci_name(dev)); + printk(KERN_INFO "PCI: skip building address cache for=%s - %s\n", + pci_name(dev), pdn->node->full_name); #endif return; } @@ -307,7 +305,7 @@ static void __pci_addr_cache_insert_device(struct pci_dev *dev) * we maintain a cache of devices that can be quickly searched. * This routine adds a device to that cache. */ -void pci_addr_cache_insert_device(struct pci_dev *dev) +static void pci_addr_cache_insert_device(struct pci_dev *dev) { unsigned long flags; @@ -350,7 +348,7 @@ restart: * the tree multiple times (once per resource). * But so what; device removal doesn't need to be that fast. */ -void pci_addr_cache_remove_device(struct pci_dev *dev) +static void pci_addr_cache_remove_device(struct pci_dev *dev) { unsigned long flags; @@ -370,8 +368,12 @@ void pci_addr_cache_remove_device(struct pci_dev *dev) */ void __init pci_addr_cache_build(void) { + struct device_node *dn; struct pci_dev *dev = NULL; + if (!eeh_subsystem_enabled) + return; + spin_lock_init(&pci_io_addr_cache_root.piar_lock); while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) { @@ -380,6 +382,10 @@ void __init pci_addr_cache_build(void) continue; } pci_addr_cache_insert_device(dev); + + /* Save the BAR's; firmware doesn't restore these after EEH reset */ + dn = pci_device_to_OF_node(dev); + eeh_save_bars(dev, PCI_DN(dn)); } #ifdef DEBUG @@ -391,22 +397,26 @@ void __init pci_addr_cache_build(void) /* --------------------------------------------------------------- */ /* Above lies the PCI Address Cache. Below lies the EEH event infrastructure */ -/** - * eeh_register_notifier - Register to find out about EEH events. - * @nb: notifier block to callback on events - */ -int eeh_register_notifier(struct notifier_block *nb) +void eeh_slot_error_detail (struct pci_dn *pdn, int severity) { - return notifier_chain_register(&eeh_notifier_chain, nb); -} + unsigned long flags; + int rc; -/** - * eeh_unregister_notifier - Unregister to an EEH event notifier. - * @nb: notifier block to callback on events - */ -int eeh_unregister_notifier(struct notifier_block *nb) -{ - return notifier_chain_unregister(&eeh_notifier_chain, nb); + /* Log the error with the rtas logger */ + spin_lock_irqsave(&slot_errbuf_lock, flags); + memset(slot_errbuf, 0, eeh_error_buf_size); + + rc = rtas_call(ibm_slot_error_detail, + 8, 1, NULL, pdn->eeh_config_addr, + BUID_HI(pdn->phb->buid), + BUID_LO(pdn->phb->buid), NULL, 0, + virt_to_phys(slot_errbuf), + eeh_error_buf_size, + severity); + + if (rc == 0) + log_error(slot_errbuf, ERR_TYPE_RTAS_LOG, 0); + spin_unlock_irqrestore(&slot_errbuf_lock, flags); } /** @@ -414,16 +424,16 @@ int eeh_unregister_notifier(struct notifier_block *nb) * @dn: device node to read * @rets: array to return results in */ -static int read_slot_reset_state(struct device_node *dn, int rets[]) +static int read_slot_reset_state(struct pci_dn *pdn, int rets[]) { int token, outputs; - struct pci_dn *pdn = dn->data; if (ibm_read_slot_reset_state2 != RTAS_UNKNOWN_SERVICE) { token = ibm_read_slot_reset_state2; outputs = 4; } else { token = ibm_read_slot_reset_state; + rets[2] = 0; /* fake PE Unavailable info */ outputs = 3; } @@ -432,87 +442,84 @@ static int read_slot_reset_state(struct device_node *dn, int rets[]) } /** - * eeh_panic - call panic() for an eeh event that cannot be handled. - * The philosophy of this routine is that it is better to panic and - * halt the OS than it is to risk possible data corruption by - * oblivious device drivers that don't know better. - * - * @dev pci device that had an eeh event - * @reset_state current reset state of the device slot + * eeh_token_to_phys - convert EEH address token to phys address + * @token i/o token, should be address in the form 0xA.... */ -static void eeh_panic(struct pci_dev *dev, int reset_state) +static inline unsigned long eeh_token_to_phys(unsigned long token) { - /* - * XXX We should create a separate sysctl for this. - * - * Since the panic_on_oops sysctl is used to halt the system - * in light of potential corruption, we can use it here. - */ - if (panic_on_oops) - panic("EEH: MMIO failure (%d) on device:%s\n", reset_state, - pci_name(dev)); - else { - __get_cpu_var(ignored_failures)++; - printk(KERN_INFO "EEH: Ignored MMIO failure (%d) on device:%s\n", - reset_state, pci_name(dev)); - } + pte_t *ptep; + unsigned long pa; + + ptep = find_linux_pte(init_mm.pgd, token); + if (!ptep) + return token; + pa = pte_pfn(*ptep) << PAGE_SHIFT; + + return pa | (token & (PAGE_SIZE-1)); } -/** - * eeh_event_handler - dispatch EEH events. The detection of a frozen - * slot can occur inside an interrupt, where it can be hard to do - * anything about it. The goal of this routine is to pull these - * detection events out of the context of the interrupt handler, and - * re-dispatch them for processing at a later time in a normal context. - * - * @dummy - unused +/** + * Return the "partitionable endpoint" (pe) under which this device lies */ -static void eeh_event_handler(void *dummy) +static struct device_node * find_device_pe(struct device_node *dn) { - unsigned long flags; - struct eeh_event *event; - - while (1) { - spin_lock_irqsave(&eeh_eventlist_lock, flags); - event = NULL; - if (!list_empty(&eeh_eventlist)) { - event = list_entry(eeh_eventlist.next, struct eeh_event, list); - list_del(&event->list); - } - spin_unlock_irqrestore(&eeh_eventlist_lock, flags); - if (event == NULL) - break; - - printk(KERN_INFO "EEH: MMIO failure (%d), notifiying device " - "%s\n", event->reset_state, - pci_name(event->dev)); + while ((dn->parent) && PCI_DN(dn->parent) && + (PCI_DN(dn->parent)->eeh_mode & EEH_MODE_SUPPORTED)) { + dn = dn->parent; + } + return dn; +} - atomic_set(&eeh_fail_count, 0); - notifier_call_chain (&eeh_notifier_chain, - EEH_NOTIFY_FREEZE, event); +/** Mark all devices that are peers of this device as failed. + * Mark the device driver too, so that it can see the failure + * immediately; this is critical, since some drivers poll + * status registers in interrupts ... If a driver is polling, + * and the slot is frozen, then the driver can deadlock in + * an interrupt context, which is bad. + */ - __get_cpu_var(slot_resets)++; +static void __eeh_mark_slot (struct device_node *dn, int mode_flag) +{ + while (dn) { + if (PCI_DN(dn)) { + PCI_DN(dn)->eeh_mode |= mode_flag; - pci_dev_put(event->dev); - kfree(event); + if (dn->child) + __eeh_mark_slot (dn->child, mode_flag); + } + dn = dn->sibling; } } -/** - * eeh_token_to_phys - convert EEH address token to phys address - * @token i/o token, should be address in the form 0xE.... - */ -static inline unsigned long eeh_token_to_phys(unsigned long token) +void eeh_mark_slot (struct device_node *dn, int mode_flag) { - pte_t *ptep; - unsigned long pa; + dn = find_device_pe (dn); + PCI_DN(dn)->eeh_mode |= mode_flag; + __eeh_mark_slot (dn->child, mode_flag); +} - ptep = find_linux_pte(init_mm.pgd, token); - if (!ptep) - return token; - pa = pte_pfn(*ptep) << PAGE_SHIFT; +static void __eeh_clear_slot (struct device_node *dn, int mode_flag) +{ + while (dn) { + if (PCI_DN(dn)) { + PCI_DN(dn)->eeh_mode &= ~mode_flag; + PCI_DN(dn)->eeh_check_count = 0; + if (dn->child) + __eeh_clear_slot (dn->child, mode_flag); + } + dn = dn->sibling; + } +} - return pa | (token & (PAGE_SIZE-1)); +void eeh_clear_slot (struct device_node *dn, int mode_flag) +{ + unsigned long flags; + spin_lock_irqsave(&confirm_error_lock, flags); + dn = find_device_pe (dn); + PCI_DN(dn)->eeh_mode &= ~mode_flag; + PCI_DN(dn)->eeh_check_count = 0; + __eeh_clear_slot (dn->child, mode_flag); + spin_unlock_irqrestore(&confirm_error_lock, flags); } /** @@ -526,7 +533,7 @@ static inline unsigned long eeh_token_to_phys(unsigned long token) * will query firmware for the EEH status. * * Returns 0 if there has not been an EEH error; otherwise returns - * a non-zero value and queues up a solt isolation event notification. + * a non-zero value and queues up a slot isolation event notification. * * It is safe to call this routine in an interrupt context. */ @@ -535,42 +542,59 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev) int ret; int rets[3]; unsigned long flags; - int rc, reset_state; - struct eeh_event *event; struct pci_dn *pdn; + int rc = 0; __get_cpu_var(total_mmio_ffs)++; if (!eeh_subsystem_enabled) return 0; - if (!dn) + if (!dn) { + __get_cpu_var(no_dn)++; return 0; - pdn = dn->data; + } + pdn = PCI_DN(dn); /* Access to IO BARs might get this far and still not want checking. */ - if (!pdn->eeh_capable || !(pdn->eeh_mode & EEH_MODE_SUPPORTED) || + if (!(pdn->eeh_mode & EEH_MODE_SUPPORTED) || pdn->eeh_mode & EEH_MODE_NOCHECK) { + __get_cpu_var(ignored_check)++; +#ifdef DEBUG + printk ("EEH:ignored check (%x) for %s %s\n", + pdn->eeh_mode, pci_name (dev), dn->full_name); +#endif return 0; } if (!pdn->eeh_config_addr) { + __get_cpu_var(no_cfg_addr)++; return 0; } - /* - * If we already have a pending isolation event for this - * slot, we know it's bad already, we don't need to check... + /* If we already have a pending isolation event for this + * slot, we know it's bad already, we don't need to check. + * Do this checking under a lock; as multiple PCI devices + * in one slot might report errors simultaneously, and we + * only want one error recovery routine running. */ + spin_lock_irqsave(&confirm_error_lock, flags); + rc = 1; if (pdn->eeh_mode & EEH_MODE_ISOLATED) { - atomic_inc(&eeh_fail_count); - if (atomic_read(&eeh_fail_count) >= EEH_MAX_FAILS) { + pdn->eeh_check_count ++; + if (pdn->eeh_check_count >= EEH_MAX_FAILS) { + printk (KERN_ERR "EEH: Device driver ignored %d bad reads, panicing\n", + pdn->eeh_check_count); + dump_stack(); + /* re-read the slot reset state */ - if (read_slot_reset_state(dn, rets) != 0) + if (read_slot_reset_state(pdn, rets) != 0) rets[0] = -1; /* reset state unknown */ - eeh_panic(dev, rets[0]); + + /* If we are here, then we hit an infinite loop. Stop. */ + panic("EEH: MMIO halt (%d) on device:%s\n", rets[0], pci_name(dev)); } - return 0; + goto dn_unlock; } /* @@ -580,66 +604,69 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev) * function zero of a multi-function device. * In any case they must share a common PHB. */ - ret = read_slot_reset_state(dn, rets); - if (!(ret == 0 && rets[1] == 1 && (rets[0] == 2 || rets[0] == 4))) { + ret = read_slot_reset_state(pdn, rets); + + /* If the call to firmware failed, punt */ + if (ret != 0) { + printk(KERN_WARNING "EEH: read_slot_reset_state() failed; rc=%d dn=%s\n", + ret, dn->full_name); __get_cpu_var(false_positives)++; - return 0; + rc = 0; + goto dn_unlock; } - /* prevent repeated reports of this failure */ - pdn->eeh_mode |= EEH_MODE_ISOLATED; - - reset_state = rets[0]; - - spin_lock_irqsave(&slot_errbuf_lock, flags); - memset(slot_errbuf, 0, eeh_error_buf_size); - - rc = rtas_call(ibm_slot_error_detail, - 8, 1, NULL, pdn->eeh_config_addr, - BUID_HI(pdn->phb->buid), - BUID_LO(pdn->phb->buid), NULL, 0, - virt_to_phys(slot_errbuf), - eeh_error_buf_size, - 1 /* Temporary Error */); - - if (rc == 0) - log_error(slot_errbuf, ERR_TYPE_RTAS_LOG, 0); - spin_unlock_irqrestore(&slot_errbuf_lock, flags); + /* If EEH is not supported on this device, punt. */ + if (rets[1] != 1) { + printk(KERN_WARNING "EEH: event on unsupported device, rc=%d dn=%s\n", + ret, dn->full_name); + __get_cpu_var(false_positives)++; + rc = 0; + goto dn_unlock; + } - printk(KERN_INFO "EEH: MMIO failure (%d) on device: %s %s\n", - rets[0], dn->name, dn->full_name); - event = kmalloc(sizeof(*event), GFP_ATOMIC); - if (event == NULL) { - eeh_panic(dev, reset_state); - return 1; - } + /* If not the kind of error we know about, punt. */ + if (rets[0] != 2 && rets[0] != 4 && rets[0] != 5) { + __get_cpu_var(false_positives)++; + rc = 0; + goto dn_unlock; + } - event->dev = dev; - event->dn = dn; - event->reset_state = reset_state; + /* Note that config-io to empty slots may fail; + * we recognize empty because they don't have children. */ + if ((rets[0] == 5) && (dn->child == NULL)) { + __get_cpu_var(false_positives)++; + rc = 0; + goto dn_unlock; + } - /* We may or may not be called in an interrupt context */ - spin_lock_irqsave(&eeh_eventlist_lock, flags); - list_add(&event->list, &eeh_eventlist); - spin_unlock_irqrestore(&eeh_eventlist_lock, flags); + __get_cpu_var(slot_resets)++; + + /* Avoid repeated reports of this failure, including problems + * with other functions on this device, and functions under + * bridges. */ + eeh_mark_slot (dn, EEH_MODE_ISOLATED); + spin_unlock_irqrestore(&confirm_error_lock, flags); + eeh_send_failure_event (dn, dev, rets[0], rets[2]); + /* Most EEH events are due to device driver bugs. Having * a stack trace will help the device-driver authors figure * out what happened. So print that out. */ - dump_stack(); - schedule_work(&eeh_event_wq); + if (rets[0] != 5) dump_stack(); + return 1; - return 0; +dn_unlock: + spin_unlock_irqrestore(&confirm_error_lock, flags); + return rc; } -EXPORT_SYMBOL(eeh_dn_check_failure); +EXPORT_SYMBOL_GPL(eeh_dn_check_failure); /** * eeh_check_failure - check if all 1's data is due to EEH slot freeze * @token i/o token, should be address in the form 0xA.... * @val value, should be all 1's (XXX why do we need this arg??) * - * Check for an eeh failure at the given token address. * Check for an EEH failure at the given token address. Call this * routine if the result of a read was all 0xff's and you want to * find out if this is due to an EEH slot freeze event. This routine @@ -656,8 +683,10 @@ unsigned long eeh_check_failure(const volatile void __iomem *token, unsigned lon /* Finding the phys addr + pci device; this is pretty quick. */ addr = eeh_token_to_phys((unsigned long __force) token); dev = pci_get_device_by_addr(addr); - if (!dev) + if (!dev) { + __get_cpu_var(no_device)++; return val; + } dn = pci_device_to_OF_node(dev); eeh_dn_check_failure (dn, dev); @@ -668,6 +697,217 @@ unsigned long eeh_check_failure(const volatile void __iomem *token, unsigned lon EXPORT_SYMBOL(eeh_check_failure); +/* ------------------------------------------------------------- */ +/* The code below deals with error recovery */ + +/** Return negative value if a permanent error, else return + * a number of milliseconds to wait until the PCI slot is + * ready to be used. + */ +static int +eeh_slot_availability(struct pci_dn *pdn) +{ + int rc; + int rets[3]; + + rc = read_slot_reset_state(pdn, rets); + + if (rc) return rc; + + if (rets[1] == 0) return -1; /* EEH is not supported */ + if (rets[0] == 0) return 0; /* Oll Korrect */ + if (rets[0] == 5) { + if (rets[2] == 0) return -1; /* permanently unavailable */ + return rets[2]; /* number of millisecs to wait */ + } + return -1; +} + +/** rtas_pci_slot_reset raises/lowers the pci #RST line + * state: 1/0 to raise/lower the #RST + * + * Clear the EEH-frozen condition on a slot. This routine + * asserts the PCI #RST line if the 'state' argument is '1', + * and drops the #RST line if 'state is '0'. This routine is + * safe to call in an interrupt context. + * + */ + +static void +rtas_pci_slot_reset(struct pci_dn *pdn, int state) +{ + int rc; + + BUG_ON (pdn==NULL); + + if (!pdn->phb) { + printk (KERN_WARNING "EEH: in slot reset, device node %s has no phb\n", + pdn->node->full_name); + return; + } + + rc = rtas_call(ibm_set_slot_reset,4,1, NULL, + pdn->eeh_config_addr, + BUID_HI(pdn->phb->buid), + BUID_LO(pdn->phb->buid), + state); + if (rc) { + printk (KERN_WARNING "EEH: Unable to reset the failed slot, (%d) #RST=%d dn=%s\n", + rc, state, pdn->node->full_name); + return; + } +} + +/** rtas_set_slot_reset -- assert the pci #RST line for 1/4 second + * dn -- device node to be reset. + */ + +void +rtas_set_slot_reset(struct pci_dn *pdn) +{ + int i, rc; + + rtas_pci_slot_reset (pdn, 1); + + /* The PCI bus requires that the reset be held high for at least + * a 100 milliseconds. We wait a bit longer 'just in case'. */ + +#define PCI_BUS_RST_HOLD_TIME_MSEC 250 + msleep (PCI_BUS_RST_HOLD_TIME_MSEC); + + /* We might get hit with another EEH freeze as soon as the + * pci slot reset line is dropped. Make sure we don't miss + * these, and clear the flag now. */ + eeh_clear_slot (pdn->node, EEH_MODE_ISOLATED); + + rtas_pci_slot_reset (pdn, 0); + + /* After a PCI slot has been reset, the PCI Express spec requires + * a 1.5 second idle time for the bus to stabilize, before starting + * up traffic. */ +#define PCI_BUS_SETTLE_TIME_MSEC 1800 + msleep (PCI_BUS_SETTLE_TIME_MSEC); + + /* Now double check with the firmware to make sure the device is + * ready to be used; if not, wait for recovery. */ + for (i=0; i<10; i++) { + rc = eeh_slot_availability (pdn); + if (rc <= 0) break; + + msleep (rc+100); + } +} + +/* ------------------------------------------------------- */ +/** Save and restore of PCI BARs + * + * Although firmware will set up BARs during boot, it doesn't + * set up device BAR's after a device reset, although it will, + * if requested, set up bridge configuration. Thus, we need to + * configure the PCI devices ourselves. + */ + +/** + * __restore_bars - Restore the Base Address Registers + * Loads the PCI configuration space base address registers, + * the expansion ROM base address, the latency timer, and etc. + * from the saved values in the device node. + */ +static inline void __restore_bars (struct pci_dn *pdn) +{ + int i; + + if (NULL==pdn->phb) return; + for (i=4; i<10; i++) { + rtas_write_config(pdn, i*4, 4, pdn->config_space[i]); + } + + /* 12 == Expansion ROM Address */ + rtas_write_config(pdn, 12*4, 4, pdn->config_space[12]); + +#define BYTE_SWAP(OFF) (8*((OFF)/4)+3-(OFF)) +#define SAVED_BYTE(OFF) (((u8 *)(pdn->config_space))[BYTE_SWAP(OFF)]) + + rtas_write_config (pdn, PCI_CACHE_LINE_SIZE, 1, + SAVED_BYTE(PCI_CACHE_LINE_SIZE)); + + rtas_write_config (pdn, PCI_LATENCY_TIMER, 1, + SAVED_BYTE(PCI_LATENCY_TIMER)); + + /* max latency, min grant, interrupt pin and line */ + rtas_write_config(pdn, 15*4, 4, pdn->config_space[15]); +} + +/** + * eeh_restore_bars - restore the PCI config space info + * + * This routine performs a recursive walk to the children + * of this device as well. + */ +void eeh_restore_bars(struct pci_dn *pdn) +{ + struct device_node *dn; + if (!pdn) + return; + + if (! pdn->eeh_is_bridge) + __restore_bars (pdn); + + dn = pdn->node->child; + while (dn) { + eeh_restore_bars (PCI_DN(dn)); + dn = dn->sibling; + } +} + +/** + * eeh_save_bars - save device bars + * + * Save the values of the device bars. Unlike the restore + * routine, this routine is *not* recursive. This is because + * PCI devices are added individuallly; but, for the restore, + * an entire slot is reset at a time. + */ +static void eeh_save_bars(struct pci_dev * pdev, struct pci_dn *pdn) +{ + int i; + + if (!pdev || !pdn ) + return; + + for (i = 0; i < 16; i++) + pci_read_config_dword(pdev, i * 4, &pdn->config_space[i]); + + if (pdev->hdr_type == PCI_HEADER_TYPE_BRIDGE) + pdn->eeh_is_bridge = 1; +} + +void +rtas_configure_bridge(struct pci_dn *pdn) +{ + int token = rtas_token ("ibm,configure-bridge"); + int rc; + + if (token == RTAS_UNKNOWN_SERVICE) + return; + rc = rtas_call(token,3,1, NULL, + pdn->eeh_config_addr, + BUID_HI(pdn->phb->buid), + BUID_LO(pdn->phb->buid)); + if (rc) { + printk (KERN_WARNING "EEH: Unable to configure device bridge (%d) for %s\n", + rc, pdn->node->full_name); + } +} + +/* ------------------------------------------------------------- */ +/* The code below deals with enabling EEH for devices during the + * early boot sequence. EEH must be enabled before any PCI probing + * can be done. + */ + +#define EEH_ENABLE 1 + struct eeh_early_enable_info { unsigned int buid_hi; unsigned int buid_lo; @@ -684,9 +924,11 @@ static void *early_enable_eeh(struct device_node *dn, void *data) u32 *device_id = (u32 *)get_property(dn, "device-id", NULL); u32 *regs; int enable; - struct pci_dn *pdn = dn->data; + struct pci_dn *pdn = PCI_DN(dn); pdn->eeh_mode = 0; + pdn->eeh_check_count = 0; + pdn->eeh_freeze_count = 0; if (status && strcmp(status, "ok") != 0) return NULL; /* ignore devices with bad status */ @@ -723,8 +965,9 @@ static void *early_enable_eeh(struct device_node *dn, void *data) /* First register entry is addr (00BBSS00) */ /* Try to enable eeh */ ret = rtas_call(ibm_set_eeh_option, 4, 1, NULL, - regs[0], info->buid_hi, info->buid_lo, - EEH_ENABLE); + regs[0], info->buid_hi, info->buid_lo, + EEH_ENABLE); + if (ret == 0) { eeh_subsystem_enabled = 1; pdn->eeh_mode |= EEH_MODE_SUPPORTED; @@ -736,7 +979,7 @@ static void *early_enable_eeh(struct device_node *dn, void *data) /* This device doesn't support EEH, but it may have an * EEH parent, in which case we mark it as supported. */ - if (dn->parent && dn->parent->data + if (dn->parent && PCI_DN(dn->parent) && (PCI_DN(dn->parent)->eeh_mode & EEH_MODE_SUPPORTED)) { /* Parent supports EEH. */ pdn->eeh_mode |= EEH_MODE_SUPPORTED; @@ -749,7 +992,7 @@ static void *early_enable_eeh(struct device_node *dn, void *data) dn->full_name); } - return NULL; + return NULL; } /* @@ -770,6 +1013,9 @@ void __init eeh_init(void) struct device_node *phb, *np; struct eeh_early_enable_info info; + spin_lock_init(&confirm_error_lock); + spin_lock_init(&slot_errbuf_lock); + np = of_find_node_by_path("/rtas"); if (np == NULL) return; @@ -797,13 +1043,11 @@ void __init eeh_init(void) for (phb = of_find_node_by_name(NULL, "pci"); phb; phb = of_find_node_by_name(phb, "pci")) { unsigned long buid; - struct pci_dn *pci; buid = get_phb_buid(phb); - if (buid == 0 || phb->data == NULL) + if (buid == 0 || PCI_DN(phb) == NULL) continue; - pci = phb->data; info.buid_lo = BUID_LO(buid); info.buid_hi = BUID_HI(buid); traverse_pci_devices(phb, early_enable_eeh, &info); @@ -832,11 +1076,13 @@ void eeh_add_device_early(struct device_node *dn) struct pci_controller *phb; struct eeh_early_enable_info info; - if (!dn || !dn->data) + if (!dn || !PCI_DN(dn)) return; phb = PCI_DN(dn)->phb; if (NULL == phb || 0 == phb->buid) { - printk(KERN_WARNING "EEH: Expected buid but found none\n"); + printk(KERN_WARNING "EEH: Expected buid but found none for %s\n", + dn->full_name); + dump_stack(); return; } @@ -844,7 +1090,7 @@ void eeh_add_device_early(struct device_node *dn) info.buid_lo = BUID_LO(phb->buid); early_enable_eeh(dn, &info); } -EXPORT_SYMBOL(eeh_add_device_early); +EXPORT_SYMBOL_GPL(eeh_add_device_early); /** * eeh_add_device_late - perform EEH initialization for the indicated pci device @@ -855,6 +1101,9 @@ EXPORT_SYMBOL(eeh_add_device_early); */ void eeh_add_device_late(struct pci_dev *dev) { + struct device_node *dn; + struct pci_dn *pdn; + if (!dev || !eeh_subsystem_enabled) return; @@ -862,9 +1111,15 @@ void eeh_add_device_late(struct pci_dev *dev) printk(KERN_DEBUG "EEH: adding device %s\n", pci_name(dev)); #endif + pci_dev_get (dev); + dn = pci_device_to_OF_node(dev); + pdn = PCI_DN(dn); + pdn->pcidev = dev; + pci_addr_cache_insert_device (dev); + eeh_save_bars(dev, pdn); } -EXPORT_SYMBOL(eeh_add_device_late); +EXPORT_SYMBOL_GPL(eeh_add_device_late); /** * eeh_remove_device - undo EEH setup for the indicated pci device @@ -875,6 +1130,7 @@ EXPORT_SYMBOL(eeh_add_device_late); */ void eeh_remove_device(struct pci_dev *dev) { + struct device_node *dn; if (!dev || !eeh_subsystem_enabled) return; @@ -883,20 +1139,29 @@ void eeh_remove_device(struct pci_dev *dev) printk(KERN_DEBUG "EEH: remove device %s\n", pci_name(dev)); #endif pci_addr_cache_remove_device(dev); + + dn = pci_device_to_OF_node(dev); + PCI_DN(dn)->pcidev = NULL; + pci_dev_put (dev); } -EXPORT_SYMBOL(eeh_remove_device); +EXPORT_SYMBOL_GPL(eeh_remove_device); static int proc_eeh_show(struct seq_file *m, void *v) { unsigned int cpu; unsigned long ffs = 0, positives = 0, failures = 0; unsigned long resets = 0; + unsigned long no_dev = 0, no_dn = 0, no_cfg = 0, no_check = 0; for_each_cpu(cpu) { ffs += per_cpu(total_mmio_ffs, cpu); positives += per_cpu(false_positives, cpu); failures += per_cpu(ignored_failures, cpu); resets += per_cpu(slot_resets, cpu); + no_dev += per_cpu(no_device, cpu); + no_dn += per_cpu(no_dn, cpu); + no_cfg += per_cpu(no_cfg_addr, cpu); + no_check += per_cpu(ignored_check, cpu); } if (0 == eeh_subsystem_enabled) { @@ -904,13 +1169,17 @@ static int proc_eeh_show(struct seq_file *m, void *v) seq_printf(m, "eeh_total_mmio_ffs=%ld\n", ffs); } else { seq_printf(m, "EEH Subsystem is enabled\n"); - seq_printf(m, "eeh_total_mmio_ffs=%ld\n" - "eeh_false_positives=%ld\n" - "eeh_ignored_failures=%ld\n" - "eeh_slot_resets=%ld\n" - "eeh_fail_count=%d\n", - ffs, positives, failures, resets, - eeh_fail_count.counter); + seq_printf(m, + "no device=%ld\n" + "no device node=%ld\n" + "no config address=%ld\n" + "check not wanted=%ld\n" + "eeh_total_mmio_ffs=%ld\n" + "eeh_false_positives=%ld\n" + "eeh_ignored_failures=%ld\n" + "eeh_slot_resets=%ld\n", + no_dev, no_dn, no_cfg, no_check, + ffs, positives, failures, resets); } return 0; @@ -932,7 +1201,7 @@ static int __init eeh_init_proc(void) { struct proc_dir_entry *e; - if (systemcfg->platform & PLATFORM_PSERIES) { + if (platform_is_pseries()) { e = create_proc_entry("ppc64/eeh", 0, NULL); if (e) e->proc_fops = &proc_eeh_operations; diff --git a/arch/powerpc/platforms/pseries/eeh_event.c b/arch/powerpc/platforms/pseries/eeh_event.c new file mode 100644 index 00000000000..92497333c2b --- /dev/null +++ b/arch/powerpc/platforms/pseries/eeh_event.c @@ -0,0 +1,155 @@ +/* + * eeh_event.c + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * Copyright (c) 2005 Linas Vepstas <linas@linas.org> + */ + +#include <linux/list.h> +#include <linux/pci.h> +#include <asm/eeh_event.h> + +/** Overview: + * EEH error states may be detected within exception handlers; + * however, the recovery processing needs to occur asynchronously + * in a normal kernel context and not an interrupt context. + * This pair of routines creates an event and queues it onto a + * work-queue, where a worker thread can drive recovery. + */ + +/* EEH event workqueue setup. */ +static spinlock_t eeh_eventlist_lock = SPIN_LOCK_UNLOCKED; +LIST_HEAD(eeh_eventlist); +static void eeh_thread_launcher(void *); +DECLARE_WORK(eeh_event_wq, eeh_thread_launcher, NULL); + +/** + * eeh_panic - call panic() for an eeh event that cannot be handled. + * The philosophy of this routine is that it is better to panic and + * halt the OS than it is to risk possible data corruption by + * oblivious device drivers that don't know better. + * + * @dev pci device that had an eeh event + * @reset_state current reset state of the device slot + */ +static void eeh_panic(struct pci_dev *dev, int reset_state) +{ + /* + * Since the panic_on_oops sysctl is used to halt the system + * in light of potential corruption, we can use it here. + */ + if (panic_on_oops) { + panic("EEH: MMIO failure (%d) on device:%s\n", reset_state, + pci_name(dev)); + } + else { + printk(KERN_INFO "EEH: Ignored MMIO failure (%d) on device:%s\n", + reset_state, pci_name(dev)); + } +} + +/** + * eeh_event_handler - dispatch EEH events. The detection of a frozen + * slot can occur inside an interrupt, where it can be hard to do + * anything about it. The goal of this routine is to pull these + * detection events out of the context of the interrupt handler, and + * re-dispatch them for processing at a later time in a normal context. + * + * @dummy - unused + */ +static int eeh_event_handler(void * dummy) +{ + unsigned long flags; + struct eeh_event *event; + + daemonize ("eehd"); + + while (1) { + set_current_state(TASK_INTERRUPTIBLE); + + spin_lock_irqsave(&eeh_eventlist_lock, flags); + event = NULL; + if (!list_empty(&eeh_eventlist)) { + event = list_entry(eeh_eventlist.next, struct eeh_event, list); + list_del(&event->list); + } + spin_unlock_irqrestore(&eeh_eventlist_lock, flags); + if (event == NULL) + break; + + printk(KERN_INFO "EEH: Detected PCI bus error on device %s\n", + pci_name(event->dev)); + + eeh_panic (event->dev, event->state); + + kfree(event); + } + + return 0; +} + +/** + * eeh_thread_launcher + * + * @dummy - unused + */ +static void eeh_thread_launcher(void *dummy) +{ + if (kernel_thread(eeh_event_handler, NULL, CLONE_KERNEL) < 0) + printk(KERN_ERR "Failed to start EEH daemon\n"); +} + +/** + * eeh_send_failure_event - generate a PCI error event + * @dev pci device + * + * This routine can be called within an interrupt context; + * the actual event will be delivered in a normal context + * (from a workqueue). + */ +int eeh_send_failure_event (struct device_node *dn, + struct pci_dev *dev, + int state, + int time_unavail) +{ + unsigned long flags; + struct eeh_event *event; + + event = kmalloc(sizeof(*event), GFP_ATOMIC); + if (event == NULL) { + printk (KERN_ERR "EEH: out of memory, event not handled\n"); + return 1; + } + + if (dev) + pci_dev_get(dev); + + event->dn = dn; + event->dev = dev; + event->state = state; + event->time_unavail = time_unavail; + + /* We may or may not be called in an interrupt context */ + spin_lock_irqsave(&eeh_eventlist_lock, flags); + list_add(&event->list, &eeh_eventlist); + spin_unlock_irqrestore(&eeh_eventlist_lock, flags); + + schedule_work(&eeh_event_wq); + + return 0; +} + +/********************** END OF FILE ******************************/ diff --git a/arch/powerpc/platforms/pseries/iommu.c b/arch/powerpc/platforms/pseries/iommu.c index fcc50bfd43f..97ba5214417 100644 --- a/arch/powerpc/platforms/pseries/iommu.c +++ b/arch/powerpc/platforms/pseries/iommu.c @@ -42,7 +42,6 @@ #include <asm/machdep.h> #include <asm/abs_addr.h> #include <asm/pSeries_reconfig.h> -#include <asm/systemcfg.h> #include <asm/firmware.h> #include <asm/tce.h> #include <asm/ppc-pci.h> @@ -582,7 +581,7 @@ void iommu_init_early_pSeries(void) return; } - if (systemcfg->platform & PLATFORM_LPAR) { + if (platform_is_lpar()) { if (firmware_has_feature(FW_FEATURE_MULTITCE)) { ppc_md.tce_build = tce_buildmulti_pSeriesLP; ppc_md.tce_free = tce_freemulti_pSeriesLP; diff --git a/arch/powerpc/platforms/pseries/pci.c b/arch/powerpc/platforms/pseries/pci.c index c198656a3bb..999a9620b5c 100644 --- a/arch/powerpc/platforms/pseries/pci.c +++ b/arch/powerpc/platforms/pseries/pci.c @@ -107,7 +107,6 @@ static void __init pSeries_request_regions(void) void __init pSeries_final_fixup(void) { - phbs_remap_io(); pSeries_request_regions(); pci_addr_cache_build(); @@ -123,7 +122,7 @@ static void fixup_winbond_82c105(struct pci_dev* dev) int i; unsigned int reg; - if (!(systemcfg->platform & PLATFORM_PSERIES)) + if (!platform_is_pseries()) return; printk("Using INTC for W82c105 IDE controller.\n"); diff --git a/arch/powerpc/platforms/pseries/reconfig.c b/arch/powerpc/platforms/pseries/reconfig.c index d7d40033945..d8864164dbe 100644 --- a/arch/powerpc/platforms/pseries/reconfig.c +++ b/arch/powerpc/platforms/pseries/reconfig.c @@ -408,7 +408,7 @@ static int proc_ppc64_create_ofdt(void) { struct proc_dir_entry *ent; - if (!(systemcfg->platform & PLATFORM_PSERIES)) + if (!platform_is_pseries()) return 0; ent = create_proc_entry("ppc64/ofdt", S_IWUSR, NULL); diff --git a/arch/powerpc/platforms/pseries/rtasd.c b/arch/powerpc/platforms/pseries/rtasd.c index e26b0420b6d..00cf331a1dc 100644 --- a/arch/powerpc/platforms/pseries/rtasd.c +++ b/arch/powerpc/platforms/pseries/rtasd.c @@ -482,10 +482,12 @@ static int __init rtas_init(void) { struct proc_dir_entry *entry; - /* No RTAS, only warn if we are on a pSeries box */ + if (!platform_is_pseries()) + return 0; + + /* No RTAS */ if (rtas_token("event-scan") == RTAS_UNKNOWN_SERVICE) { - if (systemcfg->platform & PLATFORM_PSERIES) - printk(KERN_INFO "rtasd: no event-scan on system\n"); + printk(KERN_INFO "rtasd: no event-scan on system\n"); return 1; } diff --git a/arch/ppc64/kernel/scanlog.c b/arch/powerpc/platforms/pseries/scanlog.c index 2edc947f7c4..2edc947f7c4 100644 --- a/arch/ppc64/kernel/scanlog.c +++ b/arch/powerpc/platforms/pseries/scanlog.c diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c index a093a0d4dd6..e94247c28d4 100644 --- a/arch/powerpc/platforms/pseries/setup.c +++ b/arch/powerpc/platforms/pseries/setup.c @@ -249,7 +249,7 @@ static void __init pSeries_setup_arch(void) ppc_md.idle_loop = default_idle; } - if (systemcfg->platform & PLATFORM_LPAR) + if (platform_is_lpar()) ppc_md.enable_pmcs = pseries_lpar_enable_pmcs; else ppc_md.enable_pmcs = power4_enable_pmcs; @@ -378,7 +378,7 @@ static void __init pSeries_init_early(void) fw_feature_init(); - if (systemcfg->platform & PLATFORM_LPAR) + if (platform_is_lpar()) hpte_init_lpar(); else { hpte_init_native(); @@ -388,7 +388,7 @@ static void __init pSeries_init_early(void) generic_find_legacy_serial_ports(&physport, &default_speed); - if (systemcfg->platform & PLATFORM_LPAR) + if (platform_is_lpar()) find_udbg_vterm(); else if (physport) { /* Map the uart for udbg. */ @@ -592,7 +592,7 @@ static void pseries_shared_idle(void) static int pSeries_pci_probe_mode(struct pci_bus *bus) { - if (systemcfg->platform & PLATFORM_LPAR) + if (platform_is_lpar()) return PCI_PROBE_DEVTREE; return PCI_PROBE_NORMAL; } diff --git a/arch/powerpc/platforms/pseries/smp.c b/arch/powerpc/platforms/pseries/smp.c index 7a243e8ccd7..3ba794ca328 100644 --- a/arch/powerpc/platforms/pseries/smp.c +++ b/arch/powerpc/platforms/pseries/smp.c @@ -46,6 +46,7 @@ #include <asm/rtas.h> #include <asm/pSeries_reconfig.h> #include <asm/mpic.h> +#include <asm/systemcfg.h> #include "plpar_wrappers.h" @@ -96,7 +97,7 @@ int pSeries_cpu_disable(void) int cpu = smp_processor_id(); cpu_clear(cpu, cpu_online_map); - systemcfg->processorCount--; + _systemcfg->processorCount--; /*fix boot_cpuid here*/ if (cpu == boot_cpuid) @@ -441,7 +442,7 @@ void __init smp_init_pSeries(void) smp_ops->cpu_die = pSeries_cpu_die; /* Processors can be added/removed only on LPAR */ - if (systemcfg->platform == PLATFORM_PSERIES_LPAR) + if (platform_is_lpar()) pSeries_reconfig_notifier_register(&pSeries_smp_nb); #endif diff --git a/arch/powerpc/platforms/pseries/xics.c b/arch/powerpc/platforms/pseries/xics.c index c72c86f05cb..72ac18067ec 100644 --- a/arch/powerpc/platforms/pseries/xics.c +++ b/arch/powerpc/platforms/pseries/xics.c @@ -545,7 +545,9 @@ nextnode: of_node_put(np); } - if (systemcfg->platform == PLATFORM_PSERIES) { + if (platform_is_lpar()) + ops = &pSeriesLP_ops; + else { #ifdef CONFIG_SMP for_each_cpu(i) { int hard_id; @@ -561,12 +563,11 @@ nextnode: #else xics_per_cpu[0] = ioremap(intr_base, intr_size); #endif /* CONFIG_SMP */ - } else if (systemcfg->platform == PLATFORM_PSERIES_LPAR) { - ops = &pSeriesLP_ops; } xics_8259_pic.enable = i8259_pic.enable; xics_8259_pic.disable = i8259_pic.disable; + xics_8259_pic.end = i8259_pic.end; for (i = 0; i < 16; ++i) get_irq_desc(i)->handler = &xics_8259_pic; for (; i < NR_IRQS; ++i) diff --git a/arch/powerpc/sysdev/u3_iommu.c b/arch/powerpc/sysdev/u3_iommu.c index 543d6590981..f32baf7f469 100644 --- a/arch/powerpc/sysdev/u3_iommu.c +++ b/arch/powerpc/sysdev/u3_iommu.c @@ -226,7 +226,7 @@ static void iommu_table_u3_setup(void) iommu_table_u3.it_busno = 0; iommu_table_u3.it_offset = 0; /* it_size is in number of entries */ - iommu_table_u3.it_size = dart_tablesize / sizeof(u32); + iommu_table_u3.it_size = (dart_tablesize / sizeof(u32)) >> DART_PAGE_FACTOR; /* Initialize the common IOMMU code */ iommu_table_u3.it_base = (unsigned long)dart_vbase; diff --git a/arch/powerpc/xmon/Makefile b/arch/powerpc/xmon/Makefile index 79a784f0e7a..b20312e5ed2 100644 --- a/arch/powerpc/xmon/Makefile +++ b/arch/powerpc/xmon/Makefile @@ -8,4 +8,4 @@ obj-$(CONFIG_8xx) += start_8xx.o obj-$(CONFIG_6xx) += start_32.o obj-$(CONFIG_4xx) += start_32.o obj-$(CONFIG_PPC64) += start_64.o -obj-y += xmon.o ppc-dis.o ppc-opc.o subr_prf.o setjmp.o +obj-y += xmon.o ppc-dis.o ppc-opc.o setjmp.o nonstdio.o diff --git a/arch/powerpc/xmon/nonstdio.c b/arch/powerpc/xmon/nonstdio.c new file mode 100644 index 00000000000..78765833f4c --- /dev/null +++ b/arch/powerpc/xmon/nonstdio.c @@ -0,0 +1,134 @@ +/* + * Copyright (C) 1996-2005 Paul Mackerras. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ +#include <linux/string.h> +#include <asm/time.h> +#include "nonstdio.h" + +int xmon_putchar(int c) +{ + char ch = c; + + if (c == '\n') + xmon_putchar('\r'); + return xmon_write(&ch, 1) == 1? c: -1; +} + +static char line[256]; +static char *lineptr; +static int lineleft; + +int xmon_expect(const char *str, unsigned long timeout) +{ + int c; + unsigned long t0; + + /* assume 25MHz default timebase if tb_ticks_per_sec not set yet */ + timeout *= tb_ticks_per_sec? tb_ticks_per_sec: 25000000; + t0 = get_tbl(); + do { + lineptr = line; + for (;;) { + c = xmon_read_poll(); + if (c == -1) { + if (get_tbl() - t0 > timeout) + return 0; + continue; + } + if (c == '\n') + break; + if (c != '\r' && lineptr < &line[sizeof(line) - 1]) + *lineptr++ = c; + } + *lineptr = 0; + } while (strstr(line, str) == NULL); + return 1; +} + +int xmon_getchar(void) +{ + int c; + + if (lineleft == 0) { + lineptr = line; + for (;;) { + c = xmon_readchar(); + if (c == -1 || c == 4) + break; + if (c == '\r' || c == '\n') { + *lineptr++ = '\n'; + xmon_putchar('\n'); + break; + } + switch (c) { + case 0177: + case '\b': + if (lineptr > line) { + xmon_putchar('\b'); + xmon_putchar(' '); + xmon_putchar('\b'); + --lineptr; + } + break; + case 'U' & 0x1F: + while (lineptr > line) { + xmon_putchar('\b'); + xmon_putchar(' '); + xmon_putchar('\b'); + --lineptr; + } + break; + default: + if (lineptr >= &line[sizeof(line) - 1]) + xmon_putchar('\a'); + else { + xmon_putchar(c); + *lineptr++ = c; + } + } + } + lineleft = lineptr - line; + lineptr = line; + } + if (lineleft == 0) + return -1; + --lineleft; + return *lineptr++; +} + +char *xmon_gets(char *str, int nb) +{ + char *p; + int c; + + for (p = str; p < str + nb - 1; ) { + c = xmon_getchar(); + if (c == -1) { + if (p == str) + return NULL; + break; + } + *p++ = c; + if (c == '\n') + break; + } + *p = 0; + return str; +} + +void xmon_printf(const char *format, ...) +{ + va_list args; + int n; + static char xmon_outbuf[1024]; + + va_start(args, format); + n = vsnprintf(xmon_outbuf, sizeof(xmon_outbuf), format, args); + va_end(args); + xmon_write(xmon_outbuf, n); +} diff --git a/arch/powerpc/xmon/nonstdio.h b/arch/powerpc/xmon/nonstdio.h index 84211a21c6f..47cebbd2b1b 100644 --- a/arch/powerpc/xmon/nonstdio.h +++ b/arch/powerpc/xmon/nonstdio.h @@ -1,22 +1,14 @@ -typedef int FILE; -extern FILE *xmon_stdin, *xmon_stdout; #define EOF (-1) -#define stdin xmon_stdin -#define stdout xmon_stdout + #define printf xmon_printf -#define fprintf xmon_fprintf -#define fputs xmon_fputs -#define fgets xmon_fgets #define putchar xmon_putchar -#define getchar xmon_getchar -#define putc xmon_putc -#define getc xmon_getc -#define fopen(n, m) NULL -#define fflush(f) do {} while (0) -#define fclose(f) do {} while (0) -extern char *fgets(char *, int, void *); -extern void xmon_printf(const char *, ...); -extern void xmon_fprintf(void *, const char *, ...); -extern void xmon_sprintf(char *, const char *, ...); -#define perror(s) printf("%s: no files!\n", (s)) +extern int xmon_putchar(int c); +extern int xmon_getchar(void); +extern char *xmon_gets(char *, int); +extern void xmon_printf(const char *, ...); +extern void xmon_map_scc(void); +extern int xmon_expect(const char *str, unsigned long timeout); +extern int xmon_write(void *ptr, int nb); +extern int xmon_readchar(void); +extern int xmon_read_poll(void); diff --git a/arch/powerpc/xmon/setjmp.S b/arch/powerpc/xmon/setjmp.S index f8e40dfd2bf..96a91f10e2e 100644 --- a/arch/powerpc/xmon/setjmp.S +++ b/arch/powerpc/xmon/setjmp.S @@ -14,61 +14,61 @@ _GLOBAL(xmon_setjmp) mflr r0 - STL r0,0(r3) - STL r1,SZL(r3) - STL r2,2*SZL(r3) + PPC_STL r0,0(r3) + PPC_STL r1,SZL(r3) + PPC_STL r2,2*SZL(r3) mfcr r0 - STL r0,3*SZL(r3) - STL r13,4*SZL(r3) - STL r14,5*SZL(r3) - STL r15,6*SZL(r3) - STL r16,7*SZL(r3) - STL r17,8*SZL(r3) - STL r18,9*SZL(r3) - STL r19,10*SZL(r3) - STL r20,11*SZL(r3) - STL r21,12*SZL(r3) - STL r22,13*SZL(r3) - STL r23,14*SZL(r3) - STL r24,15*SZL(r3) - STL r25,16*SZL(r3) - STL r26,17*SZL(r3) - STL r27,18*SZL(r3) - STL r28,19*SZL(r3) - STL r29,20*SZL(r3) - STL r30,21*SZL(r3) - STL r31,22*SZL(r3) + PPC_STL r0,3*SZL(r3) + PPC_STL r13,4*SZL(r3) + PPC_STL r14,5*SZL(r3) + PPC_STL r15,6*SZL(r3) + PPC_STL r16,7*SZL(r3) + PPC_STL r17,8*SZL(r3) + PPC_STL r18,9*SZL(r3) + PPC_STL r19,10*SZL(r3) + PPC_STL r20,11*SZL(r3) + PPC_STL r21,12*SZL(r3) + PPC_STL r22,13*SZL(r3) + PPC_STL r23,14*SZL(r3) + PPC_STL r24,15*SZL(r3) + PPC_STL r25,16*SZL(r3) + PPC_STL r26,17*SZL(r3) + PPC_STL r27,18*SZL(r3) + PPC_STL r28,19*SZL(r3) + PPC_STL r29,20*SZL(r3) + PPC_STL r30,21*SZL(r3) + PPC_STL r31,22*SZL(r3) li r3,0 blr _GLOBAL(xmon_longjmp) - CMPI r4,0 + PPC_LCMPI r4,0 bne 1f li r4,1 -1: LDL r13,4*SZL(r3) - LDL r14,5*SZL(r3) - LDL r15,6*SZL(r3) - LDL r16,7*SZL(r3) - LDL r17,8*SZL(r3) - LDL r18,9*SZL(r3) - LDL r19,10*SZL(r3) - LDL r20,11*SZL(r3) - LDL r21,12*SZL(r3) - LDL r22,13*SZL(r3) - LDL r23,14*SZL(r3) - LDL r24,15*SZL(r3) - LDL r25,16*SZL(r3) - LDL r26,17*SZL(r3) - LDL r27,18*SZL(r3) - LDL r28,19*SZL(r3) - LDL r29,20*SZL(r3) - LDL r30,21*SZL(r3) - LDL r31,22*SZL(r3) - LDL r0,3*SZL(r3) +1: PPC_LL r13,4*SZL(r3) + PPC_LL r14,5*SZL(r3) + PPC_LL r15,6*SZL(r3) + PPC_LL r16,7*SZL(r3) + PPC_LL r17,8*SZL(r3) + PPC_LL r18,9*SZL(r3) + PPC_LL r19,10*SZL(r3) + PPC_LL r20,11*SZL(r3) + PPC_LL r21,12*SZL(r3) + PPC_LL r22,13*SZL(r3) + PPC_LL r23,14*SZL(r3) + PPC_LL r24,15*SZL(r3) + PPC_LL r25,16*SZL(r3) + PPC_LL r26,17*SZL(r3) + PPC_LL r27,18*SZL(r3) + PPC_LL r28,19*SZL(r3) + PPC_LL r29,20*SZL(r3) + PPC_LL r30,21*SZL(r3) + PPC_LL r31,22*SZL(r3) + PPC_LL r0,3*SZL(r3) mtcrf 0x38,r0 - LDL r0,0(r3) - LDL r1,SZL(r3) - LDL r2,2*SZL(r3) + PPC_LL r0,0(r3) + PPC_LL r1,SZL(r3) + PPC_LL r2,2*SZL(r3) mtlr r0 mr r3,r4 blr @@ -84,52 +84,52 @@ _GLOBAL(xmon_longjmp) * different ABIs, though). */ _GLOBAL(xmon_save_regs) - STL r0,0*SZL(r3) - STL r2,2*SZL(r3) - STL r3,3*SZL(r3) - STL r4,4*SZL(r3) - STL r5,5*SZL(r3) - STL r6,6*SZL(r3) - STL r7,7*SZL(r3) - STL r8,8*SZL(r3) - STL r9,9*SZL(r3) - STL r10,10*SZL(r3) - STL r11,11*SZL(r3) - STL r12,12*SZL(r3) - STL r13,13*SZL(r3) - STL r14,14*SZL(r3) - STL r15,15*SZL(r3) - STL r16,16*SZL(r3) - STL r17,17*SZL(r3) - STL r18,18*SZL(r3) - STL r19,19*SZL(r3) - STL r20,20*SZL(r3) - STL r21,21*SZL(r3) - STL r22,22*SZL(r3) - STL r23,23*SZL(r3) - STL r24,24*SZL(r3) - STL r25,25*SZL(r3) - STL r26,26*SZL(r3) - STL r27,27*SZL(r3) - STL r28,28*SZL(r3) - STL r29,29*SZL(r3) - STL r30,30*SZL(r3) - STL r31,31*SZL(r3) + PPC_STL r0,0*SZL(r3) + PPC_STL r2,2*SZL(r3) + PPC_STL r3,3*SZL(r3) + PPC_STL r4,4*SZL(r3) + PPC_STL r5,5*SZL(r3) + PPC_STL r6,6*SZL(r3) + PPC_STL r7,7*SZL(r3) + PPC_STL r8,8*SZL(r3) + PPC_STL r9,9*SZL(r3) + PPC_STL r10,10*SZL(r3) + PPC_STL r11,11*SZL(r3) + PPC_STL r12,12*SZL(r3) + PPC_STL r13,13*SZL(r3) + PPC_STL r14,14*SZL(r3) + PPC_STL r15,15*SZL(r3) + PPC_STL r16,16*SZL(r3) + PPC_STL r17,17*SZL(r3) + PPC_STL r18,18*SZL(r3) + PPC_STL r19,19*SZL(r3) + PPC_STL r20,20*SZL(r3) + PPC_STL r21,21*SZL(r3) + PPC_STL r22,22*SZL(r3) + PPC_STL r23,23*SZL(r3) + PPC_STL r24,24*SZL(r3) + PPC_STL r25,25*SZL(r3) + PPC_STL r26,26*SZL(r3) + PPC_STL r27,27*SZL(r3) + PPC_STL r28,28*SZL(r3) + PPC_STL r29,29*SZL(r3) + PPC_STL r30,30*SZL(r3) + PPC_STL r31,31*SZL(r3) /* go up one stack frame for SP */ - LDL r4,0(r1) - STL r4,1*SZL(r3) + PPC_LL r4,0(r1) + PPC_STL r4,1*SZL(r3) /* get caller's LR */ - LDL r0,LRSAVE(r4) - STL r0,_NIP-STACK_FRAME_OVERHEAD(r3) - STL r0,_LINK-STACK_FRAME_OVERHEAD(r3) + PPC_LL r0,LRSAVE(r4) + PPC_STL r0,_NIP-STACK_FRAME_OVERHEAD(r3) + PPC_STL r0,_LINK-STACK_FRAME_OVERHEAD(r3) mfmsr r0 - STL r0,_MSR-STACK_FRAME_OVERHEAD(r3) + PPC_STL r0,_MSR-STACK_FRAME_OVERHEAD(r3) mfctr r0 - STL r0,_CTR-STACK_FRAME_OVERHEAD(r3) + PPC_STL r0,_CTR-STACK_FRAME_OVERHEAD(r3) mfxer r0 - STL r0,_XER-STACK_FRAME_OVERHEAD(r3) + PPC_STL r0,_XER-STACK_FRAME_OVERHEAD(r3) mfcr r0 - STL r0,_CCR-STACK_FRAME_OVERHEAD(r3) + PPC_STL r0,_CCR-STACK_FRAME_OVERHEAD(r3) li r0,0 - STL r0,_TRAP-STACK_FRAME_OVERHEAD(r3) + PPC_STL r0,_TRAP-STACK_FRAME_OVERHEAD(r3) blr diff --git a/arch/powerpc/xmon/start_32.c b/arch/powerpc/xmon/start_32.c index 69b658c0f76..c2464df4217 100644 --- a/arch/powerpc/xmon/start_32.c +++ b/arch/powerpc/xmon/start_32.c @@ -11,7 +11,6 @@ #include <linux/cuda.h> #include <linux/kernel.h> #include <linux/errno.h> -#include <linux/sysrq.h> #include <linux/bitops.h> #include <asm/xmon.h> #include <asm/prom.h> @@ -22,10 +21,11 @@ #include <asm/processor.h> #include <asm/delay.h> #include <asm/btext.h> +#include <asm/time.h> +#include "nonstdio.h" static volatile unsigned char __iomem *sccc, *sccd; unsigned int TXRDY, RXRDY, DLAB; -static int xmon_expect(const char *str, unsigned int timeout); static int use_serial; static int use_screen; @@ -33,16 +33,6 @@ static int via_modem; static int xmon_use_sccb; static struct device_node *channel_node; -#define TB_SPEED 25000000 - -static inline unsigned int readtb(void) -{ - unsigned int ret; - - asm volatile("mftb %0" : "=r" (ret) :); - return ret; -} - void buf_access(void) { if (DLAB) @@ -91,23 +81,7 @@ static unsigned long chrp_find_phys_io_base(void) } #endif /* CONFIG_PPC_CHRP */ -#ifdef CONFIG_MAGIC_SYSRQ -static void sysrq_handle_xmon(int key, struct pt_regs *regs, - struct tty_struct *tty) -{ - xmon(regs); -} - -static struct sysrq_key_op sysrq_xmon_op = -{ - .handler = sysrq_handle_xmon, - .help_msg = "Xmon", - .action_msg = "Entering xmon", -}; -#endif - -void -xmon_map_scc(void) +void xmon_map_scc(void) { #ifdef CONFIG_PPC_MULTIPLATFORM volatile unsigned char __iomem *base; @@ -217,8 +191,6 @@ xmon_map_scc(void) RXRDY = 1; DLAB = 0x80; #endif /* platform */ - - register_sysrq_key('x', &sysrq_xmon_op); } static int scc_initialized = 0; @@ -238,8 +210,7 @@ static inline void do_poll_adb(void) #endif /* CONFIG_ADB_CUDA */ } -int -xmon_write(void *handle, void *ptr, int nb) +int xmon_write(void *ptr, int nb) { char *p = ptr; int i, c, ct; @@ -311,8 +282,7 @@ static unsigned char xmon_shift_keytab[128] = "\0.\0*\0+\0\0\0\0\0/\r\0-\0" /* 0x40 - 0x4f */ "\0\0000123456789\0\0\0"; /* 0x50 - 0x5f */ -static int -xmon_get_adb_key(void) +static int xmon_get_adb_key(void) { int k, t, on; @@ -350,32 +320,21 @@ xmon_get_adb_key(void) } #endif /* CONFIG_BOOTX_TEXT */ -int -xmon_read(void *handle, void *ptr, int nb) +int xmon_readchar(void) { - char *p = ptr; - int i; - #ifdef CONFIG_BOOTX_TEXT - if (use_screen) { - for (i = 0; i < nb; ++i) - *p++ = xmon_get_adb_key(); - return i; - } + if (use_screen) + return xmon_get_adb_key(); #endif - if (!scc_initialized) - xmon_init_scc(); - for (i = 0; i < nb; ++i) { + if (!scc_initialized) + xmon_init_scc(); while ((*sccc & RXRDY) == 0) - do_poll_adb(); + do_poll_adb(); buf_access(); - *p++ = *sccd; - } - return i; + return *sccd; } -int -xmon_read_poll(void) +int xmon_read_poll(void) { if ((*sccc & RXRDY) == 0) { do_poll_adb(); @@ -395,8 +354,7 @@ static unsigned char scc_inittab[] = { 3, 0xc1, /* rx enable, 8 bits */ }; -void -xmon_init_scc(void) +void xmon_init_scc(void) { if ( _machine == _MACH_chrp ) { @@ -410,6 +368,7 @@ xmon_init_scc(void) else if ( _machine == _MACH_Pmac ) { int i, x; + unsigned long timeout; if (channel_node != 0) pmac_call_feature( @@ -424,8 +383,12 @@ xmon_init_scc(void) PMAC_FTR_MODEM_ENABLE, channel_node, 0, 1); printk(KERN_INFO "Modem powered up by debugger !\n"); - t0 = readtb(); - while (readtb() - t0 < 3*TB_SPEED) + t0 = get_tbl(); + timeout = 3 * tb_ticks_per_sec; + if (timeout == 0) + /* assume 25MHz if tb_ticks_per_sec not set */ + timeout = 75000000; + while (get_tbl() - t0 < timeout) eieio(); } /* use the B channel if requested */ @@ -447,164 +410,19 @@ xmon_init_scc(void) scc_initialized = 1; if (via_modem) { for (;;) { - xmon_write(NULL, "ATE1V1\r", 7); + xmon_write("ATE1V1\r", 7); if (xmon_expect("OK", 5)) { - xmon_write(NULL, "ATA\r", 4); + xmon_write("ATA\r", 4); if (xmon_expect("CONNECT", 40)) break; } - xmon_write(NULL, "+++", 3); + xmon_write("+++", 3); xmon_expect("OK", 3); } } } -void *xmon_stdin; -void *xmon_stdout; -void *xmon_stderr; - -int xmon_putc(int c, void *f) -{ - char ch = c; - - if (c == '\n') - xmon_putc('\r', f); - return xmon_write(f, &ch, 1) == 1? c: -1; -} - -int xmon_putchar(int c) -{ - return xmon_putc(c, xmon_stdout); -} - -int xmon_fputs(char *str, void *f) -{ - int n = strlen(str); - - return xmon_write(f, str, n) == n? 0: -1; -} - -int -xmon_readchar(void) -{ - char ch; - - for (;;) { - switch (xmon_read(xmon_stdin, &ch, 1)) { - case 1: - return ch; - case -1: - xmon_printf("read(stdin) returned -1\r\n", 0, 0); - return -1; - } - } -} - -static char line[256]; -static char *lineptr; -static int lineleft; - -int xmon_expect(const char *str, unsigned int timeout) -{ - int c; - unsigned int t0; - - timeout *= TB_SPEED; - t0 = readtb(); - do { - lineptr = line; - for (;;) { - c = xmon_read_poll(); - if (c == -1) { - if (readtb() - t0 > timeout) - return 0; - continue; - } - if (c == '\n') - break; - if (c != '\r' && lineptr < &line[sizeof(line) - 1]) - *lineptr++ = c; - } - *lineptr = 0; - } while (strstr(line, str) == NULL); - return 1; -} - -int -xmon_getchar(void) -{ - int c; - - if (lineleft == 0) { - lineptr = line; - for (;;) { - c = xmon_readchar(); - if (c == -1 || c == 4) - break; - if (c == '\r' || c == '\n') { - *lineptr++ = '\n'; - xmon_putchar('\n'); - break; - } - switch (c) { - case 0177: - case '\b': - if (lineptr > line) { - xmon_putchar('\b'); - xmon_putchar(' '); - xmon_putchar('\b'); - --lineptr; - } - break; - case 'U' & 0x1F: - while (lineptr > line) { - xmon_putchar('\b'); - xmon_putchar(' '); - xmon_putchar('\b'); - --lineptr; - } - break; - default: - if (lineptr >= &line[sizeof(line) - 1]) - xmon_putchar('\a'); - else { - xmon_putchar(c); - *lineptr++ = c; - } - } - } - lineleft = lineptr - line; - lineptr = line; - } - if (lineleft == 0) - return -1; - --lineleft; - return *lineptr++; -} - -char * -xmon_fgets(char *str, int nb, void *f) -{ - char *p; - int c; - - for (p = str; p < str + nb - 1; ) { - c = xmon_getchar(); - if (c == -1) { - if (p == str) - return NULL; - break; - } - *p++ = c; - if (c == '\n') - break; - } - *p = 0; - return str; -} - -void -xmon_enter(void) +void xmon_enter(void) { #ifdef CONFIG_ADB_PMU if (_machine == _MACH_Pmac) { @@ -613,8 +431,7 @@ xmon_enter(void) #endif } -void -xmon_leave(void) +void xmon_leave(void) { #ifdef CONFIG_ADB_PMU if (_machine == _MACH_Pmac) { diff --git a/arch/powerpc/xmon/start_64.c b/arch/powerpc/xmon/start_64.c index e50c158191e..712552c4f24 100644 --- a/arch/powerpc/xmon/start_64.c +++ b/arch/powerpc/xmon/start_64.c @@ -6,182 +6,29 @@ * as published by the Free Software Foundation; either version * 2 of the License, or (at your option) any later version. */ -#include <linux/config.h> -#include <linux/string.h> -#include <linux/kernel.h> -#include <linux/errno.h> -#include <linux/sysrq.h> -#include <linux/init.h> #include <asm/machdep.h> -#include <asm/io.h> -#include <asm/page.h> -#include <asm/prom.h> -#include <asm/processor.h> #include <asm/udbg.h> -#include <asm/system.h> #include "nonstdio.h" -#ifdef CONFIG_MAGIC_SYSRQ - -static void sysrq_handle_xmon(int key, struct pt_regs *pt_regs, - struct tty_struct *tty) -{ - /* ensure xmon is enabled */ - xmon_init(1); - debugger(pt_regs); -} - -static struct sysrq_key_op sysrq_xmon_op = +void xmon_map_scc(void) { - .handler = sysrq_handle_xmon, - .help_msg = "Xmon", - .action_msg = "Entering xmon", -}; - -static int __init setup_xmon_sysrq(void) -{ - register_sysrq_key('x', &sysrq_xmon_op); - return 0; } -__initcall(setup_xmon_sysrq); -#endif /* CONFIG_MAGIC_SYSRQ */ -int -xmon_write(void *handle, void *ptr, int nb) +int xmon_write(void *ptr, int nb) { return udbg_write(ptr, nb); } -int -xmon_read(void *handle, void *ptr, int nb) +int xmon_readchar(void) { - return udbg_read(ptr, nb); + if (udbg_getc) + return udbg_getc(); + return -1; } -int -xmon_read_poll(void) +int xmon_read_poll(void) { if (udbg_getc_poll) return udbg_getc_poll(); return -1; } - -FILE *xmon_stdin; -FILE *xmon_stdout; - -int -xmon_putc(int c, void *f) -{ - char ch = c; - - if (c == '\n') - xmon_putc('\r', f); - return xmon_write(f, &ch, 1) == 1? c: -1; -} - -int -xmon_putchar(int c) -{ - return xmon_putc(c, xmon_stdout); -} - -int -xmon_fputs(char *str, void *f) -{ - int n = strlen(str); - - return xmon_write(f, str, n) == n? 0: -1; -} - -int -xmon_readchar(void) -{ - char ch; - - for (;;) { - switch (xmon_read(xmon_stdin, &ch, 1)) { - case 1: - return ch; - case -1: - xmon_printf("read(stdin) returned -1\r\n", 0, 0); - return -1; - } - } -} - -static char line[256]; -static char *lineptr; -static int lineleft; - -int -xmon_getchar(void) -{ - int c; - - if (lineleft == 0) { - lineptr = line; - for (;;) { - c = xmon_readchar(); - if (c == -1 || c == 4) - break; - if (c == '\r' || c == '\n') { - *lineptr++ = '\n'; - xmon_putchar('\n'); - break; - } - switch (c) { - case 0177: - case '\b': - if (lineptr > line) { - xmon_putchar('\b'); - xmon_putchar(' '); - xmon_putchar('\b'); - --lineptr; - } - break; - case 'U' & 0x1F: - while (lineptr > line) { - xmon_putchar('\b'); - xmon_putchar(' '); - xmon_putchar('\b'); - --lineptr; - } - break; - default: - if (lineptr >= &line[sizeof(line) - 1]) - xmon_putchar('\a'); - else { - xmon_putchar(c); - *lineptr++ = c; - } - } - } - lineleft = lineptr - line; - lineptr = line; - } - if (lineleft == 0) - return -1; - --lineleft; - return *lineptr++; -} - -char * -xmon_fgets(char *str, int nb, void *f) -{ - char *p; - int c; - - for (p = str; p < str + nb - 1; ) { - c = xmon_getchar(); - if (c == -1) { - if (p == str) - return NULL; - break; - } - *p++ = c; - if (c == '\n') - break; - } - *p = 0; - return str; -} diff --git a/arch/powerpc/xmon/start_8xx.c b/arch/powerpc/xmon/start_8xx.c index a48bd594cf6..4c17b0486ad 100644 --- a/arch/powerpc/xmon/start_8xx.c +++ b/arch/powerpc/xmon/start_8xx.c @@ -15,273 +15,30 @@ #include <asm/8xx_immap.h> #include <asm/mpc8xx.h> #include <asm/commproc.h> +#include "nonstdio.h" -extern void xmon_printf(const char *fmt, ...); extern int xmon_8xx_write(char *str, int nb); extern int xmon_8xx_read_poll(void); extern int xmon_8xx_read_char(void); -void prom_drawhex(uint); -void prom_drawstring(const char *str); -static int use_screen = 1; /* default */ - -#define TB_SPEED 25000000 - -static inline unsigned int readtb(void) -{ - unsigned int ret; - - asm volatile("mftb %0" : "=r" (ret) :); - return ret; -} - -void buf_access(void) -{ -} - -void -xmon_map_scc(void) +void xmon_map_scc(void) { - cpmp = (cpm8xx_t *)&(((immap_t *)IMAP_ADDR)->im_cpm); - use_screen = 0; - - prom_drawstring("xmon uses serial port\n"); } -static int scc_initialized = 0; - void xmon_init_scc(void); -int -xmon_write(void *handle, void *ptr, int nb) +int xmon_write(void *ptr, int nb) { - char *p = ptr; - int i, c, ct; - - if (!scc_initialized) - xmon_init_scc(); - return(xmon_8xx_write(ptr, nb)); } -int xmon_wants_key; - -int -xmon_read(void *handle, void *ptr, int nb) +int xmon_readchar(void) { - char *p = ptr; - int i; - - if (!scc_initialized) - xmon_init_scc(); - - for (i = 0; i < nb; ++i) { - *p++ = xmon_8xx_read_char(); - } - return i; + return xmon_8xx_read_char(); } -int -xmon_read_poll(void) +int xmon_read_poll(void) { return(xmon_8xx_read_poll()); } - -void -xmon_init_scc() -{ - scc_initialized = 1; -} - -#if 0 -extern int (*prom_entry)(void *); - -int -xmon_exit(void) -{ - struct prom_args { - char *service; - } args; - - for (;;) { - args.service = "exit"; - (*prom_entry)(&args); - } -} -#endif - -void *xmon_stdin; -void *xmon_stdout; -void *xmon_stderr; - -void -xmon_init(void) -{ -} - -int -xmon_putc(int c, void *f) -{ - char ch = c; - - if (c == '\n') - xmon_putc('\r', f); - return xmon_write(f, &ch, 1) == 1? c: -1; -} - -int -xmon_putchar(int c) -{ - return xmon_putc(c, xmon_stdout); -} - -int -xmon_fputs(char *str, void *f) -{ - int n = strlen(str); - - return xmon_write(f, str, n) == n? 0: -1; -} - -int -xmon_readchar(void) -{ - char ch; - - for (;;) { - switch (xmon_read(xmon_stdin, &ch, 1)) { - case 1: - return ch; - case -1: - xmon_printf("read(stdin) returned -1\r\n", 0, 0); - return -1; - } - } -} - -static char line[256]; -static char *lineptr; -static int lineleft; - -#if 0 -int xmon_expect(const char *str, unsigned int timeout) -{ - int c; - unsigned int t0; - - timeout *= TB_SPEED; - t0 = readtb(); - do { - lineptr = line; - for (;;) { - c = xmon_read_poll(); - if (c == -1) { - if (readtb() - t0 > timeout) - return 0; - continue; - } - if (c == '\n') - break; - if (c != '\r' && lineptr < &line[sizeof(line) - 1]) - *lineptr++ = c; - } - *lineptr = 0; - } while (strstr(line, str) == NULL); - return 1; -} -#endif - -int -xmon_getchar(void) -{ - int c; - - if (lineleft == 0) { - lineptr = line; - for (;;) { - c = xmon_readchar(); - if (c == -1 || c == 4) - break; - if (c == '\r' || c == '\n') { - *lineptr++ = '\n'; - xmon_putchar('\n'); - break; - } - switch (c) { - case 0177: - case '\b': - if (lineptr > line) { - xmon_putchar('\b'); - xmon_putchar(' '); - xmon_putchar('\b'); - --lineptr; - } - break; - case 'U' & 0x1F: - while (lineptr > line) { - xmon_putchar('\b'); - xmon_putchar(' '); - xmon_putchar('\b'); - --lineptr; - } - break; - default: - if (lineptr >= &line[sizeof(line) - 1]) - xmon_putchar('\a'); - else { - xmon_putchar(c); - *lineptr++ = c; - } - } - } - lineleft = lineptr - line; - lineptr = line; - } - if (lineleft == 0) - return -1; - --lineleft; - return *lineptr++; -} - -char * -xmon_fgets(char *str, int nb, void *f) -{ - char *p; - int c; - - for (p = str; p < str + nb - 1; ) { - c = xmon_getchar(); - if (c == -1) { - if (p == str) - return 0; - break; - } - *p++ = c; - if (c == '\n') - break; - } - *p = 0; - return str; -} - -void -prom_drawhex(uint val) -{ - unsigned char buf[10]; - - int i; - for (i = 7; i >= 0; i--) - { - buf[i] = "0123456789abcdef"[val & 0x0f]; - val >>= 4; - } - buf[8] = '\0'; - xmon_fputs(buf, xmon_stdout); -} - -void -prom_drawstring(const char *str) -{ - xmon_fputs(str, xmon_stdout); -} diff --git a/arch/powerpc/xmon/subr_prf.c b/arch/powerpc/xmon/subr_prf.c deleted file mode 100644 index b48738c6dd3..00000000000 --- a/arch/powerpc/xmon/subr_prf.c +++ /dev/null @@ -1,54 +0,0 @@ -/* - * Written by Cort Dougan to replace the version originally used - * by Paul Mackerras, which came from NetBSD and thus had copyright - * conflicts with Linux. - * - * This file makes liberal use of the standard linux utility - * routines to reduce the size of the binary. We assume we can - * trust some parts of Linux inside the debugger. - * -- Cort (cort@cs.nmt.edu) - * - * Copyright (C) 1999 Cort Dougan. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/kernel.h> -#include <linux/string.h> -#include <linux/module.h> -#include <stdarg.h> -#include "nonstdio.h" - -extern int xmon_write(void *, void *, int); - -void xmon_vfprintf(void *f, const char *fmt, va_list ap) -{ - static char xmon_buf[2048]; - int n; - - n = vsprintf(xmon_buf, fmt, ap); - xmon_write(f, xmon_buf, n); -} - -void xmon_printf(const char *fmt, ...) -{ - va_list ap; - - va_start(ap, fmt); - xmon_vfprintf(stdout, fmt, ap); - va_end(ap); -} -EXPORT_SYMBOL(xmon_printf); - -void xmon_fprintf(void *f, const char *fmt, ...) -{ - va_list ap; - - va_start(ap, fmt); - xmon_vfprintf(f, fmt, ap); - va_end(ap); -} - diff --git a/arch/powerpc/xmon/xmon.c b/arch/powerpc/xmon/xmon.c index 1124f114620..cfcb2a56d66 100644 --- a/arch/powerpc/xmon/xmon.c +++ b/arch/powerpc/xmon/xmon.c @@ -1,7 +1,7 @@ /* * Routines providing a simple monitor for use on the PowerMac. * - * Copyright (C) 1996 Paul Mackerras. + * Copyright (C) 1996-2005 Paul Mackerras. * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -18,6 +18,7 @@ #include <linux/kallsyms.h> #include <linux/cpumask.h> #include <linux/module.h> +#include <linux/sysrq.h> #include <asm/ptrace.h> #include <asm/string.h> @@ -144,15 +145,10 @@ static void xmon_print_symbol(unsigned long address, const char *mid, static const char *getvecname(unsigned long vec); extern int print_insn_powerpc(unsigned long, unsigned long, int); -extern void printf(const char *fmt, ...); -extern void xmon_vfprintf(void *f, const char *fmt, va_list ap); -extern int xmon_putc(int c, void *f); -extern int putchar(int ch); extern void xmon_enter(void); extern void xmon_leave(void); -extern int xmon_read_poll(void); extern long setjmp(long *); extern void longjmp(long *, long); extern void xmon_save_regs(struct pt_regs *); @@ -748,7 +744,6 @@ cmds(struct pt_regs *excp) printf("%x:", smp_processor_id()); #endif /* CONFIG_SMP */ printf("mon> "); - fflush(stdout); flush_input(); termch = 0; cmd = skipbl(); @@ -1797,7 +1792,7 @@ memex(void) for(;;){ if (!mnoread) n = mread(adrs, val, size); - printf("%.16x%c", adrs, brev? 'r': ' '); + printf(REG"%c", adrs, brev? 'r': ' '); if (!mnoread) { if (brev) byterev(val, size); @@ -1976,17 +1971,18 @@ prdump(unsigned long adrs, long ndump) nr = mread(adrs, temp, r); adrs += nr; for (m = 0; m < r; ++m) { - if ((m & 7) == 0 && m > 0) - putchar(' '); + if ((m & (sizeof(long) - 1)) == 0 && m > 0) + putchar(' '); if (m < nr) printf("%.2x", temp[m]); else printf("%s", fault_chars[fault_type]); } - if (m <= 8) - printf(" "); - for (; m < 16; ++m) + for (; m < 16; ++m) { + if ((m & (sizeof(long) - 1)) == 0) + putchar(' '); printf(" "); + } printf(" |"); for (m = 0; m < r; ++m) { if (m < nr) { @@ -2151,7 +2147,6 @@ memzcan(void) ok = mread(a, &v, 1); if (ok && !ook) { printf("%.8x .. ", a); - fflush(stdout); } else if (!ok && ook) printf("%.8x\n", a - mskip); ook = ok; @@ -2372,7 +2367,7 @@ int inchar(void) { if (lineptr == NULL || *lineptr == 0) { - if (fgets(line, sizeof(line), stdin) == NULL) { + if (xmon_gets(line, sizeof(line)) == NULL) { lineptr = NULL; return EOF; } @@ -2526,4 +2521,29 @@ void xmon_init(int enable) __debugger_dabr_match = NULL; __debugger_fault_handler = NULL; } + xmon_map_scc(); +} + +#ifdef CONFIG_MAGIC_SYSRQ +static void sysrq_handle_xmon(int key, struct pt_regs *pt_regs, + struct tty_struct *tty) +{ + /* ensure xmon is enabled */ + xmon_init(1); + debugger(pt_regs); +} + +static struct sysrq_key_op sysrq_xmon_op = +{ + .handler = sysrq_handle_xmon, + .help_msg = "Xmon", + .action_msg = "Entering xmon", +}; + +static int __init setup_xmon_sysrq(void) +{ + register_sysrq_key('x', &sysrq_xmon_op); + return 0; } +__initcall(setup_xmon_sysrq); +#endif /* CONFIG_MAGIC_SYSRQ */ diff --git a/arch/ppc/boot/include/of1275.h b/arch/ppc/boot/include/of1275.h index 69173df76db..4ed88acfa73 100644 --- a/arch/ppc/boot/include/of1275.h +++ b/arch/ppc/boot/include/of1275.h @@ -19,6 +19,9 @@ extern prom_entry of_prom_entry; /* function declarations */ +int call_prom(const char *service, int nargs, int nret, ...); +int call_prom_ret(const char *service, int nargs, int nret, + unsigned int *rets, ...); void * claim(unsigned int virt, unsigned int size, unsigned int align); int map(unsigned int phys, unsigned int virt, unsigned int size); void enter(void); diff --git a/arch/ppc/boot/of1275/Makefile b/arch/ppc/boot/of1275/Makefile index 02e6f235d7c..0b979c00497 100644 --- a/arch/ppc/boot/of1275/Makefile +++ b/arch/ppc/boot/of1275/Makefile @@ -3,4 +3,4 @@ # lib-y := claim.o enter.o exit.o finddevice.o getprop.o ofinit.o \ - ofstdio.o read.o release.o write.o map.o + ofstdio.o read.o release.o write.o map.o call_prom.o diff --git a/arch/ppc/boot/of1275/call_prom.c b/arch/ppc/boot/of1275/call_prom.c new file mode 100644 index 00000000000..9479a3a2b8c --- /dev/null +++ b/arch/ppc/boot/of1275/call_prom.c @@ -0,0 +1,74 @@ +/* + * Copyright (C) 1996-2005 Paul Mackerras. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include "of1275.h" +#include <stdarg.h> + +int call_prom(const char *service, int nargs, int nret, ...) +{ + int i; + struct prom_args { + const char *service; + int nargs; + int nret; + unsigned int args[12]; + } args; + va_list list; + + args.service = service; + args.nargs = nargs; + args.nret = nret; + + va_start(list, nret); + for (i = 0; i < nargs; i++) + args.args[i] = va_arg(list, unsigned int); + va_end(list); + + for (i = 0; i < nret; i++) + args.args[nargs+i] = 0; + + if (of_prom_entry(&args) < 0) + return -1; + + return (nret > 0)? args.args[nargs]: 0; +} + +int call_prom_ret(const char *service, int nargs, int nret, + unsigned int *rets, ...) +{ + int i; + struct prom_args { + const char *service; + int nargs; + int nret; + unsigned int args[12]; + } args; + va_list list; + + args.service = service; + args.nargs = nargs; + args.nret = nret; + + va_start(list, rets); + for (i = 0; i < nargs; i++) + args.args[i] = va_arg(list, unsigned int); + va_end(list); + + for (i = 0; i < nret; i++) + args.args[nargs+i] = 0; + + if (of_prom_entry(&args) < 0) + return -1; + + if (rets != (void *) 0) + for (i = 1; i < nret; ++i) + rets[i-1] = args.args[nargs+i]; + + return (nret > 0)? args.args[nargs]: 0; +} diff --git a/arch/ppc/boot/of1275/claim.c b/arch/ppc/boot/of1275/claim.c index 13169a5c433..1ed3aeeff8a 100644 --- a/arch/ppc/boot/of1275/claim.c +++ b/arch/ppc/boot/of1275/claim.c @@ -9,27 +9,84 @@ */ #include "of1275.h" +#include "nonstdio.h" -void * -claim(unsigned int virt, unsigned int size, unsigned int align) +/* + * Older OF's require that when claiming a specific range of addresses, + * we claim the physical space in the /memory node and the virtual + * space in the chosen mmu node, and then do a map operation to + * map virtual to physical. + */ +static int need_map = -1; +static ihandle chosen_mmu; +static phandle memory; + +/* returns true if s2 is a prefix of s1 */ +static int string_match(const char *s1, const char *s2) +{ + for (; *s2; ++s2) + if (*s1++ != *s2) + return 0; + return 1; +} + +static int check_of_version(void) +{ + phandle oprom, chosen; + char version[64]; + + oprom = finddevice("/openprom"); + if (oprom == OF_INVALID_HANDLE) + return 0; + if (getprop(oprom, "model", version, sizeof(version)) <= 0) + return 0; + version[sizeof(version)-1] = 0; + printf("OF version = '%s'\n", version); + if (!string_match(version, "Open Firmware, 1.") + && !string_match(version, "FirmWorks,3.")) + return 0; + chosen = finddevice("/chosen"); + if (chosen == OF_INVALID_HANDLE) { + chosen = finddevice("/chosen@0"); + if (chosen == OF_INVALID_HANDLE) { + printf("no chosen\n"); + return 0; + } + } + if (getprop(chosen, "mmu", &chosen_mmu, sizeof(chosen_mmu)) <= 0) { + printf("no mmu\n"); + return 0; + } + memory = (ihandle) call_prom("open", 1, 1, "/memory"); + if (memory == OF_INVALID_HANDLE) { + memory = (ihandle) call_prom("open", 1, 1, "/memory@0"); + if (memory == OF_INVALID_HANDLE) { + printf("no memory node\n"); + return 0; + } + } + printf("old OF detected\n"); + return 1; +} + +void *claim(unsigned int virt, unsigned int size, unsigned int align) { - struct prom_args { - char *service; - int nargs; - int nret; - unsigned int virt; - unsigned int size; - unsigned int align; - void *ret; - } args; + int ret; + unsigned int result; - args.service = "claim"; - args.nargs = 3; - args.nret = 1; - args.virt = virt; - args.size = size; - args.align = align; - args.ret = (void *) 0; - (*of_prom_entry)(&args); - return args.ret; + if (need_map < 0) + need_map = check_of_version(); + if (align || !need_map) + return (void *) call_prom("claim", 3, 1, virt, size, align); + + ret = call_prom_ret("call-method", 5, 2, &result, "claim", memory, + align, size, virt); + if (ret != 0 || result == -1) + return (void *) -1; + ret = call_prom_ret("call-method", 5, 2, &result, "claim", chosen_mmu, + align, size, virt); + /* 0x12 == coherent + read/write */ + ret = call_prom("call-method", 6, 1, "map", chosen_mmu, + 0x12, size, virt, virt); + return virt; } diff --git a/arch/ppc/boot/of1275/finddevice.c b/arch/ppc/boot/of1275/finddevice.c index 2c0f7cbb793..0dcb1201b77 100644 --- a/arch/ppc/boot/of1275/finddevice.c +++ b/arch/ppc/boot/of1275/finddevice.c @@ -10,22 +10,7 @@ #include "of1275.h" -phandle -finddevice(const char *name) +phandle finddevice(const char *name) { - struct prom_args { - char *service; - int nargs; - int nret; - const char *devspec; - phandle device; - } args; - - args.service = "finddevice"; - args.nargs = 1; - args.nret = 1; - args.devspec = name; - args.device = OF_INVALID_HANDLE; - (*of_prom_entry)(&args); - return args.device; + return (phandle) call_prom("finddevice", 1, 1, name); } diff --git a/arch/ppc/boot/openfirmware/Makefile b/arch/ppc/boot/openfirmware/Makefile index 03415238fab..83a6433459c 100644 --- a/arch/ppc/boot/openfirmware/Makefile +++ b/arch/ppc/boot/openfirmware/Makefile @@ -80,8 +80,7 @@ $(obj)/note: $(utils)/mknote FORCE $(call if_changed,mknote) -$(obj)/coffcrt0.o: EXTRA_AFLAGS := -traditional -DXCOFF -$(obj)/crt0.o: EXTRA_AFLAGS := -traditional +$(obj)/coffcrt0.o: EXTRA_AFLAGS := -DXCOFF targets += coffcrt0.o crt0.o $(obj)/coffcrt0.o $(obj)/crt0.o: $(common)/crt0.S FORCE $(call if_changed_dep,as_o_S) diff --git a/arch/ppc/kernel/Makefile b/arch/ppc/kernel/Makefile index 76a55a438f2..17a4da65e27 100644 --- a/arch/ppc/kernel/Makefile +++ b/arch/ppc/kernel/Makefile @@ -12,7 +12,7 @@ extra-$(CONFIG_6xx) += idle_6xx.o extra-$(CONFIG_POWER4) += idle_power4.o extra-y += vmlinux.lds -obj-y := entry.o traps.o irq.o idle.o time.o misc.o \ +obj-y := entry.o traps.o idle.o time.o misc.o \ process.o align.o \ setup.o \ ppc_htab.o @@ -38,8 +38,7 @@ endif # These are here while we do the architecture merge else -obj-y := irq.o idle.o \ - align.o +obj-y := idle.o align.o obj-$(CONFIG_6xx) += l2cr.o cpu_setup_6xx.o obj-$(CONFIG_SOFTWARE_SUSPEND) += swsusp.o obj-$(CONFIG_MODULES) += module.o diff --git a/arch/ppc/kernel/head_booke.h b/arch/ppc/kernel/head_booke.h index aeb349b47af..f3d274c6b23 100644 --- a/arch/ppc/kernel/head_booke.h +++ b/arch/ppc/kernel/head_booke.h @@ -358,6 +358,6 @@ label: NORMAL_EXCEPTION_PROLOG; \ bne load_up_fpu; /* if from user, just load it up */ \ addi r3,r1,STACK_FRAME_OVERHEAD; \ - EXC_XFER_EE_LITE(0x800, KernelFP) + EXC_XFER_EE_LITE(0x800, kernel_fp_unavailable_exception) #endif /* __HEAD_BOOKE_H__ */ diff --git a/arch/ppc/kernel/idle.c b/arch/ppc/kernel/idle.c index 3c4e4cb6107..821a75e4560 100644 --- a/arch/ppc/kernel/idle.c +++ b/arch/ppc/kernel/idle.c @@ -63,7 +63,7 @@ void cpu_idle(void) int cpu = smp_processor_id(); for (;;) { - while (need_resched()) { + while (!need_resched()) { if (ppc_md.idle != NULL) ppc_md.idle(); else diff --git a/arch/ppc/kernel/irq.c b/arch/ppc/kernel/irq.c deleted file mode 100644 index fbb2b9f8922..00000000000 --- a/arch/ppc/kernel/irq.c +++ /dev/null @@ -1,165 +0,0 @@ -/* - * arch/ppc/kernel/irq.c - * - * Derived from arch/i386/kernel/irq.c - * Copyright (C) 1992 Linus Torvalds - * Adapted from arch/i386 by Gary Thomas - * Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org) - * Updated and modified by Cort Dougan <cort@fsmlabs.com> - * Copyright (C) 1996-2001 Cort Dougan - * Adapted for Power Macintosh by Paul Mackerras - * Copyright (C) 1996 Paul Mackerras (paulus@cs.anu.edu.au) - * Amiga/APUS changes by Jesper Skov (jskov@cygnus.co.uk). - * - * This file contains the code used by various IRQ handling routines: - * asking for different IRQ's should be done through these routines - * instead of just grabbing them. Thus setups with different IRQ numbers - * shouldn't result in any weird surprises, and installing new handlers - * should be easier. - * - * The MPC8xx has an interrupt mask in the SIU. If a bit is set, the - * interrupt is _enabled_. As expected, IRQ0 is bit 0 in the 32-bit - * mask register (of which only 16 are defined), hence the weird shifting - * and complement of the cached_irq_mask. I want to be able to stuff - * this right into the SIU SMASK register. - * Many of the prep/chrp functions are conditional compiled on CONFIG_8xx - * to reduce code space and undefined function references. - */ - -#include <linux/errno.h> -#include <linux/module.h> -#include <linux/threads.h> -#include <linux/kernel_stat.h> -#include <linux/signal.h> -#include <linux/sched.h> -#include <linux/ptrace.h> -#include <linux/ioport.h> -#include <linux/interrupt.h> -#include <linux/timex.h> -#include <linux/config.h> -#include <linux/init.h> -#include <linux/slab.h> -#include <linux/pci.h> -#include <linux/delay.h> -#include <linux/irq.h> -#include <linux/proc_fs.h> -#include <linux/random.h> -#include <linux/seq_file.h> -#include <linux/cpumask.h> -#include <linux/profile.h> -#include <linux/bitops.h> - -#include <asm/uaccess.h> -#include <asm/system.h> -#include <asm/io.h> -#include <asm/pgtable.h> -#include <asm/irq.h> -#include <asm/cache.h> -#include <asm/prom.h> -#include <asm/ptrace.h> -#include <asm/machdep.h> - -#define NR_MASK_WORDS ((NR_IRQS + 31) / 32) - -extern atomic_t ipi_recv; -extern atomic_t ipi_sent; - -#define MAXCOUNT 10000000 - -int ppc_spurious_interrupts = 0; -struct irqaction *ppc_irq_action[NR_IRQS]; -unsigned long ppc_cached_irq_mask[NR_MASK_WORDS]; -unsigned long ppc_lost_interrupts[NR_MASK_WORDS]; -atomic_t ppc_n_lost_interrupts; - -#ifdef CONFIG_TAU_INT -extern int tau_initialized; -extern int tau_interrupts(int); -#endif - -int show_interrupts(struct seq_file *p, void *v) -{ - int i = *(loff_t *) v, j; - struct irqaction * action; - unsigned long flags; - - if (i == 0) { - seq_puts(p, " "); - for (j=0; j<NR_CPUS; j++) - if (cpu_online(j)) - seq_printf(p, "CPU%d ", j); - seq_putc(p, '\n'); - } - - if (i < NR_IRQS) { - spin_lock_irqsave(&irq_desc[i].lock, flags); - action = irq_desc[i].action; - if ( !action || !action->handler ) - goto skip; - seq_printf(p, "%3d: ", i); -#ifdef CONFIG_SMP - for (j = 0; j < NR_CPUS; j++) - if (cpu_online(j)) - seq_printf(p, "%10u ", - kstat_cpu(j).irqs[i]); -#else - seq_printf(p, "%10u ", kstat_irqs(i)); -#endif /* CONFIG_SMP */ - if (irq_desc[i].handler) - seq_printf(p, " %s ", irq_desc[i].handler->typename); - else - seq_puts(p, " None "); - seq_printf(p, "%s", (irq_desc[i].status & IRQ_LEVEL) ? "Level " : "Edge "); - seq_printf(p, " %s", action->name); - for (action = action->next; action; action = action->next) - seq_printf(p, ", %s", action->name); - seq_putc(p, '\n'); -skip: - spin_unlock_irqrestore(&irq_desc[i].lock, flags); - } else if (i == NR_IRQS) { -#ifdef CONFIG_TAU_INT - if (tau_initialized){ - seq_puts(p, "TAU: "); - for (j = 0; j < NR_CPUS; j++) - if (cpu_online(j)) - seq_printf(p, "%10u ", tau_interrupts(j)); - seq_puts(p, " PowerPC Thermal Assist (cpu temp)\n"); - } -#endif -#if defined(CONFIG_SMP) && !defined(CONFIG_PPC_MERGE) - /* should this be per processor send/receive? */ - seq_printf(p, "IPI (recv/sent): %10u/%u\n", - atomic_read(&ipi_recv), atomic_read(&ipi_sent)); -#endif - seq_printf(p, "BAD: %10u\n", ppc_spurious_interrupts); - } - return 0; -} - -void do_IRQ(struct pt_regs *regs) -{ - int irq, first = 1; - irq_enter(); - - /* - * Every platform is required to implement ppc_md.get_irq. - * This function will either return an irq number or -1 to - * indicate there are no more pending. But the first time - * through the loop this means there wasn't and IRQ pending. - * The value -2 is for buggy hardware and means that this IRQ - * has already been handled. -- Tom - */ - while ((irq = ppc_md.get_irq(regs)) >= 0) { - __do_IRQ(irq, regs); - first = 0; - } - if (irq != -2 && first) - /* That's not SMP safe ... but who cares ? */ - ppc_spurious_interrupts++; - irq_exit(); -} - -void __init init_IRQ(void) -{ - ppc_md.init_IRQ(); -} diff --git a/arch/ppc/kernel/misc.S b/arch/ppc/kernel/misc.S index ae6af29938a..5e61124581d 100644 --- a/arch/ppc/kernel/misc.S +++ b/arch/ppc/kernel/misc.S @@ -497,9 +497,9 @@ END_FTR_SECTION_IFCLR(CPU_FTR_SPLIT_ID_CACHE) * and invalidate the corresponding instruction cache blocks. * This is a no-op on the 601. * - * flush_icache_range(unsigned long start, unsigned long stop) + * __flush_icache_range(unsigned long start, unsigned long stop) */ -_GLOBAL(flush_icache_range) +_GLOBAL(__flush_icache_range) BEGIN_FTR_SECTION blr /* for 601, do nothing */ END_FTR_SECTION_IFCLR(CPU_FTR_SPLIT_ID_CACHE) diff --git a/arch/ppc/kernel/pci.c b/arch/ppc/kernel/pci.c index e8f4e576750..48ed58f995c 100644 --- a/arch/ppc/kernel/pci.c +++ b/arch/ppc/kernel/pci.c @@ -62,20 +62,6 @@ struct pci_controller** hose_tail = &hose_head; static int pci_bus_count; static void -fixup_rev1_53c810(struct pci_dev* dev) -{ - /* rev 1 ncr53c810 chips don't set the class at all which means - * they don't get their resources remapped. Fix that here. - */ - - if ((dev->class == PCI_CLASS_NOT_DEFINED)) { - printk("NCR 53c810 rev 1 detected, setting PCI class.\n"); - dev->class = PCI_CLASS_STORAGE_SCSI; - } -} -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_NCR, PCI_DEVICE_ID_NCR_53C810, fixup_rev1_53c810); - -static void fixup_broken_pcnet32(struct pci_dev* dev) { if ((dev->class>>8 == PCI_CLASS_NETWORK_ETHERNET)) { diff --git a/arch/ppc/kernel/ppc_ksyms.c b/arch/ppc/kernel/ppc_ksyms.c index e0ca61b37f4..66073f77519 100644 --- a/arch/ppc/kernel/ppc_ksyms.c +++ b/arch/ppc/kernel/ppc_ksyms.c @@ -46,6 +46,7 @@ #include <asm/btext.h> #include <asm/div64.h> #include <asm/xmon.h> +#include <asm/signal.h> #ifdef CONFIG_8xx #include <asm/commproc.h> @@ -57,7 +58,6 @@ extern void machine_check_exception(struct pt_regs *regs); extern void alignment_exception(struct pt_regs *regs); extern void program_check_exception(struct pt_regs *regs); extern void single_step_exception(struct pt_regs *regs); -extern int do_signal(sigset_t *, struct pt_regs *); extern int pmac_newworld; extern int sys_sigreturn(struct pt_regs *regs); @@ -78,7 +78,6 @@ EXPORT_SYMBOL(program_check_exception); EXPORT_SYMBOL(single_step_exception); EXPORT_SYMBOL(sys_sigreturn); EXPORT_SYMBOL(ppc_n_lost_interrupts); -EXPORT_SYMBOL(ppc_lost_interrupts); EXPORT_SYMBOL(ISA_DMA_THRESHOLD); EXPORT_SYMBOL(DMA_MODE_READ); @@ -176,6 +175,7 @@ EXPORT_SYMBOL(pci_bus_to_phys); #endif /* CONFIG_PCI */ #ifdef CONFIG_NOT_COHERENT_CACHE +extern void flush_dcache_all(void); EXPORT_SYMBOL(flush_dcache_all); #endif @@ -217,9 +217,6 @@ EXPORT_SYMBOL(adb_try_handler_change); EXPORT_SYMBOL(cuda_request); EXPORT_SYMBOL(cuda_poll); #endif /* CONFIG_ADB_CUDA */ -#ifdef CONFIG_PPC_MULTIPLATFORM -EXPORT_SYMBOL(_machine); -#endif #ifdef CONFIG_PPC_PMAC EXPORT_SYMBOL(sys_ctrler); EXPORT_SYMBOL(pmac_newworld); diff --git a/arch/ppc/kernel/setup.c b/arch/ppc/kernel/setup.c index 6bcb85d2b7f..dc55e1abc45 100644 --- a/arch/ppc/kernel/setup.c +++ b/arch/ppc/kernel/setup.c @@ -76,6 +76,7 @@ unsigned int DMA_MODE_WRITE; #ifdef CONFIG_PPC_MULTIPLATFORM int _machine = 0; +EXPORT_SYMBOL(_machine); extern void prep_init(unsigned long r3, unsigned long r4, unsigned long r5, unsigned long r6, unsigned long r7); diff --git a/arch/ppc/platforms/pmac_pic.c b/arch/ppc/platforms/pmac_pic.c index 9f2d95ea856..4742bf60935 100644 --- a/arch/ppc/platforms/pmac_pic.c +++ b/arch/ppc/platforms/pmac_pic.c @@ -75,6 +75,9 @@ static DEFINE_SPINLOCK(pmac_pic_lock); #define GATWICK_IRQ_POOL_SIZE 10 static struct interrupt_info gatwick_int_pool[GATWICK_IRQ_POOL_SIZE]; +#define NR_MASK_WORDS ((NR_IRQS + 31) / 32) +static unsigned long ppc_lost_interrupts[NR_MASK_WORDS]; + /* * Mark an irq as "lost". This is only used on the pmac * since it can lose interrupts (see pmac_set_irq_mask). diff --git a/arch/ppc/platforms/prep_setup.c b/arch/ppc/platforms/prep_setup.c index 067d7d53b81..4415748071d 100644 --- a/arch/ppc/platforms/prep_setup.c +++ b/arch/ppc/platforms/prep_setup.c @@ -61,6 +61,15 @@ #include <asm/pci-bridge.h> #include <asm/todc.h> +/* prep registers for L2 */ +#define CACHECRBA 0x80000823 /* Cache configuration register address */ +#define L2CACHE_MASK 0x03 /* Mask for 2 L2 Cache bits */ +#define L2CACHE_512KB 0x00 /* 512KB */ +#define L2CACHE_256KB 0x01 /* 256KB */ +#define L2CACHE_1MB 0x02 /* 1MB */ +#define L2CACHE_NONE 0x03 /* NONE */ +#define L2CACHE_PARITY 0x08 /* Mask for L2 Cache Parity Protected bit */ + TODC_ALLOC(); unsigned char ucSystemType; diff --git a/arch/ppc64/Kconfig b/arch/ppc64/Kconfig index 29552348e58..c9d32db9d76 100644 --- a/arch/ppc64/Kconfig +++ b/arch/ppc64/Kconfig @@ -297,6 +297,10 @@ config HAVE_ARCH_EARLY_PFN_TO_NID def_bool y depends on NEED_MULTIPLE_NODES +config ARCH_MEMORY_PROBE + def_bool y + depends on MEMORY_HOTPLUG + # Some NUMA nodes have memory ranges that span # other nodes. Even though a pfn is valid and # between a node's start and end pfns, it may not diff --git a/arch/ppc64/boot/addRamDisk.c b/arch/ppc64/boot/addRamDisk.c index 7f2c0947339..c02a99952be 100644 --- a/arch/ppc64/boot/addRamDisk.c +++ b/arch/ppc64/boot/addRamDisk.c @@ -5,11 +5,59 @@ #include <sys/types.h> #include <sys/stat.h> #include <string.h> +#include <elf.h> #define ElfHeaderSize (64 * 1024) #define ElfPages (ElfHeaderSize / 4096) #define KERNELBASE (0xc000000000000000) +#define _ALIGN_UP(addr,size) (((addr)+((size)-1))&(~((size)-1))) +struct addr_range { + unsigned long long addr; + unsigned long memsize; + unsigned long offset; +}; + +static int check_elf64(void *p, int size, struct addr_range *r) +{ + Elf64_Ehdr *elf64 = p; + Elf64_Phdr *elf64ph; + + if (elf64->e_ident[EI_MAG0] != ELFMAG0 || + elf64->e_ident[EI_MAG1] != ELFMAG1 || + elf64->e_ident[EI_MAG2] != ELFMAG2 || + elf64->e_ident[EI_MAG3] != ELFMAG3 || + elf64->e_ident[EI_CLASS] != ELFCLASS64 || + elf64->e_ident[EI_DATA] != ELFDATA2MSB || + elf64->e_type != ET_EXEC || elf64->e_machine != EM_PPC64) + return 0; + + if ((elf64->e_phoff + sizeof(Elf64_Phdr)) > size) + return 0; + + elf64ph = (Elf64_Phdr *) ((unsigned long)elf64 + + (unsigned long)elf64->e_phoff); + + r->memsize = (unsigned long)elf64ph->p_memsz; + r->offset = (unsigned long)elf64ph->p_offset; + r->addr = (unsigned long long)elf64ph->p_vaddr; + +#ifdef DEBUG + printf("PPC64 ELF file, ph:\n"); + printf("p_type 0x%08x\n", elf64ph->p_type); + printf("p_flags 0x%08x\n", elf64ph->p_flags); + printf("p_offset 0x%016llx\n", elf64ph->p_offset); + printf("p_vaddr 0x%016llx\n", elf64ph->p_vaddr); + printf("p_paddr 0x%016llx\n", elf64ph->p_paddr); + printf("p_filesz 0x%016llx\n", elf64ph->p_filesz); + printf("p_memsz 0x%016llx\n", elf64ph->p_memsz); + printf("p_align 0x%016llx\n", elf64ph->p_align); + printf("... skipping 0x%08lx bytes of ELF header\n", + (unsigned long)elf64ph->p_offset); +#endif + + return 64; +} void get4k(FILE *file, char *buf ) { unsigned j; @@ -34,97 +82,92 @@ void death(const char *msg, FILE *fdesc, const char *fname) int main(int argc, char **argv) { char inbuf[4096]; - FILE *ramDisk = NULL; - FILE *sysmap = NULL; - FILE *inputVmlinux = NULL; - FILE *outputVmlinux = NULL; - - unsigned i = 0; - unsigned long ramFileLen = 0; - unsigned long ramLen = 0; - unsigned long roundR = 0; - - unsigned long sysmapFileLen = 0; - unsigned long sysmapLen = 0; - unsigned long sysmapPages = 0; - char* ptr_end = NULL; - unsigned long offset_end = 0; - - unsigned long kernelLen = 0; - unsigned long actualKernelLen = 0; - unsigned long round = 0; - unsigned long roundedKernelLen = 0; - unsigned long ramStartOffs = 0; - unsigned long ramPages = 0; - unsigned long roundedKernelPages = 0; - unsigned long hvReleaseData = 0; + struct addr_range vmlinux; + FILE *ramDisk; + FILE *inputVmlinux; + FILE *outputVmlinux; + + char *rd_name, *lx_name, *out_name; + + size_t i; + unsigned long ramFileLen; + unsigned long ramLen; + unsigned long roundR; + unsigned long offset_end; + + unsigned long kernelLen; + unsigned long actualKernelLen; + unsigned long round; + unsigned long roundedKernelLen; + unsigned long ramStartOffs; + unsigned long ramPages; + unsigned long roundedKernelPages; + unsigned long hvReleaseData; u_int32_t eyeCatcher = 0xc8a5d9c4; - unsigned long naca = 0; - unsigned long xRamDisk = 0; - unsigned long xRamDiskSize = 0; - long padPages = 0; + unsigned long naca; + unsigned long xRamDisk; + unsigned long xRamDiskSize; + long padPages; if (argc < 2) { fprintf(stderr, "Name of RAM disk file missing.\n"); exit(1); } + rd_name = argv[1]; if (argc < 3) { - fprintf(stderr, "Name of System Map input file is missing.\n"); - exit(1); - } - - if (argc < 4) { fprintf(stderr, "Name of vmlinux file missing.\n"); exit(1); } + lx_name = argv[2]; - if (argc < 5) { + if (argc < 4) { fprintf(stderr, "Name of vmlinux output file missing.\n"); exit(1); } + out_name = argv[3]; - ramDisk = fopen(argv[1], "r"); + ramDisk = fopen(rd_name, "r"); if ( ! ramDisk ) { - fprintf(stderr, "RAM disk file \"%s\" failed to open.\n", argv[1]); + fprintf(stderr, "RAM disk file \"%s\" failed to open.\n", rd_name); exit(1); } - sysmap = fopen(argv[2], "r"); - if ( ! sysmap ) { - fprintf(stderr, "System Map file \"%s\" failed to open.\n", argv[2]); - exit(1); - } - - inputVmlinux = fopen(argv[3], "r"); + inputVmlinux = fopen(lx_name, "r"); if ( ! inputVmlinux ) { - fprintf(stderr, "vmlinux file \"%s\" failed to open.\n", argv[3]); + fprintf(stderr, "vmlinux file \"%s\" failed to open.\n", lx_name); exit(1); } - outputVmlinux = fopen(argv[4], "w+"); + outputVmlinux = fopen(out_name, "w+"); if ( ! outputVmlinux ) { - fprintf(stderr, "output vmlinux file \"%s\" failed to open.\n", argv[4]); + fprintf(stderr, "output vmlinux file \"%s\" failed to open.\n", out_name); exit(1); } - - - + + i = fread(inbuf, 1, sizeof(inbuf), inputVmlinux); + if (i != sizeof(inbuf)) { + fprintf(stderr, "can not read vmlinux file %s: %u\n", lx_name, i); + exit(1); + } + + i = check_elf64(inbuf, sizeof(inbuf), &vmlinux); + if (i == 0) { + fprintf(stderr, "You must have a linux kernel specified as argv[2]\n"); + exit(1); + } + /* Input Vmlinux file */ fseek(inputVmlinux, 0, SEEK_END); kernelLen = ftell(inputVmlinux); fseek(inputVmlinux, 0, SEEK_SET); - printf("kernel file size = %d\n", kernelLen); - if ( kernelLen == 0 ) { - fprintf(stderr, "You must have a linux kernel specified as argv[3]\n"); - exit(1); - } + printf("kernel file size = %lu\n", kernelLen); actualKernelLen = kernelLen - ElfHeaderSize; - printf("actual kernel length (minus ELF header) = %d\n", actualKernelLen); + printf("actual kernel length (minus ELF header) = %lu\n", actualKernelLen); round = actualKernelLen % 4096; roundedKernelLen = actualKernelLen; @@ -134,39 +177,7 @@ int main(int argc, char **argv) roundedKernelPages = roundedKernelLen / 4096; printf("Vmlinux pages to copy = %ld/0x%lx \n", roundedKernelPages, roundedKernelPages); - - - /* Input System Map file */ - /* (needs to be processed simply to determine if we need to add pad pages due to the static variables not being included in the vmlinux) */ - fseek(sysmap, 0, SEEK_END); - sysmapFileLen = ftell(sysmap); - fseek(sysmap, 0, SEEK_SET); - printf("%s file size = %ld/0x%lx \n", argv[2], sysmapFileLen, sysmapFileLen); - - sysmapLen = sysmapFileLen; - - roundR = 4096 - (sysmapLen % 4096); - if (roundR) { - printf("Rounding System Map file up to a multiple of 4096, adding %ld/0x%lx \n", roundR, roundR); - sysmapLen += roundR; - } - printf("Rounded System Map size is %ld/0x%lx \n", sysmapLen, sysmapLen); - - /* Process the Sysmap file to determine where _end is */ - sysmapPages = sysmapLen / 4096; - /* read the whole file line by line, expect that it doesn't fail */ - while ( fgets(inbuf, 4096, sysmap) ) ; - /* search for _end in the last page of the system map */ - ptr_end = strstr(inbuf, " _end"); - if (!ptr_end) { - fprintf(stderr, "Unable to find _end in the sysmap file \n"); - fprintf(stderr, "inbuf: \n"); - fprintf(stderr, "%s \n", inbuf); - exit(1); - } - printf("Found _end in the last page of the sysmap - backing up 10 characters it looks like %s", ptr_end-10); - /* convert address of _end in system map to hex offset. */ - offset_end = (unsigned int)strtol(ptr_end-10, NULL, 16); + offset_end = _ALIGN_UP(vmlinux.memsize, 4096); /* calc how many pages we need to insert between the vmlinux and the start of the ram disk */ padPages = offset_end/4096 - roundedKernelPages; @@ -194,7 +205,7 @@ int main(int argc, char **argv) fseek(ramDisk, 0, SEEK_END); ramFileLen = ftell(ramDisk); fseek(ramDisk, 0, SEEK_SET); - printf("%s file size = %ld/0x%lx \n", argv[1], ramFileLen, ramFileLen); + printf("%s file size = %ld/0x%lx \n", rd_name, ramFileLen, ramFileLen); ramLen = ramFileLen; @@ -248,19 +259,19 @@ int main(int argc, char **argv) /* fseek to the hvReleaseData pointer */ fseek(outputVmlinux, ElfHeaderSize + 0x24, SEEK_SET); if (fread(&hvReleaseData, 4, 1, outputVmlinux) != 1) { - death("Could not read hvReleaseData pointer\n", outputVmlinux, argv[4]); + death("Could not read hvReleaseData pointer\n", outputVmlinux, out_name); } hvReleaseData = ntohl(hvReleaseData); /* Convert to native int */ - printf("hvReleaseData is at %08x\n", hvReleaseData); + printf("hvReleaseData is at %08lx\n", hvReleaseData); /* fseek to the hvReleaseData */ fseek(outputVmlinux, ElfHeaderSize + hvReleaseData, SEEK_SET); if (fread(inbuf, 0x40, 1, outputVmlinux) != 1) { - death("Could not read hvReleaseData\n", outputVmlinux, argv[4]); + death("Could not read hvReleaseData\n", outputVmlinux, out_name); } /* Check hvReleaseData sanity */ if (memcmp(inbuf, &eyeCatcher, 4) != 0) { - death("hvReleaseData is invalid\n", outputVmlinux, argv[4]); + death("hvReleaseData is invalid\n", outputVmlinux, out_name); } /* Get the naca pointer */ naca = ntohl(*((u_int32_t*) &inbuf[0x0C])) - KERNELBASE; @@ -269,13 +280,13 @@ int main(int argc, char **argv) /* fseek to the naca */ fseek(outputVmlinux, ElfHeaderSize + naca, SEEK_SET); if (fread(inbuf, 0x18, 1, outputVmlinux) != 1) { - death("Could not read naca\n", outputVmlinux, argv[4]); + death("Could not read naca\n", outputVmlinux, out_name); } xRamDisk = ntohl(*((u_int32_t *) &inbuf[0x0c])); xRamDiskSize = ntohl(*((u_int32_t *) &inbuf[0x14])); /* Make sure a RAM disk isn't already present */ if ((xRamDisk != 0) || (xRamDiskSize != 0)) { - death("RAM disk is already attached to this kernel\n", outputVmlinux, argv[4]); + death("RAM disk is already attached to this kernel\n", outputVmlinux, out_name); } /* Fill in the values */ *((u_int32_t *) &inbuf[0x0c]) = htonl(ramStartOffs); @@ -285,15 +296,15 @@ int main(int argc, char **argv) fflush(outputVmlinux); fseek(outputVmlinux, ElfHeaderSize + naca, SEEK_SET); if (fwrite(inbuf, 0x18, 1, outputVmlinux) != 1) { - death("Could not write naca\n", outputVmlinux, argv[4]); + death("Could not write naca\n", outputVmlinux, out_name); } - printf("Ram Disk of 0x%lx pages is attached to the kernel at offset 0x%08x\n", + printf("Ram Disk of 0x%lx pages is attached to the kernel at offset 0x%08lx\n", ramPages, ramStartOffs); /* Done */ fclose(outputVmlinux); /* Set permission to executable */ - chmod(argv[4], S_IRUSR|S_IWUSR|S_IXUSR|S_IRGRP|S_IXGRP|S_IROTH|S_IXOTH); + chmod(out_name, S_IRUSR|S_IWUSR|S_IXUSR|S_IRGRP|S_IXGRP|S_IROTH|S_IXOTH); return 0; } diff --git a/arch/ppc64/kernel/Makefile b/arch/ppc64/kernel/Makefile index c441aebe764..58b19f10765 100644 --- a/arch/ppc64/kernel/Makefile +++ b/arch/ppc64/kernel/Makefile @@ -11,12 +11,11 @@ obj-y := misc.o prom.o endif -obj-y += irq.o idle.o dma.o \ - align.o pacaData.o \ - udbg.o ioctl32.o \ +obj-y += idle.o dma.o \ + align.o \ + udbg.o \ rtc.o \ - cpu_setup_power4.o \ - iommu.o sysfs.o vdso.o firmware.o + iommu.o vdso.o obj-y += vdso32/ vdso64/ pci-obj-$(CONFIG_PPC_MULTIPLATFORM) += pci_dn.o pci_direct_iommu.o @@ -31,15 +30,10 @@ endif obj-$(CONFIG_PPC_PSERIES) += udbg_16550.o obj-$(CONFIG_KEXEC) += machine_kexec.o -obj-$(CONFIG_EEH) += eeh.o -obj-$(CONFIG_PROC_FS) += proc_ppc64.o obj-$(CONFIG_MODULES) += module.o ifneq ($(CONFIG_PPC_MERGE),y) obj-$(CONFIG_MODULES) += ppc_ksyms.o endif -obj-$(CONFIG_PPC_RTAS) += rtas_pci.o -obj-$(CONFIG_SCANLOG) += scanlog.o -obj-$(CONFIG_LPARCFG) += lparcfg.o obj-$(CONFIG_HVC_CONSOLE) += hvconsole.o ifneq ($(CONFIG_PPC_MERGE),y) obj-$(CONFIG_BOOTX_TEXT) += btext.o @@ -52,8 +46,6 @@ obj-$(CONFIG_PPC_MAPLE) += udbg_16550.o obj-$(CONFIG_KPROBES) += kprobes.o -CFLAGS_ioctl32.o += -Ifs/ - ifneq ($(CONFIG_PPC_MERGE),y) ifeq ($(CONFIG_PPC_ISERIES),y) arch/ppc64/kernel/head.o: arch/powerpc/kernel/lparmap.s diff --git a/arch/ppc64/kernel/asm-offsets.c b/arch/ppc64/kernel/asm-offsets.c index bce9065da6c..84ab5c18ef5 100644 --- a/arch/ppc64/kernel/asm-offsets.c +++ b/arch/ppc64/kernel/asm-offsets.c @@ -74,7 +74,6 @@ int main(void) DEFINE(ICACHEL1LINESIZE, offsetof(struct ppc64_caches, iline_size)); DEFINE(ICACHEL1LOGLINESIZE, offsetof(struct ppc64_caches, log_iline_size)); DEFINE(ICACHEL1LINESPERPAGE, offsetof(struct ppc64_caches, ilines_per_page)); - DEFINE(PLATFORM, offsetof(struct systemcfg, platform)); DEFINE(PLATFORM_LPAR, PLATFORM_LPAR); /* paca */ diff --git a/arch/ppc64/kernel/head.S b/arch/ppc64/kernel/head.S index 9e8050ea122..1c869ea72d2 100644 --- a/arch/ppc64/kernel/head.S +++ b/arch/ppc64/kernel/head.S @@ -28,7 +28,6 @@ #include <asm/processor.h> #include <asm/page.h> #include <asm/mmu.h> -#include <asm/systemcfg.h> #include <asm/ppc_asm.h> #include <asm/asm-offsets.h> #include <asm/bug.h> @@ -1701,21 +1700,9 @@ _GLOBAL(__secondary_start) HMT_MEDIUM /* Set thread priority to MEDIUM */ ld r2,PACATOC(r13) - li r6,0 - stb r6,PACAPROCENABLED(r13) - -#ifndef CONFIG_PPC_ISERIES - /* Initialize the page table pointer register. */ - LOADADDR(r6,_SDR1) - ld r6,0(r6) /* get the value of _SDR1 */ - mtspr SPRN_SDR1,r6 /* set the htab location */ -#endif - /* Initialize the first segment table (or SLB) entry */ - ld r3,PACASTABVIRT(r13) /* get addr of segment table */ -BEGIN_FTR_SECTION - bl .stab_initialize -END_FTR_SECTION_IFCLR(CPU_FTR_SLB) - bl .slb_initialize + + /* Do early setup for that CPU */ + bl .early_setup_secondary /* Initialize the kernel stack. Just a repeat for iSeries. */ LOADADDR(r3,current_set) @@ -1724,37 +1711,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_SLB) addi r1,r1,THREAD_SIZE-STACK_FRAME_OVERHEAD std r1,PACAKSAVE(r13) - ld r3,PACASTABREAL(r13) /* get raddr of segment table */ - ori r4,r3,1 /* turn on valid bit */ - -#ifdef CONFIG_PPC_ISERIES - li r0,-1 /* hypervisor call */ - li r3,1 - sldi r3,r3,63 /* 0x8000000000000000 */ - ori r3,r3,4 /* 0x8000000000000004 */ - sc /* HvCall_setASR */ -#else - /* set the ASR */ - ld r3,systemcfg@got(r2) /* r3 = ptr to systemcfg */ - ld r3,0(r3) - lwz r3,PLATFORM(r3) /* r3 = platform flags */ - andi. r3,r3,PLATFORM_LPAR /* Test if bit 0 is set (LPAR bit) */ - beq 98f /* branch if result is 0 */ - mfspr r3,SPRN_PVR - srwi r3,r3,16 - cmpwi r3,0x37 /* SStar */ - beq 97f - cmpwi r3,0x36 /* IStar */ - beq 97f - cmpwi r3,0x34 /* Pulsar */ - bne 98f -97: li r3,H_SET_ASR /* hcall = H_SET_ASR */ - HVSC /* Invoking hcall */ - b 99f -98: /* !(rpa hypervisor) || !(star) */ - mtasr r4 /* set the stab location */ -99: -#endif li r7,0 mtlr r7 @@ -1896,40 +1852,6 @@ _STATIC(start_here_multiplatform) mr r3,r31 bl .early_setup - /* set the ASR */ - ld r3,PACASTABREAL(r13) - ori r4,r3,1 /* turn on valid bit */ - ld r3,systemcfg@got(r2) /* r3 = ptr to systemcfg */ - ld r3,0(r3) - lwz r3,PLATFORM(r3) /* r3 = platform flags */ - andi. r3,r3,PLATFORM_LPAR /* Test if bit 0 is set (LPAR bit) */ - beq 98f /* branch if result is 0 */ - mfspr r3,SPRN_PVR - srwi r3,r3,16 - cmpwi r3,0x37 /* SStar */ - beq 97f - cmpwi r3,0x36 /* IStar */ - beq 97f - cmpwi r3,0x34 /* Pulsar */ - bne 98f -97: li r3,H_SET_ASR /* hcall = H_SET_ASR */ - HVSC /* Invoking hcall */ - b 99f -98: /* !(rpa hypervisor) || !(star) */ - mtasr r4 /* set the stab location */ -99: - /* Set SDR1 (hash table pointer) */ - ld r3,systemcfg@got(r2) /* r3 = ptr to systemcfg */ - ld r3,0(r3) - lwz r3,PLATFORM(r3) /* r3 = platform flags */ - /* Test if bit 0 is set (LPAR bit) */ - andi. r3,r3,PLATFORM_LPAR - bne 98f /* branch if result is !0 */ - LOADADDR(r6,_SDR1) /* Only if NOT LPAR */ - sub r6,r6,r26 - ld r6,0(r6) /* get the value of _SDR1 */ - mtspr SPRN_SDR1,r6 /* set the htab location */ -98: LOADADDR(r3,.start_here_common) SET_REG_TO_CONST(r4, MSR_KERNEL) mtspr SPRN_SRR0,r3 diff --git a/arch/ppc64/kernel/idle.c b/arch/ppc64/kernel/idle.c index 715bc0e71e0..b879d3057ef 100644 --- a/arch/ppc64/kernel/idle.c +++ b/arch/ppc64/kernel/idle.c @@ -26,7 +26,6 @@ #include <asm/processor.h> #include <asm/cputable.h> #include <asm/time.h> -#include <asm/systemcfg.h> #include <asm/machdep.h> #include <asm/smp.h> diff --git a/arch/ppc64/kernel/misc.S b/arch/ppc64/kernel/misc.S index 914632ec587..492bca6137e 100644 --- a/arch/ppc64/kernel/misc.S +++ b/arch/ppc64/kernel/misc.S @@ -78,12 +78,12 @@ _GLOBAL(call_do_softirq) mtlr r0 blr -_GLOBAL(call_handle_IRQ_event) +_GLOBAL(call___do_IRQ) mflr r0 std r0,16(r1) - stdu r1,THREAD_SIZE-112(r6) - mr r1,r6 - bl .handle_IRQ_event + stdu r1,THREAD_SIZE-112(r5) + mr r1,r5 + bl .__do_IRQ ld r1,0(r1) ld r0,16(r1) mtlr r0 diff --git a/arch/ppc64/kernel/nvram.c b/arch/ppc64/kernel/nvram.c index 4fb1a9f5060..c0fcd29918c 100644 --- a/arch/ppc64/kernel/nvram.c +++ b/arch/ppc64/kernel/nvram.c @@ -31,7 +31,6 @@ #include <asm/rtas.h> #include <asm/prom.h> #include <asm/machdep.h> -#include <asm/systemcfg.h> #undef DEBUG_NVRAM @@ -167,7 +166,7 @@ static int dev_nvram_ioctl(struct inode *inode, struct file *file, case IOC_NVRAM_GET_OFFSET: { int part, offset; - if (systemcfg->platform != PLATFORM_POWERMAC) + if (_machine != PLATFORM_POWERMAC) return -EINVAL; if (copy_from_user(&part, (void __user*)arg, sizeof(part)) != 0) return -EFAULT; @@ -450,7 +449,7 @@ static int nvram_setup_partition(void) * in our nvram, as Apple defined partitions use pretty much * all of the space */ - if (systemcfg->platform == PLATFORM_POWERMAC) + if (_machine == PLATFORM_POWERMAC) return -ENOSPC; /* see if we have an OS partition that meets our needs. diff --git a/arch/ppc64/kernel/pci.c b/arch/ppc64/kernel/pci.c index 30247ff7497..3cef1b8f57f 100644 --- a/arch/ppc64/kernel/pci.c +++ b/arch/ppc64/kernel/pci.c @@ -548,6 +548,11 @@ static int __init pcibios_init(void) if (ppc64_isabridge_dev != NULL) printk("ISA bridge at %s\n", pci_name(ppc64_isabridge_dev)); +#ifdef CONFIG_PPC_MULTIPLATFORM + /* map in PCI I/O space */ + phbs_remap_io(); +#endif + printk("PCI: Probing PCI hardware done\n"); return 0; @@ -1277,12 +1282,9 @@ long sys_pciconfig_iobase(long which, unsigned long in_bus, * G5 machines... So when something asks for bus 0 io base * (bus 0 is HT root), we return the AGP one instead. */ -#ifdef CONFIG_PPC_PMAC - if (systemcfg->platform == PLATFORM_POWERMAC && - machine_is_compatible("MacRISC4")) + if (machine_is_compatible("MacRISC4")) if (in_bus == 0) in_bus = 0xf0; -#endif /* CONFIG_PPC_PMAC */ /* That syscall isn't quite compatible with PCI domains, but it's * used on pre-domains setup. We return the first match diff --git a/arch/ppc64/kernel/pci_dn.c b/arch/ppc64/kernel/pci_dn.c index 1a443a7ada4..12c4c9e9bbc 100644 --- a/arch/ppc64/kernel/pci_dn.c +++ b/arch/ppc64/kernel/pci_dn.c @@ -43,7 +43,7 @@ static void * __devinit update_dn_pci_info(struct device_node *dn, void *data) u32 *regs; struct pci_dn *pdn; - if (phb->is_dynamic) + if (mem_init_done) pdn = kmalloc(sizeof(*pdn), GFP_KERNEL); else pdn = alloc_bootmem(sizeof(*pdn)); @@ -120,6 +120,14 @@ void *traverse_pci_devices(struct device_node *start, traverse_func pre, return NULL; } +/** + * pci_devs_phb_init_dynamic - setup pci devices under this PHB + * phb: pci-to-host bridge (top-level bridge connecting to cpu) + * + * This routine is called both during boot, (before the memory + * subsystem is set up, before kmalloc is valid) and during the + * dynamic lpar operation of adding a PHB to a running system. + */ void __devinit pci_devs_phb_init_dynamic(struct pci_controller *phb) { struct device_node * dn = (struct device_node *) phb->arch_data; @@ -201,9 +209,14 @@ static struct notifier_block pci_dn_reconfig_nb = { .notifier_call = pci_dn_reconfig_notifier, }; -/* - * Actually initialize the phbs. - * The buswalk on this phb has not happened yet. +/** + * pci_devs_phb_init - Initialize phbs and pci devs under them. + * + * This routine walks over all phb's (pci-host bridges) on the + * system, and sets up assorted pci-related structures + * (including pci info in the device node structs) for each + * pci device found underneath. This routine runs once, + * early in the boot sequence. */ void __init pci_devs_phb_init(void) { diff --git a/arch/ppc64/kernel/prom.c b/arch/ppc64/kernel/prom.c index 3402fbee62c..fbad2c36078 100644 --- a/arch/ppc64/kernel/prom.c +++ b/arch/ppc64/kernel/prom.c @@ -318,7 +318,7 @@ static int __devinit finish_node_interrupts(struct device_node *np, } /* We offset irq numbers for the u3 MPIC by 128 in PowerMac */ - if (systemcfg->platform == PLATFORM_POWERMAC && ic && ic->parent) { + if (_machine == PLATFORM_POWERMAC && ic && ic->parent) { char *name = get_property(ic->parent, "name", NULL); if (name && !strcmp(name, "u3")) np->intrs[intrcount].line += 128; @@ -1065,7 +1065,7 @@ static int __init early_init_dt_scan_chosen(unsigned long node, prop = (u32 *)of_get_flat_dt_prop(node, "linux,platform", NULL); if (prop == NULL) return 0; - systemcfg->platform = *prop; + _machine = *prop; /* check if iommu is forced on or off */ if (of_get_flat_dt_prop(node, "linux,iommu-off", NULL) != NULL) @@ -1230,11 +1230,8 @@ void __init early_init_devtree(void *params) of_scan_flat_dt(early_init_dt_scan_memory, NULL); lmb_enforce_memory_limit(memory_limit); lmb_analyze(); - systemcfg->physicalMemorySize = lmb_phys_mem_size(); lmb_reserve(0, __pa(klimit)); - DBG("Phys. mem: %lx\n", systemcfg->physicalMemorySize); - /* Reserve LMB regions used by kernel, initrd, dt, etc... */ early_reserve_mem(); @@ -1753,7 +1750,7 @@ static int of_finish_dynamic_node(struct device_node *node, /* We don't support that function on PowerMac, at least * not yet */ - if (systemcfg->platform == PLATFORM_POWERMAC) + if (_machine == PLATFORM_POWERMAC) return -ENODEV; /* fix up new node's linux_phandle field */ diff --git a/arch/ppc64/kernel/prom_init.c b/arch/ppc64/kernel/prom_init.c index e4c880dab99..6375f40b23d 100644 --- a/arch/ppc64/kernel/prom_init.c +++ b/arch/ppc64/kernel/prom_init.c @@ -1934,7 +1934,8 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4, unsigned long /* * On pSeries, inform the firmware about our capabilities */ - if (RELOC(of_platform) & PLATFORM_PSERIES) + if (RELOC(of_platform) == PLATFORM_PSERIES || + RELOC(of_platform) == PLATFORM_PSERIES_LPAR) prom_send_capabilities(); /* diff --git a/arch/ppc64/kernel/vdso.c b/arch/ppc64/kernel/vdso.c index 4aacf521e3e..1bbacac4498 100644 --- a/arch/ppc64/kernel/vdso.c +++ b/arch/ppc64/kernel/vdso.c @@ -34,6 +34,7 @@ #include <asm/machdep.h> #include <asm/cputable.h> #include <asm/sections.h> +#include <asm/systemcfg.h> #include <asm/vdso.h> #undef DEBUG @@ -179,7 +180,7 @@ static struct page * vdso_vma_nopage(struct vm_area_struct * vma, * Last page is systemcfg. */ if ((vma->vm_end - address) <= PAGE_SIZE) - pg = virt_to_page(systemcfg); + pg = virt_to_page(_systemcfg); else pg = virt_to_page(vbase + offset); @@ -604,7 +605,7 @@ void __init vdso_init(void) get_page(pg); } - get_page(virt_to_page(systemcfg)); + get_page(virt_to_page(_systemcfg)); } int in_gate_area_no_task(unsigned long addr) diff --git a/arch/sparc/kernel/cpu.c b/arch/sparc/kernel/cpu.c index 6a4ebc62193..d7bfc61d287 100644 --- a/arch/sparc/kernel/cpu.c +++ b/arch/sparc/kernel/cpu.c @@ -75,7 +75,7 @@ struct cpu_fp_info linux_sparc_fpu[] = { { 9, 3, "Fujitsu or Weitek on-chip FPU"}, }; -#define NSPARCFPU (sizeof(linux_sparc_fpu)/sizeof(struct cpu_fp_info)) +#define NSPARCFPU ARRAY_SIZE(linux_sparc_fpu) struct cpu_iu_info linux_sparc_chips[] = { /* Sun4/100, 4/200, SLC */ @@ -120,7 +120,7 @@ struct cpu_iu_info linux_sparc_chips[] = { { 0xf, 0, "UNKNOWN CPU-VENDOR/TYPE"}, }; -#define NSPARCCHIPS (sizeof(linux_sparc_chips)/sizeof(struct cpu_iu_info)) +#define NSPARCCHIPS ARRAY_SIZE(linux_sparc_chips) char *sparc_cpu_type; char *sparc_fpu_type; diff --git a/arch/sparc/kernel/pcic.c b/arch/sparc/kernel/pcic.c index 25e31d5ec99..cccfc12802e 100644 --- a/arch/sparc/kernel/pcic.c +++ b/arch/sparc/kernel/pcic.c @@ -143,7 +143,7 @@ static struct pcic_ca2irq pcic_i_jk[] = { * as several PROMs may be installed on the same physical board. */ #define SN2L_INIT(name, map) \ - { name, map, sizeof(map)/sizeof(struct pcic_ca2irq) } + { name, map, ARRAY_SIZE(map) } static struct pcic_sn2list pcic_known_sysnames[] = { SN2L_INIT("SUNW,JavaEngine1", pcic_i_je1a), /* JE1, PROM 2.32 */ diff --git a/arch/sparc/mm/fault.c b/arch/sparc/mm/fault.c index 2bbd53f3caf..9eeed3347df 100644 --- a/arch/sparc/mm/fault.c +++ b/arch/sparc/mm/fault.c @@ -33,8 +33,6 @@ #include <asm/kdebug.h> #include <asm/uaccess.h> -#define ELEMENTS(arr) (sizeof (arr)/sizeof (arr[0])) - extern int prom_node_root; /* At boot time we determine these two values necessary for setting diff --git a/arch/sparc64/kernel/cpu.c b/arch/sparc64/kernel/cpu.c index 77ef5df4e5a..00eed88ef2e 100644 --- a/arch/sparc64/kernel/cpu.c +++ b/arch/sparc64/kernel/cpu.c @@ -43,7 +43,7 @@ struct cpu_fp_info linux_sparc_fpu[] = { { 0x3e, 0x22, 0, "UltraSparc IIIi+ integrated FPU"}, }; -#define NSPARCFPU (sizeof(linux_sparc_fpu)/sizeof(struct cpu_fp_info)) +#define NSPARCFPU ARRAY_SIZE(linux_sparc_fpu) struct cpu_iu_info linux_sparc_chips[] = { { 0x17, 0x10, "TI UltraSparc I (SpitFire)"}, @@ -59,7 +59,7 @@ struct cpu_iu_info linux_sparc_chips[] = { { 0x3e, 0x22, "TI UltraSparc IIIi+ (Serrano)"}, }; -#define NSPARCCHIPS (sizeof(linux_sparc_chips)/sizeof(struct cpu_iu_info)) +#define NSPARCCHIPS ARRAY_SIZE(linux_sparc_chips) char *sparc_cpu_type = "cpu-oops"; char *sparc_fpu_type = "fpu-oops"; diff --git a/arch/sparc64/kernel/ioctl32.c b/arch/sparc64/kernel/ioctl32.c index c2e96daa5ab..e62214354bb 100644 --- a/arch/sparc64/kernel/ioctl32.c +++ b/arch/sparc64/kernel/ioctl32.c @@ -114,8 +114,6 @@ COMPATIBLE_IOCTL(FBIOGCURPOS) COMPATIBLE_IOCTL(FBIOGCURMAX) /* Little k */ /* Little v, the video4linux ioctls */ -COMPATIBLE_IOCTL(_IOR('p', 20, int[7])) /* RTCGET */ -COMPATIBLE_IOCTL(_IOW('p', 21, int[7])) /* RTCSET */ /* And these ioctls need translation */ /* Note SIOCRTMSG is no longer, so this is safe and * the user would have seen just an -EINVAL anyways. */ HANDLE_IOCTL(FBIOPUTCMAP32, fbiogetputcmap) diff --git a/arch/sparc64/kernel/sbus.c b/arch/sparc64/kernel/sbus.c index e09ddf92765..96b82505566 100644 --- a/arch/sparc64/kernel/sbus.c +++ b/arch/sparc64/kernel/sbus.c @@ -790,7 +790,7 @@ static unsigned long sysio_irq_offsets[] = { #undef bogon -#define NUM_SYSIO_OFFSETS (sizeof(sysio_irq_offsets) / sizeof(sysio_irq_offsets[0])) +#define NUM_SYSIO_OFFSETS ARRAY_SIZE(sysio_irq_offsets) /* Convert Interrupt Mapping register pointer to associated * Interrupt Clear register pointer, SYSIO specific version. diff --git a/arch/sparc64/mm/fault.c b/arch/sparc64/mm/fault.c index 3be278d916d..6f0539aa44d 100644 --- a/arch/sparc64/mm/fault.c +++ b/arch/sparc64/mm/fault.c @@ -30,8 +30,6 @@ #include <asm/sections.h> #include <asm/kdebug.h> -#define ELEMENTS(arr) (sizeof (arr)/sizeof (arr[0])) - /* * To debug kernel to catch accesses to certain virtual/physical addresses. * Mode = 0 selects physical watchpoints, mode = 1 selects virtual watchpoints. diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c index 3937adf4e5e..aa993715d64 100644 --- a/drivers/acpi/glue.c +++ b/drivers/acpi/glue.c @@ -203,6 +203,7 @@ acpi_handle acpi_get_pci_rootbridge_handle(unsigned int seg, unsigned int bus) acpi_get_devices(PCI_ROOT_HID_STRING, find_pci_rootbridge, &find, NULL); return find.handle; } +EXPORT_SYMBOL_GPL(acpi_get_pci_rootbridge_handle); /* Get device's handler per its address under its parent */ struct acpi_find_child { diff --git a/drivers/atm/horizon.c b/drivers/atm/horizon.c index 0cded046800..821c81e8cd3 100644 --- a/drivers/atm/horizon.c +++ b/drivers/atm/horizon.c @@ -1511,8 +1511,8 @@ static inline short setup_idle_tx_channel (hrz_dev * dev, hrz_vcc * vcc) { // a.k.a. prepare the channel and remember that we have done so. tx_ch_desc * tx_desc = &memmap->tx_descs[tx_channel]; - u16 rd_ptr; - u16 wr_ptr; + u32 rd_ptr; + u32 wr_ptr; u16 channel = vcc->channel; unsigned long flags; diff --git a/drivers/char/agp/ali-agp.c b/drivers/char/agp/ali-agp.c index ba54b587257..b02fc226715 100644 --- a/drivers/char/agp/ali-agp.c +++ b/drivers/char/agp/ali-agp.c @@ -389,7 +389,6 @@ static struct pci_device_id agp_ali_pci_table[] = { MODULE_DEVICE_TABLE(pci, agp_ali_pci_table); static struct pci_driver agp_ali_pci_driver = { - .owner = THIS_MODULE, .name = "agpgart-ali", .id_table = agp_ali_pci_table, .probe = agp_ali_probe, diff --git a/drivers/char/agp/amd-k7-agp.c b/drivers/char/agp/amd-k7-agp.c index 40fcd88b2ce..1f776651ac6 100644 --- a/drivers/char/agp/amd-k7-agp.c +++ b/drivers/char/agp/amd-k7-agp.c @@ -515,7 +515,6 @@ static struct pci_device_id agp_amdk7_pci_table[] = { MODULE_DEVICE_TABLE(pci, agp_amdk7_pci_table); static struct pci_driver agp_amdk7_pci_driver = { - .owner = THIS_MODULE, .name = "agpgart-amdk7", .id_table = agp_amdk7_pci_table, .probe = agp_amdk7_probe, diff --git a/drivers/char/agp/amd64-agp.c b/drivers/char/agp/amd64-agp.c index 8f748fddca9..78ce98a69f3 100644 --- a/drivers/char/agp/amd64-agp.c +++ b/drivers/char/agp/amd64-agp.c @@ -703,7 +703,6 @@ static struct pci_device_id agp_amd64_pci_table[] = { MODULE_DEVICE_TABLE(pci, agp_amd64_pci_table); static struct pci_driver agp_amd64_pci_driver = { - .owner = THIS_MODULE, .name = "agpgart-amd64", .id_table = agp_amd64_pci_table, .probe = agp_amd64_probe, diff --git a/drivers/char/agp/ati-agp.c b/drivers/char/agp/ati-agp.c index fbd41556546..53372a83b67 100644 --- a/drivers/char/agp/ati-agp.c +++ b/drivers/char/agp/ati-agp.c @@ -521,7 +521,6 @@ static struct pci_device_id agp_ati_pci_table[] = { MODULE_DEVICE_TABLE(pci, agp_ati_pci_table); static struct pci_driver agp_ati_pci_driver = { - .owner = THIS_MODULE, .name = "agpgart-ati", .id_table = agp_ati_pci_table, .probe = agp_ati_probe, diff --git a/drivers/char/agp/backend.c b/drivers/char/agp/backend.c index 73f333f491b..27bca34b4a6 100644 --- a/drivers/char/agp/backend.c +++ b/drivers/char/agp/backend.c @@ -147,6 +147,7 @@ static int agp_backend_initialize(struct agp_bridge_data *bridge) printk(KERN_ERR PFX "unable to get memory for scratch page.\n"); return -ENOMEM; } + flush_agp_mappings(); bridge->scratch_page_real = virt_to_gart(addr); bridge->scratch_page = @@ -187,9 +188,11 @@ static int agp_backend_initialize(struct agp_bridge_data *bridge) return 0; err_out: - if (bridge->driver->needs_scratch_page) + if (bridge->driver->needs_scratch_page) { bridge->driver->agp_destroy_page( gart_to_virt(bridge->scratch_page_real)); + flush_agp_mappings(); + } if (got_gatt) bridge->driver->free_gatt_table(bridge); if (got_keylist) { @@ -211,9 +214,11 @@ static void agp_backend_cleanup(struct agp_bridge_data *bridge) bridge->key_list = NULL; if (bridge->driver->agp_destroy_page && - bridge->driver->needs_scratch_page) + bridge->driver->needs_scratch_page) { bridge->driver->agp_destroy_page( gart_to_virt(bridge->scratch_page_real)); + flush_agp_mappings(); + } } /* When we remove the global variable agp_bridge from all drivers diff --git a/drivers/char/agp/efficeon-agp.c b/drivers/char/agp/efficeon-agp.c index d41e0a62e32..e7aea77a60f 100644 --- a/drivers/char/agp/efficeon-agp.c +++ b/drivers/char/agp/efficeon-agp.c @@ -429,7 +429,6 @@ static struct pci_device_id agp_efficeon_pci_table[] = { MODULE_DEVICE_TABLE(pci, agp_efficeon_pci_table); static struct pci_driver agp_efficeon_pci_driver = { - .owner = THIS_MODULE, .name = "agpgart-efficeon", .id_table = agp_efficeon_pci_table, .probe = agp_efficeon_probe, diff --git a/drivers/char/agp/generic.c b/drivers/char/agp/generic.c index c4a38715c6f..5567ce8d72b 100644 --- a/drivers/char/agp/generic.c +++ b/drivers/char/agp/generic.c @@ -57,7 +57,8 @@ int map_page_into_agp(struct page *page) { int i; i = change_page_attr(page, 1, PAGE_KERNEL_NOCACHE); - global_flush_tlb(); + /* Caller's responsibility to call global_flush_tlb() for + * performance reasons */ return i; } EXPORT_SYMBOL_GPL(map_page_into_agp); @@ -66,7 +67,8 @@ int unmap_page_from_agp(struct page *page) { int i; i = change_page_attr(page, 1, PAGE_KERNEL); - global_flush_tlb(); + /* Caller's responsibility to call global_flush_tlb() for + * performance reasons */ return i; } EXPORT_SYMBOL_GPL(unmap_page_from_agp); @@ -153,6 +155,7 @@ void agp_free_memory(struct agp_memory *curr) for (i = 0; i < curr->page_count; i++) { curr->bridge->driver->agp_destroy_page(gart_to_virt(curr->memory[i])); } + flush_agp_mappings(); } agp_free_key(curr->key); vfree(curr->memory); @@ -210,7 +213,7 @@ struct agp_memory *agp_allocate_memory(struct agp_bridge_data *bridge, new->memory[i] = virt_to_gart(addr); new->page_count++; } - new->bridge = bridge; + new->bridge = bridge; flush_agp_mappings(); diff --git a/drivers/char/agp/i460-agp.c b/drivers/char/agp/i460-agp.c index 58944cd271e..8ee19a4a6bc 100644 --- a/drivers/char/agp/i460-agp.c +++ b/drivers/char/agp/i460-agp.c @@ -111,8 +111,10 @@ static int i460_fetch_size (void) if (i460.io_page_shift != I460_IO_PAGE_SHIFT) { printk(KERN_ERR PFX - "I/O (GART) page-size %ZuKB doesn't match expected size %ZuKB\n", - 1UL << (i460.io_page_shift - 10), 1UL << (I460_IO_PAGE_SHIFT)); + "I/O (GART) page-size %luKB doesn't match expected " + "size %luKB\n", + 1UL << (i460.io_page_shift - 10), + 1UL << (I460_IO_PAGE_SHIFT)); return 0; } @@ -514,9 +516,10 @@ static void *i460_alloc_page (struct agp_bridge_data *bridge) { void *page; - if (I460_IO_PAGE_SHIFT <= PAGE_SHIFT) + if (I460_IO_PAGE_SHIFT <= PAGE_SHIFT) { page = agp_generic_alloc_page(agp_bridge); - else + global_flush_tlb(); + } else /* Returning NULL would cause problems */ /* AK: really dubious code. */ page = (void *)~0UL; @@ -525,8 +528,10 @@ static void *i460_alloc_page (struct agp_bridge_data *bridge) static void i460_destroy_page (void *page) { - if (I460_IO_PAGE_SHIFT <= PAGE_SHIFT) + if (I460_IO_PAGE_SHIFT <= PAGE_SHIFT) { agp_generic_destroy_page(page); + global_flush_tlb(); + } } #endif /* I460_LARGE_IO_PAGES */ @@ -536,7 +541,7 @@ static unsigned long i460_mask_memory (struct agp_bridge_data *bridge, { /* Make sure the returned address is a valid GATT entry */ return bridge->driver->masks[0].mask - | (((addr & ~((1 << I460_IO_PAGE_SHIFT) - 1)) & 0xffffff000) >> 12); + | (((addr & ~((1 << I460_IO_PAGE_SHIFT) - 1)) & 0xfffff000) >> 12); } struct agp_bridge_driver intel_i460_driver = { @@ -617,7 +622,6 @@ static struct pci_device_id agp_intel_i460_pci_table[] = { MODULE_DEVICE_TABLE(pci, agp_intel_i460_pci_table); static struct pci_driver agp_intel_i460_pci_driver = { - .owner = THIS_MODULE, .name = "agpgart-intel-i460", .id_table = agp_intel_i460_pci_table, .probe = agp_intel_i460_probe, diff --git a/drivers/char/agp/intel-agp.c b/drivers/char/agp/intel-agp.c index bf4cc9ffd5b..e7bed5047dc 100644 --- a/drivers/char/agp/intel-agp.c +++ b/drivers/char/agp/intel-agp.c @@ -270,6 +270,7 @@ static struct agp_memory *alloc_agpphysmem_i8xx(size_t pg_count, int type) switch (pg_count) { case 1: addr = agp_bridge->driver->agp_alloc_page(agp_bridge); + global_flush_tlb(); break; case 4: /* kludge to get 4 physical pages for ARGB cursor */ @@ -330,9 +331,11 @@ static void intel_i810_free_by_type(struct agp_memory *curr) if(curr->type == AGP_PHYS_MEMORY) { if (curr->page_count == 4) i8xx_destroy_pages(gart_to_virt(curr->memory[0])); - else + else { agp_bridge->driver->agp_destroy_page( gart_to_virt(curr->memory[0])); + global_flush_tlb(); + } vfree(curr->memory); } kfree(curr); @@ -1824,7 +1827,6 @@ static struct pci_device_id agp_intel_pci_table[] = { MODULE_DEVICE_TABLE(pci, agp_intel_pci_table); static struct pci_driver agp_intel_pci_driver = { - .owner = THIS_MODULE, .name = "agpgart-intel", .id_table = agp_intel_pci_table, .probe = agp_intel_probe, diff --git a/drivers/char/agp/nvidia-agp.c b/drivers/char/agp/nvidia-agp.c index 3aed0c5e2f9..80dafa3030b 100644 --- a/drivers/char/agp/nvidia-agp.c +++ b/drivers/char/agp/nvidia-agp.c @@ -398,7 +398,6 @@ static struct pci_device_id agp_nvidia_pci_table[] = { MODULE_DEVICE_TABLE(pci, agp_nvidia_pci_table); static struct pci_driver agp_nvidia_pci_driver = { - .owner = THIS_MODULE, .name = "agpgart-nvidia", .id_table = agp_nvidia_pci_table, .probe = agp_nvidia_probe, diff --git a/drivers/char/agp/sis-agp.c b/drivers/char/agp/sis-agp.c index a701361a889..ebc05554045 100644 --- a/drivers/char/agp/sis-agp.c +++ b/drivers/char/agp/sis-agp.c @@ -332,7 +332,6 @@ static struct pci_device_id agp_sis_pci_table[] = { MODULE_DEVICE_TABLE(pci, agp_sis_pci_table); static struct pci_driver agp_sis_pci_driver = { - .owner = THIS_MODULE, .name = "agpgart-sis", .id_table = agp_sis_pci_table, .probe = agp_sis_probe, diff --git a/drivers/char/agp/sworks-agp.c b/drivers/char/agp/sworks-agp.c index 5a5392dd125..3f8f7fa6b0f 100644 --- a/drivers/char/agp/sworks-agp.c +++ b/drivers/char/agp/sworks-agp.c @@ -545,7 +545,6 @@ static struct pci_device_id agp_serverworks_pci_table[] = { MODULE_DEVICE_TABLE(pci, agp_serverworks_pci_table); static struct pci_driver agp_serverworks_pci_driver = { - .owner = THIS_MODULE, .name = "agpgart-serverworks", .id_table = agp_serverworks_pci_table, .probe = agp_serverworks_probe, diff --git a/drivers/char/agp/uninorth-agp.c b/drivers/char/agp/uninorth-agp.c index 183c50acab2..c8255312b8c 100644 --- a/drivers/char/agp/uninorth-agp.c +++ b/drivers/char/agp/uninorth-agp.c @@ -658,7 +658,6 @@ static struct pci_device_id agp_uninorth_pci_table[] = { MODULE_DEVICE_TABLE(pci, agp_uninorth_pci_table); static struct pci_driver agp_uninorth_pci_driver = { - .owner = THIS_MODULE, .name = "agpgart-uninorth", .id_table = agp_uninorth_pci_table, .probe = agp_uninorth_probe, diff --git a/drivers/char/agp/via-agp.c b/drivers/char/agp/via-agp.c index 5d9a1370007..c847df575cf 100644 --- a/drivers/char/agp/via-agp.c +++ b/drivers/char/agp/via-agp.c @@ -518,7 +518,6 @@ MODULE_DEVICE_TABLE(pci, agp_via_pci_table); static struct pci_driver agp_via_pci_driver = { - .owner = THIS_MODULE, .name = "agpgart-via", .id_table = agp_via_pci_table, .probe = agp_via_probe, diff --git a/drivers/char/epca.c b/drivers/char/epca.c index b7a0e4d6b93..407708a001e 100644 --- a/drivers/char/epca.c +++ b/drivers/char/epca.c @@ -3113,7 +3113,6 @@ MODULE_DEVICE_TABLE(pci, epca_pci_tbl); int __init init_PCI (void) { /* Begin init_PCI */ memset (&epca_driver, 0, sizeof (epca_driver)); - epca_driver.owner = THIS_MODULE; epca_driver.name = "epca"; epca_driver.id_table = epca_pci_tbl; epca_driver.probe = epca_init_one; diff --git a/drivers/char/synclink.c b/drivers/char/synclink.c index 5d1ffa3bd4c..82c6abde68d 100644 --- a/drivers/char/synclink.c +++ b/drivers/char/synclink.c @@ -912,7 +912,6 @@ MODULE_DEVICE_TABLE(pci, synclink_pci_tbl); MODULE_LICENSE("GPL"); static struct pci_driver synclink_pci_driver = { - .owner = THIS_MODULE, .name = "synclink", .id_table = synclink_pci_tbl, .probe = synclink_init_one, diff --git a/drivers/char/synclinkmp.c b/drivers/char/synclinkmp.c index 7c063c5abc5..ee5a40be9f9 100644 --- a/drivers/char/synclinkmp.c +++ b/drivers/char/synclinkmp.c @@ -500,7 +500,6 @@ MODULE_DEVICE_TABLE(pci, synclinkmp_pci_tbl); MODULE_LICENSE("GPL"); static struct pci_driver synclinkmp_pci_driver = { - .owner = THIS_MODULE, .name = "synclinkmp", .id_table = synclinkmp_pci_tbl, .probe = synclinkmp_init_one, diff --git a/drivers/char/watchdog/pcwd_pci.c b/drivers/char/watchdog/pcwd_pci.c index d9ef55bdf88..2451edbefec 100644 --- a/drivers/char/watchdog/pcwd_pci.c +++ b/drivers/char/watchdog/pcwd_pci.c @@ -755,7 +755,6 @@ static struct pci_device_id pcipcwd_pci_tbl[] = { MODULE_DEVICE_TABLE(pci, pcipcwd_pci_tbl); static struct pci_driver pcipcwd_driver = { - .owner = THIS_MODULE, .name = WATCHDOG_NAME, .id_table = pcipcwd_pci_tbl, .probe = pcipcwd_card_init, diff --git a/drivers/char/watchdog/wdt_pci.c b/drivers/char/watchdog/wdt_pci.c index dc9370f6c34..4b3311993d4 100644 --- a/drivers/char/watchdog/wdt_pci.c +++ b/drivers/char/watchdog/wdt_pci.c @@ -711,7 +711,6 @@ MODULE_DEVICE_TABLE(pci, wdtpci_pci_tbl); static struct pci_driver wdtpci_driver = { - .owner = THIS_MODULE, .name = "wdt_pci", .id_table = wdtpci_pci_tbl, .probe = wdtpci_init_one, diff --git a/drivers/hwmon/hdaps.c b/drivers/hwmon/hdaps.c index 1e5dfc7805e..c8c84e0819f 100644 --- a/drivers/hwmon/hdaps.c +++ b/drivers/hwmon/hdaps.c @@ -60,9 +60,11 @@ #define HDAPS_POLL_PERIOD (HZ/20) /* poll for input every 1/20s */ #define HDAPS_INPUT_FUZZ 4 /* input event threshold */ +#define HDAPS_INPUT_FLAT 4 static struct timer_list hdaps_timer; static struct platform_device *pdev; +static struct input_dev *hdaps_idev; static unsigned int hdaps_invert; static u8 km_activity; static int rest_x; @@ -309,18 +311,6 @@ static struct device_driver hdaps_driver = { .resume = hdaps_resume }; -/* Input class stuff */ - -static struct input_dev hdaps_idev = { - .name = "hdaps", - .evbit = { BIT(EV_ABS) }, - .absbit = { BIT(ABS_X) | BIT(ABS_Y) }, - .absmin = { [ABS_X] = -256, [ABS_Y] = -256 }, - .absmax = { [ABS_X] = 256, [ABS_Y] = 256 }, - .absfuzz = { [ABS_X] = HDAPS_INPUT_FUZZ, [ABS_Y] = HDAPS_INPUT_FUZZ }, - .absflat = { [ABS_X] = HDAPS_INPUT_FUZZ, [ABS_Y] = HDAPS_INPUT_FUZZ }, -}; - /* * hdaps_calibrate - Set our "resting" values. Callers must hold hdaps_sem. */ @@ -342,9 +332,9 @@ static void hdaps_mousedev_poll(unsigned long unused) if (__hdaps_read_pair(HDAPS_PORT_XPOS, HDAPS_PORT_YPOS, &x, &y)) goto out; - input_report_abs(&hdaps_idev, ABS_X, x - rest_x); - input_report_abs(&hdaps_idev, ABS_Y, y - rest_y); - input_sync(&hdaps_idev); + input_report_abs(hdaps_idev, ABS_X, x - rest_x); + input_report_abs(hdaps_idev, ABS_Y, y - rest_y); + input_sync(hdaps_idev); mod_timer(&hdaps_timer, jiffies + HDAPS_POLL_PERIOD); @@ -564,12 +554,25 @@ static int __init hdaps_init(void) if (ret) goto out_device; + hdaps_idev = input_allocate_device(); + if (!hdaps_idev) { + ret = -ENOMEM; + goto out_group; + } + /* initial calibrate for the input device */ hdaps_calibrate(); /* initialize the input class */ - hdaps_idev.dev = &pdev->dev; - input_register_device(&hdaps_idev); + hdaps_idev->name = "hdaps"; + hdaps_idev->cdev.dev = &pdev->dev; + hdaps_idev->evbit[0] = BIT(EV_ABS); + input_set_abs_params(hdaps_idev, ABS_X, + -256, 256, HDAPS_INPUT_FUZZ, HDAPS_INPUT_FLAT); + input_set_abs_params(hdaps_idev, ABS_X, + -256, 256, HDAPS_INPUT_FUZZ, HDAPS_INPUT_FLAT); + + input_register_device(hdaps_idev); /* start up our timer for the input device */ init_timer(&hdaps_timer); @@ -580,6 +583,8 @@ static int __init hdaps_init(void) printk(KERN_INFO "hdaps: driver successfully loaded.\n"); return 0; +out_group: + sysfs_remove_group(&pdev->dev.kobj, &hdaps_attribute_group); out_device: platform_device_unregister(pdev); out_driver: @@ -594,7 +599,7 @@ out: static void __exit hdaps_exit(void) { del_timer_sync(&hdaps_timer); - input_unregister_device(&hdaps_idev); + input_unregister_device(hdaps_idev); sysfs_remove_group(&pdev->dev.kobj, &hdaps_attribute_group); platform_device_unregister(pdev); driver_unregister(&hdaps_driver); diff --git a/drivers/hwmon/w83627hf.c b/drivers/hwmon/w83627hf.c index 70ef926c3bd..4e9a04e1f08 100644 --- a/drivers/hwmon/w83627hf.c +++ b/drivers/hwmon/w83627hf.c @@ -180,11 +180,10 @@ superio_exit(void) #define W83781D_REG_BANK 0x4E #define W83781D_REG_CONFIG 0x40 -#define W83781D_REG_ALARM1 0x41 -#define W83781D_REG_ALARM2 0x42 -#define W83781D_REG_ALARM3 0x450 +#define W83781D_REG_ALARM1 0x459 +#define W83781D_REG_ALARM2 0x45A +#define W83781D_REG_ALARM3 0x45B -#define W83781D_REG_IRQ 0x4C #define W83781D_REG_BEEP_CONFIG 0x4D #define W83781D_REG_BEEP_INTS1 0x56 #define W83781D_REG_BEEP_INTS2 0x57 @@ -1370,13 +1369,6 @@ static void w83627hf_init_client(struct i2c_client *client) W83781D_REG_TEMP3_CONFIG, tmp & 0xfe); } } - - /* enable comparator mode for temp2 and temp3 so - alarm indication will work correctly */ - i = w83627hf_read_value(client, W83781D_REG_IRQ); - if (!(i & 0x40)) - w83627hf_write_value(client, W83781D_REG_IRQ, - i | 0x40); } /* Start monitoring */ @@ -1400,7 +1392,7 @@ static struct w83627hf_data *w83627hf_update_device(struct device *dev) /* skip missing sensors */ if (((data->type == w83697hf) && (i == 1)) || ((data->type == w83627thf || data->type == w83637hf) - && (i == 4 || i == 5))) + && (i == 5 || i == 6))) continue; data->in[i] = w83627hf_read_value(client, W83781D_REG_IN(i)); diff --git a/drivers/i2c/busses/i2c-ali1535.c b/drivers/i2c/busses/i2c-ali1535.c index ba90f5140af..3eb47890db4 100644 --- a/drivers/i2c/busses/i2c-ali1535.c +++ b/drivers/i2c/busses/i2c-ali1535.c @@ -513,7 +513,6 @@ static void __devexit ali1535_remove(struct pci_dev *dev) } static struct pci_driver ali1535_driver = { - .owner = THIS_MODULE, .name = "ali1535_smbus", .id_table = ali1535_ids, .probe = ali1535_probe, diff --git a/drivers/i2c/busses/i2c-ali1563.c b/drivers/i2c/busses/i2c-ali1563.c index f1a62d89242..e6f63208fc4 100644 --- a/drivers/i2c/busses/i2c-ali1563.c +++ b/drivers/i2c/busses/i2c-ali1563.c @@ -408,7 +408,6 @@ static struct pci_device_id __devinitdata ali1563_id_table[] = { MODULE_DEVICE_TABLE (pci, ali1563_id_table); static struct pci_driver ali1563_pci_driver = { - .owner = THIS_MODULE, .name = "ali1563_smbus", .id_table = ali1563_id_table, .probe = ali1563_probe, diff --git a/drivers/i2c/busses/i2c-ali15x3.c b/drivers/i2c/busses/i2c-ali15x3.c index 400b08ed429..7a5c0941dbc 100644 --- a/drivers/i2c/busses/i2c-ali15x3.c +++ b/drivers/i2c/busses/i2c-ali15x3.c @@ -504,7 +504,6 @@ static void __devexit ali15x3_remove(struct pci_dev *dev) } static struct pci_driver ali15x3_driver = { - .owner = THIS_MODULE, .name = "ali15x3_smbus", .id_table = ali15x3_ids, .probe = ali15x3_probe, diff --git a/drivers/i2c/busses/i2c-amd756.c b/drivers/i2c/busses/i2c-amd756.c index de035d137c3..1750dedaf4b 100644 --- a/drivers/i2c/busses/i2c-amd756.c +++ b/drivers/i2c/busses/i2c-amd756.c @@ -401,7 +401,6 @@ static void __devexit amd756_remove(struct pci_dev *dev) } static struct pci_driver amd756_driver = { - .owner = THIS_MODULE, .name = "amd756_smbus", .id_table = amd756_ids, .probe = amd756_probe, diff --git a/drivers/i2c/busses/i2c-amd8111.c b/drivers/i2c/busses/i2c-amd8111.c index f3b79a68dbe..e5ef560e686 100644 --- a/drivers/i2c/busses/i2c-amd8111.c +++ b/drivers/i2c/busses/i2c-amd8111.c @@ -384,7 +384,6 @@ static void __devexit amd8111_remove(struct pci_dev *dev) } static struct pci_driver amd8111_driver = { - .owner = THIS_MODULE, .name = "amd8111_smbus2", .id_table = amd8111_ids, .probe = amd8111_probe, diff --git a/drivers/i2c/busses/i2c-hydra.c b/drivers/i2c/busses/i2c-hydra.c index 1b5354e24bf..e0cb3b0f92f 100644 --- a/drivers/i2c/busses/i2c-hydra.c +++ b/drivers/i2c/busses/i2c-hydra.c @@ -155,7 +155,6 @@ static void __devexit hydra_remove(struct pci_dev *dev) static struct pci_driver hydra_driver = { - .owner = THIS_MODULE, .name = "hydra_smbus", .id_table = hydra_ids, .probe = hydra_probe, diff --git a/drivers/i2c/busses/i2c-i801.c b/drivers/i2c/busses/i2c-i801.c index 4f63195069d..ac3eafa8aac 100644 --- a/drivers/i2c/busses/i2c-i801.c +++ b/drivers/i2c/busses/i2c-i801.c @@ -560,7 +560,6 @@ static void __devexit i801_remove(struct pci_dev *dev) } static struct pci_driver i801_driver = { - .owner = THIS_MODULE, .name = "i801_smbus", .id_table = i801_ids, .probe = i801_probe, diff --git a/drivers/i2c/busses/i2c-i810.c b/drivers/i2c/busses/i2c-i810.c index 52bc30593bd..748be30f2ba 100644 --- a/drivers/i2c/busses/i2c-i810.c +++ b/drivers/i2c/busses/i2c-i810.c @@ -233,7 +233,6 @@ static void __devexit i810_remove(struct pci_dev *dev) } static struct pci_driver i810_driver = { - .owner = THIS_MODULE, .name = "i810_smbus", .id_table = i810_ids, .probe = i810_probe, diff --git a/drivers/i2c/busses/i2c-nforce2.c b/drivers/i2c/busses/i2c-nforce2.c index fd26036e68a..4d18e6e5f15 100644 --- a/drivers/i2c/busses/i2c-nforce2.c +++ b/drivers/i2c/busses/i2c-nforce2.c @@ -347,7 +347,6 @@ static void __devexit nforce2_remove(struct pci_dev *dev) } static struct pci_driver nforce2_driver = { - .owner = THIS_MODULE, .name = "nForce2_smbus", .id_table = nforce2_ids, .probe = nforce2_probe, diff --git a/drivers/i2c/busses/i2c-piix4.c b/drivers/i2c/busses/i2c-piix4.c index 7d63eec423f..692f4734548 100644 --- a/drivers/i2c/busses/i2c-piix4.c +++ b/drivers/i2c/busses/i2c-piix4.c @@ -462,7 +462,6 @@ static void __devexit piix4_remove(struct pci_dev *dev) } static struct pci_driver piix4_driver = { - .owner = THIS_MODULE, .name = "piix4_smbus", .id_table = piix4_ids, .probe = piix4_probe, diff --git a/drivers/i2c/busses/i2c-prosavage.c b/drivers/i2c/busses/i2c-prosavage.c index 42cb1d8ca65..9479525892e 100644 --- a/drivers/i2c/busses/i2c-prosavage.c +++ b/drivers/i2c/busses/i2c-prosavage.c @@ -301,7 +301,6 @@ static struct pci_device_id prosavage_pci_tbl[] = { MODULE_DEVICE_TABLE (pci, prosavage_pci_tbl); static struct pci_driver prosavage_driver = { - .owner = THIS_MODULE, .name = "prosavage_smbus", .id_table = prosavage_pci_tbl, .probe = prosavage_probe, diff --git a/drivers/i2c/busses/i2c-savage4.c b/drivers/i2c/busses/i2c-savage4.c index aebe87ba403..0c8518298e4 100644 --- a/drivers/i2c/busses/i2c-savage4.c +++ b/drivers/i2c/busses/i2c-savage4.c @@ -179,7 +179,6 @@ static void __devexit savage4_remove(struct pci_dev *dev) } static struct pci_driver savage4_driver = { - .owner = THIS_MODULE, .name = "savage4_smbus", .id_table = savage4_ids, .probe = savage4_probe, diff --git a/drivers/i2c/busses/i2c-sis5595.c b/drivers/i2c/busses/i2c-sis5595.c index 3ad27c3ba15..b57ab74d23e 100644 --- a/drivers/i2c/busses/i2c-sis5595.c +++ b/drivers/i2c/busses/i2c-sis5595.c @@ -398,7 +398,6 @@ static void __devexit sis5595_remove(struct pci_dev *dev) } static struct pci_driver sis5595_driver = { - .owner = THIS_MODULE, .name = "sis5595_smbus", .id_table = sis5595_ids, .probe = sis5595_probe, diff --git a/drivers/i2c/busses/i2c-sis630.c b/drivers/i2c/busses/i2c-sis630.c index 7f49e5fd3ff..acb75e28241 100644 --- a/drivers/i2c/busses/i2c-sis630.c +++ b/drivers/i2c/busses/i2c-sis630.c @@ -496,7 +496,6 @@ static void __devexit sis630_remove(struct pci_dev *dev) static struct pci_driver sis630_driver = { - .owner = THIS_MODULE, .name = "sis630_smbus", .id_table = sis630_ids, .probe = sis630_probe, diff --git a/drivers/i2c/busses/i2c-sis96x.c b/drivers/i2c/busses/i2c-sis96x.c index 6a134c09132..3024907cdaf 100644 --- a/drivers/i2c/busses/i2c-sis96x.c +++ b/drivers/i2c/busses/i2c-sis96x.c @@ -329,7 +329,6 @@ static void __devexit sis96x_remove(struct pci_dev *dev) } static struct pci_driver sis96x_driver = { - .owner = THIS_MODULE, .name = "sis96x_smbus", .id_table = sis96x_ids, .probe = sis96x_probe, diff --git a/drivers/i2c/busses/i2c-via.c b/drivers/i2c/busses/i2c-via.c index 544a38e6439..484bbacfce6 100644 --- a/drivers/i2c/busses/i2c-via.c +++ b/drivers/i2c/busses/i2c-via.c @@ -159,7 +159,6 @@ static void __devexit vt586b_remove(struct pci_dev *dev) static struct pci_driver vt586b_driver = { - .owner = THIS_MODULE, .name = "vt586b_smbus", .id_table = vt586b_ids, .probe = vt586b_probe, diff --git a/drivers/i2c/busses/i2c-viapro.c b/drivers/i2c/busses/i2c-viapro.c index c9366b50483..47e52bf2c5e 100644 --- a/drivers/i2c/busses/i2c-viapro.c +++ b/drivers/i2c/busses/i2c-viapro.c @@ -142,19 +142,18 @@ static int vt596_transaction(u8 size) /* Make sure the SMBus host is ready to start transmitting */ if ((temp = inb_p(SMBHSTSTS)) & 0x1F) { dev_dbg(&vt596_adapter.dev, "SMBus busy (0x%02x). " - "Resetting... ", temp); + "Resetting...\n", temp); outb_p(temp, SMBHSTSTS); if ((temp = inb_p(SMBHSTSTS)) & 0x1F) { - printk("Failed! (0x%02x)\n", temp); + dev_err(&vt596_adapter.dev, "SMBus reset failed! " + "(0x%02x)\n", temp); return -1; - } else { - printk("Successful!\n"); } } /* Start the transaction by setting bit 6 */ - outb_p(0x40 | (size & 0x3C), SMBHSTCNT); + outb_p(0x40 | size, SMBHSTCNT); /* We will always wait for a fraction of a second */ do { @@ -171,7 +170,7 @@ static int vt596_transaction(u8 size) if (temp & 0x10) { result = -1; dev_err(&vt596_adapter.dev, "Transaction failed (0x%02x)\n", - inb_p(SMBHSTCNT) & 0x3C); + size); } if (temp & 0x08) { @@ -180,11 +179,13 @@ static int vt596_transaction(u8 size) } if (temp & 0x04) { + int read = inb_p(SMBHSTADD) & 0x01; result = -1; - /* Quick commands are used to probe for chips, so - errors are expected, and we don't want to frighten the - user. */ - if ((inb_p(SMBHSTCNT) & 0x3C) != VT596_QUICK) + /* The quick and receive byte commands are used to probe + for chips, so errors are expected, and we don't want + to frighten the user. */ + if (!((size == VT596_QUICK && !read) || + (size == VT596_BYTE && read))) dev_err(&vt596_adapter.dev, "Transaction error!\n"); } @@ -439,7 +440,6 @@ static struct pci_device_id vt596_ids[] = { MODULE_DEVICE_TABLE(pci, vt596_ids); static struct pci_driver vt596_driver = { - .owner = THIS_MODULE, .name = "vt596_smbus", .id_table = vt596_ids, .probe = vt596_probe, @@ -462,9 +462,9 @@ static void __exit i2c_vt596_exit(void) } } -MODULE_AUTHOR( - "Frodo Looijaard <frodol@dds.nl> and " - "Philip Edelbrock <phil@netroedge.com>"); +MODULE_AUTHOR("Kyosti Malkki <kmalkki@cc.hut.fi>, " + "Mark D. Studebaker <mdsxyz123@yahoo.com> and " + "Jean Delvare <khali@linux-fr.org>"); MODULE_DESCRIPTION("vt82c596 SMBus driver"); MODULE_LICENSE("GPL"); diff --git a/drivers/i2c/busses/i2c-voodoo3.c b/drivers/i2c/busses/i2c-voodoo3.c index 650c3ebde84..b675773b0cc 100644 --- a/drivers/i2c/busses/i2c-voodoo3.c +++ b/drivers/i2c/busses/i2c-voodoo3.c @@ -225,7 +225,6 @@ static void __devexit voodoo3_remove(struct pci_dev *dev) } static struct pci_driver voodoo3_driver = { - .owner = THIS_MODULE, .name = "voodoo3_smbus", .id_table = voodoo3_ids, .probe = voodoo3_probe, diff --git a/drivers/i2c/chips/ds1337.c b/drivers/i2c/chips/ds1337.c index 01b03700741..02682fb794c 100644 --- a/drivers/i2c/chips/ds1337.c +++ b/drivers/i2c/chips/ds1337.c @@ -164,9 +164,9 @@ static int ds1337_set_datetime(struct i2c_client *client, struct rtc_time *dt) buf[1] = BIN2BCD(dt->tm_sec); buf[2] = BIN2BCD(dt->tm_min); buf[3] = BIN2BCD(dt->tm_hour); - buf[4] = BIN2BCD(dt->tm_wday) + 1; + buf[4] = BIN2BCD(dt->tm_wday + 1); buf[5] = BIN2BCD(dt->tm_mday); - buf[6] = BIN2BCD(dt->tm_mon) + 1; + buf[6] = BIN2BCD(dt->tm_mon + 1); val = dt->tm_year; if (val >= 100) { val -= 100; diff --git a/drivers/ide/Kconfig b/drivers/ide/Kconfig index a737886e39d..42e5b8175cb 100644 --- a/drivers/ide/Kconfig +++ b/drivers/ide/Kconfig @@ -539,6 +539,15 @@ config BLK_DEV_CS5530 It is safe to say Y to this question. +config BLK_DEV_CS5535 + tristate "AMD CS5535 chipset support" + depends on X86 && !X86_64 + help + Include support for UDMA on the NSC/AMD CS5535 companion chipset. + This will automatically be detected and configured if found. + + It is safe to say Y to this question. + config BLK_DEV_HPT34X tristate "HPT34X chipset support" help diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c index e83f54d37f9..f615ab75996 100644 --- a/drivers/ide/ide-floppy.c +++ b/drivers/ide/ide-floppy.c @@ -2038,11 +2038,9 @@ static int idefloppy_ioctl(struct inode *inode, struct file *file, struct ide_floppy_obj *floppy = ide_floppy_g(bdev->bd_disk); ide_drive_t *drive = floppy->drive; void __user *argp = (void __user *)arg; - int err = generic_ide_ioctl(drive, file, bdev, cmd, arg); + int err; int prevent = (arg) ? 1 : 0; idefloppy_pc_t pc; - if (err != -EINVAL) - return err; switch (cmd) { case CDROMEJECT: @@ -2094,7 +2092,7 @@ static int idefloppy_ioctl(struct inode *inode, struct file *file, case IDEFLOPPY_IOCTL_FORMAT_GET_PROGRESS: return idefloppy_get_format_progress(drive, argp); } - return -EINVAL; + return generic_ide_ioctl(drive, file, bdev, cmd, arg); } static int idefloppy_media_changed(struct gendisk *disk) diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c index 0b0aa4f5162..af7af958ab3 100644 --- a/drivers/ide/ide-iops.c +++ b/drivers/ide/ide-iops.c @@ -104,8 +104,6 @@ void default_hwif_iops (ide_hwif_t *hwif) hwif->INSL = ide_insl; } -EXPORT_SYMBOL(default_hwif_iops); - /* * MMIO operations, typically used for SATA controllers */ @@ -329,8 +327,6 @@ void default_hwif_transport(ide_hwif_t *hwif) hwif->atapi_output_bytes = atapi_output_bytes; } -EXPORT_SYMBOL(default_hwif_transport); - /* * Beginning of Taskfile OPCODE Library and feature sets. */ @@ -529,8 +525,6 @@ int wait_for_ready (ide_drive_t *drive, int timeout) return 0; } -EXPORT_SYMBOL(wait_for_ready); - /* * This routine busy-waits for the drive status to be not "busy". * It then checks the status for all of the "good" bits and none diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c index 7ec18fa3b5f..54f9639c2a8 100644 --- a/drivers/ide/ide-taskfile.c +++ b/drivers/ide/ide-taskfile.c @@ -161,8 +161,6 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task) return ide_stopped; } -EXPORT_SYMBOL(do_rw_taskfile); - /* * set_multmode_intr() is invoked on completion of a WIN_SETMULT cmd. */ diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c index 9fe19808d81..8af179b531c 100644 --- a/drivers/ide/ide.c +++ b/drivers/ide/ide.c @@ -803,6 +803,7 @@ found: hwif->irq = hw->irq; hwif->noprobe = 0; hwif->chipset = hw->chipset; + hwif->gendev.parent = hw->dev; if (!initializing) { probe_hwif_init_with_fixup(hwif, fixup); diff --git a/drivers/ide/legacy/ide-cs.c b/drivers/ide/legacy/ide-cs.c index 1dafffa7e51..ef79805218e 100644 --- a/drivers/ide/legacy/ide-cs.c +++ b/drivers/ide/legacy/ide-cs.c @@ -182,13 +182,14 @@ static void ide_detach(dev_link_t *link) } /* ide_detach */ -static int idecs_register(unsigned long io, unsigned long ctl, unsigned long irq) +static int idecs_register(unsigned long io, unsigned long ctl, unsigned long irq, struct pcmcia_device *handle) { hw_regs_t hw; memset(&hw, 0, sizeof(hw)); ide_init_hwif_ports(&hw, io, ctl, NULL); hw.irq = irq; hw.chipset = ide_pci; + hw.dev = &handle->dev; return ide_register_hw_with_fixup(&hw, NULL, ide_undecoded_slave); } @@ -327,12 +328,12 @@ static void ide_config(dev_link_t *link) /* retry registration in case device is still spinning up */ for (hd = -1, i = 0; i < 10; i++) { - hd = idecs_register(io_base, ctl_base, link->irq.AssignedIRQ); + hd = idecs_register(io_base, ctl_base, link->irq.AssignedIRQ, handle); if (hd >= 0) break; if (link->io.NumPorts1 == 0x20) { outb(0x02, ctl_base + 0x10); hd = idecs_register(io_base + 0x10, ctl_base + 0x10, - link->irq.AssignedIRQ); + link->irq.AssignedIRQ, handle); if (hd >= 0) { io_base += 0x10; ctl_base += 0x10; diff --git a/drivers/ide/pci/Makefile b/drivers/ide/pci/Makefile index af46226c179..f35d684edc2 100644 --- a/drivers/ide/pci/Makefile +++ b/drivers/ide/pci/Makefile @@ -6,6 +6,7 @@ obj-$(CONFIG_BLK_DEV_ATIIXP) += atiixp.o obj-$(CONFIG_BLK_DEV_CMD64X) += cmd64x.o obj-$(CONFIG_BLK_DEV_CS5520) += cs5520.o obj-$(CONFIG_BLK_DEV_CS5530) += cs5530.o +obj-$(CONFIG_BLK_DEV_CS5535) += cs5535.o obj-$(CONFIG_BLK_DEV_SC1200) += sc1200.o obj-$(CONFIG_BLK_DEV_CY82C693) += cy82c693.o obj-$(CONFIG_BLK_DEV_HPT34X) += hpt34x.o diff --git a/drivers/ide/pci/amd74xx.c b/drivers/ide/pci/amd74xx.c index 844a6c9fb94..21965e5ef25 100644 --- a/drivers/ide/pci/amd74xx.c +++ b/drivers/ide/pci/amd74xx.c @@ -74,6 +74,7 @@ static struct amd_ide_chip { { PCI_DEVICE_ID_NVIDIA_NFORCE_MCP04_IDE, 0x50, AMD_UDMA_133 }, { PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_IDE, 0x50, AMD_UDMA_133 }, { PCI_DEVICE_ID_NVIDIA_NFORCE_MCP55_IDE, 0x50, AMD_UDMA_133 }, + { PCI_DEVICE_ID_AMD_CS5536_IDE, 0x40, AMD_UDMA_100 }, { 0 } }; @@ -491,6 +492,7 @@ static ide_pci_device_t amd74xx_chipsets[] __devinitdata = { /* 14 */ DECLARE_NV_DEV("NFORCE-MCP04"), /* 15 */ DECLARE_NV_DEV("NFORCE-MCP51"), /* 16 */ DECLARE_NV_DEV("NFORCE-MCP55"), + /* 17 */ DECLARE_AMD_DEV("AMD5536"), }; static int __devinit amd74xx_probe(struct pci_dev *dev, const struct pci_device_id *id) @@ -527,6 +529,7 @@ static struct pci_device_id amd74xx_pci_tbl[] = { { PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP04_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 14 }, { PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 15 }, { PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP55_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 16 }, + { PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_CS5536_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 17 }, { 0, }, }; MODULE_DEVICE_TABLE(pci, amd74xx_pci_tbl); diff --git a/drivers/ide/pci/cs5535.c b/drivers/ide/pci/cs5535.c new file mode 100644 index 00000000000..6eb305197f3 --- /dev/null +++ b/drivers/ide/pci/cs5535.c @@ -0,0 +1,305 @@ +/* + * linux/drivers/ide/pci/cs5535.c + * + * Copyright (C) 2004-2005 Advanced Micro Devices, Inc. + * + * History: + * 09/20/2005 - Jaya Kumar <jayakumar.ide@gmail.com> + * - Reworked tuneproc, set_drive, misc mods to prep for mainline + * - Work was sponsored by CIS (M) Sdn Bhd. + * Ported to Kernel 2.6.11 on June 26, 2005 by + * Wolfgang Zuleger <wolfgang.zuleger@gmx.de> + * Alexander Kiausch <alex.kiausch@t-online.de> + * Originally developed by AMD for 2.4/2.6 + * + * Development of this chipset driver was funded + * by the nice folks at National Semiconductor/AMD. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as published by + * the Free Software Foundation. + * + * Documentation: + * CS5535 documentation available from AMD + */ + +#include <linux/config.h> +#include <linux/module.h> +#include <linux/pci.h> +#include <linux/ide.h> + +#include "ide-timing.h" + +#define MSR_ATAC_BASE 0x51300000 +#define ATAC_GLD_MSR_CAP (MSR_ATAC_BASE+0) +#define ATAC_GLD_MSR_CONFIG (MSR_ATAC_BASE+0x01) +#define ATAC_GLD_MSR_SMI (MSR_ATAC_BASE+0x02) +#define ATAC_GLD_MSR_ERROR (MSR_ATAC_BASE+0x03) +#define ATAC_GLD_MSR_PM (MSR_ATAC_BASE+0x04) +#define ATAC_GLD_MSR_DIAG (MSR_ATAC_BASE+0x05) +#define ATAC_IO_BAR (MSR_ATAC_BASE+0x08) +#define ATAC_RESET (MSR_ATAC_BASE+0x10) +#define ATAC_CH0D0_PIO (MSR_ATAC_BASE+0x20) +#define ATAC_CH0D0_DMA (MSR_ATAC_BASE+0x21) +#define ATAC_CH0D1_PIO (MSR_ATAC_BASE+0x22) +#define ATAC_CH0D1_DMA (MSR_ATAC_BASE+0x23) +#define ATAC_PCI_ABRTERR (MSR_ATAC_BASE+0x24) +#define ATAC_BM0_CMD_PRIM 0x00 +#define ATAC_BM0_STS_PRIM 0x02 +#define ATAC_BM0_PRD 0x04 +#define CS5535_CABLE_DETECT 0x48 + +/* Format I PIO settings. We seperate out cmd and data for safer timings */ + +static unsigned int cs5535_pio_cmd_timings[5] = +{ 0xF7F4, 0x53F3, 0x13F1, 0x5131, 0x1131 }; +static unsigned int cs5535_pio_dta_timings[5] = +{ 0xF7F4, 0xF173, 0x8141, 0x5131, 0x1131 }; + +static unsigned int cs5535_mwdma_timings[3] = +{ 0x7F0FFFF3, 0x7F035352, 0x7f024241 }; + +static unsigned int cs5535_udma_timings[5] = +{ 0x7F7436A1, 0x7F733481, 0x7F723261, 0x7F713161, 0x7F703061 }; + +/* Macros to check if the register is the reset value - reset value is an + invalid timing and indicates the register has not been set previously */ + +#define CS5535_BAD_PIO(timings) ( (timings&~0x80000000UL) == 0x00009172 ) +#define CS5535_BAD_DMA(timings) ( (timings & 0x000FFFFF) == 0x00077771 ) + +/**** + * cs5535_set_speed - Configure the chipset to the new speed + * @drive: Drive to set up + * @speed: desired speed + * + * cs5535_set_speed() configures the chipset to a new speed. + */ +static void cs5535_set_speed(ide_drive_t *drive, u8 speed) +{ + + u32 reg = 0, dummy; + int unit = drive->select.b.unit; + + + /* Set the PIO timings */ + if ((speed & XFER_MODE) == XFER_PIO) { + u8 pioa; + u8 piob; + u8 cmd; + + pioa = speed - XFER_PIO_0; + piob = ide_get_best_pio_mode(&(drive->hwif->drives[!unit]), + 255, 4, NULL); + cmd = pioa < piob ? pioa : piob; + + /* Write the speed of the current drive */ + reg = (cs5535_pio_cmd_timings[cmd] << 16) | + cs5535_pio_dta_timings[pioa]; + wrmsr(unit ? ATAC_CH0D1_PIO : ATAC_CH0D0_PIO, reg, 0); + + /* And if nessesary - change the speed of the other drive */ + rdmsr(unit ? ATAC_CH0D0_PIO : ATAC_CH0D1_PIO, reg, dummy); + + if (((reg >> 16) & cs5535_pio_cmd_timings[cmd]) != + cs5535_pio_cmd_timings[cmd]) { + reg &= 0x0000FFFF; + reg |= cs5535_pio_cmd_timings[cmd] << 16; + wrmsr(unit ? ATAC_CH0D0_PIO : ATAC_CH0D1_PIO, reg, 0); + } + + /* Set bit 31 of the DMA register for PIO format 1 timings */ + rdmsr(unit ? ATAC_CH0D1_DMA : ATAC_CH0D0_DMA, reg, dummy); + wrmsr(unit ? ATAC_CH0D1_DMA : ATAC_CH0D0_DMA, + reg | 0x80000000UL, 0); + } else { + rdmsr(unit ? ATAC_CH0D1_DMA : ATAC_CH0D0_DMA, reg, dummy); + + reg &= 0x80000000UL; /* Preserve the PIO format bit */ + + if (speed >= XFER_UDMA_0 && speed <= XFER_UDMA_7) + reg |= cs5535_udma_timings[speed - XFER_UDMA_0]; + else if (speed >= XFER_MW_DMA_0 && speed <= XFER_MW_DMA_2) + reg |= cs5535_mwdma_timings[speed - XFER_MW_DMA_0]; + else + return; + + wrmsr(unit ? ATAC_CH0D1_DMA : ATAC_CH0D0_DMA, reg, 0); + } +} + +static u8 cs5535_ratemask(ide_drive_t *drive) +{ + /* eighty93 will return 1 if it's 80core and capable of + exceeding udma2, 0 otherwise. we need ratemask to set + the max speed and if we can > udma2 then we return 2 + which selects speed_max as udma4 which is the 5535's max + speed, and 1 selects udma2 which is the max for 40c */ + if (!eighty_ninty_three(drive)) + return 1; + + return 2; +} + + +/**** + * cs5535_set_drive - Configure the drive to the new speed + * @drive: Drive to set up + * @speed: desired speed + * + * cs5535_set_drive() configures the drive and the chipset to a + * new speed. It also can be called by upper layers. + */ +static int cs5535_set_drive(ide_drive_t *drive, u8 speed) +{ + speed = ide_rate_filter(cs5535_ratemask(drive), speed); + ide_config_drive_speed(drive, speed); + cs5535_set_speed(drive, speed); + + return 0; +} + +/**** + * cs5535_tuneproc - PIO setup + * @drive: drive to set up + * @pio: mode to use (255 for 'best possible') + * + * A callback from the upper layers for PIO-only tuning. + */ +static void cs5535_tuneproc(ide_drive_t *drive, u8 xferspeed) +{ + u8 modes[] = { XFER_PIO_0, XFER_PIO_1, XFER_PIO_2, XFER_PIO_3, + XFER_PIO_4 }; + + /* cs5535 max pio is pio 4, best_pio will check the blacklist. + i think we don't need to rate_filter the incoming xferspeed + since we know we're only going to choose pio */ + xferspeed = ide_get_best_pio_mode(drive, xferspeed, 4, NULL); + ide_config_drive_speed(drive, modes[xferspeed]); + cs5535_set_speed(drive, xferspeed); +} + +static int cs5535_config_drive_for_dma(ide_drive_t *drive) +{ + u8 speed; + + speed = ide_dma_speed(drive, cs5535_ratemask(drive)); + + /* If no DMA speed was available then let dma_check hit pio */ + if (!speed) { + return 0; + } + + cs5535_set_drive(drive, speed); + return ide_dma_enable(drive); +} + +static int cs5535_dma_check(ide_drive_t *drive) +{ + ide_hwif_t *hwif = drive->hwif; + struct hd_driveid *id = drive->id; + u8 speed; + + drive->init_speed = 0; + + if ((id->capability & 1) && drive->autodma) { + if (ide_use_dma(drive)) { + if (cs5535_config_drive_for_dma(drive)) + return hwif->ide_dma_on(drive); + } + + goto fast_ata_pio; + + } else if ((id->capability & 8) || (id->field_valid & 2)) { +fast_ata_pio: + speed = ide_get_best_pio_mode(drive, 255, 4, NULL); + cs5535_set_drive(drive, speed); + return hwif->ide_dma_off_quietly(drive); + } + /* IORDY not supported */ + return 0; +} + +static u8 __devinit cs5535_cable_detect(struct pci_dev *dev) +{ + u8 bit; + + /* if a 80 wire cable was detected */ + pci_read_config_byte(dev, CS5535_CABLE_DETECT, &bit); + return (bit & 1); +} + +/**** + * init_hwif_cs5535 - Initialize one ide cannel + * @hwif: Channel descriptor + * + * This gets invoked by the IDE driver once for each channel. It + * performs channel-specific pre-initialization before drive probing. + * + */ +static void __devinit init_hwif_cs5535(ide_hwif_t *hwif) +{ + int i; + + hwif->autodma = 0; + + hwif->tuneproc = &cs5535_tuneproc; + hwif->speedproc = &cs5535_set_drive; + hwif->ide_dma_check = &cs5535_dma_check; + + hwif->atapi_dma = 1; + hwif->ultra_mask = 0x1F; + hwif->mwdma_mask = 0x07; + + + hwif->udma_four = cs5535_cable_detect(hwif->pci_dev); + + if (!noautodma) + hwif->autodma = 1; + + /* just setting autotune and not worrying about bios timings */ + for (i = 0; i < 2; i++) { + hwif->drives[i].autotune = 1; + hwif->drives[i].autodma = hwif->autodma; + } +} + +static ide_pci_device_t cs5535_chipset __devinitdata = { + .name = "CS5535", + .init_hwif = init_hwif_cs5535, + .channels = 1, + .autodma = AUTODMA, + .bootable = ON_BOARD, +}; + +static int __devinit cs5535_init_one(struct pci_dev *dev, + const struct pci_device_id *id) +{ + return ide_setup_pci_device(dev, &cs5535_chipset); +} + +static struct pci_device_id cs5535_pci_tbl[] = +{ + { PCI_VENDOR_ID_NS, PCI_DEVICE_ID_NS_CS5535_IDE, PCI_ANY_ID, + PCI_ANY_ID, 0, 0, 0}, + { 0, }, +}; + +MODULE_DEVICE_TABLE(pci, cs5535_pci_tbl); + +static struct pci_driver driver = { + .name = "CS5535_IDE", + .id_table = cs5535_pci_tbl, + .probe = cs5535_init_one, +}; + +static int __init cs5535_ide_init(void) +{ + return ide_pci_register_driver(&driver); +} + +module_init(cs5535_ide_init); + +MODULE_AUTHOR("AMD"); +MODULE_DESCRIPTION("PCI driver module for AMD/NS CS5535 IDE"); +MODULE_LICENSE("GPL"); diff --git a/drivers/ide/pci/cy82c693.c b/drivers/ide/pci/cy82c693.c index 5a33513f3dd..9f41ecd5633 100644 --- a/drivers/ide/pci/cy82c693.c +++ b/drivers/ide/pci/cy82c693.c @@ -469,7 +469,7 @@ static void __devinit init_hwif_cy82c693(ide_hwif_t *hwif) static __devinitdata ide_hwif_t *primary; -void __devinit init_iops_cy82c693(ide_hwif_t *hwif) +static void __devinit init_iops_cy82c693(ide_hwif_t *hwif) { if (PCI_FUNC(hwif->pci_dev->devfn) == 1) primary = hwif; diff --git a/drivers/ide/pci/siimage.c b/drivers/ide/pci/siimage.c index 2b9961b8813..022d244f2eb 100644 --- a/drivers/ide/pci/siimage.c +++ b/drivers/ide/pci/siimage.c @@ -701,6 +701,7 @@ static unsigned int setup_mmio_siimage (struct pci_dev *dev, const char *name) unsigned long barsize = pci_resource_len(dev, 5); u8 tmpbyte = 0; void __iomem *ioaddr; + u32 tmp, irq_mask; /* * Drop back to PIO if we can't map the mmio. Some @@ -726,6 +727,14 @@ static unsigned int setup_mmio_siimage (struct pci_dev *dev, const char *name) pci_set_drvdata(dev, (void *) ioaddr); if (pdev_is_sata(dev)) { + /* make sure IDE0/1 interrupts are not masked */ + irq_mask = (1 << 22) | (1 << 23); + tmp = readl(ioaddr + 0x48); + if (tmp & irq_mask) { + tmp &= ~irq_mask; + writel(tmp, ioaddr + 0x48); + readl(ioaddr + 0x48); /* flush */ + } writel(0, ioaddr + 0x148); writel(0, ioaddr + 0x1C8); } diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c index 18ed7765417..d4f2111d436 100644 --- a/drivers/ide/setup-pci.c +++ b/drivers/ide/setup-pci.c @@ -787,8 +787,9 @@ static int pre_init = 1; /* Before first ordered IDE scan */ static LIST_HEAD(ide_pci_drivers); /* - * ide_register_pci_driver - attach IDE driver + * __ide_register_pci_driver - attach IDE driver * @driver: pci driver + * @module: owner module of the driver * * Registers a driver with the IDE layer. The IDE layer arranges that * boot time setup is done in the expected device order and then @@ -801,15 +802,16 @@ static LIST_HEAD(ide_pci_drivers); * Returns are the same as for pci_register_driver */ -int ide_pci_register_driver(struct pci_driver *driver) +int __ide_pci_register_driver(struct pci_driver *driver, struct module *module) { if(!pre_init) - return pci_module_init(driver); + return __pci_register_driver(driver, module); + driver->driver.owner = module; list_add_tail(&driver->node, &ide_pci_drivers); return 0; } -EXPORT_SYMBOL_GPL(ide_pci_register_driver); +EXPORT_SYMBOL_GPL(__ide_pci_register_driver); /** * ide_unregister_pci_driver - unregister an IDE driver @@ -897,6 +899,6 @@ void __init ide_scan_pcibus (int scan_direction) { list_del(l); d = list_entry(l, struct pci_driver, node); - pci_register_driver(d); + __pci_register_driver(d, d->driver.owner); } } diff --git a/drivers/infiniband/core/user_mad.c b/drivers/infiniband/core/user_mad.c index aed5ca23fb2..5ea741f47fc 100644 --- a/drivers/infiniband/core/user_mad.c +++ b/drivers/infiniband/core/user_mad.c @@ -31,7 +31,7 @@ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. * - * $Id: user_mad.c 2814 2005-07-06 19:14:09Z halr $ + * $Id: user_mad.c 4010 2005-11-09 23:11:56Z roland $ */ #include <linux/module.h> @@ -110,13 +110,13 @@ struct ib_umad_device { }; struct ib_umad_file { - struct ib_umad_port *port; - struct list_head recv_list; - struct list_head port_list; - spinlock_t recv_lock; - wait_queue_head_t recv_wait; - struct ib_mad_agent *agent[IB_UMAD_MAX_AGENTS]; - struct ib_mr *mr[IB_UMAD_MAX_AGENTS]; + struct ib_umad_port *port; + struct list_head recv_list; + struct list_head port_list; + spinlock_t recv_lock; + wait_queue_head_t recv_wait; + struct ib_mad_agent *agent[IB_UMAD_MAX_AGENTS]; + int agents_dead; }; struct ib_umad_packet { @@ -145,6 +145,12 @@ static void ib_umad_release_dev(struct kref *ref) kfree(dev); } +/* caller must hold port->mutex at least for reading */ +static struct ib_mad_agent *__get_agent(struct ib_umad_file *file, int id) +{ + return file->agents_dead ? NULL : file->agent[id]; +} + static int queue_packet(struct ib_umad_file *file, struct ib_mad_agent *agent, struct ib_umad_packet *packet) @@ -152,10 +158,11 @@ static int queue_packet(struct ib_umad_file *file, int ret = 1; down_read(&file->port->mutex); + for (packet->mad.hdr.id = 0; packet->mad.hdr.id < IB_UMAD_MAX_AGENTS; packet->mad.hdr.id++) - if (agent == file->agent[packet->mad.hdr.id]) { + if (agent == __get_agent(file, packet->mad.hdr.id)) { spin_lock_irq(&file->recv_lock); list_add_tail(&packet->list, &file->recv_list); spin_unlock_irq(&file->recv_lock); @@ -327,7 +334,7 @@ static ssize_t ib_umad_write(struct file *filp, const char __user *buf, down_read(&file->port->mutex); - agent = file->agent[packet->mad.hdr.id]; + agent = __get_agent(file, packet->mad.hdr.id); if (!agent) { ret = -EINVAL; goto err_up; @@ -481,7 +488,7 @@ static int ib_umad_reg_agent(struct ib_umad_file *file, unsigned long arg) } for (agent_id = 0; agent_id < IB_UMAD_MAX_AGENTS; ++agent_id) - if (!file->agent[agent_id]) + if (!__get_agent(file, agent_id)) goto found; ret = -ENOMEM; @@ -505,29 +512,15 @@ found: goto out; } - file->agent[agent_id] = agent; - - file->mr[agent_id] = ib_get_dma_mr(agent->qp->pd, IB_ACCESS_LOCAL_WRITE); - if (IS_ERR(file->mr[agent_id])) { - ret = -ENOMEM; - goto err; - } - if (put_user(agent_id, (u32 __user *) (arg + offsetof(struct ib_user_mad_reg_req, id)))) { ret = -EFAULT; - goto err_mr; + ib_unregister_mad_agent(agent); + goto out; } + file->agent[agent_id] = agent; ret = 0; - goto out; - -err_mr: - ib_dereg_mr(file->mr[agent_id]); - -err: - file->agent[agent_id] = NULL; - ib_unregister_mad_agent(agent); out: up_write(&file->port->mutex); @@ -536,27 +529,29 @@ out: static int ib_umad_unreg_agent(struct ib_umad_file *file, unsigned long arg) { + struct ib_mad_agent *agent = NULL; u32 id; int ret = 0; - down_write(&file->port->mutex); + if (get_user(id, (u32 __user *) arg)) + return -EFAULT; - if (get_user(id, (u32 __user *) arg)) { - ret = -EFAULT; - goto out; - } + down_write(&file->port->mutex); - if (id < 0 || id >= IB_UMAD_MAX_AGENTS || !file->agent[id]) { + if (id < 0 || id >= IB_UMAD_MAX_AGENTS || !__get_agent(file, id)) { ret = -EINVAL; goto out; } - ib_dereg_mr(file->mr[id]); - ib_unregister_mad_agent(file->agent[id]); + agent = file->agent[id]; file->agent[id] = NULL; out: up_write(&file->port->mutex); + + if (agent) + ib_unregister_mad_agent(agent); + return ret; } @@ -621,23 +616,29 @@ static int ib_umad_close(struct inode *inode, struct file *filp) struct ib_umad_file *file = filp->private_data; struct ib_umad_device *dev = file->port->umad_dev; struct ib_umad_packet *packet, *tmp; + int already_dead; int i; down_write(&file->port->mutex); - for (i = 0; i < IB_UMAD_MAX_AGENTS; ++i) - if (file->agent[i]) { - ib_dereg_mr(file->mr[i]); - ib_unregister_mad_agent(file->agent[i]); - } + + already_dead = file->agents_dead; + file->agents_dead = 1; list_for_each_entry_safe(packet, tmp, &file->recv_list, list) kfree(packet); list_del(&file->port_list); - up_write(&file->port->mutex); - kfree(file); + downgrade_write(&file->port->mutex); + + if (!already_dead) + for (i = 0; i < IB_UMAD_MAX_AGENTS; ++i) + if (file->agent[i]) + ib_unregister_mad_agent(file->agent[i]); + + up_read(&file->port->mutex); + kfree(file); kref_put(&dev->ref, ib_umad_release_dev); return 0; @@ -801,7 +802,7 @@ static int ib_umad_init_port(struct ib_device *device, int port_num, goto err_class; port->sm_dev->owner = THIS_MODULE; port->sm_dev->ops = &umad_sm_fops; - kobject_set_name(&port->dev->kobj, "issm%d", port->dev_num); + kobject_set_name(&port->sm_dev->kobj, "issm%d", port->dev_num); if (cdev_add(port->sm_dev, base_dev + port->dev_num + IB_UMAD_MAX_PORTS, 1)) goto err_sm_cdev; @@ -863,14 +864,36 @@ static void ib_umad_kill_port(struct ib_umad_port *port) port->ib_dev = NULL; - list_for_each_entry(file, &port->file_list, port_list) - for (id = 0; id < IB_UMAD_MAX_AGENTS; ++id) { - if (!file->agent[id]) - continue; - ib_dereg_mr(file->mr[id]); - ib_unregister_mad_agent(file->agent[id]); - file->agent[id] = NULL; - } + /* + * Now go through the list of files attached to this port and + * unregister all of their MAD agents. We need to hold + * port->mutex while doing this to avoid racing with + * ib_umad_close(), but we can't hold the mutex for writing + * while calling ib_unregister_mad_agent(), since that might + * deadlock by calling back into queue_packet(). So we + * downgrade our lock to a read lock, and then drop and + * reacquire the write lock for the next iteration. + * + * We do list_del_init() on the file's list_head so that the + * list_del in ib_umad_close() is still OK, even after the + * file is removed from the list. + */ + while (!list_empty(&port->file_list)) { + file = list_entry(port->file_list.next, struct ib_umad_file, + port_list); + + file->agents_dead = 1; + list_del_init(&file->port_list); + + downgrade_write(&port->mutex); + + for (id = 0; id < IB_UMAD_MAX_AGENTS; ++id) + if (file->agent[id]) + ib_unregister_mad_agent(file->agent[id]); + + up_read(&port->mutex); + down_write(&port->mutex); + } up_write(&port->mutex); @@ -913,7 +936,7 @@ static void ib_umad_add_one(struct ib_device *device) err: while (--i >= s) - ib_umad_kill_port(&umad_dev->port[i]); + ib_umad_kill_port(&umad_dev->port[i - s]); kref_put(&umad_dev->ref, ib_umad_release_dev); } diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index 63a74151c60..ed45da892b1 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -708,7 +708,7 @@ ssize_t ib_uverbs_poll_cq(struct ib_uverbs_file *file, resp->wc[i].opcode = wc[i].opcode; resp->wc[i].vendor_err = wc[i].vendor_err; resp->wc[i].byte_len = wc[i].byte_len; - resp->wc[i].imm_data = wc[i].imm_data; + resp->wc[i].imm_data = (__u32 __force) wc[i].imm_data; resp->wc[i].qp_num = wc[i].qp_num; resp->wc[i].src_qp = wc[i].src_qp; resp->wc[i].wc_flags = wc[i].wc_flags; @@ -908,7 +908,12 @@ retry: if (ret) goto err_destroy; - resp.qp_handle = uobj->uobject.id; + resp.qp_handle = uobj->uobject.id; + resp.max_recv_sge = attr.cap.max_recv_sge; + resp.max_send_sge = attr.cap.max_send_sge; + resp.max_recv_wr = attr.cap.max_recv_wr; + resp.max_send_wr = attr.cap.max_send_wr; + resp.max_inline_data = attr.cap.max_inline_data; if (copy_to_user((void __user *) (unsigned long) cmd.response, &resp, sizeof resp)) { @@ -1135,7 +1140,7 @@ ssize_t ib_uverbs_post_send(struct ib_uverbs_file *file, next->num_sge = user_wr->num_sge; next->opcode = user_wr->opcode; next->send_flags = user_wr->send_flags; - next->imm_data = user_wr->imm_data; + next->imm_data = (__be32 __force) user_wr->imm_data; if (qp->qp_type == IB_QPT_UD) { next->wr.ud.ah = idr_find(&ib_uverbs_ah_idr, @@ -1701,7 +1706,6 @@ ssize_t ib_uverbs_modify_srq(struct ib_uverbs_file *file, } attr.max_wr = cmd.max_wr; - attr.max_sge = cmd.max_sge; attr.srq_limit = cmd.srq_limit; ret = ib_modify_srq(srq, &attr, cmd.attr_mask); diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c index 4186cc888ea..4c15e112736 100644 --- a/drivers/infiniband/core/verbs.c +++ b/drivers/infiniband/core/verbs.c @@ -325,16 +325,8 @@ EXPORT_SYMBOL(ib_destroy_cq); int ib_resize_cq(struct ib_cq *cq, int cqe) { - int ret; - - if (!cq->device->resize_cq) - return -ENOSYS; - - ret = cq->device->resize_cq(cq, &cqe); - if (!ret) - cq->cqe = cqe; - - return ret; + return cq->device->resize_cq ? + cq->device->resize_cq(cq, cqe) : -ENOSYS; } EXPORT_SYMBOL(ib_resize_cq); diff --git a/drivers/infiniband/hw/mthca/mthca_catas.c b/drivers/infiniband/hw/mthca/mthca_catas.c index 25ebab64bc4..c3bec7490f5 100644 --- a/drivers/infiniband/hw/mthca/mthca_catas.c +++ b/drivers/infiniband/hw/mthca/mthca_catas.c @@ -97,7 +97,7 @@ static void poll_catas(unsigned long dev_ptr) } spin_lock_irqsave(&catas_lock, flags); - if (dev->catas_err.stop) + if (!dev->catas_err.stop) mod_timer(&dev->catas_err.timer, jiffies + MTHCA_CATAS_POLL_INTERVAL); spin_unlock_irqrestore(&catas_lock, flags); diff --git a/drivers/infiniband/hw/mthca/mthca_cmd.c b/drivers/infiniband/hw/mthca/mthca_cmd.c index 49f211d55df..9ed34587fc5 100644 --- a/drivers/infiniband/hw/mthca/mthca_cmd.c +++ b/drivers/infiniband/hw/mthca/mthca_cmd.c @@ -1060,6 +1060,8 @@ int mthca_QUERY_DEV_LIM(struct mthca_dev *dev, dev_lim->hca.arbel.resize_srq = field & 1; MTHCA_GET(field, outbox, QUERY_DEV_LIM_MAX_SG_RQ_OFFSET); dev_lim->max_sg = min_t(int, field, dev_lim->max_sg); + MTHCA_GET(size, outbox, QUERY_DEV_LIM_MAX_DESC_SZ_RQ_OFFSET); + dev_lim->max_desc_sz = min_t(int, size, dev_lim->max_desc_sz); MTHCA_GET(size, outbox, QUERY_DEV_LIM_MPT_ENTRY_SZ_OFFSET); dev_lim->mpt_entry_sz = size; MTHCA_GET(field, outbox, QUERY_DEV_LIM_PBL_SZ_OFFSET); diff --git a/drivers/infiniband/hw/mthca/mthca_cq.c b/drivers/infiniband/hw/mthca/mthca_cq.c index f98e2355582..4a8adcef207 100644 --- a/drivers/infiniband/hw/mthca/mthca_cq.c +++ b/drivers/infiniband/hw/mthca/mthca_cq.c @@ -258,7 +258,7 @@ void mthca_cq_clean(struct mthca_dev *dev, u32 cqn, u32 qpn, { struct mthca_cq *cq; struct mthca_cqe *cqe; - int prod_index; + u32 prod_index; int nfreed = 0; spin_lock_irq(&dev->cq_table.lock); @@ -293,19 +293,15 @@ void mthca_cq_clean(struct mthca_dev *dev, u32 cqn, u32 qpn, * Now sweep backwards through the CQ, removing CQ entries * that match our QP by copying older entries on top of them. */ - while (prod_index > cq->cons_index) { - cqe = get_cqe(cq, (prod_index - 1) & cq->ibcq.cqe); + while ((int) --prod_index - (int) cq->cons_index >= 0) { + cqe = get_cqe(cq, prod_index & cq->ibcq.cqe); if (cqe->my_qpn == cpu_to_be32(qpn)) { if (srq) mthca_free_srq_wqe(srq, be32_to_cpu(cqe->wqe)); ++nfreed; - } - else if (nfreed) - memcpy(get_cqe(cq, (prod_index - 1 + nfreed) & - cq->ibcq.cqe), - cqe, - MTHCA_CQ_ENTRY_SIZE); - --prod_index; + } else if (nfreed) + memcpy(get_cqe(cq, (prod_index + nfreed) & cq->ibcq.cqe), + cqe, MTHCA_CQ_ENTRY_SIZE); } if (nfreed) { diff --git a/drivers/infiniband/hw/mthca/mthca_dev.h b/drivers/infiniband/hw/mthca/mthca_dev.h index e7e5d3b4f00..497ff794ef6 100644 --- a/drivers/infiniband/hw/mthca/mthca_dev.h +++ b/drivers/infiniband/hw/mthca/mthca_dev.h @@ -131,6 +131,7 @@ struct mthca_limits { int max_sg; int num_qps; int max_wqes; + int max_desc_sz; int max_qp_init_rdma; int reserved_qps; int num_srqs; @@ -154,6 +155,7 @@ struct mthca_limits { int reserved_mcgs; int num_pds; int reserved_pds; + u32 page_size_cap; u32 flags; u8 port_width_cap; }; diff --git a/drivers/infiniband/hw/mthca/mthca_main.c b/drivers/infiniband/hw/mthca/mthca_main.c index 45c6328e780..6f94b25f3ac 100644 --- a/drivers/infiniband/hw/mthca/mthca_main.c +++ b/drivers/infiniband/hw/mthca/mthca_main.c @@ -168,6 +168,7 @@ static int __devinit mthca_dev_lim(struct mthca_dev *mdev, struct mthca_dev_lim mdev->limits.max_srq_wqes = dev_lim->max_srq_sz; mdev->limits.reserved_srqs = dev_lim->reserved_srqs; mdev->limits.reserved_eecs = dev_lim->reserved_eecs; + mdev->limits.max_desc_sz = dev_lim->max_desc_sz; /* * Subtract 1 from the limit because we need to allocate a * spare CQE so the HCA HW can tell the difference between an @@ -181,6 +182,7 @@ static int __devinit mthca_dev_lim(struct mthca_dev *mdev, struct mthca_dev_lim mdev->limits.reserved_uars = dev_lim->reserved_uars; mdev->limits.reserved_pds = dev_lim->reserved_pds; mdev->limits.port_width_cap = dev_lim->max_port_width; + mdev->limits.page_size_cap = ~(u32) (dev_lim->min_page_sz - 1); mdev->limits.flags = dev_lim->flags; /* IB_DEVICE_RESIZE_MAX_WR not supported by driver. @@ -1196,7 +1198,6 @@ MODULE_DEVICE_TABLE(pci, mthca_pci_table); static struct pci_driver mthca_driver = { .name = DRV_NAME, - .owner = THIS_MODULE, .id_table = mthca_pci_table, .probe = mthca_init_one, .remove = __devexit_p(mthca_remove_one) diff --git a/drivers/infiniband/hw/mthca/mthca_provider.c b/drivers/infiniband/hw/mthca/mthca_provider.c index 6b016666826..4cc7e2846df 100644 --- a/drivers/infiniband/hw/mthca/mthca_provider.c +++ b/drivers/infiniband/hw/mthca/mthca_provider.c @@ -90,6 +90,7 @@ static int mthca_query_device(struct ib_device *ibdev, memcpy(&props->node_guid, out_mad->data + 12, 8); props->max_mr_size = ~0ull; + props->page_size_cap = mdev->limits.page_size_cap; props->max_qp = mdev->limits.num_qps - mdev->limits.reserved_qps; props->max_qp_wr = mdev->limits.max_wqes; props->max_sge = mdev->limits.max_sg; @@ -615,11 +616,11 @@ static struct ib_qp *mthca_create_qp(struct ib_pd *pd, return ERR_PTR(err); } - init_attr->cap.max_inline_data = 0; init_attr->cap.max_send_wr = qp->sq.max; init_attr->cap.max_recv_wr = qp->rq.max; init_attr->cap.max_send_sge = qp->sq.max_gs; init_attr->cap.max_recv_sge = qp->rq.max_gs; + init_attr->cap.max_inline_data = qp->max_inline_data; return &qp->ibqp; } diff --git a/drivers/infiniband/hw/mthca/mthca_provider.h b/drivers/infiniband/hw/mthca/mthca_provider.h index bcd4b01a339..1e73947b470 100644 --- a/drivers/infiniband/hw/mthca/mthca_provider.h +++ b/drivers/infiniband/hw/mthca/mthca_provider.h @@ -251,6 +251,7 @@ struct mthca_qp { struct mthca_wq sq; enum ib_sig_type sq_policy; int send_wqe_offset; + int max_inline_data; u64 *wrid; union mthca_buf queue; diff --git a/drivers/infiniband/hw/mthca/mthca_qp.c b/drivers/infiniband/hw/mthca/mthca_qp.c index 8852ea477c2..760c418d5bc 100644 --- a/drivers/infiniband/hw/mthca/mthca_qp.c +++ b/drivers/infiniband/hw/mthca/mthca_qp.c @@ -885,6 +885,48 @@ int mthca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask) return err; } +static void mthca_adjust_qp_caps(struct mthca_dev *dev, + struct mthca_pd *pd, + struct mthca_qp *qp) +{ + int max_data_size; + + /* + * Calculate the maximum size of WQE s/g segments, excluding + * the next segment and other non-data segments. + */ + max_data_size = min(dev->limits.max_desc_sz, 1 << qp->sq.wqe_shift) - + sizeof (struct mthca_next_seg); + + switch (qp->transport) { + case MLX: + max_data_size -= 2 * sizeof (struct mthca_data_seg); + break; + + case UD: + if (mthca_is_memfree(dev)) + max_data_size -= sizeof (struct mthca_arbel_ud_seg); + else + max_data_size -= sizeof (struct mthca_tavor_ud_seg); + break; + + default: + max_data_size -= sizeof (struct mthca_raddr_seg); + break; + } + + /* We don't support inline data for kernel QPs (yet). */ + if (!pd->ibpd.uobject) + qp->max_inline_data = 0; + else + qp->max_inline_data = max_data_size - MTHCA_INLINE_HEADER_SIZE; + + qp->sq.max_gs = max_data_size / sizeof (struct mthca_data_seg); + qp->rq.max_gs = (min(dev->limits.max_desc_sz, 1 << qp->rq.wqe_shift) - + sizeof (struct mthca_next_seg)) / + sizeof (struct mthca_data_seg); +} + /* * Allocate and register buffer for WQEs. qp->rq.max, sq.max, * rq.max_gs and sq.max_gs must all be assigned. @@ -902,27 +944,53 @@ static int mthca_alloc_wqe_buf(struct mthca_dev *dev, size = sizeof (struct mthca_next_seg) + qp->rq.max_gs * sizeof (struct mthca_data_seg); + if (size > dev->limits.max_desc_sz) + return -EINVAL; + for (qp->rq.wqe_shift = 6; 1 << qp->rq.wqe_shift < size; qp->rq.wqe_shift++) ; /* nothing */ - size = sizeof (struct mthca_next_seg) + - qp->sq.max_gs * sizeof (struct mthca_data_seg); + size = qp->sq.max_gs * sizeof (struct mthca_data_seg); switch (qp->transport) { case MLX: size += 2 * sizeof (struct mthca_data_seg); break; + case UD: - if (mthca_is_memfree(dev)) - size += sizeof (struct mthca_arbel_ud_seg); - else - size += sizeof (struct mthca_tavor_ud_seg); + size += mthca_is_memfree(dev) ? + sizeof (struct mthca_arbel_ud_seg) : + sizeof (struct mthca_tavor_ud_seg); break; + + case UC: + size += sizeof (struct mthca_raddr_seg); + break; + + case RC: + size += sizeof (struct mthca_raddr_seg); + /* + * An atomic op will require an atomic segment, a + * remote address segment and one scatter entry. + */ + size = max_t(int, size, + sizeof (struct mthca_atomic_seg) + + sizeof (struct mthca_raddr_seg) + + sizeof (struct mthca_data_seg)); + break; + default: - /* bind seg is as big as atomic + raddr segs */ - size += sizeof (struct mthca_bind_seg); + break; } + /* Make sure that we have enough space for a bind request */ + size = max_t(int, size, sizeof (struct mthca_bind_seg)); + + size += sizeof (struct mthca_next_seg); + + if (size > dev->limits.max_desc_sz) + return -EINVAL; + for (qp->sq.wqe_shift = 6; 1 << qp->sq.wqe_shift < size; qp->sq.wqe_shift++) ; /* nothing */ @@ -1066,6 +1134,8 @@ static int mthca_alloc_qp_common(struct mthca_dev *dev, return ret; } + mthca_adjust_qp_caps(dev, pd, qp); + /* * If this is a userspace QP, we're done now. The doorbells * will be allocated and buffers will be initialized in @@ -1486,8 +1556,8 @@ int mthca_tavor_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, } wqe += sizeof (struct mthca_atomic_seg); - size += sizeof (struct mthca_raddr_seg) / 16 + - sizeof (struct mthca_atomic_seg); + size += (sizeof (struct mthca_raddr_seg) + + sizeof (struct mthca_atomic_seg)) / 16; break; case IB_WR_RDMA_WRITE: @@ -1637,6 +1707,7 @@ int mthca_tavor_post_receive(struct ib_qp *ibqp, struct ib_recv_wr *wr, { struct mthca_dev *dev = to_mdev(ibqp->device); struct mthca_qp *qp = to_mqp(ibqp); + __be32 doorbell[2]; unsigned long flags; int err = 0; int nreq; @@ -1654,6 +1725,22 @@ int mthca_tavor_post_receive(struct ib_qp *ibqp, struct ib_recv_wr *wr, ind = qp->rq.next_ind; for (nreq = 0; wr; ++nreq, wr = wr->next) { + if (unlikely(nreq == MTHCA_TAVOR_MAX_WQES_PER_RECV_DB)) { + nreq = 0; + + doorbell[0] = cpu_to_be32((qp->rq.next_ind << qp->rq.wqe_shift) | size0); + doorbell[1] = cpu_to_be32(qp->qpn << 8); + + wmb(); + + mthca_write64(doorbell, + dev->kar + MTHCA_RECEIVE_DOORBELL, + MTHCA_GET_DOORBELL_LOCK(&dev->doorbell_lock)); + + qp->rq.head += MTHCA_TAVOR_MAX_WQES_PER_RECV_DB; + size0 = 0; + } + if (mthca_wq_overflow(&qp->rq, nreq, qp->ibqp.recv_cq)) { mthca_err(dev, "RQ %06x full (%u head, %u tail," " %d max, %d nreq)\n", qp->qpn, @@ -1711,8 +1798,6 @@ int mthca_tavor_post_receive(struct ib_qp *ibqp, struct ib_recv_wr *wr, out: if (likely(nreq)) { - __be32 doorbell[2]; - doorbell[0] = cpu_to_be32((qp->rq.next_ind << qp->rq.wqe_shift) | size0); doorbell[1] = cpu_to_be32((qp->qpn << 8) | nreq); @@ -1806,8 +1891,8 @@ int mthca_arbel_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, } wqe += sizeof (struct mthca_atomic_seg); - size += sizeof (struct mthca_raddr_seg) / 16 + - sizeof (struct mthca_atomic_seg); + size += (sizeof (struct mthca_raddr_seg) + + sizeof (struct mthca_atomic_seg)) / 16; break; case IB_WR_RDMA_READ: diff --git a/drivers/infiniband/hw/mthca/mthca_srq.c b/drivers/infiniband/hw/mthca/mthca_srq.c index 26d5161fde0..f7d234295ef 100644 --- a/drivers/infiniband/hw/mthca/mthca_srq.c +++ b/drivers/infiniband/hw/mthca/mthca_srq.c @@ -417,6 +417,7 @@ int mthca_tavor_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr, { struct mthca_dev *dev = to_mdev(ibsrq->device); struct mthca_srq *srq = to_msrq(ibsrq); + __be32 doorbell[2]; unsigned long flags; int err = 0; int first_ind; @@ -432,6 +433,25 @@ int mthca_tavor_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr, first_ind = srq->first_free; for (nreq = 0; wr; ++nreq, wr = wr->next) { + if (unlikely(nreq == MTHCA_TAVOR_MAX_WQES_PER_RECV_DB)) { + nreq = 0; + + doorbell[0] = cpu_to_be32(first_ind << srq->wqe_shift); + doorbell[1] = cpu_to_be32(srq->srqn << 8); + + /* + * Make sure that descriptors are written + * before doorbell is rung. + */ + wmb(); + + mthca_write64(doorbell, + dev->kar + MTHCA_RECEIVE_DOORBELL, + MTHCA_GET_DOORBELL_LOCK(&dev->doorbell_lock)); + + first_ind = srq->first_free; + } + ind = srq->first_free; if (ind < 0) { @@ -494,8 +514,6 @@ int mthca_tavor_post_srq_recv(struct ib_srq *ibsrq, struct ib_recv_wr *wr, } if (likely(nreq)) { - __be32 doorbell[2]; - doorbell[0] = cpu_to_be32(first_ind << srq->wqe_shift); doorbell[1] = cpu_to_be32((srq->srqn << 8) | nreq); diff --git a/drivers/infiniband/hw/mthca/mthca_wqe.h b/drivers/infiniband/hw/mthca/mthca_wqe.h index 1f4c0ff28f7..73f1c0b9021 100644 --- a/drivers/infiniband/hw/mthca/mthca_wqe.h +++ b/drivers/infiniband/hw/mthca/mthca_wqe.h @@ -49,7 +49,8 @@ enum { }; enum { - MTHCA_INVAL_LKEY = 0x100 + MTHCA_INVAL_LKEY = 0x100, + MTHCA_TAVOR_MAX_WQES_PER_RECV_DB = 256 }; struct mthca_next_seg { diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h index 0095acc0fbb..9923a15a999 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib.h +++ b/drivers/infiniband/ulp/ipoib/ipoib.h @@ -179,6 +179,7 @@ struct ipoib_dev_priv { #ifdef CONFIG_INFINIBAND_IPOIB_DEBUG struct list_head fs_list; struct dentry *mcg_dentry; + struct dentry *path_dentry; #endif }; @@ -270,7 +271,6 @@ void ipoib_mcast_dev_flush(struct net_device *dev); #ifdef CONFIG_INFINIBAND_IPOIB_DEBUG struct ipoib_mcast_iter *ipoib_mcast_iter_init(struct net_device *dev); -void ipoib_mcast_iter_free(struct ipoib_mcast_iter *iter); int ipoib_mcast_iter_next(struct ipoib_mcast_iter *iter); void ipoib_mcast_iter_read(struct ipoib_mcast_iter *iter, union ib_gid *gid, @@ -278,6 +278,11 @@ void ipoib_mcast_iter_read(struct ipoib_mcast_iter *iter, unsigned int *queuelen, unsigned int *complete, unsigned int *send_only); + +struct ipoib_path_iter *ipoib_path_iter_init(struct net_device *dev); +int ipoib_path_iter_next(struct ipoib_path_iter *iter); +void ipoib_path_iter_read(struct ipoib_path_iter *iter, + struct ipoib_path *path); #endif int ipoib_mcast_attach(struct net_device *dev, u16 mlid, @@ -299,13 +304,13 @@ void ipoib_pkey_poll(void *dev); int ipoib_pkey_dev_delay_open(struct net_device *dev); #ifdef CONFIG_INFINIBAND_IPOIB_DEBUG -int ipoib_create_debug_file(struct net_device *dev); -void ipoib_delete_debug_file(struct net_device *dev); +void ipoib_create_debug_files(struct net_device *dev); +void ipoib_delete_debug_files(struct net_device *dev); int ipoib_register_debugfs(void); void ipoib_unregister_debugfs(void); #else -static inline int ipoib_create_debug_file(struct net_device *dev) { return 0; } -static inline void ipoib_delete_debug_file(struct net_device *dev) { } +static inline void ipoib_create_debug_files(struct net_device *dev) { } +static inline void ipoib_delete_debug_files(struct net_device *dev) { } static inline int ipoib_register_debugfs(void) { return 0; } static inline void ipoib_unregister_debugfs(void) { } #endif diff --git a/drivers/infiniband/ulp/ipoib/ipoib_fs.c b/drivers/infiniband/ulp/ipoib/ipoib_fs.c index 38b150f775e..685258e3403 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_fs.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_fs.c @@ -43,6 +43,18 @@ struct file_operations; static struct dentry *ipoib_root; +static void format_gid(union ib_gid *gid, char *buf) +{ + int i, n; + + for (n = 0, i = 0; i < 8; ++i) { + n += sprintf(buf + n, "%x", + be16_to_cpu(((__be16 *) gid->raw)[i])); + if (i < 7) + buf[n++] = ':'; + } +} + static void *ipoib_mcg_seq_start(struct seq_file *file, loff_t *pos) { struct ipoib_mcast_iter *iter; @@ -54,7 +66,7 @@ static void *ipoib_mcg_seq_start(struct seq_file *file, loff_t *pos) while (n--) { if (ipoib_mcast_iter_next(iter)) { - ipoib_mcast_iter_free(iter); + kfree(iter); return NULL; } } @@ -70,7 +82,7 @@ static void *ipoib_mcg_seq_next(struct seq_file *file, void *iter_ptr, (*pos)++; if (ipoib_mcast_iter_next(iter)) { - ipoib_mcast_iter_free(iter); + kfree(iter); return NULL; } @@ -87,32 +99,32 @@ static int ipoib_mcg_seq_show(struct seq_file *file, void *iter_ptr) struct ipoib_mcast_iter *iter = iter_ptr; char gid_buf[sizeof "ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff"]; union ib_gid mgid; - int i, n; unsigned long created; unsigned int queuelen, complete, send_only; - if (iter) { - ipoib_mcast_iter_read(iter, &mgid, &created, &queuelen, - &complete, &send_only); + if (!iter) + return 0; - for (n = 0, i = 0; i < sizeof mgid / 2; ++i) { - n += sprintf(gid_buf + n, "%x", - be16_to_cpu(((__be16 *) mgid.raw)[i])); - if (i < sizeof mgid / 2 - 1) - gid_buf[n++] = ':'; - } - } + ipoib_mcast_iter_read(iter, &mgid, &created, &queuelen, + &complete, &send_only); - seq_printf(file, "GID: %*s", -(1 + (int) sizeof gid_buf), gid_buf); + format_gid(&mgid, gid_buf); seq_printf(file, - " created: %10ld queuelen: %4d complete: %d send_only: %d\n", - created, queuelen, complete, send_only); + "GID: %s\n" + " created: %10ld\n" + " queuelen: %9d\n" + " complete: %9s\n" + " send_only: %8s\n" + "\n", + gid_buf, created, queuelen, + complete ? "yes" : "no", + send_only ? "yes" : "no"); return 0; } -static struct seq_operations ipoib_seq_ops = { +static struct seq_operations ipoib_mcg_seq_ops = { .start = ipoib_mcg_seq_start, .next = ipoib_mcg_seq_next, .stop = ipoib_mcg_seq_stop, @@ -124,7 +136,7 @@ static int ipoib_mcg_open(struct inode *inode, struct file *file) struct seq_file *seq; int ret; - ret = seq_open(file, &ipoib_seq_ops); + ret = seq_open(file, &ipoib_mcg_seq_ops); if (ret) return ret; @@ -134,7 +146,7 @@ static int ipoib_mcg_open(struct inode *inode, struct file *file) return 0; } -static struct file_operations ipoib_fops = { +static struct file_operations ipoib_mcg_fops = { .owner = THIS_MODULE, .open = ipoib_mcg_open, .read = seq_read, @@ -142,25 +154,138 @@ static struct file_operations ipoib_fops = { .release = seq_release }; -int ipoib_create_debug_file(struct net_device *dev) +static void *ipoib_path_seq_start(struct seq_file *file, loff_t *pos) +{ + struct ipoib_path_iter *iter; + loff_t n = *pos; + + iter = ipoib_path_iter_init(file->private); + if (!iter) + return NULL; + + while (n--) { + if (ipoib_path_iter_next(iter)) { + kfree(iter); + return NULL; + } + } + + return iter; +} + +static void *ipoib_path_seq_next(struct seq_file *file, void *iter_ptr, + loff_t *pos) +{ + struct ipoib_path_iter *iter = iter_ptr; + + (*pos)++; + + if (ipoib_path_iter_next(iter)) { + kfree(iter); + return NULL; + } + + return iter; +} + +static void ipoib_path_seq_stop(struct seq_file *file, void *iter_ptr) +{ + /* nothing for now */ +} + +static int ipoib_path_seq_show(struct seq_file *file, void *iter_ptr) +{ + struct ipoib_path_iter *iter = iter_ptr; + char gid_buf[sizeof "ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff"]; + struct ipoib_path path; + int rate; + + if (!iter) + return 0; + + ipoib_path_iter_read(iter, &path); + + format_gid(&path.pathrec.dgid, gid_buf); + + seq_printf(file, + "GID: %s\n" + " complete: %6s\n", + gid_buf, path.pathrec.dlid ? "yes" : "no"); + + if (path.pathrec.dlid) { + rate = ib_sa_rate_enum_to_int(path.pathrec.rate) * 25; + + seq_printf(file, + " DLID: 0x%04x\n" + " SL: %12d\n" + " rate: %*d%s Gb/sec\n", + be16_to_cpu(path.pathrec.dlid), + path.pathrec.sl, + 10 - ((rate % 10) ? 2 : 0), + rate / 10, rate % 10 ? ".5" : ""); + } + + seq_putc(file, '\n'); + + return 0; +} + +static struct seq_operations ipoib_path_seq_ops = { + .start = ipoib_path_seq_start, + .next = ipoib_path_seq_next, + .stop = ipoib_path_seq_stop, + .show = ipoib_path_seq_show, +}; + +static int ipoib_path_open(struct inode *inode, struct file *file) +{ + struct seq_file *seq; + int ret; + + ret = seq_open(file, &ipoib_path_seq_ops); + if (ret) + return ret; + + seq = file->private_data; + seq->private = inode->u.generic_ip; + + return 0; +} + +static struct file_operations ipoib_path_fops = { + .owner = THIS_MODULE, + .open = ipoib_path_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release +}; + +void ipoib_create_debug_files(struct net_device *dev) { struct ipoib_dev_priv *priv = netdev_priv(dev); - char name[IFNAMSIZ + sizeof "_mcg"]; + char name[IFNAMSIZ + sizeof "_path"]; snprintf(name, sizeof name, "%s_mcg", dev->name); - priv->mcg_dentry = debugfs_create_file(name, S_IFREG | S_IRUGO, - ipoib_root, dev, &ipoib_fops); - - return priv->mcg_dentry ? 0 : -ENOMEM; + ipoib_root, dev, &ipoib_mcg_fops); + if (!priv->mcg_dentry) + ipoib_warn(priv, "failed to create mcg debug file\n"); + + snprintf(name, sizeof name, "%s_path", dev->name); + priv->path_dentry = debugfs_create_file(name, S_IFREG | S_IRUGO, + ipoib_root, dev, &ipoib_path_fops); + if (!priv->path_dentry) + ipoib_warn(priv, "failed to create path debug file\n"); } -void ipoib_delete_debug_file(struct net_device *dev) +void ipoib_delete_debug_files(struct net_device *dev) { struct ipoib_dev_priv *priv = netdev_priv(dev); if (priv->mcg_dentry) debugfs_remove(priv->mcg_dentry); + if (priv->path_dentry) + debugfs_remove(priv->path_dentry); } int ipoib_register_debugfs(void) diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index ce0296273e7..2fa30751f36 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -58,6 +58,11 @@ module_param_named(debug_level, ipoib_debug_level, int, 0644); MODULE_PARM_DESC(debug_level, "Enable debug tracing if > 0"); #endif +struct ipoib_path_iter { + struct net_device *dev; + struct ipoib_path path; +}; + static const u8 ipv4_bcast_addr[] = { 0x00, 0xff, 0xff, 0xff, 0xff, 0x12, 0x40, 0x1b, 0x00, 0x00, 0x00, 0x00, @@ -250,6 +255,64 @@ static void path_free(struct net_device *dev, struct ipoib_path *path) kfree(path); } +#ifdef CONFIG_INFINIBAND_IPOIB_DEBUG + +struct ipoib_path_iter *ipoib_path_iter_init(struct net_device *dev) +{ + struct ipoib_path_iter *iter; + + iter = kmalloc(sizeof *iter, GFP_KERNEL); + if (!iter) + return NULL; + + iter->dev = dev; + memset(iter->path.pathrec.dgid.raw, 0, 16); + + if (ipoib_path_iter_next(iter)) { + kfree(iter); + return NULL; + } + + return iter; +} + +int ipoib_path_iter_next(struct ipoib_path_iter *iter) +{ + struct ipoib_dev_priv *priv = netdev_priv(iter->dev); + struct rb_node *n; + struct ipoib_path *path; + int ret = 1; + + spin_lock_irq(&priv->lock); + + n = rb_first(&priv->path_tree); + + while (n) { + path = rb_entry(n, struct ipoib_path, rb_node); + + if (memcmp(iter->path.pathrec.dgid.raw, path->pathrec.dgid.raw, + sizeof (union ib_gid)) < 0) { + iter->path = *path; + ret = 0; + break; + } + + n = rb_next(n); + } + + spin_unlock_irq(&priv->lock); + + return ret; +} + +void ipoib_path_iter_read(struct ipoib_path_iter *iter, + struct ipoib_path *path) +{ + *path = iter->path; +} + +#endif /* CONFIG_INFINIBAND_IPOIB_DEBUG */ + void ipoib_flush_paths(struct net_device *dev) { struct ipoib_dev_priv *priv = netdev_priv(dev); @@ -763,7 +826,7 @@ void ipoib_dev_cleanup(struct net_device *dev) { struct ipoib_dev_priv *priv = netdev_priv(dev), *cpriv, *tcpriv; - ipoib_delete_debug_file(dev); + ipoib_delete_debug_files(dev); /* Delete any child interfaces first */ list_for_each_entry_safe(cpriv, tcpriv, &priv->child_intfs, list) { @@ -972,8 +1035,7 @@ static struct net_device *ipoib_add_port(const char *format, goto register_failed; } - if (ipoib_create_debug_file(priv->dev)) - goto debug_failed; + ipoib_create_debug_files(priv->dev); if (ipoib_add_pkey_attr(priv->dev)) goto sysfs_failed; @@ -987,9 +1049,7 @@ static struct net_device *ipoib_add_port(const char *format, return priv->dev; sysfs_failed: - ipoib_delete_debug_file(priv->dev); - -debug_failed: + ipoib_delete_debug_files(priv->dev); unregister_netdev(priv->dev); register_failed: diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c index 3ecf78a9493..c33ed87f9df 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c @@ -120,12 +120,8 @@ static void ipoib_mcast_free(struct ipoib_mcast *mcast) if (mcast->ah) ipoib_put_ah(mcast->ah); - while (!skb_queue_empty(&mcast->pkt_queue)) { - struct sk_buff *skb = skb_dequeue(&mcast->pkt_queue); - - skb->dev = dev; - dev_kfree_skb_any(skb); - } + while (!skb_queue_empty(&mcast->pkt_queue)) + dev_kfree_skb_any(skb_dequeue(&mcast->pkt_queue)); kfree(mcast); } @@ -317,13 +313,8 @@ ipoib_mcast_sendonly_join_complete(int status, IPOIB_GID_ARG(mcast->mcmember.mgid), status); /* Flush out any queued packets */ - while (!skb_queue_empty(&mcast->pkt_queue)) { - struct sk_buff *skb = skb_dequeue(&mcast->pkt_queue); - - skb->dev = dev; - - dev_kfree_skb_any(skb); - } + while (!skb_queue_empty(&mcast->pkt_queue)) + dev_kfree_skb_any(skb_dequeue(&mcast->pkt_queue)); /* Clear the busy flag so we try again */ clear_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags); @@ -928,21 +919,16 @@ struct ipoib_mcast_iter *ipoib_mcast_iter_init(struct net_device *dev) return NULL; iter->dev = dev; - memset(iter->mgid.raw, 0, sizeof iter->mgid); + memset(iter->mgid.raw, 0, 16); if (ipoib_mcast_iter_next(iter)) { - ipoib_mcast_iter_free(iter); + kfree(iter); return NULL; } return iter; } -void ipoib_mcast_iter_free(struct ipoib_mcast_iter *iter) -{ - kfree(iter); -} - int ipoib_mcast_iter_next(struct ipoib_mcast_iter *iter) { struct ipoib_dev_priv *priv = netdev_priv(iter->dev); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c index 332d730e60c..d280b341a37 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c @@ -113,8 +113,7 @@ int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey) priv->parent = ppriv->dev; - if (ipoib_create_debug_file(priv->dev)) - goto debug_failed; + ipoib_create_debug_files(priv->dev); if (ipoib_add_pkey_attr(priv->dev)) goto sysfs_failed; @@ -130,9 +129,7 @@ int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey) return 0; sysfs_failed: - ipoib_delete_debug_file(priv->dev); - -debug_failed: + ipoib_delete_debug_files(priv->dev); unregister_netdev(priv->dev); register_failed: diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c index 8f464271664..49fa1e4413f 100644 --- a/drivers/net/bnx2.c +++ b/drivers/net/bnx2.c @@ -2707,7 +2707,7 @@ bnx2_init_nvram(struct bnx2 *bp) if (j == entry_count) { bp->flash_info = NULL; - printk(KERN_ALERT "Unknown flash/EEPROM type.\n"); + printk(KERN_ALERT PFX "Unknown flash/EEPROM type.\n"); rc = -ENODEV; } @@ -3903,6 +3903,8 @@ bnx2_test_loopback(struct bnx2 *bp) pkt_size = 1514; skb = dev_alloc_skb(pkt_size); + if (!skb) + return -ENOMEM; packet = skb_put(skb, pkt_size); memcpy(packet, bp->mac_addr, 6); memset(packet + 6, 0x0, 8); @@ -4798,11 +4800,7 @@ bnx2_get_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom, struct bnx2 *bp = dev->priv; int rc; - if (eeprom->offset > bp->flash_info->total_size) - return -EINVAL; - - if ((eeprom->offset + eeprom->len) > bp->flash_info->total_size) - eeprom->len = bp->flash_info->total_size - eeprom->offset; + /* parameters already validated in ethtool_get_eeprom */ rc = bnx2_nvram_read(bp, eeprom->offset, eebuf, eeprom->len); @@ -4816,11 +4814,7 @@ bnx2_set_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom, struct bnx2 *bp = dev->priv; int rc; - if (eeprom->offset > bp->flash_info->total_size) - return -EINVAL; - - if ((eeprom->offset + eeprom->len) > bp->flash_info->total_size) - eeprom->len = bp->flash_info->total_size - eeprom->offset; + /* parameters already validated in ethtool_set_eeprom */ rc = bnx2_nvram_write(bp, eeprom->offset, eebuf, eeprom->len); diff --git a/drivers/net/fs_enet/fs_enet-main.c b/drivers/net/fs_enet/fs_enet-main.c index 9342d5bc7bb..f5d49a11065 100644 --- a/drivers/net/fs_enet/fs_enet-main.c +++ b/drivers/net/fs_enet/fs_enet-main.c @@ -37,6 +37,7 @@ #include <linux/ethtool.h> #include <linux/bitops.h> #include <linux/fs.h> +#include <linux/platform_device.h> #include <linux/vmalloc.h> #include <asm/pgtable.h> diff --git a/drivers/net/fs_enet/mac-fcc.c b/drivers/net/fs_enet/mac-fcc.c index a940b96433c..e67b1d06611 100644 --- a/drivers/net/fs_enet/mac-fcc.c +++ b/drivers/net/fs_enet/mac-fcc.c @@ -34,6 +34,7 @@ #include <linux/ethtool.h> #include <linux/bitops.h> #include <linux/fs.h> +#include <linux/platform_device.h> #include <asm/immap_cpm2.h> #include <asm/mpc8260.h> diff --git a/drivers/net/fs_enet/mac-fec.c b/drivers/net/fs_enet/mac-fec.c index 5ef4e845a38..2e8f4446969 100644 --- a/drivers/net/fs_enet/mac-fec.c +++ b/drivers/net/fs_enet/mac-fec.c @@ -34,6 +34,7 @@ #include <linux/ethtool.h> #include <linux/bitops.h> #include <linux/fs.h> +#include <linux/platform_device.h> #include <asm/irq.h> #include <asm/uaccess.h> diff --git a/drivers/net/fs_enet/mac-scc.c b/drivers/net/fs_enet/mac-scc.c index d8c6e9cadcf..a3897fda71f 100644 --- a/drivers/net/fs_enet/mac-scc.c +++ b/drivers/net/fs_enet/mac-scc.c @@ -34,6 +34,7 @@ #include <linux/ethtool.h> #include <linux/bitops.h> #include <linux/fs.h> +#include <linux/platform_device.h> #include <asm/irq.h> #include <asm/uaccess.h> diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index c796f41b4a5..0d765f1733b 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -2290,7 +2290,6 @@ spider_net_remove(struct pci_dev *pdev) } static struct pci_driver spider_net_driver = { - .owner = THIS_MODULE, .name = spider_net_driver_name, .id_table = spider_net_pci_tbl, .probe = spider_net_probe, diff --git a/drivers/pci/access.c b/drivers/pci/access.c index 2a42add7f56..ea16805a153 100644 --- a/drivers/pci/access.c +++ b/drivers/pci/access.c @@ -2,6 +2,8 @@ #include <linux/module.h> #include <linux/ioport.h> +#include "pci.h" + /* * This interrupt-safe spinlock protects all accesses to PCI * configuration space. diff --git a/drivers/pci/hotplug/pciehp.h b/drivers/pci/hotplug/pciehp.h index 061ead21ef1..c42b68d3aa2 100644 --- a/drivers/pci/hotplug/pciehp.h +++ b/drivers/pci/hotplug/pciehp.h @@ -32,8 +32,6 @@ #include <linux/types.h> #include <linux/pci.h> #include <linux/delay.h> -#include <asm/semaphore.h> -#include <asm/io.h> #include <linux/pcieport_if.h> #include "pci_hotplug.h" @@ -42,6 +40,7 @@ extern int pciehp_poll_mode; extern int pciehp_poll_time; extern int pciehp_debug; +extern int pciehp_force; /*#define dbg(format, arg...) do { if (pciehp_debug) printk(KERN_DEBUG "%s: " format, MY_NAME , ## arg); } while (0)*/ #define dbg(format, arg...) do { if (pciehp_debug) printk("%s: " format, MY_NAME , ## arg); } while (0) @@ -49,39 +48,20 @@ extern int pciehp_debug; #define info(format, arg...) printk(KERN_INFO "%s: " format, MY_NAME , ## arg) #define warn(format, arg...) printk(KERN_WARNING "%s: " format, MY_NAME , ## arg) -struct pci_func { - struct pci_func *next; - u8 bus; - u8 device; - u8 function; - u8 is_a_board; - u16 status; - u8 configured; - u8 switch_save; - u8 presence_save; - u32 base_length[0x06]; - u8 base_type[0x06]; - u16 reserved2; - u32 config_space[0x20]; - struct pci_resource *mem_head; - struct pci_resource *p_mem_head; - struct pci_resource *io_head; - struct pci_resource *bus_head; - struct pci_dev* pci_dev; +struct hotplug_params { + u8 cache_line_size; + u8 latency_timer; + u8 enable_serr; + u8 enable_perr; }; struct slot { struct slot *next; u8 bus; u8 device; + u16 status; u32 number; - u8 is_a_board; - u8 configured; u8 state; - u8 switch_save; - u8 presence_save; - u32 capabilities; - u16 reserved2; struct timer_list task_event; u8 hp_slot; struct controller *ctrl; @@ -90,42 +70,47 @@ struct slot { struct list_head slot_list; }; -struct pci_resource { - struct pci_resource * next; - u32 base; - u32 length; -}; - struct event_info { u32 event_type; u8 hp_slot; }; +typedef u8(*php_intr_callback_t) (u8 hp_slot, void *instance_id); + +struct php_ctlr_state_s { + struct php_ctlr_state_s *pnext; + struct pci_dev *pci_dev; + unsigned int irq; + unsigned long flags; /* spinlock's */ + u32 slot_device_offset; + u32 num_slots; + struct timer_list int_poll_timer; /* Added for poll event */ + php_intr_callback_t attention_button_callback; + php_intr_callback_t switch_change_callback; + php_intr_callback_t presence_change_callback; + php_intr_callback_t power_fault_callback; + void *callback_instance_id; + struct ctrl_reg *creg; /* Ptr to controller register space */ +}; + +#define MAX_EVENTS 10 struct controller { struct controller *next; struct semaphore crit_sect; /* critical section semaphore */ - void *hpc_ctlr_handle; /* HPC controller handle */ + struct php_ctlr_state_s *hpc_ctlr_handle; /* HPC controller handle */ int num_slots; /* Number of slots on ctlr */ int slot_num_inc; /* 1 or -1 */ - struct pci_resource *mem_head; - struct pci_resource *p_mem_head; - struct pci_resource *io_head; - struct pci_resource *bus_head; struct pci_dev *pci_dev; struct pci_bus *pci_bus; - struct event_info event_queue[10]; + struct event_info event_queue[MAX_EVENTS]; struct slot *slot; struct hpc_ops *hpc_ops; wait_queue_head_t queue; /* sleep & wake process */ u8 next_event; - u8 seg; u8 bus; u8 device; u8 function; - u8 rev; u8 slot_device_offset; - u8 add_support; - enum pci_bus_speed speed; u32 first_slot; /* First physical slot number */ /* PCIE only has 1 slot */ u8 slot_bus; /* Bus where the slots handled by this controller sit */ u8 ctrlcap; @@ -133,20 +118,6 @@ struct controller { u8 cap_base; }; -struct irq_mapping { - u8 barber_pole; - u8 valid_INT; - u8 interrupt[4]; -}; - -struct resource_lists { - struct pci_resource *mem_head; - struct pci_resource *p_mem_head; - struct pci_resource *io_head; - struct pci_resource *bus_head; - struct irq_mapping *irqs; -}; - #define INT_BUTTON_IGNORE 0 #define INT_PRESENCE_ON 1 #define INT_PRESENCE_OFF 2 @@ -200,21 +171,14 @@ struct resource_lists { * error Messages */ #define msg_initialization_err "Initialization failure, error=%d\n" -#define msg_HPC_rev_error "Unsupported revision of the PCI hot plug controller found.\n" -#define msg_HPC_non_pcie "The PCI hot plug controller is not supported by this driver.\n" -#define msg_HPC_not_supported "This system is not supported by this version of pciephd module. Upgrade to a newer version of pciehpd\n" -#define msg_unable_to_save "Unable to store PCI hot plug add resource information. This system must be rebooted before adding any PCI devices.\n" #define msg_button_on "PCI slot #%d - powering on due to button press.\n" #define msg_button_off "PCI slot #%d - powering off due to button press.\n" #define msg_button_cancel "PCI slot #%d - action canceled due to button press.\n" #define msg_button_ignore "PCI slot #%d - button press ignored. (action in progress...)\n" /* controller functions */ -extern int pciehprm_find_available_resources (struct controller *ctrl); extern int pciehp_event_start_thread (void); extern void pciehp_event_stop_thread (void); -extern struct pci_func *pciehp_slot_create (unsigned char busnumber); -extern struct pci_func *pciehp_slot_find (unsigned char bus, unsigned char device, unsigned char index); extern int pciehp_enable_slot (struct slot *slot); extern int pciehp_disable_slot (struct slot *slot); @@ -224,25 +188,17 @@ extern u8 pciehp_handle_presence_change (u8 hp_slot, void *inst_id); extern u8 pciehp_handle_power_fault (u8 hp_slot, void *inst_id); /* extern void long_delay (int delay); */ -/* resource functions */ -extern int pciehp_resource_sort_and_combine (struct pci_resource **head); - /* pci functions */ -extern int pciehp_set_irq (u8 bus_num, u8 dev_num, u8 int_pin, u8 irq_num); -/*extern int pciehp_get_bus_dev (struct controller *ctrl, u8 *bus_num, u8 *dev_num, struct slot *slot);*/ -extern int pciehp_save_config (struct controller *ctrl, int busnumber, int num_ctlr_slots, int first_device_num); -extern int pciehp_save_used_resources (struct controller *ctrl, struct pci_func * func, int flag); -extern int pciehp_save_slot_config (struct controller *ctrl, struct pci_func * new_slot); -extern void pciehp_destroy_board_resources (struct pci_func * func); -extern int pciehp_return_board_resources (struct pci_func * func, struct resource_lists * resources); -extern void pciehp_destroy_resource_list (struct resource_lists * resources); -extern int pciehp_configure_device (struct controller* ctrl, struct pci_func* func); -extern int pciehp_unconfigure_device (struct pci_func* func); +extern int pciehp_configure_device (struct slot *p_slot); +extern int pciehp_unconfigure_device (struct slot *p_slot); +extern int pciehp_get_hp_hw_control_from_firmware(struct pci_dev *dev); +extern void pciehp_get_hp_params_from_firmware(struct pci_dev *dev, + struct hotplug_params *hpp); + /* Global variables */ extern struct controller *pciehp_ctrl_list; -extern struct pci_func *pciehp_slot_list[256]; /* Inline functions */ @@ -252,12 +208,9 @@ static inline struct slot *pciehp_find_slot(struct controller *ctrl, u8 device) p_slot = ctrl->slot; - dbg("p_slot = %p\n", p_slot); - while (p_slot && (p_slot->device != device)) { tmp_slot = p_slot; p_slot = p_slot->next; - dbg("In while loop, p_slot = %p\n", p_slot); } if (p_slot == NULL) { err("ERROR: pciehp_find_slot device=0x%x\n", device); @@ -273,7 +226,6 @@ static inline int wait_for_ctrl_irq(struct controller *ctrl) DECLARE_WAITQUEUE(wait, current); - dbg("%s : start\n", __FUNCTION__); add_wait_queue(&ctrl->queue, &wait); if (!pciehp_poll_mode) /* Sleep for up to 1 second */ @@ -285,19 +237,9 @@ static inline int wait_for_ctrl_irq(struct controller *ctrl) if (signal_pending(current)) retval = -EINTR; - dbg("%s : end\n", __FUNCTION__); return retval; } -/* Puts node back in the resource list pointed to by head */ -static inline void return_resource(struct pci_resource **head, struct pci_resource *node) -{ - if (!node || !head) - return; - node->next = *head; - *head = node; -} - #define SLOT_NAME_SIZE 10 static inline void make_slot_name(char *buffer, int buffer_size, struct slot *slot) @@ -311,14 +253,7 @@ enum php_ctlr_type { ACPI }; -typedef u8(*php_intr_callback_t) (unsigned int change_id, void *instance_id); - -int pcie_init(struct controller *ctrl, struct pcie_device *dev, - php_intr_callback_t attention_button_callback, - php_intr_callback_t switch_change_callback, - php_intr_callback_t presence_change_callback, - php_intr_callback_t power_fault_callback); - +int pcie_init(struct controller *ctrl, struct pcie_device *dev); /* This has no meaning for PCI Express, as there is only 1 slot per port */ int pcie_get_ctlr_slot_config(struct controller *ctrl, diff --git a/drivers/pci/hotplug/pciehp_core.c b/drivers/pci/hotplug/pciehp_core.c index cafc7eadcf8..8df70486034 100644 --- a/drivers/pci/hotplug/pciehp_core.c +++ b/drivers/pci/hotplug/pciehp_core.c @@ -27,27 +27,20 @@ * */ -#include <linux/config.h> #include <linux/module.h> #include <linux/moduleparam.h> #include <linux/kernel.h> #include <linux/types.h> -#include <linux/proc_fs.h> -#include <linux/slab.h> -#include <linux/workqueue.h> #include <linux/pci.h> -#include <linux/init.h> -#include <asm/uaccess.h> #include "pciehp.h" -#include "pciehprm.h" #include <linux/interrupt.h> /* Global variables */ int pciehp_debug; int pciehp_poll_mode; int pciehp_poll_time; +int pciehp_force; struct controller *pciehp_ctrl_list; -struct pci_func *pciehp_slot_list[256]; #define DRIVER_VERSION "0.4" #define DRIVER_AUTHOR "Dan Zink <dan.zink@compaq.com>, Greg Kroah-Hartman <greg@kroah.com>, Dely Sy <dely.l.sy@intel.com>" @@ -60,9 +53,11 @@ MODULE_LICENSE("GPL"); module_param(pciehp_debug, bool, 0644); module_param(pciehp_poll_mode, bool, 0644); module_param(pciehp_poll_time, int, 0644); +module_param(pciehp_force, bool, 0644); MODULE_PARM_DESC(pciehp_debug, "Debugging mode enabled or not"); MODULE_PARM_DESC(pciehp_poll_mode, "Using polling mechanism for hot-plug events or not"); MODULE_PARM_DESC(pciehp_poll_time, "Polling mechanism frequency, in seconds"); +MODULE_PARM_DESC(pciehp_force, "Force pciehp, even if _OSC and OSHP are missing"); #define PCIE_MODULE_NAME "pciehp" @@ -114,8 +109,6 @@ static int init_slots(struct controller *ctrl) u32 slot_number; int result = -ENOMEM; - dbg("%s\n",__FUNCTION__); - number_of_slots = ctrl->num_slots; slot_device = ctrl->slot_device_offset; slot_number = ctrl->first_slot; @@ -370,7 +363,6 @@ static int pciehp_probe(struct pcie_device *dev, const struct pcie_port_service_ u8 value; struct pci_dev *pdev; - dbg("%s: Called by hp_drv\n", __FUNCTION__); ctrl = kmalloc(sizeof(*ctrl), GFP_KERNEL); if (!ctrl) { err("%s : out of memory\n", __FUNCTION__); @@ -378,22 +370,15 @@ static int pciehp_probe(struct pcie_device *dev, const struct pcie_port_service_ } memset(ctrl, 0, sizeof(struct controller)); - dbg("%s: DRV_thread pid = %d\n", __FUNCTION__, current->pid); - pdev = dev->port; + ctrl->pci_dev = pdev; - rc = pcie_init(ctrl, dev, - (php_intr_callback_t) pciehp_handle_attention_button, - (php_intr_callback_t) pciehp_handle_switch_change, - (php_intr_callback_t) pciehp_handle_presence_change, - (php_intr_callback_t) pciehp_handle_power_fault); + rc = pcie_init(ctrl, dev); if (rc) { dbg("%s: controller initialization failed\n", PCIE_MODULE_NAME); goto err_out_free_ctrl; } - ctrl->pci_dev = pdev; - pci_set_drvdata(pdev, ctrl); ctrl->pci_bus = kmalloc(sizeof(*ctrl->pci_bus), GFP_KERNEL); @@ -402,7 +387,6 @@ static int pciehp_probe(struct pcie_device *dev, const struct pcie_port_service_ rc = -ENOMEM; goto err_out_unmap_mmio_region; } - dbg("%s: ctrl->pci_bus %p\n", __FUNCTION__, ctrl->pci_bus); memcpy (ctrl->pci_bus, pdev->bus, sizeof (*ctrl->pci_bus)); ctrl->bus = pdev->bus->number; /* ctrl bus */ ctrl->slot_bus = pdev->subordinate->number; /* bus controlled by this HPC */ @@ -424,25 +408,6 @@ static int pciehp_probe(struct pcie_device *dev, const struct pcie_port_service_ first_device_num = ctrl->slot_device_offset; num_ctlr_slots = ctrl->num_slots; - /* Store PCI Config Space for all devices on this bus */ - dbg("%s: Before calling pciehp_save_config, ctrl->bus %x,ctrl->slot_bus %x\n", - __FUNCTION__,ctrl->bus, ctrl->slot_bus); - rc = pciehp_save_config(ctrl, ctrl->slot_bus, num_ctlr_slots, first_device_num); - if (rc) { - err("%s: unable to save PCI configuration data, error %d\n", __FUNCTION__, rc); - goto err_out_free_ctrl_bus; - } - - /* Get IO, memory, and IRQ resources for new devices */ - rc = pciehprm_find_available_resources(ctrl); - ctrl->add_support = !rc; - - if (rc) { - dbg("pciehprm_find_available_resources = %#x\n", rc); - err("unable to locate PCI configuration resources for hot plug add.\n"); - goto err_out_free_ctrl_bus; - } - /* Setup the slot information structures */ rc = init_slots(ctrl); if (rc) { @@ -451,7 +416,6 @@ static int pciehp_probe(struct pcie_device *dev, const struct pcie_port_service_ } t_slot = pciehp_find_slot(ctrl, first_device_num); - dbg("%s: t_slot %p\n", __FUNCTION__, t_slot); /* Finish setting up the hot plug ctrl device */ ctrl->next_event = 0; @@ -468,7 +432,6 @@ static int pciehp_probe(struct pcie_device *dev, const struct pcie_port_service_ down(&ctrl->crit_sect); t_slot->hpc_ops->get_adapter_status(t_slot, &value); /* Check if slot is occupied */ - dbg("%s: adpater value %x\n", __FUNCTION__, value); if ((POWER_CTRL(ctrl->ctrlcap)) && !value) { rc = t_slot->hpc_ops->power_off_slot(t_slot); /* Power off slot if not occupied*/ @@ -501,7 +464,6 @@ err_out_none: static int pcie_start_thread(void) { - int loop; int retval = 0; dbg("Initialize + Start the notification/polling mechanism \n"); @@ -512,32 +474,11 @@ static int pcie_start_thread(void) return retval; } - dbg("Initialize slot lists\n"); - /* One slot list for each bus in the system */ - for (loop = 0; loop < 256; loop++) { - pciehp_slot_list[loop] = NULL; - } - return retval; } -static inline void __exit -free_pciehp_res(struct pci_resource *res) -{ - struct pci_resource *tres; - - while (res) { - tres = res; - res = res->next; - kfree(tres); - } -} - static void __exit unload_pciehpd(void) { - struct pci_func *next; - struct pci_func *TempSlot; - int loop; struct controller *ctrl; struct controller *tctrl; @@ -546,11 +487,6 @@ static void __exit unload_pciehpd(void) while (ctrl) { cleanup_slots(ctrl); - free_pciehp_res(ctrl->io_head); - free_pciehp_res(ctrl->mem_head); - free_pciehp_res(ctrl->p_mem_head); - free_pciehp_res(ctrl->bus_head); - kfree (ctrl->pci_bus); ctrl->hpc_ops->release_ctlr(ctrl); @@ -561,20 +497,6 @@ static void __exit unload_pciehpd(void) kfree(tctrl); } - for (loop = 0; loop < 256; loop++) { - next = pciehp_slot_list[loop]; - while (next != NULL) { - free_pciehp_res(next->io_head); - free_pciehp_res(next->mem_head); - free_pciehp_res(next->p_mem_head); - free_pciehp_res(next->bus_head); - - TempSlot = next; - next = next->next; - kfree(TempSlot); - } - } - /* Stop the notification mechanism */ pciehp_event_stop_thread(); @@ -639,21 +561,16 @@ static int __init pcied_init(void) if (retval) goto error_hpc_init; - retval = pciehprm_init(PCI); - if (!retval) { - retval = pcie_port_service_register(&hpdriver_portdrv); - dbg("pcie_port_service_register = %d\n", retval); - info(DRIVER_DESC " version: " DRIVER_VERSION "\n"); - if (retval) - dbg("%s: Failure to register service\n", __FUNCTION__); - } + retval = pcie_port_service_register(&hpdriver_portdrv); + dbg("pcie_port_service_register = %d\n", retval); + info(DRIVER_DESC " version: " DRIVER_VERSION "\n"); + if (retval) + dbg("%s: Failure to register service\n", __FUNCTION__); error_hpc_init: if (retval) { - pciehprm_cleanup(); pciehp_event_stop_thread(); - } else - pciehprm_print_pirt(); + }; return retval; } @@ -663,9 +580,6 @@ static void __exit pcied_cleanup(void) dbg("unload_pciehpd()\n"); unload_pciehpd(); - pciehprm_cleanup(); - - dbg("pcie_port_service_unregister\n"); pcie_port_service_unregister(&hpdriver_portdrv); info(DRIVER_DESC " version: " DRIVER_VERSION " unloaded\n"); diff --git a/drivers/pci/hotplug/pciehp_ctrl.c b/drivers/pci/hotplug/pciehp_ctrl.c index 898f6da6f0d..5e582eca21d 100644 --- a/drivers/pci/hotplug/pciehp_ctrl.c +++ b/drivers/pci/hotplug/pciehp_ctrl.c @@ -27,25 +27,14 @@ * */ -#include <linux/config.h> #include <linux/module.h> #include <linux/kernel.h> #include <linux/types.h> -#include <linux/slab.h> -#include <linux/workqueue.h> -#include <linux/interrupt.h> -#include <linux/delay.h> -#include <linux/wait.h> #include <linux/smp_lock.h> #include <linux/pci.h> #include "../pci.h" #include "pciehp.h" -#include "pciehprm.h" -static u32 configure_new_device(struct controller *ctrl, struct pci_func *func, - u8 behind_bridge, struct resource_lists *resources, u8 bridge_bus, u8 bridge_dev); -static int configure_new_function( struct controller *ctrl, struct pci_func *func, - u8 behind_bridge, struct resource_lists *resources, u8 bridge_bus, u8 bridge_dev); static void interrupt_event_handler(struct controller *ctrl); static struct semaphore event_semaphore; /* mutex for process loop (up if something to process) */ @@ -60,22 +49,18 @@ u8 pciehp_handle_attention_button(u8 hp_slot, void *inst_id) struct slot *p_slot; u8 rc = 0; u8 getstatus; - struct pci_func *func; struct event_info *taskInfo; /* Attention Button Change */ dbg("pciehp: Attention button interrupt received.\n"); - func = pciehp_slot_find(ctrl->slot_bus, (hp_slot + ctrl->slot_device_offset), 0); - /* This is the structure that tells the worker thread what to do */ taskInfo = &(ctrl->event_queue[ctrl->next_event]); p_slot = pciehp_find_slot(ctrl, hp_slot + ctrl->slot_device_offset); - p_slot->hpc_ops->get_adapter_status(p_slot, &(func->presence_save)); p_slot->hpc_ops->get_latch_status(p_slot, &getstatus); - ctrl->next_event = (ctrl->next_event + 1) % 10; + ctrl->next_event = (ctrl->next_event + 1) % MAX_EVENTS; taskInfo->hp_slot = hp_slot; rc++; @@ -117,24 +102,20 @@ u8 pciehp_handle_switch_change(u8 hp_slot, void *inst_id) struct slot *p_slot; u8 rc = 0; u8 getstatus; - struct pci_func *func; struct event_info *taskInfo; /* Switch Change */ dbg("pciehp: Switch interrupt received.\n"); - func = pciehp_slot_find(ctrl->slot_bus, (hp_slot + ctrl->slot_device_offset), 0); - /* This is the structure that tells the worker thread * what to do */ taskInfo = &(ctrl->event_queue[ctrl->next_event]); - ctrl->next_event = (ctrl->next_event + 1) % 10; + ctrl->next_event = (ctrl->next_event + 1) % MAX_EVENTS; taskInfo->hp_slot = hp_slot; rc++; p_slot = pciehp_find_slot(ctrl, hp_slot + ctrl->slot_device_offset); - p_slot->hpc_ops->get_adapter_status(p_slot, &(func->presence_save)); p_slot->hpc_ops->get_latch_status(p_slot, &getstatus); if (getstatus) { @@ -142,14 +123,12 @@ u8 pciehp_handle_switch_change(u8 hp_slot, void *inst_id) * Switch opened */ info("Latch open on Slot(%d)\n", ctrl->first_slot + hp_slot); - func->switch_save = 0; taskInfo->event_type = INT_SWITCH_OPEN; } else { /* * Switch closed */ info("Latch close on Slot(%d)\n", ctrl->first_slot + hp_slot); - func->switch_save = 0x10; taskInfo->event_type = INT_SWITCH_CLOSE; } @@ -163,20 +142,17 @@ u8 pciehp_handle_presence_change(u8 hp_slot, void *inst_id) { struct controller *ctrl = (struct controller *) inst_id; struct slot *p_slot; - u8 rc = 0; - struct pci_func *func; + u8 presence_save, rc = 0; struct event_info *taskInfo; /* Presence Change */ dbg("pciehp: Presence/Notify input change.\n"); - func = pciehp_slot_find(ctrl->slot_bus, (hp_slot + ctrl->slot_device_offset), 0); - /* This is the structure that tells the worker thread * what to do */ taskInfo = &(ctrl->event_queue[ctrl->next_event]); - ctrl->next_event = (ctrl->next_event + 1) % 10; + ctrl->next_event = (ctrl->next_event + 1) % MAX_EVENTS; taskInfo->hp_slot = hp_slot; rc++; @@ -185,8 +161,8 @@ u8 pciehp_handle_presence_change(u8 hp_slot, void *inst_id) /* Switch is open, assume a presence change * Save the presence state */ - p_slot->hpc_ops->get_adapter_status(p_slot, &(func->presence_save)); - if (func->presence_save) { + p_slot->hpc_ops->get_adapter_status(p_slot, &presence_save); + if (presence_save) { /* * Card Present */ @@ -211,19 +187,16 @@ u8 pciehp_handle_power_fault(u8 hp_slot, void *inst_id) struct controller *ctrl = (struct controller *) inst_id; struct slot *p_slot; u8 rc = 0; - struct pci_func *func; struct event_info *taskInfo; /* power fault */ dbg("pciehp: Power fault interrupt received.\n"); - func = pciehp_slot_find(ctrl->slot_bus, (hp_slot + ctrl->slot_device_offset), 0); - /* this is the structure that tells the worker thread * what to do */ taskInfo = &(ctrl->event_queue[ctrl->next_event]); - ctrl->next_event = (ctrl->next_event + 1) % 10; + ctrl->next_event = (ctrl->next_event + 1) % MAX_EVENTS; taskInfo->hp_slot = hp_slot; rc++; @@ -234,7 +207,7 @@ u8 pciehp_handle_power_fault(u8 hp_slot, void *inst_id) * power fault Cleared */ info("Power fault cleared on Slot(%d)\n", ctrl->first_slot + hp_slot); - func->status = 0x00; + p_slot->status = 0x00; taskInfo->event_type = INT_POWER_FAULT_CLEAR; } else { /* @@ -243,7 +216,7 @@ u8 pciehp_handle_power_fault(u8 hp_slot, void *inst_id) info("Power fault on Slot(%d)\n", ctrl->first_slot + hp_slot); taskInfo->event_type = INT_POWER_FAULT; /* set power fault status for this board */ - func->status = 0xFF; + p_slot->status = 0xFF; info("power fault bit %x set\n", hp_slot); } if (rc) @@ -252,810 +225,6 @@ u8 pciehp_handle_power_fault(u8 hp_slot, void *inst_id) return rc; } - -/** - * sort_by_size: sort nodes by their length, smallest first. - * - * @head: list to sort - */ -static int sort_by_size(struct pci_resource **head) -{ - struct pci_resource *current_res; - struct pci_resource *next_res; - int out_of_order = 1; - - if (!(*head)) - return 1; - - if (!((*head)->next)) - return 0; - - while (out_of_order) { - out_of_order = 0; - - /* Special case for swapping list head */ - if (((*head)->next) && - ((*head)->length > (*head)->next->length)) { - out_of_order++; - current_res = *head; - *head = (*head)->next; - current_res->next = (*head)->next; - (*head)->next = current_res; - } - - current_res = *head; - - while (current_res->next && current_res->next->next) { - if (current_res->next->length > current_res->next->next->length) { - out_of_order++; - next_res = current_res->next; - current_res->next = current_res->next->next; - current_res = current_res->next; - next_res->next = current_res->next; - current_res->next = next_res; - } else - current_res = current_res->next; - } - } /* End of out_of_order loop */ - - return 0; -} - - -/* - * sort_by_max_size - * - * Sorts nodes on the list by their length. - * Largest first. - * - */ -static int sort_by_max_size(struct pci_resource **head) -{ - struct pci_resource *current_res; - struct pci_resource *next_res; - int out_of_order = 1; - - if (!(*head)) - return 1; - - if (!((*head)->next)) - return 0; - - while (out_of_order) { - out_of_order = 0; - - /* Special case for swapping list head */ - if (((*head)->next) && - ((*head)->length < (*head)->next->length)) { - out_of_order++; - current_res = *head; - *head = (*head)->next; - current_res->next = (*head)->next; - (*head)->next = current_res; - } - - current_res = *head; - - while (current_res->next && current_res->next->next) { - if (current_res->next->length < current_res->next->next->length) { - out_of_order++; - next_res = current_res->next; - current_res->next = current_res->next->next; - current_res = current_res->next; - next_res->next = current_res->next; - current_res->next = next_res; - } else - current_res = current_res->next; - } - } /* End of out_of_order loop */ - - return 0; -} - - -/** - * do_pre_bridge_resource_split: return one unused resource node - * @head: list to scan - * - */ -static struct pci_resource * -do_pre_bridge_resource_split(struct pci_resource **head, - struct pci_resource **orig_head, u32 alignment) -{ - struct pci_resource *prevnode = NULL; - struct pci_resource *node; - struct pci_resource *split_node; - u32 rc; - u32 temp_dword; - dbg("do_pre_bridge_resource_split\n"); - - if (!(*head) || !(*orig_head)) - return NULL; - - rc = pciehp_resource_sort_and_combine(head); - - if (rc) - return NULL; - - if ((*head)->base != (*orig_head)->base) - return NULL; - - if ((*head)->length == (*orig_head)->length) - return NULL; - - - /* If we got here, there the bridge requires some of the resource, but - * we may be able to split some off of the front - */ - node = *head; - - if (node->length & (alignment -1)) { - /* this one isn't an aligned length, so we'll make a new entry - * and split it up. - */ - split_node = kmalloc(sizeof(struct pci_resource), GFP_KERNEL); - - if (!split_node) - return NULL; - - temp_dword = (node->length | (alignment-1)) + 1 - alignment; - - split_node->base = node->base; - split_node->length = temp_dword; - - node->length -= temp_dword; - node->base += split_node->length; - - /* Put it in the list */ - *head = split_node; - split_node->next = node; - } - - if (node->length < alignment) - return NULL; - - /* Now unlink it */ - if (*head == node) { - *head = node->next; - } else { - prevnode = *head; - while (prevnode->next != node) - prevnode = prevnode->next; - - prevnode->next = node->next; - } - node->next = NULL; - - return node; -} - - -/** - * do_bridge_resource_split: return one unused resource node - * @head: list to scan - * - */ -static struct pci_resource * -do_bridge_resource_split(struct pci_resource **head, u32 alignment) -{ - struct pci_resource *prevnode = NULL; - struct pci_resource *node; - u32 rc; - u32 temp_dword; - - if (!(*head)) - return NULL; - - rc = pciehp_resource_sort_and_combine(head); - - if (rc) - return NULL; - - node = *head; - - while (node->next) { - prevnode = node; - node = node->next; - kfree(prevnode); - } - - if (node->length < alignment) { - kfree(node); - return NULL; - } - - if (node->base & (alignment - 1)) { - /* Short circuit if adjusted size is too small */ - temp_dword = (node->base | (alignment-1)) + 1; - if ((node->length - (temp_dword - node->base)) < alignment) { - kfree(node); - return NULL; - } - - node->length -= (temp_dword - node->base); - node->base = temp_dword; - } - - if (node->length & (alignment - 1)) { - /* There's stuff in use after this node */ - kfree(node); - return NULL; - } - - return node; -} - - -/* - * get_io_resource - * - * this function sorts the resource list by size and then - * returns the first node of "size" length that is not in the - * ISA aliasing window. If it finds a node larger than "size" - * it will split it up. - * - * size must be a power of two. - */ -static struct pci_resource *get_io_resource(struct pci_resource **head, u32 size) -{ - struct pci_resource *prevnode; - struct pci_resource *node; - struct pci_resource *split_node = NULL; - u32 temp_dword; - - if (!(*head)) - return NULL; - - if ( pciehp_resource_sort_and_combine(head) ) - return NULL; - - if ( sort_by_size(head) ) - return NULL; - - for (node = *head; node; node = node->next) { - if (node->length < size) - continue; - - if (node->base & (size - 1)) { - /* this one isn't base aligned properly - so we'll make a new entry and split it up */ - temp_dword = (node->base | (size-1)) + 1; - - /*/ Short circuit if adjusted size is too small */ - if ((node->length - (temp_dword - node->base)) < size) - continue; - - split_node = kmalloc(sizeof(struct pci_resource), - GFP_KERNEL); - - if (!split_node) - return NULL; - - split_node->base = node->base; - split_node->length = temp_dword - node->base; - node->base = temp_dword; - node->length -= split_node->length; - - /* Put it in the list */ - split_node->next = node->next; - node->next = split_node; - } /* End of non-aligned base */ - - /* Don't need to check if too small since we already did */ - if (node->length > size) { - /* this one is longer than we need - so we'll make a new entry and split it up */ - split_node = kmalloc(sizeof(struct pci_resource), - GFP_KERNEL); - - if (!split_node) - return NULL; - - split_node->base = node->base + size; - split_node->length = node->length - size; - node->length = size; - - /* Put it in the list */ - split_node->next = node->next; - node->next = split_node; - } /* End of too big on top end */ - - /* For IO make sure it's not in the ISA aliasing space */ - if (node->base & 0x300L) - continue; - - /* If we got here, then it is the right size - Now take it out of the list */ - if (*head == node) { - *head = node->next; - } else { - prevnode = *head; - while (prevnode->next != node) - prevnode = prevnode->next; - - prevnode->next = node->next; - } - node->next = NULL; - /* Stop looping */ - break; - } - - return node; -} - - -/* - * get_max_resource - * - * Gets the largest node that is at least "size" big from the - * list pointed to by head. It aligns the node on top and bottom - * to "size" alignment before returning it. - * J.I. modified to put max size limits of; 64M->32M->16M->8M->4M->1M - * This is needed to avoid allocating entire ACPI _CRS res to one child bridge/slot. - */ -static struct pci_resource *get_max_resource(struct pci_resource **head, u32 size) -{ - struct pci_resource *max; - struct pci_resource *temp; - struct pci_resource *split_node; - u32 temp_dword; - u32 max_size[] = { 0x4000000, 0x2000000, 0x1000000, 0x0800000, 0x0400000, 0x0200000, 0x0100000, 0x00 }; - int i; - - if (!(*head)) - return NULL; - - if (pciehp_resource_sort_and_combine(head)) - return NULL; - - if (sort_by_max_size(head)) - return NULL; - - for (max = *head;max; max = max->next) { - - /* If not big enough we could probably just bail, - instead we'll continue to the next. */ - if (max->length < size) - continue; - - if (max->base & (size - 1)) { - /* this one isn't base aligned properly - so we'll make a new entry and split it up */ - temp_dword = (max->base | (size-1)) + 1; - - /* Short circuit if adjusted size is too small */ - if ((max->length - (temp_dword - max->base)) < size) - continue; - - split_node = kmalloc(sizeof(struct pci_resource), - GFP_KERNEL); - - if (!split_node) - return NULL; - - split_node->base = max->base; - split_node->length = temp_dword - max->base; - max->base = temp_dword; - max->length -= split_node->length; - - /* Put it next in the list */ - split_node->next = max->next; - max->next = split_node; - } - - if ((max->base + max->length) & (size - 1)) { - /* this one isn't end aligned properly at the top - so we'll make a new entry and split it up */ - split_node = kmalloc(sizeof(struct pci_resource), - GFP_KERNEL); - - if (!split_node) - return NULL; - temp_dword = ((max->base + max->length) & ~(size - 1)); - split_node->base = temp_dword; - split_node->length = max->length + max->base - - split_node->base; - max->length -= split_node->length; - - /* Put it in the list */ - split_node->next = max->next; - max->next = split_node; - } - - /* Make sure it didn't shrink too much when we aligned it */ - if (max->length < size) - continue; - - for ( i = 0; max_size[i] > size; i++) { - if (max->length > max_size[i]) { - split_node = kmalloc(sizeof(struct pci_resource), - GFP_KERNEL); - if (!split_node) - break; /* return NULL; */ - split_node->base = max->base + max_size[i]; - split_node->length = max->length - max_size[i]; - max->length = max_size[i]; - /* Put it next in the list */ - split_node->next = max->next; - max->next = split_node; - break; - } - } - - /* Now take it out of the list */ - temp = (struct pci_resource*) *head; - if (temp == max) { - *head = max->next; - } else { - while (temp && temp->next != max) { - temp = temp->next; - } - - temp->next = max->next; - } - - max->next = NULL; - return max; - } - - /* If we get here, we couldn't find one */ - return NULL; -} - - -/* - * get_resource - * - * this function sorts the resource list by size and then - * returns the first node of "size" length. If it finds a node - * larger than "size" it will split it up. - * - * size must be a power of two. - */ -static struct pci_resource *get_resource(struct pci_resource **head, u32 size) -{ - struct pci_resource *prevnode; - struct pci_resource *node; - struct pci_resource *split_node; - u32 temp_dword; - - if (!(*head)) - return NULL; - - if ( pciehp_resource_sort_and_combine(head) ) - return NULL; - - if ( sort_by_size(head) ) - return NULL; - - for (node = *head; node; node = node->next) { - dbg("%s: req_size =0x%x node=%p, base=0x%x, length=0x%x\n", - __FUNCTION__, size, node, node->base, node->length); - if (node->length < size) - continue; - - if (node->base & (size - 1)) { - dbg("%s: not aligned\n", __FUNCTION__); - /* this one isn't base aligned properly - so we'll make a new entry and split it up */ - temp_dword = (node->base | (size-1)) + 1; - - /* Short circuit if adjusted size is too small */ - if ((node->length - (temp_dword - node->base)) < size) - continue; - - split_node = kmalloc(sizeof(struct pci_resource), - GFP_KERNEL); - - if (!split_node) - return NULL; - - split_node->base = node->base; - split_node->length = temp_dword - node->base; - node->base = temp_dword; - node->length -= split_node->length; - - /* Put it in the list */ - split_node->next = node->next; - node->next = split_node; - } /* End of non-aligned base */ - - /* Don't need to check if too small since we already did */ - if (node->length > size) { - dbg("%s: too big\n", __FUNCTION__); - /* this one is longer than we need - so we'll make a new entry and split it up */ - split_node = kmalloc(sizeof(struct pci_resource), - GFP_KERNEL); - - if (!split_node) - return NULL; - - split_node->base = node->base + size; - split_node->length = node->length - size; - node->length = size; - - /* Put it in the list */ - split_node->next = node->next; - node->next = split_node; - } /* End of too big on top end */ - - dbg("%s: got one!!!\n", __FUNCTION__); - /* If we got here, then it is the right size - Now take it out of the list */ - if (*head == node) { - *head = node->next; - } else { - prevnode = *head; - while (prevnode->next != node) - prevnode = prevnode->next; - - prevnode->next = node->next; - } - node->next = NULL; - /* Stop looping */ - break; - } - return node; -} - - -/* - * pciehp_resource_sort_and_combine - * - * Sorts all of the nodes in the list in ascending order by - * their base addresses. Also does garbage collection by - * combining adjacent nodes. - * - * returns 0 if success - */ -int pciehp_resource_sort_and_combine(struct pci_resource **head) -{ - struct pci_resource *node1; - struct pci_resource *node2; - int out_of_order = 1; - - dbg("%s: head = %p, *head = %p\n", __FUNCTION__, head, *head); - - if (!(*head)) - return 1; - - dbg("*head->next = %p\n",(*head)->next); - - if (!(*head)->next) - return 0; /* only one item on the list, already sorted! */ - - dbg("*head->base = 0x%x\n",(*head)->base); - dbg("*head->next->base = 0x%x\n",(*head)->next->base); - while (out_of_order) { - out_of_order = 0; - - /* Special case for swapping list head */ - if (((*head)->next) && - ((*head)->base > (*head)->next->base)) { - node1 = *head; - (*head) = (*head)->next; - node1->next = (*head)->next; - (*head)->next = node1; - out_of_order++; - } - - node1 = (*head); - - while (node1->next && node1->next->next) { - if (node1->next->base > node1->next->next->base) { - out_of_order++; - node2 = node1->next; - node1->next = node1->next->next; - node1 = node1->next; - node2->next = node1->next; - node1->next = node2; - } else - node1 = node1->next; - } - } /* End of out_of_order loop */ - - node1 = *head; - - while (node1 && node1->next) { - if ((node1->base + node1->length) == node1->next->base) { - /* Combine */ - dbg("8..\n"); - node1->length += node1->next->length; - node2 = node1->next; - node1->next = node1->next->next; - kfree(node2); - } else - node1 = node1->next; - } - - return 0; -} - - -/** - * pciehp_slot_create - Creates a node and adds it to the proper bus. - * @busnumber - bus where new node is to be located - * - * Returns pointer to the new node or NULL if unsuccessful - */ -struct pci_func *pciehp_slot_create(u8 busnumber) -{ - struct pci_func *new_slot; - struct pci_func *next; - dbg("%s: busnumber %x\n", __FUNCTION__, busnumber); - new_slot = kmalloc(sizeof(struct pci_func), GFP_KERNEL); - - if (new_slot == NULL) - return new_slot; - - memset(new_slot, 0, sizeof(struct pci_func)); - - new_slot->next = NULL; - new_slot->configured = 1; - - if (pciehp_slot_list[busnumber] == NULL) { - pciehp_slot_list[busnumber] = new_slot; - } else { - next = pciehp_slot_list[busnumber]; - while (next->next != NULL) - next = next->next; - next->next = new_slot; - } - return new_slot; -} - - -/** - * slot_remove - Removes a node from the linked list of slots. - * @old_slot: slot to remove - * - * Returns 0 if successful, !0 otherwise. - */ -static int slot_remove(struct pci_func * old_slot) -{ - struct pci_func *next; - - if (old_slot == NULL) - return 1; - - next = pciehp_slot_list[old_slot->bus]; - - if (next == NULL) - return 1; - - if (next == old_slot) { - pciehp_slot_list[old_slot->bus] = old_slot->next; - pciehp_destroy_board_resources(old_slot); - kfree(old_slot); - return 0; - } - - while ((next->next != old_slot) && (next->next != NULL)) { - next = next->next; - } - - if (next->next == old_slot) { - next->next = old_slot->next; - pciehp_destroy_board_resources(old_slot); - kfree(old_slot); - return 0; - } else - return 2; -} - - -/** - * bridge_slot_remove - Removes a node from the linked list of slots. - * @bridge: bridge to remove - * - * Returns 0 if successful, !0 otherwise. - */ -static int bridge_slot_remove(struct pci_func *bridge) -{ - u8 subordinateBus, secondaryBus; - u8 tempBus; - struct pci_func *next; - - if (bridge == NULL) - return 1; - - secondaryBus = (bridge->config_space[0x06] >> 8) & 0xFF; - subordinateBus = (bridge->config_space[0x06] >> 16) & 0xFF; - - for (tempBus = secondaryBus; tempBus <= subordinateBus; tempBus++) { - next = pciehp_slot_list[tempBus]; - - while (!slot_remove(next)) { - next = pciehp_slot_list[tempBus]; - } - } - - next = pciehp_slot_list[bridge->bus]; - - if (next == NULL) { - return 1; - } - - if (next == bridge) { - pciehp_slot_list[bridge->bus] = bridge->next; - kfree(bridge); - return 0; - } - - while ((next->next != bridge) && (next->next != NULL)) { - next = next->next; - } - - if (next->next == bridge) { - next->next = bridge->next; - kfree(bridge); - return 0; - } else - return 2; -} - - -/** - * pciehp_slot_find - Looks for a node by bus, and device, multiple functions accessed - * @bus: bus to find - * @device: device to find - * @index: is 0 for first function found, 1 for the second... - * - * Returns pointer to the node if successful, %NULL otherwise. - */ -struct pci_func *pciehp_slot_find(u8 bus, u8 device, u8 index) -{ - int found = -1; - struct pci_func *func; - - func = pciehp_slot_list[bus]; - dbg("%s: bus %x device %x index %x\n", - __FUNCTION__, bus, device, index); - if (func != NULL) { - dbg("%s: func-> bus %x device %x function %x pci_dev %p\n", - __FUNCTION__, func->bus, func->device, func->function, - func->pci_dev); - } else - dbg("%s: func == NULL\n", __FUNCTION__); - - if ((func == NULL) || ((func->device == device) && (index == 0))) - return func; - - if (func->device == device) - found++; - - while (func->next != NULL) { - func = func->next; - - dbg("%s: In while loop, func-> bus %x device %x function %x pci_dev %p\n", - __FUNCTION__, func->bus, func->device, func->function, - func->pci_dev); - if (func->device == device) - found++; - dbg("%s: while loop, found %d, index %d\n", __FUNCTION__, - found, index); - - if ((found == index) || (func->function == index)) { - dbg("%s: Found bus %x dev %x func %x\n", __FUNCTION__, - func->bus, func->device, func->function); - return func; - } - } - - return NULL; -} - -static int is_bridge(struct pci_func * func) -{ - /* Check the header type */ - if (((func->config_space[0x03] >> 16) & 0xFF) == 0x01) - return 1; - else - return 0; -} - - /* The following routines constitute the bulk of the hotplug controller logic */ @@ -1100,20 +269,17 @@ static void set_slot_off(struct controller *ctrl, struct slot * pslot) * Configures board * */ -static u32 board_added(struct pci_func * func, struct controller * ctrl) +static int board_added(struct slot *p_slot) { u8 hp_slot; - int index; - u32 temp_register = 0xFFFFFFFF; - u32 rc = 0; - struct pci_func *new_func = NULL; - struct slot *p_slot; - struct resource_lists res_lists; + int rc = 0; + struct controller *ctrl = p_slot->ctrl; - p_slot = pciehp_find_slot(ctrl, func->device); - hp_slot = func->device - ctrl->slot_device_offset; + hp_slot = p_slot->device - ctrl->slot_device_offset; - dbg("%s: func->device, slot_offset, hp_slot = %d, %d ,%d\n", __FUNCTION__, func->device, ctrl->slot_device_offset, hp_slot); + dbg("%s: slot device, slot offset, hp slot = %d, %d ,%d\n", + __FUNCTION__, p_slot->device, + ctrl->slot_device_offset, hp_slot); /* Wait for exclusive access to hardware */ down(&ctrl->crit_sect); @@ -1141,9 +307,7 @@ static u32 board_added(struct pci_func * func, struct controller * ctrl) up(&ctrl->crit_sect); /* Wait for ~1 second */ - dbg("%s: before long_delay\n", __FUNCTION__); wait_for_ctrl_irq (ctrl); - dbg("%s: afterlong_delay\n", __FUNCTION__); /* Check link training status */ rc = p_slot->hpc_ops->check_lnk_status(ctrl); @@ -1153,98 +317,47 @@ static u32 board_added(struct pci_func * func, struct controller * ctrl) return rc; } - dbg("%s: func status = %x\n", __FUNCTION__, func->status); + dbg("%s: slot status = %x\n", __FUNCTION__, p_slot->status); /* Check for a power fault */ - if (func->status == 0xFF) { + if (p_slot->status == 0xFF) { /* power fault occurred, but it was benign */ - temp_register = 0xFFFFFFFF; - dbg("%s: temp register set to %x by power fault\n", __FUNCTION__, temp_register); rc = POWER_FAILURE; - func->status = 0; - } else { - /* Get vendor/device ID u32 */ - rc = pci_bus_read_config_dword (ctrl->pci_dev->subordinate, PCI_DEVFN(func->device, func->function), - PCI_VENDOR_ID, &temp_register); - dbg("%s: pci_bus_read_config_dword returns %d\n", __FUNCTION__, rc); - dbg("%s: temp_register is %x\n", __FUNCTION__, temp_register); - - if (rc != 0) { - /* Something's wrong here */ - temp_register = 0xFFFFFFFF; - dbg("%s: temp register set to %x by error\n", __FUNCTION__, temp_register); - } - /* Preset return code. It will be changed later if things go okay. */ - rc = NO_ADAPTER_PRESENT; + p_slot->status = 0; + goto err_exit; } - /* All F's is an empty slot or an invalid board */ - if (temp_register != 0xFFFFFFFF) { /* Check for a board in the slot */ - res_lists.io_head = ctrl->io_head; - res_lists.mem_head = ctrl->mem_head; - res_lists.p_mem_head = ctrl->p_mem_head; - res_lists.bus_head = ctrl->bus_head; - res_lists.irqs = NULL; - - rc = configure_new_device(ctrl, func, 0, &res_lists, 0, 0); - dbg("%s: back from configure_new_device\n", __FUNCTION__); - - ctrl->io_head = res_lists.io_head; - ctrl->mem_head = res_lists.mem_head; - ctrl->p_mem_head = res_lists.p_mem_head; - ctrl->bus_head = res_lists.bus_head; + rc = pciehp_configure_device(p_slot); + if (rc) { + err("Cannot add device 0x%x:%x\n", p_slot->bus, + p_slot->device); + goto err_exit; + } - pciehp_resource_sort_and_combine(&(ctrl->mem_head)); - pciehp_resource_sort_and_combine(&(ctrl->p_mem_head)); - pciehp_resource_sort_and_combine(&(ctrl->io_head)); - pciehp_resource_sort_and_combine(&(ctrl->bus_head)); + p_slot->status = 0; - if (rc) { - set_slot_off(ctrl, p_slot); - return rc; - } - pciehp_save_slot_config(ctrl, func); - - func->status = 0; - func->switch_save = 0x10; - func->is_a_board = 0x01; + /* + * Some PCI Express root ports require fixup after hot-plug operation. + */ + if (pcie_mch_quirk) + pci_fixup_device(pci_fixup_final, ctrl->pci_dev); + if (PWR_LED(ctrl->ctrlcap)) { + /* Wait for exclusive access to hardware */ + down(&ctrl->crit_sect); - /* next, we will instantiate the linux pci_dev structures - * (with appropriate driver notification, if already present) - */ - index = 0; - do { - new_func = pciehp_slot_find(ctrl->slot_bus, func->device, index++); - if (new_func && !new_func->pci_dev) { - dbg("%s:call pci_hp_configure_dev, func %x\n", - __FUNCTION__, index); - pciehp_configure_device(ctrl, new_func); - } - } while (new_func); - - /* - * Some PCI Express root ports require fixup after hot-plug operation. - */ - if (pcie_mch_quirk) - pci_fixup_device(pci_fixup_final, ctrl->pci_dev); - - if (PWR_LED(ctrl->ctrlcap)) { - /* Wait for exclusive access to hardware */ - down(&ctrl->crit_sect); - - p_slot->hpc_ops->green_led_on(p_slot); + p_slot->hpc_ops->green_led_on(p_slot); - /* Wait for the command to complete */ - wait_for_ctrl_irq (ctrl); + /* Wait for the command to complete */ + wait_for_ctrl_irq (ctrl); - /* Done with exclusive hardware access */ - up(&ctrl->crit_sect); - } - } else { - set_slot_off(ctrl, p_slot); - return -1; - } + /* Done with exclusive hardware access */ + up(&ctrl->crit_sect); + } return 0; + +err_exit: + set_slot_off(ctrl, p_slot); + return -1; } @@ -1252,55 +365,25 @@ static u32 board_added(struct pci_func * func, struct controller * ctrl) * remove_board - Turns off slot and LED's * */ -static u32 remove_board(struct pci_func *func, struct controller *ctrl) +static int remove_board(struct slot *p_slot) { - int index; - u8 skip = 0; u8 device; u8 hp_slot; - u32 rc; - struct resource_lists res_lists; - struct pci_func *temp_func; - struct slot *p_slot; - - if (func == NULL) - return 1; + int rc; + struct controller *ctrl = p_slot->ctrl; - if (pciehp_unconfigure_device(func)) + if (pciehp_unconfigure_device(p_slot)) return 1; - device = func->device; + device = p_slot->device; - hp_slot = func->device - ctrl->slot_device_offset; + hp_slot = p_slot->device - ctrl->slot_device_offset; p_slot = pciehp_find_slot(ctrl, hp_slot + ctrl->slot_device_offset); dbg("In %s, hp_slot = %d\n", __FUNCTION__, hp_slot); - if ((ctrl->add_support) && - !(func->bus_head || func->mem_head || func->p_mem_head || func->io_head)) { - /* Here we check to see if we've saved any of the board's - * resources already. If so, we'll skip the attempt to - * determine what's being used. - */ - index = 0; - - temp_func = func; - - while ((temp_func = pciehp_slot_find(temp_func->bus, temp_func->device, index++))) { - if (temp_func->bus_head || temp_func->mem_head - || temp_func->p_mem_head || temp_func->io_head) { - skip = 1; - break; - } - } - - if (!skip) - rc = pciehp_save_used_resources(ctrl, func, DISABLE_CARD); - } /* Change status to shutdown */ - if (func->is_a_board) - func->status = 0x01; - func->configured = 0; + p_slot->status = 0x01; /* Wait for exclusive access to hardware */ down(&ctrl->crit_sect); @@ -1328,56 +411,6 @@ static u32 remove_board(struct pci_func *func, struct controller *ctrl) /* Done with exclusive hardware access */ up(&ctrl->crit_sect); - if (ctrl->add_support) { - while (func) { - res_lists.io_head = ctrl->io_head; - res_lists.mem_head = ctrl->mem_head; - res_lists.p_mem_head = ctrl->p_mem_head; - res_lists.bus_head = ctrl->bus_head; - - dbg("Returning resources to ctlr lists for (B/D/F) = (%#x/%#x/%#x)\n", - func->bus, func->device, func->function); - - pciehp_return_board_resources(func, &res_lists); - - ctrl->io_head = res_lists.io_head; - ctrl->mem_head = res_lists.mem_head; - ctrl->p_mem_head = res_lists.p_mem_head; - ctrl->bus_head = res_lists.bus_head; - - pciehp_resource_sort_and_combine(&(ctrl->mem_head)); - pciehp_resource_sort_and_combine(&(ctrl->p_mem_head)); - pciehp_resource_sort_and_combine(&(ctrl->io_head)); - pciehp_resource_sort_and_combine(&(ctrl->bus_head)); - - if (is_bridge(func)) { - dbg("PCI Bridge Hot-Remove s:b:d:f(%02x:%02x:%02x:%02x)\n", - ctrl->seg, func->bus, func->device, func->function); - bridge_slot_remove(func); - } else { - dbg("PCI Function Hot-Remove s:b:d:f(%02x:%02x:%02x:%02x)\n", - ctrl->seg, func->bus, func->device, func->function); - slot_remove(func); - } - - func = pciehp_slot_find(ctrl->slot_bus, device, 0); - } - - /* Setup slot structure with entry for empty slot */ - func = pciehp_slot_create(ctrl->slot_bus); - - if (func == NULL) { - return 1; - } - - func->bus = ctrl->slot_bus; - func->device = device; - func->function = 0; - func->configured = 0; - func->switch_save = 0x10; - func->is_a_board = 0; - } - return 0; } @@ -1411,13 +444,15 @@ static void pciehp_pushbutton_thread(unsigned long slot) p_slot->hpc_ops->get_power_status(p_slot, &getstatus); if (getstatus) { p_slot->state = POWEROFF_STATE; - dbg("In power_down_board, b:d(%x:%x)\n", p_slot->bus, p_slot->device); + dbg("%s: disabling bus:device(%x:%x)\n", __FUNCTION__, + p_slot->bus, p_slot->device); pciehp_disable_slot(p_slot); p_slot->state = STATIC_STATE; } else { p_slot->state = POWERON_STATE; - dbg("In add_board, b:d(%x:%x)\n", p_slot->bus, p_slot->device); + dbg("%s: adding bus:device(%x:%x)\n", __FUNCTION__, + p_slot->bus, p_slot->device); if (pciehp_enable_slot(p_slot) && PWR_LED(p_slot->ctrl->ctrlcap)) { /* Wait for exclusive access to hardware */ @@ -1459,13 +494,15 @@ static void pciehp_surprise_rm_thread(unsigned long slot) p_slot->hpc_ops->get_adapter_status(p_slot, &getstatus); if (!getstatus) { p_slot->state = POWEROFF_STATE; - dbg("In removing board, b:d(%x:%x)\n", p_slot->bus, p_slot->device); + dbg("%s: removing bus:device(%x:%x)\n", + __FUNCTION__, p_slot->bus, p_slot->device); pciehp_disable_slot(p_slot); p_slot->state = STATIC_STATE; } else { p_slot->state = POWERON_STATE; - dbg("In add_board, b:d(%x:%x)\n", p_slot->bus, p_slot->device); + dbg("%s: adding bus:device(%x:%x)\n", + __FUNCTION__, p_slot->bus, p_slot->device); if (pciehp_enable_slot(p_slot) && PWR_LED(p_slot->ctrl->ctrlcap)) { /* Wait for exclusive access to hardware */ @@ -1531,7 +568,6 @@ int pciehp_event_start_thread(void) err ("Can't start up our event thread\n"); return -1; } - dbg("Our event thread pid = %d\n", pid); return 0; } @@ -1539,9 +575,7 @@ int pciehp_event_start_thread(void) void pciehp_event_stop_thread(void) { event_finished = 1; - dbg("event_thread finish command given\n"); up(&event_semaphore); - dbg("wait for event_thread to exit\n"); down(&event_exit); } @@ -1573,7 +607,6 @@ static void interrupt_event_handler(struct controller *ctrl) { int loop = 0; int change = 1; - struct pci_func *func; u8 hp_slot; u8 getstatus; struct slot *p_slot; @@ -1581,16 +614,12 @@ static void interrupt_event_handler(struct controller *ctrl) while (change) { change = 0; - for (loop = 0; loop < 10; loop++) { + for (loop = 0; loop < MAX_EVENTS; loop++) { if (ctrl->event_queue[loop].event_type != 0) { hp_slot = ctrl->event_queue[loop].hp_slot; - func = pciehp_slot_find(ctrl->slot_bus, (hp_slot + ctrl->slot_device_offset), 0); - p_slot = pciehp_find_slot(ctrl, hp_slot + ctrl->slot_device_offset); - dbg("hp_slot %d, func %p, p_slot %p\n", hp_slot, func, p_slot); - if (ctrl->event_queue[loop].event_type == INT_BUTTON_CANCEL) { dbg("button cancel\n"); del_timer(&p_slot->task_event); @@ -1682,7 +711,6 @@ static void interrupt_event_handler(struct controller *ctrl) p_slot->task_event.function = (void (*)(unsigned long)) pushbutton_helper_thread; p_slot->task_event.data = (unsigned long) p_slot; - dbg("add_timer p_slot = %p\n", (void *) p_slot); add_timer(&p_slot->task_event); } } @@ -1737,13 +765,6 @@ int pciehp_enable_slot(struct slot *p_slot) { u8 getstatus = 0; int rc; - struct pci_func *func; - - func = pciehp_slot_find(p_slot->bus, p_slot->device, 0); - if (!func) { - dbg("%s: Error! slot NULL\n", __FUNCTION__); - return 1; - } /* Check to see if (latch closed, card present, power off) */ down(&p_slot->ctrl->crit_sect); @@ -1773,45 +794,11 @@ int pciehp_enable_slot(struct slot *p_slot) } up(&p_slot->ctrl->crit_sect); - slot_remove(func); - - func = pciehp_slot_create(p_slot->bus); - if (func == NULL) - return 1; - - func->bus = p_slot->bus; - func->device = p_slot->device; - func->function = 0; - func->configured = 0; - func->is_a_board = 1; - - /* We have to save the presence info for these slots */ - p_slot->hpc_ops->get_adapter_status(p_slot, &(func->presence_save)); p_slot->hpc_ops->get_latch_status(p_slot, &getstatus); - func->switch_save = !getstatus? 0x10:0; - rc = board_added(func, p_slot->ctrl); + rc = board_added(p_slot); if (rc) { - if (is_bridge(func)) - bridge_slot_remove(func); - else - slot_remove(func); - - /* Setup slot structure with entry for empty slot */ - func = pciehp_slot_create(p_slot->bus); - if (func == NULL) - return 1; /* Out of memory */ - - func->bus = p_slot->bus; - func->device = p_slot->device; - func->function = 0; - func->configured = 0; - func->is_a_board = 1; - - /* We have to save the presence info for these slots */ - p_slot->hpc_ops->get_adapter_status(p_slot, &(func->presence_save)); p_slot->hpc_ops->get_latch_status(p_slot, &getstatus); - func->switch_save = !getstatus? 0x10:0; } if (p_slot) @@ -1823,14 +810,8 @@ int pciehp_enable_slot(struct slot *p_slot) int pciehp_disable_slot(struct slot *p_slot) { - u8 class_code, header_type, BCR; - u8 index = 0; u8 getstatus = 0; - u32 rc = 0; int ret = 0; - unsigned int devfn; - struct pci_bus *pci_bus = p_slot->ctrl->pci_dev->subordinate; - struct pci_func *func; if (!p_slot->ctrl) return 1; @@ -1867,838 +848,8 @@ int pciehp_disable_slot(struct slot *p_slot) up(&p_slot->ctrl->crit_sect); - func = pciehp_slot_find(p_slot->bus, p_slot->device, index++); - - /* Make sure there are no video controllers here - * for all func of p_slot - */ - while (func && !rc) { - pci_bus->number = func->bus; - devfn = PCI_DEVFN(func->device, func->function); - - /* Check the Class Code */ - rc = pci_bus_read_config_byte (pci_bus, devfn, 0x0B, &class_code); - if (rc) - return rc; - - if (class_code == PCI_BASE_CLASS_DISPLAY) { - /* Display/Video adapter (not supported) */ - rc = REMOVE_NOT_SUPPORTED; - } else { - /* See if it's a bridge */ - rc = pci_bus_read_config_byte (pci_bus, devfn, PCI_HEADER_TYPE, &header_type); - if (rc) - return rc; - - /* If it's a bridge, check the VGA Enable bit */ - if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { - rc = pci_bus_read_config_byte (pci_bus, devfn, PCI_BRIDGE_CONTROL, &BCR); - if (rc) - return rc; - - /* If the VGA Enable bit is set, remove isn't supported */ - if (BCR & PCI_BRIDGE_CTL_VGA) { - rc = REMOVE_NOT_SUPPORTED; - } - } - } - - func = pciehp_slot_find(p_slot->bus, p_slot->device, index++); - } - - func = pciehp_slot_find(p_slot->bus, p_slot->device, 0); - if ((func != NULL) && !rc) { - rc = remove_board(func, p_slot->ctrl); - } else if (!rc) - rc = 1; - - if (p_slot) - update_slot_info(p_slot); - - return rc; -} - - -/** - * configure_new_device - Configures the PCI header information of one board. - * - * @ctrl: pointer to controller structure - * @func: pointer to function structure - * @behind_bridge: 1 if this is a recursive call, 0 if not - * @resources: pointer to set of resource lists - * - * Returns 0 if success - * - */ -static u32 configure_new_device(struct controller * ctrl, struct pci_func * func, - u8 behind_bridge, struct resource_lists * resources, u8 bridge_bus, u8 bridge_dev) -{ - u8 temp_byte, function, max_functions, stop_it; - int rc; - u32 ID; - struct pci_func *new_slot; - struct pci_bus lpci_bus, *pci_bus; - int index; - - new_slot = func; - - dbg("%s\n", __FUNCTION__); - memcpy(&lpci_bus, ctrl->pci_dev->subordinate, sizeof(lpci_bus)); - pci_bus = &lpci_bus; - pci_bus->number = func->bus; - - /* Check for Multi-function device */ - rc = pci_bus_read_config_byte(pci_bus, PCI_DEVFN(func->device, func->function), 0x0E, &temp_byte); - if (rc) { - dbg("%s: rc = %d\n", __FUNCTION__, rc); - return rc; - } - - if (temp_byte & 0x80) /* Multi-function device */ - max_functions = 8; - else - max_functions = 1; - - function = 0; - - do { - rc = configure_new_function(ctrl, new_slot, behind_bridge, - resources, bridge_bus, bridge_dev); - - if (rc) { - dbg("configure_new_function failed: %d\n", rc); - index = 0; - - while (new_slot) { - new_slot = pciehp_slot_find(new_slot->bus, - new_slot->device, index++); - - if (new_slot) - pciehp_return_board_resources(new_slot, - resources); - } - - return rc; - } - - function++; - - stop_it = 0; - - /* The following loop skips to the next present function - * and creates a board structure - */ - - while ((function < max_functions) && (!stop_it)) { - pci_bus_read_config_dword(pci_bus, PCI_DEVFN(func->device, function), 0x00, &ID); - - if (ID == 0xFFFFFFFF) { /* There's nothing there. */ - function++; - } else { /* There's something there */ - /* Setup slot structure. */ - new_slot = pciehp_slot_create(func->bus); - - if (new_slot == NULL) { - /* Out of memory */ - return 1; - } - - new_slot->bus = func->bus; - new_slot->device = func->device; - new_slot->function = function; - new_slot->is_a_board = 1; - new_slot->status = 0; - - stop_it++; - } - } - - } while (function < max_functions); - dbg("returning from %s\n", __FUNCTION__); - - return 0; -} - -/* - * Configuration logic that involves the hotplug data structures and - * their bookkeeping - */ - -/** - * configure_bridge: fill bridge's registers, either configure or disable it. - */ -static int -configure_bridge(struct pci_bus *pci_bus, unsigned int devfn, - struct pci_resource *mem_node, - struct pci_resource **hold_mem_node, - int base_addr, int limit_addr) -{ - u16 temp_word; - u32 rc; - - if (mem_node) { - memcpy(*hold_mem_node, mem_node, sizeof(struct pci_resource)); - mem_node->next = NULL; - - /* set Mem base and Limit registers */ - RES_CHECK(mem_node->base, 16); - temp_word = (u16)(mem_node->base >> 16); - rc = pci_bus_write_config_word(pci_bus, devfn, base_addr, temp_word); - - RES_CHECK(mem_node->base + mem_node->length - 1, 16); - temp_word = (u16)((mem_node->base + mem_node->length - 1) >> 16); - rc = pci_bus_write_config_word(pci_bus, devfn, limit_addr, temp_word); - } else { - temp_word = 0xFFFF; - rc = pci_bus_write_config_word(pci_bus, devfn, base_addr, temp_word); - - temp_word = 0x0000; - rc = pci_bus_write_config_word(pci_bus, devfn, limit_addr, temp_word); - - kfree(*hold_mem_node); - *hold_mem_node = NULL; - } - return rc; -} - -static int -configure_new_bridge(struct controller *ctrl, struct pci_func *func, - u8 behind_bridge, struct resource_lists *resources, - struct pci_bus *pci_bus) -{ - int cloop; - u8 temp_byte; - u8 device; - u16 temp_word; - u32 rc; - u32 ID; - unsigned int devfn; - struct pci_resource *mem_node; - struct pci_resource *p_mem_node; - struct pci_resource *io_node; - struct pci_resource *bus_node; - struct pci_resource *hold_mem_node; - struct pci_resource *hold_p_mem_node; - struct pci_resource *hold_IO_node; - struct pci_resource *hold_bus_node; - struct irq_mapping irqs; - struct pci_func *new_slot; - struct resource_lists temp_resources; - - devfn = PCI_DEVFN(func->device, func->function); - - /* set Primary bus */ - dbg("set Primary bus = 0x%x\n", func->bus); - rc = pci_bus_write_config_byte(pci_bus, devfn, PCI_PRIMARY_BUS, func->bus); - if (rc) - return rc; - - /* find range of busses to use */ - bus_node = get_max_resource(&resources->bus_head, 1L); - - /* If we don't have any busses to allocate, we can't continue */ - if (!bus_node) { - err("Got NO bus resource to use\n"); - return -ENOMEM; - } - dbg("Got ranges of buses to use: base:len=0x%x:%x\n", bus_node->base, bus_node->length); - - /* set Secondary bus */ - temp_byte = (u8)bus_node->base; - dbg("set Secondary bus = 0x%x\n", temp_byte); - rc = pci_bus_write_config_byte(pci_bus, devfn, PCI_SECONDARY_BUS, temp_byte); - if (rc) - return rc; - - /* set subordinate bus */ - temp_byte = (u8)(bus_node->base + bus_node->length - 1); - dbg("set subordinate bus = 0x%x\n", temp_byte); - rc = pci_bus_write_config_byte (pci_bus, devfn, PCI_SUBORDINATE_BUS, temp_byte); - if (rc) - return rc; - - /* Set HP parameters (Cache Line Size, Latency Timer) */ - rc = pciehprm_set_hpp(ctrl, func, PCI_HEADER_TYPE_BRIDGE); - if (rc) - return rc; - - /* Setup the IO, memory, and prefetchable windows */ - - io_node = get_max_resource(&(resources->io_head), 0x1000L); - if (io_node) { - dbg("io_node(base, len, next) (%x, %x, %p)\n", io_node->base, - io_node->length, io_node->next); - } - - mem_node = get_max_resource(&(resources->mem_head), 0x100000L); - if (mem_node) { - dbg("mem_node(base, len, next) (%x, %x, %p)\n", mem_node->base, - mem_node->length, mem_node->next); - } - - if (resources->p_mem_head) - p_mem_node = get_max_resource(&(resources->p_mem_head), 0x100000L); - else { - /* - * In some platform implementation, MEM and PMEM are not - * distinguished, and hence ACPI _CRS has only MEM entries - * for both MEM and PMEM. - */ - dbg("using MEM for PMEM\n"); - p_mem_node = get_max_resource(&(resources->mem_head), 0x100000L); - } - if (p_mem_node) { - dbg("p_mem_node(base, len, next) (%x, %x, %p)\n", p_mem_node->base, - p_mem_node->length, p_mem_node->next); - } - - /* set up the IRQ info */ - if (!resources->irqs) { - irqs.barber_pole = 0; - irqs.interrupt[0] = 0; - irqs.interrupt[1] = 0; - irqs.interrupt[2] = 0; - irqs.interrupt[3] = 0; - irqs.valid_INT = 0; - } else { - irqs.barber_pole = resources->irqs->barber_pole; - irqs.interrupt[0] = resources->irqs->interrupt[0]; - irqs.interrupt[1] = resources->irqs->interrupt[1]; - irqs.interrupt[2] = resources->irqs->interrupt[2]; - irqs.interrupt[3] = resources->irqs->interrupt[3]; - irqs.valid_INT = resources->irqs->valid_INT; - } - - /* set up resource lists that are now aligned on top and bottom - * for anything behind the bridge. - */ - temp_resources.bus_head = bus_node; - temp_resources.io_head = io_node; - temp_resources.mem_head = mem_node; - temp_resources.p_mem_head = p_mem_node; - temp_resources.irqs = &irqs; - - /* Make copies of the nodes we are going to pass down so that - * if there is a problem,we can just use these to free resources - */ - hold_bus_node = kmalloc(sizeof(struct pci_resource), GFP_KERNEL); - hold_IO_node = kmalloc(sizeof(struct pci_resource), GFP_KERNEL); - hold_mem_node = kmalloc(sizeof(struct pci_resource), GFP_KERNEL); - hold_p_mem_node = kmalloc(sizeof(struct pci_resource), GFP_KERNEL); - - if (!hold_bus_node || !hold_IO_node || !hold_mem_node || !hold_p_mem_node) { - kfree(hold_bus_node); - kfree(hold_IO_node); - kfree(hold_mem_node); - kfree(hold_p_mem_node); - - return 1; - } - - memcpy(hold_bus_node, bus_node, sizeof(struct pci_resource)); - - bus_node->base += 1; - bus_node->length -= 1; - bus_node->next = NULL; - - /* If we have IO resources copy them and fill in the bridge's - * IO range registers - */ - if (io_node) { - memcpy(hold_IO_node, io_node, sizeof(struct pci_resource)); - io_node->next = NULL; - - /* set IO base and Limit registers */ - RES_CHECK(io_node->base, 8); - temp_byte = (u8)(io_node->base >> 8); - rc = pci_bus_write_config_byte (pci_bus, devfn, PCI_IO_BASE, temp_byte); - - RES_CHECK(io_node->base + io_node->length - 1, 8); - temp_byte = (u8)((io_node->base + io_node->length - 1) >> 8); - rc = pci_bus_write_config_byte (pci_bus, devfn, PCI_IO_LIMIT, temp_byte); - } else { - kfree(hold_IO_node); - hold_IO_node = NULL; - } - - /* If we have memory resources copy them and fill in the bridge's - * memory range registers. Otherwise, fill in the range - * registers with values that disable them. - */ - rc = configure_bridge(pci_bus, devfn, mem_node, &hold_mem_node, - PCI_MEMORY_BASE, PCI_MEMORY_LIMIT); - - /* If we have prefetchable memory resources copy them and - * fill in the bridge's memory range registers. Otherwise, - * fill in the range registers with values that disable them. - */ - rc = configure_bridge(pci_bus, devfn, p_mem_node, &hold_p_mem_node, - PCI_PREF_MEMORY_BASE, PCI_PREF_MEMORY_LIMIT); - - /* Adjust this to compensate for extra adjustment in first loop */ - irqs.barber_pole--; - - rc = 0; - - /* Here we actually find the devices and configure them */ - for (device = 0; (device <= 0x1F) && !rc; device++) { - irqs.barber_pole = (irqs.barber_pole + 1) & 0x03; - - ID = 0xFFFFFFFF; - pci_bus->number = hold_bus_node->base; - pci_bus_read_config_dword (pci_bus, PCI_DEVFN(device, 0), PCI_VENDOR_ID, &ID); - pci_bus->number = func->bus; - - if (ID != 0xFFFFFFFF) { /* device Present */ - /* Setup slot structure. */ - new_slot = pciehp_slot_create(hold_bus_node->base); - - if (new_slot == NULL) { - /* Out of memory */ - rc = -ENOMEM; - continue; - } - - new_slot->bus = hold_bus_node->base; - new_slot->device = device; - new_slot->function = 0; - new_slot->is_a_board = 1; - new_slot->status = 0; - - rc = configure_new_device(ctrl, new_slot, 1, - &temp_resources, func->bus, - func->device); - dbg("configure_new_device rc=0x%x\n",rc); - } /* End of IF (device in slot?) */ - } /* End of FOR loop */ - - if (rc) { - pciehp_destroy_resource_list(&temp_resources); - - return_resource(&(resources->bus_head), hold_bus_node); - return_resource(&(resources->io_head), hold_IO_node); - return_resource(&(resources->mem_head), hold_mem_node); - return_resource(&(resources->p_mem_head), hold_p_mem_node); - return(rc); - } - - /* save the interrupt routing information */ - if (resources->irqs) { - resources->irqs->interrupt[0] = irqs.interrupt[0]; - resources->irqs->interrupt[1] = irqs.interrupt[1]; - resources->irqs->interrupt[2] = irqs.interrupt[2]; - resources->irqs->interrupt[3] = irqs.interrupt[3]; - resources->irqs->valid_INT = irqs.valid_INT; - } else if (!behind_bridge) { - /* We need to hook up the interrupts here */ - for (cloop = 0; cloop < 4; cloop++) { - if (irqs.valid_INT & (0x01 << cloop)) { - rc = pciehp_set_irq(func->bus, func->device, - 0x0A + cloop, irqs.interrupt[cloop]); - if (rc) { - pciehp_destroy_resource_list (&temp_resources); - return_resource(&(resources->bus_head), hold_bus_node); - return_resource(&(resources->io_head), hold_IO_node); - return_resource(&(resources->mem_head), hold_mem_node); - return_resource(&(resources->p_mem_head), hold_p_mem_node); - return rc; - } - } - } /* end of for loop */ - } - - /* Return unused bus resources - * First use the temporary node to store information for the board - */ - if (hold_bus_node && bus_node && temp_resources.bus_head) { - hold_bus_node->length = bus_node->base - hold_bus_node->base; - - hold_bus_node->next = func->bus_head; - func->bus_head = hold_bus_node; - - temp_byte = (u8)(temp_resources.bus_head->base - 1); - - /* set subordinate bus */ - dbg("re-set subordinate bus = 0x%x\n", temp_byte); - rc = pci_bus_write_config_byte (pci_bus, devfn, PCI_SUBORDINATE_BUS, temp_byte); - - if (temp_resources.bus_head->length == 0) { - kfree(temp_resources.bus_head); - temp_resources.bus_head = NULL; - } else { - dbg("return bus res of b:d(0x%x:%x) base:len(0x%x:%x)\n", - func->bus, func->device, temp_resources.bus_head->base, temp_resources.bus_head->length); - return_resource(&(resources->bus_head), temp_resources.bus_head); - } - } - - /* If we have IO space available and there is some left, - * return the unused portion - */ - if (hold_IO_node && temp_resources.io_head) { - io_node = do_pre_bridge_resource_split(&(temp_resources.io_head), - &hold_IO_node, 0x1000); - - /* Check if we were able to split something off */ - if (io_node) { - hold_IO_node->base = io_node->base + io_node->length; - - RES_CHECK(hold_IO_node->base, 8); - temp_byte = (u8)((hold_IO_node->base) >> 8); - rc = pci_bus_write_config_byte (pci_bus, devfn, PCI_IO_BASE, temp_byte); - - return_resource(&(resources->io_head), io_node); - } - - io_node = do_bridge_resource_split(&(temp_resources.io_head), 0x1000); - - /* Check if we were able to split something off */ - if (io_node) { - /* First use the temporary node to store information for the board */ - hold_IO_node->length = io_node->base - hold_IO_node->base; - - /* If we used any, add it to the board's list */ - if (hold_IO_node->length) { - hold_IO_node->next = func->io_head; - func->io_head = hold_IO_node; - - RES_CHECK(io_node->base - 1, 8); - temp_byte = (u8)((io_node->base - 1) >> 8); - rc = pci_bus_write_config_byte (pci_bus, devfn, PCI_IO_LIMIT, temp_byte); - - return_resource(&(resources->io_head), io_node); - } else { - /* it doesn't need any IO */ - temp_byte = 0x00; - rc = pci_bus_write_config_byte(pci_bus, devfn, PCI_IO_LIMIT, temp_byte); - - return_resource(&(resources->io_head), io_node); - kfree(hold_IO_node); - } - } else { - /* it used most of the range */ - hold_IO_node->next = func->io_head; - func->io_head = hold_IO_node; - } - } else if (hold_IO_node) { - /* it used the whole range */ - hold_IO_node->next = func->io_head; - func->io_head = hold_IO_node; - } - - /* If we have memory space available and there is some left, - * return the unused portion - */ - if (hold_mem_node && temp_resources.mem_head) { - mem_node = do_pre_bridge_resource_split(&(temp_resources.mem_head), &hold_mem_node, 0x100000L); - - /* Check if we were able to split something off */ - if (mem_node) { - hold_mem_node->base = mem_node->base + mem_node->length; - - RES_CHECK(hold_mem_node->base, 16); - temp_word = (u16)((hold_mem_node->base) >> 16); - rc = pci_bus_write_config_word (pci_bus, devfn, PCI_MEMORY_BASE, temp_word); - - return_resource(&(resources->mem_head), mem_node); - } - - mem_node = do_bridge_resource_split(&(temp_resources.mem_head), 0x100000L); - - /* Check if we were able to split something off */ - if (mem_node) { - /* First use the temporary node to store information for the board */ - hold_mem_node->length = mem_node->base - hold_mem_node->base; - - if (hold_mem_node->length) { - hold_mem_node->next = func->mem_head; - func->mem_head = hold_mem_node; - - /* configure end address */ - RES_CHECK(mem_node->base - 1, 16); - temp_word = (u16)((mem_node->base - 1) >> 16); - rc = pci_bus_write_config_word (pci_bus, devfn, PCI_MEMORY_LIMIT, temp_word); - - /* Return unused resources to the pool */ - return_resource(&(resources->mem_head), mem_node); - } else { - /* it doesn't need any Mem */ - temp_word = 0x0000; - rc = pci_bus_write_config_word (pci_bus, devfn, PCI_MEMORY_LIMIT, temp_word); - - return_resource(&(resources->mem_head), mem_node); - kfree(hold_mem_node); - } - } else { - /* it used most of the range */ - hold_mem_node->next = func->mem_head; - func->mem_head = hold_mem_node; - } - } else if (hold_mem_node) { - /* it used the whole range */ - hold_mem_node->next = func->mem_head; - func->mem_head = hold_mem_node; - } - - /* If we have prefetchable memory space available and there is some - * left at the end, return the unused portion - */ - if (hold_p_mem_node && temp_resources.p_mem_head) { - p_mem_node = do_pre_bridge_resource_split(&(temp_resources.p_mem_head), - &hold_p_mem_node, 0x100000L); - - /* Check if we were able to split something off */ - if (p_mem_node) { - hold_p_mem_node->base = p_mem_node->base + p_mem_node->length; - - RES_CHECK(hold_p_mem_node->base, 16); - temp_word = (u16)((hold_p_mem_node->base) >> 16); - rc = pci_bus_write_config_word (pci_bus, devfn, PCI_PREF_MEMORY_BASE, temp_word); - - return_resource(&(resources->p_mem_head), p_mem_node); - } - - p_mem_node = do_bridge_resource_split(&(temp_resources.p_mem_head), 0x100000L); - - /* Check if we were able to split something off */ - if (p_mem_node) { - /* First use the temporary node to store information for the board */ - hold_p_mem_node->length = p_mem_node->base - hold_p_mem_node->base; - - /* If we used any, add it to the board's list */ - if (hold_p_mem_node->length) { - hold_p_mem_node->next = func->p_mem_head; - func->p_mem_head = hold_p_mem_node; - - RES_CHECK(p_mem_node->base - 1, 16); - temp_word = (u16)((p_mem_node->base - 1) >> 16); - rc = pci_bus_write_config_word (pci_bus, devfn, PCI_PREF_MEMORY_LIMIT, temp_word); - - return_resource(&(resources->p_mem_head), p_mem_node); - } else { - /* it doesn't need any PMem */ - temp_word = 0x0000; - rc = pci_bus_write_config_word (pci_bus, devfn, PCI_PREF_MEMORY_LIMIT, temp_word); - - return_resource(&(resources->p_mem_head), p_mem_node); - kfree(hold_p_mem_node); - } - } else { - /* it used the most of the range */ - hold_p_mem_node->next = func->p_mem_head; - func->p_mem_head = hold_p_mem_node; - } - } else if (hold_p_mem_node) { - /* it used the whole range */ - hold_p_mem_node->next = func->p_mem_head; - func->p_mem_head = hold_p_mem_node; - } - - /* We should be configuring an IRQ and the bridge's base address - * registers if it needs them. Although we have never seen such - * a device - */ - - pciehprm_enable_card(ctrl, func, PCI_HEADER_TYPE_BRIDGE); - - dbg("PCI Bridge Hot-Added s:b:d:f(%02x:%02x:%02x:%02x)\n", ctrl->seg, func->bus, func->device, func->function); - - return rc; + ret = remove_board(p_slot); + update_slot_info(p_slot); + return ret; } -/** - * configure_new_function - Configures the PCI header information of one device - * - * @ctrl: pointer to controller structure - * @func: pointer to function structure - * @behind_bridge: 1 if this is a recursive call, 0 if not - * @resources: pointer to set of resource lists - * - * Calls itself recursively for bridged devices. - * Returns 0 if success - * - */ -static int -configure_new_function(struct controller *ctrl, struct pci_func *func, - u8 behind_bridge, struct resource_lists *resources, - u8 bridge_bus, u8 bridge_dev) -{ - int cloop; - u8 temp_byte; - u8 class_code; - u32 rc; - u32 temp_register; - u32 base; - unsigned int devfn; - struct pci_resource *mem_node; - struct pci_resource *io_node; - struct pci_bus lpci_bus, *pci_bus; - - memcpy(&lpci_bus, ctrl->pci_dev->subordinate, sizeof(lpci_bus)); - pci_bus = &lpci_bus; - pci_bus->number = func->bus; - devfn = PCI_DEVFN(func->device, func->function); - - /* Check for Bridge */ - rc = pci_bus_read_config_byte(pci_bus, devfn, PCI_HEADER_TYPE, &temp_byte); - if (rc) - return rc; - dbg("%s: bus %x dev %x func %x temp_byte = %x\n", __FUNCTION__, - func->bus, func->device, func->function, temp_byte); - - if ((temp_byte & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { /* PCI-PCI Bridge */ - rc = configure_new_bridge(ctrl, func, behind_bridge, resources, - pci_bus); - - if (rc) - return rc; - } else if ((temp_byte & 0x7F) == PCI_HEADER_TYPE_NORMAL) { - /* Standard device */ - u64 base64; - rc = pci_bus_read_config_byte(pci_bus, devfn, 0x0B, &class_code); - - if (class_code == PCI_BASE_CLASS_DISPLAY) - return DEVICE_TYPE_NOT_SUPPORTED; - - /* Figure out IO and memory needs */ - for (cloop = PCI_BASE_ADDRESS_0; cloop <= PCI_BASE_ADDRESS_5; cloop += 4) { - temp_register = 0xFFFFFFFF; - - rc = pci_bus_write_config_dword (pci_bus, devfn, cloop, temp_register); - rc = pci_bus_read_config_dword(pci_bus, devfn, cloop, &temp_register); - dbg("Bar[%x]=0x%x on bus:dev:func(0x%x:%x:%x)\n", cloop, temp_register, - func->bus, func->device, func->function); - - if (!temp_register) - continue; - - base64 = 0L; - if (temp_register & PCI_BASE_ADDRESS_SPACE_IO) { - /* Map IO */ - - /* set base = amount of IO space */ - base = temp_register & 0xFFFFFFFC; - base = ~base + 1; - - dbg("NEED IO length(0x%x)\n", base); - io_node = get_io_resource(&(resources->io_head),(ulong)base); - - /* allocate the resource to the board */ - if (io_node) { - dbg("Got IO base=0x%x(length=0x%x)\n", io_node->base, io_node->length); - base = (u32)io_node->base; - io_node->next = func->io_head; - func->io_head = io_node; - } else { - err("Got NO IO resource(length=0x%x)\n", base); - return -ENOMEM; - } - } else { /* map MEM */ - int prefetchable = 1; - struct pci_resource **res_node = &func->p_mem_head; - char *res_type_str = "PMEM"; - u32 temp_register2; - - if (!(temp_register & PCI_BASE_ADDRESS_MEM_PREFETCH)) { - prefetchable = 0; - res_node = &func->mem_head; - res_type_str++; - } - - base = temp_register & 0xFFFFFFF0; - base = ~base + 1; - - switch (temp_register & PCI_BASE_ADDRESS_MEM_TYPE_MASK) { - case PCI_BASE_ADDRESS_MEM_TYPE_32: - dbg("NEED 32 %s bar=0x%x(length=0x%x)\n", res_type_str, temp_register, base); - - if (prefetchable && resources->p_mem_head) - mem_node=get_resource(&(resources->p_mem_head), (ulong)base); - else { - if (prefetchable) - dbg("using MEM for PMEM\n"); - mem_node = get_resource(&(resources->mem_head), (ulong)base); - } - - /* allocate the resource to the board */ - if (mem_node) { - base = (u32)mem_node->base; - mem_node->next = *res_node; - *res_node = mem_node; - dbg("Got 32 %s base=0x%x(length=0x%x)\n", res_type_str, mem_node->base, - mem_node->length); - } else { - err("Got NO 32 %s resource(length=0x%x)\n", res_type_str, base); - return -ENOMEM; - } - break; - case PCI_BASE_ADDRESS_MEM_TYPE_64: - rc = pci_bus_read_config_dword(pci_bus, devfn, cloop+4, &temp_register2); - dbg("NEED 64 %s bar=0x%x:%x(length=0x%x)\n", res_type_str, temp_register2, - temp_register, base); - - if (prefetchable && resources->p_mem_head) - mem_node = get_resource(&(resources->p_mem_head), (ulong)base); - else { - if (prefetchable) - dbg("using MEM for PMEM\n"); - mem_node = get_resource(&(resources->mem_head), (ulong)base); - } - - /* allocate the resource to the board */ - if (mem_node) { - base64 = mem_node->base; - mem_node->next = *res_node; - *res_node = mem_node; - dbg("Got 64 %s base=0x%x:%x(length=%x)\n", res_type_str, (u32)(base64 >> 32), - (u32)base64, mem_node->length); - } else { - err("Got NO 64 %s resource(length=0x%x)\n", res_type_str, base); - return -ENOMEM; - } - break; - default: - dbg("reserved BAR type=0x%x\n", temp_register); - break; - } - - } - - if (base64) { - rc = pci_bus_write_config_dword(pci_bus, devfn, cloop, (u32)base64); - cloop += 4; - base64 >>= 32; - - if (base64) { - dbg("%s: high dword of base64(0x%x) set to 0\n", __FUNCTION__, (u32)base64); - base64 = 0x0L; - } - - rc = pci_bus_write_config_dword(pci_bus, devfn, cloop, (u32)base64); - } else { - rc = pci_bus_write_config_dword(pci_bus, devfn, cloop, base); - } - } /* End of base register loop */ - - /* disable ROM base Address */ - rc = pci_bus_write_config_dword (pci_bus, devfn, PCI_ROM_ADDRESS, 0x00); - - /* Set HP parameters (Cache Line Size, Latency Timer) */ - rc = pciehprm_set_hpp(ctrl, func, PCI_HEADER_TYPE_NORMAL); - if (rc) - return rc; - - pciehprm_enable_card(ctrl, func, PCI_HEADER_TYPE_NORMAL); - - dbg("PCI function Hot-Added s:b:d:f(%02x:%02x:%02x:%02x)\n", ctrl->seg, func->bus, func->device, - func->function); - } /* End of Not-A-Bridge else */ - else { - /* It's some strange type of PCI adapter (Cardbus?) */ - return DEVICE_TYPE_NOT_SUPPORTED; - } - - func->configured = 1; - - return 0; -} diff --git a/drivers/pci/hotplug/pciehp_hpc.c b/drivers/pci/hotplug/pciehp_hpc.c index 7a0e27f0e06..4a3cecca012 100644 --- a/drivers/pci/hotplug/pciehp_hpc.c +++ b/drivers/pci/hotplug/pciehp_hpc.c @@ -27,16 +27,10 @@ * */ -#include <linux/config.h> #include <linux/kernel.h> #include <linux/module.h> #include <linux/types.h> -#include <linux/slab.h> -#include <linux/vmalloc.h> -#include <linux/interrupt.h> -#include <linux/spinlock.h> #include <linux/pci.h> -#include <asm/system.h> #include "../pci.h" #include "pciehp.h" @@ -217,23 +211,6 @@ static int pcie_cap_base = 0; /* Base of the PCI Express capability item struct #define MRL_STATE 0x0020 #define PRSN_STATE 0x0040 -struct php_ctlr_state_s { - struct php_ctlr_state_s *pnext; - struct pci_dev *pci_dev; - unsigned int irq; - unsigned long flags; /* spinlock's */ - u32 slot_device_offset; - u32 num_slots; - struct timer_list int_poll_timer; /* Added for poll event */ - php_intr_callback_t attention_button_callback; - php_intr_callback_t switch_change_callback; - php_intr_callback_t presence_change_callback; - php_intr_callback_t power_fault_callback; - void *callback_instance_id; - struct ctrl_reg *creg; /* Ptr to controller register space */ -}; - - static spinlock_t hpc_event_lock; DEFINE_DBG_BUFFER /* Debug string buffer for entire HPC defined here */ @@ -297,7 +274,6 @@ static int pcie_write_cmd(struct slot *slot, u16 cmd) DBG_ENTER_ROUTINE - dbg("%s : Enter\n", __FUNCTION__); if (!php_ctlr) { err("%s: Invalid HPC controller handle!\n", __FUNCTION__); return -1; @@ -308,7 +284,6 @@ static int pcie_write_cmd(struct slot *slot, u16 cmd) err("%s : hp_register_read_word SLOT_STATUS failed\n", __FUNCTION__); return retval; } - dbg("%s : hp_register_read_word SLOT_STATUS %x\n", __FUNCTION__, slot_status); if ((slot_status & CMD_COMPLETED) == CMD_COMPLETED ) { /* After 1 sec and CMD_COMPLETED still not set, just proceed forward to issue @@ -316,14 +291,11 @@ static int pcie_write_cmd(struct slot *slot, u16 cmd) dbg("%s : CMD_COMPLETED not clear after 1 sec.\n", __FUNCTION__); } - dbg("%s: Before hp_register_write_word SLOT_CTRL %x\n", __FUNCTION__, cmd); retval = hp_register_write_word(php_ctlr->pci_dev, SLOT_CTRL(slot->ctrl->cap_base), cmd | CMD_CMPL_INTR_ENABLE); if (retval) { err("%s : hp_register_write_word SLOT_CTRL failed\n", __FUNCTION__); return retval; } - dbg("%s : hp_register_write_word SLOT_CTRL %x\n", __FUNCTION__, cmd | CMD_CMPL_INTR_ENABLE); - dbg("%s : Exit\n", __FUNCTION__); DBG_LEAVE_ROUTINE return retval; @@ -509,7 +481,6 @@ static int hpc_query_power_fault(struct slot * slot) u16 slot_status; u8 pwr_fault; int retval = 0; - u8 status; DBG_ENTER_ROUTINE @@ -521,15 +492,13 @@ static int hpc_query_power_fault(struct slot * slot) retval = hp_register_read_word(php_ctlr->pci_dev, SLOT_STATUS(slot->ctrl->cap_base), slot_status); if (retval) { - err("%s : hp_register_read_word SLOT_STATUS failed\n", __FUNCTION__); + err("%s : Cannot check for power fault\n", __FUNCTION__); return retval; } pwr_fault = (u8)((slot_status & PWR_FAULT_DETECTED) >> 1); - status = (pwr_fault != 1) ? 1 : 0; DBG_LEAVE_ROUTINE - /* Note: Logic 0 => fault */ - return status; + return pwr_fault; } static int hpc_set_attention_status(struct slot *slot, u8 value) @@ -539,7 +508,8 @@ static int hpc_set_attention_status(struct slot *slot, u8 value) u16 slot_ctrl; int rc = 0; - dbg("%s: \n", __FUNCTION__); + DBG_ENTER_ROUTINE + if (!php_ctlr) { err("%s: Invalid HPC controller handle!\n", __FUNCTION__); return -1; @@ -555,7 +525,6 @@ static int hpc_set_attention_status(struct slot *slot, u8 value) err("%s : hp_register_read_word SLOT_CTRL failed\n", __FUNCTION__); return rc; } - dbg("%s : hp_register_read_word SLOT_CTRL %x\n", __FUNCTION__, slot_ctrl); switch (value) { case 0 : /* turn off */ @@ -576,6 +545,7 @@ static int hpc_set_attention_status(struct slot *slot, u8 value) pcie_write_cmd(slot, slot_cmd); dbg("%s: SLOT_CTRL %x write cmd %x\n", __FUNCTION__, SLOT_CTRL(slot->ctrl->cap_base), slot_cmd); + DBG_LEAVE_ROUTINE return rc; } @@ -587,7 +557,8 @@ static void hpc_set_green_led_on(struct slot *slot) u16 slot_ctrl; int rc = 0; - dbg("%s: \n", __FUNCTION__); + DBG_ENTER_ROUTINE + if (!php_ctlr) { err("%s: Invalid HPC controller handle!\n", __FUNCTION__); return ; @@ -604,7 +575,6 @@ static void hpc_set_green_led_on(struct slot *slot) err("%s : hp_register_read_word SLOT_CTRL failed\n", __FUNCTION__); return; } - dbg("%s : hp_register_read_word SLOT_CTRL %x\n", __FUNCTION__, slot_ctrl); slot_cmd = (slot_ctrl & ~PWR_LED_CTRL) | 0x0100; if (!pciehp_poll_mode) slot_cmd = slot_cmd | HP_INTR_ENABLE; @@ -612,6 +582,7 @@ static void hpc_set_green_led_on(struct slot *slot) pcie_write_cmd(slot, slot_cmd); dbg("%s: SLOT_CTRL %x write cmd %x\n",__FUNCTION__, SLOT_CTRL(slot->ctrl->cap_base), slot_cmd); + DBG_LEAVE_ROUTINE return; } @@ -622,7 +593,8 @@ static void hpc_set_green_led_off(struct slot *slot) u16 slot_ctrl; int rc = 0; - dbg("%s: \n", __FUNCTION__); + DBG_ENTER_ROUTINE + if (!php_ctlr) { err("%s: Invalid HPC controller handle!\n", __FUNCTION__); return ; @@ -639,7 +611,6 @@ static void hpc_set_green_led_off(struct slot *slot) err("%s : hp_register_read_word SLOT_CTRL failed\n", __FUNCTION__); return; } - dbg("%s : hp_register_read_word SLOT_CTRL %x\n", __FUNCTION__, slot_ctrl); slot_cmd = (slot_ctrl & ~PWR_LED_CTRL) | 0x0300; @@ -648,6 +619,7 @@ static void hpc_set_green_led_off(struct slot *slot) pcie_write_cmd(slot, slot_cmd); dbg("%s: SLOT_CTRL %x write cmd %x\n", __FUNCTION__, SLOT_CTRL(slot->ctrl->cap_base), slot_cmd); + DBG_LEAVE_ROUTINE return; } @@ -658,7 +630,8 @@ static void hpc_set_green_led_blink(struct slot *slot) u16 slot_ctrl; int rc = 0; - dbg("%s: \n", __FUNCTION__); + DBG_ENTER_ROUTINE + if (!php_ctlr) { err("%s: Invalid HPC controller handle!\n", __FUNCTION__); return ; @@ -675,7 +648,6 @@ static void hpc_set_green_led_blink(struct slot *slot) err("%s : hp_register_read_word SLOT_CTRL failed\n", __FUNCTION__); return; } - dbg("%s : hp_register_read_word SLOT_CTRL %x\n", __FUNCTION__, slot_ctrl); slot_cmd = (slot_ctrl & ~PWR_LED_CTRL) | 0x0200; @@ -684,6 +656,7 @@ static void hpc_set_green_led_blink(struct slot *slot) pcie_write_cmd(slot, slot_cmd); dbg("%s: SLOT_CTRL %x write cmd %x\n",__FUNCTION__, SLOT_CTRL(slot->ctrl->cap_base), slot_cmd); + DBG_LEAVE_ROUTINE return; } @@ -780,7 +753,6 @@ static int hpc_power_on_slot(struct slot * slot) int retval = 0; DBG_ENTER_ROUTINE - dbg("%s: \n", __FUNCTION__); if (!php_ctlr) { err("%s: Invalid HPC controller handle!\n", __FUNCTION__); @@ -799,8 +771,6 @@ static int hpc_power_on_slot(struct slot * slot) err("%s : hp_register_read_word SLOT_CTRL failed\n", __FUNCTION__); return retval; } - dbg("%s: SLOT_CTRL %x, value read %xn", __FUNCTION__, SLOT_CTRL(slot->ctrl->cap_base), - slot_ctrl); slot_cmd = (slot_ctrl & ~PWR_CTRL) | POWER_ON; @@ -829,7 +799,6 @@ static int hpc_power_off_slot(struct slot * slot) int retval = 0; DBG_ENTER_ROUTINE - dbg("%s: \n", __FUNCTION__); if (!php_ctlr) { err("%s: Invalid HPC controller handle!\n", __FUNCTION__); @@ -848,8 +817,6 @@ static int hpc_power_off_slot(struct slot * slot) err("%s : hp_register_read_word SLOT_CTRL failed\n", __FUNCTION__); return retval; } - dbg("%s: SLOT_CTRL %x, value read %x\n", __FUNCTION__, SLOT_CTRL(slot->ctrl->cap_base), - slot_ctrl); slot_cmd = (slot_ctrl & ~PWR_CTRL) | POWER_OFF; @@ -924,7 +891,6 @@ static irqreturn_t pcie_isr(int IRQ, void *dev_id, struct pt_regs *regs) return IRQ_NONE; } - dbg("%s: Set Mask Hot-plug Interrupt Enable\n", __FUNCTION__); dbg("%s: hp_register_read_word SLOT_CTRL with value %x\n", __FUNCTION__, temp_word); temp_word = (temp_word & ~HP_INTR_ENABLE & ~CMD_CMPL_INTR_ENABLE) | 0x00; @@ -933,7 +899,6 @@ static irqreturn_t pcie_isr(int IRQ, void *dev_id, struct pt_regs *regs) err("%s : hp_register_write_word SLOT_CTRL failed\n", __FUNCTION__); return IRQ_NONE; } - dbg("%s: hp_register_write_word SLOT_CTRL with value %x\n", __FUNCTION__, temp_word); rc = hp_register_read_word(php_ctlr->pci_dev, SLOT_STATUS(ctrl->cap_base), slot_status); if (rc) { @@ -949,14 +914,12 @@ static irqreturn_t pcie_isr(int IRQ, void *dev_id, struct pt_regs *regs) err("%s : hp_register_write_word SLOT_STATUS failed\n", __FUNCTION__); return IRQ_NONE; } - dbg("%s: hp_register_write_word SLOT_STATUS with value %x\n", __FUNCTION__, temp_word); } if (intr_loc & CMD_COMPLETED) { /* * Command Complete Interrupt Pending */ - dbg("%s: In Command Complete Interrupt Pending\n", __FUNCTION__); wake_up_interruptible(&ctrl->queue); } @@ -989,7 +952,6 @@ static irqreturn_t pcie_isr(int IRQ, void *dev_id, struct pt_regs *regs) } dbg("%s: Unmask Hot-plug Interrupt Enable\n", __FUNCTION__); - dbg("%s: hp_register_read_word SLOT_CTRL with value %x\n", __FUNCTION__, temp_word); temp_word = (temp_word & ~HP_INTR_ENABLE) | HP_INTR_ENABLE; rc = hp_register_write_word(php_ctlr->pci_dev, SLOT_CTRL(ctrl->cap_base), temp_word); @@ -997,14 +959,12 @@ static irqreturn_t pcie_isr(int IRQ, void *dev_id, struct pt_regs *regs) err("%s : hp_register_write_word SLOT_CTRL failed\n", __FUNCTION__); return IRQ_NONE; } - dbg("%s: hp_register_write_word SLOT_CTRL with value %x\n", __FUNCTION__, temp_word); rc = hp_register_read_word(php_ctlr->pci_dev, SLOT_STATUS(ctrl->cap_base), slot_status); if (rc) { err("%s : hp_register_read_word SLOT_STATUS failed\n", __FUNCTION__); return IRQ_NONE; } - dbg("%s: hp_register_read_word SLOT_STATUS with value %x\n", __FUNCTION__, slot_status); /* Clear command complete interrupt caused by this write */ temp_word = 0x1F; @@ -1248,12 +1208,7 @@ static struct hpc_ops pciehp_hpc_ops = { .check_lnk_status = hpc_check_lnk_status, }; -int pcie_init(struct controller * ctrl, - struct pcie_device *dev, - php_intr_callback_t attention_button_callback, - php_intr_callback_t switch_change_callback, - php_intr_callback_t presence_change_callback, - php_intr_callback_t power_fault_callback) +int pcie_init(struct controller * ctrl, struct pcie_device *dev) { struct php_ctlr_state_s *php_ctlr, *p; void *instance_id = ctrl; @@ -1282,8 +1237,8 @@ int pcie_init(struct controller * ctrl, pdev = dev->port; php_ctlr->pci_dev = pdev; /* save pci_dev in context */ - dbg("%s: pdev->vendor %x pdev->device %x\n", __FUNCTION__, - pdev->vendor, pdev->device); + dbg("%s: hotplug controller vendor id 0x%x device id 0x%x\n", + __FUNCTION__, pdev->vendor, pdev->device); saved_cap_base = pcie_cap_base; @@ -1340,8 +1295,6 @@ int pcie_init(struct controller * ctrl, first = 0; } - dbg("pdev = %p: b:d:f:irq=0x%x:%x:%x:%x\n", pdev, pdev->bus->number, - PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn), dev->irq); for ( rc = 0; rc < DEVICE_COUNT_RESOURCE; rc++) if (pci_resource_len(pdev, rc) > 0) dbg("pci resource[%d] start=0x%lx(len=0x%lx)\n", rc, @@ -1359,13 +1312,12 @@ int pcie_init(struct controller * ctrl, /* find the IRQ */ php_ctlr->irq = dev->irq; - dbg("HPC interrupt = %d\n", php_ctlr->irq); /* Save interrupt callback info */ - php_ctlr->attention_button_callback = attention_button_callback; - php_ctlr->switch_change_callback = switch_change_callback; - php_ctlr->presence_change_callback = presence_change_callback; - php_ctlr->power_fault_callback = power_fault_callback; + php_ctlr->attention_button_callback = pciehp_handle_attention_button; + php_ctlr->switch_change_callback = pciehp_handle_switch_change; + php_ctlr->presence_change_callback = pciehp_handle_presence_change; + php_ctlr->power_fault_callback = pciehp_handle_power_fault; php_ctlr->callback_instance_id = instance_id; /* return PCI Controller Info */ @@ -1387,15 +1339,12 @@ int pcie_init(struct controller * ctrl, err("%s : hp_register_write_word SLOT_CTRL failed\n", __FUNCTION__); goto abort_free_ctlr; } - dbg("%s : Mask HPIE hp_register_write_word SLOT_CTRL %x\n", __FUNCTION__, temp_word); rc = hp_register_read_word(php_ctlr->pci_dev, SLOT_STATUS(ctrl->cap_base), slot_status); if (rc) { err("%s : hp_register_read_word SLOT_STATUS failed\n", __FUNCTION__); goto abort_free_ctlr; } - dbg("%s: Mask HPIE SLOT_STATUS offset %x reads slot_status %x\n", __FUNCTION__, SLOT_STATUS(ctrl->cap_base) - , slot_status); temp_word = 0x1F; /* Clear all events */ rc = hp_register_write_word(php_ctlr->pci_dev, SLOT_STATUS(ctrl->cap_base), temp_word); @@ -1403,7 +1352,6 @@ int pcie_init(struct controller * ctrl, err("%s : hp_register_write_word SLOT_STATUS failed\n", __FUNCTION__); goto abort_free_ctlr; } - dbg("%s: SLOT_STATUS offset %x writes slot_status %x\n", __FUNCTION__, SLOT_STATUS(ctrl->cap_base), temp_word); if (pciehp_poll_mode) {/* Install interrupt polling code */ /* Install and start the interrupt polling timer */ @@ -1419,13 +1367,14 @@ int pcie_init(struct controller * ctrl, } } + dbg("pciehp ctrl b:d:f:irq=0x%x:%x:%x:%x\n", pdev->bus->number, + PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn), dev->irq); + rc = hp_register_read_word(pdev, SLOT_CTRL(ctrl->cap_base), temp_word); if (rc) { err("%s : hp_register_read_word SLOT_CTRL failed\n", __FUNCTION__); goto abort_free_ctlr; } - dbg("%s: SLOT_CTRL %x value read %x\n", __FUNCTION__, SLOT_CTRL(ctrl->cap_base), temp_word); - dbg("%s: slot_cap %x\n", __FUNCTION__, slot_cap); intr_enable = intr_enable | PRSN_DETECT_ENABLE; @@ -1445,7 +1394,6 @@ int pcie_init(struct controller * ctrl, } else { temp_word = (temp_word & ~HP_INTR_ENABLE) | HP_INTR_ENABLE; } - dbg("%s: temp_word %x\n", __FUNCTION__, temp_word); /* Unmask Hot-plug Interrupt Enable for the interrupt notification mechanism case */ rc = hp_register_write_word(pdev, SLOT_CTRL(ctrl->cap_base), temp_word); @@ -1453,14 +1401,11 @@ int pcie_init(struct controller * ctrl, err("%s : hp_register_write_word SLOT_CTRL failed\n", __FUNCTION__); goto abort_free_ctlr; } - dbg("%s : Unmask HPIE hp_register_write_word SLOT_CTRL with %x\n", __FUNCTION__, temp_word); rc = hp_register_read_word(php_ctlr->pci_dev, SLOT_STATUS(ctrl->cap_base), slot_status); if (rc) { err("%s : hp_register_read_word SLOT_STATUS failed\n", __FUNCTION__); goto abort_free_ctlr; } - dbg("%s: Unmask HPIE SLOT_STATUS offset %x reads slot_status %x\n", __FUNCTION__, - SLOT_STATUS(ctrl->cap_base), slot_status); temp_word = 0x1F; /* Clear all events */ rc = hp_register_write_word(php_ctlr->pci_dev, SLOT_STATUS(ctrl->cap_base), temp_word); @@ -1468,8 +1413,16 @@ int pcie_init(struct controller * ctrl, err("%s : hp_register_write_word SLOT_STATUS failed\n", __FUNCTION__); goto abort_free_ctlr; } - dbg("%s: SLOT_STATUS offset %x writes slot_status %x\n", __FUNCTION__, SLOT_STATUS(ctrl->cap_base), temp_word); + if (pciehp_force) { + dbg("Bypassing BIOS check for pciehp use on %s\n", + pci_name(ctrl->pci_dev)); + } else { + rc = pciehp_get_hp_hw_control_from_firmware(ctrl->pci_dev); + if (rc) + goto abort_free_ctlr; + } + /* Add this HPC instance into the HPC list */ spin_lock(&list_lock); if (php_ctlr_list_head == 0) { diff --git a/drivers/pci/hotplug/pciehp_pci.c b/drivers/pci/hotplug/pciehp_pci.c index ff17d8e07e9..647673a7d22 100644 --- a/drivers/pci/hotplug/pciehp_pci.c +++ b/drivers/pci/hotplug/pciehp_pci.c @@ -27,801 +27,111 @@ * */ -#include <linux/config.h> #include <linux/module.h> #include <linux/kernel.h> #include <linux/types.h> -#include <linux/slab.h> -#include <linux/workqueue.h> -#include <linux/proc_fs.h> #include <linux/pci.h> #include "../pci.h" #include "pciehp.h" -#ifndef CONFIG_IA64 -#include "../../../arch/i386/pci/pci.h" /* horrible hack showing how processor dependant we are... */ -#endif -int pciehp_configure_device (struct controller* ctrl, struct pci_func* func) +int pciehp_configure_device(struct slot *p_slot) { - unsigned char bus; - struct pci_bus *child; - int num; - - if (func->pci_dev == NULL) - func->pci_dev = pci_find_slot(func->bus, PCI_DEVFN(func->device, func->function)); - - /* Still NULL ? Well then scan for it ! */ - if (func->pci_dev == NULL) { - dbg("%s: pci_dev still null. do pci_scan_slot\n", __FUNCTION__); - - num = pci_scan_slot(ctrl->pci_dev->subordinate, PCI_DEVFN(func->device, func->function)); - - if (num) - pci_bus_add_devices(ctrl->pci_dev->subordinate); - - func->pci_dev = pci_find_slot(func->bus, PCI_DEVFN(func->device, func->function)); - if (func->pci_dev == NULL) { - dbg("ERROR: pci_dev still null\n"); - return 0; - } + struct pci_dev *dev; + struct pci_bus *parent = p_slot->ctrl->pci_dev->subordinate; + int num, fn; + + dev = pci_find_slot(p_slot->bus, PCI_DEVFN(p_slot->device, 0)); + if (dev) { + err("Device %s already exists at %x:%x, cannot hot-add\n", + pci_name(dev), p_slot->bus, p_slot->device); + return -EINVAL; } - if (func->pci_dev->hdr_type == PCI_HEADER_TYPE_BRIDGE) { - pci_read_config_byte(func->pci_dev, PCI_SECONDARY_BUS, &bus); - child = pci_add_new_bus(func->pci_dev->bus, (func->pci_dev), bus); - pci_do_scan_bus(child); + num = pci_scan_slot(parent, PCI_DEVFN(p_slot->device, 0)); + if (num == 0) { + err("No new device found\n"); + return -ENODEV; + } + for (fn = 0; fn < 8; fn++) { + if (!(dev = pci_find_slot(p_slot->bus, + PCI_DEVFN(p_slot->device, fn)))) + continue; + if ((dev->class >> 16) == PCI_BASE_CLASS_DISPLAY) { + err("Cannot hot-add display device %s\n", + pci_name(dev)); + continue; + } + if ((dev->hdr_type == PCI_HEADER_TYPE_BRIDGE) || + (dev->hdr_type == PCI_HEADER_TYPE_CARDBUS)) { + /* Find an unused bus number for the new bridge */ + struct pci_bus *child; + unsigned char busnr, start = parent->secondary; + unsigned char end = parent->subordinate; + for (busnr = start; busnr <= end; busnr++) { + if (!pci_find_bus(pci_domain_nr(parent), + busnr)) + break; + } + if (busnr >= end) { + err("No free bus for hot-added bridge\n"); + continue; + } + child = pci_add_new_bus(parent, dev, busnr); + if (!child) { + err("Cannot add new bus for %s\n", + pci_name(dev)); + continue; + } + child->subordinate = pci_do_scan_bus(child); + pci_bus_size_bridges(child); + } + /* TBD: program firmware provided _HPP values */ + /* program_fw_provided_values(dev); */ } + pci_bus_assign_resources(parent); + pci_bus_add_devices(parent); + pci_enable_bridges(parent); return 0; } - -int pciehp_unconfigure_device(struct pci_func* func) +int pciehp_unconfigure_device(struct slot *p_slot) { int rc = 0; int j; - struct pci_bus *pbus; + u8 bctl = 0; - dbg("%s: bus/dev/func = %x/%x/%x\n", __FUNCTION__, func->bus, - func->device, func->function); - pbus = func->pci_dev->bus; + dbg("%s: bus/dev = %x/%x\n", __FUNCTION__, p_slot->bus, + p_slot->device); for (j=0; j<8 ; j++) { - struct pci_dev* temp = pci_find_slot(func->bus, - (func->device << 3) | j); - if (temp) { - pci_remove_bus_device(temp); + struct pci_dev* temp = pci_find_slot(p_slot->bus, + (p_slot->device << 3) | j); + if (!temp) + continue; + if ((temp->class >> 16) == PCI_BASE_CLASS_DISPLAY) { + err("Cannot remove display device %s\n", + pci_name(temp)); + continue; } + if (temp->hdr_type == PCI_HEADER_TYPE_BRIDGE) { + pci_read_config_byte(temp, PCI_BRIDGE_CONTROL, &bctl); + if (bctl & PCI_BRIDGE_CTL_VGA) { + err("Cannot remove display device %s\n", + pci_name(temp)); + continue; + } + } + pci_remove_bus_device(temp); } /* * Some PCI Express root ports require fixup after hot-plug operation. */ if (pcie_mch_quirk) - pci_fixup_device(pci_fixup_final, pbus->self); + pci_fixup_device(pci_fixup_final, p_slot->ctrl->pci_dev); return rc; } -/* - * pciehp_set_irq - * - * @bus_num: bus number of PCI device - * @dev_num: device number of PCI device - * @slot: pointer to u8 where slot number will be returned - */ -int pciehp_set_irq (u8 bus_num, u8 dev_num, u8 int_pin, u8 irq_num) -{ -#if defined(CONFIG_X86_32) && !defined(CONFIG_X86_IO_APIC) - int rc; - u16 temp_word; - struct pci_dev fakedev; - struct pci_bus fakebus; - - fakedev.devfn = dev_num << 3; - fakedev.bus = &fakebus; - fakebus.number = bus_num; - dbg("%s: dev %d, bus %d, pin %d, num %d\n", - __FUNCTION__, dev_num, bus_num, int_pin, irq_num); - rc = pcibios_set_irq_routing(&fakedev, int_pin - 0x0a, irq_num); - dbg("%s: rc %d\n", __FUNCTION__, rc); - if (!rc) - return !rc; - - /* set the Edge Level Control Register (ELCR) */ - temp_word = inb(0x4d0); - temp_word |= inb(0x4d1) << 8; - - temp_word |= 0x01 << irq_num; - - /* This should only be for x86 as it sets the Edge Level Control Register */ - outb((u8) (temp_word & 0xFF), 0x4d0); - outb((u8) ((temp_word & 0xFF00) >> 8), 0x4d1); -#endif - return 0; -} - -/* More PCI configuration routines; this time centered around hotplug controller */ - - -/* - * pciehp_save_config - * - * Reads configuration for all slots in a PCI bus and saves info. - * - * Note: For non-hot plug busses, the slot # saved is the device # - * - * returns 0 if success - */ -int pciehp_save_config(struct controller *ctrl, int busnumber, int num_ctlr_slots, int first_device_num) -{ - int rc; - u8 class_code; - u8 header_type; - u32 ID; - u8 secondary_bus; - struct pci_func *new_slot; - int sub_bus; - int max_functions; - int function; - u8 DevError; - int device = 0; - int cloop = 0; - int stop_it; - int index; - int is_hot_plug = num_ctlr_slots || first_device_num; - struct pci_bus lpci_bus, *pci_bus; - int FirstSupported, LastSupported; - - dbg("%s: Enter\n", __FUNCTION__); - - memcpy(&lpci_bus, ctrl->pci_dev->subordinate, sizeof(lpci_bus)); - pci_bus = &lpci_bus; - - dbg("%s: num_ctlr_slots = %d, first_device_num = %d\n", __FUNCTION__, - num_ctlr_slots, first_device_num); - - /* Decide which slots are supported */ - if (is_hot_plug) { - /********************************* - * is_hot_plug is the slot mask - *********************************/ - FirstSupported = first_device_num; - LastSupported = FirstSupported + num_ctlr_slots - 1; - } else { - FirstSupported = 0; - LastSupported = 0x1F; - } - - dbg("FirstSupported = %d, LastSupported = %d\n", FirstSupported, - LastSupported); - - /* Save PCI configuration space for all devices in supported slots */ - dbg("%s: pci_bus->number = %x\n", __FUNCTION__, pci_bus->number); - pci_bus->number = busnumber; - dbg("%s: bus = %x, dev = %x\n", __FUNCTION__, busnumber, device); - for (device = FirstSupported; device <= LastSupported; device++) { - ID = 0xFFFFFFFF; - rc = pci_bus_read_config_dword(pci_bus, PCI_DEVFN(device, 0), - PCI_VENDOR_ID, &ID); - - if (ID != 0xFFFFFFFF) { /* device in slot */ - dbg("%s: ID = %x\n", __FUNCTION__, ID); - rc = pci_bus_read_config_byte(pci_bus, PCI_DEVFN(device, 0), - 0x0B, &class_code); - if (rc) - return rc; - - rc = pci_bus_read_config_byte(pci_bus, PCI_DEVFN(device, 0), - PCI_HEADER_TYPE, &header_type); - if (rc) - return rc; - - dbg("class_code = %x, header_type = %x\n", class_code, header_type); - - /* If multi-function device, set max_functions to 8 */ - if (header_type & 0x80) - max_functions = 8; - else - max_functions = 1; - - function = 0; - - do { - DevError = 0; - dbg("%s: In do loop\n", __FUNCTION__); - - if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { /* P-P Bridge */ - /* Recurse the subordinate bus - * get the subordinate bus number - */ - rc = pci_bus_read_config_byte(pci_bus, - PCI_DEVFN(device, function), - PCI_SECONDARY_BUS, &secondary_bus); - if (rc) { - return rc; - } else { - sub_bus = (int) secondary_bus; - - /* Save secondary bus cfg spc with this recursive call. */ - rc = pciehp_save_config(ctrl, sub_bus, 0, 0); - if (rc) - return rc; - } - } - - index = 0; - new_slot = pciehp_slot_find(busnumber, device, index++); - - dbg("%s: new_slot = %p bus %x dev %x fun %x\n", - __FUNCTION__, new_slot, busnumber, device, index-1); - - while (new_slot && (new_slot->function != (u8) function)) { - new_slot = pciehp_slot_find(busnumber, device, index++); - dbg("%s: while loop, new_slot = %p bus %x dev %x fun %x\n", - __FUNCTION__, new_slot, busnumber, device, index-1); - } - if (!new_slot) { - /* Setup slot structure. */ - new_slot = pciehp_slot_create(busnumber); - dbg("%s: if, new_slot = %p bus %x dev %x fun %x\n", - __FUNCTION__, new_slot, busnumber, device, function); - - if (new_slot == NULL) - return(1); - } - - new_slot->bus = (u8) busnumber; - new_slot->device = (u8) device; - new_slot->function = (u8) function; - new_slot->is_a_board = 1; - new_slot->switch_save = 0x10; - /* In case of unsupported board */ - new_slot->status = DevError; - new_slot->pci_dev = pci_find_slot(new_slot->bus, - (new_slot->device << 3) | new_slot->function); - dbg("new_slot->pci_dev = %p\n", new_slot->pci_dev); - - for (cloop = 0; cloop < 0x20; cloop++) { - rc = pci_bus_read_config_dword(pci_bus, - PCI_DEVFN(device, function), - cloop << 2, - (u32 *) &(new_slot->config_space [cloop])); - /* dbg("new_slot->config_space[%x] = %x\n", - cloop, new_slot->config_space[cloop]); */ - if (rc) - return rc; - } - - function++; - - stop_it = 0; - - /* this loop skips to the next present function - * reading in Class Code and Header type. - */ - - while ((function < max_functions)&&(!stop_it)) { - dbg("%s: In while loop \n", __FUNCTION__); - rc = pci_bus_read_config_dword(pci_bus, - PCI_DEVFN(device, function), - PCI_VENDOR_ID, &ID); - - if (ID == 0xFFFFFFFF) { /* nothing there. */ - function++; - dbg("Nothing there\n"); - } else { /* Something there */ - rc = pci_bus_read_config_byte(pci_bus, - PCI_DEVFN(device, function), - 0x0B, &class_code); - if (rc) - return rc; - - rc = pci_bus_read_config_byte(pci_bus, - PCI_DEVFN(device, function), - PCI_HEADER_TYPE, &header_type); - if (rc) - return rc; - - dbg("class_code = %x, header_type = %x\n", class_code, header_type); - stop_it++; - } - } - - } while (function < max_functions); - /* End of IF (device in slot?) */ - } else if (is_hot_plug) { - /* Setup slot structure with entry for empty slot */ - new_slot = pciehp_slot_create(busnumber); - - if (new_slot == NULL) { - return(1); - } - dbg("new_slot = %p, bus = %x, dev = %x, fun = %x\n", new_slot, - new_slot->bus, new_slot->device, new_slot->function); - - new_slot->bus = (u8) busnumber; - new_slot->device = (u8) device; - new_slot->function = 0; - new_slot->is_a_board = 0; - new_slot->presence_save = 0; - new_slot->switch_save = 0; - } - } /* End of FOR loop */ - - dbg("%s: Exit\n", __FUNCTION__); - return(0); -} - - -/* - * pciehp_save_slot_config - * - * Saves configuration info for all PCI devices in a given slot - * including subordinate busses. - * - * returns 0 if success - */ -int pciehp_save_slot_config(struct controller *ctrl, struct pci_func * new_slot) -{ - int rc; - u8 class_code; - u8 header_type; - u32 ID; - u8 secondary_bus; - int sub_bus; - int max_functions; - int function; - int cloop = 0; - int stop_it; - struct pci_bus lpci_bus, *pci_bus; - memcpy(&lpci_bus, ctrl->pci_dev->subordinate, sizeof(lpci_bus)); - pci_bus = &lpci_bus; - pci_bus->number = new_slot->bus; - - ID = 0xFFFFFFFF; - - pci_bus_read_config_dword(pci_bus, PCI_DEVFN(new_slot->device, 0), - PCI_VENDOR_ID, &ID); - - if (ID != 0xFFFFFFFF) { /* device in slot */ - pci_bus_read_config_byte(pci_bus, PCI_DEVFN(new_slot->device, 0), - 0x0B, &class_code); - - pci_bus_read_config_byte(pci_bus, PCI_DEVFN(new_slot->device, 0), - PCI_HEADER_TYPE, &header_type); - - if (header_type & 0x80) /* Multi-function device */ - max_functions = 8; - else - max_functions = 1; - - function = 0; - - do { - if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { /* PCI-PCI Bridge */ - /* Recurse the subordinate bus */ - pci_bus_read_config_byte(pci_bus, - PCI_DEVFN(new_slot->device, function), - PCI_SECONDARY_BUS, &secondary_bus); - - sub_bus = (int) secondary_bus; - - /* Save the config headers for the secondary bus. */ - rc = pciehp_save_config(ctrl, sub_bus, 0, 0); - - if (rc) - return rc; - - } /* End of IF */ - - new_slot->status = 0; - - for (cloop = 0; cloop < 0x20; cloop++) { - pci_bus_read_config_dword(pci_bus, - PCI_DEVFN(new_slot->device, function), - cloop << 2, - (u32 *) &(new_slot->config_space [cloop])); - } - - function++; - - stop_it = 0; - - /* this loop skips to the next present function - * reading in the Class Code and the Header type. - */ - - while ((function < max_functions) && (!stop_it)) { - pci_bus_read_config_dword(pci_bus, - PCI_DEVFN(new_slot->device, function), - PCI_VENDOR_ID, &ID); - - if (ID == 0xFFFFFFFF) { /* nothing there. */ - function++; - } else { /* Something there */ - pci_bus_read_config_byte(pci_bus, - PCI_DEVFN(new_slot->device, function), - 0x0B, &class_code); - - pci_bus_read_config_byte(pci_bus, - PCI_DEVFN(new_slot->device, function), - PCI_HEADER_TYPE, &header_type); - - stop_it++; - } - } - - } while (function < max_functions); - } /* End of IF (device in slot?) */ - else { - return 2; - } - - return 0; -} - - -/* - * pciehp_save_used_resources - * - * Stores used resource information for existing boards. this is - * for boards that were in the system when this driver was loaded. - * this function is for hot plug ADD - * - * returns 0 if success - * if disable == 1(DISABLE_CARD), - * it loops for all functions of the slot and disables them. - * else, it just get resources of the function and return. - */ -int pciehp_save_used_resources(struct controller *ctrl, struct pci_func *func, int disable) -{ - u8 cloop; - u8 header_type; - u8 secondary_bus; - u8 temp_byte; - u16 command; - u16 save_command; - u16 w_base, w_length; - u32 temp_register; - u32 save_base; - u32 base, length; - u64 base64 = 0; - int index = 0; - unsigned int devfn; - struct pci_resource *mem_node = NULL; - struct pci_resource *p_mem_node = NULL; - struct pci_resource *t_mem_node; - struct pci_resource *io_node; - struct pci_resource *bus_node; - struct pci_bus lpci_bus, *pci_bus; - memcpy(&lpci_bus, ctrl->pci_dev->subordinate, sizeof(lpci_bus)); - pci_bus = &lpci_bus; - - if (disable) - func = pciehp_slot_find(func->bus, func->device, index++); - - while ((func != NULL) && func->is_a_board) { - pci_bus->number = func->bus; - devfn = PCI_DEVFN(func->device, func->function); - - /* Save the command register */ - pci_bus_read_config_word(pci_bus, devfn, PCI_COMMAND, &save_command); - - if (disable) { - /* disable card */ - command = 0x00; - pci_bus_write_config_word(pci_bus, devfn, PCI_COMMAND, command); - } - - /* Check for Bridge */ - pci_bus_read_config_byte(pci_bus, devfn, PCI_HEADER_TYPE, &header_type); - - if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { /* PCI-PCI Bridge */ - dbg("Save_used_res of PCI bridge b:d=0x%x:%x, sc=0x%x\n", - func->bus, func->device, save_command); - if (disable) { - /* Clear Bridge Control Register */ - command = 0x00; - pci_bus_write_config_word(pci_bus, devfn, PCI_BRIDGE_CONTROL, command); - } - - pci_bus_read_config_byte(pci_bus, devfn, PCI_SECONDARY_BUS, &secondary_bus); - pci_bus_read_config_byte(pci_bus, devfn, PCI_SUBORDINATE_BUS, &temp_byte); - - bus_node = kmalloc(sizeof(struct pci_resource), - GFP_KERNEL); - if (!bus_node) - return -ENOMEM; - - bus_node->base = (ulong)secondary_bus; - bus_node->length = (ulong)(temp_byte - secondary_bus + 1); - - bus_node->next = func->bus_head; - func->bus_head = bus_node; - - /* Save IO base and Limit registers */ - pci_bus_read_config_byte(pci_bus, devfn, PCI_IO_BASE, &temp_byte); - base = temp_byte; - pci_bus_read_config_byte(pci_bus, devfn, PCI_IO_LIMIT, &temp_byte); - length = temp_byte; - - if ((base <= length) && (!disable || (save_command & PCI_COMMAND_IO))) { - io_node = kmalloc(sizeof(struct pci_resource), - GFP_KERNEL); - if (!io_node) - return -ENOMEM; - - io_node->base = (ulong)(base & PCI_IO_RANGE_MASK) << 8; - io_node->length = (ulong)(length - base + 0x10) << 8; - - io_node->next = func->io_head; - func->io_head = io_node; - } - - /* Save memory base and Limit registers */ - pci_bus_read_config_word(pci_bus, devfn, PCI_MEMORY_BASE, &w_base); - pci_bus_read_config_word(pci_bus, devfn, PCI_MEMORY_LIMIT, &w_length); - - if ((w_base <= w_length) && (!disable || (save_command & PCI_COMMAND_MEMORY))) { - mem_node = kmalloc(sizeof(struct pci_resource), - GFP_KERNEL); - if (!mem_node) - return -ENOMEM; - - mem_node->base = (ulong)w_base << 16; - mem_node->length = (ulong)(w_length - w_base + 0x10) << 16; - - mem_node->next = func->mem_head; - func->mem_head = mem_node; - } - /* Save prefetchable memory base and Limit registers */ - pci_bus_read_config_word(pci_bus, devfn, PCI_PREF_MEMORY_BASE, &w_base); - pci_bus_read_config_word(pci_bus, devfn, PCI_PREF_MEMORY_LIMIT, &w_length); - - if ((w_base <= w_length) && (!disable || (save_command & PCI_COMMAND_MEMORY))) { - p_mem_node = kmalloc(sizeof(struct pci_resource), - GFP_KERNEL); - if (!p_mem_node) - return -ENOMEM; - - p_mem_node->base = (ulong)w_base << 16; - p_mem_node->length = (ulong)(w_length - w_base + 0x10) << 16; - - p_mem_node->next = func->p_mem_head; - func->p_mem_head = p_mem_node; - } - } else if ((header_type & 0x7F) == PCI_HEADER_TYPE_NORMAL) { - dbg("Save_used_res of PCI adapter b:d=0x%x:%x, sc=0x%x\n", - func->bus, func->device, save_command); - - /* Figure out IO and memory base lengths */ - for (cloop = PCI_BASE_ADDRESS_0; cloop <= PCI_BASE_ADDRESS_5; cloop += 4) { - pci_bus_read_config_dword(pci_bus, devfn, cloop, &save_base); - - temp_register = 0xFFFFFFFF; - pci_bus_write_config_dword(pci_bus, devfn, cloop, temp_register); - pci_bus_read_config_dword(pci_bus, devfn, cloop, &temp_register); - - if (!disable) - pci_bus_write_config_dword(pci_bus, devfn, cloop, save_base); - - if (!temp_register) - continue; - - base = temp_register; - - if ((base & PCI_BASE_ADDRESS_SPACE_IO) && - (!disable || (save_command & PCI_COMMAND_IO))) { - /* IO base */ - /* set temp_register = amount of IO space requested */ - base = base & 0xFFFFFFFCL; - base = (~base) + 1; - - io_node = kmalloc(sizeof (struct pci_resource), - GFP_KERNEL); - if (!io_node) - return -ENOMEM; - - io_node->base = (ulong)save_base & PCI_BASE_ADDRESS_IO_MASK; - io_node->length = (ulong)base; - dbg("sur adapter: IO bar=0x%x(length=0x%x)\n", - io_node->base, io_node->length); - - io_node->next = func->io_head; - func->io_head = io_node; - } else { /* map Memory */ - int prefetchable = 1; - /* struct pci_resources **res_node; */ - char *res_type_str = "PMEM"; - u32 temp_register2; - - t_mem_node = kmalloc(sizeof (struct pci_resource), - GFP_KERNEL); - if (!t_mem_node) - return -ENOMEM; - - if (!(base & PCI_BASE_ADDRESS_MEM_PREFETCH) && - (!disable || (save_command & PCI_COMMAND_MEMORY))) { - prefetchable = 0; - mem_node = t_mem_node; - res_type_str++; - } else - p_mem_node = t_mem_node; - - base = base & 0xFFFFFFF0L; - base = (~base) + 1; - - switch (temp_register & PCI_BASE_ADDRESS_MEM_TYPE_MASK) { - case PCI_BASE_ADDRESS_MEM_TYPE_32: - if (prefetchable) { - p_mem_node->base = (ulong)save_base & PCI_BASE_ADDRESS_MEM_MASK; - p_mem_node->length = (ulong)base; - dbg("sur adapter: 32 %s bar=0x%x(length=0x%x)\n", - res_type_str, - p_mem_node->base, - p_mem_node->length); - - p_mem_node->next = func->p_mem_head; - func->p_mem_head = p_mem_node; - } else { - mem_node->base = (ulong)save_base & PCI_BASE_ADDRESS_MEM_MASK; - mem_node->length = (ulong)base; - dbg("sur adapter: 32 %s bar=0x%x(length=0x%x)\n", - res_type_str, - mem_node->base, - mem_node->length); - - mem_node->next = func->mem_head; - func->mem_head = mem_node; - } - break; - case PCI_BASE_ADDRESS_MEM_TYPE_64: - pci_bus_read_config_dword(pci_bus, devfn, cloop+4, &temp_register2); - base64 = temp_register2; - base64 = (base64 << 32) | save_base; - - if (temp_register2) { - dbg("sur adapter: 64 %s high dword of base64(0x%x:%x) masked to 0\n", - res_type_str, temp_register2, (u32)base64); - base64 &= 0x00000000FFFFFFFFL; - } - - if (prefetchable) { - p_mem_node->base = base64 & PCI_BASE_ADDRESS_MEM_MASK; - p_mem_node->length = base; - dbg("sur adapter: 64 %s base=0x%x(len=0x%x)\n", - res_type_str, - p_mem_node->base, - p_mem_node->length); - - p_mem_node->next = func->p_mem_head; - func->p_mem_head = p_mem_node; - } else { - mem_node->base = base64 & PCI_BASE_ADDRESS_MEM_MASK; - mem_node->length = base; - dbg("sur adapter: 64 %s base=0x%x(len=0x%x)\n", - res_type_str, - mem_node->base, - mem_node->length); - - mem_node->next = func->mem_head; - func->mem_head = mem_node; - } - cloop += 4; - break; - default: - dbg("asur: reserved BAR type=0x%x\n", - temp_register); - break; - } - } - } /* End of base register loop */ - } else { /* Some other unknown header type */ - dbg("Save_used_res of PCI unknown type b:d=0x%x:%x. skip.\n", - func->bus, func->device); - } - - /* find the next device in this slot */ - if (!disable) - break; - func = pciehp_slot_find(func->bus, func->device, index++); - } - - return 0; -} - - -/** - * kfree_resource_list: release memory of all list members - * @res: resource list to free - */ -static inline void -return_resource_list(struct pci_resource **func, struct pci_resource **res) -{ - struct pci_resource *node; - struct pci_resource *t_node; - - node = *func; - *func = NULL; - while (node) { - t_node = node->next; - return_resource(res, node); - node = t_node; - } -} - -/* - * pciehp_return_board_resources - * - * this routine returns all resources allocated to a board to - * the available pool. - * - * returns 0 if success - */ -int pciehp_return_board_resources(struct pci_func * func, - struct resource_lists * resources) -{ - int rc; - - dbg("%s\n", __FUNCTION__); - - if (!func) - return 1; - - return_resource_list(&(func->io_head),&(resources->io_head)); - return_resource_list(&(func->mem_head),&(resources->mem_head)); - return_resource_list(&(func->p_mem_head),&(resources->p_mem_head)); - return_resource_list(&(func->bus_head),&(resources->bus_head)); - - rc = pciehp_resource_sort_and_combine(&(resources->mem_head)); - rc |= pciehp_resource_sort_and_combine(&(resources->p_mem_head)); - rc |= pciehp_resource_sort_and_combine(&(resources->io_head)); - rc |= pciehp_resource_sort_and_combine(&(resources->bus_head)); - - return rc; -} - -/** - * kfree_resource_list: release memory of all list members - * @res: resource list to free - */ -static inline void -kfree_resource_list(struct pci_resource **r) -{ - struct pci_resource *res, *tres; - - res = *r; - *r = NULL; - - while (res) { - tres = res; - res = res->next; - kfree(tres); - } -} - -/** - * pciehp_destroy_resource_list: put node back in the resource list - * @resources: list to put nodes back - */ -void pciehp_destroy_resource_list(struct resource_lists * resources) -{ - kfree_resource_list(&(resources->io_head)); - kfree_resource_list(&(resources->mem_head)); - kfree_resource_list(&(resources->p_mem_head)); - kfree_resource_list(&(resources->bus_head)); -} - -/** - * pciehp_destroy_board_resources: put node back in the resource list - * @resources: list to put nodes back - */ -void pciehp_destroy_board_resources(struct pci_func * func) -{ - kfree_resource_list(&(func->io_head)); - kfree_resource_list(&(func->mem_head)); - kfree_resource_list(&(func->p_mem_head)); - kfree_resource_list(&(func->bus_head)); -} diff --git a/drivers/pci/hotplug/pciehprm.h b/drivers/pci/hotplug/pciehprm.h deleted file mode 100644 index 05f20fbc5f5..00000000000 --- a/drivers/pci/hotplug/pciehprm.h +++ /dev/null @@ -1,52 +0,0 @@ -/* - * PCIEHPRM : PCIEHP Resource Manager for ACPI/non-ACPI platform - * - * Copyright (C) 1995,2001 Compaq Computer Corporation - * Copyright (C) 2001,2003 Greg Kroah-Hartman (greg@kroah.com) - * Copyright (C) 2001 IBM Corp. - * Copyright (C) 2003-2004 Intel Corporation - * - * All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or (at - * your option) any later version. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or - * NON INFRINGEMENT. See the GNU General Public License for more - * details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. - * - * Send feedback to <greg@kroah.com>, <kristen.c.accardi@intel.com> - * - */ - -#ifndef _PCIEHPRM_H_ -#define _PCIEHPRM_H_ - -#ifdef CONFIG_HOTPLUG_PCI_PCIE_PHPRM_NONACPI -#include "pciehprm_nonacpi.h" -#endif - -int pciehprm_init(enum php_ctlr_type ct); -void pciehprm_cleanup(void); -int pciehprm_print_pirt(void); -int pciehprm_find_available_resources(struct controller *ctrl); -int pciehprm_set_hpp(struct controller *ctrl, struct pci_func *func, u8 card_type); -void pciehprm_enable_card(struct controller *ctrl, struct pci_func *func, u8 card_type); - -#ifdef DEBUG -#define RES_CHECK(this, bits) \ - { if (((this) & (bits - 1))) \ - printk("%s:%d ERR: potential res loss!\n", __FUNCTION__, __LINE__); } -#else -#define RES_CHECK(this, bits) -#endif - -#endif /* _PCIEHPRM_H_ */ diff --git a/drivers/pci/hotplug/pciehprm_acpi.c b/drivers/pci/hotplug/pciehprm_acpi.c index 1406db35b08..ae244e21862 100644 --- a/drivers/pci/hotplug/pciehprm_acpi.c +++ b/drivers/pci/hotplug/pciehprm_acpi.c @@ -24,100 +24,20 @@ * */ -#include <linux/config.h> #include <linux/module.h> #include <linux/kernel.h> #include <linux/types.h> #include <linux/pci.h> -#include <linux/init.h> #include <linux/acpi.h> -#include <linux/efi.h> #include <linux/pci-acpi.h> -#include <asm/uaccess.h> -#include <asm/system.h> -#ifdef CONFIG_IA64 -#include <asm/iosapic.h> -#endif -#include <acpi/acpi.h> #include <acpi/acpi_bus.h> #include <acpi/actypes.h> #include "pciehp.h" -#include "pciehprm.h" - -#define PCI_MAX_BUS 0x100 -#define ACPI_STA_DEVICE_PRESENT 0x01 #define METHOD_NAME__SUN "_SUN" #define METHOD_NAME__HPP "_HPP" #define METHOD_NAME_OSHP "OSHP" -/* Status code for running acpi method to gain native control */ -#define NC_NOT_RUN 0 -#define OSC_NOT_EXIST 1 -#define OSC_RUN_FAILED 2 -#define OSHP_NOT_EXIST 3 -#define OSHP_RUN_FAILED 4 -#define NC_RUN_SUCCESS 5 - -#define PHP_RES_BUS 0xA0 -#define PHP_RES_IO 0xA1 -#define PHP_RES_MEM 0xA2 -#define PHP_RES_PMEM 0xA3 - -#define BRIDGE_TYPE_P2P 0x00 -#define BRIDGE_TYPE_HOST 0x01 - -/* this should go to drivers/acpi/include/ */ -struct acpi__hpp { - u8 cache_line_size; - u8 latency_timer; - u8 enable_serr; - u8 enable_perr; -}; - -struct acpi_php_slot { - struct acpi_php_slot *next; - struct acpi_bridge *bridge; - acpi_handle handle; - int seg; - int bus; - int dev; - int fun; - u32 sun; - struct pci_resource *mem_head; - struct pci_resource *p_mem_head; - struct pci_resource *io_head; - struct pci_resource *bus_head; - void *slot_ops; /* _STA, _EJx, etc */ - struct slot *slot; -}; /* per func */ - -struct acpi_bridge { - struct acpi_bridge *parent; - struct acpi_bridge *next; - struct acpi_bridge *child; - acpi_handle handle; - int seg; - int pbus; /* pdev->bus->number */ - int pdevice; /* PCI_SLOT(pdev->devfn) */ - int pfunction; /* PCI_DEVFN(pdev->devfn) */ - int bus; /* pdev->subordinate->number */ - struct acpi__hpp *_hpp; - struct acpi_php_slot *slots; - struct pci_resource *tmem_head; /* total from crs */ - struct pci_resource *tp_mem_head; /* total from crs */ - struct pci_resource *tio_head; /* total from crs */ - struct pci_resource *tbus_head; /* total from crs */ - struct pci_resource *mem_head; /* available */ - struct pci_resource *p_mem_head; /* available */ - struct pci_resource *io_head; /* available */ - struct pci_resource *bus_head; /* available */ - int scanned; - int type; -}; - -static struct acpi_bridge *acpi_bridges_head; - static u8 * acpi_path_name( acpi_handle handle) { acpi_status status; @@ -133,85 +53,43 @@ static u8 * acpi_path_name( acpi_handle handle) return path_name; } -static void acpi_get__hpp ( struct acpi_bridge *ab); -static int acpi_run_oshp ( struct acpi_bridge *ab); -static int osc_run_status = NC_NOT_RUN; -static int oshp_run_status = NC_NOT_RUN; - -static int acpi_add_slot_to_php_slots( - struct acpi_bridge *ab, - int bus_num, - acpi_handle handle, - u32 adr, - u32 sun - ) -{ - struct acpi_php_slot *aps; - static long samesun = -1; - - aps = (struct acpi_php_slot *) kmalloc (sizeof(struct acpi_php_slot), GFP_KERNEL); - if (!aps) { - err ("acpi_pciehprm: alloc for aps fail\n"); - return -1; - } - memset(aps, 0, sizeof(struct acpi_php_slot)); - - aps->handle = handle; - aps->bus = bus_num; - aps->dev = (adr >> 16) & 0xffff; - aps->fun = adr & 0xffff; - aps->sun = sun; - - aps->next = ab->slots; /* cling to the bridge */ - aps->bridge = ab; - ab->slots = aps; - - ab->scanned += 1; - if (!ab->_hpp) - acpi_get__hpp(ab); - - if (osc_run_status == OSC_NOT_EXIST) - oshp_run_status = acpi_run_oshp(ab); - - if (sun != samesun) { - info("acpi_pciehprm: Slot sun(%x) at s:b:d:f=0x%02x:%02x:%02x:%02x\n", - aps->sun, ab->seg, aps->bus, aps->dev, aps->fun); - samesun = sun; - } - return 0; -} - -static void acpi_get__hpp ( struct acpi_bridge *ab) +static acpi_status +acpi_run_hpp(acpi_handle handle, struct hotplug_params *hpp) { acpi_status status; u8 nui[4]; struct acpi_buffer ret_buf = { 0, NULL}; union acpi_object *ext_obj, *package; - u8 *path_name = acpi_path_name(ab->handle); + u8 *path_name = acpi_path_name(handle); int i, len = 0; /* get _hpp */ - status = acpi_evaluate_object(ab->handle, METHOD_NAME__HPP, NULL, &ret_buf); + status = acpi_evaluate_object(handle, METHOD_NAME__HPP, NULL, &ret_buf); switch (status) { case AE_BUFFER_OVERFLOW: ret_buf.pointer = kmalloc (ret_buf.length, GFP_KERNEL); if (!ret_buf.pointer) { - err ("acpi_pciehprm:%s alloc for _HPP fail\n", path_name); - return; + err ("%s:%s alloc for _HPP fail\n", __FUNCTION__, + path_name); + return AE_NO_MEMORY; } - status = acpi_evaluate_object(ab->handle, METHOD_NAME__HPP, NULL, &ret_buf); + status = acpi_evaluate_object(handle, METHOD_NAME__HPP, + NULL, &ret_buf); if (ACPI_SUCCESS(status)) break; default: if (ACPI_FAILURE(status)) { - err("acpi_pciehprm:%s _HPP fail=0x%x\n", path_name, status); - return; + dbg("%s:%s _HPP fail=0x%x\n", __FUNCTION__, + path_name, status); + return status; } } ext_obj = (union acpi_object *) ret_buf.pointer; if (ext_obj->type != ACPI_TYPE_PACKAGE) { - err ("acpi_pciehprm:%s _HPP obj not a package\n", path_name); + err ("%s:%s _HPP obj not a package\n", __FUNCTION__, + path_name); + status = AE_ERROR; goto free_and_return; } @@ -224,1514 +102,153 @@ static void acpi_get__hpp ( struct acpi_bridge *ab) nui[i] = (u8)ext_obj->integer.value; break; default: - err ("acpi_pciehprm:%s _HPP obj type incorrect\n", path_name); + err ("%s:%s _HPP obj type incorrect\n", __FUNCTION__, + path_name); + status = AE_ERROR; goto free_and_return; } } - ab->_hpp = kmalloc (sizeof (struct acpi__hpp), GFP_KERNEL); - if (!ab->_hpp) { - err ("acpi_pciehprm:%s alloc for _HPP failed\n", path_name); - goto free_and_return; - } - memset(ab->_hpp, 0, sizeof(struct acpi__hpp)); + hpp->cache_line_size = nui[0]; + hpp->latency_timer = nui[1]; + hpp->enable_serr = nui[2]; + hpp->enable_perr = nui[3]; - ab->_hpp->cache_line_size = nui[0]; - ab->_hpp->latency_timer = nui[1]; - ab->_hpp->enable_serr = nui[2]; - ab->_hpp->enable_perr = nui[3]; - - dbg(" _HPP: cache_line_size=0x%x\n", ab->_hpp->cache_line_size); - dbg(" _HPP: latency timer =0x%x\n", ab->_hpp->latency_timer); - dbg(" _HPP: enable SERR =0x%x\n", ab->_hpp->enable_serr); - dbg(" _HPP: enable PERR =0x%x\n", ab->_hpp->enable_perr); + dbg(" _HPP: cache_line_size=0x%x\n", hpp->cache_line_size); + dbg(" _HPP: latency timer =0x%x\n", hpp->latency_timer); + dbg(" _HPP: enable SERR =0x%x\n", hpp->enable_serr); + dbg(" _HPP: enable PERR =0x%x\n", hpp->enable_perr); free_and_return: kfree(ret_buf.pointer); + return status; } -static int acpi_run_oshp ( struct acpi_bridge *ab) +static acpi_status acpi_run_oshp(acpi_handle handle) { acpi_status status; - u8 *path_name = acpi_path_name(ab->handle); + u8 *path_name = acpi_path_name(handle); /* run OSHP */ - status = acpi_evaluate_object(ab->handle, METHOD_NAME_OSHP, NULL, NULL); + status = acpi_evaluate_object(handle, METHOD_NAME_OSHP, NULL, NULL); if (ACPI_FAILURE(status)) { - err("acpi_pciehprm:%s OSHP fails=0x%x\n", path_name, status); - oshp_run_status = (status == AE_NOT_FOUND) ? OSHP_NOT_EXIST : OSHP_RUN_FAILED; + dbg("%s:%s OSHP fails=0x%x\n", __FUNCTION__, path_name, + status); } else { - oshp_run_status = NC_RUN_SUCCESS; - dbg("acpi_pciehprm:%s OSHP passes =0x%x\n", path_name, status); - dbg("acpi_pciehprm:%s oshp_run_status =0x%x\n", path_name, oshp_run_status); - } - return oshp_run_status; -} - -static acpi_status acpi_evaluate_crs( - acpi_handle handle, - struct acpi_resource **retbuf - ) -{ - acpi_status status; - struct acpi_buffer crsbuf; - u8 *path_name = acpi_path_name(handle); - - crsbuf.length = 0; - crsbuf.pointer = NULL; - - status = acpi_get_current_resources (handle, &crsbuf); - - switch (status) { - case AE_BUFFER_OVERFLOW: - break; /* found */ - case AE_NOT_FOUND: - dbg("acpi_pciehprm:%s _CRS not found\n", path_name); - return status; - default: - err ("acpi_pciehprm:%s _CRS fail=0x%x\n", path_name, status); - return status; + dbg("%s:%s OSHP passes\n", __FUNCTION__, path_name); } - - crsbuf.pointer = kmalloc (crsbuf.length, GFP_KERNEL); - if (!crsbuf.pointer) { - err ("acpi_pciehprm: alloc %ld bytes for %s _CRS fail\n", (ulong)crsbuf.length, path_name); - return AE_NO_MEMORY; - } - - status = acpi_get_current_resources (handle, &crsbuf); - if (ACPI_FAILURE(status)) { - err("acpi_pciehprm: %s _CRS fail=0x%x.\n", path_name, status); - kfree(crsbuf.pointer); - return status; - } - - *retbuf = crsbuf.pointer; - return status; } -static void free_pci_resource ( struct pci_resource *aprh) +static int is_root_bridge(acpi_handle handle) { - struct pci_resource *res, *next; + acpi_status status; + struct acpi_device_info *info; + struct acpi_buffer buffer = {ACPI_ALLOCATE_BUFFER, NULL}; + int i; - for (res = aprh; res; res = next) { - next = res->next; - kfree(res); - } -} - -static void print_pci_resource ( struct pci_resource *aprh) -{ - struct pci_resource *res; - - for (res = aprh; res; res = res->next) - dbg(" base= 0x%x length= 0x%x\n", res->base, res->length); -} - -static void print_slot_resources( struct acpi_php_slot *aps) -{ - if (aps->bus_head) { - dbg(" BUS Resources:\n"); - print_pci_resource (aps->bus_head); - } - - if (aps->io_head) { - dbg(" IO Resources:\n"); - print_pci_resource (aps->io_head); - } - - if (aps->mem_head) { - dbg(" MEM Resources:\n"); - print_pci_resource (aps->mem_head); - } - - if (aps->p_mem_head) { - dbg(" PMEM Resources:\n"); - print_pci_resource (aps->p_mem_head); - } -} - -static void print_pci_resources( struct acpi_bridge *ab) -{ - if (ab->tbus_head) { - dbg(" Total BUS Resources:\n"); - print_pci_resource (ab->tbus_head); - } - if (ab->bus_head) { - dbg(" BUS Resources:\n"); - print_pci_resource (ab->bus_head); - } - - if (ab->tio_head) { - dbg(" Total IO Resources:\n"); - print_pci_resource (ab->tio_head); - } - if (ab->io_head) { - dbg(" IO Resources:\n"); - print_pci_resource (ab->io_head); - } - - if (ab->tmem_head) { - dbg(" Total MEM Resources:\n"); - print_pci_resource (ab->tmem_head); - } - if (ab->mem_head) { - dbg(" MEM Resources:\n"); - print_pci_resource (ab->mem_head); - } - - if (ab->tp_mem_head) { - dbg(" Total PMEM Resources:\n"); - print_pci_resource (ab->tp_mem_head); - } - if (ab->p_mem_head) { - dbg(" PMEM Resources:\n"); - print_pci_resource (ab->p_mem_head); - } - if (ab->_hpp) { - dbg(" _HPP: cache_line_size=0x%x\n", ab->_hpp->cache_line_size); - dbg(" _HPP: latency timer =0x%x\n", ab->_hpp->latency_timer); - dbg(" _HPP: enable SERR =0x%x\n", ab->_hpp->enable_serr); - dbg(" _HPP: enable PERR =0x%x\n", ab->_hpp->enable_perr); - } -} - -static int pciehprm_delete_resource( - struct pci_resource **aprh, - ulong base, - ulong size) -{ - struct pci_resource *res; - struct pci_resource *prevnode; - struct pci_resource *split_node; - ulong tbase; - - pciehp_resource_sort_and_combine(aprh); - - for (res = *aprh; res; res = res->next) { - if (res->base > base) - continue; - - if ((res->base + res->length) < (base + size)) - continue; - - if (res->base < base) { - tbase = base; - - if ((res->length - (tbase - res->base)) < size) - continue; - - split_node = (struct pci_resource *) kmalloc(sizeof(struct pci_resource), GFP_KERNEL); - if (!split_node) - return -ENOMEM; - - split_node->base = res->base; - split_node->length = tbase - res->base; - res->base = tbase; - res->length -= split_node->length; - - split_node->next = res->next; - res->next = split_node; - } - - if (res->length >= size) { - split_node = (struct pci_resource*) kmalloc(sizeof(struct pci_resource), GFP_KERNEL); - if (!split_node) - return -ENOMEM; - - split_node->base = res->base + size; - split_node->length = res->length - size; - res->length = size; - - split_node->next = res->next; - res->next = split_node; - } - - if (*aprh == res) { - *aprh = res->next; - } else { - prevnode = *aprh; - while (prevnode->next != res) - prevnode = prevnode->next; - - prevnode->next = res->next; - } - res->next = NULL; - kfree(res); - break; - } - - return 0; -} - -static int pciehprm_delete_resources( - struct pci_resource **aprh, - struct pci_resource *this - ) -{ - struct pci_resource *res; - - for (res = this; res; res = res->next) - pciehprm_delete_resource(aprh, res->base, res->length); - - return 0; -} - -static int pciehprm_add_resource( - struct pci_resource **aprh, - ulong base, - ulong size) -{ - struct pci_resource *res; - - for (res = *aprh; res; res = res->next) { - if ((res->base + res->length) == base) { - res->length += size; - size = 0L; - break; + status = acpi_get_object_info(handle, &buffer); + if (ACPI_SUCCESS(status)) { + info = buffer.pointer; + if ((info->valid & ACPI_VALID_HID) && + !strcmp(PCI_ROOT_HID_STRING, + info->hardware_id.value)) { + acpi_os_free(buffer.pointer); + return 1; + } + if (info->valid & ACPI_VALID_CID) { + for (i=0; i < info->compatibility_id.count; i++) { + if (!strcmp(PCI_ROOT_HID_STRING, + info->compatibility_id.id[i].value)) { + acpi_os_free(buffer.pointer); + return 1; + } + } } - if (res->next == *aprh) - break; } - - if (size) { - res = kmalloc(sizeof(struct pci_resource), GFP_KERNEL); - if (!res) { - err ("acpi_pciehprm: alloc for res fail\n"); - return -ENOMEM; - } - memset(res, 0, sizeof (struct pci_resource)); - - res->base = base; - res->length = size; - res->next = *aprh; - *aprh = res; - } - return 0; } -static int pciehprm_add_resources( - struct pci_resource **aprh, - struct pci_resource *this - ) -{ - struct pci_resource *res; - int rc = 0; - - for (res = this; res && !rc; res = res->next) - rc = pciehprm_add_resource(aprh, res->base, res->length); - - return rc; -} - -static void acpi_parse_io ( - struct acpi_bridge *ab, - union acpi_resource_data *data - ) +int pciehp_get_hp_hw_control_from_firmware(struct pci_dev *dev) { - struct acpi_resource_io *dataio; - dataio = (struct acpi_resource_io *) data; - - dbg("Io Resource\n"); - dbg(" %d bit decode\n", ACPI_DECODE_16 == dataio->io_decode ? 16:10); - dbg(" Range minimum base: %08X\n", dataio->min_base_address); - dbg(" Range maximum base: %08X\n", dataio->max_base_address); - dbg(" Alignment: %08X\n", dataio->alignment); - dbg(" Range Length: %08X\n", dataio->range_length); -} - -static void acpi_parse_fixed_io ( - struct acpi_bridge *ab, - union acpi_resource_data *data - ) -{ - struct acpi_resource_fixed_io *datafio; - datafio = (struct acpi_resource_fixed_io *) data; - - dbg("Fixed Io Resource\n"); - dbg(" Range base address: %08X", datafio->base_address); - dbg(" Range length: %08X", datafio->range_length); -} - -static void acpi_parse_address16_32 ( - struct acpi_bridge *ab, - union acpi_resource_data *data, - acpi_resource_type id - ) -{ - /* - * acpi_resource_address16 == acpi_resource_address32 - * acpi_resource_address16 *data16 = (acpi_resource_address16 *) data; + acpi_status status; + acpi_handle chandle, handle = DEVICE_ACPI_HANDLE(&(dev->dev)); + struct pci_dev *pdev = dev; + u8 *path_name; + /* + * Per PCI firmware specification, we should run the ACPI _OSC + * method to get control of hotplug hardware before using it. + * If an _OSC is missing, we look for an OSHP to do the same thing. + * To handle different BIOS behavior, we look for _OSC and OSHP + * within the scope of the hotplug controller and its parents, upto + * the host bridge under which this controller exists. */ - struct acpi_resource_address32 *data32 = (struct acpi_resource_address32 *) data; - struct pci_resource **aprh, **tprh; - - if (id == ACPI_RSTYPE_ADDRESS16) - dbg("acpi_pciehprm:16-Bit Address Space Resource\n"); - else - dbg("acpi_pciehprm:32-Bit Address Space Resource\n"); - - switch (data32->resource_type) { - case ACPI_MEMORY_RANGE: - dbg(" Resource Type: Memory Range\n"); - aprh = &ab->mem_head; - tprh = &ab->tmem_head; - - switch (data32->attribute.memory.cache_attribute) { - case ACPI_NON_CACHEABLE_MEMORY: - dbg(" Type Specific: Noncacheable memory\n"); - break; - case ACPI_CACHABLE_MEMORY: - dbg(" Type Specific: Cacheable memory\n"); - break; - case ACPI_WRITE_COMBINING_MEMORY: - dbg(" Type Specific: Write-combining memory\n"); - break; - case ACPI_PREFETCHABLE_MEMORY: - aprh = &ab->p_mem_head; - dbg(" Type Specific: Prefetchable memory\n"); - break; - default: - dbg(" Type Specific: Invalid cache attribute\n"); + while (!handle) { + /* + * This hotplug controller was not listed in the ACPI name + * space at all. Try to get acpi handle of parent pci bus. + */ + if (!pdev || !pdev->bus->parent) break; - } - - dbg(" Type Specific: Read%s\n", ACPI_READ_WRITE_MEMORY == data32->attribute.memory.read_write_attribute ? "/Write":" Only"); - break; - - case ACPI_IO_RANGE: - dbg(" Resource Type: I/O Range\n"); - aprh = &ab->io_head; - tprh = &ab->tio_head; - - switch (data32->attribute.io.range_attribute) { - case ACPI_NON_ISA_ONLY_RANGES: - dbg(" Type Specific: Non-ISA Io Addresses\n"); - break; - case ACPI_ISA_ONLY_RANGES: - dbg(" Type Specific: ISA Io Addresses\n"); - break; - case ACPI_ENTIRE_RANGE: - dbg(" Type Specific: ISA and non-ISA Io Addresses\n"); - break; - default: - dbg(" Type Specific: Invalid range attribute\n"); + dbg("Could not find %s in acpi namespace, trying parent\n", + pci_name(pdev)); + if (!pdev->bus->parent->self) + /* Parent must be a host bridge */ + handle = acpi_get_pci_rootbridge_handle( + pci_domain_nr(pdev->bus->parent), + pdev->bus->parent->number); + else + handle = DEVICE_ACPI_HANDLE( + &(pdev->bus->parent->self->dev)); + pdev = pdev->bus->parent->self; + } + + while (handle) { + path_name = acpi_path_name(handle); + dbg("Trying to get hotplug control for %s \n", path_name); + status = pci_osc_control_set(handle, + OSC_PCI_EXPRESS_NATIVE_HP_CONTROL); + if (status == AE_NOT_FOUND) + status = acpi_run_oshp(handle); + if (ACPI_SUCCESS(status)) { + dbg("Gained control for hotplug HW for pci %s (%s)\n", + pci_name(dev), path_name); + return 0; + } + if (is_root_bridge(handle)) break; - } - break; - - case ACPI_BUS_NUMBER_RANGE: - dbg(" Resource Type: Bus Number Range(fixed)\n"); - /* fixup to be compatible with the rest of php driver */ - data32->min_address_range++; - data32->address_length--; - aprh = &ab->bus_head; - tprh = &ab->tbus_head; - break; - default: - dbg(" Resource Type: Invalid resource type. Exiting.\n"); - return; - } - - dbg(" Resource %s\n", ACPI_CONSUMER == data32->producer_consumer ? "Consumer":"Producer"); - dbg(" %s decode\n", ACPI_SUB_DECODE == data32->decode ? "Subtractive":"Positive"); - dbg(" Min address is %s fixed\n", ACPI_ADDRESS_FIXED == data32->min_address_fixed ? "":"not"); - dbg(" Max address is %s fixed\n", ACPI_ADDRESS_FIXED == data32->max_address_fixed ? "":"not"); - dbg(" Granularity: %08X\n", data32->granularity); - dbg(" Address range min: %08X\n", data32->min_address_range); - dbg(" Address range max: %08X\n", data32->max_address_range); - dbg(" Address translation offset: %08X\n", data32->address_translation_offset); - dbg(" Address Length: %08X\n", data32->address_length); - - if (0xFF != data32->resource_source.index) { - dbg(" Resource Source Index: %X\n", data32->resource_source.index); - /* dbg(" Resource Source: %s\n", data32->resource_source.string_ptr); */ - } - - pciehprm_add_resource(aprh, data32->min_address_range, data32->address_length); -} - -static acpi_status acpi_parse_crs( - struct acpi_bridge *ab, - struct acpi_resource *crsbuf - ) -{ - acpi_status status = AE_OK; - struct acpi_resource *resource = crsbuf; - u8 count = 0; - u8 done = 0; - - while (!done) { - dbg("acpi_pciehprm: PCI bus 0x%x Resource structure %x.\n", ab->bus, count++); - switch (resource->id) { - case ACPI_RSTYPE_IRQ: - dbg("Irq -------- Resource\n"); - break; - case ACPI_RSTYPE_DMA: - dbg("DMA -------- Resource\n"); - break; - case ACPI_RSTYPE_START_DPF: - dbg("Start DPF -------- Resource\n"); - break; - case ACPI_RSTYPE_END_DPF: - dbg("End DPF -------- Resource\n"); - break; - case ACPI_RSTYPE_IO: - acpi_parse_io (ab, &resource->data); - break; - case ACPI_RSTYPE_FIXED_IO: - acpi_parse_fixed_io (ab, &resource->data); - break; - case ACPI_RSTYPE_VENDOR: - dbg("Vendor -------- Resource\n"); - break; - case ACPI_RSTYPE_END_TAG: - dbg("End_tag -------- Resource\n"); - done = 1; - break; - case ACPI_RSTYPE_MEM24: - dbg("Mem24 -------- Resource\n"); - break; - case ACPI_RSTYPE_MEM32: - dbg("Mem32 -------- Resource\n"); - break; - case ACPI_RSTYPE_FIXED_MEM32: - dbg("Fixed Mem32 -------- Resource\n"); - break; - case ACPI_RSTYPE_ADDRESS16: - acpi_parse_address16_32(ab, &resource->data, ACPI_RSTYPE_ADDRESS16); - break; - case ACPI_RSTYPE_ADDRESS32: - acpi_parse_address16_32(ab, &resource->data, ACPI_RSTYPE_ADDRESS32); - break; - case ACPI_RSTYPE_ADDRESS64: - info("Address64 -------- Resource unparsed\n"); - break; - case ACPI_RSTYPE_EXT_IRQ: - dbg("Ext Irq -------- Resource\n"); - break; - default: - dbg("Invalid -------- resource type 0x%x\n", resource->id); - break; - } - - resource = (struct acpi_resource *) ((char *)resource + resource->length); - } - - return status; -} - -static acpi_status acpi_get_crs( struct acpi_bridge *ab) -{ - acpi_status status; - struct acpi_resource *crsbuf; - - status = acpi_evaluate_crs(ab->handle, &crsbuf); - if (ACPI_SUCCESS(status)) { - status = acpi_parse_crs(ab, crsbuf); - kfree(crsbuf); - - pciehp_resource_sort_and_combine(&ab->bus_head); - pciehp_resource_sort_and_combine(&ab->io_head); - pciehp_resource_sort_and_combine(&ab->mem_head); - pciehp_resource_sort_and_combine(&ab->p_mem_head); - - pciehprm_add_resources (&ab->tbus_head, ab->bus_head); - pciehprm_add_resources (&ab->tio_head, ab->io_head); - pciehprm_add_resources (&ab->tmem_head, ab->mem_head); - pciehprm_add_resources (&ab->tp_mem_head, ab->p_mem_head); - } - - return status; -} - -/* find acpi_bridge downword from ab. */ -static struct acpi_bridge * -find_acpi_bridge_by_bus( - struct acpi_bridge *ab, - int seg, - int bus /* pdev->subordinate->number */ - ) -{ - struct acpi_bridge *lab = NULL; - - if (!ab) - return NULL; - - if ((ab->bus == bus) && (ab->seg == seg)) - return ab; - - if (ab->child) - lab = find_acpi_bridge_by_bus(ab->child, seg, bus); - - if (!lab) - if (ab->next) - lab = find_acpi_bridge_by_bus(ab->next, seg, bus); - - return lab; -} - -/* - * Build a device tree of ACPI PCI Bridges - */ -static void pciehprm_acpi_register_a_bridge ( - struct acpi_bridge **head, - struct acpi_bridge *pab, /* parent bridge to which child bridge is added */ - struct acpi_bridge *cab /* child bridge to add */ - ) -{ - struct acpi_bridge *lpab; - struct acpi_bridge *lcab; - - lpab = find_acpi_bridge_by_bus(*head, pab->seg, pab->bus); - if (!lpab) { - if (!(pab->type & BRIDGE_TYPE_HOST)) - warn("PCI parent bridge s:b(%x:%x) not in list.\n", pab->seg, pab->bus); - pab->next = *head; - *head = pab; - lpab = pab; - } - - if ((cab->type & BRIDGE_TYPE_HOST) && (pab == cab)) - return; - - lcab = find_acpi_bridge_by_bus(*head, cab->seg, cab->bus); - if (lcab) { - if ((pab->bus != lcab->parent->bus) || (lcab->bus != cab->bus)) - err("PCI child bridge s:b(%x:%x) in list with diff parent.\n", cab->seg, cab->bus); - return; - } else - lcab = cab; - - lcab->parent = lpab; - lcab->next = lpab->child; - lpab->child = lcab; -} - -static acpi_status pciehprm_acpi_build_php_slots_callback( - acpi_handle handle, - u32 Level, - void *context, - void **retval - ) -{ - ulong bus_num; - ulong seg_num; - ulong sun, adr; - ulong padr = 0; - acpi_handle phandle = NULL; - struct acpi_bridge *pab = (struct acpi_bridge *)context; - struct acpi_bridge *lab; - acpi_status status; - u8 *path_name = acpi_path_name(handle); - - /* get _SUN */ - status = acpi_evaluate_integer(handle, METHOD_NAME__SUN, NULL, &sun); - switch(status) { - case AE_NOT_FOUND: - return AE_OK; - default: - if (ACPI_FAILURE(status)) { - err("acpi_pciehprm:%s _SUN fail=0x%x\n", path_name, status); - return status; - } - } - - /* get _ADR. _ADR must exist if _SUN exists */ - status = acpi_evaluate_integer(handle, METHOD_NAME__ADR, NULL, &adr); - if (ACPI_FAILURE(status)) { - err("acpi_pciehprm:%s _ADR fail=0x%x\n", path_name, status); - return status; - } - - dbg("acpi_pciehprm:%s sun=0x%08x adr=0x%08x\n", path_name, (u32)sun, (u32)adr); - - status = acpi_get_parent(handle, &phandle); - if (ACPI_FAILURE(status)) { - err("acpi_pciehprm:%s get_parent fail=0x%x\n", path_name, status); - return (status); - } - - bus_num = pab->bus; - seg_num = pab->seg; - - if (pab->bus == bus_num) { - lab = pab; - } else { - dbg("WARN: pab is not parent\n"); - lab = find_acpi_bridge_by_bus(pab, seg_num, bus_num); - if (!lab) { - dbg("acpi_pciehprm: alloc new P2P bridge(%x) for sun(%08x)\n", (u32)bus_num, (u32)sun); - lab = (struct acpi_bridge *)kmalloc(sizeof(struct acpi_bridge), GFP_KERNEL); - if (!lab) { - err("acpi_pciehprm: alloc for ab fail\n"); - return AE_NO_MEMORY; - } - memset(lab, 0, sizeof(struct acpi_bridge)); - - lab->handle = phandle; - lab->pbus = pab->bus; - lab->pdevice = (int)(padr >> 16) & 0xffff; - lab->pfunction = (int)(padr & 0xffff); - lab->bus = (int)bus_num; - lab->scanned = 0; - lab->type = BRIDGE_TYPE_P2P; - - pciehprm_acpi_register_a_bridge (&acpi_bridges_head, pab, lab); - } else - dbg("acpi_pciehprm: found P2P bridge(%x) for sun(%08x)\n", (u32)bus_num, (u32)sun); - } - - acpi_add_slot_to_php_slots(lab, (int)bus_num, handle, (u32)adr, (u32)sun); - - return (status); -} - -static int pciehprm_acpi_build_php_slots( - struct acpi_bridge *ab, - u32 depth - ) -{ - acpi_status status; - u8 *path_name = acpi_path_name(ab->handle); - - /* Walk down this pci bridge to get _SUNs if any behind P2P */ - status = acpi_walk_namespace ( ACPI_TYPE_DEVICE, - ab->handle, - depth, - pciehprm_acpi_build_php_slots_callback, - ab, - NULL ); - if (ACPI_FAILURE(status)) { - dbg("acpi_pciehprm:%s walk for _SUN on pci bridge seg:bus(%x:%x) fail=0x%x\n", path_name, ab->seg, ab->bus, status); - return -1; - } - - return 0; -} - -static void build_a_bridge( - struct acpi_bridge *pab, - struct acpi_bridge *ab - ) -{ - u8 *path_name = acpi_path_name(ab->handle); - - pciehprm_acpi_register_a_bridge (&acpi_bridges_head, pab, ab); - - switch (ab->type) { - case BRIDGE_TYPE_HOST: - dbg("acpi_pciehprm: Registered PCI HOST Bridge(%02x) on s:b:d:f(%02x:%02x:%02x:%02x) [%s]\n", - ab->bus, ab->seg, ab->pbus, ab->pdevice, ab->pfunction, path_name); - break; - case BRIDGE_TYPE_P2P: - dbg("acpi_pciehprm: Registered PCI P2P Bridge(%02x-%02x) on s:b:d:f(%02x:%02x:%02x:%02x) [%s]\n", - ab->pbus, ab->bus, ab->seg, ab->pbus, ab->pdevice, ab->pfunction, path_name); - break; - }; - - /* build any immediate PHP slots under this pci bridge */ - pciehprm_acpi_build_php_slots(ab, 1); -} - -static struct acpi_bridge * add_p2p_bridge( - acpi_handle handle, - struct acpi_bridge *pab, /* parent */ - ulong adr - ) -{ - struct acpi_bridge *ab; - struct pci_dev *pdev; - ulong devnum, funcnum; - u8 *path_name = acpi_path_name(handle); - - ab = (struct acpi_bridge *) kmalloc (sizeof(struct acpi_bridge), GFP_KERNEL); - if (!ab) { - err("acpi_pciehprm: alloc for ab fail\n"); - return NULL; - } - memset(ab, 0, sizeof(struct acpi_bridge)); - - devnum = (adr >> 16) & 0xffff; - funcnum = adr & 0xffff; - - pdev = pci_find_slot(pab->bus, PCI_DEVFN(devnum, funcnum)); - if (!pdev || !pdev->subordinate) { - err("acpi_pciehprm:%s is not a P2P Bridge\n", path_name); - kfree(ab); - return NULL; - } - - ab->handle = handle; - ab->seg = pab->seg; - ab->pbus = pab->bus; /* or pdev->bus->number */ - ab->pdevice = devnum; /* or PCI_SLOT(pdev->devfn) */ - ab->pfunction = funcnum; /* or PCI_FUNC(pdev->devfn) */ - ab->bus = pdev->subordinate->number; - ab->scanned = 0; - ab->type = BRIDGE_TYPE_P2P; - - dbg("acpi_pciehprm: P2P(%x-%x) on pci=b:d:f(%x:%x:%x) acpi=b:d:f(%x:%x:%x) [%s]\n", - pab->bus, ab->bus, pdev->bus->number, PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn), - pab->bus, (u32)devnum, (u32)funcnum, path_name); - - build_a_bridge(pab, ab); - - return ab; -} - -static acpi_status scan_p2p_bridge( - acpi_handle handle, - u32 Level, - void *context, - void **retval - ) -{ - struct acpi_bridge *pab = (struct acpi_bridge *)context; - struct acpi_bridge *ab; - acpi_status status; - ulong adr = 0; - u8 *path_name = acpi_path_name(handle); - ulong devnum, funcnum; - struct pci_dev *pdev; - - /* get device, function */ - status = acpi_evaluate_integer(handle, METHOD_NAME__ADR, NULL, &adr); - if (ACPI_FAILURE(status)) { - if (status != AE_NOT_FOUND) - err("acpi_pciehprm:%s _ADR fail=0x%x\n", path_name, status); - return AE_OK; - } - - devnum = (adr >> 16) & 0xffff; - funcnum = adr & 0xffff; - - pdev = pci_find_slot(pab->bus, PCI_DEVFN(devnum, funcnum)); - if (!pdev) - return AE_OK; - if (!pdev->subordinate) - return AE_OK; - - ab = add_p2p_bridge(handle, pab, adr); - if (ab) { - status = acpi_walk_namespace ( ACPI_TYPE_DEVICE, - handle, - (u32)1, - scan_p2p_bridge, - ab, - NULL); + chandle = handle; + status = acpi_get_parent(chandle, &handle); if (ACPI_FAILURE(status)) - dbg("acpi_pciehprm:%s find_p2p fail=0x%x\n", path_name, status); - } - - return AE_OK; -} - -static struct acpi_bridge * add_host_bridge( - acpi_handle handle, - ulong segnum, - ulong busnum - ) -{ - ulong adr = 0; - acpi_status status; - struct acpi_bridge *ab; - u8 *path_name = acpi_path_name(handle); - - /* get device, function: host br adr is always 0000 though. */ - status = acpi_evaluate_integer(handle, METHOD_NAME__ADR, NULL, &adr); - if (ACPI_FAILURE(status)) { - err("acpi_pciehprm:%s _ADR fail=0x%x\n", path_name, status); - return NULL; - } - dbg("acpi_pciehprm: ROOT PCI seg(0x%x)bus(0x%x)dev(0x%x)func(0x%x) [%s]\n", (u32)segnum, - (u32)busnum, (u32)(adr >> 16) & 0xffff, (u32)adr & 0xffff, path_name); - - ab = (struct acpi_bridge *) kmalloc (sizeof(struct acpi_bridge), GFP_KERNEL); - if (!ab) { - err("acpi_pciehprm: alloc for ab fail\n"); - return NULL; - } - memset(ab, 0, sizeof(struct acpi_bridge)); - - ab->handle = handle; - ab->seg = (int)segnum; - ab->bus = ab->pbus = (int)busnum; - ab->pdevice = (int)(adr >> 16) & 0xffff; - ab->pfunction = (int)(adr & 0xffff); - ab->scanned = 0; - ab->type = BRIDGE_TYPE_HOST; - - /* get root pci bridge's current resources */ - status = acpi_get_crs(ab); - if (ACPI_FAILURE(status)) { - err("acpi_pciehprm:%s evaluate _CRS fail=0x%x\n", path_name, status); - kfree(ab); - return NULL; - } - - status = pci_osc_control_set (OSC_PCI_EXPRESS_NATIVE_HP_CONTROL); - if (ACPI_FAILURE(status)) { - err("%s: status %x\n", __FUNCTION__, status); - osc_run_status = (status == AE_NOT_FOUND) ? OSC_NOT_EXIST : OSC_RUN_FAILED; - } else { - osc_run_status = NC_RUN_SUCCESS; - } - dbg("%s: osc_run_status %x\n", __FUNCTION__, osc_run_status); - - build_a_bridge(ab, ab); - - return ab; -} - -static acpi_status acpi_scan_from_root_pci_callback ( - acpi_handle handle, - u32 Level, - void *context, - void **retval - ) -{ - ulong segnum = 0; - ulong busnum = 0; - acpi_status status; - struct acpi_bridge *ab; - u8 *path_name = acpi_path_name(handle); - - /* get bus number of this pci root bridge */ - status = acpi_evaluate_integer(handle, METHOD_NAME__SEG, NULL, &segnum); - if (ACPI_FAILURE(status)) { - if (status != AE_NOT_FOUND) { - err("acpi_pciehprm:%s evaluate _SEG fail=0x%x\n", path_name, status); - return status; - } - segnum = 0; - } - - /* get bus number of this pci root bridge */ - status = acpi_evaluate_integer(handle, METHOD_NAME__BBN, NULL, &busnum); - if (ACPI_FAILURE(status)) { - err("acpi_pciehprm:%s evaluate _BBN fail=0x%x\n", path_name, status); - return (status); - } - - ab = add_host_bridge(handle, segnum, busnum); - if (ab) { - status = acpi_walk_namespace ( ACPI_TYPE_DEVICE, - handle, - 1, - scan_p2p_bridge, - ab, - NULL); - if (ACPI_FAILURE(status)) - dbg("acpi_pciehprm:%s find_p2p fail=0x%x\n", path_name, status); + break; } - return AE_OK; + err("Cannot get control of hotplug hardware for pci %s\n", + pci_name(dev)); + return -1; } -static int pciehprm_acpi_scan_pci (void) +void pciehp_get_hp_params_from_firmware(struct pci_dev *dev, + struct hotplug_params *hpp) { - acpi_status status; + acpi_status status = AE_NOT_FOUND; + struct pci_dev *pdev = dev; /* - * TBD: traverse LDM device tree with the help of - * unified ACPI augmented for php device population. + * _HPP settings apply to all child buses, until another _HPP is + * encountered. If we don't find an _HPP for the input pci dev, + * look for it in the parent device scope since that would apply to + * this pci dev. If we don't find any _HPP, use hardcoded defaults */ - status = acpi_get_devices ( PCI_ROOT_HID_STRING, - acpi_scan_from_root_pci_callback, - NULL, - NULL ); - if (ACPI_FAILURE(status)) { - err("acpi_pciehprm:get_device PCI ROOT HID fail=0x%x\n", status); - return -1; - } - - return 0; -} - -int pciehprm_init(enum php_ctlr_type ctlr_type) -{ - int rc; - - if (ctlr_type != PCI) - return -ENODEV; - - dbg("pciehprm ACPI init <enter>\n"); - acpi_bridges_head = NULL; - - /* construct PCI bus:device tree of acpi_handles */ - rc = pciehprm_acpi_scan_pci(); - if (rc) - return rc; - - if ((oshp_run_status != NC_RUN_SUCCESS) && (osc_run_status != NC_RUN_SUCCESS)) { - err("Fails to gain control of native hot-plug\n"); - rc = -ENODEV; - } - - dbg("pciehprm ACPI init %s\n", (rc)?"fail":"success"); - return rc; -} - -static void free_a_slot(struct acpi_php_slot *aps) -{ - dbg(" free a php func of slot(0x%02x) on PCI b:d:f=0x%02x:%02x:%02x\n", aps->sun, aps->bus, aps->dev, aps->fun); - - free_pci_resource (aps->io_head); - free_pci_resource (aps->bus_head); - free_pci_resource (aps->mem_head); - free_pci_resource (aps->p_mem_head); - - kfree(aps); -} - -static void free_a_bridge( struct acpi_bridge *ab) -{ - struct acpi_php_slot *aps, *next; - - switch (ab->type) { - case BRIDGE_TYPE_HOST: - dbg("Free ACPI PCI HOST Bridge(%x) [%s] on s:b:d:f(%x:%x:%x:%x)\n", - ab->bus, acpi_path_name(ab->handle), ab->seg, ab->pbus, ab->pdevice, ab->pfunction); - break; - case BRIDGE_TYPE_P2P: - dbg("Free ACPI PCI P2P Bridge(%x-%x) [%s] on s:b:d:f(%x:%x:%x:%x)\n", - ab->pbus, ab->bus, acpi_path_name(ab->handle), ab->seg, ab->pbus, ab->pdevice, ab->pfunction); - break; - }; - - /* free slots first */ - for (aps = ab->slots; aps; aps = next) { - next = aps->next; - free_a_slot(aps); - } - - free_pci_resource (ab->io_head); - free_pci_resource (ab->tio_head); - free_pci_resource (ab->bus_head); - free_pci_resource (ab->tbus_head); - free_pci_resource (ab->mem_head); - free_pci_resource (ab->tmem_head); - free_pci_resource (ab->p_mem_head); - free_pci_resource (ab->tp_mem_head); - - kfree(ab); -} - -static void pciehprm_free_bridges ( struct acpi_bridge *ab) -{ - if (!ab) - return; - - if (ab->child) - pciehprm_free_bridges (ab->child); - - if (ab->next) - pciehprm_free_bridges (ab->next); - - free_a_bridge(ab); -} - -void pciehprm_cleanup(void) -{ - pciehprm_free_bridges (acpi_bridges_head); -} - -static int get_number_of_slots ( - struct acpi_bridge *ab, - int selfonly - ) -{ - struct acpi_php_slot *aps; - int prev_slot = -1; - int slot_num = 0; - - for ( aps = ab->slots; aps; aps = aps->next) - if (aps->dev != prev_slot) { - prev_slot = aps->dev; - slot_num++; - } - - if (ab->child) - slot_num += get_number_of_slots (ab->child, 0); - - if (selfonly) - return slot_num; - - if (ab->next) - slot_num += get_number_of_slots (ab->next, 0); - - return slot_num; -} - -static int print_acpi_resources (struct acpi_bridge *ab) -{ - struct acpi_php_slot *aps; - int i; - - switch (ab->type) { - case BRIDGE_TYPE_HOST: - dbg("PCI HOST Bridge (%x) [%s]\n", ab->bus, acpi_path_name(ab->handle)); - break; - case BRIDGE_TYPE_P2P: - dbg("PCI P2P Bridge (%x-%x) [%s]\n", ab->pbus, ab->bus, acpi_path_name(ab->handle)); - break; - }; - - print_pci_resources (ab); - - for ( i = -1, aps = ab->slots; aps; aps = aps->next) { - if (aps->dev == i) - continue; - dbg(" Slot sun(%x) s:b:d:f(%02x:%02x:%02x:%02x)\n", aps->sun, aps->seg, aps->bus, aps->dev, aps->fun); - print_slot_resources(aps); - i = aps->dev; - } - - if (ab->child) - print_acpi_resources (ab->child); - - if (ab->next) - print_acpi_resources (ab->next); - - return 0; -} - -int pciehprm_print_pirt(void) -{ - dbg("PCIEHPRM ACPI Slots\n"); - if (acpi_bridges_head) - print_acpi_resources (acpi_bridges_head); - - return 0; -} - -static struct acpi_php_slot * get_acpi_slot ( - struct acpi_bridge *ab, - u32 sun - ) -{ - struct acpi_php_slot *aps = NULL; - - for ( aps = ab->slots; aps; aps = aps->next) - if (aps->sun == sun) - return aps; - - if (!aps && ab->child) { - aps = (struct acpi_php_slot *)get_acpi_slot (ab->child, sun); - if (aps) - return aps; - } - - if (!aps && ab->next) { - aps = (struct acpi_php_slot *)get_acpi_slot (ab->next, sun); - if (aps) - return aps; - } - - return aps; - -} - -#if 0 -void * pciehprm_get_slot(struct slot *slot) -{ - struct acpi_bridge *ab = acpi_bridges_head; - struct acpi_php_slot *aps = get_acpi_slot (ab, slot->number); - - aps->slot = slot; - - dbg("Got acpi slot sun(%x): s:b:d:f(%x:%x:%x:%x)\n", aps->sun, aps->seg, aps->bus, aps->dev, aps->fun); - - return (void *)aps; -} -#endif - -static void pciehprm_dump_func_res( struct pci_func *fun) -{ - struct pci_func *func = fun; - - if (func->bus_head) { - dbg(": BUS Resources:\n"); - print_pci_resource (func->bus_head); - } - if (func->io_head) { - dbg(": IO Resources:\n"); - print_pci_resource (func->io_head); - } - if (func->mem_head) { - dbg(": MEM Resources:\n"); - print_pci_resource (func->mem_head); - } - if (func->p_mem_head) { - dbg(": PMEM Resources:\n"); - print_pci_resource (func->p_mem_head); - } -} - -static void pciehprm_dump_ctrl_res( struct controller *ctlr) -{ - struct controller *ctrl = ctlr; - - if (ctrl->bus_head) { - dbg(": BUS Resources:\n"); - print_pci_resource (ctrl->bus_head); - } - if (ctrl->io_head) { - dbg(": IO Resources:\n"); - print_pci_resource (ctrl->io_head); - } - if (ctrl->mem_head) { - dbg(": MEM Resources:\n"); - print_pci_resource (ctrl->mem_head); - } - if (ctrl->p_mem_head) { - dbg(": PMEM Resources:\n"); - print_pci_resource (ctrl->p_mem_head); - } -} - -static int pciehprm_get_used_resources ( - struct controller *ctrl, - struct pci_func *func - ) -{ - return pciehp_save_used_resources (ctrl, func, !DISABLE_CARD); -} - -static int configure_existing_function( - struct controller *ctrl, - struct pci_func *func - ) -{ - int rc; - - /* see how much resources the func has used. */ - rc = pciehprm_get_used_resources (ctrl, func); - - if (!rc) { - /* subtract the resources used by the func from ctrl resources */ - rc = pciehprm_delete_resources (&ctrl->bus_head, func->bus_head); - rc |= pciehprm_delete_resources (&ctrl->io_head, func->io_head); - rc |= pciehprm_delete_resources (&ctrl->mem_head, func->mem_head); - rc |= pciehprm_delete_resources (&ctrl->p_mem_head, func->p_mem_head); - if (rc) - warn("aCEF: cannot del used resources\n"); - } else - err("aCEF: cannot get used resources\n"); - - return rc; -} - -static int bind_pci_resources_to_slots ( struct controller *ctrl) -{ - struct pci_func *func, new_func; - int busn = ctrl->slot_bus; - int devn, funn; - u32 vid; - - for (devn = 0; devn < 32; devn++) { - for (funn = 0; funn < 8; funn++) { - /* - if (devn == ctrl->device && funn == ctrl->function) - continue; - */ - /* find out if this entry is for an occupied slot */ - vid = 0xFFFFFFFF; - pci_bus_read_config_dword(ctrl->pci_dev->subordinate, PCI_DEVFN(devn, funn), PCI_VENDOR_ID, &vid); - - if (vid != 0xFFFFFFFF) { - dbg("%s: vid = %x\n", __FUNCTION__, vid); - func = pciehp_slot_find(busn, devn, funn); - if (!func) { - memset(&new_func, 0, sizeof(struct pci_func)); - new_func.bus = busn; - new_func.device = devn; - new_func.function = funn; - new_func.is_a_board = 1; - configure_existing_function(ctrl, &new_func); - pciehprm_dump_func_res(&new_func); - } else { - configure_existing_function(ctrl, func); - pciehprm_dump_func_res(func); - } - dbg("aCCF:existing PCI 0x%x Func ResourceDump\n", ctrl->bus); - } - } - } - - return 0; -} - -static int bind_pci_resources( - struct controller *ctrl, - struct acpi_bridge *ab - ) -{ - int status = 0; - - if (ab->bus_head) { - dbg("bapr: BUS Resources add on PCI 0x%x\n", ab->bus); - status = pciehprm_add_resources (&ctrl->bus_head, ab->bus_head); - if (pciehprm_delete_resources (&ab->bus_head, ctrl->bus_head)) - warn("bapr: cannot sub BUS Resource on PCI 0x%x\n", ab->bus); - if (status) { - err("bapr: BUS Resource add on PCI 0x%x: fail=0x%x\n", ab->bus, status); - return status; - } - } else - info("bapr: No BUS Resource on PCI 0x%x.\n", ab->bus); - - if (ab->io_head) { - dbg("bapr: IO Resources add on PCI 0x%x\n", ab->bus); - status = pciehprm_add_resources (&ctrl->io_head, ab->io_head); - if (pciehprm_delete_resources (&ab->io_head, ctrl->io_head)) - warn("bapr: cannot sub IO Resource on PCI 0x%x\n", ab->bus); - if (status) { - err("bapr: IO Resource add on PCI 0x%x: fail=0x%x\n", ab->bus, status); - return status; - } - } else - info("bapr: No IO Resource on PCI 0x%x.\n", ab->bus); - - if (ab->mem_head) { - dbg("bapr: MEM Resources add on PCI 0x%x\n", ab->bus); - status = pciehprm_add_resources (&ctrl->mem_head, ab->mem_head); - if (pciehprm_delete_resources (&ab->mem_head, ctrl->mem_head)) - warn("bapr: cannot sub MEM Resource on PCI 0x%x\n", ab->bus); - if (status) { - err("bapr: MEM Resource add on PCI 0x%x: fail=0x%x\n", ab->bus, status); - return status; - } - } else - info("bapr: No MEM Resource on PCI 0x%x.\n", ab->bus); - - if (ab->p_mem_head) { - dbg("bapr: PMEM Resources add on PCI 0x%x\n", ab->bus); - status = pciehprm_add_resources (&ctrl->p_mem_head, ab->p_mem_head); - if (pciehprm_delete_resources (&ab->p_mem_head, ctrl->p_mem_head)) - warn("bapr: cannot sub PMEM Resource on PCI 0x%x\n", ab->bus); - if (status) { - err("bapr: PMEM Resource add on PCI 0x%x: fail=0x%x\n", ab->bus, status); - return status; - } - } else - info("bapr: No PMEM Resource on PCI 0x%x.\n", ab->bus); - - return status; -} - -static int no_pci_resources( struct acpi_bridge *ab) -{ - return !(ab->p_mem_head || ab->mem_head || ab->io_head || ab->bus_head); -} - -static int find_pci_bridge_resources ( - struct controller *ctrl, - struct acpi_bridge *ab - ) -{ - int rc = 0; - struct pci_func func; - - memset(&func, 0, sizeof(struct pci_func)); - - func.bus = ab->pbus; - func.device = ab->pdevice; - func.function = ab->pfunction; - func.is_a_board = 1; - - /* Get used resources for this PCI bridge */ - rc = pciehp_save_used_resources (ctrl, &func, !DISABLE_CARD); - - ab->io_head = func.io_head; - ab->mem_head = func.mem_head; - ab->p_mem_head = func.p_mem_head; - ab->bus_head = func.bus_head; - if (ab->bus_head) - pciehprm_delete_resource(&ab->bus_head, ctrl->pci_dev->subordinate->number, 1); - - return rc; -} - -static int get_pci_resources_from_bridge( - struct controller *ctrl, - struct acpi_bridge *ab - ) -{ - int rc = 0; - - dbg("grfb: Get Resources for PCI 0x%x from actual PCI bridge 0x%x.\n", ctrl->bus, ab->bus); - - rc = find_pci_bridge_resources (ctrl, ab); - - pciehp_resource_sort_and_combine(&ab->bus_head); - pciehp_resource_sort_and_combine(&ab->io_head); - pciehp_resource_sort_and_combine(&ab->mem_head); - pciehp_resource_sort_and_combine(&ab->p_mem_head); - - pciehprm_add_resources (&ab->tbus_head, ab->bus_head); - pciehprm_add_resources (&ab->tio_head, ab->io_head); - pciehprm_add_resources (&ab->tmem_head, ab->mem_head); - pciehprm_add_resources (&ab->tp_mem_head, ab->p_mem_head); - - return rc; -} - -static int get_pci_resources( - struct controller *ctrl, - struct acpi_bridge *ab - ) -{ - int rc = 0; - - if (no_pci_resources(ab)) { - dbg("spbr:PCI 0x%x has no resources. Get parent resources.\n", ab->bus); - rc = get_pci_resources_from_bridge(ctrl, ab); - } - - return rc; -} - -/* - * Get resources for this ctrl. - * 1. get total resources from ACPI _CRS or bridge (this ctrl) - * 2. find used resources of existing adapters - * 3. subtract used resources from total resources - */ -int pciehprm_find_available_resources( struct controller *ctrl) -{ - int rc = 0; - struct acpi_bridge *ab; - - ab = find_acpi_bridge_by_bus(acpi_bridges_head, ctrl->seg, ctrl->pci_dev->subordinate->number); - if (!ab) { - err("pfar:cannot locate acpi bridge of PCI 0x%x.\n", ctrl->pci_dev->subordinate->number); - return -1; - } - if (no_pci_resources(ab)) { - rc = get_pci_resources(ctrl, ab); - if (rc) { - err("pfar:cannot get pci resources of PCI 0x%x.\n", ctrl->pci_dev->subordinate->number); - return -1; - } - } - - rc = bind_pci_resources(ctrl, ab); - dbg("pfar:pre-Bind PCI 0x%x Ctrl Resource Dump\n", ctrl->pci_dev->subordinate->number); - pciehprm_dump_ctrl_res(ctrl); - - bind_pci_resources_to_slots (ctrl); - - dbg("pfar:post-Bind PCI 0x%x Ctrl Resource Dump\n", ctrl->pci_dev->subordinate->number); - pciehprm_dump_ctrl_res(ctrl); - - return rc; -} - -int pciehprm_set_hpp( - struct controller *ctrl, - struct pci_func *func, - u8 card_type - ) -{ - struct acpi_bridge *ab; - struct pci_bus lpci_bus, *pci_bus; - int rc = 0; - unsigned int devfn; - u8 cls= 0x08; /* default cache line size */ - u8 lt = 0x40; /* default latency timer */ - u8 ep = 0; - u8 es = 0; - - memcpy(&lpci_bus, ctrl->pci_bus, sizeof(lpci_bus)); - pci_bus = &lpci_bus; - pci_bus->number = func->bus; - devfn = PCI_DEVFN(func->device, func->function); - - ab = find_acpi_bridge_by_bus(acpi_bridges_head, ctrl->seg, ctrl->bus); - - if (ab) { - if (ab->_hpp) { - lt = (u8)ab->_hpp->latency_timer; - cls = (u8)ab->_hpp->cache_line_size; - ep = (u8)ab->_hpp->enable_perr; - es = (u8)ab->_hpp->enable_serr; - } else - dbg("_hpp: no _hpp for B/D/F=%#x/%#x/%#x. use default value\n", func->bus, func->device, func->function); - } else - dbg("_hpp: no acpi bridge for B/D/F = %#x/%#x/%#x. use default value\n", func->bus, func->device, func->function); - - - if (card_type == PCI_HEADER_TYPE_BRIDGE) { - /* set subordinate Latency Timer */ - rc |= pci_bus_write_config_byte(pci_bus, devfn, PCI_SEC_LATENCY_TIMER, lt); + while (pdev && (ACPI_FAILURE(status))) { + acpi_handle handle = DEVICE_ACPI_HANDLE(&(pdev->dev)); + if (!handle) + break; + status = acpi_run_hpp(handle, hpp); + if (!(pdev->bus->parent)) + break; + /* Check if a parent object supports _HPP */ + pdev = pdev->bus->parent->self; } - - /* set base Latency Timer */ - rc |= pci_bus_write_config_byte(pci_bus, devfn, PCI_LATENCY_TIMER, lt); - dbg(" set latency timer =0x%02x: %x\n", lt, rc); - - rc |= pci_bus_write_config_byte(pci_bus, devfn, PCI_CACHE_LINE_SIZE, cls); - dbg(" set cache_line_size=0x%02x: %x\n", cls, rc); - - return rc; } -void pciehprm_enable_card( - struct controller *ctrl, - struct pci_func *func, - u8 card_type) -{ - u16 command, cmd, bcommand, bcmd; - struct pci_bus lpci_bus, *pci_bus; - struct acpi_bridge *ab; - unsigned int devfn; - int rc; - - memcpy(&lpci_bus, ctrl->pci_bus, sizeof(lpci_bus)); - pci_bus = &lpci_bus; - pci_bus->number = func->bus; - devfn = PCI_DEVFN(func->device, func->function); - - rc = pci_bus_read_config_word(pci_bus, devfn, PCI_COMMAND, &cmd); - - if (card_type == PCI_HEADER_TYPE_BRIDGE) { - rc = pci_bus_read_config_word(pci_bus, devfn, PCI_BRIDGE_CONTROL, &bcmd); - } - - command = cmd | PCI_COMMAND_MASTER | PCI_COMMAND_INVALIDATE - | PCI_COMMAND_IO | PCI_COMMAND_MEMORY; - bcommand = bcmd | PCI_BRIDGE_CTL_NO_ISA; - - ab = find_acpi_bridge_by_bus(acpi_bridges_head, ctrl->seg, ctrl->bus); - if (ab) { - if (ab->_hpp) { - if (ab->_hpp->enable_perr) { - command |= PCI_COMMAND_PARITY; - bcommand |= PCI_BRIDGE_CTL_PARITY; - } else { - command &= ~PCI_COMMAND_PARITY; - bcommand &= ~PCI_BRIDGE_CTL_PARITY; - } - if (ab->_hpp->enable_serr) { - command |= PCI_COMMAND_SERR; - bcommand |= PCI_BRIDGE_CTL_SERR; - } else { - command &= ~PCI_COMMAND_SERR; - bcommand &= ~PCI_BRIDGE_CTL_SERR; - } - } else - dbg("no _hpp for B/D/F = %#x/%#x/%#x.\n", func->bus, func->device, func->function); - } else - dbg("no acpi bridge for B/D/F = %#x/%#x/%#x.\n", func->bus, func->device, func->function); - - if (command != cmd) { - rc = pci_bus_write_config_word(pci_bus, devfn, PCI_COMMAND, command); - } - if ((card_type == PCI_HEADER_TYPE_BRIDGE) && (bcommand != bcmd)) { - rc = pci_bus_write_config_word(pci_bus, devfn, PCI_BRIDGE_CONTROL, bcommand); - } -} diff --git a/drivers/pci/hotplug/pciehprm_nonacpi.c b/drivers/pci/hotplug/pciehprm_nonacpi.c index 76c727c74cc..29180dfe849 100644 --- a/drivers/pci/hotplug/pciehprm_nonacpi.c +++ b/drivers/pci/hotplug/pciehprm_nonacpi.c @@ -27,479 +27,21 @@ * */ -#include <linux/config.h> #include <linux/module.h> #include <linux/kernel.h> #include <linux/types.h> #include <linux/sched.h> #include <linux/pci.h> -#include <linux/init.h> #include <linux/slab.h> - -#include <asm/uaccess.h> -#ifdef CONFIG_IA64 -#include <asm/iosapic.h> -#endif - #include "pciehp.h" -#include "pciehprm.h" -#include "pciehprm_nonacpi.h" - -void pciehprm_cleanup(void) +void pciehp_get_hp_params_from_firmware(struct pci_dev *dev, + struct hotplug_params *hpp) { return; } -int pciehprm_print_pirt(void) -{ - return 0; -} - -int pciehprm_get_physical_slot_number(struct controller *ctrl, u32 *sun, u8 busnum, u8 devnum) -{ - - *sun = (u8) (ctrl->first_slot); - return 0; -} - - -static void print_pci_resource ( struct pci_resource *aprh) -{ - struct pci_resource *res; - - for (res = aprh; res; res = res->next) - dbg(" base= 0x%x length= 0x%x\n", res->base, res->length); -} - - -static void phprm_dump_func_res( struct pci_func *fun) -{ - struct pci_func *func = fun; - - if (func->bus_head) { - dbg(": BUS Resources:\n"); - print_pci_resource (func->bus_head); - } - if (func->io_head) { - dbg(": IO Resources:\n"); - print_pci_resource (func->io_head); - } - if (func->mem_head) { - dbg(": MEM Resources:\n"); - print_pci_resource (func->mem_head); - } - if (func->p_mem_head) { - dbg(": PMEM Resources:\n"); - print_pci_resource (func->p_mem_head); - } -} - -static int phprm_get_used_resources ( - struct controller *ctrl, - struct pci_func *func - ) -{ - return pciehp_save_used_resources (ctrl, func, !DISABLE_CARD); -} - -static int phprm_delete_resource( - struct pci_resource **aprh, - ulong base, - ulong size) -{ - struct pci_resource *res; - struct pci_resource *prevnode; - struct pci_resource *split_node; - ulong tbase; - - pciehp_resource_sort_and_combine(aprh); - - for (res = *aprh; res; res = res->next) { - if (res->base > base) - continue; - - if ((res->base + res->length) < (base + size)) - continue; - - if (res->base < base) { - tbase = base; - - if ((res->length - (tbase - res->base)) < size) - continue; - - split_node = (struct pci_resource *) kmalloc(sizeof(struct pci_resource), GFP_KERNEL); - if (!split_node) - return -ENOMEM; - - split_node->base = res->base; - split_node->length = tbase - res->base; - res->base = tbase; - res->length -= split_node->length; - - split_node->next = res->next; - res->next = split_node; - } - - if (res->length >= size) { - split_node = (struct pci_resource*) kmalloc(sizeof(struct pci_resource), GFP_KERNEL); - if (!split_node) - return -ENOMEM; - - split_node->base = res->base + size; - split_node->length = res->length - size; - res->length = size; - - split_node->next = res->next; - res->next = split_node; - } - - if (*aprh == res) { - *aprh = res->next; - } else { - prevnode = *aprh; - while (prevnode->next != res) - prevnode = prevnode->next; - - prevnode->next = res->next; - } - res->next = NULL; - kfree(res); - break; - } - - return 0; -} - - -static int phprm_delete_resources( - struct pci_resource **aprh, - struct pci_resource *this - ) -{ - struct pci_resource *res; - - for (res = this; res; res = res->next) - phprm_delete_resource(aprh, res->base, res->length); - - return 0; -} - - -static int configure_existing_function( - struct controller *ctrl, - struct pci_func *func - ) -{ - int rc; - - /* see how much resources the func has used. */ - rc = phprm_get_used_resources (ctrl, func); - - if (!rc) { - /* subtract the resources used by the func from ctrl resources */ - rc = phprm_delete_resources (&ctrl->bus_head, func->bus_head); - rc |= phprm_delete_resources (&ctrl->io_head, func->io_head); - rc |= phprm_delete_resources (&ctrl->mem_head, func->mem_head); - rc |= phprm_delete_resources (&ctrl->p_mem_head, func->p_mem_head); - if (rc) - warn("aCEF: cannot del used resources\n"); - } else - err("aCEF: cannot get used resources\n"); - - return rc; -} - -static int pciehprm_delete_resource( - struct pci_resource **aprh, - ulong base, - ulong size) -{ - struct pci_resource *res; - struct pci_resource *prevnode; - struct pci_resource *split_node; - ulong tbase; - - pciehp_resource_sort_and_combine(aprh); - - for (res = *aprh; res; res = res->next) { - if (res->base > base) - continue; - - if ((res->base + res->length) < (base + size)) - continue; - - if (res->base < base) { - tbase = base; - - if ((res->length - (tbase - res->base)) < size) - continue; - - split_node = (struct pci_resource *) kmalloc(sizeof(struct pci_resource), GFP_KERNEL); - if (!split_node) - return -ENOMEM; - - split_node->base = res->base; - split_node->length = tbase - res->base; - res->base = tbase; - res->length -= split_node->length; - - split_node->next = res->next; - res->next = split_node; - } - - if (res->length >= size) { - split_node = (struct pci_resource*) kmalloc(sizeof(struct pci_resource), GFP_KERNEL); - if (!split_node) - return -ENOMEM; - - split_node->base = res->base + size; - split_node->length = res->length - size; - res->length = size; - - split_node->next = res->next; - res->next = split_node; - } - - if (*aprh == res) { - *aprh = res->next; - } else { - prevnode = *aprh; - while (prevnode->next != res) - prevnode = prevnode->next; - - prevnode->next = res->next; - } - res->next = NULL; - kfree(res); - break; - } - - return 0; -} - -static int bind_pci_resources_to_slots ( struct controller *ctrl) +int pciehp_get_hp_hw_control_from_firmware(struct pci_dev *dev) { - struct pci_func *func, new_func; - int busn = ctrl->slot_bus; - int devn, funn; - u32 vid; - - for (devn = 0; devn < 32; devn++) { - for (funn = 0; funn < 8; funn++) { - /* - if (devn == ctrl->device && funn == ctrl->function) - continue; - */ - /* find out if this entry is for an occupied slot */ - vid = 0xFFFFFFFF; - - pci_bus_read_config_dword(ctrl->pci_dev->subordinate, PCI_DEVFN(devn, funn), PCI_VENDOR_ID, &vid); - - if (vid != 0xFFFFFFFF) { - dbg("%s: vid = %x bus %x dev %x fun %x\n", __FUNCTION__, - vid, busn, devn, funn); - func = pciehp_slot_find(busn, devn, funn); - dbg("%s: func = %p\n", __FUNCTION__,func); - if (!func) { - memset(&new_func, 0, sizeof(struct pci_func)); - new_func.bus = busn; - new_func.device = devn; - new_func.function = funn; - new_func.is_a_board = 1; - configure_existing_function(ctrl, &new_func); - phprm_dump_func_res(&new_func); - } else { - configure_existing_function(ctrl, func); - phprm_dump_func_res(func); - } - dbg("aCCF:existing PCI 0x%x Func ResourceDump\n", ctrl->bus); - } - } - } - return 0; } - -static void phprm_dump_ctrl_res( struct controller *ctlr) -{ - struct controller *ctrl = ctlr; - - if (ctrl->bus_head) { - dbg(": BUS Resources:\n"); - print_pci_resource (ctrl->bus_head); - } - if (ctrl->io_head) { - dbg(": IO Resources:\n"); - print_pci_resource (ctrl->io_head); - } - if (ctrl->mem_head) { - dbg(": MEM Resources:\n"); - print_pci_resource (ctrl->mem_head); - } - if (ctrl->p_mem_head) { - dbg(": PMEM Resources:\n"); - print_pci_resource (ctrl->p_mem_head); - } -} - -/* - * phprm_find_available_resources - * - * Finds available memory, IO, and IRQ resources for programming - * devices which may be added to the system - * this function is for hot plug ADD! - * - * returns 0 if success - */ -int pciehprm_find_available_resources(struct controller *ctrl) -{ - struct pci_func func; - u32 rc; - - memset(&func, 0, sizeof(struct pci_func)); - - func.bus = ctrl->bus; - func.device = ctrl->device; - func.function = ctrl->function; - func.is_a_board = 1; - - /* Get resources for this PCI bridge */ - rc = pciehp_save_used_resources (ctrl, &func, !DISABLE_CARD); - dbg("%s: pciehp_save_used_resources rc = %d\n", __FUNCTION__, rc); - - if (func.mem_head) - func.mem_head->next = ctrl->mem_head; - ctrl->mem_head = func.mem_head; - - if (func.p_mem_head) - func.p_mem_head->next = ctrl->p_mem_head; - ctrl->p_mem_head = func.p_mem_head; - - if (func.io_head) - func.io_head->next = ctrl->io_head; - ctrl->io_head = func.io_head; - - if(func.bus_head) - func.bus_head->next = ctrl->bus_head; - ctrl->bus_head = func.bus_head; - - if (ctrl->bus_head) - pciehprm_delete_resource(&ctrl->bus_head, ctrl->pci_dev->subordinate->number, 1); - - dbg("%s:pre-Bind PCI 0x%x Ctrl Resource Dump\n", __FUNCTION__, ctrl->bus); - phprm_dump_ctrl_res(ctrl); - - dbg("%s: before bind_pci_resources_to slots\n", __FUNCTION__); - - bind_pci_resources_to_slots (ctrl); - - dbg("%s:post-Bind PCI 0x%x Ctrl Resource Dump\n", __FUNCTION__, ctrl->bus); - phprm_dump_ctrl_res(ctrl); - - return (rc); -} - -int pciehprm_set_hpp( - struct controller *ctrl, - struct pci_func *func, - u8 card_type) -{ - u32 rc; - u8 temp_byte; - struct pci_bus lpci_bus, *pci_bus; - unsigned int devfn; - memcpy(&lpci_bus, ctrl->pci_bus, sizeof(lpci_bus)); - pci_bus = &lpci_bus; - pci_bus->number = func->bus; - devfn = PCI_DEVFN(func->device, func->function); - - temp_byte = 0x40; /* hard coded value for LT */ - if (card_type == PCI_HEADER_TYPE_BRIDGE) { - /* set subordinate Latency Timer */ - rc = pci_bus_write_config_byte(pci_bus, devfn, PCI_SEC_LATENCY_TIMER, temp_byte); - - if (rc) { - dbg("%s: set secondary LT error. b:d:f(%02x:%02x:%02x)\n", __FUNCTION__, - func->bus, func->device, func->function); - return rc; - } - } - - /* set base Latency Timer */ - rc = pci_bus_write_config_byte(pci_bus, devfn, PCI_LATENCY_TIMER, temp_byte); - - if (rc) { - dbg("%s: set LT error. b:d:f(%02x:%02x:%02x)\n", __FUNCTION__, func->bus, func->device, func->function); - return rc; - } - - /* set Cache Line size */ - temp_byte = 0x08; /* hard coded value for CLS */ - - rc = pci_bus_write_config_byte(pci_bus, devfn, PCI_CACHE_LINE_SIZE, temp_byte); - - if (rc) { - dbg("%s: set CLS error. b:d:f(%02x:%02x:%02x)\n", __FUNCTION__, func->bus, func->device, func->function); - } - - /* set enable_perr */ - /* set enable_serr */ - - return rc; -} - -void pciehprm_enable_card( - struct controller *ctrl, - struct pci_func *func, - u8 card_type) -{ - u16 command, bcommand; - struct pci_bus lpci_bus, *pci_bus; - unsigned int devfn; - int rc; - - memcpy(&lpci_bus, ctrl->pci_bus, sizeof(lpci_bus)); - pci_bus = &lpci_bus; - pci_bus->number = func->bus; - devfn = PCI_DEVFN(func->device, func->function); - - rc = pci_bus_read_config_word(pci_bus, devfn, PCI_COMMAND, &command); - - command |= PCI_COMMAND_PARITY | PCI_COMMAND_SERR - | PCI_COMMAND_MASTER | PCI_COMMAND_INVALIDATE - | PCI_COMMAND_IO | PCI_COMMAND_MEMORY; - - rc = pci_bus_write_config_word(pci_bus, devfn, PCI_COMMAND, command); - - if (card_type == PCI_HEADER_TYPE_BRIDGE) { - - rc = pci_bus_read_config_word(pci_bus, devfn, PCI_BRIDGE_CONTROL, &bcommand); - - bcommand |= PCI_BRIDGE_CTL_PARITY | PCI_BRIDGE_CTL_SERR - | PCI_BRIDGE_CTL_NO_ISA; - - rc = pci_bus_write_config_word(pci_bus, devfn, PCI_BRIDGE_CONTROL, bcommand); - } -} - -static int legacy_pciehprm_init_pci(void) -{ - return 0; -} - -int pciehprm_init(enum php_ctlr_type ctrl_type) -{ - int retval; - - switch (ctrl_type) { - case PCI: - retval = legacy_pciehprm_init_pci(); - break; - default: - retval = -ENODEV; - break; - } - - return retval; -} diff --git a/drivers/pci/hotplug/pciehprm_nonacpi.h b/drivers/pci/hotplug/pciehprm_nonacpi.h deleted file mode 100644 index b10603b0e95..00000000000 --- a/drivers/pci/hotplug/pciehprm_nonacpi.h +++ /dev/null @@ -1,56 +0,0 @@ -/* - * PCIEHPRM NONACPI: PHP Resource Manager for Non-ACPI/Legacy platform - * - * Copyright (C) 1995,2001 Compaq Computer Corporation - * Copyright (C) 2001 Greg Kroah-Hartman (greg@kroah.com) - * Copyright (C) 2001 IBM Corp. - * Copyright (C) 2003-2004 Intel Corporation - * - * All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or (at - * your option) any later version. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or - * NON INFRINGEMENT. See the GNU General Public License for more - * details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. - * - * Send feedback to <greg@kroah.com>, <kristen.c.accardi@intel.com> - * - */ - -#ifndef _PCIEHPRM_NONACPI_H_ -#define _PCIEHPRM_NONACPI_H_ - -struct irq_info { - u8 bus, devfn; /* bus, device and function */ - struct { - u8 link; /* IRQ line ID, chipset dependent, 0=not routed */ - u16 bitmap; /* Available IRQs */ - } __attribute__ ((packed)) irq[4]; - u8 slot; /* slot number, 0=onboard */ - u8 rfu; -} __attribute__ ((packed)); - -struct irq_routing_table { - u32 signature; /* PIRQ_SIGNATURE should be here */ - u16 version; /* PIRQ_VERSION */ - u16 size; /* Table size in bytes */ - u8 rtr_bus, rtr_devfn; /* Where the interrupt router lies */ - u16 exclusive_irqs; /* IRQs devoted exclusively to PCI usage */ - u16 rtr_vendor, rtr_device; /* Vendor and device ID of interrupt router */ - u32 miniport_data; /* Crap */ - u8 rfu[11]; - u8 checksum; /* Modulo 256 checksum must give zero */ - struct irq_info slots[0]; -} __attribute__ ((packed)); - -#endif /* _PCIEHPRM_NONACPI_H_ */ diff --git a/drivers/pci/hotplug/rpadlpar_core.c b/drivers/pci/hotplug/rpadlpar_core.c index fcb66b9a0e2..cc03609f45d 100644 --- a/drivers/pci/hotplug/rpadlpar_core.c +++ b/drivers/pci/hotplug/rpadlpar_core.c @@ -134,43 +134,6 @@ static void rpadlpar_claim_one_bus(struct pci_bus *b) rpadlpar_claim_one_bus(child_bus); } -static int pci_add_secondary_bus(struct device_node *dn, - struct pci_dev *bridge_dev) -{ - struct pci_dn *pdn = dn->data; - struct pci_controller *hose = pdn->phb; - struct pci_bus *child; - u8 sec_busno; - - /* Get busno of downstream bus */ - pci_read_config_byte(bridge_dev, PCI_SECONDARY_BUS, &sec_busno); - - /* Allocate and add to children of bridge_dev->bus */ - child = pci_add_new_bus(bridge_dev->bus, bridge_dev, sec_busno); - if (!child) { - printk(KERN_ERR "%s: could not add secondary bus\n", __FUNCTION__); - return -ENOMEM; - } - - sprintf(child->name, "PCI Bus #%02x", child->number); - - /* Fixup subordinate bridge bases and resources */ - pcibios_fixup_bus(child); - - /* Claim new bus resources */ - rpadlpar_claim_one_bus(bridge_dev->bus); - - if (hose->last_busno < child->number) - hose->last_busno = child->number; - - pdn->bussubno = child->number; - - /* ioremap() for child bus, which may or may not succeed */ - remap_bus_range(child); - - return 0; -} - static struct pci_dev *dlpar_find_new_dev(struct pci_bus *parent, struct device_node *dev_dn) { @@ -188,29 +151,41 @@ static struct pci_dev *dlpar_find_new_dev(struct pci_bus *parent, static struct pci_dev *dlpar_pci_add_bus(struct device_node *dn) { struct pci_dn *pdn = dn->data; - struct pci_controller *hose = pdn->phb; + struct pci_controller *phb = pdn->phb; struct pci_dev *dev = NULL; - /* Scan phb bus for EADS device, adding new one to bus->devices */ - if (!pci_scan_single_device(hose->bus, pdn->devfn)) { - printk(KERN_ERR "%s: found no device on bus\n", __FUNCTION__); + rpaphp_eeh_init_nodes(dn); + /* Add EADS device to PHB bus, adding new entry to bus->devices */ + dev = of_create_pci_dev(dn, phb->bus, pdn->devfn); + if (!dev) { + printk(KERN_ERR "%s: failed to create pci dev for %s\n", + __FUNCTION__, dn->full_name); return NULL; } + if (dev->hdr_type == PCI_HEADER_TYPE_BRIDGE || + dev->hdr_type == PCI_HEADER_TYPE_CARDBUS) + of_scan_pci_bridge(dn, dev); + + rpaphp_init_new_devs(dev->subordinate); + + /* Claim new bus resources */ + rpadlpar_claim_one_bus(dev->bus); + + /* ioremap() for child bus, which may or may not succeed */ + (void) remap_bus_range(dev->bus); + /* Add new devices to global lists. Register in proc, sysfs. */ - pci_bus_add_devices(hose->bus); + pci_bus_add_devices(phb->bus); /* Confirm new bridge dev was created */ - dev = dlpar_find_new_dev(hose->bus, dn); + dev = dlpar_find_new_dev(phb->bus, dn); if (dev) { if (dev->hdr_type != PCI_HEADER_TYPE_BRIDGE) { printk(KERN_ERR "%s: unexpected header type %d\n", __FUNCTION__, dev->hdr_type); return NULL; } - - if (pci_add_secondary_bus(dn, dev)) - return NULL; } return dev; @@ -219,7 +194,6 @@ static struct pci_dev *dlpar_pci_add_bus(struct device_node *dn) static int dlpar_add_pci_slot(char *drc_name, struct device_node *dn) { struct pci_dev *dev; - int rc; if (rpaphp_find_pci_bus(dn)) return -EINVAL; @@ -232,15 +206,6 @@ static int dlpar_add_pci_slot(char *drc_name, struct device_node *dn) return -EIO; } - if (dn->child) { - rc = rpaphp_config_pci_adapter(dev->subordinate); - if (rc < 0) { - printk(KERN_ERR "%s: unable to enable slot %s\n", - __FUNCTION__, drc_name); - return -EIO; - } - } - /* Add hotplug slot */ if (rpaphp_add_slot(dn)) { printk(KERN_ERR "%s: unable to add hotplug slot %s\n", @@ -306,7 +271,7 @@ static int dlpar_add_phb(char *drc_name, struct device_node *dn) { struct pci_controller *phb; - if (PCI_DN(dn)->phb) { + if (PCI_DN(dn) && PCI_DN(dn)->phb) { /* PHB already exists */ return -EINVAL; } @@ -435,6 +400,8 @@ int dlpar_remove_pci_slot(char *drc_name, struct device_node *dn) __FUNCTION__, drc_name); return -EIO; } + } else { + rpaphp_unconfig_pci_adapter(bus); } if (unmap_bus_range(bus)) { diff --git a/drivers/pci/hotplug/rpaphp.h b/drivers/pci/hotplug/rpaphp.h index 71ea5f9bb28..57ea71a7bda 100644 --- a/drivers/pci/hotplug/rpaphp.h +++ b/drivers/pci/hotplug/rpaphp.h @@ -93,6 +93,8 @@ extern int rpaphp_claim_resource(struct pci_dev *dev, int resource); extern int rpaphp_enable_pci_slot(struct slot *slot); extern int register_pci_slot(struct slot *slot); extern int rpaphp_get_pci_adapter_status(struct slot *slot, int is_init, u8 * value); +extern void rpaphp_init_new_devs(struct pci_bus *bus); +extern void rpaphp_eeh_init_nodes(struct device_node *dn); extern int rpaphp_config_pci_adapter(struct pci_bus *bus); extern int rpaphp_unconfig_pci_adapter(struct pci_bus *bus); diff --git a/drivers/pci/hotplug/rpaphp_pci.c b/drivers/pci/hotplug/rpaphp_pci.c index f7c12d7dfcf..a7859a84d1a 100644 --- a/drivers/pci/hotplug/rpaphp_pci.c +++ b/drivers/pci/hotplug/rpaphp_pci.c @@ -154,8 +154,7 @@ exit: } /* Must be called before pci_bus_add_devices */ -static void -rpaphp_fixup_new_pci_devices(struct pci_bus *bus, int fix_bus) +void rpaphp_fixup_new_pci_devices(struct pci_bus *bus, int fix_bus) { struct pci_dev *dev; @@ -184,6 +183,20 @@ rpaphp_fixup_new_pci_devices(struct pci_bus *bus, int fix_bus) } } +static void rpaphp_eeh_add_bus_device(struct pci_bus *bus) +{ + struct pci_dev *dev; + + list_for_each_entry(dev, &bus->devices, bus_list) { + eeh_add_device_late(dev); + if (dev->hdr_type == PCI_HEADER_TYPE_BRIDGE) { + struct pci_bus *subbus = dev->subordinate; + if (subbus) + rpaphp_eeh_add_bus_device (subbus); + } + } +} + static int rpaphp_pci_config_bridge(struct pci_dev *dev) { u8 sec_busno; @@ -217,6 +230,13 @@ static int rpaphp_pci_config_bridge(struct pci_dev *dev) return 0; } +void rpaphp_init_new_devs(struct pci_bus *bus) +{ + rpaphp_fixup_new_pci_devices(bus, 0); + rpaphp_eeh_add_bus_device(bus); +} +EXPORT_SYMBOL_GPL(rpaphp_init_new_devs); + /***************************************************************************** rpaphp_pci_config_slot() will configure all devices under the given slot->dn and return the the first pci_dev. @@ -233,36 +253,51 @@ rpaphp_pci_config_slot(struct pci_bus *bus) if (!dn || !dn->child) return NULL; - slotno = PCI_SLOT(PCI_DN(dn->child)->devfn); + if (systemcfg->platform == PLATFORM_PSERIES_LPAR) { + of_scan_bus(dn, bus); + if (list_empty(&bus->devices)) { + err("%s: No new device found\n", __FUNCTION__); + return NULL; + } - /* pci_scan_slot should find all children */ - num = pci_scan_slot(bus, PCI_DEVFN(slotno, 0)); - if (num) { - rpaphp_fixup_new_pci_devices(bus, 1); + rpaphp_init_new_devs(bus); pci_bus_add_devices(bus); - } - if (list_empty(&bus->devices)) { - err("%s: No new device found\n", __FUNCTION__); - return NULL; - } - list_for_each_entry(dev, &bus->devices, bus_list) { - if (dev->hdr_type == PCI_HEADER_TYPE_BRIDGE) - rpaphp_pci_config_bridge(dev); + dev = list_entry(&bus->devices, struct pci_dev, bus_list); + } else { + slotno = PCI_SLOT(PCI_DN(dn->child)->devfn); + + /* pci_scan_slot should find all children */ + num = pci_scan_slot(bus, PCI_DEVFN(slotno, 0)); + if (num) { + rpaphp_fixup_new_pci_devices(bus, 1); + pci_bus_add_devices(bus); + } + if (list_empty(&bus->devices)) { + err("%s: No new device found\n", __FUNCTION__); + return NULL; + } + list_for_each_entry(dev, &bus->devices, bus_list) { + if (dev->hdr_type == PCI_HEADER_TYPE_BRIDGE) + rpaphp_pci_config_bridge(dev); + + rpaphp_eeh_add_bus_device(bus); + } } return dev; } -static void enable_eeh(struct device_node *dn) +void rpaphp_eeh_init_nodes(struct device_node *dn) { struct device_node *sib; for (sib = dn->child; sib; sib = sib->sibling) - enable_eeh(sib); + rpaphp_eeh_init_nodes(sib); eeh_add_device_early(dn); return; } +EXPORT_SYMBOL_GPL(rpaphp_eeh_init_nodes); static void print_slot_pci_funcs(struct pci_bus *bus) { @@ -289,7 +324,7 @@ int rpaphp_config_pci_adapter(struct pci_bus *bus) if (!dn) goto exit; - enable_eeh(dn); + rpaphp_eeh_init_nodes(dn); dev = rpaphp_pci_config_slot(bus); if (!dev) { err("%s: can't find any devices.\n", __FUNCTION__); @@ -331,6 +366,7 @@ int rpaphp_unconfig_pci_adapter(struct pci_bus *bus) } return 0; } +EXPORT_SYMBOL_GPL(rpaphp_unconfig_pci_adapter); static int setup_pci_hotplug_slot_info(struct slot *slot) { @@ -444,8 +480,8 @@ int rpaphp_enable_pci_slot(struct slot *slot) retval = rpaphp_config_pci_adapter(slot->bus); if (!retval) { slot->state = CONFIGURED; - dbg("%s: PCI devices in slot[%s] has been configured\n", - __FUNCTION__, slot->name); + info("%s: devices in slot[%s] configured\n", + __FUNCTION__, slot->name); } else { slot->state = NOT_CONFIGURED; dbg("%s: no pci_dev struct for adapter in slot[%s]\n", diff --git a/drivers/pci/hotplug/shpchp_pci.c b/drivers/pci/hotplug/shpchp_pci.c index b8e95acea3b..38009bc0fd5 100644 --- a/drivers/pci/hotplug/shpchp_pci.c +++ b/drivers/pci/hotplug/shpchp_pci.c @@ -34,7 +34,7 @@ #include "../pci.h" #include "shpchp.h" -void program_fw_provided_values(struct pci_dev *dev) +static void program_fw_provided_values(struct pci_dev *dev) { u16 pci_cmd, pci_bctl; struct pci_dev *cdev; diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c index a2033552423..202b7507a35 100644 --- a/drivers/pci/msi.c +++ b/drivers/pci/msi.c @@ -23,6 +23,8 @@ #include "pci.h" #include "msi.h" +#define MSI_TARGET_CPU first_cpu(cpu_online_map) + static DEFINE_SPINLOCK(msi_lock); static struct msi_desc* msi_desc[NR_IRQS] = { [0 ... NR_IRQS-1] = NULL }; static kmem_cache_t* msi_cachep; @@ -92,6 +94,7 @@ static void set_msi_affinity(unsigned int vector, cpumask_t cpu_mask) struct msi_desc *entry; struct msg_address address; unsigned int irq = vector; + unsigned int dest_cpu = first_cpu(cpu_mask); entry = (struct msi_desc *)msi_desc[vector]; if (!entry || !entry->dev) @@ -108,9 +111,9 @@ static void set_msi_affinity(unsigned int vector, cpumask_t cpu_mask) pci_read_config_dword(entry->dev, msi_lower_address_reg(pos), &address.lo_address.value); address.lo_address.value &= MSI_ADDRESS_DEST_ID_MASK; - address.lo_address.value |= (cpu_mask_to_apicid(cpu_mask) << - MSI_TARGET_CPU_SHIFT); - entry->msi_attrib.current_cpu = cpu_mask_to_apicid(cpu_mask); + address.lo_address.value |= (cpu_physical_id(dest_cpu) << + MSI_TARGET_CPU_SHIFT); + entry->msi_attrib.current_cpu = cpu_physical_id(dest_cpu); pci_write_config_dword(entry->dev, msi_lower_address_reg(pos), address.lo_address.value); set_native_irq_info(irq, cpu_mask); @@ -123,9 +126,9 @@ static void set_msi_affinity(unsigned int vector, cpumask_t cpu_mask) address.lo_address.value = readl(entry->mask_base + offset); address.lo_address.value &= MSI_ADDRESS_DEST_ID_MASK; - address.lo_address.value |= (cpu_mask_to_apicid(cpu_mask) << - MSI_TARGET_CPU_SHIFT); - entry->msi_attrib.current_cpu = cpu_mask_to_apicid(cpu_mask); + address.lo_address.value |= (cpu_physical_id(dest_cpu) << + MSI_TARGET_CPU_SHIFT); + entry->msi_attrib.current_cpu = cpu_physical_id(dest_cpu); writel(address.lo_address.value, entry->mask_base + offset); set_native_irq_info(irq, cpu_mask); break; @@ -259,14 +262,15 @@ static void msi_data_init(struct msg_data *msi_data, static void msi_address_init(struct msg_address *msi_address) { unsigned int dest_id; + unsigned long dest_phys_id = cpu_physical_id(MSI_TARGET_CPU); memset(msi_address, 0, sizeof(struct msg_address)); msi_address->hi_address = (u32)0; dest_id = (MSI_ADDRESS_HEADER << MSI_ADDRESS_HEADER_SHIFT); - msi_address->lo_address.u.dest_mode = MSI_DEST_MODE; + msi_address->lo_address.u.dest_mode = MSI_PHYSICAL_MODE; msi_address->lo_address.u.redirection_hint = MSI_REDIRECTION_HINT_MODE; msi_address->lo_address.u.dest_id = dest_id; - msi_address->lo_address.value |= (MSI_TARGET_CPU << MSI_TARGET_CPU_SHIFT); + msi_address->lo_address.value |= (dest_phys_id << MSI_TARGET_CPU_SHIFT); } static int msi_free_vector(struct pci_dev* dev, int vector, int reassign); diff --git a/drivers/pci/pci-acpi.c b/drivers/pci/pci-acpi.c index e9e37abe1f7..a9b00cc2d88 100644 --- a/drivers/pci/pci-acpi.c +++ b/drivers/pci/pci-acpi.c @@ -91,9 +91,7 @@ acpi_query_osc ( static acpi_status acpi_run_osc ( acpi_handle handle, - u32 level, - void *context, - void **retval ) + void *context) { acpi_status status; struct acpi_object_list input; @@ -184,7 +182,7 @@ EXPORT_SYMBOL(pci_osc_support_set); * * Attempt to take control from Firmware on requested control bits. **/ -acpi_status pci_osc_control_set(u32 flags) +acpi_status pci_osc_control_set(acpi_handle handle, u32 flags) { acpi_status status; u32 ctrlset; @@ -198,10 +196,7 @@ acpi_status pci_osc_control_set(u32 flags) return AE_SUPPORT; } ctrlset_buf[OSC_CONTROL_TYPE] |= ctrlset; - status = acpi_get_devices ( PCI_ROOT_HID_STRING, - acpi_run_osc, - ctrlset_buf, - NULL ); + status = acpi_run_osc(handle, ctrlset_buf); if (ACPI_FAILURE (status)) { ctrlset_buf[OSC_CONTROL_TYPE] &= ~ctrlset; } diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index 94e68c54d27..a9046d4b8af 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -37,7 +37,7 @@ struct pci_dynid { * Adds a new dynamic pci device ID to this driver, * and causes the driver to probe for all devices again. */ -static inline ssize_t +static ssize_t store_new_id(struct device_driver *driver, const char *buf, size_t count) { struct pci_dynid *dynid; @@ -364,15 +364,16 @@ static struct kobj_type pci_driver_kobj_type = { }; /** - * pci_register_driver - register a new pci driver + * __pci_register_driver - register a new pci driver * @drv: the driver structure to register + * @owner: owner module of drv * * Adds the driver structure to the list of registered drivers. * Returns a negative value on error, otherwise 0. * If no error occurred, the driver remains registered even if * no device was claimed during registration. */ -int pci_register_driver(struct pci_driver *drv) +int __pci_register_driver(struct pci_driver *drv, struct module *owner) { int error; @@ -389,7 +390,7 @@ int pci_register_driver(struct pci_driver *drv) printk(KERN_WARNING "Warning: PCI driver %s has a struct " "device_driver shutdown method, please update!\n", drv->name); - drv->driver.owner = drv->owner; + drv->driver.owner = owner; drv->driver.kobj.ktype = &pci_driver_kobj_type; spin_lock_init(&drv->dynids.lock); @@ -526,7 +527,7 @@ postcore_initcall(pci_driver_init); EXPORT_SYMBOL(pci_match_id); EXPORT_SYMBOL(pci_match_device); -EXPORT_SYMBOL(pci_register_driver); +EXPORT_SYMBOL(__pci_register_driver); EXPORT_SYMBOL(pci_unregister_driver); EXPORT_SYMBOL(pci_dev_driver); EXPORT_SYMBOL(pci_bus_type); diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index e74d7584304..8e287a828d5 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -63,11 +63,38 @@ pci_max_busnr(void) return max; } +static int __pci_find_next_cap(struct pci_bus *bus, unsigned int devfn, u8 pos, int cap) +{ + u8 id; + int ttl = 48; + + while (ttl--) { + pci_bus_read_config_byte(bus, devfn, pos, &pos); + if (pos < 0x40) + break; + pos &= ~3; + pci_bus_read_config_byte(bus, devfn, pos + PCI_CAP_LIST_ID, + &id); + if (id == 0xff) + break; + if (id == cap) + return pos; + pos += PCI_CAP_LIST_NEXT; + } + return 0; +} + +int pci_find_next_capability(struct pci_dev *dev, u8 pos, int cap) +{ + return __pci_find_next_cap(dev->bus, dev->devfn, + pos + PCI_CAP_LIST_NEXT, cap); +} +EXPORT_SYMBOL_GPL(pci_find_next_capability); + static int __pci_bus_find_cap(struct pci_bus *bus, unsigned int devfn, u8 hdr_type, int cap) { u16 status; - u8 pos, id; - int ttl = 48; + u8 pos; pci_bus_read_config_word(bus, devfn, PCI_STATUS, &status); if (!(status & PCI_STATUS_CAP_LIST)) @@ -76,24 +103,15 @@ static int __pci_bus_find_cap(struct pci_bus *bus, unsigned int devfn, u8 hdr_ty switch (hdr_type) { case PCI_HEADER_TYPE_NORMAL: case PCI_HEADER_TYPE_BRIDGE: - pci_bus_read_config_byte(bus, devfn, PCI_CAPABILITY_LIST, &pos); + pos = PCI_CAPABILITY_LIST; break; case PCI_HEADER_TYPE_CARDBUS: - pci_bus_read_config_byte(bus, devfn, PCI_CB_CAPABILITY_LIST, &pos); + pos = PCI_CB_CAPABILITY_LIST; break; default: return 0; } - while (ttl-- && pos >= 0x40) { - pos &= ~3; - pci_bus_read_config_byte(bus, devfn, pos + PCI_CAP_LIST_ID, &id); - if (id == 0xff) - break; - if (id == cap) - return pos; - pci_bus_read_config_byte(bus, devfn, pos + PCI_CAP_LIST_NEXT, &pos); - } - return 0; + return __pci_find_next_cap(bus, devfn, pos, cap); } /** diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 5627ce1d2b3..3a4f49f4eff 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -462,11 +462,11 @@ static void __devinit quirk_vt82c686_acpi(struct pci_dev *dev) pci_read_config_word(dev, 0x70, &hm); hm &= PCI_BASE_ADDRESS_IO_MASK; - quirk_io_region(dev, hm, 128, PCI_BRIDGE_RESOURCES + 1, "vt82c868 HW-mon"); + quirk_io_region(dev, hm, 128, PCI_BRIDGE_RESOURCES + 1, "vt82c686 HW-mon"); pci_read_config_dword(dev, 0x90, &smb); smb &= PCI_BASE_ADDRESS_IO_MASK; - quirk_io_region(dev, smb, 16, PCI_BRIDGE_RESOURCES + 2, "vt82c868 SMB"); + quirk_io_region(dev, smb, 16, PCI_BRIDGE_RESOURCES + 2, "vt82c686 SMB"); } DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C686_4, quirk_vt82c686_acpi ); @@ -1243,6 +1243,21 @@ static void __devinit quirk_netmos(struct pci_dev *dev) } DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_NETMOS, PCI_ANY_ID, quirk_netmos); + +static void __devinit fixup_rev1_53c810(struct pci_dev* dev) +{ + /* rev 1 ncr53c810 chips don't set the class at all which means + * they don't get their resources remapped. Fix that here. + */ + + if (dev->class == PCI_CLASS_NOT_DEFINED) { + printk(KERN_INFO "NCR 53c810 rev 1 detected, setting PCI class.\n"); + dev->class = PCI_CLASS_STORAGE_SCSI; + } +} +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_NCR, PCI_DEVICE_ID_NCR_53C810, fixup_rev1_53c810); + + static void pci_do_fixups(struct pci_dev *dev, struct pci_fixup *f, struct pci_fixup *end) { while (f < end) { diff --git a/drivers/pcmcia/Kconfig b/drivers/pcmcia/Kconfig index ccf20039e90..309eb557f9a 100644 --- a/drivers/pcmcia/Kconfig +++ b/drivers/pcmcia/Kconfig @@ -156,7 +156,7 @@ config TCIC config PCMCIA_M8XX tristate "MPC8xx PCMCIA support" - depends on PCMCIA && PPC + depends on PCMCIA && PPC && 8xx select PCCARD_NONSTATIC help Say Y here to include support for PowerPC 8xx series PCMCIA diff --git a/drivers/pcmcia/Makefile b/drivers/pcmcia/Makefile index fe37541abbf..bcecf5133b7 100644 --- a/drivers/pcmcia/Makefile +++ b/drivers/pcmcia/Makefile @@ -25,7 +25,7 @@ obj-$(CONFIG_PD6729) += pd6729.o obj-$(CONFIG_I82365) += i82365.o obj-$(CONFIG_I82092) += i82092.o obj-$(CONFIG_TCIC) += tcic.o -obj-$(CONFIG_PCMCIA_M8XX) += m8xx_pcmcia.o +obj-$(CONFIG_PCMCIA_M8XX) += m8xx_pcmcia.o obj-$(CONFIG_HD64465_PCMCIA) += hd64465_ss.o obj-$(CONFIG_PCMCIA_SA1100) += sa11xx_core.o sa1100_cs.o obj-$(CONFIG_PCMCIA_SA1111) += sa11xx_core.o sa1111_cs.o @@ -47,10 +47,10 @@ au1x00_ss-$(CONFIG_MIPS_PB1200) += au1000_db1x00.o au1x00_ss-$(CONFIG_MIPS_PB1500) += au1000_pb1x00.o au1x00_ss-$(CONFIG_MIPS_DB1000) += au1000_db1x00.o au1x00_ss-$(CONFIG_MIPS_DB1100) += au1000_db1x00.o -au1x00_ss-$(CONFIG_MIPS_DB1200) += au1000_db1x00.o +au1x00_ss-$(CONFIG_MIPS_DB1200) += au1000_db1x00.o au1x00_ss-$(CONFIG_MIPS_DB1500) += au1000_db1x00.o au1x00_ss-$(CONFIG_MIPS_DB1550) += au1000_db1x00.o -au1x00_ss-$(CONFIG_MIPS_XXS1500) += au1000_xxs1500.o +au1x00_ss-$(CONFIG_MIPS_XXS1500) += au1000_xxs1500.o sa1111_cs-y += sa1111_generic.o sa1111_cs-$(CONFIG_ASSABET_NEPONSET) += sa1100_neponset.o diff --git a/drivers/pcmcia/au1000_db1x00.c b/drivers/pcmcia/au1000_db1x00.c index 24cfee1a412..abc13f28ba3 100644 --- a/drivers/pcmcia/au1000_db1x00.c +++ b/drivers/pcmcia/au1000_db1x00.c @@ -30,6 +30,7 @@ * */ +#include <linux/config.h> #include <linux/module.h> #include <linux/kernel.h> #include <linux/errno.h> diff --git a/drivers/pcmcia/au1000_generic.h b/drivers/pcmcia/au1000_generic.h index b0e7908392a..f2c970b5f4f 100644 --- a/drivers/pcmcia/au1000_generic.h +++ b/drivers/pcmcia/au1000_generic.h @@ -22,6 +22,8 @@ #define __ASM_AU1000_PCMCIA_H /* include the world */ +#include <linux/config.h> + #include <pcmcia/cs_types.h> #include <pcmcia/cs.h> #include <pcmcia/ss.h> diff --git a/drivers/pcmcia/au1000_pb1x00.c b/drivers/pcmcia/au1000_pb1x00.c index 86c0808d6a0..fd5522ede86 100644 --- a/drivers/pcmcia/au1000_pb1x00.c +++ b/drivers/pcmcia/au1000_pb1x00.c @@ -21,6 +21,7 @@ * with this program; if not, write to the Free Software Foundation, Inc., * 59 Temple Place - Suite 330, Boston MA 02111-1307, USA. */ +#include <linux/config.h> #include <linux/module.h> #include <linux/init.h> #include <linux/delay.h> diff --git a/drivers/pcmcia/au1000_xxs1500.c b/drivers/pcmcia/au1000_xxs1500.c index 01a895bc9a4..01874b0bb03 100644 --- a/drivers/pcmcia/au1000_xxs1500.c +++ b/drivers/pcmcia/au1000_xxs1500.c @@ -27,7 +27,6 @@ */ #include <linux/module.h> #include <linux/init.h> -#include <linux/config.h> #include <linux/delay.h> #include <linux/ioport.h> #include <linux/kernel.h> diff --git a/drivers/pcmcia/i82365.c b/drivers/pcmcia/i82365.c index 7ce455d01cc..4ddd76239b3 100644 --- a/drivers/pcmcia/i82365.c +++ b/drivers/pcmcia/i82365.c @@ -1366,6 +1366,7 @@ static int __init init_i82365(void) if (sockets == 0) { printk("not found.\n"); platform_device_unregister(&i82365_device); + release_region(i365_base, 2); driver_unregister(&i82365_driver); return -ENODEV; } diff --git a/drivers/pcmcia/m8xx_pcmcia.c b/drivers/pcmcia/m8xx_pcmcia.c index f8bed87cf2f..6d9f71cfcb3 100644 --- a/drivers/pcmcia/m8xx_pcmcia.c +++ b/drivers/pcmcia/m8xx_pcmcia.c @@ -39,7 +39,6 @@ #include <asm/io.h> #include <asm/bitops.h> -#include <asm/segment.h> #include <asm/system.h> #include <linux/kernel.h> @@ -50,6 +49,7 @@ #include <linux/ioport.h> #include <linux/delay.h> #include <linux/interrupt.h> +#include <linux/platform_device.h> #include <asm/mpc8xx.h> #include <asm/8xx_immap.h> @@ -546,29 +546,11 @@ static void m8xx_shutdown(void) free_irq(pcmcia_schlvl, NULL); } -/* copied from tcic.c */ - -static int m8xx_drv_suspend(struct device *dev, pm_message_t state, u32 level) -{ - int ret = 0; - if (level == SUSPEND_SAVE_STATE) - ret = pcmcia_socket_dev_suspend(dev, state); - return ret; -} - -static int m8xx_drv_resume(struct device *dev, u32 level) -{ - int ret = 0; - if (level == RESUME_RESTORE_STATE) - ret = pcmcia_socket_dev_resume(dev); - return ret; -} - static struct device_driver m8xx_driver = { .name = "m8xx-pcmcia", .bus = &platform_bus_type, - .suspend = m8xx_drv_suspend, - .resume = m8xx_drv_resume, + .suspend = pcmcia_socket_dev_suspend, + .resume = pcmcia_socket_dev_resume, }; static struct platform_device m8xx_device = { diff --git a/drivers/sbus/char/cpwatchdog.c b/drivers/sbus/char/cpwatchdog.c index 071ae24be89..fd2cc7782f7 100644 --- a/drivers/sbus/char/cpwatchdog.c +++ b/drivers/sbus/char/cpwatchdog.c @@ -407,7 +407,7 @@ static long wd_compat_ioctl(struct file *file, unsigned int cmd, case WIOCGSTAT: lock_kernel(); rval = wd_ioctl(file->f_dentry->d_inode, file, cmd, arg); - lock_kernel(); + unlock_kernel(); break; /* everything else is handled by the generic compat layer */ default: diff --git a/drivers/sbus/char/display7seg.c b/drivers/sbus/char/display7seg.c index 39f54213a6d..c3a51d1fae5 100644 --- a/drivers/sbus/char/display7seg.c +++ b/drivers/sbus/char/display7seg.c @@ -119,7 +119,7 @@ static long d7s_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { __u8 regs = readb(d7s_regs); __u8 ireg = 0; - int error = 0 + int error = 0; if (D7S_MINOR != iminor(file->f_dentry->d_inode)) return -ENODEV; @@ -161,7 +161,7 @@ static long d7s_ioctl(struct file *file, unsigned int cmd, unsigned long arg) writeb(regs, d7s_regs); break; }; - lock_kernel(); + unlock_kernel(); return error; } diff --git a/drivers/sbus/char/rtc.c b/drivers/sbus/char/rtc.c index 9b988baf0b5..5774bdd0e26 100644 --- a/drivers/sbus/char/rtc.c +++ b/drivers/sbus/char/rtc.c @@ -210,6 +210,27 @@ static int rtc_ioctl(struct inode *inode, struct file *file, unsigned int cmd, } } +static long rtc_compat_ioctl(struct file *file, unsigned int cmd, + unsigned long arg) +{ + int rval = -ENOIOCTLCMD; + + switch (cmd) { + /* + * These two are specific to this driver, the generic rtc ioctls + * are hanlded elsewhere. + */ + case RTCGET: + case RTCSET: + lock_kernel(); + rval = rtc_ioctl(file->f_dentry->d_inode, file, cmd, arg); + unlock_kernel(); + break; + } + + return rval; +} + static int rtc_open(struct inode *inode, struct file *file) { int ret; @@ -237,6 +258,7 @@ static struct file_operations rtc_fops = { .owner = THIS_MODULE, .llseek = no_llseek, .ioctl = rtc_ioctl, + .compat_ioctl = rtc_compat_ioctl, .open = rtc_open, .release = rtc_release, }; diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c index c888af4a456..3553da0e1cd 100644 --- a/drivers/scsi/ide-scsi.c +++ b/drivers/scsi/ide-scsi.c @@ -395,6 +395,7 @@ static int idescsi_end_request (ide_drive_t *drive, int uptodate, int nrsecs) int log = test_bit(IDESCSI_LOG_CMD, &scsi->log); struct Scsi_Host *host; u8 *scsi_buf; + int errors = rq->errors; unsigned long flags; if (!(rq->flags & (REQ_SPECIAL|REQ_SENSE))) { @@ -421,11 +422,11 @@ static int idescsi_end_request (ide_drive_t *drive, int uptodate, int nrsecs) printk (KERN_WARNING "ide-scsi: %s: timed out for %lu\n", drive->name, pc->scsi_cmd->serial_number); pc->scsi_cmd->result = DID_TIME_OUT << 16; - } else if (rq->errors >= ERROR_MAX) { + } else if (errors >= ERROR_MAX) { pc->scsi_cmd->result = DID_ERROR << 16; if (log) printk ("ide-scsi: %s: I/O error for %lu\n", drive->name, pc->scsi_cmd->serial_number); - } else if (rq->errors) { + } else if (errors) { if (log) printk ("ide-scsi: %s: check condition for %lu\n", drive->name, pc->scsi_cmd->serial_number); if (!idescsi_check_condition(drive, rq)) diff --git a/drivers/scsi/lpfc/lpfc_init.c b/drivers/scsi/lpfc/lpfc_init.c index c90723860a0..07498118359 100644 --- a/drivers/scsi/lpfc/lpfc_init.c +++ b/drivers/scsi/lpfc/lpfc_init.c @@ -1704,7 +1704,6 @@ MODULE_DEVICE_TABLE(pci, lpfc_id_table); static struct pci_driver lpfc_driver = { .name = LPFC_DRIVER_NAME, - .owner = THIS_MODULE, .id_table = lpfc_id_table, .probe = lpfc_pci_probe_one, .remove = __devexit_p(lpfc_pci_remove_one), diff --git a/drivers/usb/gadget/goku_udc.c b/drivers/usb/gadget/goku_udc.c index 654469778ab..b0f3cd63e3b 100644 --- a/drivers/usb/gadget/goku_udc.c +++ b/drivers/usb/gadget/goku_udc.c @@ -1970,7 +1970,6 @@ MODULE_DEVICE_TABLE (pci, pci_ids); static struct pci_driver goku_pci_driver = { .name = (char *) driver_name, .id_table = pci_ids, - .owner = THIS_MODULE, .probe = goku_probe, .remove = goku_remove, diff --git a/drivers/usb/gadget/net2280.c b/drivers/usb/gadget/net2280.c index 0dc6bb00bf7..c32e1f7476d 100644 --- a/drivers/usb/gadget/net2280.c +++ b/drivers/usb/gadget/net2280.c @@ -2948,7 +2948,6 @@ MODULE_DEVICE_TABLE (pci, pci_ids); static struct pci_driver net2280_pci_driver = { .name = (char *) driver_name, .id_table = pci_ids, - .owner = THIS_MODULE, .probe = net2280_probe, .remove = net2280_remove, diff --git a/drivers/usb/host/ehci-pci.c b/drivers/usb/host/ehci-pci.c index 14500885396..dfd9bd0b182 100644 --- a/drivers/usb/host/ehci-pci.c +++ b/drivers/usb/host/ehci-pci.c @@ -383,7 +383,6 @@ MODULE_DEVICE_TABLE (pci, pci_ids); static struct pci_driver ehci_pci_driver = { .name = (char *) hcd_name, .id_table = pci_ids, - .owner = THIS_MODULE, .probe = usb_hcd_pci_probe, .remove = usb_hcd_pci_remove, diff --git a/drivers/usb/host/ohci-pci.c b/drivers/usb/host/ohci-pci.c index 7ce1d9ef028..a59e536441e 100644 --- a/drivers/usb/host/ohci-pci.c +++ b/drivers/usb/host/ohci-pci.c @@ -218,7 +218,6 @@ MODULE_DEVICE_TABLE (pci, pci_ids); static struct pci_driver ohci_pci_driver = { .name = (char *) hcd_name, .id_table = pci_ids, - .owner = THIS_MODULE, .probe = usb_hcd_pci_probe, .remove = usb_hcd_pci_remove, diff --git a/drivers/usb/host/uhci-hcd.c b/drivers/usb/host/uhci-hcd.c index 15e0a511069..d33ce3982a5 100644 --- a/drivers/usb/host/uhci-hcd.c +++ b/drivers/usb/host/uhci-hcd.c @@ -831,7 +831,6 @@ MODULE_DEVICE_TABLE(pci, uhci_pci_ids); static struct pci_driver uhci_pci_driver = { .name = (char *)hcd_name, .id_table = uhci_pci_ids, - .owner = THIS_MODULE, .probe = usb_hcd_pci_probe, .remove = usb_hcd_pci_remove, diff --git a/drivers/video/backlight/corgi_bl.c b/drivers/video/backlight/corgi_bl.c index 4867498f68e..bd9a6996aee 100644 --- a/drivers/video/backlight/corgi_bl.c +++ b/drivers/video/backlight/corgi_bl.c @@ -48,6 +48,12 @@ static void corgibl_send_intensity(int intensity) corgibl_mach_set_intensity(intensity); spin_unlock_irqrestore(&bl_lock, flags); + + corgi_kick_batt = symbol_get(sharpsl_battery_kick); + if (corgi_kick_batt) { + corgi_kick_batt(); + symbol_put(sharpsl_battery_kick); + } } static void corgibl_blank(int blank) diff --git a/fs/Kconfig b/fs/Kconfig index 7d6ae369ce4..d5255e627b5 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -1601,9 +1601,10 @@ config CIFS PC operating systems. The CIFS protocol is fully supported by file servers such as Windows 2000 (including Windows 2003, NT 4 and Windows XP) as well by Samba (which provides excellent CIFS - server support for Linux and many other operating systems). Currently - you must use the smbfs client filesystem to access older SMB servers - such as Windows 9x and OS/2. + server support for Linux and many other operating systems). Limited + support for Windows ME and similar servers is provided as well. + You must use the smbfs client filesystem to access older SMB servers + such as OS/2 and DOS. The intent of the cifs module is to provide an advanced network file system client for mounting to CIFS compliant servers, @@ -1614,7 +1615,7 @@ config CIFS cifs if running only a (Samba) server. It is possible to enable both smbfs and cifs (e.g. if you are using CIFS for accessing Windows 2003 and Samba 3 servers, and smbfs for accessing old servers). If you need - to mount to Samba or Windows 2003 servers from this machine, say Y. + to mount to Samba or Windows from this machine, say Y. config CIFS_STATS bool "CIFS statistics" @@ -1623,8 +1624,22 @@ config CIFS_STATS Enabling this option will cause statistics for each server share mounted by the cifs client to be displayed in /proc/fs/cifs/Stats +config CIFS_STATS2 + bool "CIFS extended statistics" + depends on CIFS_STATS + help + Enabling this option will allow more detailed statistics on SMB + request timing to be displayed in /proc/fs/cifs/DebugData and also + allow optional logging of slow responses to dmesg (depending on the + value of /proc/fs/cifs/cifsFYI, see fs/cifs/README for more details). + These additional statistics may have a minor effect on performance + and memory utilization. + + Unless you are a developer or are doing network performance analysis + or tuning, say N. + config CIFS_XATTR - bool "CIFS extended attributes (EXPERIMENTAL)" + bool "CIFS extended attributes" depends on CIFS help Extended attributes are name:value pairs associated with inodes by @@ -1636,11 +1651,11 @@ config CIFS_XATTR prefaced by the user namespace prefix. The system namespace (used by some filesystems to store ACLs) is not supported at this time. - + If unsure, say N. config CIFS_POSIX - bool "CIFS POSIX Extensions (EXPERIMENTAL)" + bool "CIFS POSIX Extensions" depends on CIFS_XATTR help Enabling this option will cause the cifs client to attempt to @@ -1653,10 +1668,28 @@ config CIFS_POSIX config CIFS_EXPERIMENTAL bool "CIFS Experimental Features (EXPERIMENTAL)" - depends on CIFS + depends on CIFS && EXPERIMENTAL + help + Enables cifs features under testing. These features are + experimental and currently include support for writepages + (multipage writebehind performance improvements) and directory + change notification ie fcntl(F_DNOTIFY) as well as some security + improvements. Some also depend on setting at runtime the + pseudo-file /proc/fs/cifs/Experimental (which is disabled by + default). See the file fs/cifs/README for more details. + + If unsure, say N. + +config CIFS_UPCALL + bool "CIFS Kerberos/SPNEGO advanced session setup (EXPERIMENTAL)" + depends on CIFS_EXPERIMENTAL + select CONNECTOR help - Enables cifs features under testing. These features - are highly experimental. If unsure, say N. + Enables an upcall mechanism for CIFS which will be used to contact + userspace helper utilities to provide SPNEGO packaged Kerberos + tickets which are needed to mount to certain secure servers + (for which more secure Kerberos authentication is required). If + unsure, say N. config NCP_FS tristate "NCP file system support (to mount NetWare volumes)" diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index 5bab24f5905..eab3750cf30 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -2,7 +2,7 @@ Version 1.39 ------------ Defer close of a file handle slightly if pending writes depend on that file handle (this reduces the EBADF bad file handle errors that can be logged under heavy -stress on writes). +stress on writes). Modify cifs Kconfig options to expose CONFIG_CIFS_STATS2 Version 1.38 ------------ diff --git a/fs/cifs/cifs_unicode.c b/fs/cifs/cifs_unicode.c index 99a096d3f84..4e12053f080 100644 --- a/fs/cifs/cifs_unicode.c +++ b/fs/cifs/cifs_unicode.c @@ -74,10 +74,11 @@ cifs_strtoUCS(wchar_t * to, const char *from, int len, cERROR(1, ("cifs_strtoUCS: char2uni returned %d", charlen)); - to[i] = cpu_to_le16(0x003f); /* a question mark */ + /* A question mark */ + to[i] = (wchar_t)cpu_to_le16(0x003f); charlen = 1; } else - to[i] = cpu_to_le16(to[i]); + to[i] = (wchar_t)cpu_to_le16(to[i]); } diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 877095a1192..682b0235ad9 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -405,6 +405,7 @@ static struct quotactl_ops cifs_quotactl_ops = { }; #endif +#ifdef CONFIG_CIFS_EXPERIMENTAL static void cifs_umount_begin(struct super_block * sblock) { struct cifs_sb_info *cifs_sb; @@ -422,16 +423,18 @@ static void cifs_umount_begin(struct super_block * sblock) tcon->tidStatus = CifsExiting; up(&tcon->tconSem); + /* cancel_brl_requests(tcon); */ + /* cancel_notify_requests(tcon); */ if(tcon->ses && tcon->ses->server) { - cERROR(1,("wake up tasks now - umount begin not complete")); + cFYI(1,("wake up tasks now - umount begin not complete")); wake_up_all(&tcon->ses->server->request_q); } /* BB FIXME - finish add checks for tidStatus BB */ return; } - +#endif static int cifs_remount(struct super_block *sb, int *flags, char *data) { @@ -450,7 +453,9 @@ struct super_operations cifs_super_ops = { unless later we add lazy close of inodes or unless the kernel forgets to call us with the same number of releases (closes) as opens */ .show_options = cifs_show_options, -/* .umount_begin = cifs_umount_begin, */ /* BB finish in the future */ +#ifdef CONFIG_CIFS_EXPERIMENTAL + .umount_begin = cifs_umount_begin, +#endif .remount_fs = cifs_remount, }; diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index d301149b1bb..1b73f4f4c5c 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -242,11 +242,11 @@ extern int CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, const int netfid, const unsigned int count, const __u64 offset, unsigned int *nbytes, struct kvec *iov, const int nvec, const int long_op); +#endif /* CONFIG_CIFS_EXPERIMENTAL */ extern int CIFSGetSrvInodeNumber(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, __u64 * inode_number, const struct nls_table *nls_codepage, int remap_special_chars); -#endif /* CONFIG_CIFS_EXPERIMENTAL */ extern int cifs_convertUCSpath(char *target, const __le16 *source, int maxlen, const struct nls_table * codepage); extern int cifsConvertToUCS(__le16 * target, const char *source, int maxlen, diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 9312bfc5668..a53c596e108 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -2959,7 +2959,6 @@ CIFSFindClose(const int xid, struct cifsTconInfo *tcon, const __u16 searchHandle return rc; } -#ifdef CONFIG_CIFS_EXPERIMENTAL int CIFSGetSrvInodeNumber(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, @@ -3053,7 +3052,6 @@ GetInodeNumOut: goto GetInodeNumberRetry; return rc; } -#endif /* CIFS_EXPERIMENTAL */ int CIFSGetDFSRefer(const int xid, struct cifsSesInfo *ses, diff --git a/fs/cifs/cn_cifs.h b/fs/cifs/cn_cifs.h new file mode 100644 index 00000000000..ea59ccac2eb --- /dev/null +++ b/fs/cifs/cn_cifs.h @@ -0,0 +1,37 @@ +/* + * fs/cifs/cn_cifs.h + * + * Copyright (c) International Business Machines Corp., 2002 + * Author(s): Steve French (sfrench@us.ibm.com) + * + * This library is free software; you can redistribute it and/or modify + * it under the terms of the GNU Lesser General Public License as published + * by the Free Software Foundation; either version 2.1 of the License, or + * (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See + * the GNU Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public License + * along with this library; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#ifndef _CN_CIFS_H +#define _CN_CIFS_H +#ifdef CONFIG_CIFS_UPCALL +#include <linux/types.h> +#include <linux/connector.h> + +struct cifs_upcall { + char signature[4]; /* CIFS */ + enum command { + CIFS_GET_IP = 0x00000001, /* get ip address for hostname */ + CIFS_GET_SECBLOB = 0x00000002, /* get SPNEGO wrapped blob */ + } command; + /* union cifs upcall data follows */ +}; +#endif /* CIFS_UPCALL */ +#endif /* _CN_CIFS_H */ diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 450ab75d654..2cb620716bc 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -42,6 +42,7 @@ #include "ntlmssp.h" #include "nterr.h" #include "rfc1002pdu.h" +#include "cn_cifs.h" #define CIFS_PORT 445 #define RFC1001_PORT 139 diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index 912d401600f..923d071163b 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -283,7 +283,6 @@ int cifs_get_inode_info(struct inode **pinode, there Windows server or network appliances for which IndexNumber field is not guaranteed unique? */ -#ifdef CONFIG_CIFS_EXPERIMENTAL if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM){ int rc1 = 0; __u64 inode_num; @@ -299,7 +298,6 @@ int cifs_get_inode_info(struct inode **pinode, } else /* do we need cast or hash to ino? */ (*pinode)->i_ino = inode_num; } /* else ino incremented to unique num in new_inode*/ -#endif /* CIFS_EXPERIMENTAL */ insert_inode_hash(*pinode); } inode = *pinode; diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c index 1abe7343f92..4abbe860430 100644 --- a/fs/jfs/namei.c +++ b/fs/jfs/namei.c @@ -827,6 +827,7 @@ static int jfs_link(struct dentry *old_dentry, /* update object inode */ ip->i_nlink++; /* for new link */ ip->i_ctime = CURRENT_TIME; + dir->i_ctime = dir->i_mtime = CURRENT_TIME; mark_inode_dirty(dir); atomic_inc(&ip->i_count); @@ -1024,6 +1025,8 @@ static int jfs_symlink(struct inode *dip, struct dentry *dentry, insert_inode_hash(ip); mark_inode_dirty(ip); + dip->i_ctime = dip->i_mtime = CURRENT_TIME; + mark_inode_dirty(dip); /* * commit update of parent directory and link object */ diff --git a/include/asm-alpha/ide.h b/include/asm-alpha/ide.h index 68934a25931..6126afe2738 100644 --- a/include/asm-alpha/ide.h +++ b/include/asm-alpha/ide.h @@ -15,10 +15,6 @@ #include <linux/config.h> -#ifndef MAX_HWIFS -#define MAX_HWIFS CONFIG_IDE_MAX_HWIFS -#endif - #define IDE_ARCH_OBSOLETE_DEFAULTS static inline int ide_default_irq(unsigned long base) diff --git a/include/asm-arm/arch-ixp4xx/hardware.h b/include/asm-arm/arch-ixp4xx/hardware.h index 55d85eea8c1..cfb413c845f 100644 --- a/include/asm-arm/arch-ixp4xx/hardware.h +++ b/include/asm-arm/arch-ixp4xx/hardware.h @@ -44,5 +44,6 @@ extern unsigned int processor_id; #include "ixdp425.h" #include "coyote.h" #include "prpmc1100.h" +#include "nslu2.h" #endif /* _ASM_ARCH_HARDWARE_H */ diff --git a/include/asm-arm/arch-ixp4xx/irqs.h b/include/asm-arm/arch-ixp4xx/irqs.h index ca808281c7f..2cf4930372b 100644 --- a/include/asm-arm/arch-ixp4xx/irqs.h +++ b/include/asm-arm/arch-ixp4xx/irqs.h @@ -93,4 +93,11 @@ #define IRQ_COYOTE_PCI_SLOT1 IRQ_IXP4XX_GPIO11 #define IRQ_COYOTE_IDE IRQ_IXP4XX_GPIO5 +/* + * NSLU2 board IRQs + */ +#define IRQ_NSLU2_PCI_INTA IRQ_IXP4XX_GPIO11 +#define IRQ_NSLU2_PCI_INTB IRQ_IXP4XX_GPIO10 +#define IRQ_NSLU2_PCI_INTC IRQ_IXP4XX_GPIO9 + #endif diff --git a/include/asm-arm/arch-ixp4xx/nslu2.h b/include/asm-arm/arch-ixp4xx/nslu2.h new file mode 100644 index 00000000000..b8b347a559c --- /dev/null +++ b/include/asm-arm/arch-ixp4xx/nslu2.h @@ -0,0 +1,96 @@ +/* + * include/asm-arm/arch-ixp4xx/nslu2.h + * + * NSLU2 platform specific definitions + * + * Author: Mark Rakes <mrakes AT mac.com> + * Maintainers: http://www.nslu2-linux.org + * + * based on ixdp425.h: + * Copyright 2004 (c) MontaVista, Software, Inc. + * + * This file is licensed under the terms of the GNU General Public + * License version 2. This program is licensed "as is" without any + * warranty of any kind, whether express or implied. + */ + +#ifndef __ASM_ARCH_HARDWARE_H__ +#error "Do not include this directly, instead #include <asm/hardware.h>" +#endif + +#define NSLU2_FLASH_BASE IXP4XX_EXP_BUS_CS0_BASE_PHYS +#define NSLU2_FLASH_SIZE IXP4XX_EXP_BUS_CSX_REGION_SIZE + +#define NSLU2_SDA_PIN 7 +#define NSLU2_SCL_PIN 6 + +/* + * NSLU2 PCI IRQs + */ +#define NSLU2_PCI_MAX_DEV 3 +#define NSLU2_PCI_IRQ_LINES 3 + + +/* PCI controller GPIO to IRQ pin mappings */ +#define NSLU2_PCI_INTA_PIN 11 +#define NSLU2_PCI_INTB_PIN 10 +#define NSLU2_PCI_INTC_PIN 9 +#define NSLU2_PCI_INTD_PIN 8 + + +/* NSLU2 Timer */ +#define NSLU2_FREQ 66000000 +#define NSLU2_CLOCK_TICK_RATE (((NSLU2_FREQ / HZ & ~IXP4XX_OST_RELOAD_MASK) + 1) * HZ) +#define NSLU2_CLOCK_TICKS_PER_USEC ((NSLU2_CLOCK_TICK_RATE + USEC_PER_SEC/2) / USEC_PER_SEC) + +/* GPIO */ + +#define NSLU2_GPIO0 0 +#define NSLU2_GPIO1 1 +#define NSLU2_GPIO2 2 +#define NSLU2_GPIO3 3 +#define NSLU2_GPIO4 4 +#define NSLU2_GPIO5 5 +#define NSLU2_GPIO6 6 +#define NSLU2_GPIO7 7 +#define NSLU2_GPIO8 8 +#define NSLU2_GPIO9 9 +#define NSLU2_GPIO10 10 +#define NSLU2_GPIO11 11 +#define NSLU2_GPIO12 12 +#define NSLU2_GPIO13 13 +#define NSLU2_GPIO14 14 +#define NSLU2_GPIO15 15 + +/* Buttons */ + +#define NSLU2_PB_GPIO NSLU2_GPIO5 +#define NSLU2_PO_GPIO NSLU2_GPIO8 /* power off */ +#define NSLU2_RB_GPIO NSLU2_GPIO12 + +#define NSLU2_PB_IRQ IRQ_IXP4XX_GPIO5 +#define NSLU2_RB_IRQ IRQ_IXP4XX_GPIO12 + +#define NSLU2_PB_BM (1L << NSLU2_PB_GPIO) +#define NSLU2_PO_BM (1L << NSLU2_PO_GPIO) +#define NSLU2_RB_BM (1L << NSLU2_RB_GPIO) + +/* Buzzer */ + +#define NSLU2_GPIO_BUZZ 4 +#define NSLU2_BZ_BM (1L << NSLU2_GPIO_BUZZ) +/* LEDs */ + +#define NSLU2_LED_RED NSLU2_GPIO0 +#define NSLU2_LED_GRN NSLU2_GPIO1 + +#define NSLU2_LED_RED_BM (1L << NSLU2_LED_RED) +#define NSLU2_LED_GRN_BM (1L << NSLU2_LED_GRN) + +#define NSLU2_LED_DISK1 NSLU2_GPIO2 +#define NSLU2_LED_DISK2 NSLU2_GPIO3 + +#define NSLU2_LED_DISK1_BM (1L << NSLU2_GPIO2) +#define NSLU2_LED_DISK2_BM (1L << NSLU2_GPIO3) + + diff --git a/include/asm-arm/arch-omap/board-h4.h b/include/asm-arm/arch-omap/board-h4.h index d64ee9211ee..33ea29a4165 100644 --- a/include/asm-arm/arch-omap/board-h4.h +++ b/include/asm-arm/arch-omap/board-h4.h @@ -34,5 +34,11 @@ #define OMAP24XX_ETHR_START 0x08000300 #define OMAP24XX_ETHR_GPIO_IRQ 92 +#define H4_CS0_BASE 0x04000000 + +#define H4_CS0_BASE 0x04000000 + +#define H4_CS0_BASE 0x04000000 + #endif /* __ASM_ARCH_OMAP_H4_H */ diff --git a/include/asm-arm/arch-omap/board-innovator.h b/include/asm-arm/arch-omap/board-innovator.h index 79574e0ed13..b3cf33441f6 100644 --- a/include/asm-arm/arch-omap/board-innovator.h +++ b/include/asm-arm/arch-omap/board-innovator.h @@ -26,7 +26,7 @@ #ifndef __ASM_ARCH_OMAP_INNOVATOR_H #define __ASM_ARCH_OMAP_INNOVATOR_H -#if defined (CONFIG_ARCH_OMAP1510) +#if defined (CONFIG_ARCH_OMAP15XX) #ifndef OMAP_SDRAM_DEVICE #define OMAP_SDRAM_DEVICE D256M_1X16_4B @@ -44,7 +44,7 @@ void fpga_write(unsigned char val, int reg); unsigned char fpga_read(int reg); #endif -#endif /* CONFIG_ARCH_OMAP1510 */ +#endif /* CONFIG_ARCH_OMAP15XX */ #if defined (CONFIG_ARCH_OMAP16XX) diff --git a/include/asm-arm/arch-omap/clock.h b/include/asm-arm/arch-omap/clock.h new file mode 100644 index 00000000000..740c297eb11 --- /dev/null +++ b/include/asm-arm/arch-omap/clock.h @@ -0,0 +1,91 @@ +/* + * linux/include/asm-arm/arch-omap/clock.h + * + * Copyright (C) 2004 - 2005 Nokia corporation + * Written by Tuukka Tikkanen <tuukka.tikkanen@elektrobit.com> + * Based on clocks.h by Tony Lindgren, Gordon McNutt and RidgeRun, Inc + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#ifndef __ARCH_ARM_OMAP_CLOCK_H +#define __ARCH_ARM_OMAP_CLOCK_H + +struct module; + +struct clk { + struct list_head node; + struct module *owner; + const char *name; + struct clk *parent; + unsigned long rate; + __u32 flags; + void __iomem *enable_reg; + __u8 enable_bit; + __u8 rate_offset; + __u8 src_offset; + __s8 usecount; + void (*recalc)(struct clk *); + int (*set_rate)(struct clk *, unsigned long); + long (*round_rate)(struct clk *, unsigned long); + void (*init)(struct clk *); + int (*enable)(struct clk *); + void (*disable)(struct clk *); +}; + +struct clk_functions { + int (*clk_enable)(struct clk *clk); + void (*clk_disable)(struct clk *clk); + int (*clk_use)(struct clk *clk); + void (*clk_unuse)(struct clk *clk); + long (*clk_round_rate)(struct clk *clk, unsigned long rate); + int (*clk_set_rate)(struct clk *clk, unsigned long rate); + int (*clk_set_parent)(struct clk *clk, struct clk *parent); + struct clk * (*clk_get_parent)(struct clk *clk); + void (*clk_allow_idle)(struct clk *clk); + void (*clk_deny_idle)(struct clk *clk); +}; + +extern unsigned int mpurate; +extern struct list_head clocks; +extern spinlock_t clockfw_lock; + +extern int clk_init(struct clk_functions * custom_clocks); +extern int clk_register(struct clk *clk); +extern void clk_unregister(struct clk *clk); +extern void propagate_rate(struct clk *clk); +extern void followparent_recalc(struct clk * clk); +extern void clk_allow_idle(struct clk *clk); +extern void clk_deny_idle(struct clk *clk); + +/* Clock flags */ +#define RATE_CKCTL (1 << 0) /* Main fixed ratio clocks */ +#define RATE_FIXED (1 << 1) /* Fixed clock rate */ +#define RATE_PROPAGATES (1 << 2) /* Program children too */ +#define VIRTUAL_CLOCK (1 << 3) /* Composite clock from table */ +#define ALWAYS_ENABLED (1 << 4) /* Clock cannot be disabled */ +#define ENABLE_REG_32BIT (1 << 5) /* Use 32-bit access */ +#define VIRTUAL_IO_ADDRESS (1 << 6) /* Clock in virtual address */ +#define CLOCK_IDLE_CONTROL (1 << 7) +#define CLOCK_NO_IDLE_PARENT (1 << 8) +#define DELAYED_APP (1 << 9) /* Delay application of clock */ +#define CONFIG_PARTICIPANT (1 << 10) /* Fundamental clock */ +#define CM_MPU_SEL1 (1 << 11) /* Domain divider/source */ +#define CM_DSP_SEL1 (1 << 12) +#define CM_GFX_SEL1 (1 << 13) +#define CM_MODEM_SEL1 (1 << 14) +#define CM_CORE_SEL1 (1 << 15) /* Sets divider for many */ +#define CM_CORE_SEL2 (1 << 16) /* sets parent for GPT */ +#define CM_WKUP_SEL1 (1 << 17) +#define CM_PLL_SEL1 (1 << 18) +#define CM_PLL_SEL2 (1 << 19) +#define CM_SYSCLKOUT_SEL1 (1 << 20) +#define CLOCK_IN_OMAP730 (1 << 21) +#define CLOCK_IN_OMAP1510 (1 << 22) +#define CLOCK_IN_OMAP16XX (1 << 23) +#define CLOCK_IN_OMAP242X (1 << 24) +#define CLOCK_IN_OMAP243X (1 << 25) + +#endif diff --git a/include/asm-arm/arch-omap/common.h b/include/asm-arm/arch-omap/common.h index 2a676b4f13b..08d58abd821 100644 --- a/include/asm-arm/arch-omap/common.h +++ b/include/asm-arm/arch-omap/common.h @@ -31,6 +31,6 @@ struct sys_timer; extern void omap_map_common_io(void); extern struct sys_timer omap_timer; -extern void omap_serial_init(int ports[]); +extern void omap_serial_init(void); #endif /* __ARCH_ARM_MACH_OMAP_COMMON_H */ diff --git a/include/asm-arm/arch-omap/cpu.h b/include/asm-arm/arch-omap/cpu.h index 1119e2b53e7..ec7eb675d92 100644 --- a/include/asm-arm/arch-omap/cpu.h +++ b/include/asm-arm/arch-omap/cpu.h @@ -28,12 +28,7 @@ extern unsigned int system_rev; -#define OMAP_DIE_ID_0 0xfffe1800 -#define OMAP_DIE_ID_1 0xfffe1804 -#define OMAP_PRODUCTION_ID_0 0xfffe2000 -#define OMAP_PRODUCTION_ID_1 0xfffe2004 -#define OMAP32_ID_0 0xfffed400 -#define OMAP32_ID_1 0xfffed404 +#define omap2_cpu_rev() ((system_rev >> 8) & 0x0f) /* * Test if multicore OMAP support is needed @@ -50,7 +45,7 @@ extern unsigned int system_rev; # define OMAP_NAME omap730 # endif #endif -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX # ifdef OMAP_NAME # undef MULTI_OMAP1 # define MULTI_OMAP1 @@ -79,9 +74,11 @@ extern unsigned int system_rev; * Macros to group OMAP into cpu classes. * These can be used in most places. * cpu_is_omap7xx(): True for OMAP730 - * cpu_is_omap15xx(): True for OMAP1510 and OMAP5910 + * cpu_is_omap15xx(): True for OMAP1510, OMAP5910 and OMAP310 * cpu_is_omap16xx(): True for OMAP1610, OMAP5912 and OMAP1710 - * cpu_is_omap24xx(): True for OMAP2420 + * cpu_is_omap24xx(): True for OMAP2420, OMAP2422, OMAP2423, OMAP2430 + * cpu_is_omap242x(): True for OMAP2420, OMAP2422, OMAP2423 + * cpu_is_omap243x(): True for OMAP2430 */ #define GET_OMAP_CLASS (system_rev & 0xff) @@ -91,22 +88,35 @@ static inline int is_omap ##class (void) \ return (GET_OMAP_CLASS == (id)) ? 1 : 0; \ } +#define GET_OMAP_SUBCLASS ((system_rev >> 20) & 0x0fff) + +#define IS_OMAP_SUBCLASS(subclass, id) \ +static inline int is_omap ##subclass (void) \ +{ \ + return (GET_OMAP_SUBCLASS == (id)) ? 1 : 0; \ +} + IS_OMAP_CLASS(7xx, 0x07) IS_OMAP_CLASS(15xx, 0x15) IS_OMAP_CLASS(16xx, 0x16) IS_OMAP_CLASS(24xx, 0x24) +IS_OMAP_SUBCLASS(242x, 0x242) +IS_OMAP_SUBCLASS(243x, 0x243) + #define cpu_is_omap7xx() 0 #define cpu_is_omap15xx() 0 #define cpu_is_omap16xx() 0 #define cpu_is_omap24xx() 0 +#define cpu_is_omap242x() 0 +#define cpu_is_omap243x() 0 #if defined(MULTI_OMAP1) # if defined(CONFIG_ARCH_OMAP730) # undef cpu_is_omap7xx # define cpu_is_omap7xx() is_omap7xx() # endif -# if defined(CONFIG_ARCH_OMAP1510) +# if defined(CONFIG_ARCH_OMAP15XX) # undef cpu_is_omap15xx # define cpu_is_omap15xx() is_omap15xx() # endif @@ -119,7 +129,7 @@ IS_OMAP_CLASS(24xx, 0x24) # undef cpu_is_omap7xx # define cpu_is_omap7xx() 1 # endif -# if defined(CONFIG_ARCH_OMAP1510) +# if defined(CONFIG_ARCH_OMAP15XX) # undef cpu_is_omap15xx # define cpu_is_omap15xx() 1 # endif @@ -129,13 +139,18 @@ IS_OMAP_CLASS(24xx, 0x24) # endif # if defined(CONFIG_ARCH_OMAP24XX) # undef cpu_is_omap24xx +# undef cpu_is_omap242x +# undef cpu_is_omap243x # define cpu_is_omap24xx() 1 +# define cpu_is_omap242x() is_omap242x() +# define cpu_is_omap243x() is_omap243x() # endif #endif /* * Macros to detect individual cpu types. * These are only rarely needed. + * cpu_is_omap330(): True for OMAP330 * cpu_is_omap730(): True for OMAP730 * cpu_is_omap1510(): True for OMAP1510 * cpu_is_omap1610(): True for OMAP1610 @@ -144,6 +159,9 @@ IS_OMAP_CLASS(24xx, 0x24) * cpu_is_omap1621(): True for OMAP1621 * cpu_is_omap1710(): True for OMAP1710 * cpu_is_omap2420(): True for OMAP2420 + * cpu_is_omap2422(): True for OMAP2422 + * cpu_is_omap2423(): True for OMAP2423 + * cpu_is_omap2430(): True for OMAP2430 */ #define GET_OMAP_TYPE ((system_rev >> 16) & 0xffff) @@ -153,6 +171,7 @@ static inline int is_omap ##type (void) \ return (GET_OMAP_TYPE == (id)) ? 1 : 0; \ } +IS_OMAP_TYPE(310, 0x0310) IS_OMAP_TYPE(730, 0x0730) IS_OMAP_TYPE(1510, 0x1510) IS_OMAP_TYPE(1610, 0x1610) @@ -161,7 +180,11 @@ IS_OMAP_TYPE(5912, 0x1611) IS_OMAP_TYPE(1621, 0x1621) IS_OMAP_TYPE(1710, 0x1710) IS_OMAP_TYPE(2420, 0x2420) +IS_OMAP_TYPE(2422, 0x2422) +IS_OMAP_TYPE(2423, 0x2423) +IS_OMAP_TYPE(2430, 0x2430) +#define cpu_is_omap310() 0 #define cpu_is_omap730() 0 #define cpu_is_omap1510() 0 #define cpu_is_omap1610() 0 @@ -170,31 +193,33 @@ IS_OMAP_TYPE(2420, 0x2420) #define cpu_is_omap1621() 0 #define cpu_is_omap1710() 0 #define cpu_is_omap2420() 0 +#define cpu_is_omap2422() 0 +#define cpu_is_omap2423() 0 +#define cpu_is_omap2430() 0 #if defined(MULTI_OMAP1) # if defined(CONFIG_ARCH_OMAP730) # undef cpu_is_omap730 # define cpu_is_omap730() is_omap730() # endif -# if defined(CONFIG_ARCH_OMAP1510) -# undef cpu_is_omap1510 -# define cpu_is_omap1510() is_omap1510() -# endif #else # if defined(CONFIG_ARCH_OMAP730) # undef cpu_is_omap730 # define cpu_is_omap730() 1 # endif -# if defined(CONFIG_ARCH_OMAP1510) -# undef cpu_is_omap1510 -# define cpu_is_omap1510() 1 -# endif #endif /* * Whether we have MULTI_OMAP1 or not, we still need to distinguish - * between 1611B/5912 and 1710. + * between 330 vs. 1510 and 1611B/5912 vs. 1710. */ +#if defined(CONFIG_ARCH_OMAP15XX) +# undef cpu_is_omap310 +# undef cpu_is_omap1510 +# define cpu_is_omap310() is_omap310() +# define cpu_is_omap1510() is_omap1510() +#endif + #if defined(CONFIG_ARCH_OMAP16XX) # undef cpu_is_omap1610 # undef cpu_is_omap1611 @@ -208,9 +233,20 @@ IS_OMAP_TYPE(2420, 0x2420) # define cpu_is_omap1710() is_omap1710() #endif -#if defined(CONFIG_ARCH_OMAP2420) -# undef cpu_is_omap2420 -# define cpu_is_omap2420() 1 +#if defined(CONFIG_ARCH_OMAP24XX) +# undef cpu_is_omap2420 +# undef cpu_is_omap2422 +# undef cpu_is_omap2423 +# undef cpu_is_omap2430 +# define cpu_is_omap2420() is_omap2420() +# define cpu_is_omap2422() is_omap2422() +# define cpu_is_omap2423() is_omap2423() +# define cpu_is_omap2430() is_omap2430() #endif +/* Macros to detect if we have OMAP1 or OMAP2 */ +#define cpu_class_is_omap1() (cpu_is_omap730() || cpu_is_omap15xx() || \ + cpu_is_omap16xx()) +#define cpu_class_is_omap2() cpu_is_omap24xx() + #endif diff --git a/include/asm-arm/arch-omap/dma.h b/include/asm-arm/arch-omap/dma.h index 04ebef5c6e9..ccbcb580a5c 100644 --- a/include/asm-arm/arch-omap/dma.h +++ b/include/asm-arm/arch-omap/dma.h @@ -22,9 +22,109 @@ #define __ASM_ARCH_DMA_H #define MAX_DMA_ADDRESS 0xffffffff +#define MAX_DMA_CHANNELS 0 + +/* Hardware registers for omap1 */ +#define OMAP_DMA_BASE (0xfffed800) +#define OMAP_DMA_GCR (OMAP_DMA_BASE + 0x400) +#define OMAP_DMA_GSCR (OMAP_DMA_BASE + 0x404) +#define OMAP_DMA_GRST (OMAP_DMA_BASE + 0x408) +#define OMAP_DMA_HW_ID (OMAP_DMA_BASE + 0x442) +#define OMAP_DMA_PCH2_ID (OMAP_DMA_BASE + 0x444) +#define OMAP_DMA_PCH0_ID (OMAP_DMA_BASE + 0x446) +#define OMAP_DMA_PCH1_ID (OMAP_DMA_BASE + 0x448) +#define OMAP_DMA_PCHG_ID (OMAP_DMA_BASE + 0x44a) +#define OMAP_DMA_PCHD_ID (OMAP_DMA_BASE + 0x44c) +#define OMAP_DMA_CAPS_0_U (OMAP_DMA_BASE + 0x44e) +#define OMAP_DMA_CAPS_0_L (OMAP_DMA_BASE + 0x450) +#define OMAP_DMA_CAPS_1_U (OMAP_DMA_BASE + 0x452) +#define OMAP_DMA_CAPS_1_L (OMAP_DMA_BASE + 0x454) +#define OMAP_DMA_CAPS_2 (OMAP_DMA_BASE + 0x456) +#define OMAP_DMA_CAPS_3 (OMAP_DMA_BASE + 0x458) +#define OMAP_DMA_CAPS_4 (OMAP_DMA_BASE + 0x45a) +#define OMAP_DMA_PCH2_SR (OMAP_DMA_BASE + 0x460) +#define OMAP_DMA_PCH0_SR (OMAP_DMA_BASE + 0x480) +#define OMAP_DMA_PCH1_SR (OMAP_DMA_BASE + 0x482) +#define OMAP_DMA_PCHD_SR (OMAP_DMA_BASE + 0x4c0) + +/* Hardware registers for omap2 */ +#define OMAP24XX_DMA_BASE (L4_24XX_BASE + 0x56000) +#define OMAP_DMA4_REVISION (OMAP24XX_DMA_BASE + 0x00) +#define OMAP_DMA4_GCR_REG (OMAP24XX_DMA_BASE + 0x78) +#define OMAP_DMA4_IRQSTATUS_L0 (OMAP24XX_DMA_BASE + 0x08) +#define OMAP_DMA4_IRQSTATUS_L1 (OMAP24XX_DMA_BASE + 0x0c) +#define OMAP_DMA4_IRQSTATUS_L2 (OMAP24XX_DMA_BASE + 0x10) +#define OMAP_DMA4_IRQSTATUS_L3 (OMAP24XX_DMA_BASE + 0x14) +#define OMAP_DMA4_IRQENABLE_L0 (OMAP24XX_DMA_BASE + 0x18) +#define OMAP_DMA4_IRQENABLE_L1 (OMAP24XX_DMA_BASE + 0x1c) +#define OMAP_DMA4_IRQENABLE_L2 (OMAP24XX_DMA_BASE + 0x20) +#define OMAP_DMA4_IRQENABLE_L3 (OMAP24XX_DMA_BASE + 0x24) +#define OMAP_DMA4_SYSSTATUS (OMAP24XX_DMA_BASE + 0x28) +#define OMAP_DMA4_CAPS_0 (OMAP24XX_DMA_BASE + 0x64) +#define OMAP_DMA4_CAPS_2 (OMAP24XX_DMA_BASE + 0x6c) +#define OMAP_DMA4_CAPS_3 (OMAP24XX_DMA_BASE + 0x70) +#define OMAP_DMA4_CAPS_4 (OMAP24XX_DMA_BASE + 0x74) + +#ifdef CONFIG_ARCH_OMAP1 #define OMAP_LOGICAL_DMA_CH_COUNT 17 +/* Common channel specific registers for omap1 */ +#define OMAP_DMA_CSDP_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x00) +#define OMAP_DMA_CCR_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x02) +#define OMAP_DMA_CICR_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x04) +#define OMAP_DMA_CSR_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x06) +#define OMAP_DMA_CEN_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x10) +#define OMAP_DMA_CFN_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x12) +#define OMAP_DMA_CSFI_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x14) +#define OMAP_DMA_CSEI_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x16) +#define OMAP_DMA_CSAC_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x18) +#define OMAP_DMA_CDAC_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x1a) +#define OMAP_DMA_CDEI_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x1c) +#define OMAP_DMA_CDFI_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x1e) +#define OMAP_DMA_CLNK_CTRL_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x28) + +#else + +#define OMAP_LOGICAL_DMA_CH_COUNT 32 /* REVISIT: Is this 32 + 2? */ + +/* Common channel specific registers for omap2 */ +#define OMAP_DMA_CCR_REG(n) __REG32(OMAP24XX_DMA_BASE + 0x60 * (n) + 0x80) +#define OMAP_DMA_CLNK_CTRL_REG(n) __REG32(OMAP24XX_DMA_BASE + 0x60 * (n) + 0x84) +#define OMAP_DMA_CICR_REG(n) __REG32(OMAP24XX_DMA_BASE + 0x60 * (n) + 0x88) +#define OMAP_DMA_CSR_REG(n) __REG32(OMAP24XX_DMA_BASE + 0x60 * (n) + 0x8c) +#define OMAP_DMA_CSDP_REG(n) __REG32(OMAP24XX_DMA_BASE + 0x60 * (n) + 0x90) +#define OMAP_DMA_CEN_REG(n) __REG32(OMAP24XX_DMA_BASE + 0x60 * (n) + 0x94) +#define OMAP_DMA_CFN_REG(n) __REG32(OMAP24XX_DMA_BASE + 0x60 * (n) + 0x98) +#define OMAP_DMA_CSEI_REG(n) __REG32(OMAP24XX_DMA_BASE + 0x60 * (n) + 0xa4) +#define OMAP_DMA_CSFI_REG(n) __REG32(OMAP24XX_DMA_BASE + 0x60 * (n) + 0xa8) +#define OMAP_DMA_CDEI_REG(n) __REG32(OMAP24XX_DMA_BASE + 0x60 * (n) + 0xac) +#define OMAP_DMA_CDFI_REG(n) __REG32(OMAP24XX_DMA_BASE + 0x60 * (n) + 0xb0) +#define OMAP_DMA_CSAC_REG(n) __REG32(OMAP24XX_DMA_BASE + 0x60 * (n) + 0xb4) +#define OMAP_DMA_CDAC_REG(n) __REG32(OMAP24XX_DMA_BASE + 0x60 * (n) + 0xb8) + +#endif + +/* Channel specific registers only on omap1 */ +#define OMAP1_DMA_CSSA_L_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x08) +#define OMAP1_DMA_CSSA_U_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x0a) +#define OMAP1_DMA_CDSA_L_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x0c) +#define OMAP1_DMA_CDSA_U_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x0e) +#define OMAP1_DMA_COLOR_L_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x20) +#define OMAP1_DMA_CCR2_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x24) +#define OMAP1_DMA_COLOR_U_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x22) +#define OMAP1_DMA_LCH_CTRL_REG(n) __REG16(OMAP_DMA_BASE + 0x40 * (n) + 0x2a) + +/* Channel specific registers only on omap2 */ +#define OMAP2_DMA_CSSA_REG(n) __REG32(OMAP24XX_DMA_BASE + 0x60 * (n) + 0x9c) +#define OMAP2_DMA_CDSA_REG(n) __REG32(OMAP24XX_DMA_BASE + 0x60 * (n) + 0xa0) +#define OMAP2_DMA_CCEN_REG(n) __REG32(OMAP24XX_DMA_BASE + 0x60 * (n) + 0xbc) +#define OMAP2_DMA_CCFN_REG(n) __REG32(OMAP24XX_DMA_BASE + 0x60 * (n) + 0xc0) +#define OMAP2_DMA_COLOR_REG(n) __REG32(OMAP24XX_DMA_BASE + 0x60 * (n) + 0xc4) + +/*----------------------------------------------------------------------------*/ + +/* DMA channels for omap1 */ #define OMAP_DMA_NO_DEVICE 0 #define OMAP_DMA_MCSI1_TX 1 #define OMAP_DMA_MCSI1_RX 2 @@ -85,29 +185,72 @@ #define OMAP_DMA_MMC2_RX 55 #define OMAP_DMA_CRYPTO_DES_OUT 56 +/* DMA channels for 24xx */ +#define OMAP24XX_DMA_NO_DEVICE 0 +#define OMAP24XX_DMA_XTI_DMA 1 /* S_DMA_0 */ +#define OMAP24XX_DMA_EXT_NDMA_REQ0 2 /* S_DMA_1 */ +#define OMAP24XX_DMA_EXT_NDMA_REQ1 3 /* S_DMA_2 */ +#define OMAP24XX_DMA_GPMC 4 /* S_DMA_3 */ +#define OMAP24XX_DMA_GFX 5 /* S_DMA_4 */ +#define OMAP24XX_DMA_DSS 6 /* S_DMA_5 */ +#define OMAP24XX_DMA_VLYNQ_TX 7 /* S_DMA_6 */ +#define OMAP24XX_DMA_CWT 8 /* S_DMA_7 */ +#define OMAP24XX_DMA_AES_TX 9 /* S_DMA_8 */ +#define OMAP24XX_DMA_AES_RX 10 /* S_DMA_9 */ +#define OMAP24XX_DMA_DES_TX 11 /* S_DMA_10 */ +#define OMAP24XX_DMA_DES_RX 12 /* S_DMA_11 */ +#define OMAP24XX_DMA_SHA1MD5_RX 13 /* S_DMA_12 */ -#define OMAP_DMA_BASE (0xfffed800) -#define OMAP_DMA_GCR (OMAP_DMA_BASE + 0x400) -#define OMAP_DMA_GSCR (OMAP_DMA_BASE + 0x404) -#define OMAP_DMA_GRST (OMAP_DMA_BASE + 0x408) -#define OMAP_DMA_HW_ID (OMAP_DMA_BASE + 0x442) -#define OMAP_DMA_PCH2_ID (OMAP_DMA_BASE + 0x444) -#define OMAP_DMA_PCH0_ID (OMAP_DMA_BASE + 0x446) -#define OMAP_DMA_PCH1_ID (OMAP_DMA_BASE + 0x448) -#define OMAP_DMA_PCHG_ID (OMAP_DMA_BASE + 0x44a) -#define OMAP_DMA_PCHD_ID (OMAP_DMA_BASE + 0x44c) -#define OMAP_DMA_CAPS_0_U (OMAP_DMA_BASE + 0x44e) -#define OMAP_DMA_CAPS_0_L (OMAP_DMA_BASE + 0x450) -#define OMAP_DMA_CAPS_1_U (OMAP_DMA_BASE + 0x452) -#define OMAP_DMA_CAPS_1_L (OMAP_DMA_BASE + 0x454) -#define OMAP_DMA_CAPS_2 (OMAP_DMA_BASE + 0x456) -#define OMAP_DMA_CAPS_3 (OMAP_DMA_BASE + 0x458) -#define OMAP_DMA_CAPS_4 (OMAP_DMA_BASE + 0x45a) -#define OMAP_DMA_PCH2_SR (OMAP_DMA_BASE + 0x460) -#define OMAP_DMA_PCH0_SR (OMAP_DMA_BASE + 0x480) -#define OMAP_DMA_PCH1_SR (OMAP_DMA_BASE + 0x482) -#define OMAP_DMA_PCHD_SR (OMAP_DMA_BASE + 0x4c0) +#define OMAP24XX_DMA_EAC_AC_RD 17 /* S_DMA_16 */ +#define OMAP24XX_DMA_EAC_AC_WR 18 /* S_DMA_17 */ +#define OMAP24XX_DMA_EAC_MD_UL_RD 19 /* S_DMA_18 */ +#define OMAP24XX_DMA_EAC_MD_UL_WR 20 /* S_DMA_19 */ +#define OMAP24XX_DMA_EAC_MD_DL_RD 21 /* S_DMA_20 */ +#define OMAP24XX_DMA_EAC_MD_DL_WR 22 /* S_DMA_21 */ +#define OMAP24XX_DMA_EAC_BT_UL_RD 23 /* S_DMA_22 */ +#define OMAP24XX_DMA_EAC_BT_UL_WR 24 /* S_DMA_23 */ +#define OMAP24XX_DMA_EAC_BT_DL_RD 25 /* S_DMA_24 */ +#define OMAP24XX_DMA_EAC_BT_DL_WR 26 /* S_DMA_25 */ +#define OMAP24XX_DMA_I2C1_TX 27 /* S_DMA_26 */ +#define OMAP24XX_DMA_I2C1_RX 28 /* S_DMA_27 */ +#define OMAP24XX_DMA_I2C2_TX 29 /* S_DMA_28 */ +#define OMAP24XX_DMA_I2C2_RX 30 /* S_DMA_29 */ +#define OMAP24XX_DMA_MCBSP1_TX 31 /* SDMA_30 */ +#define OMAP24XX_DMA_MCBSP1_RX 32 /* SDMA_31 */ +#define OMAP24XX_DMA_MCBSP2_TX 33 /* SDMA_32 */ +#define OMAP24XX_DMA_MCBSP2_RX 34 /* SDMA_33 */ +#define OMAP24XX_DMA_SPI1_TX0 35 /* SDMA_34 */ +#define OMAP24XX_DMA_SPI1_RX0 36 /* SDMA_35 */ +#define OMAP24XX_DMA_SPI1_TX1 37 /* SDMA_36 */ +#define OMAP24XX_DMA_SPI1_RX1 38 /* SDMA_37 */ +#define OMAP24XX_DMA_SPI1_TX2 39 /* SDMA_38 */ +#define OMAP24XX_DMA_SPI1_RX2 40 /* SDMA_39 */ +#define OMAP24XX_DMA_SPI1_TX3 41 /* SDMA_40 */ +#define OMAP24XX_DMA_SPI1_RX3 42 /* SDMA_41 */ +#define OMAP24XX_DMA_SPI2_TX0 43 /* SDMA_42 */ +#define OMAP24XX_DMA_SPI2_RX0 44 /* SDMA_43 */ +#define OMAP24XX_DMA_SPI2_TX1 45 /* SDMA_44 */ +#define OMAP24XX_DMA_SPI2_RX1 46 /* SDMA_45 */ +#define OMAP24XX_DMA_UART1_TX 49 /* SDMA_48 */ +#define OMAP24XX_DMA_UART1_RX 50 /* SDMA_49 */ +#define OMAP24XX_DMA_UART2_TX 51 /* SDMA_50 */ +#define OMAP24XX_DMA_UART2_RX 52 /* SDMA_51 */ +#define OMAP24XX_DMA_UART3_TX 53 /* SDMA_52 */ +#define OMAP24XX_DMA_UART3_RX 54 /* SDMA_53 */ +#define OMAP24XX_DMA_USB_W2FC_TX0 55 /* SDMA_54 */ +#define OMAP24XX_DMA_USB_W2FC_RX0 56 /* SDMA_55 */ +#define OMAP24XX_DMA_USB_W2FC_TX1 57 /* SDMA_56 */ +#define OMAP24XX_DMA_USB_W2FC_RX1 58 /* SDMA_57 */ +#define OMAP24XX_DMA_USB_W2FC_TX2 59 /* SDMA_58 */ +#define OMAP24XX_DMA_USB_W2FC_RX2 60 /* SDMA_59 */ +#define OMAP24XX_DMA_MMC1_TX 61 /* SDMA_60 */ +#define OMAP24XX_DMA_MMC1_RX 62 /* SDMA_61 */ +#define OMAP24XX_DMA_MS 63 /* SDMA_62 */ + +/*----------------------------------------------------------------------------*/ + +/* Hardware registers for LCD DMA */ #define OMAP1510_DMA_LCD_BASE (0xfffedb00) #define OMAP1510_DMA_LCD_CTRL (OMAP1510_DMA_LCD_BASE + 0x00) #define OMAP1510_DMA_LCD_TOP_F1_L (OMAP1510_DMA_LCD_BASE + 0x02) @@ -116,7 +259,7 @@ #define OMAP1510_DMA_LCD_BOT_F1_U (OMAP1510_DMA_LCD_BASE + 0x08) #define OMAP1610_DMA_LCD_BASE (0xfffee300) -#define OMAP1610_DMA_LCD_CSDP (OMAP1610_DMA_LCD_BASE + 0xc0) +#define OMAP1610_DMA_LCD_CSDP (OMAP1610_DMA_LCD_BASE + 0xc0) #define OMAP1610_DMA_LCD_CCR (OMAP1610_DMA_LCD_BASE + 0xc2) #define OMAP1610_DMA_LCD_CTRL (OMAP1610_DMA_LCD_BASE + 0xc4) #define OMAP1610_DMA_LCD_TOP_B1_L (OMAP1610_DMA_LCD_BASE + 0xc8) @@ -134,37 +277,18 @@ #define OMAP1610_DMA_LCD_LCH_CTRL (OMAP1610_DMA_LCD_BASE + 0xea) #define OMAP1610_DMA_LCD_SRC_FI_B1_U (OMAP1610_DMA_LCD_BASE + 0xf4) - -/* Every LCh has its own set of the registers below */ -#define OMAP_DMA_CSDP(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x00) -#define OMAP_DMA_CCR(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x02) -#define OMAP_DMA_CICR(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x04) -#define OMAP_DMA_CSR(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x06) -#define OMAP_DMA_CSSA_L(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x08) -#define OMAP_DMA_CSSA_U(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x0a) -#define OMAP_DMA_CDSA_L(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x0c) -#define OMAP_DMA_CDSA_U(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x0e) -#define OMAP_DMA_CEN(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x10) -#define OMAP_DMA_CFN(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x12) -#define OMAP_DMA_CSFI(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x14) -#define OMAP_DMA_CSEI(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x16) -#define OMAP_DMA_CSAC(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x18) -#define OMAP_DMA_CDAC(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x1a) -#define OMAP_DMA_CDEI(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x1c) -#define OMAP_DMA_CDFI(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x1e) -#define OMAP_DMA_COLOR_L(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x20) -#define OMAP_DMA_COLOR_U(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x22) -#define OMAP_DMA_CCR2(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x24) -#define OMAP_DMA_CLNK_CTRL(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x28) -#define OMAP_DMA_LCH_CTRL(n) (OMAP_DMA_BASE + 0x40 * (n) + 0x2a) - -#define OMAP_DMA_TOUT_IRQ (1 << 0) +#define OMAP_DMA_TOUT_IRQ (1 << 0) /* Only on omap1 */ #define OMAP_DMA_DROP_IRQ (1 << 1) #define OMAP_DMA_HALF_IRQ (1 << 2) #define OMAP_DMA_FRAME_IRQ (1 << 3) #define OMAP_DMA_LAST_IRQ (1 << 4) #define OMAP_DMA_BLOCK_IRQ (1 << 5) -#define OMAP_DMA_SYNC_IRQ (1 << 6) +#define OMAP1_DMA_SYNC_IRQ (1 << 6) +#define OMAP2_DMA_PKT_IRQ (1 << 7) +#define OMAP2_DMA_TRANS_ERR_IRQ (1 << 8) +#define OMAP2_DMA_SECURE_ERR_IRQ (1 << 9) +#define OMAP2_DMA_SUPERVISOR_ERR_IRQ (1 << 10) +#define OMAP2_DMA_MISALIGNED_ERR_IRQ (1 << 11) #define OMAP_DMA_DATA_TYPE_S8 0x00 #define OMAP_DMA_DATA_TYPE_S16 0x01 @@ -194,6 +318,7 @@ enum { OMAP_LCD_DMA_B2_BOTTOM }; +/* REVISIT: Check if BURST_4 is really 1 (or 2) */ enum omap_dma_burst_mode { OMAP_DMA_DATA_BURST_DIS = 0, OMAP_DMA_DATA_BURST_4, @@ -206,6 +331,31 @@ enum omap_dma_color_mode { OMAP_DMA_TRANSPARENT_COPY }; +struct omap_dma_channel_params { + int data_type; /* data type 8,16,32 */ + int elem_count; /* number of elements in a frame */ + int frame_count; /* number of frames in a element */ + + int src_port; /* Only on OMAP1 REVISIT: Is this needed? */ + int src_amode; /* constant , post increment, indexed , double indexed */ + int src_start; /* source address : physical */ + int src_ei; /* source element index */ + int src_fi; /* source frame index */ + + int dst_port; /* Only on OMAP1 REVISIT: Is this needed? */ + int dst_amode; /* constant , post increment, indexed , double indexed */ + int dst_start; /* source address : physical */ + int dst_ei; /* source element index */ + int dst_fi; /* source frame index */ + + int trigger; /* trigger attached if the channel is synchronized */ + int sync_mode; /* sycn on element, frame , block or packet */ + int src_or_dst_synch; /* source synch(1) or destination synch(0) */ + + int ie; /* interrupt enabled */ +}; + + extern void omap_set_dma_priority(int dst_port, int priority); extern int omap_request_dma(int dev_id, const char *dev_name, void (* callback)(int lch, u16 ch_status, void *data), @@ -217,24 +367,30 @@ extern void omap_start_dma(int lch); extern void omap_stop_dma(int lch); extern void omap_set_dma_transfer_params(int lch, int data_type, int elem_count, int frame_count, - int sync_mode); + int sync_mode, + int dma_trigger, int src_or_dst_synch); extern void omap_set_dma_color_mode(int lch, enum omap_dma_color_mode mode, u32 color); extern void omap_set_dma_src_params(int lch, int src_port, int src_amode, - unsigned long src_start); + unsigned long src_start, + int src_ei, int src_fi); extern void omap_set_dma_src_index(int lch, int eidx, int fidx); extern void omap_set_dma_src_data_pack(int lch, int enable); extern void omap_set_dma_src_burst_mode(int lch, enum omap_dma_burst_mode burst_mode); extern void omap_set_dma_dest_params(int lch, int dest_port, int dest_amode, - unsigned long dest_start); + unsigned long dest_start, + int dst_ei, int dst_fi); extern void omap_set_dma_dest_index(int lch, int eidx, int fidx); extern void omap_set_dma_dest_data_pack(int lch, int enable); extern void omap_set_dma_dest_burst_mode(int lch, enum omap_dma_burst_mode burst_mode); +extern void omap_set_dma_params(int lch, + struct omap_dma_channel_params * params); + extern void omap_dma_link_lch (int lch_head, int lch_queue); extern void omap_dma_unlink_lch (int lch_head, int lch_queue); @@ -244,9 +400,6 @@ extern int omap_get_dma_src_addr_counter(int lch); extern void omap_clear_dma(int lch); extern int omap_dma_running(void); -/* Returns 1 if the DMA module is in OMAP1510-compatible mode, 0 otherwise */ -extern int omap_dma_in_1510_mode(void); - /* LCD DMA functions */ extern int omap_request_lcd_dma(void (* callback)(u16 status, void *data), void *data); diff --git a/include/asm-arm/arch-omap/entry-macro.S b/include/asm-arm/arch-omap/entry-macro.S index 0d29b9c56a9..f8814a84910 100644 --- a/include/asm-arm/arch-omap/entry-macro.S +++ b/include/asm-arm/arch-omap/entry-macro.S @@ -10,6 +10,20 @@ #if defined(CONFIG_ARCH_OMAP1) +#if defined(CONFIG_ARCH_OMAP730) && \ + (defined(CONFIG_ARCH_OMAP15XX) || defined(CONFIG_ARCH_OMAP16XX)) +#error "FIXME: OMAP730 doesn't support multiple-OMAP" +#elif defined(CONFIG_ARCH_OMAP730) +#define INT_IH2_IRQ INT_730_IH2_IRQ +#elif defined(CONFIG_ARCH_OMAP15XX) +#define INT_IH2_IRQ INT_1510_IH2_IRQ +#elif defined(CONFIG_ARCH_OMAP16XX) +#define INT_IH2_IRQ INT_1610_IH2_IRQ +#else +#warning "IH2 IRQ defaulted" +#define INT_IH2_IRQ INT_1510_IH2_IRQ +#endif + .macro disable_fiq .endm diff --git a/include/asm-arm/arch-omap/fpga.h b/include/asm-arm/arch-omap/fpga.h index 676807dc50e..6a883e0bdbb 100644 --- a/include/asm-arm/arch-omap/fpga.h +++ b/include/asm-arm/arch-omap/fpga.h @@ -19,7 +19,7 @@ #ifndef __ASM_ARCH_OMAP_FPGA_H #define __ASM_ARCH_OMAP_FPGA_H -#if defined(CONFIG_MACH_OMAP_INNOVATOR) && defined(CONFIG_ARCH_OMAP1510) +#if defined(CONFIG_MACH_OMAP_INNOVATOR) && defined(CONFIG_ARCH_OMAP15XX) extern void omap1510_fpga_init_irq(void); #else #define omap1510_fpga_init_irq() (0) @@ -77,6 +77,8 @@ struct h2p2_dbg_fpga { #define H2P2_DBG_FPGA_LOAD_METER_SIZE 11 #define H2P2_DBG_FPGA_LOAD_METER_MASK ((1 << H2P2_DBG_FPGA_LOAD_METER_SIZE) - 1) +#define H2P2_DBG_FPGA_P2_LED_TIMER (1 << 0) +#define H2P2_DBG_FPGA_P2_LED_IDLE (1 << 1) /* * --------------------------------------------------------------------------- diff --git a/include/asm-arm/arch-omap/gpio.h b/include/asm-arm/arch-omap/gpio.h index 74cb2b93b70..1b3885741ac 100644 --- a/include/asm-arm/arch-omap/gpio.h +++ b/include/asm-arm/arch-omap/gpio.h @@ -67,7 +67,7 @@ #define OMAP_GPIO_IRQ(nr) (OMAP_GPIO_IS_MPUIO(nr) ? \ IH_MPUIO_BASE + ((nr) & 0x0f) : \ - IH_GPIO_BASE + ((nr) & 0x3f)) + IH_GPIO_BASE + (nr)) extern int omap_gpio_init(void); /* Call from board init only */ extern int omap_request_gpio(int gpio); diff --git a/include/asm-arm/arch-omap/hardware.h b/include/asm-arm/arch-omap/hardware.h index 60201e1dd6a..5406b875c42 100644 --- a/include/asm-arm/arch-omap/hardware.h +++ b/include/asm-arm/arch-omap/hardware.h @@ -267,8 +267,6 @@ #define OMAP_LPG2_LCR (OMAP_LPG2_BASE + 0x00) #define OMAP_LPG2_PMR (OMAP_LPG2_BASE + 0x04) -#ifndef __ASSEMBLER__ - /* * --------------------------------------------------------------------------- * Processor specific defines @@ -277,13 +275,11 @@ #include "omap730.h" #include "omap1510.h" - -#ifdef CONFIG_ARCH_OMAP24XX #include "omap24xx.h" -#endif - #include "omap16xx.h" +#ifndef __ASSEMBLER__ + /* * --------------------------------------------------------------------------- * Board specific defines diff --git a/include/asm-arm/arch-omap/io.h b/include/asm-arm/arch-omap/io.h index 3d5bcd54508..f5bcc9a1aed 100644 --- a/include/asm-arm/arch-omap/io.h +++ b/include/asm-arm/arch-omap/io.h @@ -52,23 +52,33 @@ * ---------------------------------------------------------------------------- */ +#define PCIO_BASE 0 + #if defined(CONFIG_ARCH_OMAP1) + #define IO_PHYS 0xFFFB0000 -#define IO_OFFSET -0x01000000 /* Virtual IO = 0xfefb0000 */ +#define IO_OFFSET 0x01000000 /* Virtual IO = 0xfefb0000 */ #define IO_SIZE 0x40000 +#define IO_VIRT (IO_PHYS - IO_OFFSET) +#define IO_ADDRESS(pa) ((pa) - IO_OFFSET) +#define io_p2v(pa) ((pa) - IO_OFFSET) +#define io_v2p(va) ((va) + IO_OFFSET) #elif defined(CONFIG_ARCH_OMAP2) -#define IO_PHYS 0x48000000 /* L4 peripherals; other stuff has to be mapped * - * manually. */ -#define IO_OFFSET 0x90000000 /* Virtual IO = 0xd8000000 */ -#define IO_SIZE 0x08000000 -#endif -#define IO_VIRT (IO_PHYS + IO_OFFSET) -#define IO_ADDRESS(x) ((x) + IO_OFFSET) -#define PCIO_BASE 0 -#define io_p2v(x) ((x) + IO_OFFSET) -#define io_v2p(x) ((x) - IO_OFFSET) +/* We map both L3 and L4 on OMAP2 */ +#define L3_24XX_PHYS L3_24XX_BASE /* 0x68000000 */ +#define L3_24XX_VIRT 0xf8000000 +#define L3_24XX_SIZE SZ_1M /* 44kB of 128MB used, want 1MB sect */ +#define L4_24XX_PHYS L4_24XX_BASE /* 0x48000000 */ +#define L4_24XX_VIRT 0xd8000000 +#define L4_24XX_SIZE SZ_1M /* 1MB of 128MB used, want 1MB sect */ +#define IO_OFFSET 0x90000000 +#define IO_ADDRESS(pa) ((pa) + IO_OFFSET) /* Works for L3 and L4 */ +#define io_p2v(pa) ((pa) + IO_OFFSET) /* Works for L3 and L4 */ +#define io_v2p(va) ((va) - IO_OFFSET) /* Works for L3 and L4 */ + +#endif #ifndef __ASSEMBLER__ diff --git a/include/asm-arm/arch-omap/irqs.h b/include/asm-arm/arch-omap/irqs.h index 74e108ccac1..9779686bdce 100644 --- a/include/asm-arm/arch-omap/irqs.h +++ b/include/asm-arm/arch-omap/irqs.h @@ -22,8 +22,8 @@ * are different. */ -#ifndef __ASM_ARCH_OMAP1510_IRQS_H -#define __ASM_ARCH_OMAP1510_IRQS_H +#ifndef __ASM_ARCH_OMAP15XX_IRQS_H +#define __ASM_ARCH_OMAP15XX_IRQS_H /* * IRQ numbers for interrupt handler 1 @@ -31,7 +31,6 @@ * NOTE: See also the OMAP-1510 and 1610 specific IRQ numbers below * */ -#define INT_IH2_IRQ 0 #define INT_CAMERA 1 #define INT_FIQ 3 #define INT_RTDX 6 @@ -60,6 +59,7 @@ /* * OMAP-1510 specific IRQ numbers for interrupt handler 1 */ +#define INT_1510_IH2_IRQ 0 #define INT_1510_RES2 2 #define INT_1510_SPI_TX 4 #define INT_1510_SPI_RX 5 @@ -71,6 +71,7 @@ /* * OMAP-1610 specific IRQ numbers for interrupt handler 1 */ +#define INT_1610_IH2_IRQ 0 #define INT_1610_IH2_FIQ 2 #define INT_1610_McBSP2_TX 4 #define INT_1610_McBSP2_RX 5 @@ -231,6 +232,12 @@ #define INT_730_DMA_CH15 (62 + IH2_BASE) #define INT_730_NAND (63 + IH2_BASE) +#define INT_24XX_SYS_NIRQ 7 +#define INT_24XX_SDMA_IRQ0 12 +#define INT_24XX_SDMA_IRQ1 13 +#define INT_24XX_SDMA_IRQ2 14 +#define INT_24XX_SDMA_IRQ3 15 +#define INT_24XX_DSS_IRQ 25 #define INT_24XX_GPIO_BANK1 29 #define INT_24XX_GPIO_BANK2 30 #define INT_24XX_GPIO_BANK3 31 diff --git a/include/asm-arm/arch-omap/memory.h b/include/asm-arm/arch-omap/memory.h index bf545b6e0a2..df50dd53e1d 100644 --- a/include/asm-arm/arch-omap/memory.h +++ b/include/asm-arm/arch-omap/memory.h @@ -61,7 +61,7 @@ * Note that the is_lbus_device() test is not very efficient on 1510 * because of the strncmp(). */ -#ifdef CONFIG_ARCH_OMAP1510 +#ifdef CONFIG_ARCH_OMAP15XX /* * OMAP-1510 Local Bus address offset @@ -84,7 +84,7 @@ virt_to_lbus(addr) : \ __virt_to_bus(addr);}) -#endif /* CONFIG_ARCH_OMAP1510 */ +#endif /* CONFIG_ARCH_OMAP15XX */ #endif diff --git a/include/asm-arm/arch-omap/menelaus.h b/include/asm-arm/arch-omap/menelaus.h new file mode 100644 index 00000000000..46be8b8d634 --- /dev/null +++ b/include/asm-arm/arch-omap/menelaus.h @@ -0,0 +1,22 @@ +/* + * linux/include/asm-arm/arch-omap/menelaus.h + * + * Functions to access Menelaus power management chip + */ + +#ifndef __ASM_ARCH_MENELAUS_H +#define __ASM_ARCH_MENELAUS_H + +extern void menelaus_mmc_register(void (*callback)(u8 card_mask), + unsigned long data); +extern void menelaus_mmc_remove(void); +extern void menelaus_mmc_opendrain(int enable); + +#if defined(CONFIG_ARCH_OMAP24XX) && defined(CONFIG_MENELAUS) +#define omap_has_menelaus() 1 +#else +#define omap_has_menelaus() 0 +#endif + +#endif + diff --git a/include/asm-arm/arch-omap/mux.h b/include/asm-arm/arch-omap/mux.h index 1b1ad410534..13415a9aab0 100644 --- a/include/asm-arm/arch-omap/mux.h +++ b/include/asm-arm/arch-omap/mux.h @@ -4,7 +4,7 @@ * Table of the Omap register configurations for the FUNC_MUX and * PULL_DWN combinations. * - * Copyright (C) 2003 Nokia Corporation + * Copyright (C) 2003 - 2005 Nokia Corporation * * Written by Tony Lindgren <tony.lindgren@nokia.com> * @@ -58,6 +58,16 @@ .pu_pd_reg = PU_PD_SEL_##reg, \ .pu_pd_val = status, +#define MUX_REG_730(reg, mode_offset, mode) .mux_reg_name = "OMAP730_IO_CONF_"#reg, \ + .mux_reg = OMAP730_IO_CONF_##reg, \ + .mask_offset = mode_offset, \ + .mask = mode, + +#define PULL_REG_730(reg, bit, status) .pull_name = "OMAP730_IO_CONF_"#reg, \ + .pull_reg = OMAP730_IO_CONF_##reg, \ + .pull_bit = bit, \ + .pull_val = status, + #else #define MUX_REG(reg, mode_offset, mode) .mux_reg = FUNC_MUX_CTRL_##reg, \ @@ -71,6 +81,15 @@ #define PU_PD_REG(reg, status) .pu_pd_reg = PU_PD_SEL_##reg, \ .pu_pd_val = status, +#define MUX_REG_730(reg, mode_offset, mode) \ + .mux_reg = OMAP730_IO_CONF_##reg, \ + .mask_offset = mode_offset, \ + .mask = mode, + +#define PULL_REG_730(reg, bit, status) .pull_reg = OMAP730_IO_CONF_##reg, \ + .pull_bit = bit, \ + .pull_val = status, + #endif /* CONFIG_OMAP_MUX_DEBUG */ #define MUX_CFG(desc, mux_reg, mode_offset, mode, \ @@ -84,13 +103,44 @@ PU_PD_REG(pu_pd_reg, pu_pd_status) \ }, + +/* + * OMAP730 has a slightly different config for the pin mux. + * - config regs are the OMAP730_IO_CONF_x regs (see omap730.h) regs and + * not the FUNC_MUX_CTRL_x regs from hardware.h + * - for pull-up/down, only has one enable bit which is is in the same register + * as mux config + */ +#define MUX_CFG_730(desc, mux_reg, mode_offset, mode, \ + pull_reg, pull_bit, pull_status, \ + pu_pd_reg, pu_pd_status, debug_status)\ +{ \ + .name = desc, \ + .debug = debug_status, \ + MUX_REG_730(mux_reg, mode_offset, mode) \ + PULL_REG_730(mux_reg, pull_bit, pull_status) \ + PU_PD_REG(pu_pd_reg, pu_pd_status) \ +}, + +#define MUX_CFG_24XX(desc, reg_offset, mode, \ + pull_en, pull_mode, dbg) \ +{ \ + .name = desc, \ + .debug = dbg, \ + .mux_reg = reg_offset, \ + .mask = mode, \ + .pull_val = pull_en, \ + .pu_pd_val = pull_mode, \ +}, + + #define PULL_DISABLED 0 #define PULL_ENABLED 1 #define PULL_DOWN 0 #define PULL_UP 1 -typedef struct { +struct pin_config { char *name; unsigned char busy; unsigned char debug; @@ -108,13 +158,23 @@ typedef struct { const char *pu_pd_name; const unsigned int pu_pd_reg; const unsigned char pu_pd_val; -} reg_cfg_set; +}; -/* - * Lookup table for FUNC_MUX and PULL_DWN register combinations for each - * device. See also reg_cfg_table below for the register values. - */ -typedef enum { +enum omap730_index { + /* OMAP 730 keyboard */ + E2_730_KBR0, + J7_730_KBR1, + E1_730_KBR2, + F3_730_KBR3, + D2_730_KBR4, + C2_730_KBC0, + D3_730_KBC1, + E4_730_KBC2, + F4_730_KBC3, + E3_730_KBC4, +}; + +enum omap1xxx_index { /* UART1 (BT_UART_GATING)*/ UART1_TX = 0, UART1_RTS, @@ -331,245 +391,34 @@ typedef enum { V10_1610_CF_IREQ, W10_1610_CF_RESET, W11_1610_CF_CD1, -} reg_cfg_t; +}; -#if defined(__MUX_C__) && defined(CONFIG_OMAP_MUX) +enum omap24xx_index { + /* 24xx I2C */ + M19_24XX_I2C1_SCL, + L15_24XX_I2C1_SDA, + J15_24XX_I2C2_SCL, + H19_24XX_I2C2_SDA, -/* - * Table of various FUNC_MUX and PULL_DWN combinations for each device. - * See also reg_cfg_t above for the lookup table. - */ -static const reg_cfg_set __initdata_or_module -reg_cfg_table[] = { -/* - * description mux mode mux pull pull pull pu_pd pu dbg - * reg offset mode reg bit ena reg - */ -MUX_CFG("UART1_TX", 9, 21, 1, 2, 3, 0, NA, 0, 0) -MUX_CFG("UART1_RTS", 9, 12, 1, 2, 0, 0, NA, 0, 0) - -/* UART2 (COM_UART_GATING), conflicts with USB2 */ -MUX_CFG("UART2_TX", C, 27, 1, 3, 3, 0, NA, 0, 0) -MUX_CFG("UART2_RX", C, 18, 0, 3, 1, 1, NA, 0, 0) -MUX_CFG("UART2_CTS", C, 21, 0, 3, 1, 1, NA, 0, 0) -MUX_CFG("UART2_RTS", C, 24, 1, 3, 2, 0, NA, 0, 0) - -/* UART3 (GIGA_UART_GATING) */ -MUX_CFG("UART3_TX", 6, 0, 1, 0, 30, 0, NA, 0, 0) -MUX_CFG("UART3_RX", 6, 3, 0, 0, 31, 1, NA, 0, 0) -MUX_CFG("UART3_CTS", 5, 12, 2, 0, 24, 0, NA, 0, 0) -MUX_CFG("UART3_RTS", 5, 15, 2, 0, 25, 0, NA, 0, 0) -MUX_CFG("UART3_CLKREQ", 9, 27, 0, 2, 5, 0, NA, 0, 0) -MUX_CFG("UART3_BCLK", A, 0, 0, 2, 6, 0, NA, 0, 0) -MUX_CFG("Y15_1610_UART3_RTS", A, 0, 1, 2, 6, 0, NA, 0, 0) - -/* PWT & PWL, conflicts with UART3 */ -MUX_CFG("PWT", 6, 0, 2, 0, 30, 0, NA, 0, 0) -MUX_CFG("PWL", 6, 3, 1, 0, 31, 1, NA, 0, 0) - -/* USB internal master generic */ -MUX_CFG("R18_USB_VBUS", 7, 9, 2, 1, 11, 0, NA, 0, 1) -MUX_CFG("R18_1510_USB_GPIO0", 7, 9, 0, 1, 11, 1, NA, 0, 1) -/* works around erratum: W4_USB_PUEN and W4_USB_PUDIS are switched! */ -MUX_CFG("W4_USB_PUEN", D, 3, 3, 3, 5, 1, NA, 0, 1) -MUX_CFG("W4_USB_CLKO", D, 3, 1, 3, 5, 0, NA, 0, 1) -MUX_CFG("W4_USB_HIGHZ", D, 3, 4, 3, 5, 0, 3, 0, 1) -MUX_CFG("W4_GPIO58", D, 3, 7, 3, 5, 0, 3, 0, 1) - -/* USB1 master */ -MUX_CFG("USB1_SUSP", 8, 27, 2, 1, 27, 0, NA, 0, 1) -MUX_CFG("USB1_SE0", 9, 0, 2, 1, 28, 0, NA, 0, 1) -MUX_CFG("W13_1610_USB1_SE0", 9, 0, 4, 1, 28, 0, NA, 0, 1) -MUX_CFG("USB1_TXEN", 9, 3, 2, 1, 29, 0, NA, 0, 1) -MUX_CFG("USB1_TXD", 9, 24, 1, 2, 4, 0, NA, 0, 1) -MUX_CFG("USB1_VP", A, 3, 1, 2, 7, 0, NA, 0, 1) -MUX_CFG("USB1_VM", A, 6, 1, 2, 8, 0, NA, 0, 1) -MUX_CFG("USB1_RCV", A, 9, 1, 2, 9, 0, NA, 0, 1) -MUX_CFG("USB1_SPEED", A, 12, 2, 2, 10, 0, NA, 0, 1) -MUX_CFG("R13_1610_USB1_SPEED", A, 12, 5, 2, 10, 0, NA, 0, 1) -MUX_CFG("R13_1710_USB1_SEO", A, 12, 5, 2, 10, 0, NA, 0, 1) - -/* USB2 master */ -MUX_CFG("USB2_SUSP", B, 3, 1, 2, 17, 0, NA, 0, 1) -MUX_CFG("USB2_VP", B, 6, 1, 2, 18, 0, NA, 0, 1) -MUX_CFG("USB2_TXEN", B, 9, 1, 2, 19, 0, NA, 0, 1) -MUX_CFG("USB2_VM", C, 18, 1, 3, 0, 0, NA, 0, 1) -MUX_CFG("USB2_RCV", C, 21, 1, 3, 1, 0, NA, 0, 1) -MUX_CFG("USB2_SE0", C, 24, 2, 3, 2, 0, NA, 0, 1) -MUX_CFG("USB2_TXD", C, 27, 2, 3, 3, 0, NA, 0, 1) - -/* OMAP-1510 GPIO */ -MUX_CFG("R18_1510_GPIO0", 7, 9, 0, 1, 11, 1, 0, 0, 1) -MUX_CFG("R19_1510_GPIO1", 7, 6, 0, 1, 10, 1, 0, 0, 1) -MUX_CFG("M14_1510_GPIO2", 7, 3, 0, 1, 9, 1, 0, 0, 1) - -/* OMAP1610 GPIO */ -MUX_CFG("P18_1610_GPIO3", 7, 0, 0, 1, 8, 0, NA, 0, 1) -MUX_CFG("Y15_1610_GPIO17", A, 0, 7, 2, 6, 0, NA, 0, 1) - -/* OMAP-1710 GPIO */ -MUX_CFG("R18_1710_GPIO0", 7, 9, 0, 1, 11, 1, 1, 1, 1) -MUX_CFG("V2_1710_GPIO10", F, 27, 1, 4, 3, 1, 4, 1, 1) -MUX_CFG("N21_1710_GPIO14", 6, 9, 0, 1, 1, 1, 1, 1, 1) -MUX_CFG("W15_1710_GPIO40", 9, 27, 7, 2, 5, 1, 2, 1, 1) - -/* MPUIO */ -MUX_CFG("MPUIO2", 7, 18, 0, 1, 14, 1, NA, 0, 1) -MUX_CFG("N15_1610_MPUIO2", 7, 18, 0, 1, 14, 1, 1, 0, 1) -MUX_CFG("MPUIO4", 7, 15, 0, 1, 13, 1, NA, 0, 1) -MUX_CFG("MPUIO5", 7, 12, 0, 1, 12, 1, NA, 0, 1) - -MUX_CFG("T20_1610_MPUIO5", 7, 12, 0, 1, 12, 0, 3, 0, 1) -MUX_CFG("W11_1610_MPUIO6", 10, 15, 2, 3, 8, 0, 3, 0, 1) -MUX_CFG("V10_1610_MPUIO7", A, 24, 2, 2, 14, 0, 2, 0, 1) -MUX_CFG("W11_1610_MPUIO9", 10, 15, 1, 3, 8, 0, 3, 0, 1) -MUX_CFG("V10_1610_MPUIO10", A, 24, 1, 2, 14, 0, 2, 0, 1) -MUX_CFG("W10_1610_MPUIO11", A, 18, 2, 2, 11, 0, 2, 0, 1) -MUX_CFG("E20_1610_MPUIO13", 3, 21, 1, 0, 7, 0, 0, 0, 1) -MUX_CFG("U20_1610_MPUIO14", 9, 6, 6, 0, 30, 0, 0, 0, 1) -MUX_CFG("E19_1610_MPUIO15", 3, 18, 1, 0, 6, 0, 0, 0, 1) - -/* MCBSP2 */ -MUX_CFG("MCBSP2_CLKR", C, 6, 0, 2, 27, 1, NA, 0, 1) -MUX_CFG("MCBSP2_CLKX", C, 9, 0, 2, 29, 1, NA, 0, 1) -MUX_CFG("MCBSP2_DR", C, 0, 0, 2, 26, 1, NA, 0, 1) -MUX_CFG("MCBSP2_DX", C, 15, 0, 2, 31, 1, NA, 0, 1) -MUX_CFG("MCBSP2_FSR", C, 12, 0, 2, 30, 1, NA, 0, 1) -MUX_CFG("MCBSP2_FSX", C, 3, 0, 2, 27, 1, NA, 0, 1) - -/* MCBSP3 NOTE: Mode must 1 for clock */ -MUX_CFG("MCBSP3_CLKX", 9, 3, 1, 1, 29, 0, NA, 0, 1) - -/* Misc ballouts */ -MUX_CFG("BALLOUT_V8_ARMIO3", B, 18, 0, 2, 25, 1, NA, 0, 1) -MUX_CFG("N20_HDQ", 6, 18, 1, 1, 4, 0, 1, 4, 0) - -/* OMAP-1610 MMC2 */ -MUX_CFG("W8_1610_MMC2_DAT0", B, 21, 6, 2, 23, 1, 2, 1, 1) -MUX_CFG("V8_1610_MMC2_DAT1", B, 27, 6, 2, 25, 1, 2, 1, 1) -MUX_CFG("W15_1610_MMC2_DAT2", 9, 12, 6, 2, 5, 1, 2, 1, 1) -MUX_CFG("R10_1610_MMC2_DAT3", B, 18, 6, 2, 22, 1, 2, 1, 1) -MUX_CFG("Y10_1610_MMC2_CLK", B, 3, 6, 2, 17, 0, 2, 0, 1) -MUX_CFG("Y8_1610_MMC2_CMD", B, 24, 6, 2, 24, 1, 2, 1, 1) -MUX_CFG("V9_1610_MMC2_CMDDIR", B, 12, 6, 2, 20, 0, 2, 1, 1) -MUX_CFG("V5_1610_MMC2_DATDIR0", B, 15, 6, 2, 21, 0, 2, 1, 1) -MUX_CFG("W19_1610_MMC2_DATDIR1", 8, 15, 6, 1, 23, 0, 1, 1, 1) -MUX_CFG("R18_1610_MMC2_CLKIN", 7, 9, 6, 1, 11, 0, 1, 11, 1) - -/* OMAP-1610 External Trace Interface */ -MUX_CFG("M19_1610_ETM_PSTAT0", 5, 27, 1, 0, 29, 0, 0, 0, 1) -MUX_CFG("L15_1610_ETM_PSTAT1", 5, 24, 1, 0, 28, 0, 0, 0, 1) -MUX_CFG("L18_1610_ETM_PSTAT2", 5, 21, 1, 0, 27, 0, 0, 0, 1) -MUX_CFG("L19_1610_ETM_D0", 5, 18, 1, 0, 26, 0, 0, 0, 1) -MUX_CFG("J19_1610_ETM_D6", 5, 0, 1, 0, 20, 0, 0, 0, 1) -MUX_CFG("J18_1610_ETM_D7", 5, 27, 1, 0, 19, 0, 0, 0, 1) - -/* OMAP16XX GPIO */ -MUX_CFG("P20_1610_GPIO4", 6, 27, 0, 1, 7, 0, 1, 1, 1) -MUX_CFG("V9_1610_GPIO7", B, 12, 1, 2, 20, 0, 2, 1, 1) -MUX_CFG("W8_1610_GPIO9", B, 21, 0, 2, 23, 0, 2, 1, 1) -MUX_CFG("N20_1610_GPIO11", 6, 18, 0, 1, 4, 0, 1, 1, 1) -MUX_CFG("N19_1610_GPIO13", 6, 12, 0, 1, 2, 0, 1, 1, 1) -MUX_CFG("P10_1610_GPIO22", C, 0, 7, 2, 26, 0, 2, 1, 1) -MUX_CFG("V5_1610_GPIO24", B, 15, 7, 2, 21, 0, 2, 1, 1) -MUX_CFG("AA20_1610_GPIO_41", 9, 9, 7, 1, 31, 0, 1, 1, 1) -MUX_CFG("W19_1610_GPIO48", 8, 15, 7, 1, 23, 1, 1, 0, 1) -MUX_CFG("M7_1610_GPIO62", 10, 0, 0, 4, 24, 0, 4, 0, 1) -MUX_CFG("V14_16XX_GPIO37", 9, 18, 7, 2, 2, 0, 2, 2, 0) -MUX_CFG("R9_16XX_GPIO18", C, 18, 7, 3, 0, 0, 3, 0, 0) -MUX_CFG("L14_16XX_GPIO49", 6, 3, 7, 0, 31, 0, 0, 31, 0) - -/* OMAP-1610 uWire */ -MUX_CFG("V19_1610_UWIRE_SCLK", 8, 6, 0, 1, 20, 0, 1, 1, 1) -MUX_CFG("U18_1610_UWIRE_SDI", 8, 0, 0, 1, 18, 0, 1, 1, 1) -MUX_CFG("W21_1610_UWIRE_SDO", 8, 3, 0, 1, 19, 0, 1, 1, 1) -MUX_CFG("N14_1610_UWIRE_CS0", 8, 9, 1, 1, 21, 0, 1, 1, 1) -MUX_CFG("P15_1610_UWIRE_CS3", 8, 12, 1, 1, 22, 0, 1, 1, 1) -MUX_CFG("N15_1610_UWIRE_CS1", 7, 18, 2, 1, 14, 0, NA, 0, 1) - -/* OMAP-1610 Flash */ -MUX_CFG("L3_1610_FLASH_CS2B_OE",10, 6, 1, NA, 0, 0, NA, 0, 1) -MUX_CFG("M8_1610_FLASH_CS2B_WE",10, 3, 1, NA, 0, 0, NA, 0, 1) - -/* First MMC interface, same on 1510, 1610 and 1710 */ -MUX_CFG("MMC_CMD", A, 27, 0, 2, 15, 1, 2, 1, 1) -MUX_CFG("MMC_DAT1", A, 24, 0, 2, 14, 1, 2, 1, 1) -MUX_CFG("MMC_DAT2", A, 18, 0, 2, 12, 1, 2, 1, 1) -MUX_CFG("MMC_DAT0", B, 0, 0, 2, 16, 1, 2, 1, 1) -MUX_CFG("MMC_CLK", A, 21, 0, NA, 0, 0, NA, 0, 1) -MUX_CFG("MMC_DAT3", 10, 15, 0, 3, 8, 1, 3, 1, 1) -MUX_CFG("M15_1710_MMC_CLKI", 6, 21, 2, 0, 0, 0, NA, 0, 1) -MUX_CFG("P19_1710_MMC_CMDDIR", 6, 24, 6, 0, 0, 0, NA, 0, 1) -MUX_CFG("P20_1710_MMC_DATDIR0", 6, 27, 5, 0, 0, 0, NA, 0, 1) - -/* OMAP-1610 USB0 alternate configuration */ -MUX_CFG("W9_USB0_TXEN", B, 9, 5, 2, 19, 0, 2, 0, 1) -MUX_CFG("AA9_USB0_VP", B, 6, 5, 2, 18, 0, 2, 0, 1) -MUX_CFG("Y5_USB0_RCV", C, 21, 5, 3, 1, 0, 1, 0, 1) -MUX_CFG("R9_USB0_VM", C, 18, 5, 3, 0, 0, 3, 0, 1) -MUX_CFG("V6_USB0_TXD", C, 27, 5, 3, 3, 0, 3, 0, 1) -MUX_CFG("W5_USB0_SE0", C, 24, 5, 3, 2, 0, 3, 0, 1) -MUX_CFG("V9_USB0_SPEED", B, 12, 5, 2, 20, 0, 2, 0, 1) -MUX_CFG("Y10_USB0_SUSP", B, 3, 5, 2, 17, 0, 2, 0, 1) - -/* USB2 interface */ -MUX_CFG("W9_USB2_TXEN", B, 9, 1, NA, 0, 0, NA, 0, 1) -MUX_CFG("AA9_USB2_VP", B, 6, 1, NA, 0, 0, NA, 0, 1) -MUX_CFG("Y5_USB2_RCV", C, 21, 1, NA, 0, 0, NA, 0, 1) -MUX_CFG("R9_USB2_VM", C, 18, 1, NA, 0, 0, NA, 0, 1) -MUX_CFG("V6_USB2_TXD", C, 27, 2, NA, 0, 0, NA, 0, 1) -MUX_CFG("W5_USB2_SE0", C, 24, 2, NA, 0, 0, NA, 0, 1) - -/* 16XX UART */ -MUX_CFG("R13_1610_UART1_TX", A, 12, 6, 2, 10, 0, 2, 10, 1) -MUX_CFG("V14_16XX_UART1_RX", 9, 18, 0, 2, 2, 0, 2, 2, 1) -MUX_CFG("R14_1610_UART1_CTS", 9, 15, 0, 2, 1, 0, 2, 1, 1) -MUX_CFG("AA15_1610_UART1_RTS", 9, 12, 1, 2, 0, 0, 2, 0, 1) -MUX_CFG("R9_16XX_UART2_RX", C, 18, 0, 3, 0, 0, 3, 0, 1) -MUX_CFG("L14_16XX_UART3_RX", 6, 3, 0, 0, 31, 0, 0, 31, 1) - -/* I2C interface */ -MUX_CFG("I2C_SCL", 7, 24, 0, NA, 0, 0, NA, 0, 0) -MUX_CFG("I2C_SDA", 7, 27, 0, NA, 0, 0, NA, 0, 0) - -/* Keypad */ -MUX_CFG("F18_1610_KBC0", 3, 15, 0, 0, 5, 1, 0, 0, 0) -MUX_CFG("D20_1610_KBC1", 3, 12, 0, 0, 4, 1, 0, 0, 0) -MUX_CFG("D19_1610_KBC2", 3, 9, 0, 0, 3, 1, 0, 0, 0) -MUX_CFG("E18_1610_KBC3", 3, 6, 0, 0, 2, 1, 0, 0, 0) -MUX_CFG("C21_1610_KBC4", 3, 3, 0, 0, 1, 1, 0, 0, 0) -MUX_CFG("G18_1610_KBR0", 4, 0, 0, 0, 10, 1, 0, 1, 0) -MUX_CFG("F19_1610_KBR1", 3, 27, 0, 0, 9, 1, 0, 1, 0) -MUX_CFG("H14_1610_KBR2", 3, 24, 0, 0, 8, 1, 0, 1, 0) -MUX_CFG("E20_1610_KBR3", 3, 21, 0, 0, 7, 1, 0, 1, 0) -MUX_CFG("E19_1610_KBR4", 3, 18, 0, 0, 6, 1, 0, 1, 0) -MUX_CFG("N19_1610_KBR5", 6, 12, 1, 1, 2, 1, 1, 1, 0) - -/* Power management */ -MUX_CFG("T20_1610_LOW_PWR", 7, 12, 1, NA, 0, 0, NA, 0, 0) - -/* MCLK Settings */ -MUX_CFG("V5_1710_MCLK_ON", B, 15, 0, NA, 0, 0, NA, 0, 0) -MUX_CFG("V5_1710_MCLK_OFF", B, 15, 6, NA, 0, 0, NA, 0, 0) -MUX_CFG("R10_1610_MCLK_ON", B, 18, 0, NA, 22, 0, NA, 1, 0) -MUX_CFG("R10_1610_MCLK_OFF", B, 18, 6, 2, 22, 1, 2, 1, 1) - -/* CompactFlash controller, conflicts with MMC1 */ -MUX_CFG("P11_1610_CF_CD2", A, 27, 3, 2, 15, 1, 2, 1, 1) -MUX_CFG("R11_1610_CF_IOIS16", B, 0, 3, 2, 16, 1, 2, 1, 1) -MUX_CFG("V10_1610_CF_IREQ", A, 24, 3, 2, 14, 0, 2, 0, 1) -MUX_CFG("W10_1610_CF_RESET", A, 18, 3, 2, 12, 1, 2, 1, 1) -MUX_CFG("W11_1610_CF_CD1", 10, 15, 3, 3, 8, 1, 3, 1, 1) -}; + /* 24xx Menelaus interrupt */ + W19_24XX_SYS_NIRQ, -#endif /* __MUX_C__ */ + /* 24xx GPIO */ + Y20_24XX_GPIO60, + M15_24XX_GPIO92, +}; #ifdef CONFIG_OMAP_MUX /* setup pin muxing in Linux */ -extern int omap_cfg_reg(reg_cfg_t reg_cfg); +extern int omap1_mux_init(void); +extern int omap2_mux_init(void); +extern int omap_mux_register(struct pin_config * pins, unsigned long size); +extern int omap_cfg_reg(unsigned long reg_cfg); #else /* boot loader does it all (no warnings from CONFIG_OMAP_MUX_WARNINGS) */ -static inline int omap_cfg_reg(reg_cfg_t reg_cfg) { return 0; } +static inline int omap1_mux_init(void) { return 0; } +static inline int omap2_mux_init(void) { return 0; } +static inline int omap_cfg_reg(unsigned long reg_cfg) { return 0; } #endif #endif diff --git a/include/asm-arm/arch-omap/omap1510.h b/include/asm-arm/arch-omap/omap1510.h index f086a393390..c575d354850 100644 --- a/include/asm-arm/arch-omap/omap1510.h +++ b/include/asm-arm/arch-omap/omap1510.h @@ -25,8 +25,8 @@ * 675 Mass Ave, Cambridge, MA 02139, USA. */ -#ifndef __ASM_ARCH_OMAP1510_H -#define __ASM_ARCH_OMAP1510_H +#ifndef __ASM_ARCH_OMAP15XX_H +#define __ASM_ARCH_OMAP15XX_H /* * ---------------------------------------------------------------------------- @@ -44,5 +44,5 @@ #define OMAP1510_DSPREG_SIZE SZ_128K #define OMAP1510_DSPREG_START 0xE1000000 -#endif /* __ASM_ARCH_OMAP1510_H */ +#endif /* __ASM_ARCH_OMAP15XX_H */ diff --git a/include/asm-arm/arch-omap/omap24xx.h b/include/asm-arm/arch-omap/omap24xx.h index a9105466a41..6e59805fa65 100644 --- a/include/asm-arm/arch-omap/omap24xx.h +++ b/include/asm-arm/arch-omap/omap24xx.h @@ -1,15 +1,24 @@ #ifndef __ASM_ARCH_OMAP24XX_H #define __ASM_ARCH_OMAP24XX_H -#define OMAP24XX_L4_IO_BASE 0x48000000 +/* + * Please place only base defines here and put the rest in device + * specific headers. Note also that some of these defines are needed + * for omap1 to compile without adding ifdefs. + */ + +#define L4_24XX_BASE 0x48000000 +#define L3_24XX_BASE 0x68000000 /* interrupt controller */ -#define OMAP24XX_IC_BASE (OMAP24XX_L4_IO_BASE + 0xfe000) +#define OMAP24XX_IC_BASE (L4_24XX_BASE + 0xfe000) #define VA_IC_BASE IO_ADDRESS(OMAP24XX_IC_BASE) - #define OMAP24XX_IVA_INTC_BASE 0x40000000 - #define IRQ_SIR_IRQ 0x0040 +#define OMAP24XX_32KSYNCT_BASE (L4_24XX_BASE + 0x4000) +#define OMAP24XX_PRCM_BASE (L4_24XX_BASE + 0x8000) +#define OMAP24XX_SDRC_BASE (L3_24XX_BASE + 0x9000) + #endif /* __ASM_ARCH_OMAP24XX_H */ diff --git a/include/asm-arm/arch-omap/omapfb.h b/include/asm-arm/arch-omap/omapfb.h new file mode 100644 index 00000000000..4ba2622cc14 --- /dev/null +++ b/include/asm-arm/arch-omap/omapfb.h @@ -0,0 +1,281 @@ +/* + * File: include/asm-arm/arch-omap/omapfb.h + * + * Framebuffer driver for TI OMAP boards + * + * Copyright (C) 2004 Nokia Corporation + * Author: Imre Deak <imre.deak@nokia.com> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License along + * with this program; if not, write to the Free Software Foundation, Inc., + * 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + */ + +#ifndef __OMAPFB_H +#define __OMAPFB_H + +/* IOCTL commands. */ + +#define OMAP_IOW(num, dtype) _IOW('O', num, dtype) +#define OMAP_IOR(num, dtype) _IOR('O', num, dtype) +#define OMAP_IOWR(num, dtype) _IOWR('O', num, dtype) +#define OMAP_IO(num) _IO('O', num) + +#define OMAPFB_MIRROR OMAP_IOW(31, int) +#define OMAPFB_SYNC_GFX OMAP_IO(37) +#define OMAPFB_VSYNC OMAP_IO(38) +#define OMAPFB_SET_UPDATE_MODE OMAP_IOW(40, enum omapfb_update_mode) +#define OMAPFB_GET_CAPS OMAP_IOR(42, unsigned long) +#define OMAPFB_GET_UPDATE_MODE OMAP_IOW(43, enum omapfb_update_mode) +#define OMAPFB_LCD_TEST OMAP_IOW(45, int) +#define OMAPFB_CTRL_TEST OMAP_IOW(46, int) +#define OMAPFB_UPDATE_WINDOW OMAP_IOW(47, struct omapfb_update_window) +#define OMAPFB_SETUP_PLANE OMAP_IOW(48, struct omapfb_setup_plane) +#define OMAPFB_ENABLE_PLANE OMAP_IOW(49, struct omapfb_enable_plane) +#define OMAPFB_SET_COLOR_KEY OMAP_IOW(50, struct omapfb_color_key) + +#define OMAPFB_CAPS_GENERIC_MASK 0x00000fff +#define OMAPFB_CAPS_LCDC_MASK 0x00fff000 +#define OMAPFB_CAPS_PANEL_MASK 0xff000000 + +#define OMAPFB_CAPS_MANUAL_UPDATE 0x00001000 +#define OMAPFB_CAPS_SET_BACKLIGHT 0x01000000 + +/* Values from DSP must map to lower 16-bits */ +#define OMAPFB_FORMAT_MASK 0x00ff +#define OMAPFB_FORMAT_FLAG_DOUBLE 0x0100 + +enum omapfb_color_format { + OMAPFB_COLOR_RGB565 = 0, + OMAPFB_COLOR_YUV422, + OMAPFB_COLOR_YUV420, + OMAPFB_COLOR_CLUT_8BPP, + OMAPFB_COLOR_CLUT_4BPP, + OMAPFB_COLOR_CLUT_2BPP, + OMAPFB_COLOR_CLUT_1BPP, +}; + +struct omapfb_update_window { + u32 x, y; + u32 width, height; + u32 format; +}; + +enum omapfb_plane { + OMAPFB_PLANE_GFX = 0, + OMAPFB_PLANE_VID1, + OMAPFB_PLANE_VID2, +}; + +enum omapfb_channel_out { + OMAPFB_CHANNEL_OUT_LCD = 0, + OMAPFB_CHANNEL_OUT_DIGIT, +}; + +struct omapfb_setup_plane { + u8 plane; + u8 channel_out; + u32 offset; + u32 pos_x, pos_y; + u32 width, height; + u32 color_mode; +}; + +struct omapfb_enable_plane { + u8 plane; + u8 enable; +}; + +enum omapfb_color_key_type { + OMAPFB_COLOR_KEY_DISABLED = 0, + OMAPFB_COLOR_KEY_GFX_DST, + OMAPFB_COLOR_KEY_VID_SRC, +}; + +struct omapfb_color_key { + u8 channel_out; + u32 background; + u32 trans_key; + u8 key_type; +}; + +enum omapfb_update_mode { + OMAPFB_UPDATE_DISABLED = 0, + OMAPFB_AUTO_UPDATE, + OMAPFB_MANUAL_UPDATE +}; + +#ifdef __KERNEL__ + +#include <linux/completion.h> +#include <linux/interrupt.h> +#include <linux/fb.h> + +#define OMAP_LCDC_INV_VSYNC 0x0001 +#define OMAP_LCDC_INV_HSYNC 0x0002 +#define OMAP_LCDC_INV_PIX_CLOCK 0x0004 +#define OMAP_LCDC_INV_OUTPUT_EN 0x0008 +#define OMAP_LCDC_HSVS_RISING_EDGE 0x0010 +#define OMAP_LCDC_HSVS_OPPOSITE 0x0020 + +#define OMAP_LCDC_SIGNAL_MASK 0x003f + +#define OMAP_LCDC_PANEL_TFT 0x0100 + +#ifdef CONFIG_ARCH_OMAP1 +#define OMAPFB_PLANE_NUM 1 +#else +#define OMAPFB_PLANE_NUM 3 +#endif + +struct omapfb_device; + +struct lcd_panel { + const char *name; + int config; /* TFT/STN, signal inversion */ + int bpp; /* Pixel format in fb mem */ + int data_lines; /* Lines on LCD HW interface */ + + int x_res, y_res; + int pixel_clock; /* In kHz */ + int hsw; /* Horizontal synchronization + pulse width */ + int hfp; /* Horizontal front porch */ + int hbp; /* Horizontal back porch */ + int vsw; /* Vertical synchronization + pulse width */ + int vfp; /* Vertical front porch */ + int vbp; /* Vertical back porch */ + int acb; /* ac-bias pin frequency */ + int pcd; /* pixel clock divider. + Obsolete use pixel_clock instead */ + + int (*init) (struct omapfb_device *fbdev); + void (*cleanup) (void); + int (*enable) (void); + void (*disable) (void); + unsigned long (*get_caps) (void); + int (*set_bklight_level)(unsigned int level); + unsigned int (*get_bklight_level)(void); + unsigned int (*get_bklight_max) (void); + int (*run_test) (int test_num); +}; + +struct omapfb_device; + +struct extif_timings { + int cs_on_time; + int cs_off_time; + int we_on_time; + int we_off_time; + int re_on_time; + int re_off_time; + int we_cycle_time; + int re_cycle_time; + int cs_pulse_width; + int access_time; +}; + +struct lcd_ctrl_extif { + int (*init) (void); + void (*cleanup) (void); + void (*set_timings) (const struct extif_timings *timings); + void (*write_command) (u32 cmd); + u32 (*read_data) (void); + void (*write_data) (u32 data); + void (*transfer_area) (int width, int height, + void (callback)(void * data), void *data); +}; + +struct lcd_ctrl { + const char *name; + void *data; + + int (*init) (struct omapfb_device *fbdev, + int ext_mode, int req_vram_size); + void (*cleanup) (void); + void (*get_vram_layout)(unsigned long *size, + void **virt_base, + dma_addr_t *phys_base); + unsigned long (*get_caps) (void); + int (*set_update_mode)(enum omapfb_update_mode mode); + enum omapfb_update_mode (*get_update_mode)(void); + int (*setup_plane) (int plane, int channel_out, + unsigned long offset, + int screen_width, + int pos_x, int pos_y, int width, + int height, int color_mode); + int (*enable_plane) (int plane, int enable); + int (*update_window) (struct omapfb_update_window *win, + void (*callback)(void *), + void *callback_data); + void (*sync) (void); + void (*suspend) (void); + void (*resume) (void); + int (*run_test) (int test_num); + int (*setcolreg) (u_int regno, u16 red, u16 green, + u16 blue, u16 transp, + int update_hw_mem); + int (*set_color_key) (struct omapfb_color_key *ck); + +}; + +enum omapfb_state { + OMAPFB_DISABLED = 0, + OMAPFB_SUSPENDED= 99, + OMAPFB_ACTIVE = 100 +}; + +struct omapfb_device { + int state; + int ext_lcdc; /* Using external + LCD controller */ + struct semaphore rqueue_sema; + + void *vram_virt_base; + dma_addr_t vram_phys_base; + unsigned long vram_size; + + int color_mode; + int palette_size; + int mirror; + u32 pseudo_palette[17]; + + struct lcd_panel *panel; /* LCD panel */ + struct lcd_ctrl *ctrl; /* LCD controller */ + struct lcd_ctrl *int_ctrl; /* internal LCD ctrl */ + struct lcd_ctrl_extif *ext_if; /* LCD ctrl external + interface */ + struct fb_info *fb_info; + + struct device *dev; +}; + +extern struct lcd_panel h3_panel; +extern struct lcd_panel h2_panel; +extern struct lcd_panel p2_panel; +extern struct lcd_panel osk_panel; +extern struct lcd_panel innovator1610_panel; +extern struct lcd_panel innovator1510_panel; + +#ifdef CONFIG_ARCH_OMAP1 +extern struct lcd_ctrl omap1_lcd_ctrl; +#else +extern struct lcd_ctrl omap2_disp_ctrl; +#endif + +extern void omapfb_write_first_pixel(struct omapfb_device *fbdev, u16 pixval); + +#endif /* __KERNEL__ */ + +#endif /* __OMAPFB_H */ diff --git a/include/asm-arm/arch-omap/pm.h b/include/asm-arm/arch-omap/pm.h index fbd742d0c49..7c790425e36 100644 --- a/include/asm-arm/arch-omap/pm.h +++ b/include/asm-arm/arch-omap/pm.h @@ -98,7 +98,14 @@ #define OMAP1610_IDLECT3 0xfffece24 #define OMAP1610_IDLE_LOOP_REQUEST 0x0400 -#if !defined(CONFIG_ARCH_OMAP1510) && \ +#define OMAP730_IDLECT1_SLEEP_VAL 0x16c7 +#define OMAP730_IDLECT2_SLEEP_VAL 0x09c7 +#define OMAP730_IDLECT3_VAL 0x3f +#define OMAP730_IDLECT3 0xfffece24 +#define OMAP730_IDLE_LOOP_REQUEST 0x0C00 + +#if !defined(CONFIG_ARCH_OMAP730) && \ + !defined(CONFIG_ARCH_OMAP15XX) && \ !defined(CONFIG_ARCH_OMAP16XX) && \ !defined(CONFIG_ARCH_OMAP24XX) #error "Power management for this processor not implemented yet" @@ -107,8 +114,10 @@ #ifndef __ASSEMBLER__ extern void omap_pm_idle(void); extern void omap_pm_suspend(void); +extern void omap730_cpu_suspend(unsigned short, unsigned short); extern void omap1510_cpu_suspend(unsigned short, unsigned short); extern void omap1610_cpu_suspend(unsigned short, unsigned short); +extern void omap730_idle_loop_suspend(void); extern void omap1510_idle_loop_suspend(void); extern void omap1610_idle_loop_suspend(void); @@ -118,6 +127,8 @@ extern void omap_serial_wake_trigger(int enable); #define omap_serial_wake_trigger(x) {} #endif /* CONFIG_OMAP_SERIAL_WAKE */ +extern unsigned int omap730_cpu_suspend_sz; +extern unsigned int omap730_idle_loop_suspend_sz; extern unsigned int omap1510_cpu_suspend_sz; extern unsigned int omap1510_idle_loop_suspend_sz; extern unsigned int omap1610_cpu_suspend_sz; @@ -131,6 +142,10 @@ extern unsigned int omap1610_idle_loop_suspend_sz; #define ULPD_RESTORE(x) omap_writew((ulpd_sleep_save[ULPD_SLEEP_SAVE_##x]), (x)) #define ULPD_SHOW(x) ulpd_sleep_save[ULPD_SLEEP_SAVE_##x] +#define MPUI730_SAVE(x) mpui730_sleep_save[MPUI730_SLEEP_SAVE_##x] = omap_readl(x) +#define MPUI730_RESTORE(x) omap_writel((mpui730_sleep_save[MPUI730_SLEEP_SAVE_##x]), (x)) +#define MPUI730_SHOW(x) mpui730_sleep_save[MPUI730_SLEEP_SAVE_##x] + #define MPUI1510_SAVE(x) mpui1510_sleep_save[MPUI1510_SLEEP_SAVE_##x] = omap_readl(x) #define MPUI1510_RESTORE(x) omap_writel((mpui1510_sleep_save[MPUI1510_SLEEP_SAVE_##x]), (x)) #define MPUI1510_SHOW(x) mpui1510_sleep_save[MPUI1510_SLEEP_SAVE_##x] @@ -188,13 +203,34 @@ enum mpui1510_save_state { MPUI1510_SLEEP_SAVE_EMIFS_CONFIG, MPUI1510_SLEEP_SAVE_OMAP_IH1_MIR, MPUI1510_SLEEP_SAVE_OMAP_IH2_MIR, -#if defined(CONFIG_ARCH_OMAP1510) +#if defined(CONFIG_ARCH_OMAP15XX) MPUI1510_SLEEP_SAVE_SIZE #else MPUI1510_SLEEP_SAVE_SIZE = 0 #endif }; +enum mpui730_save_state { + MPUI730_SLEEP_SAVE_START = 0, + /* + * MPUI registers 32 bits + */ + MPUI730_SLEEP_SAVE_MPUI_CTRL, + MPUI730_SLEEP_SAVE_MPUI_DSP_BOOT_CONFIG, + MPUI730_SLEEP_SAVE_MPUI_DSP_API_CONFIG, + MPUI730_SLEEP_SAVE_MPUI_DSP_STATUS, + MPUI730_SLEEP_SAVE_EMIFF_SDRAM_CONFIG, + MPUI730_SLEEP_SAVE_EMIFS_CONFIG, + MPUI730_SLEEP_SAVE_OMAP_IH1_MIR, + MPUI730_SLEEP_SAVE_OMAP_IH2_0_MIR, + MPUI730_SLEEP_SAVE_OMAP_IH2_1_MIR, +#if defined(CONFIG_ARCH_OMAP730) + MPUI730_SLEEP_SAVE_SIZE +#else + MPUI730_SLEEP_SAVE_SIZE = 0 +#endif +}; + enum mpui1610_save_state { MPUI1610_SLEEP_SAVE_START = 0, /* diff --git a/include/asm-arm/arch-omap/prcm.h b/include/asm-arm/arch-omap/prcm.h new file mode 100644 index 00000000000..7b48a5cbb15 --- /dev/null +++ b/include/asm-arm/arch-omap/prcm.h @@ -0,0 +1,429 @@ +/* + * prcm.h - Access definations for use in OMAP24XX clock and power management + * + * Copyright (C) 2005 Texas Instruments, Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#ifndef __ASM_ARM_ARCH_DPM_PRCM_H +#define __ASM_ARM_ARCH_DPM_PRCM_H + +/* SET_PERFORMANCE_LEVEL PARAMETERS */ +#define PRCM_HALF_SPEED 1 +#define PRCM_FULL_SPEED 2 + +#ifndef __ASSEMBLER__ + +#define PRCM_REG32(offset) __REG32(OMAP24XX_PRCM_BASE + (offset)) + +#define PRCM_REVISION PRCM_REG32(0x000) +#define PRCM_SYSCONFIG PRCM_REG32(0x010) +#define PRCM_IRQSTATUS_MPU PRCM_REG32(0x018) +#define PRCM_IRQENABLE_MPU PRCM_REG32(0x01C) +#define PRCM_VOLTCTRL PRCM_REG32(0x050) +#define PRCM_VOLTST PRCM_REG32(0x054) +#define PRCM_CLKSRC_CTRL PRCM_REG32(0x060) +#define PRCM_CLKOUT_CTRL PRCM_REG32(0x070) +#define PRCM_CLKEMUL_CTRL PRCM_REG32(0x078) +#define PRCM_CLKCFG_CTRL PRCM_REG32(0x080) +#define PRCM_CLKCFG_STATUS PRCM_REG32(0x084) +#define PRCM_VOLTSETUP PRCM_REG32(0x090) +#define PRCM_CLKSSETUP PRCM_REG32(0x094) +#define PRCM_POLCTRL PRCM_REG32(0x098) + +/* GENERAL PURPOSE */ +#define GENERAL_PURPOSE1 PRCM_REG32(0x0B0) +#define GENERAL_PURPOSE2 PRCM_REG32(0x0B4) +#define GENERAL_PURPOSE3 PRCM_REG32(0x0B8) +#define GENERAL_PURPOSE4 PRCM_REG32(0x0BC) +#define GENERAL_PURPOSE5 PRCM_REG32(0x0C0) +#define GENERAL_PURPOSE6 PRCM_REG32(0x0C4) +#define GENERAL_PURPOSE7 PRCM_REG32(0x0C8) +#define GENERAL_PURPOSE8 PRCM_REG32(0x0CC) +#define GENERAL_PURPOSE9 PRCM_REG32(0x0D0) +#define GENERAL_PURPOSE10 PRCM_REG32(0x0D4) +#define GENERAL_PURPOSE11 PRCM_REG32(0x0D8) +#define GENERAL_PURPOSE12 PRCM_REG32(0x0DC) +#define GENERAL_PURPOSE13 PRCM_REG32(0x0E0) +#define GENERAL_PURPOSE14 PRCM_REG32(0x0E4) +#define GENERAL_PURPOSE15 PRCM_REG32(0x0E8) +#define GENERAL_PURPOSE16 PRCM_REG32(0x0EC) +#define GENERAL_PURPOSE17 PRCM_REG32(0x0F0) +#define GENERAL_PURPOSE18 PRCM_REG32(0x0F4) +#define GENERAL_PURPOSE19 PRCM_REG32(0x0F8) +#define GENERAL_PURPOSE20 PRCM_REG32(0x0FC) + +/* MPU */ +#define CM_CLKSEL_MPU PRCM_REG32(0x140) +#define CM_CLKSTCTRL_MPU PRCM_REG32(0x148) +#define RM_RSTST_MPU PRCM_REG32(0x158) +#define PM_WKDEP_MPU PRCM_REG32(0x1C8) +#define PM_EVGENCTRL_MPU PRCM_REG32(0x1D4) +#define PM_EVEGENONTIM_MPU PRCM_REG32(0x1D8) +#define PM_EVEGENOFFTIM_MPU PRCM_REG32(0x1DC) +#define PM_PWSTCTRL_MPU PRCM_REG32(0x1E0) +#define PM_PWSTST_MPU PRCM_REG32(0x1E4) + +/* CORE */ +#define CM_FCLKEN1_CORE PRCM_REG32(0x200) +#define CM_FCLKEN2_CORE PRCM_REG32(0x204) +#define CM_FCLKEN3_CORE PRCM_REG32(0x208) +#define CM_ICLKEN1_CORE PRCM_REG32(0x210) +#define CM_ICLKEN2_CORE PRCM_REG32(0x214) +#define CM_ICLKEN3_CORE PRCM_REG32(0x218) +#define CM_ICLKEN4_CORE PRCM_REG32(0x21C) +#define CM_IDLEST1_CORE PRCM_REG32(0x220) +#define CM_IDLEST2_CORE PRCM_REG32(0x224) +#define CM_IDLEST3_CORE PRCM_REG32(0x228) +#define CM_IDLEST4_CORE PRCM_REG32(0x22C) +#define CM_AUTOIDLE1_CORE PRCM_REG32(0x230) +#define CM_AUTOIDLE2_CORE PRCM_REG32(0x234) +#define CM_AUTOIDLE3_CORE PRCM_REG32(0x238) +#define CM_AUTOIDLE4_CORE PRCM_REG32(0x23C) +#define CM_CLKSEL1_CORE PRCM_REG32(0x240) +#define CM_CLKSEL2_CORE PRCM_REG32(0x244) +#define CM_CLKSTCTRL_CORE PRCM_REG32(0x248) +#define PM_WKEN1_CORE PRCM_REG32(0x2A0) +#define PM_WKEN2_CORE PRCM_REG32(0x2A4) +#define PM_WKST1_CORE PRCM_REG32(0x2B0) +#define PM_WKST2_CORE PRCM_REG32(0x2B4) +#define PM_WKDEP_CORE PRCM_REG32(0x2C8) +#define PM_PWSTCTRL_CORE PRCM_REG32(0x2E0) +#define PM_PWSTST_CORE PRCM_REG32(0x2E4) + +/* GFX */ +#define CM_FCLKEN_GFX PRCM_REG32(0x300) +#define CM_ICLKEN_GFX PRCM_REG32(0x310) +#define CM_IDLEST_GFX PRCM_REG32(0x320) +#define CM_CLKSEL_GFX PRCM_REG32(0x340) +#define CM_CLKSTCTRL_GFX PRCM_REG32(0x348) +#define RM_RSTCTRL_GFX PRCM_REG32(0x350) +#define RM_RSTST_GFX PRCM_REG32(0x358) +#define PM_WKDEP_GFX PRCM_REG32(0x3C8) +#define PM_PWSTCTRL_GFX PRCM_REG32(0x3E0) +#define PM_PWSTST_GFX PRCM_REG32(0x3E4) + +/* WAKE-UP */ +#define CM_FCLKEN_WKUP PRCM_REG32(0x400) +#define CM_ICLKEN_WKUP PRCM_REG32(0x410) +#define CM_IDLEST_WKUP PRCM_REG32(0x420) +#define CM_AUTOIDLE_WKUP PRCM_REG32(0x430) +#define CM_CLKSEL_WKUP PRCM_REG32(0x440) +#define RM_RSTCTRL_WKUP PRCM_REG32(0x450) +#define RM_RSTTIME_WKUP PRCM_REG32(0x454) +#define RM_RSTST_WKUP PRCM_REG32(0x458) +#define PM_WKEN_WKUP PRCM_REG32(0x4A0) +#define PM_WKST_WKUP PRCM_REG32(0x4B0) + +/* CLOCKS */ +#define CM_CLKEN_PLL PRCM_REG32(0x500) +#define CM_IDLEST_CKGEN PRCM_REG32(0x520) +#define CM_AUTOIDLE_PLL PRCM_REG32(0x530) +#define CM_CLKSEL1_PLL PRCM_REG32(0x540) +#define CM_CLKSEL2_PLL PRCM_REG32(0x544) + +/* DSP */ +#define CM_FCLKEN_DSP PRCM_REG32(0x800) +#define CM_ICLKEN_DSP PRCM_REG32(0x810) +#define CM_IDLEST_DSP PRCM_REG32(0x820) +#define CM_AUTOIDLE_DSP PRCM_REG32(0x830) +#define CM_CLKSEL_DSP PRCM_REG32(0x840) +#define CM_CLKSTCTRL_DSP PRCM_REG32(0x848) +#define RM_RSTCTRL_DSP PRCM_REG32(0x850) +#define RM_RSTST_DSP PRCM_REG32(0x858) +#define PM_WKEN_DSP PRCM_REG32(0x8A0) +#define PM_WKDEP_DSP PRCM_REG32(0x8C8) +#define PM_PWSTCTRL_DSP PRCM_REG32(0x8E0) +#define PM_PWSTST_DSP PRCM_REG32(0x8E4) +#define PRCM_IRQSTATUS_DSP PRCM_REG32(0x8F0) +#define PRCM_IRQENABLE_DSP PRCM_REG32(0x8F4) + +/* IVA */ +#define PRCM_IRQSTATUS_IVA PRCM_REG32(0x8F8) +#define PRCM_IRQENABLE_IVA PRCM_REG32(0x8FC) + +/* Modem on 2430 */ +#define CM_FCLKEN_MDM PRCM_REG32(0xC00) +#define CM_ICLKEN_MDM PRCM_REG32(0xC10) +#define CM_IDLEST_MDM PRCM_REG32(0xC20) +#define CM_CLKSEL_MDM PRCM_REG32(0xC40) + +/* FIXME: Move to header for 2430 */ +#define DISP_BASE (OMAP24XX_L4_IO_BASE+0x50000) +#define DISP_REG32(offset) __REG32(DISP_BASE + (offset)) + +#define OMAP24XX_GPMC_BASE (L3_24XX_BASE + 0xa000) +#define GPMC_BASE (OMAP24XX_GPMC_BASE) +#define GPMC_REG32(offset) __REG32(GPMC_BASE + (offset)) + +#define GPT1_BASE (OMAP24XX_GPT1) +#define GPT1_REG32(offset) __REG32(GPT1_BASE + (offset)) + +/* Misc sysconfig */ +#define DISPC_SYSCONFIG DISP_REG32(0x410) +#define SPI_BASE (OMAP24XX_L4_IO_BASE+0x98000) +#define MCSPI1_SYSCONFIG __REG32(SPI_BASE + 0x10) +#define MCSPI2_SYSCONFIG __REG32(SPI_BASE+0x2000 + 0x10) + +//#define DSP_MMU_SYSCONFIG 0x5A000010 +#define CAMERA_MMU_SYSCONFIG __REG32(DISP_BASE+0x2C10) +//#define IVA_MMU_SYSCONFIG 0x5D000010 +//#define DSP_DMA_SYSCONFIG 0x00FCC02C +#define CAMERA_DMA_SYSCONFIG __REG32(DISP_BASE+0x282C) +#define SYSTEM_DMA_SYSCONFIG __REG32(DISP_BASE+0x602C) +#define GPMC_SYSCONFIG GPMC_REG32(0x010) +#define MAILBOXES_SYSCONFIG __REG32(OMAP24XX_L4_IO_BASE+0x94010) +#define UART1_SYSCONFIG __REG32(OMAP24XX_L4_IO_BASE+0x6A054) +#define UART2_SYSCONFIG __REG32(OMAP24XX_L4_IO_BASE+0x6C054) +#define UART3_SYSCONFIG __REG32(OMAP24XX_L4_IO_BASE+0x6E054) +//#define IVA_SYSCONFIG 0x5C060010 +#define SDRC_SYSCONFIG __REG32(OMAP24XX_SDRC_BASE+0x10) +#define SMS_SYSCONFIG __REG32(OMAP24XX_SMS_BASE+0x10) +#define SSI_SYSCONFIG __REG32(DISP_BASE+0x8010) +//#define VLYNQ_SYSCONFIG 0x67FFFE10 + +/* rkw - good cannidates for PM_ to start what nm was trying */ +#define OMAP24XX_GPT2 (OMAP24XX_L4_IO_BASE+0x2A000) +#define OMAP24XX_GPT3 (OMAP24XX_L4_IO_BASE+0x78000) +#define OMAP24XX_GPT4 (OMAP24XX_L4_IO_BASE+0x7A000) +#define OMAP24XX_GPT5 (OMAP24XX_L4_IO_BASE+0x7C000) +#define OMAP24XX_GPT6 (OMAP24XX_L4_IO_BASE+0x7E000) +#define OMAP24XX_GPT7 (OMAP24XX_L4_IO_BASE+0x80000) +#define OMAP24XX_GPT8 (OMAP24XX_L4_IO_BASE+0x82000) +#define OMAP24XX_GPT9 (OMAP24XX_L4_IO_BASE+0x84000) +#define OMAP24XX_GPT10 (OMAP24XX_L4_IO_BASE+0x86000) +#define OMAP24XX_GPT11 (OMAP24XX_L4_IO_BASE+0x88000) +#define OMAP24XX_GPT12 (OMAP24XX_L4_IO_BASE+0x8A000) + +#define GPTIMER1_SYSCONFIG GPT1_REG32(0x010) +#define GPTIMER2_SYSCONFIG __REG32(OMAP24XX_GPT2 + 0x10) +#define GPTIMER3_SYSCONFIG __REG32(OMAP24XX_GPT3 + 0x10) +#define GPTIMER4_SYSCONFIG __REG32(OMAP24XX_GPT4 + 0x10) +#define GPTIMER5_SYSCONFIG __REG32(OMAP24XX_GPT5 + 0x10) +#define GPTIMER6_SYSCONFIG __REG32(OMAP24XX_GPT6 + 0x10) +#define GPTIMER7_SYSCONFIG __REG32(OMAP24XX_GPT7 + 0x10) +#define GPTIMER8_SYSCONFIG __REG32(OMAP24XX_GPT8 + 0x10) +#define GPTIMER9_SYSCONFIG __REG32(OMAP24XX_GPT9 + 0x10) +#define GPTIMER10_SYSCONFIG __REG32(OMAP24XX_GPT10 + 0x10) +#define GPTIMER11_SYSCONFIG __REG32(OMAP24XX_GPT11 + 0x10) +#define GPTIMER12_SYSCONFIG __REG32(OMAP24XX_GPT12 + 0x10) + +#define GPIOX_BASE(X) (OMAP24XX_GPIO_BASE+(0x2000*((X)-1))) + +#define GPIO1_SYSCONFIG __REG32((GPIOX_BASE(1)+0x10)) +#define GPIO2_SYSCONFIG __REG32((GPIOX_BASE(2)+0x10)) +#define GPIO3_SYSCONFIG __REG32((GPIOX_BASE(3)+0x10)) +#define GPIO4_SYSCONFIG __REG32((GPIOX_BASE(4)+0x10)) + +/* GP TIMER 1 */ +#define GPTIMER1_TISTAT GPT1_REG32(0x014) +#define GPTIMER1_TISR GPT1_REG32(0x018) +#define GPTIMER1_TIER GPT1_REG32(0x01C) +#define GPTIMER1_TWER GPT1_REG32(0x020) +#define GPTIMER1_TCLR GPT1_REG32(0x024) +#define GPTIMER1_TCRR GPT1_REG32(0x028) +#define GPTIMER1_TLDR GPT1_REG32(0x02C) +#define GPTIMER1_TTGR GPT1_REG32(0x030) +#define GPTIMER1_TWPS GPT1_REG32(0x034) +#define GPTIMER1_TMAR GPT1_REG32(0x038) +#define GPTIMER1_TCAR1 GPT1_REG32(0x03C) +#define GPTIMER1_TSICR GPT1_REG32(0x040) +#define GPTIMER1_TCAR2 GPT1_REG32(0x044) + +/* rkw -- base fix up please... */ +#define GPTIMER3_TISR __REG32(OMAP24XX_L4_IO_BASE+0x78018) + +/* SDRC */ +#define SDRC_DLLA_CTRL __REG32(OMAP24XX_SDRC_BASE+0x060) +#define SDRC_DLLA_STATUS __REG32(OMAP24XX_SDRC_BASE+0x064) +#define SDRC_DLLB_CTRL __REG32(OMAP24XX_SDRC_BASE+0x068) +#define SDRC_DLLB_STATUS __REG32(OMAP24XX_SDRC_BASE+0x06C) +#define SDRC_POWER __REG32(OMAP24XX_SDRC_BASE+0x070) +#define SDRC_MR_0 __REG32(OMAP24XX_SDRC_BASE+0x084) + +/* GPIO 1 */ +#define GPIO1_BASE GPIOX_BASE(1) +#define GPIO1_REG32(offset) __REG32(GPIO1_BASE + (offset)) +#define GPIO1_IRQENABLE1 GPIO1_REG32(0x01C) +#define GPIO1_IRQSTATUS1 GPIO1_REG32(0x018) +#define GPIO1_IRQENABLE2 GPIO1_REG32(0x02C) +#define GPIO1_IRQSTATUS2 GPIO1_REG32(0x028) +#define GPIO1_WAKEUPENABLE GPIO1_REG32(0x020) +#define GPIO1_RISINGDETECT GPIO1_REG32(0x048) +#define GPIO1_DATAIN GPIO1_REG32(0x038) +#define GPIO1_OE GPIO1_REG32(0x034) +#define GPIO1_DATAOUT GPIO1_REG32(0x03C) + +/* GPIO2 */ +#define GPIO2_BASE GPIOX_BASE(2) +#define GPIO2_REG32(offset) __REG32(GPIO2_BASE + (offset)) +#define GPIO2_IRQENABLE1 GPIO2_REG32(0x01C) +#define GPIO2_IRQSTATUS1 GPIO2_REG32(0x018) +#define GPIO2_IRQENABLE2 GPIO2_REG32(0x02C) +#define GPIO2_IRQSTATUS2 GPIO2_REG32(0x028) +#define GPIO2_WAKEUPENABLE GPIO2_REG32(0x020) +#define GPIO2_RISINGDETECT GPIO2_REG32(0x048) +#define GPIO2_DATAIN GPIO2_REG32(0x038) +#define GPIO2_OE GPIO2_REG32(0x034) +#define GPIO2_DATAOUT GPIO2_REG32(0x03C) + +/* GPIO 3 */ +#define GPIO3_BASE GPIOX_BASE(3) +#define GPIO3_REG32(offset) __REG32(GPIO3_BASE + (offset)) +#define GPIO3_IRQENABLE1 GPIO3_REG32(0x01C) +#define GPIO3_IRQSTATUS1 GPIO3_REG32(0x018) +#define GPIO3_IRQENABLE2 GPIO3_REG32(0x02C) +#define GPIO3_IRQSTATUS2 GPIO3_REG32(0x028) +#define GPIO3_WAKEUPENABLE GPIO3_REG32(0x020) +#define GPIO3_RISINGDETECT GPIO3_REG32(0x048) +#define GPIO3_FALLINGDETECT GPIO3_REG32(0x04C) +#define GPIO3_DATAIN GPIO3_REG32(0x038) +#define GPIO3_OE GPIO3_REG32(0x034) +#define GPIO3_DATAOUT GPIO3_REG32(0x03C) +#define GPIO3_DEBOUNCENABLE GPIO3_REG32(0x050) +#define GPIO3_DEBOUNCINGTIME GPIO3_REG32(0x054) + +/* GPIO 4 */ +#define GPIO4_BASE GPIOX_BASE(4) +#define GPIO4_REG32(offset) __REG32(GPIO4_BASE + (offset)) +#define GPIO4_IRQENABLE1 GPIO4_REG32(0x01C) +#define GPIO4_IRQSTATUS1 GPIO4_REG32(0x018) +#define GPIO4_IRQENABLE2 GPIO4_REG32(0x02C) +#define GPIO4_IRQSTATUS2 GPIO4_REG32(0x028) +#define GPIO4_WAKEUPENABLE GPIO4_REG32(0x020) +#define GPIO4_RISINGDETECT GPIO4_REG32(0x048) +#define GPIO4_FALLINGDETECT GPIO4_REG32(0x04C) +#define GPIO4_DATAIN GPIO4_REG32(0x038) +#define GPIO4_OE GPIO4_REG32(0x034) +#define GPIO4_DATAOUT GPIO4_REG32(0x03C) +#define GPIO4_DEBOUNCENABLE GPIO4_REG32(0x050) +#define GPIO4_DEBOUNCINGTIME GPIO4_REG32(0x054) + + +/* IO CONFIG */ +#define CONTROL_BASE (OMAP24XX_CTRL_BASE) +#define CONTROL_REG32(offset) __REG32(CONTROL_BASE + (offset)) + +#define CONTROL_PADCONF_SPI1_NCS2 CONTROL_REG32(0x104) +#define CONTROL_PADCONF_SYS_XTALOUT CONTROL_REG32(0x134) +#define CONTROL_PADCONF_UART1_RX CONTROL_REG32(0x0C8) +#define CONTROL_PADCONF_MCBSP1_DX CONTROL_REG32(0x10C) +#define CONTROL_PADCONF_GPMC_NCS4 CONTROL_REG32(0x090) +#define CONTROL_PADCONF_DSS_D5 CONTROL_REG32(0x0B8) +#define CONTROL_PADCONF_DSS_D9 CONTROL_REG32(0x0BC) +#define CONTROL_PADCONF_DSS_D13 CONTROL_REG32(0x0C0) +#define CONTROL_PADCONF_DSS_VSYNC CONTROL_REG32(0x0CC) + +/* CONTROL */ +#define CONTROL_DEVCONF CONTROL_REG32(0x274) + +/* INTERRUPT CONTROLLER */ +#define INTC_BASE (OMAP24XX_L4_IO_BASE+0xfe000) +#define INTC_REG32(offset) __REG32(INTC_BASE + (offset)) + +#define INTC1_U_BASE INTC_REG32(0x000) +#define INTC_MIR0 INTC_REG32(0x084) +#define INTC_MIR_SET0 INTC_REG32(0x08C) +#define INTC_MIR_CLEAR0 INTC_REG32(0x088) +#define INTC_ISR_CLEAR0 INTC_REG32(0x094) +#define INTC_MIR1 INTC_REG32(0x0A4) +#define INTC_MIR_SET1 INTC_REG32(0x0AC) +#define INTC_MIR_CLEAR1 INTC_REG32(0x0A8) +#define INTC_ISR_CLEAR1 INTC_REG32(0x0B4) +#define INTC_MIR2 INTC_REG32(0x0C4) +#define INTC_MIR_SET2 INTC_REG32(0x0CC) +#define INTC_MIR_CLEAR2 INTC_REG32(0x0C8) +#define INTC_ISR_CLEAR2 INTC_REG32(0x0D4) +#define INTC_SIR_IRQ INTC_REG32(0x040) +#define INTC_CONTROL INTC_REG32(0x048) +#define INTC_ILR11 INTC_REG32(0x12C) +#define INTC_ILR32 INTC_REG32(0x180) +#define INTC_ILR37 INTC_REG32(0x194) +#define INTC_SYSCONFIG INTC_REG32(0x010) + +/* RAM FIREWALL */ +#define RAMFW_BASE (0x68005000) +#define RAMFW_REG32(offset) __REG32(RAMFW_BASE + (offset)) + +#define RAMFW_REQINFOPERM0 RAMFW_REG32(0x048) +#define RAMFW_READPERM0 RAMFW_REG32(0x050) +#define RAMFW_WRITEPERM0 RAMFW_REG32(0x058) + +/* GPMC CS1 FPGA ON USER INTERFACE MODULE */ +//#define DEBUG_BOARD_LED_REGISTER 0x04000014 + +/* GPMC CS0 */ +#define GPMC_CONFIG1_0 GPMC_REG32(0x060) +#define GPMC_CONFIG2_0 GPMC_REG32(0x064) +#define GPMC_CONFIG3_0 GPMC_REG32(0x068) +#define GPMC_CONFIG4_0 GPMC_REG32(0x06C) +#define GPMC_CONFIG5_0 GPMC_REG32(0x070) +#define GPMC_CONFIG6_0 GPMC_REG32(0x074) +#define GPMC_CONFIG7_0 GPMC_REG32(0x078) + +/* GPMC CS1 */ +#define GPMC_CONFIG1_1 GPMC_REG32(0x090) +#define GPMC_CONFIG2_1 GPMC_REG32(0x094) +#define GPMC_CONFIG3_1 GPMC_REG32(0x098) +#define GPMC_CONFIG4_1 GPMC_REG32(0x09C) +#define GPMC_CONFIG5_1 GPMC_REG32(0x0a0) +#define GPMC_CONFIG6_1 GPMC_REG32(0x0a4) +#define GPMC_CONFIG7_1 GPMC_REG32(0x0a8) + +/* DSS */ +#define DSS_CONTROL DISP_REG32(0x040) +#define DISPC_CONTROL DISP_REG32(0x440) +#define DISPC_SYSSTATUS DISP_REG32(0x414) +#define DISPC_IRQSTATUS DISP_REG32(0x418) +#define DISPC_IRQENABLE DISP_REG32(0x41C) +#define DISPC_CONFIG DISP_REG32(0x444) +#define DISPC_DEFAULT_COLOR0 DISP_REG32(0x44C) +#define DISPC_DEFAULT_COLOR1 DISP_REG32(0x450) +#define DISPC_TRANS_COLOR0 DISP_REG32(0x454) +#define DISPC_TRANS_COLOR1 DISP_REG32(0x458) +#define DISPC_LINE_NUMBER DISP_REG32(0x460) +#define DISPC_TIMING_H DISP_REG32(0x464) +#define DISPC_TIMING_V DISP_REG32(0x468) +#define DISPC_POL_FREQ DISP_REG32(0x46C) +#define DISPC_DIVISOR DISP_REG32(0x470) +#define DISPC_SIZE_DIG DISP_REG32(0x478) +#define DISPC_SIZE_LCD DISP_REG32(0x47C) +#define DISPC_GFX_BA0 DISP_REG32(0x480) +#define DISPC_GFX_BA1 DISP_REG32(0x484) +#define DISPC_GFX_POSITION DISP_REG32(0x488) +#define DISPC_GFX_SIZE DISP_REG32(0x48C) +#define DISPC_GFX_ATTRIBUTES DISP_REG32(0x4A0) +#define DISPC_GFX_FIFO_THRESHOLD DISP_REG32(0x4A4) +#define DISPC_GFX_ROW_INC DISP_REG32(0x4AC) +#define DISPC_GFX_PIXEL_INC DISP_REG32(0x4B0) +#define DISPC_GFX_WINDOW_SKIP DISP_REG32(0x4B4) +#define DISPC_GFX_TABLE_BA DISP_REG32(0x4B8) +#define DISPC_DATA_CYCLE1 DISP_REG32(0x5D4) +#define DISPC_DATA_CYCLE2 DISP_REG32(0x5D8) +#define DISPC_DATA_CYCLE3 DISP_REG32(0x5DC) + +/* Wake up define for board */ +#define GPIO97 (1 << 1) +#define GPIO88 (1 << 24) + +#endif /* __ASSEMBLER__ */ + +#endif + + + + + diff --git a/include/asm-arm/arch-omap/sram.h b/include/asm-arm/arch-omap/sram.h new file mode 100644 index 00000000000..e72ccbf0fe0 --- /dev/null +++ b/include/asm-arm/arch-omap/sram.h @@ -0,0 +1,38 @@ +/* + * linux/include/asm-arm/arch-omap/sram.h + * + * Interface for functions that need to be run in internal SRAM + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#ifndef __ARCH_ARM_OMAP_SRAM_H +#define __ARCH_ARM_OMAP_SRAM_H + +extern void * omap_sram_push(void * start, unsigned long size); +extern void omap_sram_reprogram_clock(u32 dpllctl, u32 ckctl); + +extern void omap2_sram_ddr_init(u32 *slow_dll_ctrl, u32 fast_dll_ctrl, + u32 base_cs, u32 force_unlock); +extern void omap2_sram_reprogram_sdrc(u32 perf_level, u32 dll_val, + u32 mem_type); +extern u32 omap2_set_prcm(u32 dpll_ctrl_val, u32 sdrc_rfr_val, int bypass); + + +/* Do not use these */ +extern void sram_reprogram_clock(u32 ckctl, u32 dpllctl); +extern unsigned long sram_reprogram_clock_sz; + +extern void sram_ddr_init(u32 *slow_dll_ctrl, u32 fast_dll_ctrl, + u32 base_cs, u32 force_unlock); +extern unsigned long sram_ddr_init_sz; + +extern u32 sram_set_prcm(u32 dpll_ctrl_val, u32 sdrc_rfr_val, int bypass); +extern unsigned long sram_set_prcm_sz; + +extern void sram_reprogram_sdrc(u32 perf_level, u32 dll_val, u32 mem_type); +extern unsigned long sram_reprogram_sdrc_sz; + +#endif diff --git a/include/asm-arm/arch-omap/system.h b/include/asm-arm/arch-omap/system.h index ff37bc27e60..b43cdd2a387 100644 --- a/include/asm-arm/arch-omap/system.h +++ b/include/asm-arm/arch-omap/system.h @@ -6,18 +6,21 @@ #define __ASM_ARCH_SYSTEM_H #include <linux/config.h> #include <asm/mach-types.h> +#include <asm/hardware/clock.h> #include <asm/arch/hardware.h> -#include <asm/mach-types.h> +#include <asm/arch/prcm.h> + +#ifndef CONFIG_MACH_VOICEBLUE +#define voiceblue_reset() do {} while (0) +#endif static inline void arch_idle(void) { cpu_do_idle(); } -static inline void arch_reset(char mode) +static inline void omap1_arch_reset(char mode) { - -#ifdef CONFIG_ARCH_OMAP16XX /* * Workaround for 5912/1611b bug mentioned in sprz209d.pdf p. 28 * "Global Software Reset Affects Traffic Controller Frequency". @@ -27,13 +30,31 @@ static inline void arch_reset(char mode) DPLL_CTL); omap_writew(0x8, ARM_RSTCT1); } -#endif -#ifdef CONFIG_MACH_VOICEBLUE + if (machine_is_voiceblue()) voiceblue_reset(); else -#endif omap_writew(1, ARM_RSTCT1); } +static inline void omap2_arch_reset(char mode) +{ + u32 rate; + struct clk *vclk, *sclk; + + vclk = clk_get(NULL, "virt_prcm_set"); + sclk = clk_get(NULL, "sys_ck"); + rate = clk_get_rate(sclk); + clk_set_rate(vclk, rate); /* go to bypass for OMAP limitation */ + RM_RSTCTRL_WKUP |= 2; +} + +static inline void arch_reset(char mode) +{ + if (!cpu_is_omap24xx()) + omap1_arch_reset(mode); + else + omap2_arch_reset(mode); +} + #endif diff --git a/include/asm-arm/arch-omap/timex.h b/include/asm-arm/arch-omap/timex.h index b61ddb491e8..21f2e367185 100644 --- a/include/asm-arm/arch-omap/timex.h +++ b/include/asm-arm/arch-omap/timex.h @@ -28,6 +28,14 @@ #if !defined(__ASM_ARCH_OMAP_TIMEX_H) #define __ASM_ARCH_OMAP_TIMEX_H +/* + * OMAP 32KHz timer updates time one jiffie at a time from a secondary timer, + * and that's why the CLOCK_TICK_RATE is not 32768. + */ +#ifdef CONFIG_OMAP_32K_TIMER +#define CLOCK_TICK_RATE (CONFIG_OMAP_32K_TIMER_HZ) +#else #define CLOCK_TICK_RATE (HZ * 100000UL) +#endif #endif /* __ASM_ARCH_OMAP_TIMEX_H */ diff --git a/include/asm-arm/arch-omap/uncompress.h b/include/asm-arm/arch-omap/uncompress.h index 3545c86859c..c718264affb 100644 --- a/include/asm-arm/arch-omap/uncompress.h +++ b/include/asm-arm/arch-omap/uncompress.h @@ -36,10 +36,14 @@ putstr(const char *s) volatile u8 * uart = 0; int shift = 2; +#ifdef CONFIG_MACH_OMAP_PALMTE + return; +#endif + #ifdef CONFIG_ARCH_OMAP #ifdef CONFIG_OMAP_LL_DEBUG_UART3 uart = (volatile u8 *)(OMAP_UART3_BASE); -#elif CONFIG_OMAP_LL_DEBUG_UART2 +#elif defined(CONFIG_OMAP_LL_DEBUG_UART2) uart = (volatile u8 *)(OMAP_UART2_BASE); #else uart = (volatile u8 *)(OMAP_UART1_BASE); diff --git a/include/asm-arm/arch-pxa/sharpsl.h b/include/asm-arm/arch-pxa/sharpsl.h index 311f2bb5386..0b43495d24b 100644 --- a/include/asm-arm/arch-pxa/sharpsl.h +++ b/include/asm-arm/arch-pxa/sharpsl.h @@ -21,12 +21,18 @@ struct corgits_machinfo { void (*wait_hsync)(void); }; + /* * SharpSL Backlight */ - struct corgibl_machinfo { int max_intensity; void (*set_bl_intensity)(int intensity); }; +extern void corgibl_limit_intensity(int limit); + +/* + * SharpSL Battery/PM Driver + */ +extern void sharpsl_battery_kick(void); diff --git a/include/asm-arm/arch-pxa/ssp.h b/include/asm-arm/arch-pxa/ssp.h index 6ec67b018c0..949878c0d90 100644 --- a/include/asm-arm/arch-pxa/ssp.h +++ b/include/asm-arm/arch-pxa/ssp.h @@ -18,6 +18,11 @@ #ifndef SSP_H #define SSP_H +/* + * SSP initialisation flags + */ +#define SSP_NO_IRQ 0x1 /* don't register an irq handler in SSP driver */ + struct ssp_state { u32 cr0; u32 cr1; @@ -31,6 +36,7 @@ struct ssp_dev { u32 flags; u32 psp_flags; u32 speed; + int irq; }; int ssp_write_word(struct ssp_dev *dev, u32 data); @@ -40,7 +46,7 @@ void ssp_enable(struct ssp_dev *dev); void ssp_disable(struct ssp_dev *dev); void ssp_save_state(struct ssp_dev *dev, struct ssp_state *ssp); void ssp_restore_state(struct ssp_dev *dev, struct ssp_state *ssp); -int ssp_init(struct ssp_dev *dev, u32 port); +int ssp_init(struct ssp_dev *dev, u32 port, u32 init_flags); int ssp_config(struct ssp_dev *dev, u32 mode, u32 flags, u32 psp_flags, u32 speed); void ssp_exit(struct ssp_dev *dev); diff --git a/include/asm-i386/ide.h b/include/asm-i386/ide.h index 79dfab87135..454440193ea 100644 --- a/include/asm-i386/ide.h +++ b/include/asm-i386/ide.h @@ -41,6 +41,12 @@ static __inline__ int ide_default_irq(unsigned long base) static __inline__ unsigned long ide_default_io_base(int index) { + /* + * If PCI is present then it is not safe to poke around + * the other legacy IDE ports. Only 0x1f0 and 0x170 are + * defined compatibility mode ports for PCI. A user can + * override this using ide= but we must default safe. + */ if (pci_find_device(PCI_ANY_ID, PCI_ANY_ID, NULL) == NULL) { switch(index) { case 2: return 0x1e8; diff --git a/include/asm-i386/msi.h b/include/asm-i386/msi.h index b85393094c8..f041d4495fa 100644 --- a/include/asm-i386/msi.h +++ b/include/asm-i386/msi.h @@ -10,13 +10,6 @@ #include <mach_apic.h> #define LAST_DEVICE_VECTOR 232 -#define MSI_DEST_MODE MSI_LOGICAL_MODE -#define MSI_TARGET_CPU_SHIFT 12 - -#ifdef CONFIG_SMP -#define MSI_TARGET_CPU logical_smp_processor_id() -#else -#define MSI_TARGET_CPU cpu_to_logical_apicid(first_cpu(cpu_online_map)) -#endif +#define MSI_TARGET_CPU_SHIFT 12 #endif /* ASM_MSI_H */ diff --git a/include/asm-i386/smp.h b/include/asm-i386/smp.h index 13250199976..61d3ab9db70 100644 --- a/include/asm-i386/smp.h +++ b/include/asm-i386/smp.h @@ -45,6 +45,8 @@ extern void unlock_ipi_call_lock(void); #define MAX_APICID 256 extern u8 x86_cpu_to_apicid[]; +#define cpu_physical_id(cpu) x86_cpu_to_apicid[cpu] + #ifdef CONFIG_HOTPLUG_CPU extern void cpu_exit_clear(void); extern void cpu_uninit(void); @@ -92,6 +94,10 @@ extern int __cpu_disable(void); extern void __cpu_die(unsigned int cpu); #endif /* !__ASSEMBLY__ */ +#else /* CONFIG_SMP */ + +#define cpu_physical_id(cpu) boot_cpu_physical_apicid + #define NO_PROC_ID 0xFF /* No processor magic marker */ #endif diff --git a/include/asm-ia64/kdebug.h b/include/asm-ia64/kdebug.h index 4d376e1663f..8b01a083dde 100644 --- a/include/asm-ia64/kdebug.h +++ b/include/asm-ia64/kdebug.h @@ -22,6 +22,9 @@ * 2005-Apr Rusty Lynch <rusty.lynch@intel.com> and Anil S Keshavamurthy * <anil.s.keshavamurthy@intel.com> adopted from * include/asm-x86_64/kdebug.h + * + * 2005-Oct Keith Owens <kaos@sgi.com>. Expand notify_die to cover more + * events. */ #include <linux/notifier.h> @@ -35,13 +38,36 @@ struct die_args { int signr; }; -int register_die_notifier(struct notifier_block *nb); +extern int register_die_notifier(struct notifier_block *); +extern int unregister_die_notifier(struct notifier_block *); extern struct notifier_block *ia64die_chain; enum die_val { DIE_BREAK = 1, - DIE_SS, + DIE_FAULT, + DIE_OOPS, DIE_PAGE_FAULT, + DIE_MACHINE_HALT, + DIE_MACHINE_RESTART, + DIE_MCA_MONARCH_ENTER, + DIE_MCA_MONARCH_PROCESS, + DIE_MCA_MONARCH_LEAVE, + DIE_MCA_SLAVE_ENTER, + DIE_MCA_SLAVE_PROCESS, + DIE_MCA_SLAVE_LEAVE, + DIE_MCA_RENDZVOUS_ENTER, + DIE_MCA_RENDZVOUS_PROCESS, + DIE_MCA_RENDZVOUS_LEAVE, + DIE_INIT_MONARCH_ENTER, + DIE_INIT_MONARCH_PROCESS, + DIE_INIT_MONARCH_LEAVE, + DIE_INIT_SLAVE_ENTER, + DIE_INIT_SLAVE_PROCESS, + DIE_INIT_SLAVE_LEAVE, + DIE_KDEBUG_ENTER, + DIE_KDEBUG_LEAVE, + DIE_KDUMP_ENTER, + DIE_KDUMP_LEAVE, }; static inline int notify_die(enum die_val val, char *str, struct pt_regs *regs, diff --git a/include/asm-ia64/mmu_context.h b/include/asm-ia64/mmu_context.h index 8d6e72f7b08..b5c65081a3a 100644 --- a/include/asm-ia64/mmu_context.h +++ b/include/asm-ia64/mmu_context.h @@ -7,12 +7,13 @@ */ /* - * Routines to manage the allocation of task context numbers. Task context numbers are - * used to reduce or eliminate the need to perform TLB flushes due to context switches. - * Context numbers are implemented using ia-64 region ids. Since the IA-64 TLB does not - * consider the region number when performing a TLB lookup, we need to assign a unique - * region id to each region in a process. We use the least significant three bits in a - * region id for this purpose. + * Routines to manage the allocation of task context numbers. Task context + * numbers are used to reduce or eliminate the need to perform TLB flushes + * due to context switches. Context numbers are implemented using ia-64 + * region ids. Since the IA-64 TLB does not consider the region number when + * performing a TLB lookup, we need to assign a unique region id to each + * region in a process. We use the least significant three bits in aregion + * id for this purpose. */ #define IA64_REGION_ID_KERNEL 0 /* the kernel's region id (tlb.c depends on this being 0) */ @@ -32,13 +33,17 @@ struct ia64_ctx { spinlock_t lock; unsigned int next; /* next context number to use */ - unsigned int limit; /* next >= limit => must call wrap_mmu_context() */ - unsigned int max_ctx; /* max. context value supported by all CPUs */ + unsigned int limit; /* available free range */ + unsigned int max_ctx; /* max. context value supported by all CPUs */ + /* call wrap_mmu_context when next >= max */ + unsigned long *bitmap; /* bitmap size is max_ctx+1 */ + unsigned long *flushmap;/* pending rid to be flushed */ }; extern struct ia64_ctx ia64_ctx; DECLARE_PER_CPU(u8, ia64_need_tlb_flush); +extern void mmu_context_init (void); extern void wrap_mmu_context (struct mm_struct *mm); static inline void @@ -47,10 +52,10 @@ enter_lazy_tlb (struct mm_struct *mm, struct task_struct *tsk) } /* - * When the context counter wraps around all TLBs need to be flushed because an old - * context number might have been reused. This is signalled by the ia64_need_tlb_flush - * per-CPU variable, which is checked in the routine below. Called by activate_mm(). - * <efocht@ess.nec.de> + * When the context counter wraps around all TLBs need to be flushed because + * an old context number might have been reused. This is signalled by the + * ia64_need_tlb_flush per-CPU variable, which is checked in the routine + * below. Called by activate_mm(). <efocht@ess.nec.de> */ static inline void delayed_tlb_flush (void) @@ -60,11 +65,9 @@ delayed_tlb_flush (void) if (unlikely(__ia64_per_cpu_var(ia64_need_tlb_flush))) { spin_lock_irqsave(&ia64_ctx.lock, flags); - { - if (__ia64_per_cpu_var(ia64_need_tlb_flush)) { - local_flush_tlb_all(); - __ia64_per_cpu_var(ia64_need_tlb_flush) = 0; - } + if (__ia64_per_cpu_var(ia64_need_tlb_flush)) { + local_flush_tlb_all(); + __ia64_per_cpu_var(ia64_need_tlb_flush) = 0; } spin_unlock_irqrestore(&ia64_ctx.lock, flags); } @@ -76,20 +79,27 @@ get_mmu_context (struct mm_struct *mm) unsigned long flags; nv_mm_context_t context = mm->context; - if (unlikely(!context)) { - spin_lock_irqsave(&ia64_ctx.lock, flags); - { - /* re-check, now that we've got the lock: */ - context = mm->context; - if (context == 0) { - cpus_clear(mm->cpu_vm_mask); - if (ia64_ctx.next >= ia64_ctx.limit) - wrap_mmu_context(mm); - mm->context = context = ia64_ctx.next++; - } + if (likely(context)) + goto out; + + spin_lock_irqsave(&ia64_ctx.lock, flags); + /* re-check, now that we've got the lock: */ + context = mm->context; + if (context == 0) { + cpus_clear(mm->cpu_vm_mask); + if (ia64_ctx.next >= ia64_ctx.limit) { + ia64_ctx.next = find_next_zero_bit(ia64_ctx.bitmap, + ia64_ctx.max_ctx, ia64_ctx.next); + ia64_ctx.limit = find_next_bit(ia64_ctx.bitmap, + ia64_ctx.max_ctx, ia64_ctx.next); + if (ia64_ctx.next >= ia64_ctx.max_ctx) + wrap_mmu_context(mm); } - spin_unlock_irqrestore(&ia64_ctx.lock, flags); + mm->context = context = ia64_ctx.next++; + __set_bit(context, ia64_ctx.bitmap); } + spin_unlock_irqrestore(&ia64_ctx.lock, flags); +out: /* * Ensure we're not starting to use "context" before any old * uses of it are gone from our TLB. @@ -100,8 +110,8 @@ get_mmu_context (struct mm_struct *mm) } /* - * Initialize context number to some sane value. MM is guaranteed to be a brand-new - * address-space, so no TLB flushing is needed, ever. + * Initialize context number to some sane value. MM is guaranteed to be a + * brand-new address-space, so no TLB flushing is needed, ever. */ static inline int init_new_context (struct task_struct *p, struct mm_struct *mm) @@ -162,7 +172,10 @@ activate_context (struct mm_struct *mm) if (!cpu_isset(smp_processor_id(), mm->cpu_vm_mask)) cpu_set(smp_processor_id(), mm->cpu_vm_mask); reload_context(context); - /* in the unlikely event of a TLB-flush by another thread, redo the load: */ + /* + * in the unlikely event of a TLB-flush by another thread, + * redo the load. + */ } while (unlikely(context != mm->context)); } @@ -175,8 +188,8 @@ static inline void activate_mm (struct mm_struct *prev, struct mm_struct *next) { /* - * We may get interrupts here, but that's OK because interrupt handlers cannot - * touch user-space. + * We may get interrupts here, but that's OK because interrupt + * handlers cannot touch user-space. */ ia64_set_kr(IA64_KR_PT_BASE, __pa(next->pgd)); activate_context(next); diff --git a/include/asm-ia64/msi.h b/include/asm-ia64/msi.h index 60f2137f927..97890f7762b 100644 --- a/include/asm-ia64/msi.h +++ b/include/asm-ia64/msi.h @@ -12,9 +12,6 @@ static inline void set_intr_gate (int nr, void *func) {} #define IO_APIC_VECTOR(irq) (irq) #define ack_APIC_irq ia64_eoi -#define cpu_mask_to_apicid(mask) cpu_physical_id(first_cpu(mask)) -#define MSI_DEST_MODE MSI_PHYSICAL_MODE -#define MSI_TARGET_CPU ((ia64_getreg(_IA64_REG_CR_LID) >> 16) & 0xffff) #define MSI_TARGET_CPU_SHIFT 4 #endif /* ASM_MSI_H */ diff --git a/include/asm-ia64/tlbflush.h b/include/asm-ia64/tlbflush.h index b65c6270272..a35b323bae4 100644 --- a/include/asm-ia64/tlbflush.h +++ b/include/asm-ia64/tlbflush.h @@ -51,6 +51,7 @@ flush_tlb_mm (struct mm_struct *mm) if (!mm) return; + set_bit(mm->context, ia64_ctx.flushmap); mm->context = 0; if (atomic_read(&mm->mm_users) == 0) diff --git a/include/asm-ppc64/abs_addr.h b/include/asm-powerpc/abs_addr.h index dc3fc3fefef..18415108fc5 100644 --- a/include/asm-ppc64/abs_addr.h +++ b/include/asm-powerpc/abs_addr.h @@ -1,5 +1,5 @@ -#ifndef _ABS_ADDR_H -#define _ABS_ADDR_H +#ifndef _ASM_POWERPC_ABS_ADDR_H +#define _ASM_POWERPC_ABS_ADDR_H #include <linux/config.h> @@ -70,4 +70,4 @@ static inline unsigned long phys_to_abs(unsigned long pa) #define iseries_hv_addr(virtaddr) \ (0x8000000000000000 | virt_to_abs(virtaddr)) -#endif /* _ABS_ADDR_H */ +#endif /* _ASM_POWERPC_ABS_ADDR_H */ diff --git a/include/asm-powerpc/asm-compat.h b/include/asm-powerpc/asm-compat.h new file mode 100644 index 00000000000..8b133efc9f7 --- /dev/null +++ b/include/asm-powerpc/asm-compat.h @@ -0,0 +1,55 @@ +#ifndef _ASM_POWERPC_ASM_COMPAT_H +#define _ASM_POWERPC_ASM_COMPAT_H + +#include <linux/config.h> +#include <asm/types.h> + +#ifdef __ASSEMBLY__ +# define stringify_in_c(...) __VA_ARGS__ +# define ASM_CONST(x) x +#else +/* This version of stringify will deal with commas... */ +# define __stringify_in_c(...) #__VA_ARGS__ +# define stringify_in_c(...) __stringify_in_c(__VA_ARGS__) " " +# define __ASM_CONST(x) x##UL +# define ASM_CONST(x) __ASM_CONST(x) +#endif + +#ifdef __powerpc64__ + +/* operations for longs and pointers */ +#define PPC_LL stringify_in_c(ld) +#define PPC_STL stringify_in_c(std) +#define PPC_LCMPI stringify_in_c(cmpdi) +#define PPC_LONG stringify_in_c(.llong) +#define PPC_TLNEI stringify_in_c(tdnei) +#define PPC_LLARX stringify_in_c(ldarx) +#define PPC_STLCX stringify_in_c(stdcx.) +#define PPC_CNTLZL stringify_in_c(cntlzd) + +#else /* 32-bit */ + +/* operations for longs and pointers */ +#define PPC_LL stringify_in_c(lwz) +#define PPC_STL stringify_in_c(stw) +#define PPC_LCMPI stringify_in_c(cmpwi) +#define PPC_LONG stringify_in_c(.long) +#define PPC_TLNEI stringify_in_c(twnei) +#define PPC_LLARX stringify_in_c(lwarx) +#define PPC_STLCX stringify_in_c(stwcx.) +#define PPC_CNTLZL stringify_in_c(cntlzw) + +#endif + +#ifdef CONFIG_IBM405_ERR77 +/* Erratum #77 on the 405 means we need a sync or dcbt before every + * stwcx. The old ATOMIC_SYNC_FIX covered some but not all of this. + */ +#define PPC405_ERR77(ra,rb) stringify_in_c(dcbt ra, rb;) +#define PPC405_ERR77_SYNC stringify_in_c(sync;) +#else +#define PPC405_ERR77(ra,rb) +#define PPC405_ERR77_SYNC +#endif + +#endif /* _ASM_POWERPC_ASM_COMPAT_H */ diff --git a/include/asm-powerpc/atomic.h b/include/asm-powerpc/atomic.h index ed4b345ed75..9c0b372a46e 100644 --- a/include/asm-powerpc/atomic.h +++ b/include/asm-powerpc/atomic.h @@ -9,21 +9,13 @@ typedef struct { volatile int counter; } atomic_t; #ifdef __KERNEL__ #include <asm/synch.h> +#include <asm/asm-compat.h> #define ATOMIC_INIT(i) { (i) } #define atomic_read(v) ((v)->counter) #define atomic_set(v,i) (((v)->counter) = (i)) -/* Erratum #77 on the 405 means we need a sync or dcbt before every stwcx. - * The old ATOMIC_SYNC_FIX covered some but not all of this. - */ -#ifdef CONFIG_IBM405_ERR77 -#define PPC405_ERR77(ra,rb) "dcbt " #ra "," #rb ";" -#else -#define PPC405_ERR77(ra,rb) -#endif - static __inline__ void atomic_add(int a, atomic_t *v) { int t; @@ -205,5 +197,183 @@ static __inline__ int atomic_dec_if_positive(atomic_t *v) #define smp_mb__before_atomic_inc() smp_mb() #define smp_mb__after_atomic_inc() smp_mb() +#ifdef __powerpc64__ + +typedef struct { volatile long counter; } atomic64_t; + +#define ATOMIC64_INIT(i) { (i) } + +#define atomic64_read(v) ((v)->counter) +#define atomic64_set(v,i) (((v)->counter) = (i)) + +static __inline__ void atomic64_add(long a, atomic64_t *v) +{ + long t; + + __asm__ __volatile__( +"1: ldarx %0,0,%3 # atomic64_add\n\ + add %0,%2,%0\n\ + stdcx. %0,0,%3 \n\ + bne- 1b" + : "=&r" (t), "=m" (v->counter) + : "r" (a), "r" (&v->counter), "m" (v->counter) + : "cc"); +} + +static __inline__ long atomic64_add_return(long a, atomic64_t *v) +{ + long t; + + __asm__ __volatile__( + EIEIO_ON_SMP +"1: ldarx %0,0,%2 # atomic64_add_return\n\ + add %0,%1,%0\n\ + stdcx. %0,0,%2 \n\ + bne- 1b" + ISYNC_ON_SMP + : "=&r" (t) + : "r" (a), "r" (&v->counter) + : "cc", "memory"); + + return t; +} + +#define atomic64_add_negative(a, v) (atomic64_add_return((a), (v)) < 0) + +static __inline__ void atomic64_sub(long a, atomic64_t *v) +{ + long t; + + __asm__ __volatile__( +"1: ldarx %0,0,%3 # atomic64_sub\n\ + subf %0,%2,%0\n\ + stdcx. %0,0,%3 \n\ + bne- 1b" + : "=&r" (t), "=m" (v->counter) + : "r" (a), "r" (&v->counter), "m" (v->counter) + : "cc"); +} + +static __inline__ long atomic64_sub_return(long a, atomic64_t *v) +{ + long t; + + __asm__ __volatile__( + EIEIO_ON_SMP +"1: ldarx %0,0,%2 # atomic64_sub_return\n\ + subf %0,%1,%0\n\ + stdcx. %0,0,%2 \n\ + bne- 1b" + ISYNC_ON_SMP + : "=&r" (t) + : "r" (a), "r" (&v->counter) + : "cc", "memory"); + + return t; +} + +static __inline__ void atomic64_inc(atomic64_t *v) +{ + long t; + + __asm__ __volatile__( +"1: ldarx %0,0,%2 # atomic64_inc\n\ + addic %0,%0,1\n\ + stdcx. %0,0,%2 \n\ + bne- 1b" + : "=&r" (t), "=m" (v->counter) + : "r" (&v->counter), "m" (v->counter) + : "cc"); +} + +static __inline__ long atomic64_inc_return(atomic64_t *v) +{ + long t; + + __asm__ __volatile__( + EIEIO_ON_SMP +"1: ldarx %0,0,%1 # atomic64_inc_return\n\ + addic %0,%0,1\n\ + stdcx. %0,0,%1 \n\ + bne- 1b" + ISYNC_ON_SMP + : "=&r" (t) + : "r" (&v->counter) + : "cc", "memory"); + + return t; +} + +/* + * atomic64_inc_and_test - increment and test + * @v: pointer of type atomic64_t + * + * Atomically increments @v by 1 + * and returns true if the result is zero, or false for all + * other cases. + */ +#define atomic64_inc_and_test(v) (atomic64_inc_return(v) == 0) + +static __inline__ void atomic64_dec(atomic64_t *v) +{ + long t; + + __asm__ __volatile__( +"1: ldarx %0,0,%2 # atomic64_dec\n\ + addic %0,%0,-1\n\ + stdcx. %0,0,%2\n\ + bne- 1b" + : "=&r" (t), "=m" (v->counter) + : "r" (&v->counter), "m" (v->counter) + : "cc"); +} + +static __inline__ long atomic64_dec_return(atomic64_t *v) +{ + long t; + + __asm__ __volatile__( + EIEIO_ON_SMP +"1: ldarx %0,0,%1 # atomic64_dec_return\n\ + addic %0,%0,-1\n\ + stdcx. %0,0,%1\n\ + bne- 1b" + ISYNC_ON_SMP + : "=&r" (t) + : "r" (&v->counter) + : "cc", "memory"); + + return t; +} + +#define atomic64_sub_and_test(a, v) (atomic64_sub_return((a), (v)) == 0) +#define atomic64_dec_and_test(v) (atomic64_dec_return((v)) == 0) + +/* + * Atomically test *v and decrement if it is greater than 0. + * The function returns the old value of *v minus 1. + */ +static __inline__ long atomic64_dec_if_positive(atomic64_t *v) +{ + long t; + + __asm__ __volatile__( + EIEIO_ON_SMP +"1: ldarx %0,0,%1 # atomic64_dec_if_positive\n\ + addic. %0,%0,-1\n\ + blt- 2f\n\ + stdcx. %0,0,%1\n\ + bne- 1b" + ISYNC_ON_SMP + "\n\ +2:" : "=&r" (t) + : "r" (&v->counter) + : "cc", "memory"); + + return t; +} + +#endif /* __powerpc64__ */ + #endif /* __KERNEL__ */ #endif /* _ASM_POWERPC_ATOMIC_H_ */ diff --git a/include/asm-powerpc/bitops.h b/include/asm-powerpc/bitops.h index dc25c53704d..5727229b044 100644 --- a/include/asm-powerpc/bitops.h +++ b/include/asm-powerpc/bitops.h @@ -40,6 +40,7 @@ #include <linux/compiler.h> #include <asm/atomic.h> +#include <asm/asm-compat.h> #include <asm/synch.h> /* @@ -52,16 +53,6 @@ #define BITOP_WORD(nr) ((nr) / BITS_PER_LONG) #define BITOP_LE_SWIZZLE ((BITS_PER_LONG-1) & ~0x7) -#ifdef CONFIG_PPC64 -#define LARXL "ldarx" -#define STCXL "stdcx." -#define CNTLZL "cntlzd" -#else -#define LARXL "lwarx" -#define STCXL "stwcx." -#define CNTLZL "cntlzw" -#endif - static __inline__ void set_bit(int nr, volatile unsigned long *addr) { unsigned long old; @@ -69,10 +60,10 @@ static __inline__ void set_bit(int nr, volatile unsigned long *addr) unsigned long *p = ((unsigned long *)addr) + BITOP_WORD(nr); __asm__ __volatile__( -"1:" LARXL " %0,0,%3 # set_bit\n" +"1:" PPC_LLARX "%0,0,%3 # set_bit\n" "or %0,%0,%2\n" PPC405_ERR77(0,%3) - STCXL " %0,0,%3\n" + PPC_STLCX "%0,0,%3\n" "bne- 1b" : "=&r"(old), "=m"(*p) : "r"(mask), "r"(p), "m"(*p) @@ -86,10 +77,10 @@ static __inline__ void clear_bit(int nr, volatile unsigned long *addr) unsigned long *p = ((unsigned long *)addr) + BITOP_WORD(nr); __asm__ __volatile__( -"1:" LARXL " %0,0,%3 # set_bit\n" +"1:" PPC_LLARX "%0,0,%3 # clear_bit\n" "andc %0,%0,%2\n" PPC405_ERR77(0,%3) - STCXL " %0,0,%3\n" + PPC_STLCX "%0,0,%3\n" "bne- 1b" : "=&r"(old), "=m"(*p) : "r"(mask), "r"(p), "m"(*p) @@ -103,10 +94,10 @@ static __inline__ void change_bit(int nr, volatile unsigned long *addr) unsigned long *p = ((unsigned long *)addr) + BITOP_WORD(nr); __asm__ __volatile__( -"1:" LARXL " %0,0,%3 # set_bit\n" +"1:" PPC_LLARX "%0,0,%3 # change_bit\n" "xor %0,%0,%2\n" PPC405_ERR77(0,%3) - STCXL " %0,0,%3\n" + PPC_STLCX "%0,0,%3\n" "bne- 1b" : "=&r"(old), "=m"(*p) : "r"(mask), "r"(p), "m"(*p) @@ -122,10 +113,10 @@ static __inline__ int test_and_set_bit(unsigned long nr, __asm__ __volatile__( EIEIO_ON_SMP -"1:" LARXL " %0,0,%3 # test_and_set_bit\n" +"1:" PPC_LLARX "%0,0,%3 # test_and_set_bit\n" "or %1,%0,%2 \n" PPC405_ERR77(0,%3) - STCXL " %1,0,%3 \n" + PPC_STLCX "%1,0,%3 \n" "bne- 1b" ISYNC_ON_SMP : "=&r" (old), "=&r" (t) @@ -144,10 +135,10 @@ static __inline__ int test_and_clear_bit(unsigned long nr, __asm__ __volatile__( EIEIO_ON_SMP -"1:" LARXL " %0,0,%3 # test_and_clear_bit\n" +"1:" PPC_LLARX "%0,0,%3 # test_and_clear_bit\n" "andc %1,%0,%2 \n" PPC405_ERR77(0,%3) - STCXL " %1,0,%3 \n" + PPC_STLCX "%1,0,%3 \n" "bne- 1b" ISYNC_ON_SMP : "=&r" (old), "=&r" (t) @@ -166,10 +157,10 @@ static __inline__ int test_and_change_bit(unsigned long nr, __asm__ __volatile__( EIEIO_ON_SMP -"1:" LARXL " %0,0,%3 # test_and_change_bit\n" +"1:" PPC_LLARX "%0,0,%3 # test_and_change_bit\n" "xor %1,%0,%2 \n" PPC405_ERR77(0,%3) - STCXL " %1,0,%3 \n" + PPC_STLCX "%1,0,%3 \n" "bne- 1b" ISYNC_ON_SMP : "=&r" (old), "=&r" (t) @@ -184,9 +175,9 @@ static __inline__ void set_bits(unsigned long mask, unsigned long *addr) unsigned long old; __asm__ __volatile__( -"1:" LARXL " %0,0,%3 # set_bit\n" +"1:" PPC_LLARX "%0,0,%3 # set_bits\n" "or %0,%0,%2\n" - STCXL " %0,0,%3\n" + PPC_STLCX "%0,0,%3\n" "bne- 1b" : "=&r" (old), "=m" (*addr) : "r" (mask), "r" (addr), "m" (*addr) @@ -268,7 +259,7 @@ static __inline__ int __ilog2(unsigned long x) { int lz; - asm (CNTLZL " %0,%1" : "=r" (lz) : "r" (x)); + asm (PPC_CNTLZL "%0,%1" : "=r" (lz) : "r" (x)); return BITS_PER_LONG - 1 - lz; } diff --git a/include/asm-powerpc/bug.h b/include/asm-powerpc/bug.h index d625ee55f95..b001ecb3cd9 100644 --- a/include/asm-powerpc/bug.h +++ b/include/asm-powerpc/bug.h @@ -1,6 +1,7 @@ #ifndef _ASM_POWERPC_BUG_H #define _ASM_POWERPC_BUG_H +#include <asm/asm-compat.h> /* * Define an illegal instr to trap on the bug. * We don't use 0 because that marks the end of a function @@ -11,14 +12,6 @@ #ifndef __ASSEMBLY__ -#ifdef __powerpc64__ -#define BUG_TABLE_ENTRY ".llong" -#define BUG_TRAP_OP "tdnei" -#else -#define BUG_TABLE_ENTRY ".long" -#define BUG_TRAP_OP "twnei" -#endif /* __powerpc64__ */ - struct bug_entry { unsigned long bug_addr; long line; @@ -40,16 +33,16 @@ struct bug_entry *find_bug(unsigned long bugaddr); __asm__ __volatile__( \ "1: twi 31,0,0\n" \ ".section __bug_table,\"a\"\n" \ - "\t"BUG_TABLE_ENTRY" 1b,%0,%1,%2\n" \ + "\t"PPC_LONG" 1b,%0,%1,%2\n" \ ".previous" \ : : "i" (__LINE__), "i" (__FILE__), "i" (__FUNCTION__)); \ } while (0) #define BUG_ON(x) do { \ __asm__ __volatile__( \ - "1: "BUG_TRAP_OP" %0,0\n" \ + "1: "PPC_TLNEI" %0,0\n" \ ".section __bug_table,\"a\"\n" \ - "\t"BUG_TABLE_ENTRY" 1b,%1,%2,%3\n" \ + "\t"PPC_LONG" 1b,%1,%2,%3\n" \ ".previous" \ : : "r" ((long)(x)), "i" (__LINE__), \ "i" (__FILE__), "i" (__FUNCTION__)); \ @@ -57,9 +50,9 @@ struct bug_entry *find_bug(unsigned long bugaddr); #define WARN_ON(x) do { \ __asm__ __volatile__( \ - "1: "BUG_TRAP_OP" %0,0\n" \ + "1: "PPC_TLNEI" %0,0\n" \ ".section __bug_table,\"a\"\n" \ - "\t"BUG_TABLE_ENTRY" 1b,%1,%2,%3\n" \ + "\t"PPC_LONG" 1b,%1,%2,%3\n" \ ".previous" \ : : "r" ((long)(x)), \ "i" (__LINE__ + BUG_WARNING_TRAP), \ diff --git a/include/asm-powerpc/cache.h b/include/asm-powerpc/cache.h new file mode 100644 index 00000000000..26ce502e76e --- /dev/null +++ b/include/asm-powerpc/cache.h @@ -0,0 +1,40 @@ +#ifndef _ASM_POWERPC_CACHE_H +#define _ASM_POWERPC_CACHE_H + +#ifdef __KERNEL__ + +#include <linux/config.h> + +/* bytes per L1 cache line */ +#if defined(CONFIG_8xx) || defined(CONFIG_403GCX) +#define L1_CACHE_SHIFT 4 +#define MAX_COPY_PREFETCH 1 +#elif defined(CONFIG_PPC32) +#define L1_CACHE_SHIFT 5 +#define MAX_COPY_PREFETCH 4 +#else /* CONFIG_PPC64 */ +#define L1_CACHE_SHIFT 7 +#endif + +#define L1_CACHE_BYTES (1 << L1_CACHE_SHIFT) + +#define SMP_CACHE_BYTES L1_CACHE_BYTES +#define L1_CACHE_SHIFT_MAX 7 /* largest L1 which this arch supports */ + +#if defined(__powerpc64__) && !defined(__ASSEMBLY__) +struct ppc64_caches { + u32 dsize; /* L1 d-cache size */ + u32 dline_size; /* L1 d-cache line size */ + u32 log_dline_size; + u32 dlines_per_page; + u32 isize; /* L1 i-cache size */ + u32 iline_size; /* L1 i-cache line size */ + u32 log_iline_size; + u32 ilines_per_page; +}; + +extern struct ppc64_caches ppc64_caches; +#endif /* __powerpc64__ && ! __ASSEMBLY__ */ + +#endif /* __KERNEL__ */ +#endif /* _ASM_POWERPC_CACHE_H */ diff --git a/include/asm-ppc64/cacheflush.h b/include/asm-powerpc/cacheflush.h index ffbc08be8e5..8a740c88d93 100644 --- a/include/asm-ppc64/cacheflush.h +++ b/include/asm-powerpc/cacheflush.h @@ -1,13 +1,20 @@ -#ifndef _PPC64_CACHEFLUSH_H -#define _PPC64_CACHEFLUSH_H +/* + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ +#ifndef _ASM_POWERPC_CACHEFLUSH_H +#define _ASM_POWERPC_CACHEFLUSH_H + +#ifdef __KERNEL__ #include <linux/mm.h> #include <asm/cputable.h> /* - * No cache flushing is required when address mappings are - * changed, because the caches on PowerPCs are physically - * addressed. + * No cache flushing is required when address mappings are changed, + * because the caches on PowerPCs are physically addressed. */ #define flush_cache_all() do { } while (0) #define flush_cache_mm(mm) do { } while (0) @@ -22,27 +29,40 @@ extern void flush_dcache_page(struct page *page); #define flush_dcache_mmap_unlock(mapping) do { } while (0) extern void __flush_icache_range(unsigned long, unsigned long); +static inline void flush_icache_range(unsigned long start, unsigned long stop) +{ + if (!cpu_has_feature(CPU_FTR_COHERENT_ICACHE)) + __flush_icache_range(start, stop); +} + extern void flush_icache_user_range(struct vm_area_struct *vma, struct page *page, unsigned long addr, int len); +extern void __flush_dcache_icache(void *page_va); +extern void flush_dcache_icache_page(struct page *page); +#if defined(CONFIG_PPC32) && !defined(CONFIG_BOOKE) +extern void __flush_dcache_icache_phys(unsigned long physaddr); +#endif /* CONFIG_PPC32 && !CONFIG_BOOKE */ extern void flush_dcache_range(unsigned long start, unsigned long stop); -extern void flush_dcache_phys_range(unsigned long start, unsigned long stop); +#ifdef CONFIG_PPC32 +extern void clean_dcache_range(unsigned long start, unsigned long stop); +extern void invalidate_dcache_range(unsigned long start, unsigned long stop); +#endif /* CONFIG_PPC32 */ +#ifdef CONFIG_PPC64 extern void flush_inval_dcache_range(unsigned long start, unsigned long stop); +extern void flush_dcache_phys_range(unsigned long start, unsigned long stop); +#endif #define copy_to_user_page(vma, page, vaddr, dst, src, len) \ -do { memcpy(dst, src, len); \ - flush_icache_user_range(vma, page, vaddr, len); \ -} while (0) + do { \ + memcpy(dst, src, len); \ + flush_icache_user_range(vma, page, vaddr, len); \ + } while (0) #define copy_from_user_page(vma, page, vaddr, dst, src, len) \ memcpy(dst, src, len) -extern void __flush_dcache_icache(void *page_va); -static inline void flush_icache_range(unsigned long start, unsigned long stop) -{ - if (!cpu_has_feature(CPU_FTR_COHERENT_ICACHE)) - __flush_icache_range(start, stop); -} +#endif /* __KERNEL__ */ -#endif /* _PPC64_CACHEFLUSH_H */ +#endif /* _ASM_POWERPC_CACHEFLUSH_H */ diff --git a/include/asm-ppc64/compat.h b/include/asm-powerpc/compat.h index 6ec62cd2d1d..4db4360c4d4 100644 --- a/include/asm-ppc64/compat.h +++ b/include/asm-powerpc/compat.h @@ -1,5 +1,5 @@ -#ifndef _ASM_PPC64_COMPAT_H -#define _ASM_PPC64_COMPAT_H +#ifndef _ASM_POWERPC_COMPAT_H +#define _ASM_POWERPC_COMPAT_H /* * Architecture specific compatibility types */ @@ -49,7 +49,7 @@ struct compat_stat { compat_dev_t st_dev; compat_ino_t st_ino; compat_mode_t st_mode; - compat_nlink_t st_nlink; + compat_nlink_t st_nlink; __compat_uid32_t st_uid; __compat_gid32_t st_gid; compat_dev_t st_rdev; @@ -202,4 +202,4 @@ struct compat_shmid64_ds { compat_ulong_t __unused6; }; -#endif /* _ASM_PPC64_COMPAT_H */ +#endif /* _ASM_POWERPC_COMPAT_H */ diff --git a/include/asm-powerpc/cputable.h b/include/asm-powerpc/cputable.h index 79a0556a0ab..04e2726002c 100644 --- a/include/asm-powerpc/cputable.h +++ b/include/asm-powerpc/cputable.h @@ -2,7 +2,7 @@ #define __ASM_POWERPC_CPUTABLE_H #include <linux/config.h> -#include <asm/ppc_asm.h> /* for ASM_CONST */ +#include <asm/asm-compat.h> #define PPC_FEATURE_32 0x80000000 #define PPC_FEATURE_64 0x40000000 @@ -16,6 +16,10 @@ #define PPC_FEATURE_HAS_EFP_SINGLE 0x00400000 #define PPC_FEATURE_HAS_EFP_DOUBLE 0x00200000 #define PPC_FEATURE_NO_TB 0x00100000 +#define PPC_FEATURE_POWER4 0x00080000 +#define PPC_FEATURE_POWER5 0x00040000 +#define PPC_FEATURE_POWER5_PLUS 0x00020000 +#define PPC_FEATURE_CELL 0x00010000 #ifdef __KERNEL__ #ifndef __ASSEMBLY__ diff --git a/include/asm-powerpc/current.h b/include/asm-powerpc/current.h new file mode 100644 index 00000000000..82cd4a9ca99 --- /dev/null +++ b/include/asm-powerpc/current.h @@ -0,0 +1,27 @@ +#ifndef _ASM_POWERPC_CURRENT_H +#define _ASM_POWERPC_CURRENT_H + +/* + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +struct task_struct; + +#ifdef __powerpc64__ +#include <asm/paca.h> + +#define current (get_paca()->__current) + +#else + +/* + * We keep `current' in r2 for speed. + */ +register struct task_struct *current asm ("r2"); + +#endif + +#endif /* _ASM_POWERPC_CURRENT_H */ diff --git a/include/asm-powerpc/eeh_event.h b/include/asm-powerpc/eeh_event.h new file mode 100644 index 00000000000..d168a30b386 --- /dev/null +++ b/include/asm-powerpc/eeh_event.h @@ -0,0 +1,52 @@ +/* + * eeh_event.h + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * Copyright (c) 2005 Linas Vepstas <linas@linas.org> + */ + +#ifndef ASM_PPC64_EEH_EVENT_H +#define ASM_PPC64_EEH_EVENT_H + +/** EEH event -- structure holding pci controller data that describes + * a change in the isolation status of a PCI slot. A pointer + * to this struct is passed as the data pointer in a notify callback. + */ +struct eeh_event { + struct list_head list; + struct device_node *dn; /* struct device node */ + struct pci_dev *dev; /* affected device */ + int state; + int time_unavail; /* milliseconds until device might be available */ +}; + +/** + * eeh_send_failure_event - generate a PCI error event + * @dev pci device + * + * This routine builds a PCI error event which will be delivered + * to all listeners on the peh_notifier_chain. + * + * This routine can be called within an interrupt context; + * the actual event will be delivered in a normal context + * (from a workqueue). + */ +int eeh_send_failure_event (struct device_node *dn, + struct pci_dev *dev, + int reset_state, + int time_unavail); + +#endif /* ASM_PPC64_EEH_EVENT_H */ diff --git a/include/asm-powerpc/firmware.h b/include/asm-powerpc/firmware.h index 806c142ae9e..12fabbcb04f 100644 --- a/include/asm-powerpc/firmware.h +++ b/include/asm-powerpc/firmware.h @@ -43,6 +43,7 @@ #define FW_FEATURE_ISERIES (1UL<<21) enum { +#ifdef CONFIG_PPC64 FW_FEATURE_PSERIES_POSSIBLE = FW_FEATURE_PFT | FW_FEATURE_TCE | FW_FEATURE_SPRG0 | FW_FEATURE_DABR | FW_FEATURE_COPY | FW_FEATURE_ASR | FW_FEATURE_DEBUG | FW_FEATURE_TERM | @@ -70,6 +71,11 @@ enum { FW_FEATURE_ISERIES_ALWAYS & #endif FW_FEATURE_POSSIBLE, + +#else /* CONFIG_PPC64 */ + FW_FEATURE_POSSIBLE = 0, + FW_FEATURE_ALWAYS = 0, +#endif }; /* This is used to identify firmware features which are available diff --git a/include/asm-powerpc/futex.h b/include/asm-powerpc/futex.h index 37c94e52ab6..f0319d50b12 100644 --- a/include/asm-powerpc/futex.h +++ b/include/asm-powerpc/futex.h @@ -7,13 +7,14 @@ #include <asm/errno.h> #include <asm/synch.h> #include <asm/uaccess.h> -#include <asm/ppc_asm.h> +#include <asm/asm-compat.h> #define __futex_atomic_op(insn, ret, oldval, uaddr, oparg) \ __asm__ __volatile ( \ SYNC_ON_SMP \ "1: lwarx %0,0,%2\n" \ insn \ + PPC405_ERR77(0, %2) \ "2: stwcx. %1,0,%2\n" \ "bne- 1b\n" \ "li %1,0\n" \ @@ -23,7 +24,7 @@ ".previous\n" \ ".section __ex_table,\"a\"\n" \ ".align 3\n" \ - DATAL " 1b,4b,2b,4b\n" \ + PPC_LONG "1b,4b,2b,4b\n" \ ".previous" \ : "=&r" (oldval), "=&r" (ret) \ : "b" (uaddr), "i" (-EFAULT), "1" (oparg) \ diff --git a/include/asm-ppc64/hvcall.h b/include/asm-powerpc/hvcall.h index ab7c3cf2488..d36da61dbc5 100644 --- a/include/asm-ppc64/hvcall.h +++ b/include/asm-powerpc/hvcall.h @@ -1,5 +1,5 @@ -#ifndef _PPC64_HVCALL_H -#define _PPC64_HVCALL_H +#ifndef _ASM_POWERPC_HVCALL_H +#define _ASM_POWERPC_HVCALL_H #define HVSC .long 0x44000022 @@ -138,7 +138,7 @@ long plpar_hcall(unsigned long opcode, */ long plpar_hcall_norets(unsigned long opcode, ...); -/* +/* * Special hcall interface for ibmveth support. * Takes 8 input parms. Returns a rc and stores the * R4 return value in *out1. @@ -153,11 +153,11 @@ long plpar_hcall_8arg_2ret(unsigned long opcode, unsigned long arg7, unsigned long arg8, unsigned long *out1); - + /* plpar_hcall_4out() * - * same as plpar_hcall except with 4 output arguments. - * + * same as plpar_hcall except with 4 output arguments. + * */ long plpar_hcall_4out(unsigned long opcode, unsigned long arg1, @@ -170,4 +170,4 @@ long plpar_hcall_4out(unsigned long opcode, unsigned long *out4); #endif /* __ASSEMBLY__ */ -#endif /* _PPC64_HVCALL_H */ +#endif /* _ASM_POWERPC_HVCALL_H */ diff --git a/include/asm-powerpc/hw_irq.h b/include/asm-powerpc/hw_irq.h index c37b31b9633..26b89d859c5 100644 --- a/include/asm-powerpc/hw_irq.h +++ b/include/asm-powerpc/hw_irq.h @@ -12,7 +12,6 @@ #include <asm/processor.h> extern void timer_interrupt(struct pt_regs *); -extern void ppc_irq_dispatch_handler(struct pt_regs *regs, int irq); #ifdef CONFIG_PPC_ISERIES diff --git a/include/asm-powerpc/irq.h b/include/asm-powerpc/irq.h index b3935ea28ff..c9fbcede0ef 100644 --- a/include/asm-powerpc/irq.h +++ b/include/asm-powerpc/irq.h @@ -429,7 +429,6 @@ extern u64 ppc64_interrupt_controller; #define NR_MASK_WORDS ((NR_IRQS + 31) / 32) /* pedantic: these are long because they are used with set_bit --RR */ extern unsigned long ppc_cached_irq_mask[NR_MASK_WORDS]; -extern unsigned long ppc_lost_interrupts[NR_MASK_WORDS]; extern atomic_t ppc_n_lost_interrupts; #define virt_irq_create_mapping(x) (x) @@ -488,8 +487,8 @@ extern struct thread_info *softirq_ctx[NR_CPUS]; extern void irq_ctx_init(void); extern void call_do_softirq(struct thread_info *tp); -extern int call_handle_IRQ_event(int irq, struct pt_regs *regs, - struct irqaction *action, struct thread_info *tp); +extern int call___do_IRQ(int irq, struct pt_regs *regs, + struct thread_info *tp); #define __ARCH_HAS_DO_SOFTIRQ diff --git a/include/asm-ppc64/lppaca.h b/include/asm-powerpc/lppaca.h index 9e2a6c0649a..c1bedab1515 100644 --- a/include/asm-ppc64/lppaca.h +++ b/include/asm-powerpc/lppaca.h @@ -16,8 +16,8 @@ * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ -#ifndef _ASM_LPPACA_H -#define _ASM_LPPACA_H +#ifndef _ASM_POWERPC_LPPACA_H +#define _ASM_POWERPC_LPPACA_H //============================================================================= // @@ -28,8 +28,7 @@ //---------------------------------------------------------------------------- #include <asm/types.h> -struct lppaca -{ +struct lppaca { //============================================================================= // CACHE_LINE_1 0x0000 - 0x007F Contains read-only data // NOTE: The xDynXyz fields are fields that will be dynamically changed by @@ -129,4 +128,4 @@ struct lppaca u8 pmc_save_area[256]; // PMC interrupt Area x00-xFF }; -#endif /* _ASM_LPPACA_H */ +#endif /* _ASM_POWERPC_LPPACA_H */ diff --git a/include/asm-ppc64/paca.h b/include/asm-powerpc/paca.h index bccacd6aa93..92c765c35bd 100644 --- a/include/asm-ppc64/paca.h +++ b/include/asm-powerpc/paca.h @@ -1,11 +1,8 @@ -#ifndef _PPC64_PACA_H -#define _PPC64_PACA_H - /* - * include/asm-ppc64/paca.h + * include/asm-powerpc/paca.h * - * This control block defines the PACA which defines the processor - * specific data for each logical processor on the system. + * This control block defines the PACA which defines the processor + * specific data for each logical processor on the system. * There are some pointers defined that are utilized by PLIC. * * C 2001 PPC 64 Team, IBM Corp @@ -14,7 +11,9 @@ * modify it under the terms of the GNU General Public License * as published by the Free Software Foundation; either version * 2 of the License, or (at your option) any later version. - */ + */ +#ifndef _ASM_POWERPC_PACA_H +#define _ASM_POWERPC_PACA_H #include <linux/config.h> #include <asm/types.h> @@ -118,4 +117,4 @@ struct paca_struct { extern struct paca_struct paca[]; -#endif /* _PPC64_PACA_H */ +#endif /* _ASM_POWERPC_PACA_H */ diff --git a/include/asm-powerpc/ppc-pci.h b/include/asm-powerpc/ppc-pci.h index 13aacff755f..9896fade98a 100644 --- a/include/asm-powerpc/ppc-pci.h +++ b/include/asm-powerpc/ppc-pci.h @@ -26,6 +26,10 @@ extern unsigned long find_and_init_phbs(void); extern struct pci_dev *ppc64_isabridge_dev; /* may be NULL if no ISA bus */ +/** Bus Unit ID macros; get low and hi 32-bits of the 64-bit BUID */ +#define BUID_HI(buid) ((buid) >> 32) +#define BUID_LO(buid) ((buid) & 0xffffffff) + /* PCI device_node operations */ struct device_node; typedef void *(*traverse_func)(struct device_node *me, void *data); @@ -36,10 +40,6 @@ void pci_devs_phb_init(void); void pci_devs_phb_init_dynamic(struct pci_controller *phb); void __devinit scan_phb(struct pci_controller *hose); -/* PCI address cache management routines */ -void pci_addr_cache_insert_device(struct pci_dev *dev); -void pci_addr_cache_remove_device(struct pci_dev *dev); - /* From rtas_pci.h */ void init_pci_config_tokens (void); unsigned long get_phb_buid (struct device_node *); @@ -52,4 +52,48 @@ extern unsigned long pci_probe_only; extern unsigned long pci_assign_all_buses; extern int pci_read_irq_line(struct pci_dev *pci_dev); +/* ---- EEH internal-use-only related routines ---- */ +#ifdef CONFIG_EEH +/** + * rtas_set_slot_reset -- unfreeze a frozen slot + * + * Clear the EEH-frozen condition on a slot. This routine + * does this by asserting the PCI #RST line for 1/8th of + * a second; this routine will sleep while the adapter is + * being reset. + */ +void rtas_set_slot_reset (struct pci_dn *); + +/** + * eeh_restore_bars - Restore device configuration info. + * + * A reset of a PCI device will clear out its config space. + * This routines will restore the config space for this + * device, and is children, to values previously obtained + * from the firmware. + */ +void eeh_restore_bars(struct pci_dn *); + +/** + * rtas_configure_bridge -- firmware initialization of pci bridge + * + * Ask the firmware to configure all PCI bridges devices + * located behind the indicated node. Required after a + * pci device reset. Does essentially the same hing as + * eeh_restore_bars, but for brdges, and lets firmware + * do the work. + */ +void rtas_configure_bridge(struct pci_dn *); + +int rtas_write_config(struct pci_dn *, int where, int size, u32 val); + +/** + * mark and clear slots: find "partition endpoint" PE and set or + * clear the flags for each subnode of the PE. + */ +void eeh_mark_slot (struct device_node *dn, int mode_flag); +void eeh_clear_slot (struct device_node *dn, int mode_flag); + +#endif + #endif /* _ASM_POWERPC_PPC_PCI_H */ diff --git a/include/asm-powerpc/ppc_asm.h b/include/asm-powerpc/ppc_asm.h index c534ca41224..c27baa0563f 100644 --- a/include/asm-powerpc/ppc_asm.h +++ b/include/asm-powerpc/ppc_asm.h @@ -6,8 +6,13 @@ #include <linux/stringify.h> #include <linux/config.h> +#include <asm/asm-compat.h> -#ifdef __ASSEMBLY__ +#ifndef __ASSEMBLY__ +#error __FILE__ should only be used in assembler files +#else + +#define SZL (BITS_PER_LONG/8) /* * Macros for storing registers into and loading registers from @@ -184,12 +189,6 @@ n: oris reg,reg,(label)@h; \ ori reg,reg,(label)@l; -/* operations for longs and pointers */ -#define LDL ld -#define STL std -#define CMPI cmpdi -#define SZL 8 - /* offsets for stack frame layout */ #define LRSAVE 16 @@ -203,12 +202,6 @@ n: #define OFF(name) name@l -/* operations for longs and pointers */ -#define LDL lwz -#define STL stw -#define CMPI cmpwi -#define SZL 4 - /* offsets for stack frame layout */ #define LRSAVE 4 @@ -266,15 +259,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_601) #endif -#ifdef CONFIG_IBM405_ERR77 -#define PPC405_ERR77(ra,rb) dcbt ra, rb; -#define PPC405_ERR77_SYNC sync; -#else -#define PPC405_ERR77(ra,rb) -#define PPC405_ERR77_SYNC -#endif - - #ifdef CONFIG_IBM440EP_ERR42 #define PPC440EP_ERR42 isync #else @@ -502,17 +486,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_601) #define N_SLINE 68 #define N_SO 100 -#define ASM_CONST(x) x -#else - #define __ASM_CONST(x) x##UL - #define ASM_CONST(x) __ASM_CONST(x) - -#ifdef CONFIG_PPC64 -#define DATAL ".llong" -#else -#define DATAL ".long" -#endif - #endif /* __ASSEMBLY__ */ #endif /* _ASM_POWERPC_PPC_ASM_H */ diff --git a/include/asm-powerpc/processor.h b/include/asm-powerpc/processor.h index 1dc4bf7b52b..f6f186b56b0 100644 --- a/include/asm-powerpc/processor.h +++ b/include/asm-powerpc/processor.h @@ -17,65 +17,71 @@ #include <linux/compiler.h> #include <asm/ptrace.h> #include <asm/types.h> -#ifdef CONFIG_PPC64 -#include <asm/systemcfg.h> -#endif -#ifdef CONFIG_PPC32 -/* 32-bit platform types */ -/* We only need to define a new _MACH_xxx for machines which are part of - * a configuration which supports more than one type of different machine. - * This is currently limited to CONFIG_PPC_MULTIPLATFORM and CHRP/PReP/PMac. - * -- Tom +/* We do _not_ want to define new machine types at all, those must die + * in favor of using the device-tree + * -- BenH. */ -#define _MACH_prep 0x00000001 -#define _MACH_Pmac 0x00000002 /* pmac or pmac clone (non-chrp) */ -#define _MACH_chrp 0x00000004 /* chrp machine */ -/* see residual.h for these */ +/* Platforms codes (to be obsoleted) */ +#define PLATFORM_PSERIES 0x0100 +#define PLATFORM_PSERIES_LPAR 0x0101 +#define PLATFORM_ISERIES_LPAR 0x0201 +#define PLATFORM_LPAR 0x0001 +#define PLATFORM_POWERMAC 0x0400 +#define PLATFORM_MAPLE 0x0500 +#define PLATFORM_PREP 0x0600 +#define PLATFORM_CHRP 0x0700 +#define PLATFORM_CELL 0x1000 + +/* Compat platform codes for 32 bits */ +#define _MACH_prep PLATFORM_PREP +#define _MACH_Pmac PLATFORM_POWERMAC +#define _MACH_chrp PLATFORM_CHRP + +/* PREP sub-platform types see residual.h for these */ #define _PREP_Motorola 0x01 /* motorola prep */ #define _PREP_Firm 0x02 /* firmworks prep */ #define _PREP_IBM 0x00 /* ibm prep */ #define _PREP_Bull 0x03 /* bull prep */ -/* these are arbitrary */ +/* CHRP sub-platform types. These are arbitrary */ #define _CHRP_Motorola 0x04 /* motorola chrp, the cobra */ #define _CHRP_IBM 0x05 /* IBM chrp, the longtrail and longtrail 2 */ #define _CHRP_Pegasos 0x06 /* Genesi/bplan's Pegasos and Pegasos2 */ -#ifdef CONFIG_PPC_MULTIPLATFORM +#define platform_is_pseries() (_machine == PLATFORM_PSERIES || \ + _machine == PLATFORM_PSERIES_LPAR) +#define platform_is_lpar() (!!(_machine & PLATFORM_LPAR)) + +#if defined(CONFIG_PPC_MULTIPLATFORM) extern int _machine; +#ifdef CONFIG_PPC32 + /* what kind of prep workstation we are */ extern int _prep_type; extern int _chrp_type; /* * This is used to identify the board type from a given PReP board - * vendor. Board revision is also made available. + * vendor. Board revision is also made available. This will be moved + * elsewhere soon */ extern unsigned char ucSystemType; extern unsigned char ucBoardRev; extern unsigned char ucBoardRevMaj, ucBoardRevMin; + +#endif /* CONFIG_PPC32 */ + +#elif defined(CONFIG_PPC_ISERIES) +/* + * iSeries is soon to become MULTIPLATFORM hopefully ... + */ +#define _machine PLATFORM_ISERIES_LPAR #else #define _machine 0 #endif /* CONFIG_PPC_MULTIPLATFORM */ -#endif /* CONFIG_PPC32 */ - -#ifdef CONFIG_PPC64 -/* Platforms supported by PPC64 */ -#define PLATFORM_PSERIES 0x0100 -#define PLATFORM_PSERIES_LPAR 0x0101 -#define PLATFORM_ISERIES_LPAR 0x0201 -#define PLATFORM_LPAR 0x0001 -#define PLATFORM_POWERMAC 0x0400 -#define PLATFORM_MAPLE 0x0500 -#define PLATFORM_CELL 0x1000 - -/* Compatibility with drivers coming from PPC32 world */ -#define _machine (systemcfg->platform) -#define _MACH_Pmac PLATFORM_POWERMAC -#endif /* * Default implementation of macro that returns current diff --git a/include/asm-powerpc/reg.h b/include/asm-powerpc/reg.h index 489cf4c99c2..eb392d038ed 100644 --- a/include/asm-powerpc/reg.h +++ b/include/asm-powerpc/reg.h @@ -16,7 +16,11 @@ /* Pickup Book E specific registers. */ #if defined(CONFIG_BOOKE) || defined(CONFIG_40x) #include <asm/reg_booke.h> -#endif +#endif /* CONFIG_BOOKE || CONFIG_40x */ + +#ifdef CONFIG_8xx +#include <asm/reg_8xx.h> +#endif /* CONFIG_8xx */ #define MSR_SF_LG 63 /* Enable 64 bit mode */ #define MSR_ISF_LG 61 /* Interrupt 64b mode valid on 630 */ @@ -359,6 +363,7 @@ #define SPRN_RPA 0x3D6 /* Required Physical Address Register */ #define SPRN_SDA 0x3BF /* Sampled Data Address Register */ #define SPRN_SDR1 0x019 /* MMU Hash Base Register */ +#define SPRN_ASR 0x118 /* Address Space Register */ #define SPRN_SIA 0x3BB /* Sampled Instruction Address Register */ #define SPRN_SPRG0 0x110 /* Special Purpose Register General 0 */ #define SPRN_SPRG1 0x111 /* Special Purpose Register General 1 */ diff --git a/include/asm-ppc/cache.h b/include/asm-powerpc/reg_8xx.h index 7a157d0f4b5..e8ea346b21d 100644 --- a/include/asm-ppc/cache.h +++ b/include/asm-powerpc/reg_8xx.h @@ -1,49 +1,9 @@ /* - * include/asm-ppc/cache.h + * Contains register definitions common to PowerPC 8xx CPUs. Notice */ -#ifdef __KERNEL__ -#ifndef __ARCH_PPC_CACHE_H -#define __ARCH_PPC_CACHE_H +#ifndef _ASM_POWERPC_REG_8xx_H +#define _ASM_POWERPC_REG_8xx_H -#include <linux/config.h> - -/* bytes per L1 cache line */ -#if defined(CONFIG_8xx) || defined(CONFIG_403GCX) -#define L1_CACHE_SHIFT 4 -#define MAX_COPY_PREFETCH 1 -#elif defined(CONFIG_PPC64BRIDGE) -#define L1_CACHE_SHIFT 7 -#define MAX_COPY_PREFETCH 1 -#else -#define L1_CACHE_SHIFT 5 -#define MAX_COPY_PREFETCH 4 -#endif - -#define L1_CACHE_BYTES (1 << L1_CACHE_SHIFT) - -#define SMP_CACHE_BYTES L1_CACHE_BYTES -#define L1_CACHE_SHIFT_MAX 7 /* largest L1 which this arch supports */ - -#define L1_CACHE_ALIGN(x) (((x)+(L1_CACHE_BYTES-1))&~(L1_CACHE_BYTES-1)) -#define L1_CACHE_PAGES 8 - -#ifndef __ASSEMBLY__ -extern void clean_dcache_range(unsigned long start, unsigned long stop); -extern void flush_dcache_range(unsigned long start, unsigned long stop); -extern void invalidate_dcache_range(unsigned long start, unsigned long stop); -extern void flush_dcache_all(void); -#endif /* __ASSEMBLY__ */ - -/* prep registers for L2 */ -#define CACHECRBA 0x80000823 /* Cache configuration register address */ -#define L2CACHE_MASK 0x03 /* Mask for 2 L2 Cache bits */ -#define L2CACHE_512KB 0x00 /* 512KB */ -#define L2CACHE_256KB 0x01 /* 256KB */ -#define L2CACHE_1MB 0x02 /* 1MB */ -#define L2CACHE_NONE 0x03 /* NONE */ -#define L2CACHE_PARITY 0x08 /* Mask for L2 Cache Parity Protected bit */ - -#ifdef CONFIG_8xx /* Cache control on the MPC8xx is provided through some additional * special purpose registers. */ @@ -78,7 +38,5 @@ extern void flush_dcache_all(void); #define DC_DFWT 0x40000000 /* Data cache is forced write through */ #define DC_LES 0x20000000 /* Caches are little endian mode */ -#endif /* CONFIG_8xx */ -#endif -#endif /* __KERNEL__ */ +#endif /* _ASM_POWERPC_REG_8xx_H */ diff --git a/include/asm-ppc/signal.h b/include/asm-powerpc/signal.h index caf6ede3710..694c8d2dab8 100644 --- a/include/asm-ppc/signal.h +++ b/include/asm-powerpc/signal.h @@ -1,18 +1,11 @@ -#ifndef _ASMPPC_SIGNAL_H -#define _ASMPPC_SIGNAL_H +#ifndef _ASM_POWERPC_SIGNAL_H +#define _ASM_POWERPC_SIGNAL_H -#ifdef __KERNEL__ #include <linux/types.h> -#endif /* __KERNEL__ */ - -/* Avoid too many header ordering problems. */ -struct siginfo; - -/* Most things should be clean enough to redefine this at will, if care - is taken to make libc match. */ +#include <linux/config.h> #define _NSIG 64 -#define _NSIG_BPW 32 +#define _NSIG_BPW BITS_PER_LONG #define _NSIG_WORDS (_NSIG / _NSIG_BPW) typedef unsigned long old_sigset_t; /* at least 32 bits */ @@ -77,19 +70,19 @@ typedef struct { * SA_ONESHOT and SA_NOMASK are the historical Linux names for the Single * Unix names RESETHAND and NODEFER respectively. */ -#define SA_NOCLDSTOP 0x00000001 -#define SA_NOCLDWAIT 0x00000002 -#define SA_SIGINFO 0x00000004 -#define SA_ONSTACK 0x08000000 -#define SA_RESTART 0x10000000 -#define SA_NODEFER 0x40000000 -#define SA_RESETHAND 0x80000000 +#define SA_NOCLDSTOP 0x00000001U +#define SA_NOCLDWAIT 0x00000002U +#define SA_SIGINFO 0x00000004U +#define SA_ONSTACK 0x08000000U +#define SA_RESTART 0x10000000U +#define SA_NODEFER 0x40000000U +#define SA_RESETHAND 0x80000000U #define SA_NOMASK SA_NODEFER #define SA_ONESHOT SA_RESETHAND -#define SA_INTERRUPT 0x20000000 /* dummy -- ignored */ +#define SA_INTERRUPT 0x20000000u /* dummy -- ignored */ -#define SA_RESTORER 0x04000000 +#define SA_RESTORER 0x04000000U /* * sigaltstack controls @@ -127,10 +120,13 @@ typedef struct sigaltstack { } stack_t; #ifdef __KERNEL__ -#include <asm/sigcontext.h> +struct pt_regs; +extern int do_signal(sigset_t *oldset, struct pt_regs *regs); +extern int do_signal32(sigset_t *oldset, struct pt_regs *regs); #define ptrace_signal_deliver(regs, cookie) do { } while (0) #endif /* __KERNEL__ */ +#ifndef __powerpc64__ /* * These are parameters to dbg_sigreturn syscall. They enable or * disable certain debugging things that can be done from signal @@ -149,5 +145,6 @@ struct sig_dbg_op { /* Enable or disable branch tracing. The value sets the state. */ #define SIG_DBG_BRANCH_TRACING 2 +#endif /* ! __powerpc64__ */ -#endif +#endif /* _ASM_POWERPC_SIGNAL_H */ diff --git a/include/asm-powerpc/sparsemem.h b/include/asm-powerpc/sparsemem.h index 1c95ab99deb..58d2aab416f 100644 --- a/include/asm-powerpc/sparsemem.h +++ b/include/asm-powerpc/sparsemem.h @@ -11,6 +11,10 @@ #define MAX_PHYSADDR_BITS 38 #define MAX_PHYSMEM_BITS 36 +#ifdef CONFIG_MEMORY_HOTPLUG +extern void create_section_mapping(unsigned long start, unsigned long end); +#endif /* CONFIG_MEMORY_HOTPLUG */ + #endif /* CONFIG_SPARSEMEM */ #endif /* _ASM_POWERPC_SPARSEMEM_H */ diff --git a/include/asm-powerpc/system.h b/include/asm-powerpc/system.h index 3536a5cd7a2..5341b75c75c 100644 --- a/include/asm-powerpc/system.h +++ b/include/asm-powerpc/system.h @@ -8,7 +8,6 @@ #include <linux/kernel.h> #include <asm/hw_irq.h> -#include <asm/ppc_asm.h> #include <asm/atomic.h> /* @@ -180,6 +179,7 @@ extern struct task_struct *_switch(struct thread_struct *prev, extern unsigned int rtas_data; extern int mem_init_done; /* set on boot once kmalloc can be called */ extern unsigned long memory_limit; +extern unsigned long klimit; extern int powersave_nap; /* set if nap mode can be used in idle loop */ diff --git a/include/asm-ppc64/systemcfg.h b/include/asm-powerpc/systemcfg.h index 9b86b53129a..36b5cbe466f 100644 --- a/include/asm-ppc64/systemcfg.h +++ b/include/asm-powerpc/systemcfg.h @@ -1,7 +1,7 @@ #ifndef _SYSTEMCFG_H #define _SYSTEMCFG_H -/* +/* * Copyright (C) 2002 Peter Bergner <bergner@vnet.ibm.com>, IBM * * This program is free software; you can redistribute it and/or @@ -12,7 +12,7 @@ /* Change Activity: * 2002/09/30 : bergner : Created - * End Change Activity + * End Change Activity */ /* @@ -56,7 +56,7 @@ struct systemcfg { }; #ifdef __KERNEL__ -extern struct systemcfg *systemcfg; +extern struct systemcfg *_systemcfg; /* to be renamed */ #endif #endif /* __ASSEMBLY__ */ diff --git a/include/asm-ppc64/tce.h b/include/asm-powerpc/tce.h index d40b6b42ab3..d099d5200f9 100644 --- a/include/asm-ppc64/tce.h +++ b/include/asm-powerpc/tce.h @@ -18,8 +18,8 @@ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ -#ifndef _ASM_TCE_H -#define _ASM_TCE_H +#ifndef _ASM_POWERPC_TCE_H +#define _ASM_POWERPC_TCE_H /* * Tces come in two formats, one for the virtual bus and a different @@ -61,4 +61,4 @@ union tce_entry { }; -#endif +#endif /* _ASM_POWERPC_TCE_H */ diff --git a/include/asm-powerpc/uaccess.h b/include/asm-powerpc/uaccess.h index 33af730f0d1..3872e924cdd 100644 --- a/include/asm-powerpc/uaccess.h +++ b/include/asm-powerpc/uaccess.h @@ -120,14 +120,6 @@ struct exception_table_entry { extern long __put_user_bad(void); -#ifdef __powerpc64__ -#define __EX_TABLE_ALIGN "3" -#define __EX_TABLE_TYPE "llong" -#else -#define __EX_TABLE_ALIGN "2" -#define __EX_TABLE_TYPE "long" -#endif - /* * We don't tell gcc that we are accessing memory, but this is OK * because we do not write to any memory gcc knows about, so there @@ -142,11 +134,12 @@ extern long __put_user_bad(void); " b 2b\n" \ ".previous\n" \ ".section __ex_table,\"a\"\n" \ - " .align " __EX_TABLE_ALIGN "\n" \ - " ."__EX_TABLE_TYPE" 1b,3b\n" \ + " .balign %5\n" \ + PPC_LONG "1b,3b\n" \ ".previous" \ : "=r" (err) \ - : "r" (x), "b" (addr), "i" (-EFAULT), "0" (err)) + : "r" (x), "b" (addr), "i" (-EFAULT), "0" (err),\ + "i"(sizeof(unsigned long))) #ifdef __powerpc64__ #define __put_user_asm2(x, ptr, retval) \ @@ -162,12 +155,13 @@ extern long __put_user_bad(void); " b 3b\n" \ ".previous\n" \ ".section __ex_table,\"a\"\n" \ - " .align " __EX_TABLE_ALIGN "\n" \ - " ." __EX_TABLE_TYPE " 1b,4b\n" \ - " ." __EX_TABLE_TYPE " 2b,4b\n" \ + " .balign %5\n" \ + PPC_LONG "1b,4b\n" \ + PPC_LONG "2b,4b\n" \ ".previous" \ : "=r" (err) \ - : "r" (x), "b" (addr), "i" (-EFAULT), "0" (err)) + : "r" (x), "b" (addr), "i" (-EFAULT), "0" (err),\ + "i"(sizeof(unsigned long))) #endif /* __powerpc64__ */ #define __put_user_size(x, ptr, size, retval) \ @@ -213,11 +207,12 @@ extern long __get_user_bad(void); " b 2b\n" \ ".previous\n" \ ".section __ex_table,\"a\"\n" \ - " .align "__EX_TABLE_ALIGN "\n" \ - " ." __EX_TABLE_TYPE " 1b,3b\n" \ + " .balign %5\n" \ + PPC_LONG "1b,3b\n" \ ".previous" \ : "=r" (err), "=r" (x) \ - : "b" (addr), "i" (-EFAULT), "0" (err)) + : "b" (addr), "i" (-EFAULT), "0" (err), \ + "i"(sizeof(unsigned long))) #ifdef __powerpc64__ #define __get_user_asm2(x, addr, err) \ @@ -235,12 +230,13 @@ extern long __get_user_bad(void); " b 3b\n" \ ".previous\n" \ ".section __ex_table,\"a\"\n" \ - " .align " __EX_TABLE_ALIGN "\n" \ - " ." __EX_TABLE_TYPE " 1b,4b\n" \ - " ." __EX_TABLE_TYPE " 2b,4b\n" \ + " .balign %5\n" \ + PPC_LONG "1b,4b\n" \ + PPC_LONG "2b,4b\n" \ ".previous" \ : "=r" (err), "=&r" (x) \ - : "b" (addr), "i" (-EFAULT), "0" (err)) + : "b" (addr), "i" (-EFAULT), "0" (err), \ + "i"(sizeof(unsigned long))) #endif /* __powerpc64__ */ #define __get_user_size(x, ptr, size, retval) \ diff --git a/include/asm-powerpc/xmon.h b/include/asm-powerpc/xmon.h index ace2072d4a8..43f7129984c 100644 --- a/include/asm-powerpc/xmon.h +++ b/include/asm-powerpc/xmon.h @@ -7,7 +7,6 @@ struct pt_regs; extern int xmon(struct pt_regs *excp); extern void xmon_printf(const char *fmt, ...); extern void xmon_init(int); -extern void xmon_map_scc(void); #endif #endif diff --git a/include/asm-ppc/cacheflush.h b/include/asm-ppc/cacheflush.h deleted file mode 100644 index 6a243efb331..00000000000 --- a/include/asm-ppc/cacheflush.h +++ /dev/null @@ -1,49 +0,0 @@ -/* - * include/asm-ppc/cacheflush.h - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ -#ifdef __KERNEL__ -#ifndef _PPC_CACHEFLUSH_H -#define _PPC_CACHEFLUSH_H - -#include <linux/mm.h> - -/* - * No cache flushing is required when address mappings are - * changed, because the caches on PowerPCs are physically - * addressed. -- paulus - * Also, when SMP we use the coherency (M) bit of the - * BATs and PTEs. -- Cort - */ -#define flush_cache_all() do { } while (0) -#define flush_cache_mm(mm) do { } while (0) -#define flush_cache_range(vma, a, b) do { } while (0) -#define flush_cache_page(vma, p, pfn) do { } while (0) -#define flush_icache_page(vma, page) do { } while (0) -#define flush_cache_vmap(start, end) do { } while (0) -#define flush_cache_vunmap(start, end) do { } while (0) - -extern void flush_dcache_page(struct page *page); -#define flush_dcache_mmap_lock(mapping) do { } while (0) -#define flush_dcache_mmap_unlock(mapping) do { } while (0) - -extern void flush_icache_range(unsigned long, unsigned long); -extern void flush_icache_user_range(struct vm_area_struct *vma, - struct page *page, unsigned long addr, int len); - -#define copy_to_user_page(vma, page, vaddr, dst, src, len) \ -do { memcpy(dst, src, len); \ - flush_icache_user_range(vma, page, vaddr, len); \ -} while (0) -#define copy_from_user_page(vma, page, vaddr, dst, src, len) \ - memcpy(dst, src, len) - -extern void __flush_dcache_icache(void *page_va); -extern void __flush_dcache_icache_phys(unsigned long physaddr); -extern void flush_dcache_icache_page(struct page *page); -#endif /* _PPC_CACHEFLUSH_H */ -#endif /* __KERNEL__ */ diff --git a/include/asm-ppc/current.h b/include/asm-ppc/current.h deleted file mode 100644 index 8d41501ba10..00000000000 --- a/include/asm-ppc/current.h +++ /dev/null @@ -1,11 +0,0 @@ -#ifdef __KERNEL__ -#ifndef _PPC_CURRENT_H -#define _PPC_CURRENT_H - -/* - * We keep `current' in r2 for speed. - */ -register struct task_struct *current asm ("r2"); - -#endif /* !(_PPC_CURRENT_H) */ -#endif /* __KERNEL__ */ diff --git a/include/asm-ppc64/cache.h b/include/asm-ppc64/cache.h deleted file mode 100644 index 92140a7efbd..00000000000 --- a/include/asm-ppc64/cache.h +++ /dev/null @@ -1,36 +0,0 @@ -/* - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ -#ifndef __ARCH_PPC64_CACHE_H -#define __ARCH_PPC64_CACHE_H - -#include <asm/types.h> - -/* bytes per L1 cache line */ -#define L1_CACHE_SHIFT 7 -#define L1_CACHE_BYTES (1 << L1_CACHE_SHIFT) - -#define SMP_CACHE_BYTES L1_CACHE_BYTES -#define L1_CACHE_SHIFT_MAX 7 /* largest L1 which this arch supports */ - -#ifndef __ASSEMBLY__ - -struct ppc64_caches { - u32 dsize; /* L1 d-cache size */ - u32 dline_size; /* L1 d-cache line size */ - u32 log_dline_size; - u32 dlines_per_page; - u32 isize; /* L1 i-cache size */ - u32 iline_size; /* L1 i-cache line size */ - u32 log_iline_size; - u32 ilines_per_page; -}; - -extern struct ppc64_caches ppc64_caches; - -#endif - -#endif diff --git a/include/asm-ppc64/current.h b/include/asm-ppc64/current.h deleted file mode 100644 index 52ddc60c8b6..00000000000 --- a/include/asm-ppc64/current.h +++ /dev/null @@ -1,16 +0,0 @@ -#ifndef _PPC64_CURRENT_H -#define _PPC64_CURRENT_H - -#include <asm/paca.h> - -/* - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#define get_current() (get_paca()->__current) -#define current get_current() - -#endif /* !(_PPC64_CURRENT_H) */ diff --git a/include/asm-ppc64/eeh.h b/include/asm-ppc64/eeh.h index 40c8eb57493..89f26ab3190 100644 --- a/include/asm-ppc64/eeh.h +++ b/include/asm-ppc64/eeh.h @@ -1,4 +1,4 @@ -/* +/* * eeh.h * Copyright (C) 2001 Dave Engebretsen & Todd Inglett IBM Corporation. * @@ -6,12 +6,12 @@ * it under the terms of the GNU General Public License as published by * the Free Software Foundation; either version 2 of the License, or * (at your option) any later version. - * + * * This program is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * GNU General Public License for more details. - * + * * You should have received a copy of the GNU General Public License * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA @@ -27,8 +27,6 @@ struct pci_dev; struct device_node; -struct device_node; -struct notifier_block; #ifdef CONFIG_EEH @@ -37,6 +35,10 @@ struct notifier_block; #define EEH_MODE_NOCHECK (1<<1) #define EEH_MODE_ISOLATED (1<<2) +/* Max number of EEH freezes allowed before we consider the device + * to be permanently disabled. */ +#define EEH_MAX_ALLOWED_FREEZES 5 + void __init eeh_init(void); unsigned long eeh_check_failure(const volatile void __iomem *token, unsigned long val); @@ -59,36 +61,14 @@ void eeh_add_device_late(struct pci_dev *); * eeh_remove_device - undo EEH setup for the indicated pci device * @dev: pci device to be removed * - * This routine should be when a device is removed from a running - * system (e.g. by hotplug or dlpar). + * This routine should be called when a device is removed from + * a running system (e.g. by hotplug or dlpar). It unregisters + * the PCI device from the EEH subsystem. I/O errors affecting + * this device will no longer be detected after this call; thus, + * i/o errors affecting this slot may leave this device unusable. */ void eeh_remove_device(struct pci_dev *); -#define EEH_DISABLE 0 -#define EEH_ENABLE 1 -#define EEH_RELEASE_LOADSTORE 2 -#define EEH_RELEASE_DMA 3 - -/** - * Notifier event flags. - */ -#define EEH_NOTIFY_FREEZE 1 - -/** EEH event -- structure holding pci slot data that describes - * a change in the isolation status of a PCI slot. A pointer - * to this struct is passed as the data pointer in a notify callback. - */ -struct eeh_event { - struct list_head list; - struct pci_dev *dev; - struct device_node *dn; - int reset_state; -}; - -/** Register to find out about EEH events. */ -int eeh_register_notifier(struct notifier_block *nb); -int eeh_unregister_notifier(struct notifier_block *nb); - /** * EEH_POSSIBLE_ERROR() -- test for possible MMIO failure. * @@ -129,7 +109,7 @@ static inline void eeh_remove_device(struct pci_dev *dev) { } #define EEH_IO_ERROR_VALUE(size) (-1UL) #endif /* CONFIG_EEH */ -/* +/* * MMIO read/write operations with EEH support. */ static inline u8 eeh_readb(const volatile void __iomem *addr) diff --git a/include/asm-ppc64/mmu.h b/include/asm-ppc64/mmu.h index 4c18a5cb69f..1a7e0afa2dc 100644 --- a/include/asm-ppc64/mmu.h +++ b/include/asm-ppc64/mmu.h @@ -14,7 +14,7 @@ #define _PPC64_MMU_H_ #include <linux/config.h> -#include <asm/ppc_asm.h> /* for ASM_CONST */ +#include <asm/asm-compat.h> #include <asm/page.h> /* @@ -224,9 +224,12 @@ extern int htab_bolt_mapping(unsigned long vstart, unsigned long vend, unsigned long pstart, unsigned long mode, int psize); +extern void htab_initialize(void); +extern void htab_initialize_secondary(void); extern void hpte_init_native(void); extern void hpte_init_lpar(void); extern void hpte_init_iSeries(void); +extern void mm_init_ppc64(void); extern long pSeries_lpar_hpte_insert(unsigned long hpte_group, unsigned long va, unsigned long prpn, @@ -245,6 +248,7 @@ extern long iSeries_hpte_insert(unsigned long hpte_group, extern void stabs_alloc(void); extern void slb_initialize(void); +extern void stab_initialize(unsigned long stab); #endif /* __ASSEMBLY__ */ diff --git a/include/asm-ppc64/mmzone.h b/include/asm-ppc64/mmzone.h index 80a708e7093..15e777ce0f4 100644 --- a/include/asm-ppc64/mmzone.h +++ b/include/asm-ppc64/mmzone.h @@ -33,6 +33,9 @@ extern int numa_cpu_lookup_table[]; extern char *numa_memory_lookup_table; extern cpumask_t numa_cpumask_lookup_table[]; extern int nr_cpus_in_node[]; +#ifdef CONFIG_MEMORY_HOTPLUG +extern unsigned long max_pfn; +#endif /* 16MB regions */ #define MEMORY_INCREMENT_SHIFT 24 @@ -45,6 +48,11 @@ static inline int pa_to_nid(unsigned long pa) { int nid; +#ifdef CONFIG_MEMORY_HOTPLUG + /* kludge hot added sections default to node 0 */ + if (pa >= (max_pfn << PAGE_SHIFT)) + return 0; +#endif nid = numa_memory_lookup_table[pa >> MEMORY_INCREMENT_SHIFT]; #ifdef DEBUG_NUMA diff --git a/include/asm-ppc64/page.h b/include/asm-ppc64/page.h index 82ce187e5be..e32f1187aa2 100644 --- a/include/asm-ppc64/page.h +++ b/include/asm-ppc64/page.h @@ -11,7 +11,7 @@ */ #include <linux/config.h> -#include <asm/ppc_asm.h> /* for ASM_CONST */ +#include <asm/asm-compat.h> /* * We support either 4k or 64k software page size. When using 64k pages diff --git a/include/asm-ppc64/pci-bridge.h b/include/asm-ppc64/pci-bridge.h index 60cf8c838af..efbdaece0cf 100644 --- a/include/asm-ppc64/pci-bridge.h +++ b/include/asm-ppc64/pci-bridge.h @@ -63,7 +63,6 @@ struct pci_dn { int devfn; /* for pci devices */ int eeh_mode; /* See eeh.h for possible EEH_MODEs */ int eeh_config_addr; - int eeh_capable; /* from firmware */ int eeh_check_count; /* # times driver ignored error */ int eeh_freeze_count; /* # times this device froze up. */ int eeh_is_bridge; /* device is pci-to-pci bridge */ diff --git a/include/asm-ppc64/pgalloc.h b/include/asm-ppc64/pgalloc.h index 98da0e4262b..dcf3622d194 100644 --- a/include/asm-ppc64/pgalloc.h +++ b/include/asm-ppc64/pgalloc.h @@ -10,8 +10,8 @@ extern kmem_cache_t *pgtable_cache[]; #ifdef CONFIG_PPC_64K_PAGES #define PTE_CACHE_NUM 0 -#define PMD_CACHE_NUM 0 -#define PGD_CACHE_NUM 1 +#define PMD_CACHE_NUM 1 +#define PGD_CACHE_NUM 2 #else #define PTE_CACHE_NUM 0 #define PMD_CACHE_NUM 1 diff --git a/include/asm-ppc64/prom.h b/include/asm-ppc64/prom.h index 76bb0266d67..ddfe186589f 100644 --- a/include/asm-ppc64/prom.h +++ b/include/asm-ppc64/prom.h @@ -204,6 +204,8 @@ extern void of_detach_node(const struct device_node *); extern unsigned long prom_init(unsigned long, unsigned long, unsigned long, unsigned long, unsigned long); extern void finish_device_tree(void); +extern void unflatten_device_tree(void); +extern void early_init_devtree(void *); extern int device_is_compatible(struct device_node *device, const char *); extern int machine_is_compatible(const char *compat); extern unsigned char *get_property(struct device_node *node, const char *name, diff --git a/include/asm-ppc64/signal.h b/include/asm-ppc64/signal.h deleted file mode 100644 index 432df7dd355..00000000000 --- a/include/asm-ppc64/signal.h +++ /dev/null @@ -1,132 +0,0 @@ -#ifndef _ASMPPC64_SIGNAL_H -#define _ASMPPC64_SIGNAL_H - -#include <linux/types.h> -#include <linux/compiler.h> -#include <asm/siginfo.h> - -/* Avoid too many header ordering problems. */ -struct siginfo; - -#define _NSIG 64 -#define _NSIG_BPW 64 -#define _NSIG_WORDS (_NSIG / _NSIG_BPW) - -typedef unsigned long old_sigset_t; /* at least 32 bits */ - -typedef struct { - unsigned long sig[_NSIG_WORDS]; -} sigset_t; - -#define SIGHUP 1 -#define SIGINT 2 -#define SIGQUIT 3 -#define SIGILL 4 -#define SIGTRAP 5 -#define SIGABRT 6 -#define SIGIOT 6 -#define SIGBUS 7 -#define SIGFPE 8 -#define SIGKILL 9 -#define SIGUSR1 10 -#define SIGSEGV 11 -#define SIGUSR2 12 -#define SIGPIPE 13 -#define SIGALRM 14 -#define SIGTERM 15 -#define SIGSTKFLT 16 -#define SIGCHLD 17 -#define SIGCONT 18 -#define SIGSTOP 19 -#define SIGTSTP 20 -#define SIGTTIN 21 -#define SIGTTOU 22 -#define SIGURG 23 -#define SIGXCPU 24 -#define SIGXFSZ 25 -#define SIGVTALRM 26 -#define SIGPROF 27 -#define SIGWINCH 28 -#define SIGIO 29 -#define SIGPOLL SIGIO -/* -#define SIGLOST 29 -*/ -#define SIGPWR 30 -#define SIGSYS 31 -#define SIGUNUSED 31 - -/* These should not be considered constants from userland. */ -#define SIGRTMIN 32 -#define SIGRTMAX _NSIG - -/* - * SA_FLAGS values: - * - * SA_ONSTACK is not currently supported, but will allow sigaltstack(2). - * SA_INTERRUPT is a no-op, but left due to historical reasons. Use the - * SA_RESTART flag to get restarting signals (which were the default long ago) - * SA_NOCLDSTOP flag to turn off SIGCHLD when children stop. - * SA_RESETHAND clears the handler when the signal is delivered. - * SA_NOCLDWAIT flag on SIGCHLD to inhibit zombies. - * SA_NODEFER prevents the current signal from being masked in the handler. - * - * SA_ONESHOT and SA_NOMASK are the historical Linux names for the Single - * Unix names RESETHAND and NODEFER respectively. - */ -#define SA_NOCLDSTOP 0x00000001u -#define SA_NOCLDWAIT 0x00000002u -#define SA_SIGINFO 0x00000004u -#define SA_ONSTACK 0x08000000u -#define SA_RESTART 0x10000000u -#define SA_NODEFER 0x40000000u -#define SA_RESETHAND 0x80000000u - -#define SA_NOMASK SA_NODEFER -#define SA_ONESHOT SA_RESETHAND -#define SA_INTERRUPT 0x20000000u /* dummy -- ignored */ - -#define SA_RESTORER 0x04000000u - -/* - * sigaltstack controls - */ -#define SS_ONSTACK 1 -#define SS_DISABLE 2 - -#define MINSIGSTKSZ 2048 -#define SIGSTKSZ 8192 - -#include <asm-generic/signal.h> - -struct old_sigaction { - __sighandler_t sa_handler; - old_sigset_t sa_mask; - unsigned long sa_flags; - __sigrestore_t sa_restorer; -}; - -struct sigaction { - __sighandler_t sa_handler; - unsigned long sa_flags; - __sigrestore_t sa_restorer; - sigset_t sa_mask; /* mask last for extensibility */ -}; - -struct k_sigaction { - struct sigaction sa; -}; - -typedef struct sigaltstack { - void __user *ss_sp; - int ss_flags; - size_t ss_size; -} stack_t; - -struct pt_regs; -struct timespec; -extern int do_signal(sigset_t *oldset, struct pt_regs *regs); -extern int do_signal32(sigset_t *oldset, struct pt_regs *regs); -#define ptrace_signal_deliver(regs, cookie) do { } while (0) - -#endif /* _ASMPPC64_SIGNAL_H */ diff --git a/include/asm-ppc64/system.h b/include/asm-ppc64/system.h index 0cdd66c9f4b..bf9a6aba19c 100644 --- a/include/asm-ppc64/system.h +++ b/include/asm-ppc64/system.h @@ -149,6 +149,8 @@ struct thread_struct; extern struct task_struct * _switch(struct thread_struct *prev, struct thread_struct *next); +extern unsigned long klimit; + extern int powersave_nap; /* set if nap mode can be used in idle loop */ /* diff --git a/include/asm-sh/ide.h b/include/asm-sh/ide.h index f42cf3977a5..711dad4cb48 100644 --- a/include/asm-sh/ide.h +++ b/include/asm-sh/ide.h @@ -16,10 +16,6 @@ #include <linux/config.h> -#ifndef MAX_HWIFS -#define MAX_HWIFS CONFIG_IDE_MAX_HWIFS -#endif - #define ide_default_io_ctl(base) (0) #include <asm-generic/ide_iops.h> diff --git a/include/asm-sh64/ide.h b/include/asm-sh64/ide.h index 6fd514daa1b..852f50afe39 100644 --- a/include/asm-sh64/ide.h +++ b/include/asm-sh64/ide.h @@ -17,10 +17,6 @@ #include <linux/config.h> -#ifndef MAX_HWIFS -#define MAX_HWIFS CONFIG_IDE_MAX_HWIFS -#endif - /* Without this, the initialisation of PCI IDE cards end up calling * ide_init_hwif_ports, which won't work. */ #ifdef CONFIG_BLK_DEV_IDEPCI diff --git a/include/asm-x86_64/msi.h b/include/asm-x86_64/msi.h index 85c427e472b..356e0e82f50 100644 --- a/include/asm-x86_64/msi.h +++ b/include/asm-x86_64/msi.h @@ -11,8 +11,6 @@ #include <asm/smp.h> #define LAST_DEVICE_VECTOR 232 -#define MSI_DEST_MODE MSI_LOGICAL_MODE -#define MSI_TARGET_CPU_SHIFT 12 -#define MSI_TARGET_CPU logical_smp_processor_id() +#define MSI_TARGET_CPU_SHIFT 12 #endif /* ASM_MSI_H */ diff --git a/include/asm-x86_64/smp.h b/include/asm-x86_64/smp.h index c57ce407134..b9fb2173ef9 100644 --- a/include/asm-x86_64/smp.h +++ b/include/asm-x86_64/smp.h @@ -135,5 +135,11 @@ static __inline int logical_smp_processor_id(void) } #endif +#ifdef CONFIG_SMP +#define cpu_physical_id(cpu) x86_cpu_to_apicid[cpu] +#else +#define cpu_physical_id(cpu) boot_cpu_id +#endif + #endif diff --git a/include/linux/connector.h b/include/linux/connector.h index c5769c6585f..ad1a22c1c42 100644 --- a/include/linux/connector.h +++ b/include/linux/connector.h @@ -32,6 +32,8 @@ */ #define CN_IDX_PROC 0x1 #define CN_VAL_PROC 0x1 +#define CN_IDX_CIFS 0x2 +#define CN_VAL_CIFS 0x1 #define CN_NETLINK_USERS 1 diff --git a/include/linux/genetlink.h b/include/linux/genetlink.h new file mode 100644 index 00000000000..84f12a41dc0 --- /dev/null +++ b/include/linux/genetlink.h @@ -0,0 +1,51 @@ +#ifndef __LINUX_GENERIC_NETLINK_H +#define __LINUX_GENERIC_NETLINK_H + +#include <linux/netlink.h> + +#define GENL_NAMSIZ 16 /* length of family name */ + +#define GENL_MIN_ID NLMSG_MIN_TYPE +#define GENL_MAX_ID 1023 + +struct genlmsghdr { + __u8 cmd; + __u8 version; + __u16 reserved; +}; + +#define GENL_HDRLEN NLMSG_ALIGN(sizeof(struct genlmsghdr)) + +/* + * List of reserved static generic netlink identifiers: + */ +#define GENL_ID_GENERATE 0 +#define GENL_ID_CTRL NLMSG_MIN_TYPE + +/************************************************************************** + * Controller + **************************************************************************/ + +enum { + CTRL_CMD_UNSPEC, + CTRL_CMD_NEWFAMILY, + CTRL_CMD_DELFAMILY, + CTRL_CMD_GETFAMILY, + CTRL_CMD_NEWOPS, + CTRL_CMD_DELOPS, + CTRL_CMD_GETOPS, + __CTRL_CMD_MAX, +}; + +#define CTRL_CMD_MAX (__CTRL_CMD_MAX - 1) + +enum { + CTRL_ATTR_UNSPEC, + CTRL_ATTR_FAMILY_ID, + CTRL_ATTR_FAMILY_NAME, + __CTRL_ATTR_MAX, +}; + +#define CTRL_ATTR_MAX (__CTRL_ATTR_MAX - 1) + +#endif /* __LINUX_GENERIC_NETLINK_H */ diff --git a/include/linux/ide.h b/include/linux/ide.h index 3461abc1e85..ac8b25fa650 100644 --- a/include/linux/ide.h +++ b/include/linux/ide.h @@ -230,6 +230,7 @@ typedef struct hw_regs_s { int dma; /* our dma entry */ ide_ack_intr_t *ack_intr; /* acknowledge interrupt */ hwif_chipset_t chipset; + struct device *dev; } hw_regs_t; /* @@ -266,6 +267,10 @@ static inline void ide_std_init_ports(hw_regs_t *hw, #include <asm/ide.h> +#ifndef MAX_HWIFS +#define MAX_HWIFS CONFIG_IDE_MAX_HWIFS +#endif + /* needed on alpha, x86/x86_64, ia64, mips, ppc32 and sh */ #ifndef IDE_ARCH_OBSOLETE_DEFAULTS # define ide_default_io_base(index) (0) @@ -1324,7 +1329,8 @@ void ide_init_disk(struct gendisk *, ide_drive_t *); extern int ideprobe_init(void); extern void ide_scan_pcibus(int scan_direction) __init; -extern int ide_pci_register_driver(struct pci_driver *driver); +extern int __ide_pci_register_driver(struct pci_driver *driver, struct module *owner); +#define ide_pci_register_driver(d) __ide_pci_register_driver(d, THIS_MODULE) extern void ide_pci_unregister_driver(struct pci_driver *driver); void ide_pci_setup_ports(struct pci_dev *, struct ide_pci_device_s *, int, ata_index_t *); extern void ide_setup_pci_noise (struct pci_dev *dev, struct ide_pci_device_s *d); diff --git a/include/linux/if_ether.h b/include/linux/if_ether.h index d21c305c6c6..fe26d431de8 100644 --- a/include/linux/if_ether.h +++ b/include/linux/if_ether.h @@ -21,6 +21,8 @@ #ifndef _LINUX_IF_ETHER_H #define _LINUX_IF_ETHER_H +#include <linux/types.h> + /* * IEEE 802.3 Ethernet magic constants. The frame sizes omit the preamble * and FCS/CRC (frame check sequence). @@ -100,7 +102,7 @@ struct ethhdr { unsigned char h_dest[ETH_ALEN]; /* destination eth addr */ unsigned char h_source[ETH_ALEN]; /* source ether addr */ - unsigned short h_proto; /* packet type ID field */ + __be16 h_proto; /* packet type ID field */ } __attribute__((packed)); #ifdef __KERNEL__ diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index c6efce4a04a..936f8b76114 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -927,6 +927,13 @@ extern int netdev_max_backlog; extern int weight_p; extern int netdev_set_master(struct net_device *dev, struct net_device *master); extern int skb_checksum_help(struct sk_buff *skb, int inward); +#ifdef CONFIG_BUG +extern void netdev_rx_csum_fault(struct net_device *dev); +#else +static inline void netdev_rx_csum_fault(struct net_device *dev) +{ +} +#endif /* rx skb timestamps */ extern void net_enable_timestamp(void); extern void net_disable_timestamp(void); diff --git a/include/linux/netfilter/nf_conntrack_common.h b/include/linux/netfilter/nf_conntrack_common.h new file mode 100644 index 00000000000..6d39b518486 --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_common.h @@ -0,0 +1,159 @@ +#ifndef _NF_CONNTRACK_COMMON_H +#define _NF_CONNTRACK_COMMON_H +/* Connection state tracking for netfilter. This is separated from, + but required by, the NAT layer; it can also be used by an iptables + extension. */ +enum ip_conntrack_info +{ + /* Part of an established connection (either direction). */ + IP_CT_ESTABLISHED, + + /* Like NEW, but related to an existing connection, or ICMP error + (in either direction). */ + IP_CT_RELATED, + + /* Started a new connection to track (only + IP_CT_DIR_ORIGINAL); may be a retransmission. */ + IP_CT_NEW, + + /* >= this indicates reply direction */ + IP_CT_IS_REPLY, + + /* Number of distinct IP_CT types (no NEW in reply dirn). */ + IP_CT_NUMBER = IP_CT_IS_REPLY * 2 - 1 +}; + +/* Bitset representing status of connection. */ +enum ip_conntrack_status { + /* It's an expected connection: bit 0 set. This bit never changed */ + IPS_EXPECTED_BIT = 0, + IPS_EXPECTED = (1 << IPS_EXPECTED_BIT), + + /* We've seen packets both ways: bit 1 set. Can be set, not unset. */ + IPS_SEEN_REPLY_BIT = 1, + IPS_SEEN_REPLY = (1 << IPS_SEEN_REPLY_BIT), + + /* Conntrack should never be early-expired. */ + IPS_ASSURED_BIT = 2, + IPS_ASSURED = (1 << IPS_ASSURED_BIT), + + /* Connection is confirmed: originating packet has left box */ + IPS_CONFIRMED_BIT = 3, + IPS_CONFIRMED = (1 << IPS_CONFIRMED_BIT), + + /* Connection needs src nat in orig dir. This bit never changed. */ + IPS_SRC_NAT_BIT = 4, + IPS_SRC_NAT = (1 << IPS_SRC_NAT_BIT), + + /* Connection needs dst nat in orig dir. This bit never changed. */ + IPS_DST_NAT_BIT = 5, + IPS_DST_NAT = (1 << IPS_DST_NAT_BIT), + + /* Both together. */ + IPS_NAT_MASK = (IPS_DST_NAT | IPS_SRC_NAT), + + /* Connection needs TCP sequence adjusted. */ + IPS_SEQ_ADJUST_BIT = 6, + IPS_SEQ_ADJUST = (1 << IPS_SEQ_ADJUST_BIT), + + /* NAT initialization bits. */ + IPS_SRC_NAT_DONE_BIT = 7, + IPS_SRC_NAT_DONE = (1 << IPS_SRC_NAT_DONE_BIT), + + IPS_DST_NAT_DONE_BIT = 8, + IPS_DST_NAT_DONE = (1 << IPS_DST_NAT_DONE_BIT), + + /* Both together */ + IPS_NAT_DONE_MASK = (IPS_DST_NAT_DONE | IPS_SRC_NAT_DONE), + + /* Connection is dying (removed from lists), can not be unset. */ + IPS_DYING_BIT = 9, + IPS_DYING = (1 << IPS_DYING_BIT), +}; + +/* Connection tracking event bits */ +enum ip_conntrack_events +{ + /* New conntrack */ + IPCT_NEW_BIT = 0, + IPCT_NEW = (1 << IPCT_NEW_BIT), + + /* Expected connection */ + IPCT_RELATED_BIT = 1, + IPCT_RELATED = (1 << IPCT_RELATED_BIT), + + /* Destroyed conntrack */ + IPCT_DESTROY_BIT = 2, + IPCT_DESTROY = (1 << IPCT_DESTROY_BIT), + + /* Timer has been refreshed */ + IPCT_REFRESH_BIT = 3, + IPCT_REFRESH = (1 << IPCT_REFRESH_BIT), + + /* Status has changed */ + IPCT_STATUS_BIT = 4, + IPCT_STATUS = (1 << IPCT_STATUS_BIT), + + /* Update of protocol info */ + IPCT_PROTOINFO_BIT = 5, + IPCT_PROTOINFO = (1 << IPCT_PROTOINFO_BIT), + + /* Volatile protocol info */ + IPCT_PROTOINFO_VOLATILE_BIT = 6, + IPCT_PROTOINFO_VOLATILE = (1 << IPCT_PROTOINFO_VOLATILE_BIT), + + /* New helper for conntrack */ + IPCT_HELPER_BIT = 7, + IPCT_HELPER = (1 << IPCT_HELPER_BIT), + + /* Update of helper info */ + IPCT_HELPINFO_BIT = 8, + IPCT_HELPINFO = (1 << IPCT_HELPINFO_BIT), + + /* Volatile helper info */ + IPCT_HELPINFO_VOLATILE_BIT = 9, + IPCT_HELPINFO_VOLATILE = (1 << IPCT_HELPINFO_VOLATILE_BIT), + + /* NAT info */ + IPCT_NATINFO_BIT = 10, + IPCT_NATINFO = (1 << IPCT_NATINFO_BIT), + + /* Counter highest bit has been set */ + IPCT_COUNTER_FILLING_BIT = 11, + IPCT_COUNTER_FILLING = (1 << IPCT_COUNTER_FILLING_BIT), +}; + +enum ip_conntrack_expect_events { + IPEXP_NEW_BIT = 0, + IPEXP_NEW = (1 << IPEXP_NEW_BIT), +}; + +#ifdef __KERNEL__ +struct ip_conntrack_counter +{ + u_int32_t packets; + u_int32_t bytes; +}; + +struct ip_conntrack_stat +{ + unsigned int searched; + unsigned int found; + unsigned int new; + unsigned int invalid; + unsigned int ignore; + unsigned int delete; + unsigned int delete_list; + unsigned int insert; + unsigned int insert_failed; + unsigned int drop; + unsigned int early_drop; + unsigned int error; + unsigned int expect_new; + unsigned int expect_create; + unsigned int expect_delete; +}; + +#endif /* __KERNEL__ */ + +#endif /* _NF_CONNTRACK_COMMON_H */ diff --git a/include/linux/netfilter/nf_conntrack_ftp.h b/include/linux/netfilter/nf_conntrack_ftp.h new file mode 100644 index 00000000000..ad4a41c9ce9 --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_ftp.h @@ -0,0 +1,44 @@ +#ifndef _NF_CONNTRACK_FTP_H +#define _NF_CONNTRACK_FTP_H +/* FTP tracking. */ + +/* This enum is exposed to userspace */ +enum ip_ct_ftp_type +{ + /* PORT command from client */ + IP_CT_FTP_PORT, + /* PASV response from server */ + IP_CT_FTP_PASV, + /* EPRT command from client */ + IP_CT_FTP_EPRT, + /* EPSV response from server */ + IP_CT_FTP_EPSV, +}; + +#ifdef __KERNEL__ + +#define FTP_PORT 21 + +#define NUM_SEQ_TO_REMEMBER 2 +/* This structure exists only once per master */ +struct ip_ct_ftp_master { + /* Valid seq positions for cmd matching after newline */ + u_int32_t seq_aft_nl[IP_CT_DIR_MAX][NUM_SEQ_TO_REMEMBER]; + /* 0 means seq_match_aft_nl not set */ + int seq_aft_nl_num[IP_CT_DIR_MAX]; +}; + +struct ip_conntrack_expect; + +/* For NAT to hook in when we find a packet which describes what other + * connection we should expect. */ +extern unsigned int (*ip_nat_ftp_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + enum ip_ct_ftp_type type, + unsigned int matchoff, + unsigned int matchlen, + struct ip_conntrack_expect *exp, + u32 *seq); +#endif /* __KERNEL__ */ + +#endif /* _NF_CONNTRACK_FTP_H */ diff --git a/include/linux/netfilter/nf_conntrack_sctp.h b/include/linux/netfilter/nf_conntrack_sctp.h new file mode 100644 index 00000000000..b8994d9fd1a --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_sctp.h @@ -0,0 +1,27 @@ +#ifndef _NF_CONNTRACK_SCTP_H +#define _NF_CONNTRACK_SCTP_H +/* SCTP tracking. */ + +#include <linux/netfilter/nf_conntrack_tuple_common.h> + +enum sctp_conntrack { + SCTP_CONNTRACK_NONE, + SCTP_CONNTRACK_CLOSED, + SCTP_CONNTRACK_COOKIE_WAIT, + SCTP_CONNTRACK_COOKIE_ECHOED, + SCTP_CONNTRACK_ESTABLISHED, + SCTP_CONNTRACK_SHUTDOWN_SENT, + SCTP_CONNTRACK_SHUTDOWN_RECD, + SCTP_CONNTRACK_SHUTDOWN_ACK_SENT, + SCTP_CONNTRACK_MAX +}; + +struct ip_ct_sctp +{ + enum sctp_conntrack state; + + u_int32_t vtag[IP_CT_DIR_MAX]; + u_int32_t ttag[IP_CT_DIR_MAX]; +}; + +#endif /* _NF_CONNTRACK_SCTP_H */ diff --git a/include/linux/netfilter/nf_conntrack_tcp.h b/include/linux/netfilter/nf_conntrack_tcp.h new file mode 100644 index 00000000000..b2feeffde38 --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_tcp.h @@ -0,0 +1,56 @@ +#ifndef _NF_CONNTRACK_TCP_H +#define _NF_CONNTRACK_TCP_H +/* TCP tracking. */ + +/* This is exposed to userspace (ctnetlink) */ +enum tcp_conntrack { + TCP_CONNTRACK_NONE, + TCP_CONNTRACK_SYN_SENT, + TCP_CONNTRACK_SYN_RECV, + TCP_CONNTRACK_ESTABLISHED, + TCP_CONNTRACK_FIN_WAIT, + TCP_CONNTRACK_CLOSE_WAIT, + TCP_CONNTRACK_LAST_ACK, + TCP_CONNTRACK_TIME_WAIT, + TCP_CONNTRACK_CLOSE, + TCP_CONNTRACK_LISTEN, + TCP_CONNTRACK_MAX, + TCP_CONNTRACK_IGNORE +}; + +/* Window scaling is advertised by the sender */ +#define IP_CT_TCP_FLAG_WINDOW_SCALE 0x01 + +/* SACK is permitted by the sender */ +#define IP_CT_TCP_FLAG_SACK_PERM 0x02 + +/* This sender sent FIN first */ +#define IP_CT_TCP_FLAG_CLOSE_INIT 0x03 + +#ifdef __KERNEL__ + +struct ip_ct_tcp_state { + u_int32_t td_end; /* max of seq + len */ + u_int32_t td_maxend; /* max of ack + max(win, 1) */ + u_int32_t td_maxwin; /* max(win) */ + u_int8_t td_scale; /* window scale factor */ + u_int8_t loose; /* used when connection picked up from the middle */ + u_int8_t flags; /* per direction options */ +}; + +struct ip_ct_tcp +{ + struct ip_ct_tcp_state seen[2]; /* connection parameters per direction */ + u_int8_t state; /* state of the connection (enum tcp_conntrack) */ + /* For detecting stale connections */ + u_int8_t last_dir; /* Direction of the last packet (enum ip_conntrack_dir) */ + u_int8_t retrans; /* Number of retransmitted packets */ + u_int8_t last_index; /* Index of the last packet */ + u_int32_t last_seq; /* Last sequence number seen in dir */ + u_int32_t last_ack; /* Last sequence number seen in opposite dir */ + u_int32_t last_end; /* Last seq + len */ +}; + +#endif /* __KERNEL__ */ + +#endif /* _NF_CONNTRACK_TCP_H */ diff --git a/include/linux/netfilter/nf_conntrack_tuple_common.h b/include/linux/netfilter/nf_conntrack_tuple_common.h new file mode 100644 index 00000000000..8e145f0d61c --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_tuple_common.h @@ -0,0 +1,13 @@ +#ifndef _NF_CONNTRACK_TUPLE_COMMON_H +#define _NF_CONNTRACK_TUPLE_COMMON_H + +enum ip_conntrack_dir +{ + IP_CT_DIR_ORIGINAL, + IP_CT_DIR_REPLY, + IP_CT_DIR_MAX +}; + +#define CTINFO2DIR(ctinfo) ((ctinfo) >= IP_CT_IS_REPLY ? IP_CT_DIR_REPLY : IP_CT_DIR_ORIGINAL) + +#endif /* _NF_CONNTRACK_TUPLE_COMMON_H */ diff --git a/include/linux/netfilter/nfnetlink.h b/include/linux/netfilter/nfnetlink.h index f08e870100f..72975fa8795 100644 --- a/include/linux/netfilter/nfnetlink.h +++ b/include/linux/netfilter/nfnetlink.h @@ -146,7 +146,7 @@ extern void nfnl_unlock(void); extern int nfnetlink_subsys_register(struct nfnetlink_subsystem *n); extern int nfnetlink_subsys_unregister(struct nfnetlink_subsystem *n); -extern int nfattr_parse(struct nfattr *tb[], int maxattr, +extern void nfattr_parse(struct nfattr *tb[], int maxattr, struct nfattr *nfa, int len); #define nfattr_parse_nested(tb, max, nfa) \ diff --git a/include/linux/netfilter_ipv4/ip_conntrack.h b/include/linux/netfilter_ipv4/ip_conntrack.h index d078bb91d9e..b3432ab59a1 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack.h +++ b/include/linux/netfilter_ipv4/ip_conntrack.h @@ -1,132 +1,7 @@ #ifndef _IP_CONNTRACK_H #define _IP_CONNTRACK_H -/* Connection state tracking for netfilter. This is separated from, - but required by, the NAT layer; it can also be used by an iptables - extension. */ -enum ip_conntrack_info -{ - /* Part of an established connection (either direction). */ - IP_CT_ESTABLISHED, - - /* Like NEW, but related to an existing connection, or ICMP error - (in either direction). */ - IP_CT_RELATED, - - /* Started a new connection to track (only - IP_CT_DIR_ORIGINAL); may be a retransmission. */ - IP_CT_NEW, - - /* >= this indicates reply direction */ - IP_CT_IS_REPLY, - - /* Number of distinct IP_CT types (no NEW in reply dirn). */ - IP_CT_NUMBER = IP_CT_IS_REPLY * 2 - 1 -}; - -/* Bitset representing status of connection. */ -enum ip_conntrack_status { - /* It's an expected connection: bit 0 set. This bit never changed */ - IPS_EXPECTED_BIT = 0, - IPS_EXPECTED = (1 << IPS_EXPECTED_BIT), - - /* We've seen packets both ways: bit 1 set. Can be set, not unset. */ - IPS_SEEN_REPLY_BIT = 1, - IPS_SEEN_REPLY = (1 << IPS_SEEN_REPLY_BIT), - - /* Conntrack should never be early-expired. */ - IPS_ASSURED_BIT = 2, - IPS_ASSURED = (1 << IPS_ASSURED_BIT), - - /* Connection is confirmed: originating packet has left box */ - IPS_CONFIRMED_BIT = 3, - IPS_CONFIRMED = (1 << IPS_CONFIRMED_BIT), - - /* Connection needs src nat in orig dir. This bit never changed. */ - IPS_SRC_NAT_BIT = 4, - IPS_SRC_NAT = (1 << IPS_SRC_NAT_BIT), - - /* Connection needs dst nat in orig dir. This bit never changed. */ - IPS_DST_NAT_BIT = 5, - IPS_DST_NAT = (1 << IPS_DST_NAT_BIT), - - /* Both together. */ - IPS_NAT_MASK = (IPS_DST_NAT | IPS_SRC_NAT), - - /* Connection needs TCP sequence adjusted. */ - IPS_SEQ_ADJUST_BIT = 6, - IPS_SEQ_ADJUST = (1 << IPS_SEQ_ADJUST_BIT), - - /* NAT initialization bits. */ - IPS_SRC_NAT_DONE_BIT = 7, - IPS_SRC_NAT_DONE = (1 << IPS_SRC_NAT_DONE_BIT), - - IPS_DST_NAT_DONE_BIT = 8, - IPS_DST_NAT_DONE = (1 << IPS_DST_NAT_DONE_BIT), - - /* Both together */ - IPS_NAT_DONE_MASK = (IPS_DST_NAT_DONE | IPS_SRC_NAT_DONE), - - /* Connection is dying (removed from lists), can not be unset. */ - IPS_DYING_BIT = 9, - IPS_DYING = (1 << IPS_DYING_BIT), -}; - -/* Connection tracking event bits */ -enum ip_conntrack_events -{ - /* New conntrack */ - IPCT_NEW_BIT = 0, - IPCT_NEW = (1 << IPCT_NEW_BIT), - - /* Expected connection */ - IPCT_RELATED_BIT = 1, - IPCT_RELATED = (1 << IPCT_RELATED_BIT), - - /* Destroyed conntrack */ - IPCT_DESTROY_BIT = 2, - IPCT_DESTROY = (1 << IPCT_DESTROY_BIT), - - /* Timer has been refreshed */ - IPCT_REFRESH_BIT = 3, - IPCT_REFRESH = (1 << IPCT_REFRESH_BIT), - - /* Status has changed */ - IPCT_STATUS_BIT = 4, - IPCT_STATUS = (1 << IPCT_STATUS_BIT), - - /* Update of protocol info */ - IPCT_PROTOINFO_BIT = 5, - IPCT_PROTOINFO = (1 << IPCT_PROTOINFO_BIT), - - /* Volatile protocol info */ - IPCT_PROTOINFO_VOLATILE_BIT = 6, - IPCT_PROTOINFO_VOLATILE = (1 << IPCT_PROTOINFO_VOLATILE_BIT), - - /* New helper for conntrack */ - IPCT_HELPER_BIT = 7, - IPCT_HELPER = (1 << IPCT_HELPER_BIT), - - /* Update of helper info */ - IPCT_HELPINFO_BIT = 8, - IPCT_HELPINFO = (1 << IPCT_HELPINFO_BIT), - - /* Volatile helper info */ - IPCT_HELPINFO_VOLATILE_BIT = 9, - IPCT_HELPINFO_VOLATILE = (1 << IPCT_HELPINFO_VOLATILE_BIT), - /* NAT info */ - IPCT_NATINFO_BIT = 10, - IPCT_NATINFO = (1 << IPCT_NATINFO_BIT), - - /* Counter highest bit has been set */ - IPCT_COUNTER_FILLING_BIT = 11, - IPCT_COUNTER_FILLING = (1 << IPCT_COUNTER_FILLING_BIT), -}; - -enum ip_conntrack_expect_events { - IPEXP_NEW_BIT = 0, - IPEXP_NEW = (1 << IPEXP_NEW_BIT), -}; +#include <linux/netfilter/nf_conntrack_common.h> #ifdef __KERNEL__ #include <linux/config.h> @@ -194,12 +69,6 @@ do { \ #define IP_NF_ASSERT(x) #endif -struct ip_conntrack_counter -{ - u_int32_t packets; - u_int32_t bytes; -}; - struct ip_conntrack_helper; struct ip_conntrack @@ -426,25 +295,6 @@ static inline int is_dying(struct ip_conntrack *ct) extern unsigned int ip_conntrack_htable_size; -struct ip_conntrack_stat -{ - unsigned int searched; - unsigned int found; - unsigned int new; - unsigned int invalid; - unsigned int ignore; - unsigned int delete; - unsigned int delete_list; - unsigned int insert; - unsigned int insert_failed; - unsigned int drop; - unsigned int early_drop; - unsigned int error; - unsigned int expect_new; - unsigned int expect_create; - unsigned int expect_delete; -}; - #define CONNTRACK_STAT_INC(count) (__get_cpu_var(ip_conntrack_stat).count++) #ifdef CONFIG_IP_NF_CONNTRACK_EVENTS diff --git a/include/linux/netfilter_ipv4/ip_conntrack_ftp.h b/include/linux/netfilter_ipv4/ip_conntrack_ftp.h index 5f06429b904..63811934de4 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_ftp.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_ftp.h @@ -1,43 +1,6 @@ #ifndef _IP_CONNTRACK_FTP_H #define _IP_CONNTRACK_FTP_H -/* FTP tracking. */ -#ifdef __KERNEL__ +#include <linux/netfilter/nf_conntrack_ftp.h> -#define FTP_PORT 21 - -#endif /* __KERNEL__ */ - -enum ip_ct_ftp_type -{ - /* PORT command from client */ - IP_CT_FTP_PORT, - /* PASV response from server */ - IP_CT_FTP_PASV, - /* EPRT command from client */ - IP_CT_FTP_EPRT, - /* EPSV response from server */ - IP_CT_FTP_EPSV, -}; - -#define NUM_SEQ_TO_REMEMBER 2 -/* This structure exists only once per master */ -struct ip_ct_ftp_master { - /* Valid seq positions for cmd matching after newline */ - u_int32_t seq_aft_nl[IP_CT_DIR_MAX][NUM_SEQ_TO_REMEMBER]; - /* 0 means seq_match_aft_nl not set */ - int seq_aft_nl_num[IP_CT_DIR_MAX]; -}; - -struct ip_conntrack_expect; - -/* For NAT to hook in when we find a packet which describes what other - * connection we should expect. */ -extern unsigned int (*ip_nat_ftp_hook)(struct sk_buff **pskb, - enum ip_conntrack_info ctinfo, - enum ip_ct_ftp_type type, - unsigned int matchoff, - unsigned int matchlen, - struct ip_conntrack_expect *exp, - u32 *seq); #endif /* _IP_CONNTRACK_FTP_H */ diff --git a/include/linux/netfilter_ipv4/ip_conntrack_icmp.h b/include/linux/netfilter_ipv4/ip_conntrack_icmp.h index f1664abbe39..eed5ee3e474 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_icmp.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_icmp.h @@ -1,11 +1,6 @@ #ifndef _IP_CONNTRACK_ICMP_H #define _IP_CONNTRACK_ICMP_H -/* ICMP tracking. */ -#include <asm/atomic.h> -struct ip_ct_icmp -{ - /* Optimization: when number in == number out, forget immediately. */ - atomic_t count; -}; +#include <net/netfilter/ipv4/nf_conntrack_icmp.h> + #endif /* _IP_CONNTRACK_ICMP_H */ diff --git a/include/linux/netfilter_ipv4/ip_conntrack_sctp.h b/include/linux/netfilter_ipv4/ip_conntrack_sctp.h index 7a8d869321f..4099a041a32 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_sctp.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_sctp.h @@ -1,25 +1,6 @@ #ifndef _IP_CONNTRACK_SCTP_H #define _IP_CONNTRACK_SCTP_H -/* SCTP tracking. */ -enum sctp_conntrack { - SCTP_CONNTRACK_NONE, - SCTP_CONNTRACK_CLOSED, - SCTP_CONNTRACK_COOKIE_WAIT, - SCTP_CONNTRACK_COOKIE_ECHOED, - SCTP_CONNTRACK_ESTABLISHED, - SCTP_CONNTRACK_SHUTDOWN_SENT, - SCTP_CONNTRACK_SHUTDOWN_RECD, - SCTP_CONNTRACK_SHUTDOWN_ACK_SENT, - SCTP_CONNTRACK_MAX -}; - -struct ip_ct_sctp -{ - enum sctp_conntrack state; - - u_int32_t vtag[IP_CT_DIR_MAX]; - u_int32_t ttag[IP_CT_DIR_MAX]; -}; +#include <linux/netfilter/nf_conntrack_sctp.h> #endif /* _IP_CONNTRACK_SCTP_H */ diff --git a/include/linux/netfilter_ipv4/ip_conntrack_tcp.h b/include/linux/netfilter_ipv4/ip_conntrack_tcp.h index 16da044d97a..876b8fb17e6 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_tcp.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_tcp.h @@ -1,51 +1,6 @@ #ifndef _IP_CONNTRACK_TCP_H #define _IP_CONNTRACK_TCP_H -/* TCP tracking. */ -enum tcp_conntrack { - TCP_CONNTRACK_NONE, - TCP_CONNTRACK_SYN_SENT, - TCP_CONNTRACK_SYN_RECV, - TCP_CONNTRACK_ESTABLISHED, - TCP_CONNTRACK_FIN_WAIT, - TCP_CONNTRACK_CLOSE_WAIT, - TCP_CONNTRACK_LAST_ACK, - TCP_CONNTRACK_TIME_WAIT, - TCP_CONNTRACK_CLOSE, - TCP_CONNTRACK_LISTEN, - TCP_CONNTRACK_MAX, - TCP_CONNTRACK_IGNORE -}; - -/* Window scaling is advertised by the sender */ -#define IP_CT_TCP_FLAG_WINDOW_SCALE 0x01 - -/* SACK is permitted by the sender */ -#define IP_CT_TCP_FLAG_SACK_PERM 0x02 - -/* This sender sent FIN first */ -#define IP_CT_TCP_FLAG_CLOSE_INIT 0x03 - -struct ip_ct_tcp_state { - u_int32_t td_end; /* max of seq + len */ - u_int32_t td_maxend; /* max of ack + max(win, 1) */ - u_int32_t td_maxwin; /* max(win) */ - u_int8_t td_scale; /* window scale factor */ - u_int8_t loose; /* used when connection picked up from the middle */ - u_int8_t flags; /* per direction options */ -}; - -struct ip_ct_tcp -{ - struct ip_ct_tcp_state seen[2]; /* connection parameters per direction */ - u_int8_t state; /* state of the connection (enum tcp_conntrack) */ - /* For detecting stale connections */ - u_int8_t last_dir; /* Direction of the last packet (enum ip_conntrack_dir) */ - u_int8_t retrans; /* Number of retransmitted packets */ - u_int8_t last_index; /* Index of the last packet */ - u_int32_t last_seq; /* Last sequence number seen in dir */ - u_int32_t last_ack; /* Last sequence number seen in opposite dir */ - u_int32_t last_end; /* Last seq + len */ -}; +#include <linux/netfilter/nf_conntrack_tcp.h> #endif /* _IP_CONNTRACK_TCP_H */ diff --git a/include/linux/netfilter_ipv4/ip_conntrack_tuple.h b/include/linux/netfilter_ipv4/ip_conntrack_tuple.h index 3232db11a4e..2fdabdb4c0e 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_tuple.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_tuple.h @@ -2,6 +2,7 @@ #define _IP_CONNTRACK_TUPLE_H #include <linux/types.h> +#include <linux/netfilter/nf_conntrack_tuple_common.h> /* A `tuple' is a structure containing the information to uniquely identify a connection. ie. if two packets have the same tuple, they @@ -88,13 +89,6 @@ struct ip_conntrack_tuple (tuple)->dst.u.all = 0; \ } while (0) -enum ip_conntrack_dir -{ - IP_CT_DIR_ORIGINAL, - IP_CT_DIR_REPLY, - IP_CT_DIR_MAX -}; - #ifdef __KERNEL__ #define DUMP_TUPLE(tp) \ @@ -103,8 +97,6 @@ DEBUGP("tuple %p: %u %u.%u.%u.%u:%hu -> %u.%u.%u.%u:%hu\n", \ NIPQUAD((tp)->src.ip), ntohs((tp)->src.u.all), \ NIPQUAD((tp)->dst.ip), ntohs((tp)->dst.u.all)) -#define CTINFO2DIR(ctinfo) ((ctinfo) >= IP_CT_IS_REPLY ? IP_CT_DIR_REPLY : IP_CT_DIR_ORIGINAL) - /* If we're the first tuple, it's the original dir. */ #define DIRECTION(h) ((enum ip_conntrack_dir)(h)->tuple.dst.dir) diff --git a/include/linux/netfilter_ipv6.h b/include/linux/netfilter_ipv6.h index edcc2c6eb5c..53b2983f627 100644 --- a/include/linux/netfilter_ipv6.h +++ b/include/linux/netfilter_ipv6.h @@ -59,6 +59,7 @@ enum nf_ip6_hook_priorities { NF_IP6_PRI_FIRST = INT_MIN, + NF_IP6_PRI_CONNTRACK_DEFRAG = -400, NF_IP6_PRI_SELINUX_FIRST = -225, NF_IP6_PRI_CONNTRACK = -200, NF_IP6_PRI_BRIDGE_SABOTAGE_FORWARD = -175, diff --git a/include/linux/netlink.h b/include/linux/netlink.h index ba25ca874c2..6a2ccf78a35 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -71,7 +71,8 @@ struct nlmsghdr #define NLMSG_ALIGNTO 4 #define NLMSG_ALIGN(len) ( ((len)+NLMSG_ALIGNTO-1) & ~(NLMSG_ALIGNTO-1) ) -#define NLMSG_LENGTH(len) ((len)+NLMSG_ALIGN(sizeof(struct nlmsghdr))) +#define NLMSG_HDRLEN ((int) NLMSG_ALIGN(sizeof(struct nlmsghdr))) +#define NLMSG_LENGTH(len) ((len)+NLMSG_ALIGN(NLMSG_HDRLEN)) #define NLMSG_SPACE(len) NLMSG_ALIGN(NLMSG_LENGTH(len)) #define NLMSG_DATA(nlh) ((void*)(((char*)nlh) + NLMSG_LENGTH(0))) #define NLMSG_NEXT(nlh,len) ((len) -= NLMSG_ALIGN((nlh)->nlmsg_len), \ @@ -86,6 +87,8 @@ struct nlmsghdr #define NLMSG_DONE 0x3 /* End of a dump */ #define NLMSG_OVERRUN 0x4 /* Data lost */ +#define NLMSG_MIN_TYPE 0x10 /* < 0x10: reserved control messages */ + struct nlmsgerr { int error; @@ -108,6 +111,25 @@ enum { NETLINK_CONNECTED, }; +/* + * <------- NLA_HDRLEN ------> <-- NLA_ALIGN(payload)--> + * +---------------------+- - -+- - - - - - - - - -+- - -+ + * | Header | Pad | Payload | Pad | + * | (struct nlattr) | ing | | ing | + * +---------------------+- - -+- - - - - - - - - -+- - -+ + * <-------------- nlattr->nla_len --------------> + */ + +struct nlattr +{ + __u16 nla_len; + __u16 nla_type; +}; + +#define NLA_ALIGNTO 4 +#define NLA_ALIGN(len) (((len) + NLA_ALIGNTO - 1) & ~(NLA_ALIGNTO - 1)) +#define NLA_HDRLEN ((int) NLA_ALIGN(sizeof(struct nlattr))) + #ifdef __KERNEL__ #include <linux/capability.h> diff --git a/include/linux/pci-acpi.h b/include/linux/pci-acpi.h index 857126a36ec..4877e35ae20 100644 --- a/include/linux/pci-acpi.h +++ b/include/linux/pci-acpi.h @@ -47,14 +47,15 @@ OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL) #ifdef CONFIG_ACPI -extern acpi_status pci_osc_control_set(u32 flags); +extern acpi_status pci_osc_control_set(acpi_handle handle, u32 flags); extern acpi_status pci_osc_support_set(u32 flags); #else #if !defined(acpi_status) typedef u32 acpi_status; #define AE_ERROR (acpi_status) (0x0001) #endif -static inline acpi_status pci_osc_control_set(u32 flags) {return AE_ERROR;} +static inline acpi_status pci_osc_control_set(acpi_handle handle, u32 flags) +{return AE_ERROR;} static inline acpi_status pci_osc_support_set(u32 flags) {return AE_ERROR;} #endif diff --git a/include/linux/pci.h b/include/linux/pci.h index 3596ac94ecf..de690ca73d5 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -236,7 +236,6 @@ struct module; struct pci_driver { struct list_head node; char *name; - struct module *owner; const struct pci_device_id *id_table; /* must be non-NULL for probe to be called */ int (*probe) (struct pci_dev *dev, const struct pci_device_id *id); /* New device inserted */ void (*remove) (struct pci_dev *dev); /* Device removed (NULL if not a hot-plug capable driver) */ @@ -338,6 +337,7 @@ struct pci_dev *pci_find_device (unsigned int vendor, unsigned int device, const struct pci_dev *pci_find_device_reverse (unsigned int vendor, unsigned int device, const struct pci_dev *from); struct pci_dev *pci_find_slot (unsigned int bus, unsigned int devfn); int pci_find_capability (struct pci_dev *dev, int cap); +int pci_find_next_capability (struct pci_dev *dev, u8 pos, int cap); int pci_find_ext_capability (struct pci_dev *dev, int cap); struct pci_bus * pci_find_next_bus(const struct pci_bus *from); @@ -432,8 +432,13 @@ int pci_bus_alloc_resource(struct pci_bus *bus, struct resource *res, void *alignf_data); void pci_enable_bridges(struct pci_bus *bus); -/* New-style probing supporting hot-pluggable devices */ -int pci_register_driver(struct pci_driver *); +/* Proper probing supporting hot-pluggable devices */ +int __pci_register_driver(struct pci_driver *, struct module *); +static inline int pci_register_driver(struct pci_driver *driver) +{ + return __pci_register_driver(driver, THIS_MODULE); +} + void pci_unregister_driver(struct pci_driver *); void pci_remove_behind_bridge(struct pci_dev *); struct pci_driver *pci_dev_driver(const struct pci_dev *); @@ -547,9 +552,11 @@ static inline int pci_enable_device(struct pci_dev *dev) { return -EIO; } static inline void pci_disable_device(struct pci_dev *dev) { } static inline int pci_set_dma_mask(struct pci_dev *dev, u64 mask) { return -EIO; } static inline int pci_assign_resource(struct pci_dev *dev, int i) { return -EBUSY;} +static inline int __pci_register_driver(struct pci_driver *drv, struct module *owner) { return 0;} static inline int pci_register_driver(struct pci_driver *drv) { return 0;} static inline void pci_unregister_driver(struct pci_driver *drv) { } static inline int pci_find_capability (struct pci_dev *dev, int cap) {return 0; } +static inline int pci_find_next_capability (struct pci_dev *dev, u8 post, int cap) { return 0; } static inline int pci_find_ext_capability (struct pci_dev *dev, int cap) {return 0; } static inline const struct pci_device_id *pci_match_device(const struct pci_device_id *ids, const struct pci_dev *dev) { return NULL; } diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 9a96f058839..4e06eb0f445 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -387,6 +387,7 @@ #define PCI_DEVICE_ID_NS_SC1100_SMI 0x0511 #define PCI_DEVICE_ID_NS_SC1100_XBUS 0x0515 #define PCI_DEVICE_ID_NS_87410 0xd001 +#define PCI_DEVICE_ID_NS_CS5535_IDE 0x002d #define PCI_VENDOR_ID_TSENG 0x100c #define PCI_DEVICE_ID_TSENG_W32P_2 0x3202 @@ -487,6 +488,8 @@ #define PCI_DEVICE_ID_AMD_8151_0 0x7454 #define PCI_DEVICE_ID_AMD_8131_APIC 0x7450 +#define PCI_DEVICE_ID_AMD_CS5536_IDE 0x209A + #define PCI_VENDOR_ID_TRIDENT 0x1023 #define PCI_DEVICE_ID_TRIDENT_4DWAVE_DX 0x2000 #define PCI_DEVICE_ID_TRIDENT_4DWAVE_NX 0x2001 diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index fdfb8fe8c38..0a8ea8b3581 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -274,6 +274,9 @@ struct sk_buff { #if defined(CONFIG_IP_VS) || defined(CONFIG_IP_VS_MODULE) __u8 ipvs_property:1; #endif +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) + struct sk_buff *nfct_reasm; +#endif #ifdef CONFIG_BRIDGE_NETFILTER struct nf_bridge_info *nf_bridge; #endif @@ -1233,8 +1236,7 @@ extern unsigned int datagram_poll(struct file *file, struct socket *sock, extern int skb_copy_datagram_iovec(const struct sk_buff *from, int offset, struct iovec *to, int size); -extern int skb_copy_and_csum_datagram_iovec(const - struct sk_buff *skb, +extern int skb_copy_and_csum_datagram_iovec(struct sk_buff *skb, int hlen, struct iovec *iov); extern void skb_free_datagram(struct sock *sk, struct sk_buff *skb); @@ -1302,6 +1304,30 @@ static inline void skb_set_timestamp(struct sk_buff *skb, const struct timeval * extern void __net_timestamp(struct sk_buff *skb); +extern unsigned int __skb_checksum_complete(struct sk_buff *skb); + +/** + * skb_checksum_complete - Calculate checksum of an entire packet + * @skb: packet to process + * + * This function calculates the checksum over the entire packet plus + * the value of skb->csum. The latter can be used to supply the + * checksum of a pseudo header as used by TCP/UDP. It returns the + * checksum. + * + * For protocols that contain complete checksums such as ICMP/TCP/UDP, + * this function can be used to verify that checksum on received + * packets. In that case the function should return zero if the + * checksum is correct. In particular, this function will return zero + * if skb->ip_summed is CHECKSUM_UNNECESSARY which indicates that the + * hardware has already verified the correctness of the checksum. + */ +static inline unsigned int skb_checksum_complete(struct sk_buff *skb) +{ + return skb->ip_summed != CHECKSUM_UNNECESSARY && + __skb_checksum_complete(skb); +} + #ifdef CONFIG_NETFILTER static inline void nf_conntrack_put(struct nf_conntrack *nfct) { @@ -1313,10 +1339,26 @@ static inline void nf_conntrack_get(struct nf_conntrack *nfct) if (nfct) atomic_inc(&nfct->use); } +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) +static inline void nf_conntrack_get_reasm(struct sk_buff *skb) +{ + if (skb) + atomic_inc(&skb->users); +} +static inline void nf_conntrack_put_reasm(struct sk_buff *skb) +{ + if (skb) + kfree_skb(skb); +} +#endif static inline void nf_reset(struct sk_buff *skb) { nf_conntrack_put(skb->nfct); skb->nfct = NULL; +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) + nf_conntrack_put_reasm(skb->nfct_reasm); + skb->nfct_reasm = NULL; +#endif } #ifdef CONFIG_BRIDGE_NETFILTER diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index fc131d6602b..ab2791b3189 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -205,6 +205,7 @@ enum NET_ECONET=16, NET_SCTP=17, NET_LLC=18, + NET_NETFILTER=19, }; /* /proc/sys/kernel/random */ @@ -270,6 +271,42 @@ enum NET_UNIX_MAX_DGRAM_QLEN=3, }; +/* /proc/sys/net/netfilter */ +enum +{ + NET_NF_CONNTRACK_MAX=1, + NET_NF_CONNTRACK_TCP_TIMEOUT_SYN_SENT=2, + NET_NF_CONNTRACK_TCP_TIMEOUT_SYN_RECV=3, + NET_NF_CONNTRACK_TCP_TIMEOUT_ESTABLISHED=4, + NET_NF_CONNTRACK_TCP_TIMEOUT_FIN_WAIT=5, + NET_NF_CONNTRACK_TCP_TIMEOUT_CLOSE_WAIT=6, + NET_NF_CONNTRACK_TCP_TIMEOUT_LAST_ACK=7, + NET_NF_CONNTRACK_TCP_TIMEOUT_TIME_WAIT=8, + NET_NF_CONNTRACK_TCP_TIMEOUT_CLOSE=9, + NET_NF_CONNTRACK_UDP_TIMEOUT=10, + NET_NF_CONNTRACK_UDP_TIMEOUT_STREAM=11, + NET_NF_CONNTRACK_ICMP_TIMEOUT=12, + NET_NF_CONNTRACK_GENERIC_TIMEOUT=13, + NET_NF_CONNTRACK_BUCKETS=14, + NET_NF_CONNTRACK_LOG_INVALID=15, + NET_NF_CONNTRACK_TCP_TIMEOUT_MAX_RETRANS=16, + NET_NF_CONNTRACK_TCP_LOOSE=17, + NET_NF_CONNTRACK_TCP_BE_LIBERAL=18, + NET_NF_CONNTRACK_TCP_MAX_RETRANS=19, + NET_NF_CONNTRACK_SCTP_TIMEOUT_CLOSED=20, + NET_NF_CONNTRACK_SCTP_TIMEOUT_COOKIE_WAIT=21, + NET_NF_CONNTRACK_SCTP_TIMEOUT_COOKIE_ECHOED=22, + NET_NF_CONNTRACK_SCTP_TIMEOUT_ESTABLISHED=23, + NET_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_SENT=24, + NET_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_RECD=25, + NET_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_ACK_SENT=26, + NET_NF_CONNTRACK_COUNT=27, + NET_NF_CONNTRACK_ICMPV6_TIMEOUT=28, + NET_NF_CONNTRACK_FRAG6_TIMEOUT=29, + NET_NF_CONNTRACK_FRAG6_LOW_THRESH=30, + NET_NF_CONNTRACK_FRAG6_HIGH_THRESH=31, +}; + /* /proc/sys/net/ipv4 */ enum { @@ -353,6 +390,7 @@ enum NET_TCP_BIC_BETA=108, NET_IPV4_ICMP_ERRORS_USE_INBOUND_IFADDR=109, NET_TCP_CONG_CONTROL=110, + NET_TCP_ABC=111, }; enum { diff --git a/include/linux/tcp.h b/include/linux/tcp.h index ac4ca44c75c..0e1da6602e0 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -307,6 +307,21 @@ struct tcp_sock { struct tcp_sack_block duplicate_sack[1]; /* D-SACK block */ struct tcp_sack_block selective_acks[4]; /* The SACKS themselves*/ + struct tcp_sack_block recv_sack_cache[4]; + + /* from STCP, retrans queue hinting */ + struct sk_buff* lost_skb_hint; + + struct sk_buff *scoreboard_skb_hint; + struct sk_buff *retransmit_skb_hint; + struct sk_buff *forward_skb_hint; + struct sk_buff *fastpath_skb_hint; + + int fastpath_cnt_hint; + int lost_cnt_hint; + int retransmit_cnt_hint; + int forward_cnt_hint; + __u16 advmss; /* Advertised MSS */ __u16 prior_ssthresh; /* ssthresh saved at recovery start */ __u32 lost_out; /* Lost packets */ @@ -326,6 +341,7 @@ struct tcp_sock { __u32 snd_up; /* Urgent pointer */ __u32 total_retrans; /* Total retransmits for entire connection */ + __u32 bytes_acked; /* Appropriate Byte Counting - RFC3465 */ unsigned int keepalive_time; /* time before keep alive takes place */ unsigned int keepalive_intvl; /* time interval between keep alive probes */ diff --git a/include/net/genetlink.h b/include/net/genetlink.h new file mode 100644 index 00000000000..52d8b1a73d5 --- /dev/null +++ b/include/net/genetlink.h @@ -0,0 +1,154 @@ +#ifndef __NET_GENERIC_NETLINK_H +#define __NET_GENERIC_NETLINK_H + +#include <linux/genetlink.h> +#include <net/netlink.h> + +/** + * struct genl_family - generic netlink family + * @id: protocol family idenfitier + * @hdrsize: length of user specific header in bytes + * @name: name of family + * @version: protocol version + * @maxattr: maximum number of attributes supported + * @attrbuf: buffer to store parsed attributes + * @ops_list: list of all assigned operations + * @family_list: family list + */ +struct genl_family +{ + unsigned int id; + unsigned int hdrsize; + char name[GENL_NAMSIZ]; + unsigned int version; + unsigned int maxattr; + struct module * owner; + struct nlattr ** attrbuf; /* private */ + struct list_head ops_list; /* private */ + struct list_head family_list; /* private */ +}; + +#define GENL_ADMIN_PERM 0x01 + +/** + * struct genl_info - receiving information + * @snd_seq: sending sequence number + * @snd_pid: netlink pid of sender + * @nlhdr: netlink message header + * @genlhdr: generic netlink message header + * @userhdr: user specific header + * @attrs: netlink attributes + */ +struct genl_info +{ + u32 snd_seq; + u32 snd_pid; + struct nlmsghdr * nlhdr; + struct genlmsghdr * genlhdr; + void * userhdr; + struct nlattr ** attrs; +}; + +/** + * struct genl_ops - generic netlink operations + * @cmd: command identifier + * @flags: flags + * @policy: attribute validation policy + * @doit: standard command callback + * @dumpit: callback for dumpers + * @ops_list: operations list + */ +struct genl_ops +{ + unsigned int cmd; + unsigned int flags; + struct nla_policy *policy; + int (*doit)(struct sk_buff *skb, + struct genl_info *info); + int (*dumpit)(struct sk_buff *skb, + struct netlink_callback *cb); + struct list_head ops_list; +}; + +extern int genl_register_family(struct genl_family *family); +extern int genl_unregister_family(struct genl_family *family); +extern int genl_register_ops(struct genl_family *, struct genl_ops *ops); +extern int genl_unregister_ops(struct genl_family *, struct genl_ops *ops); + +extern struct sock *genl_sock; + +/** + * genlmsg_put - Add generic netlink header to netlink message + * @skb: socket buffer holding the message + * @pid: netlink pid the message is addressed to + * @seq: sequence number (usually the one of the sender) + * @type: netlink message type + * @hdrlen: length of the user specific header + * @flags netlink message flags + * @cmd: generic netlink command + * @version: version + * + * Returns pointer to user specific header + */ +static inline void *genlmsg_put(struct sk_buff *skb, u32 pid, u32 seq, + int type, int hdrlen, int flags, + u8 cmd, u8 version) +{ + struct nlmsghdr *nlh; + struct genlmsghdr *hdr; + + nlh = nlmsg_put(skb, pid, seq, type, GENL_HDRLEN + hdrlen, flags); + if (nlh == NULL) + return NULL; + + hdr = nlmsg_data(nlh); + hdr->cmd = cmd; + hdr->version = version; + hdr->reserved = 0; + + return (char *) hdr + GENL_HDRLEN; +} + +/** + * genlmsg_end - Finalize a generic netlink message + * @skb: socket buffer the message is stored in + * @hdr: user specific header + */ +static inline int genlmsg_end(struct sk_buff *skb, void *hdr) +{ + return nlmsg_end(skb, hdr - GENL_HDRLEN - NLMSG_HDRLEN); +} + +/** + * genlmsg_cancel - Cancel construction of a generic netlink message + * @skb: socket buffer the message is stored in + * @hdr: generic netlink message header + */ +static inline int genlmsg_cancel(struct sk_buff *skb, void *hdr) +{ + return nlmsg_cancel(skb, hdr - GENL_HDRLEN - NLMSG_HDRLEN); +} + +/** + * genlmsg_multicast - multicast a netlink message + * @skb: netlink message as socket buffer + * @pid: own netlink pid to avoid sending to yourself + * @group: multicast group id + */ +static inline int genlmsg_multicast(struct sk_buff *skb, u32 pid, + unsigned int group) +{ + return nlmsg_multicast(genl_sock, skb, pid, group); +} + +/** + * genlmsg_unicast - unicast a netlink message + * @skb: netlink message as socket buffer + * @pid: netlink pid of the destination socket + */ +static inline int genlmsg_unicast(struct sk_buff *skb, u32 pid) +{ + return nlmsg_unicast(genl_sock, skb, pid); +} + +#endif /* __NET_GENERIC_NETLINK_H */ diff --git a/include/net/netfilter/ipv4/nf_conntrack_icmp.h b/include/net/netfilter/ipv4/nf_conntrack_icmp.h new file mode 100644 index 00000000000..3dd22cff23e --- /dev/null +++ b/include/net/netfilter/ipv4/nf_conntrack_icmp.h @@ -0,0 +1,11 @@ +#ifndef _NF_CONNTRACK_ICMP_H +#define _NF_CONNTRACK_ICMP_H +/* ICMP tracking. */ +#include <asm/atomic.h> + +struct ip_ct_icmp +{ + /* Optimization: when number in == number out, forget immediately. */ + atomic_t count; +}; +#endif /* _NF_CONNTRACK_ICMP_H */ diff --git a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h new file mode 100644 index 00000000000..25b081a730e --- /dev/null +++ b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h @@ -0,0 +1,43 @@ +/* + * IPv4 support for nf_conntrack. + * + * 23 Mar 2004: Yasuyuki Kozakai @ USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - move L3 protocol dependent part from include/linux/netfilter_ipv4/ + * ip_conntarck.h + */ + +#ifndef _NF_CONNTRACK_IPV4_H +#define _NF_CONNTRACK_IPV4_H + +#ifdef CONFIG_IP_NF_NAT_NEEDED +#include <linux/netfilter_ipv4/ip_nat.h> + +/* per conntrack: nat application helper private data */ +union ip_conntrack_nat_help { + /* insert nat helper private data here */ +}; + +struct nf_conntrack_ipv4_nat { + struct ip_nat_info info; + union ip_conntrack_nat_help help; +#if defined(CONFIG_IP_NF_TARGET_MASQUERADE) || \ + defined(CONFIG_IP_NF_TARGET_MASQUERADE_MODULE) + int masq_index; +#endif +}; +#endif /* CONFIG_IP_NF_NAT_NEEDED */ + +struct nf_conntrack_ipv4 { +#ifdef CONFIG_IP_NF_NAT_NEEDED + struct nf_conntrack_ipv4_nat *nat; +#endif +}; + +/* Returns new sk_buff, or NULL */ +struct sk_buff * +nf_ct_ipv4_ct_gather_frags(struct sk_buff *skb); + +/* call to create an explicit dependency on nf_conntrack_l3proto_ipv4. */ +extern void need_ip_conntrack(void); + +#endif /*_NF_CONNTRACK_IPV4_H*/ diff --git a/include/net/netfilter/ipv6/nf_conntrack_icmpv6.h b/include/net/netfilter/ipv6/nf_conntrack_icmpv6.h new file mode 100644 index 00000000000..86591afda29 --- /dev/null +++ b/include/net/netfilter/ipv6/nf_conntrack_icmpv6.h @@ -0,0 +1,27 @@ +/* + * ICMPv6 tracking. + * + * 21 Apl 2004: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - separated from nf_conntrack_icmp.h + * + * Derived from include/linux/netfiter_ipv4/ip_conntrack_icmp.h + */ + +#ifndef _NF_CONNTRACK_ICMPV6_H +#define _NF_CONNTRACK_ICMPV6_H +#include <asm/atomic.h> + +#ifndef ICMPV6_NI_QUERY +#define ICMPV6_NI_QUERY 139 +#endif +#ifndef ICMPV6_NI_REPLY +#define ICMPV6_NI_REPLY 140 +#endif + +struct nf_ct_icmpv6 +{ + /* Optimization: when number in == number out, forget immediately. */ + atomic_t count; +}; + +#endif /* _NF_CONNTRACK_ICMPV6_H */ diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h new file mode 100644 index 00000000000..cc482561079 --- /dev/null +++ b/include/net/netfilter/nf_conntrack.h @@ -0,0 +1,354 @@ +/* + * Connection state tracking for netfilter. This is separated from, + * but required by, the (future) NAT layer; it can also be used by an iptables + * extension. + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - generalize L3 protocol dependent part. + * + * Derived from include/linux/netfiter_ipv4/ip_conntrack.h + */ + +#ifndef _NF_CONNTRACK_H +#define _NF_CONNTRACK_H + +#include <linux/netfilter/nf_conntrack_common.h> + +#ifdef __KERNEL__ +#include <linux/config.h> +#include <linux/bitops.h> +#include <linux/compiler.h> +#include <asm/atomic.h> + +#include <linux/netfilter/nf_conntrack_tcp.h> +#include <linux/netfilter/nf_conntrack_sctp.h> +#include <net/netfilter/ipv4/nf_conntrack_icmp.h> +#include <net/netfilter/ipv6/nf_conntrack_icmpv6.h> + +#include <net/netfilter/nf_conntrack_tuple.h> + +/* per conntrack: protocol private data */ +union nf_conntrack_proto { + /* insert conntrack proto private data here */ + struct ip_ct_sctp sctp; + struct ip_ct_tcp tcp; + struct ip_ct_icmp icmp; + struct nf_ct_icmpv6 icmpv6; +}; + +union nf_conntrack_expect_proto { + /* insert expect proto private data here */ +}; + +/* Add protocol helper include file here */ +#include <linux/netfilter/nf_conntrack_ftp.h> + +/* per conntrack: application helper private data */ +union nf_conntrack_help { + /* insert conntrack helper private data (master) here */ + struct ip_ct_ftp_master ct_ftp_info; +}; + +#include <linux/types.h> +#include <linux/skbuff.h> + +#ifdef CONFIG_NETFILTER_DEBUG +#define NF_CT_ASSERT(x) \ +do { \ + if (!(x)) \ + /* Wooah! I'm tripping my conntrack in a frenzy of \ + netplay... */ \ + printk("NF_CT_ASSERT: %s:%i(%s)\n", \ + __FILE__, __LINE__, __FUNCTION__); \ +} while(0) +#else +#define NF_CT_ASSERT(x) +#endif + +struct nf_conntrack_helper; + +#include <net/netfilter/ipv4/nf_conntrack_ipv4.h> +struct nf_conn +{ + /* Usage count in here is 1 for hash table/destruct timer, 1 per skb, + plus 1 for any connection(s) we are `master' for */ + struct nf_conntrack ct_general; + + /* XXX should I move this to the tail ? - Y.K */ + /* These are my tuples; original and reply */ + struct nf_conntrack_tuple_hash tuplehash[IP_CT_DIR_MAX]; + + /* Have we seen traffic both ways yet? (bitset) */ + unsigned long status; + + /* Timer function; drops refcnt when it goes off. */ + struct timer_list timeout; + +#ifdef CONFIG_NF_CT_ACCT + /* Accounting Information (same cache line as other written members) */ + struct ip_conntrack_counter counters[IP_CT_DIR_MAX]; +#endif + /* If we were expected by an expectation, this will be it */ + struct nf_conn *master; + + /* Current number of expected connections */ + unsigned int expecting; + + /* Helper. if any */ + struct nf_conntrack_helper *helper; + + /* features - nat, helper, ... used by allocating system */ + u_int32_t features; + + /* Storage reserved for other modules: */ + + union nf_conntrack_proto proto; + +#if defined(CONFIG_NF_CONNTRACK_MARK) + u_int32_t mark; +#endif + + /* These members are dynamically allocated. */ + + union nf_conntrack_help *help; + + /* Layer 3 dependent members. (ex: NAT) */ + union { + struct nf_conntrack_ipv4 *ipv4; + } l3proto; + void *data[0]; +}; + +struct nf_conntrack_expect +{ + /* Internal linked list (global expectation list) */ + struct list_head list; + + /* We expect this tuple, with the following mask */ + struct nf_conntrack_tuple tuple, mask; + + /* Function to call after setup and insertion */ + void (*expectfn)(struct nf_conn *new, + struct nf_conntrack_expect *this); + + /* The conntrack of the master connection */ + struct nf_conn *master; + + /* Timer function; deletes the expectation. */ + struct timer_list timeout; + + /* Usage count. */ + atomic_t use; + + /* Flags */ + unsigned int flags; + +#ifdef CONFIG_NF_NAT_NEEDED + /* This is the original per-proto part, used to map the + * expected connection the way the recipient expects. */ + union nf_conntrack_manip_proto saved_proto; + /* Direction relative to the master connection. */ + enum ip_conntrack_dir dir; +#endif +}; + +#define NF_CT_EXPECT_PERMANENT 0x1 + +static inline struct nf_conn * +nf_ct_tuplehash_to_ctrack(const struct nf_conntrack_tuple_hash *hash) +{ + return container_of(hash, struct nf_conn, + tuplehash[hash->tuple.dst.dir]); +} + +/* get master conntrack via master expectation */ +#define master_ct(conntr) (conntr->master) + +/* Alter reply tuple (maybe alter helper). */ +extern void +nf_conntrack_alter_reply(struct nf_conn *conntrack, + const struct nf_conntrack_tuple *newreply); + +/* Is this tuple taken? (ignoring any belonging to the given + conntrack). */ +extern int +nf_conntrack_tuple_taken(const struct nf_conntrack_tuple *tuple, + const struct nf_conn *ignored_conntrack); + +/* Return conntrack_info and tuple hash for given skb. */ +static inline struct nf_conn * +nf_ct_get(const struct sk_buff *skb, enum ip_conntrack_info *ctinfo) +{ + *ctinfo = skb->nfctinfo; + return (struct nf_conn *)skb->nfct; +} + +/* decrement reference count on a conntrack */ +static inline void nf_ct_put(struct nf_conn *ct) +{ + NF_CT_ASSERT(ct); + nf_conntrack_put(&ct->ct_general); +} + +/* call to create an explicit dependency on nf_conntrack. */ +extern void need_nf_conntrack(void); + +extern int nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse, + const struct nf_conntrack_tuple *orig); + +extern void __nf_ct_refresh_acct(struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + const struct sk_buff *skb, + unsigned long extra_jiffies, + int do_acct); + +/* Refresh conntrack for this many jiffies and do accounting */ +static inline void nf_ct_refresh_acct(struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + const struct sk_buff *skb, + unsigned long extra_jiffies) +{ + __nf_ct_refresh_acct(ct, ctinfo, skb, extra_jiffies, 1); +} + +/* Refresh conntrack for this many jiffies */ +static inline void nf_ct_refresh(struct nf_conn *ct, + const struct sk_buff *skb, + unsigned long extra_jiffies) +{ + __nf_ct_refresh_acct(ct, 0, skb, extra_jiffies, 0); +} + +/* These are for NAT. Icky. */ +/* Update TCP window tracking data when NAT mangles the packet */ +extern void nf_conntrack_tcp_update(struct sk_buff *skb, + unsigned int dataoff, + struct nf_conn *conntrack, + int dir); + +/* Call me when a conntrack is destroyed. */ +extern void (*nf_conntrack_destroyed)(struct nf_conn *conntrack); + +/* Fake conntrack entry for untracked connections */ +extern struct nf_conn nf_conntrack_untracked; + +extern int nf_ct_no_defrag; + +/* Iterate over all conntracks: if iter returns true, it's deleted. */ +extern void +nf_ct_iterate_cleanup(int (*iter)(struct nf_conn *i, void *data), void *data); +extern void nf_conntrack_free(struct nf_conn *ct); +extern struct nf_conn * +nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, + const struct nf_conntrack_tuple *repl); + +/* It's confirmed if it is, or has been in the hash table. */ +static inline int nf_ct_is_confirmed(struct nf_conn *ct) +{ + return test_bit(IPS_CONFIRMED_BIT, &ct->status); +} + +static inline int nf_ct_is_dying(struct nf_conn *ct) +{ + return test_bit(IPS_DYING_BIT, &ct->status); +} + +extern unsigned int nf_conntrack_htable_size; + +#define NF_CT_STAT_INC(count) (__get_cpu_var(nf_conntrack_stat).count++) + +#ifdef CONFIG_NF_CONNTRACK_EVENTS +#include <linux/notifier.h> +#include <linux/interrupt.h> + +struct nf_conntrack_ecache { + struct nf_conn *ct; + unsigned int events; +}; +DECLARE_PER_CPU(struct nf_conntrack_ecache, nf_conntrack_ecache); + +#define CONNTRACK_ECACHE(x) (__get_cpu_var(nf_conntrack_ecache).x) + +extern struct notifier_block *nf_conntrack_chain; +extern struct notifier_block *nf_conntrack_expect_chain; + +static inline int nf_conntrack_register_notifier(struct notifier_block *nb) +{ + return notifier_chain_register(&nf_conntrack_chain, nb); +} + +static inline int nf_conntrack_unregister_notifier(struct notifier_block *nb) +{ + return notifier_chain_unregister(&nf_conntrack_chain, nb); +} + +static inline int +nf_conntrack_expect_register_notifier(struct notifier_block *nb) +{ + return notifier_chain_register(&nf_conntrack_expect_chain, nb); +} + +static inline int +nf_conntrack_expect_unregister_notifier(struct notifier_block *nb) +{ + return notifier_chain_unregister(&nf_conntrack_expect_chain, nb); +} + +extern void nf_ct_deliver_cached_events(const struct nf_conn *ct); +extern void __nf_ct_event_cache_init(struct nf_conn *ct); + +static inline void +nf_conntrack_event_cache(enum ip_conntrack_events event, + const struct sk_buff *skb) +{ + struct nf_conn *ct = (struct nf_conn *)skb->nfct; + struct nf_conntrack_ecache *ecache; + + local_bh_disable(); + ecache = &__get_cpu_var(nf_conntrack_ecache); + if (ct != ecache->ct) + __nf_ct_event_cache_init(ct); + ecache->events |= event; + local_bh_enable(); +} + +static inline void nf_conntrack_event(enum ip_conntrack_events event, + struct nf_conn *ct) +{ + if (nf_ct_is_confirmed(ct) && !nf_ct_is_dying(ct)) + notifier_call_chain(&nf_conntrack_chain, event, ct); +} + +static inline void +nf_conntrack_expect_event(enum ip_conntrack_expect_events event, + struct nf_conntrack_expect *exp) +{ + notifier_call_chain(&nf_conntrack_expect_chain, event, exp); +} +#else /* CONFIG_NF_CONNTRACK_EVENTS */ +static inline void nf_conntrack_event_cache(enum ip_conntrack_events event, + const struct sk_buff *skb) {} +static inline void nf_conntrack_event(enum ip_conntrack_events event, + struct nf_conn *ct) {} +static inline void nf_ct_deliver_cached_events(const struct nf_conn *ct) {} +static inline void +nf_conntrack_expect_event(enum ip_conntrack_expect_events event, + struct nf_conntrack_expect *exp) {} +#endif /* CONFIG_NF_CONNTRACK_EVENTS */ + +/* no helper, no nat */ +#define NF_CT_F_BASIC 0 +/* for helper */ +#define NF_CT_F_HELP 1 +/* for nat. */ +#define NF_CT_F_NAT 2 +#define NF_CT_F_NUM 4 + +extern int +nf_conntrack_register_cache(u_int32_t features, const char *name, size_t size, + int (*init_conntrack)(struct nf_conn *, u_int32_t)); +extern void +nf_conntrack_unregister_cache(u_int32_t features); + +#endif /* __KERNEL__ */ +#endif /* _NF_CONNTRACK_H */ diff --git a/include/net/netfilter/nf_conntrack_compat.h b/include/net/netfilter/nf_conntrack_compat.h new file mode 100644 index 00000000000..3cac19fb364 --- /dev/null +++ b/include/net/netfilter/nf_conntrack_compat.h @@ -0,0 +1,108 @@ +#ifndef _NF_CONNTRACK_COMPAT_H +#define _NF_CONNTRACK_COMPAT_H + +#ifdef __KERNEL__ + +#if defined(CONFIG_IP_NF_CONNTRACK) || defined(CONFIG_IP_NF_CONNTRACK_MODULE) + +#include <linux/netfilter_ipv4/ip_conntrack.h> + +#ifdef CONFIG_IP_NF_CONNTRACK_MARK +static inline u_int32_t *nf_ct_get_mark(const struct sk_buff *skb, + u_int32_t *ctinfo) +{ + struct ip_conntrack *ct = ip_conntrack_get(skb, ctinfo); + + if (ct) + return &ct->mark; + else + return NULL; +} +#endif /* CONFIG_IP_NF_CONNTRACK_MARK */ + +#ifdef CONFIG_IP_NF_CT_ACCT +static inline struct ip_conntrack_counter * +nf_ct_get_counters(const struct sk_buff *skb) +{ + enum ip_conntrack_info ctinfo; + struct ip_conntrack *ct = ip_conntrack_get(skb, &ctinfo); + + if (ct) + return ct->counters; + else + return NULL; +} +#endif /* CONFIG_IP_NF_CT_ACCT */ + +static inline int nf_ct_is_untracked(const struct sk_buff *skb) +{ + return (skb->nfct == &ip_conntrack_untracked.ct_general); +} + +static inline void nf_ct_untrack(struct sk_buff *skb) +{ + skb->nfct = &ip_conntrack_untracked.ct_general; +} + +static inline int nf_ct_get_ctinfo(const struct sk_buff *skb, + enum ip_conntrack_info *ctinfo) +{ + struct ip_conntrack *ct = ip_conntrack_get(skb, ctinfo); + return (ct != NULL); +} + +#else /* CONFIG_IP_NF_CONNTRACK */ + +#include <net/netfilter/ipv4/nf_conntrack_ipv4.h> +#include <net/netfilter/nf_conntrack.h> + +#ifdef CONFIG_NF_CONNTRACK_MARK + +static inline u_int32_t *nf_ct_get_mark(const struct sk_buff *skb, + u_int32_t *ctinfo) +{ + struct nf_conn *ct = nf_ct_get(skb, ctinfo); + + if (ct) + return &ct->mark; + else + return NULL; +} +#endif /* CONFIG_NF_CONNTRACK_MARK */ + +#ifdef CONFIG_NF_CT_ACCT +static inline struct ip_conntrack_counter * +nf_ct_get_counters(const struct sk_buff *skb) +{ + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = nf_ct_get(skb, &ctinfo); + + if (ct) + return ct->counters; + else + return NULL; +} +#endif /* CONFIG_NF_CT_ACCT */ + +static inline int nf_ct_is_untracked(const struct sk_buff *skb) +{ + return (skb->nfct == &nf_conntrack_untracked.ct_general); +} + +static inline void nf_ct_untrack(struct sk_buff *skb) +{ + skb->nfct = &nf_conntrack_untracked.ct_general; +} + +static inline int nf_ct_get_ctinfo(const struct sk_buff *skb, + enum ip_conntrack_info *ctinfo) +{ + struct nf_conn *ct = nf_ct_get(skb, ctinfo); + return (ct != NULL); +} + +#endif /* CONFIG_IP_NF_CONNTRACK */ + +#endif /* __KERNEL__ */ + +#endif /* _NF_CONNTRACK_COMPAT_H */ diff --git a/include/net/netfilter/nf_conntrack_core.h b/include/net/netfilter/nf_conntrack_core.h new file mode 100644 index 00000000000..da254525a4c --- /dev/null +++ b/include/net/netfilter/nf_conntrack_core.h @@ -0,0 +1,76 @@ +/* + * This header is used to share core functionality between the + * standalone connection tracking module, and the compatibility layer's use + * of connection tracking. + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - generalize L3 protocol dependent part. + * + * Derived from include/linux/netfiter_ipv4/ip_conntrack_core.h + */ + +#ifndef _NF_CONNTRACK_CORE_H +#define _NF_CONNTRACK_CORE_H + +#include <linux/netfilter.h> + +/* This header is used to share core functionality between the + standalone connection tracking module, and the compatibility layer's use + of connection tracking. */ +extern unsigned int nf_conntrack_in(int pf, + unsigned int hooknum, + struct sk_buff **pskb); + +extern int nf_conntrack_init(void); +extern void nf_conntrack_cleanup(void); + +struct nf_conntrack_l3proto; +extern struct nf_conntrack_l3proto *nf_ct_find_l3proto(u_int16_t pf); +/* Like above, but you already have conntrack read lock. */ +extern struct nf_conntrack_l3proto *__nf_ct_find_l3proto(u_int16_t l3proto); + +struct nf_conntrack_protocol; + +extern int +nf_ct_get_tuple(const struct sk_buff *skb, + unsigned int nhoff, + unsigned int dataoff, + u_int16_t l3num, + u_int8_t protonum, + struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_l3proto *l3proto, + const struct nf_conntrack_protocol *protocol); + +extern int +nf_ct_invert_tuple(struct nf_conntrack_tuple *inverse, + const struct nf_conntrack_tuple *orig, + const struct nf_conntrack_l3proto *l3proto, + const struct nf_conntrack_protocol *protocol); + +/* Find a connection corresponding to a tuple. */ +extern struct nf_conntrack_tuple_hash * +nf_conntrack_find_get(const struct nf_conntrack_tuple *tuple, + const struct nf_conn *ignored_conntrack); + +extern int __nf_conntrack_confirm(struct sk_buff **pskb); + +/* Confirm a connection: returns NF_DROP if packet must be dropped. */ +static inline int nf_conntrack_confirm(struct sk_buff **pskb) +{ + struct nf_conn *ct = (struct nf_conn *)(*pskb)->nfct; + int ret = NF_ACCEPT; + + if (ct) { + if (!nf_ct_is_confirmed(ct)) + ret = __nf_conntrack_confirm(pskb); + nf_ct_deliver_cached_events(ct); + } + return ret; +} + +extern void __nf_conntrack_attach(struct sk_buff *nskb, struct sk_buff *skb); + +extern struct list_head *nf_conntrack_hash; +extern struct list_head nf_conntrack_expect_list; +extern rwlock_t nf_conntrack_lock ; +#endif /* _NF_CONNTRACK_CORE_H */ diff --git a/include/net/netfilter/nf_conntrack_helper.h b/include/net/netfilter/nf_conntrack_helper.h new file mode 100644 index 00000000000..5a66b2a3a62 --- /dev/null +++ b/include/net/netfilter/nf_conntrack_helper.h @@ -0,0 +1,51 @@ +/* + * connection tracking helpers. + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - generalize L3 protocol dependent part. + * + * Derived from include/linux/netfiter_ipv4/ip_conntrack_helper.h + */ + +#ifndef _NF_CONNTRACK_HELPER_H +#define _NF_CONNTRACK_HELPER_H +#include <net/netfilter/nf_conntrack.h> + +struct module; + +struct nf_conntrack_helper +{ + struct list_head list; /* Internal use. */ + + const char *name; /* name of the module */ + struct module *me; /* pointer to self */ + unsigned int max_expected; /* Maximum number of concurrent + * expected connections */ + unsigned int timeout; /* timeout for expecteds */ + + /* Mask of things we will help (compared against server response) */ + struct nf_conntrack_tuple tuple; + struct nf_conntrack_tuple mask; + + /* Function to call when data passes; return verdict, or -1 to + invalidate. */ + int (*help)(struct sk_buff **pskb, + unsigned int protoff, + struct nf_conn *ct, + enum ip_conntrack_info conntrackinfo); +}; + +extern int nf_conntrack_helper_register(struct nf_conntrack_helper *); +extern void nf_conntrack_helper_unregister(struct nf_conntrack_helper *); + +/* Allocate space for an expectation: this is mandatory before calling + nf_conntrack_expect_related. You will have to call put afterwards. */ +extern struct nf_conntrack_expect * +nf_conntrack_expect_alloc(struct nf_conn *master); +extern void nf_conntrack_expect_put(struct nf_conntrack_expect *exp); + +/* Add an expected connection: can have more than one per connection */ +extern int nf_conntrack_expect_related(struct nf_conntrack_expect *exp); +extern void nf_conntrack_unexpect_related(struct nf_conntrack_expect *exp); + +#endif /*_NF_CONNTRACK_HELPER_H*/ diff --git a/include/net/netfilter/nf_conntrack_l3proto.h b/include/net/netfilter/nf_conntrack_l3proto.h new file mode 100644 index 00000000000..01663e5b33d --- /dev/null +++ b/include/net/netfilter/nf_conntrack_l3proto.h @@ -0,0 +1,93 @@ +/* + * Copyright (C)2003,2004 USAGI/WIDE Project + * + * Header for use in defining a given L3 protocol for connection tracking. + * + * Author: + * Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * + * Derived from include/netfilter_ipv4/ip_conntrack_protocol.h + */ + +#ifndef _NF_CONNTRACK_L3PROTO_H +#define _NF_CONNTRACK_L3PROTO_H +#include <linux/seq_file.h> +#include <net/netfilter/nf_conntrack.h> + +struct nf_conntrack_l3proto +{ + /* Next pointer. */ + struct list_head list; + + /* L3 Protocol Family number. ex) PF_INET */ + u_int16_t l3proto; + + /* Protocol name */ + const char *name; + + /* + * Try to fill in the third arg: nhoff is offset of l3 proto + * hdr. Return true if possible. + */ + int (*pkt_to_tuple)(const struct sk_buff *skb, unsigned int nhoff, + struct nf_conntrack_tuple *tuple); + + /* + * Invert the per-proto part of the tuple: ie. turn xmit into reply. + * Some packets can't be inverted: return 0 in that case. + */ + int (*invert_tuple)(struct nf_conntrack_tuple *inverse, + const struct nf_conntrack_tuple *orig); + + /* Print out the per-protocol part of the tuple. */ + int (*print_tuple)(struct seq_file *s, + const struct nf_conntrack_tuple *); + + /* Print out the private part of the conntrack. */ + int (*print_conntrack)(struct seq_file *s, const struct nf_conn *); + + /* Returns verdict for packet, or -1 for invalid. */ + int (*packet)(struct nf_conn *conntrack, + const struct sk_buff *skb, + enum ip_conntrack_info ctinfo); + + /* + * Called when a new connection for this protocol found; + * returns TRUE if it's OK. If so, packet() called next. + */ + int (*new)(struct nf_conn *conntrack, const struct sk_buff *skb); + + /* Called when a conntrack entry is destroyed */ + void (*destroy)(struct nf_conn *conntrack); + + /* + * Called before tracking. + * *dataoff: offset of protocol header (TCP, UDP,...) in *pskb + * *protonum: protocol number + */ + int (*prepare)(struct sk_buff **pskb, unsigned int hooknum, + unsigned int *dataoff, u_int8_t *protonum); + + u_int32_t (*get_features)(const struct nf_conntrack_tuple *tuple); + + /* Module (if any) which this is connected to. */ + struct module *me; +}; + +extern struct nf_conntrack_l3proto *nf_ct_l3protos[AF_MAX]; + +/* Protocol registration. */ +extern int nf_conntrack_l3proto_register(struct nf_conntrack_l3proto *proto); +extern void nf_conntrack_l3proto_unregister(struct nf_conntrack_l3proto *proto); + +static inline struct nf_conntrack_l3proto * +nf_ct_find_l3proto(u_int16_t l3proto) +{ + return nf_ct_l3protos[l3proto]; +} + +/* Existing built-in protocols */ +extern struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4; +extern struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6; +extern struct nf_conntrack_l3proto nf_conntrack_generic_l3proto; +#endif /*_NF_CONNTRACK_L3PROTO_H*/ diff --git a/include/net/netfilter/nf_conntrack_protocol.h b/include/net/netfilter/nf_conntrack_protocol.h new file mode 100644 index 00000000000..b3afda35397 --- /dev/null +++ b/include/net/netfilter/nf_conntrack_protocol.h @@ -0,0 +1,105 @@ +/* + * Header for use in defining a given protocol for connection tracking. + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - generalized L3 protocol dependent part. + * + * Derived from include/linux/netfiter_ipv4/ip_conntrack_protcol.h + */ + +#ifndef _NF_CONNTRACK_PROTOCOL_H +#define _NF_CONNTRACK_PROTOCOL_H +#include <net/netfilter/nf_conntrack.h> + +struct seq_file; + +struct nf_conntrack_protocol +{ + /* Next pointer. */ + struct list_head list; + + /* L3 Protocol number. */ + u_int16_t l3proto; + + /* Protocol number. */ + u_int8_t proto; + + /* Protocol name */ + const char *name; + + /* Try to fill in the third arg: dataoff is offset past network protocol + hdr. Return true if possible. */ + int (*pkt_to_tuple)(const struct sk_buff *skb, + unsigned int dataoff, + struct nf_conntrack_tuple *tuple); + + /* Invert the per-proto part of the tuple: ie. turn xmit into reply. + * Some packets can't be inverted: return 0 in that case. + */ + int (*invert_tuple)(struct nf_conntrack_tuple *inverse, + const struct nf_conntrack_tuple *orig); + + /* Print out the per-protocol part of the tuple. Return like seq_* */ + int (*print_tuple)(struct seq_file *s, + const struct nf_conntrack_tuple *); + + /* Print out the private part of the conntrack. */ + int (*print_conntrack)(struct seq_file *s, const struct nf_conn *); + + /* Returns verdict for packet, or -1 for invalid. */ + int (*packet)(struct nf_conn *conntrack, + const struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info ctinfo, + int pf, + unsigned int hooknum); + + /* Called when a new connection for this protocol found; + * returns TRUE if it's OK. If so, packet() called next. */ + int (*new)(struct nf_conn *conntrack, const struct sk_buff *skb, + unsigned int dataoff); + + /* Called when a conntrack entry is destroyed */ + void (*destroy)(struct nf_conn *conntrack); + + int (*error)(struct sk_buff *skb, unsigned int dataoff, + enum ip_conntrack_info *ctinfo, + int pf, unsigned int hooknum); + + /* Module (if any) which this is connected to. */ + struct module *me; +}; + +/* Existing built-in protocols */ +extern struct nf_conntrack_protocol nf_conntrack_protocol_tcp6; +extern struct nf_conntrack_protocol nf_conntrack_protocol_udp4; +extern struct nf_conntrack_protocol nf_conntrack_protocol_udp6; +extern struct nf_conntrack_protocol nf_conntrack_generic_protocol; + +#define MAX_NF_CT_PROTO 256 +extern struct nf_conntrack_protocol **nf_ct_protos[PF_MAX]; + +extern struct nf_conntrack_protocol * +nf_ct_find_proto(u_int16_t l3proto, u_int8_t protocol); + +/* Protocol registration. */ +extern int nf_conntrack_protocol_register(struct nf_conntrack_protocol *proto); +extern void nf_conntrack_protocol_unregister(struct nf_conntrack_protocol *proto); + +/* Log invalid packets */ +extern unsigned int nf_ct_log_invalid; + +#ifdef CONFIG_SYSCTL +#ifdef DEBUG_INVALID_PACKETS +#define LOG_INVALID(proto) \ + (nf_ct_log_invalid == (proto) || nf_ct_log_invalid == IPPROTO_RAW) +#else +#define LOG_INVALID(proto) \ + ((nf_ct_log_invalid == (proto) || nf_ct_log_invalid == IPPROTO_RAW) \ + && net_ratelimit()) +#endif +#else +#define LOG_INVALID(proto) 0 +#endif /* CONFIG_SYSCTL */ + +#endif /*_NF_CONNTRACK_PROTOCOL_H*/ diff --git a/include/net/netfilter/nf_conntrack_tuple.h b/include/net/netfilter/nf_conntrack_tuple.h new file mode 100644 index 00000000000..14ce790e5c6 --- /dev/null +++ b/include/net/netfilter/nf_conntrack_tuple.h @@ -0,0 +1,190 @@ +/* + * Definitions and Declarations for tuple. + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - generalize L3 protocol dependent part. + * + * Derived from include/linux/netfiter_ipv4/ip_conntrack_tuple.h + */ + +#ifndef _NF_CONNTRACK_TUPLE_H +#define _NF_CONNTRACK_TUPLE_H + +#include <linux/netfilter/nf_conntrack_tuple_common.h> + +/* A `tuple' is a structure containing the information to uniquely + identify a connection. ie. if two packets have the same tuple, they + are in the same connection; if not, they are not. + + We divide the structure along "manipulatable" and + "non-manipulatable" lines, for the benefit of the NAT code. +*/ + +#define NF_CT_TUPLE_L3SIZE 4 + +/* The l3 protocol-specific manipulable parts of the tuple: always in + network order! */ +union nf_conntrack_man_l3proto { + u_int32_t all[NF_CT_TUPLE_L3SIZE]; + u_int32_t ip; + u_int32_t ip6[4]; +}; + +/* The protocol-specific manipulable parts of the tuple: always in + network order! */ +union nf_conntrack_man_proto +{ + /* Add other protocols here. */ + u_int16_t all; + + struct { + u_int16_t port; + } tcp; + struct { + u_int16_t port; + } udp; + struct { + u_int16_t id; + } icmp; + struct { + u_int16_t port; + } sctp; +}; + +/* The manipulable part of the tuple. */ +struct nf_conntrack_man +{ + union nf_conntrack_man_l3proto u3; + union nf_conntrack_man_proto u; + /* Layer 3 protocol */ + u_int16_t l3num; +}; + +/* This contains the information to distinguish a connection. */ +struct nf_conntrack_tuple +{ + struct nf_conntrack_man src; + + /* These are the parts of the tuple which are fixed. */ + struct { + union { + u_int32_t all[NF_CT_TUPLE_L3SIZE]; + u_int32_t ip; + u_int32_t ip6[4]; + } u3; + union { + /* Add other protocols here. */ + u_int16_t all; + + struct { + u_int16_t port; + } tcp; + struct { + u_int16_t port; + } udp; + struct { + u_int8_t type, code; + } icmp; + struct { + u_int16_t port; + } sctp; + } u; + + /* The protocol. */ + u_int8_t protonum; + + /* The direction (for tuplehash) */ + u_int8_t dir; + } dst; +}; + +/* This is optimized opposed to a memset of the whole structure. Everything we + * really care about is the source/destination unions */ +#define NF_CT_TUPLE_U_BLANK(tuple) \ + do { \ + (tuple)->src.u.all = 0; \ + (tuple)->dst.u.all = 0; \ + memset(&(tuple)->src.u3, 0, sizeof((tuple)->src.u3)); \ + memset(&(tuple)->dst.u3, 0, sizeof((tuple)->dst.u3)); \ + } while (0) + +#ifdef __KERNEL__ + +#define NF_CT_DUMP_TUPLE(tp) \ +DEBUGP("tuple %p: %u %u %04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x %hu -> %04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x %hu\n", \ + (tp), (tp)->src.l3num, (tp)->dst.protonum, \ + NIP6(*(struct in6_addr *)(tp)->src.u3.all), ntohs((tp)->src.u.all), \ + NIP6(*(struct in6_addr *)(tp)->dst.u3.all), ntohs((tp)->dst.u.all)) + +/* If we're the first tuple, it's the original dir. */ +#define NF_CT_DIRECTION(h) \ + ((enum ip_conntrack_dir)(h)->tuple.dst.dir) + +/* Connections have two entries in the hash table: one for each way */ +struct nf_conntrack_tuple_hash +{ + struct list_head list; + + struct nf_conntrack_tuple tuple; +}; + +#endif /* __KERNEL__ */ + +static inline int nf_ct_tuple_src_equal(const struct nf_conntrack_tuple *t1, + const struct nf_conntrack_tuple *t2) +{ + return (t1->src.u3.all[0] == t2->src.u3.all[0] && + t1->src.u3.all[1] == t2->src.u3.all[1] && + t1->src.u3.all[2] == t2->src.u3.all[2] && + t1->src.u3.all[3] == t2->src.u3.all[3] && + t1->src.u.all == t2->src.u.all && + t1->src.l3num == t2->src.l3num && + t1->dst.protonum == t2->dst.protonum); +} + +static inline int nf_ct_tuple_dst_equal(const struct nf_conntrack_tuple *t1, + const struct nf_conntrack_tuple *t2) +{ + return (t1->dst.u3.all[0] == t2->dst.u3.all[0] && + t1->dst.u3.all[1] == t2->dst.u3.all[1] && + t1->dst.u3.all[2] == t2->dst.u3.all[2] && + t1->dst.u3.all[3] == t2->dst.u3.all[3] && + t1->dst.u.all == t2->dst.u.all && + t1->src.l3num == t2->src.l3num && + t1->dst.protonum == t2->dst.protonum); +} + +static inline int nf_ct_tuple_equal(const struct nf_conntrack_tuple *t1, + const struct nf_conntrack_tuple *t2) +{ + return nf_ct_tuple_src_equal(t1, t2) && nf_ct_tuple_dst_equal(t1, t2); +} + +static inline int nf_ct_tuple_mask_cmp(const struct nf_conntrack_tuple *t, + const struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *mask) +{ + int count = 0; + + for (count = 0; count < NF_CT_TUPLE_L3SIZE; count++){ + if ((t->src.u3.all[count] ^ tuple->src.u3.all[count]) & + mask->src.u3.all[count]) + return 0; + } + + for (count = 0; count < NF_CT_TUPLE_L3SIZE; count++){ + if ((t->dst.u3.all[count] ^ tuple->dst.u3.all[count]) & + mask->dst.u3.all[count]) + return 0; + } + + if ((t->src.u.all ^ tuple->src.u.all) & mask->src.u.all || + (t->dst.u.all ^ tuple->dst.u.all) & mask->dst.u.all || + (t->src.l3num ^ tuple->src.l3num) & mask->src.l3num || + (t->dst.protonum ^ tuple->dst.protonum) & mask->dst.protonum) + return 0; + + return 1; +} + +#endif /* _NF_CONNTRACK_TUPLE_H */ diff --git a/include/net/netlink.h b/include/net/netlink.h new file mode 100644 index 00000000000..640c26a90cf --- /dev/null +++ b/include/net/netlink.h @@ -0,0 +1,883 @@ +#ifndef __NET_NETLINK_H +#define __NET_NETLINK_H + +#include <linux/types.h> +#include <linux/netlink.h> + +/* ======================================================================== + * Netlink Messages and Attributes Interface (As Seen On TV) + * ------------------------------------------------------------------------ + * Messages Interface + * ------------------------------------------------------------------------ + * + * Message Format: + * <--- nlmsg_total_size(payload) ---> + * <-- nlmsg_msg_size(payload) -> + * +----------+- - -+-------------+- - -+-------- - - + * | nlmsghdr | Pad | Payload | Pad | nlmsghdr + * +----------+- - -+-------------+- - -+-------- - - + * nlmsg_data(nlh)---^ ^ + * nlmsg_next(nlh)-----------------------+ + * + * Payload Format: + * <---------------------- nlmsg_len(nlh) ---------------------> + * <------ hdrlen ------> <- nlmsg_attrlen(nlh, hdrlen) -> + * +----------------------+- - -+--------------------------------+ + * | Family Header | Pad | Attributes | + * +----------------------+- - -+--------------------------------+ + * nlmsg_attrdata(nlh, hdrlen)---^ + * + * Data Structures: + * struct nlmsghdr netlink message header + * + * Message Construction: + * nlmsg_new() create a new netlink message + * nlmsg_put() add a netlink message to an skb + * nlmsg_put_answer() callback based nlmsg_put() + * nlmsg_end() finanlize netlink message + * nlmsg_cancel() cancel message construction + * nlmsg_free() free a netlink message + * + * Message Sending: + * nlmsg_multicast() multicast message to several groups + * nlmsg_unicast() unicast a message to a single socket + * + * Message Length Calculations: + * nlmsg_msg_size(payload) length of message w/o padding + * nlmsg_total_size(payload) length of message w/ padding + * nlmsg_padlen(payload) length of padding at tail + * + * Message Payload Access: + * nlmsg_data(nlh) head of message payload + * nlmsg_len(nlh) length of message payload + * nlmsg_attrdata(nlh, hdrlen) head of attributes data + * nlmsg_attrlen(nlh, hdrlen) length of attributes data + * + * Message Parsing: + * nlmsg_ok(nlh, remaining) does nlh fit into remaining bytes? + * nlmsg_next(nlh, remaining) get next netlink message + * nlmsg_parse() parse attributes of a message + * nlmsg_find_attr() find an attribute in a message + * nlmsg_for_each_msg() loop over all messages + * nlmsg_validate() validate netlink message incl. attrs + * nlmsg_for_each_attr() loop over all attributes + * + * ------------------------------------------------------------------------ + * Attributes Interface + * ------------------------------------------------------------------------ + * + * Attribute Format: + * <------- nla_total_size(payload) -------> + * <---- nla_attr_size(payload) -----> + * +----------+- - -+- - - - - - - - - +- - -+-------- - - + * | Header | Pad | Payload | Pad | Header + * +----------+- - -+- - - - - - - - - +- - -+-------- - - + * <- nla_len(nla) -> ^ + * nla_data(nla)----^ | + * nla_next(nla)-----------------------------' + * + * Data Structures: + * struct nlattr netlink attribtue header + * + * Attribute Construction: + * nla_reserve(skb, type, len) reserve skb tailroom for an attribute + * nla_put(skb, type, len, data) add attribute to skb + * + * Attribute Construction for Basic Types: + * nla_put_u8(skb, type, value) add u8 attribute to skb + * nla_put_u16(skb, type, value) add u16 attribute to skb + * nla_put_u32(skb, type, value) add u32 attribute to skb + * nla_put_u64(skb, type, value) add u64 attribute to skb + * nla_put_string(skb, type, str) add string attribute to skb + * nla_put_flag(skb, type) add flag attribute to skb + * nla_put_msecs(skb, type, jiffies) add msecs attribute to skb + * + * Exceptions Based Attribute Construction: + * NLA_PUT(skb, type, len, data) add attribute to skb + * NLA_PUT_U8(skb, type, value) add u8 attribute to skb + * NLA_PUT_U16(skb, type, value) add u16 attribute to skb + * NLA_PUT_U32(skb, type, value) add u32 attribute to skb + * NLA_PUT_U64(skb, type, value) add u64 attribute to skb + * NLA_PUT_STRING(skb, type, str) add string attribute to skb + * NLA_PUT_FLAG(skb, type) add flag attribute to skb + * NLA_PUT_MSECS(skb, type, jiffies) add msecs attribute to skb + * + * The meaning of these functions is equal to their lower case + * variants but they jump to the label nla_put_failure in case + * of a failure. + * + * Nested Attributes Construction: + * nla_nest_start(skb, type) start a nested attribute + * nla_nest_end(skb, nla) finalize a nested attribute + * nla_nest_cancel(skb, nla) cancel nested attribute construction + * + * Attribute Length Calculations: + * nla_attr_size(payload) length of attribute w/o padding + * nla_total_size(payload) length of attribute w/ padding + * nla_padlen(payload) length of padding + * + * Attribute Payload Access: + * nla_data(nla) head of attribute payload + * nla_len(nla) length of attribute payload + * + * Attribute Payload Access for Basic Types: + * nla_get_u8(nla) get payload for a u8 attribute + * nla_get_u16(nla) get payload for a u16 attribute + * nla_get_u32(nla) get payload for a u32 attribute + * nla_get_u64(nla) get payload for a u64 attribute + * nla_get_flag(nla) return 1 if flag is true + * nla_get_msecs(nla) get payload for a msecs attribute + * + * Attribute Misc: + * nla_memcpy(dest, nla, count) copy attribute into memory + * nla_memcmp(nla, data, size) compare attribute with memory area + * nla_strlcpy(dst, nla, size) copy attribute to a sized string + * nla_strcmp(nla, str) compare attribute with string + * + * Attribute Parsing: + * nla_ok(nla, remaining) does nla fit into remaining bytes? + * nla_next(nla, remaining) get next netlink attribute + * nla_validate() validate a stream of attributes + * nla_find() find attribute in stream of attributes + * nla_parse() parse and validate stream of attrs + * nla_parse_nested() parse nested attribuets + * nla_for_each_attr() loop over all attributes + *========================================================================= + */ + + /** + * Standard attribute types to specify validation policy + */ +enum { + NLA_UNSPEC, + NLA_U8, + NLA_U16, + NLA_U32, + NLA_U64, + NLA_STRING, + NLA_FLAG, + NLA_MSECS, + NLA_NESTED, + __NLA_TYPE_MAX, +}; + +#define NLA_TYPE_MAX (__NLA_TYPE_MAX - 1) + +/** + * struct nla_policy - attribute validation policy + * @type: Type of attribute or NLA_UNSPEC + * @minlen: Minimal length of payload required to be available + * + * Policies are defined as arrays of this struct, the array must be + * accessible by attribute type up to the highest identifier to be expected. + * + * Example: + * static struct nla_policy my_policy[ATTR_MAX+1] __read_mostly = { + * [ATTR_FOO] = { .type = NLA_U16 }, + * [ATTR_BAR] = { .type = NLA_STRING }, + * [ATTR_BAZ] = { .minlen = sizeof(struct mystruct) }, + * }; + */ +struct nla_policy { + u16 type; + u16 minlen; +}; + +extern void netlink_run_queue(struct sock *sk, unsigned int *qlen, + int (*cb)(struct sk_buff *, + struct nlmsghdr *, int *)); +extern void netlink_queue_skip(struct nlmsghdr *nlh, + struct sk_buff *skb); + +extern int nla_validate(struct nlattr *head, int len, int maxtype, + struct nla_policy *policy); +extern int nla_parse(struct nlattr *tb[], int maxtype, + struct nlattr *head, int len, + struct nla_policy *policy); +extern struct nlattr * nla_find(struct nlattr *head, int len, int attrtype); +extern size_t nla_strlcpy(char *dst, const struct nlattr *nla, + size_t dstsize); +extern int nla_memcpy(void *dest, struct nlattr *src, int count); +extern int nla_memcmp(const struct nlattr *nla, const void *data, + size_t size); +extern int nla_strcmp(const struct nlattr *nla, const char *str); +extern struct nlattr * __nla_reserve(struct sk_buff *skb, int attrtype, + int attrlen); +extern struct nlattr * nla_reserve(struct sk_buff *skb, int attrtype, + int attrlen); +extern void __nla_put(struct sk_buff *skb, int attrtype, + int attrlen, const void *data); +extern int nla_put(struct sk_buff *skb, int attrtype, + int attrlen, const void *data); + +/************************************************************************** + * Netlink Messages + **************************************************************************/ + +/** + * nlmsg_msg_size - length of netlink message not including padding + * @payload: length of message payload + */ +static inline int nlmsg_msg_size(int payload) +{ + return NLMSG_HDRLEN + payload; +} + +/** + * nlmsg_total_size - length of netlink message including padding + * @payload: length of message payload + */ +static inline int nlmsg_total_size(int payload) +{ + return NLMSG_ALIGN(nlmsg_msg_size(payload)); +} + +/** + * nlmsg_padlen - length of padding at the message's tail + * @payload: length of message payload + */ +static inline int nlmsg_padlen(int payload) +{ + return nlmsg_total_size(payload) - nlmsg_msg_size(payload); +} + +/** + * nlmsg_data - head of message payload + * @nlh: netlink messsage header + */ +static inline void *nlmsg_data(const struct nlmsghdr *nlh) +{ + return (unsigned char *) nlh + NLMSG_HDRLEN; +} + +/** + * nlmsg_len - length of message payload + * @nlh: netlink message header + */ +static inline int nlmsg_len(const struct nlmsghdr *nlh) +{ + return nlh->nlmsg_len - NLMSG_HDRLEN; +} + +/** + * nlmsg_attrdata - head of attributes data + * @nlh: netlink message header + * @hdrlen: length of family specific header + */ +static inline struct nlattr *nlmsg_attrdata(const struct nlmsghdr *nlh, + int hdrlen) +{ + unsigned char *data = nlmsg_data(nlh); + return (struct nlattr *) (data + NLMSG_ALIGN(hdrlen)); +} + +/** + * nlmsg_attrlen - length of attributes data + * @nlh: netlink message header + * @hdrlen: length of family specific header + */ +static inline int nlmsg_attrlen(const struct nlmsghdr *nlh, int hdrlen) +{ + return nlmsg_len(nlh) - NLMSG_ALIGN(hdrlen); +} + +/** + * nlmsg_ok - check if the netlink message fits into the remaining bytes + * @nlh: netlink message header + * @remaining: number of bytes remaining in message stream + */ +static inline int nlmsg_ok(const struct nlmsghdr *nlh, int remaining) +{ + return (remaining >= sizeof(struct nlmsghdr) && + nlh->nlmsg_len >= sizeof(struct nlmsghdr) && + nlh->nlmsg_len <= remaining); +} + +/** + * nlmsg_next - next netlink message in message stream + * @nlh: netlink message header + * @remaining: number of bytes remaining in message stream + * + * Returns the next netlink message in the message stream and + * decrements remaining by the size of the current message. + */ +static inline struct nlmsghdr *nlmsg_next(struct nlmsghdr *nlh, int *remaining) +{ + int totlen = NLMSG_ALIGN(nlh->nlmsg_len); + + *remaining -= totlen; + + return (struct nlmsghdr *) ((unsigned char *) nlh + totlen); +} + +/** + * nlmsg_parse - parse attributes of a netlink message + * @nlh: netlink message header + * @hdrlen: length of family specific header + * @tb: destination array with maxtype+1 elements + * @maxtype: maximum attribute type to be expected + * @policy: validation policy + * + * See nla_parse() + */ +static inline int nlmsg_parse(struct nlmsghdr *nlh, int hdrlen, + struct nlattr *tb[], int maxtype, + struct nla_policy *policy) +{ + if (nlh->nlmsg_len < nlmsg_msg_size(hdrlen)) + return -EINVAL; + + return nla_parse(tb, maxtype, nlmsg_attrdata(nlh, hdrlen), + nlmsg_attrlen(nlh, hdrlen), policy); +} + +/** + * nlmsg_find_attr - find a specific attribute in a netlink message + * @nlh: netlink message header + * @hdrlen: length of familiy specific header + * @attrtype: type of attribute to look for + * + * Returns the first attribute which matches the specified type. + */ +static inline struct nlattr *nlmsg_find_attr(struct nlmsghdr *nlh, + int hdrlen, int attrtype) +{ + return nla_find(nlmsg_attrdata(nlh, hdrlen), + nlmsg_attrlen(nlh, hdrlen), attrtype); +} + +/** + * nlmsg_validate - validate a netlink message including attributes + * @nlh: netlinket message header + * @hdrlen: length of familiy specific header + * @maxtype: maximum attribute type to be expected + * @policy: validation policy + */ +static inline int nlmsg_validate(struct nlmsghdr *nlh, int hdrlen, int maxtype, + struct nla_policy *policy) +{ + if (nlh->nlmsg_len < nlmsg_msg_size(hdrlen)) + return -EINVAL; + + return nla_validate(nlmsg_attrdata(nlh, hdrlen), + nlmsg_attrlen(nlh, hdrlen), maxtype, policy); +} + +/** + * nlmsg_for_each_attr - iterate over a stream of attributes + * @pos: loop counter, set to current attribute + * @nlh: netlink message header + * @hdrlen: length of familiy specific header + * @rem: initialized to len, holds bytes currently remaining in stream + */ +#define nlmsg_for_each_attr(pos, nlh, hdrlen, rem) \ + nla_for_each_attr(pos, nlmsg_attrdata(nlh, hdrlen), \ + nlmsg_attrlen(nlh, hdrlen), rem) + +#if 0 +/* FIXME: Enable once all users have been converted */ + +/** + * __nlmsg_put - Add a new netlink message to an skb + * @skb: socket buffer to store message in + * @pid: netlink process id + * @seq: sequence number of message + * @type: message type + * @payload: length of message payload + * @flags: message flags + * + * The caller is responsible to ensure that the skb provides enough + * tailroom for both the netlink header and payload. + */ +static inline struct nlmsghdr *__nlmsg_put(struct sk_buff *skb, u32 pid, + u32 seq, int type, int payload, + int flags) +{ + struct nlmsghdr *nlh; + + nlh = (struct nlmsghdr *) skb_put(skb, nlmsg_total_size(payload)); + nlh->nlmsg_type = type; + nlh->nlmsg_len = nlmsg_msg_size(payload); + nlh->nlmsg_flags = flags; + nlh->nlmsg_pid = pid; + nlh->nlmsg_seq = seq; + + memset((unsigned char *) nlmsg_data(nlh) + payload, 0, + nlmsg_padlen(payload)); + + return nlh; +} +#endif + +/** + * nlmsg_put - Add a new netlink message to an skb + * @skb: socket buffer to store message in + * @pid: netlink process id + * @seq: sequence number of message + * @type: message type + * @payload: length of message payload + * @flags: message flags + * + * Returns NULL if the tailroom of the skb is insufficient to store + * the message header and payload. + */ +static inline struct nlmsghdr *nlmsg_put(struct sk_buff *skb, u32 pid, u32 seq, + int type, int payload, int flags) +{ + if (unlikely(skb_tailroom(skb) < nlmsg_total_size(payload))) + return NULL; + + return __nlmsg_put(skb, pid, seq, type, payload, flags); +} + +/** + * nlmsg_put_answer - Add a new callback based netlink message to an skb + * @skb: socket buffer to store message in + * @cb: netlink callback + * @type: message type + * @payload: length of message payload + * @flags: message flags + * + * Returns NULL if the tailroom of the skb is insufficient to store + * the message header and payload. + */ +static inline struct nlmsghdr *nlmsg_put_answer(struct sk_buff *skb, + struct netlink_callback *cb, + int type, int payload, + int flags) +{ + return nlmsg_put(skb, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq, + type, payload, flags); +} + +/** + * nlmsg_new - Allocate a new netlink message + * @size: maximum size of message + * + * Use NLMSG_GOODSIZE if size isn't know and you need a good default size. + */ +static inline struct sk_buff *nlmsg_new(int size) +{ + return alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); +} + +/** + * nlmsg_end - Finalize a netlink message + * @skb: socket buffer the message is stored in + * @nlh: netlink message header + * + * Corrects the netlink message header to include the appeneded + * attributes. Only necessary if attributes have been added to + * the message. + * + * Returns the total data length of the skb. + */ +static inline int nlmsg_end(struct sk_buff *skb, struct nlmsghdr *nlh) +{ + nlh->nlmsg_len = skb->tail - (unsigned char *) nlh; + + return skb->len; +} + +/** + * nlmsg_cancel - Cancel construction of a netlink message + * @skb: socket buffer the message is stored in + * @nlh: netlink message header + * + * Removes the complete netlink message including all + * attributes from the socket buffer again. Returns -1. + */ +static inline int nlmsg_cancel(struct sk_buff *skb, struct nlmsghdr *nlh) +{ + skb_trim(skb, (unsigned char *) nlh - skb->data); + + return -1; +} + +/** + * nlmsg_free - free a netlink message + * @skb: socket buffer of netlink message + */ +static inline void nlmsg_free(struct sk_buff *skb) +{ + kfree_skb(skb); +} + +/** + * nlmsg_multicast - multicast a netlink message + * @sk: netlink socket to spread messages to + * @skb: netlink message as socket buffer + * @pid: own netlink pid to avoid sending to yourself + * @group: multicast group id + */ +static inline int nlmsg_multicast(struct sock *sk, struct sk_buff *skb, + u32 pid, unsigned int group) +{ + int err; + + NETLINK_CB(skb).dst_group = group; + + err = netlink_broadcast(sk, skb, pid, group, GFP_KERNEL); + if (err > 0) + err = 0; + + return err; +} + +/** + * nlmsg_unicast - unicast a netlink message + * @sk: netlink socket to spread message to + * @skb: netlink message as socket buffer + * @pid: netlink pid of the destination socket + */ +static inline int nlmsg_unicast(struct sock *sk, struct sk_buff *skb, u32 pid) +{ + int err; + + err = netlink_unicast(sk, skb, pid, MSG_DONTWAIT); + if (err > 0) + err = 0; + + return err; +} + +/** + * nlmsg_for_each_msg - iterate over a stream of messages + * @pos: loop counter, set to current message + * @head: head of message stream + * @len: length of message stream + * @rem: initialized to len, holds bytes currently remaining in stream + */ +#define nlmsg_for_each_msg(pos, head, len, rem) \ + for (pos = head, rem = len; \ + nlmsg_ok(pos, rem); \ + pos = nlmsg_next(pos, &(rem))) + +/************************************************************************** + * Netlink Attributes + **************************************************************************/ + +/** + * nla_attr_size - length of attribute not including padding + * @payload: length of payload + */ +static inline int nla_attr_size(int payload) +{ + return NLA_HDRLEN + payload; +} + +/** + * nla_total_size - total length of attribute including padding + * @payload: length of payload + */ +static inline int nla_total_size(int payload) +{ + return NLA_ALIGN(nla_attr_size(payload)); +} + +/** + * nla_padlen - length of padding at the tail of attribute + * @payload: length of payload + */ +static inline int nla_padlen(int payload) +{ + return nla_total_size(payload) - nla_attr_size(payload); +} + +/** + * nla_data - head of payload + * @nla: netlink attribute + */ +static inline void *nla_data(const struct nlattr *nla) +{ + return (char *) nla + NLA_HDRLEN; +} + +/** + * nla_len - length of payload + * @nla: netlink attribute + */ +static inline int nla_len(const struct nlattr *nla) +{ + return nla->nla_len - NLA_HDRLEN; +} + +/** + * nla_ok - check if the netlink attribute fits into the remaining bytes + * @nla: netlink attribute + * @remaining: number of bytes remaining in attribute stream + */ +static inline int nla_ok(const struct nlattr *nla, int remaining) +{ + return remaining >= sizeof(*nla) && + nla->nla_len >= sizeof(*nla) && + nla->nla_len <= remaining; +} + +/** + * nla_next - next netlink attribte in attribute stream + * @nla: netlink attribute + * @remaining: number of bytes remaining in attribute stream + * + * Returns the next netlink attribute in the attribute stream and + * decrements remaining by the size of the current attribute. + */ +static inline struct nlattr *nla_next(const struct nlattr *nla, int *remaining) +{ + int totlen = NLA_ALIGN(nla->nla_len); + + *remaining -= totlen; + return (struct nlattr *) ((char *) nla + totlen); +} + +/** + * nla_parse_nested - parse nested attributes + * @tb: destination array with maxtype+1 elements + * @maxtype: maximum attribute type to be expected + * @nla: attribute containing the nested attributes + * @policy: validation policy + * + * See nla_parse() + */ +static inline int nla_parse_nested(struct nlattr *tb[], int maxtype, + struct nlattr *nla, + struct nla_policy *policy) +{ + return nla_parse(tb, maxtype, nla_data(nla), nla_len(nla), policy); +} +/** + * nla_put_u8 - Add a u16 netlink attribute to a socket buffer + * @skb: socket buffer to add attribute to + * @attrtype: attribute type + * @value: numeric value + */ +static inline int nla_put_u8(struct sk_buff *skb, int attrtype, u8 value) +{ + return nla_put(skb, attrtype, sizeof(u8), &value); +} + +/** + * nla_put_u16 - Add a u16 netlink attribute to a socket buffer + * @skb: socket buffer to add attribute to + * @attrtype: attribute type + * @value: numeric value + */ +static inline int nla_put_u16(struct sk_buff *skb, int attrtype, u16 value) +{ + return nla_put(skb, attrtype, sizeof(u16), &value); +} + +/** + * nla_put_u32 - Add a u32 netlink attribute to a socket buffer + * @skb: socket buffer to add attribute to + * @attrtype: attribute type + * @value: numeric value + */ +static inline int nla_put_u32(struct sk_buff *skb, int attrtype, u32 value) +{ + return nla_put(skb, attrtype, sizeof(u32), &value); +} + +/** + * nla_put_64 - Add a u64 netlink attribute to a socket buffer + * @skb: socket buffer to add attribute to + * @attrtype: attribute type + * @value: numeric value + */ +static inline int nla_put_u64(struct sk_buff *skb, int attrtype, u64 value) +{ + return nla_put(skb, attrtype, sizeof(u64), &value); +} + +/** + * nla_put_string - Add a string netlink attribute to a socket buffer + * @skb: socket buffer to add attribute to + * @attrtype: attribute type + * @str: NUL terminated string + */ +static inline int nla_put_string(struct sk_buff *skb, int attrtype, + const char *str) +{ + return nla_put(skb, attrtype, strlen(str) + 1, str); +} + +/** + * nla_put_flag - Add a flag netlink attribute to a socket buffer + * @skb: socket buffer to add attribute to + * @attrtype: attribute type + */ +static inline int nla_put_flag(struct sk_buff *skb, int attrtype) +{ + return nla_put(skb, attrtype, 0, NULL); +} + +/** + * nla_put_msecs - Add a msecs netlink attribute to a socket buffer + * @skb: socket buffer to add attribute to + * @attrtype: attribute type + * @jiffies: number of msecs in jiffies + */ +static inline int nla_put_msecs(struct sk_buff *skb, int attrtype, + unsigned long jiffies) +{ + u64 tmp = jiffies_to_msecs(jiffies); + return nla_put(skb, attrtype, sizeof(u64), &tmp); +} + +#define NLA_PUT(skb, attrtype, attrlen, data) \ + do { \ + if (nla_put(skb, attrtype, attrlen, data) < 0) \ + goto nla_put_failure; \ + } while(0) + +#define NLA_PUT_TYPE(skb, type, attrtype, value) \ + do { \ + type __tmp = value; \ + NLA_PUT(skb, attrtype, sizeof(type), &__tmp); \ + } while(0) + +#define NLA_PUT_U8(skb, attrtype, value) \ + NLA_PUT_TYPE(skb, u8, attrtype, value) + +#define NLA_PUT_U16(skb, attrtype, value) \ + NLA_PUT_TYPE(skb, u16, attrtype, value) + +#define NLA_PUT_U32(skb, attrtype, value) \ + NLA_PUT_TYPE(skb, u32, attrtype, value) + +#define NLA_PUT_U64(skb, attrtype, value) \ + NLA_PUT_TYPE(skb, u64, attrtype, value) + +#define NLA_PUT_STRING(skb, attrtype, value) \ + NLA_PUT(skb, attrtype, strlen(value) + 1, value) + +#define NLA_PUT_FLAG(skb, attrtype, value) \ + NLA_PUT(skb, attrtype, 0, NULL) + +#define NLA_PUT_MSECS(skb, attrtype, jiffies) \ + NLA_PUT_U64(skb, attrtype, jiffies_to_msecs(jiffies)) + +/** + * nla_get_u32 - return payload of u32 attribute + * @nla: u32 netlink attribute + */ +static inline u32 nla_get_u32(struct nlattr *nla) +{ + return *(u32 *) nla_data(nla); +} + +/** + * nla_get_u16 - return payload of u16 attribute + * @nla: u16 netlink attribute + */ +static inline u16 nla_get_u16(struct nlattr *nla) +{ + return *(u16 *) nla_data(nla); +} + +/** + * nla_get_u8 - return payload of u8 attribute + * @nla: u8 netlink attribute + */ +static inline u8 nla_get_u8(struct nlattr *nla) +{ + return *(u8 *) nla_data(nla); +} + +/** + * nla_get_u64 - return payload of u64 attribute + * @nla: u64 netlink attribute + */ +static inline u64 nla_get_u64(struct nlattr *nla) +{ + u64 tmp; + + nla_memcpy(&tmp, nla, sizeof(tmp)); + + return tmp; +} + +/** + * nla_get_flag - return payload of flag attribute + * @nla: flag netlink attribute + */ +static inline int nla_get_flag(struct nlattr *nla) +{ + return !!nla; +} + +/** + * nla_get_msecs - return payload of msecs attribute + * @nla: msecs netlink attribute + * + * Returns the number of milliseconds in jiffies. + */ +static inline unsigned long nla_get_msecs(struct nlattr *nla) +{ + u64 msecs = nla_get_u64(nla); + + return msecs_to_jiffies((unsigned long) msecs); +} + +/** + * nla_nest_start - Start a new level of nested attributes + * @skb: socket buffer to add attributes to + * @attrtype: attribute type of container + * + * Returns the container attribute + */ +static inline struct nlattr *nla_nest_start(struct sk_buff *skb, int attrtype) +{ + struct nlattr *start = (struct nlattr *) skb->tail; + + if (nla_put(skb, attrtype, 0, NULL) < 0) + return NULL; + + return start; +} + +/** + * nla_nest_end - Finalize nesting of attributes + * @skb: socket buffer the attribtues are stored in + * @start: container attribute + * + * Corrects the container attribute header to include the all + * appeneded attributes. + * + * Returns the total data length of the skb. + */ +static inline int nla_nest_end(struct sk_buff *skb, struct nlattr *start) +{ + start->nla_len = skb->tail - (unsigned char *) start; + return skb->len; +} + +/** + * nla_nest_cancel - Cancel nesting of attributes + * @skb: socket buffer the message is stored in + * @start: container attribute + * + * Removes the container attribute and including all nested + * attributes. Returns -1. + */ +static inline int nla_nest_cancel(struct sk_buff *skb, struct nlattr *start) +{ + if (start) + skb_trim(skb, (unsigned char *) start - skb->data); + + return -1; +} + +/** + * nla_for_each_attr - iterate over a stream of attributes + * @pos: loop counter, set to current attribute + * @head: head of attribute stream + * @len: length of attribute stream + * @rem: initialized to len, holds bytes currently remaining in stream + */ +#define nla_for_each_attr(pos, head, len, rem) \ + for (pos = head, rem = len; \ + nla_ok(pos, rem); \ + pos = nla_next(pos, &(rem))) + +#endif diff --git a/include/net/sock.h b/include/net/sock.h index ff13c4cc287..982b4ecd187 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1247,6 +1247,12 @@ static inline struct page *sk_stream_alloc_page(struct sock *sk) (skb != (struct sk_buff *)&(sk)->sk_write_queue); \ skb = skb->next) +/*from STCP for fast SACK Process*/ +#define sk_stream_for_retrans_queue_from(skb, sk) \ + for (; (skb != (sk)->sk_send_head) && \ + (skb != (struct sk_buff *)&(sk)->sk_write_queue); \ + skb = skb->next) + /* * Default write policy as shown to user space via poll/select/SIGIO */ diff --git a/include/net/tcp.h b/include/net/tcp.h index c24339c4e31..0f984801197 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -27,6 +27,7 @@ #include <linux/slab.h> #include <linux/cache.h> #include <linux/percpu.h> +#include <linux/skbuff.h> #include <net/inet_connection_sock.h> #include <net/inet_timewait_sock.h> @@ -88,10 +89,10 @@ extern void tcp_time_wait(struct sock *sk, int state, int timeo); */ #define TCP_SYN_RETRIES 5 /* number of times to retry active opening a - * connection: ~180sec is RFC minumum */ + * connection: ~180sec is RFC minimum */ #define TCP_SYNACK_RETRIES 5 /* number of times to retry passive opening a - * connection: ~180sec is RFC minumum */ + * connection: ~180sec is RFC minimum */ #define TCP_ORPHAN_RETRIES 7 /* number of times to retry on an orphaned @@ -179,7 +180,7 @@ extern void tcp_time_wait(struct sock *sk, int state, int timeo); /* Flags in tp->nonagle */ #define TCP_NAGLE_OFF 1 /* Nagle's algo is disabled */ #define TCP_NAGLE_CORK 2 /* Socket is corked */ -#define TCP_NAGLE_PUSH 4 /* Cork is overriden for already queued data */ +#define TCP_NAGLE_PUSH 4 /* Cork is overridden for already queued data */ extern struct inet_timewait_death_row tcp_death_row; @@ -217,6 +218,7 @@ extern int sysctl_tcp_low_latency; extern int sysctl_tcp_nometrics_save; extern int sysctl_tcp_moderate_rcvbuf; extern int sysctl_tcp_tso_win_divisor; +extern int sysctl_tcp_abc; extern atomic_t tcp_memory_allocated; extern atomic_t tcp_sockets_allocated; @@ -550,13 +552,13 @@ extern u32 __tcp_select_window(struct sock *sk); /* TCP timestamps are only 32-bits, this causes a slight * complication on 64-bit systems since we store a snapshot - * of jiffies in the buffer control blocks below. We decidely + * of jiffies in the buffer control blocks below. We decidedly * only use of the low 32-bits of jiffies and hide the ugly * casts with the following macro. */ #define tcp_time_stamp ((__u32)(jiffies)) -/* This is what the send packet queueing engine uses to pass +/* This is what the send packet queuing engine uses to pass * TCP per-packet control information to the transmission * code. We also store the host-order sequence numbers in * here too. This is 36 bytes on 32-bit architectures, @@ -596,7 +598,7 @@ struct tcp_skb_cb { #define TCPCB_EVER_RETRANS 0x80 /* Ever retransmitted frame */ #define TCPCB_RETRANS (TCPCB_SACKED_RETRANS|TCPCB_EVER_RETRANS) -#define TCPCB_URG 0x20 /* Urgent pointer advenced here */ +#define TCPCB_URG 0x20 /* Urgent pointer advanced here */ #define TCPCB_AT_TAIL (TCPCB_URG) @@ -764,6 +766,33 @@ static inline __u32 tcp_current_ssthresh(const struct sock *sk) (tp->snd_cwnd >> 2))); } +/* + * Linear increase during slow start + */ +static inline void tcp_slow_start(struct tcp_sock *tp) +{ + if (sysctl_tcp_abc) { + /* RFC3465: Slow Start + * TCP sender SHOULD increase cwnd by the number of + * previously unacknowledged bytes ACKed by each incoming + * acknowledgment, provided the increase is not more than L + */ + if (tp->bytes_acked < tp->mss_cache) + return; + + /* We MAY increase by 2 if discovered delayed ack */ + if (sysctl_tcp_abc > 1 && tp->bytes_acked > 2*tp->mss_cache) { + if (tp->snd_cwnd < tp->snd_cwnd_clamp) + tp->snd_cwnd++; + } + } + tp->bytes_acked = 0; + + if (tp->snd_cwnd < tp->snd_cwnd_clamp) + tp->snd_cwnd++; +} + + static inline void tcp_sync_left_out(struct tcp_sock *tp) { if (tp->rx_opt.sack_ok && @@ -793,6 +822,7 @@ static inline void tcp_enter_cwr(struct sock *sk) struct tcp_sock *tp = tcp_sk(sk); tp->prior_ssthresh = 0; + tp->bytes_acked = 0; if (inet_csk(sk)->icsk_ca_state < TCP_CA_CWR) { __tcp_enter_cwr(sk); tcp_set_ca_state(sk, TCP_CA_CWR); @@ -809,6 +839,27 @@ static __inline__ __u32 tcp_max_burst(const struct tcp_sock *tp) return 3; } +/* RFC2861 Check whether we are limited by application or congestion window + * This is the inverse of cwnd check in tcp_tso_should_defer + */ +static inline int tcp_is_cwnd_limited(const struct sock *sk, u32 in_flight) +{ + const struct tcp_sock *tp = tcp_sk(sk); + u32 left; + + if (in_flight >= tp->snd_cwnd) + return 1; + + if (!(sk->sk_route_caps & NETIF_F_TSO)) + return 0; + + left = tp->snd_cwnd - in_flight; + if (sysctl_tcp_tso_win_divisor) + return left * sysctl_tcp_tso_win_divisor < tp->snd_cwnd; + else + return left <= tcp_max_burst(tp); +} + static __inline__ void tcp_minshall_update(struct tcp_sock *tp, int mss, const struct sk_buff *skb) { @@ -852,7 +903,7 @@ static __inline__ u16 tcp_v4_check(struct tcphdr *th, int len, static __inline__ int __tcp_checksum_complete(struct sk_buff *skb) { - return (unsigned short)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum)); + return __skb_checksum_complete(skb); } static __inline__ int tcp_checksum_complete(struct sk_buff *skb) @@ -1156,6 +1207,15 @@ static inline void tcp_mib_init(void) TCP_ADD_STATS_USER(TCP_MIB_MAXCONN, -1); } +/*from STCP */ +static inline void clear_all_retrans_hints(struct tcp_sock *tp){ + tp->lost_skb_hint = NULL; + tp->scoreboard_skb_hint = NULL; + tp->retransmit_skb_hint = NULL; + tp->forward_skb_hint = NULL; + tp->fastpath_skb_hint = NULL; +} + /* /proc */ enum tcp_seq_states { TCP_SEQ_STATE_LISTENING, diff --git a/include/rdma/ib_user_verbs.h b/include/rdma/ib_user_verbs.h index 072f3a2edac..5ff1490c08d 100644 --- a/include/rdma/ib_user_verbs.h +++ b/include/rdma/ib_user_verbs.h @@ -43,7 +43,7 @@ * Increment this value if any changes that break userspace ABI * compatibility are made. */ -#define IB_USER_VERBS_ABI_VERSION 3 +#define IB_USER_VERBS_ABI_VERSION 4 enum { IB_USER_VERBS_CMD_GET_CONTEXT, @@ -333,6 +333,11 @@ struct ib_uverbs_create_qp { struct ib_uverbs_create_qp_resp { __u32 qp_handle; __u32 qpn; + __u32 max_send_wr; + __u32 max_recv_wr; + __u32 max_send_sge; + __u32 max_recv_sge; + __u32 max_inline_data; }; /* @@ -552,9 +557,7 @@ struct ib_uverbs_modify_srq { __u32 srq_handle; __u32 attr_mask; __u32 max_wr; - __u32 max_sge; __u32 srq_limit; - __u32 reserved; __u64 driver_data[0]; }; diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index f72d46d54e0..a7f4c355a91 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -881,7 +881,7 @@ struct ib_device { struct ib_ucontext *context, struct ib_udata *udata); int (*destroy_cq)(struct ib_cq *cq); - int (*resize_cq)(struct ib_cq *cq, int *cqe); + int (*resize_cq)(struct ib_cq *cq, int cqe); int (*poll_cq)(struct ib_cq *cq, int num_entries, struct ib_wc *wc); int (*peek_cq)(struct ib_cq *cq, int wc_cnt); diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 5b8dd98a230..b88d4186cd7 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -155,7 +155,7 @@ int ptrace_attach(struct task_struct *task) retval = -EPERM; if (task->pid <= 1) goto bad; - if (task == current) + if (task->tgid == current->tgid) goto bad; /* the same process cannot be attached many times */ if (task->ptrace & PT_PTRACED) diff --git a/kernel/sched.c b/kernel/sched.c index ac3f5cc3bb5..b6506671b2b 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -815,7 +815,8 @@ static void activate_task(task_t *p, runqueue_t *rq, int local) } #endif - p->prio = recalc_task_prio(p, now); + if (!rt_task(p)) + p->prio = recalc_task_prio(p, now); /* * This checks to make sure it's not an uninterruptible task diff --git a/kernel/signal.c b/kernel/signal.c index 1bf3c39d610..80789a59b4d 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -1499,7 +1499,7 @@ void do_notify_parent(struct task_struct *tsk, int sig) psig = tsk->parent->sighand; spin_lock_irqsave(&psig->siglock, flags); - if (sig == SIGCHLD && + if (!tsk->ptrace && sig == SIGCHLD && (psig->action[SIGCHLD-1].sa.sa_handler == SIG_IGN || (psig->action[SIGCHLD-1].sa.sa_flags & SA_NOCLDWAIT))) { /* diff --git a/kernel/sys.c b/kernel/sys.c index c43b3e22bbd..bce933ebb29 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -1497,6 +1497,8 @@ EXPORT_SYMBOL(in_egroup_p); DECLARE_RWSEM(uts_sem); +EXPORT_SYMBOL(uts_sem); + asmlinkage long sys_newuname(struct new_utsname __user * name) { int errno = 0; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index ff81b5c6551..987225bdd66 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1330,7 +1330,7 @@ void show_free_areas(void) } else printk("\n"); - for_each_cpu(cpu) { + for_each_online_cpu(cpu) { struct per_cpu_pageset *pageset; pageset = zone_pcp(zone, cpu); diff --git a/net/core/datagram.c b/net/core/datagram.c index d219435d086..1bcfef51ac5 100644 --- a/net/core/datagram.c +++ b/net/core/datagram.c @@ -350,6 +350,20 @@ fault: return -EFAULT; } +unsigned int __skb_checksum_complete(struct sk_buff *skb) +{ + unsigned int sum; + + sum = (u16)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum)); + if (likely(!sum)) { + if (unlikely(skb->ip_summed == CHECKSUM_HW)) + netdev_rx_csum_fault(skb->dev); + skb->ip_summed = CHECKSUM_UNNECESSARY; + } + return sum; +} +EXPORT_SYMBOL(__skb_checksum_complete); + /** * skb_copy_and_csum_datagram_iovec - Copy and checkum skb to user iovec. * @skb: skbuff @@ -363,7 +377,7 @@ fault: * -EFAULT - fault during copy. Beware, in this case iovec * can be modified! */ -int skb_copy_and_csum_datagram_iovec(const struct sk_buff *skb, +int skb_copy_and_csum_datagram_iovec(struct sk_buff *skb, int hlen, struct iovec *iov) { unsigned int csum; @@ -376,8 +390,7 @@ int skb_copy_and_csum_datagram_iovec(const struct sk_buff *skb, iov++; if (iov->iov_len < chunk) { - if ((unsigned short)csum_fold(skb_checksum(skb, 0, chunk + hlen, - skb->csum))) + if (__skb_checksum_complete(skb)) goto csum_error; if (skb_copy_datagram_iovec(skb, hlen, iov, chunk)) goto fault; @@ -388,6 +401,8 @@ int skb_copy_and_csum_datagram_iovec(const struct sk_buff *skb, goto fault; if ((unsigned short)csum_fold(csum)) goto csum_error; + if (unlikely(skb->ip_summed == CHECKSUM_HW)) + netdev_rx_csum_fault(skb->dev); iov->iov_len -= chunk; iov->iov_base += chunk; } diff --git a/net/core/dev.c b/net/core/dev.c index 8d154159527..0b48e294aaf 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1108,6 +1108,18 @@ out: return ret; } +/* Take action when hardware reception checksum errors are detected. */ +#ifdef CONFIG_BUG +void netdev_rx_csum_fault(struct net_device *dev) +{ + if (net_ratelimit()) { + printk(KERN_ERR "%s: hw csum failure.\n", dev->name); + dump_stack(); + } +} +EXPORT_SYMBOL(netdev_rx_csum_fault); +#endif + #ifdef CONFIG_HIGHMEM /* Actually, we should eliminate this check as soon as we know, that: * 1. IOMMU is present and allows to map all the memory. diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 802fe11efad..49424a42a2c 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -101,16 +101,20 @@ void netpoll_queue(struct sk_buff *skb) static int checksum_udp(struct sk_buff *skb, struct udphdr *uh, unsigned short ulen, u32 saddr, u32 daddr) { - if (uh->check == 0) + unsigned int psum; + + if (uh->check == 0 || skb->ip_summed == CHECKSUM_UNNECESSARY) return 0; - if (skb->ip_summed == CHECKSUM_HW) - return csum_tcpudp_magic( - saddr, daddr, ulen, IPPROTO_UDP, skb->csum); + psum = csum_tcpudp_nofold(saddr, daddr, ulen, IPPROTO_UDP, 0); + + if (skb->ip_summed == CHECKSUM_HW && + !(u16)csum_fold(csum_add(psum, skb->csum))) + return 0; - skb->csum = csum_tcpudp_nofold(saddr, daddr, ulen, IPPROTO_UDP, 0); + skb->csum = psum; - return csum_fold(skb_checksum(skb, 0, skb->len, skb->csum)); + return __skb_checksum_complete(skb); } /* @@ -489,7 +493,7 @@ int __netpoll_rx(struct sk_buff *skb) if (ulen != len) goto out; - if (checksum_udp(skb, uh, ulen, iph->saddr, iph->daddr) < 0) + if (checksum_udp(skb, uh, ulen, iph->saddr, iph->daddr)) goto out; if (np->local_ip && np->local_ip != ntohl(iph->daddr)) goto out; diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 9bed7569ce3..8700379685e 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -49,6 +49,7 @@ #include <net/udp.h> #include <net/sock.h> #include <net/pkt_sched.h> +#include <net/netlink.h> DECLARE_MUTEX(rtnl_sem); @@ -462,11 +463,6 @@ void rtmsg_ifinfo(int type, struct net_device *dev, unsigned change) netlink_broadcast(rtnl, skb, 0, RTNLGRP_LINK, GFP_KERNEL); } -static int rtnetlink_done(struct netlink_callback *cb) -{ - return 0; -} - /* Protected by RTNL sempahore. */ static struct rtattr **rta_buf; static int rtattr_max; @@ -524,8 +520,6 @@ rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, int *errp) } if (kind == 2 && nlh->nlmsg_flags&NLM_F_DUMP) { - u32 rlen; - if (link->dumpit == NULL) link = &(rtnetlink_links[PF_UNSPEC][type]); @@ -533,14 +527,11 @@ rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, int *errp) goto err_inval; if ((*errp = netlink_dump_start(rtnl, skb, nlh, - link->dumpit, - rtnetlink_done)) != 0) { + link->dumpit, NULL)) != 0) { return -1; } - rlen = NLMSG_ALIGN(nlh->nlmsg_len); - if (rlen > skb->len) - rlen = skb->len; - skb_pull(skb, rlen); + + netlink_queue_skip(nlh, skb); return -1; } @@ -579,75 +570,13 @@ err_inval: return -1; } -/* - * Process one packet of messages. - * Malformed skbs with wrong lengths of messages are discarded silently. - */ - -static inline int rtnetlink_rcv_skb(struct sk_buff *skb) -{ - int err; - struct nlmsghdr * nlh; - - while (skb->len >= NLMSG_SPACE(0)) { - u32 rlen; - - nlh = (struct nlmsghdr *)skb->data; - if (nlh->nlmsg_len < sizeof(*nlh) || skb->len < nlh->nlmsg_len) - return 0; - rlen = NLMSG_ALIGN(nlh->nlmsg_len); - if (rlen > skb->len) - rlen = skb->len; - if (rtnetlink_rcv_msg(skb, nlh, &err)) { - /* Not error, but we must interrupt processing here: - * Note, that in this case we do not pull message - * from skb, it will be processed later. - */ - if (err == 0) - return -1; - netlink_ack(skb, nlh, err); - } else if (nlh->nlmsg_flags&NLM_F_ACK) - netlink_ack(skb, nlh, 0); - skb_pull(skb, rlen); - } - - return 0; -} - -/* - * rtnetlink input queue processing routine: - * - process as much as there was in the queue upon entry. - * - feed skbs to rtnetlink_rcv_skb, until it refuse a message, - * that will occur, when a dump started. - */ - static void rtnetlink_rcv(struct sock *sk, int len) { - unsigned int qlen = skb_queue_len(&sk->sk_receive_queue); + unsigned int qlen = 0; do { - struct sk_buff *skb; - rtnl_lock(); - - if (qlen > skb_queue_len(&sk->sk_receive_queue)) - qlen = skb_queue_len(&sk->sk_receive_queue); - - for (; qlen; qlen--) { - skb = skb_dequeue(&sk->sk_receive_queue); - if (rtnetlink_rcv_skb(skb)) { - if (skb->len) - skb_queue_head(&sk->sk_receive_queue, - skb); - else { - kfree_skb(skb); - qlen--; - } - break; - } - kfree_skb(skb); - } - + netlink_run_queue(sk, &qlen, &rtnetlink_rcv_msg); up(&rtnl_sem); netdev_run_todo(); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 95501e40100..b7d13a4fff4 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -336,6 +336,9 @@ void __kfree_skb(struct sk_buff *skb) } #ifdef CONFIG_NETFILTER nf_conntrack_put(skb->nfct); +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) + nf_conntrack_put_reasm(skb->nfct_reasm); +#endif #ifdef CONFIG_BRIDGE_NETFILTER nf_bridge_put(skb->nf_bridge); #endif @@ -414,9 +417,17 @@ struct sk_buff *skb_clone(struct sk_buff *skb, gfp_t gfp_mask) C(nfct); nf_conntrack_get(skb->nfct); C(nfctinfo); +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) + C(nfct_reasm); + nf_conntrack_get_reasm(skb->nfct_reasm); +#endif #if defined(CONFIG_IP_VS) || defined(CONFIG_IP_VS_MODULE) C(ipvs_property); #endif +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) + C(nfct_reasm); + nf_conntrack_get_reasm(skb->nfct_reasm); +#endif #ifdef CONFIG_BRIDGE_NETFILTER C(nf_bridge); nf_bridge_get(skb->nf_bridge); @@ -474,6 +485,10 @@ static void copy_skb_header(struct sk_buff *new, const struct sk_buff *old) new->nfct = old->nfct; nf_conntrack_get(old->nfct); new->nfctinfo = old->nfctinfo; +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) + new->nfct_reasm = old->nfct_reasm; + nf_conntrack_get_reasm(old->nfct_reasm); +#endif #if defined(CONFIG_IP_VS) || defined(CONFIG_IP_VS_MODULE) new->ipvs_property = old->ipvs_property; #endif diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index 175e093ec56..e3eceecd049 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -934,11 +934,11 @@ int icmp_rcv(struct sk_buff *skb) case CHECKSUM_HW: if (!(u16)csum_fold(skb->csum)) break; - LIMIT_NETDEBUG(KERN_DEBUG "icmp v4 hw csum failure\n"); + /* fall through */ case CHECKSUM_NONE: - if ((u16)csum_fold(skb_checksum(skb, 0, skb->len, 0))) + skb->csum = 0; + if (__skb_checksum_complete(skb)) goto error; - default:; } if (!pskb_pull(skb, sizeof(struct icmphdr))) diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index c6247fc8406..c04607b4921 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -872,11 +872,18 @@ int igmp_rcv(struct sk_buff *skb) return 0; } - if (!pskb_may_pull(skb, sizeof(struct igmphdr)) || - (u16)csum_fold(skb_checksum(skb, 0, len, 0))) { - in_dev_put(in_dev); - kfree_skb(skb); - return 0; + if (!pskb_may_pull(skb, sizeof(struct igmphdr))) + goto drop; + + switch (skb->ip_summed) { + case CHECKSUM_HW: + if (!(u16)csum_fold(skb->csum)) + break; + /* fall through */ + case CHECKSUM_NONE: + skb->csum = 0; + if (__skb_checksum_complete(skb)) + goto drop; } ih = skb->h.igmph; @@ -906,6 +913,8 @@ int igmp_rcv(struct sk_buff *skb) default: NETDEBUG(KERN_DEBUG "New IGMP type=%d, why we do not know about it?\n", ih->type); } + +drop: in_dev_put(in_dev); kfree_skb(skb); return 0; diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c index 71f3c7350c6..39061ed53cf 100644 --- a/net/ipv4/inet_diag.c +++ b/net/ipv4/inet_diag.c @@ -724,12 +724,6 @@ done: return skb->len; } -static int inet_diag_dump_done(struct netlink_callback *cb) -{ - return 0; -} - - static __inline__ int inet_diag_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) { @@ -760,8 +754,7 @@ inet_diag_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) goto err_inval; } return netlink_dump_start(idiagnl, skb, nlh, - inet_diag_dump, - inet_diag_dump_done); + inet_diag_dump, NULL); } else { return inet_diag_get_exact(skb, nlh); } diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index 896ce3f8f53..4e9c74b54b1 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -577,15 +577,16 @@ static int ipgre_rcv(struct sk_buff *skb) goto drop_nolock; if (flags&GRE_CSUM) { - if (skb->ip_summed == CHECKSUM_HW) { + switch (skb->ip_summed) { + case CHECKSUM_HW: csum = (u16)csum_fold(skb->csum); - if (csum) - skb->ip_summed = CHECKSUM_NONE; - } - if (skb->ip_summed == CHECKSUM_NONE) { - skb->csum = skb_checksum(skb, 0, skb->len, 0); + if (!csum) + break; + /* fall through */ + case CHECKSUM_NONE: + skb->csum = 0; + csum = __skb_checksum_complete(skb); skb->ip_summed = CHECKSUM_HW; - csum = (u16)csum_fold(skb->csum); } offset += 4; } diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index 7d917e4ce1d..9d3c8b5f327 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -5,6 +5,20 @@ menu "IP: Netfilter Configuration" depends on INET && NETFILTER +config NF_CONNTRACK_IPV4 + tristate "IPv4 support for new connection tracking (EXPERIMENTAL)" + depends on EXPERIMENTAL && NF_CONNTRACK + ---help--- + Connection tracking keeps a record of what packets have passed + through your machine, in order to figure out how they are related + into connections. + + This is IPv4 support on Layer 3 independent connection tracking. + Layer 3 independent connection tracking is experimental scheme + which generalize ip_conntrack to support other layer 3 protocols. + + To compile it as a module, choose M here. If unsure, say N. + # connection tracking, helpers and protocols config IP_NF_CONNTRACK tristate "Connection tracking (required for masq/NAT)" @@ -209,8 +223,8 @@ config IP_NF_MATCH_PKTTYPE tristate "Packet type match support" depends on IP_NF_IPTABLES help - Packet type matching allows you to match a packet by - its "class", eg. BROADCAST, MULTICAST, ... + Packet type matching allows you to match a packet by + its "class", eg. BROADCAST, MULTICAST, ... Typical usage: iptables -A INPUT -m pkttype --pkt-type broadcast -j LOG @@ -317,7 +331,8 @@ config IP_NF_MATCH_TCPMSS config IP_NF_MATCH_HELPER tristate "Helper match support" - depends on IP_NF_CONNTRACK && IP_NF_IPTABLES + depends on IP_NF_IPTABLES + depends on IP_NF_CONNTRACK || NF_CONNTRACK_IPV4 help Helper matching allows you to match packets in dynamic connections tracked by a conntrack-helper, ie. ip_conntrack_ftp @@ -326,7 +341,8 @@ config IP_NF_MATCH_HELPER config IP_NF_MATCH_STATE tristate "Connection state match support" - depends on IP_NF_CONNTRACK && IP_NF_IPTABLES + depends on IP_NF_IPTABLES + depends on IP_NF_CONNTRACK || NF_CONNTRACK_IPV4 help Connection state matching allows you to match packets based on their relationship to a tracked connection (ie. previous packets). This @@ -336,7 +352,8 @@ config IP_NF_MATCH_STATE config IP_NF_MATCH_CONNTRACK tristate "Connection tracking match support" - depends on IP_NF_CONNTRACK && IP_NF_IPTABLES + depends on IP_NF_IPTABLES + depends on IP_NF_CONNTRACK || NF_CONNTRACK_IPV4 help This is a general conntrack match module, a superset of the state match. @@ -422,7 +439,8 @@ config IP_NF_MATCH_COMMENT config IP_NF_MATCH_CONNMARK tristate 'Connection mark match support' - depends on IP_NF_CONNTRACK_MARK && IP_NF_IPTABLES + depends on IP_NF_IPTABLES + depends on IP_NF_CONNTRACK_MARK || (NF_CONNTRACK_MARK && NF_CONNTRACK_IPV4) help This option adds a `connmark' match, which allows you to match the connection mark value previously set for the session by `CONNMARK'. @@ -433,7 +451,8 @@ config IP_NF_MATCH_CONNMARK config IP_NF_MATCH_CONNBYTES tristate 'Connection byte/packet counter match support' - depends on IP_NF_CT_ACCT && IP_NF_IPTABLES + depends on IP_NF_IPTABLES + depends on IP_NF_CT_ACCT || (NF_CT_ACCT && NF_CONNTRACK_IPV4) help This option adds a `connbytes' match, which allows you to match the number of bytes and/or packets for each direction within a connection. @@ -747,7 +766,8 @@ config IP_NF_TARGET_TTL config IP_NF_TARGET_CONNMARK tristate 'CONNMARK target support' - depends on IP_NF_CONNTRACK_MARK && IP_NF_MANGLE + depends on IP_NF_MANGLE + depends on IP_NF_CONNTRACK_MARK || (NF_CONNTRACK_MARK && NF_CONNTRACK_IPV4) help This option adds a `CONNMARK' target, which allows one to manipulate the connection mark value. Similar to the MARK target, but @@ -759,7 +779,8 @@ config IP_NF_TARGET_CONNMARK config IP_NF_TARGET_CLUSTERIP tristate "CLUSTERIP target support (EXPERIMENTAL)" - depends on IP_NF_CONNTRACK_MARK && IP_NF_IPTABLES && EXPERIMENTAL + depends on IP_NF_IPTABLES && EXPERIMENTAL + depends on IP_NF_CONNTRACK_MARK || (NF_CONNTRACK_MARK && NF_CONNTRACK_IPV4) help The CLUSTERIP target allows you to build load-balancing clusters of network servers without having a dedicated load-balancing @@ -782,7 +803,7 @@ config IP_NF_RAW config IP_NF_TARGET_NOTRACK tristate 'NOTRACK target support' depends on IP_NF_RAW - depends on IP_NF_CONNTRACK + depends on IP_NF_CONNTRACK || NF_CONNTRACK_IPV4 help The NOTRACK target allows a select rule to specify which packets *not* to enter the conntrack/NAT diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index dab4b58dd31..058c48e258f 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -103,3 +103,9 @@ obj-$(CONFIG_IP_NF_ARP_MANGLE) += arpt_mangle.o obj-$(CONFIG_IP_NF_ARPFILTER) += arptable_filter.o obj-$(CONFIG_IP_NF_QUEUE) += ip_queue.o + +# objects for l3 independent conntrack +nf_conntrack_ipv4-objs := nf_conntrack_l3proto_ipv4.o nf_conntrack_proto_icmp.o + +# l3 independent conntrack +obj-$(CONFIG_NF_CONNTRACK_IPV4) += nf_conntrack_ipv4.o diff --git a/net/ipv4/netfilter/ip_conntrack_netlink.c b/net/ipv4/netfilter/ip_conntrack_netlink.c index 82a65043a8e..d2a4fec2286 100644 --- a/net/ipv4/netfilter/ip_conntrack_netlink.c +++ b/net/ipv4/netfilter/ip_conntrack_netlink.c @@ -28,11 +28,8 @@ #include <linux/netlink.h> #include <linux/spinlock.h> #include <linux/notifier.h> -#include <linux/rtnetlink.h> #include <linux/netfilter.h> -#include <linux/netfilter_ipv4.h> -#include <linux/netfilter_ipv4/ip_tables.h> #include <linux/netfilter_ipv4/ip_conntrack.h> #include <linux/netfilter_ipv4/ip_conntrack_core.h> #include <linux/netfilter_ipv4/ip_conntrack_helper.h> @@ -58,14 +55,17 @@ ctnetlink_dump_tuples_proto(struct sk_buff *skb, const struct ip_conntrack_tuple *tuple) { struct ip_conntrack_protocol *proto; + int ret = 0; NFA_PUT(skb, CTA_PROTO_NUM, sizeof(u_int8_t), &tuple->dst.protonum); proto = ip_conntrack_proto_find_get(tuple->dst.protonum); - if (proto && proto->tuple_to_nfattr) - return proto->tuple_to_nfattr(skb, tuple); + if (likely(proto && proto->tuple_to_nfattr)) { + ret = proto->tuple_to_nfattr(skb, tuple); + ip_conntrack_proto_put(proto); + } - return 0; + return ret; nfattr_failure: return -1; @@ -175,7 +175,7 @@ ctnetlink_dump_counters(struct sk_buff *skb, const struct ip_conntrack *ct, { enum ctattr_type type = dir ? CTA_COUNTERS_REPLY: CTA_COUNTERS_ORIG; struct nfattr *nest_count = NFA_NEST(skb, type); - u_int64_t tmp; + u_int32_t tmp; tmp = htonl(ct->counters[dir].packets); NFA_PUT(skb, CTA_COUNTERS32_PACKETS, sizeof(u_int32_t), &tmp); @@ -479,9 +479,7 @@ ctnetlink_parse_tuple_ip(struct nfattr *attr, struct ip_conntrack_tuple *tuple) DEBUGP("entered %s\n", __FUNCTION__); - - if (nfattr_parse_nested(tb, CTA_IP_MAX, attr) < 0) - goto nfattr_failure; + nfattr_parse_nested(tb, CTA_IP_MAX, attr); if (nfattr_bad_size(tb, CTA_IP_MAX, cta_min_ip)) return -EINVAL; @@ -497,9 +495,6 @@ ctnetlink_parse_tuple_ip(struct nfattr *attr, struct ip_conntrack_tuple *tuple) DEBUGP("leaving\n"); return 0; - -nfattr_failure: - return -1; } static const int cta_min_proto[CTA_PROTO_MAX] = { @@ -521,8 +516,7 @@ ctnetlink_parse_tuple_proto(struct nfattr *attr, DEBUGP("entered %s\n", __FUNCTION__); - if (nfattr_parse_nested(tb, CTA_PROTO_MAX, attr) < 0) - goto nfattr_failure; + nfattr_parse_nested(tb, CTA_PROTO_MAX, attr); if (nfattr_bad_size(tb, CTA_PROTO_MAX, cta_min_proto)) return -EINVAL; @@ -539,9 +533,6 @@ ctnetlink_parse_tuple_proto(struct nfattr *attr, } return ret; - -nfattr_failure: - return -1; } static inline int @@ -555,8 +546,7 @@ ctnetlink_parse_tuple(struct nfattr *cda[], struct ip_conntrack_tuple *tuple, memset(tuple, 0, sizeof(*tuple)); - if (nfattr_parse_nested(tb, CTA_TUPLE_MAX, cda[type-1]) < 0) - goto nfattr_failure; + nfattr_parse_nested(tb, CTA_TUPLE_MAX, cda[type-1]); if (!tb[CTA_TUPLE_IP-1]) return -EINVAL; @@ -583,9 +573,6 @@ ctnetlink_parse_tuple(struct nfattr *cda[], struct ip_conntrack_tuple *tuple, DEBUGP("leaving\n"); return 0; - -nfattr_failure: - return -1; } #ifdef CONFIG_IP_NF_NAT_NEEDED @@ -603,11 +590,10 @@ static int ctnetlink_parse_nat_proto(struct nfattr *attr, DEBUGP("entered %s\n", __FUNCTION__); - if (nfattr_parse_nested(tb, CTA_PROTONAT_MAX, attr) < 0) - goto nfattr_failure; + nfattr_parse_nested(tb, CTA_PROTONAT_MAX, attr); if (nfattr_bad_size(tb, CTA_PROTONAT_MAX, cta_min_protonat)) - goto nfattr_failure; + return -EINVAL; npt = ip_nat_proto_find_get(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum); if (!npt) @@ -626,9 +612,6 @@ static int ctnetlink_parse_nat_proto(struct nfattr *attr, DEBUGP("leaving\n"); return 0; - -nfattr_failure: - return -1; } static inline int @@ -642,8 +625,7 @@ ctnetlink_parse_nat(struct nfattr *cda[], memset(range, 0, sizeof(*range)); - if (nfattr_parse_nested(tb, CTA_NAT_MAX, cda[CTA_NAT-1]) < 0) - goto nfattr_failure; + nfattr_parse_nested(tb, CTA_NAT_MAX, cda[CTA_NAT-1]); if (tb[CTA_NAT_MINIP-1]) range->min_ip = *(u_int32_t *)NFA_DATA(tb[CTA_NAT_MINIP-1]); @@ -665,9 +647,6 @@ ctnetlink_parse_nat(struct nfattr *cda[], DEBUGP("leaving\n"); return 0; - -nfattr_failure: - return -1; } #endif @@ -678,8 +657,7 @@ ctnetlink_parse_help(struct nfattr *attr, char **helper_name) DEBUGP("entered %s\n", __FUNCTION__); - if (nfattr_parse_nested(tb, CTA_HELP_MAX, attr) < 0) - goto nfattr_failure; + nfattr_parse_nested(tb, CTA_HELP_MAX, attr); if (!tb[CTA_HELP_NAME-1]) return -EINVAL; @@ -687,9 +665,6 @@ ctnetlink_parse_help(struct nfattr *attr, char **helper_name) *helper_name = NFA_DATA(tb[CTA_HELP_NAME-1]); return 0; - -nfattr_failure: - return -1; } static int @@ -804,7 +779,7 @@ ctnetlink_get_conntrack(struct sock *ctnl, struct sk_buff *skb, ct = tuplehash_to_ctrack(h); err = -ENOMEM; - skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_ATOMIC); + skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (!skb2) { ip_conntrack_put(ct); return -ENOMEM; @@ -827,7 +802,7 @@ ctnetlink_get_conntrack(struct sock *ctnl, struct sk_buff *skb, free: kfree_skb(skb2); out: - return -1; + return err; } static inline int @@ -957,8 +932,7 @@ ctnetlink_change_protoinfo(struct ip_conntrack *ct, struct nfattr *cda[]) u_int16_t npt = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum; int err = 0; - if (nfattr_parse_nested(tb, CTA_PROTOINFO_MAX, attr) < 0) - goto nfattr_failure; + nfattr_parse_nested(tb, CTA_PROTOINFO_MAX, attr); proto = ip_conntrack_proto_find_get(npt); if (!proto) @@ -969,9 +943,6 @@ ctnetlink_change_protoinfo(struct ip_conntrack *ct, struct nfattr *cda[]) ip_conntrack_proto_put(proto); return err; - -nfattr_failure: - return -ENOMEM; } static int @@ -1005,6 +976,11 @@ ctnetlink_change_conntrack(struct ip_conntrack *ct, struct nfattr *cda[]) return err; } +#if defined(CONFIG_IP_NF_CONNTRACK_MARK) + if (cda[CTA_MARK-1]) + ct->mark = ntohl(*(u_int32_t *)NFA_DATA(cda[CTA_MARK-1])); +#endif + DEBUGP("all done\n"); return 0; } @@ -1048,6 +1024,11 @@ ctnetlink_create_conntrack(struct nfattr *cda[], if (ct->helper) ip_conntrack_helper_put(ct->helper); +#if defined(CONFIG_IP_NF_CONNTRACK_MARK) + if (cda[CTA_MARK-1]) + ct->mark = ntohl(*(u_int32_t *)NFA_DATA(cda[CTA_MARK-1])); +#endif + DEBUGP("conntrack with id %u inserted\n", ct->id); return 0; @@ -1312,6 +1293,14 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, if (!exp) return -ENOENT; + if (cda[CTA_EXPECT_ID-1]) { + u_int32_t id = *(u_int32_t *)NFA_DATA(cda[CTA_EXPECT_ID-1]); + if (exp->id != ntohl(id)) { + ip_conntrack_expect_put(exp); + return -ENOENT; + } + } + err = -ENOMEM; skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (!skb2) @@ -1387,7 +1376,7 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, ip_conntrack_expect_put(exp); } } - write_unlock(&ip_conntrack_lock); + write_unlock_bh(&ip_conntrack_lock); } else { /* This basically means we have to flush everything*/ write_lock_bh(&ip_conntrack_lock); @@ -1554,6 +1543,8 @@ static struct nfnetlink_subsystem ctnl_exp_subsys = { .cb = ctnl_exp_cb, }; +MODULE_ALIAS_NFNL_SUBSYS(NFNL_SUBSYS_CTNETLINK); + static int __init ctnetlink_init(void) { int ret; diff --git a/net/ipv4/netfilter/ip_conntrack_proto_icmp.c b/net/ipv4/netfilter/ip_conntrack_proto_icmp.c index 98f0015dd25..e4d6b268e8c 100644 --- a/net/ipv4/netfilter/ip_conntrack_proto_icmp.c +++ b/net/ipv4/netfilter/ip_conntrack_proto_icmp.c @@ -13,6 +13,7 @@ #include <linux/in.h> #include <linux/icmp.h> #include <linux/seq_file.h> +#include <linux/skbuff.h> #include <net/ip.h> #include <net/checksum.h> #include <linux/netfilter.h> @@ -151,13 +152,13 @@ icmp_error_message(struct sk_buff *skb, /* Not enough header? */ inside = skb_header_pointer(skb, skb->nh.iph->ihl*4, sizeof(_in), &_in); if (inside == NULL) - return NF_ACCEPT; + return -NF_ACCEPT; /* Ignore ICMP's containing fragments (shouldn't happen) */ if (inside->ip.frag_off & htons(IP_OFFSET)) { DEBUGP("icmp_error_track: fragment of proto %u\n", inside->ip.protocol); - return NF_ACCEPT; + return -NF_ACCEPT; } innerproto = ip_conntrack_proto_find_get(inside->ip.protocol); @@ -166,7 +167,7 @@ icmp_error_message(struct sk_buff *skb, if (!ip_ct_get_tuple(&inside->ip, skb, dataoff, &origtuple, innerproto)) { DEBUGP("icmp_error: ! get_tuple p=%u", inside->ip.protocol); ip_conntrack_proto_put(innerproto); - return NF_ACCEPT; + return -NF_ACCEPT; } /* Ordinarily, we'd expect the inverted tupleproto, but it's @@ -174,7 +175,7 @@ icmp_error_message(struct sk_buff *skb, if (!ip_ct_invert_tuple(&innertuple, &origtuple, innerproto)) { DEBUGP("icmp_error_track: Can't invert tuple\n"); ip_conntrack_proto_put(innerproto); - return NF_ACCEPT; + return -NF_ACCEPT; } ip_conntrack_proto_put(innerproto); @@ -190,7 +191,7 @@ icmp_error_message(struct sk_buff *skb, if (!h) { DEBUGP("icmp_error_track: no match\n"); - return NF_ACCEPT; + return -NF_ACCEPT; } /* Reverse direction from that found */ if (DIRECTION(h) != IP_CT_DIR_REPLY) @@ -230,19 +231,15 @@ icmp_error(struct sk_buff *skb, enum ip_conntrack_info *ctinfo, case CHECKSUM_HW: if (!(u16)csum_fold(skb->csum)) break; - if (LOG_INVALID(IPPROTO_ICMP)) - nf_log_packet(PF_INET, 0, skb, NULL, NULL, NULL, - "ip_ct_icmp: bad HW ICMP checksum "); - return -NF_ACCEPT; + /* fall through */ case CHECKSUM_NONE: - if ((u16)csum_fold(skb_checksum(skb, 0, skb->len, 0))) { + skb->csum = 0; + if (__skb_checksum_complete(skb)) { if (LOG_INVALID(IPPROTO_ICMP)) nf_log_packet(PF_INET, 0, skb, NULL, NULL, NULL, "ip_ct_icmp: bad ICMP checksum "); return -NF_ACCEPT; } - default: - break; } checksum_skipped: @@ -296,7 +293,8 @@ static int icmp_nfattr_to_tuple(struct nfattr *tb[], struct ip_conntrack_tuple *tuple) { if (!tb[CTA_PROTO_ICMP_TYPE-1] - || !tb[CTA_PROTO_ICMP_CODE-1]) + || !tb[CTA_PROTO_ICMP_CODE-1] + || !tb[CTA_PROTO_ICMP_ID-1]) return -1; tuple->dst.u.icmp.type = @@ -304,7 +302,7 @@ static int icmp_nfattr_to_tuple(struct nfattr *tb[], tuple->dst.u.icmp.code = *(u_int8_t *)NFA_DATA(tb[CTA_PROTO_ICMP_CODE-1]); tuple->src.u.icmp.id = - *(u_int8_t *)NFA_DATA(tb[CTA_PROTO_ICMP_ID-1]); + *(u_int16_t *)NFA_DATA(tb[CTA_PROTO_ICMP_ID-1]); return 0; } diff --git a/net/ipv4/netfilter/ip_conntrack_proto_tcp.c b/net/ipv4/netfilter/ip_conntrack_proto_tcp.c index d6701cafbcc..468c6003b4c 100644 --- a/net/ipv4/netfilter/ip_conntrack_proto_tcp.c +++ b/net/ipv4/netfilter/ip_conntrack_proto_tcp.c @@ -362,8 +362,12 @@ static int nfattr_to_tcp(struct nfattr *cda[], struct ip_conntrack *ct) struct nfattr *attr = cda[CTA_PROTOINFO_TCP-1]; struct nfattr *tb[CTA_PROTOINFO_TCP_MAX]; - if (nfattr_parse_nested(tb, CTA_PROTOINFO_TCP_MAX, attr) < 0) - goto nfattr_failure; + /* updates could not contain anything about the private + * protocol info, in that case skip the parsing */ + if (!attr) + return 0; + + nfattr_parse_nested(tb, CTA_PROTOINFO_TCP_MAX, attr); if (!tb[CTA_PROTOINFO_TCP_STATE-1]) return -EINVAL; @@ -374,9 +378,6 @@ static int nfattr_to_tcp(struct nfattr *cda[], struct ip_conntrack *ct) write_unlock_bh(&tcp_lock); return 0; - -nfattr_failure: - return -1; } #endif diff --git a/net/ipv4/netfilter/ip_nat_helper_pptp.c b/net/ipv4/netfilter/ip_nat_helper_pptp.c index ee6ab74ad3a..e546203f566 100644 --- a/net/ipv4/netfilter/ip_nat_helper_pptp.c +++ b/net/ipv4/netfilter/ip_nat_helper_pptp.c @@ -73,6 +73,7 @@ static void pptp_nat_expected(struct ip_conntrack *ct, struct ip_conntrack_tuple t; struct ip_ct_pptp_master *ct_pptp_info; struct ip_nat_pptp *nat_pptp_info; + struct ip_nat_range range; ct_pptp_info = &master->help.ct_pptp_info; nat_pptp_info = &master->nat.help.nat_pptp_info; @@ -110,7 +111,30 @@ static void pptp_nat_expected(struct ip_conntrack *ct, DEBUGP("not found!\n"); } - ip_nat_follow_master(ct, exp); + /* This must be a fresh one. */ + BUG_ON(ct->status & IPS_NAT_DONE_MASK); + + /* Change src to where master sends to */ + range.flags = IP_NAT_RANGE_MAP_IPS; + range.min_ip = range.max_ip + = ct->master->tuplehash[!exp->dir].tuple.dst.ip; + if (exp->dir == IP_CT_DIR_ORIGINAL) { + range.flags |= IP_NAT_RANGE_PROTO_SPECIFIED; + range.min = range.max = exp->saved_proto; + } + /* hook doesn't matter, but it has to do source manip */ + ip_nat_setup_info(ct, &range, NF_IP_POST_ROUTING); + + /* For DST manip, map port here to where it's expected. */ + range.flags = IP_NAT_RANGE_MAP_IPS; + range.min_ip = range.max_ip + = ct->master->tuplehash[!exp->dir].tuple.src.ip; + if (exp->dir == IP_CT_DIR_REPLY) { + range.flags |= IP_NAT_RANGE_PROTO_SPECIFIED; + range.min = range.max = exp->saved_proto; + } + /* hook doesn't matter, but it has to do destination manip */ + ip_nat_setup_info(ct, &range, NF_IP_PRE_ROUTING); } /* outbound packets == from PNS to PAC */ @@ -213,7 +237,7 @@ pptp_exp_gre(struct ip_conntrack_expect *expect_orig, /* alter expectation for PNS->PAC direction */ invert_tuplepr(&inv_t, &expect_orig->tuple); - expect_orig->saved_proto.gre.key = htons(nat_pptp_info->pac_call_id); + expect_orig->saved_proto.gre.key = htons(ct_pptp_info->pns_call_id); expect_orig->tuple.src.u.gre.key = htons(nat_pptp_info->pns_call_id); expect_orig->tuple.dst.u.gre.key = htons(ct_pptp_info->pac_call_id); expect_orig->dir = IP_CT_DIR_ORIGINAL; diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c index 9bcb398fbc1..45c52d8f4d9 100644 --- a/net/ipv4/netfilter/ipt_CLUSTERIP.c +++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c @@ -29,7 +29,7 @@ #include <linux/netfilter_ipv4/ip_tables.h> #include <linux/netfilter_ipv4/ipt_CLUSTERIP.h> -#include <linux/netfilter_ipv4/ip_conntrack.h> +#include <net/netfilter/nf_conntrack_compat.h> #define CLUSTERIP_VERSION "0.8" @@ -316,14 +316,14 @@ target(struct sk_buff **pskb, { const struct ipt_clusterip_tgt_info *cipinfo = targinfo; enum ip_conntrack_info ctinfo; - struct ip_conntrack *ct = ip_conntrack_get((*pskb), &ctinfo); - u_int32_t hash; + u_int32_t *mark, hash; /* don't need to clusterip_config_get() here, since refcount * is only decremented by destroy() - and ip_tables guarantees * that the ->target() function isn't called after ->destroy() */ - if (!ct) { + mark = nf_ct_get_mark((*pskb), &ctinfo); + if (mark == NULL) { printk(KERN_ERR "CLUSTERIP: no conntrack!\n"); /* FIXME: need to drop invalid ones, since replies * to outgoing connections of other nodes will be @@ -346,7 +346,7 @@ target(struct sk_buff **pskb, switch (ctinfo) { case IP_CT_NEW: - ct->mark = hash; + *mark = hash; break; case IP_CT_RELATED: case IP_CT_RELATED+IP_CT_IS_REPLY: @@ -363,7 +363,7 @@ target(struct sk_buff **pskb, #ifdef DEBUG_CLUSTERP DUMP_TUPLE(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); #endif - DEBUGP("hash=%u ct_hash=%u ", hash, ct->mark); + DEBUGP("hash=%u ct_hash=%u ", hash, *mark); if (!clusterip_responsible(cipinfo->config, hash)) { DEBUGP("not responsible\n"); return NF_DROP; diff --git a/net/ipv4/netfilter/ipt_CONNMARK.c b/net/ipv4/netfilter/ipt_CONNMARK.c index 05d66ab5942..8acac5a40a9 100644 --- a/net/ipv4/netfilter/ipt_CONNMARK.c +++ b/net/ipv4/netfilter/ipt_CONNMARK.c @@ -29,7 +29,7 @@ MODULE_LICENSE("GPL"); #include <linux/netfilter_ipv4/ip_tables.h> #include <linux/netfilter_ipv4/ipt_CONNMARK.h> -#include <linux/netfilter_ipv4/ip_conntrack.h> +#include <net/netfilter/nf_conntrack_compat.h> static unsigned int target(struct sk_buff **pskb, @@ -43,24 +43,24 @@ target(struct sk_buff **pskb, u_int32_t diff; u_int32_t nfmark; u_int32_t newmark; + u_int32_t ctinfo; + u_int32_t *ctmark = nf_ct_get_mark(*pskb, &ctinfo); - enum ip_conntrack_info ctinfo; - struct ip_conntrack *ct = ip_conntrack_get((*pskb), &ctinfo); - if (ct) { + if (ctmark) { switch(markinfo->mode) { case IPT_CONNMARK_SET: - newmark = (ct->mark & ~markinfo->mask) | markinfo->mark; - if (newmark != ct->mark) - ct->mark = newmark; + newmark = (*ctmark & ~markinfo->mask) | markinfo->mark; + if (newmark != *ctmark) + *ctmark = newmark; break; case IPT_CONNMARK_SAVE: - newmark = (ct->mark & ~markinfo->mask) | ((*pskb)->nfmark & markinfo->mask); - if (ct->mark != newmark) - ct->mark = newmark; + newmark = (*ctmark & ~markinfo->mask) | ((*pskb)->nfmark & markinfo->mask); + if (*ctmark != newmark) + *ctmark = newmark; break; case IPT_CONNMARK_RESTORE: nfmark = (*pskb)->nfmark; - diff = (ct->mark ^ nfmark) & markinfo->mask; + diff = (*ctmark ^ nfmark) & markinfo->mask; if (diff != 0) (*pskb)->nfmark = nfmark ^ diff; break; diff --git a/net/ipv4/netfilter/ipt_NOTRACK.c b/net/ipv4/netfilter/ipt_NOTRACK.c index a4bb9b3bc29..e3c69d072c6 100644 --- a/net/ipv4/netfilter/ipt_NOTRACK.c +++ b/net/ipv4/netfilter/ipt_NOTRACK.c @@ -5,7 +5,7 @@ #include <linux/skbuff.h> #include <linux/netfilter_ipv4/ip_tables.h> -#include <linux/netfilter_ipv4/ip_conntrack.h> +#include <net/netfilter/nf_conntrack_compat.h> static unsigned int target(struct sk_buff **pskb, @@ -23,7 +23,7 @@ target(struct sk_buff **pskb, If there is a real ct entry correspondig to this packet, it'll hang aroun till timing out. We don't deal with it for performance reasons. JK */ - (*pskb)->nfct = &ip_conntrack_untracked.ct_general; + nf_ct_untrack(*pskb); (*pskb)->nfctinfo = IP_CT_NEW; nf_conntrack_get((*pskb)->nfct); diff --git a/net/ipv4/netfilter/ipt_connbytes.c b/net/ipv4/netfilter/ipt_connbytes.c index df4a42c6da2..d68a048b717 100644 --- a/net/ipv4/netfilter/ipt_connbytes.c +++ b/net/ipv4/netfilter/ipt_connbytes.c @@ -10,7 +10,7 @@ */ #include <linux/module.h> #include <linux/skbuff.h> -#include <linux/netfilter_ipv4/ip_conntrack.h> +#include <net/netfilter/nf_conntrack_compat.h> #include <linux/netfilter_ipv4/ip_tables.h> #include <linux/netfilter_ipv4/ipt_connbytes.h> @@ -46,60 +46,59 @@ match(const struct sk_buff *skb, int *hotdrop) { const struct ipt_connbytes_info *sinfo = matchinfo; - enum ip_conntrack_info ctinfo; - struct ip_conntrack *ct; u_int64_t what = 0; /* initialize to make gcc happy */ + const struct ip_conntrack_counter *counters; - if (!(ct = ip_conntrack_get((struct sk_buff *)skb, &ctinfo))) + if (!(counters = nf_ct_get_counters(skb))) return 0; /* no match */ switch (sinfo->what) { case IPT_CONNBYTES_PKTS: switch (sinfo->direction) { case IPT_CONNBYTES_DIR_ORIGINAL: - what = ct->counters[IP_CT_DIR_ORIGINAL].packets; + what = counters[IP_CT_DIR_ORIGINAL].packets; break; case IPT_CONNBYTES_DIR_REPLY: - what = ct->counters[IP_CT_DIR_REPLY].packets; + what = counters[IP_CT_DIR_REPLY].packets; break; case IPT_CONNBYTES_DIR_BOTH: - what = ct->counters[IP_CT_DIR_ORIGINAL].packets; - what += ct->counters[IP_CT_DIR_REPLY].packets; + what = counters[IP_CT_DIR_ORIGINAL].packets; + what += counters[IP_CT_DIR_REPLY].packets; break; } break; case IPT_CONNBYTES_BYTES: switch (sinfo->direction) { case IPT_CONNBYTES_DIR_ORIGINAL: - what = ct->counters[IP_CT_DIR_ORIGINAL].bytes; + what = counters[IP_CT_DIR_ORIGINAL].bytes; break; case IPT_CONNBYTES_DIR_REPLY: - what = ct->counters[IP_CT_DIR_REPLY].bytes; + what = counters[IP_CT_DIR_REPLY].bytes; break; case IPT_CONNBYTES_DIR_BOTH: - what = ct->counters[IP_CT_DIR_ORIGINAL].bytes; - what += ct->counters[IP_CT_DIR_REPLY].bytes; + what = counters[IP_CT_DIR_ORIGINAL].bytes; + what += counters[IP_CT_DIR_REPLY].bytes; break; } break; case IPT_CONNBYTES_AVGPKT: switch (sinfo->direction) { case IPT_CONNBYTES_DIR_ORIGINAL: - what = div64_64(ct->counters[IP_CT_DIR_ORIGINAL].bytes, - ct->counters[IP_CT_DIR_ORIGINAL].packets); + what = div64_64(counters[IP_CT_DIR_ORIGINAL].bytes, + counters[IP_CT_DIR_ORIGINAL].packets); break; case IPT_CONNBYTES_DIR_REPLY: - what = div64_64(ct->counters[IP_CT_DIR_REPLY].bytes, - ct->counters[IP_CT_DIR_REPLY].packets); + what = div64_64(counters[IP_CT_DIR_REPLY].bytes, + counters[IP_CT_DIR_REPLY].packets); break; case IPT_CONNBYTES_DIR_BOTH: { u_int64_t bytes; u_int64_t pkts; - bytes = ct->counters[IP_CT_DIR_ORIGINAL].bytes + - ct->counters[IP_CT_DIR_REPLY].bytes; - pkts = ct->counters[IP_CT_DIR_ORIGINAL].packets+ - ct->counters[IP_CT_DIR_REPLY].packets; + bytes = counters[IP_CT_DIR_ORIGINAL].bytes + + counters[IP_CT_DIR_REPLY].bytes; + pkts = counters[IP_CT_DIR_ORIGINAL].packets+ + counters[IP_CT_DIR_REPLY].packets; /* FIXME_THEORETICAL: what to do if sum * overflows ? */ diff --git a/net/ipv4/netfilter/ipt_connmark.c b/net/ipv4/netfilter/ipt_connmark.c index bf8de47ce00..5306ef293b9 100644 --- a/net/ipv4/netfilter/ipt_connmark.c +++ b/net/ipv4/netfilter/ipt_connmark.c @@ -28,7 +28,7 @@ MODULE_LICENSE("GPL"); #include <linux/netfilter_ipv4/ip_tables.h> #include <linux/netfilter_ipv4/ipt_connmark.h> -#include <linux/netfilter_ipv4/ip_conntrack.h> +#include <net/netfilter/nf_conntrack_compat.h> static int match(const struct sk_buff *skb, @@ -39,12 +39,12 @@ match(const struct sk_buff *skb, int *hotdrop) { const struct ipt_connmark_info *info = matchinfo; - enum ip_conntrack_info ctinfo; - struct ip_conntrack *ct = ip_conntrack_get((struct sk_buff *)skb, &ctinfo); - if (!ct) + u_int32_t ctinfo; + const u_int32_t *ctmark = nf_ct_get_mark(skb, &ctinfo); + if (!ctmark) return 0; - return ((ct->mark & info->mask) == info->mark) ^ info->invert; + return (((*ctmark) & info->mask) == info->mark) ^ info->invert; } static int diff --git a/net/ipv4/netfilter/ipt_conntrack.c b/net/ipv4/netfilter/ipt_conntrack.c index c1d22801b7c..c8d18705469 100644 --- a/net/ipv4/netfilter/ipt_conntrack.c +++ b/net/ipv4/netfilter/ipt_conntrack.c @@ -10,7 +10,14 @@ #include <linux/module.h> #include <linux/skbuff.h> + +#if defined(CONFIG_IP_NF_CONNTRACK) || defined(CONFIG_IP_NF_CONNTRACK_MODULE) #include <linux/netfilter_ipv4/ip_conntrack.h> +#include <linux/netfilter_ipv4/ip_conntrack_tuple.h> +#else +#include <net/netfilter/nf_conntrack.h> +#endif + #include <linux/netfilter_ipv4/ip_tables.h> #include <linux/netfilter_ipv4/ipt_conntrack.h> @@ -18,6 +25,8 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR("Marc Boucher <marc@mbsi.ca>"); MODULE_DESCRIPTION("iptables connection tracking match module"); +#if defined(CONFIG_IP_NF_CONNTRACK) || defined(CONFIG_IP_NF_CONNTRACK_MODULE) + static int match(const struct sk_buff *skb, const struct net_device *in, @@ -102,6 +111,93 @@ match(const struct sk_buff *skb, return 1; } +#else /* CONFIG_IP_NF_CONNTRACK */ +static int +match(const struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + int *hotdrop) +{ + const struct ipt_conntrack_info *sinfo = matchinfo; + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + unsigned int statebit; + + ct = nf_ct_get((struct sk_buff *)skb, &ctinfo); + +#define FWINV(bool,invflg) ((bool) ^ !!(sinfo->invflags & invflg)) + + if (ct == &nf_conntrack_untracked) + statebit = IPT_CONNTRACK_STATE_UNTRACKED; + else if (ct) + statebit = IPT_CONNTRACK_STATE_BIT(ctinfo); + else + statebit = IPT_CONNTRACK_STATE_INVALID; + + if(sinfo->flags & IPT_CONNTRACK_STATE) { + if (ct) { + if(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip != + ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip) + statebit |= IPT_CONNTRACK_STATE_SNAT; + + if(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.u3.ip != + ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3.ip) + statebit |= IPT_CONNTRACK_STATE_DNAT; + } + + if (FWINV((statebit & sinfo->statemask) == 0, IPT_CONNTRACK_STATE)) + return 0; + } + + if(sinfo->flags & IPT_CONNTRACK_PROTO) { + if (!ct || FWINV(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum != sinfo->tuple[IP_CT_DIR_ORIGINAL].dst.protonum, IPT_CONNTRACK_PROTO)) + return 0; + } + + if(sinfo->flags & IPT_CONNTRACK_ORIGSRC) { + if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip&sinfo->sipmsk[IP_CT_DIR_ORIGINAL].s_addr) != sinfo->tuple[IP_CT_DIR_ORIGINAL].src.ip, IPT_CONNTRACK_ORIGSRC)) + return 0; + } + + if(sinfo->flags & IPT_CONNTRACK_ORIGDST) { + if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.u3.ip&sinfo->dipmsk[IP_CT_DIR_ORIGINAL].s_addr) != sinfo->tuple[IP_CT_DIR_ORIGINAL].dst.ip, IPT_CONNTRACK_ORIGDST)) + return 0; + } + + if(sinfo->flags & IPT_CONNTRACK_REPLSRC) { + if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3.ip&sinfo->sipmsk[IP_CT_DIR_REPLY].s_addr) != sinfo->tuple[IP_CT_DIR_REPLY].src.ip, IPT_CONNTRACK_REPLSRC)) + return 0; + } + + if(sinfo->flags & IPT_CONNTRACK_REPLDST) { + if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip&sinfo->dipmsk[IP_CT_DIR_REPLY].s_addr) != sinfo->tuple[IP_CT_DIR_REPLY].dst.ip, IPT_CONNTRACK_REPLDST)) + return 0; + } + + if(sinfo->flags & IPT_CONNTRACK_STATUS) { + if (!ct || FWINV((ct->status & sinfo->statusmask) == 0, IPT_CONNTRACK_STATUS)) + return 0; + } + + if(sinfo->flags & IPT_CONNTRACK_EXPIRES) { + unsigned long expires; + + if(!ct) + return 0; + + expires = timer_pending(&ct->timeout) ? (ct->timeout.expires - jiffies)/HZ : 0; + + if (FWINV(!(expires >= sinfo->expires_min && expires <= sinfo->expires_max), IPT_CONNTRACK_EXPIRES)) + return 0; + } + + return 1; +} + +#endif /* CONFIG_NF_IP_CONNTRACK */ + static int check(const char *tablename, const struct ipt_ip *ip, void *matchinfo, diff --git a/net/ipv4/netfilter/ipt_helper.c b/net/ipv4/netfilter/ipt_helper.c index 3e7dd014de4..bf14e1c7798 100644 --- a/net/ipv4/netfilter/ipt_helper.c +++ b/net/ipv4/netfilter/ipt_helper.c @@ -13,9 +13,15 @@ #include <linux/module.h> #include <linux/skbuff.h> #include <linux/netfilter.h> +#if defined(CONFIG_IP_NF_CONNTRACK) || defined(CONFIG_IP_NF_CONNTRACK_MODULE) #include <linux/netfilter_ipv4/ip_conntrack.h> #include <linux/netfilter_ipv4/ip_conntrack_core.h> #include <linux/netfilter_ipv4/ip_conntrack_helper.h> +#else +#include <net/netfilter/nf_conntrack.h> +#include <net/netfilter/nf_conntrack_core.h> +#include <net/netfilter/nf_conntrack_helper.h> +#endif #include <linux/netfilter_ipv4/ip_tables.h> #include <linux/netfilter_ipv4/ipt_helper.h> @@ -29,6 +35,7 @@ MODULE_DESCRIPTION("iptables helper match module"); #define DEBUGP(format, args...) #endif +#if defined(CONFIG_IP_NF_CONNTRACK) || defined(CONFIG_IP_NF_CONNTRACK_MODULE) static int match(const struct sk_buff *skb, const struct net_device *in, @@ -73,6 +80,53 @@ out_unlock: return ret; } +#else /* CONFIG_IP_NF_CONNTRACK */ + +static int +match(const struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + int *hotdrop) +{ + const struct ipt_helper_info *info = matchinfo; + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + int ret = info->invert; + + ct = nf_ct_get((struct sk_buff *)skb, &ctinfo); + if (!ct) { + DEBUGP("ipt_helper: Eek! invalid conntrack?\n"); + return ret; + } + + if (!ct->master) { + DEBUGP("ipt_helper: conntrack %p has no master\n", ct); + return ret; + } + + read_lock_bh(&nf_conntrack_lock); + if (!ct->master->helper) { + DEBUGP("ipt_helper: master ct %p has no helper\n", + exp->expectant); + goto out_unlock; + } + + DEBUGP("master's name = %s , info->name = %s\n", + ct->master->helper->name, info->name); + + if (info->name[0] == '\0') + ret ^= 1; + else + ret ^= !strncmp(ct->master->helper->name, info->name, + strlen(ct->master->helper->name)); +out_unlock: + read_unlock_bh(&nf_conntrack_lock); + return ret; +} +#endif + static int check(const char *tablename, const struct ipt_ip *ip, void *matchinfo, diff --git a/net/ipv4/netfilter/ipt_state.c b/net/ipv4/netfilter/ipt_state.c index b1511b97ea5..4d7f16b70ce 100644 --- a/net/ipv4/netfilter/ipt_state.c +++ b/net/ipv4/netfilter/ipt_state.c @@ -10,7 +10,7 @@ #include <linux/module.h> #include <linux/skbuff.h> -#include <linux/netfilter_ipv4/ip_conntrack.h> +#include <net/netfilter/nf_conntrack_compat.h> #include <linux/netfilter_ipv4/ip_tables.h> #include <linux/netfilter_ipv4/ipt_state.h> @@ -30,9 +30,9 @@ match(const struct sk_buff *skb, enum ip_conntrack_info ctinfo; unsigned int statebit; - if (skb->nfct == &ip_conntrack_untracked.ct_general) + if (nf_ct_is_untracked(skb)) statebit = IPT_STATE_UNTRACKED; - else if (!ip_conntrack_get(skb, &ctinfo)) + else if (!nf_ct_get_ctinfo(skb, &ctinfo)) statebit = IPT_STATE_INVALID; else statebit = IPT_STATE_BIT(ctinfo); diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c new file mode 100644 index 00000000000..8202c1c0afa --- /dev/null +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c @@ -0,0 +1,571 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - move L3 protocol dependent part to this file. + * 23 Mar 2004: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - add get_features() to support various size of conntrack + * structures. + * + * Derived from net/ipv4/netfilter/ip_conntrack_standalone.c + */ + +#include <linux/config.h> +#include <linux/types.h> +#include <linux/ip.h> +#include <linux/netfilter.h> +#include <linux/module.h> +#include <linux/skbuff.h> +#include <linux/icmp.h> +#include <linux/sysctl.h> +#include <net/ip.h> + +#include <linux/netfilter_ipv4.h> +#include <net/netfilter/nf_conntrack.h> +#include <net/netfilter/nf_conntrack_helper.h> +#include <net/netfilter/nf_conntrack_protocol.h> +#include <net/netfilter/nf_conntrack_l3proto.h> +#include <net/netfilter/nf_conntrack_core.h> +#include <net/netfilter/ipv4/nf_conntrack_ipv4.h> + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +DECLARE_PER_CPU(struct nf_conntrack_stat, nf_conntrack_stat); + +static int ipv4_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff, + struct nf_conntrack_tuple *tuple) +{ + u_int32_t _addrs[2], *ap; + ap = skb_header_pointer(skb, nhoff + offsetof(struct iphdr, saddr), + sizeof(u_int32_t) * 2, _addrs); + if (ap == NULL) + return 0; + + tuple->src.u3.ip = ap[0]; + tuple->dst.u3.ip = ap[1]; + + return 1; +} + +static int ipv4_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + tuple->src.u3.ip = orig->dst.u3.ip; + tuple->dst.u3.ip = orig->src.u3.ip; + + return 1; +} + +static int ipv4_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return seq_printf(s, "src=%u.%u.%u.%u dst=%u.%u.%u.%u ", + NIPQUAD(tuple->src.u3.ip), + NIPQUAD(tuple->dst.u3.ip)); +} + +static int ipv4_print_conntrack(struct seq_file *s, + const struct nf_conn *conntrack) +{ + return 0; +} + +/* Returns new sk_buff, or NULL */ +static struct sk_buff * +nf_ct_ipv4_gather_frags(struct sk_buff *skb, u_int32_t user) +{ + skb_orphan(skb); + + local_bh_disable(); + skb = ip_defrag(skb, user); + local_bh_enable(); + + if (skb) + ip_send_check(skb->nh.iph); + + return skb; +} + +static int +ipv4_prepare(struct sk_buff **pskb, unsigned int hooknum, unsigned int *dataoff, + u_int8_t *protonum) +{ + /* Never happen */ + if ((*pskb)->nh.iph->frag_off & htons(IP_OFFSET)) { + if (net_ratelimit()) { + printk(KERN_ERR "ipv4_prepare: Frag of proto %u (hook=%u)\n", + (*pskb)->nh.iph->protocol, hooknum); + } + return -NF_DROP; + } + + *dataoff = (*pskb)->nh.raw - (*pskb)->data + (*pskb)->nh.iph->ihl*4; + *protonum = (*pskb)->nh.iph->protocol; + + return NF_ACCEPT; +} + +int nat_module_is_loaded = 0; +static u_int32_t ipv4_get_features(const struct nf_conntrack_tuple *tuple) +{ + if (nat_module_is_loaded) + return NF_CT_F_NAT; + + return NF_CT_F_BASIC; +} + +static unsigned int ipv4_confirm(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + /* We've seen it coming out the other side: confirm it */ + return nf_conntrack_confirm(pskb); +} + +static unsigned int ipv4_conntrack_help(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + + /* This is where we call the helper: as the packet goes out. */ + ct = nf_ct_get(*pskb, &ctinfo); + if (ct && ct->helper) { + unsigned int ret; + ret = ct->helper->help(pskb, + (*pskb)->nh.raw - (*pskb)->data + + (*pskb)->nh.iph->ihl*4, + ct, ctinfo); + if (ret != NF_ACCEPT) + return ret; + } + return NF_ACCEPT; +} + +static unsigned int ipv4_conntrack_defrag(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ +#if !defined(CONFIG_IP_NF_NAT) && !defined(CONFIG_IP_NF_NAT_MODULE) + /* Previously seen (loopback)? Ignore. Do this before + fragment check. */ + if ((*pskb)->nfct) + return NF_ACCEPT; +#endif + + /* Gather fragments. */ + if ((*pskb)->nh.iph->frag_off & htons(IP_MF|IP_OFFSET)) { + *pskb = nf_ct_ipv4_gather_frags(*pskb, + hooknum == NF_IP_PRE_ROUTING ? + IP_DEFRAG_CONNTRACK_IN : + IP_DEFRAG_CONNTRACK_OUT); + if (!*pskb) + return NF_STOLEN; + } + return NF_ACCEPT; +} + +static unsigned int ipv4_refrag(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct rtable *rt = (struct rtable *)(*pskb)->dst; + + /* We've seen it coming out the other side: confirm */ + if (ipv4_confirm(hooknum, pskb, in, out, okfn) != NF_ACCEPT) + return NF_DROP; + + /* Local packets are never produced too large for their + interface. We degfragment them at LOCAL_OUT, however, + so we have to refragment them here. */ + if ((*pskb)->len > dst_mtu(&rt->u.dst) && + !skb_shinfo(*pskb)->tso_size) { + /* No hook can be after us, so this should be OK. */ + ip_fragment(*pskb, okfn); + return NF_STOLEN; + } + return NF_ACCEPT; +} + +static unsigned int ipv4_conntrack_in(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + return nf_conntrack_in(PF_INET, hooknum, pskb); +} + +static unsigned int ipv4_conntrack_local(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + /* root is playing with raw sockets. */ + if ((*pskb)->len < sizeof(struct iphdr) + || (*pskb)->nh.iph->ihl * 4 < sizeof(struct iphdr)) { + if (net_ratelimit()) + printk("ipt_hook: happy cracking.\n"); + return NF_ACCEPT; + } + return nf_conntrack_in(PF_INET, hooknum, pskb); +} + +/* Connection tracking may drop packets, but never alters them, so + make it the first hook. */ +static struct nf_hook_ops ipv4_conntrack_defrag_ops = { + .hook = ipv4_conntrack_defrag, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_PRE_ROUTING, + .priority = NF_IP_PRI_CONNTRACK_DEFRAG, +}; + +static struct nf_hook_ops ipv4_conntrack_in_ops = { + .hook = ipv4_conntrack_in, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_PRE_ROUTING, + .priority = NF_IP_PRI_CONNTRACK, +}; + +static struct nf_hook_ops ipv4_conntrack_defrag_local_out_ops = { + .hook = ipv4_conntrack_defrag, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_LOCAL_OUT, + .priority = NF_IP_PRI_CONNTRACK_DEFRAG, +}; + +static struct nf_hook_ops ipv4_conntrack_local_out_ops = { + .hook = ipv4_conntrack_local, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_LOCAL_OUT, + .priority = NF_IP_PRI_CONNTRACK, +}; + +/* helpers */ +static struct nf_hook_ops ipv4_conntrack_helper_out_ops = { + .hook = ipv4_conntrack_help, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_POST_ROUTING, + .priority = NF_IP_PRI_CONNTRACK_HELPER, +}; + +static struct nf_hook_ops ipv4_conntrack_helper_in_ops = { + .hook = ipv4_conntrack_help, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_LOCAL_IN, + .priority = NF_IP_PRI_CONNTRACK_HELPER, +}; + + +/* Refragmenter; last chance. */ +static struct nf_hook_ops ipv4_conntrack_out_ops = { + .hook = ipv4_refrag, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_POST_ROUTING, + .priority = NF_IP_PRI_CONNTRACK_CONFIRM, +}; + +static struct nf_hook_ops ipv4_conntrack_local_in_ops = { + .hook = ipv4_confirm, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_LOCAL_IN, + .priority = NF_IP_PRI_CONNTRACK_CONFIRM, +}; + +#ifdef CONFIG_SYSCTL +/* From nf_conntrack_proto_icmp.c */ +extern unsigned long nf_ct_icmp_timeout; +static struct ctl_table_header *nf_ct_ipv4_sysctl_header; + +static ctl_table nf_ct_sysctl_table[] = { + { + .ctl_name = NET_NF_CONNTRACK_ICMP_TIMEOUT, + .procname = "nf_conntrack_icmp_timeout", + .data = &nf_ct_icmp_timeout, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { .ctl_name = 0 } +}; + +static ctl_table nf_ct_netfilter_table[] = { + { + .ctl_name = NET_NETFILTER, + .procname = "netfilter", + .mode = 0555, + .child = nf_ct_sysctl_table, + }, + { .ctl_name = 0 } +}; + +static ctl_table nf_ct_net_table[] = { + { + .ctl_name = CTL_NET, + .procname = "net", + .mode = 0555, + .child = nf_ct_netfilter_table, + }, + { .ctl_name = 0 } +}; +#endif + +/* Fast function for those who don't want to parse /proc (and I don't + blame them). */ +/* Reversing the socket's dst/src point of view gives us the reply + mapping. */ +static int +getorigdst(struct sock *sk, int optval, void __user *user, int *len) +{ + struct inet_sock *inet = inet_sk(sk); + struct nf_conntrack_tuple_hash *h; + struct nf_conntrack_tuple tuple; + + NF_CT_TUPLE_U_BLANK(&tuple); + tuple.src.u3.ip = inet->rcv_saddr; + tuple.src.u.tcp.port = inet->sport; + tuple.dst.u3.ip = inet->daddr; + tuple.dst.u.tcp.port = inet->dport; + tuple.src.l3num = PF_INET; + tuple.dst.protonum = IPPROTO_TCP; + + /* We only do TCP at the moment: is there a better way? */ + if (strcmp(sk->sk_prot->name, "TCP")) { + DEBUGP("SO_ORIGINAL_DST: Not a TCP socket\n"); + return -ENOPROTOOPT; + } + + if ((unsigned int) *len < sizeof(struct sockaddr_in)) { + DEBUGP("SO_ORIGINAL_DST: len %u not %u\n", + *len, sizeof(struct sockaddr_in)); + return -EINVAL; + } + + h = nf_conntrack_find_get(&tuple, NULL); + if (h) { + struct sockaddr_in sin; + struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(h); + + sin.sin_family = AF_INET; + sin.sin_port = ct->tuplehash[IP_CT_DIR_ORIGINAL] + .tuple.dst.u.tcp.port; + sin.sin_addr.s_addr = ct->tuplehash[IP_CT_DIR_ORIGINAL] + .tuple.dst.u3.ip; + + DEBUGP("SO_ORIGINAL_DST: %u.%u.%u.%u %u\n", + NIPQUAD(sin.sin_addr.s_addr), ntohs(sin.sin_port)); + nf_ct_put(ct); + if (copy_to_user(user, &sin, sizeof(sin)) != 0) + return -EFAULT; + else + return 0; + } + DEBUGP("SO_ORIGINAL_DST: Can't find %u.%u.%u.%u/%u-%u.%u.%u.%u/%u.\n", + NIPQUAD(tuple.src.u3.ip), ntohs(tuple.src.u.tcp.port), + NIPQUAD(tuple.dst.u3.ip), ntohs(tuple.dst.u.tcp.port)); + return -ENOENT; +} + +static struct nf_sockopt_ops so_getorigdst = { + .pf = PF_INET, + .get_optmin = SO_ORIGINAL_DST, + .get_optmax = SO_ORIGINAL_DST+1, + .get = &getorigdst, +}; + +struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4 = { + .l3proto = PF_INET, + .name = "ipv4", + .pkt_to_tuple = ipv4_pkt_to_tuple, + .invert_tuple = ipv4_invert_tuple, + .print_tuple = ipv4_print_tuple, + .print_conntrack = ipv4_print_conntrack, + .prepare = ipv4_prepare, + .get_features = ipv4_get_features, + .me = THIS_MODULE, +}; + +extern struct nf_conntrack_protocol nf_conntrack_protocol_tcp4; +extern struct nf_conntrack_protocol nf_conntrack_protocol_udp4; +extern struct nf_conntrack_protocol nf_conntrack_protocol_icmp; +static int init_or_cleanup(int init) +{ + int ret = 0; + + if (!init) goto cleanup; + + ret = nf_register_sockopt(&so_getorigdst); + if (ret < 0) { + printk(KERN_ERR "Unable to register netfilter socket option\n"); + goto cleanup_nothing; + } + + ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_tcp4); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register tcp.\n"); + goto cleanup_sockopt; + } + + ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_udp4); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register udp.\n"); + goto cleanup_tcp; + } + + ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_icmp); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register icmp.\n"); + goto cleanup_udp; + } + + ret = nf_conntrack_l3proto_register(&nf_conntrack_l3proto_ipv4); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register ipv4\n"); + goto cleanup_icmp; + } + + ret = nf_register_hook(&ipv4_conntrack_defrag_ops); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register pre-routing defrag hook.\n"); + goto cleanup_ipv4; + } + ret = nf_register_hook(&ipv4_conntrack_defrag_local_out_ops); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register local_out defrag hook.\n"); + goto cleanup_defragops; + } + + ret = nf_register_hook(&ipv4_conntrack_in_ops); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register pre-routing hook.\n"); + goto cleanup_defraglocalops; + } + + ret = nf_register_hook(&ipv4_conntrack_local_out_ops); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register local out hook.\n"); + goto cleanup_inops; + } + + ret = nf_register_hook(&ipv4_conntrack_helper_in_ops); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register local helper hook.\n"); + goto cleanup_inandlocalops; + } + + ret = nf_register_hook(&ipv4_conntrack_helper_out_ops); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register postrouting helper hook.\n"); + goto cleanup_helperinops; + } + + ret = nf_register_hook(&ipv4_conntrack_out_ops); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register post-routing hook.\n"); + goto cleanup_helperoutops; + } + + ret = nf_register_hook(&ipv4_conntrack_local_in_ops); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register local in hook.\n"); + goto cleanup_inoutandlocalops; + } + +#ifdef CONFIG_SYSCTL + nf_ct_ipv4_sysctl_header = register_sysctl_table(nf_ct_net_table, 0); + if (nf_ct_ipv4_sysctl_header == NULL) { + printk("nf_conntrack: can't register to sysctl.\n"); + ret = -ENOMEM; + goto cleanup_localinops; + } +#endif + + /* For use by REJECT target */ + ip_ct_attach = __nf_conntrack_attach; + + return ret; + + cleanup: + synchronize_net(); + ip_ct_attach = NULL; +#ifdef CONFIG_SYSCTL + unregister_sysctl_table(nf_ct_ipv4_sysctl_header); + cleanup_localinops: +#endif + nf_unregister_hook(&ipv4_conntrack_local_in_ops); + cleanup_inoutandlocalops: + nf_unregister_hook(&ipv4_conntrack_out_ops); + cleanup_helperoutops: + nf_unregister_hook(&ipv4_conntrack_helper_out_ops); + cleanup_helperinops: + nf_unregister_hook(&ipv4_conntrack_helper_in_ops); + cleanup_inandlocalops: + nf_unregister_hook(&ipv4_conntrack_local_out_ops); + cleanup_inops: + nf_unregister_hook(&ipv4_conntrack_in_ops); + cleanup_defraglocalops: + nf_unregister_hook(&ipv4_conntrack_defrag_local_out_ops); + cleanup_defragops: + nf_unregister_hook(&ipv4_conntrack_defrag_ops); + cleanup_ipv4: + nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv4); + cleanup_icmp: + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_icmp); + cleanup_udp: + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_udp4); + cleanup_tcp: + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_tcp4); + cleanup_sockopt: + nf_unregister_sockopt(&so_getorigdst); + cleanup_nothing: + return ret; +} + +MODULE_LICENSE("GPL"); + +static int __init init(void) +{ + need_nf_conntrack(); + return init_or_cleanup(1); +} + +static void __exit fini(void) +{ + init_or_cleanup(0); +} + +module_init(init); +module_exit(fini); + +void need_ip_conntrack(void) +{ +} + +EXPORT_SYMBOL(need_ip_conntrack); +EXPORT_SYMBOL(nf_ct_ipv4_gather_frags); diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c new file mode 100644 index 00000000000..7ddb5c08f7b --- /dev/null +++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c @@ -0,0 +1,301 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - enable working with Layer 3 protocol independent connection tracking. + * + * Derived from net/ipv4/netfilter/ip_conntrack_proto_icmp.c + */ + +#include <linux/types.h> +#include <linux/sched.h> +#include <linux/timer.h> +#include <linux/netfilter.h> +#include <linux/in.h> +#include <linux/icmp.h> +#include <linux/seq_file.h> +#include <net/ip.h> +#include <net/checksum.h> +#include <linux/netfilter_ipv4.h> +#include <net/netfilter/nf_conntrack_tuple.h> +#include <net/netfilter/nf_conntrack_protocol.h> +#include <net/netfilter/nf_conntrack_core.h> + +unsigned long nf_ct_icmp_timeout = 30*HZ; + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +static int icmp_pkt_to_tuple(const struct sk_buff *skb, + unsigned int dataoff, + struct nf_conntrack_tuple *tuple) +{ + struct icmphdr _hdr, *hp; + + hp = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr); + if (hp == NULL) + return 0; + + tuple->dst.u.icmp.type = hp->type; + tuple->src.u.icmp.id = hp->un.echo.id; + tuple->dst.u.icmp.code = hp->code; + + return 1; +} + +static int icmp_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + /* Add 1; spaces filled with 0. */ + static u_int8_t invmap[] + = { [ICMP_ECHO] = ICMP_ECHOREPLY + 1, + [ICMP_ECHOREPLY] = ICMP_ECHO + 1, + [ICMP_TIMESTAMP] = ICMP_TIMESTAMPREPLY + 1, + [ICMP_TIMESTAMPREPLY] = ICMP_TIMESTAMP + 1, + [ICMP_INFO_REQUEST] = ICMP_INFO_REPLY + 1, + [ICMP_INFO_REPLY] = ICMP_INFO_REQUEST + 1, + [ICMP_ADDRESS] = ICMP_ADDRESSREPLY + 1, + [ICMP_ADDRESSREPLY] = ICMP_ADDRESS + 1}; + + if (orig->dst.u.icmp.type >= sizeof(invmap) + || !invmap[orig->dst.u.icmp.type]) + return 0; + + tuple->src.u.icmp.id = orig->src.u.icmp.id; + tuple->dst.u.icmp.type = invmap[orig->dst.u.icmp.type] - 1; + tuple->dst.u.icmp.code = orig->dst.u.icmp.code; + return 1; +} + +/* Print out the per-protocol part of the tuple. */ +static int icmp_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return seq_printf(s, "type=%u code=%u id=%u ", + tuple->dst.u.icmp.type, + tuple->dst.u.icmp.code, + ntohs(tuple->src.u.icmp.id)); +} + +/* Print out the private part of the conntrack. */ +static int icmp_print_conntrack(struct seq_file *s, + const struct nf_conn *conntrack) +{ + return 0; +} + +/* Returns verdict for packet, or -1 for invalid. */ +static int icmp_packet(struct nf_conn *ct, + const struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info ctinfo, + int pf, + unsigned int hooknum) +{ + /* Try to delete connection immediately after all replies: + won't actually vanish as we still have skb, and del_timer + means this will only run once even if count hits zero twice + (theoretically possible with SMP) */ + if (CTINFO2DIR(ctinfo) == IP_CT_DIR_REPLY) { + if (atomic_dec_and_test(&ct->proto.icmp.count) + && del_timer(&ct->timeout)) + ct->timeout.function((unsigned long)ct); + } else { + atomic_inc(&ct->proto.icmp.count); + nf_conntrack_event_cache(IPCT_PROTOINFO_VOLATILE, skb); + nf_ct_refresh_acct(ct, ctinfo, skb, nf_ct_icmp_timeout); + } + + return NF_ACCEPT; +} + +/* Called when a new connection for this protocol found. */ +static int icmp_new(struct nf_conn *conntrack, + const struct sk_buff *skb, unsigned int dataoff) +{ + static u_int8_t valid_new[] + = { [ICMP_ECHO] = 1, + [ICMP_TIMESTAMP] = 1, + [ICMP_INFO_REQUEST] = 1, + [ICMP_ADDRESS] = 1 }; + + if (conntrack->tuplehash[0].tuple.dst.u.icmp.type >= sizeof(valid_new) + || !valid_new[conntrack->tuplehash[0].tuple.dst.u.icmp.type]) { + /* Can't create a new ICMP `conn' with this. */ + DEBUGP("icmp: can't create new conn with type %u\n", + conntrack->tuplehash[0].tuple.dst.u.icmp.type); + NF_CT_DUMP_TUPLE(&conntrack->tuplehash[0].tuple); + return 0; + } + atomic_set(&conntrack->proto.icmp.count, 0); + return 1; +} + +extern struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4; +/* Returns conntrack if it dealt with ICMP, and filled in skb fields */ +static int +icmp_error_message(struct sk_buff *skb, + enum ip_conntrack_info *ctinfo, + unsigned int hooknum) +{ + struct nf_conntrack_tuple innertuple, origtuple; + struct { + struct icmphdr icmp; + struct iphdr ip; + } _in, *inside; + struct nf_conntrack_protocol *innerproto; + struct nf_conntrack_tuple_hash *h; + int dataoff; + + NF_CT_ASSERT(skb->nfct == NULL); + + /* Not enough header? */ + inside = skb_header_pointer(skb, skb->nh.iph->ihl*4, sizeof(_in), &_in); + if (inside == NULL) + return -NF_ACCEPT; + + /* Ignore ICMP's containing fragments (shouldn't happen) */ + if (inside->ip.frag_off & htons(IP_OFFSET)) { + DEBUGP("icmp_error_message: fragment of proto %u\n", + inside->ip.protocol); + return -NF_ACCEPT; + } + + innerproto = nf_ct_find_proto(PF_INET, inside->ip.protocol); + dataoff = skb->nh.iph->ihl*4 + sizeof(inside->icmp); + /* Are they talking about one of our connections? */ + if (!nf_ct_get_tuple(skb, dataoff, dataoff + inside->ip.ihl*4, PF_INET, + inside->ip.protocol, &origtuple, + &nf_conntrack_l3proto_ipv4, innerproto)) { + DEBUGP("icmp_error_message: ! get_tuple p=%u", + inside->ip.protocol); + return -NF_ACCEPT; + } + + /* Ordinarily, we'd expect the inverted tupleproto, but it's + been preserved inside the ICMP. */ + if (!nf_ct_invert_tuple(&innertuple, &origtuple, + &nf_conntrack_l3proto_ipv4, innerproto)) { + DEBUGP("icmp_error_message: no match\n"); + return -NF_ACCEPT; + } + + *ctinfo = IP_CT_RELATED; + + h = nf_conntrack_find_get(&innertuple, NULL); + if (!h) { + /* Locally generated ICMPs will match inverted if they + haven't been SNAT'ed yet */ + /* FIXME: NAT code has to handle half-done double NAT --RR */ + if (hooknum == NF_IP_LOCAL_OUT) + h = nf_conntrack_find_get(&origtuple, NULL); + + if (!h) { + DEBUGP("icmp_error_message: no match\n"); + return -NF_ACCEPT; + } + + /* Reverse direction from that found */ + if (NF_CT_DIRECTION(h) == IP_CT_DIR_REPLY) + *ctinfo += IP_CT_IS_REPLY; + } else { + if (NF_CT_DIRECTION(h) == IP_CT_DIR_REPLY) + *ctinfo += IP_CT_IS_REPLY; + } + + /* Update skb to refer to this connection */ + skb->nfct = &nf_ct_tuplehash_to_ctrack(h)->ct_general; + skb->nfctinfo = *ctinfo; + return -NF_ACCEPT; +} + +/* Small and modified version of icmp_rcv */ +static int +icmp_error(struct sk_buff *skb, unsigned int dataoff, + enum ip_conntrack_info *ctinfo, int pf, unsigned int hooknum) +{ + struct icmphdr _ih, *icmph; + + /* Not enough header? */ + icmph = skb_header_pointer(skb, skb->nh.iph->ihl*4, sizeof(_ih), &_ih); + if (icmph == NULL) { + if (LOG_INVALID(IPPROTO_ICMP)) + nf_log_packet(PF_INET, 0, skb, NULL, NULL, NULL, + "nf_ct_icmp: short packet "); + return -NF_ACCEPT; + } + + /* See ip_conntrack_proto_tcp.c */ + if (hooknum != NF_IP_PRE_ROUTING) + goto checksum_skipped; + + switch (skb->ip_summed) { + case CHECKSUM_HW: + if (!(u16)csum_fold(skb->csum)) + break; + if (LOG_INVALID(IPPROTO_ICMP)) + nf_log_packet(PF_INET, 0, skb, NULL, NULL, NULL, + "nf_ct_icmp: bad HW ICMP checksum "); + return -NF_ACCEPT; + case CHECKSUM_NONE: + if ((u16)csum_fold(skb_checksum(skb, 0, skb->len, 0))) { + if (LOG_INVALID(IPPROTO_ICMP)) + nf_log_packet(PF_INET, 0, skb, NULL, NULL, + NULL, + "nf_ct_icmp: bad ICMP checksum "); + return -NF_ACCEPT; + } + default: + break; + } + +checksum_skipped: + /* + * 18 is the highest 'known' ICMP type. Anything else is a mystery + * + * RFC 1122: 3.2.2 Unknown ICMP messages types MUST be silently + * discarded. + */ + if (icmph->type > NR_ICMP_TYPES) { + if (LOG_INVALID(IPPROTO_ICMP)) + nf_log_packet(PF_INET, 0, skb, NULL, NULL, NULL, + "nf_ct_icmp: invalid ICMP type "); + return -NF_ACCEPT; + } + + /* Need to track icmp error message? */ + if (icmph->type != ICMP_DEST_UNREACH + && icmph->type != ICMP_SOURCE_QUENCH + && icmph->type != ICMP_TIME_EXCEEDED + && icmph->type != ICMP_PARAMETERPROB + && icmph->type != ICMP_REDIRECT) + return NF_ACCEPT; + + return icmp_error_message(skb, ctinfo, hooknum); +} + +struct nf_conntrack_protocol nf_conntrack_protocol_icmp = +{ + .list = { NULL, NULL }, + .l3proto = PF_INET, + .proto = IPPROTO_ICMP, + .name = "icmp", + .pkt_to_tuple = icmp_pkt_to_tuple, + .invert_tuple = icmp_invert_tuple, + .print_tuple = icmp_print_tuple, + .print_conntrack = icmp_print_conntrack, + .packet = icmp_packet, + .new = icmp_new, + .error = icmp_error, + .destroy = NULL, + .me = NULL +}; + +EXPORT_SYMBOL(nf_conntrack_protocol_icmp); diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 65268562351..01444a02b48 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -645,6 +645,14 @@ ctl_table ipv4_table[] = { .proc_handler = &proc_tcp_congestion_control, .strategy = &sysctl_tcp_congestion_control, }, + { + .ctl_name = NET_TCP_ABC, + .procname = "tcp_abc", + .data = &sysctl_tcp_abc, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, { .ctl_name = 0 } }; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 72b7c22e1ea..9ac7a4f46bd 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1640,7 +1640,7 @@ int tcp_disconnect(struct sock *sk, int flags) } else if (tcp_need_reset(old_state) || (tp->snd_nxt != tp->write_seq && (1 << old_state) & (TCPF_CLOSING | TCPF_LAST_ACK))) { - /* The last check adjusts for discrepance of Linux wrt. RFC + /* The last check adjusts for discrepancy of Linux wrt. RFC * states */ tcp_send_active_reset(sk, gfp_any()); @@ -1669,6 +1669,7 @@ int tcp_disconnect(struct sock *sk, int flags) tp->packets_out = 0; tp->snd_ssthresh = 0x7fffffff; tp->snd_cwnd_cnt = 0; + tp->bytes_acked = 0; tcp_set_ca_state(sk, TCP_CA_Open); tcp_clear_retrans(tp); inet_csk_delack_init(sk); diff --git a/net/ipv4/tcp_bic.c b/net/ipv4/tcp_bic.c index ae35e060904..1d0cd86621b 100644 --- a/net/ipv4/tcp_bic.c +++ b/net/ipv4/tcp_bic.c @@ -217,17 +217,15 @@ static void bictcp_cong_avoid(struct sock *sk, u32 ack, bictcp_low_utilization(sk, data_acked); - if (in_flight < tp->snd_cwnd) + if (!tcp_is_cwnd_limited(sk, in_flight)) return; - if (tp->snd_cwnd <= tp->snd_ssthresh) { - /* In "safe" area, increase. */ - if (tp->snd_cwnd < tp->snd_cwnd_clamp) - tp->snd_cwnd++; - } else { + if (tp->snd_cwnd <= tp->snd_ssthresh) + tcp_slow_start(tp); + else { bictcp_update(ca, tp->snd_cwnd); - /* In dangerous area, increase slowly. + /* In dangerous area, increase slowly. * In theory this is tp->snd_cwnd += 1 / tp->snd_cwnd */ if (tp->snd_cwnd_cnt >= ca->cnt) { diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index bbf2d6624e8..c7cc62c8dc1 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -186,24 +186,32 @@ void tcp_reno_cong_avoid(struct sock *sk, u32 ack, u32 rtt, u32 in_flight, { struct tcp_sock *tp = tcp_sk(sk); - if (in_flight < tp->snd_cwnd) + if (!tcp_is_cwnd_limited(sk, in_flight)) return; - if (tp->snd_cwnd <= tp->snd_ssthresh) { - /* In "safe" area, increase. */ - if (tp->snd_cwnd < tp->snd_cwnd_clamp) - tp->snd_cwnd++; - } else { - /* In dangerous area, increase slowly. - * In theory this is tp->snd_cwnd += 1 / tp->snd_cwnd - */ - if (tp->snd_cwnd_cnt >= tp->snd_cwnd) { - if (tp->snd_cwnd < tp->snd_cwnd_clamp) - tp->snd_cwnd++; - tp->snd_cwnd_cnt = 0; - } else - tp->snd_cwnd_cnt++; - } + /* In "safe" area, increase. */ + if (tp->snd_cwnd <= tp->snd_ssthresh) + tcp_slow_start(tp); + + /* In dangerous area, increase slowly. */ + else if (sysctl_tcp_abc) { + /* RFC3465: Apppriate Byte Count + * increase once for each full cwnd acked + */ + if (tp->bytes_acked >= tp->snd_cwnd*tp->mss_cache) { + tp->bytes_acked -= tp->snd_cwnd*tp->mss_cache; + if (tp->snd_cwnd < tp->snd_cwnd_clamp) + tp->snd_cwnd++; + } + } else { + /* In theory this is tp->snd_cwnd += 1 / tp->snd_cwnd */ + if (tp->snd_cwnd_cnt >= tp->snd_cwnd) { + if (tp->snd_cwnd < tp->snd_cwnd_clamp) + tp->snd_cwnd++; + tp->snd_cwnd_cnt = 0; + } else + tp->snd_cwnd_cnt++; + } } EXPORT_SYMBOL_GPL(tcp_reno_cong_avoid); diff --git a/net/ipv4/tcp_highspeed.c b/net/ipv4/tcp_highspeed.c index 6acc04bde08..82b3c189bd7 100644 --- a/net/ipv4/tcp_highspeed.c +++ b/net/ipv4/tcp_highspeed.c @@ -111,18 +111,17 @@ static void hstcp_init(struct sock *sk) } static void hstcp_cong_avoid(struct sock *sk, u32 adk, u32 rtt, - u32 in_flight, int good) + u32 in_flight, u32 pkts_acked) { struct tcp_sock *tp = tcp_sk(sk); struct hstcp *ca = inet_csk_ca(sk); - if (in_flight < tp->snd_cwnd) + if (!tcp_is_cwnd_limited(sk, in_flight)) return; - if (tp->snd_cwnd <= tp->snd_ssthresh) { - if (tp->snd_cwnd < tp->snd_cwnd_clamp) - tp->snd_cwnd++; - } else { + if (tp->snd_cwnd <= tp->snd_ssthresh) + tcp_slow_start(tp); + else { /* Update AIMD parameters */ if (tp->snd_cwnd > hstcp_aimd_vals[ca->ai].cwnd) { while (tp->snd_cwnd > hstcp_aimd_vals[ca->ai].cwnd && diff --git a/net/ipv4/tcp_htcp.c b/net/ipv4/tcp_htcp.c index e47b37984e9..3284cfb993e 100644 --- a/net/ipv4/tcp_htcp.c +++ b/net/ipv4/tcp_htcp.c @@ -207,14 +207,13 @@ static void htcp_cong_avoid(struct sock *sk, u32 ack, u32 rtt, struct tcp_sock *tp = tcp_sk(sk); struct htcp *ca = inet_csk_ca(sk); - if (in_flight < tp->snd_cwnd) + if (!tcp_is_cwnd_limited(sk, in_flight)) return; - if (tp->snd_cwnd <= tp->snd_ssthresh) { - /* In "safe" area, increase. */ - if (tp->snd_cwnd < tp->snd_cwnd_clamp) - tp->snd_cwnd++; - } else { + if (tp->snd_cwnd <= tp->snd_ssthresh) + tcp_slow_start(tp); + else { + measure_rtt(sk); /* keep track of number of round-trip times since last backoff event */ @@ -224,7 +223,7 @@ static void htcp_cong_avoid(struct sock *sk, u32 ack, u32 rtt, htcp_alpha_update(ca); } - /* In dangerous area, increase slowly. + /* In dangerous area, increase slowly. * In theory this is tp->snd_cwnd += alpha / tp->snd_cwnd */ if ((tp->snd_cwnd_cnt++ * ca->alpha)>>7 >= tp->snd_cwnd) { diff --git a/net/ipv4/tcp_hybla.c b/net/ipv4/tcp_hybla.c index 77add63623d..40dbb387751 100644 --- a/net/ipv4/tcp_hybla.c +++ b/net/ipv4/tcp_hybla.c @@ -100,12 +100,12 @@ static void hybla_cong_avoid(struct sock *sk, u32 ack, u32 rtt, ca->minrtt = tp->srtt; } + if (!tcp_is_cwnd_limited(sk, in_flight)) + return; + if (!ca->hybla_en) return tcp_reno_cong_avoid(sk, ack, rtt, in_flight, flag); - if (in_flight < tp->snd_cwnd) - return; - if (ca->rho == 0) hybla_recalc_param(sk); diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 3e98b57578d..40a26b7157b 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -42,7 +42,7 @@ * Andi Kleen : Moved open_request checking here * and process RSTs for open_requests. * Andi Kleen : Better prune_queue, and other fixes. - * Andrey Savochkin: Fix RTT measurements in the presnce of + * Andrey Savochkin: Fix RTT measurements in the presence of * timestamps. * Andrey Savochkin: Check sequence numbers correctly when * removing SACKs due to in sequence incoming @@ -89,6 +89,7 @@ int sysctl_tcp_frto; int sysctl_tcp_nometrics_save; int sysctl_tcp_moderate_rcvbuf = 1; +int sysctl_tcp_abc = 1; #define FLAG_DATA 0x01 /* Incoming frame contained data. */ #define FLAG_WIN_UPDATE 0x02 /* Incoming ACK was a window update. */ @@ -223,7 +224,7 @@ static void tcp_fixup_sndbuf(struct sock *sk) * of receiver window. Check #2. * * The scheme does not work when sender sends good segments opening - * window and then starts to feed us spagetti. But it should work + * window and then starts to feed us spaghetti. But it should work * in common situations. Otherwise, we have to rely on queue collapsing. */ @@ -233,7 +234,7 @@ static int __tcp_grow_window(const struct sock *sk, struct tcp_sock *tp, { /* Optimize this! */ int truesize = tcp_win_from_space(skb->truesize)/2; - int window = tcp_full_space(sk)/2; + int window = tcp_win_from_space(sysctl_tcp_rmem[2])/2; while (tp->rcv_ssthresh <= window) { if (truesize <= skb->len) @@ -277,7 +278,7 @@ static void tcp_fixup_rcvbuf(struct sock *sk) int rcvmem = tp->advmss + MAX_TCP_HEADER + 16 + sizeof(struct sk_buff); /* Try to select rcvbuf so that 4 mss-sized segments - * will fit to window and correspoding skbs will fit to our rcvbuf. + * will fit to window and corresponding skbs will fit to our rcvbuf. * (was 3; 4 is minimum to allow fast retransmit to work.) */ while (tcp_win_from_space(rcvmem) < tp->advmss) @@ -286,7 +287,7 @@ static void tcp_fixup_rcvbuf(struct sock *sk) sk->sk_rcvbuf = min(4 * rcvmem, sysctl_tcp_rmem[2]); } -/* 4. Try to fixup all. It is made iimediately after connection enters +/* 4. Try to fixup all. It is made immediately after connection enters * established state. */ static void tcp_init_buffer_space(struct sock *sk) @@ -326,37 +327,18 @@ static void tcp_init_buffer_space(struct sock *sk) static void tcp_clamp_window(struct sock *sk, struct tcp_sock *tp) { struct inet_connection_sock *icsk = inet_csk(sk); - struct sk_buff *skb; - unsigned int app_win = tp->rcv_nxt - tp->copied_seq; - int ofo_win = 0; icsk->icsk_ack.quick = 0; - skb_queue_walk(&tp->out_of_order_queue, skb) { - ofo_win += skb->len; - } - - /* If overcommit is due to out of order segments, - * do not clamp window. Try to expand rcvbuf instead. - */ - if (ofo_win) { - if (sk->sk_rcvbuf < sysctl_tcp_rmem[2] && - !(sk->sk_userlocks & SOCK_RCVBUF_LOCK) && - !tcp_memory_pressure && - atomic_read(&tcp_memory_allocated) < sysctl_tcp_mem[0]) - sk->sk_rcvbuf = min(atomic_read(&sk->sk_rmem_alloc), - sysctl_tcp_rmem[2]); + if (sk->sk_rcvbuf < sysctl_tcp_rmem[2] && + !(sk->sk_userlocks & SOCK_RCVBUF_LOCK) && + !tcp_memory_pressure && + atomic_read(&tcp_memory_allocated) < sysctl_tcp_mem[0]) { + sk->sk_rcvbuf = min(atomic_read(&sk->sk_rmem_alloc), + sysctl_tcp_rmem[2]); } - if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf) { - app_win += ofo_win; - if (atomic_read(&sk->sk_rmem_alloc) >= 2 * sk->sk_rcvbuf) - app_win >>= 1; - if (app_win > icsk->icsk_ack.rcv_mss) - app_win -= icsk->icsk_ack.rcv_mss; - app_win = max(app_win, 2U*tp->advmss); - + if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf) tp->rcv_ssthresh = min(tp->window_clamp, 2U*tp->advmss); - } } /* Receiver "autotuning" code. @@ -385,8 +367,8 @@ static void tcp_rcv_rtt_update(struct tcp_sock *tp, u32 sample, int win_dep) * are stalled on filesystem I/O. * * Also, since we are only going for a minimum in the - * non-timestamp case, we do not smoothe things out - * else with timestamps disabled convergance takes too + * non-timestamp case, we do not smoother things out + * else with timestamps disabled convergence takes too * long. */ if (!win_dep) { @@ -395,7 +377,7 @@ static void tcp_rcv_rtt_update(struct tcp_sock *tp, u32 sample, int win_dep) } else if (m < new_sample) new_sample = m << 3; } else { - /* No previous mesaure. */ + /* No previous measure. */ new_sample = m << 3; } @@ -524,7 +506,7 @@ static void tcp_event_data_recv(struct sock *sk, struct tcp_sock *tp, struct sk_ if (icsk->icsk_ack.ato > icsk->icsk_rto) icsk->icsk_ack.ato = icsk->icsk_rto; } else if (m > icsk->icsk_rto) { - /* Too long gap. Apparently sender falled to + /* Too long gap. Apparently sender failed to * restart window, so that we send ACKs quickly. */ tcp_incr_quickack(sk); @@ -548,10 +530,9 @@ static void tcp_event_data_recv(struct sock *sk, struct tcp_sock *tp, struct sk_ * To save cycles in the RFC 1323 implementation it was better to break * it up into three procedures. -- erics */ -static void tcp_rtt_estimator(struct sock *sk, const __u32 mrtt, u32 *usrtt) +static void tcp_rtt_estimator(struct sock *sk, const __u32 mrtt) { struct tcp_sock *tp = tcp_sk(sk); - const struct inet_connection_sock *icsk = inet_csk(sk); long m = mrtt; /* RTT */ /* The following amusing code comes from Jacobson's @@ -565,7 +546,7 @@ static void tcp_rtt_estimator(struct sock *sk, const __u32 mrtt, u32 *usrtt) * * Funny. This algorithm seems to be very broken. * These formulae increase RTO, when it should be decreased, increase - * too slowly, when it should be incresed fastly, decrease too fastly + * too slowly, when it should be increased fastly, decrease too fastly * etc. I guess in BSD RTO takes ONE value, so that it is absolutely * does not matter how to _calculate_ it. Seems, it was trap * that VJ failed to avoid. 8) @@ -610,9 +591,6 @@ static void tcp_rtt_estimator(struct sock *sk, const __u32 mrtt, u32 *usrtt) tp->mdev_max = tp->rttvar = max(tp->mdev, TCP_RTO_MIN); tp->rtt_seq = tp->snd_nxt; } - - if (icsk->icsk_ca_ops->rtt_sample) - icsk->icsk_ca_ops->rtt_sample(sk, *usrtt); } /* Calculate rto without backoff. This is the second half of Van Jacobson's @@ -629,14 +607,14 @@ static inline void tcp_set_rto(struct sock *sk) * at least by solaris and freebsd. "Erratic ACKs" has _nothing_ * to do with delayed acks, because at cwnd>2 true delack timeout * is invisible. Actually, Linux-2.4 also generates erratic - * ACKs in some curcumstances. + * ACKs in some circumstances. */ inet_csk(sk)->icsk_rto = (tp->srtt >> 3) + tp->rttvar; /* 2. Fixups made earlier cannot be right. * If we do not estimate RTO correctly without them, * all the algo is pure shit and should be replaced - * with correct one. It is exaclty, which we pretend to do. + * with correct one. It is exactly, which we pretend to do. */ } @@ -794,7 +772,7 @@ static void tcp_init_metrics(struct sock *sk) * to make it more realistic. * * A bit of theory. RTT is time passed after "normal" sized packet - * is sent until it is ACKed. In normal curcumstances sending small + * is sent until it is ACKed. In normal circumstances sending small * packets force peer to delay ACKs and calculation is correct too. * The algorithm is adaptive and, provided we follow specs, it * NEVER underestimate RTT. BUT! If peer tries to make some clever @@ -919,18 +897,32 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_ int prior_fackets; u32 lost_retrans = 0; int flag = 0; + int dup_sack = 0; int i; if (!tp->sacked_out) tp->fackets_out = 0; prior_fackets = tp->fackets_out; - for (i=0; i<num_sacks; i++, sp++) { - struct sk_buff *skb; - __u32 start_seq = ntohl(sp->start_seq); - __u32 end_seq = ntohl(sp->end_seq); - int fack_count = 0; - int dup_sack = 0; + /* SACK fastpath: + * if the only SACK change is the increase of the end_seq of + * the first block then only apply that SACK block + * and use retrans queue hinting otherwise slowpath */ + flag = 1; + for (i = 0; i< num_sacks; i++) { + __u32 start_seq = ntohl(sp[i].start_seq); + __u32 end_seq = ntohl(sp[i].end_seq); + + if (i == 0){ + if (tp->recv_sack_cache[i].start_seq != start_seq) + flag = 0; + } else { + if ((tp->recv_sack_cache[i].start_seq != start_seq) || + (tp->recv_sack_cache[i].end_seq != end_seq)) + flag = 0; + } + tp->recv_sack_cache[i].start_seq = start_seq; + tp->recv_sack_cache[i].end_seq = end_seq; /* Check for D-SACK. */ if (i == 0) { @@ -962,15 +954,58 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_ if (before(ack, prior_snd_una - tp->max_window)) return 0; } + } + + if (flag) + num_sacks = 1; + else { + int j; + tp->fastpath_skb_hint = NULL; + + /* order SACK blocks to allow in order walk of the retrans queue */ + for (i = num_sacks-1; i > 0; i--) { + for (j = 0; j < i; j++){ + if (after(ntohl(sp[j].start_seq), + ntohl(sp[j+1].start_seq))){ + sp[j].start_seq = htonl(tp->recv_sack_cache[j+1].start_seq); + sp[j].end_seq = htonl(tp->recv_sack_cache[j+1].end_seq); + sp[j+1].start_seq = htonl(tp->recv_sack_cache[j].start_seq); + sp[j+1].end_seq = htonl(tp->recv_sack_cache[j].end_seq); + } + + } + } + } + + /* clear flag as used for different purpose in following code */ + flag = 0; + + for (i=0; i<num_sacks; i++, sp++) { + struct sk_buff *skb; + __u32 start_seq = ntohl(sp->start_seq); + __u32 end_seq = ntohl(sp->end_seq); + int fack_count; + + /* Use SACK fastpath hint if valid */ + if (tp->fastpath_skb_hint) { + skb = tp->fastpath_skb_hint; + fack_count = tp->fastpath_cnt_hint; + } else { + skb = sk->sk_write_queue.next; + fack_count = 0; + } /* Event "B" in the comment above. */ if (after(end_seq, tp->high_seq)) flag |= FLAG_DATA_LOST; - sk_stream_for_retrans_queue(skb, sk) { + sk_stream_for_retrans_queue_from(skb, sk) { int in_sack, pcount; u8 sacked; + tp->fastpath_skb_hint = skb; + tp->fastpath_cnt_hint = fack_count; + /* The retransmission queue is always in order, so * we can short-circuit the walk early. */ @@ -1045,6 +1080,9 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_ TCP_SKB_CB(skb)->sacked &= ~(TCPCB_LOST|TCPCB_SACKED_RETRANS); tp->lost_out -= tcp_skb_pcount(skb); tp->retrans_out -= tcp_skb_pcount(skb); + + /* clear lost hint */ + tp->retransmit_skb_hint = NULL; } } else { /* New sack for not retransmitted frame, @@ -1057,6 +1095,9 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_ if (sacked & TCPCB_LOST) { TCP_SKB_CB(skb)->sacked &= ~TCPCB_LOST; tp->lost_out -= tcp_skb_pcount(skb); + + /* clear lost hint */ + tp->retransmit_skb_hint = NULL; } } @@ -1080,6 +1121,7 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_ (TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_RETRANS)) { TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS; tp->retrans_out -= tcp_skb_pcount(skb); + tp->retransmit_skb_hint = NULL; } } } @@ -1107,6 +1149,9 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_ TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS; tp->retrans_out -= tcp_skb_pcount(skb); + /* clear lost hint */ + tp->retransmit_skb_hint = NULL; + if (!(TCP_SKB_CB(skb)->sacked&(TCPCB_LOST|TCPCB_SACKED_ACKED))) { tp->lost_out += tcp_skb_pcount(skb); TCP_SKB_CB(skb)->sacked |= TCPCB_LOST; @@ -1214,6 +1259,8 @@ static void tcp_enter_frto_loss(struct sock *sk) tcp_set_ca_state(sk, TCP_CA_Loss); tp->high_seq = tp->frto_highmark; TCP_ECN_queue_cwr(tp); + + clear_all_retrans_hints(tp); } void tcp_clear_retrans(struct tcp_sock *tp) @@ -1251,6 +1298,7 @@ void tcp_enter_loss(struct sock *sk, int how) tp->snd_cwnd_cnt = 0; tp->snd_cwnd_stamp = tcp_time_stamp; + tp->bytes_acked = 0; tcp_clear_retrans(tp); /* Push undo marker, if it was plain RTO and nothing @@ -1279,6 +1327,8 @@ void tcp_enter_loss(struct sock *sk, int how) tcp_set_ca_state(sk, TCP_CA_Loss); tp->high_seq = tp->snd_nxt; TCP_ECN_queue_cwr(tp); + + clear_all_retrans_hints(tp); } static int tcp_check_sack_reneging(struct sock *sk) @@ -1503,17 +1553,37 @@ static void tcp_mark_head_lost(struct sock *sk, struct tcp_sock *tp, int packets, u32 high_seq) { struct sk_buff *skb; - int cnt = packets; + int cnt; - BUG_TRAP(cnt <= tp->packets_out); + BUG_TRAP(packets <= tp->packets_out); + if (tp->lost_skb_hint) { + skb = tp->lost_skb_hint; + cnt = tp->lost_cnt_hint; + } else { + skb = sk->sk_write_queue.next; + cnt = 0; + } - sk_stream_for_retrans_queue(skb, sk) { - cnt -= tcp_skb_pcount(skb); - if (cnt < 0 || after(TCP_SKB_CB(skb)->end_seq, high_seq)) + sk_stream_for_retrans_queue_from(skb, sk) { + /* TODO: do this better */ + /* this is not the most efficient way to do this... */ + tp->lost_skb_hint = skb; + tp->lost_cnt_hint = cnt; + cnt += tcp_skb_pcount(skb); + if (cnt > packets || after(TCP_SKB_CB(skb)->end_seq, high_seq)) break; if (!(TCP_SKB_CB(skb)->sacked&TCPCB_TAGBITS)) { TCP_SKB_CB(skb)->sacked |= TCPCB_LOST; tp->lost_out += tcp_skb_pcount(skb); + + /* clear xmit_retransmit_queue hints + * if this is beyond hint */ + if(tp->retransmit_skb_hint != NULL && + before(TCP_SKB_CB(skb)->seq, + TCP_SKB_CB(tp->retransmit_skb_hint)->seq)) { + + tp->retransmit_skb_hint = NULL; + } } } tcp_sync_left_out(tp); @@ -1540,13 +1610,28 @@ static void tcp_update_scoreboard(struct sock *sk, struct tcp_sock *tp) if (tcp_head_timedout(sk, tp)) { struct sk_buff *skb; - sk_stream_for_retrans_queue(skb, sk) { - if (tcp_skb_timedout(sk, skb) && - !(TCP_SKB_CB(skb)->sacked&TCPCB_TAGBITS)) { + skb = tp->scoreboard_skb_hint ? tp->scoreboard_skb_hint + : sk->sk_write_queue.next; + + sk_stream_for_retrans_queue_from(skb, sk) { + if (!tcp_skb_timedout(sk, skb)) + break; + + if (!(TCP_SKB_CB(skb)->sacked&TCPCB_TAGBITS)) { TCP_SKB_CB(skb)->sacked |= TCPCB_LOST; tp->lost_out += tcp_skb_pcount(skb); + + /* clear xmit_retrans hint */ + if (tp->retransmit_skb_hint && + before(TCP_SKB_CB(skb)->seq, + TCP_SKB_CB(tp->retransmit_skb_hint)->seq)) + + tp->retransmit_skb_hint = NULL; } } + + tp->scoreboard_skb_hint = skb; + tcp_sync_left_out(tp); } } @@ -1626,6 +1711,10 @@ static void tcp_undo_cwr(struct sock *sk, const int undo) } tcp_moderate_cwnd(tp); tp->snd_cwnd_stamp = tcp_time_stamp; + + /* There is something screwy going on with the retrans hints after + an undo */ + clear_all_retrans_hints(tp); } static inline int tcp_may_undo(struct tcp_sock *tp) @@ -1709,6 +1798,9 @@ static int tcp_try_undo_loss(struct sock *sk, struct tcp_sock *tp) sk_stream_for_retrans_queue(skb, sk) { TCP_SKB_CB(skb)->sacked &= ~TCPCB_LOST; } + + clear_all_retrans_hints(tp); + DBGUNDO(sk, tp, "partial loss"); tp->lost_out = 0; tp->left_out = tp->sacked_out; @@ -1908,6 +2000,7 @@ tcp_fastretrans_alert(struct sock *sk, u32 prior_snd_una, TCP_ECN_queue_cwr(tp); } + tp->bytes_acked = 0; tp->snd_cwnd_cnt = 0; tcp_set_ca_state(sk, TCP_CA_Recovery); } @@ -1919,9 +2012,9 @@ tcp_fastretrans_alert(struct sock *sk, u32 prior_snd_una, } /* Read draft-ietf-tcplw-high-performance before mucking - * with this code. (Superceeds RFC1323) + * with this code. (Supersedes RFC1323) */ -static void tcp_ack_saw_tstamp(struct sock *sk, u32 *usrtt, int flag) +static void tcp_ack_saw_tstamp(struct sock *sk, int flag) { /* RTTM Rule: A TSecr value received in a segment is used to * update the averaged RTT measurement only if the segment @@ -1932,7 +2025,7 @@ static void tcp_ack_saw_tstamp(struct sock *sk, u32 *usrtt, int flag) * 1998/04/10 Andrey V. Savochkin <saw@msu.ru> * * Changed: reset backoff as soon as we see the first valid sample. - * If we do not, we get strongly overstimated rto. With timestamps + * If we do not, we get strongly overestimated rto. With timestamps * samples are accepted even from very old segments: f.e., when rtt=1 * increases to 8, we retransmit 5 times and after 8 seconds delayed * answer arrives rto becomes 120 seconds! If at least one of segments @@ -1940,13 +2033,13 @@ static void tcp_ack_saw_tstamp(struct sock *sk, u32 *usrtt, int flag) */ struct tcp_sock *tp = tcp_sk(sk); const __u32 seq_rtt = tcp_time_stamp - tp->rx_opt.rcv_tsecr; - tcp_rtt_estimator(sk, seq_rtt, usrtt); + tcp_rtt_estimator(sk, seq_rtt); tcp_set_rto(sk); inet_csk(sk)->icsk_backoff = 0; tcp_bound_rto(sk); } -static void tcp_ack_no_tstamp(struct sock *sk, u32 seq_rtt, u32 *usrtt, int flag) +static void tcp_ack_no_tstamp(struct sock *sk, u32 seq_rtt, int flag) { /* We don't have a timestamp. Can only use * packets that are not retransmitted to determine @@ -1960,21 +2053,21 @@ static void tcp_ack_no_tstamp(struct sock *sk, u32 seq_rtt, u32 *usrtt, int flag if (flag & FLAG_RETRANS_DATA_ACKED) return; - tcp_rtt_estimator(sk, seq_rtt, usrtt); + tcp_rtt_estimator(sk, seq_rtt); tcp_set_rto(sk); inet_csk(sk)->icsk_backoff = 0; tcp_bound_rto(sk); } static inline void tcp_ack_update_rtt(struct sock *sk, const int flag, - const s32 seq_rtt, u32 *usrtt) + const s32 seq_rtt) { const struct tcp_sock *tp = tcp_sk(sk); /* Note that peer MAY send zero echo. In this case it is ignored. (rfc1323) */ if (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr) - tcp_ack_saw_tstamp(sk, usrtt, flag); + tcp_ack_saw_tstamp(sk, flag); else if (seq_rtt >= 0) - tcp_ack_no_tstamp(sk, seq_rtt, usrtt, flag); + tcp_ack_no_tstamp(sk, seq_rtt, flag); } static inline void tcp_cong_avoid(struct sock *sk, u32 ack, u32 rtt, @@ -2054,20 +2147,27 @@ static int tcp_tso_acked(struct sock *sk, struct sk_buff *skb, return acked; } +static inline u32 tcp_usrtt(const struct sk_buff *skb) +{ + struct timeval tv, now; + + do_gettimeofday(&now); + skb_get_timestamp(skb, &tv); + return (now.tv_sec - tv.tv_sec) * 1000000 + (now.tv_usec - tv.tv_usec); +} /* Remove acknowledged frames from the retransmission queue. */ -static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p, s32 *seq_usrtt) +static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p) { struct tcp_sock *tp = tcp_sk(sk); + const struct inet_connection_sock *icsk = inet_csk(sk); struct sk_buff *skb; __u32 now = tcp_time_stamp; int acked = 0; __s32 seq_rtt = -1; - struct timeval usnow; u32 pkts_acked = 0; - - if (seq_usrtt) - do_gettimeofday(&usnow); + void (*rtt_sample)(struct sock *sk, u32 usrtt) + = icsk->icsk_ca_ops->rtt_sample; while ((skb = skb_peek(&sk->sk_write_queue)) && skb != sk->sk_send_head) { @@ -2107,16 +2207,11 @@ static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p, s32 *seq_usrtt tp->retrans_out -= tcp_skb_pcount(skb); acked |= FLAG_RETRANS_DATA_ACKED; seq_rtt = -1; - } else if (seq_rtt < 0) + } else if (seq_rtt < 0) { seq_rtt = now - scb->when; - if (seq_usrtt) { - struct timeval tv; - - skb_get_timestamp(skb, &tv); - *seq_usrtt = (usnow.tv_sec - tv.tv_sec) * 1000000 - + (usnow.tv_usec - tv.tv_usec); + if (rtt_sample) + (*rtt_sample)(sk, tcp_usrtt(skb)); } - if (sacked & TCPCB_SACKED_ACKED) tp->sacked_out -= tcp_skb_pcount(skb); if (sacked & TCPCB_LOST) @@ -2126,17 +2221,20 @@ static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p, s32 *seq_usrtt !before(scb->end_seq, tp->snd_up)) tp->urg_mode = 0; } - } else if (seq_rtt < 0) + } else if (seq_rtt < 0) { seq_rtt = now - scb->when; + if (rtt_sample) + (*rtt_sample)(sk, tcp_usrtt(skb)); + } tcp_dec_pcount_approx(&tp->fackets_out, skb); tcp_packets_out_dec(tp, skb); __skb_unlink(skb, &sk->sk_write_queue); sk_stream_free_skb(sk, skb); + clear_all_retrans_hints(tp); } if (acked&FLAG_ACKED) { - const struct inet_connection_sock *icsk = inet_csk(sk); - tcp_ack_update_rtt(sk, acked, seq_rtt, seq_usrtt); + tcp_ack_update_rtt(sk, acked, seq_rtt); tcp_ack_packets_out(sk, tp); if (icsk->icsk_ca_ops->pkts_acked) @@ -2284,7 +2382,7 @@ static void tcp_process_frto(struct sock *sk, u32 prior_snd_una) } /* F-RTO affects on two new ACKs following RTO. - * At latest on third ACK the TCP behavor is back to normal. + * At latest on third ACK the TCP behavior is back to normal. */ tp->frto_counter = (tp->frto_counter + 1) % 3; } @@ -2299,7 +2397,6 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) u32 ack = TCP_SKB_CB(skb)->ack_seq; u32 prior_in_flight; s32 seq_rtt; - s32 seq_usrtt = 0; int prior_packets; /* If the ack is newer than sent or older than previous acks @@ -2311,6 +2408,9 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) if (before(ack, prior_snd_una)) goto old_ack; + if (sysctl_tcp_abc && icsk->icsk_ca_state < TCP_CA_CWR) + tp->bytes_acked += ack - prior_snd_una; + if (!(flag&FLAG_SLOWPATH) && after(ack, prior_snd_una)) { /* Window is constant, pure forward advance. * No more checks are required. @@ -2352,14 +2452,13 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) prior_in_flight = tcp_packets_in_flight(tp); /* See if we can take anything off of the retransmit queue. */ - flag |= tcp_clean_rtx_queue(sk, &seq_rtt, - icsk->icsk_ca_ops->rtt_sample ? &seq_usrtt : NULL); + flag |= tcp_clean_rtx_queue(sk, &seq_rtt); if (tp->frto_counter) tcp_process_frto(sk, prior_snd_una); if (tcp_ack_is_dubious(sk, flag)) { - /* Advanve CWND, if state allows this. */ + /* Advance CWND, if state allows this. */ if ((flag & FLAG_DATA_ACKED) && tcp_may_raise_cwnd(sk, flag)) tcp_cong_avoid(sk, ack, seq_rtt, prior_in_flight, 0); tcp_fastretrans_alert(sk, prior_snd_una, prior_packets, flag); @@ -3148,7 +3247,7 @@ tcp_collapse(struct sock *sk, struct sk_buff_head *list, { struct sk_buff *skb; - /* First, check that queue is collapsable and find + /* First, check that queue is collapsible and find * the point where collapsing can be useful. */ for (skb = head; skb != tail; ) { /* No new bits? It is possible on ofo queue. */ @@ -3456,7 +3555,7 @@ static __inline__ void tcp_ack_snd_check(struct sock *sk) /* * This routine is only called when we have urgent data - * signalled. Its the 'slow' part of tcp_urg. It could be + * signaled. Its the 'slow' part of tcp_urg. It could be * moved inline now as tcp_urg is only called from one * place. We handle URGent data wrong. We have to - as * BSD still doesn't use the correction from RFC961. @@ -3501,7 +3600,7 @@ static void tcp_check_urg(struct sock * sk, struct tcphdr * th) * urgent. To do this requires some care. We cannot just ignore * tp->copied_seq since we would read the last urgent byte again * as data, nor can we alter copied_seq until this data arrives - * or we break the sematics of SIOCATMARK (and thus sockatmark()) + * or we break the semantics of SIOCATMARK (and thus sockatmark()) * * NOTE. Double Dutch. Rendering to plain English: author of comment * above did something sort of send("A", MSG_OOB); send("B", MSG_OOB); @@ -3646,7 +3745,7 @@ int tcp_rcv_established(struct sock *sk, struct sk_buff *skb, tp->rx_opt.saw_tstamp = 0; /* pred_flags is 0xS?10 << 16 + snd_wnd - * if header_predition is to be made + * if header_prediction is to be made * 'S' will always be tp->tcp_header_len >> 2 * '?' will be 0 for the fast path, otherwise pred_flags is 0 to * turn it off (when there are holes in the receive @@ -4242,7 +4341,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb, */ if (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr && !tp->srtt) - tcp_ack_saw_tstamp(sk, NULL, 0); + tcp_ack_saw_tstamp(sk, 0); if (tp->rx_opt.tstamp_ok) tp->advmss -= TCPOLEN_TSTAMP_ALIGNED; @@ -4372,6 +4471,7 @@ discard: EXPORT_SYMBOL(sysctl_tcp_ecn); EXPORT_SYMBOL(sysctl_tcp_reordering); +EXPORT_SYMBOL(sysctl_tcp_abc); EXPORT_SYMBOL(tcp_parse_options); EXPORT_SYMBOL(tcp_rcv_established); EXPORT_SYMBOL(tcp_rcv_state_process); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 634dabb558f..4d5021e1929 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -39,7 +39,7 @@ * request_sock handling and moved * most of it into the af independent code. * Added tail drop and some other bugfixes. - * Added new listen sematics. + * Added new listen semantics. * Mike McLagan : Routing by source * Juan Jose Ciarlante: ip_dynaddr bits * Andi Kleen: various fixes. @@ -1110,24 +1110,18 @@ static struct sock *tcp_v4_hnd_req(struct sock *sk, struct sk_buff *skb) static int tcp_v4_checksum_init(struct sk_buff *skb) { if (skb->ip_summed == CHECKSUM_HW) { - skb->ip_summed = CHECKSUM_UNNECESSARY; if (!tcp_v4_check(skb->h.th, skb->len, skb->nh.iph->saddr, - skb->nh.iph->daddr, skb->csum)) + skb->nh.iph->daddr, skb->csum)) { + skb->ip_summed = CHECKSUM_UNNECESSARY; return 0; - - LIMIT_NETDEBUG(KERN_DEBUG "hw tcp v4 csum failed\n"); - skb->ip_summed = CHECKSUM_NONE; + } } + + skb->csum = csum_tcpudp_nofold(skb->nh.iph->saddr, skb->nh.iph->daddr, + skb->len, IPPROTO_TCP, 0); + if (skb->len <= 76) { - if (tcp_v4_check(skb->h.th, skb->len, skb->nh.iph->saddr, - skb->nh.iph->daddr, - skb_checksum(skb, 0, skb->len, 0))) - return -1; - skb->ip_summed = CHECKSUM_UNNECESSARY; - } else { - skb->csum = ~tcp_v4_check(skb->h.th, skb->len, - skb->nh.iph->saddr, - skb->nh.iph->daddr, 0); + return __skb_checksum_complete(skb); } return 0; } @@ -1216,10 +1210,10 @@ int tcp_v4_rcv(struct sk_buff *skb) /* An explanation is required here, I think. * Packet length and doff are validated by header prediction, - * provided case of th->doff==0 is elimineted. + * provided case of th->doff==0 is eliminated. * So, we defer the checks. */ if ((skb->ip_summed != CHECKSUM_UNNECESSARY && - tcp_v4_checksum_init(skb) < 0)) + tcp_v4_checksum_init(skb))) goto bad_packet; th = skb->h.th; diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index b1a63b2c6b4..1b66a2ac432 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -158,7 +158,7 @@ kill_with_rst: /* I am shamed, but failed to make it more elegant. * Yes, it is direct reference to IP, which is impossible * to generalize to IPv6. Taking into account that IPv6 - * do not undertsnad recycling in any case, it not + * do not understand recycling in any case, it not * a big problem in practice. --ANK */ if (tw->tw_family == AF_INET && tcp_death_row.sysctl_tw_recycle && tcptw->tw_ts_recent_stamp && @@ -194,7 +194,7 @@ kill_with_rst: /* In window segment, it may be only reset or bare ack. */ if (th->rst) { - /* This is TIME_WAIT assasination, in two flavors. + /* This is TIME_WAIT assassination, in two flavors. * Oh well... nobody has a sufficient solution to this * protocol bug yet. */ @@ -380,6 +380,7 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct request_sock *req, */ newtp->snd_cwnd = 2; newtp->snd_cwnd_cnt = 0; + newtp->bytes_acked = 0; newtp->frto_counter = 0; newtp->frto_highmark = 0; @@ -550,7 +551,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, /* RFC793 page 36: "If the connection is in any non-synchronized state ... * and the incoming segment acknowledges something not yet - * sent (the segment carries an unaccaptable ACK) ... + * sent (the segment carries an unacceptable ACK) ... * a reset is sent." * * Invalid ACK: reset will be sent by listening socket diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index b907456a79f..029c70dfb58 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -436,6 +436,8 @@ int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, unsigned int mss u16 flags; BUG_ON(len > skb->len); + + clear_all_retrans_hints(tp); nsize = skb_headlen(skb) - len; if (nsize < 0) nsize = 0; @@ -599,7 +601,7 @@ int tcp_trim_head(struct sock *sk, struct sk_buff *skb, u32 len) for TCP options, but includes only bare TCP header. tp->rx_opt.mss_clamp is mss negotiated at connection setup. - It is minumum of user_mss and mss received with SYN. + It is minimum of user_mss and mss received with SYN. It also does not include TCP options. tp->pmtu_cookie is last pmtu, seen by this function. @@ -1171,7 +1173,7 @@ u32 __tcp_select_window(struct sock *sk) { struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); - /* MSS for the peer's data. Previous verions used mss_clamp + /* MSS for the peer's data. Previous versions used mss_clamp * here. I don't know if the value based on our guesses * of peer's MSS is better for the performance. It's more correct * but may be worse for the performance because of rcv_mss @@ -1260,7 +1262,10 @@ static void tcp_retrans_try_collapse(struct sock *sk, struct sk_buff *skb, int m BUG_ON(tcp_skb_pcount(skb) != 1 || tcp_skb_pcount(next_skb) != 1); - /* Ok. We will be able to collapse the packet. */ + /* changing transmit queue under us so clear hints */ + clear_all_retrans_hints(tp); + + /* Ok. We will be able to collapse the packet. */ __skb_unlink(next_skb, &sk->sk_write_queue); memcpy(skb_put(skb, next_skb_size), next_skb->data, next_skb_size); @@ -1330,6 +1335,8 @@ void tcp_simple_retransmit(struct sock *sk) } } + clear_all_retrans_hints(tp); + if (!lost) return; @@ -1361,7 +1368,7 @@ int tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb) int err; /* Do not sent more than we queued. 1/4 is reserved for possible - * copying overhead: frgagmentation, tunneling, mangling etc. + * copying overhead: fragmentation, tunneling, mangling etc. */ if (atomic_read(&sk->sk_wmem_alloc) > min(sk->sk_wmem_queued + (sk->sk_wmem_queued >> 2), sk->sk_sndbuf)) @@ -1468,13 +1475,25 @@ void tcp_xmit_retransmit_queue(struct sock *sk) const struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; - int packet_cnt = tp->lost_out; + int packet_cnt; + + if (tp->retransmit_skb_hint) { + skb = tp->retransmit_skb_hint; + packet_cnt = tp->retransmit_cnt_hint; + }else{ + skb = sk->sk_write_queue.next; + packet_cnt = 0; + } /* First pass: retransmit lost packets. */ - if (packet_cnt) { - sk_stream_for_retrans_queue(skb, sk) { + if (tp->lost_out) { + sk_stream_for_retrans_queue_from(skb, sk) { __u8 sacked = TCP_SKB_CB(skb)->sacked; + /* we could do better than to assign each time */ + tp->retransmit_skb_hint = skb; + tp->retransmit_cnt_hint = packet_cnt; + /* Assume this retransmit will generate * only one packet for congestion window * calculation purposes. This works because @@ -1485,10 +1504,12 @@ void tcp_xmit_retransmit_queue(struct sock *sk) if (tcp_packets_in_flight(tp) >= tp->snd_cwnd) return; - if (sacked&TCPCB_LOST) { + if (sacked & TCPCB_LOST) { if (!(sacked&(TCPCB_SACKED_ACKED|TCPCB_SACKED_RETRANS))) { - if (tcp_retransmit_skb(sk, skb)) + if (tcp_retransmit_skb(sk, skb)) { + tp->retransmit_skb_hint = NULL; return; + } if (icsk->icsk_ca_state != TCP_CA_Loss) NET_INC_STATS_BH(LINUX_MIB_TCPFASTRETRANS); else @@ -1501,8 +1522,8 @@ void tcp_xmit_retransmit_queue(struct sock *sk) TCP_RTO_MAX); } - packet_cnt -= tcp_skb_pcount(skb); - if (packet_cnt <= 0) + packet_cnt += tcp_skb_pcount(skb); + if (packet_cnt >= tp->lost_out) break; } } @@ -1528,9 +1549,18 @@ void tcp_xmit_retransmit_queue(struct sock *sk) if (tcp_may_send_now(sk, tp)) return; - packet_cnt = 0; + if (tp->forward_skb_hint) { + skb = tp->forward_skb_hint; + packet_cnt = tp->forward_cnt_hint; + } else{ + skb = sk->sk_write_queue.next; + packet_cnt = 0; + } + + sk_stream_for_retrans_queue_from(skb, sk) { + tp->forward_cnt_hint = packet_cnt; + tp->forward_skb_hint = skb; - sk_stream_for_retrans_queue(skb, sk) { /* Similar to the retransmit loop above we * can pretend that the retransmitted SKB * we send out here will be composed of one @@ -1547,8 +1577,10 @@ void tcp_xmit_retransmit_queue(struct sock *sk) continue; /* Ok, retransmit it. */ - if (tcp_retransmit_skb(sk, skb)) + if (tcp_retransmit_skb(sk, skb)) { + tp->forward_skb_hint = NULL; break; + } if (skb == skb_peek(&sk->sk_write_queue)) inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, @@ -2058,3 +2090,4 @@ EXPORT_SYMBOL(tcp_connect); EXPORT_SYMBOL(tcp_make_synack); EXPORT_SYMBOL(tcp_simple_retransmit); EXPORT_SYMBOL(tcp_sync_mss); +EXPORT_SYMBOL(sysctl_tcp_tso_win_divisor); diff --git a/net/ipv4/tcp_scalable.c b/net/ipv4/tcp_scalable.c index 327770bf552..26d7486ee50 100644 --- a/net/ipv4/tcp_scalable.c +++ b/net/ipv4/tcp_scalable.c @@ -20,20 +20,20 @@ static void tcp_scalable_cong_avoid(struct sock *sk, u32 ack, u32 rtt, u32 in_flight, int flag) { struct tcp_sock *tp = tcp_sk(sk); - if (in_flight < tp->snd_cwnd) + + if (!tcp_is_cwnd_limited(sk, in_flight)) return; - if (tp->snd_cwnd <= tp->snd_ssthresh) { - tp->snd_cwnd++; - } else { + if (tp->snd_cwnd <= tp->snd_ssthresh) + tcp_slow_start(tp); + else { tp->snd_cwnd_cnt++; if (tp->snd_cwnd_cnt > min(tp->snd_cwnd, TCP_SCALABLE_AI_CNT)){ - tp->snd_cwnd++; + if (tp->snd_cwnd < tp->snd_cwnd_clamp) + tp->snd_cwnd++; tp->snd_cwnd_cnt = 0; } } - tp->snd_cwnd = min_t(u32, tp->snd_cwnd, tp->snd_cwnd_clamp); - tp->snd_cwnd_stamp = tcp_time_stamp; } static u32 tcp_scalable_ssthresh(struct sock *sk) diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 415ee47ac1c..e1880959614 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -58,7 +58,7 @@ static void tcp_write_err(struct sock *sk) * to prevent DoS attacks. It is called when a retransmission timeout * or zero probe timeout occurs on orphaned socket. * - * Criterium is still not confirmed experimentally and may change. + * Criteria is still not confirmed experimentally and may change. * We kill the socket, if: * 1. If number of orphaned sockets exceeds an administratively configured * limit. @@ -132,7 +132,7 @@ static int tcp_write_timeout(struct sock *sk) hole detection. :-( It is place to make it. It is not made. I do not want - to make it. It is disguisting. It does not work in any + to make it. It is disgusting. It does not work in any case. Let me to cite the same draft, which requires for us to implement this: diff --git a/net/ipv4/tcp_vegas.c b/net/ipv4/tcp_vegas.c index 93c5f92070f..b7d296a8ac6 100644 --- a/net/ipv4/tcp_vegas.c +++ b/net/ipv4/tcp_vegas.c @@ -236,8 +236,7 @@ static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, /* We don't have enough RTT samples to do the Vegas * calculation, so we'll behave like Reno. */ - if (tp->snd_cwnd > tp->snd_ssthresh) - tp->snd_cwnd++; + tcp_reno_cong_avoid(sk, ack, seq_rtt, in_flight, flag); } else { u32 rtt, target_cwnd, diff; @@ -275,7 +274,7 @@ static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, */ diff = (old_wnd << V_PARAM_SHIFT) - target_cwnd; - if (tp->snd_cwnd < tp->snd_ssthresh) { + if (tp->snd_cwnd <= tp->snd_ssthresh) { /* Slow start. */ if (diff > gamma) { /* Going too fast. Time to slow down @@ -295,6 +294,7 @@ static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, V_PARAM_SHIFT)+1); } + tcp_slow_start(tp); } else { /* Congestion avoidance. */ u32 next_snd_cwnd; @@ -327,37 +327,17 @@ static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, else if (next_snd_cwnd < tp->snd_cwnd) tp->snd_cwnd--; } - } - /* Wipe the slate clean for the next RTT. */ - vegas->cntRTT = 0; - vegas->minRTT = 0x7fffffff; + if (tp->snd_cwnd < 2) + tp->snd_cwnd = 2; + else if (tp->snd_cwnd > tp->snd_cwnd_clamp) + tp->snd_cwnd = tp->snd_cwnd_clamp; + } } - /* The following code is executed for every ack we receive, - * except for conditions checked in should_advance_cwnd() - * before the call to tcp_cong_avoid(). Mainly this means that - * we only execute this code if the ack actually acked some - * data. - */ - - /* If we are in slow start, increase our cwnd in response to this ACK. - * (If we are not in slow start then we are in congestion avoidance, - * and adjust our congestion window only once per RTT. See the code - * above.) - */ - if (tp->snd_cwnd <= tp->snd_ssthresh) - tp->snd_cwnd++; - - /* to keep cwnd from growing without bound */ - tp->snd_cwnd = min_t(u32, tp->snd_cwnd, tp->snd_cwnd_clamp); - - /* Make sure that we are never so timid as to reduce our cwnd below - * 2 MSS. - * - * Going below 2 MSS would risk huge delayed ACKs from our receiver. - */ - tp->snd_cwnd = max(tp->snd_cwnd, 2U); + /* Wipe the slate clean for the next RTT. */ + vegas->cntRTT = 0; + vegas->minRTT = 0x7fffffff; } /* Extract info for Tcp socket info provided via netlink. */ diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index e0bd1013cb0..2422a5f7195 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -761,7 +761,7 @@ int udp_ioctl(struct sock *sk, int cmd, unsigned long arg) static __inline__ int __udp_checksum_complete(struct sk_buff *skb) { - return (unsigned short)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum)); + return __skb_checksum_complete(skb); } static __inline__ int udp_checksum_complete(struct sk_buff *skb) @@ -1100,11 +1100,8 @@ static int udp_checksum_init(struct sk_buff *skb, struct udphdr *uh, if (uh->check == 0) { skb->ip_summed = CHECKSUM_UNNECESSARY; } else if (skb->ip_summed == CHECKSUM_HW) { - skb->ip_summed = CHECKSUM_UNNECESSARY; if (!udp_check(uh, ulen, saddr, daddr, skb->csum)) - return 0; - LIMIT_NETDEBUG(KERN_DEBUG "udp v4 hw csum failure.\n"); - skb->ip_summed = CHECKSUM_NONE; + skb->ip_summed = CHECKSUM_UNNECESSARY; } if (skb->ip_summed != CHECKSUM_UNNECESSARY) skb->csum = csum_tcpudp_nofold(saddr, daddr, ulen, IPPROTO_UDP, 0); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index b7a5f51238b..ddcf7754eec 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1022,6 +1022,7 @@ int ipv6_dev_get_saddr(struct net_device *daddr_dev, continue; } +#ifdef CONFIG_IPV6_PRIVACY /* Rule 7: Prefer public address * Note: prefer temprary address if use_tempaddr >= 2 */ @@ -1042,7 +1043,7 @@ int ipv6_dev_get_saddr(struct net_device *daddr_dev, if (hiscore.attrs & IPV6_SADDR_SCORE_PRIVACY) continue; } - +#endif /* Rule 8: Use longest matching prefix */ if (hiscore.rule < 8) hiscore.matchlen = ipv6_addr_diff(&ifa_result->addr, daddr); diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 23e540365a1..1bdf0fb8bf8 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -585,17 +585,16 @@ static int icmpv6_rcv(struct sk_buff **pskb, unsigned int *nhoffp) daddr = &skb->nh.ipv6h->daddr; /* Perform checksum. */ - if (skb->ip_summed == CHECKSUM_HW) { - skb->ip_summed = CHECKSUM_UNNECESSARY; - if (csum_ipv6_magic(saddr, daddr, skb->len, IPPROTO_ICMPV6, - skb->csum)) { - LIMIT_NETDEBUG(KERN_DEBUG "ICMPv6 hw checksum failed\n"); - skb->ip_summed = CHECKSUM_NONE; - } - } - if (skb->ip_summed == CHECKSUM_NONE) { - if (csum_ipv6_magic(saddr, daddr, skb->len, IPPROTO_ICMPV6, - skb_checksum(skb, 0, skb->len, 0))) { + switch (skb->ip_summed) { + case CHECKSUM_HW: + if (!csum_ipv6_magic(saddr, daddr, skb->len, IPPROTO_ICMPV6, + skb->csum)) + break; + /* fall through */ + case CHECKSUM_NONE: + skb->csum = ~csum_ipv6_magic(saddr, daddr, skb->len, + IPPROTO_ICMPV6, 0); + if (__skb_checksum_complete(skb)) { LIMIT_NETDEBUG(KERN_DEBUG "ICMPv6 checksum failed [%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x > %04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x]\n", NIP6(*saddr), NIP6(*daddr)); goto discard_it; diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c index 6e348042693..a6026d2787d 100644 --- a/net/ipv6/ip6_input.c +++ b/net/ipv6/ip6_input.c @@ -176,6 +176,11 @@ resubmit: if (ipprot->flags & INET6_PROTO_FINAL) { struct ipv6hdr *hdr; + /* Free reference early: we don't need it any more, + and it may hold ip_conntrack module loaded + indefinitely. */ + nf_reset(skb); + skb_postpull_rcsum(skb, skb->nh.raw, skb->h.raw - skb->nh.raw); hdr = skb->nh.ipv6h; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index dbd9767b32e..c1fa693511a 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -441,9 +441,15 @@ static void ip6_copy_metadata(struct sk_buff *to, struct sk_buff *from) #ifdef CONFIG_NETFILTER to->nfmark = from->nfmark; /* Connection association is same as pre-frag packet */ + nf_conntrack_put(to->nfct); to->nfct = from->nfct; nf_conntrack_get(to->nfct); to->nfctinfo = from->nfctinfo; +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) + nf_conntrack_put_reasm(to->nfct_reasm); + to->nfct_reasm = from->nfct_reasm; + nf_conntrack_get_reasm(to->nfct_reasm); +#endif #ifdef CONFIG_BRIDGE_NETFILTER nf_bridge_put(to->nf_bridge); to->nf_bridge = from->nf_bridge; diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index e6b0e3954c0..e315d0f80af 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -525,6 +525,7 @@ ip6ip6_rcv(struct sk_buff **pskb, unsigned int *nhoffp) if ((t = ip6ip6_tnl_lookup(&ipv6h->saddr, &ipv6h->daddr)) != NULL) { if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) { + read_unlock(&ip6ip6_lock); kfree_skb(skb); return 0; } diff --git a/net/ipv6/netfilter/Kconfig b/net/ipv6/netfilter/Kconfig index bb7ccfe33f2..971ba60bf6e 100644 --- a/net/ipv6/netfilter/Kconfig +++ b/net/ipv6/netfilter/Kconfig @@ -278,5 +278,19 @@ config IP6_NF_RAW If you want to compile it as a module, say M here and read <file:Documentation/modules.txt>. If unsure, say `N'. +config NF_CONNTRACK_IPV6 + tristate "IPv6 support for new connection tracking (EXPERIMENTAL)" + depends on EXPERIMENTAL && NF_CONNTRACK + ---help--- + Connection tracking keeps a record of what packets have passed + through your machine, in order to figure out how they are related + into connections. + + This is IPv6 support on Layer 3 independent connection tracking. + Layer 3 independent connection tracking is experimental scheme + which generalize ip_conntrack to support other layer 3 protocols. + + To compile it as a module, choose M here. If unsure, say N. + endmenu diff --git a/net/ipv6/netfilter/Makefile b/net/ipv6/netfilter/Makefile index 2b2c370e8b1..9ab5b2ca1f5 100644 --- a/net/ipv6/netfilter/Makefile +++ b/net/ipv6/netfilter/Makefile @@ -27,3 +27,9 @@ obj-$(CONFIG_IP6_NF_TARGET_LOG) += ip6t_LOG.o obj-$(CONFIG_IP6_NF_RAW) += ip6table_raw.o obj-$(CONFIG_IP6_NF_MATCH_HL) += ip6t_hl.o obj-$(CONFIG_IP6_NF_TARGET_REJECT) += ip6t_REJECT.o + +# objects for l3 independent conntrack +nf_conntrack_ipv6-objs := nf_conntrack_l3proto_ipv6.o nf_conntrack_proto_icmpv6.o nf_conntrack_reasm.o + +# l3 independent conntrack +obj-$(CONFIG_NF_CONNTRACK_IPV6) += nf_conntrack_ipv6.o diff --git a/net/ipv6/netfilter/ip6t_MARK.c b/net/ipv6/netfilter/ip6t_MARK.c index 0c7584f9217..eab8fb864ee 100644 --- a/net/ipv6/netfilter/ip6t_MARK.c +++ b/net/ipv6/netfilter/ip6t_MARK.c @@ -56,9 +56,9 @@ checkentry(const char *tablename, return 1; } -static struct ip6t_target ip6t_mark_reg = { - .name = "MARK", - .target = target, +static struct ip6t_target ip6t_mark_reg = { + .name = "MARK", + .target = target, .checkentry = checkentry, .me = THIS_MODULE }; diff --git a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c new file mode 100644 index 00000000000..e2c90b3a807 --- /dev/null +++ b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c @@ -0,0 +1,556 @@ +/* + * Copyright (C)2004 USAGI/WIDE Project + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * Author: + * Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - support Layer 3 protocol independent connection tracking. + * Based on the original ip_conntrack code which had the following + * copyright information: + * (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org> + * + * 23 Mar 2004: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - add get_features() to support various size of conntrack + * structures. + */ + +#include <linux/config.h> +#include <linux/types.h> +#include <linux/ipv6.h> +#include <linux/in6.h> +#include <linux/netfilter.h> +#include <linux/module.h> +#include <linux/skbuff.h> +#include <linux/icmp.h> +#include <linux/sysctl.h> +#include <net/ipv6.h> + +#include <linux/netfilter_ipv6.h> +#include <net/netfilter/nf_conntrack.h> +#include <net/netfilter/nf_conntrack_helper.h> +#include <net/netfilter/nf_conntrack_protocol.h> +#include <net/netfilter/nf_conntrack_l3proto.h> +#include <net/netfilter/nf_conntrack_core.h> + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +DECLARE_PER_CPU(struct ip_conntrack_stat, nf_conntrack_stat); + +static int ipv6_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff, + struct nf_conntrack_tuple *tuple) +{ + u_int32_t _addrs[8], *ap; + + ap = skb_header_pointer(skb, nhoff + offsetof(struct ipv6hdr, saddr), + sizeof(_addrs), _addrs); + if (ap == NULL) + return 0; + + memcpy(tuple->src.u3.ip6, ap, sizeof(tuple->src.u3.ip6)); + memcpy(tuple->dst.u3.ip6, ap + 4, sizeof(tuple->dst.u3.ip6)); + + return 1; +} + +static int ipv6_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + memcpy(tuple->src.u3.ip6, orig->dst.u3.ip6, sizeof(tuple->src.u3.ip6)); + memcpy(tuple->dst.u3.ip6, orig->src.u3.ip6, sizeof(tuple->dst.u3.ip6)); + + return 1; +} + +static int ipv6_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return seq_printf(s, "src=%x:%x:%x:%x:%x:%x:%x:%x dst=%x:%x:%x:%x:%x:%x:%x:%x ", + NIP6(*((struct in6_addr *)tuple->src.u3.ip6)), + NIP6(*((struct in6_addr *)tuple->dst.u3.ip6))); +} + +static int ipv6_print_conntrack(struct seq_file *s, + const struct nf_conn *conntrack) +{ + return 0; +} + +/* + * Based on ipv6_skip_exthdr() in net/ipv6/exthdr.c + * + * This function parses (probably truncated) exthdr set "hdr" + * of length "len". "nexthdrp" initially points to some place, + * where type of the first header can be found. + * + * It skips all well-known exthdrs, and returns pointer to the start + * of unparsable area i.e. the first header with unknown type. + * if success, *nexthdr is updated by type/protocol of this header. + * + * NOTES: - it may return pointer pointing beyond end of packet, + * if the last recognized header is truncated in the middle. + * - if packet is truncated, so that all parsed headers are skipped, + * it returns -1. + * - if packet is fragmented, return pointer of the fragment header. + * - ESP is unparsable for now and considered like + * normal payload protocol. + * - Note also special handling of AUTH header. Thanks to IPsec wizards. + */ + +int nf_ct_ipv6_skip_exthdr(struct sk_buff *skb, int start, u8 *nexthdrp, + int len) +{ + u8 nexthdr = *nexthdrp; + + while (ipv6_ext_hdr(nexthdr)) { + struct ipv6_opt_hdr hdr; + int hdrlen; + + if (len < (int)sizeof(struct ipv6_opt_hdr)) + return -1; + if (nexthdr == NEXTHDR_NONE) + break; + if (nexthdr == NEXTHDR_FRAGMENT) + break; + if (skb_copy_bits(skb, start, &hdr, sizeof(hdr))) + BUG(); + if (nexthdr == NEXTHDR_AUTH) + hdrlen = (hdr.hdrlen+2)<<2; + else + hdrlen = ipv6_optlen(&hdr); + + nexthdr = hdr.nexthdr; + len -= hdrlen; + start += hdrlen; + } + + *nexthdrp = nexthdr; + return start; +} + +static int +ipv6_prepare(struct sk_buff **pskb, unsigned int hooknum, unsigned int *dataoff, + u_int8_t *protonum) +{ + unsigned int extoff; + unsigned char pnum; + int protoff; + + extoff = (u8*)((*pskb)->nh.ipv6h + 1) - (*pskb)->data; + pnum = (*pskb)->nh.ipv6h->nexthdr; + + protoff = nf_ct_ipv6_skip_exthdr(*pskb, extoff, &pnum, + (*pskb)->len - extoff); + + /* + * (protoff == (*pskb)->len) mean that the packet doesn't have no data + * except of IPv6 & ext headers. but it's tracked anyway. - YK + */ + if ((protoff < 0) || (protoff > (*pskb)->len)) { + DEBUGP("ip6_conntrack_core: can't find proto in pkt\n"); + NF_CT_STAT_INC(error); + NF_CT_STAT_INC(invalid); + return -NF_ACCEPT; + } + + *dataoff = protoff; + *protonum = pnum; + return NF_ACCEPT; +} + +static u_int32_t ipv6_get_features(const struct nf_conntrack_tuple *tuple) +{ + return NF_CT_F_BASIC; +} + +static unsigned int ipv6_confirm(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + + /* This is where we call the helper: as the packet goes out. */ + ct = nf_ct_get(*pskb, &ctinfo); + if (ct && ct->helper) { + unsigned int ret, protoff; + unsigned int extoff = (u8*)((*pskb)->nh.ipv6h + 1) + - (*pskb)->data; + unsigned char pnum = (*pskb)->nh.ipv6h->nexthdr; + + protoff = nf_ct_ipv6_skip_exthdr(*pskb, extoff, &pnum, + (*pskb)->len - extoff); + if (protoff < 0 || protoff > (*pskb)->len || + pnum == NEXTHDR_FRAGMENT) { + DEBUGP("proto header not found\n"); + return NF_ACCEPT; + } + + ret = ct->helper->help(pskb, protoff, ct, ctinfo); + if (ret != NF_ACCEPT) + return ret; + } + + /* We've seen it coming out the other side: confirm it */ + + return nf_conntrack_confirm(pskb); +} + +extern struct sk_buff *nf_ct_frag6_gather(struct sk_buff *skb); +extern void nf_ct_frag6_output(unsigned int hooknum, struct sk_buff *skb, + struct net_device *in, + struct net_device *out, + int (*okfn)(struct sk_buff *)); +static unsigned int ipv6_defrag(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct sk_buff *reasm; + + /* Previously seen (loopback)? */ + if ((*pskb)->nfct) + return NF_ACCEPT; + + reasm = nf_ct_frag6_gather(*pskb); + + /* queued */ + if (reasm == NULL) + return NF_STOLEN; + + /* error occured or not fragmented */ + if (reasm == *pskb) + return NF_ACCEPT; + + nf_ct_frag6_output(hooknum, reasm, (struct net_device *)in, + (struct net_device *)out, okfn); + + return NF_STOLEN; +} + +static unsigned int ipv6_conntrack_in(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct sk_buff *reasm = (*pskb)->nfct_reasm; + + /* This packet is fragmented and has reassembled packet. */ + if (reasm) { + /* Reassembled packet isn't parsed yet ? */ + if (!reasm->nfct) { + unsigned int ret; + + ret = nf_conntrack_in(PF_INET6, hooknum, &reasm); + if (ret != NF_ACCEPT) + return ret; + } + nf_conntrack_get(reasm->nfct); + (*pskb)->nfct = reasm->nfct; + return NF_ACCEPT; + } + + return nf_conntrack_in(PF_INET6, hooknum, pskb); +} + +static unsigned int ipv6_conntrack_local(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + /* root is playing with raw sockets. */ + if ((*pskb)->len < sizeof(struct ipv6hdr)) { + if (net_ratelimit()) + printk("ipv6_conntrack_local: packet too short\n"); + return NF_ACCEPT; + } + return ipv6_conntrack_in(hooknum, pskb, in, out, okfn); +} + +/* Connection tracking may drop packets, but never alters them, so + make it the first hook. */ +static struct nf_hook_ops ipv6_conntrack_defrag_ops = { + .hook = ipv6_defrag, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_IP6_PRE_ROUTING, + .priority = NF_IP6_PRI_CONNTRACK_DEFRAG, +}; + +static struct nf_hook_ops ipv6_conntrack_in_ops = { + .hook = ipv6_conntrack_in, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_IP6_PRE_ROUTING, + .priority = NF_IP6_PRI_CONNTRACK, +}; + +static struct nf_hook_ops ipv6_conntrack_local_out_ops = { + .hook = ipv6_conntrack_local, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_IP6_LOCAL_OUT, + .priority = NF_IP6_PRI_CONNTRACK, +}; + +static struct nf_hook_ops ipv6_conntrack_defrag_local_out_ops = { + .hook = ipv6_defrag, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_IP6_LOCAL_OUT, + .priority = NF_IP6_PRI_CONNTRACK_DEFRAG, +}; + +/* Refragmenter; last chance. */ +static struct nf_hook_ops ipv6_conntrack_out_ops = { + .hook = ipv6_confirm, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_IP6_POST_ROUTING, + .priority = NF_IP6_PRI_LAST, +}; + +static struct nf_hook_ops ipv6_conntrack_local_in_ops = { + .hook = ipv6_confirm, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_IP6_LOCAL_IN, + .priority = NF_IP6_PRI_LAST-1, +}; + +#ifdef CONFIG_SYSCTL + +/* From nf_conntrack_proto_icmpv6.c */ +extern unsigned long nf_ct_icmpv6_timeout; + +/* From nf_conntrack_frag6.c */ +extern unsigned long nf_ct_frag6_timeout; +extern unsigned long nf_ct_frag6_low_thresh; +extern unsigned long nf_ct_frag6_high_thresh; + +static struct ctl_table_header *nf_ct_ipv6_sysctl_header; + +static ctl_table nf_ct_sysctl_table[] = { + { + .ctl_name = NET_NF_CONNTRACK_ICMPV6_TIMEOUT, + .procname = "nf_conntrack_icmpv6_timeout", + .data = &nf_ct_icmpv6_timeout, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_FRAG6_TIMEOUT, + .procname = "nf_conntrack_frag6_timeout", + .data = &nf_ct_frag6_timeout, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_FRAG6_LOW_THRESH, + .procname = "nf_conntrack_frag6_low_thresh", + .data = &nf_ct_frag6_low_thresh, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_FRAG6_HIGH_THRESH, + .procname = "nf_conntrack_frag6_high_thresh", + .data = &nf_ct_frag6_high_thresh, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { .ctl_name = 0 } +}; + +static ctl_table nf_ct_netfilter_table[] = { + { + .ctl_name = NET_NETFILTER, + .procname = "netfilter", + .mode = 0555, + .child = nf_ct_sysctl_table, + }, + { .ctl_name = 0 } +}; + +static ctl_table nf_ct_net_table[] = { + { + .ctl_name = CTL_NET, + .procname = "net", + .mode = 0555, + .child = nf_ct_netfilter_table, + }, + { .ctl_name = 0 } +}; +#endif + +struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6 = { + .l3proto = PF_INET6, + .name = "ipv6", + .pkt_to_tuple = ipv6_pkt_to_tuple, + .invert_tuple = ipv6_invert_tuple, + .print_tuple = ipv6_print_tuple, + .print_conntrack = ipv6_print_conntrack, + .prepare = ipv6_prepare, + .get_features = ipv6_get_features, + .me = THIS_MODULE, +}; + +extern struct nf_conntrack_protocol nf_conntrack_protocol_tcp6; +extern struct nf_conntrack_protocol nf_conntrack_protocol_udp6; +extern struct nf_conntrack_protocol nf_conntrack_protocol_icmpv6; +extern int nf_ct_frag6_init(void); +extern void nf_ct_frag6_cleanup(void); +static int init_or_cleanup(int init) +{ + int ret = 0; + + if (!init) goto cleanup; + + ret = nf_ct_frag6_init(); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't initialize frag6.\n"); + goto cleanup_nothing; + } + ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_tcp6); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register tcp.\n"); + goto cleanup_frag6; + } + + ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_udp6); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register udp.\n"); + goto cleanup_tcp; + } + + ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_icmpv6); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register icmpv6.\n"); + goto cleanup_udp; + } + + ret = nf_conntrack_l3proto_register(&nf_conntrack_l3proto_ipv6); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register ipv6\n"); + goto cleanup_icmpv6; + } + + ret = nf_register_hook(&ipv6_conntrack_defrag_ops); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register pre-routing defrag " + "hook.\n"); + goto cleanup_ipv6; + } + + ret = nf_register_hook(&ipv6_conntrack_defrag_local_out_ops); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register local_out defrag " + "hook.\n"); + goto cleanup_defragops; + } + + ret = nf_register_hook(&ipv6_conntrack_in_ops); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register pre-routing hook.\n"); + goto cleanup_defraglocalops; + } + + ret = nf_register_hook(&ipv6_conntrack_local_out_ops); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register local out hook.\n"); + goto cleanup_inops; + } + + ret = nf_register_hook(&ipv6_conntrack_out_ops); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register post-routing hook.\n"); + goto cleanup_inandlocalops; + } + + ret = nf_register_hook(&ipv6_conntrack_local_in_ops); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register local in hook.\n"); + goto cleanup_inoutandlocalops; + } + +#ifdef CONFIG_SYSCTL + nf_ct_ipv6_sysctl_header = register_sysctl_table(nf_ct_net_table, 0); + if (nf_ct_ipv6_sysctl_header == NULL) { + printk("nf_conntrack: can't register to sysctl.\n"); + ret = -ENOMEM; + goto cleanup_localinops; + } +#endif + return ret; + + cleanup: + synchronize_net(); +#ifdef CONFIG_SYSCTL + unregister_sysctl_table(nf_ct_ipv6_sysctl_header); + cleanup_localinops: +#endif + nf_unregister_hook(&ipv6_conntrack_local_in_ops); + cleanup_inoutandlocalops: + nf_unregister_hook(&ipv6_conntrack_out_ops); + cleanup_inandlocalops: + nf_unregister_hook(&ipv6_conntrack_local_out_ops); + cleanup_inops: + nf_unregister_hook(&ipv6_conntrack_in_ops); + cleanup_defraglocalops: + nf_unregister_hook(&ipv6_conntrack_defrag_local_out_ops); + cleanup_defragops: + nf_unregister_hook(&ipv6_conntrack_defrag_ops); + cleanup_ipv6: + nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv6); + cleanup_icmpv6: + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_icmpv6); + cleanup_udp: + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_udp6); + cleanup_tcp: + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_tcp6); + cleanup_frag6: + nf_ct_frag6_cleanup(); + cleanup_nothing: + return ret; +} + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Yasuyuki KOZAKAI @USAGI <yasuyuki.kozakai@toshiba.co.jp>"); + +static int __init init(void) +{ + need_nf_conntrack(); + return init_or_cleanup(1); +} + +static void __exit fini(void) +{ + init_or_cleanup(0); +} + +module_init(init); +module_exit(fini); + +void need_ip6_conntrack(void) +{ +} + +EXPORT_SYMBOL(need_ip6_conntrack); diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c new file mode 100644 index 00000000000..c0f1da5497a --- /dev/null +++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c @@ -0,0 +1,272 @@ +/* + * Copyright (C)2003,2004 USAGI/WIDE Project + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * Author: + * Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - ICMPv6 tracking support. Derived from the original ip_conntrack code + * net/ipv4/netfilter/ip_conntrack_proto_icmp.c which had the following + * copyright information: + * (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org> + */ + +#include <linux/types.h> +#include <linux/sched.h> +#include <linux/timer.h> +#include <linux/module.h> +#include <linux/netfilter.h> +#include <linux/in6.h> +#include <linux/icmpv6.h> +#include <linux/ipv6.h> +#include <net/ipv6.h> +#include <net/ip6_checksum.h> +#include <linux/seq_file.h> +#include <linux/netfilter_ipv6.h> +#include <net/netfilter/nf_conntrack_tuple.h> +#include <net/netfilter/nf_conntrack_protocol.h> +#include <net/netfilter/nf_conntrack_core.h> +#include <net/netfilter/ipv6/nf_conntrack_icmpv6.h> + +unsigned long nf_ct_icmpv6_timeout = 30*HZ; + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +static int icmpv6_pkt_to_tuple(const struct sk_buff *skb, + unsigned int dataoff, + struct nf_conntrack_tuple *tuple) +{ + struct icmp6hdr _hdr, *hp; + + hp = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr); + if (hp == NULL) + return 0; + tuple->dst.u.icmp.type = hp->icmp6_type; + tuple->src.u.icmp.id = hp->icmp6_identifier; + tuple->dst.u.icmp.code = hp->icmp6_code; + + return 1; +} + +static int icmpv6_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + /* Add 1; spaces filled with 0. */ + static u_int8_t invmap[] = { + [ICMPV6_ECHO_REQUEST - 128] = ICMPV6_ECHO_REPLY + 1, + [ICMPV6_ECHO_REPLY - 128] = ICMPV6_ECHO_REQUEST + 1, + [ICMPV6_NI_QUERY - 128] = ICMPV6_NI_QUERY + 1, + [ICMPV6_NI_REPLY - 128] = ICMPV6_NI_REPLY +1 + }; + + __u8 type = orig->dst.u.icmp.type - 128; + if (type >= sizeof(invmap) || !invmap[type]) + return 0; + + tuple->src.u.icmp.id = orig->src.u.icmp.id; + tuple->dst.u.icmp.type = invmap[type] - 1; + tuple->dst.u.icmp.code = orig->dst.u.icmp.code; + return 1; +} + +/* Print out the per-protocol part of the tuple. */ +static int icmpv6_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return seq_printf(s, "type=%u code=%u id=%u ", + tuple->dst.u.icmp.type, + tuple->dst.u.icmp.code, + ntohs(tuple->src.u.icmp.id)); +} + +/* Print out the private part of the conntrack. */ +static int icmpv6_print_conntrack(struct seq_file *s, + const struct nf_conn *conntrack) +{ + return 0; +} + +/* Returns verdict for packet, or -1 for invalid. */ +static int icmpv6_packet(struct nf_conn *ct, + const struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info ctinfo, + int pf, + unsigned int hooknum) +{ + /* Try to delete connection immediately after all replies: + won't actually vanish as we still have skb, and del_timer + means this will only run once even if count hits zero twice + (theoretically possible with SMP) */ + if (CTINFO2DIR(ctinfo) == IP_CT_DIR_REPLY) { + if (atomic_dec_and_test(&ct->proto.icmp.count) + && del_timer(&ct->timeout)) + ct->timeout.function((unsigned long)ct); + } else { + atomic_inc(&ct->proto.icmp.count); + nf_conntrack_event_cache(IPCT_PROTOINFO_VOLATILE, skb); + nf_ct_refresh_acct(ct, ctinfo, skb, nf_ct_icmpv6_timeout); + } + + return NF_ACCEPT; +} + +/* Called when a new connection for this protocol found. */ +static int icmpv6_new(struct nf_conn *conntrack, + const struct sk_buff *skb, + unsigned int dataoff) +{ + static u_int8_t valid_new[] = { + [ICMPV6_ECHO_REQUEST - 128] = 1, + [ICMPV6_NI_QUERY - 128] = 1 + }; + + if (conntrack->tuplehash[0].tuple.dst.u.icmp.type - 128 >= sizeof(valid_new) + || !valid_new[conntrack->tuplehash[0].tuple.dst.u.icmp.type - 128]) { + /* Can't create a new ICMPv6 `conn' with this. */ + DEBUGP("icmp: can't create new conn with type %u\n", + conntrack->tuplehash[0].tuple.dst.u.icmp.type); + NF_CT_DUMP_TUPLE(&conntrack->tuplehash[0].tuple); + return 0; + } + atomic_set(&conntrack->proto.icmp.count, 0); + return 1; +} + +extern int +nf_ct_ipv6_skip_exthdr(struct sk_buff *skb, int start, u8 *nexthdrp, int len); +extern struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6; +static int +icmpv6_error_message(struct sk_buff *skb, + unsigned int icmp6off, + enum ip_conntrack_info *ctinfo, + unsigned int hooknum) +{ + struct nf_conntrack_tuple intuple, origtuple; + struct nf_conntrack_tuple_hash *h; + struct icmp6hdr _hdr, *hp; + unsigned int inip6off; + struct nf_conntrack_protocol *inproto; + u_int8_t inprotonum; + unsigned int inprotoff; + + NF_CT_ASSERT(skb->nfct == NULL); + + hp = skb_header_pointer(skb, icmp6off, sizeof(_hdr), &_hdr); + if (hp == NULL) { + DEBUGP("icmpv6_error: Can't get ICMPv6 hdr.\n"); + return -NF_ACCEPT; + } + + inip6off = icmp6off + sizeof(_hdr); + if (skb_copy_bits(skb, inip6off+offsetof(struct ipv6hdr, nexthdr), + &inprotonum, sizeof(inprotonum)) != 0) { + DEBUGP("icmpv6_error: Can't get nexthdr in inner IPv6 header.\n"); + return -NF_ACCEPT; + } + inprotoff = nf_ct_ipv6_skip_exthdr(skb, + inip6off + sizeof(struct ipv6hdr), + &inprotonum, + skb->len - inip6off + - sizeof(struct ipv6hdr)); + + if ((inprotoff < 0) || (inprotoff > skb->len) || + (inprotonum == NEXTHDR_FRAGMENT)) { + DEBUGP("icmpv6_error: Can't get protocol header in ICMPv6 payload.\n"); + return -NF_ACCEPT; + } + + inproto = nf_ct_find_proto(PF_INET6, inprotonum); + + /* Are they talking about one of our connections? */ + if (!nf_ct_get_tuple(skb, inip6off, inprotoff, PF_INET6, inprotonum, + &origtuple, &nf_conntrack_l3proto_ipv6, inproto)) { + DEBUGP("icmpv6_error: Can't get tuple\n"); + return -NF_ACCEPT; + } + + /* Ordinarily, we'd expect the inverted tupleproto, but it's + been preserved inside the ICMP. */ + if (!nf_ct_invert_tuple(&intuple, &origtuple, + &nf_conntrack_l3proto_ipv6, inproto)) { + DEBUGP("icmpv6_error: Can't invert tuple\n"); + return -NF_ACCEPT; + } + + *ctinfo = IP_CT_RELATED; + + h = nf_conntrack_find_get(&intuple, NULL); + if (!h) { + DEBUGP("icmpv6_error: no match\n"); + return -NF_ACCEPT; + } else { + if (NF_CT_DIRECTION(h) == IP_CT_DIR_REPLY) + *ctinfo += IP_CT_IS_REPLY; + } + + /* Update skb to refer to this connection */ + skb->nfct = &nf_ct_tuplehash_to_ctrack(h)->ct_general; + skb->nfctinfo = *ctinfo; + return -NF_ACCEPT; +} + +static int +icmpv6_error(struct sk_buff *skb, unsigned int dataoff, + enum ip_conntrack_info *ctinfo, int pf, unsigned int hooknum) +{ + struct icmp6hdr _ih, *icmp6h; + + icmp6h = skb_header_pointer(skb, dataoff, sizeof(_ih), &_ih); + if (icmp6h == NULL) { + if (LOG_INVALID(IPPROTO_ICMPV6)) + nf_log_packet(PF_INET6, 0, skb, NULL, NULL, NULL, + "nf_ct_icmpv6: short packet "); + return -NF_ACCEPT; + } + + if (hooknum != NF_IP6_PRE_ROUTING) + goto skipped; + + /* Ignore it if the checksum's bogus. */ + if (csum_ipv6_magic(&skb->nh.ipv6h->saddr, &skb->nh.ipv6h->daddr, + skb->len - dataoff, IPPROTO_ICMPV6, + skb_checksum(skb, dataoff, + skb->len - dataoff, 0))) { + nf_log_packet(PF_INET6, 0, skb, NULL, NULL, NULL, + "nf_ct_icmpv6: ICMPv6 checksum failed\n"); + return -NF_ACCEPT; + } + +skipped: + + /* is not error message ? */ + if (icmp6h->icmp6_type >= 128) + return NF_ACCEPT; + + return icmpv6_error_message(skb, dataoff, ctinfo, hooknum); +} + +struct nf_conntrack_protocol nf_conntrack_protocol_icmpv6 = +{ + .l3proto = PF_INET6, + .proto = IPPROTO_ICMPV6, + .name = "icmpv6", + .pkt_to_tuple = icmpv6_pkt_to_tuple, + .invert_tuple = icmpv6_invert_tuple, + .print_tuple = icmpv6_print_tuple, + .print_conntrack = icmpv6_print_conntrack, + .packet = icmpv6_packet, + .new = icmpv6_new, + .error = icmpv6_error, +}; + +EXPORT_SYMBOL(nf_conntrack_protocol_icmpv6); diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c new file mode 100644 index 00000000000..7640b9bb769 --- /dev/null +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -0,0 +1,885 @@ +/* + * IPv6 fragment reassembly for connection tracking + * + * Copyright (C)2004 USAGI/WIDE Project + * + * Author: + * Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * + * Based on: net/ipv6/reassembly.c + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include <linux/config.h> +#include <linux/errno.h> +#include <linux/types.h> +#include <linux/string.h> +#include <linux/socket.h> +#include <linux/sockios.h> +#include <linux/jiffies.h> +#include <linux/net.h> +#include <linux/list.h> +#include <linux/netdevice.h> +#include <linux/in6.h> +#include <linux/ipv6.h> +#include <linux/icmpv6.h> +#include <linux/random.h> +#include <linux/jhash.h> + +#include <net/sock.h> +#include <net/snmp.h> + +#include <net/ipv6.h> +#include <net/protocol.h> +#include <net/transp_v6.h> +#include <net/rawv6.h> +#include <net/ndisc.h> +#include <net/addrconf.h> +#include <linux/sysctl.h> +#include <linux/netfilter.h> +#include <linux/netfilter_ipv6.h> +#include <linux/kernel.h> +#include <linux/module.h> + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +#define NF_CT_FRAG6_HIGH_THRESH 262144 /* == 256*1024 */ +#define NF_CT_FRAG6_LOW_THRESH 196608 /* == 192*1024 */ +#define NF_CT_FRAG6_TIMEOUT IPV6_FRAG_TIMEOUT + +int nf_ct_frag6_high_thresh = 256*1024; +int nf_ct_frag6_low_thresh = 192*1024; +int nf_ct_frag6_timeout = IPV6_FRAG_TIMEOUT; + +struct nf_ct_frag6_skb_cb +{ + struct inet6_skb_parm h; + int offset; + struct sk_buff *orig; +}; + +#define NFCT_FRAG6_CB(skb) ((struct nf_ct_frag6_skb_cb*)((skb)->cb)) + +struct nf_ct_frag6_queue +{ + struct nf_ct_frag6_queue *next; + struct list_head lru_list; /* lru list member */ + + __u32 id; /* fragment id */ + struct in6_addr saddr; + struct in6_addr daddr; + + spinlock_t lock; + atomic_t refcnt; + struct timer_list timer; /* expire timer */ + struct sk_buff *fragments; + int len; + int meat; + struct timeval stamp; + unsigned int csum; + __u8 last_in; /* has first/last segment arrived? */ +#define COMPLETE 4 +#define FIRST_IN 2 +#define LAST_IN 1 + __u16 nhoffset; + struct nf_ct_frag6_queue **pprev; +}; + +/* Hash table. */ + +#define FRAG6Q_HASHSZ 64 + +static struct nf_ct_frag6_queue *nf_ct_frag6_hash[FRAG6Q_HASHSZ]; +static rwlock_t nf_ct_frag6_lock = RW_LOCK_UNLOCKED; +static u32 nf_ct_frag6_hash_rnd; +static LIST_HEAD(nf_ct_frag6_lru_list); +int nf_ct_frag6_nqueues = 0; + +static __inline__ void __fq_unlink(struct nf_ct_frag6_queue *fq) +{ + if (fq->next) + fq->next->pprev = fq->pprev; + *fq->pprev = fq->next; + list_del(&fq->lru_list); + nf_ct_frag6_nqueues--; +} + +static __inline__ void fq_unlink(struct nf_ct_frag6_queue *fq) +{ + write_lock(&nf_ct_frag6_lock); + __fq_unlink(fq); + write_unlock(&nf_ct_frag6_lock); +} + +static unsigned int ip6qhashfn(u32 id, struct in6_addr *saddr, + struct in6_addr *daddr) +{ + u32 a, b, c; + + a = saddr->s6_addr32[0]; + b = saddr->s6_addr32[1]; + c = saddr->s6_addr32[2]; + + a += JHASH_GOLDEN_RATIO; + b += JHASH_GOLDEN_RATIO; + c += nf_ct_frag6_hash_rnd; + __jhash_mix(a, b, c); + + a += saddr->s6_addr32[3]; + b += daddr->s6_addr32[0]; + c += daddr->s6_addr32[1]; + __jhash_mix(a, b, c); + + a += daddr->s6_addr32[2]; + b += daddr->s6_addr32[3]; + c += id; + __jhash_mix(a, b, c); + + return c & (FRAG6Q_HASHSZ - 1); +} + +static struct timer_list nf_ct_frag6_secret_timer; +int nf_ct_frag6_secret_interval = 10 * 60 * HZ; + +static void nf_ct_frag6_secret_rebuild(unsigned long dummy) +{ + unsigned long now = jiffies; + int i; + + write_lock(&nf_ct_frag6_lock); + get_random_bytes(&nf_ct_frag6_hash_rnd, sizeof(u32)); + for (i = 0; i < FRAG6Q_HASHSZ; i++) { + struct nf_ct_frag6_queue *q; + + q = nf_ct_frag6_hash[i]; + while (q) { + struct nf_ct_frag6_queue *next = q->next; + unsigned int hval = ip6qhashfn(q->id, + &q->saddr, + &q->daddr); + + if (hval != i) { + /* Unlink. */ + if (q->next) + q->next->pprev = q->pprev; + *q->pprev = q->next; + + /* Relink to new hash chain. */ + if ((q->next = nf_ct_frag6_hash[hval]) != NULL) + q->next->pprev = &q->next; + nf_ct_frag6_hash[hval] = q; + q->pprev = &nf_ct_frag6_hash[hval]; + } + + q = next; + } + } + write_unlock(&nf_ct_frag6_lock); + + mod_timer(&nf_ct_frag6_secret_timer, now + nf_ct_frag6_secret_interval); +} + +atomic_t nf_ct_frag6_mem = ATOMIC_INIT(0); + +/* Memory Tracking Functions. */ +static inline void frag_kfree_skb(struct sk_buff *skb) +{ + atomic_sub(skb->truesize, &nf_ct_frag6_mem); + if (NFCT_FRAG6_CB(skb)->orig) + kfree_skb(NFCT_FRAG6_CB(skb)->orig); + + kfree_skb(skb); +} + +static inline void frag_free_queue(struct nf_ct_frag6_queue *fq) +{ + atomic_sub(sizeof(struct nf_ct_frag6_queue), &nf_ct_frag6_mem); + kfree(fq); +} + +static inline struct nf_ct_frag6_queue *frag_alloc_queue(void) +{ + struct nf_ct_frag6_queue *fq = kmalloc(sizeof(struct nf_ct_frag6_queue), GFP_ATOMIC); + + if (!fq) + return NULL; + atomic_add(sizeof(struct nf_ct_frag6_queue), &nf_ct_frag6_mem); + return fq; +} + +/* Destruction primitives. */ + +/* Complete destruction of fq. */ +static void nf_ct_frag6_destroy(struct nf_ct_frag6_queue *fq) +{ + struct sk_buff *fp; + + BUG_TRAP(fq->last_in&COMPLETE); + BUG_TRAP(del_timer(&fq->timer) == 0); + + /* Release all fragment data. */ + fp = fq->fragments; + while (fp) { + struct sk_buff *xp = fp->next; + + frag_kfree_skb(fp); + fp = xp; + } + + frag_free_queue(fq); +} + +static __inline__ void fq_put(struct nf_ct_frag6_queue *fq) +{ + if (atomic_dec_and_test(&fq->refcnt)) + nf_ct_frag6_destroy(fq); +} + +/* Kill fq entry. It is not destroyed immediately, + * because caller (and someone more) holds reference count. + */ +static __inline__ void fq_kill(struct nf_ct_frag6_queue *fq) +{ + if (del_timer(&fq->timer)) + atomic_dec(&fq->refcnt); + + if (!(fq->last_in & COMPLETE)) { + fq_unlink(fq); + atomic_dec(&fq->refcnt); + fq->last_in |= COMPLETE; + } +} + +static void nf_ct_frag6_evictor(void) +{ + struct nf_ct_frag6_queue *fq; + struct list_head *tmp; + + for (;;) { + if (atomic_read(&nf_ct_frag6_mem) <= nf_ct_frag6_low_thresh) + return; + read_lock(&nf_ct_frag6_lock); + if (list_empty(&nf_ct_frag6_lru_list)) { + read_unlock(&nf_ct_frag6_lock); + return; + } + tmp = nf_ct_frag6_lru_list.next; + fq = list_entry(tmp, struct nf_ct_frag6_queue, lru_list); + atomic_inc(&fq->refcnt); + read_unlock(&nf_ct_frag6_lock); + + spin_lock(&fq->lock); + if (!(fq->last_in&COMPLETE)) + fq_kill(fq); + spin_unlock(&fq->lock); + + fq_put(fq); + } +} + +static void nf_ct_frag6_expire(unsigned long data) +{ + struct nf_ct_frag6_queue *fq = (struct nf_ct_frag6_queue *) data; + + spin_lock(&fq->lock); + + if (fq->last_in & COMPLETE) + goto out; + + fq_kill(fq); + +out: + spin_unlock(&fq->lock); + fq_put(fq); +} + +/* Creation primitives. */ + + +static struct nf_ct_frag6_queue *nf_ct_frag6_intern(unsigned int hash, + struct nf_ct_frag6_queue *fq_in) +{ + struct nf_ct_frag6_queue *fq; + + write_lock(&nf_ct_frag6_lock); +#ifdef CONFIG_SMP + for (fq = nf_ct_frag6_hash[hash]; fq; fq = fq->next) { + if (fq->id == fq_in->id && + !ipv6_addr_cmp(&fq_in->saddr, &fq->saddr) && + !ipv6_addr_cmp(&fq_in->daddr, &fq->daddr)) { + atomic_inc(&fq->refcnt); + write_unlock(&nf_ct_frag6_lock); + fq_in->last_in |= COMPLETE; + fq_put(fq_in); + return fq; + } + } +#endif + fq = fq_in; + + if (!mod_timer(&fq->timer, jiffies + nf_ct_frag6_timeout)) + atomic_inc(&fq->refcnt); + + atomic_inc(&fq->refcnt); + if ((fq->next = nf_ct_frag6_hash[hash]) != NULL) + fq->next->pprev = &fq->next; + nf_ct_frag6_hash[hash] = fq; + fq->pprev = &nf_ct_frag6_hash[hash]; + INIT_LIST_HEAD(&fq->lru_list); + list_add_tail(&fq->lru_list, &nf_ct_frag6_lru_list); + nf_ct_frag6_nqueues++; + write_unlock(&nf_ct_frag6_lock); + return fq; +} + + +static struct nf_ct_frag6_queue * +nf_ct_frag6_create(unsigned int hash, u32 id, struct in6_addr *src, struct in6_addr *dst) +{ + struct nf_ct_frag6_queue *fq; + + if ((fq = frag_alloc_queue()) == NULL) { + DEBUGP("Can't alloc new queue\n"); + goto oom; + } + + memset(fq, 0, sizeof(struct nf_ct_frag6_queue)); + + fq->id = id; + ipv6_addr_copy(&fq->saddr, src); + ipv6_addr_copy(&fq->daddr, dst); + + init_timer(&fq->timer); + fq->timer.function = nf_ct_frag6_expire; + fq->timer.data = (long) fq; + fq->lock = SPIN_LOCK_UNLOCKED; + atomic_set(&fq->refcnt, 1); + + return nf_ct_frag6_intern(hash, fq); + +oom: + return NULL; +} + +static __inline__ struct nf_ct_frag6_queue * +fq_find(u32 id, struct in6_addr *src, struct in6_addr *dst) +{ + struct nf_ct_frag6_queue *fq; + unsigned int hash = ip6qhashfn(id, src, dst); + + read_lock(&nf_ct_frag6_lock); + for (fq = nf_ct_frag6_hash[hash]; fq; fq = fq->next) { + if (fq->id == id && + !ipv6_addr_cmp(src, &fq->saddr) && + !ipv6_addr_cmp(dst, &fq->daddr)) { + atomic_inc(&fq->refcnt); + read_unlock(&nf_ct_frag6_lock); + return fq; + } + } + read_unlock(&nf_ct_frag6_lock); + + return nf_ct_frag6_create(hash, id, src, dst); +} + + +static int nf_ct_frag6_queue(struct nf_ct_frag6_queue *fq, struct sk_buff *skb, + struct frag_hdr *fhdr, int nhoff) +{ + struct sk_buff *prev, *next; + int offset, end; + + if (fq->last_in & COMPLETE) { + DEBUGP("Allready completed\n"); + goto err; + } + + offset = ntohs(fhdr->frag_off) & ~0x7; + end = offset + (ntohs(skb->nh.ipv6h->payload_len) - + ((u8 *) (fhdr + 1) - (u8 *) (skb->nh.ipv6h + 1))); + + if ((unsigned int)end > IPV6_MAXPLEN) { + DEBUGP("offset is too large.\n"); + return -1; + } + + if (skb->ip_summed == CHECKSUM_HW) + skb->csum = csum_sub(skb->csum, + csum_partial(skb->nh.raw, + (u8*)(fhdr + 1) - skb->nh.raw, + 0)); + + /* Is this the final fragment? */ + if (!(fhdr->frag_off & htons(IP6_MF))) { + /* If we already have some bits beyond end + * or have different end, the segment is corrupted. + */ + if (end < fq->len || + ((fq->last_in & LAST_IN) && end != fq->len)) { + DEBUGP("already received last fragment\n"); + goto err; + } + fq->last_in |= LAST_IN; + fq->len = end; + } else { + /* Check if the fragment is rounded to 8 bytes. + * Required by the RFC. + */ + if (end & 0x7) { + /* RFC2460 says always send parameter problem in + * this case. -DaveM + */ + DEBUGP("the end of this fragment is not rounded to 8 bytes.\n"); + return -1; + } + if (end > fq->len) { + /* Some bits beyond end -> corruption. */ + if (fq->last_in & LAST_IN) { + DEBUGP("last packet already reached.\n"); + goto err; + } + fq->len = end; + } + } + + if (end == offset) + goto err; + + /* Point into the IP datagram 'data' part. */ + if (!pskb_pull(skb, (u8 *) (fhdr + 1) - skb->data)) { + DEBUGP("queue: message is too short.\n"); + goto err; + } + if (end-offset < skb->len) { + if (pskb_trim(skb, end - offset)) { + DEBUGP("Can't trim\n"); + goto err; + } + if (skb->ip_summed != CHECKSUM_UNNECESSARY) + skb->ip_summed = CHECKSUM_NONE; + } + + /* Find out which fragments are in front and at the back of us + * in the chain of fragments so far. We must know where to put + * this fragment, right? + */ + prev = NULL; + for (next = fq->fragments; next != NULL; next = next->next) { + if (NFCT_FRAG6_CB(next)->offset >= offset) + break; /* bingo! */ + prev = next; + } + + /* We found where to put this one. Check for overlap with + * preceding fragment, and, if needed, align things so that + * any overlaps are eliminated. + */ + if (prev) { + int i = (NFCT_FRAG6_CB(prev)->offset + prev->len) - offset; + + if (i > 0) { + offset += i; + if (end <= offset) { + DEBUGP("overlap\n"); + goto err; + } + if (!pskb_pull(skb, i)) { + DEBUGP("Can't pull\n"); + goto err; + } + if (skb->ip_summed != CHECKSUM_UNNECESSARY) + skb->ip_summed = CHECKSUM_NONE; + } + } + + /* Look for overlap with succeeding segments. + * If we can merge fragments, do it. + */ + while (next && NFCT_FRAG6_CB(next)->offset < end) { + /* overlap is 'i' bytes */ + int i = end - NFCT_FRAG6_CB(next)->offset; + + if (i < next->len) { + /* Eat head of the next overlapped fragment + * and leave the loop. The next ones cannot overlap. + */ + DEBUGP("Eat head of the overlapped parts.: %d", i); + if (!pskb_pull(next, i)) + goto err; + + /* next fragment */ + NFCT_FRAG6_CB(next)->offset += i; + fq->meat -= i; + if (next->ip_summed != CHECKSUM_UNNECESSARY) + next->ip_summed = CHECKSUM_NONE; + break; + } else { + struct sk_buff *free_it = next; + + /* Old fragmnet is completely overridden with + * new one drop it. + */ + next = next->next; + + if (prev) + prev->next = next; + else + fq->fragments = next; + + fq->meat -= free_it->len; + frag_kfree_skb(free_it); + } + } + + NFCT_FRAG6_CB(skb)->offset = offset; + + /* Insert this fragment in the chain of fragments. */ + skb->next = next; + if (prev) + prev->next = skb; + else + fq->fragments = skb; + + skb->dev = NULL; + skb_get_timestamp(skb, &fq->stamp); + fq->meat += skb->len; + atomic_add(skb->truesize, &nf_ct_frag6_mem); + + /* The first fragment. + * nhoffset is obtained from the first fragment, of course. + */ + if (offset == 0) { + fq->nhoffset = nhoff; + fq->last_in |= FIRST_IN; + } + write_lock(&nf_ct_frag6_lock); + list_move_tail(&fq->lru_list, &nf_ct_frag6_lru_list); + write_unlock(&nf_ct_frag6_lock); + return 0; + +err: + return -1; +} + +/* + * Check if this packet is complete. + * Returns NULL on failure by any reason, and pointer + * to current nexthdr field in reassembled frame. + * + * It is called with locked fq, and caller must check that + * queue is eligible for reassembly i.e. it is not COMPLETE, + * the last and the first frames arrived and all the bits are here. + */ +static struct sk_buff * +nf_ct_frag6_reasm(struct nf_ct_frag6_queue *fq, struct net_device *dev) +{ + struct sk_buff *fp, *op, *head = fq->fragments; + int payload_len; + + fq_kill(fq); + + BUG_TRAP(head != NULL); + BUG_TRAP(NFCT_FRAG6_CB(head)->offset == 0); + + /* Unfragmented part is taken from the first segment. */ + payload_len = (head->data - head->nh.raw) - sizeof(struct ipv6hdr) + fq->len - sizeof(struct frag_hdr); + if (payload_len > IPV6_MAXPLEN) { + DEBUGP("payload len is too large.\n"); + goto out_oversize; + } + + /* Head of list must not be cloned. */ + if (skb_cloned(head) && pskb_expand_head(head, 0, 0, GFP_ATOMIC)) { + DEBUGP("skb is cloned but can't expand head"); + goto out_oom; + } + + /* If the first fragment is fragmented itself, we split + * it to two chunks: the first with data and paged part + * and the second, holding only fragments. */ + if (skb_shinfo(head)->frag_list) { + struct sk_buff *clone; + int i, plen = 0; + + if ((clone = alloc_skb(0, GFP_ATOMIC)) == NULL) { + DEBUGP("Can't alloc skb\n"); + goto out_oom; + } + clone->next = head->next; + head->next = clone; + skb_shinfo(clone)->frag_list = skb_shinfo(head)->frag_list; + skb_shinfo(head)->frag_list = NULL; + for (i=0; i<skb_shinfo(head)->nr_frags; i++) + plen += skb_shinfo(head)->frags[i].size; + clone->len = clone->data_len = head->data_len - plen; + head->data_len -= clone->len; + head->len -= clone->len; + clone->csum = 0; + clone->ip_summed = head->ip_summed; + + NFCT_FRAG6_CB(clone)->orig = NULL; + atomic_add(clone->truesize, &nf_ct_frag6_mem); + } + + /* We have to remove fragment header from datagram and to relocate + * header in order to calculate ICV correctly. */ + head->nh.raw[fq->nhoffset] = head->h.raw[0]; + memmove(head->head + sizeof(struct frag_hdr), head->head, + (head->data - head->head) - sizeof(struct frag_hdr)); + head->mac.raw += sizeof(struct frag_hdr); + head->nh.raw += sizeof(struct frag_hdr); + + skb_shinfo(head)->frag_list = head->next; + head->h.raw = head->data; + skb_push(head, head->data - head->nh.raw); + atomic_sub(head->truesize, &nf_ct_frag6_mem); + + for (fp=head->next; fp; fp = fp->next) { + head->data_len += fp->len; + head->len += fp->len; + if (head->ip_summed != fp->ip_summed) + head->ip_summed = CHECKSUM_NONE; + else if (head->ip_summed == CHECKSUM_HW) + head->csum = csum_add(head->csum, fp->csum); + head->truesize += fp->truesize; + atomic_sub(fp->truesize, &nf_ct_frag6_mem); + } + + head->next = NULL; + head->dev = dev; + skb_set_timestamp(head, &fq->stamp); + head->nh.ipv6h->payload_len = htons(payload_len); + + /* Yes, and fold redundant checksum back. 8) */ + if (head->ip_summed == CHECKSUM_HW) + head->csum = csum_partial(head->nh.raw, head->h.raw-head->nh.raw, head->csum); + + fq->fragments = NULL; + + /* all original skbs are linked into the NFCT_FRAG6_CB(head).orig */ + fp = skb_shinfo(head)->frag_list; + if (NFCT_FRAG6_CB(fp)->orig == NULL) + /* at above code, head skb is divided into two skbs. */ + fp = fp->next; + + op = NFCT_FRAG6_CB(head)->orig; + for (; fp; fp = fp->next) { + struct sk_buff *orig = NFCT_FRAG6_CB(fp)->orig; + + op->next = orig; + op = orig; + NFCT_FRAG6_CB(fp)->orig = NULL; + } + + return head; + +out_oversize: + if (net_ratelimit()) + printk(KERN_DEBUG "nf_ct_frag6_reasm: payload len = %d\n", payload_len); + goto out_fail; +out_oom: + if (net_ratelimit()) + printk(KERN_DEBUG "nf_ct_frag6_reasm: no memory for reassembly\n"); +out_fail: + return NULL; +} + +/* + * find the header just before Fragment Header. + * + * if success return 0 and set ... + * (*prevhdrp): the value of "Next Header Field" in the header + * just before Fragment Header. + * (*prevhoff): the offset of "Next Header Field" in the header + * just before Fragment Header. + * (*fhoff) : the offset of Fragment Header. + * + * Based on ipv6_skip_hdr() in net/ipv6/exthdr.c + * + */ +static int +find_prev_fhdr(struct sk_buff *skb, u8 *prevhdrp, int *prevhoff, int *fhoff) +{ + u8 nexthdr = skb->nh.ipv6h->nexthdr; + u8 prev_nhoff = (u8 *)&skb->nh.ipv6h->nexthdr - skb->data; + int start = (u8 *)(skb->nh.ipv6h+1) - skb->data; + int len = skb->len - start; + u8 prevhdr = NEXTHDR_IPV6; + + while (nexthdr != NEXTHDR_FRAGMENT) { + struct ipv6_opt_hdr hdr; + int hdrlen; + + if (!ipv6_ext_hdr(nexthdr)) { + return -1; + } + if (len < (int)sizeof(struct ipv6_opt_hdr)) { + DEBUGP("too short\n"); + return -1; + } + if (nexthdr == NEXTHDR_NONE) { + DEBUGP("next header is none\n"); + return -1; + } + if (skb_copy_bits(skb, start, &hdr, sizeof(hdr))) + BUG(); + if (nexthdr == NEXTHDR_AUTH) + hdrlen = (hdr.hdrlen+2)<<2; + else + hdrlen = ipv6_optlen(&hdr); + + prevhdr = nexthdr; + prev_nhoff = start; + + nexthdr = hdr.nexthdr; + len -= hdrlen; + start += hdrlen; + } + + if (len < 0) + return -1; + + *prevhdrp = prevhdr; + *prevhoff = prev_nhoff; + *fhoff = start; + + return 0; +} + +struct sk_buff *nf_ct_frag6_gather(struct sk_buff *skb) +{ + struct sk_buff *clone; + struct net_device *dev = skb->dev; + struct frag_hdr *fhdr; + struct nf_ct_frag6_queue *fq; + struct ipv6hdr *hdr; + int fhoff, nhoff; + u8 prevhdr; + struct sk_buff *ret_skb = NULL; + + /* Jumbo payload inhibits frag. header */ + if (skb->nh.ipv6h->payload_len == 0) { + DEBUGP("payload len = 0\n"); + return skb; + } + + if (find_prev_fhdr(skb, &prevhdr, &nhoff, &fhoff) < 0) + return skb; + + clone = skb_clone(skb, GFP_ATOMIC); + if (clone == NULL) { + DEBUGP("Can't clone skb\n"); + return skb; + } + + NFCT_FRAG6_CB(clone)->orig = skb; + + if (!pskb_may_pull(clone, fhoff + sizeof(*fhdr))) { + DEBUGP("message is too short.\n"); + goto ret_orig; + } + + clone->h.raw = clone->data + fhoff; + hdr = clone->nh.ipv6h; + fhdr = (struct frag_hdr *)clone->h.raw; + + if (!(fhdr->frag_off & htons(0xFFF9))) { + DEBUGP("Invalid fragment offset\n"); + /* It is not a fragmented frame */ + goto ret_orig; + } + + if (atomic_read(&nf_ct_frag6_mem) > nf_ct_frag6_high_thresh) + nf_ct_frag6_evictor(); + + fq = fq_find(fhdr->identification, &hdr->saddr, &hdr->daddr); + if (fq == NULL) { + DEBUGP("Can't find and can't create new queue\n"); + goto ret_orig; + } + + spin_lock(&fq->lock); + + if (nf_ct_frag6_queue(fq, clone, fhdr, nhoff) < 0) { + spin_unlock(&fq->lock); + DEBUGP("Can't insert skb to queue\n"); + fq_put(fq); + goto ret_orig; + } + + if (fq->last_in == (FIRST_IN|LAST_IN) && fq->meat == fq->len) { + ret_skb = nf_ct_frag6_reasm(fq, dev); + if (ret_skb == NULL) + DEBUGP("Can't reassemble fragmented packets\n"); + } + spin_unlock(&fq->lock); + + fq_put(fq); + return ret_skb; + +ret_orig: + kfree_skb(clone); + return skb; +} + +void nf_ct_frag6_output(unsigned int hooknum, struct sk_buff *skb, + struct net_device *in, struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct sk_buff *s, *s2; + + for (s = NFCT_FRAG6_CB(skb)->orig; s;) { + nf_conntrack_put_reasm(s->nfct_reasm); + nf_conntrack_get_reasm(skb); + s->nfct_reasm = skb; + + s2 = s->next; + NF_HOOK_THRESH(PF_INET6, hooknum, s, in, out, okfn, + NF_IP6_PRI_CONNTRACK_DEFRAG + 1); + s = s2; + } + nf_conntrack_put_reasm(skb); +} + +int nf_ct_frag6_kfree_frags(struct sk_buff *skb) +{ + struct sk_buff *s, *s2; + + for (s = NFCT_FRAG6_CB(skb)->orig; s; s = s2) { + + s2 = s->next; + kfree_skb(s); + } + + kfree_skb(skb); + + return 0; +} + +int nf_ct_frag6_init(void) +{ + nf_ct_frag6_hash_rnd = (u32) ((num_physpages ^ (num_physpages>>7)) ^ + (jiffies ^ (jiffies >> 6))); + + init_timer(&nf_ct_frag6_secret_timer); + nf_ct_frag6_secret_timer.function = nf_ct_frag6_secret_rebuild; + nf_ct_frag6_secret_timer.expires = jiffies + + nf_ct_frag6_secret_interval; + add_timer(&nf_ct_frag6_secret_timer); + + return 0; +} + +void nf_ct_frag6_cleanup(void) +{ + del_timer(&nf_ct_frag6_secret_timer); + nf_ct_frag6_evictor(); +} diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index a1265a320b1..8e9628f1c4c 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -174,8 +174,10 @@ int ipv6_raw_deliver(struct sk_buff *skb, int nexthdr) struct sk_buff *clone = skb_clone(skb, GFP_ATOMIC); /* Not releasing hash table! */ - if (clone) + if (clone) { + nf_reset(clone); rawv6_rcv(sk, clone); + } } sk = __raw_v6_lookup(sk_next(sk), nexthdr, daddr, saddr, IP6CB(skb)->iif); @@ -296,13 +298,10 @@ void rawv6_err(struct sock *sk, struct sk_buff *skb, static inline int rawv6_rcv_skb(struct sock * sk, struct sk_buff * skb) { if ((raw6_sk(sk)->checksum || sk->sk_filter) && - skb->ip_summed != CHECKSUM_UNNECESSARY) { - if ((unsigned short)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum))) { - /* FIXME: increment a raw6 drops counter here */ - kfree_skb(skb); - return 0; - } - skb->ip_summed = CHECKSUM_UNNECESSARY; + skb_checksum_complete(skb)) { + /* FIXME: increment a raw6 drops counter here */ + kfree_skb(skb); + return 0; } /* Charge it to the socket. */ @@ -335,32 +334,25 @@ int rawv6_rcv(struct sock *sk, struct sk_buff *skb) if (!rp->checksum) skb->ip_summed = CHECKSUM_UNNECESSARY; - if (skb->ip_summed != CHECKSUM_UNNECESSARY) { - if (skb->ip_summed == CHECKSUM_HW) { - skb_postpull_rcsum(skb, skb->nh.raw, - skb->h.raw - skb->nh.raw); + if (skb->ip_summed == CHECKSUM_HW) { + skb_postpull_rcsum(skb, skb->nh.raw, + skb->h.raw - skb->nh.raw); + if (!csum_ipv6_magic(&skb->nh.ipv6h->saddr, + &skb->nh.ipv6h->daddr, + skb->len, inet->num, skb->csum)) skb->ip_summed = CHECKSUM_UNNECESSARY; - if (csum_ipv6_magic(&skb->nh.ipv6h->saddr, - &skb->nh.ipv6h->daddr, - skb->len, inet->num, skb->csum)) { - LIMIT_NETDEBUG(KERN_DEBUG "raw v6 hw csum failure.\n"); - skb->ip_summed = CHECKSUM_NONE; - } - } - if (skb->ip_summed == CHECKSUM_NONE) - skb->csum = ~csum_ipv6_magic(&skb->nh.ipv6h->saddr, - &skb->nh.ipv6h->daddr, - skb->len, inet->num, 0); } + if (skb->ip_summed != CHECKSUM_UNNECESSARY) + skb->csum = ~csum_ipv6_magic(&skb->nh.ipv6h->saddr, + &skb->nh.ipv6h->daddr, + skb->len, inet->num, 0); if (inet->hdrincl) { - if (skb->ip_summed != CHECKSUM_UNNECESSARY && - (unsigned short)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum))) { + if (skb_checksum_complete(skb)) { /* FIXME: increment a raw6 drops counter here */ kfree_skb(skb); return 0; } - skb->ip_summed = CHECKSUM_UNNECESSARY; } rawv6_rcv_skb(sk, skb); @@ -405,7 +397,7 @@ static int rawv6_recvmsg(struct kiocb *iocb, struct sock *sk, if (skb->ip_summed==CHECKSUM_UNNECESSARY) { err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied); } else if (msg->msg_flags&MSG_TRUNC) { - if ((unsigned short)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum))) + if (__skb_checksum_complete(skb)) goto csum_copy_err; err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied); } else { diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 227e99ed510..f7f42c3e96c 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1710,7 +1710,7 @@ static void fib6_dump_end(struct netlink_callback *cb) static int fib6_dump_done(struct netlink_callback *cb) { fib6_dump_end(cb); - return cb->done(cb); + return cb->done ? cb->done(cb) : 0; } int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb) diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index d746d3b27ef..62c0e5bd931 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1401,20 +1401,18 @@ out: static int tcp_v6_checksum_init(struct sk_buff *skb) { if (skb->ip_summed == CHECKSUM_HW) { - skb->ip_summed = CHECKSUM_UNNECESSARY; if (!tcp_v6_check(skb->h.th,skb->len,&skb->nh.ipv6h->saddr, - &skb->nh.ipv6h->daddr,skb->csum)) + &skb->nh.ipv6h->daddr,skb->csum)) { + skb->ip_summed = CHECKSUM_UNNECESSARY; return 0; - LIMIT_NETDEBUG(KERN_DEBUG "hw tcp v6 csum failed\n"); + } } + + skb->csum = ~tcp_v6_check(skb->h.th,skb->len,&skb->nh.ipv6h->saddr, + &skb->nh.ipv6h->daddr, 0); + if (skb->len <= 76) { - if (tcp_v6_check(skb->h.th,skb->len,&skb->nh.ipv6h->saddr, - &skb->nh.ipv6h->daddr,skb_checksum(skb, 0, skb->len, 0))) - return -1; - skb->ip_summed = CHECKSUM_UNNECESSARY; - } else { - skb->csum = ~tcp_v6_check(skb->h.th,skb->len,&skb->nh.ipv6h->saddr, - &skb->nh.ipv6h->daddr,0); + return __skb_checksum_complete(skb); } return 0; } @@ -1575,7 +1573,7 @@ static int tcp_v6_rcv(struct sk_buff **pskb, unsigned int *nhoffp) goto discard_it; if ((skb->ip_summed != CHECKSUM_UNNECESSARY && - tcp_v6_checksum_init(skb) < 0)) + tcp_v6_checksum_init(skb))) goto bad_packet; th = skb->h.th; diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index bf9519341fd..e671153b47b 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -248,7 +248,7 @@ try_again: err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov, copied); } else if (msg->msg_flags&MSG_TRUNC) { - if ((unsigned short)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum))) + if (__skb_checksum_complete(skb)) goto csum_copy_err; err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov, copied); @@ -363,13 +363,10 @@ static inline int udpv6_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) return -1; } - if (skb->ip_summed != CHECKSUM_UNNECESSARY) { - if ((unsigned short)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum))) { - UDP6_INC_STATS_BH(UDP_MIB_INERRORS); - kfree_skb(skb); - return 0; - } - skb->ip_summed = CHECKSUM_UNNECESSARY; + if (skb_checksum_complete(skb)) { + UDP6_INC_STATS_BH(UDP_MIB_INERRORS); + kfree_skb(skb); + return 0; } if (sock_queue_rcv_skb(sk,skb)<0) { @@ -491,13 +488,10 @@ static int udpv6_rcv(struct sk_buff **pskb, unsigned int *nhoffp) uh = skb->h.uh; } - if (skb->ip_summed==CHECKSUM_HW) { + if (skb->ip_summed == CHECKSUM_HW && + !csum_ipv6_magic(saddr, daddr, ulen, IPPROTO_UDP, skb->csum)) skb->ip_summed = CHECKSUM_UNNECESSARY; - if (csum_ipv6_magic(saddr, daddr, ulen, IPPROTO_UDP, skb->csum)) { - LIMIT_NETDEBUG(KERN_DEBUG "udp v6 hw csum failure.\n"); - skb->ip_summed = CHECKSUM_NONE; - } - } + if (skb->ip_summed != CHECKSUM_UNNECESSARY) skb->csum = ~csum_ipv6_magic(saddr, daddr, ulen, IPPROTO_UDP, 0); @@ -521,8 +515,7 @@ static int udpv6_rcv(struct sk_buff **pskb, unsigned int *nhoffp) if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) goto discard; - if (skb->ip_summed != CHECKSUM_UNNECESSARY && - (unsigned short)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum))) + if (skb_checksum_complete(skb)) goto discard; UDP6_INC_STATS_BH(UDP_MIB_NOPORTS); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 8296b38bf27..a84f9221e5f 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -1,3 +1,6 @@ +menu "Core Netfilter Configuration" + depends on NET && NETFILTER + config NETFILTER_NETLINK tristate "Netfilter netlink interface" help @@ -22,3 +25,74 @@ config NETFILTER_NETLINK_LOG and is also scheduled to replace the old syslog-based ipt_LOG and ip6t_LOG modules. +config NF_CONNTRACK + tristate "Layer 3 Independent Connection tracking (EXPERIMENTAL)" + depends on EXPERIMENTAL && IP_NF_CONNTRACK=n + default n + ---help--- + Connection tracking keeps a record of what packets have passed + through your machine, in order to figure out how they are related + into connections. + + Layer 3 independent connection tracking is experimental scheme + which generalize ip_conntrack to support other layer 3 protocols. + + To compile it as a module, choose M here. If unsure, say N. + +config NF_CT_ACCT + bool "Connection tracking flow accounting" + depends on NF_CONNTRACK + help + If this option is enabled, the connection tracking code will + keep per-flow packet and byte counters. + + Those counters can be used for flow-based accounting or the + `connbytes' match. + + If unsure, say `N'. + +config NF_CONNTRACK_MARK + bool 'Connection mark tracking support' + depends on NF_CONNTRACK + help + This option enables support for connection marks, used by the + `CONNMARK' target and `connmark' match. Similar to the mark value + of packets, but this mark value is kept in the conntrack session + instead of the individual packets. + +config NF_CONNTRACK_EVENTS + bool "Connection tracking events" + depends on NF_CONNTRACK + help + If this option is enabled, the connection tracking code will + provide a notifier chain that can be used by other kernel code + to get notified aboutchanges in the connection tracking state. + + If unsure, say `N'. + +config NF_CT_PROTO_SCTP + tristate 'SCTP protocol on new connection tracking support (EXPERIMENTAL)' + depends on EXPERIMENTAL && NF_CONNTRACK + default n + help + With this option enabled, the layer 3 independent connection + tracking code will be able to do state tracking on SCTP connections. + + If you want to compile it as a module, say M here and read + Documentation/modules.txt. If unsure, say `N'. + +config NF_CONNTRACK_FTP + tristate "FTP support on new connection tracking (EXPERIMENTAL)" + depends on EXPERIMENTAL && NF_CONNTRACK + help + Tracking FTP connections is problematic: special helpers are + required for tracking them, and doing masquerading and other forms + of Network Address Translation on them. + + This is FTP support on Layer 3 independent connection tracking. + Layer 3 independent connection tracking is experimental scheme + which generalize ip_conntrack to support other layer 3 protocols. + + To compile it as a module, choose M here. If unsure, say N. + +endmenu diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index b3b44f8b415..55f019ad2c0 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -5,3 +5,11 @@ obj-$(CONFIG_NETFILTER) = netfilter.o obj-$(CONFIG_NETFILTER_NETLINK) += nfnetlink.o obj-$(CONFIG_NETFILTER_NETLINK_QUEUE) += nfnetlink_queue.o obj-$(CONFIG_NETFILTER_NETLINK_LOG) += nfnetlink_log.o + +nf_conntrack-objs := nf_conntrack_core.o nf_conntrack_standalone.o nf_conntrack_l3proto_generic.o nf_conntrack_proto_generic.o nf_conntrack_proto_tcp.o nf_conntrack_proto_udp.o + +obj-$(CONFIG_NF_CONNTRACK) += nf_conntrack.o +obj-$(CONFIG_NF_CONNTRACK_FTP) += nf_conntrack_ftp.o + +# SCTP protocol connection tracking +obj-$(CONFIG_NF_CT_PROTO_SCTP) += nf_conntrack_proto_sctp.o diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c new file mode 100644 index 00000000000..9a67c796b38 --- /dev/null +++ b/net/netfilter/nf_conntrack_core.c @@ -0,0 +1,1538 @@ +/* Connection state tracking for netfilter. This is separated from, + but required by, the NAT layer; it can also be used by an iptables + extension. */ + +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2005 Netfilter Core Team <coreteam@netfilter.org> + * (C) 2003,2004 USAGI/WIDE Project <http://www.linux-ipv6.org> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * 23 Apr 2001: Harald Welte <laforge@gnumonks.org> + * - new API and handling of conntrack/nat helpers + * - now capable of multiple expectations for one master + * 16 Jul 2002: Harald Welte <laforge@gnumonks.org> + * - add usage/reference counts to ip_conntrack_expect + * - export ip_conntrack[_expect]_{find_get,put} functions + * 16 Dec 2003: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - generalize L3 protocol denendent part. + * 23 Mar 2004: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - add support various size of conntrack structures. + * + * Derived from net/ipv4/netfilter/ip_conntrack_core.c + */ + +#include <linux/config.h> +#include <linux/types.h> +#include <linux/netfilter.h> +#include <linux/module.h> +#include <linux/skbuff.h> +#include <linux/proc_fs.h> +#include <linux/vmalloc.h> +#include <linux/stddef.h> +#include <linux/slab.h> +#include <linux/random.h> +#include <linux/jhash.h> +#include <linux/err.h> +#include <linux/percpu.h> +#include <linux/moduleparam.h> +#include <linux/notifier.h> +#include <linux/kernel.h> +#include <linux/netdevice.h> +#include <linux/socket.h> + +/* This rwlock protects the main hash table, protocol/helper/expected + registrations, conntrack timers*/ +#define ASSERT_READ_LOCK(x) +#define ASSERT_WRITE_LOCK(x) + +#include <net/netfilter/nf_conntrack.h> +#include <net/netfilter/nf_conntrack_l3proto.h> +#include <net/netfilter/nf_conntrack_protocol.h> +#include <net/netfilter/nf_conntrack_helper.h> +#include <net/netfilter/nf_conntrack_core.h> +#include <linux/netfilter_ipv4/listhelp.h> + +#define NF_CONNTRACK_VERSION "0.4.1" + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +DEFINE_RWLOCK(nf_conntrack_lock); + +/* nf_conntrack_standalone needs this */ +atomic_t nf_conntrack_count = ATOMIC_INIT(0); + +void (*nf_conntrack_destroyed)(struct nf_conn *conntrack) = NULL; +LIST_HEAD(nf_conntrack_expect_list); +struct nf_conntrack_protocol **nf_ct_protos[PF_MAX]; +struct nf_conntrack_l3proto *nf_ct_l3protos[PF_MAX]; +static LIST_HEAD(helpers); +unsigned int nf_conntrack_htable_size = 0; +int nf_conntrack_max; +struct list_head *nf_conntrack_hash; +static kmem_cache_t *nf_conntrack_expect_cachep; +struct nf_conn nf_conntrack_untracked; +unsigned int nf_ct_log_invalid; +static LIST_HEAD(unconfirmed); +static int nf_conntrack_vmalloc; + +#ifdef CONFIG_NF_CONNTRACK_EVENTS +struct notifier_block *nf_conntrack_chain; +struct notifier_block *nf_conntrack_expect_chain; + +DEFINE_PER_CPU(struct nf_conntrack_ecache, nf_conntrack_ecache); + +/* deliver cached events and clear cache entry - must be called with locally + * disabled softirqs */ +static inline void +__nf_ct_deliver_cached_events(struct nf_conntrack_ecache *ecache) +{ + DEBUGP("ecache: delivering events for %p\n", ecache->ct); + if (nf_ct_is_confirmed(ecache->ct) && !nf_ct_is_dying(ecache->ct) + && ecache->events) + notifier_call_chain(&nf_conntrack_chain, ecache->events, + ecache->ct); + + ecache->events = 0; + nf_ct_put(ecache->ct); + ecache->ct = NULL; +} + +/* Deliver all cached events for a particular conntrack. This is called + * by code prior to async packet handling for freeing the skb */ +void nf_ct_deliver_cached_events(const struct nf_conn *ct) +{ + struct nf_conntrack_ecache *ecache; + + local_bh_disable(); + ecache = &__get_cpu_var(nf_conntrack_ecache); + if (ecache->ct == ct) + __nf_ct_deliver_cached_events(ecache); + local_bh_enable(); +} + +/* Deliver cached events for old pending events, if current conntrack != old */ +void __nf_ct_event_cache_init(struct nf_conn *ct) +{ + struct nf_conntrack_ecache *ecache; + + /* take care of delivering potentially old events */ + ecache = &__get_cpu_var(nf_conntrack_ecache); + BUG_ON(ecache->ct == ct); + if (ecache->ct) + __nf_ct_deliver_cached_events(ecache); + /* initialize for this conntrack/packet */ + ecache->ct = ct; + nf_conntrack_get(&ct->ct_general); +} + +/* flush the event cache - touches other CPU's data and must not be called + * while packets are still passing through the code */ +static void nf_ct_event_cache_flush(void) +{ + struct nf_conntrack_ecache *ecache; + int cpu; + + for_each_cpu(cpu) { + ecache = &per_cpu(nf_conntrack_ecache, cpu); + if (ecache->ct) + nf_ct_put(ecache->ct); + } +} +#else +static inline void nf_ct_event_cache_flush(void) {} +#endif /* CONFIG_NF_CONNTRACK_EVENTS */ + +DEFINE_PER_CPU(struct ip_conntrack_stat, nf_conntrack_stat); +EXPORT_PER_CPU_SYMBOL(nf_conntrack_stat); + +/* + * This scheme offers various size of "struct nf_conn" dependent on + * features(helper, nat, ...) + */ + +#define NF_CT_FEATURES_NAMELEN 256 +static struct { + /* name of slab cache. printed in /proc/slabinfo */ + char *name; + + /* size of slab cache */ + size_t size; + + /* slab cache pointer */ + kmem_cache_t *cachep; + + /* allocated slab cache + modules which uses this slab cache */ + int use; + + /* Initialization */ + int (*init_conntrack)(struct nf_conn *, u_int32_t); + +} nf_ct_cache[NF_CT_F_NUM]; + +/* protect members of nf_ct_cache except of "use" */ +DEFINE_RWLOCK(nf_ct_cache_lock); + +/* This avoids calling kmem_cache_create() with same name simultaneously */ +DECLARE_MUTEX(nf_ct_cache_mutex); + +extern struct nf_conntrack_protocol nf_conntrack_generic_protocol; +struct nf_conntrack_protocol * +nf_ct_find_proto(u_int16_t l3proto, u_int8_t protocol) +{ + if (unlikely(nf_ct_protos[l3proto] == NULL)) + return &nf_conntrack_generic_protocol; + + return nf_ct_protos[l3proto][protocol]; +} + +static int nf_conntrack_hash_rnd_initted; +static unsigned int nf_conntrack_hash_rnd; + +static u_int32_t __hash_conntrack(const struct nf_conntrack_tuple *tuple, + unsigned int size, unsigned int rnd) +{ + unsigned int a, b; + a = jhash((void *)tuple->src.u3.all, sizeof(tuple->src.u3.all), + ((tuple->src.l3num) << 16) | tuple->dst.protonum); + b = jhash((void *)tuple->dst.u3.all, sizeof(tuple->dst.u3.all), + (tuple->src.u.all << 16) | tuple->dst.u.all); + + return jhash_2words(a, b, rnd) % size; +} + +static inline u_int32_t hash_conntrack(const struct nf_conntrack_tuple *tuple) +{ + return __hash_conntrack(tuple, nf_conntrack_htable_size, + nf_conntrack_hash_rnd); +} + +/* Initialize "struct nf_conn" which has spaces for helper */ +static int +init_conntrack_for_helper(struct nf_conn *conntrack, u_int32_t features) +{ + + conntrack->help = (union nf_conntrack_help *) + (((unsigned long)conntrack->data + + (__alignof__(union nf_conntrack_help) - 1)) + & (~((unsigned long)(__alignof__(union nf_conntrack_help) -1)))); + return 0; +} + +int nf_conntrack_register_cache(u_int32_t features, const char *name, + size_t size, + int (*init)(struct nf_conn *, u_int32_t)) +{ + int ret = 0; + char *cache_name; + kmem_cache_t *cachep; + + DEBUGP("nf_conntrack_register_cache: features=0x%x, name=%s, size=%d\n", + features, name, size); + + if (features < NF_CT_F_BASIC || features >= NF_CT_F_NUM) { + DEBUGP("nf_conntrack_register_cache: invalid features.: 0x%x\n", + features); + return -EINVAL; + } + + down(&nf_ct_cache_mutex); + + write_lock_bh(&nf_ct_cache_lock); + /* e.g: multiple helpers are loaded */ + if (nf_ct_cache[features].use > 0) { + DEBUGP("nf_conntrack_register_cache: already resisterd.\n"); + if ((!strncmp(nf_ct_cache[features].name, name, + NF_CT_FEATURES_NAMELEN)) + && nf_ct_cache[features].size == size + && nf_ct_cache[features].init_conntrack == init) { + DEBUGP("nf_conntrack_register_cache: reusing.\n"); + nf_ct_cache[features].use++; + ret = 0; + } else + ret = -EBUSY; + + write_unlock_bh(&nf_ct_cache_lock); + up(&nf_ct_cache_mutex); + return ret; + } + write_unlock_bh(&nf_ct_cache_lock); + + /* + * The memory space for name of slab cache must be alive until + * cache is destroyed. + */ + cache_name = kmalloc(sizeof(char)*NF_CT_FEATURES_NAMELEN, GFP_ATOMIC); + if (cache_name == NULL) { + DEBUGP("nf_conntrack_register_cache: can't alloc cache_name\n"); + ret = -ENOMEM; + goto out_up_mutex; + } + + if (strlcpy(cache_name, name, NF_CT_FEATURES_NAMELEN) + >= NF_CT_FEATURES_NAMELEN) { + printk("nf_conntrack_register_cache: name too long\n"); + ret = -EINVAL; + goto out_free_name; + } + + cachep = kmem_cache_create(cache_name, size, 0, 0, + NULL, NULL); + if (!cachep) { + printk("nf_conntrack_register_cache: Can't create slab cache " + "for the features = 0x%x\n", features); + ret = -ENOMEM; + goto out_free_name; + } + + write_lock_bh(&nf_ct_cache_lock); + nf_ct_cache[features].use = 1; + nf_ct_cache[features].size = size; + nf_ct_cache[features].init_conntrack = init; + nf_ct_cache[features].cachep = cachep; + nf_ct_cache[features].name = cache_name; + write_unlock_bh(&nf_ct_cache_lock); + + goto out_up_mutex; + +out_free_name: + kfree(cache_name); +out_up_mutex: + up(&nf_ct_cache_mutex); + return ret; +} + +/* FIXME: In the current, only nf_conntrack_cleanup() can call this function. */ +void nf_conntrack_unregister_cache(u_int32_t features) +{ + kmem_cache_t *cachep; + char *name; + + /* + * This assures that kmem_cache_create() isn't called before destroying + * slab cache. + */ + DEBUGP("nf_conntrack_unregister_cache: 0x%04x\n", features); + down(&nf_ct_cache_mutex); + + write_lock_bh(&nf_ct_cache_lock); + if (--nf_ct_cache[features].use > 0) { + write_unlock_bh(&nf_ct_cache_lock); + up(&nf_ct_cache_mutex); + return; + } + cachep = nf_ct_cache[features].cachep; + name = nf_ct_cache[features].name; + nf_ct_cache[features].cachep = NULL; + nf_ct_cache[features].name = NULL; + nf_ct_cache[features].init_conntrack = NULL; + nf_ct_cache[features].size = 0; + write_unlock_bh(&nf_ct_cache_lock); + + synchronize_net(); + + kmem_cache_destroy(cachep); + kfree(name); + + up(&nf_ct_cache_mutex); +} + +int +nf_ct_get_tuple(const struct sk_buff *skb, + unsigned int nhoff, + unsigned int dataoff, + u_int16_t l3num, + u_int8_t protonum, + struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_l3proto *l3proto, + const struct nf_conntrack_protocol *protocol) +{ + NF_CT_TUPLE_U_BLANK(tuple); + + tuple->src.l3num = l3num; + if (l3proto->pkt_to_tuple(skb, nhoff, tuple) == 0) + return 0; + + tuple->dst.protonum = protonum; + tuple->dst.dir = IP_CT_DIR_ORIGINAL; + + return protocol->pkt_to_tuple(skb, dataoff, tuple); +} + +int +nf_ct_invert_tuple(struct nf_conntrack_tuple *inverse, + const struct nf_conntrack_tuple *orig, + const struct nf_conntrack_l3proto *l3proto, + const struct nf_conntrack_protocol *protocol) +{ + NF_CT_TUPLE_U_BLANK(inverse); + + inverse->src.l3num = orig->src.l3num; + if (l3proto->invert_tuple(inverse, orig) == 0) + return 0; + + inverse->dst.dir = !orig->dst.dir; + + inverse->dst.protonum = orig->dst.protonum; + return protocol->invert_tuple(inverse, orig); +} + +/* nf_conntrack_expect helper functions */ +static void nf_ct_unlink_expect(struct nf_conntrack_expect *exp) +{ + ASSERT_WRITE_LOCK(&nf_conntrack_lock); + NF_CT_ASSERT(!timer_pending(&exp_timeout)); + list_del(&exp->list); + NF_CT_STAT_INC(expect_delete); + exp->master->expecting--; + nf_conntrack_expect_put(exp); +} + +static void expectation_timed_out(unsigned long ul_expect) +{ + struct nf_conntrack_expect *exp = (void *)ul_expect; + + write_lock_bh(&nf_conntrack_lock); + nf_ct_unlink_expect(exp); + write_unlock_bh(&nf_conntrack_lock); + nf_conntrack_expect_put(exp); +} + +/* If an expectation for this connection is found, it gets delete from + * global list then returned. */ +static struct nf_conntrack_expect * +find_expectation(const struct nf_conntrack_tuple *tuple) +{ + struct nf_conntrack_expect *i; + + list_for_each_entry(i, &nf_conntrack_expect_list, list) { + /* If master is not in hash table yet (ie. packet hasn't left + this machine yet), how can other end know about expected? + Hence these are not the droids you are looking for (if + master ct never got confirmed, we'd hold a reference to it + and weird things would happen to future packets). */ + if (nf_ct_tuple_mask_cmp(tuple, &i->tuple, &i->mask) + && nf_ct_is_confirmed(i->master)) { + if (i->flags & NF_CT_EXPECT_PERMANENT) { + atomic_inc(&i->use); + return i; + } else if (del_timer(&i->timeout)) { + nf_ct_unlink_expect(i); + return i; + } + } + } + return NULL; +} + +/* delete all expectations for this conntrack */ +static void remove_expectations(struct nf_conn *ct) +{ + struct nf_conntrack_expect *i, *tmp; + + /* Optimization: most connection never expect any others. */ + if (ct->expecting == 0) + return; + + list_for_each_entry_safe(i, tmp, &nf_conntrack_expect_list, list) { + if (i->master == ct && del_timer(&i->timeout)) { + nf_ct_unlink_expect(i); + nf_conntrack_expect_put(i); + } + } +} + +static void +clean_from_lists(struct nf_conn *ct) +{ + unsigned int ho, hr; + + DEBUGP("clean_from_lists(%p)\n", ct); + ASSERT_WRITE_LOCK(&nf_conntrack_lock); + + ho = hash_conntrack(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + hr = hash_conntrack(&ct->tuplehash[IP_CT_DIR_REPLY].tuple); + LIST_DELETE(&nf_conntrack_hash[ho], &ct->tuplehash[IP_CT_DIR_ORIGINAL]); + LIST_DELETE(&nf_conntrack_hash[hr], &ct->tuplehash[IP_CT_DIR_REPLY]); + + /* Destroy all pending expectations */ + remove_expectations(ct); +} + +static void +destroy_conntrack(struct nf_conntrack *nfct) +{ + struct nf_conn *ct = (struct nf_conn *)nfct; + struct nf_conntrack_l3proto *l3proto; + struct nf_conntrack_protocol *proto; + + DEBUGP("destroy_conntrack(%p)\n", ct); + NF_CT_ASSERT(atomic_read(&nfct->use) == 0); + NF_CT_ASSERT(!timer_pending(&ct->timeout)); + + nf_conntrack_event(IPCT_DESTROY, ct); + set_bit(IPS_DYING_BIT, &ct->status); + + /* To make sure we don't get any weird locking issues here: + * destroy_conntrack() MUST NOT be called with a write lock + * to nf_conntrack_lock!!! -HW */ + l3proto = nf_ct_find_l3proto(ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.l3num); + if (l3proto && l3proto->destroy) + l3proto->destroy(ct); + + proto = nf_ct_find_proto(ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.l3num, + ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.protonum); + if (proto && proto->destroy) + proto->destroy(ct); + + if (nf_conntrack_destroyed) + nf_conntrack_destroyed(ct); + + write_lock_bh(&nf_conntrack_lock); + /* Expectations will have been removed in clean_from_lists, + * except TFTP can create an expectation on the first packet, + * before connection is in the list, so we need to clean here, + * too. */ + remove_expectations(ct); + + /* We overload first tuple to link into unconfirmed list. */ + if (!nf_ct_is_confirmed(ct)) { + BUG_ON(list_empty(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list)); + list_del(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list); + } + + NF_CT_STAT_INC(delete); + write_unlock_bh(&nf_conntrack_lock); + + if (ct->master) + nf_ct_put(ct->master); + + DEBUGP("destroy_conntrack: returning ct=%p to slab\n", ct); + nf_conntrack_free(ct); +} + +static void death_by_timeout(unsigned long ul_conntrack) +{ + struct nf_conn *ct = (void *)ul_conntrack; + + write_lock_bh(&nf_conntrack_lock); + /* Inside lock so preempt is disabled on module removal path. + * Otherwise we can get spurious warnings. */ + NF_CT_STAT_INC(delete_list); + clean_from_lists(ct); + write_unlock_bh(&nf_conntrack_lock); + nf_ct_put(ct); +} + +static inline int +conntrack_tuple_cmp(const struct nf_conntrack_tuple_hash *i, + const struct nf_conntrack_tuple *tuple, + const struct nf_conn *ignored_conntrack) +{ + ASSERT_READ_LOCK(&nf_conntrack_lock); + return nf_ct_tuplehash_to_ctrack(i) != ignored_conntrack + && nf_ct_tuple_equal(tuple, &i->tuple); +} + +static struct nf_conntrack_tuple_hash * +__nf_conntrack_find(const struct nf_conntrack_tuple *tuple, + const struct nf_conn *ignored_conntrack) +{ + struct nf_conntrack_tuple_hash *h; + unsigned int hash = hash_conntrack(tuple); + + ASSERT_READ_LOCK(&nf_conntrack_lock); + list_for_each_entry(h, &nf_conntrack_hash[hash], list) { + if (conntrack_tuple_cmp(h, tuple, ignored_conntrack)) { + NF_CT_STAT_INC(found); + return h; + } + NF_CT_STAT_INC(searched); + } + + return NULL; +} + +/* Find a connection corresponding to a tuple. */ +struct nf_conntrack_tuple_hash * +nf_conntrack_find_get(const struct nf_conntrack_tuple *tuple, + const struct nf_conn *ignored_conntrack) +{ + struct nf_conntrack_tuple_hash *h; + + read_lock_bh(&nf_conntrack_lock); + h = __nf_conntrack_find(tuple, ignored_conntrack); + if (h) + atomic_inc(&nf_ct_tuplehash_to_ctrack(h)->ct_general.use); + read_unlock_bh(&nf_conntrack_lock); + + return h; +} + +/* Confirm a connection given skb; places it in hash table */ +int +__nf_conntrack_confirm(struct sk_buff **pskb) +{ + unsigned int hash, repl_hash; + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + + ct = nf_ct_get(*pskb, &ctinfo); + + /* ipt_REJECT uses nf_conntrack_attach to attach related + ICMP/TCP RST packets in other direction. Actual packet + which created connection will be IP_CT_NEW or for an + expected connection, IP_CT_RELATED. */ + if (CTINFO2DIR(ctinfo) != IP_CT_DIR_ORIGINAL) + return NF_ACCEPT; + + hash = hash_conntrack(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + repl_hash = hash_conntrack(&ct->tuplehash[IP_CT_DIR_REPLY].tuple); + + /* We're not in hash table, and we refuse to set up related + connections for unconfirmed conns. But packet copies and + REJECT will give spurious warnings here. */ + /* NF_CT_ASSERT(atomic_read(&ct->ct_general.use) == 1); */ + + /* No external references means noone else could have + confirmed us. */ + NF_CT_ASSERT(!nf_ct_is_confirmed(ct)); + DEBUGP("Confirming conntrack %p\n", ct); + + write_lock_bh(&nf_conntrack_lock); + + /* See if there's one in the list already, including reverse: + NAT could have grabbed it without realizing, since we're + not in the hash. If there is, we lost race. */ + if (!LIST_FIND(&nf_conntrack_hash[hash], + conntrack_tuple_cmp, + struct nf_conntrack_tuple_hash *, + &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple, NULL) + && !LIST_FIND(&nf_conntrack_hash[repl_hash], + conntrack_tuple_cmp, + struct nf_conntrack_tuple_hash *, + &ct->tuplehash[IP_CT_DIR_REPLY].tuple, NULL)) { + /* Remove from unconfirmed list */ + list_del(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list); + + list_prepend(&nf_conntrack_hash[hash], + &ct->tuplehash[IP_CT_DIR_ORIGINAL]); + list_prepend(&nf_conntrack_hash[repl_hash], + &ct->tuplehash[IP_CT_DIR_REPLY]); + /* Timer relative to confirmation time, not original + setting time, otherwise we'd get timer wrap in + weird delay cases. */ + ct->timeout.expires += jiffies; + add_timer(&ct->timeout); + atomic_inc(&ct->ct_general.use); + set_bit(IPS_CONFIRMED_BIT, &ct->status); + NF_CT_STAT_INC(insert); + write_unlock_bh(&nf_conntrack_lock); + if (ct->helper) + nf_conntrack_event_cache(IPCT_HELPER, *pskb); +#ifdef CONFIG_NF_NAT_NEEDED + if (test_bit(IPS_SRC_NAT_DONE_BIT, &ct->status) || + test_bit(IPS_DST_NAT_DONE_BIT, &ct->status)) + nf_conntrack_event_cache(IPCT_NATINFO, *pskb); +#endif + nf_conntrack_event_cache(master_ct(ct) ? + IPCT_RELATED : IPCT_NEW, *pskb); + return NF_ACCEPT; + } + + NF_CT_STAT_INC(insert_failed); + write_unlock_bh(&nf_conntrack_lock); + return NF_DROP; +} + +/* Returns true if a connection correspondings to the tuple (required + for NAT). */ +int +nf_conntrack_tuple_taken(const struct nf_conntrack_tuple *tuple, + const struct nf_conn *ignored_conntrack) +{ + struct nf_conntrack_tuple_hash *h; + + read_lock_bh(&nf_conntrack_lock); + h = __nf_conntrack_find(tuple, ignored_conntrack); + read_unlock_bh(&nf_conntrack_lock); + + return h != NULL; +} + +/* There's a small race here where we may free a just-assured + connection. Too bad: we're in trouble anyway. */ +static inline int unreplied(const struct nf_conntrack_tuple_hash *i) +{ + return !(test_bit(IPS_ASSURED_BIT, + &nf_ct_tuplehash_to_ctrack(i)->status)); +} + +static int early_drop(struct list_head *chain) +{ + /* Traverse backwards: gives us oldest, which is roughly LRU */ + struct nf_conntrack_tuple_hash *h; + struct nf_conn *ct = NULL; + int dropped = 0; + + read_lock_bh(&nf_conntrack_lock); + h = LIST_FIND_B(chain, unreplied, struct nf_conntrack_tuple_hash *); + if (h) { + ct = nf_ct_tuplehash_to_ctrack(h); + atomic_inc(&ct->ct_general.use); + } + read_unlock_bh(&nf_conntrack_lock); + + if (!ct) + return dropped; + + if (del_timer(&ct->timeout)) { + death_by_timeout((unsigned long)ct); + dropped = 1; + NF_CT_STAT_INC(early_drop); + } + nf_ct_put(ct); + return dropped; +} + +static inline int helper_cmp(const struct nf_conntrack_helper *i, + const struct nf_conntrack_tuple *rtuple) +{ + return nf_ct_tuple_mask_cmp(rtuple, &i->tuple, &i->mask); +} + +static struct nf_conntrack_helper * +nf_ct_find_helper(const struct nf_conntrack_tuple *tuple) +{ + return LIST_FIND(&helpers, helper_cmp, + struct nf_conntrack_helper *, + tuple); +} + +static struct nf_conn * +__nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, + const struct nf_conntrack_tuple *repl, + const struct nf_conntrack_l3proto *l3proto) +{ + struct nf_conn *conntrack = NULL; + u_int32_t features = 0; + + if (!nf_conntrack_hash_rnd_initted) { + get_random_bytes(&nf_conntrack_hash_rnd, 4); + nf_conntrack_hash_rnd_initted = 1; + } + + if (nf_conntrack_max + && atomic_read(&nf_conntrack_count) >= nf_conntrack_max) { + unsigned int hash = hash_conntrack(orig); + /* Try dropping from this hash chain. */ + if (!early_drop(&nf_conntrack_hash[hash])) { + if (net_ratelimit()) + printk(KERN_WARNING + "nf_conntrack: table full, dropping" + " packet.\n"); + return ERR_PTR(-ENOMEM); + } + } + + /* find features needed by this conntrack. */ + features = l3proto->get_features(orig); + read_lock_bh(&nf_conntrack_lock); + if (nf_ct_find_helper(repl) != NULL) + features |= NF_CT_F_HELP; + read_unlock_bh(&nf_conntrack_lock); + + DEBUGP("nf_conntrack_alloc: features=0x%x\n", features); + + read_lock_bh(&nf_ct_cache_lock); + + if (!nf_ct_cache[features].use) { + DEBUGP("nf_conntrack_alloc: not supported features = 0x%x\n", + features); + goto out; + } + + conntrack = kmem_cache_alloc(nf_ct_cache[features].cachep, GFP_ATOMIC); + if (conntrack == NULL) { + DEBUGP("nf_conntrack_alloc: Can't alloc conntrack from cache\n"); + goto out; + } + + memset(conntrack, 0, nf_ct_cache[features].size); + conntrack->features = features; + if (nf_ct_cache[features].init_conntrack && + nf_ct_cache[features].init_conntrack(conntrack, features) < 0) { + DEBUGP("nf_conntrack_alloc: failed to init\n"); + kmem_cache_free(nf_ct_cache[features].cachep, conntrack); + conntrack = NULL; + goto out; + } + + atomic_set(&conntrack->ct_general.use, 1); + conntrack->ct_general.destroy = destroy_conntrack; + conntrack->tuplehash[IP_CT_DIR_ORIGINAL].tuple = *orig; + conntrack->tuplehash[IP_CT_DIR_REPLY].tuple = *repl; + /* Don't set timer yet: wait for confirmation */ + init_timer(&conntrack->timeout); + conntrack->timeout.data = (unsigned long)conntrack; + conntrack->timeout.function = death_by_timeout; + + atomic_inc(&nf_conntrack_count); +out: + read_unlock_bh(&nf_ct_cache_lock); + return conntrack; +} + +struct nf_conn *nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, + const struct nf_conntrack_tuple *repl) +{ + struct nf_conntrack_l3proto *l3proto; + + l3proto = nf_ct_find_l3proto(orig->src.l3num); + return __nf_conntrack_alloc(orig, repl, l3proto); +} + +void nf_conntrack_free(struct nf_conn *conntrack) +{ + u_int32_t features = conntrack->features; + NF_CT_ASSERT(features >= NF_CT_F_BASIC && features < NF_CT_F_NUM); + DEBUGP("nf_conntrack_free: features = 0x%x, conntrack=%p\n", features, + conntrack); + kmem_cache_free(nf_ct_cache[features].cachep, conntrack); + atomic_dec(&nf_conntrack_count); +} + +/* Allocate a new conntrack: we return -ENOMEM if classification + failed due to stress. Otherwise it really is unclassifiable. */ +static struct nf_conntrack_tuple_hash * +init_conntrack(const struct nf_conntrack_tuple *tuple, + struct nf_conntrack_l3proto *l3proto, + struct nf_conntrack_protocol *protocol, + struct sk_buff *skb, + unsigned int dataoff) +{ + struct nf_conn *conntrack; + struct nf_conntrack_tuple repl_tuple; + struct nf_conntrack_expect *exp; + + if (!nf_ct_invert_tuple(&repl_tuple, tuple, l3proto, protocol)) { + DEBUGP("Can't invert tuple.\n"); + return NULL; + } + + conntrack = __nf_conntrack_alloc(tuple, &repl_tuple, l3proto); + if (conntrack == NULL || IS_ERR(conntrack)) { + DEBUGP("Can't allocate conntrack.\n"); + return (struct nf_conntrack_tuple_hash *)conntrack; + } + + if (!protocol->new(conntrack, skb, dataoff)) { + nf_conntrack_free(conntrack); + DEBUGP("init conntrack: can't track with proto module\n"); + return NULL; + } + + write_lock_bh(&nf_conntrack_lock); + exp = find_expectation(tuple); + + if (exp) { + DEBUGP("conntrack: expectation arrives ct=%p exp=%p\n", + conntrack, exp); + /* Welcome, Mr. Bond. We've been expecting you... */ + __set_bit(IPS_EXPECTED_BIT, &conntrack->status); + conntrack->master = exp->master; +#ifdef CONFIG_NF_CONNTRACK_MARK + conntrack->mark = exp->master->mark; +#endif + nf_conntrack_get(&conntrack->master->ct_general); + NF_CT_STAT_INC(expect_new); + } else { + conntrack->helper = nf_ct_find_helper(&repl_tuple); + + NF_CT_STAT_INC(new); + } + + /* Overload tuple linked list to put us in unconfirmed list. */ + list_add(&conntrack->tuplehash[IP_CT_DIR_ORIGINAL].list, &unconfirmed); + + write_unlock_bh(&nf_conntrack_lock); + + if (exp) { + if (exp->expectfn) + exp->expectfn(conntrack, exp); + nf_conntrack_expect_put(exp); + } + + return &conntrack->tuplehash[IP_CT_DIR_ORIGINAL]; +} + +/* On success, returns conntrack ptr, sets skb->nfct and ctinfo */ +static inline struct nf_conn * +resolve_normal_ct(struct sk_buff *skb, + unsigned int dataoff, + u_int16_t l3num, + u_int8_t protonum, + struct nf_conntrack_l3proto *l3proto, + struct nf_conntrack_protocol *proto, + int *set_reply, + enum ip_conntrack_info *ctinfo) +{ + struct nf_conntrack_tuple tuple; + struct nf_conntrack_tuple_hash *h; + struct nf_conn *ct; + + if (!nf_ct_get_tuple(skb, (unsigned int)(skb->nh.raw - skb->data), + dataoff, l3num, protonum, &tuple, l3proto, + proto)) { + DEBUGP("resolve_normal_ct: Can't get tuple\n"); + return NULL; + } + + /* look for tuple match */ + h = nf_conntrack_find_get(&tuple, NULL); + if (!h) { + h = init_conntrack(&tuple, l3proto, proto, skb, dataoff); + if (!h) + return NULL; + if (IS_ERR(h)) + return (void *)h; + } + ct = nf_ct_tuplehash_to_ctrack(h); + + /* It exists; we have (non-exclusive) reference. */ + if (NF_CT_DIRECTION(h) == IP_CT_DIR_REPLY) { + *ctinfo = IP_CT_ESTABLISHED + IP_CT_IS_REPLY; + /* Please set reply bit if this packet OK */ + *set_reply = 1; + } else { + /* Once we've had two way comms, always ESTABLISHED. */ + if (test_bit(IPS_SEEN_REPLY_BIT, &ct->status)) { + DEBUGP("nf_conntrack_in: normal packet for %p\n", ct); + *ctinfo = IP_CT_ESTABLISHED; + } else if (test_bit(IPS_EXPECTED_BIT, &ct->status)) { + DEBUGP("nf_conntrack_in: related packet for %p\n", ct); + *ctinfo = IP_CT_RELATED; + } else { + DEBUGP("nf_conntrack_in: new packet for %p\n", ct); + *ctinfo = IP_CT_NEW; + } + *set_reply = 0; + } + skb->nfct = &ct->ct_general; + skb->nfctinfo = *ctinfo; + return ct; +} + +unsigned int +nf_conntrack_in(int pf, unsigned int hooknum, struct sk_buff **pskb) +{ + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + struct nf_conntrack_l3proto *l3proto; + struct nf_conntrack_protocol *proto; + unsigned int dataoff; + u_int8_t protonum; + int set_reply = 0; + int ret; + + /* Previously seen (loopback or untracked)? Ignore. */ + if ((*pskb)->nfct) { + NF_CT_STAT_INC(ignore); + return NF_ACCEPT; + } + + l3proto = nf_ct_find_l3proto((u_int16_t)pf); + if ((ret = l3proto->prepare(pskb, hooknum, &dataoff, &protonum)) <= 0) { + DEBUGP("not prepared to track yet or error occured\n"); + return -ret; + } + + proto = nf_ct_find_proto((u_int16_t)pf, protonum); + + /* It may be an special packet, error, unclean... + * inverse of the return code tells to the netfilter + * core what to do with the packet. */ + if (proto->error != NULL && + (ret = proto->error(*pskb, dataoff, &ctinfo, pf, hooknum)) <= 0) { + NF_CT_STAT_INC(error); + NF_CT_STAT_INC(invalid); + return -ret; + } + + ct = resolve_normal_ct(*pskb, dataoff, pf, protonum, l3proto, proto, + &set_reply, &ctinfo); + if (!ct) { + /* Not valid part of a connection */ + NF_CT_STAT_INC(invalid); + return NF_ACCEPT; + } + + if (IS_ERR(ct)) { + /* Too stressed to deal. */ + NF_CT_STAT_INC(drop); + return NF_DROP; + } + + NF_CT_ASSERT((*pskb)->nfct); + + ret = proto->packet(ct, *pskb, dataoff, ctinfo, pf, hooknum); + if (ret < 0) { + /* Invalid: inverse of the return code tells + * the netfilter core what to do */ + DEBUGP("nf_conntrack_in: Can't track with proto module\n"); + nf_conntrack_put((*pskb)->nfct); + (*pskb)->nfct = NULL; + NF_CT_STAT_INC(invalid); + return -ret; + } + + if (set_reply && !test_and_set_bit(IPS_SEEN_REPLY_BIT, &ct->status)) + nf_conntrack_event_cache(IPCT_STATUS, *pskb); + + return ret; +} + +int nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse, + const struct nf_conntrack_tuple *orig) +{ + return nf_ct_invert_tuple(inverse, orig, + nf_ct_find_l3proto(orig->src.l3num), + nf_ct_find_proto(orig->src.l3num, + orig->dst.protonum)); +} + +/* Would two expected things clash? */ +static inline int expect_clash(const struct nf_conntrack_expect *a, + const struct nf_conntrack_expect *b) +{ + /* Part covered by intersection of masks must be unequal, + otherwise they clash */ + struct nf_conntrack_tuple intersect_mask; + int count; + + intersect_mask.src.l3num = a->mask.src.l3num & b->mask.src.l3num; + intersect_mask.src.u.all = a->mask.src.u.all & b->mask.src.u.all; + intersect_mask.dst.u.all = a->mask.dst.u.all & b->mask.dst.u.all; + intersect_mask.dst.protonum = a->mask.dst.protonum + & b->mask.dst.protonum; + + for (count = 0; count < NF_CT_TUPLE_L3SIZE; count++){ + intersect_mask.src.u3.all[count] = + a->mask.src.u3.all[count] & b->mask.src.u3.all[count]; + } + + for (count = 0; count < NF_CT_TUPLE_L3SIZE; count++){ + intersect_mask.dst.u3.all[count] = + a->mask.dst.u3.all[count] & b->mask.dst.u3.all[count]; + } + + return nf_ct_tuple_mask_cmp(&a->tuple, &b->tuple, &intersect_mask); +} + +static inline int expect_matches(const struct nf_conntrack_expect *a, + const struct nf_conntrack_expect *b) +{ + return a->master == b->master + && nf_ct_tuple_equal(&a->tuple, &b->tuple) + && nf_ct_tuple_equal(&a->mask, &b->mask); +} + +/* Generally a bad idea to call this: could have matched already. */ +void nf_conntrack_unexpect_related(struct nf_conntrack_expect *exp) +{ + struct nf_conntrack_expect *i; + + write_lock_bh(&nf_conntrack_lock); + /* choose the the oldest expectation to evict */ + list_for_each_entry_reverse(i, &nf_conntrack_expect_list, list) { + if (expect_matches(i, exp) && del_timer(&i->timeout)) { + nf_ct_unlink_expect(i); + write_unlock_bh(&nf_conntrack_lock); + nf_conntrack_expect_put(i); + return; + } + } + write_unlock_bh(&nf_conntrack_lock); +} + +/* We don't increase the master conntrack refcount for non-fulfilled + * conntracks. During the conntrack destruction, the expectations are + * always killed before the conntrack itself */ +struct nf_conntrack_expect *nf_conntrack_expect_alloc(struct nf_conn *me) +{ + struct nf_conntrack_expect *new; + + new = kmem_cache_alloc(nf_conntrack_expect_cachep, GFP_ATOMIC); + if (!new) { + DEBUGP("expect_related: OOM allocating expect\n"); + return NULL; + } + new->master = me; + atomic_set(&new->use, 1); + return new; +} + +void nf_conntrack_expect_put(struct nf_conntrack_expect *exp) +{ + if (atomic_dec_and_test(&exp->use)) + kmem_cache_free(nf_conntrack_expect_cachep, exp); +} + +static void nf_conntrack_expect_insert(struct nf_conntrack_expect *exp) +{ + atomic_inc(&exp->use); + exp->master->expecting++; + list_add(&exp->list, &nf_conntrack_expect_list); + + init_timer(&exp->timeout); + exp->timeout.data = (unsigned long)exp; + exp->timeout.function = expectation_timed_out; + exp->timeout.expires = jiffies + exp->master->helper->timeout * HZ; + add_timer(&exp->timeout); + + atomic_inc(&exp->use); + NF_CT_STAT_INC(expect_create); +} + +/* Race with expectations being used means we could have none to find; OK. */ +static void evict_oldest_expect(struct nf_conn *master) +{ + struct nf_conntrack_expect *i; + + list_for_each_entry_reverse(i, &nf_conntrack_expect_list, list) { + if (i->master == master) { + if (del_timer(&i->timeout)) { + nf_ct_unlink_expect(i); + nf_conntrack_expect_put(i); + } + break; + } + } +} + +static inline int refresh_timer(struct nf_conntrack_expect *i) +{ + if (!del_timer(&i->timeout)) + return 0; + + i->timeout.expires = jiffies + i->master->helper->timeout*HZ; + add_timer(&i->timeout); + return 1; +} + +int nf_conntrack_expect_related(struct nf_conntrack_expect *expect) +{ + struct nf_conntrack_expect *i; + int ret; + + DEBUGP("nf_conntrack_expect_related %p\n", related_to); + DEBUGP("tuple: "); NF_CT_DUMP_TUPLE(&expect->tuple); + DEBUGP("mask: "); NF_CT_DUMP_TUPLE(&expect->mask); + + write_lock_bh(&nf_conntrack_lock); + list_for_each_entry(i, &nf_conntrack_expect_list, list) { + if (expect_matches(i, expect)) { + /* Refresh timer: if it's dying, ignore.. */ + if (refresh_timer(i)) { + ret = 0; + goto out; + } + } else if (expect_clash(i, expect)) { + ret = -EBUSY; + goto out; + } + } + /* Will be over limit? */ + if (expect->master->helper->max_expected && + expect->master->expecting >= expect->master->helper->max_expected) + evict_oldest_expect(expect->master); + + nf_conntrack_expect_insert(expect); + nf_conntrack_expect_event(IPEXP_NEW, expect); + ret = 0; +out: + write_unlock_bh(&nf_conntrack_lock); + return ret; +} + +/* Alter reply tuple (maybe alter helper). This is for NAT, and is + implicitly racy: see __nf_conntrack_confirm */ +void nf_conntrack_alter_reply(struct nf_conn *conntrack, + const struct nf_conntrack_tuple *newreply) +{ + write_lock_bh(&nf_conntrack_lock); + /* Should be unconfirmed, so not in hash table yet */ + NF_CT_ASSERT(!nf_ct_is_confirmed(conntrack)); + + DEBUGP("Altering reply tuple of %p to ", conntrack); + NF_CT_DUMP_TUPLE(newreply); + + conntrack->tuplehash[IP_CT_DIR_REPLY].tuple = *newreply; + if (!conntrack->master && conntrack->expecting == 0) + conntrack->helper = nf_ct_find_helper(newreply); + write_unlock_bh(&nf_conntrack_lock); +} + +int nf_conntrack_helper_register(struct nf_conntrack_helper *me) +{ + int ret; + BUG_ON(me->timeout == 0); + + ret = nf_conntrack_register_cache(NF_CT_F_HELP, "nf_conntrack:help", + sizeof(struct nf_conn) + + sizeof(union nf_conntrack_help) + + __alignof__(union nf_conntrack_help), + init_conntrack_for_helper); + if (ret < 0) { + printk(KERN_ERR "nf_conntrack_helper_reigster: Unable to create slab cache for conntracks\n"); + return ret; + } + write_lock_bh(&nf_conntrack_lock); + list_prepend(&helpers, me); + write_unlock_bh(&nf_conntrack_lock); + + return 0; +} + +static inline int unhelp(struct nf_conntrack_tuple_hash *i, + const struct nf_conntrack_helper *me) +{ + if (nf_ct_tuplehash_to_ctrack(i)->helper == me) { + nf_conntrack_event(IPCT_HELPER, nf_ct_tuplehash_to_ctrack(i)); + nf_ct_tuplehash_to_ctrack(i)->helper = NULL; + } + return 0; +} + +void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me) +{ + unsigned int i; + struct nf_conntrack_expect *exp, *tmp; + + /* Need write lock here, to delete helper. */ + write_lock_bh(&nf_conntrack_lock); + LIST_DELETE(&helpers, me); + + /* Get rid of expectations */ + list_for_each_entry_safe(exp, tmp, &nf_conntrack_expect_list, list) { + if (exp->master->helper == me && del_timer(&exp->timeout)) { + nf_ct_unlink_expect(exp); + nf_conntrack_expect_put(exp); + } + } + + /* Get rid of expecteds, set helpers to NULL. */ + LIST_FIND_W(&unconfirmed, unhelp, struct nf_conntrack_tuple_hash*, me); + for (i = 0; i < nf_conntrack_htable_size; i++) + LIST_FIND_W(&nf_conntrack_hash[i], unhelp, + struct nf_conntrack_tuple_hash *, me); + write_unlock_bh(&nf_conntrack_lock); + + /* Someone could be still looking at the helper in a bh. */ + synchronize_net(); +} + +/* Refresh conntrack for this many jiffies and do accounting if do_acct is 1 */ +void __nf_ct_refresh_acct(struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + const struct sk_buff *skb, + unsigned long extra_jiffies, + int do_acct) +{ + int event = 0; + + NF_CT_ASSERT(ct->timeout.data == (unsigned long)ct); + NF_CT_ASSERT(skb); + + write_lock_bh(&nf_conntrack_lock); + + /* If not in hash table, timer will not be active yet */ + if (!nf_ct_is_confirmed(ct)) { + ct->timeout.expires = extra_jiffies; + event = IPCT_REFRESH; + } else { + /* Need del_timer for race avoidance (may already be dying). */ + if (del_timer(&ct->timeout)) { + ct->timeout.expires = jiffies + extra_jiffies; + add_timer(&ct->timeout); + event = IPCT_REFRESH; + } + } + +#ifdef CONFIG_NF_CT_ACCT + if (do_acct) { + ct->counters[CTINFO2DIR(ctinfo)].packets++; + ct->counters[CTINFO2DIR(ctinfo)].bytes += + skb->len - (unsigned int)(skb->nh.raw - skb->data); + if ((ct->counters[CTINFO2DIR(ctinfo)].packets & 0x80000000) + || (ct->counters[CTINFO2DIR(ctinfo)].bytes & 0x80000000)) + event |= IPCT_COUNTER_FILLING; + } +#endif + + write_unlock_bh(&nf_conntrack_lock); + + /* must be unlocked when calling event cache */ + if (event) + nf_conntrack_event_cache(event, skb); +} + +/* Used by ipt_REJECT and ip6t_REJECT. */ +void __nf_conntrack_attach(struct sk_buff *nskb, struct sk_buff *skb) +{ + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + + /* This ICMP is in reverse direction to the packet which caused it */ + ct = nf_ct_get(skb, &ctinfo); + if (CTINFO2DIR(ctinfo) == IP_CT_DIR_ORIGINAL) + ctinfo = IP_CT_RELATED + IP_CT_IS_REPLY; + else + ctinfo = IP_CT_RELATED; + + /* Attach to new skbuff, and increment count */ + nskb->nfct = &ct->ct_general; + nskb->nfctinfo = ctinfo; + nf_conntrack_get(nskb->nfct); +} + +static inline int +do_iter(const struct nf_conntrack_tuple_hash *i, + int (*iter)(struct nf_conn *i, void *data), + void *data) +{ + return iter(nf_ct_tuplehash_to_ctrack(i), data); +} + +/* Bring out ya dead! */ +static struct nf_conntrack_tuple_hash * +get_next_corpse(int (*iter)(struct nf_conn *i, void *data), + void *data, unsigned int *bucket) +{ + struct nf_conntrack_tuple_hash *h = NULL; + + write_lock_bh(&nf_conntrack_lock); + for (; *bucket < nf_conntrack_htable_size; (*bucket)++) { + h = LIST_FIND_W(&nf_conntrack_hash[*bucket], do_iter, + struct nf_conntrack_tuple_hash *, iter, data); + if (h) + break; + } + if (!h) + h = LIST_FIND_W(&unconfirmed, do_iter, + struct nf_conntrack_tuple_hash *, iter, data); + if (h) + atomic_inc(&nf_ct_tuplehash_to_ctrack(h)->ct_general.use); + write_unlock_bh(&nf_conntrack_lock); + + return h; +} + +void +nf_ct_iterate_cleanup(int (*iter)(struct nf_conn *i, void *data), void *data) +{ + struct nf_conntrack_tuple_hash *h; + unsigned int bucket = 0; + + while ((h = get_next_corpse(iter, data, &bucket)) != NULL) { + struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(h); + /* Time to push up daises... */ + if (del_timer(&ct->timeout)) + death_by_timeout((unsigned long)ct); + /* ... else the timer will get him soon. */ + + nf_ct_put(ct); + } +} + +static int kill_all(struct nf_conn *i, void *data) +{ + return 1; +} + +static void free_conntrack_hash(struct list_head *hash, int vmalloced, int size) +{ + if (vmalloced) + vfree(hash); + else + free_pages((unsigned long)hash, + get_order(sizeof(struct list_head) * size)); +} + +/* Mishearing the voices in his head, our hero wonders how he's + supposed to kill the mall. */ +void nf_conntrack_cleanup(void) +{ + int i; + + /* This makes sure all current packets have passed through + netfilter framework. Roll on, two-stage module + delete... */ + synchronize_net(); + + nf_ct_event_cache_flush(); + i_see_dead_people: + nf_ct_iterate_cleanup(kill_all, NULL); + if (atomic_read(&nf_conntrack_count) != 0) { + schedule(); + goto i_see_dead_people; + } + + for (i = 0; i < NF_CT_F_NUM; i++) { + if (nf_ct_cache[i].use == 0) + continue; + + NF_CT_ASSERT(nf_ct_cache[i].use == 1); + nf_ct_cache[i].use = 1; + nf_conntrack_unregister_cache(i); + } + kmem_cache_destroy(nf_conntrack_expect_cachep); + free_conntrack_hash(nf_conntrack_hash, nf_conntrack_vmalloc, + nf_conntrack_htable_size); +} + +static struct list_head *alloc_hashtable(int size, int *vmalloced) +{ + struct list_head *hash; + unsigned int i; + + *vmalloced = 0; + hash = (void*)__get_free_pages(GFP_KERNEL, + get_order(sizeof(struct list_head) + * size)); + if (!hash) { + *vmalloced = 1; + printk(KERN_WARNING "nf_conntrack: falling back to vmalloc.\n"); + hash = vmalloc(sizeof(struct list_head) * size); + } + + if (hash) + for (i = 0; i < size; i++) + INIT_LIST_HEAD(&hash[i]); + + return hash; +} + +int set_hashsize(const char *val, struct kernel_param *kp) +{ + int i, bucket, hashsize, vmalloced; + int old_vmalloced, old_size; + int rnd; + struct list_head *hash, *old_hash; + struct nf_conntrack_tuple_hash *h; + + /* On boot, we can set this without any fancy locking. */ + if (!nf_conntrack_htable_size) + return param_set_uint(val, kp); + + hashsize = simple_strtol(val, NULL, 0); + if (!hashsize) + return -EINVAL; + + hash = alloc_hashtable(hashsize, &vmalloced); + if (!hash) + return -ENOMEM; + + /* We have to rehahs for the new table anyway, so we also can + * use a newrandom seed */ + get_random_bytes(&rnd, 4); + + write_lock_bh(&nf_conntrack_lock); + for (i = 0; i < nf_conntrack_htable_size; i++) { + while (!list_empty(&nf_conntrack_hash[i])) { + h = list_entry(nf_conntrack_hash[i].next, + struct nf_conntrack_tuple_hash, list); + list_del(&h->list); + bucket = __hash_conntrack(&h->tuple, hashsize, rnd); + list_add_tail(&h->list, &hash[bucket]); + } + } + old_size = nf_conntrack_htable_size; + old_vmalloced = nf_conntrack_vmalloc; + old_hash = nf_conntrack_hash; + + nf_conntrack_htable_size = hashsize; + nf_conntrack_vmalloc = vmalloced; + nf_conntrack_hash = hash; + nf_conntrack_hash_rnd = rnd; + write_unlock_bh(&nf_conntrack_lock); + + free_conntrack_hash(old_hash, old_vmalloced, old_size); + return 0; +} + +module_param_call(hashsize, set_hashsize, param_get_uint, + &nf_conntrack_htable_size, 0600); + +int __init nf_conntrack_init(void) +{ + unsigned int i; + int ret; + + /* Idea from tcp.c: use 1/16384 of memory. On i386: 32MB + * machine has 256 buckets. >= 1GB machines have 8192 buckets. */ + if (!nf_conntrack_htable_size) { + nf_conntrack_htable_size + = (((num_physpages << PAGE_SHIFT) / 16384) + / sizeof(struct list_head)); + if (num_physpages > (1024 * 1024 * 1024 / PAGE_SIZE)) + nf_conntrack_htable_size = 8192; + if (nf_conntrack_htable_size < 16) + nf_conntrack_htable_size = 16; + } + nf_conntrack_max = 8 * nf_conntrack_htable_size; + + printk("nf_conntrack version %s (%u buckets, %d max)\n", + NF_CONNTRACK_VERSION, nf_conntrack_htable_size, + nf_conntrack_max); + + nf_conntrack_hash = alloc_hashtable(nf_conntrack_htable_size, + &nf_conntrack_vmalloc); + if (!nf_conntrack_hash) { + printk(KERN_ERR "Unable to create nf_conntrack_hash\n"); + goto err_out; + } + + ret = nf_conntrack_register_cache(NF_CT_F_BASIC, "nf_conntrack:basic", + sizeof(struct nf_conn), NULL); + if (ret < 0) { + printk(KERN_ERR "Unable to create nf_conn slab cache\n"); + goto err_free_hash; + } + + nf_conntrack_expect_cachep = kmem_cache_create("nf_conntrack_expect", + sizeof(struct nf_conntrack_expect), + 0, 0, NULL, NULL); + if (!nf_conntrack_expect_cachep) { + printk(KERN_ERR "Unable to create nf_expect slab cache\n"); + goto err_free_conntrack_slab; + } + + /* Don't NEED lock here, but good form anyway. */ + write_lock_bh(&nf_conntrack_lock); + for (i = 0; i < PF_MAX; i++) + nf_ct_l3protos[i] = &nf_conntrack_generic_l3proto; + write_unlock_bh(&nf_conntrack_lock); + + /* Set up fake conntrack: + - to never be deleted, not in any hashes */ + atomic_set(&nf_conntrack_untracked.ct_general.use, 1); + /* - and look it like as a confirmed connection */ + set_bit(IPS_CONFIRMED_BIT, &nf_conntrack_untracked.status); + + return ret; + +err_free_conntrack_slab: + nf_conntrack_unregister_cache(NF_CT_F_BASIC); +err_free_hash: + free_conntrack_hash(nf_conntrack_hash, nf_conntrack_vmalloc, + nf_conntrack_htable_size); +err_out: + return -ENOMEM; +} diff --git a/net/netfilter/nf_conntrack_ftp.c b/net/netfilter/nf_conntrack_ftp.c new file mode 100644 index 00000000000..65080e269f2 --- /dev/null +++ b/net/netfilter/nf_conntrack_ftp.c @@ -0,0 +1,698 @@ +/* FTP extension for connection tracking. */ + +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org> + * (C) 2003,2004 USAGI/WIDE Project <http://www.linux-ipv6.org> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - enable working with Layer 3 protocol independent connection tracking. + * - track EPRT and EPSV commands with IPv6 address. + * + * Derived from net/ipv4/netfilter/ip_conntrack_ftp.c + */ + +#include <linux/config.h> +#include <linux/module.h> +#include <linux/moduleparam.h> +#include <linux/netfilter.h> +#include <linux/ip.h> +#include <linux/ipv6.h> +#include <linux/ctype.h> +#include <net/checksum.h> +#include <net/tcp.h> + +#include <net/netfilter/nf_conntrack.h> +#include <net/netfilter/nf_conntrack_helper.h> +#include <linux/netfilter/nf_conntrack_ftp.h> + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Rusty Russell <rusty@rustcorp.com.au>"); +MODULE_DESCRIPTION("ftp connection tracking helper"); + +/* This is slow, but it's simple. --RR */ +static char *ftp_buffer; + +static DEFINE_SPINLOCK(nf_ftp_lock); + +#define MAX_PORTS 8 +static u_int16_t ports[MAX_PORTS]; +static unsigned int ports_c; +module_param_array(ports, ushort, &ports_c, 0400); + +static int loose; +module_param(loose, int, 0600); + +unsigned int (*nf_nat_ftp_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + enum ip_ct_ftp_type type, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conntrack_expect *exp, + u32 *seq); +EXPORT_SYMBOL_GPL(nf_nat_ftp_hook); + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +static int try_rfc959(const char *, size_t, struct nf_conntrack_man *, char); +static int try_eprt(const char *, size_t, struct nf_conntrack_man *, char); +static int try_epsv_response(const char *, size_t, struct nf_conntrack_man *, + char); + +static struct ftp_search { + enum ip_conntrack_dir dir; + const char *pattern; + size_t plen; + char skip; + char term; + enum ip_ct_ftp_type ftptype; + int (*getnum)(const char *, size_t, struct nf_conntrack_man *, char); +} search[] = { + { + IP_CT_DIR_ORIGINAL, + "PORT", sizeof("PORT") - 1, ' ', '\r', + IP_CT_FTP_PORT, + try_rfc959, + }, + { + IP_CT_DIR_REPLY, + "227 ", sizeof("227 ") - 1, '(', ')', + IP_CT_FTP_PASV, + try_rfc959, + }, + { + IP_CT_DIR_ORIGINAL, + "EPRT", sizeof("EPRT") - 1, ' ', '\r', + IP_CT_FTP_EPRT, + try_eprt, + }, + { + IP_CT_DIR_REPLY, + "229 ", sizeof("229 ") - 1, '(', ')', + IP_CT_FTP_EPSV, + try_epsv_response, + }, +}; + +/* This code is based on inet_pton() in glibc-2.2.4 */ +static int +get_ipv6_addr(const char *src, size_t dlen, struct in6_addr *dst, u_int8_t term) +{ + static const char xdigits[] = "0123456789abcdef"; + u_int8_t tmp[16], *tp, *endp, *colonp; + int ch, saw_xdigit; + u_int32_t val; + size_t clen = 0; + + tp = memset(tmp, '\0', sizeof(tmp)); + endp = tp + sizeof(tmp); + colonp = NULL; + + /* Leading :: requires some special handling. */ + if (*src == ':'){ + if (*++src != ':') { + DEBUGP("invalid \":\" at the head of addr\n"); + return 0; + } + clen++; + } + + saw_xdigit = 0; + val = 0; + while ((clen < dlen) && (*src != term)) { + const char *pch; + + ch = tolower(*src++); + clen++; + + pch = strchr(xdigits, ch); + if (pch != NULL) { + val <<= 4; + val |= (pch - xdigits); + if (val > 0xffff) + return 0; + + saw_xdigit = 1; + continue; + } + if (ch != ':') { + DEBUGP("get_ipv6_addr: invalid char. \'%c\'\n", ch); + return 0; + } + + if (!saw_xdigit) { + if (colonp) { + DEBUGP("invalid location of \"::\".\n"); + return 0; + } + colonp = tp; + continue; + } else if (*src == term) { + DEBUGP("trancated IPv6 addr\n"); + return 0; + } + + if (tp + 2 > endp) + return 0; + *tp++ = (u_int8_t) (val >> 8) & 0xff; + *tp++ = (u_int8_t) val & 0xff; + + saw_xdigit = 0; + val = 0; + continue; + } + if (saw_xdigit) { + if (tp + 2 > endp) + return 0; + *tp++ = (u_int8_t) (val >> 8) & 0xff; + *tp++ = (u_int8_t) val & 0xff; + } + if (colonp != NULL) { + /* + * Since some memmove()'s erroneously fail to handle + * overlapping regions, we'll do the shift by hand. + */ + const int n = tp - colonp; + int i; + + if (tp == endp) + return 0; + + for (i = 1; i <= n; i++) { + endp[- i] = colonp[n - i]; + colonp[n - i] = 0; + } + tp = endp; + } + if (tp != endp || (*src != term)) + return 0; + + memcpy(dst->s6_addr, tmp, sizeof(dst->s6_addr)); + return clen; +} + +static int try_number(const char *data, size_t dlen, u_int32_t array[], + int array_size, char sep, char term) +{ + u_int32_t i, len; + + memset(array, 0, sizeof(array[0])*array_size); + + /* Keep data pointing at next char. */ + for (i = 0, len = 0; len < dlen && i < array_size; len++, data++) { + if (*data >= '0' && *data <= '9') { + array[i] = array[i]*10 + *data - '0'; + } + else if (*data == sep) + i++; + else { + /* Unexpected character; true if it's the + terminator and we're finished. */ + if (*data == term && i == array_size - 1) + return len; + + DEBUGP("Char %u (got %u nums) `%u' unexpected\n", + len, i, *data); + return 0; + } + } + DEBUGP("Failed to fill %u numbers separated by %c\n", array_size, sep); + + return 0; +} + +/* Returns 0, or length of numbers: 192,168,1,1,5,6 */ +static int try_rfc959(const char *data, size_t dlen, + struct nf_conntrack_man *cmd, char term) +{ + int length; + u_int32_t array[6]; + + length = try_number(data, dlen, array, 6, ',', term); + if (length == 0) + return 0; + + cmd->u3.ip = htonl((array[0] << 24) | (array[1] << 16) | + (array[2] << 8) | array[3]); + cmd->u.tcp.port = htons((array[4] << 8) | array[5]); + return length; +} + +/* Grab port: number up to delimiter */ +static int get_port(const char *data, int start, size_t dlen, char delim, + u_int16_t *port) +{ + u_int16_t tmp_port = 0; + int i; + + for (i = start; i < dlen; i++) { + /* Finished? */ + if (data[i] == delim) { + if (tmp_port == 0) + break; + *port = htons(tmp_port); + DEBUGP("get_port: return %d\n", tmp_port); + return i + 1; + } + else if (data[i] >= '0' && data[i] <= '9') + tmp_port = tmp_port*10 + data[i] - '0'; + else { /* Some other crap */ + DEBUGP("get_port: invalid char.\n"); + break; + } + } + return 0; +} + +/* Returns 0, or length of numbers: |1|132.235.1.2|6275| or |2|3ffe::1|6275| */ +static int try_eprt(const char *data, size_t dlen, struct nf_conntrack_man *cmd, + char term) +{ + char delim; + int length; + + /* First character is delimiter, then "1" for IPv4 or "2" for IPv6, + then delimiter again. */ + if (dlen <= 3) { + DEBUGP("EPRT: too short\n"); + return 0; + } + delim = data[0]; + if (isdigit(delim) || delim < 33 || delim > 126 || data[2] != delim) { + DEBUGP("try_eprt: invalid delimitter.\n"); + return 0; + } + + if ((cmd->l3num == PF_INET && data[1] != '1') || + (cmd->l3num == PF_INET6 && data[1] != '2')) { + DEBUGP("EPRT: invalid protocol number.\n"); + return 0; + } + + DEBUGP("EPRT: Got %c%c%c\n", delim, data[1], delim); + + if (data[1] == '1') { + u_int32_t array[4]; + + /* Now we have IP address. */ + length = try_number(data + 3, dlen - 3, array, 4, '.', delim); + if (length != 0) + cmd->u3.ip = htonl((array[0] << 24) | (array[1] << 16) + | (array[2] << 8) | array[3]); + } else { + /* Now we have IPv6 address. */ + length = get_ipv6_addr(data + 3, dlen - 3, + (struct in6_addr *)cmd->u3.ip6, delim); + } + + if (length == 0) + return 0; + DEBUGP("EPRT: Got IP address!\n"); + /* Start offset includes initial "|1|", and trailing delimiter */ + return get_port(data, 3 + length + 1, dlen, delim, &cmd->u.tcp.port); +} + +/* Returns 0, or length of numbers: |||6446| */ +static int try_epsv_response(const char *data, size_t dlen, + struct nf_conntrack_man *cmd, char term) +{ + char delim; + + /* Three delimiters. */ + if (dlen <= 3) return 0; + delim = data[0]; + if (isdigit(delim) || delim < 33 || delim > 126 + || data[1] != delim || data[2] != delim) + return 0; + + return get_port(data, 3, dlen, delim, &cmd->u.tcp.port); +} + +/* Return 1 for match, 0 for accept, -1 for partial. */ +static int find_pattern(const char *data, size_t dlen, + const char *pattern, size_t plen, + char skip, char term, + unsigned int *numoff, + unsigned int *numlen, + struct nf_conntrack_man *cmd, + int (*getnum)(const char *, size_t, + struct nf_conntrack_man *, char)) +{ + size_t i; + + DEBUGP("find_pattern `%s': dlen = %u\n", pattern, dlen); + if (dlen == 0) + return 0; + + if (dlen <= plen) { + /* Short packet: try for partial? */ + if (strnicmp(data, pattern, dlen) == 0) + return -1; + else return 0; + } + + if (strnicmp(data, pattern, plen) != 0) { +#if 0 + size_t i; + + DEBUGP("ftp: string mismatch\n"); + for (i = 0; i < plen; i++) { + DEBUGP("ftp:char %u `%c'(%u) vs `%c'(%u)\n", + i, data[i], data[i], + pattern[i], pattern[i]); + } +#endif + return 0; + } + + DEBUGP("Pattern matches!\n"); + /* Now we've found the constant string, try to skip + to the 'skip' character */ + for (i = plen; data[i] != skip; i++) + if (i == dlen - 1) return -1; + + /* Skip over the last character */ + i++; + + DEBUGP("Skipped up to `%c'!\n", skip); + + *numoff = i; + *numlen = getnum(data + i, dlen - i, cmd, term); + if (!*numlen) + return -1; + + DEBUGP("Match succeeded!\n"); + return 1; +} + +/* Look up to see if we're just after a \n. */ +static int find_nl_seq(u32 seq, const struct ip_ct_ftp_master *info, int dir) +{ + unsigned int i; + + for (i = 0; i < info->seq_aft_nl_num[dir]; i++) + if (info->seq_aft_nl[dir][i] == seq) + return 1; + return 0; +} + +/* We don't update if it's older than what we have. */ +static void update_nl_seq(u32 nl_seq, struct ip_ct_ftp_master *info, int dir, + struct sk_buff *skb) +{ + unsigned int i, oldest = NUM_SEQ_TO_REMEMBER; + + /* Look for oldest: if we find exact match, we're done. */ + for (i = 0; i < info->seq_aft_nl_num[dir]; i++) { + if (info->seq_aft_nl[dir][i] == nl_seq) + return; + + if (oldest == info->seq_aft_nl_num[dir] + || before(info->seq_aft_nl[dir][i], oldest)) + oldest = i; + } + + if (info->seq_aft_nl_num[dir] < NUM_SEQ_TO_REMEMBER) { + info->seq_aft_nl[dir][info->seq_aft_nl_num[dir]++] = nl_seq; + nf_conntrack_event_cache(IPCT_HELPINFO_VOLATILE, skb); + } else if (oldest != NUM_SEQ_TO_REMEMBER) { + info->seq_aft_nl[dir][oldest] = nl_seq; + nf_conntrack_event_cache(IPCT_HELPINFO_VOLATILE, skb); + } +} + +static int help(struct sk_buff **pskb, + unsigned int protoff, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo) +{ + unsigned int dataoff, datalen; + struct tcphdr _tcph, *th; + char *fb_ptr; + int ret; + u32 seq; + int dir = CTINFO2DIR(ctinfo); + unsigned int matchlen, matchoff; + struct ip_ct_ftp_master *ct_ftp_info = &ct->help->ct_ftp_info; + struct nf_conntrack_expect *exp; + struct nf_conntrack_man cmd = {}; + + unsigned int i; + int found = 0, ends_in_nl; + + /* Until there's been traffic both ways, don't look in packets. */ + if (ctinfo != IP_CT_ESTABLISHED + && ctinfo != IP_CT_ESTABLISHED+IP_CT_IS_REPLY) { + DEBUGP("ftp: Conntrackinfo = %u\n", ctinfo); + return NF_ACCEPT; + } + + th = skb_header_pointer(*pskb, protoff, sizeof(_tcph), &_tcph); + if (th == NULL) + return NF_ACCEPT; + + dataoff = protoff + th->doff * 4; + /* No data? */ + if (dataoff >= (*pskb)->len) { + DEBUGP("ftp: dataoff(%u) >= skblen(%u)\n", dataoff, + (*pskb)->len); + return NF_ACCEPT; + } + datalen = (*pskb)->len - dataoff; + + spin_lock_bh(&nf_ftp_lock); + fb_ptr = skb_header_pointer(*pskb, dataoff, datalen, ftp_buffer); + BUG_ON(fb_ptr == NULL); + + ends_in_nl = (fb_ptr[datalen - 1] == '\n'); + seq = ntohl(th->seq) + datalen; + + /* Look up to see if we're just after a \n. */ + if (!find_nl_seq(ntohl(th->seq), ct_ftp_info, dir)) { + /* Now if this ends in \n, update ftp info. */ + DEBUGP("nf_conntrack_ftp_help: wrong seq pos %s(%u) or %s(%u)\n", + ct_ftp_info->seq_aft_nl_num[dir] > 0 ? "" : "(UNSET)", + ct_ftp_info->seq_aft_nl[dir][0], + ct_ftp_info->seq_aft_nl_num[dir] > 1 ? "" : "(UNSET)", + ct_ftp_info->seq_aft_nl[dir][1]); + ret = NF_ACCEPT; + goto out_update_nl; + } + + /* Initialize IP/IPv6 addr to expected address (it's not mentioned + in EPSV responses) */ + cmd.l3num = ct->tuplehash[dir].tuple.src.l3num; + memcpy(cmd.u3.all, &ct->tuplehash[dir].tuple.src.u3.all, + sizeof(cmd.u3.all)); + + for (i = 0; i < ARRAY_SIZE(search); i++) { + if (search[i].dir != dir) continue; + + found = find_pattern(fb_ptr, datalen, + search[i].pattern, + search[i].plen, + search[i].skip, + search[i].term, + &matchoff, &matchlen, + &cmd, + search[i].getnum); + if (found) break; + } + if (found == -1) { + /* We don't usually drop packets. After all, this is + connection tracking, not packet filtering. + However, it is necessary for accurate tracking in + this case. */ + if (net_ratelimit()) + printk("conntrack_ftp: partial %s %u+%u\n", + search[i].pattern, + ntohl(th->seq), datalen); + ret = NF_DROP; + goto out; + } else if (found == 0) { /* No match */ + ret = NF_ACCEPT; + goto out_update_nl; + } + + DEBUGP("conntrack_ftp: match `%.*s' (%u bytes at %u)\n", + (int)matchlen, fb_ptr + matchoff, + matchlen, ntohl(th->seq) + matchoff); + + exp = nf_conntrack_expect_alloc(ct); + if (exp == NULL) { + ret = NF_DROP; + goto out; + } + + /* We refer to the reverse direction ("!dir") tuples here, + * because we're expecting something in the other direction. + * Doesn't matter unless NAT is happening. */ + exp->tuple.dst.u3 = ct->tuplehash[!dir].tuple.dst.u3; + + /* Update the ftp info */ + if ((cmd.l3num == ct->tuplehash[dir].tuple.src.l3num) && + memcmp(&cmd.u3.all, &ct->tuplehash[dir].tuple.src.u3.all, + sizeof(cmd.u3.all))) { + /* Enrico Scholz's passive FTP to partially RNAT'd ftp + server: it really wants us to connect to a + different IP address. Simply don't record it for + NAT. */ + if (cmd.l3num == PF_INET) { + DEBUGP("conntrack_ftp: NOT RECORDING: %u,%u,%u,%u != %u.%u.%u.%u\n", + NIPQUAD(cmd.u3.ip), + NIPQUAD(ct->tuplehash[dir].tuple.src.u3.ip)); + } else { + DEBUGP("conntrack_ftp: NOT RECORDING: %x:%x:%x:%x:%x:%x:%x:%x != %x:%x:%x:%x:%x:%x:%x:%x\n", + NIP6(*((struct in6_addr *)cmd.u3.ip6)), + NIP6(*((struct in6_addr *)ct->tuplehash[dir] + .tuple.src.u3.ip6))); + } + + /* Thanks to Cristiano Lincoln Mattos + <lincoln@cesar.org.br> for reporting this potential + problem (DMZ machines opening holes to internal + networks, or the packet filter itself). */ + if (!loose) { + ret = NF_ACCEPT; + goto out_put_expect; + } + memcpy(&exp->tuple.dst.u3, &cmd.u3.all, + sizeof(exp->tuple.dst.u3)); + } + + exp->tuple.src.u3 = ct->tuplehash[!dir].tuple.src.u3; + exp->tuple.src.l3num = cmd.l3num; + exp->tuple.src.u.tcp.port = 0; + exp->tuple.dst.u.tcp.port = cmd.u.tcp.port; + exp->tuple.dst.protonum = IPPROTO_TCP; + + exp->mask = (struct nf_conntrack_tuple) + { .src = { .l3num = 0xFFFF, + .u = { .tcp = { 0 }}, + }, + .dst = { .protonum = 0xFF, + .u = { .tcp = { 0xFFFF }}, + }, + }; + if (cmd.l3num == PF_INET) { + exp->mask.src.u3.ip = 0xFFFFFFFF; + exp->mask.dst.u3.ip = 0xFFFFFFFF; + } else { + memset(exp->mask.src.u3.ip6, 0xFF, + sizeof(exp->mask.src.u3.ip6)); + memset(exp->mask.dst.u3.ip6, 0xFF, + sizeof(exp->mask.src.u3.ip6)); + } + + exp->expectfn = NULL; + exp->flags = 0; + + /* Now, NAT might want to mangle the packet, and register the + * (possibly changed) expectation itself. */ + if (nf_nat_ftp_hook) + ret = nf_nat_ftp_hook(pskb, ctinfo, search[i].ftptype, + matchoff, matchlen, exp, &seq); + else { + /* Can't expect this? Best to drop packet now. */ + if (nf_conntrack_expect_related(exp) != 0) + ret = NF_DROP; + else + ret = NF_ACCEPT; + } + +out_put_expect: + nf_conntrack_expect_put(exp); + +out_update_nl: + /* Now if this ends in \n, update ftp info. Seq may have been + * adjusted by NAT code. */ + if (ends_in_nl) + update_nl_seq(seq, ct_ftp_info, dir, *pskb); + out: + spin_unlock_bh(&nf_ftp_lock); + return ret; +} + +static struct nf_conntrack_helper ftp[MAX_PORTS][2]; +static char ftp_names[MAX_PORTS][2][sizeof("ftp-65535")]; + +/* don't make this __exit, since it's called from __init ! */ +static void fini(void) +{ + int i, j; + for (i = 0; i < ports_c; i++) { + for (j = 0; j < 2; j++) { + if (ftp[i][j].me == NULL) + continue; + + DEBUGP("nf_ct_ftp: unregistering helper for pf: %d " + "port: %d\n", + ftp[i][j].tuple.src.l3num, ports[i]); + nf_conntrack_helper_unregister(&ftp[i][j]); + } + } + + kfree(ftp_buffer); +} + +static int __init init(void) +{ + int i, j = -1, ret = 0; + char *tmpname; + + ftp_buffer = kmalloc(65536, GFP_KERNEL); + if (!ftp_buffer) + return -ENOMEM; + + if (ports_c == 0) + ports[ports_c++] = FTP_PORT; + + /* FIXME should be configurable whether IPv4 and IPv6 FTP connections + are tracked or not - YK */ + for (i = 0; i < ports_c; i++) { + memset(&ftp[i], 0, sizeof(struct nf_conntrack_helper)); + + ftp[i][0].tuple.src.l3num = PF_INET; + ftp[i][1].tuple.src.l3num = PF_INET6; + for (j = 0; j < 2; j++) { + ftp[i][j].tuple.src.u.tcp.port = htons(ports[i]); + ftp[i][j].tuple.dst.protonum = IPPROTO_TCP; + ftp[i][j].mask.src.u.tcp.port = 0xFFFF; + ftp[i][j].mask.dst.protonum = 0xFF; + ftp[i][j].max_expected = 1; + ftp[i][j].timeout = 5 * 60; /* 5 Minutes */ + ftp[i][j].me = THIS_MODULE; + ftp[i][j].help = help; + tmpname = &ftp_names[i][j][0]; + if (ports[i] == FTP_PORT) + sprintf(tmpname, "ftp"); + else + sprintf(tmpname, "ftp-%d", ports[i]); + ftp[i][j].name = tmpname; + + DEBUGP("nf_ct_ftp: registering helper for pf: %d " + "port: %d\n", + ftp[i][j].tuple.src.l3num, ports[i]); + ret = nf_conntrack_helper_register(&ftp[i][j]); + if (ret) { + printk("nf_ct_ftp: failed to register helper " + " for pf: %d port: %d\n", + ftp[i][j].tuple.src.l3num, ports[i]); + fini(); + return ret; + } + } + } + + return 0; +} + +module_init(init); +module_exit(fini); diff --git a/net/netfilter/nf_conntrack_l3proto_generic.c b/net/netfilter/nf_conntrack_l3proto_generic.c new file mode 100644 index 00000000000..7de4f06c63c --- /dev/null +++ b/net/netfilter/nf_conntrack_l3proto_generic.c @@ -0,0 +1,98 @@ +/* + * (C) 2003,2004 USAGI/WIDE Project <http://www.linux-ipv6.org> + * + * Based largely upon the original ip_conntrack code which + * had the following copyright information: + * + * (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * Author: + * Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + */ + +#include <linux/config.h> +#include <linux/types.h> +#include <linux/ip.h> +#include <linux/netfilter.h> +#include <linux/module.h> +#include <linux/skbuff.h> +#include <linux/icmp.h> +#include <linux/sysctl.h> +#include <net/ip.h> + +#include <linux/netfilter_ipv4.h> +#include <net/netfilter/nf_conntrack.h> +#include <net/netfilter/nf_conntrack_protocol.h> +#include <net/netfilter/nf_conntrack_l3proto.h> +#include <net/netfilter/nf_conntrack_core.h> +#include <net/netfilter/ipv4/nf_conntrack_ipv4.h> + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +DECLARE_PER_CPU(struct nf_conntrack_stat, nf_conntrack_stat); + +static int generic_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff, + struct nf_conntrack_tuple *tuple) +{ + memset(&tuple->src.u3, 0, sizeof(tuple->src.u3)); + memset(&tuple->dst.u3, 0, sizeof(tuple->dst.u3)); + + return 1; +} + +static int generic_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + memset(&tuple->src.u3, 0, sizeof(tuple->src.u3)); + memset(&tuple->dst.u3, 0, sizeof(tuple->dst.u3)); + + return 1; +} + +static int generic_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return 0; +} + +static int generic_print_conntrack(struct seq_file *s, + const struct nf_conn *conntrack) +{ + return 0; +} + +static int +generic_prepare(struct sk_buff **pskb, unsigned int hooknum, + unsigned int *dataoff, u_int8_t *protonum) +{ + /* Never track !!! */ + return -NF_ACCEPT; +} + + +static u_int32_t generic_get_features(const struct nf_conntrack_tuple *tuple) + +{ + return NF_CT_F_BASIC; +} + +struct nf_conntrack_l3proto nf_conntrack_generic_l3proto = { + .l3proto = PF_UNSPEC, + .name = "unknown", + .pkt_to_tuple = generic_pkt_to_tuple, + .invert_tuple = generic_invert_tuple, + .print_tuple = generic_print_tuple, + .print_conntrack = generic_print_conntrack, + .prepare = generic_prepare, + .get_features = generic_get_features, + .me = THIS_MODULE, +}; diff --git a/net/netfilter/nf_conntrack_proto_generic.c b/net/netfilter/nf_conntrack_proto_generic.c new file mode 100644 index 00000000000..36425f6c833 --- /dev/null +++ b/net/netfilter/nf_conntrack_proto_generic.c @@ -0,0 +1,85 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - enable working with L3 protocol independent connection tracking. + * + * Derived from net/ipv4/netfilter/ip_conntrack_proto_generic.c + */ + +#include <linux/types.h> +#include <linux/sched.h> +#include <linux/timer.h> +#include <linux/netfilter.h> +#include <net/netfilter/nf_conntrack_protocol.h> + +unsigned long nf_ct_generic_timeout = 600*HZ; + +static int generic_pkt_to_tuple(const struct sk_buff *skb, + unsigned int dataoff, + struct nf_conntrack_tuple *tuple) +{ + tuple->src.u.all = 0; + tuple->dst.u.all = 0; + + return 1; +} + +static int generic_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + tuple->src.u.all = 0; + tuple->dst.u.all = 0; + + return 1; +} + +/* Print out the per-protocol part of the tuple. */ +static int generic_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return 0; +} + +/* Print out the private part of the conntrack. */ +static int generic_print_conntrack(struct seq_file *s, + const struct nf_conn *state) +{ + return 0; +} + +/* Returns verdict for packet, or -1 for invalid. */ +static int packet(struct nf_conn *conntrack, + const struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info ctinfo, + int pf, + unsigned int hooknum) +{ + nf_ct_refresh_acct(conntrack, ctinfo, skb, nf_ct_generic_timeout); + return NF_ACCEPT; +} + +/* Called when a new connection for this protocol found. */ +static int new(struct nf_conn *conntrack, const struct sk_buff *skb, + unsigned int dataoff) +{ + return 1; +} + +struct nf_conntrack_protocol nf_conntrack_generic_protocol = +{ + .l3proto = PF_UNSPEC, + .proto = 0, + .name = "unknown", + .pkt_to_tuple = generic_pkt_to_tuple, + .invert_tuple = generic_invert_tuple, + .print_tuple = generic_print_tuple, + .print_conntrack = generic_print_conntrack, + .packet = packet, + .new = new, +}; diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c new file mode 100644 index 00000000000..3a600f77b4e --- /dev/null +++ b/net/netfilter/nf_conntrack_proto_sctp.c @@ -0,0 +1,670 @@ +/* + * Connection tracking protocol helper module for SCTP. + * + * SCTP is defined in RFC 2960. References to various sections in this code + * are to this RFC. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * 17 Oct 2004: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - enable working with L3 protocol independent connection tracking. + * + * Derived from net/ipv4/ip_conntrack_sctp.c + */ + +/* + * Added support for proc manipulation of timeouts. + */ + +#include <linux/types.h> +#include <linux/sched.h> +#include <linux/timer.h> +#include <linux/netfilter.h> +#include <linux/module.h> +#include <linux/in.h> +#include <linux/ip.h> +#include <linux/sctp.h> +#include <linux/string.h> +#include <linux/seq_file.h> + +#include <net/netfilter/nf_conntrack.h> +#include <net/netfilter/nf_conntrack_protocol.h> + +#if 0 +#define DEBUGP(format, ...) printk(format, ## __VA_ARGS__) +#else +#define DEBUGP(format, args...) +#endif + +/* Protects conntrack->proto.sctp */ +static DEFINE_RWLOCK(sctp_lock); + +/* FIXME: Examine ipfilter's timeouts and conntrack transitions more + closely. They're more complex. --RR + + And so for me for SCTP :D -Kiran */ + +static const char *sctp_conntrack_names[] = { + "NONE", + "CLOSED", + "COOKIE_WAIT", + "COOKIE_ECHOED", + "ESTABLISHED", + "SHUTDOWN_SENT", + "SHUTDOWN_RECD", + "SHUTDOWN_ACK_SENT", +}; + +#define SECS * HZ +#define MINS * 60 SECS +#define HOURS * 60 MINS +#define DAYS * 24 HOURS + +static unsigned long nf_ct_sctp_timeout_closed = 10 SECS; +static unsigned long nf_ct_sctp_timeout_cookie_wait = 3 SECS; +static unsigned long nf_ct_sctp_timeout_cookie_echoed = 3 SECS; +static unsigned long nf_ct_sctp_timeout_established = 5 DAYS; +static unsigned long nf_ct_sctp_timeout_shutdown_sent = 300 SECS / 1000; +static unsigned long nf_ct_sctp_timeout_shutdown_recd = 300 SECS / 1000; +static unsigned long nf_ct_sctp_timeout_shutdown_ack_sent = 3 SECS; + +static unsigned long * sctp_timeouts[] += { NULL, /* SCTP_CONNTRACK_NONE */ + &nf_ct_sctp_timeout_closed, /* SCTP_CONNTRACK_CLOSED */ + &nf_ct_sctp_timeout_cookie_wait, /* SCTP_CONNTRACK_COOKIE_WAIT */ + &nf_ct_sctp_timeout_cookie_echoed, /* SCTP_CONNTRACK_COOKIE_ECHOED */ + &nf_ct_sctp_timeout_established, /* SCTP_CONNTRACK_ESTABLISHED */ + &nf_ct_sctp_timeout_shutdown_sent, /* SCTP_CONNTRACK_SHUTDOWN_SENT */ + &nf_ct_sctp_timeout_shutdown_recd, /* SCTP_CONNTRACK_SHUTDOWN_RECD */ + &nf_ct_sctp_timeout_shutdown_ack_sent /* SCTP_CONNTRACK_SHUTDOWN_ACK_SENT */ + }; + +#define sNO SCTP_CONNTRACK_NONE +#define sCL SCTP_CONNTRACK_CLOSED +#define sCW SCTP_CONNTRACK_COOKIE_WAIT +#define sCE SCTP_CONNTRACK_COOKIE_ECHOED +#define sES SCTP_CONNTRACK_ESTABLISHED +#define sSS SCTP_CONNTRACK_SHUTDOWN_SENT +#define sSR SCTP_CONNTRACK_SHUTDOWN_RECD +#define sSA SCTP_CONNTRACK_SHUTDOWN_ACK_SENT +#define sIV SCTP_CONNTRACK_MAX + +/* + These are the descriptions of the states: + +NOTE: These state names are tantalizingly similar to the states of an +SCTP endpoint. But the interpretation of the states is a little different, +considering that these are the states of the connection and not of an end +point. Please note the subtleties. -Kiran + +NONE - Nothing so far. +COOKIE WAIT - We have seen an INIT chunk in the original direction, or also + an INIT_ACK chunk in the reply direction. +COOKIE ECHOED - We have seen a COOKIE_ECHO chunk in the original direction. +ESTABLISHED - We have seen a COOKIE_ACK in the reply direction. +SHUTDOWN_SENT - We have seen a SHUTDOWN chunk in the original direction. +SHUTDOWN_RECD - We have seen a SHUTDOWN chunk in the reply directoin. +SHUTDOWN_ACK_SENT - We have seen a SHUTDOWN_ACK chunk in the direction opposite + to that of the SHUTDOWN chunk. +CLOSED - We have seen a SHUTDOWN_COMPLETE chunk in the direction of + the SHUTDOWN chunk. Connection is closed. +*/ + +/* TODO + - I have assumed that the first INIT is in the original direction. + This messes things when an INIT comes in the reply direction in CLOSED + state. + - Check the error type in the reply dir before transitioning from +cookie echoed to closed. + - Sec 5.2.4 of RFC 2960 + - Multi Homing support. +*/ + +/* SCTP conntrack state transitions */ +static enum sctp_conntrack sctp_conntracks[2][9][SCTP_CONNTRACK_MAX] = { + { +/* ORIGINAL */ +/* sNO, sCL, sCW, sCE, sES, sSS, sSR, sSA */ +/* init */ {sCW, sCW, sCW, sCE, sES, sSS, sSR, sSA}, +/* init_ack */ {sCL, sCL, sCW, sCE, sES, sSS, sSR, sSA}, +/* abort */ {sCL, sCL, sCL, sCL, sCL, sCL, sCL, sCL}, +/* shutdown */ {sCL, sCL, sCW, sCE, sSS, sSS, sSR, sSA}, +/* shutdown_ack */ {sSA, sCL, sCW, sCE, sES, sSA, sSA, sSA}, +/* error */ {sCL, sCL, sCW, sCE, sES, sSS, sSR, sSA},/* Cant have Stale cookie*/ +/* cookie_echo */ {sCL, sCL, sCE, sCE, sES, sSS, sSR, sSA},/* 5.2.4 - Big TODO */ +/* cookie_ack */ {sCL, sCL, sCW, sCE, sES, sSS, sSR, sSA},/* Cant come in orig dir */ +/* shutdown_comp*/ {sCL, sCL, sCW, sCE, sES, sSS, sSR, sCL} + }, + { +/* REPLY */ +/* sNO, sCL, sCW, sCE, sES, sSS, sSR, sSA */ +/* init */ {sIV, sCL, sCW, sCE, sES, sSS, sSR, sSA},/* INIT in sCL Big TODO */ +/* init_ack */ {sIV, sCL, sCW, sCE, sES, sSS, sSR, sSA}, +/* abort */ {sIV, sCL, sCL, sCL, sCL, sCL, sCL, sCL}, +/* shutdown */ {sIV, sCL, sCW, sCE, sSR, sSS, sSR, sSA}, +/* shutdown_ack */ {sIV, sCL, sCW, sCE, sES, sSA, sSA, sSA}, +/* error */ {sIV, sCL, sCW, sCL, sES, sSS, sSR, sSA}, +/* cookie_echo */ {sIV, sCL, sCW, sCE, sES, sSS, sSR, sSA},/* Cant come in reply dir */ +/* cookie_ack */ {sIV, sCL, sCW, sES, sES, sSS, sSR, sSA}, +/* shutdown_comp*/ {sIV, sCL, sCW, sCE, sES, sSS, sSR, sCL} + } +}; + +static int sctp_pkt_to_tuple(const struct sk_buff *skb, + unsigned int dataoff, + struct nf_conntrack_tuple *tuple) +{ + sctp_sctphdr_t _hdr, *hp; + + DEBUGP(__FUNCTION__); + DEBUGP("\n"); + + /* Actually only need first 8 bytes. */ + hp = skb_header_pointer(skb, dataoff, 8, &_hdr); + if (hp == NULL) + return 0; + + tuple->src.u.sctp.port = hp->source; + tuple->dst.u.sctp.port = hp->dest; + return 1; +} + +static int sctp_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + DEBUGP(__FUNCTION__); + DEBUGP("\n"); + + tuple->src.u.sctp.port = orig->dst.u.sctp.port; + tuple->dst.u.sctp.port = orig->src.u.sctp.port; + return 1; +} + +/* Print out the per-protocol part of the tuple. */ +static int sctp_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + DEBUGP(__FUNCTION__); + DEBUGP("\n"); + + return seq_printf(s, "sport=%hu dport=%hu ", + ntohs(tuple->src.u.sctp.port), + ntohs(tuple->dst.u.sctp.port)); +} + +/* Print out the private part of the conntrack. */ +static int sctp_print_conntrack(struct seq_file *s, + const struct nf_conn *conntrack) +{ + enum sctp_conntrack state; + + DEBUGP(__FUNCTION__); + DEBUGP("\n"); + + read_lock_bh(&sctp_lock); + state = conntrack->proto.sctp.state; + read_unlock_bh(&sctp_lock); + + return seq_printf(s, "%s ", sctp_conntrack_names[state]); +} + +#define for_each_sctp_chunk(skb, sch, _sch, offset, dataoff, count) \ +for (offset = dataoff + sizeof(sctp_sctphdr_t), count = 0; \ + offset < skb->len && \ + (sch = skb_header_pointer(skb, offset, sizeof(_sch), &_sch)); \ + offset += (htons(sch->length) + 3) & ~3, count++) + +/* Some validity checks to make sure the chunks are fine */ +static int do_basic_checks(struct nf_conn *conntrack, + const struct sk_buff *skb, + unsigned int dataoff, + char *map) +{ + u_int32_t offset, count; + sctp_chunkhdr_t _sch, *sch; + int flag; + + DEBUGP(__FUNCTION__); + DEBUGP("\n"); + + flag = 0; + + for_each_sctp_chunk (skb, sch, _sch, offset, dataoff, count) { + DEBUGP("Chunk Num: %d Type: %d\n", count, sch->type); + + if (sch->type == SCTP_CID_INIT + || sch->type == SCTP_CID_INIT_ACK + || sch->type == SCTP_CID_SHUTDOWN_COMPLETE) { + flag = 1; + } + + /* Cookie Ack/Echo chunks not the first OR + Init / Init Ack / Shutdown compl chunks not the only chunks */ + if ((sch->type == SCTP_CID_COOKIE_ACK + || sch->type == SCTP_CID_COOKIE_ECHO + || flag) + && count !=0 ) { + DEBUGP("Basic checks failed\n"); + return 1; + } + + if (map) { + set_bit(sch->type, (void *)map); + } + } + + DEBUGP("Basic checks passed\n"); + return 0; +} + +static int new_state(enum ip_conntrack_dir dir, + enum sctp_conntrack cur_state, + int chunk_type) +{ + int i; + + DEBUGP(__FUNCTION__); + DEBUGP("\n"); + + DEBUGP("Chunk type: %d\n", chunk_type); + + switch (chunk_type) { + case SCTP_CID_INIT: + DEBUGP("SCTP_CID_INIT\n"); + i = 0; break; + case SCTP_CID_INIT_ACK: + DEBUGP("SCTP_CID_INIT_ACK\n"); + i = 1; break; + case SCTP_CID_ABORT: + DEBUGP("SCTP_CID_ABORT\n"); + i = 2; break; + case SCTP_CID_SHUTDOWN: + DEBUGP("SCTP_CID_SHUTDOWN\n"); + i = 3; break; + case SCTP_CID_SHUTDOWN_ACK: + DEBUGP("SCTP_CID_SHUTDOWN_ACK\n"); + i = 4; break; + case SCTP_CID_ERROR: + DEBUGP("SCTP_CID_ERROR\n"); + i = 5; break; + case SCTP_CID_COOKIE_ECHO: + DEBUGP("SCTP_CID_COOKIE_ECHO\n"); + i = 6; break; + case SCTP_CID_COOKIE_ACK: + DEBUGP("SCTP_CID_COOKIE_ACK\n"); + i = 7; break; + case SCTP_CID_SHUTDOWN_COMPLETE: + DEBUGP("SCTP_CID_SHUTDOWN_COMPLETE\n"); + i = 8; break; + default: + /* Other chunks like DATA, SACK, HEARTBEAT and + its ACK do not cause a change in state */ + DEBUGP("Unknown chunk type, Will stay in %s\n", + sctp_conntrack_names[cur_state]); + return cur_state; + } + + DEBUGP("dir: %d cur_state: %s chunk_type: %d new_state: %s\n", + dir, sctp_conntrack_names[cur_state], chunk_type, + sctp_conntrack_names[sctp_conntracks[dir][i][cur_state]]); + + return sctp_conntracks[dir][i][cur_state]; +} + +/* Returns verdict for packet, or -1 for invalid. */ +static int sctp_packet(struct nf_conn *conntrack, + const struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info ctinfo, + int pf, + unsigned int hooknum) +{ + enum sctp_conntrack newconntrack, oldsctpstate; + sctp_sctphdr_t _sctph, *sh; + sctp_chunkhdr_t _sch, *sch; + u_int32_t offset, count; + char map[256 / sizeof (char)] = {0}; + + DEBUGP(__FUNCTION__); + DEBUGP("\n"); + + sh = skb_header_pointer(skb, dataoff, sizeof(_sctph), &_sctph); + if (sh == NULL) + return -1; + + if (do_basic_checks(conntrack, skb, dataoff, map) != 0) + return -1; + + /* Check the verification tag (Sec 8.5) */ + if (!test_bit(SCTP_CID_INIT, (void *)map) + && !test_bit(SCTP_CID_SHUTDOWN_COMPLETE, (void *)map) + && !test_bit(SCTP_CID_COOKIE_ECHO, (void *)map) + && !test_bit(SCTP_CID_ABORT, (void *)map) + && !test_bit(SCTP_CID_SHUTDOWN_ACK, (void *)map) + && (sh->vtag != conntrack->proto.sctp.vtag[CTINFO2DIR(ctinfo)])) { + DEBUGP("Verification tag check failed\n"); + return -1; + } + + oldsctpstate = newconntrack = SCTP_CONNTRACK_MAX; + for_each_sctp_chunk (skb, sch, _sch, offset, dataoff, count) { + write_lock_bh(&sctp_lock); + + /* Special cases of Verification tag check (Sec 8.5.1) */ + if (sch->type == SCTP_CID_INIT) { + /* Sec 8.5.1 (A) */ + if (sh->vtag != 0) { + write_unlock_bh(&sctp_lock); + return -1; + } + } else if (sch->type == SCTP_CID_ABORT) { + /* Sec 8.5.1 (B) */ + if (!(sh->vtag == conntrack->proto.sctp.vtag[CTINFO2DIR(ctinfo)]) + && !(sh->vtag == conntrack->proto.sctp.vtag + [1 - CTINFO2DIR(ctinfo)])) { + write_unlock_bh(&sctp_lock); + return -1; + } + } else if (sch->type == SCTP_CID_SHUTDOWN_COMPLETE) { + /* Sec 8.5.1 (C) */ + if (!(sh->vtag == conntrack->proto.sctp.vtag[CTINFO2DIR(ctinfo)]) + && !(sh->vtag == conntrack->proto.sctp.vtag + [1 - CTINFO2DIR(ctinfo)] + && (sch->flags & 1))) { + write_unlock_bh(&sctp_lock); + return -1; + } + } else if (sch->type == SCTP_CID_COOKIE_ECHO) { + /* Sec 8.5.1 (D) */ + if (!(sh->vtag == conntrack->proto.sctp.vtag[CTINFO2DIR(ctinfo)])) { + write_unlock_bh(&sctp_lock); + return -1; + } + } + + oldsctpstate = conntrack->proto.sctp.state; + newconntrack = new_state(CTINFO2DIR(ctinfo), oldsctpstate, sch->type); + + /* Invalid */ + if (newconntrack == SCTP_CONNTRACK_MAX) { + DEBUGP("nf_conntrack_sctp: Invalid dir=%i ctype=%u conntrack=%u\n", + CTINFO2DIR(ctinfo), sch->type, oldsctpstate); + write_unlock_bh(&sctp_lock); + return -1; + } + + /* If it is an INIT or an INIT ACK note down the vtag */ + if (sch->type == SCTP_CID_INIT + || sch->type == SCTP_CID_INIT_ACK) { + sctp_inithdr_t _inithdr, *ih; + + ih = skb_header_pointer(skb, offset + sizeof(sctp_chunkhdr_t), + sizeof(_inithdr), &_inithdr); + if (ih == NULL) { + write_unlock_bh(&sctp_lock); + return -1; + } + DEBUGP("Setting vtag %x for dir %d\n", + ih->init_tag, !CTINFO2DIR(ctinfo)); + conntrack->proto.sctp.vtag[!CTINFO2DIR(ctinfo)] = ih->init_tag; + } + + conntrack->proto.sctp.state = newconntrack; + if (oldsctpstate != newconntrack) + nf_conntrack_event_cache(IPCT_PROTOINFO, skb); + write_unlock_bh(&sctp_lock); + } + + nf_ct_refresh_acct(conntrack, ctinfo, skb, *sctp_timeouts[newconntrack]); + + if (oldsctpstate == SCTP_CONNTRACK_COOKIE_ECHOED + && CTINFO2DIR(ctinfo) == IP_CT_DIR_REPLY + && newconntrack == SCTP_CONNTRACK_ESTABLISHED) { + DEBUGP("Setting assured bit\n"); + set_bit(IPS_ASSURED_BIT, &conntrack->status); + nf_conntrack_event_cache(IPCT_STATUS, skb); + } + + return NF_ACCEPT; +} + +/* Called when a new connection for this protocol found. */ +static int sctp_new(struct nf_conn *conntrack, const struct sk_buff *skb, + unsigned int dataoff) +{ + enum sctp_conntrack newconntrack; + sctp_sctphdr_t _sctph, *sh; + sctp_chunkhdr_t _sch, *sch; + u_int32_t offset, count; + char map[256 / sizeof (char)] = {0}; + + DEBUGP(__FUNCTION__); + DEBUGP("\n"); + + sh = skb_header_pointer(skb, dataoff, sizeof(_sctph), &_sctph); + if (sh == NULL) + return 0; + + if (do_basic_checks(conntrack, skb, dataoff, map) != 0) + return 0; + + /* If an OOTB packet has any of these chunks discard (Sec 8.4) */ + if ((test_bit (SCTP_CID_ABORT, (void *)map)) + || (test_bit (SCTP_CID_SHUTDOWN_COMPLETE, (void *)map)) + || (test_bit (SCTP_CID_COOKIE_ACK, (void *)map))) { + return 0; + } + + newconntrack = SCTP_CONNTRACK_MAX; + for_each_sctp_chunk (skb, sch, _sch, offset, dataoff, count) { + /* Don't need lock here: this conntrack not in circulation yet */ + newconntrack = new_state(IP_CT_DIR_ORIGINAL, + SCTP_CONNTRACK_NONE, sch->type); + + /* Invalid: delete conntrack */ + if (newconntrack == SCTP_CONNTRACK_MAX) { + DEBUGP("nf_conntrack_sctp: invalid new deleting.\n"); + return 0; + } + + /* Copy the vtag into the state info */ + if (sch->type == SCTP_CID_INIT) { + if (sh->vtag == 0) { + sctp_inithdr_t _inithdr, *ih; + + ih = skb_header_pointer(skb, offset + sizeof(sctp_chunkhdr_t), + sizeof(_inithdr), &_inithdr); + if (ih == NULL) + return 0; + + DEBUGP("Setting vtag %x for new conn\n", + ih->init_tag); + + conntrack->proto.sctp.vtag[IP_CT_DIR_REPLY] = + ih->init_tag; + } else { + /* Sec 8.5.1 (A) */ + return 0; + } + } + /* If it is a shutdown ack OOTB packet, we expect a return + shutdown complete, otherwise an ABORT Sec 8.4 (5) and (8) */ + else { + DEBUGP("Setting vtag %x for new conn OOTB\n", + sh->vtag); + conntrack->proto.sctp.vtag[IP_CT_DIR_REPLY] = sh->vtag; + } + + conntrack->proto.sctp.state = newconntrack; + } + + return 1; +} + +struct nf_conntrack_protocol nf_conntrack_protocol_sctp4 = { + .l3proto = PF_INET, + .proto = IPPROTO_SCTP, + .name = "sctp", + .pkt_to_tuple = sctp_pkt_to_tuple, + .invert_tuple = sctp_invert_tuple, + .print_tuple = sctp_print_tuple, + .print_conntrack = sctp_print_conntrack, + .packet = sctp_packet, + .new = sctp_new, + .destroy = NULL, + .me = THIS_MODULE +}; + +struct nf_conntrack_protocol nf_conntrack_protocol_sctp6 = { + .l3proto = PF_INET6, + .proto = IPPROTO_SCTP, + .name = "sctp", + .pkt_to_tuple = sctp_pkt_to_tuple, + .invert_tuple = sctp_invert_tuple, + .print_tuple = sctp_print_tuple, + .print_conntrack = sctp_print_conntrack, + .packet = sctp_packet, + .new = sctp_new, + .destroy = NULL, + .me = THIS_MODULE +}; + +#ifdef CONFIG_SYSCTL +static ctl_table nf_ct_sysctl_table[] = { + { + .ctl_name = NET_NF_CONNTRACK_SCTP_TIMEOUT_CLOSED, + .procname = "nf_conntrack_sctp_timeout_closed", + .data = &nf_ct_sctp_timeout_closed, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_SCTP_TIMEOUT_COOKIE_WAIT, + .procname = "nf_conntrack_sctp_timeout_cookie_wait", + .data = &nf_ct_sctp_timeout_cookie_wait, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_SCTP_TIMEOUT_COOKIE_ECHOED, + .procname = "nf_conntrack_sctp_timeout_cookie_echoed", + .data = &nf_ct_sctp_timeout_cookie_echoed, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_SCTP_TIMEOUT_ESTABLISHED, + .procname = "nf_conntrack_sctp_timeout_established", + .data = &nf_ct_sctp_timeout_established, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_SENT, + .procname = "nf_conntrack_sctp_timeout_shutdown_sent", + .data = &nf_ct_sctp_timeout_shutdown_sent, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_RECD, + .procname = "nf_conntrack_sctp_timeout_shutdown_recd", + .data = &nf_ct_sctp_timeout_shutdown_recd, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_ACK_SENT, + .procname = "nf_conntrack_sctp_timeout_shutdown_ack_sent", + .data = &nf_ct_sctp_timeout_shutdown_ack_sent, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { .ctl_name = 0 } +}; + +static ctl_table nf_ct_netfilter_table[] = { + { + .ctl_name = NET_NETFILTER, + .procname = "netfilter", + .mode = 0555, + .child = nf_ct_sysctl_table, + }, + { .ctl_name = 0 } +}; + +static ctl_table nf_ct_net_table[] = { + { + .ctl_name = CTL_NET, + .procname = "net", + .mode = 0555, + .child = nf_ct_netfilter_table, + }, + { .ctl_name = 0 } +}; + +static struct ctl_table_header *nf_ct_sysctl_header; +#endif + +int __init init(void) +{ + int ret; + + ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_sctp4); + if (ret) { + printk("nf_conntrack_proto_sctp4: protocol register failed\n"); + goto out; + } + ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_sctp6); + if (ret) { + printk("nf_conntrack_proto_sctp6: protocol register failed\n"); + goto cleanup_sctp4; + } + +#ifdef CONFIG_SYSCTL + nf_ct_sysctl_header = register_sysctl_table(nf_ct_net_table, 0); + if (nf_ct_sysctl_header == NULL) { + printk("nf_conntrack_proto_sctp: can't register to sysctl.\n"); + goto cleanup; + } +#endif + + return ret; + +#ifdef CONFIG_SYSCTL + cleanup: + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_sctp6); +#endif + cleanup_sctp4: + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_sctp4); + out: + DEBUGP("SCTP conntrack module loading %s\n", + ret ? "failed": "succeeded"); + return ret; +} + +void __exit fini(void) +{ + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_sctp6); + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_sctp4); +#ifdef CONFIG_SYSCTL + unregister_sysctl_table(nf_ct_sysctl_header); +#endif + DEBUGP("SCTP conntrack module unloaded\n"); +} + +module_init(init); +module_exit(fini); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Kiran Kumar Immidi"); +MODULE_DESCRIPTION("Netfilter connection tracking protocol helper for SCTP"); diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c new file mode 100644 index 00000000000..83d90dd624f --- /dev/null +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -0,0 +1,1162 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>: + * - Real stateful connection tracking + * - Modified state transitions table + * - Window scaling support added + * - SACK support added + * + * Willy Tarreau: + * - State table bugfixes + * - More robust state changes + * - Tuning timer parameters + * + * 27 Oct 2004: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - genelized Layer 3 protocol part. + * + * Derived from net/ipv4/netfilter/ip_conntrack_proto_tcp.c + * + * version 2.2 + */ + +#include <linux/config.h> +#include <linux/types.h> +#include <linux/sched.h> +#include <linux/timer.h> +#include <linux/netfilter.h> +#include <linux/module.h> +#include <linux/in.h> +#include <linux/tcp.h> +#include <linux/spinlock.h> +#include <linux/skbuff.h> +#include <linux/ipv6.h> +#include <net/ip6_checksum.h> + +#include <net/tcp.h> + +#include <linux/netfilter.h> +#include <linux/netfilter_ipv4.h> +#include <linux/netfilter_ipv6.h> +#include <net/netfilter/nf_conntrack.h> +#include <net/netfilter/nf_conntrack_protocol.h> + +#if 0 +#define DEBUGP printk +#define DEBUGP_VARS +#else +#define DEBUGP(format, args...) +#endif + +/* Protects conntrack->proto.tcp */ +static DEFINE_RWLOCK(tcp_lock); + +/* "Be conservative in what you do, + be liberal in what you accept from others." + If it's non-zero, we mark only out of window RST segments as INVALID. */ +int nf_ct_tcp_be_liberal = 0; + +/* When connection is picked up from the middle, how many packets are required + to pass in each direction when we assume we are in sync - if any side uses + window scaling, we lost the game. + If it is set to zero, we disable picking up already established + connections. */ +int nf_ct_tcp_loose = 3; + +/* Max number of the retransmitted packets without receiving an (acceptable) + ACK from the destination. If this number is reached, a shorter timer + will be started. */ +int nf_ct_tcp_max_retrans = 3; + + /* FIXME: Examine ipfilter's timeouts and conntrack transitions more + closely. They're more complex. --RR */ + +static const char *tcp_conntrack_names[] = { + "NONE", + "SYN_SENT", + "SYN_RECV", + "ESTABLISHED", + "FIN_WAIT", + "CLOSE_WAIT", + "LAST_ACK", + "TIME_WAIT", + "CLOSE", + "LISTEN" +}; + +#define SECS * HZ +#define MINS * 60 SECS +#define HOURS * 60 MINS +#define DAYS * 24 HOURS + +unsigned long nf_ct_tcp_timeout_syn_sent = 2 MINS; +unsigned long nf_ct_tcp_timeout_syn_recv = 60 SECS; +unsigned long nf_ct_tcp_timeout_established = 5 DAYS; +unsigned long nf_ct_tcp_timeout_fin_wait = 2 MINS; +unsigned long nf_ct_tcp_timeout_close_wait = 60 SECS; +unsigned long nf_ct_tcp_timeout_last_ack = 30 SECS; +unsigned long nf_ct_tcp_timeout_time_wait = 2 MINS; +unsigned long nf_ct_tcp_timeout_close = 10 SECS; + +/* RFC1122 says the R2 limit should be at least 100 seconds. + Linux uses 15 packets as limit, which corresponds + to ~13-30min depending on RTO. */ +unsigned long nf_ct_tcp_timeout_max_retrans = 5 MINS; + +static unsigned long * tcp_timeouts[] += { NULL, /* TCP_CONNTRACK_NONE */ + &nf_ct_tcp_timeout_syn_sent, /* TCP_CONNTRACK_SYN_SENT, */ + &nf_ct_tcp_timeout_syn_recv, /* TCP_CONNTRACK_SYN_RECV, */ + &nf_ct_tcp_timeout_established, /* TCP_CONNTRACK_ESTABLISHED, */ + &nf_ct_tcp_timeout_fin_wait, /* TCP_CONNTRACK_FIN_WAIT, */ + &nf_ct_tcp_timeout_close_wait, /* TCP_CONNTRACK_CLOSE_WAIT, */ + &nf_ct_tcp_timeout_last_ack, /* TCP_CONNTRACK_LAST_ACK, */ + &nf_ct_tcp_timeout_time_wait, /* TCP_CONNTRACK_TIME_WAIT, */ + &nf_ct_tcp_timeout_close, /* TCP_CONNTRACK_CLOSE, */ + NULL, /* TCP_CONNTRACK_LISTEN */ + }; + +#define sNO TCP_CONNTRACK_NONE +#define sSS TCP_CONNTRACK_SYN_SENT +#define sSR TCP_CONNTRACK_SYN_RECV +#define sES TCP_CONNTRACK_ESTABLISHED +#define sFW TCP_CONNTRACK_FIN_WAIT +#define sCW TCP_CONNTRACK_CLOSE_WAIT +#define sLA TCP_CONNTRACK_LAST_ACK +#define sTW TCP_CONNTRACK_TIME_WAIT +#define sCL TCP_CONNTRACK_CLOSE +#define sLI TCP_CONNTRACK_LISTEN +#define sIV TCP_CONNTRACK_MAX +#define sIG TCP_CONNTRACK_IGNORE + +/* What TCP flags are set from RST/SYN/FIN/ACK. */ +enum tcp_bit_set { + TCP_SYN_SET, + TCP_SYNACK_SET, + TCP_FIN_SET, + TCP_ACK_SET, + TCP_RST_SET, + TCP_NONE_SET, +}; + +/* + * The TCP state transition table needs a few words... + * + * We are the man in the middle. All the packets go through us + * but might get lost in transit to the destination. + * It is assumed that the destinations can't receive segments + * we haven't seen. + * + * The checked segment is in window, but our windows are *not* + * equivalent with the ones of the sender/receiver. We always + * try to guess the state of the current sender. + * + * The meaning of the states are: + * + * NONE: initial state + * SYN_SENT: SYN-only packet seen + * SYN_RECV: SYN-ACK packet seen + * ESTABLISHED: ACK packet seen + * FIN_WAIT: FIN packet seen + * CLOSE_WAIT: ACK seen (after FIN) + * LAST_ACK: FIN seen (after FIN) + * TIME_WAIT: last ACK seen + * CLOSE: closed connection + * + * LISTEN state is not used. + * + * Packets marked as IGNORED (sIG): + * if they may be either invalid or valid + * and the receiver may send back a connection + * closing RST or a SYN/ACK. + * + * Packets marked as INVALID (sIV): + * if they are invalid + * or we do not support the request (simultaneous open) + */ +static enum tcp_conntrack tcp_conntracks[2][6][TCP_CONNTRACK_MAX] = { + { +/* ORIGINAL */ +/* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sLI */ +/*syn*/ { sSS, sSS, sIG, sIG, sIG, sIG, sIG, sSS, sSS, sIV }, +/* + * sNO -> sSS Initialize a new connection + * sSS -> sSS Retransmitted SYN + * sSR -> sIG Late retransmitted SYN? + * sES -> sIG Error: SYNs in window outside the SYN_SENT state + * are errors. Receiver will reply with RST + * and close the connection. + * Or we are not in sync and hold a dead connection. + * sFW -> sIG + * sCW -> sIG + * sLA -> sIG + * sTW -> sSS Reopened connection (RFC 1122). + * sCL -> sSS + */ +/* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sLI */ +/*synack*/ { sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV }, +/* + * A SYN/ACK from the client is always invalid: + * - either it tries to set up a simultaneous open, which is + * not supported; + * - or the firewall has just been inserted between the two hosts + * during the session set-up. The SYN will be retransmitted + * by the true client (or it'll time out). + */ +/* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sLI */ +/*fin*/ { sIV, sIV, sFW, sFW, sLA, sLA, sLA, sTW, sCL, sIV }, +/* + * sNO -> sIV Too late and no reason to do anything... + * sSS -> sIV Client migth not send FIN in this state: + * we enforce waiting for a SYN/ACK reply first. + * sSR -> sFW Close started. + * sES -> sFW + * sFW -> sLA FIN seen in both directions, waiting for + * the last ACK. + * Migth be a retransmitted FIN as well... + * sCW -> sLA + * sLA -> sLA Retransmitted FIN. Remain in the same state. + * sTW -> sTW + * sCL -> sCL + */ +/* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sLI */ +/*ack*/ { sES, sIV, sES, sES, sCW, sCW, sTW, sTW, sCL, sIV }, +/* + * sNO -> sES Assumed. + * sSS -> sIV ACK is invalid: we haven't seen a SYN/ACK yet. + * sSR -> sES Established state is reached. + * sES -> sES :-) + * sFW -> sCW Normal close request answered by ACK. + * sCW -> sCW + * sLA -> sTW Last ACK detected. + * sTW -> sTW Retransmitted last ACK. Remain in the same state. + * sCL -> sCL + */ +/* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sLI */ +/*rst*/ { sIV, sCL, sCL, sCL, sCL, sCL, sCL, sCL, sCL, sIV }, +/*none*/ { sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV } + }, + { +/* REPLY */ +/* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sLI */ +/*syn*/ { sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV }, +/* + * sNO -> sIV Never reached. + * sSS -> sIV Simultaneous open, not supported + * sSR -> sIV Simultaneous open, not supported. + * sES -> sIV Server may not initiate a connection. + * sFW -> sIV + * sCW -> sIV + * sLA -> sIV + * sTW -> sIV Reopened connection, but server may not do it. + * sCL -> sIV + */ +/* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sLI */ +/*synack*/ { sIV, sSR, sSR, sIG, sIG, sIG, sIG, sIG, sIG, sIV }, +/* + * sSS -> sSR Standard open. + * sSR -> sSR Retransmitted SYN/ACK. + * sES -> sIG Late retransmitted SYN/ACK? + * sFW -> sIG Might be SYN/ACK answering ignored SYN + * sCW -> sIG + * sLA -> sIG + * sTW -> sIG + * sCL -> sIG + */ +/* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sLI */ +/*fin*/ { sIV, sIV, sFW, sFW, sLA, sLA, sLA, sTW, sCL, sIV }, +/* + * sSS -> sIV Server might not send FIN in this state. + * sSR -> sFW Close started. + * sES -> sFW + * sFW -> sLA FIN seen in both directions. + * sCW -> sLA + * sLA -> sLA Retransmitted FIN. + * sTW -> sTW + * sCL -> sCL + */ +/* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sLI */ +/*ack*/ { sIV, sIV, sSR, sES, sCW, sCW, sTW, sTW, sCL, sIV }, +/* + * sSS -> sIV Might be a half-open connection. + * sSR -> sSR Might answer late resent SYN. + * sES -> sES :-) + * sFW -> sCW Normal close request answered by ACK. + * sCW -> sCW + * sLA -> sTW Last ACK detected. + * sTW -> sTW Retransmitted last ACK. + * sCL -> sCL + */ +/* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sLI */ +/*rst*/ { sIV, sCL, sCL, sCL, sCL, sCL, sCL, sCL, sCL, sIV }, +/*none*/ { sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV } + } +}; + +static int tcp_pkt_to_tuple(const struct sk_buff *skb, + unsigned int dataoff, + struct nf_conntrack_tuple *tuple) +{ + struct tcphdr _hdr, *hp; + + /* Actually only need first 8 bytes. */ + hp = skb_header_pointer(skb, dataoff, 8, &_hdr); + if (hp == NULL) + return 0; + + tuple->src.u.tcp.port = hp->source; + tuple->dst.u.tcp.port = hp->dest; + + return 1; +} + +static int tcp_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + tuple->src.u.tcp.port = orig->dst.u.tcp.port; + tuple->dst.u.tcp.port = orig->src.u.tcp.port; + return 1; +} + +/* Print out the per-protocol part of the tuple. */ +static int tcp_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return seq_printf(s, "sport=%hu dport=%hu ", + ntohs(tuple->src.u.tcp.port), + ntohs(tuple->dst.u.tcp.port)); +} + +/* Print out the private part of the conntrack. */ +static int tcp_print_conntrack(struct seq_file *s, + const struct nf_conn *conntrack) +{ + enum tcp_conntrack state; + + read_lock_bh(&tcp_lock); + state = conntrack->proto.tcp.state; + read_unlock_bh(&tcp_lock); + + return seq_printf(s, "%s ", tcp_conntrack_names[state]); +} + +static unsigned int get_conntrack_index(const struct tcphdr *tcph) +{ + if (tcph->rst) return TCP_RST_SET; + else if (tcph->syn) return (tcph->ack ? TCP_SYNACK_SET : TCP_SYN_SET); + else if (tcph->fin) return TCP_FIN_SET; + else if (tcph->ack) return TCP_ACK_SET; + else return TCP_NONE_SET; +} + +/* TCP connection tracking based on 'Real Stateful TCP Packet Filtering + in IP Filter' by Guido van Rooij. + + http://www.nluug.nl/events/sane2000/papers.html + http://www.iae.nl/users/guido/papers/tcp_filtering.ps.gz + + The boundaries and the conditions are changed according to RFC793: + the packet must intersect the window (i.e. segments may be + after the right or before the left edge) and thus receivers may ACK + segments after the right edge of the window. + + td_maxend = max(sack + max(win,1)) seen in reply packets + td_maxwin = max(max(win, 1)) + (sack - ack) seen in sent packets + td_maxwin += seq + len - sender.td_maxend + if seq + len > sender.td_maxend + td_end = max(seq + len) seen in sent packets + + I. Upper bound for valid data: seq <= sender.td_maxend + II. Lower bound for valid data: seq + len >= sender.td_end - receiver.td_maxwin + III. Upper bound for valid ack: sack <= receiver.td_end + IV. Lower bound for valid ack: ack >= receiver.td_end - MAXACKWINDOW + + where sack is the highest right edge of sack block found in the packet. + + The upper bound limit for a valid ack is not ignored - + we doesn't have to deal with fragments. +*/ + +static inline __u32 segment_seq_plus_len(__u32 seq, + size_t len, + unsigned int dataoff, + struct tcphdr *tcph) +{ + /* XXX Should I use payload length field in IP/IPv6 header ? + * - YK */ + return (seq + len - dataoff - tcph->doff*4 + + (tcph->syn ? 1 : 0) + (tcph->fin ? 1 : 0)); +} + +/* Fixme: what about big packets? */ +#define MAXACKWINCONST 66000 +#define MAXACKWINDOW(sender) \ + ((sender)->td_maxwin > MAXACKWINCONST ? (sender)->td_maxwin \ + : MAXACKWINCONST) + +/* + * Simplified tcp_parse_options routine from tcp_input.c + */ +static void tcp_options(const struct sk_buff *skb, + unsigned int dataoff, + struct tcphdr *tcph, + struct ip_ct_tcp_state *state) +{ + unsigned char buff[(15 * 4) - sizeof(struct tcphdr)]; + unsigned char *ptr; + int length = (tcph->doff*4) - sizeof(struct tcphdr); + + if (!length) + return; + + ptr = skb_header_pointer(skb, dataoff + sizeof(struct tcphdr), + length, buff); + BUG_ON(ptr == NULL); + + state->td_scale = + state->flags = 0; + + while (length > 0) { + int opcode=*ptr++; + int opsize; + + switch (opcode) { + case TCPOPT_EOL: + return; + case TCPOPT_NOP: /* Ref: RFC 793 section 3.1 */ + length--; + continue; + default: + opsize=*ptr++; + if (opsize < 2) /* "silly options" */ + return; + if (opsize > length) + break; /* don't parse partial options */ + + if (opcode == TCPOPT_SACK_PERM + && opsize == TCPOLEN_SACK_PERM) + state->flags |= IP_CT_TCP_FLAG_SACK_PERM; + else if (opcode == TCPOPT_WINDOW + && opsize == TCPOLEN_WINDOW) { + state->td_scale = *(u_int8_t *)ptr; + + if (state->td_scale > 14) { + /* See RFC1323 */ + state->td_scale = 14; + } + state->flags |= + IP_CT_TCP_FLAG_WINDOW_SCALE; + } + ptr += opsize - 2; + length -= opsize; + } + } +} + +static void tcp_sack(const struct sk_buff *skb, unsigned int dataoff, + struct tcphdr *tcph, __u32 *sack) +{ + unsigned char buff[(15 * 4) - sizeof(struct tcphdr)]; + unsigned char *ptr; + int length = (tcph->doff*4) - sizeof(struct tcphdr); + __u32 tmp; + + if (!length) + return; + + ptr = skb_header_pointer(skb, dataoff + sizeof(struct tcphdr), + length, buff); + BUG_ON(ptr == NULL); + + /* Fast path for timestamp-only option */ + if (length == TCPOLEN_TSTAMP_ALIGNED*4 + && *(__u32 *)ptr == + __constant_ntohl((TCPOPT_NOP << 24) + | (TCPOPT_NOP << 16) + | (TCPOPT_TIMESTAMP << 8) + | TCPOLEN_TIMESTAMP)) + return; + + while (length > 0) { + int opcode = *ptr++; + int opsize, i; + + switch (opcode) { + case TCPOPT_EOL: + return; + case TCPOPT_NOP: /* Ref: RFC 793 section 3.1 */ + length--; + continue; + default: + opsize = *ptr++; + if (opsize < 2) /* "silly options" */ + return; + if (opsize > length) + break; /* don't parse partial options */ + + if (opcode == TCPOPT_SACK + && opsize >= (TCPOLEN_SACK_BASE + + TCPOLEN_SACK_PERBLOCK) + && !((opsize - TCPOLEN_SACK_BASE) + % TCPOLEN_SACK_PERBLOCK)) { + for (i = 0; + i < (opsize - TCPOLEN_SACK_BASE); + i += TCPOLEN_SACK_PERBLOCK) { + memcpy(&tmp, (__u32 *)(ptr + i) + 1, + sizeof(__u32)); + tmp = ntohl(tmp); + + if (after(tmp, *sack)) + *sack = tmp; + } + return; + } + ptr += opsize - 2; + length -= opsize; + } + } +} + +static int tcp_in_window(struct ip_ct_tcp *state, + enum ip_conntrack_dir dir, + unsigned int index, + const struct sk_buff *skb, + unsigned int dataoff, + struct tcphdr *tcph, + int pf) +{ + struct ip_ct_tcp_state *sender = &state->seen[dir]; + struct ip_ct_tcp_state *receiver = &state->seen[!dir]; + __u32 seq, ack, sack, end, win, swin; + int res; + + /* + * Get the required data from the packet. + */ + seq = ntohl(tcph->seq); + ack = sack = ntohl(tcph->ack_seq); + win = ntohs(tcph->window); + end = segment_seq_plus_len(seq, skb->len, dataoff, tcph); + + if (receiver->flags & IP_CT_TCP_FLAG_SACK_PERM) + tcp_sack(skb, dataoff, tcph, &sack); + + DEBUGP("tcp_in_window: START\n"); + DEBUGP("tcp_in_window: src=%u.%u.%u.%u:%hu dst=%u.%u.%u.%u:%hu " + "seq=%u ack=%u sack=%u win=%u end=%u\n", + NIPQUAD(iph->saddr), ntohs(tcph->source), + NIPQUAD(iph->daddr), ntohs(tcph->dest), + seq, ack, sack, win, end); + DEBUGP("tcp_in_window: sender end=%u maxend=%u maxwin=%u scale=%i " + "receiver end=%u maxend=%u maxwin=%u scale=%i\n", + sender->td_end, sender->td_maxend, sender->td_maxwin, + sender->td_scale, + receiver->td_end, receiver->td_maxend, receiver->td_maxwin, + receiver->td_scale); + + if (sender->td_end == 0) { + /* + * Initialize sender data. + */ + if (tcph->syn && tcph->ack) { + /* + * Outgoing SYN-ACK in reply to a SYN. + */ + sender->td_end = + sender->td_maxend = end; + sender->td_maxwin = (win == 0 ? 1 : win); + + tcp_options(skb, dataoff, tcph, sender); + /* + * RFC 1323: + * Both sides must send the Window Scale option + * to enable window scaling in either direction. + */ + if (!(sender->flags & IP_CT_TCP_FLAG_WINDOW_SCALE + && receiver->flags & IP_CT_TCP_FLAG_WINDOW_SCALE)) + sender->td_scale = + receiver->td_scale = 0; + } else { + /* + * We are in the middle of a connection, + * its history is lost for us. + * Let's try to use the data from the packet. + */ + sender->td_end = end; + sender->td_maxwin = (win == 0 ? 1 : win); + sender->td_maxend = end + sender->td_maxwin; + } + } else if (((state->state == TCP_CONNTRACK_SYN_SENT + && dir == IP_CT_DIR_ORIGINAL) + || (state->state == TCP_CONNTRACK_SYN_RECV + && dir == IP_CT_DIR_REPLY)) + && after(end, sender->td_end)) { + /* + * RFC 793: "if a TCP is reinitialized ... then it need + * not wait at all; it must only be sure to use sequence + * numbers larger than those recently used." + */ + sender->td_end = + sender->td_maxend = end; + sender->td_maxwin = (win == 0 ? 1 : win); + + tcp_options(skb, dataoff, tcph, sender); + } + + if (!(tcph->ack)) { + /* + * If there is no ACK, just pretend it was set and OK. + */ + ack = sack = receiver->td_end; + } else if (((tcp_flag_word(tcph) & (TCP_FLAG_ACK|TCP_FLAG_RST)) == + (TCP_FLAG_ACK|TCP_FLAG_RST)) + && (ack == 0)) { + /* + * Broken TCP stacks, that set ACK in RST packets as well + * with zero ack value. + */ + ack = sack = receiver->td_end; + } + + if (seq == end + && (!tcph->rst + || (seq == 0 && state->state == TCP_CONNTRACK_SYN_SENT))) + /* + * Packets contains no data: we assume it is valid + * and check the ack value only. + * However RST segments are always validated by their + * SEQ number, except when seq == 0 (reset sent answering + * SYN. + */ + seq = end = sender->td_end; + + DEBUGP("tcp_in_window: src=%u.%u.%u.%u:%hu dst=%u.%u.%u.%u:%hu " + "seq=%u ack=%u sack =%u win=%u end=%u\n", + NIPQUAD(iph->saddr), ntohs(tcph->source), + NIPQUAD(iph->daddr), ntohs(tcph->dest), + seq, ack, sack, win, end); + DEBUGP("tcp_in_window: sender end=%u maxend=%u maxwin=%u scale=%i " + "receiver end=%u maxend=%u maxwin=%u scale=%i\n", + sender->td_end, sender->td_maxend, sender->td_maxwin, + sender->td_scale, + receiver->td_end, receiver->td_maxend, receiver->td_maxwin, + receiver->td_scale); + + DEBUGP("tcp_in_window: I=%i II=%i III=%i IV=%i\n", + before(seq, sender->td_maxend + 1), + after(end, sender->td_end - receiver->td_maxwin - 1), + before(sack, receiver->td_end + 1), + after(ack, receiver->td_end - MAXACKWINDOW(sender))); + + if (sender->loose || receiver->loose || + (before(seq, sender->td_maxend + 1) && + after(end, sender->td_end - receiver->td_maxwin - 1) && + before(sack, receiver->td_end + 1) && + after(ack, receiver->td_end - MAXACKWINDOW(sender)))) { + /* + * Take into account window scaling (RFC 1323). + */ + if (!tcph->syn) + win <<= sender->td_scale; + + /* + * Update sender data. + */ + swin = win + (sack - ack); + if (sender->td_maxwin < swin) + sender->td_maxwin = swin; + if (after(end, sender->td_end)) + sender->td_end = end; + /* + * Update receiver data. + */ + if (after(end, sender->td_maxend)) + receiver->td_maxwin += end - sender->td_maxend; + if (after(sack + win, receiver->td_maxend - 1)) { + receiver->td_maxend = sack + win; + if (win == 0) + receiver->td_maxend++; + } + + /* + * Check retransmissions. + */ + if (index == TCP_ACK_SET) { + if (state->last_dir == dir + && state->last_seq == seq + && state->last_ack == ack + && state->last_end == end) + state->retrans++; + else { + state->last_dir = dir; + state->last_seq = seq; + state->last_ack = ack; + state->last_end = end; + state->retrans = 0; + } + } + /* + * Close the window of disabled window tracking :-) + */ + if (sender->loose) + sender->loose--; + + res = 1; + } else { + if (LOG_INVALID(IPPROTO_TCP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_tcp: %s ", + before(seq, sender->td_maxend + 1) ? + after(end, sender->td_end - receiver->td_maxwin - 1) ? + before(sack, receiver->td_end + 1) ? + after(ack, receiver->td_end - MAXACKWINDOW(sender)) ? "BUG" + : "ACK is under the lower bound (possible overly delayed ACK)" + : "ACK is over the upper bound (ACKed data not seen yet)" + : "SEQ is under the lower bound (already ACKed data retransmitted)" + : "SEQ is over the upper bound (over the window of the receiver)"); + + res = nf_ct_tcp_be_liberal; + } + + DEBUGP("tcp_in_window: res=%i sender end=%u maxend=%u maxwin=%u " + "receiver end=%u maxend=%u maxwin=%u\n", + res, sender->td_end, sender->td_maxend, sender->td_maxwin, + receiver->td_end, receiver->td_maxend, receiver->td_maxwin); + + return res; +} + +#ifdef CONFIG_IP_NF_NAT_NEEDED +/* Update sender->td_end after NAT successfully mangled the packet */ +/* Caller must linearize skb at tcp header. */ +void nf_conntrack_tcp_update(struct sk_buff *skb, + unsigned int dataoff, + struct nf_conn *conntrack, + int dir) +{ + struct tcphdr *tcph = (void *)skb->data + dataoff; + __u32 end; +#ifdef DEBUGP_VARS + struct ip_ct_tcp_state *sender = &conntrack->proto.tcp.seen[dir]; + struct ip_ct_tcp_state *receiver = &conntrack->proto.tcp.seen[!dir]; +#endif + + end = segment_seq_plus_len(ntohl(tcph->seq), skb->len, dataoff, tcph); + + write_lock_bh(&tcp_lock); + /* + * We have to worry for the ack in the reply packet only... + */ + if (after(end, conntrack->proto.tcp.seen[dir].td_end)) + conntrack->proto.tcp.seen[dir].td_end = end; + conntrack->proto.tcp.last_end = end; + write_unlock_bh(&tcp_lock); + DEBUGP("tcp_update: sender end=%u maxend=%u maxwin=%u scale=%i " + "receiver end=%u maxend=%u maxwin=%u scale=%i\n", + sender->td_end, sender->td_maxend, sender->td_maxwin, + sender->td_scale, + receiver->td_end, receiver->td_maxend, receiver->td_maxwin, + receiver->td_scale); +} + +#endif + +#define TH_FIN 0x01 +#define TH_SYN 0x02 +#define TH_RST 0x04 +#define TH_PUSH 0x08 +#define TH_ACK 0x10 +#define TH_URG 0x20 +#define TH_ECE 0x40 +#define TH_CWR 0x80 + +/* table of valid flag combinations - ECE and CWR are always valid */ +static u8 tcp_valid_flags[(TH_FIN|TH_SYN|TH_RST|TH_PUSH|TH_ACK|TH_URG) + 1] = +{ + [TH_SYN] = 1, + [TH_SYN|TH_ACK] = 1, + [TH_SYN|TH_ACK|TH_PUSH] = 1, + [TH_RST] = 1, + [TH_RST|TH_ACK] = 1, + [TH_RST|TH_ACK|TH_PUSH] = 1, + [TH_FIN|TH_ACK] = 1, + [TH_ACK] = 1, + [TH_ACK|TH_PUSH] = 1, + [TH_ACK|TH_URG] = 1, + [TH_ACK|TH_URG|TH_PUSH] = 1, + [TH_FIN|TH_ACK|TH_PUSH] = 1, + [TH_FIN|TH_ACK|TH_URG] = 1, + [TH_FIN|TH_ACK|TH_URG|TH_PUSH] = 1, +}; + +/* Protect conntrack agaist broken packets. Code taken from ipt_unclean.c. */ +static int tcp_error(struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info *ctinfo, + int pf, + unsigned int hooknum, + int(*csum)(const struct sk_buff *,unsigned int)) +{ + struct tcphdr _tcph, *th; + unsigned int tcplen = skb->len - dataoff; + u_int8_t tcpflags; + + /* Smaller that minimal TCP header? */ + th = skb_header_pointer(skb, dataoff, sizeof(_tcph), &_tcph); + if (th == NULL) { + if (LOG_INVALID(IPPROTO_TCP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_tcp: short packet "); + return -NF_ACCEPT; + } + + /* Not whole TCP header or malformed packet */ + if (th->doff*4 < sizeof(struct tcphdr) || tcplen < th->doff*4) { + if (LOG_INVALID(IPPROTO_TCP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_tcp: truncated/malformed packet "); + return -NF_ACCEPT; + } + + /* Checksum invalid? Ignore. + * We skip checking packets on the outgoing path + * because the semantic of CHECKSUM_HW is different there + * and moreover root might send raw packets. + */ + /* FIXME: Source route IP option packets --RR */ + if (((pf == PF_INET && hooknum == NF_IP_PRE_ROUTING) || + (pf == PF_INET6 && hooknum == NF_IP6_PRE_ROUTING)) + && skb->ip_summed != CHECKSUM_UNNECESSARY + && csum(skb, dataoff)) { + if (LOG_INVALID(IPPROTO_TCP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_tcp: bad TCP checksum "); + return -NF_ACCEPT; + } + + /* Check TCP flags. */ + tcpflags = (((u_int8_t *)th)[13] & ~(TH_ECE|TH_CWR)); + if (!tcp_valid_flags[tcpflags]) { + if (LOG_INVALID(IPPROTO_TCP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_tcp: invalid TCP flag combination "); + return -NF_ACCEPT; + } + + return NF_ACCEPT; +} + +static int csum4(const struct sk_buff *skb, unsigned int dataoff) +{ + return csum_tcpudp_magic(skb->nh.iph->saddr, skb->nh.iph->daddr, + skb->len - dataoff, IPPROTO_TCP, + skb->ip_summed == CHECKSUM_HW ? skb->csum + : skb_checksum(skb, dataoff, + skb->len - dataoff, 0)); +} + +static int csum6(const struct sk_buff *skb, unsigned int dataoff) +{ + return csum_ipv6_magic(&skb->nh.ipv6h->saddr, &skb->nh.ipv6h->daddr, + skb->len - dataoff, IPPROTO_TCP, + skb->ip_summed == CHECKSUM_HW ? skb->csum + : skb_checksum(skb, dataoff, skb->len - dataoff, + 0)); +} + +static int tcp_error4(struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info *ctinfo, + int pf, + unsigned int hooknum) +{ + return tcp_error(skb, dataoff, ctinfo, pf, hooknum, csum4); +} + +static int tcp_error6(struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info *ctinfo, + int pf, + unsigned int hooknum) +{ + return tcp_error(skb, dataoff, ctinfo, pf, hooknum, csum6); +} + +/* Returns verdict for packet, or -1 for invalid. */ +static int tcp_packet(struct nf_conn *conntrack, + const struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info ctinfo, + int pf, + unsigned int hooknum) +{ + enum tcp_conntrack new_state, old_state; + enum ip_conntrack_dir dir; + struct tcphdr *th, _tcph; + unsigned long timeout; + unsigned int index; + + th = skb_header_pointer(skb, dataoff, sizeof(_tcph), &_tcph); + BUG_ON(th == NULL); + + write_lock_bh(&tcp_lock); + old_state = conntrack->proto.tcp.state; + dir = CTINFO2DIR(ctinfo); + index = get_conntrack_index(th); + new_state = tcp_conntracks[dir][index][old_state]; + + switch (new_state) { + case TCP_CONNTRACK_IGNORE: + /* Either SYN in ORIGINAL + * or SYN/ACK in REPLY. */ + if (index == TCP_SYNACK_SET + && conntrack->proto.tcp.last_index == TCP_SYN_SET + && conntrack->proto.tcp.last_dir != dir + && ntohl(th->ack_seq) == + conntrack->proto.tcp.last_end) { + /* This SYN/ACK acknowledges a SYN that we earlier + * ignored as invalid. This means that the client and + * the server are both in sync, while the firewall is + * not. We kill this session and block the SYN/ACK so + * that the client cannot but retransmit its SYN and + * thus initiate a clean new session. + */ + write_unlock_bh(&tcp_lock); + if (LOG_INVALID(IPPROTO_TCP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_tcp: killing out of sync session "); + if (del_timer(&conntrack->timeout)) + conntrack->timeout.function((unsigned long) + conntrack); + return -NF_DROP; + } + conntrack->proto.tcp.last_index = index; + conntrack->proto.tcp.last_dir = dir; + conntrack->proto.tcp.last_seq = ntohl(th->seq); + conntrack->proto.tcp.last_end = + segment_seq_plus_len(ntohl(th->seq), skb->len, dataoff, th); + + write_unlock_bh(&tcp_lock); + if (LOG_INVALID(IPPROTO_TCP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_tcp: invalid packed ignored "); + return NF_ACCEPT; + case TCP_CONNTRACK_MAX: + /* Invalid packet */ + DEBUGP("nf_ct_tcp: Invalid dir=%i index=%u ostate=%u\n", + dir, get_conntrack_index(th), + old_state); + write_unlock_bh(&tcp_lock); + if (LOG_INVALID(IPPROTO_TCP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_tcp: invalid state "); + return -NF_ACCEPT; + case TCP_CONNTRACK_SYN_SENT: + if (old_state < TCP_CONNTRACK_TIME_WAIT) + break; + if ((conntrack->proto.tcp.seen[dir].flags & + IP_CT_TCP_FLAG_CLOSE_INIT) + || after(ntohl(th->seq), + conntrack->proto.tcp.seen[dir].td_end)) { + /* Attempt to reopen a closed connection. + * Delete this connection and look up again. */ + write_unlock_bh(&tcp_lock); + if (del_timer(&conntrack->timeout)) + conntrack->timeout.function((unsigned long) + conntrack); + return -NF_REPEAT; + } + case TCP_CONNTRACK_CLOSE: + if (index == TCP_RST_SET + && test_bit(IPS_SEEN_REPLY_BIT, &conntrack->status) + && conntrack->proto.tcp.last_index == TCP_SYN_SET + && ntohl(th->ack_seq) == conntrack->proto.tcp.last_end) { + /* RST sent to invalid SYN we had let trough + * SYN was in window then, tear down connection. + * We skip window checking, because packet might ACK + * segments we ignored in the SYN. */ + goto in_window; + } + /* Just fall trough */ + default: + /* Keep compilers happy. */ + break; + } + + if (!tcp_in_window(&conntrack->proto.tcp, dir, index, + skb, dataoff, th, pf)) { + write_unlock_bh(&tcp_lock); + return -NF_ACCEPT; + } + in_window: + /* From now on we have got in-window packets */ + conntrack->proto.tcp.last_index = index; + + DEBUGP("tcp_conntracks: src=%u.%u.%u.%u:%hu dst=%u.%u.%u.%u:%hu " + "syn=%i ack=%i fin=%i rst=%i old=%i new=%i\n", + NIPQUAD(iph->saddr), ntohs(th->source), + NIPQUAD(iph->daddr), ntohs(th->dest), + (th->syn ? 1 : 0), (th->ack ? 1 : 0), + (th->fin ? 1 : 0), (th->rst ? 1 : 0), + old_state, new_state); + + conntrack->proto.tcp.state = new_state; + if (old_state != new_state + && (new_state == TCP_CONNTRACK_FIN_WAIT + || new_state == TCP_CONNTRACK_CLOSE)) + conntrack->proto.tcp.seen[dir].flags |= IP_CT_TCP_FLAG_CLOSE_INIT; + timeout = conntrack->proto.tcp.retrans >= nf_ct_tcp_max_retrans + && *tcp_timeouts[new_state] > nf_ct_tcp_timeout_max_retrans + ? nf_ct_tcp_timeout_max_retrans : *tcp_timeouts[new_state]; + write_unlock_bh(&tcp_lock); + + nf_conntrack_event_cache(IPCT_PROTOINFO_VOLATILE, skb); + if (new_state != old_state) + nf_conntrack_event_cache(IPCT_PROTOINFO, skb); + + if (!test_bit(IPS_SEEN_REPLY_BIT, &conntrack->status)) { + /* If only reply is a RST, we can consider ourselves not to + have an established connection: this is a fairly common + problem case, so we can delete the conntrack + immediately. --RR */ + if (th->rst) { + if (del_timer(&conntrack->timeout)) + conntrack->timeout.function((unsigned long) + conntrack); + return NF_ACCEPT; + } + } else if (!test_bit(IPS_ASSURED_BIT, &conntrack->status) + && (old_state == TCP_CONNTRACK_SYN_RECV + || old_state == TCP_CONNTRACK_ESTABLISHED) + && new_state == TCP_CONNTRACK_ESTABLISHED) { + /* Set ASSURED if we see see valid ack in ESTABLISHED + after SYN_RECV or a valid answer for a picked up + connection. */ + set_bit(IPS_ASSURED_BIT, &conntrack->status); + nf_conntrack_event_cache(IPCT_STATUS, skb); + } + nf_ct_refresh_acct(conntrack, ctinfo, skb, timeout); + + return NF_ACCEPT; +} + +/* Called when a new connection for this protocol found. */ +static int tcp_new(struct nf_conn *conntrack, + const struct sk_buff *skb, + unsigned int dataoff) +{ + enum tcp_conntrack new_state; + struct tcphdr *th, _tcph; +#ifdef DEBUGP_VARS + struct ip_ct_tcp_state *sender = &conntrack->proto.tcp.seen[0]; + struct ip_ct_tcp_state *receiver = &conntrack->proto.tcp.seen[1]; +#endif + + th = skb_header_pointer(skb, dataoff, sizeof(_tcph), &_tcph); + BUG_ON(th == NULL); + + /* Don't need lock here: this conntrack not in circulation yet */ + new_state + = tcp_conntracks[0][get_conntrack_index(th)] + [TCP_CONNTRACK_NONE]; + + /* Invalid: delete conntrack */ + if (new_state >= TCP_CONNTRACK_MAX) { + DEBUGP("nf_ct_tcp: invalid new deleting.\n"); + return 0; + } + + if (new_state == TCP_CONNTRACK_SYN_SENT) { + /* SYN packet */ + conntrack->proto.tcp.seen[0].td_end = + segment_seq_plus_len(ntohl(th->seq), skb->len, + dataoff, th); + conntrack->proto.tcp.seen[0].td_maxwin = ntohs(th->window); + if (conntrack->proto.tcp.seen[0].td_maxwin == 0) + conntrack->proto.tcp.seen[0].td_maxwin = 1; + conntrack->proto.tcp.seen[0].td_maxend = + conntrack->proto.tcp.seen[0].td_end; + + tcp_options(skb, dataoff, th, &conntrack->proto.tcp.seen[0]); + conntrack->proto.tcp.seen[1].flags = 0; + conntrack->proto.tcp.seen[0].loose = + conntrack->proto.tcp.seen[1].loose = 0; + } else if (nf_ct_tcp_loose == 0) { + /* Don't try to pick up connections. */ + return 0; + } else { + /* + * We are in the middle of a connection, + * its history is lost for us. + * Let's try to use the data from the packet. + */ + conntrack->proto.tcp.seen[0].td_end = + segment_seq_plus_len(ntohl(th->seq), skb->len, + dataoff, th); + conntrack->proto.tcp.seen[0].td_maxwin = ntohs(th->window); + if (conntrack->proto.tcp.seen[0].td_maxwin == 0) + conntrack->proto.tcp.seen[0].td_maxwin = 1; + conntrack->proto.tcp.seen[0].td_maxend = + conntrack->proto.tcp.seen[0].td_end + + conntrack->proto.tcp.seen[0].td_maxwin; + conntrack->proto.tcp.seen[0].td_scale = 0; + + /* We assume SACK. Should we assume window scaling too? */ + conntrack->proto.tcp.seen[0].flags = + conntrack->proto.tcp.seen[1].flags = IP_CT_TCP_FLAG_SACK_PERM; + conntrack->proto.tcp.seen[0].loose = + conntrack->proto.tcp.seen[1].loose = nf_ct_tcp_loose; + } + + conntrack->proto.tcp.seen[1].td_end = 0; + conntrack->proto.tcp.seen[1].td_maxend = 0; + conntrack->proto.tcp.seen[1].td_maxwin = 1; + conntrack->proto.tcp.seen[1].td_scale = 0; + + /* tcp_packet will set them */ + conntrack->proto.tcp.state = TCP_CONNTRACK_NONE; + conntrack->proto.tcp.last_index = TCP_NONE_SET; + + DEBUGP("tcp_new: sender end=%u maxend=%u maxwin=%u scale=%i " + "receiver end=%u maxend=%u maxwin=%u scale=%i\n", + sender->td_end, sender->td_maxend, sender->td_maxwin, + sender->td_scale, + receiver->td_end, receiver->td_maxend, receiver->td_maxwin, + receiver->td_scale); + return 1; +} + +struct nf_conntrack_protocol nf_conntrack_protocol_tcp4 = +{ + .l3proto = PF_INET, + .proto = IPPROTO_TCP, + .name = "tcp", + .pkt_to_tuple = tcp_pkt_to_tuple, + .invert_tuple = tcp_invert_tuple, + .print_tuple = tcp_print_tuple, + .print_conntrack = tcp_print_conntrack, + .packet = tcp_packet, + .new = tcp_new, + .error = tcp_error4, +}; + +struct nf_conntrack_protocol nf_conntrack_protocol_tcp6 = +{ + .l3proto = PF_INET6, + .proto = IPPROTO_TCP, + .name = "tcp", + .pkt_to_tuple = tcp_pkt_to_tuple, + .invert_tuple = tcp_invert_tuple, + .print_tuple = tcp_print_tuple, + .print_conntrack = tcp_print_conntrack, + .packet = tcp_packet, + .new = tcp_new, + .error = tcp_error6, +}; + +EXPORT_SYMBOL(nf_conntrack_protocol_tcp4); +EXPORT_SYMBOL(nf_conntrack_protocol_tcp6); diff --git a/net/netfilter/nf_conntrack_proto_udp.c b/net/netfilter/nf_conntrack_proto_udp.c new file mode 100644 index 00000000000..3cae7ce420d --- /dev/null +++ b/net/netfilter/nf_conntrack_proto_udp.c @@ -0,0 +1,216 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - enable working with Layer 3 protocol independent connection tracking. + * + * Derived from net/ipv4/netfilter/ip_conntrack_proto_udp.c + */ + +#include <linux/types.h> +#include <linux/sched.h> +#include <linux/timer.h> +#include <linux/module.h> +#include <linux/netfilter.h> +#include <linux/udp.h> +#include <linux/seq_file.h> +#include <linux/skbuff.h> +#include <linux/ipv6.h> +#include <net/ip6_checksum.h> +#include <net/checksum.h> +#include <linux/netfilter.h> +#include <linux/netfilter_ipv4.h> +#include <linux/netfilter_ipv6.h> +#include <net/netfilter/nf_conntrack_protocol.h> + +unsigned long nf_ct_udp_timeout = 30*HZ; +unsigned long nf_ct_udp_timeout_stream = 180*HZ; + +static int udp_pkt_to_tuple(const struct sk_buff *skb, + unsigned int dataoff, + struct nf_conntrack_tuple *tuple) +{ + struct udphdr _hdr, *hp; + + /* Actually only need first 8 bytes. */ + hp = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr); + if (hp == NULL) + return 0; + + tuple->src.u.udp.port = hp->source; + tuple->dst.u.udp.port = hp->dest; + + return 1; +} + +static int udp_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + tuple->src.u.udp.port = orig->dst.u.udp.port; + tuple->dst.u.udp.port = orig->src.u.udp.port; + return 1; +} + +/* Print out the per-protocol part of the tuple. */ +static int udp_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return seq_printf(s, "sport=%hu dport=%hu ", + ntohs(tuple->src.u.udp.port), + ntohs(tuple->dst.u.udp.port)); +} + +/* Print out the private part of the conntrack. */ +static int udp_print_conntrack(struct seq_file *s, + const struct nf_conn *conntrack) +{ + return 0; +} + +/* Returns verdict for packet, and may modify conntracktype */ +static int udp_packet(struct nf_conn *conntrack, + const struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info ctinfo, + int pf, + unsigned int hooknum) +{ + /* If we've seen traffic both ways, this is some kind of UDP + stream. Extend timeout. */ + if (test_bit(IPS_SEEN_REPLY_BIT, &conntrack->status)) { + nf_ct_refresh_acct(conntrack, ctinfo, skb, + nf_ct_udp_timeout_stream); + /* Also, more likely to be important, and not a probe */ + if (!test_and_set_bit(IPS_ASSURED_BIT, &conntrack->status)) + nf_conntrack_event_cache(IPCT_STATUS, skb); + } else + nf_ct_refresh_acct(conntrack, ctinfo, skb, nf_ct_udp_timeout); + + return NF_ACCEPT; +} + +/* Called when a new connection for this protocol found. */ +static int udp_new(struct nf_conn *conntrack, const struct sk_buff *skb, + unsigned int dataoff) +{ + return 1; +} + +static int udp_error(struct sk_buff *skb, unsigned int dataoff, + enum ip_conntrack_info *ctinfo, + int pf, + unsigned int hooknum, + int (*csum)(const struct sk_buff *, unsigned int)) +{ + unsigned int udplen = skb->len - dataoff; + struct udphdr _hdr, *hdr; + + /* Header is too small? */ + hdr = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr); + if (hdr == NULL) { + if (LOG_INVALID(IPPROTO_UDP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_udp: short packet "); + return -NF_ACCEPT; + } + + /* Truncated/malformed packets */ + if (ntohs(hdr->len) > udplen || ntohs(hdr->len) < sizeof(*hdr)) { + if (LOG_INVALID(IPPROTO_UDP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_udp: truncated/malformed packet "); + return -NF_ACCEPT; + } + + /* Packet with no checksum */ + if (!hdr->check) + return NF_ACCEPT; + + /* Checksum invalid? Ignore. + * We skip checking packets on the outgoing path + * because the semantic of CHECKSUM_HW is different there + * and moreover root might send raw packets. + * FIXME: Source route IP option packets --RR */ + if (((pf == PF_INET && hooknum == NF_IP_PRE_ROUTING) || + (pf == PF_INET6 && hooknum == NF_IP6_PRE_ROUTING)) + && skb->ip_summed != CHECKSUM_UNNECESSARY + && csum(skb, dataoff)) { + if (LOG_INVALID(IPPROTO_UDP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_udp: bad UDP checksum "); + return -NF_ACCEPT; + } + + return NF_ACCEPT; +} + +static int csum4(const struct sk_buff *skb, unsigned int dataoff) +{ + return csum_tcpudp_magic(skb->nh.iph->saddr, skb->nh.iph->daddr, + skb->len - dataoff, IPPROTO_UDP, + skb->ip_summed == CHECKSUM_HW ? skb->csum + : skb_checksum(skb, dataoff, + skb->len - dataoff, 0)); +} + +static int csum6(const struct sk_buff *skb, unsigned int dataoff) +{ + return csum_ipv6_magic(&skb->nh.ipv6h->saddr, &skb->nh.ipv6h->daddr, + skb->len - dataoff, IPPROTO_UDP, + skb->ip_summed == CHECKSUM_HW ? skb->csum + : skb_checksum(skb, dataoff, skb->len - dataoff, + 0)); +} + +static int udp_error4(struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info *ctinfo, + int pf, + unsigned int hooknum) +{ + return udp_error(skb, dataoff, ctinfo, pf, hooknum, csum4); +} + +static int udp_error6(struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info *ctinfo, + int pf, + unsigned int hooknum) +{ + return udp_error(skb, dataoff, ctinfo, pf, hooknum, csum6); +} + +struct nf_conntrack_protocol nf_conntrack_protocol_udp4 = +{ + .l3proto = PF_INET, + .proto = IPPROTO_UDP, + .name = "udp", + .pkt_to_tuple = udp_pkt_to_tuple, + .invert_tuple = udp_invert_tuple, + .print_tuple = udp_print_tuple, + .print_conntrack = udp_print_conntrack, + .packet = udp_packet, + .new = udp_new, + .error = udp_error4, +}; + +struct nf_conntrack_protocol nf_conntrack_protocol_udp6 = +{ + .l3proto = PF_INET6, + .proto = IPPROTO_UDP, + .name = "udp", + .pkt_to_tuple = udp_pkt_to_tuple, + .invert_tuple = udp_invert_tuple, + .print_tuple = udp_print_tuple, + .print_conntrack = udp_print_conntrack, + .packet = udp_packet, + .new = udp_new, + .error = udp_error6, +}; + +EXPORT_SYMBOL(nf_conntrack_protocol_udp4); +EXPORT_SYMBOL(nf_conntrack_protocol_udp6); diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c new file mode 100644 index 00000000000..45224db4fe2 --- /dev/null +++ b/net/netfilter/nf_conntrack_standalone.c @@ -0,0 +1,869 @@ +/* This file contains all the functions required for the standalone + nf_conntrack module. + + These are not required by the compatibility layer. +*/ + +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - generalize L3 protocol dependent part. + * + * Derived from net/ipv4/netfilter/ip_conntrack_standalone.c + */ + +#include <linux/config.h> +#include <linux/types.h> +#include <linux/netfilter.h> +#include <linux/module.h> +#include <linux/skbuff.h> +#include <linux/proc_fs.h> +#include <linux/seq_file.h> +#include <linux/percpu.h> +#include <linux/netdevice.h> +#ifdef CONFIG_SYSCTL +#include <linux/sysctl.h> +#endif + +#define ASSERT_READ_LOCK(x) +#define ASSERT_WRITE_LOCK(x) + +#include <net/netfilter/nf_conntrack.h> +#include <net/netfilter/nf_conntrack_l3proto.h> +#include <net/netfilter/nf_conntrack_protocol.h> +#include <net/netfilter/nf_conntrack_core.h> +#include <net/netfilter/nf_conntrack_helper.h> +#include <linux/netfilter_ipv4/listhelp.h> + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +MODULE_LICENSE("GPL"); + +extern atomic_t nf_conntrack_count; +DECLARE_PER_CPU(struct ip_conntrack_stat, nf_conntrack_stat); + +static int kill_l3proto(struct nf_conn *i, void *data) +{ + return (i->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num == + ((struct nf_conntrack_l3proto *)data)->l3proto); +} + +static int kill_proto(struct nf_conn *i, void *data) +{ + struct nf_conntrack_protocol *proto; + proto = (struct nf_conntrack_protocol *)data; + return (i->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum == + proto->proto) && + (i->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num == + proto->l3proto); +} + +#ifdef CONFIG_PROC_FS +static int +print_tuple(struct seq_file *s, const struct nf_conntrack_tuple *tuple, + struct nf_conntrack_l3proto *l3proto, + struct nf_conntrack_protocol *proto) +{ + return l3proto->print_tuple(s, tuple) || proto->print_tuple(s, tuple); +} + +#ifdef CONFIG_NF_CT_ACCT +static unsigned int +seq_print_counters(struct seq_file *s, + const struct ip_conntrack_counter *counter) +{ + return seq_printf(s, "packets=%llu bytes=%llu ", + (unsigned long long)counter->packets, + (unsigned long long)counter->bytes); +} +#else +#define seq_print_counters(x, y) 0 +#endif + +struct ct_iter_state { + unsigned int bucket; +}; + +static struct list_head *ct_get_first(struct seq_file *seq) +{ + struct ct_iter_state *st = seq->private; + + for (st->bucket = 0; + st->bucket < nf_conntrack_htable_size; + st->bucket++) { + if (!list_empty(&nf_conntrack_hash[st->bucket])) + return nf_conntrack_hash[st->bucket].next; + } + return NULL; +} + +static struct list_head *ct_get_next(struct seq_file *seq, struct list_head *head) +{ + struct ct_iter_state *st = seq->private; + + head = head->next; + while (head == &nf_conntrack_hash[st->bucket]) { + if (++st->bucket >= nf_conntrack_htable_size) + return NULL; + head = nf_conntrack_hash[st->bucket].next; + } + return head; +} + +static struct list_head *ct_get_idx(struct seq_file *seq, loff_t pos) +{ + struct list_head *head = ct_get_first(seq); + + if (head) + while (pos && (head = ct_get_next(seq, head))) + pos--; + return pos ? NULL : head; +} + +static void *ct_seq_start(struct seq_file *seq, loff_t *pos) +{ + read_lock_bh(&nf_conntrack_lock); + return ct_get_idx(seq, *pos); +} + +static void *ct_seq_next(struct seq_file *s, void *v, loff_t *pos) +{ + (*pos)++; + return ct_get_next(s, v); +} + +static void ct_seq_stop(struct seq_file *s, void *v) +{ + read_unlock_bh(&nf_conntrack_lock); +} + +/* return 0 on success, 1 in case of error */ +static int ct_seq_show(struct seq_file *s, void *v) +{ + const struct nf_conntrack_tuple_hash *hash = v; + const struct nf_conn *conntrack = nf_ct_tuplehash_to_ctrack(hash); + struct nf_conntrack_l3proto *l3proto; + struct nf_conntrack_protocol *proto; + + ASSERT_READ_LOCK(&nf_conntrack_lock); + NF_CT_ASSERT(conntrack); + + /* we only want to print DIR_ORIGINAL */ + if (NF_CT_DIRECTION(hash)) + return 0; + + l3proto = nf_ct_find_l3proto(conntrack->tuplehash[IP_CT_DIR_ORIGINAL] + .tuple.src.l3num); + + NF_CT_ASSERT(l3proto); + proto = nf_ct_find_proto(conntrack->tuplehash[IP_CT_DIR_ORIGINAL] + .tuple.src.l3num, + conntrack->tuplehash[IP_CT_DIR_ORIGINAL] + .tuple.dst.protonum); + NF_CT_ASSERT(proto); + + if (seq_printf(s, "%-8s %u %-8s %u %ld ", + l3proto->name, + conntrack->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num, + proto->name, + conntrack->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum, + timer_pending(&conntrack->timeout) + ? (long)(conntrack->timeout.expires - jiffies)/HZ : 0) != 0) + return -ENOSPC; + + if (l3proto->print_conntrack(s, conntrack)) + return -ENOSPC; + + if (proto->print_conntrack(s, conntrack)) + return -ENOSPC; + + if (print_tuple(s, &conntrack->tuplehash[IP_CT_DIR_ORIGINAL].tuple, + l3proto, proto)) + return -ENOSPC; + + if (seq_print_counters(s, &conntrack->counters[IP_CT_DIR_ORIGINAL])) + return -ENOSPC; + + if (!(test_bit(IPS_SEEN_REPLY_BIT, &conntrack->status))) + if (seq_printf(s, "[UNREPLIED] ")) + return -ENOSPC; + + if (print_tuple(s, &conntrack->tuplehash[IP_CT_DIR_REPLY].tuple, + l3proto, proto)) + return -ENOSPC; + + if (seq_print_counters(s, &conntrack->counters[IP_CT_DIR_REPLY])) + return -ENOSPC; + + if (test_bit(IPS_ASSURED_BIT, &conntrack->status)) + if (seq_printf(s, "[ASSURED] ")) + return -ENOSPC; + +#if defined(CONFIG_NF_CONNTRACK_MARK) + if (seq_printf(s, "mark=%u ", conntrack->mark)) + return -ENOSPC; +#endif + + if (seq_printf(s, "use=%u\n", atomic_read(&conntrack->ct_general.use))) + return -ENOSPC; + + return 0; +} + +static struct seq_operations ct_seq_ops = { + .start = ct_seq_start, + .next = ct_seq_next, + .stop = ct_seq_stop, + .show = ct_seq_show +}; + +static int ct_open(struct inode *inode, struct file *file) +{ + struct seq_file *seq; + struct ct_iter_state *st; + int ret; + + st = kmalloc(sizeof(struct ct_iter_state), GFP_KERNEL); + if (st == NULL) + return -ENOMEM; + ret = seq_open(file, &ct_seq_ops); + if (ret) + goto out_free; + seq = file->private_data; + seq->private = st; + memset(st, 0, sizeof(struct ct_iter_state)); + return ret; +out_free: + kfree(st); + return ret; +} + +static struct file_operations ct_file_ops = { + .owner = THIS_MODULE, + .open = ct_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_private, +}; + +/* expects */ +static void *exp_seq_start(struct seq_file *s, loff_t *pos) +{ + struct list_head *e = &nf_conntrack_expect_list; + loff_t i; + + /* strange seq_file api calls stop even if we fail, + * thus we need to grab lock since stop unlocks */ + read_lock_bh(&nf_conntrack_lock); + + if (list_empty(e)) + return NULL; + + for (i = 0; i <= *pos; i++) { + e = e->next; + if (e == &nf_conntrack_expect_list) + return NULL; + } + return e; +} + +static void *exp_seq_next(struct seq_file *s, void *v, loff_t *pos) +{ + struct list_head *e = v; + + ++*pos; + e = e->next; + + if (e == &nf_conntrack_expect_list) + return NULL; + + return e; +} + +static void exp_seq_stop(struct seq_file *s, void *v) +{ + read_unlock_bh(&nf_conntrack_lock); +} + +static int exp_seq_show(struct seq_file *s, void *v) +{ + struct nf_conntrack_expect *expect = v; + + if (expect->timeout.function) + seq_printf(s, "%ld ", timer_pending(&expect->timeout) + ? (long)(expect->timeout.expires - jiffies)/HZ : 0); + else + seq_printf(s, "- "); + seq_printf(s, "l3proto = %u proto=%u ", + expect->tuple.src.l3num, + expect->tuple.dst.protonum); + print_tuple(s, &expect->tuple, + nf_ct_find_l3proto(expect->tuple.src.l3num), + nf_ct_find_proto(expect->tuple.src.l3num, + expect->tuple.dst.protonum)); + return seq_putc(s, '\n'); +} + +static struct seq_operations exp_seq_ops = { + .start = exp_seq_start, + .next = exp_seq_next, + .stop = exp_seq_stop, + .show = exp_seq_show +}; + +static int exp_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &exp_seq_ops); +} + +static struct file_operations exp_file_ops = { + .owner = THIS_MODULE, + .open = exp_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release +}; + +static void *ct_cpu_seq_start(struct seq_file *seq, loff_t *pos) +{ + int cpu; + + if (*pos == 0) + return SEQ_START_TOKEN; + + for (cpu = *pos-1; cpu < NR_CPUS; ++cpu) { + if (!cpu_possible(cpu)) + continue; + *pos = cpu + 1; + return &per_cpu(nf_conntrack_stat, cpu); + } + + return NULL; +} + +static void *ct_cpu_seq_next(struct seq_file *seq, void *v, loff_t *pos) +{ + int cpu; + + for (cpu = *pos; cpu < NR_CPUS; ++cpu) { + if (!cpu_possible(cpu)) + continue; + *pos = cpu + 1; + return &per_cpu(nf_conntrack_stat, cpu); + } + + return NULL; +} + +static void ct_cpu_seq_stop(struct seq_file *seq, void *v) +{ +} + +static int ct_cpu_seq_show(struct seq_file *seq, void *v) +{ + unsigned int nr_conntracks = atomic_read(&nf_conntrack_count); + struct ip_conntrack_stat *st = v; + + if (v == SEQ_START_TOKEN) { + seq_printf(seq, "entries searched found new invalid ignore delete delete_list insert insert_failed drop early_drop icmp_error expect_new expect_create expect_delete\n"); + return 0; + } + + seq_printf(seq, "%08x %08x %08x %08x %08x %08x %08x %08x " + "%08x %08x %08x %08x %08x %08x %08x %08x \n", + nr_conntracks, + st->searched, + st->found, + st->new, + st->invalid, + st->ignore, + st->delete, + st->delete_list, + st->insert, + st->insert_failed, + st->drop, + st->early_drop, + st->error, + + st->expect_new, + st->expect_create, + st->expect_delete + ); + return 0; +} + +static struct seq_operations ct_cpu_seq_ops = { + .start = ct_cpu_seq_start, + .next = ct_cpu_seq_next, + .stop = ct_cpu_seq_stop, + .show = ct_cpu_seq_show, +}; + +static int ct_cpu_seq_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &ct_cpu_seq_ops); +} + +static struct file_operations ct_cpu_seq_fops = { + .owner = THIS_MODULE, + .open = ct_cpu_seq_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_private, +}; +#endif /* CONFIG_PROC_FS */ + +/* Sysctl support */ + +#ifdef CONFIG_SYSCTL + +/* From nf_conntrack_core.c */ +extern int nf_conntrack_max; +extern unsigned int nf_conntrack_htable_size; + +/* From nf_conntrack_proto_tcp.c */ +extern unsigned long nf_ct_tcp_timeout_syn_sent; +extern unsigned long nf_ct_tcp_timeout_syn_recv; +extern unsigned long nf_ct_tcp_timeout_established; +extern unsigned long nf_ct_tcp_timeout_fin_wait; +extern unsigned long nf_ct_tcp_timeout_close_wait; +extern unsigned long nf_ct_tcp_timeout_last_ack; +extern unsigned long nf_ct_tcp_timeout_time_wait; +extern unsigned long nf_ct_tcp_timeout_close; +extern unsigned long nf_ct_tcp_timeout_max_retrans; +extern int nf_ct_tcp_loose; +extern int nf_ct_tcp_be_liberal; +extern int nf_ct_tcp_max_retrans; + +/* From nf_conntrack_proto_udp.c */ +extern unsigned long nf_ct_udp_timeout; +extern unsigned long nf_ct_udp_timeout_stream; + +/* From nf_conntrack_proto_generic.c */ +extern unsigned long nf_ct_generic_timeout; + +/* Log invalid packets of a given protocol */ +static int log_invalid_proto_min = 0; +static int log_invalid_proto_max = 255; + +static struct ctl_table_header *nf_ct_sysctl_header; + +static ctl_table nf_ct_sysctl_table[] = { + { + .ctl_name = NET_NF_CONNTRACK_MAX, + .procname = "nf_conntrack_max", + .data = &nf_conntrack_max, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = NET_NF_CONNTRACK_COUNT, + .procname = "nf_conntrack_count", + .data = &nf_conntrack_count, + .maxlen = sizeof(int), + .mode = 0444, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = NET_NF_CONNTRACK_BUCKETS, + .procname = "nf_conntrack_buckets", + .data = &nf_conntrack_htable_size, + .maxlen = sizeof(unsigned int), + .mode = 0444, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_TIMEOUT_SYN_SENT, + .procname = "nf_conntrack_tcp_timeout_syn_sent", + .data = &nf_ct_tcp_timeout_syn_sent, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_TIMEOUT_SYN_RECV, + .procname = "nf_conntrack_tcp_timeout_syn_recv", + .data = &nf_ct_tcp_timeout_syn_recv, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_TIMEOUT_ESTABLISHED, + .procname = "nf_conntrack_tcp_timeout_established", + .data = &nf_ct_tcp_timeout_established, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_TIMEOUT_FIN_WAIT, + .procname = "nf_conntrack_tcp_timeout_fin_wait", + .data = &nf_ct_tcp_timeout_fin_wait, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_TIMEOUT_CLOSE_WAIT, + .procname = "nf_conntrack_tcp_timeout_close_wait", + .data = &nf_ct_tcp_timeout_close_wait, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_TIMEOUT_LAST_ACK, + .procname = "nf_conntrack_tcp_timeout_last_ack", + .data = &nf_ct_tcp_timeout_last_ack, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_TIMEOUT_TIME_WAIT, + .procname = "nf_conntrack_tcp_timeout_time_wait", + .data = &nf_ct_tcp_timeout_time_wait, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_TIMEOUT_CLOSE, + .procname = "nf_conntrack_tcp_timeout_close", + .data = &nf_ct_tcp_timeout_close, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_UDP_TIMEOUT, + .procname = "nf_conntrack_udp_timeout", + .data = &nf_ct_udp_timeout, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_UDP_TIMEOUT_STREAM, + .procname = "nf_conntrack_udp_timeout_stream", + .data = &nf_ct_udp_timeout_stream, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_GENERIC_TIMEOUT, + .procname = "nf_conntrack_generic_timeout", + .data = &nf_ct_generic_timeout, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_LOG_INVALID, + .procname = "nf_conntrack_log_invalid", + .data = &nf_ct_log_invalid, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_minmax, + .strategy = &sysctl_intvec, + .extra1 = &log_invalid_proto_min, + .extra2 = &log_invalid_proto_max, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_TIMEOUT_MAX_RETRANS, + .procname = "nf_conntrack_tcp_timeout_max_retrans", + .data = &nf_ct_tcp_timeout_max_retrans, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_LOOSE, + .procname = "nf_conntrack_tcp_loose", + .data = &nf_ct_tcp_loose, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_BE_LIBERAL, + .procname = "nf_conntrack_tcp_be_liberal", + .data = &nf_ct_tcp_be_liberal, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_MAX_RETRANS, + .procname = "nf_conntrack_tcp_max_retrans", + .data = &nf_ct_tcp_max_retrans, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + + { .ctl_name = 0 } +}; + +#define NET_NF_CONNTRACK_MAX 2089 + +static ctl_table nf_ct_netfilter_table[] = { + { + .ctl_name = NET_NETFILTER, + .procname = "netfilter", + .mode = 0555, + .child = nf_ct_sysctl_table, + }, + { + .ctl_name = NET_NF_CONNTRACK_MAX, + .procname = "nf_conntrack_max", + .data = &nf_conntrack_max, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { .ctl_name = 0 } +}; + +static ctl_table nf_ct_net_table[] = { + { + .ctl_name = CTL_NET, + .procname = "net", + .mode = 0555, + .child = nf_ct_netfilter_table, + }, + { .ctl_name = 0 } +}; +EXPORT_SYMBOL(nf_ct_log_invalid); +#endif /* CONFIG_SYSCTL */ + +static int init_or_cleanup(int init) +{ +#ifdef CONFIG_PROC_FS + struct proc_dir_entry *proc, *proc_exp, *proc_stat; +#endif + int ret = 0; + + if (!init) goto cleanup; + + ret = nf_conntrack_init(); + if (ret < 0) + goto cleanup_nothing; + +#ifdef CONFIG_PROC_FS + proc = proc_net_fops_create("nf_conntrack", 0440, &ct_file_ops); + if (!proc) goto cleanup_init; + + proc_exp = proc_net_fops_create("nf_conntrack_expect", 0440, + &exp_file_ops); + if (!proc_exp) goto cleanup_proc; + + proc_stat = create_proc_entry("nf_conntrack", S_IRUGO, proc_net_stat); + if (!proc_stat) + goto cleanup_proc_exp; + + proc_stat->proc_fops = &ct_cpu_seq_fops; + proc_stat->owner = THIS_MODULE; +#endif +#ifdef CONFIG_SYSCTL + nf_ct_sysctl_header = register_sysctl_table(nf_ct_net_table, 0); + if (nf_ct_sysctl_header == NULL) { + printk("nf_conntrack: can't register to sysctl.\n"); + ret = -ENOMEM; + goto cleanup_proc_stat; + } +#endif + + return ret; + + cleanup: +#ifdef CONFIG_SYSCTL + unregister_sysctl_table(nf_ct_sysctl_header); + cleanup_proc_stat: +#endif +#ifdef CONFIG_PROC_FS + proc_net_remove("nf_conntrack_stat"); + cleanup_proc_exp: + proc_net_remove("nf_conntrack_expect"); + cleanup_proc: + proc_net_remove("nf_conntrack"); + cleanup_init: +#endif /* CNFIG_PROC_FS */ + nf_conntrack_cleanup(); + cleanup_nothing: + return ret; +} + +int nf_conntrack_l3proto_register(struct nf_conntrack_l3proto *proto) +{ + int ret = 0; + + write_lock_bh(&nf_conntrack_lock); + if (nf_ct_l3protos[proto->l3proto] != &nf_conntrack_generic_l3proto) { + ret = -EBUSY; + goto out; + } + nf_ct_l3protos[proto->l3proto] = proto; +out: + write_unlock_bh(&nf_conntrack_lock); + + return ret; +} + +void nf_conntrack_l3proto_unregister(struct nf_conntrack_l3proto *proto) +{ + write_lock_bh(&nf_conntrack_lock); + nf_ct_l3protos[proto->l3proto] = &nf_conntrack_generic_l3proto; + write_unlock_bh(&nf_conntrack_lock); + + /* Somebody could be still looking at the proto in bh. */ + synchronize_net(); + + /* Remove all contrack entries for this protocol */ + nf_ct_iterate_cleanup(kill_l3proto, proto); +} + +/* FIXME: Allow NULL functions and sub in pointers to generic for + them. --RR */ +int nf_conntrack_protocol_register(struct nf_conntrack_protocol *proto) +{ + int ret = 0; + +retry: + write_lock_bh(&nf_conntrack_lock); + if (nf_ct_protos[proto->l3proto]) { + if (nf_ct_protos[proto->l3proto][proto->proto] + != &nf_conntrack_generic_protocol) { + ret = -EBUSY; + goto out_unlock; + } + } else { + /* l3proto may be loaded latter. */ + struct nf_conntrack_protocol **proto_array; + int i; + + write_unlock_bh(&nf_conntrack_lock); + + proto_array = (struct nf_conntrack_protocol **) + kmalloc(MAX_NF_CT_PROTO * + sizeof(struct nf_conntrack_protocol *), + GFP_KERNEL); + if (proto_array == NULL) { + ret = -ENOMEM; + goto out; + } + for (i = 0; i < MAX_NF_CT_PROTO; i++) + proto_array[i] = &nf_conntrack_generic_protocol; + + write_lock_bh(&nf_conntrack_lock); + if (nf_ct_protos[proto->l3proto]) { + /* bad timing, but no problem */ + write_unlock_bh(&nf_conntrack_lock); + kfree(proto_array); + } else { + nf_ct_protos[proto->l3proto] = proto_array; + write_unlock_bh(&nf_conntrack_lock); + } + + /* + * Just once because array is never freed until unloading + * nf_conntrack.ko + */ + goto retry; + } + + nf_ct_protos[proto->l3proto][proto->proto] = proto; + +out_unlock: + write_unlock_bh(&nf_conntrack_lock); +out: + return ret; +} + +void nf_conntrack_protocol_unregister(struct nf_conntrack_protocol *proto) +{ + write_lock_bh(&nf_conntrack_lock); + nf_ct_protos[proto->l3proto][proto->proto] + = &nf_conntrack_generic_protocol; + write_unlock_bh(&nf_conntrack_lock); + + /* Somebody could be still looking at the proto in bh. */ + synchronize_net(); + + /* Remove all contrack entries for this protocol */ + nf_ct_iterate_cleanup(kill_proto, proto); +} + +static int __init init(void) +{ + return init_or_cleanup(1); +} + +static void __exit fini(void) +{ + init_or_cleanup(0); +} + +module_init(init); +module_exit(fini); + +/* Some modules need us, but don't depend directly on any symbol. + They should call this. */ +void need_nf_conntrack(void) +{ +} + +#ifdef CONFIG_NF_CONNTRACK_EVENTS +EXPORT_SYMBOL_GPL(nf_conntrack_chain); +EXPORT_SYMBOL_GPL(nf_conntrack_expect_chain); +EXPORT_SYMBOL_GPL(nf_conntrack_register_notifier); +EXPORT_SYMBOL_GPL(nf_conntrack_unregister_notifier); +EXPORT_SYMBOL_GPL(__nf_ct_event_cache_init); +EXPORT_PER_CPU_SYMBOL_GPL(nf_conntrack_ecache); +EXPORT_SYMBOL_GPL(nf_ct_deliver_cached_events); +#endif +EXPORT_SYMBOL(nf_conntrack_l3proto_register); +EXPORT_SYMBOL(nf_conntrack_l3proto_unregister); +EXPORT_SYMBOL(nf_conntrack_protocol_register); +EXPORT_SYMBOL(nf_conntrack_protocol_unregister); +EXPORT_SYMBOL(nf_ct_invert_tuplepr); +EXPORT_SYMBOL(nf_conntrack_alter_reply); +EXPORT_SYMBOL(nf_conntrack_destroyed); +EXPORT_SYMBOL(need_nf_conntrack); +EXPORT_SYMBOL(nf_conntrack_helper_register); +EXPORT_SYMBOL(nf_conntrack_helper_unregister); +EXPORT_SYMBOL(nf_ct_iterate_cleanup); +EXPORT_SYMBOL(__nf_ct_refresh_acct); +EXPORT_SYMBOL(nf_ct_protos); +EXPORT_SYMBOL(nf_ct_find_proto); +EXPORT_SYMBOL(nf_ct_l3protos); +EXPORT_SYMBOL(nf_conntrack_expect_alloc); +EXPORT_SYMBOL(nf_conntrack_expect_put); +EXPORT_SYMBOL(nf_conntrack_expect_related); +EXPORT_SYMBOL(nf_conntrack_unexpect_related); +EXPORT_SYMBOL(nf_conntrack_tuple_taken); +EXPORT_SYMBOL(nf_conntrack_htable_size); +EXPORT_SYMBOL(nf_conntrack_lock); +EXPORT_SYMBOL(nf_conntrack_hash); +EXPORT_SYMBOL(nf_conntrack_untracked); +EXPORT_SYMBOL_GPL(nf_conntrack_find_get); +#ifdef CONFIG_IP_NF_NAT_NEEDED +EXPORT_SYMBOL(nf_conntrack_tcp_update); +#endif +EXPORT_SYMBOL(__nf_conntrack_confirm); +EXPORT_SYMBOL(nf_ct_get_tuple); +EXPORT_SYMBOL(nf_ct_invert_tuple); +EXPORT_SYMBOL(nf_conntrack_in); +EXPORT_SYMBOL(__nf_conntrack_attach); diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c index 4bc27a6334c..83f4c53030f 100644 --- a/net/netfilter/nfnetlink.c +++ b/net/netfilter/nfnetlink.c @@ -128,7 +128,7 @@ void __nfa_fill(struct sk_buff *skb, int attrtype, int attrlen, memset(NFA_DATA(nfa) + attrlen, 0, NFA_ALIGN(size) - size); } -int nfattr_parse(struct nfattr *tb[], int maxattr, struct nfattr *nfa, int len) +void nfattr_parse(struct nfattr *tb[], int maxattr, struct nfattr *nfa, int len) { memset(tb, 0, sizeof(struct nfattr *) * maxattr); @@ -138,8 +138,6 @@ int nfattr_parse(struct nfattr *tb[], int maxattr, struct nfattr *nfa, int len) tb[flavor-1] = nfa; nfa = NFA_NEXT(nfa, len); } - - return 0; } /** @@ -242,15 +240,18 @@ static inline int nfnetlink_rcv_msg(struct sk_buff *skb, ss = nfnetlink_get_subsys(type); if (!ss) { #ifdef CONFIG_KMOD - /* don't call nfnl_shunlock, since it would reenter - * with further packet processing */ - up(&nfnl_sem); - request_module("nfnetlink-subsys-%d", NFNL_SUBSYS_ID(type)); - nfnl_shlock(); - ss = nfnetlink_get_subsys(type); + if (cap_raised(NETLINK_CB(skb).eff_cap, CAP_NET_ADMIN)) { + /* don't call nfnl_shunlock, since it would reenter + * with further packet processing */ + up(&nfnl_sem); + request_module("nfnetlink-subsys-%d", + NFNL_SUBSYS_ID(type)); + nfnl_shlock(); + ss = nfnetlink_get_subsys(type); + } if (!ss) #endif - goto err_inval; + goto err_inval; } nc = nfnetlink_find_client(type, ss); diff --git a/net/netlink/Makefile b/net/netlink/Makefile index 39d9c2dcd03..e3589c2de49 100644 --- a/net/netlink/Makefile +++ b/net/netlink/Makefile @@ -2,4 +2,4 @@ # Makefile for the netlink driver. # -obj-y := af_netlink.o +obj-y := af_netlink.o attr.o genetlink.o diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 5ca283537bc..8c38ee6d255 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -58,6 +58,7 @@ #include <net/sock.h> #include <net/scm.h> +#include <net/netlink.h> #define Nprintk(a...) #define NLGRPSZ(x) (ALIGN(x, sizeof(unsigned long) * 8) / 8) @@ -427,7 +428,8 @@ static int netlink_release(struct socket *sock) spin_lock(&nlk->cb_lock); if (nlk->cb) { - nlk->cb->done(nlk->cb); + if (nlk->cb->done) + nlk->cb->done(nlk->cb); netlink_destroy_callback(nlk->cb); nlk->cb = NULL; } @@ -1322,7 +1324,8 @@ static int netlink_dump(struct sock *sk) skb_queue_tail(&sk->sk_receive_queue, skb); sk->sk_data_ready(sk, skb->len); - cb->done(cb); + if (cb->done) + cb->done(cb); nlk->cb = NULL; spin_unlock(&nlk->cb_lock); @@ -1409,6 +1412,94 @@ void netlink_ack(struct sk_buff *in_skb, struct nlmsghdr *nlh, int err) netlink_unicast(in_skb->sk, skb, NETLINK_CB(in_skb).pid, MSG_DONTWAIT); } +static int netlink_rcv_skb(struct sk_buff *skb, int (*cb)(struct sk_buff *, + struct nlmsghdr *, int *)) +{ + unsigned int total_len; + struct nlmsghdr *nlh; + int err; + + while (skb->len >= nlmsg_total_size(0)) { + nlh = (struct nlmsghdr *) skb->data; + + if (skb->len < nlh->nlmsg_len) + return 0; + + total_len = min(NLMSG_ALIGN(nlh->nlmsg_len), skb->len); + + if (cb(skb, nlh, &err) < 0) { + /* Not an error, but we have to interrupt processing + * here. Note: that in this case we do not pull + * message from skb, it will be processed later. + */ + if (err == 0) + return -1; + netlink_ack(skb, nlh, err); + } else if (nlh->nlmsg_flags & NLM_F_ACK) + netlink_ack(skb, nlh, 0); + + skb_pull(skb, total_len); + } + + return 0; +} + +/** + * nelink_run_queue - Process netlink receive queue. + * @sk: Netlink socket containing the queue + * @qlen: Place to store queue length upon entry + * @cb: Callback function invoked for each netlink message found + * + * Processes as much as there was in the queue upon entry and invokes + * a callback function for each netlink message found. The callback + * function may refuse a message by returning a negative error code + * but setting the error pointer to 0 in which case this function + * returns with a qlen != 0. + * + * qlen must be initialized to 0 before the initial entry, afterwards + * the function may be called repeatedly until qlen reaches 0. + */ +void netlink_run_queue(struct sock *sk, unsigned int *qlen, + int (*cb)(struct sk_buff *, struct nlmsghdr *, int *)) +{ + struct sk_buff *skb; + + if (!*qlen || *qlen > skb_queue_len(&sk->sk_receive_queue)) + *qlen = skb_queue_len(&sk->sk_receive_queue); + + for (; *qlen; (*qlen)--) { + skb = skb_dequeue(&sk->sk_receive_queue); + if (netlink_rcv_skb(skb, cb)) { + if (skb->len) + skb_queue_head(&sk->sk_receive_queue, skb); + else { + kfree_skb(skb); + (*qlen)--; + } + break; + } + + kfree_skb(skb); + } +} + +/** + * netlink_queue_skip - Skip netlink message while processing queue. + * @nlh: Netlink message to be skipped + * @skb: Socket buffer containing the netlink messages. + * + * Pulls the given netlink message off the socket buffer so the next + * call to netlink_queue_run() will not reconsider the message. + */ +void netlink_queue_skip(struct nlmsghdr *nlh, struct sk_buff *skb) +{ + int msglen = NLMSG_ALIGN(nlh->nlmsg_len); + + if (msglen > skb->len) + msglen = skb->len; + + skb_pull(skb, msglen); +} #ifdef CONFIG_PROC_FS struct nl_seq_iter { @@ -1657,6 +1748,8 @@ out: core_initcall(netlink_proto_init); EXPORT_SYMBOL(netlink_ack); +EXPORT_SYMBOL(netlink_run_queue); +EXPORT_SYMBOL(netlink_queue_skip); EXPORT_SYMBOL(netlink_broadcast); EXPORT_SYMBOL(netlink_dump_start); EXPORT_SYMBOL(netlink_kernel_create); diff --git a/net/netlink/attr.c b/net/netlink/attr.c new file mode 100644 index 00000000000..fffef4ab276 --- /dev/null +++ b/net/netlink/attr.c @@ -0,0 +1,328 @@ +/* + * NETLINK Netlink attributes + * + * Authors: Thomas Graf <tgraf@suug.ch> + * Alexey Kuznetsov <kuznet@ms2.inr.ac.ru> + */ + +#include <linux/config.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/jiffies.h> +#include <linux/netdevice.h> +#include <linux/skbuff.h> +#include <linux/string.h> +#include <linux/types.h> +#include <net/netlink.h> + +static u16 nla_attr_minlen[NLA_TYPE_MAX+1] __read_mostly = { + [NLA_U8] = sizeof(u8), + [NLA_U16] = sizeof(u16), + [NLA_U32] = sizeof(u32), + [NLA_U64] = sizeof(u64), + [NLA_STRING] = 1, + [NLA_NESTED] = NLA_HDRLEN, +}; + +static int validate_nla(struct nlattr *nla, int maxtype, + struct nla_policy *policy) +{ + struct nla_policy *pt; + int minlen = 0; + + if (nla->nla_type <= 0 || nla->nla_type > maxtype) + return 0; + + pt = &policy[nla->nla_type]; + + BUG_ON(pt->type > NLA_TYPE_MAX); + + if (pt->minlen) + minlen = pt->minlen; + else if (pt->type != NLA_UNSPEC) + minlen = nla_attr_minlen[pt->type]; + + if (pt->type == NLA_FLAG && nla_len(nla) > 0) + return -ERANGE; + + if (nla_len(nla) < minlen) + return -ERANGE; + + return 0; +} + +/** + * nla_validate - Validate a stream of attributes + * @head: head of attribute stream + * @len: length of attribute stream + * @maxtype: maximum attribute type to be expected + * @policy: validation policy + * + * Validates all attributes in the specified attribute stream against the + * specified policy. Attributes with a type exceeding maxtype will be + * ignored. See documenation of struct nla_policy for more details. + * + * Returns 0 on success or a negative error code. + */ +int nla_validate(struct nlattr *head, int len, int maxtype, + struct nla_policy *policy) +{ + struct nlattr *nla; + int rem, err; + + nla_for_each_attr(nla, head, len, rem) { + err = validate_nla(nla, maxtype, policy); + if (err < 0) + goto errout; + } + + err = 0; +errout: + return err; +} + +/** + * nla_parse - Parse a stream of attributes into a tb buffer + * @tb: destination array with maxtype+1 elements + * @maxtype: maximum attribute type to be expected + * @head: head of attribute stream + * @len: length of attribute stream + * + * Parses a stream of attributes and stores a pointer to each attribute in + * the tb array accessable via the attribute type. Attributes with a type + * exceeding maxtype will be silently ignored for backwards compatibility + * reasons. policy may be set to NULL if no validation is required. + * + * Returns 0 on success or a negative error code. + */ +int nla_parse(struct nlattr *tb[], int maxtype, struct nlattr *head, int len, + struct nla_policy *policy) +{ + struct nlattr *nla; + int rem, err; + + memset(tb, 0, sizeof(struct nlattr *) * (maxtype + 1)); + + nla_for_each_attr(nla, head, len, rem) { + u16 type = nla->nla_type; + + if (type > 0 && type <= maxtype) { + if (policy) { + err = validate_nla(nla, maxtype, policy); + if (err < 0) + goto errout; + } + + tb[type] = nla; + } + } + + if (unlikely(rem > 0)) + printk(KERN_WARNING "netlink: %d bytes leftover after parsing " + "attributes.\n", rem); + + err = 0; +errout: + return err; +} + +/** + * nla_find - Find a specific attribute in a stream of attributes + * @head: head of attribute stream + * @len: length of attribute stream + * @attrtype: type of attribute to look for + * + * Returns the first attribute in the stream matching the specified type. + */ +struct nlattr *nla_find(struct nlattr *head, int len, int attrtype) +{ + struct nlattr *nla; + int rem; + + nla_for_each_attr(nla, head, len, rem) + if (nla->nla_type == attrtype) + return nla; + + return NULL; +} + +/** + * nla_strlcpy - Copy string attribute payload into a sized buffer + * @dst: where to copy the string to + * @src: attribute to copy the string from + * @dstsize: size of destination buffer + * + * Copies at most dstsize - 1 bytes into the destination buffer. + * The result is always a valid NUL-terminated string. Unlike + * strlcpy the destination buffer is always padded out. + * + * Returns the length of the source buffer. + */ +size_t nla_strlcpy(char *dst, const struct nlattr *nla, size_t dstsize) +{ + size_t srclen = nla_len(nla); + char *src = nla_data(nla); + + if (srclen > 0 && src[srclen - 1] == '\0') + srclen--; + + if (dstsize > 0) { + size_t len = (srclen >= dstsize) ? dstsize - 1 : srclen; + + memset(dst, 0, dstsize); + memcpy(dst, src, len); + } + + return srclen; +} + +/** + * nla_memcpy - Copy a netlink attribute into another memory area + * @dest: where to copy to memcpy + * @src: netlink attribute to copy from + * @count: size of the destination area + * + * Note: The number of bytes copied is limited by the length of + * attribute's payload. memcpy + * + * Returns the number of bytes copied. + */ +int nla_memcpy(void *dest, struct nlattr *src, int count) +{ + int minlen = min_t(int, count, nla_len(src)); + + memcpy(dest, nla_data(src), minlen); + + return minlen; +} + +/** + * nla_memcmp - Compare an attribute with sized memory area + * @nla: netlink attribute + * @data: memory area + * @size: size of memory area + */ +int nla_memcmp(const struct nlattr *nla, const void *data, + size_t size) +{ + int d = nla_len(nla) - size; + + if (d == 0) + d = memcmp(nla_data(nla), data, size); + + return d; +} + +/** + * nla_strcmp - Compare a string attribute against a string + * @nla: netlink string attribute + * @str: another string + */ +int nla_strcmp(const struct nlattr *nla, const char *str) +{ + int len = strlen(str) + 1; + int d = nla_len(nla) - len; + + if (d == 0) + d = memcmp(nla_data(nla), str, len); + + return d; +} + +/** + * __nla_reserve - reserve room for attribute on the skb + * @skb: socket buffer to reserve room on + * @attrtype: attribute type + * @attrlen: length of attribute payload + * + * Adds a netlink attribute header to a socket buffer and reserves + * room for the payload but does not copy it. + * + * The caller is responsible to ensure that the skb provides enough + * tailroom for the attribute header and payload. + */ +struct nlattr *__nla_reserve(struct sk_buff *skb, int attrtype, int attrlen) +{ + struct nlattr *nla; + + nla = (struct nlattr *) skb_put(skb, nla_total_size(attrlen)); + nla->nla_type = attrtype; + nla->nla_len = nla_attr_size(attrlen); + + memset((unsigned char *) nla + nla->nla_len, 0, nla_padlen(attrlen)); + + return nla; +} + +/** + * nla_reserve - reserve room for attribute on the skb + * @skb: socket buffer to reserve room on + * @attrtype: attribute type + * @attrlen: length of attribute payload + * + * Adds a netlink attribute header to a socket buffer and reserves + * room for the payload but does not copy it. + * + * Returns NULL if the tailroom of the skb is insufficient to store + * the attribute header and payload. + */ +struct nlattr *nla_reserve(struct sk_buff *skb, int attrtype, int attrlen) +{ + if (unlikely(skb_tailroom(skb) < nla_total_size(attrlen))) + return NULL; + + return __nla_reserve(skb, attrtype, attrlen); +} + +/** + * __nla_put - Add a netlink attribute to a socket buffer + * @skb: socket buffer to add attribute to + * @attrtype: attribute type + * @attrlen: length of attribute payload + * @data: head of attribute payload + * + * The caller is responsible to ensure that the skb provides enough + * tailroom for the attribute header and payload. + */ +void __nla_put(struct sk_buff *skb, int attrtype, int attrlen, + const void *data) +{ + struct nlattr *nla; + + nla = __nla_reserve(skb, attrtype, attrlen); + memcpy(nla_data(nla), data, attrlen); +} + + +/** + * nla_put - Add a netlink attribute to a socket buffer + * @skb: socket buffer to add attribute to + * @attrtype: attribute type + * @attrlen: length of attribute payload + * @data: head of attribute payload + * + * Returns -1 if the tailroom of the skb is insufficient to store + * the attribute header and payload. + */ +int nla_put(struct sk_buff *skb, int attrtype, int attrlen, const void *data) +{ + if (unlikely(skb_tailroom(skb) < nla_total_size(attrlen))) + return -1; + + __nla_put(skb, attrtype, attrlen, data); + return 0; +} + + +EXPORT_SYMBOL(nla_validate); +EXPORT_SYMBOL(nla_parse); +EXPORT_SYMBOL(nla_find); +EXPORT_SYMBOL(nla_strlcpy); +EXPORT_SYMBOL(__nla_reserve); +EXPORT_SYMBOL(nla_reserve); +EXPORT_SYMBOL(__nla_put); +EXPORT_SYMBOL(nla_put); +EXPORT_SYMBOL(nla_memcpy); +EXPORT_SYMBOL(nla_memcmp); +EXPORT_SYMBOL(nla_strcmp); diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c new file mode 100644 index 00000000000..287cfcc5695 --- /dev/null +++ b/net/netlink/genetlink.c @@ -0,0 +1,579 @@ +/* + * NETLINK Generic Netlink Family + * + * Authors: Jamal Hadi Salim + * Thomas Graf <tgraf@suug.ch> + */ + +#include <linux/config.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/types.h> +#include <linux/socket.h> +#include <linux/string.h> +#include <linux/skbuff.h> +#include <net/sock.h> +#include <net/genetlink.h> + +struct sock *genl_sock = NULL; + +static DECLARE_MUTEX(genl_sem); /* serialization of message processing */ + +static void genl_lock(void) +{ + down(&genl_sem); +} + +static int genl_trylock(void) +{ + return down_trylock(&genl_sem); +} + +static void genl_unlock(void) +{ + up(&genl_sem); + + if (genl_sock && genl_sock->sk_receive_queue.qlen) + genl_sock->sk_data_ready(genl_sock, 0); +} + +#define GENL_FAM_TAB_SIZE 16 +#define GENL_FAM_TAB_MASK (GENL_FAM_TAB_SIZE - 1) + +static struct list_head family_ht[GENL_FAM_TAB_SIZE]; + +static int genl_ctrl_event(int event, void *data); + +static inline unsigned int genl_family_hash(unsigned int id) +{ + return id & GENL_FAM_TAB_MASK; +} + +static inline struct list_head *genl_family_chain(unsigned int id) +{ + return &family_ht[genl_family_hash(id)]; +} + +static struct genl_family *genl_family_find_byid(unsigned int id) +{ + struct genl_family *f; + + list_for_each_entry(f, genl_family_chain(id), family_list) + if (f->id == id) + return f; + + return NULL; +} + +static struct genl_family *genl_family_find_byname(char *name) +{ + struct genl_family *f; + int i; + + for (i = 0; i < GENL_FAM_TAB_SIZE; i++) + list_for_each_entry(f, genl_family_chain(i), family_list) + if (strcmp(f->name, name) == 0) + return f; + + return NULL; +} + +static struct genl_ops *genl_get_cmd(u8 cmd, struct genl_family *family) +{ + struct genl_ops *ops; + + list_for_each_entry(ops, &family->ops_list, ops_list) + if (ops->cmd == cmd) + return ops; + + return NULL; +} + +/* Of course we are going to have problems once we hit + * 2^16 alive types, but that can only happen by year 2K +*/ +static inline u16 genl_generate_id(void) +{ + static u16 id_gen_idx; + int overflowed = 0; + + do { + if (id_gen_idx == 0) + id_gen_idx = GENL_MIN_ID; + + if (++id_gen_idx > GENL_MAX_ID) { + if (!overflowed) { + overflowed = 1; + id_gen_idx = 0; + continue; + } else + return 0; + } + + } while (genl_family_find_byid(id_gen_idx)); + + return id_gen_idx; +} + +/** + * genl_register_ops - register generic netlink operations + * @family: generic netlink family + * @ops: operations to be registered + * + * Registers the specified operations and assigns them to the specified + * family. Either a doit or dumpit callback must be specified or the + * operation will fail. Only one operation structure per command + * identifier may be registered. + * + * See include/net/genetlink.h for more documenation on the operations + * structure. + * + * Returns 0 on success or a negative error code. + */ +int genl_register_ops(struct genl_family *family, struct genl_ops *ops) +{ + int err = -EINVAL; + + if (ops->dumpit == NULL && ops->doit == NULL) + goto errout; + + if (genl_get_cmd(ops->cmd, family)) { + err = -EEXIST; + goto errout; + } + + genl_lock(); + list_add_tail(&ops->ops_list, &family->ops_list); + genl_unlock(); + + genl_ctrl_event(CTRL_CMD_NEWOPS, ops); + err = 0; +errout: + return err; +} + +/** + * genl_unregister_ops - unregister generic netlink operations + * @family: generic netlink family + * @ops: operations to be unregistered + * + * Unregisters the specified operations and unassigns them from the + * specified family. The operation blocks until the current message + * processing has finished and doesn't start again until the + * unregister process has finished. + * + * Note: It is not necessary to unregister all operations before + * unregistering the family, unregistering the family will cause + * all assigned operations to be unregistered automatically. + * + * Returns 0 on success or a negative error code. + */ +int genl_unregister_ops(struct genl_family *family, struct genl_ops *ops) +{ + struct genl_ops *rc; + + genl_lock(); + list_for_each_entry(rc, &family->ops_list, ops_list) { + if (rc == ops) { + list_del(&ops->ops_list); + genl_unlock(); + genl_ctrl_event(CTRL_CMD_DELOPS, ops); + return 0; + } + } + genl_unlock(); + + return -ENOENT; +} + +/** + * genl_register_family - register a generic netlink family + * @family: generic netlink family + * + * Registers the specified family after validating it first. Only one + * family may be registered with the same family name or identifier. + * The family id may equal GENL_ID_GENERATE causing an unique id to + * be automatically generated and assigned. + * + * Return 0 on success or a negative error code. + */ +int genl_register_family(struct genl_family *family) +{ + int err = -EINVAL; + + if (family->id && family->id < GENL_MIN_ID) + goto errout; + + if (family->id > GENL_MAX_ID) + goto errout; + + INIT_LIST_HEAD(&family->ops_list); + + genl_lock(); + + if (genl_family_find_byname(family->name)) { + err = -EEXIST; + goto errout_locked; + } + + if (genl_family_find_byid(family->id)) { + err = -EEXIST; + goto errout_locked; + } + + if (!try_module_get(family->owner)) { + err = -EBUSY; + goto errout_locked; + } + + if (family->id == GENL_ID_GENERATE) { + u16 newid = genl_generate_id(); + + if (!newid) { + err = -ENOMEM; + goto errout_locked; + } + + family->id = newid; + } + + if (family->maxattr) { + family->attrbuf = kmalloc((family->maxattr+1) * + sizeof(struct nlattr *), GFP_KERNEL); + if (family->attrbuf == NULL) { + err = -ENOMEM; + goto errout; + } + } else + family->attrbuf = NULL; + + list_add_tail(&family->family_list, genl_family_chain(family->id)); + genl_unlock(); + + genl_ctrl_event(CTRL_CMD_NEWFAMILY, family); + + return 0; + +errout_locked: + genl_unlock(); +errout: + return err; +} + +/** + * genl_unregister_family - unregister generic netlink family + * @family: generic netlink family + * + * Unregisters the specified family. + * + * Returns 0 on success or a negative error code. + */ +int genl_unregister_family(struct genl_family *family) +{ + struct genl_family *rc; + + genl_lock(); + + list_for_each_entry(rc, genl_family_chain(family->id), family_list) { + if (family->id != rc->id || strcmp(rc->name, family->name)) + continue; + + list_del(&rc->family_list); + INIT_LIST_HEAD(&family->ops_list); + genl_unlock(); + + module_put(family->owner); + kfree(family->attrbuf); + genl_ctrl_event(CTRL_CMD_DELFAMILY, family); + return 0; + } + + genl_unlock(); + + return -ENOENT; +} + +static inline int genl_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, + int *errp) +{ + struct genl_ops *ops; + struct genl_family *family; + struct genl_info info; + struct genlmsghdr *hdr = nlmsg_data(nlh); + int hdrlen, err = -EINVAL; + + if (!(nlh->nlmsg_flags & NLM_F_REQUEST)) + goto ignore; + + if (nlh->nlmsg_type < NLMSG_MIN_TYPE) + goto ignore; + + family = genl_family_find_byid(nlh->nlmsg_type); + if (family == NULL) { + err = -ENOENT; + goto errout; + } + + hdrlen = GENL_HDRLEN + family->hdrsize; + if (nlh->nlmsg_len < nlmsg_msg_size(hdrlen)) + goto errout; + + ops = genl_get_cmd(hdr->cmd, family); + if (ops == NULL) { + err = -EOPNOTSUPP; + goto errout; + } + + if ((ops->flags & GENL_ADMIN_PERM) && security_netlink_recv(skb)) { + err = -EPERM; + goto errout; + } + + if (nlh->nlmsg_flags & NLM_F_DUMP) { + if (ops->dumpit == NULL) { + err = -EOPNOTSUPP; + goto errout; + } + + *errp = err = netlink_dump_start(genl_sock, skb, nlh, + ops->dumpit, NULL); + if (err == 0) + skb_pull(skb, min(NLMSG_ALIGN(nlh->nlmsg_len), + skb->len)); + return -1; + } + + if (ops->doit == NULL) { + err = -EOPNOTSUPP; + goto errout; + } + + if (family->attrbuf) { + err = nlmsg_parse(nlh, hdrlen, family->attrbuf, family->maxattr, + ops->policy); + if (err < 0) + goto errout; + } + + info.snd_seq = nlh->nlmsg_seq; + info.snd_pid = NETLINK_CB(skb).pid; + info.nlhdr = nlh; + info.genlhdr = nlmsg_data(nlh); + info.userhdr = nlmsg_data(nlh) + GENL_HDRLEN; + info.attrs = family->attrbuf; + + *errp = err = ops->doit(skb, &info); + return err; + +ignore: + return 0; + +errout: + *errp = err; + return -1; +} + +static void genl_rcv(struct sock *sk, int len) +{ + unsigned int qlen = 0; + + do { + if (genl_trylock()) + return; + netlink_run_queue(sk, &qlen, &genl_rcv_msg); + genl_unlock(); + } while (qlen && genl_sock && genl_sock->sk_receive_queue.qlen); +} + +/************************************************************************** + * Controller + **************************************************************************/ + +static int ctrl_fill_info(struct genl_family *family, u32 pid, u32 seq, + u32 flags, struct sk_buff *skb, u8 cmd) +{ + void *hdr; + + hdr = genlmsg_put(skb, pid, seq, GENL_ID_CTRL, 0, flags, cmd, + family->version); + if (hdr == NULL) + return -1; + + NLA_PUT_STRING(skb, CTRL_ATTR_FAMILY_NAME, family->name); + NLA_PUT_U16(skb, CTRL_ATTR_FAMILY_ID, family->id); + + return genlmsg_end(skb, hdr); + +nla_put_failure: + return genlmsg_cancel(skb, hdr); +} + +static int ctrl_dumpfamily(struct sk_buff *skb, struct netlink_callback *cb) +{ + + int i, n = 0; + struct genl_family *rt; + int chains_to_skip = cb->args[0]; + int fams_to_skip = cb->args[1]; + + for (i = 0; i < GENL_FAM_TAB_SIZE; i++) { + if (i < chains_to_skip) + continue; + n = 0; + list_for_each_entry(rt, genl_family_chain(i), family_list) { + if (++n < fams_to_skip) + continue; + if (ctrl_fill_info(rt, NETLINK_CB(cb->skb).pid, + cb->nlh->nlmsg_seq, NLM_F_MULTI, + skb, CTRL_CMD_NEWFAMILY) < 0) + goto errout; + } + + fams_to_skip = 0; + } + +errout: + cb->args[0] = i; + cb->args[1] = n; + + return skb->len; +} + +static struct sk_buff *ctrl_build_msg(struct genl_family *family, u32 pid, + int seq, int cmd) +{ + struct sk_buff *skb; + int err; + + skb = nlmsg_new(NLMSG_GOODSIZE); + if (skb == NULL) + return ERR_PTR(-ENOBUFS); + + err = ctrl_fill_info(family, pid, seq, 0, skb, cmd); + if (err < 0) { + nlmsg_free(skb); + return ERR_PTR(err); + } + + return skb; +} + +static struct nla_policy ctrl_policy[CTRL_ATTR_MAX+1] __read_mostly = { + [CTRL_ATTR_FAMILY_ID] = { .type = NLA_U16 }, + [CTRL_ATTR_FAMILY_NAME] = { .type = NLA_STRING }, +}; + +static int ctrl_getfamily(struct sk_buff *skb, struct genl_info *info) +{ + struct sk_buff *msg; + struct genl_family *res = NULL; + int err = -EINVAL; + + if (info->attrs[CTRL_ATTR_FAMILY_ID]) { + u16 id = nla_get_u16(info->attrs[CTRL_ATTR_FAMILY_ID]); + res = genl_family_find_byid(id); + } + + if (info->attrs[CTRL_ATTR_FAMILY_NAME]) { + char name[GENL_NAMSIZ]; + + if (nla_strlcpy(name, info->attrs[CTRL_ATTR_FAMILY_NAME], + GENL_NAMSIZ) >= GENL_NAMSIZ) + goto errout; + + res = genl_family_find_byname(name); + } + + if (res == NULL) { + err = -ENOENT; + goto errout; + } + + msg = ctrl_build_msg(res, info->snd_pid, info->snd_seq, + CTRL_CMD_NEWFAMILY); + if (IS_ERR(msg)) { + err = PTR_ERR(msg); + goto errout; + } + + err = genlmsg_unicast(msg, info->snd_pid); +errout: + return err; +} + +static int genl_ctrl_event(int event, void *data) +{ + struct sk_buff *msg; + + if (genl_sock == NULL) + return 0; + + switch (event) { + case CTRL_CMD_NEWFAMILY: + case CTRL_CMD_DELFAMILY: + msg = ctrl_build_msg(data, 0, 0, event); + if (IS_ERR(msg)) + return PTR_ERR(msg); + + genlmsg_multicast(msg, 0, GENL_ID_CTRL); + break; + } + + return 0; +} + +static struct genl_ops genl_ctrl_ops = { + .cmd = CTRL_CMD_GETFAMILY, + .doit = ctrl_getfamily, + .dumpit = ctrl_dumpfamily, + .policy = ctrl_policy, +}; + +static struct genl_family genl_ctrl = { + .id = GENL_ID_CTRL, + .name = "nlctrl", + .version = 0x1, + .maxattr = CTRL_ATTR_MAX, + .owner = THIS_MODULE, +}; + +static int __init genl_init(void) +{ + int i, err; + + for (i = 0; i < GENL_FAM_TAB_SIZE; i++) + INIT_LIST_HEAD(&family_ht[i]); + + err = genl_register_family(&genl_ctrl); + if (err < 0) + goto errout; + + err = genl_register_ops(&genl_ctrl, &genl_ctrl_ops); + if (err < 0) + goto errout_register; + + netlink_set_nonroot(NETLINK_GENERIC, NL_NONROOT_RECV); + genl_sock = netlink_kernel_create(NETLINK_GENERIC, GENL_MAX_ID, + genl_rcv, THIS_MODULE); + if (genl_sock == NULL) { + panic("GENL: Cannot initialize generic netlink\n"); + return -ENOMEM; + } + + return 0; + +errout_register: + genl_unregister_family(&genl_ctrl); +errout: + panic("GENL: Cannot register controller: %d\n", err); + return err; +} + +subsys_initcall(genl_init); + +EXPORT_SYMBOL(genl_sock); +EXPORT_SYMBOL(genl_register_ops); +EXPORT_SYMBOL(genl_unregister_ops); +EXPORT_SYMBOL(genl_register_family); +EXPORT_SYMBOL(genl_unregister_family); diff --git a/net/rxrpc/transport.c b/net/rxrpc/transport.c index 122c086ee2d..dbe6105e83a 100644 --- a/net/rxrpc/transport.c +++ b/net/rxrpc/transport.c @@ -23,6 +23,7 @@ #include <linux/in.h> #include <linux/in6.h> #include <linux/icmp.h> +#include <linux/skbuff.h> #include <net/sock.h> #include <net/ip.h> #if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE) @@ -475,15 +476,11 @@ void rxrpc_trans_receive_packet(struct rxrpc_transport *trans) /* we'll probably need to checksum it (didn't call * sock_recvmsg) */ - if (pkt->ip_summed != CHECKSUM_UNNECESSARY) { - if ((unsigned short) - csum_fold(skb_checksum(pkt, 0, pkt->len, - pkt->csum))) { - kfree_skb(pkt); - rxrpc_krxiod_queue_transport(trans); - _leave(" CSUM failed"); - return; - } + if (skb_checksum_complete(pkt)) { + kfree_skb(pkt); + rxrpc_krxiod_queue_transport(trans); + _leave(" CSUM failed"); + return; } addr = pkt->nh.iph->saddr; diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 702ede309b0..61c3abeacca 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -55,6 +55,7 @@ static void call_bind(struct rpc_task *task); static void call_bind_status(struct rpc_task *task); static void call_transmit(struct rpc_task *task); static void call_status(struct rpc_task *task); +static void call_transmit_status(struct rpc_task *task); static void call_refresh(struct rpc_task *task); static void call_refreshresult(struct rpc_task *task); static void call_timeout(struct rpc_task *task); @@ -672,6 +673,18 @@ call_allocate(struct rpc_task *task) rpc_exit(task, -ERESTARTSYS); } +static inline int +rpc_task_need_encode(struct rpc_task *task) +{ + return task->tk_rqstp->rq_snd_buf.len == 0; +} + +static inline void +rpc_task_force_reencode(struct rpc_task *task) +{ + task->tk_rqstp->rq_snd_buf.len = 0; +} + /* * 3. Encode arguments of an RPC call */ @@ -867,12 +880,14 @@ call_transmit(struct rpc_task *task) if (task->tk_status != 0) return; /* Encode here so that rpcsec_gss can use correct sequence number. */ - if (task->tk_rqstp->rq_bytes_sent == 0) { + if (rpc_task_need_encode(task)) { + task->tk_rqstp->rq_bytes_sent = 0; call_encode(task); /* Did the encode result in an error condition? */ if (task->tk_status != 0) goto out_nosend; } + task->tk_action = call_transmit_status; xprt_transmit(task); if (task->tk_status < 0) return; @@ -884,6 +899,7 @@ call_transmit(struct rpc_task *task) out_nosend: /* release socket write lock before attempting to handle error */ xprt_abort_transmit(task); + rpc_task_force_reencode(task); } /* @@ -915,7 +931,6 @@ call_status(struct rpc_task *task) break; case -ECONNREFUSED: case -ENOTCONN: - req->rq_bytes_sent = 0; if (clnt->cl_autobind) clnt->cl_port = 0; task->tk_action = call_bind; @@ -937,7 +952,18 @@ call_status(struct rpc_task *task) } /* - * 6a. Handle RPC timeout + * 6a. Handle transmission errors. + */ +static void +call_transmit_status(struct rpc_task *task) +{ + if (task->tk_status != -EAGAIN) + rpc_task_force_reencode(task); + call_status(task); +} + +/* + * 6b. Handle RPC timeout * We do not release the request slot, so we keep using the * same XID for all retransmits. */ diff --git a/net/sunrpc/socklib.c b/net/sunrpc/socklib.c index 8f97e90f36c..eb330d4f66d 100644 --- a/net/sunrpc/socklib.c +++ b/net/sunrpc/socklib.c @@ -6,6 +6,9 @@ * Copyright (C) 1995, 1996 Olaf Kirch <okir@monad.swb.de> */ +#include <linux/compiler.h> +#include <linux/netdevice.h> +#include <linux/skbuff.h> #include <linux/types.h> #include <linux/pagemap.h> #include <linux/udp.h> @@ -165,6 +168,8 @@ int csum_partial_copy_to_xdr(struct xdr_buf *xdr, struct sk_buff *skb) return -1; if ((unsigned short)csum_fold(desc.csum)) return -1; + if (unlikely(skb->ip_summed == CHECKSUM_HW)) + netdev_rx_csum_fault(skb->dev); return 0; no_checksum: if (xdr_partial_copy_from_skb(xdr, 0, &desc, skb_read_bits) < 0) diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index f16e7cdd615..e50e7cf4373 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -623,12 +623,9 @@ svc_udp_recvfrom(struct svc_rqst *rqstp) /* we can use it in-place */ rqstp->rq_arg.head[0].iov_base = skb->data + sizeof(struct udphdr); rqstp->rq_arg.head[0].iov_len = len; - if (skb->ip_summed != CHECKSUM_UNNECESSARY) { - if ((unsigned short)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum))) { - skb_free_datagram(svsk->sk_sk, skb); - return 0; - } - skb->ip_summed = CHECKSUM_UNNECESSARY; + if (skb_checksum_complete(skb)) { + skb_free_datagram(svsk->sk_sk, skb); + return 0; } rqstp->rq_skbuff = skb; } diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index c35336a0f71..0cdd9a07e04 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -18,7 +18,6 @@ #include <linux/string.h> #include <linux/net.h> #include <linux/skbuff.h> -#include <linux/netlink.h> #include <linux/rtnetlink.h> #include <linux/pfkeyv2.h> #include <linux/ipsec.h> @@ -26,6 +25,7 @@ #include <linux/security.h> #include <net/sock.h> #include <net/xfrm.h> +#include <net/netlink.h> #include <asm/uaccess.h> static struct sock *xfrm_nl; @@ -948,11 +948,6 @@ static struct xfrm_link { [XFRM_MSG_FLUSHPOLICY - XFRM_MSG_BASE] = { .doit = xfrm_flush_policy }, }; -static int xfrm_done(struct netlink_callback *cb) -{ - return 0; -} - static int xfrm_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, int *errp) { struct rtattr *xfrma[XFRMA_MAX]; @@ -984,20 +979,15 @@ static int xfrm_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, int *err if ((type == (XFRM_MSG_GETSA - XFRM_MSG_BASE) || type == (XFRM_MSG_GETPOLICY - XFRM_MSG_BASE)) && (nlh->nlmsg_flags & NLM_F_DUMP)) { - u32 rlen; - if (link->dump == NULL) goto err_einval; if ((*errp = netlink_dump_start(xfrm_nl, skb, nlh, - link->dump, - xfrm_done)) != 0) { + link->dump, NULL)) != 0) { return -1; } - rlen = NLMSG_ALIGN(nlh->nlmsg_len); - if (rlen > skb->len) - rlen = skb->len; - skb_pull(skb, rlen); + + netlink_queue_skip(nlh, skb); return -1; } @@ -1032,60 +1022,13 @@ err_einval: return -1; } -static int xfrm_user_rcv_skb(struct sk_buff *skb) -{ - int err; - struct nlmsghdr *nlh; - - while (skb->len >= NLMSG_SPACE(0)) { - u32 rlen; - - nlh = (struct nlmsghdr *) skb->data; - if (nlh->nlmsg_len < sizeof(*nlh) || - skb->len < nlh->nlmsg_len) - return 0; - rlen = NLMSG_ALIGN(nlh->nlmsg_len); - if (rlen > skb->len) - rlen = skb->len; - if (xfrm_user_rcv_msg(skb, nlh, &err) < 0) { - if (err == 0) - return -1; - netlink_ack(skb, nlh, err); - } else if (nlh->nlmsg_flags & NLM_F_ACK) - netlink_ack(skb, nlh, 0); - skb_pull(skb, rlen); - } - - return 0; -} - static void xfrm_netlink_rcv(struct sock *sk, int len) { - unsigned int qlen = skb_queue_len(&sk->sk_receive_queue); + unsigned int qlen = 0; do { - struct sk_buff *skb; - down(&xfrm_cfg_sem); - - if (qlen > skb_queue_len(&sk->sk_receive_queue)) - qlen = skb_queue_len(&sk->sk_receive_queue); - - for (; qlen; qlen--) { - skb = skb_dequeue(&sk->sk_receive_queue); - if (xfrm_user_rcv_skb(skb)) { - if (skb->len) - skb_queue_head(&sk->sk_receive_queue, - skb); - else { - kfree_skb(skb); - qlen--; - } - break; - } - kfree_skb(skb); - } - + netlink_run_queue(sk, &qlen, &xfrm_user_rcv_msg); up(&xfrm_cfg_sem); } while (qlen); diff --git a/sound/pci/ad1889.c b/sound/pci/ad1889.c index e72ccd1a004..1fdae678a34 100644 --- a/sound/pci/ad1889.c +++ b/sound/pci/ad1889.c @@ -1067,7 +1067,6 @@ MODULE_DEVICE_TABLE(pci, snd_ad1889_ids); static struct pci_driver ad1889_pci = { .name = "AD1889 Audio", - .owner = THIS_MODULE, .id_table = snd_ad1889_ids, .probe = snd_ad1889_probe, .remove = __devexit_p(snd_ad1889_remove), diff --git a/sound/pci/ali5451/ali5451.c b/sound/pci/ali5451/ali5451.c index 4e76c4a636d..feffbe73387 100644 --- a/sound/pci/ali5451/ali5451.c +++ b/sound/pci/ali5451/ali5451.c @@ -2403,7 +2403,6 @@ static void __devexit snd_ali_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "ALI 5451", - .owner = THIS_MODULE, .id_table = snd_ali_ids, .probe = snd_ali_probe, .remove = __devexit_p(snd_ali_remove), diff --git a/sound/pci/als4000.c b/sound/pci/als4000.c index 7c61561f297..1904df65026 100644 --- a/sound/pci/als4000.c +++ b/sound/pci/als4000.c @@ -768,7 +768,6 @@ static void __devexit snd_card_als4000_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "ALS4000", - .owner = THIS_MODULE, .id_table = snd_als4000_ids, .probe = snd_card_als4000_probe, .remove = __devexit_p(snd_card_als4000_remove), diff --git a/sound/pci/atiixp.c b/sound/pci/atiixp.c index f5dad9248e3..8bae10d9352 100644 --- a/sound/pci/atiixp.c +++ b/sound/pci/atiixp.c @@ -1635,7 +1635,6 @@ static void __devexit snd_atiixp_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "ATI IXP AC97 controller", - .owner = THIS_MODULE, .id_table = snd_atiixp_ids, .probe = snd_atiixp_probe, .remove = __devexit_p(snd_atiixp_remove), diff --git a/sound/pci/atiixp_modem.c b/sound/pci/atiixp_modem.c index 0cf20207957..3174b662541 100644 --- a/sound/pci/atiixp_modem.c +++ b/sound/pci/atiixp_modem.c @@ -1309,7 +1309,6 @@ static void __devexit snd_atiixp_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "ATI IXP MC97 controller", - .owner = THIS_MODULE, .id_table = snd_atiixp_ids, .probe = snd_atiixp_probe, .remove = __devexit_p(snd_atiixp_remove), diff --git a/sound/pci/au88x0/au88x0.c b/sound/pci/au88x0/au88x0.c index 6af3b13f2fd..d965609d8b3 100644 --- a/sound/pci/au88x0/au88x0.c +++ b/sound/pci/au88x0/au88x0.c @@ -373,7 +373,6 @@ static void __devexit snd_vortex_remove(struct pci_dev *pci) // pci_driver definition static struct pci_driver driver = { .name = CARD_NAME_SHORT, - .owner = THIS_MODULE, .id_table = snd_vortex_ids, .probe = snd_vortex_probe, .remove = __devexit_p(snd_vortex_remove), diff --git a/sound/pci/azt3328.c b/sound/pci/azt3328.c index da99b1be2e8..ab737d6df41 100644 --- a/sound/pci/azt3328.c +++ b/sound/pci/azt3328.c @@ -1838,7 +1838,6 @@ snd_azf3328_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "AZF3328", - .owner = THIS_MODULE, .id_table = snd_azf3328_ids, .probe = snd_azf3328_probe, .remove = __devexit_p(snd_azf3328_remove), diff --git a/sound/pci/bt87x.c b/sound/pci/bt87x.c index 01d98eeb242..8feca228c6d 100644 --- a/sound/pci/bt87x.c +++ b/sound/pci/bt87x.c @@ -897,14 +897,13 @@ static void __devexit snd_bt87x_remove(struct pci_dev *pci) /* default entries for all Bt87x cards - it's not exported */ /* driver_data is set to 0 to call detection */ static struct pci_device_id snd_bt87x_default_ids[] = { - BT_DEVICE(878, PCI_ANY_ID, PCI_ANY_ID, 0), - BT_DEVICE(879, PCI_ANY_ID, PCI_ANY_ID, 0), + BT_DEVICE(PCI_DEVICE_ID_BROOKTREE_878, PCI_ANY_ID, PCI_ANY_ID, 0), + BT_DEVICE(PCI_DEVICE_ID_BROOKTREE_879, PCI_ANY_ID, PCI_ANY_ID, 0), { } }; static struct pci_driver driver = { .name = "Bt87x", - .owner = THIS_MODULE, .id_table = snd_bt87x_ids, .probe = snd_bt87x_probe, .remove = __devexit_p(snd_bt87x_remove), diff --git a/sound/pci/ca0106/ca0106_main.c b/sound/pci/ca0106/ca0106_main.c index ee58d16002e..389d967c97f 100644 --- a/sound/pci/ca0106/ca0106_main.c +++ b/sound/pci/ca0106/ca0106_main.c @@ -1499,7 +1499,6 @@ MODULE_DEVICE_TABLE(pci, snd_ca0106_ids); // pci_driver definition static struct pci_driver driver = { .name = "CA0106", - .owner = THIS_MODULE, .id_table = snd_ca0106_ids, .probe = snd_ca0106_probe, .remove = __devexit_p(snd_ca0106_remove), diff --git a/sound/pci/cmipci.c b/sound/pci/cmipci.c index 57e8e433d56..db605373b3b 100644 --- a/sound/pci/cmipci.c +++ b/sound/pci/cmipci.c @@ -3053,7 +3053,6 @@ static void __devexit snd_cmipci_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "C-Media PCI", - .owner = THIS_MODULE, .id_table = snd_cmipci_ids, .probe = snd_cmipci_probe, .remove = __devexit_p(snd_cmipci_remove), diff --git a/sound/pci/cs4281.c b/sound/pci/cs4281.c index aea2c47712f..034ff3755a3 100644 --- a/sound/pci/cs4281.c +++ b/sound/pci/cs4281.c @@ -2106,7 +2106,6 @@ static int cs4281_resume(snd_card_t *card) static struct pci_driver driver = { .name = "CS4281", - .owner = THIS_MODULE, .id_table = snd_cs4281_ids, .probe = snd_cs4281_probe, .remove = __devexit_p(snd_cs4281_remove), diff --git a/sound/pci/cs46xx/cs46xx.c b/sound/pci/cs46xx/cs46xx.c index 32b4f8465ce..b9fff4ee6f9 100644 --- a/sound/pci/cs46xx/cs46xx.c +++ b/sound/pci/cs46xx/cs46xx.c @@ -163,7 +163,6 @@ static void __devexit snd_card_cs46xx_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "Sound Fusion CS46xx", - .owner = THIS_MODULE, .id_table = snd_cs46xx_ids, .probe = snd_card_cs46xx_probe, .remove = __devexit_p(snd_card_cs46xx_remove), diff --git a/sound/pci/emu10k1/emu10k1.c b/sound/pci/emu10k1/emu10k1.c index dd1ea9d3b81..78270f8710f 100644 --- a/sound/pci/emu10k1/emu10k1.c +++ b/sound/pci/emu10k1/emu10k1.c @@ -223,7 +223,6 @@ static void __devexit snd_card_emu10k1_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "EMU10K1_Audigy", - .owner = THIS_MODULE, .id_table = snd_emu10k1_ids, .probe = snd_card_emu10k1_probe, .remove = __devexit_p(snd_card_emu10k1_remove), diff --git a/sound/pci/emu10k1/emu10k1x.c b/sound/pci/emu10k1/emu10k1x.c index cbb689474e7..795577716a5 100644 --- a/sound/pci/emu10k1/emu10k1x.c +++ b/sound/pci/emu10k1/emu10k1x.c @@ -1613,7 +1613,6 @@ MODULE_DEVICE_TABLE(pci, snd_emu10k1x_ids); // pci_driver definition static struct pci_driver driver = { .name = "EMU10K1X", - .owner = THIS_MODULE, .id_table = snd_emu10k1x_ids, .probe = snd_emu10k1x_probe, .remove = __devexit_p(snd_emu10k1x_remove), diff --git a/sound/pci/ens1370.c b/sound/pci/ens1370.c index 92ff7c510f2..2daa575f43f 100644 --- a/sound/pci/ens1370.c +++ b/sound/pci/ens1370.c @@ -2386,7 +2386,6 @@ static void __devexit snd_audiopci_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = DRIVER_NAME, - .owner = THIS_MODULE, .id_table = snd_audiopci_ids, .probe = snd_audiopci_probe, .remove = __devexit_p(snd_audiopci_remove), diff --git a/sound/pci/es1938.c b/sound/pci/es1938.c index 78f90defcab..c134f48152b 100644 --- a/sound/pci/es1938.c +++ b/sound/pci/es1938.c @@ -1758,7 +1758,6 @@ static void __devexit snd_es1938_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "ESS ES1938 (Solo-1)", - .owner = THIS_MODULE, .id_table = snd_es1938_ids, .probe = snd_es1938_probe, .remove = __devexit_p(snd_es1938_remove), diff --git a/sound/pci/es1968.c b/sound/pci/es1968.c index ac8294e21cc..50079dc9074 100644 --- a/sound/pci/es1968.c +++ b/sound/pci/es1968.c @@ -2761,7 +2761,6 @@ static void __devexit snd_es1968_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "ES1968 (ESS Maestro)", - .owner = THIS_MODULE, .id_table = snd_es1968_ids, .probe = snd_es1968_probe, .remove = __devexit_p(snd_es1968_remove), diff --git a/sound/pci/fm801.c b/sound/pci/fm801.c index 4c7c8d225c7..4e1d3434888 100644 --- a/sound/pci/fm801.c +++ b/sound/pci/fm801.c @@ -1459,7 +1459,6 @@ static void __devexit snd_card_fm801_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "FM801", - .owner = THIS_MODULE, .id_table = snd_fm801_ids, .probe = snd_card_fm801_probe, .remove = __devexit_p(snd_card_fm801_remove), diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index 9d1412a9f2f..ed525c03c99 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -1616,7 +1616,6 @@ MODULE_DEVICE_TABLE(pci, azx_ids); /* pci_driver definition */ static struct pci_driver driver = { .name = "HDA Intel", - .owner = THIS_MODULE, .id_table = azx_ids, .probe = azx_probe, .remove = __devexit_p(azx_remove), diff --git a/sound/pci/ice1712/ice1712.c b/sound/pci/ice1712/ice1712.c index 5aca37798c3..bd71bf42454 100644 --- a/sound/pci/ice1712/ice1712.c +++ b/sound/pci/ice1712/ice1712.c @@ -2735,7 +2735,6 @@ static void __devexit snd_ice1712_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "ICE1712", - .owner = THIS_MODULE, .id_table = snd_ice1712_ids, .probe = snd_ice1712_probe, .remove = __devexit_p(snd_ice1712_remove), diff --git a/sound/pci/ice1712/ice1724.c b/sound/pci/ice1712/ice1724.c index 5b4293f5a65..0b5389ee26d 100644 --- a/sound/pci/ice1712/ice1724.c +++ b/sound/pci/ice1712/ice1724.c @@ -2332,7 +2332,6 @@ static void __devexit snd_vt1724_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "ICE1724", - .owner = THIS_MODULE, .id_table = snd_vt1724_ids, .probe = snd_vt1724_probe, .remove = __devexit_p(snd_vt1724_remove), diff --git a/sound/pci/intel8x0.c b/sound/pci/intel8x0.c index 0801083f32d..cf7801d2dd1 100644 --- a/sound/pci/intel8x0.c +++ b/sound/pci/intel8x0.c @@ -2876,7 +2876,6 @@ static void __devexit snd_intel8x0_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "Intel ICH", - .owner = THIS_MODULE, .id_table = snd_intel8x0_ids, .probe = snd_intel8x0_probe, .remove = __devexit_p(snd_intel8x0_remove), diff --git a/sound/pci/intel8x0m.c b/sound/pci/intel8x0m.c index acfb197a833..a42091860da 100644 --- a/sound/pci/intel8x0m.c +++ b/sound/pci/intel8x0m.c @@ -1317,7 +1317,6 @@ static void __devexit snd_intel8x0m_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "Intel ICH Modem", - .owner = THIS_MODULE, .id_table = snd_intel8x0m_ids, .probe = snd_intel8x0m_probe, .remove = __devexit_p(snd_intel8x0m_remove), diff --git a/sound/pci/korg1212/korg1212.c b/sound/pci/korg1212/korg1212.c index 5561fd4091e..a110d664f62 100644 --- a/sound/pci/korg1212/korg1212.c +++ b/sound/pci/korg1212/korg1212.c @@ -2534,7 +2534,6 @@ static void __devexit snd_korg1212_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "korg1212", - .owner = THIS_MODULE, .id_table = snd_korg1212_ids, .probe = snd_korg1212_probe, .remove = __devexit_p(snd_korg1212_remove), diff --git a/sound/pci/maestro3.c b/sound/pci/maestro3.c index 99eb76c56f8..ede7a75bfe0 100644 --- a/sound/pci/maestro3.c +++ b/sound/pci/maestro3.c @@ -2858,7 +2858,6 @@ static void __devexit snd_m3_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "Maestro3", - .owner = THIS_MODULE, .id_table = snd_m3_ids, .probe = snd_m3_probe, .remove = __devexit_p(snd_m3_remove), diff --git a/sound/pci/mixart/mixart.c b/sound/pci/mixart/mixart.c index c341c99ec78..b3090a13eda 100644 --- a/sound/pci/mixart/mixart.c +++ b/sound/pci/mixart/mixart.c @@ -1423,7 +1423,6 @@ static void __devexit snd_mixart_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "Digigram miXart", - .owner = THIS_MODULE, .id_table = snd_mixart_ids, .probe = snd_mixart_probe, .remove = __devexit_p(snd_mixart_remove), diff --git a/sound/pci/nm256/nm256.c b/sound/pci/nm256/nm256.c index e7aa1517845..089d23b4a00 100644 --- a/sound/pci/nm256/nm256.c +++ b/sound/pci/nm256/nm256.c @@ -1673,7 +1673,6 @@ static void __devexit snd_nm256_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "NeoMagic 256", - .owner = THIS_MODULE, .id_table = snd_nm256_ids, .probe = snd_nm256_probe, .remove = __devexit_p(snd_nm256_remove), diff --git a/sound/pci/rme32.c b/sound/pci/rme32.c index e6627b0e38e..783df7625c1 100644 --- a/sound/pci/rme32.c +++ b/sound/pci/rme32.c @@ -2012,7 +2012,6 @@ static void __devexit snd_rme32_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "RME Digi32", - .owner = THIS_MODULE, .id_table = snd_rme32_ids, .probe = snd_rme32_probe, .remove = __devexit_p(snd_rme32_remove), diff --git a/sound/pci/rme96.c b/sound/pci/rme96.c index 0eddeb16a10..6d422ef6499 100644 --- a/sound/pci/rme96.c +++ b/sound/pci/rme96.c @@ -2413,7 +2413,6 @@ static void __devexit snd_rme96_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "RME Digi96", - .owner = THIS_MODULE, .id_table = snd_rme96_ids, .probe = snd_rme96_probe, .remove = __devexit_p(snd_rme96_remove), diff --git a/sound/pci/rme9652/hdsp.c b/sound/pci/rme9652/hdsp.c index 845158b01b0..d15ffb3e9b0 100644 --- a/sound/pci/rme9652/hdsp.c +++ b/sound/pci/rme9652/hdsp.c @@ -5062,7 +5062,6 @@ static void __devexit snd_hdsp_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "RME Hammerfall DSP", - .owner = THIS_MODULE, .id_table = snd_hdsp_ids, .probe = snd_hdsp_probe, .remove = __devexit_p(snd_hdsp_remove), diff --git a/sound/pci/rme9652/hdspm.c b/sound/pci/rme9652/hdspm.c index 60a1141f132..a1aef6f6767 100644 --- a/sound/pci/rme9652/hdspm.c +++ b/sound/pci/rme9652/hdspm.c @@ -3639,7 +3639,6 @@ static void __devexit snd_hdspm_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "RME Hammerfall DSP MADI", - .owner = THIS_MODULE, .id_table = snd_hdspm_ids, .probe = snd_hdspm_probe, .remove = __devexit_p(snd_hdspm_remove), diff --git a/sound/pci/rme9652/rme9652.c b/sound/pci/rme9652/rme9652.c index 59fcef9b6b8..f9d0c126c21 100644 --- a/sound/pci/rme9652/rme9652.c +++ b/sound/pci/rme9652/rme9652.c @@ -2654,7 +2654,6 @@ static void __devexit snd_rme9652_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "RME Digi9652 (Hammerfall)", - .owner = THIS_MODULE, .id_table = snd_rme9652_ids, .probe = snd_rme9652_probe, .remove = __devexit_p(snd_rme9652_remove), diff --git a/sound/pci/sonicvibes.c b/sound/pci/sonicvibes.c index 9a35474aad0..e92ef3ae2ca 100644 --- a/sound/pci/sonicvibes.c +++ b/sound/pci/sonicvibes.c @@ -1502,7 +1502,6 @@ static void __devexit snd_sonic_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "S3 SonicVibes", - .owner = THIS_MODULE, .id_table = snd_sonic_ids, .probe = snd_sonic_probe, .remove = __devexit_p(snd_sonic_remove), diff --git a/sound/pci/trident/trident.c b/sound/pci/trident/trident.c index a8ca8e17853..940d531575c 100644 --- a/sound/pci/trident/trident.c +++ b/sound/pci/trident/trident.c @@ -177,7 +177,6 @@ static void __devexit snd_trident_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "Trident4DWaveAudio", - .owner = THIS_MODULE, .id_table = snd_trident_ids, .probe = snd_trident_probe, .remove = __devexit_p(snd_trident_remove), diff --git a/sound/pci/via82xx.c b/sound/pci/via82xx.c index 523eace250f..fad2a2413bf 100644 --- a/sound/pci/via82xx.c +++ b/sound/pci/via82xx.c @@ -2478,7 +2478,6 @@ static void __devexit snd_via82xx_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "VIA 82xx Audio", - .owner = THIS_MODULE, .id_table = snd_via82xx_ids, .probe = snd_via82xx_probe, .remove = __devexit_p(snd_via82xx_remove), diff --git a/sound/pci/via82xx_modem.c b/sound/pci/via82xx_modem.c index 011f0fb63bf..b83660bd05b 100644 --- a/sound/pci/via82xx_modem.c +++ b/sound/pci/via82xx_modem.c @@ -1198,7 +1198,6 @@ static void __devexit snd_via82xx_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "VIA 82xx Modem", - .owner = THIS_MODULE, .id_table = snd_via82xx_modem_ids, .probe = snd_via82xx_probe, .remove = __devexit_p(snd_via82xx_remove), diff --git a/sound/pci/vx222/vx222.c b/sound/pci/vx222/vx222.c index 2a7ad9dec02..dca6bd2c758 100644 --- a/sound/pci/vx222/vx222.c +++ b/sound/pci/vx222/vx222.c @@ -252,7 +252,6 @@ static void __devexit snd_vx222_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "Digigram VX222", - .owner = THIS_MODULE, .id_table = snd_vx222_ids, .probe = snd_vx222_probe, .remove = __devexit_p(snd_vx222_remove), diff --git a/sound/pci/ymfpci/ymfpci.c b/sound/pci/ymfpci/ymfpci.c index 1bbba32517f..d013237205d 100644 --- a/sound/pci/ymfpci/ymfpci.c +++ b/sound/pci/ymfpci/ymfpci.c @@ -344,7 +344,6 @@ static void __devexit snd_card_ymfpci_remove(struct pci_dev *pci) static struct pci_driver driver = { .name = "Yamaha DS-XG PCI", - .owner = THIS_MODULE, .id_table = snd_ymfpci_ids, .probe = snd_card_ymfpci_probe, .remove = __devexit_p(snd_card_ymfpci_remove), |