From 9f14517bd6f38cf4b48d742a0ac5db9d17edecba Mon Sep 17 00:00:00 2001 From: John Stultz Date: Wed, 1 Jun 2011 00:32:50 -0700 Subject: [PATCH 001/439] clocksource: tile: convert to use clocksource_register_hz Convert tile to use clocksource_register_hz. CC: Thomas Gleixner Signed-off-by: John Stultz Signed-off-by: Chris Metcalf --- arch/tile/kernel/time.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/arch/tile/kernel/time.c b/arch/tile/kernel/time.c index c4be58cc5d50..f6f50f2a5e37 100644 --- a/arch/tile/kernel/time.c +++ b/arch/tile/kernel/time.c @@ -78,7 +78,6 @@ static struct clocksource cycle_counter_cs = { .rating = 300, .read = clocksource_get_cycles, .mask = CLOCKSOURCE_MASK(64), - .shift = 22, /* typical value, e.g. x86 tsc uses this */ .flags = CLOCK_SOURCE_IS_CONTINUOUS, }; @@ -91,8 +90,6 @@ void __init setup_clock(void) cycles_per_sec = hv_sysconf(HV_SYSCONF_CPU_SPEED); sched_clock_mult = clocksource_hz2mult(cycles_per_sec, SCHED_CLOCK_SHIFT); - cycle_counter_cs.mult = - clocksource_hz2mult(cycles_per_sec, cycle_counter_cs.shift); } void __init calibrate_delay(void) @@ -107,7 +104,7 @@ void __init calibrate_delay(void) void __init time_init(void) { /* Initialize and register the clock source. */ - clocksource_register(&cycle_counter_cs); + clocksource_register_hz(&cycle_counter_cs, cycles_per_sec); /* Start up the tile-timer interrupt source on the boot cpu. */ setup_tile_timer(); From 2c007a9d8cb018a3ce2106342f2aff2abca2a3bd Mon Sep 17 00:00:00 2001 From: Chris Metcalf Date: Fri, 3 Jun 2011 17:36:18 -0400 Subject: [PATCH 002/439] tile: use generic-y format for one-line asm-generic headers This lets us remove a lot of one-line wrapper header files. See commit d8ecc5cd8e227bc318513b5306ae88a474b8886d for context. Signed-off-by: Chris Metcalf --- arch/tile/include/asm/Kbuild | 38 +++++++++++++++++++++++ arch/tile/include/asm/bug.h | 1 - arch/tile/include/asm/bugs.h | 1 - arch/tile/include/asm/cputime.h | 1 - arch/tile/include/asm/device.h | 1 - arch/tile/include/asm/div64.h | 1 - arch/tile/include/asm/emergency-restart.h | 1 - arch/tile/include/asm/errno.h | 1 - arch/tile/include/asm/fb.h | 1 - arch/tile/include/asm/fcntl.h | 1 - arch/tile/include/asm/ioctl.h | 1 - arch/tile/include/asm/ioctls.h | 1 - arch/tile/include/asm/ipc.h | 1 - arch/tile/include/asm/ipcbuf.h | 1 - arch/tile/include/asm/irq_regs.h | 1 - arch/tile/include/asm/kdebug.h | 1 - arch/tile/include/asm/local.h | 1 - arch/tile/include/asm/module.h | 1 - arch/tile/include/asm/msgbuf.h | 1 - arch/tile/include/asm/mutex.h | 1 - arch/tile/include/asm/param.h | 1 - arch/tile/include/asm/parport.h | 1 - arch/tile/include/asm/poll.h | 1 - arch/tile/include/asm/posix_types.h | 1 - arch/tile/include/asm/resource.h | 1 - arch/tile/include/asm/scatterlist.h | 1 - arch/tile/include/asm/sembuf.h | 1 - arch/tile/include/asm/serial.h | 1 - arch/tile/include/asm/shmbuf.h | 1 - arch/tile/include/asm/shmparam.h | 1 - arch/tile/include/asm/socket.h | 1 - arch/tile/include/asm/sockios.h | 1 - arch/tile/include/asm/statfs.h | 1 - arch/tile/include/asm/termbits.h | 1 - arch/tile/include/asm/termios.h | 1 - arch/tile/include/asm/types.h | 1 - arch/tile/include/asm/ucontext.h | 1 - arch/tile/include/asm/xor.h | 1 - 38 files changed, 38 insertions(+), 37 deletions(-) delete mode 100644 arch/tile/include/asm/bug.h delete mode 100644 arch/tile/include/asm/bugs.h delete mode 100644 arch/tile/include/asm/cputime.h delete mode 100644 arch/tile/include/asm/device.h delete mode 100644 arch/tile/include/asm/div64.h delete mode 100644 arch/tile/include/asm/emergency-restart.h delete mode 100644 arch/tile/include/asm/errno.h delete mode 100644 arch/tile/include/asm/fb.h delete mode 100644 arch/tile/include/asm/fcntl.h delete mode 100644 arch/tile/include/asm/ioctl.h delete mode 100644 arch/tile/include/asm/ioctls.h delete mode 100644 arch/tile/include/asm/ipc.h delete mode 100644 arch/tile/include/asm/ipcbuf.h delete mode 100644 arch/tile/include/asm/irq_regs.h delete mode 100644 arch/tile/include/asm/kdebug.h delete mode 100644 arch/tile/include/asm/local.h delete mode 100644 arch/tile/include/asm/module.h delete mode 100644 arch/tile/include/asm/msgbuf.h delete mode 100644 arch/tile/include/asm/mutex.h delete mode 100644 arch/tile/include/asm/param.h delete mode 100644 arch/tile/include/asm/parport.h delete mode 100644 arch/tile/include/asm/poll.h delete mode 100644 arch/tile/include/asm/posix_types.h delete mode 100644 arch/tile/include/asm/resource.h delete mode 100644 arch/tile/include/asm/scatterlist.h delete mode 100644 arch/tile/include/asm/sembuf.h delete mode 100644 arch/tile/include/asm/serial.h delete mode 100644 arch/tile/include/asm/shmbuf.h delete mode 100644 arch/tile/include/asm/shmparam.h delete mode 100644 arch/tile/include/asm/socket.h delete mode 100644 arch/tile/include/asm/sockios.h delete mode 100644 arch/tile/include/asm/statfs.h delete mode 100644 arch/tile/include/asm/termbits.h delete mode 100644 arch/tile/include/asm/termios.h delete mode 100644 arch/tile/include/asm/types.h delete mode 100644 arch/tile/include/asm/ucontext.h delete mode 100644 arch/tile/include/asm/xor.h diff --git a/arch/tile/include/asm/Kbuild b/arch/tile/include/asm/Kbuild index 849ab2fa1f5c..aec60dc06007 100644 --- a/arch/tile/include/asm/Kbuild +++ b/arch/tile/include/asm/Kbuild @@ -2,3 +2,41 @@ include include/asm-generic/Kbuild.asm header-y += ucontext.h header-y += hardwall.h + +generic-y += bug.h +generic-y += bugs.h +generic-y += cputime.h +generic-y += device.h +generic-y += div64.h +generic-y += emergency-restart.h +generic-y += errno.h +generic-y += fb.h +generic-y += fcntl.h +generic-y += ioctl.h +generic-y += ioctls.h +generic-y += ipc.h +generic-y += ipcbuf.h +generic-y += irq_regs.h +generic-y += kdebug.h +generic-y += local.h +generic-y += module.h +generic-y += msgbuf.h +generic-y += mutex.h +generic-y += param.h +generic-y += parport.h +generic-y += poll.h +generic-y += posix_types.h +generic-y += resource.h +generic-y += scatterlist.h +generic-y += sembuf.h +generic-y += serial.h +generic-y += shmbuf.h +generic-y += shmparam.h +generic-y += socket.h +generic-y += sockios.h +generic-y += statfs.h +generic-y += termbits.h +generic-y += termios.h +generic-y += types.h +generic-y += ucontext.h +generic-y += xor.h diff --git a/arch/tile/include/asm/bug.h b/arch/tile/include/asm/bug.h deleted file mode 100644 index b12fd89e42e9..000000000000 --- a/arch/tile/include/asm/bug.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/bugs.h b/arch/tile/include/asm/bugs.h deleted file mode 100644 index 61791e1ad9f5..000000000000 --- a/arch/tile/include/asm/bugs.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/cputime.h b/arch/tile/include/asm/cputime.h deleted file mode 100644 index 6d68ad7e0ea3..000000000000 --- a/arch/tile/include/asm/cputime.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/device.h b/arch/tile/include/asm/device.h deleted file mode 100644 index f0a4c256403b..000000000000 --- a/arch/tile/include/asm/device.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/div64.h b/arch/tile/include/asm/div64.h deleted file mode 100644 index 6cd978cefb28..000000000000 --- a/arch/tile/include/asm/div64.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/emergency-restart.h b/arch/tile/include/asm/emergency-restart.h deleted file mode 100644 index 3711bd9d50bd..000000000000 --- a/arch/tile/include/asm/emergency-restart.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/errno.h b/arch/tile/include/asm/errno.h deleted file mode 100644 index 4c82b503d92f..000000000000 --- a/arch/tile/include/asm/errno.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/fb.h b/arch/tile/include/asm/fb.h deleted file mode 100644 index 3a4988e8df45..000000000000 --- a/arch/tile/include/asm/fb.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/fcntl.h b/arch/tile/include/asm/fcntl.h deleted file mode 100644 index 46ab12db5739..000000000000 --- a/arch/tile/include/asm/fcntl.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/ioctl.h b/arch/tile/include/asm/ioctl.h deleted file mode 100644 index b279fe06dfe5..000000000000 --- a/arch/tile/include/asm/ioctl.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/ioctls.h b/arch/tile/include/asm/ioctls.h deleted file mode 100644 index ec34c760665e..000000000000 --- a/arch/tile/include/asm/ioctls.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/ipc.h b/arch/tile/include/asm/ipc.h deleted file mode 100644 index a46e3d9c2a3f..000000000000 --- a/arch/tile/include/asm/ipc.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/ipcbuf.h b/arch/tile/include/asm/ipcbuf.h deleted file mode 100644 index 84c7e51cb6d0..000000000000 --- a/arch/tile/include/asm/ipcbuf.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/irq_regs.h b/arch/tile/include/asm/irq_regs.h deleted file mode 100644 index 3dd9c0b70270..000000000000 --- a/arch/tile/include/asm/irq_regs.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/kdebug.h b/arch/tile/include/asm/kdebug.h deleted file mode 100644 index 6ece1b037665..000000000000 --- a/arch/tile/include/asm/kdebug.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/local.h b/arch/tile/include/asm/local.h deleted file mode 100644 index c11c530f74d0..000000000000 --- a/arch/tile/include/asm/local.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/module.h b/arch/tile/include/asm/module.h deleted file mode 100644 index 1e4b79fe8584..000000000000 --- a/arch/tile/include/asm/module.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/msgbuf.h b/arch/tile/include/asm/msgbuf.h deleted file mode 100644 index 809134c644a6..000000000000 --- a/arch/tile/include/asm/msgbuf.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/mutex.h b/arch/tile/include/asm/mutex.h deleted file mode 100644 index ff6101aa2c71..000000000000 --- a/arch/tile/include/asm/mutex.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/param.h b/arch/tile/include/asm/param.h deleted file mode 100644 index 965d45427975..000000000000 --- a/arch/tile/include/asm/param.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/parport.h b/arch/tile/include/asm/parport.h deleted file mode 100644 index cf252af64590..000000000000 --- a/arch/tile/include/asm/parport.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/poll.h b/arch/tile/include/asm/poll.h deleted file mode 100644 index c98509d3149e..000000000000 --- a/arch/tile/include/asm/poll.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/posix_types.h b/arch/tile/include/asm/posix_types.h deleted file mode 100644 index 22cae6230ceb..000000000000 --- a/arch/tile/include/asm/posix_types.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/resource.h b/arch/tile/include/asm/resource.h deleted file mode 100644 index 04bc4db8921b..000000000000 --- a/arch/tile/include/asm/resource.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/scatterlist.h b/arch/tile/include/asm/scatterlist.h deleted file mode 100644 index 35d786fe93ae..000000000000 --- a/arch/tile/include/asm/scatterlist.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/sembuf.h b/arch/tile/include/asm/sembuf.h deleted file mode 100644 index 7673b83cfef7..000000000000 --- a/arch/tile/include/asm/sembuf.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/serial.h b/arch/tile/include/asm/serial.h deleted file mode 100644 index a0cb0caff152..000000000000 --- a/arch/tile/include/asm/serial.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/shmbuf.h b/arch/tile/include/asm/shmbuf.h deleted file mode 100644 index 83c05fc2de38..000000000000 --- a/arch/tile/include/asm/shmbuf.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/shmparam.h b/arch/tile/include/asm/shmparam.h deleted file mode 100644 index 93f30deb95d0..000000000000 --- a/arch/tile/include/asm/shmparam.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/socket.h b/arch/tile/include/asm/socket.h deleted file mode 100644 index 6b71384b9d8b..000000000000 --- a/arch/tile/include/asm/socket.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/sockios.h b/arch/tile/include/asm/sockios.h deleted file mode 100644 index def6d4746ee7..000000000000 --- a/arch/tile/include/asm/sockios.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/statfs.h b/arch/tile/include/asm/statfs.h deleted file mode 100644 index 0b91fe198c20..000000000000 --- a/arch/tile/include/asm/statfs.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/termbits.h b/arch/tile/include/asm/termbits.h deleted file mode 100644 index 3935b106de79..000000000000 --- a/arch/tile/include/asm/termbits.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/termios.h b/arch/tile/include/asm/termios.h deleted file mode 100644 index 280d78a9d966..000000000000 --- a/arch/tile/include/asm/termios.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/types.h b/arch/tile/include/asm/types.h deleted file mode 100644 index b9e79bc580dd..000000000000 --- a/arch/tile/include/asm/types.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/ucontext.h b/arch/tile/include/asm/ucontext.h deleted file mode 100644 index 9bc07b9f30fb..000000000000 --- a/arch/tile/include/asm/ucontext.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/tile/include/asm/xor.h b/arch/tile/include/asm/xor.h deleted file mode 100644 index c82eb12a5b18..000000000000 --- a/arch/tile/include/asm/xor.h +++ /dev/null @@ -1 +0,0 @@ -#include From ea41b1e5440442cea5c029b192e9ebbe68e423f6 Mon Sep 17 00:00:00 2001 From: Chris Metcalf Date: Fri, 10 Jun 2011 10:14:03 -0400 Subject: [PATCH 003/439] ioctl-number.txt: add the tile hardwall ioctl range Signed-off-by: Chris Metcalf --- Documentation/ioctl/ioctl-number.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/ioctl/ioctl-number.txt b/Documentation/ioctl/ioctl-number.txt index 3a46e360496d..b7dc3747b711 100644 --- a/Documentation/ioctl/ioctl-number.txt +++ b/Documentation/ioctl/ioctl-number.txt @@ -292,6 +292,7 @@ Code Seq#(hex) Include File Comments 0xA0 all linux/sdp/sdp.h Industrial Device Project +0xA2 00-0F arch/tile/include/asm/hardwall.h 0xA3 80-8F Port ACL in development: 0xA3 90-9F linux/dtlk.h From dbcb4a1a3f16702918caa4d4ab7062965050a780 Mon Sep 17 00:00:00 2001 From: Chris Metcalf Date: Fri, 10 Jun 2011 13:07:48 -0400 Subject: [PATCH 004/439] arch/tile: add hypervisor-based character driver for SPI flash ROM The first version of this patch proposed an arch/tile/drivers/ directory, but the consensus was that this was probably a poor choice for a place to group Tilera-specific drivers, and that in any case grouping by platform was discouraged, and grouping by function was preferred. This version of the patch addresses various issues raised in the community, primarily the absence of sysfs integration. The sysfs integration now handles passing information on sector size, page size, and total partition size to userspace as well. In addition, we now use a single "struct cdev" to manage all the partition minor devices, and dynamically discover the correct number of partitions from the hypervisor rather than using a module_param with a default value. This driver has no particular "peer" drivers it can be grouped with. It is sort of like an MTD driver for SPI ROM, but it doesn't group well with the other MTD devices since it relies on hypervisor virtualization to handle many of the irritating aspects of flash ROM management: sector awareness, background read for sub-sector writes, bit examination to determine whether a sector erase needs to be issued, etc. It is in fact more like an EEPROM driver, but the hypervisor virtualization does require a "flush" command if you wish to commit a sector write prior to writing to a different sector, and this is sufficiently different from generic I2C/SPI EEPROMs that as a result it doesn't group well with them either. The simple character device is already in use by a range of Tilera SPI ROM management tools, as well as by customers. In addition, using the simple character device actually simplifies the userspace tools, since they don't need to manage sector erase, background read, etc. This both simplifies the code (since we can uniformly manage plain files and the SPI ROM) as well as makes the user code portable to non-Linux platforms that don't offer the same MTD ioctls. Signed-off-by: Chris Metcalf Reviewed-by: Arnd Bergmann --- arch/tile/include/hv/drv_srom_intf.h | 41 +++ drivers/char/Kconfig | 11 + drivers/char/Makefile | 2 + drivers/char/tile-srom.c | 481 +++++++++++++++++++++++++++ 4 files changed, 535 insertions(+) create mode 100644 arch/tile/include/hv/drv_srom_intf.h create mode 100644 drivers/char/tile-srom.c diff --git a/arch/tile/include/hv/drv_srom_intf.h b/arch/tile/include/hv/drv_srom_intf.h new file mode 100644 index 000000000000..6395faa6d9e6 --- /dev/null +++ b/arch/tile/include/hv/drv_srom_intf.h @@ -0,0 +1,41 @@ +/* + * Copyright 2011 Tilera 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, version 2. + * + * 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. + */ + +/** + * @file drv_srom_intf.h + * Interface definitions for the SPI Flash ROM driver. + */ + +#ifndef _SYS_HV_INCLUDE_DRV_SROM_INTF_H +#define _SYS_HV_INCLUDE_DRV_SROM_INTF_H + +/** Read this offset to get the total device size. */ +#define SROM_TOTAL_SIZE_OFF 0xF0000000 + +/** Read this offset to get the device sector size. */ +#define SROM_SECTOR_SIZE_OFF 0xF0000004 + +/** Read this offset to get the device page size. */ +#define SROM_PAGE_SIZE_OFF 0xF0000008 + +/** Write this offset to flush any pending writes. */ +#define SROM_FLUSH_OFF 0xF1000000 + +/** Write this offset, plus the byte offset of the start of a sector, to + * erase a sector. Any write data is ignored, but there must be at least + * one byte of write data. Only applies when the driver is in MTD mode. + */ +#define SROM_ERASE_OFF 0xF2000000 + +#endif /* _SYS_HV_INCLUDE_DRV_SROM_INTF_H */ diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig index 49502bc5360a..423fd56bf612 100644 --- a/drivers/char/Kconfig +++ b/drivers/char/Kconfig @@ -616,5 +616,16 @@ config MSM_SMD_PKT Enables userspace clients to read and write to some packet SMD ports via device interface for MSM chipset. +config TILE_SROM + bool "Character-device access via hypervisor to the Tilera SPI ROM" + depends on TILE + default y + ---help--- + This device provides character-level read-write access + to the SROM, typically via the "0", "1", and "2" devices + in /dev/srom/. The Tilera hypervisor makes the flash + device appear much like a simple EEPROM, and knows + how to partition a single ROM for multiple purposes. + endmenu diff --git a/drivers/char/Makefile b/drivers/char/Makefile index 7a00672bd85d..32762ba769c2 100644 --- a/drivers/char/Makefile +++ b/drivers/char/Makefile @@ -63,3 +63,5 @@ obj-$(CONFIG_RAMOOPS) += ramoops.o obj-$(CONFIG_JS_RTC) += js-rtc.o js-rtc-y = rtc.o + +obj-$(CONFIG_TILE_SROM) += tile-srom.o diff --git a/drivers/char/tile-srom.c b/drivers/char/tile-srom.c new file mode 100644 index 000000000000..cf3ee008dca2 --- /dev/null +++ b/drivers/char/tile-srom.c @@ -0,0 +1,481 @@ +/* + * Copyright 2011 Tilera 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, version 2. + * + * 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. + * + * SPI Flash ROM driver + * + * This source code is derived from code provided in "Linux Device + * Drivers, Third Edition", by Jonathan Corbet, Alessandro Rubini, and + * Greg Kroah-Hartman, published by O'Reilly Media, Inc. + */ + +#include +#include +#include +#include /* printk() */ +#include /* kmalloc() */ +#include /* everything... */ +#include /* error codes */ +#include /* size_t */ +#include +#include /* O_ACCMODE */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +/* + * Size of our hypervisor I/O requests. We break up large transfers + * so that we don't spend large uninterrupted spans of time in the + * hypervisor. Erasing an SROM sector takes a significant fraction of + * a second, so if we allowed the user to, say, do one I/O to write the + * entire ROM, we'd get soft lockup timeouts, or worse. + */ +#define SROM_CHUNK_SIZE ((size_t)4096) + +/* + * When hypervisor is busy (e.g. erasing), poll the status periodically. + */ + +/* + * Interval to poll the state in msec + */ +#define SROM_WAIT_TRY_INTERVAL 20 + +/* + * Maximum times to poll the state + */ +#define SROM_MAX_WAIT_TRY_TIMES 1000 + +struct srom_dev { + int hv_devhdl; /* Handle for hypervisor device */ + u32 total_size; /* Size of this device */ + u32 sector_size; /* Size of a sector */ + u32 page_size; /* Size of a page */ + struct mutex lock; /* Allow only one accessor at a time */ +}; + +static int srom_major; /* Dynamic major by default */ +module_param(srom_major, int, 0); +MODULE_AUTHOR("Tilera Corporation"); +MODULE_LICENSE("GPL"); + +static int srom_devs; /* Number of SROM partitions */ +static struct cdev srom_cdev; +static struct class *srom_class; +static struct srom_dev *srom_devices; + +/* + * Handle calling the hypervisor and managing EAGAIN/EBUSY. + */ + +static ssize_t _srom_read(int hv_devhdl, void *buf, + loff_t off, size_t count) +{ + int retval, retries = SROM_MAX_WAIT_TRY_TIMES; + for (;;) { + retval = hv_dev_pread(hv_devhdl, 0, (HV_VirtAddr)buf, + count, off); + if (retval >= 0) + return retval; + if (retval == HV_EAGAIN) + continue; + if (retval == HV_EBUSY && --retries > 0) { + msleep(SROM_WAIT_TRY_INTERVAL); + continue; + } + pr_err("_srom_read: error %d\n", retval); + return -EIO; + } +} + +static ssize_t _srom_write(int hv_devhdl, const void *buf, + loff_t off, size_t count) +{ + int retval, retries = SROM_MAX_WAIT_TRY_TIMES; + for (;;) { + retval = hv_dev_pwrite(hv_devhdl, 0, (HV_VirtAddr)buf, + count, off); + if (retval >= 0) + return retval; + if (retval == HV_EAGAIN) + continue; + if (retval == HV_EBUSY && --retries > 0) { + msleep(SROM_WAIT_TRY_INTERVAL); + continue; + } + pr_err("_srom_write: error %d\n", retval); + return -EIO; + } +} + +/** + * srom_open() - Device open routine. + * @inode: Inode for this device. + * @filp: File for this specific open of the device. + * + * Returns zero, or an error code. + */ +static int srom_open(struct inode *inode, struct file *filp) +{ + filp->private_data = &srom_devices[iminor(inode)]; + return 0; +} + + +/** + * srom_release() - Device release routine. + * @inode: Inode for this device. + * @filp: File for this specific open of the device. + * + * Returns zero, or an error code. + */ +static int srom_release(struct inode *inode, struct file *filp) +{ + struct srom_dev *srom = filp->private_data; + char dummy; + + /* Make sure we've flushed anything written to the ROM. */ + mutex_lock(&srom->lock); + if (srom->hv_devhdl >= 0) + _srom_write(srom->hv_devhdl, &dummy, SROM_FLUSH_OFF, 1); + mutex_unlock(&srom->lock); + + filp->private_data = NULL; + + return 0; +} + + +/** + * srom_read() - Read data from the device. + * @filp: File for this specific open of the device. + * @buf: User's data buffer. + * @count: Number of bytes requested. + * @f_pos: File position. + * + * Returns number of bytes read, or an error code. + */ +static ssize_t srom_read(struct file *filp, char __user *buf, + size_t count, loff_t *f_pos) +{ + int retval = 0; + void *kernbuf; + struct srom_dev *srom = filp->private_data; + + kernbuf = kmalloc(SROM_CHUNK_SIZE, GFP_KERNEL); + if (!kernbuf) + return -ENOMEM; + + if (mutex_lock_interruptible(&srom->lock)) { + retval = -ERESTARTSYS; + kfree(kernbuf); + return retval; + } + + while (count) { + int hv_retval; + int bytes_this_pass = min(count, SROM_CHUNK_SIZE); + + hv_retval = _srom_read(srom->hv_devhdl, kernbuf, + *f_pos, bytes_this_pass); + if (hv_retval > 0) { + if (copy_to_user(buf, kernbuf, hv_retval) != 0) { + retval = -EFAULT; + break; + } + } else if (hv_retval <= 0) { + if (retval == 0) + retval = hv_retval; + break; + } + + retval += hv_retval; + *f_pos += hv_retval; + buf += hv_retval; + count -= hv_retval; + } + + mutex_unlock(&srom->lock); + kfree(kernbuf); + + return retval; +} + +/** + * srom_write() - Write data to the device. + * @filp: File for this specific open of the device. + * @buf: User's data buffer. + * @count: Number of bytes requested. + * @f_pos: File position. + * + * Returns number of bytes written, or an error code. + */ +static ssize_t srom_write(struct file *filp, const char __user *buf, + size_t count, loff_t *f_pos) +{ + int retval = 0; + void *kernbuf; + struct srom_dev *srom = filp->private_data; + + kernbuf = kmalloc(SROM_CHUNK_SIZE, GFP_KERNEL); + if (!kernbuf) + return -ENOMEM; + + if (mutex_lock_interruptible(&srom->lock)) { + retval = -ERESTARTSYS; + kfree(kernbuf); + return retval; + } + + while (count) { + int hv_retval; + int bytes_this_pass = min(count, SROM_CHUNK_SIZE); + + if (copy_from_user(kernbuf, buf, bytes_this_pass) != 0) { + retval = -EFAULT; + break; + } + + hv_retval = _srom_write(srom->hv_devhdl, kernbuf, + *f_pos, bytes_this_pass); + if (hv_retval <= 0) { + if (retval == 0) + retval = hv_retval; + break; + } + + retval += hv_retval; + *f_pos += hv_retval; + buf += hv_retval; + count -= hv_retval; + } + + mutex_unlock(&srom->lock); + kfree(kernbuf); + + return retval; +} + +/* Provide our own implementation so we can use srom->total_size. */ +loff_t srom_llseek(struct file *filp, loff_t offset, int origin) +{ + struct srom_dev *srom = filp->private_data; + + if (mutex_lock_interruptible(&srom->lock)) + return -ERESTARTSYS; + + switch (origin) { + case SEEK_END: + offset += srom->total_size; + break; + case SEEK_CUR: + offset += filp->f_pos; + break; + } + + if (offset < 0 || offset > srom->total_size) { + offset = -EINVAL; + } else { + filp->f_pos = offset; + filp->f_version = 0; + } + + mutex_unlock(&srom->lock); + + return offset; +} + +static ssize_t total_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct srom_dev *srom = dev_get_drvdata(dev); + return sprintf(buf, "%u\n", srom->total_size); +} + +static ssize_t sector_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct srom_dev *srom = dev_get_drvdata(dev); + return sprintf(buf, "%u\n", srom->sector_size); +} + +static ssize_t page_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct srom_dev *srom = dev_get_drvdata(dev); + return sprintf(buf, "%u\n", srom->page_size); +} + +static struct device_attribute srom_dev_attrs[] = { + __ATTR(total_size, S_IRUGO, total_show, NULL), + __ATTR(sector_size, S_IRUGO, sector_show, NULL), + __ATTR(page_size, S_IRUGO, page_show, NULL), + __ATTR_NULL +}; + +static char *srom_devnode(struct device *dev, mode_t *mode) +{ + *mode = S_IRUGO | S_IWUSR; + return kasprintf(GFP_KERNEL, "srom/%s", dev_name(dev)); +} + +/* + * The fops + */ +static const struct file_operations srom_fops = { + .owner = THIS_MODULE, + .llseek = srom_llseek, + .read = srom_read, + .write = srom_write, + .open = srom_open, + .release = srom_release, +}; + +/** + * srom_setup_minor() - Initialize per-minor information. + * @srom: Per-device SROM state. + * @index: Device to set up. + */ +static int srom_setup_minor(struct srom_dev *srom, int index) +{ + struct device *dev; + int devhdl = srom->hv_devhdl; + + mutex_init(&srom->lock); + + if (_srom_read(devhdl, &srom->total_size, + SROM_TOTAL_SIZE_OFF, sizeof(srom->total_size)) < 0) + return -EIO; + if (_srom_read(devhdl, &srom->sector_size, + SROM_SECTOR_SIZE_OFF, sizeof(srom->sector_size)) < 0) + return -EIO; + if (_srom_read(devhdl, &srom->page_size, + SROM_PAGE_SIZE_OFF, sizeof(srom->page_size)) < 0) + return -EIO; + + dev = device_create(srom_class, &platform_bus, + MKDEV(srom_major, index), srom, "%d", index); + return IS_ERR(dev) ? PTR_ERR(dev) : 0; +} + +/** srom_init() - Initialize the driver's module. */ +static int srom_init(void) +{ + int result, i; + dev_t dev = MKDEV(srom_major, 0); + + /* + * Start with a plausible number of partitions; the krealloc() call + * below will yield about log(srom_devs) additional allocations. + */ + srom_devices = kzalloc(4 * sizeof(struct srom_dev), GFP_KERNEL); + + /* Discover the number of srom partitions. */ + for (i = 0; ; i++) { + int devhdl; + char buf[20]; + struct srom_dev *new_srom_devices = + krealloc(srom_devices, (i+1) * sizeof(struct srom_dev), + GFP_KERNEL | __GFP_ZERO); + if (!new_srom_devices) { + result = -ENOMEM; + goto fail_mem; + } + srom_devices = new_srom_devices; + sprintf(buf, "srom/0/%d", i); + devhdl = hv_dev_open((HV_VirtAddr)buf, 0); + if (devhdl < 0) { + if (devhdl != HV_ENODEV) + pr_notice("srom/%d: hv_dev_open failed: %d.\n", + i, devhdl); + break; + } + srom_devices[i].hv_devhdl = devhdl; + } + srom_devs = i; + + /* Bail out early if we have no partitions at all. */ + if (srom_devs == 0) { + result = -ENODEV; + goto fail_mem; + } + + /* Register our major, and accept a dynamic number. */ + if (srom_major) + result = register_chrdev_region(dev, srom_devs, "srom"); + else { + result = alloc_chrdev_region(&dev, 0, srom_devs, "srom"); + srom_major = MAJOR(dev); + } + if (result < 0) + goto fail_mem; + + /* Register a character device. */ + cdev_init(&srom_cdev, &srom_fops); + srom_cdev.owner = THIS_MODULE; + srom_cdev.ops = &srom_fops; + result = cdev_add(&srom_cdev, dev, srom_devs); + if (result < 0) + goto fail_chrdev; + + /* Create a sysfs class. */ + srom_class = class_create(THIS_MODULE, "srom"); + if (IS_ERR(srom_class)) { + result = PTR_ERR(srom_class); + goto fail_cdev; + } + srom_class->dev_attrs = srom_dev_attrs; + srom_class->devnode = srom_devnode; + + /* Do per-partition initialization */ + for (i = 0; i < srom_devs; i++) { + result = srom_setup_minor(srom_devices + i, i); + if (result < 0) + goto fail_class; + } + + return 0; + +fail_class: + for (i = 0; i < srom_devs; i++) + device_destroy(srom_class, MKDEV(srom_major, i)); + class_destroy(srom_class); +fail_cdev: + cdev_del(&srom_cdev); +fail_chrdev: + unregister_chrdev_region(dev, srom_devs); +fail_mem: + kfree(srom_devices); + return result; +} + +/** srom_cleanup() - Clean up the driver's module. */ +static void srom_cleanup(void) +{ + int i; + for (i = 0; i < srom_devs; i++) + device_destroy(srom_class, MKDEV(srom_major, i)); + class_destroy(srom_class); + cdev_del(&srom_cdev); + unregister_chrdev_region(MKDEV(srom_major, 0), srom_devs); + kfree(srom_devices); +} + +module_init(srom_init); +module_exit(srom_cleanup); From cf8e98d15361f8c594da00a3f7a500787fc1a426 Mon Sep 17 00:00:00 2001 From: Chris Metcalf Date: Wed, 15 Jun 2011 10:35:38 -0400 Subject: [PATCH 005/439] arch/tile: remove useless set_fixmap_nocache() macro TILE doesn't support PAGE_KERNEL_NOCACHE so the macro isn't useful; it's a copy-and-paste from the first version of this header in 2007. Signed-off-by: Chris Metcalf --- arch/tile/include/asm/fixmap.h | 6 ------ 1 file changed, 6 deletions(-) diff --git a/arch/tile/include/asm/fixmap.h b/arch/tile/include/asm/fixmap.h index 51537ff9265a..c66f7933beaa 100644 --- a/arch/tile/include/asm/fixmap.h +++ b/arch/tile/include/asm/fixmap.h @@ -75,12 +75,6 @@ extern void __set_fixmap(enum fixed_addresses idx, #define set_fixmap(idx, phys) \ __set_fixmap(idx, phys, PAGE_KERNEL) -/* - * Some hardware wants to get fixmapped without caching. - */ -#define set_fixmap_nocache(idx, phys) \ - __set_fixmap(idx, phys, PAGE_KERNEL_NOCACHE) - #define clear_fixmap(idx) \ __set_fixmap(idx, 0, __pgprot(0)) From ca7cc5110a313a609da40ae948978a585352564b Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Wed, 13 Jul 2011 13:14:13 +0800 Subject: [PATCH 006/439] ACPI, APEI, ERST, Prevent erst_dbg from loading if ERST is disabled erst_dbg module can not work when ERST is disabled. So disable module loading to provide clearer information to user. Signed-off-by: Huang Ying Signed-off-by: Len Brown --- drivers/acpi/apei/erst-dbg.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/acpi/apei/erst-dbg.c b/drivers/acpi/apei/erst-dbg.c index a4cfb64c86a1..cb04aa43c33a 100644 --- a/drivers/acpi/apei/erst-dbg.c +++ b/drivers/acpi/apei/erst-dbg.c @@ -213,6 +213,10 @@ static struct miscdevice erst_dbg_dev = { static __init int erst_dbg_init(void) { + if (erst_disable) { + pr_info(ERST_DBG_PFX "ERST support is disabled.\n"); + return -ENODEV; + } return misc_register(&erst_dbg_dev); } From d37afc50e618271839f001ea653949eefc728167 Mon Sep 17 00:00:00 2001 From: Chen Gong Date: Wed, 13 Jul 2011 13:14:14 +0800 Subject: [PATCH 007/439] ACPI, APEI, ERST, Fix erst-dbg long record reading issue When we debug ERST table with erst-dbg, if the error record in ERST table is too long(>4K), it can't be read out. So this patch increases the buffer size to 16K to ensure such error records can be read from ERST table. Signed-off-by: Chen Gong Signed-off-by: Huang Ying Signed-off-by: Len Brown --- drivers/acpi/apei/erst-dbg.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/acpi/apei/erst-dbg.c b/drivers/acpi/apei/erst-dbg.c index cb04aa43c33a..903549df809b 100644 --- a/drivers/acpi/apei/erst-dbg.c +++ b/drivers/acpi/apei/erst-dbg.c @@ -33,7 +33,7 @@ #define ERST_DBG_PFX "ERST DBG: " -#define ERST_DBG_RECORD_LEN_MAX 4096 +#define ERST_DBG_RECORD_LEN_MAX 0x4000 static void *erst_dbg_buf; static unsigned int erst_dbg_buf_len; From 5588340d46a484da53bbce8136184d9c7fbc259c Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Wed, 13 Jul 2011 13:14:15 +0800 Subject: [PATCH 008/439] ACPI, APEI, GHES, Do not ratelimit fatal error printk before panic printk is used by GHES to report hardware errors. Normally, the printk will be ratelimited to avoid too many hardware error reports in kernel log. Because there may be thousands or even millions of corrected hardware errors during system running. That is different for fatal hardware error, because system will go panic as soon as possible, there will be no more than several error records. And these error records are valuable for system fault diagnosis, so they should not be ratelimited. Signed-off-by: Huang Ying Signed-off-by: Len Brown --- drivers/acpi/apei/ghes.c | 25 ++++++++++++++----------- 1 file changed, 14 insertions(+), 11 deletions(-) diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c index f703b2881153..f339c0f8369c 100644 --- a/drivers/acpi/apei/ghes.c +++ b/drivers/acpi/apei/ghes.c @@ -360,11 +360,8 @@ static void ghes_do_proc(struct ghes *ghes) } } -static void ghes_print_estatus(const char *pfx, struct ghes *ghes) +static void __ghes_print_estatus(const char *pfx, struct ghes *ghes) { - /* Not more than 2 messages every 5 seconds */ - static DEFINE_RATELIMIT_STATE(ratelimit, 5*HZ, 2); - if (pfx == NULL) { if (ghes_severity(ghes->estatus->error_severity) <= GHES_SEV_CORRECTED) @@ -372,12 +369,18 @@ static void ghes_print_estatus(const char *pfx, struct ghes *ghes) else pfx = KERN_ERR HW_ERR; } - if (__ratelimit(&ratelimit)) { - printk( - "%s""Hardware error from APEI Generic Hardware Error Source: %d\n", - pfx, ghes->generic->header.source_id); - apei_estatus_print(pfx, ghes->estatus); - } + printk("%s""Hardware error from APEI Generic Hardware Error Source: %d\n", + pfx, ghes->generic->header.source_id); + apei_estatus_print(pfx, ghes->estatus); +} + +static void ghes_print_estatus(const char *pfx, struct ghes *ghes) +{ + /* Not more than 2 messages every 5 seconds */ + static DEFINE_RATELIMIT_STATE(ratelimit, 5*HZ, 2); + + if (__ratelimit(&ratelimit)) + __ghes_print_estatus(pfx, ghes); } static int ghes_proc(struct ghes *ghes) @@ -476,7 +479,7 @@ static int ghes_notify_nmi(struct notifier_block *this, if (sev_global >= GHES_SEV_PANIC) { oops_begin(); - ghes_print_estatus(KERN_EMERG HW_ERR, ghes_global); + __ghes_print_estatus(KERN_EMERG HW_ERR, ghes_global); /* reboot to log the error! */ if (panic_timeout == 0) panic_timeout = ghes_panic_timeout; From eecf2f7124834dd1cad21807526a8ea031ba8217 Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Wed, 13 Jul 2011 13:14:16 +0800 Subject: [PATCH 009/439] ACPI, APEI, Add apei_exec_run_optional Some actions in APEI ERST and EINJ tables are optional, for example, ACPI_EINJ_BEGIN_OPERATION action is used to do some preparation for error injection, and firmware may choose to do nothing here. While some other actions are mandatory, for example, firmware must provide ACPI_EINJ_GET_ERROR_TYPE implementation. Original implementation treats all actions as optional (that is, can have no instructions), that may cause issue if firmware does not provide some mandatory actions. To fix this, this patch adds apei_exec_run_optional, which should be used for optional actions. The original apei_exec_run should be used for mandatory actions. Cc: Thomas Renninger Signed-off-by: Huang Ying Signed-off-by: Len Brown --- drivers/acpi/apei/apei-base.c | 9 +++++---- drivers/acpi/apei/apei-internal.h | 13 ++++++++++++- 2 files changed, 17 insertions(+), 5 deletions(-) diff --git a/drivers/acpi/apei/apei-base.c b/drivers/acpi/apei/apei-base.c index 4a904a4bf05f..0714194229da 100644 --- a/drivers/acpi/apei/apei-base.c +++ b/drivers/acpi/apei/apei-base.c @@ -157,9 +157,10 @@ EXPORT_SYMBOL_GPL(apei_exec_noop); * Interpret the specified action. Go through whole action table, * execute all instructions belong to the action. */ -int apei_exec_run(struct apei_exec_context *ctx, u8 action) +int __apei_exec_run(struct apei_exec_context *ctx, u8 action, + bool optional) { - int rc; + int rc = -ENOENT; u32 i, ip; struct acpi_whea_header *entry; apei_exec_ins_func_t run; @@ -198,9 +199,9 @@ rewind: goto rewind; } - return 0; + return !optional && rc < 0 ? rc : 0; } -EXPORT_SYMBOL_GPL(apei_exec_run); +EXPORT_SYMBOL_GPL(__apei_exec_run); typedef int (*apei_exec_entry_func_t)(struct apei_exec_context *ctx, struct acpi_whea_header *entry, diff --git a/drivers/acpi/apei/apei-internal.h b/drivers/acpi/apei/apei-internal.h index ef0581f2094d..f286cf753f32 100644 --- a/drivers/acpi/apei/apei-internal.h +++ b/drivers/acpi/apei/apei-internal.h @@ -50,7 +50,18 @@ static inline u64 apei_exec_ctx_get_output(struct apei_exec_context *ctx) return ctx->value; } -int apei_exec_run(struct apei_exec_context *ctx, u8 action); +int __apei_exec_run(struct apei_exec_context *ctx, u8 action, bool optional); + +static inline int apei_exec_run(struct apei_exec_context *ctx, u8 action) +{ + return __apei_exec_run(ctx, action, 0); +} + +/* It is optional whether the firmware provides the action */ +static inline int apei_exec_run_optional(struct apei_exec_context *ctx, u8 action) +{ + return __apei_exec_run(ctx, action, 1); +} /* Common instruction implementation */ From 392913de7cc7446531922f29c0a4382d8d09626c Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Wed, 13 Jul 2011 13:14:17 +0800 Subject: [PATCH 010/439] ACPI, APEI, Use apei_exec_run_optional in APEI EINJ and ERST This patch changes APEI EINJ and ERST to use apei_exec_run for mandatory actions, and apei_exec_run_optional for optional actions. Cc: Thomas Renninger Signed-off-by: Huang Ying Signed-off-by: Len Brown --- drivers/acpi/apei/einj.c | 4 ++-- drivers/acpi/apei/erst.c | 12 ++++++------ 2 files changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/acpi/apei/einj.c b/drivers/acpi/apei/einj.c index f74b2ea11f21..ab821f1cfa1f 100644 --- a/drivers/acpi/apei/einj.c +++ b/drivers/acpi/apei/einj.c @@ -285,7 +285,7 @@ static int __einj_error_inject(u32 type, u64 param1, u64 param2) einj_exec_ctx_init(&ctx); - rc = apei_exec_run(&ctx, ACPI_EINJ_BEGIN_OPERATION); + rc = apei_exec_run_optional(&ctx, ACPI_EINJ_BEGIN_OPERATION); if (rc) return rc; apei_exec_ctx_set_input(&ctx, type); @@ -323,7 +323,7 @@ static int __einj_error_inject(u32 type, u64 param1, u64 param2) rc = __einj_error_trigger(trigger_paddr); if (rc) return rc; - rc = apei_exec_run(&ctx, ACPI_EINJ_END_OPERATION); + rc = apei_exec_run_optional(&ctx, ACPI_EINJ_END_OPERATION); return rc; } diff --git a/drivers/acpi/apei/erst.c b/drivers/acpi/apei/erst.c index e6cef8e1b534..c8ba9da14e43 100644 --- a/drivers/acpi/apei/erst.c +++ b/drivers/acpi/apei/erst.c @@ -642,7 +642,7 @@ static int __erst_write_to_storage(u64 offset) int rc; erst_exec_ctx_init(&ctx); - rc = apei_exec_run(&ctx, ACPI_ERST_BEGIN_WRITE); + rc = apei_exec_run_optional(&ctx, ACPI_ERST_BEGIN_WRITE); if (rc) return rc; apei_exec_ctx_set_input(&ctx, offset); @@ -666,7 +666,7 @@ static int __erst_write_to_storage(u64 offset) if (rc) return rc; val = apei_exec_ctx_get_output(&ctx); - rc = apei_exec_run(&ctx, ACPI_ERST_END); + rc = apei_exec_run_optional(&ctx, ACPI_ERST_END); if (rc) return rc; @@ -681,7 +681,7 @@ static int __erst_read_from_storage(u64 record_id, u64 offset) int rc; erst_exec_ctx_init(&ctx); - rc = apei_exec_run(&ctx, ACPI_ERST_BEGIN_READ); + rc = apei_exec_run_optional(&ctx, ACPI_ERST_BEGIN_READ); if (rc) return rc; apei_exec_ctx_set_input(&ctx, offset); @@ -709,7 +709,7 @@ static int __erst_read_from_storage(u64 record_id, u64 offset) if (rc) return rc; val = apei_exec_ctx_get_output(&ctx); - rc = apei_exec_run(&ctx, ACPI_ERST_END); + rc = apei_exec_run_optional(&ctx, ACPI_ERST_END); if (rc) return rc; @@ -724,7 +724,7 @@ static int __erst_clear_from_storage(u64 record_id) int rc; erst_exec_ctx_init(&ctx); - rc = apei_exec_run(&ctx, ACPI_ERST_BEGIN_CLEAR); + rc = apei_exec_run_optional(&ctx, ACPI_ERST_BEGIN_CLEAR); if (rc) return rc; apei_exec_ctx_set_input(&ctx, record_id); @@ -748,7 +748,7 @@ static int __erst_clear_from_storage(u64 record_id) if (rc) return rc; val = apei_exec_ctx_get_output(&ctx); - rc = apei_exec_run(&ctx, ACPI_ERST_END); + rc = apei_exec_run_optional(&ctx, ACPI_ERST_END); if (rc) return rc; From 86cd47334b00b6aa9b5d0ebf389a6fe76f21c641 Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Wed, 13 Jul 2011 13:14:18 +0800 Subject: [PATCH 011/439] ACPI, APEI, GHES, Prevent GHES to be built as module GHES (Generic Hardware Error Source) is used to process hardware error notification in firmware first mode. But because firmware first mode can be turned on but can not be turned off, it is unreasonable to unload the GHES module with firmware first mode turned on. To avoid confusion, this patch makes GHES can be enabled/disabled in configuration time, but not built as module and unloaded at run time. Signed-off-by: Huang Ying Reviewed-by: Andi Kleen Reviewed-by: Matthew Garrett Signed-off-by: Len Brown --- drivers/acpi/apei/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/acpi/apei/Kconfig b/drivers/acpi/apei/Kconfig index f739a70b1c70..3f45dde17aec 100644 --- a/drivers/acpi/apei/Kconfig +++ b/drivers/acpi/apei/Kconfig @@ -10,7 +10,7 @@ config ACPI_APEI error injection. config ACPI_APEI_GHES - tristate "APEI Generic Hardware Error Source" + bool "APEI Generic Hardware Error Source" depends on ACPI_APEI && X86 select ACPI_HED help From b6a9501658530d8b8374e37f1edb549039a8a260 Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Wed, 13 Jul 2011 13:14:19 +0800 Subject: [PATCH 012/439] ACPI, APEI, GHES, Support disable GHES at boot time Some machine may have broken firmware so that GHES and firmware first mode should be disabled. This patch adds support to that. Signed-off-by: Huang Ying Reviewed-by: Andi Kleen Reviewed-by: Matthew Garrett Signed-off-by: Len Brown --- drivers/acpi/apei/ghes.c | 8 ++++++++ drivers/acpi/apei/hest.c | 17 +++++++++-------- include/acpi/apei.h | 1 + 3 files changed, 18 insertions(+), 8 deletions(-) diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c index f339c0f8369c..b142b94bf8b2 100644 --- a/drivers/acpi/apei/ghes.c +++ b/drivers/acpi/apei/ghes.c @@ -77,6 +77,9 @@ struct ghes { }; }; +int ghes_disable; +module_param_named(disable, ghes_disable, bool, 0); + static int ghes_panic_timeout __read_mostly = 30; /* @@ -665,6 +668,11 @@ static int __init ghes_init(void) return -EINVAL; } + if (ghes_disable) { + pr_info(GHES_PFX "GHES is not enabled!\n"); + return -EINVAL; + } + rc = ghes_ioremap_init(); if (rc) goto err; diff --git a/drivers/acpi/apei/hest.c b/drivers/acpi/apei/hest.c index 181bc2f7bb74..05fee06f4d6e 100644 --- a/drivers/acpi/apei/hest.c +++ b/drivers/acpi/apei/hest.c @@ -231,16 +231,17 @@ void __init acpi_hest_init(void) goto err; } - rc = apei_hest_parse(hest_parse_ghes_count, &ghes_count); - if (rc) - goto err; - - rc = hest_ghes_dev_register(ghes_count); - if (!rc) { - pr_info(HEST_PFX "Table parsing has been initialized.\n"); - return; + if (!ghes_disable) { + rc = apei_hest_parse(hest_parse_ghes_count, &ghes_count); + if (rc) + goto err; + rc = hest_ghes_dev_register(ghes_count); + if (rc) + goto err; } + pr_info(HEST_PFX "Table parsing has been initialized.\n"); + return; err: hest_disable = 1; } diff --git a/include/acpi/apei.h b/include/acpi/apei.h index e67b523a50e1..d40bc5521fcb 100644 --- a/include/acpi/apei.h +++ b/include/acpi/apei.h @@ -18,6 +18,7 @@ extern int hest_disable; extern int erst_disable; +extern int ghes_disable; #ifdef CONFIG_ACPI_APEI void __init acpi_hest_init(void); From eccddd32ced0df8f9130024157bf8d37df860d76 Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Wed, 13 Jul 2011 13:14:20 +0800 Subject: [PATCH 013/439] ACPI, APEI, Add APEI bit support in generic _OSC call In APEI firmware first mode, hardware error is reported by hardware to firmware firstly, then firmware reports the error to Linux in a GHES error record via POLL/SCI/IRQ/NMI etc. This may result in some issues if OS has no full APEI support. So some firmware implementation will work in a back-compatible mode by default. Where firmware will only notify OS in old-fashion, without GHES record. For example, for a fatal hardware error, only NMI is signaled, no GHES record. To gain full APEI power on these machines, APEI bit in generic _OSC call can be specified to tell firmware that Linux has full APEI support. This patch adds the APEI bit support in generic _OSC call. Signed-off-by: Huang Ying Reviewed-by: Andi Kleen Reviewed-by: Matthew Garrett Signed-off-by: Len Brown --- drivers/acpi/bus.c | 16 ++++++++++++++-- include/linux/acpi.h | 2 ++ 2 files changed, 16 insertions(+), 2 deletions(-) diff --git a/drivers/acpi/bus.c b/drivers/acpi/bus.c index d1e06c182cdb..43ce3b0c4921 100644 --- a/drivers/acpi/bus.c +++ b/drivers/acpi/bus.c @@ -39,6 +39,7 @@ #include #include #include +#include #include #include @@ -519,6 +520,7 @@ out_kfree: } EXPORT_SYMBOL(acpi_run_osc); +bool osc_sb_apei_support_acked; static u8 sb_uuid_str[] = "0811B06E-4A27-44F9-8D60-3CBBC22E7B48"; static void acpi_bus_osc_support(void) { @@ -541,11 +543,21 @@ static void acpi_bus_osc_support(void) #if defined(CONFIG_ACPI_PROCESSOR) || defined(CONFIG_ACPI_PROCESSOR_MODULE) capbuf[OSC_SUPPORT_TYPE] |= OSC_SB_PPC_OST_SUPPORT; #endif + +#ifdef CONFIG_ACPI_APEI_GHES + if (!ghes_disable) + capbuf[OSC_SUPPORT_TYPE] |= OSC_SB_APEI_SUPPORT; +#endif if (ACPI_FAILURE(acpi_get_handle(NULL, "\\_SB", &handle))) return; - if (ACPI_SUCCESS(acpi_run_osc(handle, &context))) + if (ACPI_SUCCESS(acpi_run_osc(handle, &context))) { + u32 *capbuf_ret = context.ret.pointer; + if (context.ret.length > OSC_SUPPORT_TYPE) + osc_sb_apei_support_acked = + capbuf_ret[OSC_SUPPORT_TYPE] & OSC_SB_APEI_SUPPORT; kfree(context.ret.pointer); - /* do we need to check the returned cap? Sounds no */ + } + /* do we need to check other returned cap? Sounds no */ } /* -------------------------------------------------------------------------- diff --git a/include/linux/acpi.h b/include/linux/acpi.h index 1deb2a73c2da..e19527de6a93 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -280,6 +280,8 @@ acpi_status acpi_run_osc(acpi_handle handle, struct acpi_osc_context *context); #define OSC_SB_CPUHP_OST_SUPPORT 8 #define OSC_SB_APEI_SUPPORT 16 +extern bool osc_sb_apei_support_acked; + /* PCI defined _OSC bits */ /* _OSC DW1 Definition (OS Support Fields) */ #define OSC_EXT_PCI_CONFIG_SUPPORT 1 From 9fb0bfe1408d5506b7b83d13d1eed573fd71d67d Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Wed, 13 Jul 2011 13:14:21 +0800 Subject: [PATCH 014/439] ACPI, APEI, Add WHEA _OSC support APEI firmware first mode must be turned on explicitly on some machines, otherwise there may be no GHES hardware error record for hardware error notification. APEI bit in generic _OSC call can be used to do that, but on some machine, a special WHEA _OSC call must be used. This patch adds the support to that WHEA _OSC call. Signed-off-by: Huang Ying Reviewed-by: Andi Kleen Reviewed-by: Matthew Garrett Signed-off-by: Len Brown --- drivers/acpi/apei/apei-base.c | 26 ++++++++++++++++++++++++++ drivers/acpi/apei/apei-internal.h | 2 ++ drivers/acpi/apei/ghes.c | 10 ++++++++++ 3 files changed, 38 insertions(+) diff --git a/drivers/acpi/apei/apei-base.c b/drivers/acpi/apei/apei-base.c index 0714194229da..8041248fce9b 100644 --- a/drivers/acpi/apei/apei-base.c +++ b/drivers/acpi/apei/apei-base.c @@ -604,3 +604,29 @@ struct dentry *apei_get_debugfs_dir(void) return dapei; } EXPORT_SYMBOL_GPL(apei_get_debugfs_dir); + +int apei_osc_setup(void) +{ + static u8 whea_uuid_str[] = "ed855e0c-6c90-47bf-a62a-26de0fc5ad5c"; + acpi_handle handle; + u32 capbuf[3]; + struct acpi_osc_context context = { + .uuid_str = whea_uuid_str, + .rev = 1, + .cap.length = sizeof(capbuf), + .cap.pointer = capbuf, + }; + + capbuf[OSC_QUERY_TYPE] = OSC_QUERY_ENABLE; + capbuf[OSC_SUPPORT_TYPE] = 0; + capbuf[OSC_CONTROL_TYPE] = 0; + + if (ACPI_FAILURE(acpi_get_handle(NULL, "\\_SB", &handle)) + || ACPI_FAILURE(acpi_run_osc(handle, &context))) + return -EIO; + else { + kfree(context.ret.pointer); + return 0; + } +} +EXPORT_SYMBOL_GPL(apei_osc_setup); diff --git a/drivers/acpi/apei/apei-internal.h b/drivers/acpi/apei/apei-internal.h index f286cf753f32..f57050e7a5e7 100644 --- a/drivers/acpi/apei/apei-internal.h +++ b/drivers/acpi/apei/apei-internal.h @@ -124,4 +124,6 @@ void apei_estatus_print(const char *pfx, const struct acpi_hest_generic_status *estatus); int apei_estatus_check_header(const struct acpi_hest_generic_status *estatus); int apei_estatus_check(const struct acpi_hest_generic_status *estatus); + +int apei_osc_setup(void); #endif diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c index b142b94bf8b2..b1390a61cde1 100644 --- a/drivers/acpi/apei/ghes.c +++ b/drivers/acpi/apei/ghes.c @@ -681,6 +681,16 @@ static int __init ghes_init(void) if (rc) goto err_ioremap_exit; + rc = apei_osc_setup(); + if (rc == 0 && osc_sb_apei_support_acked) + pr_info(GHES_PFX "APEI firmware first mode is enabled by APEI bit and WHEA _OSC.\n"); + else if (rc == 0 && !osc_sb_apei_support_acked) + pr_info(GHES_PFX "APEI firmware first mode is enabled by WHEA _OSC.\n"); + else if (rc && osc_sb_apei_support_acked) + pr_info(GHES_PFX "APEI firmware first mode is enabled by APEI bit.\n"); + else + pr_info(GHES_PFX "Failed to enable APEI firmware first mode.\n"); + return 0; err_ioremap_exit: ghes_ioremap_exit(); From c8cefe307d79c57b32ca1d4c9ebff528f8dd914c Mon Sep 17 00:00:00 2001 From: Bob Moore Date: Tue, 14 Jun 2011 10:42:53 +0800 Subject: [PATCH 015/439] ACPICA: Load operator: re-instate most restrictions on incoming table signature Now, only allow "SSDT" "OEM", and a null signature. Signed-off-by: Bob Moore Signed-off-by: Lin Ming Signed-off-by: Len Brown --- drivers/acpi/acpica/tbinstal.c | 27 ++++++++++++++++++++++----- 1 file changed, 22 insertions(+), 5 deletions(-) diff --git a/drivers/acpi/acpica/tbinstal.c b/drivers/acpi/acpica/tbinstal.c index 48db0944ce4a..62365f6075dd 100644 --- a/drivers/acpi/acpica/tbinstal.c +++ b/drivers/acpi/acpica/tbinstal.c @@ -126,12 +126,29 @@ acpi_tb_add_table(struct acpi_table_desc *table_desc, u32 *table_index) } /* - * Originally, we checked the table signature for "SSDT" or "PSDT" here. - * Next, we added support for OEMx tables, signature "OEM". - * Valid tables were encountered with a null signature, so we've just - * given up on validating the signature, since it seems to be a waste - * of code. The original code was removed (05/2008). + * Validate the incoming table signature. + * + * 1) Originally, we checked the table signature for "SSDT" or "PSDT". + * 2) We added support for OEMx tables, signature "OEM". + * 3) Valid tables were encountered with a null signature, so we just + * gave up on validating the signature, (05/2008). + * 4) We encountered non-AML tables such as the MADT, which caused + * interpreter errors and kernel faults. So now, we once again allow + * only "SSDT", "OEMx", and now, also a null signature. (05/2011). */ + if ((table_desc->pointer->signature[0] != 0x00) && + (!ACPI_COMPARE_NAME(table_desc->pointer->signature, ACPI_SIG_SSDT)) + && (ACPI_STRNCMP(table_desc->pointer->signature, "OEM", 3))) { + ACPI_ERROR((AE_INFO, + "Table has invalid signature [%4.4s] (0x%8.8X), must be SSDT or OEMx", + acpi_ut_valid_acpi_name(*(u32 *)table_desc-> + pointer-> + signature) ? table_desc-> + pointer->signature : "????", + *(u32 *)table_desc->pointer->signature)); + + return_ACPI_STATUS(AE_BAD_SIGNATURE); + } (void)acpi_ut_acquire_mutex(ACPI_MTX_TABLES); From f631f9cafbc47bbfe5eb2aa831cd323175298448 Mon Sep 17 00:00:00 2001 From: Bob Moore Date: Tue, 14 Jun 2011 10:44:51 +0800 Subject: [PATCH 016/439] ACPICA: Add missing _TDL to list of predefined names Affects both iASL and core ACPICA. Signed-off-by: Bob Moore Signed-off-by: Lin Ming Signed-off-by: Len Brown --- drivers/acpi/acpica/acpredef.h | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/acpi/acpica/acpredef.h b/drivers/acpi/acpica/acpredef.h index 94e73c97cf85..c445cca490ea 100644 --- a/drivers/acpi/acpica/acpredef.h +++ b/drivers/acpi/acpica/acpredef.h @@ -468,6 +468,7 @@ static const union acpi_predefined_info predefined_names[] = {{"_SWS", 0, ACPI_RTYPE_INTEGER}}, {{"_TC1", 0, ACPI_RTYPE_INTEGER}}, {{"_TC2", 0, ACPI_RTYPE_INTEGER}}, + {{"_TDL", 0, ACPI_RTYPE_INTEGER}}, {{"_TIP", 1, ACPI_RTYPE_INTEGER}}, {{"_TIV", 1, ACPI_RTYPE_INTEGER}}, {{"_TMP", 0, ACPI_RTYPE_INTEGER}}, From 80f40ce0f10e87d9a27f1e29325c6f39245fbcb1 Mon Sep 17 00:00:00 2001 From: Bob Moore Date: Tue, 14 Jun 2011 10:45:57 +0800 Subject: [PATCH 017/439] ACPICA: Update to version 20110527 Version 20110527. Signed-off-by: Bob Moore Signed-off-by: Lin Ming Signed-off-by: Len Brown --- include/acpi/acpixf.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/acpi/acpixf.h b/include/acpi/acpixf.h index 2ed0a8486c19..06853f76d81b 100644 --- a/include/acpi/acpixf.h +++ b/include/acpi/acpixf.h @@ -47,7 +47,7 @@ /* Current ACPICA subsystem version in YYYYMMDD format */ -#define ACPI_CA_VERSION 0x20110413 +#define ACPI_CA_VERSION 0x20110527 #include "actypes.h" #include "actbl.h" From d57b23ad0ca7a46931e4d98eb6b4b73b112f0969 Mon Sep 17 00:00:00 2001 From: Bob Moore Date: Mon, 4 Jul 2011 08:24:03 +0000 Subject: [PATCH 018/439] ACPICA: Add option to disable method return value validation and repair Runtime option can be used to disable return value repair if this is causing a problem on a particular machine. Signed-off-by: Bob Moore Signed-off-by: Lin Ming Signed-off-by: Len Brown --- drivers/acpi/acpica/acglobal.h | 6 ++++++ drivers/acpi/acpica/nspredef.c | 18 ++++++++++++------ include/acpi/acpixf.h | 1 + 3 files changed, 19 insertions(+), 6 deletions(-) diff --git a/drivers/acpi/acpica/acglobal.h b/drivers/acpi/acpica/acglobal.h index 73863d86f022..76dc02f15574 100644 --- a/drivers/acpi/acpica/acglobal.h +++ b/drivers/acpi/acpica/acglobal.h @@ -126,6 +126,12 @@ u8 ACPI_INIT_GLOBAL(acpi_gbl_copy_dsdt_locally, FALSE); */ u8 ACPI_INIT_GLOBAL(acpi_gbl_truncate_io_addresses, FALSE); +/* + * Disable runtime checking and repair of values returned by control methods. + * Use only if the repair is causing a problem on a particular machine. + */ +u8 ACPI_INIT_GLOBAL(acpi_gbl_disable_auto_repair, FALSE); + /* acpi_gbl_FADT is a local copy of the FADT, converted to a common format. */ struct acpi_table_fadt acpi_gbl_FADT; diff --git a/drivers/acpi/acpica/nspredef.c b/drivers/acpi/acpica/nspredef.c index 9fb03fa8ffde..607a9a035262 100644 --- a/drivers/acpi/acpica/nspredef.c +++ b/drivers/acpi/acpica/nspredef.c @@ -193,14 +193,20 @@ acpi_ns_check_predefined_names(struct acpi_namespace_node *node, } /* - * 1) We have a return value, but if one wasn't expected, just exit, this is - * not a problem. For example, if the "Implicit Return" feature is - * enabled, methods will always return a value. + * Return value validation and possible repair. * - * 2) If the return value can be of any type, then we cannot perform any - * validation, exit. + * 1) Don't perform return value validation/repair if this feature + * has been disabled via a global option. + * + * 2) We have a return value, but if one wasn't expected, just exit, + * this is not a problem. For example, if the "Implicit Return" + * feature is enabled, methods will always return a value. + * + * 3) If the return value can be of any type, then we cannot perform + * any validation, just exit. */ - if ((!predefined->info.expected_btypes) || + if (acpi_gbl_disable_auto_repair || + (!predefined->info.expected_btypes) || (predefined->info.expected_btypes == ACPI_RTYPE_ALL)) { goto cleanup; } diff --git a/include/acpi/acpixf.h b/include/acpi/acpixf.h index 06853f76d81b..d42d7ced3943 100644 --- a/include/acpi/acpixf.h +++ b/include/acpi/acpixf.h @@ -69,6 +69,7 @@ extern u32 acpi_gbl_trace_flags; extern u32 acpi_gbl_enable_aml_debug_object; extern u8 acpi_gbl_copy_dsdt_locally; extern u8 acpi_gbl_truncate_io_addresses; +extern u8 acpi_gbl_disable_auto_repair; extern u32 acpi_current_gpe_count; extern struct acpi_table_fadt acpi_gbl_FADT; From 8f9c91273e36e5762c617c23e4fd48d5172e0dac Mon Sep 17 00:00:00 2001 From: Fenghua Yu Date: Mon, 4 Jul 2011 08:36:16 +0000 Subject: [PATCH 019/439] ACPICA: Do not repair _TSS return package if _PSS is present We can only sort the _TSS return package if there is no _PSS in the same scope. This is because if _PSS is present, the ACPI specification dictates that the _TSS Power Dissipation field is to be ignored, and therefore some BIOSs leave garbage values in the _TSS Power field(s). In this case, it is best to just return the _TSS package as-is. Reported-by: Fenghua Yu Signed-off-by: Fenghua Yu Signed-off-by: Bob Moore Signed-off-by: Lin Ming Signed-off-by: Len Brown --- drivers/acpi/acpica/aclocal.h | 1 + drivers/acpi/acpica/nspredef.c | 1 + drivers/acpi/acpica/nsrepair2.c | 15 +++++++++++++++ 3 files changed, 17 insertions(+) diff --git a/drivers/acpi/acpica/aclocal.h b/drivers/acpi/acpica/aclocal.h index c7f743ca395b..5552125d8340 100644 --- a/drivers/acpi/acpica/aclocal.h +++ b/drivers/acpi/acpica/aclocal.h @@ -357,6 +357,7 @@ struct acpi_predefined_data { char *pathname; const union acpi_predefined_info *predefined; union acpi_operand_object *parent_package; + struct acpi_namespace_node *node; u32 flags; u8 node_flags; }; diff --git a/drivers/acpi/acpica/nspredef.c b/drivers/acpi/acpica/nspredef.c index 607a9a035262..c845c8089f39 100644 --- a/drivers/acpi/acpica/nspredef.c +++ b/drivers/acpi/acpica/nspredef.c @@ -218,6 +218,7 @@ acpi_ns_check_predefined_names(struct acpi_namespace_node *node, goto cleanup; } data->predefined = predefined; + data->node = node; data->node_flags = node->flags; data->pathname = pathname; diff --git a/drivers/acpi/acpica/nsrepair2.c b/drivers/acpi/acpica/nsrepair2.c index 973883babee1..024c4f263f87 100644 --- a/drivers/acpi/acpica/nsrepair2.c +++ b/drivers/acpi/acpica/nsrepair2.c @@ -503,6 +503,21 @@ acpi_ns_repair_TSS(struct acpi_predefined_data *data, { union acpi_operand_object *return_object = *return_object_ptr; acpi_status status; + struct acpi_namespace_node *node; + + /* + * We can only sort the _TSS return package if there is no _PSS in the + * same scope. This is because if _PSS is present, the ACPI specification + * dictates that the _TSS Power Dissipation field is to be ignored, and + * therefore some BIOSs leave garbage values in the _TSS Power field(s). + * In this case, it is best to just return the _TSS package as-is. + * (May, 2011) + */ + status = + acpi_ns_get_node(data->node, "^_PSS", ACPI_NS_NO_UPSEARCH, &node); + if (ACPI_SUCCESS(status)) { + return (AE_OK); + } status = acpi_ns_check_sorted_list(data, return_object, 5, 1, ACPI_SORT_DESCENDING, From 3eb208f0a36cf7a86953afa7a4eb6776294e6768 Mon Sep 17 00:00:00 2001 From: Bob Moore Date: Mon, 4 Jul 2011 08:38:51 +0000 Subject: [PATCH 020/439] ACPICA: Update to version 20110623 Version 20110623. Signed-off-by: Bob Moore Signed-off-by: Lin Ming Signed-off-by: Len Brown --- include/acpi/acpixf.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/acpi/acpixf.h b/include/acpi/acpixf.h index d42d7ced3943..f554a9313b43 100644 --- a/include/acpi/acpixf.h +++ b/include/acpi/acpixf.h @@ -47,7 +47,7 @@ /* Current ACPICA subsystem version in YYYYMMDD format */ -#define ACPI_CA_VERSION 0x20110527 +#define ACPI_CA_VERSION 0x20110623 #include "actypes.h" #include "actbl.h" From 1dd5c715e5b7524da8c1030f5cf1ea903e45c457 Mon Sep 17 00:00:00 2001 From: Lan Tianyu Date: Fri, 1 Jul 2011 16:03:15 +0800 Subject: [PATCH 021/439] ACPI / SBS: Add getting state operation in the acpi_sbs_battery_get_property() https://bugzilla.kernel.org/show_bug.cgi?id=24492 Signed-off-by: Lan Tianyu Signed-off-by: Len Brown --- drivers/acpi/sbs.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/acpi/sbs.c b/drivers/acpi/sbs.c index 51ae3794ec7f..01dad1bedd17 100644 --- a/drivers/acpi/sbs.c +++ b/drivers/acpi/sbs.c @@ -130,6 +130,9 @@ struct acpi_sbs { #define to_acpi_sbs(x) container_of(x, struct acpi_sbs, charger) +static int acpi_sbs_remove(struct acpi_device *device, int type); +static int acpi_battery_get_state(struct acpi_battery *battery); + static inline int battery_scale(int log) { int scale = 1; @@ -195,6 +198,8 @@ static int acpi_sbs_battery_get_property(struct power_supply *psy, if ((!battery->present) && psp != POWER_SUPPLY_PROP_PRESENT) return -ENODEV; + + acpi_battery_get_state(battery); switch (psp) { case POWER_SUPPLY_PROP_STATUS: if (battery->rate_now < 0) @@ -903,8 +908,6 @@ static void acpi_sbs_callback(void *context) } } -static int acpi_sbs_remove(struct acpi_device *device, int type); - static int acpi_sbs_add(struct acpi_device *device) { struct acpi_sbs *sbs; From e4108292cc5b5ca07abc83af31a78338362810ca Mon Sep 17 00:00:00 2001 From: Lan Tianyu Date: Fri, 1 Jul 2011 16:03:39 +0800 Subject: [PATCH 022/439] ACPI / SBS: Correct the value of power_now and power_avg in the sysfs https://bugzilla.kernel.org/show_bug.cgi?id=24492 Signed-off-by: Lan Tianyu Signed-off-by: Len Brown --- drivers/acpi/sbs.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/acpi/sbs.c b/drivers/acpi/sbs.c index 01dad1bedd17..c517ec8106e9 100644 --- a/drivers/acpi/sbs.c +++ b/drivers/acpi/sbs.c @@ -230,11 +230,17 @@ static int acpi_sbs_battery_get_property(struct power_supply *psy, case POWER_SUPPLY_PROP_POWER_NOW: val->intval = abs(battery->rate_now) * acpi_battery_ipscale(battery) * 1000; + val->intval *= (acpi_battery_mode(battery)) ? + (battery->voltage_now * + acpi_battery_vscale(battery) / 1000) : 1; break; case POWER_SUPPLY_PROP_CURRENT_AVG: case POWER_SUPPLY_PROP_POWER_AVG: val->intval = abs(battery->rate_avg) * acpi_battery_ipscale(battery) * 1000; + val->intval *= (acpi_battery_mode(battery)) ? + (battery->voltage_now * + acpi_battery_vscale(battery) / 1000) : 1; break; case POWER_SUPPLY_PROP_CAPACITY: val->intval = battery->state_of_charge; From ae6f61870490c10a0b0436e5afffa00c9dacffef Mon Sep 17 00:00:00 2001 From: Lan Tianyu Date: Thu, 30 Jun 2011 11:32:40 +0800 Subject: [PATCH 023/439] ACPI / Battery: Add the power unit macro This patch is cosmetic only, and makes no functional change. Signed-off-by: Lan Tianyu Signed-off-by: Len Brown --- drivers/acpi/battery.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c index fcc13ac0aa18..611434f413db 100644 --- a/drivers/acpi/battery.c +++ b/drivers/acpi/battery.c @@ -55,6 +55,9 @@ #define ACPI_BATTERY_NOTIFY_INFO 0x81 #define ACPI_BATTERY_NOTIFY_THRESHOLD 0x82 +/* Battery power unit: 0 means mW, 1 means mA */ +#define ACPI_BATTERY_POWER_UNIT_MA 1 + #define _COMPONENT ACPI_BATTERY_COMPONENT ACPI_MODULE_NAME("battery"); @@ -301,7 +304,8 @@ static enum power_supply_property energy_battery_props[] = { #ifdef CONFIG_ACPI_PROCFS_POWER inline char *acpi_battery_units(struct acpi_battery *battery) { - return (battery->power_unit)?"mA":"mW"; + return (battery->power_unit == ACPI_BATTERY_POWER_UNIT_MA) ? + "mA" : "mW"; } #endif @@ -544,7 +548,7 @@ static int sysfs_add_battery(struct acpi_battery *battery) { int result; - if (battery->power_unit) { + if (battery->power_unit == ACPI_BATTERY_POWER_UNIT_MA) { battery->bat.properties = charge_battery_props; battery->bat.num_properties = ARRAY_SIZE(charge_battery_props); @@ -575,7 +579,8 @@ static void sysfs_remove_battery(struct acpi_battery *battery) static void acpi_battery_quirks(struct acpi_battery *battery) { - if (dmi_name_in_vendors("Acer") && battery->power_unit) { + if (dmi_name_in_vendors("Acer") && + battery->power_unit == ACPI_BATTERY_POWER_UNIT_MA) { set_bit(ACPI_BATTERY_QUIRK_SIGNED16_CURRENT, &battery->flags); } } From 55003b2105a4578736f3e868fbaa889bb1ff3ce0 Mon Sep 17 00:00:00 2001 From: Lan Tianyu Date: Thu, 30 Jun 2011 11:33:12 +0800 Subject: [PATCH 024/439] ACPI / Battery: Change 16-bit signed negative battery current into correct value This patch is for some machines which report the battery current as a 16-bit signed negative when it is charging. This is caused by DSDT bug. The commit bc76f90b8a5cf4aceedf210d08d5e8292f820cec has resolved the problem for Acer laptops. But some other machines also have such problem. https://bugzilla.kernel.org/show_bug.cgi?id=33722 Since it is improper that the current is above 32A on laptops whether on AC or on battery, this patch is to check the current and take its absolute value as current and producing a message when it is negative in s16. Remove Acer quirk, as this workaround handles Acer too. Signed-off-by: Lan Tianyu Signed-off-by: Len Brown --- drivers/acpi/battery.c | 26 ++++++++++---------------- 1 file changed, 10 insertions(+), 16 deletions(-) diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c index 611434f413db..ff1ff70da384 100644 --- a/drivers/acpi/battery.c +++ b/drivers/acpi/battery.c @@ -94,11 +94,6 @@ MODULE_DEVICE_TABLE(acpi, battery_device_ids); enum { ACPI_BATTERY_ALARM_PRESENT, ACPI_BATTERY_XINFO_PRESENT, - /* For buggy DSDTs that report negative 16-bit values for either - * charging or discharging current and/or report 0 as 65536 - * due to bad math. - */ - ACPI_BATTERY_QUIRK_SIGNED16_CURRENT, ACPI_BATTERY_QUIRK_PERCENTAGE_CAPACITY, }; @@ -465,9 +460,17 @@ static int acpi_battery_get_state(struct acpi_battery *battery) battery->update_time = jiffies; kfree(buffer.pointer); - if (test_bit(ACPI_BATTERY_QUIRK_SIGNED16_CURRENT, &battery->flags) && - battery->rate_now != -1) + /* For buggy DSDTs that report negative 16-bit values for either + * charging or discharging current and/or report 0 as 65536 + * due to bad math. + */ + if (battery->power_unit == ACPI_BATTERY_POWER_UNIT_MA && + battery->rate_now != ACPI_BATTERY_VALUE_UNKNOWN && + (s16)(battery->rate_now) < 0) { battery->rate_now = abs((s16)battery->rate_now); + printk_once(KERN_WARNING FW_BUG "battery: (dis)charge rate" + " invalid.\n"); + } if (test_bit(ACPI_BATTERY_QUIRK_PERCENTAGE_CAPACITY, &battery->flags) && battery->capacity_now >= 0 && battery->capacity_now <= 100) @@ -577,14 +580,6 @@ static void sysfs_remove_battery(struct acpi_battery *battery) battery->bat.dev = NULL; } -static void acpi_battery_quirks(struct acpi_battery *battery) -{ - if (dmi_name_in_vendors("Acer") && - battery->power_unit == ACPI_BATTERY_POWER_UNIT_MA) { - set_bit(ACPI_BATTERY_QUIRK_SIGNED16_CURRENT, &battery->flags); - } -} - /* * According to the ACPI spec, some kinds of primary batteries can * report percentage battery remaining capacity directly to OS. @@ -628,7 +623,6 @@ static int acpi_battery_update(struct acpi_battery *battery) result = acpi_battery_get_info(battery); if (result) return result; - acpi_battery_quirks(battery); acpi_battery_init_alarm(battery); } if (!battery->bat.dev) From 7b78622d0f9df9f274a319eea1494240119d3734 Mon Sep 17 00:00:00 2001 From: Lan Tianyu Date: Thu, 30 Jun 2011 11:33:27 +0800 Subject: [PATCH 025/439] ACPI / Battery: Rename acpi_battery_quirks2 with acpi_battery_quirks This patch is cosmetic only, and makes no functional change. Since the acpi_battery_quirks has been deleted, rename acpi_battery_quirks2 with acpi_battery_quirks to clean the code. Signed-off-by: Lan Tianyu Signed-off-by: Len Brown --- drivers/acpi/battery.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c index ff1ff70da384..ba7926fce4fb 100644 --- a/drivers/acpi/battery.c +++ b/drivers/acpi/battery.c @@ -592,7 +592,7 @@ static void sysfs_remove_battery(struct acpi_battery *battery) * * Handle this correctly so that they won't break userspace. */ -static void acpi_battery_quirks2(struct acpi_battery *battery) +static void acpi_battery_quirks(struct acpi_battery *battery) { if (test_bit(ACPI_BATTERY_QUIRK_PERCENTAGE_CAPACITY, &battery->flags)) return ; @@ -628,7 +628,7 @@ static int acpi_battery_update(struct acpi_battery *battery) if (!battery->bat.dev) sysfs_add_battery(battery); result = acpi_battery_get_state(battery); - acpi_battery_quirks2(battery); + acpi_battery_quirks(battery); return result; } From d5a5911b3278bad6515a9958f7318f74d534ef64 Mon Sep 17 00:00:00 2001 From: Lan Tianyu Date: Thu, 30 Jun 2011 11:33:40 +0800 Subject: [PATCH 026/439] ACPI / Battery: Add the hibernation process in the battery_notify() The Commit 25be58215 has added a PM notifier to refresh the sys in order to deal with the unit change of the Battery Present Rate. But it just consided the suspend situation. The problem also will happen during the hibernation according the bug 28192. https://bugzilla.kernel.org/show_bug.cgi?id=28192 This patch adds the hibernation process and fix the bug. Signed-off-by: Lan Tianyu Signed-off-by: Len Brown --- drivers/acpi/battery.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c index ba7926fce4fb..2fe7cfd95680 100644 --- a/drivers/acpi/battery.c +++ b/drivers/acpi/battery.c @@ -947,6 +947,7 @@ static int battery_notify(struct notifier_block *nb, struct acpi_battery *battery = container_of(nb, struct acpi_battery, pm_nb); switch (mode) { + case PM_POST_HIBERNATION: case PM_POST_SUSPEND: sysfs_remove_battery(battery); sysfs_add_battery(battery); From 6e17fb6aa1a67afa1827ae317c3594040f055730 Mon Sep 17 00:00:00 2001 From: Lan Tianyu Date: Thu, 30 Jun 2011 11:33:58 +0800 Subject: [PATCH 027/439] ACPI / Battery: Add the check before refresh sysfs in the battery_notify() In the commit 25be5821, add the refresh sysfs when system resumes from suspending. But it didn't check that the battery exists. This will cause battery sysfs files added when the battery doesn't exist. This patch add the check before refreshing. https://bugzilla.kernel.org/show_bug.cgi?id=35642 Signed-off-by: Lan Tianyu Signed-off-by: Len Brown --- drivers/acpi/battery.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c index 2fe7cfd95680..4ba339d0ea19 100644 --- a/drivers/acpi/battery.c +++ b/drivers/acpi/battery.c @@ -949,8 +949,10 @@ static int battery_notify(struct notifier_block *nb, switch (mode) { case PM_POST_HIBERNATION: case PM_POST_SUSPEND: - sysfs_remove_battery(battery); - sysfs_add_battery(battery); + if (battery->bat.dev) { + sysfs_remove_battery(battery); + sysfs_add_battery(battery); + } break; } From 9c921c22a7f33397a6774d7fa076db9b6a0fd669 Mon Sep 17 00:00:00 2001 From: Lan Tianyu Date: Thu, 30 Jun 2011 11:34:12 +0800 Subject: [PATCH 028/439] ACPI / Battery: Resolve the race condition in the sysfs_remove_battery() Use battery->lock in sysfs_remove_battery() to make checking, removing, and clearing bat.dev atomic. This is necessary because sysfs_remove_battery() may be invoked concurrently from different paths. https://bugzilla.kernel.org/show_bug.cgi?id=35642 Signed-off-by: Lan Tianyu Signed-off-by: Len Brown --- drivers/acpi/battery.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c index 4ba339d0ea19..40bf01d42cc3 100644 --- a/drivers/acpi/battery.c +++ b/drivers/acpi/battery.c @@ -573,11 +573,16 @@ static int sysfs_add_battery(struct acpi_battery *battery) static void sysfs_remove_battery(struct acpi_battery *battery) { - if (!battery->bat.dev) + mutex_lock(&battery->lock); + if (!battery->bat.dev) { + mutex_unlock(&battery->lock); return; + } + device_remove_file(battery->bat.dev, &alarm_attr); power_supply_unregister(&battery->bat); battery->bat.dev = NULL; + mutex_unlock(&battery->lock); } /* From e545b55a1e980cbb6a158886286106bbf39722b1 Mon Sep 17 00:00:00 2001 From: Jon Mason Date: Sun, 19 Jun 2011 18:51:37 -0500 Subject: [PATCH 029/439] ACPI: fix 80 char overflow Trivial fix for 80 char line overflow in drivers/acpi/pci_root.c Signed-off-by: Jon Mason Signed-off-by: Len Brown --- drivers/acpi/pci_root.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c index d06078d660ad..2672c798272f 100644 --- a/drivers/acpi/pci_root.c +++ b/drivers/acpi/pci_root.c @@ -485,7 +485,8 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device) root->secondary.end = 0xFF; printk(KERN_WARNING FW_BUG PREFIX "no secondary bus range in _CRS\n"); - status = acpi_evaluate_integer(device->handle, METHOD_NAME__BBN, NULL, &bus); + status = acpi_evaluate_integer(device->handle, METHOD_NAME__BBN, + NULL, &bus); if (ACPI_SUCCESS(status)) root->secondary.start = bus; else if (status == AE_NOT_FOUND) From d8926bb3badd36670fecf2de4a062c78bc37430b Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Wed, 13 Jul 2011 10:38:47 -0700 Subject: [PATCH 030/439] btrfs: don't BUG_ON btrfs_alloc_path() errors This patch fixes many callers of btrfs_alloc_path() which BUG_ON allocation failure. All the sites that are fixed in this patch were checked by me to be fairly trivial to fix because of at least one of two criteria: - Callers of the function catch errors from it already so bubbling the error up will be handled. - Callers of the function might BUG_ON any nonzero return code in which case there is no behavior changed (but we still got to remove a BUG_ON) The following functions were updated: btrfs_lookup_extent, alloc_reserved_tree_block, btrfs_remove_block_group, btrfs_lookup_csums_range, btrfs_csum_file_blocks, btrfs_mark_extent_written, btrfs_inode_by_name, btrfs_new_inode, btrfs_symlink, insert_reserved_file_extent, and run_delalloc_nocow Signed-off-by: Mark Fasheh --- fs/btrfs/extent-tree.c | 12 +++++++++--- fs/btrfs/file-item.c | 7 +++++-- fs/btrfs/file.c | 3 ++- fs/btrfs/inode.c | 18 +++++++++++++----- 4 files changed, 29 insertions(+), 11 deletions(-) diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 71cd456fdb60..aa91773fe31b 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -667,7 +667,9 @@ int btrfs_lookup_extent(struct btrfs_root *root, u64 start, u64 len) struct btrfs_path *path; path = btrfs_alloc_path(); - BUG_ON(!path); + if (!path) + return -ENOMEM; + key.objectid = start; key.offset = len; btrfs_set_key_type(&key, BTRFS_EXTENT_ITEM_KEY); @@ -5494,7 +5496,8 @@ static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans, u32 size = sizeof(*extent_item) + sizeof(*block_info) + sizeof(*iref); path = btrfs_alloc_path(); - BUG_ON(!path); + if (!path) + return -ENOMEM; path->leave_spinning = 1; ret = btrfs_insert_empty_item(trans, fs_info->extent_root, path, @@ -7162,7 +7165,10 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, spin_unlock(&cluster->refill_lock); path = btrfs_alloc_path(); - BUG_ON(!path); + if (!path) { + ret = -ENOMEM; + goto out; + } inode = lookup_free_space_inode(root, block_group, path); if (!IS_ERR(inode)) { diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c index 90d4ee52cd45..f92ff0ed6e03 100644 --- a/fs/btrfs/file-item.c +++ b/fs/btrfs/file-item.c @@ -282,7 +282,8 @@ int btrfs_lookup_csums_range(struct btrfs_root *root, u64 start, u64 end, u16 csum_size = btrfs_super_csum_size(&root->fs_info->super_copy); path = btrfs_alloc_path(); - BUG_ON(!path); + if (!path) + return -ENOMEM; if (search_commit) { path->skip_locking = 1; @@ -672,7 +673,9 @@ int btrfs_csum_file_blocks(struct btrfs_trans_handle *trans, btrfs_super_csum_size(&root->fs_info->super_copy); path = btrfs_alloc_path(); - BUG_ON(!path); + if (!path) + return -ENOMEM; + sector_sum = sums->sums; again: next_offset = (u64)-1; diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index fa4ef18b66b1..23d1d811e2b2 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -855,7 +855,8 @@ int btrfs_mark_extent_written(struct btrfs_trans_handle *trans, btrfs_drop_extent_cache(inode, start, end - 1, 0); path = btrfs_alloc_path(); - BUG_ON(!path); + if (!path) + return -ENOMEM; again: recow = 0; split = start; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 3601f0aebddf..8be7d7a759ce 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1070,7 +1070,8 @@ static noinline int run_delalloc_nocow(struct inode *inode, u64 ino = btrfs_ino(inode); path = btrfs_alloc_path(); - BUG_ON(!path); + if (!path) + return -ENOMEM; nolock = is_free_space_inode(root, inode); @@ -1644,7 +1645,8 @@ static int insert_reserved_file_extent(struct btrfs_trans_handle *trans, int ret; path = btrfs_alloc_path(); - BUG_ON(!path); + if (!path) + return -ENOMEM; path->leave_spinning = 1; @@ -3713,7 +3715,8 @@ static int btrfs_inode_by_name(struct inode *dir, struct dentry *dentry, int ret = 0; path = btrfs_alloc_path(); - BUG_ON(!path); + if (!path) + return -ENOMEM; di = btrfs_lookup_dir_item(NULL, root, path, btrfs_ino(dir), name, namelen, 0); @@ -4438,7 +4441,8 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, int owner; path = btrfs_alloc_path(); - BUG_ON(!path); + if (!path) + return ERR_PTR(-ENOMEM); inode = new_inode(root->fs_info->sb); if (!inode) { @@ -7194,7 +7198,11 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, goto out_unlock; path = btrfs_alloc_path(); - BUG_ON(!path); + if (!path) { + err = -ENOMEM; + drop_inode = 1; + goto out_unlock; + } key.objectid = btrfs_ino(inode); key.offset = 0; btrfs_set_key_type(&key, BTRFS_EXTENT_DATA_KEY); From 1e5063d093b5f024ae35bf835ca07463de2c1a1f Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Tue, 12 Jul 2011 10:46:06 -0700 Subject: [PATCH 031/439] btrfs: Don't BUG_ON alloc_path errors in replay_one_buffer() The two ->process_func call sites in tree-log.c which were ignoring a return code have also been updated to gracefully exit as well. Signed-off-by: Mark Fasheh --- fs/btrfs/tree-log.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index 4ce8a9f41d1e..f3cacc079102 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -1617,7 +1617,8 @@ static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb, return 0; path = btrfs_alloc_path(); - BUG_ON(!path); + if (!path) + return -ENOMEM; nritems = btrfs_header_nritems(eb); for (i = 0; i < nritems; i++) { @@ -1723,7 +1724,9 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans, return -ENOMEM; if (*level == 1) { - wc->process_func(root, next, wc, ptr_gen); + ret = wc->process_func(root, next, wc, ptr_gen); + if (ret) + return ret; path->slots[*level]++; if (wc->free) { @@ -1788,8 +1791,11 @@ static noinline int walk_up_log_tree(struct btrfs_trans_handle *trans, parent = path->nodes[*level + 1]; root_owner = btrfs_header_owner(parent); - wc->process_func(root, path->nodes[*level], wc, + ret = wc->process_func(root, path->nodes[*level], wc, btrfs_header_generation(path->nodes[*level])); + if (ret) + return ret; + if (wc->free) { struct extent_buffer *next; From 0eb0e19cde6f01397ef8c0e094e44beb75c62a1e Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Tue, 12 Jul 2011 16:44:10 -0700 Subject: [PATCH 032/439] btrfs: Don't BUG_ON alloc_path errors in btrfs_truncate_inode_items I moved the path allocation up a few lines to the top of the function so that we couldn't get into the state where we've dropped delayed items and the extent cache but fail due to -ENOMEM. Signed-off-by: Mark Fasheh --- fs/btrfs/inode.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 8be7d7a759ce..a0faf7d7f0e0 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -3172,6 +3172,11 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans, BUG_ON(new_size > 0 && min_type != BTRFS_EXTENT_DATA_KEY); + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + path->reada = -1; + if (root->ref_cows || root == root->fs_info->tree_root) btrfs_drop_extent_cache(inode, new_size & (~mask), (u64)-1, 0); @@ -3184,10 +3189,6 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans, if (min_type == 0 && root == BTRFS_I(inode)->root) btrfs_kill_delayed_inode_items(inode); - path = btrfs_alloc_path(); - BUG_ON(!path); - path->reada = -1; - key.objectid = ino; key.offset = (u64)-1; key.type = (u8)-1; From 1748f843a0190ef4332d03a64263f383af72682b Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Tue, 12 Jul 2011 11:25:31 -0700 Subject: [PATCH 033/439] btrfs: Don't BUG_ON alloc_path errors in btrfs_read_locked_inode btrfs_iget() also needed an update so that errors from btrfs_locked_inode() are caught and bubbled back up. Signed-off-by: Mark Fasheh --- fs/btrfs/inode.c | 22 +++++++++++++++++----- 1 file changed, 17 insertions(+), 5 deletions(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index a0faf7d7f0e0..88829993db6c 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -2518,7 +2518,9 @@ static void btrfs_read_locked_inode(struct inode *inode) filled = true; path = btrfs_alloc_path(); - BUG_ON(!path); + if (!path) + goto make_bad; + path->leave_spinning = 1; memcpy(&location, &BTRFS_I(inode)->location, sizeof(location)); @@ -3973,6 +3975,7 @@ struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location, struct btrfs_root *root, int *new) { struct inode *inode; + int bad_inode = 0; inode = btrfs_iget_locked(s, location->objectid, root); if (!inode) @@ -3982,10 +3985,19 @@ struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location, BTRFS_I(inode)->root = root; memcpy(&BTRFS_I(inode)->location, location, sizeof(*location)); btrfs_read_locked_inode(inode); - inode_tree_add(inode); - unlock_new_inode(inode); - if (new) - *new = 1; + if (!is_bad_inode(inode)) { + inode_tree_add(inode); + unlock_new_inode(inode); + if (new) + *new = 1; + } else { + bad_inode = 1; + } + } + + if (bad_inode) { + iput(inode); + inode = ERR_PTR(-ESTALE); } return inode; From 17e9f796bd92cddec17d781c459376f82340fa44 Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Tue, 12 Jul 2011 11:10:23 -0700 Subject: [PATCH 034/439] btrfs: Don't BUG_ON alloc_path errors in btrfs_balance() Dealing with this seems trivial - the only caller of btrfs_balance() is btrfs_ioctl() which passes the error code directly back to userspace. There also isn't much state to unwind (if I'm wrong about this point, we can always safely move the allocation to the top of btrfs_balance() anyway). Signed-off-by: Mark Fasheh --- fs/btrfs/volumes.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 19450bc53632..530a2fcea1ef 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -2061,8 +2061,10 @@ int btrfs_balance(struct btrfs_root *dev_root) /* step two, relocate all the chunks */ path = btrfs_alloc_path(); - BUG_ON(!path); - + if (!path) { + ret = -ENOMEM; + goto error; + } key.objectid = BTRFS_FIRST_CHUNK_TREE_OBJECTID; key.offset = (u64)-1; key.type = BTRFS_CHUNK_ITEM_KEY; From e4c0d0e22ce5cf84b3993b5af6c2dac08d52f06b Mon Sep 17 00:00:00 2001 From: Len Brown Date: Fri, 15 Jul 2011 23:39:00 -0400 Subject: [PATCH 035/439] tools/power x86_energy_perf_policy: fix print of uninitialized string Looks like I was going to stick the brand string in the verbose ouput, but didn't get around to it. Signed-off-by: Len Brown --- .../x86/x86_energy_perf_policy/x86_energy_perf_policy.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/tools/power/x86/x86_energy_perf_policy/x86_energy_perf_policy.c b/tools/power/x86/x86_energy_perf_policy/x86_energy_perf_policy.c index 2618ef2ba31f..33c5c7ee148f 100644 --- a/tools/power/x86/x86_energy_perf_policy/x86_energy_perf_policy.c +++ b/tools/power/x86/x86_energy_perf_policy/x86_energy_perf_policy.c @@ -137,7 +137,6 @@ void cmdline(int argc, char **argv) void validate_cpuid(void) { unsigned int eax, ebx, ecx, edx, max_level; - char brand[16]; unsigned int fms, family, model, stepping; eax = ebx = ecx = edx = 0; @@ -160,8 +159,8 @@ void validate_cpuid(void) model += ((fms >> 16) & 0xf) << 4; if (verbose > 1) - printf("CPUID %s %d levels family:model:stepping " - "0x%x:%x:%x (%d:%d:%d)\n", brand, max_level, + printf("CPUID %d levels family:model:stepping " + "0x%x:%x:%x (%d:%d:%d)\n", max_level, family, model, stepping, family, model, stepping); if (!(edx & (1 << 5))) { From d7f6169a0d32002657886fee561c641acddb9a75 Mon Sep 17 00:00:00 2001 From: Stefan Assmann Date: Fri, 15 Jul 2011 14:52:30 +0200 Subject: [PATCH 036/439] ACPI: fix CONFIG_X86_REROUTE_FOR_BROKEN_BOOT_IRQS The following was observed by Steve Rostedt on 3.0.0-rc5 Backtrace: irq 16: nobody cared (try booting with the "irqpoll" option) Pid: 65, comm: irq/16-uhci_hcd Not tainted 3.0.0-rc5-test+ #94 Call Trace: [] __report_bad_irq+0x37/0xc1 [] note_interrupt+0x14e/0x1c9 [] ? irq_thread_fn+0x3c/0x3c [] irq_thread+0xf6/0x1b1 [] ? irq_finalize_oneshot+0xb3/0xb3 [] kthread+0x9f/0xa7 [] kernel_thread_helper+0x4/0x10 [] ? finish_task_switch+0x7b/0xc0 [] ? retint_restore_args+0x13/0x13 [] ? __init_kthread_worker+0x5a/0x5a [] ? gs_change+0x13/0x13 handlers: [] irq_default_primary_handler threaded [] usb_hcd_irq [] irq_default_primary_handler threaded [] usb_hcd_irq Disabling IRQ #16 The problem being that a device triggers boot interrupts (due to threaded interrupt handling and masking of the IO-APIC), which are forwarded to the PIRQ line of the device. These interrupts are not handled on the PIRQ line because the interrupt handler is not present there. This should have already been fixed by CONFIG_X86_REROUTE_FOR_BROKEN_BOOT_IRQS. However some parts of the quirk got lost in the ACPI merge. This is a resent of the patch proposed in 2009. See http://lkml.org/lkml/2009/9/7/192 Signed-off-by: Stefan Assmann Tested-by: Steven Rostedt Signed-off-by: Len Brown --- drivers/acpi/pci_irq.c | 58 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 58 insertions(+) diff --git a/drivers/acpi/pci_irq.c b/drivers/acpi/pci_irq.c index f907cfbfa13c..7f9eba9a0b02 100644 --- a/drivers/acpi/pci_irq.c +++ b/drivers/acpi/pci_irq.c @@ -303,6 +303,61 @@ void acpi_pci_irq_del_prt(struct pci_bus *bus) /* -------------------------------------------------------------------------- PCI Interrupt Routing Support -------------------------------------------------------------------------- */ +#ifdef CONFIG_X86_IO_APIC +extern int noioapicquirk; +extern int noioapicreroute; + +static int bridge_has_boot_interrupt_variant(struct pci_bus *bus) +{ + struct pci_bus *bus_it; + + for (bus_it = bus ; bus_it ; bus_it = bus_it->parent) { + if (!bus_it->self) + return 0; + if (bus_it->self->irq_reroute_variant) + return bus_it->self->irq_reroute_variant; + } + return 0; +} + +/* + * Some chipsets (e.g. Intel 6700PXH) generate a legacy INTx when the IRQ + * entry in the chipset's IO-APIC is masked (as, e.g. the RT kernel does + * during interrupt handling). When this INTx generation cannot be disabled, + * we reroute these interrupts to their legacy equivalent to get rid of + * spurious interrupts. + */ +static int acpi_reroute_boot_interrupt(struct pci_dev *dev, + struct acpi_prt_entry *entry) +{ + if (noioapicquirk || noioapicreroute) { + return 0; + } else { + switch (bridge_has_boot_interrupt_variant(dev->bus)) { + case 0: + /* no rerouting necessary */ + return 0; + case INTEL_IRQ_REROUTE_VARIANT: + /* + * Remap according to INTx routing table in 6700PXH + * specs, intel order number 302628-002, section + * 2.15.2. Other chipsets (80332, ...) have the same + * mapping and are handled here as well. + */ + dev_info(&dev->dev, "PCI IRQ %d -> rerouted to legacy " + "IRQ %d\n", entry->index, + (entry->index % 4) + 16); + entry->index = (entry->index % 4) + 16; + return 1; + default: + dev_warn(&dev->dev, "Cannot reroute IRQ %d to legacy " + "IRQ: unknown mapping\n", entry->index); + return -1; + } + } +} +#endif /* CONFIG_X86_IO_APIC */ + static struct acpi_prt_entry *acpi_pci_irq_lookup(struct pci_dev *dev, int pin) { struct acpi_prt_entry *entry; @@ -311,6 +366,9 @@ static struct acpi_prt_entry *acpi_pci_irq_lookup(struct pci_dev *dev, int pin) entry = acpi_pci_irq_find_prt_entry(dev, pin); if (entry) { +#ifdef CONFIG_X86_IO_APIC + acpi_reroute_boot_interrupt(dev, entry); +#endif /* CONFIG_X86_IO_APIC */ ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Found %s[%c] _PRT entry\n", pci_name(dev), pin_name(pin))); return entry; From 9c8b04be443b33939f374a811c82abeebe0a61d1 Mon Sep 17 00:00:00 2001 From: Vasiliy Kulikov Date: Sat, 25 Jun 2011 21:07:52 +0400 Subject: [PATCH 037/439] ACPI: constify ops structs Structs battery_file, acpi_dock_ops, file_operations, thermal_cooling_device_ops, thermal_zone_device_ops, kernel_param_ops are not changed in runtime. It is safe to make them const. register_hotplug_dock_device() was altered to take const "ops" argument to respect acpi_dock_ops' const notion. Signed-off-by: Vasiliy Kulikov Acked-by: Jeff Garzik Signed-off-by: Len Brown --- drivers/acpi/battery.c | 2 +- drivers/acpi/dock.c | 4 ++-- drivers/acpi/ec_sys.c | 2 +- drivers/acpi/fan.c | 2 +- drivers/acpi/processor_thermal.c | 2 +- drivers/acpi/sysfs.c | 4 ++-- drivers/acpi/thermal.c | 2 +- drivers/acpi/video.c | 2 +- drivers/ata/libata-acpi.c | 4 ++-- drivers/pci/hotplug/acpiphp_glue.c | 2 +- include/acpi/acpi_drivers.h | 2 +- include/acpi/processor.h | 2 +- 12 files changed, 15 insertions(+), 15 deletions(-) diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c index fcc13ac0aa18..746b47507e9f 100644 --- a/drivers/acpi/battery.c +++ b/drivers/acpi/battery.c @@ -863,7 +863,7 @@ DECLARE_FILE_FUNCTIONS(alarm); }, \ } -static struct battery_file { +static const struct battery_file { struct file_operations ops; mode_t mode; const char *name; diff --git a/drivers/acpi/dock.c b/drivers/acpi/dock.c index 1864ad3cf895..19a61136d848 100644 --- a/drivers/acpi/dock.c +++ b/drivers/acpi/dock.c @@ -77,7 +77,7 @@ struct dock_dependent_device { struct list_head list; struct list_head hotplug_list; acpi_handle handle; - struct acpi_dock_ops *ops; + const struct acpi_dock_ops *ops; void *context; }; @@ -589,7 +589,7 @@ EXPORT_SYMBOL_GPL(unregister_dock_notifier); * the dock driver after _DCK is executed. */ int -register_hotplug_dock_device(acpi_handle handle, struct acpi_dock_ops *ops, +register_hotplug_dock_device(acpi_handle handle, const struct acpi_dock_ops *ops, void *context) { struct dock_dependent_device *dd; diff --git a/drivers/acpi/ec_sys.c b/drivers/acpi/ec_sys.c index 05b44201a614..22f918bacd35 100644 --- a/drivers/acpi/ec_sys.c +++ b/drivers/acpi/ec_sys.c @@ -92,7 +92,7 @@ static ssize_t acpi_ec_write_io(struct file *f, const char __user *buf, return count; } -static struct file_operations acpi_ec_io_ops = { +static const struct file_operations acpi_ec_io_ops = { .owner = THIS_MODULE, .open = acpi_ec_open_io, .read = acpi_ec_read_io, diff --git a/drivers/acpi/fan.c b/drivers/acpi/fan.c index 467479f07c1f..0f0356ca1a9e 100644 --- a/drivers/acpi/fan.c +++ b/drivers/acpi/fan.c @@ -110,7 +110,7 @@ fan_set_cur_state(struct thermal_cooling_device *cdev, unsigned long state) return result; } -static struct thermal_cooling_device_ops fan_cooling_ops = { +static const struct thermal_cooling_device_ops fan_cooling_ops = { .get_max_state = fan_get_max_state, .get_cur_state = fan_get_cur_state, .set_cur_state = fan_set_cur_state, diff --git a/drivers/acpi/processor_thermal.c b/drivers/acpi/processor_thermal.c index 79cb65332894..870550d6a4bf 100644 --- a/drivers/acpi/processor_thermal.c +++ b/drivers/acpi/processor_thermal.c @@ -244,7 +244,7 @@ processor_set_cur_state(struct thermal_cooling_device *cdev, return result; } -struct thermal_cooling_device_ops processor_cooling_ops = { +const struct thermal_cooling_device_ops processor_cooling_ops = { .get_max_state = processor_get_max_state, .get_cur_state = processor_get_cur_state, .set_cur_state = processor_set_cur_state, diff --git a/drivers/acpi/sysfs.c b/drivers/acpi/sysfs.c index 77255f250dbb..c538d0ef10ff 100644 --- a/drivers/acpi/sysfs.c +++ b/drivers/acpi/sysfs.c @@ -149,12 +149,12 @@ static int param_get_debug_level(char *buffer, const struct kernel_param *kp) return result; } -static struct kernel_param_ops param_ops_debug_layer = { +static const struct kernel_param_ops param_ops_debug_layer = { .set = param_set_uint, .get = param_get_debug_layer, }; -static struct kernel_param_ops param_ops_debug_level = { +static const struct kernel_param_ops param_ops_debug_level = { .set = param_set_uint, .get = param_get_debug_level, }; diff --git a/drivers/acpi/thermal.c b/drivers/acpi/thermal.c index 2607e17b520f..48fbc647b178 100644 --- a/drivers/acpi/thermal.c +++ b/drivers/acpi/thermal.c @@ -812,7 +812,7 @@ acpi_thermal_unbind_cooling_device(struct thermal_zone_device *thermal, thermal_zone_unbind_cooling_device); } -static struct thermal_zone_device_ops acpi_thermal_zone_ops = { +static const struct thermal_zone_device_ops acpi_thermal_zone_ops = { .bind = acpi_thermal_bind_cooling_device, .unbind = acpi_thermal_unbind_cooling_device, .get_temp = thermal_get_temp, diff --git a/drivers/acpi/video.c b/drivers/acpi/video.c index db39e9e607d8..c6f9ef8d9ccb 100644 --- a/drivers/acpi/video.c +++ b/drivers/acpi/video.c @@ -308,7 +308,7 @@ video_set_cur_state(struct thermal_cooling_device *cooling_dev, unsigned long st return acpi_video_device_lcd_set_level(video, level); } -static struct thermal_cooling_device_ops video_cooling_ops = { +static const struct thermal_cooling_device_ops video_cooling_ops = { .get_max_state = video_get_max_state, .get_cur_state = video_get_cur_state, .set_cur_state = video_set_cur_state, diff --git a/drivers/ata/libata-acpi.c b/drivers/ata/libata-acpi.c index a791b8ce6294..993d40620f91 100644 --- a/drivers/ata/libata-acpi.c +++ b/drivers/ata/libata-acpi.c @@ -218,12 +218,12 @@ static void ata_acpi_dev_uevent(acpi_handle handle, u32 event, void *data) ata_acpi_uevent(dev->link->ap, dev, event); } -static struct acpi_dock_ops ata_acpi_dev_dock_ops = { +static const struct acpi_dock_ops ata_acpi_dev_dock_ops = { .handler = ata_acpi_dev_notify_dock, .uevent = ata_acpi_dev_uevent, }; -static struct acpi_dock_ops ata_acpi_ap_dock_ops = { +static const struct acpi_dock_ops ata_acpi_ap_dock_ops = { .handler = ata_acpi_ap_notify_dock, .uevent = ata_acpi_ap_uevent, }; diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c index a70fa89f76fd..220285760b68 100644 --- a/drivers/pci/hotplug/acpiphp_glue.c +++ b/drivers/pci/hotplug/acpiphp_glue.c @@ -110,7 +110,7 @@ static int post_dock_fixups(struct notifier_block *nb, unsigned long val, } -static struct acpi_dock_ops acpiphp_dock_ops = { +static const struct acpi_dock_ops acpiphp_dock_ops = { .handler = handle_hotplug_event_func, }; diff --git a/include/acpi/acpi_drivers.h b/include/acpi/acpi_drivers.h index 3090471b2a5e..e49c36d38d7e 100644 --- a/include/acpi/acpi_drivers.h +++ b/include/acpi/acpi_drivers.h @@ -128,7 +128,7 @@ extern int is_dock_device(acpi_handle handle); extern int register_dock_notifier(struct notifier_block *nb); extern void unregister_dock_notifier(struct notifier_block *nb); extern int register_hotplug_dock_device(acpi_handle handle, - struct acpi_dock_ops *ops, + const struct acpi_dock_ops *ops, void *context); extern void unregister_hotplug_dock_device(acpi_handle handle); #else diff --git a/include/acpi/processor.h b/include/acpi/processor.h index ba4928cae473..67055f180330 100644 --- a/include/acpi/processor.h +++ b/include/acpi/processor.h @@ -337,7 +337,7 @@ extern struct cpuidle_driver acpi_idle_driver; /* in processor_thermal.c */ int acpi_processor_get_limit_info(struct acpi_processor *pr); -extern struct thermal_cooling_device_ops processor_cooling_ops; +extern const struct thermal_cooling_device_ops processor_cooling_ops; #ifdef CONFIG_CPU_FREQ void acpi_thermal_cpufreq_init(void); void acpi_thermal_cpufreq_exit(void); From 4996c02306a25def1d352ec8e8f48895bbc7dea9 Mon Sep 17 00:00:00 2001 From: Takao Indoh Date: Thu, 14 Jul 2011 18:05:21 -0400 Subject: [PATCH 038/439] ACPI: introduce "acpi_rsdp=" parameter for kdump There is a problem with putting the first kernel in EFI virtual mode, it is that when the second kernel comes up it tries to initialize the EFI again and once we have put EFI in virtual mode we can not really do that. Actually, EFI is not necessary for kdump, we can boot the second kernel with "noefi" parameter, but the boot will mostly fail because 2nd kernel cannot find RSDP. In this situation, we introduced "acpi_rsdp=" kernel parameter, so that kexec-tools can pass the "noefi acpi_rsdp=X" to the second kernel to make kdump works. The physical address of the RSDP can be got from sysfs(/sys/firmware/efi/systab). Signed-off-by: Takao Indoh Reviewed-by: WANG Cong Reviewed-by: Konrad Rzeszutek Wilk Signed-off-by: Len Brown --- Documentation/kernel-parameters.txt | 5 +++++ drivers/acpi/osl.c | 15 +++++++++++++++ 2 files changed, 20 insertions(+) diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index aa47be71df4c..37713bcf7778 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -163,6 +163,11 @@ bytes respectively. Such letter suffixes can also be entirely omitted. See also Documentation/power/pm.txt, pci=noacpi + acpi_rsdp= [ACPI,EFI,KEXEC] + Pass the RSDP address to the kernel, mostly used + on machines running EFI runtime service to boot the + second kernel for kdump. + acpi_apic_instance= [ACPI, IOAPIC] Format: 2: use 2nd APIC table, if available diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c index 52ca9649d769..27cd1408f6d8 100644 --- a/drivers/acpi/osl.c +++ b/drivers/acpi/osl.c @@ -237,8 +237,23 @@ void acpi_os_vprintf(const char *fmt, va_list args) #endif } +#ifdef CONFIG_KEXEC +static unsigned long acpi_rsdp; +static int __init setup_acpi_rsdp(char *arg) +{ + acpi_rsdp = simple_strtoul(arg, NULL, 16); + return 0; +} +early_param("acpi_rsdp", setup_acpi_rsdp); +#endif + acpi_physical_address __init acpi_os_get_root_pointer(void) { +#ifdef CONFIG_KEXEC + if (acpi_rsdp) + return acpi_rsdp; +#endif + if (efi_enabled) { if (efi.acpi20 != EFI_INVALID_TABLE_ADDR) return efi.acpi20; From e80bba4b5108c6479379740201b0a5d9da5ffbac Mon Sep 17 00:00:00 2001 From: Stefan Hajnoczi Date: Tue, 12 Jul 2011 09:03:28 +0100 Subject: [PATCH 039/439] ACPI / Battery: avoid acpi_battery_add() use-after-free When acpi_battery_add_fs() fails the error handling code does not clean up completely. Moreover, it does not return resulting in a use-after-free. Signed-off-by: Stefan Hajnoczi Signed-off-by: Len Brown --- drivers/acpi/battery.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c index 40bf01d42cc3..c771768f57c8 100644 --- a/drivers/acpi/battery.c +++ b/drivers/acpi/battery.c @@ -986,21 +986,27 @@ static int acpi_battery_add(struct acpi_device *device) #ifdef CONFIG_ACPI_PROCFS_POWER result = acpi_battery_add_fs(device); #endif - if (!result) { - printk(KERN_INFO PREFIX "%s Slot [%s] (battery %s)\n", - ACPI_BATTERY_DEVICE_NAME, acpi_device_bid(device), - device->status.battery_present ? "present" : "absent"); - } else { + if (result) { #ifdef CONFIG_ACPI_PROCFS_POWER acpi_battery_remove_fs(device); #endif - kfree(battery); + goto fail; } + printk(KERN_INFO PREFIX "%s Slot [%s] (battery %s)\n", + ACPI_BATTERY_DEVICE_NAME, acpi_device_bid(device), + device->status.battery_present ? "present" : "absent"); + battery->pm_nb.notifier_call = battery_notify; register_pm_notifier(&battery->pm_nb); return result; + +fail: + sysfs_remove_battery(battery); + mutex_destroy(&battery->lock); + kfree(battery); + return result; } static int acpi_battery_remove(struct acpi_device *device, int type) From eb03cb02b74df6dd0b653d5f6d976f16a434dfaf Mon Sep 17 00:00:00 2001 From: Stefan Hajnoczi Date: Tue, 12 Jul 2011 09:03:29 +0100 Subject: [PATCH 040/439] ACPI / Battery: propagate sysfs error in acpi_battery_add() Make sure the error return from sysfs_add_battery() is checked and propagated out from acpi_battery_add(). Signed-off-by: Stefan Hajnoczi Signed-off-by: Len Brown --- drivers/acpi/battery.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c index c771768f57c8..ffce2f06dd8d 100644 --- a/drivers/acpi/battery.c +++ b/drivers/acpi/battery.c @@ -630,8 +630,11 @@ static int acpi_battery_update(struct acpi_battery *battery) return result; acpi_battery_init_alarm(battery); } - if (!battery->bat.dev) - sysfs_add_battery(battery); + if (!battery->bat.dev) { + result = sysfs_add_battery(battery); + if (result) + return result; + } result = acpi_battery_get_state(battery); acpi_battery_quirks(battery); return result; @@ -982,7 +985,9 @@ static int acpi_battery_add(struct acpi_device *device) if (ACPI_SUCCESS(acpi_get_handle(battery->device->handle, "_BIX", &handle))) set_bit(ACPI_BATTERY_XINFO_PRESENT, &battery->flags); - acpi_battery_update(battery); + result = acpi_battery_update(battery); + if (result) + goto fail; #ifdef CONFIG_ACPI_PROCFS_POWER result = acpi_battery_add_fs(device); #endif From d9ba5fe76d604514444b1ea0a19f38c6196a46e3 Mon Sep 17 00:00:00 2001 From: Carlos O'Donell Date: Fri, 8 Jul 2011 17:27:00 -0400 Subject: [PATCH 041/439] [PARISC] Fix futex support Implements futex op support and makes futex cmpxchg atomic. Tested on 64-bit SMP kernel running on 2 x PA8700s. [jejb: checkpatch fixes] Signed-off-by: Carlos O'Donell Tested-by: John David Anglin Cc: stable@kernel.org Signed-off-by: James Bottomley --- arch/parisc/include/asm/futex.h | 66 ++++++++++++++++++++++++++++++--- 1 file changed, 60 insertions(+), 6 deletions(-) diff --git a/arch/parisc/include/asm/futex.h b/arch/parisc/include/asm/futex.h index 67a33cc27ef2..2388bdb32832 100644 --- a/arch/parisc/include/asm/futex.h +++ b/arch/parisc/include/asm/futex.h @@ -5,11 +5,14 @@ #include #include +#include #include static inline int futex_atomic_op_inuser (int encoded_op, u32 __user *uaddr) { + unsigned long int flags; + u32 val; int op = (encoded_op >> 28) & 7; int cmp = (encoded_op >> 24) & 15; int oparg = (encoded_op << 8) >> 20; @@ -18,21 +21,58 @@ futex_atomic_op_inuser (int encoded_op, u32 __user *uaddr) if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28)) oparg = 1 << oparg; - if (! access_ok (VERIFY_WRITE, uaddr, sizeof(u32))) + if (!access_ok(VERIFY_WRITE, uaddr, sizeof(*uaddr))) return -EFAULT; pagefault_disable(); + _atomic_spin_lock_irqsave(uaddr, flags); + switch (op) { case FUTEX_OP_SET: + /* *(int *)UADDR2 = OPARG; */ + ret = get_user(oldval, uaddr); + if (!ret) + ret = put_user(oparg, uaddr); + break; case FUTEX_OP_ADD: + /* *(int *)UADDR2 += OPARG; */ + ret = get_user(oldval, uaddr); + if (!ret) { + val = oldval + oparg; + ret = put_user(val, uaddr); + } + break; case FUTEX_OP_OR: + /* *(int *)UADDR2 |= OPARG; */ + ret = get_user(oldval, uaddr); + if (!ret) { + val = oldval | oparg; + ret = put_user(val, uaddr); + } + break; case FUTEX_OP_ANDN: + /* *(int *)UADDR2 &= ~OPARG; */ + ret = get_user(oldval, uaddr); + if (!ret) { + val = oldval & ~oparg; + ret = put_user(val, uaddr); + } + break; case FUTEX_OP_XOR: + /* *(int *)UADDR2 ^= OPARG; */ + ret = get_user(oldval, uaddr); + if (!ret) { + val = oldval ^ oparg; + ret = put_user(val, uaddr); + } + break; default: ret = -ENOSYS; } + _atomic_spin_unlock_irqrestore(uaddr, flags); + pagefault_enable(); if (!ret) { @@ -54,7 +94,9 @@ static inline int futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr, u32 oldval, u32 newval) { + int ret; u32 val; + unsigned long flags; /* futex.c wants to do a cmpxchg_inatomic on kernel NULL, which is * our gateway page, and causes no end of trouble... @@ -65,12 +107,24 @@ futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr, if (!access_ok(VERIFY_WRITE, uaddr, sizeof(u32))) return -EFAULT; - if (get_user(val, uaddr)) - return -EFAULT; - if (val == oldval && put_user(newval, uaddr)) - return -EFAULT; + /* HPPA has no cmpxchg in hardware and therefore the + * best we can do here is use an array of locks. The + * lock selected is based on a hash of the userspace + * address. This should scale to a couple of CPUs. + */ + + _atomic_spin_lock_irqsave(uaddr, flags); + + ret = get_user(val, uaddr); + + if (!ret && val == oldval) + ret = put_user(newval, uaddr); + *uval = val; - return 0; + + _atomic_spin_unlock_irqrestore(uaddr, flags); + + return ret; } #endif /*__KERNEL__*/ From 7676e345824f162191b1fe2058ad948a6cf91c20 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Thu, 19 May 2011 21:32:57 +0900 Subject: [PATCH 042/439] IPVS: Free resources on module removal This resolves a panic on module removal. Reported-by: Dave Jones Acked-by: Julian Anastasov Signed-off-by: Hans Schillstrom Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_ctl.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 699c79a55657..a178cb345848 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -3771,6 +3771,7 @@ err_sock: void ip_vs_control_cleanup(void) { EnterFunction(2); + unregister_netdevice_notifier(&ip_vs_dst_notifier); ip_vs_genl_unregister(); nf_unregister_sockopt(&ip_vs_sockopts); LeaveFunction(2); From a5071c2fd91f320bc40952df59517053a7346fa9 Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Tue, 19 Jul 2011 15:38:56 -0700 Subject: [PATCH 043/439] drm/i915: provide more error output when mode sets fail If a mode set fails we may get a message from drm_crtc_helper if we're lucky, but it won't tell us anything about *why* we failed to set a mode. So add a few DRM_ERRORs for the cases that shouldn't happen so we can debug things more easily. Signed-off-by: Jesse Barnes Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_display.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index b5b15bda71d9..261ffe47a5d2 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -2096,7 +2096,7 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, /* no fb bound */ if (!crtc->fb) { - DRM_DEBUG_KMS("No FB bound\n"); + DRM_ERROR("No FB bound\n"); return 0; } @@ -2105,6 +2105,7 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, case 1: break; default: + DRM_ERROR("no plane for crtc\n"); return -EINVAL; } @@ -2114,6 +2115,7 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, NULL); if (ret != 0) { mutex_unlock(&dev->struct_mutex); + DRM_ERROR("pin & fence failed\n"); return ret; } @@ -2142,6 +2144,7 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, if (ret) { i915_gem_object_unpin(to_intel_framebuffer(crtc->fb)->obj); mutex_unlock(&dev->struct_mutex); + DRM_ERROR("failed to update base address\n"); return ret; } From f0b69efc29b024747a88ce020dada425e3193d5a Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Tue, 19 Jul 2011 16:21:40 -0700 Subject: [PATCH 044/439] drm/i915: Skip GPU wait for scanout pin while wedged Failing to pin a scanout buffer will most likely lead to a black screen, so if the GPU is wedged, then just let the pin happen and hope that things work out OK. v2: Just ignore any error from i915_gem_object_wait_rendering, as suggested by Chris Wilson Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/i915_gem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index e9d1d5c3a696..e46f2734acc5 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -3113,7 +3113,7 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj, if (pipelined != obj->ring) { ret = i915_gem_object_wait_rendering(obj); - if (ret) + if (ret == -ERESTARTSYS) return ret; } From 842d452985300f4ec14c68cb86046e8a1a3b7251 Mon Sep 17 00:00:00 2001 From: Ole Henrik Jahren Date: Fri, 22 Jul 2011 15:56:01 +0200 Subject: [PATCH 045/439] drm/i915: Fix typo in DRM_I915_OVERLAY_PUT_IMAGE ioctl define Because of a typo, calling ioctl with DRM_IOCTL_I915_OVERLAY_PUT_IMAGE is broken if the macro is used directly. When using libdrm the bug is not hit, since libdrm handles the ioctl encoding internally. The typo also leads to the .cmd and .cmd_drv fields of the drm_ioctl structure for DRM_I915_OVERLAY_PUT_IMAGE having inconsistent content. Signed-off-by: Ole Henrik Jahren Acked-by: Daniel Vetter Cc: stable@kernel.org Signed-off-by: Keith Packard --- include/drm/i915_drm.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/drm/i915_drm.h b/include/drm/i915_drm.h index c4d6dbfa3ff4..28c0d114cb52 100644 --- a/include/drm/i915_drm.h +++ b/include/drm/i915_drm.h @@ -237,7 +237,7 @@ typedef struct _drm_i915_sarea { #define DRM_IOCTL_I915_GEM_GET_APERTURE DRM_IOR (DRM_COMMAND_BASE + DRM_I915_GEM_GET_APERTURE, struct drm_i915_gem_get_aperture) #define DRM_IOCTL_I915_GET_PIPE_FROM_CRTC_ID DRM_IOWR(DRM_COMMAND_BASE + DRM_I915_GET_PIPE_FROM_CRTC_ID, struct drm_i915_get_pipe_from_crtc_id) #define DRM_IOCTL_I915_GEM_MADVISE DRM_IOWR(DRM_COMMAND_BASE + DRM_I915_GEM_MADVISE, struct drm_i915_gem_madvise) -#define DRM_IOCTL_I915_OVERLAY_PUT_IMAGE DRM_IOW(DRM_COMMAND_BASE + DRM_IOCTL_I915_OVERLAY_ATTRS, struct drm_intel_overlay_put_image) +#define DRM_IOCTL_I915_OVERLAY_PUT_IMAGE DRM_IOW(DRM_COMMAND_BASE + DRM_I915_OVERLAY_PUT_IMAGE, struct drm_intel_overlay_put_image) #define DRM_IOCTL_I915_OVERLAY_ATTRS DRM_IOWR(DRM_COMMAND_BASE + DRM_I915_OVERLAY_ATTRS, struct drm_intel_overlay_attrs) /* Allow drivers to submit batchbuffers directly to hardware, relying From f3234706a77bd6e1592ae71fb3268e04cb030dba Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Fri, 22 Jul 2011 10:44:39 -0700 Subject: [PATCH 046/439] drm/i915: Initialize RCS ring status page address in intel_render_ring_init_dri Physically-addressed hardware status pages are initialized early in the driver load process by i915_init_phys_hws. For UMS environments, the ring structure is not initialized until the X server starts. At that point, the entire ring structure is re-initialized with all new values. Any values set in the ring structure (including ring->status_page.page_addr) will be lost when the ring is re-initialized. This patch moves the initialization of the status_page.page_addr value to intel_render_ring_init_dri. Signed-off-by: Keith Packard Cc: stable@kernel.org --- drivers/gpu/drm/i915/i915_dma.c | 6 ++---- drivers/gpu/drm/i915/intel_ringbuffer.c | 3 +++ 2 files changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c index 296fbd66f0e1..7eef6e11d9ac 100644 --- a/drivers/gpu/drm/i915/i915_dma.c +++ b/drivers/gpu/drm/i915/i915_dma.c @@ -61,7 +61,6 @@ static void i915_write_hws_pga(struct drm_device *dev) static int i915_init_phys_hws(struct drm_device *dev) { drm_i915_private_t *dev_priv = dev->dev_private; - struct intel_ring_buffer *ring = LP_RING(dev_priv); /* Program Hardware Status Page */ dev_priv->status_page_dmah = @@ -71,10 +70,9 @@ static int i915_init_phys_hws(struct drm_device *dev) DRM_ERROR("Can not allocate hardware status page\n"); return -ENOMEM; } - ring->status_page.page_addr = - (void __force __iomem *)dev_priv->status_page_dmah->vaddr; - memset_io(ring->status_page.page_addr, 0, PAGE_SIZE); + memset_io((void __force __iomem *)dev_priv->status_page_dmah->vaddr, + 0, PAGE_SIZE); i915_write_hws_pga(dev); diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 95c4b1429935..1f61fc7b754c 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -1319,6 +1319,9 @@ int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size) ring->get_seqno = pc_render_get_seqno; } + if (!I915_NEED_GFX_HWS(dev)) + ring->status_page.page_addr = dev_priv->status_page_dmah->vaddr; + ring->dev = dev; INIT_LIST_HEAD(&ring->active_list); INIT_LIST_HEAD(&ring->request_list); From 9c54c0dd948d715ccfd79e97d852f80eeb53254a Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Wed, 15 Jun 2011 23:32:33 +0200 Subject: [PATCH 047/439] drm/i915: load the LUT before pipe enable on ILK+ Per the specs and to address https://bugs.freedesktop.org/show_bug.cgi?id=36888. Signed-off-by: Jesse Barnes Cc: stable@kernel.org Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_display.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 0f1c799afea1..5609c065aaf4 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -2699,14 +2699,18 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc) I915_WRITE(PF_WIN_SZ(pipe), dev_priv->pch_pf_size); } + /* + * On ILK+ LUT must be loaded before the pipe is running but with + * clocks enabled + */ + intel_crtc_load_lut(crtc); + intel_enable_pipe(dev_priv, pipe, is_pch_port); intel_enable_plane(dev_priv, plane, pipe); if (is_pch_port) ironlake_pch_enable(crtc); - intel_crtc_load_lut(crtc); - mutex_lock(&dev->struct_mutex); intel_update_fbc(dev); mutex_unlock(&dev->struct_mutex); From 3baef940f289f08e4aada1fd481ab9ee3f070144 Mon Sep 17 00:00:00 2001 From: Dave Jiang Date: Fri, 22 Jul 2011 14:04:56 -0700 Subject: [PATCH 048/439] ioat: Adding PCI IDs for IOAT devices on SandyBridge platforms Adding to pci_id.h and the device table for ioat. Signed-off-by: Dave Jiang Signed-off-by: Dan Williams --- drivers/dma/ioat/pci.c | 11 +++++++++++ include/linux/pci_ids.h | 10 ++++++++++ 2 files changed, 21 insertions(+) diff --git a/drivers/dma/ioat/pci.c b/drivers/dma/ioat/pci.c index fab37d1cf48d..5e3a40f79945 100644 --- a/drivers/dma/ioat/pci.c +++ b/drivers/dma/ioat/pci.c @@ -72,6 +72,17 @@ static struct pci_device_id ioat_pci_tbl[] = { { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_JSF8) }, { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_JSF9) }, + { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_SNB0) }, + { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_SNB1) }, + { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_SNB2) }, + { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_SNB3) }, + { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_SNB4) }, + { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_SNB5) }, + { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_SNB6) }, + { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_SNB7) }, + { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_SNB8) }, + { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_IOAT_SNB9) }, + { 0, } }; MODULE_DEVICE_TABLE(pci, ioat_pci_tbl); diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index f8910e155566..a8571b36ae97 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2699,6 +2699,16 @@ #define PCI_DEVICE_ID_INTEL_ICH10_5 0x3a60 #define PCI_DEVICE_ID_INTEL_5_3400_SERIES_LPC_MIN 0x3b00 #define PCI_DEVICE_ID_INTEL_5_3400_SERIES_LPC_MAX 0x3b1f +#define PCI_DEVICE_ID_INTEL_IOAT_SNB0 0x3c20 +#define PCI_DEVICE_ID_INTEL_IOAT_SNB1 0x3c21 +#define PCI_DEVICE_ID_INTEL_IOAT_SNB2 0x3c22 +#define PCI_DEVICE_ID_INTEL_IOAT_SNB3 0x3c23 +#define PCI_DEVICE_ID_INTEL_IOAT_SNB4 0x3c24 +#define PCI_DEVICE_ID_INTEL_IOAT_SNB5 0x3c25 +#define PCI_DEVICE_ID_INTEL_IOAT_SNB6 0x3c26 +#define PCI_DEVICE_ID_INTEL_IOAT_SNB7 0x3c27 +#define PCI_DEVICE_ID_INTEL_IOAT_SNB8 0x3c2e +#define PCI_DEVICE_ID_INTEL_IOAT_SNB9 0x3c2f #define PCI_DEVICE_ID_INTEL_IOAT_SNB 0x402f #define PCI_DEVICE_ID_INTEL_5100_16 0x65f0 #define PCI_DEVICE_ID_INTEL_5100_21 0x65f5 From 9b487ced8a903f1028fcfaef1e6406acc91fe0fa Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 7 Jun 2011 15:26:33 -0700 Subject: [PATCH 049/439] Avoid section type conflict in dma/ioat/dma_v3.c const __read_mostly is not legal and causes section type conflicts. That's because the read.mostly section is not read only. Simply drop the __read_mostly designation. Signed-off-by: Andi Kleen [drop __read_mostly instead of const] Signed-off-by: Dan Williams --- drivers/dma/ioat/dma_v3.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/dma/ioat/dma_v3.c b/drivers/dma/ioat/dma_v3.c index d845dc4b7103..fec8664f2f23 100644 --- a/drivers/dma/ioat/dma_v3.c +++ b/drivers/dma/ioat/dma_v3.c @@ -73,10 +73,10 @@ /* provide a lookup table for setting the source address in the base or * extended descriptor of an xor or pq descriptor */ -static const u8 xor_idx_to_desc __read_mostly = 0xd0; -static const u8 xor_idx_to_field[] __read_mostly = { 1, 4, 5, 6, 7, 0, 1, 2 }; -static const u8 pq_idx_to_desc __read_mostly = 0xf8; -static const u8 pq_idx_to_field[] __read_mostly = { 1, 4, 5, 0, 1, 2, 4, 5 }; +static const u8 xor_idx_to_desc = 0xd0; +static const u8 xor_idx_to_field[] = { 1, 4, 5, 6, 7, 0, 1, 2 }; +static const u8 pq_idx_to_desc = 0xf8; +static const u8 pq_idx_to_field[] = { 1, 4, 5, 0, 1, 2, 4, 5 }; static dma_addr_t xor_get_src(struct ioat_raw_descriptor *descs[2], int idx) { From 97cdd7101079adc3c626d159c62d43de949516c8 Mon Sep 17 00:00:00 2001 From: Adam Jackson Date: Tue, 12 Jul 2011 17:38:00 -0400 Subject: [PATCH 050/439] drm/i915/dp: Zero the DPCD data before connection probe Signed-off-by: Adam Jackson Reviewed-by: Keith Packard Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_dp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index e2aced6eec4c..de24f314eda9 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -1673,6 +1673,7 @@ intel_dp_detect(struct drm_connector *connector, bool force) struct edid *edid = NULL; intel_dp->has_audio = false; + memset(intel_dp->dpcd, 0, sizeof(intel_dp->dpcd)); if (HAS_PCH_SPLIT(dev)) status = ironlake_dp_detect(intel_dp); From 1b9be9d09d85b3697418dc444db30d069203ff7d Mon Sep 17 00:00:00 2001 From: Adam Jackson Date: Tue, 12 Jul 2011 17:38:01 -0400 Subject: [PATCH 051/439] drm/i915/dp: Move DPCD dump to common code instead of PCH-only No reason not to see this on g4x, after all. Signed-off-by: Adam Jackson Reviewed-by: Keith Packard Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_dp.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index de24f314eda9..0be85a00bb52 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -1615,8 +1615,6 @@ ironlake_dp_detect(struct intel_dp *intel_dp) sizeof (intel_dp->dpcd)); if (ret && intel_dp->dpcd[DP_DPCD_REV] != 0) status = connector_status_connected; - DRM_DEBUG_KMS("DPCD: %hx%hx%hx%hx\n", intel_dp->dpcd[0], - intel_dp->dpcd[1], intel_dp->dpcd[2], intel_dp->dpcd[3]); return status; } @@ -1679,6 +1677,10 @@ intel_dp_detect(struct drm_connector *connector, bool force) status = ironlake_dp_detect(intel_dp); else status = g4x_dp_detect(intel_dp); + + DRM_DEBUG_KMS("DPCD: %hx%hx%hx%hx\n", intel_dp->dpcd[0], + intel_dp->dpcd[1], intel_dp->dpcd[2], intel_dp->dpcd[3]); + if (status != connector_status_connected) return status; From 9de88e6e89a2222061af8e1448f6f346e3413fc8 Mon Sep 17 00:00:00 2001 From: Adam Jackson Date: Tue, 12 Jul 2011 17:38:02 -0400 Subject: [PATCH 052/439] drm/i915/dp: Read more DPCD registers on connection probe For parity with radeon and nouveau, and also because I suspect we're going to need it to get format-conversion dongles right. Signed-off-by: Adam Jackson Reviewed-by: Keith Packard Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_dp.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 0be85a00bb52..2f0566b12f71 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -52,7 +52,7 @@ struct intel_dp { uint32_t color_range; uint8_t link_bw; uint8_t lane_count; - uint8_t dpcd[4]; + uint8_t dpcd[8]; struct i2c_adapter adapter; struct i2c_algo_dp_aux_data algo; bool is_pch_edp; @@ -1678,8 +1678,10 @@ intel_dp_detect(struct drm_connector *connector, bool force) else status = g4x_dp_detect(intel_dp); - DRM_DEBUG_KMS("DPCD: %hx%hx%hx%hx\n", intel_dp->dpcd[0], - intel_dp->dpcd[1], intel_dp->dpcd[2], intel_dp->dpcd[3]); + DRM_DEBUG_KMS("DPCD: %hx%hx%hx%hx%hx%hx%hx%hx\n", intel_dp->dpcd[0], + intel_dp->dpcd[1], intel_dp->dpcd[2], intel_dp->dpcd[3], + intel_dp->dpcd[4], intel_dp->dpcd[5], intel_dp->dpcd[6], + intel_dp->dpcd[7]); if (status != connector_status_connected) return status; From ac66ae8346fff704301e24ac55da1d76020660b2 Mon Sep 17 00:00:00 2001 From: Adam Jackson Date: Tue, 12 Jul 2011 17:38:03 -0400 Subject: [PATCH 053/439] drm/i915/dp: Better hexdump of DPCD %hx alone prints 0 as "0", not "00". Signed-off-by: Adam Jackson Reviewed-by: Keith Packard Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_dp.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 2f0566b12f71..13bddd34c057 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -1678,10 +1678,10 @@ intel_dp_detect(struct drm_connector *connector, bool force) else status = g4x_dp_detect(intel_dp); - DRM_DEBUG_KMS("DPCD: %hx%hx%hx%hx%hx%hx%hx%hx\n", intel_dp->dpcd[0], - intel_dp->dpcd[1], intel_dp->dpcd[2], intel_dp->dpcd[3], - intel_dp->dpcd[4], intel_dp->dpcd[5], intel_dp->dpcd[6], - intel_dp->dpcd[7]); + DRM_DEBUG_KMS("DPCD: %02hx%02hx%02hx%02hx%02hx%02hx%02hx%02hx\n", + intel_dp->dpcd[0], intel_dp->dpcd[1], intel_dp->dpcd[2], + intel_dp->dpcd[3], intel_dp->dpcd[4], intel_dp->dpcd[5], + intel_dp->dpcd[6], intel_dp->dpcd[7]); if (status != connector_status_connected) return status; From 71ba9000e673d6171a52f2a8b14e0419087f7199 Mon Sep 17 00:00:00 2001 From: Adam Jackson Date: Tue, 12 Jul 2011 17:38:04 -0400 Subject: [PATCH 054/439] drm/i915/dp: Retry DPCD fetch on G4X too Signed-off-by: Adam Jackson Reviewed-by: Keith Packard Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_dp.c | 33 +++++++++++++++------------------ 1 file changed, 15 insertions(+), 18 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 13bddd34c057..9a0c3ca1ffa9 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -1595,11 +1595,23 @@ intel_dp_check_link_status(struct intel_dp *intel_dp) } } +static enum drm_connector_status +i915_dp_detect_common(struct intel_dp *intel_dp) +{ + enum drm_connector_status status = connector_status_disconnected; + + if (intel_dp_aux_native_read_retry(intel_dp, 0x000, intel_dp->dpcd, + sizeof (intel_dp->dpcd)) && + (intel_dp->dpcd[DP_DPCD_REV] != 0)) + status = connector_status_connected; + + return status; +} + static enum drm_connector_status ironlake_dp_detect(struct intel_dp *intel_dp) { enum drm_connector_status status; - bool ret; /* Can't disconnect eDP, but you can close the lid... */ if (is_edp(intel_dp)) { @@ -1609,13 +1621,7 @@ ironlake_dp_detect(struct intel_dp *intel_dp) return status; } - status = connector_status_disconnected; - ret = intel_dp_aux_native_read_retry(intel_dp, - 0x000, intel_dp->dpcd, - sizeof (intel_dp->dpcd)); - if (ret && intel_dp->dpcd[DP_DPCD_REV] != 0) - status = connector_status_connected; - return status; + return i915_dp_detect_common(intel_dp); } static enum drm_connector_status @@ -1623,7 +1629,6 @@ g4x_dp_detect(struct intel_dp *intel_dp) { struct drm_device *dev = intel_dp->base.base.dev; struct drm_i915_private *dev_priv = dev->dev_private; - enum drm_connector_status status; uint32_t temp, bit; switch (intel_dp->output_reg) { @@ -1645,15 +1650,7 @@ g4x_dp_detect(struct intel_dp *intel_dp) if ((temp & bit) == 0) return connector_status_disconnected; - status = connector_status_disconnected; - if (intel_dp_aux_native_read(intel_dp, 0x000, intel_dp->dpcd, - sizeof (intel_dp->dpcd)) == sizeof (intel_dp->dpcd)) - { - if (intel_dp->dpcd[DP_DPCD_REV] != 0) - status = connector_status_connected; - } - - return status; + return i915_dp_detect_common(intel_dp); } /** From a2cab1b24a4ea75a68fa21bfb7d5b1a45121583c Mon Sep 17 00:00:00 2001 From: Adam Jackson Date: Tue, 12 Jul 2011 17:38:05 -0400 Subject: [PATCH 055/439] drm/i915/dp: Explicitly request 8/10 channel coding It's not clear what a sink would do if you wrote zero to this register - which I guess would mean "I don't support any channel encodings, good luck" - but let's not find out. Signed-off-by: Adam Jackson Reviewed-by: Keith Packard Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_dp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 9a0c3ca1ffa9..1c3a36feaf69 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -769,6 +769,7 @@ intel_dp_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode, memset(intel_dp->link_configuration, 0, DP_LINK_CONFIGURATION_SIZE); intel_dp->link_configuration[0] = intel_dp->link_bw; intel_dp->link_configuration[1] = intel_dp->lane_count; + intel_dp->link_configuration[8] = DP_SET_ANSI_8B10B; /* * Check for DPCD version > 1.1 and enhanced framing support From 92b8e897f6e7ba4aa10037ebd8186f85d39330d0 Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Tue, 12 Jul 2011 10:57:59 -0700 Subject: [PATCH 056/439] btrfs: Don't BUG_ON alloc_path errors in find_next_chunk I also removed the BUG_ON from error return of find_next_chunk in init_first_rw_device(). It turns out that the only caller of init_first_rw_device() also BUGS on any nonzero return so no actual behavior change has occurred here. do_chunk_alloc() also needed an update since it calls btrfs_alloc_chunk() which can now return -ENOMEM. Instead of setting space_info->full on any error from btrfs_alloc_chunk() I catch and return every error value _except_ -ENOSPC. Thanks goes to Tsutomu Itoh for pointing that issue out. Signed-off-by: Mark Fasheh --- fs/btrfs/extent-tree.c | 4 ++++ fs/btrfs/volumes.c | 6 ++++-- 2 files changed, 8 insertions(+), 2 deletions(-) diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index aa91773fe31b..f6af4236e59b 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -3277,6 +3277,9 @@ again: } ret = btrfs_alloc_chunk(trans, extent_root, flags); + if (ret < 0 && ret != -ENOSPC) + goto out; + spin_lock(&space_info->lock); if (ret) space_info->full = 1; @@ -3286,6 +3289,7 @@ again: space_info->force_alloc = CHUNK_ALLOC_NO_FORCE; space_info->chunk_alloc = 0; spin_unlock(&space_info->lock); +out: mutex_unlock(&extent_root->fs_info->chunk_mutex); return ret; } diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 530a2fcea1ef..90d956c17d92 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -1037,7 +1037,8 @@ static noinline int find_next_chunk(struct btrfs_root *root, struct btrfs_key found_key; path = btrfs_alloc_path(); - BUG_ON(!path); + if (!path) + return -ENOMEM; key.objectid = objectid; key.offset = (u64)-1; @@ -2663,7 +2664,8 @@ static noinline int init_first_rw_device(struct btrfs_trans_handle *trans, ret = find_next_chunk(fs_info->chunk_root, BTRFS_FIRST_CHUNK_TREE_OBJECTID, &chunk_offset); - BUG_ON(ret); + if (ret) + return ret; alloc_profile = BTRFS_BLOCK_GROUP_METADATA | (fs_info->metadata_alloc_profile & From 38a1a919535742af677303271eb4ff731547b706 Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Wed, 13 Jul 2011 10:59:59 -0700 Subject: [PATCH 057/439] btrfs: don't BUG_ON allocation errors in btrfs_drop_snapshot In addition to properly handling allocation failure from btrfs_alloc_path, I also fixed up the kzalloc error handling code immediately below it. Signed-off-by: Mark Fasheh --- fs/btrfs/extent-tree.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index f6af4236e59b..6bce721e7bbc 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -6272,10 +6272,14 @@ int btrfs_drop_snapshot(struct btrfs_root *root, int level; path = btrfs_alloc_path(); - BUG_ON(!path); + if (!path) + return -ENOMEM; wc = kzalloc(sizeof(*wc), GFP_NOFS); - BUG_ON(!wc); + if (!wc) { + btrfs_free_path(path); + return -ENOMEM; + } trans = btrfs_start_transaction(tree_root, 0); BUG_ON(IS_ERR(trans)); From a65e34c79c88895766ab1f8a5afa451eed26622b Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Mon, 25 Jul 2011 10:04:56 -0700 Subject: [PATCH 058/439] drm/i915: Hold mode_config->mutex during hotplug processing Hotplug detection is a mode setting operation and must hold the struct_mutex or risk colliding with other mode setting operations. In particular, the display port hotplug function attempts to re-train the link if the monitor is supposed to be running when plugged back in. If that happens while mode setting is underway, the link will get scrambled, leaving it in an inconsistent state. This is a special case -- usually the driver mode setting entry points are covered by the upper level DRM code, but in this case the function is invoked as a work function not under the control of DRM. Signed-off-by: Keith Packard Cc: stable@kernel.org Reviewed-by: Jesse Barnes --- drivers/gpu/drm/i915/i915_irq.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 3b03f85ea627..9da2a2c99953 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -306,6 +306,7 @@ static void i915_hotplug_work_func(struct work_struct *work) struct drm_mode_config *mode_config = &dev->mode_config; struct intel_encoder *encoder; + mutex_lock(&mode_config->mutex); DRM_DEBUG_KMS("running encoder hotplug functions\n"); list_for_each_entry(encoder, &mode_config->encoder_list, base.head) @@ -314,6 +315,8 @@ static void i915_hotplug_work_func(struct work_struct *work) /* Just fire off a uevent and let userspace tell us what to do */ drm_helper_hpd_irq_event(dev); + + mutex_unlock(&mode_config->mutex); } static void i915_handle_rps_change(struct drm_device *dev) From 302983e9059e9ef5de3ca7671918eeb237c5971e Mon Sep 17 00:00:00 2001 From: Adam Jackson Date: Wed, 13 Jul 2011 16:32:32 -0400 Subject: [PATCH 059/439] drm/i915/pch: Fix integer math bugs in panel fitting MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Consider a 1600x900 panel, upscaling a 1360x768 mode, full-aspect. The old math would give you: scaled_width = 1600 * 768; /* 1228800 */ scaled_height = 1360 * 900; /* 1224000 */ if (scaled_width > scaled_height) { /* pillarbox, and true */ width = 1224000 / 768; /* int(1593.75) = 1593 */ x = (1600 - 1593 + 1) / 2; /* 4 */ y = 0; height = 768; } /* ... */ This is broken. The total width of scanout would then be 1593 + 4 + 4, or 1601, which is wider than the panel itself. The hardware very dutifully implements this, and you end up with a black 45° diagonal from the top-left corner to the bottom edge of the screen. It's a cool effect and all, but not what you wanted. Similar things happen for the letterbox case. The problem is that you have an integer number of pixels, which means it's usually impossible to upscale equally on both axes. 1360/768 is 1.7708, 1600/900 is 1.7777. Since we're constrained on the one axis, the other one wants to come out as an even number of pixels (the panel is almost certainly even on both axes, and the x/y offsets will be applied on both sides). In the math above, if 'width' comes out even, rounding down is correct; if it's odd, you'd rather round up. So just increment width/height in those cases. Tested on a Lenovo T500 (Ironlake). Signed-off-by: Adam Jackson Tested-By: Daniel Manrique Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=38851 Reviewed-by: Chris Wilson Cc: stable@kernel.org Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_panel.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c index a06ff07a4d3b..05f500cd9c24 100644 --- a/drivers/gpu/drm/i915/intel_panel.c +++ b/drivers/gpu/drm/i915/intel_panel.c @@ -83,11 +83,15 @@ intel_pch_panel_fitting(struct drm_device *dev, u32 scaled_height = mode->hdisplay * adjusted_mode->vdisplay; if (scaled_width > scaled_height) { /* pillar */ width = scaled_height / mode->vdisplay; + if (width & 1) + width++; x = (adjusted_mode->hdisplay - width + 1) / 2; y = 0; height = adjusted_mode->vdisplay; } else if (scaled_width < scaled_height) { /* letter */ height = scaled_width / mode->hdisplay; + if (height & 1) + height++; y = (adjusted_mode->vdisplay - height + 1) / 2; x = 0; width = adjusted_mode->hdisplay; From 81055854d096959898fdc17ed11729eb019eff07 Mon Sep 17 00:00:00 2001 From: Adam Jackson Date: Thu, 21 Jul 2011 17:48:37 -0400 Subject: [PATCH 060/439] drm/i915/dp: Explicitly disable symbol scrambling while training The DP spec says training patterns 1 and 2 are to be sent non-scrambled, and the GPU docs claim that happens (or at least, there's no explicit scrambling control). But the sink may be confused if we don't explicitly tell it what we're doing, so play it safe. Signed-off-by: Adam Jackson Reviewed-by: Jesse Barnes Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_dp.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 1c3a36feaf69..8aecb072466b 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -1370,7 +1370,8 @@ intel_dp_start_link_train(struct intel_dp *intel_dp) reg = DP | DP_LINK_TRAIN_PAT_1; if (!intel_dp_set_link_train(intel_dp, reg, - DP_TRAINING_PATTERN_1)) + DP_TRAINING_PATTERN_1 | + DP_LINK_SCRAMBLING_DISABLE)) break; /* Set training pattern 1 */ @@ -1445,7 +1446,8 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp) /* channel eq pattern */ if (!intel_dp_set_link_train(intel_dp, reg, - DP_TRAINING_PATTERN_2)) + DP_TRAINING_PATTERN_2 | + DP_LINK_SCRAMBLING_DISABLE)) break; udelay(400); From e85194641bec56179dcf5e1704ce5c6bf30340c6 Mon Sep 17 00:00:00 2001 From: Adam Jackson Date: Thu, 21 Jul 2011 17:48:38 -0400 Subject: [PATCH 061/439] drm/i915/dp: Don't turn CPT DP ports on too early The docs say the port has to come on in training pattern 1; at this point, though, ->DP is in normal mode. The intent here is to wait until the port is in fact sending data, but that doesn't happen since we've broken the sequence the hardware expects, and the vblank wait will time out and kvetch in the log. Signed-off-by: Adam Jackson Reviewed-by: Jesse Barnes Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_dp.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 8aecb072466b..dcc7ae6d4141 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -1334,10 +1334,16 @@ intel_dp_start_link_train(struct intel_dp *intel_dp) u32 reg; uint32_t DP = intel_dp->DP; - /* Enable output, wait for it to become active */ - I915_WRITE(intel_dp->output_reg, intel_dp->DP); - POSTING_READ(intel_dp->output_reg); - intel_wait_for_vblank(dev, intel_crtc->pipe); + /* + * On CPT we have to enable the port in training pattern 1, which + * will happen below in intel_dp_set_link_train. Otherwise, enable + * the port and wait for it to become active. + */ + if (!HAS_PCH_CPT(dev)) { + I915_WRITE(intel_dp->output_reg, intel_dp->DP); + POSTING_READ(intel_dp->output_reg); + intel_wait_for_vblank(dev, intel_crtc->pipe); + } /* Write the link configuration data */ intel_dp_aux_native_write(intel_dp, DP_LINK_BW_SET, From 9ea71503a8ed9184d2d0b8ccc4d269d05f7940ae Mon Sep 17 00:00:00 2001 From: Shawn Bohrer Date: Thu, 30 Jun 2011 11:21:32 -0500 Subject: [PATCH 062/439] futex: Fix regression with read only mappings commit 7485d0d3758e8e6491a5c9468114e74dc050785d (futexes: Remove rw parameter from get_futex_key()) in 2.6.33 fixed two problems: First, It prevented a loop when encountering a ZERO_PAGE. Second, it fixed RW MAP_PRIVATE futex operations by forcing the COW to occur by unconditionally performing a write access get_user_pages_fast() to get the page. The commit also introduced a user-mode regression in that it broke futex operations on read-only memory maps. For example, this breaks workloads that have one or more reader processes doing a FUTEX_WAIT on a futex within a read only shared file mapping, and a writer processes that has a writable mapping issuing the FUTEX_WAKE. This fixes the regression for valid futex operations on RO mappings by trying a RO get_user_pages_fast() when the RW get_user_pages_fast() fails. This change makes it necessary to also check for invalid use cases, such as anonymous RO mappings (which can never change) and the ZERO_PAGE which the commit referenced above was written to address. This patch does restore the original behavior with RO MAP_PRIVATE mappings, which have inherent user-mode usage problems and don't really make sense. With this patch performing a FUTEX_WAIT within a RO MAP_PRIVATE mapping will be successfully woken provided another process updates the region of the underlying mapped file. However, the mmap() man page states that for a MAP_PRIVATE mapping: It is unspecified whether changes made to the file after the mmap() call are visible in the mapped region. So user-mode users attempting to use futex operations on RO MAP_PRIVATE mappings are depending on unspecified behavior. Additionally a RO MAP_PRIVATE mapping could fail to wake up in the following case. Thread-A: call futex(FUTEX_WAIT, memory-region-A). get_futex_key() return inode based key. sleep on the key Thread-B: call mprotect(PROT_READ|PROT_WRITE, memory-region-A) Thread-B: write memory-region-A. COW happen. This process's memory-region-A become related to new COWed private (ie PageAnon=1) page. Thread-B: call futex(FUETX_WAKE, memory-region-A). get_futex_key() return mm based key. IOW, we fail to wake up Thread-A. Once again doing something like this is just silly and users who do something like this get what they deserve. While RO MAP_PRIVATE mappings are nonsensical, checking for a private mapping requires walking the vmas and was deemed too costly to avoid a userspace hang. This Patch is based on Peter Zijlstra's initial patch with modifications to only allow RO mappings for futex operations that need VERIFY_READ access. Reported-by: David Oliver Signed-off-by: Shawn Bohrer Acked-by: Peter Zijlstra Signed-off-by: Darren Hart Cc: KOSAKI Motohiro Cc: peterz@infradead.org Cc: eric.dumazet@gmail.com Cc: zvonler@rgmadvisors.com Cc: hughd@google.com Link: http://lkml.kernel.org/r/1309450892-30676-1-git-send-email-sbohrer@rgmadvisors.com Cc: stable@kernel.org Signed-off-by: Thomas Gleixner --- kernel/futex.c | 54 +++++++++++++++++++++++++++++++++++++++----------- 1 file changed, 42 insertions(+), 12 deletions(-) diff --git a/kernel/futex.c b/kernel/futex.c index fe28dc282eae..70bb54bd2468 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -218,6 +218,8 @@ static void drop_futex_key_refs(union futex_key *key) * @uaddr: virtual address of the futex * @fshared: 0 for a PROCESS_PRIVATE futex, 1 for PROCESS_SHARED * @key: address where result is stored. + * @rw: mapping needs to be read/write (values: VERIFY_READ, + * VERIFY_WRITE) * * Returns a negative error code or 0 * The key words are stored in *key on success. @@ -229,12 +231,12 @@ static void drop_futex_key_refs(union futex_key *key) * lock_page() might sleep, the caller should not hold a spinlock. */ static int -get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key) +get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key, int rw) { unsigned long address = (unsigned long)uaddr; struct mm_struct *mm = current->mm; struct page *page, *page_head; - int err; + int err, ro = 0; /* * The futex address must be "naturally" aligned. @@ -262,8 +264,18 @@ get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key) again: err = get_user_pages_fast(address, 1, 1, &page); + /* + * If write access is not required (eg. FUTEX_WAIT), try + * and get read-only access. + */ + if (err == -EFAULT && rw == VERIFY_READ) { + err = get_user_pages_fast(address, 1, 0, &page); + ro = 1; + } if (err < 0) return err; + else + err = 0; #ifdef CONFIG_TRANSPARENT_HUGEPAGE page_head = page; @@ -305,6 +317,13 @@ again: if (!page_head->mapping) { unlock_page(page_head); put_page(page_head); + /* + * ZERO_PAGE pages don't have a mapping. Avoid a busy loop + * trying to find one. RW mapping would have COW'd (and thus + * have a mapping) so this page is RO and won't ever change. + */ + if ((page_head == ZERO_PAGE(address))) + return -EFAULT; goto again; } @@ -316,6 +335,15 @@ again: * the object not the particular process. */ if (PageAnon(page_head)) { + /* + * A RO anonymous page will never change and thus doesn't make + * sense for futex operations. + */ + if (ro) { + err = -EFAULT; + goto out; + } + key->both.offset |= FUT_OFF_MMSHARED; /* ref taken on mm */ key->private.mm = mm; key->private.address = address; @@ -327,9 +355,10 @@ again: get_futex_key_refs(key); +out: unlock_page(page_head); put_page(page_head); - return 0; + return err; } static inline void put_futex_key(union futex_key *key) @@ -940,7 +969,7 @@ futex_wake(u32 __user *uaddr, unsigned int flags, int nr_wake, u32 bitset) if (!bitset) return -EINVAL; - ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &key); + ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &key, VERIFY_READ); if (unlikely(ret != 0)) goto out; @@ -986,10 +1015,10 @@ futex_wake_op(u32 __user *uaddr1, unsigned int flags, u32 __user *uaddr2, int ret, op_ret; retry: - ret = get_futex_key(uaddr1, flags & FLAGS_SHARED, &key1); + ret = get_futex_key(uaddr1, flags & FLAGS_SHARED, &key1, VERIFY_READ); if (unlikely(ret != 0)) goto out; - ret = get_futex_key(uaddr2, flags & FLAGS_SHARED, &key2); + ret = get_futex_key(uaddr2, flags & FLAGS_SHARED, &key2, VERIFY_WRITE); if (unlikely(ret != 0)) goto out_put_key1; @@ -1243,10 +1272,11 @@ retry: pi_state = NULL; } - ret = get_futex_key(uaddr1, flags & FLAGS_SHARED, &key1); + ret = get_futex_key(uaddr1, flags & FLAGS_SHARED, &key1, VERIFY_READ); if (unlikely(ret != 0)) goto out; - ret = get_futex_key(uaddr2, flags & FLAGS_SHARED, &key2); + ret = get_futex_key(uaddr2, flags & FLAGS_SHARED, &key2, + requeue_pi ? VERIFY_WRITE : VERIFY_READ); if (unlikely(ret != 0)) goto out_put_key1; @@ -1790,7 +1820,7 @@ static int futex_wait_setup(u32 __user *uaddr, u32 val, unsigned int flags, * while the syscall executes. */ retry: - ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &q->key); + ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &q->key, VERIFY_READ); if (unlikely(ret != 0)) return ret; @@ -1941,7 +1971,7 @@ static int futex_lock_pi(u32 __user *uaddr, unsigned int flags, int detect, } retry: - ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &q.key); + ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &q.key, VERIFY_WRITE); if (unlikely(ret != 0)) goto out; @@ -2060,7 +2090,7 @@ retry: if ((uval & FUTEX_TID_MASK) != vpid) return -EPERM; - ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &key); + ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &key, VERIFY_WRITE); if (unlikely(ret != 0)) goto out; @@ -2249,7 +2279,7 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, debug_rt_mutex_init_waiter(&rt_waiter); rt_waiter.task = NULL; - ret = get_futex_key(uaddr2, flags & FLAGS_SHARED, &key2); + ret = get_futex_key(uaddr2, flags & FLAGS_SHARED, &key2, VERIFY_WRITE); if (unlikely(ret != 0)) goto out; From 4a28a3fa8c37193b105807193b3695048f0050a4 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 27 Jul 2011 12:58:17 +0300 Subject: [PATCH 063/439] iscsi-target: Fix NULL dereference on allocation failure This patch fixes a bug in iscsi_target_init_negotiation() where the "goto out" path dereferences "login" which is NULL upon a memory allocation failure. Signed-off-by: Dan Carpenter Signed-off-by: Nicholas Bellinger --- drivers/target/iscsi/iscsi_target_nego.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/target/iscsi/iscsi_target_nego.c b/drivers/target/iscsi/iscsi_target_nego.c index 713a4d23557a..4d087ac11067 100644 --- a/drivers/target/iscsi/iscsi_target_nego.c +++ b/drivers/target/iscsi/iscsi_target_nego.c @@ -978,7 +978,7 @@ struct iscsi_login *iscsi_target_init_negotiation( pr_err("Unable to allocate memory for struct iscsi_login.\n"); iscsit_tx_login_rsp(conn, ISCSI_STATUS_CLS_TARGET_ERR, ISCSI_LOGIN_STATUS_NO_RESOURCES); - goto out; + return NULL; } login->req = kzalloc(ISCSI_HDR_LEN, GFP_KERNEL); From 8f50c7f5d63b0e4a29f6f96e8cfaca782e0c458d Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 27 Jul 2011 14:11:43 +0300 Subject: [PATCH 064/439] iscsi-target: strlen() doesn't count the terminator This patch fixes an off by one check in iscsit_add_tiqn() because the NULL terminator isn't taken into consideration. Signed-off-by: Dan Carpenter Signed-off-by: Nicholas Bellinger --- drivers/target/iscsi/iscsi_target.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c index 14c81c4265bd..396c79d8569c 100644 --- a/drivers/target/iscsi/iscsi_target.c +++ b/drivers/target/iscsi/iscsi_target.c @@ -120,7 +120,7 @@ struct iscsi_tiqn *iscsit_add_tiqn(unsigned char *buf) struct iscsi_tiqn *tiqn = NULL; int ret; - if (strlen(buf) > ISCSI_IQN_LEN) { + if (strlen(buf) >= ISCSI_IQN_LEN) { pr_err("Target IQN exceeds %d bytes\n", ISCSI_IQN_LEN); return ERR_PTR(-EINVAL); From 76f1928e8831961a692459924cfac0ca1b0d303a Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Wed, 27 Jul 2011 12:16:22 -0700 Subject: [PATCH 065/439] iscsi-target: Fix uninitialized usage of cmd->pad_bytes This patch fixes an uninitialized usage of cmd->pad_bytes inside of iscsit_handle_text_cmd() introduced during a v4.1 change to use cmd members instead of local pad_bytes variables. Reported-by: Dan Carpenter Signed-off-by: Nicholas Bellinger --- drivers/target/iscsi/iscsi_target.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c index 396c79d8569c..3ab8abca29c3 100644 --- a/drivers/target/iscsi/iscsi_target.c +++ b/drivers/target/iscsi/iscsi_target.c @@ -1857,7 +1857,7 @@ static int iscsit_handle_text_cmd( char *text_ptr, *text_in; int cmdsn_ret, niov = 0, rx_got, rx_size; u32 checksum = 0, data_crc = 0, payload_length; - u32 padding = 0, text_length = 0; + u32 padding = 0, pad_bytes = 0, text_length = 0; struct iscsi_cmd *cmd; struct kvec iov[3]; struct iscsi_text *hdr; @@ -1896,7 +1896,7 @@ static int iscsit_handle_text_cmd( padding = ((-payload_length) & 3); if (padding != 0) { - iov[niov].iov_base = cmd->pad_bytes; + iov[niov].iov_base = &pad_bytes; iov[niov++].iov_len = padding; rx_size += padding; pr_debug("Receiving %u additional bytes" @@ -1917,7 +1917,7 @@ static int iscsit_handle_text_cmd( if (conn->conn_ops->DataDigest) { iscsit_do_crypto_hash_buf(&conn->conn_rx_hash, text_in, text_length, - padding, cmd->pad_bytes, + padding, (u8 *)&pad_bytes, (u8 *)&data_crc); if (checksum != data_crc) { From 7bbb654e9c77a8692425dbd2c704ad245b86ec8a Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Wed, 27 Jul 2011 12:37:03 -0700 Subject: [PATCH 066/439] iscsi-target: Fix snprintf usage with MAX_PORTAL_LEN This patch makes lio_target_call_addnptotpg() use sprintf() with MAX_PORTAL_LEN + 1 to address the following smatch warning: drivers/target/iscsi/iscsi_target_configfs.c +184 lio_target_call_addnptotpg(21) error: snprintf() chops off the last chars of 'name': 257 vs 256 Reported-by: Dan Carpenter Signed-off-by: Nicholas Bellinger --- drivers/target/iscsi/iscsi_target_configfs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/target/iscsi/iscsi_target_configfs.c b/drivers/target/iscsi/iscsi_target_configfs.c index 32bb92c44450..f095e65b1ccf 100644 --- a/drivers/target/iscsi/iscsi_target_configfs.c +++ b/drivers/target/iscsi/iscsi_target_configfs.c @@ -181,7 +181,7 @@ struct se_tpg_np *lio_target_call_addnptotpg( return ERR_PTR(-EOVERFLOW); } memset(buf, 0, MAX_PORTAL_LEN + 1); - snprintf(buf, MAX_PORTAL_LEN, "%s", name); + snprintf(buf, MAX_PORTAL_LEN + 1, "%s", name); memset(&sockaddr, 0, sizeof(struct __kernel_sockaddr_storage)); From aadcec010127a20599500c7007fbe9998744be41 Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Wed, 27 Jul 2011 20:13:22 +0000 Subject: [PATCH 067/439] iscsi-target: Fix CONFIG_SMP=n and CONFIG_MODULES=n build failure This patch fixes the following CONFIG_SMP=n and CONFIG_MODULES=n build failure, because iscsit_thread_get_cpumask() is defined as a macro in iscsi_target.c, but needed by iscsi_target_login.c drivers/built-in.o: In function `iscsi_post_login_handler': iscsi_target_login.c:(.text+0x13a315): undefined reference to `iscsit_thread_get_cpumask' iscsi_target_login.c:(.text+0x13a4b4): undefined reference to `iscsit_thread_get_cpumask' make: *** [.tmp_vmlinux1] Error 1 Reported-by: Randy Dunlap Signed-off-by: Nicholas Bellinger --- drivers/target/iscsi/iscsi_target.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c index 3ab8abca29c3..c24fb10de60b 100644 --- a/drivers/target/iscsi/iscsi_target.c +++ b/drivers/target/iscsi/iscsi_target.c @@ -3468,7 +3468,12 @@ static inline void iscsit_thread_check_cpumask( } #else -#define iscsit_thread_get_cpumask(X) ({}) + +void iscsit_thread_get_cpumask(struct iscsi_conn *conn) +{ + return; +} + #define iscsit_thread_check_cpumask(X, Y, Z) ({}) #endif /* CONFIG_SMP */ From b9f5edc25036779bbff1272f8f94860d2d7a483b Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Wed, 27 Jul 2011 20:21:15 +0000 Subject: [PATCH 068/439] MAINTAINERS: Add target-devel list for drivers/target/ Reported-by: Randy Dunlap Signed-off-by: Nicholas Bellinger --- MAINTAINERS | 1 + 1 file changed, 1 insertion(+) diff --git a/MAINTAINERS b/MAINTAINERS index 41ec646d8a98..fcd29e36b84d 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -6109,6 +6109,7 @@ F: include/linux/sysv_fs.h TARGET SUBSYSTEM M: Nicholas A. Bellinger L: linux-scsi@vger.kernel.org +L: target-devel@vger.kernel.org L: http://groups.google.com/group/linux-iscsi-target-dev W: http://www.linux-iscsi.org T: git git://git.kernel.org/pub/scm/linux/kernel/git/nab/lio-core-2.6.git master From 5db0753ba5aabcd9fa298029f03b32b3c96b5a39 Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Wed, 27 Jul 2011 22:18:52 -0700 Subject: [PATCH 069/439] target: Fix WRITE_SAME_16 lba assignment breakage This patch fixes a bug in WRITE_SAME_16 LBA assignment where get_unaligned_be16() is incorrectly being used instead of get_unaligned_be64() for a 64-bit LBA. This was introduced with: commit a1d8b49abd60ba5d09e7c968731abcb0f8f1cbf6 Author: Andy Grover Date: Mon May 2 17:12:10 2011 -0700 target: Updates from AGrover and HCH (round 3) (target: inline struct se_transport_task into struct se_cmd) Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_transport.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c index c75a01a1c475..ff7fcf8366a0 100644 --- a/drivers/target/target_core_transport.c +++ b/drivers/target/target_core_transport.c @@ -3324,7 +3324,7 @@ static int transport_generic_cmd_sequencer( goto out_invalid_cdb_field; } - cmd->t_task_lba = get_unaligned_be16(&cdb[2]); + cmd->t_task_lba = get_unaligned_be64(&cdb[2]); passthrough = (dev->transport->transport_type == TRANSPORT_PLUGIN_PHBA_PDEV); /* From ca9380fd68514c7bc952282c1b4fc70607e9fe43 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Thu, 28 Jul 2011 14:46:05 +0200 Subject: [PATCH 070/439] ALSA: sound/core/pcm_compat.c: adjust array index Convert array index from the loop bound to the loop index. A simplified version of the semantic patch that fixes this problem is as follows: (http://coccinelle.lip6.fr/) // @@ expression e1,e2,ar; @@ for(e1 = 0; e1 < e2; e1++) { <... ar[ - e2 + e1 ] ...> } // Signed-off-by: Julia Lawall Cc: Signed-off-by: Takashi Iwai --- sound/core/pcm_compat.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/core/pcm_compat.c b/sound/core/pcm_compat.c index 5fb2e28e796f..91cdf9435fec 100644 --- a/sound/core/pcm_compat.c +++ b/sound/core/pcm_compat.c @@ -342,7 +342,7 @@ static int snd_pcm_ioctl_xfern_compat(struct snd_pcm_substream *substream, kfree(bufs); return -EFAULT; } - bufs[ch] = compat_ptr(ptr); + bufs[i] = compat_ptr(ptr); bufptr++; } if (dir == SNDRV_PCM_STREAM_PLAYBACK) From d52a10d003694e48d28cf0218db16372ed61f1bd Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 28 Jul 2011 09:48:19 -0700 Subject: [PATCH 071/439] target: iscsi_target depends on NET iscsi target code uses lots on network interface functions, so it should depend on NET. Fixes many build errors when NET is not enabled: ERROR: "kernel_sendmsg" [drivers/target/iscsi/iscsi_target_mod.ko] undefined! ERROR: "in_aton" [drivers/target/iscsi/iscsi_target_mod.ko] undefined! ERROR: "sock_release" [drivers/target/iscsi/iscsi_target_mod.ko] undefined! ERROR: "kernel_listen" [drivers/target/iscsi/iscsi_target_mod.ko] undefined! ERROR: "kernel_setsockopt" [drivers/target/iscsi/iscsi_target_mod.ko] undefined! ERROR: "kernel_recvmsg" [drivers/target/iscsi/iscsi_target_mod.ko] undefined! ERROR: "kernel_accept" [drivers/target/iscsi/iscsi_target_mod.ko] undefined! ERROR: "sock_create" [drivers/target/iscsi/iscsi_target_mod.ko] undefined! ERROR: "kernel_bind" [drivers/target/iscsi/iscsi_target_mod.ko] undefined! ERROR: "in6_pton" [drivers/target/iscsi/iscsi_target_mod.ko] undefined! Signed-off-by: Randy Dunlap Signed-off-by: Nicholas Bellinger --- drivers/target/iscsi/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/target/iscsi/Kconfig b/drivers/target/iscsi/Kconfig index 564ff4e0dbc4..8345fb457a40 100644 --- a/drivers/target/iscsi/Kconfig +++ b/drivers/target/iscsi/Kconfig @@ -1,5 +1,6 @@ config ISCSI_TARGET tristate "Linux-iSCSI.org iSCSI Target Mode Stack" + depends on NET select CRYPTO select CRYPTO_CRC32C select CRYPTO_CRC32C_INTEL if X86 From 40ee3381dd1010432acc13e907329029096c5bfc Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Thu, 28 Jul 2011 15:31:19 -0700 Subject: [PATCH 072/439] drm/i915: Fixup for 'Hold mode_config->mutex during hotplug' drm_helper_hpd_irq_event queues another work proc to go and deliver the user-space event, and that function also wants to hold the config mutex, so we shouldn't hold the mutex across the drm_helper_hpd_irq_event call. Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/i915_irq.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 9da2a2c99953..9b1d669f7d4b 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -313,10 +313,10 @@ static void i915_hotplug_work_func(struct work_struct *work) if (encoder->hot_plug) encoder->hot_plug(encoder); + mutex_unlock(&mode_config->mutex); + /* Just fire off a uevent and let userspace tell us what to do */ drm_helper_hpd_irq_event(dev); - - mutex_unlock(&mode_config->mutex); } static void i915_handle_rps_change(struct drm_device *dev) From 92fd8fd13b7570f6a8ba519c4e8ec98f10a86ce9 Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Mon, 25 Jul 2011 19:50:10 -0700 Subject: [PATCH 073/439] drm/i915: Use dp_detect_common in hotplug helper function This uses the common dpcd reading routine, i915_dp_detect_common, instead of open-coding a call to intel_dp_aux_native_read. Besides reducing duplicated code, this also gains the read retries which may be necessary when a cable is first plugged back in and the link needs to be retrained. Signed-off-by: Keith Packard Reviewed-by: Jesse Barnes Reviewed-by: Adam Jackson --- drivers/gpu/drm/i915/intel_dp.c | 39 ++++++++++++++++----------------- 1 file changed, 19 insertions(+), 20 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index dcc7ae6d4141..45db8106ab12 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -1567,6 +1567,20 @@ intel_dp_link_down(struct intel_dp *intel_dp) POSTING_READ(intel_dp->output_reg); } +static enum drm_connector_status +i915_dp_detect_common(struct intel_dp *intel_dp) +{ + enum drm_connector_status status = connector_status_disconnected; + + if (intel_dp_aux_native_read_retry(intel_dp, 0x000, intel_dp->dpcd, + sizeof (intel_dp->dpcd)) && + (intel_dp->dpcd[DP_DPCD_REV] != 0)) { + status = connector_status_connected; + } + + return status; +} + /* * According to DP spec * 5.1.2: @@ -1579,44 +1593,29 @@ intel_dp_link_down(struct intel_dp *intel_dp) static void intel_dp_check_link_status(struct intel_dp *intel_dp) { - int ret; - if (!intel_dp->base.base.crtc) return; + /* Try to read receiver status if the link appears to be up */ if (!intel_dp_get_link_status(intel_dp)) { intel_dp_link_down(intel_dp); return; } - /* Try to read receiver status if the link appears to be up */ - ret = intel_dp_aux_native_read(intel_dp, - 0x000, intel_dp->dpcd, - sizeof (intel_dp->dpcd)); - if (ret != sizeof(intel_dp->dpcd)) { + /* Now read the DPCD to see if it's actually running */ + if (i915_dp_detect_common(intel_dp) != connector_status_connected) { intel_dp_link_down(intel_dp); return; } if (!intel_channel_eq_ok(intel_dp)) { + DRM_DEBUG_KMS("%s: channel EQ not ok, retraining\n", + drm_get_encoder_name(&intel_dp->base.base)); intel_dp_start_link_train(intel_dp); intel_dp_complete_link_train(intel_dp); } } -static enum drm_connector_status -i915_dp_detect_common(struct intel_dp *intel_dp) -{ - enum drm_connector_status status = connector_status_disconnected; - - if (intel_dp_aux_native_read_retry(intel_dp, 0x000, intel_dp->dpcd, - sizeof (intel_dp->dpcd)) && - (intel_dp->dpcd[DP_DPCD_REV] != 0)) - status = connector_status_connected; - - return status; -} - static enum drm_connector_status ironlake_dp_detect(struct intel_dp *intel_dp) { From 26d61aad7a46115628341e9eb95433f30efef21a Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Mon, 25 Jul 2011 20:01:09 -0700 Subject: [PATCH 074/439] drm/i915: Rename i915_dp_detect_common to intel_dp_get_dpcd This describes the function better, allowing it to be used where the DPCD value is relevant. Signed-off-by: Keith Packard Reviewed-by: Jesse Barnes Reviewed-by: Adam Jackson --- drivers/gpu/drm/i915/intel_dp.c | 24 +++++++++++++++--------- 1 file changed, 15 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 45db8106ab12..41674e1a26c7 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -1567,18 +1567,16 @@ intel_dp_link_down(struct intel_dp *intel_dp) POSTING_READ(intel_dp->output_reg); } -static enum drm_connector_status -i915_dp_detect_common(struct intel_dp *intel_dp) +static bool +intel_dp_get_dpcd(struct intel_dp *intel_dp) { - enum drm_connector_status status = connector_status_disconnected; - if (intel_dp_aux_native_read_retry(intel_dp, 0x000, intel_dp->dpcd, sizeof (intel_dp->dpcd)) && (intel_dp->dpcd[DP_DPCD_REV] != 0)) { - status = connector_status_connected; + return true; } - return status; + return false; } /* @@ -1603,7 +1601,7 @@ intel_dp_check_link_status(struct intel_dp *intel_dp) } /* Now read the DPCD to see if it's actually running */ - if (i915_dp_detect_common(intel_dp) != connector_status_connected) { + if (!intel_dp_get_dpcd(intel_dp)) { intel_dp_link_down(intel_dp); return; } @@ -1616,6 +1614,14 @@ intel_dp_check_link_status(struct intel_dp *intel_dp) } } +static enum drm_connector_status +intel_dp_detect_dpcd(struct intel_dp *intel_dp) +{ + if (intel_dp_get_dpcd(intel_dp)) + return connector_status_connected; + return connector_status_disconnected; +} + static enum drm_connector_status ironlake_dp_detect(struct intel_dp *intel_dp) { @@ -1629,7 +1635,7 @@ ironlake_dp_detect(struct intel_dp *intel_dp) return status; } - return i915_dp_detect_common(intel_dp); + return intel_dp_detect_dpcd(intel_dp); } static enum drm_connector_status @@ -1658,7 +1664,7 @@ g4x_dp_detect(struct intel_dp *intel_dp) if ((temp & bit) == 0) return connector_status_disconnected; - return i915_dp_detect_common(intel_dp); + return intel_dp_detect_dpcd(intel_dp); } /** From 59f3e272d788305e16098f0b18309919c9216d67 Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Mon, 25 Jul 2011 20:01:56 -0700 Subject: [PATCH 075/439] drm/i915: In intel_dp_init, replace read of DPCD with intel_dp_get_dpcd Eliminates an open-coded read and also gains the retry behaviour of intel_dp_get_dpcd, which seems like a good idea. Signed-off-by: Keith Packard Reviewed-by: Jesse Barnes Reviewed-by: Adam Jackson --- drivers/gpu/drm/i915/intel_dp.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 41674e1a26c7..9f134d2b7bec 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -2015,7 +2015,7 @@ intel_dp_init(struct drm_device *dev, int output_reg) /* Cache some DPCD data in the eDP case */ if (is_edp(intel_dp)) { - int ret; + bool ret; u32 pp_on, pp_div; pp_on = I915_READ(PCH_PP_ON_DELAYS); @@ -2028,11 +2028,9 @@ intel_dp_init(struct drm_device *dev, int output_reg) dev_priv->panel_t12 *= 100; /* t12 in 100ms units */ ironlake_edp_panel_vdd_on(intel_dp); - ret = intel_dp_aux_native_read(intel_dp, DP_DPCD_REV, - intel_dp->dpcd, - sizeof(intel_dp->dpcd)); + ret = intel_dp_get_dpcd(intel_dp); ironlake_edp_panel_vdd_off(intel_dp); - if (ret == sizeof(intel_dp->dpcd)) { + if (ret) { if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11) dev_priv->no_aux_handshake = intel_dp->dpcd[DP_MAX_DOWNSPREAD] & From f0575e92974d328e8816ed89704c985a7d7d90ac Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Mon, 25 Jul 2011 22:12:43 -0700 Subject: [PATCH 076/439] drm/i915: DP_PIPE_ENABLED must check transcoder on CPT Display port pipe selection on CPT is not done with a bit in the output register, rather it is controlled by a couple of bits in the separate transcoder register which indicate which display port output is connected to the transcoder. This patch replaces the simplistic macro DP_PIPE_ENABLED with the rather more complicated function dp_pipe_enabled which checks the output register to see if that is enabled, and then goes on to either check the output register pipe selection bit (on non-CPT) or the transcoder DP selection bits (on CPT). Before this patch, any time the mode of pipe A was changed, any display port outputs on pipe B would get disabled as intel_disable_pch_ports would ensure that the mode setting operation could occur on pipe A without interference from other outputs connected to that pch port Signed-off-by: Keith Packard Reviewed-by: Jesse Barnes Reviewed-by: Adam Jackson --- drivers/gpu/drm/i915/i915_reg.h | 3 -- drivers/gpu/drm/i915/intel_display.c | 45 +++++++++++++++++++++------- 2 files changed, 34 insertions(+), 14 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 5d5def756c9e..f7315653a842 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -2083,9 +2083,6 @@ #define DP_PIPEB_SELECT (1 << 30) #define DP_PIPE_MASK (1 << 30) -#define DP_PIPE_ENABLED(V, P) \ - (((V) & (DP_PIPE_MASK | DP_PORT_EN)) == ((P) << 30 | DP_PORT_EN)) - /* Link training mode - select a suitable mode for each stage */ #define DP_LINK_TRAIN_PAT_1 (0 << 28) #define DP_LINK_TRAIN_PAT_2 (1 << 28) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 5609c065aaf4..fc26cb4c9ba3 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -979,11 +979,29 @@ static void assert_transcoder_disabled(struct drm_i915_private *dev_priv, pipe_name(pipe)); } +static bool dp_pipe_enabled(struct drm_i915_private *dev_priv, enum pipe pipe, + int reg, u32 port_sel, u32 val) +{ + if ((val & DP_PORT_EN) == 0) + return false; + + if (HAS_PCH_CPT(dev_priv->dev)) { + u32 trans_dp_ctl_reg = TRANS_DP_CTL(pipe); + u32 trans_dp_ctl = I915_READ(trans_dp_ctl_reg); + if ((trans_dp_ctl & TRANS_DP_PORT_SEL_MASK) != port_sel) + return false; + } else { + if ((val & DP_PIPE_MASK) != (pipe << 30)) + return false; + } + return true; +} + static void assert_pch_dp_disabled(struct drm_i915_private *dev_priv, - enum pipe pipe, int reg) + enum pipe pipe, int reg, u32 port_sel) { u32 val = I915_READ(reg); - WARN(DP_PIPE_ENABLED(val, pipe), + WARN(dp_pipe_enabled(dev_priv, pipe, reg, port_sel, val), "PCH DP (0x%08x) enabled on transcoder %c, should be disabled\n", reg, pipe_name(pipe)); } @@ -1003,9 +1021,9 @@ static void assert_pch_ports_disabled(struct drm_i915_private *dev_priv, int reg; u32 val; - assert_pch_dp_disabled(dev_priv, pipe, PCH_DP_B); - assert_pch_dp_disabled(dev_priv, pipe, PCH_DP_C); - assert_pch_dp_disabled(dev_priv, pipe, PCH_DP_D); + assert_pch_dp_disabled(dev_priv, pipe, PCH_DP_B, TRANS_DP_PORT_SEL_B); + assert_pch_dp_disabled(dev_priv, pipe, PCH_DP_C, TRANS_DP_PORT_SEL_C); + assert_pch_dp_disabled(dev_priv, pipe, PCH_DP_D, TRANS_DP_PORT_SEL_D); reg = PCH_ADPA; val = I915_READ(reg); @@ -1334,19 +1352,24 @@ static void intel_disable_plane(struct drm_i915_private *dev_priv, } static void disable_pch_dp(struct drm_i915_private *dev_priv, - enum pipe pipe, int reg) + enum pipe pipe, int reg, u32 port_sel) { u32 val = I915_READ(reg); - if (DP_PIPE_ENABLED(val, pipe)) + if (dp_pipe_enabled(dev_priv, pipe, reg, port_sel, val)) { + DRM_DEBUG_KMS("Disabling pch dp %x on pipe %d\n", reg, pipe); I915_WRITE(reg, val & ~DP_PORT_EN); + } } static void disable_pch_hdmi(struct drm_i915_private *dev_priv, enum pipe pipe, int reg) { u32 val = I915_READ(reg); - if (HDMI_PIPE_ENABLED(val, pipe)) + if (HDMI_PIPE_ENABLED(val, pipe)) { + DRM_DEBUG_KMS("Disabling pch HDMI %x on pipe %d\n", + reg, pipe); I915_WRITE(reg, val & ~PORT_ENABLE); + } } /* Disable any ports connected to this transcoder */ @@ -1358,9 +1381,9 @@ static void intel_disable_pch_ports(struct drm_i915_private *dev_priv, val = I915_READ(PCH_PP_CONTROL); I915_WRITE(PCH_PP_CONTROL, val | PANEL_UNLOCK_REGS); - disable_pch_dp(dev_priv, pipe, PCH_DP_B); - disable_pch_dp(dev_priv, pipe, PCH_DP_C); - disable_pch_dp(dev_priv, pipe, PCH_DP_D); + disable_pch_dp(dev_priv, pipe, PCH_DP_B, TRANS_DP_PORT_SEL_B); + disable_pch_dp(dev_priv, pipe, PCH_DP_C, TRANS_DP_PORT_SEL_C); + disable_pch_dp(dev_priv, pipe, PCH_DP_D, TRANS_DP_PORT_SEL_D); reg = PCH_ADPA; val = I915_READ(reg); From d2b996ac698aebb28557355857927b8b934bb4f9 Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Mon, 25 Jul 2011 22:37:51 -0700 Subject: [PATCH 077/439] Revert and fix "drm/i915/dp: remove DPMS mode tracking from DP" This reverts commit 885a50147f00a8a80108904bf58a18af357717f3. We actually *do* need to track DPMS state so that on hotplug, we don't retrain the link until DPMS is disabled. However, that code had avery small bug -- it wouldn't set the dpms_mode at mode set time, and so link retraining would not actually occur on monitor hotplug until the monitor had gone through a DPMS off/DPMS on cycle. Signed-off-by: Keith Packard Tested-by: Andrew Lutomirski --- drivers/gpu/drm/i915/intel_dp.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 9f134d2b7bec..449364187bc0 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -50,6 +50,7 @@ struct intel_dp { bool has_audio; int force_audio; uint32_t color_range; + int dpms_mode; uint8_t link_bw; uint8_t lane_count; uint8_t dpcd[8]; @@ -1011,6 +1012,8 @@ static void intel_dp_commit(struct drm_encoder *encoder) if (is_edp(intel_dp)) ironlake_edp_backlight_on(dev); + + intel_dp->dpms_mode = DRM_MODE_DPMS_ON; } static void @@ -1045,6 +1048,7 @@ intel_dp_dpms(struct drm_encoder *encoder, int mode) if (is_edp(intel_dp)) ironlake_edp_backlight_on(dev); } + intel_dp->dpms_mode = mode; } /* @@ -1591,6 +1595,9 @@ intel_dp_get_dpcd(struct intel_dp *intel_dp) static void intel_dp_check_link_status(struct intel_dp *intel_dp) { + if (intel_dp->dpms_mode != DRM_MODE_DPMS_ON) + return; + if (!intel_dp->base.base.crtc) return; @@ -1939,6 +1946,7 @@ intel_dp_init(struct drm_device *dev, int output_reg) return; intel_dp->output_reg = output_reg; + intel_dp->dpms_mode = -1; intel_connector = kzalloc(sizeof(struct intel_connector), GFP_KERNEL); if (!intel_connector) { From 120eced9efe7fdb5123db4ea47e9adee9b66284e Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Wed, 27 Jul 2011 01:21:40 -0700 Subject: [PATCH 078/439] drm/i915: Set crtc DPMS mode to ON in intel_crtc_mode_set This corrects the DPMS mode tracking so that the DPMS code will actually turn the CRTC off the next time the screen saves. Signed-off-by: Keith Packard Reviewed-by: Jesse Barnes --- drivers/gpu/drm/i915/intel_display.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index fc26cb4c9ba3..d9b8c15998d9 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -5273,6 +5273,8 @@ static int intel_crtc_mode_set(struct drm_crtc *crtc, drm_vblank_post_modeset(dev, pipe); + intel_crtc->dpms_mode = DRM_MODE_DPMS_ON; + return ret; } From 3bcf603f6d5d18bd9d076dc280de71f48add4101 Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Wed, 27 Jul 2011 11:51:40 -0700 Subject: [PATCH 079/439] drm/i915: apply timing generator bug workaround on CPT and PPT On CougarPoint and PantherPoint PCH chips, the timing generator may fail to start after DP training completes. This is due to a bug in the FDI autotraining detect logic (which will stall the timing generator and re-enable it once training completes), so disable it to avoid silent DP mode setting failures. Signed-off-by: Jesse Barnes Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/i915_reg.h | 5 +++++ drivers/gpu/drm/i915/intel_display.c | 4 ++++ 2 files changed, 9 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index f7315653a842..00bd51070314 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -3072,6 +3072,11 @@ #define TRANS_6BPC (2<<5) #define TRANS_12BPC (3<<5) +#define _TRANSA_CHICKEN2 0xf0064 +#define _TRANSB_CHICKEN2 0xf1064 +#define TRANS_CHICKEN2(pipe) _PIPE(pipe, _TRANSA_CHICKEN2, _TRANSB_CHICKEN2) +#define TRANS_AUTOTRAIN_GEN_STALL_DIS (1<<31) + #define SOUTH_CHICKEN2 0xc2004 #define DPLS_EDP_PPS_FIX_DIS (1<<0) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index d9b8c15998d9..502efc37b074 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -7524,6 +7524,7 @@ static void ibx_init_clock_gating(struct drm_device *dev) static void cpt_init_clock_gating(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; + int pipe; /* * On Ibex Peak and Cougar Point, we need to disable clock @@ -7533,6 +7534,9 @@ static void cpt_init_clock_gating(struct drm_device *dev) I915_WRITE(SOUTH_DSPCLK_GATE_D, PCH_DPLSUNIT_CLOCK_GATE_DISABLE); I915_WRITE(SOUTH_CHICKEN2, I915_READ(SOUTH_CHICKEN2) | DPLS_EDP_PPS_FIX_DIS); + /* Without this, mode sets may fail silently on FDI */ + for_each_pipe(pipe) + I915_WRITE(TRANS_CHICKEN2(pipe), TRANS_AUTOTRAIN_GEN_STALL_DIS); } static void ironlake_teardown_rc6(struct drm_device *dev) From 2704cf5fbd248871a745d210733c6319959d2b0c Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Thu, 28 Jul 2011 11:52:45 -0700 Subject: [PATCH 080/439] drm/i915: flush plane control changes on ILK+ as well After writing to the plane control reg we need to write to the surface reg to trigger the double buffered register latch. On previous chipsets, writing to DSPADDR was enough, but on ILK+ DSPSURF is the reg that triggers the double buffer latch. v2: write DSPADDR too to cover pre-965 chipsets v3: use flush_display_plane instead, that's what it's for v4: send the right patch Signed-off-by: Jesse Barnes Tested-by: Keith Packard Reviewed-by: Keith Packard Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_display.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 502efc37b074..a7a7b672c4c2 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -1323,8 +1323,8 @@ static void intel_enable_plane(struct drm_i915_private *dev_priv, static void intel_flush_display_plane(struct drm_i915_private *dev_priv, enum plane plane) { - u32 reg = DSPADDR(plane); - I915_WRITE(reg, I915_READ(reg)); + I915_WRITE(DSPADDR(plane), I915_READ(DSPADDR(plane))); + I915_WRITE(DSPSURF(plane), I915_READ(DSPSURF(plane))); } /** From d74362c9e45689d8d7e3d4bcf6681c4358ef4f2e Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Thu, 28 Jul 2011 14:47:14 -0700 Subject: [PATCH 081/439] drm/i915: Flush other plane register writes Writes to the plane control register are buffered in the chip until a write to the DSPADDR (pre-965) or DSPSURF (post-965) register occurs. This patch adds flushes in: intel_enable_plane gen6_init_clock_gating ivybridge_init_clock_gating Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_display.c | 31 ++++++++++++++++------------ 1 file changed, 18 insertions(+), 13 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index a7a7b672c4c2..8f7ed7304269 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -1290,6 +1290,17 @@ static void intel_disable_pipe(struct drm_i915_private *dev_priv, intel_wait_for_pipe_off(dev_priv->dev, pipe); } +/* + * Plane regs are double buffered, going from enabled->disabled needs a + * trigger in order to latch. The display address reg provides this. + */ +static void intel_flush_display_plane(struct drm_i915_private *dev_priv, + enum plane plane) +{ + I915_WRITE(DSPADDR(plane), I915_READ(DSPADDR(plane))); + I915_WRITE(DSPSURF(plane), I915_READ(DSPSURF(plane))); +} + /** * intel_enable_plane - enable a display plane on a given pipe * @dev_priv: i915 private structure @@ -1313,20 +1324,10 @@ static void intel_enable_plane(struct drm_i915_private *dev_priv, return; I915_WRITE(reg, val | DISPLAY_PLANE_ENABLE); + intel_flush_display_plane(dev_priv, plane); intel_wait_for_vblank(dev_priv->dev, pipe); } -/* - * Plane regs are double buffered, going from enabled->disabled needs a - * trigger in order to latch. The display address reg provides this. - */ -static void intel_flush_display_plane(struct drm_i915_private *dev_priv, - enum plane plane) -{ - I915_WRITE(DSPADDR(plane), I915_READ(DSPADDR(plane))); - I915_WRITE(DSPSURF(plane), I915_READ(DSPSURF(plane))); -} - /** * intel_disable_plane - disable a display plane * @dev_priv: i915 private structure @@ -7418,10 +7419,12 @@ static void gen6_init_clock_gating(struct drm_device *dev) ILK_DPARB_CLK_GATE | ILK_DPFD_CLK_GATE); - for_each_pipe(pipe) + for_each_pipe(pipe) { I915_WRITE(DSPCNTR(pipe), I915_READ(DSPCNTR(pipe)) | DISPPLANE_TRICKLE_FEED_DISABLE); + intel_flush_display_plane(dev_priv, pipe); + } } static void ivybridge_init_clock_gating(struct drm_device *dev) @@ -7438,10 +7441,12 @@ static void ivybridge_init_clock_gating(struct drm_device *dev) I915_WRITE(ILK_DSPCLK_GATE, IVB_VRHUNIT_CLK_GATE); - for_each_pipe(pipe) + for_each_pipe(pipe) { I915_WRITE(DSPCNTR(pipe), I915_READ(DSPCNTR(pipe)) | DISPPLANE_TRICKLE_FEED_DISABLE); + intel_flush_display_plane(dev_priv, pipe); + } } static void g4x_init_clock_gating(struct drm_device *dev) From cb0e093162d7b6589c2217a00e2abfef686b32d6 Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Thu, 28 Jul 2011 14:50:30 -0700 Subject: [PATCH 082/439] drm/i915: fix CB tuning check for ILK+ CB tuning is needed to handle potential process variations that might cause clock jitter for certain PLL settings. However, we were setting it incorrectly since we were using the wrong M value as a check (M1 when we needed to use the whole M value). Fix it up, making my HDMI attached display a little prettier (used to have occasional dots crawl across the display). Signed-off-by: Jesse Barnes Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_display.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 8f7ed7304269..32c8c956d975 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -4994,7 +4994,7 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, } else if (is_sdvo && is_tv) factor = 20; - if (clock.m1 < factor * clock.n) + if (clock.m < factor * clock.n) fp |= FP_CB_TUNE; dpll = 0; From 4203223a1aed862b4445fdcd260d6139603a51d9 Mon Sep 17 00:00:00 2001 From: Tushar Gohad Date: Thu, 28 Jul 2011 10:36:20 +0000 Subject: [PATCH 083/439] xfrm: Fix key lengths for rfc3686(ctr(aes)) Fix the min and max bit lengths for AES-CTR (RFC3686) keys. The number of bits in key spec is the key length (128/256) plus 32 bits of nonce. This change takes care of the "Invalid key length" errors reported by setkey when specifying 288 bit keys for aes-ctr. Signed-off-by: Tushar Gohad Acked-by: Herbert Xu Signed-off-by: David S. Miller --- net/xfrm/xfrm_algo.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/xfrm/xfrm_algo.c b/net/xfrm/xfrm_algo.c index 58064d9e565d..791ab2e77f3f 100644 --- a/net/xfrm/xfrm_algo.c +++ b/net/xfrm/xfrm_algo.c @@ -462,8 +462,8 @@ static struct xfrm_algo_desc ealg_list[] = { .desc = { .sadb_alg_id = SADB_X_EALG_AESCTR, .sadb_alg_ivlen = 8, - .sadb_alg_minbits = 128, - .sadb_alg_maxbits = 256 + .sadb_alg_minbits = 160, + .sadb_alg_maxbits = 288 } }, }; From fe66101f14813b77d84f6450d51772a2af2b81a1 Mon Sep 17 00:00:00 2001 From: Klement Fish Date: Thu, 28 Jul 2011 06:03:22 +0000 Subject: [PATCH 084/439] sis190: Rx filter init is needed for MAC address change. Addresses https://bugzilla.kernel.org/show_bug.cgi?id=34552 Signed-off-by: Klement Fish Acked-by: Francois Romieu Signed-off-by: David S. Miller --- drivers/net/sis190.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/drivers/net/sis190.c b/drivers/net/sis190.c index 8ad7bfbaa3af..3c0f1312b391 100644 --- a/drivers/net/sis190.c +++ b/drivers/net/sis190.c @@ -1825,6 +1825,16 @@ static int sis190_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) generic_mii_ioctl(&tp->mii_if, if_mii(ifr), cmd, NULL); } +static int sis190_mac_addr(struct net_device *dev, void *p) +{ + int rc; + + rc = eth_mac_addr(dev, p); + if (!rc) + sis190_init_rxfilter(dev); + return rc; +} + static const struct net_device_ops sis190_netdev_ops = { .ndo_open = sis190_open, .ndo_stop = sis190_close, @@ -1833,7 +1843,7 @@ static const struct net_device_ops sis190_netdev_ops = { .ndo_tx_timeout = sis190_tx_timeout, .ndo_set_multicast_list = sis190_set_rx_mode, .ndo_change_mtu = eth_change_mtu, - .ndo_set_mac_address = eth_mac_addr, + .ndo_set_mac_address = sis190_mac_addr, .ndo_validate_addr = eth_validate_addr, #ifdef CONFIG_NET_POLL_CONTROLLER .ndo_poll_controller = sis190_netpoll, From c173bfac2428bd6c5da5d7cc8942b8b85d0819d4 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Thu, 28 Jul 2011 10:54:23 +0000 Subject: [PATCH 085/439] MAINTAINERS: orphan FrameRelay DLCI Mike McLagan hasn't contributed in many years and his email bounces. Signed-off-by: Joe Perches Cc: David Miller Cc: Stephen Hemminger Signed-off-by: Andrew Morton Signed-off-by: David S. Miller --- MAINTAINERS | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 1d2e79db0f58..a99fed1a3d87 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2637,9 +2637,8 @@ S: Maintained F: arch/x86/math-emu/ FRAME RELAY DLCI/FRAD (Sangoma drivers too) -M: Mike McLagan L: netdev@vger.kernel.org -S: Maintained +S: Orphan F: drivers/net/wan/dlci.c F: drivers/net/wan/sdla.c From 93a3aa25933461d76141179fc94aa32d5f9d954a Mon Sep 17 00:00:00 2001 From: Lennart Sorensen Date: Thu, 28 Jul 2011 13:18:11 +0000 Subject: [PATCH 086/439] r8169: Add support for D-Link 530T rev C1 (Kernel Bug 38862) The D-Link DGE-530T rev C1 is a re-badged Realtek 8169 named DLG10028C, unlike the previous revisions which were skge based. It is probably the same as the discontinued DGE-528T (0x4300) other than the PCI ID. The PCI ID is 0x1186:0x4302. Adding it to r8169.c where 0x1186:0x4300 is already found makes the card be detected and work. This fixes https://bugzilla.kernel.org/show_bug.cgi?id=38862 Signed-off-by: Len Sorensen Signed-off-by: David S. Miller --- drivers/net/r8169.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c index 7d9c650f395e..c77286edba4d 100644 --- a/drivers/net/r8169.c +++ b/drivers/net/r8169.c @@ -239,6 +239,7 @@ static DEFINE_PCI_DEVICE_TABLE(rtl8169_pci_tbl) = { { PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8168), 0, 0, RTL_CFG_1 }, { PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8169), 0, 0, RTL_CFG_0 }, { PCI_DEVICE(PCI_VENDOR_ID_DLINK, 0x4300), 0, 0, RTL_CFG_0 }, + { PCI_DEVICE(PCI_VENDOR_ID_DLINK, 0x4302), 0, 0, RTL_CFG_0 }, { PCI_DEVICE(PCI_VENDOR_ID_AT, 0xc107), 0, 0, RTL_CFG_0 }, { PCI_DEVICE(0x16ec, 0x0116), 0, 0, RTL_CFG_0 }, { PCI_VENDOR_ID_LINKSYS, 0x1032, From 177c27bf05d0ea508e65afdbe4b6998c81e46af5 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 28 Jul 2011 06:54:36 +0000 Subject: [PATCH 087/439] net: fix new sunrpc kernel-doc warning Fix new kernel-doc warning in sunrpc: Warning(net/sunrpc/xprt.c:196): No description found for parameter 'xprt' Signed-off-by: Randy Dunlap Signed-off-by: David S. Miller --- net/sunrpc/xprt.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index 9b6a4d1ea8f8..f4385e45a5fc 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -187,6 +187,7 @@ EXPORT_SYMBOL_GPL(xprt_load_transport); /** * xprt_reserve_xprt - serialize write access to transports * @task: task that is requesting access to the transport + * @xprt: pointer to the target transport * * This prevents mixing the payload of separate requests, and prevents * transport connects from colliding with writes. No congestion control From 9e8f90dfe58eb19140bc66655170c7aef9eadbf2 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Thu, 28 Jul 2011 18:26:32 -0700 Subject: [PATCH 088/439] proc_fork_connector: a lockless ->real_parent usage is not safe proc_fork_connector() uses ->real_parent lockless. This is not safe if copy_process() was called with CLONE_THREAD or CLONE_PARENT, in this case the parent != current can go away at any moment. Signed-off-by: Oleg Nesterov Cc: Vladimir Zapolskiy Cc: "David S. Miller" Cc: Evgeniy Polyakov Cc: Evgeniy Polyakov Signed-off-by: Andrew Morton Signed-off-by: David S. Miller --- drivers/connector/cn_proc.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/connector/cn_proc.c b/drivers/connector/cn_proc.c index 3ee1fdb31ea7..e55814bc0d06 100644 --- a/drivers/connector/cn_proc.c +++ b/drivers/connector/cn_proc.c @@ -57,6 +57,7 @@ void proc_fork_connector(struct task_struct *task) struct proc_event *ev; __u8 buffer[CN_PROC_MSG_SIZE]; struct timespec ts; + struct task_struct *parent; if (atomic_read(&proc_event_num_listeners) < 1) return; @@ -67,8 +68,11 @@ void proc_fork_connector(struct task_struct *task) ktime_get_ts(&ts); /* get high res monotonic timestamp */ put_unaligned(timespec_to_ns(&ts), (__u64 *)&ev->timestamp_ns); ev->what = PROC_EVENT_FORK; - ev->event_data.fork.parent_pid = task->real_parent->pid; - ev->event_data.fork.parent_tgid = task->real_parent->tgid; + rcu_read_lock(); + parent = rcu_dereference(task->real_parent); + ev->event_data.fork.parent_pid = parent->pid; + ev->event_data.fork.parent_tgid = parent->tgid; + rcu_read_unlock(); ev->event_data.fork.child_pid = task->pid; ev->event_data.fork.child_tgid = task->tgid; From 700d1ef33ff1d9a582b4a1dc23a130049f239942 Mon Sep 17 00:00:00 2001 From: Adrian Knoth Date: Fri, 29 Jul 2011 03:11:02 +0200 Subject: [PATCH 089/439] ALSA: hdspm - Provide MADI speed mode selector on RME MADI and MADIface When running in slave mode (no clock master), there is no way to determine the real wirespeed on the MADI link (single/double/quad speed). Like physical gear, simply provide the user with a tristate switch to select the appropriate format. Signed-off-by: Adrian Knoth Signed-off-by: Takashi Iwai --- sound/pci/rme9652/hdspm.c | 91 ++++++++++++++++++++++++++++++++++++++- 1 file changed, 89 insertions(+), 2 deletions(-) diff --git a/sound/pci/rme9652/hdspm.c b/sound/pci/rme9652/hdspm.c index af130ee0c45d..d21964994168 100644 --- a/sound/pci/rme9652/hdspm.c +++ b/sound/pci/rme9652/hdspm.c @@ -3415,6 +3415,91 @@ static int snd_hdspm_put_qs_wire(struct snd_kcontrol *kcontrol, return change; } +#define HDSPM_MADI_SPEEDMODE(xname, xindex) \ +{ .iface = SNDRV_CTL_ELEM_IFACE_MIXER, \ + .name = xname, \ + .index = xindex, \ + .info = snd_hdspm_info_madi_speedmode, \ + .get = snd_hdspm_get_madi_speedmode, \ + .put = snd_hdspm_put_madi_speedmode \ +} + +static int hdspm_madi_speedmode(struct hdspm *hdspm) +{ + if (hdspm->control_register & HDSPM_QuadSpeed) + return 2; + if (hdspm->control_register & HDSPM_DoubleSpeed) + return 1; + return 0; +} + +static int hdspm_set_madi_speedmode(struct hdspm *hdspm, int mode) +{ + hdspm->control_register &= ~(HDSPM_DoubleSpeed | HDSPM_QuadSpeed); + switch (mode) { + case 0: + break; + case 1: + hdspm->control_register |= HDSPM_DoubleSpeed; + break; + case 2: + hdspm->control_register |= HDSPM_QuadSpeed; + break; + } + hdspm_write(hdspm, HDSPM_controlRegister, hdspm->control_register); + + return 0; +} + +static int snd_hdspm_info_madi_speedmode(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_info *uinfo) +{ + static char *texts[] = { "Single", "Double", "Quad" }; + + uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED; + uinfo->count = 1; + uinfo->value.enumerated.items = 3; + + if (uinfo->value.enumerated.item >= uinfo->value.enumerated.items) + uinfo->value.enumerated.item = + uinfo->value.enumerated.items - 1; + strcpy(uinfo->value.enumerated.name, + texts[uinfo->value.enumerated.item]); + + return 0; +} + +static int snd_hdspm_get_madi_speedmode(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct hdspm *hdspm = snd_kcontrol_chip(kcontrol); + + spin_lock_irq(&hdspm->lock); + ucontrol->value.enumerated.item[0] = hdspm_madi_speedmode(hdspm); + spin_unlock_irq(&hdspm->lock); + return 0; +} + +static int snd_hdspm_put_madi_speedmode(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct hdspm *hdspm = snd_kcontrol_chip(kcontrol); + int change; + int val; + + if (!snd_hdspm_use_is_exclusive(hdspm)) + return -EBUSY; + val = ucontrol->value.integer.value[0]; + if (val < 0) + val = 0; + if (val > 2) + val = 2; + spin_lock_irq(&hdspm->lock); + change = val != hdspm_madi_speedmode(hdspm); + hdspm_set_madi_speedmode(hdspm, val); + spin_unlock_irq(&hdspm->lock); + return change; +} #define HDSPM_MIXER(xname, xindex) \ { .iface = SNDRV_CTL_ELEM_IFACE_HWDEP, \ @@ -4289,7 +4374,8 @@ static struct snd_kcontrol_new snd_hdspm_controls_madi[] = { HDSPM_TX_64("TX 64 channels mode", 0), HDSPM_C_TMS("Clear Track Marker", 0), HDSPM_SAFE_MODE("Safe Mode", 0), - HDSPM_INPUT_SELECT("Input Select", 0) + HDSPM_INPUT_SELECT("Input Select", 0), + HDSPM_MADI_SPEEDMODE("MADI Speed Mode", 0) }; @@ -4302,7 +4388,8 @@ static struct snd_kcontrol_new snd_hdspm_controls_madiface[] = { HDSPM_SYNC_CHECK("MADI SyncCheck", 0), HDSPM_TX_64("TX 64 channels mode", 0), HDSPM_C_TMS("Clear Track Marker", 0), - HDSPM_SAFE_MODE("Safe Mode", 0) + HDSPM_SAFE_MODE("Safe Mode", 0), + HDSPM_MADI_SPEEDMODE("MADI Speed Mode", 0) }; static struct snd_kcontrol_new snd_hdspm_controls_aio[] = { From d12c51d8299667464e31d545acc4ebb7031d024c Mon Sep 17 00:00:00 2001 From: Adrian Knoth Date: Fri, 29 Jul 2011 03:11:03 +0200 Subject: [PATCH 090/439] ALSA: hdspm - Fix reported external sample rate on RME MADI and MADIface In slave mode, the card can only detect the base frequency (32..48kHz) on the MADI link (exception: 96k frames), so the real external sample rate is this base frequency multiplied by 1, 2 or 4 depending on the speed mode. This patch enables 64..192kHz sample rates in clock slave mode, which failed before due to an alleged sample rate mismatch between the MADI link (e.g., 48kHz) and the application in DS/QS mode (e.g., 96kHz, 192kHz). Signed-off-by: Adrian Knoth Signed-off-by: Takashi Iwai --- sound/pci/rme9652/hdspm.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/sound/pci/rme9652/hdspm.c b/sound/pci/rme9652/hdspm.c index d21964994168..88ae27431cb2 100644 --- a/sound/pci/rme9652/hdspm.c +++ b/sound/pci/rme9652/hdspm.c @@ -1217,6 +1217,22 @@ static int hdspm_external_sample_rate(struct hdspm *hdspm) rate = 0; break; } + + /* QS and DS rates normally can not be detected + * automatically by the card. Only exception is MADI + * in 96k frame mode. + * + * So if we read SS values (32 .. 48k), check for + * user-provided DS/QS bits in the control register + * and multiply the base frequency accordingly. + */ + if (rate <= 48000) { + if (hdspm->control_register & HDSPM_QuadSpeed) + rate *= 4; + else if (hdspm->control_register & + HDSPM_DoubleSpeed) + rate *= 2; + } } break; } From 5f8b4d53d7efe00ecb7d96c8cc0b4a44e130f174 Mon Sep 17 00:00:00 2001 From: Adrian Knoth Date: Fri, 29 Jul 2011 03:11:04 +0200 Subject: [PATCH 091/439] ALSA: hdspm - Add firmware revision 0xcc for RME MADI Apparently, there are multiple old firmware revisions in the wild for the PCI RME MADI cards. Just add them to the list of supported devices and treat them like their modern counterparts. Signed-off-by: Adrian Knoth Signed-off-by: Takashi Iwai --- sound/pci/rme9652/hdspm.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/sound/pci/rme9652/hdspm.c b/sound/pci/rme9652/hdspm.c index 88ae27431cb2..6edc67ced905 100644 --- a/sound/pci/rme9652/hdspm.c +++ b/sound/pci/rme9652/hdspm.c @@ -521,6 +521,7 @@ MODULE_SUPPORTED_DEVICE("{{RME HDSPM-MADI}}"); #define HDSPM_DMA_AREA_KILOBYTES (HDSPM_DMA_AREA_BYTES/1024) /* revisions >= 230 indicate AES32 card */ +#define HDSPM_MADI_ANCIENT_REV 204 #define HDSPM_MADI_OLD_REV 207 #define HDSPM_MADI_REV 210 #define HDSPM_RAYDAT_REV 211 @@ -6484,6 +6485,7 @@ static int __devinit snd_hdspm_create(struct snd_card *card, switch (hdspm->firmware_rev) { case HDSPM_MADI_REV: case HDSPM_MADI_OLD_REV: + case HDSPM_MADI_ANCIENT_REV: hdspm->io_type = MADI; hdspm->card_name = "RME MADI"; hdspm->midiPorts = 3; From 06132fdf63c7d9acc19f136623bf2201f49e73b5 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 28 Jul 2011 12:26:16 +0100 Subject: [PATCH 092/439] ASoC: Fix txx9aclc.c build 552d1ef6b5a98d7b95959d5b139071e3c90cebf1 [ASoC: core - Optimise and refactor pcm_new() to pass only rtd] breaks compilation of txx9aclc.c: CC [M] sound/soc/txx9/txx9aclc.o /home/ralf/src/linux/linux-mips/sound/soc/txx9/txx9aclc.c: In function 'txx9aclc_pcm_new': /home/ralf/src/linux/linux-mips/sound/soc/txx9/txx9aclc.c:318:3: error: 'card' undeclared (first use in this function) /home/ralf/src/linux/linux-mips/sound/soc/txx9/txx9aclc.c:318:3: note: each undeclared identifier is reported only once for each function it appears in make[5]: *** [sound/soc/txx9/txx9aclc.o] Error 1 Fixed by providing a definition for card. Signed-off-by: Ralf Baechle Signed-off-by: Takashi Iwai --- sound/soc/txx9/txx9aclc.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/soc/txx9/txx9aclc.c b/sound/soc/txx9/txx9aclc.c index 34aa972669ed..3de99af8cb82 100644 --- a/sound/soc/txx9/txx9aclc.c +++ b/sound/soc/txx9/txx9aclc.c @@ -290,6 +290,7 @@ static void txx9aclc_pcm_free_dma_buffers(struct snd_pcm *pcm) static int txx9aclc_pcm_new(struct snd_soc_pcm_runtime *rtd) { + struct snd_card *card = rtd->card->snd_card; struct snd_soc_dai *dai = rtd->cpu_dai; struct snd_pcm *pcm = rtd->pcm; struct platform_device *pdev = to_platform_device(dai->platform->dev); From ac85fe8b21248054851e05bfaa352562e5b06dd3 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Thu, 28 Jul 2011 23:31:26 -0700 Subject: [PATCH 093/439] sparc: Sanitize cpu feature detection and reporting. Instead of evaluating the cpu features for ELF_HWCAP every exec, calculate it once at boot time. Add AV_SPARC_* capability flag bits, compatible with what Solaris reports to applications. Report these capabilities once in the kernel log, and also via /proc/cpuinfo in a new "cpucaps" entry. If available, fetch the cpu features from the machine description 'hwcap-list' property of the 'cpu' node. Signed-off-by: David S. Miller --- arch/sparc/include/asm/elf_64.h | 63 ++++++-------- arch/sparc/kernel/cpu.c | 1 + arch/sparc/kernel/kernel.h | 6 ++ arch/sparc/kernel/setup_64.c | 149 ++++++++++++++++++++++++++++++++ 4 files changed, 184 insertions(+), 35 deletions(-) diff --git a/arch/sparc/include/asm/elf_64.h b/arch/sparc/include/asm/elf_64.h index 64f7a00b3747..7df8b7f544d4 100644 --- a/arch/sparc/include/asm/elf_64.h +++ b/arch/sparc/include/asm/elf_64.h @@ -59,15 +59,33 @@ #define R_SPARC_6 45 /* Bits present in AT_HWCAP, primarily for Sparc32. */ +#define HWCAP_SPARC_FLUSH 0x00000001 +#define HWCAP_SPARC_STBAR 0x00000002 +#define HWCAP_SPARC_SWAP 0x00000004 +#define HWCAP_SPARC_MULDIV 0x00000008 +#define HWCAP_SPARC_V9 0x00000010 +#define HWCAP_SPARC_ULTRA3 0x00000020 +#define HWCAP_SPARC_BLKINIT 0x00000040 +#define HWCAP_SPARC_N2 0x00000080 -#define HWCAP_SPARC_FLUSH 1 /* CPU supports flush instruction. */ -#define HWCAP_SPARC_STBAR 2 -#define HWCAP_SPARC_SWAP 4 -#define HWCAP_SPARC_MULDIV 8 -#define HWCAP_SPARC_V9 16 -#define HWCAP_SPARC_ULTRA3 32 -#define HWCAP_SPARC_BLKINIT 64 -#define HWCAP_SPARC_N2 128 +/* Solaris compatible AT_HWCAP bits. */ +#define AV_SPARC_MUL32 0x00000100 /* 32x32 multiply is efficient */ +#define AV_SPARC_DIV32 0x00000200 /* 32x32 divide is efficient */ +#define AV_SPARC_FSMULD 0x00000400 /* 'fsmuld' is efficient */ +#define AV_SPARC_V8PLUS 0x00000800 /* v9 insn available to 32bit */ +#define AV_SPARC_POPC 0x00001000 /* 'popc' is efficient */ +#define AV_SPARC_VIS 0x00002000 /* VIS insns available */ +#define AV_SPARC_VIS2 0x00004000 /* VIS2 insns available */ +#define AV_SPARC_ASI_BLK_INIT 0x00008000 /* block init ASIs available */ +#define AV_SPARC_FMAF 0x00010000 /* fused multiply-add */ +#define AV_SPARC_VIS3 0x00020000 /* VIS3 insns available */ +#define AV_SPARC_HPC 0x00040000 /* HPC insns available */ +#define AV_SPARC_RANDOM 0x00080000 /* 'random' insn available */ +#define AV_SPARC_TRANS 0x00100000 /* transaction insns available */ +#define AV_SPARC_FJFMAU 0x00200000 /* unfused multiply-add */ +#define AV_SPARC_IMA 0x00400000 /* integer multiply-add */ +#define AV_SPARC_ASI_CACHE_SPARING \ + 0x00800000 /* cache sparing ASIs available */ #define CORE_DUMP_USE_REGSET @@ -162,33 +180,8 @@ typedef struct { #define ELF_ET_DYN_BASE 0x0000010000000000UL #define COMPAT_ELF_ET_DYN_BASE 0x0000000070000000UL - -/* This yields a mask that user programs can use to figure out what - instruction set this cpu supports. */ - -/* On Ultra, we support all of the v8 capabilities. */ -static inline unsigned int sparc64_elf_hwcap(void) -{ - unsigned int cap = (HWCAP_SPARC_FLUSH | HWCAP_SPARC_STBAR | - HWCAP_SPARC_SWAP | HWCAP_SPARC_MULDIV | - HWCAP_SPARC_V9); - - if (tlb_type == cheetah || tlb_type == cheetah_plus) - cap |= HWCAP_SPARC_ULTRA3; - else if (tlb_type == hypervisor) { - if (sun4v_chip_type == SUN4V_CHIP_NIAGARA1 || - sun4v_chip_type == SUN4V_CHIP_NIAGARA2 || - sun4v_chip_type == SUN4V_CHIP_NIAGARA3) - cap |= HWCAP_SPARC_BLKINIT; - if (sun4v_chip_type == SUN4V_CHIP_NIAGARA2 || - sun4v_chip_type == SUN4V_CHIP_NIAGARA3) - cap |= HWCAP_SPARC_N2; - } - - return cap; -} - -#define ELF_HWCAP sparc64_elf_hwcap() +extern unsigned long sparc64_elf_hwcap; +#define ELF_HWCAP sparc64_elf_hwcap /* This yields a string that ld.so will use to load implementation specific libraries for optimization. This is more specific in diff --git a/arch/sparc/kernel/cpu.c b/arch/sparc/kernel/cpu.c index 17cf290dc2bc..9810fd881058 100644 --- a/arch/sparc/kernel/cpu.c +++ b/arch/sparc/kernel/cpu.c @@ -396,6 +396,7 @@ static int show_cpuinfo(struct seq_file *m, void *__unused) , cpu_data(0).clock_tick #endif ); + cpucap_info(m); #ifdef CONFIG_SMP smp_bogo(m); #endif diff --git a/arch/sparc/kernel/kernel.h b/arch/sparc/kernel/kernel.h index 6f6544cfa0ef..8325d7759381 100644 --- a/arch/sparc/kernel/kernel.h +++ b/arch/sparc/kernel/kernel.h @@ -10,6 +10,12 @@ extern const char *sparc_pmu_type; extern unsigned int fsr_storage; extern int ncpus_probed; +#ifdef CONFIG_SPARC64 +/* setup_64.c */ +struct seq_file; +extern void cpucap_info(struct seq_file *); +#endif + #ifdef CONFIG_SPARC32 /* cpu.c */ extern void cpu_probe(void); diff --git a/arch/sparc/kernel/setup_64.c b/arch/sparc/kernel/setup_64.c index c4dd0999da86..242dbb3f31d2 100644 --- a/arch/sparc/kernel/setup_64.c +++ b/arch/sparc/kernel/setup_64.c @@ -29,6 +29,7 @@ #include #include #include +#include #include #include @@ -46,6 +47,8 @@ #include #include #include +#include +#include #ifdef CONFIG_IP_PNP #include @@ -278,6 +281,151 @@ void __init boot_cpu_id_too_large(int cpu) } #endif +/* On Ultra, we support all of the v8 capabilities. */ +unsigned long sparc64_elf_hwcap = (HWCAP_SPARC_FLUSH | HWCAP_SPARC_STBAR | + HWCAP_SPARC_SWAP | HWCAP_SPARC_MULDIV | + HWCAP_SPARC_V9); +EXPORT_SYMBOL(sparc64_elf_hwcap); + +static const char *hwcaps[] = { + "flush", "stbar", "swap", "muldiv", "v9", + "ultra3", "blkinit", "n2", + + /* These strings are as they appear in the machine description + * 'hwcap-list' property for cpu nodes. + */ + "mul32", "div32", "fsmuld", "v8plus", "popc", "vis", "vis2", + "ASIBlkInit", "fmaf", "vis3", "hpc", "random", "trans", "fjfmau", + "ima", "cspare", +}; + +void cpucap_info(struct seq_file *m) +{ + unsigned long caps = sparc64_elf_hwcap; + int i, printed = 0; + + seq_puts(m, "cpucaps\t\t: "); + for (i = 0; i < ARRAY_SIZE(hwcaps); i++) { + unsigned long bit = 1UL << i; + if (caps & bit) { + seq_printf(m, "%s%s", + printed ? "," : "", hwcaps[i]); + printed++; + } + } + seq_putc(m, '\n'); +} + +static void __init report_hwcaps(unsigned long caps) +{ + int i, printed = 0; + + printk(KERN_INFO "CPU CAPS: ["); + for (i = 0; i < ARRAY_SIZE(hwcaps); i++) { + unsigned long bit = 1UL << i; + if (caps & bit) { + printk(KERN_CONT "%s%s", + printed ? "," : "", hwcaps[i]); + if (++printed == 8) { + printk(KERN_CONT "]\n"); + printk(KERN_INFO "CPU CAPS: ["); + printed = 0; + } + } + } + printk(KERN_CONT "]\n"); +} + +static unsigned long __init mdesc_cpu_hwcap_list(void) +{ + struct mdesc_handle *hp; + unsigned long caps = 0; + const char *prop; + int len; + u64 pn; + + hp = mdesc_grab(); + if (!hp) + return 0; + + pn = mdesc_node_by_name(hp, MDESC_NODE_NULL, "cpu"); + if (pn == MDESC_NODE_NULL) + goto out; + + prop = mdesc_get_property(hp, pn, "hwcap-list", &len); + if (!prop) + goto out; + + while (len) { + int i, plen; + + for (i = 0; i < ARRAY_SIZE(hwcaps); i++) { + unsigned long bit = 1UL << i; + + if (!strcmp(prop, hwcaps[i])) { + caps |= bit; + break; + } + } + + plen = strlen(prop) + 1; + prop += plen; + len -= plen; + } + +out: + mdesc_release(hp); + return caps; +} + +/* This yields a mask that user programs can use to figure out what + * instruction set this cpu supports. + */ +static void __init init_sparc64_elf_hwcap(void) +{ + unsigned long cap = sparc64_elf_hwcap; + unsigned long mdesc_caps; + + if (tlb_type == cheetah || tlb_type == cheetah_plus) + cap |= HWCAP_SPARC_ULTRA3; + else if (tlb_type == hypervisor) { + if (sun4v_chip_type == SUN4V_CHIP_NIAGARA1 || + sun4v_chip_type == SUN4V_CHIP_NIAGARA2 || + sun4v_chip_type == SUN4V_CHIP_NIAGARA3) + cap |= HWCAP_SPARC_BLKINIT; + if (sun4v_chip_type == SUN4V_CHIP_NIAGARA2 || + sun4v_chip_type == SUN4V_CHIP_NIAGARA3) + cap |= HWCAP_SPARC_N2; + } + + cap |= (AV_SPARC_MUL32 | AV_SPARC_DIV32 | AV_SPARC_V8PLUS); + + mdesc_caps = mdesc_cpu_hwcap_list(); + if (!mdesc_caps) { + if (tlb_type == spitfire) + cap |= AV_SPARC_VIS; + if (tlb_type == cheetah || tlb_type == cheetah_plus) + cap |= AV_SPARC_VIS | AV_SPARC_VIS2; + if (tlb_type == cheetah_plus) + cap |= AV_SPARC_POPC; + if (tlb_type == hypervisor) { + if (sun4v_chip_type == SUN4V_CHIP_NIAGARA1) + cap |= AV_SPARC_ASI_BLK_INIT; + if (sun4v_chip_type == SUN4V_CHIP_NIAGARA2 || + sun4v_chip_type == SUN4V_CHIP_NIAGARA3) + cap |= (AV_SPARC_VIS | AV_SPARC_VIS2 | + AV_SPARC_ASI_BLK_INIT | + AV_SPARC_POPC); + if (sun4v_chip_type == SUN4V_CHIP_NIAGARA3) + cap |= (AV_SPARC_VIS3 | AV_SPARC_HPC | + AV_SPARC_FMAF); + } + } + sparc64_elf_hwcap = cap | mdesc_caps; + + report_hwcaps(sparc64_elf_hwcap); +} + void __init setup_arch(char **cmdline_p) { /* Initialize PROM console and command line. */ @@ -337,6 +485,7 @@ void __init setup_arch(char **cmdline_p) init_cur_cpu_trap(current_thread_info()); paging_init(); + init_sparc64_elf_hwcap(); } extern int stop_a_enabled; From 548c210fbffdb008a80fa41ff0cb3965f185583d Mon Sep 17 00:00:00 2001 From: John David Anglin Date: Sat, 11 Jun 2011 14:42:06 -0400 Subject: [PATCH 094/439] [PARISC] fix return type of __atomic64_add_return The return type of __atomic64_add_return of should be s64 or long, not int. This fixes the atomic64 test failure that I previously reported. Signed-off-by: John David Anglin Cc: stable@kernel.org Signed-off-by: James Bottomley --- arch/parisc/include/asm/atomic.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/parisc/include/asm/atomic.h b/arch/parisc/include/asm/atomic.h index f81955934aeb..26fd1146dda6 100644 --- a/arch/parisc/include/asm/atomic.h +++ b/arch/parisc/include/asm/atomic.h @@ -259,10 +259,10 @@ static __inline__ int atomic_add_unless(atomic_t *v, int a, int u) #define ATOMIC64_INIT(i) ((atomic64_t) { (i) }) -static __inline__ int +static __inline__ s64 __atomic64_add_return(s64 i, atomic64_t *v) { - int ret; + s64 ret; unsigned long flags; _atomic_spin_lock_irqsave(v, flags); From b055c8f3ef9f7bc6ba415d900f298d7801a9d1d4 Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Fri, 8 Jul 2011 11:31:57 -0700 Subject: [PATCH 095/439] drm/i915/hdmi: send AVI info frames on ILK+ as well On Ironlake and above, we have per-transcoder DIP registers, so use them for sending DIPs like AVI infoframes on ILK and above. Signed-off-by: Jesse Barnes Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/i915_reg.h | 14 +++++++++++ drivers/gpu/drm/i915/intel_hdmi.c | 39 ++++++++++++++++++++++++++++++- 2 files changed, 52 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 00bd51070314..30d8aae9ac4c 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -3020,6 +3020,20 @@ #define _TRANSA_DP_LINK_M2 0xe0048 #define _TRANSA_DP_LINK_N2 0xe004c +/* Per-transcoder DIP controls */ + +#define _VIDEO_DIP_CTL_A 0xe0200 +#define _VIDEO_DIP_DATA_A 0xe0208 +#define _VIDEO_DIP_GCP_A 0xe0210 + +#define _VIDEO_DIP_CTL_B 0xe1200 +#define _VIDEO_DIP_DATA_B 0xe1208 +#define _VIDEO_DIP_GCP_B 0xe1210 + +#define TVIDEO_DIP_CTL(pipe) _PIPE(pipe, _VIDEO_DIP_CTL_A, _VIDEO_DIP_CTL_B) +#define TVIDEO_DIP_DATA(pipe) _PIPE(pipe, _VIDEO_DIP_DATA_A, _VIDEO_DIP_DATA_B) +#define TVIDEO_DIP_GCP(pipe) _PIPE(pipe, _VIDEO_DIP_GCP_A, _VIDEO_DIP_GCP_B) + #define _TRANS_HTOTAL_B 0xe1000 #define _TRANS_HBLANK_B 0xe1004 #define _TRANS_HSYNC_B 0xe1008 diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c index aa0a8e83142e..c220255ef1d9 100644 --- a/drivers/gpu/drm/i915/intel_hdmi.c +++ b/drivers/gpu/drm/i915/intel_hdmi.c @@ -112,6 +112,40 @@ static void intel_hdmi_set_avi_infoframe(struct drm_encoder *encoder) VIDEO_DIP_ENABLE_AVI); } +static void intel_ironlake_hdmi_set_avi_infoframe(struct drm_encoder *encoder) +{ + struct dip_infoframe avi_if = { + .type = DIP_TYPE_AVI, + .ver = DIP_VERSION_AVI, + .len = DIP_LEN_AVI, + }; + uint32_t *data = (uint32_t *)&avi_if; + struct drm_device *dev = encoder->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); + struct drm_crtc *crtc = encoder->crtc; + struct intel_crtc *intel_crtc = to_intel_crtc(crtc); + int reg = TVIDEO_DIP_CTL(intel_crtc->pipe); + unsigned i; + + if (!intel_hdmi->has_hdmi_sink) + return; + + intel_wait_for_vblank(dev, intel_crtc->pipe); + + I915_WRITE(reg, VIDEO_DIP_SELECT_AVI); + + intel_dip_infoframe_csum(&avi_if); + for (i = 0; i < sizeof(avi_if); i += 4) { + I915_WRITE(TVIDEO_DIP_DATA(intel_crtc->pipe), *data); + data++; + } + + I915_WRITE(reg, VIDEO_DIP_ENABLE | VIDEO_DIP_SELECT_AVI | + VIDEO_DIP_FREQ_VSYNC | (DIP_LEN_AVI << 8) | + VIDEO_DIP_ENABLE_AVI); +} + static void intel_hdmi_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) @@ -149,7 +183,10 @@ static void intel_hdmi_mode_set(struct drm_encoder *encoder, I915_WRITE(intel_hdmi->sdvox_reg, sdvox); POSTING_READ(intel_hdmi->sdvox_reg); - intel_hdmi_set_avi_infoframe(encoder); + if (HAS_PCH_SPLIT(dev)) + intel_ironlake_hdmi_set_avi_infoframe(encoder); + else + intel_hdmi_set_avi_infoframe(encoder); } static void intel_hdmi_dpms(struct drm_encoder *encoder, int mode) From e0e3fb482105c65ce6f5480a86092e966a29ed79 Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Fri, 29 Jul 2011 14:45:21 -0700 Subject: [PATCH 096/439] drm/i915: Ignore GPU wedged errors while pinning scanout buffers Failing to pin a scanout buffer will most likely lead to a black screen, so if the GPU is wedged, then just let the pin happen and hope that things work out OK. Signed-off-by: Keith Packard Reviewed-by: Chris Wilson --- drivers/gpu/drm/i915/i915_gem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index a087e1bf0c2f..d5c7c7bccef5 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -3019,7 +3019,7 @@ i915_gem_object_set_to_display_plane(struct drm_i915_gem_object *obj, /* Currently, we are always called from an non-interruptible context. */ if (pipelined != obj->ring) { ret = i915_gem_object_wait_rendering(obj); - if (ret) + if (ret == -ERESTARTSYS) return ret; } From b066254fee2b0b4d1323295f8ae34c9442222165 Mon Sep 17 00:00:00 2001 From: Pieterjan Camerlynck Date: Tue, 26 Jul 2011 16:23:54 +0200 Subject: [PATCH 097/439] i915: add Dell OptiPlex FX170 to intel_no_lvds The Dell OptiPlex FX170 claims to have LVDS, but doesn't. Signed-off-by: Pieterjan Camerlynck Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_lvds.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c index b28f7bd9f88a..2e8ddfcba40c 100644 --- a/drivers/gpu/drm/i915/intel_lvds.c +++ b/drivers/gpu/drm/i915/intel_lvds.c @@ -688,6 +688,14 @@ static const struct dmi_system_id intel_no_lvds[] = { DMI_MATCH(DMI_PRODUCT_NAME, "Studio Hybrid 140g"), }, }, + { + .callback = intel_no_lvds_dmi_callback, + .ident = "Dell OptiPlex FX170", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex FX170"), + }, + }, { .callback = intel_no_lvds_dmi_callback, .ident = "AOpen Mini PC", From 358733e9047cafcc185ca19b8c369c659ac0c4cf Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Wed, 27 Jul 2011 11:53:01 -0700 Subject: [PATCH 098/439] drm/i915: add GPU max frequency control file Mainly for use in debugging and benchmarking, this file allows the user to control the max frequency used by the GPU. Frequency may still vary based on workload (if the frequency is set to higher than the minimum) but won't go over the newly set value. Signed-off-by: Jesse Barnes Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/i915_debugfs.c | 90 +++++++++++++++++++++++++++++ 1 file changed, 90 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index 0a893f7400fa..782b781df9e0 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -1300,6 +1300,76 @@ static const struct file_operations i915_wedged_fops = { .llseek = default_llseek, }; +static int +i915_max_freq_open(struct inode *inode, + struct file *filp) +{ + filp->private_data = inode->i_private; + return 0; +} + +static ssize_t +i915_max_freq_read(struct file *filp, + char __user *ubuf, + size_t max, + loff_t *ppos) +{ + struct drm_device *dev = filp->private_data; + drm_i915_private_t *dev_priv = dev->dev_private; + char buf[80]; + int len; + + len = snprintf(buf, sizeof (buf), + "max freq: %d\n", dev_priv->max_delay * 50); + + if (len > sizeof (buf)) + len = sizeof (buf); + + return simple_read_from_buffer(ubuf, max, ppos, buf, len); +} + +static ssize_t +i915_max_freq_write(struct file *filp, + const char __user *ubuf, + size_t cnt, + loff_t *ppos) +{ + struct drm_device *dev = filp->private_data; + struct drm_i915_private *dev_priv = dev->dev_private; + char buf[20]; + int val = 1; + + if (cnt > 0) { + if (cnt > sizeof (buf) - 1) + return -EINVAL; + + if (copy_from_user(buf, ubuf, cnt)) + return -EFAULT; + buf[cnt] = 0; + + val = simple_strtoul(buf, NULL, 0); + } + + DRM_DEBUG_DRIVER("Manually setting max freq to %d\n", val); + + /* + * Turbo will still be enabled, but won't go above the set value. + */ + dev_priv->max_delay = val / 50; + + gen6_set_rps(dev, val / 50); + + return cnt; +} + +static const struct file_operations i915_max_freq_fops = { + .owner = THIS_MODULE, + .open = i915_max_freq_open, + .read = i915_max_freq_read, + .write = i915_max_freq_write, + .llseek = default_llseek, +}; + /* As the drm_debugfs_init() routines are called before dev->dev_private is * allocated we need to hook into the minor for release. */ static int @@ -1399,6 +1469,21 @@ static int i915_forcewake_create(struct dentry *root, struct drm_minor *minor) return drm_add_fake_info_node(minor, ent, &i915_forcewake_fops); } +static int i915_max_freq_create(struct dentry *root, struct drm_minor *minor) +{ + struct drm_device *dev = minor->dev; + struct dentry *ent; + + ent = debugfs_create_file("i915_max_freq", + S_IRUGO | S_IWUSR, + root, dev, + &i915_max_freq_fops); + if (IS_ERR(ent)) + return PTR_ERR(ent); + + return drm_add_fake_info_node(minor, ent, &i915_max_freq_fops); +} + static struct drm_info_list i915_debugfs_list[] = { {"i915_capabilities", i915_capabilities, 0}, {"i915_gem_objects", i915_gem_object_info, 0}, @@ -1449,6 +1534,9 @@ int i915_debugfs_init(struct drm_minor *minor) return ret; ret = i915_forcewake_create(minor->debugfs_root, minor); + if (ret) + return ret; + ret = i915_max_freq_create(minor->debugfs_root, minor); if (ret) return ret; @@ -1465,6 +1553,8 @@ void i915_debugfs_cleanup(struct drm_minor *minor) 1, minor); drm_debugfs_remove_files((struct drm_info_list *) &i915_wedged_fops, 1, minor); + drm_debugfs_remove_files((struct drm_info_list *) &i915_max_freq_fops, + 1, minor); } #endif /* CONFIG_DEBUG_FS */ From 013a41ec541d5daa0c9f2b5126d2e820902c052d Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Tue, 19 Jul 2011 15:38:56 -0700 Subject: [PATCH 099/439] drm/i915: provide more error output when mode sets fail If a mode set fails we may get a message from drm_crtc_helper if we're lucky, but it won't tell us anything about *why* we failed to set a mode. So add a few DRM_ERRORs for the cases that shouldn't happen so we can debug things more easily. Signed-off-by: Jesse Barnes Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_display.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 32c8c956d975..d2fb8cf3de0d 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -1958,7 +1958,7 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, /* no fb bound */ if (!crtc->fb) { - DRM_DEBUG_KMS("No FB bound\n"); + DRM_ERROR("No FB bound\n"); return 0; } @@ -1967,6 +1967,7 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, case 1: break; default: + DRM_ERROR("no plane for crtc\n"); return -EINVAL; } @@ -1976,6 +1977,7 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, NULL); if (ret != 0) { mutex_unlock(&dev->struct_mutex); + DRM_ERROR("pin & fence failed\n"); return ret; } @@ -2004,6 +2006,7 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, if (ret) { i915_gem_object_unpin(to_intel_framebuffer(crtc->fb)->obj); mutex_unlock(&dev->struct_mutex); + DRM_ERROR("failed to update base address\n"); return ret; } From 070d329ae52e2fde341771d753a5b728145881f4 Mon Sep 17 00:00:00 2001 From: Michel Alexandre Salim Date: Thu, 28 Jul 2011 18:52:06 +0200 Subject: [PATCH 100/439] drm/i915: Add quirk to disable SSC on Sony Vaio Y2 Using the new quirk added to support disabling SSC on Lenovo U160 (#36656, commit 435793dfb8aec7b2e19f72d5bce8a22fd0b57839), also register the Vaio as a special case and disable SSC for it. This patch fixes #34437 on fdo bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=34437 Signed-off-by: Michel Alexandre Salim Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_display.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index d2fb8cf3de0d..3d2900cbf431 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -7892,6 +7892,9 @@ struct intel_quirk intel_quirks[] = { /* Lenovo U160 cannot use SSC on LVDS */ { 0x0046, 0x17aa, 0x3920, quirk_ssc_force_disable }, + + /* Sony Vaio Y cannot use SSC on LVDS */ + { 0x0046, 0x104d, 0x9076, quirk_ssc_force_disable }, }; static void intel_init_quirks(struct drm_device *dev) From 291427f5fdadec6e4be2924172e83588880e1539 Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Fri, 29 Jul 2011 12:42:37 -0700 Subject: [PATCH 101/439] drm/i915: apply phase pointer override on SNB+ too These bits moved around on SNB and above. v2: again with the git send-email fail v3: add macros for getting per-pipe override & enable bits v4: enable phase sync pointer on SNB and IVB configs as well Signed-off-by: Jesse Barnes Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/i915_reg.h | 5 +++++ drivers/gpu/drm/i915/intel_display.c | 31 ++++++++++++++++++++++++++++ 2 files changed, 36 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 30d8aae9ac4c..a7f7a347c700 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -3091,6 +3091,11 @@ #define TRANS_CHICKEN2(pipe) _PIPE(pipe, _TRANSA_CHICKEN2, _TRANSB_CHICKEN2) #define TRANS_AUTOTRAIN_GEN_STALL_DIS (1<<31) +#define SOUTH_CHICKEN1 0xc2000 +#define FDIA_PHASE_SYNC_SHIFT_OVR 19 +#define FDIA_PHASE_SYNC_SHIFT_EN 18 +#define FDI_PHASE_SYNC_OVR(pipe) (1<<(FDIA_PHASE_SYNC_SHIFT_OVR - ((pipe) * 2))) +#define FDI_PHASE_SYNC_EN(pipe) (1<<(FDIA_PHASE_SYNC_SHIFT_EN - ((pipe) * 2))) #define SOUTH_CHICKEN2 0xc2004 #define DPLS_EDP_PPS_FIX_DIS (1<<0) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 3d2900cbf431..53164606918f 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -2113,6 +2113,18 @@ static void intel_fdi_normal_train(struct drm_crtc *crtc) FDI_FE_ERRC_ENABLE); } +static void cpt_phase_pointer_enable(struct drm_device *dev, int pipe) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + u32 flags = I915_READ(SOUTH_CHICKEN1); + + flags |= FDI_PHASE_SYNC_OVR(pipe); + I915_WRITE(SOUTH_CHICKEN1, flags); /* once to unlock... */ + flags |= FDI_PHASE_SYNC_EN(pipe); + I915_WRITE(SOUTH_CHICKEN1, flags); /* then again to enable */ + POSTING_READ(SOUTH_CHICKEN1); +} + /* The FDI link training functions for ILK/Ibexpeak. */ static void ironlake_fdi_link_train(struct drm_crtc *crtc) { @@ -2263,6 +2275,9 @@ static void gen6_fdi_link_train(struct drm_crtc *crtc) POSTING_READ(reg); udelay(150); + if (HAS_PCH_CPT(dev)) + cpt_phase_pointer_enable(dev, pipe); + for (i = 0; i < 4; i++ ) { reg = FDI_TX_CTL(pipe); temp = I915_READ(reg); @@ -2379,6 +2394,9 @@ static void ivb_manual_fdi_link_train(struct drm_crtc *crtc) POSTING_READ(reg); udelay(150); + if (HAS_PCH_CPT(dev)) + cpt_phase_pointer_enable(dev, pipe); + for (i = 0; i < 4; i++ ) { reg = FDI_TX_CTL(pipe); temp = I915_READ(reg); @@ -2488,6 +2506,17 @@ static void ironlake_fdi_pll_enable(struct drm_crtc *crtc) } } +static void cpt_phase_pointer_disable(struct drm_device *dev, int pipe) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + u32 flags = I915_READ(SOUTH_CHICKEN1); + + flags &= ~(FDI_PHASE_SYNC_EN(pipe)); + I915_WRITE(SOUTH_CHICKEN1, flags); /* once to disable... */ + flags &= ~(FDI_PHASE_SYNC_OVR(pipe)); + I915_WRITE(SOUTH_CHICKEN1, flags); /* then again to lock */ + POSTING_READ(SOUTH_CHICKEN1); +} static void ironlake_fdi_disable(struct drm_crtc *crtc) { struct drm_device *dev = crtc->dev; @@ -2517,6 +2546,8 @@ static void ironlake_fdi_disable(struct drm_crtc *crtc) I915_WRITE(FDI_RX_CHICKEN(pipe), I915_READ(FDI_RX_CHICKEN(pipe) & ~FDI_RX_PHASE_SYNC_POINTER_EN)); + } else if (HAS_PCH_CPT(dev)) { + cpt_phase_pointer_disable(dev, pipe); } /* still set train pattern 1 */ From cda2bb78c24de7674eafa3210314dc75bed344a6 Mon Sep 17 00:00:00 2001 From: Adam Jackson Date: Tue, 26 Jul 2011 16:53:06 -0400 Subject: [PATCH 102/439] drm/i915/pch: Save/restore PCH_PORT_HOTPLUG across suspend At least on a Lenovo X220 the HPD bits of this are enabled at boot but cleared after resume, which means plug interrupts stop working. This also happens to fix DP displays re-lighting on resume. I'm quite certain that's an accident: the first DP link train inevitably fails on that machine, and it's only serendipity that we're getting multiple plug interrupts and the second train works. But I shall take my victories where I get them. Signed-off-by: Adam Jackson Tested-by: Keith Packard Reviewed-by: Keith Packard Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/i915_drv.h | 1 + drivers/gpu/drm/i915/i915_suspend.c | 2 ++ 2 files changed, 3 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index ce7914c4c044..e0d0e278f62d 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -541,6 +541,7 @@ typedef struct drm_i915_private { u32 savePIPEB_LINK_M1; u32 savePIPEB_LINK_N1; u32 saveMCHBAR_RENDER_STANDBY; + u32 savePCH_PORT_HOTPLUG; struct { /** Bridge to intel-gtt-ko */ diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c index 5257cfc34c35..27693c05c6d2 100644 --- a/drivers/gpu/drm/i915/i915_suspend.c +++ b/drivers/gpu/drm/i915/i915_suspend.c @@ -814,6 +814,7 @@ int i915_save_state(struct drm_device *dev) dev_priv->saveFDI_RXB_IMR = I915_READ(_FDI_RXB_IMR); dev_priv->saveMCHBAR_RENDER_STANDBY = I915_READ(RSTDBYCTL); + dev_priv->savePCH_PORT_HOTPLUG = I915_READ(PCH_PORT_HOTPLUG); } else { dev_priv->saveIER = I915_READ(IER); dev_priv->saveIMR = I915_READ(IMR); @@ -865,6 +866,7 @@ int i915_restore_state(struct drm_device *dev) I915_WRITE(GTIMR, dev_priv->saveGTIMR); I915_WRITE(_FDI_RXA_IMR, dev_priv->saveFDI_RXA_IMR); I915_WRITE(_FDI_RXB_IMR, dev_priv->saveFDI_RXB_IMR); + I915_WRITE(PCH_PORT_HOTPLUG, dev_priv->savePCH_PORT_HOTPLUG); } else { I915_WRITE(IER, dev_priv->saveIER); I915_WRITE(IMR, dev_priv->saveIMR); From bb0c5ed6ec523199e34e81dcef8e987507553b63 Mon Sep 17 00:00:00 2001 From: Zhang Rui Date: Sat, 30 Jul 2011 01:40:48 -0400 Subject: [PATCH 103/439] ACPI: DMI workaround for Asus A8N-SLI Premium and Asus A8N-SLI DELUX MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit DMI workaround for A8N-SLI Premium and A8N-SLI DELUXE to enable the s3 suspend old ordering. http://bugzilla.kernel.org/show_bug.cgi?id=9528 Tested-by: Heiko Ettelbrück Tested-by: Brian Beardall Signed-off-by: Zhang Rui Signed-off-by: Len Brown --- drivers/acpi/sleep.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/drivers/acpi/sleep.c b/drivers/acpi/sleep.c index 6c949602cbd1..3ed80b2ca907 100644 --- a/drivers/acpi/sleep.c +++ b/drivers/acpi/sleep.c @@ -428,6 +428,22 @@ static struct dmi_system_id __initdata acpisleep_dmi_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "1000 Series"), }, }, + { + .callback = init_old_suspend_ordering, + .ident = "Asus A8N-SLI DELUXE", + .matches = { + DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."), + DMI_MATCH(DMI_BOARD_NAME, "A8N-SLI DELUXE"), + }, + }, + { + .callback = init_old_suspend_ordering, + .ident = "Asus A8N-SLI Premium", + .matches = { + DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."), + DMI_MATCH(DMI_BOARD_NAME, "A8N-SLI Premium"), + }, + }, {}, }; #endif /* CONFIG_SUSPEND */ From ec2cf68e024b4f66df11683147d63e07db6ee875 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Fri, 29 Jul 2011 21:14:12 -0700 Subject: [PATCH 104/439] ALSA: rtctimer.c needs module.h rtctimer.c uses interfaces from linux/module.h, so it should include that file. This fixes build errors. Signed-off-by: Randy Dunlap Signed-off-by: Takashi Iwai --- sound/core/rtctimer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/core/rtctimer.c b/sound/core/rtctimer.c index 0851cd13e303..e85e72baff9e 100644 --- a/sound/core/rtctimer.c +++ b/sound/core/rtctimer.c @@ -22,7 +22,7 @@ #include #include -#include +#include #include #include #include From dd8ae59d48790d5c25f47ebbe502c8ca379fdde0 Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Sat, 30 Jul 2011 05:03:58 -0700 Subject: [PATCH 105/439] target: Fix bug for transport_generic_wait_for_tasks with direct operation This patch fixes a bug in transport_handle_cdb_direct() usage with target_core where transport_generic_wait_for_tasks() was bypassing active I/O + usage of cmd->t_transport_stop_comp because cmd->t_transport_active=1 was not being set before dispatching with transport_generic_new_cmd(). The fix follows existing usage in transport_generic_handle_cdb*() -> transport_add_cmd_to_queue() and set these directly, as well as handle transport_generic_new_cmd() exceptions for QUEUE_FULL and CHECK_CONDITION instead of propigating up to RX context fabric code. The bug was manifesting itself with the following SLUB poison overwritten warnings with iscsi-target v4.1 LUNs using the new process context direct operation during session reinstatement with active I/O exception handling: [885410.498267] ============================================================================= [885410.621622] BUG lio_cmd_cache: Poison overwritten [885410.621791] ----------------------------------------------------------------------------- [885410.621792] [885410.623420] INFO: 0xffff880000cf3750-0xffff880000cf378d. First byte 0x6a instead of 0x6b [885410.626332] INFO: Allocated in iscsit_allocate_cmd+0x1c/0xd4 [iscsi_target_mod] age=345 cpu=1 pid=22554 [885411.855189] INFO: Freed in iscsit_release_cmd+0x208/0x217 [iscsi_target_mod] age=1410 cpu=1 pid=22554 [885411.856048] INFO: Slab 0xffffea000002d480 objects=22 used=0 fp=0xffff880000cf7300 flags=0x4080 [885411.856368] INFO: Object 0xffff880000cf33c0 @offset=13248 fp=0xffff880000cf6780 [885411.955678] Pid: 22554, comm: iscsi_trx Not tainted 3.0.0-rc7+ #30 [885411.956040] Call Trace: [885411.957029] [] print_trailer+0x12e/0x137 [885412.752879] [] check_bytes_and_report+0xb9/0xfd [885412.754933] [] check_object+0xb5/0x192 [885412.755099] [] __free_slab+0x96/0x13a [885412.757008] [] discard_slab+0x41/0x43 [885412.758171] [] __slab_free+0xf3/0xfe [885412.761027] [] ? iscsit_release_cmd+0x208/0x217 [iscsi_target_mod] [885412.761354] [] kmem_cache_free+0x6f/0xac [885412.761536] [] iscsit_release_cmd+0x208/0x217 [iscsi_target_mod] [885412.762056] [] ? iblock_free_task+0x34/0x39 [target_core_iblock] [885412.762368] [] lio_release_cmd+0x10/0x12 [iscsi_target_mod] [885412.764129] [] transport_release_cmd+0x2f/0x33 [target_core_mod] [885412.805024] [] transport_generic_remove+0xb6/0xc3 [target_core_mod] [885412.806424] [] ? try_to_wake_up+0x1bd/0x1bd [885412.809033] [] transport_generic_free_cmd+0x75/0x7d [target_core_mod] [885412.810066] [] transport_generic_wait_for_tasks+0x21c/0x22b [target_core_mod] [885412.811056] [] ? mutex_lock+0x11/0x32 [885412.813059] [] ? mutex_lock+0x11/0x32 [885412.813200] [] iscsit_close_connection+0x1d5/0x63a [iscsi_target_mod] [885412.813517] [] iscsit_take_action_for_connection_exit+0xdb/0xe0 [iscsi_target_mod] [885412.813851] [] iscsi_target_rx_thread+0x11f6/0x1221 [iscsi_target_mod] [885412.829024] [] ? pick_next_task_fair+0xbe/0x10e [885412.831010] [] ? iscsit_handle_scsi_cmd+0x91d/0x91d [iscsi_target_mod] [885412.833011] [] ? iscsit_handle_scsi_cmd+0x91d/0x91d [iscsi_target_mod] [885412.835010] [] kthread+0x7d/0x85 [885412.837022] [] kernel_thread_helper+0x4/0x10 [885412.838008] [] ? kthread_worker_fn+0x145/0x145 [885412.840047] [] ? gs_change+0x13/0x13 [885412.842007] FIX lio_cmd_cache: Restoring 0xffff880000cf3750-0xffff880000cf378d=0x6 Cc: Christoph Hellwig Cc: Andy Grover Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_transport.c | 31 ++++++++++++++++++++++++-- 1 file changed, 29 insertions(+), 2 deletions(-) diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c index ff7fcf8366a0..89760329d5d0 100644 --- a/drivers/target/target_core_transport.c +++ b/drivers/target/target_core_transport.c @@ -1747,6 +1747,8 @@ int transport_generic_handle_cdb( } EXPORT_SYMBOL(transport_generic_handle_cdb); +static void transport_generic_request_failure(struct se_cmd *, + struct se_device *, int, int); /* * Used by fabric module frontends to queue tasks directly. * Many only be used from process context only @@ -1754,6 +1756,8 @@ EXPORT_SYMBOL(transport_generic_handle_cdb); int transport_handle_cdb_direct( struct se_cmd *cmd) { + int ret; + if (!cmd->se_lun) { dump_stack(); pr_err("cmd->se_lun is NULL\n"); @@ -1765,8 +1769,31 @@ int transport_handle_cdb_direct( " from interrupt context\n"); return -EINVAL; } - - return transport_generic_new_cmd(cmd); + /* + * Set TRANSPORT_NEW_CMD state and cmd->t_transport_active=1 following + * transport_generic_handle_cdb*() -> transport_add_cmd_to_queue() + * in existing usage to ensure that outstanding descriptors are handled + * correctly during shutdown via transport_generic_wait_for_tasks() + * + * Also, we don't take cmd->t_state_lock here as we only expect + * this to be called for initial descriptor submission. + */ + cmd->t_state = TRANSPORT_NEW_CMD; + atomic_set(&cmd->t_transport_active, 1); + /* + * transport_generic_new_cmd() is already handling QUEUE_FULL, + * so follow TRANSPORT_NEW_CMD processing thread context usage + * and call transport_generic_request_failure() if necessary.. + */ + ret = transport_generic_new_cmd(cmd); + if (ret == -EAGAIN) + return 0; + else if (ret < 0) { + cmd->transport_error_status = ret; + transport_generic_request_failure(cmd, NULL, 0, + (cmd->data_direction != DMA_TO_DEVICE)); + } + return 0; } EXPORT_SYMBOL(transport_handle_cdb_direct); From ab3d0abe2e4c1f164af7a6cc3694fcb8c24a57ee Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sat, 30 Jul 2011 11:53:47 -0700 Subject: [PATCH 106/439] Input: psmouse - hgpk.c needs module.h hgpk.c uses interfaces from linux/module.h, so it should include that file. This fixes build errors. Signed-off-by: Randy Dunlap Signed-off-by: Dmitry Torokhov --- drivers/input/mouse/hgpk.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/input/mouse/hgpk.c b/drivers/input/mouse/hgpk.c index 95577c15ae56..4d17d9f3320b 100644 --- a/drivers/input/mouse/hgpk.c +++ b/drivers/input/mouse/hgpk.c @@ -32,6 +32,7 @@ #define DEBUG #include #include +#include #include #include #include From 2501ec97663d84cfbc8fd8848c382f89c3bf8d1d Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Sat, 30 Jul 2011 11:55:38 -0700 Subject: [PATCH 107/439] Input: kxtj9 - explicitly include module.h We need to explicitly include module.h since some of its facilities are used. Signed-off-by: Stephen Rothwell Signed-off-by: Dmitry Torokhov --- drivers/input/misc/kxtj9.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/input/misc/kxtj9.c b/drivers/input/misc/kxtj9.c index c456f63b6bae..783597a9a64a 100644 --- a/drivers/input/misc/kxtj9.c +++ b/drivers/input/misc/kxtj9.c @@ -21,6 +21,7 @@ #include #include #include +#include #include #include #include From 3f27757a1182f5e3feff7425b1c3e43f3e466724 Mon Sep 17 00:00:00 2001 From: Rakesh Iyer Date: Sat, 30 Jul 2011 12:01:48 -0700 Subject: [PATCH 108/439] Input: tegra-kbc - fix computation of polling time Fix a constant definition and computation of polling time. [dtor@mail.ru: switched to using DIV_ROUND_UP as was suggested by Thierry Reding ] Signed-off-by: Rakesh Iyer Signed-off-by: Dmitry Torokhov --- drivers/input/keyboard/tegra-kbc.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/input/keyboard/tegra-kbc.c b/drivers/input/keyboard/tegra-kbc.c index da3828fc2c09..f270447ba951 100644 --- a/drivers/input/keyboard/tegra-kbc.c +++ b/drivers/input/keyboard/tegra-kbc.c @@ -19,6 +19,7 @@ * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ +#include #include #include #include @@ -37,7 +38,7 @@ #define KBC_ROW_SCAN_DLY 5 /* KBC uses a 32KHz clock so a cycle = 1/32Khz */ -#define KBC_CYCLE_USEC 32 +#define KBC_CYCLE_MS 32 /* KBC Registers */ @@ -647,7 +648,7 @@ static int __devinit tegra_kbc_probe(struct platform_device *pdev) debounce_cnt = min(pdata->debounce_cnt, KBC_MAX_DEBOUNCE_CNT); scan_time_rows = (KBC_ROW_SCAN_TIME + debounce_cnt) * num_rows; kbc->repoll_dly = KBC_ROW_SCAN_DLY + scan_time_rows + pdata->repeat_cnt; - kbc->repoll_dly = ((kbc->repoll_dly * KBC_CYCLE_USEC) + 999) / 1000; + kbc->repoll_dly = DIV_ROUND_UP(kbc->repoll_dly, KBC_CYCLE_MS); input_dev->name = pdev->name; input_dev->id.bustype = BUS_HOST; From 52db9819ac96b0c5a4e075d836cf21dc529cbce4 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Sat, 30 Jul 2011 12:05:13 -0700 Subject: [PATCH 109/439] Input: lm8323 - add missing device_remove_file for dev_attr_time Add missing device_remove_file() for dev_attr_time in lm8323_remove(). Also calling device_remove_file() in lm8323_probe() error path to remove sysfs attribute file. Signed-off-by: Axel Lin Signed-off-by: Dmitry Torokhov --- drivers/input/keyboard/lm8323.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/drivers/input/keyboard/lm8323.c b/drivers/input/keyboard/lm8323.c index ab0acaf7fe8f..756348a7f93a 100644 --- a/drivers/input/keyboard/lm8323.c +++ b/drivers/input/keyboard/lm8323.c @@ -754,8 +754,11 @@ fail3: device_remove_file(&client->dev, &dev_attr_disable_kp); fail2: while (--pwm >= 0) - if (lm->pwm[pwm].enabled) + if (lm->pwm[pwm].enabled) { + device_remove_file(lm->pwm[pwm].cdev.dev, + &dev_attr_time); led_classdev_unregister(&lm->pwm[pwm].cdev); + } fail1: input_free_device(idev); kfree(lm); @@ -775,8 +778,10 @@ static int __devexit lm8323_remove(struct i2c_client *client) device_remove_file(&lm->client->dev, &dev_attr_disable_kp); for (i = 0; i < 3; i++) - if (lm->pwm[i].enabled) + if (lm->pwm[i].enabled) { + device_remove_file(lm->pwm[i].cdev.dev, &dev_attr_time); led_classdev_unregister(&lm->pwm[i].cdev); + } kfree(lm); From 1f4bb066433322f6f189b084ceebdfb4add77292 Mon Sep 17 00:00:00 2001 From: Tobias Klauser Date: Sat, 30 Jul 2011 12:08:10 -0700 Subject: [PATCH 110/439] Input: gpio_keys - return proper error code if memory allocation fails Return -ENOMEM if kzalloc fails in gpio_keys_get_devtree_pdata(). Signed-off-by: Tobias Klauser Signed-off-by: Dmitry Torokhov --- drivers/input/keyboard/gpio_keys.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/input/keyboard/gpio_keys.c b/drivers/input/keyboard/gpio_keys.c index ce281d152275..67df91af8424 100644 --- a/drivers/input/keyboard/gpio_keys.c +++ b/drivers/input/keyboard/gpio_keys.c @@ -483,7 +483,7 @@ static int gpio_keys_get_devtree_pdata(struct device *dev, buttons = kzalloc(pdata->nbuttons * (sizeof *buttons), GFP_KERNEL); if (!buttons) - return -ENODEV; + return -ENOMEM; pp = NULL; i = 0; From 205e9a2106b934ea39049bab28f0896c17a2cb30 Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Fri, 29 Jul 2011 18:37:02 +0400 Subject: [PATCH 111/439] [PARISC] wire up sendmmsg syscall Cc: stable@kernel.org Signed-off-by: James Bottomley --- arch/parisc/include/asm/unistd.h | 3 ++- arch/parisc/kernel/syscall_table.S | 1 + 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/parisc/include/asm/unistd.h b/arch/parisc/include/asm/unistd.h index 3392de3e7be0..d61de64f990a 100644 --- a/arch/parisc/include/asm/unistd.h +++ b/arch/parisc/include/asm/unistd.h @@ -821,8 +821,9 @@ #define __NR_open_by_handle_at (__NR_Linux + 326) #define __NR_syncfs (__NR_Linux + 327) #define __NR_setns (__NR_Linux + 328) +#define __NR_sendmmsg (__NR_Linux + 329) -#define __NR_Linux_syscalls (__NR_setns + 1) +#define __NR_Linux_syscalls (__NR_sendmmsg + 1) #define __IGNORE_select /* newselect */ diff --git a/arch/parisc/kernel/syscall_table.S b/arch/parisc/kernel/syscall_table.S index 34a4f5a2fffb..e66366fd2abc 100644 --- a/arch/parisc/kernel/syscall_table.S +++ b/arch/parisc/kernel/syscall_table.S @@ -427,6 +427,7 @@ ENTRY_COMP(open_by_handle_at) ENTRY_SAME(syncfs) ENTRY_SAME(setns) + ENTRY_COMP(sendmmsg) /* Nothing yet */ From 71ff069c3d6a2b23eaf516bc714a20ce0cdc3609 Mon Sep 17 00:00:00 2001 From: Shawn Guo Date: Sun, 31 Jul 2011 19:56:10 -0700 Subject: [PATCH 112/439] Input: mma8450 - add device tree probe support It adds device tree probe support for mma8450 driver. Signed-off-by: Shawn Guo Acked-by: Eric Miao Acked-by: Grant Likely Signed-off-by: Dmitry Torokhov --- .../devicetree/bindings/input/fsl-mma8450.txt | 11 +++++++++++ drivers/input/misc/mma8450.c | 8 ++++++++ 2 files changed, 19 insertions(+) create mode 100644 Documentation/devicetree/bindings/input/fsl-mma8450.txt diff --git a/Documentation/devicetree/bindings/input/fsl-mma8450.txt b/Documentation/devicetree/bindings/input/fsl-mma8450.txt new file mode 100644 index 000000000000..a00c94ccbdee --- /dev/null +++ b/Documentation/devicetree/bindings/input/fsl-mma8450.txt @@ -0,0 +1,11 @@ +* Freescale MMA8450 3-Axis Accelerometer + +Required properties: +- compatible : "fsl,mma8450". + +Example: + +accelerometer: mma8450@1c { + compatible = "fsl,mma8450"; + reg = <0x1c>; +}; diff --git a/drivers/input/misc/mma8450.c b/drivers/input/misc/mma8450.c index 20f8f9284f02..6c76cf792991 100644 --- a/drivers/input/misc/mma8450.c +++ b/drivers/input/misc/mma8450.c @@ -24,6 +24,7 @@ #include #include #include +#include #define MMA8450_DRV_NAME "mma8450" @@ -229,10 +230,17 @@ static const struct i2c_device_id mma8450_id[] = { }; MODULE_DEVICE_TABLE(i2c, mma8450_id); +static const struct of_device_id mma8450_dt_ids[] = { + { .compatible = "fsl,mma8450", }, + { /* sentinel */ } +}; +MODULE_DEVICE_TABLE(i2c, mma8450_dt_ids); + static struct i2c_driver mma8450_driver = { .driver = { .name = MMA8450_DRV_NAME, .owner = THIS_MODULE, + .of_match_table = mma8450_dt_ids, }, .probe = mma8450_probe, .remove = __devexit_p(mma8450_remove), From dc889f18646325d07eb24ef634f082d60b0beebb Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Sun, 31 Jul 2011 23:16:43 +0200 Subject: [PATCH 113/439] ALSA: asihpi - Don't leak firmware if mem alloc fails We leak the memory allocated to 'firmware' when we fail to release_firmware() after a kmalloc() failure in hpi_dsp_code_open(). This patch should take care of the leak. Signed-off-by: Jesper Juhl Signed-off-by: Takashi Iwai --- sound/pci/asihpi/hpidspcd.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/sound/pci/asihpi/hpidspcd.c b/sound/pci/asihpi/hpidspcd.c index 3a7afa31c1d8..71d32c868c92 100644 --- a/sound/pci/asihpi/hpidspcd.c +++ b/sound/pci/asihpi/hpidspcd.c @@ -43,6 +43,7 @@ short hpi_dsp_code_open(u32 adapter, void *os_data, struct dsp_code *dsp_code, struct pci_dev *dev = os_data; struct code_header header; char fw_name[20]; + short err_ret = HPI_ERROR_DSP_FILE_NOT_FOUND; int err; sprintf(fw_name, "asihpi/dsp%04x.bin", adapter); @@ -85,8 +86,10 @@ short hpi_dsp_code_open(u32 adapter, void *os_data, struct dsp_code *dsp_code, HPI_DEBUG_LOG(DEBUG, "dsp code %s opened\n", fw_name); dsp_code->pvt = kmalloc(sizeof(*dsp_code->pvt), GFP_KERNEL); - if (!dsp_code->pvt) - return HPI_ERROR_MEMORY_ALLOC; + if (!dsp_code->pvt) { + err_ret = HPI_ERROR_MEMORY_ALLOC; + goto error2; + } dsp_code->pvt->dev = dev; dsp_code->pvt->firmware = firmware; @@ -99,7 +102,7 @@ error2: release_firmware(firmware); error1: dsp_code->block_length = 0; - return HPI_ERROR_DSP_FILE_NOT_FOUND; + return err_ret; } /*-------------------------------------------------------------------*/ From 2a1690d8cd5b7a4271f993f59a4484d41e2884c3 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 1 Aug 2011 02:23:15 -0700 Subject: [PATCH 114/439] MAINTAINERS: Remove Pekka Savola from ipv6. Signed-off-by: David S. Miller --- MAINTAINERS | 1 - 1 file changed, 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index a99fed1a3d87..5d4ceba0e960 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4443,7 +4443,6 @@ F: include/linux/netdevice.h NETWORKING [IPv4/IPv6] M: "David S. Miller" M: Alexey Kuznetsov -M: "Pekka Savola (ipv6)" M: James Morris M: Hideaki YOSHIFUJI M: Patrick McHardy From a1889c0d2039a53ae04abb9f20c62500bd312bf3 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Thu, 28 Jul 2011 02:46:01 +0000 Subject: [PATCH 115/439] net: adjust array index Convert array index from the loop bound to the loop index. A simplified version of the semantic patch that fixes this problem is as follows: (http://coccinelle.lip6.fr/) // @@ expression e1,e2,ar; @@ for(e1 = 0; e1 < e2; e1++) { <... ar[ - e2 + e1 ] ...> } // Signed-off-by: Julia Lawall Signed-off-by: David S. Miller --- net/ipv4/igmp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index f1d27f6c9351..283c0a26e03f 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -1718,7 +1718,7 @@ static int ip_mc_add_src(struct in_device *in_dev, __be32 *pmca, int sfmode, pmc->sfcount[sfmode]--; for (j=0; jsfcount[MCAST_EXCLUDE] != 0)) { #ifdef CONFIG_IP_MULTICAST struct ip_sf_list *psf; From 956837f7c954443f426a82ba6f17b33488cf9a0c Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Thu, 28 Jul 2011 02:46:03 +0000 Subject: [PATCH 116/439] drivers/net/niu.c: adjust array index Convert array index from the loop bound to the loop index. A simplified version of the semantic patch that fixes this problem is as follows: (http://coccinelle.lip6.fr/) // @@ expression e1,e2,ar; @@ for(e1 = 0; e1 < e2; e1++) { <... ar[ - e2 + e1 ] ...> } // Signed-off-by: Julia Lawall Signed-off-by: David S. Miller --- drivers/net/niu.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/niu.c b/drivers/net/niu.c index cd6c2317e29e..ed47585a6862 100644 --- a/drivers/net/niu.c +++ b/drivers/net/niu.c @@ -9201,7 +9201,7 @@ static int __devinit niu_ldg_init(struct niu *np) first_chan = 0; for (i = 0; i < port; i++) - first_chan += parent->rxchan_per_port[port]; + first_chan += parent->rxchan_per_port[i]; num_chan = parent->rxchan_per_port[port]; for (i = first_chan; i < (first_chan + num_chan); i++) { @@ -9217,7 +9217,7 @@ static int __devinit niu_ldg_init(struct niu *np) first_chan = 0; for (i = 0; i < port; i++) - first_chan += parent->txchan_per_port[port]; + first_chan += parent->txchan_per_port[i]; num_chan = parent->txchan_per_port[port]; for (i = first_chan; i < (first_chan + num_chan); i++) { err = niu_ldg_assign_ldn(np, parent, From e1738bd9cecc5c867b0e2996470c1ff20f66ba79 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 29 Jul 2011 19:22:42 +0000 Subject: [PATCH 117/439] sch_sfq: fix sfq_enqueue() commit 8efa88540635 (sch_sfq: avoid giving spurious NET_XMIT_CN signals) forgot to call qdisc_tree_decrease_qlen() to signal upper levels that a packet (from another flow) was dropped, leading to various problems. With help from Michal Soltys and Michal Pokrywka, who did a bisection. Bugzilla ref: https://bugzilla.kernel.org/show_bug.cgi?id=39372 Debian ref: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=631945 Reported-by: Lucas Bocchi Reported-and-bisected-by: Michal Pokrywka Signed-off-by: Eric Dumazet CC: Michal Soltys Acked-by: Patrick McHardy Signed-off-by: David S. Miller --- net/sched/sch_sfq.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c index 4536ee64383e..4f5510e2bd6f 100644 --- a/net/sched/sch_sfq.c +++ b/net/sched/sch_sfq.c @@ -410,7 +410,12 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc *sch) /* Return Congestion Notification only if we dropped a packet * from this flow. */ - return (qlen != slot->qlen) ? NET_XMIT_CN : NET_XMIT_SUCCESS; + if (qlen != slot->qlen) + return NET_XMIT_CN; + + /* As we dropped a packet, better let upper stack know this */ + qdisc_tree_decrease_qlen(sch, 1); + return NET_XMIT_SUCCESS; } static struct sk_buff * From 28c1b6d60e3ad0aecf48aaefc6995904e2ab1b9e Mon Sep 17 00:00:00 2001 From: Felipe Balbi Date: Fri, 22 Jul 2011 22:47:21 +0300 Subject: [PATCH 118/439] of: address: use resource_size helper that should be the approved way of calculating the size of resources. No functional changes. Cc: Grant Likely Signed-off-by: Felipe Balbi Signed-off-by: Grant Likely --- drivers/of/address.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/of/address.c b/drivers/of/address.c index da1f4b9605df..72c33fbe451d 100644 --- a/drivers/of/address.c +++ b/drivers/of/address.c @@ -610,6 +610,6 @@ void __iomem *of_iomap(struct device_node *np, int index) if (of_address_to_resource(np, index, &res)) return NULL; - return ioremap(res.start, 1 + res.end - res.start); + return ioremap(res.start, resource_size(&res)); } EXPORT_SYMBOL(of_iomap); From 5f9288af065537be1fe6d1d5bd54ba51d6df721a Mon Sep 17 00:00:00 2001 From: Michal Simek Date: Wed, 13 Jul 2011 15:46:21 +0200 Subject: [PATCH 119/439] input: xilinx_ps2: Add missing of_address.h header Add missing header. Error log: CC drivers/input/serio/xilinx_ps2.o drivers/input/serio/xilinx_ps2.c: In function 'xps2_of_probe': drivers/input/serio/xilinx_ps2.c:249: error: implicit declaration of function 'of_address_to_resource' Signed-off-by: Michal Simek Signed-off-by: Grant Likely --- drivers/input/serio/xilinx_ps2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/input/serio/xilinx_ps2.c b/drivers/input/serio/xilinx_ps2.c index 80baa53da5b1..d64c5a43aaad 100644 --- a/drivers/input/serio/xilinx_ps2.c +++ b/drivers/input/serio/xilinx_ps2.c @@ -23,7 +23,7 @@ #include #include #include - +#include #include #include From c1227340ca65c2069222a956a68b6842d460c4f4 Mon Sep 17 00:00:00 2001 From: Felix Fietkau Date: Wed, 27 Jul 2011 15:01:02 +0200 Subject: [PATCH 120/439] ath9k: initialize tx chainmask before testing channel tx power values With an uninitialized chainmask, the per-channel power will only contain the power limits for a single chain instead of the combined tx power. Signed-off-by: Felix Fietkau Cc: stable@kernel.org Signed-off-by: John W. Linville --- drivers/net/wireless/ath/ath9k/init.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/wireless/ath/ath9k/init.c b/drivers/net/wireless/ath/ath9k/init.c index dceaa4a82811..2e9634641c46 100644 --- a/drivers/net/wireless/ath/ath9k/init.c +++ b/drivers/net/wireless/ath/ath9k/init.c @@ -669,8 +669,10 @@ static void ath9k_init_band_txpower(struct ath_softc *sc, int band) static void ath9k_init_txpower_limits(struct ath_softc *sc) { struct ath_hw *ah = sc->sc_ah; + struct ath_common *common = ath9k_hw_common(sc->sc_ah); struct ath9k_channel *curchan = ah->curchan; + ah->txchainmask = common->tx_chainmask; if (ah->caps.hw_caps & ATH9K_HW_CAP_2GHZ) ath9k_init_band_txpower(sc, IEEE80211_BAND_2GHZ); if (ah->caps.hw_caps & ATH9K_HW_CAP_5GHZ) From 17e859a899712d16c3e70b045d61ad9e02c53f8a Mon Sep 17 00:00:00 2001 From: Stanislaw Gruszka Date: Wed, 27 Jul 2011 15:37:43 +0200 Subject: [PATCH 121/439] iwlegacy: set tx power after rxon_assoc If settings of tx power was deferred during scan or changing channel we have to setup them during commit rxon. Fix problem on 3945 (4965 already has this fix). Optimize code to apply tx settings only when tx power was actually changed. Cc: stable@kernel.org # 2.6.39+ Signed-off-by: Stanislaw Gruszka Signed-off-by: John W. Linville --- drivers/net/wireless/iwlegacy/iwl-3945.c | 6 +++++- drivers/net/wireless/iwlegacy/iwl-4965.c | 8 ++++++-- 2 files changed, 11 insertions(+), 3 deletions(-) diff --git a/drivers/net/wireless/iwlegacy/iwl-3945.c b/drivers/net/wireless/iwlegacy/iwl-3945.c index dab67a12d73b..73fe3cdf796b 100644 --- a/drivers/net/wireless/iwlegacy/iwl-3945.c +++ b/drivers/net/wireless/iwlegacy/iwl-3945.c @@ -1746,7 +1746,11 @@ int iwl3945_commit_rxon(struct iwl_priv *priv, struct iwl_rxon_context *ctx) } memcpy(active_rxon, staging_rxon, sizeof(*active_rxon)); - + /* + * We do not commit tx power settings while channel changing, + * do it now if tx power changed. + */ + iwl_legacy_set_tx_power(priv, priv->tx_power_next, false); return 0; } diff --git a/drivers/net/wireless/iwlegacy/iwl-4965.c b/drivers/net/wireless/iwlegacy/iwl-4965.c index bd4b000733f7..ecdc6e557428 100644 --- a/drivers/net/wireless/iwlegacy/iwl-4965.c +++ b/drivers/net/wireless/iwlegacy/iwl-4965.c @@ -1235,7 +1235,12 @@ static int iwl4965_commit_rxon(struct iwl_priv *priv, struct iwl_rxon_context *c memcpy(active_rxon, &ctx->staging, sizeof(*active_rxon)); iwl_legacy_print_rx_config_cmd(priv, ctx); - goto set_tx_power; + /* + * We do not commit tx power settings while channel changing, + * do it now if tx power changed. + */ + iwl_legacy_set_tx_power(priv, priv->tx_power_next, false); + return 0; } /* If we are currently associated and the new config requires @@ -1315,7 +1320,6 @@ static int iwl4965_commit_rxon(struct iwl_priv *priv, struct iwl_rxon_context *c iwl4965_init_sensitivity(priv); -set_tx_power: /* If we issue a new RXON command which required a tune then we must * send a new TXPOWER command or we won't be able to Tx any frames */ ret = iwl_legacy_set_tx_power(priv, priv->tx_power_next, true); From 84404623da45aac04595a8f5760a58df0e955d87 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Fri, 29 Jul 2011 11:52:18 +0300 Subject: [PATCH 122/439] cfg80211: off by one in nl80211_trigger_scan() The test is off by one so we'd read past the end of the wiphy->bands[] array on the next line. Signed-off-by: Dan Carpenter Signed-off-by: John W. Linville --- net/wireless/nl80211.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 28d2aa109bee..e83e7fee3bc0 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -3464,7 +3464,7 @@ static int nl80211_trigger_scan(struct sk_buff *skb, struct genl_info *info) tmp) { enum ieee80211_band band = nla_type(attr); - if (band < 0 || band > IEEE80211_NUM_BANDS) { + if (band < 0 || band >= IEEE80211_NUM_BANDS) { err = -EINVAL; goto out_free; } From d4930086bdd0c08a8b3a4d66a9c702297cb74a99 Mon Sep 17 00:00:00 2001 From: Stanislaw Gruszka Date: Fri, 29 Jul 2011 15:59:08 +0200 Subject: [PATCH 123/439] ath9k: skip ->config_pci_powersave() if PCIe port has ASPM disabled We receive many bug reports about system hang during suspend/resume when ath9k driver is in use. Adrian Chadd remarked that this problem happens on systems that have ASPM disabled. To do not hit the bug, skip doing ->config_pci_powersave magic if PCIe downstream port device, which ath9k device is connected to, has ASPM disabled. Bug was introduced by: commit 53bc7aa08b48e5cd745f986731cc7dc24eef2a9f Author: Vivek Natarajan Date: Mon Apr 5 14:48:04 2010 +0530 ath9k: Add support for newer AR9285 chipsets. Patch should address: https://bugzilla.kernel.org/show_bug.cgi?id=37462 https://bugzilla.kernel.org/show_bug.cgi?id=37082 https://bugzilla.redhat.com/show_bug.cgi?id=697157 however I did not receive confirmation about that, except from Camilo Mesias, whose system stops hang regularly with this patch (but still hangs from time to time, but this is probably some other bug). Tested-by: Camilo Mesias Cc: stable@kernel.org # 2.6.35+ Signed-off-by: Stanislaw Gruszka Signed-off-by: John W. Linville --- drivers/net/wireless/ath/ath9k/ar9002_hw.c | 6 +---- drivers/net/wireless/ath/ath9k/ar9003_hw.c | 6 +---- drivers/net/wireless/ath/ath9k/hw.c | 11 +++++++-- drivers/net/wireless/ath/ath9k/hw.h | 3 ++- drivers/net/wireless/ath/ath9k/pci.c | 27 ++++++++++++++++++++++ 5 files changed, 40 insertions(+), 13 deletions(-) diff --git a/drivers/net/wireless/ath/ath9k/ar9002_hw.c b/drivers/net/wireless/ath/ath9k/ar9002_hw.c index 9ff7c30573b8..44d9d8d56490 100644 --- a/drivers/net/wireless/ath/ath9k/ar9002_hw.c +++ b/drivers/net/wireless/ath/ath9k/ar9002_hw.c @@ -309,11 +309,7 @@ static void ar9002_hw_configpcipowersave(struct ath_hw *ah, u8 i; u32 val; - if (ah->is_pciexpress != true) - return; - - /* Do not touch SerDes registers */ - if (ah->config.pcie_powersave_enable == 2) + if (ah->is_pciexpress != true || ah->aspm_enabled != true) return; /* Nothing to do on restore for 11N */ diff --git a/drivers/net/wireless/ath/ath9k/ar9003_hw.c b/drivers/net/wireless/ath/ath9k/ar9003_hw.c index 8efdec247c02..ad2bb2bf4e8a 100644 --- a/drivers/net/wireless/ath/ath9k/ar9003_hw.c +++ b/drivers/net/wireless/ath/ath9k/ar9003_hw.c @@ -519,11 +519,7 @@ static void ar9003_hw_configpcipowersave(struct ath_hw *ah, int restore, int power_off) { - if (ah->is_pciexpress != true) - return; - - /* Do not touch SerDes registers */ - if (ah->config.pcie_powersave_enable == 2) + if (ah->is_pciexpress != true || ah->aspm_enabled != true) return; /* Nothing to do on restore for 11N */ diff --git a/drivers/net/wireless/ath/ath9k/hw.c b/drivers/net/wireless/ath/ath9k/hw.c index 8006ce0c7357..8dcefe74f4c3 100644 --- a/drivers/net/wireless/ath/ath9k/hw.c +++ b/drivers/net/wireless/ath/ath9k/hw.c @@ -318,6 +318,14 @@ static void ath9k_hw_disablepcie(struct ath_hw *ah) REG_WRITE(ah, AR_PCIE_SERDES2, 0x00000000); } +static void ath9k_hw_aspm_init(struct ath_hw *ah) +{ + struct ath_common *common = ath9k_hw_common(ah); + + if (common->bus_ops->aspm_init) + common->bus_ops->aspm_init(common); +} + /* This should work for all families including legacy */ static bool ath9k_hw_chip_test(struct ath_hw *ah) { @@ -378,7 +386,6 @@ static void ath9k_hw_init_config(struct ath_hw *ah) ah->config.additional_swba_backoff = 0; ah->config.ack_6mb = 0x0; ah->config.cwm_ignore_extcca = 0; - ah->config.pcie_powersave_enable = 0; ah->config.pcie_clock_req = 0; ah->config.pcie_waen = 0; ah->config.analog_shiftreg = 1; @@ -598,7 +605,7 @@ static int __ath9k_hw_init(struct ath_hw *ah) if (ah->is_pciexpress) - ath9k_hw_configpcipowersave(ah, 0, 0); + ath9k_hw_aspm_init(ah); else ath9k_hw_disablepcie(ah); diff --git a/drivers/net/wireless/ath/ath9k/hw.h b/drivers/net/wireless/ath/ath9k/hw.h index 6acd0f975ae1..c79889036ec4 100644 --- a/drivers/net/wireless/ath/ath9k/hw.h +++ b/drivers/net/wireless/ath/ath9k/hw.h @@ -219,7 +219,6 @@ struct ath9k_ops_config { int additional_swba_backoff; int ack_6mb; u32 cwm_ignore_extcca; - u8 pcie_powersave_enable; bool pcieSerDesWrite; u8 pcie_clock_req; u32 pcie_waen; @@ -673,6 +672,7 @@ struct ath_hw { bool sw_mgmt_crypto; bool is_pciexpress; + bool aspm_enabled; bool is_monitoring; bool need_an_top2_fixup; u16 tx_trig_level; @@ -874,6 +874,7 @@ struct ath_bus_ops { bool (*eeprom_read)(struct ath_common *common, u32 off, u16 *data); void (*bt_coex_prep)(struct ath_common *common); void (*extn_synch_en)(struct ath_common *common); + void (*aspm_init)(struct ath_common *common); }; static inline struct ath_common *ath9k_hw_common(struct ath_hw *ah) diff --git a/drivers/net/wireless/ath/ath9k/pci.c b/drivers/net/wireless/ath/ath9k/pci.c index 3bad0b2cf9a3..be4ea1329813 100644 --- a/drivers/net/wireless/ath/ath9k/pci.c +++ b/drivers/net/wireless/ath/ath9k/pci.c @@ -16,6 +16,7 @@ #include #include +#include #include #include "ath9k.h" @@ -115,12 +116,38 @@ static void ath_pci_extn_synch_enable(struct ath_common *common) pci_write_config_byte(pdev, sc->sc_ah->caps.pcie_lcr_offset, lnkctl); } +static void ath_pci_aspm_init(struct ath_common *common) +{ + struct ath_softc *sc = (struct ath_softc *) common->priv; + struct ath_hw *ah = sc->sc_ah; + struct pci_dev *pdev = to_pci_dev(sc->dev); + struct pci_dev *parent; + int pos; + u8 aspm; + + if (!pci_is_pcie(pdev)) + return; + + parent = pdev->bus->self; + if (WARN_ON(!parent)) + return; + + pos = pci_pcie_cap(parent); + pci_read_config_byte(parent, pos + PCI_EXP_LNKCTL, &aspm); + if (aspm & (PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1)) { + ah->aspm_enabled = true; + /* Initialize PCIe PM and SERDES registers. */ + ath9k_hw_configpcipowersave(ah, 0, 0); + } +} + static const struct ath_bus_ops ath_pci_bus_ops = { .ath_bus_type = ATH_PCI, .read_cachesize = ath_pci_read_cachesize, .eeprom_read = ath_pci_eeprom_read, .bt_coex_prep = ath_pci_bt_coex_prep, .extn_synch_en = ath_pci_extn_synch_enable, + .aspm_init = ath_pci_aspm_init, }; static int ath_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id) From b6b67df3f24c45af0012ee3c8af2f62ca083ae18 Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Fri, 29 Jul 2011 10:53:12 -0500 Subject: [PATCH 124/439] rtlwifi: Fix kernel oops on ARM SOC This driver uses information from the self member of the pci_bus struct to get information regarding the bridge to which the PCIe device is attached. Unfortunately, this member is not established on all architectures, which leads to a kernel oops. Skipping the entire block that uses the self member to determine the bridge vendor will only affect RTL8192DE devices as that driver sets the ASPM support flag differently when the bridge vendor is Intel. If the self member is available, there is no functional change. This patch fixes Bugzilla No. 40212. Reported-by: Hubert Liao Signed-off-by: Larry Finger Cc: Stable [back to 2.6.38] Signed-off-by: John W. Linville --- drivers/net/wireless/rtlwifi/pci.c | 20 +++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-) diff --git a/drivers/net/wireless/rtlwifi/pci.c b/drivers/net/wireless/rtlwifi/pci.c index 5efd57833489..56f12358389d 100644 --- a/drivers/net/wireless/rtlwifi/pci.c +++ b/drivers/net/wireless/rtlwifi/pci.c @@ -1696,15 +1696,17 @@ static bool _rtl_pci_find_adapter(struct pci_dev *pdev, pcipriv->ndis_adapter.devnumber = PCI_SLOT(pdev->devfn); pcipriv->ndis_adapter.funcnumber = PCI_FUNC(pdev->devfn); - /*find bridge info */ - pcipriv->ndis_adapter.pcibridge_vendorid = bridge_pdev->vendor; - for (tmp = 0; tmp < PCI_BRIDGE_VENDOR_MAX; tmp++) { - if (bridge_pdev->vendor == pcibridge_vendors[tmp]) { - pcipriv->ndis_adapter.pcibridge_vendor = tmp; - RT_TRACE(rtlpriv, COMP_INIT, DBG_DMESG, - ("Pci Bridge Vendor is found index: %d\n", - tmp)); - break; + if (bridge_pdev) { + /*find bridge info if available */ + pcipriv->ndis_adapter.pcibridge_vendorid = bridge_pdev->vendor; + for (tmp = 0; tmp < PCI_BRIDGE_VENDOR_MAX; tmp++) { + if (bridge_pdev->vendor == pcibridge_vendors[tmp]) { + pcipriv->ndis_adapter.pcibridge_vendor = tmp; + RT_TRACE(rtlpriv, COMP_INIT, DBG_DMESG, + ("Pci Bridge Vendor is found index:" + " %d\n", tmp)); + break; + } } } From b52398b6e4522176dd125722c72c301015d24520 Mon Sep 17 00:00:00 2001 From: Stanislaw Gruszka Date: Sat, 30 Jul 2011 13:32:56 +0200 Subject: [PATCH 125/439] rt2x00: rt2800: fix zeroing skb structure We should clear skb->data not skb itself. Bug was introduced by: commit 0b8004aa12d13ec750d102ba4082a95f0107c649 "rt2x00: Properly reserve room for descriptors in skbs". Cc: stable@kernel.org # 2.6.36+ Signed-off-by: Stanislaw Gruszka Acked-by: Gertjan van Wingerde Acked-by: Ivo van Doorn Signed-off-by: John W. Linville --- drivers/net/wireless/rt2x00/rt2800lib.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/net/wireless/rt2x00/rt2800lib.c b/drivers/net/wireless/rt2x00/rt2800lib.c index 75d2c6cc93eb..f94d669ed832 100644 --- a/drivers/net/wireless/rt2x00/rt2800lib.c +++ b/drivers/net/wireless/rt2x00/rt2800lib.c @@ -703,8 +703,7 @@ void rt2800_write_beacon(struct queue_entry *entry, struct txentry_desc *txdesc) /* * Add space for the TXWI in front of the skb. */ - skb_push(entry->skb, TXWI_DESC_SIZE); - memset(entry->skb, 0, TXWI_DESC_SIZE); + memset(skb_push(entry->skb, TXWI_DESC_SIZE), 0, TXWI_DESC_SIZE); /* * Register descriptor details in skb frame descriptor. From a0f98dde11a1afe9fbf5c98f57968e086e98b6f5 Mon Sep 17 00:00:00 2001 From: Wanlong Gao Date: Mon, 18 Jul 2011 12:19:35 +0000 Subject: [PATCH 126/439] Btrfs:don't check the return value of __btrfs_add_inode_defrag Don't need to check the return value of __btrfs_add_inode_defrag(), since it will always return 0. Signed-off-by: Wanlong Gao Signed-off-by: Chris Mason --- fs/btrfs/file.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 41ca5fdaee6c..7f134a730efb 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -74,7 +74,7 @@ struct inode_defrag { * If an existing record is found the defrag item you * pass in is freed */ -static int __btrfs_add_inode_defrag(struct inode *inode, +static void __btrfs_add_inode_defrag(struct inode *inode, struct inode_defrag *defrag) { struct btrfs_root *root = BTRFS_I(inode)->root; @@ -106,11 +106,11 @@ static int __btrfs_add_inode_defrag(struct inode *inode, BTRFS_I(inode)->in_defrag = 1; rb_link_node(&defrag->rb_node, parent, p); rb_insert_color(&defrag->rb_node, &root->fs_info->defrag_inodes); - return 0; + return; exists: kfree(defrag); - return 0; + return; } @@ -123,7 +123,6 @@ int btrfs_add_inode_defrag(struct btrfs_trans_handle *trans, { struct btrfs_root *root = BTRFS_I(inode)->root; struct inode_defrag *defrag; - int ret = 0; u64 transid; if (!btrfs_test_opt(root, AUTO_DEFRAG)) @@ -150,9 +149,9 @@ int btrfs_add_inode_defrag(struct btrfs_trans_handle *trans, spin_lock(&root->fs_info->defrag_inodes_lock); if (!BTRFS_I(inode)->in_defrag) - ret = __btrfs_add_inode_defrag(inode, defrag); + __btrfs_add_inode_defrag(inode, defrag); spin_unlock(&root->fs_info->defrag_inodes_lock); - return ret; + return 0; } /* From b532402e4d147e4f409c4e7f50d4413e8450101d Mon Sep 17 00:00:00 2001 From: Tsutomu Itoh Date: Tue, 19 Jul 2011 07:27:20 +0000 Subject: [PATCH 127/439] Btrfs: return error to caller when btrfs_unlink() failes When btrfs_unlink_inode() and btrfs_orphan_add() in btrfs_unlink() are error, the error code is returned to the caller instead of BUG_ON(). Signed-off-by: Tsutomu Itoh Signed-off-by: Chris Mason --- fs/btrfs/extent-tree.c | 3 ++- fs/btrfs/inode.c | 10 +++++++--- 2 files changed, 9 insertions(+), 4 deletions(-) diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 55bddffede73..db2057f262ad 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -7201,7 +7201,8 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, inode = lookup_free_space_inode(root, block_group, path); if (!IS_ERR(inode)) { - btrfs_orphan_add(trans, inode); + ret = btrfs_orphan_add(trans, inode); + BUG_ON(ret); clear_nlink(inode); /* One for the block groups ref */ spin_lock(&block_group->lock); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 4360ccb191b1..5da43aea205f 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -2217,7 +2217,8 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode) if (!root->orphan_block_rsv) { block_rsv = btrfs_alloc_block_rsv(root); - BUG_ON(!block_rsv); + if (!block_rsv) + return -ENOMEM; } spin_lock(&root->orphan_lock); @@ -3002,13 +3003,16 @@ static int btrfs_unlink(struct inode *dir, struct dentry *dentry) ret = btrfs_unlink_inode(trans, root, dir, dentry->d_inode, dentry->d_name.name, dentry->d_name.len); - BUG_ON(ret); + if (ret) + goto out; if (inode->i_nlink == 0) { ret = btrfs_orphan_add(trans, inode); - BUG_ON(ret); + if (ret) + goto out; } +out: nr = trans->blocks_used; __unlink_end_trans(trans, root); btrfs_btree_balance_dirty(root, nr); From b6973aa62253f3791ef6fa5e9f9de099645fc2bd Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Wed, 20 Jul 2011 03:46:35 +0000 Subject: [PATCH 128/439] Btrfs: fix readahead in file defrag We passed the wrong value to btrfs_force_ra(). Fix this by changing the argument of btrfs_force_ra() from last_index to nr_page. Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/ctree.h | 11 ++++++++--- fs/btrfs/inode.c | 13 ------------- 2 files changed, 8 insertions(+), 16 deletions(-) diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 3be57c611040..7ac4d25fd4f8 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -2520,6 +2520,14 @@ int btrfs_lookup_csums_range(struct btrfs_root *root, u64 start, u64 end, #define PageChecked PageFsMisc #endif +/* This forces readahead on a given range of bytes in an inode */ +static inline void btrfs_force_ra(struct address_space *mapping, + struct file_ra_state *ra, struct file *file, + pgoff_t offset, unsigned long req_size) +{ + page_cache_sync_readahead(mapping, ra, file, offset, req_size); +} + struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry); int btrfs_set_inode_index(struct inode *dir, u64 *index); int btrfs_unlink_inode(struct btrfs_trans_handle *trans, @@ -2548,9 +2556,6 @@ int btrfs_create_subvol_root(struct btrfs_trans_handle *trans, int btrfs_merge_bio_hook(struct page *page, unsigned long offset, size_t size, struct bio *bio, unsigned long bio_flags); -unsigned long btrfs_force_ra(struct address_space *mapping, - struct file_ra_state *ra, struct file *file, - pgoff_t offset, pgoff_t last_index); int btrfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf); int btrfs_readpage(struct file *file, struct page *page); void btrfs_evict_inode(struct inode *inode); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 5da43aea205f..69e448eddf09 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -6697,19 +6697,6 @@ int btrfs_create_subvol_root(struct btrfs_trans_handle *trans, return 0; } -/* helper function for file defrag and space balancing. This - * forces readahead on a given range of bytes in an inode - */ -unsigned long btrfs_force_ra(struct address_space *mapping, - struct file_ra_state *ra, struct file *file, - pgoff_t offset, pgoff_t last_index) -{ - pgoff_t req_size = last_index - offset + 1; - - page_cache_sync_readahead(mapping, ra, file, offset, req_size); - return offset + req_size; -} - struct inode *btrfs_alloc_inode(struct super_block *sb) { struct btrfs_inode *ei; From 1bf85046e493c88be1c1bad9084428373089f618 Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Thu, 21 Jul 2011 16:56:09 +0000 Subject: [PATCH 129/439] btrfs: Make extent-io callbacks that never fail return void The set/clear bit and the extent split/merge hooks only ever return 0. Changing them to return void simplifies the error handling cases later. This patch changes the hook prototypes, the single implementation of each, and the functions that call them to return void instead. Since all four of these hooks execute under a spinlock, they're necessarily simple. Signed-off-by: Jeff Mahoney Signed-off-by: Chris Mason --- fs/btrfs/extent_io.c | 52 ++++++++++++-------------------------------- fs/btrfs/extent_io.h | 18 +++++++-------- fs/btrfs/inode.c | 26 ++++++++++------------ 3 files changed, 34 insertions(+), 62 deletions(-) diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 5bbdb243bb6f..789d0b23048f 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -254,14 +254,14 @@ static void merge_cb(struct extent_io_tree *tree, struct extent_state *new, * * This should be called with the tree lock held. */ -static int merge_state(struct extent_io_tree *tree, - struct extent_state *state) +static void merge_state(struct extent_io_tree *tree, + struct extent_state *state) { struct extent_state *other; struct rb_node *other_node; if (state->state & (EXTENT_IOBITS | EXTENT_BOUNDARY)) - return 0; + return; other_node = rb_prev(&state->rb_node); if (other_node) { @@ -287,19 +287,13 @@ static int merge_state(struct extent_io_tree *tree, free_extent_state(other); } } - - return 0; } -static int set_state_cb(struct extent_io_tree *tree, +static void set_state_cb(struct extent_io_tree *tree, struct extent_state *state, int *bits) { - if (tree->ops && tree->ops->set_bit_hook) { - return tree->ops->set_bit_hook(tree->mapping->host, - state, bits); - } - - return 0; + if (tree->ops && tree->ops->set_bit_hook) + tree->ops->set_bit_hook(tree->mapping->host, state, bits); } static void clear_state_cb(struct extent_io_tree *tree, @@ -325,7 +319,6 @@ static int insert_state(struct extent_io_tree *tree, { struct rb_node *node; int bits_to_set = *bits & ~EXTENT_CTLBITS; - int ret; if (end < start) { printk(KERN_ERR "btrfs end < start %llu %llu\n", @@ -335,9 +328,7 @@ static int insert_state(struct extent_io_tree *tree, } state->start = start; state->end = end; - ret = set_state_cb(tree, state, bits); - if (ret) - return ret; + set_state_cb(tree, state, bits); if (bits_to_set & EXTENT_DIRTY) tree->dirty_bytes += end - start + 1; @@ -357,13 +348,11 @@ static int insert_state(struct extent_io_tree *tree, return 0; } -static int split_cb(struct extent_io_tree *tree, struct extent_state *orig, +static void split_cb(struct extent_io_tree *tree, struct extent_state *orig, u64 split) { if (tree->ops && tree->ops->split_extent_hook) - return tree->ops->split_extent_hook(tree->mapping->host, - orig, split); - return 0; + tree->ops->split_extent_hook(tree->mapping->host, orig, split); } /* @@ -670,23 +659,18 @@ out: return 0; } -static int set_state_bits(struct extent_io_tree *tree, +static void set_state_bits(struct extent_io_tree *tree, struct extent_state *state, int *bits) { - int ret; int bits_to_set = *bits & ~EXTENT_CTLBITS; - ret = set_state_cb(tree, state, bits); - if (ret) - return ret; + set_state_cb(tree, state, bits); if ((bits_to_set & EXTENT_DIRTY) && !(state->state & EXTENT_DIRTY)) { u64 range = state->end - state->start + 1; tree->dirty_bytes += range; } state->state |= bits_to_set; - - return 0; } static void cache_state(struct extent_state *state, @@ -779,9 +763,7 @@ hit_next: goto out; } - err = set_state_bits(tree, state, &bits); - if (err) - goto out; + set_state_bits(tree, state, &bits); cache_state(state, cached_state); merge_state(tree, state); @@ -830,9 +812,7 @@ hit_next: if (err) goto out; if (state->end <= end) { - err = set_state_bits(tree, state, &bits); - if (err) - goto out; + set_state_bits(tree, state, &bits); cache_state(state, cached_state); merge_state(tree, state); if (last_end == (u64)-1) @@ -893,11 +873,7 @@ hit_next: err = split_state(tree, state, prealloc, end + 1); BUG_ON(err == -EEXIST); - err = set_state_bits(tree, prealloc, &bits); - if (err) { - prealloc = NULL; - goto out; - } + set_state_bits(tree, prealloc, &bits); cache_state(prealloc, cached_state); merge_state(tree, prealloc); prealloc = NULL; diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index 21a7ca9e7282..d6871dc7524f 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -76,15 +76,15 @@ struct extent_io_ops { struct extent_state *state); int (*writepage_end_io_hook)(struct page *page, u64 start, u64 end, struct extent_state *state, int uptodate); - int (*set_bit_hook)(struct inode *inode, struct extent_state *state, - int *bits); - int (*clear_bit_hook)(struct inode *inode, struct extent_state *state, - int *bits); - int (*merge_extent_hook)(struct inode *inode, - struct extent_state *new, - struct extent_state *other); - int (*split_extent_hook)(struct inode *inode, - struct extent_state *orig, u64 split); + void (*set_bit_hook)(struct inode *inode, struct extent_state *state, + int *bits); + void (*clear_bit_hook)(struct inode *inode, struct extent_state *state, + int *bits); + void (*merge_extent_hook)(struct inode *inode, + struct extent_state *new, + struct extent_state *other); + void (*split_extent_hook)(struct inode *inode, + struct extent_state *orig, u64 split); int (*write_cache_pages_lock_hook)(struct page *page); }; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 69e448eddf09..34195f9fc6bb 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1283,17 +1283,16 @@ static int run_delalloc_range(struct inode *inode, struct page *locked_page, return ret; } -static int btrfs_split_extent_hook(struct inode *inode, - struct extent_state *orig, u64 split) +static void btrfs_split_extent_hook(struct inode *inode, + struct extent_state *orig, u64 split) { /* not delalloc, ignore it */ if (!(orig->state & EXTENT_DELALLOC)) - return 0; + return; spin_lock(&BTRFS_I(inode)->lock); BTRFS_I(inode)->outstanding_extents++; spin_unlock(&BTRFS_I(inode)->lock); - return 0; } /* @@ -1302,18 +1301,17 @@ static int btrfs_split_extent_hook(struct inode *inode, * extents, such as when we are doing sequential writes, so we can properly * account for the metadata space we'll need. */ -static int btrfs_merge_extent_hook(struct inode *inode, - struct extent_state *new, - struct extent_state *other) +static void btrfs_merge_extent_hook(struct inode *inode, + struct extent_state *new, + struct extent_state *other) { /* not delalloc, ignore it */ if (!(other->state & EXTENT_DELALLOC)) - return 0; + return; spin_lock(&BTRFS_I(inode)->lock); BTRFS_I(inode)->outstanding_extents--; spin_unlock(&BTRFS_I(inode)->lock); - return 0; } /* @@ -1321,8 +1319,8 @@ static int btrfs_merge_extent_hook(struct inode *inode, * bytes in this file, and to maintain the list of inodes that * have pending delalloc work to be done. */ -static int btrfs_set_bit_hook(struct inode *inode, - struct extent_state *state, int *bits) +static void btrfs_set_bit_hook(struct inode *inode, + struct extent_state *state, int *bits) { /* @@ -1352,14 +1350,13 @@ static int btrfs_set_bit_hook(struct inode *inode, } spin_unlock(&root->fs_info->delalloc_lock); } - return 0; } /* * extent_io.c clear_bit_hook, see set_bit_hook for why */ -static int btrfs_clear_bit_hook(struct inode *inode, - struct extent_state *state, int *bits) +static void btrfs_clear_bit_hook(struct inode *inode, + struct extent_state *state, int *bits) { /* * set_bit and clear bit hooks normally require _irqsave/restore @@ -1396,7 +1393,6 @@ static int btrfs_clear_bit_hook(struct inode *inode, } spin_unlock(&root->fs_info->delalloc_lock); } - return 0; } /* From 61cfea9bb84d41b09e59822c33b3eb27acc48358 Mon Sep 17 00:00:00 2001 From: WuBo Date: Tue, 26 Jul 2011 03:30:11 +0000 Subject: [PATCH 130/439] Btrfs: Protect the readonly flag of block group The access for ro in btrfs_block_group_cache should be protected because of the racy lock in relocation. Signed-off-by: Wu Bo Signed-off-by: Chris Mason --- fs/btrfs/extent-tree.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index db2057f262ad..a4db88da125e 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -6549,8 +6549,6 @@ static int set_block_group_ro(struct btrfs_block_group_cache *cache, int force) u64 min_allocable_bytes; int ret = -ENOSPC; - if (cache->ro) - return 0; /* * We need some metadata space and system metadata space for @@ -6566,6 +6564,12 @@ static int set_block_group_ro(struct btrfs_block_group_cache *cache, int force) spin_lock(&sinfo->lock); spin_lock(&cache->lock); + + if (cache->ro) { + ret = 0; + goto out; + } + num_bytes = cache->key.offset - cache->reserved - cache->pinned - cache->bytes_super - btrfs_block_group_used(&cache->item); @@ -6579,7 +6583,7 @@ static int set_block_group_ro(struct btrfs_block_group_cache *cache, int force) cache->ro = 1; ret = 0; } - +out: spin_unlock(&cache->lock); spin_unlock(&sinfo->lock); return ret; From ff1f2b4407454d926d8b143bf37e4dce66026a28 Mon Sep 17 00:00:00 2001 From: liubo Date: Wed, 27 Jul 2011 09:49:18 +0000 Subject: [PATCH 131/439] Btrfs: fix oops while writing data to SSD partitions Here I have a two SSD-partitions btrfs, and they are defaultly set to "data=raid0, metadata=raid1", then I try to fill my btrfs partition till "No space left on device", via "dd if=/dev/zero of=/mnt/btrfs/tmp". I get an oops panic from kernel BUG at fs/btrfs/extent-tree.c:5199!, which refers to find_free_extent's BUG_ON(index != get_block_group_index(block_group)); In SSD mode, in order to find enough space to alloc, we may check the block_group cache which has been checked sometime before, but the index is not updated, where it hits the BUG_ON. Signed-off-by: Liu Bo Acked-by: Josef Bacik Signed-off-by: Chris Mason --- fs/btrfs/extent-tree.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index a4db88da125e..b172b7fa4889 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -5079,7 +5079,9 @@ have_block_group: * group is does point to and try again */ if (!last_ptr_loop && last_ptr->block_group && - last_ptr->block_group != block_group) { + last_ptr->block_group != block_group && + index <= + get_block_group_index(last_ptr->block_group)) { btrfs_put_block_group(block_group); block_group = last_ptr->block_group; From bf5f32ecb6caac52b4d1c083251b3dd4f40a0b7a Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Thu, 14 Jul 2011 21:23:06 +0000 Subject: [PATCH 132/439] btrfs: make btrfs_set_root_node void This is fairly trivial - btrfs_set_root_node() - always returns zero so we can just make it void. All callers ignore the return code now anyway. I also made sure to check that none of the functions that btrfs_set_root_node() calls returns an error that we might have needed to catch and pass back. Signed-off-by: Mark Fasheh Signed-off-by: Chris Mason --- fs/btrfs/ctree.h | 4 ++-- fs/btrfs/root-tree.c | 5 ++--- 2 files changed, 4 insertions(+), 5 deletions(-) diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 7ac4d25fd4f8..34ce4140bb41 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -2406,8 +2406,8 @@ int btrfs_find_last_root(struct btrfs_root *root, u64 objectid, struct btrfs_root_item *item, struct btrfs_key *key); int btrfs_find_dead_roots(struct btrfs_root *root, u64 objectid); int btrfs_find_orphan_roots(struct btrfs_root *tree_root); -int btrfs_set_root_node(struct btrfs_root_item *item, - struct extent_buffer *node); +void btrfs_set_root_node(struct btrfs_root_item *item, + struct extent_buffer *node); void btrfs_check_and_init_root_item(struct btrfs_root_item *item); /* dir-item.c */ diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c index ebe45443de06..f4099904565a 100644 --- a/fs/btrfs/root-tree.c +++ b/fs/btrfs/root-tree.c @@ -71,13 +71,12 @@ out: return ret; } -int btrfs_set_root_node(struct btrfs_root_item *item, - struct extent_buffer *node) +void btrfs_set_root_node(struct btrfs_root_item *item, + struct extent_buffer *node) { btrfs_set_root_bytenr(item, node->start); btrfs_set_root_level(item, btrfs_header_level(node)); btrfs_set_root_generation(item, btrfs_header_generation(node)); - return 0; } /* From b783e62d9620445d1ca3de76badf0468cdd329ba Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Wed, 13 Jul 2011 15:03:50 +0000 Subject: [PATCH 133/439] Btrfs: don't print the leaf if we had an error In __btrfs_free_extent we will print the leaf if we fail to find the extent we wanted, but the problem is if we get an error we won't have a leaf so often this leads to a NULL pointer dereference and we lose the error that actually occurred. So only print the leaf if ret > 0, which means we didn't find the item we were looking for but we didn't error either. This way the error is preserved. Signed-off-by: Josef Bacik Signed-off-by: Chris Mason --- fs/btrfs/extent-tree.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index b172b7fa4889..66bac226944e 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -4462,7 +4462,9 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, printk(KERN_ERR "umm, got %d back from search" ", was looking for %llu\n", ret, (unsigned long long)bytenr); - btrfs_print_leaf(extent_root, path->nodes[0]); + if (ret > 0) + btrfs_print_leaf(extent_root, + path->nodes[0]); } BUG_ON(ret); extent_slot = path->slots[0]; From 77906a5075a4eb767026c2e07b1a412d08aea5be Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Thu, 14 Jul 2011 03:16:00 +0000 Subject: [PATCH 134/439] Btrfs: copy string correctly in INO_LOOKUP ioctl Memory areas [ptr, ptr+total_len] and [name, name+total_len] may overlap, so it's wrong to use memcpy(). Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/ioctl.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index fd252fff4c66..2bb08862a4f6 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -1757,11 +1757,10 @@ static noinline int btrfs_search_path_in_tree(struct btrfs_fs_info *info, key.objectid = key.offset; key.offset = (u64)-1; dirid = key.objectid; - } if (ptr < name) goto out; - memcpy(name, ptr, total_len); + memmove(name, ptr, total_len); name[total_len]='\0'; ret = 0; out: From e55179b3d7d41d83fd6b5f59325f4a8d8ac9700a Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Thu, 14 Jul 2011 03:16:47 +0000 Subject: [PATCH 135/439] Btrfs: check the nodatasum flag when writing compressed files If mounting with nodatasum option, we won't csum file data for general write or direct-io write, and this rule should also be applied when writing compressed files. Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/compression.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c index bfe42b03eaf9..8ec5d86f1734 100644 --- a/fs/btrfs/compression.c +++ b/fs/btrfs/compression.c @@ -338,6 +338,7 @@ int btrfs_submit_compressed_write(struct inode *inode, u64 start, u64 first_byte = disk_start; struct block_device *bdev; int ret; + int skip_sum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM; WARN_ON(start & ((u64)PAGE_CACHE_SIZE - 1)); cb = kmalloc(compressed_bio_size(root, compressed_len), GFP_NOFS); @@ -392,8 +393,11 @@ int btrfs_submit_compressed_write(struct inode *inode, u64 start, ret = btrfs_bio_wq_end_io(root->fs_info, bio, 0); BUG_ON(ret); - ret = btrfs_csum_one_bio(root, inode, bio, start, 1); - BUG_ON(ret); + if (!skip_sum) { + ret = btrfs_csum_one_bio(root, inode, bio, + start, 1); + BUG_ON(ret); + } ret = btrfs_map_bio(root, WRITE, bio, 0, 1); BUG_ON(ret); @@ -418,8 +422,10 @@ int btrfs_submit_compressed_write(struct inode *inode, u64 start, ret = btrfs_bio_wq_end_io(root->fs_info, bio, 0); BUG_ON(ret); - ret = btrfs_csum_one_bio(root, inode, bio, start, 1); - BUG_ON(ret); + if (!skip_sum) { + ret = btrfs_csum_one_bio(root, inode, bio, start, 1); + BUG_ON(ret); + } ret = btrfs_map_bio(root, WRITE, bio, 0, 1); BUG_ON(ret); From 72d63ed6427cf233e2b352c0b80c3e5c5a444986 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Thu, 14 Jul 2011 03:17:00 +0000 Subject: [PATCH 136/439] Btrfs: use wait_event() Use wait_event() when possible to avoid code duplication. Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/transaction.c | 59 +++++------------------------------------- 1 file changed, 7 insertions(+), 52 deletions(-) diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index eb55863bb4ae..ff5549fe4624 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -216,17 +216,11 @@ static void wait_current_trans(struct btrfs_root *root) spin_lock(&root->fs_info->trans_lock); cur_trans = root->fs_info->running_transaction; if (cur_trans && cur_trans->blocked) { - DEFINE_WAIT(wait); atomic_inc(&cur_trans->use_count); spin_unlock(&root->fs_info->trans_lock); - while (1) { - prepare_to_wait(&root->fs_info->transaction_wait, &wait, - TASK_UNINTERRUPTIBLE); - if (!cur_trans->blocked) - break; - schedule(); - } - finish_wait(&root->fs_info->transaction_wait, &wait); + + wait_event(root->fs_info->transaction_wait, + !cur_trans->blocked); put_transaction(cur_trans); } else { spin_unlock(&root->fs_info->trans_lock); @@ -360,15 +354,7 @@ struct btrfs_trans_handle *btrfs_start_ioctl_transaction(struct btrfs_root *root static noinline int wait_for_commit(struct btrfs_root *root, struct btrfs_transaction *commit) { - DEFINE_WAIT(wait); - while (!commit->commit_done) { - prepare_to_wait(&commit->commit_wait, &wait, - TASK_UNINTERRUPTIBLE); - if (commit->commit_done) - break; - schedule(); - } - finish_wait(&commit->commit_wait, &wait); + wait_event(commit->commit_wait, commit->commit_done); return 0; } @@ -1085,22 +1071,7 @@ int btrfs_transaction_blocked(struct btrfs_fs_info *info) static void wait_current_trans_commit_start(struct btrfs_root *root, struct btrfs_transaction *trans) { - DEFINE_WAIT(wait); - - if (trans->in_commit) - return; - - while (1) { - prepare_to_wait(&root->fs_info->transaction_blocked_wait, &wait, - TASK_UNINTERRUPTIBLE); - if (trans->in_commit) { - finish_wait(&root->fs_info->transaction_blocked_wait, - &wait); - break; - } - schedule(); - finish_wait(&root->fs_info->transaction_blocked_wait, &wait); - } + wait_event(root->fs_info->transaction_blocked_wait, trans->in_commit); } /* @@ -1110,24 +1081,8 @@ static void wait_current_trans_commit_start(struct btrfs_root *root, static void wait_current_trans_commit_start_and_unblock(struct btrfs_root *root, struct btrfs_transaction *trans) { - DEFINE_WAIT(wait); - - if (trans->commit_done || (trans->in_commit && !trans->blocked)) - return; - - while (1) { - prepare_to_wait(&root->fs_info->transaction_wait, &wait, - TASK_UNINTERRUPTIBLE); - if (trans->commit_done || - (trans->in_commit && !trans->blocked)) { - finish_wait(&root->fs_info->transaction_wait, - &wait); - break; - } - schedule(); - finish_wait(&root->fs_info->transaction_wait, - &wait); - } + wait_event(root->fs_info->transaction_wait, + trans->commit_done || (trans->in_commit && !trans->blocked)); } /* From b9c8300c2ac354d850159f301d5b3ead13854cdd Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Thu, 14 Jul 2011 03:17:14 +0000 Subject: [PATCH 137/439] Btrfs: remove a BUG_ON() in btrfs_commit_transaction() wait_for_commit() always returns 0. Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/transaction.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index ff5549fe4624..7dc36fab4afc 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -351,11 +351,10 @@ struct btrfs_trans_handle *btrfs_start_ioctl_transaction(struct btrfs_root *root } /* wait for a transaction commit to be fully complete */ -static noinline int wait_for_commit(struct btrfs_root *root, +static noinline void wait_for_commit(struct btrfs_root *root, struct btrfs_transaction *commit) { wait_event(commit->commit_wait, commit->commit_done); - return 0; } int btrfs_wait_for_commit(struct btrfs_root *root, u64 transid) @@ -1189,8 +1188,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans, atomic_inc(&cur_trans->use_count); btrfs_end_transaction(trans, root); - ret = wait_for_commit(root, cur_trans); - BUG_ON(ret); + wait_for_commit(root, cur_trans); put_transaction(cur_trans); From 15de900d08bb132833c8622610ddb1be660a6018 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Thu, 14 Jul 2011 03:17:27 +0000 Subject: [PATCH 138/439] Btrfs: remove remaining ref-cache code Since commit f2a97a9dbd86eb1ef956bdf20e05c507b32beb96 ("btrfs: remove all unused functions"), there's no extern functions at all in ref-cache.c, so just remove the remaining dead code. Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/ref-cache.c | 68 -------------------------------------------- fs/btrfs/ref-cache.h | 52 --------------------------------- 2 files changed, 120 deletions(-) delete mode 100644 fs/btrfs/ref-cache.c delete mode 100644 fs/btrfs/ref-cache.h diff --git a/fs/btrfs/ref-cache.c b/fs/btrfs/ref-cache.c deleted file mode 100644 index 82d569cb6267..000000000000 --- a/fs/btrfs/ref-cache.c +++ /dev/null @@ -1,68 +0,0 @@ -/* - * Copyright (C) 2008 Oracle. 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 v2 as published by the Free Software Foundation. - * - * 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 021110-1307, USA. - */ - -#include -#include -#include -#include "ctree.h" -#include "ref-cache.h" -#include "transaction.h" - -static struct rb_node *tree_insert(struct rb_root *root, u64 bytenr, - struct rb_node *node) -{ - struct rb_node **p = &root->rb_node; - struct rb_node *parent = NULL; - struct btrfs_leaf_ref *entry; - - while (*p) { - parent = *p; - entry = rb_entry(parent, struct btrfs_leaf_ref, rb_node); - - if (bytenr < entry->bytenr) - p = &(*p)->rb_left; - else if (bytenr > entry->bytenr) - p = &(*p)->rb_right; - else - return parent; - } - - entry = rb_entry(node, struct btrfs_leaf_ref, rb_node); - rb_link_node(node, parent, p); - rb_insert_color(node, root); - return NULL; -} - -static struct rb_node *tree_search(struct rb_root *root, u64 bytenr) -{ - struct rb_node *n = root->rb_node; - struct btrfs_leaf_ref *entry; - - while (n) { - entry = rb_entry(n, struct btrfs_leaf_ref, rb_node); - WARN_ON(!entry->in_tree); - - if (bytenr < entry->bytenr) - n = n->rb_left; - else if (bytenr > entry->bytenr) - n = n->rb_right; - else - return n; - } - return NULL; -} diff --git a/fs/btrfs/ref-cache.h b/fs/btrfs/ref-cache.h deleted file mode 100644 index 24f7001f6387..000000000000 --- a/fs/btrfs/ref-cache.h +++ /dev/null @@ -1,52 +0,0 @@ -/* - * Copyright (C) 2008 Oracle. 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 v2 as published by the Free Software Foundation. - * - * 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 021110-1307, USA. - */ -#ifndef __REFCACHE__ -#define __REFCACHE__ - -struct btrfs_extent_info { - /* bytenr and num_bytes find the extent in the extent allocation tree */ - u64 bytenr; - u64 num_bytes; - - /* objectid and offset find the back reference for the file */ - u64 objectid; - u64 offset; -}; - -struct btrfs_leaf_ref { - struct rb_node rb_node; - struct btrfs_leaf_ref_tree *tree; - int in_tree; - atomic_t usage; - - u64 root_gen; - u64 bytenr; - u64 owner; - u64 generation; - int nritems; - - struct list_head list; - struct btrfs_extent_info extents[]; -}; - -static inline size_t btrfs_leaf_ref_size(int nr_extents) -{ - return sizeof(struct btrfs_leaf_ref) + - sizeof(struct btrfs_extent_info) * nr_extents; -} -#endif From 9b89d95a143bb0a9abc4ba0fdcdda78211930f1a Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Thu, 14 Jul 2011 03:17:39 +0000 Subject: [PATCH 139/439] Btrfs: make acl functions really no-op if acl is not enabled So there's no overhead for something we don't use. Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/Makefile | 4 +++- fs/btrfs/acl.c | 17 ----------------- fs/btrfs/ctree.h | 15 ++++++++++++--- 3 files changed, 15 insertions(+), 21 deletions(-) diff --git a/fs/btrfs/Makefile b/fs/btrfs/Makefile index 9b72dcf1cd25..40e6ac08c21f 100644 --- a/fs/btrfs/Makefile +++ b/fs/btrfs/Makefile @@ -6,5 +6,7 @@ btrfs-y += super.o ctree.o extent-tree.o print-tree.o root-tree.o dir-item.o \ transaction.o inode.o file.o tree-defrag.o \ extent_map.o sysfs.o struct-funcs.o xattr.o ordered-data.o \ extent_io.o volumes.o async-thread.o ioctl.o locking.o orphan.o \ - export.o tree-log.o acl.o free-space-cache.o zlib.o lzo.o \ + export.o tree-log.o free-space-cache.o zlib.o lzo.o \ compression.o delayed-ref.o relocation.o delayed-inode.o scrub.o + +btrfs-$(CONFIG_BTRFS_FS_POSIX_ACL) += acl.o diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c index f66fc9959733..b206d4c2720b 100644 --- a/fs/btrfs/acl.c +++ b/fs/btrfs/acl.c @@ -28,8 +28,6 @@ #include "btrfs_inode.h" #include "xattr.h" -#ifdef CONFIG_BTRFS_FS_POSIX_ACL - static struct posix_acl *btrfs_get_acl(struct inode *inode, int type) { int size; @@ -318,18 +316,3 @@ const struct xattr_handler btrfs_xattr_acl_access_handler = { .get = btrfs_xattr_acl_get, .set = btrfs_xattr_acl_set, }; - -#else /* CONFIG_BTRFS_FS_POSIX_ACL */ - -int btrfs_acl_chmod(struct inode *inode) -{ - return 0; -} - -int btrfs_init_acl(struct btrfs_trans_handle *trans, - struct inode *inode, struct inode *dir) -{ - return 0; -} - -#endif /* CONFIG_BTRFS_FS_POSIX_ACL */ diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 34ce4140bb41..a6263bdab818 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -2650,12 +2650,21 @@ do { \ /* acl.c */ #ifdef CONFIG_BTRFS_FS_POSIX_ACL int btrfs_check_acl(struct inode *inode, int mask, unsigned int flags); -#else -#define btrfs_check_acl NULL -#endif int btrfs_init_acl(struct btrfs_trans_handle *trans, struct inode *inode, struct inode *dir); int btrfs_acl_chmod(struct inode *inode); +#else +#define btrfs_check_acl NULL +static inline int btrfs_init_acl(struct btrfs_trans_handle *trans, + struct inode *inode, struct inode *dir) +{ + return 0; +} +static inline int btrfs_acl_chmod(struct inode *inode) +{ + return 0; +} +#endif /* relocation.c */ int btrfs_relocate_block_group(struct btrfs_root *root, u64 group_start); From 85d85a743da894029723e002eb556ceeebc03658 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Thu, 14 Jul 2011 03:17:52 +0000 Subject: [PATCH 140/439] Btrfs: remove redundant code for dir item lookup When we search a dir item with a specific hash code, we can just return NULL without further checking if btrfs_search_slot() returns 1. Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/dir-item.c | 30 ++---------------------------- 1 file changed, 2 insertions(+), 28 deletions(-) diff --git a/fs/btrfs/dir-item.c b/fs/btrfs/dir-item.c index c360a848d97f..31d84e78129b 100644 --- a/fs/btrfs/dir-item.c +++ b/fs/btrfs/dir-item.c @@ -198,8 +198,6 @@ struct btrfs_dir_item *btrfs_lookup_dir_item(struct btrfs_trans_handle *trans, struct btrfs_key key; int ins_len = mod < 0 ? -1 : 0; int cow = mod != 0; - struct btrfs_key found_key; - struct extent_buffer *leaf; key.objectid = dir; btrfs_set_key_type(&key, BTRFS_DIR_ITEM_KEY); @@ -209,18 +207,7 @@ struct btrfs_dir_item *btrfs_lookup_dir_item(struct btrfs_trans_handle *trans, ret = btrfs_search_slot(trans, root, &key, path, ins_len, cow); if (ret < 0) return ERR_PTR(ret); - if (ret > 0) { - if (path->slots[0] == 0) - return NULL; - path->slots[0]--; - } - - leaf = path->nodes[0]; - btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]); - - if (found_key.objectid != dir || - btrfs_key_type(&found_key) != BTRFS_DIR_ITEM_KEY || - found_key.offset != key.offset) + if (ret > 0) return NULL; return btrfs_match_dir_item_name(root, path, name, name_len); @@ -315,8 +302,6 @@ struct btrfs_dir_item *btrfs_lookup_xattr(struct btrfs_trans_handle *trans, struct btrfs_key key; int ins_len = mod < 0 ? -1 : 0; int cow = mod != 0; - struct btrfs_key found_key; - struct extent_buffer *leaf; key.objectid = dir; btrfs_set_key_type(&key, BTRFS_XATTR_ITEM_KEY); @@ -324,18 +309,7 @@ struct btrfs_dir_item *btrfs_lookup_xattr(struct btrfs_trans_handle *trans, ret = btrfs_search_slot(trans, root, &key, path, ins_len, cow); if (ret < 0) return ERR_PTR(ret); - if (ret > 0) { - if (path->slots[0] == 0) - return NULL; - path->slots[0]--; - } - - leaf = path->nodes[0]; - btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]); - - if (found_key.objectid != dir || - btrfs_key_type(&found_key) != BTRFS_XATTR_ITEM_KEY || - found_key.offset != key.offset) + if (ret > 0) return NULL; return btrfs_match_dir_item_name(root, path, name, name_len); From 7e016a038e829c7d1271e1d57b8002860bbdf0db Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Thu, 14 Jul 2011 03:18:03 +0000 Subject: [PATCH 141/439] Btrfs: clean up search_extent_mapping() rb_node returned by __tree_search() can be a valid pointer or NULL, but won't be some errno. Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/extent_map.c | 17 +++-------------- 1 file changed, 3 insertions(+), 14 deletions(-) diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index 2d0410344ea3..911a9db801e0 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -379,23 +379,12 @@ struct extent_map *search_extent_mapping(struct extent_map_tree *tree, em = rb_entry(next, struct extent_map, rb_node); goto found; } - if (!rb_node) { - em = NULL; - goto out; - } - if (IS_ERR(rb_node)) { - em = ERR_CAST(rb_node); - goto out; - } + if (!rb_node) + return NULL; + em = rb_entry(rb_node, struct extent_map, rb_node); - goto found; - - em = NULL; - goto out; - found: atomic_inc(&em->refs); -out: return em; } From ed64f06652210b4a52fe0ea65ac43f9c6af1d988 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Thu, 14 Jul 2011 03:18:15 +0000 Subject: [PATCH 142/439] Btrfs: clean up code for extent_map lookup lookup_extent_map() and search_extent_map() can share most of code. Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/extent_map.c | 87 +++++++++++++++---------------------------- 1 file changed, 30 insertions(+), 57 deletions(-) diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index 911a9db801e0..df7a803005fb 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -299,6 +299,34 @@ static u64 range_end(u64 start, u64 len) return start + len; } +struct extent_map *__lookup_extent_mapping(struct extent_map_tree *tree, + u64 start, u64 len, int strict) +{ + struct extent_map *em; + struct rb_node *rb_node; + struct rb_node *prev = NULL; + struct rb_node *next = NULL; + u64 end = range_end(start, len); + + rb_node = __tree_search(&tree->map, start, &prev, &next); + if (!rb_node) { + if (prev) + rb_node = prev; + else if (next) + rb_node = next; + else + return NULL; + } + + em = rb_entry(rb_node, struct extent_map, rb_node); + + if (strict && !(end > em->start && start < extent_map_end(em))) + return NULL; + + atomic_inc(&em->refs); + return em; +} + /** * lookup_extent_mapping - lookup extent_map * @tree: tree to lookup in @@ -313,42 +341,7 @@ static u64 range_end(u64 start, u64 len) struct extent_map *lookup_extent_mapping(struct extent_map_tree *tree, u64 start, u64 len) { - struct extent_map *em; - struct rb_node *rb_node; - struct rb_node *prev = NULL; - struct rb_node *next = NULL; - u64 end = range_end(start, len); - - rb_node = __tree_search(&tree->map, start, &prev, &next); - if (!rb_node && prev) { - em = rb_entry(prev, struct extent_map, rb_node); - if (end > em->start && start < extent_map_end(em)) - goto found; - } - if (!rb_node && next) { - em = rb_entry(next, struct extent_map, rb_node); - if (end > em->start && start < extent_map_end(em)) - goto found; - } - if (!rb_node) { - em = NULL; - goto out; - } - if (IS_ERR(rb_node)) { - em = ERR_CAST(rb_node); - goto out; - } - em = rb_entry(rb_node, struct extent_map, rb_node); - if (end > em->start && start < extent_map_end(em)) - goto found; - - em = NULL; - goto out; - -found: - atomic_inc(&em->refs); -out: - return em; + return __lookup_extent_mapping(tree, start, len, 1); } /** @@ -365,27 +358,7 @@ out: struct extent_map *search_extent_mapping(struct extent_map_tree *tree, u64 start, u64 len) { - struct extent_map *em; - struct rb_node *rb_node; - struct rb_node *prev = NULL; - struct rb_node *next = NULL; - - rb_node = __tree_search(&tree->map, start, &prev, &next); - if (!rb_node && prev) { - em = rb_entry(prev, struct extent_map, rb_node); - goto found; - } - if (!rb_node && next) { - em = rb_entry(next, struct extent_map, rb_node); - goto found; - } - if (!rb_node) - return NULL; - - em = rb_entry(rb_node, struct extent_map, rb_node); -found: - atomic_inc(&em->refs); - return em; + return __lookup_extent_mapping(tree, start, len, 0); } /** From 4d2c8f62f12a6652db67cc0c1f4a4a498b05ddbc Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Thu, 14 Jul 2011 03:18:33 +0000 Subject: [PATCH 143/439] Btrfs: clean up code for merging extent maps unpin_extent_cache() and add_extent_mapping() shares the same code that merges extent maps. Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/extent_map.c | 59 +++++++++++++++---------------------------- 1 file changed, 21 insertions(+), 38 deletions(-) diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index df7a803005fb..7c97b3301459 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -183,22 +183,10 @@ static int mergable_maps(struct extent_map *prev, struct extent_map *next) return 0; } -int unpin_extent_cache(struct extent_map_tree *tree, u64 start, u64 len) +static void try_merge_map(struct extent_map_tree *tree, struct extent_map *em) { - int ret = 0; struct extent_map *merge = NULL; struct rb_node *rb; - struct extent_map *em; - - write_lock(&tree->lock); - em = lookup_extent_mapping(tree, start, len); - - WARN_ON(!em || em->start != start); - - if (!em) - goto out; - - clear_bit(EXTENT_FLAG_PINNED, &em->flags); if (em->start != 0) { rb = rb_prev(&em->rb_node); @@ -225,6 +213,24 @@ int unpin_extent_cache(struct extent_map_tree *tree, u64 start, u64 len) merge->in_tree = 0; free_extent_map(merge); } +} + +int unpin_extent_cache(struct extent_map_tree *tree, u64 start, u64 len) +{ + int ret = 0; + struct extent_map *em; + + write_lock(&tree->lock); + em = lookup_extent_mapping(tree, start, len); + + WARN_ON(!em || em->start != start); + + if (!em) + goto out; + + clear_bit(EXTENT_FLAG_PINNED, &em->flags); + + try_merge_map(tree, em); free_extent_map(em); out: @@ -247,7 +253,6 @@ int add_extent_mapping(struct extent_map_tree *tree, struct extent_map *em) { int ret = 0; - struct extent_map *merge = NULL; struct rb_node *rb; struct extent_map *exist; @@ -263,30 +268,8 @@ int add_extent_mapping(struct extent_map_tree *tree, goto out; } atomic_inc(&em->refs); - if (em->start != 0) { - rb = rb_prev(&em->rb_node); - if (rb) - merge = rb_entry(rb, struct extent_map, rb_node); - if (rb && mergable_maps(merge, em)) { - em->start = merge->start; - em->len += merge->len; - em->block_len += merge->block_len; - em->block_start = merge->block_start; - merge->in_tree = 0; - rb_erase(&merge->rb_node, &tree->map); - free_extent_map(merge); - } - } - rb = rb_next(&em->rb_node); - if (rb) - merge = rb_entry(rb, struct extent_map, rb_node); - if (rb && mergable_maps(em, merge)) { - em->len += merge->len; - em->block_len += merge->len; - rb_erase(&merge->rb_node, &tree->map); - merge->in_tree = 0; - free_extent_map(merge); - } + + try_merge_map(tree, em); out: return ret; } From 3a6d457ec79d4cdf2313189b4e852e53f2b8d2b2 Mon Sep 17 00:00:00 2001 From: Xiao Guangrong Date: Thu, 14 Jul 2011 03:18:52 +0000 Subject: [PATCH 144/439] Btrfs: remove unused members from struct extent_state These members are not used at all. Signed-off-by: Xiao Guangrong Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/extent_io.h | 2 -- 1 file changed, 2 deletions(-) diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index d6871dc7524f..7b2f0c3e7929 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -108,8 +108,6 @@ struct extent_state { wait_queue_head_t wq; atomic_t refs; unsigned long state; - u64 split_start; - u64 split_end; /* for use by the FS */ u64 private; From 3150b6996934455d7f2da243939d80a9b839085a Mon Sep 17 00:00:00 2001 From: Xiao Guangrong Date: Thu, 14 Jul 2011 03:19:08 +0000 Subject: [PATCH 145/439] Btrfs: clean up for insert_state() Don't duplicate set_state_bits(). Signed-off-by: Xiao Guangrong Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/extent_io.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 789d0b23048f..0d69425d0197 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -303,6 +303,9 @@ static void clear_state_cb(struct extent_io_tree *tree, tree->ops->clear_bit_hook(tree->mapping->host, state, bits); } +static void set_state_bits(struct extent_io_tree *tree, + struct extent_state *state, int *bits); + /* * insert an extent_state struct into the tree. 'bits' are set on the * struct before it is inserted. @@ -318,7 +321,6 @@ static int insert_state(struct extent_io_tree *tree, int *bits) { struct rb_node *node; - int bits_to_set = *bits & ~EXTENT_CTLBITS; if (end < start) { printk(KERN_ERR "btrfs end < start %llu %llu\n", @@ -328,11 +330,9 @@ static int insert_state(struct extent_io_tree *tree, } state->start = start; state->end = end; - set_state_cb(tree, state, bits); - if (bits_to_set & EXTENT_DIRTY) - tree->dirty_bytes += end - start + 1; - state->state |= bits_to_set; + set_state_bits(tree, state, bits); + node = tree_insert(&tree->state, end, &state->rb_node); if (node) { struct extent_state *found; From ded91f0814a349e36b3ca0eff497f816ad518fba Mon Sep 17 00:00:00 2001 From: Xiao Guangrong Date: Thu, 14 Jul 2011 03:19:27 +0000 Subject: [PATCH 146/439] Btrfs: clean up for wait_extent_bit() We can just use cond_resched_lock(). Signed-off-by: Xiao Guangrong Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/extent_io.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 0d69425d0197..a0811e639f58 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -648,11 +648,7 @@ again: if (start > end) break; - if (need_resched()) { - spin_unlock(&tree->lock); - cond_resched(); - spin_lock(&tree->lock); - } + cond_resched_lock(&tree->lock); } out: spin_unlock(&tree->lock); From 69261c4b6a394ead1b5ca2966c908469355eb603 Mon Sep 17 00:00:00 2001 From: Xiao Guangrong Date: Thu, 14 Jul 2011 03:19:45 +0000 Subject: [PATCH 147/439] Btrfs: clean up for find_first_extent_bit() find_first_extent_bit() and find_first_extent_bit_state() share most of the code, and we can just make the former call the latter. Signed-off-by: Xiao Guangrong Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/extent_io.c | 64 +++++++++++++++++--------------------------- 1 file changed, 24 insertions(+), 40 deletions(-) diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index a0811e639f58..a59ebc608ccc 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -1031,46 +1031,6 @@ static int set_range_writeback(struct extent_io_tree *tree, u64 start, u64 end) return 0; } -/* - * find the first offset in the io tree with 'bits' set. zero is - * returned if we find something, and *start_ret and *end_ret are - * set to reflect the state struct that was found. - * - * If nothing was found, 1 is returned, < 0 on error - */ -int find_first_extent_bit(struct extent_io_tree *tree, u64 start, - u64 *start_ret, u64 *end_ret, int bits) -{ - struct rb_node *node; - struct extent_state *state; - int ret = 1; - - spin_lock(&tree->lock); - /* - * this search will find all the extents that end after - * our range starts. - */ - node = tree_search(tree, start); - if (!node) - goto out; - - while (1) { - state = rb_entry(node, struct extent_state, rb_node); - if (state->end >= start && (state->state & bits)) { - *start_ret = state->start; - *end_ret = state->end; - ret = 0; - break; - } - node = rb_next(node); - if (!node) - break; - } -out: - spin_unlock(&tree->lock); - return ret; -} - /* find the first state struct with 'bits' set after 'start', and * return it. tree->lock must be held. NULL will returned if * nothing was found after 'start' @@ -1102,6 +1062,30 @@ out: return NULL; } +/* + * find the first offset in the io tree with 'bits' set. zero is + * returned if we find something, and *start_ret and *end_ret are + * set to reflect the state struct that was found. + * + * If nothing was found, 1 is returned, < 0 on error + */ +int find_first_extent_bit(struct extent_io_tree *tree, u64 start, + u64 *start_ret, u64 *end_ret, int bits) +{ + struct extent_state *state; + int ret = 1; + + spin_lock(&tree->lock); + state = find_first_extent_bit_state(tree, start, bits); + if (state) { + *start_ret = state->start; + *end_ret = state->end; + ret = 0; + } + spin_unlock(&tree->lock); + return ret; +} + /* * find a contiguous range of bytes in the file marked as delalloc, not * more than 'max_bytes'. start and end are used to return the range, From 341d14f161a475ebdbc9adff1f7e681e1185dee9 Mon Sep 17 00:00:00 2001 From: Mitch Harder Date: Tue, 12 Jul 2011 19:43:45 +0000 Subject: [PATCH 148/439] Btrfs: Remove unused variable 'last_index' in file.c The variable 'last_index' is calculated in the __btrfs_buffered_write function and passed as a parameter to the prepare_pages function, but is not used anywhere in the prepare_pages function. Remove instances of 'last_index' in these functions. Signed-off-by: Mitch Harder Signed-off-by: Chris Mason --- fs/btrfs/file.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 7f134a730efb..010aec8be824 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -1059,7 +1059,7 @@ static int prepare_uptodate_page(struct page *page, u64 pos) static noinline int prepare_pages(struct btrfs_root *root, struct file *file, struct page **pages, size_t num_pages, loff_t pos, unsigned long first_index, - unsigned long last_index, size_t write_bytes) + size_t write_bytes) { struct extent_state *cached_state = NULL; int i; @@ -1159,7 +1159,6 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file, struct btrfs_root *root = BTRFS_I(inode)->root; struct page **pages = NULL; unsigned long first_index; - unsigned long last_index; size_t num_written = 0; int nrptrs; int ret = 0; @@ -1172,7 +1171,6 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file, return -ENOMEM; first_index = pos >> PAGE_CACHE_SHIFT; - last_index = (pos + iov_iter_count(i)) >> PAGE_CACHE_SHIFT; while (iov_iter_count(i) > 0) { size_t offset = pos & (PAGE_CACHE_SIZE - 1); @@ -1206,8 +1204,7 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file, * contents of pages from loop to loop */ ret = prepare_pages(root, file, pages, num_pages, - pos, first_index, last_index, - write_bytes); + pos, first_index, write_bytes); if (ret) { btrfs_delalloc_release_space(inode, num_pages << PAGE_CACHE_SHIFT); From 0d10ee2e6deb5c8409ae65b970846344897d5e4e Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Mon, 1 Aug 2011 14:37:36 -0400 Subject: [PATCH 149/439] Btrfs: don't call writepages from within write_full_page When doing a writepage we call writepages to try and write out any other dirty pages in the area. This could cause problems where we commit a transaction and then have somebody else dirtying metadata in the area as we could end up writing out a lot more than we care about, which could cause latency on anybody who is waiting for the transaction to completely finish committing. Thanks, Signed-off-by: Josef Bacik Signed-off-by: Chris Mason --- fs/btrfs/extent_io.c | 10 ---------- 1 file changed, 10 deletions(-) diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index a59ebc608ccc..8491712b1dad 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2502,7 +2502,6 @@ int extent_write_full_page(struct extent_io_tree *tree, struct page *page, struct writeback_control *wbc) { int ret; - struct address_space *mapping = page->mapping; struct extent_page_data epd = { .bio = NULL, .tree = tree, @@ -2510,18 +2509,9 @@ int extent_write_full_page(struct extent_io_tree *tree, struct page *page, .extent_locked = 0, .sync_io = wbc->sync_mode == WB_SYNC_ALL, }; - struct writeback_control wbc_writepages = { - .sync_mode = wbc->sync_mode, - .older_than_this = NULL, - .nr_to_write = 64, - .range_start = page_offset(page) + PAGE_CACHE_SIZE, - .range_end = (loff_t)-1, - }; ret = __extent_writepage(page, wbc, &epd); - extent_write_cache_pages(tree, mapping, &wbc_writepages, - __extent_writepage, &epd, flush_write_bio); flush_epd_write_bio(&epd); return ret; } From 62ac41a6e443ef26b9de862c6e20c088e2b04dde Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Thu, 28 Jul 2011 12:55:14 -0700 Subject: [PATCH 150/439] drm/i915: don't use uninitialized EDID bpc values when picking pipe bpp The EDID parser will zero out the bpc value, and the driver needs to handle that case. In our picker, we'll just ignore 0 values as far as bpp picking goes. Fixes https://bugs.freedesktop.org/show_bug.cgi?id=39323. Signed-off-by: Jesse Barnes Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_display.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 2bf5bb63fe41..35364e68a091 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -4600,7 +4600,9 @@ static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc, if (connector->encoder != encoder) continue; - if (connector->display_info.bpc < display_bpc) { + /* Don't use an invalid EDID bpc value */ + if (connector->display_info.bpc && + connector->display_info.bpc < display_bpc) { DRM_DEBUG_DRIVER("clamping display bpc (was %d) to EDID reported max of %d\n", display_bpc, connector->display_info.bpc); display_bpc = connector->display_info.bpc; } @@ -5215,7 +5217,8 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, temp |= PIPE_12BPC; break; default: - WARN(1, "intel_choose_pipe_bpp returned invalid value\n"); + WARN(1, "intel_choose_pipe_bpp returned invalid value %d\n", + pipe_bpp); temp |= PIPE_8BPC; pipe_bpp = 24; break; From 11bee43ebba0bfc92165c059f6e9869197ea8889 Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Mon, 1 Aug 2011 15:02:20 -0700 Subject: [PATCH 151/439] drm/i915/dp: wait for previous AUX channel activity to clear Before initiating a new read or write on the DP AUX channel, wait for any outstanding activity to complete. This may happen during normal retry behavior. If the wait fails (i.e. after 1ms the AUX channel is still busy) dump a backtrace to make the caller easier to spot. v2: use msleep instead, and timeout after 3ms (only ever saw 1 retry with msleep in testing) v3: fix backtrace check to trigger if the 3ms wait times out Fixes https://bugs.freedesktop.org/show_bug.cgi?id=38136. Signed-off-by: Jesse Barnes Reviewed-by: Keith Packard Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_dp.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 449364187bc0..ba72fbcbbbc9 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -315,9 +315,17 @@ intel_dp_aux_ch(struct intel_dp *intel_dp, else precharge = 5; - if (I915_READ(ch_ctl) & DP_AUX_CH_CTL_SEND_BUSY) { - DRM_ERROR("dp_aux_ch not started status 0x%08x\n", - I915_READ(ch_ctl)); + /* Try to wait for any previous AUX channel activity */ + for (try = 0; try < 3; try++) { + status = I915_READ(ch_ctl); + if ((status & DP_AUX_CH_CTL_SEND_BUSY) == 0) + break; + msleep(1); + } + + if (try == 3) { + WARN(1, "dp_aux_ch not started status 0x%08x\n", + I915_READ(ch_ctl)); return -EBUSY; } From a08af810cdc29d2ca930e8a869d3d01744c392d8 Mon Sep 17 00:00:00 2001 From: Chas Williams Date: Mon, 1 Aug 2011 17:56:14 -0700 Subject: [PATCH 152/439] atm: br2864: sent packets truncated in VC routed mode Reported-by: Pascal Hambourg Signed-off-by: Chas Williams Signed-off-by: David S. Miller --- net/atm/br2684.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/net/atm/br2684.c b/net/atm/br2684.c index 2252c2085dac..52cfd0c3ea71 100644 --- a/net/atm/br2684.c +++ b/net/atm/br2684.c @@ -242,8 +242,6 @@ static int br2684_xmit_vcc(struct sk_buff *skb, struct net_device *dev, if (brdev->payload == p_bridged) { skb_push(skb, 2); memset(skb->data, 0, 2); - } else { /* p_routed */ - skb_pull(skb, ETH_HLEN); } } skb_debug(skb); From 87a0874cf19f1bc9bd25bd7d053a0ea25ccf8373 Mon Sep 17 00:00:00 2001 From: Paul Moore Date: Mon, 1 Aug 2011 11:10:26 +0000 Subject: [PATCH 153/439] doc: Update the MAINTAINERS info for Paul Moore My @hp.com will no longer be valid starting August 5, 2011 so an update is necessary. My new email address is employer independent so we don't have to worry about doing this again any time soon. Signed-off-by: Paul Moore Signed-off-by: Paul Moore Signed-off-by: David S. Miller --- MAINTAINERS | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 5d4ceba0e960..ca87f470e548 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4395,10 +4395,10 @@ F: net/*/netfilter/ F: net/netfilter/ NETLABEL -M: Paul Moore +M: Paul Moore W: http://netlabel.sf.net L: netdev@vger.kernel.org -S: Supported +S: Maintained F: Documentation/netlabel/ F: include/net/netlabel.h F: net/netlabel/ @@ -4455,7 +4455,7 @@ F: include/net/ip* F: arch/x86/net/* NETWORKING [LABELED] (NetLabel, CIPSO, Labeled IPsec, SECMARK) -M: Paul Moore +M: Paul Moore L: netdev@vger.kernel.org S: Maintained From 82c21bfab41a77bc01affe21bea9727d776774a7 Mon Sep 17 00:00:00 2001 From: Paul Moore Date: Mon, 1 Aug 2011 11:10:33 +0000 Subject: [PATCH 154/439] doc: Update the email address for Paul Moore in various source files My @hp.com will no longer be valid starting August 5, 2011 so an update is necessary. My new email address is employer independent so we don't have to worry about doing this again any time soon. Signed-off-by: Paul Moore Signed-off-by: Paul Moore Signed-off-by: David S. Miller --- include/net/cipso_ipv4.h | 2 +- include/net/netlabel.h | 2 +- net/netlabel/Makefile | 2 -- net/netlabel/netlabel_addrlist.c | 2 +- net/netlabel/netlabel_addrlist.h | 2 +- net/netlabel/netlabel_cipso_v4.c | 2 +- net/netlabel/netlabel_cipso_v4.h | 2 +- net/netlabel/netlabel_domainhash.c | 2 +- net/netlabel/netlabel_domainhash.h | 2 +- net/netlabel/netlabel_kapi.c | 2 +- net/netlabel/netlabel_mgmt.c | 2 +- net/netlabel/netlabel_mgmt.h | 2 +- net/netlabel/netlabel_unlabeled.c | 2 +- net/netlabel/netlabel_unlabeled.h | 2 +- net/netlabel/netlabel_user.c | 2 +- net/netlabel/netlabel_user.h | 2 +- security/selinux/hooks.c | 2 +- security/selinux/include/netif.h | 2 +- security/selinux/include/netlabel.h | 2 +- security/selinux/include/netnode.h | 2 +- security/selinux/include/netport.h | 2 +- security/selinux/netif.c | 2 +- security/selinux/netlabel.c | 2 +- security/selinux/netnode.c | 2 +- security/selinux/netport.c | 2 +- security/selinux/selinuxfs.c | 2 +- security/selinux/ss/ebitmap.c | 2 +- security/selinux/ss/mls.c | 2 +- security/selinux/ss/mls.h | 2 +- security/selinux/ss/policydb.c | 2 +- security/selinux/ss/services.c | 2 +- security/smack/smack_lsm.c | 2 +- 32 files changed, 31 insertions(+), 33 deletions(-) diff --git a/include/net/cipso_ipv4.h b/include/net/cipso_ipv4.h index 3b938743514b..9808877c2ab9 100644 --- a/include/net/cipso_ipv4.h +++ b/include/net/cipso_ipv4.h @@ -8,7 +8,7 @@ * have chosen to adopt the protocol and over the years it has become a * de-facto standard for labeled networking. * - * Author: Paul Moore + * Author: Paul Moore * */ diff --git a/include/net/netlabel.h b/include/net/netlabel.h index f21a16ee3705..f67440970d7e 100644 --- a/include/net/netlabel.h +++ b/include/net/netlabel.h @@ -4,7 +4,7 @@ * The NetLabel system manages static and dynamic label mappings for network * protocols such as CIPSO and RIPSO. * - * Author: Paul Moore + * Author: Paul Moore * */ diff --git a/net/netlabel/Makefile b/net/netlabel/Makefile index ea750e9df65f..d2732fc952e2 100644 --- a/net/netlabel/Makefile +++ b/net/netlabel/Makefile @@ -1,8 +1,6 @@ # # Makefile for the NetLabel subsystem. # -# Feb 9, 2006, Paul Moore -# # base objects obj-y := netlabel_user.o netlabel_kapi.o diff --git a/net/netlabel/netlabel_addrlist.c b/net/netlabel/netlabel_addrlist.c index c0519139679e..96b749dacc34 100644 --- a/net/netlabel/netlabel_addrlist.c +++ b/net/netlabel/netlabel_addrlist.c @@ -6,7 +6,7 @@ * system manages static and dynamic label mappings for network protocols such * as CIPSO and RIPSO. * - * Author: Paul Moore + * Author: Paul Moore * */ diff --git a/net/netlabel/netlabel_addrlist.h b/net/netlabel/netlabel_addrlist.h index 2b9644e19de0..fdbc1d2c7352 100644 --- a/net/netlabel/netlabel_addrlist.h +++ b/net/netlabel/netlabel_addrlist.h @@ -6,7 +6,7 @@ * system manages static and dynamic label mappings for network protocols such * as CIPSO and RIPSO. * - * Author: Paul Moore + * Author: Paul Moore * */ diff --git a/net/netlabel/netlabel_cipso_v4.c b/net/netlabel/netlabel_cipso_v4.c index dd53a36d89af..6bf878335d94 100644 --- a/net/netlabel/netlabel_cipso_v4.c +++ b/net/netlabel/netlabel_cipso_v4.c @@ -5,7 +5,7 @@ * NetLabel system manages static and dynamic label mappings for network * protocols such as CIPSO and RIPSO. * - * Author: Paul Moore + * Author: Paul Moore * */ diff --git a/net/netlabel/netlabel_cipso_v4.h b/net/netlabel/netlabel_cipso_v4.h index af7f3355103e..d24d774bfd62 100644 --- a/net/netlabel/netlabel_cipso_v4.h +++ b/net/netlabel/netlabel_cipso_v4.h @@ -5,7 +5,7 @@ * NetLabel system manages static and dynamic label mappings for network * protocols such as CIPSO and RIPSO. * - * Author: Paul Moore + * Author: Paul Moore * */ diff --git a/net/netlabel/netlabel_domainhash.c b/net/netlabel/netlabel_domainhash.c index 2aa975e5452d..7d8083cde34f 100644 --- a/net/netlabel/netlabel_domainhash.c +++ b/net/netlabel/netlabel_domainhash.c @@ -6,7 +6,7 @@ * system manages static and dynamic label mappings for network protocols such * as CIPSO and RIPSO. * - * Author: Paul Moore + * Author: Paul Moore * */ diff --git a/net/netlabel/netlabel_domainhash.h b/net/netlabel/netlabel_domainhash.h index 0261dda3f2d2..bfcc0f7024c5 100644 --- a/net/netlabel/netlabel_domainhash.h +++ b/net/netlabel/netlabel_domainhash.h @@ -6,7 +6,7 @@ * system manages static and dynamic label mappings for network protocols such * as CIPSO and RIPSO. * - * Author: Paul Moore + * Author: Paul Moore * */ diff --git a/net/netlabel/netlabel_kapi.c b/net/netlabel/netlabel_kapi.c index b528dd928d3c..58107d060846 100644 --- a/net/netlabel/netlabel_kapi.c +++ b/net/netlabel/netlabel_kapi.c @@ -5,7 +5,7 @@ * system manages static and dynamic label mappings for network protocols such * as CIPSO and RIPSO. * - * Author: Paul Moore + * Author: Paul Moore * */ diff --git a/net/netlabel/netlabel_mgmt.c b/net/netlabel/netlabel_mgmt.c index dff8a0809245..bfa555869775 100644 --- a/net/netlabel/netlabel_mgmt.c +++ b/net/netlabel/netlabel_mgmt.c @@ -5,7 +5,7 @@ * NetLabel system manages static and dynamic label mappings for network * protocols such as CIPSO and RIPSO. * - * Author: Paul Moore + * Author: Paul Moore * */ diff --git a/net/netlabel/netlabel_mgmt.h b/net/netlabel/netlabel_mgmt.h index 8db37f4c10f7..5a9f31ce5799 100644 --- a/net/netlabel/netlabel_mgmt.h +++ b/net/netlabel/netlabel_mgmt.h @@ -5,7 +5,7 @@ * NetLabel system manages static and dynamic label mappings for network * protocols such as CIPSO and RIPSO. * - * Author: Paul Moore + * Author: Paul Moore * */ diff --git a/net/netlabel/netlabel_unlabeled.c b/net/netlabel/netlabel_unlabeled.c index f1ecf848e3ac..e6e823656f9d 100644 --- a/net/netlabel/netlabel_unlabeled.c +++ b/net/netlabel/netlabel_unlabeled.c @@ -5,7 +5,7 @@ * NetLabel system. The NetLabel system manages static and dynamic label * mappings for network protocols such as CIPSO and RIPSO. * - * Author: Paul Moore + * Author: Paul Moore * */ diff --git a/net/netlabel/netlabel_unlabeled.h b/net/netlabel/netlabel_unlabeled.h index 0bc8dc3f9e3c..700af49022a0 100644 --- a/net/netlabel/netlabel_unlabeled.h +++ b/net/netlabel/netlabel_unlabeled.h @@ -5,7 +5,7 @@ * NetLabel system. The NetLabel system manages static and dynamic label * mappings for network protocols such as CIPSO and RIPSO. * - * Author: Paul Moore + * Author: Paul Moore * */ diff --git a/net/netlabel/netlabel_user.c b/net/netlabel/netlabel_user.c index a3fd75ac3fa5..9fae63f10298 100644 --- a/net/netlabel/netlabel_user.c +++ b/net/netlabel/netlabel_user.c @@ -5,7 +5,7 @@ * NetLabel system manages static and dynamic label mappings for network * protocols such as CIPSO and RIPSO. * - * Author: Paul Moore + * Author: Paul Moore * */ diff --git a/net/netlabel/netlabel_user.h b/net/netlabel/netlabel_user.h index f4fc4c9ad567..81969785e279 100644 --- a/net/netlabel/netlabel_user.h +++ b/net/netlabel/netlabel_user.h @@ -5,7 +5,7 @@ * NetLabel system manages static and dynamic label mappings for network * protocols such as CIPSO and RIPSO. * - * Author: Paul Moore + * Author: Paul Moore * */ diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index a38316b2e3f6..266a2292451d 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -14,7 +14,7 @@ * Copyright (C) 2004-2005 Trusted Computer Solutions, Inc. * * Copyright (C) 2006, 2007, 2009 Hewlett-Packard Development Company, L.P. - * Paul Moore + * Paul Moore * Copyright (C) 2007 Hitachi Software Engineering Co., Ltd. * Yuichi Nakamura * diff --git a/security/selinux/include/netif.h b/security/selinux/include/netif.h index ce23edd128b3..43d507242b42 100644 --- a/security/selinux/include/netif.h +++ b/security/selinux/include/netif.h @@ -8,7 +8,7 @@ * * Copyright (C) 2003 Red Hat, Inc., James Morris * Copyright (C) 2007 Hewlett-Packard Development Company, L.P. - * Paul Moore, + * Paul Moore * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2, diff --git a/security/selinux/include/netlabel.h b/security/selinux/include/netlabel.h index cf2f628e6e28..8c59b8f150e8 100644 --- a/security/selinux/include/netlabel.h +++ b/security/selinux/include/netlabel.h @@ -1,7 +1,7 @@ /* * SELinux interface to the NetLabel subsystem * - * Author : Paul Moore + * Author: Paul Moore * */ diff --git a/security/selinux/include/netnode.h b/security/selinux/include/netnode.h index 1b94450d11d2..df7a5ed6c694 100644 --- a/security/selinux/include/netnode.h +++ b/security/selinux/include/netnode.h @@ -6,7 +6,7 @@ * needed to reduce the lookup overhead since most of these queries happen on * a per-packet basis. * - * Author: Paul Moore + * Author: Paul Moore * */ diff --git a/security/selinux/include/netport.h b/security/selinux/include/netport.h index 8991752eaf93..4d965b83d735 100644 --- a/security/selinux/include/netport.h +++ b/security/selinux/include/netport.h @@ -5,7 +5,7 @@ * mapping is maintained as part of the normal policy but a fast cache is * needed to reduce the lookup overhead. * - * Author: Paul Moore + * Author: Paul Moore * */ diff --git a/security/selinux/netif.c b/security/selinux/netif.c index 58cc481c93d5..326f22cbe405 100644 --- a/security/selinux/netif.c +++ b/security/selinux/netif.c @@ -8,7 +8,7 @@ * * Copyright (C) 2003 Red Hat, Inc., James Morris * Copyright (C) 2007 Hewlett-Packard Development Company, L.P. - * Paul Moore + * Paul Moore * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2, diff --git a/security/selinux/netlabel.c b/security/selinux/netlabel.c index c3bf3ed07b06..da4b8b233280 100644 --- a/security/selinux/netlabel.c +++ b/security/selinux/netlabel.c @@ -4,7 +4,7 @@ * This file provides the necessary glue to tie NetLabel into the SELinux * subsystem. * - * Author: Paul Moore + * Author: Paul Moore * */ diff --git a/security/selinux/netnode.c b/security/selinux/netnode.c index 8b691a863186..3bf46abaa688 100644 --- a/security/selinux/netnode.c +++ b/security/selinux/netnode.c @@ -6,7 +6,7 @@ * needed to reduce the lookup overhead since most of these queries happen on * a per-packet basis. * - * Author: Paul Moore + * Author: Paul Moore * * This code is heavily based on the "netif" concept originally developed by * James Morris diff --git a/security/selinux/netport.c b/security/selinux/netport.c index ae76e298de7d..0b62bd112461 100644 --- a/security/selinux/netport.c +++ b/security/selinux/netport.c @@ -5,7 +5,7 @@ * mapping is maintained as part of the normal policy but a fast cache is * needed to reduce the lookup overhead. * - * Author: Paul Moore + * Author: Paul Moore * * This code is heavily based on the "netif" concept originally developed by * James Morris diff --git a/security/selinux/selinuxfs.c b/security/selinux/selinuxfs.c index de7900ef53da..55d92cbb177a 100644 --- a/security/selinux/selinuxfs.c +++ b/security/selinux/selinuxfs.c @@ -2,7 +2,7 @@ * * Added conditional policy language extensions * - * Updated: Hewlett-Packard + * Updated: Hewlett-Packard * * Added support for the policy capability bitmap * diff --git a/security/selinux/ss/ebitmap.c b/security/selinux/ss/ebitmap.c index d42951fcbe87..30f119b1d1ec 100644 --- a/security/selinux/ss/ebitmap.c +++ b/security/selinux/ss/ebitmap.c @@ -4,7 +4,7 @@ * Author : Stephen Smalley, */ /* - * Updated: Hewlett-Packard + * Updated: Hewlett-Packard * * Added support to import/export the NetLabel category bitmap * diff --git a/security/selinux/ss/mls.c b/security/selinux/ss/mls.c index e96174216bc9..fbf9c5816c71 100644 --- a/security/selinux/ss/mls.c +++ b/security/selinux/ss/mls.c @@ -11,7 +11,7 @@ * Copyright (C) 2004-2006 Trusted Computer Solutions, Inc. */ /* - * Updated: Hewlett-Packard + * Updated: Hewlett-Packard * * Added support to import/export the MLS label from NetLabel * diff --git a/security/selinux/ss/mls.h b/security/selinux/ss/mls.h index 037bf9d82d41..e4369e3e6366 100644 --- a/security/selinux/ss/mls.h +++ b/security/selinux/ss/mls.h @@ -11,7 +11,7 @@ * Copyright (C) 2004-2006 Trusted Computer Solutions, Inc. */ /* - * Updated: Hewlett-Packard + * Updated: Hewlett-Packard * * Added support to import/export the MLS label from NetLabel * diff --git a/security/selinux/ss/policydb.c b/security/selinux/ss/policydb.c index d246aca3f4fb..2381d0ded228 100644 --- a/security/selinux/ss/policydb.c +++ b/security/selinux/ss/policydb.c @@ -13,7 +13,7 @@ * * Added conditional policy language extensions * - * Updated: Hewlett-Packard + * Updated: Hewlett-Packard * * Added support for the policy capability bitmap * diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c index 973e00e34fa9..f6917bc0aa05 100644 --- a/security/selinux/ss/services.c +++ b/security/selinux/ss/services.c @@ -13,7 +13,7 @@ * * Added conditional policy language extensions * - * Updated: Hewlett-Packard + * Updated: Hewlett-Packard * * Added support for NetLabel * Added support for the policy capability bitmap diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c index f375eb2e1957..b9c5e149903b 100644 --- a/security/smack/smack_lsm.c +++ b/security/smack/smack_lsm.c @@ -9,7 +9,7 @@ * * Copyright (C) 2007 Casey Schaufler * Copyright (C) 2009 Hewlett-Packard Development Company, L.P. - * Paul Moore + * Paul Moore * Copyright (C) 2010 Nokia Corporation * * This program is free software; you can redistribute it and/or modify From 8fb4e1399137dbbf7f1071bc723b5eff8ad46d1a Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Mon, 1 Aug 2011 17:59:44 -0700 Subject: [PATCH 155/439] Documentation/bonding.txt: Update to 3.x version numbers Signed-off-by: Jesper Juhl Signed-off-by: David S. Miller --- Documentation/networking/bonding.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/networking/bonding.txt b/Documentation/networking/bonding.txt index 675612ff41ae..5dd960d75174 100644 --- a/Documentation/networking/bonding.txt +++ b/Documentation/networking/bonding.txt @@ -599,7 +599,7 @@ num_unsol_na affect only the active-backup mode. These options were added for bonding versions 3.3.0 and 3.4.0 respectively. - From Linux 2.6.40 and bonding version 3.7.1, these notifications + From Linux 3.0 and bonding version 3.7.1, these notifications are generated by the ipv4 and ipv6 code and the numbers of repetitions cannot be set independently. From 22019b17821ab3543090827491e465c5816cbcd6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 29 Jul 2011 18:37:31 +0000 Subject: [PATCH 156/439] net: add kerneldoc to skb_copy_bits() Since skb_copy_bits() is called from assembly, add a fat comment to make clear we should think twice before changing its prototype. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/core/skbuff.c | 17 +++++++++++++++-- 1 file changed, 15 insertions(+), 2 deletions(-) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 2beda824636e..27002dffe7ed 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1369,8 +1369,21 @@ pull_pages: } EXPORT_SYMBOL(__pskb_pull_tail); -/* Copy some data bits from skb to kernel buffer. */ - +/** + * skb_copy_bits - copy bits from skb to kernel buffer + * @skb: source skb + * @offset: offset in source + * @to: destination buffer + * @len: number of bytes to copy + * + * Copy the specified number of bytes from the source skb to the + * destination buffer. + * + * CAUTION ! : + * If its prototype is ever changed, + * check arch/{*}/net/{*}.S files, + * since it is called from BPF assembly code. + */ int skb_copy_bits(const struct sk_buff *skb, int offset, void *to, int len) { int start = skb_headlen(skb); From da5aa861bea09197e6ae4d7c46618616064891e4 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Tue, 2 Aug 2011 02:17:48 +0200 Subject: [PATCH 157/439] fix block device fallout from ->fsync() changes blkdev_fsync() needs to write pages in pagecache... Signed-off-by: Rafael J. Wysocki Signed-off-by: Al Viro --- fs/block_dev.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/fs/block_dev.c b/fs/block_dev.c index f28680553288..ff77262e887c 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -387,6 +387,10 @@ int blkdev_fsync(struct file *filp, loff_t start, loff_t end, int datasync) struct inode *bd_inode = filp->f_mapping->host; struct block_device *bdev = I_BDEV(bd_inode); int error; + + error = filemap_write_and_wait_range(filp->f_mapping, start, end); + if (error) + return error; /* * There is no need to serialise calls to blkdev_issue_flush with From 8997621bb2daaf19a4e9d82f118224159d8054e2 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Tue, 2 Aug 2011 00:45:48 -0400 Subject: [PATCH 158/439] ACPI print OSI(Linux) warning only once This message gets repeated on some machines: https://bugzilla.kernel.org/show_bug.cgi?id=29292 Signed-off-by: Len Brown --- drivers/acpi/osl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c index 52ca9649d769..37659804226c 100644 --- a/drivers/acpi/osl.c +++ b/drivers/acpi/osl.c @@ -155,7 +155,7 @@ static u32 acpi_osi_handler(acpi_string interface, u32 supported) { if (!strcmp("Linux", interface)) { - printk(KERN_NOTICE FW_BUG PREFIX + printk_once(KERN_NOTICE FW_BUG PREFIX "BIOS _OSI(Linux) query %s%s\n", osi_linux.enable ? "honored" : "ignored", osi_linux.cmdline ? " via cmdline" : From 08f984c7f7f569b66acbbd163676b4bc7f64addc Mon Sep 17 00:00:00 2001 From: Eliot Blennerhassett Date: Tue, 2 Aug 2011 09:44:24 +1200 Subject: [PATCH 159/439] ALSA: asihpi - Clarify adapter index validity check Avoids assigning possibly invalid address to pa, even if it is never dereferenced. Correct error response to reflect request object/function ids. Signed-off-by: Eliot Blennerhassett Signed-off-by: Takashi Iwai --- sound/pci/asihpi/hpioctl.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-) diff --git a/sound/pci/asihpi/hpioctl.c b/sound/pci/asihpi/hpioctl.c index 9683f84ecdc8..a32502e796de 100644 --- a/sound/pci/asihpi/hpioctl.c +++ b/sound/pci/asihpi/hpioctl.c @@ -177,16 +177,21 @@ long asihpi_hpi_ioctl(struct file *file, unsigned int cmd, unsigned long arg) } else { u16 __user *ptr = NULL; u32 size = 0; - + u32 adapter_present; /* -1=no data 0=read from user mem, 1=write to user mem */ int wrflag = -1; - u32 adapter = hm->h.adapter_index; - struct hpi_adapter *pa = &adapters[adapter]; + struct hpi_adapter *pa; - if ((adapter >= HPI_MAX_ADAPTERS) || (!pa->type)) { - hpi_init_response(&hr->r0, HPI_OBJ_ADAPTER, - HPI_ADAPTER_OPEN, - HPI_ERROR_BAD_ADAPTER_NUMBER); + if (hm->h.adapter_index < HPI_MAX_ADAPTERS) { + pa = &adapters[hm->h.adapter_index]; + adapter_present = pa->type; + } else { + adapter_present = 0; + } + + if (!adapter_present) { + hpi_init_response(&hr->r0, hm->h.object, + hm->h.function, HPI_ERROR_BAD_ADAPTER_NUMBER); uncopied_bytes = copy_to_user(puhr, hr, sizeof(hr->h)); From bea1906620ce72b63f83735c4cc2642b25ec54ae Mon Sep 17 00:00:00 2001 From: David Engraf Date: Wed, 20 Jul 2011 15:03:39 +0200 Subject: [PATCH 160/439] watchdog: shwdt: fix usage of mod_timer Fix the usage of mod_timer() and make the driver usable. mod_timer() must be called with an absolute timeout in jiffies. The old implementation used a relative timeout thus the hardware watchdog was never triggered. Signed-off-by: David Engraf Signed-off-by: Paul Mundt Signed-off-by: Wim Van sebroeck Signed-off-by: Andrew Morton Cc: stable --- drivers/watchdog/shwdt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/watchdog/shwdt.c b/drivers/watchdog/shwdt.c index db84f2322d1a..a267dc078daf 100644 --- a/drivers/watchdog/shwdt.c +++ b/drivers/watchdog/shwdt.c @@ -64,7 +64,7 @@ * misses its deadline, the kernel timer will allow the WDT to overflow. */ static int clock_division_ratio = WTCSR_CKS_4096; -#define next_ping_period(cks) msecs_to_jiffies(cks - 4) +#define next_ping_period(cks) (jiffies + msecs_to_jiffies(cks - 4)) static const struct watchdog_info sh_wdt_info; static struct platform_device *sh_wdt_dev; From 6b01d30eef64456ad9e261d2173266a3244da8e1 Mon Sep 17 00:00:00 2001 From: Mart Gerrits Date: Sat, 30 Jul 2011 16:59:12 +0200 Subject: [PATCH 161/439] watchdog: Fix POST failure on ASUS P5N32-E SLI and similar boards At present the module does not unset the NO_REBOOT bit upon shutdown, this causes the BIOS to fail the POST once and reset. During the next boot it displays the following error message: ***** Warning: System BOOT Fail ***** Your system last boot fail or POST interrupted. Please enter setup to load default and reboot again. Press F1 to continue, DEL to enter SETUP With this patch the NO_REBOOT flag will be unset on shutdown and thus stop this failure from occurring. Tested on 'ASUS P5N32-E SLI with BIOS revision 1801' and 'ASUS P5N32-E SLI PLUS with BIOS revision 1502'. Signed-off-by: Mart Gerrits Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/nv_tco.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/watchdog/nv_tco.c b/drivers/watchdog/nv_tco.c index afa78a54711e..809f41c30c44 100644 --- a/drivers/watchdog/nv_tco.c +++ b/drivers/watchdog/nv_tco.c @@ -458,7 +458,15 @@ static int __devexit nv_tco_remove(struct platform_device *dev) static void nv_tco_shutdown(struct platform_device *dev) { + u32 val; + tco_timer_stop(); + + /* Some BIOSes fail the POST (once) if the NO_REBOOT flag is not + * unset during shutdown. */ + pci_read_config_dword(tco_pci, MCP51_SMBUS_SETUP_B, &val); + val &= ~MCP51_SMBUS_SETUP_B_TCO_REBOOT; + pci_write_config_dword(tco_pci, MCP51_SMBUS_SETUP_B, val); } static struct platform_driver nv_tco_driver = { From 7abdd34dbf58bf271db7ee6551f2ff72204a83fd Mon Sep 17 00:00:00 2001 From: Josh Boyer Date: Sat, 30 Jul 2011 10:18:48 -0400 Subject: [PATCH 162/439] watchdog: Cleanup WATCHDOG_CORE help text The newly added WATCHDOG_CORE option is a bool, but the help text suggests it can be built as a module. Fix it up. Signed-off-by: Josh Boyer Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/Kconfig | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig index f441726ddf2b..86b0735e6aa0 100644 --- a/drivers/watchdog/Kconfig +++ b/drivers/watchdog/Kconfig @@ -36,9 +36,6 @@ config WATCHDOG_CORE and gives them the /dev/watchdog interface (and later also the sysfs interface). - To compile this driver as a module, choose M here: the module will - be called watchdog. - config WATCHDOG_NOWAYOUT bool "Disable watchdog shutdown on close" help From 87b7ba3d24a25cf18aece447de27d7804fa9668c Mon Sep 17 00:00:00 2001 From: Vladislav Zolotarov Date: Tue, 2 Aug 2011 01:35:43 -0700 Subject: [PATCH 163/439] bnx2x: Prevent restarting Tx during bnx2x_nic_unload Tx queues were stopped before bp->state was changed to a value different from BNX2X_STATE_OPEN, which allowed the bnx2x_tx_int() called from the NAPI context to re-enable it. This then allowed the netdev->ndo_start_xmit() to be called in the middle of the function reset and rings freeing. This patch changes bp->state to a value different from BNX2X_STATE_OPEN BEFORE disabling the Tx queues in order to restore the broken protection against the above race in the bnx2x_tx_int(). Signed-off-by: Vladislav Zolotarov Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_cmn.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/net/bnx2x/bnx2x_cmn.c b/drivers/net/bnx2x/bnx2x_cmn.c index 5b0dba6d4efa..d724a18b5285 100644 --- a/drivers/net/bnx2x/bnx2x_cmn.c +++ b/drivers/net/bnx2x/bnx2x_cmn.c @@ -1989,14 +1989,20 @@ int bnx2x_nic_unload(struct bnx2x *bp, int unload_mode) return -EINVAL; } + /* + * It's important to set the bp->state to the value different from + * BNX2X_STATE_OPEN and only then stop the Tx. Otherwise bnx2x_tx_int() + * may restart the Tx from the NAPI context (see bnx2x_tx_int()). + */ + bp->state = BNX2X_STATE_CLOSING_WAIT4_HALT; + smp_mb(); + /* Stop Tx */ bnx2x_tx_disable(bp); #ifdef BCM_CNIC bnx2x_cnic_notify(bp, CNIC_CTL_STOP_CMD); #endif - bp->state = BNX2X_STATE_CLOSING_WAIT4_HALT; - smp_mb(); bp->rx_mode = BNX2X_RX_MODE_NONE; From 762a80d9fc9f690a3a35983f3b4619a220650808 Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Tue, 2 Aug 2011 12:32:00 +0100 Subject: [PATCH 164/439] dm snapshot: flush disk cache when merging This patch makes dm-snapshot flush disk cache when writing metadata for merging snapshot. Without cache flushing the disk may reorder metadata write and other data writes and there is a possibility of data corruption in case of power fault. Cc: stable@kernel.org Signed-off-by: Mikulas Patocka Signed-off-by: Alasdair G Kergon --- drivers/md/dm-snap-persistent.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/md/dm-snap-persistent.c b/drivers/md/dm-snap-persistent.c index 135c2f1fdbfc..e4ecadf0548a 100644 --- a/drivers/md/dm-snap-persistent.c +++ b/drivers/md/dm-snap-persistent.c @@ -753,7 +753,7 @@ static int persistent_commit_merge(struct dm_exception_store *store, for (i = 0; i < nr_merged; i++) clear_exception(ps, ps->current_committed - 1 - i); - r = area_io(ps, WRITE); + r = area_io(ps, WRITE_FLUSH_FUA); if (r < 0) return r; From 286f367dad40beb3234a18c17391d03ba939a7f3 Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Tue, 2 Aug 2011 12:32:00 +0100 Subject: [PATCH 165/439] dm mpath: fix potential NULL pointer in feature arg processing Avoid dereferencing a NULL pointer if the number of feature arguments supplied is fewer than indicated. Signed-off-by: Mike Snitzer Signed-off-by: Alasdair G Kergon Cc: stable@kernel.org --- drivers/md/dm-mpath.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index c3547016f0f1..adf851a081bd 100644 --- a/drivers/md/dm-mpath.c +++ b/drivers/md/dm-mpath.c @@ -807,6 +807,11 @@ static int parse_features(struct arg_set *as, struct multipath *m) if (!argc) return 0; + if (argc > as->argc) { + ti->error = "not enough arguments for features"; + return -EINVAL; + } + do { param_name = shift(as); argc--; From bb91bc7bacb906c9f3a9b22744c53fa7564b51ba Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Tue, 2 Aug 2011 12:32:01 +0100 Subject: [PATCH 166/439] dm io: flush cpu cache with vmapped io For normal kernel pages, CPU cache is synchronized by the dma layer. However, this is not done for pages allocated with vmalloc. If we do I/O to/from vmallocated pages, we must synchronize CPU cache explicitly. Prior to doing I/O on vmallocated page we must call flush_kernel_vmap_range to flush dirty cache on the virtual address. After finished read we must call invalidate_kernel_vmap_range to invalidate cache on the virtual address, so that accesses to the virtual address return newly read data and not stale data from CPU cache. This patch fixes metadata corruption on dm-snapshots on PA-RISC and possibly other architectures with caches indexed by virtual address. Cc: stable Signed-off-by: Mikulas Patocka Signed-off-by: Alasdair G Kergon --- drivers/md/dm-io.c | 29 +++++++++++++++++++++++++++-- 1 file changed, 27 insertions(+), 2 deletions(-) diff --git a/drivers/md/dm-io.c b/drivers/md/dm-io.c index 2067288f61f9..ad2eba40e319 100644 --- a/drivers/md/dm-io.c +++ b/drivers/md/dm-io.c @@ -38,6 +38,8 @@ struct io { struct dm_io_client *client; io_notify_fn callback; void *context; + void *vma_invalidate_address; + unsigned long vma_invalidate_size; } __attribute__((aligned(DM_IO_MAX_REGIONS))); static struct kmem_cache *_dm_io_cache; @@ -116,6 +118,10 @@ static void dec_count(struct io *io, unsigned int region, int error) set_bit(region, &io->error_bits); if (atomic_dec_and_test(&io->count)) { + if (io->vma_invalidate_size) + invalidate_kernel_vmap_range(io->vma_invalidate_address, + io->vma_invalidate_size); + if (io->sleeper) wake_up_process(io->sleeper); @@ -159,6 +165,9 @@ struct dpages { unsigned context_u; void *context_ptr; + + void *vma_invalidate_address; + unsigned long vma_invalidate_size; }; /* @@ -377,6 +386,9 @@ static int sync_io(struct dm_io_client *client, unsigned int num_regions, io->sleeper = current; io->client = client; + io->vma_invalidate_address = dp->vma_invalidate_address; + io->vma_invalidate_size = dp->vma_invalidate_size; + dispatch_io(rw, num_regions, where, dp, io, 1); while (1) { @@ -415,13 +427,21 @@ static int async_io(struct dm_io_client *client, unsigned int num_regions, io->callback = fn; io->context = context; + io->vma_invalidate_address = dp->vma_invalidate_address; + io->vma_invalidate_size = dp->vma_invalidate_size; + dispatch_io(rw, num_regions, where, dp, io, 0); return 0; } -static int dp_init(struct dm_io_request *io_req, struct dpages *dp) +static int dp_init(struct dm_io_request *io_req, struct dpages *dp, + unsigned long size) { /* Set up dpages based on memory type */ + + dp->vma_invalidate_address = NULL; + dp->vma_invalidate_size = 0; + switch (io_req->mem.type) { case DM_IO_PAGE_LIST: list_dp_init(dp, io_req->mem.ptr.pl, io_req->mem.offset); @@ -432,6 +452,11 @@ static int dp_init(struct dm_io_request *io_req, struct dpages *dp) break; case DM_IO_VMA: + flush_kernel_vmap_range(io_req->mem.ptr.vma, size); + if ((io_req->bi_rw & RW_MASK) == READ) { + dp->vma_invalidate_address = io_req->mem.ptr.vma; + dp->vma_invalidate_size = size; + } vm_dp_init(dp, io_req->mem.ptr.vma); break; @@ -460,7 +485,7 @@ int dm_io(struct dm_io_request *io_req, unsigned num_regions, int r; struct dpages dp; - r = dp_init(io_req, &dp); + r = dp_init(io_req, &dp, (unsigned long)where->count << SECTOR_SHIFT); if (r) return r; From d15b774c2920d55e3d58275c97fbe3adc3afde38 Mon Sep 17 00:00:00 2001 From: Alasdair G Kergon Date: Tue, 2 Aug 2011 12:32:01 +0100 Subject: [PATCH 167/439] dm: fix idr leak on module removal Destroy _minor_idr when unloading the core dm module. (Found by kmemleak.) Cc: stable@kernel.org Signed-off-by: Alasdair G Kergon --- drivers/md/dm.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 0cf68b478878..41abc6dd481b 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -37,6 +37,8 @@ static const char *_name = DM_NAME; static unsigned int major = 0; static unsigned int _major = 0; +static DEFINE_IDR(_minor_idr); + static DEFINE_SPINLOCK(_minor_lock); /* * For bio-based dm. @@ -313,6 +315,12 @@ static void __exit dm_exit(void) while (i--) _exits[i](); + + /* + * Should be empty by this point. + */ + idr_remove_all(&_minor_idr); + idr_destroy(&_minor_idr); } /* @@ -1705,8 +1713,6 @@ static int dm_any_congested(void *congested_data, int bdi_bits) /*----------------------------------------------------------------- * An IDR is used to keep track of allocated minor numbers. *---------------------------------------------------------------*/ -static DEFINE_IDR(_minor_idr); - static void free_minor(int minor) { spin_lock(&_minor_lock); From 283a8328ca5b987e547848de8ff0e28edcfb9e08 Mon Sep 17 00:00:00 2001 From: Alasdair G Kergon Date: Tue, 2 Aug 2011 12:32:01 +0100 Subject: [PATCH 168/439] dm: suppress endian warnings Suppress sparse warnings about cpu_to_le32() by using __le32 types for on-disk data etc. Signed-off-by: Alasdair G Kergon --- drivers/md/dm-crypt.c | 6 +-- drivers/md/dm-log.c | 20 ++++++---- drivers/md/dm-snap-persistent.c | 71 ++++++++++++++++++--------------- 3 files changed, 54 insertions(+), 43 deletions(-) diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index bae6c4e23d3f..f5406766ece3 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -239,7 +239,7 @@ static int crypt_iv_plain_gen(struct crypt_config *cc, u8 *iv, struct dm_crypt_request *dmreq) { memset(iv, 0, cc->iv_size); - *(u32 *)iv = cpu_to_le32(dmreq->iv_sector & 0xffffffff); + *(__le32 *)iv = cpu_to_le32(dmreq->iv_sector & 0xffffffff); return 0; } @@ -248,7 +248,7 @@ static int crypt_iv_plain64_gen(struct crypt_config *cc, u8 *iv, struct dm_crypt_request *dmreq) { memset(iv, 0, cc->iv_size); - *(u64 *)iv = cpu_to_le64(dmreq->iv_sector); + *(__le64 *)iv = cpu_to_le64(dmreq->iv_sector); return 0; } @@ -415,7 +415,7 @@ static int crypt_iv_essiv_gen(struct crypt_config *cc, u8 *iv, struct crypto_cipher *essiv_tfm = this_crypt_config(cc)->iv_private; memset(iv, 0, cc->iv_size); - *(u64 *)iv = cpu_to_le64(dmreq->iv_sector); + *(__le64 *)iv = cpu_to_le64(dmreq->iv_sector); crypto_cipher_encrypt_one(essiv_tfm, iv, iv); return 0; diff --git a/drivers/md/dm-log.c b/drivers/md/dm-log.c index 948e3f4925bf..5f06fb687408 100644 --- a/drivers/md/dm-log.c +++ b/drivers/md/dm-log.c @@ -197,15 +197,21 @@ EXPORT_SYMBOL(dm_dirty_log_destroy); #define MIRROR_DISK_VERSION 2 #define LOG_OFFSET 2 -struct log_header { - uint32_t magic; +struct log_header_disk { + __le32 magic; /* * Simple, incrementing version. no backward * compatibility. */ + __le32 version; + __le64 nr_regions; +} __packed; + +struct log_header_core { + uint32_t magic; uint32_t version; - sector_t nr_regions; + uint64_t nr_regions; }; struct log_c { @@ -239,10 +245,10 @@ struct log_c { int log_dev_failed; int log_dev_flush_failed; struct dm_dev *log_dev; - struct log_header header; + struct log_header_core header; struct dm_io_region header_location; - struct log_header *disk_header; + struct log_header_disk *disk_header; }; /* @@ -271,14 +277,14 @@ static inline void log_clear_bit(struct log_c *l, /*---------------------------------------------------------------- * Header IO *--------------------------------------------------------------*/ -static void header_to_disk(struct log_header *core, struct log_header *disk) +static void header_to_disk(struct log_header_core *core, struct log_header_disk *disk) { disk->magic = cpu_to_le32(core->magic); disk->version = cpu_to_le32(core->version); disk->nr_regions = cpu_to_le64(core->nr_regions); } -static void header_from_disk(struct log_header *core, struct log_header *disk) +static void header_from_disk(struct log_header_core *core, struct log_header_disk *disk) { core->magic = le32_to_cpu(disk->magic); core->version = le32_to_cpu(disk->version); diff --git a/drivers/md/dm-snap-persistent.c b/drivers/md/dm-snap-persistent.c index e4ecadf0548a..39becbec4dfe 100644 --- a/drivers/md/dm-snap-persistent.c +++ b/drivers/md/dm-snap-persistent.c @@ -58,25 +58,30 @@ #define NUM_SNAPSHOT_HDR_CHUNKS 1 struct disk_header { - uint32_t magic; + __le32 magic; /* * Is this snapshot valid. There is no way of recovering * an invalid snapshot. */ - uint32_t valid; + __le32 valid; /* * Simple, incrementing version. no backward * compatibility. */ - uint32_t version; + __le32 version; /* In sectors */ - uint32_t chunk_size; -}; + __le32 chunk_size; +} __packed; struct disk_exception { + __le64 old_chunk; + __le64 new_chunk; +} __packed; + +struct core_exception { uint64_t old_chunk; uint64_t new_chunk; }; @@ -396,32 +401,32 @@ static struct disk_exception *get_exception(struct pstore *ps, uint32_t index) } static void read_exception(struct pstore *ps, - uint32_t index, struct disk_exception *result) + uint32_t index, struct core_exception *result) { - struct disk_exception *e = get_exception(ps, index); + struct disk_exception *de = get_exception(ps, index); /* copy it */ - result->old_chunk = le64_to_cpu(e->old_chunk); - result->new_chunk = le64_to_cpu(e->new_chunk); + result->old_chunk = le64_to_cpu(de->old_chunk); + result->new_chunk = le64_to_cpu(de->new_chunk); } static void write_exception(struct pstore *ps, - uint32_t index, struct disk_exception *de) + uint32_t index, struct core_exception *e) { - struct disk_exception *e = get_exception(ps, index); + struct disk_exception *de = get_exception(ps, index); /* copy it */ - e->old_chunk = cpu_to_le64(de->old_chunk); - e->new_chunk = cpu_to_le64(de->new_chunk); + de->old_chunk = cpu_to_le64(e->old_chunk); + de->new_chunk = cpu_to_le64(e->new_chunk); } static void clear_exception(struct pstore *ps, uint32_t index) { - struct disk_exception *e = get_exception(ps, index); + struct disk_exception *de = get_exception(ps, index); /* clear it */ - e->old_chunk = 0; - e->new_chunk = 0; + de->old_chunk = 0; + de->new_chunk = 0; } /* @@ -437,13 +442,13 @@ static int insert_exceptions(struct pstore *ps, { int r; unsigned int i; - struct disk_exception de; + struct core_exception e; /* presume the area is full */ *full = 1; for (i = 0; i < ps->exceptions_per_area; i++) { - read_exception(ps, i, &de); + read_exception(ps, i, &e); /* * If the new_chunk is pointing at the start of @@ -451,7 +456,7 @@ static int insert_exceptions(struct pstore *ps, * is we know that we've hit the end of the * exceptions. Therefore the area is not full. */ - if (de.new_chunk == 0LL) { + if (e.new_chunk == 0LL) { ps->current_committed = i; *full = 0; break; @@ -460,13 +465,13 @@ static int insert_exceptions(struct pstore *ps, /* * Keep track of the start of the free chunks. */ - if (ps->next_free <= de.new_chunk) - ps->next_free = de.new_chunk + 1; + if (ps->next_free <= e.new_chunk) + ps->next_free = e.new_chunk + 1; /* * Otherwise we add the exception to the snapshot. */ - r = callback(callback_context, de.old_chunk, de.new_chunk); + r = callback(callback_context, e.old_chunk, e.new_chunk); if (r) return r; } @@ -641,12 +646,12 @@ static void persistent_commit_exception(struct dm_exception_store *store, { unsigned int i; struct pstore *ps = get_info(store); - struct disk_exception de; + struct core_exception ce; struct commit_callback *cb; - de.old_chunk = e->old_chunk; - de.new_chunk = e->new_chunk; - write_exception(ps, ps->current_committed++, &de); + ce.old_chunk = e->old_chunk; + ce.new_chunk = e->new_chunk; + write_exception(ps, ps->current_committed++, &ce); /* * Add the callback to the back of the array. This code @@ -701,7 +706,7 @@ static int persistent_prepare_merge(struct dm_exception_store *store, chunk_t *last_new_chunk) { struct pstore *ps = get_info(store); - struct disk_exception de; + struct core_exception ce; int nr_consecutive; int r; @@ -722,9 +727,9 @@ static int persistent_prepare_merge(struct dm_exception_store *store, ps->current_committed = ps->exceptions_per_area; } - read_exception(ps, ps->current_committed - 1, &de); - *last_old_chunk = de.old_chunk; - *last_new_chunk = de.new_chunk; + read_exception(ps, ps->current_committed - 1, &ce); + *last_old_chunk = ce.old_chunk; + *last_new_chunk = ce.new_chunk; /* * Find number of consecutive chunks within the current area, @@ -733,9 +738,9 @@ static int persistent_prepare_merge(struct dm_exception_store *store, for (nr_consecutive = 1; nr_consecutive < ps->current_committed; nr_consecutive++) { read_exception(ps, ps->current_committed - 1 - nr_consecutive, - &de); - if (de.old_chunk != *last_old_chunk - nr_consecutive || - de.new_chunk != *last_new_chunk - nr_consecutive) + &ce); + if (ce.old_chunk != *last_old_chunk - nr_consecutive || + ce.new_chunk != *last_new_chunk - nr_consecutive) break; } From 936688d7eb0f39be96c5791be1a04994cc8d6aa0 Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Tue, 2 Aug 2011 12:32:01 +0100 Subject: [PATCH 169/439] dm table: fix discard support Remove 'discards_supported' from the dm_table structure. The same information can be easily discovered from the table's target(s) in dm_table_supports_discards(). Before this fix dm_table_supports_discards() would skip checking the individual targets' 'discards_supported' flag if any one target in the table didn't set num_discard_requests > 0. Now the per-target 'discards_supported' flag is effective at insuring the final DM device advertises discard support. But, to be clear, targets that don't support discards (!num_discard_requests) will not receive discard requests. Also DMWARN if a target sets 'discards_supported' override but forgets to set 'num_discard_requests'. Signed-off-by: Mike Snitzer Signed-off-by: Alasdair G Kergon --- drivers/md/dm-table.c | 15 +++++++-------- drivers/md/dm.c | 3 ++- 2 files changed, 9 insertions(+), 9 deletions(-) diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index bfe9c2333cea..3909fa259f55 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -54,7 +54,6 @@ struct dm_table { sector_t *highs; struct dm_target *targets; - unsigned discards_supported:1; unsigned integrity_supported:1; /* @@ -209,7 +208,6 @@ int dm_table_create(struct dm_table **result, fmode_t mode, INIT_LIST_HEAD(&t->devices); INIT_LIST_HEAD(&t->target_callbacks); atomic_set(&t->holders, 0); - t->discards_supported = 1; if (!num_targets) num_targets = KEYS_PER_NODE; @@ -791,8 +789,9 @@ int dm_table_add_target(struct dm_table *t, const char *type, t->highs[t->num_targets++] = tgt->begin + tgt->len - 1; - if (!tgt->num_discard_requests) - t->discards_supported = 0; + if (!tgt->num_discard_requests && tgt->discards_supported) + DMWARN("%s: %s: ignoring discards_supported because num_discard_requests is zero.", + dm_device_name(t->md), type); return 0; @@ -1359,19 +1358,19 @@ bool dm_table_supports_discards(struct dm_table *t) struct dm_target *ti; unsigned i = 0; - if (!t->discards_supported) - return 0; - /* * Unless any target used by the table set discards_supported, * require at least one underlying device to support discards. * t->devices includes internal dm devices such as mirror logs * so we need to use iterate_devices here, which targets - * supporting discard must provide. + * supporting discard selectively must provide. */ while (i < dm_table_get_num_targets(t)) { ti = dm_table_get_target(t, i++); + if (!ti->num_discard_requests) + continue; + if (ti->discards_supported) return 1; diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 41abc6dd481b..aeb0fa1ccfe4 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -1179,7 +1179,8 @@ static int __clone_and_map_discard(struct clone_info *ci) /* * Even though the device advertised discard support, - * reconfiguration might have changed that since the + * that does not mean every target supports it, and + * reconfiguration might also have changed that since the * check was performed. */ if (!ti->num_discard_requests) From c8f543e0786785d8c7118fc2878e42bc0193a88d Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Tue, 2 Aug 2011 12:32:01 +0100 Subject: [PATCH 170/439] dm log: clean up bit little endian bitops Using __test_and_{set,clear}_bit_le() with ignoring its return value can be replaced with __{set,clear}_bit_le(). This also removes unnecessary casts. Signed-off-by: Akinobu Mita Signed-off-by: Alasdair G Kergon --- drivers/md/dm-log.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/drivers/md/dm-log.c b/drivers/md/dm-log.c index 5f06fb687408..306ce123d6b4 100644 --- a/drivers/md/dm-log.c +++ b/drivers/md/dm-log.c @@ -257,20 +257,20 @@ struct log_c { */ static inline int log_test_bit(uint32_t *bs, unsigned bit) { - return test_bit_le(bit, (unsigned long *) bs) ? 1 : 0; + return test_bit_le(bit, bs) ? 1 : 0; } static inline void log_set_bit(struct log_c *l, uint32_t *bs, unsigned bit) { - __test_and_set_bit_le(bit, (unsigned long *) bs); + __set_bit_le(bit, bs); l->touched_cleaned = 1; } static inline void log_clear_bit(struct log_c *l, uint32_t *bs, unsigned bit) { - __test_and_clear_bit_le(bit, (unsigned long *) bs); + __clear_bit_le(bit, bs); l->touched_dirtied = 1; } @@ -745,8 +745,7 @@ static int core_get_resync_work(struct dm_dirty_log *log, region_t *region) return 0; do { - *region = find_next_zero_bit_le( - (unsigned long *) lc->sync_bits, + *region = find_next_zero_bit_le(lc->sync_bits, lc->region_count, lc->sync_search); lc->sync_search = *region + 1; From 6c9b27ab08aaf46426515b8b858ad9c60731c7a1 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Tue, 2 Aug 2011 12:32:02 +0100 Subject: [PATCH 171/439] dm log: userspace use list_move Replace list_del() followed by list_add() with list_move(). Signed-off-by: Kirill A. Shutemov Signed-off-by: Alasdair G Kergon --- drivers/md/dm-log-userspace-base.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/md/dm-log-userspace-base.c b/drivers/md/dm-log-userspace-base.c index aa2e0c374ab3..1021c8986011 100644 --- a/drivers/md/dm-log-userspace-base.c +++ b/drivers/md/dm-log-userspace-base.c @@ -394,8 +394,7 @@ static int flush_by_group(struct log_c *lc, struct list_head *flush_list) group[count] = fe->region; count++; - list_del(&fe->list); - list_add(&fe->list, &tmp_list); + list_move(&fe->list, &tmp_list); type = fe->type; if (count >= MAX_FLUSH_GROUP_COUNT) From e29e65aacbd9e628378084905cbcf62a9fa4a8cc Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Tue, 2 Aug 2011 12:32:02 +0100 Subject: [PATCH 172/439] dm: use vzalloc Use vzalloc() instead of vmalloc()+memset(). Signed-off-by: Joe Perches Signed-off-by: Alasdair G Kergon --- drivers/md/dm-log.c | 3 +-- drivers/md/dm-snap-persistent.c | 3 +-- drivers/md/dm-table.c | 4 +--- 3 files changed, 3 insertions(+), 7 deletions(-) diff --git a/drivers/md/dm-log.c b/drivers/md/dm-log.c index 306ce123d6b4..3b52bb72bd1f 100644 --- a/drivers/md/dm-log.c +++ b/drivers/md/dm-log.c @@ -492,7 +492,7 @@ static int create_log_context(struct dm_dirty_log *log, struct dm_target *ti, memset(lc->sync_bits, (sync == NOSYNC) ? -1 : 0, bitset_size); lc->sync_count = (sync == NOSYNC) ? region_count : 0; - lc->recovering_bits = vmalloc(bitset_size); + lc->recovering_bits = vzalloc(bitset_size); if (!lc->recovering_bits) { DMWARN("couldn't allocate sync bitset"); vfree(lc->sync_bits); @@ -504,7 +504,6 @@ static int create_log_context(struct dm_dirty_log *log, struct dm_target *ti, kfree(lc); return -ENOMEM; } - memset(lc->recovering_bits, 0, bitset_size); lc->sync_search = 0; log->context = lc; diff --git a/drivers/md/dm-snap-persistent.c b/drivers/md/dm-snap-persistent.c index 39becbec4dfe..1a0acb8abdf6 100644 --- a/drivers/md/dm-snap-persistent.c +++ b/drivers/md/dm-snap-persistent.c @@ -174,10 +174,9 @@ static int alloc_area(struct pstore *ps) if (!ps->area) goto err_area; - ps->zero_area = vmalloc(len); + ps->zero_area = vzalloc(len); if (!ps->zero_area) goto err_zero_area; - memset(ps->zero_area, 0, len); ps->header_area = vmalloc(len); if (!ps->header_area) diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index 3909fa259f55..8dc67555e736 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -153,9 +153,7 @@ void *dm_vcalloc(unsigned long nmemb, unsigned long elem_size) return NULL; size = nmemb * elem_size; - addr = vmalloc(size); - if (addr) - memset(addr, 0, size); + addr = vzalloc(size); return addr; } From 4622afb3f50e03ce6da42002e7ed3675dfafc187 Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Tue, 2 Aug 2011 12:32:02 +0100 Subject: [PATCH 173/439] dm kcopyd: remove offset field from job structure The offset field in struct kcopyd_job is always zero so remove it. Signed-off-by: Mikulas Patocka Signed-off-by: Alasdair G Kergon --- drivers/md/dm-kcopyd.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/drivers/md/dm-kcopyd.c b/drivers/md/dm-kcopyd.c index 320401dec104..e7926fa1eef2 100644 --- a/drivers/md/dm-kcopyd.c +++ b/drivers/md/dm-kcopyd.c @@ -224,7 +224,6 @@ struct kcopyd_job { unsigned int num_dests; struct dm_io_region dests[DM_KCOPYD_MAX_REGIONS]; - sector_t offset; unsigned int nr_pages; struct page_list *pages; @@ -380,7 +379,7 @@ static int run_io_job(struct kcopyd_job *job) .bi_rw = job->rw, .mem.type = DM_IO_PAGE_LIST, .mem.ptr.pl = job->pages, - .mem.offset = job->offset, + .mem.offset = 0, .notify.fn = complete_io, .notify.context = job, .client = job->kc->io_client, @@ -398,8 +397,7 @@ static int run_pages_job(struct kcopyd_job *job) { int r; - job->nr_pages = dm_div_up(job->dests[0].count + job->offset, - PAGE_SIZE >> 9); + job->nr_pages = dm_div_up(job->dests[0].count, PAGE_SIZE >> 9); r = kcopyd_get_pages(job->kc, job->nr_pages, &job->pages); if (!r) { /* this job is ready for io */ @@ -602,7 +600,6 @@ int dm_kcopyd_copy(struct dm_kcopyd_client *kc, struct dm_io_region *from, job->num_dests = num_dests; memcpy(&job->dests, dests, sizeof(*dests) * num_dests); - job->offset = 0; job->nr_pages = 0; job->pages = NULL; From 5bf45a3dcdba9ff43959f7b5b44523fab254c19c Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Tue, 2 Aug 2011 12:32:02 +0100 Subject: [PATCH 174/439] dm kcopyd: remove nr_pages field from job structure The nr_pages field in struct kcopyd_job is only used temporarily in run_pages_job() to count the number of required pages. We can use a local variable instead. Signed-off-by: Mikulas Patocka Signed-off-by: Alasdair G Kergon --- drivers/md/dm-kcopyd.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/md/dm-kcopyd.c b/drivers/md/dm-kcopyd.c index e7926fa1eef2..98725e119324 100644 --- a/drivers/md/dm-kcopyd.c +++ b/drivers/md/dm-kcopyd.c @@ -224,7 +224,6 @@ struct kcopyd_job { unsigned int num_dests; struct dm_io_region dests[DM_KCOPYD_MAX_REGIONS]; - unsigned int nr_pages; struct page_list *pages; /* @@ -396,9 +395,9 @@ static int run_io_job(struct kcopyd_job *job) static int run_pages_job(struct kcopyd_job *job) { int r; + unsigned nr_pages = dm_div_up(job->dests[0].count, PAGE_SIZE >> 9); - job->nr_pages = dm_div_up(job->dests[0].count, PAGE_SIZE >> 9); - r = kcopyd_get_pages(job->kc, job->nr_pages, &job->pages); + r = kcopyd_get_pages(job->kc, nr_pages, &job->pages); if (!r) { /* this job is ready for io */ push(&job->kc->io_jobs, job); @@ -600,7 +599,6 @@ int dm_kcopyd_copy(struct dm_kcopyd_client *kc, struct dm_io_region *from, job->num_dests = num_dests; memcpy(&job->dests, dests, sizeof(*dests) * num_dests); - job->nr_pages = 0; job->pages = NULL; job->fn = fn; From aa3f0794d279cd154ac100f92ff3904ea1f56de2 Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Tue, 2 Aug 2011 12:32:03 +0100 Subject: [PATCH 175/439] dm snapshot: remove unused definitions Remove a couple of unused #defines. Signed-off-by: Mikulas Patocka Signed-off-by: Alasdair G Kergon --- drivers/md/dm-snap.c | 10 ---------- 1 file changed, 10 deletions(-) diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c index 9ecff5f3023a..9d6daa0eaff6 100644 --- a/drivers/md/dm-snap.c +++ b/drivers/md/dm-snap.c @@ -29,16 +29,6 @@ static const char dm_snapshot_merge_target_name[] = "snapshot-merge"; #define dm_target_is_snapshot_merge(ti) \ ((ti)->type->name == dm_snapshot_merge_target_name) -/* - * The percentage increment we will wake up users at - */ -#define WAKE_UP_PERCENT 5 - -/* - * kcopyd priority of snapshot operations - */ -#define SNAPSHOT_COPY_PRIORITY 2 - /* * The size of the mempool used to track chunks in use. */ From a2d2b0345a0f30c169b7d08b8cebdd4853fcb0f8 Mon Sep 17 00:00:00 2001 From: Jonathan Brassow Date: Tue, 2 Aug 2011 12:32:03 +0100 Subject: [PATCH 176/439] dm snapshot: style cleanups Coding style cleanups. Signed-off-by: Alasdair G Kergon Signed-off-by: Jonathan Brassow --- drivers/md/dm-snap-persistent.c | 4 ++-- drivers/md/dm-snap.c | 14 ++++++-------- 2 files changed, 8 insertions(+), 10 deletions(-) diff --git a/drivers/md/dm-snap-persistent.c b/drivers/md/dm-snap-persistent.c index 1a0acb8abdf6..d1f1d7017103 100644 --- a/drivers/md/dm-snap-persistent.c +++ b/drivers/md/dm-snap-persistent.c @@ -567,7 +567,7 @@ static int persistent_read_metadata(struct dm_exception_store *store, ps->exceptions_per_area = (ps->store->chunk_size << SECTOR_SHIFT) / sizeof(struct disk_exception); ps->callbacks = dm_vcalloc(ps->exceptions_per_area, - sizeof(*ps->callbacks)); + sizeof(*ps->callbacks)); if (!ps->callbacks) return -ENOMEM; @@ -674,7 +674,7 @@ static void persistent_commit_exception(struct dm_exception_store *store, * If we completely filled the current area, then wipe the next one. */ if ((ps->current_committed == ps->exceptions_per_area) && - zero_disk_area(ps, ps->current_area + 1)) + zero_disk_area(ps, ps->current_area + 1)) ps->valid = 0; /* diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c index 9d6daa0eaff6..94dee05dd28e 100644 --- a/drivers/md/dm-snap.c +++ b/drivers/md/dm-snap.c @@ -1045,8 +1045,7 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv) s = kmalloc(sizeof(*s), GFP_KERNEL); if (!s) { - ti->error = "Cannot allocate snapshot context private " - "structure"; + ti->error = "Cannot allocate private snapshot structure"; r = -ENOMEM; goto bad; } @@ -1405,7 +1404,7 @@ static void pending_complete(struct dm_snap_pending_exception *pe, int success) */ dm_insert_exception(&s->complete, e); - out: +out: dm_remove_exception(&pe->e); snapshot_bios = bio_list_get(&pe->snapshot_bios); origin_bios = bio_list_get(&pe->origin_bios); @@ -1470,8 +1469,7 @@ static void start_copy(struct dm_snap_pending_exception *pe) dest.count = src.count; /* Hand over to kcopyd */ - dm_kcopyd_copy(s->kcopyd_client, - &src, 1, &dest, 0, copy_callback, pe); + dm_kcopyd_copy(s->kcopyd_client, &src, 1, &dest, 0, copy_callback, pe); } static struct dm_snap_pending_exception * @@ -1618,9 +1616,9 @@ static int snapshot_map(struct dm_target *ti, struct bio *bio, map_context->ptr = track_chunk(s, chunk); } - out_unlock: +out_unlock: up_write(&s->lock); - out: +out: return r; } @@ -1964,7 +1962,7 @@ static int __origin_write(struct list_head *snapshots, sector_t sector, pe_to_start_now = pe; } - next_snapshot: +next_snapshot: up_write(&snap->lock); if (pe_to_start_now) { From 13c87583ea4e867211fc3e7edab750c353c47c95 Mon Sep 17 00:00:00 2001 From: Jonathan Brassow Date: Tue, 2 Aug 2011 12:32:03 +0100 Subject: [PATCH 177/439] dm raid: cleanup parameter handling Re-order the parameters so they are handled consistently in the same order where defined, parsed and output. Only include rebuild parameters in the STATUSTYPE_TABLE output if they were supplied in the original table line. Correct the parameter count when outputting rebuild: there are two words, not one. Use case-independent checks for keywords (as in other device-mapper targets). Signed-off-by: Jonathan Brassow Signed-off-by: Alasdair G Kergon --- drivers/md/dm-raid.c | 42 +++++++++++++++++++++++------------------- 1 file changed, 23 insertions(+), 19 deletions(-) diff --git a/drivers/md/dm-raid.c b/drivers/md/dm-raid.c index e5d8904fc8f6..efa960ff5ba4 100644 --- a/drivers/md/dm-raid.c +++ b/drivers/md/dm-raid.c @@ -43,13 +43,14 @@ struct raid_dev { /* * Flags for rs->print_flags field. */ -#define DMPF_DAEMON_SLEEP 0x1 -#define DMPF_MAX_WRITE_BEHIND 0x2 -#define DMPF_SYNC 0x4 -#define DMPF_NOSYNC 0x8 -#define DMPF_STRIPE_CACHE 0x10 -#define DMPF_MIN_RECOVERY_RATE 0x20 -#define DMPF_MAX_RECOVERY_RATE 0x40 +#define DMPF_SYNC 0x1 +#define DMPF_NOSYNC 0x2 +#define DMPF_REBUILD 0x4 +#define DMPF_DAEMON_SLEEP 0x8 +#define DMPF_MIN_RECOVERY_RATE 0x10 +#define DMPF_MAX_RECOVERY_RATE 0x20 +#define DMPF_MAX_WRITE_BEHIND 0x40 +#define DMPF_STRIPE_CACHE 0x80 struct raid_set { struct dm_target *ti; @@ -275,13 +276,13 @@ static int parse_raid_params(struct raid_set *rs, char **argv, set_bit(In_sync, &rs->dev[i].rdev.flags); for (i = 0; i < num_raid_params; i++) { - if (!strcmp(argv[i], "nosync")) { + if (!strcasecmp(argv[i], "nosync")) { rs->md.recovery_cp = MaxSector; rs->print_flags |= DMPF_NOSYNC; rs->md.flags |= MD_SYNC_STATE_FORCED; continue; } - if (!strcmp(argv[i], "sync")) { + if (!strcasecmp(argv[i], "sync")) { rs->md.recovery_cp = 0; rs->print_flags |= DMPF_SYNC; rs->md.flags |= MD_SYNC_STATE_FORCED; @@ -300,7 +301,7 @@ static int parse_raid_params(struct raid_set *rs, char **argv, return -EINVAL; } - if (!strcmp(key, "rebuild")) { + if (!strcasecmp(key, "rebuild")) { if (++rebuild_cnt > rs->raid_type->parity_devs) { rs->ti->error = "Too many rebuild drives given"; return -EINVAL; @@ -311,7 +312,8 @@ static int parse_raid_params(struct raid_set *rs, char **argv, } clear_bit(In_sync, &rs->dev[value].rdev.flags); rs->dev[value].rdev.recovery_offset = 0; - } else if (!strcmp(key, "max_write_behind")) { + rs->print_flags |= DMPF_REBUILD; + } else if (!strcasecmp(key, "max_write_behind")) { rs->print_flags |= DMPF_MAX_WRITE_BEHIND; /* @@ -324,14 +326,14 @@ static int parse_raid_params(struct raid_set *rs, char **argv, return -EINVAL; } rs->md.bitmap_info.max_write_behind = value; - } else if (!strcmp(key, "daemon_sleep")) { + } else if (!strcasecmp(key, "daemon_sleep")) { rs->print_flags |= DMPF_DAEMON_SLEEP; if (!value || (value > MAX_SCHEDULE_TIMEOUT)) { rs->ti->error = "daemon sleep period out of range"; return -EINVAL; } rs->md.bitmap_info.daemon_sleep = value; - } else if (!strcmp(key, "stripe_cache")) { + } else if (!strcasecmp(key, "stripe_cache")) { rs->print_flags |= DMPF_STRIPE_CACHE; /* @@ -348,14 +350,14 @@ static int parse_raid_params(struct raid_set *rs, char **argv, rs->ti->error = "Bad stripe_cache size"; return -EINVAL; } - } else if (!strcmp(key, "min_recovery_rate")) { + } else if (!strcasecmp(key, "min_recovery_rate")) { rs->print_flags |= DMPF_MIN_RECOVERY_RATE; if (value > INT_MAX) { rs->ti->error = "min_recovery_rate out of range"; return -EINVAL; } rs->md.sync_speed_min = (int)value; - } else if (!strcmp(key, "max_recovery_rate")) { + } else if (!strcasecmp(key, "max_recovery_rate")) { rs->print_flags |= DMPF_MAX_RECOVERY_RATE; if (value > INT_MAX) { rs->ti->error = "max_recovery_rate out of range"; @@ -547,11 +549,12 @@ static int raid_status(struct dm_target *ti, status_type_t type, case STATUSTYPE_TABLE: /* The string you would use to construct this array */ for (i = 0; i < rs->md.raid_disks; i++) - if (rs->dev[i].data_dev && + if ((rs->print_flags & DMPF_REBUILD) && + rs->dev[i].data_dev && !test_bit(In_sync, &rs->dev[i].rdev.flags)) - raid_param_cnt++; /* for rebuilds */ + raid_param_cnt += 2; /* for rebuilds */ - raid_param_cnt += (hweight64(rs->print_flags) * 2); + raid_param_cnt += (hweight64(rs->print_flags & ~DMPF_REBUILD) * 2); if (rs->print_flags & (DMPF_SYNC | DMPF_NOSYNC)) raid_param_cnt--; @@ -565,7 +568,8 @@ static int raid_status(struct dm_target *ti, status_type_t type, DMEMIT(" nosync"); for (i = 0; i < rs->md.raid_disks; i++) - if (rs->dev[i].data_dev && + if ((rs->print_flags & DMPF_REBUILD) && + rs->dev[i].data_dev && !test_bit(In_sync, &rs->dev[i].rdev.flags)) DMEMIT(" rebuild %u", i); From 2ca4c92f58f9386e080b26f9ccd78c9ca5825a42 Mon Sep 17 00:00:00 2001 From: Alasdair G Kergon Date: Tue, 2 Aug 2011 12:32:03 +0100 Subject: [PATCH 178/439] dm ioctl: prevent empty message Detect invalid empty messages in core dm instead of requiring every target to check this. Signed-off-by: Alasdair G Kergon --- drivers/md/dm-ioctl.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index 4cacdad2270a..1622a6bc0bfd 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -1402,6 +1402,11 @@ static int target_message(struct dm_ioctl *param, size_t param_size) goto out; } + if (!argc) { + DMWARN("Empty message received."); + goto out; + } + table = dm_get_live_table(md); if (!table) goto out_argv; From 3e8dbb7f3966c80d77272f8b4f430babc99f6595 Mon Sep 17 00:00:00 2001 From: Alasdair G Kergon Date: Tue, 2 Aug 2011 12:32:03 +0100 Subject: [PATCH 179/439] dm raid: tidy includes A dm target only needs to use include/linux dm headers. Signed-off-by: Alasdair G Kergon --- drivers/md/dm-raid.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/md/dm-raid.c b/drivers/md/dm-raid.c index efa960ff5ba4..d4e95b2e39f6 100644 --- a/drivers/md/dm-raid.c +++ b/drivers/md/dm-raid.c @@ -9,9 +9,10 @@ #include "md.h" #include "raid5.h" -#include "dm.h" #include "bitmap.h" +#include + #define DM_MSG_PREFIX "raid" /* From 08649012545cfb116798260352547cf4d47064ec Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Tue, 2 Aug 2011 12:32:04 +0100 Subject: [PATCH 180/439] dm table: clean dm_get_device and move exports There is no need for __table_get_device to be factored out. Also move the exports to the end of their respective functions. Signed-off-by: Mike Snitzer Signed-off-by: Alasdair G Kergon --- drivers/md/dm-table.c | 33 +++++++++++++-------------------- 1 file changed, 13 insertions(+), 20 deletions(-) diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index 8dc67555e736..d58b200a3531 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -157,6 +157,7 @@ void *dm_vcalloc(unsigned long nmemb, unsigned long elem_size) return addr; } +EXPORT_SYMBOL(dm_vcalloc); /* * highs, and targets are managed as dynamic arrays during a @@ -277,6 +278,7 @@ void dm_table_get(struct dm_table *t) { atomic_inc(&t->holders); } +EXPORT_SYMBOL(dm_table_get); void dm_table_put(struct dm_table *t) { @@ -286,6 +288,7 @@ void dm_table_put(struct dm_table *t) smp_mb__before_atomic_dec(); atomic_dec(&t->holders); } +EXPORT_SYMBOL(dm_table_put); /* * Checks to see if we need to extend highs or targets. @@ -451,13 +454,14 @@ static int upgrade_mode(struct dm_dev_internal *dd, fmode_t new_mode, * Add a device to the list, or just increment the usage count if * it's already present. */ -static int __table_get_device(struct dm_table *t, struct dm_target *ti, - const char *path, fmode_t mode, struct dm_dev **result) +int dm_get_device(struct dm_target *ti, const char *path, fmode_t mode, + struct dm_dev **result) { int r; dev_t uninitialized_var(dev); struct dm_dev_internal *dd; unsigned int major, minor; + struct dm_table *t = ti->table; BUG_ON(!t); @@ -505,6 +509,7 @@ static int __table_get_device(struct dm_table *t, struct dm_target *ti, *result = &dd->dm_dev; return 0; } +EXPORT_SYMBOL(dm_get_device); int dm_set_device_limits(struct dm_target *ti, struct dm_dev *dev, sector_t start, sector_t len, void *data) @@ -543,15 +548,8 @@ int dm_set_device_limits(struct dm_target *ti, struct dm_dev *dev, } EXPORT_SYMBOL_GPL(dm_set_device_limits); -int dm_get_device(struct dm_target *ti, const char *path, fmode_t mode, - struct dm_dev **result) -{ - return __table_get_device(ti->table, ti, path, mode, result); -} - - /* - * Decrement a devices use count and remove it if necessary. + * Decrement a device's use count and remove it if necessary. */ void dm_put_device(struct dm_target *ti, struct dm_dev *d) { @@ -564,6 +562,7 @@ void dm_put_device(struct dm_target *ti, struct dm_dev *d) kfree(dd); } } +EXPORT_SYMBOL(dm_put_device); /* * Checks to see if the target joins onto the end of the table. @@ -1074,11 +1073,13 @@ void dm_table_event(struct dm_table *t) t->event_fn(t->event_context); mutex_unlock(&_event_lock); } +EXPORT_SYMBOL(dm_table_event); sector_t dm_table_get_size(struct dm_table *t) { return t->num_targets ? (t->highs[t->num_targets - 1] + 1) : 0; } +EXPORT_SYMBOL(dm_table_get_size); struct dm_target *dm_table_get_target(struct dm_table *t, unsigned int index) { @@ -1234,6 +1235,7 @@ fmode_t dm_table_get_mode(struct dm_table *t) { return t->mode; } +EXPORT_SYMBOL(dm_table_get_mode); static void suspend_targets(struct dm_table *t, unsigned postsuspend) { @@ -1342,6 +1344,7 @@ struct mapped_device *dm_table_get_md(struct dm_table *t) { return t->md; } +EXPORT_SYMBOL(dm_table_get_md); static int device_discard_capable(struct dm_target *ti, struct dm_dev *dev, sector_t start, sector_t len, void *data) @@ -1379,13 +1382,3 @@ bool dm_table_supports_discards(struct dm_table *t) return 0; } - -EXPORT_SYMBOL(dm_vcalloc); -EXPORT_SYMBOL(dm_get_device); -EXPORT_SYMBOL(dm_put_device); -EXPORT_SYMBOL(dm_table_event); -EXPORT_SYMBOL(dm_table_get_size); -EXPORT_SYMBOL(dm_table_get_mode); -EXPORT_SYMBOL(dm_table_get_md); -EXPORT_SYMBOL(dm_table_put); -EXPORT_SYMBOL(dm_table_get); From d5b9dd04bd74b774b8e8d93ced7a0d15ad403fa9 Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Tue, 2 Aug 2011 12:32:04 +0100 Subject: [PATCH 181/439] dm: ignore merge_bvec for snapshots when safe Add a new flag DMF_MERGE_IS_OPTIONAL to struct mapped_device to indicate whether the device can accept bios larger than the size its merge function returns. When set, use this to send large bios to snapshots which can split them if necessary. Snapshot I/O may be significantly fragmented and this approach seems to improve peformance. Before the patch, dm_set_device_limits restricted bio size to page size if the underlying device had a merge function and the target didn't provide a merge function. After the patch, dm_set_device_limits restricts bio size to page size if the underlying device has a merge function, doesn't have DMF_MERGE_IS_OPTIONAL flag and the target doesn't provide a merge function. The snapshot target can't provide a merge function because when the merge function is called, it is impossible to determine where the bio will be remapped. Previously this led us to impose a 4k limit, which we can now remove if the snapshot store is located on a device without a merge function. Together with another patch for optimizing full chunk writes, it improves performance from 29MB/s to 40MB/s when writing to the filesystem on snapshot store. If the snapshot store is placed on a non-dm device with a merge function (such as md-raid), device mapper still limits all bios to page size. Signed-off-by: Mikulas Patocka Signed-off-by: Alasdair G Kergon --- drivers/md/dm-table.c | 3 +-- drivers/md/dm.c | 61 +++++++++++++++++++++++++++++++++++++++++++ drivers/md/dm.h | 2 ++ 3 files changed, 64 insertions(+), 2 deletions(-) diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index d58b200a3531..03516c7f5be1 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -540,8 +540,7 @@ int dm_set_device_limits(struct dm_target *ti, struct dm_dev *dev, * If not we'll force DM to use PAGE_SIZE or * smaller I/O, just to be safe. */ - - if (q->merge_bvec_fn && !ti->type->merge) + if (dm_queue_merge_is_compulsory(q) && !ti->type->merge) blk_limits_max_hw_sectors(limits, (unsigned int) (PAGE_SIZE >> 9)); return 0; diff --git a/drivers/md/dm.c b/drivers/md/dm.c index aeb0fa1ccfe4..1000eaf984ef 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -111,6 +111,7 @@ EXPORT_SYMBOL_GPL(dm_get_rq_mapinfo); #define DMF_FREEING 3 #define DMF_DELETING 4 #define DMF_NOFLUSH_SUSPENDING 5 +#define DMF_MERGE_IS_OPTIONAL 6 /* * Work processed by per-device workqueue. @@ -1992,6 +1993,59 @@ static void __set_size(struct mapped_device *md, sector_t size) i_size_write(md->bdev->bd_inode, (loff_t)size << SECTOR_SHIFT); } +/* + * Return 1 if the queue has a compulsory merge_bvec_fn function. + * + * If this function returns 0, then the device is either a non-dm + * device without a merge_bvec_fn, or it is a dm device that is + * able to split any bios it receives that are too big. + */ +int dm_queue_merge_is_compulsory(struct request_queue *q) +{ + struct mapped_device *dev_md; + + if (!q->merge_bvec_fn) + return 0; + + if (q->make_request_fn == dm_request) { + dev_md = q->queuedata; + if (test_bit(DMF_MERGE_IS_OPTIONAL, &dev_md->flags)) + return 0; + } + + return 1; +} + +static int dm_device_merge_is_compulsory(struct dm_target *ti, + struct dm_dev *dev, sector_t start, + sector_t len, void *data) +{ + struct block_device *bdev = dev->bdev; + struct request_queue *q = bdev_get_queue(bdev); + + return dm_queue_merge_is_compulsory(q); +} + +/* + * Return 1 if it is acceptable to ignore merge_bvec_fn based + * on the properties of the underlying devices. + */ +static int dm_table_merge_is_optional(struct dm_table *table) +{ + unsigned i = 0; + struct dm_target *ti; + + while (i < dm_table_get_num_targets(table)) { + ti = dm_table_get_target(table, i++); + + if (ti->type->iterate_devices && + ti->type->iterate_devices(ti, dm_device_merge_is_compulsory, NULL)) + return 0; + } + + return 1; +} + /* * Returns old map, which caller must destroy. */ @@ -2002,6 +2056,7 @@ static struct dm_table *__bind(struct mapped_device *md, struct dm_table *t, struct request_queue *q = md->queue; sector_t size; unsigned long flags; + int merge_is_optional; size = dm_table_get_size(t); @@ -2027,10 +2082,16 @@ static struct dm_table *__bind(struct mapped_device *md, struct dm_table *t, __bind_mempools(md, t); + merge_is_optional = dm_table_merge_is_optional(t); + write_lock_irqsave(&md->map_lock, flags); old_map = md->map; md->map = t; dm_table_set_restrictions(t, q, limits); + if (merge_is_optional) + set_bit(DMF_MERGE_IS_OPTIONAL, &md->flags); + else + clear_bit(DMF_MERGE_IS_OPTIONAL, &md->flags); write_unlock_irqrestore(&md->map_lock, flags); return old_map; diff --git a/drivers/md/dm.h b/drivers/md/dm.h index 1aaf16746da8..6745dbd278a4 100644 --- a/drivers/md/dm.h +++ b/drivers/md/dm.h @@ -66,6 +66,8 @@ int dm_table_alloc_md_mempools(struct dm_table *t); void dm_table_free_md_mempools(struct dm_table *t); struct dm_md_mempools *dm_table_get_md_mempools(struct dm_table *t); +int dm_queue_merge_is_compulsory(struct request_queue *q); + void dm_lock_md_type(struct mapped_device *md); void dm_unlock_md_type(struct mapped_device *md); void dm_set_md_type(struct mapped_device *md, unsigned type); From a6e50b409d3f9e0833e69c3c9cca822e8fa4adbb Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Tue, 2 Aug 2011 12:32:04 +0100 Subject: [PATCH 182/439] dm snapshot: skip reading origin when overwriting complete chunk If we write a full chunk in the snapshot, skip reading the origin device because the whole chunk will be overwritten anyway. This patch changes the snapshot write logic when a full chunk is written. In this case: 1. allocate the exception 2. dispatch the bio (but don't report the bio completion to device mapper) 3. write the exception record 4. report bio completed Callbacks must be done through the kcopyd thread, because callbacks must not race with each other. So we create two new functions: dm_kcopyd_prepare_callback: allocate a job structure and prepare the callback. (This function must not be called from interrupt context.) dm_kcopyd_do_callback: submit callback. (This function may be called from interrupt context.) Performance test (on snapshots with 4k chunk size): without the patch: non-direct-io sequential write (dd): 17.7MB/s direct-io sequential write (dd): 20.9MB/s non-direct-io random write (mkfs.ext2): 0.44s with the patch: non-direct-io sequential write (dd): 26.5MB/s direct-io sequential write (dd): 33.2MB/s non-direct-io random write (mkfs.ext2): 0.27s Signed-off-by: Mikulas Patocka Signed-off-by: Alasdair G Kergon --- drivers/md/dm-kcopyd.c | 31 ++++++++++++++++++++ drivers/md/dm-snap.c | 60 +++++++++++++++++++++++++++++++++++++-- include/linux/dm-kcopyd.h | 15 ++++++++++ 3 files changed, 103 insertions(+), 3 deletions(-) diff --git a/drivers/md/dm-kcopyd.c b/drivers/md/dm-kcopyd.c index 98725e119324..f82147029636 100644 --- a/drivers/md/dm-kcopyd.c +++ b/drivers/md/dm-kcopyd.c @@ -617,6 +617,37 @@ int dm_kcopyd_copy(struct dm_kcopyd_client *kc, struct dm_io_region *from, } EXPORT_SYMBOL(dm_kcopyd_copy); +void *dm_kcopyd_prepare_callback(struct dm_kcopyd_client *kc, + dm_kcopyd_notify_fn fn, void *context) +{ + struct kcopyd_job *job; + + job = mempool_alloc(kc->job_pool, GFP_NOIO); + + memset(job, 0, sizeof(struct kcopyd_job)); + job->kc = kc; + job->fn = fn; + job->context = context; + + atomic_inc(&kc->nr_jobs); + + return job; +} +EXPORT_SYMBOL(dm_kcopyd_prepare_callback); + +void dm_kcopyd_do_callback(void *j, int read_err, unsigned long write_err) +{ + struct kcopyd_job *job = j; + struct dm_kcopyd_client *kc = job->kc; + + job->read_err = read_err; + job->write_err = write_err; + + push(&kc->complete_jobs, job); + wake(kc); +} +EXPORT_SYMBOL(dm_kcopyd_do_callback); + /* * Cancels a kcopyd job, eg. someone might be deactivating a * mirror. diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c index 94dee05dd28e..6f758870fc19 100644 --- a/drivers/md/dm-snap.c +++ b/drivers/md/dm-snap.c @@ -170,6 +170,13 @@ struct dm_snap_pending_exception { * kcopyd. */ int started; + + /* + * For writing a complete chunk, bypassing the copy. + */ + struct bio *full_bio; + bio_end_io_t *full_bio_end_io; + void *full_bio_private; }; /* @@ -1369,6 +1376,7 @@ static void pending_complete(struct dm_snap_pending_exception *pe, int success) struct dm_snapshot *s = pe->snap; struct bio *origin_bios = NULL; struct bio *snapshot_bios = NULL; + struct bio *full_bio = NULL; int error = 0; if (!success) { @@ -1408,6 +1416,11 @@ out: dm_remove_exception(&pe->e); snapshot_bios = bio_list_get(&pe->snapshot_bios); origin_bios = bio_list_get(&pe->origin_bios); + full_bio = pe->full_bio; + if (full_bio) { + full_bio->bi_end_io = pe->full_bio_end_io; + full_bio->bi_private = pe->full_bio_private; + } free_pending_exception(pe); increment_pending_exceptions_done_count(); @@ -1415,10 +1428,15 @@ out: up_write(&s->lock); /* Submit any pending write bios */ - if (error) + if (error) { + if (full_bio) + bio_io_error(full_bio); error_bios(snapshot_bios); - else + } else { + if (full_bio) + bio_endio(full_bio, 0); flush_bios(snapshot_bios); + } retry_origin_bios(s, origin_bios); } @@ -1472,6 +1490,32 @@ static void start_copy(struct dm_snap_pending_exception *pe) dm_kcopyd_copy(s->kcopyd_client, &src, 1, &dest, 0, copy_callback, pe); } +static void full_bio_end_io(struct bio *bio, int error) +{ + void *callback_data = bio->bi_private; + + dm_kcopyd_do_callback(callback_data, 0, error ? 1 : 0); +} + +static void start_full_bio(struct dm_snap_pending_exception *pe, + struct bio *bio) +{ + struct dm_snapshot *s = pe->snap; + void *callback_data; + + pe->full_bio = bio; + pe->full_bio_end_io = bio->bi_end_io; + pe->full_bio_private = bio->bi_private; + + callback_data = dm_kcopyd_prepare_callback(s->kcopyd_client, + copy_callback, pe); + + bio->bi_end_io = full_bio_end_io; + bio->bi_private = callback_data; + + generic_make_request(bio); +} + static struct dm_snap_pending_exception * __lookup_pending_exception(struct dm_snapshot *s, chunk_t chunk) { @@ -1507,6 +1551,7 @@ __find_pending_exception(struct dm_snapshot *s, bio_list_init(&pe->origin_bios); bio_list_init(&pe->snapshot_bios); pe->started = 0; + pe->full_bio = NULL; if (s->store->type->prepare_exception(s->store, &pe->e)) { free_pending_exception(pe); @@ -1600,10 +1645,19 @@ static int snapshot_map(struct dm_target *ti, struct bio *bio, } remap_exception(s, &pe->e, bio, chunk); - bio_list_add(&pe->snapshot_bios, bio); r = DM_MAPIO_SUBMITTED; + if (!pe->started && + bio->bi_size == (s->store->chunk_size << SECTOR_SHIFT)) { + pe->started = 1; + up_write(&s->lock); + start_full_bio(pe, bio); + goto out; + } + + bio_list_add(&pe->snapshot_bios, bio); + if (!pe->started) { /* this is protected by snap->lock */ pe->started = 1; diff --git a/include/linux/dm-kcopyd.h b/include/linux/dm-kcopyd.h index 298d587e349b..5e54458e920f 100644 --- a/include/linux/dm-kcopyd.h +++ b/include/linux/dm-kcopyd.h @@ -42,5 +42,20 @@ int dm_kcopyd_copy(struct dm_kcopyd_client *kc, struct dm_io_region *from, unsigned num_dests, struct dm_io_region *dests, unsigned flags, dm_kcopyd_notify_fn fn, void *context); +/* + * Prepare a callback and submit it via the kcopyd thread. + * + * dm_kcopyd_prepare_callback allocates a callback structure and returns it. + * It must not be called from interrupt context. + * The returned value should be passed into dm_kcopyd_do_callback. + * + * dm_kcopyd_do_callback submits the callback. + * It may be called from interrupt context. + * The callback is issued from the kcopyd thread. + */ +void *dm_kcopyd_prepare_callback(struct dm_kcopyd_client *kc, + dm_kcopyd_notify_fn fn, void *context); +void dm_kcopyd_do_callback(void *job, int read_err, unsigned long write_err); + #endif /* __KERNEL__ */ #endif /* _LINUX_DM_KCOPYD_H */ From 498f0103ea13123e007660def9072a0b7dd1c599 Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Tue, 2 Aug 2011 12:32:04 +0100 Subject: [PATCH 183/439] dm table: share target argument parsing functions Move multipath target argument parsing code into dm-table so other targets can share it. Signed-off-by: Mike Snitzer Signed-off-by: Alasdair G Kergon --- drivers/md/dm-crypt.c | 7 +- drivers/md/dm-mpath.c | 152 ++++++++++------------------------ drivers/md/dm-table.c | 57 +++++++++++++ include/linux/device-mapper.h | 43 ++++++++++ 4 files changed, 147 insertions(+), 112 deletions(-) diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index f5406766ece3..b79e7472a9b3 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -30,7 +30,6 @@ #include #define DM_MSG_PREFIX "crypt" -#define MESG_STR(x) x, sizeof(x) /* * context holding the current state of a multi-part conversion @@ -1770,12 +1769,12 @@ static int crypt_message(struct dm_target *ti, unsigned argc, char **argv) if (argc < 2) goto error; - if (!strnicmp(argv[0], MESG_STR("key"))) { + if (!strcasecmp(argv[0], "key")) { if (!test_bit(DM_CRYPT_SUSPENDED, &cc->flags)) { DMWARN("not suspended during key manipulation."); return -EINVAL; } - if (argc == 3 && !strnicmp(argv[1], MESG_STR("set"))) { + if (argc == 3 && !strcasecmp(argv[1], "set")) { ret = crypt_set_key(cc, argv[2]); if (ret) return ret; @@ -1783,7 +1782,7 @@ static int crypt_message(struct dm_target *ti, unsigned argc, char **argv) ret = cc->iv_gen_ops->init(cc); return ret; } - if (argc == 2 && !strnicmp(argv[1], MESG_STR("wipe"))) { + if (argc == 2 && !strcasecmp(argv[1], "wipe")) { if (cc->iv_gen_ops && cc->iv_gen_ops->wipe) { ret = cc->iv_gen_ops->wipe(cc); if (ret) diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index adf851a081bd..5e0090ef4182 100644 --- a/drivers/md/dm-mpath.c +++ b/drivers/md/dm-mpath.c @@ -22,7 +22,6 @@ #include #define DM_MSG_PREFIX "multipath" -#define MESG_STR(x) x, sizeof(x) #define DM_PG_INIT_DELAY_MSECS 2000 #define DM_PG_INIT_DELAY_DEFAULT ((unsigned) -1) @@ -505,80 +504,29 @@ static void trigger_event(struct work_struct *work) * <#paths> <#per-path selector args> * [ []* ]+ ]+ *---------------------------------------------------------------*/ -struct param { - unsigned min; - unsigned max; - char *error; -}; - -static int read_param(struct param *param, char *str, unsigned *v, char **error) -{ - if (!str || - (sscanf(str, "%u", v) != 1) || - (*v < param->min) || - (*v > param->max)) { - *error = param->error; - return -EINVAL; - } - - return 0; -} - -struct arg_set { - unsigned argc; - char **argv; -}; - -static char *shift(struct arg_set *as) -{ - char *r; - - if (as->argc) { - as->argc--; - r = *as->argv; - as->argv++; - return r; - } - - return NULL; -} - -static void consume(struct arg_set *as, unsigned n) -{ - BUG_ON (as->argc < n); - as->argc -= n; - as->argv += n; -} - -static int parse_path_selector(struct arg_set *as, struct priority_group *pg, +static int parse_path_selector(struct dm_arg_set *as, struct priority_group *pg, struct dm_target *ti) { int r; struct path_selector_type *pst; unsigned ps_argc; - static struct param _params[] = { + static struct dm_arg _args[] = { {0, 1024, "invalid number of path selector args"}, }; - pst = dm_get_path_selector(shift(as)); + pst = dm_get_path_selector(dm_shift_arg(as)); if (!pst) { ti->error = "unknown path selector type"; return -EINVAL; } - r = read_param(_params, shift(as), &ps_argc, &ti->error); + r = dm_read_arg_group(_args, as, &ps_argc, &ti->error); if (r) { dm_put_path_selector(pst); return -EINVAL; } - if (ps_argc > as->argc) { - dm_put_path_selector(pst); - ti->error = "not enough arguments for path selector"; - return -EINVAL; - } - r = pst->create(&pg->ps, ps_argc, as->argv); if (r) { dm_put_path_selector(pst); @@ -587,12 +535,12 @@ static int parse_path_selector(struct arg_set *as, struct priority_group *pg, } pg->ps.type = pst; - consume(as, ps_argc); + dm_consume_args(as, ps_argc); return 0; } -static struct pgpath *parse_path(struct arg_set *as, struct path_selector *ps, +static struct pgpath *parse_path(struct dm_arg_set *as, struct path_selector *ps, struct dm_target *ti) { int r; @@ -609,7 +557,7 @@ static struct pgpath *parse_path(struct arg_set *as, struct path_selector *ps, if (!p) return ERR_PTR(-ENOMEM); - r = dm_get_device(ti, shift(as), dm_table_get_mode(ti->table), + r = dm_get_device(ti, dm_shift_arg(as), dm_table_get_mode(ti->table), &p->path.dev); if (r) { ti->error = "error getting device"; @@ -660,16 +608,16 @@ static struct pgpath *parse_path(struct arg_set *as, struct path_selector *ps, return ERR_PTR(r); } -static struct priority_group *parse_priority_group(struct arg_set *as, +static struct priority_group *parse_priority_group(struct dm_arg_set *as, struct multipath *m) { - static struct param _params[] = { + static struct dm_arg _args[] = { {1, 1024, "invalid number of paths"}, {0, 1024, "invalid number of selector args"} }; int r; - unsigned i, nr_selector_args, nr_params; + unsigned i, nr_selector_args, nr_args; struct priority_group *pg; struct dm_target *ti = m->ti; @@ -693,26 +641,26 @@ static struct priority_group *parse_priority_group(struct arg_set *as, /* * read the paths */ - r = read_param(_params, shift(as), &pg->nr_pgpaths, &ti->error); + r = dm_read_arg(_args, as, &pg->nr_pgpaths, &ti->error); if (r) goto bad; - r = read_param(_params + 1, shift(as), &nr_selector_args, &ti->error); + r = dm_read_arg(_args + 1, as, &nr_selector_args, &ti->error); if (r) goto bad; - nr_params = 1 + nr_selector_args; + nr_args = 1 + nr_selector_args; for (i = 0; i < pg->nr_pgpaths; i++) { struct pgpath *pgpath; - struct arg_set path_args; + struct dm_arg_set path_args; - if (as->argc < nr_params) { + if (as->argc < nr_args) { ti->error = "not enough path parameters"; r = -EINVAL; goto bad; } - path_args.argc = nr_params; + path_args.argc = nr_args; path_args.argv = as->argv; pgpath = parse_path(&path_args, &pg->ps, ti); @@ -723,7 +671,7 @@ static struct priority_group *parse_priority_group(struct arg_set *as, pgpath->pg = pg; list_add_tail(&pgpath->list, &pg->pgpaths); - consume(as, nr_params); + dm_consume_args(as, nr_args); } return pg; @@ -733,28 +681,23 @@ static struct priority_group *parse_priority_group(struct arg_set *as, return ERR_PTR(r); } -static int parse_hw_handler(struct arg_set *as, struct multipath *m) +static int parse_hw_handler(struct dm_arg_set *as, struct multipath *m) { unsigned hw_argc; int ret; struct dm_target *ti = m->ti; - static struct param _params[] = { + static struct dm_arg _args[] = { {0, 1024, "invalid number of hardware handler args"}, }; - if (read_param(_params, shift(as), &hw_argc, &ti->error)) + if (dm_read_arg_group(_args, as, &hw_argc, &ti->error)) return -EINVAL; if (!hw_argc) return 0; - if (hw_argc > as->argc) { - ti->error = "not enough arguments for hardware handler"; - return -EINVAL; - } - - m->hw_handler_name = kstrdup(shift(as), GFP_KERNEL); + m->hw_handler_name = kstrdup(dm_shift_arg(as), GFP_KERNEL); request_module("scsi_dh_%s", m->hw_handler_name); if (scsi_dh_handler_exist(m->hw_handler_name) == 0) { ti->error = "unknown hardware handler type"; @@ -778,7 +721,7 @@ static int parse_hw_handler(struct arg_set *as, struct multipath *m) for (i = 0, p+=j+1; i <= hw_argc - 2; i++, p+=j+1) j = sprintf(p, "%s", as->argv[i]); } - consume(as, hw_argc - 1); + dm_consume_args(as, hw_argc - 1); return 0; fail: @@ -787,52 +730,45 @@ fail: return ret; } -static int parse_features(struct arg_set *as, struct multipath *m) +static int parse_features(struct dm_arg_set *as, struct multipath *m) { int r; unsigned argc; struct dm_target *ti = m->ti; - const char *param_name; + const char *arg_name; - static struct param _params[] = { + static struct dm_arg _args[] = { {0, 5, "invalid number of feature args"}, {1, 50, "pg_init_retries must be between 1 and 50"}, {0, 60000, "pg_init_delay_msecs must be between 0 and 60000"}, }; - r = read_param(_params, shift(as), &argc, &ti->error); + r = dm_read_arg_group(_args, as, &argc, &ti->error); if (r) return -EINVAL; if (!argc) return 0; - if (argc > as->argc) { - ti->error = "not enough arguments for features"; - return -EINVAL; - } - do { - param_name = shift(as); + arg_name = dm_shift_arg(as); argc--; - if (!strnicmp(param_name, MESG_STR("queue_if_no_path"))) { + if (!strcasecmp(arg_name, "queue_if_no_path")) { r = queue_if_no_path(m, 1, 0); continue; } - if (!strnicmp(param_name, MESG_STR("pg_init_retries")) && + if (!strcasecmp(arg_name, "pg_init_retries") && (argc >= 1)) { - r = read_param(_params + 1, shift(as), - &m->pg_init_retries, &ti->error); + r = dm_read_arg(_args + 1, as, &m->pg_init_retries, &ti->error); argc--; continue; } - if (!strnicmp(param_name, MESG_STR("pg_init_delay_msecs")) && + if (!strcasecmp(arg_name, "pg_init_delay_msecs") && (argc >= 1)) { - r = read_param(_params + 2, shift(as), - &m->pg_init_delay_msecs, &ti->error); + r = dm_read_arg(_args + 2, as, &m->pg_init_delay_msecs, &ti->error); argc--; continue; } @@ -847,15 +783,15 @@ static int parse_features(struct arg_set *as, struct multipath *m) static int multipath_ctr(struct dm_target *ti, unsigned int argc, char **argv) { - /* target parameters */ - static struct param _params[] = { + /* target arguments */ + static struct dm_arg _args[] = { {0, 1024, "invalid number of priority groups"}, {0, 1024, "invalid initial priority group number"}, }; int r; struct multipath *m; - struct arg_set as; + struct dm_arg_set as; unsigned pg_count = 0; unsigned next_pg_num; @@ -876,11 +812,11 @@ static int multipath_ctr(struct dm_target *ti, unsigned int argc, if (r) goto bad; - r = read_param(_params, shift(&as), &m->nr_priority_groups, &ti->error); + r = dm_read_arg(_args, &as, &m->nr_priority_groups, &ti->error); if (r) goto bad; - r = read_param(_params + 1, shift(&as), &next_pg_num, &ti->error); + r = dm_read_arg(_args + 1, &as, &next_pg_num, &ti->error); if (r) goto bad; @@ -1510,10 +1446,10 @@ static int multipath_message(struct dm_target *ti, unsigned argc, char **argv) } if (argc == 1) { - if (!strnicmp(argv[0], MESG_STR("queue_if_no_path"))) { + if (!strcasecmp(argv[0], "queue_if_no_path")) { r = queue_if_no_path(m, 1, 0); goto out; - } else if (!strnicmp(argv[0], MESG_STR("fail_if_no_path"))) { + } else if (!strcasecmp(argv[0], "fail_if_no_path")) { r = queue_if_no_path(m, 0, 0); goto out; } @@ -1524,18 +1460,18 @@ static int multipath_message(struct dm_target *ti, unsigned argc, char **argv) goto out; } - if (!strnicmp(argv[0], MESG_STR("disable_group"))) { + if (!strcasecmp(argv[0], "disable_group")) { r = bypass_pg_num(m, argv[1], 1); goto out; - } else if (!strnicmp(argv[0], MESG_STR("enable_group"))) { + } else if (!strcasecmp(argv[0], "enable_group")) { r = bypass_pg_num(m, argv[1], 0); goto out; - } else if (!strnicmp(argv[0], MESG_STR("switch_group"))) { + } else if (!strcasecmp(argv[0], "switch_group")) { r = switch_pg_num(m, argv[1]); goto out; - } else if (!strnicmp(argv[0], MESG_STR("reinstate_path"))) + } else if (!strcasecmp(argv[0], "reinstate_path")) action = reinstate_path; - else if (!strnicmp(argv[0], MESG_STR("fail_path"))) + else if (!strcasecmp(argv[0], "fail_path")) action = fail_path; else { DMWARN("Unrecognised multipath message received."); diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index 03516c7f5be1..259ce99302fc 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -797,6 +797,63 @@ int dm_table_add_target(struct dm_table *t, const char *type, return r; } +/* + * Target argument parsing helpers. + */ +static int validate_next_arg(struct dm_arg *arg, struct dm_arg_set *arg_set, + unsigned *value, char **error, unsigned grouped) +{ + const char *arg_str = dm_shift_arg(arg_set); + + if (!arg_str || + (sscanf(arg_str, "%u", value) != 1) || + (*value < arg->min) || + (*value > arg->max) || + (grouped && arg_set->argc < *value)) { + *error = arg->error; + return -EINVAL; + } + + return 0; +} + +int dm_read_arg(struct dm_arg *arg, struct dm_arg_set *arg_set, + unsigned *value, char **error) +{ + return validate_next_arg(arg, arg_set, value, error, 0); +} +EXPORT_SYMBOL(dm_read_arg); + +int dm_read_arg_group(struct dm_arg *arg, struct dm_arg_set *arg_set, + unsigned *value, char **error) +{ + return validate_next_arg(arg, arg_set, value, error, 1); +} +EXPORT_SYMBOL(dm_read_arg_group); + +const char *dm_shift_arg(struct dm_arg_set *as) +{ + char *r; + + if (as->argc) { + as->argc--; + r = *as->argv; + as->argv++; + return r; + } + + return NULL; +} +EXPORT_SYMBOL(dm_shift_arg); + +void dm_consume_args(struct dm_arg_set *as, unsigned num_args) +{ + BUG_ON(as->argc < num_args); + as->argc -= num_args; + as->argv += num_args; +} +EXPORT_SYMBOL(dm_consume_args); + static int dm_table_set_type(struct dm_table *t) { unsigned i; diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h index 4427e0454051..3fa1f3d90ce0 100644 --- a/include/linux/device-mapper.h +++ b/include/linux/device-mapper.h @@ -208,6 +208,49 @@ struct dm_target_callbacks { int dm_register_target(struct target_type *t); void dm_unregister_target(struct target_type *t); +/* + * Target argument parsing. + */ +struct dm_arg_set { + unsigned argc; + char **argv; +}; + +/* + * The minimum and maximum value of a numeric argument, together with + * the error message to use if the number is found to be outside that range. + */ +struct dm_arg { + unsigned min; + unsigned max; + char *error; +}; + +/* + * Validate the next argument, either returning it as *value or, if invalid, + * returning -EINVAL and setting *error. + */ +int dm_read_arg(struct dm_arg *arg, struct dm_arg_set *arg_set, + unsigned *value, char **error); + +/* + * Process the next argument as the start of a group containing between + * arg->min and arg->max further arguments. Either return the size as + * *num_args or, if invalid, return -EINVAL and set *error. + */ +int dm_read_arg_group(struct dm_arg *arg, struct dm_arg_set *arg_set, + unsigned *num_args, char **error); + +/* + * Return the current argument and shift to the next. + */ +const char *dm_shift_arg(struct dm_arg_set *as); + +/* + * Move through num_args arguments. + */ +void dm_consume_args(struct dm_arg_set *as, unsigned num_args); + /*----------------------------------------------------------------- * Functions for creating and manipulating mapped devices. * Drop the reference with dm_put when you finish with the object. From 30e4171bfe3d1c49689803338005cc0071dddaff Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Tue, 2 Aug 2011 12:32:05 +0100 Subject: [PATCH 184/439] dm flakey: use dm_target_offset and support discards Use dm_target_offset() and support discards. Signed-off-by: Mike Snitzer Signed-off-by: Alasdair G Kergon --- drivers/md/dm-flakey.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/md/dm-flakey.c b/drivers/md/dm-flakey.c index ea790623c30b..70a69b2f93d2 100644 --- a/drivers/md/dm-flakey.c +++ b/drivers/md/dm-flakey.c @@ -79,6 +79,7 @@ static int flakey_ctr(struct dm_target *ti, unsigned int argc, char **argv) } ti->num_flush_requests = 1; + ti->num_discard_requests = 1; ti->private = fc; return 0; @@ -99,7 +100,7 @@ static sector_t flakey_map_sector(struct dm_target *ti, sector_t bi_sector) { struct flakey_c *fc = ti->private; - return fc->start + (bi_sector - ti->begin); + return fc->start + dm_target_offset(ti, bi_sector); } static void flakey_map_bio(struct dm_target *ti, struct bio *bio) From dfd068b01f02653c6650f1c0eda443b2655d1471 Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Tue, 2 Aug 2011 12:32:05 +0100 Subject: [PATCH 185/439] dm flakey: support feature args Add the ability to specify arbitrary feature flags when creating a flakey target. This code uses the same target argument helpers that the multipath target does. Also remove the superfluous 'dm-flakey' prefixes from the error messages, as they already contain the prefix 'flakey'. Signed-off-by: Mike Snitzer Signed-off-by: Alasdair G Kergon --- drivers/md/dm-flakey.c | 87 ++++++++++++++++++++++++++++++++---------- 1 file changed, 66 insertions(+), 21 deletions(-) diff --git a/drivers/md/dm-flakey.c b/drivers/md/dm-flakey.c index 70a69b2f93d2..dd963dc3ca08 100644 --- a/drivers/md/dm-flakey.c +++ b/drivers/md/dm-flakey.c @@ -27,54 +27,99 @@ struct flakey_c { unsigned down_interval; }; +static int parse_features(struct dm_arg_set *as, struct dm_target *ti) +{ + int r; + unsigned argc; + const char *arg_name; + + static struct dm_arg _args[] = { + {0, 0, "Invalid number of feature args"}, + }; + + /* No feature arguments supplied. */ + if (!as->argc) + return 0; + + r = dm_read_arg_group(_args, as, &argc, &ti->error); + if (r) + return -EINVAL; + + while (argc && !r) { + arg_name = dm_shift_arg(as); + argc--; + + ti->error = "Unrecognised flakey feature requested"; + r = -EINVAL; + } + + return r; +} + /* - * Construct a flakey mapping: + * Construct a flakey mapping: + * [<#feature args> []*] */ static int flakey_ctr(struct dm_target *ti, unsigned int argc, char **argv) { - struct flakey_c *fc; - unsigned long long tmp; + static struct dm_arg _args[] = { + {0, UINT_MAX, "Invalid up interval"}, + {0, UINT_MAX, "Invalid down interval"}, + }; - if (argc != 4) { - ti->error = "dm-flakey: Invalid argument count"; + int r; + struct flakey_c *fc; + unsigned long long tmpll; + struct dm_arg_set as; + const char *devname; + + as.argc = argc; + as.argv = argv; + + if (argc < 4) { + ti->error = "Invalid argument count"; return -EINVAL; } fc = kmalloc(sizeof(*fc), GFP_KERNEL); if (!fc) { - ti->error = "dm-flakey: Cannot allocate linear context"; + ti->error = "Cannot allocate linear context"; return -ENOMEM; } fc->start_time = jiffies; - if (sscanf(argv[1], "%llu", &tmp) != 1) { - ti->error = "dm-flakey: Invalid device sector"; - goto bad; - } - fc->start = tmp; + devname = dm_shift_arg(&as); - if (sscanf(argv[2], "%u", &fc->up_interval) != 1) { - ti->error = "dm-flakey: Invalid up interval"; + if (sscanf(dm_shift_arg(&as), "%llu", &tmpll) != 1) { + ti->error = "Invalid device sector"; goto bad; } + fc->start = tmpll; - if (sscanf(argv[3], "%u", &fc->down_interval) != 1) { - ti->error = "dm-flakey: Invalid down interval"; + r = dm_read_arg(_args, &as, &fc->up_interval, &ti->error); + if (r) + goto bad; + + r = dm_read_arg(_args, &as, &fc->down_interval, &ti->error); + if (r) goto bad; - } if (!(fc->up_interval + fc->down_interval)) { - ti->error = "dm-flakey: Total (up + down) interval is zero"; + ti->error = "Total (up + down) interval is zero"; goto bad; } if (fc->up_interval + fc->down_interval < fc->up_interval) { - ti->error = "dm-flakey: Interval overflow"; + ti->error = "Interval overflow"; goto bad; } - if (dm_get_device(ti, argv[0], dm_table_get_mode(ti->table), &fc->dev)) { - ti->error = "dm-flakey: Device lookup failed"; + r = parse_features(&as, ti); + if (r) + goto bad; + + if (dm_get_device(ti, devname, dm_table_get_mode(ti->table), &fc->dev)) { + ti->error = "Device lookup failed"; goto bad; } @@ -178,7 +223,7 @@ static int flakey_iterate_devices(struct dm_target *ti, iterate_devices_callout_ static struct target_type flakey_target = { .name = "flakey", - .version = {1, 1, 0}, + .version = {1, 2, 0}, .module = THIS_MODULE, .ctr = flakey_ctr, .dtr = flakey_dtr, From b26f5e3d7127487e934758c1fbe05d683b082cb0 Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Tue, 2 Aug 2011 12:32:05 +0100 Subject: [PATCH 186/439] dm flakey: add drop_writes Add 'drop_writes' option to drop writes silently while the device is 'down'. Reads are not touched. Signed-off-by: Mike Snitzer Signed-off-by: Alasdair G Kergon --- Documentation/device-mapper/dm-flakey.txt | 29 +++++++--- drivers/md/dm-flakey.c | 65 +++++++++++++++++++---- 2 files changed, 79 insertions(+), 15 deletions(-) diff --git a/Documentation/device-mapper/dm-flakey.txt b/Documentation/device-mapper/dm-flakey.txt index c8efdfd19a65..1b66c868ee7e 100644 --- a/Documentation/device-mapper/dm-flakey.txt +++ b/Documentation/device-mapper/dm-flakey.txt @@ -1,17 +1,34 @@ dm-flakey ========= -This target is the same as the linear target except that it returns I/O -errors periodically. It's been found useful in simulating failing -devices for testing purposes. +This target is the same as the linear target except that it exhibits +unreliable behaviour periodically. It's been found useful in simulating +failing devices for testing purposes. Starting from the time the table is loaded, the device is available for - seconds, then returns errors for seconds, -and then this cycle repeats. + seconds, then exhibits unreliable behaviour for seconds, and then this cycle repeats. -Parameters: +Also, consider using this in combination with the dm-delay target too, +which can delay reads and writes and/or send them to different +underlying devices. + +Table parameters +---------------- + \ + [ []] + +Mandatory parameters: : Full pathname to the underlying block-device, or a "major:minor" device-number. : Starting sector within the device. : Number of seconds device is available. : Number of seconds device returns errors. + +Optional feature parameters: + If no feature parameters are present, during the periods of + unreliability, all I/O returns errors. + + drop_writes: + All write I/O is silently ignored. + Read I/O is handled correctly. diff --git a/drivers/md/dm-flakey.c b/drivers/md/dm-flakey.c index dd963dc3ca08..e7c4c2a64f4b 100644 --- a/drivers/md/dm-flakey.c +++ b/drivers/md/dm-flakey.c @@ -25,16 +25,22 @@ struct flakey_c { sector_t start; unsigned up_interval; unsigned down_interval; + unsigned long flags; }; -static int parse_features(struct dm_arg_set *as, struct dm_target *ti) +enum feature_flag_bits { + DROP_WRITES +}; + +static int parse_features(struct dm_arg_set *as, struct flakey_c *fc, + struct dm_target *ti) { int r; unsigned argc; const char *arg_name; static struct dm_arg _args[] = { - {0, 0, "Invalid number of feature args"}, + {0, 1, "Invalid number of feature args"}, }; /* No feature arguments supplied. */ @@ -49,6 +55,18 @@ static int parse_features(struct dm_arg_set *as, struct dm_target *ti) arg_name = dm_shift_arg(as); argc--; + /* + * drop_writes + */ + if (!strcasecmp(arg_name, "drop_writes")) { + if (test_and_set_bit(DROP_WRITES, &fc->flags)) { + ti->error = "Feature drop_writes duplicated"; + return -EINVAL; + } + + continue; + } + ti->error = "Unrecognised flakey feature requested"; r = -EINVAL; } @@ -59,6 +77,9 @@ static int parse_features(struct dm_arg_set *as, struct dm_target *ti) /* * Construct a flakey mapping: * [<#feature args> []*] + * + * Feature args: + * [drop_writes] */ static int flakey_ctr(struct dm_target *ti, unsigned int argc, char **argv) { @@ -81,7 +102,7 @@ static int flakey_ctr(struct dm_target *ti, unsigned int argc, char **argv) return -EINVAL; } - fc = kmalloc(sizeof(*fc), GFP_KERNEL); + fc = kzalloc(sizeof(*fc), GFP_KERNEL); if (!fc) { ti->error = "Cannot allocate linear context"; return -ENOMEM; @@ -114,7 +135,7 @@ static int flakey_ctr(struct dm_target *ti, unsigned int argc, char **argv) goto bad; } - r = parse_features(&as, ti); + r = parse_features(&as, fc, ti); if (r) goto bad; @@ -162,12 +183,31 @@ static int flakey_map(struct dm_target *ti, struct bio *bio, { struct flakey_c *fc = ti->private; unsigned elapsed; + unsigned rw; /* Are we alive ? */ elapsed = (jiffies - fc->start_time) / HZ; - if (elapsed % (fc->up_interval + fc->down_interval) >= fc->up_interval) - return -EIO; + if (elapsed % (fc->up_interval + fc->down_interval) >= fc->up_interval) { + rw = bio_data_dir(bio); + /* + * Drop writes. Map reads as normal. + */ + if (test_bit(DROP_WRITES, &fc->flags)) { + if (rw == WRITE) { + bio_endio(bio, 0); + return DM_MAPIO_SUBMITTED; + } + goto map_bio; + } + + /* + * Default setting errors all I/O. + */ + return -EIO; + } + +map_bio: flakey_map_bio(ti, bio); return DM_MAPIO_REMAPPED; @@ -176,7 +216,9 @@ static int flakey_map(struct dm_target *ti, struct bio *bio, static int flakey_status(struct dm_target *ti, status_type_t type, char *result, unsigned int maxlen) { + unsigned sz = 0; struct flakey_c *fc = ti->private; + unsigned drop_writes; switch (type) { case STATUSTYPE_INFO: @@ -184,9 +226,14 @@ static int flakey_status(struct dm_target *ti, status_type_t type, break; case STATUSTYPE_TABLE: - snprintf(result, maxlen, "%s %llu %u %u", fc->dev->name, - (unsigned long long)fc->start, fc->up_interval, - fc->down_interval); + DMEMIT("%s %llu %u %u ", fc->dev->name, + (unsigned long long)fc->start, fc->up_interval, + fc->down_interval); + + drop_writes = test_bit(DROP_WRITES, &fc->flags); + DMEMIT("%u ", drop_writes); + if (drop_writes) + DMEMIT("drop_writes "); break; } return 0; From a3998799fb4df0b0af8271a7d50c4269032397aa Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Tue, 2 Aug 2011 12:32:06 +0100 Subject: [PATCH 187/439] dm flakey: add corrupt_bio_byte feature Add corrupt_bio_byte feature to simulate corruption by overwriting a byte at a specified position with a specified value during intervals when the device is "down". Signed-off-by: Mike Snitzer Signed-off-by: Alasdair G Kergon --- Documentation/device-mapper/dm-flakey.txt | 19 +++ drivers/md/dm-flakey.c | 155 +++++++++++++++++++--- 2 files changed, 159 insertions(+), 15 deletions(-) diff --git a/Documentation/device-mapper/dm-flakey.txt b/Documentation/device-mapper/dm-flakey.txt index 1b66c868ee7e..6ff5c2327227 100644 --- a/Documentation/device-mapper/dm-flakey.txt +++ b/Documentation/device-mapper/dm-flakey.txt @@ -32,3 +32,22 @@ Optional feature parameters: drop_writes: All write I/O is silently ignored. Read I/O is handled correctly. + + corrupt_bio_byte : + During , replace of the data of + each matching bio with . + + : The offset of the byte to replace. + Counting starts at 1, to replace the first byte. + : Either 'r' to corrupt reads or 'w' to corrupt writes. + 'w' is incompatible with drop_writes. + : The value (from 0-255) to write. + : Perform the replacement only if bio->bi_rw has all the + selected flags set. + +Examples: + corrupt_bio_byte 32 r 1 0 + - replaces the 32nd byte of READ bios with the value 1 + + corrupt_bio_byte 224 w 0 32 + - replaces the 224th byte of REQ_META (=32) bios with the value 0 diff --git a/drivers/md/dm-flakey.c b/drivers/md/dm-flakey.c index e7c4c2a64f4b..89f73ca22cfa 100644 --- a/drivers/md/dm-flakey.c +++ b/drivers/md/dm-flakey.c @@ -1,6 +1,6 @@ /* * Copyright (C) 2003 Sistina Software (UK) Limited. - * Copyright (C) 2004, 2010 Red Hat, Inc. All rights reserved. + * Copyright (C) 2004, 2010-2011 Red Hat, Inc. All rights reserved. * * This file is released under the GPL. */ @@ -15,6 +15,9 @@ #define DM_MSG_PREFIX "flakey" +#define all_corrupt_bio_flags_match(bio, fc) \ + (((bio)->bi_rw & (fc)->corrupt_bio_flags) == (fc)->corrupt_bio_flags) + /* * Flakey: Used for testing only, simulates intermittent, * catastrophic device failure. @@ -26,6 +29,10 @@ struct flakey_c { unsigned up_interval; unsigned down_interval; unsigned long flags; + unsigned corrupt_bio_byte; + unsigned corrupt_bio_rw; + unsigned corrupt_bio_value; + unsigned corrupt_bio_flags; }; enum feature_flag_bits { @@ -40,7 +47,10 @@ static int parse_features(struct dm_arg_set *as, struct flakey_c *fc, const char *arg_name; static struct dm_arg _args[] = { - {0, 1, "Invalid number of feature args"}, + {0, 6, "Invalid number of feature args"}, + {1, UINT_MAX, "Invalid corrupt bio byte"}, + {0, 255, "Invalid corrupt value to write into bio byte (0-255)"}, + {0, UINT_MAX, "Invalid corrupt bio flags mask"}, }; /* No feature arguments supplied. */ @@ -49,9 +59,9 @@ static int parse_features(struct dm_arg_set *as, struct flakey_c *fc, r = dm_read_arg_group(_args, as, &argc, &ti->error); if (r) - return -EINVAL; + return r; - while (argc && !r) { + while (argc) { arg_name = dm_shift_arg(as); argc--; @@ -67,11 +77,61 @@ static int parse_features(struct dm_arg_set *as, struct flakey_c *fc, continue; } + /* + * corrupt_bio_byte + */ + if (!strcasecmp(arg_name, "corrupt_bio_byte")) { + if (!argc) + ti->error = "Feature corrupt_bio_byte requires parameters"; + + r = dm_read_arg(_args + 1, as, &fc->corrupt_bio_byte, &ti->error); + if (r) + return r; + argc--; + + /* + * Direction r or w? + */ + arg_name = dm_shift_arg(as); + if (!strcasecmp(arg_name, "w")) + fc->corrupt_bio_rw = WRITE; + else if (!strcasecmp(arg_name, "r")) + fc->corrupt_bio_rw = READ; + else { + ti->error = "Invalid corrupt bio direction (r or w)"; + return -EINVAL; + } + argc--; + + /* + * Value of byte (0-255) to write in place of correct one. + */ + r = dm_read_arg(_args + 2, as, &fc->corrupt_bio_value, &ti->error); + if (r) + return r; + argc--; + + /* + * Only corrupt bios with these flags set. + */ + r = dm_read_arg(_args + 3, as, &fc->corrupt_bio_flags, &ti->error); + if (r) + return r; + argc--; + + continue; + } + ti->error = "Unrecognised flakey feature requested"; - r = -EINVAL; + return -EINVAL; } - return r; + if (test_bit(DROP_WRITES, &fc->flags) && (fc->corrupt_bio_rw == WRITE)) { + ti->error = "drop_writes is incompatible with corrupt_bio_byte with the WRITE flag set"; + return -EINVAL; + } + + return 0; } /* @@ -80,6 +140,11 @@ static int parse_features(struct dm_arg_set *as, struct flakey_c *fc, * * Feature args: * [drop_writes] + * [corrupt_bio_byte ] + * + * Nth_byte starts from 1 for the first byte. + * Direction is r for READ or w for WRITE. + * bio_flags is ignored if 0. */ static int flakey_ctr(struct dm_target *ti, unsigned int argc, char **argv) { @@ -178,31 +243,64 @@ static void flakey_map_bio(struct dm_target *ti, struct bio *bio) bio->bi_sector = flakey_map_sector(ti, bio->bi_sector); } +static void corrupt_bio_data(struct bio *bio, struct flakey_c *fc) +{ + unsigned bio_bytes = bio_cur_bytes(bio); + char *data = bio_data(bio); + + /* + * Overwrite the Nth byte of the data returned. + */ + if (data && bio_bytes >= fc->corrupt_bio_byte) { + data[fc->corrupt_bio_byte - 1] = fc->corrupt_bio_value; + + DMDEBUG("Corrupting data bio=%p by writing %u to byte %u " + "(rw=%c bi_rw=%lu bi_sector=%llu cur_bytes=%u)\n", + bio, fc->corrupt_bio_value, fc->corrupt_bio_byte, + (bio_data_dir(bio) == WRITE) ? 'w' : 'r', + bio->bi_rw, (unsigned long long)bio->bi_sector, bio_bytes); + } +} + static int flakey_map(struct dm_target *ti, struct bio *bio, union map_info *map_context) { struct flakey_c *fc = ti->private; unsigned elapsed; - unsigned rw; /* Are we alive ? */ elapsed = (jiffies - fc->start_time) / HZ; if (elapsed % (fc->up_interval + fc->down_interval) >= fc->up_interval) { - rw = bio_data_dir(bio); + /* + * Flag this bio as submitted while down. + */ + map_context->ll = 1; /* - * Drop writes. Map reads as normal. + * Map reads as normal. + */ + if (bio_data_dir(bio) == READ) + goto map_bio; + + /* + * Drop writes? */ if (test_bit(DROP_WRITES, &fc->flags)) { - if (rw == WRITE) { - bio_endio(bio, 0); - return DM_MAPIO_SUBMITTED; - } + bio_endio(bio, 0); + return DM_MAPIO_SUBMITTED; + } + + /* + * Corrupt matching writes. + */ + if (fc->corrupt_bio_byte && (fc->corrupt_bio_rw == WRITE)) { + if (all_corrupt_bio_flags_match(bio, fc)) + corrupt_bio_data(bio, fc); goto map_bio; } /* - * Default setting errors all I/O. + * By default, error all I/O. */ return -EIO; } @@ -213,6 +311,24 @@ map_bio: return DM_MAPIO_REMAPPED; } +static int flakey_end_io(struct dm_target *ti, struct bio *bio, + int error, union map_info *map_context) +{ + struct flakey_c *fc = ti->private; + unsigned bio_submitted_while_down = map_context->ll; + + /* + * Corrupt successful READs while in down state. + * If flags were specified, only corrupt those that match. + */ + if (!error && bio_submitted_while_down && + (bio_data_dir(bio) == READ) && (fc->corrupt_bio_rw == READ) && + all_corrupt_bio_flags_match(bio, fc)) + corrupt_bio_data(bio, fc); + + return error; +} + static int flakey_status(struct dm_target *ti, status_type_t type, char *result, unsigned int maxlen) { @@ -231,9 +347,17 @@ static int flakey_status(struct dm_target *ti, status_type_t type, fc->down_interval); drop_writes = test_bit(DROP_WRITES, &fc->flags); - DMEMIT("%u ", drop_writes); + DMEMIT("%u ", drop_writes + (fc->corrupt_bio_byte > 0) * 5); + if (drop_writes) DMEMIT("drop_writes "); + + if (fc->corrupt_bio_byte) + DMEMIT("corrupt_bio_byte %u %c %u %u ", + fc->corrupt_bio_byte, + (fc->corrupt_bio_rw == WRITE) ? 'w' : 'r', + fc->corrupt_bio_value, fc->corrupt_bio_flags); + break; } return 0; @@ -275,6 +399,7 @@ static struct target_type flakey_target = { .ctr = flakey_ctr, .dtr = flakey_dtr, .map = flakey_map, + .end_io = flakey_end_io, .status = flakey_status, .ioctl = flakey_ioctl, .merge = flakey_merge, From 0ddf9644cc26e74ed671525e61a17bdbebf18da6 Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Tue, 2 Aug 2011 12:32:06 +0100 Subject: [PATCH 188/439] dm ioctl: fill in device parameters in more ioctls Move parameter filling from find_device to __find_device_hash_cell. This patch causes ioctls using __find_device_hash_cell (DM_DEV_REMOVE_CMD, DM_DEV_SUSPEND_CMD - resume, DM_TABLE_CLEAR_CMD) to return device parameters, bringing them into line with the other ioctls. Signed-off-by: Mikulas Patocka Signed-off-by: Alasdair G Kergon --- drivers/md/dm-ioctl.c | 63 +++++++++++++++++++++++----------------- include/linux/dm-ioctl.h | 4 +-- 2 files changed, 38 insertions(+), 29 deletions(-) diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index 1622a6bc0bfd..99d38a6925a4 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -719,24 +719,49 @@ static int dev_create(struct dm_ioctl *param, size_t param_size) static struct hash_cell *__find_device_hash_cell(struct dm_ioctl *param) { struct mapped_device *md; - void *mdptr = NULL; + struct hash_cell *hc = NULL; - if (*param->uuid) - return __get_uuid_cell(param->uuid); + if (*param->uuid) { + hc = __get_uuid_cell(param->uuid); + if (!hc) + return NULL; + goto fill_params; + } - if (*param->name) - return __get_name_cell(param->name); + if (*param->name) { + hc = __get_name_cell(param->name); + if (!hc) + return NULL; + goto fill_params; + } md = dm_get_md(huge_decode_dev(param->dev)); if (!md) - goto out; + return NULL; - mdptr = dm_get_mdptr(md); - if (!mdptr) + hc = dm_get_mdptr(md); + if (!hc) { dm_put(md); + return NULL; + } -out: - return mdptr; +fill_params: + /* + * Sneakily write in both the name and the uuid + * while we have the cell. + */ + strlcpy(param->name, hc->name, sizeof(param->name)); + if (hc->uuid) + strlcpy(param->uuid, hc->uuid, sizeof(param->uuid)); + else + param->uuid[0] = '\0'; + + if (hc->new_map) + param->flags |= DM_INACTIVE_PRESENT_FLAG; + else + param->flags &= ~DM_INACTIVE_PRESENT_FLAG; + + return hc; } static struct mapped_device *find_device(struct dm_ioctl *param) @@ -746,24 +771,8 @@ static struct mapped_device *find_device(struct dm_ioctl *param) down_read(&_hash_lock); hc = __find_device_hash_cell(param); - if (hc) { + if (hc) md = hc->md; - - /* - * Sneakily write in both the name and the uuid - * while we have the cell. - */ - strlcpy(param->name, hc->name, sizeof(param->name)); - if (hc->uuid) - strlcpy(param->uuid, hc->uuid, sizeof(param->uuid)); - else - param->uuid[0] = '\0'; - - if (hc->new_map) - param->flags |= DM_INACTIVE_PRESENT_FLAG; - else - param->flags &= ~DM_INACTIVE_PRESENT_FLAG; - } up_read(&_hash_lock); return md; diff --git a/include/linux/dm-ioctl.h b/include/linux/dm-ioctl.h index 3708455ee6c3..0cb8eff76bd6 100644 --- a/include/linux/dm-ioctl.h +++ b/include/linux/dm-ioctl.h @@ -267,9 +267,9 @@ enum { #define DM_DEV_SET_GEOMETRY _IOWR(DM_IOCTL, DM_DEV_SET_GEOMETRY_CMD, struct dm_ioctl) #define DM_VERSION_MAJOR 4 -#define DM_VERSION_MINOR 20 +#define DM_VERSION_MINOR 21 #define DM_VERSION_PATCHLEVEL 0 -#define DM_VERSION_EXTRA "-ioctl (2011-02-02)" +#define DM_VERSION_EXTRA "-ioctl (2011-07-06)" /* Status bits */ #define DM_READONLY_FLAG (1 << 0) /* In/Out */ From ba2e19b0f4ccd6920fe175a86521ff18ede260cb Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Tue, 2 Aug 2011 12:32:06 +0100 Subject: [PATCH 189/439] dm ioctl: introduce __get_dev_cell Move logic to find device based on major/minor number to a separate function __get_dev_cell (similar to __get_uuid_cell and __get_name_cell). This makes the function __find_device_hash_cell more straightforward. Signed-off-by: Mikulas Patocka Signed-off-by: Alasdair G Kergon --- drivers/md/dm-ioctl.c | 41 ++++++++++++++++++++++++----------------- 1 file changed, 24 insertions(+), 17 deletions(-) diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index 99d38a6925a4..4c350914f4a0 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -128,6 +128,24 @@ static struct hash_cell *__get_uuid_cell(const char *str) return NULL; } +static struct hash_cell *__get_dev_cell(uint64_t dev) +{ + struct mapped_device *md; + struct hash_cell *hc; + + md = dm_get_md(huge_decode_dev(dev)); + if (!md) + return NULL; + + hc = dm_get_mdptr(md); + if (!hc) { + dm_put(md); + return NULL; + } + + return hc; +} + /*----------------------------------------------------------------- * Inserting, removing and renaming a device. *---------------------------------------------------------------*/ @@ -718,34 +736,23 @@ static int dev_create(struct dm_ioctl *param, size_t param_size) */ static struct hash_cell *__find_device_hash_cell(struct dm_ioctl *param) { - struct mapped_device *md; struct hash_cell *hc = NULL; if (*param->uuid) { hc = __get_uuid_cell(param->uuid); if (!hc) return NULL; - goto fill_params; - } - - if (*param->name) { + } else if (*param->name) { hc = __get_name_cell(param->name); if (!hc) return NULL; - goto fill_params; - } - - md = dm_get_md(huge_decode_dev(param->dev)); - if (!md) + } else if (param->dev) { + hc = __get_dev_cell(param->dev); + if (!hc) + return NULL; + } else return NULL; - hc = dm_get_mdptr(md); - if (!hc) { - dm_put(md); - return NULL; - } - -fill_params: /* * Sneakily write in both the name and the uuid * while we have the cell. From 759dea204cce9f1fc2a5d00ea25211299fc7a4a0 Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Tue, 2 Aug 2011 12:32:06 +0100 Subject: [PATCH 190/439] dm ioctl: forbid multiple device specifiers Exactly one of name, uuid or device must be specified when referencing an existing device. This removes the ambiguity (risking the wrong device being updated) if two conflicting parameters were specified. Previously one parameter got used and any others were ignored silently. Signed-off-by: Mikulas Patocka Signed-off-by: Alasdair G Kergon --- drivers/md/dm-ioctl.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index 4c350914f4a0..2e9a3ca37bdd 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -739,10 +739,16 @@ static struct hash_cell *__find_device_hash_cell(struct dm_ioctl *param) struct hash_cell *hc = NULL; if (*param->uuid) { + if (*param->name || param->dev) + return NULL; + hc = __get_uuid_cell(param->uuid); if (!hc) return NULL; } else if (*param->name) { + if (param->dev) + return NULL; + hc = __get_name_cell(param->name); if (!hc) return NULL; From c0a2fa1ef1057a1e9450d6f055f1cde2ad4f85a2 Mon Sep 17 00:00:00 2001 From: Jonathan Brassow Date: Tue, 2 Aug 2011 12:32:06 +0100 Subject: [PATCH 191/439] dm raid: improve table parameters documentation Add more information about some dm-raid table parameters and clarify how parameters are printed when 'dmsetup table' is issued. Signed-off-by: Jonathan Brassow Signed-off-by: Alasdair G Kergon --- Documentation/device-mapper/dm-raid.txt | 114 +++++++++++++++--------- 1 file changed, 73 insertions(+), 41 deletions(-) diff --git a/Documentation/device-mapper/dm-raid.txt b/Documentation/device-mapper/dm-raid.txt index 33b6b7071ac8..4f9dd3cecc11 100644 --- a/Documentation/device-mapper/dm-raid.txt +++ b/Documentation/device-mapper/dm-raid.txt @@ -1,49 +1,75 @@ -Device-mapper RAID (dm-raid) is a bridge from DM to MD. It -provides a way to use device-mapper interfaces to access the MD RAID -drivers. +dm-raid +------- -As with all device-mapper targets, the nominal public interfaces are the -constructor (CTR) tables and the status outputs (both STATUSTYPE_INFO -and STATUSTYPE_TABLE). The CTR table looks like the following: +The device-mapper RAID (dm-raid) target provides a bridge from DM to MD. +It allows the MD RAID drivers to be accessed using a device-mapper +interface. -1: raid \ -2: <#raid_params> \ -3: <#raid_devs> .. +The target is named "raid" and it accepts the following parameters: -Line 1 contains the standard first three arguments to any device-mapper -target - the start, length, and target type fields. The target type in -this case is "raid". + <#raid_params> \ + <#raid_devs> [.. ] -Line 2 contains the arguments that define the particular raid -type/personality/level, the required arguments for that raid type, and -any optional arguments. Possible raid types include: raid4, raid5_la, -raid5_ls, raid5_rs, raid6_zr, raid6_nr, and raid6_nc. (raid1 is -planned for the future.) The list of required and optional parameters -is the same for all the current raid types. The required parameters are -positional, while the optional parameters are given as key/value pairs. -The possible parameters are as follows: - Chunk size in sectors. - [[no]sync] Force/Prevent RAID initialization - [rebuild ] Rebuild the drive indicated by the index - [daemon_sleep ] Time between bitmap daemon work to clear bits - [min_recovery_rate ] Throttle RAID initialization - [max_recovery_rate ] Throttle RAID initialization - [max_write_behind ] See '-write-behind=' (man mdadm) - [stripe_cache ] Stripe cache size for higher RAIDs +: + raid4 RAID4 dedicated parity disk + raid5_la RAID5 left asymmetric + - rotating parity 0 with data continuation + raid5_ra RAID5 right asymmetric + - rotating parity N with data continuation + raid5_ls RAID5 left symmetric + - rotating parity 0 with data restart + raid5_rs RAID5 right symmetric + - rotating parity N with data restart + raid6_zr RAID6 zero restart + - rotating parity zero (left-to-right) with data restart + raid6_nr RAID6 N restart + - rotating parity N (right-to-left) with data restart + raid6_nc RAID6 N continue + - rotating parity N (right-to-left) with data continuation -Line 3 contains the list of devices that compose the array in -metadata/data device pairs. If the metadata is stored separately, a '-' -is given for the metadata device position. If a drive has failed or is -missing at creation time, a '-' can be given for both the metadata and -data drives for a given position. + Refererence: Chapter 4 of + http://www.snia.org/sites/default/files/SNIA_DDF_Technical_Position_v2.0.pdf -NB. Currently all metadata devices must be specified as '-'. +<#raid_params>: The number of parameters that follow. -Examples: + consists of + Mandatory parameters: + : Chunk size in sectors. This parameter is often known as + "stripe size". It is the only mandatory parameter and + is placed first. + + followed by optional parameters (in any order): + [sync|nosync] Force or prevent RAID initialization. + + [rebuild ] Rebuild drive number idx (first drive is 0). + + [daemon_sleep ] + Interval between runs of the bitmap daemon that + clear bits. A longer interval means less bitmap I/O but + resyncing after a failure is likely to take longer. + + [min_recovery_rate ] Throttle RAID initialization + [max_recovery_rate ] Throttle RAID initialization + [max_write_behind ] See '-write-behind=' (man mdadm) + [stripe_cache ] Stripe cache size (higher RAIDs only) + +<#raid_devs>: The number of devices composing the array. + Each device consists of two entries. The first is the device + containing the metadata (if any); the second is the one containing the + data. Currently, separate metadata devices are not supported and '-' + is required in place of the metadata device. + + If a drive has failed or is missing at creation time, a '-' can be + given for both the metadata and data drives for a given position. + + +Example tables +-------------- # RAID4 - 4 data drives, 1 parity # No metadata devices specified to hold superblock/bitmap info # Chunk size of 1MiB # (Lines separated for easy reading) + 0 1960893648 raid \ raid4 1 2048 \ 5 - 8:17 - 8:33 - 8:49 - 8:65 - 8:81 @@ -51,20 +77,26 @@ Examples: # RAID4 - 4 data drives, 1 parity (no metadata devices) # Chunk size of 1MiB, force RAID initialization, # min recovery rate at 20 kiB/sec/disk + 0 1960893648 raid \ raid4 4 2048 min_recovery_rate 20 sync\ 5 - 8:17 - 8:33 - 8:49 - 8:65 - 8:81 -Performing a 'dmsetup table' should display the CTR table used to -construct the mapping (with possible reordering of optional -parameters). +'dmsetup table' displays the table used to construct the mapping. +The optional parameters will always be printed in the order listed +above with "sync" or "nosync" always output ahead of the other +arguments, regardless of the order used when originally loading the table. -Performing a 'dmsetup status' will yield information on the state and -health of the array. The output is as follows: +'dmsetup status' yields information on the state and health of the +array. +The output is as follows: 1: raid \ 2: <#devices> <1 health char for each dev> -Line 1 is standard DM output. Line 2 is best shown by example: +Line 1 is the standard output produced by device-mapper. +Line 2 is produced by the raid target, and best explained by example: 0 1960893648 raid raid4 5 AAAAA 2/490221568 Here we can see the RAID type is raid4, there are 5 devices - all of which are 'A'live, and the array is 2/490221568 complete with recovery. +Faulty or missing devices are marked 'D'. Devices that are out-of-sync +are marked 'a'. From c1084561bb85da3630540ebe951749a8cd8fc714 Mon Sep 17 00:00:00 2001 From: Jonathan Brassow Date: Tue, 2 Aug 2011 12:32:07 +0100 Subject: [PATCH 192/439] dm raid: add region_size parameter Allow the user to specify the region_size. Ensures that the supplied value meets md's constraints, viz. the number of regions does not exceed 2^21. Signed-off-by: Jonathan Brassow Signed-off-by: Alasdair G Kergon --- Documentation/device-mapper/dm-raid.txt | 4 ++ drivers/md/dm-raid.c | 82 ++++++++++++++++++++++++- 2 files changed, 83 insertions(+), 3 deletions(-) diff --git a/Documentation/device-mapper/dm-raid.txt b/Documentation/device-mapper/dm-raid.txt index 4f9dd3cecc11..be4419a30781 100644 --- a/Documentation/device-mapper/dm-raid.txt +++ b/Documentation/device-mapper/dm-raid.txt @@ -52,6 +52,10 @@ The target is named "raid" and it accepts the following parameters: [max_recovery_rate ] Throttle RAID initialization [max_write_behind ] See '-write-behind=' (man mdadm) [stripe_cache ] Stripe cache size (higher RAIDs only) + [region_size ] + The region_size multiplied by the number of regions is the + logical size of the array. The bitmap records the device + synchronisation state for each region. <#raid_devs>: The number of devices composing the array. Each device consists of two entries. The first is the device diff --git a/drivers/md/dm-raid.c b/drivers/md/dm-raid.c index d4e95b2e39f6..a8a1915a450d 100644 --- a/drivers/md/dm-raid.c +++ b/drivers/md/dm-raid.c @@ -52,7 +52,7 @@ struct raid_dev { #define DMPF_MAX_RECOVERY_RATE 0x20 #define DMPF_MAX_WRITE_BEHIND 0x40 #define DMPF_STRIPE_CACHE 0x80 - +#define DMPF_REGION_SIZE 0X100 struct raid_set { struct dm_target *ti; @@ -236,6 +236,67 @@ static int dev_parms(struct raid_set *rs, char **argv) return 0; } +/* + * validate_region_size + * @rs + * @region_size: region size in sectors. If 0, pick a size (4MiB default). + * + * Set rs->md.bitmap_info.chunksize (which really refers to 'region size'). + * Ensure that (ti->len/region_size < 2^21) - required by MD bitmap. + * + * Returns: 0 on success, -EINVAL on failure. + */ +static int validate_region_size(struct raid_set *rs, unsigned long region_size) +{ + unsigned long min_region_size = rs->ti->len / (1 << 21); + + if (!region_size) { + /* + * Choose a reasonable default. All figures in sectors. + */ + if (min_region_size > (1 << 13)) { + DMINFO("Choosing default region size of %lu sectors", + region_size); + region_size = min_region_size; + } else { + DMINFO("Choosing default region size of 4MiB"); + region_size = 1 << 13; /* sectors */ + } + } else { + /* + * Validate user-supplied value. + */ + if (region_size > rs->ti->len) { + rs->ti->error = "Supplied region size is too large"; + return -EINVAL; + } + + if (region_size < min_region_size) { + DMERR("Supplied region_size (%lu sectors) below minimum (%lu)", + region_size, min_region_size); + rs->ti->error = "Supplied region size is too small"; + return -EINVAL; + } + + if (!is_power_of_2(region_size)) { + rs->ti->error = "Region size is not a power of 2"; + return -EINVAL; + } + + if (region_size < rs->md.chunk_sectors) { + rs->ti->error = "Region size is smaller than the chunk size"; + return -EINVAL; + } + } + + /* + * Convert sectors to bytes. + */ + rs->md.bitmap_info.chunksize = (region_size << 9); + + return 0; +} + /* * Possible arguments are... * RAID456: @@ -249,12 +310,13 @@ static int dev_parms(struct raid_set *rs, char **argv) * [max_recovery_rate ] Throttle RAID initialization * [max_write_behind ] See '-write-behind=' (man mdadm) * [stripe_cache ] Stripe cache size for higher RAIDs + * [region_size ] Defines granularity of bitmap */ static int parse_raid_params(struct raid_set *rs, char **argv, unsigned num_raid_params) { unsigned i, rebuild_cnt = 0; - unsigned long value; + unsigned long value, region_size = 0; char *key; /* @@ -365,6 +427,9 @@ static int parse_raid_params(struct raid_set *rs, char **argv, return -EINVAL; } rs->md.sync_speed_max = (int)value; + } else if (!strcasecmp(key, "region_size")) { + rs->print_flags |= DMPF_REGION_SIZE; + region_size = value; } else { DMERR("Unable to parse RAID parameter: %s", key); rs->ti->error = "Unable to parse RAID parameters"; @@ -372,6 +437,14 @@ static int parse_raid_params(struct raid_set *rs, char **argv, } } + if (validate_region_size(rs, region_size)) + return -EINVAL; + + if (rs->md.chunk_sectors) + rs->ti->split_io = rs->md.chunk_sectors; + else + rs->ti->split_io = region_size; + /* Assume there are no metadata devices until the drives are parsed */ rs->md.persistent = 0; rs->md.external = 1; @@ -469,7 +542,6 @@ static int raid_ctr(struct dm_target *ti, unsigned argc, char **argv) goto bad; INIT_WORK(&rs->md.event_work, do_table_event); - ti->split_io = rs->md.chunk_sectors; ti->private = rs; mutex_lock(&rs->md.reconfig_mutex); @@ -596,6 +668,10 @@ static int raid_status(struct dm_target *ti, status_type_t type, conf ? conf->max_nr_stripes * 2 : 0); } + if (rs->print_flags & DMPF_REGION_SIZE) + DMEMIT(" region_size %lu", + rs->md.bitmap_info.chunksize >> 9); + DMEMIT(" %d", rs->md.raid_disks); for (i = 0; i < rs->md.raid_disks; i++) { DMEMIT(" -"); /* metadata device */ From 46bed2b5c16bb7c82e1088d7ae75fb958c8a8c4e Mon Sep 17 00:00:00 2001 From: Jonathan Brassow Date: Tue, 2 Aug 2011 12:32:07 +0100 Subject: [PATCH 193/439] dm raid: add write_mostly parameter Add the write_mostly parameter to RAID1 dm-raid tables. This allows the user to set the WriteMostly flag on a RAID1 device that should normally be avoided for read I/O. Signed-off-by: Jonathan Brassow Signed-off-by: Alasdair G Kergon --- Documentation/device-mapper/dm-raid.txt | 4 +++- drivers/md/dm-raid.c | 26 ++++++++++++++++++++++++- 2 files changed, 28 insertions(+), 2 deletions(-) diff --git a/Documentation/device-mapper/dm-raid.txt b/Documentation/device-mapper/dm-raid.txt index be4419a30781..a7d1c4abc927 100644 --- a/Documentation/device-mapper/dm-raid.txt +++ b/Documentation/device-mapper/dm-raid.txt @@ -50,6 +50,7 @@ The target is named "raid" and it accepts the following parameters: [min_recovery_rate ] Throttle RAID initialization [max_recovery_rate ] Throttle RAID initialization + [write_mostly ] Drive index is write-mostly [max_write_behind ] See '-write-behind=' (man mdadm) [stripe_cache ] Stripe cache size (higher RAIDs only) [region_size ] @@ -87,9 +88,10 @@ Example tables 5 - 8:17 - 8:33 - 8:49 - 8:65 - 8:81 'dmsetup table' displays the table used to construct the mapping. -The optional parameters will always be printed in the order listed +The optional parameters are always printed in the order listed above with "sync" or "nosync" always output ahead of the other arguments, regardless of the order used when originally loading the table. +Arguments that can be repeated are ordered by value. 'dmsetup status' yields information on the state and health of the array. diff --git a/drivers/md/dm-raid.c b/drivers/md/dm-raid.c index a8a1915a450d..88143a0303d2 100644 --- a/drivers/md/dm-raid.c +++ b/drivers/md/dm-raid.c @@ -308,6 +308,7 @@ static int validate_region_size(struct raid_set *rs, unsigned long region_size) * [daemon_sleep ] Time between bitmap daemon work to clear bits * [min_recovery_rate ] Throttle RAID initialization * [max_recovery_rate ] Throttle RAID initialization + * [write_mostly ] Indicate a write mostly drive via index * [max_write_behind ] See '-write-behind=' (man mdadm) * [stripe_cache ] Stripe cache size for higher RAIDs * [region_size ] Defines granularity of bitmap @@ -376,7 +377,21 @@ static int parse_raid_params(struct raid_set *rs, char **argv, clear_bit(In_sync, &rs->dev[value].rdev.flags); rs->dev[value].rdev.recovery_offset = 0; rs->print_flags |= DMPF_REBUILD; + } else if (!strcasecmp(key, "write_mostly")) { + if (rs->raid_type->level != 1) { + rs->ti->error = "write_mostly option is only valid for RAID1"; + return -EINVAL; + } + if (value > rs->md.raid_disks) { + rs->ti->error = "Invalid write_mostly drive index given"; + return -EINVAL; + } + set_bit(WriteMostly, &rs->dev[value].rdev.flags); } else if (!strcasecmp(key, "max_write_behind")) { + if (rs->raid_type->level != 1) { + rs->ti->error = "max_write_behind option is only valid for RAID1"; + return -EINVAL; + } rs->print_flags |= DMPF_MAX_WRITE_BEHIND; /* @@ -621,11 +636,15 @@ static int raid_status(struct dm_target *ti, status_type_t type, break; case STATUSTYPE_TABLE: /* The string you would use to construct this array */ - for (i = 0; i < rs->md.raid_disks; i++) + for (i = 0; i < rs->md.raid_disks; i++) { if ((rs->print_flags & DMPF_REBUILD) && rs->dev[i].data_dev && !test_bit(In_sync, &rs->dev[i].rdev.flags)) raid_param_cnt += 2; /* for rebuilds */ + if (rs->dev[i].data_dev && + test_bit(WriteMostly, &rs->dev[i].rdev.flags)) + raid_param_cnt += 2; + } raid_param_cnt += (hweight64(rs->print_flags & ~DMPF_REBUILD) * 2); if (rs->print_flags & (DMPF_SYNC | DMPF_NOSYNC)) @@ -656,6 +675,11 @@ static int raid_status(struct dm_target *ti, status_type_t type, if (rs->print_flags & DMPF_MAX_RECOVERY_RATE) DMEMIT(" max_recovery_rate %d", rs->md.sync_speed_max); + for (i = 0; i < rs->md.raid_disks; i++) + if (rs->dev[i].data_dev && + test_bit(WriteMostly, &rs->dev[i].rdev.flags)) + DMEMIT(" write_mostly %u", i); + if (rs->print_flags & DMPF_MAX_WRITE_BEHIND) DMEMIT(" max_write_behind %lu", rs->md.bitmap_info.max_write_behind); From b12d437b73d32203a41fde0d407e91812c866844 Mon Sep 17 00:00:00 2001 From: Jonathan Brassow Date: Tue, 2 Aug 2011 12:32:07 +0100 Subject: [PATCH 194/439] dm raid: support metadata devices Add the ability to parse and use metadata devices to dm-raid. Although not strictly required, without the metadata devices, many features of RAID are unavailable. They are used to store a superblock and bitmap. The role, or position in the array, of each device must be recorded in its superblock. This is to help with fault handling, array reshaping, and sanity checks. RAID 4/5/6 devices must be loaded in a specific order: in this way, the 'array_position' field helps validate the correctness of the mapping when it is loaded. It can be used during reshaping to identify which devices are added/removed. Fault handling is impossible without this field. For example, when a device fails it is recorded in the superblock. If this is a RAID1 device and the offending device is removed from the array, there must be a way during subsequent array assembly to determine that the failed device was the one removed. This is done by correlating the 'array_position' field and the bit-field variable 'failed_devices'. Signed-off-by: Jonathan Brassow Signed-off-by: Alasdair G Kergon --- Documentation/device-mapper/dm-raid.txt | 12 +- drivers/md/Kconfig | 5 +- drivers/md/dm-raid.c | 421 +++++++++++++++++++++++- 3 files changed, 413 insertions(+), 25 deletions(-) diff --git a/Documentation/device-mapper/dm-raid.txt b/Documentation/device-mapper/dm-raid.txt index a7d1c4abc927..2a8c11331d2d 100644 --- a/Documentation/device-mapper/dm-raid.txt +++ b/Documentation/device-mapper/dm-raid.txt @@ -11,6 +11,7 @@ The target is named "raid" and it accepts the following parameters: <#raid_devs> [.. ] : + raid1 RAID1 mirroring raid4 RAID4 dedicated parity disk raid5_la RAID5 left asymmetric - rotating parity 0 with data continuation @@ -61,8 +62,7 @@ The target is named "raid" and it accepts the following parameters: <#raid_devs>: The number of devices composing the array. Each device consists of two entries. The first is the device containing the metadata (if any); the second is the one containing the - data. Currently, separate metadata devices are not supported and '-' - is required in place of the metadata device. + data. If a drive has failed or is missing at creation time, a '-' can be given for both the metadata and data drives for a given position. @@ -70,7 +70,7 @@ The target is named "raid" and it accepts the following parameters: Example tables -------------- -# RAID4 - 4 data drives, 1 parity +# RAID4 - 4 data drives, 1 parity (no metadata devices) # No metadata devices specified to hold superblock/bitmap info # Chunk size of 1MiB # (Lines separated for easy reading) @@ -79,13 +79,13 @@ Example tables raid4 1 2048 \ 5 - 8:17 - 8:33 - 8:49 - 8:65 - 8:81 -# RAID4 - 4 data drives, 1 parity (no metadata devices) +# RAID4 - 4 data drives, 1 parity (with metadata devices) # Chunk size of 1MiB, force RAID initialization, # min recovery rate at 20 kiB/sec/disk 0 1960893648 raid \ - raid4 4 2048 min_recovery_rate 20 sync\ - 5 - 8:17 - 8:33 - 8:49 - 8:65 - 8:81 + raid4 4 2048 sync min_recovery_rate 20 \ + 5 8:17 8:18 8:33 8:34 8:49 8:50 8:65 8:66 8:81 8:82 'dmsetup table' displays the table used to construct the mapping. The optional parameters are always printed in the order listed diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig index 8420129fc5ee..f75a66e7d312 100644 --- a/drivers/md/Kconfig +++ b/drivers/md/Kconfig @@ -241,12 +241,13 @@ config DM_MIRROR needed for live data migration tools such as 'pvmove'. config DM_RAID - tristate "RAID 4/5/6 target (EXPERIMENTAL)" + tristate "RAID 1/4/5/6 target (EXPERIMENTAL)" depends on BLK_DEV_DM && EXPERIMENTAL + select MD_RAID1 select MD_RAID456 select BLK_DEV_MD ---help--- - A dm target that supports RAID4, RAID5 and RAID6 mappings + A dm target that supports RAID1, RAID4, RAID5 and RAID6 mappings A RAID-5 set of N drives with a capacity of C MB per drive provides the capacity of C * (N - 1) MB, and protects against a failure diff --git a/drivers/md/dm-raid.c b/drivers/md/dm-raid.c index 88143a0303d2..69873806c50c 100644 --- a/drivers/md/dm-raid.c +++ b/drivers/md/dm-raid.c @@ -16,12 +16,10 @@ #define DM_MSG_PREFIX "raid" /* - * If the MD doesn't support MD_SYNC_STATE_FORCED yet, then - * make it so the flag doesn't set anything. + * The following flags are used by dm-raid.c to set up the array state. + * They must be cleared before md_run is called. */ -#ifndef MD_SYNC_STATE_FORCED -#define MD_SYNC_STATE_FORCED 0 -#endif +#define FirstUse 10 /* rdev flag */ struct raid_dev { /* @@ -149,9 +147,16 @@ static void context_free(struct raid_set *rs) { int i; - for (i = 0; i < rs->md.raid_disks; i++) + for (i = 0; i < rs->md.raid_disks; i++) { + if (rs->dev[i].meta_dev) + dm_put_device(rs->ti, rs->dev[i].meta_dev); + if (rs->dev[i].rdev.sb_page) + put_page(rs->dev[i].rdev.sb_page); + rs->dev[i].rdev.sb_page = NULL; + rs->dev[i].rdev.sb_loaded = 0; if (rs->dev[i].data_dev) dm_put_device(rs->ti, rs->dev[i].data_dev); + } kfree(rs); } @@ -161,7 +166,16 @@ static void context_free(struct raid_set *rs) * : meta device name or '-' if missing * : data device name or '-' if missing * - * This code parses those words. + * The following are permitted: + * - - + * - + * + * + * The following is not allowed: + * - + * + * This code parses those words. If there is a failure, + * the caller must use context_free to unwind the operations. */ static int dev_parms(struct raid_set *rs, char **argv) { @@ -184,8 +198,16 @@ static int dev_parms(struct raid_set *rs, char **argv) rs->dev[i].rdev.mddev = &rs->md; if (strcmp(argv[0], "-")) { - rs->ti->error = "Metadata devices not supported"; - return -EINVAL; + ret = dm_get_device(rs->ti, argv[0], + dm_table_get_mode(rs->ti->table), + &rs->dev[i].meta_dev); + rs->ti->error = "RAID metadata device lookup failure"; + if (ret) + return ret; + + rs->dev[i].rdev.sb_page = alloc_page(GFP_KERNEL); + if (!rs->dev[i].rdev.sb_page) + return -ENOMEM; } if (!strcmp(argv[1], "-")) { @@ -195,6 +217,10 @@ static int dev_parms(struct raid_set *rs, char **argv) return -EINVAL; } + rs->ti->error = "No data device supplied with metadata device"; + if (rs->dev[i].meta_dev) + return -EINVAL; + continue; } @@ -206,6 +232,10 @@ static int dev_parms(struct raid_set *rs, char **argv) return ret; } + if (rs->dev[i].meta_dev) { + metadata_available = 1; + rs->dev[i].rdev.meta_bdev = rs->dev[i].meta_dev->bdev; + } rs->dev[i].rdev.bdev = rs->dev[i].data_dev->bdev; list_add(&rs->dev[i].rdev.same_set, &rs->md.disks); if (!test_bit(In_sync, &rs->dev[i].rdev.flags)) @@ -333,23 +363,40 @@ static int parse_raid_params(struct raid_set *rs, char **argv, argv++; num_raid_params--; + /* + * We set each individual device as In_sync with a completed + * 'recovery_offset'. If there has been a device failure or + * replacement then one of the following cases applies: + * + * 1) User specifies 'rebuild'. + * - Device is reset when param is read. + * 2) A new device is supplied. + * - No matching superblock found, resets device. + * 3) Device failure was transient and returns on reload. + * - Failure noticed, resets device for bitmap replay. + * 4) Device hadn't completed recovery after previous failure. + * - Superblock is read and overrides recovery_offset. + * + * What is found in the superblocks of the devices is always + * authoritative, unless 'rebuild' or '[no]sync' was specified. + */ + for (i = 0; i < rs->md.raid_disks; i++) { + set_bit(In_sync, &rs->dev[i].rdev.flags); + rs->dev[i].rdev.recovery_offset = MaxSector; + } + /* * Second, parse the unordered optional arguments */ - for (i = 0; i < rs->md.raid_disks; i++) - set_bit(In_sync, &rs->dev[i].rdev.flags); - for (i = 0; i < num_raid_params; i++) { if (!strcasecmp(argv[i], "nosync")) { rs->md.recovery_cp = MaxSector; rs->print_flags |= DMPF_NOSYNC; - rs->md.flags |= MD_SYNC_STATE_FORCED; continue; } if (!strcasecmp(argv[i], "sync")) { rs->md.recovery_cp = 0; rs->print_flags |= DMPF_SYNC; - rs->md.flags |= MD_SYNC_STATE_FORCED; continue; } @@ -481,14 +528,345 @@ static int raid_is_congested(struct dm_target_callbacks *cb, int bits) return md_raid5_congested(&rs->md, bits); } +/* + * This structure is never routinely used by userspace, unlike md superblocks. + * Devices with this superblock should only ever be accessed via device-mapper. + */ +#define DM_RAID_MAGIC 0x64526D44 +struct dm_raid_superblock { + __le32 magic; /* "DmRd" */ + __le32 features; /* Used to indicate possible future changes */ + + __le32 num_devices; /* Number of devices in this array. (Max 64) */ + __le32 array_position; /* The position of this drive in the array */ + + __le64 events; /* Incremented by md when superblock updated */ + __le64 failed_devices; /* Bit field of devices to indicate failures */ + + /* + * This offset tracks the progress of the repair or replacement of + * an individual drive. + */ + __le64 disk_recovery_offset; + + /* + * This offset tracks the progress of the initial array + * synchronisation/parity calculation. + */ + __le64 array_resync_offset; + + /* + * RAID characteristics + */ + __le32 level; + __le32 layout; + __le32 stripe_sectors; + + __u8 pad[452]; /* Round struct to 512 bytes. */ + /* Always set to 0 when writing. */ +} __packed; + +static int read_disk_sb(mdk_rdev_t *rdev, int size) +{ + BUG_ON(!rdev->sb_page); + + if (rdev->sb_loaded) + return 0; + + if (!sync_page_io(rdev, 0, size, rdev->sb_page, READ, 1)) { + DMERR("Failed to read device superblock"); + return -EINVAL; + } + + rdev->sb_loaded = 1; + + return 0; +} + +static void super_sync(mddev_t *mddev, mdk_rdev_t *rdev) +{ + mdk_rdev_t *r, *t; + uint64_t failed_devices; + struct dm_raid_superblock *sb; + + sb = page_address(rdev->sb_page); + failed_devices = le64_to_cpu(sb->failed_devices); + + rdev_for_each(r, t, mddev) + if ((r->raid_disk >= 0) && test_bit(Faulty, &r->flags)) + failed_devices |= (1ULL << r->raid_disk); + + memset(sb, 0, sizeof(*sb)); + + sb->magic = cpu_to_le32(DM_RAID_MAGIC); + sb->features = cpu_to_le32(0); /* No features yet */ + + sb->num_devices = cpu_to_le32(mddev->raid_disks); + sb->array_position = cpu_to_le32(rdev->raid_disk); + + sb->events = cpu_to_le64(mddev->events); + sb->failed_devices = cpu_to_le64(failed_devices); + + sb->disk_recovery_offset = cpu_to_le64(rdev->recovery_offset); + sb->array_resync_offset = cpu_to_le64(mddev->recovery_cp); + + sb->level = cpu_to_le32(mddev->level); + sb->layout = cpu_to_le32(mddev->layout); + sb->stripe_sectors = cpu_to_le32(mddev->chunk_sectors); +} + +/* + * super_load + * + * This function creates a superblock if one is not found on the device + * and will decide which superblock to use if there's a choice. + * + * Return: 1 if use rdev, 0 if use refdev, -Exxx otherwise + */ +static int super_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev) +{ + int ret; + struct dm_raid_superblock *sb; + struct dm_raid_superblock *refsb; + uint64_t events_sb, events_refsb; + + rdev->sb_start = 0; + rdev->sb_size = sizeof(*sb); + + ret = read_disk_sb(rdev, rdev->sb_size); + if (ret) + return ret; + + sb = page_address(rdev->sb_page); + if (sb->magic != cpu_to_le32(DM_RAID_MAGIC)) { + super_sync(rdev->mddev, rdev); + + set_bit(FirstUse, &rdev->flags); + + /* Force writing of superblocks to disk */ + set_bit(MD_CHANGE_DEVS, &rdev->mddev->flags); + + /* Any superblock is better than none, choose that if given */ + return refdev ? 0 : 1; + } + + if (!refdev) + return 1; + + events_sb = le64_to_cpu(sb->events); + + refsb = page_address(refdev->sb_page); + events_refsb = le64_to_cpu(refsb->events); + + return (events_sb > events_refsb) ? 1 : 0; +} + +static int super_init_validation(mddev_t *mddev, mdk_rdev_t *rdev) +{ + int role; + struct raid_set *rs = container_of(mddev, struct raid_set, md); + uint64_t events_sb; + uint64_t failed_devices; + struct dm_raid_superblock *sb; + uint32_t new_devs = 0; + uint32_t rebuilds = 0; + mdk_rdev_t *r, *t; + struct dm_raid_superblock *sb2; + + sb = page_address(rdev->sb_page); + events_sb = le64_to_cpu(sb->events); + failed_devices = le64_to_cpu(sb->failed_devices); + + /* + * Initialise to 1 if this is a new superblock. + */ + mddev->events = events_sb ? : 1; + + /* + * Reshaping is not currently allowed + */ + if ((le32_to_cpu(sb->level) != mddev->level) || + (le32_to_cpu(sb->layout) != mddev->layout) || + (le32_to_cpu(sb->stripe_sectors) != mddev->chunk_sectors)) { + DMERR("Reshaping arrays not yet supported."); + return -EINVAL; + } + + /* We can only change the number of devices in RAID1 right now */ + if ((rs->raid_type->level != 1) && + (le32_to_cpu(sb->num_devices) != mddev->raid_disks)) { + DMERR("Reshaping arrays not yet supported."); + return -EINVAL; + } + + if (!(rs->print_flags & (DMPF_SYNC | DMPF_NOSYNC))) + mddev->recovery_cp = le64_to_cpu(sb->array_resync_offset); + + /* + * During load, we set FirstUse if a new superblock was written. + * There are two reasons we might not have a superblock: + * 1) The array is brand new - in which case, all of the + * devices must have their In_sync bit set. Also, + * recovery_cp must be 0, unless forced. + * 2) This is a new device being added to an old array + * and the new device needs to be rebuilt - in which + * case the In_sync bit will /not/ be set and + * recovery_cp must be MaxSector. + */ + rdev_for_each(r, t, mddev) { + if (!test_bit(In_sync, &r->flags)) { + if (!test_bit(FirstUse, &r->flags)) + DMERR("Superblock area of " + "rebuild device %d should have been " + "cleared.", r->raid_disk); + set_bit(FirstUse, &r->flags); + rebuilds++; + } else if (test_bit(FirstUse, &r->flags)) + new_devs++; + } + + if (!rebuilds) { + if (new_devs == mddev->raid_disks) { + DMINFO("Superblocks created for new array"); + set_bit(MD_ARRAY_FIRST_USE, &mddev->flags); + } else if (new_devs) { + DMERR("New device injected " + "into existing array without 'rebuild' " + "parameter specified"); + return -EINVAL; + } + } else if (new_devs) { + DMERR("'rebuild' devices cannot be " + "injected into an array with other first-time devices"); + return -EINVAL; + } else if (mddev->recovery_cp != MaxSector) { + DMERR("'rebuild' specified while array is not in-sync"); + return -EINVAL; + } + + /* + * Now we set the Faulty bit for those devices that are + * recorded in the superblock as failed. + */ + rdev_for_each(r, t, mddev) { + if (!r->sb_page) + continue; + sb2 = page_address(r->sb_page); + sb2->failed_devices = 0; + + /* + * Check for any device re-ordering. + */ + if (!test_bit(FirstUse, &r->flags) && (r->raid_disk >= 0)) { + role = le32_to_cpu(sb2->array_position); + if (role != r->raid_disk) { + if (rs->raid_type->level != 1) { + rs->ti->error = "Cannot change device " + "positions in RAID array"; + return -EINVAL; + } + DMINFO("RAID1 device #%d now at position #%d", + role, r->raid_disk); + } + + /* + * Partial recovery is performed on + * returning failed devices. + */ + if (failed_devices & (1 << role)) + set_bit(Faulty, &r->flags); + } + } + + return 0; +} + +static int super_validate(mddev_t *mddev, mdk_rdev_t *rdev) +{ + struct dm_raid_superblock *sb = page_address(rdev->sb_page); + + /* + * If mddev->events is not set, we know we have not yet initialized + * the array. + */ + if (!mddev->events && super_init_validation(mddev, rdev)) + return -EINVAL; + + mddev->bitmap_info.offset = 4096 >> 9; /* Enable bitmap creation */ + rdev->mddev->bitmap_info.default_offset = 4096 >> 9; + if (!test_bit(FirstUse, &rdev->flags)) { + rdev->recovery_offset = le64_to_cpu(sb->disk_recovery_offset); + if (rdev->recovery_offset != MaxSector) + clear_bit(In_sync, &rdev->flags); + } + + /* + * If a device comes back, set it as not In_sync and no longer faulty. + */ + if (test_bit(Faulty, &rdev->flags)) { + clear_bit(Faulty, &rdev->flags); + clear_bit(In_sync, &rdev->flags); + rdev->saved_raid_disk = rdev->raid_disk; + rdev->recovery_offset = 0; + } + + clear_bit(FirstUse, &rdev->flags); + + return 0; +} + +/* + * Analyse superblocks and select the freshest. + */ +static int analyse_superblocks(struct dm_target *ti, struct raid_set *rs) +{ + int ret; + mdk_rdev_t *rdev, *freshest, *tmp; + mddev_t *mddev = &rs->md; + + freshest = NULL; + rdev_for_each(rdev, tmp, mddev) { + if (!rdev->meta_bdev) + continue; + + ret = super_load(rdev, freshest); + + switch (ret) { + case 1: + freshest = rdev; + break; + case 0: + break; + default: + ti->error = "Failed to load superblock"; + return ret; + } + } + + if (!freshest) + return 0; + + /* + * Validation of the freshest device provides the source of + * validation for the remaining devices. + */ + ti->error = "Unable to assemble array: Invalid superblocks"; + if (super_validate(mddev, freshest)) + return -EINVAL; + + rdev_for_each(rdev, tmp, mddev) + if ((rdev != freshest) && super_validate(mddev, rdev)) + return -EINVAL; + + return 0; +} + /* * Construct a RAID4/5/6 mapping: * Args: * <#raid_params> \ * <#raid_devs> { .. } * - * ** metadata devices are not supported yet, use '-' instead ** - * * varies by . See 'parse_raid_params' for * details on possible . */ @@ -556,6 +934,11 @@ static int raid_ctr(struct dm_target *ti, unsigned argc, char **argv) if (ret) goto bad; + rs->md.sync_super = super_sync; + ret = analyse_superblocks(ti, rs); + if (ret) + goto bad; + INIT_WORK(&rs->md.event_work, do_table_event); ti->private = rs; @@ -698,7 +1081,10 @@ static int raid_status(struct dm_target *ti, status_type_t type, DMEMIT(" %d", rs->md.raid_disks); for (i = 0; i < rs->md.raid_disks; i++) { - DMEMIT(" -"); /* metadata device */ + if (rs->dev[i].meta_dev) + DMEMIT(" %s", rs->dev[i].meta_dev->name); + else + DMEMIT(" -"); if (rs->dev[i].data_dev) DMEMIT(" %s", rs->dev[i].data_dev->name); @@ -755,6 +1141,7 @@ static void raid_resume(struct dm_target *ti) { struct raid_set *rs = ti->private; + bitmap_load(&rs->md); mddev_resume(&rs->md); } From 327372797c88b24953f454cd51a3734c02697bdd Mon Sep 17 00:00:00 2001 From: Jonathan Brassow Date: Tue, 2 Aug 2011 12:32:07 +0100 Subject: [PATCH 195/439] dm raid: add md raid1 support Support the MD RAID1 personality through dm-raid. Signed-off-by: Jonathan Brassow Signed-off-by: Alasdair G Kergon --- drivers/md/dm-raid.c | 49 +++++++++++++++++++++++++++++++++++--------- 1 file changed, 39 insertions(+), 10 deletions(-) diff --git a/drivers/md/dm-raid.c b/drivers/md/dm-raid.c index 69873806c50c..a002dd85db1e 100644 --- a/drivers/md/dm-raid.c +++ b/drivers/md/dm-raid.c @@ -8,6 +8,7 @@ #include #include "md.h" +#include "raid1.h" #include "raid5.h" #include "bitmap.h" @@ -72,6 +73,7 @@ static struct raid_type { const unsigned level; /* RAID level. */ const unsigned algorithm; /* RAID algorithm. */ } raid_types[] = { + {"raid1", "RAID1 (mirroring)", 0, 2, 1, 0 /* NONE */}, {"raid4", "RAID4 (dedicated parity disk)", 1, 2, 5, ALGORITHM_PARITY_0}, {"raid5_la", "RAID5 (left asymmetric)", 1, 2, 5, ALGORITHM_LEFT_ASYMMETRIC}, {"raid5_ra", "RAID5 (right asymmetric)", 1, 2, 5, ALGORITHM_RIGHT_ASYMMETRIC}, @@ -105,7 +107,8 @@ static struct raid_set *context_alloc(struct dm_target *ti, struct raid_type *ra } sectors_per_dev = ti->len; - if (sector_div(sectors_per_dev, (raid_devs - raid_type->parity_devs))) { + if ((raid_type->level > 1) && + sector_div(sectors_per_dev, (raid_devs - raid_type->parity_devs))) { ti->error = "Target length not divisible by number of data devices"; return ERR_PTR(-EINVAL); } @@ -329,13 +332,16 @@ static int validate_region_size(struct raid_set *rs, unsigned long region_size) /* * Possible arguments are... - * RAID456: * [optional_args] * - * Optional args: - * [[no]sync] Force or prevent recovery of the entire array + * Argument definitions + * The number of sectors per disk that + * will form the "stripe" + * [[no]sync] Force or prevent recovery of the + * entire array * [rebuild ] Rebuild the drive indicated by the index - * [daemon_sleep ] Time between bitmap daemon work to clear bits + * [daemon_sleep ] Time between bitmap daemon work to + * clear bits * [min_recovery_rate ] Throttle RAID initialization * [max_recovery_rate ] Throttle RAID initialization * [write_mostly ] Indicate a write mostly drive via index @@ -352,11 +358,21 @@ static int parse_raid_params(struct raid_set *rs, char **argv, /* * First, parse the in-order required arguments + * "chunk_size" is the only argument of this type. */ - if ((strict_strtoul(argv[0], 10, &value) < 0) || - !is_power_of_2(value) || (value < 8)) { + if ((strict_strtoul(argv[0], 10, &value) < 0)) { rs->ti->error = "Bad chunk size"; return -EINVAL; + } else if (rs->raid_type->level == 1) { + if (value) + DMERR("Ignoring chunk size parameter for RAID 1"); + value = 0; + } else if (!is_power_of_2(value)) { + rs->ti->error = "Chunk size must be a power of 2"; + return -EINVAL; + } else if (value < 8) { + rs->ti->error = "Chunk size value is too small"; + return -EINVAL; } rs->md.new_chunk_sectors = rs->md.chunk_sectors = value; @@ -413,8 +429,12 @@ static int parse_raid_params(struct raid_set *rs, char **argv, } if (!strcasecmp(key, "rebuild")) { - if (++rebuild_cnt > rs->raid_type->parity_devs) { - rs->ti->error = "Too many rebuild drives given"; + rebuild_cnt++; + if (((rs->raid_type->level != 1) && + (rebuild_cnt > rs->raid_type->parity_devs)) || + ((rs->raid_type->level == 1) && + (rebuild_cnt > (rs->md.raid_disks - 1)))) { + rs->ti->error = "Too many rebuild devices specified for given RAID type"; return -EINVAL; } if (value > rs->md.raid_disks) { @@ -507,6 +527,11 @@ static int parse_raid_params(struct raid_set *rs, char **argv, else rs->ti->split_io = region_size; + if (rs->md.chunk_sectors) + rs->ti->split_io = rs->md.chunk_sectors; + else + rs->ti->split_io = region_size; + /* Assume there are no metadata devices until the drives are parsed */ rs->md.persistent = 0; rs->md.external = 1; @@ -525,6 +550,9 @@ static int raid_is_congested(struct dm_target_callbacks *cb, int bits) { struct raid_set *rs = container_of(cb, struct raid_set, callbacks); + if (rs->raid_type->level == 1) + return md_raid1_congested(&rs->md, bits); + return md_raid5_congested(&rs->md, bits); } @@ -955,6 +983,7 @@ static int raid_ctr(struct dm_target *ti, unsigned argc, char **argv) rs->callbacks.congested_fn = raid_is_congested; dm_table_add_target_callbacks(ti->table, &rs->callbacks); + mddev_suspend(&rs->md); return 0; bad: @@ -1147,7 +1176,7 @@ static void raid_resume(struct dm_target *ti) static struct target_type raid_target = { .name = "raid", - .version = {1, 0, 0}, + .version = {1, 1, 0}, .module = THIS_MODULE, .ctr = raid_ctr, .dtr = raid_dtr, From 772ae5f54d69c38a5e3c4352c5fdbdaff141af21 Mon Sep 17 00:00:00 2001 From: Milan Broz Date: Tue, 2 Aug 2011 12:32:08 +0100 Subject: [PATCH 196/439] dm crypt: optionally support discard requests Add optional parameter field to dmcrypt table and support "allow_discards" option. Discard requests bypass crypt queue processing. Bio is simple remapped to underlying device. Note that discard will be never enabled by default because of security consequences. It is up to the administrator to enable it for encrypted devices. (Note that userspace cryptsetup does not understand new optional parameters yet. Support for this will come later. Until then, you should use 'dmsetup' to enable and disable this.) Signed-off-by: Milan Broz Signed-off-by: Alasdair G Kergon --- Documentation/device-mapper/dm-crypt.txt | 21 +++++++++- drivers/md/dm-crypt.c | 49 ++++++++++++++++++++++-- 2 files changed, 65 insertions(+), 5 deletions(-) diff --git a/Documentation/device-mapper/dm-crypt.txt b/Documentation/device-mapper/dm-crypt.txt index 6b5c42dbbe84..2c656ae43ba7 100644 --- a/Documentation/device-mapper/dm-crypt.txt +++ b/Documentation/device-mapper/dm-crypt.txt @@ -4,7 +4,8 @@ dm-crypt Device-Mapper's "crypt" target provides transparent encryption of block devices using the kernel crypto API. -Parameters: +Parameters: \ + [<#opt_params> ] Encryption cipher and an optional IV generation mode. @@ -37,6 +38,24 @@ Parameters: Starting sector within the device where the encrypted data begins. +<#opt_params> + Number of optional parameters. If there are no optional parameters, + the optional paramaters section can be skipped or #opt_params can be zero. + Otherwise #opt_params is the number of following arguments. + + Example of optional parameters section: + 1 allow_discards + +allow_discards + Block discard requests (a.k.a. TRIM) are passed through the crypt device. + The default is to ignore discard requests. + + WARNING: Assess the specific security risks carefully before enabling this + option. For example, allowing discards on encrypted devices may lead to + the leak of information about the ciphertext device (filesystem type, + used space etc.) if the discarded blocks can be located easily on the + device later. + Example scripts =============== LUKS (Linux Unified Key Setup) is now the preferred way to set up disk diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index b79e7472a9b3..49da55c1528a 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -1574,11 +1574,17 @@ bad_mem: static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv) { struct crypt_config *cc; - unsigned int key_size; + unsigned int key_size, opt_params; unsigned long long tmpll; int ret; + struct dm_arg_set as; + const char *opt_string; - if (argc != 5) { + static struct dm_arg _args[] = { + {0, 1, "Invalid number of feature args"}, + }; + + if (argc < 5) { ti->error = "Not enough arguments"; return -EINVAL; } @@ -1647,6 +1653,30 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv) } cc->start = tmpll; + argv += 5; + argc -= 5; + + /* Optional parameters */ + if (argc) { + as.argc = argc; + as.argv = argv; + + ret = dm_read_arg_group(_args, &as, &opt_params, &ti->error); + if (ret) + goto bad; + + opt_string = dm_shift_arg(&as); + + if (opt_params == 1 && opt_string && + !strcasecmp(opt_string, "allow_discards")) + ti->num_discard_requests = 1; + else if (opt_params) { + ret = -EINVAL; + ti->error = "Invalid feature arguments"; + goto bad; + } + } + ret = -ENOMEM; cc->io_queue = alloc_workqueue("kcryptd_io", WQ_NON_REENTRANT| @@ -1681,9 +1711,16 @@ static int crypt_map(struct dm_target *ti, struct bio *bio, struct dm_crypt_io *io; struct crypt_config *cc; - if (bio->bi_rw & REQ_FLUSH) { + /* + * If bio is REQ_FLUSH or REQ_DISCARD, just bypass crypt queues. + * - for REQ_FLUSH device-mapper core ensures that no IO is in-flight + * - for REQ_DISCARD caller must use flush if IO ordering matters + */ + if (unlikely(bio->bi_rw & (REQ_FLUSH | REQ_DISCARD))) { cc = ti->private; bio->bi_bdev = cc->dev->bdev; + if (bio_sectors(bio)) + bio->bi_sector = cc->start + dm_target_offset(ti, bio->bi_sector); return DM_MAPIO_REMAPPED; } @@ -1726,6 +1763,10 @@ static int crypt_status(struct dm_target *ti, status_type_t type, DMEMIT(" %llu %s %llu", (unsigned long long)cc->iv_offset, cc->dev->name, (unsigned long long)cc->start); + + if (ti->num_discard_requests) + DMEMIT(" 1 allow_discards"); + break; } return 0; @@ -1822,7 +1863,7 @@ static int crypt_iterate_devices(struct dm_target *ti, static struct target_type crypt_target = { .name = "crypt", - .version = {1, 10, 0}, + .version = {1, 11, 0}, .module = THIS_MODULE, .ctr = crypt_ctr, .dtr = crypt_dtr, From ed8b752bccf2560e305e25125721d2f0ac759e88 Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Tue, 2 Aug 2011 12:32:08 +0100 Subject: [PATCH 197/439] dm table: set flush capability based on underlying devices DM has always advertised both REQ_FLUSH and REQ_FUA flush capabilities regardless of whether or not a given DM device's underlying devices also advertised a need for them. Block's flush-merge changes from 2.6.39 have proven to be more costly for DM devices. Performance regressions have been reported even when DM's underlying devices do not advertise that they have a write cache. Fix the performance regressions by configuring a DM device's flushing capabilities based on those of the underlying devices' capabilities. Signed-off-by: Mike Snitzer Signed-off-by: Alasdair G Kergon --- drivers/md/dm-table.c | 43 +++++++++++++++++++++++++++++++++++++++++++ drivers/md/dm.c | 1 - 2 files changed, 43 insertions(+), 1 deletion(-) diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index 259ce99302fc..986b8754bb08 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -1248,9 +1248,45 @@ static void dm_table_set_integrity(struct dm_table *t) blk_get_integrity(template_disk)); } +static int device_flush_capable(struct dm_target *ti, struct dm_dev *dev, + sector_t start, sector_t len, void *data) +{ + unsigned flush = (*(unsigned *)data); + struct request_queue *q = bdev_get_queue(dev->bdev); + + return q && (q->flush_flags & flush); +} + +static bool dm_table_supports_flush(struct dm_table *t, unsigned flush) +{ + struct dm_target *ti; + unsigned i = 0; + + /* + * Require at least one underlying device to support flushes. + * t->devices includes internal dm devices such as mirror logs + * so we need to use iterate_devices here, which targets + * supporting flushes must provide. + */ + while (i < dm_table_get_num_targets(t)) { + ti = dm_table_get_target(t, i++); + + if (!ti->num_flush_requests) + continue; + + if (ti->type->iterate_devices && + ti->type->iterate_devices(ti, device_flush_capable, &flush)) + return 1; + } + + return 0; +} + void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q, struct queue_limits *limits) { + unsigned flush = 0; + /* * Copy table's limits to the DM device's request_queue */ @@ -1261,6 +1297,13 @@ void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q, else queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, q); + if (dm_table_supports_flush(t, REQ_FLUSH)) { + flush |= REQ_FLUSH; + if (dm_table_supports_flush(t, REQ_FUA)) + flush |= REQ_FUA; + } + blk_queue_flush(q, flush); + dm_table_set_integrity(t); /* diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 1000eaf984ef..52b39f335bb3 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -1808,7 +1808,6 @@ static void dm_init_md_queue(struct mapped_device *md) blk_queue_make_request(md->queue, dm_request); blk_queue_bounce_limit(md->queue, BLK_BOUNCE_ANY); blk_queue_merge_bvec(md->queue, dm_merge_bvec); - blk_queue_flush(md->queue, REQ_FLUSH | REQ_FUA); } /* From 50658b6602dad6dc76c226917f5d8cec0f680eab Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Tue, 2 Aug 2011 11:29:24 +0200 Subject: [PATCH 198/439] spi/pl022: remove function cannot exit The remove function in the PL022 driver cannot abort the remove function any way, so restructure the code so as not to make that assumption. Remove will now proceed no matter whether it can stop the transfer queue or not. Reported-by: Russell King Signed-off-by: Linus Walleij Signed-off-by: Grant Likely --- drivers/spi/spi-pl022.c | 11 +++-------- 1 file changed, 3 insertions(+), 8 deletions(-) diff --git a/drivers/spi/spi-pl022.c b/drivers/spi/spi-pl022.c index eba88c749fb1..730b4a37b823 100644 --- a/drivers/spi/spi-pl022.c +++ b/drivers/spi/spi-pl022.c @@ -2267,17 +2267,13 @@ static int __devexit pl022_remove(struct amba_device *adev) { struct pl022 *pl022 = amba_get_drvdata(adev); - int status = 0; + if (!pl022) return 0; /* Remove the queue */ - status = destroy_queue(pl022); - if (status != 0) { - dev_err(&adev->dev, - "queue remove failed (%d)\n", status); - return status; - } + if (destroy_queue(pl022) != 0) + dev_err(&adev->dev, "queue remove failed\n"); load_ssp_default_config(pl022); pl022_dma_remove(pl022); free_irq(adev->irq[0], pl022); @@ -2289,7 +2285,6 @@ pl022_remove(struct amba_device *adev) spi_unregister_master(pl022->master); spi_master_put(pl022->master); amba_set_drvdata(adev, NULL); - dev_dbg(&adev->dev, "remove succeeded\n"); return 0; } From aac285c6cb9622f1cc05ed162604bf5ad2da9a8d Mon Sep 17 00:00:00 2001 From: Jamie Iles Date: Tue, 2 Aug 2011 15:45:07 +0100 Subject: [PATCH 199/439] of: constify property name parameters for helper functions The helper functions for reading u32 integers, u32 arrays and strings should have the property name as a const pointer. Cc: Grant Likely Signed-off-by: Jamie Iles Signed-off-by: Grant Likely --- drivers/of/base.c | 7 ++++--- include/linux/of.h | 15 +++++++++------ 2 files changed, 13 insertions(+), 9 deletions(-) diff --git a/drivers/of/base.c b/drivers/of/base.c index 02ed36719def..3ff22e32b602 100644 --- a/drivers/of/base.c +++ b/drivers/of/base.c @@ -610,8 +610,9 @@ EXPORT_SYMBOL(of_find_node_by_phandle); * * The out_value is modified only if a valid u32 value can be decoded. */ -int of_property_read_u32_array(const struct device_node *np, char *propname, - u32 *out_values, size_t sz) +int of_property_read_u32_array(const struct device_node *np, + const char *propname, u32 *out_values, + size_t sz) { struct property *prop = of_find_property(np, propname, NULL); const __be32 *val; @@ -645,7 +646,7 @@ EXPORT_SYMBOL_GPL(of_property_read_u32_array); * * The out_string pointer is modified only if a valid string can be decoded. */ -int of_property_read_string(struct device_node *np, char *propname, +int of_property_read_string(struct device_node *np, const char *propname, const char **out_string) { struct property *prop = of_find_property(np, propname, NULL); diff --git a/include/linux/of.h b/include/linux/of.h index bd716f8908de..0085bb01c041 100644 --- a/include/linux/of.h +++ b/include/linux/of.h @@ -196,12 +196,13 @@ extern struct property *of_find_property(const struct device_node *np, const char *name, int *lenp); extern int of_property_read_u32_array(const struct device_node *np, - char *propname, + const char *propname, u32 *out_values, size_t sz); -extern int of_property_read_string(struct device_node *np, char *propname, - const char **out_string); +extern int of_property_read_string(struct device_node *np, + const char *propname, + const char **out_string); extern int of_device_is_compatible(const struct device_node *device, const char *); extern int of_device_is_available(const struct device_node *device); @@ -242,13 +243,15 @@ static inline bool of_have_populated_dt(void) } static inline int of_property_read_u32_array(const struct device_node *np, - char *propname, u32 *out_values, size_t sz) + const char *propname, + u32 *out_values, size_t sz) { return -ENOSYS; } static inline int of_property_read_string(struct device_node *np, - char *propname, const char **out_string) + const char *propname, + const char **out_string) { return -ENOSYS; } @@ -256,7 +259,7 @@ static inline int of_property_read_string(struct device_node *np, #endif /* CONFIG_OF */ static inline int of_property_read_u32(const struct device_node *np, - char *propname, + const char *propname, u32 *out_value) { return of_property_read_u32_array(np, propname, out_value, 1); From d945fa0da7db9ace61fbb2140cfdb1e2d2b24e7e Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Tue, 2 Aug 2011 14:13:26 +0900 Subject: [PATCH 200/439] MAINTAINERS: Add keyword match for of_match_table to device tree section If a patch is working with an of_match_table it's probably adding an OF binding to a driver in which case the binding ought to be reviewed by the device tree folks to make sure that things like the device matches are set up correctly. Signed-off-by: Mark Brown Signed-off-by: Grant Likely --- MAINTAINERS | 1 + 1 file changed, 1 insertion(+) diff --git a/MAINTAINERS b/MAINTAINERS index c9c6324a7a9f..0d2fcda465eb 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4722,6 +4722,7 @@ S: Maintained F: drivers/of F: include/linux/of*.h K: of_get_property +K: of_match_table OPENRISC ARCHITECTURE M: Jonas Bonn From cc1a93e68f6c0d736b771f0746e8e4186f483fdc Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Mon, 1 Aug 2011 12:46:57 -0700 Subject: [PATCH 201/439] iwlagn: sysfs couldn't find the priv pointer This bug has been introduced by: d593411084a56124aa9d80aafa15db8463b2d8f7 Author: Emmanuel Grumbach Date: Mon Jul 11 10:48:51 2011 +0300 iwlagn: simplify the bus architecture Revert part of the buggy patch: dev_get_drvdata will now return iwl_priv as it did before the patch. Signed-off-by: Emmanuel Grumbach Tested-by: Daniel Halperin Signed-off-by: Wey-Yi Guy Signed-off-by: John W. Linville --- drivers/net/wireless/iwlwifi/iwl-pci.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) diff --git a/drivers/net/wireless/iwlwifi/iwl-pci.c b/drivers/net/wireless/iwlwifi/iwl-pci.c index fb7e436b40c7..69d4ec467dca 100644 --- a/drivers/net/wireless/iwlwifi/iwl-pci.c +++ b/drivers/net/wireless/iwlwifi/iwl-pci.c @@ -134,6 +134,7 @@ static void iwl_pci_apm_config(struct iwl_bus *bus) static void iwl_pci_set_drv_data(struct iwl_bus *bus, void *drv_data) { bus->drv_data = drv_data; + pci_set_drvdata(IWL_BUS_GET_PCI_DEV(bus), drv_data); } static void iwl_pci_get_hw_id(struct iwl_bus *bus, char buf[], @@ -454,8 +455,6 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) pci_write_config_word(pdev, PCI_COMMAND, pci_cmd); } - pci_set_drvdata(pdev, bus); - bus->dev = &pdev->dev; bus->irq = pdev->irq; bus->ops = &pci_ops; @@ -494,11 +493,12 @@ static void iwl_pci_down(struct iwl_bus *bus) static void __devexit iwl_pci_remove(struct pci_dev *pdev) { - struct iwl_bus *bus = pci_get_drvdata(pdev); + struct iwl_priv *priv = pci_get_drvdata(pdev); + void *bus_specific = priv->bus->bus_specific; - iwl_remove(bus->drv_data); + iwl_remove(priv); - iwl_pci_down(bus); + iwl_pci_down(bus_specific); } #ifdef CONFIG_PM @@ -506,20 +506,20 @@ static void __devexit iwl_pci_remove(struct pci_dev *pdev) static int iwl_pci_suspend(struct device *device) { struct pci_dev *pdev = to_pci_dev(device); - struct iwl_bus *bus = pci_get_drvdata(pdev); + struct iwl_priv *priv = pci_get_drvdata(pdev); /* Before you put code here, think about WoWLAN. You cannot check here * whether WoWLAN is enabled or not, and your code will run even if * WoWLAN is enabled - don't kill the NIC, someone may need it in Sx. */ - return iwl_suspend(bus->drv_data); + return iwl_suspend(priv); } static int iwl_pci_resume(struct device *device) { struct pci_dev *pdev = to_pci_dev(device); - struct iwl_bus *bus = pci_get_drvdata(pdev); + struct iwl_priv *priv = pci_get_drvdata(pdev); /* Before you put code here, think about WoWLAN. You cannot check here * whether WoWLAN is enabled or not, and your code will run even if @@ -532,7 +532,7 @@ static int iwl_pci_resume(struct device *device) */ pci_write_config_byte(pdev, PCI_CFG_RETRY_TIMEOUT, 0x00); - return iwl_resume(bus->drv_data); + return iwl_resume(priv); } static SIMPLE_DEV_PM_OPS(iwl_dev_pm_ops, iwl_pci_suspend, iwl_pci_resume); From 449f94eadc91d69fa044fa92ef45c02e86559e13 Mon Sep 17 00:00:00 2001 From: Helmut Schaa Date: Tue, 2 Aug 2011 11:43:14 +0200 Subject: [PATCH 202/439] rt2x00: Fix compilation without CONFIG_RT2X00_LIB_CRYPTO This was introduced by commit 77b5621bac4a56b83b9081f48d4e7d1accdde400 (rt2x00: Don't use queue entry as parameter when creating TX descriptor.) Signed-off-by: Helmut Schaa Acked-by: Gertjan van Wingerde Acked-by: Ivo van Doorn Reported-by: Randy Dunlap Acked-by: Randy Dunlap Signed-off-by: John W. Linville --- drivers/net/wireless/rt2x00/rt2x00lib.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/rt2x00/rt2x00lib.h b/drivers/net/wireless/rt2x00/rt2x00lib.h index 15cdc7e57fc4..4cdf247a870d 100644 --- a/drivers/net/wireless/rt2x00/rt2x00lib.h +++ b/drivers/net/wireless/rt2x00/rt2x00lib.h @@ -355,7 +355,8 @@ static inline enum cipher rt2x00crypto_key_to_cipher(struct ieee80211_key_conf * return CIPHER_NONE; } -static inline void rt2x00crypto_create_tx_descriptor(struct queue_entry *entry, +static inline void rt2x00crypto_create_tx_descriptor(struct rt2x00_dev *rt2x00dev, + struct sk_buff *skb, struct txentry_desc *txdesc) { } From 00898a47269ae5e6dda04defad00234b96692d95 Mon Sep 17 00:00:00 2001 From: Stanislaw Gruszka Date: Tue, 2 Aug 2011 13:29:02 +0200 Subject: [PATCH 203/439] rt2x00: fix usage of NULL queue We may call rt2x00queue_pause_queue(queue) with queue == NULL. Bug was introduced by commit 62fe778412b36791b7897cfa139342906fbbf07b "rt2x00: Fix stuck queue in tx failure case" . Cc: stable@kernel.org # 3.0+ Signed-off-by: Stanislaw Gruszka Acked-by: Ivo van Doorn Acked-by: Gertjan van Wingerde Signed-off-by: John W. Linville --- drivers/net/wireless/rt2x00/rt2x00mac.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/rt2x00/rt2x00mac.c b/drivers/net/wireless/rt2x00/rt2x00mac.c index 8efab3983528..4ccf23805973 100644 --- a/drivers/net/wireless/rt2x00/rt2x00mac.c +++ b/drivers/net/wireless/rt2x00/rt2x00mac.c @@ -113,7 +113,7 @@ void rt2x00mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb) * due to possible race conditions in mac80211. */ if (!test_bit(DEVICE_STATE_PRESENT, &rt2x00dev->flags)) - goto exit_fail; + goto exit_free_skb; /* * Use the ATIM queue if appropriate and present. @@ -127,7 +127,7 @@ void rt2x00mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb) ERROR(rt2x00dev, "Attempt to send packet over invalid queue %d.\n" "Please file bug report to %s.\n", qid, DRV_PROJECT); - goto exit_fail; + goto exit_free_skb; } /* @@ -159,6 +159,7 @@ void rt2x00mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb) exit_fail: rt2x00queue_pause_queue(queue); + exit_free_skb: dev_kfree_skb_any(skb); } EXPORT_SYMBOL_GPL(rt2x00mac_tx); From f35291082294ca6737953bbe4e9491ede04ab822 Mon Sep 17 00:00:00 2001 From: Wey-Yi Guy Date: Tue, 2 Aug 2011 09:08:37 -0700 Subject: [PATCH 204/439] iwlagn: 5000 do not support idle mode 5000 series has issue supporting power save idle mode: commit 9dc2153315650eae220898668b6aa56a25c130be iwlwifi: always support idle mode for agn devices For agn devices, always support idle mode which help power consumption in idle unassociated state. the above changes cause 5000 become not stable when power management is "on" http://bugzilla.intellinuxwireless.org/show_bug.cgi?id=2312 Cc: stable@kernel.org #2.6.39, #3.0.0 Reported-by: Devin J Pohly Signed-off-by: Wey-Yi Guy Signed-off-by: John W. Linville --- drivers/net/wireless/iwlwifi/iwl-5000.c | 1 + drivers/net/wireless/iwlwifi/iwl-core.h | 2 ++ drivers/net/wireless/iwlwifi/iwl-power.c | 3 ++- 3 files changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/iwlwifi/iwl-5000.c b/drivers/net/wireless/iwlwifi/iwl-5000.c index 3eeb12ebe6e9..c95cefd529dc 100644 --- a/drivers/net/wireless/iwlwifi/iwl-5000.c +++ b/drivers/net/wireless/iwlwifi/iwl-5000.c @@ -365,6 +365,7 @@ static struct iwl_base_params iwl5000_base_params = { .chain_noise_scale = 1000, .wd_timeout = IWL_LONG_WD_TIMEOUT, .max_event_log_size = 512, + .no_idle_support = true, }; static struct iwl_ht_params iwl5000_ht_params = { .ht_greenfield_support = true, diff --git a/drivers/net/wireless/iwlwifi/iwl-core.h b/drivers/net/wireless/iwlwifi/iwl-core.h index 3e6bb734dcb7..02817a438550 100644 --- a/drivers/net/wireless/iwlwifi/iwl-core.h +++ b/drivers/net/wireless/iwlwifi/iwl-core.h @@ -135,6 +135,7 @@ struct iwl_mod_params { * @temperature_kelvin: temperature report by uCode in kelvin * @max_event_log_size: size of event log buffer size for ucode event logging * @shadow_reg_enable: HW shadhow register bit + * @no_idle_support: do not support idle mode */ struct iwl_base_params { int eeprom_size; @@ -156,6 +157,7 @@ struct iwl_base_params { bool temperature_kelvin; u32 max_event_log_size; const bool shadow_reg_enable; + const bool no_idle_support; }; /* * @advanced_bt_coexist: support advanced bt coexist diff --git a/drivers/net/wireless/iwlwifi/iwl-power.c b/drivers/net/wireless/iwlwifi/iwl-power.c index 3ec619c6881c..cd64df05f9ed 100644 --- a/drivers/net/wireless/iwlwifi/iwl-power.c +++ b/drivers/net/wireless/iwlwifi/iwl-power.c @@ -349,7 +349,8 @@ static void iwl_power_build_cmd(struct iwl_priv *priv, if (priv->wowlan) iwl_static_sleep_cmd(priv, cmd, IWL_POWER_INDEX_5, dtimper); - else if (priv->hw->conf.flags & IEEE80211_CONF_IDLE) + else if (!priv->cfg->base_params->no_idle_support && + priv->hw->conf.flags & IEEE80211_CONF_IDLE) iwl_static_sleep_cmd(priv, cmd, IWL_POWER_INDEX_5, 20); else if (iwl_tt_is_low_power_state(priv)) { /* in thermal throttling low power state */ From ab92402af04c151c26541eb28e1c26286a429db5 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Thu, 28 Jul 2011 13:48:40 -0700 Subject: [PATCH 205/439] thermal: hide CONFIG_THERMAL_HWMON It's about time to revert 16d752397301b9 ("thermal: Create CONFIG_THERMAL_HWMON=n"). Anybody running a kernel >= 2.6.40 would also be running a recent enough version of lm-sensors. Actually having CONFIG_THERMAL_HWMON is pretty convenient so instead of dropping it, we keep it but hide it. Signed-off-by: Jean Delvare Cc: Rene Herman Acked-by: Guenter Roeck Signed-off-by: Andrew Morton Signed-off-by: Len Brown --- Documentation/feature-removal-schedule.txt | 9 --------- drivers/thermal/Kconfig | 8 ++------ 2 files changed, 2 insertions(+), 15 deletions(-) diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index b1c921c27519..951143dd0684 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -310,15 +310,6 @@ Who: Ravikiran Thirumalai --------------------------- -What: CONFIG_THERMAL_HWMON -When: January 2009 -Why: This option was introduced just to allow older lm-sensors userspace - to keep working over the upgrade to 2.6.26. At the scheduled time of - removal fixed lm-sensors (2.x or 3.x) should be readily available. -Who: Rene Herman - ---------------------------- - What: Code that is now under CONFIG_WIRELESS_EXT_SYSFS (in net/core/net-sysfs.c) When: After the only user (hal) has seen a release with the patches diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig index bf7c687519ef..f7f71b2d3101 100644 --- a/drivers/thermal/Kconfig +++ b/drivers/thermal/Kconfig @@ -14,11 +14,7 @@ menuconfig THERMAL If you want this support, you should say Y or M here. config THERMAL_HWMON - bool "Hardware monitoring support" + bool depends on THERMAL depends on HWMON=y || HWMON=THERMAL - help - The generic thermal sysfs driver's hardware monitoring support - requires a 2.10.7/3.0.2 or later lm-sensors userspace. - - Say Y if your user-space is new enough. + default y From 0d97d7a494d43be77f57e688369be0aae33d1ade Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Thu, 28 Jul 2011 13:48:41 -0700 Subject: [PATCH 206/439] thermal: split hwmon lookup to a separate function We'll soon need to reuse it. Signed-off-by: Jean Delvare Cc: Rene Herman Acked-by: Guenter Roeck Signed-off-by: Andrew Morton Signed-off-by: Len Brown --- drivers/thermal/thermal_sys.c | 29 +++++++++++++++++++++-------- 1 file changed, 21 insertions(+), 8 deletions(-) diff --git a/drivers/thermal/thermal_sys.c b/drivers/thermal/thermal_sys.c index 0b1c82ad6805..b6353d61fac5 100644 --- a/drivers/thermal/thermal_sys.c +++ b/drivers/thermal/thermal_sys.c @@ -469,6 +469,22 @@ temp_crit_show(struct device *dev, struct device_attribute *attr, } +static struct thermal_hwmon_device * +thermal_hwmon_lookup_by_type(const struct thermal_zone_device *tz) +{ + struct thermal_hwmon_device *hwmon; + + mutex_lock(&thermal_list_lock); + list_for_each_entry(hwmon, &thermal_hwmon_list, node) + if (!strcmp(hwmon->type, tz->type)) { + mutex_unlock(&thermal_list_lock); + return hwmon; + } + mutex_unlock(&thermal_list_lock); + + return NULL; +} + static int thermal_add_hwmon_sysfs(struct thermal_zone_device *tz) { @@ -476,14 +492,11 @@ thermal_add_hwmon_sysfs(struct thermal_zone_device *tz) int new_hwmon_device = 1; int result; - mutex_lock(&thermal_list_lock); - list_for_each_entry(hwmon, &thermal_hwmon_list, node) - if (!strcmp(hwmon->type, tz->type)) { - new_hwmon_device = 0; - mutex_unlock(&thermal_list_lock); - goto register_sys_interface; - } - mutex_unlock(&thermal_list_lock); + hwmon = thermal_hwmon_lookup_by_type(tz); + if (hwmon) { + new_hwmon_device = 0; + goto register_sys_interface; + } hwmon = kzalloc(sizeof(struct thermal_hwmon_device), GFP_KERNEL); if (!hwmon) From 31f5396ad3bde23c8416e8d23ba425e27f413314 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Thu, 28 Jul 2011 13:48:42 -0700 Subject: [PATCH 207/439] thermal: make THERMAL_HWMON implementation fully internal THERMAL_HWMON is implemented inside the thermal_sys driver and has no effect on drivers implementing thermal zones, so they shouldn't see anything related to it in . Making the THERMAL_HWMON implementation fully internal has two advantages beyond the cleaner design: * This avoids rebuilding all thermal drivers if the THERMAL_HWMON implementation changes, or if CONFIG_THERMAL_HWMON gets enabled or disabled. * This avoids breaking the thermal kABI in these cases too, which should make distributions happy. The only drawback I can see is slightly higher memory fragmentation, as the number of kzalloc() calls will increase by one per thermal zone. But I doubt it will be a problem in practice, as I've never seen a system with more than two thermal zones. Signed-off-by: Jean Delvare Cc: Rene Herman Acked-by: Guenter Roeck Signed-off-by: Andrew Morton Signed-off-by: Len Brown --- drivers/thermal/thermal_sys.c | 117 ++++++++++++++++++++++++++-------- include/linux/thermal.h | 22 ------- 2 files changed, 92 insertions(+), 47 deletions(-) diff --git a/drivers/thermal/thermal_sys.c b/drivers/thermal/thermal_sys.c index b6353d61fac5..708f8e92771a 100644 --- a/drivers/thermal/thermal_sys.c +++ b/drivers/thermal/thermal_sys.c @@ -420,6 +420,29 @@ thermal_cooling_device_trip_point_show(struct device *dev, /* hwmon sys I/F */ #include + +/* thermal zone devices with the same type share one hwmon device */ +struct thermal_hwmon_device { + char type[THERMAL_NAME_LENGTH]; + struct device *device; + int count; + struct list_head tz_list; + struct list_head node; +}; + +struct thermal_hwmon_attr { + struct device_attribute attr; + char name[16]; +}; + +/* one temperature input for each thermal zone */ +struct thermal_hwmon_temp { + struct list_head hwmon_node; + struct thermal_zone_device *tz; + struct thermal_hwmon_attr temp_input; /* hwmon sys attr */ + struct thermal_hwmon_attr temp_crit; /* hwmon sys attr */ +}; + static LIST_HEAD(thermal_hwmon_list); static ssize_t @@ -437,9 +460,10 @@ temp_input_show(struct device *dev, struct device_attribute *attr, char *buf) int ret; struct thermal_hwmon_attr *hwmon_attr = container_of(attr, struct thermal_hwmon_attr, attr); - struct thermal_zone_device *tz - = container_of(hwmon_attr, struct thermal_zone_device, + struct thermal_hwmon_temp *temp + = container_of(hwmon_attr, struct thermal_hwmon_temp, temp_input); + struct thermal_zone_device *tz = temp->tz; ret = tz->ops->get_temp(tz, &temperature); @@ -455,9 +479,10 @@ temp_crit_show(struct device *dev, struct device_attribute *attr, { struct thermal_hwmon_attr *hwmon_attr = container_of(attr, struct thermal_hwmon_attr, attr); - struct thermal_zone_device *tz - = container_of(hwmon_attr, struct thermal_zone_device, + struct thermal_hwmon_temp *temp + = container_of(hwmon_attr, struct thermal_hwmon_temp, temp_crit); + struct thermal_zone_device *tz = temp->tz; long temperature; int ret; @@ -485,10 +510,29 @@ thermal_hwmon_lookup_by_type(const struct thermal_zone_device *tz) return NULL; } +/* Find the temperature input matching a given thermal zone */ +static struct thermal_hwmon_temp * +thermal_hwmon_lookup_temp(const struct thermal_hwmon_device *hwmon, + const struct thermal_zone_device *tz) +{ + struct thermal_hwmon_temp *temp; + + mutex_lock(&thermal_list_lock); + list_for_each_entry(temp, &hwmon->tz_list, hwmon_node) + if (temp->tz == tz) { + mutex_unlock(&thermal_list_lock); + return temp; + } + mutex_unlock(&thermal_list_lock); + + return NULL; +} + static int thermal_add_hwmon_sysfs(struct thermal_zone_device *tz) { struct thermal_hwmon_device *hwmon; + struct thermal_hwmon_temp *temp; int new_hwmon_device = 1; int result; @@ -515,30 +559,36 @@ thermal_add_hwmon_sysfs(struct thermal_zone_device *tz) goto free_mem; register_sys_interface: - tz->hwmon = hwmon; + temp = kzalloc(sizeof(struct thermal_hwmon_temp), GFP_KERNEL); + if (!temp) { + result = -ENOMEM; + goto unregister_name; + } + + temp->tz = tz; hwmon->count++; - snprintf(tz->temp_input.name, THERMAL_NAME_LENGTH, + snprintf(temp->temp_input.name, THERMAL_NAME_LENGTH, "temp%d_input", hwmon->count); - tz->temp_input.attr.attr.name = tz->temp_input.name; - tz->temp_input.attr.attr.mode = 0444; - tz->temp_input.attr.show = temp_input_show; - sysfs_attr_init(&tz->temp_input.attr.attr); - result = device_create_file(hwmon->device, &tz->temp_input.attr); + temp->temp_input.attr.attr.name = temp->temp_input.name; + temp->temp_input.attr.attr.mode = 0444; + temp->temp_input.attr.show = temp_input_show; + sysfs_attr_init(&temp->temp_input.attr.attr); + result = device_create_file(hwmon->device, &temp->temp_input.attr); if (result) - goto unregister_name; + goto free_temp_mem; if (tz->ops->get_crit_temp) { unsigned long temperature; if (!tz->ops->get_crit_temp(tz, &temperature)) { - snprintf(tz->temp_crit.name, THERMAL_NAME_LENGTH, + snprintf(temp->temp_crit.name, THERMAL_NAME_LENGTH, "temp%d_crit", hwmon->count); - tz->temp_crit.attr.attr.name = tz->temp_crit.name; - tz->temp_crit.attr.attr.mode = 0444; - tz->temp_crit.attr.show = temp_crit_show; - sysfs_attr_init(&tz->temp_crit.attr.attr); + temp->temp_crit.attr.attr.name = temp->temp_crit.name; + temp->temp_crit.attr.attr.mode = 0444; + temp->temp_crit.attr.show = temp_crit_show; + sysfs_attr_init(&temp->temp_crit.attr.attr); result = device_create_file(hwmon->device, - &tz->temp_crit.attr); + &temp->temp_crit.attr); if (result) goto unregister_input; } @@ -547,13 +597,15 @@ thermal_add_hwmon_sysfs(struct thermal_zone_device *tz) mutex_lock(&thermal_list_lock); if (new_hwmon_device) list_add_tail(&hwmon->node, &thermal_hwmon_list); - list_add_tail(&tz->hwmon_node, &hwmon->tz_list); + list_add_tail(&temp->hwmon_node, &hwmon->tz_list); mutex_unlock(&thermal_list_lock); return 0; unregister_input: - device_remove_file(hwmon->device, &tz->temp_input.attr); + device_remove_file(hwmon->device, &temp->temp_input.attr); + free_temp_mem: + kfree(temp); unregister_name: if (new_hwmon_device) { device_remove_file(hwmon->device, &dev_attr_name); @@ -569,15 +621,30 @@ thermal_add_hwmon_sysfs(struct thermal_zone_device *tz) static void thermal_remove_hwmon_sysfs(struct thermal_zone_device *tz) { - struct thermal_hwmon_device *hwmon = tz->hwmon; + struct thermal_hwmon_device *hwmon; + struct thermal_hwmon_temp *temp; - tz->hwmon = NULL; - device_remove_file(hwmon->device, &tz->temp_input.attr); + hwmon = thermal_hwmon_lookup_by_type(tz); + if (unlikely(!hwmon)) { + /* Should never happen... */ + dev_dbg(&tz->device, "hwmon device lookup failed!\n"); + return; + } + + temp = thermal_hwmon_lookup_temp(hwmon, tz); + if (unlikely(!temp)) { + /* Should never happen... */ + dev_dbg(&tz->device, "temperature input lookup failed!\n"); + return; + } + + device_remove_file(hwmon->device, &temp->temp_input.attr); if (tz->ops->get_crit_temp) - device_remove_file(hwmon->device, &tz->temp_crit.attr); + device_remove_file(hwmon->device, &temp->temp_crit.attr); mutex_lock(&thermal_list_lock); - list_del(&tz->hwmon_node); + list_del(&temp->hwmon_node); + kfree(temp); if (!list_empty(&hwmon->tz_list)) { mutex_unlock(&thermal_list_lock); return; diff --git a/include/linux/thermal.h b/include/linux/thermal.h index d3ec89fb4122..47b4a27e6e97 100644 --- a/include/linux/thermal.h +++ b/include/linux/thermal.h @@ -85,22 +85,6 @@ struct thermal_cooling_device { ((long)t-2732+5)/10 : ((long)t-2732-5)/10) #define CELSIUS_TO_KELVIN(t) ((t)*10+2732) -#if defined(CONFIG_THERMAL_HWMON) -/* thermal zone devices with the same type share one hwmon device */ -struct thermal_hwmon_device { - char type[THERMAL_NAME_LENGTH]; - struct device *device; - int count; - struct list_head tz_list; - struct list_head node; -}; - -struct thermal_hwmon_attr { - struct device_attribute attr; - char name[16]; -}; -#endif - struct thermal_zone_device { int id; char type[THERMAL_NAME_LENGTH]; @@ -120,12 +104,6 @@ struct thermal_zone_device { struct mutex lock; /* protect cooling devices list */ struct list_head node; struct delayed_work poll_queue; -#if defined(CONFIG_THERMAL_HWMON) - struct list_head hwmon_node; - struct thermal_hwmon_device *hwmon; - struct thermal_hwmon_attr temp_input; /* hwmon sys attr */ - struct thermal_hwmon_attr temp_crit; /* hwmon sys attr */ -#endif }; /* Adding event notification support elements */ #define THERMAL_GENL_FAMILY_NAME "thermal_event" From f52e00c668669c9c290e84adf859c76db6d92a5a Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Thu, 28 Jul 2011 13:48:43 -0700 Subject: [PATCH 208/439] ACPI: remove NID_INVAL b552a8c56db8 ("ACPI: remove NID_INVAL") removed the left over uses of NID_INVAL, but didn't actually remove the definition. Remove it. Signed-off-by: David Rientjes Signed-off-by: Andrew Morton Signed-off-by: Len Brown --- include/linux/acpi.h | 1 - 1 file changed, 1 deletion(-) diff --git a/include/linux/acpi.h b/include/linux/acpi.h index 1deb2a73c2da..2312e850aab8 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -238,7 +238,6 @@ extern int acpi_paddr_to_node(u64 start_addr, u64 size); extern int pnpacpi_disabled; #define PXM_INVAL (-1) -#define NID_INVAL (-1) int acpi_check_resource_conflict(const struct resource *res); From aa165971c2923d05988f920c978e438dbc7b0de6 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Thu, 28 Jul 2011 13:48:43 -0700 Subject: [PATCH 209/439] ACPI: add missing _OSI strings Linux supports some optional features, but it should notify the BIOS about them via the _OSI method. Currently Linux doesn't notify any, which might make such features not work because the BIOS doesn't know about them. Jarosz has a system which needs this to make ACPI processor aggregator device work. Reported-by: "Jarosz, Sebastian" Signed-off-by: Shaohua Li Acked-by: Matthew Garrett Signed-off-by: Andrew Morton Signed-off-by: Len Brown --- drivers/acpi/osl.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c index 372f9b70f7f4..0e4785703d45 100644 --- a/drivers/acpi/osl.c +++ b/drivers/acpi/osl.c @@ -1083,7 +1083,13 @@ struct osi_setup_entry { bool enable; }; -static struct osi_setup_entry __initdata osi_setup_entries[OSI_STRING_ENTRIES_MAX]; +static struct osi_setup_entry __initdata + osi_setup_entries[OSI_STRING_ENTRIES_MAX] = { + {"Module Device", true}, + {"Processor Device", true}, + {"3.0 _SCP Extensions", true}, + {"Processor Aggregator Device", true}, +}; void __init acpi_osi_setup(char *str) { From d1afa65ca59d4e6a5f1a8c1ab9bfa73f2fa8b777 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Tue, 2 Aug 2011 12:35:04 +0200 Subject: [PATCH 210/439] arch/tile/mm/init.c: trivial: use BUG_ON Use BUG_ON(x) rather than if(x) BUG(); The semantic patch that fixes this problem is as follows: (http://coccinelle.lip6.fr/) // @@ identifier x; @@ -if (x) BUG(); +BUG_ON(x); @@ identifier x; @@ -if (!x) BUG(); +BUG_ON(!x); // Signed-off-by: Julia Lawall Signed-off-by: Chris Metcalf --- arch/tile/mm/init.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/arch/tile/mm/init.c b/arch/tile/mm/init.c index 4e10c4023028..7309988c9794 100644 --- a/arch/tile/mm/init.c +++ b/arch/tile/mm/init.c @@ -836,8 +836,7 @@ void __init mem_init(void) #endif #ifdef CONFIG_FLATMEM - if (!mem_map) - BUG(); + BUG_ON(!mem_map); #endif #ifdef CONFIG_HIGHMEM From 6c8111e9a0e73ef1e58a1bf0a10c23ee1512e7a2 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Tue, 2 Aug 2011 17:15:33 -0400 Subject: [PATCH 211/439] ACPI: delete stale reference in kernel-parameters.txt Says for acpi= See also Documentation/power/pm.txt, pci=noacpi but this file does not exist Reported-by: Alan Cox Signed-off-by: Len Brown --- Documentation/kernel-parameters.txt | 2 -- 1 file changed, 2 deletions(-) diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index aa47be71df4c..53800c2712e6 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -161,8 +161,6 @@ bytes respectively. Such letter suffixes can also be entirely omitted. rsdt -- prefer RSDT over (default) XSDT copy_dsdt -- copy DSDT to memory - See also Documentation/power/pm.txt, pci=noacpi - acpi_apic_instance= [ACPI, IOAPIC] Format: 2: use 2nd APIC table, if available From b728a5c806fb36f9adebf2a862bbd015e074afca Mon Sep 17 00:00:00 2001 From: Tony Luck Date: Tue, 2 Aug 2011 15:08:30 -0700 Subject: [PATCH 212/439] efivars: fix warnings when CONFIG_PSTORE=n MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit drivers/firmware/efivars.c:161: warning: ‘utf16_strlen’ defined but not used utf16_strlen() is only used inside CONFIG_PSTORE - make this "static inline" to shut the compiler up [thanks to hpa for the suggestion]. drivers/firmware/efivars.c:602: warning: initialization from incompatible pointer type Between v1 and v2 of this patch series we decided to make the "part" number unsigned - but missed fixing the stub version of efi_pstore_write() Acked-by: Matthew Garrett Acked-by: Mike Waychison Signed-off-by: Tony Luck --- drivers/firmware/efivars.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/firmware/efivars.c b/drivers/firmware/efivars.c index eacb05e6cfb3..eb80b549ed8d 100644 --- a/drivers/firmware/efivars.c +++ b/drivers/firmware/efivars.c @@ -157,7 +157,7 @@ utf16_strnlen(efi_char16_t *s, size_t maxlength) return length; } -static unsigned long +static inline unsigned long utf16_strlen(efi_char16_t *s) { return utf16_strnlen(s, ~0UL); @@ -580,8 +580,8 @@ static ssize_t efi_pstore_read(u64 *id, enum pstore_type_id *type, return -1; } -static u64 efi_pstore_write(enum pstore_type_id type, int part, size_t size, - struct pstore_info *psi) +static u64 efi_pstore_write(enum pstore_type_id type, unsigned int part, + size_t size, struct pstore_info *psi) { return 0; } From d30c4b7a87e8b19583d5ef1402d9b38f51e30f44 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Sun, 31 Jul 2011 18:19:33 -0400 Subject: [PATCH 213/439] tools/power turbostat: fit output into 80 columns on snb-ep Reduce columns for package number to 1. If you can afford more than 9 packages, you can also afford a terminal with more than 80 columns:-) Also shave a column also off the package C-states Signed-off-by: Len Brown --- tools/power/x86/turbostat/turbostat.c | 46 +++++++++++++-------------- 1 file changed, 23 insertions(+), 23 deletions(-) diff --git a/tools/power/x86/turbostat/turbostat.c b/tools/power/x86/turbostat/turbostat.c index 6d8ef4a3a9b5..8b2d37b59c9e 100644 --- a/tools/power/x86/turbostat/turbostat.c +++ b/tools/power/x86/turbostat/turbostat.c @@ -128,34 +128,34 @@ unsigned long long get_msr(int cpu, off_t offset) void print_header(void) { if (show_pkg) - fprintf(stderr, "pkg "); + fprintf(stderr, "pk"); if (show_core) - fprintf(stderr, "core"); + fprintf(stderr, " cr"); if (show_cpu) fprintf(stderr, " CPU"); if (do_nhm_cstates) - fprintf(stderr, " %%c0 "); + fprintf(stderr, " %%c0 "); if (has_aperf) - fprintf(stderr, " GHz"); + fprintf(stderr, " GHz"); fprintf(stderr, " TSC"); if (do_nhm_cstates) - fprintf(stderr, " %%c1 "); + fprintf(stderr, " %%c1"); if (do_nhm_cstates) - fprintf(stderr, " %%c3 "); + fprintf(stderr, " %%c3"); if (do_nhm_cstates) - fprintf(stderr, " %%c6 "); + fprintf(stderr, " %%c6"); if (do_snb_cstates) - fprintf(stderr, " %%c7 "); + fprintf(stderr, " %%c7"); if (do_snb_cstates) - fprintf(stderr, " %%pc2 "); + fprintf(stderr, " %%pc2"); if (do_nhm_cstates) - fprintf(stderr, " %%pc3 "); + fprintf(stderr, " %%pc3"); if (do_nhm_cstates) - fprintf(stderr, " %%pc6 "); + fprintf(stderr, " %%pc6"); if (do_snb_cstates) - fprintf(stderr, " %%pc7 "); + fprintf(stderr, " %%pc7"); if (extra_msr_offset) - fprintf(stderr, " MSR 0x%x ", extra_msr_offset); + fprintf(stderr, " MSR 0x%x ", extra_msr_offset); putc('\n', stderr); } @@ -194,14 +194,14 @@ void print_cnt(struct counters *p) /* topology columns, print blanks on 1st (average) line */ if (p == cnt_average) { if (show_pkg) - fprintf(stderr, " "); + fprintf(stderr, " "); if (show_core) fprintf(stderr, " "); if (show_cpu) fprintf(stderr, " "); } else { if (show_pkg) - fprintf(stderr, "%4d", p->pkg); + fprintf(stderr, "%d", p->pkg); if (show_core) fprintf(stderr, "%4d", p->core); if (show_cpu) @@ -241,22 +241,22 @@ void print_cnt(struct counters *p) if (!skip_c1) fprintf(stderr, "%7.2f", 100.0 * p->c1/p->tsc); else - fprintf(stderr, " ****"); + fprintf(stderr, " ****"); } if (do_nhm_cstates) - fprintf(stderr, "%7.2f", 100.0 * p->c3/p->tsc); + fprintf(stderr, " %6.2f", 100.0 * p->c3/p->tsc); if (do_nhm_cstates) - fprintf(stderr, "%7.2f", 100.0 * p->c6/p->tsc); + fprintf(stderr, " %6.2f", 100.0 * p->c6/p->tsc); if (do_snb_cstates) - fprintf(stderr, "%7.2f", 100.0 * p->c7/p->tsc); + fprintf(stderr, " %6.2f", 100.0 * p->c7/p->tsc); if (do_snb_cstates) - fprintf(stderr, "%7.2f", 100.0 * p->pc2/p->tsc); + fprintf(stderr, " %5.2f", 100.0 * p->pc2/p->tsc); if (do_nhm_cstates) - fprintf(stderr, "%7.2f", 100.0 * p->pc3/p->tsc); + fprintf(stderr, " %5.2f", 100.0 * p->pc3/p->tsc); if (do_nhm_cstates) - fprintf(stderr, "%7.2f", 100.0 * p->pc6/p->tsc); + fprintf(stderr, " %5.2f", 100.0 * p->pc6/p->tsc); if (do_snb_cstates) - fprintf(stderr, "%7.2f", 100.0 * p->pc7/p->tsc); + fprintf(stderr, " %5.2f", 100.0 * p->pc7/p->tsc); if (extra_msr_offset) fprintf(stderr, " 0x%016llx", p->extra_msr); putc('\n', stderr); From c28aa38567101bad4e020f4392df41d0bf6c165c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?fran=C3=A7ois=20romieu?= Date: Tue, 2 Aug 2011 03:53:43 +0000 Subject: [PATCH 214/439] r8169 : MAC address change fix for the 8168e-vl. Addresses https://bugzilla.kernel.org/show_bug.cgi?id=39252 Hayes suggested that the usual MAC{0, 4} register writes be completed with writes to extended GigaMAC registers : - 0xe0 .. 0xe5 - 0xf2 .. 0xf7 Registers 0xf0 and 0xf1 should be set to 0. Signed-off-by: Francois Romieu Cc: Hayes Wang Signed-off-by: David S. Miller --- drivers/net/r8169.c | 27 +++++++++++++++++++++++++++ 1 file changed, 27 insertions(+) diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c index c77286edba4d..02339b3352e7 100644 --- a/drivers/net/r8169.c +++ b/drivers/net/r8169.c @@ -1092,6 +1092,21 @@ rtl_w1w0_eri(void __iomem *ioaddr, int addr, u32 mask, u32 p, u32 m, int type) rtl_eri_write(ioaddr, addr, mask, (val & ~m) | p, type); } +struct exgmac_reg { + u16 addr; + u16 mask; + u32 val; +}; + +static void rtl_write_exgmac_batch(void __iomem *ioaddr, + const struct exgmac_reg *r, int len) +{ + while (len-- > 0) { + rtl_eri_write(ioaddr, r->addr, r->mask, r->val, ERIAR_EXGMAC); + r++; + } +} + static u8 rtl8168d_efuse_read(void __iomem *ioaddr, int reg_addr) { u8 value = 0xff; @@ -3117,6 +3132,18 @@ static void rtl_rar_set(struct rtl8169_private *tp, u8 *addr) RTL_W32(MAC0, low); RTL_R32(MAC0); + if (tp->mac_version == RTL_GIGA_MAC_VER_34) { + const struct exgmac_reg e[] = { + { .addr = 0xe0, ERIAR_MASK_1111, .val = low }, + { .addr = 0xe4, ERIAR_MASK_1111, .val = high }, + { .addr = 0xf0, ERIAR_MASK_1111, .val = low << 16 }, + { .addr = 0xf4, ERIAR_MASK_1111, .val = high << 16 | + low >> 16 }, + }; + + rtl_write_exgmac_batch(ioaddr, e, ARRAY_SIZE(e)); + } + RTL_W8(Cfg9346, Cfg9346_Lock); spin_unlock_irq(&tp->lock); From fc05a78efb8e91e884017bb0bc43f690aa5b4dcd Mon Sep 17 00:00:00 2001 From: Steve French Date: Wed, 3 Aug 2011 03:17:43 +0000 Subject: [PATCH 215/439] Revert "cifs: advertise the right receive buffer size to the server" This reverts commit c4d3396b261473ded6f370edd1e79ba34e089d7e. Problems discovered with readdir to Samba due to not accounting for header size properly with this change --- fs/cifs/sess.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index 243d58720513..d3e619692ee0 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -124,8 +124,7 @@ static __u32 cifs_ssetup_hdr(struct cifs_ses *ses, SESSION_SETUP_ANDX *pSMB) /* that we use in next few lines */ /* Note that header is initialized to zero in header_assemble */ pSMB->req.AndXCommand = 0xFF; - pSMB->req.MaxBufferSize = cpu_to_le16(min_t(u32, CIFSMaxBufSize - 4, - USHRT_MAX)); + pSMB->req.MaxBufferSize = cpu_to_le16(ses->server->maxBuf); pSMB->req.MaxMpxCount = cpu_to_le16(ses->server->maxReq); pSMB->req.VcNumber = get_next_vcnum(ses); From b80289833463215d2f3d1d72cf735fc7ba78da57 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Tue, 26 Jul 2011 12:20:18 -0400 Subject: [PATCH 216/439] cifs: demote DFS referral lookup errors to cFYI cifs: demote DFS referral lookup errors to cFYI Now that we call into this routine on every mount, anyone who doesn't have the upcall configured will get multiple printks about failed lookups. Reported-and-Tested-by: Martijn Uffing Signed-off-by: Jeff Layton Signed-off-by: Steve French --- fs/cifs/cifs_dfs_ref.c | 5 +++-- fs/cifs/dns_resolve.c | 4 ++-- 2 files changed, 5 insertions(+), 4 deletions(-) diff --git a/fs/cifs/cifs_dfs_ref.c b/fs/cifs/cifs_dfs_ref.c index 8d8f28c94c0f..6873bb634a97 100644 --- a/fs/cifs/cifs_dfs_ref.c +++ b/fs/cifs/cifs_dfs_ref.c @@ -141,10 +141,11 @@ char *cifs_compose_mount_options(const char *sb_mountdata, rc = dns_resolve_server_name_to_ip(*devname, &srvIP); if (rc < 0) { - cERROR(1, "%s: Failed to resolve server part of %s to IP: %d", - __func__, *devname, rc); + cFYI(1, "%s: Failed to resolve server part of %s to IP: %d", + __func__, *devname, rc); goto compose_mount_options_err; } + /* md_len = strlen(...) + 12 for 'sep+prefixpath=' * assuming that we have 'unc=' and 'ip=' in * the original sb_mountdata diff --git a/fs/cifs/dns_resolve.c b/fs/cifs/dns_resolve.c index 548f06230a6d..1d2d91d9bf65 100644 --- a/fs/cifs/dns_resolve.c +++ b/fs/cifs/dns_resolve.c @@ -79,8 +79,8 @@ dns_resolve_server_name_to_ip(const char *unc, char **ip_addr) /* Perform the upcall */ rc = dns_query(NULL, hostname, len, NULL, ip_addr, NULL); if (rc < 0) - cERROR(1, "%s: unable to resolve: %*.*s", - __func__, len, len, hostname); + cFYI(1, "%s: unable to resolve: %*.*s", + __func__, len, len, hostname); else cFYI(1, "%s: resolved: %*.*s to %s", __func__, len, len, hostname, *ip_addr); From e95ade083939dcb4b0c51c1a2c8504ea9ef3d6ef Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 1 Aug 2011 18:18:57 -0700 Subject: [PATCH 217/439] sparc: Minor tweaks to Niagara page copy/clear. Don't use floating point on Niagara2, use the traditional plain Niagara code instead. Unroll Niagara loops to 128 bytes for copy, and 256 bytes for clear. Signed-off-by: David S. Miller --- arch/sparc/kernel/head_64.S | 2 +- arch/sparc/lib/Makefile | 2 +- arch/sparc/lib/NG2page.S | 61 ------------------- arch/sparc/lib/NGpage.S | 114 ++++++++++++++++++++++++------------ 4 files changed, 77 insertions(+), 102 deletions(-) delete mode 100644 arch/sparc/lib/NG2page.S diff --git a/arch/sparc/kernel/head_64.S b/arch/sparc/kernel/head_64.S index c752603a7c0d..0eac1b2fc53d 100644 --- a/arch/sparc/kernel/head_64.S +++ b/arch/sparc/kernel/head_64.S @@ -559,7 +559,7 @@ niagara2_patch: nop call niagara_patch_bzero nop - call niagara2_patch_pageops + call niagara_patch_pageops nop ba,a,pt %xcc, 80f diff --git a/arch/sparc/lib/Makefile b/arch/sparc/lib/Makefile index 7f01b8fce8bc..c25f94d28df8 100644 --- a/arch/sparc/lib/Makefile +++ b/arch/sparc/lib/Makefile @@ -31,7 +31,7 @@ lib-$(CONFIG_SPARC64) += NGmemcpy.o NGcopy_from_user.o NGcopy_to_user.o lib-$(CONFIG_SPARC64) += NGpatch.o NGpage.o NGbzero.o lib-$(CONFIG_SPARC64) += NG2memcpy.o NG2copy_from_user.o NG2copy_to_user.o -lib-$(CONFIG_SPARC64) += NG2patch.o NG2page.o +lib-$(CONFIG_SPARC64) += NG2patch.o lib-$(CONFIG_SPARC64) += GENmemcpy.o GENcopy_from_user.o GENcopy_to_user.o lib-$(CONFIG_SPARC64) += GENpatch.o GENpage.o GENbzero.o diff --git a/arch/sparc/lib/NG2page.S b/arch/sparc/lib/NG2page.S deleted file mode 100644 index 73b6b7c72cbf..000000000000 --- a/arch/sparc/lib/NG2page.S +++ /dev/null @@ -1,61 +0,0 @@ -/* NG2page.S: Niagara-2 optimized clear and copy page. - * - * Copyright (C) 2007 (davem@davemloft.net) - */ - -#include -#include -#include - - .text - .align 32 - - /* This is heavily simplified from the sun4u variants - * because Niagara-2 does not have any D-cache aliasing issues. - */ -NG2copy_user_page: /* %o0=dest, %o1=src, %o2=vaddr */ - prefetch [%o1 + 0x00], #one_read - prefetch [%o1 + 0x40], #one_read - VISEntryHalf - set PAGE_SIZE, %g7 - sub %o0, %o1, %g3 -1: stxa %g0, [%o1 + %g3] ASI_BLK_INIT_QUAD_LDD_P - subcc %g7, 64, %g7 - ldda [%o1] ASI_BLK_P, %f0 - stda %f0, [%o1 + %g3] ASI_BLK_P - add %o1, 64, %o1 - bne,pt %xcc, 1b - prefetch [%o1 + 0x40], #one_read - membar #Sync - VISExitHalf - retl - nop - -#define BRANCH_ALWAYS 0x10680000 -#define NOP 0x01000000 -#define NG_DO_PATCH(OLD, NEW) \ - sethi %hi(NEW), %g1; \ - or %g1, %lo(NEW), %g1; \ - sethi %hi(OLD), %g2; \ - or %g2, %lo(OLD), %g2; \ - sub %g1, %g2, %g1; \ - sethi %hi(BRANCH_ALWAYS), %g3; \ - sll %g1, 11, %g1; \ - srl %g1, 11 + 2, %g1; \ - or %g3, %lo(BRANCH_ALWAYS), %g3; \ - or %g3, %g1, %g3; \ - stw %g3, [%g2]; \ - sethi %hi(NOP), %g3; \ - or %g3, %lo(NOP), %g3; \ - stw %g3, [%g2 + 0x4]; \ - flush %g2; - - .globl niagara2_patch_pageops - .type niagara2_patch_pageops,#function -niagara2_patch_pageops: - NG_DO_PATCH(copy_user_page, NG2copy_user_page) - NG_DO_PATCH(_clear_page, NGclear_page) - NG_DO_PATCH(clear_user_page, NGclear_user_page) - retl - nop - .size niagara2_patch_pageops,.-niagara2_patch_pageops diff --git a/arch/sparc/lib/NGpage.S b/arch/sparc/lib/NGpage.S index 428920de05ba..b9e790b9c6b8 100644 --- a/arch/sparc/lib/NGpage.S +++ b/arch/sparc/lib/NGpage.S @@ -16,55 +16,91 @@ */ NGcopy_user_page: /* %o0=dest, %o1=src, %o2=vaddr */ - prefetch [%o1 + 0x00], #one_read - mov 8, %g1 - mov 16, %g2 - mov 24, %g3 + save %sp, -192, %sp + rd %asi, %g3 + wr %g0, ASI_BLK_INIT_QUAD_LDD_P, %asi set PAGE_SIZE, %g7 + prefetch [%i1 + 0x00], #one_read + prefetch [%i1 + 0x40], #one_read -1: ldda [%o1 + %g0] ASI_BLK_INIT_QUAD_LDD_P, %o2 - ldda [%o1 + %g2] ASI_BLK_INIT_QUAD_LDD_P, %o4 - prefetch [%o1 + 0x40], #one_read - add %o1, 32, %o1 - stxa %o2, [%o0 + %g0] ASI_BLK_INIT_QUAD_LDD_P - stxa %o3, [%o0 + %g1] ASI_BLK_INIT_QUAD_LDD_P - ldda [%o1 + %g0] ASI_BLK_INIT_QUAD_LDD_P, %o2 - stxa %o4, [%o0 + %g2] ASI_BLK_INIT_QUAD_LDD_P - stxa %o5, [%o0 + %g3] ASI_BLK_INIT_QUAD_LDD_P - ldda [%o1 + %g2] ASI_BLK_INIT_QUAD_LDD_P, %o4 - add %o1, 32, %o1 - add %o0, 32, %o0 - stxa %o2, [%o0 + %g0] ASI_BLK_INIT_QUAD_LDD_P - stxa %o3, [%o0 + %g1] ASI_BLK_INIT_QUAD_LDD_P - stxa %o4, [%o0 + %g2] ASI_BLK_INIT_QUAD_LDD_P - stxa %o5, [%o0 + %g3] ASI_BLK_INIT_QUAD_LDD_P - subcc %g7, 64, %g7 +1: prefetch [%i1 + 0x80], #one_read + prefetch [%i1 + 0xc0], #one_read + ldda [%i1 + 0x00] %asi, %o2 + ldda [%i1 + 0x10] %asi, %o4 + ldda [%i1 + 0x20] %asi, %l2 + ldda [%i1 + 0x30] %asi, %l4 + stxa %o2, [%i0 + 0x00] %asi + stxa %o3, [%i0 + 0x08] %asi + stxa %o4, [%i0 + 0x10] %asi + stxa %o5, [%i0 + 0x18] %asi + stxa %l2, [%i0 + 0x20] %asi + stxa %l3, [%i0 + 0x28] %asi + stxa %l4, [%i0 + 0x30] %asi + stxa %l5, [%i0 + 0x38] %asi + ldda [%i1 + 0x40] %asi, %o2 + ldda [%i1 + 0x50] %asi, %o4 + ldda [%i1 + 0x60] %asi, %l2 + ldda [%i1 + 0x70] %asi, %l4 + stxa %o2, [%i0 + 0x40] %asi + stxa %o3, [%i0 + 0x48] %asi + stxa %o4, [%i0 + 0x50] %asi + stxa %o5, [%i0 + 0x58] %asi + stxa %l2, [%i0 + 0x60] %asi + stxa %l3, [%i0 + 0x68] %asi + stxa %l4, [%i0 + 0x70] %asi + stxa %l5, [%i0 + 0x78] %asi + add %i1, 128, %i1 + subcc %g7, 128, %g7 bne,pt %xcc, 1b - add %o0, 32, %o0 + add %i0, 128, %i0 + wr %g3, 0x0, %asi membar #Sync - retl - nop + ret + restore - .globl NGclear_page, NGclear_user_page + .align 32 NGclear_page: /* %o0=dest */ NGclear_user_page: /* %o0=dest, %o1=vaddr */ - mov 8, %g1 - mov 16, %g2 - mov 24, %g3 + rd %asi, %g3 + wr %g0, ASI_BLK_INIT_QUAD_LDD_P, %asi set PAGE_SIZE, %g7 -1: stxa %g0, [%o0 + %g0] ASI_BLK_INIT_QUAD_LDD_P - stxa %g0, [%o0 + %g1] ASI_BLK_INIT_QUAD_LDD_P - stxa %g0, [%o0 + %g2] ASI_BLK_INIT_QUAD_LDD_P - stxa %g0, [%o0 + %g3] ASI_BLK_INIT_QUAD_LDD_P - add %o0, 32, %o0 - stxa %g0, [%o0 + %g0] ASI_BLK_INIT_QUAD_LDD_P - stxa %g0, [%o0 + %g1] ASI_BLK_INIT_QUAD_LDD_P - stxa %g0, [%o0 + %g2] ASI_BLK_INIT_QUAD_LDD_P - stxa %g0, [%o0 + %g3] ASI_BLK_INIT_QUAD_LDD_P - subcc %g7, 64, %g7 +1: stxa %g0, [%o0 + 0x00] %asi + stxa %g0, [%o0 + 0x08] %asi + stxa %g0, [%o0 + 0x10] %asi + stxa %g0, [%o0 + 0x18] %asi + stxa %g0, [%o0 + 0x20] %asi + stxa %g0, [%o0 + 0x28] %asi + stxa %g0, [%o0 + 0x30] %asi + stxa %g0, [%o0 + 0x38] %asi + stxa %g0, [%o0 + 0x40] %asi + stxa %g0, [%o0 + 0x48] %asi + stxa %g0, [%o0 + 0x50] %asi + stxa %g0, [%o0 + 0x58] %asi + stxa %g0, [%o0 + 0x60] %asi + stxa %g0, [%o0 + 0x68] %asi + stxa %g0, [%o0 + 0x70] %asi + stxa %g0, [%o0 + 0x78] %asi + stxa %g0, [%o0 + 0x80] %asi + stxa %g0, [%o0 + 0x88] %asi + stxa %g0, [%o0 + 0x90] %asi + stxa %g0, [%o0 + 0x98] %asi + stxa %g0, [%o0 + 0xa0] %asi + stxa %g0, [%o0 + 0xa8] %asi + stxa %g0, [%o0 + 0xb0] %asi + stxa %g0, [%o0 + 0xb8] %asi + stxa %g0, [%o0 + 0xc0] %asi + stxa %g0, [%o0 + 0xc8] %asi + stxa %g0, [%o0 + 0xd0] %asi + stxa %g0, [%o0 + 0xd8] %asi + stxa %g0, [%o0 + 0xe0] %asi + stxa %g0, [%o0 + 0xe8] %asi + stxa %g0, [%o0 + 0xf0] %asi + stxa %g0, [%o0 + 0xf8] %asi + subcc %g7, 256, %g7 bne,pt %xcc, 1b - add %o0, 32, %o0 + add %o0, 256, %o0 + wr %g3, 0x0, %asi membar #Sync retl nop From ef7c4d4675d2a9206f913f26ca1a5cd41bff9d41 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 29 Jul 2011 09:42:07 -0700 Subject: [PATCH 218/439] sparc: Use popc if possible for hweight routines. Just like powerpc, we code patch at boot time. Signed-off-by: David S. Miller --- arch/sparc/include/asm/bitops_64.h | 42 +++--------------------- arch/sparc/kernel/entry.h | 7 ++++ arch/sparc/kernel/setup_64.c | 27 ++++++++++++++++ arch/sparc/kernel/sparc_ksyms_64.c | 7 ++++ arch/sparc/kernel/vmlinux.lds.S | 6 +++- arch/sparc/lib/Makefile | 2 +- arch/sparc/lib/hweight.S | 51 ++++++++++++++++++++++++++++++ 7 files changed, 102 insertions(+), 40 deletions(-) create mode 100644 arch/sparc/lib/hweight.S diff --git a/arch/sparc/include/asm/bitops_64.h b/arch/sparc/include/asm/bitops_64.h index 325e295d60de..3588807baa6e 100644 --- a/arch/sparc/include/asm/bitops_64.h +++ b/arch/sparc/include/asm/bitops_64.h @@ -42,45 +42,11 @@ extern void change_bit(unsigned long nr, volatile unsigned long *addr); * of bits set) of a N-bit word */ -#ifdef ULTRA_HAS_POPULATION_COUNT +extern unsigned long __arch_hweight64(__u64 w); +extern unsigned int __arch_hweight32(unsigned int w); +extern unsigned int __arch_hweight16(unsigned int w); +extern unsigned int __arch_hweight8(unsigned int w); -static inline unsigned int __arch_hweight64(unsigned long w) -{ - unsigned int res; - - __asm__ ("popc %1,%0" : "=r" (res) : "r" (w)); - return res; -} - -static inline unsigned int __arch_hweight32(unsigned int w) -{ - unsigned int res; - - __asm__ ("popc %1,%0" : "=r" (res) : "r" (w & 0xffffffff)); - return res; -} - -static inline unsigned int __arch_hweight16(unsigned int w) -{ - unsigned int res; - - __asm__ ("popc %1,%0" : "=r" (res) : "r" (w & 0xffff)); - return res; -} - -static inline unsigned int __arch_hweight8(unsigned int w) -{ - unsigned int res; - - __asm__ ("popc %1,%0" : "=r" (res) : "r" (w & 0xff)); - return res; -} - -#else - -#include - -#endif #include #include #endif /* __KERNEL__ */ diff --git a/arch/sparc/kernel/entry.h b/arch/sparc/kernel/entry.h index d1f1361c4167..16d1545d84fc 100644 --- a/arch/sparc/kernel/entry.h +++ b/arch/sparc/kernel/entry.h @@ -42,6 +42,13 @@ extern void fpsave(unsigned long *fpregs, unsigned long *fsr, extern void fpload(unsigned long *fpregs, unsigned long *fsr); #else /* CONFIG_SPARC32 */ +struct popc_3insn_patch_entry { + unsigned int addr; + unsigned int insns[3]; +}; +extern struct popc_3insn_patch_entry __popc_3insn_patch, + __popc_3insn_patch_end; + extern void __init per_cpu_patch(void); extern void __init sun4v_patch(void); extern void __init boot_cpu_id_too_large(int cpu); diff --git a/arch/sparc/kernel/setup_64.c b/arch/sparc/kernel/setup_64.c index 242dbb3f31d2..26d114187e10 100644 --- a/arch/sparc/kernel/setup_64.c +++ b/arch/sparc/kernel/setup_64.c @@ -272,6 +272,30 @@ void __init sun4v_patch(void) sun4v_hvapi_init(); } +static void __init popc_patch(void) +{ + struct popc_3insn_patch_entry *p3; + + p3 = &__popc_3insn_patch; + while (p3 < &__popc_3insn_patch_end) { + unsigned long addr = p3->addr; + + *(unsigned int *) (addr + 0) = p3->insns[0]; + wmb(); + __asm__ __volatile__("flush %0" : : "r" (addr + 0)); + + *(unsigned int *) (addr + 4) = p3->insns[1]; + wmb(); + __asm__ __volatile__("flush %0" : : "r" (addr + 4)); + + *(unsigned int *) (addr + 8) = p3->insns[2]; + wmb(); + __asm__ __volatile__("flush %0" : : "r" (addr + 4)); + + p3++; + } +} + #ifdef CONFIG_SMP void __init boot_cpu_id_too_large(int cpu) { @@ -424,6 +448,9 @@ static void __init init_sparc64_elf_hwcap(void) sparc64_elf_hwcap = cap | mdesc_caps; report_hwcaps(sparc64_elf_hwcap); + + if (sparc64_elf_hwcap & AV_SPARC_POPC) + popc_patch(); } void __init setup_arch(char **cmdline_p) diff --git a/arch/sparc/kernel/sparc_ksyms_64.c b/arch/sparc/kernel/sparc_ksyms_64.c index 372ad59c4cba..d0ee65aced0d 100644 --- a/arch/sparc/kernel/sparc_ksyms_64.c +++ b/arch/sparc/kernel/sparc_ksyms_64.c @@ -8,6 +8,7 @@ #include #include #include +#include #include #include @@ -38,5 +39,11 @@ EXPORT_SYMBOL(sun4v_niagara_setperf); EXPORT_SYMBOL(sun4v_niagara2_getperf); EXPORT_SYMBOL(sun4v_niagara2_setperf); +/* from hweight.S */ +EXPORT_SYMBOL(__arch_hweight8); +EXPORT_SYMBOL(__arch_hweight16); +EXPORT_SYMBOL(__arch_hweight32); +EXPORT_SYMBOL(__arch_hweight64); + /* Exporting a symbol from /init/main.c */ EXPORT_SYMBOL(saved_command_line); diff --git a/arch/sparc/kernel/vmlinux.lds.S b/arch/sparc/kernel/vmlinux.lds.S index c0220759003e..de20c14625eb 100644 --- a/arch/sparc/kernel/vmlinux.lds.S +++ b/arch/sparc/kernel/vmlinux.lds.S @@ -107,7 +107,11 @@ SECTIONS *(.sun4v_2insn_patch) __sun4v_2insn_patch_end = .; } - + .popc_3insn_patch : { + __popc_3insn_patch = .; + *(.popc_3insn_patch) + __popc_3insn_patch_end = .; + } PERCPU_SECTION(SMP_CACHE_BYTES) . = ALIGN(PAGE_SIZE); diff --git a/arch/sparc/lib/Makefile b/arch/sparc/lib/Makefile index c25f94d28df8..11c850a9b66f 100644 --- a/arch/sparc/lib/Makefile +++ b/arch/sparc/lib/Makefile @@ -37,7 +37,7 @@ lib-$(CONFIG_SPARC64) += GENmemcpy.o GENcopy_from_user.o GENcopy_to_user.o lib-$(CONFIG_SPARC64) += GENpatch.o GENpage.o GENbzero.o lib-$(CONFIG_SPARC64) += copy_in_user.o user_fixup.o memmove.o -lib-$(CONFIG_SPARC64) += mcount.o ipcsum.o xor.o +lib-$(CONFIG_SPARC64) += mcount.o ipcsum.o xor.o hweight.o obj-y += iomap.o obj-$(CONFIG_SPARC32) += atomic32.o diff --git a/arch/sparc/lib/hweight.S b/arch/sparc/lib/hweight.S new file mode 100644 index 000000000000..95414e0a6808 --- /dev/null +++ b/arch/sparc/lib/hweight.S @@ -0,0 +1,51 @@ +#include + + .text + .align 32 +ENTRY(__arch_hweight8) + ba,pt %xcc, __sw_hweight8 + nop + nop +ENDPROC(__arch_hweight8) + .section .popc_3insn_patch, "ax" + .word __arch_hweight8 + sllx %o0, 64-8, %g1 + retl + popc %g1, %o0 + .previous + +ENTRY(__arch_hweight16) + ba,pt %xcc, __sw_hweight16 + nop + nop +ENDPROC(__arch_hweight16) + .section .popc_3insn_patch, "ax" + .word __arch_hweight16 + sllx %o0, 64-16, %g1 + retl + popc %g1, %o0 + .previous + +ENTRY(__arch_hweight32) + ba,pt %xcc, __sw_hweight32 + nop + nop +ENDPROC(__arch_hweight32) + .section .popc_3insn_patch, "ax" + .word __arch_hweight32 + sllx %o0, 64-32, %g1 + retl + popc %g1, %o0 + .previous + +ENTRY(__arch_hweight64) + ba,pt %xcc, __sw_hweight64 + nop + nop +ENDPROC(__arch_hweight64) + .section .popc_3insn_patch, "ax" + .word __arch_hweight64 + retl + popc %o0, %o0 + nop + .previous From d600cbed0fe8fceec04500824f638dfe4996c653 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 1 Aug 2011 19:41:12 -0700 Subject: [PATCH 219/439] sparc: Use hweight64() in popc emulation. Signed-off-by: David S. Miller --- arch/sparc/kernel/unaligned_64.c | 15 ++++----------- 1 file changed, 4 insertions(+), 11 deletions(-) diff --git a/arch/sparc/kernel/unaligned_64.c b/arch/sparc/kernel/unaligned_64.c index 35cff1673aa4..76e4ac1a13e1 100644 --- a/arch/sparc/kernel/unaligned_64.c +++ b/arch/sparc/kernel/unaligned_64.c @@ -22,6 +22,7 @@ #include #include #include +#include #include enum direction { @@ -373,16 +374,11 @@ asmlinkage void kernel_unaligned_trap(struct pt_regs *regs, unsigned int insn) } } -static char popc_helper[] = { -0, 1, 1, 2, 1, 2, 2, 3, -1, 2, 2, 3, 2, 3, 3, 4, -}; - int handle_popc(u32 insn, struct pt_regs *regs) { - u64 value; - int ret, i, rd = ((insn >> 25) & 0x1f); int from_kernel = (regs->tstate & TSTATE_PRIV) != 0; + int ret, rd = ((insn >> 25) & 0x1f); + u64 value; perf_sw_event(PERF_COUNT_SW_EMULATION_FAULTS, 1, regs, 0); if (insn & 0x2000) { @@ -392,10 +388,7 @@ int handle_popc(u32 insn, struct pt_regs *regs) maybe_flush_windows(0, insn & 0x1f, rd, from_kernel); value = fetch_reg(insn & 0x1f, regs); } - for (ret = 0, i = 0; i < 16; i++) { - ret += popc_helper[value & 0xf]; - value >>= 4; - } + ret = hweight64(value); if (rd < 16) { if (rd) regs->u_regs[rd] = ret; From e2eb9f8158ead43a88c0f0b4d74257b1be938a18 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 1 Aug 2011 22:45:18 -0700 Subject: [PATCH 220/439] sparc: Add some missing hypervisor API groups. Signed-off-by: David S. Miller --- arch/sparc/include/asm/hypervisor.h | 6 ++++++ arch/sparc/kernel/hvapi.c | 6 ++++++ 2 files changed, 12 insertions(+) diff --git a/arch/sparc/include/asm/hypervisor.h b/arch/sparc/include/asm/hypervisor.h index 7a5f80df15d0..2b7da27ff17d 100644 --- a/arch/sparc/include/asm/hypervisor.h +++ b/arch/sparc/include/asm/hypervisor.h @@ -2940,11 +2940,17 @@ extern unsigned long sun4v_ncs_request(unsigned long request, #define HV_GRP_CORE 0x0001 #define HV_GRP_INTR 0x0002 #define HV_GRP_SOFT_STATE 0x0003 +#define HV_GRP_TM 0x0080 #define HV_GRP_PCI 0x0100 #define HV_GRP_LDOM 0x0101 #define HV_GRP_SVC_CHAN 0x0102 #define HV_GRP_NCS 0x0103 #define HV_GRP_RNG 0x0104 +#define HV_GRP_PBOOT 0x0105 +#define HV_GRP_TPM 0x0107 +#define HV_GRP_SDIO 0x0108 +#define HV_GRP_SDIO_ERR 0x0109 +#define HV_GRP_REBOOT_DATA 0x0110 #define HV_GRP_NIAG_PERF 0x0200 #define HV_GRP_FIRE_PERF 0x0201 #define HV_GRP_N2_CPU 0x0202 diff --git a/arch/sparc/kernel/hvapi.c b/arch/sparc/kernel/hvapi.c index d306e648c33c..c2d055d8ba9e 100644 --- a/arch/sparc/kernel/hvapi.c +++ b/arch/sparc/kernel/hvapi.c @@ -28,11 +28,17 @@ static struct api_info api_table[] = { { .group = HV_GRP_CORE, .flags = FLAG_PRE_API }, { .group = HV_GRP_INTR, }, { .group = HV_GRP_SOFT_STATE, }, + { .group = HV_GRP_TM, }, { .group = HV_GRP_PCI, .flags = FLAG_PRE_API }, { .group = HV_GRP_LDOM, }, { .group = HV_GRP_SVC_CHAN, .flags = FLAG_PRE_API }, { .group = HV_GRP_NCS, .flags = FLAG_PRE_API }, { .group = HV_GRP_RNG, }, + { .group = HV_GRP_PBOOT, }, + { .group = HV_GRP_TPM, }, + { .group = HV_GRP_SDIO, }, + { .group = HV_GRP_SDIO_ERR, }, + { .group = HV_GRP_REBOOT_DATA, }, { .group = HV_GRP_NIAG_PERF, .flags = FLAG_PRE_API }, { .group = HV_GRP_FIRE_PERF, }, { .group = HV_GRP_N2_CPU, }, From ea5e7447ea9d555558e0f13798f5143dd51a915a Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 1 Aug 2011 23:27:17 -0700 Subject: [PATCH 221/439] sparc: Set reboot-cmd using reboot data hypervisor call if available. Signed-off-by: David S. Miller --- arch/sparc/include/asm/hypervisor.h | 7 +++++++ arch/sparc/kernel/ds.c | 30 +++++++++++++++++++++++++++-- arch/sparc/kernel/hvcalls.S | 7 +++++++ arch/sparc/kernel/kernel.h | 9 +++++++++ arch/sparc/kernel/sstate.c | 9 ++------- 5 files changed, 53 insertions(+), 9 deletions(-) diff --git a/arch/sparc/include/asm/hypervisor.h b/arch/sparc/include/asm/hypervisor.h index 2b7da27ff17d..015a761eaa32 100644 --- a/arch/sparc/include/asm/hypervisor.h +++ b/arch/sparc/include/asm/hypervisor.h @@ -2927,6 +2927,13 @@ extern unsigned long sun4v_ncs_request(unsigned long request, #define HV_FAST_FIRE_GET_PERFREG 0x120 #define HV_FAST_FIRE_SET_PERFREG 0x121 +#define HV_FAST_REBOOT_DATA_SET 0x172 + +#ifndef __ASSEMBLY__ +extern unsigned long sun4v_reboot_data_set(unsigned long ra, + unsigned long len); +#endif + /* Function numbers for HV_CORE_TRAP. */ #define HV_CORE_SET_VER 0x00 #define HV_CORE_PUTCHAR 0x01 diff --git a/arch/sparc/kernel/ds.c b/arch/sparc/kernel/ds.c index dd1342c0a3be..490e5418740d 100644 --- a/arch/sparc/kernel/ds.c +++ b/arch/sparc/kernel/ds.c @@ -15,12 +15,15 @@ #include #include +#include #include #include #include #include #include +#include "kernel.h" + #define DRV_MODULE_NAME "ds" #define PFX DRV_MODULE_NAME ": " #define DRV_MODULE_VERSION "1.0" @@ -828,18 +831,32 @@ void ldom_set_var(const char *var, const char *value) } } +static char full_boot_str[256] __attribute__((aligned(32))); +static int reboot_data_supported; + void ldom_reboot(const char *boot_command) { /* Don't bother with any of this if the boot_command * is empty. */ if (boot_command && strlen(boot_command)) { - char full_boot_str[256]; + unsigned long len; strcpy(full_boot_str, "boot "); strcpy(full_boot_str + strlen("boot "), boot_command); + len = strlen(full_boot_str); - ldom_set_var("reboot-command", full_boot_str); + if (reboot_data_supported) { + unsigned long ra = kimage_addr_to_ra(full_boot_str); + unsigned long hv_ret; + + hv_ret = sun4v_reboot_data_set(ra, len); + if (hv_ret != HV_EOK) + pr_err("SUN4V: Unable to set reboot data " + "hv_ret=%lu\n", hv_ret); + } else { + ldom_set_var("reboot-command", full_boot_str); + } } sun4v_mach_sir(); } @@ -1237,6 +1254,15 @@ static struct vio_driver ds_driver = { static int __init ds_init(void) { + unsigned long hv_ret, major, minor; + + hv_ret = sun4v_get_version(HV_GRP_REBOOT_DATA, &major, &minor); + if (hv_ret == HV_EOK) { + pr_info("SUN4V: Reboot data supported (maj=%lu,min=%lu).\n", + major, minor); + reboot_data_supported = 1; + } + kthread_run(ds_thread, NULL, "kldomd"); return vio_register_driver(&ds_driver); diff --git a/arch/sparc/kernel/hvcalls.S b/arch/sparc/kernel/hvcalls.S index 8a5f35ffb15e..58d60de4d65b 100644 --- a/arch/sparc/kernel/hvcalls.S +++ b/arch/sparc/kernel/hvcalls.S @@ -798,3 +798,10 @@ ENTRY(sun4v_niagara2_setperf) retl nop ENDPROC(sun4v_niagara2_setperf) + +ENTRY(sun4v_reboot_data_set) + mov HV_FAST_REBOOT_DATA_SET, %o5 + ta HV_FAST_TRAP + retl + nop +ENDPROC(sun4v_reboot_data_set) diff --git a/arch/sparc/kernel/kernel.h b/arch/sparc/kernel/kernel.h index 8325d7759381..fd6c36b1df74 100644 --- a/arch/sparc/kernel/kernel.h +++ b/arch/sparc/kernel/kernel.h @@ -4,6 +4,8 @@ #include #include +#include +#include /* cpu.c */ extern const char *sparc_pmu_type; @@ -14,6 +16,13 @@ extern int ncpus_probed; /* setup_64.c */ struct seq_file; extern void cpucap_info(struct seq_file *); + +static inline unsigned long kimage_addr_to_ra(const char *p) +{ + unsigned long val = (unsigned long) p; + + return kern_base + (val - KERNBASE); +} #endif #ifdef CONFIG_SPARC32 diff --git a/arch/sparc/kernel/sstate.c b/arch/sparc/kernel/sstate.c index 8cdbe5946b43..c59af546f522 100644 --- a/arch/sparc/kernel/sstate.c +++ b/arch/sparc/kernel/sstate.c @@ -14,15 +14,10 @@ #include #include +#include "kernel.h" + static int hv_supports_soft_state; -static unsigned long kimage_addr_to_ra(const char *p) -{ - unsigned long val = (unsigned long) p; - - return kern_base + (val - KERNBASE); -} - static void do_set_sstate(unsigned long state, const char *msg) { unsigned long err; From 56d205cc5c0a3032a605121d4253e111193bf923 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 2 Aug 2011 20:23:34 -0700 Subject: [PATCH 222/439] sparc: Use popc when possible for ffs/__ffs/ffz. Signed-off-by: David S. Miller --- arch/sparc/include/asm/bitops_64.h | 7 +-- arch/sparc/kernel/entry.h | 7 +++ arch/sparc/kernel/setup_64.c | 34 +++++++----- arch/sparc/kernel/sparc_ksyms_64.c | 4 ++ arch/sparc/kernel/vmlinux.lds.S | 5 ++ arch/sparc/lib/Makefile | 2 +- arch/sparc/lib/ffs.S | 84 ++++++++++++++++++++++++++++++ 7 files changed, 127 insertions(+), 16 deletions(-) create mode 100644 arch/sparc/lib/ffs.S diff --git a/arch/sparc/include/asm/bitops_64.h b/arch/sparc/include/asm/bitops_64.h index 3588807baa6e..29011cc0e4be 100644 --- a/arch/sparc/include/asm/bitops_64.h +++ b/arch/sparc/include/asm/bitops_64.h @@ -26,16 +26,17 @@ extern void change_bit(unsigned long nr, volatile unsigned long *addr); #define smp_mb__before_clear_bit() barrier() #define smp_mb__after_clear_bit() barrier() -#include -#include #include #include #include #ifdef __KERNEL__ +extern int ffs(int x); +extern unsigned long __ffs(unsigned long); + +#include #include -#include /* * hweightN: returns the hamming weight (i.e. the number diff --git a/arch/sparc/kernel/entry.h b/arch/sparc/kernel/entry.h index 16d1545d84fc..e27f8ea8656e 100644 --- a/arch/sparc/kernel/entry.h +++ b/arch/sparc/kernel/entry.h @@ -49,6 +49,13 @@ struct popc_3insn_patch_entry { extern struct popc_3insn_patch_entry __popc_3insn_patch, __popc_3insn_patch_end; +struct popc_6insn_patch_entry { + unsigned int addr; + unsigned int insns[6]; +}; +extern struct popc_6insn_patch_entry __popc_6insn_patch, + __popc_6insn_patch_end; + extern void __init per_cpu_patch(void); extern void __init sun4v_patch(void); extern void __init boot_cpu_id_too_large(int cpu); diff --git a/arch/sparc/kernel/setup_64.c b/arch/sparc/kernel/setup_64.c index 26d114187e10..3e9daea1653d 100644 --- a/arch/sparc/kernel/setup_64.c +++ b/arch/sparc/kernel/setup_64.c @@ -275,25 +275,35 @@ void __init sun4v_patch(void) static void __init popc_patch(void) { struct popc_3insn_patch_entry *p3; + struct popc_6insn_patch_entry *p6; p3 = &__popc_3insn_patch; while (p3 < &__popc_3insn_patch_end) { - unsigned long addr = p3->addr; + unsigned long i, addr = p3->addr; - *(unsigned int *) (addr + 0) = p3->insns[0]; - wmb(); - __asm__ __volatile__("flush %0" : : "r" (addr + 0)); - - *(unsigned int *) (addr + 4) = p3->insns[1]; - wmb(); - __asm__ __volatile__("flush %0" : : "r" (addr + 4)); - - *(unsigned int *) (addr + 8) = p3->insns[2]; - wmb(); - __asm__ __volatile__("flush %0" : : "r" (addr + 4)); + for (i = 0; i < 3; i++) { + *(unsigned int *) (addr + (i * 4)) = p3->insns[i]; + wmb(); + __asm__ __volatile__("flush %0" + : : "r" (addr + (i * 4))); + } p3++; } + + p6 = &__popc_6insn_patch; + while (p6 < &__popc_6insn_patch_end) { + unsigned long i, addr = p6->addr; + + for (i = 0; i < 6; i++) { + *(unsigned int *) (addr + (i * 4)) = p6->insns[i]; + wmb(); + __asm__ __volatile__("flush %0" + : : "r" (addr + (i * 4))); + } + + p6++; + } } #ifdef CONFIG_SMP diff --git a/arch/sparc/kernel/sparc_ksyms_64.c b/arch/sparc/kernel/sparc_ksyms_64.c index d0ee65aced0d..83b47ab02d96 100644 --- a/arch/sparc/kernel/sparc_ksyms_64.c +++ b/arch/sparc/kernel/sparc_ksyms_64.c @@ -45,5 +45,9 @@ EXPORT_SYMBOL(__arch_hweight16); EXPORT_SYMBOL(__arch_hweight32); EXPORT_SYMBOL(__arch_hweight64); +/* from ffs_ffz.S */ +EXPORT_SYMBOL(ffs); +EXPORT_SYMBOL(__ffs); + /* Exporting a symbol from /init/main.c */ EXPORT_SYMBOL(saved_command_line); diff --git a/arch/sparc/kernel/vmlinux.lds.S b/arch/sparc/kernel/vmlinux.lds.S index de20c14625eb..94a954892d3f 100644 --- a/arch/sparc/kernel/vmlinux.lds.S +++ b/arch/sparc/kernel/vmlinux.lds.S @@ -112,6 +112,11 @@ SECTIONS *(.popc_3insn_patch) __popc_3insn_patch_end = .; } + .popc_6insn_patch : { + __popc_6insn_patch = .; + *(.popc_6insn_patch) + __popc_6insn_patch_end = .; + } PERCPU_SECTION(SMP_CACHE_BYTES) . = ALIGN(PAGE_SIZE); diff --git a/arch/sparc/lib/Makefile b/arch/sparc/lib/Makefile index 11c850a9b66f..a3fc4375a150 100644 --- a/arch/sparc/lib/Makefile +++ b/arch/sparc/lib/Makefile @@ -37,7 +37,7 @@ lib-$(CONFIG_SPARC64) += GENmemcpy.o GENcopy_from_user.o GENcopy_to_user.o lib-$(CONFIG_SPARC64) += GENpatch.o GENpage.o GENbzero.o lib-$(CONFIG_SPARC64) += copy_in_user.o user_fixup.o memmove.o -lib-$(CONFIG_SPARC64) += mcount.o ipcsum.o xor.o hweight.o +lib-$(CONFIG_SPARC64) += mcount.o ipcsum.o xor.o hweight.o ffs.o obj-y += iomap.o obj-$(CONFIG_SPARC32) += atomic32.o diff --git a/arch/sparc/lib/ffs.S b/arch/sparc/lib/ffs.S new file mode 100644 index 000000000000..b39389f69899 --- /dev/null +++ b/arch/sparc/lib/ffs.S @@ -0,0 +1,84 @@ +#include + + .register %g2,#scratch + + .text + .align 32 + +ENTRY(ffs) + brnz,pt %o0, 1f + mov 1, %o1 + retl + clr %o0 + nop + nop +ENTRY(__ffs) + sllx %o0, 32, %g1 /* 1 */ + srlx %o0, 32, %g2 + + clr %o1 /* 2 */ + movrz %g1, %g2, %o0 + + movrz %g1, 32, %o1 /* 3 */ +1: clr %o2 + + sllx %o0, (64 - 16), %g1 /* 4 */ + srlx %o0, 16, %g2 + + movrz %g1, %g2, %o0 /* 5 */ + clr %o3 + + movrz %g1, 16, %o2 /* 6 */ + clr %o4 + + and %o0, 0xff, %g1 /* 7 */ + srlx %o0, 8, %g2 + + movrz %g1, %g2, %o0 /* 8 */ + clr %o5 + + movrz %g1, 8, %o3 /* 9 */ + add %o2, %o1, %o2 + + and %o0, 0xf, %g1 /* 10 */ + srlx %o0, 4, %g2 + + movrz %g1, %g2, %o0 /* 11 */ + add %o2, %o3, %o2 + + movrz %g1, 4, %o4 /* 12 */ + + and %o0, 0x3, %g1 /* 13 */ + srlx %o0, 2, %g2 + + movrz %g1, %g2, %o0 /* 14 */ + add %o2, %o4, %o2 + + movrz %g1, 2, %o5 /* 15 */ + + and %o0, 0x1, %g1 /* 16 */ + + add %o2, %o5, %o2 /* 17 */ + xor %g1, 0x1, %g1 + + retl /* 18 */ + add %o2, %g1, %o0 +ENDPROC(ffs) +ENDPROC(__ffs) + + .section .popc_6insn_patch, "ax" + .word ffs + brz,pn %o0, 98f + neg %o0, %g1 + xnor %o0, %g1, %o1 + popc %o1, %o0 +98: retl + nop + .word __ffs + neg %o0, %g1 + xnor %o0, %g1, %o1 + popc %o1, %o0 + retl + sub %o0, 1, %o0 + nop + .previous From 951c0d660a7c35286e401ca6d6ef38c9d49643c7 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 3 Aug 2011 00:47:21 -0400 Subject: [PATCH 223/439] get rid of boilerplate switches in posix_acl.h the only potentially subtle thing here: get_cached_acl() is never called with the second argument other than ACL_TYPE_{ACCESS,DEFAULT}. IOW, that return ERR_PTR(-EINVAL) in there might as well be BUG(). Signed-off-by: Al Viro --- include/linux/posix_acl.h | 62 +++++++++++++-------------------------- 1 file changed, 20 insertions(+), 42 deletions(-) diff --git a/include/linux/posix_acl.h b/include/linux/posix_acl.h index 951bba82d50d..a9c2fb29be96 100644 --- a/include/linux/posix_acl.h +++ b/include/linux/posix_acl.h @@ -84,20 +84,22 @@ extern struct posix_acl *get_posix_acl(struct inode *, int); extern int set_posix_acl(struct inode *, int, struct posix_acl *); #ifdef CONFIG_FS_POSIX_ACL -static inline struct posix_acl *get_cached_acl(struct inode *inode, int type) +static inline struct posix_acl **acl_by_type(struct inode *inode, int type) { - struct posix_acl **p, *acl; switch (type) { case ACL_TYPE_ACCESS: - p = &inode->i_acl; - break; + return &inode->i_acl; case ACL_TYPE_DEFAULT: - p = &inode->i_default_acl; - break; + return &inode->i_default_acl; default: - return ERR_PTR(-EINVAL); + BUG(); } - acl = ACCESS_ONCE(*p); +} + +static inline struct posix_acl *get_cached_acl(struct inode *inode, int type) +{ + struct posix_acl **p = acl_by_type(inode, type); + struct posix_acl *acl = ACCESS_ONCE(*p); if (acl) { spin_lock(&inode->i_lock); acl = *p; @@ -110,18 +112,8 @@ static inline struct posix_acl *get_cached_acl(struct inode *inode, int type) static inline int negative_cached_acl(struct inode *inode, int type) { - struct posix_acl **p, *acl; - switch (type) { - case ACL_TYPE_ACCESS: - p = &inode->i_acl; - break; - case ACL_TYPE_DEFAULT: - p = &inode->i_default_acl; - break; - default: - BUG(); - } - acl = ACCESS_ONCE(*p); + struct posix_acl **p = acl_by_type(inode, type); + struct posix_acl *acl = ACCESS_ONCE(*p); if (acl) return 0; return 1; @@ -131,18 +123,11 @@ static inline void set_cached_acl(struct inode *inode, int type, struct posix_acl *acl) { - struct posix_acl *old = NULL; + struct posix_acl **p = acl_by_type(inode, type); + struct posix_acl *old; spin_lock(&inode->i_lock); - switch (type) { - case ACL_TYPE_ACCESS: - old = inode->i_acl; - inode->i_acl = posix_acl_dup(acl); - break; - case ACL_TYPE_DEFAULT: - old = inode->i_default_acl; - inode->i_default_acl = posix_acl_dup(acl); - break; - } + old = *p; + *p = posix_dup_acl(acl); spin_unlock(&inode->i_lock); if (old != ACL_NOT_CACHED) posix_acl_release(old); @@ -150,18 +135,11 @@ static inline void set_cached_acl(struct inode *inode, static inline void forget_cached_acl(struct inode *inode, int type) { - struct posix_acl *old = NULL; + struct posix_acl **p = acl_by_type(inode, type); + struct posix_acl *old; spin_lock(&inode->i_lock); - switch (type) { - case ACL_TYPE_ACCESS: - old = inode->i_acl; - inode->i_acl = ACL_NOT_CACHED; - break; - case ACL_TYPE_DEFAULT: - old = inode->i_default_acl; - inode->i_default_acl = ACL_NOT_CACHED; - break; - } + old = *p; + *p = ACL_NOT_CACHED; spin_unlock(&inode->i_lock); if (old != ACL_NOT_CACHED) posix_acl_release(old); From 3567866bf26190d1e734c975c907eb06e923ba23 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 2 Aug 2011 21:32:13 -0400 Subject: [PATCH 224/439] RCUify freeing acls, let check_acl() go ahead in RCU mode if acl is cached Signed-off-by: Al Viro --- fs/namei.c | 17 ++++++----------- include/linux/posix_acl.h | 18 +++++++++--------- 2 files changed, 15 insertions(+), 20 deletions(-) diff --git a/fs/namei.c b/fs/namei.c index 445fd5da11fa..3d607bd80e09 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -179,19 +179,14 @@ static int check_acl(struct inode *inode, int mask) #ifdef CONFIG_FS_POSIX_ACL struct posix_acl *acl; - /* - * Under RCU walk, we cannot even do a "get_cached_acl()", - * because that involves locking and getting a refcount on - * a cached ACL. - * - * So the only case we handle during RCU walking is the - * case of a cached "no ACL at all", which needs no locks - * or refcounts. - */ if (mask & MAY_NOT_BLOCK) { - if (negative_cached_acl(inode, ACL_TYPE_ACCESS)) + acl = get_cached_acl_rcu(inode, ACL_TYPE_ACCESS); + if (!acl) return -EAGAIN; - return -ECHILD; + /* no ->get_acl() calls in RCU mode... */ + if (acl == ACL_NOT_CACHED) + return -ECHILD; + return posix_acl_permission(inode, acl, mask); } acl = get_cached_acl(inode, ACL_TYPE_ACCESS); diff --git a/include/linux/posix_acl.h b/include/linux/posix_acl.h index a9c2fb29be96..b7681102a4b9 100644 --- a/include/linux/posix_acl.h +++ b/include/linux/posix_acl.h @@ -9,6 +9,7 @@ #define __LINUX_POSIX_ACL_H #include +#include #define ACL_UNDEFINED_ID (-1) @@ -38,7 +39,10 @@ struct posix_acl_entry { }; struct posix_acl { - atomic_t a_refcount; + union { + atomic_t a_refcount; + struct rcu_head a_rcu; + }; unsigned int a_count; struct posix_acl_entry a_entries[0]; }; @@ -65,7 +69,7 @@ static inline void posix_acl_release(struct posix_acl *acl) { if (acl && atomic_dec_and_test(&acl->a_refcount)) - kfree(acl); + kfree_rcu(acl, a_rcu); } @@ -110,13 +114,9 @@ static inline struct posix_acl *get_cached_acl(struct inode *inode, int type) return acl; } -static inline int negative_cached_acl(struct inode *inode, int type) +static inline struct posix_acl *get_cached_acl_rcu(struct inode *inode, int type) { - struct posix_acl **p = acl_by_type(inode, type); - struct posix_acl *acl = ACCESS_ONCE(*p); - if (acl) - return 0; - return 1; + return rcu_dereference(*acl_by_type(inode, type)); } static inline void set_cached_acl(struct inode *inode, @@ -127,7 +127,7 @@ static inline void set_cached_acl(struct inode *inode, struct posix_acl *old; spin_lock(&inode->i_lock); old = *p; - *p = posix_dup_acl(acl); + rcu_assign_pointer(*p, posix_acl_dup(acl)); spin_unlock(&inode->i_lock); if (old != ACL_NOT_CACHED) posix_acl_release(old); From 0a7a8fff7b8a3bc2d3528af07c9c88083250303d Mon Sep 17 00:00:00 2001 From: Tobias Klauser Date: Tue, 2 Aug 2011 15:40:22 -0700 Subject: [PATCH 225/439] Input: gpio_keys - fix two typos in devicetree documentation Signed-off-by: Tobias Klauser Signed-off-by: Dmitry Torokhov --- Documentation/devicetree/bindings/gpio/gpio_keys.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/gpio/gpio_keys.txt b/Documentation/devicetree/bindings/gpio/gpio_keys.txt index 7190c99d7611..5c2c02140a62 100644 --- a/Documentation/devicetree/bindings/gpio/gpio_keys.txt +++ b/Documentation/devicetree/bindings/gpio/gpio_keys.txt @@ -10,7 +10,7 @@ Optional properties: Each button (key) is represented as a sub-node of "gpio-keys": Subnode properties: - - gpios: OF devcie-tree gpio specificatin. + - gpios: OF device-tree gpio specification. - label: Descriptive name of the key. - linux,code: Keycode to emit. From 4fecc2088fd02ce8a6580c12822987d3c6bab9b8 Mon Sep 17 00:00:00 2001 From: Michael Hennerich Date: Tue, 2 Aug 2011 15:41:37 -0700 Subject: [PATCH 226/439] Input: ad7879 - fix deficient device disable Input close or device disable should not interact with the exported gpiolib functionality. However that's the case. __ad7879_disable() clears the entire AD7879_REG_CTRL2, while it should just power down the ADC and its reference. Signed-off-by: Michael Hennerich Signed-off-by: Dmitry Torokhov --- drivers/input/touchscreen/ad7879.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/input/touchscreen/ad7879.c b/drivers/input/touchscreen/ad7879.c index bc3b5187f3a3..131f9d1c921b 100644 --- a/drivers/input/touchscreen/ad7879.c +++ b/drivers/input/touchscreen/ad7879.c @@ -249,12 +249,14 @@ static void __ad7879_enable(struct ad7879 *ts) static void __ad7879_disable(struct ad7879 *ts) { + u16 reg = (ts->cmd_crtl2 & ~AD7879_PM(-1)) | + AD7879_PM(AD7879_PM_SHUTDOWN); disable_irq(ts->irq); if (del_timer_sync(&ts->timer)) ad7879_ts_event_release(ts); - ad7879_write(ts, AD7879_REG_CTRL2, AD7879_PM(AD7879_PM_SHUTDOWN)); + ad7879_write(ts, AD7879_REG_CTRL2, reg); } From dcd998ccdbf74a7d8fe0f0a44e85da1ed5975946 Mon Sep 17 00:00:00 2001 From: Kiran Patil Date: Wed, 3 Aug 2011 09:20:01 +0000 Subject: [PATCH 227/439] tcm_fc: Handle DDP/SW fc_frame_payload_get failures in ft_recv_write_data Problem: HW DDP context was not invalidated in case of ABORTS, etc... This leads to the problem where memory pages which are used for DDP as user descriptor could get reused for some other purpose (such as to satisfy new memory allocation request either by kernel or user mode threads) and since HW DDP context was not invalidated, HW continue to write to those pages, hence causing memory corruption. Fix: Either on incoming ABORTS or due to exchange time out, allowed the target to cleanup HW DDP context if it was setup for respective ft_cmd. Added new function to perform this cleanup, furthur it can be enhanced for other cleanup activity. Fix ft_recv_write_data() to properly handle fc_frame_payload_get to return pointer to payload if it exist. If there is no payload which is most common case (+ve case in case if DDP is working as expected, it will return NULL. Yes, scope of buf is limited to printk. Invalidation of HW context (which is done inside ft_invl_hw_context() is necessary in SUCCESS and FAILURE case of DDP. Hence invalidation is DONE as long as there was DDP setup (whether it worked correctly or not, NOTE: For some reason, if there is any error w.r.t DDP such as out of order packet reception, HW simply post the full packet in rx queue. Signed-off-by: Kiran Patil Cc: Robert W Love Signed-off-by: Nicholas A. Bellinger --- drivers/target/tcm_fc/tcm_fc.h | 5 ++ drivers/target/tcm_fc/tfc_cmd.c | 1 + drivers/target/tcm_fc/tfc_io.c | 127 +++++++++++++++++++------------- 3 files changed, 81 insertions(+), 52 deletions(-) diff --git a/drivers/target/tcm_fc/tcm_fc.h b/drivers/target/tcm_fc/tcm_fc.h index f7fff7ed63c3..bd4fe21a23b8 100644 --- a/drivers/target/tcm_fc/tcm_fc.h +++ b/drivers/target/tcm_fc/tcm_fc.h @@ -187,4 +187,9 @@ void ft_dump_cmd(struct ft_cmd *, const char *caller); ssize_t ft_format_wwn(char *, size_t, u64); +/* + * Underlying HW specific helper function + */ +void ft_invl_hw_context(struct ft_cmd *); + #endif /* __TCM_FC_H__ */ diff --git a/drivers/target/tcm_fc/tfc_cmd.c b/drivers/target/tcm_fc/tfc_cmd.c index a9e9a31da11d..03977e8996d8 100644 --- a/drivers/target/tcm_fc/tfc_cmd.c +++ b/drivers/target/tcm_fc/tfc_cmd.c @@ -320,6 +320,7 @@ static void ft_recv_seq(struct fc_seq *sp, struct fc_frame *fp, void *arg) default: pr_debug("%s: unhandled frame r_ctl %x\n", __func__, fh->fh_r_ctl); + ft_invl_hw_context(cmd); fc_frame_free(fp); transport_generic_free_cmd(&cmd->se_cmd, 0, 0); break; diff --git a/drivers/target/tcm_fc/tfc_io.c b/drivers/target/tcm_fc/tfc_io.c index 11e6483fc127..06943eeb3c84 100644 --- a/drivers/target/tcm_fc/tfc_io.c +++ b/drivers/target/tcm_fc/tfc_io.c @@ -214,62 +214,49 @@ void ft_recv_write_data(struct ft_cmd *cmd, struct fc_frame *fp) if (!(ntoh24(fh->fh_f_ctl) & FC_FC_REL_OFF)) goto drop; - /* - * Doesn't expect even single byte of payload. Payload - * is expected to be copied directly to user buffers - * due to DDP (Large Rx offload) feature, hence - * BUG_ON if BUF is non-NULL - */ - buf = fc_frame_payload_get(fp, 1); - if (cmd->was_ddp_setup && buf) { - pr_debug("%s: When DDP was setup, not expected to" - "receive frame with payload, Payload shall be" - "copied directly to buffer instead of coming " - "via. legacy receive queues\n", __func__); - BUG_ON(buf); + f_ctl = ntoh24(fh->fh_f_ctl); + ep = fc_seq_exch(seq); + lport = ep->lp; + if (cmd->was_ddp_setup) { + BUG_ON(!ep); + BUG_ON(!lport); } /* - * If ft_cmd indicated 'ddp_setup', in that case only the last frame - * should come with 'TSI bit being set'. If 'TSI bit is not set and if - * data frame appears here, means error condition. In both the cases - * release the DDP context (ddp_put) and in error case, as well - * initiate error recovery mechanism. + * Doesn't expect payload if DDP is setup. Payload + * is expected to be copied directly to user buffers + * due to DDP (Large Rx offload), */ - ep = fc_seq_exch(seq); - if (cmd->was_ddp_setup) { - BUG_ON(!ep); - lport = ep->lp; - BUG_ON(!lport); - } - if (cmd->was_ddp_setup && ep->xid != FC_XID_UNKNOWN) { - f_ctl = ntoh24(fh->fh_f_ctl); - /* - * If TSI bit set in f_ctl, means last write data frame is - * received successfully where payload is posted directly - * to user buffer and only the last frame's header is posted - * in legacy receive queue - */ - if (f_ctl & FC_FC_SEQ_INIT) { /* TSI bit set in FC frame */ - cmd->write_data_len = lport->tt.ddp_done(lport, - ep->xid); - goto last_frame; - } else { - /* - * Updating the write_data_len may be meaningless at - * this point, but just in case if required in future - * for debugging or any other purpose - */ - pr_err("%s: Received frame with TSI bit not" - " being SET, dropping the frame, " - "cmd->sg <%p>, cmd->sg_cnt <0x%x>\n", - __func__, cmd->sg, cmd->sg_cnt); - cmd->write_data_len = lport->tt.ddp_done(lport, - ep->xid); - lport->tt.seq_exch_abort(cmd->seq, 0); - goto drop; - } - } + buf = fc_frame_payload_get(fp, 1); + if (buf) + pr_err("%s: xid 0x%x, f_ctl 0x%x, cmd->sg %p, " + "cmd->sg_cnt 0x%x. DDP was setup" + " hence not expected to receive frame with " + "payload, Frame will be dropped if " + "'Sequence Initiative' bit in f_ctl is " + "not set\n", __func__, ep->xid, f_ctl, + cmd->sg, cmd->sg_cnt); + /* + * Invalidate HW DDP context if it was setup for respective + * command. Invalidation of HW DDP context is requited in both + * situation (success and error). + */ + ft_invl_hw_context(cmd); + + /* + * If "Sequence Initiative (TSI)" bit set in f_ctl, means last + * write data frame is received successfully where payload is + * posted directly to user buffer and only the last frame's + * header is posted in receive queue. + * + * If "Sequence Initiative (TSI)" bit is not set, means error + * condition w.r.t. DDP, hence drop the packet and let explict + * ABORTS from other end of exchange timer trigger the recovery. + */ + if (f_ctl & FC_FC_SEQ_INIT) + goto last_frame; + else + goto drop; rel_off = ntohl(fh->fh_parm_offset); frame_len = fr_len(fp); @@ -332,3 +319,39 @@ last_frame: drop: fc_frame_free(fp); } + +/* + * Handle and cleanup any HW specific resources if + * received ABORTS, errors, timeouts. + */ +void ft_invl_hw_context(struct ft_cmd *cmd) +{ + struct fc_seq *seq = cmd->seq; + struct fc_exch *ep = NULL; + struct fc_lport *lport = NULL; + + BUG_ON(!cmd); + + /* Cleanup the DDP context in HW if DDP was setup */ + if (cmd->was_ddp_setup && seq) { + ep = fc_seq_exch(seq); + if (ep) { + lport = ep->lp; + if (lport && (ep->xid <= lport->lro_xid)) + /* + * "ddp_done" trigger invalidation of HW + * specific DDP context + */ + cmd->write_data_len = lport->tt.ddp_done(lport, + ep->xid); + + /* + * Resetting same variable to indicate HW's + * DDP context has been invalidated to avoid + * re_invalidation of same context (context is + * identified using ep->xid) + */ + cmd->was_ddp_setup = 0; + } + } +} From 9d5b36be64b42058e7fd12b71266bbf2bb7600fc Mon Sep 17 00:00:00 2001 From: Yaniv Rosner Date: Tue, 2 Aug 2011 22:59:10 +0000 Subject: [PATCH 228/439] bnx2x: Fix missing pause on for 578xx When link speed is 1G and below, pause weren't sent due to missing pause setting in the UMAC. Signed-off-by: Yaniv Rosner Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_link.c | 6 ++++++ drivers/net/bnx2x/bnx2x_reg.h | 2 ++ 2 files changed, 8 insertions(+) diff --git a/drivers/net/bnx2x/bnx2x_link.c b/drivers/net/bnx2x/bnx2x_link.c index bcd8f0038628..aa9958e37445 100644 --- a/drivers/net/bnx2x/bnx2x_link.c +++ b/drivers/net/bnx2x/bnx2x_link.c @@ -1546,6 +1546,12 @@ static void bnx2x_umac_enable(struct link_params *params, vars->line_speed); break; } + if (!(vars->flow_ctrl & BNX2X_FLOW_CTRL_TX)) + val |= UMAC_COMMAND_CONFIG_REG_IGNORE_TX_PAUSE; + + if (!(vars->flow_ctrl & BNX2X_FLOW_CTRL_RX)) + val |= UMAC_COMMAND_CONFIG_REG_PAUSE_IGNORE; + REG_WR(bp, umac_base + UMAC_REG_COMMAND_CONFIG, val); udelay(50); diff --git a/drivers/net/bnx2x/bnx2x_reg.h b/drivers/net/bnx2x/bnx2x_reg.h index 02461fef8751..d84642aca450 100644 --- a/drivers/net/bnx2x/bnx2x_reg.h +++ b/drivers/net/bnx2x/bnx2x_reg.h @@ -4771,9 +4771,11 @@ The fields are: [4:0] - tail pointer; 10:5] - Link List size; 15:11] - header pointer. */ #define UCM_REG_XX_TABLE 0xe0300 +#define UMAC_COMMAND_CONFIG_REG_IGNORE_TX_PAUSE (0x1<<28) #define UMAC_COMMAND_CONFIG_REG_LOOP_ENA (0x1<<15) #define UMAC_COMMAND_CONFIG_REG_NO_LGTH_CHECK (0x1<<24) #define UMAC_COMMAND_CONFIG_REG_PAD_EN (0x1<<5) +#define UMAC_COMMAND_CONFIG_REG_PAUSE_IGNORE (0x1<<8) #define UMAC_COMMAND_CONFIG_REG_PROMIS_EN (0x1<<4) #define UMAC_COMMAND_CONFIG_REG_RX_ENA (0x1<<1) #define UMAC_COMMAND_CONFIG_REG_SW_RESET (0x1<<13) From b507766205f85d6d69892287e346a7c264a216b4 Mon Sep 17 00:00:00 2001 From: Yaniv Rosner Date: Tue, 2 Aug 2011 22:59:18 +0000 Subject: [PATCH 229/439] bnx2x: Fix chip hanging due to TX pipe stall. Fix a problem in which the 578xx chip hangs after running traffic, and then pulling the network cable. This occurs since TX pipe is stalled due to missing XON indication towards the NIG. Signed-off-by: Yaniv Rosner Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_link.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/drivers/net/bnx2x/bnx2x_link.c b/drivers/net/bnx2x/bnx2x_link.c index aa9958e37445..0a7091d3de29 100644 --- a/drivers/net/bnx2x/bnx2x_link.c +++ b/drivers/net/bnx2x/bnx2x_link.c @@ -1667,10 +1667,20 @@ static void bnx2x_xmac_disable(struct link_params *params) { u8 port = params->port; struct bnx2x *bp = params->bp; - u32 xmac_base = (port) ? GRCBASE_XMAC1 : GRCBASE_XMAC0; + u32 pfc_ctrl, xmac_base = (port) ? GRCBASE_XMAC1 : GRCBASE_XMAC0; if (REG_RD(bp, MISC_REG_RESET_REG_2) & MISC_REGISTERS_RESET_REG_2_XMAC) { + /* + * Send an indication to change the state in the NIG back to XON + * Clearing this bit enables the next set of this bit to get + * rising edge + */ + pfc_ctrl = REG_RD(bp, xmac_base + XMAC_REG_PFC_CTRL_HI); + REG_WR(bp, xmac_base + XMAC_REG_PFC_CTRL_HI, + (pfc_ctrl & ~(1<<1))); + REG_WR(bp, xmac_base + XMAC_REG_PFC_CTRL_HI, + (pfc_ctrl | (1<<1))); DP(NETIF_MSG_LINK, "Disable XMAC on port %x\n", port); REG_WR(bp, xmac_base + XMAC_REG_CTRL, 0); usleep_range(1000, 1000); From de6f3377d2da3b384ca3d716ffb8687ad175788a Mon Sep 17 00:00:00 2001 From: Yaniv Rosner Date: Tue, 2 Aug 2011 22:59:25 +0000 Subject: [PATCH 230/439] bnx2x: Fix remote fault handling Fix couple of issues of remote fault detection and handling: Link may go down due to remote fault indications during link establishment. Possible link down after primary function migration. Remote fault was not detected on 578xx. Signed-off-by: Yaniv Rosner Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_hsi.h | 2 + drivers/net/bnx2x/bnx2x_link.c | 110 ++++++++++++++++++++------------- drivers/net/bnx2x/bnx2x_link.h | 2 +- 3 files changed, 69 insertions(+), 45 deletions(-) diff --git a/drivers/net/bnx2x/bnx2x_hsi.h b/drivers/net/bnx2x/bnx2x_hsi.h index 06727f32e505..dc24de40e336 100644 --- a/drivers/net/bnx2x/bnx2x_hsi.h +++ b/drivers/net/bnx2x/bnx2x_hsi.h @@ -1204,6 +1204,8 @@ struct drv_port_mb { #define LINK_STATUS_PFC_ENABLED 0x20000000 + #define LINK_STATUS_PHYSICAL_LINK_FLAG 0x40000000 + u32 port_stx; u32 stat_nig_timer; diff --git a/drivers/net/bnx2x/bnx2x_link.c b/drivers/net/bnx2x/bnx2x_link.c index 0a7091d3de29..8e68e1582ec2 100644 --- a/drivers/net/bnx2x/bnx2x_link.c +++ b/drivers/net/bnx2x/bnx2x_link.c @@ -1745,6 +1745,10 @@ static int bnx2x_emac_enable(struct link_params *params, DP(NETIF_MSG_LINK, "enabling EMAC\n"); + /* Disable BMAC */ + REG_WR(bp, GRCBASE_MISC + MISC_REGISTERS_RESET_REG_2_CLEAR, + (MISC_REGISTERS_RESET_REG_2_RST_BMAC0 << port)); + /* enable emac and not bmac */ REG_WR(bp, NIG_REG_EGRESS_EMAC0_PORT + port*4, 1); @@ -2599,12 +2603,6 @@ static int bnx2x_bmac1_enable(struct link_params *params, REG_WR_DMAE(bp, bmac_addr + BIGMAC_REGISTER_RX_LLFC_MSG_FLDS, wb_data, 2); - if (vars->phy_flags & PHY_TX_ERROR_CHECK_FLAG) { - REG_RD_DMAE(bp, bmac_addr + BIGMAC_REGISTER_RX_LSS_STATUS, - wb_data, 2); - if (wb_data[0] > 0) - return -ESRCH; - } return 0; } @@ -2670,16 +2668,6 @@ static int bnx2x_bmac2_enable(struct link_params *params, udelay(30); bnx2x_update_pfc_bmac2(params, vars, is_lb); - if (vars->phy_flags & PHY_TX_ERROR_CHECK_FLAG) { - REG_RD_DMAE(bp, bmac_addr + BIGMAC2_REGISTER_RX_LSS_STAT, - wb_data, 2); - if (wb_data[0] > 0) { - DP(NETIF_MSG_LINK, "Got bad LSS status 0x%x\n", - wb_data[0]); - return -ESRCH; - } - } - return 0; } @@ -4369,6 +4357,9 @@ void bnx2x_link_status_update(struct link_params *params, vars->link_up = (vars->link_status & LINK_STATUS_LINK_UP); vars->phy_flags = PHY_XGXS_FLAG; + if (vars->link_status & LINK_STATUS_PHYSICAL_LINK_FLAG) + vars->phy_flags |= PHY_PHYSICAL_LINK_FLAG; + if (vars->link_up) { DP(NETIF_MSG_LINK, "phy link up\n"); @@ -4460,6 +4451,8 @@ void bnx2x_link_status_update(struct link_params *params, /* indicate no mac active */ vars->mac_type = MAC_TYPE_NONE; + if (vars->link_status & LINK_STATUS_PHYSICAL_LINK_FLAG) + vars->phy_flags |= PHY_HALF_OPEN_CONN_FLAG; } /* Sync media type */ @@ -6176,6 +6169,7 @@ static int bnx2x_update_link_down(struct link_params *params, /* update shared memory */ vars->link_status &= ~(LINK_STATUS_SPEED_AND_DUPLEX_MASK | LINK_STATUS_LINK_UP | + LINK_STATUS_PHYSICAL_LINK_FLAG | LINK_STATUS_AUTO_NEGOTIATE_COMPLETE | LINK_STATUS_RX_FLOW_CONTROL_FLAG_MASK | LINK_STATUS_TX_FLOW_CONTROL_FLAG_MASK | @@ -6213,7 +6207,8 @@ static int bnx2x_update_link_up(struct link_params *params, u8 port = params->port; int rc = 0; - vars->link_status |= LINK_STATUS_LINK_UP; + vars->link_status |= (LINK_STATUS_LINK_UP | + LINK_STATUS_PHYSICAL_LINK_FLAG); vars->phy_flags |= PHY_PHYSICAL_LINK_FLAG; if (vars->flow_ctrl & BNX2X_FLOW_CTRL_TX) @@ -8132,7 +8127,6 @@ void bnx2x_handle_module_detect_int(struct link_params *params) offsetof(struct shmem_region, dev_info. port_feature_config[params->port]. config)); - bnx2x_set_gpio_int(bp, gpio_num, MISC_REGISTERS_GPIO_INT_OUTPUT_SET, gpio_port); @@ -8141,8 +8135,9 @@ void bnx2x_handle_module_detect_int(struct link_params *params) * Disable transmit for this module */ phy->media_type = ETH_PHY_NOT_PRESENT; - if ((val & PORT_FEAT_CFG_OPT_MDL_ENFRCMNT_MASK) == - PORT_FEAT_CFG_OPT_MDL_ENFRCMNT_DISABLE_TX_LASER) + if (((val & PORT_FEAT_CFG_OPT_MDL_ENFRCMNT_MASK) == + PORT_FEAT_CFG_OPT_MDL_ENFRCMNT_DISABLE_TX_LASER) || + CHIP_IS_E3(bp)) bnx2x_sfp_set_transmitter(params, phy, 0); } } @@ -8244,9 +8239,6 @@ static u8 bnx2x_8706_config_init(struct bnx2x_phy *phy, u16 cnt, val, tmp1; struct bnx2x *bp = params->bp; - /* SPF+ PHY: Set flag to check for Tx error */ - vars->phy_flags = PHY_TX_ERROR_CHECK_FLAG; - bnx2x_set_gpio(bp, MISC_REGISTERS_GPIO_2, MISC_REGISTERS_GPIO_OUTPUT_HIGH, params->port); /* HW reset */ @@ -8430,9 +8422,6 @@ static int bnx2x_8726_config_init(struct bnx2x_phy *phy, struct bnx2x *bp = params->bp; DP(NETIF_MSG_LINK, "Initializing BCM8726\n"); - /* SPF+ PHY: Set flag to check for Tx error */ - vars->phy_flags = PHY_TX_ERROR_CHECK_FLAG; - bnx2x_cl45_write(bp, phy, MDIO_PMA_DEVAD, MDIO_PMA_REG_CTRL, 1<<15); bnx2x_wait_reset_complete(bp, phy, params); @@ -8601,9 +8590,6 @@ static int bnx2x_8727_config_init(struct bnx2x_phy *phy, struct bnx2x *bp = params->bp; /* Enable PMD link, MOD_ABS_FLT, and 1G link alarm */ - /* SPF+ PHY: Set flag to check for Tx error */ - vars->phy_flags = PHY_TX_ERROR_CHECK_FLAG; - bnx2x_wait_reset_complete(bp, phy, params); rx_alarm_ctrl_val = (1<<2) | (1<<5) ; /* Should be 0x6 to enable XS on Tx side. */ @@ -10619,7 +10605,8 @@ static struct bnx2x_phy phy_warpcore = { .type = PORT_HW_CFG_XGXS_EXT_PHY_TYPE_DIRECT, .addr = 0xff, .def_md_devad = 0, - .flags = FLAGS_HW_LOCK_REQUIRED, + .flags = (FLAGS_HW_LOCK_REQUIRED | + FLAGS_TX_ERROR_CHECK), .rx_preemphasis = {0xffff, 0xffff, 0xffff, 0xffff}, .tx_preemphasis = {0xffff, 0xffff, 0xffff, 0xffff}, .mdio_ctrl = 0, @@ -10745,7 +10732,8 @@ static struct bnx2x_phy phy_8706 = { .type = PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8706, .addr = 0xff, .def_md_devad = 0, - .flags = FLAGS_INIT_XGXS_FIRST, + .flags = (FLAGS_INIT_XGXS_FIRST | + FLAGS_TX_ERROR_CHECK), .rx_preemphasis = {0xffff, 0xffff, 0xffff, 0xffff}, .tx_preemphasis = {0xffff, 0xffff, 0xffff, 0xffff}, .mdio_ctrl = 0, @@ -10776,7 +10764,8 @@ static struct bnx2x_phy phy_8726 = { .addr = 0xff, .def_md_devad = 0, .flags = (FLAGS_HW_LOCK_REQUIRED | - FLAGS_INIT_XGXS_FIRST), + FLAGS_INIT_XGXS_FIRST | + FLAGS_TX_ERROR_CHECK), .rx_preemphasis = {0xffff, 0xffff, 0xffff, 0xffff}, .tx_preemphasis = {0xffff, 0xffff, 0xffff, 0xffff}, .mdio_ctrl = 0, @@ -10807,7 +10796,8 @@ static struct bnx2x_phy phy_8727 = { .type = PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM8727, .addr = 0xff, .def_md_devad = 0, - .flags = FLAGS_FAN_FAILURE_DET_REQ, + .flags = (FLAGS_FAN_FAILURE_DET_REQ | + FLAGS_TX_ERROR_CHECK), .rx_preemphasis = {0xffff, 0xffff, 0xffff, 0xffff}, .tx_preemphasis = {0xffff, 0xffff, 0xffff, 0xffff}, .mdio_ctrl = 0, @@ -12194,10 +12184,6 @@ static void bnx2x_analyze_link_error(struct link_params *params, u8 led_mode; u32 half_open_conn = (vars->phy_flags & PHY_HALF_OPEN_CONN_FLAG) > 0; - /*DP(NETIF_MSG_LINK, "CHECK LINK: %x half_open:%x-> lss:%x\n", - vars->link_up, - half_open_conn, lss_status);*/ - if ((lss_status ^ half_open_conn) == 0) return; @@ -12210,6 +12196,7 @@ static void bnx2x_analyze_link_error(struct link_params *params, * b. Update link_vars->link_up */ if (lss_status) { + DP(NETIF_MSG_LINK, "Remote Fault detected !!!\n"); vars->link_status &= ~LINK_STATUS_LINK_UP; vars->link_up = 0; vars->phy_flags |= PHY_HALF_OPEN_CONN_FLAG; @@ -12219,6 +12206,7 @@ static void bnx2x_analyze_link_error(struct link_params *params, */ led_mode = LED_MODE_OFF; } else { + DP(NETIF_MSG_LINK, "Remote Fault cleared\n"); vars->link_status |= LINK_STATUS_LINK_UP; vars->link_up = 1; vars->phy_flags &= ~PHY_HALF_OPEN_CONN_FLAG; @@ -12235,6 +12223,15 @@ static void bnx2x_analyze_link_error(struct link_params *params, bnx2x_notify_link_changed(bp); } +/****************************************************************************** +* Description: +* This function checks for half opened connection change indication. +* When such change occurs, it calls the bnx2x_analyze_link_error +* to check if Remote Fault is set or cleared. Reception of remote fault +* status message in the MAC indicates that the peer's MAC has detected +* a fault, for example, due to break in the TX side of fiber. +* +******************************************************************************/ static void bnx2x_check_half_open_conn(struct link_params *params, struct link_vars *vars) { @@ -12245,9 +12242,28 @@ static void bnx2x_check_half_open_conn(struct link_params *params, if ((vars->phy_flags & PHY_PHYSICAL_LINK_FLAG) == 0) return; - if (!CHIP_IS_E3(bp) && + if (CHIP_IS_E3(bp) && (REG_RD(bp, MISC_REG_RESET_REG_2) & - (MISC_REGISTERS_RESET_REG_2_RST_BMAC0 << params->port))) { + (MISC_REGISTERS_RESET_REG_2_XMAC))) { + /* Check E3 XMAC */ + /* + * Note that link speed cannot be queried here, since it may be + * zero while link is down. In case UMAC is active, LSS will + * simply not be set + */ + mac_base = (params->port) ? GRCBASE_XMAC1 : GRCBASE_XMAC0; + + /* Clear stick bits (Requires rising edge) */ + REG_WR(bp, mac_base + XMAC_REG_CLEAR_RX_LSS_STATUS, 0); + REG_WR(bp, mac_base + XMAC_REG_CLEAR_RX_LSS_STATUS, + XMAC_CLEAR_RX_LSS_STATUS_REG_CLEAR_LOCAL_FAULT_STATUS | + XMAC_CLEAR_RX_LSS_STATUS_REG_CLEAR_REMOTE_FAULT_STATUS); + if (REG_RD(bp, mac_base + XMAC_REG_RX_LSS_STATUS)) + lss_status = 1; + + bnx2x_analyze_link_error(params, vars, lss_status); + } else if (REG_RD(bp, MISC_REG_RESET_REG_2) & + (MISC_REGISTERS_RESET_REG_2_RST_BMAC0 << params->port)) { /* Check E1X / E2 BMAC */ u32 lss_status_reg; u32 wb_data[2]; @@ -12269,14 +12285,20 @@ static void bnx2x_check_half_open_conn(struct link_params *params, void bnx2x_period_func(struct link_params *params, struct link_vars *vars) { struct bnx2x *bp = params->bp; + u16 phy_idx; if (!params) { - DP(NETIF_MSG_LINK, "Ininitliazed params !\n"); + DP(NETIF_MSG_LINK, "Uninitialized params !\n"); return; } - /* DP(NETIF_MSG_LINK, "Periodic called vars->phy_flags 0x%x speed 0x%x - RESET_REG_2 0x%x\n", vars->phy_flags, vars->line_speed, - REG_RD(bp, MISC_REG_RESET_REG_2)); */ - bnx2x_check_half_open_conn(params, vars); + + for (phy_idx = INT_PHY; phy_idx < MAX_PHYS; phy_idx++) { + if (params->phy[phy_idx].flags & FLAGS_TX_ERROR_CHECK) { + bnx2x_set_aer_mmd(params, ¶ms->phy[phy_idx]); + bnx2x_check_half_open_conn(params, vars); + break; + } + } + if (CHIP_IS_E3(bp)) bnx2x_check_over_curr(params, vars); } diff --git a/drivers/net/bnx2x/bnx2x_link.h b/drivers/net/bnx2x/bnx2x_link.h index 6a7708d5da37..7ee6b51f4fb8 100644 --- a/drivers/net/bnx2x/bnx2x_link.h +++ b/drivers/net/bnx2x/bnx2x_link.h @@ -145,6 +145,7 @@ struct bnx2x_phy { #define FLAGS_SFP_NOT_APPROVED (1<<7) #define FLAGS_MDC_MDIO_WA (1<<8) #define FLAGS_DUMMY_READ (1<<9) +#define FLAGS_TX_ERROR_CHECK (1<<12) /* preemphasis values for the rx side */ u16 rx_preemphasis[4]; @@ -276,7 +277,6 @@ struct link_vars { #define PHY_PHYSICAL_LINK_FLAG (1<<2) #define PHY_HALF_OPEN_CONN_FLAG (1<<3) #define PHY_OVER_CURRENT_FLAG (1<<4) -#define PHY_TX_ERROR_CHECK_FLAG (1<<5) u8 mac_type; #define MAC_TYPE_NONE 0 From 157fa283a7cb5bc6a55dd4e0daf6eeef66adf354 Mon Sep 17 00:00:00 2001 From: Yaniv Rosner Date: Tue, 2 Aug 2011 22:59:32 +0000 Subject: [PATCH 231/439] bnx2x: Fix BCM578xx-B0 MDIO access Fix MDIO access to Warpcore on new chip version of 578xx. Signed-off-by: Yaniv Rosner Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_link.c | 16 ++++++++++++++-- drivers/net/bnx2x/bnx2x_link.h | 1 + drivers/net/bnx2x/bnx2x_reg.h | 4 +++- 3 files changed, 18 insertions(+), 3 deletions(-) diff --git a/drivers/net/bnx2x/bnx2x_link.c b/drivers/net/bnx2x/bnx2x_link.c index 8e68e1582ec2..45255bdc5136 100644 --- a/drivers/net/bnx2x/bnx2x_link.c +++ b/drivers/net/bnx2x/bnx2x_link.c @@ -2953,7 +2953,9 @@ static int bnx2x_cl45_read(struct bnx2x *bp, struct bnx2x_phy *phy, u32 val; u16 i; int rc = 0; - + if (phy->flags & FLAGS_MDC_MDIO_WA_B0) + bnx2x_bits_en(bp, phy->mdio_ctrl + EMAC_REG_EMAC_MDIO_STATUS, + EMAC_MDIO_STATUS_10MB); /* address */ val = ((phy->addr << 21) | (devad << 16) | reg | EMAC_MDIO_COMM_COMMAND_ADDRESS | @@ -3007,6 +3009,9 @@ static int bnx2x_cl45_read(struct bnx2x *bp, struct bnx2x_phy *phy, } } + if (phy->flags & FLAGS_MDC_MDIO_WA_B0) + bnx2x_bits_dis(bp, phy->mdio_ctrl + EMAC_REG_EMAC_MDIO_STATUS, + EMAC_MDIO_STATUS_10MB); return rc; } @@ -3016,6 +3021,9 @@ static int bnx2x_cl45_write(struct bnx2x *bp, struct bnx2x_phy *phy, u32 tmp; u8 i; int rc = 0; + if (phy->flags & FLAGS_MDC_MDIO_WA_B0) + bnx2x_bits_en(bp, phy->mdio_ctrl + EMAC_REG_EMAC_MDIO_STATUS, + EMAC_MDIO_STATUS_10MB); /* address */ @@ -3069,7 +3077,9 @@ static int bnx2x_cl45_write(struct bnx2x *bp, struct bnx2x_phy *phy, bnx2x_cl45_read(bp, phy, devad, 0xf, &temp_val); } } - + if (phy->flags & FLAGS_MDC_MDIO_WA_B0) + bnx2x_bits_dis(bp, phy->mdio_ctrl + EMAC_REG_EMAC_MDIO_STATUS, + EMAC_MDIO_STATUS_10MB); return rc; } @@ -11118,6 +11128,8 @@ static int bnx2x_populate_int_phy(struct bnx2x *bp, u32 shmem_base, u8 port, */ if (CHIP_REV(bp) == CHIP_REV_Ax) phy->flags |= FLAGS_MDC_MDIO_WA; + else + phy->flags |= FLAGS_MDC_MDIO_WA_B0; } else { switch (switch_cfg) { case SWITCH_CFG_1G: diff --git a/drivers/net/bnx2x/bnx2x_link.h b/drivers/net/bnx2x/bnx2x_link.h index 7ee6b51f4fb8..c12db6da213e 100644 --- a/drivers/net/bnx2x/bnx2x_link.h +++ b/drivers/net/bnx2x/bnx2x_link.h @@ -145,6 +145,7 @@ struct bnx2x_phy { #define FLAGS_SFP_NOT_APPROVED (1<<7) #define FLAGS_MDC_MDIO_WA (1<<8) #define FLAGS_DUMMY_READ (1<<9) +#define FLAGS_MDC_MDIO_WA_B0 (1<<10) #define FLAGS_TX_ERROR_CHECK (1<<12) /* preemphasis values for the rx side */ diff --git a/drivers/net/bnx2x/bnx2x_reg.h b/drivers/net/bnx2x/bnx2x_reg.h index d84642aca450..27b5ecb11830 100644 --- a/drivers/net/bnx2x/bnx2x_reg.h +++ b/drivers/net/bnx2x/bnx2x_reg.h @@ -5624,8 +5624,9 @@ #define EMAC_MDIO_COMM_START_BUSY (1L<<29) #define EMAC_MDIO_MODE_AUTO_POLL (1L<<4) #define EMAC_MDIO_MODE_CLAUSE_45 (1L<<31) -#define EMAC_MDIO_MODE_CLOCK_CNT (0x3fL<<16) +#define EMAC_MDIO_MODE_CLOCK_CNT (0x3ffL<<16) #define EMAC_MDIO_MODE_CLOCK_CNT_BITSHIFT 16 +#define EMAC_MDIO_STATUS_10MB (1L<<1) #define EMAC_MODE_25G_MODE (1L<<5) #define EMAC_MODE_HALF_DUPLEX (1L<<1) #define EMAC_MODE_PORT_GMII (2L<<2) @@ -5636,6 +5637,7 @@ #define EMAC_REG_EMAC_MAC_MATCH 0x10 #define EMAC_REG_EMAC_MDIO_COMM 0xac #define EMAC_REG_EMAC_MDIO_MODE 0xb4 +#define EMAC_REG_EMAC_MDIO_STATUS 0xb0 #define EMAC_REG_EMAC_MODE 0x0 #define EMAC_REG_EMAC_RX_MODE 0xc8 #define EMAC_REG_EMAC_RX_MTU_SIZE 0x9c From 793bd450370bf85cd63cccaff5e2f1a62908a52f Mon Sep 17 00:00:00 2001 From: Yaniv Rosner Date: Tue, 2 Aug 2011 22:59:40 +0000 Subject: [PATCH 232/439] bnx2x: Fix LED behavior This fix resolve two problems seen regarding LED: 1. LED doesn't flash during port identification. 2. Traffic LED sometimes do not blink. Signed-off-by: Yaniv Rosner Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_link.c | 24 +++++++++++++++++------- 1 file changed, 17 insertions(+), 7 deletions(-) diff --git a/drivers/net/bnx2x/bnx2x_link.c b/drivers/net/bnx2x/bnx2x_link.c index 45255bdc5136..d8f6eff2d6b4 100644 --- a/drivers/net/bnx2x/bnx2x_link.c +++ b/drivers/net/bnx2x/bnx2x_link.c @@ -5922,20 +5922,30 @@ int bnx2x_set_led(struct link_params *params, tmp = EMAC_RD(bp, EMAC_REG_EMAC_LED); EMAC_WR(bp, EMAC_REG_EMAC_LED, (tmp | EMAC_LED_OVERRIDE)); - return rc; + /* + * return here without enabling traffic + * LED blink andsetting rate in ON mode. + * In oper mode, enabling LED blink + * and setting rate is needed. + */ + if (mode == LED_MODE_ON) + return rc; } - } else if (SINGLE_MEDIA_DIRECT(params) && - (CHIP_IS_E1x(bp) || - CHIP_IS_E2(bp))) { + } else if (SINGLE_MEDIA_DIRECT(params)) { /* * This is a work-around for HW issue found when link * is up in CL73 */ - REG_WR(bp, NIG_REG_LED_MODE_P0 + port*4, 0); REG_WR(bp, NIG_REG_LED_10G_P0 + port*4, 1); - } else { + if (CHIP_IS_E1x(bp) || + CHIP_IS_E2(bp) || + (mode == LED_MODE_ON)) + REG_WR(bp, NIG_REG_LED_MODE_P0 + port*4, 0); + else + REG_WR(bp, NIG_REG_LED_MODE_P0 + port*4, + hw_led_mode); + } else REG_WR(bp, NIG_REG_LED_MODE_P0 + port*4, hw_led_mode); - } REG_WR(bp, NIG_REG_LED_CONTROL_OVERRIDE_TRAFFIC_P0 + port*4, 0); /* Set blinking rate to ~15.9Hz */ From 19af03a3c8cb1e07e31c070dfde9fac2e5e7796c Mon Sep 17 00:00:00 2001 From: Yaniv Rosner Date: Tue, 2 Aug 2011 22:59:47 +0000 Subject: [PATCH 233/439] bnx2x: Fix link issue with DAC over 578xx Fix no-link issue on BCM578xx when direct attached cable is connected since Warpcore microcode restart was missing to re-read the new mode. Signed-off-by: Yaniv Rosner Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_link.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/bnx2x/bnx2x_link.c b/drivers/net/bnx2x/bnx2x_link.c index d8f6eff2d6b4..6a3fdf8ace8d 100644 --- a/drivers/net/bnx2x/bnx2x_link.c +++ b/drivers/net/bnx2x/bnx2x_link.c @@ -8029,6 +8029,9 @@ static void bnx2x_warpcore_set_limiting_mode(struct link_params *params, bnx2x_cl45_read(bp, phy, MDIO_WC_DEVAD, MDIO_WC_REG_UC_INFO_B1_FIRMWARE_MODE, &val); + /* Restart microcode to re-read the new mode */ + bnx2x_warpcore_reset_lane(bp, phy, 1); + bnx2x_warpcore_reset_lane(bp, phy, 0); } From fd38f73eb936f9d9f28e4f7ff598cc0780e09424 Mon Sep 17 00:00:00 2001 From: Yaniv Rosner Date: Tue, 2 Aug 2011 22:59:53 +0000 Subject: [PATCH 234/439] bnx2x: Fix BCM84833 link BCM84833 fail to link due to incorrect auto-negotiation setting. Signed-off-by: Yaniv Rosner Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_link.c | 21 ++++++++++++--------- 1 file changed, 12 insertions(+), 9 deletions(-) diff --git a/drivers/net/bnx2x/bnx2x_link.c b/drivers/net/bnx2x/bnx2x_link.c index 6a3fdf8ace8d..cf8b8d13ab6c 100644 --- a/drivers/net/bnx2x/bnx2x_link.c +++ b/drivers/net/bnx2x/bnx2x_link.c @@ -9268,7 +9268,13 @@ static int bnx2x_848xx_cmn_config_init(struct bnx2x_phy *phy, if (phy->req_duplex == DUPLEX_FULL) autoneg_val |= (1<<8); - bnx2x_cl45_write(bp, phy, + /* + * Always write this if this is not 84833. + * For 84833, write it only when it's a forced speed. + */ + if ((phy->type != PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM84833) || + ((autoneg_val & (1<<12)) == 0)) + bnx2x_cl45_write(bp, phy, MDIO_AN_DEVAD, MDIO_AN_REG_8481_LEGACY_MII_CTRL, autoneg_val); @@ -9282,13 +9288,12 @@ static int bnx2x_848xx_cmn_config_init(struct bnx2x_phy *phy, bnx2x_cl45_write(bp, phy, MDIO_AN_DEVAD, MDIO_AN_REG_CTRL, 0x3200); - } else if (phy->req_line_speed != SPEED_10 && - phy->req_line_speed != SPEED_100) { + } else bnx2x_cl45_write(bp, phy, MDIO_AN_DEVAD, MDIO_AN_REG_8481_10GBASE_T_AN_CTRL, 1); - } + /* Save spirom version */ bnx2x_save_848xx_spirom_version(phy, params); @@ -9781,11 +9786,9 @@ static void bnx2x_848x3_link_reset(struct bnx2x_phy *phy, bnx2x_cl45_read(bp, phy, MDIO_CTL_DEVAD, 0x400f, &val16); - /* Put to low power mode on newer FW */ - if ((val16 & 0x303f) > 0x1009) - bnx2x_cl45_write(bp, phy, - MDIO_PMA_DEVAD, - MDIO_PMA_REG_CTRL, 0x800); + bnx2x_cl45_write(bp, phy, + MDIO_PMA_DEVAD, + MDIO_PMA_REG_CTRL, 0x800); } } From d2059a061164120a1e44a0ca46fe08044d6d7c2d Mon Sep 17 00:00:00 2001 From: Yaniv Rosner Date: Tue, 2 Aug 2011 23:00:00 +0000 Subject: [PATCH 235/439] bnx2x: Fix BCM54618se invalid link indication After resetting BCM54618se, link partner would still see link since the PHY wasn't put into low-power state. Signed-off-by: Yaniv Rosner Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_link.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/net/bnx2x/bnx2x_link.c b/drivers/net/bnx2x/bnx2x_link.c index cf8b8d13ab6c..5e6f3513862c 100644 --- a/drivers/net/bnx2x/bnx2x_link.c +++ b/drivers/net/bnx2x/bnx2x_link.c @@ -10219,8 +10219,15 @@ static void bnx2x_54618se_link_reset(struct bnx2x_phy *phy, u32 cfg_pin; u8 port; - /* This works with E3 only, no need to check the chip - before determining the port. */ + /* + * In case of no EPIO routed to reset the GPHY, put it + * in low power mode. + */ + bnx2x_cl22_write(bp, phy, MDIO_PMA_REG_CTRL, 0x800); + /* + * This works with E3 only, no need to check the chip + * before determining the port. + */ port = params->port; cfg_pin = (REG_RD(bp, params->shmem_base + offsetof(struct shmem_region, From afad009ad76ece72a3c9629bbc08f14459b9bba7 Mon Sep 17 00:00:00 2001 From: Yaniv Rosner Date: Tue, 2 Aug 2011 23:00:06 +0000 Subject: [PATCH 236/439] bnx2x: Fix BCM578xx MAC test Signed-off-by: Yaniv Rosner Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_link.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/drivers/net/bnx2x/bnx2x_link.c b/drivers/net/bnx2x/bnx2x_link.c index 5e6f3513862c..01fb92c37a9f 100644 --- a/drivers/net/bnx2x/bnx2x_link.c +++ b/drivers/net/bnx2x/bnx2x_link.c @@ -11541,13 +11541,12 @@ void bnx2x_init_xmac_loopback(struct link_params *params, * Set WC to loopback mode since link is required to provide clock * to the XMAC in 20G mode */ - if (vars->line_speed == SPEED_20000) { - bnx2x_set_aer_mmd(params, ¶ms->phy[0]); - bnx2x_warpcore_reset_lane(bp, ¶ms->phy[0], 0); - params->phy[INT_PHY].config_loopback( + bnx2x_set_aer_mmd(params, ¶ms->phy[0]); + bnx2x_warpcore_reset_lane(bp, ¶ms->phy[0], 0); + params->phy[INT_PHY].config_loopback( ¶ms->phy[INT_PHY], params); - } + bnx2x_xmac_enable(params, vars, 1); REG_WR(bp, NIG_REG_EGRESS_DRAIN0_MODE + params->port*4, 0); } From 28f4881cbf9ce285edfc245a8990af36d21c062f Mon Sep 17 00:00:00 2001 From: Yaniv Rosner Date: Tue, 2 Aug 2011 23:00:12 +0000 Subject: [PATCH 237/439] bnx2x: Clear MDIO access warning during first driver load Signed-off-by: Yaniv Rosner Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x_link.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/net/bnx2x/bnx2x_link.c b/drivers/net/bnx2x/bnx2x_link.c index 01fb92c37a9f..d45b1555a602 100644 --- a/drivers/net/bnx2x/bnx2x_link.c +++ b/drivers/net/bnx2x/bnx2x_link.c @@ -11724,12 +11724,16 @@ int bnx2x_link_reset(struct link_params *params, struct link_vars *vars, bnx2x_set_led(params, vars, LED_MODE_OFF, 0); if (reset_ext_phy) { + bnx2x_set_mdio_clk(bp, params->chip_id, port); for (phy_index = EXT_PHY1; phy_index < params->num_phys; phy_index++) { - if (params->phy[phy_index].link_reset) + if (params->phy[phy_index].link_reset) { + bnx2x_set_aer_mmd(params, + ¶ms->phy[phy_index]); params->phy[phy_index].link_reset( ¶ms->phy[phy_index], params); + } if (params->phy[phy_index].flags & FLAGS_REARM_LATCH_SIGNAL) clear_latch_ind = 1; From f2c31e32b378a6653f8de606149d963baf11d7d3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 29 Jul 2011 19:00:53 +0000 Subject: [PATCH 238/439] net: fix NULL dereferences in check_peer_redir() Gergely Kalman reported crashes in check_peer_redir(). It appears commit f39925dbde778 (ipv4: Cache learned redirect information in inetpeer.) added a race, leading to possible NULL ptr dereference. Since we can now change dst neighbour, we should make sure a reader can safely use a neighbour. Add RCU protection to dst neighbour, and make sure check_peer_redir() can be called safely by different cpus in parallel. As neighbours are already freed after one RCU grace period, this patch should not add typical RCU penalty (cache cold effects) Many thanks to Gergely for providing a pretty report pointing to the bug. Reported-by: Gergely Kalman Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/dst.h | 17 +++++++++++++---- net/ipv4/ip_output.c | 10 ++++++++-- net/ipv4/route.c | 14 ++++++++------ net/ipv6/addrconf.c | 2 +- net/ipv6/ip6_fib.c | 2 +- net/ipv6/ip6_output.c | 13 +++++++++++-- net/ipv6/route.c | 35 +++++++++++++++++++++++++---------- 7 files changed, 67 insertions(+), 26 deletions(-) diff --git a/include/net/dst.h b/include/net/dst.h index 29e255796ce1..13d507d69ddb 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -37,7 +37,7 @@ struct dst_entry { unsigned long _metrics; unsigned long expires; struct dst_entry *path; - struct neighbour *_neighbour; + struct neighbour __rcu *_neighbour; #ifdef CONFIG_XFRM struct xfrm_state *xfrm; #else @@ -88,12 +88,17 @@ struct dst_entry { static inline struct neighbour *dst_get_neighbour(struct dst_entry *dst) { - return dst->_neighbour; + return rcu_dereference(dst->_neighbour); +} + +static inline struct neighbour *dst_get_neighbour_raw(struct dst_entry *dst) +{ + return rcu_dereference_raw(dst->_neighbour); } static inline void dst_set_neighbour(struct dst_entry *dst, struct neighbour *neigh) { - dst->_neighbour = neigh; + rcu_assign_pointer(dst->_neighbour, neigh); } extern u32 *dst_cow_metrics_generic(struct dst_entry *dst, unsigned long old); @@ -382,8 +387,12 @@ static inline void dst_rcu_free(struct rcu_head *head) static inline void dst_confirm(struct dst_entry *dst) { if (dst) { - struct neighbour *n = dst_get_neighbour(dst); + struct neighbour *n; + + rcu_read_lock(); + n = dst_get_neighbour(dst); neigh_confirm(n); + rcu_read_unlock(); } } diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index ccaaa851ab42..77d3eded665a 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -204,9 +204,15 @@ static inline int ip_finish_output2(struct sk_buff *skb) skb = skb2; } + rcu_read_lock(); neigh = dst_get_neighbour(dst); - if (neigh) - return neigh_output(neigh, skb); + if (neigh) { + int res = neigh_output(neigh, skb); + + rcu_read_unlock(); + return res; + } + rcu_read_unlock(); if (net_ratelimit()) printk(KERN_DEBUG "ip_finish_output2: No header cache and no neighbour!\n"); diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 1730689f560e..6afc4eb50591 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1628,16 +1628,18 @@ static int check_peer_redir(struct dst_entry *dst, struct inet_peer *peer) { struct rtable *rt = (struct rtable *) dst; __be32 orig_gw = rt->rt_gateway; - struct neighbour *n; + struct neighbour *n, *old_n; dst_confirm(&rt->dst); - neigh_release(dst_get_neighbour(&rt->dst)); - dst_set_neighbour(&rt->dst, NULL); - rt->rt_gateway = peer->redirect_learned.a4; - rt_bind_neighbour(rt); - n = dst_get_neighbour(&rt->dst); + + n = ipv4_neigh_lookup(&rt->dst, &rt->rt_gateway); + if (IS_ERR(n)) + return PTR_ERR(n); + old_n = xchg(&rt->dst._neighbour, n); + if (old_n) + neigh_release(old_n); if (!n || !(n->nud_state & NUD_VALID)) { if (n) neigh_event_send(n, NULL); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index a55500cc0b29..f012ebd87b43 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -656,7 +656,7 @@ ipv6_add_addr(struct inet6_dev *idev, const struct in6_addr *addr, int pfxlen, * layer address of our nexhop router */ - if (dst_get_neighbour(&rt->dst) == NULL) + if (dst_get_neighbour_raw(&rt->dst) == NULL) ifa->flags &= ~IFA_F_OPTIMISTIC; ifa->idev = idev; diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 54a4678955bf..320d91d20ad7 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -1455,7 +1455,7 @@ static int fib6_age(struct rt6_info *rt, void *arg) RT6_TRACE("aging clone %p\n", rt); return -1; } else if ((rt->rt6i_flags & RTF_GATEWAY) && - (!(dst_get_neighbour(&rt->dst)->flags & NTF_ROUTER))) { + (!(dst_get_neighbour_raw(&rt->dst)->flags & NTF_ROUTER))) { RT6_TRACE("purging route %p via non-router but gateway\n", rt); return -1; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 32e5339db0c8..4c882cf4e8a1 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -135,10 +135,15 @@ static int ip6_finish_output2(struct sk_buff *skb) skb->len); } + rcu_read_lock(); neigh = dst_get_neighbour(dst); - if (neigh) - return neigh_output(neigh, skb); + if (neigh) { + int res = neigh_output(neigh, skb); + rcu_read_unlock(); + return res; + } + rcu_read_unlock(); IP6_INC_STATS_BH(dev_net(dst->dev), ip6_dst_idev(dst), IPSTATS_MIB_OUTNOROUTES); kfree_skb(skb); @@ -975,12 +980,14 @@ static int ip6_dst_lookup_tail(struct sock *sk, * dst entry and replace it instead with the * dst entry of the nexthop router */ + rcu_read_lock(); n = dst_get_neighbour(*dst); if (n && !(n->nud_state & NUD_VALID)) { struct inet6_ifaddr *ifp; struct flowi6 fl_gw6; int redirect; + rcu_read_unlock(); ifp = ipv6_get_ifaddr(net, &fl6->saddr, (*dst)->dev, 1); @@ -1000,6 +1007,8 @@ static int ip6_dst_lookup_tail(struct sock *sk, if ((err = (*dst)->error)) goto out_err_release; } + } else { + rcu_read_unlock(); } #endif diff --git a/net/ipv6/route.c b/net/ipv6/route.c index e8987da06667..9e69eb0ec6dd 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -364,7 +364,7 @@ out: #ifdef CONFIG_IPV6_ROUTER_PREF static void rt6_probe(struct rt6_info *rt) { - struct neighbour *neigh = rt ? dst_get_neighbour(&rt->dst) : NULL; + struct neighbour *neigh; /* * Okay, this does not seem to be appropriate * for now, however, we need to check if it @@ -373,8 +373,10 @@ static void rt6_probe(struct rt6_info *rt) * Router Reachability Probe MUST be rate-limited * to no more than one per minute. */ + rcu_read_lock(); + neigh = rt ? dst_get_neighbour(&rt->dst) : NULL; if (!neigh || (neigh->nud_state & NUD_VALID)) - return; + goto out; read_lock_bh(&neigh->lock); if (!(neigh->nud_state & NUD_VALID) && time_after(jiffies, neigh->updated + rt->rt6i_idev->cnf.rtr_probe_interval)) { @@ -387,8 +389,11 @@ static void rt6_probe(struct rt6_info *rt) target = (struct in6_addr *)&neigh->primary_key; addrconf_addr_solict_mult(target, &mcaddr); ndisc_send_ns(rt->rt6i_dev, NULL, target, &mcaddr, NULL); - } else + } else { read_unlock_bh(&neigh->lock); + } +out: + rcu_read_unlock(); } #else static inline void rt6_probe(struct rt6_info *rt) @@ -412,8 +417,11 @@ static inline int rt6_check_dev(struct rt6_info *rt, int oif) static inline int rt6_check_neigh(struct rt6_info *rt) { - struct neighbour *neigh = dst_get_neighbour(&rt->dst); + struct neighbour *neigh; int m; + + rcu_read_lock(); + neigh = dst_get_neighbour(&rt->dst); if (rt->rt6i_flags & RTF_NONEXTHOP || !(rt->rt6i_flags & RTF_GATEWAY)) m = 1; @@ -430,6 +438,7 @@ static inline int rt6_check_neigh(struct rt6_info *rt) read_unlock_bh(&neigh->lock); } else m = 0; + rcu_read_unlock(); return m; } @@ -769,7 +778,7 @@ static struct rt6_info *rt6_alloc_clone(struct rt6_info *ort, rt->rt6i_dst.plen = 128; rt->rt6i_flags |= RTF_CACHE; rt->dst.flags |= DST_HOST; - dst_set_neighbour(&rt->dst, neigh_clone(dst_get_neighbour(&ort->dst))); + dst_set_neighbour(&rt->dst, neigh_clone(dst_get_neighbour_raw(&ort->dst))); } return rt; } @@ -803,7 +812,7 @@ restart: dst_hold(&rt->dst); read_unlock_bh(&table->tb6_lock); - if (!dst_get_neighbour(&rt->dst) && !(rt->rt6i_flags & RTF_NONEXTHOP)) + if (!dst_get_neighbour_raw(&rt->dst) && !(rt->rt6i_flags & RTF_NONEXTHOP)) nrt = rt6_alloc_cow(rt, &fl6->daddr, &fl6->saddr); else if (!(rt->dst.flags & DST_HOST)) nrt = rt6_alloc_clone(rt, &fl6->daddr); @@ -1587,7 +1596,7 @@ void rt6_redirect(const struct in6_addr *dest, const struct in6_addr *src, dst_confirm(&rt->dst); /* Duplicate redirect: silently ignore. */ - if (neigh == dst_get_neighbour(&rt->dst)) + if (neigh == dst_get_neighbour_raw(&rt->dst)) goto out; nrt = ip6_rt_copy(rt, dest); @@ -1682,7 +1691,7 @@ again: 1. It is connected route. Action: COW 2. It is gatewayed route or NONEXTHOP route. Action: clone it. */ - if (!dst_get_neighbour(&rt->dst) && !(rt->rt6i_flags & RTF_NONEXTHOP)) + if (!dst_get_neighbour_raw(&rt->dst) && !(rt->rt6i_flags & RTF_NONEXTHOP)) nrt = rt6_alloc_cow(rt, daddr, saddr); else nrt = rt6_alloc_clone(rt, daddr); @@ -2326,6 +2335,7 @@ static int rt6_fill_node(struct net *net, struct nlmsghdr *nlh; long expires; u32 table; + struct neighbour *n; if (prefix) { /* user wants prefix routes only */ if (!(rt->rt6i_flags & RTF_PREFIX_RT)) { @@ -2414,8 +2424,11 @@ static int rt6_fill_node(struct net *net, if (rtnetlink_put_metrics(skb, dst_metrics_ptr(&rt->dst)) < 0) goto nla_put_failure; - if (dst_get_neighbour(&rt->dst)) - NLA_PUT(skb, RTA_GATEWAY, 16, &dst_get_neighbour(&rt->dst)->primary_key); + rcu_read_lock(); + n = dst_get_neighbour(&rt->dst); + if (n) + NLA_PUT(skb, RTA_GATEWAY, 16, &n->primary_key); + rcu_read_unlock(); if (rt->dst.dev) NLA_PUT_U32(skb, RTA_OIF, rt->rt6i_dev->ifindex); @@ -2608,12 +2621,14 @@ static int rt6_info_route(struct rt6_info *rt, void *p_arg) #else seq_puts(m, "00000000000000000000000000000000 00 "); #endif + rcu_read_lock(); n = dst_get_neighbour(&rt->dst); if (n) { seq_printf(m, "%pi6", n->primary_key); } else { seq_puts(m, "00000000000000000000000000000000"); } + rcu_read_unlock(); seq_printf(m, " %08x %08x %08x %08x %8s\n", rt->rt6i_metric, atomic_read(&rt->dst.__refcnt), rt->dst.__use, rt->rt6i_flags, From 750f463a749e28464151ad26938d11b07b1c43cb Mon Sep 17 00:00:00 2001 From: Shawn Guo Date: Wed, 3 Aug 2011 11:28:14 +0100 Subject: [PATCH 239/439] dt: add of_alias_scan and of_alias_get_id The patch adds function of_alias_scan to populate a global lookup table with the properties of 'aliases' node and function of_alias_get_id for drivers to find alias id from the lookup table. Signed-off-by: Shawn Guo [grant.likely: add locking and rework parse loop] Signed-off-by: Grant Likely --- drivers/of/base.c | 130 +++++++++++++++++++++++++++++++++++++++++ drivers/of/fdt.c | 4 +- include/linux/of.h | 8 +++ include/linux/of_fdt.h | 1 - 4 files changed, 141 insertions(+), 2 deletions(-) diff --git a/drivers/of/base.c b/drivers/of/base.c index 3ff22e32b602..fb28b5af733b 100644 --- a/drivers/of/base.c +++ b/drivers/of/base.c @@ -17,14 +17,39 @@ * as published by the Free Software Foundation; either version * 2 of the License, or (at your option) any later version. */ +#include #include #include #include #include #include +/** + * struct alias_prop - Alias property in 'aliases' node + * @link: List node to link the structure in aliases_lookup list + * @alias: Alias property name + * @np: Pointer to device_node that the alias stands for + * @id: Index value from end of alias name + * @stem: Alias string without the index + * + * The structure represents one alias property of 'aliases' node as + * an entry in aliases_lookup list. + */ +struct alias_prop { + struct list_head link; + const char *alias; + struct device_node *np; + int id; + char stem[0]; +}; + +static LIST_HEAD(aliases_lookup); + struct device_node *allnodes; struct device_node *of_chosen; +struct device_node *of_aliases; + +static DEFINE_MUTEX(of_aliases_mutex); /* use when traversing tree through the allnext, child, sibling, * or parent members of struct device_node. @@ -988,3 +1013,108 @@ out_unlock: } #endif /* defined(CONFIG_OF_DYNAMIC) */ +static void of_alias_add(struct alias_prop *ap, struct device_node *np, + int id, const char *stem, int stem_len) +{ + ap->id = id; + ap->np = np; + strncpy(ap->stem, stem, stem_len); + ap->stem[stem_len] = 0; + list_add_tail(&ap->link, &aliases_lookup); + pr_debug("adding DT alias:%s: stem=%s id=%i node=%s\n", + ap->alias, ap->stem, ap->id, np ? np->full_name : NULL); +} + +/** + * of_alias_scan() - Scan all properties of 'aliases' node + * + * The function scans all the properties of 'aliases' node and populate + * the global lookup table with the properties. It returns the + * number of alias_prop found, or error code in error case. + */ +__init void of_alias_scan(void) +{ + struct property *pp; + + if (!of_aliases) + return; + + for_each_property(pp, of_aliases->properties) { + const char *start = pp->name; + const char *end = start + strlen(start); + struct device_node *np; + struct alias_prop *ap; + int id, len; + + /* Skip those we do not want to proceed */ + if (!strcmp(pp->name, "name") || + !strcmp(pp->name, "phandle") || + !strcmp(pp->name, "linux,phandle")) + continue; + + np = of_find_node_by_path(pp->value); + if (!np) + continue; + + /* walk alias backwards to extract the id and 'stem' string */ + while (isdigit(*(end-1)) && end > start) + end--; + len = end - start; + id = strlen(end) ? simple_strtoul(end, NULL, 10) : -1; + + /* Allocate an alias_prop with enough space for the stem */ + ap = early_init_dt_alloc_memory_arch(sizeof(*ap) + len + 1, 4); + if (!ap) + continue; + ap->alias = start; + of_alias_add(ap, np, id, start, len); + } +} + +/** + * of_alias_get_id() - Get alias id for the given device_node + * @np: Pointer to the given device_node + * @stem: Alias stem of the given device_node + * + * The function travels the lookup table to get alias id for the given + * device_node and alias stem. It returns the alias id if find it. + * If not, dynamically creates one in the lookup table and returns it, + * or returns error code if fail to create. + */ +int of_alias_get_id(struct device_node *np, const char *stem) +{ + struct alias_prop *app; + int id = 0; + bool found = false; + + mutex_lock(&of_aliases_mutex); + list_for_each_entry(app, &aliases_lookup, link) { + if (strcmp(app->stem, stem) != 0) + continue; + + if (np == app->np) { + found = true; + id = app->id; + break; + } + + if (id <= app->id) + id = app->id + 1; + } + + /* If an id is not found, then allocate a new one */ + if (!found) { + app = kzalloc(sizeof(*app) + strlen(stem) + 1, 4); + if (!app) { + id = -ENODEV; + goto out; + } + of_alias_add(app, np, id, stem, strlen(stem)); + } + + out: + mutex_unlock(&of_aliases_mutex); + + return id; +} +EXPORT_SYMBOL_GPL(of_alias_get_id); diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c index 65200af29c52..13d6d3a96b31 100644 --- a/drivers/of/fdt.c +++ b/drivers/of/fdt.c @@ -707,10 +707,12 @@ void __init unflatten_device_tree(void) __unflatten_device_tree(initial_boot_params, &allnodes, early_init_dt_alloc_memory_arch); - /* Get pointer to OF "/chosen" node for use everywhere */ + /* Get pointer to "/chosen" and "/aliasas" nodes for use everywhere */ of_chosen = of_find_node_by_path("/chosen"); if (of_chosen == NULL) of_chosen = of_find_node_by_path("/chosen@0"); + of_aliases = of_find_node_by_path("/aliases"); + of_alias_scan(); } #endif /* CONFIG_OF_EARLY_FLATTREE */ diff --git a/include/linux/of.h b/include/linux/of.h index 0085bb01c041..bc3dc6399547 100644 --- a/include/linux/of.h +++ b/include/linux/of.h @@ -68,6 +68,7 @@ struct device_node { /* Pointer for first entry in chain of all nodes. */ extern struct device_node *allnodes; extern struct device_node *of_chosen; +extern struct device_node *of_aliases; extern rwlock_t devtree_lock; static inline bool of_have_populated_dt(void) @@ -209,6 +210,9 @@ extern int of_device_is_available(const struct device_node *device); extern const void *of_get_property(const struct device_node *node, const char *name, int *lenp); +#define for_each_property(pp, properties) \ + for (pp = properties; pp != NULL; pp = pp->next) + extern int of_n_addr_cells(struct device_node *np); extern int of_n_size_cells(struct device_node *np); extern const struct of_device_id *of_match_node( @@ -221,6 +225,10 @@ extern int of_parse_phandles_with_args(struct device_node *np, const char *list_name, const char *cells_name, int index, struct device_node **out_node, const void **out_args); +extern void *early_init_dt_alloc_memory_arch(u64 size, u64 align); +extern void of_alias_scan(void); +extern int of_alias_get_id(struct device_node *np, const char *stem); + extern int of_machine_is_compatible(const char *compat); extern int prom_add_property(struct device_node* np, struct property* prop); diff --git a/include/linux/of_fdt.h b/include/linux/of_fdt.h index c84d900fbbb3..b74b74ffe0e7 100644 --- a/include/linux/of_fdt.h +++ b/include/linux/of_fdt.h @@ -97,7 +97,6 @@ extern void early_init_dt_check_for_initrd(unsigned long node); extern int early_init_dt_scan_memory(unsigned long node, const char *uname, int depth, void *data); extern void early_init_dt_add_memory_arch(u64 base, u64 size); -extern void * early_init_dt_alloc_memory_arch(u64 size, u64 align); extern u64 dt_mem_next_cell(int s, __be32 **cellp); /* From be8d97a540cec5d272b1e08e27741a3c3aa38f66 Mon Sep 17 00:00:00 2001 From: Jan Glauber Date: Wed, 3 Aug 2011 16:44:17 +0200 Subject: [PATCH 240/439] [S390] qdio: 2nd stage retry on SIGA-W busy conditions The SIGA-W may return with the busy bit set which means the device was blocked. The busy loop which retries the SIGA-W for 100us may not be long enough when running under a heavily loaded hypervisor. Extend the retry mechanism by adding a longer second stage which retries the SIGA-W for up to 10s. In difference to the first retry loop the second stage is using mdelay to stop the cpu between the retries and thereby avoid additional preassure in on the hypervisor. If the second stage retry is successfull a device reset is avoided. Signed-off-by: Jan Glauber Signed-off-by: Heiko Carstens --- drivers/s390/cio/qdio.h | 2 ++ drivers/s390/cio/qdio_main.c | 21 ++++++++++++++++++--- 2 files changed, 20 insertions(+), 3 deletions(-) diff --git a/drivers/s390/cio/qdio.h b/drivers/s390/cio/qdio.h index 7bc643f3f5ab..e5c966462c5a 100644 --- a/drivers/s390/cio/qdio.h +++ b/drivers/s390/cio/qdio.h @@ -14,6 +14,8 @@ #include "chsc.h" #define QDIO_BUSY_BIT_PATIENCE (100 << 12) /* 100 microseconds */ +#define QDIO_BUSY_BIT_RETRY_DELAY 10 /* 10 milliseconds */ +#define QDIO_BUSY_BIT_RETRIES 1000 /* = 10s retry time */ #define QDIO_INPUT_THRESHOLD (500 << 12) /* 500 microseconds */ /* diff --git a/drivers/s390/cio/qdio_main.c b/drivers/s390/cio/qdio_main.c index e58169c32474..288c9140290e 100644 --- a/drivers/s390/cio/qdio_main.c +++ b/drivers/s390/cio/qdio_main.c @@ -313,7 +313,7 @@ static int qdio_siga_output(struct qdio_q *q, unsigned int *busy_bit) unsigned long schid = *((u32 *) &q->irq_ptr->schid); unsigned int fc = QDIO_SIGA_WRITE; u64 start_time = 0; - int cc; + int retries = 0, cc; if (is_qebsm(q)) { schid = q->irq_ptr->sch_token; @@ -325,6 +325,7 @@ again: /* hipersocket busy condition */ if (unlikely(*busy_bit)) { WARN_ON(queue_type(q) != QDIO_IQDIO_QFMT || cc != 2); + retries++; if (!start_time) { start_time = get_clock(); @@ -333,6 +334,11 @@ again: if ((get_clock() - start_time) < QDIO_BUSY_BIT_PATIENCE) goto again; } + if (retries) { + DBF_DEV_EVENT(DBF_WARN, q->irq_ptr, + "%4x cc2 BB1:%1d", SCH_NO(q), q->nr); + DBF_DEV_EVENT(DBF_WARN, q->irq_ptr, "count:%u", retries); + } return cc; } @@ -728,13 +734,14 @@ static inline int qdio_outbound_q_moved(struct qdio_q *q) static int qdio_kick_outbound_q(struct qdio_q *q) { + int retries = 0, cc; unsigned int busy_bit; - int cc; if (!need_siga_out(q)) return 0; DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "siga-w:%1d", q->nr); +retry: qperf_inc(q, siga_write); cc = qdio_siga_output(q, &busy_bit); @@ -743,7 +750,11 @@ static int qdio_kick_outbound_q(struct qdio_q *q) break; case 2: if (busy_bit) { - DBF_ERROR("%4x cc2 REP:%1d", SCH_NO(q), q->nr); + while (++retries < QDIO_BUSY_BIT_RETRIES) { + mdelay(QDIO_BUSY_BIT_RETRY_DELAY); + goto retry; + } + DBF_ERROR("%4x cc2 BBC:%1d", SCH_NO(q), q->nr); cc |= QDIO_ERROR_SIGA_BUSY; } else DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "siga-w cc2:%1d", q->nr); @@ -753,6 +764,10 @@ static int qdio_kick_outbound_q(struct qdio_q *q) DBF_ERROR("%4x SIGA-W:%1d", SCH_NO(q), cc); break; } + if (retries) { + DBF_ERROR("%4x cc2 BB2:%1d", SCH_NO(q), q->nr); + DBF_ERROR("count:%u", retries); + } return cc; } From 944291de33b26a8b403f13f5eb0cc51fb982aa1e Mon Sep 17 00:00:00 2001 From: Jan Glauber Date: Wed, 3 Aug 2011 16:44:18 +0200 Subject: [PATCH 241/439] [S390] missing return in page_table_alloc_pgste MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix the following compile warning for !CONFIG_PGSTE: CC arch/s390/mm/pgtable.o arch/s390/mm/pgtable.c: In function ‘page_table_alloc_pgste’: arch/s390/mm/pgtable.c:531:1: warning: no return statement in function returning non-void [-Wreturn-type] Signed-off-by: Jan Glauber Signed-off-by: Heiko Carstens --- arch/s390/mm/pgtable.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/s390/mm/pgtable.c b/arch/s390/mm/pgtable.c index 2adb23938a7f..4d1f2bce87b3 100644 --- a/arch/s390/mm/pgtable.c +++ b/arch/s390/mm/pgtable.c @@ -528,6 +528,7 @@ static inline void page_table_free_pgste(unsigned long *table) static inline unsigned long *page_table_alloc_pgste(struct mm_struct *mm, unsigned long vmaddr) { + return NULL; } static inline void page_table_free_pgste(unsigned long *table) From 7dd6b3343fdc190712d1620ee8848d25c4c77c33 Mon Sep 17 00:00:00 2001 From: Michael Holzheu Date: Wed, 3 Aug 2011 16:44:19 +0200 Subject: [PATCH 242/439] [S390] Add PSW restart shutdown trigger With this patch a new S390 shutdown trigger "restart" is added. If under z/VM "systerm restart" is entered or under the HMC the "PSW restart" button is pressed, the PSW located at 0 (31 bit) or 0x1a0 (64 bit) bit is loaded. Now we execute do_restart() that processes the restart action that is defined under /sys/firmware/shutdown_actions/on_restart. Currently the following actions are possible: reipl (default), stop, vmcmd, dump, and dump_reipl. Signed-off-by: Michael Holzheu Signed-off-by: Heiko Carstens --- arch/s390/include/asm/lowcore.h | 11 ++++++++-- arch/s390/include/asm/system.h | 1 + arch/s390/kernel/asm-offsets.c | 1 + arch/s390/kernel/entry.S | 28 +++++++++++++++++++++++ arch/s390/kernel/entry64.S | 20 +++++++++++++++++ arch/s390/kernel/ipl.c | 39 ++++++++++++++++++++++++++++++++- arch/s390/kernel/setup.c | 24 +++++++++++++++++++- arch/s390/kernel/smp.c | 11 +++++++++- arch/s390/mm/maccess.c | 16 ++++++++++++++ 9 files changed, 146 insertions(+), 5 deletions(-) diff --git a/arch/s390/include/asm/lowcore.h b/arch/s390/include/asm/lowcore.h index f26280d9e88d..e85c911aabf0 100644 --- a/arch/s390/include/asm/lowcore.h +++ b/arch/s390/include/asm/lowcore.h @@ -18,6 +18,7 @@ void system_call(void); void pgm_check_handler(void); void mcck_int_handler(void); void io_int_handler(void); +void psw_restart_int_handler(void); #ifdef CONFIG_32BIT @@ -150,7 +151,10 @@ struct _lowcore { */ __u32 ipib; /* 0x0e00 */ __u32 ipib_checksum; /* 0x0e04 */ - __u8 pad_0x0e08[0x0f00-0x0e08]; /* 0x0e08 */ + + /* 64 bit save area */ + __u64 save_area_64; /* 0x0e08 */ + __u8 pad_0x0e10[0x0f00-0x0e10]; /* 0x0e10 */ /* Extended facility list */ __u64 stfle_fac_list[32]; /* 0x0f00 */ @@ -286,7 +290,10 @@ struct _lowcore { */ __u64 ipib; /* 0x0e00 */ __u32 ipib_checksum; /* 0x0e08 */ - __u8 pad_0x0e0c[0x0f00-0x0e0c]; /* 0x0e0c */ + + /* 64 bit save area */ + __u64 save_area_64; /* 0x0e0c */ + __u8 pad_0x0e14[0x0f00-0x0e14]; /* 0x0e14 */ /* Extended facility list */ __u64 stfle_fac_list[32]; /* 0x0f00 */ diff --git a/arch/s390/include/asm/system.h b/arch/s390/include/asm/system.h index d382629a0172..6582f69f2389 100644 --- a/arch/s390/include/asm/system.h +++ b/arch/s390/include/asm/system.h @@ -113,6 +113,7 @@ extern void pfault_fini(void); extern void cmma_init(void); extern int memcpy_real(void *, void *, size_t); +extern void copy_to_absolute_zero(void *dest, void *src, size_t count); #define finish_arch_switch(prev) do { \ set_fs(current->thread.mm_segment); \ diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c index 05d8f38734ec..199a537d4f36 100644 --- a/arch/s390/kernel/asm-offsets.c +++ b/arch/s390/kernel/asm-offsets.c @@ -142,6 +142,7 @@ int main(void) DEFINE(__LC_FPREGS_SAVE_AREA, offsetof(struct _lowcore, floating_pt_save_area)); DEFINE(__LC_GPREGS_SAVE_AREA, offsetof(struct _lowcore, gpregs_save_area)); DEFINE(__LC_CREGS_SAVE_AREA, offsetof(struct _lowcore, cregs_save_area)); + DEFINE(__LC_SAVE_AREA_64, offsetof(struct _lowcore, save_area_64)); #ifdef CONFIG_32BIT DEFINE(SAVE_AREA_BASE, offsetof(struct _lowcore, extended_save_area_addr)); #else /* CONFIG_32BIT */ diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index 3eab7cfab07c..02ec8fe7d03f 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -849,6 +849,34 @@ restart_crash: restart_go: #endif +# +# PSW restart interrupt handler +# +ENTRY(psw_restart_int_handler) + st %r15,__LC_SAVE_AREA_64(%r0) # save r15 + basr %r15,0 +0: l %r15,.Lrestart_stack-0b(%r15) # load restart stack + l %r15,0(%r15) + ahi %r15,-SP_SIZE # make room for pt_regs + stm %r0,%r14,SP_R0(%r15) # store gprs %r0-%r14 to stack + mvc SP_R15(4,%r15),__LC_SAVE_AREA_64(%r0)# store saved %r15 to stack + mvc SP_PSW(8,%r15),__LC_RST_OLD_PSW(%r0) # store restart old psw + xc __SF_BACKCHAIN(4,%r15),__SF_BACKCHAIN(%r15) # set backchain to 0 + basr %r14,0 +1: l %r14,.Ldo_restart-1b(%r14) + basr %r14,%r14 + + basr %r14,0 # load disabled wait PSW if +2: lpsw restart_psw_crash-2b(%r14) # do_restart returns + .align 4 +.Ldo_restart: + .long do_restart +.Lrestart_stack: + .long restart_stack + .align 8 +restart_psw_crash: + .long 0x000a0000,0x00000000 + restart_psw_crash + .section .kprobes.text, "ax" #ifdef CONFIG_CHECK_STACK diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S index 7a0fd426ca92..5f729d627cef 100644 --- a/arch/s390/kernel/entry64.S +++ b/arch/s390/kernel/entry64.S @@ -865,6 +865,26 @@ restart_crash: restart_go: #endif +# +# PSW restart interrupt handler +# +ENTRY(psw_restart_int_handler) + stg %r15,__LC_SAVE_AREA_64(%r0) # save r15 + larl %r15,restart_stack # load restart stack + lg %r15,0(%r15) + aghi %r15,-SP_SIZE # make room for pt_regs + stmg %r0,%r14,SP_R0(%r15) # store gprs %r0-%r14 to stack + mvc SP_R15(8,%r15),__LC_SAVE_AREA_64(%r0)# store saved %r15 to stack + mvc SP_PSW(16,%r15),__LC_RST_OLD_PSW(%r0)# store restart old psw + xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) # set backchain to 0 + brasl %r14,do_restart + + larl %r14,restart_psw_crash # load disabled wait PSW if + lpswe 0(%r14) # do_restart returns + .align 8 +restart_psw_crash: + .quad 0x0002000080000000,0x0000000000000000 + restart_psw_crash + .section .kprobes.text, "ax" #ifdef CONFIG_CHECK_STACK diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c index a689070be287..059c59027ac0 100644 --- a/arch/s390/kernel/ipl.c +++ b/arch/s390/kernel/ipl.c @@ -45,11 +45,13 @@ * - halt * - power off * - reipl + * - restart */ #define ON_PANIC_STR "on_panic" #define ON_HALT_STR "on_halt" #define ON_POFF_STR "on_poff" #define ON_REIPL_STR "on_reboot" +#define ON_RESTART_STR "on_restart" struct shutdown_action; struct shutdown_trigger { @@ -1544,17 +1546,20 @@ static char vmcmd_on_reboot[128]; static char vmcmd_on_panic[128]; static char vmcmd_on_halt[128]; static char vmcmd_on_poff[128]; +static char vmcmd_on_restart[128]; DEFINE_IPL_ATTR_STR_RW(vmcmd, on_reboot, "%s\n", "%s\n", vmcmd_on_reboot); DEFINE_IPL_ATTR_STR_RW(vmcmd, on_panic, "%s\n", "%s\n", vmcmd_on_panic); DEFINE_IPL_ATTR_STR_RW(vmcmd, on_halt, "%s\n", "%s\n", vmcmd_on_halt); DEFINE_IPL_ATTR_STR_RW(vmcmd, on_poff, "%s\n", "%s\n", vmcmd_on_poff); +DEFINE_IPL_ATTR_STR_RW(vmcmd, on_restart, "%s\n", "%s\n", vmcmd_on_restart); static struct attribute *vmcmd_attrs[] = { &sys_vmcmd_on_reboot_attr.attr, &sys_vmcmd_on_panic_attr.attr, &sys_vmcmd_on_halt_attr.attr, &sys_vmcmd_on_poff_attr.attr, + &sys_vmcmd_on_restart_attr.attr, NULL, }; @@ -1576,6 +1581,8 @@ static void vmcmd_run(struct shutdown_trigger *trigger) cmd = vmcmd_on_halt; else if (strcmp(trigger->name, ON_POFF_STR) == 0) cmd = vmcmd_on_poff; + else if (strcmp(trigger->name, ON_RESTART_STR) == 0) + cmd = vmcmd_on_restart; else return; @@ -1707,6 +1714,34 @@ static void do_panic(void) stop_run(&on_panic_trigger); } +/* on restart */ + +static struct shutdown_trigger on_restart_trigger = {ON_RESTART_STR, + &reipl_action}; + +static ssize_t on_restart_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + return sprintf(page, "%s\n", on_restart_trigger.action->name); +} + +static ssize_t on_restart_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t len) +{ + return set_trigger(buf, &on_restart_trigger, len); +} + +static struct kobj_attribute on_restart_attr = + __ATTR(on_restart, 0644, on_restart_show, on_restart_store); + +void do_restart(void) +{ + smp_send_stop(); + on_restart_trigger.action->fn(&on_restart_trigger); + stop_run(&on_restart_trigger); +} + /* on halt */ static struct shutdown_trigger on_halt_trigger = {ON_HALT_STR, &stop_action}; @@ -1783,7 +1818,9 @@ static void __init shutdown_triggers_init(void) if (sysfs_create_file(&shutdown_actions_kset->kobj, &on_poff_attr.attr)) goto fail; - + if (sysfs_create_file(&shutdown_actions_kset->kobj, + &on_restart_attr.attr)) + goto fail; return; fail: panic("shutdown_triggers_init failed\n"); diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 0c35dee10b00..18640ff0aa1a 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -346,7 +346,7 @@ setup_lowcore(void) lc = __alloc_bootmem_low(LC_PAGES * PAGE_SIZE, LC_PAGES * PAGE_SIZE, 0); lc->restart_psw.mask = PSW_BASE_BITS | PSW_DEFAULT_KEY; lc->restart_psw.addr = - PSW_ADDR_AMODE | (unsigned long) restart_int_handler; + PSW_ADDR_AMODE | (unsigned long) psw_restart_int_handler; if (user_mode != HOME_SPACE_MODE) lc->restart_psw.mask |= PSW_ASC_HOME; lc->external_new_psw.mask = psw_kernel_bits; @@ -529,6 +529,27 @@ static void __init setup_memory_end(void) memory_end = memory_size; } +void *restart_stack __attribute__((__section__(".data"))); + +/* + * Setup new PSW and allocate stack for PSW restart interrupt + */ +static void __init setup_restart_psw(void) +{ + psw_t psw; + + restart_stack = __alloc_bootmem(ASYNC_SIZE, ASYNC_SIZE, 0); + restart_stack += ASYNC_SIZE; + + /* + * Setup restart PSW for absolute zero lowcore. This is necesary + * if PSW restart is done on an offline CPU that has lowcore zero + */ + psw.mask = PSW_BASE_BITS | PSW_DEFAULT_KEY; + psw.addr = PSW_ADDR_AMODE | (unsigned long) psw_restart_int_handler; + copy_to_absolute_zero(&S390_lowcore.restart_psw, &psw, sizeof(psw)); +} + static void __init setup_memory(void) { @@ -792,6 +813,7 @@ setup_arch(char **cmdline_p) setup_addressing_mode(); setup_memory(); setup_resources(); + setup_restart_psw(); setup_lowcore(); cpu_init(); diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index a6d85c0a7f20..70f4b9604575 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -468,6 +468,11 @@ int __cpuinit start_secondary(void *cpuvoid) ipi_call_lock(); set_cpu_online(smp_processor_id(), true); ipi_call_unlock(); + __ctl_clear_bit(0, 28); /* Disable lowcore protection */ + S390_lowcore.restart_psw.mask = PSW_BASE_BITS | PSW_DEFAULT_KEY; + S390_lowcore.restart_psw.addr = + PSW_ADDR_AMODE | (unsigned long) psw_restart_int_handler; + __ctl_set_bit(0, 28); /* Enable lowcore protection */ /* Switch on interrupts */ local_irq_enable(); /* cpu_idle will call schedule for us */ @@ -507,7 +512,11 @@ static int __cpuinit smp_alloc_lowcore(int cpu) memset((char *)lowcore + 512, 0, sizeof(*lowcore) - 512); lowcore->async_stack = async_stack + ASYNC_SIZE; lowcore->panic_stack = panic_stack + PAGE_SIZE; - + lowcore->restart_psw.mask = PSW_BASE_BITS | PSW_DEFAULT_KEY; + lowcore->restart_psw.addr = + PSW_ADDR_AMODE | (unsigned long) restart_int_handler; + if (user_mode != HOME_SPACE_MODE) + lowcore->restart_psw.mask |= PSW_ASC_HOME; #ifndef CONFIG_64BIT if (MACHINE_HAS_IEEE) { unsigned long save_area; diff --git a/arch/s390/mm/maccess.c b/arch/s390/mm/maccess.c index 51e5cd9b906a..5dbbaa6e594c 100644 --- a/arch/s390/mm/maccess.c +++ b/arch/s390/mm/maccess.c @@ -85,3 +85,19 @@ int memcpy_real(void *dest, void *src, size_t count) arch_local_irq_restore(flags); return rc; } + +/* + * Copy memory to absolute zero + */ +void copy_to_absolute_zero(void *dest, void *src, size_t count) +{ + unsigned long cr0; + + BUG_ON((unsigned long) dest + count >= sizeof(struct _lowcore)); + preempt_disable(); + __ctl_store(cr0, 0, 0); + __ctl_clear_bit(0, 28); /* disable lowcore protection */ + memcpy_real(dest + store_prefix(), src, count); + __ctl_load(cr0, 0, 0); + preempt_enable(); +} From e4258d55bff06780cd424c671b576a90acc1592f Mon Sep 17 00:00:00 2001 From: Stefan Weinhuber Date: Wed, 3 Aug 2011 16:44:20 +0200 Subject: [PATCH 243/439] [S390] dasd: use vmalloc for statistics input buffer The size of the buffer that is used to store DASD statistics input strings depends on the user input. If the input string is to large, the write operation could fail with -ENOMEM. To avoid this, use vmalloc instead of kmalloc. Signed-off-by: Stefan Weinhuber Signed-off-by: Heiko Carstens --- drivers/s390/block/dasd.c | 9 +++++---- drivers/s390/block/dasd_proc.c | 4 ++-- 2 files changed, 7 insertions(+), 6 deletions(-) diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c index 432444af7ee4..a1d3ddba99cc 100644 --- a/drivers/s390/block/dasd.c +++ b/drivers/s390/block/dasd.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include @@ -888,11 +889,11 @@ char *dasd_get_user_string(const char __user *user_buf, size_t user_len) { char *buffer; - buffer = kmalloc(user_len + 1, GFP_KERNEL); + buffer = vmalloc(user_len + 1); if (buffer == NULL) return ERR_PTR(-ENOMEM); if (copy_from_user(buffer, user_buf, user_len) != 0) { - kfree(buffer); + vfree(buffer); return ERR_PTR(-EFAULT); } /* got the string, now strip linefeed. */ @@ -930,7 +931,7 @@ static ssize_t dasd_stats_write(struct file *file, dasd_profile_off(prof); } else rc = -EINVAL; - kfree(buffer); + vfree(buffer); return rc; } @@ -1042,7 +1043,7 @@ static ssize_t dasd_stats_global_write(struct file *file, dasd_global_profile_level = DASD_PROFILE_OFF; } else rc = -EINVAL; - kfree(buffer); + vfree(buffer); return rc; } diff --git a/drivers/s390/block/dasd_proc.c b/drivers/s390/block/dasd_proc.c index 6c3c5364d082..e12989fff4ff 100644 --- a/drivers/s390/block/dasd_proc.c +++ b/drivers/s390/block/dasd_proc.c @@ -312,14 +312,14 @@ static ssize_t dasd_stats_proc_write(struct file *file, pr_info("The statistics have been reset\n"); } else goto out_parse_error; - kfree(buffer); + vfree(buffer); return user_len; out_parse_error: rc = -EINVAL; pr_warning("%s is not a supported value for /proc/dasd/statistics\n", str); out_error: - kfree(buffer); + vfree(buffer); return rc; #else pr_warning("/proc/dasd/statistics: is not activated in this kernel\n"); From ef1daec8da2c04b0c6e91a34b9cac1aad33c6692 Mon Sep 17 00:00:00 2001 From: Michael Holzheu Date: Wed, 3 Aug 2011 16:44:21 +0200 Subject: [PATCH 244/439] [S390] Export store_status() function For kdump we need a store status function to save the registers for the current CPU. Therefore this patch exports a function "store_status()". In addition to that now also floating point registers are saved correctly. Signed-off-by: Michael Holzheu Signed-off-by: Heiko Carstens --- arch/s390/kernel/reipl64.S | 80 ++++++++++++++++++++++++++++---------- 1 file changed, 60 insertions(+), 20 deletions(-) diff --git a/arch/s390/kernel/reipl64.S b/arch/s390/kernel/reipl64.S index 78eb7cfbd3d1..e690975403f4 100644 --- a/arch/s390/kernel/reipl64.S +++ b/arch/s390/kernel/reipl64.S @@ -1,5 +1,5 @@ /* - * Copyright IBM Corp 2000,2009 + * Copyright IBM Corp 2000,2011 * Author(s): Holger Smolinski , * Denis Joseph Barrow, */ @@ -7,6 +7,64 @@ #include #include +# +# store_status +# +# Prerequisites to run this function: +# - Prefix register is set to zero +# - Original prefix register is stored in "dump_prefix_page" +# - Lowcore protection is off +# +ENTRY(store_status) + /* Save register one and load save area base */ + stg %r1,__LC_SAVE_AREA_64(%r0) + lghi %r1,SAVE_AREA_BASE + /* General purpose registers */ + stmg %r0,%r15,__LC_GPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + lg %r2,__LC_SAVE_AREA_64(%r0) + stg %r2,__LC_GPREGS_SAVE_AREA-SAVE_AREA_BASE+8(%r1) + /* Control registers */ + stctg %c0,%c15,__LC_CREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + /* Access registers */ + stam %a0,%a15,__LC_AREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + /* Floating point registers */ + std %f0, 0x00 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f1, 0x08 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f2, 0x10 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f3, 0x18 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f4, 0x20 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f5, 0x28 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f6, 0x30 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f7, 0x38 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f8, 0x40 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f9, 0x48 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f10,0x50 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f11,0x58 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f12,0x60 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f13,0x68 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f14,0x70 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + std %f15,0x78 + __LC_FPREGS_SAVE_AREA-SAVE_AREA_BASE(%r1) + /* Floating point control register */ + stfpc __LC_FP_CREG_SAVE_AREA-SAVE_AREA_BASE(%r1) + /* CPU timer */ + stpt __LC_CPU_TIMER_SAVE_AREA-SAVE_AREA_BASE(%r1) + /* Saved prefix register */ + larl %r2,dump_prefix_page + mvc __LC_PREFIX_SAVE_AREA-SAVE_AREA_BASE(4,%r1),0(%r2) + /* Clock comparator - seven bytes */ + larl %r2,.Lclkcmp + stckc 0(%r2) + mvc __LC_CLOCK_COMP_SAVE_AREA-SAVE_AREA_BASE + 1(7,%r1),1(%r2) + /* Program status word */ + epsw %r2,%r3 + st %r2,__LC_PSW_SAVE_AREA-SAVE_AREA_BASE + 0(%r1) + st %r3,__LC_PSW_SAVE_AREA-SAVE_AREA_BASE + 4(%r1) + larl %r2,store_status + stg %r2,__LC_PSW_SAVE_AREA-SAVE_AREA_BASE + 8(%r1) + br %r14 +.align 8 +.Lclkcmp: .quad 0x0000000000000000 + # # do_reipl_asm # Parameter: r2 = schid of reipl device @@ -15,22 +73,7 @@ ENTRY(do_reipl_asm) basr %r13,0 .Lpg0: lpswe .Lnewpsw-.Lpg0(%r13) -.Lpg1: # do store status of all registers - - stg %r1,.Lregsave-.Lpg0(%r13) - lghi %r1,0x1000 - stmg %r0,%r15,__LC_GPREGS_SAVE_AREA-0x1000(%r1) - lg %r0,.Lregsave-.Lpg0(%r13) - stg %r0,__LC_GPREGS_SAVE_AREA-0x1000+8(%r1) - stctg %c0,%c15,__LC_CREGS_SAVE_AREA-0x1000(%r1) - stam %a0,%a15,__LC_AREGS_SAVE_AREA-0x1000(%r1) - lg %r10,.Ldump_pfx-.Lpg0(%r13) - mvc __LC_PREFIX_SAVE_AREA-0x1000(4,%r1),0(%r10) - stfpc __LC_FP_CREG_SAVE_AREA-0x1000(%r1) - stckc .Lclkcmp-.Lpg0(%r13) - mvc __LC_CLOCK_COMP_SAVE_AREA-0x1000(7,%r1),.Lclkcmp-.Lpg0(%r13) - stpt __LC_CPU_TIMER_SAVE_AREA-0x1000(%r1) - stg %r13, __LC_PSW_SAVE_AREA-0x1000+8(%r1) +.Lpg1: brasl %r14,store_status lctlg %c6,%c6,.Lall-.Lpg0(%r13) lgr %r1,%r2 @@ -67,10 +110,7 @@ ENTRY(do_reipl_asm) st %r14,.Ldispsw+12-.Lpg0(%r13) lpswe .Ldispsw-.Lpg0(%r13) .align 8 -.Lclkcmp: .quad 0x0000000000000000 .Lall: .quad 0x00000000ff000000 -.Ldump_pfx: .quad dump_prefix_page -.Lregsave: .quad 0x0000000000000000 .align 16 /* * These addresses have to be 31 bit otherwise From 9dc7356ee1266d6a69fc80f28d006f71c20bf172 Mon Sep 17 00:00:00 2001 From: Michael Holzheu Date: Wed, 3 Aug 2011 16:44:22 +0200 Subject: [PATCH 245/439] [S390] Use diagnose 308 for system reset The diagnose 308 call is the prefered method for clearing all ongoing I/O. Therefore if it is available we use it instead of doing a manual reset. Signed-off-by: Michael Holzheu Signed-off-by: Heiko Carstens --- arch/s390/include/asm/ipl.h | 1 + arch/s390/kernel/base.S | 36 ++++++++++++++++++++++++++++++++++++ arch/s390/kernel/ipl.c | 6 ++++++ 3 files changed, 43 insertions(+) diff --git a/arch/s390/include/asm/ipl.h b/arch/s390/include/asm/ipl.h index 5e95d95450b3..97cc4403fabf 100644 --- a/arch/s390/include/asm/ipl.h +++ b/arch/s390/include/asm/ipl.h @@ -167,5 +167,6 @@ enum diag308_rc { }; extern int diag308(unsigned long subcode, void *addr); +extern void diag308_reset(void); #endif /* _ASM_S390_IPL_H */ diff --git a/arch/s390/kernel/base.S b/arch/s390/kernel/base.S index 209938c1dfc8..255435663bf8 100644 --- a/arch/s390/kernel/base.S +++ b/arch/s390/kernel/base.S @@ -76,6 +76,42 @@ s390_base_pgm_handler_fn: .quad 0 .previous +# +# Calls diag 308 subcode 1 and continues execution +# +# The following conditions must be ensured before calling this function: +# * Prefix register = 0 +# * Lowcore protection is disabled +# +ENTRY(diag308_reset) + larl %r4,.Lctlregs # Save control registers + stctg %c0,%c15,0(%r4) + larl %r4,.Lrestart_psw # Setup restart PSW at absolute 0 + lghi %r3,0 + lg %r4,0(%r4) # Save PSW + sturg %r4,%r3 # Use sturg, because of large pages + lghi %r1,1 + diag %r1,%r1,0x308 +.Lrestart_part2: + lhi %r0,0 # Load r0 with zero + lhi %r1,2 # Use mode 2 = ESAME (dump) + sigp %r1,%r0,0x12 # Switch to ESAME mode + sam64 # Switch to 64 bit addressing mode + larl %r4,.Lctlregs # Restore control registers + lctlg %c0,%c15,0(%r4) + br %r14 +.align 16 +.Lrestart_psw: + .long 0x00080000,0x80000000 + .Lrestart_part2 + + .section .bss +.align 8 +.Lctlregs: + .rept 16 + .quad 0 + .endr + .previous + #else /* CONFIG_64BIT */ ENTRY(s390_base_mcck_handler) diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c index 059c59027ac0..04361d5a4279 100644 --- a/arch/s390/kernel/ipl.c +++ b/arch/s390/kernel/ipl.c @@ -1996,6 +1996,12 @@ static void do_reset_calls(void) { struct reset_call *reset; +#ifdef CONFIG_64BIT + if (diag308_set_works) { + diag308_reset(); + return; + } +#endif list_for_each_entry(reset, &rcall, list) reset->fn(); } From 34cd551a31cd4e35cd3f9ed9f13bb46b4ee98508 Mon Sep 17 00:00:00 2001 From: Stefan Haberland Date: Wed, 3 Aug 2011 16:44:23 +0200 Subject: [PATCH 246/439] [S390] dasd: check if raw track access is supported To use raw track access some special storage server commands are needed. Older storage hardware may not support these commands. So check if raw track access is possible while setting the DASD online. Signed-off-by: Stefan Haberland Signed-off-by: Martin Schwidefsky Signed-off-by: Heiko Carstens --- drivers/s390/block/dasd_eckd.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/s390/block/dasd_eckd.c b/drivers/s390/block/dasd_eckd.c index 30fb979d684d..6e835c9fdfcb 100644 --- a/drivers/s390/block/dasd_eckd.c +++ b/drivers/s390/block/dasd_eckd.c @@ -1461,6 +1461,15 @@ dasd_eckd_check_characteristics(struct dasd_device *device) "Read device characteristic failed, rc=%d", rc); goto out_err3; } + + if ((device->features & DASD_FEATURE_USERAW) && + !(private->rdc_data.facilities.RT_in_LR)) { + dev_err(&device->cdev->dev, "The storage server does not " + "support raw-track access\n"); + rc = -EINVAL; + goto out_err3; + } + /* find the valid cylinder size */ if (private->rdc_data.no_cyl == LV_COMPAT_CYL && private->rdc_data.long_no_cyl) From 3a81b171429113ae0af2920f5d4ecfe571230627 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 3 Aug 2011 16:44:24 +0200 Subject: [PATCH 247/439] [S390] Add support for IBM zEnterprise 114 Just fix up the Kconfig description and the elf platform. Signed-off-by: Heiko Carstens --- arch/s390/Kconfig | 8 ++++---- arch/s390/kernel/setup.c | 1 + 2 files changed, 5 insertions(+), 4 deletions(-) diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index c03fef7a9c22..17dbb4318261 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -273,11 +273,11 @@ config MARCH_Z10 on older machines. config MARCH_Z196 - bool "IBM zEnterprise 196" + bool "IBM zEnterprise 114 and 196" help - Select this to enable optimizations for IBM zEnterprise 196 - (2817 series). The kernel will be slightly faster but will not work - on older machines. + Select this to enable optimizations for IBM zEnterprise 114 and 196 + (2818 and 2817 series). The kernel will be slightly faster but will + not work on older machines. endchoice diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 18640ff0aa1a..7b371c37061d 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -752,6 +752,7 @@ static void __init setup_hwcaps(void) strcpy(elf_platform, "z10"); break; case 0x2817: + case 0x2818: strcpy(elf_platform, "z196"); break; } From 7a0e42f168337d4af8fe230c1043cb04786c04c2 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 3 Aug 2011 16:44:25 +0200 Subject: [PATCH 248/439] [S390] asm offsets: fix coding style Because of readability reasons we ignore the 80 character line limit in asm offsets. Just one line per define, nothing else. Signed-off-by: Heiko Carstens --- arch/s390/kernel/asm-offsets.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c index 199a537d4f36..532fd4322156 100644 --- a/arch/s390/kernel/asm-offsets.c +++ b/arch/s390/kernel/asm-offsets.c @@ -27,12 +27,9 @@ int main(void) BLANK(); DEFINE(__TASK_pid, offsetof(struct task_struct, pid)); BLANK(); - DEFINE(__THREAD_per_cause, - offsetof(struct task_struct, thread.per_event.cause)); - DEFINE(__THREAD_per_address, - offsetof(struct task_struct, thread.per_event.address)); - DEFINE(__THREAD_per_paid, - offsetof(struct task_struct, thread.per_event.paid)); + DEFINE(__THREAD_per_cause, offsetof(struct task_struct, thread.per_event.cause)); + DEFINE(__THREAD_per_address, offsetof(struct task_struct, thread.per_event.address)); + DEFINE(__THREAD_per_paid, offsetof(struct task_struct, thread.per_event.paid)); BLANK(); DEFINE(__TI_task, offsetof(struct thread_info, task)); DEFINE(__TI_domain, offsetof(struct thread_info, exec_domain)); From 391c62feb1798b6d31bd88076eae649b091ad8bf Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 3 Aug 2011 16:44:26 +0200 Subject: [PATCH 249/439] [S390] signal: convert to use set_current_blocked() Convert to use set_current_blocked() like x86. Signed-off-by: Heiko Carstens --- arch/s390/kernel/compat_signal.c | 43 +++++++---------------- arch/s390/kernel/signal.c | 59 ++++++++++---------------------- 2 files changed, 30 insertions(+), 72 deletions(-) diff --git a/arch/s390/kernel/compat_signal.c b/arch/s390/kernel/compat_signal.c index eee999853a7c..a9a285b8c4ad 100644 --- a/arch/s390/kernel/compat_signal.c +++ b/arch/s390/kernel/compat_signal.c @@ -380,20 +380,13 @@ asmlinkage long sys32_sigreturn(void) goto badframe; if (__copy_from_user(&set.sig, &frame->sc.oldmask, _SIGMASK_COPY_SIZE32)) goto badframe; - sigdelsetmask(&set, ~_BLOCKABLE); - spin_lock_irq(¤t->sighand->siglock); - current->blocked = set; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - + set_current_blocked(&set); if (restore_sigregs32(regs, &frame->sregs)) goto badframe; if (restore_sigregs_gprs_high(regs, frame->gprs_high)) goto badframe; - return regs->gprs[2]; - badframe: force_sig(SIGSEGV, current); return 0; @@ -413,31 +406,22 @@ asmlinkage long sys32_rt_sigreturn(void) goto badframe; if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set))) goto badframe; - sigdelsetmask(&set, ~_BLOCKABLE); - spin_lock_irq(¤t->sighand->siglock); - current->blocked = set; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - + set_current_blocked(&set); if (restore_sigregs32(regs, &frame->uc.uc_mcontext)) goto badframe; if (restore_sigregs_gprs_high(regs, frame->gprs_high)) goto badframe; - err = __get_user(ss_sp, &frame->uc.uc_stack.ss_sp); st.ss_sp = compat_ptr(ss_sp); err |= __get_user(st.ss_size, &frame->uc.uc_stack.ss_size); err |= __get_user(st.ss_flags, &frame->uc.uc_stack.ss_flags); if (err) goto badframe; - set_fs (KERNEL_DS); do_sigaltstack((stack_t __force __user *)&st, NULL, regs->gprs[15]); set_fs (old_fs); - return regs->gprs[2]; - badframe: force_sig(SIGSEGV, current); return 0; @@ -605,10 +589,10 @@ give_sigsegv: * OK, we're invoking a handler */ -int -handle_signal32(unsigned long sig, struct k_sigaction *ka, - siginfo_t *info, sigset_t *oldset, struct pt_regs * regs) +int handle_signal32(unsigned long sig, struct k_sigaction *ka, + siginfo_t *info, sigset_t *oldset, struct pt_regs *regs) { + sigset_t blocked; int ret; /* Set up the stack frame */ @@ -616,15 +600,12 @@ handle_signal32(unsigned long sig, struct k_sigaction *ka, ret = setup_rt_frame32(sig, ka, info, oldset, regs); else ret = setup_frame32(sig, ka, oldset, regs); - - if (ret == 0) { - spin_lock_irq(¤t->sighand->siglock); - sigorsets(¤t->blocked,¤t->blocked,&ka->sa.sa_mask); - if (!(ka->sa.sa_flags & SA_NODEFER)) - sigaddset(¤t->blocked,sig); - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - } - return ret; + if (ret) + return ret; + sigorsets(&blocked, ¤t->blocked, &ka->sa.sa_mask); + if (!(ka->sa.sa_flags & SA_NODEFER)) + sigaddset(&blocked, sig); + set_current_blocked(&blocked); + return 0; } diff --git a/arch/s390/kernel/signal.c b/arch/s390/kernel/signal.c index abbb3c3c7aab..38a88b6128be 100644 --- a/arch/s390/kernel/signal.c +++ b/arch/s390/kernel/signal.c @@ -57,17 +57,15 @@ typedef struct */ SYSCALL_DEFINE3(sigsuspend, int, history0, int, history1, old_sigset_t, mask) { - mask &= _BLOCKABLE; - spin_lock_irq(¤t->sighand->siglock); - current->saved_sigmask = current->blocked; - siginitset(¤t->blocked, mask); - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); + sigset_t blocked; + current->saved_sigmask = current->blocked; + mask &= _BLOCKABLE; + siginitset(&blocked, mask); + set_current_blocked(&blocked); set_current_state(TASK_INTERRUPTIBLE); schedule(); set_thread_flag(TIF_RESTORE_SIGMASK); - return -ERESTARTNOHAND; } @@ -172,18 +170,11 @@ SYSCALL_DEFINE0(sigreturn) goto badframe; if (__copy_from_user(&set.sig, &frame->sc.oldmask, _SIGMASK_COPY_SIZE)) goto badframe; - sigdelsetmask(&set, ~_BLOCKABLE); - spin_lock_irq(¤t->sighand->siglock); - current->blocked = set; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - + set_current_blocked(&set); if (restore_sigregs(regs, &frame->sregs)) goto badframe; - return regs->gprs[2]; - badframe: force_sig(SIGSEGV, current); return 0; @@ -199,21 +190,14 @@ SYSCALL_DEFINE0(rt_sigreturn) goto badframe; if (__copy_from_user(&set.sig, &frame->uc.uc_sigmask, sizeof(set))) goto badframe; - sigdelsetmask(&set, ~_BLOCKABLE); - spin_lock_irq(¤t->sighand->siglock); - current->blocked = set; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - + set_current_blocked(&set); if (restore_sigregs(regs, &frame->uc.uc_mcontext)) goto badframe; - if (do_sigaltstack(&frame->uc.uc_stack, NULL, regs->gprs[15]) == -EFAULT) goto badframe; return regs->gprs[2]; - badframe: force_sig(SIGSEGV, current); return 0; @@ -385,14 +369,11 @@ give_sigsegv: return -EFAULT; } -/* - * OK, we're invoking a handler - */ - -static int -handle_signal(unsigned long sig, struct k_sigaction *ka, - siginfo_t *info, sigset_t *oldset, struct pt_regs * regs) +static int handle_signal(unsigned long sig, struct k_sigaction *ka, + siginfo_t *info, sigset_t *oldset, + struct pt_regs *regs) { + sigset_t blocked; int ret; /* Set up the stack frame */ @@ -400,17 +381,13 @@ handle_signal(unsigned long sig, struct k_sigaction *ka, ret = setup_rt_frame(sig, ka, info, oldset, regs); else ret = setup_frame(sig, ka, oldset, regs); - - if (ret == 0) { - spin_lock_irq(¤t->sighand->siglock); - sigorsets(¤t->blocked,¤t->blocked,&ka->sa.sa_mask); - if (!(ka->sa.sa_flags & SA_NODEFER)) - sigaddset(¤t->blocked,sig); - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - } - - return ret; + if (ret) + return ret; + sigorsets(&blocked, ¤t->blocked, &ka->sa.sa_mask); + if (!(ka->sa.sa_flags & SA_NODEFER)) + sigaddset(&blocked, sig); + set_current_blocked(&blocked); + return 0; } /* From cc34321d5806b7919531a1d951bb6deb62c163c9 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 3 Aug 2011 16:44:27 +0200 Subject: [PATCH 250/439] [S390] cpu hotplug: on cpu start wait until being marked active This is the same as fd8a7de1 "x86: cpu-hotplug: Prevent softirq wakeup on wrong CPU". Unlike on x86 this doesn't fix a bug on s390 since we do not have threaded interrupt handlers. However we want to keep the same initialization order like on x86. This should prevent bugs caused by code which assumes (and relies on) the init order is the same on each architecture. Signed-off-by: Heiko Carstens --- arch/s390/kernel/smp.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 70f4b9604575..86371fdd68e7 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -473,7 +473,12 @@ int __cpuinit start_secondary(void *cpuvoid) S390_lowcore.restart_psw.addr = PSW_ADDR_AMODE | (unsigned long) psw_restart_int_handler; __ctl_set_bit(0, 28); /* Enable lowcore protection */ - /* Switch on interrupts */ + /* + * Wait until the cpu which brought this one up marked it + * active before enabling interrupts. + */ + while (!cpumask_test_cpu(smp_processor_id(), cpu_active_mask)) + cpu_relax(); local_irq_enable(); /* cpu_idle will call schedule for us */ cpu_idle(); From ada5ed5484f492e4eb13d788c6c077cf243f53e6 Mon Sep 17 00:00:00 2001 From: Mathias Krause Date: Wed, 3 Aug 2011 16:44:28 +0200 Subject: [PATCH 251/439] [S390] exec: remove redundant set_fs(USER_DS) The address limit is already set in flush_old_exec() so those calls to set_fs(USER_DS) are redundant. Signed-off-by: Mathias Krause Signed-off-by: Heiko Carstens --- arch/s390/include/asm/processor.h | 2 -- 1 file changed, 2 deletions(-) diff --git a/arch/s390/include/asm/processor.h b/arch/s390/include/asm/processor.h index 55dfcc8bdc0d..a4b6229e5d4b 100644 --- a/arch/s390/include/asm/processor.h +++ b/arch/s390/include/asm/processor.h @@ -119,14 +119,12 @@ struct stack_frame { * Do necessary setup to start up a new thread. */ #define start_thread(regs, new_psw, new_stackp) do { \ - set_fs(USER_DS); \ regs->psw.mask = psw_user_bits; \ regs->psw.addr = new_psw | PSW_ADDR_AMODE; \ regs->gprs[15] = new_stackp; \ } while (0) #define start_thread31(regs, new_psw, new_stackp) do { \ - set_fs(USER_DS); \ regs->psw.mask = psw_user32_bits; \ regs->psw.addr = new_psw | PSW_ADDR_AMODE; \ regs->gprs[15] = new_stackp; \ From d5f4d113cb9dac182ba44eb605ec4a2f1fdfd3ec Mon Sep 17 00:00:00 2001 From: Peter Huewe Date: Wed, 3 Aug 2011 16:44:29 +0200 Subject: [PATCH 252/439] [S390] sclp_async: Use kstrtoul_from_user This patch replaces the code for getting an unsigned long from a userspace buffer by a simple call to kstroul_from_user. This makes it easier to read and less error prone. Signed-off-by: Peter Huewe Signed-off-by: Heiko Carstens --- drivers/s390/char/sclp_async.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) diff --git a/drivers/s390/char/sclp_async.c b/drivers/s390/char/sclp_async.c index 7ad30e72f868..5f9f929e891c 100644 --- a/drivers/s390/char/sclp_async.c +++ b/drivers/s390/char/sclp_async.c @@ -82,12 +82,9 @@ static int proc_handler_callhome(struct ctl_table *ctl, int write, return -EFAULT; } else { len = *count; - rc = copy_from_user(buf, buffer, sizeof(buf)); - if (rc != 0) - return -EFAULT; - buf[sizeof(buf) - 1] = '\0'; - if (strict_strtoul(buf, 0, &val) != 0) - return -EINVAL; + rc = kstrtoul_from_user(buffer, len, 0, &val); + if (rc) + return rc; if (val != 0 && val != 1) return -EINVAL; callhome_enabled = val; From af6df871ba87ff93739babb16d26160c37c43183 Mon Sep 17 00:00:00 2001 From: Peter Huewe Date: Wed, 3 Aug 2011 16:44:30 +0200 Subject: [PATCH 253/439] [S390] qdio: Use kstrtoul_from_user This patch replaces the code for getting an unsigned long from a userspace buffer by a simple call to kstroul_from_user. This makes it easier to read and less error prone. Signed-off-by: Peter Huewe Acked-by: Jan Glauber Signed-off-by: Heiko Carstens --- drivers/s390/cio/qdio_debug.c | 10 ++-------- 1 file changed, 2 insertions(+), 8 deletions(-) diff --git a/drivers/s390/cio/qdio_debug.c b/drivers/s390/cio/qdio_debug.c index f8b03a636e49..0e615cb912d0 100644 --- a/drivers/s390/cio/qdio_debug.c +++ b/drivers/s390/cio/qdio_debug.c @@ -188,19 +188,13 @@ static ssize_t qperf_seq_write(struct file *file, const char __user *ubuf, struct qdio_irq *irq_ptr = seq->private; struct qdio_q *q; unsigned long val; - char buf[8]; int ret, i; if (!irq_ptr) return 0; - if (count >= sizeof(buf)) - return -EINVAL; - if (copy_from_user(&buf, ubuf, count)) - return -EFAULT; - buf[count] = 0; - ret = strict_strtoul(buf, 10, &val); - if (ret < 0) + ret = kstrtoul_from_user(ubuf, count, 10, &val); + if (ret) return ret; switch (val) { From b7f275042f5d69b6c31975341ce5f06cf2aff95f Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 3 Aug 2011 16:44:31 +0200 Subject: [PATCH 254/439] [S390] smp: remove pointless comments in startup_secondary() Remove pointless comments in startup_secondary(). There is not too much value in having comments like e.g. "call cpu notifiers" just before a call to notify_cpu*(). Signed-off-by: Heiko Carstens --- arch/s390/kernel/smp.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 86371fdd68e7..6ab16ac64d29 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -452,19 +452,13 @@ out: */ int __cpuinit start_secondary(void *cpuvoid) { - /* Setup the cpu */ cpu_init(); preempt_disable(); - /* Enable TOD clock interrupts on the secondary cpu. */ init_cpu_timer(); - /* Enable cpu timer interrupts on the secondary cpu. */ init_cpu_vtimer(); - /* Enable pfault pseudo page faults on this cpu. */ pfault_init(); - /* call cpu notifiers */ notify_cpu_starting(smp_processor_id()); - /* Mark this cpu as online */ ipi_call_lock(); set_cpu_online(smp_processor_id(), true); ipi_call_unlock(); From 9e8ed3ae924b65ab5f088fe63ee6f4326f04590f Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Wed, 3 Aug 2011 16:44:32 +0200 Subject: [PATCH 255/439] [S390] signal: use set_restore_sigmask() helper We should call set_restore_sigmask() instead of directly setting TIF_RESTORE_SIGMASK. This change should have been done three years earlier... see 4e4c22 "signals: add set_restore_sigmask". Signed-off-by: Heiko Carstens --- arch/s390/kernel/signal.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/s390/kernel/signal.c b/arch/s390/kernel/signal.c index 38a88b6128be..9a40e1cc5ec3 100644 --- a/arch/s390/kernel/signal.c +++ b/arch/s390/kernel/signal.c @@ -65,7 +65,7 @@ SYSCALL_DEFINE3(sigsuspend, int, history0, int, history1, old_sigset_t, mask) set_current_blocked(&blocked); set_current_state(TASK_INTERRUPTIBLE); schedule(); - set_thread_flag(TIF_RESTORE_SIGMASK); + set_restore_sigmask(); return -ERESTARTNOHAND; } From df013ffb8119c89f062ab05b7f544704315db47b Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Wed, 13 Jul 2011 13:14:22 +0800 Subject: [PATCH 256/439] Add Kconfig option ARCH_HAVE_NMI_SAFE_CMPXCHG cmpxchg() is widely used by lockless code, including NMI-safe lockless code. But on some architectures, the cmpxchg() implementation is not NMI-safe, on these architectures the lockless code may need a spin_trylock_irqsave() based implementation. This patch adds a Kconfig option: ARCH_HAVE_NMI_SAFE_CMPXCHG, so that NMI-safe lockless code can depend on it or provide different implementation according to it. On many architectures, cmpxchg is only NMI-safe for several specific operand sizes. So, ARCH_HAVE_NMI_SAFE_CMPXCHG define in this patch only guarantees cmpxchg is NMI-safe for sizeof(unsigned long). Signed-off-by: Huang Ying Acked-by: Mike Frysinger Acked-by: Paul Mundt Acked-by: Hans-Christian Egtvedt Acked-by: Benjamin Herrenschmidt Acked-by: Chris Metcalf Acked-by: Richard Henderson CC: Mikael Starvik Acked-by: David Howells CC: Yoshinori Sato CC: Tony Luck CC: Hirokazu Takata CC: Geert Uytterhoeven CC: Michal Simek Acked-by: Ralf Baechle CC: Kyle McMartin CC: Martin Schwidefsky CC: Chen Liqin CC: "David S. Miller" CC: Ingo Molnar CC: Chris Zankel Signed-off-by: Len Brown --- arch/Kconfig | 3 +++ arch/alpha/Kconfig | 1 + arch/avr32/Kconfig | 1 + arch/frv/Kconfig | 1 + arch/ia64/Kconfig | 1 + arch/m68k/Kconfig | 1 + arch/parisc/Kconfig | 1 + arch/powerpc/Kconfig | 1 + arch/s390/Kconfig | 1 + arch/sh/Kconfig | 1 + arch/sparc/Kconfig | 1 + arch/tile/Kconfig | 1 + arch/x86/Kconfig | 1 + 13 files changed, 15 insertions(+) diff --git a/arch/Kconfig b/arch/Kconfig index 26b0e2397a57..4b0669cbb3b0 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -178,4 +178,7 @@ config HAVE_ARCH_MUTEX_CPU_RELAX config HAVE_RCU_TABLE_FREE bool +config ARCH_HAVE_NMI_SAFE_CMPXCHG + bool + source "kernel/gcov/Kconfig" diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig index 60219bf94198..23f8d8c32886 100644 --- a/arch/alpha/Kconfig +++ b/arch/alpha/Kconfig @@ -13,6 +13,7 @@ config ALPHA select AUTO_IRQ_AFFINITY if SMP select GENERIC_IRQ_SHOW select ARCH_WANT_OPTIONAL_GPIOLIB + select ARCH_HAVE_NMI_SAFE_CMPXCHG help The Alpha is a 64-bit general-purpose processor designed and marketed by the Digital Equipment Corporation of blessed memory, diff --git a/arch/avr32/Kconfig b/arch/avr32/Kconfig index e9d689b7c833..197e96f70405 100644 --- a/arch/avr32/Kconfig +++ b/arch/avr32/Kconfig @@ -10,6 +10,7 @@ config AVR32 select GENERIC_IRQ_PROBE select HARDIRQS_SW_RESEND select GENERIC_IRQ_SHOW + select ARCH_HAVE_NMI_SAFE_CMPXCHG help AVR32 is a high-performance 32-bit RISC microprocessor core, designed for cost-sensitive embedded applications, with particular diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig index cb884e489425..bad27a6ff407 100644 --- a/arch/frv/Kconfig +++ b/arch/frv/Kconfig @@ -7,6 +7,7 @@ config FRV select HAVE_PERF_EVENTS select HAVE_GENERIC_HARDIRQS select GENERIC_IRQ_SHOW + select ARCH_HAVE_NMI_SAFE_CMPXCHG config ZONE_DMA bool diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index 38280ef4a2af..b1227ddc672a 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -27,6 +27,7 @@ config IA64 select GENERIC_PENDING_IRQ if SMP select IRQ_PER_CPU select GENERIC_IRQ_SHOW + select ARCH_HAVE_NMI_SAFE_CMPXCHG default y help The Itanium Processor Family is Intel's 64-bit successor to diff --git a/arch/m68k/Kconfig b/arch/m68k/Kconfig index d66e34c718d0..effe8f0fa31e 100644 --- a/arch/m68k/Kconfig +++ b/arch/m68k/Kconfig @@ -6,6 +6,7 @@ config M68K select GENERIC_ATOMIC64 if MMU select HAVE_GENERIC_HARDIRQS if !MMU select GENERIC_IRQ_SHOW if !MMU + select ARCH_HAVE_NMI_SAFE_CMPXCHG if RMW_INSNS config RWSEM_GENERIC_SPINLOCK bool diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig index 65adc86a230e..e077b0bf56ca 100644 --- a/arch/parisc/Kconfig +++ b/arch/parisc/Kconfig @@ -15,6 +15,7 @@ config PARISC select HAVE_GENERIC_HARDIRQS select GENERIC_IRQ_PROBE select IRQ_PER_CPU + select ARCH_HAVE_NMI_SAFE_CMPXCHG help The PA-RISC microprocessor is designed by Hewlett-Packard and used diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 2729c6663d8a..e55f754dd50d 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -134,6 +134,7 @@ config PPC select GENERIC_IRQ_SHOW_LEVEL select HAVE_RCU_TABLE_FREE if SMP select HAVE_SYSCALL_TRACEPOINTS + select ARCH_HAVE_NMI_SAFE_CMPXCHG config EARLY_PRINTK bool diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index c03fef7a9c22..0f98bbddade5 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -81,6 +81,7 @@ config S390 select INIT_ALL_POSSIBLE select HAVE_IRQ_WORK select HAVE_PERF_EVENTS + select ARCH_HAVE_NMI_SAFE_CMPXCHG select HAVE_KERNEL_GZIP select HAVE_KERNEL_BZIP2 select HAVE_KERNEL_LZMA diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig index bbdeb48bbf8e..66405eb21766 100644 --- a/arch/sh/Kconfig +++ b/arch/sh/Kconfig @@ -11,6 +11,7 @@ config SUPERH select HAVE_DMA_ATTRS select HAVE_IRQ_WORK select HAVE_PERF_EVENTS + select ARCH_HAVE_NMI_SAFE_CMPXCHG if (GUSA_RB || CPU_SH4A) select PERF_USE_VMALLOC select HAVE_KERNEL_GZIP select HAVE_KERNEL_BZIP2 diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig index 253986bd6bb6..a63cddfb3a15 100644 --- a/arch/sparc/Kconfig +++ b/arch/sparc/Kconfig @@ -53,6 +53,7 @@ config SPARC64 select HAVE_PERF_EVENTS select PERF_USE_VMALLOC select IRQ_PREFLOW_FASTEOI + select ARCH_HAVE_NMI_SAFE_CMPXCHG config ARCH_DEFCONFIG string diff --git a/arch/tile/Kconfig b/arch/tile/Kconfig index 0249b8b4db54..b30f71ac0d06 100644 --- a/arch/tile/Kconfig +++ b/arch/tile/Kconfig @@ -12,6 +12,7 @@ config TILE select GENERIC_PENDING_IRQ if SMP select GENERIC_IRQ_SHOW select SYS_HYPERVISOR + select ARCH_HAVE_NMI_SAFE_CMPXCHG if !M386 # FIXME: investigate whether we need/want these options. # select HAVE_IOREMAP_PROT diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index da349723d411..a680a60898de 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -70,6 +70,7 @@ config X86 select IRQ_FORCED_THREADING select USE_GENERIC_SMP_HELPERS if SMP select HAVE_BPF_JIT if (X86_64 && NET) + select ARCH_HAVE_NMI_SAFE_CMPXCHG config INSTRUCTION_DECODER def_bool (KPROBES || PERF_EVENTS) From f49f23abf3dd786ddcac1c1e7db3c2013b07413f Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Wed, 13 Jul 2011 13:14:23 +0800 Subject: [PATCH 257/439] lib, Add lock-less NULL terminated single list Cmpxchg is used to implement adding new entry to the list, deleting all entries from the list, deleting first entry of the list and some other operations. Because this is a single list, so the tail can not be accessed in O(1). If there are multiple producers and multiple consumers, llist_add can be used in producers and llist_del_all can be used in consumers. They can work simultaneously without lock. But llist_del_first can not be used here. Because llist_del_first depends on list->first->next does not changed if list->first is not changed during its operation, but llist_del_first, llist_add, llist_add (or llist_del_all, llist_add, llist_add) sequence in another consumer may violate that. If there are multiple producers and one consumer, llist_add can be used in producers and llist_del_all or llist_del_first can be used in the consumer. This can be summarized as follow: | add | del_first | del_all add | - | - | - del_first | | L | L del_all | | | - Where "-" stands for no lock is needed, while "L" stands for lock is needed. The list entries deleted via llist_del_all can be traversed with traversing function such as llist_for_each etc. But the list entries can not be traversed safely before deleted from the list. The order of deleted entries is from the newest to the oldest added one. If you want to traverse from the oldest to the newest, you must reverse the order by yourself before traversing. The basic atomic operation of this list is cmpxchg on long. On architectures that don't have NMI-safe cmpxchg implementation, the list can NOT be used in NMI handler. So code uses the list in NMI handler should depend on CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG. Signed-off-by: Huang Ying Reviewed-by: Andi Kleen Reviewed-by: Mathieu Desnoyers Cc: Andrew Morton Signed-off-by: Len Brown --- include/linux/llist.h | 126 +++++++++++++++++++++++++++++++++++++++++ lib/Kconfig | 3 + lib/Makefile | 2 + lib/llist.c | 129 ++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 260 insertions(+) create mode 100644 include/linux/llist.h create mode 100644 lib/llist.c diff --git a/include/linux/llist.h b/include/linux/llist.h new file mode 100644 index 000000000000..aa0c8b5b3cd0 --- /dev/null +++ b/include/linux/llist.h @@ -0,0 +1,126 @@ +#ifndef LLIST_H +#define LLIST_H +/* + * Lock-less NULL terminated single linked list + * + * If there are multiple producers and multiple consumers, llist_add + * can be used in producers and llist_del_all can be used in + * consumers. They can work simultaneously without lock. But + * llist_del_first can not be used here. Because llist_del_first + * depends on list->first->next does not changed if list->first is not + * changed during its operation, but llist_del_first, llist_add, + * llist_add (or llist_del_all, llist_add, llist_add) sequence in + * another consumer may violate that. + * + * If there are multiple producers and one consumer, llist_add can be + * used in producers and llist_del_all or llist_del_first can be used + * in the consumer. + * + * This can be summarized as follow: + * + * | add | del_first | del_all + * add | - | - | - + * del_first | | L | L + * del_all | | | - + * + * Where "-" stands for no lock is needed, while "L" stands for lock + * is needed. + * + * The list entries deleted via llist_del_all can be traversed with + * traversing function such as llist_for_each etc. But the list + * entries can not be traversed safely before deleted from the list. + * The order of deleted entries is from the newest to the oldest added + * one. If you want to traverse from the oldest to the newest, you + * must reverse the order by yourself before traversing. + * + * The basic atomic operation of this list is cmpxchg on long. On + * architectures that don't have NMI-safe cmpxchg implementation, the + * list can NOT be used in NMI handler. So code uses the list in NMI + * handler should depend on CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG. + */ + +struct llist_head { + struct llist_node *first; +}; + +struct llist_node { + struct llist_node *next; +}; + +#define LLIST_HEAD_INIT(name) { NULL } +#define LLIST_HEAD(name) struct llist_head name = LLIST_HEAD_INIT(name) + +/** + * init_llist_head - initialize lock-less list head + * @head: the head for your lock-less list + */ +static inline void init_llist_head(struct llist_head *list) +{ + list->first = NULL; +} + +/** + * llist_entry - get the struct of this entry + * @ptr: the &struct llist_node pointer. + * @type: the type of the struct this is embedded in. + * @member: the name of the llist_node within the struct. + */ +#define llist_entry(ptr, type, member) \ + container_of(ptr, type, member) + +/** + * llist_for_each - iterate over some deleted entries of a lock-less list + * @pos: the &struct llist_node to use as a loop cursor + * @node: the first entry of deleted list entries + * + * In general, some entries of the lock-less list can be traversed + * safely only after being deleted from list, so start with an entry + * instead of list head. + * + * If being used on entries deleted from lock-less list directly, the + * traverse order is from the newest to the oldest added entry. If + * you want to traverse from the oldest to the newest, you must + * reverse the order by yourself before traversing. + */ +#define llist_for_each(pos, node) \ + for ((pos) = (node); pos; (pos) = (pos)->next) + +/** + * llist_for_each_entry - iterate over some deleted entries of lock-less list of given type + * @pos: the type * to use as a loop cursor. + * @node: the fist entry of deleted list entries. + * @member: the name of the llist_node with the struct. + * + * In general, some entries of the lock-less list can be traversed + * safely only after being removed from list, so start with an entry + * instead of list head. + * + * If being used on entries deleted from lock-less list directly, the + * traverse order is from the newest to the oldest added entry. If + * you want to traverse from the oldest to the newest, you must + * reverse the order by yourself before traversing. + */ +#define llist_for_each_entry(pos, node, member) \ + for ((pos) = llist_entry((node), typeof(*(pos)), member); \ + &(pos)->member != NULL; \ + (pos) = llist_entry((pos)->member.next, typeof(*(pos)), member)) + +/** + * llist_empty - tests whether a lock-less list is empty + * @head: the list to test + * + * Not guaranteed to be accurate or up to date. Just a quick way to + * test whether the list is empty without deleting something from the + * list. + */ +static inline int llist_empty(const struct llist_head *head) +{ + return ACCESS_ONCE(head->first) == NULL; +} + +void llist_add(struct llist_node *new, struct llist_head *head); +void llist_add_batch(struct llist_node *new_first, struct llist_node *new_last, + struct llist_head *head); +struct llist_node *llist_del_first(struct llist_head *head); +struct llist_node *llist_del_all(struct llist_head *head); +#endif /* LLIST_H */ diff --git a/lib/Kconfig b/lib/Kconfig index 830181cc7a83..25c19678a30f 100644 --- a/lib/Kconfig +++ b/lib/Kconfig @@ -262,4 +262,7 @@ config AVERAGE If unsure, say N. +config LLIST + bool + endmenu diff --git a/lib/Makefile b/lib/Makefile index 6b597fdb1898..d770b817202e 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -112,6 +112,8 @@ obj-$(CONFIG_AVERAGE) += average.o obj-$(CONFIG_CPU_RMAP) += cpu_rmap.o +obj-$(CONFIG_LLIST) += llist.o + hostprogs-y := gen_crc32table clean-files := crc32table.h diff --git a/lib/llist.c b/lib/llist.c new file mode 100644 index 000000000000..da445724fa1f --- /dev/null +++ b/lib/llist.c @@ -0,0 +1,129 @@ +/* + * Lock-less NULL terminated single linked list + * + * The basic atomic operation of this list is cmpxchg on long. On + * architectures that don't have NMI-safe cmpxchg implementation, the + * list can NOT be used in NMI handler. So code uses the list in NMI + * handler should depend on CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG. + * + * Copyright 2010,2011 Intel Corp. + * Author: Huang Ying + * + * 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; + * + * 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 +#include +#include +#include + +#include + +/** + * llist_add - add a new entry + * @new: new entry to be added + * @head: the head for your lock-less list + */ +void llist_add(struct llist_node *new, struct llist_head *head) +{ + struct llist_node *entry, *old_entry; + +#ifndef CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG + BUG_ON(in_nmi()); +#endif + + entry = head->first; + do { + old_entry = entry; + new->next = entry; + cpu_relax(); + } while ((entry = cmpxchg(&head->first, old_entry, new)) != old_entry); +} +EXPORT_SYMBOL_GPL(llist_add); + +/** + * llist_add_batch - add several linked entries in batch + * @new_first: first entry in batch to be added + * @new_last: last entry in batch to be added + * @head: the head for your lock-less list + */ +void llist_add_batch(struct llist_node *new_first, struct llist_node *new_last, + struct llist_head *head) +{ + struct llist_node *entry, *old_entry; + +#ifndef CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG + BUG_ON(in_nmi()); +#endif + + entry = head->first; + do { + old_entry = entry; + new_last->next = entry; + cpu_relax(); + } while ((entry = cmpxchg(&head->first, old_entry, new_first)) != old_entry); +} +EXPORT_SYMBOL_GPL(llist_add_batch); + +/** + * llist_del_first - delete the first entry of lock-less list + * @head: the head for your lock-less list + * + * If list is empty, return NULL, otherwise, return the first entry + * deleted, this is the newest added one. + * + * Only one llist_del_first user can be used simultaneously with + * multiple llist_add users without lock. Because otherwise + * llist_del_first, llist_add, llist_add (or llist_del_all, llist_add, + * llist_add) sequence in another user may change @head->first->next, + * but keep @head->first. If multiple consumers are needed, please + * use llist_del_all or use lock between consumers. + */ +struct llist_node *llist_del_first(struct llist_head *head) +{ + struct llist_node *entry, *old_entry, *next; + +#ifndef CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG + BUG_ON(in_nmi()); +#endif + + entry = head->first; + do { + if (entry == NULL) + return NULL; + old_entry = entry; + next = entry->next; + cpu_relax(); + } while ((entry = cmpxchg(&head->first, old_entry, next)) != old_entry); + + return entry; +} +EXPORT_SYMBOL_GPL(llist_del_first); + +/** + * llist_del_all - delete all entries from lock-less list + * @head: the head of lock-less list to delete all entries + * + * If list is empty, return NULL, otherwise, delete all entries and + * return the pointer to the first entry. The order of entries + * deleted is from the newest to the oldest added one. + */ +struct llist_node *llist_del_all(struct llist_head *head) +{ +#ifndef CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG + BUG_ON(in_nmi()); +#endif + + return xchg(&head->first, NULL); +} +EXPORT_SYMBOL_GPL(llist_del_all); From 7f184275aa306046fe7edcbef3229754f0d97402 Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Wed, 13 Jul 2011 13:14:24 +0800 Subject: [PATCH 258/439] lib, Make gen_pool memory allocator lockless This version of the gen_pool memory allocator supports lockless operation. This makes it safe to use in NMI handlers and other special unblockable contexts that could otherwise deadlock on locks. This is implemented by using atomic operations and retries on any conflicts. The disadvantage is that there may be livelocks in extreme cases. For better scalability, one gen_pool allocator can be used for each CPU. The lockless operation only works if there is enough memory available. If new memory is added to the pool a lock has to be still taken. So any user relying on locklessness has to ensure that sufficient memory is preallocated. The basic atomic operation of this allocator is cmpxchg on long. On architectures that don't have NMI-safe cmpxchg implementation, the allocator can NOT be used in NMI handler. So code uses the allocator in NMI handler should depend on CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG. Signed-off-by: Huang Ying Reviewed-by: Andi Kleen Reviewed-by: Mathieu Desnoyers Cc: Andrew Morton Signed-off-by: Len Brown --- include/linux/bitmap.h | 1 + include/linux/genalloc.h | 34 ++++- lib/bitmap.c | 2 - lib/genalloc.c | 300 +++++++++++++++++++++++++++++++-------- 4 files changed, 272 insertions(+), 65 deletions(-) diff --git a/include/linux/bitmap.h b/include/linux/bitmap.h index dcafe0bf0005..907dd58aa228 100644 --- a/include/linux/bitmap.h +++ b/include/linux/bitmap.h @@ -145,6 +145,7 @@ extern void bitmap_release_region(unsigned long *bitmap, int pos, int order); extern int bitmap_allocate_region(unsigned long *bitmap, int pos, int order); extern void bitmap_copy_le(void *dst, const unsigned long *src, int nbits); +#define BITMAP_FIRST_WORD_MASK(start) (~0UL << ((start) % BITS_PER_LONG)) #define BITMAP_LAST_WORD_MASK(nbits) \ ( \ ((nbits) % BITS_PER_LONG) ? \ diff --git a/include/linux/genalloc.h b/include/linux/genalloc.h index 5bbebda78b02..5e98eeb2af3b 100644 --- a/include/linux/genalloc.h +++ b/include/linux/genalloc.h @@ -1,8 +1,26 @@ /* - * Basic general purpose allocator for managing special purpose memory - * not managed by the regular kmalloc/kfree interface. - * Uses for this includes on-device special memory, uncached memory - * etc. + * Basic general purpose allocator for managing special purpose + * memory, for example, memory that is not managed by the regular + * kmalloc/kfree interface. Uses for this includes on-device special + * memory, uncached memory etc. + * + * It is safe to use the allocator in NMI handlers and other special + * unblockable contexts that could otherwise deadlock on locks. This + * is implemented by using atomic operations and retries on any + * conflicts. The disadvantage is that there may be livelocks in + * extreme cases. For better scalability, one allocator can be used + * for each CPU. + * + * The lockless operation only works if there is enough memory + * available. If new memory is added to the pool a lock has to be + * still taken. So any user relying on locklessness has to ensure + * that sufficient memory is preallocated. + * + * The basic atomic operation of this allocator is cmpxchg on long. + * On architectures that don't have NMI-safe cmpxchg implementation, + * the allocator can NOT be used in NMI handler. So code uses the + * allocator in NMI handler should depend on + * CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG. * * This source code is licensed under the GNU General Public License, * Version 2. See the file COPYING for more details. @@ -15,7 +33,7 @@ * General purpose special memory pool descriptor. */ struct gen_pool { - rwlock_t lock; + spinlock_t lock; struct list_head chunks; /* list of chunks in this pool */ int min_alloc_order; /* minimum allocation order */ }; @@ -24,8 +42,8 @@ struct gen_pool { * General purpose special memory pool chunk descriptor. */ struct gen_pool_chunk { - spinlock_t lock; struct list_head next_chunk; /* next chunk in pool */ + atomic_t avail; phys_addr_t phys_addr; /* physical starting address of memory chunk */ unsigned long start_addr; /* starting address of memory chunk */ unsigned long end_addr; /* ending address of memory chunk */ @@ -56,4 +74,8 @@ static inline int gen_pool_add(struct gen_pool *pool, unsigned long addr, extern void gen_pool_destroy(struct gen_pool *); extern unsigned long gen_pool_alloc(struct gen_pool *, size_t); extern void gen_pool_free(struct gen_pool *, unsigned long, size_t); +extern void gen_pool_for_each_chunk(struct gen_pool *, + void (*)(struct gen_pool *, struct gen_pool_chunk *, void *), void *); +extern size_t gen_pool_avail(struct gen_pool *); +extern size_t gen_pool_size(struct gen_pool *); #endif /* __GENALLOC_H__ */ diff --git a/lib/bitmap.c b/lib/bitmap.c index 3f3b68199d74..e3c9e999501e 100644 --- a/lib/bitmap.c +++ b/lib/bitmap.c @@ -271,8 +271,6 @@ int __bitmap_weight(const unsigned long *bitmap, int bits) } EXPORT_SYMBOL(__bitmap_weight); -#define BITMAP_FIRST_WORD_MASK(start) (~0UL << ((start) % BITS_PER_LONG)) - void bitmap_set(unsigned long *map, int start, int nr) { unsigned long *p = map + BIT_WORD(start); diff --git a/lib/genalloc.c b/lib/genalloc.c index 577ddf805975..f352cc42f4f8 100644 --- a/lib/genalloc.c +++ b/lib/genalloc.c @@ -1,8 +1,26 @@ /* - * Basic general purpose allocator for managing special purpose memory - * not managed by the regular kmalloc/kfree interface. - * Uses for this includes on-device special memory, uncached memory - * etc. + * Basic general purpose allocator for managing special purpose + * memory, for example, memory that is not managed by the regular + * kmalloc/kfree interface. Uses for this includes on-device special + * memory, uncached memory etc. + * + * It is safe to use the allocator in NMI handlers and other special + * unblockable contexts that could otherwise deadlock on locks. This + * is implemented by using atomic operations and retries on any + * conflicts. The disadvantage is that there may be livelocks in + * extreme cases. For better scalability, one allocator can be used + * for each CPU. + * + * The lockless operation only works if there is enough memory + * available. If new memory is added to the pool a lock has to be + * still taken. So any user relying on locklessness has to ensure + * that sufficient memory is preallocated. + * + * The basic atomic operation of this allocator is cmpxchg on long. + * On architectures that don't have NMI-safe cmpxchg implementation, + * the allocator can NOT be used in NMI handler. So code uses the + * allocator in NMI handler should depend on + * CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG. * * Copyright 2005 (C) Jes Sorensen * @@ -13,8 +31,109 @@ #include #include #include +#include +#include #include +static int set_bits_ll(unsigned long *addr, unsigned long mask_to_set) +{ + unsigned long val, nval; + + nval = *addr; + do { + val = nval; + if (val & mask_to_set) + return -EBUSY; + cpu_relax(); + } while ((nval = cmpxchg(addr, val, val | mask_to_set)) != val); + + return 0; +} + +static int clear_bits_ll(unsigned long *addr, unsigned long mask_to_clear) +{ + unsigned long val, nval; + + nval = *addr; + do { + val = nval; + if ((val & mask_to_clear) != mask_to_clear) + return -EBUSY; + cpu_relax(); + } while ((nval = cmpxchg(addr, val, val & ~mask_to_clear)) != val); + + return 0; +} + +/* + * bitmap_set_ll - set the specified number of bits at the specified position + * @map: pointer to a bitmap + * @start: a bit position in @map + * @nr: number of bits to set + * + * Set @nr bits start from @start in @map lock-lessly. Several users + * can set/clear the same bitmap simultaneously without lock. If two + * users set the same bit, one user will return remain bits, otherwise + * return 0. + */ +static int bitmap_set_ll(unsigned long *map, int start, int nr) +{ + unsigned long *p = map + BIT_WORD(start); + const int size = start + nr; + int bits_to_set = BITS_PER_LONG - (start % BITS_PER_LONG); + unsigned long mask_to_set = BITMAP_FIRST_WORD_MASK(start); + + while (nr - bits_to_set >= 0) { + if (set_bits_ll(p, mask_to_set)) + return nr; + nr -= bits_to_set; + bits_to_set = BITS_PER_LONG; + mask_to_set = ~0UL; + p++; + } + if (nr) { + mask_to_set &= BITMAP_LAST_WORD_MASK(size); + if (set_bits_ll(p, mask_to_set)) + return nr; + } + + return 0; +} + +/* + * bitmap_clear_ll - clear the specified number of bits at the specified position + * @map: pointer to a bitmap + * @start: a bit position in @map + * @nr: number of bits to set + * + * Clear @nr bits start from @start in @map lock-lessly. Several users + * can set/clear the same bitmap simultaneously without lock. If two + * users clear the same bit, one user will return remain bits, + * otherwise return 0. + */ +static int bitmap_clear_ll(unsigned long *map, int start, int nr) +{ + unsigned long *p = map + BIT_WORD(start); + const int size = start + nr; + int bits_to_clear = BITS_PER_LONG - (start % BITS_PER_LONG); + unsigned long mask_to_clear = BITMAP_FIRST_WORD_MASK(start); + + while (nr - bits_to_clear >= 0) { + if (clear_bits_ll(p, mask_to_clear)) + return nr; + nr -= bits_to_clear; + bits_to_clear = BITS_PER_LONG; + mask_to_clear = ~0UL; + p++; + } + if (nr) { + mask_to_clear &= BITMAP_LAST_WORD_MASK(size); + if (clear_bits_ll(p, mask_to_clear)) + return nr; + } + + return 0; +} /** * gen_pool_create - create a new special memory pool @@ -30,7 +149,7 @@ struct gen_pool *gen_pool_create(int min_alloc_order, int nid) pool = kmalloc_node(sizeof(struct gen_pool), GFP_KERNEL, nid); if (pool != NULL) { - rwlock_init(&pool->lock); + spin_lock_init(&pool->lock); INIT_LIST_HEAD(&pool->chunks); pool->min_alloc_order = min_alloc_order; } @@ -63,14 +182,14 @@ int gen_pool_add_virt(struct gen_pool *pool, unsigned long virt, phys_addr_t phy if (unlikely(chunk == NULL)) return -ENOMEM; - spin_lock_init(&chunk->lock); chunk->phys_addr = phys; chunk->start_addr = virt; chunk->end_addr = virt + size; + atomic_set(&chunk->avail, size); - write_lock(&pool->lock); - list_add(&chunk->next_chunk, &pool->chunks); - write_unlock(&pool->lock); + spin_lock(&pool->lock); + list_add_rcu(&chunk->next_chunk, &pool->chunks); + spin_unlock(&pool->lock); return 0; } @@ -85,19 +204,19 @@ EXPORT_SYMBOL(gen_pool_add_virt); */ phys_addr_t gen_pool_virt_to_phys(struct gen_pool *pool, unsigned long addr) { - struct list_head *_chunk; struct gen_pool_chunk *chunk; + phys_addr_t paddr = -1; - read_lock(&pool->lock); - list_for_each(_chunk, &pool->chunks) { - chunk = list_entry(_chunk, struct gen_pool_chunk, next_chunk); - - if (addr >= chunk->start_addr && addr < chunk->end_addr) - return chunk->phys_addr + addr - chunk->start_addr; + rcu_read_lock(); + list_for_each_entry_rcu(chunk, &pool->chunks, next_chunk) { + if (addr >= chunk->start_addr && addr < chunk->end_addr) { + paddr = chunk->phys_addr + (addr - chunk->start_addr); + break; + } } - read_unlock(&pool->lock); + rcu_read_unlock(); - return -1; + return paddr; } EXPORT_SYMBOL(gen_pool_virt_to_phys); @@ -115,7 +234,6 @@ void gen_pool_destroy(struct gen_pool *pool) int order = pool->min_alloc_order; int bit, end_bit; - list_for_each_safe(_chunk, _next_chunk, &pool->chunks) { chunk = list_entry(_chunk, struct gen_pool_chunk, next_chunk); list_del(&chunk->next_chunk); @@ -137,44 +255,50 @@ EXPORT_SYMBOL(gen_pool_destroy); * @size: number of bytes to allocate from the pool * * Allocate the requested number of bytes from the specified pool. - * Uses a first-fit algorithm. + * Uses a first-fit algorithm. Can not be used in NMI handler on + * architectures without NMI-safe cmpxchg implementation. */ unsigned long gen_pool_alloc(struct gen_pool *pool, size_t size) { - struct list_head *_chunk; struct gen_pool_chunk *chunk; - unsigned long addr, flags; + unsigned long addr = 0; int order = pool->min_alloc_order; - int nbits, start_bit, end_bit; + int nbits, start_bit = 0, end_bit, remain; + +#ifndef CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG + BUG_ON(in_nmi()); +#endif if (size == 0) return 0; nbits = (size + (1UL << order) - 1) >> order; - - read_lock(&pool->lock); - list_for_each(_chunk, &pool->chunks) { - chunk = list_entry(_chunk, struct gen_pool_chunk, next_chunk); + rcu_read_lock(); + list_for_each_entry_rcu(chunk, &pool->chunks, next_chunk) { + if (size > atomic_read(&chunk->avail)) + continue; end_bit = (chunk->end_addr - chunk->start_addr) >> order; - - spin_lock_irqsave(&chunk->lock, flags); - start_bit = bitmap_find_next_zero_area(chunk->bits, end_bit, 0, - nbits, 0); - if (start_bit >= end_bit) { - spin_unlock_irqrestore(&chunk->lock, flags); +retry: + start_bit = bitmap_find_next_zero_area(chunk->bits, end_bit, + start_bit, nbits, 0); + if (start_bit >= end_bit) continue; + remain = bitmap_set_ll(chunk->bits, start_bit, nbits); + if (remain) { + remain = bitmap_clear_ll(chunk->bits, start_bit, + nbits - remain); + BUG_ON(remain); + goto retry; } addr = chunk->start_addr + ((unsigned long)start_bit << order); - - bitmap_set(chunk->bits, start_bit, nbits); - spin_unlock_irqrestore(&chunk->lock, flags); - read_unlock(&pool->lock); - return addr; + size = nbits << order; + atomic_sub(size, &chunk->avail); + break; } - read_unlock(&pool->lock); - return 0; + rcu_read_unlock(); + return addr; } EXPORT_SYMBOL(gen_pool_alloc); @@ -184,33 +308,95 @@ EXPORT_SYMBOL(gen_pool_alloc); * @addr: starting address of memory to free back to pool * @size: size in bytes of memory to free * - * Free previously allocated special memory back to the specified pool. + * Free previously allocated special memory back to the specified + * pool. Can not be used in NMI handler on architectures without + * NMI-safe cmpxchg implementation. */ void gen_pool_free(struct gen_pool *pool, unsigned long addr, size_t size) { - struct list_head *_chunk; struct gen_pool_chunk *chunk; - unsigned long flags; int order = pool->min_alloc_order; - int bit, nbits; + int start_bit, nbits, remain; + +#ifndef CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG + BUG_ON(in_nmi()); +#endif nbits = (size + (1UL << order) - 1) >> order; - - read_lock(&pool->lock); - list_for_each(_chunk, &pool->chunks) { - chunk = list_entry(_chunk, struct gen_pool_chunk, next_chunk); - + rcu_read_lock(); + list_for_each_entry_rcu(chunk, &pool->chunks, next_chunk) { if (addr >= chunk->start_addr && addr < chunk->end_addr) { BUG_ON(addr + size > chunk->end_addr); - spin_lock_irqsave(&chunk->lock, flags); - bit = (addr - chunk->start_addr) >> order; - while (nbits--) - __clear_bit(bit++, chunk->bits); - spin_unlock_irqrestore(&chunk->lock, flags); - break; + start_bit = (addr - chunk->start_addr) >> order; + remain = bitmap_clear_ll(chunk->bits, start_bit, nbits); + BUG_ON(remain); + size = nbits << order; + atomic_add(size, &chunk->avail); + rcu_read_unlock(); + return; } } - BUG_ON(nbits > 0); - read_unlock(&pool->lock); + rcu_read_unlock(); + BUG(); } EXPORT_SYMBOL(gen_pool_free); + +/** + * gen_pool_for_each_chunk - call func for every chunk of generic memory pool + * @pool: the generic memory pool + * @func: func to call + * @data: additional data used by @func + * + * Call @func for every chunk of generic memory pool. The @func is + * called with rcu_read_lock held. + */ +void gen_pool_for_each_chunk(struct gen_pool *pool, + void (*func)(struct gen_pool *pool, struct gen_pool_chunk *chunk, void *data), + void *data) +{ + struct gen_pool_chunk *chunk; + + rcu_read_lock(); + list_for_each_entry_rcu(chunk, &(pool)->chunks, next_chunk) + func(pool, chunk, data); + rcu_read_unlock(); +} +EXPORT_SYMBOL(gen_pool_for_each_chunk); + +/** + * gen_pool_avail - get available free space of the pool + * @pool: pool to get available free space + * + * Return available free space of the specified pool. + */ +size_t gen_pool_avail(struct gen_pool *pool) +{ + struct gen_pool_chunk *chunk; + size_t avail = 0; + + rcu_read_lock(); + list_for_each_entry_rcu(chunk, &pool->chunks, next_chunk) + avail += atomic_read(&chunk->avail); + rcu_read_unlock(); + return avail; +} +EXPORT_SYMBOL_GPL(gen_pool_avail); + +/** + * gen_pool_size - get size in bytes of memory managed by the pool + * @pool: pool to get size + * + * Return size in bytes of memory managed by the pool. + */ +size_t gen_pool_size(struct gen_pool *pool) +{ + struct gen_pool_chunk *chunk; + size_t size = 0; + + rcu_read_lock(); + list_for_each_entry_rcu(chunk, &pool->chunks, next_chunk) + size += chunk->end_addr - chunk->start_addr; + rcu_read_unlock(); + return size; +} +EXPORT_SYMBOL_GPL(gen_pool_size); From 67eb2e99076708cc790019a6a08ca3e0ae130a3a Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Wed, 13 Jul 2011 13:14:25 +0800 Subject: [PATCH 259/439] ACPI, APEI, GHES, printk support for recoverable error via NMI Some APEI GHES recoverable errors are reported via NMI, but printk is not safe in NMI context. To solve the issue, a lock-less memory allocator is used to allocate memory in NMI handler, save the error record into the allocated memory, put the error record into a lock-less list. On the other hand, an irq_work is used to delay the operation from NMI context to IRQ context. The irq_work IRQ handler will remove nodes from lock-less list, printk the error record and do some further processing include recovery operation, then free the memory. Signed-off-by: Huang Ying Signed-off-by: Len Brown --- drivers/acpi/apei/Kconfig | 2 + drivers/acpi/apei/ghes.c | 209 ++++++++++++++++++++++++++++++++++---- 2 files changed, 193 insertions(+), 18 deletions(-) diff --git a/drivers/acpi/apei/Kconfig b/drivers/acpi/apei/Kconfig index 3f45dde17aec..35596eaaca17 100644 --- a/drivers/acpi/apei/Kconfig +++ b/drivers/acpi/apei/Kconfig @@ -13,6 +13,8 @@ config ACPI_APEI_GHES bool "APEI Generic Hardware Error Source" depends on ACPI_APEI && X86 select ACPI_HED + select LLIST + select GENERIC_ALLOCATOR help Generic Hardware Error Source provides a way to report platform hardware errors (such as that from chipset). It diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c index b1390a61cde1..d1a40218e17e 100644 --- a/drivers/acpi/apei/ghes.c +++ b/drivers/acpi/apei/ghes.c @@ -12,7 +12,7 @@ * For more information about Generic Hardware Error Source, please * refer to ACPI Specification version 4.0, section 17.3.2.6 * - * Copyright 2010 Intel Corp. + * Copyright 2010,2011 Intel Corp. * Author: Huang Ying * * This program is free software; you can redistribute it and/or @@ -42,6 +42,9 @@ #include #include #include +#include +#include +#include #include #include #include @@ -53,6 +56,15 @@ #define GHES_PFX "GHES: " #define GHES_ESTATUS_MAX_SIZE 65536 +#define GHES_ESOURCE_PREALLOC_MAX_SIZE 65536 + +#define GHES_ESTATUS_POOL_MIN_ALLOC_ORDER 3 + +#define GHES_ESTATUS_NODE_LEN(estatus_len) \ + (sizeof(struct ghes_estatus_node) + (estatus_len)) +#define GHES_ESTATUS_FROM_NODE(estatus_node) \ + ((struct acpi_hest_generic_status *) \ + ((struct ghes_estatus_node *)(estatus_node) + 1)) /* * One struct ghes is created for each generic hardware error source. @@ -77,6 +89,11 @@ struct ghes { }; }; +struct ghes_estatus_node { + struct llist_node llnode; + struct acpi_hest_generic *generic; +}; + int ghes_disable; module_param_named(disable, ghes_disable, bool, 0); @@ -124,6 +141,19 @@ static struct vm_struct *ghes_ioremap_area; static DEFINE_RAW_SPINLOCK(ghes_ioremap_lock_nmi); static DEFINE_SPINLOCK(ghes_ioremap_lock_irq); +/* + * printk is not safe in NMI context. So in NMI handler, we allocate + * required memory from lock-less memory allocator + * (ghes_estatus_pool), save estatus into it, put them into lock-less + * list (ghes_estatus_llist), then delay printk into IRQ context via + * irq_work (ghes_proc_irq_work). ghes_estatus_size_request record + * required pool size by all NMI error source. + */ +static struct gen_pool *ghes_estatus_pool; +static unsigned long ghes_estatus_pool_size_request; +static struct llist_head ghes_estatus_llist; +static struct irq_work ghes_proc_irq_work; + static int ghes_ioremap_init(void) { ghes_ioremap_area = __get_vm_area(PAGE_SIZE * GHES_IOREMAP_PAGES, @@ -183,6 +213,55 @@ static void ghes_iounmap_irq(void __iomem *vaddr_ptr) __flush_tlb_one(vaddr); } +static int ghes_estatus_pool_init(void) +{ + ghes_estatus_pool = gen_pool_create(GHES_ESTATUS_POOL_MIN_ALLOC_ORDER, -1); + if (!ghes_estatus_pool) + return -ENOMEM; + return 0; +} + +static void ghes_estatus_pool_free_chunk_page(struct gen_pool *pool, + struct gen_pool_chunk *chunk, + void *data) +{ + free_page(chunk->start_addr); +} + +static void ghes_estatus_pool_exit(void) +{ + gen_pool_for_each_chunk(ghes_estatus_pool, + ghes_estatus_pool_free_chunk_page, NULL); + gen_pool_destroy(ghes_estatus_pool); +} + +static int ghes_estatus_pool_expand(unsigned long len) +{ + unsigned long i, pages, size, addr; + int ret; + + ghes_estatus_pool_size_request += PAGE_ALIGN(len); + size = gen_pool_size(ghes_estatus_pool); + if (size >= ghes_estatus_pool_size_request) + return 0; + pages = (ghes_estatus_pool_size_request - size) / PAGE_SIZE; + for (i = 0; i < pages; i++) { + addr = __get_free_page(GFP_KERNEL); + if (!addr) + return -ENOMEM; + ret = gen_pool_add(ghes_estatus_pool, addr, PAGE_SIZE, -1); + if (ret) + return ret; + } + + return 0; +} + +static void ghes_estatus_pool_shrink(unsigned long len) +{ + ghes_estatus_pool_size_request -= PAGE_ALIGN(len); +} + static struct ghes *ghes_new(struct acpi_hest_generic *generic) { struct ghes *ghes; @@ -344,13 +423,13 @@ static void ghes_clear_estatus(struct ghes *ghes) ghes->flags &= ~GHES_TO_CLEAR; } -static void ghes_do_proc(struct ghes *ghes) +static void ghes_do_proc(const struct acpi_hest_generic_status *estatus) { int sev, processed = 0; struct acpi_hest_generic_data *gdata; - sev = ghes_severity(ghes->estatus->error_severity); - apei_estatus_for_each_section(ghes->estatus, gdata) { + sev = ghes_severity(estatus->error_severity); + apei_estatus_for_each_section(estatus, gdata) { #ifdef CONFIG_X86_MCE if (!uuid_le_cmp(*(uuid_le *)gdata->section_type, CPER_SEC_PLATFORM_MEM)) { @@ -363,27 +442,37 @@ static void ghes_do_proc(struct ghes *ghes) } } -static void __ghes_print_estatus(const char *pfx, struct ghes *ghes) +static void __ghes_print_estatus(const char *pfx, + const struct acpi_hest_generic *generic, + const struct acpi_hest_generic_status *estatus) { if (pfx == NULL) { - if (ghes_severity(ghes->estatus->error_severity) <= + if (ghes_severity(estatus->error_severity) <= GHES_SEV_CORRECTED) pfx = KERN_WARNING HW_ERR; else pfx = KERN_ERR HW_ERR; } printk("%s""Hardware error from APEI Generic Hardware Error Source: %d\n", - pfx, ghes->generic->header.source_id); - apei_estatus_print(pfx, ghes->estatus); + pfx, generic->header.source_id); + apei_estatus_print(pfx, estatus); } -static void ghes_print_estatus(const char *pfx, struct ghes *ghes) +static void ghes_print_estatus(const char *pfx, + const struct acpi_hest_generic *generic, + const struct acpi_hest_generic_status *estatus) { /* Not more than 2 messages every 5 seconds */ - static DEFINE_RATELIMIT_STATE(ratelimit, 5*HZ, 2); + static DEFINE_RATELIMIT_STATE(ratelimit_corrected, 5*HZ, 2); + static DEFINE_RATELIMIT_STATE(ratelimit_uncorrected, 5*HZ, 2); + struct ratelimit_state *ratelimit; - if (__ratelimit(&ratelimit)) - __ghes_print_estatus(pfx, ghes); + if (ghes_severity(estatus->error_severity) <= GHES_SEV_CORRECTED) + ratelimit = &ratelimit_corrected; + else + ratelimit = &ratelimit_uncorrected; + if (__ratelimit(ratelimit)) + __ghes_print_estatus(pfx, generic, estatus); } static int ghes_proc(struct ghes *ghes) @@ -393,8 +482,8 @@ static int ghes_proc(struct ghes *ghes) rc = ghes_read_estatus(ghes, 0); if (rc) goto out; - ghes_print_estatus(NULL, ghes); - ghes_do_proc(ghes); + ghes_print_estatus(NULL, ghes->generic, ghes->estatus); + ghes_do_proc(ghes->estatus); out: ghes_clear_estatus(ghes); @@ -453,6 +542,40 @@ static int ghes_notify_sci(struct notifier_block *this, return ret; } +static void ghes_proc_in_irq(struct irq_work *irq_work) +{ + struct llist_node *llnode, *next, *tail = NULL; + struct ghes_estatus_node *estatus_node; + struct acpi_hest_generic_status *estatus; + u32 len, node_len; + + /* + * Because the time order of estatus in list is reversed, + * revert it back to proper order. + */ + llnode = llist_del_all(&ghes_estatus_llist); + while (llnode) { + next = llnode->next; + llnode->next = tail; + tail = llnode; + llnode = next; + } + llnode = tail; + while (llnode) { + next = llnode->next; + estatus_node = llist_entry(llnode, struct ghes_estatus_node, + llnode); + estatus = GHES_ESTATUS_FROM_NODE(estatus_node); + len = apei_estatus_len(estatus); + node_len = GHES_ESTATUS_NODE_LEN(len); + ghes_do_proc(estatus); + ghes_print_estatus(NULL, estatus_node->generic, estatus); + gen_pool_free(ghes_estatus_pool, (unsigned long)estatus_node, + node_len); + llnode = next; + } +} + static int ghes_notify_nmi(struct notifier_block *this, unsigned long cmd, void *data) { @@ -482,7 +605,8 @@ static int ghes_notify_nmi(struct notifier_block *this, if (sev_global >= GHES_SEV_PANIC) { oops_begin(); - __ghes_print_estatus(KERN_EMERG HW_ERR, ghes_global); + __ghes_print_estatus(KERN_EMERG HW_ERR, ghes_global->generic, + ghes_global->estatus); /* reboot to log the error! */ if (panic_timeout == 0) panic_timeout = ghes_panic_timeout; @@ -490,12 +614,31 @@ static int ghes_notify_nmi(struct notifier_block *this, } list_for_each_entry_rcu(ghes, &ghes_nmi, list) { +#ifdef CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG + u32 len, node_len; + struct ghes_estatus_node *estatus_node; + struct acpi_hest_generic_status *estatus; +#endif if (!(ghes->flags & GHES_TO_CLEAR)) continue; - /* Do not print estatus because printk is not NMI safe */ - ghes_do_proc(ghes); +#ifdef CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG + /* Save estatus for further processing in IRQ context */ + len = apei_estatus_len(ghes->estatus); + node_len = GHES_ESTATUS_NODE_LEN(len); + estatus_node = (void *)gen_pool_alloc(ghes_estatus_pool, + node_len); + if (estatus_node) { + estatus_node->generic = ghes->generic; + estatus = GHES_ESTATUS_FROM_NODE(estatus_node); + memcpy(estatus, ghes->estatus, len); + llist_add(&estatus_node->llnode, &ghes_estatus_llist); + } +#endif ghes_clear_estatus(ghes); } +#ifdef CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG + irq_work_queue(&ghes_proc_irq_work); +#endif out: raw_spin_unlock(&ghes_nmi_lock); @@ -510,10 +653,26 @@ static struct notifier_block ghes_notifier_nmi = { .notifier_call = ghes_notify_nmi, }; +static unsigned long ghes_esource_prealloc_size( + const struct acpi_hest_generic *generic) +{ + unsigned long block_length, prealloc_records, prealloc_size; + + block_length = min_t(unsigned long, generic->error_block_length, + GHES_ESTATUS_MAX_SIZE); + prealloc_records = max_t(unsigned long, + generic->records_to_preallocate, 1); + prealloc_size = min_t(unsigned long, block_length * prealloc_records, + GHES_ESOURCE_PREALLOC_MAX_SIZE); + + return prealloc_size; +} + static int __devinit ghes_probe(struct platform_device *ghes_dev) { struct acpi_hest_generic *generic; struct ghes *ghes = NULL; + unsigned long len; int rc = -EINVAL; generic = *(struct acpi_hest_generic **)ghes_dev->dev.platform_data; @@ -579,6 +738,8 @@ static int __devinit ghes_probe(struct platform_device *ghes_dev) mutex_unlock(&ghes_list_mutex); break; case ACPI_HEST_NOTIFY_NMI: + len = ghes_esource_prealloc_size(generic); + ghes_estatus_pool_expand(len); mutex_lock(&ghes_list_mutex); if (list_empty(&ghes_nmi)) register_die_notifier(&ghes_notifier_nmi); @@ -603,6 +764,7 @@ static int __devexit ghes_remove(struct platform_device *ghes_dev) { struct ghes *ghes; struct acpi_hest_generic *generic; + unsigned long len; ghes = platform_get_drvdata(ghes_dev); generic = ghes->generic; @@ -633,6 +795,8 @@ static int __devexit ghes_remove(struct platform_device *ghes_dev) * freed after NMI handler finishes. */ synchronize_rcu(); + len = ghes_esource_prealloc_size(generic); + ghes_estatus_pool_shrink(len); break; default: BUG(); @@ -673,14 +837,20 @@ static int __init ghes_init(void) return -EINVAL; } + init_irq_work(&ghes_proc_irq_work, ghes_proc_in_irq); + rc = ghes_ioremap_init(); if (rc) goto err; - rc = platform_driver_register(&ghes_platform_driver); + rc = ghes_estatus_pool_init(); if (rc) goto err_ioremap_exit; + rc = platform_driver_register(&ghes_platform_driver); + if (rc) + goto err_pool_exit; + rc = apei_osc_setup(); if (rc == 0 && osc_sb_apei_support_acked) pr_info(GHES_PFX "APEI firmware first mode is enabled by APEI bit and WHEA _OSC.\n"); @@ -692,6 +862,8 @@ static int __init ghes_init(void) pr_info(GHES_PFX "Failed to enable APEI firmware first mode.\n"); return 0; +err_pool_exit: + ghes_estatus_pool_exit(); err_ioremap_exit: ghes_ioremap_exit(); err: @@ -701,6 +873,7 @@ err: static void __exit ghes_exit(void) { platform_driver_unregister(&ghes_platform_driver); + ghes_estatus_pool_exit(); ghes_ioremap_exit(); } From 152cef40a808d3034e383465b3f7d6783613e458 Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Wed, 13 Jul 2011 13:14:26 +0800 Subject: [PATCH 260/439] ACPI, APEI, GHES, Error records content based throttle printk is used by GHES to report hardware errors. Ratelimit is enforced on the printk to avoid too many hardware error reports in kernel log. Because there may be thousands or even millions of corrected hardware errors during system running. Currently, a simple scheme is used. That is, the total number of hardware error reporting is ratelimited. This may cause some issues in practice. For example, there are two kinds of hardware errors occurred in system. One is corrected memory error, because the fault memory address is accessed frequently, there may be hundreds error report per-second. The other is corrected PCIe AER error, it will be reported once per-second. Because they share one ratelimit control structure, it is highly possible that only memory error is reported. To avoid the above issue, an error record content based throttle algorithm is implemented in the patch. Where after the first successful reporting, all error records that are same are throttled for some time, to let other kinds of error records have the opportunity to be reported. In above example, the memory errors will be throttled for some time, after being printked. Then the PCIe AER error will be printked successfully. Signed-off-by: Huang Ying Signed-off-by: Len Brown --- drivers/acpi/apei/ghes.c | 184 +++++++++++++++++++++++++++++++++++++-- 1 file changed, 177 insertions(+), 7 deletions(-) diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c index d1a40218e17e..931410d31a96 100644 --- a/drivers/acpi/apei/ghes.c +++ b/drivers/acpi/apei/ghes.c @@ -60,6 +60,21 @@ #define GHES_ESTATUS_POOL_MIN_ALLOC_ORDER 3 +/* This is just an estimation for memory pool allocation */ +#define GHES_ESTATUS_CACHE_AVG_SIZE 512 + +#define GHES_ESTATUS_CACHES_SIZE 4 + +#define GHES_ESTATUS_IN_CACHE_MAX_NSEC (10 * NSEC_PER_SEC) +/* Prevent too many caches are allocated because of RCU */ +#define GHES_ESTATUS_CACHE_ALLOCED_MAX (GHES_ESTATUS_CACHES_SIZE * 3 / 2) + +#define GHES_ESTATUS_CACHE_LEN(estatus_len) \ + (sizeof(struct ghes_estatus_cache) + (estatus_len)) +#define GHES_ESTATUS_FROM_CACHE(estatus_cache) \ + ((struct acpi_hest_generic_status *) \ + ((struct ghes_estatus_cache *)(estatus_cache) + 1)) + #define GHES_ESTATUS_NODE_LEN(estatus_len) \ (sizeof(struct ghes_estatus_node) + (estatus_len)) #define GHES_ESTATUS_FROM_NODE(estatus_node) \ @@ -94,6 +109,14 @@ struct ghes_estatus_node { struct acpi_hest_generic *generic; }; +struct ghes_estatus_cache { + u32 estatus_len; + atomic_t count; + struct acpi_hest_generic *generic; + unsigned long long time_in; + struct rcu_head rcu; +}; + int ghes_disable; module_param_named(disable, ghes_disable, bool, 0); @@ -154,6 +177,9 @@ static unsigned long ghes_estatus_pool_size_request; static struct llist_head ghes_estatus_llist; static struct irq_work ghes_proc_irq_work; +struct ghes_estatus_cache *ghes_estatus_caches[GHES_ESTATUS_CACHES_SIZE]; +static atomic_t ghes_estatus_cache_alloced; + static int ghes_ioremap_init(void) { ghes_ioremap_area = __get_vm_area(PAGE_SIZE * GHES_IOREMAP_PAGES, @@ -458,9 +484,9 @@ static void __ghes_print_estatus(const char *pfx, apei_estatus_print(pfx, estatus); } -static void ghes_print_estatus(const char *pfx, - const struct acpi_hest_generic *generic, - const struct acpi_hest_generic_status *estatus) +static int ghes_print_estatus(const char *pfx, + const struct acpi_hest_generic *generic, + const struct acpi_hest_generic_status *estatus) { /* Not more than 2 messages every 5 seconds */ static DEFINE_RATELIMIT_STATE(ratelimit_corrected, 5*HZ, 2); @@ -471,8 +497,137 @@ static void ghes_print_estatus(const char *pfx, ratelimit = &ratelimit_corrected; else ratelimit = &ratelimit_uncorrected; - if (__ratelimit(ratelimit)) + if (__ratelimit(ratelimit)) { __ghes_print_estatus(pfx, generic, estatus); + return 1; + } + return 0; +} + +/* + * GHES error status reporting throttle, to report more kinds of + * errors, instead of just most frequently occurred errors. + */ +static int ghes_estatus_cached(struct acpi_hest_generic_status *estatus) +{ + u32 len; + int i, cached = 0; + unsigned long long now; + struct ghes_estatus_cache *cache; + struct acpi_hest_generic_status *cache_estatus; + + len = apei_estatus_len(estatus); + rcu_read_lock(); + for (i = 0; i < GHES_ESTATUS_CACHES_SIZE; i++) { + cache = rcu_dereference(ghes_estatus_caches[i]); + if (cache == NULL) + continue; + if (len != cache->estatus_len) + continue; + cache_estatus = GHES_ESTATUS_FROM_CACHE(cache); + if (memcmp(estatus, cache_estatus, len)) + continue; + atomic_inc(&cache->count); + now = sched_clock(); + if (now - cache->time_in < GHES_ESTATUS_IN_CACHE_MAX_NSEC) + cached = 1; + break; + } + rcu_read_unlock(); + return cached; +} + +static struct ghes_estatus_cache *ghes_estatus_cache_alloc( + struct acpi_hest_generic *generic, + struct acpi_hest_generic_status *estatus) +{ + int alloced; + u32 len, cache_len; + struct ghes_estatus_cache *cache; + struct acpi_hest_generic_status *cache_estatus; + + alloced = atomic_add_return(1, &ghes_estatus_cache_alloced); + if (alloced > GHES_ESTATUS_CACHE_ALLOCED_MAX) { + atomic_dec(&ghes_estatus_cache_alloced); + return NULL; + } + len = apei_estatus_len(estatus); + cache_len = GHES_ESTATUS_CACHE_LEN(len); + cache = (void *)gen_pool_alloc(ghes_estatus_pool, cache_len); + if (!cache) { + atomic_dec(&ghes_estatus_cache_alloced); + return NULL; + } + cache_estatus = GHES_ESTATUS_FROM_CACHE(cache); + memcpy(cache_estatus, estatus, len); + cache->estatus_len = len; + atomic_set(&cache->count, 0); + cache->generic = generic; + cache->time_in = sched_clock(); + return cache; +} + +static void ghes_estatus_cache_free(struct ghes_estatus_cache *cache) +{ + u32 len; + + len = apei_estatus_len(GHES_ESTATUS_FROM_CACHE(cache)); + len = GHES_ESTATUS_CACHE_LEN(len); + gen_pool_free(ghes_estatus_pool, (unsigned long)cache, len); + atomic_dec(&ghes_estatus_cache_alloced); +} + +static void ghes_estatus_cache_rcu_free(struct rcu_head *head) +{ + struct ghes_estatus_cache *cache; + + cache = container_of(head, struct ghes_estatus_cache, rcu); + ghes_estatus_cache_free(cache); +} + +static void ghes_estatus_cache_add( + struct acpi_hest_generic *generic, + struct acpi_hest_generic_status *estatus) +{ + int i, slot = -1, count; + unsigned long long now, duration, period, max_period = 0; + struct ghes_estatus_cache *cache, *slot_cache = NULL, *new_cache; + + new_cache = ghes_estatus_cache_alloc(generic, estatus); + if (new_cache == NULL) + return; + rcu_read_lock(); + now = sched_clock(); + for (i = 0; i < GHES_ESTATUS_CACHES_SIZE; i++) { + cache = rcu_dereference(ghes_estatus_caches[i]); + if (cache == NULL) { + slot = i; + slot_cache = NULL; + break; + } + duration = now - cache->time_in; + if (duration >= GHES_ESTATUS_IN_CACHE_MAX_NSEC) { + slot = i; + slot_cache = cache; + break; + } + count = atomic_read(&cache->count); + period = duration / (count + 1); + if (period > max_period) { + max_period = period; + slot = i; + slot_cache = cache; + } + } + /* new_cache must be put into array after its contents are written */ + smp_wmb(); + if (slot != -1 && cmpxchg(ghes_estatus_caches + slot, + slot_cache, new_cache) == slot_cache) { + if (slot_cache) + call_rcu(&slot_cache->rcu, ghes_estatus_cache_rcu_free); + } else + ghes_estatus_cache_free(new_cache); + rcu_read_unlock(); } static int ghes_proc(struct ghes *ghes) @@ -482,9 +637,11 @@ static int ghes_proc(struct ghes *ghes) rc = ghes_read_estatus(ghes, 0); if (rc) goto out; - ghes_print_estatus(NULL, ghes->generic, ghes->estatus); + if (!ghes_estatus_cached(ghes->estatus)) { + if (ghes_print_estatus(NULL, ghes->generic, ghes->estatus)) + ghes_estatus_cache_add(ghes->generic, ghes->estatus); + } ghes_do_proc(ghes->estatus); - out: ghes_clear_estatus(ghes); return 0; @@ -546,6 +703,7 @@ static void ghes_proc_in_irq(struct irq_work *irq_work) { struct llist_node *llnode, *next, *tail = NULL; struct ghes_estatus_node *estatus_node; + struct acpi_hest_generic *generic; struct acpi_hest_generic_status *estatus; u32 len, node_len; @@ -569,7 +727,11 @@ static void ghes_proc_in_irq(struct irq_work *irq_work) len = apei_estatus_len(estatus); node_len = GHES_ESTATUS_NODE_LEN(len); ghes_do_proc(estatus); - ghes_print_estatus(NULL, estatus_node->generic, estatus); + if (!ghes_estatus_cached(estatus)) { + generic = estatus_node->generic; + if (ghes_print_estatus(NULL, generic, estatus)) + ghes_estatus_cache_add(generic, estatus); + } gen_pool_free(ghes_estatus_pool, (unsigned long)estatus_node, node_len); llnode = next; @@ -622,6 +784,8 @@ static int ghes_notify_nmi(struct notifier_block *this, if (!(ghes->flags & GHES_TO_CLEAR)) continue; #ifdef CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG + if (ghes_estatus_cached(ghes->estatus)) + goto next; /* Save estatus for further processing in IRQ context */ len = apei_estatus_len(ghes->estatus); node_len = GHES_ESTATUS_NODE_LEN(len); @@ -633,6 +797,7 @@ static int ghes_notify_nmi(struct notifier_block *this, memcpy(estatus, ghes->estatus, len); llist_add(&estatus_node->llnode, &ghes_estatus_llist); } +next: #endif ghes_clear_estatus(ghes); } @@ -847,6 +1012,11 @@ static int __init ghes_init(void) if (rc) goto err_ioremap_exit; + rc = ghes_estatus_pool_expand(GHES_ESTATUS_CACHE_AVG_SIZE * + GHES_ESTATUS_CACHE_ALLOCED_MAX); + if (rc) + goto err_pool_exit; + rc = platform_driver_register(&ghes_platform_driver); if (rc) goto err_pool_exit; From ea8f5fb8a71fddaf5f3a17100d3247855701f732 Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Wed, 13 Jul 2011 13:14:27 +0800 Subject: [PATCH 261/439] HWPoison: add memory_failure_queue() memory_failure() is the entry point for HWPoison memory error recovery. It must be called in process context. But commonly hardware memory errors are notified via MCE or NMI, so some delayed execution mechanism must be used. In MCE handler, a work queue + ring buffer mechanism is used. In addition to MCE, now APEI (ACPI Platform Error Interface) GHES (Generic Hardware Error Source) can be used to report memory errors too. To add support to APEI GHES memory recovery, a mechanism similar to that of MCE is implemented. memory_failure_queue() is the new entry point that can be called in IRQ context. The next step is to make MCE handler uses this interface too. Signed-off-by: Huang Ying Cc: Andi Kleen Cc: Wu Fengguang Cc: Andrew Morton Signed-off-by: Len Brown --- include/linux/mm.h | 1 + mm/memory-failure.c | 92 +++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 93 insertions(+) diff --git a/include/linux/mm.h b/include/linux/mm.h index 9670f71d7be9..303108eed28c 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1633,6 +1633,7 @@ enum mf_flags { }; extern void memory_failure(unsigned long pfn, int trapno); extern int __memory_failure(unsigned long pfn, int trapno, int flags); +extern void memory_failure_queue(unsigned long pfn, int trapno, int flags); extern int unpoison_memory(unsigned long pfn); extern int sysctl_memory_failure_early_kill; extern int sysctl_memory_failure_recovery; diff --git a/mm/memory-failure.c b/mm/memory-failure.c index 740c4f52059c..2b43ba051ac9 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -53,6 +53,7 @@ #include #include #include +#include #include "internal.h" int sysctl_memory_failure_early_kill __read_mostly = 0; @@ -1178,6 +1179,97 @@ void memory_failure(unsigned long pfn, int trapno) __memory_failure(pfn, trapno, 0); } +#define MEMORY_FAILURE_FIFO_ORDER 4 +#define MEMORY_FAILURE_FIFO_SIZE (1 << MEMORY_FAILURE_FIFO_ORDER) + +struct memory_failure_entry { + unsigned long pfn; + int trapno; + int flags; +}; + +struct memory_failure_cpu { + DECLARE_KFIFO(fifo, struct memory_failure_entry, + MEMORY_FAILURE_FIFO_SIZE); + spinlock_t lock; + struct work_struct work; +}; + +static DEFINE_PER_CPU(struct memory_failure_cpu, memory_failure_cpu); + +/** + * memory_failure_queue - Schedule handling memory failure of a page. + * @pfn: Page Number of the corrupted page + * @trapno: Trap number reported in the signal to user space. + * @flags: Flags for memory failure handling + * + * This function is called by the low level hardware error handler + * when it detects hardware memory corruption of a page. It schedules + * the recovering of error page, including dropping pages, killing + * processes etc. + * + * The function is primarily of use for corruptions that + * happen outside the current execution context (e.g. when + * detected by a background scrubber) + * + * Can run in IRQ context. + */ +void memory_failure_queue(unsigned long pfn, int trapno, int flags) +{ + struct memory_failure_cpu *mf_cpu; + unsigned long proc_flags; + struct memory_failure_entry entry = { + .pfn = pfn, + .trapno = trapno, + .flags = flags, + }; + + mf_cpu = &get_cpu_var(memory_failure_cpu); + spin_lock_irqsave(&mf_cpu->lock, proc_flags); + if (kfifo_put(&mf_cpu->fifo, &entry)) + schedule_work_on(smp_processor_id(), &mf_cpu->work); + else + pr_err("Memory failure: buffer overflow when queuing memory failure at 0x%#lx\n", + pfn); + spin_unlock_irqrestore(&mf_cpu->lock, proc_flags); + put_cpu_var(memory_failure_cpu); +} +EXPORT_SYMBOL_GPL(memory_failure_queue); + +static void memory_failure_work_func(struct work_struct *work) +{ + struct memory_failure_cpu *mf_cpu; + struct memory_failure_entry entry = { 0, }; + unsigned long proc_flags; + int gotten; + + mf_cpu = &__get_cpu_var(memory_failure_cpu); + for (;;) { + spin_lock_irqsave(&mf_cpu->lock, proc_flags); + gotten = kfifo_get(&mf_cpu->fifo, &entry); + spin_unlock_irqrestore(&mf_cpu->lock, proc_flags); + if (!gotten) + break; + __memory_failure(entry.pfn, entry.trapno, entry.flags); + } +} + +static int __init memory_failure_init(void) +{ + struct memory_failure_cpu *mf_cpu; + int cpu; + + for_each_possible_cpu(cpu) { + mf_cpu = &per_cpu(memory_failure_cpu, cpu); + spin_lock_init(&mf_cpu->lock); + INIT_KFIFO(mf_cpu->fifo); + INIT_WORK(&mf_cpu->work, memory_failure_work_func); + } + + return 0; +} +core_initcall(memory_failure_init); + /** * unpoison_memory - Unpoison a previously poisoned page * @pfn: Page number of the to be unpoisoned page From ba61ca4aab47441f1c6cec28a9a6aa0489fd1df3 Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Wed, 13 Jul 2011 13:14:28 +0800 Subject: [PATCH 262/439] ACPI, APEI, GHES: Add hardware memory error recovery support memory_failure_queue() is called when recoverable memory errors are notified by firmware to do the recovery work. Signed-off-by: Huang Ying Signed-off-by: Len Brown --- drivers/acpi/apei/Kconfig | 7 +++++++ drivers/acpi/apei/ghes.c | 24 +++++++++++++++++------- 2 files changed, 24 insertions(+), 7 deletions(-) diff --git a/drivers/acpi/apei/Kconfig b/drivers/acpi/apei/Kconfig index 35596eaaca17..c34aa51af4ee 100644 --- a/drivers/acpi/apei/Kconfig +++ b/drivers/acpi/apei/Kconfig @@ -32,6 +32,13 @@ config ACPI_APEI_PCIEAER PCIe AER errors may be reported via APEI firmware first mode. Turn on this option to enable the corresponding support. +config ACPI_APEI_MEMORY_FAILURE + bool "APEI memory error recovering support" + depends on ACPI_APEI && MEMORY_FAILURE + help + Memory errors may be reported via APEI firmware first mode. + Turn on this option to enable the memory recovering support. + config ACPI_APEI_EINJ tristate "APEI Error INJection (EINJ)" depends on ACPI_APEI && DEBUG_FS diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c index 931410d31a96..e92c47c46f91 100644 --- a/drivers/acpi/apei/ghes.c +++ b/drivers/acpi/apei/ghes.c @@ -451,20 +451,30 @@ static void ghes_clear_estatus(struct ghes *ghes) static void ghes_do_proc(const struct acpi_hest_generic_status *estatus) { - int sev, processed = 0; + int sev, sec_sev; struct acpi_hest_generic_data *gdata; sev = ghes_severity(estatus->error_severity); apei_estatus_for_each_section(estatus, gdata) { -#ifdef CONFIG_X86_MCE + sec_sev = ghes_severity(gdata->error_severity); if (!uuid_le_cmp(*(uuid_le *)gdata->section_type, CPER_SEC_PLATFORM_MEM)) { - apei_mce_report_mem_error( - sev == GHES_SEV_CORRECTED, - (struct cper_sec_mem_err *)(gdata+1)); - processed = 1; - } + struct cper_sec_mem_err *mem_err; + mem_err = (struct cper_sec_mem_err *)(gdata+1); +#ifdef CONFIG_X86_MCE + apei_mce_report_mem_error(sev == GHES_SEV_CORRECTED, + mem_err); #endif +#ifdef CONFIG_ACPI_APEI_MEMORY_FAILURE + if (sev == GHES_SEV_RECOVERABLE && + sec_sev == GHES_SEV_RECOVERABLE && + mem_err->validation_bits & CPER_MEM_VALID_PHYSICAL_ADDRESS) { + unsigned long pfn; + pfn = mem_err->physical_addr >> PAGE_SHIFT; + memory_failure_queue(pfn, 0, 0); + } +#endif + } } } From a7e09d450b2e0b068e850d103b6ee1af537d1910 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Sat, 16 Jul 2011 18:14:21 -0400 Subject: [PATCH 263/439] ACPI: APEI build fix as GHES is optional... When # CONFIG_ACPI_APEI_GHES is not set: (.init.text+0x4c22): undefined reference to `ghes_disable' Reported-by: Randy Dunlap Acked-by: Randy Dunlap Signed-off-by: Len Brown --- drivers/acpi/bus.c | 2 -- include/acpi/apei.h | 4 ++++ 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/acpi/bus.c b/drivers/acpi/bus.c index 43ce3b0c4921..437ddbf0c49a 100644 --- a/drivers/acpi/bus.c +++ b/drivers/acpi/bus.c @@ -544,10 +544,8 @@ static void acpi_bus_osc_support(void) capbuf[OSC_SUPPORT_TYPE] |= OSC_SB_PPC_OST_SUPPORT; #endif -#ifdef CONFIG_ACPI_APEI_GHES if (!ghes_disable) capbuf[OSC_SUPPORT_TYPE] |= OSC_SB_APEI_SUPPORT; -#endif if (ACPI_FAILURE(acpi_get_handle(NULL, "\\_SB", &handle))) return; if (ACPI_SUCCESS(acpi_run_osc(handle, &context))) { diff --git a/include/acpi/apei.h b/include/acpi/apei.h index d40bc5521fcb..51a527d24a8a 100644 --- a/include/acpi/apei.h +++ b/include/acpi/apei.h @@ -18,7 +18,11 @@ extern int hest_disable; extern int erst_disable; +#ifdef CONFIG_ACPI_APEI_GHES extern int ghes_disable; +#else +#define ghes_disable 1 +#endif #ifdef CONFIG_ACPI_APEI void __init acpi_hest_init(void); From 70cb6e1da00db6c9212e6fd69bd96fd41c797077 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Tue, 2 Aug 2011 18:00:21 -0400 Subject: [PATCH 264/439] APEI GHES: 32-bit buildfix MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit drivers/acpi/apei/ghes.c:542: warning: integer overflow in expression drivers/acpi/apei/ghes.c:619: warning: integer overflow in expression ghes.c:(.text+0x46289): undefined reference to `__udivdi3'   in function ghes_estatus_cache_add(). Reported-by: Randy Dunlap Signed-off-by: Len Brown --- drivers/acpi/apei/ghes.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c index e92c47c46f91..0784f99a4665 100644 --- a/drivers/acpi/apei/ghes.c +++ b/drivers/acpi/apei/ghes.c @@ -65,7 +65,7 @@ #define GHES_ESTATUS_CACHES_SIZE 4 -#define GHES_ESTATUS_IN_CACHE_MAX_NSEC (10 * NSEC_PER_SEC) +#define GHES_ESTATUS_IN_CACHE_MAX_NSEC 10000000000ULL /* Prevent too many caches are allocated because of RCU */ #define GHES_ESTATUS_CACHE_ALLOCED_MAX (GHES_ESTATUS_CACHES_SIZE * 3 / 2) @@ -622,7 +622,8 @@ static void ghes_estatus_cache_add( break; } count = atomic_read(&cache->count); - period = duration / (count + 1); + period = duration; + do_div(period, (count + 1)); if (period > max_period) { max_period = period; slot = i; From c3e6088e1036f8084bc7444b38437da136b7588b Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Wed, 20 Jul 2011 16:09:29 +0800 Subject: [PATCH 265/439] ACPI, APEI, EINJ Param support is disabled by default EINJ parameter support is only usable for some specific BIOS. Originally, it is expected to have no harm for BIOS does not support it. But now, we found it will cause issue (memory overwriting) for some BIOS. So param support is disabled by default and only enabled when newly added module parameter named "param_extension" is explicitly specified. Signed-off-by: Huang Ying Cc: Matthew Garrett Acked-by: Don Zickus Acked-by: Tony Luck Signed-off-by: Len Brown --- Documentation/acpi/apei/einj.txt | 11 +++++++-- drivers/acpi/apei/einj.c | 39 ++++++++++++++++++++------------ 2 files changed, 33 insertions(+), 17 deletions(-) diff --git a/Documentation/acpi/apei/einj.txt b/Documentation/acpi/apei/einj.txt index dfab71848dc8..5cc699ba5453 100644 --- a/Documentation/acpi/apei/einj.txt +++ b/Documentation/acpi/apei/einj.txt @@ -48,12 +48,19 @@ directory apei/einj. The following files are provided. - param1 This file is used to set the first error parameter value. Effect of parameter depends on error_type specified. For memory error, this is - physical memory address. + physical memory address. Only available if param_extension module + parameter is specified. - param2 This file is used to set the second error parameter value. Effect of parameter depends on error_type specified. For memory error, this is - physical memory address mask. + physical memory address mask. Only available if param_extension + module parameter is specified. + +Injecting parameter support is a BIOS version specific extension, that +is, it only works on some BIOS version. If you want to use it, please +make sure your BIOS version has the proper support and specify +"param_extension=y" in module parameter. For more information about EINJ, please refer to ACPI specification version 4.0, section 17.5. diff --git a/drivers/acpi/apei/einj.c b/drivers/acpi/apei/einj.c index ab821f1cfa1f..589b96c38704 100644 --- a/drivers/acpi/apei/einj.c +++ b/drivers/acpi/apei/einj.c @@ -46,7 +46,8 @@ * Some BIOSes allow parameters to the SET_ERROR_TYPE entries in the * EINJ table through an unpublished extension. Use with caution as * most will ignore the parameter and make their own choice of address - * for error injection. + * for error injection. This extension is used only if + * param_extension module parameter is specified. */ struct einj_parameter { u64 type; @@ -65,6 +66,9 @@ struct einj_parameter { ((struct acpi_whea_header *)((char *)(tab) + \ sizeof(struct acpi_table_einj))) +static bool param_extension; +module_param(param_extension, bool, 0); + static struct acpi_table_einj *einj_tab; static struct apei_resources einj_resources; @@ -489,14 +493,6 @@ static int __init einj_init(void) einj_debug_dir, NULL, &error_type_fops); if (!fentry) goto err_cleanup; - fentry = debugfs_create_x64("param1", S_IRUSR | S_IWUSR, - einj_debug_dir, &error_param1); - if (!fentry) - goto err_cleanup; - fentry = debugfs_create_x64("param2", S_IRUSR | S_IWUSR, - einj_debug_dir, &error_param2); - if (!fentry) - goto err_cleanup; fentry = debugfs_create_file("error_inject", S_IWUSR, einj_debug_dir, NULL, &error_inject_fops); if (!fentry) @@ -513,12 +509,23 @@ static int __init einj_init(void) rc = apei_exec_pre_map_gars(&ctx); if (rc) goto err_release; - param_paddr = einj_get_parameter_address(); - if (param_paddr) { - einj_param = ioremap(param_paddr, sizeof(*einj_param)); - rc = -ENOMEM; - if (!einj_param) - goto err_unmap; + if (param_extension) { + param_paddr = einj_get_parameter_address(); + if (param_paddr) { + einj_param = ioremap(param_paddr, sizeof(*einj_param)); + rc = -ENOMEM; + if (!einj_param) + goto err_unmap; + fentry = debugfs_create_x64("param1", S_IRUSR | S_IWUSR, + einj_debug_dir, &error_param1); + if (!fentry) + goto err_unmap; + fentry = debugfs_create_x64("param2", S_IRUSR | S_IWUSR, + einj_debug_dir, &error_param2); + if (!fentry) + goto err_unmap; + } else + pr_warn(EINJ_PFX "Parameter extension is not supported.\n"); } pr_info(EINJ_PFX "Error INJection is initialized.\n"); @@ -526,6 +533,8 @@ static int __init einj_init(void) return 0; err_unmap: + if (einj_param) + iounmap(einj_param); apei_exec_post_unmap_gars(&ctx); err_release: apei_resources_release(&einj_resources); From 4edd17a25c99f34bd7a75c1daf31afe840237da8 Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Wed, 3 Aug 2011 10:34:58 -0700 Subject: [PATCH 266/439] Revert "drm/i915/dp: Zero the DPCD data before connection probe" This reverts commit 97cdd7101079adc3c626d159c62d43de949516c8. Clearing the dpcd data means that if the fetch fails, any previous data will be lost. On eDP, this is no fun as we only fetch dpcd at init time, so the memset will destroy that the next time through. --- drivers/gpu/drm/i915/intel_dp.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index ba72fbcbbbc9..2f901c05ea2c 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -1697,7 +1697,6 @@ intel_dp_detect(struct drm_connector *connector, bool force) struct edid *edid = NULL; intel_dp->has_audio = false; - memset(intel_dp->dpcd, 0, sizeof(intel_dp->dpcd)); if (HAS_PCH_SPLIT(dev)) status = ironlake_dp_detect(intel_dp); From 4e20fa65a3ea789510eed1a15deb9e8aab2b8202 Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Wed, 3 Aug 2011 10:52:24 -0700 Subject: [PATCH 267/439] drm/i915: Try enabling RC6 by default (again) Jesse Barnes and I found a couple of issues where incorrect mode setting would cause problems with RC6 enabled. We're hopeful that fixing those will resolve the outstanding issues with a few machines that had trouble before 3.0 with rc6. Cc: Pekka Enberg Cc: Francesco Allertsen Cc: Ted Phelps Cc: Gu Rui Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=38567 Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=38332 Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/i915_drv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index ce045a8cf82c..60e4b9ef618b 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -62,7 +62,7 @@ module_param_named(semaphores, i915_semaphores, int, 0600); MODULE_PARM_DESC(semaphores, "Use semaphores for inter-ring sync (default: false)"); -unsigned int i915_enable_rc6 __read_mostly = 0; +unsigned int i915_enable_rc6 __read_mostly = 1; module_param_named(i915_enable_rc6, i915_enable_rc6, int, 0600); MODULE_PARM_DESC(i915_enable_rc6, "Enable power-saving render C-state 6 (default: true)"); From 27a26b775ab89ff65d307de180e466d4c1215186 Mon Sep 17 00:00:00 2001 From: Huang Weiyi Date: Sat, 23 Jul 2011 16:21:57 +0800 Subject: [PATCH 268/439] xen/pciback: remove duplicated #include Remove duplicated #include('s) in drivers/xen/xen-pciback/xenbus.c Signed-off-by: Huang Weiyi Signed-off-by: Konrad Rzeszutek Wilk --- drivers/xen/xen-pciback/xenbus.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/xen/xen-pciback/xenbus.c b/drivers/xen/xen-pciback/xenbus.c index 206c4ce030bc..978d2c6f5dca 100644 --- a/drivers/xen/xen-pciback/xenbus.c +++ b/drivers/xen/xen-pciback/xenbus.c @@ -11,7 +11,6 @@ #include #include #include -#include #include "pciback.h" #define DRV_NAME "xen-pciback" From f24144c0c397fe3697b5b563c4a86d4f1fb185cc Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Fri, 22 Jul 2011 14:00:06 -0400 Subject: [PATCH 269/439] xen/grant: Fix compile warning. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit drivers/xen/grant-table.c:85: warning: ‘rc’ may be used uninitialized in this function Signed-off-by: Konrad Rzeszutek Wilk --- drivers/xen/grant-table.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c index fd725cde6ad1..4f44b347b24a 100644 --- a/drivers/xen/grant-table.c +++ b/drivers/xen/grant-table.c @@ -82,7 +82,7 @@ static inline grant_ref_t *__gnttab_entry(grant_ref_t entry) static int get_free_entries(unsigned count) { unsigned long flags; - int ref, rc; + int ref, rc = 0; grant_ref_t head; spin_lock_irqsave(&gnttab_list_lock, flags); From 61077b2c50476ac85ced8a56fc6b5aaa3aa9c980 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Fri, 22 Jul 2011 14:01:16 -0400 Subject: [PATCH 270/439] xen/balloon: Fix compile errors - missing header files. With a specific enough .config file compile errors show for missing workqueue declarations. Reported-by: Randy Dunlap Signed-off-by: Konrad Rzeszutek Wilk --- drivers/xen/xen-selfballoon.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/xen/xen-selfballoon.c b/drivers/xen/xen-selfballoon.c index 010937b5a7c9..1b4afd81f872 100644 --- a/drivers/xen/xen-selfballoon.c +++ b/drivers/xen/xen-selfballoon.c @@ -70,10 +70,10 @@ #include #include #include - +#include #include - #include +#include /* Enable/disable with sysfs. */ static int xen_selfballooning_enabled __read_mostly; From 65d4b248114e18b3d805c7ecb88d9ea64dd978e4 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Sat, 30 Jul 2011 11:21:09 -0400 Subject: [PATCH 271/439] xen/self-balloon: Add dependency on tmem. Without enabling CONFIG_XEN_TMEM we get this: drivers/xen/xen-selfballoon.c:461: undefined reference to `tmem_enabled' Signed-off-by: Konrad Rzeszutek Wilk --- drivers/xen/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/xen/Kconfig b/drivers/xen/Kconfig index f815283667af..5f7ff8e2fc14 100644 --- a/drivers/xen/Kconfig +++ b/drivers/xen/Kconfig @@ -11,7 +11,7 @@ config XEN_BALLOON config XEN_SELFBALLOONING bool "Dynamically self-balloon kernel memory to target" - depends on XEN && XEN_BALLOON && CLEANCACHE && SWAP + depends on XEN && XEN_BALLOON && CLEANCACHE && SWAP && XEN_TMEM default n help Self-ballooning dynamically balloons available kernel memory driven From db9481c0476c6475d058ac7ecebb5a822b43cc99 Mon Sep 17 00:00:00 2001 From: Mathias Krause Date: Wed, 3 Aug 2011 14:57:11 -0400 Subject: [PATCH 272/439] ext4: use kzalloc in ext4_kzalloc() Commit 9933fc0i (ext4: introduce ext4_kvmalloc(), ext4_kzalloc(), and ext4_kvfree()) intruduced wrappers around k*alloc/vmalloc but introduced a typo for ext4_kzalloc() by not using kzalloc() but kmalloc(). Signed-off-by: Mathias Krause Signed-off-by: "Theodore Ts'o" --- fs/ext4/super.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/ext4/super.c b/fs/ext4/super.c index e2d88baf91d3..4687fea0c00f 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -124,7 +124,7 @@ void *ext4_kvzalloc(size_t size, gfp_t flags) { void *ret; - ret = kmalloc(size, flags); + ret = kzalloc(size, flags); if (!ret) ret = __vmalloc(size, flags | __GFP_ZERO, PAGE_KERNEL); return ret; From 0193e072268fe62c4b19ad4b05cd0d4b23c43bb9 Mon Sep 17 00:00:00 2001 From: Pavel Shilovsky Date: Wed, 3 Aug 2011 23:12:18 +0400 Subject: [PATCH 273/439] CIFS: Fix missing a decrement of inFlight value if we failed on getting mid entry in cifs_call_async. Cc: stable@kernel.org Signed-off-by: Pavel Shilovsky Signed-off-by: Steve French --- fs/cifs/transport.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 147aa22c3c3a..c1b9c4b10739 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -362,6 +362,8 @@ cifs_call_async(struct TCP_Server_Info *server, struct kvec *iov, mid = AllocMidQEntry(hdr, server); if (mid == NULL) { mutex_unlock(&server->srv_mutex); + atomic_dec(&server->inFlight); + wake_up(&server->request_q); return -ENOMEM; } From 6dccf9c508d5d773859df1cc2dce75c5b19e35a0 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Tue, 12 Jul 2011 22:29:32 -0400 Subject: [PATCH 274/439] mrst_pmu: driver for Intel Moorestown Power Management Unit The Moorestown (MRST) Power Management Unit (PMU) driver directs the SOC power states in the "Langwell" south complex (SCU). It hooks pci_platform_pm_ops[] and thus observes all PCI ".set_state" requests. For devices in the SC, the pmu driver translates those PCI requests into the appropriate commands for the SCU. The PMU driver helps implement S0i3, a deep system idle power idle state. Entry into S0i3 is via cpuidle, just like regular processor c-states. S0i3 depends on pre-conditions including uni-processor, graphics off, and certain IO devices in the SC must be off. If those pre-conditions are met, then the PMU allows cpuidle to enter S0i3, otherwise such requests are demoted, either to Atom C4 or Atom C6. This driver is based on prototype work by Bruce Flemming, Illyas Mansoor, Rajeev D. Muralidhar, Vishwesh M. Rudramuni, Hari Seshadri and Sujith Thomas. The current driver also includes contributions from H. Peter Anvin, Arjan van de Ven, Kristen Accardi, and Yong Wang. Thanks for additional review feedback from Alan Cox and Randy Dunlap. Acked-by: Alan Cox Acked-by: H. Peter Anvin Signed-off-by: Len Brown --- MAINTAINERS | 6 + arch/x86/platform/mrst/Makefile | 1 + arch/x86/platform/mrst/pmu.c | 817 ++++++++++++++++++++++++++++++++ arch/x86/platform/mrst/pmu.h | 234 +++++++++ 4 files changed, 1058 insertions(+) create mode 100644 arch/x86/platform/mrst/pmu.c create mode 100644 arch/x86/platform/mrst/pmu.h diff --git a/MAINTAINERS b/MAINTAINERS index 187282da9213..d37b387b5109 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3344,6 +3344,12 @@ F: drivers/net/ixgb/ F: drivers/net/ixgbe/ F: drivers/net/ixgbevf/ +INTEL MRST PMU DRIVER +M: Len Brown +L: linux-pm@lists.linux-foundation.org +S: Supported +F: arch/x86/platform/mrst/pmu.* + INTEL PRO/WIRELESS 2100 NETWORK CONNECTION SUPPORT L: linux-wireless@vger.kernel.org S: Orphan diff --git a/arch/x86/platform/mrst/Makefile b/arch/x86/platform/mrst/Makefile index f61ccdd49341..1ea38775a6d3 100644 --- a/arch/x86/platform/mrst/Makefile +++ b/arch/x86/platform/mrst/Makefile @@ -1,3 +1,4 @@ obj-$(CONFIG_X86_MRST) += mrst.o obj-$(CONFIG_X86_MRST) += vrtc.o obj-$(CONFIG_EARLY_PRINTK_MRST) += early_printk_mrst.o +obj-$(CONFIG_X86_MRST) += pmu.o diff --git a/arch/x86/platform/mrst/pmu.c b/arch/x86/platform/mrst/pmu.c new file mode 100644 index 000000000000..9281da7d91bd --- /dev/null +++ b/arch/x86/platform/mrst/pmu.c @@ -0,0 +1,817 @@ +/* + * mrst/pmu.c - driver for MRST Power Management Unit + * + * Copyright (c) 2011, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope 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., + * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "pmu.h" + +#define IPCMSG_FW_REVISION 0xF4 + +struct mrst_device { + u16 pci_dev_num; /* DEBUG only */ + u16 lss; + u16 latest_request; + unsigned int pci_state_counts[PCI_D3cold + 1]; /* DEBUG only */ +}; + +/* + * comlete list of MRST PCI devices + */ +static struct mrst_device mrst_devs[] = { +/* 0 */ { 0x0800, LSS_SPI0 }, /* Moorestown SPI Ctrl 0 */ +/* 1 */ { 0x0801, LSS_SPI1 }, /* Moorestown SPI Ctrl 1 */ +/* 2 */ { 0x0802, LSS_I2C0 }, /* Moorestown I2C 0 */ +/* 3 */ { 0x0803, LSS_I2C1 }, /* Moorestown I2C 1 */ +/* 4 */ { 0x0804, LSS_I2C2 }, /* Moorestown I2C 2 */ +/* 5 */ { 0x0805, LSS_KBD }, /* Moorestown Keyboard Ctrl */ +/* 6 */ { 0x0806, LSS_USB_HC }, /* Moorestown USB Ctrl */ +/* 7 */ { 0x0807, LSS_SD_HC0 }, /* Moorestown SD Host Ctrl 0 */ +/* 8 */ { 0x0808, LSS_SD_HC1 }, /* Moorestown SD Host Ctrl 1 */ +/* 9 */ { 0x0809, LSS_NAND }, /* Moorestown NAND Ctrl */ +/* 10 */ { 0x080a, LSS_AUDIO }, /* Moorestown Audio Ctrl */ +/* 11 */ { 0x080b, LSS_IMAGING }, /* Moorestown ISP */ +/* 12 */ { 0x080c, LSS_SECURITY }, /* Moorestown Security Controller */ +/* 13 */ { 0x080d, LSS_DISPLAY }, /* Moorestown External Displays */ +/* 14 */ { 0x080e, 0 }, /* Moorestown SCU IPC */ +/* 15 */ { 0x080f, LSS_GPIO }, /* Moorestown GPIO Controller */ +/* 16 */ { 0x0810, 0 }, /* Moorestown Power Management Unit */ +/* 17 */ { 0x0811, LSS_USB_OTG }, /* Moorestown OTG Ctrl */ +/* 18 */ { 0x0812, LSS_SPI2 }, /* Moorestown SPI Ctrl 2 */ +/* 19 */ { 0x0813, 0 }, /* Moorestown SC DMA */ +/* 20 */ { 0x0814, LSS_AUDIO_LPE }, /* Moorestown LPE DMA */ +/* 21 */ { 0x0815, LSS_AUDIO_SSP }, /* Moorestown SSP0 */ + +/* 22 */ { 0x084F, LSS_SD_HC2 }, /* Moorestown SD Host Ctrl 2 */ + +/* 23 */ { 0x4102, 0 }, /* Lincroft */ +/* 24 */ { 0x4110, 0 }, /* Lincroft */ +}; + +/* n.b. We ignore PCI-id 0x815 in LSS9 b/c MeeGo has no driver for it */ +static u16 mrst_lss9_pci_ids[] = {0x080a, 0x0814, 0}; +static u16 mrst_lss10_pci_ids[] = {0x0800, 0x0801, 0x0802, 0x0803, + 0x0804, 0x0805, 0x080f, 0}; + +/* handle concurrent SMP invokations of pmu_pci_set_power_state() */ +static spinlock_t mrst_pmu_power_state_lock; + +static unsigned int wake_counters[MRST_NUM_LSS]; /* DEBUG only */ +static unsigned int pmu_irq_stats[INT_INVALID + 1]; /* DEBUG only */ + +static int graphics_is_off; +static int lss_s0i3_enabled; +static bool mrst_pmu_s0i3_enable; + +/* debug counters */ +static u32 pmu_wait_ready_calls; +static u32 pmu_wait_ready_udelays; +static u32 pmu_wait_ready_udelays_max; +static u32 pmu_wait_done_calls; +static u32 pmu_wait_done_udelays; +static u32 pmu_wait_done_udelays_max; +static u32 pmu_set_power_state_entry; +static u32 pmu_set_power_state_send_cmd; + +static struct mrst_device *pci_id_2_mrst_dev(u16 pci_dev_num) +{ + int index = 0; + + if ((pci_dev_num >= 0x0800) && (pci_dev_num <= 0x815)) + index = pci_dev_num - 0x800; + else if (pci_dev_num == 0x084F) + index = 22; + else if (pci_dev_num == 0x4102) + index = 23; + else if (pci_dev_num == 0x4110) + index = 24; + + if (pci_dev_num != mrst_devs[index].pci_dev_num) { + WARN_ONCE(1, FW_BUG "Unknown PCI device 0x%04X\n", pci_dev_num); + return 0; + } + + return &mrst_devs[index]; +} + +/** + * mrst_pmu_validate_cstates + * @dev: cpuidle_device + * + * Certain states are not appropriate for governor to pick in some cases. + * This function will be called as cpuidle_device's prepare callback and + * thus tells governor to ignore such states when selecting the next state + * to enter. + */ + +#define IDLE_STATE4_IS_C6 4 +#define IDLE_STATE5_IS_S0I3 5 + +int mrst_pmu_invalid_cstates(void) +{ + int cpu = smp_processor_id(); + + /* + * Demote to C4 if the PMU is busy. + * Since LSS changes leave the busy bit clear... + * busy means either the PMU is waiting for an ACK-C6 that + * isn't coming due to an MWAIT that returned immediately; + * or we returned from S0i3 successfully, and the PMU + * is not done sending us interrupts. + */ + if (pmu_read_busy_status()) + return 1 << IDLE_STATE4_IS_C6 | 1 << IDLE_STATE5_IS_S0I3; + + /* + * Disallow S0i3 if: PMU is not initialized, or CPU1 is active, + * or if device LSS is insufficient, or the GPU is active, + * or if it has been explicitly disabled. + */ + if (!pmu_reg || !cpumask_equal(cpu_online_mask, cpumask_of(cpu)) || + !lss_s0i3_enabled || !graphics_is_off || !mrst_pmu_s0i3_enable) + return 1 << IDLE_STATE5_IS_S0I3; + else + return 0; +} + +/* + * pmu_update_wake_counters(): read PM_WKS, update wake_counters[] + * DEBUG only. + */ +static void pmu_update_wake_counters(void) +{ + int lss; + u32 wake_status; + + wake_status = pmu_read_wks(); + + for (lss = 0; lss < MRST_NUM_LSS; ++lss) { + if (wake_status & (1 << lss)) + wake_counters[lss]++; + } +} + +int mrst_pmu_s0i3_entry(void) +{ + int status; + + /* Clear any possible error conditions */ + pmu_write_ics(0x300); + + /* set wake control to current D-states */ + pmu_write_wssc(S0I3_SSS_TARGET); + + status = mrst_s0i3_entry(PM_S0I3_COMMAND, &pmu_reg->pm_cmd); + pmu_update_wake_counters(); + return status; +} + +/* poll for maximum of 5ms for busy bit to clear */ +static int pmu_wait_ready(void) +{ + int udelays; + + pmu_wait_ready_calls++; + + for (udelays = 0; udelays < 500; ++udelays) { + if (udelays > pmu_wait_ready_udelays_max) + pmu_wait_ready_udelays_max = udelays; + + if (pmu_read_busy_status() == 0) + return 0; + + udelay(10); + pmu_wait_ready_udelays++; + } + + /* + * if this fires, observe + * /sys/kernel/debug/mrst_pmu_wait_ready_calls + * /sys/kernel/debug/mrst_pmu_wait_ready_udelays + */ + WARN_ONCE(1, "SCU not ready for 5ms"); + return -EBUSY; +} +/* poll for maximum of 50ms us for busy bit to clear */ +static int pmu_wait_done(void) +{ + int udelays; + + pmu_wait_done_calls++; + + for (udelays = 0; udelays < 500; ++udelays) { + if (udelays > pmu_wait_done_udelays_max) + pmu_wait_done_udelays_max = udelays; + + if (pmu_read_busy_status() == 0) + return 0; + + udelay(100); + pmu_wait_done_udelays++; + } + + /* + * if this fires, observe + * /sys/kernel/debug/mrst_pmu_wait_done_calls + * /sys/kernel/debug/mrst_pmu_wait_done_udelays + */ + WARN_ONCE(1, "SCU not done for 50ms"); + return -EBUSY; +} + +u32 mrst_pmu_msi_is_disabled(void) +{ + return pmu_msi_is_disabled(); +} + +void mrst_pmu_enable_msi(void) +{ + pmu_msi_enable(); +} + +/** + * pmu_irq - pmu driver interrupt handler + * Context: interrupt context + */ +static irqreturn_t pmu_irq(int irq, void *dummy) +{ + union pmu_pm_ics pmu_ics; + + pmu_ics.value = pmu_read_ics(); + + if (!pmu_ics.bits.pending) + return IRQ_NONE; + + switch (pmu_ics.bits.cause) { + case INT_SPURIOUS: + case INT_CMD_DONE: + case INT_CMD_ERR: + case INT_WAKE_RX: + case INT_SS_ERROR: + case INT_S0IX_MISS: + case INT_NO_ACKC6: + pmu_irq_stats[pmu_ics.bits.cause]++; + break; + default: + pmu_irq_stats[INT_INVALID]++; + } + + pmu_write_ics(pmu_ics.value); /* Clear pending interrupt */ + + return IRQ_HANDLED; +} + +/* + * Translate PCI power management to MRST LSS D-states + */ +static int pci_2_mrst_state(int lss, pci_power_t pci_state) +{ + switch (pci_state) { + case PCI_D0: + if (SSMSK(D0i1, lss) & D0I1_ACG_SSS_TARGET) + return D0i1; + else + return D0; + case PCI_D1: + return D0i1; + case PCI_D2: + return D0i2; + case PCI_D3hot: + case PCI_D3cold: + return D0i3; + default: + WARN(1, "pci_state %d\n", pci_state); + return 0; + } +} + +static int pmu_issue_command(u32 pm_ssc) +{ + union pmu_pm_set_cfg_cmd_t command; + + if (pmu_read_busy_status()) { + pr_debug("pmu is busy, Operation not permitted\n"); + return -1; + } + + /* + * enable interrupts in PMU so that interrupts are + * propagated when ioc bit for a particular set + * command is set + */ + + pmu_irq_enable(); + + /* Configure the sub systems for pmu2 */ + + pmu_write_ssc(pm_ssc); + + /* + * Send the set config command for pmu its configured + * for mode CM_IMMEDIATE & hence with No Trigger + */ + + command.pmu2_params.d_param.cfg_mode = CM_IMMEDIATE; + command.pmu2_params.d_param.cfg_delay = 0; + command.pmu2_params.d_param.rsvd = 0; + + /* construct the command to send SET_CFG to particular PMU */ + command.pmu2_params.d_param.cmd = SET_CFG_CMD; + command.pmu2_params.d_param.ioc = 0; + command.pmu2_params.d_param.mode_id = 0; + command.pmu2_params.d_param.sys_state = SYS_STATE_S0I0; + + /* write the value of PM_CMD into particular PMU */ + pr_debug("pmu command being written %x\n", + command.pmu_pm_set_cfg_cmd_value); + + pmu_write_cmd(command.pmu_pm_set_cfg_cmd_value); + + return 0; +} + +static u16 pmu_min_lss_pci_req(u16 *ids, u16 pci_state) +{ + u16 existing_request; + int i; + + for (i = 0; ids[i]; ++i) { + struct mrst_device *mrst_dev; + + mrst_dev = pci_id_2_mrst_dev(ids[i]); + if (unlikely(!mrst_dev)) + continue; + + existing_request = mrst_dev->latest_request; + if (existing_request < pci_state) + pci_state = existing_request; + } + return pci_state; +} + +/** + * pmu_pci_set_power_state - Callback function is used by all the PCI devices + * for a platform specific device power on/shutdown. + */ + +int pmu_pci_set_power_state(struct pci_dev *pdev, pci_power_t pci_state) +{ + u32 old_sss, new_sss; + int status = 0; + struct mrst_device *mrst_dev; + + pmu_set_power_state_entry++; + + BUG_ON(pdev->vendor != PCI_VENDOR_ID_INTEL); + BUG_ON(pci_state < PCI_D0 || pci_state > PCI_D3cold); + + mrst_dev = pci_id_2_mrst_dev(pdev->device); + if (unlikely(!mrst_dev)) + return -ENODEV; + + mrst_dev->pci_state_counts[pci_state]++; /* count invocations */ + + /* PMU driver calls self as part of PCI initialization, ignore */ + if (pdev->device == PCI_DEV_ID_MRST_PMU) + return 0; + + BUG_ON(!pmu_reg); /* SW bug if called before initialized */ + + spin_lock(&mrst_pmu_power_state_lock); + + if (pdev->d3_delay) { + dev_dbg(&pdev->dev, "d3_delay %d, should be 0\n", + pdev->d3_delay); + pdev->d3_delay = 0; + } + /* + * If Lincroft graphics, simply remember state + */ + if ((pdev->class >> 16) == PCI_BASE_CLASS_DISPLAY + && !((pdev->class & PCI_SUB_CLASS_MASK) >> 8)) { + if (pci_state == PCI_D0) + graphics_is_off = 0; + else + graphics_is_off = 1; + goto ret; + } + + if (!mrst_dev->lss) + goto ret; /* device with no LSS */ + + if (mrst_dev->latest_request == pci_state) + goto ret; /* no change */ + + mrst_dev->latest_request = pci_state; /* record latest request */ + + /* + * LSS9 and LSS10 contain multiple PCI devices. + * Use the lowest numbered (highest power) state in the LSS + */ + if (mrst_dev->lss == 9) + pci_state = pmu_min_lss_pci_req(mrst_lss9_pci_ids, pci_state); + else if (mrst_dev->lss == 10) + pci_state = pmu_min_lss_pci_req(mrst_lss10_pci_ids, pci_state); + + status = pmu_wait_ready(); + if (status) + goto ret; + + old_sss = pmu_read_sss(); + new_sss = old_sss & ~SSMSK(3, mrst_dev->lss); + new_sss |= SSMSK(pci_2_mrst_state(mrst_dev->lss, pci_state), + mrst_dev->lss); + + if (new_sss == old_sss) + goto ret; /* nothing to do */ + + pmu_set_power_state_send_cmd++; + + status = pmu_issue_command(new_sss); + + if (unlikely(status != 0)) { + dev_err(&pdev->dev, "Failed to Issue a PM command\n"); + goto ret; + } + + if (pmu_wait_done()) + goto ret; + + lss_s0i3_enabled = + ((pmu_read_sss() & S0I3_SSS_TARGET) == S0I3_SSS_TARGET); +ret: + spin_unlock(&mrst_pmu_power_state_lock); + return status; +} + +#ifdef CONFIG_DEBUG_FS +static char *d0ix_names[] = {"D0", "D0i1", "D0i2", "D0i3"}; + +static inline const char *d0ix_name(int state) +{ + return d0ix_names[(int) state]; +} + +static int debug_mrst_pmu_show(struct seq_file *s, void *unused) +{ + struct pci_dev *pdev = NULL; + u32 cur_pmsss; + int lss; + + seq_printf(s, "0x%08X D0I1_ACG_SSS_TARGET\n", D0I1_ACG_SSS_TARGET); + + cur_pmsss = pmu_read_sss(); + + seq_printf(s, "0x%08X S0I3_SSS_TARGET\n", S0I3_SSS_TARGET); + + seq_printf(s, "0x%08X Current SSS ", cur_pmsss); + seq_printf(s, lss_s0i3_enabled ? "\n" : "[BLOCKS s0i3]\n"); + + if (cpumask_equal(cpu_online_mask, cpumask_of(0))) + seq_printf(s, "cpu0 is only cpu online\n"); + else + seq_printf(s, "cpu0 is NOT only cpu online [BLOCKS S0i3]\n"); + + seq_printf(s, "GFX: %s\n", graphics_is_off ? "" : "[BLOCKS s0i3]"); + + + for_each_pci_dev(pdev) { + int pos; + u16 pmcsr; + struct mrst_device *mrst_dev; + int i; + + mrst_dev = pci_id_2_mrst_dev(pdev->device); + + seq_printf(s, "%s %04x/%04X %-16.16s ", + dev_name(&pdev->dev), + pdev->vendor, pdev->device, + dev_driver_string(&pdev->dev)); + + if (unlikely (!mrst_dev)) { + seq_printf(s, " UNKNOWN\n"); + continue; + } + + if (mrst_dev->lss) + seq_printf(s, "LSS %2d %-4s ", mrst_dev->lss, + d0ix_name(((cur_pmsss >> + (mrst_dev->lss * 2)) & 0x3))); + else + seq_printf(s, " "); + + /* PCI PM config space setting */ + pos = pci_find_capability(pdev, PCI_CAP_ID_PM); + if (pos != 0) { + pci_read_config_word(pdev, pos + PCI_PM_CTRL, &pmcsr); + seq_printf(s, "PCI-%-4s", + pci_power_name(pmcsr & PCI_PM_CTRL_STATE_MASK)); + } else { + seq_printf(s, " "); + } + + seq_printf(s, " %s ", pci_power_name(mrst_dev->latest_request)); + for (i = 0; i <= PCI_D3cold; ++i) + seq_printf(s, "%d ", mrst_dev->pci_state_counts[i]); + + if (mrst_dev->lss) { + unsigned int lssmask; + + lssmask = SSMSK(D0i3, mrst_dev->lss); + + if ((lssmask & S0I3_SSS_TARGET) && + ((lssmask & cur_pmsss) != + (lssmask & S0I3_SSS_TARGET))) + seq_printf(s , "[BLOCKS s0i3]"); + } + + seq_printf(s, "\n"); + } + seq_printf(s, "Wake Counters:\n"); + for (lss = 0; lss < MRST_NUM_LSS; ++lss) + seq_printf(s, "LSS%d %d\n", lss, wake_counters[lss]); + + seq_printf(s, "Interrupt Counters:\n"); + seq_printf(s, + "INT_SPURIOUS \t%8u\n" "INT_CMD_DONE \t%8u\n" + "INT_CMD_ERR \t%8u\n" "INT_WAKE_RX \t%8u\n" + "INT_SS_ERROR \t%8u\n" "INT_S0IX_MISS\t%8u\n" + "INT_NO_ACKC6 \t%8u\n" "INT_INVALID \t%8u\n", + pmu_irq_stats[INT_SPURIOUS], pmu_irq_stats[INT_CMD_DONE], + pmu_irq_stats[INT_CMD_ERR], pmu_irq_stats[INT_WAKE_RX], + pmu_irq_stats[INT_SS_ERROR], pmu_irq_stats[INT_S0IX_MISS], + pmu_irq_stats[INT_NO_ACKC6], pmu_irq_stats[INT_INVALID]); + + seq_printf(s, "mrst_pmu_wait_ready_calls %8d\n", + pmu_wait_ready_calls); + seq_printf(s, "mrst_pmu_wait_ready_udelays %8d\n", + pmu_wait_ready_udelays); + seq_printf(s, "mrst_pmu_wait_ready_udelays_max %8d\n", + pmu_wait_ready_udelays_max); + seq_printf(s, "mrst_pmu_wait_done_calls %8d\n", + pmu_wait_done_calls); + seq_printf(s, "mrst_pmu_wait_done_udelays %8d\n", + pmu_wait_done_udelays); + seq_printf(s, "mrst_pmu_wait_done_udelays_max %8d\n", + pmu_wait_done_udelays_max); + seq_printf(s, "mrst_pmu_set_power_state_entry %8d\n", + pmu_set_power_state_entry); + seq_printf(s, "mrst_pmu_set_power_state_send_cmd %8d\n", + pmu_set_power_state_send_cmd); + seq_printf(s, "SCU busy: %d\n", pmu_read_busy_status()); + + return 0; +} + +static int debug_mrst_pmu_open(struct inode *inode, struct file *file) +{ + return single_open(file, debug_mrst_pmu_show, NULL); +} + +static const struct file_operations devices_state_operations = { + .open = debug_mrst_pmu_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; +#endif /* DEBUG_FS */ + +/* + * Validate SCU PCI shim PCI vendor capability byte + * against LSS hard-coded in mrst_devs[] above. + * DEBUG only. + */ +static void pmu_scu_firmware_debug(void) +{ + struct pci_dev *pdev = NULL; + + for_each_pci_dev(pdev) { + struct mrst_device *mrst_dev; + u8 pci_config_lss; + int pos; + + mrst_dev = pci_id_2_mrst_dev(pdev->device); + if (unlikely(!mrst_dev)) { + printk(KERN_ERR FW_BUG "pmu: Unknown " + "PCI device 0x%04X\n", pdev->device); + continue; + } + + if (mrst_dev->lss == 0) + continue; /* no LSS in our table */ + + pos = pci_find_capability(pdev, PCI_CAP_ID_VNDR); + if (!pos != 0) { + printk(KERN_ERR FW_BUG "pmu: 0x%04X " + "missing PCI Vendor Capability\n", + pdev->device); + continue; + } + pci_read_config_byte(pdev, pos + 4, &pci_config_lss); + if (!(pci_config_lss & PCI_VENDOR_CAP_LOG_SS_MASK)) { + printk(KERN_ERR FW_BUG "pmu: 0x%04X " + "invalid PCI Vendor Capability 0x%x " + " expected LSS 0x%X\n", + pdev->device, pci_config_lss, mrst_dev->lss); + continue; + } + pci_config_lss &= PCI_VENDOR_CAP_LOG_ID_MASK; + + if (mrst_dev->lss == pci_config_lss) + continue; + + printk(KERN_ERR FW_BUG "pmu: 0x%04X LSS = %d, expected %d\n", + pdev->device, pci_config_lss, mrst_dev->lss); + } +} + +/** + * pmu_probe + */ +static int __devinit pmu_probe(struct pci_dev *pdev, + const struct pci_device_id *pci_id) +{ + int ret; + struct mrst_pmu_reg *pmu; + + /* Init the device */ + ret = pci_enable_device(pdev); + if (ret) { + dev_err(&pdev->dev, "Unable to Enable PCI device\n"); + return ret; + } + + ret = pci_request_regions(pdev, MRST_PMU_DRV_NAME); + if (ret < 0) { + dev_err(&pdev->dev, "Cannot obtain PCI resources, aborting\n"); + goto out_err1; + } + + /* Map the memory of PMU reg base */ + pmu = pci_iomap(pdev, 0, 0); + if (!pmu) { + dev_err(&pdev->dev, "Unable to map the PMU address space\n"); + ret = -ENOMEM; + goto out_err2; + } + +#ifdef CONFIG_DEBUG_FS + /* /sys/kernel/debug/mrst_pmu */ + (void) debugfs_create_file("mrst_pmu", S_IFREG | S_IRUGO, + NULL, NULL, &devices_state_operations); +#endif + pmu_reg = pmu; /* success */ + + if (request_irq(pdev->irq, pmu_irq, 0, MRST_PMU_DRV_NAME, NULL)) { + dev_err(&pdev->dev, "Registering isr has failed\n"); + ret = -1; + goto out_err3; + } + + pmu_scu_firmware_debug(); + + pmu_write_wkc(S0I3_WAKE_SOURCES); /* Enable S0i3 wakeup sources */ + + pmu_wait_ready(); + + pmu_write_ssc(D0I1_ACG_SSS_TARGET); /* Enable Auto-Clock_Gating */ + pmu_write_cmd(0x201); + + spin_lock_init(&mrst_pmu_power_state_lock); + + /* Enable the hardware interrupt */ + pmu_irq_enable(); + return 0; + +out_err3: + free_irq(pdev->irq, NULL); + pci_iounmap(pdev, pmu_reg); + pmu_reg = NULL; +out_err2: + pci_release_region(pdev, 0); +out_err1: + pci_disable_device(pdev); + return ret; +} + +static void __devexit pmu_remove(struct pci_dev *pdev) +{ + dev_err(&pdev->dev, "Mid PM pmu_remove called\n"); + + /* Freeing up the irq */ + free_irq(pdev->irq, NULL); + + pci_iounmap(pdev, pmu_reg); + pmu_reg = NULL; + + /* disable the current PCI device */ + pci_release_region(pdev, 0); + pci_disable_device(pdev); +} + +static DEFINE_PCI_DEVICE_TABLE(pmu_pci_ids) = { + { PCI_VDEVICE(INTEL, PCI_DEV_ID_MRST_PMU), 0 }, + { } +}; + +MODULE_DEVICE_TABLE(pci, pmu_pci_ids); + +static struct pci_driver driver = { + .name = MRST_PMU_DRV_NAME, + .id_table = pmu_pci_ids, + .probe = pmu_probe, + .remove = __devexit_p(pmu_remove), +}; + +/** + * pmu_pci_register - register the PMU driver as PCI device + */ +static int __init pmu_pci_register(void) +{ + return pci_register_driver(&driver); +} + +/* Register and probe via fs_initcall() to preceed device_initcall() */ +fs_initcall(pmu_pci_register); + +static void __exit mid_pci_cleanup(void) +{ + pci_unregister_driver(&driver); +} + +static int ia_major; +static int ia_minor; + +static int pmu_sfi_parse_oem(struct sfi_table_header *table) +{ + struct sfi_table_simple *sb; + + sb = (struct sfi_table_simple *)table; + ia_major = (sb->pentry[1] >> 0) & 0xFFFF; + ia_minor = (sb->pentry[1] >> 16) & 0xFFFF; + printk(KERN_INFO "mrst_pmu: IA FW version v%x.%x\n", + ia_major, ia_minor); + + return 0; +} + +static int __init scu_fw_check(void) +{ + int ret; + u32 fw_version; + + if (!pmu_reg) + return 0; /* this driver didn't probe-out */ + + sfi_table_parse("OEMB", NULL, NULL, pmu_sfi_parse_oem); + + if (ia_major < 0x6005 || ia_minor < 0x1525) { + WARN(1, "mrst_pmu: IA FW version too old\n"); + return -1; + } + + ret = intel_scu_ipc_command(IPCMSG_FW_REVISION, 0, NULL, 0, + &fw_version, 1); + + if (ret) { + WARN(1, "mrst_pmu: IPC FW version? %d\n", ret); + } else { + int scu_major = (fw_version >> 8) & 0xFF; + int scu_minor = (fw_version >> 0) & 0xFF; + + printk(KERN_INFO "mrst_pmu: firmware v%x\n", fw_version); + + if ((scu_major >= 0xC0) && (scu_minor >= 0x49)) { + printk(KERN_INFO "mrst_pmu: enabling S0i3\n"); + mrst_pmu_s0i3_enable = true; + } else { + WARN(1, "mrst_pmu: S0i3 disabled, old firmware %X.%X", + scu_major, scu_minor); + } + } + return 0; +} +late_initcall(scu_fw_check); +module_exit(mid_pci_cleanup); diff --git a/arch/x86/platform/mrst/pmu.h b/arch/x86/platform/mrst/pmu.h new file mode 100644 index 000000000000..bfbfe64b167b --- /dev/null +++ b/arch/x86/platform/mrst/pmu.h @@ -0,0 +1,234 @@ +/* + * mrst/pmu.h - private definitions for MRST Power Management Unit mrst/pmu.c + * + * Copyright (c) 2011, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope 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., + * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. + */ + +#ifndef _MRST_PMU_H_ +#define _MRST_PMU_H_ + +#define PCI_DEV_ID_MRST_PMU 0x0810 +#define MRST_PMU_DRV_NAME "mrst_pmu" +#define PCI_SUB_CLASS_MASK 0xFF00 + +#define PCI_VENDOR_CAP_LOG_ID_MASK 0x7F +#define PCI_VENDOR_CAP_LOG_SS_MASK 0x80 + +#define SUB_SYS_ALL_D0I1 0x01155555 +#define S0I3_WAKE_SOURCES 0x00001FFF + +#define PM_S0I3_COMMAND \ + ((0 << 31) | /* Reserved */ \ + (0 << 30) | /* Core must be idle */ \ + (0xc2 << 22) | /* ACK C6 trigger */ \ + (3 << 19) | /* Trigger on DMI message */ \ + (3 << 16) | /* Enter S0i3 */ \ + (0 << 13) | /* Numeric mode ID (sw) */ \ + (3 << 9) | /* Trigger mode */ \ + (0 << 8) | /* Do not interrupt */ \ + (1 << 0)) /* Set configuration */ + +#define LSS_DMI 0 +#define LSS_SD_HC0 1 +#define LSS_SD_HC1 2 +#define LSS_NAND 3 +#define LSS_IMAGING 4 +#define LSS_SECURITY 5 +#define LSS_DISPLAY 6 +#define LSS_USB_HC 7 +#define LSS_USB_OTG 8 +#define LSS_AUDIO 9 +#define LSS_AUDIO_LPE 9 +#define LSS_AUDIO_SSP 9 +#define LSS_I2C0 10 +#define LSS_I2C1 10 +#define LSS_I2C2 10 +#define LSS_KBD 10 +#define LSS_SPI0 10 +#define LSS_SPI1 10 +#define LSS_SPI2 10 +#define LSS_GPIO 10 +#define LSS_SRAM 11 /* used by SCU, do not touch */ +#define LSS_SD_HC2 12 +/* LSS hardware bits 15,14,13 are hardwired to 0, thus unusable */ +#define MRST_NUM_LSS 13 + +#define MIN(a, b) (((a) < (b)) ? (a) : (b)) + +#define SSMSK(mask, lss) ((mask) << ((lss) * 2)) +#define D0 0 +#define D0i1 1 +#define D0i2 2 +#define D0i3 3 + +#define S0I3_SSS_TARGET ( \ + SSMSK(D0i1, LSS_DMI) | \ + SSMSK(D0i3, LSS_SD_HC0) | \ + SSMSK(D0i3, LSS_SD_HC1) | \ + SSMSK(D0i3, LSS_NAND) | \ + SSMSK(D0i3, LSS_SD_HC2) | \ + SSMSK(D0i3, LSS_IMAGING) | \ + SSMSK(D0i3, LSS_SECURITY) | \ + SSMSK(D0i3, LSS_DISPLAY) | \ + SSMSK(D0i3, LSS_USB_HC) | \ + SSMSK(D0i3, LSS_USB_OTG) | \ + SSMSK(D0i3, LSS_AUDIO) | \ + SSMSK(D0i1, LSS_I2C0)) + +/* + * D0i1 on Langwell is Autonomous Clock Gating (ACG). + * Enable ACG on every LSS except camera and audio + */ +#define D0I1_ACG_SSS_TARGET \ + (SUB_SYS_ALL_D0I1 & ~SSMSK(D0i1, LSS_IMAGING) & ~SSMSK(D0i1, LSS_AUDIO)) + +enum cm_mode { + CM_NOP, /* ignore the config mode value */ + CM_IMMEDIATE, + CM_DELAY, + CM_TRIGGER, + CM_INVALID +}; + +enum sys_state { + SYS_STATE_S0I0, + SYS_STATE_S0I1, + SYS_STATE_S0I2, + SYS_STATE_S0I3, + SYS_STATE_S3, + SYS_STATE_S5 +}; + +#define SET_CFG_CMD 1 + +enum int_status { + INT_SPURIOUS = 0, + INT_CMD_DONE = 1, + INT_CMD_ERR = 2, + INT_WAKE_RX = 3, + INT_SS_ERROR = 4, + INT_S0IX_MISS = 5, + INT_NO_ACKC6 = 6, + INT_INVALID = 7, +}; + +/* PMU register interface */ +static struct mrst_pmu_reg { + u32 pm_sts; /* 0x00 */ + u32 pm_cmd; /* 0x04 */ + u32 pm_ics; /* 0x08 */ + u32 _resv1; /* 0x0C */ + u32 pm_wkc[2]; /* 0x10 */ + u32 pm_wks[2]; /* 0x18 */ + u32 pm_ssc[4]; /* 0x20 */ + u32 pm_sss[4]; /* 0x30 */ + u32 pm_wssc[4]; /* 0x40 */ + u32 pm_c3c4; /* 0x50 */ + u32 pm_c5c6; /* 0x54 */ + u32 pm_msi_disable; /* 0x58 */ +} *pmu_reg; + +static inline u32 pmu_read_sts(void) { return readl(&pmu_reg->pm_sts); } +static inline u32 pmu_read_ics(void) { return readl(&pmu_reg->pm_ics); } +static inline u32 pmu_read_wks(void) { return readl(&pmu_reg->pm_wks[0]); } +static inline u32 pmu_read_sss(void) { return readl(&pmu_reg->pm_sss[0]); } + +static inline void pmu_write_cmd(u32 arg) { writel(arg, &pmu_reg->pm_cmd); } +static inline void pmu_write_ics(u32 arg) { writel(arg, &pmu_reg->pm_ics); } +static inline void pmu_write_wkc(u32 arg) { writel(arg, &pmu_reg->pm_wkc[0]); } +static inline void pmu_write_ssc(u32 arg) { writel(arg, &pmu_reg->pm_ssc[0]); } +static inline void pmu_write_wssc(u32 arg) + { writel(arg, &pmu_reg->pm_wssc[0]); } + +static inline void pmu_msi_enable(void) { writel(0, &pmu_reg->pm_msi_disable); } +static inline u32 pmu_msi_is_disabled(void) + { return readl(&pmu_reg->pm_msi_disable); } + +union pmu_pm_ics { + struct { + u32 cause:8; + u32 enable:1; + u32 pending:1; + u32 reserved:22; + } bits; + u32 value; +}; + +static inline void pmu_irq_enable(void) +{ + union pmu_pm_ics pmu_ics; + + pmu_ics.value = pmu_read_ics(); + pmu_ics.bits.enable = 1; + pmu_write_ics(pmu_ics.value); +} + +union pmu_pm_status { + struct { + u32 pmu_rev:8; + u32 pmu_busy:1; + u32 mode_id:4; + u32 Reserved:19; + } pmu_status_parts; + u32 pmu_status_value; +}; + +static inline int pmu_read_busy_status(void) +{ + union pmu_pm_status result; + + result.pmu_status_value = pmu_read_sts(); + + return result.pmu_status_parts.pmu_busy; +} + +/* pmu set config parameters */ +struct cfg_delay_param_t { + u32 cmd:8; + u32 ioc:1; + u32 cfg_mode:4; + u32 mode_id:3; + u32 sys_state:3; + u32 cfg_delay:8; + u32 rsvd:5; +}; + +struct cfg_trig_param_t { + u32 cmd:8; + u32 ioc:1; + u32 cfg_mode:4; + u32 mode_id:3; + u32 sys_state:3; + u32 cfg_trig_type:3; + u32 cfg_trig_val:8; + u32 cmbi:1; + u32 rsvd1:1; +}; + +union pmu_pm_set_cfg_cmd_t { + union { + struct cfg_delay_param_t d_param; + struct cfg_trig_param_t t_param; + } pmu2_params; + u32 pmu_pm_set_cfg_cmd_value; +}; + +#ifdef FUTURE_PATCH +extern int mrst_s0i3_entry(u32 regval, u32 *regaddr); +#else +static inline int mrst_s0i3_entry(u32 regval, u32 *regaddr) { return -1; } +#endif +#endif From 62027aea23fcd14478abdddd3b74a4e0f5fb2984 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Fri, 1 Apr 2011 18:13:10 -0400 Subject: [PATCH 275/439] cpuidle: create bootparam "cpuidle.off=1" useful for disabling cpuidle to fall back to architecture-default idle loop cpuidle drivers and governors will fail to register. on x86 they'll say so: intel_idle: intel_idle yielding to (null) ACPI: acpi_idle yielding to (null) Signed-off-by: Len Brown --- Documentation/kernel-parameters.txt | 3 +++ drivers/cpuidle/cpuidle.c | 10 ++++++++++ drivers/cpuidle/cpuidle.h | 1 + drivers/cpuidle/driver.c | 3 +++ drivers/cpuidle/governor.c | 3 +++ 5 files changed, 20 insertions(+) diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index aa47be71df4c..9b8e62d75a0c 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -546,6 +546,9 @@ bytes respectively. Such letter suffixes can also be entirely omitted. /proc//coredump_filter. See also Documentation/filesystems/proc.txt. + cpuidle.off=1 [CPU_IDLE] + disable the cpuidle sub-system + cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver Format: ,,,[,] diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c index bf5092455a8f..faae2c357bab 100644 --- a/drivers/cpuidle/cpuidle.c +++ b/drivers/cpuidle/cpuidle.c @@ -28,6 +28,12 @@ LIST_HEAD(cpuidle_detected_devices); static void (*pm_idle_old)(void); static int enabled_devices; +static int off __read_mostly; + +int cpuidle_disabled(void) +{ + return off; +} #if defined(CONFIG_ARCH_HAS_CPU_IDLE_WAIT) static void cpuidle_kick_cpus(void) @@ -427,6 +433,9 @@ static int __init cpuidle_init(void) { int ret; + if (cpuidle_disabled()) + return -ENODEV; + pm_idle_old = pm_idle; ret = cpuidle_add_class_sysfs(&cpu_sysdev_class); @@ -438,4 +447,5 @@ static int __init cpuidle_init(void) return 0; } +module_param(off, int, 0444); core_initcall(cpuidle_init); diff --git a/drivers/cpuidle/cpuidle.h b/drivers/cpuidle/cpuidle.h index 33e50d556f17..38c3fd8b9d76 100644 --- a/drivers/cpuidle/cpuidle.h +++ b/drivers/cpuidle/cpuidle.h @@ -13,6 +13,7 @@ extern struct list_head cpuidle_governors; extern struct list_head cpuidle_detected_devices; extern struct mutex cpuidle_lock; extern spinlock_t cpuidle_driver_lock; +extern int cpuidle_disabled(void); /* idle loop */ extern void cpuidle_install_idle_handler(void); diff --git a/drivers/cpuidle/driver.c b/drivers/cpuidle/driver.c index fd1601e3d125..3f7e3cedd133 100644 --- a/drivers/cpuidle/driver.c +++ b/drivers/cpuidle/driver.c @@ -26,6 +26,9 @@ int cpuidle_register_driver(struct cpuidle_driver *drv) if (!drv) return -EINVAL; + if (cpuidle_disabled()) + return -ENODEV; + spin_lock(&cpuidle_driver_lock); if (cpuidle_curr_driver) { spin_unlock(&cpuidle_driver_lock); diff --git a/drivers/cpuidle/governor.c b/drivers/cpuidle/governor.c index 724c164d31c9..ea2f8e7aa24a 100644 --- a/drivers/cpuidle/governor.c +++ b/drivers/cpuidle/governor.c @@ -81,6 +81,9 @@ int cpuidle_register_governor(struct cpuidle_governor *gov) if (!gov || !gov->select) return -EINVAL; + if (cpuidle_disabled()) + return -ENODEV; + mutex_lock(&cpuidle_lock); if (__cpuidle_find_governor(gov->name) == NULL) { ret = 0; From d91ee5863b71e8c90eaf6035bff3078a85e2e7b5 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Fri, 1 Apr 2011 18:28:35 -0400 Subject: [PATCH 276/439] cpuidle: replace xen access to x86 pm_idle and default_idle When a Xen Dom0 kernel boots on a hypervisor, it gets access to the raw-hardware ACPI tables. While it parses the idle tables for the hypervisor's beneift, it uses HLT for its own idle. Rather than have xen scribble on pm_idle and access default_idle, have it simply disable_cpuidle() so acpi_idle will not load and architecture default HLT will be used. cc: xen-devel@lists.xensource.com Tested-by: Konrad Rzeszutek Wilk Acked-by: H. Peter Anvin Signed-off-by: Len Brown --- arch/x86/xen/setup.c | 3 ++- drivers/cpuidle/cpuidle.c | 4 ++++ include/linux/cpuidle.h | 2 ++ 3 files changed, 8 insertions(+), 1 deletion(-) diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c index 60aeeb56948f..a9627e2e3295 100644 --- a/arch/x86/xen/setup.c +++ b/arch/x86/xen/setup.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include @@ -426,7 +427,7 @@ void __init xen_arch_setup(void) #ifdef CONFIG_X86_32 boot_cpu_data.hlt_works_ok = 1; #endif - pm_idle = default_idle; + disable_cpuidle(); boot_option_idle_override = IDLE_HALT; fiddle_vdso(); diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c index faae2c357bab..041df0b056b2 100644 --- a/drivers/cpuidle/cpuidle.c +++ b/drivers/cpuidle/cpuidle.c @@ -34,6 +34,10 @@ int cpuidle_disabled(void) { return off; } +void disable_cpuidle(void) +{ + off = 1; +} #if defined(CONFIG_ARCH_HAS_CPU_IDLE_WAIT) static void cpuidle_kick_cpus(void) diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h index 36719ead50e8..b89f67da919e 100644 --- a/include/linux/cpuidle.h +++ b/include/linux/cpuidle.h @@ -122,6 +122,7 @@ struct cpuidle_driver { }; #ifdef CONFIG_CPU_IDLE +extern void disable_cpuidle(void); extern int cpuidle_register_driver(struct cpuidle_driver *drv); struct cpuidle_driver *cpuidle_get_driver(void); @@ -135,6 +136,7 @@ extern int cpuidle_enable_device(struct cpuidle_device *dev); extern void cpuidle_disable_device(struct cpuidle_device *dev); #else +static inline void disable_cpuidle(void) { } static inline int cpuidle_register_driver(struct cpuidle_driver *drv) {return -ENODEV; } From 4bfc8288bc4a64529c5547d17349a2a1f4675507 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Wed, 30 Mar 2011 23:52:29 -0400 Subject: [PATCH 277/439] x86 idle: move mwait_idle_with_hints() to where it is used ...and make it static no functional change cc: x86@kernel.org Acked-by: H. Peter Anvin Signed-off-by: Len Brown --- arch/x86/include/asm/processor.h | 2 -- arch/x86/kernel/acpi/cstate.c | 23 +++++++++++++++++++++++ arch/x86/kernel/process.c | 23 ----------------------- 3 files changed, 23 insertions(+), 25 deletions(-) diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h index 219371546afd..0d1171c97729 100644 --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -751,8 +751,6 @@ static inline void __sti_mwait(unsigned long eax, unsigned long ecx) :: "a" (eax), "c" (ecx)); } -extern void mwait_idle_with_hints(unsigned long eax, unsigned long ecx); - extern void select_idle_routine(const struct cpuinfo_x86 *c); extern void init_amd_e400_c1e_mask(void); diff --git a/arch/x86/kernel/acpi/cstate.c b/arch/x86/kernel/acpi/cstate.c index 5812404a0d4c..f50e7fb2a201 100644 --- a/arch/x86/kernel/acpi/cstate.c +++ b/arch/x86/kernel/acpi/cstate.c @@ -149,6 +149,29 @@ int acpi_processor_ffh_cstate_probe(unsigned int cpu, } EXPORT_SYMBOL_GPL(acpi_processor_ffh_cstate_probe); +/* + * This uses new MONITOR/MWAIT instructions on P4 processors with PNI, + * which can obviate IPI to trigger checking of need_resched. + * We execute MONITOR against need_resched and enter optimized wait state + * through MWAIT. Whenever someone changes need_resched, we would be woken + * up from MWAIT (without an IPI). + * + * New with Core Duo processors, MWAIT can take some hints based on CPU + * capability. + */ +void mwait_idle_with_hints(unsigned long ax, unsigned long cx) +{ + if (!need_resched()) { + if (this_cpu_has(X86_FEATURE_CLFLUSH_MONITOR)) + clflush((void *)¤t_thread_info()->flags); + + __monitor((void *)¤t_thread_info()->flags, 0, 0); + smp_mb(); + if (!need_resched()) + __mwait(ax, cx); + } +} + void acpi_processor_ffh_cstate_enter(struct acpi_processor_cx *cx) { unsigned int cpu = smp_processor_id(); diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index e1ba8cb24e4e..e7e3b019c439 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -438,29 +438,6 @@ void cpu_idle_wait(void) } EXPORT_SYMBOL_GPL(cpu_idle_wait); -/* - * This uses new MONITOR/MWAIT instructions on P4 processors with PNI, - * which can obviate IPI to trigger checking of need_resched. - * We execute MONITOR against need_resched and enter optimized wait state - * through MWAIT. Whenever someone changes need_resched, we would be woken - * up from MWAIT (without an IPI). - * - * New with Core Duo processors, MWAIT can take some hints based on CPU - * capability. - */ -void mwait_idle_with_hints(unsigned long ax, unsigned long cx) -{ - if (!need_resched()) { - if (this_cpu_has(X86_FEATURE_CLFLUSH_MONITOR)) - clflush((void *)¤t_thread_info()->flags); - - __monitor((void *)¤t_thread_info()->flags, 0, 0); - smp_mb(); - if (!need_resched()) - __mwait(ax, cx); - } -} - /* Default MONITOR/MWAIT with no hints, used for default C1 state */ static void mwait_idle(void) { From a0bfa1373859e9d11dc92561a8667588803e42d8 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Fri, 1 Apr 2011 19:34:59 -0400 Subject: [PATCH 278/439] cpuidle: stop depending on pm_idle cpuidle users should call cpuidle_call_idle() directly rather than via (pm_idle)() function pointer. Architecture may choose to continue using (pm_idle)(), but cpuidle need not depend on it: my_arch_cpu_idle() ... if(cpuidle_call_idle()) pm_idle(); cc: Kevin Hilman cc: Paul Mundt cc: x86@kernel.org Acked-by: H. Peter Anvin Signed-off-by: Len Brown --- arch/arm/kernel/process.c | 4 +++- arch/sh/kernel/idle.c | 6 ++++-- arch/x86/kernel/process_32.c | 4 +++- arch/x86/kernel/process_64.c | 4 +++- drivers/cpuidle/cpuidle.c | 38 +++++++++++++++++------------------- include/linux/cpuidle.h | 2 ++ 6 files changed, 33 insertions(+), 25 deletions(-) diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index 5e1e54197227..d7ee0d4c072d 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -30,6 +30,7 @@ #include #include #include +#include #include #include @@ -196,7 +197,8 @@ void cpu_idle(void) cpu_relax(); } else { stop_critical_timings(); - pm_idle(); + if (cpuidle_call_idle()) + pm_idle(); start_critical_timings(); /* * This will eventually be removed - pm_idle diff --git a/arch/sh/kernel/idle.c b/arch/sh/kernel/idle.c index 425d604e3a28..9c7099ebfe14 100644 --- a/arch/sh/kernel/idle.c +++ b/arch/sh/kernel/idle.c @@ -16,12 +16,13 @@ #include #include #include +#include #include #include #include #include -void (*pm_idle)(void) = NULL; +static void (*pm_idle)(void); static int hlt_counter; @@ -100,7 +101,8 @@ void cpu_idle(void) local_irq_disable(); /* Don't trace irqs off for idle */ stop_critical_timings(); - pm_idle(); + if (cpuidle_call_idle()) + pm_idle(); /* * Sanity check to ensure that pm_idle() returns * with IRQs enabled diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index a3d0dc59067b..7a3b65107a27 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -38,6 +38,7 @@ #include #include #include +#include #include #include @@ -109,7 +110,8 @@ void cpu_idle(void) local_irq_disable(); /* Don't trace irqs off for idle */ stop_critical_timings(); - pm_idle(); + if (cpuidle_idle_call()) + pm_idle(); start_critical_timings(); } tick_nohz_restart_sched_tick(); diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index ca6f7ab8df33..f693e44e1bf6 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -37,6 +37,7 @@ #include #include #include +#include #include #include @@ -136,7 +137,8 @@ void cpu_idle(void) enter_idle(); /* Don't trace irqs off for idle */ stop_critical_timings(); - pm_idle(); + if (cpuidle_idle_call()) + pm_idle(); start_critical_timings(); /* In many cases the interrupt that ended idle diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c index 041df0b056b2..d4c542372886 100644 --- a/drivers/cpuidle/cpuidle.c +++ b/drivers/cpuidle/cpuidle.c @@ -25,10 +25,10 @@ DEFINE_PER_CPU(struct cpuidle_device *, cpuidle_devices); DEFINE_MUTEX(cpuidle_lock); LIST_HEAD(cpuidle_detected_devices); -static void (*pm_idle_old)(void); static int enabled_devices; static int off __read_mostly; +static int initialized __read_mostly; int cpuidle_disabled(void) { @@ -56,25 +56,23 @@ static int __cpuidle_register_device(struct cpuidle_device *dev); * cpuidle_idle_call - the main idle loop * * NOTE: no locks or semaphores should be used here + * return non-zero on failure */ -static void cpuidle_idle_call(void) +int cpuidle_idle_call(void) { struct cpuidle_device *dev = __this_cpu_read(cpuidle_devices); struct cpuidle_state *target_state; int next_state; + if (off) + return -ENODEV; + + if (!initialized) + return -ENODEV; + /* check if the device is ready */ - if (!dev || !dev->enabled) { - if (pm_idle_old) - pm_idle_old(); - else -#if defined(CONFIG_ARCH_HAS_DEFAULT_IDLE) - default_idle(); -#else - local_irq_enable(); -#endif - return; - } + if (!dev || !dev->enabled) + return -EBUSY; #if 0 /* shows regressions, re-enable for 2.6.29 */ @@ -99,7 +97,7 @@ static void cpuidle_idle_call(void) next_state = cpuidle_curr_governor->select(dev); if (need_resched()) { local_irq_enable(); - return; + return 0; } target_state = &dev->states[next_state]; @@ -124,6 +122,8 @@ static void cpuidle_idle_call(void) /* give the governor an opportunity to reflect on the outcome */ if (cpuidle_curr_governor->reflect) cpuidle_curr_governor->reflect(dev); + + return 0; } /** @@ -131,10 +131,10 @@ static void cpuidle_idle_call(void) */ void cpuidle_install_idle_handler(void) { - if (enabled_devices && (pm_idle != cpuidle_idle_call)) { + if (enabled_devices) { /* Make sure all changes finished before we switch to new idle */ smp_wmb(); - pm_idle = cpuidle_idle_call; + initialized = 1; } } @@ -143,8 +143,8 @@ void cpuidle_install_idle_handler(void) */ void cpuidle_uninstall_idle_handler(void) { - if (enabled_devices && pm_idle_old && (pm_idle != pm_idle_old)) { - pm_idle = pm_idle_old; + if (enabled_devices) { + initialized = 0; cpuidle_kick_cpus(); } } @@ -440,8 +440,6 @@ static int __init cpuidle_init(void) if (cpuidle_disabled()) return -ENODEV; - pm_idle_old = pm_idle; - ret = cpuidle_add_class_sysfs(&cpu_sysdev_class); if (ret) return ret; diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h index b89f67da919e..b51629e15cfc 100644 --- a/include/linux/cpuidle.h +++ b/include/linux/cpuidle.h @@ -123,6 +123,7 @@ struct cpuidle_driver { #ifdef CONFIG_CPU_IDLE extern void disable_cpuidle(void); +extern int cpuidle_idle_call(void); extern int cpuidle_register_driver(struct cpuidle_driver *drv); struct cpuidle_driver *cpuidle_get_driver(void); @@ -137,6 +138,7 @@ extern void cpuidle_disable_device(struct cpuidle_device *dev); #else static inline void disable_cpuidle(void) { } +static inline int cpuidle_idle_call(void) { return -ENODEV; } static inline int cpuidle_register_driver(struct cpuidle_driver *drv) {return -ENODEV; } From 06fa0a883a01a34a0449ec116c5288c1d196b4b0 Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin Date: Wed, 3 Aug 2011 16:38:59 -0700 Subject: [PATCH 279/439] mlx4: Fixing Ethernet unicast packet steering For older FW versions, fixing the usage of per port Mac table. For each port we must define the base QP number, which is passed to the HW. Setting the correct value in SET_PORT FW command to enable the steering. Reported-by: Roland Dreier Tested-by: Roland Dreier Signed-off-by: Yevgeny Petrilin Signed-off-by: David S. Miller --- drivers/net/mlx4/en_port.c | 2 +- drivers/net/mlx4/main.c | 2 ++ 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/mlx4/en_port.c b/drivers/net/mlx4/en_port.c index 5e7109178061..5ada5b469112 100644 --- a/drivers/net/mlx4/en_port.c +++ b/drivers/net/mlx4/en_port.c @@ -128,7 +128,7 @@ int mlx4_SET_PORT_qpn_calc(struct mlx4_dev *dev, u8 port, u32 base_qpn, memset(context, 0, sizeof *context); context->base_qpn = cpu_to_be32(base_qpn); - context->n_mac = 0x7; + context->n_mac = 0x2; context->promisc = cpu_to_be32(promisc << SET_PORT_PROMISC_SHIFT | base_qpn); context->mcast = cpu_to_be32(m_promisc << SET_PORT_MC_PROMISC_SHIFT | diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c index c94b3426d355..f0ee35df4dd7 100644 --- a/drivers/net/mlx4/main.c +++ b/drivers/net/mlx4/main.c @@ -1117,6 +1117,8 @@ static int mlx4_init_port_info(struct mlx4_dev *dev, int port) info->port = port; mlx4_init_mac_table(dev, &info->mac_table); mlx4_init_vlan_table(dev, &info->vlan_table); + info->base_qpn = dev->caps.reserved_qps_base[MLX4_QP_REGION_ETH_ADDR] + + (port - 1) * (1 << log_num_mac); sprintf(info->dev_name, "mlx4_port%d", port); info->port_attr.attr.name = info->dev_name; From dd48c085c1cdf9446f92826f1fd451167fb6c2fd Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Wed, 3 Aug 2011 16:21:01 -0700 Subject: [PATCH 280/439] fault-injection: add ability to export fault_attr in arbitrary directory init_fault_attr_dentries() is used to export fault_attr via debugfs. But it can only export it in debugfs root directory. Per Forlin is working on mmc_fail_request which adds support to inject data errors after a completed host transfer in MMC subsystem. The fault_attr for mmc_fail_request should be defined per mmc host and export it in debugfs directory per mmc host like /sys/kernel/debug/mmc0/mmc_fail_request. init_fault_attr_dentries() doesn't help for mmc_fail_request. So this introduces fault_create_debugfs_attr() which is able to create a directory in the arbitrary directory and replace init_fault_attr_dentries(). [akpm@linux-foundation.org: extraneous semicolon, per Randy] Signed-off-by: Akinobu Mita Tested-by: Per Forlin Cc: Jens Axboe Cc: Christoph Lameter Cc: Pekka Enberg Cc: Matt Mackall Cc: Randy Dunlap Cc: Stephen Rothwell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- .../fault-injection/fault-injection.txt | 3 +-- block/blk-core.c | 6 ++++-- block/blk-timeout.c | 5 ++++- include/linux/fault-inject.h | 18 +++++------------ lib/fault-inject.c | 20 +++++++------------ mm/failslab.c | 14 ++++++------- mm/page_alloc.c | 13 +++++------- 7 files changed, 33 insertions(+), 46 deletions(-) diff --git a/Documentation/fault-injection/fault-injection.txt b/Documentation/fault-injection/fault-injection.txt index 7be15e44d481..82a5d250d75e 100644 --- a/Documentation/fault-injection/fault-injection.txt +++ b/Documentation/fault-injection/fault-injection.txt @@ -143,8 +143,7 @@ o provide a way to configure fault attributes failslab, fail_page_alloc, and fail_make_request use this way. Helper functions: - init_fault_attr_dentries(entries, attr, name); - void cleanup_fault_attr_dentries(entries); + fault_create_debugfs_attr(name, parent, attr); - module parameters diff --git a/block/blk-core.c b/block/blk-core.c index b850bedad229..b627558c461f 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -1368,8 +1368,10 @@ static bool should_fail_request(struct hd_struct *part, unsigned int bytes) static int __init fail_make_request_debugfs(void) { - return init_fault_attr_dentries(&fail_make_request, - "fail_make_request"); + struct dentry *dir = fault_create_debugfs_attr("fail_make_request", + NULL, &fail_make_request); + + return IS_ERR(dir) ? PTR_ERR(dir) : 0; } late_initcall(fail_make_request_debugfs); diff --git a/block/blk-timeout.c b/block/blk-timeout.c index 4f0c06c7a338..780354888958 100644 --- a/block/blk-timeout.c +++ b/block/blk-timeout.c @@ -28,7 +28,10 @@ int blk_should_fake_timeout(struct request_queue *q) static int __init fail_io_timeout_debugfs(void) { - return init_fault_attr_dentries(&fail_io_timeout, "fail_io_timeout"); + struct dentry *dir = fault_create_debugfs_attr("fail_io_timeout", + NULL, &fail_io_timeout); + + return IS_ERR(dir) ? PTR_ERR(dir) : 0; } late_initcall(fail_io_timeout_debugfs); diff --git a/include/linux/fault-inject.h b/include/linux/fault-inject.h index 3ff060ac7810..c6f996f2abb6 100644 --- a/include/linux/fault-inject.h +++ b/include/linux/fault-inject.h @@ -25,10 +25,6 @@ struct fault_attr { unsigned long reject_end; unsigned long count; - -#ifdef CONFIG_FAULT_INJECTION_DEBUG_FS - struct dentry *dir; -#endif }; #define FAULT_ATTR_INITIALIZER { \ @@ -45,19 +41,15 @@ bool should_fail(struct fault_attr *attr, ssize_t size); #ifdef CONFIG_FAULT_INJECTION_DEBUG_FS -int init_fault_attr_dentries(struct fault_attr *attr, const char *name); -void cleanup_fault_attr_dentries(struct fault_attr *attr); +struct dentry *fault_create_debugfs_attr(const char *name, + struct dentry *parent, struct fault_attr *attr); #else /* CONFIG_FAULT_INJECTION_DEBUG_FS */ -static inline int init_fault_attr_dentries(struct fault_attr *attr, - const char *name) -{ - return -ENODEV; -} - -static inline void cleanup_fault_attr_dentries(struct fault_attr *attr) +static inline struct dentry *fault_create_debugfs_attr(const char *name, + struct dentry *parent, struct fault_attr *attr) { + return ERR_PTR(-ENODEV); } #endif /* CONFIG_FAULT_INJECTION_DEBUG_FS */ diff --git a/lib/fault-inject.c b/lib/fault-inject.c index 2577b121c7c1..f193b7796449 100644 --- a/lib/fault-inject.c +++ b/lib/fault-inject.c @@ -197,21 +197,15 @@ static struct dentry *debugfs_create_atomic_t(const char *name, mode_t mode, return debugfs_create_file(name, mode, parent, value, &fops_atomic_t); } -void cleanup_fault_attr_dentries(struct fault_attr *attr) -{ - debugfs_remove_recursive(attr->dir); -} - -int init_fault_attr_dentries(struct fault_attr *attr, const char *name) +struct dentry *fault_create_debugfs_attr(const char *name, + struct dentry *parent, struct fault_attr *attr) { mode_t mode = S_IFREG | S_IRUSR | S_IWUSR; struct dentry *dir; - dir = debugfs_create_dir(name, NULL); + dir = debugfs_create_dir(name, parent); if (!dir) - return -ENOMEM; - - attr->dir = dir; + return ERR_PTR(-ENOMEM); if (!debugfs_create_ul("probability", mode, dir, &attr->probability)) goto fail; @@ -243,11 +237,11 @@ int init_fault_attr_dentries(struct fault_attr *attr, const char *name) #endif /* CONFIG_FAULT_INJECTION_STACKTRACE_FILTER */ - return 0; + return dir; fail: - debugfs_remove_recursive(attr->dir); + debugfs_remove_recursive(dir); - return -ENOMEM; + return ERR_PTR(-ENOMEM); } #endif /* CONFIG_FAULT_INJECTION_DEBUG_FS */ diff --git a/mm/failslab.c b/mm/failslab.c index 1ce58c201dca..0dd7b8fec71c 100644 --- a/mm/failslab.c +++ b/mm/failslab.c @@ -34,23 +34,23 @@ __setup("failslab=", setup_failslab); #ifdef CONFIG_FAULT_INJECTION_DEBUG_FS static int __init failslab_debugfs_init(void) { + struct dentry *dir; mode_t mode = S_IFREG | S_IRUSR | S_IWUSR; - int err; - err = init_fault_attr_dentries(&failslab.attr, "failslab"); - if (err) - return err; + dir = fault_create_debugfs_attr("failslab", NULL, &failslab.attr); + if (IS_ERR(dir)) + return PTR_ERR(dir); - if (!debugfs_create_bool("ignore-gfp-wait", mode, failslab.attr.dir, + if (!debugfs_create_bool("ignore-gfp-wait", mode, dir, &failslab.ignore_gfp_wait)) goto fail; - if (!debugfs_create_bool("cache-filter", mode, failslab.attr.dir, + if (!debugfs_create_bool("cache-filter", mode, dir, &failslab.cache_filter)) goto fail; return 0; fail: - cleanup_fault_attr_dentries(&failslab.attr); + debugfs_remove_recursive(dir); return -ENOMEM; } diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 1dbcf8888f14..6e8ecb6e021c 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1409,14 +1409,11 @@ static int __init fail_page_alloc_debugfs(void) { mode_t mode = S_IFREG | S_IRUSR | S_IWUSR; struct dentry *dir; - int err; - err = init_fault_attr_dentries(&fail_page_alloc.attr, - "fail_page_alloc"); - if (err) - return err; - - dir = fail_page_alloc.attr.dir; + dir = fault_create_debugfs_attr("fail_page_alloc", NULL, + &fail_page_alloc.attr); + if (IS_ERR(dir)) + return PTR_ERR(dir); if (!debugfs_create_bool("ignore-gfp-wait", mode, dir, &fail_page_alloc.ignore_gfp_wait)) @@ -1430,7 +1427,7 @@ static int __init fail_page_alloc_debugfs(void) return 0; fail: - cleanup_fault_attr_dentries(&fail_page_alloc.attr); + debugfs_remove_recursive(dir); return -ENOMEM; } From 12b3e038e5bb4860c17d001e92a6fa9964c0a7b9 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Daniel=20Gl=C3=B6ckner?= Date: Wed, 3 Aug 2011 16:21:02 -0700 Subject: [PATCH 281/439] rtc-omap: fix initialization of control register MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit As the comment explains, the intention of the code is to clear the OMAP_RTC_CTRL_MODE_12_24 bit, but instead it only clears the OMAP_RTC_CTRL_SPLIT and OMAP_RTC_CTRL_AUTO_COMP bits, which should be kept. OMAP_RTC_CTRL_DISABLE, OMAP_RTC_CTRL_SET_32_COUNTER, OMAP_RTC_CTRL_TEST, and OMAP_RTC_CTRL_ROUND_30S are also better off being cleared. Signed-off-by: Daniel Glöckner Cc: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/rtc-omap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/rtc/rtc-omap.c b/drivers/rtc/rtc-omap.c index bcae8dd41496..7789002bdd5c 100644 --- a/drivers/rtc/rtc-omap.c +++ b/drivers/rtc/rtc-omap.c @@ -368,7 +368,7 @@ static int __init omap_rtc_probe(struct platform_device *pdev) pr_info("%s: already running\n", pdev->name); /* force to 24 hour mode */ - new_ctrl = reg & ~(OMAP_RTC_CTRL_SPLIT|OMAP_RTC_CTRL_AUTO_COMP); + new_ctrl = reg & (OMAP_RTC_CTRL_SPLIT|OMAP_RTC_CTRL_AUTO_COMP); new_ctrl |= OMAP_RTC_CTRL_STOP; /* BOARD-SPECIFIC CUSTOMIZATION CAN GO HERE: From dfc428b656c4693a2334a8d9865b430beddb562a Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Wed, 3 Aug 2011 16:21:04 -0700 Subject: [PATCH 282/439] taskstats: add_del_listener() shouldn't use the wrong node 1. Commit 26c4caea9d69 "don't allow duplicate entries in listener mode" changed add_del_listener(REGISTER) so that "next_cpu:" can reuse the listener allocated for the previous cpu, this doesn't look exactly right even if minor. Change the code to kfree() in the already-registered case, this case is unlikely anyway so the extra kmalloc_node() shouldn't hurt but looke more correct and clean. 2. use the plain list_for_each_entry() instead of _safe() to scan listeners->list. 3. Remove the unneeded INIT_LIST_HEAD(&s->list), we are going to list_add(&s->list). Signed-off-by: Oleg Nesterov Reviewed-by: Vasiliy Kulikov Cc: Balbir Singh Reviewed-by: Jerome Marchand Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/taskstats.c | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-) diff --git a/kernel/taskstats.c b/kernel/taskstats.c index d1db2880d1cf..a09a54936f19 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -291,30 +291,28 @@ static int add_del_listener(pid_t pid, const struct cpumask *mask, int isadd) if (!cpumask_subset(mask, cpu_possible_mask)) return -EINVAL; - s = NULL; if (isadd == REGISTER) { for_each_cpu(cpu, mask) { - if (!s) - s = kmalloc_node(sizeof(struct listener), - GFP_KERNEL, cpu_to_node(cpu)); + s = kmalloc_node(sizeof(struct listener), + GFP_KERNEL, cpu_to_node(cpu)); if (!s) goto cleanup; + s->pid = pid; - INIT_LIST_HEAD(&s->list); s->valid = 1; listeners = &per_cpu(listener_array, cpu); down_write(&listeners->sem); - list_for_each_entry_safe(s2, tmp, &listeners->list, list) { + list_for_each_entry(s2, &listeners->list, list) { if (s2->pid == pid) - goto next_cpu; + goto exists; } list_add(&s->list, &listeners->list); s = NULL; -next_cpu: +exists: up_write(&listeners->sem); + kfree(s); /* nop if NULL */ } - kfree(s); return 0; } From a7295898a1d2e501427f557111c2b4bdfc90b1ed Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Wed, 3 Aug 2011 16:21:05 -0700 Subject: [PATCH 283/439] taskstats: add_del_listener() should ignore !valid listeners When send_cpu_listeners() finds the orphaned listener it marks it as !valid and drops listeners->sem. Before it takes this sem for writing, s->pid can be reused and add_del_listener() can wrongly try to re-use this entry. Change add_del_listener() to check ->valid = T. Signed-off-by: Oleg Nesterov Reviewed-by: Vasiliy Kulikov Acked-by: Balbir Singh Cc: Jerome Marchand Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/taskstats.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/taskstats.c b/kernel/taskstats.c index a09a54936f19..e19ce1454ee1 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -304,7 +304,7 @@ static int add_del_listener(pid_t pid, const struct cpumask *mask, int isadd) listeners = &per_cpu(listener_array, cpu); down_write(&listeners->sem); list_for_each_entry(s2, &listeners->list, list) { - if (s2->pid == pid) + if (s2->pid == pid && s2->valid) goto exists; } list_add(&s->list, &listeners->list); From 88eca0207cf1574328c3ce8c3be537a9317261bb Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Wed, 3 Aug 2011 16:21:06 -0700 Subject: [PATCH 284/439] ida: simplified functions for id allocation The current hyper-optimized functions are overkill if you simply want to allocate an id for a device. Create versions which use an internal lock. In followup patches, numerous drivers are converted to use this interface. Thanks to Tejun for feedback. Signed-off-by: Rusty Russell Acked-by: Tejun Heo Acked-by: Jonathan Cameron Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/idr.h | 4 +++ lib/idr.c | 67 +++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 71 insertions(+) diff --git a/include/linux/idr.h b/include/linux/idr.h index 13a801f3d028..255491cf522e 100644 --- a/include/linux/idr.h +++ b/include/linux/idr.h @@ -146,6 +146,10 @@ void ida_remove(struct ida *ida, int id); void ida_destroy(struct ida *ida); void ida_init(struct ida *ida); +int ida_simple_get(struct ida *ida, unsigned int start, unsigned int end, + gfp_t gfp_mask); +void ida_simple_remove(struct ida *ida, unsigned int id); + void __init idr_init_cache(void); #endif /* __IDR_H__ */ diff --git a/lib/idr.c b/lib/idr.c index e15502e8b21e..db040ce3fa73 100644 --- a/lib/idr.c +++ b/lib/idr.c @@ -34,8 +34,10 @@ #include #include #include +#include static struct kmem_cache *idr_layer_cache; +static DEFINE_SPINLOCK(simple_ida_lock); static struct idr_layer *get_from_free_list(struct idr *idp) { @@ -925,6 +927,71 @@ void ida_destroy(struct ida *ida) } EXPORT_SYMBOL(ida_destroy); +/** + * ida_simple_get - get a new id. + * @ida: the (initialized) ida. + * @start: the minimum id (inclusive, < 0x8000000) + * @end: the maximum id (exclusive, < 0x8000000 or 0) + * @gfp_mask: memory allocation flags + * + * Allocates an id in the range start <= id < end, or returns -ENOSPC. + * On memory allocation failure, returns -ENOMEM. + * + * Use ida_simple_remove() to get rid of an id. + */ +int ida_simple_get(struct ida *ida, unsigned int start, unsigned int end, + gfp_t gfp_mask) +{ + int ret, id; + unsigned int max; + + BUG_ON((int)start < 0); + BUG_ON((int)end < 0); + + if (end == 0) + max = 0x80000000; + else { + BUG_ON(end < start); + max = end - 1; + } + +again: + if (!ida_pre_get(ida, gfp_mask)) + return -ENOMEM; + + spin_lock(&simple_ida_lock); + ret = ida_get_new_above(ida, start, &id); + if (!ret) { + if (id > max) { + ida_remove(ida, id); + ret = -ENOSPC; + } else { + ret = id; + } + } + spin_unlock(&simple_ida_lock); + + if (unlikely(ret == -EAGAIN)) + goto again; + + return ret; +} +EXPORT_SYMBOL(ida_simple_get); + +/** + * ida_simple_remove - remove an allocated id. + * @ida: the (initialized) ida. + * @id: the id returned by ida_simple_get. + */ +void ida_simple_remove(struct ida *ida, unsigned int id) +{ + BUG_ON((int)id < 0); + spin_lock(&simple_ida_lock); + ida_remove(ida, id); + spin_unlock(&simple_ida_lock); +} +EXPORT_SYMBOL(ida_simple_remove); + /** * ida_init - initialize ida handle * @ida: ida handle From 2af14162656b81bea9e03e76d7c5f1787cc86ea6 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 3 Aug 2011 16:21:07 -0700 Subject: [PATCH 285/439] fs/dcache.c: fix new kernel-doc warning Fix new kernel-doc warning in fs/dcache.c: Warning(fs/dcache.c:797): No description found for parameter 'sb' Signed-off-by: Randy Dunlap Cc: Al Viro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/dcache.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/dcache.c b/fs/dcache.c index 2347cdb15abb..c83cae19161e 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -795,6 +795,7 @@ relock: /** * prune_dcache_sb - shrink the dcache + * @sb: superblock * @nr_to_scan: number of entries to try to free * * Attempt to shrink the superblock dcache LRU by @nr_to_scan entries. This is From f2d34fd9435c7e60cb5189d036efe9abfc911862 Mon Sep 17 00:00:00 2001 From: Will Drewry Date: Wed, 3 Aug 2011 16:21:08 -0700 Subject: [PATCH 286/439] Documentation: add pointer to name_to_dev_t for root= values Update kernel-parameters.txt to point users to the authoritative comment for name_to_dev_t. In addition, updates other places where some name_to_dev_t behavior was discussed. All other references to root= appear to be for explicit sample usage or just side comments when discussing other kernel parameters. Signed-off-by: Will Drewry Cc: Kay Sievers Cc: Randy Dunlap Cc: Namhyung Kim Cc: Trond Myklebust Cc: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/frv/booting.txt | 13 ++++++++++--- Documentation/kernel-parameters.txt | 1 + Documentation/m68k/kernel-options.txt | 14 ++++++++++++++ 3 files changed, 25 insertions(+), 3 deletions(-) diff --git a/Documentation/frv/booting.txt b/Documentation/frv/booting.txt index ace200b7c214..37c4d84a0e57 100644 --- a/Documentation/frv/booting.txt +++ b/Documentation/frv/booting.txt @@ -106,13 +106,20 @@ separated by spaces: To use the first on-chip serial port at baud rate 115200, no parity, 8 bits, and no flow control. - (*) root=/dev/ + (*) root= - This specifies the device upon which the root filesystem resides. For - example: + This specifies the device upon which the root filesystem resides. It + may be specified by major and minor number, device path, or even + partition uuid, if supported. For example: /dev/nfs NFS root filesystem /dev/mtdblock3 Fourth RedBoot partition on the System Flash + PARTUUID=00112233-4455-6677-8899-AABBCCDDEEFF/PARTNROFF=1 + first partition after the partition with the given UUID + 253:0 Device with major 253 and minor 0 + + Authoritative information can be found in + "Documentation/kernel-parameters.txt". (*) rw diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index d57f00d1bb25..865e39f1850c 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -2245,6 +2245,7 @@ bytes respectively. Such letter suffixes can also be entirely omitted. ro [KNL] Mount root device read-only on boot root= [KNL] Root filesystem + See name_to_dev_t comment in init/do_mounts.c. rootdelay= [KNL] Delay (in seconds) to pause before attempting to mount the root filesystem diff --git a/Documentation/m68k/kernel-options.txt b/Documentation/m68k/kernel-options.txt index c93bed66e25d..97d45f276fe6 100644 --- a/Documentation/m68k/kernel-options.txt +++ b/Documentation/m68k/kernel-options.txt @@ -129,6 +129,20 @@ decimal 11 is the major of SCSI CD-ROMs, and the minor 0 stands for the first of these. You can find out all valid major numbers by looking into include/linux/major.h. +In addition to major and minor numbers, if the device containing your +root partition uses a partition table format with unique partition +identifiers, then you may use them. For instance, +"root=PARTUUID=00112233-4455-6677-8899-AABBCCDDEEFF". It is also +possible to reference another partition on the same device using a +known partition UUID as the starting point. For example, +if partition 5 of the device has the UUID of +00112233-4455-6677-8899-AABBCCDDEEFF then partition 3 may be found as +follows: + PARTUUID=00112233-4455-6677-8899-AABBCCDDEEFF/PARTNROFF=-2 + +Authoritative information can be found in +"Documentation/kernel-parameters.txt". + 2.2) ro, rw ----------- From f7b9fcbbc3b8593ff7dc587f90c2fe90a2fd1e6f Mon Sep 17 00:00:00 2001 From: Sergiu Iordache Date: Wed, 3 Aug 2011 16:21:09 -0700 Subject: [PATCH 287/439] ramoops: update module parameters Update the module parameters when platform data is used. This means that they can be read from /sys/module/ramoops/parameters in order to parse the memory area. Signed-off-by: Sergiu Iordache Cc: Marco Stornelli Cc: Seiji Aguchi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/ramoops.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/char/ramoops.c b/drivers/char/ramoops.c index fca0c51bbc90..810aff9e750f 100644 --- a/drivers/char/ramoops.c +++ b/drivers/char/ramoops.c @@ -147,6 +147,14 @@ static int __init ramoops_probe(struct platform_device *pdev) cxt->phys_addr = pdata->mem_address; cxt->record_size = pdata->record_size; cxt->dump_oops = pdata->dump_oops; + /* + * Update the module parameter variables as well so they are visible + * through /sys/module/ramoops/parameters/ + */ + mem_size = pdata->mem_size; + mem_address = pdata->mem_address; + record_size = pdata->record_size; + dump_oops = pdata->dump_oops; if (!request_mem_region(cxt->phys_addr, cxt->size, "ramoops")) { pr_err("request mem region failed\n"); From 3dab1bce8e840606e76f1bad5a009328e2aa35c7 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 3 Aug 2011 16:21:09 -0700 Subject: [PATCH 288/439] mm: page_alloc: increase __GFP_BITS_SHIFT to include __GFP_OTHER_NODE __GFP_OTHER_NODE is used for NUMA allocations on behalf of other nodes. It's supposed to be passed through from the page allocator to zone_statistics(), but it never gets there as gfp_allowed_mask is not wide enough and masks out the flag early in the allocation path. The result is an accounting glitch where successful NUMA allocations by-agent are not properly attributed as local. Increase __GFP_BITS_SHIFT so that it includes __GFP_OTHER_NODE. Signed-off-by: Johannes Weiner Acked-by: Andi Kleen Reviewed-by: Minchan Kim Acked-by: Mel Gorman Reviewed-by: Michal Hocko Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/gfp.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/gfp.h b/include/linux/gfp.h index cb4089254f01..3a76faf6a3ee 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -92,7 +92,7 @@ struct vm_area_struct; */ #define __GFP_NOTRACK_FALSE_POSITIVE (__GFP_NOTRACK) -#define __GFP_BITS_SHIFT 23 /* Room for 23 __GFP_FOO bits */ +#define __GFP_BITS_SHIFT 24 /* Room for N __GFP_FOO bits */ #define __GFP_BITS_MASK ((__force gfp_t)((1 << __GFP_BITS_SHIFT) - 1)) /* This equals 0, but use constants in case they ever change */ From 1560ffe62a9d53a51faeec7417becfba4f2a0d18 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 3 Aug 2011 16:21:11 -0700 Subject: [PATCH 289/439] tpm_tis: fix build when ACPI is not enabled Fix tpm_tis.c build when CONFIG_ACPI is not enabled by providing a stub function. Fixes many build errors/warnings: drivers/char/tpm/tpm_tis.c:89: error: dereferencing pointer to incomplete type drivers/char/tpm/tpm_tis.c:89: warning: type defaults to 'int' in declaration of 'type name' drivers/char/tpm/tpm_tis.c:89: error: request for member 'list' in something not a structure or union ... Signed-off-by: Randy Dunlap Cc: Leendert van Doorn Cc: James Morris Cc: Ingo Molnar Cc: Stephen Rothwell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/tpm/tpm_tis.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/char/tpm/tpm_tis.c b/drivers/char/tpm/tpm_tis.c index 7fc2f108f490..3f4051a7c5a7 100644 --- a/drivers/char/tpm/tpm_tis.c +++ b/drivers/char/tpm/tpm_tis.c @@ -80,7 +80,7 @@ enum tis_defaults { static LIST_HEAD(tis_chips); static DEFINE_SPINLOCK(tis_lock); -#ifdef CONFIG_PNP +#if defined(CONFIG_PNP) && defined(CONFIG_ACPI) static int is_itpm(struct pnp_dev *dev) { struct acpi_device *acpi = pnp_acpi_device(dev); @@ -93,6 +93,11 @@ static int is_itpm(struct pnp_dev *dev) return 0; } +#else +static inline int is_itpm(struct pnp_dev *dev) +{ + return 0; +} #endif static int check_locality(struct tpm_chip *chip, int l) From d4969213f9e75ec1bfa6ea65c279c64cab7d1bd6 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Wed, 3 Aug 2011 16:21:12 -0700 Subject: [PATCH 290/439] cris: fix a build error in kernel/fork.c Fix this error: kernel/fork.c:267: error: implicit declaration of function 'alloc_thread_info_node' This is due to renaming alloc_thread_info() to alloc_thread_info_node(). [akpm@linux-foundation.org: coding-style fixes] Reported-by: Geert Uytterhoeven Signed-off-by: WANG Cong Cc: Mikael Starvik Cc: Jesper Nilsson Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/cris/include/asm/thread_info.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/cris/include/asm/thread_info.h b/arch/cris/include/asm/thread_info.h index 29b74a105830..332f19c54557 100644 --- a/arch/cris/include/asm/thread_info.h +++ b/arch/cris/include/asm/thread_info.h @@ -11,8 +11,6 @@ #ifdef __KERNEL__ -#define __HAVE_ARCH_THREAD_INFO_ALLOCATOR - #ifndef __ASSEMBLY__ #include #include @@ -67,8 +65,10 @@ struct thread_info { #define init_thread_info (init_thread_union.thread_info) +#define __HAVE_ARCH_THREAD_INFO_ALLOCATOR /* thread information allocation */ -#define alloc_thread_info(tsk, node) ((struct thread_info *) __get_free_pages(GFP_KERNEL,1)) +#define alloc_thread_info_node(tsk, node) \ + ((struct thread_info *) __get_free_pages(GFP_KERNEL, 1)) #define free_thread_info(ti) free_pages((unsigned long) (ti), 1) #endif /* !__ASSEMBLY__ */ From 4b851d88192c22cf77418a0b4c45b5c789276837 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Wed, 3 Aug 2011 16:21:14 -0700 Subject: [PATCH 291/439] cris: fix a build error in sync_serial_open() Fix: arch/cris/arch-v10/drivers/sync_serial.c:628: error: 'ret' undeclared (first use in this function) 'ret' should be 'err'. Reported-by: Geert Uytterhoeven Signed-off-by: WANG Cong Cc: Mikael Starvik Cc: Jesper Nilsson Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/cris/arch-v10/drivers/sync_serial.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/cris/arch-v10/drivers/sync_serial.c b/arch/cris/arch-v10/drivers/sync_serial.c index 850265373611..d2ad9e7a1738 100644 --- a/arch/cris/arch-v10/drivers/sync_serial.c +++ b/arch/cris/arch-v10/drivers/sync_serial.c @@ -625,11 +625,11 @@ static int sync_serial_open(struct inode *inode, struct file *file) *R_IRQ_MASK1_SET = 1 << port->data_avail_bit; DEBUG(printk(KERN_DEBUG "sser%d rec started\n", dev)); } - ret = 0; + err = 0; out: mutex_unlock(&sync_serial_mutex); - return ret; + return err; } static int sync_serial_release(struct inode *inode, struct file *file) From b4bc281266e84e9a432b588ebdcef5fb94dc8ecb Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Wed, 3 Aug 2011 16:21:14 -0700 Subject: [PATCH 292/439] cris: fix the prototype of sync_serial_ioctl() Fix: arch/cris/arch-v10/drivers/sync_serial.c:961: error: conflicting types for 'sync_serial_ioctl' Reported-by: Geert Uytterhoeven Signed-off-by: WANG Cong Cc: Mikael Starvik Cc: Jesper Nilsson Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/cris/arch-v10/drivers/sync_serial.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/cris/arch-v10/drivers/sync_serial.c b/arch/cris/arch-v10/drivers/sync_serial.c index d2ad9e7a1738..466af40c5822 100644 --- a/arch/cris/arch-v10/drivers/sync_serial.c +++ b/arch/cris/arch-v10/drivers/sync_serial.c @@ -158,7 +158,7 @@ static int sync_serial_open(struct inode *inode, struct file *file); static int sync_serial_release(struct inode *inode, struct file *file); static unsigned int sync_serial_poll(struct file *filp, poll_table *wait); -static int sync_serial_ioctl(struct file *file, +static long sync_serial_ioctl(struct file *file, unsigned int cmd, unsigned long arg); static ssize_t sync_serial_write(struct file *file, const char *buf, size_t count, loff_t *ppos); From 1646ec9db75e151b0479dbfaf972f741d0476ec7 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Wed, 3 Aug 2011 16:21:15 -0700 Subject: [PATCH 293/439] cris: add missing declaration of kgdb_init() and breakpoint() Fix: arch/cris/arch-v10/kernel/irq.c:239: error: implicit declaration of function 'kgdb_init' arch/cris/arch-v10/kernel/irq.c:240: error: implicit declaration of function 'breakpoint' Declare these two functions. Reported-by: Geert Uytterhoeven Signed-off-by: WANG Cong Cc: Mikael Starvik Cc: Jesper Nilsson Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/cris/arch-v10/kernel/irq.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/cris/arch-v10/kernel/irq.c b/arch/cris/arch-v10/kernel/irq.c index 907cfb5a873d..ba0e5965d6e3 100644 --- a/arch/cris/arch-v10/kernel/irq.c +++ b/arch/cris/arch-v10/kernel/irq.c @@ -20,6 +20,9 @@ #define crisv10_mask_irq(irq_nr) (*R_VECT_MASK_CLR = 1 << (irq_nr)); #define crisv10_unmask_irq(irq_nr) (*R_VECT_MASK_SET = 1 << (irq_nr)); +extern void kgdb_init(void); +extern void breakpoint(void); + /* don't use set_int_vector, it bypasses the linux interrupt handlers. it is * global just so that the kernel gdb can use it. */ From 4c4dd903e72507c853b8c2b04b22e0d0c721fd93 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Wed, 3 Aug 2011 16:21:16 -0700 Subject: [PATCH 294/439] drivers/video/backlight/aat2870_bl.c: fix error checking for backlight_device_register backlight_device_register() returns ERR_PTR() on error. Signed-off-by: Axel Lin Cc: Richard Purdie Cc: Jin Park Cc: Samuel Ortiz Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/backlight/aat2870_bl.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/video/backlight/aat2870_bl.c b/drivers/video/backlight/aat2870_bl.c index 4952a617563d..f13a3f7e2e8f 100644 --- a/drivers/video/backlight/aat2870_bl.c +++ b/drivers/video/backlight/aat2870_bl.c @@ -158,10 +158,10 @@ static int aat2870_bl_probe(struct platform_device *pdev) props.type = BACKLIGHT_RAW; bd = backlight_device_register("aat2870-backlight", &pdev->dev, aat2870_bl, &aat2870_bl_ops, &props); - if (!bd) { + if (IS_ERR(bd)) { dev_err(&pdev->dev, "Failed allocate memory for backlight device\n"); - ret = -ENOMEM; + ret = PTR_ERR(bd); goto out_kfree; } From 5d6f921b42749d1a70441685b7a4f2801e12ebfb Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Wed, 3 Aug 2011 16:21:17 -0700 Subject: [PATCH 295/439] drivers/video/backlight/aat2870_bl.c: fix setting max_current - Current implementation tests wrong value for setting aat2870_bl->max_current. - In the current implementation, we cannot differentiate between 2 cases: a) if pdata->max_current is not set , or b) pdata->max_current is set to AAT2870_CURRENT_0_45 (which is also 0). Fix it by setting AAT2870_CURRENT_0_45 to be 1 and adjust the equation in aat2870_brightness() accordingly. Signed-off-by: Axel Lin Cc: Richard Purdie Cc: Samuel Ortiz Tested-by: Jin Park Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/backlight/aat2870_bl.c | 4 ++-- include/linux/mfd/aat2870.h | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/video/backlight/aat2870_bl.c b/drivers/video/backlight/aat2870_bl.c index f13a3f7e2e8f..331f1ef1dad5 100644 --- a/drivers/video/backlight/aat2870_bl.c +++ b/drivers/video/backlight/aat2870_bl.c @@ -44,7 +44,7 @@ static inline int aat2870_brightness(struct aat2870_bl_driver_data *aat2870_bl, struct backlight_device *bd = aat2870_bl->bd; int val; - val = brightness * aat2870_bl->max_current; + val = brightness * (aat2870_bl->max_current - 1); val /= bd->props.max_brightness; return val; @@ -175,7 +175,7 @@ static int aat2870_bl_probe(struct platform_device *pdev) else aat2870_bl->channels = AAT2870_BL_CH_ALL; - if (pdata->max_brightness > 0) + if (pdata->max_current > 0) aat2870_bl->max_current = pdata->max_current; else aat2870_bl->max_current = AAT2870_CURRENT_27_9; diff --git a/include/linux/mfd/aat2870.h b/include/linux/mfd/aat2870.h index 89212df05622..f7316c29bdec 100644 --- a/include/linux/mfd/aat2870.h +++ b/include/linux/mfd/aat2870.h @@ -89,7 +89,7 @@ enum aat2870_id { /* Backlight current magnitude (mA) */ enum aat2870_current { - AAT2870_CURRENT_0_45, + AAT2870_CURRENT_0_45 = 1, AAT2870_CURRENT_0_90, AAT2870_CURRENT_1_80, AAT2870_CURRENT_2_70, From 70d327198a434edb95b3d858bc8010b8add28e3e Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 3 Aug 2011 16:21:17 -0700 Subject: [PATCH 296/439] drivers/video/backlight/aat2870_bl.c: make it buildable as a module i386 allmodconfig: drivers/built-in.o: In function `aat2870_bl_remove': aat2870_bl.c:(.text+0x414f9): undefined reference to `backlight_device_unregister' drivers/built-in.o: In function `aat2870_bl_probe': aat2870_bl.c:(.text+0x418fc): undefined reference to `backlight_device_register' aat2870_bl.c:(.text+0x41a31): undefined reference to `backlight_device_unregiste Cc: Jin Park Cc: Samuel Ortiz Cc: Axel Lin Cc: Richard Purdie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/backlight/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/video/backlight/Kconfig b/drivers/video/backlight/Kconfig index 69407e72aac1..278aeaa92505 100644 --- a/drivers/video/backlight/Kconfig +++ b/drivers/video/backlight/Kconfig @@ -336,7 +336,7 @@ config BACKLIGHT_PCF50633 enable its driver. config BACKLIGHT_AAT2870 - bool "AnalogicTech AAT2870 Backlight" + tristate "AnalogicTech AAT2870 Backlight" depends on BACKLIGHT_CLASS_DEVICE && MFD_AAT2870_CORE help If you have a AnalogicTech AAT2870 say Y to enable the From 6328650bb4d854a7dc1498d1c0048b838b0d340c Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 3 Aug 2011 16:21:18 -0700 Subject: [PATCH 297/439] radix_tree: exceptional entries and indices A patchset to extend tmpfs to MAX_LFS_FILESIZE by abandoning its peculiar swap vector, instead keeping a file's swap entries in the same radix tree as its struct page pointers: thus saving memory, and simplifying its code and locking. This patch: The radix_tree is used by several subsystems for different purposes. A major use is to store the struct page pointers of a file's pagecache for memory management. But what if mm wanted to store something other than page pointers there too? The low bit of a radix_tree entry is already used to denote an indirect pointer, for internal use, and the unlikely radix_tree_deref_retry() case. Define the next bit as denoting an exceptional entry, and supply inline functions radix_tree_exception() to return non-0 in either unlikely case, and radix_tree_exceptional_entry() to return non-0 in the second case. If a subsystem already uses radix_tree with that bit set, no problem: it does not affect internal workings at all, but is defined for the convenience of those storing well-aligned pointers in the radix_tree. The radix_tree_gang_lookups have an implicit assumption that the caller can deduce the offset of each entry returned e.g. by the page->index of a struct page. But that may not be feasible for some kinds of item to be stored there. radix_tree_gang_lookup_slot() allow for an optional indices argument, output array in which to return those offsets. The same could be added to other radix_tree_gang_lookups, but for now keep it to the only one for which we need it. Signed-off-by: Hugh Dickins Acked-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/radix-tree.h | 36 +++++++++++++++++++++++++++++++++--- lib/radix-tree.c | 29 +++++++++++++++++++---------- mm/filemap.c | 4 ++-- 3 files changed, 54 insertions(+), 15 deletions(-) diff --git a/include/linux/radix-tree.h b/include/linux/radix-tree.h index 23241c2fecce..b7edf8251455 100644 --- a/include/linux/radix-tree.h +++ b/include/linux/radix-tree.h @@ -39,7 +39,15 @@ * when it is shrunk, before we rcu free the node. See shrink code for * details. */ -#define RADIX_TREE_INDIRECT_PTR 1 +#define RADIX_TREE_INDIRECT_PTR 1 +/* + * A common use of the radix tree is to store pointers to struct pages; + * but shmem/tmpfs needs also to store swap entries in the same tree: + * those are marked as exceptional entries to distinguish them. + * EXCEPTIONAL_ENTRY tests the bit, EXCEPTIONAL_SHIFT shifts content past it. + */ +#define RADIX_TREE_EXCEPTIONAL_ENTRY 2 +#define RADIX_TREE_EXCEPTIONAL_SHIFT 2 #define radix_tree_indirect_to_ptr(ptr) \ radix_tree_indirect_to_ptr((void __force *)(ptr)) @@ -173,6 +181,28 @@ static inline int radix_tree_deref_retry(void *arg) return unlikely((unsigned long)arg & RADIX_TREE_INDIRECT_PTR); } +/** + * radix_tree_exceptional_entry - radix_tree_deref_slot gave exceptional entry? + * @arg: value returned by radix_tree_deref_slot + * Returns: 0 if well-aligned pointer, non-0 if exceptional entry. + */ +static inline int radix_tree_exceptional_entry(void *arg) +{ + /* Not unlikely because radix_tree_exception often tested first */ + return (unsigned long)arg & RADIX_TREE_EXCEPTIONAL_ENTRY; +} + +/** + * radix_tree_exception - radix_tree_deref_slot returned either exception? + * @arg: value returned by radix_tree_deref_slot + * Returns: 0 if well-aligned pointer, non-0 if either kind of exception. + */ +static inline int radix_tree_exception(void *arg) +{ + return unlikely((unsigned long)arg & + (RADIX_TREE_INDIRECT_PTR | RADIX_TREE_EXCEPTIONAL_ENTRY)); +} + /** * radix_tree_replace_slot - replace item in a slot * @pslot: pointer to slot, returned by radix_tree_lookup_slot @@ -194,8 +224,8 @@ void *radix_tree_delete(struct radix_tree_root *, unsigned long); unsigned int radix_tree_gang_lookup(struct radix_tree_root *root, void **results, unsigned long first_index, unsigned int max_items); -unsigned int -radix_tree_gang_lookup_slot(struct radix_tree_root *root, void ***results, +unsigned int radix_tree_gang_lookup_slot(struct radix_tree_root *root, + void ***results, unsigned long *indices, unsigned long first_index, unsigned int max_items); unsigned long radix_tree_next_hole(struct radix_tree_root *root, unsigned long index, unsigned long max_scan); diff --git a/lib/radix-tree.c b/lib/radix-tree.c index 7ea2e033d715..348eaefbed74 100644 --- a/lib/radix-tree.c +++ b/lib/radix-tree.c @@ -823,8 +823,8 @@ unsigned long radix_tree_prev_hole(struct radix_tree_root *root, EXPORT_SYMBOL(radix_tree_prev_hole); static unsigned int -__lookup(struct radix_tree_node *slot, void ***results, unsigned long index, - unsigned int max_items, unsigned long *next_index) +__lookup(struct radix_tree_node *slot, void ***results, unsigned long *indices, + unsigned long index, unsigned int max_items, unsigned long *next_index) { unsigned int nr_found = 0; unsigned int shift, height; @@ -857,12 +857,16 @@ __lookup(struct radix_tree_node *slot, void ***results, unsigned long index, /* Bottom level: grab some items */ for (i = index & RADIX_TREE_MAP_MASK; i < RADIX_TREE_MAP_SIZE; i++) { - index++; if (slot->slots[i]) { - results[nr_found++] = &(slot->slots[i]); - if (nr_found == max_items) + results[nr_found] = &(slot->slots[i]); + if (indices) + indices[nr_found] = index; + if (++nr_found == max_items) { + index++; goto out; + } } + index++; } out: *next_index = index; @@ -918,8 +922,8 @@ radix_tree_gang_lookup(struct radix_tree_root *root, void **results, if (cur_index > max_index) break; - slots_found = __lookup(node, (void ***)results + ret, cur_index, - max_items - ret, &next_index); + slots_found = __lookup(node, (void ***)results + ret, NULL, + cur_index, max_items - ret, &next_index); nr_found = 0; for (i = 0; i < slots_found; i++) { struct radix_tree_node *slot; @@ -944,6 +948,7 @@ EXPORT_SYMBOL(radix_tree_gang_lookup); * radix_tree_gang_lookup_slot - perform multiple slot lookup on radix tree * @root: radix tree root * @results: where the results of the lookup are placed + * @indices: where their indices should be placed (but usually NULL) * @first_index: start the lookup from this key * @max_items: place up to this many items at *results * @@ -958,7 +963,8 @@ EXPORT_SYMBOL(radix_tree_gang_lookup); * protection, radix_tree_deref_slot may fail requiring a retry. */ unsigned int -radix_tree_gang_lookup_slot(struct radix_tree_root *root, void ***results, +radix_tree_gang_lookup_slot(struct radix_tree_root *root, + void ***results, unsigned long *indices, unsigned long first_index, unsigned int max_items) { unsigned long max_index; @@ -974,6 +980,8 @@ radix_tree_gang_lookup_slot(struct radix_tree_root *root, void ***results, if (first_index > 0) return 0; results[0] = (void **)&root->rnode; + if (indices) + indices[0] = 0; return 1; } node = indirect_to_ptr(node); @@ -987,8 +995,9 @@ radix_tree_gang_lookup_slot(struct radix_tree_root *root, void ***results, if (cur_index > max_index) break; - slots_found = __lookup(node, results + ret, cur_index, - max_items - ret, &next_index); + slots_found = __lookup(node, results + ret, + indices ? indices + ret : NULL, + cur_index, max_items - ret, &next_index); ret += slots_found; if (next_index == 0) break; diff --git a/mm/filemap.c b/mm/filemap.c index 867d40222ec7..b83aebfd0a00 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -840,7 +840,7 @@ unsigned find_get_pages(struct address_space *mapping, pgoff_t start, rcu_read_lock(); restart: nr_found = radix_tree_gang_lookup_slot(&mapping->page_tree, - (void ***)pages, start, nr_pages); + (void ***)pages, NULL, start, nr_pages); ret = 0; for (i = 0; i < nr_found; i++) { struct page *page; @@ -903,7 +903,7 @@ unsigned find_get_pages_contig(struct address_space *mapping, pgoff_t index, rcu_read_lock(); restart: nr_found = radix_tree_gang_lookup_slot(&mapping->page_tree, - (void ***)pages, index, nr_pages); + (void ***)pages, NULL, index, nr_pages); ret = 0; for (i = 0; i < nr_found; i++) { struct page *page; From a2c16d6cb0e478812829ca84aeabd02e36af35eb Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 3 Aug 2011 16:21:19 -0700 Subject: [PATCH 298/439] mm: let swap use exceptional entries If swap entries are to be stored along with struct page pointers in a radix tree, they need to be distinguished as exceptional entries. Most of the handling of swap entries in radix tree will be contained in shmem.c, but a few functions in filemap.c's common code need to check for their appearance: find_get_page(), find_lock_page(), find_get_pages() and find_get_pages_contig(). So as not to slow their fast paths, tuck those checks inside the existing checks for unlikely radix_tree_deref_slot(); except for find_lock_page(), where it is an added test. And make it a BUG in find_get_pages_tag(), which is not applied to tmpfs files. A part of the reason for eliminating shmem_readpage() earlier, was to minimize the places where common code would need to allow for swap entries. The swp_entry_t known to swapfile.c must be massaged into a slightly different form when stored in the radix tree, just as it gets massaged into a pte_t when stored in page tables. In an i386 kernel this limits its information (type and page offset) to 30 bits: given 32 "types" of swapfile and 4kB pagesize, that's a maximum swapfile size of 128GB. Which is less than the 512GB we previously allowed with X86_PAE (where the swap entry can occupy the entire upper 32 bits of a pte_t), but not a new limitation on 32-bit without PAE; and there's not a new limitation on 64-bit (where swap filesize is already limited to 16TB by a 32-bit page offset). Thirty areas of 128GB is probably still enough swap for a 64GB 32-bit machine. Provide swp_to_radix_entry() and radix_to_swp_entry() conversions, and enforce filesize limit in read_swap_header(), just as for ptes. Signed-off-by: Hugh Dickins Acked-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/swapops.h | 23 +++++++++++++++++++ mm/filemap.c | 49 ++++++++++++++++++++++++++--------------- mm/swapfile.c | 20 ++++++++++------- 3 files changed, 66 insertions(+), 26 deletions(-) diff --git a/include/linux/swapops.h b/include/linux/swapops.h index cd42e30b7c6e..2189d3ffc85d 100644 --- a/include/linux/swapops.h +++ b/include/linux/swapops.h @@ -1,3 +1,8 @@ +#ifndef _LINUX_SWAPOPS_H +#define _LINUX_SWAPOPS_H + +#include + /* * swapcache pages are stored in the swapper_space radix tree. We want to * get good packing density in that tree, so the index should be dense in @@ -76,6 +81,22 @@ static inline pte_t swp_entry_to_pte(swp_entry_t entry) return __swp_entry_to_pte(arch_entry); } +static inline swp_entry_t radix_to_swp_entry(void *arg) +{ + swp_entry_t entry; + + entry.val = (unsigned long)arg >> RADIX_TREE_EXCEPTIONAL_SHIFT; + return entry; +} + +static inline void *swp_to_radix_entry(swp_entry_t entry) +{ + unsigned long value; + + value = entry.val << RADIX_TREE_EXCEPTIONAL_SHIFT; + return (void *)(value | RADIX_TREE_EXCEPTIONAL_ENTRY); +} + #ifdef CONFIG_MIGRATION static inline swp_entry_t make_migration_entry(struct page *page, int write) { @@ -169,3 +190,5 @@ static inline int non_swap_entry(swp_entry_t entry) return 0; } #endif + +#endif /* _LINUX_SWAPOPS_H */ diff --git a/mm/filemap.c b/mm/filemap.c index b83aebfd0a00..76bfb6460f57 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -714,9 +714,12 @@ repeat: page = radix_tree_deref_slot(pagep); if (unlikely(!page)) goto out; - if (radix_tree_deref_retry(page)) + if (radix_tree_exception(page)) { + if (radix_tree_exceptional_entry(page)) + goto out; + /* radix_tree_deref_retry(page) */ goto repeat; - + } if (!page_cache_get_speculative(page)) goto repeat; @@ -753,7 +756,7 @@ struct page *find_lock_page(struct address_space *mapping, pgoff_t offset) repeat: page = find_get_page(mapping, offset); - if (page) { + if (page && !radix_tree_exception(page)) { lock_page(page); /* Has the page been truncated? */ if (unlikely(page->mapping != mapping)) { @@ -849,11 +852,14 @@ repeat: if (unlikely(!page)) continue; - /* - * This can only trigger when the entry at index 0 moves out - * of or back to the root: none yet gotten, safe to restart. - */ - if (radix_tree_deref_retry(page)) { + if (radix_tree_exception(page)) { + if (radix_tree_exceptional_entry(page)) + continue; + /* + * radix_tree_deref_retry(page): + * can only trigger when entry at index 0 moves out of + * or back to root: none yet gotten, safe to restart. + */ WARN_ON(start | i); goto restart; } @@ -912,12 +918,16 @@ repeat: if (unlikely(!page)) continue; - /* - * This can only trigger when the entry at index 0 moves out - * of or back to the root: none yet gotten, safe to restart. - */ - if (radix_tree_deref_retry(page)) + if (radix_tree_exception(page)) { + if (radix_tree_exceptional_entry(page)) + break; + /* + * radix_tree_deref_retry(page): + * can only trigger when entry at index 0 moves out of + * or back to root: none yet gotten, safe to restart. + */ goto restart; + } if (!page_cache_get_speculative(page)) goto repeat; @@ -977,12 +987,15 @@ repeat: if (unlikely(!page)) continue; - /* - * This can only trigger when the entry at index 0 moves out - * of or back to the root: none yet gotten, safe to restart. - */ - if (radix_tree_deref_retry(page)) + if (radix_tree_exception(page)) { + BUG_ON(radix_tree_exceptional_entry(page)); + /* + * radix_tree_deref_retry(page): + * can only trigger when entry at index 0 moves out of + * or back to root: none yet gotten, safe to restart. + */ goto restart; + } if (!page_cache_get_speculative(page)) goto repeat; diff --git a/mm/swapfile.c b/mm/swapfile.c index 1b8c33907242..17bc224bce68 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -1924,20 +1924,24 @@ static unsigned long read_swap_header(struct swap_info_struct *p, /* * Find out how many pages are allowed for a single swap - * device. There are two limiting factors: 1) the number of - * bits for the swap offset in the swp_entry_t type and - * 2) the number of bits in the a swap pte as defined by - * the different architectures. In order to find the - * largest possible bit mask a swap entry with swap type 0 + * device. There are three limiting factors: 1) the number + * of bits for the swap offset in the swp_entry_t type, and + * 2) the number of bits in the swap pte as defined by the + * the different architectures, and 3) the number of free bits + * in an exceptional radix_tree entry. In order to find the + * largest possible bit mask, a swap entry with swap type 0 * and swap offset ~0UL is created, encoded to a swap pte, - * decoded to a swp_entry_t again and finally the swap + * decoded to a swp_entry_t again, and finally the swap * offset is extracted. This will mask all the bits from * the initial ~0UL mask that can't be encoded in either * the swp_entry_t or the architecture definition of a - * swap pte. + * swap pte. Then the same is done for a radix_tree entry. */ maxpages = swp_offset(pte_to_swp_entry( - swp_entry_to_pte(swp_entry(0, ~0UL)))) + 1; + swp_entry_to_pte(swp_entry(0, ~0UL)))); + maxpages = swp_offset(radix_to_swp_entry( + swp_to_radix_entry(swp_entry(0, maxpages)))) + 1; + if (maxpages > swap_header->info.last_page) { maxpages = swap_header->info.last_page + 1; /* p->max is an unsigned int: don't overflow it */ From 285b2c4fdd69ea73b4762785d8c6be83b6c074a6 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 3 Aug 2011 16:21:20 -0700 Subject: [PATCH 299/439] tmpfs: demolish old swap vector support The maximum size of a shmem/tmpfs file has been limited by the maximum size of its triple-indirect swap vector. With 4kB page size, maximum filesize was just over 2TB on a 32-bit kernel, but sadly one eighth of that on a 64-bit kernel. (With 8kB page size, maximum filesize was just over 4TB on a 64-bit kernel, but 16TB on a 32-bit kernel, MAX_LFS_FILESIZE being then more restrictive than swap vector layout.) It's a shame that tmpfs should be more restrictive than ramfs, and this limitation has now been noticed. Add another level to the swap vector? No, it became obscure and hard to maintain, once I complicated it to make use of highmem pages nine years ago: better choose another way. Surely, if 2.4 had had the radix tree pagecache introduced in 2.5, then tmpfs would never have invented its own peculiar radix tree: we would have fitted swap entries into the common radix tree instead, in much the same way as we fit swap entries into page tables. And why should each file have a separate radix tree for its pages and for its swap entries? The swap entries are required precisely where and when the pages are not. We want to put them together in a single radix tree: which can then avoid much of the locking which was needed to prevent them from being exchanged underneath us. This also avoids the waste of memory devoted to swap vectors, first in the shmem_inode itself, then at least two more pages once a file grew beyond 16 data pages (pages accounted by df and du, but not by memcg). Allocated upfront, to avoid allocation when under swapping pressure, but pure waste when CONFIG_SWAP is not set - I have never spattered around the ifdefs to prevent that, preferring this move to sharing the common radix tree instead. There are three downsides to sharing the radix tree. One, that it binds tmpfs more tightly to the rest of mm, either requiring knowledge of swap entries in radix tree there, or duplication of its code here in shmem.c. I believe that the simplications and memory savings (and probable higher performance, not yet measured) justify that. Two, that on HIGHMEM systems with SWAP enabled, it's the lowmem radix nodes that cannot be freed under memory pressure - whereas before it was the less precious highmem swap vector pages that could not be freed. I'm hoping that 64-bit has now been accessible for long enough, that the highmem argument has grown much less persuasive. Three, that swapoff is slower than it used to be on tmpfs files, since it's using a simple generic mechanism not tailored to it: I find this noticeable, and shall want to improve, but maybe nobody else will notice. So... now remove most of the old swap vector code from shmem.c. But, for the moment, keep the simple i_direct vector of 16 pages, with simple accessors shmem_put_swap() and shmem_get_swap(), as a toy implementation to help mark where swap needs to be handled in subsequent patches. Signed-off-by: Hugh Dickins Acked-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/shmem_fs.h | 2 - mm/shmem.c | 782 +++++---------------------------------- 2 files changed, 84 insertions(+), 700 deletions(-) diff --git a/include/linux/shmem_fs.h b/include/linux/shmem_fs.h index aa08fa8fd79b..80b695213fdb 100644 --- a/include/linux/shmem_fs.h +++ b/include/linux/shmem_fs.h @@ -17,9 +17,7 @@ struct shmem_inode_info { unsigned long flags; unsigned long alloced; /* data pages alloced to file */ unsigned long swapped; /* subtotal assigned to swap */ - unsigned long next_index; /* highest alloced index + 1 */ struct shared_policy policy; /* NUMA memory alloc policy */ - struct page *i_indirect; /* top indirect blocks page */ union { swp_entry_t i_direct[SHMEM_NR_DIRECT]; /* first blocks */ char inline_symlink[SHMEM_SYMLINK_INLINE_LEN]; diff --git a/mm/shmem.c b/mm/shmem.c index 5cc21f8b4cd3..5574b00ca771 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -66,37 +66,9 @@ static struct vfsmount *shm_mnt; #include #include -/* - * The maximum size of a shmem/tmpfs file is limited by the maximum size of - * its triple-indirect swap vector - see illustration at shmem_swp_entry(). - * - * With 4kB page size, maximum file size is just over 2TB on a 32-bit kernel, - * but one eighth of that on a 64-bit kernel. With 8kB page size, maximum - * file size is just over 4TB on a 64-bit kernel, but 16TB on a 32-bit kernel, - * MAX_LFS_FILESIZE being then more restrictive than swap vector layout. - * - * We use / and * instead of shifts in the definitions below, so that the swap - * vector can be tested with small even values (e.g. 20) for ENTRIES_PER_PAGE. - */ -#define ENTRIES_PER_PAGE (PAGE_CACHE_SIZE/sizeof(unsigned long)) -#define ENTRIES_PER_PAGEPAGE ((unsigned long long)ENTRIES_PER_PAGE*ENTRIES_PER_PAGE) - -#define SHMSWP_MAX_INDEX (SHMEM_NR_DIRECT + (ENTRIES_PER_PAGEPAGE/2) * (ENTRIES_PER_PAGE+1)) -#define SHMSWP_MAX_BYTES (SHMSWP_MAX_INDEX << PAGE_CACHE_SHIFT) - -#define SHMEM_MAX_BYTES min_t(unsigned long long, SHMSWP_MAX_BYTES, MAX_LFS_FILESIZE) -#define SHMEM_MAX_INDEX ((unsigned long)((SHMEM_MAX_BYTES+1) >> PAGE_CACHE_SHIFT)) - #define BLOCKS_PER_PAGE (PAGE_CACHE_SIZE/512) #define VM_ACCT(size) (PAGE_CACHE_ALIGN(size) >> PAGE_SHIFT) -/* info->flags needs VM_flags to handle pagein/truncate races efficiently */ -#define SHMEM_PAGEIN VM_READ -#define SHMEM_TRUNCATE VM_WRITE - -/* Definition to limit shmem_truncate's steps between cond_rescheds */ -#define LATENCY_LIMIT 64 - /* Pretend that each entry is of this size in directory's i_size */ #define BOGO_DIRENT_SIZE 20 @@ -107,7 +79,7 @@ struct shmem_xattr { char value[0]; }; -/* Flag allocation requirements to shmem_getpage and shmem_swp_alloc */ +/* Flag allocation requirements to shmem_getpage */ enum sgp_type { SGP_READ, /* don't exceed i_size, don't allocate page */ SGP_CACHE, /* don't exceed i_size, may allocate page */ @@ -137,56 +109,6 @@ static inline int shmem_getpage(struct inode *inode, pgoff_t index, mapping_gfp_mask(inode->i_mapping), fault_type); } -static inline struct page *shmem_dir_alloc(gfp_t gfp_mask) -{ - /* - * The above definition of ENTRIES_PER_PAGE, and the use of - * BLOCKS_PER_PAGE on indirect pages, assume PAGE_CACHE_SIZE: - * might be reconsidered if it ever diverges from PAGE_SIZE. - * - * Mobility flags are masked out as swap vectors cannot move - */ - return alloc_pages((gfp_mask & ~GFP_MOVABLE_MASK) | __GFP_ZERO, - PAGE_CACHE_SHIFT-PAGE_SHIFT); -} - -static inline void shmem_dir_free(struct page *page) -{ - __free_pages(page, PAGE_CACHE_SHIFT-PAGE_SHIFT); -} - -static struct page **shmem_dir_map(struct page *page) -{ - return (struct page **)kmap_atomic(page, KM_USER0); -} - -static inline void shmem_dir_unmap(struct page **dir) -{ - kunmap_atomic(dir, KM_USER0); -} - -static swp_entry_t *shmem_swp_map(struct page *page) -{ - return (swp_entry_t *)kmap_atomic(page, KM_USER1); -} - -static inline void shmem_swp_balance_unmap(void) -{ - /* - * When passing a pointer to an i_direct entry, to code which - * also handles indirect entries and so will shmem_swp_unmap, - * we must arrange for the preempt count to remain in balance. - * What kmap_atomic of a lowmem page does depends on config - * and architecture, so pretend to kmap_atomic some lowmem page. - */ - (void) kmap_atomic(ZERO_PAGE(0), KM_USER1); -} - -static inline void shmem_swp_unmap(swp_entry_t *entry) -{ - kunmap_atomic(entry, KM_USER1); -} - static inline struct shmem_sb_info *SHMEM_SB(struct super_block *sb) { return sb->s_fs_info; @@ -303,468 +225,56 @@ static void shmem_recalc_inode(struct inode *inode) } } -/** - * shmem_swp_entry - find the swap vector position in the info structure - * @info: info structure for the inode - * @index: index of the page to find - * @page: optional page to add to the structure. Has to be preset to - * all zeros - * - * If there is no space allocated yet it will return NULL when - * page is NULL, else it will use the page for the needed block, - * setting it to NULL on return to indicate that it has been used. - * - * The swap vector is organized the following way: - * - * There are SHMEM_NR_DIRECT entries directly stored in the - * shmem_inode_info structure. So small files do not need an addional - * allocation. - * - * For pages with index > SHMEM_NR_DIRECT there is the pointer - * i_indirect which points to a page which holds in the first half - * doubly indirect blocks, in the second half triple indirect blocks: - * - * For an artificial ENTRIES_PER_PAGE = 4 this would lead to the - * following layout (for SHMEM_NR_DIRECT == 16): - * - * i_indirect -> dir --> 16-19 - * | +-> 20-23 - * | - * +-->dir2 --> 24-27 - * | +-> 28-31 - * | +-> 32-35 - * | +-> 36-39 - * | - * +-->dir3 --> 40-43 - * +-> 44-47 - * +-> 48-51 - * +-> 52-55 - */ -static swp_entry_t *shmem_swp_entry(struct shmem_inode_info *info, unsigned long index, struct page **page) +static void shmem_put_swap(struct shmem_inode_info *info, pgoff_t index, + swp_entry_t swap) { - unsigned long offset; - struct page **dir; - struct page *subdir; - - if (index < SHMEM_NR_DIRECT) { - shmem_swp_balance_unmap(); - return info->i_direct+index; - } - if (!info->i_indirect) { - if (page) { - info->i_indirect = *page; - *page = NULL; - } - return NULL; /* need another page */ - } - - index -= SHMEM_NR_DIRECT; - offset = index % ENTRIES_PER_PAGE; - index /= ENTRIES_PER_PAGE; - dir = shmem_dir_map(info->i_indirect); - - if (index >= ENTRIES_PER_PAGE/2) { - index -= ENTRIES_PER_PAGE/2; - dir += ENTRIES_PER_PAGE/2 + index/ENTRIES_PER_PAGE; - index %= ENTRIES_PER_PAGE; - subdir = *dir; - if (!subdir) { - if (page) { - *dir = *page; - *page = NULL; - } - shmem_dir_unmap(dir); - return NULL; /* need another page */ - } - shmem_dir_unmap(dir); - dir = shmem_dir_map(subdir); - } - - dir += index; - subdir = *dir; - if (!subdir) { - if (!page || !(subdir = *page)) { - shmem_dir_unmap(dir); - return NULL; /* need a page */ - } - *dir = subdir; - *page = NULL; - } - shmem_dir_unmap(dir); - return shmem_swp_map(subdir) + offset; + if (index < SHMEM_NR_DIRECT) + info->i_direct[index] = swap; } -static void shmem_swp_set(struct shmem_inode_info *info, swp_entry_t *entry, unsigned long value) +static swp_entry_t shmem_get_swap(struct shmem_inode_info *info, pgoff_t index) { - long incdec = value? 1: -1; - - entry->val = value; - info->swapped += incdec; - if ((unsigned long)(entry - info->i_direct) >= SHMEM_NR_DIRECT) { - struct page *page = kmap_atomic_to_page(entry); - set_page_private(page, page_private(page) + incdec); - } + return (index < SHMEM_NR_DIRECT) ? + info->i_direct[index] : (swp_entry_t){0}; } -/** - * shmem_swp_alloc - get the position of the swap entry for the page. - * @info: info structure for the inode - * @index: index of the page to find - * @sgp: check and recheck i_size? skip allocation? - * @gfp: gfp mask to use for any page allocation - * - * If the entry does not exist, allocate it. - */ -static swp_entry_t *shmem_swp_alloc(struct shmem_inode_info *info, - unsigned long index, enum sgp_type sgp, gfp_t gfp) -{ - struct inode *inode = &info->vfs_inode; - struct shmem_sb_info *sbinfo = SHMEM_SB(inode->i_sb); - struct page *page = NULL; - swp_entry_t *entry; - - if (sgp != SGP_WRITE && - ((loff_t) index << PAGE_CACHE_SHIFT) >= i_size_read(inode)) - return ERR_PTR(-EINVAL); - - while (!(entry = shmem_swp_entry(info, index, &page))) { - if (sgp == SGP_READ) - return shmem_swp_map(ZERO_PAGE(0)); - /* - * Test used_blocks against 1 less max_blocks, since we have 1 data - * page (and perhaps indirect index pages) yet to allocate: - * a waste to allocate index if we cannot allocate data. - */ - if (sbinfo->max_blocks) { - if (percpu_counter_compare(&sbinfo->used_blocks, - sbinfo->max_blocks - 1) >= 0) - return ERR_PTR(-ENOSPC); - percpu_counter_inc(&sbinfo->used_blocks); - inode->i_blocks += BLOCKS_PER_PAGE; - } - - spin_unlock(&info->lock); - page = shmem_dir_alloc(gfp); - spin_lock(&info->lock); - - if (!page) { - shmem_free_blocks(inode, 1); - return ERR_PTR(-ENOMEM); - } - if (sgp != SGP_WRITE && - ((loff_t) index << PAGE_CACHE_SHIFT) >= i_size_read(inode)) { - entry = ERR_PTR(-EINVAL); - break; - } - if (info->next_index <= index) - info->next_index = index + 1; - } - if (page) { - /* another task gave its page, or truncated the file */ - shmem_free_blocks(inode, 1); - shmem_dir_free(page); - } - if (info->next_index <= index && !IS_ERR(entry)) - info->next_index = index + 1; - return entry; -} - -/** - * shmem_free_swp - free some swap entries in a directory - * @dir: pointer to the directory - * @edir: pointer after last entry of the directory - * @punch_lock: pointer to spinlock when needed for the holepunch case - */ -static int shmem_free_swp(swp_entry_t *dir, swp_entry_t *edir, - spinlock_t *punch_lock) -{ - spinlock_t *punch_unlock = NULL; - swp_entry_t *ptr; - int freed = 0; - - for (ptr = dir; ptr < edir; ptr++) { - if (ptr->val) { - if (unlikely(punch_lock)) { - punch_unlock = punch_lock; - punch_lock = NULL; - spin_lock(punch_unlock); - if (!ptr->val) - continue; - } - free_swap_and_cache(*ptr); - *ptr = (swp_entry_t){0}; - freed++; - } - } - if (punch_unlock) - spin_unlock(punch_unlock); - return freed; -} - -static int shmem_map_and_free_swp(struct page *subdir, int offset, - int limit, struct page ***dir, spinlock_t *punch_lock) -{ - swp_entry_t *ptr; - int freed = 0; - - ptr = shmem_swp_map(subdir); - for (; offset < limit; offset += LATENCY_LIMIT) { - int size = limit - offset; - if (size > LATENCY_LIMIT) - size = LATENCY_LIMIT; - freed += shmem_free_swp(ptr+offset, ptr+offset+size, - punch_lock); - if (need_resched()) { - shmem_swp_unmap(ptr); - if (*dir) { - shmem_dir_unmap(*dir); - *dir = NULL; - } - cond_resched(); - ptr = shmem_swp_map(subdir); - } - } - shmem_swp_unmap(ptr); - return freed; -} - -static void shmem_free_pages(struct list_head *next) -{ - struct page *page; - int freed = 0; - - do { - page = container_of(next, struct page, lru); - next = next->next; - shmem_dir_free(page); - freed++; - if (freed >= LATENCY_LIMIT) { - cond_resched(); - freed = 0; - } - } while (next); -} - -void shmem_truncate_range(struct inode *inode, loff_t start, loff_t end) +void shmem_truncate_range(struct inode *inode, loff_t lstart, loff_t lend) { + struct address_space *mapping = inode->i_mapping; struct shmem_inode_info *info = SHMEM_I(inode); - unsigned long idx; - unsigned long size; - unsigned long limit; - unsigned long stage; - unsigned long diroff; - struct page **dir; - struct page *topdir; - struct page *middir; - struct page *subdir; - swp_entry_t *ptr; - LIST_HEAD(pages_to_free); - long nr_pages_to_free = 0; - long nr_swaps_freed = 0; - int offset; - int freed; - int punch_hole; - spinlock_t *needs_lock; - spinlock_t *punch_lock; - unsigned long upper_limit; + pgoff_t start = (lstart + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; + pgoff_t end = (lend >> PAGE_CACHE_SHIFT); + pgoff_t index; + swp_entry_t swap; - truncate_inode_pages_range(inode->i_mapping, start, end); + truncate_inode_pages_range(mapping, lstart, lend); - inode->i_ctime = inode->i_mtime = CURRENT_TIME; - idx = (start + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; - if (idx >= info->next_index) - return; + if (end > SHMEM_NR_DIRECT) + end = SHMEM_NR_DIRECT; spin_lock(&info->lock); - info->flags |= SHMEM_TRUNCATE; - if (likely(end == (loff_t) -1)) { - limit = info->next_index; - upper_limit = SHMEM_MAX_INDEX; - info->next_index = idx; - needs_lock = NULL; - punch_hole = 0; - } else { - if (end + 1 >= inode->i_size) { /* we may free a little more */ - limit = (inode->i_size + PAGE_CACHE_SIZE - 1) >> - PAGE_CACHE_SHIFT; - upper_limit = SHMEM_MAX_INDEX; - } else { - limit = (end + 1) >> PAGE_CACHE_SHIFT; - upper_limit = limit; - } - needs_lock = &info->lock; - punch_hole = 1; - } - - topdir = info->i_indirect; - if (topdir && idx <= SHMEM_NR_DIRECT && !punch_hole) { - info->i_indirect = NULL; - nr_pages_to_free++; - list_add(&topdir->lru, &pages_to_free); - } - spin_unlock(&info->lock); - - if (info->swapped && idx < SHMEM_NR_DIRECT) { - ptr = info->i_direct; - size = limit; - if (size > SHMEM_NR_DIRECT) - size = SHMEM_NR_DIRECT; - nr_swaps_freed = shmem_free_swp(ptr+idx, ptr+size, needs_lock); - } - - /* - * If there are no indirect blocks or we are punching a hole - * below indirect blocks, nothing to be done. - */ - if (!topdir || limit <= SHMEM_NR_DIRECT) - goto done2; - - /* - * The truncation case has already dropped info->lock, and we're safe - * because i_size and next_index have already been lowered, preventing - * access beyond. But in the punch_hole case, we still need to take - * the lock when updating the swap directory, because there might be - * racing accesses by shmem_getpage(SGP_CACHE), shmem_unuse_inode or - * shmem_writepage. However, whenever we find we can remove a whole - * directory page (not at the misaligned start or end of the range), - * we first NULLify its pointer in the level above, and then have no - * need to take the lock when updating its contents: needs_lock and - * punch_lock (either pointing to info->lock or NULL) manage this. - */ - - upper_limit -= SHMEM_NR_DIRECT; - limit -= SHMEM_NR_DIRECT; - idx = (idx > SHMEM_NR_DIRECT)? (idx - SHMEM_NR_DIRECT): 0; - offset = idx % ENTRIES_PER_PAGE; - idx -= offset; - - dir = shmem_dir_map(topdir); - stage = ENTRIES_PER_PAGEPAGE/2; - if (idx < ENTRIES_PER_PAGEPAGE/2) { - middir = topdir; - diroff = idx/ENTRIES_PER_PAGE; - } else { - dir += ENTRIES_PER_PAGE/2; - dir += (idx - ENTRIES_PER_PAGEPAGE/2)/ENTRIES_PER_PAGEPAGE; - while (stage <= idx) - stage += ENTRIES_PER_PAGEPAGE; - middir = *dir; - if (*dir) { - diroff = ((idx - ENTRIES_PER_PAGEPAGE/2) % - ENTRIES_PER_PAGEPAGE) / ENTRIES_PER_PAGE; - if (!diroff && !offset && upper_limit >= stage) { - if (needs_lock) { - spin_lock(needs_lock); - *dir = NULL; - spin_unlock(needs_lock); - needs_lock = NULL; - } else - *dir = NULL; - nr_pages_to_free++; - list_add(&middir->lru, &pages_to_free); - } - shmem_dir_unmap(dir); - dir = shmem_dir_map(middir); - } else { - diroff = 0; - offset = 0; - idx = stage; + for (index = start; index < end; index++) { + swap = shmem_get_swap(info, index); + if (swap.val) { + free_swap_and_cache(swap); + shmem_put_swap(info, index, (swp_entry_t){0}); + info->swapped--; } } - for (; idx < limit; idx += ENTRIES_PER_PAGE, diroff++) { - if (unlikely(idx == stage)) { - shmem_dir_unmap(dir); - dir = shmem_dir_map(topdir) + - ENTRIES_PER_PAGE/2 + idx/ENTRIES_PER_PAGEPAGE; - while (!*dir) { - dir++; - idx += ENTRIES_PER_PAGEPAGE; - if (idx >= limit) - goto done1; - } - stage = idx + ENTRIES_PER_PAGEPAGE; - middir = *dir; - if (punch_hole) - needs_lock = &info->lock; - if (upper_limit >= stage) { - if (needs_lock) { - spin_lock(needs_lock); - *dir = NULL; - spin_unlock(needs_lock); - needs_lock = NULL; - } else - *dir = NULL; - nr_pages_to_free++; - list_add(&middir->lru, &pages_to_free); - } - shmem_dir_unmap(dir); - cond_resched(); - dir = shmem_dir_map(middir); - diroff = 0; - } - punch_lock = needs_lock; - subdir = dir[diroff]; - if (subdir && !offset && upper_limit-idx >= ENTRIES_PER_PAGE) { - if (needs_lock) { - spin_lock(needs_lock); - dir[diroff] = NULL; - spin_unlock(needs_lock); - punch_lock = NULL; - } else - dir[diroff] = NULL; - nr_pages_to_free++; - list_add(&subdir->lru, &pages_to_free); - } - if (subdir && page_private(subdir) /* has swap entries */) { - size = limit - idx; - if (size > ENTRIES_PER_PAGE) - size = ENTRIES_PER_PAGE; - freed = shmem_map_and_free_swp(subdir, - offset, size, &dir, punch_lock); - if (!dir) - dir = shmem_dir_map(middir); - nr_swaps_freed += freed; - if (offset || punch_lock) { - spin_lock(&info->lock); - set_page_private(subdir, - page_private(subdir) - freed); - spin_unlock(&info->lock); - } else - BUG_ON(page_private(subdir) != freed); - } - offset = 0; - } -done1: - shmem_dir_unmap(dir); -done2: - if (inode->i_mapping->nrpages && (info->flags & SHMEM_PAGEIN)) { + if (mapping->nrpages) { + spin_unlock(&info->lock); /* - * Call truncate_inode_pages again: racing shmem_unuse_inode - * may have swizzled a page in from swap since - * truncate_pagecache or generic_delete_inode did it, before we - * lowered next_index. Also, though shmem_getpage checks - * i_size before adding to cache, no recheck after: so fix the - * narrow window there too. + * A page may have meanwhile sneaked in from swap. */ - truncate_inode_pages_range(inode->i_mapping, start, end); + truncate_inode_pages_range(mapping, lstart, lend); + spin_lock(&info->lock); } - spin_lock(&info->lock); - info->flags &= ~SHMEM_TRUNCATE; - info->swapped -= nr_swaps_freed; - if (nr_pages_to_free) - shmem_free_blocks(inode, nr_pages_to_free); shmem_recalc_inode(inode); spin_unlock(&info->lock); - /* - * Empty swap vector directory pages to be freed? - */ - if (!list_empty(&pages_to_free)) { - pages_to_free.prev->next = NULL; - shmem_free_pages(pages_to_free.next); - } + inode->i_ctime = inode->i_mtime = CURRENT_TIME; } EXPORT_SYMBOL_GPL(shmem_truncate_range); @@ -797,19 +307,6 @@ static int shmem_setattr(struct dentry *dentry, struct iattr *attr) if (page) unlock_page(page); } - /* - * Reset SHMEM_PAGEIN flag so that shmem_truncate can - * detect if any pages might have been added to cache - * after truncate_inode_pages. But we needn't bother - * if it's being fully truncated to zero-length: the - * nrpages check is efficient enough in that case. - */ - if (newsize) { - struct shmem_inode_info *info = SHMEM_I(inode); - spin_lock(&info->lock); - info->flags &= ~SHMEM_PAGEIN; - spin_unlock(&info->lock); - } } if (newsize != oldsize) { i_size_write(inode, newsize); @@ -859,106 +356,28 @@ static void shmem_evict_inode(struct inode *inode) end_writeback(inode); } -static inline int shmem_find_swp(swp_entry_t entry, swp_entry_t *dir, swp_entry_t *edir) -{ - swp_entry_t *ptr; - - for (ptr = dir; ptr < edir; ptr++) { - if (ptr->val == entry.val) - return ptr - dir; - } - return -1; -} - static int shmem_unuse_inode(struct shmem_inode_info *info, swp_entry_t entry, struct page *page) { - struct address_space *mapping; + struct address_space *mapping = info->vfs_inode.i_mapping; unsigned long idx; - unsigned long size; - unsigned long limit; - unsigned long stage; - struct page **dir; - struct page *subdir; - swp_entry_t *ptr; - int offset; int error; - idx = 0; - ptr = info->i_direct; - spin_lock(&info->lock); - if (!info->swapped) { - list_del_init(&info->swaplist); - goto lost2; - } - limit = info->next_index; - size = limit; - if (size > SHMEM_NR_DIRECT) - size = SHMEM_NR_DIRECT; - offset = shmem_find_swp(entry, ptr, ptr+size); - if (offset >= 0) { - shmem_swp_balance_unmap(); - goto found; - } - if (!info->i_indirect) - goto lost2; - - dir = shmem_dir_map(info->i_indirect); - stage = SHMEM_NR_DIRECT + ENTRIES_PER_PAGEPAGE/2; - - for (idx = SHMEM_NR_DIRECT; idx < limit; idx += ENTRIES_PER_PAGE, dir++) { - if (unlikely(idx == stage)) { - shmem_dir_unmap(dir-1); - if (cond_resched_lock(&info->lock)) { - /* check it has not been truncated */ - if (limit > info->next_index) { - limit = info->next_index; - if (idx >= limit) - goto lost2; - } - } - dir = shmem_dir_map(info->i_indirect) + - ENTRIES_PER_PAGE/2 + idx/ENTRIES_PER_PAGEPAGE; - while (!*dir) { - dir++; - idx += ENTRIES_PER_PAGEPAGE; - if (idx >= limit) - goto lost1; - } - stage = idx + ENTRIES_PER_PAGEPAGE; - subdir = *dir; - shmem_dir_unmap(dir); - dir = shmem_dir_map(subdir); - } - subdir = *dir; - if (subdir && page_private(subdir)) { - ptr = shmem_swp_map(subdir); - size = limit - idx; - if (size > ENTRIES_PER_PAGE) - size = ENTRIES_PER_PAGE; - offset = shmem_find_swp(entry, ptr, ptr+size); - shmem_swp_unmap(ptr); - if (offset >= 0) { - shmem_dir_unmap(dir); - ptr = shmem_swp_map(subdir); - goto found; - } - } - } -lost1: - shmem_dir_unmap(dir-1); -lost2: - spin_unlock(&info->lock); + for (idx = 0; idx < SHMEM_NR_DIRECT; idx++) + if (shmem_get_swap(info, idx).val == entry.val) + goto found; return 0; found: - idx += offset; - ptr += offset; + spin_lock(&info->lock); + if (shmem_get_swap(info, idx).val != entry.val) { + spin_unlock(&info->lock); + return 0; + } /* * Move _head_ to start search for next from here. * But be careful: shmem_evict_inode checks list_empty without taking * mutex, and there's an instant in list_move_tail when info->swaplist - * would appear empty, if it were the only one on shmem_swaplist. We - * could avoid doing it if inode NULL; or use this minor optimization. + * would appear empty, if it were the only one on shmem_swaplist. */ if (shmem_swaplist.next != &info->swaplist) list_move_tail(&shmem_swaplist, &info->swaplist); @@ -968,19 +387,17 @@ found: * but also to hold up shmem_evict_inode(): so inode cannot be freed * beneath us (pagelock doesn't help until the page is in pagecache). */ - mapping = info->vfs_inode.i_mapping; error = add_to_page_cache_locked(page, mapping, idx, GFP_NOWAIT); /* which does mem_cgroup_uncharge_cache_page on error */ if (error != -ENOMEM) { delete_from_swap_cache(page); set_page_dirty(page); - info->flags |= SHMEM_PAGEIN; - shmem_swp_set(info, ptr, 0); + shmem_put_swap(info, idx, (swp_entry_t){0}); + info->swapped--; swap_free(entry); error = 1; /* not an error, but entry was found */ } - shmem_swp_unmap(ptr); spin_unlock(&info->lock); return error; } @@ -1017,7 +434,14 @@ int shmem_unuse(swp_entry_t entry, struct page *page) mutex_lock(&shmem_swaplist_mutex); list_for_each_safe(p, next, &shmem_swaplist) { info = list_entry(p, struct shmem_inode_info, swaplist); - found = shmem_unuse_inode(info, entry, page); + if (!info->swapped) { + spin_lock(&info->lock); + if (!info->swapped) + list_del_init(&info->swaplist); + spin_unlock(&info->lock); + } + if (info->swapped) + found = shmem_unuse_inode(info, entry, page); cond_resched(); if (found) break; @@ -1041,7 +465,7 @@ out: static int shmem_writepage(struct page *page, struct writeback_control *wbc) { struct shmem_inode_info *info; - swp_entry_t *entry, swap; + swp_entry_t swap, oswap; struct address_space *mapping; unsigned long index; struct inode *inode; @@ -1067,6 +491,15 @@ static int shmem_writepage(struct page *page, struct writeback_control *wbc) WARN_ON_ONCE(1); /* Still happens? Tell us about it! */ goto redirty; } + + /* + * Just for this patch, we have a toy implementation, + * which can swap out only the first SHMEM_NR_DIRECT pages: + * for simple demonstration of where we need to think about swap. + */ + if (index >= SHMEM_NR_DIRECT) + goto redirty; + swap = get_swap_page(); if (!swap.val) goto redirty; @@ -1087,22 +520,19 @@ static int shmem_writepage(struct page *page, struct writeback_control *wbc) spin_lock(&info->lock); mutex_unlock(&shmem_swaplist_mutex); - if (index >= info->next_index) { - BUG_ON(!(info->flags & SHMEM_TRUNCATE)); - goto unlock; - } - entry = shmem_swp_entry(info, index, NULL); - if (entry->val) { + oswap = shmem_get_swap(info, index); + if (oswap.val) { WARN_ON_ONCE(1); /* Still happens? Tell us about it! */ - free_swap_and_cache(*entry); - shmem_swp_set(info, entry, 0); + free_swap_and_cache(oswap); + shmem_put_swap(info, index, (swp_entry_t){0}); + info->swapped--; } shmem_recalc_inode(inode); if (add_to_swap_cache(page, swap, GFP_ATOMIC) == 0) { delete_from_page_cache(page); - shmem_swp_set(info, entry, swap.val); - shmem_swp_unmap(entry); + shmem_put_swap(info, index, swap); + info->swapped++; swap_shmem_alloc(swap); spin_unlock(&info->lock); BUG_ON(page_mapped(page)); @@ -1110,13 +540,7 @@ static int shmem_writepage(struct page *page, struct writeback_control *wbc) return 0; } - shmem_swp_unmap(entry); -unlock: spin_unlock(&info->lock); - /* - * add_to_swap_cache() doesn't return -EEXIST, so we can safely - * clear SWAP_HAS_CACHE flag. - */ swapcache_free(swap, NULL); redirty: set_page_dirty(page); @@ -1230,12 +654,10 @@ static int shmem_getpage_gfp(struct inode *inode, pgoff_t idx, struct shmem_sb_info *sbinfo; struct page *page; struct page *prealloc_page = NULL; - swp_entry_t *entry; swp_entry_t swap; int error; - int ret; - if (idx >= SHMEM_MAX_INDEX) + if (idx > (MAX_LFS_FILESIZE >> PAGE_CACHE_SHIFT)) return -EFBIG; repeat: page = find_lock_page(mapping, idx); @@ -1272,37 +694,22 @@ repeat: spin_lock(&info->lock); shmem_recalc_inode(inode); - entry = shmem_swp_alloc(info, idx, sgp, gfp); - if (IS_ERR(entry)) { - spin_unlock(&info->lock); - error = PTR_ERR(entry); - goto out; - } - swap = *entry; - + swap = shmem_get_swap(info, idx); if (swap.val) { /* Look it up and read it in.. */ page = lookup_swap_cache(swap); if (!page) { - shmem_swp_unmap(entry); spin_unlock(&info->lock); /* here we actually do the io */ if (fault_type) *fault_type |= VM_FAULT_MAJOR; page = shmem_swapin(swap, gfp, info, idx); if (!page) { - spin_lock(&info->lock); - entry = shmem_swp_alloc(info, idx, sgp, gfp); - if (IS_ERR(entry)) - error = PTR_ERR(entry); - else { - if (entry->val == swap.val) - error = -ENOMEM; - shmem_swp_unmap(entry); - } - spin_unlock(&info->lock); - if (error) + swp_entry_t nswap = shmem_get_swap(info, idx); + if (nswap.val == swap.val) { + error = -ENOMEM; goto out; + } goto repeat; } wait_on_page_locked(page); @@ -1312,14 +719,12 @@ repeat: /* We have to do this with page locked to prevent races */ if (!trylock_page(page)) { - shmem_swp_unmap(entry); spin_unlock(&info->lock); wait_on_page_locked(page); page_cache_release(page); goto repeat; } if (PageWriteback(page)) { - shmem_swp_unmap(entry); spin_unlock(&info->lock); wait_on_page_writeback(page); unlock_page(page); @@ -1327,7 +732,6 @@ repeat: goto repeat; } if (!PageUptodate(page)) { - shmem_swp_unmap(entry); spin_unlock(&info->lock); unlock_page(page); page_cache_release(page); @@ -1338,7 +742,6 @@ repeat: error = add_to_page_cache_locked(page, mapping, idx, GFP_NOWAIT); if (error) { - shmem_swp_unmap(entry); spin_unlock(&info->lock); if (error == -ENOMEM) { /* @@ -1358,16 +761,14 @@ repeat: goto repeat; } - info->flags |= SHMEM_PAGEIN; - shmem_swp_set(info, entry, 0); - shmem_swp_unmap(entry); delete_from_swap_cache(page); + shmem_put_swap(info, idx, (swp_entry_t){0}); + info->swapped--; spin_unlock(&info->lock); set_page_dirty(page); swap_free(swap); } else if (sgp == SGP_READ) { - shmem_swp_unmap(entry); page = find_get_page(mapping, idx); if (page && !trylock_page(page)) { spin_unlock(&info->lock); @@ -1378,7 +779,6 @@ repeat: spin_unlock(&info->lock); } else if (prealloc_page) { - shmem_swp_unmap(entry); sbinfo = SHMEM_SB(inode->i_sb); if (sbinfo->max_blocks) { if (percpu_counter_compare(&sbinfo->used_blocks, @@ -1393,34 +793,24 @@ repeat: page = prealloc_page; prealloc_page = NULL; - entry = shmem_swp_alloc(info, idx, sgp, gfp); - if (IS_ERR(entry)) - error = PTR_ERR(entry); - else { - swap = *entry; - shmem_swp_unmap(entry); - } - ret = error || swap.val; - if (ret) + swap = shmem_get_swap(info, idx); + if (swap.val) mem_cgroup_uncharge_cache_page(page); else - ret = add_to_page_cache_lru(page, mapping, + error = add_to_page_cache_lru(page, mapping, idx, GFP_NOWAIT); /* * At add_to_page_cache_lru() failure, * uncharge will be done automatically. */ - if (ret) { + if (swap.val || error) { shmem_unacct_blocks(info->flags, 1); shmem_free_blocks(inode, 1); spin_unlock(&info->lock); page_cache_release(page); - if (error) - goto out; goto repeat; } - info->flags |= SHMEM_PAGEIN; info->alloced++; spin_unlock(&info->lock); clear_highpage(page); @@ -2627,7 +2017,7 @@ int shmem_fill_super(struct super_block *sb, void *data, int silent) goto failed; sbinfo->free_inodes = sbinfo->max_inodes; - sb->s_maxbytes = SHMEM_MAX_BYTES; + sb->s_maxbytes = MAX_LFS_FILESIZE; sb->s_blocksize = PAGE_CACHE_SIZE; sb->s_blocksize_bits = PAGE_CACHE_SHIFT; sb->s_magic = TMPFS_MAGIC; @@ -2863,7 +2253,7 @@ out4: void mem_cgroup_get_shmem_target(struct inode *inode, pgoff_t pgoff, struct page **pagep, swp_entry_t *ent) { - swp_entry_t entry = { .val = 0 }, *ptr; + swp_entry_t entry = { .val = 0 }; struct page *page = NULL; struct shmem_inode_info *info = SHMEM_I(inode); @@ -2871,16 +2261,13 @@ void mem_cgroup_get_shmem_target(struct inode *inode, pgoff_t pgoff, goto out; spin_lock(&info->lock); - ptr = shmem_swp_entry(info, pgoff, NULL); #ifdef CONFIG_SWAP - if (ptr && ptr->val) { - entry.val = ptr->val; + entry = shmem_get_swap(info, pgoff); + if (entry.val) page = find_get_page(&swapper_space, entry.val); - } else + else #endif page = find_get_page(inode->i_mapping, pgoff); - if (ptr) - shmem_swp_unmap(ptr); spin_unlock(&info->lock); out: *pagep = page; @@ -2963,7 +2350,6 @@ out: #define shmem_get_inode(sb, dir, mode, dev, flags) ramfs_get_inode(sb, dir, mode, dev) #define shmem_acct_size(flags, size) 0 #define shmem_unacct_size(flags, size) do {} while (0) -#define SHMEM_MAX_BYTES MAX_LFS_FILESIZE #endif /* CONFIG_SHMEM */ @@ -2987,7 +2373,7 @@ struct file *shmem_file_setup(const char *name, loff_t size, unsigned long flags if (IS_ERR(shm_mnt)) return (void *)shm_mnt; - if (size < 0 || size > SHMEM_MAX_BYTES) + if (size < 0 || size > MAX_LFS_FILESIZE) return ERR_PTR(-EINVAL); if (shmem_acct_size(flags, size)) From 41ffe5d5ceef7f7ff2ff18e320d88ca6d629efaf Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 3 Aug 2011 16:21:21 -0700 Subject: [PATCH 300/439] tmpfs: miscellaneous trivial cleanups While it's at its least, make a number of boring nitpicky cleanups to shmem.c, mostly for consistency of variable naming. Things like "swap" instead of "entry", "pgoff_t index" instead of "unsigned long idx". And since everything else here is prefixed "shmem_", better change init_tmpfs() to shmem_init(). Signed-off-by: Hugh Dickins Acked-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/shmem_fs.h | 2 +- init/main.c | 2 +- mm/shmem.c | 216 +++++++++++++++++++-------------------- 3 files changed, 109 insertions(+), 111 deletions(-) diff --git a/include/linux/shmem_fs.h b/include/linux/shmem_fs.h index 80b695213fdb..3f05795dcf7b 100644 --- a/include/linux/shmem_fs.h +++ b/include/linux/shmem_fs.h @@ -47,7 +47,7 @@ static inline struct shmem_inode_info *SHMEM_I(struct inode *inode) /* * Functions in mm/shmem.c called directly from elsewhere: */ -extern int init_tmpfs(void); +extern int shmem_init(void); extern int shmem_fill_super(struct super_block *sb, void *data, int silent); extern struct file *shmem_file_setup(const char *name, loff_t size, unsigned long flags); diff --git a/init/main.c b/init/main.c index d7211faed2ad..1952d37e4ecb 100644 --- a/init/main.c +++ b/init/main.c @@ -715,7 +715,7 @@ static void __init do_basic_setup(void) { cpuset_init_smp(); usermodehelper_init(); - init_tmpfs(); + shmem_init(); driver_init(); init_irq_proc(); do_ctors(); diff --git a/mm/shmem.c b/mm/shmem.c index 5574b00ca771..24e95ac16053 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -28,7 +28,6 @@ #include #include #include -#include #include static struct vfsmount *shm_mnt; @@ -51,6 +50,7 @@ static struct vfsmount *shm_mnt; #include #include #include +#include #include #include #include @@ -63,7 +63,6 @@ static struct vfsmount *shm_mnt; #include #include -#include #include #define BLOCKS_PER_PAGE (PAGE_CACHE_SIZE/512) @@ -201,7 +200,7 @@ static void shmem_free_inode(struct super_block *sb) } /** - * shmem_recalc_inode - recalculate the size of an inode + * shmem_recalc_inode - recalculate the block usage of an inode * @inode: inode to recalc * * We have to calculate the free blocks since the mm can drop @@ -356,19 +355,20 @@ static void shmem_evict_inode(struct inode *inode) end_writeback(inode); } -static int shmem_unuse_inode(struct shmem_inode_info *info, swp_entry_t entry, struct page *page) +static int shmem_unuse_inode(struct shmem_inode_info *info, + swp_entry_t swap, struct page *page) { struct address_space *mapping = info->vfs_inode.i_mapping; - unsigned long idx; + pgoff_t index; int error; - for (idx = 0; idx < SHMEM_NR_DIRECT; idx++) - if (shmem_get_swap(info, idx).val == entry.val) + for (index = 0; index < SHMEM_NR_DIRECT; index++) + if (shmem_get_swap(info, index).val == swap.val) goto found; return 0; found: spin_lock(&info->lock); - if (shmem_get_swap(info, idx).val != entry.val) { + if (shmem_get_swap(info, index).val != swap.val) { spin_unlock(&info->lock); return 0; } @@ -387,15 +387,15 @@ found: * but also to hold up shmem_evict_inode(): so inode cannot be freed * beneath us (pagelock doesn't help until the page is in pagecache). */ - error = add_to_page_cache_locked(page, mapping, idx, GFP_NOWAIT); + error = add_to_page_cache_locked(page, mapping, index, GFP_NOWAIT); /* which does mem_cgroup_uncharge_cache_page on error */ if (error != -ENOMEM) { delete_from_swap_cache(page); set_page_dirty(page); - shmem_put_swap(info, idx, (swp_entry_t){0}); + shmem_put_swap(info, index, (swp_entry_t){0}); info->swapped--; - swap_free(entry); + swap_free(swap); error = 1; /* not an error, but entry was found */ } spin_unlock(&info->lock); @@ -405,9 +405,9 @@ found: /* * shmem_unuse() search for an eventually swapped out shmem page. */ -int shmem_unuse(swp_entry_t entry, struct page *page) +int shmem_unuse(swp_entry_t swap, struct page *page) { - struct list_head *p, *next; + struct list_head *this, *next; struct shmem_inode_info *info; int found = 0; int error; @@ -432,8 +432,8 @@ int shmem_unuse(swp_entry_t entry, struct page *page) radix_tree_preload_end(); mutex_lock(&shmem_swaplist_mutex); - list_for_each_safe(p, next, &shmem_swaplist) { - info = list_entry(p, struct shmem_inode_info, swaplist); + list_for_each_safe(this, next, &shmem_swaplist) { + info = list_entry(this, struct shmem_inode_info, swaplist); if (!info->swapped) { spin_lock(&info->lock); if (!info->swapped) @@ -441,7 +441,7 @@ int shmem_unuse(swp_entry_t entry, struct page *page) spin_unlock(&info->lock); } if (info->swapped) - found = shmem_unuse_inode(info, entry, page); + found = shmem_unuse_inode(info, swap, page); cond_resched(); if (found) break; @@ -467,7 +467,7 @@ static int shmem_writepage(struct page *page, struct writeback_control *wbc) struct shmem_inode_info *info; swp_entry_t swap, oswap; struct address_space *mapping; - unsigned long index; + pgoff_t index; struct inode *inode; BUG_ON(!PageLocked(page)); @@ -577,35 +577,33 @@ static struct mempolicy *shmem_get_sbmpol(struct shmem_sb_info *sbinfo) } #endif /* CONFIG_TMPFS */ -static struct page *shmem_swapin(swp_entry_t entry, gfp_t gfp, - struct shmem_inode_info *info, unsigned long idx) +static struct page *shmem_swapin(swp_entry_t swap, gfp_t gfp, + struct shmem_inode_info *info, pgoff_t index) { struct mempolicy mpol, *spol; struct vm_area_struct pvma; - struct page *page; spol = mpol_cond_copy(&mpol, - mpol_shared_policy_lookup(&info->policy, idx)); + mpol_shared_policy_lookup(&info->policy, index)); /* Create a pseudo vma that just contains the policy */ pvma.vm_start = 0; - pvma.vm_pgoff = idx; + pvma.vm_pgoff = index; pvma.vm_ops = NULL; pvma.vm_policy = spol; - page = swapin_readahead(entry, gfp, &pvma, 0); - return page; + return swapin_readahead(swap, gfp, &pvma, 0); } static struct page *shmem_alloc_page(gfp_t gfp, - struct shmem_inode_info *info, unsigned long idx) + struct shmem_inode_info *info, pgoff_t index) { struct vm_area_struct pvma; /* Create a pseudo vma that just contains the policy */ pvma.vm_start = 0; - pvma.vm_pgoff = idx; + pvma.vm_pgoff = index; pvma.vm_ops = NULL; - pvma.vm_policy = mpol_shared_policy_lookup(&info->policy, idx); + pvma.vm_policy = mpol_shared_policy_lookup(&info->policy, index); /* * alloc_page_vma() will drop the shared policy reference @@ -614,19 +612,19 @@ static struct page *shmem_alloc_page(gfp_t gfp, } #else /* !CONFIG_NUMA */ #ifdef CONFIG_TMPFS -static inline void shmem_show_mpol(struct seq_file *seq, struct mempolicy *p) +static inline void shmem_show_mpol(struct seq_file *seq, struct mempolicy *mpol) { } #endif /* CONFIG_TMPFS */ -static inline struct page *shmem_swapin(swp_entry_t entry, gfp_t gfp, - struct shmem_inode_info *info, unsigned long idx) +static inline struct page *shmem_swapin(swp_entry_t swap, gfp_t gfp, + struct shmem_inode_info *info, pgoff_t index) { - return swapin_readahead(entry, gfp, NULL, 0); + return swapin_readahead(swap, gfp, NULL, 0); } static inline struct page *shmem_alloc_page(gfp_t gfp, - struct shmem_inode_info *info, unsigned long idx) + struct shmem_inode_info *info, pgoff_t index) { return alloc_page(gfp); } @@ -646,7 +644,7 @@ static inline struct mempolicy *shmem_get_sbmpol(struct shmem_sb_info *sbinfo) * vm. If we swap it in we mark it dirty since we also free the swap * entry since a page cannot live in both the swap and page cache */ -static int shmem_getpage_gfp(struct inode *inode, pgoff_t idx, +static int shmem_getpage_gfp(struct inode *inode, pgoff_t index, struct page **pagep, enum sgp_type sgp, gfp_t gfp, int *fault_type) { struct address_space *mapping = inode->i_mapping; @@ -657,10 +655,10 @@ static int shmem_getpage_gfp(struct inode *inode, pgoff_t idx, swp_entry_t swap; int error; - if (idx > (MAX_LFS_FILESIZE >> PAGE_CACHE_SHIFT)) + if (index > (MAX_LFS_FILESIZE >> PAGE_CACHE_SHIFT)) return -EFBIG; repeat: - page = find_lock_page(mapping, idx); + page = find_lock_page(mapping, index); if (page) { /* * Once we can get the page lock, it must be uptodate: @@ -681,7 +679,7 @@ repeat: radix_tree_preload_end(); if (sgp != SGP_READ && !prealloc_page) { - prealloc_page = shmem_alloc_page(gfp, info, idx); + prealloc_page = shmem_alloc_page(gfp, info, index); if (prealloc_page) { SetPageSwapBacked(prealloc_page); if (mem_cgroup_cache_charge(prealloc_page, @@ -694,7 +692,7 @@ repeat: spin_lock(&info->lock); shmem_recalc_inode(inode); - swap = shmem_get_swap(info, idx); + swap = shmem_get_swap(info, index); if (swap.val) { /* Look it up and read it in.. */ page = lookup_swap_cache(swap); @@ -703,9 +701,9 @@ repeat: /* here we actually do the io */ if (fault_type) *fault_type |= VM_FAULT_MAJOR; - page = shmem_swapin(swap, gfp, info, idx); + page = shmem_swapin(swap, gfp, info, index); if (!page) { - swp_entry_t nswap = shmem_get_swap(info, idx); + swp_entry_t nswap = shmem_get_swap(info, index); if (nswap.val == swap.val) { error = -ENOMEM; goto out; @@ -740,7 +738,7 @@ repeat: } error = add_to_page_cache_locked(page, mapping, - idx, GFP_NOWAIT); + index, GFP_NOWAIT); if (error) { spin_unlock(&info->lock); if (error == -ENOMEM) { @@ -762,14 +760,14 @@ repeat: } delete_from_swap_cache(page); - shmem_put_swap(info, idx, (swp_entry_t){0}); + shmem_put_swap(info, index, (swp_entry_t){0}); info->swapped--; spin_unlock(&info->lock); set_page_dirty(page); swap_free(swap); } else if (sgp == SGP_READ) { - page = find_get_page(mapping, idx); + page = find_get_page(mapping, index); if (page && !trylock_page(page)) { spin_unlock(&info->lock); wait_on_page_locked(page); @@ -793,12 +791,12 @@ repeat: page = prealloc_page; prealloc_page = NULL; - swap = shmem_get_swap(info, idx); + swap = shmem_get_swap(info, index); if (swap.val) mem_cgroup_uncharge_cache_page(page); else error = add_to_page_cache_lru(page, mapping, - idx, GFP_NOWAIT); + index, GFP_NOWAIT); /* * At add_to_page_cache_lru() failure, * uncharge will be done automatically. @@ -841,7 +839,7 @@ nospace: * but must also avoid reporting a spurious ENOSPC while working on a * full tmpfs. */ - page = find_get_page(mapping, idx); + page = find_get_page(mapping, index); spin_unlock(&info->lock); if (page) { page_cache_release(page); @@ -872,20 +870,20 @@ static int shmem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) } #ifdef CONFIG_NUMA -static int shmem_set_policy(struct vm_area_struct *vma, struct mempolicy *new) +static int shmem_set_policy(struct vm_area_struct *vma, struct mempolicy *mpol) { - struct inode *i = vma->vm_file->f_path.dentry->d_inode; - return mpol_set_shared_policy(&SHMEM_I(i)->policy, vma, new); + struct inode *inode = vma->vm_file->f_path.dentry->d_inode; + return mpol_set_shared_policy(&SHMEM_I(inode)->policy, vma, mpol); } static struct mempolicy *shmem_get_policy(struct vm_area_struct *vma, unsigned long addr) { - struct inode *i = vma->vm_file->f_path.dentry->d_inode; - unsigned long idx; + struct inode *inode = vma->vm_file->f_path.dentry->d_inode; + pgoff_t index; - idx = ((addr - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff; - return mpol_shared_policy_lookup(&SHMEM_I(i)->policy, idx); + index = ((addr - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff; + return mpol_shared_policy_lookup(&SHMEM_I(inode)->policy, index); } #endif @@ -1016,7 +1014,8 @@ static void do_shmem_file_read(struct file *filp, loff_t *ppos, read_descriptor_ { struct inode *inode = filp->f_path.dentry->d_inode; struct address_space *mapping = inode->i_mapping; - unsigned long index, offset; + pgoff_t index; + unsigned long offset; enum sgp_type sgp = SGP_READ; /* @@ -1032,7 +1031,8 @@ static void do_shmem_file_read(struct file *filp, loff_t *ppos, read_descriptor_ for (;;) { struct page *page = NULL; - unsigned long end_index, nr, ret; + pgoff_t end_index; + unsigned long nr, ret; loff_t i_size = i_size_read(inode); end_index = i_size >> PAGE_CACHE_SHIFT; @@ -1270,8 +1270,9 @@ static int shmem_statfs(struct dentry *dentry, struct kstatfs *buf) buf->f_namelen = NAME_MAX; if (sbinfo->max_blocks) { buf->f_blocks = sbinfo->max_blocks; - buf->f_bavail = buf->f_bfree = - sbinfo->max_blocks - percpu_counter_sum(&sbinfo->used_blocks); + buf->f_bavail = + buf->f_bfree = sbinfo->max_blocks - + percpu_counter_sum(&sbinfo->used_blocks); } if (sbinfo->max_inodes) { buf->f_files = sbinfo->max_inodes; @@ -1480,8 +1481,8 @@ static void *shmem_follow_link_inline(struct dentry *dentry, struct nameidata *n static void *shmem_follow_link(struct dentry *dentry, struct nameidata *nd) { struct page *page = NULL; - int res = shmem_getpage(dentry->d_inode, 0, &page, SGP_READ, NULL); - nd_set_link(nd, res ? ERR_PTR(res) : kmap(page)); + int error = shmem_getpage(dentry->d_inode, 0, &page, SGP_READ, NULL); + nd_set_link(nd, error ? ERR_PTR(error) : kmap(page)); if (page) unlock_page(page); return page; @@ -1592,7 +1593,6 @@ out: return err; } - static const struct xattr_handler *shmem_xattr_handlers[] = { #ifdef CONFIG_TMPFS_POSIX_ACL &generic_acl_access_handler, @@ -2052,14 +2052,14 @@ static struct kmem_cache *shmem_inode_cachep; static struct inode *shmem_alloc_inode(struct super_block *sb) { - struct shmem_inode_info *p; - p = (struct shmem_inode_info *)kmem_cache_alloc(shmem_inode_cachep, GFP_KERNEL); - if (!p) + struct shmem_inode_info *info; + info = kmem_cache_alloc(shmem_inode_cachep, GFP_KERNEL); + if (!info) return NULL; - return &p->vfs_inode; + return &info->vfs_inode; } -static void shmem_i_callback(struct rcu_head *head) +static void shmem_destroy_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); INIT_LIST_HEAD(&inode->i_dentry); @@ -2072,25 +2072,24 @@ static void shmem_destroy_inode(struct inode *inode) /* only struct inode is valid if it's an inline symlink */ mpol_free_shared_policy(&SHMEM_I(inode)->policy); } - call_rcu(&inode->i_rcu, shmem_i_callback); + call_rcu(&inode->i_rcu, shmem_destroy_callback); } -static void init_once(void *foo) +static void shmem_init_inode(void *foo) { - struct shmem_inode_info *p = (struct shmem_inode_info *) foo; - - inode_init_once(&p->vfs_inode); + struct shmem_inode_info *info = foo; + inode_init_once(&info->vfs_inode); } -static int init_inodecache(void) +static int shmem_init_inodecache(void) { shmem_inode_cachep = kmem_cache_create("shmem_inode_cache", sizeof(struct shmem_inode_info), - 0, SLAB_PANIC, init_once); + 0, SLAB_PANIC, shmem_init_inode); return 0; } -static void destroy_inodecache(void) +static void shmem_destroy_inodecache(void) { kmem_cache_destroy(shmem_inode_cachep); } @@ -2187,21 +2186,20 @@ static const struct vm_operations_struct shmem_vm_ops = { #endif }; - static struct dentry *shmem_mount(struct file_system_type *fs_type, int flags, const char *dev_name, void *data) { return mount_nodev(fs_type, flags, data, shmem_fill_super); } -static struct file_system_type tmpfs_fs_type = { +static struct file_system_type shmem_fs_type = { .owner = THIS_MODULE, .name = "tmpfs", .mount = shmem_mount, .kill_sb = kill_litter_super, }; -int __init init_tmpfs(void) +int __init shmem_init(void) { int error; @@ -2209,18 +2207,18 @@ int __init init_tmpfs(void) if (error) goto out4; - error = init_inodecache(); + error = shmem_init_inodecache(); if (error) goto out3; - error = register_filesystem(&tmpfs_fs_type); + error = register_filesystem(&shmem_fs_type); if (error) { printk(KERN_ERR "Could not register tmpfs\n"); goto out2; } - shm_mnt = vfs_kern_mount(&tmpfs_fs_type, MS_NOUSER, - tmpfs_fs_type.name, NULL); + shm_mnt = vfs_kern_mount(&shmem_fs_type, MS_NOUSER, + shmem_fs_type.name, NULL); if (IS_ERR(shm_mnt)) { error = PTR_ERR(shm_mnt); printk(KERN_ERR "Could not kern_mount tmpfs\n"); @@ -2229,9 +2227,9 @@ int __init init_tmpfs(void) return 0; out1: - unregister_filesystem(&tmpfs_fs_type); + unregister_filesystem(&shmem_fs_type); out2: - destroy_inodecache(); + shmem_destroy_inodecache(); out3: bdi_destroy(&shmem_backing_dev_info); out4: @@ -2241,37 +2239,37 @@ out4: #ifdef CONFIG_CGROUP_MEM_RES_CTLR /** - * mem_cgroup_get_shmem_target - find a page or entry assigned to the shmem file + * mem_cgroup_get_shmem_target - find page or swap assigned to the shmem file * @inode: the inode to be searched - * @pgoff: the offset to be searched + * @index: the page offset to be searched * @pagep: the pointer for the found page to be stored - * @ent: the pointer for the found swap entry to be stored + * @swapp: the pointer for the found swap entry to be stored * * If a page is found, refcount of it is incremented. Callers should handle * these refcount. */ -void mem_cgroup_get_shmem_target(struct inode *inode, pgoff_t pgoff, - struct page **pagep, swp_entry_t *ent) +void mem_cgroup_get_shmem_target(struct inode *inode, pgoff_t index, + struct page **pagep, swp_entry_t *swapp) { - swp_entry_t entry = { .val = 0 }; - struct page *page = NULL; struct shmem_inode_info *info = SHMEM_I(inode); + struct page *page = NULL; + swp_entry_t swap = {0}; - if ((pgoff << PAGE_CACHE_SHIFT) >= i_size_read(inode)) + if ((index << PAGE_CACHE_SHIFT) >= i_size_read(inode)) goto out; spin_lock(&info->lock); #ifdef CONFIG_SWAP - entry = shmem_get_swap(info, pgoff); - if (entry.val) - page = find_get_page(&swapper_space, entry.val); + swap = shmem_get_swap(info, index); + if (swap.val) + page = find_get_page(&swapper_space, swap.val); else #endif - page = find_get_page(inode->i_mapping, pgoff); + page = find_get_page(inode->i_mapping, index); spin_unlock(&info->lock); out: *pagep = page; - *ent = entry; + *swapp = swap; } #endif @@ -2288,23 +2286,23 @@ out: #include -static struct file_system_type tmpfs_fs_type = { +static struct file_system_type shmem_fs_type = { .name = "tmpfs", .mount = ramfs_mount, .kill_sb = kill_litter_super, }; -int __init init_tmpfs(void) +int __init shmem_init(void) { - BUG_ON(register_filesystem(&tmpfs_fs_type) != 0); + BUG_ON(register_filesystem(&shmem_fs_type) != 0); - shm_mnt = kern_mount(&tmpfs_fs_type); + shm_mnt = kern_mount(&shmem_fs_type); BUG_ON(IS_ERR(shm_mnt)); return 0; } -int shmem_unuse(swp_entry_t entry, struct page *page) +int shmem_unuse(swp_entry_t swap, struct page *page) { return 0; } @@ -2314,34 +2312,34 @@ int shmem_lock(struct file *file, int lock, struct user_struct *user) return 0; } -void shmem_truncate_range(struct inode *inode, loff_t start, loff_t end) +void shmem_truncate_range(struct inode *inode, loff_t lstart, loff_t lend) { - truncate_inode_pages_range(inode->i_mapping, start, end); + truncate_inode_pages_range(inode->i_mapping, lstart, lend); } EXPORT_SYMBOL_GPL(shmem_truncate_range); #ifdef CONFIG_CGROUP_MEM_RES_CTLR /** - * mem_cgroup_get_shmem_target - find a page or entry assigned to the shmem file + * mem_cgroup_get_shmem_target - find page or swap assigned to the shmem file * @inode: the inode to be searched - * @pgoff: the offset to be searched + * @index: the page offset to be searched * @pagep: the pointer for the found page to be stored - * @ent: the pointer for the found swap entry to be stored + * @swapp: the pointer for the found swap entry to be stored * * If a page is found, refcount of it is incremented. Callers should handle * these refcount. */ -void mem_cgroup_get_shmem_target(struct inode *inode, pgoff_t pgoff, - struct page **pagep, swp_entry_t *ent) +void mem_cgroup_get_shmem_target(struct inode *inode, pgoff_t index, + struct page **pagep, swp_entry_t *swapp) { struct page *page = NULL; - if ((pgoff << PAGE_CACHE_SHIFT) >= i_size_read(inode)) + if ((index << PAGE_CACHE_SHIFT) >= i_size_read(inode)) goto out; - page = find_get_page(inode->i_mapping, pgoff); + page = find_get_page(inode->i_mapping, index); out: *pagep = page; - *ent = (swp_entry_t){ .val = 0 }; + *swapp = (swp_entry_t){0}; } #endif From bda97eab0cc9c6385b9f26abdda6459f630f4513 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 3 Aug 2011 16:21:21 -0700 Subject: [PATCH 301/439] tmpfs: copy truncate_inode_pages_range Bring truncate.c's code for truncate_inode_pages_range() inline into shmem_truncate_range(), replacing its first call (there's a followup call below, but leave that one, it will disappear next). Don't play with it yet, apart from leaving out the cleancache flush, and (importantly) the nrpages == 0 skip, and moving shmem_setattr()'s partial page preparation into its partial page handling. Signed-off-by: Hugh Dickins Acked-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/shmem.c | 99 +++++++++++++++++++++++++++++++++++++++++++----------- 1 file changed, 79 insertions(+), 20 deletions(-) diff --git a/mm/shmem.c b/mm/shmem.c index 24e95ac16053..e101c211ed1f 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -50,6 +50,7 @@ static struct vfsmount *shm_mnt; #include #include #include +#include #include #include #include @@ -242,11 +243,88 @@ void shmem_truncate_range(struct inode *inode, loff_t lstart, loff_t lend) struct address_space *mapping = inode->i_mapping; struct shmem_inode_info *info = SHMEM_I(inode); pgoff_t start = (lstart + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; + unsigned partial = lstart & (PAGE_CACHE_SIZE - 1); pgoff_t end = (lend >> PAGE_CACHE_SHIFT); + struct pagevec pvec; pgoff_t index; swp_entry_t swap; + int i; - truncate_inode_pages_range(mapping, lstart, lend); + BUG_ON((lend & (PAGE_CACHE_SIZE - 1)) != (PAGE_CACHE_SIZE - 1)); + + pagevec_init(&pvec, 0); + index = start; + while (index <= end && pagevec_lookup(&pvec, mapping, index, + min(end - index, (pgoff_t)PAGEVEC_SIZE - 1) + 1)) { + mem_cgroup_uncharge_start(); + for (i = 0; i < pagevec_count(&pvec); i++) { + struct page *page = pvec.pages[i]; + + /* We rely upon deletion not changing page->index */ + index = page->index; + if (index > end) + break; + + if (!trylock_page(page)) + continue; + WARN_ON(page->index != index); + if (PageWriteback(page)) { + unlock_page(page); + continue; + } + truncate_inode_page(mapping, page); + unlock_page(page); + } + pagevec_release(&pvec); + mem_cgroup_uncharge_end(); + cond_resched(); + index++; + } + + if (partial) { + struct page *page = NULL; + shmem_getpage(inode, start - 1, &page, SGP_READ, NULL); + if (page) { + zero_user_segment(page, partial, PAGE_CACHE_SIZE); + set_page_dirty(page); + unlock_page(page); + page_cache_release(page); + } + } + + index = start; + for ( ; ; ) { + cond_resched(); + if (!pagevec_lookup(&pvec, mapping, index, + min(end - index, (pgoff_t)PAGEVEC_SIZE - 1) + 1)) { + if (index == start) + break; + index = start; + continue; + } + if (index == start && pvec.pages[0]->index > end) { + pagevec_release(&pvec); + break; + } + mem_cgroup_uncharge_start(); + for (i = 0; i < pagevec_count(&pvec); i++) { + struct page *page = pvec.pages[i]; + + /* We rely upon deletion not changing page->index */ + index = page->index; + if (index > end) + break; + + lock_page(page); + WARN_ON(page->index != index); + wait_on_page_writeback(page); + truncate_inode_page(mapping, page); + unlock_page(page); + } + pagevec_release(&pvec); + mem_cgroup_uncharge_end(); + index++; + } if (end > SHMEM_NR_DIRECT) end = SHMEM_NR_DIRECT; @@ -289,24 +367,7 @@ static int shmem_setattr(struct dentry *dentry, struct iattr *attr) if (S_ISREG(inode->i_mode) && (attr->ia_valid & ATTR_SIZE)) { loff_t oldsize = inode->i_size; loff_t newsize = attr->ia_size; - struct page *page = NULL; - if (newsize < oldsize) { - /* - * If truncating down to a partial page, then - * if that page is already allocated, hold it - * in memory until the truncation is over, so - * truncate_partial_page cannot miss it were - * it assigned to swap. - */ - if (newsize & (PAGE_CACHE_SIZE-1)) { - (void) shmem_getpage(inode, - newsize >> PAGE_CACHE_SHIFT, - &page, SGP_READ, NULL); - if (page) - unlock_page(page); - } - } if (newsize != oldsize) { i_size_write(inode, newsize); inode->i_ctime = inode->i_mtime = CURRENT_TIME; @@ -318,8 +379,6 @@ static int shmem_setattr(struct dentry *dentry, struct iattr *attr) /* unmap again to remove racily COWed private pages */ unmap_mapping_range(inode->i_mapping, holebegin, 0, 1); } - if (page) - page_cache_release(page); } setattr_copy(inode, attr); From 7a5d0fbb29936fad7f17b1cb001b0c33a5f13328 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 3 Aug 2011 16:21:22 -0700 Subject: [PATCH 302/439] tmpfs: convert shmem_truncate_range to radix-swap Disable the toy swapping implementation in shmem_writepage() - it's hard to support two schemes at once - and convert shmem_truncate_range() to a lockless gang lookup of swap entries along with pages, freeing both. Since the second loop tightens its noose until all entries of either kind have been squeezed out (and we shall make sure that there's not an instant when neither is visible), there is no longer a need for yet another pass below. shmem_radix_tree_replace() compensates for the lockless lookup by checking that the expected entry is in place, under lock, before replacing it. Here it just deletes, but will be used in later patches to substitute swap entry for page or page for swap entry. Signed-off-by: Hugh Dickins Acked-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/shmem.c | 194 ++++++++++++++++++++++++++++++++++++++++------------- 1 file changed, 147 insertions(+), 47 deletions(-) diff --git a/mm/shmem.c b/mm/shmem.c index e101c211ed1f..4439b7d55819 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -238,6 +238,111 @@ static swp_entry_t shmem_get_swap(struct shmem_inode_info *info, pgoff_t index) info->i_direct[index] : (swp_entry_t){0}; } +/* + * Replace item expected in radix tree by a new item, while holding tree lock. + */ +static int shmem_radix_tree_replace(struct address_space *mapping, + pgoff_t index, void *expected, void *replacement) +{ + void **pslot; + void *item = NULL; + + VM_BUG_ON(!expected); + pslot = radix_tree_lookup_slot(&mapping->page_tree, index); + if (pslot) + item = radix_tree_deref_slot_protected(pslot, + &mapping->tree_lock); + if (item != expected) + return -ENOENT; + if (replacement) + radix_tree_replace_slot(pslot, replacement); + else + radix_tree_delete(&mapping->page_tree, index); + return 0; +} + +/* + * Like find_get_pages, but collecting swap entries as well as pages. + */ +static unsigned shmem_find_get_pages_and_swap(struct address_space *mapping, + pgoff_t start, unsigned int nr_pages, + struct page **pages, pgoff_t *indices) +{ + unsigned int i; + unsigned int ret; + unsigned int nr_found; + + rcu_read_lock(); +restart: + nr_found = radix_tree_gang_lookup_slot(&mapping->page_tree, + (void ***)pages, indices, start, nr_pages); + ret = 0; + for (i = 0; i < nr_found; i++) { + struct page *page; +repeat: + page = radix_tree_deref_slot((void **)pages[i]); + if (unlikely(!page)) + continue; + if (radix_tree_exception(page)) { + if (radix_tree_exceptional_entry(page)) + goto export; + /* radix_tree_deref_retry(page) */ + goto restart; + } + if (!page_cache_get_speculative(page)) + goto repeat; + + /* Has the page moved? */ + if (unlikely(page != *((void **)pages[i]))) { + page_cache_release(page); + goto repeat; + } +export: + indices[ret] = indices[i]; + pages[ret] = page; + ret++; + } + if (unlikely(!ret && nr_found)) + goto restart; + rcu_read_unlock(); + return ret; +} + +/* + * Remove swap entry from radix tree, free the swap and its page cache. + */ +static int shmem_free_swap(struct address_space *mapping, + pgoff_t index, void *radswap) +{ + int error; + + spin_lock_irq(&mapping->tree_lock); + error = shmem_radix_tree_replace(mapping, index, radswap, NULL); + spin_unlock_irq(&mapping->tree_lock); + if (!error) + free_swap_and_cache(radix_to_swp_entry(radswap)); + return error; +} + +/* + * Pagevec may contain swap entries, so shuffle up pages before releasing. + */ +static void shmem_pagevec_release(struct pagevec *pvec) +{ + int i, j; + + for (i = 0, j = 0; i < pagevec_count(pvec); i++) { + struct page *page = pvec->pages[i]; + if (!radix_tree_exceptional_entry(page)) + pvec->pages[j++] = page; + } + pvec->nr = j; + pagevec_release(pvec); +} + +/* + * Remove range of pages and swap entries from radix tree, and free them. + */ void shmem_truncate_range(struct inode *inode, loff_t lstart, loff_t lend) { struct address_space *mapping = inode->i_mapping; @@ -246,36 +351,44 @@ void shmem_truncate_range(struct inode *inode, loff_t lstart, loff_t lend) unsigned partial = lstart & (PAGE_CACHE_SIZE - 1); pgoff_t end = (lend >> PAGE_CACHE_SHIFT); struct pagevec pvec; + pgoff_t indices[PAGEVEC_SIZE]; + long nr_swaps_freed = 0; pgoff_t index; - swp_entry_t swap; int i; BUG_ON((lend & (PAGE_CACHE_SIZE - 1)) != (PAGE_CACHE_SIZE - 1)); pagevec_init(&pvec, 0); index = start; - while (index <= end && pagevec_lookup(&pvec, mapping, index, - min(end - index, (pgoff_t)PAGEVEC_SIZE - 1) + 1)) { + while (index <= end) { + pvec.nr = shmem_find_get_pages_and_swap(mapping, index, + min(end - index, (pgoff_t)PAGEVEC_SIZE - 1) + 1, + pvec.pages, indices); + if (!pvec.nr) + break; mem_cgroup_uncharge_start(); for (i = 0; i < pagevec_count(&pvec); i++) { struct page *page = pvec.pages[i]; - /* We rely upon deletion not changing page->index */ - index = page->index; + index = indices[i]; if (index > end) break; - if (!trylock_page(page)) - continue; - WARN_ON(page->index != index); - if (PageWriteback(page)) { - unlock_page(page); + if (radix_tree_exceptional_entry(page)) { + nr_swaps_freed += !shmem_free_swap(mapping, + index, page); continue; } - truncate_inode_page(mapping, page); + + if (!trylock_page(page)) + continue; + if (page->mapping == mapping) { + VM_BUG_ON(PageWriteback(page)); + truncate_inode_page(mapping, page); + } unlock_page(page); } - pagevec_release(&pvec); + shmem_pagevec_release(&pvec); mem_cgroup_uncharge_end(); cond_resched(); index++; @@ -295,59 +408,47 @@ void shmem_truncate_range(struct inode *inode, loff_t lstart, loff_t lend) index = start; for ( ; ; ) { cond_resched(); - if (!pagevec_lookup(&pvec, mapping, index, - min(end - index, (pgoff_t)PAGEVEC_SIZE - 1) + 1)) { + pvec.nr = shmem_find_get_pages_and_swap(mapping, index, + min(end - index, (pgoff_t)PAGEVEC_SIZE - 1) + 1, + pvec.pages, indices); + if (!pvec.nr) { if (index == start) break; index = start; continue; } - if (index == start && pvec.pages[0]->index > end) { - pagevec_release(&pvec); + if (index == start && indices[0] > end) { + shmem_pagevec_release(&pvec); break; } mem_cgroup_uncharge_start(); for (i = 0; i < pagevec_count(&pvec); i++) { struct page *page = pvec.pages[i]; - /* We rely upon deletion not changing page->index */ - index = page->index; + index = indices[i]; if (index > end) break; + if (radix_tree_exceptional_entry(page)) { + nr_swaps_freed += !shmem_free_swap(mapping, + index, page); + continue; + } + lock_page(page); - WARN_ON(page->index != index); - wait_on_page_writeback(page); - truncate_inode_page(mapping, page); + if (page->mapping == mapping) { + VM_BUG_ON(PageWriteback(page)); + truncate_inode_page(mapping, page); + } unlock_page(page); } - pagevec_release(&pvec); + shmem_pagevec_release(&pvec); mem_cgroup_uncharge_end(); index++; } - if (end > SHMEM_NR_DIRECT) - end = SHMEM_NR_DIRECT; - spin_lock(&info->lock); - for (index = start; index < end; index++) { - swap = shmem_get_swap(info, index); - if (swap.val) { - free_swap_and_cache(swap); - shmem_put_swap(info, index, (swp_entry_t){0}); - info->swapped--; - } - } - - if (mapping->nrpages) { - spin_unlock(&info->lock); - /* - * A page may have meanwhile sneaked in from swap. - */ - truncate_inode_pages_range(mapping, lstart, lend); - spin_lock(&info->lock); - } - + info->swapped -= nr_swaps_freed; shmem_recalc_inode(inode); spin_unlock(&info->lock); @@ -552,11 +653,10 @@ static int shmem_writepage(struct page *page, struct writeback_control *wbc) } /* - * Just for this patch, we have a toy implementation, - * which can swap out only the first SHMEM_NR_DIRECT pages: - * for simple demonstration of where we need to think about swap. + * Disable even the toy swapping implementation, while we convert + * functions one by one to having swap entries in the radix tree. */ - if (index >= SHMEM_NR_DIRECT) + if (index < ULONG_MAX) goto redirty; swap = get_swap_page(); From 46f65ec15c6878a2b4a49f6e01b20b201b46a9e4 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 3 Aug 2011 16:21:23 -0700 Subject: [PATCH 303/439] tmpfs: convert shmem_unuse_inode to radix-swap Convert shmem_unuse_inode() to use a lockless gang lookup of the radix tree, searching for matching swap. This is somewhat slower than the old method: because of repeated radix tree descents, because of copying entries up, but probably most because the old method noted and skipped once a vector page was cleared of swap. Perhaps we can devise a use of radix tree tagging to achieve that later. shmem_add_to_page_cache() uses shmem_radix_tree_replace() to compensate for the lockless lookup by checking that the expected entry is in place, under lock. It is not very satisfactory to be copying this much from add_to_page_cache_locked(), but I think easier to sell than insisting that every caller of add_to_page_cache*() go through the extras. Signed-off-by: Hugh Dickins Acked-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/shmem.c | 133 ++++++++++++++++++++++++++++++++++++++++++----------- 1 file changed, 107 insertions(+), 26 deletions(-) diff --git a/mm/shmem.c b/mm/shmem.c index 4439b7d55819..174f97188e8a 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -261,6 +261,55 @@ static int shmem_radix_tree_replace(struct address_space *mapping, return 0; } +/* + * Like add_to_page_cache_locked, but error if expected item has gone. + */ +static int shmem_add_to_page_cache(struct page *page, + struct address_space *mapping, + pgoff_t index, gfp_t gfp, void *expected) +{ + int error; + + VM_BUG_ON(!PageLocked(page)); + VM_BUG_ON(!PageSwapBacked(page)); + + error = mem_cgroup_cache_charge(page, current->mm, + gfp & GFP_RECLAIM_MASK); + if (error) + goto out; + if (!expected) + error = radix_tree_preload(gfp & GFP_RECLAIM_MASK); + if (!error) { + page_cache_get(page); + page->mapping = mapping; + page->index = index; + + spin_lock_irq(&mapping->tree_lock); + if (!expected) + error = radix_tree_insert(&mapping->page_tree, + index, page); + else + error = shmem_radix_tree_replace(mapping, index, + expected, page); + if (!error) { + mapping->nrpages++; + __inc_zone_page_state(page, NR_FILE_PAGES); + __inc_zone_page_state(page, NR_SHMEM); + spin_unlock_irq(&mapping->tree_lock); + } else { + page->mapping = NULL; + spin_unlock_irq(&mapping->tree_lock); + page_cache_release(page); + } + if (!expected) + radix_tree_preload_end(); + } + if (error) + mem_cgroup_uncharge_cache_page(page); +out: + return error; +} + /* * Like find_get_pages, but collecting swap entries as well as pages. */ @@ -308,6 +357,42 @@ export: return ret; } +/* + * Lockless lookup of swap entry in radix tree, avoiding refcount on pages. + */ +static pgoff_t shmem_find_swap(struct address_space *mapping, void *radswap) +{ + void **slots[PAGEVEC_SIZE]; + pgoff_t indices[PAGEVEC_SIZE]; + unsigned int nr_found; + +restart: + nr_found = 1; + indices[0] = -1; + while (nr_found) { + pgoff_t index = indices[nr_found - 1] + 1; + unsigned int i; + + rcu_read_lock(); + nr_found = radix_tree_gang_lookup_slot(&mapping->page_tree, + slots, indices, index, PAGEVEC_SIZE); + for (i = 0; i < nr_found; i++) { + void *item = radix_tree_deref_slot(slots[i]); + if (radix_tree_deref_retry(item)) { + rcu_read_unlock(); + goto restart; + } + if (item == radswap) { + rcu_read_unlock(); + return indices[i]; + } + } + rcu_read_unlock(); + cond_resched(); + } + return -1; +} + /* * Remove swap entry from radix tree, free the swap and its page cache. */ @@ -515,23 +600,21 @@ static void shmem_evict_inode(struct inode *inode) end_writeback(inode); } +/* + * If swap found in inode, free it and move page from swapcache to filecache. + */ static int shmem_unuse_inode(struct shmem_inode_info *info, swp_entry_t swap, struct page *page) { struct address_space *mapping = info->vfs_inode.i_mapping; + void *radswap; pgoff_t index; int error; - for (index = 0; index < SHMEM_NR_DIRECT; index++) - if (shmem_get_swap(info, index).val == swap.val) - goto found; - return 0; -found: - spin_lock(&info->lock); - if (shmem_get_swap(info, index).val != swap.val) { - spin_unlock(&info->lock); + radswap = swp_to_radix_entry(swap); + index = shmem_find_swap(mapping, radswap); + if (index == -1) return 0; - } /* * Move _head_ to start search for next from here. @@ -547,23 +630,30 @@ found: * but also to hold up shmem_evict_inode(): so inode cannot be freed * beneath us (pagelock doesn't help until the page is in pagecache). */ - error = add_to_page_cache_locked(page, mapping, index, GFP_NOWAIT); + error = shmem_add_to_page_cache(page, mapping, index, + GFP_NOWAIT, radswap); /* which does mem_cgroup_uncharge_cache_page on error */ if (error != -ENOMEM) { + /* + * Truncation and eviction use free_swap_and_cache(), which + * only does trylock page: if we raced, best clean up here. + */ delete_from_swap_cache(page); set_page_dirty(page); - shmem_put_swap(info, index, (swp_entry_t){0}); - info->swapped--; - swap_free(swap); + if (!error) { + spin_lock(&info->lock); + info->swapped--; + spin_unlock(&info->lock); + swap_free(swap); + } error = 1; /* not an error, but entry was found */ } - spin_unlock(&info->lock); return error; } /* - * shmem_unuse() search for an eventually swapped out shmem page. + * Search through swapped inodes to find and replace swap by page. */ int shmem_unuse(swp_entry_t swap, struct page *page) { @@ -576,20 +666,12 @@ int shmem_unuse(swp_entry_t swap, struct page *page) * Charge page using GFP_KERNEL while we can wait, before taking * the shmem_swaplist_mutex which might hold up shmem_writepage(). * Charged back to the user (not to caller) when swap account is used. - * add_to_page_cache() will be called with GFP_NOWAIT. + * shmem_add_to_page_cache() will be called with GFP_NOWAIT. */ error = mem_cgroup_cache_charge(page, current->mm, GFP_KERNEL); if (error) goto out; - /* - * Try to preload while we can wait, to not make a habit of - * draining atomic reserves; but don't latch on to this cpu, - * it's okay if sometimes we get rescheduled after this. - */ - error = radix_tree_preload(GFP_KERNEL); - if (error) - goto uncharge; - radix_tree_preload_end(); + /* No radix_tree_preload: swap entry keeps a place for page in tree */ mutex_lock(&shmem_swaplist_mutex); list_for_each_safe(this, next, &shmem_swaplist) { @@ -608,7 +690,6 @@ int shmem_unuse(swp_entry_t swap, struct page *page) } mutex_unlock(&shmem_swaplist_mutex); -uncharge: if (!found) mem_cgroup_uncharge_cache_page(page); if (found < 0) From 54af60421822bb9cb664dd5cd7aac46c01ccfcf8 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 3 Aug 2011 16:21:24 -0700 Subject: [PATCH 304/439] tmpfs: convert shmem_getpage_gfp to radix-swap Convert shmem_getpage_gfp(), the engine-room of shmem, to expect page or swap entry returned from radix tree by find_lock_page(). Whereas the repetitive old method proceeded mainly under info->lock, dropping and repeating whenever one of the conditions needed was not met, now we can proceed without it, leaving shmem_add_to_page_cache() to check for a race. This way there is no need to preallocate a page, no need for an early radix_tree_preload(), no need for mem_cgroup_shmem_charge_fallback(). Move the error unwinding down to the bottom instead of repeating it throughout. ENOSPC handling is a little different from before: there is no longer any race between find_lock_page() and finding swap, but we can arrive at ENOSPC before calling shmem_recalc_inode(), which might occasionally discover freed space. Be stricter to check i_size before returning. info->lock is used for little but alloced, swapped, i_blocks updates. Move i_blocks updates out from under the max_blocks check, so even an unlimited size=0 mount can show accurate du. Signed-off-by: Hugh Dickins Acked-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/shmem.c | 263 +++++++++++++++++++++++------------------------------ 1 file changed, 114 insertions(+), 149 deletions(-) diff --git a/mm/shmem.c b/mm/shmem.c index 174f97188e8a..92f01d7cc150 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -166,15 +166,6 @@ static struct backing_dev_info shmem_backing_dev_info __read_mostly = { static LIST_HEAD(shmem_swaplist); static DEFINE_MUTEX(shmem_swaplist_mutex); -static void shmem_free_blocks(struct inode *inode, long pages) -{ - struct shmem_sb_info *sbinfo = SHMEM_SB(inode->i_sb); - if (sbinfo->max_blocks) { - percpu_counter_add(&sbinfo->used_blocks, -pages); - inode->i_blocks -= pages*BLOCKS_PER_PAGE; - } -} - static int shmem_reserve_inode(struct super_block *sb) { struct shmem_sb_info *sbinfo = SHMEM_SB(sb); @@ -219,9 +210,12 @@ static void shmem_recalc_inode(struct inode *inode) freed = info->alloced - info->swapped - inode->i_mapping->nrpages; if (freed > 0) { + struct shmem_sb_info *sbinfo = SHMEM_SB(inode->i_sb); + if (sbinfo->max_blocks) + percpu_counter_add(&sbinfo->used_blocks, -freed); info->alloced -= freed; + inode->i_blocks -= freed * BLOCKS_PER_PAGE; shmem_unacct_blocks(info->flags, freed); - shmem_free_blocks(inode, freed); } } @@ -888,205 +882,180 @@ static int shmem_getpage_gfp(struct inode *inode, pgoff_t index, struct page **pagep, enum sgp_type sgp, gfp_t gfp, int *fault_type) { struct address_space *mapping = inode->i_mapping; - struct shmem_inode_info *info = SHMEM_I(inode); + struct shmem_inode_info *info; struct shmem_sb_info *sbinfo; struct page *page; - struct page *prealloc_page = NULL; swp_entry_t swap; int error; + int once = 0; if (index > (MAX_LFS_FILESIZE >> PAGE_CACHE_SHIFT)) return -EFBIG; repeat: + swap.val = 0; page = find_lock_page(mapping, index); - if (page) { + if (radix_tree_exceptional_entry(page)) { + swap = radix_to_swp_entry(page); + page = NULL; + } + + if (sgp != SGP_WRITE && + ((loff_t)index << PAGE_CACHE_SHIFT) >= i_size_read(inode)) { + error = -EINVAL; + goto failed; + } + + if (page || (sgp == SGP_READ && !swap.val)) { /* * Once we can get the page lock, it must be uptodate: * if there were an error in reading back from swap, * the page would not be inserted into the filecache. */ - BUG_ON(!PageUptodate(page)); - goto done; + BUG_ON(page && !PageUptodate(page)); + *pagep = page; + return 0; } /* - * Try to preload while we can wait, to not make a habit of - * draining atomic reserves; but don't latch on to this cpu. + * Fast cache lookup did not find it: + * bring it back from swap or allocate. */ - error = radix_tree_preload(gfp & GFP_RECLAIM_MASK); - if (error) - goto out; - radix_tree_preload_end(); + info = SHMEM_I(inode); + sbinfo = SHMEM_SB(inode->i_sb); - if (sgp != SGP_READ && !prealloc_page) { - prealloc_page = shmem_alloc_page(gfp, info, index); - if (prealloc_page) { - SetPageSwapBacked(prealloc_page); - if (mem_cgroup_cache_charge(prealloc_page, - current->mm, GFP_KERNEL)) { - page_cache_release(prealloc_page); - prealloc_page = NULL; - } - } - } - - spin_lock(&info->lock); - shmem_recalc_inode(inode); - swap = shmem_get_swap(info, index); if (swap.val) { /* Look it up and read it in.. */ page = lookup_swap_cache(swap); if (!page) { - spin_unlock(&info->lock); /* here we actually do the io */ if (fault_type) *fault_type |= VM_FAULT_MAJOR; page = shmem_swapin(swap, gfp, info, index); if (!page) { - swp_entry_t nswap = shmem_get_swap(info, index); - if (nswap.val == swap.val) { - error = -ENOMEM; - goto out; - } - goto repeat; + error = -ENOMEM; + goto failed; } - wait_on_page_locked(page); - page_cache_release(page); - goto repeat; } /* We have to do this with page locked to prevent races */ - if (!trylock_page(page)) { - spin_unlock(&info->lock); - wait_on_page_locked(page); - page_cache_release(page); - goto repeat; - } - if (PageWriteback(page)) { - spin_unlock(&info->lock); - wait_on_page_writeback(page); - unlock_page(page); - page_cache_release(page); - goto repeat; - } + lock_page(page); if (!PageUptodate(page)) { - spin_unlock(&info->lock); - unlock_page(page); - page_cache_release(page); error = -EIO; - goto out; + goto failed; + } + wait_on_page_writeback(page); + + /* Someone may have already done it for us */ + if (page->mapping) { + if (page->mapping == mapping && + page->index == index) + goto done; + error = -EEXIST; + goto failed; } - error = add_to_page_cache_locked(page, mapping, - index, GFP_NOWAIT); - if (error) { - spin_unlock(&info->lock); - if (error == -ENOMEM) { - /* - * reclaim from proper memory cgroup and - * call memcg's OOM if needed. - */ - error = mem_cgroup_shmem_charge_fallback( - page, current->mm, gfp); - if (error) { - unlock_page(page); - page_cache_release(page); - goto out; - } - } - unlock_page(page); - page_cache_release(page); - goto repeat; - } + error = shmem_add_to_page_cache(page, mapping, index, + gfp, swp_to_radix_entry(swap)); + if (error) + goto failed; + + spin_lock(&info->lock); + info->swapped--; + shmem_recalc_inode(inode); + spin_unlock(&info->lock); delete_from_swap_cache(page); - shmem_put_swap(info, index, (swp_entry_t){0}); - info->swapped--; - spin_unlock(&info->lock); set_page_dirty(page); swap_free(swap); - } else if (sgp == SGP_READ) { - page = find_get_page(mapping, index); - if (page && !trylock_page(page)) { - spin_unlock(&info->lock); - wait_on_page_locked(page); - page_cache_release(page); - goto repeat; + } else { + if (shmem_acct_block(info->flags)) { + error = -ENOSPC; + goto failed; } - spin_unlock(&info->lock); - - } else if (prealloc_page) { - sbinfo = SHMEM_SB(inode->i_sb); if (sbinfo->max_blocks) { if (percpu_counter_compare(&sbinfo->used_blocks, - sbinfo->max_blocks) >= 0 || - shmem_acct_block(info->flags)) - goto nospace; + sbinfo->max_blocks) >= 0) { + error = -ENOSPC; + goto unacct; + } percpu_counter_inc(&sbinfo->used_blocks); - inode->i_blocks += BLOCKS_PER_PAGE; - } else if (shmem_acct_block(info->flags)) - goto nospace; - - page = prealloc_page; - prealloc_page = NULL; - - swap = shmem_get_swap(info, index); - if (swap.val) - mem_cgroup_uncharge_cache_page(page); - else - error = add_to_page_cache_lru(page, mapping, - index, GFP_NOWAIT); - /* - * At add_to_page_cache_lru() failure, - * uncharge will be done automatically. - */ - if (swap.val || error) { - shmem_unacct_blocks(info->flags, 1); - shmem_free_blocks(inode, 1); - spin_unlock(&info->lock); - page_cache_release(page); - goto repeat; } + page = shmem_alloc_page(gfp, info, index); + if (!page) { + error = -ENOMEM; + goto decused; + } + + SetPageSwapBacked(page); + __set_page_locked(page); + error = shmem_add_to_page_cache(page, mapping, index, + gfp, NULL); + if (error) + goto decused; + lru_cache_add_anon(page); + + spin_lock(&info->lock); info->alloced++; + inode->i_blocks += BLOCKS_PER_PAGE; + shmem_recalc_inode(inode); spin_unlock(&info->lock); + clear_highpage(page); flush_dcache_page(page); SetPageUptodate(page); if (sgp == SGP_DIRTY) set_page_dirty(page); - - } else { - spin_unlock(&info->lock); - error = -ENOMEM; - goto out; } done: - *pagep = page; - error = 0; -out: - if (prealloc_page) { - mem_cgroup_uncharge_cache_page(prealloc_page); - page_cache_release(prealloc_page); + /* Perhaps the file has been truncated since we checked */ + if (sgp != SGP_WRITE && + ((loff_t)index << PAGE_CACHE_SHIFT) >= i_size_read(inode)) { + error = -EINVAL; + goto trunc; } - return error; + *pagep = page; + return 0; -nospace: /* - * Perhaps the page was brought in from swap between find_lock_page - * and taking info->lock? We allow for that at add_to_page_cache_lru, - * but must also avoid reporting a spurious ENOSPC while working on a - * full tmpfs. + * Error recovery. */ - page = find_get_page(mapping, index); +trunc: + ClearPageDirty(page); + delete_from_page_cache(page); + spin_lock(&info->lock); + info->alloced--; + inode->i_blocks -= BLOCKS_PER_PAGE; spin_unlock(&info->lock); +decused: + if (sbinfo->max_blocks) + percpu_counter_add(&sbinfo->used_blocks, -1); +unacct: + shmem_unacct_blocks(info->flags, 1); +failed: + if (swap.val && error != -EINVAL) { + struct page *test = find_get_page(mapping, index); + if (test && !radix_tree_exceptional_entry(test)) + page_cache_release(test); + /* Have another try if the entry has changed */ + if (test != swp_to_radix_entry(swap)) + error = -EEXIST; + } if (page) { + unlock_page(page); page_cache_release(page); + } + if (error == -ENOSPC && !once++) { + info = SHMEM_I(inode); + spin_lock(&info->lock); + shmem_recalc_inode(inode); + spin_unlock(&info->lock); goto repeat; } - error = -ENOSPC; - goto out; + if (error == -EEXIST) + goto repeat; + return error; } static int shmem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) @@ -1095,9 +1064,6 @@ static int shmem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) int error; int ret = VM_FAULT_LOCKED; - if (((loff_t)vmf->pgoff << PAGE_CACHE_SHIFT) >= i_size_read(inode)) - return VM_FAULT_SIGBUS; - error = shmem_getpage(inode, vmf->pgoff, &vmf->page, SGP_CACHE, &ret); if (error) return ((error == -ENOMEM) ? VM_FAULT_OOM : VM_FAULT_SIGBUS); @@ -2164,8 +2130,7 @@ static int shmem_remount_fs(struct super_block *sb, int *flags, char *data) if (config.max_inodes < inodes) goto out; /* - * Those tests also disallow limited->unlimited while any are in - * use, so i_blocks will always be zero when max_blocks is zero; + * Those tests disallow limited->unlimited while any are in use; * but we must separately disallow unlimited->limited, because * in that case we have no record of how much is already in use. */ From aa3b189551ad8e5cc1d9c663735c131650238278 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 3 Aug 2011 16:21:24 -0700 Subject: [PATCH 305/439] tmpfs: convert mem_cgroup shmem to radix-swap Remove mem_cgroup_shmem_charge_fallback(): it was only required when we had to move swappage to filecache with GFP_NOWAIT. Remove the GFP_NOWAIT special case from mem_cgroup_cache_charge(), by moving its call out from shmem_add_to_page_cache() to two of thats three callers. But leave it doing mem_cgroup_uncharge_cache_page() on error: although asymmetrical, it's easier for all 3 callers to handle. These two changes would also be appropriate if anyone were to start using shmem_read_mapping_page_gfp() with GFP_NOWAIT. Remove mem_cgroup_get_shmem_target(): mc_handle_file_pte() can test radix_tree_exceptional_entry() to get what it needs for itself. Signed-off-by: Hugh Dickins Acked-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memcontrol.h | 8 ---- include/linux/shmem_fs.h | 2 - mm/memcontrol.c | 68 +++++-------------------------- mm/shmem.c | 83 +++++--------------------------------- 4 files changed, 21 insertions(+), 140 deletions(-) diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index b96600786913..3b535db00a94 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -86,8 +86,6 @@ extern void mem_cgroup_uncharge_end(void); extern void mem_cgroup_uncharge_page(struct page *page); extern void mem_cgroup_uncharge_cache_page(struct page *page); -extern int mem_cgroup_shmem_charge_fallback(struct page *page, - struct mm_struct *mm, gfp_t gfp_mask); extern void mem_cgroup_out_of_memory(struct mem_cgroup *mem, gfp_t gfp_mask); int task_in_mem_cgroup(struct task_struct *task, const struct mem_cgroup *mem); @@ -225,12 +223,6 @@ static inline void mem_cgroup_uncharge_cache_page(struct page *page) { } -static inline int mem_cgroup_shmem_charge_fallback(struct page *page, - struct mm_struct *mm, gfp_t gfp_mask) -{ - return 0; -} - static inline void mem_cgroup_add_lru_list(struct page *page, int lru) { } diff --git a/include/linux/shmem_fs.h b/include/linux/shmem_fs.h index 3f05795dcf7b..0c8e952df594 100644 --- a/include/linux/shmem_fs.h +++ b/include/linux/shmem_fs.h @@ -57,8 +57,6 @@ extern struct page *shmem_read_mapping_page_gfp(struct address_space *mapping, pgoff_t index, gfp_t gfp_mask); extern void shmem_truncate_range(struct inode *inode, loff_t start, loff_t end); extern int shmem_unuse(swp_entry_t entry, struct page *page); -extern void mem_cgroup_get_shmem_target(struct inode *inode, pgoff_t pgoff, - struct page **pagep, swp_entry_t *ent); static inline struct page *shmem_read_mapping_page( struct address_space *mapping, pgoff_t index) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 5f84d2351ddb..f4ec4e7ca4cd 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -35,7 +35,6 @@ #include #include #include -#include #include #include #include @@ -2873,30 +2872,6 @@ int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm, return 0; if (PageCompound(page)) return 0; - /* - * Corner case handling. This is called from add_to_page_cache() - * in usual. But some FS (shmem) precharges this page before calling it - * and call add_to_page_cache() with GFP_NOWAIT. - * - * For GFP_NOWAIT case, the page may be pre-charged before calling - * add_to_page_cache(). (See shmem.c) check it here and avoid to call - * charge twice. (It works but has to pay a bit larger cost.) - * And when the page is SwapCache, it should take swap information - * into account. This is under lock_page() now. - */ - if (!(gfp_mask & __GFP_WAIT)) { - struct page_cgroup *pc; - - pc = lookup_page_cgroup(page); - if (!pc) - return 0; - lock_page_cgroup(pc); - if (PageCgroupUsed(pc)) { - unlock_page_cgroup(pc); - return 0; - } - unlock_page_cgroup(pc); - } if (unlikely(!mm)) mm = &init_mm; @@ -3486,31 +3461,6 @@ void mem_cgroup_end_migration(struct mem_cgroup *mem, cgroup_release_and_wakeup_rmdir(&mem->css); } -/* - * A call to try to shrink memory usage on charge failure at shmem's swapin. - * Calling hierarchical_reclaim is not enough because we should update - * last_oom_jiffies to prevent pagefault_out_of_memory from invoking global OOM. - * Moreover considering hierarchy, we should reclaim from the mem_over_limit, - * not from the memcg which this page would be charged to. - * try_charge_swapin does all of these works properly. - */ -int mem_cgroup_shmem_charge_fallback(struct page *page, - struct mm_struct *mm, - gfp_t gfp_mask) -{ - struct mem_cgroup *mem; - int ret; - - if (mem_cgroup_disabled()) - return 0; - - ret = mem_cgroup_try_charge_swapin(mm, page, gfp_mask, &mem); - if (!ret) - mem_cgroup_cancel_charge_swapin(mem); /* it does !mem check */ - - return ret; -} - #ifdef CONFIG_DEBUG_VM static struct page_cgroup *lookup_page_cgroup_used(struct page *page) { @@ -5330,15 +5280,17 @@ static struct page *mc_handle_file_pte(struct vm_area_struct *vma, pgoff = pte_to_pgoff(ptent); /* page is moved even if it's not RSS of this task(page-faulted). */ - if (!mapping_cap_swap_backed(mapping)) { /* normal file */ - page = find_get_page(mapping, pgoff); - } else { /* shmem/tmpfs file. we should take account of swap too. */ - swp_entry_t ent; - mem_cgroup_get_shmem_target(inode, pgoff, &page, &ent); - if (do_swap_account) - entry->val = ent.val; - } + page = find_get_page(mapping, pgoff); +#ifdef CONFIG_SWAP + /* shmem/tmpfs may report page out on swap: account for that too. */ + if (radix_tree_exceptional_entry(page)) { + swp_entry_t swap = radix_to_swp_entry(page); + if (do_swap_account) + *entry = swap; + page = find_get_page(&swapper_space, swap.val); + } +#endif return page; } diff --git a/mm/shmem.c b/mm/shmem.c index 92f01d7cc150..13ef2d7e912d 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -262,15 +262,11 @@ static int shmem_add_to_page_cache(struct page *page, struct address_space *mapping, pgoff_t index, gfp_t gfp, void *expected) { - int error; + int error = 0; VM_BUG_ON(!PageLocked(page)); VM_BUG_ON(!PageSwapBacked(page)); - error = mem_cgroup_cache_charge(page, current->mm, - gfp & GFP_RECLAIM_MASK); - if (error) - goto out; if (!expected) error = radix_tree_preload(gfp & GFP_RECLAIM_MASK); if (!error) { @@ -300,7 +296,6 @@ static int shmem_add_to_page_cache(struct page *page, } if (error) mem_cgroup_uncharge_cache_page(page); -out: return error; } @@ -660,7 +655,6 @@ int shmem_unuse(swp_entry_t swap, struct page *page) * Charge page using GFP_KERNEL while we can wait, before taking * the shmem_swaplist_mutex which might hold up shmem_writepage(). * Charged back to the user (not to caller) when swap account is used. - * shmem_add_to_page_cache() will be called with GFP_NOWAIT. */ error = mem_cgroup_cache_charge(page, current->mm, GFP_KERNEL); if (error) @@ -954,8 +948,11 @@ repeat: goto failed; } - error = shmem_add_to_page_cache(page, mapping, index, - gfp, swp_to_radix_entry(swap)); + error = mem_cgroup_cache_charge(page, current->mm, + gfp & GFP_RECLAIM_MASK); + if (!error) + error = shmem_add_to_page_cache(page, mapping, index, + gfp, swp_to_radix_entry(swap)); if (error) goto failed; @@ -990,8 +987,11 @@ repeat: SetPageSwapBacked(page); __set_page_locked(page); - error = shmem_add_to_page_cache(page, mapping, index, - gfp, NULL); + error = mem_cgroup_cache_charge(page, current->mm, + gfp & GFP_RECLAIM_MASK); + if (!error) + error = shmem_add_to_page_cache(page, mapping, index, + gfp, NULL); if (error) goto decused; lru_cache_add_anon(page); @@ -2442,42 +2442,6 @@ out4: return error; } -#ifdef CONFIG_CGROUP_MEM_RES_CTLR -/** - * mem_cgroup_get_shmem_target - find page or swap assigned to the shmem file - * @inode: the inode to be searched - * @index: the page offset to be searched - * @pagep: the pointer for the found page to be stored - * @swapp: the pointer for the found swap entry to be stored - * - * If a page is found, refcount of it is incremented. Callers should handle - * these refcount. - */ -void mem_cgroup_get_shmem_target(struct inode *inode, pgoff_t index, - struct page **pagep, swp_entry_t *swapp) -{ - struct shmem_inode_info *info = SHMEM_I(inode); - struct page *page = NULL; - swp_entry_t swap = {0}; - - if ((index << PAGE_CACHE_SHIFT) >= i_size_read(inode)) - goto out; - - spin_lock(&info->lock); -#ifdef CONFIG_SWAP - swap = shmem_get_swap(info, index); - if (swap.val) - page = find_get_page(&swapper_space, swap.val); - else -#endif - page = find_get_page(inode->i_mapping, index); - spin_unlock(&info->lock); -out: - *pagep = page; - *swapp = swap; -} -#endif - #else /* !CONFIG_SHMEM */ /* @@ -2523,31 +2487,6 @@ void shmem_truncate_range(struct inode *inode, loff_t lstart, loff_t lend) } EXPORT_SYMBOL_GPL(shmem_truncate_range); -#ifdef CONFIG_CGROUP_MEM_RES_CTLR -/** - * mem_cgroup_get_shmem_target - find page or swap assigned to the shmem file - * @inode: the inode to be searched - * @index: the page offset to be searched - * @pagep: the pointer for the found page to be stored - * @swapp: the pointer for the found swap entry to be stored - * - * If a page is found, refcount of it is incremented. Callers should handle - * these refcount. - */ -void mem_cgroup_get_shmem_target(struct inode *inode, pgoff_t index, - struct page **pagep, swp_entry_t *swapp) -{ - struct page *page = NULL; - - if ((index << PAGE_CACHE_SHIFT) >= i_size_read(inode)) - goto out; - page = find_get_page(inode->i_mapping, index); -out: - *pagep = page; - *swapp = (swp_entry_t){0}; -} -#endif - #define shmem_vm_ops generic_file_vm_ops #define shmem_file_operations ramfs_file_operations #define shmem_get_inode(sb, dir, mode, dev, flags) ramfs_get_inode(sb, dir, mode, dev) From 6922c0c7abd387374255801f7739624867e8acad Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 3 Aug 2011 16:21:25 -0700 Subject: [PATCH 306/439] tmpfs: convert shmem_writepage and enable swap Convert shmem_writepage() to use shmem_delete_from_page_cache() to use shmem_radix_tree_replace() to substitute swap entry for page pointer atomically in the radix tree. As with shmem_add_to_page_cache(), it's not entirely satisfactory to be copying such code from delete_from_swap_cache, but again judged easier to sell than making its other callers go through the extras. Remove the toy implementation's shmem_put_swap() and shmem_get_swap(), now unreferenced, and the hack to disable swap: it's now good to go. The way things have worked out, info->lock no longer helps to guard the shmem_swaplist: we increment swapped under shmem_swaplist_mutex only. That global mutex exclusion between shmem_writepage() and shmem_unuse() is not pretty, and we ought to find another way; but it's been forced on us by recent race discoveries, not a consequence of this patchset. And what has become of the WARN_ON_ONCE(1) free_swap_and_cache() if a swap entry was found already present? That's no longer possible, the (unknown) one inserting this page into filecache would hit the swap entry occupying that slot. Signed-off-by: Hugh Dickins Acked-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/shmem.c | 88 +++++++++++++++++++++++------------------------------- 1 file changed, 37 insertions(+), 51 deletions(-) diff --git a/mm/shmem.c b/mm/shmem.c index 13ef2d7e912d..0f094a258526 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -6,7 +6,8 @@ * 2000-2001 Christoph Rohland * 2000-2001 SAP AG * 2002 Red Hat Inc. - * Copyright (C) 2002-2005 Hugh Dickins. + * Copyright (C) 2002-2011 Hugh Dickins. + * Copyright (C) 2011 Google Inc. * Copyright (C) 2002-2005 VERITAS Software Corporation. * Copyright (C) 2004 Andi Kleen, SuSE Labs * @@ -219,19 +220,6 @@ static void shmem_recalc_inode(struct inode *inode) } } -static void shmem_put_swap(struct shmem_inode_info *info, pgoff_t index, - swp_entry_t swap) -{ - if (index < SHMEM_NR_DIRECT) - info->i_direct[index] = swap; -} - -static swp_entry_t shmem_get_swap(struct shmem_inode_info *info, pgoff_t index) -{ - return (index < SHMEM_NR_DIRECT) ? - info->i_direct[index] : (swp_entry_t){0}; -} - /* * Replace item expected in radix tree by a new item, while holding tree lock. */ @@ -299,6 +287,25 @@ static int shmem_add_to_page_cache(struct page *page, return error; } +/* + * Like delete_from_page_cache, but substitutes swap for page. + */ +static void shmem_delete_from_page_cache(struct page *page, void *radswap) +{ + struct address_space *mapping = page->mapping; + int error; + + spin_lock_irq(&mapping->tree_lock); + error = shmem_radix_tree_replace(mapping, page->index, page, radswap); + page->mapping = NULL; + mapping->nrpages--; + __dec_zone_page_state(page, NR_FILE_PAGES); + __dec_zone_page_state(page, NR_SHMEM); + spin_unlock_irq(&mapping->tree_lock); + page_cache_release(page); + BUG_ON(error); +} + /* * Like find_get_pages, but collecting swap entries as well as pages. */ @@ -664,14 +671,10 @@ int shmem_unuse(swp_entry_t swap, struct page *page) mutex_lock(&shmem_swaplist_mutex); list_for_each_safe(this, next, &shmem_swaplist) { info = list_entry(this, struct shmem_inode_info, swaplist); - if (!info->swapped) { - spin_lock(&info->lock); - if (!info->swapped) - list_del_init(&info->swaplist); - spin_unlock(&info->lock); - } if (info->swapped) found = shmem_unuse_inode(info, swap, page); + else + list_del_init(&info->swaplist); cond_resched(); if (found) break; @@ -694,10 +697,10 @@ out: static int shmem_writepage(struct page *page, struct writeback_control *wbc) { struct shmem_inode_info *info; - swp_entry_t swap, oswap; struct address_space *mapping; - pgoff_t index; struct inode *inode; + swp_entry_t swap; + pgoff_t index; BUG_ON(!PageLocked(page)); mapping = page->mapping; @@ -720,55 +723,38 @@ static int shmem_writepage(struct page *page, struct writeback_control *wbc) WARN_ON_ONCE(1); /* Still happens? Tell us about it! */ goto redirty; } - - /* - * Disable even the toy swapping implementation, while we convert - * functions one by one to having swap entries in the radix tree. - */ - if (index < ULONG_MAX) - goto redirty; - swap = get_swap_page(); if (!swap.val) goto redirty; /* * Add inode to shmem_unuse()'s list of swapped-out inodes, - * if it's not already there. Do it now because we cannot take - * mutex while holding spinlock, and must do so before the page - * is moved to swap cache, when its pagelock no longer protects + * if it's not already there. Do it now before the page is + * moved to swap cache, when its pagelock no longer protects * the inode from eviction. But don't unlock the mutex until - * we've taken the spinlock, because shmem_unuse_inode() will - * prune a !swapped inode from the swaplist under both locks. + * we've incremented swapped, because shmem_unuse_inode() will + * prune a !swapped inode from the swaplist under this mutex. */ mutex_lock(&shmem_swaplist_mutex); if (list_empty(&info->swaplist)) list_add_tail(&info->swaplist, &shmem_swaplist); - spin_lock(&info->lock); - mutex_unlock(&shmem_swaplist_mutex); - - oswap = shmem_get_swap(info, index); - if (oswap.val) { - WARN_ON_ONCE(1); /* Still happens? Tell us about it! */ - free_swap_and_cache(oswap); - shmem_put_swap(info, index, (swp_entry_t){0}); - info->swapped--; - } - shmem_recalc_inode(inode); - if (add_to_swap_cache(page, swap, GFP_ATOMIC) == 0) { - delete_from_page_cache(page); - shmem_put_swap(info, index, swap); - info->swapped++; swap_shmem_alloc(swap); + shmem_delete_from_page_cache(page, swp_to_radix_entry(swap)); + + spin_lock(&info->lock); + info->swapped++; + shmem_recalc_inode(inode); spin_unlock(&info->lock); + + mutex_unlock(&shmem_swaplist_mutex); BUG_ON(page_mapped(page)); swap_writepage(page, wbc); return 0; } - spin_unlock(&info->lock); + mutex_unlock(&shmem_swaplist_mutex); swapcache_free(swap, NULL); redirty: set_page_dirty(page); From 69f07ec938712b58755add82dd3d0b35f01317cc Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 3 Aug 2011 16:21:26 -0700 Subject: [PATCH 307/439] tmpfs: use kmemdup for short symlinks But we've not yet removed the old swp_entry_t i_direct[16] from shmem_inode_info. That's because it was still being shared with the inline symlink. Remove it now (saving 64 or 128 bytes from shmem inode size), and use kmemdup() for short symlinks, say, those up to 128 bytes. I wonder why mpol_free_shared_policy() is done in shmem_destroy_inode() rather than shmem_evict_inode(), where we usually do such freeing? I guess it doesn't matter, and I'm not into NUMA mpol testing right now. Signed-off-by: Hugh Dickins Acked-by: Rik van Riel Reviewed-by: Pekka Enberg Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/shmem_fs.h | 11 +++-------- mm/shmem.c | 31 ++++++++++++++++++------------- 2 files changed, 21 insertions(+), 21 deletions(-) diff --git a/include/linux/shmem_fs.h b/include/linux/shmem_fs.h index 0c8e952df594..9291ac3cc627 100644 --- a/include/linux/shmem_fs.h +++ b/include/linux/shmem_fs.h @@ -8,20 +8,15 @@ /* inode in-kernel data */ -#define SHMEM_NR_DIRECT 16 - -#define SHMEM_SYMLINK_INLINE_LEN (SHMEM_NR_DIRECT * sizeof(swp_entry_t)) - struct shmem_inode_info { spinlock_t lock; unsigned long flags; unsigned long alloced; /* data pages alloced to file */ - unsigned long swapped; /* subtotal assigned to swap */ - struct shared_policy policy; /* NUMA memory alloc policy */ union { - swp_entry_t i_direct[SHMEM_NR_DIRECT]; /* first blocks */ - char inline_symlink[SHMEM_SYMLINK_INLINE_LEN]; + unsigned long swapped; /* subtotal assigned to swap */ + char *symlink; /* unswappable short symlink */ }; + struct shared_policy policy; /* NUMA memory alloc policy */ struct list_head swaplist; /* chain of maybes on swap */ struct list_head xattr_list; /* list of shmem_xattr */ struct inode vfs_inode; diff --git a/mm/shmem.c b/mm/shmem.c index 0f094a258526..3a5be0feb6af 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -73,6 +73,9 @@ static struct vfsmount *shm_mnt; /* Pretend that each entry is of this size in directory's i_size */ #define BOGO_DIRENT_SIZE 20 +/* Symlink up to this size is kmalloc'ed instead of using a swappable page */ +#define SHORT_SYMLINK_LEN 128 + struct shmem_xattr { struct list_head list; /* anchored by shmem_inode_info->xattr_list */ char *name; /* xattr name */ @@ -585,7 +588,8 @@ static void shmem_evict_inode(struct inode *inode) list_del_init(&info->swaplist); mutex_unlock(&shmem_swaplist_mutex); } - } + } else + kfree(info->symlink); list_for_each_entry_safe(xattr, nxattr, &info->xattr_list, list) { kfree(xattr->name); @@ -1173,7 +1177,7 @@ static struct inode *shmem_get_inode(struct super_block *sb, const struct inode #ifdef CONFIG_TMPFS static const struct inode_operations shmem_symlink_inode_operations; -static const struct inode_operations shmem_symlink_inline_operations; +static const struct inode_operations shmem_short_symlink_operations; static int shmem_write_begin(struct file *file, struct address_space *mapping, @@ -1638,10 +1642,13 @@ static int shmem_symlink(struct inode *dir, struct dentry *dentry, const char *s info = SHMEM_I(inode); inode->i_size = len-1; - if (len <= SHMEM_SYMLINK_INLINE_LEN) { - /* do it inline */ - memcpy(info->inline_symlink, symname, len); - inode->i_op = &shmem_symlink_inline_operations; + if (len <= SHORT_SYMLINK_LEN) { + info->symlink = kmemdup(symname, len, GFP_KERNEL); + if (!info->symlink) { + iput(inode); + return -ENOMEM; + } + inode->i_op = &shmem_short_symlink_operations; } else { error = shmem_getpage(inode, 0, &page, SGP_WRITE, NULL); if (error) { @@ -1664,9 +1671,9 @@ static int shmem_symlink(struct inode *dir, struct dentry *dentry, const char *s return 0; } -static void *shmem_follow_link_inline(struct dentry *dentry, struct nameidata *nd) +static void *shmem_follow_short_symlink(struct dentry *dentry, struct nameidata *nd) { - nd_set_link(nd, SHMEM_I(dentry->d_inode)->inline_symlink); + nd_set_link(nd, SHMEM_I(dentry->d_inode)->symlink); return NULL; } @@ -1914,9 +1921,9 @@ static ssize_t shmem_listxattr(struct dentry *dentry, char *buffer, size_t size) } #endif /* CONFIG_TMPFS_XATTR */ -static const struct inode_operations shmem_symlink_inline_operations = { +static const struct inode_operations shmem_short_symlink_operations = { .readlink = generic_readlink, - .follow_link = shmem_follow_link_inline, + .follow_link = shmem_follow_short_symlink, #ifdef CONFIG_TMPFS_XATTR .setxattr = shmem_setxattr, .getxattr = shmem_getxattr, @@ -2259,10 +2266,8 @@ static void shmem_destroy_callback(struct rcu_head *head) static void shmem_destroy_inode(struct inode *inode) { - if ((inode->i_mode & S_IFMT) == S_IFREG) { - /* only struct inode is valid if it's an inline symlink */ + if ((inode->i_mode & S_IFMT) == S_IFREG) mpol_free_shared_policy(&SHMEM_I(inode)->policy); - } call_rcu(&inode->i_rcu, shmem_destroy_callback); } From 31475dd611209413bace21651a400afb91d0bd9d Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 3 Aug 2011 16:21:27 -0700 Subject: [PATCH 308/439] mm: a few small updates for radix-swap Remove PageSwapBacked (!page_is_file_cache) cases from add_to_page_cache_locked() and add_to_page_cache_lru(): those pages now go through shmem_add_to_page_cache(). Remove a comment on maximum tmpfs size from fsstack_copy_inode_size(), and add a comment on swap entries to invalidate_mapping_pages(). And mincore_page() uses find_get_page() on what might be shmem or a tmpfs file: allow for a radix_tree_exceptional_entry(), and proceed to find_get_page() on swapper_space if so (oh, swapper_space needs #ifdef). Signed-off-by: Hugh Dickins Acked-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/stack.c | 5 +---- mm/filemap.c | 21 +++------------------ mm/mincore.c | 10 ++++++---- mm/truncate.c | 8 ++++++++ 4 files changed, 18 insertions(+), 26 deletions(-) diff --git a/fs/stack.c b/fs/stack.c index 4a6f7f440658..b4f2ab48a61f 100644 --- a/fs/stack.c +++ b/fs/stack.c @@ -29,10 +29,7 @@ void fsstack_copy_inode_size(struct inode *dst, struct inode *src) * * We don't actually know what locking is used at the lower level; * but if it's a filesystem that supports quotas, it will be using - * i_lock as in inode_add_bytes(). tmpfs uses other locking, and - * its 32-bit is (just) able to exceed 2TB i_size with the aid of - * holes; but its i_blocks cannot carry into the upper long without - * almost 2TB swap - let's ignore that case. + * i_lock as in inode_add_bytes(). */ if (sizeof(i_blocks) > sizeof(long)) spin_lock(&src->i_lock); diff --git a/mm/filemap.c b/mm/filemap.c index 76bfb6460f57..96778faf82d5 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -33,7 +33,6 @@ #include #include /* for BUG_ON(!in_atomic()) only */ #include -#include /* for page_is_file_cache() */ #include #include "internal.h" @@ -462,6 +461,7 @@ int add_to_page_cache_locked(struct page *page, struct address_space *mapping, int error; VM_BUG_ON(!PageLocked(page)); + VM_BUG_ON(PageSwapBacked(page)); error = mem_cgroup_cache_charge(page, current->mm, gfp_mask & GFP_RECLAIM_MASK); @@ -479,8 +479,6 @@ int add_to_page_cache_locked(struct page *page, struct address_space *mapping, if (likely(!error)) { mapping->nrpages++; __inc_zone_page_state(page, NR_FILE_PAGES); - if (PageSwapBacked(page)) - __inc_zone_page_state(page, NR_SHMEM); spin_unlock_irq(&mapping->tree_lock); } else { page->mapping = NULL; @@ -502,22 +500,9 @@ int add_to_page_cache_lru(struct page *page, struct address_space *mapping, { int ret; - /* - * Splice_read and readahead add shmem/tmpfs pages into the page cache - * before shmem_readpage has a chance to mark them as SwapBacked: they - * need to go on the anon lru below, and mem_cgroup_cache_charge - * (called in add_to_page_cache) needs to know where they're going too. - */ - if (mapping_cap_swap_backed(mapping)) - SetPageSwapBacked(page); - ret = add_to_page_cache(page, mapping, offset, gfp_mask); - if (ret == 0) { - if (page_is_file_cache(page)) - lru_cache_add_file(page); - else - lru_cache_add_anon(page); - } + if (ret == 0) + lru_cache_add_file(page); return ret; } EXPORT_SYMBOL_GPL(add_to_page_cache_lru); diff --git a/mm/mincore.c b/mm/mincore.c index a4e6b9d75c76..733f1829b0dc 100644 --- a/mm/mincore.c +++ b/mm/mincore.c @@ -69,12 +69,14 @@ static unsigned char mincore_page(struct address_space *mapping, pgoff_t pgoff) * file will not get a swp_entry_t in its pte, but rather it is like * any other file mapping (ie. marked !present and faulted in with * tmpfs's .fault). So swapped out tmpfs mappings are tested here. - * - * However when tmpfs moves the page from pagecache and into swapcache, - * it is still in core, but the find_get_page below won't find it. - * No big deal, but make a note of it. */ page = find_get_page(mapping, pgoff); +#ifdef CONFIG_SWAP + if (radix_tree_exceptional_entry(page)) { + swp_entry_t swap = radix_to_swp_entry(page); + page = find_get_page(&swapper_space, swap.val); + } +#endif if (page) { present = PageUptodate(page); page_cache_release(page); diff --git a/mm/truncate.c b/mm/truncate.c index 232eb2736a79..b40ac6d4e86e 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -336,6 +336,14 @@ unsigned long invalidate_mapping_pages(struct address_space *mapping, unsigned long count = 0; int i; + /* + * Note: this function may get called on a shmem/tmpfs mapping: + * pagevec_lookup() might then return 0 prematurely (because it + * got a gangful of swap entries); but it's hardly worth worrying + * about - it can rarely have anything to free from such a mapping + * (most pages are dirty), and already skips over any difficulties. + */ + pagevec_init(&pvec, 0); while (index <= end && pagevec_lookup(&pvec, mapping, index, min(end - index, (pgoff_t)PAGEVEC_SIZE - 1) + 1)) { From e504f3fdd63d486d45b18009e5a65f2e329acb0a Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 3 Aug 2011 16:21:27 -0700 Subject: [PATCH 309/439] tmpfs radix_tree: locate_item to speed up swapoff We have already acknowledged that swapoff of a tmpfs file is slower than it was before conversion to the generic radix_tree: a little slower there will be acceptable, if the hotter paths are faster. But it was a shock to find swapoff of a 500MB file 20 times slower on my laptop, taking 10 minutes; and at that rate it significantly slows down my testing. Now, most of that turned out to be overhead from PROVE_LOCKING and PROVE_RCU: without those it was only 4 times slower than before; and more realistic tests on other machines don't fare as badly. I've tried a number of things to improve it, including tagging the swap entries, then doing lookup by tag: I'd expected that to halve the time, but in practice it's erratic, and often counter-productive. The only change I've so far found to make a consistent improvement, is to short-circuit the way we go back and forth, gang lookup packing entries into the array supplied, then shmem scanning that array for the target entry. Scanning in place doubles the speed, so it's now only twice as slow as before (or three times slower when the PROVEs are on). So, add radix_tree_locate_item() as an expedient, once-off, single-caller hack to do the lookup directly in place. #ifdef it on CONFIG_SHMEM and CONFIG_SWAP, as much to document its limited applicability as save space in other configurations. And, sadly, #include sched.h for cond_resched(). Signed-off-by: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/radix-tree.h | 1 + lib/radix-tree.c | 92 ++++++++++++++++++++++++++++++++++++++ mm/shmem.c | 38 +--------------- 3 files changed, 94 insertions(+), 37 deletions(-) diff --git a/include/linux/radix-tree.h b/include/linux/radix-tree.h index b7edf8251455..9d4539c52e53 100644 --- a/include/linux/radix-tree.h +++ b/include/linux/radix-tree.h @@ -252,6 +252,7 @@ unsigned long radix_tree_range_tag_if_tagged(struct radix_tree_root *root, unsigned long nr_to_tag, unsigned int fromtag, unsigned int totag); int radix_tree_tagged(struct radix_tree_root *root, unsigned int tag); +unsigned long radix_tree_locate_item(struct radix_tree_root *root, void *item); static inline void radix_tree_preload_end(void) { diff --git a/lib/radix-tree.c b/lib/radix-tree.c index 348eaefbed74..a2f9da59c197 100644 --- a/lib/radix-tree.c +++ b/lib/radix-tree.c @@ -1203,6 +1203,98 @@ radix_tree_gang_lookup_tag_slot(struct radix_tree_root *root, void ***results, } EXPORT_SYMBOL(radix_tree_gang_lookup_tag_slot); +#if defined(CONFIG_SHMEM) && defined(CONFIG_SWAP) +#include /* for cond_resched() */ + +/* + * This linear search is at present only useful to shmem_unuse_inode(). + */ +static unsigned long __locate(struct radix_tree_node *slot, void *item, + unsigned long index, unsigned long *found_index) +{ + unsigned int shift, height; + unsigned long i; + + height = slot->height; + shift = (height-1) * RADIX_TREE_MAP_SHIFT; + + for ( ; height > 1; height--) { + i = (index >> shift) & RADIX_TREE_MAP_MASK; + for (;;) { + if (slot->slots[i] != NULL) + break; + index &= ~((1UL << shift) - 1); + index += 1UL << shift; + if (index == 0) + goto out; /* 32-bit wraparound */ + i++; + if (i == RADIX_TREE_MAP_SIZE) + goto out; + } + + shift -= RADIX_TREE_MAP_SHIFT; + slot = rcu_dereference_raw(slot->slots[i]); + if (slot == NULL) + goto out; + } + + /* Bottom level: check items */ + for (i = 0; i < RADIX_TREE_MAP_SIZE; i++) { + if (slot->slots[i] == item) { + *found_index = index + i; + index = 0; + goto out; + } + } + index += RADIX_TREE_MAP_SIZE; +out: + return index; +} + +/** + * radix_tree_locate_item - search through radix tree for item + * @root: radix tree root + * @item: item to be found + * + * Returns index where item was found, or -1 if not found. + * Caller must hold no lock (since this time-consuming function needs + * to be preemptible), and must check afterwards if item is still there. + */ +unsigned long radix_tree_locate_item(struct radix_tree_root *root, void *item) +{ + struct radix_tree_node *node; + unsigned long max_index; + unsigned long cur_index = 0; + unsigned long found_index = -1; + + do { + rcu_read_lock(); + node = rcu_dereference_raw(root->rnode); + if (!radix_tree_is_indirect_ptr(node)) { + rcu_read_unlock(); + if (node == item) + found_index = 0; + break; + } + + node = indirect_to_ptr(node); + max_index = radix_tree_maxindex(node->height); + if (cur_index > max_index) + break; + + cur_index = __locate(node, item, cur_index, &found_index); + rcu_read_unlock(); + cond_resched(); + } while (cur_index != 0 && cur_index <= max_index); + + return found_index; +} +#else +unsigned long radix_tree_locate_item(struct radix_tree_root *root, void *item) +{ + return -1; +} +#endif /* CONFIG_SHMEM && CONFIG_SWAP */ /** * radix_tree_shrink - shrink height of a radix tree to minimal diff --git a/mm/shmem.c b/mm/shmem.c index 3a5be0feb6af..1c702f6f1241 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -356,42 +356,6 @@ export: return ret; } -/* - * Lockless lookup of swap entry in radix tree, avoiding refcount on pages. - */ -static pgoff_t shmem_find_swap(struct address_space *mapping, void *radswap) -{ - void **slots[PAGEVEC_SIZE]; - pgoff_t indices[PAGEVEC_SIZE]; - unsigned int nr_found; - -restart: - nr_found = 1; - indices[0] = -1; - while (nr_found) { - pgoff_t index = indices[nr_found - 1] + 1; - unsigned int i; - - rcu_read_lock(); - nr_found = radix_tree_gang_lookup_slot(&mapping->page_tree, - slots, indices, index, PAGEVEC_SIZE); - for (i = 0; i < nr_found; i++) { - void *item = radix_tree_deref_slot(slots[i]); - if (radix_tree_deref_retry(item)) { - rcu_read_unlock(); - goto restart; - } - if (item == radswap) { - rcu_read_unlock(); - return indices[i]; - } - } - rcu_read_unlock(); - cond_resched(); - } - return -1; -} - /* * Remove swap entry from radix tree, free the swap and its page cache. */ @@ -612,7 +576,7 @@ static int shmem_unuse_inode(struct shmem_inode_info *info, int error; radswap = swp_to_radix_entry(swap); - index = shmem_find_swap(mapping, radswap); + index = radix_tree_locate_item(&mapping->page_tree, radswap); if (index == -1) return 0; From 8079b1c859c44f27d63da4951f5038a16589a563 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 3 Aug 2011 16:21:28 -0700 Subject: [PATCH 310/439] mm: clarify the radix_tree exceptional cases Make the radix_tree exceptional cases, mostly in filemap.c, clearer. It's hard to devise a suitable snappy name that illuminates the use by shmem/tmpfs for swap, while keeping filemap/pagecache/radix_tree generality. And akpm points out that /* radix_tree_deref_retry(page) */ comments look like calls that have been commented out for unknown reason. Skirt the naming difficulty by rearranging these blocks to handle the transient radix_tree_deref_retry(page) case first; then just explain the remaining shmem/tmpfs swap case in a comment. Signed-off-by: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/filemap.c | 66 ++++++++++++++++++++++++++++++++++------------------ mm/mincore.c | 1 + mm/shmem.c | 12 ++++++---- 3 files changed, 53 insertions(+), 26 deletions(-) diff --git a/mm/filemap.c b/mm/filemap.c index 96778faf82d5..645a080ba4df 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -700,10 +700,14 @@ repeat: if (unlikely(!page)) goto out; if (radix_tree_exception(page)) { - if (radix_tree_exceptional_entry(page)) - goto out; - /* radix_tree_deref_retry(page) */ - goto repeat; + if (radix_tree_deref_retry(page)) + goto repeat; + /* + * Otherwise, shmem/tmpfs must be storing a swap entry + * here as an exceptional entry: so return it without + * attempting to raise page count. + */ + goto out; } if (!page_cache_get_speculative(page)) goto repeat; @@ -838,15 +842,21 @@ repeat: continue; if (radix_tree_exception(page)) { - if (radix_tree_exceptional_entry(page)) - continue; + if (radix_tree_deref_retry(page)) { + /* + * Transient condition which can only trigger + * when entry at index 0 moves out of or back + * to root: none yet gotten, safe to restart. + */ + WARN_ON(start | i); + goto restart; + } /* - * radix_tree_deref_retry(page): - * can only trigger when entry at index 0 moves out of - * or back to root: none yet gotten, safe to restart. + * Otherwise, shmem/tmpfs must be storing a swap entry + * here as an exceptional entry: so skip over it - + * we only reach this from invalidate_mapping_pages(). */ - WARN_ON(start | i); - goto restart; + continue; } if (!page_cache_get_speculative(page)) @@ -904,14 +914,20 @@ repeat: continue; if (radix_tree_exception(page)) { - if (radix_tree_exceptional_entry(page)) - break; + if (radix_tree_deref_retry(page)) { + /* + * Transient condition which can only trigger + * when entry at index 0 moves out of or back + * to root: none yet gotten, safe to restart. + */ + goto restart; + } /* - * radix_tree_deref_retry(page): - * can only trigger when entry at index 0 moves out of - * or back to root: none yet gotten, safe to restart. + * Otherwise, shmem/tmpfs must be storing a swap entry + * here as an exceptional entry: so stop looking for + * contiguous pages. */ - goto restart; + break; } if (!page_cache_get_speculative(page)) @@ -973,13 +989,19 @@ repeat: continue; if (radix_tree_exception(page)) { - BUG_ON(radix_tree_exceptional_entry(page)); + if (radix_tree_deref_retry(page)) { + /* + * Transient condition which can only trigger + * when entry at index 0 moves out of or back + * to root: none yet gotten, safe to restart. + */ + goto restart; + } /* - * radix_tree_deref_retry(page): - * can only trigger when entry at index 0 moves out of - * or back to root: none yet gotten, safe to restart. + * This function is never used on a shmem/tmpfs + * mapping, so a swap entry won't be found here. */ - goto restart; + BUG(); } if (!page_cache_get_speculative(page)) diff --git a/mm/mincore.c b/mm/mincore.c index 733f1829b0dc..636a86876ff2 100644 --- a/mm/mincore.c +++ b/mm/mincore.c @@ -72,6 +72,7 @@ static unsigned char mincore_page(struct address_space *mapping, pgoff_t pgoff) */ page = find_get_page(mapping, pgoff); #ifdef CONFIG_SWAP + /* shmem/tmpfs may return swap: account for swapcache page too. */ if (radix_tree_exceptional_entry(page)) { swp_entry_t swap = radix_to_swp_entry(page); page = find_get_page(&swapper_space, swap.val); diff --git a/mm/shmem.c b/mm/shmem.c index 1c702f6f1241..32f6763f16fb 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -332,10 +332,14 @@ repeat: if (unlikely(!page)) continue; if (radix_tree_exception(page)) { - if (radix_tree_exceptional_entry(page)) - goto export; - /* radix_tree_deref_retry(page) */ - goto restart; + if (radix_tree_deref_retry(page)) + goto restart; + /* + * Otherwise, we must be storing a swap entry + * here as an exceptional entry: so return it + * without attempting to raise page count. + */ + goto export; } if (!page_cache_get_speculative(page)) goto repeat; From 206506ccf04b6790d11553a0c8595d1bf65790fe Mon Sep 17 00:00:00 2001 From: "Robert P. J. Day" Date: Wed, 3 Aug 2011 16:21:29 -0700 Subject: [PATCH 311/439] tmpfs: expand "help" to explain value of TMPFS_POSIX_ACL Expand the fs/Kconfig "help" info to clarify why it's a bad idea to deselect the TMPFS_POSIX_ACL config variable. Signed-off-by: Robert P. J. Day Acked-by: Randy Dunlap Acked-by: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/Kconfig | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) diff --git a/fs/Kconfig b/fs/Kconfig index 19891aab9c6e..9fe0b349f4cd 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -127,14 +127,21 @@ config TMPFS_POSIX_ACL select TMPFS_XATTR select GENERIC_ACL help - POSIX Access Control Lists (ACLs) support permissions for users and - groups beyond the owner/group/world scheme. + POSIX Access Control Lists (ACLs) support additional access rights + for users and groups beyond the standard owner/group/world scheme, + and this option selects support for ACLs specifically for tmpfs + filesystems. + + If you've selected TMPFS, it's possible that you'll also need + this option as there are a number of Linux distros that require + POSIX ACL support under /dev for certain features to work properly. + For example, some distros need this feature for ALSA-related /dev + files for sound to work properly. In short, if you're not sure, + say Y. To learn more about Access Control Lists, visit the POSIX ACLs for Linux website . - If you don't know what Access Control Lists are, say N. - config TMPFS_XATTR bool "Tmpfs extended attributes" depends on TMPFS From ebec9a7bf11f843b0602b06c402f04bf4213b35a Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Wed, 3 Aug 2011 09:22:54 -0700 Subject: [PATCH 312/439] drm: track CEA version number if present Drivers need to know the CEA version number in addition to other display info (like whether the display is an HDMI sink) before enabling certain features. So track the CEA version number in the display info structure. Signed-off-by: Jesse Barnes Signed-off-by: Keith Packard --- drivers/gpu/drm/drm_edid.c | 9 +++++++++ include/drm/drm_crtc.h | 2 ++ 2 files changed, 11 insertions(+) diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c index 09292193dafe..e12f8b0cec94 100644 --- a/drivers/gpu/drm/drm_edid.c +++ b/drivers/gpu/drm/drm_edid.c @@ -1439,6 +1439,8 @@ EXPORT_SYMBOL(drm_detect_monitor_audio); static void drm_add_display_info(struct edid *edid, struct drm_display_info *info) { + u8 *edid_ext; + info->width_mm = edid->width_cm * 10; info->height_mm = edid->height_cm * 10; @@ -1483,6 +1485,13 @@ static void drm_add_display_info(struct edid *edid, info->color_formats = DRM_COLOR_FORMAT_YCRCB444; if (info->color_formats & DRM_EDID_FEATURE_RGB_YCRCB422) info->color_formats = DRM_COLOR_FORMAT_YCRCB422; + + /* Get data from CEA blocks if present */ + edid_ext = drm_find_cea_extension(edid); + if (!edid_ext) + return; + + info->cea_rev = edid_ext[1]; } /** diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h index 33d12f87f0e0..d515bc8f4ba7 100644 --- a/include/drm/drm_crtc.h +++ b/include/drm/drm_crtc.h @@ -205,6 +205,8 @@ struct drm_display_info { enum subpixel_order subpixel_order; u32 color_formats; + u8 cea_rev; + char *raw_edid; /* if any */ }; From 45187ace97f7b3deb559b25348ccb7e301c158c9 Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Wed, 3 Aug 2011 09:22:55 -0700 Subject: [PATCH 313/439] drm/i915/hdmi: split infoframe setting from infoframe type code This makes it easier to add support for other infoframes (e.g. SPD, vendor specific). Signed-off-by: Jesse Barnes Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/i915_reg.h | 1 + drivers/gpu/drm/i915/intel_drv.h | 2 + drivers/gpu/drm/i915/intel_hdmi.c | 163 ++++++++++++++++++++---------- 3 files changed, 113 insertions(+), 53 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index be67a596eee5..56cae72a5207 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -1506,6 +1506,7 @@ #define VIDEO_DIP_SELECT_AVI (0 << 19) #define VIDEO_DIP_SELECT_VENDOR (1 << 19) #define VIDEO_DIP_SELECT_SPD (3 << 19) +#define VIDEO_DIP_SELECT_MASK (3 << 19) #define VIDEO_DIP_FREQ_ONCE (0 << 16) #define VIDEO_DIP_FREQ_VSYNC (1 << 16) #define VIDEO_DIP_FREQ_2VSYNC (2 << 16) diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index 6e990f9760ef..2f47dcf0b367 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -178,6 +178,8 @@ struct intel_crtc { #define to_intel_encoder(x) container_of(x, struct intel_encoder, base) #define to_intel_framebuffer(x) container_of(x, struct intel_framebuffer, base) +#define DIP_HEADER_SIZE 5 + #define DIP_TYPE_AVI 0x82 #define DIP_VERSION_AVI 0x2 #define DIP_LEN_AVI 13 diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c index 105e2b892f3a..d43cebdea1ee 100644 --- a/drivers/gpu/drm/i915/intel_hdmi.c +++ b/drivers/gpu/drm/i915/intel_hdmi.c @@ -45,6 +45,8 @@ struct intel_hdmi { bool has_hdmi_sink; bool has_audio; int force_audio; + void (*write_infoframe)(struct drm_encoder *encoder, + struct dip_infoframe *frame); }; static struct intel_hdmi *enc_to_intel_hdmi(struct drm_encoder *encoder) @@ -58,37 +60,70 @@ static struct intel_hdmi *intel_attached_hdmi(struct drm_connector *connector) struct intel_hdmi, base); } -void intel_dip_infoframe_csum(struct dip_infoframe *avi_if) +void intel_dip_infoframe_csum(struct dip_infoframe *frame) { - uint8_t *data = (uint8_t *)avi_if; + uint8_t *data = (uint8_t *)frame; uint8_t sum = 0; unsigned i; - avi_if->checksum = 0; - avi_if->ecc = 0; + frame->checksum = 0; + frame->ecc = 0; - for (i = 0; i < sizeof(*avi_if); i++) + /* Header isn't part of the checksum */ + for (i = 5; i < frame->len; i++) sum += data[i]; - avi_if->checksum = 0x100 - sum; + frame->checksum = 0x100 - sum; } -static void intel_hdmi_set_avi_infoframe(struct drm_encoder *encoder) +static u32 intel_infoframe_index(struct dip_infoframe *frame) { - struct dip_infoframe avi_if = { - .type = DIP_TYPE_AVI, - .ver = DIP_VERSION_AVI, - .len = DIP_LEN_AVI, - }; - uint32_t *data = (uint32_t *)&avi_if; + u32 flags = 0; + + switch (frame->type) { + case DIP_TYPE_AVI: + flags |= VIDEO_DIP_SELECT_AVI; + break; + case DIP_TYPE_SPD: + flags |= VIDEO_DIP_SELECT_SPD; + break; + default: + DRM_DEBUG_DRIVER("unknown info frame type %d\n", frame->type); + break; + } + + return flags; +} + +static u32 intel_infoframe_flags(struct dip_infoframe *frame) +{ + u32 flags = 0; + + switch (frame->type) { + case DIP_TYPE_AVI: + flags |= VIDEO_DIP_ENABLE_AVI | VIDEO_DIP_FREQ_VSYNC; + break; + case DIP_TYPE_SPD: + flags |= VIDEO_DIP_ENABLE_SPD | VIDEO_DIP_FREQ_2VSYNC; + break; + default: + DRM_DEBUG_DRIVER("unknown info frame type %d\n", frame->type); + break; + } + + return flags; +} + +static void i9xx_write_infoframe(struct drm_encoder *encoder, + struct dip_infoframe *frame) +{ + uint32_t *data = (uint32_t *)frame; struct drm_device *dev = encoder->dev; struct drm_i915_private *dev_priv = dev->dev_private; struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); - u32 port; - unsigned i; + u32 port, flags, val = I915_READ(VIDEO_DIP_CTL); + unsigned i, len = DIP_HEADER_SIZE + frame->len; - if (!intel_hdmi->has_hdmi_sink) - return; /* XXX first guess at handling video port, is this corrent? */ if (intel_hdmi->sdvox_reg == SDVOB) @@ -98,52 +133,72 @@ static void intel_hdmi_set_avi_infoframe(struct drm_encoder *encoder) else return; - I915_WRITE(VIDEO_DIP_CTL, VIDEO_DIP_ENABLE | port | - VIDEO_DIP_SELECT_AVI | VIDEO_DIP_FREQ_VSYNC); + flags = intel_infoframe_index(frame); - intel_dip_infoframe_csum(&avi_if); - for (i = 0; i < sizeof(avi_if); i += 4) { + val &= ~VIDEO_DIP_SELECT_MASK; + + I915_WRITE(VIDEO_DIP_CTL, val | port | flags); + + for (i = 0; i < len; i += 4) { I915_WRITE(VIDEO_DIP_DATA, *data); data++; } - I915_WRITE(VIDEO_DIP_CTL, VIDEO_DIP_ENABLE | port | - VIDEO_DIP_SELECT_AVI | VIDEO_DIP_FREQ_VSYNC | - VIDEO_DIP_ENABLE_AVI); + flags |= intel_infoframe_flags(frame); + + I915_WRITE(VIDEO_DIP_CTL, VIDEO_DIP_ENABLE | val | port | flags); } -static void intel_ironlake_hdmi_set_avi_infoframe(struct drm_encoder *encoder) +static void ironlake_write_infoframe(struct drm_encoder *encoder, + struct dip_infoframe *frame) +{ + uint32_t *data = (uint32_t *)frame; + struct drm_device *dev = encoder->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + struct drm_crtc *crtc = encoder->crtc; + struct intel_crtc *intel_crtc = to_intel_crtc(crtc); + int reg = TVIDEO_DIP_CTL(intel_crtc->pipe); + unsigned i, len = DIP_HEADER_SIZE + frame->len; + u32 flags, val = I915_READ(reg); + + intel_wait_for_vblank(dev, intel_crtc->pipe); + + flags = intel_infoframe_index(frame); + + val &= ~VIDEO_DIP_SELECT_MASK; + + I915_WRITE(reg, val | flags); + + for (i = 0; i < len; i += 4) { + I915_WRITE(TVIDEO_DIP_DATA(intel_crtc->pipe), *data); + data++; + } + + flags |= intel_infoframe_flags(frame); + + I915_WRITE(reg, VIDEO_DIP_ENABLE | val | flags); +} +static void intel_set_infoframe(struct drm_encoder *encoder, + struct dip_infoframe *frame) +{ + struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); + + if (!intel_hdmi->has_hdmi_sink) + return; + + intel_dip_infoframe_csum(frame); + intel_hdmi->write_infoframe(encoder, frame); +} + +static void intel_hdmi_set_avi_infoframe(struct drm_encoder *encoder) { struct dip_infoframe avi_if = { .type = DIP_TYPE_AVI, .ver = DIP_VERSION_AVI, .len = DIP_LEN_AVI, }; - uint32_t *data = (uint32_t *)&avi_if; - struct drm_device *dev = encoder->dev; - struct drm_i915_private *dev_priv = dev->dev_private; - struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); - struct drm_crtc *crtc = encoder->crtc; - struct intel_crtc *intel_crtc = to_intel_crtc(crtc); - int reg = TVIDEO_DIP_CTL(intel_crtc->pipe); - unsigned i; - if (!intel_hdmi->has_hdmi_sink) - return; - - intel_wait_for_vblank(dev, intel_crtc->pipe); - - I915_WRITE(reg, VIDEO_DIP_SELECT_AVI); - - intel_dip_infoframe_csum(&avi_if); - for (i = 0; i < sizeof(avi_if); i += 4) { - I915_WRITE(TVIDEO_DIP_DATA(intel_crtc->pipe), *data); - data++; - } - - I915_WRITE(reg, VIDEO_DIP_ENABLE | VIDEO_DIP_SELECT_AVI | - VIDEO_DIP_FREQ_VSYNC | (DIP_LEN_AVI << 8) | - VIDEO_DIP_ENABLE_AVI); + intel_set_infoframe(encoder, &avi_if); } static void intel_hdmi_mode_set(struct drm_encoder *encoder, @@ -189,10 +244,7 @@ static void intel_hdmi_mode_set(struct drm_encoder *encoder, I915_WRITE(intel_hdmi->sdvox_reg, sdvox); POSTING_READ(intel_hdmi->sdvox_reg); - if (HAS_PCH_SPLIT(dev)) - intel_ironlake_hdmi_set_avi_infoframe(encoder); - else - intel_hdmi_set_avi_infoframe(encoder); + intel_hdmi_set_avi_infoframe(encoder); } static void intel_hdmi_dpms(struct drm_encoder *encoder, int mode) @@ -470,6 +522,11 @@ void intel_hdmi_init(struct drm_device *dev, int sdvox_reg) intel_hdmi->sdvox_reg = sdvox_reg; + if (!HAS_PCH_SPLIT(dev)) + intel_hdmi->write_infoframe = i9xx_write_infoframe; + else + intel_hdmi->write_infoframe = ironlake_write_infoframe; + drm_encoder_helper_add(&intel_encoder->base, &intel_hdmi_helper_funcs); intel_hdmi_add_properties(intel_hdmi, connector); From c0864cb39c68696e80657360eba63da5e743b7aa Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Wed, 3 Aug 2011 09:22:56 -0700 Subject: [PATCH 314/439] drm/i915/hdmi: HDMI source product description infoframe support Set an SPD infoframe if the sink supports it. Signed-off-by: Jesse Barnes Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_drv.h | 21 +++++++++++++++++++++ drivers/gpu/drm/i915/intel_hdmi.c | 16 ++++++++++++++++ 2 files changed, 37 insertions(+) diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index 2f47dcf0b367..7b330e76a435 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -184,6 +184,22 @@ struct intel_crtc { #define DIP_VERSION_AVI 0x2 #define DIP_LEN_AVI 13 +#define DIP_TYPE_SPD 0x3 +#define DIP_VERSION_SPD 0x1 +#define DIP_LEN_SPD 25 +#define DIP_SPD_UNKNOWN 0 +#define DIP_SPD_DSTB 0x1 +#define DIP_SPD_DVDP 0x2 +#define DIP_SPD_DVHS 0x3 +#define DIP_SPD_HDDVR 0x4 +#define DIP_SPD_DVC 0x5 +#define DIP_SPD_DSC 0x6 +#define DIP_SPD_VCD 0x7 +#define DIP_SPD_GAME 0x8 +#define DIP_SPD_PC 0x9 +#define DIP_SPD_BD 0xa +#define DIP_SPD_SCD 0xb + struct dip_infoframe { uint8_t type; /* HB0 */ uint8_t ver; /* HB1 */ @@ -208,6 +224,11 @@ struct dip_infoframe { uint16_t left_bar_end; uint16_t right_bar_start; } avi; + struct { + uint8_t vn[8]; + uint8_t pd[16]; + uint8_t sdi; + } spd; uint8_t payload[27]; } __attribute__ ((packed)) body; } __attribute__((packed)); diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c index d43cebdea1ee..226ba830f383 100644 --- a/drivers/gpu/drm/i915/intel_hdmi.c +++ b/drivers/gpu/drm/i915/intel_hdmi.c @@ -201,6 +201,21 @@ static void intel_hdmi_set_avi_infoframe(struct drm_encoder *encoder) intel_set_infoframe(encoder, &avi_if); } +static void intel_hdmi_set_spd_infoframe(struct drm_encoder *encoder) +{ + struct dip_infoframe spd_if; + + memset(&spd_if, 0, sizeof(spd_if)); + spd_if.type = DIP_TYPE_SPD; + spd_if.ver = DIP_VERSION_SPD; + spd_if.len = DIP_LEN_SPD; + strcpy(spd_if.body.spd.vn, "Intel"); + strcpy(spd_if.body.spd.pd, "Integrated gfx"); + spd_if.body.spd.sdi = DIP_SPD_PC; + + intel_set_infoframe(encoder, &spd_if); +} + static void intel_hdmi_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) @@ -245,6 +260,7 @@ static void intel_hdmi_mode_set(struct drm_encoder *encoder, POSTING_READ(intel_hdmi->sdvox_reg); intel_hdmi_set_avi_infoframe(encoder); + intel_hdmi_set_spd_infoframe(encoder); } static void intel_hdmi_dpms(struct drm_encoder *encoder, int mode) From 07b7ddd9b7f17a567e3ac2b33a4dffcb2a4524e0 Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Wed, 3 Aug 2011 11:28:44 -0700 Subject: [PATCH 315/439] drm/i915: allow cache sharing policy control Expose the SNB+ cache sharing policy register in debugfs. The new file, i915_cache_sharing, has 4 values, 0-3, with 0 being "max uncore resources" and 3 being the minimum. Exposing this control should make benchmarking easier and help us choose a good default. Signed-off-by: Jesse Barnes Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/i915_debugfs.c | 99 +++++++++++++++++++++++++++++ drivers/gpu/drm/i915/i915_reg.h | 8 +++ 2 files changed, 107 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index 5bf7bf57c641..a8ab6263e0d7 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -1408,6 +1408,85 @@ static const struct file_operations i915_max_freq_fops = { .llseek = default_llseek, }; +static int +i915_cache_sharing_open(struct inode *inode, + struct file *filp) +{ + filp->private_data = inode->i_private; + return 0; +} + +static ssize_t +i915_cache_sharing_read(struct file *filp, + char __user *ubuf, + size_t max, + loff_t *ppos) +{ + struct drm_device *dev = filp->private_data; + drm_i915_private_t *dev_priv = dev->dev_private; + char buf[80]; + u32 snpcr; + int len; + + mutex_lock(&dev_priv->dev->struct_mutex); + snpcr = I915_READ(GEN6_MBCUNIT_SNPCR); + mutex_unlock(&dev_priv->dev->struct_mutex); + + len = snprintf(buf, sizeof (buf), + "%d\n", (snpcr & GEN6_MBC_SNPCR_MASK) >> + GEN6_MBC_SNPCR_SHIFT); + + if (len > sizeof (buf)) + len = sizeof (buf); + + return simple_read_from_buffer(ubuf, max, ppos, buf, len); +} + +static ssize_t +i915_cache_sharing_write(struct file *filp, + const char __user *ubuf, + size_t cnt, + loff_t *ppos) +{ + struct drm_device *dev = filp->private_data; + struct drm_i915_private *dev_priv = dev->dev_private; + char buf[20]; + u32 snpcr; + int val = 1; + + if (cnt > 0) { + if (cnt > sizeof (buf) - 1) + return -EINVAL; + + if (copy_from_user(buf, ubuf, cnt)) + return -EFAULT; + buf[cnt] = 0; + + val = simple_strtoul(buf, NULL, 0); + } + + if (val < 0 || val > 3) + return -EINVAL; + + DRM_DEBUG_DRIVER("Manually setting uncore sharing to %d\n", val); + + /* Update the cache sharing policy here as well */ + snpcr = I915_READ(GEN6_MBCUNIT_SNPCR); + snpcr &= ~GEN6_MBC_SNPCR_MASK; + snpcr |= (val << GEN6_MBC_SNPCR_SHIFT); + I915_WRITE(GEN6_MBCUNIT_SNPCR, snpcr); + + return cnt; +} + +static const struct file_operations i915_cache_sharing_fops = { + .owner = THIS_MODULE, + .open = i915_cache_sharing_open, + .read = i915_cache_sharing_read, + .write = i915_cache_sharing_write, + .llseek = default_llseek, +}; + /* As the drm_debugfs_init() routines are called before dev->dev_private is * allocated we need to hook into the minor for release. */ static int @@ -1522,6 +1601,21 @@ static int i915_max_freq_create(struct dentry *root, struct drm_minor *minor) return drm_add_fake_info_node(minor, ent, &i915_max_freq_fops); } +static int i915_cache_sharing_create(struct dentry *root, struct drm_minor *minor) +{ + struct drm_device *dev = minor->dev; + struct dentry *ent; + + ent = debugfs_create_file("i915_cache_sharing", + S_IRUGO | S_IWUSR, + root, dev, + &i915_cache_sharing_fops); + if (IS_ERR(ent)) + return PTR_ERR(ent); + + return drm_add_fake_info_node(minor, ent, &i915_cache_sharing_fops); +} + static struct drm_info_list i915_debugfs_list[] = { {"i915_capabilities", i915_capabilities, 0}, {"i915_gem_objects", i915_gem_object_info, 0}, @@ -1576,6 +1670,9 @@ int i915_debugfs_init(struct drm_minor *minor) if (ret) return ret; ret = i915_max_freq_create(minor->debugfs_root, minor); + if (ret) + return ret; + ret = i915_cache_sharing_create(minor->debugfs_root, minor); if (ret) return ret; @@ -1594,6 +1691,8 @@ void i915_debugfs_cleanup(struct drm_minor *minor) 1, minor); drm_debugfs_remove_files((struct drm_info_list *) &i915_max_freq_fops, 1, minor); + drm_debugfs_remove_files((struct drm_info_list *) &i915_cache_sharing_fops, + 1, minor); } #endif /* CONFIG_DEBUG_FS */ diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 56cae72a5207..d1331f771e2f 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -78,6 +78,14 @@ #define GRDOM_RENDER (1<<2) #define GRDOM_MEDIA (3<<2) +#define GEN6_MBCUNIT_SNPCR 0x900c /* for LLC config */ +#define GEN6_MBC_SNPCR_SHIFT 21 +#define GEN6_MBC_SNPCR_MASK (3<<21) +#define GEN6_MBC_SNPCR_MAX (0<<21) +#define GEN6_MBC_SNPCR_MED (1<<21) +#define GEN6_MBC_SNPCR_LOW (2<<21) +#define GEN6_MBC_SNPCR_MIN (3<<21) /* only 1/16th of the cache is shared */ + #define GEN6_GDRST 0x941c #define GEN6_GRDOM_FULL (1 << 0) #define GEN6_GRDOM_RENDER (1 << 1) From 33a30ed4bdccd95ed84a1a20c1fef8ac89788ce5 Mon Sep 17 00:00:00 2001 From: Vasiliy Kulikov Date: Wed, 3 Aug 2011 22:26:55 +0400 Subject: [PATCH 316/439] shm: fix wrong tests Commit 4c677e2eefdb ("shm: optimize locking and ipc_namespace getting") introduced a copy-paste bug. Due to the bug cycle optimizations were disabled. Signed-off-by: Vasiliy Kulikov Signed-off-by: Linus Torvalds --- ipc/shm.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/ipc/shm.c b/ipc/shm.c index 9fb044f3b345..7efff043ffbf 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -294,7 +294,7 @@ static int shm_try_destroy_orphaned(int id, void *p, void *data) void shm_destroy_orphaned(struct ipc_namespace *ns) { down_write(&shm_ids(ns).rw_mutex); - if (&shm_ids(ns).in_use) + if (shm_ids(ns).in_use) idr_for_each(&shm_ids(ns).ipcs_idr, &shm_try_destroy_orphaned, ns); up_write(&shm_ids(ns).rw_mutex); } @@ -306,7 +306,7 @@ void exit_shm(struct task_struct *task) /* Destroy all already created segments, but not mapped yet */ down_write(&shm_ids(ns).rw_mutex); - if (&shm_ids(ns).in_use) + if (shm_ids(ns).in_use) idr_for_each(&shm_ids(ns).ipcs_idr, &shm_try_destroy_current, ns); up_write(&shm_ids(ns).rw_mutex); } From 298507d4d2cff2248e84afcf646b697301294442 Mon Sep 17 00:00:00 2001 From: Vasiliy Kulikov Date: Wed, 3 Aug 2011 22:28:26 +0400 Subject: [PATCH 317/439] shm: optimize exit_shm() We may optimistically check .in_use == 0 without holding the rw_mutex: it's the common case, and if it's zero, there certainly won't be any segments associated with us. After taking the lock, the idr_for_each() will do the right thing, so we could now drop the re-check inside the lock without any real cost. But it won't hurt. Signed-off-by: Vasiliy Kulikov Signed-off-by: Linus Torvalds --- ipc/shm.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/ipc/shm.c b/ipc/shm.c index 7efff043ffbf..b5bae9d945b6 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -304,6 +304,9 @@ void exit_shm(struct task_struct *task) { struct ipc_namespace *ns = task->nsproxy->ipc_ns; + if (shm_ids(ns).in_use == 0) + return; + /* Destroy all already created segments, but not mapped yet */ down_write(&shm_ids(ns).rw_mutex); if (shm_ids(ns).in_use) From 5ee5a07ce3a54de3d1192f8c9c2378d51a51e3bd Mon Sep 17 00:00:00 2001 From: Jon Mason Date: Wed, 3 Aug 2011 06:42:42 +0000 Subject: [PATCH 318/439] irda: use PCI_VENDOR_ID_* Use PCI_VENDOR_ID_* from pci_ids.h instead of creating #define locally. Signed-off-by: Jon Mason Signed-off-by: David S. Miller --- drivers/net/irda/smsc-ircc2.c | 18 ++++++++---------- 1 file changed, 8 insertions(+), 10 deletions(-) diff --git a/drivers/net/irda/smsc-ircc2.c b/drivers/net/irda/smsc-ircc2.c index 954f6e938fb7..8b1c3484d271 100644 --- a/drivers/net/irda/smsc-ircc2.c +++ b/drivers/net/irda/smsc-ircc2.c @@ -2405,8 +2405,6 @@ static int __init smsc_superio_lpc(unsigned short cfg_base) * addresses making a subsystem device table necessary. */ #ifdef CONFIG_PCI -#define PCIID_VENDOR_INTEL 0x8086 -#define PCIID_VENDOR_ALI 0x10b9 static struct smsc_ircc_subsystem_configuration subsystem_configurations[] __initdata = { /* * Subsystems needing entries: @@ -2416,7 +2414,7 @@ static struct smsc_ircc_subsystem_configuration subsystem_configurations[] __ini */ { /* Guessed entry */ - .vendor = PCIID_VENDOR_INTEL, /* Intel 82801DBM LPC bridge */ + .vendor = PCI_VENDOR_ID_INTEL, /* Intel 82801DBM LPC bridge */ .device = 0x24cc, .subvendor = 0x103c, .subdevice = 0x08bc, @@ -2429,7 +2427,7 @@ static struct smsc_ircc_subsystem_configuration subsystem_configurations[] __ini .name = "HP nx5000 family", }, { - .vendor = PCIID_VENDOR_INTEL, /* Intel 82801DBM LPC bridge */ + .vendor = PCI_VENDOR_ID_INTEL, /* Intel 82801DBM LPC bridge */ .device = 0x24cc, .subvendor = 0x103c, .subdevice = 0x088c, @@ -2443,7 +2441,7 @@ static struct smsc_ircc_subsystem_configuration subsystem_configurations[] __ini .name = "HP nc8000 family", }, { - .vendor = PCIID_VENDOR_INTEL, /* Intel 82801DBM LPC bridge */ + .vendor = PCI_VENDOR_ID_INTEL, /* Intel 82801DBM LPC bridge */ .device = 0x24cc, .subvendor = 0x103c, .subdevice = 0x0890, @@ -2456,7 +2454,7 @@ static struct smsc_ircc_subsystem_configuration subsystem_configurations[] __ini .name = "HP nc6000 family", }, { - .vendor = PCIID_VENDOR_INTEL, /* Intel 82801DBM LPC bridge */ + .vendor = PCI_VENDOR_ID_INTEL, /* Intel 82801DBM LPC bridge */ .device = 0x24cc, .subvendor = 0x0e11, .subdevice = 0x0860, @@ -2471,7 +2469,7 @@ static struct smsc_ircc_subsystem_configuration subsystem_configurations[] __ini }, { /* Intel 82801DB/DBL (ICH4/ICH4-L) LPC Interface Bridge */ - .vendor = PCIID_VENDOR_INTEL, + .vendor = PCI_VENDOR_ID_INTEL, .device = 0x24c0, .subvendor = 0x1179, .subdevice = 0xffff, /* 0xffff is "any" */ @@ -2484,7 +2482,7 @@ static struct smsc_ircc_subsystem_configuration subsystem_configurations[] __ini .name = "Toshiba laptop with Intel 82801DB/DBL LPC bridge", }, { - .vendor = PCIID_VENDOR_INTEL, /* Intel 82801CAM ISA bridge */ + .vendor = PCI_VENDOR_ID_INTEL, /* Intel 82801CAM ISA bridge */ .device = 0x248c, .subvendor = 0x1179, .subdevice = 0xffff, /* 0xffff is "any" */ @@ -2498,7 +2496,7 @@ static struct smsc_ircc_subsystem_configuration subsystem_configurations[] __ini }, { /* 82801DBM (ICH4-M) LPC Interface Bridge */ - .vendor = PCIID_VENDOR_INTEL, + .vendor = PCI_VENDOR_ID_INTEL, .device = 0x24cc, .subvendor = 0x1179, .subdevice = 0xffff, /* 0xffff is "any" */ @@ -2512,7 +2510,7 @@ static struct smsc_ircc_subsystem_configuration subsystem_configurations[] __ini }, { /* ALi M1533/M1535 PCI to ISA Bridge [Aladdin IV/V/V+] */ - .vendor = PCIID_VENDOR_ALI, + .vendor = PCI_VENDOR_ID_AL, .device = 0x1533, .subvendor = 0x1179, .subdevice = 0xffff, /* 0xffff is "any" */ From d0b0c8c79b26a1f48e1e92a2c073655159e72b7c Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Fri, 22 Jul 2011 14:20:46 -0700 Subject: [PATCH 319/439] ioat: fix xor_idx_to_desc For versions of the device that implement operation-types 0x87, 0x88 (IOAT_OP_XOR, IOAT_OP_XOR_VAL) this map determines whether a given source is located in the base or extended descriptor. Source addresses 6 through 8 require an extended descriptor, hence 0xe0, not 0xd0. No shipping hardware currently implements these operation types. Reported-by: Evgueni Smogailov Signed-off-by: Dan Williams --- drivers/dma/ioat/dma_v3.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/dma/ioat/dma_v3.c b/drivers/dma/ioat/dma_v3.c index fec8664f2f23..f519c93a61e7 100644 --- a/drivers/dma/ioat/dma_v3.c +++ b/drivers/dma/ioat/dma_v3.c @@ -73,7 +73,7 @@ /* provide a lookup table for setting the source address in the base or * extended descriptor of an xor or pq descriptor */ -static const u8 xor_idx_to_desc = 0xd0; +static const u8 xor_idx_to_desc = 0xe0; static const u8 xor_idx_to_field[] = { 1, 4, 5, 6, 7, 0, 1, 2 }; static const u8 pq_idx_to_desc = 0xf8; static const u8 pq_idx_to_field[] = { 1, 4, 5, 0, 1, 2, 4, 5 }; From 21ef4b8b7a7d59a995bf44382de38c95587767d4 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Wed, 20 Jul 2011 11:32:28 +0800 Subject: [PATCH 320/439] dmaengine: use DEFINE_IDR for static initialization We could use DEFINE_IDR for statically allocated idr that allow us to save a few lines of code. And also remove unneeded mutex_init() for dma_list_mutex, as dma_list_mutex is initialized automatically by DEFINE_MUTEX(). Signed-off-by: Axel Lin Signed-off-by: Dan Williams --- drivers/dma/dmaengine.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c index 8bcb15fb959d..bc11eebff44d 100644 --- a/drivers/dma/dmaengine.c +++ b/drivers/dma/dmaengine.c @@ -61,9 +61,9 @@ #include static DEFINE_MUTEX(dma_list_mutex); +static DEFINE_IDR(dma_idr); static LIST_HEAD(dma_device_list); static long dmaengine_ref_count; -static struct idr dma_idr; /* --- sysfs implementation --- */ @@ -1049,8 +1049,6 @@ EXPORT_SYMBOL_GPL(dma_run_dependencies); static int __init dma_bus_init(void) { - idr_init(&dma_idr); - mutex_init(&dma_list_mutex); return class_register(&dma_devclass); } arch_initcall(dma_bus_init); From 33f35f2a4ee3abfc0f87990058aa1b6b5092f725 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 3 Aug 2011 22:00:38 -1000 Subject: [PATCH 321/439] x86: don't include xen/xen.h in unless XEN is enabled Dmitry Kasatkin reports: "kernel-devel package with kernel headers have no directory if XEN is disabled. Modules which inclide asm/io.h won't compile. XEN related content is behind the CONFIG_XEN flag in the io.h. And should be also behind CONFIG_XEN flag." So move the include of down into the section that is conditional on CONFIG_XEN. Reported-by: Dmitry Kasatkin Signed-off-by: Linus Torvalds --- arch/x86/include/asm/io.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/arch/x86/include/asm/io.h b/arch/x86/include/asm/io.h index d02804d650c4..d8e8eefbe24c 100644 --- a/arch/x86/include/asm/io.h +++ b/arch/x86/include/asm/io.h @@ -40,8 +40,6 @@ #include #include -#include - #define build_mmio_read(name, size, type, reg, barrier) \ static inline type name(const volatile void __iomem *addr) \ { type ret; asm volatile("mov" size " %1,%0":reg (ret) \ @@ -334,6 +332,7 @@ extern void fixup_early_ioremap(void); extern bool is_early_ioremap_ptep(pte_t *ptep); #ifdef CONFIG_XEN +#include struct bio_vec; extern bool xen_biovec_phys_mergeable(const struct bio_vec *vec1, From 288d5abec8314ae50fe6692f324b0444acae8486 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 3 Aug 2011 22:03:29 -1000 Subject: [PATCH 322/439] Boot up with usermodehelper disabled The core device layer sends tons of uevent notifications for each device it finds, and if the kernel has been built with a non-empty CONFIG_UEVENT_HELPER_PATH that will make us try to execute the usermode helper binary for all these events very early in the boot. Not only won't the root filesystem even be mounted at that point, we literally won't have necessarily even initialized all the process handling data structures at that point, which causes no end of silly problems even when the usermode helper doesn't actually succeed in executing. So just use our existing infrastructure to disable the usermodehelpers to make the kernel start out with them disabled. We enable them when we've at least initialized stuff a bit. Problems related to an uninitialized init_ipc_ns.ids[IPC_SHM_IDS].rw_mutex reported by various people. Reported-by: Manuel Lauss Reported-by: Richard Weinberger Reported-by: Marc Zyngier Acked-by: Kay Sievers Cc: Andrew Morton Cc: Vasiliy Kulikov Cc: Greg KH Signed-off-by: Linus Torvalds --- init/main.c | 5 ++++- kernel/kmod.c | 2 +- 2 files changed, 5 insertions(+), 2 deletions(-) diff --git a/init/main.c b/init/main.c index 1952d37e4ecb..9c51ee7adf3d 100644 --- a/init/main.c +++ b/init/main.c @@ -369,9 +369,12 @@ static noinline void __init_refok rest_init(void) init_idle_bootup_task(current); preempt_enable_no_resched(); schedule(); - preempt_disable(); + + /* At this point, we can enable user mode helper functionality */ + usermodehelper_enable(); /* Call into cpu_idle with preempt disabled */ + preempt_disable(); cpu_idle(); } diff --git a/kernel/kmod.c b/kernel/kmod.c index 47613dfb7b28..ddc7644c1305 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -274,7 +274,7 @@ static void __call_usermodehelper(struct work_struct *work) * (used for preventing user land processes from being created after the user * land has been frozen during a system-wide hibernation or suspend operation). */ -static int usermodehelper_disabled; +static int usermodehelper_disabled = 1; /* Number of helpers running */ static atomic_t running_helpers = ATOMIC_INIT(0); From 7d36b26be0f3c6b86e3ab7e1539e42f3a3bc79ca Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 26 Jul 2011 13:13:44 +0200 Subject: [PATCH 323/439] lockdep: Fix trace_hardirqs_on_caller() Commit dd4e5d3ac4a ("lockdep: Fix trace_[soft,hard]irqs_[on,off]() recursion") made a bit of a mess of the various checks and error conditions. In particular it moved the check for !irqs_disabled() before the spurious enable test, resulting in some warnings. Reported-by: Arnaud Lacombe Reported-by: Dave Jones Reported-and-tested-by: Sergey Senozhatsky Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1311679697.24752.28.camel@twins Signed-off-by: Ingo Molnar --- kernel/lockdep.c | 30 ++++++++++++++++-------------- 1 file changed, 16 insertions(+), 14 deletions(-) diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 3956f5149e25..74ca247a4d4f 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -2485,23 +2485,9 @@ static void __trace_hardirqs_on_caller(unsigned long ip) { struct task_struct *curr = current; - if (DEBUG_LOCKS_WARN_ON(unlikely(early_boot_irqs_disabled))) - return; - - if (unlikely(curr->hardirqs_enabled)) { - /* - * Neither irq nor preemption are disabled here - * so this is racy by nature but losing one hit - * in a stat is not a big deal. - */ - __debug_atomic_inc(redundant_hardirqs_on); - return; - } /* we'll do an OFF -> ON transition: */ curr->hardirqs_enabled = 1; - if (DEBUG_LOCKS_WARN_ON(current->hardirq_context)) - return; /* * We are going to turn hardirqs on, so set the * usage bit for all held locks: @@ -2529,9 +2515,25 @@ void trace_hardirqs_on_caller(unsigned long ip) if (unlikely(!debug_locks || current->lockdep_recursion)) return; + if (unlikely(current->hardirqs_enabled)) { + /* + * Neither irq nor preemption are disabled here + * so this is racy by nature but losing one hit + * in a stat is not a big deal. + */ + __debug_atomic_inc(redundant_hardirqs_on); + return; + } + if (DEBUG_LOCKS_WARN_ON(!irqs_disabled())) return; + if (DEBUG_LOCKS_WARN_ON(unlikely(early_boot_irqs_disabled))) + return; + + if (DEBUG_LOCKS_WARN_ON(current->hardirq_context)) + return; + current->lockdep_recursion = 1; __trace_hardirqs_on_caller(ip); current->lockdep_recursion = 0; From 70a0686a72c7a7e554b404ca11406ceec709d425 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 25 Jul 2011 12:09:59 +0200 Subject: [PATCH 324/439] lockdep: Fix up warning On Sun, 2011-07-24 at 21:06 -0400, Arnaud Lacombe wrote: > /src/linux/linux/kernel/lockdep.c: In function 'mark_held_locks': > /src/linux/linux/kernel/lockdep.c:2471:31: warning: comparison of > distinct pointer types lacks a cast The warning is harmless in this case, but the below makes it go away. Reported-by: Arnaud Lacombe Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1311588599.2617.56.camel@laptop Signed-off-by: Ingo Molnar --- kernel/lockdep.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 74ca247a4d4f..5903586f32a0 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -2468,7 +2468,7 @@ mark_held_locks(struct task_struct *curr, enum mark_type mark) BUG_ON(usage_bit >= LOCK_USAGE_STATES); - if (hlock_class(hlock)->key == &__lockdep_no_validate__) + if (hlock_class(hlock)->key == __lockdep_no_validate__.subkeys) continue; if (!mark_lock(curr, hlock, usage_bit)) From 83835b3d9aec8e9f666d8223d8a386814f756266 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 22 Jul 2011 15:26:05 +0200 Subject: [PATCH 325/439] slab, lockdep: Annotate slab -> rcu -> debug_object -> slab Lockdep thinks there's lock recursion through: kmem_cache_free() cache_flusharray() spin_lock(&l3->list_lock) <----------------. free_block() | slab_destroy() | call_rcu() | debug_object_activate() | debug_object_init() | __debug_object_init() | kmem_cache_alloc() | cache_alloc_refill() | spin_lock(&l3->list_lock) --' Now debug objects doesn't use SLAB_DESTROY_BY_RCU and hence there is no actual possibility of recursing. Luckily debug objects marks it slab with SLAB_DEBUG_OBJECTS so we can identify the thing. Mark all SLAB_DEBUG_OBJECTS (all one!) slab caches with a special lockdep key so that lockdep sees its a different cachep. Also add a WARN on trying to create a SLAB_DESTROY_BY_RCU | SLAB_DEBUG_OBJECTS cache, to avoid possible future trouble. Reported-and-tested-by: Sebastian Siewior [ fixes to the initial patch ] Reported-by: Thomas Gleixner Acked-by: Pekka Enberg Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1311341165.27400.58.camel@twins Signed-off-by: Ingo Molnar --- mm/slab.c | 86 +++++++++++++++++++++++++++++++++++++++++++------------ 1 file changed, 68 insertions(+), 18 deletions(-) diff --git a/mm/slab.c b/mm/slab.c index 95947400702b..0703578fb4d7 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -622,6 +622,51 @@ int slab_is_available(void) static struct lock_class_key on_slab_l3_key; static struct lock_class_key on_slab_alc_key; +static struct lock_class_key debugobj_l3_key; +static struct lock_class_key debugobj_alc_key; + +static void slab_set_lock_classes(struct kmem_cache *cachep, + struct lock_class_key *l3_key, struct lock_class_key *alc_key, + int q) +{ + struct array_cache **alc; + struct kmem_list3 *l3; + int r; + + l3 = cachep->nodelists[q]; + if (!l3) + return; + + lockdep_set_class(&l3->list_lock, l3_key); + alc = l3->alien; + /* + * FIXME: This check for BAD_ALIEN_MAGIC + * should go away when common slab code is taught to + * work even without alien caches. + * Currently, non NUMA code returns BAD_ALIEN_MAGIC + * for alloc_alien_cache, + */ + if (!alc || (unsigned long)alc == BAD_ALIEN_MAGIC) + return; + for_each_node(r) { + if (alc[r]) + lockdep_set_class(&alc[r]->lock, alc_key); + } +} + +static void slab_set_debugobj_lock_classes_node(struct kmem_cache *cachep, int node) +{ + slab_set_lock_classes(cachep, &debugobj_l3_key, &debugobj_alc_key, node); +} + +static void slab_set_debugobj_lock_classes(struct kmem_cache *cachep) +{ + int node; + + for_each_online_node(node) + slab_set_debugobj_lock_classes_node(cachep, node); +} + static void init_node_lock_keys(int q) { struct cache_sizes *s = malloc_sizes; @@ -630,29 +675,14 @@ static void init_node_lock_keys(int q) return; for (s = malloc_sizes; s->cs_size != ULONG_MAX; s++) { - struct array_cache **alc; struct kmem_list3 *l3; - int r; l3 = s->cs_cachep->nodelists[q]; if (!l3 || OFF_SLAB(s->cs_cachep)) continue; - lockdep_set_class(&l3->list_lock, &on_slab_l3_key); - alc = l3->alien; - /* - * FIXME: This check for BAD_ALIEN_MAGIC - * should go away when common slab code is taught to - * work even without alien caches. - * Currently, non NUMA code returns BAD_ALIEN_MAGIC - * for alloc_alien_cache, - */ - if (!alc || (unsigned long)alc == BAD_ALIEN_MAGIC) - continue; - for_each_node(r) { - if (alc[r]) - lockdep_set_class(&alc[r]->lock, - &on_slab_alc_key); - } + + slab_set_lock_classes(s->cs_cachep, &on_slab_l3_key, + &on_slab_alc_key, q); } } @@ -671,6 +701,14 @@ static void init_node_lock_keys(int q) static inline void init_lock_keys(void) { } + +static void slab_set_debugobj_lock_classes_node(struct kmem_cache *cachep, int node) +{ +} + +static void slab_set_debugobj_lock_classes(struct kmem_cache *cachep) +{ +} #endif /* @@ -1264,6 +1302,8 @@ static int __cpuinit cpuup_prepare(long cpu) spin_unlock_irq(&l3->list_lock); kfree(shared); free_alien_cache(alien); + if (cachep->flags & SLAB_DEBUG_OBJECTS) + slab_set_debugobj_lock_classes_node(cachep, node); } init_node_lock_keys(node); @@ -2426,6 +2466,16 @@ kmem_cache_create (const char *name, size_t size, size_t align, goto oops; } + if (flags & SLAB_DEBUG_OBJECTS) { + /* + * Would deadlock through slab_destroy()->call_rcu()-> + * debug_object_activate()->kmem_cache_alloc(). + */ + WARN_ON_ONCE(flags & SLAB_DESTROY_BY_RCU); + + slab_set_debugobj_lock_classes(cachep); + } + /* cache setup completed, link it into the list */ list_add(&cachep->next, &cache_chain); oops: From f59de8992aa6dc85e81aadc26b0f69e17809721d Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 14 Jul 2011 15:19:09 +0200 Subject: [PATCH 326/439] lockdep: Clear whole lockdep_map on initialization lockdep_init_map() only initializes parts of lockdep_map and triggers kmemcheck warning when it is copied as a whole. There isn't anything to be gained by clearing selectively. memset() the whole structure and remove loop for ->class_cache[] clearing. Addresses https://bugzilla.kernel.org/show_bug.cgi?id=35532 Signed-off-by: Tejun Heo Reported-and-tested-by: Christian Casteyde Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=35532 Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110714131909.GJ3455@htj.dyndns.org Signed-off-by: Ingo Molnar --- kernel/lockdep.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 5903586f32a0..8c24294e477f 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -2874,10 +2874,7 @@ static int mark_lock(struct task_struct *curr, struct held_lock *this, void lockdep_init_map(struct lockdep_map *lock, const char *name, struct lock_class_key *key, int subclass) { - int i; - - for (i = 0; i < NR_LOCKDEP_CACHING_CLASSES; i++) - lock->class_cache[i] = NULL; + memset(lock, 0, sizeof(*lock)); #ifdef CONFIG_LOCK_STAT lock->cpu = raw_smp_processor_id(); From 30765b92ada267c5395fc788623cb15233276f5c Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 28 Jul 2011 23:22:56 +0200 Subject: [PATCH 327/439] slab, lockdep: Annotate the locks before using them Fernando found we hit the regular OFF_SLAB 'recursion' before we annotate the locks, cure this. The relevant portion of the stack-trace: > [ 0.000000] [] rt_spin_lock+0x50/0x56 > [ 0.000000] [] __cache_free+0x43/0xc3 > [ 0.000000] [] kmem_cache_free+0x6c/0xdc > [ 0.000000] [] slab_destroy+0x4f/0x53 > [ 0.000000] [] free_block+0x94/0xc1 > [ 0.000000] [] do_tune_cpucache+0x10b/0x2bb > [ 0.000000] [] enable_cpucache+0x7b/0xa7 > [ 0.000000] [] kmem_cache_init_late+0x1f/0x61 > [ 0.000000] [] start_kernel+0x24c/0x363 > [ 0.000000] [] i386_start_kernel+0xa9/0xaf Reported-by: Fernando Lopez-Lezcano Acked-by: Pekka Enberg Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1311888176.2617.379.camel@laptop Signed-off-by: Ingo Molnar --- mm/slab.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/mm/slab.c b/mm/slab.c index 0703578fb4d7..6d90a091fdca 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -1666,6 +1666,9 @@ void __init kmem_cache_init_late(void) { struct kmem_cache *cachep; + /* Annotate slab for lockdep -- annotate the malloc caches */ + init_lock_keys(); + /* 6) resize the head arrays to their final sizes */ mutex_lock(&cache_chain_mutex); list_for_each_entry(cachep, &cache_chain, next) @@ -1676,9 +1679,6 @@ void __init kmem_cache_init_late(void) /* Done! */ g_cpucache_up = FULL; - /* Annotate slab for lockdep -- annotate the malloc caches */ - init_lock_keys(); - /* * Register a cpu startup notifier callback that initializes * cpu_cache_get for all new cpus From 89e9aad65ffad96c3a35ff979a61a65761873951 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 4 Aug 2011 01:29:51 -0700 Subject: [PATCH 328/439] ide: Fix irq flags madness commit ec1a123 (IDE: pass IRQ flags to the IDE core) introduced the bogosity of passing unfiltered resource->flags to the irq_flags which are used for request_irq. It results in random bits set (especially IORESOURCE_IRQ which maps to IRQF_PER_CPU). Filter the bits proper. Signed-off-by: Thomas Gleixner Signed-off-by: David S. Miller --- drivers/ide/ide_platform.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/ide/ide_platform.c b/drivers/ide/ide_platform.c index 542603b394e4..962693b10a1c 100644 --- a/drivers/ide/ide_platform.c +++ b/drivers/ide/ide_platform.c @@ -19,6 +19,7 @@ #include #include #include +#include #include static void __devinit plat_ide_setup_ports(struct ide_hw *hw, @@ -95,7 +96,10 @@ static int __devinit plat_ide_probe(struct platform_device *pdev) plat_ide_setup_ports(&hw, base, alt_base, pdata, res_irq->start); hw.dev = &pdev->dev; - d.irq_flags = res_irq->flags; + d.irq_flags = res_irq->flags & IRQF_TRIGGER_MASK; + if (res_irq->flags & IORESOURCE_IRQ_SHAREABLE) + d.irq_flags |= IRQF_SHARED; + if (mmio) d.host_flags |= IDE_HFLAG_MMIO; From 0302899e144296d6ce8cb3679a9a42d5c6436910 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Thu, 4 Aug 2011 01:30:34 -0700 Subject: [PATCH 329/439] drivers/ide/cy82c693.c: Add missing pci_dev_put Pci_get_slot calls pci_dev_get, so pci_dev_put is needed before leaving the function in the case where pci_get_slot is locally used. The semantic match that finds this problem is as follows: (http://coccinelle.lip6.fr/) // @@ local idexpression x; expression e; @@ *x = pci_get_slot(...) ... when != true x == NULL when != pci_dev_put(x) when != e = x when != if (x != NULL) {<+... pci_dev_put(x); ...+>} *return ...; // Signed-off-by: Julia Lawall Signed-off-by: David S. Miller --- drivers/ide/cy82c693.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/ide/cy82c693.c b/drivers/ide/cy82c693.c index 3be60da52123..67cbcfa35122 100644 --- a/drivers/ide/cy82c693.c +++ b/drivers/ide/cy82c693.c @@ -141,6 +141,8 @@ static void cy82c693_set_pio_mode(ide_hwif_t *hwif, ide_drive_t *drive) pci_write_config_byte(dev, CY82_IDE_SLAVE_IOW, time_16); pci_write_config_byte(dev, CY82_IDE_SLAVE_8BIT, time_8); } + if (hwif->index > 0) + pci_dev_put(dev); } static void __devinit init_iops_cy82c693(ide_hwif_t *hwif) From fabb5bd96d060da961d74531a0900da4da5ad2d4 Mon Sep 17 00:00:00 2001 From: Sam Ravnborg Date: Thu, 4 Aug 2011 01:35:12 -0700 Subject: [PATCH 330/439] sparc: use kbuild-generic support for true asm-generic header files Signed-off-by: Sam Ravnborg Signed-off-by: David S. Miller --- arch/sparc/include/asm/Kbuild | 5 +++++ arch/sparc/include/asm/div64.h | 1 - arch/sparc/include/asm/irq_regs.h | 1 - arch/sparc/include/asm/local.h | 6 ------ arch/sparc/include/asm/local64.h | 1 - 5 files changed, 5 insertions(+), 9 deletions(-) delete mode 100644 arch/sparc/include/asm/div64.h delete mode 100644 arch/sparc/include/asm/irq_regs.h delete mode 100644 arch/sparc/include/asm/local.h delete mode 100644 arch/sparc/include/asm/local64.h diff --git a/arch/sparc/include/asm/Kbuild b/arch/sparc/include/asm/Kbuild index 3c93f08ce187..2c2e38821f60 100644 --- a/arch/sparc/include/asm/Kbuild +++ b/arch/sparc/include/asm/Kbuild @@ -16,3 +16,8 @@ header-y += traps.h header-y += uctx.h header-y += utrap.h header-y += watchdog.h + +generic-y += div64.h +generic-y += local64.h +generic-y += irq_regs.h +generic-y += local.h diff --git a/arch/sparc/include/asm/div64.h b/arch/sparc/include/asm/div64.h deleted file mode 100644 index 6cd978cefb28..000000000000 --- a/arch/sparc/include/asm/div64.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/sparc/include/asm/irq_regs.h b/arch/sparc/include/asm/irq_regs.h deleted file mode 100644 index 3dd9c0b70270..000000000000 --- a/arch/sparc/include/asm/irq_regs.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/arch/sparc/include/asm/local.h b/arch/sparc/include/asm/local.h deleted file mode 100644 index bc80815a435c..000000000000 --- a/arch/sparc/include/asm/local.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef _SPARC_LOCAL_H -#define _SPARC_LOCAL_H - -#include - -#endif diff --git a/arch/sparc/include/asm/local64.h b/arch/sparc/include/asm/local64.h deleted file mode 100644 index 36c93b5cc239..000000000000 --- a/arch/sparc/include/asm/local64.h +++ /dev/null @@ -1 +0,0 @@ -#include From 36c35416a94f5632c3addad05217ff02c39b3b61 Mon Sep 17 00:00:00 2001 From: Giuseppe Scrivano Date: Wed, 3 Aug 2011 22:10:29 +0000 Subject: [PATCH 331/439] cdc_ncm: fix endianness problem. Fix a misusage of the struct usb_cdc_notification to pass arguments to the usb_control_msg function. The usb_control_msg function expects host endian arguments but usb_cdc_notification stores these values as little endian. Now usb_control_msg is directly invoked with host endian values. Signed-off-by: Giuseppe Scrivano Signed-off-by: David S. Miller --- drivers/net/usb/cdc_ncm.c | 156 ++++++++++++++------------------------ 1 file changed, 56 insertions(+), 100 deletions(-) diff --git a/drivers/net/usb/cdc_ncm.c b/drivers/net/usb/cdc_ncm.c index fd622a66ebbf..a03336e086d5 100644 --- a/drivers/net/usb/cdc_ncm.c +++ b/drivers/net/usb/cdc_ncm.c @@ -53,7 +53,7 @@ #include #include -#define DRIVER_VERSION "01-June-2011" +#define DRIVER_VERSION "04-Aug-2011" /* CDC NCM subclass 3.2.1 */ #define USB_CDC_NCM_NDP16_LENGTH_MIN 0x10 @@ -163,35 +163,8 @@ cdc_ncm_get_drvinfo(struct net_device *net, struct ethtool_drvinfo *info) usb_make_path(dev->udev, info->bus_info, sizeof(info->bus_info)); } -static int -cdc_ncm_do_request(struct cdc_ncm_ctx *ctx, struct usb_cdc_notification *req, - void *data, u16 flags, u16 *actlen, u16 timeout) -{ - int err; - - err = usb_control_msg(ctx->udev, (req->bmRequestType & USB_DIR_IN) ? - usb_rcvctrlpipe(ctx->udev, 0) : - usb_sndctrlpipe(ctx->udev, 0), - req->bNotificationType, req->bmRequestType, - req->wValue, - req->wIndex, data, - req->wLength, timeout); - - if (err < 0) { - if (actlen) - *actlen = 0; - return err; - } - - if (actlen) - *actlen = err; - - return 0; -} - static u8 cdc_ncm_setup(struct cdc_ncm_ctx *ctx) { - struct usb_cdc_notification req; u32 val; u8 flags; u8 iface_no; @@ -200,14 +173,14 @@ static u8 cdc_ncm_setup(struct cdc_ncm_ctx *ctx) iface_no = ctx->control->cur_altsetting->desc.bInterfaceNumber; - req.bmRequestType = USB_TYPE_CLASS | USB_DIR_IN | USB_RECIP_INTERFACE; - req.bNotificationType = USB_CDC_GET_NTB_PARAMETERS; - req.wValue = 0; - req.wIndex = cpu_to_le16(iface_no); - req.wLength = cpu_to_le16(sizeof(ctx->ncm_parm)); - - err = cdc_ncm_do_request(ctx, &req, &ctx->ncm_parm, 0, NULL, 1000); - if (err) { + err = usb_control_msg(ctx->udev, + usb_rcvctrlpipe(ctx->udev, 0), + USB_CDC_GET_NTB_PARAMETERS, + USB_TYPE_CLASS | USB_DIR_IN + | USB_RECIP_INTERFACE, + 0, iface_no, &ctx->ncm_parm, + sizeof(ctx->ncm_parm), 10000); + if (err < 0) { pr_debug("failed GET_NTB_PARAMETERS\n"); return 1; } @@ -253,31 +226,26 @@ static u8 cdc_ncm_setup(struct cdc_ncm_ctx *ctx) /* inform device about NTB input size changes */ if (ctx->rx_max != le32_to_cpu(ctx->ncm_parm.dwNtbInMaxSize)) { - req.bmRequestType = USB_TYPE_CLASS | USB_DIR_OUT | - USB_RECIP_INTERFACE; - req.bNotificationType = USB_CDC_SET_NTB_INPUT_SIZE; - req.wValue = 0; - req.wIndex = cpu_to_le16(iface_no); if (flags & USB_CDC_NCM_NCAP_NTB_INPUT_SIZE) { struct usb_cdc_ncm_ndp_input_size ndp_in_sz; - - req.wLength = 8; - ndp_in_sz.dwNtbInMaxSize = cpu_to_le32(ctx->rx_max); - ndp_in_sz.wNtbInMaxDatagrams = - cpu_to_le16(CDC_NCM_DPT_DATAGRAMS_MAX); - ndp_in_sz.wReserved = 0; - err = cdc_ncm_do_request(ctx, &req, &ndp_in_sz, 0, NULL, - 1000); + err = usb_control_msg(ctx->udev, + usb_sndctrlpipe(ctx->udev, 0), + USB_CDC_SET_NTB_INPUT_SIZE, + USB_TYPE_CLASS | USB_DIR_OUT + | USB_RECIP_INTERFACE, + 0, iface_no, &ndp_in_sz, 8, 1000); } else { __le32 dwNtbInMaxSize = cpu_to_le32(ctx->rx_max); - - req.wLength = 4; - err = cdc_ncm_do_request(ctx, &req, &dwNtbInMaxSize, 0, - NULL, 1000); + err = usb_control_msg(ctx->udev, + usb_sndctrlpipe(ctx->udev, 0), + USB_CDC_SET_NTB_INPUT_SIZE, + USB_TYPE_CLASS | USB_DIR_OUT + | USB_RECIP_INTERFACE, + 0, iface_no, &dwNtbInMaxSize, 4, 1000); } - if (err) + if (err < 0) pr_debug("Setting NTB Input Size failed\n"); } @@ -332,29 +300,24 @@ static u8 cdc_ncm_setup(struct cdc_ncm_ctx *ctx) /* set CRC Mode */ if (flags & USB_CDC_NCM_NCAP_CRC_MODE) { - req.bmRequestType = USB_TYPE_CLASS | USB_DIR_OUT | - USB_RECIP_INTERFACE; - req.bNotificationType = USB_CDC_SET_CRC_MODE; - req.wValue = cpu_to_le16(USB_CDC_NCM_CRC_NOT_APPENDED); - req.wIndex = cpu_to_le16(iface_no); - req.wLength = 0; - - err = cdc_ncm_do_request(ctx, &req, NULL, 0, NULL, 1000); - if (err) + err = usb_control_msg(ctx->udev, usb_sndctrlpipe(ctx->udev, 0), + USB_CDC_SET_CRC_MODE, + USB_TYPE_CLASS | USB_DIR_OUT + | USB_RECIP_INTERFACE, + USB_CDC_NCM_CRC_NOT_APPENDED, + iface_no, NULL, 0, 1000); + if (err < 0) pr_debug("Setting CRC mode off failed\n"); } /* set NTB format, if both formats are supported */ if (ntb_fmt_supported & USB_CDC_NCM_NTH32_SIGN) { - req.bmRequestType = USB_TYPE_CLASS | USB_DIR_OUT | - USB_RECIP_INTERFACE; - req.bNotificationType = USB_CDC_SET_NTB_FORMAT; - req.wValue = cpu_to_le16(USB_CDC_NCM_NTB16_FORMAT); - req.wIndex = cpu_to_le16(iface_no); - req.wLength = 0; - - err = cdc_ncm_do_request(ctx, &req, NULL, 0, NULL, 1000); - if (err) + err = usb_control_msg(ctx->udev, usb_sndctrlpipe(ctx->udev, 0), + USB_CDC_SET_NTB_FORMAT, USB_TYPE_CLASS + | USB_DIR_OUT | USB_RECIP_INTERFACE, + USB_CDC_NCM_NTB16_FORMAT, + iface_no, NULL, 0, 1000); + if (err < 0) pr_debug("Setting NTB format to 16-bit failed\n"); } @@ -364,17 +327,13 @@ static u8 cdc_ncm_setup(struct cdc_ncm_ctx *ctx) if (flags & USB_CDC_NCM_NCAP_MAX_DATAGRAM_SIZE) { __le16 max_datagram_size; u16 eth_max_sz = le16_to_cpu(ctx->ether_desc->wMaxSegmentSize); - - req.bmRequestType = USB_TYPE_CLASS | USB_DIR_IN | - USB_RECIP_INTERFACE; - req.bNotificationType = USB_CDC_GET_MAX_DATAGRAM_SIZE; - req.wValue = 0; - req.wIndex = cpu_to_le16(iface_no); - req.wLength = cpu_to_le16(2); - - err = cdc_ncm_do_request(ctx, &req, &max_datagram_size, 0, NULL, - 1000); - if (err) { + err = usb_control_msg(ctx->udev, usb_rcvctrlpipe(ctx->udev, 0), + USB_CDC_GET_MAX_DATAGRAM_SIZE, + USB_TYPE_CLASS | USB_DIR_IN + | USB_RECIP_INTERFACE, + 0, iface_no, &max_datagram_size, + 2, 1000); + if (err < 0) { pr_debug("GET_MAX_DATAGRAM_SIZE failed, use size=%u\n", CDC_NCM_MIN_DATAGRAM_SIZE); } else { @@ -395,17 +354,15 @@ static u8 cdc_ncm_setup(struct cdc_ncm_ctx *ctx) CDC_NCM_MIN_DATAGRAM_SIZE; /* if value changed, update device */ - req.bmRequestType = USB_TYPE_CLASS | USB_DIR_OUT | - USB_RECIP_INTERFACE; - req.bNotificationType = USB_CDC_SET_MAX_DATAGRAM_SIZE; - req.wValue = 0; - req.wIndex = cpu_to_le16(iface_no); - req.wLength = 2; - max_datagram_size = cpu_to_le16(ctx->max_datagram_size); - - err = cdc_ncm_do_request(ctx, &req, &max_datagram_size, - 0, NULL, 1000); - if (err) + err = usb_control_msg(ctx->udev, + usb_sndctrlpipe(ctx->udev, 0), + USB_CDC_SET_MAX_DATAGRAM_SIZE, + USB_TYPE_CLASS | USB_DIR_OUT + | USB_RECIP_INTERFACE, + 0, + iface_no, &max_datagram_size, + 2, 1000); + if (err < 0) pr_debug("SET_MAX_DATAGRAM_SIZE failed\n"); } @@ -671,7 +628,7 @@ cdc_ncm_fill_tx_frame(struct cdc_ncm_ctx *ctx, struct sk_buff *skb) u32 rem; u32 offset; u32 last_offset; - u16 n = 0; + u16 n = 0, index; u8 ready2send = 0; /* if there is a remaining skb, it gets priority */ @@ -859,8 +816,8 @@ cdc_ncm_fill_tx_frame(struct cdc_ncm_ctx *ctx, struct sk_buff *skb) cpu_to_le16(sizeof(ctx->tx_ncm.nth16)); ctx->tx_ncm.nth16.wSequence = cpu_to_le16(ctx->tx_seq); ctx->tx_ncm.nth16.wBlockLength = cpu_to_le16(last_offset); - ctx->tx_ncm.nth16.wNdpIndex = ALIGN(sizeof(struct usb_cdc_ncm_nth16), - ctx->tx_ndp_modulus); + index = ALIGN(sizeof(struct usb_cdc_ncm_nth16), ctx->tx_ndp_modulus); + ctx->tx_ncm.nth16.wNdpIndex = cpu_to_le16(index); memcpy(skb_out->data, &(ctx->tx_ncm.nth16), sizeof(ctx->tx_ncm.nth16)); ctx->tx_seq++; @@ -873,12 +830,11 @@ cdc_ncm_fill_tx_frame(struct cdc_ncm_ctx *ctx, struct sk_buff *skb) ctx->tx_ncm.ndp16.wLength = cpu_to_le16(rem); ctx->tx_ncm.ndp16.wNextNdpIndex = 0; /* reserved */ - memcpy(((u8 *)skb_out->data) + ctx->tx_ncm.nth16.wNdpIndex, + memcpy(((u8 *)skb_out->data) + index, &(ctx->tx_ncm.ndp16), sizeof(ctx->tx_ncm.ndp16)); - memcpy(((u8 *)skb_out->data) + ctx->tx_ncm.nth16.wNdpIndex + - sizeof(ctx->tx_ncm.ndp16), + memcpy(((u8 *)skb_out->data) + index + sizeof(ctx->tx_ncm.ndp16), &(ctx->tx_ncm.dpe16), (ctx->tx_curr_frame_num + 1) * sizeof(struct usb_cdc_ncm_dpe16)); From d3e614577198757d5854caa912e88f2d4296479b Mon Sep 17 00:00:00 2001 From: Tord Andersson Date: Wed, 3 Aug 2011 22:11:47 +0000 Subject: [PATCH 332/439] macb: restore wrap bit when performing underrun cleanup When TX underrun occurs, a cleanup is performed that marks all buffers as used. As a side effect it also clears the wrap bit in the last buffer. This patch will restore the wrap bit. Signed-off-by: Tord Andersson Acked-by: Nicolas Ferre Signed-off-by: David S. Miller --- drivers/net/macb.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/macb.c b/drivers/net/macb.c index 0fcdc25699d8..dc4e305a1087 100644 --- a/drivers/net/macb.c +++ b/drivers/net/macb.c @@ -322,6 +322,9 @@ static void macb_tx(struct macb *bp) for (i = 0; i < TX_RING_SIZE; i++) bp->tx_ring[i].ctrl = MACB_BIT(TX_USED); + /* Add wrap bit */ + bp->tx_ring[TX_RING_SIZE - 1].ctrl |= MACB_BIT(TX_WRAP); + /* free transmit buffer in upper layer*/ for (tail = bp->tx_tail; tail != head; tail = NEXT_TX(tail)) { struct ring_info *rp = &bp->tx_skb[tail]; From a61b582954183e93a3dc3a5cf6bfd2e2c3b40aba Mon Sep 17 00:00:00 2001 From: Josip Rodin Date: Thu, 4 Aug 2011 02:47:40 -0700 Subject: [PATCH 333/439] sparc: Fix __atomic_add_unless() return value. Signed-off-by: David S. Miller --- arch/sparc/lib/atomic32.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/sparc/lib/atomic32.c b/arch/sparc/lib/atomic32.c index 8600eb2461b5..1d32b54089aa 100644 --- a/arch/sparc/lib/atomic32.c +++ b/arch/sparc/lib/atomic32.c @@ -65,7 +65,7 @@ int __atomic_add_unless(atomic_t *v, int a, int u) if (ret != u) v->counter += a; spin_unlock_irqrestore(ATOMIC_HASH(v), flags); - return ret != u; + return ret; } EXPORT_SYMBOL(__atomic_add_unless); From 9e191b22c91873c09c722d7f956ab9f3276f9a37 Mon Sep 17 00:00:00 2001 From: Grant Likely Date: Thu, 4 Aug 2011 10:47:40 +0100 Subject: [PATCH 334/439] dt: remove of_alias_get_id() reference of_alias_get_id() is broken and being reverted. Remove the reference to it and replace with a single incrementing id number. There is no risk of regression here on the imx driver since the imx change to use of_alias_get_id() is commit 22698aa2, "serial/imx: add device tree probe support" which is new for v3.1, and it won't get used unless CONFIG_OF is enabled and the board is booted using a device tree. A single incrementing integer is sufficient for now. Signed-off-by: Grant Likely Acked-by: Shawn Guo --- drivers/tty/serial/imx.c | 13 ++++--------- 1 file changed, 4 insertions(+), 9 deletions(-) diff --git a/drivers/tty/serial/imx.c b/drivers/tty/serial/imx.c index 827db7654594..7e91b3d368cd 100644 --- a/drivers/tty/serial/imx.c +++ b/drivers/tty/serial/imx.c @@ -1286,22 +1286,17 @@ static int serial_imx_resume(struct platform_device *dev) static int serial_imx_probe_dt(struct imx_port *sport, struct platform_device *pdev) { + static int portnum = 0; struct device_node *np = pdev->dev.of_node; const struct of_device_id *of_id = of_match_device(imx_uart_dt_ids, &pdev->dev); - int ret; if (!np) return -ENODEV; - ret = of_alias_get_id(np, "serial"); - if (ret < 0) { - pr_err("%s: failed to get alias id, errno %d\n", - __func__, ret); - return -ENODEV; - } else { - sport->port.line = ret; - } + sport->port.line = portnum++; + if (sport->port.line >= UART_NR) + return -EINVAL; if (of_get_property(np, "fsl,uart-has-rtscts", NULL)) sport->have_rtscts = 1; From fe55c1844a1c106e9d9d3dd27cbfcf8caeb9e77e Mon Sep 17 00:00:00 2001 From: Grant Likely Date: Thu, 4 Aug 2011 10:27:32 +0100 Subject: [PATCH 335/439] Revert "dt: add of_alias_scan and of_alias_get_id" This reverts commit 750f463a749e28464151ad26938d11b07b1c43cb. of_alias_* still needs work to be generalized for 'promtree' dt platforms, and to no implicitly create entries for available ids. Signed-off-by: Grant Likely --- drivers/of/base.c | 130 ----------------------------------------- drivers/of/fdt.c | 4 +- include/linux/of.h | 8 --- include/linux/of_fdt.h | 1 + 4 files changed, 2 insertions(+), 141 deletions(-) diff --git a/drivers/of/base.c b/drivers/of/base.c index fb28b5af733b..3ff22e32b602 100644 --- a/drivers/of/base.c +++ b/drivers/of/base.c @@ -17,39 +17,14 @@ * as published by the Free Software Foundation; either version * 2 of the License, or (at your option) any later version. */ -#include #include #include #include #include #include -/** - * struct alias_prop - Alias property in 'aliases' node - * @link: List node to link the structure in aliases_lookup list - * @alias: Alias property name - * @np: Pointer to device_node that the alias stands for - * @id: Index value from end of alias name - * @stem: Alias string without the index - * - * The structure represents one alias property of 'aliases' node as - * an entry in aliases_lookup list. - */ -struct alias_prop { - struct list_head link; - const char *alias; - struct device_node *np; - int id; - char stem[0]; -}; - -static LIST_HEAD(aliases_lookup); - struct device_node *allnodes; struct device_node *of_chosen; -struct device_node *of_aliases; - -static DEFINE_MUTEX(of_aliases_mutex); /* use when traversing tree through the allnext, child, sibling, * or parent members of struct device_node. @@ -1013,108 +988,3 @@ out_unlock: } #endif /* defined(CONFIG_OF_DYNAMIC) */ -static void of_alias_add(struct alias_prop *ap, struct device_node *np, - int id, const char *stem, int stem_len) -{ - ap->id = id; - ap->np = np; - strncpy(ap->stem, stem, stem_len); - ap->stem[stem_len] = 0; - list_add_tail(&ap->link, &aliases_lookup); - pr_debug("adding DT alias:%s: stem=%s id=%i node=%s\n", - ap->alias, ap->stem, ap->id, np ? np->full_name : NULL); -} - -/** - * of_alias_scan() - Scan all properties of 'aliases' node - * - * The function scans all the properties of 'aliases' node and populate - * the global lookup table with the properties. It returns the - * number of alias_prop found, or error code in error case. - */ -__init void of_alias_scan(void) -{ - struct property *pp; - - if (!of_aliases) - return; - - for_each_property(pp, of_aliases->properties) { - const char *start = pp->name; - const char *end = start + strlen(start); - struct device_node *np; - struct alias_prop *ap; - int id, len; - - /* Skip those we do not want to proceed */ - if (!strcmp(pp->name, "name") || - !strcmp(pp->name, "phandle") || - !strcmp(pp->name, "linux,phandle")) - continue; - - np = of_find_node_by_path(pp->value); - if (!np) - continue; - - /* walk alias backwards to extract the id and 'stem' string */ - while (isdigit(*(end-1)) && end > start) - end--; - len = end - start; - id = strlen(end) ? simple_strtoul(end, NULL, 10) : -1; - - /* Allocate an alias_prop with enough space for the stem */ - ap = early_init_dt_alloc_memory_arch(sizeof(*ap) + len + 1, 4); - if (!ap) - continue; - ap->alias = start; - of_alias_add(ap, np, id, start, len); - } -} - -/** - * of_alias_get_id() - Get alias id for the given device_node - * @np: Pointer to the given device_node - * @stem: Alias stem of the given device_node - * - * The function travels the lookup table to get alias id for the given - * device_node and alias stem. It returns the alias id if find it. - * If not, dynamically creates one in the lookup table and returns it, - * or returns error code if fail to create. - */ -int of_alias_get_id(struct device_node *np, const char *stem) -{ - struct alias_prop *app; - int id = 0; - bool found = false; - - mutex_lock(&of_aliases_mutex); - list_for_each_entry(app, &aliases_lookup, link) { - if (strcmp(app->stem, stem) != 0) - continue; - - if (np == app->np) { - found = true; - id = app->id; - break; - } - - if (id <= app->id) - id = app->id + 1; - } - - /* If an id is not found, then allocate a new one */ - if (!found) { - app = kzalloc(sizeof(*app) + strlen(stem) + 1, 4); - if (!app) { - id = -ENODEV; - goto out; - } - of_alias_add(app, np, id, stem, strlen(stem)); - } - - out: - mutex_unlock(&of_aliases_mutex); - - return id; -} -EXPORT_SYMBOL_GPL(of_alias_get_id); diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c index 13d6d3a96b31..65200af29c52 100644 --- a/drivers/of/fdt.c +++ b/drivers/of/fdt.c @@ -707,12 +707,10 @@ void __init unflatten_device_tree(void) __unflatten_device_tree(initial_boot_params, &allnodes, early_init_dt_alloc_memory_arch); - /* Get pointer to "/chosen" and "/aliasas" nodes for use everywhere */ + /* Get pointer to OF "/chosen" node for use everywhere */ of_chosen = of_find_node_by_path("/chosen"); if (of_chosen == NULL) of_chosen = of_find_node_by_path("/chosen@0"); - of_aliases = of_find_node_by_path("/aliases"); - of_alias_scan(); } #endif /* CONFIG_OF_EARLY_FLATTREE */ diff --git a/include/linux/of.h b/include/linux/of.h index bc3dc6399547..0085bb01c041 100644 --- a/include/linux/of.h +++ b/include/linux/of.h @@ -68,7 +68,6 @@ struct device_node { /* Pointer for first entry in chain of all nodes. */ extern struct device_node *allnodes; extern struct device_node *of_chosen; -extern struct device_node *of_aliases; extern rwlock_t devtree_lock; static inline bool of_have_populated_dt(void) @@ -210,9 +209,6 @@ extern int of_device_is_available(const struct device_node *device); extern const void *of_get_property(const struct device_node *node, const char *name, int *lenp); -#define for_each_property(pp, properties) \ - for (pp = properties; pp != NULL; pp = pp->next) - extern int of_n_addr_cells(struct device_node *np); extern int of_n_size_cells(struct device_node *np); extern const struct of_device_id *of_match_node( @@ -225,10 +221,6 @@ extern int of_parse_phandles_with_args(struct device_node *np, const char *list_name, const char *cells_name, int index, struct device_node **out_node, const void **out_args); -extern void *early_init_dt_alloc_memory_arch(u64 size, u64 align); -extern void of_alias_scan(void); -extern int of_alias_get_id(struct device_node *np, const char *stem); - extern int of_machine_is_compatible(const char *compat); extern int prom_add_property(struct device_node* np, struct property* prop); diff --git a/include/linux/of_fdt.h b/include/linux/of_fdt.h index b74b74ffe0e7..c84d900fbbb3 100644 --- a/include/linux/of_fdt.h +++ b/include/linux/of_fdt.h @@ -97,6 +97,7 @@ extern void early_init_dt_check_for_initrd(unsigned long node); extern int early_init_dt_scan_memory(unsigned long node, const char *uname, int depth, void *data); extern void early_init_dt_add_memory_arch(u64 base, u64 size); +extern void * early_init_dt_alloc_memory_arch(u64 size, u64 align); extern u64 dt_mem_next_cell(int s, __be32 **cellp); /* From 945a51517cc0bd9e461f2018624dfc1faef9ddee Mon Sep 17 00:00:00 2001 From: Jesse Brandeburg Date: Wed, 20 Jul 2011 00:56:21 +0000 Subject: [PATCH 336/439] intel drivers: repair missing flush operations after review of all intel drivers, found several instances where drivers had the incorrect pattern of: memory mapped write(); delay(); which should always be: memory mapped write(); write flush(); /* aka memory mapped read */ delay(); explanation: The reason for including the flush is that writes can be held (posted) in PCI/PCIe bridges, but the read always has to complete synchronously and therefore has to flush all pending writes to a device. If a write is held and followed by a delay, the delay means nothing because the write may not have reached hardware (maybe even not until the next read) Signed-off-by: Jesse Brandeburg Tested-by: Aaron Brown Signed-off-by: Jeff Kirsher --- drivers/net/e1000/e1000_ethtool.c | 6 ++++++ drivers/net/e1000/e1000_hw.c | 3 +++ drivers/net/e1000e/es2lan.c | 2 ++ drivers/net/e1000e/ethtool.c | 9 +++++++++ drivers/net/e1000e/ich8lan.c | 4 ++++ drivers/net/e1000e/lib.c | 1 + drivers/net/e1000e/phy.c | 2 ++ drivers/net/igb/e1000_nvm.c | 1 + drivers/net/igb/igb_ethtool.c | 5 +++++ drivers/net/igb/igb_main.c | 2 ++ drivers/net/igbvf/netdev.c | 2 ++ drivers/net/ixgb/ixgb_ee.c | 9 +++++++++ drivers/net/ixgb/ixgb_hw.c | 2 ++ drivers/net/ixgbe/ixgbe_common.c | 1 + drivers/net/ixgbe/ixgbe_ethtool.c | 5 +++++ drivers/net/ixgbe/ixgbe_main.c | 1 + drivers/net/ixgbe/ixgbe_phy.c | 3 +++ drivers/net/ixgbe/ixgbe_x540.c | 1 + 18 files changed, 59 insertions(+) diff --git a/drivers/net/e1000/e1000_ethtool.c b/drivers/net/e1000/e1000_ethtool.c index c5f0f04219f3..5548d464261a 100644 --- a/drivers/net/e1000/e1000_ethtool.c +++ b/drivers/net/e1000/e1000_ethtool.c @@ -838,6 +838,7 @@ static int e1000_intr_test(struct e1000_adapter *adapter, u64 *data) /* Disable all the interrupts */ ew32(IMC, 0xFFFFFFFF); + E1000_WRITE_FLUSH(); msleep(10); /* Test each interrupt */ @@ -856,6 +857,7 @@ static int e1000_intr_test(struct e1000_adapter *adapter, u64 *data) adapter->test_icr = 0; ew32(IMC, mask); ew32(ICS, mask); + E1000_WRITE_FLUSH(); msleep(10); if (adapter->test_icr & mask) { @@ -873,6 +875,7 @@ static int e1000_intr_test(struct e1000_adapter *adapter, u64 *data) adapter->test_icr = 0; ew32(IMS, mask); ew32(ICS, mask); + E1000_WRITE_FLUSH(); msleep(10); if (!(adapter->test_icr & mask)) { @@ -890,6 +893,7 @@ static int e1000_intr_test(struct e1000_adapter *adapter, u64 *data) adapter->test_icr = 0; ew32(IMC, ~mask & 0x00007FFF); ew32(ICS, ~mask & 0x00007FFF); + E1000_WRITE_FLUSH(); msleep(10); if (adapter->test_icr) { @@ -901,6 +905,7 @@ static int e1000_intr_test(struct e1000_adapter *adapter, u64 *data) /* Disable all the interrupts */ ew32(IMC, 0xFFFFFFFF); + E1000_WRITE_FLUSH(); msleep(10); /* Unhook test interrupt handler */ @@ -1394,6 +1399,7 @@ static int e1000_run_loopback_test(struct e1000_adapter *adapter) if (unlikely(++k == txdr->count)) k = 0; } ew32(TDT, k); + E1000_WRITE_FLUSH(); msleep(200); time = jiffies; /* set the start time for the receive */ good_cnt = 0; diff --git a/drivers/net/e1000/e1000_hw.c b/drivers/net/e1000/e1000_hw.c index 1698622af434..8545c7aa93eb 100644 --- a/drivers/net/e1000/e1000_hw.c +++ b/drivers/net/e1000/e1000_hw.c @@ -446,6 +446,7 @@ s32 e1000_reset_hw(struct e1000_hw *hw) /* Must reset the PHY before resetting the MAC */ if ((hw->mac_type == e1000_82541) || (hw->mac_type == e1000_82547)) { ew32(CTRL, (ctrl | E1000_CTRL_PHY_RST)); + E1000_WRITE_FLUSH(); msleep(5); } @@ -3752,6 +3753,7 @@ static s32 e1000_acquire_eeprom(struct e1000_hw *hw) /* Clear SK and CS */ eecd &= ~(E1000_EECD_CS | E1000_EECD_SK); ew32(EECD, eecd); + E1000_WRITE_FLUSH(); udelay(1); } @@ -3824,6 +3826,7 @@ static void e1000_release_eeprom(struct e1000_hw *hw) eecd &= ~E1000_EECD_SK; /* Lower SCK */ ew32(EECD, eecd); + E1000_WRITE_FLUSH(); udelay(hw->eeprom.delay_usec); } else if (hw->eeprom.type == e1000_eeprom_microwire) { diff --git a/drivers/net/e1000e/es2lan.c b/drivers/net/e1000e/es2lan.c index c0ecb2d9fdb7..e4f42257c24c 100644 --- a/drivers/net/e1000e/es2lan.c +++ b/drivers/net/e1000e/es2lan.c @@ -1313,6 +1313,7 @@ static s32 e1000_read_kmrn_reg_80003es2lan(struct e1000_hw *hw, u32 offset, kmrnctrlsta = ((offset << E1000_KMRNCTRLSTA_OFFSET_SHIFT) & E1000_KMRNCTRLSTA_OFFSET) | E1000_KMRNCTRLSTA_REN; ew32(KMRNCTRLSTA, kmrnctrlsta); + e1e_flush(); udelay(2); @@ -1347,6 +1348,7 @@ static s32 e1000_write_kmrn_reg_80003es2lan(struct e1000_hw *hw, u32 offset, kmrnctrlsta = ((offset << E1000_KMRNCTRLSTA_OFFSET_SHIFT) & E1000_KMRNCTRLSTA_OFFSET) | data; ew32(KMRNCTRLSTA, kmrnctrlsta); + e1e_flush(); udelay(2); diff --git a/drivers/net/e1000e/ethtool.c b/drivers/net/e1000e/ethtool.c index cb1a3623253e..72756e4e6448 100644 --- a/drivers/net/e1000e/ethtool.c +++ b/drivers/net/e1000e/ethtool.c @@ -964,6 +964,7 @@ static int e1000_intr_test(struct e1000_adapter *adapter, u64 *data) /* Disable all the interrupts */ ew32(IMC, 0xFFFFFFFF); + e1e_flush(); usleep_range(10000, 20000); /* Test each interrupt */ @@ -996,6 +997,7 @@ static int e1000_intr_test(struct e1000_adapter *adapter, u64 *data) adapter->test_icr = 0; ew32(IMC, mask); ew32(ICS, mask); + e1e_flush(); usleep_range(10000, 20000); if (adapter->test_icr & mask) { @@ -1014,6 +1016,7 @@ static int e1000_intr_test(struct e1000_adapter *adapter, u64 *data) adapter->test_icr = 0; ew32(IMS, mask); ew32(ICS, mask); + e1e_flush(); usleep_range(10000, 20000); if (!(adapter->test_icr & mask)) { @@ -1032,6 +1035,7 @@ static int e1000_intr_test(struct e1000_adapter *adapter, u64 *data) adapter->test_icr = 0; ew32(IMC, ~mask & 0x00007FFF); ew32(ICS, ~mask & 0x00007FFF); + e1e_flush(); usleep_range(10000, 20000); if (adapter->test_icr) { @@ -1043,6 +1047,7 @@ static int e1000_intr_test(struct e1000_adapter *adapter, u64 *data) /* Disable all the interrupts */ ew32(IMC, 0xFFFFFFFF); + e1e_flush(); usleep_range(10000, 20000); /* Unhook test interrupt handler */ @@ -1276,6 +1281,7 @@ static int e1000_integrated_phy_loopback(struct e1000_adapter *adapter) E1000_CTRL_FD); /* Force Duplex to FULL */ ew32(CTRL, ctrl_reg); + e1e_flush(); udelay(500); return 0; @@ -1418,6 +1424,7 @@ static int e1000_set_82571_fiber_loopback(struct e1000_adapter *adapter) */ #define E1000_SERDES_LB_ON 0x410 ew32(SCTL, E1000_SERDES_LB_ON); + e1e_flush(); usleep_range(10000, 20000); return 0; @@ -1513,6 +1520,7 @@ static void e1000_loopback_cleanup(struct e1000_adapter *adapter) hw->phy.media_type == e1000_media_type_internal_serdes) { #define E1000_SERDES_LB_OFF 0x400 ew32(SCTL, E1000_SERDES_LB_OFF); + e1e_flush(); usleep_range(10000, 20000); break; } @@ -1592,6 +1600,7 @@ static int e1000_run_loopback_test(struct e1000_adapter *adapter) k = 0; } ew32(TDT, k); + e1e_flush(); msleep(200); time = jiffies; /* set the start time for the receive */ good_cnt = 0; diff --git a/drivers/net/e1000e/ich8lan.c b/drivers/net/e1000e/ich8lan.c index c1752124f3cd..f7a75c1531ad 100644 --- a/drivers/net/e1000e/ich8lan.c +++ b/drivers/net/e1000e/ich8lan.c @@ -283,6 +283,7 @@ static void e1000_toggle_lanphypc_value_ich8lan(struct e1000_hw *hw) ctrl |= E1000_CTRL_LANPHYPC_OVERRIDE; ctrl &= ~E1000_CTRL_LANPHYPC_VALUE; ew32(CTRL, ctrl); + e1e_flush(); udelay(10); ctrl &= ~E1000_CTRL_LANPHYPC_OVERRIDE; ew32(CTRL, ctrl); @@ -1230,9 +1231,11 @@ s32 e1000_configure_k1_ich8lan(struct e1000_hw *hw, bool k1_enable) ew32(CTRL, reg); ew32(CTRL_EXT, ctrl_ext | E1000_CTRL_EXT_SPD_BYPS); + e1e_flush(); udelay(20); ew32(CTRL, ctrl_reg); ew32(CTRL_EXT, ctrl_ext); + e1e_flush(); udelay(20); out: @@ -3090,6 +3093,7 @@ static s32 e1000_reset_hw_ich8lan(struct e1000_hw *hw) ret_val = e1000_acquire_swflag_ich8lan(hw); e_dbg("Issuing a global reset to ich8lan\n"); ew32(CTRL, (ctrl | E1000_CTRL_RST)); + /* cannot issue a flush here because it hangs the hardware */ msleep(20); if (!ret_val) diff --git a/drivers/net/e1000e/lib.c b/drivers/net/e1000e/lib.c index 65580b405942..7898a67d6505 100644 --- a/drivers/net/e1000e/lib.c +++ b/drivers/net/e1000e/lib.c @@ -1986,6 +1986,7 @@ static s32 e1000_ready_nvm_eeprom(struct e1000_hw *hw) /* Clear SK and CS */ eecd &= ~(E1000_EECD_CS | E1000_EECD_SK); ew32(EECD, eecd); + e1e_flush(); udelay(1); /* diff --git a/drivers/net/e1000e/phy.c b/drivers/net/e1000e/phy.c index 2a6ee13285b1..8666476cb9be 100644 --- a/drivers/net/e1000e/phy.c +++ b/drivers/net/e1000e/phy.c @@ -537,6 +537,7 @@ static s32 __e1000_read_kmrn_reg(struct e1000_hw *hw, u32 offset, u16 *data, kmrnctrlsta = ((offset << E1000_KMRNCTRLSTA_OFFSET_SHIFT) & E1000_KMRNCTRLSTA_OFFSET) | E1000_KMRNCTRLSTA_REN; ew32(KMRNCTRLSTA, kmrnctrlsta); + e1e_flush(); udelay(2); @@ -609,6 +610,7 @@ static s32 __e1000_write_kmrn_reg(struct e1000_hw *hw, u32 offset, u16 data, kmrnctrlsta = ((offset << E1000_KMRNCTRLSTA_OFFSET_SHIFT) & E1000_KMRNCTRLSTA_OFFSET) | data; ew32(KMRNCTRLSTA, kmrnctrlsta); + e1e_flush(); udelay(2); diff --git a/drivers/net/igb/e1000_nvm.c b/drivers/net/igb/e1000_nvm.c index 7dcd65cede56..40407124e722 100644 --- a/drivers/net/igb/e1000_nvm.c +++ b/drivers/net/igb/e1000_nvm.c @@ -285,6 +285,7 @@ static s32 igb_ready_nvm_eeprom(struct e1000_hw *hw) /* Clear SK and CS */ eecd &= ~(E1000_EECD_CS | E1000_EECD_SK); wr32(E1000_EECD, eecd); + wrfl(); udelay(1); timeout = NVM_MAX_RETRY_SPI; diff --git a/drivers/net/igb/igb_ethtool.c b/drivers/net/igb/igb_ethtool.c index ff244ce803ce..414b0225be89 100644 --- a/drivers/net/igb/igb_ethtool.c +++ b/drivers/net/igb/igb_ethtool.c @@ -1225,6 +1225,7 @@ static int igb_intr_test(struct igb_adapter *adapter, u64 *data) /* Disable all the interrupts */ wr32(E1000_IMC, ~0); + wrfl(); msleep(10); /* Define all writable bits for ICS */ @@ -1268,6 +1269,7 @@ static int igb_intr_test(struct igb_adapter *adapter, u64 *data) wr32(E1000_IMC, mask); wr32(E1000_ICS, mask); + wrfl(); msleep(10); if (adapter->test_icr & mask) { @@ -1289,6 +1291,7 @@ static int igb_intr_test(struct igb_adapter *adapter, u64 *data) wr32(E1000_IMS, mask); wr32(E1000_ICS, mask); + wrfl(); msleep(10); if (!(adapter->test_icr & mask)) { @@ -1310,6 +1313,7 @@ static int igb_intr_test(struct igb_adapter *adapter, u64 *data) wr32(E1000_IMC, ~mask); wr32(E1000_ICS, ~mask); + wrfl(); msleep(10); if (adapter->test_icr & mask) { @@ -1321,6 +1325,7 @@ static int igb_intr_test(struct igb_adapter *adapter, u64 *data) /* Disable all the interrupts */ wr32(E1000_IMC, ~0); + wrfl(); msleep(10); /* Unhook test interrupt handler */ diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c index dc599059512a..ae3937e8cdef 100644 --- a/drivers/net/igb/igb_main.c +++ b/drivers/net/igb/igb_main.c @@ -1052,6 +1052,7 @@ msi_only: kfree(adapter->vf_data); adapter->vf_data = NULL; wr32(E1000_IOVCTL, E1000_IOVCTL_REUSE_VFQ); + wrfl(); msleep(100); dev_info(&adapter->pdev->dev, "IOV Disabled\n"); } @@ -2198,6 +2199,7 @@ static void __devexit igb_remove(struct pci_dev *pdev) kfree(adapter->vf_data); adapter->vf_data = NULL; wr32(E1000_IOVCTL, E1000_IOVCTL_REUSE_VFQ); + wrfl(); msleep(100); dev_info(&pdev->dev, "IOV Disabled\n"); } diff --git a/drivers/net/igbvf/netdev.c b/drivers/net/igbvf/netdev.c index 1330c8e932da..40ed066e3ef4 100644 --- a/drivers/net/igbvf/netdev.c +++ b/drivers/net/igbvf/netdev.c @@ -1226,6 +1226,7 @@ static void igbvf_configure_tx(struct igbvf_adapter *adapter) /* disable transmits */ txdctl = er32(TXDCTL(0)); ew32(TXDCTL(0), txdctl & ~E1000_TXDCTL_QUEUE_ENABLE); + e1e_flush(); msleep(10); /* Setup the HW Tx Head and Tail descriptor pointers */ @@ -1306,6 +1307,7 @@ static void igbvf_configure_rx(struct igbvf_adapter *adapter) /* disable receives */ rxdctl = er32(RXDCTL(0)); ew32(RXDCTL(0), rxdctl & ~E1000_RXDCTL_QUEUE_ENABLE); + e1e_flush(); msleep(10); rdlen = rx_ring->count * sizeof(union e1000_adv_rx_desc); diff --git a/drivers/net/ixgb/ixgb_ee.c b/drivers/net/ixgb/ixgb_ee.c index c982ab9f9005..38b362b67857 100644 --- a/drivers/net/ixgb/ixgb_ee.c +++ b/drivers/net/ixgb/ixgb_ee.c @@ -57,6 +57,7 @@ ixgb_raise_clock(struct ixgb_hw *hw, */ *eecd_reg = *eecd_reg | IXGB_EECD_SK; IXGB_WRITE_REG(hw, EECD, *eecd_reg); + IXGB_WRITE_FLUSH(hw); udelay(50); } @@ -75,6 +76,7 @@ ixgb_lower_clock(struct ixgb_hw *hw, */ *eecd_reg = *eecd_reg & ~IXGB_EECD_SK; IXGB_WRITE_REG(hw, EECD, *eecd_reg); + IXGB_WRITE_FLUSH(hw); udelay(50); } @@ -112,6 +114,7 @@ ixgb_shift_out_bits(struct ixgb_hw *hw, eecd_reg |= IXGB_EECD_DI; IXGB_WRITE_REG(hw, EECD, eecd_reg); + IXGB_WRITE_FLUSH(hw); udelay(50); @@ -206,21 +209,25 @@ ixgb_standby_eeprom(struct ixgb_hw *hw) /* Deselect EEPROM */ eecd_reg &= ~(IXGB_EECD_CS | IXGB_EECD_SK); IXGB_WRITE_REG(hw, EECD, eecd_reg); + IXGB_WRITE_FLUSH(hw); udelay(50); /* Clock high */ eecd_reg |= IXGB_EECD_SK; IXGB_WRITE_REG(hw, EECD, eecd_reg); + IXGB_WRITE_FLUSH(hw); udelay(50); /* Select EEPROM */ eecd_reg |= IXGB_EECD_CS; IXGB_WRITE_REG(hw, EECD, eecd_reg); + IXGB_WRITE_FLUSH(hw); udelay(50); /* Clock low */ eecd_reg &= ~IXGB_EECD_SK; IXGB_WRITE_REG(hw, EECD, eecd_reg); + IXGB_WRITE_FLUSH(hw); udelay(50); } @@ -239,11 +246,13 @@ ixgb_clock_eeprom(struct ixgb_hw *hw) /* Rising edge of clock */ eecd_reg |= IXGB_EECD_SK; IXGB_WRITE_REG(hw, EECD, eecd_reg); + IXGB_WRITE_FLUSH(hw); udelay(50); /* Falling edge of clock */ eecd_reg &= ~IXGB_EECD_SK; IXGB_WRITE_REG(hw, EECD, eecd_reg); + IXGB_WRITE_FLUSH(hw); udelay(50); } diff --git a/drivers/net/ixgb/ixgb_hw.c b/drivers/net/ixgb/ixgb_hw.c index 6cb2e42ff4c1..3d61a9e4faf7 100644 --- a/drivers/net/ixgb/ixgb_hw.c +++ b/drivers/net/ixgb/ixgb_hw.c @@ -149,6 +149,7 @@ ixgb_adapter_stop(struct ixgb_hw *hw) */ IXGB_WRITE_REG(hw, RCTL, IXGB_READ_REG(hw, RCTL) & ~IXGB_RCTL_RXEN); IXGB_WRITE_REG(hw, TCTL, IXGB_READ_REG(hw, TCTL) & ~IXGB_TCTL_TXEN); + IXGB_WRITE_FLUSH(hw); msleep(IXGB_DELAY_BEFORE_RESET); /* Issue a global reset to the MAC. This will reset the chip's @@ -1220,6 +1221,7 @@ ixgb_optics_reset_bcm(struct ixgb_hw *hw) ctrl &= ~IXGB_CTRL0_SDP2; ctrl |= IXGB_CTRL0_SDP3; IXGB_WRITE_REG(hw, CTRL0, ctrl); + IXGB_WRITE_FLUSH(hw); /* SerDes needs extra delay */ msleep(IXGB_SUN_PHY_RESET_DELAY); diff --git a/drivers/net/ixgbe/ixgbe_common.c b/drivers/net/ixgbe/ixgbe_common.c index 777051f54e53..fc1375f26fe5 100644 --- a/drivers/net/ixgbe/ixgbe_common.c +++ b/drivers/net/ixgbe/ixgbe_common.c @@ -2632,6 +2632,7 @@ s32 ixgbe_blink_led_start_generic(struct ixgbe_hw *hw, u32 index) autoc_reg |= IXGBE_AUTOC_AN_RESTART; autoc_reg |= IXGBE_AUTOC_FLU; IXGBE_WRITE_REG(hw, IXGBE_AUTOC, autoc_reg); + IXGBE_WRITE_FLUSH(hw); usleep_range(10000, 20000); } diff --git a/drivers/net/ixgbe/ixgbe_ethtool.c b/drivers/net/ixgbe/ixgbe_ethtool.c index dc649553a0a6..82d4244c6e10 100644 --- a/drivers/net/ixgbe/ixgbe_ethtool.c +++ b/drivers/net/ixgbe/ixgbe_ethtool.c @@ -1378,6 +1378,7 @@ static int ixgbe_intr_test(struct ixgbe_adapter *adapter, u64 *data) /* Disable all the interrupts */ IXGBE_WRITE_REG(&adapter->hw, IXGBE_EIMC, 0xFFFFFFFF); + IXGBE_WRITE_FLUSH(&adapter->hw); usleep_range(10000, 20000); /* Test each interrupt */ @@ -1398,6 +1399,7 @@ static int ixgbe_intr_test(struct ixgbe_adapter *adapter, u64 *data) ~mask & 0x00007FFF); IXGBE_WRITE_REG(&adapter->hw, IXGBE_EICS, ~mask & 0x00007FFF); + IXGBE_WRITE_FLUSH(&adapter->hw); usleep_range(10000, 20000); if (adapter->test_icr & mask) { @@ -1415,6 +1417,7 @@ static int ixgbe_intr_test(struct ixgbe_adapter *adapter, u64 *data) adapter->test_icr = 0; IXGBE_WRITE_REG(&adapter->hw, IXGBE_EIMS, mask); IXGBE_WRITE_REG(&adapter->hw, IXGBE_EICS, mask); + IXGBE_WRITE_FLUSH(&adapter->hw); usleep_range(10000, 20000); if (!(adapter->test_icr &mask)) { @@ -1435,6 +1438,7 @@ static int ixgbe_intr_test(struct ixgbe_adapter *adapter, u64 *data) ~mask & 0x00007FFF); IXGBE_WRITE_REG(&adapter->hw, IXGBE_EICS, ~mask & 0x00007FFF); + IXGBE_WRITE_FLUSH(&adapter->hw); usleep_range(10000, 20000); if (adapter->test_icr) { @@ -1446,6 +1450,7 @@ static int ixgbe_intr_test(struct ixgbe_adapter *adapter, u64 *data) /* Disable all the interrupts */ IXGBE_WRITE_REG(&adapter->hw, IXGBE_EIMC, 0xFFFFFFFF); + IXGBE_WRITE_FLUSH(&adapter->hw); usleep_range(10000, 20000); /* Unhook test interrupt handler */ diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c index 1be617545dc9..26b132bca43e 100644 --- a/drivers/net/ixgbe/ixgbe_main.c +++ b/drivers/net/ixgbe/ixgbe_main.c @@ -184,6 +184,7 @@ static inline void ixgbe_disable_sriov(struct ixgbe_adapter *adapter) vmdctl = IXGBE_READ_REG(hw, IXGBE_VT_CTL); vmdctl &= ~IXGBE_VT_CTL_POOL_MASK; IXGBE_WRITE_REG(hw, IXGBE_VT_CTL, vmdctl); + IXGBE_WRITE_FLUSH(hw); /* take a breather then clean up driver data */ msleep(100); diff --git a/drivers/net/ixgbe/ixgbe_phy.c b/drivers/net/ixgbe/ixgbe_phy.c index 735f686c3b36..f7ca3511b9fe 100644 --- a/drivers/net/ixgbe/ixgbe_phy.c +++ b/drivers/net/ixgbe/ixgbe_phy.c @@ -1585,6 +1585,7 @@ static s32 ixgbe_raise_i2c_clk(struct ixgbe_hw *hw, u32 *i2cctl) *i2cctl |= IXGBE_I2C_CLK_OUT; IXGBE_WRITE_REG(hw, IXGBE_I2CCTL, *i2cctl); + IXGBE_WRITE_FLUSH(hw); /* SCL rise time (1000ns) */ udelay(IXGBE_I2C_T_RISE); @@ -1605,6 +1606,7 @@ static void ixgbe_lower_i2c_clk(struct ixgbe_hw *hw, u32 *i2cctl) *i2cctl &= ~IXGBE_I2C_CLK_OUT; IXGBE_WRITE_REG(hw, IXGBE_I2CCTL, *i2cctl); + IXGBE_WRITE_FLUSH(hw); /* SCL fall time (300ns) */ udelay(IXGBE_I2C_T_FALL); @@ -1628,6 +1630,7 @@ static s32 ixgbe_set_i2c_data(struct ixgbe_hw *hw, u32 *i2cctl, bool data) *i2cctl &= ~IXGBE_I2C_DATA_OUT; IXGBE_WRITE_REG(hw, IXGBE_I2CCTL, *i2cctl); + IXGBE_WRITE_FLUSH(hw); /* Data rise/fall (1000ns/300ns) and set-up time (250ns) */ udelay(IXGBE_I2C_T_RISE + IXGBE_I2C_T_FALL + IXGBE_I2C_T_SU_DATA); diff --git a/drivers/net/ixgbe/ixgbe_x540.c b/drivers/net/ixgbe/ixgbe_x540.c index bec30ed91adc..2696c78e9f46 100644 --- a/drivers/net/ixgbe/ixgbe_x540.c +++ b/drivers/net/ixgbe/ixgbe_x540.c @@ -162,6 +162,7 @@ mac_reset_top: ctrl_ext = IXGBE_READ_REG(hw, IXGBE_CTRL_EXT); ctrl_ext |= IXGBE_CTRL_EXT_PFRSTD; IXGBE_WRITE_REG(hw, IXGBE_CTRL_EXT, ctrl_ext); + IXGBE_WRITE_FLUSH(hw); msleep(50); From b9e06f70dc186f8353cc593f2b4609383b3be7a9 Mon Sep 17 00:00:00 2001 From: Bruce Allan Date: Fri, 22 Jul 2011 06:21:41 +0000 Subject: [PATCH 337/439] e1000e: remove unnecessary check for NULL pointer The array shadow_ram is never NULL. Signed-off-by: Bruce Allan Tested-by: Aaron Brown Signed-off-by: Jeff Kirsher --- drivers/net/e1000e/ich8lan.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/net/e1000e/ich8lan.c b/drivers/net/e1000e/ich8lan.c index f7a75c1531ad..4e36978b8fd8 100644 --- a/drivers/net/e1000e/ich8lan.c +++ b/drivers/net/e1000e/ich8lan.c @@ -2137,8 +2137,7 @@ static s32 e1000_read_nvm_ich8lan(struct e1000_hw *hw, u16 offset, u16 words, ret_val = 0; for (i = 0; i < words; i++) { - if ((dev_spec->shadow_ram) && - (dev_spec->shadow_ram[offset+i].modified)) { + if (dev_spec->shadow_ram[offset+i].modified) { data[i] = dev_spec->shadow_ram[offset+i].value; } else { ret_val = e1000_read_flash_word_ich8lan(hw, From 9fb7a5f77b26dedfcfa4e3a36fe207f818662bee Mon Sep 17 00:00:00 2001 From: Bruce Allan Date: Fri, 29 Jul 2011 05:52:51 +0000 Subject: [PATCH 338/439] e1000e: minor re-order of #include files The recent commit a6b7a407 when back-ported to the out-of-tree e1000e driver caused a compilation error on older kernels which required a re-ordering of the #include files. This cosmetic patch syncs the two drivers for easier maintainability. Signed-off-by: Bruce Allan Tested-by: Aaron Brown Signed-off-by: Jeff Kirsher --- drivers/net/e1000e/ethtool.c | 2 +- drivers/net/e1000e/netdev.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/e1000e/ethtool.c b/drivers/net/e1000e/ethtool.c index 72756e4e6448..06d88f316dce 100644 --- a/drivers/net/e1000e/ethtool.c +++ b/drivers/net/e1000e/ethtool.c @@ -28,8 +28,8 @@ /* ethtool support for e1000 */ -#include #include +#include #include #include #include diff --git a/drivers/net/e1000e/netdev.c b/drivers/net/e1000e/netdev.c index 4353ad56cf16..ab4be80f7ab5 100644 --- a/drivers/net/e1000e/netdev.c +++ b/drivers/net/e1000e/netdev.c @@ -31,12 +31,12 @@ #include #include #include -#include #include #include #include #include #include +#include #include #include #include From 6d337dce664b6872ddf1655f6b1fcab76ce35b08 Mon Sep 17 00:00:00 2001 From: Carolyn Wyborny Date: Thu, 7 Jul 2011 00:24:56 +0000 Subject: [PATCH 339/439] igb: fix WOL on second port of i350 device This patch fixes a problem where WOL would fail on second port of i350 device. Reported-by: Martin Wilck Reported-by: Stefan Assmann Signed-off-by: Carolyn Wyborny Tested-by: Aaron Brown Signed-off-by: Jeff Kirsher --- drivers/net/igb/igb_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c index ae3937e8cdef..40d4c405fd7e 100644 --- a/drivers/net/igb/igb_main.c +++ b/drivers/net/igb/igb_main.c @@ -2023,7 +2023,7 @@ static int __devinit igb_probe(struct pci_dev *pdev, if (hw->bus.func == 0) hw->nvm.ops.read(hw, NVM_INIT_CONTROL3_PORT_A, 1, &eeprom_data); - else if (hw->mac.type == e1000_82580) + else if (hw->mac.type >= e1000_82580) hw->nvm.ops.read(hw, NVM_INIT_CONTROL3_PORT_A + NVM_82580_LAN_FUNC_OFFSET(hw->bus.func), 1, &eeprom_data); From 2a72c31ee4aa31b6a762390e4811a0edf5eefcef Mon Sep 17 00:00:00 2001 From: Don Skidmore Date: Wed, 20 Jul 2011 02:27:05 +0000 Subject: [PATCH 340/439] ixgbe: fix __ixgbe_notify_dca() bail out code The way __ixgbe_notify_dca() was currently set up it would not be possible to add a requester. Both cases of the IXGBE_FLAG_DCA_ENABLED bit being on and off would lead to the function exiting for a DCA_PROVIDER_ADD. Signed-off-by: Don Skidmore Tested-by: Phil Schmitt Signed-off-by: Jeff Kirsher --- drivers/net/ixgbe/ixgbe_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c index 26b132bca43e..e86297b32733 100644 --- a/drivers/net/ixgbe/ixgbe_main.c +++ b/drivers/net/ixgbe/ixgbe_main.c @@ -1006,7 +1006,7 @@ static int __ixgbe_notify_dca(struct device *dev, void *data) struct ixgbe_adapter *adapter = dev_get_drvdata(dev); unsigned long event = *(unsigned long *)data; - if (!(adapter->flags & IXGBE_FLAG_DCA_ENABLED)) + if (!(adapter->flags & IXGBE_FLAG_DCA_CAPABLE)) return 0; switch (event) { From b57e35bd0e545181c94405ce35b89000aed56cc5 Mon Sep 17 00:00:00 2001 From: Emil Tantilov Date: Thu, 28 Jul 2011 06:17:04 +0000 Subject: [PATCH 341/439] ixgbe: fix PHY link setup for 82599 Fix pointer to setup_link for 82599. This resolves some link issues when advertising modes unsupported by the link partner. Signed-off-by: Emil Tantilov Tested-by: Phil Schmitt Signed-off-by: Jeff Kirsher --- drivers/net/ixgbe/ixgbe_82599.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ixgbe/ixgbe_82599.c b/drivers/net/ixgbe/ixgbe_82599.c index 3b3dd4df4c5c..34f30ec79c2e 100644 --- a/drivers/net/ixgbe/ixgbe_82599.c +++ b/drivers/net/ixgbe/ixgbe_82599.c @@ -213,6 +213,7 @@ static s32 ixgbe_init_phy_ops_82599(struct ixgbe_hw *hw) switch (hw->phy.type) { case ixgbe_phy_tn: phy->ops.check_link = &ixgbe_check_phy_link_tnx; + phy->ops.setup_link = &ixgbe_setup_phy_link_tnx; phy->ops.get_firmware_version = &ixgbe_get_phy_firmware_version_tnx; break; From b8709894f248b569d7a0f32cc5e84fa14f7585f0 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 27 Jul 2011 04:17:25 +0000 Subject: [PATCH 342/439] drm/radeon/kms: fix version comment due to merge timing Compute cs support was actually added in 2.11.0 rather than 2.10.0, but the patch was written prior. Update comment to match. Signed-off-by: Alex Deucher Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon_drv.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c index 85f033f19a8a..e71d2ed7fa11 100644 --- a/drivers/gpu/drm/radeon/radeon_drv.c +++ b/drivers/gpu/drm/radeon/radeon_drv.c @@ -50,8 +50,8 @@ * 2.7.0 - fixups for r600 2D tiling support. (no external ABI change), add eg dyn gpr regs * 2.8.0 - pageflip support, r500 US_FORMAT regs. r500 ARGB2101010 colorbuf, r300->r500 CMASK, clock crystal query * 2.9.0 - r600 tiling (s3tc,rgtc) working, SET_PREDICATION packet 3 on r600 + eg, backend query - * 2.10.0 - fusion 2D tiling, initial compute support for the CS checker - * 2.11.0 - backend map + * 2.10.0 - fusion 2D tiling + * 2.11.0 - backend map, initial compute support for the CS checker */ #define KMS_DRIVER_MAJOR 2 #define KMS_DRIVER_MINOR 11 From 884988398f43378a170026184648ce8ff14dcc9f Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 27 Jul 2011 09:53:40 +0000 Subject: [PATCH 343/439] drm/radeon: off by one in check_reg() functions This off by one range check was copy and pasted a couple places. It's not really harmful, but we should fix it anyway. Signed-off-by: Dan Carpenter Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/evergreen_cs.c | 2 +- drivers/gpu/drm/radeon/r600_cs.c | 3 +-- 2 files changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/radeon/evergreen_cs.c b/drivers/gpu/drm/radeon/evergreen_cs.c index 189e86522b5b..a134790903d3 100644 --- a/drivers/gpu/drm/radeon/evergreen_cs.c +++ b/drivers/gpu/drm/radeon/evergreen_cs.c @@ -428,7 +428,7 @@ static inline int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u3 last_reg = ARRAY_SIZE(evergreen_reg_safe_bm); i = (reg >> 7); - if (i > last_reg) { + if (i >= last_reg) { dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx); return -EINVAL; } diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c index db8ef1905d5f..cf83aa05a684 100644 --- a/drivers/gpu/drm/radeon/r600_cs.c +++ b/drivers/gpu/drm/radeon/r600_cs.c @@ -915,12 +915,11 @@ static inline int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx { struct r600_cs_track *track = (struct r600_cs_track *)p->track; struct radeon_cs_reloc *reloc; - u32 last_reg = ARRAY_SIZE(r600_reg_safe_bm); u32 m, i, tmp, *ib; int r; i = (reg >> 7); - if (i > last_reg) { + if (i >= ARRAY_SIZE(r600_reg_safe_bm)) { dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx); return -EINVAL; } From c41b9ee901bb2c7e3eacfa7e171de50c15d61c0b Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Sat, 30 Jul 2011 18:12:24 +0000 Subject: [PATCH 344/439] drm/radeon/kms: add thermal chip quirk for asus 9600xt The board has an lm63 compatible thermal chip, but no thermal chip entry in the vbios tables. Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=39513 Signed-off-by: Alex Deucher Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon_combios.c | 21 ++++++++++++++++++++- 1 file changed, 20 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/radeon/radeon_combios.c b/drivers/gpu/drm/radeon/radeon_combios.c index a74217cd192f..e0138b674aca 100644 --- a/drivers/gpu/drm/radeon/radeon_combios.c +++ b/drivers/gpu/drm/radeon/radeon_combios.c @@ -2557,6 +2557,7 @@ void radeon_combios_get_power_modes(struct radeon_device *rdev) u16 offset, misc, misc2 = 0; u8 rev, blocks, tmp; int state_index = 0; + struct radeon_i2c_bus_rec i2c_bus; rdev->pm.default_power_state_index = -1; @@ -2575,7 +2576,6 @@ void radeon_combios_get_power_modes(struct radeon_device *rdev) offset = combios_get_table_offset(dev, COMBIOS_OVERDRIVE_INFO_TABLE); if (offset) { u8 thermal_controller = 0, gpio = 0, i2c_addr = 0, clk_bit = 0, data_bit = 0; - struct radeon_i2c_bus_rec i2c_bus; rev = RBIOS8(offset); @@ -2617,6 +2617,25 @@ void radeon_combios_get_power_modes(struct radeon_device *rdev) i2c_new_device(&rdev->pm.i2c_bus->adapter, &info); } } + } else { + /* boards with a thermal chip, but no overdrive table */ + + /* Asus 9600xt has an f75375 on the monid bus */ + if ((dev->pdev->device == 0x4152) && + (dev->pdev->subsystem_vendor == 0x1043) && + (dev->pdev->subsystem_device == 0xc002)) { + i2c_bus = combios_setup_i2c_bus(rdev, DDC_MONID, 0, 0); + rdev->pm.i2c_bus = radeon_i2c_lookup(rdev, &i2c_bus); + if (rdev->pm.i2c_bus) { + struct i2c_board_info info = { }; + const char *name = "f75375"; + info.addr = 0x28; + strlcpy(info.type, name, sizeof(info.type)); + i2c_new_device(&rdev->pm.i2c_bus->adapter, &info); + DRM_INFO("Possible %s thermal controller at 0x%02x\n", + name, info.addr); + } + } } if (rdev->flags & RADEON_IS_MOBILITY) { From 4d5cb60d3f6c396899a515e0e667d0f855ed66cc Mon Sep 17 00:00:00 2001 From: Emil Velikov Date: Sat, 30 Jul 2011 14:26:01 +0000 Subject: [PATCH 345/439] drm/debugfs: Initialise empty variable [airlied: move char declaration] Signed-off-by: Emil Velikov Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_debugfs.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/drm_debugfs.c b/drivers/gpu/drm/drm_debugfs.c index 9d8c892d07c9..9d2668a50872 100644 --- a/drivers/gpu/drm/drm_debugfs.c +++ b/drivers/gpu/drm/drm_debugfs.c @@ -90,7 +90,6 @@ int drm_debugfs_create_files(struct drm_info_list *files, int count, struct drm_device *dev = minor->dev; struct dentry *ent; struct drm_info_node *tmp; - char name[64]; int i, ret; for (i = 0; i < count; i++) { @@ -108,6 +107,9 @@ int drm_debugfs_create_files(struct drm_info_list *files, int count, ent = debugfs_create_file(files[i].name, S_IFREG | S_IRUGO, root, tmp, &drm_debugfs_fops); if (!ent) { + char name[64]; + strncpy(name, root->d_name.name, + min(root->d_name.len, 64U)); DRM_ERROR("Cannot create /sys/kernel/debug/dri/%s/%s\n", name, files[i].name); kfree(tmp); From 69ad2ffe57a81c7b700e337f066b8b39c3655910 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Fernando=20Luis=20V=C3=83=C6=92=C3=86=E2=80=99=C3=83?= =?UTF-8?q?=E2=80=9A=C3=82=C2=A1zquez=20Cao?= Date: Thu, 28 Jul 2011 05:44:04 +0000 Subject: [PATCH 346/439] drm/radeon: clean reg header files Reg header files are generated so they are not cleaned automagically. They need to be added to the clean-files list. Signed-off-by: Fernando Luis Vazquez Cao Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/Makefile | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/radeon/Makefile b/drivers/gpu/drm/radeon/Makefile index 3896ef811102..9f363e0c4b60 100644 --- a/drivers/gpu/drm/radeon/Makefile +++ b/drivers/gpu/drm/radeon/Makefile @@ -5,6 +5,7 @@ ccflags-y := -Iinclude/drm hostprogs-y := mkregtable +clean-files := rn50_reg_safe.h r100_reg_safe.h r200_reg_safe.h rv515_reg_safe.h r300_reg_safe.h r420_reg_safe.h rs600_reg_safe.h r600_reg_safe.h evergreen_reg_safe.h cayman_reg_safe.h quiet_cmd_mkregtable = MKREGTABLE $@ cmd_mkregtable = $(obj)/mkregtable $< > $@ From 816985d4f96cda5159b6b65f3442964725030ae9 Mon Sep 17 00:00:00 2001 From: Bojan Prtvar Date: Wed, 3 Aug 2011 19:51:19 +0000 Subject: [PATCH 347/439] drm/radeon: fix potential NULL dereference in drivers/gpu/drm/radeon/atom.c kzalloc() can return NULL, so I added check for it Signed-off-by: Bojan Prtvar Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/atom.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/radeon/atom.c b/drivers/gpu/drm/radeon/atom.c index ebdb0fdb8348..e88c64417a8a 100644 --- a/drivers/gpu/drm/radeon/atom.c +++ b/drivers/gpu/drm/radeon/atom.c @@ -1245,6 +1245,9 @@ struct atom_context *atom_parse(struct card_info *card, void *bios) char name[512]; int i; + if (!ctx) + return NULL; + ctx->card = card; ctx->bios = bios; From e1c44acc8cabda6b38864f25e809b306c4d2d790 Mon Sep 17 00:00:00 2001 From: Joonyoung Shim Date: Thu, 4 Aug 2011 05:41:00 +0000 Subject: [PATCH 348/439] drm: Fix irq install error handling The registered irq should be unregistered by free_irq() if irq_postinstall() returns the error after request_irq() is called successfully. [airlied: add vga switcheroo disable] Signed-off-by: Joonyoung Shim Signed-off-by: Kyungmin Park Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_irq.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c index 2022a5c966bb..9be574fbcc2f 100644 --- a/drivers/gpu/drm/drm_irq.c +++ b/drivers/gpu/drm/drm_irq.c @@ -368,6 +368,9 @@ int drm_irq_install(struct drm_device *dev) mutex_lock(&dev->struct_mutex); dev->irq_enabled = 0; mutex_unlock(&dev->struct_mutex); + if (!drm_core_check_feature(dev, DRIVER_MODESET)) + vga_client_register(dev->pdev, NULL, NULL, NULL); + free_irq(drm_dev_to_irq(dev), dev); } return ret; From 5037f8acf448dd0de9868dc7410f45879d3d1a1b Mon Sep 17 00:00:00 2001 From: Joonyoung Shim Date: Thu, 4 Aug 2011 05:41:01 +0000 Subject: [PATCH 349/439] drm: Add NULL check about irq functions The struct drm_driver has some function pointers for irq. They are gpu specific and some functions aren't essential things. This can prevents creation of unnecessary dummy function for irq. Signed-off-by: Joonyoung Shim Signed-off-by: Kyungmin Park Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_irq.c | 23 +++++++++++++++-------- 1 file changed, 15 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c index 9be574fbcc2f..3830e9e478c0 100644 --- a/drivers/gpu/drm/drm_irq.c +++ b/drivers/gpu/drm/drm_irq.c @@ -291,11 +291,14 @@ static void drm_irq_vgaarb_nokms(void *cookie, bool state) if (!dev->irq_enabled) return; - if (state) - dev->driver->irq_uninstall(dev); - else { - dev->driver->irq_preinstall(dev); - dev->driver->irq_postinstall(dev); + if (state) { + if (dev->driver->irq_uninstall) + dev->driver->irq_uninstall(dev); + } else { + if (dev->driver->irq_preinstall) + dev->driver->irq_preinstall(dev); + if (dev->driver->irq_postinstall) + dev->driver->irq_postinstall(dev); } } @@ -338,7 +341,8 @@ int drm_irq_install(struct drm_device *dev) DRM_DEBUG("irq=%d\n", drm_dev_to_irq(dev)); /* Before installing handler */ - dev->driver->irq_preinstall(dev); + if (dev->driver->irq_preinstall) + dev->driver->irq_preinstall(dev); /* Install handler */ if (drm_core_check_feature(dev, DRIVER_IRQ_SHARED)) @@ -363,7 +367,9 @@ int drm_irq_install(struct drm_device *dev) vga_client_register(dev->pdev, (void *)dev, drm_irq_vgaarb_nokms, NULL); /* After installing handler */ - ret = dev->driver->irq_postinstall(dev); + if (dev->driver->irq_postinstall) + ret = dev->driver->irq_postinstall(dev); + if (ret < 0) { mutex_lock(&dev->struct_mutex); dev->irq_enabled = 0; @@ -416,7 +422,8 @@ int drm_irq_uninstall(struct drm_device *dev) if (!drm_core_check_feature(dev, DRIVER_MODESET)) vga_client_register(dev->pdev, NULL, NULL, NULL); - dev->driver->irq_uninstall(dev); + if (dev->driver->irq_uninstall) + dev->driver->irq_uninstall(dev); free_irq(drm_dev_to_irq(dev), dev); From 051963d4832ed61e5ae74f5330b0a94489e101b9 Mon Sep 17 00:00:00 2001 From: Thomas Reim Date: Fri, 29 Jul 2011 14:28:57 +0000 Subject: [PATCH 350/439] drm: Separate EDID Header Check from EDID Block Check Provides function drm_edid_header_is_valid() for EDID header check and replaces EDID header check part of function drm_edid_block_valid() by a call of drm_edid_header_is_valid(). This is a prerequisite to extend DDC probing, e. g. in function radeon_ddc_probe() for Radeon devices, by a central EDID header check. Tested for kernel 2.6.35, 2.6.38 and 3.0 Cc: Signed-off-by: Thomas Reim Reviewed-by: Alex Deucher Acked-by: Stephen Michaels Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_edid.c | 24 ++++++++++++++++++------ include/drm/drm_crtc.h | 1 + 2 files changed, 19 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c index 7496d245f28c..7425e5c9bd75 100644 --- a/drivers/gpu/drm/drm_edid.c +++ b/drivers/gpu/drm/drm_edid.c @@ -127,6 +127,23 @@ static const u8 edid_header[] = { 0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x00 }; + /* + * Sanity check the header of the base EDID block. Return 8 if the header + * is perfect, down to 0 if it's totally wrong. + */ +int drm_edid_header_is_valid(const u8 *raw_edid) +{ + int i, score = 0; + + for (i = 0; i < sizeof(edid_header); i++) + if (raw_edid[i] == edid_header[i]) + score++; + + return score; +} +EXPORT_SYMBOL(drm_edid_header_is_valid); + + /* * Sanity check the EDID block (base or extension). Return 0 if the block * doesn't check out, or 1 if it's valid. @@ -139,12 +156,7 @@ drm_edid_block_valid(u8 *raw_edid) struct edid *edid = (struct edid *)raw_edid; if (raw_edid[0] == 0x00) { - int score = 0; - - for (i = 0; i < sizeof(edid_header); i++) - if (raw_edid[i] == edid_header[i]) - score++; - + int score = drm_edid_header_is_valid(raw_edid); if (score == 8) ; else if (score >= 6) { DRM_DEBUG("Fixing EDID header, your hardware may be failing\n"); diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h index d515bc8f4ba7..44335e57eaaa 100644 --- a/include/drm/drm_crtc.h +++ b/include/drm/drm_crtc.h @@ -804,6 +804,7 @@ extern struct drm_display_mode *drm_gtf_mode_complex(struct drm_device *dev, extern int drm_add_modes_noedid(struct drm_connector *connector, int hdisplay, int vdisplay); +extern int drm_edid_header_is_valid(const u8 *raw_edid); extern bool drm_edid_is_valid(struct edid *edid); struct drm_display_mode *drm_mode_find_dmt(struct drm_device *dev, int hsize, int vsize, int fresh); From e384fab8c6f3ca88600bcb2ebdf0eb2f90864fab Mon Sep 17 00:00:00 2001 From: Thomas Reim Date: Fri, 29 Jul 2011 14:28:58 +0000 Subject: [PATCH 351/439] drm/radeon: Extended DDC Probing for Connectors with Improperly Wired DDC Lines (here: Asus M2A-VM HDMI) Some integrated ATI Radeon chipset implementations with add-on HDMI card (e. g. Asus M2A-VM HDMI) indicate the availability of a DDC even when the add-on card is not plugged in or HDMI is disabled in BIOS setup. In this case, drm_get_edid() and drm_edid_block_valid() periodically dump data and kernel errors into system log files and onto terminals. For these connectors DDC probing is extended by a check for a correct EDID header. Only in case a valid EDID header is also found, the (HDMI or DVI) connector will be used by the Radeon driver. This prevents the kernel driver from useless flooding of logs and terminal sessions with EDID dumps and error messages. This patch adds a flag 'requires_extended_probe' to the radeon_connector structure. In function radeon_connector_needs_extended_probe() this flag can be set on a chipset family/vendor/connector type specific basis. In addition, function radeon_ddc_probe() has been adapted to perform extended DDC probing if required by the connector's flag. Requires function drm_edid_header_is_valid() in DRM module provided by [PATCH] drm: Separate EDID Header Check from EDID Block Check. Tested for kernel 2.6.35, 2.6.38 and 3.0 on Asus M2A-VM HDMI board BugLink: https://bugzilla.redhat.com/show_bug.cgi?id=668196 BugLink: http://bugs.launchpad.net/bugs/7228066 Cc: Signed-off-by: Thomas Reim Reviewed-by: Alex Deucher Acked-by: Stephen Michaels Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon_connectors.c | 45 ++++++++++++++++++++-- drivers/gpu/drm/radeon/radeon_display.c | 9 +++++ drivers/gpu/drm/radeon/radeon_i2c.c | 32 +++++++++++---- drivers/gpu/drm/radeon/radeon_mode.h | 6 ++- 4 files changed, 80 insertions(+), 12 deletions(-) diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index 9792d4ffdc86..6a0f1befbc15 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -430,6 +430,36 @@ int radeon_connector_set_property(struct drm_connector *connector, struct drm_pr return 0; } +/* + * Some integrated ATI Radeon chipset implementations (e. g. + * Asus M2A-VM HDMI) may indicate the availability of a DDC, + * even when there's no monitor connected. For these connectors + * following DDC probe extension will be applied: check also for the + * availability of EDID with at least a correct EDID header. Only then, + * DDC is assumed to be available. This prevents drm_get_edid() and + * drm_edid_block_valid() from periodically dumping data and kernel + * errors into the logs and onto the terminal. + */ +static bool radeon_connector_needs_extended_probe(struct radeon_device *dev, + uint32_t supported_device, + int connector_type) +{ + /* Asus M2A-VM HDMI board sends data to i2c bus even, + * if HDMI add-on card is not plugged in or HDMI is disabled in + * BIOS. Valid DDC can only be assumed, if also a valid EDID header + * can be retrieved via i2c bus during DDC probe */ + if ((dev->pdev->device == 0x791e) && + (dev->pdev->subsystem_vendor == 0x1043) && + (dev->pdev->subsystem_device == 0x826d)) { + if ((connector_type == DRM_MODE_CONNECTOR_HDMIA) && + (supported_device == ATOM_DEVICE_DFP2_SUPPORT)) + return true; + } + + /* Default: no EDID header probe required for DDC probing */ + return false; +} + static void radeon_fixup_lvds_native_mode(struct drm_encoder *encoder, struct drm_connector *connector) { @@ -661,7 +691,8 @@ radeon_vga_detect(struct drm_connector *connector, bool force) ret = connector_status_disconnected; if (radeon_connector->ddc_bus) - dret = radeon_ddc_probe(radeon_connector); + dret = radeon_ddc_probe(radeon_connector, + radeon_connector->requires_extended_probe); if (dret) { if (radeon_connector->edid) { kfree(radeon_connector->edid); @@ -833,7 +864,8 @@ radeon_dvi_detect(struct drm_connector *connector, bool force) bool dret = false; if (radeon_connector->ddc_bus) - dret = radeon_ddc_probe(radeon_connector); + dret = radeon_ddc_probe(radeon_connector, + radeon_connector->requires_extended_probe); if (dret) { if (radeon_connector->edid) { kfree(radeon_connector->edid); @@ -1251,7 +1283,8 @@ radeon_dp_detect(struct drm_connector *connector, bool force) if (radeon_dp_getdpcd(radeon_connector)) ret = connector_status_connected; } else { - if (radeon_ddc_probe(radeon_connector)) + if (radeon_ddc_probe(radeon_connector, + radeon_connector->requires_extended_probe)) ret = connector_status_connected; } } @@ -1406,6 +1439,9 @@ radeon_add_atom_connector(struct drm_device *dev, radeon_connector->shared_ddc = shared_ddc; radeon_connector->connector_object_id = connector_object_id; radeon_connector->hpd = *hpd; + radeon_connector->requires_extended_probe = + radeon_connector_needs_extended_probe(rdev, supported_device, + connector_type); radeon_connector->router = *router; if (router->ddc_valid || router->cd_valid) { radeon_connector->router_bus = radeon_i2c_lookup(rdev, &router->i2c_info); @@ -1752,6 +1788,9 @@ radeon_add_legacy_connector(struct drm_device *dev, radeon_connector->devices = supported_device; radeon_connector->connector_object_id = connector_object_id; radeon_connector->hpd = *hpd; + radeon_connector->requires_extended_probe = + radeon_connector_needs_extended_probe(rdev, supported_device, + connector_type); switch (connector_type) { case DRM_MODE_CONNECTOR_VGA: drm_connector_init(dev, &radeon_connector->base, &radeon_vga_connector_funcs, connector_type); diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c index 28f4655905bc..1a858944e4f3 100644 --- a/drivers/gpu/drm/radeon/radeon_display.c +++ b/drivers/gpu/drm/radeon/radeon_display.c @@ -751,8 +751,17 @@ static int radeon_ddc_dump(struct drm_connector *connector) if (!radeon_connector->ddc_bus) return -1; edid = drm_get_edid(connector, &radeon_connector->ddc_bus->adapter); + /* Log EDID retrieval status here. In particular with regard to + * connectors with requires_extended_probe flag set, that will prevent + * function radeon_dvi_detect() to fetch EDID on this connector, + * as long as there is no valid EDID header found */ if (edid) { + DRM_INFO("Radeon display connector %s: Found valid EDID", + drm_get_connector_name(connector)); kfree(edid); + } else { + DRM_INFO("Radeon display connector %s: No monitor connected or invalid EDID", + drm_get_connector_name(connector)); } return ret; } diff --git a/drivers/gpu/drm/radeon/radeon_i2c.c b/drivers/gpu/drm/radeon/radeon_i2c.c index 781196db792f..6c111c1fa3f9 100644 --- a/drivers/gpu/drm/radeon/radeon_i2c.c +++ b/drivers/gpu/drm/radeon/radeon_i2c.c @@ -32,17 +32,17 @@ * radeon_ddc_probe * */ -bool radeon_ddc_probe(struct radeon_connector *radeon_connector) +bool radeon_ddc_probe(struct radeon_connector *radeon_connector, bool requires_extended_probe) { - u8 out_buf[] = { 0x0, 0x0}; - u8 buf[2]; + u8 out = 0x0; + u8 buf[8]; int ret; struct i2c_msg msgs[] = { { .addr = 0x50, .flags = 0, .len = 1, - .buf = out_buf, + .buf = &out, }, { .addr = 0x50, @@ -52,15 +52,31 @@ bool radeon_ddc_probe(struct radeon_connector *radeon_connector) } }; + /* Read 8 bytes from i2c for extended probe of EDID header */ + if (requires_extended_probe) + msgs[1].len = 8; + /* on hw with routers, select right port */ if (radeon_connector->router.ddc_valid) radeon_router_select_ddc_port(radeon_connector); ret = i2c_transfer(&radeon_connector->ddc_bus->adapter, msgs, 2); - if (ret == 2) - return true; - - return false; + if (ret != 2) + /* Couldn't find an accessible DDC on this connector */ + return false; + if (requires_extended_probe) { + /* Probe also for valid EDID header + * EDID header starts with: + * 0x00,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0x00. + * Only the first 6 bytes must be valid as + * drm_edid_block_valid() can fix the last 2 bytes */ + if (drm_edid_header_is_valid(buf) < 6) { + /* Couldn't find an accessible EDID on this + * connector */ + return false; + } + } + return true; } /* bit banging i2c */ diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h index 6df4e3cec0c2..d09031c03e26 100644 --- a/drivers/gpu/drm/radeon/radeon_mode.h +++ b/drivers/gpu/drm/radeon/radeon_mode.h @@ -438,6 +438,9 @@ struct radeon_connector { struct radeon_i2c_chan *ddc_bus; /* some systems have an hdmi and vga port with a shared ddc line */ bool shared_ddc; + /* for some Radeon chip families we apply an additional EDID header + check as part of the DDC probe */ + bool requires_extended_probe; bool use_digital; /* we need to mind the EDID between detect and get modes due to analog/digital/tvencoder */ @@ -514,7 +517,8 @@ extern void radeon_i2c_put_byte(struct radeon_i2c_chan *i2c, u8 val); extern void radeon_router_select_ddc_port(struct radeon_connector *radeon_connector); extern void radeon_router_select_cd_port(struct radeon_connector *radeon_connector); -extern bool radeon_ddc_probe(struct radeon_connector *radeon_connector); +extern bool radeon_ddc_probe(struct radeon_connector *radeon_connector, + bool requires_extended_probe); extern int radeon_ddc_get_modes(struct radeon_connector *radeon_connector); extern struct drm_encoder *radeon_best_encoder(struct drm_connector *connector); From d522d9cc5bdd41214084383fc3e6d882f6916a78 Mon Sep 17 00:00:00 2001 From: Thomas Reim Date: Fri, 29 Jul 2011 14:28:59 +0000 Subject: [PATCH 352/439] drm/radeon: Log Subsystem Vendor and Device Information Log PCI subsystem vendor and subsystem device ID in addition to PCI vendor and device ID during kernel mode initialisation. This helps to better identify radeon devices of third-party vendors, e. g. for bug analysis. Tested for kernel 2.6.35, 2.6.38 and 3.0 on Asus M2A-VM HDMI board Cc: Signed-off-by: Thomas Reim Reviewed-by: Alex Deucher Acked-by: Stephen Michaels Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon_device.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c index 7cfaa7e2f3b5..440e6ecccc40 100644 --- a/drivers/gpu/drm/radeon/radeon_device.c +++ b/drivers/gpu/drm/radeon/radeon_device.c @@ -704,8 +704,9 @@ int radeon_device_init(struct radeon_device *rdev, rdev->gpu_lockup = false; rdev->accel_working = false; - DRM_INFO("initializing kernel modesetting (%s 0x%04X:0x%04X).\n", - radeon_family_name[rdev->family], pdev->vendor, pdev->device); + DRM_INFO("initializing kernel modesetting (%s 0x%04X:0x%04X 0x%04X:0x%04X).\n", + radeon_family_name[rdev->family], pdev->vendor, pdev->device, + pdev->subsystem_vendor, pdev->subsystem_device); /* mutex initialization are all done here so we * can recall function without having locking issues */ From a81b31e9fc98e067b7e7f1244861c97e44268e2d Mon Sep 17 00:00:00 2001 From: Thomas Reim Date: Fri, 29 Jul 2011 14:29:00 +0000 Subject: [PATCH 353/439] drm/radeon: Extended DDC Probing for ECS A740GM-M DVI-D Connector ECS A740GM-M with ATI RADEON 2100 sends data to i2c bus for a DVI connector that is not implemented/existent on the board. Fix by applying extented DDC probing for this connector. Requires [PATCH] drm/radeon: Extended DDC Probing for Connectors with Improperly Wired DDC Lines Tested for kernel 2.6.38 on Asus ECS A740GM-M board BugLink: http://bugs.launchpad.net/bugs/810926 Cc: Signed-off-by: Thomas Reim Reviewed-by: Alex Deucher Acked-by: Stephen Michaels Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon_connectors.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index 6a0f1befbc15..6d6b5f16bc09 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -455,6 +455,15 @@ static bool radeon_connector_needs_extended_probe(struct radeon_device *dev, (supported_device == ATOM_DEVICE_DFP2_SUPPORT)) return true; } + /* ECS A740GM-M with ATI RADEON 2100 sends data to i2c bus + * for a DVI connector that is not implemented */ + if ((dev->pdev->device == 0x796e) && + (dev->pdev->subsystem_vendor == 0x1019) && + (dev->pdev->subsystem_device == 0x2615)) { + if ((connector_type == DRM_MODE_CONNECTOR_DVID) && + (supported_device == ATOM_DEVICE_DFP2_SUPPORT)) + return true; + } /* Default: no EDID header probe required for DDC probing */ return false; From 82de9a0cc34bc1640c4f133f13d62a765596b2b9 Mon Sep 17 00:00:00 2001 From: Arnaud Lacombe Date: Thu, 4 Aug 2011 10:39:44 -0400 Subject: [PATCH 354/439] eisa/pci_eisa.c: fix BUG introduced by 005bdad7b80 While `pci_eisa_driver' still refer `pci_eisa_init', the .probe() function should not be called after init memory release, as pointed out by commit 74b9a297. The structure is still referenced in the drivers subsystem, and can be accesseed through sysfs, so the modpost warning is a false positive. Mark it as such. In the same time, the warning referenced in 005bdad7b80 did only mention `pci_eisa_driver', not `pci_eisa_pci_tbl', so remove its marking. Broken-by: Arnaud Lacombe (in 005bdad7b80) Reported-by: Tetsuo Handa Signed-off-by: Arnaud Lacombe Signed-off-by: Linus Torvalds --- drivers/eisa/pci_eisa.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/eisa/pci_eisa.c b/drivers/eisa/pci_eisa.c index 30da70d06a6d..cdae207028a7 100644 --- a/drivers/eisa/pci_eisa.c +++ b/drivers/eisa/pci_eisa.c @@ -45,13 +45,13 @@ static int __init pci_eisa_init(struct pci_dev *pdev, return 0; } -static struct pci_device_id __initdata pci_eisa_pci_tbl[] = { +static struct pci_device_id pci_eisa_pci_tbl[] = { { PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_BRIDGE_EISA << 8, 0xffff00, 0 }, { 0, } }; -static struct pci_driver __initdata pci_eisa_driver = { +static struct pci_driver __refdata pci_eisa_driver = { .name = "pci_eisa", .id_table = pci_eisa_pci_tbl, .probe = pci_eisa_init, From 1e9ea2656b656edd3c8de98675bbc0340211b5bd Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Wed, 3 Aug 2011 09:43:44 -0700 Subject: [PATCH 355/439] xen/tracing: it looks like we wanted CONFIG_FTRACE Apparently we wanted CONFIG_FTRACE rather the CONFIG_FUNCTION_TRACER. Reported-by: Sander Eikelenboom Tested-by: Sander Eikelenboom Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Konrad Rzeszutek Wilk --- arch/x86/xen/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/xen/Makefile b/arch/x86/xen/Makefile index 45e94aca5bce..3326204e251f 100644 --- a/arch/x86/xen/Makefile +++ b/arch/x86/xen/Makefile @@ -15,7 +15,7 @@ obj-y := enlighten.o setup.o multicalls.o mmu.o irq.o \ grant-table.o suspend.o platform-pci-unplug.o \ p2m.o -obj-$(CONFIG_FUNCTION_TRACER) += trace.o +obj-$(CONFIG_FTRACE) += trace.o obj-$(CONFIG_SMP) += smp.o obj-$(CONFIG_PARAVIRT_SPINLOCKS)+= spinlock.o From 8f3c5883d840d079a5d2db20893b5ac8ba453b40 Mon Sep 17 00:00:00 2001 From: Igor Mammedov Date: Tue, 2 Aug 2011 11:45:24 +0200 Subject: [PATCH 356/439] xen: Fix printk() format in xen/setup.c Use correct format specifier for unsigned long. Signed-off-by: Igor Mammedov Signed-off-by: Konrad Rzeszutek Wilk --- arch/x86/xen/setup.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c index 60aeeb56948f..e4056321b243 100644 --- a/arch/x86/xen/setup.c +++ b/arch/x86/xen/setup.c @@ -113,7 +113,7 @@ static unsigned long __init xen_release_chunk(phys_addr_t start_addr, len++; } } - printk(KERN_CONT "%ld pages freed\n", len); + printk(KERN_CONT "%lu pages freed\n", len); return len; } @@ -139,7 +139,7 @@ static unsigned long __init xen_return_unused_memory(unsigned long max_pfn, if (last_end < max_addr) released += xen_release_chunk(last_end, max_addr); - printk(KERN_INFO "released %ld pages of unused memory\n", released); + printk(KERN_INFO "released %lu pages of unused memory\n", released); return released; } From 98f531da8479eec3d828adc7f0865baaab99a543 Mon Sep 17 00:00:00 2001 From: Igor Mammedov Date: Tue, 2 Aug 2011 11:45:25 +0200 Subject: [PATCH 357/439] xen: Fix misleading WARN message at xen_release_chunk WARN message should not complain "Failed to release memory %lx-%lx err=%d\n" ^^^^^^^ about range when it fails to release just one page, instead it should say what pfn is not freed. In addition line: printk(KERN_INFO "xen_release_chunk: looking at area pfn %lx-%lx: " ... printk(KERN_CONT "%lu pages freed\n", len); will be broken if WARN in between this line is fired. So fix it by using a single printk for this. Signed-off-by: Igor Mammedov Signed-off-by: Konrad Rzeszutek Wilk --- arch/x86/xen/setup.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c index e4056321b243..02ffd9e48c9f 100644 --- a/arch/x86/xen/setup.c +++ b/arch/x86/xen/setup.c @@ -92,8 +92,6 @@ static unsigned long __init xen_release_chunk(phys_addr_t start_addr, if (end <= start) return 0; - printk(KERN_INFO "xen_release_chunk: looking at area pfn %lx-%lx: ", - start, end); for(pfn = start; pfn < end; pfn++) { unsigned long mfn = pfn_to_mfn(pfn); @@ -106,14 +104,14 @@ static unsigned long __init xen_release_chunk(phys_addr_t start_addr, ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, &reservation); - WARN(ret != 1, "Failed to release memory %lx-%lx err=%d\n", - start, end, ret); + WARN(ret != 1, "Failed to release pfn %lx err=%d\n", pfn, ret); if (ret == 1) { __set_phys_to_machine(pfn, INVALID_P2M_ENTRY); len++; } } - printk(KERN_CONT "%lu pages freed\n", len); + printk(KERN_INFO "Freeing %lx-%lx pfn range: %lu pages freed\n", + start, end, len); return len; } From 655b16128482fd12808f77a6799eea5419c93709 Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Sun, 29 May 2011 10:57:47 +0300 Subject: [PATCH 358/439] nfs_xdr: Move nfs4_string definition out of #ifdef CONFIG_NFS_V4 exofs file system wants to use pnfs_osd_xdr.h file instead of redefining pnfs-objects types in it's private "pnfs.h" headr. Before we do the switch we must make sure pnfs_osd_xdr.h is compilable also under NFS versions smaller than 4.1. Since now it is needed regardless of version, by the exofs code. nfs4_string is not the only nfs4 type out in the global scope. Ack-by: Trond Myklebust Signed-off-by: Boaz Harrosh --- include/linux/nfs_xdr.h | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 00848d86ffb2..d9f5e8d6ca84 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -759,6 +759,11 @@ struct nfs3_getaclres { struct posix_acl * acl_default; }; +struct nfs4_string { + unsigned int len; + char *data; +}; + #ifdef CONFIG_NFS_V4 typedef u64 clientid4; @@ -949,11 +954,6 @@ struct nfs4_server_caps_res { struct nfs4_sequence_res seq_res; }; -struct nfs4_string { - unsigned int len; - char *data; -}; - #define NFS4_PATHNAME_MAXCOMPONENTS 512 struct nfs4_pathname { unsigned int ncomponents; From 26ae93c2dc7152463d319c28768f242a11a54620 Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Tue, 2 Feb 2010 15:56:53 +0200 Subject: [PATCH 359/439] exofs: Remove pnfs-osd private definitions Now that pnfs-osd has hit mainline we can remove exofs's private header. (And the FIXME comment) Signed-off-by: Boaz Harrosh --- fs/exofs/exofs.h | 6 +----- fs/exofs/pnfs.h | 45 --------------------------------------------- 2 files changed, 1 insertion(+), 50 deletions(-) delete mode 100644 fs/exofs/pnfs.h diff --git a/fs/exofs/exofs.h b/fs/exofs/exofs.h index c965806c2821..e103dbd86fe1 100644 --- a/fs/exofs/exofs.h +++ b/fs/exofs/exofs.h @@ -36,13 +36,9 @@ #include #include #include +#include #include "common.h" -/* FIXME: Remove once pnfs hits mainline - * #include - */ -#include "pnfs.h" - #define EXOFS_ERR(fmt, a...) printk(KERN_ERR "exofs: " fmt, ##a) #ifdef CONFIG_EXOFS_DEBUG diff --git a/fs/exofs/pnfs.h b/fs/exofs/pnfs.h deleted file mode 100644 index c52e9888b8ab..000000000000 --- a/fs/exofs/pnfs.h +++ /dev/null @@ -1,45 +0,0 @@ -/* - * Copyright (C) 2008, 2009 - * Boaz Harrosh - * - * This file is part of exofs. - * - * exofs 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. - * - */ - -/* FIXME: Remove this file once pnfs hits mainline */ - -#ifndef __EXOFS_PNFS_H__ -#define __EXOFS_PNFS_H__ - -#if ! defined(__PNFS_OSD_XDR_H__) - -enum pnfs_iomode { - IOMODE_READ = 1, - IOMODE_RW = 2, - IOMODE_ANY = 3, -}; - -/* Layout Structure */ -enum pnfs_osd_raid_algorithm4 { - PNFS_OSD_RAID_0 = 1, - PNFS_OSD_RAID_4 = 2, - PNFS_OSD_RAID_5 = 3, - PNFS_OSD_RAID_PQ = 4 /* Reed-Solomon P+Q */ -}; - -struct pnfs_osd_data_map { - u32 odm_num_comps; - u64 odm_stripe_unit; - u32 odm_group_width; - u32 odm_group_depth; - u32 odm_mirror_cnt; - u32 odm_raid_algorithm; -}; - -#endif /* ! defined(__PNFS_OSD_XDR_H__) */ - -#endif /* __EXOFS_PNFS_H__ */ From 6d4073e88132259485ef1b2c88daa5e50c95789c Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Wed, 27 Jul 2011 17:51:53 -0700 Subject: [PATCH 360/439] exofs: BUG: Avoid sbi realloc Since the beginning we realloced the sbi structure when a bigger then one device table was specified. (I know that was really stupid). Then much later when "register bdi" was added (By Jens) it was registering the pointer to sbi->bdi before the realloc. We never saw this problem because up till now the realloc did not do anything since the device table was small enough to fit in the original allocation. But once we starting testing with large device tables (Bigger then 28) we noticed the crash of writeback operating on a deallocated pointer. * Avoid the all mess by allocating the device-table as a second array and get rid of the variable-sized structure and the rest of this mess. * Take the chance to clean near by structures and comments. * Add a needed dprint on startup to indicate the loaded layout. * Also move the bdi registration to the very end because it will only fail in a low memory, which will probably fail before hand. There are many more likely causes to not load before that. This way the error handling is made simpler. (Just doing this would be enough to fix the BUG) Signed-off-by: Boaz Harrosh --- fs/exofs/exofs.h | 9 +++------ fs/exofs/super.c | 45 +++++++++++++++++++++++++++------------------ 2 files changed, 30 insertions(+), 24 deletions(-) diff --git a/fs/exofs/exofs.h b/fs/exofs/exofs.h index e103dbd86fe1..9f62349a5a5c 100644 --- a/fs/exofs/exofs.h +++ b/fs/exofs/exofs.h @@ -66,13 +66,14 @@ struct exofs_layout { enum exofs_inode_layout_gen_functions lay_func; unsigned s_numdevs; /* Num of devices in array */ - struct osd_dev *s_ods[0]; /* Variable length */ + struct osd_dev **s_ods; /* osd_dev array */ }; /* * our extension to the in-memory superblock */ struct exofs_sb_info { + struct backing_dev_info bdi; /* register our bdi with VFS */ struct exofs_sb_stats s_ess; /* Written often, pre-allocate*/ int s_timeout; /* timeout for OSD operations */ uint64_t s_nextid; /* highest object ID used */ @@ -81,15 +82,11 @@ struct exofs_sb_info { u32 s_next_generation; /* next gen # to use */ atomic_t s_curr_pending; /* number of pending commands */ uint8_t s_cred[OSD_CAP_LEN]; /* credential for the fscb */ - struct backing_dev_info bdi; /* register our bdi with VFS */ struct pnfs_osd_data_map data_map; /* Default raid to use * FIXME: Needed ? */ -/* struct exofs_layout dir_layout;*/ /* Default dir layout */ - struct exofs_layout layout; /* Default files layout, - * contains the variable osd_dev - * array. Keep last */ + struct exofs_layout layout; /* Default files layout */ struct osd_dev *_min_one_dev[1]; /* Place holder for one dev */ }; diff --git a/fs/exofs/super.c b/fs/exofs/super.c index c57beddcc217..a747c871d313 100644 --- a/fs/exofs/super.c +++ b/fs/exofs/super.c @@ -393,6 +393,8 @@ void exofs_free_sbi(struct exofs_sb_info *sbi) osduld_put_device(od); } } + if (sbi->layout.s_ods != sbi->_min_one_dev) + kfree(sbi->layout.s_ods); kfree(sbi); } @@ -501,6 +503,15 @@ static int _read_and_match_data_map(struct exofs_sb_info *sbi, unsigned numdevs, return -EINVAL; } + EXOFS_DBGMSG("exofs: layout: " + "num_comps=%u stripe_unit=0x%x group_width=%u " + "group_depth=0x%llx mirrors_p1=%u raid_algorithm=%u\n", + numdevs, + sbi->layout.stripe_unit, + sbi->layout.group_width, + _LLU(sbi->layout.group_depth), + sbi->layout.mirrors_p1, + sbi->data_map.odm_raid_algorithm); return 0; } @@ -547,11 +558,10 @@ static int exofs_devs_2_odi(struct exofs_dt_device_info *dt_dev, return !(odi->systemid_len || odi->osdname_len); } -static int exofs_read_lookup_dev_table(struct exofs_sb_info **psbi, +static int exofs_read_lookup_dev_table(struct exofs_sb_info *sbi, + struct osd_dev *fscb_od, unsigned table_count) { - struct exofs_sb_info *sbi = *psbi; - struct osd_dev *fscb_od; struct osd_obj_id obj = {.partition = sbi->layout.s_pid, .id = EXOFS_DEVTABLE_ID}; struct exofs_device_table *dt; @@ -567,8 +577,6 @@ static int exofs_read_lookup_dev_table(struct exofs_sb_info **psbi, return -ENOMEM; } - fscb_od = sbi->layout.s_ods[0]; - sbi->layout.s_ods[0] = NULL; sbi->layout.s_numdevs = 0; ret = exofs_read_kern(fscb_od, sbi->s_cred, &obj, 0, dt, table_bytes); if (unlikely(ret)) { @@ -590,14 +598,13 @@ static int exofs_read_lookup_dev_table(struct exofs_sb_info **psbi, if (likely(numdevs > 1)) { unsigned size = numdevs * sizeof(sbi->layout.s_ods[0]); - sbi = krealloc(sbi, sizeof(*sbi) + size, GFP_KERNEL); - if (unlikely(!sbi)) { + sbi->layout.s_ods = kzalloc(size, GFP_KERNEL); + if (unlikely(!sbi->layout.s_ods)) { + EXOFS_ERR("ERROR: faild allocating Device array[%d]\n", + numdevs); ret = -ENOMEM; goto out; } - memset(&sbi->layout.s_ods[1], 0, - size - sizeof(sbi->layout.s_ods[0])); - *psbi = sbi; } for (i = 0; i < numdevs; i++) { @@ -684,10 +691,6 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent) if (!sbi) return -ENOMEM; - ret = bdi_setup_and_register(&sbi->bdi, "exofs", BDI_CAP_MAP_COPY); - if (ret) - goto free_bdi; - /* use mount options to fill superblock */ if (opts->is_osdname) { struct osd_dev_info odi = {.systemid_len = 0}; @@ -709,7 +712,7 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent) sbi->layout.group_width = 1; sbi->layout.group_depth = -1; sbi->layout.group_count = 1; - sbi->layout.s_ods[0] = od; + sbi->layout.s_ods = sbi->_min_one_dev; sbi->layout.s_numdevs = 1; sbi->layout.s_pid = opts->pid; sbi->s_timeout = opts->timeout; @@ -757,9 +760,11 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent) table_count = le64_to_cpu(fscb.s_dev_table_count); if (table_count) { - ret = exofs_read_lookup_dev_table(&sbi, table_count); + ret = exofs_read_lookup_dev_table(sbi, od, table_count); if (unlikely(ret)) goto free_sbi; + } else { + sbi->layout.s_ods[0] = od; } __sbi_read_stats(sbi); @@ -793,6 +798,12 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent) goto free_sbi; } + ret = bdi_setup_and_register(&sbi->bdi, "exofs", BDI_CAP_MAP_COPY); + if (ret) { + EXOFS_DBGMSG("Failed to bdi_setup_and_register\n"); + goto free_sbi; + } + _exofs_print_device("Mounting", opts->dev_name, sbi->layout.s_ods[0], sbi->layout.s_pid); if (opts->is_osdname) @@ -800,8 +811,6 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent) return 0; free_sbi: - bdi_destroy(&sbi->bdi); -free_bdi: EXOFS_ERR("Unable to mount exofs on %s pid=0x%llx err=%d\n", opts->dev_name, sbi->layout.s_pid, ret); exofs_free_sbi(sbi); From 9ce730475e1b950d78a69c1be3410109c103ac98 Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Wed, 3 Aug 2011 20:18:01 -0700 Subject: [PATCH 361/439] exofs: Small cleanup of exofs_fill_super Small cleanup that unifies duplicated code used in both the error and success cases Signed-off-by: Boaz Harrosh --- fs/exofs/super.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/fs/exofs/super.c b/fs/exofs/super.c index a747c871d313..65fe5debe11c 100644 --- a/fs/exofs/super.c +++ b/fs/exofs/super.c @@ -698,6 +698,8 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent) odi.osdname_len = strlen(opts->dev_name); odi.osdname = (u8 *)opts->dev_name; od = osduld_info_lookup(&odi); + kfree(opts->dev_name); + opts->dev_name = NULL; } else { od = osduld_path_lookup(opts->dev_name); } @@ -806,16 +808,12 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent) _exofs_print_device("Mounting", opts->dev_name, sbi->layout.s_ods[0], sbi->layout.s_pid); - if (opts->is_osdname) - kfree(opts->dev_name); return 0; free_sbi: EXOFS_ERR("Unable to mount exofs on %s pid=0x%llx err=%d\n", opts->dev_name, sbi->layout.s_pid, ret); exofs_free_sbi(sbi); - if (opts->is_osdname) - kfree(opts->dev_name); return ret; } From 16f75bb35d54b44356f496272c013f7ace5fa698 Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Wed, 3 Aug 2011 20:44:16 -0700 Subject: [PATCH 362/439] exofs: Fix truncate for the raid-groups case In the general raid-group case the truncate was wrong in that it did not also fix the object length of the neighboring groups. There are two bad cases in the old code: 1. Space that should be freed was not. 2. If a file That was big is truncated small, then made bigger again, the holes would not contain zeros but could expose old data. (If the growing of the file expands to more than a full groups cycle + group size (> S + T)) Signed-off-by: Boaz Harrosh --- fs/exofs/ios.c | 73 ++++++++++++++++++++++++++++++++++++-------------- 1 file changed, 53 insertions(+), 20 deletions(-) diff --git a/fs/exofs/ios.c b/fs/exofs/ios.c index f74a2ec027a6..fbb47ba2cd71 100644 --- a/fs/exofs/ios.c +++ b/fs/exofs/ios.c @@ -305,20 +305,21 @@ int exofs_check_io(struct exofs_io_state *ios, u64 *resid) struct _striping_info { u64 obj_offset; u64 group_length; + u64 M; /* for truncate */ unsigned dev; unsigned unit_off; }; -static void _calc_stripe_info(struct exofs_io_state *ios, u64 file_offset, +static void _calc_stripe_info(struct exofs_layout *layout, u64 file_offset, struct _striping_info *si) { - u32 stripe_unit = ios->layout->stripe_unit; - u32 group_width = ios->layout->group_width; - u64 group_depth = ios->layout->group_depth; + u32 stripe_unit = layout->stripe_unit; + u32 group_width = layout->group_width; + u64 group_depth = layout->group_depth; u32 U = stripe_unit * group_width; u64 T = U * group_depth; - u64 S = T * ios->layout->group_count; + u64 S = T * layout->group_count; u64 M = div64_u64(file_offset, S); /* @@ -333,7 +334,7 @@ static void _calc_stripe_info(struct exofs_io_state *ios, u64 file_offset, /* "H - (N * U)" is just "H % U" so it's bound to u32 */ si->dev = (u32)(H - (N * U)) / stripe_unit + G * group_width; - si->dev *= ios->layout->mirrors_p1; + si->dev *= layout->mirrors_p1; div_u64_rem(file_offset, stripe_unit, &si->unit_off); @@ -341,6 +342,7 @@ static void _calc_stripe_info(struct exofs_io_state *ios, u64 file_offset, (M * group_depth * stripe_unit); si->group_length = T - H; + si->M = M; } static int _add_stripe_unit(struct exofs_io_state *ios, unsigned *cur_pg, @@ -454,7 +456,7 @@ static int _prepare_for_striping(struct exofs_io_state *ios) if (ios->kern_buff) { struct exofs_per_dev_state *per_dev = &ios->per_dev[0]; - _calc_stripe_info(ios, ios->offset, &si); + _calc_stripe_info(ios->layout, ios->offset, &si); per_dev->offset = si.obj_offset; per_dev->dev = si.dev; @@ -468,7 +470,7 @@ static int _prepare_for_striping(struct exofs_io_state *ios) } while (length) { - _calc_stripe_info(ios, offset, &si); + _calc_stripe_info(ios->layout, offset, &si); if (length < si.group_length) si.group_length = length; @@ -745,6 +747,31 @@ static int _truncate_mirrors(struct exofs_io_state *ios, unsigned cur_comp, return 0; } +struct _trunc_info { + struct _striping_info si; + u64 prev_group_obj_off; + u64 next_group_obj_off; + + unsigned first_group_dev; + unsigned nex_group_dev; + unsigned max_devs; +}; + +void _calc_trunk_info(struct exofs_layout *layout, u64 file_offset, + struct _trunc_info *ti) +{ + unsigned stripe_unit = layout->stripe_unit; + + _calc_stripe_info(layout, file_offset, &ti->si); + + ti->prev_group_obj_off = ti->si.M * stripe_unit; + ti->next_group_obj_off = ti->si.M ? (ti->si.M - 1) * stripe_unit : 0; + + ti->first_group_dev = ti->si.dev - (ti->si.dev % layout->group_width); + ti->nex_group_dev = ti->first_group_dev + layout->group_width; + ti->max_devs = layout->group_width * layout->group_count; +} + int exofs_oi_truncate(struct exofs_i_info *oi, u64 size) { struct exofs_sb_info *sbi = oi->vfs_inode.i_sb->s_fs_info; @@ -753,14 +780,16 @@ int exofs_oi_truncate(struct exofs_i_info *oi, u64 size) struct osd_attr attr; __be64 newsize; } *size_attrs; - struct _striping_info si; + struct _trunc_info ti; int i, ret; ret = exofs_get_io_state(&sbi->layout, &ios); if (unlikely(ret)) return ret; - size_attrs = kcalloc(ios->layout->group_width, sizeof(*size_attrs), + _calc_trunk_info(ios->layout, size, &ti); + + size_attrs = kcalloc(ti.max_devs, sizeof(*size_attrs), GFP_KERNEL); if (unlikely(!size_attrs)) { ret = -ENOMEM; @@ -769,26 +798,30 @@ int exofs_oi_truncate(struct exofs_i_info *oi, u64 size) ios->obj.id = exofs_oi_objno(oi); ios->cred = oi->i_cred; - ios->numdevs = ios->layout->s_numdevs; - _calc_stripe_info(ios, size, &si); - for (i = 0; i < ios->layout->group_width; ++i) { + for (i = 0; i < ti.max_devs; ++i) { struct exofs_trunc_attr *size_attr = &size_attrs[i]; u64 obj_size; - if (i < si.dev) - obj_size = si.obj_offset + - ios->layout->stripe_unit - si.unit_off; - else if (i == si.dev) - obj_size = si.obj_offset; - else /* i > si.dev */ - obj_size = si.obj_offset - si.unit_off; + if (i < ti.first_group_dev) + obj_size = ti.prev_group_obj_off; + else if (i >= ti.nex_group_dev) + obj_size = ti.next_group_obj_off; + else if (i < ti.si.dev) /* dev within this group */ + obj_size = ti.si.obj_offset + + ios->layout->stripe_unit - ti.si.unit_off; + else if (i == ti.si.dev) + obj_size = ti.si.obj_offset; + else /* i > ti.dev */ + obj_size = ti.si.obj_offset - ti.si.unit_off; size_attr->newsize = cpu_to_be64(obj_size); size_attr->attr = g_attr_logical_length; size_attr->attr.val_ptr = &size_attr->newsize; + EXOFS_DBGMSG("trunc(0x%llx) obj_offset=0x%llx dev=%d\n", + _LLU(ios->obj.id), _LLU(obj_size), i); ret = _truncate_mirrors(ios, i * ios->layout->mirrors_p1, &size_attr->attr); if (unlikely(ret)) From 1eb9a4b8a3c8a141cf2ab27309f089923b69c707 Mon Sep 17 00:00:00 2001 From: Matthew Garrett Date: Wed, 6 Jul 2011 16:48:49 -0400 Subject: [PATCH 363/439] efi: Fix argument types for SetVariable() for ia64 The spec says this takes uint32 for attributes, not uintn. Signed-off-by: Matthew Garrett Signed-off-by: Tony Luck --- arch/ia64/kernel/efi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/ia64/kernel/efi.c b/arch/ia64/kernel/efi.c index 6fc03aff046c..c38d22e5e902 100644 --- a/arch/ia64/kernel/efi.c +++ b/arch/ia64/kernel/efi.c @@ -156,7 +156,7 @@ prefix##_get_next_variable (unsigned long *name_size, efi_char16_t *name, \ #define STUB_SET_VARIABLE(prefix, adjust_arg) \ static efi_status_t \ prefix##_set_variable (efi_char16_t *name, efi_guid_t *vendor, \ - unsigned long attr, unsigned long data_size, \ + u32 attr, unsigned long data_size, \ void *data) \ { \ struct ia64_fpreg fr[6]; \ From cbc158d6bfa1990f7869717bb5270867c66068d1 Mon Sep 17 00:00:00 2001 From: David Brown Date: Thu, 4 Aug 2011 09:24:31 -0700 Subject: [PATCH 364/439] cpuidle: Consistent spelling of cpuidle_idle_call() Commit a0bfa1373859e9d11dc92561a8667588803e42d8 mispells cpuidle_idle_call() on ARM and SH code. Fix this to be consistent. Cc: Kevin Hilman Cc: Paul Mundt Cc: x86@kernel.org Cc: Len Brown Signed-off-by: David Brown [ Also done by Mark Brown - th ebug has been around forever, and was noticed in -next, but the idle tree never picked it up. Bad bad bad ] Signed-off-by: Linus Torvalds --- arch/arm/kernel/process.c | 2 +- arch/sh/kernel/idle.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index d7ee0d4c072d..1a347f481e5e 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -197,7 +197,7 @@ void cpu_idle(void) cpu_relax(); } else { stop_critical_timings(); - if (cpuidle_call_idle()) + if (cpuidle_idle_call()) pm_idle(); start_critical_timings(); /* diff --git a/arch/sh/kernel/idle.c b/arch/sh/kernel/idle.c index 3c45de1db716..32114e0941ae 100644 --- a/arch/sh/kernel/idle.c +++ b/arch/sh/kernel/idle.c @@ -101,7 +101,7 @@ void cpu_idle(void) local_irq_disable(); /* Don't trace irqs off for idle */ stop_critical_timings(); - if (cpuidle_call_idle()) + if (cpuidle_idle_call()) pm_idle(); /* * Sanity check to ensure that pm_idle() returns From 140d0b2108faebc77c6523296e211e509cb9f5f9 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 4 Aug 2011 19:35:59 -1000 Subject: [PATCH 365/439] Do 'shm_init_ns()' in an early pure_initcall This isn't really critical any more, since other patches (commit 298507d4d2cf: "shm: optimize exit_shm()") have caused us to not actually need to touch the rw_mutex unless there are actual shm segments associated with the namespace, but we really should do tne shm_init_ns() earlier than we do now. This, together with commit 288d5abec831 ("Boot up with usermodehelper disabled") will mean that we really do initialize the initial ipc namespace data structure before we run any tasks. Tested-by: Marc Zyngier Signed-off-by: Linus Torvalds --- ipc/shm.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/ipc/shm.c b/ipc/shm.c index b5bae9d945b6..02ecf2c078fc 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -105,9 +105,16 @@ void shm_exit_ns(struct ipc_namespace *ns) } #endif -void __init shm_init (void) +static int __init ipc_ns_init(void) { shm_init_ns(&init_ipc_ns); + return 0; +} + +pure_initcall(ipc_ns_init); + +void __init shm_init (void) +{ ipc_init_proc_interface("sysvipc/shm", #if BITS_PER_LONG <= 32 " key shmid perms size cpid lpid nattch uid gid cuid cgid atime dtime ctime rss swap\n", From 9076d0e7e02b98f7a65df10d1956326c8d8ba61a Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 5 Aug 2011 00:53:57 -0700 Subject: [PATCH 366/439] sparc: Access kernel TSB using physical addressing when possible. On sun4v this is basically required since we point the hypervisor and the TSB walking hardware at these tables using physical addressing too. Signed-off-by: David S. Miller --- arch/sparc/include/asm/tsb.h | 51 ++++++++++++++++----------------- arch/sparc/kernel/ktlb.S | 24 ++++++++-------- arch/sparc/kernel/vmlinux.lds.S | 10 +++++++ arch/sparc/mm/init_64.c | 40 +++++++++++++++++++++++++- 4 files changed, 85 insertions(+), 40 deletions(-) diff --git a/arch/sparc/include/asm/tsb.h b/arch/sparc/include/asm/tsb.h index 83c571d8c8a7..1a8afd1ad04f 100644 --- a/arch/sparc/include/asm/tsb.h +++ b/arch/sparc/include/asm/tsb.h @@ -133,29 +133,6 @@ extern struct tsb_phys_patch_entry __tsb_phys_patch, __tsb_phys_patch_end; sub TSB, 0x8, TSB; \ TSB_STORE(TSB, TAG); -#define KTSB_LOAD_QUAD(TSB, REG) \ - ldda [TSB] ASI_NUCLEUS_QUAD_LDD, REG; - -#define KTSB_STORE(ADDR, VAL) \ - stxa VAL, [ADDR] ASI_N; - -#define KTSB_LOCK_TAG(TSB, REG1, REG2) \ -99: lduwa [TSB] ASI_N, REG1; \ - sethi %hi(TSB_TAG_LOCK_HIGH), REG2;\ - andcc REG1, REG2, %g0; \ - bne,pn %icc, 99b; \ - nop; \ - casa [TSB] ASI_N, REG1, REG2;\ - cmp REG1, REG2; \ - bne,pn %icc, 99b; \ - nop; \ - -#define KTSB_WRITE(TSB, TTE, TAG) \ - add TSB, 0x8, TSB; \ - stxa TTE, [TSB] ASI_N; \ - sub TSB, 0x8, TSB; \ - stxa TAG, [TSB] ASI_N; - /* Do a kernel page table walk. Leaves physical PTE pointer in * REG1. Jumps to FAIL_LABEL on early page table walk termination. * VADDR will not be clobbered, but REG2 will. @@ -239,6 +216,8 @@ extern struct tsb_phys_patch_entry __tsb_phys_patch, __tsb_phys_patch_end; (KERNEL_TSB_SIZE_BYTES / 16) #define KERNEL_TSB4M_NENTRIES 4096 +#define KTSB_PHYS_SHIFT 15 + /* Do a kernel TSB lookup at tl>0 on VADDR+TAG, branch to OK_LABEL * on TSB hit. REG1, REG2, REG3, and REG4 are used as temporaries * and the found TTE will be left in REG1. REG3 and REG4 must @@ -247,13 +226,22 @@ extern struct tsb_phys_patch_entry __tsb_phys_patch, __tsb_phys_patch_end; * VADDR and TAG will be preserved and not clobbered by this macro. */ #define KERN_TSB_LOOKUP_TL1(VADDR, TAG, REG1, REG2, REG3, REG4, OK_LABEL) \ - sethi %hi(swapper_tsb), REG1; \ +661: sethi %hi(swapper_tsb), REG1; \ or REG1, %lo(swapper_tsb), REG1; \ + .section .swapper_tsb_phys_patch, "ax"; \ + .word 661b; \ + .previous; \ +661: nop; \ + .section .tsb_ldquad_phys_patch, "ax"; \ + .word 661b; \ + sllx REG1, KTSB_PHYS_SHIFT, REG1; \ + sllx REG1, KTSB_PHYS_SHIFT, REG1; \ + .previous; \ srlx VADDR, PAGE_SHIFT, REG2; \ and REG2, (KERNEL_TSB_NENTRIES - 1), REG2; \ sllx REG2, 4, REG2; \ add REG1, REG2, REG2; \ - KTSB_LOAD_QUAD(REG2, REG3); \ + TSB_LOAD_QUAD(REG2, REG3); \ cmp REG3, TAG; \ be,a,pt %xcc, OK_LABEL; \ mov REG4, REG1; @@ -263,12 +251,21 @@ extern struct tsb_phys_patch_entry __tsb_phys_patch, __tsb_phys_patch_end; * we can make use of that for the index computation. */ #define KERN_TSB4M_LOOKUP_TL1(TAG, REG1, REG2, REG3, REG4, OK_LABEL) \ - sethi %hi(swapper_4m_tsb), REG1; \ +661: sethi %hi(swapper_4m_tsb), REG1; \ or REG1, %lo(swapper_4m_tsb), REG1; \ + .section .swapper_4m_tsb_phys_patch, "ax"; \ + .word 661b; \ + .previous; \ +661: nop; \ + .section .tsb_ldquad_phys_patch, "ax"; \ + .word 661b; \ + sllx REG1, KTSB_PHYS_SHIFT, REG1; \ + sllx REG1, KTSB_PHYS_SHIFT, REG1; \ + .previous; \ and TAG, (KERNEL_TSB4M_NENTRIES - 1), REG2; \ sllx REG2, 4, REG2; \ add REG1, REG2, REG2; \ - KTSB_LOAD_QUAD(REG2, REG3); \ + TSB_LOAD_QUAD(REG2, REG3); \ cmp REG3, TAG; \ be,a,pt %xcc, OK_LABEL; \ mov REG4, REG1; diff --git a/arch/sparc/kernel/ktlb.S b/arch/sparc/kernel/ktlb.S index 1d361477d7d6..79f310364849 100644 --- a/arch/sparc/kernel/ktlb.S +++ b/arch/sparc/kernel/ktlb.S @@ -47,16 +47,16 @@ kvmap_itlb_tsb_miss: kvmap_itlb_vmalloc_addr: KERN_PGTABLE_WALK(%g4, %g5, %g2, kvmap_itlb_longpath) - KTSB_LOCK_TAG(%g1, %g2, %g7) + TSB_LOCK_TAG(%g1, %g2, %g7) /* Load and check PTE. */ ldxa [%g5] ASI_PHYS_USE_EC, %g5 mov 1, %g7 sllx %g7, TSB_TAG_INVALID_BIT, %g7 brgez,a,pn %g5, kvmap_itlb_longpath - KTSB_STORE(%g1, %g7) + TSB_STORE(%g1, %g7) - KTSB_WRITE(%g1, %g5, %g6) + TSB_WRITE(%g1, %g5, %g6) /* fallthrough to TLB load */ @@ -102,9 +102,9 @@ kvmap_itlb_longpath: kvmap_itlb_obp: OBP_TRANS_LOOKUP(%g4, %g5, %g2, %g3, kvmap_itlb_longpath) - KTSB_LOCK_TAG(%g1, %g2, %g7) + TSB_LOCK_TAG(%g1, %g2, %g7) - KTSB_WRITE(%g1, %g5, %g6) + TSB_WRITE(%g1, %g5, %g6) ba,pt %xcc, kvmap_itlb_load nop @@ -112,17 +112,17 @@ kvmap_itlb_obp: kvmap_dtlb_obp: OBP_TRANS_LOOKUP(%g4, %g5, %g2, %g3, kvmap_dtlb_longpath) - KTSB_LOCK_TAG(%g1, %g2, %g7) + TSB_LOCK_TAG(%g1, %g2, %g7) - KTSB_WRITE(%g1, %g5, %g6) + TSB_WRITE(%g1, %g5, %g6) ba,pt %xcc, kvmap_dtlb_load nop .align 32 kvmap_dtlb_tsb4m_load: - KTSB_LOCK_TAG(%g1, %g2, %g7) - KTSB_WRITE(%g1, %g5, %g6) + TSB_LOCK_TAG(%g1, %g2, %g7) + TSB_WRITE(%g1, %g5, %g6) ba,pt %xcc, kvmap_dtlb_load nop @@ -222,16 +222,16 @@ kvmap_linear_patch: kvmap_dtlb_vmalloc_addr: KERN_PGTABLE_WALK(%g4, %g5, %g2, kvmap_dtlb_longpath) - KTSB_LOCK_TAG(%g1, %g2, %g7) + TSB_LOCK_TAG(%g1, %g2, %g7) /* Load and check PTE. */ ldxa [%g5] ASI_PHYS_USE_EC, %g5 mov 1, %g7 sllx %g7, TSB_TAG_INVALID_BIT, %g7 brgez,a,pn %g5, kvmap_dtlb_longpath - KTSB_STORE(%g1, %g7) + TSB_STORE(%g1, %g7) - KTSB_WRITE(%g1, %g5, %g6) + TSB_WRITE(%g1, %g5, %g6) /* fallthrough to TLB load */ diff --git a/arch/sparc/kernel/vmlinux.lds.S b/arch/sparc/kernel/vmlinux.lds.S index 94a954892d3f..0e1605697b49 100644 --- a/arch/sparc/kernel/vmlinux.lds.S +++ b/arch/sparc/kernel/vmlinux.lds.S @@ -107,6 +107,16 @@ SECTIONS *(.sun4v_2insn_patch) __sun4v_2insn_patch_end = .; } + .swapper_tsb_phys_patch : { + __swapper_tsb_phys_patch = .; + *(.swapper_tsb_phys_patch) + __swapper_tsb_phys_patch_end = .; + } + .swapper_4m_tsb_phys_patch : { + __swapper_4m_tsb_phys_patch = .; + *(.swapper_4m_tsb_phys_patch) + __swapper_4m_tsb_phys_patch_end = .; + } .popc_3insn_patch : { __popc_3insn_patch = .; *(.popc_3insn_patch) diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c index 3fd8e18bed80..adfac23d976a 100644 --- a/arch/sparc/mm/init_64.c +++ b/arch/sparc/mm/init_64.c @@ -1597,6 +1597,42 @@ static void __init tsb_phys_patch(void) static struct hv_tsb_descr ktsb_descr[NUM_KTSB_DESCR]; extern struct tsb swapper_tsb[KERNEL_TSB_NENTRIES]; +static void patch_one_ktsb_phys(unsigned int *start, unsigned int *end, unsigned long pa) +{ + pa >>= KTSB_PHYS_SHIFT; + + while (start < end) { + unsigned int *ia = (unsigned int *)(unsigned long)*start; + + ia[0] = (ia[0] & ~0x3fffff) | (pa >> 10); + __asm__ __volatile__("flush %0" : : "r" (ia)); + + ia[1] = (ia[1] & ~0x3ff) | (pa & 0x3ff); + __asm__ __volatile__("flush %0" : : "r" (ia + 1)); + + start++; + } +} + +static void ktsb_phys_patch(void) +{ + extern unsigned int __swapper_tsb_phys_patch; + extern unsigned int __swapper_tsb_phys_patch_end; + extern unsigned int __swapper_4m_tsb_phys_patch; + extern unsigned int __swapper_4m_tsb_phys_patch_end; + unsigned long ktsb_pa; + + ktsb_pa = kern_base + ((unsigned long)&swapper_tsb[0] - KERNBASE); + patch_one_ktsb_phys(&__swapper_tsb_phys_patch, + &__swapper_tsb_phys_patch_end, ktsb_pa); +#ifndef CONFIG_DEBUG_PAGEALLOC + ktsb_pa = (kern_base + + ((unsigned long)&swapper_4m_tsb[0] - KERNBASE)); + patch_one_ktsb_phys(&__swapper_4m_tsb_phys_patch, + &__swapper_4m_tsb_phys_patch_end, ktsb_pa); +#endif +} + static void __init sun4v_ktsb_init(void) { unsigned long ktsb_pa; @@ -1716,8 +1752,10 @@ void __init paging_init(void) sun4u_pgprot_init(); if (tlb_type == cheetah_plus || - tlb_type == hypervisor) + tlb_type == hypervisor) { tsb_phys_patch(); + ktsb_phys_patch(); + } if (tlb_type == hypervisor) { sun4v_patch_tlb_handlers(); From 961f65fc41cdc1f9099a6075258816c0db98e390 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 5 Aug 2011 02:38:27 -0700 Subject: [PATCH 367/439] sparc: Size mondo queues more sanely. There is currently no upper limit on the mondo queue sizes we'll use, which guarentees that we'll eventually his page allocation limits, and thus allocation failures, due to MAX_ORDER. Cap the sizes sanely, current limits are: CPU MONDO 2 * max_possible_cpus DEV MONDO 256 (basically NR_IRQS) RES MONDO 128 NRES MONDO 4 Signed-off-by: David S. Miller --- arch/sparc/kernel/mdesc.c | 30 +++++++++++++++++++++++------- 1 file changed, 23 insertions(+), 7 deletions(-) diff --git a/arch/sparc/kernel/mdesc.c b/arch/sparc/kernel/mdesc.c index 42f28c7420e1..acaebb63c4fd 100644 --- a/arch/sparc/kernel/mdesc.c +++ b/arch/sparc/kernel/mdesc.c @@ -508,6 +508,8 @@ const char *mdesc_node_name(struct mdesc_handle *hp, u64 node) } EXPORT_SYMBOL(mdesc_node_name); +static u64 max_cpus = 64; + static void __init report_platform_properties(void) { struct mdesc_handle *hp = mdesc_grab(); @@ -543,8 +545,10 @@ static void __init report_platform_properties(void) if (v) printk("PLATFORM: watchdog-max-timeout [%llu ms]\n", *v); v = mdesc_get_property(hp, pn, "max-cpus", NULL); - if (v) - printk("PLATFORM: max-cpus [%llu]\n", *v); + if (v) { + max_cpus = *v; + printk("PLATFORM: max-cpus [%llu]\n", max_cpus); + } #ifdef CONFIG_SMP { @@ -715,7 +719,7 @@ static void __cpuinit set_proc_ids(struct mdesc_handle *hp) } static void __cpuinit get_one_mondo_bits(const u64 *p, unsigned int *mask, - unsigned char def) + unsigned long def, unsigned long max) { u64 val; @@ -726,6 +730,9 @@ static void __cpuinit get_one_mondo_bits(const u64 *p, unsigned int *mask, if (!val || val >= 64) goto use_default; + if (val > max) + val = max; + *mask = ((1U << val) * 64U) - 1U; return; @@ -736,19 +743,28 @@ use_default: static void __cpuinit get_mondo_data(struct mdesc_handle *hp, u64 mp, struct trap_per_cpu *tb) { + static int printed; const u64 *val; val = mdesc_get_property(hp, mp, "q-cpu-mondo-#bits", NULL); - get_one_mondo_bits(val, &tb->cpu_mondo_qmask, 7); + get_one_mondo_bits(val, &tb->cpu_mondo_qmask, 7, ilog2(max_cpus * 2)); val = mdesc_get_property(hp, mp, "q-dev-mondo-#bits", NULL); - get_one_mondo_bits(val, &tb->dev_mondo_qmask, 7); + get_one_mondo_bits(val, &tb->dev_mondo_qmask, 7, 8); val = mdesc_get_property(hp, mp, "q-resumable-#bits", NULL); - get_one_mondo_bits(val, &tb->resum_qmask, 6); + get_one_mondo_bits(val, &tb->resum_qmask, 6, 7); val = mdesc_get_property(hp, mp, "q-nonresumable-#bits", NULL); - get_one_mondo_bits(val, &tb->nonresum_qmask, 2); + get_one_mondo_bits(val, &tb->nonresum_qmask, 2, 2); + if (!printed++) { + pr_info("SUN4V: Mondo queue sizes " + "[cpu(%u) dev(%u) r(%u) nr(%u)]\n", + tb->cpu_mondo_qmask + 1, + tb->dev_mondo_qmask + 1, + tb->resum_qmask + 1, + tb->nonresum_qmask + 1); + } } static void * __cpuinit mdesc_iterate_over_cpus(void *(*func)(struct mdesc_handle *, u64, int, void *), void *arg, cpumask_t *mask) From 39060a07781b4930656752943cf5d66376d0533c Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Fri, 5 Aug 2011 10:56:29 +0100 Subject: [PATCH 368/439] Revert "drm/i915: Try enabling RC6 by default (again)" This reverts commit 4e20fa65a3ea789510eed1a15deb9e8aab2b8202. Francesco Allertsen still has a broken configuration. Signed-off-by: Dave Airlie --- drivers/gpu/drm/i915/i915_drv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 60e4b9ef618b..ce045a8cf82c 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -62,7 +62,7 @@ module_param_named(semaphores, i915_semaphores, int, 0600); MODULE_PARM_DESC(semaphores, "Use semaphores for inter-ring sync (default: false)"); -unsigned int i915_enable_rc6 __read_mostly = 1; +unsigned int i915_enable_rc6 __read_mostly = 0; module_param_named(i915_enable_rc6, i915_enable_rc6, int, 0600); MODULE_PARM_DESC(i915_enable_rc6, "Enable power-saving render C-state 6 (default: true)"); From 728ffb86f10873aaf4abd26dde691ee40ae731fe Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Thu, 4 Aug 2011 14:07:38 +0000 Subject: [PATCH 369/439] net: sendmmsg should only return an error if no messages were sent sendmmsg uses a similar error return strategy as recvmmsg but it turns out to be a confusing way to communicate errors. The current code stores the error code away and returns it on the next sendmmsg call. This means a call with completely valid arguments could get an error from a previous call. Change things so we only return an error if no datagrams could be sent. If less than the requested number of messages were sent, the application must retry starting at the first failed one and if the problem is persistent the error will be returned. This matches the behaviour of other syscalls like read/write - it is not an error if less than the requested number of elements are sent. Signed-off-by: Anton Blanchard Cc: stable [3.0+] Signed-off-by: David S. Miller --- net/socket.c | 27 +++------------------------ 1 file changed, 3 insertions(+), 24 deletions(-) diff --git a/net/socket.c b/net/socket.c index b1cbbcd92558..e4ed2359eb43 100644 --- a/net/socket.c +++ b/net/socket.c @@ -2005,12 +2005,9 @@ int __sys_sendmmsg(int fd, struct mmsghdr __user *mmsg, unsigned int vlen, if (!sock) return err; - err = sock_error(sock->sk); - if (err) - goto out_put; - entry = mmsg; compat_entry = (struct compat_mmsghdr __user *)mmsg; + err = 0; while (datagrams < vlen) { /* @@ -2037,30 +2034,12 @@ int __sys_sendmmsg(int fd, struct mmsghdr __user *mmsg, unsigned int vlen, ++datagrams; } -out_put: fput_light(sock->file, fput_needed); - if (err == 0) + /* We only return an error if no datagrams were able to be sent */ + if (datagrams != 0) return datagrams; - if (datagrams != 0) { - /* - * We may send less entries than requested (vlen) if the - * sock is non blocking... - */ - if (err != -EAGAIN) { - /* - * ... or if sendmsg returns an error after we - * send some datagrams, where we record the - * error to return on the next call or if the - * app asks about it using getsockopt(SO_ERROR). - */ - sock->sk->sk_err = -err; - } - - return datagrams; - } - return err; } From 98382f419f32d2c12d021943b87dea555677144b Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Thu, 4 Aug 2011 14:07:39 +0000 Subject: [PATCH 370/439] net: Cap number of elements for sendmmsg To limit the amount of time we can spend in sendmmsg, cap the number of elements to UIO_MAXIOV (currently 1024). For error handling an application using sendmmsg needs to retry at the first unsent message, so capping is simpler and requires less application logic than returning EINVAL. Signed-off-by: Anton Blanchard Cc: stable [3.0+] Signed-off-by: David S. Miller --- net/socket.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/socket.c b/net/socket.c index e4ed2359eb43..b5c6de4f268a 100644 --- a/net/socket.c +++ b/net/socket.c @@ -1999,6 +1999,9 @@ int __sys_sendmmsg(int fd, struct mmsghdr __user *mmsg, unsigned int vlen, struct compat_mmsghdr __user *compat_entry; struct msghdr msg_sys; + if (vlen > UIO_MAXIOV) + vlen = UIO_MAXIOV; + datagrams = 0; sock = sockfd_lookup_light(fd, &err, &fput_needed); From c71d8ebe7a4496fb7231151cb70a6baa0cb56f9a Mon Sep 17 00:00:00 2001 From: Tetsuo Handa Date: Thu, 4 Aug 2011 14:07:40 +0000 Subject: [PATCH 371/439] net: Fix security_socket_sendmsg() bypass problem. The sendmmsg() introduced by commit 228e548e "net: Add sendmmsg socket system call" is capable of sending to multiple different destination addresses. SMACK is using destination's address for checking sendmsg() permission. However, security_socket_sendmsg() is called for only once even if multiple different destination addresses are passed to sendmmsg(). Therefore, we need to call security_socket_sendmsg() for each destination address rather than only the first destination address. Since calling security_socket_sendmsg() every time when only single destination address was passed to sendmmsg() is a waste of time, omit calling security_socket_sendmsg() unless destination address of previous datagram and that of current datagram differs. Signed-off-by: Tetsuo Handa Acked-by: Anton Blanchard Cc: stable [3.0+] Signed-off-by: David S. Miller --- net/socket.c | 43 ++++++++++++++++++++++++++++++++++--------- 1 file changed, 34 insertions(+), 9 deletions(-) diff --git a/net/socket.c b/net/socket.c index b5c6de4f268a..24a77400b65e 100644 --- a/net/socket.c +++ b/net/socket.c @@ -1871,8 +1871,14 @@ SYSCALL_DEFINE2(shutdown, int, fd, int, how) #define COMPAT_NAMELEN(msg) COMPAT_MSG(msg, msg_namelen) #define COMPAT_FLAGS(msg) COMPAT_MSG(msg, msg_flags) +struct used_address { + struct sockaddr_storage name; + unsigned int name_len; +}; + static int __sys_sendmsg(struct socket *sock, struct msghdr __user *msg, - struct msghdr *msg_sys, unsigned flags, int nosec) + struct msghdr *msg_sys, unsigned flags, + struct used_address *used_address) { struct compat_msghdr __user *msg_compat = (struct compat_msghdr __user *)msg; @@ -1953,8 +1959,28 @@ static int __sys_sendmsg(struct socket *sock, struct msghdr __user *msg, if (sock->file->f_flags & O_NONBLOCK) msg_sys->msg_flags |= MSG_DONTWAIT; - err = (nosec ? sock_sendmsg_nosec : sock_sendmsg)(sock, msg_sys, - total_len); + /* + * If this is sendmmsg() and current destination address is same as + * previously succeeded address, omit asking LSM's decision. + * used_address->name_len is initialized to UINT_MAX so that the first + * destination address never matches. + */ + if (used_address && used_address->name_len == msg_sys->msg_namelen && + !memcmp(&used_address->name, msg->msg_name, + used_address->name_len)) { + err = sock_sendmsg_nosec(sock, msg_sys, total_len); + goto out_freectl; + } + err = sock_sendmsg(sock, msg_sys, total_len); + /* + * If this is sendmmsg() and sending to current destination address was + * successful, remember it. + */ + if (used_address && err >= 0) { + used_address->name_len = msg_sys->msg_namelen; + memcpy(&used_address->name, msg->msg_name, + used_address->name_len); + } out_freectl: if (ctl_buf != ctl) @@ -1979,7 +2005,7 @@ SYSCALL_DEFINE3(sendmsg, int, fd, struct msghdr __user *, msg, unsigned, flags) if (!sock) goto out; - err = __sys_sendmsg(sock, msg, &msg_sys, flags, 0); + err = __sys_sendmsg(sock, msg, &msg_sys, flags, NULL); fput_light(sock->file, fput_needed); out: @@ -1998,6 +2024,7 @@ int __sys_sendmmsg(int fd, struct mmsghdr __user *mmsg, unsigned int vlen, struct mmsghdr __user *entry; struct compat_mmsghdr __user *compat_entry; struct msghdr msg_sys; + struct used_address used_address; if (vlen > UIO_MAXIOV) vlen = UIO_MAXIOV; @@ -2008,24 +2035,22 @@ int __sys_sendmmsg(int fd, struct mmsghdr __user *mmsg, unsigned int vlen, if (!sock) return err; + used_address.name_len = UINT_MAX; entry = mmsg; compat_entry = (struct compat_mmsghdr __user *)mmsg; err = 0; while (datagrams < vlen) { - /* - * No need to ask LSM for more than the first datagram. - */ if (MSG_CMSG_COMPAT & flags) { err = __sys_sendmsg(sock, (struct msghdr __user *)compat_entry, - &msg_sys, flags, datagrams); + &msg_sys, flags, &used_address); if (err < 0) break; err = __put_user(err, &compat_entry->msg_len); ++compat_entry; } else { err = __sys_sendmsg(sock, (struct msghdr __user *)entry, - &msg_sys, flags, datagrams); + &msg_sys, flags, &used_address); if (err < 0) break; err = put_user(err, &entry->msg_len); From 20e72a44098641f0c4de34a31287a93e006afb5b Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin Date: Thu, 4 Aug 2011 01:05:12 +0000 Subject: [PATCH 372/439] mlx4: decreasing ref count when removing mac For older FW versions, when a Mac address removed from Mac table, we should set 0 for reference count for the corresponding Mac index. Fixes a bug where removing Mac from the table still left that entry as invalid. Signed-off-by: Yevgeny Petrilin Tested-by: Roland Dreier Signed-off-by: David S. Miller --- drivers/net/mlx4/port.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/net/mlx4/port.c b/drivers/net/mlx4/port.c index 1f95afda6841..609e0ec14cee 100644 --- a/drivers/net/mlx4/port.c +++ b/drivers/net/mlx4/port.c @@ -258,9 +258,12 @@ void mlx4_unregister_mac(struct mlx4_dev *dev, u8 port, int qpn) if (validate_index(dev, table, index)) goto out; - table->entries[index] = 0; - mlx4_set_port_mac_table(dev, port, table->entries); - --table->total; + /* Check whether this address has reference count */ + if (!(--table->refs[index])) { + table->entries[index] = 0; + mlx4_set_port_mac_table(dev, port, table->entries); + --table->total; + } out: mutex_unlock(&table->mutex); } From c15fea2d8ca834dae491339c47e4fb3c81428190 Mon Sep 17 00:00:00 2001 From: Max Matveev Date: Fri, 5 Aug 2011 03:56:30 -0700 Subject: [PATCH 373/439] ipv6: check for IPv4 mapped addresses when connecting IPv6 sockets When support for binding to 'mapped INADDR_ANY (::ffff.0.0.0.0)' was added in 0f8d3c7ac3693d7b6c731bf2159273a59bf70e12 the rest of the code wasn't told so now it's possible to bind IPv6 datagram socket to ::ffff.0.0.0.0, connect it to another IPv4 address and it will all work except for getsockhame() which does not return the local address as expected. To give getsockname() something to work with check for 'mapped INADDR_ANY' when connecting and update the in-core source addresses appropriately. Signed-off-by: Max Matveev Signed-off-by: David S. Miller --- net/ipv6/datagram.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c index 16560336eb72..9ef1831746ef 100644 --- a/net/ipv6/datagram.c +++ b/net/ipv6/datagram.c @@ -33,6 +33,11 @@ #include #include +static inline int ipv6_mapped_addr_any(const struct in6_addr *a) +{ + return (ipv6_addr_v4mapped(a) && (a->s6_addr32[3] == 0)); +} + int ip6_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) { struct sockaddr_in6 *usin = (struct sockaddr_in6 *) uaddr; @@ -102,10 +107,12 @@ ipv4_connected: ipv6_addr_set_v4mapped(inet->inet_daddr, &np->daddr); - if (ipv6_addr_any(&np->saddr)) + if (ipv6_addr_any(&np->saddr) || + ipv6_mapped_addr_any(&np->saddr)) ipv6_addr_set_v4mapped(inet->inet_saddr, &np->saddr); - if (ipv6_addr_any(&np->rcv_saddr)) { + if (ipv6_addr_any(&np->rcv_saddr) || + ipv6_mapped_addr_any(&np->rcv_saddr)) { ipv6_addr_set_v4mapped(inet->inet_rcv_saddr, &np->rcv_saddr); if (sk->sk_prot->rehash) From c00c8aa2d976e9ed1d12a57b42d6e9b27efb7abe Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Thu, 4 Aug 2011 18:42:10 -0400 Subject: [PATCH 374/439] xen/trace: Fix compile error when CONFIG_XEN_PRIVILEGED_GUEST is not set MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit with CONFIG_XEN and CONFIG_FTRACE set we get this: arch/x86/xen/trace.c:22: error: ‘__HYPERVISOR_console_io’ undeclared here (not in a function) arch/x86/xen/trace.c:22: error: array index in initializer not of integer type arch/x86/xen/trace.c:22: error: (near initialization for ‘xen_hypercall_names’) arch/x86/xen/trace.c:23: error: ‘__HYPERVISOR_physdev_op_compat’ undeclared here (not in a function) Issue was that the definitions of __HYPERVISOR were not pulled if CONFIG_XEN_PRIVILEGED_GUEST was not set. Reported-by: Randy Dunlap Acked-by: Randy Dunlap Acked-by: Ingo Molnar Signed-off-by: Konrad Rzeszutek Wilk --- arch/x86/xen/trace.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/xen/trace.c b/arch/x86/xen/trace.c index 734beba2a08c..520022d1a181 100644 --- a/arch/x86/xen/trace.c +++ b/arch/x86/xen/trace.c @@ -1,4 +1,5 @@ #include +#include #define N(x) [__HYPERVISOR_##x] = "("#x")" static const char *xen_hypercall_names[] = { From f9e8c45002cacad536b338dfa9e910e341a49c31 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Fri, 5 Aug 2011 10:28:01 -0400 Subject: [PATCH 375/439] cifs: convert prefixpath delimiters in cifs_build_path_to_root Regression from 2.6.39... The delimiters in the prefixpath are not being converted based on whether posix paths are in effect. Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=727834 Reported-and-Tested-by: Iain Arnell Reported-by: Patrick Oltmann Cc: Pavel Shilovsky Cc: stable@kernel.org Signed-off-by: Jeff Layton Signed-off-by: Steve French --- fs/cifs/inode.c | 14 ++------------ 1 file changed, 2 insertions(+), 12 deletions(-) diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index 9b018c8334fa..a7b2dcd4a53e 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -764,20 +764,10 @@ char *cifs_build_path_to_root(struct smb_vol *vol, struct cifs_sb_info *cifs_sb, if (full_path == NULL) return full_path; - if (dfsplen) { + if (dfsplen) strncpy(full_path, tcon->treeName, dfsplen); - /* switch slash direction in prepath depending on whether - * windows or posix style path names - */ - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) { - int i; - for (i = 0; i < dfsplen; i++) { - if (full_path[i] == '\\') - full_path[i] = '/'; - } - } - } strncpy(full_path + dfsplen, vol->prepath, pplen); + convert_delimiter(full_path, CIFS_DIR_SEP(cifs_sb)); full_path[dfsplen + pplen] = 0; /* add trailing null */ return full_path; } From 80975d21aae2136ccae1ce914a1602dc1d8b0795 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Fri, 5 Aug 2011 09:02:40 -0400 Subject: [PATCH 376/439] cifs: cope with negative dentries in cifs_get_root The loop around lookup_one_len doesn't handle the case where it might return a negative dentry, which can cause an oops on the next pass through the loop. Check for that and break out of the loop with an error of -ENOENT if there is one. Fixes the panic reported here: https://bugzilla.redhat.com/show_bug.cgi?id=727927 Reported-by: TR Bentley Reported-by: Iain Arnell Cc: Al Viro Cc: stable@kernel.org Signed-off-by: Jeff Layton Signed-off-by: Steve French --- fs/cifs/cifsfs.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 212e5629cc1d..f93eb948d071 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -563,6 +563,10 @@ cifs_get_root(struct smb_vol *vol, struct super_block *sb) mutex_unlock(&dir->i_mutex); dput(dentry); dentry = child; + if (!dentry->d_inode) { + dput(dentry); + dentry = ERR_PTR(-ENOENT); + } } while (!IS_ERR(dentry)); _FreeXid(xid); kfree(full_path); From 7b8aca65db5dd1aaa6dc1e11f6bfcc0ecd6bc8a4 Mon Sep 17 00:00:00 2001 From: "Lee, Chun-Yi" Date: Tue, 31 May 2011 14:52:22 +0800 Subject: [PATCH 377/439] acer-wmi: schedule threeg and interface sysfs for feature removal we can now autodetect internal 3G device and already have the threeg rfkill device. So, we plan to remove threeg sysfs support for it's no longer necessary. We also plan to remove interface sysfs file that exposed which ACPI-WMI interface that was used by acer-wmi driver. It will replaced by information log when acer-wmi initial. We keep it around for userspace compatibility reasons, schedule removal in 2012. Cc: Carlos Corbacho Cc: Matthew Garrett Cc: Dmitry Torokhov Cc: Corentin Chary Acked-by: Thomas Renninger Signed-off-by: Lee, Chun-Yi Signed-off-by: Matthew Garrett --- Documentation/feature-removal-schedule.txt | 11 +++++++++++ drivers/platform/x86/acer-wmi.c | 11 +++++++++-- 2 files changed, 20 insertions(+), 2 deletions(-) diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 43f48098220d..c4a6e148732a 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -581,3 +581,14 @@ Why: This driver has been superseded by g_mass_storage. Who: Alan Stern ---------------------------- + +What: threeg and interface sysfs files in /sys/devices/platform/acer-wmi +When: 2012 +Why: In 3.0, we can now autodetect internal 3G device and already have + the threeg rfkill device. So, we plan to remove threeg sysfs support + for it's no longer necessary. + + We also plan to remove interface sysfs file that exposed which ACPI-WMI + interface that was used by acer-wmi driver. It will replaced by + information log when acer-wmi initial. +Who: Lee, Chun-Yi diff --git a/drivers/platform/x86/acer-wmi.c b/drivers/platform/x86/acer-wmi.c index e1c4938b301b..089db868799e 100644 --- a/drivers/platform/x86/acer-wmi.c +++ b/drivers/platform/x86/acer-wmi.c @@ -1400,6 +1400,9 @@ static ssize_t show_bool_threeg(struct device *dev, { u32 result; \ acpi_status status; + + pr_info("This threeg sysfs will be removed in 2012" + " - used by: %s\n", current->comm); if (wmi_has_guid(WMID_GUID3)) status = wmid3_get_device_status(&result, ACER_WMID3_GDS_THREEG); @@ -1415,8 +1418,10 @@ static ssize_t set_bool_threeg(struct device *dev, { u32 tmp = simple_strtoul(buf, NULL, 10); acpi_status status = set_u32(tmp, ACER_CAP_THREEG); - if (ACPI_FAILURE(status)) - return -EINVAL; + pr_info("This threeg sysfs will be removed in 2012" + " - used by: %s\n", current->comm); + if (ACPI_FAILURE(status)) + return -EINVAL; return count; } static DEVICE_ATTR(threeg, S_IRUGO | S_IWUSR, show_bool_threeg, @@ -1425,6 +1430,8 @@ static DEVICE_ATTR(threeg, S_IRUGO | S_IWUSR, show_bool_threeg, static ssize_t show_interface(struct device *dev, struct device_attribute *attr, char *buf) { + pr_info("This interface sysfs will be removed in 2012" + " - used by: %s\n", current->comm); switch (interface->type) { case ACER_AMW0: return sprintf(buf, "AMW0\n"); From 2605d753e488330f61000f1b2dc72d1668fba4ac Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Wed, 8 Jun 2011 14:56:42 +0800 Subject: [PATCH 378/439] platform-drivers-x86: dell-laptop: Remove unneeded mutex_init() for buffer_mutex DEFINE_MUTEX() will automatically initialize buffer_mutex, no need to call mutex_init() in dell_init(). Signed-off-by: Axel Lin Signed-off-by: Matthew Garrett --- drivers/platform/x86/dell-laptop.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/platform/x86/dell-laptop.c b/drivers/platform/x86/dell-laptop.c index e39ab1d3ed87..f31fa4efa725 100644 --- a/drivers/platform/x86/dell-laptop.c +++ b/drivers/platform/x86/dell-laptop.c @@ -612,7 +612,6 @@ static int __init dell_init(void) if (!bufferpage) goto fail_buffer; buffer = page_address(bufferpage); - mutex_init(&buffer_mutex); ret = dell_setup_rfkill(); From b06862ba6b98bf05f92772bbe36971a483e35fdf Mon Sep 17 00:00:00 2001 From: Julien Valroff Date: Fri, 24 Jun 2011 08:47:17 -0400 Subject: [PATCH 379/439] acerhdf: add support for Aspire 1810TZ BIOS v1.3314 Would you please consider applying the following patch adding support for the Aspire 1810TZ BIOS v.1.3314 version to the acerhdf module and avoids the following error: acerhdf: unknown (unsupported) BIOS version Acer/Aspire 1810TZ/v1.3314, ple= ase report, aborting! Not sure about the other Aspire models, but it seems at least 1810T should also be updated. Signed-off-by: Julien Valroff --- drivers/platform/x86/acerhdf.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/platform/x86/acerhdf.c b/drivers/platform/x86/acerhdf.c index fca3489218b7..d339756608af 100644 --- a/drivers/platform/x86/acerhdf.c +++ b/drivers/platform/x86/acerhdf.c @@ -182,6 +182,7 @@ static const struct bios_settings_t bios_tbl[] = { {"Acer", "Aspire 1810T", "v1.3308", 0x55, 0x58, {0x9e, 0x00} }, {"Acer", "Aspire 1810TZ", "v1.3310", 0x55, 0x58, {0x9e, 0x00} }, {"Acer", "Aspire 1810T", "v1.3310", 0x55, 0x58, {0x9e, 0x00} }, + {"Acer", "Aspire 1810TZ", "v1.3314", 0x55, 0x58, {0x9e, 0x00} }, /* Acer 531 */ {"Acer", "AO531h", "v0.3201", 0x55, 0x58, {0x20, 0x00} }, /* Gateway */ From 1a04d8ffc04c10fc50124f311d4c8c391f9a04ca Mon Sep 17 00:00:00 2001 From: Seth Forshee Date: Tue, 21 Jun 2011 12:00:32 -0500 Subject: [PATCH 380/439] acer-wmi: Add support for Aspire 1830 wlan hotkey Signed-off-by: Seth Forshee Signed-off-by: Matthew Garrett --- drivers/platform/x86/acer-wmi.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/platform/x86/acer-wmi.c b/drivers/platform/x86/acer-wmi.c index 089db868799e..fa7153020ca1 100644 --- a/drivers/platform/x86/acer-wmi.c +++ b/drivers/platform/x86/acer-wmi.c @@ -99,6 +99,7 @@ enum acer_wmi_event_ids { static const struct key_entry acer_wmi_keymap[] = { {KE_KEY, 0x01, {KEY_WLAN} }, /* WiFi */ {KE_KEY, 0x03, {KEY_WLAN} }, /* WiFi */ + {KE_KEY, 0x04, {KEY_WLAN} }, /* WiFi */ {KE_KEY, 0x12, {KEY_BLUETOOTH} }, /* BT */ {KE_KEY, 0x21, {KEY_PROG1} }, /* Backup */ {KE_KEY, 0x22, {KEY_PROG2} }, /* Arcade */ From 38803141bf6ccf8f20d05c6f48e6e12c3650ce9a Mon Sep 17 00:00:00 2001 From: "Lee, Chun-Yi" Date: Fri, 10 Jun 2011 15:24:26 +0800 Subject: [PATCH 381/439] msi-laptop: add MSI U270 netbook to module alias and scm list After test, msi-laptop driver also can support MSI U270 netbook. So, add MSI U270's dmi information to module alias and scm table for support this machine. Tested on MSI U270 netbook. Cc: Matthew Garrett Cc: Dmitry Torokhov Cc: Corentin Chary Cc: Thomas Renninger Signed-off-by: Lee, Chun-Yi Signed-off-by: Matthew Garrett --- drivers/platform/x86/msi-laptop.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/platform/x86/msi-laptop.c b/drivers/platform/x86/msi-laptop.c index 3ff629df9f01..f204643c5052 100644 --- a/drivers/platform/x86/msi-laptop.c +++ b/drivers/platform/x86/msi-laptop.c @@ -538,6 +538,15 @@ static struct dmi_system_id __initdata msi_load_scm_models_dmi_table[] = { }, .callback = dmi_check_cb }, + { + .ident = "MSI U270", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, + "Micro-Star International Co., Ltd."), + DMI_MATCH(DMI_PRODUCT_NAME, "U270 series"), + }, + .callback = dmi_check_cb + }, { } }; @@ -996,3 +1005,4 @@ MODULE_ALIAS("dmi:*:svnMICRO-STARINTERNATIONAL*:pnMS-N034:*"); MODULE_ALIAS("dmi:*:svnMICRO-STARINTERNATIONAL*:pnMS-N051:*"); MODULE_ALIAS("dmi:*:svnMICRO-STARINTERNATIONAL*:pnMS-N014:*"); MODULE_ALIAS("dmi:*:svnMicro-StarInternational*:pnCR620:*"); +MODULE_ALIAS("dmi:*:svnMicro-StarInternational*:pnU270series:*"); From 33009557bd9397c446a59e4cc91059a8e84c046b Mon Sep 17 00:00:00 2001 From: Andy Lutomirski Date: Tue, 24 May 2011 15:16:43 -0400 Subject: [PATCH 382/439] Add KEY_MICMUTE and enable it on Lenovo X220 I suspect that this works on T410. Signed-off-by: Andy Lutomirski Signed-off-by: Matthew Garrett --- drivers/platform/x86/thinkpad_acpi.c | 11 ++++++++++- include/linux/input.h | 2 ++ 2 files changed, 12 insertions(+), 1 deletion(-) diff --git a/drivers/platform/x86/thinkpad_acpi.c b/drivers/platform/x86/thinkpad_acpi.c index 26c5b117df22..7bd829f247eb 100644 --- a/drivers/platform/x86/thinkpad_acpi.c +++ b/drivers/platform/x86/thinkpad_acpi.c @@ -3185,9 +3185,18 @@ static int __init hotkey_init(struct ibm_init_struct *iibm) KEY_VENDOR, /* 0x17: Thinkpad/AccessIBM/Lenovo */ + /* (assignments unknown, please report if found) */ + KEY_UNKNOWN, KEY_UNKNOWN, + + /* + * The mic mute button only sends 0x1a. It does not + * automatically mute the mic or change the mute light. + */ + KEY_MICMUTE, /* 0x1a: Mic mute (since ?400 or so) */ + /* (assignments unknown, please report if found) */ KEY_UNKNOWN, KEY_UNKNOWN, KEY_UNKNOWN, KEY_UNKNOWN, - KEY_UNKNOWN, KEY_UNKNOWN, KEY_UNKNOWN, KEY_UNKNOWN, + KEY_UNKNOWN, }, }; diff --git a/include/linux/input.h b/include/linux/input.h index 068784e17972..a637e7814334 100644 --- a/include/linux/input.h +++ b/include/linux/input.h @@ -438,6 +438,8 @@ struct input_keymap_entry { #define KEY_WIMAX 246 #define KEY_RFKILL 247 /* Key that controls all radios */ +#define KEY_MICMUTE 248 /* Mute / unmute the microphone */ + /* Code 255 is reserved for special needs of AT keyboard driver */ #define BTN_MISC 0x100 From 3371f48167e04017125dd08cc1f70fa93d2f2e17 Mon Sep 17 00:00:00 2001 From: Ike Panhc Date: Thu, 30 Jun 2011 19:50:40 +0800 Subject: [PATCH 383/439] ideapad: define cfg bits and create sysfs node for cfg Create /sys/devices/platform/ideapad/cfg for showing cfg value. Signed-off-by: Ike Panhc Signed-off-by: Matthew Garrett --- .../ABI/testing/sysfs-platform-ideapad-laptop | 17 +++++++ drivers/platform/x86/ideapad-laptop.c | 51 +++++++++++++------ 2 files changed, 52 insertions(+), 16 deletions(-) diff --git a/Documentation/ABI/testing/sysfs-platform-ideapad-laptop b/Documentation/ABI/testing/sysfs-platform-ideapad-laptop index 807fca2ae2a4..ff53183c3848 100644 --- a/Documentation/ABI/testing/sysfs-platform-ideapad-laptop +++ b/Documentation/ABI/testing/sysfs-platform-ideapad-laptop @@ -4,3 +4,20 @@ KernelVersion: 2.6.37 Contact: "Ike Panhc " Description: Control the power of camera module. 1 means on, 0 means off. + +What: /sys/devices/platform/ideapad/cfg +Date: Jun 2011 +KernelVersion: 3.1 +Contact: "Ike Panhc " +Description: + Ideapad capability bits. + Bit 8-10: 1 - Intel graphic only + 2 - ATI graphic only + 3 - Nvidia graphic only + 4 - Intel and ATI graphic + 5 - Intel and Nvidia graphic + Bit 16: Bluetooth exist (1 for exist) + Bit 17: 3G exist (1 for exist) + Bit 18: Wifi exist (1 for exist) + Bit 19: Camera exist (1 for exist) + diff --git a/drivers/platform/x86/ideapad-laptop.c b/drivers/platform/x86/ideapad-laptop.c index bfdda33feb26..42b9ba7e903c 100644 --- a/drivers/platform/x86/ideapad-laptop.c +++ b/drivers/platform/x86/ideapad-laptop.c @@ -35,10 +35,15 @@ #define IDEAPAD_RFKILL_DEV_NUM (3) +#define CFG_BT_BIT (16) +#define CFG_3G_BIT (17) +#define CFG_WIFI_BIT (18) + struct ideapad_private { struct rfkill *rfk[IDEAPAD_RFKILL_DEV_NUM]; struct platform_device *platform_device; struct input_dev *inputdev; + unsigned long cfg; }; static acpi_handle ideapad_handle; @@ -155,7 +160,7 @@ static int write_ec_cmd(acpi_handle handle, int cmd, unsigned long data) } /* - * camera power + * sysfs */ static ssize_t show_ideapad_cam(struct device *dev, struct device_attribute *attr, @@ -186,6 +191,27 @@ static ssize_t store_ideapad_cam(struct device *dev, static DEVICE_ATTR(camera_power, 0644, show_ideapad_cam, store_ideapad_cam); +static ssize_t show_ideapad_cfg(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct ideapad_private *priv = dev_get_drvdata(dev); + + return sprintf(buf, "0x%.8lX\n", priv->cfg); +} + +static DEVICE_ATTR(cfg, 0444, show_ideapad_cfg, NULL); + +static struct attribute *ideapad_attributes[] = { + &dev_attr_camera_power.attr, + &dev_attr_cfg.attr, + NULL +}; + +static struct attribute_group ideapad_attribute_group = { + .attrs = ideapad_attributes +}; + /* * Rfkill */ @@ -197,9 +223,9 @@ struct ideapad_rfk_data { }; const struct ideapad_rfk_data ideapad_rfk_data[] = { - { "ideapad_wlan", 18, 0x15, RFKILL_TYPE_WLAN }, - { "ideapad_bluetooth", 16, 0x17, RFKILL_TYPE_BLUETOOTH }, - { "ideapad_3g", 17, 0x20, RFKILL_TYPE_WWAN }, + { "ideapad_wlan", CFG_WIFI_BIT, 0x15, RFKILL_TYPE_WLAN }, + { "ideapad_bluetooth", CFG_BT_BIT, 0x17, RFKILL_TYPE_BLUETOOTH }, + { "ideapad_3g", CFG_3G_BIT, 0x20, RFKILL_TYPE_WWAN }, }; static int ideapad_rfk_set(void *data, bool blocked) @@ -280,15 +306,6 @@ static void __devexit ideapad_unregister_rfkill(struct acpi_device *adevice, /* * Platform device */ -static struct attribute *ideapad_attributes[] = { - &dev_attr_camera_power.attr, - NULL -}; - -static struct attribute_group ideapad_attribute_group = { - .attrs = ideapad_attributes -}; - static int __devinit ideapad_platform_init(struct ideapad_private *priv) { int result; @@ -393,10 +410,11 @@ MODULE_DEVICE_TABLE(acpi, ideapad_device_ids); static int __devinit ideapad_acpi_add(struct acpi_device *adevice) { - int ret, i, cfg; + int ret, i; + unsigned long cfg; struct ideapad_private *priv; - if (read_method_int(adevice->handle, "_CFG", &cfg)) + if (read_method_int(adevice->handle, "_CFG", (int *)&cfg)) return -ENODEV; priv = kzalloc(sizeof(*priv), GFP_KERNEL); @@ -404,6 +422,7 @@ static int __devinit ideapad_acpi_add(struct acpi_device *adevice) return -ENOMEM; dev_set_drvdata(&adevice->dev, priv); ideapad_handle = adevice->handle; + priv->cfg = cfg; ret = ideapad_platform_init(priv); if (ret) @@ -414,7 +433,7 @@ static int __devinit ideapad_acpi_add(struct acpi_device *adevice) goto input_failed; for (i = 0; i < IDEAPAD_RFKILL_DEV_NUM; i++) { - if (test_bit(ideapad_rfk_data[i].cfgbit, (unsigned long *)&cfg)) + if (test_bit(ideapad_rfk_data[i].cfgbit, &cfg)) ideapad_register_rfkill(adevice, i); else priv->rfk[i] = NULL; From a84511f7fbeb37e26aacb9c72f5a21ffc24e909e Mon Sep 17 00:00:00 2001 From: Ike Panhc Date: Thu, 30 Jun 2011 19:50:47 +0800 Subject: [PATCH 384/439] ideapad: let camera_power node invisiable if no camera Signed-off-by: Ike Panhc Signed-off-by: Matthew Garrett --- drivers/platform/x86/ideapad-laptop.c | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) diff --git a/drivers/platform/x86/ideapad-laptop.c b/drivers/platform/x86/ideapad-laptop.c index 42b9ba7e903c..1612abde5188 100644 --- a/drivers/platform/x86/ideapad-laptop.c +++ b/drivers/platform/x86/ideapad-laptop.c @@ -38,6 +38,7 @@ #define CFG_BT_BIT (16) #define CFG_3G_BIT (17) #define CFG_WIFI_BIT (18) +#define CFG_CAMERA_BIT (19) struct ideapad_private { struct rfkill *rfk[IDEAPAD_RFKILL_DEV_NUM]; @@ -208,7 +209,24 @@ static struct attribute *ideapad_attributes[] = { NULL }; +static mode_t ideapad_is_visible(struct kobject *kobj, + struct attribute *attr, + int idx) +{ + struct device *dev = container_of(kobj, struct device, kobj); + struct ideapad_private *priv = dev_get_drvdata(dev); + bool supported; + + if (attr == &dev_attr_camera_power.attr) + supported = test_bit(CFG_CAMERA_BIT, &(priv->cfg)); + else + supported = true; + + return supported ? attr->mode : 0; +} + static struct attribute_group ideapad_attribute_group = { + .is_visible = ideapad_is_visible, .attrs = ideapad_attributes }; From a4ecbb8ae7be16497db2f984ee7a3ffec0f164c3 Mon Sep 17 00:00:00 2001 From: Ike Panhc Date: Thu, 30 Jun 2011 19:50:52 +0800 Subject: [PATCH 385/439] ideapad: add backlight driver When acpi_backlight=vendor in cmdline or no backlight support in acpi video device, ideapad-laptop will register backlight device and control brightness and backlight power via the command in VPC2004. Signed-off-by: Ike Panhc Signed-off-by: Matthew Garrett --- drivers/platform/x86/ideapad-laptop.c | 123 ++++++++++++++++++++++++-- 1 file changed, 117 insertions(+), 6 deletions(-) diff --git a/drivers/platform/x86/ideapad-laptop.c b/drivers/platform/x86/ideapad-laptop.c index 1612abde5188..8811c68f809d 100644 --- a/drivers/platform/x86/ideapad-laptop.c +++ b/drivers/platform/x86/ideapad-laptop.c @@ -32,6 +32,8 @@ #include #include #include +#include +#include #define IDEAPAD_RFKILL_DEV_NUM (3) @@ -44,6 +46,7 @@ struct ideapad_private { struct rfkill *rfk[IDEAPAD_RFKILL_DEV_NUM]; struct platform_device *platform_device; struct input_dev *inputdev; + struct backlight_device *blightdev; unsigned long cfg; }; @@ -309,8 +312,7 @@ static int __devinit ideapad_register_rfkill(struct acpi_device *adevice, return 0; } -static void __devexit ideapad_unregister_rfkill(struct acpi_device *adevice, - int dev) +static void ideapad_unregister_rfkill(struct acpi_device *adevice, int dev) { struct ideapad_private *priv = dev_get_drvdata(&adevice->dev); @@ -417,6 +419,98 @@ static void ideapad_input_report(struct ideapad_private *priv, sparse_keymap_report_event(priv->inputdev, scancode, 1, true); } +/* + * backlight + */ +static int ideapad_backlight_get_brightness(struct backlight_device *blightdev) +{ + unsigned long now; + + if (read_ec_data(ideapad_handle, 0x12, &now)) + return -EIO; + return now; +} + +static int ideapad_backlight_update_status(struct backlight_device *blightdev) +{ + if (write_ec_cmd(ideapad_handle, 0x13, blightdev->props.brightness)) + return -EIO; + if (write_ec_cmd(ideapad_handle, 0x33, + blightdev->props.power == FB_BLANK_POWERDOWN ? 0 : 1)) + return -EIO; + + return 0; +} + +static const struct backlight_ops ideapad_backlight_ops = { + .get_brightness = ideapad_backlight_get_brightness, + .update_status = ideapad_backlight_update_status, +}; + +static int ideapad_backlight_init(struct ideapad_private *priv) +{ + struct backlight_device *blightdev; + struct backlight_properties props; + unsigned long max, now, power; + + if (read_ec_data(ideapad_handle, 0x11, &max)) + return -EIO; + if (read_ec_data(ideapad_handle, 0x12, &now)) + return -EIO; + if (read_ec_data(ideapad_handle, 0x18, &power)) + return -EIO; + + memset(&props, 0, sizeof(struct backlight_properties)); + props.max_brightness = max; + props.type = BACKLIGHT_PLATFORM; + blightdev = backlight_device_register("ideapad", + &priv->platform_device->dev, + priv, + &ideapad_backlight_ops, + &props); + if (IS_ERR(blightdev)) { + pr_err("Could not register backlight device\n"); + return PTR_ERR(blightdev); + } + + priv->blightdev = blightdev; + blightdev->props.brightness = now; + blightdev->props.power = power ? FB_BLANK_UNBLANK : FB_BLANK_POWERDOWN; + backlight_update_status(blightdev); + + return 0; +} + +static void ideapad_backlight_exit(struct ideapad_private *priv) +{ + if (priv->blightdev) + backlight_device_unregister(priv->blightdev); + priv->blightdev = NULL; +} + +static void ideapad_backlight_notify_power(struct ideapad_private *priv) +{ + unsigned long power; + struct backlight_device *blightdev = priv->blightdev; + + if (read_ec_data(ideapad_handle, 0x18, &power)) + return; + blightdev->props.power = power ? FB_BLANK_UNBLANK : FB_BLANK_POWERDOWN; +} + +static void ideapad_backlight_notify_brightness(struct ideapad_private *priv) +{ + unsigned long now; + + /* if we control brightness via acpi video driver */ + if (priv->blightdev == NULL) { + read_ec_data(ideapad_handle, 0x12, &now); + return; + } + + backlight_force_update(priv->blightdev, BACKLIGHT_UPDATE_HOTKEY); +} + /* * module init/exit */ @@ -458,8 +552,17 @@ static int __devinit ideapad_acpi_add(struct acpi_device *adevice) } ideapad_sync_rfk_state(adevice); + if (!acpi_video_backlight_support()) { + ret = ideapad_backlight_init(priv); + if (ret && ret != -ENODEV) + goto backlight_failed; + } + return 0; +backlight_failed: + for (i = 0; i < IDEAPAD_RFKILL_DEV_NUM; i++) + ideapad_unregister_rfkill(adevice, i); input_failed: ideapad_platform_exit(priv); platform_failed: @@ -472,6 +575,7 @@ static int __devexit ideapad_acpi_remove(struct acpi_device *adevice, int type) struct ideapad_private *priv = dev_get_drvdata(&adevice->dev); int i; + ideapad_backlight_exit(priv); for (i = 0; i < IDEAPAD_RFKILL_DEV_NUM; i++) ideapad_unregister_rfkill(adevice, i); ideapad_input_exit(priv); @@ -496,12 +600,19 @@ static void ideapad_acpi_notify(struct acpi_device *adevice, u32 event) vpc1 = (vpc2 << 8) | vpc1; for (vpc_bit = 0; vpc_bit < 16; vpc_bit++) { if (test_bit(vpc_bit, &vpc1)) { - if (vpc_bit == 9) + switch (vpc_bit) { + case 9: ideapad_sync_rfk_state(adevice); - else if (vpc_bit == 4) - read_ec_data(handle, 0x12, &vpc2); - else + break; + case 4: + ideapad_backlight_notify_brightness(priv); + break; + case 2: + ideapad_backlight_notify_power(priv); + break; + default: ideapad_input_report(priv, vpc_bit); + } } } } From 49979d091d1847823c064301da1ec173619ddd92 Mon Sep 17 00:00:00 2001 From: Corentin Chary Date: Fri, 1 Jul 2011 11:34:26 +0200 Subject: [PATCH 386/439] asus-wmi: fix hwmon/pwm1 The code was completly broken, and should never had been sent to the kernel. That's what happens when you write code without hardware to test it. Cc: stable@kernel.org Signed-off-by: Corentin Chary Signed-off-by: Matthew Garrett --- drivers/platform/x86/asus-wmi.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/platform/x86/asus-wmi.c b/drivers/platform/x86/asus-wmi.c index 65b66aa44c78..bcb9fb41ddc4 100644 --- a/drivers/platform/x86/asus-wmi.c +++ b/drivers/platform/x86/asus-wmi.c @@ -797,8 +797,8 @@ exit: * Hwmon device */ static ssize_t asus_hwmon_pwm1(struct device *dev, - struct device_attribute *attr, - char *buf) + struct device_attribute *attr, + char *buf) { struct asus_wmi *asus = dev_get_drvdata(dev); u32 value; @@ -809,7 +809,7 @@ static ssize_t asus_hwmon_pwm1(struct device *dev, if (err < 0) return err; - value |= 0xFF; + value &= 0xFF; if (value == 1) /* Low Speed */ value = 85; @@ -869,7 +869,7 @@ static mode_t asus_hwmon_sysfs_is_visible(struct kobject *kobj, * - reverved bits are non-zero * - sfun and presence bit are not set */ - if (value != ASUS_WMI_UNSUPPORTED_METHOD || value & 0xFFF80000 + if (value == ASUS_WMI_UNSUPPORTED_METHOD || value & 0xFFF80000 || (!asus->sfun && !(value & ASUS_WMI_DSTS_PRESENCE_BIT))) ok = false; } @@ -904,6 +904,7 @@ static int asus_wmi_hwmon_init(struct asus_wmi *asus) pr_err("Could not register asus hwmon device\n"); return PTR_ERR(hwmon); } + dev_set_drvdata(hwmon, asus); asus->hwmon_device = hwmon; result = sysfs_create_group(&hwmon->kobj, &hwmon_attribute_group); if (result) From c4453f6a7ceff330ff37a9712ee8305fc3f8e9b6 Mon Sep 17 00:00:00 2001 From: Seth Forshee Date: Fri, 1 Jul 2011 11:34:27 +0200 Subject: [PATCH 387/439] asus-wmi: Add callback for hotkey filtering This is required for the T101MT home key, which behaves differently than other hotkeys. Signed-off-by: Seth Forshee Signed-off-by: Corentin Chary Signed-off-by: Matthew Garrett --- drivers/platform/x86/asus-wmi.c | 12 +++++++++++- drivers/platform/x86/asus-wmi.h | 6 ++++++ 2 files changed, 17 insertions(+), 1 deletion(-) diff --git a/drivers/platform/x86/asus-wmi.c b/drivers/platform/x86/asus-wmi.c index bcb9fb41ddc4..cea9fdcad64c 100644 --- a/drivers/platform/x86/asus-wmi.c +++ b/drivers/platform/x86/asus-wmi.c @@ -1061,6 +1061,8 @@ static void asus_wmi_notify(u32 value, void *context) acpi_status status; int code; int orig_code; + unsigned int key_value = 1; + bool autorelease = 1; status = wmi_get_event_data(value, &response); if (status != AE_OK) { @@ -1076,6 +1078,13 @@ static void asus_wmi_notify(u32 value, void *context) code = obj->integer.value; orig_code = code; + if (asus->driver->key_filter) { + asus->driver->key_filter(asus->driver, &code, &key_value, + &autorelease); + if (code == ASUS_WMI_KEY_IGNORE) + goto exit; + } + if (code >= NOTIFY_BRNUP_MIN && code <= NOTIFY_BRNUP_MAX) code = NOTIFY_BRNUP_MIN; else if (code >= NOTIFY_BRNDOWN_MIN && @@ -1085,7 +1094,8 @@ static void asus_wmi_notify(u32 value, void *context) if (code == NOTIFY_BRNUP_MIN || code == NOTIFY_BRNDOWN_MIN) { if (!acpi_video_backlight_support()) asus_wmi_backlight_notify(asus, orig_code); - } else if (!sparse_keymap_report_event(asus->inputdev, code, 1, true)) + } else if (!sparse_keymap_report_event(asus->inputdev, code, + key_value, autorelease)) pr_info("Unknown key %x pressed\n", code); exit: diff --git a/drivers/platform/x86/asus-wmi.h b/drivers/platform/x86/asus-wmi.h index c044522c8766..4da61033bcb9 100644 --- a/drivers/platform/x86/asus-wmi.h +++ b/drivers/platform/x86/asus-wmi.h @@ -29,6 +29,8 @@ #include +#define ASUS_WMI_KEY_IGNORE (-1) + struct module; struct key_entry; struct asus_wmi; @@ -44,6 +46,10 @@ struct asus_wmi_driver { const struct key_entry *keymap; const char *input_name; const char *input_phys; + /* Returns new code, value, and autorelease values in arguments. + * Return ASUS_WMI_KEY_IGNORE in code if event should be ignored. */ + void (*key_filter) (struct asus_wmi_driver *driver, int *code, + unsigned int *value, bool *autorelease); int (*probe) (struct platform_device *device); void (*quirks) (struct asus_wmi_driver *driver); From 6ae8b807371ba8d5b42648721c528582295136f7 Mon Sep 17 00:00:00 2001 From: Seth Forshee Date: Fri, 1 Jul 2011 11:34:28 +0200 Subject: [PATCH 388/439] eeepc-wmi: Add support for T101MT Home/Express Gate key This key is different than other hotkeys, having seperate scan codes for press, release, and hold, so it requires some special filtering. Press and release events are passed on, and hold events are ignored since sparse-keymap does not support hardware autorepeat. Note that "Home" in the context of this button doesn't mean the same thing as the usual Home key, and it really isn't clear at all what is meant by "Home". The manufacurer's description of the button indicates that it should launch some sort of touch screen settings interface on short press and apply a desktop rotation on long press. Signed-off-by: Seth Forshee Signed-off-by: Corentin Chary Signed-off-by: Matthew Garrett --- drivers/platform/x86/eeepc-wmi.c | 26 ++++++++++++++++++++++++++ 1 file changed, 26 insertions(+) diff --git a/drivers/platform/x86/eeepc-wmi.c b/drivers/platform/x86/eeepc-wmi.c index 4aa867a9b88b..f41a977efa94 100644 --- a/drivers/platform/x86/eeepc-wmi.c +++ b/drivers/platform/x86/eeepc-wmi.c @@ -56,6 +56,11 @@ MODULE_PARM_DESC(hotplug_wireless, "If your laptop needs that, please report to " "acpi4asus-user@lists.sourceforge.net."); +/* Values for T101MT "Home" key */ +#define HOME_PRESS 0xe4 +#define HOME_HOLD 0xea +#define HOME_RELEASE 0xe5 + static const struct key_entry eeepc_wmi_keymap[] = { /* Sleep already handled via generic ACPI code */ { KE_KEY, 0x30, { KEY_VOLUMEUP } }, @@ -71,6 +76,7 @@ static const struct key_entry eeepc_wmi_keymap[] = { { KE_KEY, 0xcc, { KEY_SWITCHVIDEOMODE } }, { KE_KEY, 0xe0, { KEY_PROG1 } }, /* Task Manager */ { KE_KEY, 0xe1, { KEY_F14 } }, /* Change Resolution */ + { KE_KEY, HOME_PRESS, { KEY_CONFIG } }, /* Home/Express gate key */ { KE_KEY, 0xe8, { KEY_SCREENLOCK } }, { KE_KEY, 0xe9, { KEY_BRIGHTNESS_ZERO } }, { KE_KEY, 0xeb, { KEY_CAMERA_ZOOMOUT } }, @@ -81,6 +87,25 @@ static const struct key_entry eeepc_wmi_keymap[] = { { KE_END, 0}, }; +static void eeepc_wmi_key_filter(struct asus_wmi_driver *asus_wmi, int *code, + unsigned int *value, bool *autorelease) +{ + switch (*code) { + case HOME_PRESS: + *value = 1; + *autorelease = 0; + break; + case HOME_HOLD: + *code = ASUS_WMI_KEY_IGNORE; + break; + case HOME_RELEASE: + *code = HOME_PRESS; + *value = 0; + *autorelease = 0; + break; + } +} + static acpi_status eeepc_wmi_parse_device(acpi_handle handle, u32 level, void *context, void **retval) { @@ -151,6 +176,7 @@ static struct asus_wmi_driver asus_wmi_driver = { .keymap = eeepc_wmi_keymap, .input_name = "Eee PC WMI hotkeys", .input_phys = EEEPC_WMI_FILE "/input0", + .key_filter = eeepc_wmi_key_filter, .probe = eeepc_wmi_probe, .quirks = eeepc_wmi_quirks, }; From 39bbde059973d3337ae5f2fae5097d52599415e4 Mon Sep 17 00:00:00 2001 From: Seth Forshee Date: Mon, 4 Jul 2011 09:49:20 +0200 Subject: [PATCH 389/439] asus-wmi: Enable autorepeat for hotkey input device The T101MT Home/Express Gate key autorepeats in hardware, but sparse-keymap does not support hardware autorepeat. Enable the input core's software autorepeat to emulate the hardware behavior. Normal hotkeys are autoreleased, so the behavior of these keys will not be affected. Signed-off-by: Seth Forshee Signed-off-by: Corentin Chary Signed-off-by: Matthew Garrett --- drivers/platform/x86/asus-wmi.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/platform/x86/asus-wmi.c b/drivers/platform/x86/asus-wmi.c index cea9fdcad64c..0fce7de77b1c 100644 --- a/drivers/platform/x86/asus-wmi.c +++ b/drivers/platform/x86/asus-wmi.c @@ -205,6 +205,7 @@ static int asus_wmi_input_init(struct asus_wmi *asus) asus->inputdev->phys = asus->driver->input_phys; asus->inputdev->id.bustype = BUS_HOST; asus->inputdev->dev.parent = &asus->platform_device->dev; + set_bit(EV_REP, asus->inputdev->evbit); err = sparse_keymap_setup(asus->inputdev, asus->driver->keymap, NULL); if (err) From 57d5c8e742dbb8cdc794d9e0fee769c9533072af Mon Sep 17 00:00:00 2001 From: Corentin Chary Date: Fri, 1 Jul 2011 11:34:30 +0200 Subject: [PATCH 390/439] asus-wmi: second part of the version starts at 16 not 8 Signed-off-by: Corentin Chary Signed-off-by: Matthew Garrett --- drivers/platform/x86/asus-wmi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/platform/x86/asus-wmi.c b/drivers/platform/x86/asus-wmi.c index 0fce7de77b1c..fa2b395145a6 100644 --- a/drivers/platform/x86/asus-wmi.c +++ b/drivers/platform/x86/asus-wmi.c @@ -1246,7 +1246,7 @@ static int asus_wmi_platform_init(struct asus_wmi *asus) /* We don't know yet what to do with this version... */ if (!asus_wmi_evaluate_method(ASUS_WMI_METHODID_SPEC, 0, 0x9, &rv)) { - pr_info("BIOS WMI version: %d.%d", rv >> 8, rv & 0xFF); + pr_info("BIOS WMI version: %d.%d", rv >> 16, rv & 0xFF); asus->spec = rv; } From e9809c0b9670656855655d8ed6dc33718ec12ba2 Mon Sep 17 00:00:00 2001 From: Corentin Chary Date: Fri, 1 Jul 2011 11:34:31 +0200 Subject: [PATCH 391/439] asus-wmi: add keyboard backlight support Based on a patch from Nate Weibley. . Cc: Nate Weibley Signed-off-by: Corentin Chary Signed-off-by: Matthew Garrett --- drivers/platform/x86/asus-wmi.c | 130 +++++++++++++++++++++++++++----- 1 file changed, 113 insertions(+), 17 deletions(-) diff --git a/drivers/platform/x86/asus-wmi.c b/drivers/platform/x86/asus-wmi.c index fa2b395145a6..83a7a8101fd6 100644 --- a/drivers/platform/x86/asus-wmi.c +++ b/drivers/platform/x86/asus-wmi.c @@ -66,6 +66,8 @@ MODULE_LICENSE("GPL"); #define NOTIFY_BRNUP_MAX 0x1f #define NOTIFY_BRNDOWN_MIN 0x20 #define NOTIFY_BRNDOWN_MAX 0x2e +#define NOTIFY_KBD_BRTUP 0xc4 +#define NOTIFY_KBD_BRTDWN 0xc5 /* WMI Methods */ #define ASUS_WMI_METHODID_SPEC 0x43455053 /* BIOS SPECification */ @@ -174,8 +176,11 @@ struct asus_wmi { struct led_classdev tpd_led; int tpd_led_wk; + struct led_classdev kbd_led; + int kbd_led_wk; struct workqueue_struct *led_workqueue; struct work_struct tpd_led_work; + struct work_struct kbd_led_work; struct asus_rfkill wlan; struct asus_rfkill bluetooth; @@ -360,30 +365,79 @@ static enum led_brightness tpd_led_get(struct led_classdev *led_cdev) return read_tpd_led_state(asus); } -static int asus_wmi_led_init(struct asus_wmi *asus) +static void kbd_led_update(struct work_struct *work) { - int rv; + int ctrl_param = 0; + struct asus_wmi *asus; - if (read_tpd_led_state(asus) < 0) - return 0; + asus = container_of(work, struct asus_wmi, kbd_led_work); - asus->led_workqueue = create_singlethread_workqueue("led_workqueue"); - if (!asus->led_workqueue) - return -ENOMEM; - INIT_WORK(&asus->tpd_led_work, tpd_led_update); + /* + * bits 0-2: level + * bit 7: light on/off + */ + if (asus->kbd_led_wk > 0) + ctrl_param = 0x80 | (asus->kbd_led_wk & 0x7F); - asus->tpd_led.name = "asus::touchpad"; - asus->tpd_led.brightness_set = tpd_led_set; - asus->tpd_led.brightness_get = tpd_led_get; - asus->tpd_led.max_brightness = 1; + asus_wmi_set_devstate(ASUS_WMI_DEVID_KBD_BACKLIGHT, ctrl_param, NULL); +} - rv = led_classdev_register(&asus->platform_device->dev, &asus->tpd_led); - if (rv) { - destroy_workqueue(asus->led_workqueue); - return rv; +static int kbd_led_read(struct asus_wmi *asus, int *level, int *env) +{ + int retval; + + /* + * bits 0-2: level + * bit 7: light on/off + * bit 8-10: environment (0: dark, 1: normal, 2: light) + * bit 17: status unknown + */ + retval = asus_wmi_get_devstate_bits(asus, ASUS_WMI_DEVID_KBD_BACKLIGHT, + 0xFFFF); + + if (retval == 0x8000) + retval = -ENODEV; + + if (retval >= 0) { + if (level) + *level = retval & 0x80 ? retval & 0x7F : 0; + if (env) + *env = (retval >> 8) & 0x7F; + retval = 0; } - return 0; + return retval; +} + +static void kbd_led_set(struct led_classdev *led_cdev, + enum led_brightness value) +{ + struct asus_wmi *asus; + + asus = container_of(led_cdev, struct asus_wmi, kbd_led); + + if (value > asus->kbd_led.max_brightness) + value = asus->kbd_led.max_brightness; + else if (value < 0) + value = 0; + + asus->kbd_led_wk = value; + queue_work(asus->led_workqueue, &asus->kbd_led_work); +} + +static enum led_brightness kbd_led_get(struct led_classdev *led_cdev) +{ + struct asus_wmi *asus; + int retval, value; + + asus = container_of(led_cdev, struct asus_wmi, kbd_led); + + retval = kbd_led_read(asus, &value, NULL); + + if (retval < 0) + return retval; + + return value; } static void asus_wmi_led_exit(struct asus_wmi *asus) @@ -394,6 +448,48 @@ static void asus_wmi_led_exit(struct asus_wmi *asus) destroy_workqueue(asus->led_workqueue); } +static int asus_wmi_led_init(struct asus_wmi *asus) +{ + int rv = 0; + + asus->led_workqueue = create_singlethread_workqueue("led_workqueue"); + if (!asus->led_workqueue) + return -ENOMEM; + + if (read_tpd_led_state(asus) >= 0) { + INIT_WORK(&asus->tpd_led_work, tpd_led_update); + + asus->tpd_led.name = "asus::touchpad"; + asus->tpd_led.brightness_set = tpd_led_set; + asus->tpd_led.brightness_get = tpd_led_get; + asus->tpd_led.max_brightness = 1; + + rv = led_classdev_register(&asus->platform_device->dev, + &asus->tpd_led); + if (rv) + goto error; + } + + if (kbd_led_read(asus, NULL, NULL) >= 0) { + INIT_WORK(&asus->kbd_led_work, kbd_led_update); + + asus->kbd_led.name = "asus::kbd_backlight"; + asus->kbd_led.brightness_set = kbd_led_set; + asus->kbd_led.brightness_get = kbd_led_get; + asus->kbd_led.max_brightness = 3; + + rv = led_classdev_register(&asus->platform_device->dev, + &asus->kbd_led); + } + +error: + if (rv) + asus_wmi_led_exit(asus); + + return rv; +} + + /* * PCI hotplug (for wlan rfkill) */ From 8fe8c25ef92ffc773a49c4b0dab2938a6fd64489 Mon Sep 17 00:00:00 2001 From: Corentin Chary Date: Fri, 1 Jul 2011 11:34:32 +0200 Subject: [PATCH 392/439] asus-wmi: fix section mismatch Signed-off-by: Corentin Chary Signed-off-by: Matthew Garrett --- drivers/platform/x86/asus-wmi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/platform/x86/asus-wmi.c b/drivers/platform/x86/asus-wmi.c index 83a7a8101fd6..41a96f2b136e 100644 --- a/drivers/platform/x86/asus-wmi.c +++ b/drivers/platform/x86/asus-wmi.c @@ -1712,7 +1712,7 @@ static int asus_wmi_probe(struct platform_device *pdev) static bool used; -int asus_wmi_register_driver(struct asus_wmi_driver *driver) +int __init_or_module asus_wmi_register_driver(struct asus_wmi_driver *driver) { struct platform_driver *platform_driver; struct platform_device *platform_device; From af965e973276b015d1b7d9ee78ed8d8df9d6bc98 Mon Sep 17 00:00:00 2001 From: Corentin Chary Date: Fri, 1 Jul 2011 11:34:34 +0200 Subject: [PATCH 393/439] asus-wmi: fix keyboard backlight detection Signed-off-by: Corentin Chary Signed-off-by: Matthew Garrett --- drivers/platform/x86/asus-wmi.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/platform/x86/asus-wmi.c b/drivers/platform/x86/asus-wmi.c index 41a96f2b136e..4b0384e140f8 100644 --- a/drivers/platform/x86/asus-wmi.c +++ b/drivers/platform/x86/asus-wmi.c @@ -395,8 +395,9 @@ static int kbd_led_read(struct asus_wmi *asus, int *level, int *env) retval = asus_wmi_get_devstate_bits(asus, ASUS_WMI_DEVID_KBD_BACKLIGHT, 0xFFFF); + /* Unknown status is considered as off */ if (retval == 0x8000) - retval = -ENODEV; + retval = 0; if (retval >= 0) { if (level) From 79ec1172c29ed521f1652b0b44dceb84ba115541 Mon Sep 17 00:00:00 2001 From: Corentin Chary Date: Fri, 1 Jul 2011 11:34:35 +0200 Subject: [PATCH 394/439] asus-wmi: add some device ids Signed-off-by: Corentin Chary Signed-off-by: Matthew Garrett --- drivers/platform/x86/asus-wmi.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/platform/x86/asus-wmi.c b/drivers/platform/x86/asus-wmi.c index 4b0384e140f8..e60385df4af2 100644 --- a/drivers/platform/x86/asus-wmi.c +++ b/drivers/platform/x86/asus-wmi.c @@ -95,6 +95,7 @@ MODULE_LICENSE("GPL"); /* Wireless */ #define ASUS_WMI_DEVID_HW_SWITCH 0x00010001 #define ASUS_WMI_DEVID_WIRELESS_LED 0x00010002 +#define ASUS_WMI_DEVID_CWAP 0x00010003 #define ASUS_WMI_DEVID_WLAN 0x00010011 #define ASUS_WMI_DEVID_BLUETOOTH 0x00010013 #define ASUS_WMI_DEVID_GPS 0x00010015 @@ -104,6 +105,12 @@ MODULE_LICENSE("GPL"); /* Leds */ /* 0x000200XX and 0x000400XX */ +#define ASUS_WMI_DEVID_LED1 0x00020011 +#define ASUS_WMI_DEVID_LED2 0x00020012 +#define ASUS_WMI_DEVID_LED3 0x00020013 +#define ASUS_WMI_DEVID_LED4 0x00020014 +#define ASUS_WMI_DEVID_LED5 0x00020015 +#define ASUS_WMI_DEVID_LED6 0x00020016 /* Backlight and Brightness */ #define ASUS_WMI_DEVID_BACKLIGHT 0x00050011 From 6118b8adb50c23714d5be089965082bbbb48d831 Mon Sep 17 00:00:00 2001 From: Corentin Chary Date: Fri, 1 Jul 2011 11:34:36 +0200 Subject: [PATCH 395/439] asus-wmi: add thermal sensor Signed-off-by: Corentin Chary Signed-off-by: Matthew Garrett --- drivers/platform/x86/asus-wmi.c | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+) diff --git a/drivers/platform/x86/asus-wmi.c b/drivers/platform/x86/asus-wmi.c index e60385df4af2..d01f767c7b69 100644 --- a/drivers/platform/x86/asus-wmi.c +++ b/drivers/platform/x86/asus-wmi.c @@ -44,6 +44,7 @@ #include #include #include +#include #include #include @@ -930,7 +931,26 @@ static ssize_t asus_hwmon_pwm1(struct device *dev, return sprintf(buf, "%d\n", value); } +static ssize_t asus_hwmon_temp1(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct asus_wmi *asus = dev_get_drvdata(dev); + u32 value; + int err; + + err = asus_wmi_get_devstate(asus, ASUS_WMI_DEVID_THERMAL_CTRL, &value); + + if (err < 0) + return err; + + value = KELVIN_TO_CELSIUS((value & 0xFFFF)) * 1000; + + return sprintf(buf, "%d\n", value); +} + static SENSOR_DEVICE_ATTR(pwm1, S_IRUGO, asus_hwmon_pwm1, NULL, 0); +static SENSOR_DEVICE_ATTR(temp1_input, S_IRUGO, asus_hwmon_temp1, NULL, 0); static ssize_t show_name(struct device *dev, struct device_attribute *attr, char *buf) @@ -941,6 +961,7 @@ static SENSOR_DEVICE_ATTR(name, S_IRUGO, show_name, NULL, 0); static struct attribute *hwmon_attributes[] = { &sensor_dev_attr_pwm1.dev_attr.attr, + &sensor_dev_attr_temp1_input.dev_attr.attr, &sensor_dev_attr_name.dev_attr.attr, NULL }; From e02431d6f5e8f1023ce0cfbaf70ddf4afae924d8 Mon Sep 17 00:00:00 2001 From: Corentin Chary Date: Fri, 1 Jul 2011 11:34:37 +0200 Subject: [PATCH 396/439] asus-wmi: check for temp1 presence Signed-off-by: Corentin Chary Signed-off-by: Matthew Garrett --- drivers/platform/x86/asus-wmi.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/platform/x86/asus-wmi.c b/drivers/platform/x86/asus-wmi.c index d01f767c7b69..e50872632c5a 100644 --- a/drivers/platform/x86/asus-wmi.c +++ b/drivers/platform/x86/asus-wmi.c @@ -967,7 +967,7 @@ static struct attribute *hwmon_attributes[] = { }; static mode_t asus_hwmon_sysfs_is_visible(struct kobject *kobj, - struct attribute *attr, int idx) + struct attribute *attr, int idx) { struct device *dev = container_of(kobj, struct device, kobj); struct platform_device *pdev = to_platform_device(dev->parent); @@ -978,6 +978,8 @@ static mode_t asus_hwmon_sysfs_is_visible(struct kobject *kobj, if (attr == &sensor_dev_attr_pwm1.dev_attr.attr) dev_id = ASUS_WMI_DEVID_FAN_CTRL; + else if (attr == &sensor_dev_attr_temp1_input.dev_attr.attr) + dev_id = ASUS_WMI_DEVID_THERMAL_CTRL; if (dev_id != -1) { int err = asus_wmi_get_devstate(asus, dev_id, &value); @@ -998,6 +1000,10 @@ static mode_t asus_hwmon_sysfs_is_visible(struct kobject *kobj, if (value == ASUS_WMI_UNSUPPORTED_METHOD || value & 0xFFF80000 || (!asus->sfun && !(value & ASUS_WMI_DSTS_PRESENCE_BIT))) ok = false; + } else if (dev_id == ASUS_WMI_DEVID_THERMAL_CTRL) { + /* If value is zero, something is clearly wrong */ + if (value == 0) + ok = false; } return ok ? attr->mode : 0; From 3df5fdadf6400373a696bb14e27d4771e5f6afb3 Mon Sep 17 00:00:00 2001 From: Corentin Chary Date: Fri, 1 Jul 2011 11:34:38 +0200 Subject: [PATCH 397/439] asus-wmi: return proper value in store_cpufv() Cc: stable@kernel.org Signed-off-by: Corentin Chary Signed-off-by: Matthew Garrett --- drivers/platform/x86/asus-wmi.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/platform/x86/asus-wmi.c b/drivers/platform/x86/asus-wmi.c index e50872632c5a..3b9fb91e25ff 100644 --- a/drivers/platform/x86/asus-wmi.c +++ b/drivers/platform/x86/asus-wmi.c @@ -1307,14 +1307,18 @@ ASUS_WMI_CREATE_DEVICE_ATTR(cardr, 0644, ASUS_WMI_DEVID_CARDREADER); static ssize_t store_cpufv(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - int value; + int value, rv; if (!count || sscanf(buf, "%i", &value) != 1) return -EINVAL; if (value < 0 || value > 2) return -EINVAL; - return asus_wmi_evaluate_method(ASUS_WMI_METHODID_CFVS, value, 0, NULL); + rv = asus_wmi_evaluate_method(ASUS_WMI_METHODID_CFVS, value, 0, NULL); + if (rv < 0) + return rv; + + return count; } static DEVICE_ATTR(cpufv, S_IRUGO | S_IWUSR, NULL, store_cpufv); From fddbfed595b307f9dddc7a86ddfbcbcb80141e28 Mon Sep 17 00:00:00 2001 From: Corentin Chary Date: Fri, 1 Jul 2011 11:34:39 +0200 Subject: [PATCH 398/439] asus-wmi: add CWAP support and clarify the meaning of WAPF bits ref: http://dev.iksaif.net/projects/3/wiki/Asus-laptop_WAPF Signed-off-by: Corentin Chary Signed-off-by: Matthew Garrett --- drivers/platform/x86/asus-laptop.c | 9 ++++----- drivers/platform/x86/asus-nb-wmi.c | 27 +++++++++++++++++++++++---- drivers/platform/x86/asus-wmi.c | 6 ++++++ drivers/platform/x86/asus-wmi.h | 1 + drivers/platform/x86/eeepc-wmi.c | 1 + 5 files changed, 35 insertions(+), 9 deletions(-) diff --git a/drivers/platform/x86/asus-laptop.c b/drivers/platform/x86/asus-laptop.c index d65df92e2acc..fa6d7ec68b26 100644 --- a/drivers/platform/x86/asus-laptop.c +++ b/drivers/platform/x86/asus-laptop.c @@ -70,11 +70,10 @@ MODULE_LICENSE("GPL"); * WAPF defines the behavior of the Fn+Fx wlan key * The significance of values is yet to be found, but * most of the time: - * 0x0 will do nothing - * 0x1 will allow to control the device with Fn+Fx key. - * 0x4 will send an ACPI event (0x88) while pressing the Fn+Fx key - * 0x5 like 0x1 or 0x4 - * So, if something doesn't work as you want, just try other values =) + * Bit | Bluetooth | WLAN + * 0 | Hardware | Hardware + * 1 | Hardware | Software + * 4 | Software | Software */ static uint wapf = 1; module_param(wapf, uint, 0444); diff --git a/drivers/platform/x86/asus-nb-wmi.c b/drivers/platform/x86/asus-nb-wmi.c index 0580d99b0798..b0859d4183e8 100644 --- a/drivers/platform/x86/asus-nb-wmi.c +++ b/drivers/platform/x86/asus-nb-wmi.c @@ -38,6 +38,24 @@ MODULE_LICENSE("GPL"); MODULE_ALIAS("wmi:"ASUS_NB_WMI_EVENT_GUID); +/* + * WAPF defines the behavior of the Fn+Fx wlan key + * The significance of values is yet to be found, but + * most of the time: + * Bit | Bluetooth | WLAN + * 0 | Hardware | Hardware + * 1 | Hardware | Software + * 4 | Software | Software + */ +static uint wapf; +module_param(wapf, uint, 0444); +MODULE_PARM_DESC(wapf, "WAPF value"); + +static void asus_nb_wmi_quirks(struct asus_wmi_driver *driver) +{ + driver->wapf = wapf; +} + static const struct key_entry asus_nb_wmi_keymap[] = { { KE_KEY, 0x30, { KEY_VOLUMEUP } }, { KE_KEY, 0x31, { KEY_VOLUMEDOWN } }, @@ -53,16 +71,16 @@ static const struct key_entry asus_nb_wmi_keymap[] = { { KE_KEY, 0x51, { KEY_WWW } }, { KE_KEY, 0x55, { KEY_CALC } }, { KE_KEY, 0x5C, { KEY_F15 } }, /* Power Gear key */ - { KE_KEY, 0x5D, { KEY_WLAN } }, - { KE_KEY, 0x5E, { KEY_WLAN } }, - { KE_KEY, 0x5F, { KEY_WLAN } }, + { KE_KEY, 0x5D, { KEY_WLAN } }, /* Wireless console Toggle */ + { KE_KEY, 0x5E, { KEY_WLAN } }, /* Wireless console Enable */ + { KE_KEY, 0x5F, { KEY_WLAN } }, /* Wireless console Disable */ { KE_KEY, 0x60, { KEY_SWITCHVIDEOMODE } }, { KE_KEY, 0x61, { KEY_SWITCHVIDEOMODE } }, { KE_KEY, 0x62, { KEY_SWITCHVIDEOMODE } }, { KE_KEY, 0x63, { KEY_SWITCHVIDEOMODE } }, { KE_KEY, 0x6B, { KEY_TOUCHPAD_TOGGLE } }, - { KE_KEY, 0x7E, { KEY_BLUETOOTH } }, { KE_KEY, 0x7D, { KEY_BLUETOOTH } }, + { KE_KEY, 0x7E, { KEY_BLUETOOTH } }, { KE_KEY, 0x82, { KEY_CAMERA } }, { KE_KEY, 0x88, { KEY_RFKILL } }, { KE_KEY, 0x8A, { KEY_PROG1 } }, @@ -81,6 +99,7 @@ static struct asus_wmi_driver asus_nb_wmi_driver = { .keymap = asus_nb_wmi_keymap, .input_name = "Asus WMI hotkeys", .input_phys = ASUS_NB_WMI_FILE "/input0", + .quirks = asus_nb_wmi_quirks, }; diff --git a/drivers/platform/x86/asus-wmi.c b/drivers/platform/x86/asus-wmi.c index 3b9fb91e25ff..8787dfacf11d 100644 --- a/drivers/platform/x86/asus-wmi.c +++ b/drivers/platform/x86/asus-wmi.c @@ -1413,6 +1413,12 @@ static int asus_wmi_platform_init(struct asus_wmi *asus) return -ENODEV; } + /* CWAP allow to define the behavior of the Fn+F2 key, + * this method doesn't seems to be present on Eee PCs */ + if (asus->driver->wapf >= 0) + asus_wmi_set_devstate(ASUS_WMI_DEVID_CWAP, + asus->driver->wapf, NULL); + return asus_wmi_sysfs_init(asus->platform_device); } diff --git a/drivers/platform/x86/asus-wmi.h b/drivers/platform/x86/asus-wmi.h index 4da61033bcb9..8147c10161cc 100644 --- a/drivers/platform/x86/asus-wmi.h +++ b/drivers/platform/x86/asus-wmi.h @@ -37,6 +37,7 @@ struct asus_wmi; struct asus_wmi_driver { bool hotplug_wireless; + int wapf; const char *name; struct module *owner; diff --git a/drivers/platform/x86/eeepc-wmi.c b/drivers/platform/x86/eeepc-wmi.c index f41a977efa94..9f6e64302b45 100644 --- a/drivers/platform/x86/eeepc-wmi.c +++ b/drivers/platform/x86/eeepc-wmi.c @@ -166,6 +166,7 @@ static void eeepc_dmi_check(struct asus_wmi_driver *driver) static void eeepc_wmi_quirks(struct asus_wmi_driver *driver) { driver->hotplug_wireless = hotplug_wireless; + driver->wapf = -1; eeepc_dmi_check(driver); } From 43be8bde1fdfbe3f4dedfd04ca4f9d6dd1e04b46 Mon Sep 17 00:00:00 2001 From: Corentin Chary Date: Fri, 1 Jul 2011 11:34:40 +0200 Subject: [PATCH 399/439] asus-wmi: add gps rfkill support Signed-off-by: Corentin Chary Signed-off-by: Matthew Garrett --- drivers/platform/x86/asus-wmi.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/drivers/platform/x86/asus-wmi.c b/drivers/platform/x86/asus-wmi.c index 8787dfacf11d..57514deb8e2b 100644 --- a/drivers/platform/x86/asus-wmi.c +++ b/drivers/platform/x86/asus-wmi.c @@ -194,6 +194,7 @@ struct asus_wmi { struct asus_rfkill bluetooth; struct asus_rfkill wimax; struct asus_rfkill wwan3g; + struct asus_rfkill gps; struct hotplug_slot *hotplug_slot; struct mutex hotplug_lock; @@ -835,6 +836,11 @@ static void asus_wmi_rfkill_exit(struct asus_wmi *asus) rfkill_destroy(asus->wwan3g.rfkill); asus->wwan3g.rfkill = NULL; } + if (asus->gps.rfkill) { + rfkill_unregister(asus->gps.rfkill); + rfkill_destroy(asus->gps.rfkill); + asus->gps.rfkill = NULL; + } } static int asus_wmi_rfkill_init(struct asus_wmi *asus) @@ -869,6 +875,12 @@ static int asus_wmi_rfkill_init(struct asus_wmi *asus) if (result && result != -ENODEV) goto exit; + result = asus_new_rfkill(asus, &asus->gps, "asus-gps", + RFKILL_TYPE_GPS, ASUS_WMI_DEVID_GPS); + + if (result && result != -ENODEV) + goto exit; + if (!asus->driver->hotplug_wireless) goto exit; @@ -1721,6 +1733,10 @@ static int asus_hotk_restore(struct device *device) bl = !asus_wmi_get_devstate_simple(asus, ASUS_WMI_DEVID_WWAN3G); rfkill_set_sw_state(asus->wwan3g.rfkill, bl); } + if (asus->gps.rfkill) { + bl = !asus_wmi_get_devstate_simple(asus, ASUS_WMI_DEVID_GPS); + rfkill_set_sw_state(asus->gps.rfkill, bl); + } return 0; } From a912d329102aa13c572989c6adc103685c9d24ae Mon Sep 17 00:00:00 2001 From: Corentin Chary Date: Fri, 1 Jul 2011 11:34:41 +0200 Subject: [PATCH 400/439] asus-wmi: add uwb rfkill support Signed-off-by: Corentin Chary Signed-off-by: Matthew Garrett --- drivers/platform/x86/asus-wmi.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/drivers/platform/x86/asus-wmi.c b/drivers/platform/x86/asus-wmi.c index 57514deb8e2b..95cba9ebf6c0 100644 --- a/drivers/platform/x86/asus-wmi.c +++ b/drivers/platform/x86/asus-wmi.c @@ -195,6 +195,7 @@ struct asus_wmi { struct asus_rfkill wimax; struct asus_rfkill wwan3g; struct asus_rfkill gps; + struct asus_rfkill uwb; struct hotplug_slot *hotplug_slot; struct mutex hotplug_lock; @@ -841,6 +842,11 @@ static void asus_wmi_rfkill_exit(struct asus_wmi *asus) rfkill_destroy(asus->gps.rfkill); asus->gps.rfkill = NULL; } + if (asus->uwb.rfkill) { + rfkill_unregister(asus->uwb.rfkill); + rfkill_destroy(asus->uwb.rfkill); + asus->uwb.rfkill = NULL; + } } static int asus_wmi_rfkill_init(struct asus_wmi *asus) @@ -881,6 +887,12 @@ static int asus_wmi_rfkill_init(struct asus_wmi *asus) if (result && result != -ENODEV) goto exit; + result = asus_new_rfkill(asus, &asus->uwb, "asus-uwb", + RFKILL_TYPE_UWB, ASUS_WMI_DEVID_UWB); + + if (result && result != -ENODEV) + goto exit; + if (!asus->driver->hotplug_wireless) goto exit; @@ -1737,6 +1749,10 @@ static int asus_hotk_restore(struct device *device) bl = !asus_wmi_get_devstate_simple(asus, ASUS_WMI_DEVID_GPS); rfkill_set_sw_state(asus->gps.rfkill, bl); } + if (asus->uwb.rfkill) { + bl = !asus_wmi_get_devstate_simple(asus, ASUS_WMI_DEVID_UWB); + rfkill_set_sw_state(asus->uwb.rfkill, bl); + } return 0; } From 4e2441c046737d3fb452ae46f0e4b12e18cf440d Mon Sep 17 00:00:00 2001 From: J Witteveen Date: Sun, 3 Jul 2011 13:15:44 +0200 Subject: [PATCH 401/439] Samsung Laptop platform driver: support N510 The N510 benefits from this code as well. Below is a patch to include support. Signed-off-by: Jouke Witteveen Signed-off-by: Matthew Garrett --- drivers/platform/x86/samsung-laptop.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/platform/x86/samsung-laptop.c b/drivers/platform/x86/samsung-laptop.c index d347116d150e..e12c88e92c4e 100644 --- a/drivers/platform/x86/samsung-laptop.c +++ b/drivers/platform/x86/samsung-laptop.c @@ -520,6 +520,16 @@ static struct dmi_system_id __initdata samsung_dmi_table[] = { }, .callback = dmi_check_cb, }, + { + .ident = "N510", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, + "SAMSUNG ELECTRONICS CO., LTD."), + DMI_MATCH(DMI_PRODUCT_NAME, "N510"), + DMI_MATCH(DMI_BOARD_NAME, "N510"), + }, + .callback = dmi_check_cb, + }, { .ident = "X125", .matches = { From 94ee48b9287c5c22c90754a48dbcf80b2df99d5f Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Wed, 6 Jul 2011 11:05:20 +0800 Subject: [PATCH 402/439] platform-drivers-x86: msi-wmi: add missing sparse_keymap_free in msi_wmi_init error path Signed-off-by: Axel Lin Signed-off-by: Matthew Garrett --- drivers/platform/x86/msi-wmi.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/platform/x86/msi-wmi.c b/drivers/platform/x86/msi-wmi.c index c832e3356cd6..6f40bf202dc7 100644 --- a/drivers/platform/x86/msi-wmi.c +++ b/drivers/platform/x86/msi-wmi.c @@ -272,6 +272,7 @@ static int __init msi_wmi_init(void) err_free_backlight: backlight_device_unregister(backlight); err_free_input: + sparse_keymap_free(msi_wmi_input_dev); input_unregister_device(msi_wmi_input_dev); err_uninstall_notifier: wmi_remove_notify_handler(MSIWMI_EVENT_GUID); From 03f8952cf63b6059e56074be6ed450dc7739cdee Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Wed, 6 Jul 2011 23:40:51 +0800 Subject: [PATCH 403/439] platform-drivers-x86: intel_mid_thermal: fix memory leak The memory for td_info which is allocated in initialize_sensor() should be properly kfreed in mid_thermal_probe() error patch and mid_thermal_remove(). Signed-off-by: Axel Lin Signed-off-by: Matthew Garrett --- drivers/platform/x86/intel_mid_thermal.c | 26 +++++++++++++++++------- 1 file changed, 19 insertions(+), 7 deletions(-) diff --git a/drivers/platform/x86/intel_mid_thermal.c b/drivers/platform/x86/intel_mid_thermal.c index 3a578323122b..ccd7b1f83519 100644 --- a/drivers/platform/x86/intel_mid_thermal.c +++ b/drivers/platform/x86/intel_mid_thermal.c @@ -493,20 +493,30 @@ static int mid_thermal_probe(struct platform_device *pdev) /* Register each sensor with the generic thermal framework*/ for (i = 0; i < MSIC_THERMAL_SENSORS; i++) { + struct thermal_device_info *td_info = initialize_sensor(i); + + if (!td_info) { + ret = -ENOMEM; + goto err; + } pinfo->tzd[i] = thermal_zone_device_register(name[i], - 0, initialize_sensor(i), &tzd_ops, 0, 0, 0, 0); - if (IS_ERR(pinfo->tzd[i])) - goto reg_fail; + 0, td_info, &tzd_ops, 0, 0, 0, 0); + if (IS_ERR(pinfo->tzd[i])) { + kfree(td_info); + ret = PTR_ERR(pinfo->tzd[i]); + goto err; + } } pinfo->pdev = pdev; platform_set_drvdata(pdev, pinfo); return 0; -reg_fail: - ret = PTR_ERR(pinfo->tzd[i]); - while (--i >= 0) +err: + while (--i >= 0) { + kfree(pinfo->tzd[i]->devdata); thermal_zone_device_unregister(pinfo->tzd[i]); + } configure_adc(0); kfree(pinfo); return ret; @@ -524,8 +534,10 @@ static int mid_thermal_remove(struct platform_device *pdev) int i; struct platform_info *pinfo = platform_get_drvdata(pdev); - for (i = 0; i < MSIC_THERMAL_SENSORS; i++) + for (i = 0; i < MSIC_THERMAL_SENSORS; i++) { + kfree(pinfo->tzd[i]->devdata); thermal_zone_device_unregister(pinfo->tzd[i]); + } kfree(pinfo); platform_set_drvdata(pdev, NULL); From c8bb2ebd6267d6c1288201344d7e7ee987d74467 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Thu, 7 Jul 2011 10:05:49 +0800 Subject: [PATCH 404/439] platform-drivers-x86: intel_menlow: add missing return AE_OK for intel_menlow_register_sensor() Otherwise, the error path will always be executed. Signed-off-by: Axel Lin Signed-off-by: Matthew Garrett --- drivers/platform/x86/intel_menlow.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/platform/x86/intel_menlow.c b/drivers/platform/x86/intel_menlow.c index 809adea4965f..abddc83e9fd7 100644 --- a/drivers/platform/x86/intel_menlow.c +++ b/drivers/platform/x86/intel_menlow.c @@ -477,6 +477,8 @@ static acpi_status intel_menlow_register_sensor(acpi_handle handle, u32 lvl, return AE_ERROR; } + return AE_OK; + aux1_not_found: if (status == AE_NOT_FOUND) return AE_OK; From b73210aec44ddf73ca856cbca2d25daeccf47025 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Thu, 7 Jul 2011 10:21:15 +0800 Subject: [PATCH 405/439] platform-drivers-x86: intel_rar_register: convert to DEFINE_PCI_DEVICE_TABLE And also remove unused variable 'my_id_table'. Signed-off-by: Axel Lin Signed-off-by: Matthew Garrett --- drivers/platform/x86/intel_rar_register.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/platform/x86/intel_rar_register.c b/drivers/platform/x86/intel_rar_register.c index bde47e9080cd..c8a6aed45277 100644 --- a/drivers/platform/x86/intel_rar_register.c +++ b/drivers/platform/x86/intel_rar_register.c @@ -637,15 +637,13 @@ end_function: return error; } -const struct pci_device_id rar_pci_id_tbl[] = { +static DEFINE_PCI_DEVICE_TABLE(rar_pci_id_tbl) = { { PCI_VDEVICE(INTEL, 0x4110) }, { 0 } }; MODULE_DEVICE_TABLE(pci, rar_pci_id_tbl); -const struct pci_device_id *my_id_table = rar_pci_id_tbl; - /* field for registering driver to PCI device */ static struct pci_driver rar_pci_driver = { .name = "rar_register_driver", From daa7769641019e9a0da180170cc1cd1b8ebd1641 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Thu, 7 Jul 2011 10:22:46 +0800 Subject: [PATCH 406/439] platform-drivers-x86: intel_scu_ipc: convert to DEFINE_PCI_DEVICE_TABLE Signed-off-by: Axel Lin Signed-off-by: Matthew Garrett --- drivers/platform/x86/intel_scu_ipc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/platform/x86/intel_scu_ipc.c b/drivers/platform/x86/intel_scu_ipc.c index 940accbe28d3..c86665369a22 100644 --- a/drivers/platform/x86/intel_scu_ipc.c +++ b/drivers/platform/x86/intel_scu_ipc.c @@ -725,7 +725,7 @@ static void ipc_remove(struct pci_dev *pdev) intel_scu_devices_destroy(); } -static const struct pci_device_id pci_ids[] = { +static DEFINE_PCI_DEVICE_TABLE(pci_ids) = { {PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x080e)}, {PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x082a)}, { 0,} From 78542e18335e35136c2dcb6c904f3f554241ae97 Mon Sep 17 00:00:00 2001 From: Frederick van der Wyck Date: Thu, 7 Jul 2011 22:05:25 +0100 Subject: [PATCH 407/439] Platform: Samsung Q10 backlight driver This adds backlight control on the Samsung Q10 laptop, which does not support the SABI interface. Also tested successfully on the Dell Latitude X200. Signed-off-by: Frederick van der Wyck Signed-off-by: Matthew Garrett --- drivers/platform/x86/Kconfig | 8 ++ drivers/platform/x86/Makefile | 1 + drivers/platform/x86/samsung-q10.c | 196 +++++++++++++++++++++++++++++ 3 files changed, 205 insertions(+) create mode 100644 drivers/platform/x86/samsung-q10.c diff --git a/drivers/platform/x86/Kconfig b/drivers/platform/x86/Kconfig index 45e0191c35dd..1e88d4785321 100644 --- a/drivers/platform/x86/Kconfig +++ b/drivers/platform/x86/Kconfig @@ -769,4 +769,12 @@ config INTEL_OAKTRAIL enable/disable the Camera, WiFi, BT etc. devices. If in doubt, say Y here; it will only load on supported platforms. +config SAMSUNG_Q10 + tristate "Samsung Q10 Extras" + depends on SERIO_I8042 + select BACKLIGHT_CLASS_DEVICE + ---help--- + This driver provides support for backlight control on Samsung Q10 + and related laptops, including Dell Latitude X200. + endif # X86_PLATFORM_DEVICES diff --git a/drivers/platform/x86/Makefile b/drivers/platform/x86/Makefile index afc1f832aa67..293a320d9faa 100644 --- a/drivers/platform/x86/Makefile +++ b/drivers/platform/x86/Makefile @@ -44,3 +44,4 @@ obj-$(CONFIG_SAMSUNG_LAPTOP) += samsung-laptop.o obj-$(CONFIG_MXM_WMI) += mxm-wmi.o obj-$(CONFIG_INTEL_MID_POWER_BUTTON) += intel_mid_powerbtn.o obj-$(CONFIG_INTEL_OAKTRAIL) += intel_oaktrail.o +obj-$(CONFIG_SAMSUNG_Q10) += samsung-q10.o diff --git a/drivers/platform/x86/samsung-q10.c b/drivers/platform/x86/samsung-q10.c new file mode 100644 index 000000000000..54cf3d63fe8d --- /dev/null +++ b/drivers/platform/x86/samsung-q10.c @@ -0,0 +1,196 @@ +/* + * Driver for Samsung Q10 and related laptops: controls the backlight + * + * Copyright (c) 2011 Frederick van der Wyck + * + * 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 +#include +#include +#include +#include +#include +#include + +#define SAMSUNGQ10_BL_MAX_INTENSITY 255 +#define SAMSUNGQ10_BL_DEFAULT_INTENSITY 185 + +#define SAMSUNGQ10_BL_8042_CMD 0xbe +#define SAMSUNGQ10_BL_8042_DATA { 0x89, 0x91 } + +static int samsungq10_bl_brightness; + +static bool force; +module_param(force, bool, 0); +MODULE_PARM_DESC(force, + "Disable the DMI check and force the driver to be loaded"); + +static int samsungq10_bl_set_intensity(struct backlight_device *bd) +{ + + int brightness = bd->props.brightness; + unsigned char c[3] = SAMSUNGQ10_BL_8042_DATA; + + c[2] = (unsigned char)brightness; + i8042_lock_chip(); + i8042_command(c, (0x30 << 8) | SAMSUNGQ10_BL_8042_CMD); + i8042_unlock_chip(); + samsungq10_bl_brightness = brightness; + + return 0; +} + +static int samsungq10_bl_get_intensity(struct backlight_device *bd) +{ + return samsungq10_bl_brightness; +} + +static const struct backlight_ops samsungq10_bl_ops = { + .get_brightness = samsungq10_bl_get_intensity, + .update_status = samsungq10_bl_set_intensity, +}; + +#ifdef CONFIG_PM_SLEEP +static int samsungq10_suspend(struct device *dev) +{ + return 0; +} + +static int samsungq10_resume(struct device *dev) +{ + + struct backlight_device *bd = dev_get_drvdata(dev); + + samsungq10_bl_set_intensity(bd); + return 0; +} +#else +#define samsungq10_suspend NULL +#define samsungq10_resume NULL +#endif + +static SIMPLE_DEV_PM_OPS(samsungq10_pm_ops, + samsungq10_suspend, samsungq10_resume); + +static int __devinit samsungq10_probe(struct platform_device *pdev) +{ + + struct backlight_properties props; + struct backlight_device *bd; + + memset(&props, 0, sizeof(struct backlight_properties)); + props.type = BACKLIGHT_PLATFORM; + props.max_brightness = SAMSUNGQ10_BL_MAX_INTENSITY; + bd = backlight_device_register("samsung", &pdev->dev, NULL, + &samsungq10_bl_ops, &props); + if (IS_ERR(bd)) + return PTR_ERR(bd); + + platform_set_drvdata(pdev, bd); + + bd->props.brightness = SAMSUNGQ10_BL_DEFAULT_INTENSITY; + samsungq10_bl_set_intensity(bd); + + return 0; +} + +static int __devexit samsungq10_remove(struct platform_device *pdev) +{ + + struct backlight_device *bd = platform_get_drvdata(pdev); + + bd->props.brightness = SAMSUNGQ10_BL_DEFAULT_INTENSITY; + samsungq10_bl_set_intensity(bd); + + backlight_device_unregister(bd); + + return 0; +} + +static struct platform_driver samsungq10_driver = { + .driver = { + .name = KBUILD_MODNAME, + .owner = THIS_MODULE, + .pm = &samsungq10_pm_ops, + }, + .probe = samsungq10_probe, + .remove = __devexit_p(samsungq10_remove), +}; + +static struct platform_device *samsungq10_device; + +static int __init dmi_check_callback(const struct dmi_system_id *id) +{ + printk(KERN_INFO KBUILD_MODNAME ": found model '%s'\n", id->ident); + return 0; +} + +static struct dmi_system_id __initdata samsungq10_dmi_table[] = { + { + .ident = "Samsung Q10", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Samsung"), + DMI_MATCH(DMI_PRODUCT_NAME, "SQ10"), + }, + .callback = dmi_check_callback, + }, + { + .ident = "Samsung Q20", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "SAMSUNG Electronics"), + DMI_MATCH(DMI_PRODUCT_NAME, "SENS Q20"), + }, + .callback = dmi_check_callback, + }, + { + .ident = "Samsung Q25", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "SAMSUNG Electronics"), + DMI_MATCH(DMI_PRODUCT_NAME, "NQ25"), + }, + .callback = dmi_check_callback, + }, + { + .ident = "Dell Latitude X200", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Computer Corporation"), + DMI_MATCH(DMI_PRODUCT_NAME, "X200"), + }, + .callback = dmi_check_callback, + }, + { }, +}; +MODULE_DEVICE_TABLE(dmi, samsungq10_dmi_table); + +static int __init samsungq10_init(void) +{ + if (!force && !dmi_check_system(samsungq10_dmi_table)) + return -ENODEV; + + samsungq10_device = platform_create_bundle(&samsungq10_driver, + samsungq10_probe, + NULL, 0, NULL, 0); + + if (IS_ERR(samsungq10_device)) + return PTR_ERR(samsungq10_device); + + return 0; +} + +static void __exit samsungq10_exit(void) +{ + platform_device_unregister(samsungq10_device); + platform_driver_unregister(&samsungq10_driver); +} + +module_init(samsungq10_init); +module_exit(samsungq10_exit); + +MODULE_AUTHOR("Frederick van der Wyck "); +MODULE_DESCRIPTION("Samsung Q10 Driver"); +MODULE_LICENSE("GPL"); From 046f828b410837d304137cec9e4e0d72e50fc454 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Mon, 18 Jul 2011 16:08:21 +0800 Subject: [PATCH 408/439] platform-drivers-x86: samsung-q10: make dmi_check_callback return 1 We only care about if there is any successful match from the dmi table or no match at all, we can make dmi_check_system return immediately if we have a successful match instead of iterate thorough the whole table. Signed-off-by: Axel Lin Signed-off-by: Matthew Garrett --- drivers/platform/x86/samsung-q10.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/platform/x86/samsung-q10.c b/drivers/platform/x86/samsung-q10.c index 54cf3d63fe8d..1e54ae74274c 100644 --- a/drivers/platform/x86/samsung-q10.c +++ b/drivers/platform/x86/samsung-q10.c @@ -127,7 +127,7 @@ static struct platform_device *samsungq10_device; static int __init dmi_check_callback(const struct dmi_system_id *id) { printk(KERN_INFO KBUILD_MODNAME ": found model '%s'\n", id->ident); - return 0; + return 1; } static struct dmi_system_id __initdata samsungq10_dmi_table[] = { From f1566f0dc07ec9b5409b348070f5a700032d7881 Mon Sep 17 00:00:00 2001 From: Seth Forshee Date: Mon, 1 Aug 2011 15:46:10 -0500 Subject: [PATCH 409/439] dell-wmi: Add keys for Dell XPS L502X All of these keys are being reported on the keyboard controller but are also generating WMI events. Add them to the legacy keymap to silence the noise. BugLink: http://bugs.launchpad.net/bugs/815914 Signed-off-by: Seth Forshee Signed-off-by: Matthew Garrett --- drivers/platform/x86/dell-wmi.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/platform/x86/dell-wmi.c b/drivers/platform/x86/dell-wmi.c index ce790827e199..fa9a2171cc13 100644 --- a/drivers/platform/x86/dell-wmi.c +++ b/drivers/platform/x86/dell-wmi.c @@ -54,6 +54,8 @@ MODULE_ALIAS("wmi:"DELL_EVENT_GUID); */ static const struct key_entry dell_wmi_legacy_keymap[] __initconst = { + { KE_IGNORE, 0x003a, { KEY_CAPSLOCK } }, + { KE_KEY, 0xe045, { KEY_PROG1 } }, { KE_KEY, 0xe009, { KEY_EJECTCD } }, @@ -85,6 +87,11 @@ static const struct key_entry dell_wmi_legacy_keymap[] __initconst = { { KE_IGNORE, 0xe013, { KEY_RESERVED } }, { KE_IGNORE, 0xe020, { KEY_MUTE } }, + + /* Shortcut and audio panel keys */ + { KE_IGNORE, 0xe025, { KEY_RESERVED } }, + { KE_IGNORE, 0xe026, { KEY_RESERVED } }, + { KE_IGNORE, 0xe02e, { KEY_VOLUMEDOWN } }, { KE_IGNORE, 0xe030, { KEY_VOLUMEUP } }, { KE_IGNORE, 0xe033, { KEY_KBDILLUMUP } }, @@ -92,6 +99,9 @@ static const struct key_entry dell_wmi_legacy_keymap[] __initconst = { { KE_IGNORE, 0xe03a, { KEY_CAPSLOCK } }, { KE_IGNORE, 0xe045, { KEY_NUMLOCK } }, { KE_IGNORE, 0xe046, { KEY_SCROLLLOCK } }, + { KE_IGNORE, 0xe0f7, { KEY_MUTE } }, + { KE_IGNORE, 0xe0f8, { KEY_VOLUMEDOWN } }, + { KE_IGNORE, 0xe0f9, { KEY_VOLUMEUP } }, { KE_END, 0 } }; From 78a7539b881eb557494a7c810625c0307b27296c Mon Sep 17 00:00:00 2001 From: Thomas Courbon Date: Wed, 20 Jul 2011 22:57:44 +0200 Subject: [PATCH 410/439] Platform: fix samsung-laptop DMI identification for N150/N210/220/N230 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Some samsung latop of the N150/N2{10,20,30} serie are badly detected by the samsung-laptop platform driver, see bug # 36082. It appears that N230 identifies itself as N150/N210/N220/N230 whereas the other identify themselves as N150/N210/220. This patch attemtp fix #36082 allowing correct identification for all the said netbook model. Reported-by: Daniel Eklöf Signed-off-by: Thomas Courbon Signed-off-by: Matthew Garrett --- drivers/platform/x86/samsung-laptop.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/platform/x86/samsung-laptop.c b/drivers/platform/x86/samsung-laptop.c index e12c88e92c4e..359163011044 100644 --- a/drivers/platform/x86/samsung-laptop.c +++ b/drivers/platform/x86/samsung-laptop.c @@ -610,6 +610,16 @@ static struct dmi_system_id __initdata samsung_dmi_table[] = { }, .callback = dmi_check_cb, }, + { + .ident = "N150/N210/N220", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, + "SAMSUNG ELECTRONICS CO., LTD."), + DMI_MATCH(DMI_PRODUCT_NAME, "N150/N210/N220"), + DMI_MATCH(DMI_BOARD_NAME, "N150/N210/N220"), + }, + .callback = dmi_check_cb, + }, { .ident = "N150/N210/N220/N230", .matches = { From 70fda70a724c737b0df4195763f9124d181fe64d Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Fri, 22 Jul 2011 09:21:36 -0700 Subject: [PATCH 411/439] x86 driver: fix typo in TDP override enabling When enabling turbo, we need to set both the TDC and TDP bits. IIRC only the TDC one actually matters, but fix it up anyway since the current code is confusing. Signed-off-by: Jesse Barnes Signed-off-by: Matthew Garrett --- drivers/platform/x86/intel_ips.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/platform/x86/intel_ips.c b/drivers/platform/x86/intel_ips.c index 5ffe7c398148..809a3ae943c6 100644 --- a/drivers/platform/x86/intel_ips.c +++ b/drivers/platform/x86/intel_ips.c @@ -403,7 +403,7 @@ static void ips_cpu_raise(struct ips_driver *ips) thm_writew(THM_MPCPC, (new_tdp_limit * 10) / 8); - turbo_override |= TURBO_TDC_OVR_EN | TURBO_TDC_OVR_EN; + turbo_override |= TURBO_TDC_OVR_EN | TURBO_TDP_OVR_EN; wrmsrl(TURBO_POWER_CURRENT_LIMIT, turbo_override); turbo_override &= ~TURBO_TDP_MASK; @@ -438,7 +438,7 @@ static void ips_cpu_lower(struct ips_driver *ips) thm_writew(THM_MPCPC, (new_limit * 10) / 8); - turbo_override |= TURBO_TDC_OVR_EN | TURBO_TDC_OVR_EN; + turbo_override |= TURBO_TDC_OVR_EN | TURBO_TDP_OVR_EN; wrmsrl(TURBO_POWER_CURRENT_LIMIT, turbo_override); turbo_override &= ~TURBO_TDP_MASK; From 7451a55af08d40fd6b28802f4e26a13264114430 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Wed, 27 Jul 2011 15:27:34 +0800 Subject: [PATCH 412/439] platform-drivers-x86: ideapad-laptop: add missing ideapad_input_exit in ideapad_acpi_add error path In the case of ideapad_backlight_init() failure, we need to free the resources allocated by ideapad_input_init(). Aslo drop __devexit annotation for ideapad_input_exit() because we also call it in ideapad_acpi_add() error path. Signed-off-by: Axel Lin Signed-off-by: Matthew Garrett --- drivers/platform/x86/ideapad-laptop.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/platform/x86/ideapad-laptop.c b/drivers/platform/x86/ideapad-laptop.c index 8811c68f809d..0c595410e788 100644 --- a/drivers/platform/x86/ideapad-laptop.c +++ b/drivers/platform/x86/ideapad-laptop.c @@ -406,7 +406,7 @@ err_free_dev: return error; } -static void __devexit ideapad_input_exit(struct ideapad_private *priv) +static void ideapad_input_exit(struct ideapad_private *priv) { sparse_keymap_free(priv->inputdev); input_unregister_device(priv->inputdev); @@ -563,6 +563,7 @@ static int __devinit ideapad_acpi_add(struct acpi_device *adevice) backlight_failed: for (i = 0; i < IDEAPAD_RFKILL_DEV_NUM; i++) ideapad_unregister_rfkill(adevice, i); + ideapad_input_exit(priv); input_failed: ideapad_platform_exit(priv); platform_failed: From 2c3422d9c475c62adae82743d8507651342925ae Mon Sep 17 00:00:00 2001 From: "Anton V. Boyarshinov" Date: Thu, 28 Jul 2011 18:05:35 +0400 Subject: [PATCH 413/439] acerhdf.c: spaces in aliased changed to * It seems that aliases shouldn't contain spaces, as module-init-tools uses them as delimeters in module.alias file Signed-off-by: Anton V. Boyarshinov Signed-off-by: Matthew Garrett --- drivers/platform/x86/acerhdf.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/platform/x86/acerhdf.c b/drivers/platform/x86/acerhdf.c index d339756608af..760c6d7624fe 100644 --- a/drivers/platform/x86/acerhdf.c +++ b/drivers/platform/x86/acerhdf.c @@ -704,15 +704,15 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR("Peter Feuerer"); MODULE_DESCRIPTION("Aspire One temperature and fan driver"); MODULE_ALIAS("dmi:*:*Acer*:pnAOA*:"); -MODULE_ALIAS("dmi:*:*Acer*:pnAspire 1410*:"); -MODULE_ALIAS("dmi:*:*Acer*:pnAspire 1810*:"); +MODULE_ALIAS("dmi:*:*Acer*:pnAspire*1410*:"); +MODULE_ALIAS("dmi:*:*Acer*:pnAspire*1810*:"); MODULE_ALIAS("dmi:*:*Acer*:pnAO531*:"); MODULE_ALIAS("dmi:*:*Gateway*:pnAOA*:"); MODULE_ALIAS("dmi:*:*Gateway*:pnLT31*:"); -MODULE_ALIAS("dmi:*:*Packard Bell*:pnAOA*:"); -MODULE_ALIAS("dmi:*:*Packard Bell*:pnDOA*:"); -MODULE_ALIAS("dmi:*:*Packard Bell*:pnDOTMU*:"); -MODULE_ALIAS("dmi:*:*Packard Bell*:pnDOTMA*:"); +MODULE_ALIAS("dmi:*:*Packard*Bell*:pnAOA*:"); +MODULE_ALIAS("dmi:*:*Packard*Bell*:pnDOA*:"); +MODULE_ALIAS("dmi:*:*Packard*Bell*:pnDOTMU*:"); +MODULE_ALIAS("dmi:*:*Packard*Bell*:pnDOTMA*:"); module_init(acerhdf_init); module_exit(acerhdf_exit); From 15b956a0b5651bbb1217ec374fdd67291dabb2af Mon Sep 17 00:00:00 2001 From: "Lee, Chun-Yi" Date: Sat, 30 Jul 2011 17:00:45 +0800 Subject: [PATCH 414/439] acer-wmi: support Lenovo ideapad S205 wifi switch The AMW0 function in acer-wmi works on Lenovo ideapad S205 for control the wifi hardware state. We also found there have a 0x78 EC register exposes the state of wifi hardware switch on the machine. So, add this patch to support Lenovo ideapad S205 wifi hardware switch in acer-wmi driver. Reference: bko#37892 https://bugzilla.kernel.org/show_bug.cgi?id=37892 Cc: Carlos Corbacho Cc: Matthew Garrett Cc: Dmitry Torokhov Cc: Corentin Chary Cc: Thomas Renninger Tested-by: Florian Heyer Signed-off-by: Lee, Chun-Yi Signed-off-by: Matthew Garrett --- drivers/platform/x86/acer-wmi.c | 28 ++++++++++++++++++++++++++-- 1 file changed, 26 insertions(+), 2 deletions(-) diff --git a/drivers/platform/x86/acer-wmi.c b/drivers/platform/x86/acer-wmi.c index fa7153020ca1..af2bb20cb2fb 100644 --- a/drivers/platform/x86/acer-wmi.c +++ b/drivers/platform/x86/acer-wmi.c @@ -305,6 +305,10 @@ static struct quirk_entry quirk_fujitsu_amilo_li_1718 = { .wireless = 2, }; +static struct quirk_entry quirk_lenovo_ideapad_s205 = { + .wireless = 3, +}; + /* The Aspire One has a dummy ACPI-WMI interface - disable it */ static struct dmi_system_id __devinitdata acer_blacklist[] = { { @@ -451,6 +455,15 @@ static struct dmi_system_id acer_quirks[] = { }, .driver_data = &quirk_medion_md_98300, }, + { + .callback = dmi_matched, + .ident = "Lenovo Ideapad S205", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), + DMI_MATCH(DMI_PRODUCT_NAME, "10382LG"), + }, + .driver_data = &quirk_lenovo_ideapad_s205, + }, {} }; @@ -543,6 +556,12 @@ struct wmi_interface *iface) return AE_ERROR; *value = result & 0x1; return AE_OK; + case 3: + err = ec_read(0x78, &result); + if (err) + return AE_ERROR; + *value = result & 0x1; + return AE_OK; default: err = ec_read(0xA, &result); if (err) @@ -1267,8 +1286,13 @@ static void acer_rfkill_update(struct work_struct *ignored) acpi_status status; status = get_u32(&state, ACER_CAP_WIRELESS); - if (ACPI_SUCCESS(status)) - rfkill_set_sw_state(wireless_rfkill, !state); + if (ACPI_SUCCESS(status)) { + if (quirks->wireless == 3) { + rfkill_set_hw_state(wireless_rfkill, !state); + } else { + rfkill_set_sw_state(wireless_rfkill, !state); + } + } if (has_cap(ACER_CAP_BLUETOOTH)) { status = get_u32(&state, ACER_CAP_BLUETOOTH); From fe202fde50a986a8510c62a76dc8733c1a8fac86 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Fri, 5 Aug 2011 21:45:11 +0200 Subject: [PATCH 415/439] PM / Domains: Fix pm_genpd_poweron() The local variable ret is defined twice in pm_genpd_poweron(), which causes this function to always return 0, even if the PM domain's .power_on() callback fails, in which case an error code should be returned. Remove the wrong second definition of ret and additionally remove an unnecessary definition of wait from pm_genpd_poweron(). Signed-off-by: Rafael J. Wysocki --- drivers/base/power/domain.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/base/power/domain.c b/drivers/base/power/domain.c index be8714aa9dd6..e18566a0fedd 100644 --- a/drivers/base/power/domain.c +++ b/drivers/base/power/domain.c @@ -80,7 +80,6 @@ static void genpd_set_active(struct generic_pm_domain *genpd) int pm_genpd_poweron(struct generic_pm_domain *genpd) { struct generic_pm_domain *parent = genpd->parent; - DEFINE_WAIT(wait); int ret = 0; start: @@ -112,7 +111,7 @@ int pm_genpd_poweron(struct generic_pm_domain *genpd) } if (genpd->power_on) { - int ret = genpd->power_on(genpd); + ret = genpd->power_on(genpd); if (ret) goto out; } From 02b26774afebb2d62695ba3230319d70d8c6cc2d Mon Sep 17 00:00:00 2001 From: Kevin Hilman Date: Fri, 5 Aug 2011 21:45:20 +0200 Subject: [PATCH 416/439] PM / Runtime: Allow _put_sync() from interrupts-disabled context Currently the use of pm_runtime_put_sync() is not safe from interrupts-disabled context because rpm_idle() will release the spinlock and enable interrupts for the idle callbacks. This enables interrupts during a time where interrupts were expected to be disabled, and can have strange side effects on drivers that expected interrupts to be disabled. This is not a bug since the documentation clearly states that only _put_sync_suspend() is safe in IRQ-safe mode. However, pm_runtime_put_sync() could be made safe when in IRQ-safe mode by releasing the spinlock but not re-enabling interrupts, which is what this patch aims to do. Problem was found when using some buggy drivers that set pm_runtime_irq_safe() and used _put_sync() in interrupts-disabled context. Reported-by: Colin Cross Tested-by: Nishanth Menon Signed-off-by: Kevin Hilman Signed-off-by: Rafael J. Wysocki --- Documentation/power/runtime_pm.txt | 10 +++++----- drivers/base/power/runtime.c | 10 ++++++++-- 2 files changed, 13 insertions(+), 7 deletions(-) diff --git a/Documentation/power/runtime_pm.txt b/Documentation/power/runtime_pm.txt index 14dd3c6ad97e..4ce5450ab6e8 100644 --- a/Documentation/power/runtime_pm.txt +++ b/Documentation/power/runtime_pm.txt @@ -54,11 +54,10 @@ referred to as subsystem-level callbacks in what follows. By default, the callbacks are always invoked in process context with interrupts enabled. However, subsystems can use the pm_runtime_irq_safe() helper function to tell the PM core that a device's ->runtime_suspend() and ->runtime_resume() -callbacks should be invoked in atomic context with interrupts disabled -(->runtime_idle() is still invoked the default way). This implies that these -callback routines must not block or sleep, but it also means that the -synchronous helper functions listed at the end of Section 4 can be used within -an interrupt handler or in an atomic context. +callbacks should be invoked in atomic context with interrupts disabled. +This implies that these callback routines must not block or sleep, but it also +means that the synchronous helper functions listed at the end of Section 4 can +be used within an interrupt handler or in an atomic context. The subsystem-level suspend callback is _entirely_ _responsible_ for handling the suspend of the device as appropriate, which may, but need not include @@ -483,6 +482,7 @@ pm_runtime_suspend() pm_runtime_autosuspend() pm_runtime_resume() pm_runtime_get_sync() +pm_runtime_put_sync() pm_runtime_put_sync_suspend() 5. Runtime PM Initialization, Device Probing and Removal diff --git a/drivers/base/power/runtime.c b/drivers/base/power/runtime.c index 8dc247c974af..acb3f83b8079 100644 --- a/drivers/base/power/runtime.c +++ b/drivers/base/power/runtime.c @@ -226,11 +226,17 @@ static int rpm_idle(struct device *dev, int rpmflags) callback = NULL; if (callback) { - spin_unlock_irq(&dev->power.lock); + if (dev->power.irq_safe) + spin_unlock(&dev->power.lock); + else + spin_unlock_irq(&dev->power.lock); callback(dev); - spin_lock_irq(&dev->power.lock); + if (dev->power.irq_safe) + spin_lock(&dev->power.lock); + else + spin_lock_irq(&dev->power.lock); } dev->power.idle_notification = false; From 69d94ec6d83d84044252d9ba03f6a8970816e350 Mon Sep 17 00:00:00 2001 From: Sergey Senozhatsky Date: Sat, 6 Aug 2011 01:34:08 +0300 Subject: [PATCH 417/439] Battery: sysfs_remove_battery(): possible circular locking Commit 9c921c22a7f33397a6774d7fa076db9b6a0fd669 Author: Lan Tianyu ACPI / Battery: Resolve the race condition in the sysfs_remove_battery() fixed BUG https://bugzilla.kernel.org/show_bug.cgi?id=35642 , but as a side effect made lockdep unhappy with sysfs_remove_battery(): [14818.477168] [14818.477170] ======================================================= [14818.477200] [ INFO: possible circular locking dependency detected ] [14818.477221] 3.1.0-dbg-07865-g1280ea8-dirty #668 [14818.477236] ------------------------------------------------------- [14818.477257] s2ram/1599 is trying to acquire lock: [14818.477276] (s_active#8){++++.+}, at: [] sysfs_addrm_finish+0x31/0x5a [14818.477323] [14818.477325] but task is already holding lock: [14818.477350] (&battery->lock){+.+.+.}, at: [] sysfs_remove_battery+0x10/0x4b [battery] [14818.477395] [14818.477397] which lock already depends on the new lock. [14818.477399] [..] [14818.479121] stack backtrace: [14818.479148] Pid: 1599, comm: s2ram Not tainted 3.1.0-dbg-07865-g1280ea8-dirty #668 [14818.479175] Call Trace: [14818.479198] [] print_circular_bug+0x293/0x2a4 [14818.479228] [] __lock_acquire+0xfe4/0x164b [14818.479260] [] ? sysfs_addrm_finish+0x31/0x5a [14818.479288] [] lock_acquire+0x138/0x1ac [14818.479316] [] ? sysfs_addrm_finish+0x31/0x5a [14818.479345] [] sysfs_deactivate+0x9b/0xec [14818.479373] [] ? sysfs_addrm_finish+0x31/0x5a [14818.479405] [] sysfs_addrm_finish+0x31/0x5a [14818.479433] [] sysfs_hash_and_remove+0x54/0x77 [14818.479461] [] sysfs_remove_file+0x12/0x14 [14818.479488] [] device_remove_file+0x12/0x14 [14818.479516] [] device_del+0x119/0x17c [14818.479542] [] device_unregister+0xe/0x1a [14818.479570] [] power_supply_unregister+0x23/0x27 [14818.479601] [] sysfs_remove_battery+0x34/0x4b [battery] [14818.479632] [] battery_notify+0x2c/0x3a [battery] [14818.479662] [] notifier_call_chain+0x74/0xa1 [14818.479692] [] __blocking_notifier_call_chain+0x6c/0x89 [14818.479722] [] blocking_notifier_call_chain+0xf/0x11 [14818.479751] [] pm_notifier_call_chain+0x15/0x27 [14818.479770] [] enter_state+0xa7/0xd5 [14818.479782] [] state_store+0xaa/0xc0 [14818.479795] [] ? pm_async_store+0x45/0x45 [14818.479807] [] kobj_attr_store+0x17/0x19 [14818.479820] [] sysfs_write_file+0x103/0x13f [14818.479834] [] vfs_write+0xad/0x13d [14818.479847] [] sys_write+0x45/0x6c [14818.479860] [] system_call_fastpath+0x16/0x1b This patch introduces separate lock to struct acpi_battery to grab in sysfs_remove_battery() instead of battery->lock. So fix by Lan Tianyu is still there, we just grab independent lock. Signed-off-by: Sergey Senozhatsky Tested-by: Lan Tianyu Signed-off-by: Len Brown --- drivers/acpi/battery.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c index ffce2f06dd8d..8953b7087091 100644 --- a/drivers/acpi/battery.c +++ b/drivers/acpi/battery.c @@ -99,6 +99,7 @@ enum { struct acpi_battery { struct mutex lock; + struct mutex sysfs_lock; struct power_supply bat; struct acpi_device *device; struct notifier_block pm_nb; @@ -573,16 +574,16 @@ static int sysfs_add_battery(struct acpi_battery *battery) static void sysfs_remove_battery(struct acpi_battery *battery) { - mutex_lock(&battery->lock); + mutex_lock(&battery->sysfs_lock); if (!battery->bat.dev) { - mutex_unlock(&battery->lock); + mutex_unlock(&battery->sysfs_lock); return; } device_remove_file(battery->bat.dev, &alarm_attr); power_supply_unregister(&battery->bat); battery->bat.dev = NULL; - mutex_unlock(&battery->lock); + mutex_unlock(&battery->sysfs_lock); } /* @@ -982,6 +983,7 @@ static int acpi_battery_add(struct acpi_device *device) strcpy(acpi_device_class(device), ACPI_BATTERY_CLASS); device->driver_data = battery; mutex_init(&battery->lock); + mutex_init(&battery->sysfs_lock); if (ACPI_SUCCESS(acpi_get_handle(battery->device->handle, "_BIX", &handle))) set_bit(ACPI_BATTERY_XINFO_PRESENT, &battery->flags); @@ -1010,6 +1012,7 @@ static int acpi_battery_add(struct acpi_device *device) fail: sysfs_remove_battery(battery); mutex_destroy(&battery->lock); + mutex_destroy(&battery->sysfs_lock); kfree(battery); return result; } @@ -1027,6 +1030,7 @@ static int acpi_battery_remove(struct acpi_device *device, int type) #endif sysfs_remove_battery(battery); mutex_destroy(&battery->lock); + mutex_destroy(&battery->sysfs_lock); kfree(battery); return 0; } From 0785a8e87be0202744d8681363aecbd4ffbb5f5a Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sat, 6 Aug 2011 05:26:35 -0700 Subject: [PATCH 418/439] sparc: Fix build with DEBUG_PAGEALLOC enabled. arch/sparc/mm/init_64.c:1622:22: error: unused variable '__swapper_4m_tsb_phys_patch_end' [-Werror=unused-variable] arch/sparc/mm/init_64.c:1621:22: error: unused variable '__swapper_4m_tsb_phys_patch' [-Werror=unused-variable] Signed-off-by: David S. Miller --- arch/sparc/mm/init_64.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c index adfac23d976a..581531dbc8b5 100644 --- a/arch/sparc/mm/init_64.c +++ b/arch/sparc/mm/init_64.c @@ -1618,18 +1618,20 @@ static void ktsb_phys_patch(void) { extern unsigned int __swapper_tsb_phys_patch; extern unsigned int __swapper_tsb_phys_patch_end; - extern unsigned int __swapper_4m_tsb_phys_patch; - extern unsigned int __swapper_4m_tsb_phys_patch_end; unsigned long ktsb_pa; ktsb_pa = kern_base + ((unsigned long)&swapper_tsb[0] - KERNBASE); patch_one_ktsb_phys(&__swapper_tsb_phys_patch, &__swapper_tsb_phys_patch_end, ktsb_pa); #ifndef CONFIG_DEBUG_PAGEALLOC + { + extern unsigned int __swapper_4m_tsb_phys_patch; + extern unsigned int __swapper_4m_tsb_phys_patch_end; ktsb_pa = (kern_base + ((unsigned long)&swapper_4m_tsb[0] - KERNBASE)); patch_one_ktsb_phys(&__swapper_4m_tsb_phys_patch, &__swapper_4m_tsb_phys_patch_end, ktsb_pa); + } #endif } From 1eb19a12bd2214cdcad5273d472b062a4ba97fa1 Mon Sep 17 00:00:00 2001 From: Mandeep Singh Baines Date: Fri, 5 Aug 2011 18:46:27 -0700 Subject: [PATCH 419/439] lib/sha1: use the git implementation of SHA-1 For ChromiumOS, we use SHA-1 to verify the integrity of the root filesystem. The speed of the kernel sha-1 implementation has a major impact on our boot performance. To improve boot performance, we investigated using the heavily optimized sha-1 implementation used in git. With the git sha-1 implementation, we see a 11.7% improvement in boot time. 10 reboots, remove slowest/fastest. Before: Mean: 6.58 seconds Stdev: 0.14 After (with git sha-1, this patch): Mean: 5.89 seconds Stdev: 0.07 The other cool thing about the git SHA-1 implementation is that it only needs 64 bytes of stack for the workspace while the original kernel implementation needed 320 bytes. Signed-off-by: Mandeep Singh Baines Cc: Ramsay Jones Cc: Nicolas Pitre Cc: Herbert Xu Cc: David S. Miller Cc: linux-crypto@vger.kernel.org Signed-off-by: Linus Torvalds --- include/linux/cryptohash.h | 2 +- lib/sha1.c | 198 ++++++++++++++++++++++++++++--------- 2 files changed, 152 insertions(+), 48 deletions(-) diff --git a/include/linux/cryptohash.h b/include/linux/cryptohash.h index ec78a4bbe1d5..f9452185b019 100644 --- a/include/linux/cryptohash.h +++ b/include/linux/cryptohash.h @@ -3,7 +3,7 @@ #define SHA_DIGEST_WORDS 5 #define SHA_MESSAGE_BYTES (512 /*bits*/ / 8) -#define SHA_WORKSPACE_WORDS 80 +#define SHA_WORKSPACE_WORDS 16 void sha_init(__u32 *buf); void sha_transform(__u32 *digest, const char *data, __u32 *W); diff --git a/lib/sha1.c b/lib/sha1.c index 4c45fd50e913..f33271dd00cb 100644 --- a/lib/sha1.c +++ b/lib/sha1.c @@ -1,31 +1,72 @@ /* - * SHA transform algorithm, originally taken from code written by - * Peter Gutmann, and placed in the public domain. + * SHA1 routine optimized to do word accesses rather than byte accesses, + * and to avoid unnecessary copies into the context array. + * + * This was based on the git SHA1 implementation. */ #include #include -#include +#include +#include -/* The SHA f()-functions. */ +/* + * If you have 32 registers or more, the compiler can (and should) + * try to change the array[] accesses into registers. However, on + * machines with less than ~25 registers, that won't really work, + * and at least gcc will make an unholy mess of it. + * + * So to avoid that mess which just slows things down, we force + * the stores to memory to actually happen (we might be better off + * with a 'W(t)=(val);asm("":"+m" (W(t))' there instead, as + * suggested by Artur Skawina - that will also make gcc unable to + * try to do the silly "optimize away loads" part because it won't + * see what the value will be). + * + * Ben Herrenschmidt reports that on PPC, the C version comes close + * to the optimized asm with this (ie on PPC you don't want that + * 'volatile', since there are lots of registers). + * + * On ARM we get the best code generation by forcing a full memory barrier + * between each SHA_ROUND, otherwise gcc happily get wild with spilling and + * the stack frame size simply explode and performance goes down the drain. + */ -#define f1(x,y,z) (z ^ (x & (y ^ z))) /* x ? y : z */ -#define f2(x,y,z) (x ^ y ^ z) /* XOR */ -#define f3(x,y,z) ((x & y) + (z & (x ^ y))) /* majority */ +#ifdef CONFIG_X86 + #define setW(x, val) (*(volatile __u32 *)&W(x) = (val)) +#elif defined(CONFIG_ARM) + #define setW(x, val) do { W(x) = (val); __asm__("":::"memory"); } while (0) +#else + #define setW(x, val) (W(x) = (val)) +#endif -/* The SHA Mysterious Constants */ +/* This "rolls" over the 512-bit array */ +#define W(x) (array[(x)&15]) -#define K1 0x5A827999L /* Rounds 0-19: sqrt(2) * 2^30 */ -#define K2 0x6ED9EBA1L /* Rounds 20-39: sqrt(3) * 2^30 */ -#define K3 0x8F1BBCDCL /* Rounds 40-59: sqrt(5) * 2^30 */ -#define K4 0xCA62C1D6L /* Rounds 60-79: sqrt(10) * 2^30 */ +/* + * Where do we get the source from? The first 16 iterations get it from + * the input data, the next mix it from the 512-bit array. + */ +#define SHA_SRC(t) get_unaligned_be32((__u32 *)data + t) +#define SHA_MIX(t) rol32(W(t+13) ^ W(t+8) ^ W(t+2) ^ W(t), 1) + +#define SHA_ROUND(t, input, fn, constant, A, B, C, D, E) do { \ + __u32 TEMP = input(t); setW(t, TEMP); \ + E += TEMP + rol32(A,5) + (fn) + (constant); \ + B = ror32(B, 2); } while (0) + +#define T_0_15(t, A, B, C, D, E) SHA_ROUND(t, SHA_SRC, (((C^D)&B)^D) , 0x5a827999, A, B, C, D, E ) +#define T_16_19(t, A, B, C, D, E) SHA_ROUND(t, SHA_MIX, (((C^D)&B)^D) , 0x5a827999, A, B, C, D, E ) +#define T_20_39(t, A, B, C, D, E) SHA_ROUND(t, SHA_MIX, (B^C^D) , 0x6ed9eba1, A, B, C, D, E ) +#define T_40_59(t, A, B, C, D, E) SHA_ROUND(t, SHA_MIX, ((B&C)+(D&(B^C))) , 0x8f1bbcdc, A, B, C, D, E ) +#define T_60_79(t, A, B, C, D, E) SHA_ROUND(t, SHA_MIX, (B^C^D) , 0xca62c1d6, A, B, C, D, E ) /** * sha_transform - single block SHA1 transform * * @digest: 160 bit digest to update * @data: 512 bits of data to hash - * @W: 80 words of workspace (see note) + * @array: 16 words of workspace (see note) * * This function generates a SHA1 digest for a single 512-bit block. * Be warned, it does not handle padding and message digest, do not @@ -36,47 +77,111 @@ * to clear the workspace. This is left to the caller to avoid * unnecessary clears between chained hashing operations. */ -void sha_transform(__u32 *digest, const char *in, __u32 *W) +void sha_transform(__u32 *digest, const char *data, __u32 *array) { - __u32 a, b, c, d, e, t, i; + __u32 A, B, C, D, E; - for (i = 0; i < 16; i++) - W[i] = be32_to_cpu(((const __be32 *)in)[i]); + A = digest[0]; + B = digest[1]; + C = digest[2]; + D = digest[3]; + E = digest[4]; - for (i = 0; i < 64; i++) - W[i+16] = rol32(W[i+13] ^ W[i+8] ^ W[i+2] ^ W[i], 1); + /* Round 1 - iterations 0-16 take their input from 'data' */ + T_0_15( 0, A, B, C, D, E); + T_0_15( 1, E, A, B, C, D); + T_0_15( 2, D, E, A, B, C); + T_0_15( 3, C, D, E, A, B); + T_0_15( 4, B, C, D, E, A); + T_0_15( 5, A, B, C, D, E); + T_0_15( 6, E, A, B, C, D); + T_0_15( 7, D, E, A, B, C); + T_0_15( 8, C, D, E, A, B); + T_0_15( 9, B, C, D, E, A); + T_0_15(10, A, B, C, D, E); + T_0_15(11, E, A, B, C, D); + T_0_15(12, D, E, A, B, C); + T_0_15(13, C, D, E, A, B); + T_0_15(14, B, C, D, E, A); + T_0_15(15, A, B, C, D, E); - a = digest[0]; - b = digest[1]; - c = digest[2]; - d = digest[3]; - e = digest[4]; + /* Round 1 - tail. Input from 512-bit mixing array */ + T_16_19(16, E, A, B, C, D); + T_16_19(17, D, E, A, B, C); + T_16_19(18, C, D, E, A, B); + T_16_19(19, B, C, D, E, A); - for (i = 0; i < 20; i++) { - t = f1(b, c, d) + K1 + rol32(a, 5) + e + W[i]; - e = d; d = c; c = rol32(b, 30); b = a; a = t; - } + /* Round 2 */ + T_20_39(20, A, B, C, D, E); + T_20_39(21, E, A, B, C, D); + T_20_39(22, D, E, A, B, C); + T_20_39(23, C, D, E, A, B); + T_20_39(24, B, C, D, E, A); + T_20_39(25, A, B, C, D, E); + T_20_39(26, E, A, B, C, D); + T_20_39(27, D, E, A, B, C); + T_20_39(28, C, D, E, A, B); + T_20_39(29, B, C, D, E, A); + T_20_39(30, A, B, C, D, E); + T_20_39(31, E, A, B, C, D); + T_20_39(32, D, E, A, B, C); + T_20_39(33, C, D, E, A, B); + T_20_39(34, B, C, D, E, A); + T_20_39(35, A, B, C, D, E); + T_20_39(36, E, A, B, C, D); + T_20_39(37, D, E, A, B, C); + T_20_39(38, C, D, E, A, B); + T_20_39(39, B, C, D, E, A); - for (; i < 40; i ++) { - t = f2(b, c, d) + K2 + rol32(a, 5) + e + W[i]; - e = d; d = c; c = rol32(b, 30); b = a; a = t; - } + /* Round 3 */ + T_40_59(40, A, B, C, D, E); + T_40_59(41, E, A, B, C, D); + T_40_59(42, D, E, A, B, C); + T_40_59(43, C, D, E, A, B); + T_40_59(44, B, C, D, E, A); + T_40_59(45, A, B, C, D, E); + T_40_59(46, E, A, B, C, D); + T_40_59(47, D, E, A, B, C); + T_40_59(48, C, D, E, A, B); + T_40_59(49, B, C, D, E, A); + T_40_59(50, A, B, C, D, E); + T_40_59(51, E, A, B, C, D); + T_40_59(52, D, E, A, B, C); + T_40_59(53, C, D, E, A, B); + T_40_59(54, B, C, D, E, A); + T_40_59(55, A, B, C, D, E); + T_40_59(56, E, A, B, C, D); + T_40_59(57, D, E, A, B, C); + T_40_59(58, C, D, E, A, B); + T_40_59(59, B, C, D, E, A); - for (; i < 60; i ++) { - t = f3(b, c, d) + K3 + rol32(a, 5) + e + W[i]; - e = d; d = c; c = rol32(b, 30); b = a; a = t; - } + /* Round 4 */ + T_60_79(60, A, B, C, D, E); + T_60_79(61, E, A, B, C, D); + T_60_79(62, D, E, A, B, C); + T_60_79(63, C, D, E, A, B); + T_60_79(64, B, C, D, E, A); + T_60_79(65, A, B, C, D, E); + T_60_79(66, E, A, B, C, D); + T_60_79(67, D, E, A, B, C); + T_60_79(68, C, D, E, A, B); + T_60_79(69, B, C, D, E, A); + T_60_79(70, A, B, C, D, E); + T_60_79(71, E, A, B, C, D); + T_60_79(72, D, E, A, B, C); + T_60_79(73, C, D, E, A, B); + T_60_79(74, B, C, D, E, A); + T_60_79(75, A, B, C, D, E); + T_60_79(76, E, A, B, C, D); + T_60_79(77, D, E, A, B, C); + T_60_79(78, C, D, E, A, B); + T_60_79(79, B, C, D, E, A); - for (; i < 80; i ++) { - t = f2(b, c, d) + K4 + rol32(a, 5) + e + W[i]; - e = d; d = c; c = rol32(b, 30); b = a; a = t; - } - - digest[0] += a; - digest[1] += b; - digest[2] += c; - digest[3] += d; - digest[4] += e; + digest[0] += A; + digest[1] += B; + digest[2] += C; + digest[3] += D; + digest[4] += E; } EXPORT_SYMBOL(sha_transform); @@ -92,4 +197,3 @@ void sha_init(__u32 *buf) buf[3] = 0x10325476; buf[4] = 0xc3d2e1f0; } - From c21427043dec93d40e3a1af970831d1f5f15ce5d Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sat, 6 Aug 2011 11:43:08 -0700 Subject: [PATCH 420/439] oom_ajd: don't use WARN_ONCE, just use printk_once WARN_ONCE() is very annoying, in that it shows the stack trace that we don't care about at all, and also triggers various user-level "kernel oopsed" logic that we really don't care about. And it's not like the user can do anything about the applications (sshd) in question, it's a distro issue. Requested-by: Andi Kleen (and many others) Signed-off-by: Linus Torvalds --- fs/proc/base.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/proc/base.c b/fs/proc/base.c index 08e3eccf9a12..73a562bf7266 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -1118,7 +1118,7 @@ static ssize_t oom_adjust_write(struct file *file, const char __user *buf, * Warn that /proc/pid/oom_adj is deprecated, see * Documentation/feature-removal-schedule.txt. */ - WARN_ONCE(1, "%s (%d): /proc/%d/oom_adj is deprecated, please use /proc/%d/oom_score_adj instead.\n", + printk_once(KERN_WARNING "%s (%d): /proc/%d/oom_adj is deprecated, please use /proc/%d/oom_score_adj instead.\n", current->comm, task_pid_nr(current), task_pid_nr(task), task_pid_nr(task)); task->signal->oom_adj = oom_adjust; From 1117f72ea0217ba0cc19f05adbbd8b9a397f5ab7 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sat, 6 Aug 2011 11:51:33 -0700 Subject: [PATCH 421/439] vfs: show O_CLOEXE bit properly in /proc//fdinfo/ files The CLOEXE bit is magical, and for performance (and semantic) reasons we don't actually maintain it in the file descriptor itself, but in a separate bit array. Which means that when we show f_flags, the CLOEXE status is shown incorrectly: we show the status not as it is now, but as it was when the file was opened. Fix that by looking up the bit properly in the 'fdt->close_on_exec' bit array. Uli needs this in order to re-implement the pfiles program: "For normal file descriptors (not sockets) this was the last piece of information which wasn't available. This is all part of my 'give Solaris users no reason to not switch' effort. I intend to offer the code to the util-linux-ng maintainers." Requested-by: Ulrich Drepper Signed-off-by: Linus Torvalds --- fs/proc/base.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/fs/proc/base.c b/fs/proc/base.c index 73a562bf7266..5eb02069e1b8 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -1919,6 +1919,14 @@ static int proc_fd_info(struct inode *inode, struct path *path, char *info) spin_lock(&files->file_lock); file = fcheck_files(files, fd); if (file) { + unsigned int f_flags; + struct fdtable *fdt; + + fdt = files_fdtable(files); + f_flags = file->f_flags & ~O_CLOEXEC; + if (FD_ISSET(fd, fdt->close_on_exec)) + f_flags |= O_CLOEXEC; + if (path) { *path = file->f_path; path_get(&file->f_path); @@ -1928,7 +1936,7 @@ static int proc_fd_info(struct inode *inode, struct path *path, char *info) "pos:\t%lli\n" "flags:\t0%o\n", (long long) file->f_pos, - file->f_flags); + f_flags); spin_unlock(&files->file_lock); put_files_struct(files); return 0; From 6f76b6fcaa6025bc9b2c00e055c7ccd39730568d Mon Sep 17 00:00:00 2001 From: Josh Triplett Date: Wed, 3 Aug 2011 12:19:07 -0700 Subject: [PATCH 422/439] CodingStyle: Document the exception of not splitting user-visible strings, for grepping Patch reviewers now recommend not splitting long user-visible strings, such as printk messages, even if they exceed 80 columns. This avoids breaking grep. However, that recommendation did not actually appear anywhere in Documentation/CodingStyle. See, for example, the thread at http://news.gmane.org/find-root.php?message_id=%3c1312215262.11635.15.camel%40Joe%2dLaptop%3e Signed-off-by: Josh Triplett Signed-off-by: Linus Torvalds --- Documentation/CodingStyle | 21 ++++++--------------- 1 file changed, 6 insertions(+), 15 deletions(-) diff --git a/Documentation/CodingStyle b/Documentation/CodingStyle index fa6e25b94a54..c940239d9678 100644 --- a/Documentation/CodingStyle +++ b/Documentation/CodingStyle @@ -80,22 +80,13 @@ available tools. The limit on the length of lines is 80 columns and this is a strongly preferred limit. -Statements longer than 80 columns will be broken into sensible chunks. -Descendants are always substantially shorter than the parent and are placed -substantially to the right. The same applies to function headers with a long -argument list. Long strings are as well broken into shorter strings. The -only exception to this is where exceeding 80 columns significantly increases -readability and does not hide information. +Statements longer than 80 columns will be broken into sensible chunks, unless +exceeding 80 columns significantly increases readability and does not hide +information. Descendants are always substantially shorter than the parent and +are placed substantially to the right. The same applies to function headers +with a long argument list. However, never break user-visible strings such as +printk messages, because that breaks the ability to grep for them. -void fun(int a, int b, int c) -{ - if (condition) - printk(KERN_WARNING "Warning this is a long printk with " - "3 parameters a: %u b: %u " - "c: %u \n", a, b, c); - else - next_statement; -} Chapter 3: Placing Braces and Spaces From 4b00e4b3940eabb38adeec0823751820fe2d6fda Mon Sep 17 00:00:00 2001 From: John Stanley Date: Wed, 3 Aug 2011 20:41:00 -0400 Subject: [PATCH 423/439] savagedb: Fix typo causing regression in savage4 series video chip detection Two additional savage4 variants were added, but the S3_SAVAGE4_SERIES macro was incompletely modified, resulting in a false positive detection of a savage4 card regardless of which savage card is actually present. For non-savage4 series cards, such as a Savage/IX-MV card, this results in garbled video and/or a hard-hang at boot time. Fix this by changing an '||' to an '&&' in the S3_SAVAGE4_SERIES macro. Signed-off-by: John P. Stanley Reviewed-by: Tormod Volden [ The macros have incomplete parenthesis too, but whatever .. -Linus ] Signed-off-by: Linus Torvalds --- drivers/video/savage/savagefb.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/video/savage/savagefb.h b/drivers/video/savage/savagefb.h index 32549d177b19..dcaab9012ca2 100644 --- a/drivers/video/savage/savagefb.h +++ b/drivers/video/savage/savagefb.h @@ -55,7 +55,7 @@ #define S3_SAVAGE3D_SERIES(chip) ((chip>=S3_SAVAGE3D) && (chip<=S3_SAVAGE_MX)) -#define S3_SAVAGE4_SERIES(chip) ((chip>=S3_SAVAGE4) || (chip<=S3_PROSAVAGEDDR)) +#define S3_SAVAGE4_SERIES(chip) ((chip>=S3_SAVAGE4) && (chip<=S3_PROSAVAGEDDR)) #define S3_SAVAGE_MOBILE_SERIES(chip) ((chip==S3_SAVAGE_MX) || (chip==S3_SUPERSAVAGE)) From bc0b96b54a21246e377122d54569eef71cec535f Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 3 Aug 2011 19:45:10 -0700 Subject: [PATCH 424/439] crypto: Move md5_transform to lib/md5.c We are going to use this for TCP/IP sequence number and fragment ID generation. Signed-off-by: David S. Miller --- crypto/md5.c | 92 +----------------------------------- include/linux/cryptohash.h | 5 ++ lib/Makefile | 2 +- lib/md5.c | 95 ++++++++++++++++++++++++++++++++++++++ 4 files changed, 102 insertions(+), 92 deletions(-) create mode 100644 lib/md5.c diff --git a/crypto/md5.c b/crypto/md5.c index 30efc7dad891..7febeaab923b 100644 --- a/crypto/md5.c +++ b/crypto/md5.c @@ -21,99 +21,9 @@ #include #include #include +#include #include -#define F1(x, y, z) (z ^ (x & (y ^ z))) -#define F2(x, y, z) F1(z, x, y) -#define F3(x, y, z) (x ^ y ^ z) -#define F4(x, y, z) (y ^ (x | ~z)) - -#define MD5STEP(f, w, x, y, z, in, s) \ - (w += f(x, y, z) + in, w = (w<>(32-s)) + x) - -static void md5_transform(u32 *hash, u32 const *in) -{ - u32 a, b, c, d; - - a = hash[0]; - b = hash[1]; - c = hash[2]; - d = hash[3]; - - MD5STEP(F1, a, b, c, d, in[0] + 0xd76aa478, 7); - MD5STEP(F1, d, a, b, c, in[1] + 0xe8c7b756, 12); - MD5STEP(F1, c, d, a, b, in[2] + 0x242070db, 17); - MD5STEP(F1, b, c, d, a, in[3] + 0xc1bdceee, 22); - MD5STEP(F1, a, b, c, d, in[4] + 0xf57c0faf, 7); - MD5STEP(F1, d, a, b, c, in[5] + 0x4787c62a, 12); - MD5STEP(F1, c, d, a, b, in[6] + 0xa8304613, 17); - MD5STEP(F1, b, c, d, a, in[7] + 0xfd469501, 22); - MD5STEP(F1, a, b, c, d, in[8] + 0x698098d8, 7); - MD5STEP(F1, d, a, b, c, in[9] + 0x8b44f7af, 12); - MD5STEP(F1, c, d, a, b, in[10] + 0xffff5bb1, 17); - MD5STEP(F1, b, c, d, a, in[11] + 0x895cd7be, 22); - MD5STEP(F1, a, b, c, d, in[12] + 0x6b901122, 7); - MD5STEP(F1, d, a, b, c, in[13] + 0xfd987193, 12); - MD5STEP(F1, c, d, a, b, in[14] + 0xa679438e, 17); - MD5STEP(F1, b, c, d, a, in[15] + 0x49b40821, 22); - - MD5STEP(F2, a, b, c, d, in[1] + 0xf61e2562, 5); - MD5STEP(F2, d, a, b, c, in[6] + 0xc040b340, 9); - MD5STEP(F2, c, d, a, b, in[11] + 0x265e5a51, 14); - MD5STEP(F2, b, c, d, a, in[0] + 0xe9b6c7aa, 20); - MD5STEP(F2, a, b, c, d, in[5] + 0xd62f105d, 5); - MD5STEP(F2, d, a, b, c, in[10] + 0x02441453, 9); - MD5STEP(F2, c, d, a, b, in[15] + 0xd8a1e681, 14); - MD5STEP(F2, b, c, d, a, in[4] + 0xe7d3fbc8, 20); - MD5STEP(F2, a, b, c, d, in[9] + 0x21e1cde6, 5); - MD5STEP(F2, d, a, b, c, in[14] + 0xc33707d6, 9); - MD5STEP(F2, c, d, a, b, in[3] + 0xf4d50d87, 14); - MD5STEP(F2, b, c, d, a, in[8] + 0x455a14ed, 20); - MD5STEP(F2, a, b, c, d, in[13] + 0xa9e3e905, 5); - MD5STEP(F2, d, a, b, c, in[2] + 0xfcefa3f8, 9); - MD5STEP(F2, c, d, a, b, in[7] + 0x676f02d9, 14); - MD5STEP(F2, b, c, d, a, in[12] + 0x8d2a4c8a, 20); - - MD5STEP(F3, a, b, c, d, in[5] + 0xfffa3942, 4); - MD5STEP(F3, d, a, b, c, in[8] + 0x8771f681, 11); - MD5STEP(F3, c, d, a, b, in[11] + 0x6d9d6122, 16); - MD5STEP(F3, b, c, d, a, in[14] + 0xfde5380c, 23); - MD5STEP(F3, a, b, c, d, in[1] + 0xa4beea44, 4); - MD5STEP(F3, d, a, b, c, in[4] + 0x4bdecfa9, 11); - MD5STEP(F3, c, d, a, b, in[7] + 0xf6bb4b60, 16); - MD5STEP(F3, b, c, d, a, in[10] + 0xbebfbc70, 23); - MD5STEP(F3, a, b, c, d, in[13] + 0x289b7ec6, 4); - MD5STEP(F3, d, a, b, c, in[0] + 0xeaa127fa, 11); - MD5STEP(F3, c, d, a, b, in[3] + 0xd4ef3085, 16); - MD5STEP(F3, b, c, d, a, in[6] + 0x04881d05, 23); - MD5STEP(F3, a, b, c, d, in[9] + 0xd9d4d039, 4); - MD5STEP(F3, d, a, b, c, in[12] + 0xe6db99e5, 11); - MD5STEP(F3, c, d, a, b, in[15] + 0x1fa27cf8, 16); - MD5STEP(F3, b, c, d, a, in[2] + 0xc4ac5665, 23); - - MD5STEP(F4, a, b, c, d, in[0] + 0xf4292244, 6); - MD5STEP(F4, d, a, b, c, in[7] + 0x432aff97, 10); - MD5STEP(F4, c, d, a, b, in[14] + 0xab9423a7, 15); - MD5STEP(F4, b, c, d, a, in[5] + 0xfc93a039, 21); - MD5STEP(F4, a, b, c, d, in[12] + 0x655b59c3, 6); - MD5STEP(F4, d, a, b, c, in[3] + 0x8f0ccc92, 10); - MD5STEP(F4, c, d, a, b, in[10] + 0xffeff47d, 15); - MD5STEP(F4, b, c, d, a, in[1] + 0x85845dd1, 21); - MD5STEP(F4, a, b, c, d, in[8] + 0x6fa87e4f, 6); - MD5STEP(F4, d, a, b, c, in[15] + 0xfe2ce6e0, 10); - MD5STEP(F4, c, d, a, b, in[6] + 0xa3014314, 15); - MD5STEP(F4, b, c, d, a, in[13] + 0x4e0811a1, 21); - MD5STEP(F4, a, b, c, d, in[4] + 0xf7537e82, 6); - MD5STEP(F4, d, a, b, c, in[11] + 0xbd3af235, 10); - MD5STEP(F4, c, d, a, b, in[2] + 0x2ad7d2bb, 15); - MD5STEP(F4, b, c, d, a, in[9] + 0xeb86d391, 21); - - hash[0] += a; - hash[1] += b; - hash[2] += c; - hash[3] += d; -} - /* XXX: this stuff can be optimized */ static inline void le32_to_cpu_array(u32 *buf, unsigned int words) { diff --git a/include/linux/cryptohash.h b/include/linux/cryptohash.h index ec78a4bbe1d5..d2984fbebbc0 100644 --- a/include/linux/cryptohash.h +++ b/include/linux/cryptohash.h @@ -8,6 +8,11 @@ void sha_init(__u32 *buf); void sha_transform(__u32 *digest, const char *data, __u32 *W); +#define MD5_DIGEST_WORDS 4 +#define MD5_MESSAGE_BYTES 64 + +void md5_transform(__u32 *hash, __u32 const *in); + __u32 half_md4_transform(__u32 buf[4], __u32 const in[8]); #endif diff --git a/lib/Makefile b/lib/Makefile index 6457af4a7caf..d5d175c8a6ca 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -10,7 +10,7 @@ endif lib-y := ctype.o string.o vsprintf.o cmdline.o \ rbtree.o radix-tree.o dump_stack.o timerqueue.o\ idr.o int_sqrt.o extable.o prio_tree.o \ - sha1.o irq_regs.o reciprocal_div.o argv_split.o \ + sha1.o md5.o irq_regs.o reciprocal_div.o argv_split.o \ proportions.o prio_heap.o ratelimit.o show_mem.o \ is_single_threaded.o plist.o decompress.o find_next_bit.o diff --git a/lib/md5.c b/lib/md5.c new file mode 100644 index 000000000000..c777180e1f2f --- /dev/null +++ b/lib/md5.c @@ -0,0 +1,95 @@ +#include +#include +#include + +#define F1(x, y, z) (z ^ (x & (y ^ z))) +#define F2(x, y, z) F1(z, x, y) +#define F3(x, y, z) (x ^ y ^ z) +#define F4(x, y, z) (y ^ (x | ~z)) + +#define MD5STEP(f, w, x, y, z, in, s) \ + (w += f(x, y, z) + in, w = (w<>(32-s)) + x) + +void md5_transform(__u32 *hash, __u32 const *in) +{ + u32 a, b, c, d; + + a = hash[0]; + b = hash[1]; + c = hash[2]; + d = hash[3]; + + MD5STEP(F1, a, b, c, d, in[0] + 0xd76aa478, 7); + MD5STEP(F1, d, a, b, c, in[1] + 0xe8c7b756, 12); + MD5STEP(F1, c, d, a, b, in[2] + 0x242070db, 17); + MD5STEP(F1, b, c, d, a, in[3] + 0xc1bdceee, 22); + MD5STEP(F1, a, b, c, d, in[4] + 0xf57c0faf, 7); + MD5STEP(F1, d, a, b, c, in[5] + 0x4787c62a, 12); + MD5STEP(F1, c, d, a, b, in[6] + 0xa8304613, 17); + MD5STEP(F1, b, c, d, a, in[7] + 0xfd469501, 22); + MD5STEP(F1, a, b, c, d, in[8] + 0x698098d8, 7); + MD5STEP(F1, d, a, b, c, in[9] + 0x8b44f7af, 12); + MD5STEP(F1, c, d, a, b, in[10] + 0xffff5bb1, 17); + MD5STEP(F1, b, c, d, a, in[11] + 0x895cd7be, 22); + MD5STEP(F1, a, b, c, d, in[12] + 0x6b901122, 7); + MD5STEP(F1, d, a, b, c, in[13] + 0xfd987193, 12); + MD5STEP(F1, c, d, a, b, in[14] + 0xa679438e, 17); + MD5STEP(F1, b, c, d, a, in[15] + 0x49b40821, 22); + + MD5STEP(F2, a, b, c, d, in[1] + 0xf61e2562, 5); + MD5STEP(F2, d, a, b, c, in[6] + 0xc040b340, 9); + MD5STEP(F2, c, d, a, b, in[11] + 0x265e5a51, 14); + MD5STEP(F2, b, c, d, a, in[0] + 0xe9b6c7aa, 20); + MD5STEP(F2, a, b, c, d, in[5] + 0xd62f105d, 5); + MD5STEP(F2, d, a, b, c, in[10] + 0x02441453, 9); + MD5STEP(F2, c, d, a, b, in[15] + 0xd8a1e681, 14); + MD5STEP(F2, b, c, d, a, in[4] + 0xe7d3fbc8, 20); + MD5STEP(F2, a, b, c, d, in[9] + 0x21e1cde6, 5); + MD5STEP(F2, d, a, b, c, in[14] + 0xc33707d6, 9); + MD5STEP(F2, c, d, a, b, in[3] + 0xf4d50d87, 14); + MD5STEP(F2, b, c, d, a, in[8] + 0x455a14ed, 20); + MD5STEP(F2, a, b, c, d, in[13] + 0xa9e3e905, 5); + MD5STEP(F2, d, a, b, c, in[2] + 0xfcefa3f8, 9); + MD5STEP(F2, c, d, a, b, in[7] + 0x676f02d9, 14); + MD5STEP(F2, b, c, d, a, in[12] + 0x8d2a4c8a, 20); + + MD5STEP(F3, a, b, c, d, in[5] + 0xfffa3942, 4); + MD5STEP(F3, d, a, b, c, in[8] + 0x8771f681, 11); + MD5STEP(F3, c, d, a, b, in[11] + 0x6d9d6122, 16); + MD5STEP(F3, b, c, d, a, in[14] + 0xfde5380c, 23); + MD5STEP(F3, a, b, c, d, in[1] + 0xa4beea44, 4); + MD5STEP(F3, d, a, b, c, in[4] + 0x4bdecfa9, 11); + MD5STEP(F3, c, d, a, b, in[7] + 0xf6bb4b60, 16); + MD5STEP(F3, b, c, d, a, in[10] + 0xbebfbc70, 23); + MD5STEP(F3, a, b, c, d, in[13] + 0x289b7ec6, 4); + MD5STEP(F3, d, a, b, c, in[0] + 0xeaa127fa, 11); + MD5STEP(F3, c, d, a, b, in[3] + 0xd4ef3085, 16); + MD5STEP(F3, b, c, d, a, in[6] + 0x04881d05, 23); + MD5STEP(F3, a, b, c, d, in[9] + 0xd9d4d039, 4); + MD5STEP(F3, d, a, b, c, in[12] + 0xe6db99e5, 11); + MD5STEP(F3, c, d, a, b, in[15] + 0x1fa27cf8, 16); + MD5STEP(F3, b, c, d, a, in[2] + 0xc4ac5665, 23); + + MD5STEP(F4, a, b, c, d, in[0] + 0xf4292244, 6); + MD5STEP(F4, d, a, b, c, in[7] + 0x432aff97, 10); + MD5STEP(F4, c, d, a, b, in[14] + 0xab9423a7, 15); + MD5STEP(F4, b, c, d, a, in[5] + 0xfc93a039, 21); + MD5STEP(F4, a, b, c, d, in[12] + 0x655b59c3, 6); + MD5STEP(F4, d, a, b, c, in[3] + 0x8f0ccc92, 10); + MD5STEP(F4, c, d, a, b, in[10] + 0xffeff47d, 15); + MD5STEP(F4, b, c, d, a, in[1] + 0x85845dd1, 21); + MD5STEP(F4, a, b, c, d, in[8] + 0x6fa87e4f, 6); + MD5STEP(F4, d, a, b, c, in[15] + 0xfe2ce6e0, 10); + MD5STEP(F4, c, d, a, b, in[6] + 0xa3014314, 15); + MD5STEP(F4, b, c, d, a, in[13] + 0x4e0811a1, 21); + MD5STEP(F4, a, b, c, d, in[4] + 0xf7537e82, 6); + MD5STEP(F4, d, a, b, c, in[11] + 0xbd3af235, 10); + MD5STEP(F4, c, d, a, b, in[2] + 0x2ad7d2bb, 15); + MD5STEP(F4, b, c, d, a, in[9] + 0xeb86d391, 21); + + hash[0] += a; + hash[1] += b; + hash[2] += c; + hash[3] += d; +} +EXPORT_SYMBOL(md5_transform); From 6e5714eaf77d79ae1c8b47e3e040ff5411b717ec Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 3 Aug 2011 20:50:44 -0700 Subject: [PATCH 425/439] net: Compute protocol sequence numbers and fragment IDs using MD5. Computers have become a lot faster since we compromised on the partial MD4 hash which we use currently for performance reasons. MD5 is a much safer choice, and is inline with both RFC1948 and other ISS generators (OpenBSD, Solaris, etc.) Furthermore, only having 24-bits of the sequence number be truly unpredictable is a very serious limitation. So the periodic regeneration and 8-bit counter have been removed. We compute and use a full 32-bit sequence number. For ipv6, DCCP was found to use a 32-bit truncated initial sequence number (it needs 43-bits) and that is fixed here as well. Reported-by: Dan Kaminsky Tested-by: Willy Tarreau Signed-off-by: David S. Miller --- drivers/char/random.c | 349 +---------------------- include/linux/random.h | 12 - include/net/secure_seq.h | 20 ++ net/core/Makefile | 2 +- net/core/secure_seq.c | 184 ++++++++++++ net/dccp/ipv4.c | 1 + net/dccp/ipv6.c | 9 +- net/ipv4/inet_hashtables.c | 1 + net/ipv4/inetpeer.c | 1 + net/ipv4/netfilter/nf_nat_proto_common.c | 1 + net/ipv4/route.c | 1 + net/ipv4/tcp_ipv4.c | 1 + net/ipv6/inet6_hashtables.c | 1 + net/ipv6/tcp_ipv6.c | 1 + 14 files changed, 223 insertions(+), 361 deletions(-) create mode 100644 include/net/secure_seq.h create mode 100644 net/core/secure_seq.c diff --git a/drivers/char/random.c b/drivers/char/random.c index 729281961f22..c35a785005b0 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -1300,345 +1300,14 @@ ctl_table random_table[] = { }; #endif /* CONFIG_SYSCTL */ -/******************************************************************** - * - * Random functions for networking - * - ********************************************************************/ +static u32 random_int_secret[MD5_MESSAGE_BYTES / 4] ____cacheline_aligned; -/* - * TCP initial sequence number picking. This uses the random number - * generator to pick an initial secret value. This value is hashed - * along with the TCP endpoint information to provide a unique - * starting point for each pair of TCP endpoints. This defeats - * attacks which rely on guessing the initial TCP sequence number. - * This algorithm was suggested by Steve Bellovin. - * - * Using a very strong hash was taking an appreciable amount of the total - * TCP connection establishment time, so this is a weaker hash, - * compensated for by changing the secret periodically. - */ - -/* F, G and H are basic MD4 functions: selection, majority, parity */ -#define F(x, y, z) ((z) ^ ((x) & ((y) ^ (z)))) -#define G(x, y, z) (((x) & (y)) + (((x) ^ (y)) & (z))) -#define H(x, y, z) ((x) ^ (y) ^ (z)) - -/* - * The generic round function. The application is so specific that - * we don't bother protecting all the arguments with parens, as is generally - * good macro practice, in favor of extra legibility. - * Rotation is separate from addition to prevent recomputation - */ -#define ROUND(f, a, b, c, d, x, s) \ - (a += f(b, c, d) + x, a = (a << s) | (a >> (32 - s))) -#define K1 0 -#define K2 013240474631UL -#define K3 015666365641UL - -#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) - -static __u32 twothirdsMD4Transform(__u32 const buf[4], __u32 const in[12]) +static int __init random_int_secret_init(void) { - __u32 a = buf[0], b = buf[1], c = buf[2], d = buf[3]; - - /* Round 1 */ - ROUND(F, a, b, c, d, in[ 0] + K1, 3); - ROUND(F, d, a, b, c, in[ 1] + K1, 7); - ROUND(F, c, d, a, b, in[ 2] + K1, 11); - ROUND(F, b, c, d, a, in[ 3] + K1, 19); - ROUND(F, a, b, c, d, in[ 4] + K1, 3); - ROUND(F, d, a, b, c, in[ 5] + K1, 7); - ROUND(F, c, d, a, b, in[ 6] + K1, 11); - ROUND(F, b, c, d, a, in[ 7] + K1, 19); - ROUND(F, a, b, c, d, in[ 8] + K1, 3); - ROUND(F, d, a, b, c, in[ 9] + K1, 7); - ROUND(F, c, d, a, b, in[10] + K1, 11); - ROUND(F, b, c, d, a, in[11] + K1, 19); - - /* Round 2 */ - ROUND(G, a, b, c, d, in[ 1] + K2, 3); - ROUND(G, d, a, b, c, in[ 3] + K2, 5); - ROUND(G, c, d, a, b, in[ 5] + K2, 9); - ROUND(G, b, c, d, a, in[ 7] + K2, 13); - ROUND(G, a, b, c, d, in[ 9] + K2, 3); - ROUND(G, d, a, b, c, in[11] + K2, 5); - ROUND(G, c, d, a, b, in[ 0] + K2, 9); - ROUND(G, b, c, d, a, in[ 2] + K2, 13); - ROUND(G, a, b, c, d, in[ 4] + K2, 3); - ROUND(G, d, a, b, c, in[ 6] + K2, 5); - ROUND(G, c, d, a, b, in[ 8] + K2, 9); - ROUND(G, b, c, d, a, in[10] + K2, 13); - - /* Round 3 */ - ROUND(H, a, b, c, d, in[ 3] + K3, 3); - ROUND(H, d, a, b, c, in[ 7] + K3, 9); - ROUND(H, c, d, a, b, in[11] + K3, 11); - ROUND(H, b, c, d, a, in[ 2] + K3, 15); - ROUND(H, a, b, c, d, in[ 6] + K3, 3); - ROUND(H, d, a, b, c, in[10] + K3, 9); - ROUND(H, c, d, a, b, in[ 1] + K3, 11); - ROUND(H, b, c, d, a, in[ 5] + K3, 15); - ROUND(H, a, b, c, d, in[ 9] + K3, 3); - ROUND(H, d, a, b, c, in[ 0] + K3, 9); - ROUND(H, c, d, a, b, in[ 4] + K3, 11); - ROUND(H, b, c, d, a, in[ 8] + K3, 15); - - return buf[1] + b; /* "most hashed" word */ - /* Alternative: return sum of all words? */ -} -#endif - -#undef ROUND -#undef F -#undef G -#undef H -#undef K1 -#undef K2 -#undef K3 - -/* This should not be decreased so low that ISNs wrap too fast. */ -#define REKEY_INTERVAL (300 * HZ) -/* - * Bit layout of the tcp sequence numbers (before adding current time): - * bit 24-31: increased after every key exchange - * bit 0-23: hash(source,dest) - * - * The implementation is similar to the algorithm described - * in the Appendix of RFC 1185, except that - * - it uses a 1 MHz clock instead of a 250 kHz clock - * - it performs a rekey every 5 minutes, which is equivalent - * to a (source,dest) tulple dependent forward jump of the - * clock by 0..2^(HASH_BITS+1) - * - * Thus the average ISN wraparound time is 68 minutes instead of - * 4.55 hours. - * - * SMP cleanup and lock avoidance with poor man's RCU. - * Manfred Spraul - * - */ -#define COUNT_BITS 8 -#define COUNT_MASK ((1 << COUNT_BITS) - 1) -#define HASH_BITS 24 -#define HASH_MASK ((1 << HASH_BITS) - 1) - -static struct keydata { - __u32 count; /* already shifted to the final position */ - __u32 secret[12]; -} ____cacheline_aligned ip_keydata[2]; - -static unsigned int ip_cnt; - -static void rekey_seq_generator(struct work_struct *work); - -static DECLARE_DELAYED_WORK(rekey_work, rekey_seq_generator); - -/* - * Lock avoidance: - * The ISN generation runs lockless - it's just a hash over random data. - * State changes happen every 5 minutes when the random key is replaced. - * Synchronization is performed by having two copies of the hash function - * state and rekey_seq_generator always updates the inactive copy. - * The copy is then activated by updating ip_cnt. - * The implementation breaks down if someone blocks the thread - * that processes SYN requests for more than 5 minutes. Should never - * happen, and even if that happens only a not perfectly compliant - * ISN is generated, nothing fatal. - */ -static void rekey_seq_generator(struct work_struct *work) -{ - struct keydata *keyptr = &ip_keydata[1 ^ (ip_cnt & 1)]; - - get_random_bytes(keyptr->secret, sizeof(keyptr->secret)); - keyptr->count = (ip_cnt & COUNT_MASK) << HASH_BITS; - smp_wmb(); - ip_cnt++; - schedule_delayed_work(&rekey_work, - round_jiffies_relative(REKEY_INTERVAL)); -} - -static inline struct keydata *get_keyptr(void) -{ - struct keydata *keyptr = &ip_keydata[ip_cnt & 1]; - - smp_rmb(); - - return keyptr; -} - -static __init int seqgen_init(void) -{ - rekey_seq_generator(NULL); + get_random_bytes(random_int_secret, sizeof(random_int_secret)); return 0; } -late_initcall(seqgen_init); - -#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) -__u32 secure_tcpv6_sequence_number(__be32 *saddr, __be32 *daddr, - __be16 sport, __be16 dport) -{ - __u32 seq; - __u32 hash[12]; - struct keydata *keyptr = get_keyptr(); - - /* The procedure is the same as for IPv4, but addresses are longer. - * Thus we must use twothirdsMD4Transform. - */ - - memcpy(hash, saddr, 16); - hash[4] = ((__force u16)sport << 16) + (__force u16)dport; - memcpy(&hash[5], keyptr->secret, sizeof(__u32) * 7); - - seq = twothirdsMD4Transform((const __u32 *)daddr, hash) & HASH_MASK; - seq += keyptr->count; - - seq += ktime_to_ns(ktime_get_real()); - - return seq; -} -EXPORT_SYMBOL(secure_tcpv6_sequence_number); -#endif - -/* The code below is shamelessly stolen from secure_tcp_sequence_number(). - * All blames to Andrey V. Savochkin . - */ -__u32 secure_ip_id(__be32 daddr) -{ - struct keydata *keyptr; - __u32 hash[4]; - - keyptr = get_keyptr(); - - /* - * Pick a unique starting offset for each IP destination. - * The dest ip address is placed in the starting vector, - * which is then hashed with random data. - */ - hash[0] = (__force __u32)daddr; - hash[1] = keyptr->secret[9]; - hash[2] = keyptr->secret[10]; - hash[3] = keyptr->secret[11]; - - return half_md4_transform(hash, keyptr->secret); -} - -__u32 secure_ipv6_id(const __be32 daddr[4]) -{ - const struct keydata *keyptr; - __u32 hash[4]; - - keyptr = get_keyptr(); - - hash[0] = (__force __u32)daddr[0]; - hash[1] = (__force __u32)daddr[1]; - hash[2] = (__force __u32)daddr[2]; - hash[3] = (__force __u32)daddr[3]; - - return half_md4_transform(hash, keyptr->secret); -} - -#ifdef CONFIG_INET - -__u32 secure_tcp_sequence_number(__be32 saddr, __be32 daddr, - __be16 sport, __be16 dport) -{ - __u32 seq; - __u32 hash[4]; - struct keydata *keyptr = get_keyptr(); - - /* - * Pick a unique starting offset for each TCP connection endpoints - * (saddr, daddr, sport, dport). - * Note that the words are placed into the starting vector, which is - * then mixed with a partial MD4 over random data. - */ - hash[0] = (__force u32)saddr; - hash[1] = (__force u32)daddr; - hash[2] = ((__force u16)sport << 16) + (__force u16)dport; - hash[3] = keyptr->secret[11]; - - seq = half_md4_transform(hash, keyptr->secret) & HASH_MASK; - seq += keyptr->count; - /* - * As close as possible to RFC 793, which - * suggests using a 250 kHz clock. - * Further reading shows this assumes 2 Mb/s networks. - * For 10 Mb/s Ethernet, a 1 MHz clock is appropriate. - * For 10 Gb/s Ethernet, a 1 GHz clock should be ok, but - * we also need to limit the resolution so that the u32 seq - * overlaps less than one time per MSL (2 minutes). - * Choosing a clock of 64 ns period is OK. (period of 274 s) - */ - seq += ktime_to_ns(ktime_get_real()) >> 6; - - return seq; -} - -/* Generate secure starting point for ephemeral IPV4 transport port search */ -u32 secure_ipv4_port_ephemeral(__be32 saddr, __be32 daddr, __be16 dport) -{ - struct keydata *keyptr = get_keyptr(); - u32 hash[4]; - - /* - * Pick a unique starting offset for each ephemeral port search - * (saddr, daddr, dport) and 48bits of random data. - */ - hash[0] = (__force u32)saddr; - hash[1] = (__force u32)daddr; - hash[2] = (__force u32)dport ^ keyptr->secret[10]; - hash[3] = keyptr->secret[11]; - - return half_md4_transform(hash, keyptr->secret); -} -EXPORT_SYMBOL_GPL(secure_ipv4_port_ephemeral); - -#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) -u32 secure_ipv6_port_ephemeral(const __be32 *saddr, const __be32 *daddr, - __be16 dport) -{ - struct keydata *keyptr = get_keyptr(); - u32 hash[12]; - - memcpy(hash, saddr, 16); - hash[4] = (__force u32)dport; - memcpy(&hash[5], keyptr->secret, sizeof(__u32) * 7); - - return twothirdsMD4Transform((const __u32 *)daddr, hash); -} -#endif - -#if defined(CONFIG_IP_DCCP) || defined(CONFIG_IP_DCCP_MODULE) -/* Similar to secure_tcp_sequence_number but generate a 48 bit value - * bit's 32-47 increase every key exchange - * 0-31 hash(source, dest) - */ -u64 secure_dccp_sequence_number(__be32 saddr, __be32 daddr, - __be16 sport, __be16 dport) -{ - u64 seq; - __u32 hash[4]; - struct keydata *keyptr = get_keyptr(); - - hash[0] = (__force u32)saddr; - hash[1] = (__force u32)daddr; - hash[2] = ((__force u16)sport << 16) + (__force u16)dport; - hash[3] = keyptr->secret[11]; - - seq = half_md4_transform(hash, keyptr->secret); - seq |= ((u64)keyptr->count) << (32 - HASH_BITS); - - seq += ktime_to_ns(ktime_get_real()); - seq &= (1ull << 48) - 1; - - return seq; -} -EXPORT_SYMBOL(secure_dccp_sequence_number); -#endif - -#endif /* CONFIG_INET */ - +late_initcall(random_int_secret_init); /* * Get a random word for internal kernel use only. Similar to urandom but @@ -1646,17 +1315,15 @@ EXPORT_SYMBOL(secure_dccp_sequence_number); * value is not cryptographically secure but for several uses the cost of * depleting entropy is too high */ -DEFINE_PER_CPU(__u32 [4], get_random_int_hash); +DEFINE_PER_CPU(__u32 [MD5_DIGEST_WORDS], get_random_int_hash); unsigned int get_random_int(void) { - struct keydata *keyptr; __u32 *hash = get_cpu_var(get_random_int_hash); - int ret; + unsigned int ret; - keyptr = get_keyptr(); hash[0] += current->pid + jiffies + get_cycles(); - - ret = half_md4_transform(hash, keyptr->secret); + md5_transform(hash, random_int_secret); + ret = hash[0]; put_cpu_var(get_random_int_hash); return ret; diff --git a/include/linux/random.h b/include/linux/random.h index ce29a040c8dc..d13059f3ea32 100644 --- a/include/linux/random.h +++ b/include/linux/random.h @@ -57,18 +57,6 @@ extern void add_interrupt_randomness(int irq); extern void get_random_bytes(void *buf, int nbytes); void generate_random_uuid(unsigned char uuid_out[16]); -extern __u32 secure_ip_id(__be32 daddr); -extern __u32 secure_ipv6_id(const __be32 daddr[4]); -extern u32 secure_ipv4_port_ephemeral(__be32 saddr, __be32 daddr, __be16 dport); -extern u32 secure_ipv6_port_ephemeral(const __be32 *saddr, const __be32 *daddr, - __be16 dport); -extern __u32 secure_tcp_sequence_number(__be32 saddr, __be32 daddr, - __be16 sport, __be16 dport); -extern __u32 secure_tcpv6_sequence_number(__be32 *saddr, __be32 *daddr, - __be16 sport, __be16 dport); -extern u64 secure_dccp_sequence_number(__be32 saddr, __be32 daddr, - __be16 sport, __be16 dport); - #ifndef MODULE extern const struct file_operations random_fops, urandom_fops; #endif diff --git a/include/net/secure_seq.h b/include/net/secure_seq.h new file mode 100644 index 000000000000..d97f6892c019 --- /dev/null +++ b/include/net/secure_seq.h @@ -0,0 +1,20 @@ +#ifndef _NET_SECURE_SEQ +#define _NET_SECURE_SEQ + +#include + +extern __u32 secure_ip_id(__be32 daddr); +extern __u32 secure_ipv6_id(const __be32 daddr[4]); +extern u32 secure_ipv4_port_ephemeral(__be32 saddr, __be32 daddr, __be16 dport); +extern u32 secure_ipv6_port_ephemeral(const __be32 *saddr, const __be32 *daddr, + __be16 dport); +extern __u32 secure_tcp_sequence_number(__be32 saddr, __be32 daddr, + __be16 sport, __be16 dport); +extern __u32 secure_tcpv6_sequence_number(__be32 *saddr, __be32 *daddr, + __be16 sport, __be16 dport); +extern u64 secure_dccp_sequence_number(__be32 saddr, __be32 daddr, + __be16 sport, __be16 dport); +extern u64 secure_dccpv6_sequence_number(__be32 *saddr, __be32 *daddr, + __be16 sport, __be16 dport); + +#endif /* _NET_SECURE_SEQ */ diff --git a/net/core/Makefile b/net/core/Makefile index 8a04dd22cf77..0d357b1c4e57 100644 --- a/net/core/Makefile +++ b/net/core/Makefile @@ -3,7 +3,7 @@ # obj-y := sock.o request_sock.o skbuff.o iovec.o datagram.o stream.o scm.o \ - gen_stats.o gen_estimator.o net_namespace.o + gen_stats.o gen_estimator.o net_namespace.o secure_seq.o obj-$(CONFIG_SYSCTL) += sysctl_net_core.o diff --git a/net/core/secure_seq.c b/net/core/secure_seq.c new file mode 100644 index 000000000000..45329d7c9dd9 --- /dev/null +++ b/net/core/secure_seq.c @@ -0,0 +1,184 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +static u32 net_secret[MD5_MESSAGE_BYTES / 4] ____cacheline_aligned; + +static int __init net_secret_init(void) +{ + get_random_bytes(net_secret, sizeof(net_secret)); + return 0; +} +late_initcall(net_secret_init); + +static u32 seq_scale(u32 seq) +{ + /* + * As close as possible to RFC 793, which + * suggests using a 250 kHz clock. + * Further reading shows this assumes 2 Mb/s networks. + * For 10 Mb/s Ethernet, a 1 MHz clock is appropriate. + * For 10 Gb/s Ethernet, a 1 GHz clock should be ok, but + * we also need to limit the resolution so that the u32 seq + * overlaps less than one time per MSL (2 minutes). + * Choosing a clock of 64 ns period is OK. (period of 274 s) + */ + return seq + (ktime_to_ns(ktime_get_real()) >> 6); +} + +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) +__u32 secure_tcpv6_sequence_number(__be32 *saddr, __be32 *daddr, + __be16 sport, __be16 dport) +{ + u32 secret[MD5_MESSAGE_BYTES / 4]; + u32 hash[MD5_DIGEST_WORDS]; + u32 i; + + memcpy(hash, saddr, 16); + for (i = 0; i < 4; i++) + secret[i] = net_secret[i] + daddr[i]; + secret[4] = net_secret[4] + + (((__force u16)sport << 16) + (__force u16)dport); + for (i = 5; i < MD5_MESSAGE_BYTES / 4; i++) + secret[i] = net_secret[i]; + + md5_transform(hash, secret); + + return seq_scale(hash[0]); +} +EXPORT_SYMBOL(secure_tcpv6_sequence_number); + +u32 secure_ipv6_port_ephemeral(const __be32 *saddr, const __be32 *daddr, + __be16 dport) +{ + u32 secret[MD5_MESSAGE_BYTES / 4]; + u32 hash[MD5_DIGEST_WORDS]; + u32 i; + + memcpy(hash, saddr, 16); + for (i = 0; i < 4; i++) + secret[i] = net_secret[i] + (__force u32) daddr[i]; + secret[4] = net_secret[4] + (__force u32)dport; + for (i = 5; i < MD5_MESSAGE_BYTES / 4; i++) + secret[i] = net_secret[i]; + + md5_transform(hash, secret); + + return hash[0]; +} +#endif + +#ifdef CONFIG_INET +__u32 secure_ip_id(__be32 daddr) +{ + u32 hash[MD5_DIGEST_WORDS]; + + hash[0] = (__force __u32) daddr; + hash[1] = net_secret[13]; + hash[2] = net_secret[14]; + hash[3] = net_secret[15]; + + md5_transform(hash, net_secret); + + return hash[0]; +} + +__u32 secure_ipv6_id(const __be32 daddr[4]) +{ + __u32 hash[4]; + + memcpy(hash, daddr, 16); + md5_transform(hash, net_secret); + + return hash[0]; +} + +__u32 secure_tcp_sequence_number(__be32 saddr, __be32 daddr, + __be16 sport, __be16 dport) +{ + u32 hash[MD5_DIGEST_WORDS]; + + hash[0] = (__force u32)saddr; + hash[1] = (__force u32)daddr; + hash[2] = ((__force u16)sport << 16) + (__force u16)dport; + hash[3] = net_secret[15]; + + md5_transform(hash, net_secret); + + return seq_scale(hash[0]); +} + +u32 secure_ipv4_port_ephemeral(__be32 saddr, __be32 daddr, __be16 dport) +{ + u32 hash[MD5_DIGEST_WORDS]; + + hash[0] = (__force u32)saddr; + hash[1] = (__force u32)daddr; + hash[2] = (__force u32)dport ^ net_secret[14]; + hash[3] = net_secret[15]; + + md5_transform(hash, net_secret); + + return hash[0]; +} +EXPORT_SYMBOL_GPL(secure_ipv4_port_ephemeral); +#endif + +#if defined(CONFIG_IP_DCCP) || defined(CONFIG_IP_DCCP_MODULE) +u64 secure_dccp_sequence_number(__be32 saddr, __be32 daddr, + __be16 sport, __be16 dport) +{ + u32 hash[MD5_DIGEST_WORDS]; + u64 seq; + + hash[0] = (__force u32)saddr; + hash[1] = (__force u32)daddr; + hash[2] = ((__force u16)sport << 16) + (__force u16)dport; + hash[3] = net_secret[15]; + + md5_transform(hash, net_secret); + + seq = hash[0] | (((u64)hash[1]) << 32); + seq += ktime_to_ns(ktime_get_real()); + seq &= (1ull << 48) - 1; + + return seq; +} +EXPORT_SYMBOL(secure_dccp_sequence_number); + +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) +u64 secure_dccpv6_sequence_number(__be32 *saddr, __be32 *daddr, + __be16 sport, __be16 dport) +{ + u32 secret[MD5_MESSAGE_BYTES / 4]; + u32 hash[MD5_DIGEST_WORDS]; + u64 seq; + u32 i; + + memcpy(hash, saddr, 16); + for (i = 0; i < 4; i++) + secret[i] = net_secret[i] + daddr[i]; + secret[4] = net_secret[4] + + (((__force u16)sport << 16) + (__force u16)dport); + for (i = 5; i < MD5_MESSAGE_BYTES / 4; i++) + secret[i] = net_secret[i]; + + md5_transform(hash, secret); + + seq = hash[0] | (((u64)hash[1]) << 32); + seq += ktime_to_ns(ktime_get_real()); + seq &= (1ull << 48) - 1; + + return seq; +} +EXPORT_SYMBOL(secure_dccpv6_sequence_number); +#endif +#endif diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 8c36adfd1919..332639b56f4d 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -26,6 +26,7 @@ #include #include #include +#include #include "ackvec.h" #include "ccid.h" diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 8dc4348774a5..b74f76117dcf 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -29,6 +29,7 @@ #include #include #include +#include #include "dccp.h" #include "ipv6.h" @@ -69,13 +70,7 @@ static inline void dccp_v6_send_check(struct sock *sk, struct sk_buff *skb) dh->dccph_checksum = dccp_v6_csum_finish(skb, &np->saddr, &np->daddr); } -static inline __u32 secure_dccpv6_sequence_number(__be32 *saddr, __be32 *daddr, - __be16 sport, __be16 dport ) -{ - return secure_tcpv6_sequence_number(saddr, daddr, sport, dport); -} - -static inline __u32 dccp_v6_init_sequence(struct sk_buff *skb) +static inline __u64 dccp_v6_init_sequence(struct sk_buff *skb) { return secure_dccpv6_sequence_number(ipv6_hdr(skb)->daddr.s6_addr32, ipv6_hdr(skb)->saddr.s6_addr32, diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index 3c0369a3a663..984ec656b03b 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -21,6 +21,7 @@ #include #include +#include #include /* diff --git a/net/ipv4/inetpeer.c b/net/ipv4/inetpeer.c index e38213817d0a..86f13c67ea85 100644 --- a/net/ipv4/inetpeer.c +++ b/net/ipv4/inetpeer.c @@ -19,6 +19,7 @@ #include #include #include +#include /* * Theory of operations. diff --git a/net/ipv4/netfilter/nf_nat_proto_common.c b/net/ipv4/netfilter/nf_nat_proto_common.c index 3e61faf23a9a..f52d41ea0690 100644 --- a/net/ipv4/netfilter/nf_nat_proto_common.c +++ b/net/ipv4/netfilter/nf_nat_proto_common.c @@ -12,6 +12,7 @@ #include #include +#include #include #include #include diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 6afc4eb50591..e3dec1c9f09d 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -109,6 +109,7 @@ #include #endif #include +#include #define RT_FL_TOS(oldflp4) \ ((u32)(oldflp4->flowi4_tos & (IPTOS_RT_MASK | RTO_ONLINK))) diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 955b8e65b69e..1c12b8ec849d 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -72,6 +72,7 @@ #include #include #include +#include #include #include diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c index b53197233709..73f1a00a96af 100644 --- a/net/ipv6/inet6_hashtables.c +++ b/net/ipv6/inet6_hashtables.c @@ -20,6 +20,7 @@ #include #include #include +#include #include int __inet6_hash(struct sock *sk, struct inet_timewait_sock *tw) diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 78aa53492b3e..d1fb63f4aeb7 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -61,6 +61,7 @@ #include #include #include +#include #include From e1042ba0991aab80ced34f7dade6ec25f22b4304 Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Tue, 16 Nov 2010 20:09:58 +0200 Subject: [PATCH 426/439] exofs: Add offset/length to exofs_get_io_state In future raid code we will need to know the IO offset/length and if it's a read or write to determine some of the array sizes we'll need. So add a new exofs_get_rw_state() API for use when writeing/reading. All other simple cases are left using the old way. The major change to this is that now we need to call exofs_get_io_state later at inode.c::read_exec and inode.c::write_exec when we actually know these things. So this patch is kept separate so I can test things apart from other changes. Signed-off-by: Boaz Harrosh --- fs/exofs/exofs.h | 6 +++++- fs/exofs/inode.c | 34 +++++++++++++++++++++------------- fs/exofs/ios.c | 14 ++++++++++++-- 3 files changed, 38 insertions(+), 16 deletions(-) diff --git a/fs/exofs/exofs.h b/fs/exofs/exofs.h index 9f62349a5a5c..fd913ddfd48b 100644 --- a/fs/exofs/exofs.h +++ b/fs/exofs/exofs.h @@ -109,7 +109,7 @@ static inline osd_id exofs_oi_objno(struct exofs_i_info *oi) } struct exofs_io_state; -typedef void (*exofs_io_done_fn)(struct exofs_io_state *or, void *private); +typedef void (*exofs_io_done_fn)(struct exofs_io_state *ios, void *private); struct exofs_io_state { struct kref kref; @@ -137,6 +137,8 @@ struct exofs_io_state { unsigned out_attr_len; struct osd_attr *out_attr; + bool reading; + /* Variable array of size numdevs */ unsigned numdevs; struct exofs_per_dev_state { @@ -218,6 +220,8 @@ void exofs_make_credential(u8 cred_a[OSD_CAP_LEN], int exofs_read_kern(struct osd_dev *od, u8 *cred, struct osd_obj_id *obj, u64 offset, void *p, unsigned length); +int exofs_get_rw_state(struct exofs_layout *layout, bool is_reading, + u64 offset, u64 length, struct exofs_io_state **ios); int exofs_get_io_state(struct exofs_layout *layout, struct exofs_io_state **ios); void exofs_put_io_state(struct exofs_io_state *ios); diff --git a/fs/exofs/inode.c b/fs/exofs/inode.c index 8472c098445d..ba9f0bedcbaf 100644 --- a/fs/exofs/inode.c +++ b/fs/exofs/inode.c @@ -110,13 +110,6 @@ static int pcol_try_alloc(struct page_collect *pcol) { unsigned pages; - if (!pcol->ios) { /* First time allocate io_state */ - int ret = exofs_get_io_state(&pcol->sbi->layout, &pcol->ios); - - if (ret) - return ret; - } - /* TODO: easily support bio chaining */ pages = exofs_max_io_pages(&pcol->sbi->layout, pcol->expected_pages); @@ -269,17 +262,25 @@ static void _unlock_pcol_pages(struct page_collect *pcol, int ret, int rw) static int read_exec(struct page_collect *pcol) { struct exofs_i_info *oi = exofs_i(pcol->inode); - struct exofs_io_state *ios = pcol->ios; + struct exofs_io_state *ios; struct page_collect *pcol_copy = NULL; int ret; if (!pcol->pages) return 0; + if (!pcol->ios) { + int ret = exofs_get_rw_state(&pcol->sbi->layout, true, + pcol->pg_first << PAGE_CACHE_SHIFT, + pcol->length, &pcol->ios); + + if (ret) + return ret; + } + + ios = pcol->ios; ios->pages = pcol->pages; ios->nr_pages = pcol->nr_pages; - ios->length = pcol->length; - ios->offset = pcol->pg_first << PAGE_CACHE_SHIFT; if (pcol->read_4_write) { exofs_oi_read(oi, pcol->ios); @@ -507,13 +508,21 @@ static void writepages_done(struct exofs_io_state *ios, void *p) static int write_exec(struct page_collect *pcol) { struct exofs_i_info *oi = exofs_i(pcol->inode); - struct exofs_io_state *ios = pcol->ios; + struct exofs_io_state *ios; struct page_collect *pcol_copy = NULL; int ret; if (!pcol->pages) return 0; + BUG_ON(pcol->ios); + ret = exofs_get_rw_state(&pcol->sbi->layout, false, + pcol->pg_first << PAGE_CACHE_SHIFT, + pcol->length, &pcol->ios); + + if (unlikely(ret)) + goto err; + pcol_copy = kmalloc(sizeof(*pcol_copy), GFP_KERNEL); if (!pcol_copy) { EXOFS_ERR("write_exec: Failed to kmalloc(pcol)\n"); @@ -523,10 +532,9 @@ static int write_exec(struct page_collect *pcol) *pcol_copy = *pcol; + ios = pcol->ios; ios->pages = pcol_copy->pages; ios->nr_pages = pcol_copy->nr_pages; - ios->offset = pcol_copy->pg_first << PAGE_CACHE_SHIFT; - ios->length = pcol_copy->length; ios->done = writepages_done; ios->private = pcol_copy; diff --git a/fs/exofs/ios.c b/fs/exofs/ios.c index fbb47ba2cd71..096405e51b94 100644 --- a/fs/exofs/ios.c +++ b/fs/exofs/ios.c @@ -69,8 +69,8 @@ out: return ret; } -int exofs_get_io_state(struct exofs_layout *layout, - struct exofs_io_state **pios) +int exofs_get_rw_state(struct exofs_layout *layout, bool is_reading, + u64 offset, u64 length, struct exofs_io_state **pios) { struct exofs_io_state *ios; @@ -87,10 +87,20 @@ int exofs_get_io_state(struct exofs_layout *layout, ios->layout = layout; ios->obj.partition = layout->s_pid; + ios->offset = offset; + ios->length = length; + ios->reading = is_reading; + *pios = ios; return 0; } +int exofs_get_io_state(struct exofs_layout *layout, + struct exofs_io_state **ios) +{ + return exofs_get_rw_state(layout, true, 0, 0, ios); +} + void exofs_put_io_state(struct exofs_io_state *ios) { if (ios) { From 85e44df4748670a1a7d8441b2d75843cdebc478a Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Mon, 16 May 2011 15:26:47 +0300 Subject: [PATCH 427/439] exofs: Move exofs specific osd operations out of ios.c ios.c will be moving to an external library, for use by the objects-layout-driver. Remove from it some exofs specific functions. Also g_attr_logical_length is used both by inode.c and ios.c move definition to the later, to keep it independent Signed-off-by: Boaz Harrosh --- fs/exofs/exofs.h | 17 +++++++----- fs/exofs/inode.c | 3 --- fs/exofs/ios.c | 64 +++------------------------------------------ fs/exofs/super.c | 68 +++++++++++++++++++++++++++++++++++++++++++++--- 4 files changed, 79 insertions(+), 73 deletions(-) diff --git a/fs/exofs/exofs.h b/fs/exofs/exofs.h index fd913ddfd48b..5a77fc76433f 100644 --- a/fs/exofs/exofs.h +++ b/fs/exofs/exofs.h @@ -205,6 +205,14 @@ static inline struct exofs_i_info *exofs_i(struct inode *inode) */ unsigned exofs_layout_od_id(struct exofs_layout *layout, osd_id obj_no, unsigned layout_index); + +static inline struct osd_dev *exofs_ios_od(struct exofs_io_state *ios, + unsigned layout_index) +{ + return ios->layout->s_ods[ + exofs_layout_od_id(ios->layout, ios->obj.id, layout_index)]; +} + /* * Maximum count of links to a file */ @@ -215,11 +223,6 @@ unsigned exofs_layout_od_id(struct exofs_layout *layout, *************************/ /* ios.c */ -void exofs_make_credential(u8 cred_a[OSD_CAP_LEN], - const struct osd_obj_id *obj); -int exofs_read_kern(struct osd_dev *od, u8 *cred, struct osd_obj_id *obj, - u64 offset, void *p, unsigned length); - int exofs_get_rw_state(struct exofs_layout *layout, bool is_reading, u64 offset, u64 length, struct exofs_io_state **ios); int exofs_get_io_state(struct exofs_layout *layout, @@ -234,6 +237,7 @@ int exofs_sbi_write(struct exofs_io_state *ios); int exofs_sbi_read(struct exofs_io_state *ios); int extract_attr_from_ios(struct exofs_io_state *ios, struct osd_attr *attr); +extern const struct osd_attr g_attr_logical_length; int exofs_oi_truncate(struct exofs_i_info *oi, u64 new_len); static inline int exofs_oi_write(struct exofs_i_info *oi, @@ -278,6 +282,8 @@ int exofs_set_link(struct inode *, struct exofs_dir_entry *, struct page *, struct inode *); /* super.c */ +void exofs_make_credential(u8 cred_a[OSD_CAP_LEN], + const struct osd_obj_id *obj); int exofs_sbi_write_stats(struct exofs_sb_info *sbi); /********************* @@ -292,7 +298,6 @@ extern const struct file_operations exofs_file_operations; /* inode.c */ extern const struct address_space_operations exofs_aops; -extern const struct osd_attr g_attr_logical_length; /* namei.c */ extern const struct inode_operations exofs_dir_inode_operations; diff --git a/fs/exofs/inode.c b/fs/exofs/inode.c index ba9f0bedcbaf..abcdac993039 100644 --- a/fs/exofs/inode.c +++ b/fs/exofs/inode.c @@ -852,9 +852,6 @@ static inline int exofs_inode_is_fast_symlink(struct inode *inode) return S_ISLNK(inode->i_mode) && (oi->i_data[0] != 0); } -const struct osd_attr g_attr_logical_length = ATTR_DEF( - OSD_APAGE_OBJECT_INFORMATION, OSD_ATTR_OI_LOGICAL_LENGTH, 8); - static int _do_truncate(struct inode *inode, loff_t newsize) { struct exofs_i_info *oi = exofs_i(inode); diff --git a/fs/exofs/ios.c b/fs/exofs/ios.c index 096405e51b94..21d6130b462e 100644 --- a/fs/exofs/ios.c +++ b/fs/exofs/ios.c @@ -31,44 +31,6 @@ #define EXOFS_DBGMSG2(M...) do {} while (0) /* #define EXOFS_DBGMSG2 EXOFS_DBGMSG */ -void exofs_make_credential(u8 cred_a[OSD_CAP_LEN], const struct osd_obj_id *obj) -{ - osd_sec_init_nosec_doall_caps(cred_a, obj, false, true); -} - -int exofs_read_kern(struct osd_dev *od, u8 *cred, struct osd_obj_id *obj, - u64 offset, void *p, unsigned length) -{ - struct osd_request *or = osd_start_request(od, GFP_KERNEL); -/* struct osd_sense_info osi = {.key = 0};*/ - int ret; - - if (unlikely(!or)) { - EXOFS_DBGMSG("%s: osd_start_request failed.\n", __func__); - return -ENOMEM; - } - ret = osd_req_read_kern(or, obj, offset, p, length); - if (unlikely(ret)) { - EXOFS_DBGMSG("%s: osd_req_read_kern failed.\n", __func__); - goto out; - } - - ret = osd_finalize_request(or, 0, cred, NULL); - if (unlikely(ret)) { - EXOFS_DBGMSG("Failed to osd_finalize_request() => %d\n", ret); - goto out; - } - - ret = osd_execute_request(or); - if (unlikely(ret)) - EXOFS_DBGMSG("osd_execute_request() => %d\n", ret); - /* osd_req_decode_sense(or, ret); */ - -out: - osd_end_request(or); - return ret; -} - int exofs_get_rw_state(struct exofs_layout *layout, bool is_reading, u64 offset, u64 length, struct exofs_io_state **pios) { @@ -119,29 +81,6 @@ void exofs_put_io_state(struct exofs_io_state *ios) } } -unsigned exofs_layout_od_id(struct exofs_layout *layout, - osd_id obj_no, unsigned layout_index) -{ -/* switch (layout->lay_func) { - case LAYOUT_MOVING_WINDOW: - {*/ - unsigned dev_mod = obj_no; - - return (layout_index + dev_mod * layout->mirrors_p1) % - layout->s_numdevs; -/* } - case LAYOUT_FUNC_IMPLICT: - return layout->devs[layout_index]; - }*/ -} - -static inline struct osd_dev *exofs_ios_od(struct exofs_io_state *ios, - unsigned layout_index) -{ - return ios->layout->s_ods[ - exofs_layout_od_id(ios->layout, ios->obj.id, layout_index)]; -} - static void _sync_done(struct exofs_io_state *ios, void *p) { struct completion *waiting = p; @@ -844,3 +783,6 @@ out: exofs_put_io_state(ios); return ret; } + +const struct osd_attr g_attr_logical_length = ATTR_DEF( + OSD_APAGE_OBJECT_INFORMATION, OSD_ATTR_OI_LOGICAL_LENGTH, 8); diff --git a/fs/exofs/super.c b/fs/exofs/super.c index 65fe5debe11c..8783f3d33c4a 100644 --- a/fs/exofs/super.c +++ b/fs/exofs/super.c @@ -40,6 +40,8 @@ #include "exofs.h" +#define EXOFS_DBGMSG2(M...) do {} while (0) + /****************************************************************************** * MOUNT OPTIONS *****************************************************************************/ @@ -208,10 +210,64 @@ static void destroy_inodecache(void) } /****************************************************************************** - * SUPERBLOCK FUNCTIONS + * Some osd helpers *****************************************************************************/ -static const struct super_operations exofs_sops; -static const struct export_operations exofs_export_ops; +void exofs_make_credential(u8 cred_a[OSD_CAP_LEN], const struct osd_obj_id *obj) +{ + osd_sec_init_nosec_doall_caps(cred_a, obj, false, true); +} + +static int exofs_read_kern(struct osd_dev *od, u8 *cred, struct osd_obj_id *obj, + u64 offset, void *p, unsigned length) +{ + struct osd_request *or = osd_start_request(od, GFP_KERNEL); +/* struct osd_sense_info osi = {.key = 0};*/ + int ret; + + if (unlikely(!or)) { + EXOFS_DBGMSG("%s: osd_start_request failed.\n", __func__); + return -ENOMEM; + } + ret = osd_req_read_kern(or, obj, offset, p, length); + if (unlikely(ret)) { + EXOFS_DBGMSG("%s: osd_req_read_kern failed.\n", __func__); + goto out; + } + + ret = osd_finalize_request(or, 0, cred, NULL); + if (unlikely(ret)) { + EXOFS_DBGMSG("Failed to osd_finalize_request() => %d\n", ret); + goto out; + } + + ret = osd_execute_request(or); + if (unlikely(ret)) + EXOFS_DBGMSG("osd_execute_request() => %d\n", ret); + /* osd_req_decode_sense(or, ret); */ + +out: + osd_end_request(or); + EXOFS_DBGMSG2("read_kern(0x%llx) offset=0x%llx " + "length=0x%llx dev=%p ret=>%d\n", + _LLU(obj->id), _LLU(offset), _LLU(length), od, ret); + return ret; +} + +unsigned exofs_layout_od_id(struct exofs_layout *layout, + osd_id obj_no, unsigned layout_index) +{ +/* switch (layout->lay_func) { + case LAYOUT_MOVING_WINDOW: + {*/ + unsigned dev_mod = obj_no; + + return (layout_index + dev_mod * layout->mirrors_p1) % + layout->s_numdevs; +/* } + case LAYOUT_FUNC_IMPLICT: + return layout->devs[layout_index]; + }*/ +} static const struct osd_attr g_attr_sb_stats = ATTR_DEF( EXOFS_APAGE_SB_DATA, @@ -308,6 +364,12 @@ int exofs_sbi_write_stats(struct exofs_sb_info *sbi) return ret; } +/****************************************************************************** + * SUPERBLOCK FUNCTIONS + *****************************************************************************/ +static const struct super_operations exofs_sops; +static const struct export_operations exofs_export_ops; + /* * Write the superblock to the OSD */ From 9e9db45649eb5d3ee5622fdad741914ecf1016a0 Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Fri, 5 Aug 2011 15:06:04 -0700 Subject: [PATCH 428/439] exofs: ios: Move to a per inode components & device-table Exofs raid engine was saving on memory space by having a single layout-info, single pid, and a single device-table, global to the filesystem. Then passing a credential and object_id info at the io_state level, private for each inode. It would also devise this contraption of rotating the device table view for each inode->ino to spread out the device usage. This is not compatible with the pnfs-objects standard, demanding that each inode can have it's own layout-info, device-table, and each object component it's own pid, oid and creds. So: Bring exofs raid engine to be usable for generic pnfs-objects use by: * Define an exofs_comp structure that holds obj_id and credential info. * Break up exofs_layout struct to an exofs_components structure that holds a possible array of exofs_comp and the array of devices + the size of the arrays. * Add a "comps" parameter to get_io_state() that specifies the ids creds and device array to use for each IO. This enables to keep the layout global, but the device-table view, creds and IDs at the inode level. It only adds two 64bit to each inode, since some of these members already existed in another form. * ios raid engine now access layout-info and comps-info through the passed pointers. Everything is pre-prepared by caller for generic access of these structures and arrays. At the exofs Level: * Super block holds an exofs_components struct that holds the device array, previously in layout. The devices there are in device-table order. The device-array is twice bigger and repeats the device-table twice so now each inode's device array can point to a random device and have a round-robin view of the table, making it compatible to previous exofs versions. * Each inode has an exofs_components struct that is initialized at load time, with it's own view of the device table IDs and creds. When doing IO this gets passed to the io_state together with the layout. While preforming this change. Bugs where found where credentials with the wrong IDs where used to access the different SB objects (super.c). As well as some dead code. It was never noticed because the target we use does not check the credentials. Signed-off-by: Boaz Harrosh --- fs/exofs/exofs.h | 104 ++++++++++++++++++++---------------- fs/exofs/inode.c | 59 ++++++++++---------- fs/exofs/ios.c | 102 +++++++++++++++++++++-------------- fs/exofs/super.c | 136 ++++++++++++++++++++++++----------------------- 4 files changed, 218 insertions(+), 183 deletions(-) diff --git a/fs/exofs/exofs.h b/fs/exofs/exofs.h index 5a77fc76433f..39ad50165162 100644 --- a/fs/exofs/exofs.h +++ b/fs/exofs/exofs.h @@ -52,9 +52,12 @@ /* u64 has problems with printk this will cast it to unsigned long long */ #define _LLU(x) (unsigned long long)(x) -struct exofs_layout { - osd_id s_pid; /* partition ID of file system*/ +struct exofs_comp { + struct osd_obj_id obj; + u8 cred[OSD_CAP_LEN]; +}; +struct exofs_layout { /* Our way of looking at the data_map */ unsigned stripe_unit; unsigned mirrors_p1; @@ -62,11 +65,18 @@ struct exofs_layout { unsigned group_width; u64 group_depth; unsigned group_count; +}; - enum exofs_inode_layout_gen_functions lay_func; - - unsigned s_numdevs; /* Num of devices in array */ - struct osd_dev **s_ods; /* osd_dev array */ +struct exofs_components { + unsigned numdevs; /* Num of devices in array */ + /* If @single_comp == EC_SINGLE_COMP, @comps points to a single + * component. else there are @numdevs components + */ + enum EC_COMP_USAGE { + EC_SINGLE_COMP = 0, EC_MULTPLE_COMPS = 0xffffffff + } single_comp; + struct exofs_comp *comps; + struct osd_dev **ods; /* osd_dev array */ }; /* @@ -81,12 +91,13 @@ struct exofs_sb_info { spinlock_t s_next_gen_lock; /* spinlock for gen # update */ u32 s_next_generation; /* next gen # to use */ atomic_t s_curr_pending; /* number of pending commands */ - uint8_t s_cred[OSD_CAP_LEN]; /* credential for the fscb */ struct pnfs_osd_data_map data_map; /* Default raid to use * FIXME: Needed ? */ struct exofs_layout layout; /* Default files layout */ + struct exofs_comp one_comp; /* id & cred of partition id=0*/ + struct exofs_components comps; /* comps for the partition */ struct osd_dev *_min_one_dev[1]; /* Place holder for one dev */ }; @@ -100,7 +111,8 @@ struct exofs_i_info { uint32_t i_data[EXOFS_IDATA];/*short symlink names and device #s*/ uint32_t i_dir_start_lookup; /* which page to start lookup */ uint64_t i_commit_size; /* the object's written length */ - uint8_t i_cred[OSD_CAP_LEN];/* all-powerful credential */ + struct exofs_comp one_comp; /* same component for all devices */ + struct exofs_components comps; /* inode view of the device table */ }; static inline osd_id exofs_oi_objno(struct exofs_i_info *oi) @@ -118,8 +130,7 @@ struct exofs_io_state { exofs_io_done_fn done; struct exofs_layout *layout; - struct osd_obj_id obj; - u8 *cred; + struct exofs_components *comps; /* Global read/write IO*/ loff_t offset; @@ -199,20 +210,6 @@ static inline struct exofs_i_info *exofs_i(struct inode *inode) return container_of(inode, struct exofs_i_info, vfs_inode); } -/* - * Given a layout, object_number and stripe_index return the associated global - * dev_index - */ -unsigned exofs_layout_od_id(struct exofs_layout *layout, - osd_id obj_no, unsigned layout_index); - -static inline struct osd_dev *exofs_ios_od(struct exofs_io_state *ios, - unsigned layout_index) -{ - return ios->layout->s_ods[ - exofs_layout_od_id(ios->layout, ios->obj.id, layout_index)]; -} - /* * Maximum count of links to a file */ @@ -223,10 +220,13 @@ static inline struct osd_dev *exofs_ios_od(struct exofs_io_state *ios, *************************/ /* ios.c */ -int exofs_get_rw_state(struct exofs_layout *layout, bool is_reading, - u64 offset, u64 length, struct exofs_io_state **ios); -int exofs_get_io_state(struct exofs_layout *layout, - struct exofs_io_state **ios); +int exofs_get_rw_state(struct exofs_layout *layout, + struct exofs_components *comps, + bool is_reading, u64 offset, u64 length, + struct exofs_io_state **ios); +int exofs_get_io_state(struct exofs_layout *layout, + struct exofs_components *comps, + struct exofs_io_state **ios); void exofs_put_io_state(struct exofs_io_state *ios); int exofs_check_io(struct exofs_io_state *ios, u64 *resid); @@ -235,27 +235,12 @@ int exofs_sbi_create(struct exofs_io_state *ios); int exofs_sbi_remove(struct exofs_io_state *ios); int exofs_sbi_write(struct exofs_io_state *ios); int exofs_sbi_read(struct exofs_io_state *ios); +int exofs_truncate(struct exofs_layout *layout, struct exofs_components *comps, + u64 size); int extract_attr_from_ios(struct exofs_io_state *ios, struct osd_attr *attr); extern const struct osd_attr g_attr_logical_length; -int exofs_oi_truncate(struct exofs_i_info *oi, u64 new_len); -static inline int exofs_oi_write(struct exofs_i_info *oi, - struct exofs_io_state *ios) -{ - ios->obj.id = exofs_oi_objno(oi); - ios->cred = oi->i_cred; - return exofs_sbi_write(ios); -} - -static inline int exofs_oi_read(struct exofs_i_info *oi, - struct exofs_io_state *ios) -{ - ios->obj.id = exofs_oi_objno(oi); - ios->cred = oi->i_cred; - return exofs_sbi_read(ios); -} - /* inode.c */ unsigned exofs_max_io_pages(struct exofs_layout *layout, unsigned expected_pages); @@ -307,4 +292,33 @@ extern const struct inode_operations exofs_special_inode_operations; extern const struct inode_operations exofs_symlink_inode_operations; extern const struct inode_operations exofs_fast_symlink_inode_operations; +/* exofs_init_comps will initialize an exofs_components device array + * pointing to a single exofs_comp struct, and a round-robin view + * of the device table. + * The first device of each inode is the [inode->ino % num_devices] + * and the rest of the devices sequentially following where the + * first device is after the last device. + * It is assumed that the global device array at @sbi is twice + * bigger and that the device table repeats twice. + * See: exofs_read_lookup_dev_table() + */ +static inline void exofs_init_comps(struct exofs_components *comps, + struct exofs_comp *one_comp, + struct exofs_sb_info *sbi, osd_id oid) +{ + unsigned dev_mod = (unsigned)oid, first_dev; + + one_comp->obj.partition = sbi->one_comp.obj.partition; + one_comp->obj.id = oid; + exofs_make_credential(one_comp->cred, &one_comp->obj); + + comps->numdevs = sbi->comps.numdevs; + comps->single_comp = EC_SINGLE_COMP; + comps->comps = one_comp; + + /* Round robin device view of the table */ + first_dev = (dev_mod * sbi->layout.mirrors_p1) % sbi->comps.numdevs; + comps->ods = sbi->comps.ods + first_dev; +} + #endif diff --git a/fs/exofs/inode.c b/fs/exofs/inode.c index abcdac993039..3cc24f25844d 100644 --- a/fs/exofs/inode.c +++ b/fs/exofs/inode.c @@ -270,7 +270,7 @@ static int read_exec(struct page_collect *pcol) return 0; if (!pcol->ios) { - int ret = exofs_get_rw_state(&pcol->sbi->layout, true, + int ret = exofs_get_rw_state(&pcol->sbi->layout, &oi->comps, true, pcol->pg_first << PAGE_CACHE_SHIFT, pcol->length, &pcol->ios); @@ -283,7 +283,7 @@ static int read_exec(struct page_collect *pcol) ios->nr_pages = pcol->nr_pages; if (pcol->read_4_write) { - exofs_oi_read(oi, pcol->ios); + exofs_sbi_read(pcol->ios); return __readpages_done(pcol); } @@ -296,14 +296,14 @@ static int read_exec(struct page_collect *pcol) *pcol_copy = *pcol; ios->done = readpages_done; ios->private = pcol_copy; - ret = exofs_oi_read(oi, ios); + ret = exofs_sbi_read(ios); if (unlikely(ret)) goto err; atomic_inc(&pcol->sbi->s_curr_pending); EXOFS_DBGMSG2("read_exec obj=0x%llx start=0x%llx length=0x%lx\n", - ios->obj.id, _LLU(ios->offset), pcol->length); + oi->one_comp.obj.id, _LLU(ios->offset), pcol->length); /* pages ownership was passed to pcol_copy */ _pcol_reset(pcol); @@ -516,7 +516,7 @@ static int write_exec(struct page_collect *pcol) return 0; BUG_ON(pcol->ios); - ret = exofs_get_rw_state(&pcol->sbi->layout, false, + ret = exofs_get_rw_state(&pcol->sbi->layout, &oi->comps, false, pcol->pg_first << PAGE_CACHE_SHIFT, pcol->length, &pcol->ios); @@ -538,9 +538,9 @@ static int write_exec(struct page_collect *pcol) ios->done = writepages_done; ios->private = pcol_copy; - ret = exofs_oi_write(oi, ios); + ret = exofs_sbi_write(ios); if (unlikely(ret)) { - EXOFS_ERR("write_exec: exofs_oi_write() Failed\n"); + EXOFS_ERR("write_exec: exofs_sbi_write() Failed\n"); goto err; } @@ -855,11 +855,12 @@ static inline int exofs_inode_is_fast_symlink(struct inode *inode) static int _do_truncate(struct inode *inode, loff_t newsize) { struct exofs_i_info *oi = exofs_i(inode); + struct exofs_sb_info *sbi = inode->i_sb->s_fs_info; int ret; inode->i_mtime = inode->i_ctime = CURRENT_TIME; - ret = exofs_oi_truncate(oi, (u64)newsize); + ret = exofs_truncate(&sbi->layout, &oi->comps, (u64)newsize); if (likely(!ret)) truncate_setsize(inode, newsize); @@ -926,18 +927,14 @@ static int exofs_get_inode(struct super_block *sb, struct exofs_i_info *oi, struct exofs_on_disk_inode_layout *layout; int ret; - ret = exofs_get_io_state(&sbi->layout, &ios); + ret = exofs_get_io_state(&sbi->layout, &oi->comps, &ios); if (unlikely(ret)) { EXOFS_ERR("%s: exofs_get_io_state failed.\n", __func__); return ret; } - ios->obj.id = exofs_oi_objno(oi); - exofs_make_credential(oi->i_cred, &ios->obj); - ios->cred = oi->i_cred; - - attrs[1].len = exofs_on_disk_inode_layout_size(sbi->layout.s_numdevs); - attrs[2].len = exofs_on_disk_inode_layout_size(sbi->layout.s_numdevs); + attrs[1].len = exofs_on_disk_inode_layout_size(sbi->comps.numdevs); + attrs[2].len = exofs_on_disk_inode_layout_size(sbi->comps.numdevs); ios->in_attr = attrs; ios->in_attr_len = ARRAY_SIZE(attrs); @@ -945,7 +942,7 @@ static int exofs_get_inode(struct super_block *sb, struct exofs_i_info *oi, ret = exofs_sbi_read(ios); if (unlikely(ret)) { EXOFS_ERR("object(0x%llx) corrupted, return empty file=>%d\n", - _LLU(ios->obj.id), ret); + _LLU(oi->one_comp.obj.id), ret); memset(inode, 0, sizeof(*inode)); inode->i_mode = 0040000 | (0777 & ~022); /* If object is lost on target we might as well enable it's @@ -1021,6 +1018,8 @@ struct inode *exofs_iget(struct super_block *sb, unsigned long ino) return inode; oi = exofs_i(inode); __oi_init(oi); + exofs_init_comps(&oi->comps, &oi->one_comp, sb->s_fs_info, + exofs_oi_objno(oi)); /* read the inode from the osd */ ret = exofs_get_inode(sb, oi, &fcb); @@ -1126,7 +1125,8 @@ static void create_done(struct exofs_io_state *ios, void *p) if (unlikely(ret)) { EXOFS_ERR("object=0x%llx creation failed in pid=0x%llx", - _LLU(exofs_oi_objno(oi)), _LLU(sbi->layout.s_pid)); + _LLU(exofs_oi_objno(oi)), + _LLU(oi->one_comp.obj.partition)); /*TODO: When FS is corrupted creation can fail, object already * exist. Get rid of this asynchronous creation, if exist * increment the obj counter and try the next object. Until we @@ -1145,14 +1145,13 @@ static void create_done(struct exofs_io_state *ios, void *p) */ struct inode *exofs_new_inode(struct inode *dir, int mode) { - struct super_block *sb; + struct super_block *sb = dir->i_sb; + struct exofs_sb_info *sbi = sb->s_fs_info; struct inode *inode; struct exofs_i_info *oi; - struct exofs_sb_info *sbi; struct exofs_io_state *ios; int ret; - sb = dir->i_sb; inode = new_inode(sb); if (!inode) return ERR_PTR(-ENOMEM); @@ -1162,8 +1161,6 @@ struct inode *exofs_new_inode(struct inode *dir, int mode) set_obj_2bcreated(oi); - sbi = sb->s_fs_info; - inode->i_mapping->backing_dev_info = sb->s_bdi; inode_init_owner(inode, dir, mode); inode->i_ino = sbi->s_nextid++; @@ -1175,22 +1172,21 @@ struct inode *exofs_new_inode(struct inode *dir, int mode) spin_unlock(&sbi->s_next_gen_lock); insert_inode_hash(inode); + exofs_init_comps(&oi->comps, &oi->one_comp, sb->s_fs_info, + exofs_oi_objno(oi)); exofs_sbi_write_stats(sbi); /* Make sure new sbi->s_nextid is on disk */ mark_inode_dirty(inode); - ret = exofs_get_io_state(&sbi->layout, &ios); + ret = exofs_get_io_state(&sbi->layout, &oi->comps, &ios); if (unlikely(ret)) { EXOFS_ERR("exofs_new_inode: exofs_get_io_state failed\n"); return ERR_PTR(ret); } - ios->obj.id = exofs_oi_objno(oi); - exofs_make_credential(oi->i_cred, &ios->obj); - ios->done = create_done; ios->private = inode; - ios->cred = oi->i_cred; + ret = exofs_sbi_create(ios); if (ret) { exofs_put_io_state(ios); @@ -1271,7 +1267,7 @@ static int exofs_update_inode(struct inode *inode, int do_sync) } else memcpy(fcb->i_data, oi->i_data, sizeof(fcb->i_data)); - ret = exofs_get_io_state(&sbi->layout, &ios); + ret = exofs_get_io_state(&sbi->layout, &oi->comps, &ios); if (unlikely(ret)) { EXOFS_ERR("%s: exofs_get_io_state failed.\n", __func__); goto free_args; @@ -1290,7 +1286,7 @@ static int exofs_update_inode(struct inode *inode, int do_sync) ios->private = args; } - ret = exofs_oi_write(oi, ios); + ret = exofs_sbi_write(ios); if (!do_sync && !ret) { atomic_inc(&sbi->s_curr_pending); goto out; /* deallocation in updatei_done */ @@ -1354,16 +1350,15 @@ void exofs_evict_inode(struct inode *inode) /* ignore the error, attempt a remove anyway */ /* Now Remove the OSD objects */ - ret = exofs_get_io_state(&sbi->layout, &ios); + ret = exofs_get_io_state(&sbi->layout, &oi->comps, &ios); if (unlikely(ret)) { EXOFS_ERR("%s: exofs_get_io_state failed\n", __func__); return; } - ios->obj.id = exofs_oi_objno(oi); ios->done = delete_done; ios->private = sbi; - ios->cred = oi->i_cred; + ret = exofs_sbi_remove(ios); if (ret) { EXOFS_ERR("%s: exofs_sbi_remove failed\n", __func__); diff --git a/fs/exofs/ios.c b/fs/exofs/ios.c index 21d6130b462e..f9d5c5a823fb 100644 --- a/fs/exofs/ios.c +++ b/fs/exofs/ios.c @@ -31,24 +31,41 @@ #define EXOFS_DBGMSG2(M...) do {} while (0) /* #define EXOFS_DBGMSG2 EXOFS_DBGMSG */ -int exofs_get_rw_state(struct exofs_layout *layout, bool is_reading, - u64 offset, u64 length, struct exofs_io_state **pios) +static u8 *_ios_cred(struct exofs_io_state *ios, unsigned index) +{ + return ios->comps->comps[index & ios->comps->single_comp].cred; +} + +static struct osd_obj_id *_ios_obj(struct exofs_io_state *ios, unsigned index) +{ + return &ios->comps->comps[index & ios->comps->single_comp].obj; +} + +static struct osd_dev *_ios_od(struct exofs_io_state *ios, unsigned index) +{ + return ios->comps->ods[index]; +} + +int exofs_get_rw_state(struct exofs_layout *layout, + struct exofs_components *comps, + bool is_reading, u64 offset, u64 length, + struct exofs_io_state **pios) { struct exofs_io_state *ios; /*TODO: Maybe use kmem_cach per sbi of size * exofs_io_state_size(layout->s_numdevs) */ - ios = kzalloc(exofs_io_state_size(layout->s_numdevs), GFP_KERNEL); + ios = kzalloc(exofs_io_state_size(comps->numdevs), GFP_KERNEL); if (unlikely(!ios)) { EXOFS_DBGMSG("Failed kzalloc bytes=%d\n", - exofs_io_state_size(layout->s_numdevs)); + exofs_io_state_size(comps->numdevs)); *pios = NULL; return -ENOMEM; } ios->layout = layout; - ios->obj.partition = layout->s_pid; + ios->comps = comps; ios->offset = offset; ios->length = length; ios->reading = is_reading; @@ -58,9 +75,10 @@ int exofs_get_rw_state(struct exofs_layout *layout, bool is_reading, } int exofs_get_io_state(struct exofs_layout *layout, + struct exofs_components *comps, struct exofs_io_state **ios) { - return exofs_get_rw_state(layout, true, 0, 0, ios); + return exofs_get_rw_state(layout, comps, true, 0, 0, ios); } void exofs_put_io_state(struct exofs_io_state *ios) @@ -119,7 +137,7 @@ static int exofs_io_execute(struct exofs_io_state *ios) if (unlikely(!or)) continue; - ret = osd_finalize_request(or, 0, ios->cred, NULL); + ret = osd_finalize_request(or, 0, _ios_cred(ios, i), NULL); if (unlikely(ret)) { EXOFS_DBGMSG("Failed to osd_finalize_request() => %d\n", ret); @@ -300,7 +318,7 @@ static int _add_stripe_unit(struct exofs_io_state *ios, unsigned *cur_pg, { unsigned pg = *cur_pg; struct request_queue *q = - osd_request_queue(exofs_ios_od(ios, per_dev->dev)); + osd_request_queue(_ios_od(ios, per_dev->dev)); per_dev->length += cur_len; @@ -440,10 +458,10 @@ int exofs_sbi_create(struct exofs_io_state *ios) { int i, ret; - for (i = 0; i < ios->layout->s_numdevs; i++) { + for (i = 0; i < ios->comps->numdevs; i++) { struct osd_request *or; - or = osd_start_request(exofs_ios_od(ios, i), GFP_KERNEL); + or = osd_start_request(_ios_od(ios, i), GFP_KERNEL); if (unlikely(!or)) { EXOFS_ERR("%s: osd_start_request failed\n", __func__); ret = -ENOMEM; @@ -452,7 +470,7 @@ int exofs_sbi_create(struct exofs_io_state *ios) ios->per_dev[i].or = or; ios->numdevs++; - osd_req_create_object(or, &ios->obj); + osd_req_create_object(or, _ios_obj(ios, i)); } ret = exofs_io_execute(ios); @@ -464,10 +482,10 @@ int exofs_sbi_remove(struct exofs_io_state *ios) { int i, ret; - for (i = 0; i < ios->layout->s_numdevs; i++) { + for (i = 0; i < ios->comps->numdevs; i++) { struct osd_request *or; - or = osd_start_request(exofs_ios_od(ios, i), GFP_KERNEL); + or = osd_start_request(_ios_od(ios, i), GFP_KERNEL); if (unlikely(!or)) { EXOFS_ERR("%s: osd_start_request failed\n", __func__); ret = -ENOMEM; @@ -476,7 +494,7 @@ int exofs_sbi_remove(struct exofs_io_state *ios) ios->per_dev[i].or = or; ios->numdevs++; - osd_req_remove_object(or, &ios->obj); + osd_req_remove_object(or, _ios_obj(ios, i)); } ret = exofs_io_execute(ios); @@ -498,7 +516,7 @@ static int _sbi_write_mirror(struct exofs_io_state *ios, int cur_comp) struct exofs_per_dev_state *per_dev = &ios->per_dev[cur_comp]; struct osd_request *or; - or = osd_start_request(exofs_ios_od(ios, dev), GFP_KERNEL); + or = osd_start_request(_ios_od(ios, dev), GFP_KERNEL); if (unlikely(!or)) { EXOFS_ERR("%s: osd_start_request failed\n", __func__); ret = -ENOMEM; @@ -533,25 +551,29 @@ static int _sbi_write_mirror(struct exofs_io_state *ios, int cur_comp) bio->bi_rw |= REQ_WRITE; } - osd_req_write(or, &ios->obj, per_dev->offset, bio, - per_dev->length); + osd_req_write(or, _ios_obj(ios, dev), per_dev->offset, + bio, per_dev->length); EXOFS_DBGMSG("write(0x%llx) offset=0x%llx " "length=0x%llx dev=%d\n", - _LLU(ios->obj.id), _LLU(per_dev->offset), + _LLU(_ios_obj(ios, dev)->id), + _LLU(per_dev->offset), _LLU(per_dev->length), dev); } else if (ios->kern_buff) { - ret = osd_req_write_kern(or, &ios->obj, per_dev->offset, - ios->kern_buff, ios->length); + ret = osd_req_write_kern(or, _ios_obj(ios, dev), + per_dev->offset, + ios->kern_buff, ios->length); if (unlikely(ret)) goto out; EXOFS_DBGMSG2("write_kern(0x%llx) offset=0x%llx " "length=0x%llx dev=%d\n", - _LLU(ios->obj.id), _LLU(per_dev->offset), + _LLU(_ios_obj(ios, dev)->id), + _LLU(per_dev->offset), _LLU(ios->length), dev); } else { - osd_req_set_attributes(or, &ios->obj); + osd_req_set_attributes(or, _ios_obj(ios, dev)); EXOFS_DBGMSG2("obj(0x%llx) set_attributes=%d dev=%d\n", - _LLU(ios->obj.id), ios->out_attr_len, dev); + _LLU(_ios_obj(ios, dev)->id), + ios->out_attr_len, dev); } if (ios->out_attr) @@ -590,13 +612,14 @@ static int _sbi_read_mirror(struct exofs_io_state *ios, unsigned cur_comp) { struct osd_request *or; struct exofs_per_dev_state *per_dev = &ios->per_dev[cur_comp]; - unsigned first_dev = (unsigned)ios->obj.id; + struct osd_obj_id *obj = _ios_obj(ios, cur_comp); + unsigned first_dev = (unsigned)obj->id; if (ios->pages && !per_dev->length) return 0; /* Just an empty slot */ first_dev = per_dev->dev + first_dev % ios->layout->mirrors_p1; - or = osd_start_request(exofs_ios_od(ios, first_dev), GFP_KERNEL); + or = osd_start_request(_ios_od(ios, first_dev), GFP_KERNEL); if (unlikely(!or)) { EXOFS_ERR("%s: osd_start_request failed\n", __func__); return -ENOMEM; @@ -604,25 +627,26 @@ static int _sbi_read_mirror(struct exofs_io_state *ios, unsigned cur_comp) per_dev->or = or; if (ios->pages) { - osd_req_read(or, &ios->obj, per_dev->offset, + osd_req_read(or, obj, per_dev->offset, per_dev->bio, per_dev->length); EXOFS_DBGMSG("read(0x%llx) offset=0x%llx length=0x%llx" - " dev=%d\n", _LLU(ios->obj.id), + " dev=%d\n", _LLU(obj->id), _LLU(per_dev->offset), _LLU(per_dev->length), first_dev); } else if (ios->kern_buff) { - int ret = osd_req_read_kern(or, &ios->obj, per_dev->offset, + int ret = osd_req_read_kern(or, obj, per_dev->offset, ios->kern_buff, ios->length); EXOFS_DBGMSG2("read_kern(0x%llx) offset=0x%llx " "length=0x%llx dev=%d ret=>%d\n", - _LLU(ios->obj.id), _LLU(per_dev->offset), + _LLU(obj->id), _LLU(per_dev->offset), _LLU(ios->length), first_dev, ret); if (unlikely(ret)) return ret; } else { - osd_req_get_attributes(or, &ios->obj); + osd_req_get_attributes(or, obj); EXOFS_DBGMSG2("obj(0x%llx) get_attributes=%d dev=%d\n", - _LLU(ios->obj.id), ios->in_attr_len, first_dev); + _LLU(obj->id), + ios->in_attr_len, first_dev); } if (ios->out_attr) osd_req_add_set_attr_list(or, ios->out_attr, ios->out_attr_len); @@ -682,14 +706,14 @@ static int _truncate_mirrors(struct exofs_io_state *ios, unsigned cur_comp, struct exofs_per_dev_state *per_dev = &ios->per_dev[cur_comp]; struct osd_request *or; - or = osd_start_request(exofs_ios_od(ios, cur_comp), GFP_KERNEL); + or = osd_start_request(_ios_od(ios, cur_comp), GFP_KERNEL); if (unlikely(!or)) { EXOFS_ERR("%s: osd_start_request failed\n", __func__); return -ENOMEM; } per_dev->or = or; - osd_req_set_attributes(or, &ios->obj); + osd_req_set_attributes(or, _ios_obj(ios, cur_comp)); osd_req_add_set_attr_list(or, attr, 1); } @@ -721,9 +745,9 @@ void _calc_trunk_info(struct exofs_layout *layout, u64 file_offset, ti->max_devs = layout->group_width * layout->group_count; } -int exofs_oi_truncate(struct exofs_i_info *oi, u64 size) +int exofs_truncate(struct exofs_layout *layout, struct exofs_components *comps, + u64 size) { - struct exofs_sb_info *sbi = oi->vfs_inode.i_sb->s_fs_info; struct exofs_io_state *ios; struct exofs_trunc_attr { struct osd_attr attr; @@ -732,7 +756,7 @@ int exofs_oi_truncate(struct exofs_i_info *oi, u64 size) struct _trunc_info ti; int i, ret; - ret = exofs_get_io_state(&sbi->layout, &ios); + ret = exofs_get_io_state(layout, comps, &ios); if (unlikely(ret)) return ret; @@ -745,9 +769,7 @@ int exofs_oi_truncate(struct exofs_i_info *oi, u64 size) goto out; } - ios->obj.id = exofs_oi_objno(oi); - ios->cred = oi->i_cred; - ios->numdevs = ios->layout->s_numdevs; + ios->numdevs = ios->comps->numdevs; for (i = 0; i < ti.max_devs; ++i) { struct exofs_trunc_attr *size_attr = &size_attrs[i]; @@ -770,7 +792,7 @@ int exofs_oi_truncate(struct exofs_i_info *oi, u64 size) size_attr->attr.val_ptr = &size_attr->newsize; EXOFS_DBGMSG("trunc(0x%llx) obj_offset=0x%llx dev=%d\n", - _LLU(ios->obj.id), _LLU(obj_size), i); + _LLU(comps->comps->obj.id), _LLU(obj_size), i); ret = _truncate_mirrors(ios, i * ios->layout->mirrors_p1, &size_attr->attr); if (unlikely(ret)) diff --git a/fs/exofs/super.c b/fs/exofs/super.c index 8783f3d33c4a..4d6877967c87 100644 --- a/fs/exofs/super.c +++ b/fs/exofs/super.c @@ -253,22 +253,6 @@ out: return ret; } -unsigned exofs_layout_od_id(struct exofs_layout *layout, - osd_id obj_no, unsigned layout_index) -{ -/* switch (layout->lay_func) { - case LAYOUT_MOVING_WINDOW: - {*/ - unsigned dev_mod = obj_no; - - return (layout_index + dev_mod * layout->mirrors_p1) % - layout->s_numdevs; -/* } - case LAYOUT_FUNC_IMPLICT: - return layout->devs[layout_index]; - }*/ -} - static const struct osd_attr g_attr_sb_stats = ATTR_DEF( EXOFS_APAGE_SB_DATA, EXOFS_ATTR_SB_STATS, @@ -282,14 +266,12 @@ static int __sbi_read_stats(struct exofs_sb_info *sbi) struct exofs_io_state *ios; int ret; - ret = exofs_get_io_state(&sbi->layout, &ios); + ret = exofs_get_io_state(&sbi->layout, &sbi->comps, &ios); if (unlikely(ret)) { EXOFS_ERR("%s: exofs_get_io_state failed.\n", __func__); return ret; } - ios->cred = sbi->s_cred; - ios->in_attr = attrs; ios->in_attr_len = ARRAY_SIZE(attrs); @@ -339,7 +321,7 @@ int exofs_sbi_write_stats(struct exofs_sb_info *sbi) struct exofs_io_state *ios; int ret; - ret = exofs_get_io_state(&sbi->layout, &ios); + ret = exofs_get_io_state(&sbi->layout, &sbi->comps, &ios); if (unlikely(ret)) { EXOFS_ERR("%s: exofs_get_io_state failed.\n", __func__); return ret; @@ -349,7 +331,7 @@ int exofs_sbi_write_stats(struct exofs_sb_info *sbi) sbi->s_ess.s_numfiles = cpu_to_le64(sbi->s_numfiles); attrs[0].val_ptr = &sbi->s_ess; - ios->cred = sbi->s_cred; + ios->done = stats_done; ios->private = sbi; ios->out_attr = attrs; @@ -377,6 +359,8 @@ int exofs_sync_fs(struct super_block *sb, int wait) { struct exofs_sb_info *sbi; struct exofs_fscb *fscb; + struct exofs_comp one_comp; + struct exofs_components comps; struct exofs_io_state *ios; int ret = -ENOMEM; @@ -393,7 +377,10 @@ int exofs_sync_fs(struct super_block *sb, int wait) * version). Otherwise the exofs_fscb is read-only from mkfs time. All * the writeable info is set in exofs_sbi_write_stats() above. */ - ret = exofs_get_io_state(&sbi->layout, &ios); + + exofs_init_comps(&comps, &one_comp, sbi, EXOFS_SUPER_ID); + + ret = exofs_get_io_state(&sbi->layout, &comps, &ios); if (unlikely(ret)) goto out; @@ -407,10 +394,8 @@ int exofs_sync_fs(struct super_block *sb, int wait) fscb->s_newfs = 0; fscb->s_version = EXOFS_FSCB_VER; - ios->obj.id = EXOFS_SUPER_ID; ios->offset = 0; ios->kern_buff = fscb; - ios->cred = sbi->s_cred; ret = exofs_sbi_write(ios); if (unlikely(ret)) @@ -446,17 +431,17 @@ static void _exofs_print_device(const char *msg, const char *dev_path, void exofs_free_sbi(struct exofs_sb_info *sbi) { - while (sbi->layout.s_numdevs) { - int i = --sbi->layout.s_numdevs; - struct osd_dev *od = sbi->layout.s_ods[i]; + while (sbi->comps.numdevs) { + int i = --sbi->comps.numdevs; + struct osd_dev *od = sbi->comps.ods[i]; if (od) { - sbi->layout.s_ods[i] = NULL; + sbi->comps.ods[i] = NULL; osduld_put_device(od); } } - if (sbi->layout.s_ods != sbi->_min_one_dev) - kfree(sbi->layout.s_ods); + if (sbi->comps.ods != sbi->_min_one_dev) + kfree(sbi->comps.ods); kfree(sbi); } @@ -483,8 +468,8 @@ static void exofs_put_super(struct super_block *sb) msecs_to_jiffies(100)); } - _exofs_print_device("Unmounting", NULL, sbi->layout.s_ods[0], - sbi->layout.s_pid); + _exofs_print_device("Unmounting", NULL, sbi->comps.ods[0], + sbi->one_comp.obj.partition); bdi_destroy(&sbi->bdi); exofs_free_sbi(sbi); @@ -624,8 +609,7 @@ static int exofs_read_lookup_dev_table(struct exofs_sb_info *sbi, struct osd_dev *fscb_od, unsigned table_count) { - struct osd_obj_id obj = {.partition = sbi->layout.s_pid, - .id = EXOFS_DEVTABLE_ID}; + struct exofs_comp comp; struct exofs_device_table *dt; unsigned table_bytes = table_count * sizeof(dt->dt_dev_table[0]) + sizeof(*dt); @@ -639,8 +623,14 @@ static int exofs_read_lookup_dev_table(struct exofs_sb_info *sbi, return -ENOMEM; } - sbi->layout.s_numdevs = 0; - ret = exofs_read_kern(fscb_od, sbi->s_cred, &obj, 0, dt, table_bytes); + sbi->comps.numdevs = 0; + + comp.obj.partition = sbi->one_comp.obj.partition; + comp.obj.id = EXOFS_DEVTABLE_ID; + exofs_make_credential(comp.cred, &comp.obj); + + ret = exofs_read_kern(fscb_od, comp.cred, &comp.obj, 0, dt, + table_bytes); if (unlikely(ret)) { EXOFS_ERR("ERROR: reading device table\n"); goto out; @@ -658,10 +648,13 @@ static int exofs_read_lookup_dev_table(struct exofs_sb_info *sbi, goto out; if (likely(numdevs > 1)) { - unsigned size = numdevs * sizeof(sbi->layout.s_ods[0]); + unsigned size = numdevs * sizeof(sbi->comps.ods[0]); - sbi->layout.s_ods = kzalloc(size, GFP_KERNEL); - if (unlikely(!sbi->layout.s_ods)) { + /* Twice bigger table: See exofs_init_comps() and below + * comment + */ + sbi->comps.ods = kzalloc(size + size - 1, GFP_KERNEL); + if (unlikely(!sbi->comps.ods)) { EXOFS_ERR("ERROR: faild allocating Device array[%d]\n", numdevs); ret = -ENOMEM; @@ -688,8 +681,8 @@ static int exofs_read_lookup_dev_table(struct exofs_sb_info *sbi, * line. We always keep them in device-table order. */ if (fscb_od && osduld_device_same(fscb_od, &odi)) { - sbi->layout.s_ods[i] = fscb_od; - ++sbi->layout.s_numdevs; + sbi->comps.ods[i] = fscb_od; + ++sbi->comps.numdevs; fscb_od = NULL; continue; } @@ -702,13 +695,13 @@ static int exofs_read_lookup_dev_table(struct exofs_sb_info *sbi, goto out; } - sbi->layout.s_ods[i] = od; - ++sbi->layout.s_numdevs; + sbi->comps.ods[i] = od; + ++sbi->comps.numdevs; /* Read the fscb of the other devices to make sure the FS * partition is there. */ - ret = exofs_read_kern(od, sbi->s_cred, &obj, 0, &fscb, + ret = exofs_read_kern(od, comp.cred, &comp.obj, 0, &fscb, sizeof(fscb)); if (unlikely(ret)) { EXOFS_ERR("ERROR: Malformed participating device " @@ -725,13 +718,22 @@ static int exofs_read_lookup_dev_table(struct exofs_sb_info *sbi, out: kfree(dt); - if (unlikely(!ret && fscb_od)) { - EXOFS_ERR( - "ERROR: Bad device-table container device not present\n"); - osduld_put_device(fscb_od); - ret = -EINVAL; - } + if (likely(!ret)) { + unsigned numdevs = sbi->comps.numdevs; + if (unlikely(fscb_od)) { + EXOFS_ERR("ERROR: Bad device-table container device not present\n"); + osduld_put_device(fscb_od); + return -EINVAL; + } + /* exofs round-robins the device table view according to inode + * number. We hold a: twice bigger table hence inodes can point + * to any device and have a sequential view of the table + * starting at this device. See exofs_init_comps() + */ + for (i = 0; i < numdevs - 1; ++i) + sbi->comps.ods[i + numdevs] = sbi->comps.ods[i]; + } return ret; } @@ -745,7 +747,7 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent) struct exofs_sb_info *sbi; /*extended info */ struct osd_dev *od; /* Master device */ struct exofs_fscb fscb; /*on-disk superblock info */ - struct osd_obj_id obj; + struct exofs_comp comp; unsigned table_count; int ret; @@ -776,11 +778,16 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent) sbi->layout.group_width = 1; sbi->layout.group_depth = -1; sbi->layout.group_count = 1; - sbi->layout.s_ods = sbi->_min_one_dev; - sbi->layout.s_numdevs = 1; - sbi->layout.s_pid = opts->pid; sbi->s_timeout = opts->timeout; + sbi->one_comp.obj.partition = opts->pid; + sbi->one_comp.obj.id = 0; + exofs_make_credential(sbi->one_comp.cred, &sbi->one_comp.obj); + sbi->comps.numdevs = 1; + sbi->comps.single_comp = EC_SINGLE_COMP; + sbi->comps.comps = &sbi->one_comp; + sbi->comps.ods = sbi->_min_one_dev; + /* fill in some other data by hand */ memset(sb->s_id, 0, sizeof(sb->s_id)); strcpy(sb->s_id, "exofs"); @@ -791,11 +798,11 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent) sb->s_bdev = NULL; sb->s_dev = 0; - obj.partition = sbi->layout.s_pid; - obj.id = EXOFS_SUPER_ID; - exofs_make_credential(sbi->s_cred, &obj); + comp.obj.partition = sbi->one_comp.obj.partition; + comp.obj.id = EXOFS_SUPER_ID; + exofs_make_credential(comp.cred, &comp.obj); - ret = exofs_read_kern(od, sbi->s_cred, &obj, 0, &fscb, sizeof(fscb)); + ret = exofs_read_kern(od, comp.cred, &comp.obj, 0, &fscb, sizeof(fscb)); if (unlikely(ret)) goto free_sbi; @@ -828,7 +835,7 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent) if (unlikely(ret)) goto free_sbi; } else { - sbi->layout.s_ods[0] = od; + sbi->comps.ods[0] = od; } __sbi_read_stats(sbi); @@ -868,13 +875,13 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent) goto free_sbi; } - _exofs_print_device("Mounting", opts->dev_name, sbi->layout.s_ods[0], - sbi->layout.s_pid); + _exofs_print_device("Mounting", opts->dev_name, sbi->comps.ods[0], + sbi->one_comp.obj.partition); return 0; free_sbi: EXOFS_ERR("Unable to mount exofs on %s pid=0x%llx err=%d\n", - opts->dev_name, sbi->layout.s_pid, ret); + opts->dev_name, sbi->one_comp.obj.partition, ret); exofs_free_sbi(sbi); return ret; } @@ -915,17 +922,14 @@ static int exofs_statfs(struct dentry *dentry, struct kstatfs *buf) }; uint64_t capacity = ULLONG_MAX; uint64_t used = ULLONG_MAX; - uint8_t cred_a[OSD_CAP_LEN]; int ret; - ret = exofs_get_io_state(&sbi->layout, &ios); + ret = exofs_get_io_state(&sbi->layout, &sbi->comps, &ios); if (ret) { EXOFS_DBGMSG("exofs_get_io_state failed.\n"); return ret; } - exofs_make_credential(cred_a, &ios->obj); - ios->cred = sbi->s_cred; ios->in_attr = attrs; ios->in_attr_len = ARRAY_SIZE(attrs); From 8ff660ab85f524bdc7652eb5d38aaef1d66aa9c7 Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Sat, 6 Aug 2011 19:26:31 -0700 Subject: [PATCH 429/439] exofs: Rename raid engine from exofs/ios.c => ore ORE stands for "Objects Raid Engine" This patch is a mechanical rename of everything that was in ios.c and its API declaration to an ore.c and an osd_ore.h header. The ore engine will later be used by the pnfs objects layout driver. * File ios.c => ore.c * Declaration of types and API are moved from exofs.h to a new osd_ore.h * All used types are prefixed by ore_ from their exofs_ name. * Shift includes from exofs.h to osd_ore.h so osd_ore.h is independent, include it from exofs.h. Other than a pure rename there are no other changes. Next patch will move the ore into it's own module and will export the API to be used by exofs and later the layout driver Signed-off-by: Boaz Harrosh --- fs/exofs/Kbuild | 2 +- fs/exofs/exofs.h | 119 +++------------------------- fs/exofs/inode.c | 88 ++++++++++----------- fs/exofs/{ios.c => ore.c} | 160 ++++++++++++++++++++------------------ fs/exofs/super.c | 56 ++++++------- include/scsi/osd_ore.h | 125 +++++++++++++++++++++++++++++ 6 files changed, 295 insertions(+), 255 deletions(-) rename fs/exofs/{ios.c => ore.c} (79%) create mode 100644 include/scsi/osd_ore.h diff --git a/fs/exofs/Kbuild b/fs/exofs/Kbuild index 2d0f757fda3e..c1af2217b087 100644 --- a/fs/exofs/Kbuild +++ b/fs/exofs/Kbuild @@ -12,5 +12,5 @@ # Kbuild - Gets included from the Kernels Makefile and build system # -exofs-y := ios.o inode.o file.o symlink.o namei.o dir.o super.o +exofs-y := ore.o inode.o file.o symlink.o namei.o dir.o super.o obj-$(CONFIG_EXOFS_FS) += exofs.o diff --git a/fs/exofs/exofs.h b/fs/exofs/exofs.h index 39ad50165162..f4e442ec7445 100644 --- a/fs/exofs/exofs.h +++ b/fs/exofs/exofs.h @@ -36,7 +36,8 @@ #include #include #include -#include +#include + #include "common.h" #define EXOFS_ERR(fmt, a...) printk(KERN_ERR "exofs: " fmt, ##a) @@ -52,33 +53,6 @@ /* u64 has problems with printk this will cast it to unsigned long long */ #define _LLU(x) (unsigned long long)(x) -struct exofs_comp { - struct osd_obj_id obj; - u8 cred[OSD_CAP_LEN]; -}; - -struct exofs_layout { - /* Our way of looking at the data_map */ - unsigned stripe_unit; - unsigned mirrors_p1; - - unsigned group_width; - u64 group_depth; - unsigned group_count; -}; - -struct exofs_components { - unsigned numdevs; /* Num of devices in array */ - /* If @single_comp == EC_SINGLE_COMP, @comps points to a single - * component. else there are @numdevs components - */ - enum EC_COMP_USAGE { - EC_SINGLE_COMP = 0, EC_MULTPLE_COMPS = 0xffffffff - } single_comp; - struct exofs_comp *comps; - struct osd_dev **ods; /* osd_dev array */ -}; - /* * our extension to the in-memory superblock */ @@ -95,9 +69,9 @@ struct exofs_sb_info { struct pnfs_osd_data_map data_map; /* Default raid to use * FIXME: Needed ? */ - struct exofs_layout layout; /* Default files layout */ - struct exofs_comp one_comp; /* id & cred of partition id=0*/ - struct exofs_components comps; /* comps for the partition */ + struct ore_layout layout; /* Default files layout */ + struct ore_comp one_comp; /* id & cred of partition id=0*/ + struct ore_components comps; /* comps for the partition */ struct osd_dev *_min_one_dev[1]; /* Place holder for one dev */ }; @@ -111,8 +85,8 @@ struct exofs_i_info { uint32_t i_data[EXOFS_IDATA];/*short symlink names and device #s*/ uint32_t i_dir_start_lookup; /* which page to start lookup */ uint64_t i_commit_size; /* the object's written length */ - struct exofs_comp one_comp; /* same component for all devices */ - struct exofs_components comps; /* inode view of the device table */ + struct ore_comp one_comp; /* same component for all devices */ + struct ore_components comps; /* inode view of the device table */ }; static inline osd_id exofs_oi_objno(struct exofs_i_info *oi) @@ -120,53 +94,6 @@ static inline osd_id exofs_oi_objno(struct exofs_i_info *oi) return oi->vfs_inode.i_ino + EXOFS_OBJ_OFF; } -struct exofs_io_state; -typedef void (*exofs_io_done_fn)(struct exofs_io_state *ios, void *private); - -struct exofs_io_state { - struct kref kref; - - void *private; - exofs_io_done_fn done; - - struct exofs_layout *layout; - struct exofs_components *comps; - - /* Global read/write IO*/ - loff_t offset; - unsigned long length; - void *kern_buff; - - struct page **pages; - unsigned nr_pages; - unsigned pgbase; - unsigned pages_consumed; - - /* Attributes */ - unsigned in_attr_len; - struct osd_attr *in_attr; - unsigned out_attr_len; - struct osd_attr *out_attr; - - bool reading; - - /* Variable array of size numdevs */ - unsigned numdevs; - struct exofs_per_dev_state { - struct osd_request *or; - struct bio *bio; - loff_t offset; - unsigned length; - unsigned dev; - } per_dev[]; -}; - -static inline unsigned exofs_io_state_size(unsigned numdevs) -{ - return sizeof(struct exofs_io_state) + - sizeof(struct exofs_per_dev_state) * numdevs; -} - /* * our inode flags */ @@ -219,30 +146,8 @@ static inline struct exofs_i_info *exofs_i(struct inode *inode) * function declarations * *************************/ -/* ios.c */ -int exofs_get_rw_state(struct exofs_layout *layout, - struct exofs_components *comps, - bool is_reading, u64 offset, u64 length, - struct exofs_io_state **ios); -int exofs_get_io_state(struct exofs_layout *layout, - struct exofs_components *comps, - struct exofs_io_state **ios); -void exofs_put_io_state(struct exofs_io_state *ios); - -int exofs_check_io(struct exofs_io_state *ios, u64 *resid); - -int exofs_sbi_create(struct exofs_io_state *ios); -int exofs_sbi_remove(struct exofs_io_state *ios); -int exofs_sbi_write(struct exofs_io_state *ios); -int exofs_sbi_read(struct exofs_io_state *ios); -int exofs_truncate(struct exofs_layout *layout, struct exofs_components *comps, - u64 size); - -int extract_attr_from_ios(struct exofs_io_state *ios, struct osd_attr *attr); -extern const struct osd_attr g_attr_logical_length; - /* inode.c */ -unsigned exofs_max_io_pages(struct exofs_layout *layout, +unsigned exofs_max_io_pages(struct ore_layout *layout, unsigned expected_pages); int exofs_setattr(struct dentry *, struct iattr *); int exofs_write_begin(struct file *file, struct address_space *mapping, @@ -292,8 +197,8 @@ extern const struct inode_operations exofs_special_inode_operations; extern const struct inode_operations exofs_symlink_inode_operations; extern const struct inode_operations exofs_fast_symlink_inode_operations; -/* exofs_init_comps will initialize an exofs_components device array - * pointing to a single exofs_comp struct, and a round-robin view +/* exofs_init_comps will initialize an ore_components device array + * pointing to a single ore_comp struct, and a round-robin view * of the device table. * The first device of each inode is the [inode->ino % num_devices] * and the rest of the devices sequentially following where the @@ -302,8 +207,8 @@ extern const struct inode_operations exofs_fast_symlink_inode_operations; * bigger and that the device table repeats twice. * See: exofs_read_lookup_dev_table() */ -static inline void exofs_init_comps(struct exofs_components *comps, - struct exofs_comp *one_comp, +static inline void exofs_init_comps(struct ore_components *comps, + struct ore_comp *one_comp, struct exofs_sb_info *sbi, osd_id oid) { unsigned dev_mod = (unsigned)oid, first_dev; diff --git a/fs/exofs/inode.c b/fs/exofs/inode.c index 3cc24f25844d..f39a38fc2349 100644 --- a/fs/exofs/inode.c +++ b/fs/exofs/inode.c @@ -43,7 +43,7 @@ enum { BIO_MAX_PAGES_KMALLOC = PAGE_SIZE / sizeof(struct page *), }; -unsigned exofs_max_io_pages(struct exofs_layout *layout, +unsigned exofs_max_io_pages(struct ore_layout *layout, unsigned expected_pages) { unsigned pages = min_t(unsigned, expected_pages, MAX_PAGES_KMALLOC); @@ -58,7 +58,7 @@ struct page_collect { struct exofs_sb_info *sbi; struct inode *inode; unsigned expected_pages; - struct exofs_io_state *ios; + struct ore_io_state *ios; struct page **pages; unsigned alloc_pages; @@ -133,7 +133,7 @@ static void pcol_free(struct page_collect *pcol) pcol->pages = NULL; if (pcol->ios) { - exofs_put_io_state(pcol->ios); + ore_put_io_state(pcol->ios); pcol->ios = NULL; } } @@ -193,7 +193,7 @@ static int __readpages_done(struct page_collect *pcol) u64 resid; u64 good_bytes; u64 length = 0; - int ret = exofs_check_io(pcol->ios, &resid); + int ret = ore_check_io(pcol->ios, &resid); if (likely(!ret)) good_bytes = pcol->length; @@ -234,7 +234,7 @@ static int __readpages_done(struct page_collect *pcol) } /* callback of async reads */ -static void readpages_done(struct exofs_io_state *ios, void *p) +static void readpages_done(struct ore_io_state *ios, void *p) { struct page_collect *pcol = p; @@ -262,7 +262,7 @@ static void _unlock_pcol_pages(struct page_collect *pcol, int ret, int rw) static int read_exec(struct page_collect *pcol) { struct exofs_i_info *oi = exofs_i(pcol->inode); - struct exofs_io_state *ios; + struct ore_io_state *ios; struct page_collect *pcol_copy = NULL; int ret; @@ -270,7 +270,7 @@ static int read_exec(struct page_collect *pcol) return 0; if (!pcol->ios) { - int ret = exofs_get_rw_state(&pcol->sbi->layout, &oi->comps, true, + int ret = ore_get_rw_state(&pcol->sbi->layout, &oi->comps, true, pcol->pg_first << PAGE_CACHE_SHIFT, pcol->length, &pcol->ios); @@ -283,7 +283,7 @@ static int read_exec(struct page_collect *pcol) ios->nr_pages = pcol->nr_pages; if (pcol->read_4_write) { - exofs_sbi_read(pcol->ios); + ore_read(pcol->ios); return __readpages_done(pcol); } @@ -296,7 +296,7 @@ static int read_exec(struct page_collect *pcol) *pcol_copy = *pcol; ios->done = readpages_done; ios->private = pcol_copy; - ret = exofs_sbi_read(ios); + ret = ore_read(ios); if (unlikely(ret)) goto err; @@ -458,14 +458,14 @@ static int exofs_readpage(struct file *file, struct page *page) } /* Callback for osd_write. All writes are asynchronous */ -static void writepages_done(struct exofs_io_state *ios, void *p) +static void writepages_done(struct ore_io_state *ios, void *p) { struct page_collect *pcol = p; int i; u64 resid; u64 good_bytes; u64 length = 0; - int ret = exofs_check_io(ios, &resid); + int ret = ore_check_io(ios, &resid); atomic_dec(&pcol->sbi->s_curr_pending); @@ -508,7 +508,7 @@ static void writepages_done(struct exofs_io_state *ios, void *p) static int write_exec(struct page_collect *pcol) { struct exofs_i_info *oi = exofs_i(pcol->inode); - struct exofs_io_state *ios; + struct ore_io_state *ios; struct page_collect *pcol_copy = NULL; int ret; @@ -516,7 +516,7 @@ static int write_exec(struct page_collect *pcol) return 0; BUG_ON(pcol->ios); - ret = exofs_get_rw_state(&pcol->sbi->layout, &oi->comps, false, + ret = ore_get_rw_state(&pcol->sbi->layout, &oi->comps, false, pcol->pg_first << PAGE_CACHE_SHIFT, pcol->length, &pcol->ios); @@ -538,9 +538,9 @@ static int write_exec(struct page_collect *pcol) ios->done = writepages_done; ios->private = pcol_copy; - ret = exofs_sbi_write(ios); + ret = ore_write(ios); if (unlikely(ret)) { - EXOFS_ERR("write_exec: exofs_sbi_write() Failed\n"); + EXOFS_ERR("write_exec: ore_write() Failed\n"); goto err; } @@ -860,7 +860,7 @@ static int _do_truncate(struct inode *inode, loff_t newsize) inode->i_mtime = inode->i_ctime = CURRENT_TIME; - ret = exofs_truncate(&sbi->layout, &oi->comps, (u64)newsize); + ret = ore_truncate(&sbi->layout, &oi->comps, (u64)newsize); if (likely(!ret)) truncate_setsize(inode, newsize); @@ -923,13 +923,13 @@ static int exofs_get_inode(struct super_block *sb, struct exofs_i_info *oi, [1] = g_attr_inode_file_layout, [2] = g_attr_inode_dir_layout, }; - struct exofs_io_state *ios; + struct ore_io_state *ios; struct exofs_on_disk_inode_layout *layout; int ret; - ret = exofs_get_io_state(&sbi->layout, &oi->comps, &ios); + ret = ore_get_io_state(&sbi->layout, &oi->comps, &ios); if (unlikely(ret)) { - EXOFS_ERR("%s: exofs_get_io_state failed.\n", __func__); + EXOFS_ERR("%s: ore_get_io_state failed.\n", __func__); return ret; } @@ -939,7 +939,7 @@ static int exofs_get_inode(struct super_block *sb, struct exofs_i_info *oi, ios->in_attr = attrs; ios->in_attr_len = ARRAY_SIZE(attrs); - ret = exofs_sbi_read(ios); + ret = ore_read(ios); if (unlikely(ret)) { EXOFS_ERR("object(0x%llx) corrupted, return empty file=>%d\n", _LLU(oi->one_comp.obj.id), ret); @@ -992,7 +992,7 @@ static int exofs_get_inode(struct super_block *sb, struct exofs_i_info *oi, } out: - exofs_put_io_state(ios); + ore_put_io_state(ios); return ret; } @@ -1111,15 +1111,15 @@ int __exofs_wait_obj_created(struct exofs_i_info *oi) * set the obj_created flag so that other methods know that the object exists on * the OSD. */ -static void create_done(struct exofs_io_state *ios, void *p) +static void create_done(struct ore_io_state *ios, void *p) { struct inode *inode = p; struct exofs_i_info *oi = exofs_i(inode); struct exofs_sb_info *sbi = inode->i_sb->s_fs_info; int ret; - ret = exofs_check_io(ios, NULL); - exofs_put_io_state(ios); + ret = ore_check_io(ios, NULL); + ore_put_io_state(ios); atomic_dec(&sbi->s_curr_pending); @@ -1149,7 +1149,7 @@ struct inode *exofs_new_inode(struct inode *dir, int mode) struct exofs_sb_info *sbi = sb->s_fs_info; struct inode *inode; struct exofs_i_info *oi; - struct exofs_io_state *ios; + struct ore_io_state *ios; int ret; inode = new_inode(sb); @@ -1178,18 +1178,18 @@ struct inode *exofs_new_inode(struct inode *dir, int mode) mark_inode_dirty(inode); - ret = exofs_get_io_state(&sbi->layout, &oi->comps, &ios); + ret = ore_get_io_state(&sbi->layout, &oi->comps, &ios); if (unlikely(ret)) { - EXOFS_ERR("exofs_new_inode: exofs_get_io_state failed\n"); + EXOFS_ERR("exofs_new_inode: ore_get_io_state failed\n"); return ERR_PTR(ret); } ios->done = create_done; ios->private = inode; - ret = exofs_sbi_create(ios); + ret = ore_create(ios); if (ret) { - exofs_put_io_state(ios); + ore_put_io_state(ios); return ERR_PTR(ret); } atomic_inc(&sbi->s_curr_pending); @@ -1208,11 +1208,11 @@ struct updatei_args { /* * Callback function from exofs_update_inode(). */ -static void updatei_done(struct exofs_io_state *ios, void *p) +static void updatei_done(struct ore_io_state *ios, void *p) { struct updatei_args *args = p; - exofs_put_io_state(ios); + ore_put_io_state(ios); atomic_dec(&args->sbi->s_curr_pending); @@ -1228,7 +1228,7 @@ static int exofs_update_inode(struct inode *inode, int do_sync) struct exofs_i_info *oi = exofs_i(inode); struct super_block *sb = inode->i_sb; struct exofs_sb_info *sbi = sb->s_fs_info; - struct exofs_io_state *ios; + struct ore_io_state *ios; struct osd_attr attr; struct exofs_fcb *fcb; struct updatei_args *args; @@ -1267,9 +1267,9 @@ static int exofs_update_inode(struct inode *inode, int do_sync) } else memcpy(fcb->i_data, oi->i_data, sizeof(fcb->i_data)); - ret = exofs_get_io_state(&sbi->layout, &oi->comps, &ios); + ret = ore_get_io_state(&sbi->layout, &oi->comps, &ios); if (unlikely(ret)) { - EXOFS_ERR("%s: exofs_get_io_state failed.\n", __func__); + EXOFS_ERR("%s: ore_get_io_state failed.\n", __func__); goto free_args; } @@ -1286,13 +1286,13 @@ static int exofs_update_inode(struct inode *inode, int do_sync) ios->private = args; } - ret = exofs_sbi_write(ios); + ret = ore_write(ios); if (!do_sync && !ret) { atomic_inc(&sbi->s_curr_pending); goto out; /* deallocation in updatei_done */ } - exofs_put_io_state(ios); + ore_put_io_state(ios); free_args: kfree(args); out: @@ -1311,11 +1311,11 @@ int exofs_write_inode(struct inode *inode, struct writeback_control *wbc) * Callback function from exofs_delete_inode() - don't have much cleaning up to * do. */ -static void delete_done(struct exofs_io_state *ios, void *p) +static void delete_done(struct ore_io_state *ios, void *p) { struct exofs_sb_info *sbi = p; - exofs_put_io_state(ios); + ore_put_io_state(ios); atomic_dec(&sbi->s_curr_pending); } @@ -1330,7 +1330,7 @@ void exofs_evict_inode(struct inode *inode) struct exofs_i_info *oi = exofs_i(inode); struct super_block *sb = inode->i_sb; struct exofs_sb_info *sbi = sb->s_fs_info; - struct exofs_io_state *ios; + struct ore_io_state *ios; int ret; truncate_inode_pages(&inode->i_data, 0); @@ -1350,19 +1350,19 @@ void exofs_evict_inode(struct inode *inode) /* ignore the error, attempt a remove anyway */ /* Now Remove the OSD objects */ - ret = exofs_get_io_state(&sbi->layout, &oi->comps, &ios); + ret = ore_get_io_state(&sbi->layout, &oi->comps, &ios); if (unlikely(ret)) { - EXOFS_ERR("%s: exofs_get_io_state failed\n", __func__); + EXOFS_ERR("%s: ore_get_io_state failed\n", __func__); return; } ios->done = delete_done; ios->private = sbi; - ret = exofs_sbi_remove(ios); + ret = ore_remove(ios); if (ret) { - EXOFS_ERR("%s: exofs_sbi_remove failed\n", __func__); - exofs_put_io_state(ios); + EXOFS_ERR("%s: ore_remove failed\n", __func__); + ore_put_io_state(ios); return; } atomic_inc(&sbi->s_curr_pending); diff --git a/fs/exofs/ios.c b/fs/exofs/ore.c similarity index 79% rename from fs/exofs/ios.c rename to fs/exofs/ore.c index f9d5c5a823fb..3a000084cd21 100644 --- a/fs/exofs/ios.c +++ b/fs/exofs/ore.c @@ -23,43 +23,54 @@ */ #include -#include #include -#include "exofs.h" +#include -#define EXOFS_DBGMSG2(M...) do {} while (0) -/* #define EXOFS_DBGMSG2 EXOFS_DBGMSG */ +#define ORE_ERR(fmt, a...) printk(KERN_ERR "ore: " fmt, ##a) -static u8 *_ios_cred(struct exofs_io_state *ios, unsigned index) +#ifdef CONFIG_EXOFS_DEBUG +#define ORE_DBGMSG(fmt, a...) \ + printk(KERN_NOTICE "ore @%s:%d: " fmt, __func__, __LINE__, ##a) +#else +#define ORE_DBGMSG(fmt, a...) \ + do { if (0) printk(fmt, ##a); } while (0) +#endif + +/* u64 has problems with printk this will cast it to unsigned long long */ +#define _LLU(x) (unsigned long long)(x) + +#define ORE_DBGMSG2(M...) do {} while (0) +/* #define ORE_DBGMSG2 ORE_DBGMSG */ + +static u8 *_ios_cred(struct ore_io_state *ios, unsigned index) { return ios->comps->comps[index & ios->comps->single_comp].cred; } -static struct osd_obj_id *_ios_obj(struct exofs_io_state *ios, unsigned index) +static struct osd_obj_id *_ios_obj(struct ore_io_state *ios, unsigned index) { return &ios->comps->comps[index & ios->comps->single_comp].obj; } -static struct osd_dev *_ios_od(struct exofs_io_state *ios, unsigned index) +static struct osd_dev *_ios_od(struct ore_io_state *ios, unsigned index) { return ios->comps->ods[index]; } -int exofs_get_rw_state(struct exofs_layout *layout, - struct exofs_components *comps, - bool is_reading, u64 offset, u64 length, - struct exofs_io_state **pios) +int ore_get_rw_state(struct ore_layout *layout, struct ore_components *comps, + bool is_reading, u64 offset, u64 length, + struct ore_io_state **pios) { - struct exofs_io_state *ios; + struct ore_io_state *ios; /*TODO: Maybe use kmem_cach per sbi of size * exofs_io_state_size(layout->s_numdevs) */ - ios = kzalloc(exofs_io_state_size(comps->numdevs), GFP_KERNEL); + ios = kzalloc(ore_io_state_size(comps->numdevs), GFP_KERNEL); if (unlikely(!ios)) { - EXOFS_DBGMSG("Failed kzalloc bytes=%d\n", - exofs_io_state_size(comps->numdevs)); + ORE_DBGMSG("Failed kzalloc bytes=%d\n", + ore_io_state_size(comps->numdevs)); *pios = NULL; return -ENOMEM; } @@ -74,20 +85,19 @@ int exofs_get_rw_state(struct exofs_layout *layout, return 0; } -int exofs_get_io_state(struct exofs_layout *layout, - struct exofs_components *comps, - struct exofs_io_state **ios) +int ore_get_io_state(struct ore_layout *layout, struct ore_components *comps, + struct ore_io_state **ios) { - return exofs_get_rw_state(layout, comps, true, 0, 0, ios); + return ore_get_rw_state(layout, comps, true, 0, 0, ios); } -void exofs_put_io_state(struct exofs_io_state *ios) +void ore_put_io_state(struct ore_io_state *ios) { if (ios) { unsigned i; for (i = 0; i < ios->numdevs; i++) { - struct exofs_per_dev_state *per_dev = &ios->per_dev[i]; + struct ore_per_dev_state *per_dev = &ios->per_dev[i]; if (per_dev->or) osd_end_request(per_dev->or); @@ -99,7 +109,7 @@ void exofs_put_io_state(struct exofs_io_state *ios) } } -static void _sync_done(struct exofs_io_state *ios, void *p) +static void _sync_done(struct ore_io_state *ios, void *p) { struct completion *waiting = p; @@ -108,20 +118,20 @@ static void _sync_done(struct exofs_io_state *ios, void *p) static void _last_io(struct kref *kref) { - struct exofs_io_state *ios = container_of( - kref, struct exofs_io_state, kref); + struct ore_io_state *ios = container_of( + kref, struct ore_io_state, kref); ios->done(ios, ios->private); } static void _done_io(struct osd_request *or, void *p) { - struct exofs_io_state *ios = p; + struct ore_io_state *ios = p; kref_put(&ios->kref, _last_io); } -static int exofs_io_execute(struct exofs_io_state *ios) +static int ore_io_execute(struct ore_io_state *ios) { DECLARE_COMPLETION_ONSTACK(wait); bool sync = (ios->done == NULL); @@ -139,7 +149,7 @@ static int exofs_io_execute(struct exofs_io_state *ios) ret = osd_finalize_request(or, 0, _ios_cred(ios, i), NULL); if (unlikely(ret)) { - EXOFS_DBGMSG("Failed to osd_finalize_request() => %d\n", + ORE_DBGMSG("Failed to osd_finalize_request() => %d\n", ret); return ret; } @@ -161,7 +171,7 @@ static int exofs_io_execute(struct exofs_io_state *ios) if (sync) { wait_for_completion(&wait); - ret = exofs_check_io(ios, NULL); + ret = ore_check_io(ios, NULL); } return ret; } @@ -181,7 +191,7 @@ static void _clear_bio(struct bio *bio) } } -int exofs_check_io(struct exofs_io_state *ios, u64 *resid) +int ore_check_io(struct ore_io_state *ios, u64 *resid) { enum osd_err_priority acumulated_osd_err = 0; int acumulated_lin_err = 0; @@ -202,7 +212,7 @@ int exofs_check_io(struct exofs_io_state *ios, u64 *resid) if (OSD_ERR_PRI_CLEAR_PAGES == osi.osd_err_pri) { /* start read offset passed endof file */ _clear_bio(ios->per_dev[i].bio); - EXOFS_DBGMSG("start read offset passed end of file " + ORE_DBGMSG("start read offset passed end of file " "offset=0x%llx, length=0x%llx\n", _LLU(ios->per_dev[i].offset), _LLU(ios->per_dev[i].length)); @@ -277,7 +287,7 @@ struct _striping_info { unsigned unit_off; }; -static void _calc_stripe_info(struct exofs_layout *layout, u64 file_offset, +static void _calc_stripe_info(struct ore_layout *layout, u64 file_offset, struct _striping_info *si) { u32 stripe_unit = layout->stripe_unit; @@ -312,8 +322,8 @@ static void _calc_stripe_info(struct exofs_layout *layout, u64 file_offset, si->M = M; } -static int _add_stripe_unit(struct exofs_io_state *ios, unsigned *cur_pg, - unsigned pgbase, struct exofs_per_dev_state *per_dev, +static int _add_stripe_unit(struct ore_io_state *ios, unsigned *cur_pg, + unsigned pgbase, struct ore_per_dev_state *per_dev, int cur_len) { unsigned pg = *cur_pg; @@ -330,7 +340,7 @@ static int _add_stripe_unit(struct exofs_io_state *ios, unsigned *cur_pg, per_dev->bio = bio_kmalloc(GFP_KERNEL, bio_size); if (unlikely(!per_dev->bio)) { - EXOFS_DBGMSG("Failed to allocate BIO size=%u\n", + ORE_DBGMSG("Failed to allocate BIO size=%u\n", bio_size); return -ENOMEM; } @@ -356,7 +366,7 @@ static int _add_stripe_unit(struct exofs_io_state *ios, unsigned *cur_pg, return 0; } -static int _prepare_one_group(struct exofs_io_state *ios, u64 length, +static int _prepare_one_group(struct ore_io_state *ios, u64 length, struct _striping_info *si) { unsigned stripe_unit = ios->layout->stripe_unit; @@ -369,7 +379,7 @@ static int _prepare_one_group(struct exofs_io_state *ios, u64 length, int ret = 0; while (length) { - struct exofs_per_dev_state *per_dev = &ios->per_dev[dev]; + struct ore_per_dev_state *per_dev = &ios->per_dev[dev]; unsigned cur_len, page_off = 0; if (!per_dev->length) { @@ -412,7 +422,7 @@ out: return ret; } -static int _prepare_for_striping(struct exofs_io_state *ios) +static int _prepare_for_striping(struct ore_io_state *ios) { u64 length = ios->length; u64 offset = ios->offset; @@ -421,7 +431,7 @@ static int _prepare_for_striping(struct exofs_io_state *ios) if (!ios->pages) { if (ios->kern_buff) { - struct exofs_per_dev_state *per_dev = &ios->per_dev[0]; + struct ore_per_dev_state *per_dev = &ios->per_dev[0]; _calc_stripe_info(ios->layout, ios->offset, &si); per_dev->offset = si.obj_offset; @@ -454,7 +464,7 @@ out: return ret; } -int exofs_sbi_create(struct exofs_io_state *ios) +int ore_create(struct ore_io_state *ios) { int i, ret; @@ -463,7 +473,7 @@ int exofs_sbi_create(struct exofs_io_state *ios) or = osd_start_request(_ios_od(ios, i), GFP_KERNEL); if (unlikely(!or)) { - EXOFS_ERR("%s: osd_start_request failed\n", __func__); + ORE_ERR("%s: osd_start_request failed\n", __func__); ret = -ENOMEM; goto out; } @@ -472,13 +482,13 @@ int exofs_sbi_create(struct exofs_io_state *ios) osd_req_create_object(or, _ios_obj(ios, i)); } - ret = exofs_io_execute(ios); + ret = ore_io_execute(ios); out: return ret; } -int exofs_sbi_remove(struct exofs_io_state *ios) +int ore_remove(struct ore_io_state *ios) { int i, ret; @@ -487,7 +497,7 @@ int exofs_sbi_remove(struct exofs_io_state *ios) or = osd_start_request(_ios_od(ios, i), GFP_KERNEL); if (unlikely(!or)) { - EXOFS_ERR("%s: osd_start_request failed\n", __func__); + ORE_ERR("%s: osd_start_request failed\n", __func__); ret = -ENOMEM; goto out; } @@ -496,15 +506,15 @@ int exofs_sbi_remove(struct exofs_io_state *ios) osd_req_remove_object(or, _ios_obj(ios, i)); } - ret = exofs_io_execute(ios); + ret = ore_io_execute(ios); out: return ret; } -static int _sbi_write_mirror(struct exofs_io_state *ios, int cur_comp) +static int _write_mirror(struct ore_io_state *ios, int cur_comp) { - struct exofs_per_dev_state *master_dev = &ios->per_dev[cur_comp]; + struct ore_per_dev_state *master_dev = &ios->per_dev[cur_comp]; unsigned dev = ios->per_dev[cur_comp].dev; unsigned last_comp = cur_comp + ios->layout->mirrors_p1; int ret = 0; @@ -513,12 +523,12 @@ static int _sbi_write_mirror(struct exofs_io_state *ios, int cur_comp) return 0; /* Just an empty slot */ for (; cur_comp < last_comp; ++cur_comp, ++dev) { - struct exofs_per_dev_state *per_dev = &ios->per_dev[cur_comp]; + struct ore_per_dev_state *per_dev = &ios->per_dev[cur_comp]; struct osd_request *or; or = osd_start_request(_ios_od(ios, dev), GFP_KERNEL); if (unlikely(!or)) { - EXOFS_ERR("%s: osd_start_request failed\n", __func__); + ORE_ERR("%s: osd_start_request failed\n", __func__); ret = -ENOMEM; goto out; } @@ -532,7 +542,7 @@ static int _sbi_write_mirror(struct exofs_io_state *ios, int cur_comp) bio = bio_kmalloc(GFP_KERNEL, master_dev->bio->bi_max_vecs); if (unlikely(!bio)) { - EXOFS_DBGMSG( + ORE_DBGMSG( "Failed to allocate BIO size=%u\n", master_dev->bio->bi_max_vecs); ret = -ENOMEM; @@ -553,7 +563,7 @@ static int _sbi_write_mirror(struct exofs_io_state *ios, int cur_comp) osd_req_write(or, _ios_obj(ios, dev), per_dev->offset, bio, per_dev->length); - EXOFS_DBGMSG("write(0x%llx) offset=0x%llx " + ORE_DBGMSG("write(0x%llx) offset=0x%llx " "length=0x%llx dev=%d\n", _LLU(_ios_obj(ios, dev)->id), _LLU(per_dev->offset), @@ -564,14 +574,14 @@ static int _sbi_write_mirror(struct exofs_io_state *ios, int cur_comp) ios->kern_buff, ios->length); if (unlikely(ret)) goto out; - EXOFS_DBGMSG2("write_kern(0x%llx) offset=0x%llx " + ORE_DBGMSG2("write_kern(0x%llx) offset=0x%llx " "length=0x%llx dev=%d\n", _LLU(_ios_obj(ios, dev)->id), _LLU(per_dev->offset), _LLU(ios->length), dev); } else { osd_req_set_attributes(or, _ios_obj(ios, dev)); - EXOFS_DBGMSG2("obj(0x%llx) set_attributes=%d dev=%d\n", + ORE_DBGMSG2("obj(0x%llx) set_attributes=%d dev=%d\n", _LLU(_ios_obj(ios, dev)->id), ios->out_attr_len, dev); } @@ -589,7 +599,7 @@ out: return ret; } -int exofs_sbi_write(struct exofs_io_state *ios) +int ore_write(struct ore_io_state *ios) { int i; int ret; @@ -599,19 +609,19 @@ int exofs_sbi_write(struct exofs_io_state *ios) return ret; for (i = 0; i < ios->numdevs; i += ios->layout->mirrors_p1) { - ret = _sbi_write_mirror(ios, i); + ret = _write_mirror(ios, i); if (unlikely(ret)) return ret; } - ret = exofs_io_execute(ios); + ret = ore_io_execute(ios); return ret; } -static int _sbi_read_mirror(struct exofs_io_state *ios, unsigned cur_comp) +static int _read_mirror(struct ore_io_state *ios, unsigned cur_comp) { struct osd_request *or; - struct exofs_per_dev_state *per_dev = &ios->per_dev[cur_comp]; + struct ore_per_dev_state *per_dev = &ios->per_dev[cur_comp]; struct osd_obj_id *obj = _ios_obj(ios, cur_comp); unsigned first_dev = (unsigned)obj->id; @@ -621,7 +631,7 @@ static int _sbi_read_mirror(struct exofs_io_state *ios, unsigned cur_comp) first_dev = per_dev->dev + first_dev % ios->layout->mirrors_p1; or = osd_start_request(_ios_od(ios, first_dev), GFP_KERNEL); if (unlikely(!or)) { - EXOFS_ERR("%s: osd_start_request failed\n", __func__); + ORE_ERR("%s: osd_start_request failed\n", __func__); return -ENOMEM; } per_dev->or = or; @@ -629,14 +639,14 @@ static int _sbi_read_mirror(struct exofs_io_state *ios, unsigned cur_comp) if (ios->pages) { osd_req_read(or, obj, per_dev->offset, per_dev->bio, per_dev->length); - EXOFS_DBGMSG("read(0x%llx) offset=0x%llx length=0x%llx" + ORE_DBGMSG("read(0x%llx) offset=0x%llx length=0x%llx" " dev=%d\n", _LLU(obj->id), _LLU(per_dev->offset), _LLU(per_dev->length), first_dev); } else if (ios->kern_buff) { int ret = osd_req_read_kern(or, obj, per_dev->offset, ios->kern_buff, ios->length); - EXOFS_DBGMSG2("read_kern(0x%llx) offset=0x%llx " + ORE_DBGMSG2("read_kern(0x%llx) offset=0x%llx " "length=0x%llx dev=%d ret=>%d\n", _LLU(obj->id), _LLU(per_dev->offset), _LLU(ios->length), first_dev, ret); @@ -644,7 +654,7 @@ static int _sbi_read_mirror(struct exofs_io_state *ios, unsigned cur_comp) return ret; } else { osd_req_get_attributes(or, obj); - EXOFS_DBGMSG2("obj(0x%llx) get_attributes=%d dev=%d\n", + ORE_DBGMSG2("obj(0x%llx) get_attributes=%d dev=%d\n", _LLU(obj->id), ios->in_attr_len, first_dev); } @@ -657,7 +667,7 @@ static int _sbi_read_mirror(struct exofs_io_state *ios, unsigned cur_comp) return 0; } -int exofs_sbi_read(struct exofs_io_state *ios) +int ore_read(struct ore_io_state *ios) { int i; int ret; @@ -667,16 +677,16 @@ int exofs_sbi_read(struct exofs_io_state *ios) return ret; for (i = 0; i < ios->numdevs; i += ios->layout->mirrors_p1) { - ret = _sbi_read_mirror(ios, i); + ret = _read_mirror(ios, i); if (unlikely(ret)) return ret; } - ret = exofs_io_execute(ios); + ret = ore_io_execute(ios); return ret; } -int extract_attr_from_ios(struct exofs_io_state *ios, struct osd_attr *attr) +int extract_attr_from_ios(struct ore_io_state *ios, struct osd_attr *attr) { struct osd_attr cur_attr = {.attr_page = 0}; /* start with zeros */ void *iter = NULL; @@ -697,18 +707,18 @@ int extract_attr_from_ios(struct exofs_io_state *ios, struct osd_attr *attr) return -EIO; } -static int _truncate_mirrors(struct exofs_io_state *ios, unsigned cur_comp, +static int _truncate_mirrors(struct ore_io_state *ios, unsigned cur_comp, struct osd_attr *attr) { int last_comp = cur_comp + ios->layout->mirrors_p1; for (; cur_comp < last_comp; ++cur_comp) { - struct exofs_per_dev_state *per_dev = &ios->per_dev[cur_comp]; + struct ore_per_dev_state *per_dev = &ios->per_dev[cur_comp]; struct osd_request *or; or = osd_start_request(_ios_od(ios, cur_comp), GFP_KERNEL); if (unlikely(!or)) { - EXOFS_ERR("%s: osd_start_request failed\n", __func__); + ORE_ERR("%s: osd_start_request failed\n", __func__); return -ENOMEM; } per_dev->or = or; @@ -730,7 +740,7 @@ struct _trunc_info { unsigned max_devs; }; -void _calc_trunk_info(struct exofs_layout *layout, u64 file_offset, +void _calc_trunk_info(struct ore_layout *layout, u64 file_offset, struct _trunc_info *ti) { unsigned stripe_unit = layout->stripe_unit; @@ -745,10 +755,10 @@ void _calc_trunk_info(struct exofs_layout *layout, u64 file_offset, ti->max_devs = layout->group_width * layout->group_count; } -int exofs_truncate(struct exofs_layout *layout, struct exofs_components *comps, +int ore_truncate(struct ore_layout *layout, struct ore_components *comps, u64 size) { - struct exofs_io_state *ios; + struct ore_io_state *ios; struct exofs_trunc_attr { struct osd_attr attr; __be64 newsize; @@ -756,7 +766,7 @@ int exofs_truncate(struct exofs_layout *layout, struct exofs_components *comps, struct _trunc_info ti; int i, ret; - ret = exofs_get_io_state(layout, comps, &ios); + ret = ore_get_io_state(layout, comps, &ios); if (unlikely(ret)) return ret; @@ -791,18 +801,18 @@ int exofs_truncate(struct exofs_layout *layout, struct exofs_components *comps, size_attr->attr = g_attr_logical_length; size_attr->attr.val_ptr = &size_attr->newsize; - EXOFS_DBGMSG("trunc(0x%llx) obj_offset=0x%llx dev=%d\n", + ORE_DBGMSG("trunc(0x%llx) obj_offset=0x%llx dev=%d\n", _LLU(comps->comps->obj.id), _LLU(obj_size), i); ret = _truncate_mirrors(ios, i * ios->layout->mirrors_p1, &size_attr->attr); if (unlikely(ret)) goto out; } - ret = exofs_io_execute(ios); + ret = ore_io_execute(ios); out: kfree(size_attrs); - exofs_put_io_state(ios); + ore_put_io_state(ios); return ret; } diff --git a/fs/exofs/super.c b/fs/exofs/super.c index 4d6877967c87..274894053b02 100644 --- a/fs/exofs/super.c +++ b/fs/exofs/super.c @@ -263,19 +263,19 @@ static int __sbi_read_stats(struct exofs_sb_info *sbi) struct osd_attr attrs[] = { [0] = g_attr_sb_stats, }; - struct exofs_io_state *ios; + struct ore_io_state *ios; int ret; - ret = exofs_get_io_state(&sbi->layout, &sbi->comps, &ios); + ret = ore_get_io_state(&sbi->layout, &sbi->comps, &ios); if (unlikely(ret)) { - EXOFS_ERR("%s: exofs_get_io_state failed.\n", __func__); + EXOFS_ERR("%s: ore_get_io_state failed.\n", __func__); return ret; } ios->in_attr = attrs; ios->in_attr_len = ARRAY_SIZE(attrs); - ret = exofs_sbi_read(ios); + ret = ore_read(ios); if (unlikely(ret)) { EXOFS_ERR("Error reading super_block stats => %d\n", ret); goto out; @@ -302,13 +302,13 @@ static int __sbi_read_stats(struct exofs_sb_info *sbi) } out: - exofs_put_io_state(ios); + ore_put_io_state(ios); return ret; } -static void stats_done(struct exofs_io_state *ios, void *p) +static void stats_done(struct ore_io_state *ios, void *p) { - exofs_put_io_state(ios); + ore_put_io_state(ios); /* Good thanks nothing to do anymore */ } @@ -318,12 +318,12 @@ int exofs_sbi_write_stats(struct exofs_sb_info *sbi) struct osd_attr attrs[] = { [0] = g_attr_sb_stats, }; - struct exofs_io_state *ios; + struct ore_io_state *ios; int ret; - ret = exofs_get_io_state(&sbi->layout, &sbi->comps, &ios); + ret = ore_get_io_state(&sbi->layout, &sbi->comps, &ios); if (unlikely(ret)) { - EXOFS_ERR("%s: exofs_get_io_state failed.\n", __func__); + EXOFS_ERR("%s: ore_get_io_state failed.\n", __func__); return ret; } @@ -337,10 +337,10 @@ int exofs_sbi_write_stats(struct exofs_sb_info *sbi) ios->out_attr = attrs; ios->out_attr_len = ARRAY_SIZE(attrs); - ret = exofs_sbi_write(ios); + ret = ore_write(ios); if (unlikely(ret)) { - EXOFS_ERR("%s: exofs_sbi_write failed.\n", __func__); - exofs_put_io_state(ios); + EXOFS_ERR("%s: ore_write failed.\n", __func__); + ore_put_io_state(ios); } return ret; @@ -359,9 +359,9 @@ int exofs_sync_fs(struct super_block *sb, int wait) { struct exofs_sb_info *sbi; struct exofs_fscb *fscb; - struct exofs_comp one_comp; - struct exofs_components comps; - struct exofs_io_state *ios; + struct ore_comp one_comp; + struct ore_components comps; + struct ore_io_state *ios; int ret = -ENOMEM; fscb = kmalloc(sizeof(*fscb), GFP_KERNEL); @@ -380,7 +380,7 @@ int exofs_sync_fs(struct super_block *sb, int wait) exofs_init_comps(&comps, &one_comp, sbi, EXOFS_SUPER_ID); - ret = exofs_get_io_state(&sbi->layout, &comps, &ios); + ret = ore_get_io_state(&sbi->layout, &comps, &ios); if (unlikely(ret)) goto out; @@ -397,9 +397,9 @@ int exofs_sync_fs(struct super_block *sb, int wait) ios->offset = 0; ios->kern_buff = fscb; - ret = exofs_sbi_write(ios); + ret = ore_write(ios); if (unlikely(ret)) - EXOFS_ERR("%s: exofs_sbi_write failed.\n", __func__); + EXOFS_ERR("%s: ore_write failed.\n", __func__); else sb->s_dirt = 0; @@ -407,7 +407,7 @@ int exofs_sync_fs(struct super_block *sb, int wait) unlock_super(sb); out: EXOFS_DBGMSG("s_nextid=0x%llx ret=%d\n", _LLU(sbi->s_nextid), ret); - exofs_put_io_state(ios); + ore_put_io_state(ios); kfree(fscb); return ret; } @@ -562,7 +562,7 @@ static int _read_and_match_data_map(struct exofs_sb_info *sbi, unsigned numdevs, return 0; } -static unsigned __ra_pages(struct exofs_layout *layout) +static unsigned __ra_pages(struct ore_layout *layout) { const unsigned _MIN_RA = 32; /* min 128K read-ahead */ unsigned ra_pages = layout->group_width * layout->stripe_unit / @@ -609,7 +609,7 @@ static int exofs_read_lookup_dev_table(struct exofs_sb_info *sbi, struct osd_dev *fscb_od, unsigned table_count) { - struct exofs_comp comp; + struct ore_comp comp; struct exofs_device_table *dt; unsigned table_bytes = table_count * sizeof(dt->dt_dev_table[0]) + sizeof(*dt); @@ -747,7 +747,7 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent) struct exofs_sb_info *sbi; /*extended info */ struct osd_dev *od; /* Master device */ struct exofs_fscb fscb; /*on-disk superblock info */ - struct exofs_comp comp; + struct ore_comp comp; unsigned table_count; int ret; @@ -913,7 +913,7 @@ static int exofs_statfs(struct dentry *dentry, struct kstatfs *buf) { struct super_block *sb = dentry->d_sb; struct exofs_sb_info *sbi = sb->s_fs_info; - struct exofs_io_state *ios; + struct ore_io_state *ios; struct osd_attr attrs[] = { ATTR_DEF(OSD_APAGE_PARTITION_QUOTAS, OSD_ATTR_PQ_CAPACITY_QUOTA, sizeof(__be64)), @@ -924,16 +924,16 @@ static int exofs_statfs(struct dentry *dentry, struct kstatfs *buf) uint64_t used = ULLONG_MAX; int ret; - ret = exofs_get_io_state(&sbi->layout, &sbi->comps, &ios); + ret = ore_get_io_state(&sbi->layout, &sbi->comps, &ios); if (ret) { - EXOFS_DBGMSG("exofs_get_io_state failed.\n"); + EXOFS_DBGMSG("ore_get_io_state failed.\n"); return ret; } ios->in_attr = attrs; ios->in_attr_len = ARRAY_SIZE(attrs); - ret = exofs_sbi_read(ios); + ret = ore_read(ios); if (unlikely(ret)) goto out; @@ -962,7 +962,7 @@ static int exofs_statfs(struct dentry *dentry, struct kstatfs *buf) buf->f_namelen = EXOFS_NAME_LEN; out: - exofs_put_io_state(ios); + ore_put_io_state(ios); return ret; } diff --git a/include/scsi/osd_ore.h b/include/scsi/osd_ore.h new file mode 100644 index 000000000000..c5c5e008e6de --- /dev/null +++ b/include/scsi/osd_ore.h @@ -0,0 +1,125 @@ +/* + * Copyright (C) 2011 + * Boaz Harrosh + * + * Public Declarations of the ORE API + * + * This file is part of the ORE (Object Raid Engine) library. + * + * ORE 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. (GPL v2) + * + * ORE 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 the ORE; if not, write to the Free Software + * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __ORE_H__ +#define __ORE_H__ + +#include +#include +#include +#include + +struct ore_comp { + struct osd_obj_id obj; + u8 cred[OSD_CAP_LEN]; +}; + +struct ore_layout { + /* Our way of looking at the data_map */ + unsigned stripe_unit; + unsigned mirrors_p1; + + unsigned group_width; + u64 group_depth; + unsigned group_count; +}; + +struct ore_components { + unsigned numdevs; /* Num of devices in array */ + /* If @single_comp == EC_SINGLE_COMP, @comps points to a single + * component. else there are @numdevs components + */ + enum EC_COMP_USAGE { + EC_SINGLE_COMP = 0, EC_MULTPLE_COMPS = 0xffffffff + } single_comp; + struct ore_comp *comps; + struct osd_dev **ods; /* osd_dev array */ +}; + +struct ore_io_state; +typedef void (*ore_io_done_fn)(struct ore_io_state *ios, void *private); + +struct ore_io_state { + struct kref kref; + + void *private; + ore_io_done_fn done; + + struct ore_layout *layout; + struct ore_components *comps; + + /* Global read/write IO*/ + loff_t offset; + unsigned long length; + void *kern_buff; + + struct page **pages; + unsigned nr_pages; + unsigned pgbase; + unsigned pages_consumed; + + /* Attributes */ + unsigned in_attr_len; + struct osd_attr *in_attr; + unsigned out_attr_len; + struct osd_attr *out_attr; + + bool reading; + + /* Variable array of size numdevs */ + unsigned numdevs; + struct ore_per_dev_state { + struct osd_request *or; + struct bio *bio; + loff_t offset; + unsigned length; + unsigned dev; + } per_dev[]; +}; + +static inline unsigned ore_io_state_size(unsigned numdevs) +{ + return sizeof(struct ore_io_state) + + sizeof(struct ore_per_dev_state) * numdevs; +} + +/* ore.c */ +int ore_get_rw_state(struct ore_layout *layout, struct ore_components *comps, + bool is_reading, u64 offset, u64 length, + struct ore_io_state **ios); +int ore_get_io_state(struct ore_layout *layout, struct ore_components *comps, + struct ore_io_state **ios); +void ore_put_io_state(struct ore_io_state *ios); + +int ore_check_io(struct ore_io_state *ios, u64 *resid); + +int ore_create(struct ore_io_state *ios); +int ore_remove(struct ore_io_state *ios); +int ore_write(struct ore_io_state *ios); +int ore_read(struct ore_io_state *ios); +int ore_truncate(struct ore_layout *layout, struct ore_components *comps, + u64 size); + +int extract_attr_from_ios(struct ore_io_state *ios, struct osd_attr *attr); + +extern const struct osd_attr g_attr_logical_length; + +#endif From cf283ade08c454e884394a4720f22421dd33a715 Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Sat, 6 Aug 2011 19:22:06 -0700 Subject: [PATCH 430/439] ore: Make ore its own module Export everything from ore need exporting. Change Kbuild and Kconfig to build ore.ko as an independent module. Import ore from exofs Signed-off-by: Boaz Harrosh --- fs/exofs/Kbuild | 5 ++++- fs/exofs/Kconfig | 4 ++++ fs/exofs/ore.c | 15 +++++++++++++++ 3 files changed, 23 insertions(+), 1 deletion(-) diff --git a/fs/exofs/Kbuild b/fs/exofs/Kbuild index c1af2217b087..c5a5855a6c44 100644 --- a/fs/exofs/Kbuild +++ b/fs/exofs/Kbuild @@ -12,5 +12,8 @@ # Kbuild - Gets included from the Kernels Makefile and build system # -exofs-y := ore.o inode.o file.o symlink.o namei.o dir.o super.o +# ore module library +obj-$(CONFIG_ORE) += ore.o + +exofs-y := inode.o file.o symlink.o namei.o dir.o super.o obj-$(CONFIG_EXOFS_FS) += exofs.o diff --git a/fs/exofs/Kconfig b/fs/exofs/Kconfig index 86194b2f799d..70bae4149291 100644 --- a/fs/exofs/Kconfig +++ b/fs/exofs/Kconfig @@ -1,6 +1,10 @@ +config ORE + tristate + config EXOFS_FS tristate "exofs: OSD based file system support" depends on SCSI_OSD_ULD + select ORE help EXOFS is a file system that uses an OSD storage device, as its backing storage. diff --git a/fs/exofs/ore.c b/fs/exofs/ore.c index 3a000084cd21..25305af88198 100644 --- a/fs/exofs/ore.c +++ b/fs/exofs/ore.c @@ -43,6 +43,10 @@ #define ORE_DBGMSG2(M...) do {} while (0) /* #define ORE_DBGMSG2 ORE_DBGMSG */ +MODULE_AUTHOR("Boaz Harrosh "); +MODULE_DESCRIPTION("Objects Raid Engine ore.ko"); +MODULE_LICENSE("GPL"); + static u8 *_ios_cred(struct ore_io_state *ios, unsigned index) { return ios->comps->comps[index & ios->comps->single_comp].cred; @@ -84,12 +88,14 @@ int ore_get_rw_state(struct ore_layout *layout, struct ore_components *comps, *pios = ios; return 0; } +EXPORT_SYMBOL(ore_get_rw_state); int ore_get_io_state(struct ore_layout *layout, struct ore_components *comps, struct ore_io_state **ios) { return ore_get_rw_state(layout, comps, true, 0, 0, ios); } +EXPORT_SYMBOL(ore_get_io_state); void ore_put_io_state(struct ore_io_state *ios) { @@ -108,6 +114,7 @@ void ore_put_io_state(struct ore_io_state *ios) kfree(ios); } } +EXPORT_SYMBOL(ore_put_io_state); static void _sync_done(struct ore_io_state *ios, void *p) { @@ -236,6 +243,7 @@ int ore_check_io(struct ore_io_state *ios, u64 *resid) return acumulated_lin_err; } +EXPORT_SYMBOL(ore_check_io); /* * L - logical offset into the file @@ -487,6 +495,7 @@ int ore_create(struct ore_io_state *ios) out: return ret; } +EXPORT_SYMBOL(ore_create); int ore_remove(struct ore_io_state *ios) { @@ -511,6 +520,7 @@ int ore_remove(struct ore_io_state *ios) out: return ret; } +EXPORT_SYMBOL(ore_remove); static int _write_mirror(struct ore_io_state *ios, int cur_comp) { @@ -617,6 +627,7 @@ int ore_write(struct ore_io_state *ios) ret = ore_io_execute(ios); return ret; } +EXPORT_SYMBOL(ore_write); static int _read_mirror(struct ore_io_state *ios, unsigned cur_comp) { @@ -685,6 +696,7 @@ int ore_read(struct ore_io_state *ios) ret = ore_io_execute(ios); return ret; } +EXPORT_SYMBOL(ore_read); int extract_attr_from_ios(struct ore_io_state *ios, struct osd_attr *attr) { @@ -706,6 +718,7 @@ int extract_attr_from_ios(struct ore_io_state *ios, struct osd_attr *attr) return -EIO; } +EXPORT_SYMBOL(extract_attr_from_ios); static int _truncate_mirrors(struct ore_io_state *ios, unsigned cur_comp, struct osd_attr *attr) @@ -815,6 +828,8 @@ out: ore_put_io_state(ios); return ret; } +EXPORT_SYMBOL(ore_truncate); const struct osd_attr g_attr_logical_length = ATTR_DEF( OSD_APAGE_OBJECT_INFORMATION, OSD_ATTR_OI_LOGICAL_LENGTH, 8); +EXPORT_SYMBOL(g_attr_logical_length); From 830c0f0edca67403d361fe976a25b17356c11f19 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sat, 6 Aug 2011 22:41:50 -0700 Subject: [PATCH 431/439] vfs: renumber DCACHE_xyz flags, remove some stale ones Gcc tends to generate better code with small integers, including the DCACHE_xyz flag tests - so move the common ones to be first in the list. Also just remove the unused DCACHE_INOTIFY_PARENT_WATCHED and DCACHE_AUTOFS_PENDING values, their users no longer exists in the source tree. And add a "unlikely()" to the DCACHE_OP_COMPARE test, since we want the common case to be a nice straight-line fall-through. Signed-off-by: Linus Torvalds --- fs/dcache.c | 2 +- include/linux/dcache.h | 30 +++++++++++++----------------- 2 files changed, 14 insertions(+), 18 deletions(-) diff --git a/fs/dcache.c b/fs/dcache.c index c83cae19161e..a88948b8bd17 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -1729,7 +1729,7 @@ seqretry: */ if (read_seqcount_retry(&dentry->d_seq, *seq)) goto seqretry; - if (parent->d_flags & DCACHE_OP_COMPARE) { + if (unlikely(parent->d_flags & DCACHE_OP_COMPARE)) { if (parent->d_op->d_compare(parent, *inode, dentry, i, tlen, tname, name)) diff --git a/include/linux/dcache.h b/include/linux/dcache.h index d37d2a793099..62157c03caf7 100644 --- a/include/linux/dcache.h +++ b/include/linux/dcache.h @@ -180,12 +180,12 @@ struct dentry_operations { */ /* d_flags entries */ -#define DCACHE_AUTOFS_PENDING 0x0001 /* autofs: "under construction" */ -#define DCACHE_NFSFS_RENAMED 0x0002 - /* this dentry has been "silly renamed" and has to be deleted on the last - * dput() */ +#define DCACHE_OP_HASH 0x0001 +#define DCACHE_OP_COMPARE 0x0002 +#define DCACHE_OP_REVALIDATE 0x0004 +#define DCACHE_OP_DELETE 0x0008 -#define DCACHE_DISCONNECTED 0x0004 +#define DCACHE_DISCONNECTED 0x0010 /* This dentry is possibly not currently connected to the dcache tree, in * which case its parent will either be itself, or will have this flag as * well. nfsd will not use a dentry with this bit set, but will first @@ -196,22 +196,18 @@ struct dentry_operations { * dentry into place and return that dentry rather than the passed one, * typically using d_splice_alias. */ -#define DCACHE_REFERENCED 0x0008 /* Recently used, don't discard. */ -#define DCACHE_RCUACCESS 0x0010 /* Entry has ever been RCU-visible */ -#define DCACHE_INOTIFY_PARENT_WATCHED 0x0020 - /* Parent inode is watched by inotify */ - -#define DCACHE_COOKIE 0x0040 /* For use by dcookie subsystem */ -#define DCACHE_FSNOTIFY_PARENT_WATCHED 0x0080 - /* Parent inode is watched by some fsnotify listener */ +#define DCACHE_REFERENCED 0x0020 /* Recently used, don't discard. */ +#define DCACHE_RCUACCESS 0x0040 /* Entry has ever been RCU-visible */ #define DCACHE_CANT_MOUNT 0x0100 #define DCACHE_GENOCIDE 0x0200 -#define DCACHE_OP_HASH 0x1000 -#define DCACHE_OP_COMPARE 0x2000 -#define DCACHE_OP_REVALIDATE 0x4000 -#define DCACHE_OP_DELETE 0x8000 +#define DCACHE_NFSFS_RENAMED 0x1000 + /* this dentry has been "silly renamed" and has to be deleted on the last + * dput() */ +#define DCACHE_COOKIE 0x2000 /* For use by dcookie subsystem */ +#define DCACHE_FSNOTIFY_PARENT_WATCHED 0x4000 + /* Parent inode is watched by some fsnotify listener */ #define DCACHE_MOUNTED 0x10000 /* is a mountpoint */ #define DCACHE_NEED_AUTOMOUNT 0x20000 /* handle automount on this dir */ From 3ddcd0569cd68f00f3beae9a7959b72918bb91f4 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sat, 6 Aug 2011 22:45:50 -0700 Subject: [PATCH 432/439] vfs: optimize inode cache access patterns The inode structure layout is largely random, and some of the vfs paths really do care. The path lookup in particular is already quite D$ intensive, and profiles show that accessing the 'inode->i_op->xyz' fields is quite costly. We already optimized the dcache to not unnecessarily load the d_op structure for members that are often NULL using the DCACHE_OP_xyz bits in dentry->d_flags, and this does something very similar for the inode ops that are used during pathname lookup. It also re-orders the fields so that the fields accessed by 'stat' are together at the beginning of the inode structure, and roughly in the order accessed. The effect of this seems to be in the 1-2% range for an empty kernel "make -j" run (which is fairly kernel-intensive, mostly in filename lookup), so it's visible. The numbers are fairly noisy, though, and likely depend a lot on exact microarchitecture. So there's more tuning to be done. Signed-off-by: Linus Torvalds --- fs/inode.c | 1 + fs/namei.c | 76 ++++++++++++++++++++++++++++++++++++++++------ fs/stat.c | 4 +-- include/linux/fs.h | 59 +++++++++++++++++++++-------------- 4 files changed, 106 insertions(+), 34 deletions(-) diff --git a/fs/inode.c b/fs/inode.c index 5aab80dc008c..73920d555c88 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -143,6 +143,7 @@ int inode_init_always(struct super_block *sb, struct inode *inode) inode->i_op = &empty_iops; inode->i_fop = &empty_fops; inode->i_nlink = 1; + inode->i_opflags = 0; inode->i_uid = 0; inode->i_gid = 0; atomic_set(&inode->i_writecount, 0); diff --git a/fs/namei.c b/fs/namei.c index 3d607bd80e09..4a98bf154d88 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -308,6 +308,26 @@ int generic_permission(struct inode *inode, int mask) return -EACCES; } +/* + * We _really_ want to just do "generic_permission()" without + * even looking at the inode->i_op values. So we keep a cache + * flag in inode->i_opflags, that says "this has not special + * permission function, use the fast case". + */ +static inline int do_inode_permission(struct inode *inode, int mask) +{ + if (unlikely(!(inode->i_opflags & IOP_FASTPERM))) { + if (likely(inode->i_op->permission)) + return inode->i_op->permission(inode, mask); + + /* This gets set once for the inode lifetime */ + spin_lock(&inode->i_lock); + inode->i_opflags |= IOP_FASTPERM; + spin_unlock(&inode->i_lock); + } + return generic_permission(inode, mask); +} + /** * inode_permission - check for access rights to a given inode * @inode: inode to check permission on @@ -322,7 +342,7 @@ int inode_permission(struct inode *inode, int mask) { int retval; - if (mask & MAY_WRITE) { + if (unlikely(mask & MAY_WRITE)) { umode_t mode = inode->i_mode; /* @@ -339,11 +359,7 @@ int inode_permission(struct inode *inode, int mask) return -EACCES; } - if (inode->i_op->permission) - retval = inode->i_op->permission(inode, mask); - else - retval = generic_permission(inode, mask); - + retval = do_inode_permission(inode, mask); if (retval) return retval; @@ -1245,6 +1261,26 @@ static void terminate_walk(struct nameidata *nd) } } +/* + * Do we need to follow links? We _really_ want to be able + * to do this check without having to look at inode->i_op, + * so we keep a cache of "no, this doesn't need follow_link" + * for the common case. + */ +static inline int do_follow_link(struct inode *inode, int follow) +{ + if (unlikely(!(inode->i_opflags & IOP_NOFOLLOW))) { + if (likely(inode->i_op->follow_link)) + return follow; + + /* This gets set once for the inode lifetime */ + spin_lock(&inode->i_lock); + inode->i_opflags |= IOP_NOFOLLOW; + spin_unlock(&inode->i_lock); + } + return 0; +} + static inline int walk_component(struct nameidata *nd, struct path *path, struct qstr *name, int type, int follow) { @@ -1267,7 +1303,7 @@ static inline int walk_component(struct nameidata *nd, struct path *path, terminate_walk(nd); return -ENOENT; } - if (unlikely(inode->i_op->follow_link) && follow) { + if (do_follow_link(inode, follow)) { if (nd->flags & LOOKUP_RCU) { if (unlikely(unlazy_walk(nd, path->dentry))) { terminate_walk(nd); @@ -1319,6 +1355,26 @@ static inline int nested_symlink(struct path *path, struct nameidata *nd) return res; } +/* + * We really don't want to look at inode->i_op->lookup + * when we don't have to. So we keep a cache bit in + * the inode ->i_opflags field that says "yes, we can + * do lookup on this inode". + */ +static inline int can_lookup(struct inode *inode) +{ + if (likely(inode->i_opflags & IOP_LOOKUP)) + return 1; + if (likely(!inode->i_op->lookup)) + return 0; + + /* We do this once for the lifetime of the inode */ + spin_lock(&inode->i_lock); + inode->i_opflags |= IOP_LOOKUP; + spin_unlock(&inode->i_lock); + return 1; +} + /* * Name resolution. * This is the basic name resolution function, turning a pathname into @@ -1398,10 +1454,10 @@ static int link_path_walk(const char *name, struct nameidata *nd) if (err) return err; } + if (can_lookup(nd->inode)) + continue; err = -ENOTDIR; - if (!nd->inode->i_op->lookup) - break; - continue; + break; /* here ends the main loop */ last_component: diff --git a/fs/stat.c b/fs/stat.c index 961039121cb8..ba5316ffac61 100644 --- a/fs/stat.c +++ b/fs/stat.c @@ -27,12 +27,12 @@ void generic_fillattr(struct inode *inode, struct kstat *stat) stat->uid = inode->i_uid; stat->gid = inode->i_gid; stat->rdev = inode->i_rdev; + stat->size = i_size_read(inode); stat->atime = inode->i_atime; stat->mtime = inode->i_mtime; stat->ctime = inode->i_ctime; - stat->size = i_size_read(inode); - stat->blocks = inode->i_blocks; stat->blksize = (1 << inode->i_blkbits); + stat->blocks = inode->i_blocks; } EXPORT_SYMBOL(generic_fillattr); diff --git a/include/linux/fs.h b/include/linux/fs.h index 786b3b1113cf..178cdb4f1d4a 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -738,22 +738,54 @@ static inline int mapping_writably_mapped(struct address_space *mapping) struct posix_acl; #define ACL_NOT_CACHED ((void *)(-1)) +#define IOP_FASTPERM 0x0001 +#define IOP_LOOKUP 0x0002 +#define IOP_NOFOLLOW 0x0004 + +/* + * Keep mostly read-only and often accessed (especially for + * the RCU path lookup and 'stat' data) fields at the beginning + * of the 'struct inode' + */ struct inode { - /* RCU path lookup touches following: */ umode_t i_mode; + unsigned short i_opflags; uid_t i_uid; gid_t i_gid; + unsigned int i_flags; + +#ifdef CONFIG_FS_POSIX_ACL + struct posix_acl *i_acl; + struct posix_acl *i_default_acl; +#endif + const struct inode_operations *i_op; struct super_block *i_sb; + struct address_space *i_mapping; - spinlock_t i_lock; /* i_blocks, i_bytes, maybe i_size */ - unsigned int i_flags; - unsigned long i_state; #ifdef CONFIG_SECURITY void *i_security; #endif - struct mutex i_mutex; + /* Stat data, not accessed from path walking */ + unsigned long i_ino; + unsigned int i_nlink; + dev_t i_rdev; + loff_t i_size; + struct timespec i_atime; + struct timespec i_mtime; + struct timespec i_ctime; + unsigned int i_blkbits; + blkcnt_t i_blocks; + +#ifdef __NEED_I_SIZE_ORDERED + seqcount_t i_size_seqcount; +#endif + + /* Misc */ + unsigned long i_state; + spinlock_t i_lock; /* i_blocks, i_bytes, maybe i_size */ + struct mutex i_mutex; unsigned long dirtied_when; /* jiffies of first dirtying */ @@ -765,25 +797,12 @@ struct inode { struct list_head i_dentry; struct rcu_head i_rcu; }; - unsigned long i_ino; atomic_t i_count; - unsigned int i_nlink; - dev_t i_rdev; - unsigned int i_blkbits; u64 i_version; - loff_t i_size; -#ifdef __NEED_I_SIZE_ORDERED - seqcount_t i_size_seqcount; -#endif - struct timespec i_atime; - struct timespec i_mtime; - struct timespec i_ctime; - blkcnt_t i_blocks; unsigned short i_bytes; atomic_t i_dio_count; const struct file_operations *i_fop; /* former ->i_op->default_file_ops */ struct file_lock *i_flock; - struct address_space *i_mapping; struct address_space i_data; #ifdef CONFIG_QUOTA struct dquot *i_dquot[MAXQUOTAS]; @@ -806,10 +825,6 @@ struct inode { atomic_t i_readcount; /* struct files open RO */ #endif atomic_t i_writecount; -#ifdef CONFIG_FS_POSIX_ACL - struct posix_acl *i_acl; - struct posix_acl *i_default_acl; -#endif void *i_private; /* fs or device private pointer */ }; From 206b1d09a56dcd2db1052245c4131879c410eaf8 Mon Sep 17 00:00:00 2001 From: Ari Savolainen Date: Sat, 6 Aug 2011 19:43:07 +0300 Subject: [PATCH 433/439] Fix POSIX ACL permission check After commit 3567866bf261: "RCUify freeing acls, let check_acl() go ahead in RCU mode if acl is cached" posix_acl_permission is being called with an unsupported flag and the permission check fails. This patch fixes the issue. Signed-off-by: Ari Savolainen Signed-off-by: Al Viro --- fs/namei.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/namei.c b/fs/namei.c index 4a98bf154d88..fc13609a354a 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -186,7 +186,7 @@ static int check_acl(struct inode *inode, int mask) /* no ->get_acl() calls in RCU mode... */ if (acl == ACL_NOT_CACHED) return -ECHILD; - return posix_acl_permission(inode, acl, mask); + return posix_acl_permission(inode, acl, mask & ~MAY_NOT_BLOCK); } acl = get_cached_acl(inode, ACL_TYPE_ACCESS); From 7813b94a54987571082ff19e9d87eabbfec23b4e Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 7 Aug 2011 09:53:20 -0700 Subject: [PATCH 434/439] vfs: rename 'do_follow_link' to 'should_follow_link' Al points out that the do_follow_link() helper function really is misnamed - it's about whether we should try to follow a symlink or not, not about actually doing the following. Signed-off-by: Linus Torvalds --- fs/namei.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/namei.c b/fs/namei.c index fc13609a354a..2826db35dc25 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1267,7 +1267,7 @@ static void terminate_walk(struct nameidata *nd) * so we keep a cache of "no, this doesn't need follow_link" * for the common case. */ -static inline int do_follow_link(struct inode *inode, int follow) +static inline int should_follow_link(struct inode *inode, int follow) { if (unlikely(!(inode->i_opflags & IOP_NOFOLLOW))) { if (likely(inode->i_op->follow_link)) @@ -1303,7 +1303,7 @@ static inline int walk_component(struct nameidata *nd, struct path *path, terminate_walk(nd); return -ENOENT; } - if (do_follow_link(inode, follow)) { + if (should_follow_link(inode, follow)) { if (nd->flags & LOOKUP_RCU) { if (unlikely(unlazy_walk(nd, path->dentry))) { terminate_walk(nd); From 3295514841c2112d94451ba5deaf54f5afb78ea9 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 7 Aug 2011 18:55:11 +0100 Subject: [PATCH 435/439] fix rcu annotations noise in cred.h task->cred is declared as __rcu, and access to other tasks' ->cred is, indeed, protected. Access to current->cred does not need rcu_dereference() at all, since only the task itself can change its ->cred. sparse, of course, has no way of knowing that... Add force-cast in current_cred(), make current_fsuid() et.al. use it. Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- include/linux/cred.h | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/include/linux/cred.h b/include/linux/cred.h index 48e82af1159b..98f46efbe2d2 100644 --- a/include/linux/cred.h +++ b/include/linux/cred.h @@ -265,10 +265,11 @@ static inline void put_cred(const struct cred *_cred) /** * current_cred - Access the current task's subjective credentials * - * Access the subjective credentials of the current task. + * Access the subjective credentials of the current task. RCU-safe, + * since nobody else can modify it. */ #define current_cred() \ - (current->cred) + (*(__force struct cred **)¤t->cred) /** * __task_cred - Access a task's objective credentials @@ -307,7 +308,7 @@ static inline void put_cred(const struct cred *_cred) ({ \ struct user_struct *__u; \ struct cred *__cred; \ - __cred = (struct cred *) current_cred(); \ + __cred = current_cred(); \ __u = get_uid(__cred->user); \ __u; \ }) @@ -322,7 +323,7 @@ static inline void put_cred(const struct cred *_cred) ({ \ struct group_info *__groups; \ struct cred *__cred; \ - __cred = (struct cred *) current_cred(); \ + __cred = current_cred(); \ __groups = get_group_info(__cred->group_info); \ __groups; \ }) @@ -341,7 +342,7 @@ static inline void put_cred(const struct cred *_cred) #define current_cred_xxx(xxx) \ ({ \ - current->cred->xxx; \ + current_cred()->xxx; \ }) #define current_uid() (current_cred_xxx(uid)) From 4d4487140d34c1b9b321889d2d209321b0da6643 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 7 Aug 2011 14:07:03 -0700 Subject: [PATCH 436/439] arm: remove "optimized" SHA1 routines Since commit 1eb19a12bd22 ("lib/sha1: use the git implementation of SHA-1"), the ARM SHA1 routines no longer work. The reason? They depended on the larger 320-byte workspace, and now the sha1 workspace is just 16 words (64 bytes). So the assembly version would overwrite the stack randomly. The optimized asm version is also probably slower than the new improved C version, so there's no reason to keep it around. At least that was the case in git, where what appears to be the same assembly language version was removed two years ago because the optimized C BLK_SHA1 code was faster. Reported-and-tested-by: Joachim Eastwood Cc: Andreas Schwab Cc: Nicolas Pitre Signed-off-by: Linus Torvalds --- arch/arm/lib/Makefile | 2 +- arch/arm/lib/sha1.S | 211 ------------------------------------------ 2 files changed, 1 insertion(+), 212 deletions(-) delete mode 100644 arch/arm/lib/sha1.S diff --git a/arch/arm/lib/Makefile b/arch/arm/lib/Makefile index 59ff42ddf0ae..cf73a7f742dd 100644 --- a/arch/arm/lib/Makefile +++ b/arch/arm/lib/Makefile @@ -12,7 +12,7 @@ lib-y := backtrace.o changebit.o csumipv6.o csumpartial.o \ strchr.o strrchr.o \ testchangebit.o testclearbit.o testsetbit.o \ ashldi3.o ashrdi3.o lshrdi3.o muldi3.o \ - ucmpdi2.o lib1funcs.o div64.o sha1.o \ + ucmpdi2.o lib1funcs.o div64.o \ io-readsb.o io-writesb.o io-readsl.o io-writesl.o mmu-y := clear_user.o copy_page.o getuser.o putuser.o diff --git a/arch/arm/lib/sha1.S b/arch/arm/lib/sha1.S deleted file mode 100644 index eb0edb80d7b8..000000000000 --- a/arch/arm/lib/sha1.S +++ /dev/null @@ -1,211 +0,0 @@ -/* - * linux/arch/arm/lib/sha1.S - * - * SHA transform optimized for ARM - * - * Copyright: (C) 2005 by Nicolas Pitre - * Created: September 17, 2005 - * - * 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. - * - * The reference implementation for this code is linux/lib/sha1.c - */ - -#include - - .text - - -/* - * void sha_transform(__u32 *digest, const char *in, __u32 *W) - * - * Note: the "in" ptr may be unaligned. - */ - -ENTRY(sha_transform) - - stmfd sp!, {r4 - r8, lr} - - @ for (i = 0; i < 16; i++) - @ W[i] = be32_to_cpu(in[i]); - -#ifdef __ARMEB__ - mov r4, r0 - mov r0, r2 - mov r2, #64 - bl memcpy - mov r2, r0 - mov r0, r4 -#else - mov r3, r2 - mov lr, #16 -1: ldrb r4, [r1], #1 - ldrb r5, [r1], #1 - ldrb r6, [r1], #1 - ldrb r7, [r1], #1 - subs lr, lr, #1 - orr r5, r5, r4, lsl #8 - orr r6, r6, r5, lsl #8 - orr r7, r7, r6, lsl #8 - str r7, [r3], #4 - bne 1b -#endif - - @ for (i = 0; i < 64; i++) - @ W[i+16] = ror(W[i+13] ^ W[i+8] ^ W[i+2] ^ W[i], 31); - - sub r3, r2, #4 - mov lr, #64 -2: ldr r4, [r3, #4]! - subs lr, lr, #1 - ldr r5, [r3, #8] - ldr r6, [r3, #32] - ldr r7, [r3, #52] - eor r4, r4, r5 - eor r4, r4, r6 - eor r4, r4, r7 - mov r4, r4, ror #31 - str r4, [r3, #64] - bne 2b - - /* - * The SHA functions are: - * - * f1(B,C,D) = (D ^ (B & (C ^ D))) - * f2(B,C,D) = (B ^ C ^ D) - * f3(B,C,D) = ((B & C) | (D & (B | C))) - * - * Then the sub-blocks are processed as follows: - * - * A' = ror(A, 27) + f(B,C,D) + E + K + *W++ - * B' = A - * C' = ror(B, 2) - * D' = C - * E' = D - * - * We therefore unroll each loop 5 times to avoid register shuffling. - * Also the ror for C (and also D and E which are successivelyderived - * from it) is applied in place to cut on an additional mov insn for - * each round. - */ - - .macro sha_f1, A, B, C, D, E - ldr r3, [r2], #4 - eor ip, \C, \D - add \E, r1, \E, ror #2 - and ip, \B, ip, ror #2 - add \E, \E, \A, ror #27 - eor ip, ip, \D, ror #2 - add \E, \E, r3 - add \E, \E, ip - .endm - - .macro sha_f2, A, B, C, D, E - ldr r3, [r2], #4 - add \E, r1, \E, ror #2 - eor ip, \B, \C, ror #2 - add \E, \E, \A, ror #27 - eor ip, ip, \D, ror #2 - add \E, \E, r3 - add \E, \E, ip - .endm - - .macro sha_f3, A, B, C, D, E - ldr r3, [r2], #4 - add \E, r1, \E, ror #2 - orr ip, \B, \C, ror #2 - add \E, \E, \A, ror #27 - and ip, ip, \D, ror #2 - add \E, \E, r3 - and r3, \B, \C, ror #2 - orr ip, ip, r3 - add \E, \E, ip - .endm - - ldmia r0, {r4 - r8} - - mov lr, #4 - ldr r1, .L_sha_K + 0 - - /* adjust initial values */ - mov r6, r6, ror #30 - mov r7, r7, ror #30 - mov r8, r8, ror #30 - -3: subs lr, lr, #1 - sha_f1 r4, r5, r6, r7, r8 - sha_f1 r8, r4, r5, r6, r7 - sha_f1 r7, r8, r4, r5, r6 - sha_f1 r6, r7, r8, r4, r5 - sha_f1 r5, r6, r7, r8, r4 - bne 3b - - ldr r1, .L_sha_K + 4 - mov lr, #4 - -4: subs lr, lr, #1 - sha_f2 r4, r5, r6, r7, r8 - sha_f2 r8, r4, r5, r6, r7 - sha_f2 r7, r8, r4, r5, r6 - sha_f2 r6, r7, r8, r4, r5 - sha_f2 r5, r6, r7, r8, r4 - bne 4b - - ldr r1, .L_sha_K + 8 - mov lr, #4 - -5: subs lr, lr, #1 - sha_f3 r4, r5, r6, r7, r8 - sha_f3 r8, r4, r5, r6, r7 - sha_f3 r7, r8, r4, r5, r6 - sha_f3 r6, r7, r8, r4, r5 - sha_f3 r5, r6, r7, r8, r4 - bne 5b - - ldr r1, .L_sha_K + 12 - mov lr, #4 - -6: subs lr, lr, #1 - sha_f2 r4, r5, r6, r7, r8 - sha_f2 r8, r4, r5, r6, r7 - sha_f2 r7, r8, r4, r5, r6 - sha_f2 r6, r7, r8, r4, r5 - sha_f2 r5, r6, r7, r8, r4 - bne 6b - - ldmia r0, {r1, r2, r3, ip, lr} - add r4, r1, r4 - add r5, r2, r5 - add r6, r3, r6, ror #2 - add r7, ip, r7, ror #2 - add r8, lr, r8, ror #2 - stmia r0, {r4 - r8} - - ldmfd sp!, {r4 - r8, pc} - -ENDPROC(sha_transform) - - .align 2 -.L_sha_K: - .word 0x5a827999, 0x6ed9eba1, 0x8f1bbcdc, 0xca62c1d6 - - -/* - * void sha_init(__u32 *buf) - */ - - .align 2 -.L_sha_initial_digest: - .word 0x67452301, 0xefcdab89, 0x98badcfe, 0x10325476, 0xc3d2e1f0 - -ENTRY(sha_init) - - str lr, [sp, #-4]! - adr r1, .L_sha_initial_digest - ldmia r1, {r1, r2, r3, ip, lr} - stmia r0, {r1, r2, r3, ip, lr} - ldr pc, [sp], #4 - -ENDPROC(sha_init) From f23c126bfabef88c201c8cc56bd3ccd9d59c51e0 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 7 Aug 2011 15:49:11 -0700 Subject: [PATCH 437/439] arm: remove stale export of 'sha_transform' The generic library code already exports the generic function, this was left-over from the ARM-specific version that just got removed. Signed-off-by: Linus Torvalds --- arch/arm/kernel/armksyms.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/arch/arm/kernel/armksyms.c b/arch/arm/kernel/armksyms.c index acca35aebe28..aeef960ff795 100644 --- a/arch/arm/kernel/armksyms.c +++ b/arch/arm/kernel/armksyms.c @@ -112,9 +112,6 @@ EXPORT_SYMBOL(__put_user_4); EXPORT_SYMBOL(__put_user_8); #endif - /* crypto hash */ -EXPORT_SYMBOL(sha_transform); - /* gcc lib functions */ EXPORT_SYMBOL(__ashldi3); EXPORT_SYMBOL(__ashrdi3); From fc97114b8d67819fadcc5af855da9a3e6a6a329b Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 8 Aug 2011 00:26:50 +0200 Subject: [PATCH 438/439] sh: Fix boot crash related to SCI Commit d006199e72a9 ("serial: sh-sci: Regtype probing doesn't need to be fatal.") made sci_init_single() return when sci_probe_regmap() succeeds, although it should return when sci_probe_regmap() fails. This causes systems using the serial sh-sci driver to crash during boot. Fix the problem by using the right return condition. Signed-off-by: Rafael J. Wysocki Signed-off-by: Linus Torvalds --- drivers/tty/serial/sh-sci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/tty/serial/sh-sci.c b/drivers/tty/serial/sh-sci.c index d0a56235c50e..2ec57b2fb278 100644 --- a/drivers/tty/serial/sh-sci.c +++ b/drivers/tty/serial/sh-sci.c @@ -1889,7 +1889,7 @@ static int __devinit sci_init_single(struct platform_device *dev, if (p->regtype == SCIx_PROBE_REGTYPE) { ret = sci_probe_regmap(p); - if (unlikely(!ret)) + if (unlikely(ret)) return ret; } From 322a8b034003c0d46d39af85bf24fee27b902f48 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 7 Aug 2011 18:23:30 -0700 Subject: [PATCH 439/439] Linux 3.1-rc1 --- Makefile | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/Makefile b/Makefile index f676d15cd348..b4ca4e111c9a 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 3 -PATCHLEVEL = 0 +PATCHLEVEL = 1 SUBLEVEL = 0 -EXTRAVERSION = +EXTRAVERSION = -rc1 NAME = Sneaky Weasel # *DOCUMENTATION*