From 5070437cd99511f69ae561f2ab417142a47a85ec Mon Sep 17 00:00:00 2001 From: Lars-Peter Clausen Date: Thu, 21 Oct 2010 17:55:01 +0200 Subject: power_supply: Add gpio charger driver This patch adds a simple driver for chargers indicating their online status through a GPIO pin. Signed-off-by: Lars-Peter Clausen Signed-off-by: Anton Vorontsov --- include/linux/power/gpio-charger.h | 41 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 41 insertions(+) create mode 100644 include/linux/power/gpio-charger.h (limited to 'include/linux') diff --git a/include/linux/power/gpio-charger.h b/include/linux/power/gpio-charger.h new file mode 100644 index 000000000000..de1dfe09a03d --- /dev/null +++ b/include/linux/power/gpio-charger.h @@ -0,0 +1,41 @@ +/* + * Copyright (C) 2010, Lars-Peter Clausen + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + * + * You should have received a copy of the GNU General Public License along + * with this program; if not, write to the Free Software Foundation, Inc., + * 675 Mass Ave, Cambridge, MA 02139, USA. + * + */ + +#ifndef __LINUX_POWER_GPIO_CHARGER_H__ +#define __LINUX_POWER_GPIO_CHARGER_H__ + +#include +#include + +/** + * struct gpio_charger_platform_data - platform_data for gpio_charger devices + * @name: Name for the chargers power_supply device + * @type: Type of the charger + * @gpio: GPIO which is used to indicate the chargers status + * @gpio_active_low: Should be set to 1 if the GPIO is active low otherwise 0 + * @supplied_to: Array of battery names to which this chargers supplies power + * @num_supplicants: Number of entries in the supplied_to array + */ +struct gpio_charger_platform_data { + const char *name; + enum power_supply_type type; + + int gpio; + int gpio_active_low; + + char **supplied_to; + size_t num_supplicants; +}; + +#endif -- cgit v1.2.3 From 5b275ce27077d6463ca28c9671dce7c2c1f622e2 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Thu, 11 Nov 2010 15:27:29 +0000 Subject: thermal: make ops constant And while touching that function definition do something about the disaster of formatting there. Signed-off-by: Alan Cox Acked-by: Zhang Rui Signed-off-by: Len Brown --- drivers/thermal/thermal_sys.c | 17 +++++------------ include/linux/thermal.h | 15 +++++---------- 2 files changed, 10 insertions(+), 22 deletions(-) (limited to 'include/linux') diff --git a/drivers/thermal/thermal_sys.c b/drivers/thermal/thermal_sys.c index 13c72c629329..bde3477c1c6b 100644 --- a/drivers/thermal/thermal_sys.c +++ b/drivers/thermal/thermal_sys.c @@ -823,11 +823,8 @@ static struct class thermal_class = { * @devdata: device private data. * @ops: standard thermal cooling devices callbacks. */ -struct thermal_cooling_device *thermal_cooling_device_register(char *type, - void *devdata, - struct - thermal_cooling_device_ops - *ops) +struct thermal_cooling_device *thermal_cooling_device_register( + char *type, void *devdata, const struct thermal_cooling_device_ops *ops) { struct thermal_cooling_device *cdev; struct thermal_zone_device *pos; @@ -1048,13 +1045,9 @@ EXPORT_SYMBOL(thermal_zone_device_update); * section 11.1.5.1 of the ACPI specification 3.0. */ struct thermal_zone_device *thermal_zone_device_register(char *type, - int trips, - void *devdata, struct - thermal_zone_device_ops - *ops, int tc1, int - tc2, - int passive_delay, - int polling_delay) + int trips, void *devdata, + const struct thermal_zone_device_ops *ops, + int tc1, int tc2, int passive_delay, int polling_delay) { struct thermal_zone_device *tz; struct thermal_cooling_device *pos; diff --git a/include/linux/thermal.h b/include/linux/thermal.h index 1de8b9eb841b..06626904daa0 100644 --- a/include/linux/thermal.h +++ b/include/linux/thermal.h @@ -77,7 +77,7 @@ struct thermal_cooling_device { char type[THERMAL_NAME_LENGTH]; struct device device; void *devdata; - struct thermal_cooling_device_ops *ops; + const struct thermal_cooling_device_ops *ops; struct list_head node; }; @@ -114,7 +114,7 @@ struct thermal_zone_device { int last_temperature; bool passive; unsigned int forced_passive; - struct thermal_zone_device_ops *ops; + const struct thermal_zone_device_ops *ops; struct list_head cooling_devices; struct idr idr; struct mutex lock; /* protect cooling devices list */ @@ -129,11 +129,8 @@ struct thermal_zone_device { }; struct thermal_zone_device *thermal_zone_device_register(char *, int, void *, - struct - thermal_zone_device_ops - *, int tc1, int tc2, - int passive_freq, - int polling_freq); + const struct thermal_zone_device_ops *, int tc1, int tc2, + int passive_freq, int polling_freq); void thermal_zone_device_unregister(struct thermal_zone_device *); int thermal_zone_bind_cooling_device(struct thermal_zone_device *, int, @@ -142,9 +139,7 @@ int thermal_zone_unbind_cooling_device(struct thermal_zone_device *, int, struct thermal_cooling_device *); void thermal_zone_device_update(struct thermal_zone_device *); struct thermal_cooling_device *thermal_cooling_device_register(char *, void *, - struct - thermal_cooling_device_ops - *); + const struct thermal_cooling_device_ops *); void thermal_cooling_device_unregister(struct thermal_cooling_device *); #endif /* __THERMAL_H__ */ -- cgit v1.2.3 From 3e3198f1adda8e0fbd499bde806781237d6c841f Mon Sep 17 00:00:00 2001 From: Roman Tereshonkov Date: Wed, 3 Nov 2010 12:55:19 +0200 Subject: mtd: onenand: add option and variable for cache program feature A new option is added for cache program feature. A new variable ongoing is introduced for onenand_chip to handle the multi-command cache program operation. Signed-off-by: Roman Tereshonkov Acked-by: Kyungmin Park Signed-off-by: Artem Bityutskiy Signed-off-by: David Woodhouse --- include/linux/mtd/onenand.h | 12 ++++++++++++ 1 file changed, 12 insertions(+) (limited to 'include/linux') diff --git a/include/linux/mtd/onenand.h b/include/linux/mtd/onenand.h index 0c8815bfae1c..6da3fe314828 100644 --- a/include/linux/mtd/onenand.h +++ b/include/linux/mtd/onenand.h @@ -137,6 +137,14 @@ struct onenand_chip { void *bbm; void *priv; + + /* + * Shows that the current operation is composed + * of sequence of commands. For example, cache program. + * Such command status OnGo bit is checked at the end of + * sequence. + */ + unsigned int ongoing; }; /* @@ -171,6 +179,9 @@ struct onenand_chip { #define ONENAND_IS_2PLANE(this) (0) #endif +#define ONENAND_IS_CACHE_PROGRAM(this) \ + (this->options & ONENAND_HAS_CACHE_PROGRAM) + /* Check byte access in OneNAND */ #define ONENAND_CHECK_BYTE_ACCESS(addr) (addr & 0x1) @@ -181,6 +192,7 @@ struct onenand_chip { #define ONENAND_HAS_UNLOCK_ALL (0x0002) #define ONENAND_HAS_2PLANE (0x0004) #define ONENAND_HAS_4KB_PAGE (0x0008) +#define ONENAND_HAS_CACHE_PROGRAM (0x0010) #define ONENAND_SKIP_UNLOCK_CHECK (0x0100) #define ONENAND_PAGEBUF_ALLOC (0x1000) #define ONENAND_OOBBUF_ALLOC (0x2000) -- cgit v1.2.3 From 65ab220c30bc2120eaa39753cadec68616e3f906 Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Mon, 8 Nov 2010 11:17:54 +0530 Subject: mtd: fsmc.h: including linux/io.h for definition of readl Signed-off-by: Viresh Kumar Signed-off-by: Artem Bityutskiy Signed-off-by: David Woodhouse --- include/linux/mtd/fsmc.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/mtd/fsmc.h b/include/linux/mtd/fsmc.h index 5d2556700ec2..e210b87e6cd6 100644 --- a/include/linux/mtd/fsmc.h +++ b/include/linux/mtd/fsmc.h @@ -16,6 +16,7 @@ #ifndef __MTD_FSMC_H #define __MTD_FSMC_H +#include #include #include #include -- cgit v1.2.3 From cc31822250236ec173bb2aa149ebe2ba35405db2 Mon Sep 17 00:00:00 2001 From: Guillaume LECERF Date: Wed, 17 Nov 2010 12:35:50 +0100 Subject: mtd: cfi_fixup: remove unused 'param' parameter The 'param' parameter has never been used since its introduction, so simply remove it. Signed-off-by: Guillaume LECERF Signed-off-by: Artem Bityutskiy Signed-off-by: David Woodhouse --- drivers/mtd/chips/cfi_cmdset_0001.c | 54 ++++++++++----------- drivers/mtd/chips/cfi_cmdset_0002.c | 94 ++++++++++++++++++------------------- drivers/mtd/chips/cfi_util.c | 2 +- drivers/mtd/chips/fwh_lock.h | 2 +- include/linux/mtd/cfi.h | 3 +- 5 files changed, 77 insertions(+), 78 deletions(-) (limited to 'include/linux') diff --git a/drivers/mtd/chips/cfi_cmdset_0001.c b/drivers/mtd/chips/cfi_cmdset_0001.c index ad9268b44416..44cbfc093ecc 100644 --- a/drivers/mtd/chips/cfi_cmdset_0001.c +++ b/drivers/mtd/chips/cfi_cmdset_0001.c @@ -162,7 +162,7 @@ static void cfi_tell_features(struct cfi_pri_intelext *extp) #endif /* Atmel chips don't use the same PRI format as Intel chips */ -static void fixup_convert_atmel_pri(struct mtd_info *mtd, void *param) +static void fixup_convert_atmel_pri(struct mtd_info *mtd) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; @@ -202,7 +202,7 @@ static void fixup_convert_atmel_pri(struct mtd_info *mtd, void *param) cfi->cfiq->BufWriteTimeoutMax = 0; } -static void fixup_at49bv640dx_lock(struct mtd_info *mtd, void *param) +static void fixup_at49bv640dx_lock(struct mtd_info *mtd) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; @@ -214,7 +214,7 @@ static void fixup_at49bv640dx_lock(struct mtd_info *mtd, void *param) #ifdef CMDSET0001_DISABLE_ERASE_SUSPEND_ON_WRITE /* Some Intel Strata Flash prior to FPO revision C has bugs in this area */ -static void fixup_intel_strataflash(struct mtd_info *mtd, void* param) +static void fixup_intel_strataflash(struct mtd_info *mtd) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; @@ -227,7 +227,7 @@ static void fixup_intel_strataflash(struct mtd_info *mtd, void* param) #endif #ifdef CMDSET0001_DISABLE_WRITE_SUSPEND -static void fixup_no_write_suspend(struct mtd_info *mtd, void* param) +static void fixup_no_write_suspend(struct mtd_info *mtd) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; @@ -240,7 +240,7 @@ static void fixup_no_write_suspend(struct mtd_info *mtd, void* param) } #endif -static void fixup_st_m28w320ct(struct mtd_info *mtd, void* param) +static void fixup_st_m28w320ct(struct mtd_info *mtd) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; @@ -249,7 +249,7 @@ static void fixup_st_m28w320ct(struct mtd_info *mtd, void* param) cfi->cfiq->BufWriteTimeoutMax = 0; /* Not supported */ } -static void fixup_st_m28w320cb(struct mtd_info *mtd, void* param) +static void fixup_st_m28w320cb(struct mtd_info *mtd) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; @@ -259,7 +259,7 @@ static void fixup_st_m28w320cb(struct mtd_info *mtd, void* param) (cfi->cfiq->EraseRegionInfo[1] & 0xffff0000) | 0x3e; }; -static void fixup_use_point(struct mtd_info *mtd, void *param) +static void fixup_use_point(struct mtd_info *mtd) { struct map_info *map = mtd->priv; if (!mtd->point && map_is_linear(map)) { @@ -268,7 +268,7 @@ static void fixup_use_point(struct mtd_info *mtd, void *param) } } -static void fixup_use_write_buffers(struct mtd_info *mtd, void *param) +static void fixup_use_write_buffers(struct mtd_info *mtd) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; @@ -282,7 +282,7 @@ static void fixup_use_write_buffers(struct mtd_info *mtd, void *param) /* * Some chips power-up with all sectors locked by default. */ -static void fixup_unlock_powerup_lock(struct mtd_info *mtd, void *param) +static void fixup_unlock_powerup_lock(struct mtd_info *mtd) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; @@ -295,31 +295,31 @@ static void fixup_unlock_powerup_lock(struct mtd_info *mtd, void *param) } static struct cfi_fixup cfi_fixup_table[] = { - { CFI_MFR_ATMEL, CFI_ID_ANY, fixup_convert_atmel_pri, NULL }, - { CFI_MFR_ATMEL, AT49BV640D, fixup_at49bv640dx_lock, NULL }, - { CFI_MFR_ATMEL, AT49BV640DT, fixup_at49bv640dx_lock, NULL }, + { CFI_MFR_ATMEL, CFI_ID_ANY, fixup_convert_atmel_pri }, + { CFI_MFR_ATMEL, AT49BV640D, fixup_at49bv640dx_lock }, + { CFI_MFR_ATMEL, AT49BV640DT, fixup_at49bv640dx_lock }, #ifdef CMDSET0001_DISABLE_ERASE_SUSPEND_ON_WRITE - { CFI_MFR_ANY, CFI_ID_ANY, fixup_intel_strataflash, NULL }, + { CFI_MFR_ANY, CFI_ID_ANY, fixup_intel_strataflash }, #endif #ifdef CMDSET0001_DISABLE_WRITE_SUSPEND - { CFI_MFR_ANY, CFI_ID_ANY, fixup_no_write_suspend, NULL }, + { CFI_MFR_ANY, CFI_ID_ANY, fixup_no_write_suspend }, #endif #if !FORCE_WORD_WRITE - { CFI_MFR_ANY, CFI_ID_ANY, fixup_use_write_buffers, NULL }, + { CFI_MFR_ANY, CFI_ID_ANY, fixup_use_write_buffers }, #endif - { CFI_MFR_ST, 0x00ba, /* M28W320CT */ fixup_st_m28w320ct, NULL }, - { CFI_MFR_ST, 0x00bb, /* M28W320CB */ fixup_st_m28w320cb, NULL }, - { CFI_MFR_INTEL, CFI_ID_ANY, fixup_unlock_powerup_lock, NULL, }, - { 0, 0, NULL, NULL } + { CFI_MFR_ST, 0x00ba, /* M28W320CT */ fixup_st_m28w320ct }, + { CFI_MFR_ST, 0x00bb, /* M28W320CB */ fixup_st_m28w320cb }, + { CFI_MFR_INTEL, CFI_ID_ANY, fixup_unlock_powerup_lock }, + { 0, 0, NULL } }; static struct cfi_fixup jedec_fixup_table[] = { - { CFI_MFR_INTEL, I82802AB, fixup_use_fwh_lock, NULL, }, - { CFI_MFR_INTEL, I82802AC, fixup_use_fwh_lock, NULL, }, - { CFI_MFR_ST, M50LPW080, fixup_use_fwh_lock, NULL, }, - { CFI_MFR_ST, M50FLW080A, fixup_use_fwh_lock, NULL, }, - { CFI_MFR_ST, M50FLW080B, fixup_use_fwh_lock, NULL, }, - { 0, 0, NULL, NULL } + { CFI_MFR_INTEL, I82802AB, fixup_use_fwh_lock }, + { CFI_MFR_INTEL, I82802AC, fixup_use_fwh_lock }, + { CFI_MFR_ST, M50LPW080, fixup_use_fwh_lock }, + { CFI_MFR_ST, M50FLW080A, fixup_use_fwh_lock }, + { CFI_MFR_ST, M50FLW080B, fixup_use_fwh_lock }, + { 0, 0, NULL } }; static struct cfi_fixup fixup_table[] = { /* The CFI vendor ids and the JEDEC vendor IDs appear @@ -327,8 +327,8 @@ static struct cfi_fixup fixup_table[] = { * well. This table is to pick all cases where * we know that is the case. */ - { CFI_MFR_ANY, CFI_ID_ANY, fixup_use_point, NULL }, - { 0, 0, NULL, NULL } + { CFI_MFR_ANY, CFI_ID_ANY, fixup_use_point }, + { 0, 0, NULL } }; static void cfi_fixup_major_minor(struct cfi_private *cfi, diff --git a/drivers/mtd/chips/cfi_cmdset_0002.c b/drivers/mtd/chips/cfi_cmdset_0002.c index 3b8e32d87977..9d68ab919f31 100644 --- a/drivers/mtd/chips/cfi_cmdset_0002.c +++ b/drivers/mtd/chips/cfi_cmdset_0002.c @@ -134,7 +134,7 @@ static void cfi_tell_features(struct cfi_pri_amdstd *extp) #ifdef AMD_BOOTLOC_BUG /* Wheee. Bring me the head of someone at AMD. */ -static void fixup_amd_bootblock(struct mtd_info *mtd, void* param) +static void fixup_amd_bootblock(struct mtd_info *mtd) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; @@ -186,7 +186,7 @@ static void fixup_amd_bootblock(struct mtd_info *mtd, void* param) } #endif -static void fixup_use_write_buffers(struct mtd_info *mtd, void *param) +static void fixup_use_write_buffers(struct mtd_info *mtd) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; @@ -197,7 +197,7 @@ static void fixup_use_write_buffers(struct mtd_info *mtd, void *param) } /* Atmel chips don't use the same PRI format as AMD chips */ -static void fixup_convert_atmel_pri(struct mtd_info *mtd, void *param) +static void fixup_convert_atmel_pri(struct mtd_info *mtd) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; @@ -228,14 +228,14 @@ static void fixup_convert_atmel_pri(struct mtd_info *mtd, void *param) cfi->cfiq->BufWriteTimeoutMax = 0; } -static void fixup_use_secsi(struct mtd_info *mtd, void *param) +static void fixup_use_secsi(struct mtd_info *mtd) { /* Setup for chips with a secsi area */ mtd->read_user_prot_reg = cfi_amdstd_secsi_read; mtd->read_fact_prot_reg = cfi_amdstd_secsi_read; } -static void fixup_use_erase_chip(struct mtd_info *mtd, void *param) +static void fixup_use_erase_chip(struct mtd_info *mtd) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; @@ -250,7 +250,7 @@ static void fixup_use_erase_chip(struct mtd_info *mtd, void *param) * Some Atmel chips (e.g. the AT49BV6416) power-up with all sectors * locked by default. */ -static void fixup_use_atmel_lock(struct mtd_info *mtd, void *param) +static void fixup_use_atmel_lock(struct mtd_info *mtd) { mtd->lock = cfi_atmel_lock; mtd->unlock = cfi_atmel_unlock; @@ -271,7 +271,7 @@ static void fixup_old_sst_eraseregion(struct mtd_info *mtd) cfi->cfiq->NumEraseRegions = 1; } -static void fixup_sst39vf(struct mtd_info *mtd, void *param) +static void fixup_sst39vf(struct mtd_info *mtd) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; @@ -282,7 +282,7 @@ static void fixup_sst39vf(struct mtd_info *mtd, void *param) cfi->addr_unlock2 = 0x2AAA; } -static void fixup_sst39vf_rev_b(struct mtd_info *mtd, void *param) +static void fixup_sst39vf_rev_b(struct mtd_info *mtd) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; @@ -295,12 +295,12 @@ static void fixup_sst39vf_rev_b(struct mtd_info *mtd, void *param) cfi->sector_erase_cmd = CMD(0x50); } -static void fixup_sst38vf640x_sectorsize(struct mtd_info *mtd, void *param) +static void fixup_sst38vf640x_sectorsize(struct mtd_info *mtd) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; - fixup_sst39vf_rev_b(mtd, param); + fixup_sst39vf_rev_b(mtd); /* * CFI reports 1024 sectors (0x03ff+1) of 64KBytes (0x0100*256) where @@ -310,7 +310,7 @@ static void fixup_sst38vf640x_sectorsize(struct mtd_info *mtd, void *param) pr_warning("%s: Bad 38VF640x CFI data; adjusting sector size from 64 to 8KiB\n", mtd->name); } -static void fixup_s29gl064n_sectors(struct mtd_info *mtd, void *param) +static void fixup_s29gl064n_sectors(struct mtd_info *mtd) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; @@ -321,7 +321,7 @@ static void fixup_s29gl064n_sectors(struct mtd_info *mtd, void *param) } } -static void fixup_s29gl032n_sectors(struct mtd_info *mtd, void *param) +static void fixup_s29gl032n_sectors(struct mtd_info *mtd) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; @@ -334,47 +334,47 @@ static void fixup_s29gl032n_sectors(struct mtd_info *mtd, void *param) /* Used to fix CFI-Tables of chips without Extended Query Tables */ static struct cfi_fixup cfi_nopri_fixup_table[] = { - { CFI_MFR_SST, 0x234A, fixup_sst39vf, NULL, }, /* SST39VF1602 */ - { CFI_MFR_SST, 0x234B, fixup_sst39vf, NULL, }, /* SST39VF1601 */ - { CFI_MFR_SST, 0x235A, fixup_sst39vf, NULL, }, /* SST39VF3202 */ - { CFI_MFR_SST, 0x235B, fixup_sst39vf, NULL, }, /* SST39VF3201 */ - { CFI_MFR_SST, 0x235C, fixup_sst39vf_rev_b, NULL, }, /* SST39VF3202B */ - { CFI_MFR_SST, 0x235D, fixup_sst39vf_rev_b, NULL, }, /* SST39VF3201B */ - { CFI_MFR_SST, 0x236C, fixup_sst39vf_rev_b, NULL, }, /* SST39VF6402B */ - { CFI_MFR_SST, 0x236D, fixup_sst39vf_rev_b, NULL, }, /* SST39VF6401B */ - { 0, 0, NULL, NULL } + { CFI_MFR_SST, 0x234a, fixup_sst39vf }, /* SST39VF1602 */ + { CFI_MFR_SST, 0x234b, fixup_sst39vf }, /* SST39VF1601 */ + { CFI_MFR_SST, 0x235a, fixup_sst39vf }, /* SST39VF3202 */ + { CFI_MFR_SST, 0x235b, fixup_sst39vf }, /* SST39VF3201 */ + { CFI_MFR_SST, 0x235c, fixup_sst39vf_rev_b }, /* SST39VF3202B */ + { CFI_MFR_SST, 0x235d, fixup_sst39vf_rev_b }, /* SST39VF3201B */ + { CFI_MFR_SST, 0x236c, fixup_sst39vf_rev_b }, /* SST39VF6402B */ + { CFI_MFR_SST, 0x236d, fixup_sst39vf_rev_b }, /* SST39VF6401B */ + { 0, 0, NULL } }; static struct cfi_fixup cfi_fixup_table[] = { - { CFI_MFR_ATMEL, CFI_ID_ANY, fixup_convert_atmel_pri, NULL }, + { CFI_MFR_ATMEL, CFI_ID_ANY, fixup_convert_atmel_pri }, #ifdef AMD_BOOTLOC_BUG - { CFI_MFR_AMD, CFI_ID_ANY, fixup_amd_bootblock, NULL }, - { CFI_MFR_MACRONIX, CFI_ID_ANY, fixup_amd_bootblock, NULL }, + { CFI_MFR_AMD, CFI_ID_ANY, fixup_amd_bootblock }, + { CFI_MFR_MACRONIX, CFI_ID_ANY, fixup_amd_bootblock }, #endif - { CFI_MFR_AMD, 0x0050, fixup_use_secsi, NULL, }, - { CFI_MFR_AMD, 0x0053, fixup_use_secsi, NULL, }, - { CFI_MFR_AMD, 0x0055, fixup_use_secsi, NULL, }, - { CFI_MFR_AMD, 0x0056, fixup_use_secsi, NULL, }, - { CFI_MFR_AMD, 0x005C, fixup_use_secsi, NULL, }, - { CFI_MFR_AMD, 0x005F, fixup_use_secsi, NULL, }, - { CFI_MFR_AMD, 0x0c01, fixup_s29gl064n_sectors, NULL, }, - { CFI_MFR_AMD, 0x1301, fixup_s29gl064n_sectors, NULL, }, - { CFI_MFR_AMD, 0x1a00, fixup_s29gl032n_sectors, NULL, }, - { CFI_MFR_AMD, 0x1a01, fixup_s29gl032n_sectors, NULL, }, - { CFI_MFR_SST, 0x536A, fixup_sst38vf640x_sectorsize, NULL, }, /* SST38VF6402 */ - { CFI_MFR_SST, 0x536B, fixup_sst38vf640x_sectorsize, NULL, }, /* SST38VF6401 */ - { CFI_MFR_SST, 0x536C, fixup_sst38vf640x_sectorsize, NULL, }, /* SST38VF6404 */ - { CFI_MFR_SST, 0x536D, fixup_sst38vf640x_sectorsize, NULL, }, /* SST38VF6403 */ + { CFI_MFR_AMD, 0x0050, fixup_use_secsi }, + { CFI_MFR_AMD, 0x0053, fixup_use_secsi }, + { CFI_MFR_AMD, 0x0055, fixup_use_secsi }, + { CFI_MFR_AMD, 0x0056, fixup_use_secsi }, + { CFI_MFR_AMD, 0x005C, fixup_use_secsi }, + { CFI_MFR_AMD, 0x005F, fixup_use_secsi }, + { CFI_MFR_AMD, 0x0c01, fixup_s29gl064n_sectors }, + { CFI_MFR_AMD, 0x1301, fixup_s29gl064n_sectors }, + { CFI_MFR_AMD, 0x1a00, fixup_s29gl032n_sectors }, + { CFI_MFR_AMD, 0x1a01, fixup_s29gl032n_sectors }, + { CFI_MFR_SST, 0x536a, fixup_sst38vf640x_sectorsize }, /* SST38VF6402 */ + { CFI_MFR_SST, 0x536b, fixup_sst38vf640x_sectorsize }, /* SST38VF6401 */ + { CFI_MFR_SST, 0x536c, fixup_sst38vf640x_sectorsize }, /* SST38VF6404 */ + { CFI_MFR_SST, 0x536d, fixup_sst38vf640x_sectorsize }, /* SST38VF6403 */ #if !FORCE_WORD_WRITE - { CFI_MFR_ANY, CFI_ID_ANY, fixup_use_write_buffers, NULL, }, + { CFI_MFR_ANY, CFI_ID_ANY, fixup_use_write_buffers }, #endif - { 0, 0, NULL, NULL } + { 0, 0, NULL } }; static struct cfi_fixup jedec_fixup_table[] = { - { CFI_MFR_SST, SST49LF004B, fixup_use_fwh_lock, NULL, }, - { CFI_MFR_SST, SST49LF040B, fixup_use_fwh_lock, NULL, }, - { CFI_MFR_SST, SST49LF008A, fixup_use_fwh_lock, NULL, }, - { 0, 0, NULL, NULL } + { CFI_MFR_SST, SST49LF004B, fixup_use_fwh_lock }, + { CFI_MFR_SST, SST49LF040B, fixup_use_fwh_lock }, + { CFI_MFR_SST, SST49LF008A, fixup_use_fwh_lock }, + { 0, 0, NULL } }; static struct cfi_fixup fixup_table[] = { @@ -383,9 +383,9 @@ static struct cfi_fixup fixup_table[] = { * well. This table is to pick all cases where * we know that is the case. */ - { CFI_MFR_ANY, CFI_ID_ANY, fixup_use_erase_chip, NULL }, - { CFI_MFR_ATMEL, AT49BV6416, fixup_use_atmel_lock, NULL }, - { 0, 0, NULL, NULL } + { CFI_MFR_ANY, CFI_ID_ANY, fixup_use_erase_chip }, + { CFI_MFR_ATMEL, AT49BV6416, fixup_use_atmel_lock }, + { 0, 0, NULL } }; diff --git a/drivers/mtd/chips/cfi_util.c b/drivers/mtd/chips/cfi_util.c index 360525c637d2..6ae3d111e1e7 100644 --- a/drivers/mtd/chips/cfi_util.c +++ b/drivers/mtd/chips/cfi_util.c @@ -156,7 +156,7 @@ void cfi_fixup(struct mtd_info *mtd, struct cfi_fixup *fixups) for (f=fixups; f->fixup; f++) { if (((f->mfr == CFI_MFR_ANY) || (f->mfr == cfi->mfr)) && ((f->id == CFI_ID_ANY) || (f->id == cfi->id))) { - f->fixup(mtd, f->param); + f->fixup(mtd); } } } diff --git a/drivers/mtd/chips/fwh_lock.h b/drivers/mtd/chips/fwh_lock.h index d18064977192..5e3cc80128aa 100644 --- a/drivers/mtd/chips/fwh_lock.h +++ b/drivers/mtd/chips/fwh_lock.h @@ -98,7 +98,7 @@ static int fwh_unlock_varsize(struct mtd_info *mtd, loff_t ofs, uint64_t len) return ret; } -static void fixup_use_fwh_lock(struct mtd_info *mtd, void *param) +static void fixup_use_fwh_lock(struct mtd_info *mtd) { printk(KERN_NOTICE "using fwh lock/unlock method\n"); /* Setup for the chips with the fwh lock method */ diff --git a/include/linux/mtd/cfi.h b/include/linux/mtd/cfi.h index 4dd0c2cd7659..a9baee6864af 100644 --- a/include/linux/mtd/cfi.h +++ b/include/linux/mtd/cfi.h @@ -527,8 +527,7 @@ struct cfi_extquery *cfi_read_pri(struct map_info *map, uint16_t adr, uint16_t s struct cfi_fixup { uint16_t mfr; uint16_t id; - void (*fixup)(struct mtd_info *mtd, void* param); - void* param; + void (*fixup)(struct mtd_info *mtd); }; #define CFI_MFR_ANY 0xFFFF -- cgit v1.2.3 From 1534b8b09757190ce6e97fa97f9ad77c49082cd8 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 18 Nov 2010 15:02:21 -0800 Subject: mtd: fix nand kernel-doc warnings Warning(include/linux/mtd/nand.h:543): No description found for parameter 'badblockbits' Warning(drivers/mtd/nand/nand_bbt.c:1101): No description found for parameter 'mtd' Signed-off-by: Randy Dunlap Cc: David Woodhouse Cc: linux-mtd@lists.infradead.org Signed-off-by: Artem Bityutskiy Signed-off-by: David Woodhouse --- drivers/mtd/nand/nand_bbt.c | 3 ++- include/linux/mtd/nand.h | 2 ++ 2 files changed, 4 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/mtd/nand/nand_bbt.c b/drivers/mtd/nand/nand_bbt.c index 586b981f0e61..6ebd869993aa 100644 --- a/drivers/mtd/nand/nand_bbt.c +++ b/drivers/mtd/nand/nand_bbt.c @@ -1092,7 +1092,8 @@ static void mark_bbt_region(struct mtd_info *mtd, struct nand_bbt_descr *td) /** * verify_bbt_descr - verify the bad block description - * @bd: the table to verify + * @mtd: MTD device structure + * @bd: the table to verify * * This functions performs a few sanity checks on the bad block description * table. diff --git a/include/linux/mtd/nand.h b/include/linux/mtd/nand.h index 63e17d01fde9..1f489b247a29 100644 --- a/include/linux/mtd/nand.h +++ b/include/linux/mtd/nand.h @@ -448,6 +448,8 @@ struct nand_buffers { * See the defines for further explanation. * @badblockpos: [INTERN] position of the bad block marker in the oob * area. + * @badblockbits: [INTERN] number of bits to left-shift the bad block + * number * @cellinfo: [INTERN] MLC/multichip data from chip ident * @numchips: [INTERN] number of physical chips * @chipsize: [INTERN] the size of one chip for multichip arrays -- cgit v1.2.3 From a7e93dcd9aacb3ef4acfcc4310577f3ae0741821 Mon Sep 17 00:00:00 2001 From: Roman Tereshonkov Date: Tue, 23 Nov 2010 14:17:17 +0200 Subject: mtd: fix master device identification for mtd repartition Function mtd_has_master renamed as mtd_is_partition to follow the function logic. The patch fixes the problem of checking the right mtd device for partition creation. To delete partition checking is not needed here so as it is done in mtd_del_partition. By master we consider the mtd device which does not belong to any partition. Signed-off-by: Roman Tereshonkov Signed-off-by: Artem Bityutskiy Signed-off-by: David Woodhouse --- drivers/mtd/mtdchar.c | 8 ++++---- drivers/mtd/mtdpart.c | 10 +++++----- include/linux/mtd/partitions.h | 2 +- 3 files changed, 10 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/drivers/mtd/mtdchar.c b/drivers/mtd/mtdchar.c index cad8fcc7b239..16de17b5b829 100644 --- a/drivers/mtd/mtdchar.c +++ b/drivers/mtd/mtdchar.c @@ -522,10 +522,6 @@ static int mtd_blkpg_ioctl(struct mtd_info *mtd, if (!capable(CAP_SYS_ADMIN)) return -EPERM; - /* Only master mtd device must be used to control partitions */ - if (!mtd_is_master(mtd)) - return -EINVAL; - if (copy_from_user(&a, arg, sizeof(struct blkpg_ioctl_arg))) return -EFAULT; @@ -535,6 +531,10 @@ static int mtd_blkpg_ioctl(struct mtd_info *mtd, switch (a.op) { case BLKPG_ADD_PARTITION: + /* Only master mtd device must be used to add partitions */ + if (mtd_is_partition(mtd)) + return -EINVAL; + return mtd_add_partition(mtd, p.devname, p.start, p.length); case BLKPG_DEL_PARTITION: diff --git a/drivers/mtd/mtdpart.c b/drivers/mtd/mtdpart.c index 79e3689f1e16..1047ff0a4f55 100644 --- a/drivers/mtd/mtdpart.c +++ b/drivers/mtd/mtdpart.c @@ -720,19 +720,19 @@ int parse_mtd_partitions(struct mtd_info *master, const char **types, } EXPORT_SYMBOL_GPL(parse_mtd_partitions); -int mtd_is_master(struct mtd_info *mtd) +int mtd_is_partition(struct mtd_info *mtd) { struct mtd_part *part; - int nopart = 0; + int ispart = 0; mutex_lock(&mtd_partitions_mutex); list_for_each_entry(part, &mtd_partitions, list) if (&part->mtd == mtd) { - nopart = 1; + ispart = 1; break; } mutex_unlock(&mtd_partitions_mutex); - return nopart; + return ispart; } -EXPORT_SYMBOL_GPL(mtd_is_master); +EXPORT_SYMBOL_GPL(mtd_is_partition); diff --git a/include/linux/mtd/partitions.h b/include/linux/mtd/partitions.h index 2b54316591d2..4a0a8ba90a72 100644 --- a/include/linux/mtd/partitions.h +++ b/include/linux/mtd/partitions.h @@ -89,7 +89,7 @@ static inline int mtd_has_cmdlinepart(void) { return 1; } static inline int mtd_has_cmdlinepart(void) { return 0; } #endif -int mtd_is_master(struct mtd_info *mtd); +int mtd_is_partition(struct mtd_info *mtd); int mtd_add_partition(struct mtd_info *master, char *name, long long offset, long long length); int mtd_del_partition(struct mtd_info *master, int partno); -- cgit v1.2.3 From 593cd8711221c9661dbf9beb2fb42fecca03e693 Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Mon, 29 Nov 2010 13:52:19 +0100 Subject: mtd: FSMC NAND use the PrimeCell identifier macros The FSMC actually has a standard ARM PrimeCell ID register, and the "revision" part of that register contains the thing that the code is looking at. Reuse the infrastructure from the AMBA bus abstraction and rid local defines. Signed-off-by: Linus Walleij Signed-off-by: David Woodhouse --- drivers/mtd/nand/fsmc_nand.c | 23 ++++++++++++++++++++--- include/linux/mtd/fsmc.h | 19 ------------------- 2 files changed, 20 insertions(+), 22 deletions(-) (limited to 'include/linux') diff --git a/drivers/mtd/nand/fsmc_nand.c b/drivers/mtd/nand/fsmc_nand.c index 5af4b3c122f5..205b10b9f9b9 100644 --- a/drivers/mtd/nand/fsmc_nand.c +++ b/drivers/mtd/nand/fsmc_nand.c @@ -31,6 +31,7 @@ #include #include #include +#include #include static struct nand_ecclayout fsmc_ecc1_layout = { @@ -184,8 +185,9 @@ const char *part_probes[] = { "cmdlinepart", NULL }; #endif /** - * struct fsmc_nand_data - atructure for FSMC NAND device state + * struct fsmc_nand_data - structure for FSMC NAND device state * + * @pid: Part ID on the AMBA PrimeCell format * @mtd: MTD info for a NAND flash. * @nand: Chip related info for a NAND flash. * @partitions: Partition info for a NAND Flash. @@ -201,6 +203,7 @@ const char *part_probes[] = { "cmdlinepart", NULL }; * @regs_va: FSMC regs base address. */ struct fsmc_nand_data { + u32 pid; struct mtd_info mtd; struct nand_chip nand; struct mtd_partition *partitions; @@ -541,6 +544,8 @@ static int __init fsmc_nand_probe(struct platform_device *pdev) struct fsmc_regs *regs; struct resource *res; int ret = 0; + u32 pid; + int i; if (!pdata) { dev_err(&pdev->dev, "platform data is NULL\n"); @@ -630,6 +635,18 @@ static int __init fsmc_nand_probe(struct platform_device *pdev) if (ret) goto err_probe1; + /* + * This device ID is actually a common AMBA ID as used on the + * AMBA PrimeCell bus. However it is not a PrimeCell. + */ + for (pid = 0, i = 0; i < 4; i++) + pid |= (readl(host->regs_va + resource_size(res) - 0x20 + 4 * i) & 255) << (i * 8); + host->pid = pid; + dev_info(&pdev->dev, "FSMC device partno %03x, manufacturer %02x, " + "revision %02x, config %02x\n", + AMBA_PART_BITS(pid), AMBA_MANF_BITS(pid), + AMBA_REV_BITS(pid), AMBA_CONFIG_BITS(pid)); + host->bank = pdata->bank; host->select_chip = pdata->select_bank; regs = host->regs_va; @@ -657,7 +674,7 @@ static int __init fsmc_nand_probe(struct platform_device *pdev) fsmc_nand_setup(regs, host->bank, nand->options & NAND_BUSWIDTH_16); - if (get_fsmc_version(host->regs_va) == FSMC_VER8) { + if (AMBA_REV_BITS(host->pid) >= 8) { nand->ecc.read_page = fsmc_read_page_hwecc; nand->ecc.calculate = fsmc_read_hwecc_ecc4; nand->ecc.correct = fsmc_correct_data; @@ -677,7 +694,7 @@ static int __init fsmc_nand_probe(struct platform_device *pdev) goto err_probe; } - if (get_fsmc_version(host->regs_va) == FSMC_VER8) { + if (AMBA_REV_BITS(host->pid) >= 8) { if (host->mtd.writesize == 512) { nand->ecc.layout = &fsmc_ecc4_sp_layout; host->ecc_place = &fsmc_ecc4_sp_place; diff --git a/include/linux/mtd/fsmc.h b/include/linux/mtd/fsmc.h index e210b87e6cd6..96e8e67a053e 100644 --- a/include/linux/mtd/fsmc.h +++ b/include/linux/mtd/fsmc.h @@ -115,25 +115,6 @@ struct fsmc_regs { #define FSMC_THOLD_4 (4 << 16) #define FSMC_THIZ_1 (1 << 24) -/* peripid2 register definitions */ -#define FSMC_REVISION_MSK (0xf) -#define FSMC_REVISION_SHFT (0x4) - -#define FSMC_VER1 1 -#define FSMC_VER2 2 -#define FSMC_VER3 3 -#define FSMC_VER4 4 -#define FSMC_VER5 5 -#define FSMC_VER6 6 -#define FSMC_VER7 7 -#define FSMC_VER8 8 - -static inline uint32_t get_fsmc_version(struct fsmc_regs *regs) -{ - return (readl(®s->peripid2) >> FSMC_REVISION_SHFT) & - FSMC_REVISION_MSK; -} - /* * There are 13 bytes of ecc for every 512 byte block in FSMC version 8 * and it has to be read consecutively and immediately after the 512 -- cgit v1.2.3 From b5602e86432aaf0cc90dd207bf74e3a2bfb5078b Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Mon, 29 Nov 2010 13:52:27 +0100 Subject: mtd: FSMC NAND fix obvious speling errors Fix spelling in the interface file. Signed-off-by: Linus Walleij Signed-off-by: David Woodhouse --- include/linux/mtd/fsmc.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mtd/fsmc.h b/include/linux/mtd/fsmc.h index 96e8e67a053e..6987995ad3cf 100644 --- a/include/linux/mtd/fsmc.h +++ b/include/linux/mtd/fsmc.h @@ -28,7 +28,7 @@ /* * The placement of the Command Latch Enable (CLE) and - * Address Latch Enable (ALE) is twised around in the + * Address Latch Enable (ALE) is twisted around in the * SPEAR310 implementation. */ #if defined(CONFIG_MACH_SPEAR310) @@ -63,7 +63,7 @@ struct fsmc_nor_bank_regs { /* ctrl_tim register definitions */ -struct fsms_nand_bank_regs { +struct fsmc_nand_bank_regs { uint32_t pc; uint32_t sts; uint32_t comm; @@ -79,7 +79,7 @@ struct fsms_nand_bank_regs { struct fsmc_regs { struct fsmc_nor_bank_regs nor_bank_regs[FSMC_MAX_NOR_BANKS]; uint8_t reserved_1[0x40 - 0x20]; - struct fsms_nand_bank_regs bank_regs[FSMC_MAX_NAND_BANKS]; + struct fsmc_nand_bank_regs bank_regs[FSMC_MAX_NAND_BANKS]; uint8_t reserved_2[0xfe0 - 0xc0]; uint32_t peripid0; /* 0xfe0 */ uint32_t peripid1; /* 0xfe4 */ -- cgit v1.2.3 From c9aa308fd5c373faeda588cfb02b04f116904613 Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Tue, 7 Dec 2010 10:22:29 +0800 Subject: Add CPER PCIe error section structure and constants definition On some machine, PCIe error is reported via APEI (ACPI Platform Error Interface). The error data is passed from firmware to Linux via CPER PCIe error section structure. This patch adds CPER PCIe error section structure and constants definition. Signed-off-by: Huang Ying Signed-off-by: Len Brown --- include/linux/cper.h | 86 +++++++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 82 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/cper.h b/include/linux/cper.h index bf972f81e2a7..3104aaff5dd0 100644 --- a/include/linux/cper.h +++ b/include/linux/cper.h @@ -39,10 +39,12 @@ * Severity difinition for error_severity in struct cper_record_header * and section_severity in struct cper_section_descriptor */ -#define CPER_SEV_RECOVERABLE 0x0 -#define CPER_SEV_FATAL 0x1 -#define CPER_SEV_CORRECTED 0x2 -#define CPER_SEV_INFORMATIONAL 0x3 +enum { + CPER_SEV_RECOVERABLE, + CPER_SEV_FATAL, + CPER_SEV_CORRECTED, + CPER_SEV_INFORMATIONAL, +}; /* * Validation bits difinition for validation_bits in struct @@ -201,6 +203,47 @@ UUID_LE(0x036F84E1, 0x7F37, 0x428c, 0xA7, 0x9E, 0x57, 0x5F, \ 0xDF, 0xAA, 0x84, 0xEC) +#define CPER_PROC_VALID_TYPE 0x0001 +#define CPER_PROC_VALID_ISA 0x0002 +#define CPER_PROC_VALID_ERROR_TYPE 0x0004 +#define CPER_PROC_VALID_OPERATION 0x0008 +#define CPER_PROC_VALID_FLAGS 0x0010 +#define CPER_PROC_VALID_LEVEL 0x0020 +#define CPER_PROC_VALID_VERSION 0x0040 +#define CPER_PROC_VALID_BRAND_INFO 0x0080 +#define CPER_PROC_VALID_ID 0x0100 +#define CPER_PROC_VALID_TARGET_ADDRESS 0x0200 +#define CPER_PROC_VALID_REQUESTOR_ID 0x0400 +#define CPER_PROC_VALID_RESPONDER_ID 0x0800 +#define CPER_PROC_VALID_IP 0x1000 + +#define CPER_MEM_VALID_ERROR_STATUS 0x0001 +#define CPER_MEM_VALID_PHYSICAL_ADDRESS 0x0002 +#define CPER_MEM_VALID_PHYSICAL_ADDRESS_MASK 0x0004 +#define CPER_MEM_VALID_NODE 0x0008 +#define CPER_MEM_VALID_CARD 0x0010 +#define CPER_MEM_VALID_MODULE 0x0020 +#define CPER_MEM_VALID_BANK 0x0040 +#define CPER_MEM_VALID_DEVICE 0x0080 +#define CPER_MEM_VALID_ROW 0x0100 +#define CPER_MEM_VALID_COLUMN 0x0200 +#define CPER_MEM_VALID_BIT_POSITION 0x0400 +#define CPER_MEM_VALID_REQUESTOR_ID 0x0800 +#define CPER_MEM_VALID_RESPONDER_ID 0x1000 +#define CPER_MEM_VALID_TARGET_ID 0x2000 +#define CPER_MEM_VALID_ERROR_TYPE 0x4000 + +#define CPER_PCIE_VALID_PORT_TYPE 0x0001 +#define CPER_PCIE_VALID_VERSION 0x0002 +#define CPER_PCIE_VALID_COMMAND_STATUS 0x0004 +#define CPER_PCIE_VALID_DEVICE_ID 0x0008 +#define CPER_PCIE_VALID_SERIAL_NUMBER 0x0010 +#define CPER_PCIE_VALID_BRIDGE_CONTROL_STATUS 0x0020 +#define CPER_PCIE_VALID_CAPABILITY 0x0040 +#define CPER_PCIE_VALID_AER_INFO 0x0080 + +#define CPER_PCIE_SLOT_SHIFT 3 + /* * All tables and structs must be byte-packed to match CPER * specification, since the tables are provided by the system BIOS @@ -306,6 +349,41 @@ struct cper_sec_mem_err { __u8 error_type; }; +struct cper_sec_pcie { + __u64 validation_bits; + __u32 port_type; + struct { + __u8 minor; + __u8 major; + __u8 reserved[2]; + } version; + __u16 command; + __u16 status; + __u32 reserved; + struct { + __u16 vendor_id; + __u16 device_id; + __u8 class_code[3]; + __u8 function; + __u8 device; + __u16 segment; + __u8 bus; + __u8 secondary_bus; + __u16 slot; + __u8 reserved; + } device_id; + struct { + __u32 lower; + __u32 upper; + } serial_number; + struct { + __u16 secondary_status; + __u16 control; + } bridge; + __u8 capability[60]; + __u8 aer_info[96]; +}; + /* Reset to default packing */ #pragma pack() -- cgit v1.2.3 From 16f4232ce4d6855361b4eb56262f4a202295c978 Mon Sep 17 00:00:00 2001 From: Zhao Yakui Date: Wed, 8 Dec 2010 10:10:16 +0800 Subject: IPMI: Add one interface to get more info of low-level IPMI device The IPMI smi_watcher will be used to catch the IPMI interface as they come or go. In order to communicate with the correct IPMI device, it should be confirmed whether it is what we wanted especially on the system with multiple IPMI devices. But the new_smi callback function of smi_watcher provides very limited info(only the interface number and dev pointer) and there is no detailed info about the low level interface. For example: which mechansim registers the IPMI interface(ACPI, PCI, DMI and so on). This is to add one interface that can get more info of low-level IPMI device. For example: the ACPI device handle will be returned for the pnp_acpi IPMI device. Signed-off-by: Zhao Yakui Signed-off-by: Corey Minyard Signed-off-by: Len Brown --- drivers/char/ipmi/ipmi_msghandler.c | 27 ++++++++++++++++++++++++++ drivers/char/ipmi/ipmi_si_intf.c | 20 +++++++++++++++---- include/linux/ipmi.h | 38 +++++++++++++++++++++++++++++++++++++ include/linux/ipmi_smi.h | 8 ++++++++ 4 files changed, 89 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index 2fe72f8edf44..38223e93aa98 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -970,6 +970,33 @@ out_kfree: } EXPORT_SYMBOL(ipmi_create_user); +int ipmi_get_smi_info(int if_num, struct ipmi_smi_info *data) +{ + int rv = 0; + ipmi_smi_t intf; + struct ipmi_smi_handlers *handlers; + + mutex_lock(&ipmi_interfaces_mutex); + list_for_each_entry_rcu(intf, &ipmi_interfaces, link) { + if (intf->intf_num == if_num) + goto found; + } + /* Not found, return an error */ + rv = -EINVAL; + mutex_unlock(&ipmi_interfaces_mutex); + return rv; + +found: + handlers = intf->handlers; + rv = -ENOSYS; + if (handlers->get_smi_info) + rv = handlers->get_smi_info(intf->send_info, data); + mutex_unlock(&ipmi_interfaces_mutex); + + return rv; +} +EXPORT_SYMBOL(ipmi_get_smi_info); + static void free_user(struct kref *ref) { ipmi_user_t user = container_of(ref, struct ipmi_user, refcount); diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index 035da9e64a17..945ae4d5d21f 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -57,6 +57,7 @@ #include #include #include +#include #include #include #include "ipmi_si_sm.h" @@ -107,10 +108,6 @@ enum si_type { }; static char *si_to_str[] = { "kcs", "smic", "bt" }; -enum ipmi_addr_src { - SI_INVALID = 0, SI_HOTMOD, SI_HARDCODED, SI_SPMI, SI_ACPI, SI_SMBIOS, - SI_PCI, SI_DEVICETREE, SI_DEFAULT -}; static char *ipmi_addr_src_to_str[] = { NULL, "hotmod", "hardcoded", "SPMI", "ACPI", "SMBIOS", "PCI", "device-tree", "default" }; @@ -291,6 +288,7 @@ struct smi_info { struct task_struct *thread; struct list_head link; + union ipmi_smi_info_union addr_info; }; #define smi_inc_stat(smi, stat) \ @@ -1186,6 +1184,18 @@ static int smi_start_processing(void *send_info, return 0; } +static int get_smi_info(void *send_info, struct ipmi_smi_info *data) +{ + struct smi_info *smi = send_info; + + data->addr_src = smi->addr_source; + data->dev = smi->dev; + data->addr_info = smi->addr_info; + get_device(smi->dev); + + return 0; +} + static void set_maintenance_mode(void *send_info, int enable) { struct smi_info *smi_info = send_info; @@ -1197,6 +1207,7 @@ static void set_maintenance_mode(void *send_info, int enable) static struct ipmi_smi_handlers handlers = { .owner = THIS_MODULE, .start_processing = smi_start_processing, + .get_smi_info = get_smi_info, .sender = sender, .request_events = request_events, .set_maintenance_mode = set_maintenance_mode, @@ -2156,6 +2167,7 @@ static int __devinit ipmi_pnp_probe(struct pnp_dev *dev, printk(KERN_INFO PFX "probing via ACPI\n"); handle = acpi_dev->handle; + info->addr_info.acpi_info.acpi_handle = handle; /* _IFT tells us the interface type: KCS, BT, etc */ status = acpi_evaluate_integer(handle, "_IFT", NULL, &tmp); diff --git a/include/linux/ipmi.h b/include/linux/ipmi.h index 65aae34759de..045f2f275cd0 100644 --- a/include/linux/ipmi.h +++ b/include/linux/ipmi.h @@ -454,6 +454,44 @@ unsigned int ipmi_addr_length(int addr_type); /* Validate that the given IPMI address is valid. */ int ipmi_validate_addr(struct ipmi_addr *addr, int len); +/* + * How did the IPMI driver find out about the device? + */ +enum ipmi_addr_src { + SI_INVALID = 0, SI_HOTMOD, SI_HARDCODED, SI_SPMI, SI_ACPI, SI_SMBIOS, + SI_PCI, SI_DEVICETREE, SI_DEFAULT +}; + +union ipmi_smi_info_union { + /* + * the acpi_info element is defined for the SI_ACPI + * address type + */ + struct { + void *acpi_handle; + } acpi_info; +}; + +struct ipmi_smi_info { + enum ipmi_addr_src addr_src; + + /* + * Base device for the interface. Don't forget to put this when + * you are done. + */ + struct device *dev; + + /* + * The addr_info provides more detailed info for some IPMI + * devices, depending on the addr_src. Currently only SI_ACPI + * info is provided. + */ + union ipmi_smi_info_union addr_info; +}; + +/* This is to get the private info of ipmi_smi_t */ +extern int ipmi_get_smi_info(int if_num, struct ipmi_smi_info *data); + #endif /* __KERNEL__ */ diff --git a/include/linux/ipmi_smi.h b/include/linux/ipmi_smi.h index 4b48318ac542..906590aa6907 100644 --- a/include/linux/ipmi_smi.h +++ b/include/linux/ipmi_smi.h @@ -39,6 +39,7 @@ #include #include #include +#include /* This files describes the interface for IPMI system management interface drivers to bind into the IPMI message handler. */ @@ -86,6 +87,13 @@ struct ipmi_smi_handlers { int (*start_processing)(void *send_info, ipmi_smi_t new_intf); + /* + * Get the detailed private info of the low level interface and store + * it into the structure of ipmi_smi_data. For example: the + * ACPI device handle will be returned for the pnp_acpi IPMI device. + */ + int (*get_smi_info)(void *send_info, struct ipmi_smi_info *data); + /* Called to enqueue an SMI message to be sent. This operation is not allowed to fail. If an error occurs, it should report back the error in a received message. It may -- cgit v1.2.3 From 7c96aef75949a56ec427fc6a2522dace2af33605 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 15 Nov 2010 11:27:01 +1100 Subject: sunrpc: remove xpt_pool The xpt_pool field is only used for reporting BUGs. And it isn't used correctly. In particular, when it is cleared in svc_xprt_received before XPT_BUSY is cleared, there is no guarantee that either the compiler or the CPU might not re-order to two assignments, just setting xpt_pool to NULL after XPT_BUSY is cleared. If a different cpu were running svc_xprt_enqueue at this moment, it might see XPT_BUSY clear and then xpt_pool non-NULL, and so BUG. This could be fixed by calling smp_mb__before_clear_bit() before the clear_bit. However as xpt_pool isn't really used, it seems safest to simply remove xpt_pool. Another alternate would be to change the clear_bit to clear_bit_unlock, and the test_and_set_bit to test_and_set_bit_lock. Signed-off-by: NeilBrown Signed-off-by: J. Bruce Fields --- include/linux/sunrpc/svc_xprt.h | 1 - net/sunrpc/svc_xprt.c | 5 ----- 2 files changed, 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/svc_xprt.h b/include/linux/sunrpc/svc_xprt.h index aea0d438e3c7..c8f81da15c7e 100644 --- a/include/linux/sunrpc/svc_xprt.h +++ b/include/linux/sunrpc/svc_xprt.h @@ -63,7 +63,6 @@ struct svc_xprt { #define XPT_LISTENER 11 /* listening endpoint */ #define XPT_CACHE_AUTH 12 /* cache auth info */ - struct svc_pool *xpt_pool; /* current pool iff queued */ struct svc_serv *xpt_server; /* service for transport */ atomic_t xpt_reserved; /* space on outq that is rsvd */ struct mutex xpt_mutex; /* to serialize sending data */ diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index 5a75d23645c8..5eae53b1e306 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -351,8 +351,6 @@ void svc_xprt_enqueue(struct svc_xprt *xprt) dprintk("svc: transport %p busy, not enqueued\n", xprt); goto out_unlock; } - BUG_ON(xprt->xpt_pool != NULL); - xprt->xpt_pool = pool; if (!list_empty(&pool->sp_threads)) { rqstp = list_entry(pool->sp_threads.next, @@ -370,13 +368,11 @@ void svc_xprt_enqueue(struct svc_xprt *xprt) rqstp->rq_reserved = serv->sv_max_mesg; atomic_add(rqstp->rq_reserved, &xprt->xpt_reserved); pool->sp_stats.threads_woken++; - BUG_ON(xprt->xpt_pool != pool); wake_up(&rqstp->rq_wait); } else { dprintk("svc: transport %p put into queue\n", xprt); list_add_tail(&xprt->xpt_ready, &pool->sp_sockets); pool->sp_stats.sockets_queued++; - BUG_ON(xprt->xpt_pool != pool); } out_unlock: @@ -415,7 +411,6 @@ static struct svc_xprt *svc_xprt_dequeue(struct svc_pool *pool) void svc_xprt_received(struct svc_xprt *xprt) { BUG_ON(!test_bit(XPT_BUSY, &xprt->xpt_flags)); - xprt->xpt_pool = NULL; /* As soon as we clear busy, the xprt could be closed and * 'put', so we need a reference to call svc_xprt_enqueue with: */ -- cgit v1.2.3 From 04f4ad16b231abbfde34c762697ad035a3af0b5f Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Thu, 16 Dec 2010 09:51:13 -0500 Subject: nfsd4: implement secinfo_no_name Implementation of this operation is mandatory for NFSv4.1. Signed-off-by: J. Bruce Fields --- fs/nfsd/nfs4proc.c | 27 +++++++++++++++++++++++++++ fs/nfsd/nfs4xdr.c | 15 +++++++++++++-- fs/nfsd/xdr4.h | 5 +++++ include/linux/nfs4.h | 3 +++ 4 files changed, 48 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 095431a35722..f80c3997d24c 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -779,6 +779,29 @@ nfsd4_secinfo(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, return err; } +static __be32 +nfsd4_secinfo_no_name(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, + struct nfsd4_secinfo_no_name *sin) +{ + __be32 err; + + switch (sin->sin_style) { + case NFS4_SECINFO_STYLE4_CURRENT_FH: + break; + case NFS4_SECINFO_STYLE4_PARENT: + err = nfsd4_do_lookupp(rqstp, &cstate->current_fh); + if (err) + return err; + break; + default: + return nfserr_inval; + } + exp_get(cstate->current_fh.fh_export); + sin->sin_exp = cstate->current_fh.fh_export; + fh_put(&cstate->current_fh); + return nfs_ok; +} + static __be32 nfsd4_setattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_setattr *setattr) @@ -1327,6 +1350,10 @@ static struct nfsd4_operation nfsd4_ops[] = { .op_flags = ALLOWED_WITHOUT_FH, .op_name = "OP_RECLAIM_COMPLETE", }, + [OP_SECINFO_NO_NAME] = { + .op_func = (nfsd4op_func)nfsd4_secinfo_no_name, + .op_name = "OP_SECINFO_NO_NAME", + }, }; static const char *nfsd4_op_name(unsigned opnum) diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 71d7d339e44a..b543b2410b54 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -846,6 +846,17 @@ nfsd4_decode_secinfo(struct nfsd4_compoundargs *argp, DECODE_TAIL; } +static __be32 +nfsd4_decode_secinfo_no_name(struct nfsd4_compoundargs *argp, + struct nfsd4_secinfo_no_name *sin) +{ + DECODE_HEAD; + + READ_BUF(4); + READ32(sin->sin_style); + DECODE_TAIL; +} + static __be32 nfsd4_decode_setattr(struct nfsd4_compoundargs *argp, struct nfsd4_setattr *setattr) { @@ -1358,7 +1369,7 @@ static nfsd4_dec nfsd41_dec_ops[] = { [OP_LAYOUTCOMMIT] = (nfsd4_dec)nfsd4_decode_notsupp, [OP_LAYOUTGET] = (nfsd4_dec)nfsd4_decode_notsupp, [OP_LAYOUTRETURN] = (nfsd4_dec)nfsd4_decode_notsupp, - [OP_SECINFO_NO_NAME] = (nfsd4_dec)nfsd4_decode_notsupp, + [OP_SECINFO_NO_NAME] = (nfsd4_dec)nfsd4_decode_secinfo_no_name, [OP_SEQUENCE] = (nfsd4_dec)nfsd4_decode_sequence, [OP_SET_SSV] = (nfsd4_dec)nfsd4_decode_notsupp, [OP_TEST_STATEID] = (nfsd4_dec)nfsd4_decode_notsupp, @@ -3162,7 +3173,7 @@ static nfsd4_enc nfsd4_enc_ops[] = { [OP_LAYOUTCOMMIT] = (nfsd4_enc)nfsd4_encode_noop, [OP_LAYOUTGET] = (nfsd4_enc)nfsd4_encode_noop, [OP_LAYOUTRETURN] = (nfsd4_enc)nfsd4_encode_noop, - [OP_SECINFO_NO_NAME] = (nfsd4_enc)nfsd4_encode_noop, + [OP_SECINFO_NO_NAME] = (nfsd4_enc)nfsd4_encode_secinfo, [OP_SEQUENCE] = (nfsd4_enc)nfsd4_encode_sequence, [OP_SET_SSV] = (nfsd4_enc)nfsd4_encode_noop, [OP_TEST_STATEID] = (nfsd4_enc)nfsd4_encode_noop, diff --git a/fs/nfsd/xdr4.h b/fs/nfsd/xdr4.h index 60fce3dc5cb5..799c30c3b495 100644 --- a/fs/nfsd/xdr4.h +++ b/fs/nfsd/xdr4.h @@ -311,6 +311,11 @@ struct nfsd4_secinfo { struct svc_export *si_exp; /* response */ }; +struct nfsd4_secinfo_no_name { + u32 sin_style; /* request */ + struct svc_export *sin_exp; /* response */ +}; + struct nfsd4_setattr { stateid_t sa_stateid; /* request */ u32 sa_bmval[3]; /* request */ diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h index 4925b22219d2..26afa3021ed6 100644 --- a/include/linux/nfs4.h +++ b/include/linux/nfs4.h @@ -136,6 +136,9 @@ #define SEQ4_STATUS_CB_PATH_DOWN_SESSION 0x00000200 #define SEQ4_STATUS_BACKCHANNEL_FAULT 0x00000400 +#define NFS4_SECINFO_STYLE4_CURRENT_FH 0 +#define NFS4_SECINFO_STYLE4_PARENT 1 + #define NFS4_MAX_UINT64 (~(u64)0) /* An NFS4 sessions server must support at least NFS4_MAX_OPS operations. -- cgit v1.2.3 From c66ae9bb4dcaac78cc5e30d0ce7ff2bf3dcb48d9 Mon Sep 17 00:00:00 2001 From: Vasily Khoruzhick Date: Mon, 13 Dec 2010 12:26:21 +0200 Subject: s3c_adc_battery: Add gpio_inverted field to pdata Add support for inverted gpio_charge_finished values. This change is necessary for H1940 support. Signed-off-by: Vasily Khoruzhick Signed-off-by: Anton Vorontsov --- drivers/power/s3c_adc_battery.c | 12 +++++++++--- include/linux/s3c_adc_battery.h | 1 + 2 files changed, 10 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/drivers/power/s3c_adc_battery.c b/drivers/power/s3c_adc_battery.c index fe16b482e912..7bc5bfe55ce9 100644 --- a/drivers/power/s3c_adc_battery.c +++ b/drivers/power/s3c_adc_battery.c @@ -112,6 +112,13 @@ static int calc_full_volt(int volt_val, int cur_val, int impedance) return volt_val + cur_val * impedance / 1000; } +static int charge_finished(struct s3c_adc_bat *bat) +{ + return bat->pdata->gpio_inverted ? + !gpio_get_value(bat->pdata->gpio_charge_finished) : + gpio_get_value(bat->pdata->gpio_charge_finished); +} + static int s3c_adc_bat_get_property(struct power_supply *psy, enum power_supply_property psp, union power_supply_propval *val) @@ -140,7 +147,7 @@ static int s3c_adc_bat_get_property(struct power_supply *psy, if (bat->cable_plugged && ((bat->pdata->gpio_charge_finished < 0) || - !gpio_get_value(bat->pdata->gpio_charge_finished))) { + !charge_finished(bat))) { lut = bat->pdata->lut_acin; lut_size = bat->pdata->lut_acin_cnt; } @@ -236,8 +243,7 @@ static void s3c_adc_bat_work(struct work_struct *work) } } else { if ((bat->pdata->gpio_charge_finished >= 0) && is_plugged) { - is_charged = gpio_get_value( - main_bat.pdata->gpio_charge_finished); + is_charged = charge_finished(&main_bat); if (is_charged) { if (bat->pdata->disable_charger) bat->pdata->disable_charger(); diff --git a/include/linux/s3c_adc_battery.h b/include/linux/s3c_adc_battery.h index dbce22faa660..fbe58b7e63eb 100644 --- a/include/linux/s3c_adc_battery.h +++ b/include/linux/s3c_adc_battery.h @@ -14,6 +14,7 @@ struct s3c_adc_bat_pdata { void (*disable_charger)(void); int gpio_charge_finished; + int gpio_inverted; const struct s3c_adc_bat_thresh *lut_noac; unsigned int lut_noac_cnt; -- cgit v1.2.3 From 00aaaef9a51a1a25c5d6d52ce510772f149a0eb0 Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Thu, 11 Nov 2010 15:46:54 +0800 Subject: PCI: MSI: Move MSI-X entry definition to pci_regs.h Then it can be used by others. Reviewed-by: Hidetoshi Seto Reviewed-by: Matthew Wilcox Signed-off-by: Sheng Yang Signed-off-by: Jesse Barnes --- drivers/pci/msi.h | 6 ------ include/linux/pci_regs.h | 7 +++++++ 2 files changed, 7 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/drivers/pci/msi.h b/drivers/pci/msi.h index feff3bee6fe5..65c42f80f23e 100644 --- a/drivers/pci/msi.h +++ b/drivers/pci/msi.h @@ -6,12 +6,6 @@ #ifndef MSI_H #define MSI_H -#define PCI_MSIX_ENTRY_SIZE 16 -#define PCI_MSIX_ENTRY_LOWER_ADDR 0 -#define PCI_MSIX_ENTRY_UPPER_ADDR 4 -#define PCI_MSIX_ENTRY_DATA 8 -#define PCI_MSIX_ENTRY_VECTOR_CTRL 12 - #define msi_control_reg(base) (base + PCI_MSI_FLAGS) #define msi_lower_address_reg(base) (base + PCI_MSI_ADDRESS_LO) #define msi_upper_address_reg(base) (base + PCI_MSI_ADDRESS_HI) diff --git a/include/linux/pci_regs.h b/include/linux/pci_regs.h index af83076c31a6..b21d33e4bc2a 100644 --- a/include/linux/pci_regs.h +++ b/include/linux/pci_regs.h @@ -309,6 +309,13 @@ #define PCI_MSIX_PBA 8 #define PCI_MSIX_FLAGS_BIRMASK (7 << 0) +/* MSI-X entry's format */ +#define PCI_MSIX_ENTRY_SIZE 16 +#define PCI_MSIX_ENTRY_LOWER_ADDR 0 +#define PCI_MSIX_ENTRY_UPPER_ADDR 4 +#define PCI_MSIX_ENTRY_DATA 8 +#define PCI_MSIX_ENTRY_VECTOR_CTRL 12 + /* CompactPCI Hotswap Register */ #define PCI_CHSWP_CSR 2 /* Control and Status Register */ -- cgit v1.2.3 From 8d805286968811223cca002134ba3d81244d5313 Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Thu, 11 Nov 2010 15:46:55 +0800 Subject: PCI: Add mask bit definition for MSI-X table Then we can use it instead of magic number 1. Reviewed-by: Hidetoshi Seto Cc: Matthew Wilcox Signed-off-by: Sheng Yang Signed-off-by: Jesse Barnes --- drivers/pci/msi.c | 5 +++-- include/linux/pci_regs.h | 1 + 2 files changed, 4 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c index 7c24dcef2989..44b0aeee83e5 100644 --- a/drivers/pci/msi.c +++ b/drivers/pci/msi.c @@ -168,8 +168,9 @@ static u32 __msix_mask_irq(struct msi_desc *desc, u32 flag) u32 mask_bits = desc->masked; unsigned offset = desc->msi_attrib.entry_nr * PCI_MSIX_ENTRY_SIZE + PCI_MSIX_ENTRY_VECTOR_CTRL; - mask_bits &= ~1; - mask_bits |= flag; + mask_bits &= ~PCI_MSIX_ENTRY_CTRL_MASKBIT; + if (flag) + mask_bits |= PCI_MSIX_ENTRY_CTRL_MASKBIT; writel(mask_bits, desc->mask_base + offset); return mask_bits; diff --git a/include/linux/pci_regs.h b/include/linux/pci_regs.h index b21d33e4bc2a..d4f2c80a6c3e 100644 --- a/include/linux/pci_regs.h +++ b/include/linux/pci_regs.h @@ -315,6 +315,7 @@ #define PCI_MSIX_ENTRY_UPPER_ADDR 4 #define PCI_MSIX_ENTRY_DATA 8 #define PCI_MSIX_ENTRY_VECTOR_CTRL 12 +#define PCI_MSIX_ENTRY_CTRL_MASKBIT 1 /* CompactPCI Hotswap Register */ -- cgit v1.2.3 From 2f671e2dbff6eb5ef4e2600adbec550c13b8fe72 Mon Sep 17 00:00:00 2001 From: Matthew Garrett Date: Mon, 6 Dec 2010 14:00:56 -0500 Subject: PCI: Disable ASPM if BIOS asks us to We currently refuse to touch the ASPM registers if the BIOS tells us that ASPM isn't supported. This can cause problems if the BIOS has (for any reason) enabled ASPM on some devices anyway. Change the code such that we explicitly clear ASPM if the FADT indicates that ASPM isn't supported, and make sure we tidy up appropriately on device removal in order to deal with the hotplug case. If ASPM is disabled because the BIOS doesn't hand over control then we won't touch the registers. Signed-off-by: Matthew Garrett Signed-off-by: Jesse Barnes --- drivers/pci/pci-acpi.c | 1 + drivers/pci/pcie/aspm.c | 21 +++++++++++++++++---- include/linux/pci-aspm.h | 5 ++++- 3 files changed, 22 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/drivers/pci/pci-acpi.c b/drivers/pci/pci-acpi.c index 24e19c594e57..d7ea699b8ddc 100644 --- a/drivers/pci/pci-acpi.c +++ b/drivers/pci/pci-acpi.c @@ -399,6 +399,7 @@ static int __init acpi_pci_init(void) if (acpi_gbl_FADT.boot_flags & ACPI_FADT_NO_ASPM) { printk(KERN_INFO"ACPI FADT declares the system doesn't support PCIe ASPM, so disable it\n"); + pcie_clear_aspm(); pcie_no_aspm(); } diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c index 71222814c1ec..3188cd96b338 100644 --- a/drivers/pci/pcie/aspm.c +++ b/drivers/pci/pcie/aspm.c @@ -68,7 +68,7 @@ struct pcie_link_state { struct aspm_latency acceptable[8]; }; -static int aspm_disabled, aspm_force; +static int aspm_disabled, aspm_force, aspm_clear_state; static DEFINE_MUTEX(aspm_lock); static LIST_HEAD(link_list); @@ -139,7 +139,7 @@ static void pcie_set_clkpm(struct pcie_link_state *link, int enable) { /* Don't enable Clock PM if the link is not Clock PM capable */ if (!link->clkpm_capable && enable) - return; + enable = 0; /* Need nothing if the specified equals to current state */ if (link->clkpm_enabled == enable) return; @@ -498,6 +498,10 @@ static int pcie_aspm_sanity_check(struct pci_dev *pdev) struct pci_dev *child; int pos; u32 reg32; + + if (aspm_clear_state) + return -EINVAL; + /* * Some functions in a slot might not all be PCIe functions, * very strange. Disable ASPM for the whole slot @@ -563,12 +567,15 @@ void pcie_aspm_init_link_state(struct pci_dev *pdev) struct pcie_link_state *link; int blacklist = !!pcie_aspm_sanity_check(pdev); - if (aspm_disabled || !pci_is_pcie(pdev) || pdev->link_state) + if (!pci_is_pcie(pdev) || pdev->link_state) return; if (pdev->pcie_type != PCI_EXP_TYPE_ROOT_PORT && pdev->pcie_type != PCI_EXP_TYPE_DOWNSTREAM) return; + if (aspm_disabled && !aspm_clear_state) + return; + /* VIA has a strange chipset, root port is under a bridge */ if (pdev->pcie_type == PCI_EXP_TYPE_ROOT_PORT && pdev->bus->self) @@ -641,7 +648,7 @@ void pcie_aspm_exit_link_state(struct pci_dev *pdev) struct pci_dev *parent = pdev->bus->self; struct pcie_link_state *link, *root, *parent_link; - if (aspm_disabled || !pci_is_pcie(pdev) || + if ((aspm_disabled && !aspm_clear_state) || !pci_is_pcie(pdev) || !parent || !parent->link_state) return; if ((parent->pcie_type != PCI_EXP_TYPE_ROOT_PORT) && @@ -899,6 +906,12 @@ static int __init pcie_aspm_disable(char *str) __setup("pcie_aspm=", pcie_aspm_disable); +void pcie_clear_aspm(void) +{ + if (!aspm_force) + aspm_clear_state = 1; +} + void pcie_no_aspm(void) { if (!aspm_force) diff --git a/include/linux/pci-aspm.h b/include/linux/pci-aspm.h index 91ba0b338b47..ce6810512c66 100644 --- a/include/linux/pci-aspm.h +++ b/include/linux/pci-aspm.h @@ -27,6 +27,7 @@ extern void pcie_aspm_init_link_state(struct pci_dev *pdev); extern void pcie_aspm_exit_link_state(struct pci_dev *pdev); extern void pcie_aspm_pm_state_change(struct pci_dev *pdev); extern void pci_disable_link_state(struct pci_dev *pdev, int state); +extern void pcie_clear_aspm(void); extern void pcie_no_aspm(void); #else static inline void pcie_aspm_init_link_state(struct pci_dev *pdev) @@ -41,7 +42,9 @@ static inline void pcie_aspm_pm_state_change(struct pci_dev *pdev) static inline void pci_disable_link_state(struct pci_dev *pdev, int state) { } - +static inline void pcie_clear_aspm(void) +{ +} static inline void pcie_no_aspm(void) { } -- cgit v1.2.3 From 1d3c16a818e992c199844954d95c17fd7ce6cbba Mon Sep 17 00:00:00 2001 From: Jon Mason Date: Tue, 30 Nov 2010 17:43:26 -0600 Subject: PCI: make pci_restore_state return void pci_restore_state only ever returns 0, thus there is no benefit in having it return any value. Also, a large majority of the callers do not check the return code of pci_restore_state. Make the pci_restore_state a void return and avoid the overhead. Acked-by: Mauro Carvalho Chehab Signed-off-by: Jon Mason Signed-off-by: Jesse Barnes --- drivers/media/video/cafe_ccic.c | 4 +--- drivers/net/myri10ge/myri10ge.c | 4 +--- drivers/net/sfc/falcon.c | 25 +++++-------------------- drivers/net/skge.c | 4 +--- drivers/net/sky2.c | 5 +---- drivers/net/wireless/rt2x00/rt2x00pci.c | 4 ++-- drivers/pci/pci-driver.c | 3 ++- drivers/pci/pci.c | 7 ++----- drivers/scsi/ipr.c | 8 +------- drivers/scsi/pmcraid.c | 7 +------ drivers/staging/sm7xx/smtcfb.c | 2 +- include/linux/pci.h | 8 +++----- sound/pci/cs5535audio/cs5535audio_pm.c | 7 +------ 13 files changed, 22 insertions(+), 66 deletions(-) (limited to 'include/linux') diff --git a/drivers/media/video/cafe_ccic.c b/drivers/media/video/cafe_ccic.c index 0dfff50891e4..737bb877e962 100644 --- a/drivers/media/video/cafe_ccic.c +++ b/drivers/media/video/cafe_ccic.c @@ -2186,9 +2186,7 @@ static int cafe_pci_resume(struct pci_dev *pdev) struct cafe_camera *cam = to_cam(v4l2_dev); int ret = 0; - ret = pci_restore_state(pdev); - if (ret) - return ret; + pci_restore_state(pdev); ret = pci_enable_device(pdev); if (ret) { diff --git a/drivers/net/myri10ge/myri10ge.c b/drivers/net/myri10ge/myri10ge.c index 8524cc40ec57..d3c4a374a92e 100644 --- a/drivers/net/myri10ge/myri10ge.c +++ b/drivers/net/myri10ge/myri10ge.c @@ -3403,9 +3403,7 @@ static int myri10ge_resume(struct pci_dev *pdev) return -EIO; } - status = pci_restore_state(pdev); - if (status) - return status; + pci_restore_state(pdev); status = pci_enable_device(pdev); if (status) { diff --git a/drivers/net/sfc/falcon.c b/drivers/net/sfc/falcon.c index 267019bb2b15..1763b9a7fd8e 100644 --- a/drivers/net/sfc/falcon.c +++ b/drivers/net/sfc/falcon.c @@ -1066,22 +1066,9 @@ static int falcon_reset_hw(struct efx_nic *efx, enum reset_type method) /* Restore PCI configuration if needed */ if (method == RESET_TYPE_WORLD) { - if (efx_nic_is_dual_func(efx)) { - rc = pci_restore_state(nic_data->pci_dev2); - if (rc) { - netif_err(efx, drv, efx->net_dev, - "failed to restore PCI config for " - "the secondary function\n"); - goto fail3; - } - } - rc = pci_restore_state(efx->pci_dev); - if (rc) { - netif_err(efx, drv, efx->net_dev, - "failed to restore PCI config for the " - "primary function\n"); - goto fail4; - } + if (efx_nic_is_dual_func(efx)) + pci_restore_state(nic_data->pci_dev2); + pci_restore_state(efx->pci_dev); netif_dbg(efx, drv, efx->net_dev, "successfully restored PCI config\n"); } @@ -1092,7 +1079,7 @@ static int falcon_reset_hw(struct efx_nic *efx, enum reset_type method) rc = -ETIMEDOUT; netif_err(efx, hw, efx->net_dev, "timed out waiting for hardware reset\n"); - goto fail5; + goto fail3; } netif_dbg(efx, hw, efx->net_dev, "hardware reset complete\n"); @@ -1100,11 +1087,9 @@ static int falcon_reset_hw(struct efx_nic *efx, enum reset_type method) /* pci_save_state() and pci_restore_state() MUST be called in pairs */ fail2: -fail3: pci_restore_state(efx->pci_dev); fail1: -fail4: -fail5: +fail3: return rc; } diff --git a/drivers/net/skge.c b/drivers/net/skge.c index 220e0398f1d5..61553af38e65 100644 --- a/drivers/net/skge.c +++ b/drivers/net/skge.c @@ -4087,9 +4087,7 @@ static int skge_resume(struct pci_dev *pdev) if (err) goto out; - err = pci_restore_state(pdev); - if (err) - goto out; + pci_restore_state(pdev); err = skge_reset(hw); if (err) diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index d6577084ce70..be3aee782760 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -4969,10 +4969,7 @@ static int sky2_resume(struct pci_dev *pdev) if (err) goto out; - err = pci_restore_state(pdev); - if (err) - goto out; - + pci_restore_state(pdev); pci_enable_wake(pdev, PCI_D0, 0); /* Re-enable all clocks */ diff --git a/drivers/net/wireless/rt2x00/rt2x00pci.c b/drivers/net/wireless/rt2x00/rt2x00pci.c index 2449d785cf8d..4fd4c33de6ae 100644 --- a/drivers/net/wireless/rt2x00/rt2x00pci.c +++ b/drivers/net/wireless/rt2x00/rt2x00pci.c @@ -356,12 +356,12 @@ int rt2x00pci_resume(struct pci_dev *pci_dev) struct rt2x00_dev *rt2x00dev = hw->priv; if (pci_set_power_state(pci_dev, PCI_D0) || - pci_enable_device(pci_dev) || - pci_restore_state(pci_dev)) { + pci_enable_device(pci_dev)) { ERROR(rt2x00dev, "Failed to resume device.\n"); return -EIO; } + pci_restore_state(pci_dev); return rt2x00lib_resume(rt2x00dev); } EXPORT_SYMBOL_GPL(rt2x00pci_resume); diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index 8a6f797de8e5..80e551ee640b 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -449,7 +449,8 @@ static int pci_restore_standard_config(struct pci_dev *pci_dev) return error; } - return pci_restore_state(pci_dev); + pci_restore_state(pci_dev); + return 0; } static void pci_pm_default_resume_early(struct pci_dev *pci_dev) diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 710c8a29be0d..6762dcae90ab 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -937,14 +937,13 @@ pci_save_state(struct pci_dev *dev) * pci_restore_state - Restore the saved state of a PCI device * @dev: - PCI device that we're dealing with */ -int -pci_restore_state(struct pci_dev *dev) +void pci_restore_state(struct pci_dev *dev) { int i; u32 val; if (!dev->state_saved) - return 0; + return; /* PCI Express register must be restored first */ pci_restore_pcie_state(dev); @@ -968,8 +967,6 @@ pci_restore_state(struct pci_dev *dev) pci_restore_iov_state(dev); dev->state_saved = false; - - return 0; } static int do_pci_enable_device(struct pci_dev *dev, int bars) diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c index 5bbaee597e88..524d586c3147 100644 --- a/drivers/scsi/ipr.c +++ b/drivers/scsi/ipr.c @@ -7487,16 +7487,10 @@ static int ipr_reset_restore_cfg_space(struct ipr_cmnd *ipr_cmd) { struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg; volatile u32 int_reg; - int rc; ENTER; ioa_cfg->pdev->state_saved = true; - rc = pci_restore_state(ioa_cfg->pdev); - - if (rc != PCIBIOS_SUCCESSFUL) { - ipr_cmd->s.ioasa.hdr.ioasc = cpu_to_be32(IPR_IOASC_PCI_ACCESS_ERROR); - return IPR_RC_JOB_CONTINUE; - } + pci_restore_state(ioa_cfg->pdev); if (ipr_set_pcix_cmd_reg(ioa_cfg)) { ipr_cmd->s.ioasa.hdr.ioasc = cpu_to_be32(IPR_IOASC_PCI_ACCESS_ERROR); diff --git a/drivers/scsi/pmcraid.c b/drivers/scsi/pmcraid.c index 300d59f389da..321cf3ae8630 100644 --- a/drivers/scsi/pmcraid.c +++ b/drivers/scsi/pmcraid.c @@ -2228,12 +2228,7 @@ static void pmcraid_ioa_reset(struct pmcraid_cmd *cmd) /* Once either bist or pci reset is done, restore PCI config * space. If this fails, proceed with hard reset again */ - if (pci_restore_state(pinstance->pdev)) { - pmcraid_info("config-space error resetting again\n"); - pinstance->ioa_state = IOA_STATE_IN_RESET_ALERT; - pmcraid_reset_alert(cmd); - break; - } + pci_restore_state(pinstance->pdev); /* fail all pending commands */ pmcraid_fail_outstanding_cmds(pinstance); diff --git a/drivers/staging/sm7xx/smtcfb.c b/drivers/staging/sm7xx/smtcfb.c index 24f47d6388f4..7162dee3b0d8 100644 --- a/drivers/staging/sm7xx/smtcfb.c +++ b/drivers/staging/sm7xx/smtcfb.c @@ -1071,7 +1071,7 @@ static int __maybe_unused smtcfb_resume(struct pci_dev *pdev) /* when resuming, restore pci data and fb cursor */ if (pdev->dev.power.power_state.event != PM_EVENT_FREEZE) { retv = pci_set_power_state(pdev, PCI_D0); - retv = pci_restore_state(pdev); + pci_restore_state(pdev); if (pci_enable_device(pdev)) return -1; pci_set_master(pdev); diff --git a/include/linux/pci.h b/include/linux/pci.h index 7454408c41b6..63cbadce337e 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -806,7 +806,7 @@ size_t pci_get_rom_size(struct pci_dev *pdev, void __iomem *rom, size_t size); /* Power management related routines */ int pci_save_state(struct pci_dev *dev); -int pci_restore_state(struct pci_dev *dev); +void pci_restore_state(struct pci_dev *dev); int __pci_complete_power_transition(struct pci_dev *dev, pci_power_t state); int pci_set_power_state(struct pci_dev *dev, pci_power_t state); pci_power_t pci_choose_state(struct pci_dev *dev, pm_message_t state); @@ -1168,10 +1168,8 @@ static inline int pci_save_state(struct pci_dev *dev) return 0; } -static inline int pci_restore_state(struct pci_dev *dev) -{ - return 0; -} +static inline void pci_restore_state(struct pci_dev *dev) +{ } static inline int pci_set_power_state(struct pci_dev *dev, pci_power_t state) { diff --git a/sound/pci/cs5535audio/cs5535audio_pm.c b/sound/pci/cs5535audio/cs5535audio_pm.c index a3301cc4ab82..185b00088320 100644 --- a/sound/pci/cs5535audio/cs5535audio_pm.c +++ b/sound/pci/cs5535audio/cs5535audio_pm.c @@ -90,12 +90,7 @@ int snd_cs5535audio_resume(struct pci_dev *pci) int i; pci_set_power_state(pci, PCI_D0); - if (pci_restore_state(pci) < 0) { - printk(KERN_ERR "cs5535audio: pci_restore_state failed, " - "disabling device\n"); - snd_card_disconnect(card); - return -EIO; - } + pci_restore_state(pci); if (pci_enable_device(pci) < 0) { printk(KERN_ERR "cs5535audio: pci_enable_device failed, " "disabling device\n"); -- cgit v1.2.3 From 9b444b36fee16d2aaae9cc91ce594ecb15d922a9 Mon Sep 17 00:00:00 2001 From: Seth Heasley Date: Wed, 17 Nov 2010 12:12:08 -0700 Subject: x86/PCI: irq and pci_ids patch for Intel Patsburg This patch adds an additional LPC Controller DeviceID for the Intel Patsburg PCH. Signed-off-by: Seth Heasley Signed-off-by: Jesse Barnes --- arch/x86/pci/irq.c | 3 ++- include/linux/pci_ids.h | 3 ++- 2 files changed, 4 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/arch/x86/pci/irq.c b/arch/x86/pci/irq.c index 9f9bfb705cf9..87e6c8323117 100644 --- a/arch/x86/pci/irq.c +++ b/arch/x86/pci/irq.c @@ -589,7 +589,8 @@ static __init int intel_router_probe(struct irq_router *r, struct pci_dev *route case PCI_DEVICE_ID_INTEL_ICH10_1: case PCI_DEVICE_ID_INTEL_ICH10_2: case PCI_DEVICE_ID_INTEL_ICH10_3: - case PCI_DEVICE_ID_INTEL_PATSBURG_LPC: + case PCI_DEVICE_ID_INTEL_PATSBURG_LPC_0: + case PCI_DEVICE_ID_INTEL_PATSBURG_LPC_1: r->name = "PIIX/ICH"; r->get = pirq_piix_get; r->set = pirq_piix_set; diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index cb845c16ad7d..d830106827fd 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2468,7 +2468,8 @@ #define PCI_DEVICE_ID_INTEL_COUGARPOINT_LPC_MIN 0x1c41 #define PCI_DEVICE_ID_INTEL_COUGARPOINT_LPC_MAX 0x1c5f #define PCI_DEVICE_ID_INTEL_PATSBURG_SMBUS 0x1d22 -#define PCI_DEVICE_ID_INTEL_PATSBURG_LPC 0x1d40 +#define PCI_DEVICE_ID_INTEL_PATSBURG_LPC_0 0x1d40 +#define PCI_DEVICE_ID_INTEL_PATSBURG_LPC_1 0x1d41 #define PCI_DEVICE_ID_INTEL_82801AA_0 0x2410 #define PCI_DEVICE_ID_INTEL_82801AA_1 0x2411 #define PCI_DEVICE_ID_INTEL_82801AA_3 0x2413 -- cgit v1.2.3 From fe31e69740eddc7316071ed5165fed6703c8cd12 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sun, 19 Dec 2010 15:57:16 +0100 Subject: PCI/PCIe: Clear Root PME Status bits early during system resume I noticed that PCI Express PMEs don't work on my Toshiba Portege R500 after the system has been woken up from a sleep state by a PME (through Wake-on-LAN). After some investigation it turned out that the BIOS didn't clear the Root PME Status bit in the root port that received the wakeup PME and since the Requester ID was also set in the port's Root Status register, any subsequent PMEs didn't trigger interrupts. This problem can be avoided by clearing the Root PME Status bits in all PCI Express root ports during early resume. For this purpose, add an early resume routine to the PCIe port driver and make this driver be always registered, even if pci_ports_disable is set (in which case the driver's only function is to provide the early resume callback). Signed-off-by: Rafael J. Wysocki Signed-off-by: Jesse Barnes --- drivers/pci/pcie/pme.c | 27 ++++----------------------- drivers/pci/pcie/portdrv.h | 2 ++ drivers/pci/pcie/portdrv_core.c | 25 ++++++++++++++----------- drivers/pci/pcie/portdrv_pci.c | 37 +++++++++++++++++++++++++++++++++---- include/linux/pci_regs.h | 2 ++ 5 files changed, 55 insertions(+), 38 deletions(-) (limited to 'include/linux') diff --git a/drivers/pci/pcie/pme.c b/drivers/pci/pcie/pme.c index 2f3c90407227..073f0308c6b2 100644 --- a/drivers/pci/pcie/pme.c +++ b/drivers/pci/pcie/pme.c @@ -26,9 +26,6 @@ #include "../pci.h" #include "portdrv.h" -#define PCI_EXP_RTSTA_PME 0x10000 /* PME status */ -#define PCI_EXP_RTSTA_PENDING 0x20000 /* PME pending */ - /* * If this switch is set, MSI will not be used for PCIe PME signaling. This * causes the PCIe port driver to use INTx interrupts only, but it turns out @@ -73,22 +70,6 @@ void pcie_pme_interrupt_enable(struct pci_dev *dev, bool enable) pci_write_config_word(dev, rtctl_pos, rtctl); } -/** - * pcie_pme_clear_status - Clear root port PME interrupt status. - * @dev: PCIe root port or event collector. - */ -static void pcie_pme_clear_status(struct pci_dev *dev) -{ - int rtsta_pos; - u32 rtsta; - - rtsta_pos = pci_pcie_cap(dev) + PCI_EXP_RTSTA; - - pci_read_config_dword(dev, rtsta_pos, &rtsta); - rtsta |= PCI_EXP_RTSTA_PME; - pci_write_config_dword(dev, rtsta_pos, rtsta); -} - /** * pcie_pme_walk_bus - Scan a PCI bus for devices asserting PME#. * @bus: PCI bus to scan. @@ -253,7 +234,7 @@ static void pcie_pme_work_fn(struct work_struct *work) * Clear PME status of the port. If there are other * pending PMEs, the status will be set again. */ - pcie_pme_clear_status(port); + pcie_clear_root_pme_status(port); spin_unlock_irq(&data->lock); pcie_pme_handle_request(port, rtsta & 0xffff); @@ -378,7 +359,7 @@ static int pcie_pme_probe(struct pcie_device *srv) port = srv->port; pcie_pme_interrupt_enable(port, false); - pcie_pme_clear_status(port); + pcie_clear_root_pme_status(port); ret = request_irq(srv->irq, pcie_pme_irq, IRQF_SHARED, "PCIe PME", srv); if (ret) { @@ -402,7 +383,7 @@ static int pcie_pme_suspend(struct pcie_device *srv) spin_lock_irq(&data->lock); pcie_pme_interrupt_enable(port, false); - pcie_pme_clear_status(port); + pcie_clear_root_pme_status(port); data->noirq = true; spin_unlock_irq(&data->lock); @@ -422,7 +403,7 @@ static int pcie_pme_resume(struct pcie_device *srv) spin_lock_irq(&data->lock); data->noirq = false; - pcie_pme_clear_status(port); + pcie_clear_root_pme_status(port); pcie_pme_interrupt_enable(port, true); spin_unlock_irq(&data->lock); diff --git a/drivers/pci/pcie/portdrv.h b/drivers/pci/pcie/portdrv.h index 7b5aba0a3291..8fcc03598b29 100644 --- a/drivers/pci/pcie/portdrv.h +++ b/drivers/pci/pcie/portdrv.h @@ -35,6 +35,8 @@ extern void pcie_port_bus_unregister(void); struct pci_dev; +extern void pcie_clear_root_pme_status(struct pci_dev *dev); + #ifdef CONFIG_PCIE_PME extern bool pcie_pme_msi_disabled; diff --git a/drivers/pci/pcie/portdrv_core.c b/drivers/pci/pcie/portdrv_core.c index a9c222d79ebc..5130d0d22390 100644 --- a/drivers/pci/pcie/portdrv_core.c +++ b/drivers/pci/pcie/portdrv_core.c @@ -241,17 +241,17 @@ static int get_port_device_capability(struct pci_dev *dev) int cap_mask; int err; + if (pcie_ports_disabled) + return 0; + err = pcie_port_platform_notify(dev, &cap_mask); - if (pcie_ports_auto) { - if (err) { - pcie_no_aspm(); - return 0; - } - } else { + if (!pcie_ports_auto) { cap_mask = PCIE_PORT_SERVICE_PME | PCIE_PORT_SERVICE_HP | PCIE_PORT_SERVICE_VC; if (pci_aer_available()) cap_mask |= PCIE_PORT_SERVICE_AER; + } else if (err) { + return 0; } pos = pci_pcie_cap(dev); @@ -349,15 +349,18 @@ int pcie_port_device_register(struct pci_dev *dev) int status, capabilities, i, nr_service; int irqs[PCIE_PORT_DEVICE_MAXSERVICES]; - /* Get and check PCI Express port services */ - capabilities = get_port_device_capability(dev); - if (!capabilities) - return -ENODEV; - /* Enable PCI Express port device */ status = pci_enable_device(dev); if (status) return status; + + /* Get and check PCI Express port services */ + capabilities = get_port_device_capability(dev); + if (!capabilities) { + pcie_no_aspm(); + return 0; + } + pci_set_master(dev); /* * Initialize service irqs. Don't use service devices that diff --git a/drivers/pci/pcie/portdrv_pci.c b/drivers/pci/pcie/portdrv_pci.c index f9033e190fb6..e0610bda1dea 100644 --- a/drivers/pci/pcie/portdrv_pci.c +++ b/drivers/pci/pcie/portdrv_pci.c @@ -57,6 +57,22 @@ __setup("pcie_ports=", pcie_port_setup); /* global data */ +/** + * pcie_clear_root_pme_status - Clear root port PME interrupt status. + * @dev: PCIe root port or event collector. + */ +void pcie_clear_root_pme_status(struct pci_dev *dev) +{ + int rtsta_pos; + u32 rtsta; + + rtsta_pos = pci_pcie_cap(dev) + PCI_EXP_RTSTA; + + pci_read_config_dword(dev, rtsta_pos, &rtsta); + rtsta |= PCI_EXP_RTSTA_PME; + pci_write_config_dword(dev, rtsta_pos, rtsta); +} + static int pcie_portdrv_restore_config(struct pci_dev *dev) { int retval; @@ -69,6 +85,20 @@ static int pcie_portdrv_restore_config(struct pci_dev *dev) } #ifdef CONFIG_PM +static int pcie_port_resume_noirq(struct device *dev) +{ + struct pci_dev *pdev = to_pci_dev(dev); + + /* + * Some BIOSes forget to clear Root PME Status bits after system wakeup + * which breaks ACPI-based runtime wakeup on PCI Express, so clear those + * bits now just in case (shouldn't hurt). + */ + if(pdev->pcie_type == PCI_EXP_TYPE_ROOT_PORT) + pcie_clear_root_pme_status(pdev); + return 0; +} + static const struct dev_pm_ops pcie_portdrv_pm_ops = { .suspend = pcie_port_device_suspend, .resume = pcie_port_device_resume, @@ -76,6 +106,7 @@ static const struct dev_pm_ops pcie_portdrv_pm_ops = { .thaw = pcie_port_device_resume, .poweroff = pcie_port_device_suspend, .restore = pcie_port_device_resume, + .resume_noirq = pcie_port_resume_noirq, }; #define PCIE_PORTDRV_PM_OPS (&pcie_portdrv_pm_ops) @@ -327,10 +358,8 @@ static int __init pcie_portdrv_init(void) { int retval; - if (pcie_ports_disabled) { - pcie_no_aspm(); - return -EACCES; - } + if (pcie_ports_disabled) + return pci_register_driver(&pcie_portdriver); dmi_check_system(pcie_portdrv_dmi_table); diff --git a/include/linux/pci_regs.h b/include/linux/pci_regs.h index d4f2c80a6c3e..5b7e6b1ba54f 100644 --- a/include/linux/pci_regs.h +++ b/include/linux/pci_regs.h @@ -504,6 +504,8 @@ #define PCI_EXP_RTCTL_CRSSVE 0x10 /* CRS Software Visibility Enable */ #define PCI_EXP_RTCAP 30 /* Root Capabilities */ #define PCI_EXP_RTSTA 32 /* Root Status */ +#define PCI_EXP_RTSTA_PME 0x10000 /* PME status */ +#define PCI_EXP_RTSTA_PENDING 0x20000 /* PME pending */ #define PCI_EXP_DEVCAP2 36 /* Device Capabilities 2 */ #define PCI_EXP_DEVCAP2_ARI 0x20 /* Alternative Routing-ID */ #define PCI_EXP_DEVCTL2 40 /* Device Control 2 */ -- cgit v1.2.3 From bdd5f05d91e8ae68075b812ce244c918d3d752cd Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Tue, 4 Jan 2011 13:31:45 -0500 Subject: SUNRPC: Remove more code when NFSD_DEPRECATED is not configured Signed-off-by: NeilBrown [bfields@redhat.com: moved svcauth_unix_purge outside ifdef's.] Signed-off-by: J. Bruce Fields --- include/linux/sunrpc/cache.h | 3 +++ net/sunrpc/svcauth_unix.c | 12 ++++++++++++ 2 files changed, 15 insertions(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/cache.h b/include/linux/sunrpc/cache.h index 6950c981882d..c2aebe8eeffd 100644 --- a/include/linux/sunrpc/cache.h +++ b/include/linux/sunrpc/cache.h @@ -255,10 +255,13 @@ static inline time_t get_expiry(char **bpp) return rv - boot.tv_sec; } +#ifdef CONFIG_NFSD_DEPRECATED static inline void sunrpc_invalidate(struct cache_head *h, struct cache_detail *detail) { h->expiry_time = seconds_since_boot() - 1; detail->nextcheck = seconds_since_boot(); } +#endif /* CONFIG_NFSD_DEPRECATED */ + #endif /* _LINUX_SUNRPC_CACHE_H_ */ diff --git a/net/sunrpc/svcauth_unix.c b/net/sunrpc/svcauth_unix.c index 59a7c524a8b1..30916b06c12b 100644 --- a/net/sunrpc/svcauth_unix.c +++ b/net/sunrpc/svcauth_unix.c @@ -30,7 +30,9 @@ struct unix_domain { struct auth_domain h; +#ifdef CONFIG_NFSD_DEPRECATED int addr_changes; +#endif /* CONFIG_NFSD_DEPRECATED */ /* other stuff later */ }; @@ -64,7 +66,9 @@ struct auth_domain *unix_domain_find(char *name) return NULL; } new->h.flavour = &svcauth_unix; +#ifdef CONFIG_NFSD_DEPRECATED new->addr_changes = 0; +#endif /* CONFIG_NFSD_DEPRECATED */ rv = auth_domain_lookup(name, &new->h); } } @@ -91,7 +95,9 @@ struct ip_map { char m_class[8]; /* e.g. "nfsd" */ struct in6_addr m_addr; struct unix_domain *m_client; +#ifdef CONFIG_NFSD_DEPRECATED int m_add_change; +#endif /* CONFIG_NFSD_DEPRECATED */ }; static void ip_map_put(struct kref *kref) @@ -145,7 +151,9 @@ static void update(struct cache_head *cnew, struct cache_head *citem) kref_get(&item->m_client->h.ref); new->m_client = item->m_client; +#ifdef CONFIG_NFSD_DEPRECATED new->m_add_change = item->m_add_change; +#endif /* CONFIG_NFSD_DEPRECATED */ } static struct cache_head *ip_map_alloc(void) { @@ -330,6 +338,7 @@ static int __ip_map_update(struct cache_detail *cd, struct ip_map *ipm, ip.h.flags = 0; if (!udom) set_bit(CACHE_NEGATIVE, &ip.h.flags); +#ifdef CONFIG_NFSD_DEPRECATED else { ip.m_add_change = udom->addr_changes; /* if this is from the legacy set_client system call, @@ -338,6 +347,7 @@ static int __ip_map_update(struct cache_detail *cd, struct ip_map *ipm, if (expiry == NEVER) ip.m_add_change++; } +#endif /* CONFIG_NFSD_DEPRECATED */ ip.h.expiry_time = expiry; ch = sunrpc_cache_update(cd, &ip.h, &ipm->h, hash_str(ipm->m_class, IP_HASHBITS) ^ @@ -357,6 +367,7 @@ static inline int ip_map_update(struct net *net, struct ip_map *ipm, return __ip_map_update(sn->ip_map_cache, ipm, udom, expiry); } +#ifdef CONFIG_NFSD_DEPRECATED int auth_unix_add_addr(struct net *net, struct in6_addr *addr, struct auth_domain *dom) { struct unix_domain *udom; @@ -411,6 +422,7 @@ struct auth_domain *auth_unix_lookup(struct net *net, struct in6_addr *addr) return rv; } EXPORT_SYMBOL_GPL(auth_unix_lookup); +#endif /* CONFIG_NFSD_DEPRECATED */ void svcauth_unix_purge(void) { -- cgit v1.2.3 From 9e701c610923aaeac8b38b9202a686d1cc9ee35d Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Sun, 2 Jan 2011 21:56:36 -0500 Subject: svcrpc: simpler request dropping Currently we use -EAGAIN returns to determine when to drop a deferred request. On its own, that is error-prone, as it makes us treat -EAGAIN returns from other functions specially to prevent inadvertent dropping. So, use a flag on the request instead. Returning an error on request deferral is still required, to prevent further processing, but we no longer need worry that an error return on its own could result in a drop. Signed-off-by: J. Bruce Fields --- fs/nfsd/nfssvc.c | 2 +- include/linux/sunrpc/svc.h | 1 + net/sunrpc/svc.c | 3 ++- net/sunrpc/svc_xprt.c | 1 + 4 files changed, 5 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index 2bae1d86f5f2..18743c4d8bca 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -608,7 +608,7 @@ nfsd_dispatch(struct svc_rqst *rqstp, __be32 *statp) /* Now call the procedure handler, and encode NFS status. */ nfserr = proc->pc_func(rqstp, rqstp->rq_argp, rqstp->rq_resp); nfserr = map_new_errors(rqstp->rq_vers, nfserr); - if (nfserr == nfserr_dropit) { + if (nfserr == nfserr_dropit || rqstp->rq_dropme) { dprintk("nfsd: Dropping request; may be revisited later\n"); nfsd_cache_update(rqstp, RC_NOCACHE, NULL); return 0; diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index 5a3085b9b394..d45c482b6444 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -269,6 +269,7 @@ struct svc_rqst { struct cache_req rq_chandle; /* handle passed to caches for * request delaying */ + bool rq_dropme; /* Catering to nfsd */ struct auth_domain * rq_client; /* RPC peer info */ struct auth_domain * rq_gssclient; /* "gss/"-style peer info */ diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 6359c42c4941..df1931f8ae98 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -1005,6 +1005,7 @@ svc_process_common(struct svc_rqst *rqstp, struct kvec *argv, struct kvec *resv) rqstp->rq_splice_ok = 1; /* Will be turned off only when NFSv4 Sessions are used */ rqstp->rq_usedeferral = 1; + rqstp->rq_dropme = false; /* Setup reply header */ rqstp->rq_xprt->xpt_ops->xpo_prep_reply_hdr(rqstp); @@ -1106,7 +1107,7 @@ svc_process_common(struct svc_rqst *rqstp, struct kvec *argv, struct kvec *resv) *statp = procp->pc_func(rqstp, rqstp->rq_argp, rqstp->rq_resp); /* Encode reply */ - if (*statp == rpc_drop_reply) { + if (rqstp->rq_dropme) { if (procp->pc_release) procp->pc_release(rqstp, NULL, rqstp->rq_resp); goto dropit; diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index 5eae53b1e306..173f3b975d49 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -1019,6 +1019,7 @@ static struct cache_deferred_req *svc_defer(struct cache_req *req) } svc_xprt_get(rqstp->rq_xprt); dr->xprt = rqstp->rq_xprt; + rqstp->rq_dropme = true; dr->handle.revisit = svc_revisit; return &dr->handle; -- cgit v1.2.3 From c45821d263a8a5109d69a9e8942b8d65bcd5f31a Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Sun, 31 Oct 2010 00:04:44 -0400 Subject: locks: eliminate fl_mylease callback The nfs server only supports read delegations for now, so we don't care how conflicts are determined. All we care is that unlocks are recognized as matching the leases they are meant to remove. After the last patch, a comparison of struct files will work for that purpose. So we no longer need this callback. Signed-off-by: J. Bruce Fields --- fs/locks.c | 8 +------- fs/nfsd/nfs4state.c | 21 +-------------------- include/linux/fs.h | 1 - 3 files changed, 2 insertions(+), 28 deletions(-) (limited to 'include/linux') diff --git a/fs/locks.c b/fs/locks.c index 8729347bcd1a..5cb65062281a 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -444,15 +444,9 @@ static void lease_release_private_callback(struct file_lock *fl) fl->fl_file->f_owner.signum = 0; } -static int lease_mylease_callback(struct file_lock *fl, struct file_lock *try) -{ - return fl->fl_file == try->fl_file; -} - static const struct lock_manager_operations lease_manager_ops = { .fl_break = lease_break_callback, .fl_release_private = lease_release_private_callback, - .fl_mylease = lease_mylease_callback, .fl_change = lease_modify, }; @@ -1405,7 +1399,7 @@ int generic_setlease(struct file *filp, long arg, struct file_lock **flp) for (before = &inode->i_flock; ((fl = *before) != NULL) && IS_LEASE(fl); before = &fl->fl_next) { - if (lease->fl_lmops->fl_mylease(fl, lease)) + if (fl->fl_file == lease->fl_file) my_before = before; else if (fl->fl_type == (F_INPROGRESS | F_UNLCK)) /* diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index cbe1b81c147d..87d4c48b6069 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -2295,24 +2295,6 @@ void nfsd_break_deleg_cb(struct file_lock *fl) nfsd4_cb_recall(dp); } -/* - * Called from setlease() with lock_flocks() held - */ -static -int nfsd_same_client_deleg_cb(struct file_lock *onlist, struct file_lock *try) -{ - struct nfs4_delegation *onlistd = - (struct nfs4_delegation *)onlist->fl_owner; - struct nfs4_delegation *tryd = - (struct nfs4_delegation *)try->fl_owner; - - if (onlist->fl_lmops != try->fl_lmops) - return 0; - - return onlistd->dl_client == tryd->dl_client; -} - - static int nfsd_change_deleg_cb(struct file_lock **onlist, int arg) { @@ -2324,7 +2306,6 @@ int nfsd_change_deleg_cb(struct file_lock **onlist, int arg) static const struct lock_manager_operations nfsd_lease_mng_ops = { .fl_break = nfsd_break_deleg_cb, - .fl_mylease = nfsd_same_client_deleg_cb, .fl_change = nfsd_change_deleg_cb, }; @@ -2630,7 +2611,7 @@ nfs4_open_delegation(struct svc_fh *fh, struct nfsd4_open *open, struct nfs4_sta dp->dl_flock = fl; /* vfs_setlease checks to see if delegation should be handed out. - * the lock_manager callbacks fl_mylease and fl_change are used + * the lock_manager callback fl_change is used */ if ((status = vfs_setlease(fl->fl_file, fl->fl_type, &fl))) { dprintk("NFSD: setlease failed [%d], no delegation\n", status); diff --git a/include/linux/fs.h b/include/linux/fs.h index 090f0eacde29..73ce69f728e1 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1059,7 +1059,6 @@ struct lock_manager_operations { int (*fl_grant)(struct file_lock *, struct file_lock *, int); void (*fl_release_private)(struct file_lock *); void (*fl_break)(struct file_lock *); - int (*fl_mylease)(struct file_lock *, struct file_lock *); int (*fl_change)(struct file_lock **, int); }; -- cgit v1.2.3 From 2ca72e17e5acb1052c35c9faba609c2289ce7a92 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Tue, 4 Jan 2011 17:37:15 -0500 Subject: nfsd4: move idmap and acl header files into fs/nfsd These are internal nfsd interfaces. Signed-off-by: J. Bruce Fields --- fs/nfsd/acl.h | 61 +++++++++++++++++++++++++++++++++++++++++++ fs/nfsd/idmap.h | 64 ++++++++++++++++++++++++++++++++++++++++++++++ fs/nfsd/nfs4acl.c | 2 +- fs/nfsd/nfs4idmap.c | 2 +- fs/nfsd/nfs4xdr.c | 5 ++-- fs/nfsd/nfsctl.c | 2 +- fs/nfsd/vfs.c | 4 +-- include/linux/nfs4_acl.h | 61 ------------------------------------------- include/linux/nfsd_idmap.h | 64 ---------------------------------------------- 9 files changed, 133 insertions(+), 132 deletions(-) create mode 100644 fs/nfsd/acl.h create mode 100644 fs/nfsd/idmap.h delete mode 100644 include/linux/nfs4_acl.h delete mode 100644 include/linux/nfsd_idmap.h (limited to 'include/linux') diff --git a/fs/nfsd/acl.h b/fs/nfsd/acl.h new file mode 100644 index 000000000000..c9c05a78e9bb --- /dev/null +++ b/fs/nfsd/acl.h @@ -0,0 +1,61 @@ +/* + * include/linux/nfs4_acl.c + * + * Common NFSv4 ACL handling definitions. + * + * Copyright (c) 2002 The Regents of the University of Michigan. + * All rights reserved. + * + * Marius Aamodt Eriksen + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the name of the University nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED + * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF + * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING + * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef LINUX_NFS4_ACL_H +#define LINUX_NFS4_ACL_H + +#include + +/* Maximum ACL we'll accept from client; chosen (somewhat arbitrarily) to + * fit in a page: */ +#define NFS4_ACL_MAX 170 + +struct nfs4_acl *nfs4_acl_new(int); +int nfs4_acl_get_whotype(char *, u32); +int nfs4_acl_write_who(int who, char *p); +int nfs4_acl_permission(struct nfs4_acl *acl, uid_t owner, gid_t group, + uid_t who, u32 mask); + +#define NFS4_ACL_TYPE_DEFAULT 0x01 +#define NFS4_ACL_DIR 0x02 +#define NFS4_ACL_OWNER 0x04 + +struct nfs4_acl *nfs4_acl_posix_to_nfsv4(struct posix_acl *, + struct posix_acl *, unsigned int flags); +int nfs4_acl_nfsv4_to_posix(struct nfs4_acl *, struct posix_acl **, + struct posix_acl **, unsigned int flags); + +#endif /* LINUX_NFS4_ACL_H */ diff --git a/fs/nfsd/idmap.h b/fs/nfsd/idmap.h new file mode 100644 index 000000000000..d4a2ac18bd4c --- /dev/null +++ b/fs/nfsd/idmap.h @@ -0,0 +1,64 @@ +/* + * include/linux/nfsd_idmap.h + * + * Mapping of UID to name and vice versa. + * + * Copyright (c) 2002, 2003 The Regents of the University of + * Michigan. All rights reserved. +> * + * Marius Aamodt Eriksen + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the name of the University nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED + * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF + * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING + * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef LINUX_NFSD_IDMAP_H +#define LINUX_NFSD_IDMAP_H + +#include +#include + +/* XXX from linux/nfs_idmap.h */ +#define IDMAP_NAMESZ 128 + +#ifdef CONFIG_NFSD_V4 +int nfsd_idmap_init(void); +void nfsd_idmap_shutdown(void); +#else +static inline int nfsd_idmap_init(void) +{ + return 0; +} +static inline void nfsd_idmap_shutdown(void) +{ +} +#endif + +int nfsd_map_name_to_uid(struct svc_rqst *, const char *, size_t, __u32 *); +int nfsd_map_name_to_gid(struct svc_rqst *, const char *, size_t, __u32 *); +int nfsd_map_uid_to_name(struct svc_rqst *, __u32, char *); +int nfsd_map_gid_to_name(struct svc_rqst *, __u32, char *); + +#endif /* LINUX_NFSD_IDMAP_H */ diff --git a/fs/nfsd/nfs4acl.c b/fs/nfsd/nfs4acl.c index e48052615159..ad88f1c0a4c3 100644 --- a/fs/nfsd/nfs4acl.c +++ b/fs/nfsd/nfs4acl.c @@ -36,7 +36,7 @@ #include #include -#include +#include "acl.h" /* mode bit translations: */ diff --git a/fs/nfsd/nfs4idmap.c b/fs/nfsd/nfs4idmap.c index 844960fd0395..cbd599732765 100644 --- a/fs/nfsd/nfs4idmap.c +++ b/fs/nfsd/nfs4idmap.c @@ -33,10 +33,10 @@ */ #include -#include #include #include #include +#include "idmap.h" /* * Cache entry diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 364aae7d5998..2a0814d0ab1a 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -44,13 +44,14 @@ #include #include #include -#include -#include #include +#include "idmap.h" +#include "acl.h" #include "xdr4.h" #include "vfs.h" + #define NFSDDBG_FACILITY NFSDDBG_XDR /* diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 6840ec3ceecf..33b3e2b06779 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -8,12 +8,12 @@ #include #include -#include #include #include #include #include +#include "idmap.h" #include "nfsd.h" #include "cache.h" diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 6a3af2ff3afe..b991125ce4a5 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -35,8 +35,8 @@ #endif /* CONFIG_NFSD_V3 */ #ifdef CONFIG_NFSD_V4 -#include -#include +#include "acl.h" +#include "idmap.h" #endif /* CONFIG_NFSD_V4 */ #include "nfsd.h" diff --git a/include/linux/nfs4_acl.h b/include/linux/nfs4_acl.h deleted file mode 100644 index c9c05a78e9bb..000000000000 --- a/include/linux/nfs4_acl.h +++ /dev/null @@ -1,61 +0,0 @@ -/* - * include/linux/nfs4_acl.c - * - * Common NFSv4 ACL handling definitions. - * - * Copyright (c) 2002 The Regents of the University of Michigan. - * All rights reserved. - * - * Marius Aamodt Eriksen - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * - * 1. Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * 2. Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in the - * documentation and/or other materials provided with the distribution. - * 3. Neither the name of the University nor the names of its - * contributors may be used to endorse or promote products derived - * from this software without specific prior written permission. - * - * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED - * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF - * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE - * DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE - * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR - * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF - * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR - * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF - * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING - * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS - * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - */ - -#ifndef LINUX_NFS4_ACL_H -#define LINUX_NFS4_ACL_H - -#include - -/* Maximum ACL we'll accept from client; chosen (somewhat arbitrarily) to - * fit in a page: */ -#define NFS4_ACL_MAX 170 - -struct nfs4_acl *nfs4_acl_new(int); -int nfs4_acl_get_whotype(char *, u32); -int nfs4_acl_write_who(int who, char *p); -int nfs4_acl_permission(struct nfs4_acl *acl, uid_t owner, gid_t group, - uid_t who, u32 mask); - -#define NFS4_ACL_TYPE_DEFAULT 0x01 -#define NFS4_ACL_DIR 0x02 -#define NFS4_ACL_OWNER 0x04 - -struct nfs4_acl *nfs4_acl_posix_to_nfsv4(struct posix_acl *, - struct posix_acl *, unsigned int flags); -int nfs4_acl_nfsv4_to_posix(struct nfs4_acl *, struct posix_acl **, - struct posix_acl **, unsigned int flags); - -#endif /* LINUX_NFS4_ACL_H */ diff --git a/include/linux/nfsd_idmap.h b/include/linux/nfsd_idmap.h deleted file mode 100644 index d4a2ac18bd4c..000000000000 --- a/include/linux/nfsd_idmap.h +++ /dev/null @@ -1,64 +0,0 @@ -/* - * include/linux/nfsd_idmap.h - * - * Mapping of UID to name and vice versa. - * - * Copyright (c) 2002, 2003 The Regents of the University of - * Michigan. All rights reserved. -> * - * Marius Aamodt Eriksen - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * - * 1. Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * 2. Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in the - * documentation and/or other materials provided with the distribution. - * 3. Neither the name of the University nor the names of its - * contributors may be used to endorse or promote products derived - * from this software without specific prior written permission. - * - * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED - * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF - * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE - * DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE - * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR - * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF - * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR - * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF - * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING - * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS - * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - */ - -#ifndef LINUX_NFSD_IDMAP_H -#define LINUX_NFSD_IDMAP_H - -#include -#include - -/* XXX from linux/nfs_idmap.h */ -#define IDMAP_NAMESZ 128 - -#ifdef CONFIG_NFSD_V4 -int nfsd_idmap_init(void); -void nfsd_idmap_shutdown(void); -#else -static inline int nfsd_idmap_init(void) -{ - return 0; -} -static inline void nfsd_idmap_shutdown(void) -{ -} -#endif - -int nfsd_map_name_to_uid(struct svc_rqst *, const char *, size_t, __u32 *); -int nfsd_map_name_to_gid(struct svc_rqst *, const char *, size_t, __u32 *); -int nfsd_map_uid_to_name(struct svc_rqst *, __u32, char *); -int nfsd_map_gid_to_name(struct svc_rqst *, __u32, char *); - -#endif /* LINUX_NFSD_IDMAP_H */ -- cgit v1.2.3 From 91aa5fadb831e7b6ea473a526a6b49c6dc4819ce Mon Sep 17 00:00:00 2001 From: Russell King - ARM Linux Date: Mon, 3 Jan 2011 22:31:04 +0000 Subject: ARM: PL08x: fix atomic_t usage and tx_submit() return value range The last_issued variable uses an atomic type, which is only incremented inside a protected region, and then read. Everywhere else only reads the value, so it isn't using atomic_t correctly, and it doesn't even need to. Moreover, the DMA engine code provides us with a variable for this already - chan.cookie. Use chan.cookie instead. Also, avoid negative dma_cookie_t values - negative returns from tx_submit() mean failure, yet in reality we always succeed. Restart from cookie 1, just like other DMA engine drivers do. Signed-off-by: Russell King Acked-by: Linus Walleij Signed-off-by: Dan Williams --- drivers/dma/amba-pl08x.c | 18 +++++++++--------- include/linux/amba/pl08x.h | 1 - 2 files changed, 9 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/drivers/dma/amba-pl08x.c b/drivers/dma/amba-pl08x.c index 0809810f9e7a..5d9a15652dba 100644 --- a/drivers/dma/amba-pl08x.c +++ b/drivers/dma/amba-pl08x.c @@ -74,7 +74,6 @@ #include #include #include -#include #include #include @@ -1082,8 +1081,10 @@ static dma_cookie_t pl08x_tx_submit(struct dma_async_tx_descriptor *tx) { struct pl08x_dma_chan *plchan = to_pl08x_chan(tx->chan); - atomic_inc(&plchan->last_issued); - tx->cookie = atomic_read(&plchan->last_issued); + plchan->chan.cookie += 1; + if (plchan->chan.cookie < 0) + plchan->chan.cookie = 1; + tx->cookie = plchan->chan.cookie; /* This unlock follows the lock in the prep() function */ spin_unlock_irqrestore(&plchan->lock, plchan->lockflags); @@ -1115,7 +1116,7 @@ pl08x_dma_tx_status(struct dma_chan *chan, enum dma_status ret; u32 bytesleft = 0; - last_used = atomic_read(&plchan->last_issued); + last_used = plchan->chan.cookie; last_complete = plchan->lc; ret = dma_async_is_complete(cookie, last_complete, last_used); @@ -1131,7 +1132,7 @@ pl08x_dma_tx_status(struct dma_chan *chan, /* * This cookie not complete yet */ - last_used = atomic_read(&plchan->last_issued); + last_used = plchan->chan.cookie; last_complete = plchan->lc; /* Get number of bytes left in the active transactions and queue */ @@ -1641,8 +1642,7 @@ static void pl08x_tasklet(unsigned long data) /* * Update last completed */ - plchan->lc = - (plchan->at->tx.cookie); + plchan->lc = plchan->at->tx.cookie; /* * Callback to signal completion @@ -1820,8 +1820,8 @@ static int pl08x_dma_init_virtual_channels(struct pl08x_driver_data *pl08x, chan->name); chan->chan.device = dmadev; - atomic_set(&chan->last_issued, 0); - chan->lc = atomic_read(&chan->last_issued); + chan->chan.cookie = 0; + chan->lc = 0; spin_lock_init(&chan->lock); INIT_LIST_HEAD(&chan->desc_list); diff --git a/include/linux/amba/pl08x.h b/include/linux/amba/pl08x.h index 521a0f8974ac..4ae62b4684f9 100644 --- a/include/linux/amba/pl08x.h +++ b/include/linux/amba/pl08x.h @@ -174,7 +174,6 @@ struct pl08x_dma_chan { struct pl08x_channel_data *cd; dma_addr_t runtime_addr; enum dma_data_direction runtime_direction; - atomic_t last_issued; dma_cookie_t lc; struct list_head desc_list; struct pl08x_txd *at; -- cgit v1.2.3 From 7cb72ad959b16ac594118977b7954a7d2ec7a052 Mon Sep 17 00:00:00 2001 From: Russell King - ARM Linux Date: Mon, 3 Jan 2011 22:35:28 +0000 Subject: ARM: PL08x: avoid 'void *' struct fields when we can type them properly Avoid using 'void *' struct fields when the structs are not defined in linux/amba/pl08x.h - instead, forward declare the struct names, and use these instead. This ensures we have proper typechecking. Signed-off-by: Russell King Acked-by: Linus Walleij Signed-off-by: Dan Williams --- drivers/dma/amba-pl08x.c | 16 ++++++++-------- include/linux/amba/pl08x.h | 5 ++++- 2 files changed, 12 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/drivers/dma/amba-pl08x.c b/drivers/dma/amba-pl08x.c index 707fa081c9fa..fada97873d7d 100644 --- a/drivers/dma/amba-pl08x.c +++ b/drivers/dma/amba-pl08x.c @@ -107,7 +107,7 @@ struct vendor_data { * An LLI struct - see PL08x TRM. Note that next uses bit[0] as a bus bit, * start & end do not - their bus bit info is in cctl. */ -struct lli { +struct pl08x_lli { dma_addr_t src; dma_addr_t dst; dma_addr_t next; @@ -160,7 +160,7 @@ struct pl08x_driver_data { /* Maximum times we call dma_pool_alloc on this pool without freeing */ #define PL08X_MAX_ALLOCS 0x40 -#define MAX_NUM_TSFR_LLIS (PL08X_LLI_TSFR_SIZE/sizeof(struct lli)) +#define MAX_NUM_TSFR_LLIS (PL08X_LLI_TSFR_SIZE/sizeof(struct pl08x_lli)) #define PL08X_ALIGN 8 static inline struct pl08x_dma_chan *to_pl08x_chan(struct dma_chan *chan) @@ -354,8 +354,8 @@ static u32 pl08x_getbytes_chan(struct pl08x_dma_chan *plchan) * currently active transaction. */ if (ch && txd) { - struct lli *llis_va = txd->llis_va; - struct lli *llis_bus = (struct lli *) txd->llis_bus; + struct pl08x_lli *llis_va = txd->llis_va; + struct pl08x_lli *llis_bus = (struct pl08x_lli *) txd->llis_bus; u32 clli = readl(ch->base + PL080_CH_LLI); /* First get the bytes in the current active LLI */ @@ -558,8 +558,8 @@ static int pl08x_fill_lli_for_desc(struct pl08x_driver_data *pl08x, struct pl08x_txd *txd, int num_llis, int len, u32 cctl, u32 *remainder) { - struct lli *llis_va = txd->llis_va; - struct lli *llis_bus = (struct lli *) txd->llis_bus; + struct pl08x_lli *llis_va = txd->llis_va; + struct pl08x_lli *llis_bus = (struct pl08x_lli *) txd->llis_bus; BUG_ON(num_llis >= MAX_NUM_TSFR_LLIS); @@ -620,8 +620,8 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, u32 cctl; int max_bytes_per_lli; int total_bytes = 0; - struct lli *llis_va; - struct lli *llis_bus; + struct pl08x_lli *llis_va; + struct pl08x_lli *llis_bus; txd->llis_va = dma_pool_alloc(pl08x->pool, GFP_NOWAIT, &txd->llis_bus); diff --git a/include/linux/amba/pl08x.h b/include/linux/amba/pl08x.h index 4ae62b4684f9..3ecc20fce26a 100644 --- a/include/linux/amba/pl08x.h +++ b/include/linux/amba/pl08x.h @@ -22,6 +22,9 @@ #include #include +struct pl08x_lli; +struct pl08x_driver_data; + /** * struct pl08x_channel_data - data structure to pass info between * platform and PL08x driver regarding channel configuration @@ -179,7 +182,7 @@ struct pl08x_dma_chan { struct pl08x_txd *at; unsigned long lockflags; spinlock_t lock; - void *host; + struct pl08x_driver_data *host; enum pl08x_dma_chan_state state; bool slave; struct pl08x_txd *waiting; -- cgit v1.2.3 From cace658572ba5d1075f3891e823130a66f3e330f Mon Sep 17 00:00:00 2001 From: Russell King - ARM Linux Date: Mon, 3 Jan 2011 22:37:31 +0000 Subject: ARM: PL08x: use 'size_t' for lengths Use size_t for variables denoting lengths throughout, and use the 'z' qualifier for printing the value. For safety, add a BUG_ON() in pl08x_fill_lli_for_desc() to catch the remainder potentially becoming negative. Signed-off-by: Russell King Acked-by: Linus Walleij Signed-off-by: Dan Williams --- drivers/dma/amba-pl08x.c | 40 +++++++++++++++++----------------------- include/linux/amba/pl08x.h | 4 ++-- 2 files changed, 19 insertions(+), 25 deletions(-) (limited to 'include/linux') diff --git a/drivers/dma/amba-pl08x.c b/drivers/dma/amba-pl08x.c index 760b71eec84c..fa78697790c0 100644 --- a/drivers/dma/amba-pl08x.c +++ b/drivers/dma/amba-pl08x.c @@ -342,7 +342,7 @@ static u32 pl08x_getbytes_chan(struct pl08x_dma_chan *plchan) struct pl08x_txd *txdi = NULL; struct pl08x_txd *txd; unsigned long flags; - u32 bytes = 0; + size_t bytes = 0; spin_lock_irqsave(&plchan->lock, flags); @@ -470,7 +470,7 @@ static inline unsigned int pl08x_get_bytes_for_cctl(unsigned int coded) } static inline u32 pl08x_cctl_bits(u32 cctl, u8 srcwidth, u8 dstwidth, - u32 tsize) + size_t tsize) { u32 retbits = cctl; @@ -583,6 +583,8 @@ static int pl08x_fill_lli_for_desc(struct pl08x_driver_data *pl08x, if (cctl & PL080_CONTROL_DST_INCR) txd->dstbus.addr += len; + BUG_ON(*remainder < len); + *remainder -= len; return num_llis + 1; @@ -591,7 +593,7 @@ static int pl08x_fill_lli_for_desc(struct pl08x_driver_data *pl08x, /* * Return number of bytes to fill to boundary, or len */ -static inline u32 pl08x_pre_boundary(u32 addr, u32 len) +static inline size_t pl08x_pre_boundary(u32 addr, size_t len) { u32 boundary; @@ -614,11 +616,11 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, { struct pl08x_channel_data *cd = txd->cd; struct pl08x_bus_data *mbus, *sbus; - u32 remainder; + size_t remainder; int num_llis = 0; u32 cctl; - int max_bytes_per_lli; - int total_bytes = 0; + size_t max_bytes_per_lli; + size_t total_bytes = 0; struct pl08x_lli *llis_va; txd->llis_va = dma_pool_alloc(pl08x->pool, GFP_NOWAIT, @@ -686,13 +688,13 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, max_bytes_per_lli = min(txd->srcbus.buswidth, txd->dstbus.buswidth) * PL080_CONTROL_TRANSFER_SIZE_MASK; dev_vdbg(&pl08x->adev->dev, - "%s max bytes per lli = %d\n", + "%s max bytes per lli = %zu\n", __func__, max_bytes_per_lli); /* We need to count this down to zero */ remainder = txd->len; dev_vdbg(&pl08x->adev->dev, - "%s remainder = %d\n", + "%s remainder = %zu\n", __func__, remainder); /* @@ -760,9 +762,7 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, * width left */ while (remainder > (mbus->buswidth - 1)) { - int lli_len, target_len; - int tsize; - int odd_bytes; + size_t lli_len, target_len, tsize, odd_bytes; /* * If enough left try to send max possible, @@ -805,7 +805,7 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, if (lli_len <= 0) { dev_err(&pl08x->adev->dev, - "%s lli_len is %d, <= 0\n", + "%s lli_len is %zu, <= 0\n", __func__, lli_len); return 0; } @@ -853,7 +853,7 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, if (target_len != lli_len) { dev_vdbg(&pl08x->adev->dev, - "%s can't send what we want. Desired 0x%08x, lli of 0x%08x bytes in txd of 0x%08x\n", + "%s can't send what we want. Desired 0x%08zx, lli of 0x%08zx bytes in txd of 0x%08zx\n", __func__, target_len, lli_len, txd->len); } @@ -863,7 +863,7 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, tsize); dev_vdbg(&pl08x->adev->dev, - "%s fill lli with single lli chunk of size 0x%08x (remainder 0x%08x)\n", + "%s fill lli with single lli chunk of size 0x%08zx (remainder 0x%08zx)\n", __func__, lli_len, remainder); num_llis = pl08x_fill_lli_for_desc(pl08x, txd, num_llis, lli_len, cctl, @@ -882,7 +882,7 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, && (remainder); j++) { cctl = pl08x_cctl_bits(cctl, 1, 1, 1); dev_vdbg(&pl08x->adev->dev, - "%s align with boundary, single byte (remain 0x%08x)\n", + "%s align with boundary, single byte (remain 0x%08zx)\n", __func__, remainder); num_llis = pl08x_fill_lli_for_desc(pl08x, @@ -896,16 +896,10 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, /* * Send any odd bytes */ - if (remainder < 0) { - dev_err(&pl08x->adev->dev, "%s remainder not fitted 0x%08x bytes\n", - __func__, remainder); - return 0; - } - while (remainder) { cctl = pl08x_cctl_bits(cctl, 1, 1, 1); dev_vdbg(&pl08x->adev->dev, - "%s align with boundary, single odd byte (remain %d)\n", + "%s align with boundary, single odd byte (remain %zu)\n", __func__, remainder); num_llis = pl08x_fill_lli_for_desc(pl08x, txd, num_llis, 1, cctl, &remainder); @@ -914,7 +908,7 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, } if (total_bytes != txd->len) { dev_err(&pl08x->adev->dev, - "%s size of encoded lli:s don't match total txd, transferred 0x%08x from size 0x%08x\n", + "%s size of encoded lli:s don't match total txd, transferred 0x%08zx from size 0x%08zx\n", __func__, total_bytes, txd->len); return 0; } diff --git a/include/linux/amba/pl08x.h b/include/linux/amba/pl08x.h index 3ecc20fce26a..2c834ed5f41f 100644 --- a/include/linux/amba/pl08x.h +++ b/include/linux/amba/pl08x.h @@ -77,7 +77,7 @@ struct pl08x_bus_data { dma_addr_t addr; u8 maxwidth; u8 buswidth; - u32 fill_bytes; + size_t fill_bytes; }; /** @@ -113,7 +113,7 @@ struct pl08x_txd { enum dma_data_direction direction; struct pl08x_bus_data srcbus; struct pl08x_bus_data dstbus; - int len; + size_t len; dma_addr_t llis_bus; void *llis_va; struct pl08x_channel_data *cd; -- cgit v1.2.3 From 19524d77ec34faf58d313ba34fb755ef6e159216 Mon Sep 17 00:00:00 2001 From: Russell King - ARM Linux Date: Mon, 3 Jan 2011 22:39:13 +0000 Subject: ARM: PL08x: avoid duplicating registers in txd and phychan structures As we now have all the code accessing the phychan {csrc,cdst,clli,cctl, ccfg} members in one function, there's no point storing the data into the struct. Get rid of the struct members. Re-order the register dump in the dev_dbg() to reflect the order we write the registers to the DMA device. The txd {csrc,cdst,clli,cctl} values are duplicates of the lli[0] values, so there's no point duplicating these either. Program the DMAC registers directly from the lli[0] values. Signed-off-by: Russell King Acked-by: Linus Walleij Signed-off-by: Dan Williams --- drivers/dma/amba-pl08x.c | 57 ++++++++++++++++------------------------------ include/linux/amba/pl08x.h | 13 ----------- 2 files changed, 19 insertions(+), 51 deletions(-) (limited to 'include/linux') diff --git a/drivers/dma/amba-pl08x.c b/drivers/dma/amba-pl08x.c index c025a4b4bae4..a1a18bde6b7f 100644 --- a/drivers/dma/amba-pl08x.c +++ b/drivers/dma/amba-pl08x.c @@ -193,33 +193,25 @@ static void pl08x_start_txd(struct pl08x_dma_chan *plchan, { struct pl08x_driver_data *pl08x = plchan->host; struct pl08x_phy_chan *phychan = plchan->phychan; - u32 val; + struct pl08x_lli *lli = &txd->llis_va[0]; + u32 val, ccfg; plchan->at = txd; - /* Copy the basic control register calculated at transfer config */ - phychan->csrc = txd->csrc; - phychan->cdst = txd->cdst; - phychan->clli = txd->clli; - phychan->cctl = txd->cctl; - /* Assign the signal to the proper control registers */ - phychan->ccfg = plchan->cd->ccfg; - phychan->ccfg &= ~PL080_CONFIG_SRC_SEL_MASK; - phychan->ccfg &= ~PL080_CONFIG_DST_SEL_MASK; + ccfg = plchan->cd->ccfg; + ccfg &= ~(PL080_CONFIG_SRC_SEL_MASK | PL080_CONFIG_DST_SEL_MASK); + /* If it wasn't set from AMBA, ignore it */ if (txd->direction == DMA_TO_DEVICE) /* Select signal as destination */ - phychan->ccfg |= - (phychan->signal << PL080_CONFIG_DST_SEL_SHIFT); + ccfg |= phychan->signal << PL080_CONFIG_DST_SEL_SHIFT; else if (txd->direction == DMA_FROM_DEVICE) /* Select signal as source */ - phychan->ccfg |= - (phychan->signal << PL080_CONFIG_SRC_SEL_SHIFT); - /* Always enable error interrupts */ - phychan->ccfg |= PL080_CONFIG_ERR_IRQ_MASK; - /* Always enable terminal interrupts */ - phychan->ccfg |= PL080_CONFIG_TC_IRQ_MASK; + ccfg |= phychan->signal << PL080_CONFIG_SRC_SEL_SHIFT; + + /* Always enable error and terminal interrupts */ + ccfg |= PL080_CONFIG_ERR_IRQ_MASK | PL080_CONFIG_TC_IRQ_MASK; /* Wait for channel inactive */ while (pl08x_phy_channel_busy(phychan)) @@ -227,19 +219,15 @@ static void pl08x_start_txd(struct pl08x_dma_chan *plchan, dev_vdbg(&pl08x->adev->dev, "WRITE channel %d: csrc=0x%08x, cdst=0x%08x, " - "cctl=0x%08x, clli=0x%08x, ccfg=0x%08x\n", - phychan->id, - phychan->csrc, - phychan->cdst, - phychan->cctl, - phychan->clli, - phychan->ccfg); - - writel(phychan->csrc, phychan->base + PL080_CH_SRC_ADDR); - writel(phychan->cdst, phychan->base + PL080_CH_DST_ADDR); - writel(phychan->clli, phychan->base + PL080_CH_LLI); - writel(phychan->cctl, phychan->base + PL080_CH_CONTROL); - writel(phychan->ccfg, phychan->base + PL080_CH_CONFIG); + "clli=0x%08x, cctl=0x%08x, ccfg=0x%08x\n", + phychan->id, lli->src, lli->dst, lli->lli, lli->cctl, + ccfg); + + writel(lli->src, phychan->base + PL080_CH_SRC_ADDR); + writel(lli->dst, phychan->base + PL080_CH_DST_ADDR); + writel(lli->lli, phychan->base + PL080_CH_LLI); + writel(lli->cctl, phychan->base + PL080_CH_CONTROL); + writel(ccfg, phychan->base + PL080_CH_CONFIG); /* Enable the DMA channel */ /* Do not access config register until channel shows as disabled */ @@ -920,13 +908,6 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, */ llis_va[num_llis - 1].cctl |= PL080_CONTROL_TC_IRQ_EN; - /* Now store the channel register values */ - txd->csrc = llis_va[0].src; - txd->cdst = llis_va[0].dst; - txd->clli = llis_va[0].lli; - txd->cctl = llis_va[0].cctl; - /* ccfg will be set at physical channel allocation time */ - #ifdef VERBOSE_DEBUG { int i; diff --git a/include/linux/amba/pl08x.h b/include/linux/amba/pl08x.h index 2c834ed5f41f..29d974562df9 100644 --- a/include/linux/amba/pl08x.h +++ b/include/linux/amba/pl08x.h @@ -95,11 +95,6 @@ struct pl08x_phy_chan { spinlock_t lock; int signal; struct pl08x_dma_chan *serving; - u32 csrc; - u32 cdst; - u32 clli; - u32 cctl; - u32 ccfg; }; /** @@ -118,14 +113,6 @@ struct pl08x_txd { void *llis_va; struct pl08x_channel_data *cd; bool active; - /* - * Settings to be put into the physical channel when we - * trigger this txd - */ - u32 csrc; - u32 cdst; - u32 clli; - u32 cctl; }; /** -- cgit v1.2.3 From 4983a04fd2562986360b646b378f267308bc22c0 Mon Sep 17 00:00:00 2001 From: Russell King - ARM Linux Date: Mon, 3 Jan 2011 22:39:33 +0000 Subject: ARM: PL08x: move ccfg into txd structure The ccfg register is used to configure the channel parameters - the type and direction of transfer, the flow control signal and IRQ mask enables. The type and direction of transfer is known in the relevent prep_* function where a txd is created. The IRQ mask enables are always set, and the flow control signals are always set when we start processing a txd according to phychan->signal. If we store the ccfg value in the txd structure, we can avoid modifying platform data - and even having it in platform data at all. So, remove it from platform data too. Signed-off-by: Russell King Acked-by: Linus Walleij Signed-off-by: Dan Williams --- drivers/dma/amba-pl08x.c | 28 ++++++++++++---------------- include/linux/amba/pl08x.h | 6 +++++- 2 files changed, 17 insertions(+), 17 deletions(-) (limited to 'include/linux') diff --git a/drivers/dma/amba-pl08x.c b/drivers/dma/amba-pl08x.c index a1a18bde6b7f..75f9e2d4b032 100644 --- a/drivers/dma/amba-pl08x.c +++ b/drivers/dma/amba-pl08x.c @@ -194,15 +194,11 @@ static void pl08x_start_txd(struct pl08x_dma_chan *plchan, struct pl08x_driver_data *pl08x = plchan->host; struct pl08x_phy_chan *phychan = plchan->phychan; struct pl08x_lli *lli = &txd->llis_va[0]; - u32 val, ccfg; + u32 val, ccfg = txd->ccfg; plchan->at = txd; - /* Assign the signal to the proper control registers */ - ccfg = plchan->cd->ccfg; - ccfg &= ~(PL080_CONFIG_SRC_SEL_MASK | PL080_CONFIG_DST_SEL_MASK); - - /* If it wasn't set from AMBA, ignore it */ + /* Assign the flow control signal to this channel */ if (txd->direction == DMA_TO_DEVICE) /* Select signal as destination */ ccfg |= phychan->signal << PL080_CONFIG_DST_SEL_SHIFT; @@ -210,9 +206,6 @@ static void pl08x_start_txd(struct pl08x_dma_chan *plchan, /* Select signal as source */ ccfg |= phychan->signal << PL080_CONFIG_SRC_SEL_SHIFT; - /* Always enable error and terminal interrupts */ - ccfg |= PL080_CONFIG_ERR_IRQ_MASK | PL080_CONFIG_TC_IRQ_MASK; - /* Wait for channel inactive */ while (pl08x_phy_channel_busy(phychan)) cpu_relax(); @@ -1161,8 +1154,6 @@ static void dma_set_runtime_config(struct dma_chan *chan, enum dma_slave_buswidth addr_width; u32 maxburst; u32 cctl = 0; - /* Mask out all except src and dst channel */ - u32 ccfg = cd->ccfg & 0x000003DEU; int i; /* Transfer direction */ @@ -1170,13 +1161,11 @@ static void dma_set_runtime_config(struct dma_chan *chan, if (config->direction == DMA_TO_DEVICE) { plchan->runtime_addr = config->dst_addr; cctl |= PL080_CONTROL_SRC_INCR; - ccfg |= PL080_FLOW_MEM2PER << PL080_CONFIG_FLOW_CONTROL_SHIFT; addr_width = config->dst_addr_width; maxburst = config->dst_maxburst; } else if (config->direction == DMA_FROM_DEVICE) { plchan->runtime_addr = config->src_addr; cctl |= PL080_CONTROL_DST_INCR; - ccfg |= PL080_FLOW_PER2MEM << PL080_CONFIG_FLOW_CONTROL_SHIFT; addr_width = config->src_addr_width; maxburst = config->src_maxburst; } else { @@ -1226,16 +1215,15 @@ static void dma_set_runtime_config(struct dma_chan *chan, /* Modify the default channel data to fit PrimeCell request */ cd->cctl = cctl; - cd->ccfg = ccfg; dev_dbg(&pl08x->adev->dev, "configured channel %s (%s) for %s, data width %d, " - "maxburst %d words, LE, CCTL=0x%08x, CCFG=0x%08x\n", + "maxburst %d words, LE, CCTL=0x%08x\n", dma_chan_name(chan), plchan->name, (config->direction == DMA_FROM_DEVICE) ? "RX" : "TX", addr_width, maxburst, - cctl, ccfg); + cctl); } /* @@ -1340,6 +1328,10 @@ static struct pl08x_txd *pl08x_get_txd(struct pl08x_dma_chan *plchan) dma_async_tx_descriptor_init(&txd->tx, &plchan->chan); txd->tx.tx_submit = pl08x_tx_submit; INIT_LIST_HEAD(&txd->node); + + /* Always enable error and terminal interrupts */ + txd->ccfg = PL080_CONFIG_ERR_IRQ_MASK | + PL080_CONFIG_TC_IRQ_MASK; } return txd; } @@ -1369,6 +1361,8 @@ static struct dma_async_tx_descriptor *pl08x_prep_dma_memcpy( /* Set platform data for m2m */ txd->cd = &pl08x->pd->memcpy_channel; + txd->ccfg |= PL080_FLOW_MEM2MEM << PL080_CONFIG_FLOW_CONTROL_SHIFT; + /* Both to be incremented or the code will break */ txd->cd->cctl |= PL080_CONTROL_SRC_INCR | PL080_CONTROL_DST_INCR; txd->len = len; @@ -1424,12 +1418,14 @@ static struct dma_async_tx_descriptor *pl08x_prep_slave_sg( */ txd->direction = direction; if (direction == DMA_TO_DEVICE) { + txd->ccfg |= PL080_FLOW_MEM2PER << PL080_CONFIG_FLOW_CONTROL_SHIFT; txd->srcbus.addr = sgl->dma_address; if (plchan->runtime_addr) txd->dstbus.addr = plchan->runtime_addr; else txd->dstbus.addr = plchan->cd->addr; } else if (direction == DMA_FROM_DEVICE) { + txd->ccfg |= PL080_FLOW_PER2MEM << PL080_CONFIG_FLOW_CONTROL_SHIFT; if (plchan->runtime_addr) txd->srcbus.addr = plchan->runtime_addr; else diff --git a/include/linux/amba/pl08x.h b/include/linux/amba/pl08x.h index 29d974562df9..8e74cb1845dd 100644 --- a/include/linux/amba/pl08x.h +++ b/include/linux/amba/pl08x.h @@ -58,7 +58,6 @@ struct pl08x_channel_data { int max_signal; u32 muxval; u32 cctl; - u32 ccfg; dma_addr_t addr; bool circular_buffer; bool single; @@ -113,6 +112,11 @@ struct pl08x_txd { void *llis_va; struct pl08x_channel_data *cd; bool active; + /* + * Settings to be put into the physical channel when we + * trigger this txd. Other registers are in llis_va[0]. + */ + u32 ccfg; }; /** -- cgit v1.2.3 From 70b5ed6b6d72cd8b1a3d4b7b878a0dd132bec7ba Mon Sep 17 00:00:00 2001 From: Russell King - ARM Linux Date: Mon, 3 Jan 2011 22:40:13 +0000 Subject: ARM: PL08x: move default cctl into txd structure Rather than modifying platform data while preparing a transfer, copy the cctl value into the txd structure and modify the value there. Signed-off-by: Russell King Acked-by: Linus Walleij Signed-off-by: Dan Williams --- drivers/dma/amba-pl08x.c | 21 ++++++--------------- include/linux/amba/pl08x.h | 3 ++- 2 files changed, 8 insertions(+), 16 deletions(-) (limited to 'include/linux') diff --git a/drivers/dma/amba-pl08x.c b/drivers/dma/amba-pl08x.c index f0a29885cb83..6d224d41be2d 100644 --- a/drivers/dma/amba-pl08x.c +++ b/drivers/dma/amba-pl08x.c @@ -577,7 +577,6 @@ static inline size_t pl08x_pre_boundary(u32 addr, size_t len) static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, struct pl08x_txd *txd) { - struct pl08x_channel_data *cd = txd->cd; struct pl08x_bus_data *mbus, *sbus; size_t remainder; int num_llis = 0; @@ -595,17 +594,8 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, pl08x->pool_ctr++; - /* - * Initialize bus values for this transfer - * from the passed optimal values - */ - if (!cd) { - dev_err(&pl08x->adev->dev, "%s no channel data\n", __func__); - return 0; - } - - /* Get the default CCTL from the platform data */ - cctl = cd->cctl; + /* Get the default CCTL */ + cctl = txd->cctl; /* * On the PL080 we have two bus masters and we @@ -1358,11 +1348,11 @@ static struct dma_async_tx_descriptor *pl08x_prep_dma_memcpy( txd->dstbus.addr = dest; /* Set platform data for m2m */ - txd->cd = &pl08x->pd->memcpy_channel; txd->ccfg |= PL080_FLOW_MEM2MEM << PL080_CONFIG_FLOW_CONTROL_SHIFT; + txd->cctl = pl08x->pd->memcpy_channel.cctl; /* Both to be incremented or the code will break */ - txd->cd->cctl |= PL080_CONTROL_SRC_INCR | PL080_CONTROL_DST_INCR; + txd->cctl |= PL080_CONTROL_SRC_INCR | PL080_CONTROL_DST_INCR; txd->len = len; ret = pl08x_prep_channel_resources(plchan, txd); @@ -1415,6 +1405,8 @@ static struct dma_async_tx_descriptor *pl08x_prep_slave_sg( * channel target address dynamically at runtime. */ txd->direction = direction; + txd->cctl = plchan->cd->cctl; + if (direction == DMA_TO_DEVICE) { txd->ccfg |= PL080_FLOW_MEM2PER << PL080_CONFIG_FLOW_CONTROL_SHIFT; txd->srcbus.addr = sgl->dma_address; @@ -1434,7 +1426,6 @@ static struct dma_async_tx_descriptor *pl08x_prep_slave_sg( "%s direction unsupported\n", __func__); return NULL; } - txd->cd = plchan->cd; txd->len = sgl->length; ret = pl08x_prep_channel_resources(plchan, txd); diff --git a/include/linux/amba/pl08x.h b/include/linux/amba/pl08x.h index 8e74cb1845dd..8d9083067d3d 100644 --- a/include/linux/amba/pl08x.h +++ b/include/linux/amba/pl08x.h @@ -110,8 +110,9 @@ struct pl08x_txd { size_t len; dma_addr_t llis_bus; void *llis_va; - struct pl08x_channel_data *cd; bool active; + /* Default cctl value for LLIs */ + u32 cctl; /* * Settings to be put into the physical channel when we * trigger this txd. Other registers are in llis_va[0]. -- cgit v1.2.3 From 30749cb4a40f02a199640011e5ab5c5f60b8482e Mon Sep 17 00:00:00 2001 From: Russell King - ARM Linux Date: Mon, 3 Jan 2011 22:41:13 +0000 Subject: ARM: PL08x: allow AHB master port selection to be configured Platforms need to be able to control which AHB master interface is used, as each AHB master interface may be asymetric. Allow the interfaces used for fetching LLIs, memory, and each peripheral to be configured individually. Signed-off-by: Russell King Acked-by: Linus Walleij Signed-off-by: Dan Williams --- drivers/dma/amba-pl08x.c | 77 ++++++++++++++++++++++++++-------------------- include/linux/amba/pl08x.h | 15 +++++++-- 2 files changed, 56 insertions(+), 36 deletions(-) (limited to 'include/linux') diff --git a/drivers/dma/amba-pl08x.c b/drivers/dma/amba-pl08x.c index 4ee0ab19b1b6..986c12775b0b 100644 --- a/drivers/dma/amba-pl08x.c +++ b/drivers/dma/amba-pl08x.c @@ -126,6 +126,8 @@ struct pl08x_lli { * @phy_chans: array of data for the physical channels * @pool: a pool for the LLI descriptors * @pool_ctr: counter of LLIs in the pool + * @lli_buses: bitmask to or in to LLI pointer selecting AHB port for LLI fetches + * @mem_buses: set to indicate memory transfers on AHB2. * @lock: a spinlock for this struct */ struct pl08x_driver_data { @@ -138,6 +140,8 @@ struct pl08x_driver_data { struct pl08x_phy_chan *phy_chans; struct dma_pool *pool; int pool_ctr; + u8 lli_buses; + u8 mem_buses; spinlock_t lock; }; @@ -526,20 +530,12 @@ static int pl08x_fill_lli_for_desc(struct pl08x_driver_data *pl08x, BUG_ON(num_llis >= MAX_NUM_TSFR_LLIS); - llis_va[num_llis].cctl = cctl; - llis_va[num_llis].src = txd->srcbus.addr; - llis_va[num_llis].dst = txd->dstbus.addr; - - /* - * On versions with dual masters, you can optionally AND on - * PL080_LLI_LM_AHB2 to the LLI to tell the hardware to read - * in new LLIs with that controller, but we always try to - * choose AHB1 to point into memory. The idea is to have AHB2 - * fixed on the peripheral and AHB1 messing around in the - * memory. So we don't manipulate this bit currently. - */ - + llis_va[num_llis].cctl = cctl; + llis_va[num_llis].src = txd->srcbus.addr; + llis_va[num_llis].dst = txd->dstbus.addr; llis_va[num_llis].lli = llis_bus + (num_llis + 1) * sizeof(struct pl08x_lli); + if (pl08x->lli_buses & PL08X_AHB2) + llis_va[num_llis].lli |= PL080_LLI_LM_AHB2; if (cctl & PL080_CONTROL_SRC_INCR) txd->srcbus.addr += len; @@ -639,13 +635,6 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, pl08x_choose_master_bus(&txd->srcbus, &txd->dstbus, &mbus, &sbus, cctl); - - /* - * The lowest bit of the LLI register - * is also used to indicate which master to - * use for reading the LLIs. - */ - if (txd->len < mbus->buswidth) { /* * Less than a bus width available @@ -1282,6 +1271,23 @@ static int pl08x_prep_channel_resources(struct pl08x_dma_chan *plchan, return 0; } +/* + * Given the source and destination available bus masks, select which + * will be routed to each port. We try to have source and destination + * on separate ports, but always respect the allowable settings. + */ +static u32 pl08x_select_bus(struct pl08x_driver_data *pl08x, u8 src, u8 dst) +{ + u32 cctl = 0; + + if (!(dst & PL08X_AHB1) || ((dst & PL08X_AHB2) && (src & PL08X_AHB1))) + cctl |= PL080_CONTROL_DST_AHB2; + if (!(src & PL08X_AHB1) || ((src & PL08X_AHB2) && !(dst & PL08X_AHB2))) + cctl |= PL080_CONTROL_SRC_AHB2; + + return cctl; +} + static struct pl08x_txd *pl08x_get_txd(struct pl08x_dma_chan *plchan) { struct pl08x_txd *txd = kzalloc(sizeof(struct pl08x_txd), GFP_NOWAIT); @@ -1330,15 +1336,9 @@ static struct dma_async_tx_descriptor *pl08x_prep_dma_memcpy( /* Both to be incremented or the code will break */ txd->cctl |= PL080_CONTROL_SRC_INCR | PL080_CONTROL_DST_INCR; - /* - * On the PL080 we have two bus masters and we should select one for - * source and one for destination. We try to use AHB2 for the bus - * which does not increment (typically the peripheral) else we just - * choose something. - */ if (pl08x->vd->dualmaster) - /* Source increments, use AHB2 for destination */ - txd->cctl |= PL080_CONTROL_DST_AHB2; + txd->cctl |= pl08x_select_bus(pl08x, + pl08x->mem_buses, pl08x->mem_buses); ret = pl08x_prep_channel_resources(plchan, txd); if (ret) @@ -1359,6 +1359,7 @@ static struct dma_async_tx_descriptor *pl08x_prep_slave_sg( struct pl08x_dma_chan *plchan = to_pl08x_chan(chan); struct pl08x_driver_data *pl08x = plchan->host; struct pl08x_txd *txd; + u8 src_buses, dst_buses; int ret; /* @@ -1403,31 +1404,31 @@ static struct dma_async_tx_descriptor *pl08x_prep_slave_sg( if (direction == DMA_TO_DEVICE) { txd->ccfg |= PL080_FLOW_MEM2PER << PL080_CONFIG_FLOW_CONTROL_SHIFT; txd->cctl |= PL080_CONTROL_SRC_INCR; - if (pl08x->vd->dualmaster) - /* Source increments, use AHB2 for destination */ - txd->cctl |= PL080_CONTROL_DST_AHB2; txd->srcbus.addr = sgl->dma_address; if (plchan->runtime_addr) txd->dstbus.addr = plchan->runtime_addr; else txd->dstbus.addr = plchan->cd->addr; + src_buses = pl08x->mem_buses; + dst_buses = plchan->cd->periph_buses; } else if (direction == DMA_FROM_DEVICE) { txd->ccfg |= PL080_FLOW_PER2MEM << PL080_CONFIG_FLOW_CONTROL_SHIFT; txd->cctl |= PL080_CONTROL_DST_INCR; - if (pl08x->vd->dualmaster) - /* Destination increments, use AHB2 for source */ - txd->cctl |= PL080_CONTROL_SRC_AHB2; if (plchan->runtime_addr) txd->srcbus.addr = plchan->runtime_addr; else txd->srcbus.addr = plchan->cd->addr; txd->dstbus.addr = sgl->dma_address; + src_buses = plchan->cd->periph_buses; + dst_buses = pl08x->mem_buses; } else { dev_err(&pl08x->adev->dev, "%s direction unsupported\n", __func__); return NULL; } + txd->cctl |= pl08x_select_bus(pl08x, src_buses, dst_buses); + ret = pl08x_prep_channel_resources(plchan, txd); if (ret) return NULL; @@ -1879,6 +1880,14 @@ static int pl08x_probe(struct amba_device *adev, struct amba_id *id) pl08x->adev = adev; pl08x->vd = vd; + /* By default, AHB1 only. If dualmaster, from platform */ + pl08x->lli_buses = PL08X_AHB1; + pl08x->mem_buses = PL08X_AHB1; + if (pl08x->vd->dualmaster) { + pl08x->lli_buses = pl08x->pd->lli_buses; + pl08x->mem_buses = pl08x->pd->mem_buses; + } + /* A DMA memory pool for LLIs, align on 1-byte boundary */ pl08x->pool = dma_pool_create(DRIVER_NAME, &pl08x->adev->dev, PL08X_LLI_TSFR_SIZE, PL08X_ALIGN, 0); diff --git a/include/linux/amba/pl08x.h b/include/linux/amba/pl08x.h index 8d9083067d3d..f858651027fd 100644 --- a/include/linux/amba/pl08x.h +++ b/include/linux/amba/pl08x.h @@ -25,6 +25,12 @@ struct pl08x_lli; struct pl08x_driver_data; +/* Bitmasks for selecting AHB ports for DMA transfers */ +enum { + PL08X_AHB1 = (1 << 0), + PL08X_AHB2 = (1 << 1) +}; + /** * struct pl08x_channel_data - data structure to pass info between * platform and PL08x driver regarding channel configuration @@ -51,6 +57,8 @@ struct pl08x_driver_data; * round round round) * @single: the device connected to this channel will request single * DMA transfers, not bursts. (Bursts are default.) + * @periph_buses: the device connected to this channel is accessible via + * these buses (use PL08X_AHB1 | PL08X_AHB2). */ struct pl08x_channel_data { char *bus_id; @@ -61,6 +69,7 @@ struct pl08x_channel_data { dma_addr_t addr; bool circular_buffer; bool single; + u8 periph_buses; }; /** @@ -193,8 +202,8 @@ struct pl08x_dma_chan { * less than zero, else it returns the allocated signal number * @put_signal: indicate to the platform that this physical signal is not * running any DMA transfer and multiplexing can be recycled - * @bus_bit_lli: Bit[0] of the address indicated which AHB bus master the - * LLI addresses are on 0/1 Master 1/2. + * @lli_buses: buses which LLIs can be fetched from: PL08X_AHB1 | PL08X_AHB2 + * @mem_buses: buses which memory can be accessed from: PL08X_AHB1 | PL08X_AHB2 */ struct pl08x_platform_data { struct pl08x_channel_data *slave_channels; @@ -202,6 +211,8 @@ struct pl08x_platform_data { struct pl08x_channel_data memcpy_channel; int (*get_signal)(struct pl08x_dma_chan *); void (*put_signal)(struct pl08x_dma_chan *); + u8 lli_buses; + u8 mem_buses; }; #ifdef CONFIG_AMBA_PL08X -- cgit v1.2.3 From d7244e9a27a3da27d62aabf560ee828d7991493e Mon Sep 17 00:00:00 2001 From: Russell King - ARM Linux Date: Mon, 3 Jan 2011 22:43:35 +0000 Subject: ARM: PL08x: shrink srcbus/dstbus in txd structure We only need to store the dma address. Signed-off-by: Russell King Acked-by: Linus Walleij Signed-off-by: Dan Williams --- drivers/dma/amba-pl08x.c | 20 ++++++++++---------- include/linux/amba/pl08x.h | 4 ++-- 2 files changed, 12 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/drivers/dma/amba-pl08x.c b/drivers/dma/amba-pl08x.c index 39970c5038d6..8d8d33ce5684 100644 --- a/drivers/dma/amba-pl08x.c +++ b/drivers/dma/amba-pl08x.c @@ -595,8 +595,8 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, bd.txd = txd; bd.pl08x = pl08x; - bd.srcbus.addr = txd->srcbus.addr; - bd.dstbus.addr = txd->dstbus.addr; + bd.srcbus.addr = txd->src_addr; + bd.dstbus.addr = txd->dst_addr; /* Find maximum width of the source bus */ bd.srcbus.maxwidth = @@ -1313,8 +1313,8 @@ static struct dma_async_tx_descriptor *pl08x_prep_dma_memcpy( } txd->direction = DMA_NONE; - txd->srcbus.addr = src; - txd->dstbus.addr = dest; + txd->src_addr = src; + txd->dst_addr = dest; txd->len = len; /* Set platform data for m2m */ @@ -1393,21 +1393,21 @@ static struct dma_async_tx_descriptor *pl08x_prep_slave_sg( if (direction == DMA_TO_DEVICE) { txd->ccfg |= PL080_FLOW_MEM2PER << PL080_CONFIG_FLOW_CONTROL_SHIFT; txd->cctl |= PL080_CONTROL_SRC_INCR; - txd->srcbus.addr = sgl->dma_address; + txd->src_addr = sgl->dma_address; if (plchan->runtime_addr) - txd->dstbus.addr = plchan->runtime_addr; + txd->dst_addr = plchan->runtime_addr; else - txd->dstbus.addr = plchan->cd->addr; + txd->dst_addr = plchan->cd->addr; src_buses = pl08x->mem_buses; dst_buses = plchan->cd->periph_buses; } else if (direction == DMA_FROM_DEVICE) { txd->ccfg |= PL080_FLOW_PER2MEM << PL080_CONFIG_FLOW_CONTROL_SHIFT; txd->cctl |= PL080_CONTROL_DST_INCR; if (plchan->runtime_addr) - txd->srcbus.addr = plchan->runtime_addr; + txd->src_addr = plchan->runtime_addr; else - txd->srcbus.addr = plchan->cd->addr; - txd->dstbus.addr = sgl->dma_address; + txd->src_addr = plchan->cd->addr; + txd->dst_addr = sgl->dma_address; src_buses = plchan->cd->periph_buses; dst_buses = pl08x->mem_buses; } else { diff --git a/include/linux/amba/pl08x.h b/include/linux/amba/pl08x.h index f858651027fd..4e9e71860e1c 100644 --- a/include/linux/amba/pl08x.h +++ b/include/linux/amba/pl08x.h @@ -114,8 +114,8 @@ struct pl08x_txd { struct dma_async_tx_descriptor tx; struct list_head node; enum dma_data_direction direction; - struct pl08x_bus_data srcbus; - struct pl08x_bus_data dstbus; + dma_addr_t src_addr; + dma_addr_t dst_addr; size_t len; dma_addr_t llis_bus; void *llis_va; -- cgit v1.2.3 From 15c17232fbd1f7687c740c3c26f9e7f337bd9e36 Mon Sep 17 00:00:00 2001 From: Russell King - ARM Linux Date: Mon, 3 Jan 2011 22:44:36 +0000 Subject: ARM: PL08x: rename 'desc_list' as 'pend_list' This 'desc_list' is actually a list of pending descriptors, so name it after its function (pending list) rather than what it contains (descriptors). Signed-off-by: Russell King Acked-by: Linus Walleij Signed-off-by: Dan Williams --- drivers/dma/amba-pl08x.c | 20 ++++++++++---------- include/linux/amba/pl08x.h | 4 ++-- 2 files changed, 12 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/drivers/dma/amba-pl08x.c b/drivers/dma/amba-pl08x.c index fb469dedcdf3..433b9e747f75 100644 --- a/drivers/dma/amba-pl08x.c +++ b/drivers/dma/amba-pl08x.c @@ -348,9 +348,9 @@ static u32 pl08x_getbytes_chan(struct pl08x_dma_chan *plchan) } /* Sum up all queued transactions */ - if (!list_empty(&plchan->desc_list)) { + if (!list_empty(&plchan->pend_list)) { struct pl08x_txd *txdi; - list_for_each_entry(txdi, &plchan->desc_list, node) { + list_for_each_entry(txdi, &plchan->pend_list, node) { bytes += txdi->len; } } @@ -880,9 +880,9 @@ static void pl08x_free_txd_list(struct pl08x_driver_data *pl08x, struct pl08x_txd *txdi = NULL; struct pl08x_txd *next; - if (!list_empty(&plchan->desc_list)) { + if (!list_empty(&plchan->pend_list)) { list_for_each_entry_safe(txdi, - next, &plchan->desc_list, node) { + next, &plchan->pend_list, node) { list_del(&txdi->node); pl08x_free_txd(pl08x, txdi); } @@ -1183,10 +1183,10 @@ static void pl08x_issue_pending(struct dma_chan *chan) } /* Take the first element in the queue and execute it */ - if (!list_empty(&plchan->desc_list)) { + if (!list_empty(&plchan->pend_list)) { struct pl08x_txd *next; - next = list_first_entry(&plchan->desc_list, + next = list_first_entry(&plchan->pend_list, struct pl08x_txd, node); list_del(&next->node); @@ -1213,7 +1213,7 @@ static int pl08x_prep_channel_resources(struct pl08x_dma_chan *plchan, spin_lock_irqsave(&plchan->lock, plchan->lockflags); - list_add_tail(&txd->node, &plchan->desc_list); + list_add_tail(&txd->node, &plchan->pend_list); /* * See if we already have a physical channel allocated, @@ -1571,10 +1571,10 @@ static void pl08x_tasklet(unsigned long data) * If a new descriptor is queued, set it up * plchan->at is NULL here */ - if (!list_empty(&plchan->desc_list)) { + if (!list_empty(&plchan->pend_list)) { struct pl08x_txd *next; - next = list_first_entry(&plchan->desc_list, + next = list_first_entry(&plchan->pend_list, struct pl08x_txd, node); list_del(&next->node); @@ -1736,7 +1736,7 @@ static int pl08x_dma_init_virtual_channels(struct pl08x_driver_data *pl08x, chan->lc = 0; spin_lock_init(&chan->lock); - INIT_LIST_HEAD(&chan->desc_list); + INIT_LIST_HEAD(&chan->pend_list); tasklet_init(&chan->tasklet, pl08x_tasklet, (unsigned long) chan); diff --git a/include/linux/amba/pl08x.h b/include/linux/amba/pl08x.h index 4e9e71860e1c..08a9024e2d2f 100644 --- a/include/linux/amba/pl08x.h +++ b/include/linux/amba/pl08x.h @@ -158,7 +158,7 @@ enum pl08x_dma_chan_state { * @runtime_direction: current direction of this channel according to * runtime config * @lc: last completed transaction on this channel - * @desc_list: queued transactions pending on this channel + * @pend_list: queued transactions pending on this channel * @at: active transaction on this channel * @lockflags: sometimes we let a lock last between two function calls, * especially prep/submit, and then we need to store the IRQ flags @@ -179,7 +179,7 @@ struct pl08x_dma_chan { dma_addr_t runtime_addr; enum dma_data_direction runtime_direction; dma_cookie_t lc; - struct list_head desc_list; + struct list_head pend_list; struct pl08x_txd *at; unsigned long lockflags; spinlock_t lock; -- cgit v1.2.3 From 8087aacda040bdbf84940712d132ce80c30b9d5d Mon Sep 17 00:00:00 2001 From: Russell King - ARM Linux Date: Mon, 3 Jan 2011 22:45:17 +0000 Subject: ARM: PL08x: introduce 'phychan_hold' to hold on to physical channels Introduce 'phychan_hold' to hold on to physical DMA channels while we're preparing a new descriptor for it. This will be incremented when we allocate a physical channel and set the MUX registers during the preparation of the TXD, and will only be decremented when the TXD is submitted. This prevents the physical channel being given up before the new TXD is placed on the queue. Signed-off-by: Russell King Acked-by: Linus Walleij Signed-off-by: Dan Williams --- drivers/dma/amba-pl08x.c | 11 +++++++++++ include/linux/amba/pl08x.h | 3 +++ 2 files changed, 14 insertions(+) (limited to 'include/linux') diff --git a/drivers/dma/amba-pl08x.c b/drivers/dma/amba-pl08x.c index 650e2bbc7aad..bf6f7d02c9f6 100644 --- a/drivers/dma/amba-pl08x.c +++ b/drivers/dma/amba-pl08x.c @@ -959,6 +959,7 @@ static int prep_phy_channel(struct pl08x_dma_chan *plchan, ch->signal, plchan->name); + plchan->phychan_hold++; plchan->phychan = ch; return 0; @@ -998,6 +999,8 @@ static dma_cookie_t pl08x_tx_submit(struct dma_async_tx_descriptor *tx) /* Do this memcpy whenever there is a channel ready */ plchan->state = PL08X_CHAN_WAITING; plchan->waiting = txd; + } else { + plchan->phychan_hold--; } /* This unlock follows the lock in the prep() function */ @@ -1585,6 +1588,7 @@ static void pl08x_tasklet(unsigned long data) */ plchan->lc = txd->tx.cookie; } + /* * If a new descriptor is queued, set it up * plchan->at is NULL here @@ -1598,6 +1602,12 @@ static void pl08x_tasklet(unsigned long data) list_del(&next->node); pl08x_start_txd(plchan, next); + } else if (plchan->phychan_hold) { + /* + * This channel is still in use - we have a new txd being + * prepared and will soon be queued. Don't give up the + * physical channel. + */ } else { struct pl08x_dma_chan *waiting = NULL; @@ -1625,6 +1635,7 @@ static void pl08x_tasklet(unsigned long data) ret = prep_phy_channel(waiting, waiting->waiting); BUG_ON(ret); + waiting->phychan_hold--; waiting->state = PL08X_CHAN_RUNNING; waiting->waiting = NULL; pl08x_issue_pending(&waiting->chan); diff --git a/include/linux/amba/pl08x.h b/include/linux/amba/pl08x.h index 08a9024e2d2f..95b76ea1829f 100644 --- a/include/linux/amba/pl08x.h +++ b/include/linux/amba/pl08x.h @@ -151,6 +151,8 @@ enum pl08x_dma_chan_state { * struct pl08x_dma_chan - this structure wraps a DMA ENGINE channel * @chan: wrappped abstract channel * @phychan: the physical channel utilized by this channel, if there is one + * @phychan_hold: if non-zero, hold on to the physical channel even if we + * have no pending entries * @tasklet: tasklet scheduled by the IRQ to handle actual work etc * @name: name of channel * @cd: channel platform data @@ -173,6 +175,7 @@ enum pl08x_dma_chan_state { struct pl08x_dma_chan { struct dma_chan chan; struct pl08x_phy_chan *phychan; + int phychan_hold; struct tasklet_struct tasklet; char *name; struct pl08x_channel_data *cd; -- cgit v1.2.3 From c370e594efe2993620d24d41a78f325102e99d1c Mon Sep 17 00:00:00 2001 From: Russell King - ARM Linux Date: Mon, 3 Jan 2011 22:45:37 +0000 Subject: ARM: PL08x: fix locking between prepare function and submit function The PL08x driver holds on to the channel lock with interrupts disabled between the prepare and the subsequent submit API functions. This means that the locking state when the prepare function returns is dependent on whether it suceeeds or not. It did this to ensure that the physical channel wasn't released, and as it used to add the descriptor onto the pending list at prepare time rather than submit time. Now that we have reorganized the code to remove those reasons, we can now safely release the spinlock at the end of preparation and reacquire it in our submit function. Signed-off-by: Russell King Acked-by: Linus Walleij Signed-off-by: Dan Williams --- drivers/dma/amba-pl08x.c | 28 +++++++++------------------- include/linux/amba/pl08x.h | 4 ---- 2 files changed, 9 insertions(+), 23 deletions(-) (limited to 'include/linux') diff --git a/drivers/dma/amba-pl08x.c b/drivers/dma/amba-pl08x.c index bf6f7d02c9f6..1c9f712520d6 100644 --- a/drivers/dma/amba-pl08x.c +++ b/drivers/dma/amba-pl08x.c @@ -981,6 +981,9 @@ static dma_cookie_t pl08x_tx_submit(struct dma_async_tx_descriptor *tx) { struct pl08x_dma_chan *plchan = to_pl08x_chan(tx->chan); struct pl08x_txd *txd = to_pl08x_txd(tx); + unsigned long flags; + + spin_lock_irqsave(&plchan->lock, flags); plchan->chan.cookie += 1; if (plchan->chan.cookie < 0) @@ -1003,8 +1006,7 @@ static dma_cookie_t pl08x_tx_submit(struct dma_async_tx_descriptor *tx) plchan->phychan_hold--; } - /* This unlock follows the lock in the prep() function */ - spin_unlock_irqrestore(&plchan->lock, plchan->lockflags); + spin_unlock_irqrestore(&plchan->lock, flags); return tx->cookie; } @@ -1225,9 +1227,9 @@ static void pl08x_issue_pending(struct dma_chan *chan) static int pl08x_prep_channel_resources(struct pl08x_dma_chan *plchan, struct pl08x_txd *txd) { - int num_llis; struct pl08x_driver_data *pl08x = plchan->host; - int ret; + unsigned long flags; + int num_llis, ret; num_llis = pl08x_fill_llis_for_desc(pl08x, txd); if (!num_llis) { @@ -1235,7 +1237,7 @@ static int pl08x_prep_channel_resources(struct pl08x_dma_chan *plchan, return -EINVAL; } - spin_lock_irqsave(&plchan->lock, plchan->lockflags); + spin_lock_irqsave(&plchan->lock, flags); /* * See if we already have a physical channel allocated, @@ -1258,7 +1260,7 @@ static int pl08x_prep_channel_resources(struct pl08x_dma_chan *plchan, if (plchan->slave) { pl08x_free_txd_list(pl08x, plchan); pl08x_free_txd(pl08x, txd); - spin_unlock_irqrestore(&plchan->lock, plchan->lockflags); + spin_unlock_irqrestore(&plchan->lock, flags); return -EBUSY; } } else @@ -1272,11 +1274,7 @@ static int pl08x_prep_channel_resources(struct pl08x_dma_chan *plchan, if (plchan->state == PL08X_CHAN_IDLE) plchan->state = PL08X_CHAN_PAUSED; - /* - * Notice that we leave plchan->lock locked on purpose: - * it will be unlocked in the subsequent tx_submit() - * call. This is a consequence of the current API. - */ + spin_unlock_irqrestore(&plchan->lock, flags); return 0; } @@ -1355,10 +1353,6 @@ static struct dma_async_tx_descriptor *pl08x_prep_dma_memcpy( ret = pl08x_prep_channel_resources(plchan, txd); if (ret) return NULL; - /* - * NB: the channel lock is held at this point so tx_submit() - * must be called in direct succession. - */ return &txd->tx; } @@ -1444,10 +1438,6 @@ static struct dma_async_tx_descriptor *pl08x_prep_slave_sg( ret = pl08x_prep_channel_resources(plchan, txd); if (ret) return NULL; - /* - * NB: the channel lock is held at this point so tx_submit() - * must be called in direct succession. - */ return &txd->tx; } diff --git a/include/linux/amba/pl08x.h b/include/linux/amba/pl08x.h index 95b76ea1829f..933b4ed12be5 100644 --- a/include/linux/amba/pl08x.h +++ b/include/linux/amba/pl08x.h @@ -162,9 +162,6 @@ enum pl08x_dma_chan_state { * @lc: last completed transaction on this channel * @pend_list: queued transactions pending on this channel * @at: active transaction on this channel - * @lockflags: sometimes we let a lock last between two function calls, - * especially prep/submit, and then we need to store the IRQ flags - * in the channel state, here * @lock: a lock for this channel data * @host: a pointer to the host (internal use) * @state: whether the channel is idle, paused, running etc @@ -184,7 +181,6 @@ struct pl08x_dma_chan { dma_cookie_t lc; struct list_head pend_list; struct pl08x_txd *at; - unsigned long lockflags; spinlock_t lock; struct pl08x_driver_data *host; enum pl08x_dma_chan_state state; -- cgit v1.2.3 From cf24dc85ff29a41abd8e73730e5feb22b2666bd3 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Fri, 19 Feb 2010 15:39:52 +0100 Subject: mtd: OneNAND: add enable / disable methods to onenand_chip Add enable / disable methods called from get_device() / release_device(). These can be used, for example, to allow the driver to prevent the voltage regulator from being put to sleep while OneNAND is in use. Signed-off-by: Adrian Hunter Signed-off-by: David Woodhouse --- drivers/mtd/onenand/onenand_base.c | 4 ++++ include/linux/mtd/onenand.h | 2 ++ 2 files changed, 6 insertions(+) (limited to 'include/linux') diff --git a/drivers/mtd/onenand/onenand_base.c b/drivers/mtd/onenand/onenand_base.c index 2d7c90de3144..2edef5868ace 100644 --- a/drivers/mtd/onenand/onenand_base.c +++ b/drivers/mtd/onenand/onenand_base.c @@ -948,6 +948,8 @@ static int onenand_get_device(struct mtd_info *mtd, int new_state) if (this->state == FL_READY) { this->state = new_state; spin_unlock(&this->chip_lock); + if (new_state != FL_PM_SUSPENDED && this->enable) + this->enable(mtd); break; } if (new_state == FL_PM_SUSPENDED) { @@ -974,6 +976,8 @@ static void onenand_release_device(struct mtd_info *mtd) { struct onenand_chip *this = mtd->priv; + if (this->state != FL_PM_SUSPENDED && this->disable) + this->disable(mtd); /* Release the chip */ spin_lock(&this->chip_lock); this->state = FL_READY; diff --git a/include/linux/mtd/onenand.h b/include/linux/mtd/onenand.h index 6da3fe314828..ae418e41d8f5 100644 --- a/include/linux/mtd/onenand.h +++ b/include/linux/mtd/onenand.h @@ -118,6 +118,8 @@ struct onenand_chip { int (*chip_probe)(struct mtd_info *mtd); int (*block_markbad)(struct mtd_info *mtd, loff_t ofs); int (*scan_bbt)(struct mtd_info *mtd); + int (*enable)(struct mtd_info *mtd); + int (*disable)(struct mtd_info *mtd); struct completion complete; int irq; -- cgit v1.2.3 From 0e4ca7e5101e7f4054452b8d71c535eec64a187b Mon Sep 17 00:00:00 2001 From: Anatolij Gustschin Date: Thu, 16 Dec 2010 23:42:14 +0100 Subject: mtd: add writebufsize field to mtd_info struct This field will be used to indicate the write buffer size of the MTD device. UBI will set it's minimal I/O unit size (min_io_size) to the indicated write buffer size. By this change we intend to fix failed recovery of UBIFS partitions we currently observe on NOR flash when mounting the partition after unclean unmount. Currently the min_io_size is set to mtd->writesize (which is 1 byte for NOR flash). But flash programming is often done from prepared write buffer containing multiple bytes and is performed in one programming operation which could be interrupted by a power cut or a system reset causing corrupted (partially written) areas in a flash sector. Knowing the size of potentially corrupted areas UBIFS scanning and recovery algorithms are able to perform successful recovery. In case of NOR flash minimal I/O size must be equal to the maximal size of the write buffer used by embedded flash programming algorithm. In case of NAND flash mtd->writebufsize should be equivalent to mtd->writesize. The subsequent patches will add mtd->writebufsize initialization where needed. Signed-off-by: Anatolij Gustschin Signed-off-by: Artem Bityutskiy Signed-off-by: David Woodhouse --- include/linux/mtd/mtd.h | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'include/linux') diff --git a/include/linux/mtd/mtd.h b/include/linux/mtd/mtd.h index fe8d77ebec13..9d5306bad117 100644 --- a/include/linux/mtd/mtd.h +++ b/include/linux/mtd/mtd.h @@ -144,6 +144,17 @@ struct mtd_info { */ uint32_t writesize; + /* + * Size of the write buffer used by the MTD. MTD devices having a write + * buffer can write multiple writesize chunks at a time. E.g. while + * writing 4 * writesize bytes to a device with 2 * writesize bytes + * buffer the MTD driver can (but doesn't have to) do 2 writesize + * operations, but not 4. Currently, all NANDs have writebufsize + * equivalent to writesize (NAND page size). Some NOR flashes do have + * writebufsize greater than writesize. + */ + uint32_t writebufsize; + uint32_t oobsize; // Amount of OOB data per block (e.g. 16) uint32_t oobavail; // Available OOB bytes per block -- cgit v1.2.3 From 26fcaf60fe3861409eb4c455c5c0d0f00f599b08 Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Fri, 7 Jan 2011 01:42:31 +0100 Subject: PM: Fix oops in suspend/hibernate code related to failing ioremap() When ioremap() fails (which might happen for some reason), we nicely oops in suspend_nvs_save() due to NULL dereference by memcpy() in there. Fail gracefully instead. Signed-off-by: Jiri Slaby Signed-off-by: Rafael J. Wysocki Signed-off-by: Len Brown --- drivers/acpi/sleep.c | 5 ++--- include/linux/suspend.h | 4 ++-- kernel/power/nvs.c | 8 +++++++- 3 files changed, 11 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/drivers/acpi/sleep.c b/drivers/acpi/sleep.c index febb153b5a68..d8bca6c90719 100644 --- a/drivers/acpi/sleep.c +++ b/drivers/acpi/sleep.c @@ -124,8 +124,7 @@ static int acpi_pm_freeze(void) static int acpi_pm_pre_suspend(void) { acpi_pm_freeze(); - suspend_nvs_save(); - return 0; + return suspend_nvs_save(); } /** @@ -151,7 +150,7 @@ static int acpi_pm_prepare(void) { int error = __acpi_pm_prepare(); if (!error) - acpi_pm_pre_suspend(); + error = acpi_pm_pre_suspend(); return error; } diff --git a/include/linux/suspend.h b/include/linux/suspend.h index 26697514c5ec..acb7d911bb0c 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -262,7 +262,7 @@ static inline bool system_entering_hibernation(void) { return false; } extern int suspend_nvs_register(unsigned long start, unsigned long size); extern int suspend_nvs_alloc(void); extern void suspend_nvs_free(void); -extern void suspend_nvs_save(void); +extern int suspend_nvs_save(void); extern void suspend_nvs_restore(void); #else /* CONFIG_SUSPEND_NVS */ static inline int suspend_nvs_register(unsigned long a, unsigned long b) @@ -271,7 +271,7 @@ static inline int suspend_nvs_register(unsigned long a, unsigned long b) } static inline int suspend_nvs_alloc(void) { return 0; } static inline void suspend_nvs_free(void) {} -static inline void suspend_nvs_save(void) {} +static inline int suspend_nvs_save(void) {} static inline void suspend_nvs_restore(void) {} #endif /* CONFIG_SUSPEND_NVS */ diff --git a/kernel/power/nvs.c b/kernel/power/nvs.c index 1836db60bbb6..57c6fabbb6b6 100644 --- a/kernel/power/nvs.c +++ b/kernel/power/nvs.c @@ -105,7 +105,7 @@ int suspend_nvs_alloc(void) /** * suspend_nvs_save - save NVS memory regions */ -void suspend_nvs_save(void) +int suspend_nvs_save(void) { struct nvs_page *entry; @@ -114,8 +114,14 @@ void suspend_nvs_save(void) list_for_each_entry(entry, &nvs_list, node) if (entry->data) { entry->kaddr = ioremap(entry->phys_start, entry->size); + if (!entry->kaddr) { + suspend_nvs_free(); + return -ENOMEM; + } memcpy(entry->data, entry->kaddr, entry->size); } + + return 0; } /** -- cgit v1.2.3 From 976513dbfc1547c7b1822566923058655f0c32fd Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Fri, 7 Jan 2011 01:43:44 +0100 Subject: PM / ACPI: Move NVS saving and restoring code to drivers/acpi The saving of the ACPI NVS area during hibernation and suspend and restoring it during the subsequent resume is entirely specific to ACPI, so move it to drivers/acpi and drop the CONFIG_SUSPEND_NVS configuration option which is redundant. Signed-off-by: Rafael J. Wysocki Signed-off-by: Len Brown --- arch/x86/kernel/e820.c | 1 + drivers/acpi/Makefile | 2 +- drivers/acpi/internal.h | 8 +++ drivers/acpi/nvs.c | 142 ++++++++++++++++++++++++++++++++++++++++++++++++ include/linux/acpi.h | 9 +++ include/linux/suspend.h | 17 ------ kernel/power/Kconfig | 5 -- kernel/power/Makefile | 1 - kernel/power/nvs.c | 142 ------------------------------------------------ 9 files changed, 161 insertions(+), 166 deletions(-) create mode 100644 drivers/acpi/nvs.c delete mode 100644 kernel/power/nvs.c (limited to 'include/linux') diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index 0c2b7ef7a34d..294f26da0c0c 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include diff --git a/drivers/acpi/Makefile b/drivers/acpi/Makefile index 3d031d02e54b..9cc9f2c4da79 100644 --- a/drivers/acpi/Makefile +++ b/drivers/acpi/Makefile @@ -24,7 +24,7 @@ acpi-y += atomicio.o # sleep related files acpi-y += wakeup.o acpi-y += sleep.o -acpi-$(CONFIG_ACPI_SLEEP) += proc.o +acpi-$(CONFIG_ACPI_SLEEP) += proc.o nvs.o # diff --git a/drivers/acpi/internal.h b/drivers/acpi/internal.h index a212bfeddf8c..7c23b76e8eca 100644 --- a/drivers/acpi/internal.h +++ b/drivers/acpi/internal.h @@ -82,8 +82,16 @@ extern int acpi_sleep_init(void); #ifdef CONFIG_ACPI_SLEEP int acpi_sleep_proc_init(void); +int suspend_nvs_alloc(void); +void suspend_nvs_free(void); +int suspend_nvs_save(void); +void suspend_nvs_restore(void); #else static inline int acpi_sleep_proc_init(void) { return 0; } +static inline int suspend_nvs_alloc(void) { return 0; } +static inline void suspend_nvs_free(void) {} +static inline int suspend_nvs_save(void) {} +static inline void suspend_nvs_restore(void) {} #endif #endif /* _ACPI_INTERNAL_H_ */ diff --git a/drivers/acpi/nvs.c b/drivers/acpi/nvs.c new file mode 100644 index 000000000000..57c6fabbb6b6 --- /dev/null +++ b/drivers/acpi/nvs.c @@ -0,0 +1,142 @@ +/* + * linux/kernel/power/hibernate_nvs.c - Routines for handling NVS memory + * + * Copyright (C) 2008,2009 Rafael J. Wysocki , Novell Inc. + * + * This file is released under the GPLv2. + */ + +#include +#include +#include +#include +#include +#include + +/* + * Platforms, like ACPI, may want us to save some memory used by them during + * suspend and to restore the contents of this memory during the subsequent + * resume. The code below implements a mechanism allowing us to do that. + */ + +struct nvs_page { + unsigned long phys_start; + unsigned int size; + void *kaddr; + void *data; + struct list_head node; +}; + +static LIST_HEAD(nvs_list); + +/** + * suspend_nvs_register - register platform NVS memory region to save + * @start - physical address of the region + * @size - size of the region + * + * The NVS region need not be page-aligned (both ends) and we arrange + * things so that the data from page-aligned addresses in this region will + * be copied into separate RAM pages. + */ +int suspend_nvs_register(unsigned long start, unsigned long size) +{ + struct nvs_page *entry, *next; + + while (size > 0) { + unsigned int nr_bytes; + + entry = kzalloc(sizeof(struct nvs_page), GFP_KERNEL); + if (!entry) + goto Error; + + list_add_tail(&entry->node, &nvs_list); + entry->phys_start = start; + nr_bytes = PAGE_SIZE - (start & ~PAGE_MASK); + entry->size = (size < nr_bytes) ? size : nr_bytes; + + start += entry->size; + size -= entry->size; + } + return 0; + + Error: + list_for_each_entry_safe(entry, next, &nvs_list, node) { + list_del(&entry->node); + kfree(entry); + } + return -ENOMEM; +} + +/** + * suspend_nvs_free - free data pages allocated for saving NVS regions + */ +void suspend_nvs_free(void) +{ + struct nvs_page *entry; + + list_for_each_entry(entry, &nvs_list, node) + if (entry->data) { + free_page((unsigned long)entry->data); + entry->data = NULL; + if (entry->kaddr) { + iounmap(entry->kaddr); + entry->kaddr = NULL; + } + } +} + +/** + * suspend_nvs_alloc - allocate memory necessary for saving NVS regions + */ +int suspend_nvs_alloc(void) +{ + struct nvs_page *entry; + + list_for_each_entry(entry, &nvs_list, node) { + entry->data = (void *)__get_free_page(GFP_KERNEL); + if (!entry->data) { + suspend_nvs_free(); + return -ENOMEM; + } + } + return 0; +} + +/** + * suspend_nvs_save - save NVS memory regions + */ +int suspend_nvs_save(void) +{ + struct nvs_page *entry; + + printk(KERN_INFO "PM: Saving platform NVS memory\n"); + + list_for_each_entry(entry, &nvs_list, node) + if (entry->data) { + entry->kaddr = ioremap(entry->phys_start, entry->size); + if (!entry->kaddr) { + suspend_nvs_free(); + return -ENOMEM; + } + memcpy(entry->data, entry->kaddr, entry->size); + } + + return 0; +} + +/** + * suspend_nvs_restore - restore NVS memory regions + * + * This function is going to be called with interrupts disabled, so it + * cannot iounmap the virtual addresses used to access the NVS region. + */ +void suspend_nvs_restore(void) +{ + struct nvs_page *entry; + + printk(KERN_INFO "PM: Restoring platform NVS memory\n"); + + list_for_each_entry(entry, &nvs_list, node) + if (entry->data) + memcpy(entry->kaddr, entry->data, entry->size); +} diff --git a/include/linux/acpi.h b/include/linux/acpi.h index 67c91b4418b0..fa7ed6a983d0 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -254,6 +254,15 @@ void __init acpi_old_suspend_ordering(void); void __init acpi_nvs_nosave(void); #endif /* CONFIG_PM_SLEEP */ +#ifdef CONFIG_ACPI_SLEEP +int suspend_nvs_register(unsigned long start, unsigned long size); +#else +static inline int suspend_nvs_register(unsigned long a, unsigned long b) +{ + return 0; +} +#endif + struct acpi_osc_context { char *uuid_str; /* uuid string */ int rev; diff --git a/include/linux/suspend.h b/include/linux/suspend.h index acb7d911bb0c..0e288e3c37be 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -258,23 +258,6 @@ static inline int hibernate(void) { return -ENOSYS; } static inline bool system_entering_hibernation(void) { return false; } #endif /* CONFIG_HIBERNATION */ -#ifdef CONFIG_SUSPEND_NVS -extern int suspend_nvs_register(unsigned long start, unsigned long size); -extern int suspend_nvs_alloc(void); -extern void suspend_nvs_free(void); -extern int suspend_nvs_save(void); -extern void suspend_nvs_restore(void); -#else /* CONFIG_SUSPEND_NVS */ -static inline int suspend_nvs_register(unsigned long a, unsigned long b) -{ - return 0; -} -static inline int suspend_nvs_alloc(void) { return 0; } -static inline void suspend_nvs_free(void) {} -static inline int suspend_nvs_save(void) {} -static inline void suspend_nvs_restore(void) {} -#endif /* CONFIG_SUSPEND_NVS */ - #ifdef CONFIG_PM_SLEEP void save_processor_state(void); void restore_processor_state(void); diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index a5aff3ebad38..265729966ece 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -100,13 +100,9 @@ config PM_SLEEP_ADVANCED_DEBUG depends on PM_ADVANCED_DEBUG default n -config SUSPEND_NVS - bool - config SUSPEND bool "Suspend to RAM and standby" depends on PM && ARCH_SUSPEND_POSSIBLE - select SUSPEND_NVS if HAS_IOMEM default y ---help--- Allow the system to enter sleep states in which main memory is @@ -140,7 +136,6 @@ config HIBERNATION depends on PM && SWAP && ARCH_HIBERNATION_POSSIBLE select LZO_COMPRESS select LZO_DECOMPRESS - select SUSPEND_NVS if HAS_IOMEM ---help--- Enable the suspend to disk (STD) functionality, which is usually called "hibernation" in user interfaces. STD checkpoints the diff --git a/kernel/power/Makefile b/kernel/power/Makefile index f9063c6b185d..120a15823325 100644 --- a/kernel/power/Makefile +++ b/kernel/power/Makefile @@ -10,6 +10,5 @@ obj-$(CONFIG_SUSPEND) += suspend.o obj-$(CONFIG_PM_TEST_SUSPEND) += suspend_test.o obj-$(CONFIG_HIBERNATION) += hibernate.o snapshot.o swap.o user.o \ block_io.o -obj-$(CONFIG_SUSPEND_NVS) += nvs.o obj-$(CONFIG_MAGIC_SYSRQ) += poweroff.o diff --git a/kernel/power/nvs.c b/kernel/power/nvs.c deleted file mode 100644 index 57c6fabbb6b6..000000000000 --- a/kernel/power/nvs.c +++ /dev/null @@ -1,142 +0,0 @@ -/* - * linux/kernel/power/hibernate_nvs.c - Routines for handling NVS memory - * - * Copyright (C) 2008,2009 Rafael J. Wysocki , Novell Inc. - * - * This file is released under the GPLv2. - */ - -#include -#include -#include -#include -#include -#include - -/* - * Platforms, like ACPI, may want us to save some memory used by them during - * suspend and to restore the contents of this memory during the subsequent - * resume. The code below implements a mechanism allowing us to do that. - */ - -struct nvs_page { - unsigned long phys_start; - unsigned int size; - void *kaddr; - void *data; - struct list_head node; -}; - -static LIST_HEAD(nvs_list); - -/** - * suspend_nvs_register - register platform NVS memory region to save - * @start - physical address of the region - * @size - size of the region - * - * The NVS region need not be page-aligned (both ends) and we arrange - * things so that the data from page-aligned addresses in this region will - * be copied into separate RAM pages. - */ -int suspend_nvs_register(unsigned long start, unsigned long size) -{ - struct nvs_page *entry, *next; - - while (size > 0) { - unsigned int nr_bytes; - - entry = kzalloc(sizeof(struct nvs_page), GFP_KERNEL); - if (!entry) - goto Error; - - list_add_tail(&entry->node, &nvs_list); - entry->phys_start = start; - nr_bytes = PAGE_SIZE - (start & ~PAGE_MASK); - entry->size = (size < nr_bytes) ? size : nr_bytes; - - start += entry->size; - size -= entry->size; - } - return 0; - - Error: - list_for_each_entry_safe(entry, next, &nvs_list, node) { - list_del(&entry->node); - kfree(entry); - } - return -ENOMEM; -} - -/** - * suspend_nvs_free - free data pages allocated for saving NVS regions - */ -void suspend_nvs_free(void) -{ - struct nvs_page *entry; - - list_for_each_entry(entry, &nvs_list, node) - if (entry->data) { - free_page((unsigned long)entry->data); - entry->data = NULL; - if (entry->kaddr) { - iounmap(entry->kaddr); - entry->kaddr = NULL; - } - } -} - -/** - * suspend_nvs_alloc - allocate memory necessary for saving NVS regions - */ -int suspend_nvs_alloc(void) -{ - struct nvs_page *entry; - - list_for_each_entry(entry, &nvs_list, node) { - entry->data = (void *)__get_free_page(GFP_KERNEL); - if (!entry->data) { - suspend_nvs_free(); - return -ENOMEM; - } - } - return 0; -} - -/** - * suspend_nvs_save - save NVS memory regions - */ -int suspend_nvs_save(void) -{ - struct nvs_page *entry; - - printk(KERN_INFO "PM: Saving platform NVS memory\n"); - - list_for_each_entry(entry, &nvs_list, node) - if (entry->data) { - entry->kaddr = ioremap(entry->phys_start, entry->size); - if (!entry->kaddr) { - suspend_nvs_free(); - return -ENOMEM; - } - memcpy(entry->data, entry->kaddr, entry->size); - } - - return 0; -} - -/** - * suspend_nvs_restore - restore NVS memory regions - * - * This function is going to be called with interrupts disabled, so it - * cannot iounmap the virtual addresses used to access the NVS region. - */ -void suspend_nvs_restore(void) -{ - struct nvs_page *entry; - - printk(KERN_INFO "PM: Restoring platform NVS memory\n"); - - list_for_each_entry(entry, &nvs_list, node) - if (entry->data) - memcpy(entry->kaddr, entry->data, entry->size); -} -- cgit v1.2.3 From 45007fd590c6b099cec5d36ea7056b00f8b4916a Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 6 Jan 2011 22:36:46 +0100 Subject: nl80211: add/fix mesh docs Some mesh attribute/command docs are missing or have errors in the name so they don't match, fix all of them. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/linux/nl80211.h | 20 +++++++++++++++----- 1 file changed, 15 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h index 2b89b712565b..821ffb954f14 100644 --- a/include/linux/nl80211.h +++ b/include/linux/nl80211.h @@ -148,6 +148,10 @@ * @NL80211_CMD_SET_MPATH: Set mesh path attributes for mesh path to * destination %NL80211_ATTR_MAC on the interface identified by * %NL80211_ATTR_IFINDEX. + * @NL80211_CMD_NEW_MPATH: Create a new mesh path for the destination given by + * %NL80211_ATTR_MAC via %NL80211_ATTR_MPATH_NEXT_HOP. + * @NL80211_CMD_DEL_MPATH: Delete a mesh path to the destination given by + * %NL80211_ATTR_MAC. * @NL80211_CMD_NEW_PATH: Add a mesh path with given attributes to the * the interface identified by %NL80211_ATTR_IFINDEX. * @NL80211_CMD_DEL_PATH: Remove a mesh path identified by %NL80211_ATTR_MAC @@ -612,7 +616,7 @@ enum nl80211_commands { * consisting of a nested array. * * @NL80211_ATTR_MESH_ID: mesh id (1-32 bytes). - * @NL80211_ATTR_PLINK_ACTION: action to perform on the mesh peer link. + * @NL80211_ATTR_STA_PLINK_ACTION: action to perform on the mesh peer link. * @NL80211_ATTR_MPATH_NEXT_HOP: MAC address of the next hop for a mesh path. * @NL80211_ATTR_MPATH_INFO: information about a mesh_path, part of mesh path * info given for %NL80211_CMD_GET_MPATH, nested attribute described at @@ -879,7 +883,9 @@ enum nl80211_commands { * See &enum nl80211_key_default_types. * * @NL80211_ATTR_MESH_SETUP: Optional mesh setup parameters. These cannot be - * changed once the mesh is active. + * changed once the mesh is active. + * @NL80211_ATTR_MESH_CONFIG: Mesh configuration parameters, a nested attribute + * containing attributes from &enum nl80211_meshconf_params. * * @NL80211_ATTR_MAX: highest attribute number currently defined * @__NL80211_ATTR_AFTER_LAST: internal use @@ -1225,8 +1231,6 @@ enum nl80211_rate_info { * @NL80211_STA_INFO_INACTIVE_TIME: time since last activity (u32, msecs) * @NL80211_STA_INFO_RX_BYTES: total received bytes (u32, from this station) * @NL80211_STA_INFO_TX_BYTES: total transmitted bytes (u32, to this station) - * @__NL80211_STA_INFO_AFTER_LAST: internal - * @NL80211_STA_INFO_MAX: highest possible station info attribute * @NL80211_STA_INFO_SIGNAL: signal strength of last received PPDU (u8, dBm) * @NL80211_STA_INFO_TX_BITRATE: current unicast tx rate, nested attribute * containing info as possible, see &enum nl80211_sta_info_txrate. @@ -1236,6 +1240,11 @@ enum nl80211_rate_info { * @NL80211_STA_INFO_TX_RETRIES: total retries (u32, to this station) * @NL80211_STA_INFO_TX_FAILED: total failed packets (u32, to this station) * @NL80211_STA_INFO_SIGNAL_AVG: signal strength average (u8, dBm) + * @NL80211_STA_INFO_LLID: the station's mesh LLID + * @NL80211_STA_INFO_PLID: the station's mesh PLID + * @NL80211_STA_INFO_PLINK_STATE: peer link state for the station + * @__NL80211_STA_INFO_AFTER_LAST: internal + * @NL80211_STA_INFO_MAX: highest possible station info attribute */ enum nl80211_sta_info { __NL80211_STA_INFO_INVALID, @@ -1626,7 +1635,7 @@ enum nl80211_mntr_flags { * @NL80211_MESHCONF_HWMP_NET_DIAM_TRVS_TIME: The interval of time (in TUs) * that it takes for an HWMP information element to propagate across the mesh * - * @NL80211_MESHCONF_ROOTMODE: whether root mode is enabled or not + * @NL80211_MESHCONF_HWMP_ROOTMODE: whether root mode is enabled or not * * @NL80211_MESHCONF_ELEMENT_TTL: specifies the value of TTL field set at a * source mesh point for path selection elements. @@ -1678,6 +1687,7 @@ enum nl80211_meshconf_params { * element that vendors will use to identify the path selection methods and * metrics in use. * + * @NL80211_MESH_SETUP_ATTR_MAX: highest possible mesh setup attribute number * @__NL80211_MESH_SETUP_ATTR_AFTER_LAST: Internal use */ enum nl80211_mesh_setup_params { -- cgit v1.2.3 From e159489baa717dbae70f9903770a6a4990865887 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Sun, 9 Jan 2011 23:32:15 +0100 Subject: workqueue: relax lockdep annotation on flush_work() Currently, the lockdep annotation in flush_work() requires exclusive access on the workqueue the target work is queued on and triggers warning if a work is trying to flush another work on the same workqueue; however, this is no longer true as workqueues can now execute multiple works concurrently. This patch adds lock_map_acquire_read() and make process_one_work() hold read access to the workqueue while executing a work and start_flush_work() check for write access if concurrnecy level is one or the workqueue has a rescuer (as only one execution resource - the rescuer - is guaranteed to be available under memory pressure), and read access if higher. This better represents what's going on and removes spurious lockdep warnings which are triggered by fake dependency chain created through flush_work(). * Peter pointed out that flushing another work from a WQ_MEM_RECLAIM wq breaks forward progress guarantee under memory pressure. Condition check accordingly updated. Signed-off-by: Tejun Heo Reported-by: "Rafael J. Wysocki" Tested-by: "Rafael J. Wysocki" Cc: Peter Zijlstra Cc: stable@kernel.org --- include/linux/lockdep.h | 3 +++ kernel/workqueue.c | 14 ++++++++++++-- 2 files changed, 15 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index 71c09b26c759..9f19430c7d07 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -522,12 +522,15 @@ static inline void print_irqtrace_events(struct task_struct *curr) #ifdef CONFIG_DEBUG_LOCK_ALLOC # ifdef CONFIG_PROVE_LOCKING # define lock_map_acquire(l) lock_acquire(l, 0, 0, 0, 2, NULL, _THIS_IP_) +# define lock_map_acquire_read(l) lock_acquire(l, 0, 0, 2, 2, NULL, _THIS_IP_) # else # define lock_map_acquire(l) lock_acquire(l, 0, 0, 0, 1, NULL, _THIS_IP_) +# define lock_map_acquire_read(l) lock_acquire(l, 0, 0, 2, 1, NULL, _THIS_IP_) # endif # define lock_map_release(l) lock_release(l, 1, _THIS_IP_) #else # define lock_map_acquire(l) do { } while (0) +# define lock_map_acquire_read(l) do { } while (0) # define lock_map_release(l) do { } while (0) #endif diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 8ee6ec82f88a..930c2390b77e 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -1840,7 +1840,7 @@ __acquires(&gcwq->lock) spin_unlock_irq(&gcwq->lock); work_clear_pending(work); - lock_map_acquire(&cwq->wq->lockdep_map); + lock_map_acquire_read(&cwq->wq->lockdep_map); lock_map_acquire(&lockdep_map); trace_workqueue_execute_start(work); f(work); @@ -2384,8 +2384,18 @@ static bool start_flush_work(struct work_struct *work, struct wq_barrier *barr, insert_wq_barrier(cwq, barr, work, worker); spin_unlock_irq(&gcwq->lock); - lock_map_acquire(&cwq->wq->lockdep_map); + /* + * If @max_active is 1 or rescuer is in use, flushing another work + * item on the same workqueue may lead to deadlock. Make sure the + * flusher is not running on the same workqueue by verifying write + * access. + */ + if (cwq->wq->saved_max_active == 1 || cwq->wq->flags & WQ_RESCUER) + lock_map_acquire(&cwq->wq->lockdep_map); + else + lock_map_acquire_read(&cwq->wq->lockdep_map); lock_map_release(&cwq->wq->lockdep_map); + return true; already_gone: spin_unlock_irq(&gcwq->lock); -- cgit v1.2.3 From 925268a06dc2b1ff7bfcc37419a6827a0e739639 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Tue, 11 Jan 2011 16:44:01 +0900 Subject: memory hotplug: one more lock on memory hotplug Now, memory_hotplug_(un)lock() is used for add/remove/offline pages for avoiding races with hibernation. But this should be held in online_pages(), too. It seems asymmetric. There are cases where one has to avoid a race with memory hotplug notifier and his own local code, and hotplug v.s. hotplug. This will add a generic solution for avoiding races. In other view, having lock here has no big impacts. online pages is tend to be done by udev script at el against each memory section one by one. Then, it's better to have lock here, too. Cc: # 2.6.37 Reviewed-by: Christoph Lameter Acked-by: David Rientjes Signed-off-by: KAMEZAWA Hiroyuki Signed-off-by: Pekka Enberg --- include/linux/memory_hotplug.h | 6 ++++++ mm/memory_hotplug.c | 4 ++++ 2 files changed, 10 insertions(+) (limited to 'include/linux') diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h index 31c237a00c48..12b9eb5a36c3 100644 --- a/include/linux/memory_hotplug.h +++ b/include/linux/memory_hotplug.h @@ -161,6 +161,12 @@ extern void register_page_bootmem_info_node(struct pglist_data *pgdat); extern void put_page_bootmem(struct page *page); #endif +/* + * Lock for memory hotplug guarantees 1) all callbacks for memory hotplug + * notifier will be called under this. 2) offline/online/add/remove memory + * will not run simultaneously. + */ + void lock_memory_hotplug(void); void unlock_memory_hotplug(void); diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index 2c6523af5473..83163c096a75 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -407,6 +407,7 @@ int online_pages(unsigned long pfn, unsigned long nr_pages) int ret; struct memory_notify arg; + lock_memory_hotplug(); arg.start_pfn = pfn; arg.nr_pages = nr_pages; arg.status_change_nid = -1; @@ -419,6 +420,7 @@ int online_pages(unsigned long pfn, unsigned long nr_pages) ret = notifier_to_errno(ret); if (ret) { memory_notify(MEM_CANCEL_ONLINE, &arg); + unlock_memory_hotplug(); return ret; } /* @@ -443,6 +445,7 @@ int online_pages(unsigned long pfn, unsigned long nr_pages) printk(KERN_DEBUG "online_pages %lx at %lx failed\n", nr_pages, pfn); memory_notify(MEM_CANCEL_ONLINE, &arg); + unlock_memory_hotplug(); return ret; } @@ -467,6 +470,7 @@ int online_pages(unsigned long pfn, unsigned long nr_pages) if (onlined_pages) memory_notify(MEM_ONLINE, &arg); + unlock_memory_hotplug(); return 0; } -- cgit v1.2.3 From 1d1bc8f2074f0b728dfca2a3c16f2f5a3f298ffc Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Mon, 4 Oct 2010 23:12:59 -0400 Subject: nfsd4: support BIND_CONN_TO_SESSION Basic xdr and processing for BIND_CONN_TO_SESSION. This adds a connection to the list of connections associated with a session. Signed-off-by: J. Bruce Fields --- fs/nfsd/nfs4proc.c | 9 ++++++-- fs/nfsd/nfs4state.c | 59 ++++++++++++++++++++++++++++++++++++++++++++++------ fs/nfsd/nfs4xdr.c | 34 ++++++++++++++++++++++++++++-- fs/nfsd/state.h | 5 +++++ fs/nfsd/xdr4.h | 2 ++ include/linux/nfs4.h | 3 +++ 6 files changed, 102 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index fd6694b49e1c..db52546143d1 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -1004,8 +1004,8 @@ static const char *nfsd4_op_name(unsigned opnum); * Also note, enforced elsewhere: * - SEQUENCE other than as first op results in * NFS4ERR_SEQUENCE_POS. (Enforced in nfsd4_sequence().) - * - BIND_CONN_TO_SESSION must be the only op in its compound - * (Will be enforced in nfsd4_bind_conn_to_session().) + * - BIND_CONN_TO_SESSION must be the only op in its compound. + * (Enforced in nfsd4_bind_conn_to_session().) * - DESTROY_SESSION must be the final operation in a compound, if * sessionid's in SEQUENCE and DESTROY_SESSION are the same. * (Enforced in nfsd4_destroy_session().) @@ -1326,6 +1326,11 @@ static struct nfsd4_operation nfsd4_ops[] = { .op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP, .op_name = "OP_EXCHANGE_ID", }, + [OP_BIND_CONN_TO_SESSION] = { + .op_func = (nfsd4op_func)nfsd4_bind_conn_to_session, + .op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP, + .op_name = "OP_BIND_CONN_TO_SESSION", + }, [OP_CREATE_SESSION] = { .op_func = (nfsd4op_func)nfsd4_create_session, .op_flags = ALLOWED_WITHOUT_FH | ALLOWED_AS_FIRST_OP, diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 3cf9900d5f32..956174f488a7 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -679,15 +679,12 @@ static int nfsd4_register_conn(struct nfsd4_conn *conn) return register_xpt_user(conn->cn_xprt, &conn->cn_xpt_user); } -static __be32 nfsd4_new_conn(struct svc_rqst *rqstp, struct nfsd4_session *ses) +static __be32 nfsd4_new_conn(struct svc_rqst *rqstp, struct nfsd4_session *ses, u32 dir) { struct nfsd4_conn *conn; - u32 flags = NFS4_CDFC4_FORE; int ret; - if (ses->se_flags & SESSION4_BACK_CHAN) - flags |= NFS4_CDFC4_BACK; - conn = alloc_conn(rqstp, flags); + conn = alloc_conn(rqstp, dir); if (!conn) return nfserr_jukebox; nfsd4_hash_conn(conn, ses); @@ -698,6 +695,17 @@ static __be32 nfsd4_new_conn(struct svc_rqst *rqstp, struct nfsd4_session *ses) return nfs_ok; } +static __be32 nfsd4_new_conn_from_crses(struct svc_rqst *rqstp, struct nfsd4_session *ses) +{ + u32 dir = NFS4_CDFC4_FORE; + + if (ses->se_flags & SESSION4_BACK_CHAN) + dir |= NFS4_CDFC4_BACK; + + return nfsd4_new_conn(rqstp, ses, dir); +} + +/* must be called under client_lock */ static void nfsd4_del_conns(struct nfsd4_session *s) { struct nfs4_client *clp = s->se_client; @@ -776,7 +784,7 @@ static struct nfsd4_session *alloc_init_session(struct svc_rqst *rqstp, struct n spin_unlock(&clp->cl_lock); spin_unlock(&client_lock); - status = nfsd4_new_conn(rqstp, new); + status = nfsd4_new_conn_from_crses(rqstp, new); /* whoops: benny points out, status is ignored! (err, or bogus) */ if (status) { free_session(&new->se_ref); @@ -1597,6 +1605,45 @@ static bool nfsd4_last_compound_op(struct svc_rqst *rqstp) return argp->opcnt == resp->opcnt; } +static __be32 nfsd4_map_bcts_dir(u32 *dir) +{ + switch (*dir) { + case NFS4_CDFC4_FORE: + case NFS4_CDFC4_BACK: + return nfs_ok; + case NFS4_CDFC4_FORE_OR_BOTH: + case NFS4_CDFC4_BACK_OR_BOTH: + *dir = NFS4_CDFC4_BOTH; + return nfs_ok; + }; + return nfserr_inval; +} + +__be32 nfsd4_bind_conn_to_session(struct svc_rqst *rqstp, + struct nfsd4_compound_state *cstate, + struct nfsd4_bind_conn_to_session *bcts) +{ + __be32 status; + + if (!nfsd4_last_compound_op(rqstp)) + return nfserr_not_only_op; + spin_lock(&client_lock); + cstate->session = find_in_sessionid_hashtbl(&bcts->sessionid); + /* Sorta weird: we only need the refcnt'ing because new_conn acquires + * client_lock iself: */ + if (cstate->session) { + nfsd4_get_session(cstate->session); + atomic_inc(&cstate->session->se_client->cl_refcount); + } + spin_unlock(&client_lock); + if (!cstate->session) + return nfserr_badsession; + + status = nfsd4_map_bcts_dir(&bcts->dir); + nfsd4_new_conn(rqstp, cstate->session, bcts->dir); + return nfs_ok; +} + static bool nfsd4_compound_in_session(struct nfsd4_session *session, struct nfs4_sessionid *sid) { if (!session) diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index ca3786905dec..4ff2c9e0b276 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -421,6 +421,21 @@ nfsd4_decode_access(struct nfsd4_compoundargs *argp, struct nfsd4_access *access DECODE_TAIL; } +static __be32 nfsd4_decode_bind_conn_to_session(struct nfsd4_compoundargs *argp, struct nfsd4_bind_conn_to_session *bcts) +{ + DECODE_HEAD; + u32 dummy; + + READ_BUF(NFS4_MAX_SESSIONID_LEN + 8); + COPYMEM(bcts->sessionid.data, NFS4_MAX_SESSIONID_LEN); + READ32(bcts->dir); + /* XXX: Perhaps Tom Tucker could help us figure out how we + * should be using ctsa_use_conn_in_rdma_mode: */ + READ32(dummy); + + DECODE_TAIL; +} + static __be32 nfsd4_decode_close(struct nfsd4_compoundargs *argp, struct nfsd4_close *close) { @@ -1359,7 +1374,7 @@ static nfsd4_dec nfsd41_dec_ops[] = { /* new operations for NFSv4.1 */ [OP_BACKCHANNEL_CTL] = (nfsd4_dec)nfsd4_decode_notsupp, - [OP_BIND_CONN_TO_SESSION]= (nfsd4_dec)nfsd4_decode_notsupp, + [OP_BIND_CONN_TO_SESSION]= (nfsd4_dec)nfsd4_decode_bind_conn_to_session, [OP_EXCHANGE_ID] = (nfsd4_dec)nfsd4_decode_exchange_id, [OP_CREATE_SESSION] = (nfsd4_dec)nfsd4_decode_create_session, [OP_DESTROY_SESSION] = (nfsd4_dec)nfsd4_decode_destroy_session, @@ -2383,6 +2398,21 @@ nfsd4_encode_access(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_ return nfserr; } +static __be32 nfsd4_encode_bind_conn_to_session(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_bind_conn_to_session *bcts) +{ + __be32 *p; + + if (!nfserr) { + RESERVE_SPACE(NFS4_MAX_SESSIONID_LEN + 8); + WRITEMEM(bcts->sessionid.data, NFS4_MAX_SESSIONID_LEN); + WRITE32(bcts->dir); + /* XXX: ? */ + WRITE32(0); + ADJUST_ARGS(); + } + return nfserr; +} + static __be32 nfsd4_encode_close(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_close *close) { @@ -3174,7 +3204,7 @@ static nfsd4_enc nfsd4_enc_ops[] = { /* NFSv4.1 operations */ [OP_BACKCHANNEL_CTL] = (nfsd4_enc)nfsd4_encode_noop, - [OP_BIND_CONN_TO_SESSION] = (nfsd4_enc)nfsd4_encode_noop, + [OP_BIND_CONN_TO_SESSION] = (nfsd4_enc)nfsd4_encode_bind_conn_to_session, [OP_EXCHANGE_ID] = (nfsd4_enc)nfsd4_encode_exchange_id, [OP_CREATE_SESSION] = (nfsd4_enc)nfsd4_encode_create_session, [OP_DESTROY_SESSION] = (nfsd4_enc)nfsd4_encode_destroy_session, diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h index cf6dc83fd545..442f6d8e024c 100644 --- a/fs/nfsd/state.h +++ b/fs/nfsd/state.h @@ -148,6 +148,11 @@ struct nfsd4_create_session { u32 gid; }; +struct nfsd4_bind_conn_to_session { + struct nfs4_sessionid sessionid; + u32 dir; +}; + /* The single slot clientid cache structure */ struct nfsd4_clid_slot { u32 sl_seqid; diff --git a/fs/nfsd/xdr4.h b/fs/nfsd/xdr4.h index 799c30c3b495..3a7aa4d98c1f 100644 --- a/fs/nfsd/xdr4.h +++ b/fs/nfsd/xdr4.h @@ -427,6 +427,7 @@ struct nfsd4_op { /* NFSv4.1 */ struct nfsd4_exchange_id exchange_id; + struct nfsd4_bind_conn_to_session bind_conn_to_session; struct nfsd4_create_session create_session; struct nfsd4_destroy_session destroy_session; struct nfsd4_sequence sequence; @@ -523,6 +524,7 @@ extern __be32 nfsd4_replay_cache_entry(struct nfsd4_compoundres *resp, struct nfsd4_sequence *seq); extern __be32 nfsd4_exchange_id(struct svc_rqst *rqstp, struct nfsd4_compound_state *, struct nfsd4_exchange_id *); +extern __be32 nfsd4_bind_conn_to_session(struct svc_rqst *, struct nfsd4_compound_state *, struct nfsd4_bind_conn_to_session *); extern __be32 nfsd4_create_session(struct svc_rqst *, struct nfsd4_compound_state *, struct nfsd4_create_session *); diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h index 26afa3021ed6..a5918938e41e 100644 --- a/include/linux/nfs4.h +++ b/include/linux/nfs4.h @@ -65,6 +65,9 @@ #define NFS4_CDFC4_FORE 0x1 #define NFS4_CDFC4_BACK 0x2 +#define NFS4_CDFC4_BOTH 0x3 +#define NFS4_CDFC4_FORE_OR_BOTH 0x3 +#define NFS4_CDFC4_BACK_OR_BOTH 0x7 #define NFS4_SET_TO_SERVER_TIME 0 #define NFS4_SET_TO_CLIENT_TIME 1 -- cgit v1.2.3 From d75faea330dbd1873c9094e9926ae306590c0998 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Tue, 30 Nov 2010 19:15:01 -0500 Subject: rpc: move sk_bc_xprt to svc_xprt This seems obviously transport-level information even if it's currently used only by the server socket code. Signed-off-by: J. Bruce Fields --- include/linux/sunrpc/svc_xprt.h | 1 + include/linux/sunrpc/svcsock.h | 1 - net/sunrpc/svcsock.c | 10 ++++++---- net/sunrpc/xprtsock.c | 2 +- 4 files changed, 8 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/svc_xprt.h b/include/linux/sunrpc/svc_xprt.h index c8f81da15c7e..7ad9751a0d87 100644 --- a/include/linux/sunrpc/svc_xprt.h +++ b/include/linux/sunrpc/svc_xprt.h @@ -79,6 +79,7 @@ struct svc_xprt { struct list_head xpt_users; /* callbacks on free */ struct net *xpt_net; + struct rpc_xprt *xpt_bc_xprt; /* NFSv4.1 backchannel */ }; static inline void unregister_xpt_user(struct svc_xprt *xpt, struct svc_xpt_user *u) diff --git a/include/linux/sunrpc/svcsock.h b/include/linux/sunrpc/svcsock.h index 1b353a76c304..04dba23c59f2 100644 --- a/include/linux/sunrpc/svcsock.h +++ b/include/linux/sunrpc/svcsock.h @@ -28,7 +28,6 @@ struct svc_sock { /* private TCP part */ u32 sk_reclen; /* length of record */ u32 sk_tcplen; /* current read length */ - struct rpc_xprt *sk_bc_xprt; /* NFSv4.1 backchannel xprt */ }; /* diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 52bd1130e83a..3bb400f86eae 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -987,15 +987,17 @@ static int svc_process_calldir(struct svc_sock *svsk, struct svc_rqst *rqstp, vec[0] = rqstp->rq_arg.head[0]; } else { /* REPLY */ - if (svsk->sk_bc_xprt) - req = xprt_lookup_rqst(svsk->sk_bc_xprt, xid); + struct rpc_xprt *bc_xprt = svsk->sk_xprt.xpt_bc_xprt; + + if (bc_xprt) + req = xprt_lookup_rqst(bc_xprt, xid); if (!req) { printk(KERN_NOTICE "%s: Got unrecognized reply: " - "calldir 0x%x sk_bc_xprt %p xid %08x\n", + "calldir 0x%x xpt_bc_xprt %p xid %08x\n", __func__, ntohl(calldir), - svsk->sk_bc_xprt, xid); + bc_xprt, xid); vec[0] = rqstp->rq_arg.head[0]; goto out; } diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index dfcab5ac65af..18dc42eb5597 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -2379,9 +2379,9 @@ static struct rpc_xprt *xs_setup_bc_tcp(struct xprt_create *args) * The backchannel uses the same socket connection as the * forechannel */ + args->bc_xprt->xpt_bc_xprt = xprt; xprt->bc_xprt = args->bc_xprt; bc_sock = container_of(args->bc_xprt, struct svc_sock, sk_xprt); - bc_sock->sk_bc_xprt = xprt; transport->sock = bc_sock->sk_sock; transport->inet = bc_sock->sk_sk; -- cgit v1.2.3 From f0418aa4b1103f959d64dc18273efa04ee0140e9 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 8 Dec 2010 13:48:19 -0500 Subject: rpc: allow xprt_class->setup to return a preexisting xprt This allows us to reuse the xprt associated with a server connection if one has already been set up. Signed-off-by: J. Bruce Fields --- include/linux/sunrpc/xprt.h | 1 + net/sunrpc/xprt.c | 3 +++ net/sunrpc/xprtsock.c | 18 +++++++++--------- 3 files changed, 13 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h index 89d10d279a20..bef0f535f746 100644 --- a/include/linux/sunrpc/xprt.h +++ b/include/linux/sunrpc/xprt.h @@ -321,6 +321,7 @@ void xprt_conditional_disconnect(struct rpc_xprt *xprt, unsigned int cookie); #define XPRT_CLOSING (6) #define XPRT_CONNECTION_ABORT (7) #define XPRT_CONNECTION_CLOSE (8) +#define XPRT_INITIALIZED (9) static inline void xprt_set_connected(struct rpc_xprt *xprt) { diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index 749ad15ae305..856274d7e85c 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -1102,6 +1102,9 @@ found: -PTR_ERR(xprt)); return xprt; } + if (test_and_set_bit(XPRT_INITIALIZED, &xprt->state)) + /* ->setup returned a pre-initialized xprt: */ + return xprt; spin_lock_init(&xprt->transport_lock); spin_lock_init(&xprt->reserve_lock); diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 0ef4dd4414ec..ee091c869fcd 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -2359,6 +2359,15 @@ static struct rpc_xprt *xs_setup_bc_tcp(struct xprt_create *args) struct svc_sock *bc_sock; struct rpc_xprt *ret; + if (args->bc_xprt->xpt_bc_xprt) { + /* + * This server connection already has a backchannel + * export; we can't create a new one, as we wouldn't be + * able to match replies based on xid any more. So, + * reuse the already-existing one: + */ + return args->bc_xprt->xpt_bc_xprt; + } xprt = xs_setup_xprt(args, xprt_tcp_slot_table_entries); if (IS_ERR(xprt)) return xprt; @@ -2396,15 +2405,6 @@ static struct rpc_xprt *xs_setup_bc_tcp(struct xprt_create *args) xprt->address_strings[RPC_DISPLAY_PORT], xprt->address_strings[RPC_DISPLAY_PROTO]); - /* - * The backchannel uses the same socket connection as the - * forechannel - */ - if (args->bc_xprt->xpt_bc_xprt) { - /* XXX: actually, want to catch this case... */ - ret = ERR_PTR(-EINVAL); - goto out_err; - } /* * Once we've associated a backchannel xprt with a connection, * we want to keep it around as long as long as the connection -- cgit v1.2.3 From 4cb18728709683c91a5f6f8d5f337bfb498b089a Mon Sep 17 00:00:00 2001 From: "R.Durgadoss" Date: Wed, 27 Oct 2010 03:33:29 +0530 Subject: thermal: Add event notification to thermal framework This patch adds event notification support to the generic thermal sysfs framework in the kernel. The notification is in the form of a netlink event. Signed-off-by: R.Durgadoss Signed-off-by: Len Brown --- Documentation/ABI/stable/thermal-notification | 4 + Documentation/thermal/sysfs-api.txt | 12 +++ drivers/thermal/Kconfig | 1 + drivers/thermal/thermal_sys.c | 103 +++++++++++++++++++++++++- include/linux/thermal.h | 32 ++++++++ 5 files changed, 151 insertions(+), 1 deletion(-) create mode 100644 Documentation/ABI/stable/thermal-notification (limited to 'include/linux') diff --git a/Documentation/ABI/stable/thermal-notification b/Documentation/ABI/stable/thermal-notification new file mode 100644 index 000000000000..9723e8b7aeb3 --- /dev/null +++ b/Documentation/ABI/stable/thermal-notification @@ -0,0 +1,4 @@ +What: A notification mechanism for thermal related events +Description: + This interface enables notification for thermal related events. + The notification is in the form of a netlink event. diff --git a/Documentation/thermal/sysfs-api.txt b/Documentation/thermal/sysfs-api.txt index cb3d15bc1aeb..b61e46f449aa 100644 --- a/Documentation/thermal/sysfs-api.txt +++ b/Documentation/thermal/sysfs-api.txt @@ -278,3 +278,15 @@ method, the sys I/F structure will be built like this: |---name: acpitz |---temp1_input: 37000 |---temp1_crit: 100000 + +4. Event Notification + +The framework includes a simple notification mechanism, in the form of a +netlink event. Netlink socket initialization is done during the _init_ +of the framework. Drivers which intend to use the notification mechanism +just need to call generate_netlink_event() with two arguments viz +(originator, event). Typically the originator will be an integer assigned +to a thermal_zone_device when it registers itself with the framework. The +event will be one of:{THERMAL_AUX0, THERMAL_AUX1, THERMAL_CRITICAL, +THERMAL_DEV_FAULT}. Notification can be sent when the current temperature +crosses any of the configured thresholds. diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig index bf7c687519ef..f7a5dba3ca23 100644 --- a/drivers/thermal/Kconfig +++ b/drivers/thermal/Kconfig @@ -4,6 +4,7 @@ menuconfig THERMAL tristate "Generic Thermal sysfs driver" + depends on NET help Generic Thermal Sysfs driver offers a generic mechanism for thermal management. Usually it's made up of one or more thermal diff --git a/drivers/thermal/thermal_sys.c b/drivers/thermal/thermal_sys.c index 13c72c629329..760e045c93c8 100644 --- a/drivers/thermal/thermal_sys.c +++ b/drivers/thermal/thermal_sys.c @@ -32,6 +32,8 @@ #include #include #include +#include +#include MODULE_AUTHOR("Zhang Rui"); MODULE_DESCRIPTION("Generic thermal management sysfs support"); @@ -58,6 +60,22 @@ static LIST_HEAD(thermal_tz_list); static LIST_HEAD(thermal_cdev_list); static DEFINE_MUTEX(thermal_list_lock); +static unsigned int thermal_event_seqnum; + +static struct genl_family thermal_event_genl_family = { + .id = GENL_ID_GENERATE, + .name = THERMAL_GENL_FAMILY_NAME, + .version = THERMAL_GENL_VERSION, + .maxattr = THERMAL_GENL_ATTR_MAX, +}; + +static struct genl_multicast_group thermal_event_mcgrp = { + .name = THERMAL_GENL_MCAST_GROUP_NAME, +}; + +static int genetlink_init(void); +static void genetlink_exit(void); + static int get_idr(struct idr *idr, struct mutex *lock, int *id) { int err; @@ -1214,6 +1232,82 @@ void thermal_zone_device_unregister(struct thermal_zone_device *tz) EXPORT_SYMBOL(thermal_zone_device_unregister); +int generate_netlink_event(u32 orig, enum events event) +{ + struct sk_buff *skb; + struct nlattr *attr; + struct thermal_genl_event *thermal_event; + void *msg_header; + int size; + int result; + + /* allocate memory */ + size = nla_total_size(sizeof(struct thermal_genl_event)) + \ + nla_total_size(0); + + skb = genlmsg_new(size, GFP_ATOMIC); + if (!skb) + return -ENOMEM; + + /* add the genetlink message header */ + msg_header = genlmsg_put(skb, 0, thermal_event_seqnum++, + &thermal_event_genl_family, 0, + THERMAL_GENL_CMD_EVENT); + if (!msg_header) { + nlmsg_free(skb); + return -ENOMEM; + } + + /* fill the data */ + attr = nla_reserve(skb, THERMAL_GENL_ATTR_EVENT, \ + sizeof(struct thermal_genl_event)); + + if (!attr) { + nlmsg_free(skb); + return -EINVAL; + } + + thermal_event = nla_data(attr); + if (!thermal_event) { + nlmsg_free(skb); + return -EINVAL; + } + + memset(thermal_event, 0, sizeof(struct thermal_genl_event)); + + thermal_event->orig = orig; + thermal_event->event = event; + + /* send multicast genetlink message */ + result = genlmsg_end(skb, msg_header); + if (result < 0) { + nlmsg_free(skb); + return result; + } + + result = genlmsg_multicast(skb, 0, thermal_event_mcgrp.id, GFP_ATOMIC); + if (result) + printk(KERN_INFO "failed to send netlink event:%d", result); + + return result; +} +EXPORT_SYMBOL(generate_netlink_event); + +static int genetlink_init(void) +{ + int result; + + result = genl_register_family(&thermal_event_genl_family); + if (result) + return result; + + result = genl_register_mc_group(&thermal_event_genl_family, + &thermal_event_mcgrp); + if (result) + genl_unregister_family(&thermal_event_genl_family); + return result; +} + static int __init thermal_init(void) { int result = 0; @@ -1225,9 +1319,15 @@ static int __init thermal_init(void) mutex_destroy(&thermal_idr_lock); mutex_destroy(&thermal_list_lock); } + result = genetlink_init(); return result; } +static void genetlink_exit(void) +{ + genl_unregister_family(&thermal_event_genl_family); +} + static void __exit thermal_exit(void) { class_unregister(&thermal_class); @@ -1235,7 +1335,8 @@ static void __exit thermal_exit(void) idr_destroy(&thermal_cdev_idr); mutex_destroy(&thermal_idr_lock); mutex_destroy(&thermal_list_lock); + genetlink_exit(); } -subsys_initcall(thermal_init); +fs_initcall(thermal_init); module_exit(thermal_exit); diff --git a/include/linux/thermal.h b/include/linux/thermal.h index 1de8b9eb841b..784644961279 100644 --- a/include/linux/thermal.h +++ b/include/linux/thermal.h @@ -127,6 +127,37 @@ struct thermal_zone_device { struct thermal_hwmon_attr temp_crit; /* hwmon sys attr */ #endif }; +/* Adding event notification support elements */ +#define THERMAL_GENL_FAMILY_NAME "thermal_event" +#define THERMAL_GENL_VERSION 0x01 +#define THERMAL_GENL_MCAST_GROUP_NAME "thermal_mc_group" + +enum events { + THERMAL_AUX0, + THERMAL_AUX1, + THERMAL_CRITICAL, + THERMAL_DEV_FAULT, +}; + +struct thermal_genl_event { + u32 orig; + enum events event; +}; +/* attributes of thermal_genl_family */ +enum { + THERMAL_GENL_ATTR_UNSPEC, + THERMAL_GENL_ATTR_EVENT, + __THERMAL_GENL_ATTR_MAX, +}; +#define THERMAL_GENL_ATTR_MAX (__THERMAL_GENL_ATTR_MAX - 1) + +/* commands supported by the thermal_genl_family */ +enum { + THERMAL_GENL_CMD_UNSPEC, + THERMAL_GENL_CMD_EVENT, + __THERMAL_GENL_CMD_MAX, +}; +#define THERMAL_GENL_CMD_MAX (__THERMAL_GENL_CMD_MAX - 1) struct thermal_zone_device *thermal_zone_device_register(char *, int, void *, struct @@ -146,5 +177,6 @@ struct thermal_cooling_device *thermal_cooling_device_register(char *, void *, thermal_cooling_device_ops *); void thermal_cooling_device_unregister(struct thermal_cooling_device *); +extern int generate_netlink_event(u32 orig, enum events event); #endif /* __THERMAL_H__ */ -- cgit v1.2.3 From d247632c08c674864d438733280422ddb7130ff8 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Wed, 12 Jan 2011 02:34:59 -0500 Subject: cpuidle: delete NOP CPUIDLE_FLAG_POLL it serves no purpose Signed-off-by: Len Brown --- drivers/cpuidle/cpuidle.c | 2 +- include/linux/cpuidle.h | 1 - 2 files changed, 1 insertion(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c index 37e446041306..dd5f1eafd6f2 100644 --- a/drivers/cpuidle/cpuidle.c +++ b/drivers/cpuidle/cpuidle.c @@ -186,7 +186,7 @@ static void poll_idle_init(struct cpuidle_device *dev) state->exit_latency = 0; state->target_residency = 0; state->power_usage = -1; - state->flags = CPUIDLE_FLAG_POLL; + state->flags = 0; state->enter = poll_idle; } #else diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h index 1be416bbbb82..dee32853b0da 100644 --- a/include/linux/cpuidle.h +++ b/include/linux/cpuidle.h @@ -48,7 +48,6 @@ struct cpuidle_state { /* Idle State Flags */ #define CPUIDLE_FLAG_TIME_VALID (0x01) /* is residency time measurable? */ #define CPUIDLE_FLAG_CHECK_BM (0x02) /* BM activity will exit state */ -#define CPUIDLE_FLAG_POLL (0x10) /* no latency, no savings */ #define CPUIDLE_FLAG_SHALLOW (0x20) /* low latency, minimal savings */ #define CPUIDLE_FLAG_BALANCED (0x40) /* medium latency, moderate savings */ #define CPUIDLE_FLAG_DEEP (0x80) /* high latency, large savings */ -- cgit v1.2.3 From 642f11c53f17aee991d97d14e6922172849ef227 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Wed, 12 Jan 2011 02:45:00 -0500 Subject: cpuidle: delete unused CPUIDLE_FLAG_SHALLOW, BALANCED, DEEP definitions Signed-off-by: Len Brown --- include/linux/cpuidle.h | 3 --- 1 file changed, 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h index dee32853b0da..c25295337382 100644 --- a/include/linux/cpuidle.h +++ b/include/linux/cpuidle.h @@ -48,9 +48,6 @@ struct cpuidle_state { /* Idle State Flags */ #define CPUIDLE_FLAG_TIME_VALID (0x01) /* is residency time measurable? */ #define CPUIDLE_FLAG_CHECK_BM (0x02) /* BM activity will exit state */ -#define CPUIDLE_FLAG_SHALLOW (0x20) /* low latency, minimal savings */ -#define CPUIDLE_FLAG_BALANCED (0x40) /* medium latency, moderate savings */ -#define CPUIDLE_FLAG_DEEP (0x80) /* high latency, large savings */ #define CPUIDLE_FLAG_IGNORE (0x100) /* ignore during this idle period */ #define CPUIDLE_FLAG_TLB_FLUSHED (0x200) /* tlb will be flushed */ -- cgit v1.2.3 From 956d033fb2eb3f8818260cdf01644bf4dc1a9e33 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Wed, 12 Jan 2011 02:51:20 -0500 Subject: cpuidle: CPUIDLE_FLAG_TLB_FLUSHED is specific to intel_idle Signed-off-by: Len Brown --- drivers/idle/intel_idle.c | 8 ++++++++ include/linux/cpuidle.h | 1 - 2 files changed, 8 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c index 21d387132dbc..8256309deaad 100644 --- a/drivers/idle/intel_idle.c +++ b/drivers/idle/intel_idle.c @@ -81,6 +81,14 @@ static int intel_idle(struct cpuidle_device *dev, struct cpuidle_state *state); static struct cpuidle_state *cpuidle_state_table; +/* + * Set this flag for states where the HW flushes the TLB for us + * and so we don't need cross-calls to keep it consistent. + * If this flag is set, SW flushes the TLB, so even if the + * HW doesn't do the flushing, this flag is safe to use. + */ +#define CPUIDLE_FLAG_TLB_FLUSHED 0x10000 + /* * States are indexed by the cstate number, * which is also the index into the MWAIT hint array. diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h index c25295337382..6be722c725d5 100644 --- a/include/linux/cpuidle.h +++ b/include/linux/cpuidle.h @@ -49,7 +49,6 @@ struct cpuidle_state { #define CPUIDLE_FLAG_TIME_VALID (0x01) /* is residency time measurable? */ #define CPUIDLE_FLAG_CHECK_BM (0x02) /* BM activity will exit state */ #define CPUIDLE_FLAG_IGNORE (0x100) /* ignore during this idle period */ -#define CPUIDLE_FLAG_TLB_FLUSHED (0x200) /* tlb will be flushed */ #define CPUIDLE_DRIVER_FLAGS_MASK (0xFFFF0000) -- cgit v1.2.3 From 5392083748a340f68052c0b83a7ce28b10324972 Mon Sep 17 00:00:00 2001 From: Len Brown Date: Wed, 12 Jan 2011 02:56:15 -0500 Subject: cpuidle: CPUIDLE_FLAG_CHECK_BM is omap3_idle specific Signed-off-by: Len Brown --- arch/arm/mach-omap2/cpuidle34xx.c | 2 ++ include/linux/cpuidle.h | 1 - 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/arch/arm/mach-omap2/cpuidle34xx.c b/arch/arm/mach-omap2/cpuidle34xx.c index 0d50b45d041c..f9be4e3078bc 100644 --- a/arch/arm/mach-omap2/cpuidle34xx.c +++ b/arch/arm/mach-omap2/cpuidle34xx.c @@ -47,6 +47,8 @@ #define OMAP3_STATE_MAX OMAP3_STATE_C7 +#define CPUIDLE_FLAG_CHECK_BM 0x10000 /* use omap3_enter_idle_bm() */ + struct omap3_processor_cx { u8 valid; u8 type; diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h index 6be722c725d5..36719ead50e8 100644 --- a/include/linux/cpuidle.h +++ b/include/linux/cpuidle.h @@ -47,7 +47,6 @@ struct cpuidle_state { /* Idle State Flags */ #define CPUIDLE_FLAG_TIME_VALID (0x01) /* is residency time measurable? */ -#define CPUIDLE_FLAG_CHECK_BM (0x02) /* BM activity will exit state */ #define CPUIDLE_FLAG_IGNORE (0x100) /* ignore during this idle period */ #define CPUIDLE_DRIVER_FLAGS_MASK (0xFFFF0000) -- cgit v1.2.3 From f00c9e44ad1a9660fe8cd3ca15b6cd9497172eab Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Wed, 15 Sep 2010 17:38:58 +0200 Subject: quota: Fix deadlock during path resolution As Al Viro pointed out path resolution during Q_QUOTAON calls to quotactl is prone to deadlocks. We hold s_umount semaphore for reading during the path resolution and resolution itself may need to acquire the semaphore for writing when e. g. autofs mountpoint is passed. Solve the problem by performing the resolution before we get hold of the superblock (and thus s_umount semaphore). The whole thing is complicated by the fact that some filesystems (OCFS2) ignore the path argument. So to distinguish between filesystem which want the path and which do not we introduce new .quota_on_meta callback which does not get the path. OCFS2 then uses this callback instead of old .quota_on. CC: Al Viro CC: Christoph Hellwig CC: Ted Ts'o CC: Joel Becker Signed-off-by: Jan Kara --- fs/ext3/super.c | 25 +++++++------------------ fs/ext4/super.c | 25 +++++++------------------ fs/ocfs2/super.c | 5 ++--- fs/quota/dquot.c | 18 ++---------------- fs/quota/quota.c | 41 +++++++++++++++++++++++++++-------------- fs/reiserfs/super.c | 17 ++++++----------- include/linux/quota.h | 5 ++++- include/linux/quotaops.h | 4 +--- 8 files changed, 56 insertions(+), 84 deletions(-) (limited to 'include/linux') diff --git a/fs/ext3/super.c b/fs/ext3/super.c index b7d0554631e4..0e0d391626be 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -755,7 +755,7 @@ static int ext3_release_dquot(struct dquot *dquot); static int ext3_mark_dquot_dirty(struct dquot *dquot); static int ext3_write_info(struct super_block *sb, int type); static int ext3_quota_on(struct super_block *sb, int type, int format_id, - char *path); + struct path *path); static int ext3_quota_on_mount(struct super_block *sb, int type); static ssize_t ext3_quota_read(struct super_block *sb, int type, char *data, size_t len, loff_t off); @@ -2885,27 +2885,20 @@ static int ext3_quota_on_mount(struct super_block *sb, int type) * Standard function to be called on quota_on */ static int ext3_quota_on(struct super_block *sb, int type, int format_id, - char *name) + struct path *path) { int err; - struct path path; if (!test_opt(sb, QUOTA)) return -EINVAL; - err = kern_path(name, LOOKUP_FOLLOW, &path); - if (err) - return err; - /* Quotafile not on the same filesystem? */ - if (path.mnt->mnt_sb != sb) { - path_put(&path); + if (path->mnt->mnt_sb != sb) return -EXDEV; - } /* Journaling quota? */ if (EXT3_SB(sb)->s_qf_names[type]) { /* Quotafile not of fs root? */ - if (path.dentry->d_parent != sb->s_root) + if (path->dentry->d_parent != sb->s_root) ext3_msg(sb, KERN_WARNING, "warning: Quota file not on filesystem root. " "Journaled quota will not work."); @@ -2915,7 +2908,7 @@ static int ext3_quota_on(struct super_block *sb, int type, int format_id, * When we journal data on quota file, we have to flush journal to see * all updates to the file when we bypass pagecache... */ - if (ext3_should_journal_data(path.dentry->d_inode)) { + if (ext3_should_journal_data(path->dentry->d_inode)) { /* * We don't need to lock updates but journal_flush() could * otherwise be livelocked... @@ -2923,15 +2916,11 @@ static int ext3_quota_on(struct super_block *sb, int type, int format_id, journal_lock_updates(EXT3_SB(sb)->s_journal); err = journal_flush(EXT3_SB(sb)->s_journal); journal_unlock_updates(EXT3_SB(sb)->s_journal); - if (err) { - path_put(&path); + if (err) return err; - } } - err = dquot_quota_on_path(sb, type, format_id, &path); - path_put(&path); - return err; + return dquot_quota_on(sb, type, format_id, path); } /* Read data from quotafile - avoid pagecache and such because we cannot afford diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 29c80f6d8b27..0f10ccd6bfc0 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1162,7 +1162,7 @@ static int ext4_release_dquot(struct dquot *dquot); static int ext4_mark_dquot_dirty(struct dquot *dquot); static int ext4_write_info(struct super_block *sb, int type); static int ext4_quota_on(struct super_block *sb, int type, int format_id, - char *path); + struct path *path); static int ext4_quota_off(struct super_block *sb, int type); static int ext4_quota_on_mount(struct super_block *sb, int type); static ssize_t ext4_quota_read(struct super_block *sb, int type, char *data, @@ -4566,27 +4566,20 @@ static int ext4_quota_on_mount(struct super_block *sb, int type) * Standard function to be called on quota_on */ static int ext4_quota_on(struct super_block *sb, int type, int format_id, - char *name) + struct path *path) { int err; - struct path path; if (!test_opt(sb, QUOTA)) return -EINVAL; - err = kern_path(name, LOOKUP_FOLLOW, &path); - if (err) - return err; - /* Quotafile not on the same filesystem? */ - if (path.mnt->mnt_sb != sb) { - path_put(&path); + if (path->mnt->mnt_sb != sb) return -EXDEV; - } /* Journaling quota? */ if (EXT4_SB(sb)->s_qf_names[type]) { /* Quotafile not in fs root? */ - if (path.dentry->d_parent != sb->s_root) + if (path->dentry->d_parent != sb->s_root) ext4_msg(sb, KERN_WARNING, "Quota file not on filesystem root. " "Journaled quota will not work"); @@ -4597,7 +4590,7 @@ static int ext4_quota_on(struct super_block *sb, int type, int format_id, * all updates to the file when we bypass pagecache... */ if (EXT4_SB(sb)->s_journal && - ext4_should_journal_data(path.dentry->d_inode)) { + ext4_should_journal_data(path->dentry->d_inode)) { /* * We don't need to lock updates but journal_flush() could * otherwise be livelocked... @@ -4605,15 +4598,11 @@ static int ext4_quota_on(struct super_block *sb, int type, int format_id, jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal); err = jbd2_journal_flush(EXT4_SB(sb)->s_journal); jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal); - if (err) { - path_put(&path); + if (err) return err; - } } - err = dquot_quota_on_path(sb, type, format_id, &path); - path_put(&path); - return err; + return dquot_quota_on(sb, type, format_id, path); } static int ext4_quota_off(struct super_block *sb, int type) diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 17ff46fa8a10..31c3ffd2f8d0 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -993,8 +993,7 @@ static void ocfs2_disable_quotas(struct ocfs2_super *osb) } /* Handle quota on quotactl */ -static int ocfs2_quota_on(struct super_block *sb, int type, int format_id, - char *path) +static int ocfs2_quota_on(struct super_block *sb, int type, int format_id) { unsigned int feature[MAXQUOTAS] = { OCFS2_FEATURE_RO_COMPAT_USRQUOTA, OCFS2_FEATURE_RO_COMPAT_GRPQUOTA}; @@ -1013,7 +1012,7 @@ static int ocfs2_quota_off(struct super_block *sb, int type) } static const struct quotactl_ops ocfs2_quotactl_ops = { - .quota_on = ocfs2_quota_on, + .quota_on_meta = ocfs2_quota_on, .quota_off = ocfs2_quota_off, .quota_sync = dquot_quota_sync, .get_info = dquot_get_dqinfo, diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c index 84becd3e4772..a2a622e079f0 100644 --- a/fs/quota/dquot.c +++ b/fs/quota/dquot.c @@ -2189,8 +2189,8 @@ int dquot_resume(struct super_block *sb, int type) } EXPORT_SYMBOL(dquot_resume); -int dquot_quota_on_path(struct super_block *sb, int type, int format_id, - struct path *path) +int dquot_quota_on(struct super_block *sb, int type, int format_id, + struct path *path) { int error = security_quota_on(path->dentry); if (error) @@ -2204,20 +2204,6 @@ int dquot_quota_on_path(struct super_block *sb, int type, int format_id, DQUOT_LIMITS_ENABLED); return error; } -EXPORT_SYMBOL(dquot_quota_on_path); - -int dquot_quota_on(struct super_block *sb, int type, int format_id, char *name) -{ - struct path path; - int error; - - error = kern_path(name, LOOKUP_FOLLOW, &path); - if (!error) { - error = dquot_quota_on_path(sb, type, format_id, &path); - path_put(&path); - } - return error; -} EXPORT_SYMBOL(dquot_quota_on); /* diff --git a/fs/quota/quota.c b/fs/quota/quota.c index b299961e1edb..b34bdb25490c 100644 --- a/fs/quota/quota.c +++ b/fs/quota/quota.c @@ -64,18 +64,15 @@ static int quota_sync_all(int type) } static int quota_quotaon(struct super_block *sb, int type, int cmd, qid_t id, - void __user *addr) + struct path *path) { - char *pathname; - int ret = -ENOSYS; - - pathname = getname(addr); - if (IS_ERR(pathname)) - return PTR_ERR(pathname); - if (sb->s_qcop->quota_on) - ret = sb->s_qcop->quota_on(sb, type, id, pathname); - putname(pathname); - return ret; + if (!sb->s_qcop->quota_on && !sb->s_qcop->quota_on_meta) + return -ENOSYS; + if (sb->s_qcop->quota_on_meta) + return sb->s_qcop->quota_on_meta(sb, type, id); + if (IS_ERR(path)) + return PTR_ERR(path); + return sb->s_qcop->quota_on(sb, type, id, path); } static int quota_getfmt(struct super_block *sb, int type, void __user *addr) @@ -241,7 +238,7 @@ static int quota_getxquota(struct super_block *sb, int type, qid_t id, /* Copy parameters and call proper function */ static int do_quotactl(struct super_block *sb, int type, int cmd, qid_t id, - void __user *addr) + void __user *addr, struct path *path) { int ret; @@ -256,7 +253,7 @@ static int do_quotactl(struct super_block *sb, int type, int cmd, qid_t id, switch (cmd) { case Q_QUOTAON: - return quota_quotaon(sb, type, cmd, id, addr); + return quota_quotaon(sb, type, cmd, id, path); case Q_QUOTAOFF: if (!sb->s_qcop->quota_off) return -ENOSYS; @@ -335,6 +332,7 @@ SYSCALL_DEFINE4(quotactl, unsigned int, cmd, const char __user *, special, { uint cmds, type; struct super_block *sb = NULL; + struct path path, *pathp = NULL; int ret; cmds = cmd >> SUBCMDSHIFT; @@ -351,12 +349,27 @@ SYSCALL_DEFINE4(quotactl, unsigned int, cmd, const char __user *, special, return -ENODEV; } + /* + * Path for quotaon has to be resolved before grabbing superblock + * because that gets s_umount sem which is also possibly needed by path + * resolution (think about autofs) and thus deadlocks could arise. + */ + if (cmds == Q_QUOTAON) { + ret = user_path_at(AT_FDCWD, addr, LOOKUP_FOLLOW, &path); + if (ret) + pathp = ERR_PTR(ret); + else + pathp = &path; + } + sb = quotactl_block(special); if (IS_ERR(sb)) return PTR_ERR(sb); - ret = do_quotactl(sb, type, cmds, id, addr); + ret = do_quotactl(sb, type, cmds, id, addr, pathp); drop_super(sb); + if (pathp && !IS_ERR(pathp)) + path_put(pathp); return ret; } diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 2575682a9ead..0aab04f46827 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -632,7 +632,7 @@ static int reiserfs_acquire_dquot(struct dquot *); static int reiserfs_release_dquot(struct dquot *); static int reiserfs_mark_dquot_dirty(struct dquot *); static int reiserfs_write_info(struct super_block *, int); -static int reiserfs_quota_on(struct super_block *, int, int, char *); +static int reiserfs_quota_on(struct super_block *, int, int, struct path *); static const struct dquot_operations reiserfs_quota_operations = { .write_dquot = reiserfs_write_dquot, @@ -2048,25 +2048,21 @@ static int reiserfs_quota_on_mount(struct super_block *sb, int type) * Standard function to be called on quota_on */ static int reiserfs_quota_on(struct super_block *sb, int type, int format_id, - char *name) + struct path *path) { int err; - struct path path; struct inode *inode; struct reiserfs_transaction_handle th; if (!(REISERFS_SB(sb)->s_mount_opt & (1 << REISERFS_QUOTA))) return -EINVAL; - err = kern_path(name, LOOKUP_FOLLOW, &path); - if (err) - return err; /* Quotafile not on the same filesystem? */ - if (path.mnt->mnt_sb != sb) { + if (path->mnt->mnt_sb != sb) { err = -EXDEV; goto out; } - inode = path.dentry->d_inode; + inode = path->dentry->d_inode; /* We must not pack tails for quota files on reiserfs for quota IO to work */ if (!(REISERFS_I(inode)->i_flags & i_nopack_mask)) { err = reiserfs_unpack(inode, NULL); @@ -2082,7 +2078,7 @@ static int reiserfs_quota_on(struct super_block *sb, int type, int format_id, /* Journaling quota? */ if (REISERFS_SB(sb)->s_qf_names[type]) { /* Quotafile not of fs root? */ - if (path.dentry->d_parent != sb->s_root) + if (path->dentry->d_parent != sb->s_root) reiserfs_warning(sb, "super-6521", "Quota file not on filesystem root. " "Journalled quota will not work."); @@ -2101,9 +2097,8 @@ static int reiserfs_quota_on(struct super_block *sb, int type, int format_id, if (err) goto out; } - err = dquot_quota_on_path(sb, type, format_id, &path); + err = dquot_quota_on(sb, type, format_id, path); out: - path_put(&path); return err; } diff --git a/include/linux/quota.h b/include/linux/quota.h index 94c1f03b50eb..9a85412e0db6 100644 --- a/include/linux/quota.h +++ b/include/linux/quota.h @@ -322,9 +322,12 @@ struct dquot_operations { qsize_t *(*get_reserved_space) (struct inode *); }; +struct path; + /* Operations handling requests from userspace */ struct quotactl_ops { - int (*quota_on)(struct super_block *, int, int, char *); + int (*quota_on)(struct super_block *, int, int, struct path *); + int (*quota_on_meta)(struct super_block *, int, int); int (*quota_off)(struct super_block *, int); int (*quota_sync)(struct super_block *, int, int); int (*get_info)(struct super_block *, int, struct if_dqinfo *); diff --git a/include/linux/quotaops.h b/include/linux/quotaops.h index 223b14cd129c..eb354f6f26b3 100644 --- a/include/linux/quotaops.h +++ b/include/linux/quotaops.h @@ -76,11 +76,9 @@ int dquot_mark_dquot_dirty(struct dquot *dquot); int dquot_file_open(struct inode *inode, struct file *file); -int dquot_quota_on(struct super_block *sb, int type, int format_id, - char *path); int dquot_enable(struct inode *inode, int type, int format_id, unsigned int flags); -int dquot_quota_on_path(struct super_block *sb, int type, int format_id, +int dquot_quota_on(struct super_block *sb, int type, int format_id, struct path *path); int dquot_quota_on_mount(struct super_block *sb, char *qf_name, int format_id, int type); -- cgit v1.2.3 From 2fc72c7b84002ffb3c66918e2a7b0ee607d8b5aa Mon Sep 17 00:00:00 2001 From: KOVACS Krisztian Date: Wed, 12 Jan 2011 20:25:08 +0100 Subject: netfilter: fix compilation when conntrack is disabled but tproxy is enabled The IPv6 tproxy patches split IPv6 defragmentation off of conntrack, but failed to update the #ifdef stanzas guarding the defragmentation related fields and code in skbuff and conntrack related code in nf_defrag_ipv6.c. This patch adds the required #ifdefs so that IPv6 tproxy can truly be used without connection tracking. Original report: http://marc.info/?l=linux-netdev&m=129010118516341&w=2 Reported-by: Randy Dunlap Acked-by: Randy Dunlap Signed-off-by: KOVACS Krisztian Signed-off-by: Pablo Neira Ayuso --- include/linux/skbuff.h | 15 +++++++++++++++ include/net/netfilter/ipv6/nf_conntrack_ipv6.h | 10 ---------- include/net/netfilter/ipv6/nf_defrag_ipv6.h | 10 ++++++++++ net/core/skbuff.c | 2 ++ net/ipv6/netfilter/nf_defrag_ipv6_hooks.c | 8 +++++++- 5 files changed, 34 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 20ec0a64cb9f..bf221d65d9ad 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -255,6 +255,11 @@ typedef unsigned int sk_buff_data_t; typedef unsigned char *sk_buff_data_t; #endif +#if defined(CONFIG_NF_DEFRAG_IPV4) || defined(CONFIG_NF_DEFRAG_IPV4_MODULE) || \ + defined(CONFIG_NF_DEFRAG_IPV6) || defined(CONFIG_NF_DEFRAG_IPV6_MODULE) +#define NET_SKBUFF_NF_DEFRAG_NEEDED 1 +#endif + /** * struct sk_buff - socket buffer * @next: Next buffer in list @@ -362,6 +367,8 @@ struct sk_buff { void (*destructor)(struct sk_buff *skb); #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) struct nf_conntrack *nfct; +#endif +#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED struct sk_buff *nfct_reasm; #endif #ifdef CONFIG_BRIDGE_NETFILTER @@ -2057,6 +2064,8 @@ static inline void nf_conntrack_get(struct nf_conntrack *nfct) if (nfct) atomic_inc(&nfct->use); } +#endif +#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED static inline void nf_conntrack_get_reasm(struct sk_buff *skb) { if (skb) @@ -2085,6 +2094,8 @@ static inline void nf_reset(struct sk_buff *skb) #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) nf_conntrack_put(skb->nfct); skb->nfct = NULL; +#endif +#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED nf_conntrack_put_reasm(skb->nfct_reasm); skb->nfct_reasm = NULL; #endif @@ -2101,6 +2112,8 @@ static inline void __nf_copy(struct sk_buff *dst, const struct sk_buff *src) dst->nfct = src->nfct; nf_conntrack_get(src->nfct); dst->nfctinfo = src->nfctinfo; +#endif +#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED dst->nfct_reasm = src->nfct_reasm; nf_conntrack_get_reasm(src->nfct_reasm); #endif @@ -2114,6 +2127,8 @@ static inline void nf_copy(struct sk_buff *dst, const struct sk_buff *src) { #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) nf_conntrack_put(dst->nfct); +#endif +#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED nf_conntrack_put_reasm(dst->nfct_reasm); #endif #ifdef CONFIG_BRIDGE_NETFILTER diff --git a/include/net/netfilter/ipv6/nf_conntrack_ipv6.h b/include/net/netfilter/ipv6/nf_conntrack_ipv6.h index 1ee717eb5b09..a4c993685795 100644 --- a/include/net/netfilter/ipv6/nf_conntrack_ipv6.h +++ b/include/net/netfilter/ipv6/nf_conntrack_ipv6.h @@ -7,16 +7,6 @@ extern struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp6; extern struct nf_conntrack_l4proto nf_conntrack_l4proto_udp6; extern struct nf_conntrack_l4proto nf_conntrack_l4proto_icmpv6; -extern int nf_ct_frag6_init(void); -extern void nf_ct_frag6_cleanup(void); -extern struct sk_buff *nf_ct_frag6_gather(struct sk_buff *skb, u32 user); -extern void nf_ct_frag6_output(unsigned int hooknum, struct sk_buff *skb, - struct net_device *in, - struct net_device *out, - int (*okfn)(struct sk_buff *)); - -struct inet_frags_ctl; - #include extern struct ctl_table nf_ct_ipv6_sysctl_table[]; diff --git a/include/net/netfilter/ipv6/nf_defrag_ipv6.h b/include/net/netfilter/ipv6/nf_defrag_ipv6.h index 94dd54d76b48..fd79c9a1779d 100644 --- a/include/net/netfilter/ipv6/nf_defrag_ipv6.h +++ b/include/net/netfilter/ipv6/nf_defrag_ipv6.h @@ -3,4 +3,14 @@ extern void nf_defrag_ipv6_enable(void); +extern int nf_ct_frag6_init(void); +extern void nf_ct_frag6_cleanup(void); +extern struct sk_buff *nf_ct_frag6_gather(struct sk_buff *skb, u32 user); +extern void nf_ct_frag6_output(unsigned int hooknum, struct sk_buff *skb, + struct net_device *in, + struct net_device *out, + int (*okfn)(struct sk_buff *)); + +struct inet_frags_ctl; + #endif /* _NF_DEFRAG_IPV6_H */ diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 19d6c21220fd..d31bb36ae0dc 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -380,6 +380,8 @@ static void skb_release_head_state(struct sk_buff *skb) } #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) nf_conntrack_put(skb->nfct); +#endif +#ifdef NET_SKBUFF_NF_DEFRAG_NEEDED nf_conntrack_put_reasm(skb->nfct_reasm); #endif #ifdef CONFIG_BRIDGE_NETFILTER diff --git a/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c b/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c index 99abfb53bab9..97c5b21b9674 100644 --- a/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c +++ b/net/ipv6/netfilter/nf_defrag_ipv6_hooks.c @@ -19,13 +19,15 @@ #include #include +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) #include #include #include #include #include -#include #include +#endif +#include #include static enum ip6_defrag_users nf_ct6_defrag_user(unsigned int hooknum, @@ -33,8 +35,10 @@ static enum ip6_defrag_users nf_ct6_defrag_user(unsigned int hooknum, { u16 zone = NF_CT_DEFAULT_ZONE; +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) if (skb->nfct) zone = nf_ct_zone((struct nf_conn *)skb->nfct); +#endif #ifdef CONFIG_BRIDGE_NETFILTER if (skb->nf_bridge && @@ -56,9 +60,11 @@ static unsigned int ipv6_defrag(unsigned int hooknum, { struct sk_buff *reasm; +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) /* Previously seen (loopback)? */ if (skb->nfct && !nf_ct_is_template((struct nf_conn *)skb->nfct)) return NF_ACCEPT; +#endif reasm = nf_ct_frag6_gather(skb, nf_ct6_defrag_user(hooknum, skb)); /* queued */ -- cgit v1.2.3 From 6fed05c9c9812b5882bc708f4da4fa8d5df2875c Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 12 Jan 2011 22:03:20 +0100 Subject: ACPI / PM: Fix build problems for !CONFIG_ACPI related to NVS rework The recent rework of the NVS saving/restoring code introduced two build issues for !CONFIG_ACPI, a warning in drivers/acpi/internal.h and an error in arch/x86/kernel/e820.c. Fix them by providing suitable static inline definitions of the relevant functions. Signed-off-by: Rafael J. Wysocki Acked-by: Randy Dunlap Signed-off-by: Len Brown --- drivers/acpi/internal.h | 2 +- include/linux/acpi.h | 19 ++++++++++--------- 2 files changed, 11 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/drivers/acpi/internal.h b/drivers/acpi/internal.h index 7c23b76e8eca..6d69bbedf97a 100644 --- a/drivers/acpi/internal.h +++ b/drivers/acpi/internal.h @@ -90,7 +90,7 @@ void suspend_nvs_restore(void); static inline int acpi_sleep_proc_init(void) { return 0; } static inline int suspend_nvs_alloc(void) { return 0; } static inline void suspend_nvs_free(void) {} -static inline int suspend_nvs_save(void) {} +static inline int suspend_nvs_save(void) { return 0; } static inline void suspend_nvs_restore(void) {} #endif diff --git a/include/linux/acpi.h b/include/linux/acpi.h index fa7ed6a983d0..eb176bb1b15b 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -254,15 +254,6 @@ void __init acpi_old_suspend_ordering(void); void __init acpi_nvs_nosave(void); #endif /* CONFIG_PM_SLEEP */ -#ifdef CONFIG_ACPI_SLEEP -int suspend_nvs_register(unsigned long start, unsigned long size); -#else -static inline int suspend_nvs_register(unsigned long a, unsigned long b) -{ - return 0; -} -#endif - struct acpi_osc_context { char *uuid_str; /* uuid string */ int rev; @@ -361,4 +352,14 @@ static inline int acpi_table_parse(char *id, return -1; } #endif /* !CONFIG_ACPI */ + +#ifdef CONFIG_ACPI_SLEEP +int suspend_nvs_register(unsigned long start, unsigned long size); +#else +static inline int suspend_nvs_register(unsigned long a, unsigned long b) +{ + return 0; +} +#endif + #endif /*_LINUX_ACPI_H*/ -- cgit v1.2.3 From da995a8aee044bc5d0847e19e351cd48a2cb8bcc Mon Sep 17 00:00:00 2001 From: Aleksey Senin Date: Thu, 2 Dec 2010 11:44:49 +0000 Subject: IB/mlx4: Handle protocol field in multicast table The newest device firmware stores IB vs. Ethernet protocol in two bits in members_count field of multicast group table (0: Infiniband, 1: Ethernet). When changing the QP members count for a multicast group, it important not to reset this information. When calling multicast attach first time, the protocol type should be specified. In this patch we always set it IB, but in the future we will handle Ethernet too. When looking for a QP, the protocol type shoud be checked too. Signed-off-by: Aleksey Senin Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mlx4/main.c | 9 +++++---- drivers/net/mlx4/mcg.c | 23 +++++++++++++---------- include/linux/mlx4/device.h | 10 ++++++++-- include/linux/mlx4/driver.h | 6 +----- 4 files changed, 27 insertions(+), 21 deletions(-) (limited to 'include/linux') diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c index d68d849ab866..c7a6213c6996 100644 --- a/drivers/infiniband/hw/mlx4/main.c +++ b/drivers/infiniband/hw/mlx4/main.c @@ -623,8 +623,9 @@ static int mlx4_ib_mcg_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid) struct mlx4_ib_dev *mdev = to_mdev(ibqp->device); struct mlx4_ib_qp *mqp = to_mqp(ibqp); - err = mlx4_multicast_attach(mdev->dev, &mqp->mqp, gid->raw, !!(mqp->flags & - MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK)); + err = mlx4_multicast_attach(mdev->dev, &mqp->mqp, gid->raw, + !!(mqp->flags & MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK), + MLX4_PROTOCOL_IB); if (err) return err; @@ -635,7 +636,7 @@ static int mlx4_ib_mcg_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid) return 0; err_add: - mlx4_multicast_detach(mdev->dev, &mqp->mqp, gid->raw); + mlx4_multicast_detach(mdev->dev, &mqp->mqp, gid->raw, MLX4_PROTOCOL_IB); return err; } @@ -665,7 +666,7 @@ static int mlx4_ib_mcg_detach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid) struct mlx4_ib_gid_entry *ge; err = mlx4_multicast_detach(mdev->dev, - &mqp->mqp, gid->raw); + &mqp->mqp, gid->raw, MLX4_PROTOCOL_IB); if (err) return err; diff --git a/drivers/net/mlx4/mcg.c b/drivers/net/mlx4/mcg.c index c4f88b7ef7b6..79cf42db2ea9 100644 --- a/drivers/net/mlx4/mcg.c +++ b/drivers/net/mlx4/mcg.c @@ -95,7 +95,8 @@ static int mlx4_MGID_HASH(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *mailbox * entry in hash chain and *mgm holds end of hash chain. */ static int find_mgm(struct mlx4_dev *dev, - u8 *gid, struct mlx4_cmd_mailbox *mgm_mailbox, + u8 *gid, enum mlx4_protocol protocol, + struct mlx4_cmd_mailbox *mgm_mailbox, u16 *hash, int *prev, int *index) { struct mlx4_cmd_mailbox *mailbox; @@ -134,7 +135,8 @@ static int find_mgm(struct mlx4_dev *dev, return err; } - if (!memcmp(mgm->gid, gid, 16)) + if (!memcmp(mgm->gid, gid, 16) && + be32_to_cpu(mgm->members_count) >> 30 == protocol) return err; *prev = *index; @@ -146,7 +148,7 @@ static int find_mgm(struct mlx4_dev *dev, } int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16], - int block_mcast_loopback) + int block_mcast_loopback, enum mlx4_protocol protocol) { struct mlx4_priv *priv = mlx4_priv(dev); struct mlx4_cmd_mailbox *mailbox; @@ -165,7 +167,7 @@ int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16], mutex_lock(&priv->mcg_table.mutex); - err = find_mgm(dev, gid, mailbox, &hash, &prev, &index); + err = find_mgm(dev, gid, protocol, mailbox, &hash, &prev, &index); if (err) goto out; @@ -187,7 +189,7 @@ int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16], memcpy(mgm->gid, gid, 16); } - members_count = be32_to_cpu(mgm->members_count); + members_count = be32_to_cpu(mgm->members_count) & 0xffffff; if (members_count == MLX4_QP_PER_MGM) { mlx4_err(dev, "MGM at index %x is full.\n", index); err = -ENOMEM; @@ -207,7 +209,7 @@ int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16], else mgm->qp[members_count++] = cpu_to_be32(qp->qpn & MGM_QPN_MASK); - mgm->members_count = cpu_to_be32(members_count); + mgm->members_count = cpu_to_be32(members_count | (u32) protocol << 30); err = mlx4_WRITE_MCG(dev, index, mailbox); if (err) @@ -242,7 +244,8 @@ out: } EXPORT_SYMBOL_GPL(mlx4_multicast_attach); -int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16]) +int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16], + enum mlx4_protocol protocol) { struct mlx4_priv *priv = mlx4_priv(dev); struct mlx4_cmd_mailbox *mailbox; @@ -260,7 +263,7 @@ int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16]) mutex_lock(&priv->mcg_table.mutex); - err = find_mgm(dev, gid, mailbox, &hash, &prev, &index); + err = find_mgm(dev, gid, protocol, mailbox, &hash, &prev, &index); if (err) goto out; @@ -270,7 +273,7 @@ int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16]) goto out; } - members_count = be32_to_cpu(mgm->members_count); + members_count = be32_to_cpu(mgm->members_count) & 0xffffff; for (loc = -1, i = 0; i < members_count; ++i) if ((be32_to_cpu(mgm->qp[i]) & MGM_QPN_MASK) == qp->qpn) loc = i; @@ -282,7 +285,7 @@ int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16]) } - mgm->members_count = cpu_to_be32(--members_count); + mgm->members_count = cpu_to_be32(--members_count | (u32) protocol << 30); mgm->qp[loc] = mgm->qp[i - 1]; mgm->qp[i - 1] = 0; diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h index a7b15bc7648e..049214642036 100644 --- a/include/linux/mlx4/device.h +++ b/include/linux/mlx4/device.h @@ -144,6 +144,11 @@ enum { MLX4_STAT_RATE_OFFSET = 5 }; +enum mlx4_protocol { + MLX4_PROTOCOL_IB, + MLX4_PROTOCOL_EN, +}; + enum { MLX4_MTT_FLAG_PRESENT = 1 }; @@ -500,8 +505,9 @@ int mlx4_INIT_PORT(struct mlx4_dev *dev, int port); int mlx4_CLOSE_PORT(struct mlx4_dev *dev, int port); int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16], - int block_mcast_loopback); -int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16]); + int block_mcast_loopback, enum mlx4_protocol protocol); +int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16], + enum mlx4_protocol protocol); int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *index); void mlx4_unregister_mac(struct mlx4_dev *dev, u8 port, int index); diff --git a/include/linux/mlx4/driver.h b/include/linux/mlx4/driver.h index f407cd4bfb34..e1eebf78caba 100644 --- a/include/linux/mlx4/driver.h +++ b/include/linux/mlx4/driver.h @@ -34,6 +34,7 @@ #define MLX4_DRIVER_H #include +#include struct mlx4_dev; @@ -44,11 +45,6 @@ enum mlx4_dev_event { MLX4_DEV_EVENT_PORT_REINIT, }; -enum mlx4_protocol { - MLX4_PROTOCOL_IB, - MLX4_PROTOCOL_EN, -}; - struct mlx4_interface { void * (*add) (struct mlx4_dev *dev); void (*remove)(struct mlx4_dev *dev, void *context); -- cgit v1.2.3 From 84c89557a302e18414a011cc52b1abd034860743 Mon Sep 17 00:00:00 2001 From: Peter Jones Date: Thu, 13 Jan 2011 19:59:47 +0000 Subject: dm ioctl: allow rename to fill empty uuid Allow the uuid of a mapped device to be set after device creation. Previously the uuid (which is optional) could only be set by DM_DEV_CREATE. If no uuid was supplied it could not be set later. Sometimes it's necessary to create the device before the uuid is known, and in such cases the uuid must be filled in after the creation. This patch extends DM_DEV_RENAME to accept a uuid accompanied by a new flag DM_UUID_FLAG. This can only be done once and if no uuid was previously supplied. It cannot be used to change an existing uuid. DM_VERSION_MINOR is also bumped to 19 to indicate this interface extension is available. Signed-off-by: Peter Jones Signed-off-by: Jonathan Brassow Signed-off-by: Alasdair G Kergon --- drivers/md/dm-ioctl.c | 103 +++++++++++++++++++++++++++++++++++------------ include/linux/dm-ioctl.h | 12 ++++-- 2 files changed, 87 insertions(+), 28 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index 4b54618b4159..f3481d922206 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -295,19 +295,55 @@ retry: DMWARN("remove_all left %d open device(s)", dev_skipped); } +/* + * Set the uuid of a hash_cell that isn't already set. + */ +static void __set_cell_uuid(struct hash_cell *hc, char *new_uuid) +{ + mutex_lock(&dm_hash_cells_mutex); + hc->uuid = new_uuid; + mutex_unlock(&dm_hash_cells_mutex); + + list_add(&hc->uuid_list, _uuid_buckets + hash_str(new_uuid)); +} + +/* + * Changes the name of a hash_cell and returns the old name for + * the caller to free. + */ +static char *__change_cell_name(struct hash_cell *hc, char *new_name) +{ + char *old_name; + + /* + * Rename and move the name cell. + */ + list_del(&hc->name_list); + old_name = hc->name; + + mutex_lock(&dm_hash_cells_mutex); + hc->name = new_name; + mutex_unlock(&dm_hash_cells_mutex); + + list_add(&hc->name_list, _name_buckets + hash_str(new_name)); + + return old_name; +} + static struct mapped_device *dm_hash_rename(struct dm_ioctl *param, const char *new) { - char *new_name, *old_name; + char *new_data, *old_name = NULL; struct hash_cell *hc; struct dm_table *table; struct mapped_device *md; + unsigned change_uuid = (param->flags & DM_UUID_FLAG) ? 1 : 0; /* * duplicate new. */ - new_name = kstrdup(new, GFP_KERNEL); - if (!new_name) + new_data = kstrdup(new, GFP_KERNEL); + if (!new_data) return ERR_PTR(-ENOMEM); down_write(&_hash_lock); @@ -315,13 +351,19 @@ static struct mapped_device *dm_hash_rename(struct dm_ioctl *param, /* * Is new free ? */ - hc = __get_name_cell(new); + if (change_uuid) + hc = __get_uuid_cell(new); + else + hc = __get_name_cell(new); + if (hc) { - DMWARN("asked to rename to an already-existing name %s -> %s", + DMWARN("Unable to change %s on mapped device %s to one that " + "already exists: %s", + change_uuid ? "uuid" : "name", param->name, new); dm_put(hc->md); up_write(&_hash_lock); - kfree(new_name); + kfree(new_data); return ERR_PTR(-EBUSY); } @@ -330,22 +372,30 @@ static struct mapped_device *dm_hash_rename(struct dm_ioctl *param, */ hc = __get_name_cell(param->name); if (!hc) { - DMWARN("asked to rename a non-existent device %s -> %s", - param->name, new); + DMWARN("Unable to rename non-existent device, %s to %s%s", + param->name, change_uuid ? "uuid " : "", new); up_write(&_hash_lock); - kfree(new_name); + kfree(new_data); return ERR_PTR(-ENXIO); } /* - * rename and move the name cell. + * Does this device already have a uuid? */ - list_del(&hc->name_list); - old_name = hc->name; - mutex_lock(&dm_hash_cells_mutex); - hc->name = new_name; - mutex_unlock(&dm_hash_cells_mutex); - list_add(&hc->name_list, _name_buckets + hash_str(new_name)); + if (change_uuid && hc->uuid) { + DMWARN("Unable to change uuid of mapped device %s to %s " + "because uuid is already set to %s", + param->name, new, hc->uuid); + dm_put(hc->md); + up_write(&_hash_lock); + kfree(new_data); + return ERR_PTR(-EINVAL); + } + + if (change_uuid) + __set_cell_uuid(hc, new_data); + else + old_name = __change_cell_name(hc, new_data); /* * Wake up any dm event waiters. @@ -774,21 +824,24 @@ static int invalid_str(char *str, void *end) static int dev_rename(struct dm_ioctl *param, size_t param_size) { int r; - char *new_name = (char *) param + param->data_start; + char *new_data = (char *) param + param->data_start; struct mapped_device *md; + unsigned change_uuid = (param->flags & DM_UUID_FLAG) ? 1 : 0; - if (new_name < param->data || - invalid_str(new_name, (void *) param + param_size) || - strlen(new_name) > DM_NAME_LEN - 1) { - DMWARN("Invalid new logical volume name supplied."); + if (new_data < param->data || + invalid_str(new_data, (void *) param + param_size) || + strlen(new_data) > (change_uuid ? DM_UUID_LEN - 1 : DM_NAME_LEN - 1)) { + DMWARN("Invalid new mapped device name or uuid string supplied."); return -EINVAL; } - r = check_name(new_name); - if (r) - return r; + if (!change_uuid) { + r = check_name(new_data); + if (r) + return r; + } - md = dm_hash_rename(param, new_name); + md = dm_hash_rename(param, new_data); if (IS_ERR(md)) return PTR_ERR(md); diff --git a/include/linux/dm-ioctl.h b/include/linux/dm-ioctl.h index 49eab360d5d4..e453e1174ae1 100644 --- a/include/linux/dm-ioctl.h +++ b/include/linux/dm-ioctl.h @@ -44,7 +44,7 @@ * Remove a device, destroy any tables. * * DM_DEV_RENAME: - * Rename a device. + * Rename a device or set its uuid if none was previously supplied. * * DM_SUSPEND: * This performs both suspend and resume, depending which flag is @@ -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 18 +#define DM_VERSION_MINOR 19 #define DM_VERSION_PATCHLEVEL 0 -#define DM_VERSION_EXTRA "-ioctl (2010-06-29)" +#define DM_VERSION_EXTRA "-ioctl (2010-10-14)" /* Status bits */ #define DM_READONLY_FLAG (1 << 0) /* In/Out */ @@ -322,4 +322,10 @@ enum { */ #define DM_UEVENT_GENERATED_FLAG (1 << 13) /* Out */ +/* + * If set, rename changes the uuid not the name. Only permitted + * if no uuid was previously supplied: an existing uuid cannot be changed. + */ +#define DM_UUID_FLAG (1 << 14) /* In */ + #endif /* _LINUX_DM_IOCTL_H */ -- cgit v1.2.3 From 86a54a4802df10d23ccd655e2083e812fe990243 Mon Sep 17 00:00:00 2001 From: Jonathan Brassow Date: Thu, 13 Jan 2011 19:59:52 +0000 Subject: dm log userspace: add version number to comms This patch adds a 'version' field to the 'dm_ulog_request' structure. The 'version' field is taken from a portion of the unused 'padding' field in the 'dm_ulog_request' structure. This was done to avoid changing the size of the structure and possibly disrupting backwards compatibility. The version number will help notify user-space daemons when a change has been made to the kernel/userspace log API. Signed-off-by: Jonathan Brassow Signed-off-by: Mike Snitzer Signed-off-by: Alasdair G Kergon --- drivers/md/dm-log-userspace-base.c | 6 ++++-- drivers/md/dm-log-userspace-transfer.c | 1 + include/linux/dm-log-userspace.h | 13 ++++++++++++- 3 files changed, 17 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/dm-log-userspace-base.c b/drivers/md/dm-log-userspace-base.c index 31e1687e7bf6..aa2e0c374ab3 100644 --- a/drivers/md/dm-log-userspace-base.c +++ b/drivers/md/dm-log-userspace-base.c @@ -12,6 +12,8 @@ #include "dm-log-userspace-transfer.h" +#define DM_LOG_USERSPACE_VSN "1.1.0" + struct flush_entry { int type; region_t region; @@ -765,7 +767,7 @@ static int __init userspace_dirty_log_init(void) return r; } - DMINFO("version 1.0.0 loaded"); + DMINFO("version " DM_LOG_USERSPACE_VSN " loaded"); return 0; } @@ -775,7 +777,7 @@ static void __exit userspace_dirty_log_exit(void) dm_ulog_tfr_exit(); mempool_destroy(flush_entry_pool); - DMINFO("version 1.0.0 unloaded"); + DMINFO("version " DM_LOG_USERSPACE_VSN " unloaded"); return; } diff --git a/drivers/md/dm-log-userspace-transfer.c b/drivers/md/dm-log-userspace-transfer.c index 075cbcf8a9f5..049eaf12aaab 100644 --- a/drivers/md/dm-log-userspace-transfer.c +++ b/drivers/md/dm-log-userspace-transfer.c @@ -198,6 +198,7 @@ resend: memset(tfr, 0, DM_ULOG_PREALLOCED_SIZE - sizeof(struct cn_msg)); memcpy(tfr->uuid, uuid, DM_UUID_LEN); + tfr->version = DM_ULOG_REQUEST_VERSION; tfr->luid = luid; tfr->seq = dm_ulog_seq++; diff --git a/include/linux/dm-log-userspace.h b/include/linux/dm-log-userspace.h index 0c3c3a2110c4..eeace7d3ff15 100644 --- a/include/linux/dm-log-userspace.h +++ b/include/linux/dm-log-userspace.h @@ -370,6 +370,16 @@ #define DM_ULOG_REQUEST_TYPE(request_type) \ (DM_ULOG_REQUEST_MASK & (request_type)) +/* + * DM_ULOG_REQUEST_VERSION is incremented when there is a + * change to the way information is passed between kernel + * and userspace. This could be a structure change of + * dm_ulog_request or a change in the way requests are + * issued/handled. Changes are outlined here: + * version 1: Initial implementation + */ +#define DM_ULOG_REQUEST_VERSION 1 + struct dm_ulog_request { /* * The local unique identifier (luid) and the universally unique @@ -383,8 +393,9 @@ struct dm_ulog_request { */ uint64_t luid; char uuid[DM_UUID_LEN]; - char padding[7]; /* Padding because DM_UUID_LEN = 129 */ + char padding[3]; /* Padding because DM_UUID_LEN = 129 */ + uint32_t version; /* See DM_ULOG_REQUEST_VERSION */ int32_t error; /* Used to report back processing errors */ uint32_t seq; /* Sequence number for request */ -- cgit v1.2.3 From 9c4376de98719d2768dd919553843de34bb094a6 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 13 Jan 2011 19:59:58 +0000 Subject: dm: use non reentrant workqueues if equivalent kmirrord_wq, kcopyd_work and md->wq are created per dm instance and serve only a single work item from the dm instance, so non-reentrant workqueues would provide the same ordering guarantees as ordered ones while allowing CPU affinity and use of the workqueues for other purposes. Switch them to non-reentrant workqueues. Signed-off-by: Tejun Heo Signed-off-by: Mike Snitzer Signed-off-by: Alasdair G Kergon --- drivers/md/dm-kcopyd.c | 3 ++- drivers/md/dm-raid1.c | 5 +++-- drivers/md/dm.c | 3 ++- include/linux/dm-ioctl.h | 4 ++-- 4 files changed, 9 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/dm-kcopyd.c b/drivers/md/dm-kcopyd.c index 63d67169c7f4..924f5f0084c2 100644 --- a/drivers/md/dm-kcopyd.c +++ b/drivers/md/dm-kcopyd.c @@ -672,7 +672,8 @@ int dm_kcopyd_client_create(unsigned int nr_pages, goto bad_slab; INIT_WORK(&kc->kcopyd_work, do_work); - kc->kcopyd_wq = alloc_ordered_workqueue("kcopyd", WQ_MEM_RECLAIM); + kc->kcopyd_wq = alloc_workqueue("kcopyd", + WQ_NON_REENTRANT | WQ_MEM_RECLAIM, 0); if (!kc->kcopyd_wq) goto bad_workqueue; diff --git a/drivers/md/dm-raid1.c b/drivers/md/dm-raid1.c index 39917430f9f8..dee326775c60 100644 --- a/drivers/md/dm-raid1.c +++ b/drivers/md/dm-raid1.c @@ -1085,7 +1085,8 @@ static int mirror_ctr(struct dm_target *ti, unsigned int argc, char **argv) ti->num_flush_requests = 1; ti->num_discard_requests = 1; - ms->kmirrord_wq = alloc_ordered_workqueue("kmirrord", WQ_MEM_RECLAIM); + ms->kmirrord_wq = alloc_workqueue("kmirrord", + WQ_NON_REENTRANT | WQ_MEM_RECLAIM, 0); if (!ms->kmirrord_wq) { DMERR("couldn't start kmirrord"); r = -ENOMEM; @@ -1414,7 +1415,7 @@ static int mirror_iterate_devices(struct dm_target *ti, static struct target_type mirror_target = { .name = "mirror", - .version = {1, 12, 0}, + .version = {1, 12, 1}, .module = THIS_MODULE, .ctr = mirror_ctr, .dtr = mirror_dtr, diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 39aaa9290128..e504bb40d60e 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -1883,7 +1883,8 @@ static struct mapped_device *alloc_dev(int minor) add_disk(md->disk); format_dev_t(md->name, MKDEV(_major, minor)); - md->wq = alloc_ordered_workqueue("kdmflush", WQ_MEM_RECLAIM); + md->wq = alloc_workqueue("kdmflush", + WQ_NON_REENTRANT | WQ_MEM_RECLAIM, 0); if (!md->wq) goto bad_thread; diff --git a/include/linux/dm-ioctl.h b/include/linux/dm-ioctl.h index e453e1174ae1..78bbf47bbb96 100644 --- a/include/linux/dm-ioctl.h +++ b/include/linux/dm-ioctl.h @@ -268,8 +268,8 @@ enum { #define DM_VERSION_MAJOR 4 #define DM_VERSION_MINOR 19 -#define DM_VERSION_PATCHLEVEL 0 -#define DM_VERSION_EXTRA "-ioctl (2010-10-14)" +#define DM_VERSION_PATCHLEVEL 1 +#define DM_VERSION_EXTRA "-ioctl (2011-01-07)" /* Status bits */ #define DM_READONLY_FLAG (1 << 0) /* In/Out */ -- cgit v1.2.3 From 9d357b0787bb3c91835d5e658c3bda178f9ca419 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 13 Jan 2011 20:00:01 +0000 Subject: dm: introduce target callbacks and congestion callback DM currently implements congestion checking by checking on congestion in each component device. For raid456 we need to also check if the stripe cache is congested. Add per-target congestion checker callback support. Extending the target_callbacks structure with additional callback functions allows for establishing multiple callbacks per-target (a callback is also needed for unplug). Cc: linux-raid@vger.kernel.org Signed-off-by: NeilBrown Signed-off-by: Jonathan Brassow Signed-off-by: Mike Snitzer Signed-off-by: Alasdair G Kergon --- drivers/md/dm-table.c | 14 ++++++++++++++ include/linux/device-mapper.h | 11 +++++++++++ 2 files changed, 25 insertions(+) (limited to 'include/linux') diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index 985c20a4f30e..7e2ec3c05550 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -71,6 +71,8 @@ struct dm_table { void *event_context; struct dm_md_mempools *mempools; + + struct list_head target_callbacks; }; /* @@ -204,6 +206,7 @@ int dm_table_create(struct dm_table **result, fmode_t mode, return -ENOMEM; INIT_LIST_HEAD(&t->devices); + INIT_LIST_HEAD(&t->target_callbacks); atomic_set(&t->holders, 0); t->discards_supported = 1; @@ -1225,10 +1228,17 @@ int dm_table_resume_targets(struct dm_table *t) return 0; } +void dm_table_add_target_callbacks(struct dm_table *t, struct dm_target_callbacks *cb) +{ + list_add(&cb->list, &t->target_callbacks); +} +EXPORT_SYMBOL_GPL(dm_table_add_target_callbacks); + int dm_table_any_congested(struct dm_table *t, int bdi_bits) { struct dm_dev_internal *dd; struct list_head *devices = dm_table_get_devices(t); + struct dm_target_callbacks *cb; int r = 0; list_for_each_entry(dd, devices, list) { @@ -1243,6 +1253,10 @@ int dm_table_any_congested(struct dm_table *t, int bdi_bits) bdevname(dd->dm_dev.bdev, b)); } + list_for_each_entry(cb, &t->target_callbacks, list) + if (cb->congested_fn) + r |= cb->congested_fn(cb, bdi_bits); + return r; } diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h index 2970022faa63..4b1c63d478ab 100644 --- a/include/linux/device-mapper.h +++ b/include/linux/device-mapper.h @@ -193,6 +193,12 @@ struct dm_target { char *error; }; +/* Each target can link one of these into the table */ +struct dm_target_callbacks { + struct list_head list; + int (*congested_fn) (struct dm_target_callbacks *, int); +}; + int dm_register_target(struct target_type *t); void dm_unregister_target(struct target_type *t); @@ -268,6 +274,11 @@ int dm_table_create(struct dm_table **result, fmode_t mode, int dm_table_add_target(struct dm_table *t, const char *type, sector_t start, sector_t len, char *params); +/* + * Target_ctr should call this if it needs to add any callbacks. + */ +void dm_table_add_target_callbacks(struct dm_table *t, struct dm_target_callbacks *cb); + /* * Finally call this to make the table ready for use. */ -- cgit v1.2.3 From 99d03c141b40914b67d63c9d23b8da4386422ed7 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 13 Jan 2011 20:00:02 +0000 Subject: dm: per target unplug callback support Add per-target unplug callback support. Cc: linux-raid@vger.kernel.org Signed-off-by: NeilBrown Signed-off-by: Jonathan Brassow Signed-off-by: Mike Snitzer Signed-off-by: Alasdair G Kergon --- drivers/md/dm-table.c | 5 +++++ include/linux/device-mapper.h | 1 + 2 files changed, 6 insertions(+) (limited to 'include/linux') diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index 7e2ec3c05550..dffa0ac7c4f0 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -1278,6 +1278,7 @@ void dm_table_unplug_all(struct dm_table *t) { struct dm_dev_internal *dd; struct list_head *devices = dm_table_get_devices(t); + struct dm_target_callbacks *cb; list_for_each_entry(dd, devices, list) { struct request_queue *q = bdev_get_queue(dd->dm_dev.bdev); @@ -1290,6 +1291,10 @@ void dm_table_unplug_all(struct dm_table *t) dm_device_name(t->md), bdevname(dd->dm_dev.bdev, b)); } + + list_for_each_entry(cb, &t->target_callbacks, list) + if (cb->unplug_fn) + cb->unplug_fn(cb); } struct mapped_device *dm_table_get_md(struct dm_table *t) diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h index 4b1c63d478ab..272496d1fae4 100644 --- a/include/linux/device-mapper.h +++ b/include/linux/device-mapper.h @@ -197,6 +197,7 @@ struct dm_target { struct dm_target_callbacks { struct list_head list; int (*congested_fn) (struct dm_target_callbacks *, int); + void (*unplug_fn)(struct dm_target_callbacks *); }; int dm_register_target(struct target_type *t); -- cgit v1.2.3 From 8d661f1e462d50bd83de87ee628aaf820ce3c66c Mon Sep 17 00:00:00 2001 From: Amitkumar Karwar Date: Tue, 11 Jan 2011 16:14:24 -0800 Subject: ieee80211: correct IEEE80211_ADDBA_PARAM_BUF_SIZE_MASK macro It is defined in include/linux/ieee80211.h. As per IEEE spec. bit6 to bit15 in block ack parameter represents buffer size. So the bitmask should be 0xFFC0. Signed-off-by: Amitkumar Karwar Signed-off-by: Bing Zhao Cc: stable@kernel.org Reviewed-by: Johannes Berg Signed-off-by: John W. Linville --- include/linux/ieee80211.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index 6042228954a7..294169e31364 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -959,7 +959,7 @@ struct ieee80211_ht_info { /* block-ack parameters */ #define IEEE80211_ADDBA_PARAM_POLICY_MASK 0x0002 #define IEEE80211_ADDBA_PARAM_TID_MASK 0x003C -#define IEEE80211_ADDBA_PARAM_BUF_SIZE_MASK 0xFFA0 +#define IEEE80211_ADDBA_PARAM_BUF_SIZE_MASK 0xFFC0 #define IEEE80211_DELBA_PARAM_TID_MASK 0xF000 #define IEEE80211_DELBA_PARAM_INITIATOR_MASK 0x0800 -- cgit v1.2.3 From d8a3515e2a9523f8ed56d1f4537d16338bda2bc2 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 13 Jan 2011 17:26:46 -0800 Subject: Revert "gpiolib: annotate gpio-intialization with __must_check" This reverts commit 0fdae42d361bbb431ca0ab0efed5126a94821177, which wasn't really supposed to go in, and causes lots of annoying warnings. Quoth Andrew: "Complete brainfart - I meant to drop that patch ages ago." Quoth Greg: "Ick, yeah, that patch isn't ok to go in as-is, all of the callers need to be fixed up first, which is what I thought we had agreed on..." Reported-by: Stephen Rothwell Acked-by: Andrew Morton Acked-by: Greg KH Signed-off-by: Linus Torvalds --- Documentation/gpio.txt | 2 +- include/asm-generic/gpio.h | 10 +++++----- include/linux/gpio.h | 6 +++--- 3 files changed, 9 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/Documentation/gpio.txt b/Documentation/gpio.txt index a492d92bb098..792faa3c06cf 100644 --- a/Documentation/gpio.txt +++ b/Documentation/gpio.txt @@ -135,7 +135,7 @@ setting up a platform_device using the GPIO, is mark its direction: int gpio_direction_input(unsigned gpio); int gpio_direction_output(unsigned gpio, int value); -The return value is zero for success, else a negative errno. It must +The return value is zero for success, else a negative errno. It should be checked, since the get/set calls don't have error returns and since misconfiguration is possible. You should normally issue these calls from a task context. However, for spinlock-safe GPIOs it's OK to use them diff --git a/include/asm-generic/gpio.h b/include/asm-generic/gpio.h index 6098cae2af8e..ff5c66080c8c 100644 --- a/include/asm-generic/gpio.h +++ b/include/asm-generic/gpio.h @@ -147,11 +147,11 @@ extern struct gpio_chip *gpiochip_find(void *data, /* Always use the library code for GPIO management calls, * or when sleeping may be involved. */ -extern int __must_check gpio_request(unsigned gpio, const char *label); +extern int gpio_request(unsigned gpio, const char *label); extern void gpio_free(unsigned gpio); -extern int __must_check gpio_direction_input(unsigned gpio); -extern int __must_check gpio_direction_output(unsigned gpio, int value); +extern int gpio_direction_input(unsigned gpio); +extern int gpio_direction_output(unsigned gpio, int value); extern int gpio_set_debounce(unsigned gpio, unsigned debounce); @@ -192,8 +192,8 @@ struct gpio { const char *label; }; -extern int __must_check gpio_request_one(unsigned gpio, unsigned long flags, const char *label); -extern int __must_check gpio_request_array(struct gpio *array, size_t num); +extern int gpio_request_one(unsigned gpio, unsigned long flags, const char *label); +extern int gpio_request_array(struct gpio *array, size_t num); extern void gpio_free_array(struct gpio *array, size_t num); #ifdef CONFIG_GPIO_SYSFS diff --git a/include/linux/gpio.h b/include/linux/gpio.h index f79d67f413e4..4b47ed96f131 100644 --- a/include/linux/gpio.h +++ b/include/linux/gpio.h @@ -30,7 +30,7 @@ static inline int gpio_is_valid(int number) return 0; } -static inline int __must_check gpio_request(unsigned gpio, const char *label) +static inline int gpio_request(unsigned gpio, const char *label) { return -ENOSYS; } @@ -62,12 +62,12 @@ static inline void gpio_free_array(struct gpio *array, size_t num) WARN_ON(1); } -static inline int __must_check gpio_direction_input(unsigned gpio) +static inline int gpio_direction_input(unsigned gpio) { return -ENOSYS; } -static inline int __must_check gpio_direction_output(unsigned gpio, int value) +static inline int gpio_direction_output(unsigned gpio, int value) { return -ENOSYS; } -- cgit v1.2.3 From 6c9ae009b298753a3baf71298d676a68b5a10c8f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 13 Jan 2011 15:45:38 -0800 Subject: irq: use per_cpu kstat_irqs Use modern per_cpu API to increment {soft|hard}irq counters, and use per_cpu allocation for (struct irq_desc)->kstats_irq instead of an array. This gives better SMP/NUMA locality and saves few instructions per irq. With small nr_cpuids values (8 for example), kstats_irq was a small array (less than L1_CACHE_BYTES), potentially source of false sharing. In the !CONFIG_SPARSE_IRQ case, remove the huge, NUMA/cache unfriendly kstat_irqs_all[NR_IRQS][NR_CPUS] array. Note: we still populate kstats_irq for all possible irqs in early_irq_init(). We probably could use on-demand allocations. (Code included in alloc_descs()). Problem is not all IRQS are used with a prior alloc_descs() call. kstat_irqs_this_cpu() is not used anymore, remove it. Signed-off-by: Eric Dumazet Reviewed-by: Christoph Lameter Cc: Ingo Molnar Cc: Andi Kleen Cc: Tejun Heo Cc: Thomas Gleixner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/irqdesc.h | 2 +- include/linux/kernel_stat.h | 19 +++++++++---------- kernel/irq/irqdesc.c | 40 ++++++++++++++++++++++++++++++---------- 3 files changed, 40 insertions(+), 21 deletions(-) (limited to 'include/linux') diff --git a/include/linux/irqdesc.h b/include/linux/irqdesc.h index 979c68cc7458..6a64c6fa81af 100644 --- a/include/linux/irqdesc.h +++ b/include/linux/irqdesc.h @@ -57,7 +57,7 @@ struct irq_desc { #endif struct timer_rand_state *timer_rand_state; - unsigned int *kstat_irqs; + unsigned int __percpu *kstat_irqs; irq_flow_handler_t handle_irq; struct irqaction *action; /* IRQ action list */ unsigned int status; /* IRQ status */ diff --git a/include/linux/kernel_stat.h b/include/linux/kernel_stat.h index 44e83ba12b5b..0cce2db580c3 100644 --- a/include/linux/kernel_stat.h +++ b/include/linux/kernel_stat.h @@ -46,16 +46,14 @@ DECLARE_PER_CPU(struct kernel_stat, kstat); extern unsigned long long nr_context_switches(void); #ifndef CONFIG_GENERIC_HARDIRQS -#define kstat_irqs_this_cpu(irq) \ - (this_cpu_read(kstat.irqs[irq]) struct irq_desc; static inline void kstat_incr_irqs_this_cpu(unsigned int irq, struct irq_desc *desc) { - kstat_this_cpu.irqs[irq]++; - kstat_this_cpu.irqs_sum++; + __this_cpu_inc(kstat.irqs[irq]); + __this_cpu_inc(kstat.irqs_sum); } static inline unsigned int kstat_irqs_cpu(unsigned int irq, int cpu) @@ -65,17 +63,18 @@ static inline unsigned int kstat_irqs_cpu(unsigned int irq, int cpu) #else #include extern unsigned int kstat_irqs_cpu(unsigned int irq, int cpu); -#define kstat_irqs_this_cpu(DESC) \ - ((DESC)->kstat_irqs[smp_processor_id()]) -#define kstat_incr_irqs_this_cpu(irqno, DESC) do {\ - ((DESC)->kstat_irqs[smp_processor_id()]++);\ - kstat_this_cpu.irqs_sum++; } while (0) + +#define kstat_incr_irqs_this_cpu(irqno, DESC) \ +do { \ + __this_cpu_inc(*(DESC)->kstat_irqs); \ + __this_cpu_inc(kstat.irqs_sum); \ +} while (0) #endif static inline void kstat_incr_softirqs_this_cpu(unsigned int irq) { - kstat_this_cpu.softirqs[irq]++; + __this_cpu_inc(kstat.softirqs[irq]); } static inline unsigned int kstat_softirqs_cpu(unsigned int irq, int cpu) diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c index 9988d03797f5..282f20230e67 100644 --- a/kernel/irq/irqdesc.c +++ b/kernel/irq/irqdesc.c @@ -72,6 +72,8 @@ static inline int desc_node(struct irq_desc *desc) { return 0; } static void desc_set_defaults(unsigned int irq, struct irq_desc *desc, int node) { + int cpu; + desc->irq_data.irq = irq; desc->irq_data.chip = &no_irq_chip; desc->irq_data.chip_data = NULL; @@ -83,7 +85,8 @@ static void desc_set_defaults(unsigned int irq, struct irq_desc *desc, int node) desc->irq_count = 0; desc->irqs_unhandled = 0; desc->name = NULL; - memset(desc->kstat_irqs, 0, nr_cpu_ids * sizeof(*(desc->kstat_irqs))); + for_each_possible_cpu(cpu) + *per_cpu_ptr(desc->kstat_irqs, cpu) = 0; desc_smp_init(desc, node); } @@ -133,8 +136,7 @@ static struct irq_desc *alloc_desc(int irq, int node) if (!desc) return NULL; /* allocate based on nr_cpu_ids */ - desc->kstat_irqs = kzalloc_node(nr_cpu_ids * sizeof(*desc->kstat_irqs), - gfp, node); + desc->kstat_irqs = alloc_percpu(unsigned int); if (!desc->kstat_irqs) goto err_desc; @@ -149,7 +151,7 @@ static struct irq_desc *alloc_desc(int irq, int node) return desc; err_kstat: - kfree(desc->kstat_irqs); + free_percpu(desc->kstat_irqs); err_desc: kfree(desc); return NULL; @@ -166,7 +168,7 @@ static void free_desc(unsigned int irq) mutex_unlock(&sparse_irq_lock); free_masks(desc); - kfree(desc->kstat_irqs); + free_percpu(desc->kstat_irqs); kfree(desc); } @@ -234,7 +236,6 @@ struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned_in_smp = { } }; -static unsigned int kstat_irqs_all[NR_IRQS][NR_CPUS]; int __init early_irq_init(void) { int count, i, node = first_online_node; @@ -250,7 +251,8 @@ int __init early_irq_init(void) for (i = 0; i < count; i++) { desc[i].irq_data.irq = i; desc[i].irq_data.chip = &no_irq_chip; - desc[i].kstat_irqs = kstat_irqs_all[i]; + /* TODO : do this allocation on-demand ... */ + desc[i].kstat_irqs = alloc_percpu(unsigned int); alloc_masks(desc + i, GFP_KERNEL, node); desc_smp_init(desc + i, node); lockdep_set_class(&desc[i].lock, &irq_desc_lock_class); @@ -275,6 +277,22 @@ static void free_desc(unsigned int irq) static inline int alloc_descs(unsigned int start, unsigned int cnt, int node) { +#if defined(CONFIG_KSTAT_IRQS_ONDEMAND) + struct irq_desc *desc; + unsigned int i; + + for (i = 0; i < cnt; i++) { + desc = irq_to_desc(start + i); + if (desc && !desc->kstat_irqs) { + unsigned int __percpu *stats = alloc_percpu(unsigned int); + + if (!stats) + return -1; + if (cmpxchg(&desc->kstat_irqs, NULL, stats) != NULL) + free_percpu(stats); + } + } +#endif return start; } #endif /* !CONFIG_SPARSE_IRQ */ @@ -391,7 +409,9 @@ void dynamic_irq_cleanup(unsigned int irq) unsigned int kstat_irqs_cpu(unsigned int irq, int cpu) { struct irq_desc *desc = irq_to_desc(irq); - return desc ? desc->kstat_irqs[cpu] : 0; + + return desc && desc->kstat_irqs ? + *per_cpu_ptr(desc->kstat_irqs, cpu) : 0; } #ifdef CONFIG_GENERIC_HARDIRQS @@ -401,10 +421,10 @@ unsigned int kstat_irqs(unsigned int irq) int cpu; int sum = 0; - if (!desc) + if (!desc || !desc->kstat_irqs) return 0; for_each_possible_cpu(cpu) - sum += desc->kstat_irqs[cpu]; + sum += *per_cpu_ptr(desc->kstat_irqs, cpu); return sum; } #endif /* CONFIG_GENERIC_HARDIRQS */ -- cgit v1.2.3 From 43bb40c9e3aa51a3b038c9df2c9afb4d4685614d Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Thu, 13 Jan 2011 15:45:40 -0800 Subject: sched: remove long deprecated CLONE_STOPPED flag This warning was added in commit bdff746a3915 ("clone: prepare to recycle CLONE_STOPPED") three years ago. 2.6.26 came and went. As far as I know, no-one is actually using CLONE_STOPPED. Signed-off-by: Dave Jones Cc: Peter Zijlstra Cc: Ingo Molnar Cc: Tejun Heo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 3 ++- kernel/fork.c | 28 +--------------------------- 2 files changed, 3 insertions(+), 28 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index 96e23215e276..07402530fc70 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -21,7 +21,8 @@ #define CLONE_DETACHED 0x00400000 /* Unused, ignored */ #define CLONE_UNTRACED 0x00800000 /* set if the tracing process can't force CLONE_PTRACE on this clone */ #define CLONE_CHILD_SETTID 0x01000000 /* set the TID in the child */ -#define CLONE_STOPPED 0x02000000 /* Start in stopped state */ +/* 0x02000000 was previously the unused CLONE_STOPPED (Start in stopped state) + and is now available for re-use. */ #define CLONE_NEWUTS 0x04000000 /* New utsname group? */ #define CLONE_NEWIPC 0x08000000 /* New ipcs */ #define CLONE_NEWUSER 0x10000000 /* New user namespace */ diff --git a/kernel/fork.c b/kernel/fork.c index d9b44f20b6b0..76a1fdd80bdf 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1409,23 +1409,6 @@ long do_fork(unsigned long clone_flags, return -EPERM; } - /* - * We hope to recycle these flags after 2.6.26 - */ - if (unlikely(clone_flags & CLONE_STOPPED)) { - static int __read_mostly count = 100; - - if (count > 0 && printk_ratelimit()) { - char comm[TASK_COMM_LEN]; - - count--; - printk(KERN_INFO "fork(): process `%s' used deprecated " - "clone flags 0x%lx\n", - get_task_comm(comm, current), - clone_flags & CLONE_STOPPED); - } - } - /* * When called from kernel_thread, don't do user tracing stuff. */ @@ -1464,16 +1447,7 @@ long do_fork(unsigned long clone_flags, */ p->flags &= ~PF_STARTING; - if (unlikely(clone_flags & CLONE_STOPPED)) { - /* - * We'll start up with an immediate SIGSTOP. - */ - sigaddset(&p->pending.signal, SIGSTOP); - set_tsk_thread_flag(p, TIF_SIGPENDING); - __set_task_state(p, TASK_STOPPED); - } else { - wake_up_new_task(p, clone_flags); - } + wake_up_new_task(p, clone_flags); tracehook_report_clone_complete(trace, regs, clone_flags, nr, p); -- cgit v1.2.3 From 88f5acf88ae6a9778f6d25d0d5d7ec2d57764a97 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Thu, 13 Jan 2011 15:45:41 -0800 Subject: mm: page allocator: adjust the per-cpu counter threshold when memory is low Commit aa45484 ("calculate a better estimate of NR_FREE_PAGES when memory is low") noted that watermarks were based on the vmstat NR_FREE_PAGES. To avoid synchronization overhead, these counters are maintained on a per-cpu basis and drained both periodically and when a threshold is above a threshold. On large CPU systems, the difference between the estimate and real value of NR_FREE_PAGES can be very high. The system can get into a case where pages are allocated far below the min watermark potentially causing livelock issues. The commit solved the problem by taking a better reading of NR_FREE_PAGES when memory was low. Unfortately, as reported by Shaohua Li this accurate reading can consume a large amount of CPU time on systems with many sockets due to cache line bouncing. This patch takes a different approach. For large machines where counter drift might be unsafe and while kswapd is awake, the per-cpu thresholds for the target pgdat are reduced to limit the level of drift to what should be a safe level. This incurs a performance penalty in heavy memory pressure by a factor that depends on the workload and the machine but the machine should function correctly without accidentally exhausting all memory on a node. There is an additional cost when kswapd wakes and sleeps but the event is not expected to be frequent - in Shaohua's test case, there was one recorded sleep and wake event at least. To ensure that kswapd wakes up, a safe version of zone_watermark_ok() is introduced that takes a more accurate reading of NR_FREE_PAGES when called from wakeup_kswapd, when deciding whether it is really safe to go back to sleep in sleeping_prematurely() and when deciding if a zone is really balanced or not in balance_pgdat(). We are still using an expensive function but limiting how often it is called. When the test case is reproduced, the time spent in the watermark functions is reduced. The following report is on the percentage of time spent cumulatively spent in the functions zone_nr_free_pages(), zone_watermark_ok(), __zone_watermark_ok(), zone_watermark_ok_safe(), zone_page_state_snapshot(), zone_page_state(). vanilla 11.6615% disable-threshold 0.2584% David said: : We had to pull aa454840 "mm: page allocator: calculate a better estimate : of NR_FREE_PAGES when memory is low and kswapd is awake" from 2.6.36 : internally because tests showed that it would cause the machine to stall : as the result of heavy kswapd activity. I merged it back with this fix as : it is pending in the -mm tree and it solves the issue we were seeing, so I : definitely think this should be pushed to -stable (and I would seriously : consider it for 2.6.37 inclusion even at this late date). Signed-off-by: Mel Gorman Reported-by: Shaohua Li Reviewed-by: Christoph Lameter Tested-by: Nicolas Bareil Cc: David Rientjes Cc: Kyle McMartin Cc: [2.6.37.1, 2.6.36.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mmzone.h | 10 +++----- include/linux/vmstat.h | 5 ++++ mm/mmzone.c | 21 ---------------- mm/page_alloc.c | 35 ++++++++++++++++++++------ mm/vmscan.c | 23 +++++++++-------- mm/vmstat.c | 68 +++++++++++++++++++++++++++++++++++++++++++++++++- 6 files changed, 115 insertions(+), 47 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 39c24ebe9cfd..48906629335c 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -458,12 +458,6 @@ static inline int zone_is_oom_locked(const struct zone *zone) return test_bit(ZONE_OOM_LOCKED, &zone->flags); } -#ifdef CONFIG_SMP -unsigned long zone_nr_free_pages(struct zone *zone); -#else -#define zone_nr_free_pages(zone) zone_page_state(zone, NR_FREE_PAGES) -#endif /* CONFIG_SMP */ - /* * The "priority" of VM scanning is how much of the queues we will scan in one * go. A value of 12 for DEF_PRIORITY implies that we will scan 1/4096th of the @@ -661,7 +655,9 @@ typedef struct pglist_data { extern struct mutex zonelists_mutex; void build_all_zonelists(void *data); void wakeup_kswapd(struct zone *zone, int order); -int zone_watermark_ok(struct zone *z, int order, unsigned long mark, +bool zone_watermark_ok(struct zone *z, int order, unsigned long mark, + int classzone_idx, int alloc_flags); +bool zone_watermark_ok_safe(struct zone *z, int order, unsigned long mark, int classzone_idx, int alloc_flags); enum memmap_context { MEMMAP_EARLY, diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h index eaaea37b3b75..e4cc21cf5870 100644 --- a/include/linux/vmstat.h +++ b/include/linux/vmstat.h @@ -254,6 +254,8 @@ extern void dec_zone_state(struct zone *, enum zone_stat_item); extern void __dec_zone_state(struct zone *, enum zone_stat_item); void refresh_cpu_vm_stats(int); +void reduce_pgdat_percpu_threshold(pg_data_t *pgdat); +void restore_pgdat_percpu_threshold(pg_data_t *pgdat); #else /* CONFIG_SMP */ /* @@ -298,6 +300,9 @@ static inline void __dec_zone_page_state(struct page *page, #define dec_zone_page_state __dec_zone_page_state #define mod_zone_page_state __mod_zone_page_state +static inline void reduce_pgdat_percpu_threshold(pg_data_t *pgdat) { } +static inline void restore_pgdat_percpu_threshold(pg_data_t *pgdat) { } + static inline void refresh_cpu_vm_stats(int cpu) { } #endif diff --git a/mm/mmzone.c b/mm/mmzone.c index e35bfb82c855..f5b7d1760213 100644 --- a/mm/mmzone.c +++ b/mm/mmzone.c @@ -87,24 +87,3 @@ int memmap_valid_within(unsigned long pfn, return 1; } #endif /* CONFIG_ARCH_HAS_HOLES_MEMORYMODEL */ - -#ifdef CONFIG_SMP -/* Called when a more accurate view of NR_FREE_PAGES is needed */ -unsigned long zone_nr_free_pages(struct zone *zone) -{ - unsigned long nr_free_pages = zone_page_state(zone, NR_FREE_PAGES); - - /* - * While kswapd is awake, it is considered the zone is under some - * memory pressure. Under pressure, there is a risk that - * per-cpu-counter-drift will allow the min watermark to be breached - * potentially causing a live-lock. While kswapd is awake and - * free pages are low, get a better estimate for free pages - */ - if (nr_free_pages < zone->percpu_drift_mark && - !waitqueue_active(&zone->zone_pgdat->kswapd_wait)) - return zone_page_state_snapshot(zone, NR_FREE_PAGES); - - return nr_free_pages; -} -#endif /* CONFIG_SMP */ diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 826ba6922e84..22a1bb7723e4 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1460,24 +1460,24 @@ static inline int should_fail_alloc_page(gfp_t gfp_mask, unsigned int order) #endif /* CONFIG_FAIL_PAGE_ALLOC */ /* - * Return 1 if free pages are above 'mark'. This takes into account the order + * Return true if free pages are above 'mark'. This takes into account the order * of the allocation. */ -int zone_watermark_ok(struct zone *z, int order, unsigned long mark, - int classzone_idx, int alloc_flags) +static bool __zone_watermark_ok(struct zone *z, int order, unsigned long mark, + int classzone_idx, int alloc_flags, long free_pages) { /* free_pages my go negative - that's OK */ long min = mark; - long free_pages = zone_nr_free_pages(z) - (1 << order) + 1; int o; + free_pages -= (1 << order) + 1; if (alloc_flags & ALLOC_HIGH) min -= min / 2; if (alloc_flags & ALLOC_HARDER) min -= min / 4; if (free_pages <= min + z->lowmem_reserve[classzone_idx]) - return 0; + return false; for (o = 0; o < order; o++) { /* At the next order, this order's pages become unavailable */ free_pages -= z->free_area[o].nr_free << o; @@ -1486,9 +1486,28 @@ int zone_watermark_ok(struct zone *z, int order, unsigned long mark, min >>= 1; if (free_pages <= min) - return 0; + return false; } - return 1; + return true; +} + +bool zone_watermark_ok(struct zone *z, int order, unsigned long mark, + int classzone_idx, int alloc_flags) +{ + return __zone_watermark_ok(z, order, mark, classzone_idx, alloc_flags, + zone_page_state(z, NR_FREE_PAGES)); +} + +bool zone_watermark_ok_safe(struct zone *z, int order, unsigned long mark, + int classzone_idx, int alloc_flags) +{ + long free_pages = zone_page_state(z, NR_FREE_PAGES); + + if (z->percpu_drift_mark && free_pages < z->percpu_drift_mark) + free_pages = zone_page_state_snapshot(z, NR_FREE_PAGES); + + return __zone_watermark_ok(z, order, mark, classzone_idx, alloc_flags, + free_pages); } #ifdef CONFIG_NUMA @@ -2442,7 +2461,7 @@ void show_free_areas(void) " all_unreclaimable? %s" "\n", zone->name, - K(zone_nr_free_pages(zone)), + K(zone_page_state(zone, NR_FREE_PAGES)), K(min_wmark_pages(zone)), K(low_wmark_pages(zone)), K(high_wmark_pages(zone)), diff --git a/mm/vmscan.c b/mm/vmscan.c index 9ca587c69274..5da4295e7d67 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -2143,7 +2143,7 @@ static int sleeping_prematurely(pg_data_t *pgdat, int order, long remaining) if (zone->all_unreclaimable) continue; - if (!zone_watermark_ok(zone, order, high_wmark_pages(zone), + if (!zone_watermark_ok_safe(zone, order, high_wmark_pages(zone), 0, 0)) return 1; } @@ -2230,7 +2230,7 @@ loop_again: shrink_active_list(SWAP_CLUSTER_MAX, zone, &sc, priority, 0); - if (!zone_watermark_ok(zone, order, + if (!zone_watermark_ok_safe(zone, order, high_wmark_pages(zone), 0, 0)) { end_zone = i; break; @@ -2276,7 +2276,7 @@ loop_again: * We put equal pressure on every zone, unless one * zone has way too many pages free already. */ - if (!zone_watermark_ok(zone, order, + if (!zone_watermark_ok_safe(zone, order, 8*high_wmark_pages(zone), end_zone, 0)) shrink_zone(priority, zone, &sc); reclaim_state->reclaimed_slab = 0; @@ -2297,7 +2297,7 @@ loop_again: total_scanned > sc.nr_reclaimed + sc.nr_reclaimed / 2) sc.may_writepage = 1; - if (!zone_watermark_ok(zone, order, + if (!zone_watermark_ok_safe(zone, order, high_wmark_pages(zone), end_zone, 0)) { all_zones_ok = 0; /* @@ -2305,7 +2305,7 @@ loop_again: * means that we have a GFP_ATOMIC allocation * failure risk. Hurry up! */ - if (!zone_watermark_ok(zone, order, + if (!zone_watermark_ok_safe(zone, order, min_wmark_pages(zone), end_zone, 0)) has_under_min_watermark_zone = 1; } else { @@ -2448,7 +2448,9 @@ static int kswapd(void *p) */ if (!sleeping_prematurely(pgdat, order, remaining)) { trace_mm_vmscan_kswapd_sleep(pgdat->node_id); + restore_pgdat_percpu_threshold(pgdat); schedule(); + reduce_pgdat_percpu_threshold(pgdat); } else { if (remaining) count_vm_event(KSWAPD_LOW_WMARK_HIT_QUICKLY); @@ -2487,16 +2489,17 @@ void wakeup_kswapd(struct zone *zone, int order) if (!populated_zone(zone)) return; - pgdat = zone->zone_pgdat; - if (zone_watermark_ok(zone, order, low_wmark_pages(zone), 0, 0)) + if (!cpuset_zone_allowed_hardwall(zone, GFP_KERNEL)) return; + pgdat = zone->zone_pgdat; if (pgdat->kswapd_max_order < order) pgdat->kswapd_max_order = order; - trace_mm_vmscan_wakeup_kswapd(pgdat->node_id, zone_idx(zone), order); - if (!cpuset_zone_allowed_hardwall(zone, GFP_KERNEL)) - return; if (!waitqueue_active(&pgdat->kswapd_wait)) return; + if (zone_watermark_ok_safe(zone, order, low_wmark_pages(zone), 0, 0)) + return; + + trace_mm_vmscan_wakeup_kswapd(pgdat->node_id, zone_idx(zone), order); wake_up_interruptible(&pgdat->kswapd_wait); } diff --git a/mm/vmstat.c b/mm/vmstat.c index 312d728976f1..bc0f095791b4 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -83,6 +83,30 @@ EXPORT_SYMBOL(vm_stat); #ifdef CONFIG_SMP +static int calculate_pressure_threshold(struct zone *zone) +{ + int threshold; + int watermark_distance; + + /* + * As vmstats are not up to date, there is drift between the estimated + * and real values. For high thresholds and a high number of CPUs, it + * is possible for the min watermark to be breached while the estimated + * value looks fine. The pressure threshold is a reduced value such + * that even the maximum amount of drift will not accidentally breach + * the min watermark + */ + watermark_distance = low_wmark_pages(zone) - min_wmark_pages(zone); + threshold = max(1, (int)(watermark_distance / num_online_cpus())); + + /* + * Maximum threshold is 125 + */ + threshold = min(125, threshold); + + return threshold; +} + static int calculate_threshold(struct zone *zone) { int threshold; @@ -161,6 +185,48 @@ static void refresh_zone_stat_thresholds(void) } } +void reduce_pgdat_percpu_threshold(pg_data_t *pgdat) +{ + struct zone *zone; + int cpu; + int threshold; + int i; + + get_online_cpus(); + for (i = 0; i < pgdat->nr_zones; i++) { + zone = &pgdat->node_zones[i]; + if (!zone->percpu_drift_mark) + continue; + + threshold = calculate_pressure_threshold(zone); + for_each_online_cpu(cpu) + per_cpu_ptr(zone->pageset, cpu)->stat_threshold + = threshold; + } + put_online_cpus(); +} + +void restore_pgdat_percpu_threshold(pg_data_t *pgdat) +{ + struct zone *zone; + int cpu; + int threshold; + int i; + + get_online_cpus(); + for (i = 0; i < pgdat->nr_zones; i++) { + zone = &pgdat->node_zones[i]; + if (!zone->percpu_drift_mark) + continue; + + threshold = calculate_threshold(zone); + for_each_online_cpu(cpu) + per_cpu_ptr(zone->pageset, cpu)->stat_threshold + = threshold; + } + put_online_cpus(); +} + /* * For use when we know that interrupts are disabled. */ @@ -911,7 +977,7 @@ static void zoneinfo_show_print(struct seq_file *m, pg_data_t *pgdat, "\n scanned %lu" "\n spanned %lu" "\n present %lu", - zone_nr_free_pages(zone), + zone_page_state(zone, NR_FREE_PAGES), min_wmark_pages(zone), low_wmark_pages(zone), high_wmark_pages(zone), -- cgit v1.2.3 From b44129b30652c8771db2265939bb8b463724043d Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Thu, 13 Jan 2011 15:45:43 -0800 Subject: mm: vmstat: use a single setter function and callback for adjusting percpu thresholds reduce_pgdat_percpu_threshold() and restore_pgdat_percpu_threshold() exist to adjust the per-cpu vmstat thresholds while kswapd is awake to avoid errors due to counter drift. The functions duplicate some code so this patch replaces them with a single set_pgdat_percpu_threshold() that takes a callback function to calculate the desired threshold as a parameter. [akpm@linux-foundation.org: readability tweak] [kosaki.motohiro@jp.fujitsu.com: set_pgdat_percpu_threshold(): don't use for_each_online_cpu] Signed-off-by: Mel Gorman Reviewed-by: Christoph Lameter Reviewed-by: KAMEZAWA Hiroyuki Signed-off-by: KOSAKI Motohiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/vmstat.h | 10 ++++++---- mm/vmscan.c | 19 +++++++++++++++++-- mm/vmstat.c | 36 +++++++----------------------------- 3 files changed, 30 insertions(+), 35 deletions(-) (limited to 'include/linux') diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h index e4cc21cf5870..833e676d6d92 100644 --- a/include/linux/vmstat.h +++ b/include/linux/vmstat.h @@ -254,8 +254,11 @@ extern void dec_zone_state(struct zone *, enum zone_stat_item); extern void __dec_zone_state(struct zone *, enum zone_stat_item); void refresh_cpu_vm_stats(int); -void reduce_pgdat_percpu_threshold(pg_data_t *pgdat); -void restore_pgdat_percpu_threshold(pg_data_t *pgdat); + +int calculate_pressure_threshold(struct zone *zone); +int calculate_normal_threshold(struct zone *zone); +void set_pgdat_percpu_threshold(pg_data_t *pgdat, + int (*calculate_pressure)(struct zone *)); #else /* CONFIG_SMP */ /* @@ -300,8 +303,7 @@ static inline void __dec_zone_page_state(struct page *page, #define dec_zone_page_state __dec_zone_page_state #define mod_zone_page_state __mod_zone_page_state -static inline void reduce_pgdat_percpu_threshold(pg_data_t *pgdat) { } -static inline void restore_pgdat_percpu_threshold(pg_data_t *pgdat) { } +#define set_pgdat_percpu_threshold(pgdat, callback) { } static inline void refresh_cpu_vm_stats(int cpu) { } #endif diff --git a/mm/vmscan.c b/mm/vmscan.c index 5da4295e7d67..86f8c3418795 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -2448,9 +2448,24 @@ static int kswapd(void *p) */ if (!sleeping_prematurely(pgdat, order, remaining)) { trace_mm_vmscan_kswapd_sleep(pgdat->node_id); - restore_pgdat_percpu_threshold(pgdat); + + /* + * vmstat counters are not perfectly + * accurate and the estimated value + * for counters such as NR_FREE_PAGES + * can deviate from the true value by + * nr_online_cpus * threshold. To + * avoid the zone watermarks being + * breached while under pressure, we + * reduce the per-cpu vmstat threshold + * while kswapd is awake and restore + * them before going back to sleep. + */ + set_pgdat_percpu_threshold(pgdat, + calculate_normal_threshold); schedule(); - reduce_pgdat_percpu_threshold(pgdat); + set_pgdat_percpu_threshold(pgdat, + calculate_pressure_threshold); } else { if (remaining) count_vm_event(KSWAPD_LOW_WMARK_HIT_QUICKLY); diff --git a/mm/vmstat.c b/mm/vmstat.c index bc0f095791b4..751a65e00aac 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -83,7 +83,7 @@ EXPORT_SYMBOL(vm_stat); #ifdef CONFIG_SMP -static int calculate_pressure_threshold(struct zone *zone) +int calculate_pressure_threshold(struct zone *zone) { int threshold; int watermark_distance; @@ -107,7 +107,7 @@ static int calculate_pressure_threshold(struct zone *zone) return threshold; } -static int calculate_threshold(struct zone *zone) +int calculate_normal_threshold(struct zone *zone) { int threshold; int mem; /* memory in 128 MB units */ @@ -166,7 +166,7 @@ static void refresh_zone_stat_thresholds(void) for_each_populated_zone(zone) { unsigned long max_drift, tolerate_drift; - threshold = calculate_threshold(zone); + threshold = calculate_normal_threshold(zone); for_each_online_cpu(cpu) per_cpu_ptr(zone->pageset, cpu)->stat_threshold @@ -185,46 +185,24 @@ static void refresh_zone_stat_thresholds(void) } } -void reduce_pgdat_percpu_threshold(pg_data_t *pgdat) +void set_pgdat_percpu_threshold(pg_data_t *pgdat, + int (*calculate_pressure)(struct zone *)) { struct zone *zone; int cpu; int threshold; int i; - get_online_cpus(); - for (i = 0; i < pgdat->nr_zones; i++) { - zone = &pgdat->node_zones[i]; - if (!zone->percpu_drift_mark) - continue; - - threshold = calculate_pressure_threshold(zone); - for_each_online_cpu(cpu) - per_cpu_ptr(zone->pageset, cpu)->stat_threshold - = threshold; - } - put_online_cpus(); -} - -void restore_pgdat_percpu_threshold(pg_data_t *pgdat) -{ - struct zone *zone; - int cpu; - int threshold; - int i; - - get_online_cpus(); for (i = 0; i < pgdat->nr_zones; i++) { zone = &pgdat->node_zones[i]; if (!zone->percpu_drift_mark) continue; - threshold = calculate_threshold(zone); - for_each_online_cpu(cpu) + threshold = (*calculate_pressure)(zone); + for_each_possible_cpu(cpu) per_cpu_ptr(zone->pageset, cpu)->stat_threshold = threshold; } - put_online_cpus(); } /* -- cgit v1.2.3 From 3e7d344970673c5334cf7b5bb27c8c0942b06126 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Thu, 13 Jan 2011 15:45:56 -0800 Subject: mm: vmscan: reclaim order-0 and use compaction instead of lumpy reclaim Lumpy reclaim is disruptive. It reclaims a large number of pages and ignores the age of the pages it reclaims. This can incur significant stalls and potentially increase the number of major faults. Compaction has reached the point where it is considered reasonably stable (meaning it has passed a lot of testing) and is a potential candidate for displacing lumpy reclaim. This patch introduces an alternative to lumpy reclaim whe compaction is available called reclaim/compaction. The basic operation is very simple - instead of selecting a contiguous range of pages to reclaim, a number of order-0 pages are reclaimed and then compaction is later by either kswapd (compact_zone_order()) or direct compaction (__alloc_pages_direct_compact()). [akpm@linux-foundation.org: fix build] [akpm@linux-foundation.org: use conventional task_struct naming] Signed-off-by: Mel Gorman Cc: Andrea Arcangeli Cc: KOSAKI Motohiro Cc: Rik van Riel Acked-by: Johannes Weiner Cc: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/compaction.h | 14 +++++++ include/linux/kernel.h | 7 ++++ mm/compaction.c | 89 ++++++++++++++++++++++++--------------- mm/migrate.c | 17 ++++++++ mm/page_alloc.c | 16 +++++++ mm/vmscan.c | 102 ++++++++++++++++++++++++++++++++++++++------- 6 files changed, 196 insertions(+), 49 deletions(-) (limited to 'include/linux') diff --git a/include/linux/compaction.h b/include/linux/compaction.h index 5ac51552d908..2592883d862d 100644 --- a/include/linux/compaction.h +++ b/include/linux/compaction.h @@ -22,6 +22,9 @@ extern int sysctl_extfrag_handler(struct ctl_table *table, int write, extern int fragmentation_index(struct zone *zone, unsigned int order); extern unsigned long try_to_compact_pages(struct zonelist *zonelist, int order, gfp_t gfp_mask, nodemask_t *mask); +extern unsigned long compaction_suitable(struct zone *zone, int order); +extern unsigned long compact_zone_order(struct zone *zone, int order, + gfp_t gfp_mask); /* Do not skip compaction more than 64 times */ #define COMPACT_MAX_DEFER_SHIFT 6 @@ -59,6 +62,17 @@ static inline unsigned long try_to_compact_pages(struct zonelist *zonelist, return COMPACT_CONTINUE; } +static inline unsigned long compaction_suitable(struct zone *zone, int order) +{ + return COMPACT_SKIPPED; +} + +static inline unsigned long compact_zone_order(struct zone *zone, int order, + gfp_t gfp_mask) +{ + return 0; +} + static inline void defer_compaction(struct zone *zone) { } diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 57dac7022b63..5a9d9059520b 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -600,6 +600,13 @@ struct sysinfo { #define NUMA_BUILD 0 #endif +/* This helps us avoid #ifdef CONFIG_COMPACTION */ +#ifdef CONFIG_COMPACTION +#define COMPACTION_BUILD 1 +#else +#define COMPACTION_BUILD 0 +#endif + /* Rebuild everything on CONFIG_FTRACE_MCOUNT_RECORD */ #ifdef CONFIG_FTRACE_MCOUNT_RECORD # define REBUILD_DUE_TO_FTRACE_MCOUNT_RECORD diff --git a/mm/compaction.c b/mm/compaction.c index 20011a850fef..8fe917ec7c11 100644 --- a/mm/compaction.c +++ b/mm/compaction.c @@ -384,10 +384,62 @@ static int compact_finished(struct zone *zone, return COMPACT_CONTINUE; } +/* + * compaction_suitable: Is this suitable to run compaction on this zone now? + * Returns + * COMPACT_SKIPPED - If there are too few free pages for compaction + * COMPACT_PARTIAL - If the allocation would succeed without compaction + * COMPACT_CONTINUE - If compaction should run now + */ +unsigned long compaction_suitable(struct zone *zone, int order) +{ + int fragindex; + unsigned long watermark; + + /* + * Watermarks for order-0 must be met for compaction. Note the 2UL. + * This is because during migration, copies of pages need to be + * allocated and for a short time, the footprint is higher + */ + watermark = low_wmark_pages(zone) + (2UL << order); + if (!zone_watermark_ok(zone, 0, watermark, 0, 0)) + return COMPACT_SKIPPED; + + /* + * fragmentation index determines if allocation failures are due to + * low memory or external fragmentation + * + * index of -1 implies allocations might succeed dependingon watermarks + * index towards 0 implies failure is due to lack of memory + * index towards 1000 implies failure is due to fragmentation + * + * Only compact if a failure would be due to fragmentation. + */ + fragindex = fragmentation_index(zone, order); + if (fragindex >= 0 && fragindex <= sysctl_extfrag_threshold) + return COMPACT_SKIPPED; + + if (fragindex == -1 && zone_watermark_ok(zone, order, watermark, 0, 0)) + return COMPACT_PARTIAL; + + return COMPACT_CONTINUE; +} + static int compact_zone(struct zone *zone, struct compact_control *cc) { int ret; + ret = compaction_suitable(zone, cc->order); + switch (ret) { + case COMPACT_PARTIAL: + case COMPACT_SKIPPED: + /* Compaction is likely to fail */ + return ret; + case COMPACT_CONTINUE: + /* Fall through to compaction */ + ; + } + /* Setup to move all movable pages to the end of the zone */ cc->migrate_pfn = zone->zone_start_pfn; cc->free_pfn = cc->migrate_pfn + zone->spanned_pages; @@ -429,7 +481,7 @@ static int compact_zone(struct zone *zone, struct compact_control *cc) return ret; } -static unsigned long compact_zone_order(struct zone *zone, +unsigned long compact_zone_order(struct zone *zone, int order, gfp_t gfp_mask) { struct compact_control cc = { @@ -462,7 +514,6 @@ unsigned long try_to_compact_pages(struct zonelist *zonelist, enum zone_type high_zoneidx = gfp_zone(gfp_mask); int may_enter_fs = gfp_mask & __GFP_FS; int may_perform_io = gfp_mask & __GFP_IO; - unsigned long watermark; struct zoneref *z; struct zone *zone; int rc = COMPACT_SKIPPED; @@ -480,43 +531,13 @@ unsigned long try_to_compact_pages(struct zonelist *zonelist, /* Compact each zone in the list */ for_each_zone_zonelist_nodemask(zone, z, zonelist, high_zoneidx, nodemask) { - int fragindex; int status; - /* - * Watermarks for order-0 must be met for compaction. Note - * the 2UL. This is because during migration, copies of - * pages need to be allocated and for a short time, the - * footprint is higher - */ - watermark = low_wmark_pages(zone) + (2UL << order); - if (!zone_watermark_ok(zone, 0, watermark, 0, 0)) - continue; - - /* - * fragmentation index determines if allocation failures are - * due to low memory or external fragmentation - * - * index of -1 implies allocations might succeed depending - * on watermarks - * index towards 0 implies failure is due to lack of memory - * index towards 1000 implies failure is due to fragmentation - * - * Only compact if a failure would be due to fragmentation. - */ - fragindex = fragmentation_index(zone, order); - if (fragindex >= 0 && fragindex <= sysctl_extfrag_threshold) - continue; - - if (fragindex == -1 && zone_watermark_ok(zone, order, watermark, 0, 0)) { - rc = COMPACT_PARTIAL; - break; - } - status = compact_zone_order(zone, order, gfp_mask); rc = max(status, rc); - if (zone_watermark_ok(zone, order, watermark, 0, 0)) + /* If a normal allocation would succeed, stop compacting */ + if (zone_watermark_ok(zone, order, low_wmark_pages(zone), 0, 0)) break; } diff --git a/mm/migrate.c b/mm/migrate.c index 6ae8a66a7045..94875b265928 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -639,6 +639,23 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private, if (!trylock_page(page)) { if (!force) goto move_newpage; + + /* + * It's not safe for direct compaction to call lock_page. + * For example, during page readahead pages are added locked + * to the LRU. Later, when the IO completes the pages are + * marked uptodate and unlocked. However, the queueing + * could be merging multiple pages for one bio (e.g. + * mpage_readpages). If an allocation happens for the + * second or third page, the process can end up locking + * the same page twice and deadlocking. Rather than + * trying to be clever about what pages can be locked, + * avoid the use of lock_page for direct compaction + * altogether. + */ + if (current->flags & PF_MEMALLOC) + goto move_newpage; + lock_page(page); } diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 22a1bb7723e4..03a66a31bfcd 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1815,12 +1815,15 @@ __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order, int migratetype, unsigned long *did_some_progress) { struct page *page; + struct task_struct *tsk = current; if (!order || compaction_deferred(preferred_zone)) return NULL; + tsk->flags |= PF_MEMALLOC; *did_some_progress = try_to_compact_pages(zonelist, order, gfp_mask, nodemask); + tsk->flags &= ~PF_MEMALLOC; if (*did_some_progress != COMPACT_SKIPPED) { /* Page migration frees to the PCP lists but we want merging */ @@ -2121,6 +2124,19 @@ rebalance: /* Wait for some write requests to complete then retry */ wait_iff_congested(preferred_zone, BLK_RW_ASYNC, HZ/50); goto rebalance; + } else { + /* + * High-order allocations do not necessarily loop after + * direct reclaim and reclaim/compaction depends on compaction + * being called after reclaim so call directly if necessary + */ + page = __alloc_pages_direct_compact(gfp_mask, order, + zonelist, high_zoneidx, + nodemask, + alloc_flags, preferred_zone, + migratetype, &did_some_progress); + if (page) + goto got_pg; } nopage: diff --git a/mm/vmscan.c b/mm/vmscan.c index 3464312bde07..10ebd74a423c 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -32,6 +32,7 @@ #include #include #include +#include #include #include #include @@ -59,12 +60,15 @@ * LUMPY_MODE_CONTIGRECLAIM: For high-order allocations, take a reference * page from the LRU and reclaim all pages within a * naturally aligned range + * LUMPY_MODE_COMPACTION: For high-order allocations, reclaim a number of + * order-0 pages and then compact the zone */ typedef unsigned __bitwise__ lumpy_mode; #define LUMPY_MODE_SINGLE ((__force lumpy_mode)0x01u) #define LUMPY_MODE_ASYNC ((__force lumpy_mode)0x02u) #define LUMPY_MODE_SYNC ((__force lumpy_mode)0x04u) #define LUMPY_MODE_CONTIGRECLAIM ((__force lumpy_mode)0x08u) +#define LUMPY_MODE_COMPACTION ((__force lumpy_mode)0x10u) struct scan_control { /* Incremented by the number of inactive pages that were scanned */ @@ -286,18 +290,20 @@ static void set_lumpy_reclaim_mode(int priority, struct scan_control *sc, lumpy_mode syncmode = sync ? LUMPY_MODE_SYNC : LUMPY_MODE_ASYNC; /* - * Some reclaim have alredy been failed. No worth to try synchronous - * lumpy reclaim. + * Initially assume we are entering either lumpy reclaim or + * reclaim/compaction.Depending on the order, we will either set the + * sync mode or just reclaim order-0 pages later. */ - if (sync && sc->lumpy_reclaim_mode & LUMPY_MODE_SINGLE) - return; + if (COMPACTION_BUILD) + sc->lumpy_reclaim_mode = LUMPY_MODE_COMPACTION; + else + sc->lumpy_reclaim_mode = LUMPY_MODE_CONTIGRECLAIM; /* - * If we need a large contiguous chunk of memory, or have - * trouble getting a small set of contiguous pages, we - * will reclaim both active and inactive pages. + * Avoid using lumpy reclaim or reclaim/compaction if possible by + * restricting when its set to either costly allocations or when + * under memory pressure */ - sc->lumpy_reclaim_mode = LUMPY_MODE_CONTIGRECLAIM; if (sc->order > PAGE_ALLOC_COSTLY_ORDER) sc->lumpy_reclaim_mode |= syncmode; else if (sc->order && priority < DEF_PRIORITY - 2) @@ -1385,8 +1391,8 @@ shrink_inactive_list(unsigned long nr_to_scan, struct zone *zone, if (scanning_global_lru(sc)) { nr_taken = isolate_pages_global(nr_to_scan, &page_list, &nr_scanned, sc->order, - sc->lumpy_reclaim_mode & LUMPY_MODE_SINGLE ? - ISOLATE_INACTIVE : ISOLATE_BOTH, + sc->lumpy_reclaim_mode & LUMPY_MODE_CONTIGRECLAIM ? + ISOLATE_BOTH : ISOLATE_INACTIVE, zone, 0, file); zone->pages_scanned += nr_scanned; if (current_is_kswapd()) @@ -1398,8 +1404,8 @@ shrink_inactive_list(unsigned long nr_to_scan, struct zone *zone, } else { nr_taken = mem_cgroup_isolate_pages(nr_to_scan, &page_list, &nr_scanned, sc->order, - sc->lumpy_reclaim_mode & LUMPY_MODE_SINGLE ? - ISOLATE_INACTIVE : ISOLATE_BOTH, + sc->lumpy_reclaim_mode & LUMPY_MODE_CONTIGRECLAIM ? + ISOLATE_BOTH : ISOLATE_INACTIVE, zone, sc->mem_cgroup, 0, file); /* @@ -1814,6 +1820,57 @@ out: } } +/* + * Reclaim/compaction depends on a number of pages being freed. To avoid + * disruption to the system, a small number of order-0 pages continue to be + * rotated and reclaimed in the normal fashion. However, by the time we get + * back to the allocator and call try_to_compact_zone(), we ensure that + * there are enough free pages for it to be likely successful + */ +static inline bool should_continue_reclaim(struct zone *zone, + unsigned long nr_reclaimed, + unsigned long nr_scanned, + struct scan_control *sc) +{ + unsigned long pages_for_compaction; + unsigned long inactive_lru_pages; + + /* If not in reclaim/compaction mode, stop */ + if (!(sc->lumpy_reclaim_mode & LUMPY_MODE_COMPACTION)) + return false; + + /* + * If we failed to reclaim and have scanned the full list, stop. + * NOTE: Checking just nr_reclaimed would exit reclaim/compaction far + * faster but obviously would be less likely to succeed + * allocation. If this is desirable, use GFP_REPEAT to decide + * if both reclaimed and scanned should be checked or just + * reclaimed + */ + if (!nr_reclaimed && !nr_scanned) + return false; + + /* + * If we have not reclaimed enough pages for compaction and the + * inactive lists are large enough, continue reclaiming + */ + pages_for_compaction = (2UL << sc->order); + inactive_lru_pages = zone_nr_lru_pages(zone, sc, LRU_INACTIVE_ANON) + + zone_nr_lru_pages(zone, sc, LRU_INACTIVE_FILE); + if (sc->nr_reclaimed < pages_for_compaction && + inactive_lru_pages > pages_for_compaction) + return true; + + /* If compaction would go ahead or the allocation would succeed, stop */ + switch (compaction_suitable(zone, sc->order)) { + case COMPACT_PARTIAL: + case COMPACT_CONTINUE: + return false; + default: + return true; + } +} + /* * This is a basic per-zone page freer. Used by both kswapd and direct reclaim. */ @@ -1823,9 +1880,12 @@ static void shrink_zone(int priority, struct zone *zone, unsigned long nr[NR_LRU_LISTS]; unsigned long nr_to_scan; enum lru_list l; - unsigned long nr_reclaimed = sc->nr_reclaimed; + unsigned long nr_reclaimed; unsigned long nr_to_reclaim = sc->nr_to_reclaim; + unsigned long nr_scanned = sc->nr_scanned; +restart: + nr_reclaimed = 0; get_scan_count(zone, sc, nr, priority); while (nr[LRU_INACTIVE_ANON] || nr[LRU_ACTIVE_FILE] || @@ -1851,8 +1911,7 @@ static void shrink_zone(int priority, struct zone *zone, if (nr_reclaimed >= nr_to_reclaim && priority < DEF_PRIORITY) break; } - - sc->nr_reclaimed = nr_reclaimed; + sc->nr_reclaimed += nr_reclaimed; /* * Even if we did not try to evict anon pages at all, we want to @@ -1861,6 +1920,11 @@ static void shrink_zone(int priority, struct zone *zone, if (inactive_anon_is_low(zone, sc)) shrink_active_list(SWAP_CLUSTER_MAX, zone, sc, priority, 0); + /* reclaim/compaction might need reclaim to continue */ + if (should_continue_reclaim(zone, nr_reclaimed, + sc->nr_scanned - nr_scanned, sc)) + goto restart; + throttle_vm_writeout(sc->gfp_mask); } @@ -2307,6 +2371,14 @@ loop_again: total_scanned > sc.nr_reclaimed + sc.nr_reclaimed / 2) sc.may_writepage = 1; + /* + * Compact the zone for higher orders to reduce + * latencies for higher-order allocations that + * would ordinarily call try_to_compact_pages() + */ + if (sc.order > PAGE_ALLOC_COSTLY_ORDER) + compact_zone_order(zone, sc.order, sc.gfp_mask); + if (!zone_watermark_ok_safe(zone, order, high_wmark_pages(zone), end_zone, 0)) { all_zones_ok = 0; -- cgit v1.2.3 From 77f1fe6b08b13a87391549c8a820ddc817b6f50e Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Thu, 13 Jan 2011 15:45:57 -0800 Subject: mm: migration: allow migration to operate asynchronously and avoid synchronous compaction in the faster path Migration synchronously waits for writeback if the initial passes fails. Callers of memory compaction do not necessarily want this behaviour if the caller is latency sensitive or expects that synchronous migration is not going to have a significantly better success rate. This patch adds a sync parameter to migrate_pages() allowing the caller to indicate if wait_on_page_writeback() is allowed within migration or not. For reclaim/compaction, try_to_compact_pages() is first called asynchronously, direct reclaim runs and then try_to_compact_pages() is called synchronously as there is a greater expectation that it'll succeed. [akpm@linux-foundation.org: build/merge fix] Signed-off-by: Mel Gorman Cc: Andrea Arcangeli Cc: KOSAKI Motohiro Cc: Rik van Riel Acked-by: Johannes Weiner Cc: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/compaction.h | 10 ++++++---- include/linux/migrate.h | 12 ++++++++---- mm/compaction.c | 14 ++++++++++---- mm/memory-failure.c | 8 +++++--- mm/memory_hotplug.c | 3 ++- mm/mempolicy.c | 4 ++-- mm/migrate.c | 22 +++++++++++++--------- mm/page_alloc.c | 21 +++++++++++++++------ mm/vmscan.c | 3 ++- 9 files changed, 63 insertions(+), 34 deletions(-) (limited to 'include/linux') diff --git a/include/linux/compaction.h b/include/linux/compaction.h index 2592883d862d..72cba4034785 100644 --- a/include/linux/compaction.h +++ b/include/linux/compaction.h @@ -21,10 +21,11 @@ extern int sysctl_extfrag_handler(struct ctl_table *table, int write, extern int fragmentation_index(struct zone *zone, unsigned int order); extern unsigned long try_to_compact_pages(struct zonelist *zonelist, - int order, gfp_t gfp_mask, nodemask_t *mask); + int order, gfp_t gfp_mask, nodemask_t *mask, + bool sync); extern unsigned long compaction_suitable(struct zone *zone, int order); extern unsigned long compact_zone_order(struct zone *zone, int order, - gfp_t gfp_mask); + gfp_t gfp_mask, bool sync); /* Do not skip compaction more than 64 times */ #define COMPACT_MAX_DEFER_SHIFT 6 @@ -57,7 +58,8 @@ static inline bool compaction_deferred(struct zone *zone) #else static inline unsigned long try_to_compact_pages(struct zonelist *zonelist, - int order, gfp_t gfp_mask, nodemask_t *nodemask) + int order, gfp_t gfp_mask, nodemask_t *nodemask, + bool sync) { return COMPACT_CONTINUE; } @@ -68,7 +70,7 @@ static inline unsigned long compaction_suitable(struct zone *zone, int order) } static inline unsigned long compact_zone_order(struct zone *zone, int order, - gfp_t gfp_mask) + gfp_t gfp_mask, bool sync) { return 0; } diff --git a/include/linux/migrate.h b/include/linux/migrate.h index 085527fb8261..fa31902803fd 100644 --- a/include/linux/migrate.h +++ b/include/linux/migrate.h @@ -13,9 +13,11 @@ extern void putback_lru_pages(struct list_head *l); extern int migrate_page(struct address_space *, struct page *, struct page *); extern int migrate_pages(struct list_head *l, new_page_t x, - unsigned long private, int offlining); + unsigned long private, int offlining, + bool sync); extern int migrate_huge_pages(struct list_head *l, new_page_t x, - unsigned long private, int offlining); + unsigned long private, int offlining, + bool sync); extern int fail_migrate_page(struct address_space *, struct page *, struct page *); @@ -33,9 +35,11 @@ extern int migrate_huge_page_move_mapping(struct address_space *mapping, static inline void putback_lru_pages(struct list_head *l) {} static inline int migrate_pages(struct list_head *l, new_page_t x, - unsigned long private, int offlining) { return -ENOSYS; } + unsigned long private, int offlining, + bool sync) { return -ENOSYS; } static inline int migrate_huge_pages(struct list_head *l, new_page_t x, - unsigned long private, int offlining) { return -ENOSYS; } + unsigned long private, int offlining, + bool sync) { return -ENOSYS; } static inline int migrate_prep(void) { return -ENOSYS; } static inline int migrate_prep_local(void) { return -ENOSYS; } diff --git a/mm/compaction.c b/mm/compaction.c index 8fe917ec7c11..47fca1069343 100644 --- a/mm/compaction.c +++ b/mm/compaction.c @@ -33,6 +33,7 @@ struct compact_control { unsigned long nr_migratepages; /* Number of pages to migrate */ unsigned long free_pfn; /* isolate_freepages search base */ unsigned long migrate_pfn; /* isolate_migratepages search base */ + bool sync; /* Synchronous migration */ /* Account for isolated anon and file pages */ unsigned long nr_anon; @@ -455,7 +456,8 @@ static int compact_zone(struct zone *zone, struct compact_control *cc) nr_migrate = cc->nr_migratepages; migrate_pages(&cc->migratepages, compaction_alloc, - (unsigned long)cc, 0); + (unsigned long)cc, 0, + cc->sync); update_nr_listpages(cc); nr_remaining = cc->nr_migratepages; @@ -482,7 +484,8 @@ static int compact_zone(struct zone *zone, struct compact_control *cc) } unsigned long compact_zone_order(struct zone *zone, - int order, gfp_t gfp_mask) + int order, gfp_t gfp_mask, + bool sync) { struct compact_control cc = { .nr_freepages = 0, @@ -490,6 +493,7 @@ unsigned long compact_zone_order(struct zone *zone, .order = order, .migratetype = allocflags_to_migratetype(gfp_mask), .zone = zone, + .sync = sync, }; INIT_LIST_HEAD(&cc.freepages); INIT_LIST_HEAD(&cc.migratepages); @@ -505,11 +509,13 @@ int sysctl_extfrag_threshold = 500; * @order: The order of the current allocation * @gfp_mask: The GFP mask of the current allocation * @nodemask: The allowed nodes to allocate from + * @sync: Whether migration is synchronous or not * * This is the main entry point for direct page compaction. */ unsigned long try_to_compact_pages(struct zonelist *zonelist, - int order, gfp_t gfp_mask, nodemask_t *nodemask) + int order, gfp_t gfp_mask, nodemask_t *nodemask, + bool sync) { enum zone_type high_zoneidx = gfp_zone(gfp_mask); int may_enter_fs = gfp_mask & __GFP_FS; @@ -533,7 +539,7 @@ unsigned long try_to_compact_pages(struct zonelist *zonelist, nodemask) { int status; - status = compact_zone_order(zone, order, gfp_mask); + status = compact_zone_order(zone, order, gfp_mask, sync); rc = max(status, rc); /* If a normal allocation would succeed, stop compacting */ diff --git a/mm/memory-failure.c b/mm/memory-failure.c index 46ab2c044b0e..2323a8039a98 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -1290,9 +1290,10 @@ static int soft_offline_huge_page(struct page *page, int flags) /* Keep page count to indicate a given hugepage is isolated. */ list_add(&hpage->lru, &pagelist); - ret = migrate_huge_pages(&pagelist, new_page, MPOL_MF_MOVE_ALL, 0); + ret = migrate_huge_pages(&pagelist, new_page, MPOL_MF_MOVE_ALL, 0, + true); if (ret) { - putback_lru_pages(&pagelist); + putback_lru_pages(&pagelist); pr_debug("soft offline: %#lx: migration failed %d, type %lx\n", pfn, ret, page->flags); if (ret > 0) @@ -1413,7 +1414,8 @@ int soft_offline_page(struct page *page, int flags) LIST_HEAD(pagelist); list_add(&page->lru, &pagelist); - ret = migrate_pages(&pagelist, new_page, MPOL_MF_MOVE_ALL, 0); + ret = migrate_pages(&pagelist, new_page, MPOL_MF_MOVE_ALL, + 0, true); if (ret) { pr_info("soft offline: %#lx: migration failed %d, type %lx\n", pfn, ret, page->flags); diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index 2c6523af5473..584fc5588fdd 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -733,7 +733,8 @@ do_migrate_range(unsigned long start_pfn, unsigned long end_pfn) goto out; } /* this function returns # of failed pages */ - ret = migrate_pages(&source, hotremove_migrate_alloc, 0, 1); + ret = migrate_pages(&source, hotremove_migrate_alloc, 0, + 1, true); if (ret) putback_lru_pages(&source); } diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 11ff260fb282..9db274593086 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -935,7 +935,7 @@ static int migrate_to_node(struct mm_struct *mm, int source, int dest, return PTR_ERR(vma); if (!list_empty(&pagelist)) { - err = migrate_pages(&pagelist, new_node_page, dest, 0); + err = migrate_pages(&pagelist, new_node_page, dest, 0, true); if (err) putback_lru_pages(&pagelist); } @@ -1155,7 +1155,7 @@ static long do_mbind(unsigned long start, unsigned long len, if (!list_empty(&pagelist)) { nr_failed = migrate_pages(&pagelist, new_vma_page, - (unsigned long)vma, 0); + (unsigned long)vma, 0, true); if (nr_failed) putback_lru_pages(&pagelist); } diff --git a/mm/migrate.c b/mm/migrate.c index 94875b265928..dc47f6c40353 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -614,7 +614,7 @@ static int move_to_new_page(struct page *newpage, struct page *page, * to the newly allocated page in newpage. */ static int unmap_and_move(new_page_t get_new_page, unsigned long private, - struct page *page, int force, int offlining) + struct page *page, int force, int offlining, bool sync) { int rc = 0; int *result = NULL; @@ -682,7 +682,7 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private, BUG_ON(charge); if (PageWriteback(page)) { - if (!force) + if (!force || !sync) goto uncharge; wait_on_page_writeback(page); } @@ -827,7 +827,7 @@ move_newpage: */ static int unmap_and_move_huge_page(new_page_t get_new_page, unsigned long private, struct page *hpage, - int force, int offlining) + int force, int offlining, bool sync) { int rc = 0; int *result = NULL; @@ -841,7 +841,7 @@ static int unmap_and_move_huge_page(new_page_t get_new_page, rc = -EAGAIN; if (!trylock_page(hpage)) { - if (!force) + if (!force || !sync) goto out; lock_page(hpage); } @@ -909,7 +909,8 @@ out: * Return: Number of pages not migrated or error code. */ int migrate_pages(struct list_head *from, - new_page_t get_new_page, unsigned long private, int offlining) + new_page_t get_new_page, unsigned long private, int offlining, + bool sync) { int retry = 1; int nr_failed = 0; @@ -929,7 +930,8 @@ int migrate_pages(struct list_head *from, cond_resched(); rc = unmap_and_move(get_new_page, private, - page, pass > 2, offlining); + page, pass > 2, offlining, + sync); switch(rc) { case -ENOMEM: @@ -958,7 +960,8 @@ out: } int migrate_huge_pages(struct list_head *from, - new_page_t get_new_page, unsigned long private, int offlining) + new_page_t get_new_page, unsigned long private, int offlining, + bool sync) { int retry = 1; int nr_failed = 0; @@ -974,7 +977,8 @@ int migrate_huge_pages(struct list_head *from, cond_resched(); rc = unmap_and_move_huge_page(get_new_page, - private, page, pass > 2, offlining); + private, page, pass > 2, offlining, + sync); switch(rc) { case -ENOMEM: @@ -1107,7 +1111,7 @@ set_status: err = 0; if (!list_empty(&pagelist)) { err = migrate_pages(&pagelist, new_page_node, - (unsigned long)pm, 0); + (unsigned long)pm, 0, true); if (err) putback_lru_pages(&pagelist); } diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 03a66a31bfcd..0fd486467b4b 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1812,7 +1812,8 @@ static struct page * __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order, struct zonelist *zonelist, enum zone_type high_zoneidx, nodemask_t *nodemask, int alloc_flags, struct zone *preferred_zone, - int migratetype, unsigned long *did_some_progress) + int migratetype, unsigned long *did_some_progress, + bool sync_migration) { struct page *page; struct task_struct *tsk = current; @@ -1822,7 +1823,7 @@ __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order, tsk->flags |= PF_MEMALLOC; *did_some_progress = try_to_compact_pages(zonelist, order, gfp_mask, - nodemask); + nodemask, sync_migration); tsk->flags &= ~PF_MEMALLOC; if (*did_some_progress != COMPACT_SKIPPED) { @@ -1859,7 +1860,8 @@ static inline struct page * __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order, struct zonelist *zonelist, enum zone_type high_zoneidx, nodemask_t *nodemask, int alloc_flags, struct zone *preferred_zone, - int migratetype, unsigned long *did_some_progress) + int migratetype, unsigned long *did_some_progress, + bool sync_migration) { return NULL; } @@ -2001,6 +2003,7 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order, unsigned long pages_reclaimed = 0; unsigned long did_some_progress; struct task_struct *p = current; + bool sync_migration = false; /* * In the slowpath, we sanity check order to avoid ever trying to @@ -2063,14 +2066,19 @@ rebalance: if (test_thread_flag(TIF_MEMDIE) && !(gfp_mask & __GFP_NOFAIL)) goto nopage; - /* Try direct compaction */ + /* + * Try direct compaction. The first pass is asynchronous. Subsequent + * attempts after direct reclaim are synchronous + */ page = __alloc_pages_direct_compact(gfp_mask, order, zonelist, high_zoneidx, nodemask, alloc_flags, preferred_zone, - migratetype, &did_some_progress); + migratetype, &did_some_progress, + sync_migration); if (page) goto got_pg; + sync_migration = true; /* Try direct reclaim and then allocating */ page = __alloc_pages_direct_reclaim(gfp_mask, order, @@ -2134,7 +2142,8 @@ rebalance: zonelist, high_zoneidx, nodemask, alloc_flags, preferred_zone, - migratetype, &did_some_progress); + migratetype, &did_some_progress, + sync_migration); if (page) goto got_pg; } diff --git a/mm/vmscan.c b/mm/vmscan.c index 10ebd74a423c..8320d115c85d 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -2377,7 +2377,8 @@ loop_again: * would ordinarily call try_to_compact_pages() */ if (sc.order > PAGE_ALLOC_COSTLY_ORDER) - compact_zone_order(zone, sc.order, sc.gfp_mask); + compact_zone_order(zone, sc.order, sc.gfp_mask, + false); if (!zone_watermark_ok_safe(zone, order, high_wmark_pages(zone), end_zone, 0)) { -- cgit v1.2.3 From 7f0f24967b0349798803260b2e4bf347cffa1990 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Thu, 13 Jan 2011 15:45:58 -0800 Subject: mm: migration: cleanup migrate_pages API by matching types for offlining and sync With the introduction of the boolean sync parameter, the API looks a little inconsistent as offlining is still an int. Convert offlining to a bool for the sake of being tidy. Signed-off-by: Mel Gorman Cc: Andrea Arcangeli Cc: KOSAKI Motohiro Cc: Rik van Riel Acked-by: Johannes Weiner Cc: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/migrate.h | 8 ++++---- mm/compaction.c | 2 +- mm/memory_hotplug.c | 2 +- mm/mempolicy.c | 6 ++++-- mm/migrate.c | 8 ++++---- 5 files changed, 14 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/migrate.h b/include/linux/migrate.h index fa31902803fd..e39aeecfe9a2 100644 --- a/include/linux/migrate.h +++ b/include/linux/migrate.h @@ -13,10 +13,10 @@ extern void putback_lru_pages(struct list_head *l); extern int migrate_page(struct address_space *, struct page *, struct page *); extern int migrate_pages(struct list_head *l, new_page_t x, - unsigned long private, int offlining, + unsigned long private, bool offlining, bool sync); extern int migrate_huge_pages(struct list_head *l, new_page_t x, - unsigned long private, int offlining, + unsigned long private, bool offlining, bool sync); extern int fail_migrate_page(struct address_space *, @@ -35,10 +35,10 @@ extern int migrate_huge_page_move_mapping(struct address_space *mapping, static inline void putback_lru_pages(struct list_head *l) {} static inline int migrate_pages(struct list_head *l, new_page_t x, - unsigned long private, int offlining, + unsigned long private, bool offlining, bool sync) { return -ENOSYS; } static inline int migrate_huge_pages(struct list_head *l, new_page_t x, - unsigned long private, int offlining, + unsigned long private, bool offlining, bool sync) { return -ENOSYS; } static inline int migrate_prep(void) { return -ENOSYS; } diff --git a/mm/compaction.c b/mm/compaction.c index 47fca1069343..e005a30e968c 100644 --- a/mm/compaction.c +++ b/mm/compaction.c @@ -456,7 +456,7 @@ static int compact_zone(struct zone *zone, struct compact_control *cc) nr_migrate = cc->nr_migratepages; migrate_pages(&cc->migratepages, compaction_alloc, - (unsigned long)cc, 0, + (unsigned long)cc, false, cc->sync); update_nr_listpages(cc); nr_remaining = cc->nr_migratepages; diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index 584fc5588fdd..a2832c092509 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -734,7 +734,7 @@ do_migrate_range(unsigned long start_pfn, unsigned long end_pfn) } /* this function returns # of failed pages */ ret = migrate_pages(&source, hotremove_migrate_alloc, 0, - 1, true); + true, true); if (ret) putback_lru_pages(&source); } diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 9db274593086..7ee55af8d79c 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -935,7 +935,8 @@ static int migrate_to_node(struct mm_struct *mm, int source, int dest, return PTR_ERR(vma); if (!list_empty(&pagelist)) { - err = migrate_pages(&pagelist, new_node_page, dest, 0, true); + err = migrate_pages(&pagelist, new_node_page, dest, + false, true); if (err) putback_lru_pages(&pagelist); } @@ -1155,7 +1156,8 @@ static long do_mbind(unsigned long start, unsigned long len, if (!list_empty(&pagelist)) { nr_failed = migrate_pages(&pagelist, new_vma_page, - (unsigned long)vma, 0, true); + (unsigned long)vma, + false, true); if (nr_failed) putback_lru_pages(&pagelist); } diff --git a/mm/migrate.c b/mm/migrate.c index dc47f6c40353..690d0de993af 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -614,7 +614,7 @@ static int move_to_new_page(struct page *newpage, struct page *page, * to the newly allocated page in newpage. */ static int unmap_and_move(new_page_t get_new_page, unsigned long private, - struct page *page, int force, int offlining, bool sync) + struct page *page, int force, bool offlining, bool sync) { int rc = 0; int *result = NULL; @@ -827,7 +827,7 @@ move_newpage: */ static int unmap_and_move_huge_page(new_page_t get_new_page, unsigned long private, struct page *hpage, - int force, int offlining, bool sync) + int force, bool offlining, bool sync) { int rc = 0; int *result = NULL; @@ -909,7 +909,7 @@ out: * Return: Number of pages not migrated or error code. */ int migrate_pages(struct list_head *from, - new_page_t get_new_page, unsigned long private, int offlining, + new_page_t get_new_page, unsigned long private, bool offlining, bool sync) { int retry = 1; @@ -960,7 +960,7 @@ out: } int migrate_huge_pages(struct list_head *from, - new_page_t get_new_page, unsigned long private, int offlining, + new_page_t get_new_page, unsigned long private, bool offlining, bool sync) { int retry = 1; -- cgit v1.2.3 From e5a5623b28198aa91ea71ee5d3846757fc76bc87 Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Thu, 13 Jan 2011 15:46:00 -0800 Subject: mm: remove unused get_vm_area_node get_vm_area_node() is unused in the kernel and can thus be removed. Signed-off-by: David Rientjes Cc: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/vmalloc.h | 3 --- mm/vmalloc.c | 7 ------- 2 files changed, 10 deletions(-) (limited to 'include/linux') diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h index 44b54f619ac6..cb73c755fac8 100644 --- a/include/linux/vmalloc.h +++ b/include/linux/vmalloc.h @@ -90,9 +90,6 @@ extern struct vm_struct *__get_vm_area_caller(unsigned long size, unsigned long flags, unsigned long start, unsigned long end, void *caller); -extern struct vm_struct *get_vm_area_node(unsigned long size, - unsigned long flags, int node, - gfp_t gfp_mask); extern struct vm_struct *remove_vm_area(const void *addr); extern int map_vm_area(struct vm_struct *area, pgprot_t prot, diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 48245af07b10..78ec9d8bc57c 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -1315,13 +1315,6 @@ struct vm_struct *get_vm_area_caller(unsigned long size, unsigned long flags, -1, GFP_KERNEL, caller); } -struct vm_struct *get_vm_area_node(unsigned long size, unsigned long flags, - int node, gfp_t gfp_mask) -{ - return __get_vm_area_node(size, 1, flags, VMALLOC_START, VMALLOC_END, - node, gfp_mask, __builtin_return_address(0)); -} - static struct vm_struct *find_vm_area(const void *addr) { struct vmap_area *va; -- cgit v1.2.3 From ec3f64fc9c196a304c4b7db3e1ff56d640628509 Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Thu, 13 Jan 2011 15:46:01 -0800 Subject: mm: remove gfp mask from pcpu_get_vm_areas pcpu_get_vm_areas() only uses GFP_KERNEL allocations, so remove the gfp_t formal and use the mask internally. Signed-off-by: David Rientjes Cc: Christoph Lameter Cc: Tejun Heo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/vmalloc.h | 2 +- mm/percpu-vm.c | 2 +- mm/vmalloc.c | 21 +++++++++------------ 3 files changed, 11 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h index cb73c755fac8..c7348b8d0a81 100644 --- a/include/linux/vmalloc.h +++ b/include/linux/vmalloc.h @@ -117,7 +117,7 @@ extern __init void vm_area_register_early(struct vm_struct *vm, size_t align); #ifdef CONFIG_SMP struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets, const size_t *sizes, int nr_vms, - size_t align, gfp_t gfp_mask); + size_t align); void pcpu_free_vm_areas(struct vm_struct **vms, int nr_vms); #endif diff --git a/mm/percpu-vm.c b/mm/percpu-vm.c index 7d9c1d0ebd3f..ea534960a04b 100644 --- a/mm/percpu-vm.c +++ b/mm/percpu-vm.c @@ -421,7 +421,7 @@ static struct pcpu_chunk *pcpu_create_chunk(void) return NULL; vms = pcpu_get_vm_areas(pcpu_group_offsets, pcpu_group_sizes, - pcpu_nr_groups, pcpu_atom_size, GFP_KERNEL); + pcpu_nr_groups, pcpu_atom_size); if (!vms) { pcpu_free_chunk(chunk); return NULL; diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 78ec9d8bc57c..f67546636322 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -2196,17 +2196,16 @@ static unsigned long pvm_determine_end(struct vmap_area **pnext, * @sizes: array containing size of each area * @nr_vms: the number of areas to allocate * @align: alignment, all entries in @offsets and @sizes must be aligned to this - * @gfp_mask: allocation mask * * Returns: kmalloc'd vm_struct pointer array pointing to allocated * vm_structs on success, %NULL on failure * * Percpu allocator wants to use congruent vm areas so that it can * maintain the offsets among percpu areas. This function allocates - * congruent vmalloc areas for it. These areas tend to be scattered - * pretty far, distance between two areas easily going up to - * gigabytes. To avoid interacting with regular vmallocs, these areas - * are allocated from top. + * congruent vmalloc areas for it with GFP_KERNEL. These areas tend to + * be scattered pretty far, distance between two areas easily going up + * to gigabytes. To avoid interacting with regular vmallocs, these + * areas are allocated from top. * * Despite its complicated look, this allocator is rather simple. It * does everything top-down and scans areas from the end looking for @@ -2217,7 +2216,7 @@ static unsigned long pvm_determine_end(struct vmap_area **pnext, */ struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets, const size_t *sizes, int nr_vms, - size_t align, gfp_t gfp_mask) + size_t align) { const unsigned long vmalloc_start = ALIGN(VMALLOC_START, align); const unsigned long vmalloc_end = VMALLOC_END & ~(align - 1); @@ -2227,8 +2226,6 @@ struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets, unsigned long base, start, end, last_end; bool purged = false; - gfp_mask &= GFP_RECLAIM_MASK; - /* verify parameters and allocate data structures */ BUG_ON(align & ~PAGE_MASK || !is_power_of_2(align)); for (last_area = 0, area = 0; area < nr_vms; area++) { @@ -2261,14 +2258,14 @@ struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets, return NULL; } - vms = kzalloc(sizeof(vms[0]) * nr_vms, gfp_mask); - vas = kzalloc(sizeof(vas[0]) * nr_vms, gfp_mask); + vms = kzalloc(sizeof(vms[0]) * nr_vms, GFP_KERNEL); + vas = kzalloc(sizeof(vas[0]) * nr_vms, GFP_KERNEL); if (!vas || !vms) goto err_free; for (area = 0; area < nr_vms; area++) { - vas[area] = kzalloc(sizeof(struct vmap_area), gfp_mask); - vms[area] = kzalloc(sizeof(struct vm_struct), gfp_mask); + vas[area] = kzalloc(sizeof(struct vmap_area), GFP_KERNEL); + vms[area] = kzalloc(sizeof(struct vm_struct), GFP_KERNEL); if (!vas[area] || !vms[area]) goto err_free; } -- cgit v1.2.3 From d0a21265dfb5fa8ae54e90d0fb6d1c215b10a28a Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Thu, 13 Jan 2011 15:46:02 -0800 Subject: mm: unify module_alloc code for vmalloc Four architectures (arm, mips, sparc, x86) use __vmalloc_area() for module_init(). Much of the code is duplicated and can be generalized in a globally accessible function, __vmalloc_node_range(). __vmalloc_node() now calls into __vmalloc_node_range() with a range of [VMALLOC_START, VMALLOC_END) for functionally equivalent behavior. Each architecture may then use __vmalloc_node_range() directly to remove the duplication of code. Signed-off-by: David Rientjes Cc: Christoph Lameter Cc: Russell King Cc: Ralf Baechle Cc: "David S. Miller" Cc: Ingo Molnar Cc: "H. Peter Anvin" Cc: Thomas Gleixner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/kernel/module.c | 14 +++---------- arch/mips/kernel/module.c | 14 +++---------- arch/sparc/kernel/module.c | 14 ++++--------- arch/x86/kernel/module.c | 17 ++++------------ include/linux/vmalloc.h | 5 +++-- mm/vmalloc.c | 50 +++++++++++++++++++++++++++------------------- 6 files changed, 46 insertions(+), 68 deletions(-) (limited to 'include/linux') diff --git a/arch/arm/kernel/module.c b/arch/arm/kernel/module.c index 0c1bb68ff4a8..2cfe8161b478 100644 --- a/arch/arm/kernel/module.c +++ b/arch/arm/kernel/module.c @@ -38,17 +38,9 @@ #ifdef CONFIG_MMU void *module_alloc(unsigned long size) { - struct vm_struct *area; - - size = PAGE_ALIGN(size); - if (!size) - return NULL; - - area = __get_vm_area(size, VM_ALLOC, MODULES_VADDR, MODULES_END); - if (!area) - return NULL; - - return __vmalloc_area(area, GFP_KERNEL, PAGE_KERNEL_EXEC); + return __vmalloc_node_range(size, 1, MODULES_VADDR, MODULES_END, + GFP_KERNEL, PAGE_KERNEL_EXEC, -1, + __builtin_return_address(0)); } #else /* CONFIG_MMU */ void *module_alloc(unsigned long size) diff --git a/arch/mips/kernel/module.c b/arch/mips/kernel/module.c index 6f51dda87fce..d87a72e9fac7 100644 --- a/arch/mips/kernel/module.c +++ b/arch/mips/kernel/module.c @@ -46,17 +46,9 @@ static DEFINE_SPINLOCK(dbe_lock); void *module_alloc(unsigned long size) { #ifdef MODULE_START - struct vm_struct *area; - - size = PAGE_ALIGN(size); - if (!size) - return NULL; - - area = __get_vm_area(size, VM_ALLOC, MODULE_START, MODULE_END); - if (!area) - return NULL; - - return __vmalloc_area(area, GFP_KERNEL, PAGE_KERNEL); + return __vmalloc_node_range(size, 1, MODULE_START, MODULE_END, + GFP_KERNEL, PAGE_KERNEL, -1, + __builtin_return_address(0)); #else if (size == 0) return NULL; diff --git a/arch/sparc/kernel/module.c b/arch/sparc/kernel/module.c index ee3c7dde8d9f..8d348c474a2f 100644 --- a/arch/sparc/kernel/module.c +++ b/arch/sparc/kernel/module.c @@ -23,17 +23,11 @@ static void *module_map(unsigned long size) { - struct vm_struct *area; - - size = PAGE_ALIGN(size); - if (!size || size > MODULES_LEN) - return NULL; - - area = __get_vm_area(size, VM_ALLOC, MODULES_VADDR, MODULES_END); - if (!area) + if (PAGE_ALIGN(size) > MODULES_LEN) return NULL; - - return __vmalloc_area(area, GFP_KERNEL, PAGE_KERNEL); + return __vmalloc_node_range(size, 1, MODULES_VADDR, MODULES_END, + GFP_KERNEL, PAGE_KERNEL, -1, + __builtin_return_address(0)); } static char *dot2underscore(char *name) diff --git a/arch/x86/kernel/module.c b/arch/x86/kernel/module.c index 8f2956091735..ab23f1ad4bf1 100644 --- a/arch/x86/kernel/module.c +++ b/arch/x86/kernel/module.c @@ -37,20 +37,11 @@ void *module_alloc(unsigned long size) { - struct vm_struct *area; - - if (!size) - return NULL; - size = PAGE_ALIGN(size); - if (size > MODULES_LEN) + if (PAGE_ALIGN(size) > MODULES_LEN) return NULL; - - area = __get_vm_area(size, VM_ALLOC, MODULES_VADDR, MODULES_END); - if (!area) - return NULL; - - return __vmalloc_area(area, GFP_KERNEL | __GFP_HIGHMEM, - PAGE_KERNEL_EXEC); + return __vmalloc_node_range(size, 1, MODULES_VADDR, MODULES_END, + GFP_KERNEL | __GFP_HIGHMEM, PAGE_KERNEL_EXEC, + -1, __builtin_return_address(0)); } /* Free memory returned from module_alloc */ diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h index c7348b8d0a81..4ed6fcd6b726 100644 --- a/include/linux/vmalloc.h +++ b/include/linux/vmalloc.h @@ -59,8 +59,9 @@ extern void *vmalloc_exec(unsigned long size); extern void *vmalloc_32(unsigned long size); extern void *vmalloc_32_user(unsigned long size); extern void *__vmalloc(unsigned long size, gfp_t gfp_mask, pgprot_t prot); -extern void *__vmalloc_area(struct vm_struct *area, gfp_t gfp_mask, - pgprot_t prot); +extern void *__vmalloc_node_range(unsigned long size, unsigned long align, + unsigned long start, unsigned long end, gfp_t gfp_mask, + pgprot_t prot, int node, void *caller); extern void vfree(const void *addr); extern void *vmap(struct page **pages, unsigned int count, diff --git a/mm/vmalloc.c b/mm/vmalloc.c index f67546636322..284346ee0e91 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -1530,25 +1530,12 @@ fail: return NULL; } -void *__vmalloc_area(struct vm_struct *area, gfp_t gfp_mask, pgprot_t prot) -{ - void *addr = __vmalloc_area_node(area, gfp_mask, prot, -1, - __builtin_return_address(0)); - - /* - * A ref_count = 3 is needed because the vm_struct and vmap_area - * structures allocated in the __get_vm_area_node() function contain - * references to the virtual address of the vmalloc'ed block. - */ - kmemleak_alloc(addr, area->size - PAGE_SIZE, 3, gfp_mask); - - return addr; -} - /** - * __vmalloc_node - allocate virtually contiguous memory + * __vmalloc_node_range - allocate virtually contiguous memory * @size: allocation size * @align: desired alignment + * @start: vm area range start + * @end: vm area range end * @gfp_mask: flags for the page level allocator * @prot: protection mask for the allocated pages * @node: node to use for allocation or -1 @@ -1558,9 +1545,9 @@ void *__vmalloc_area(struct vm_struct *area, gfp_t gfp_mask, pgprot_t prot) * allocator with @gfp_mask flags. Map them into contiguous * kernel virtual space, using a pagetable protection of @prot. */ -static void *__vmalloc_node(unsigned long size, unsigned long align, - gfp_t gfp_mask, pgprot_t prot, - int node, void *caller) +void *__vmalloc_node_range(unsigned long size, unsigned long align, + unsigned long start, unsigned long end, gfp_t gfp_mask, + pgprot_t prot, int node, void *caller) { struct vm_struct *area; void *addr; @@ -1570,8 +1557,8 @@ static void *__vmalloc_node(unsigned long size, unsigned long align, if (!size || (size >> PAGE_SHIFT) > totalram_pages) return NULL; - area = __get_vm_area_node(size, align, VM_ALLOC, VMALLOC_START, - VMALLOC_END, node, gfp_mask, caller); + area = __get_vm_area_node(size, align, VM_ALLOC, start, end, node, + gfp_mask, caller); if (!area) return NULL; @@ -1588,6 +1575,27 @@ static void *__vmalloc_node(unsigned long size, unsigned long align, return addr; } +/** + * __vmalloc_node - allocate virtually contiguous memory + * @size: allocation size + * @align: desired alignment + * @gfp_mask: flags for the page level allocator + * @prot: protection mask for the allocated pages + * @node: node to use for allocation or -1 + * @caller: caller's return address + * + * Allocate enough pages to cover @size from the page level + * allocator with @gfp_mask flags. Map them into contiguous + * kernel virtual space, using a pagetable protection of @prot. + */ +static void *__vmalloc_node(unsigned long size, unsigned long align, + gfp_t gfp_mask, pgprot_t prot, + int node, void *caller) +{ + return __vmalloc_node_range(size, align, VMALLOC_START, VMALLOC_END, + gfp_mask, prot, node, caller); +} + void *__vmalloc(unsigned long size, gfp_t gfp_mask, pgprot_t prot) { return __vmalloc_node(size, 1, gfp_mask, prot, -1, -- cgit v1.2.3 From dabb16f639820267b3850d804571c70bd93d4e07 Mon Sep 17 00:00:00 2001 From: Mandeep Singh Baines Date: Thu, 13 Jan 2011 15:46:05 -0800 Subject: oom: allow a non-CAP_SYS_RESOURCE proces to oom_score_adj down We'd like to be able to oom_score_adj a process up/down as it enters/leaves the foreground. Currently, it is not possible to oom_adj down without CAP_SYS_RESOURCE. This patch allows a task to decrease its oom_score_adj back to the value that a CAP_SYS_RESOURCE thread set it to or its inherited value at fork. Assuming the thread that has forked it has oom_score_adj of 0, each process could decrease it back from 0 upon activation unless a CAP_SYS_RESOURCE thread elevated it to something higher. Alternative considered: * a setuid binary * a daemon with CAP_SYS_RESOURCE Since you don't wan't all processes to be able to reduce their oom_adj, a setuid or daemon implementation would be complex. The alternatives also have much higher overhead. This patch updated from original patch based on feedback from David Rientjes. Signed-off-by: Mandeep Singh Baines Acked-by: David Rientjes Cc: KAMEZAWA Hiroyuki Cc: KOSAKI Motohiro Cc: Rik van Riel Cc: Ying Han Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/filesystems/proc.txt | 4 ++++ fs/proc/base.c | 4 +++- include/linux/sched.h | 2 ++ kernel/fork.c | 1 + 4 files changed, 10 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt index ef757fca470b..23cae6548d3a 100644 --- a/Documentation/filesystems/proc.txt +++ b/Documentation/filesystems/proc.txt @@ -1323,6 +1323,10 @@ scaled linearly with /proc//oom_score_adj. Writing to /proc//oom_score_adj or /proc//oom_adj will change the other with its scaled value. +The value of /proc//oom_score_adj may be reduced no lower than the last +value set by a CAP_SYS_RESOURCE process. To reduce the value any lower +requires CAP_SYS_RESOURCE. + NOTICE: /proc//oom_adj is deprecated and will be removed, please see Documentation/feature-removal-schedule.txt. diff --git a/fs/proc/base.c b/fs/proc/base.c index 93f1cdd5d3d7..9d096e82b201 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -1151,7 +1151,7 @@ static ssize_t oom_score_adj_write(struct file *file, const char __user *buf, goto err_task_lock; } - if (oom_score_adj < task->signal->oom_score_adj && + if (oom_score_adj < task->signal->oom_score_adj_min && !capable(CAP_SYS_RESOURCE)) { err = -EACCES; goto err_sighand; @@ -1164,6 +1164,8 @@ static ssize_t oom_score_adj_write(struct file *file, const char __user *buf, atomic_dec(&task->mm->oom_disable_count); } task->signal->oom_score_adj = oom_score_adj; + if (has_capability_noaudit(current, CAP_SYS_RESOURCE)) + task->signal->oom_score_adj_min = oom_score_adj; /* * Scale /proc/pid/oom_adj appropriately ensuring that OOM_DISABLE is * always attainable. diff --git a/include/linux/sched.h b/include/linux/sched.h index 07402530fc70..f23b5bb6f52e 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -634,6 +634,8 @@ struct signal_struct { int oom_adj; /* OOM kill score adjustment (bit shift) */ int oom_score_adj; /* OOM kill score adjustment */ + int oom_score_adj_min; /* OOM kill score adjustment minimum value. + * Only settable by CAP_SYS_RESOURCE. */ struct mutex cred_guard_mutex; /* guard against foreign influences on * credential calculations diff --git a/kernel/fork.c b/kernel/fork.c index 76a1fdd80bdf..1499607e4da2 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -910,6 +910,7 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk) sig->oom_adj = current->signal->oom_adj; sig->oom_score_adj = current->signal->oom_score_adj; + sig->oom_score_adj_min = current->signal->oom_score_adj_min; mutex_init(&sig->cred_guard_mutex); -- cgit v1.2.3 From 212260aa07135b327752dc02625c68cf4ce04caf Mon Sep 17 00:00:00 2001 From: Rik van Riel Date: Thu, 13 Jan 2011 15:46:06 -0800 Subject: mm: clear PageError bit in msync & fsync Temporary IO failures, eg. due to loss of both multipath paths, can permanently leave the PageError bit set on a page, resulting in msync or fsync returning -EIO over and over again, even if IO is now getting to the disk correctly. We already clear the AS_ENOSPC and AS_IO bits in mapping->flags in the filemap_fdatawait_range function. Also clearing the PageError bit on the page allows subsequent msync or fsync calls on this file to return without an error, if the subsequent IO succeeds. Unfortunately data written out in the msync or fsync call that returned -EIO can still get lost, because the page dirty bit appears to not get restored on IO error. However, the alternative could be potentially all of memory filling up with uncleanable dirty pages, hanging the system, so there is no nice choice here... Signed-off-by: Rik van Riel Acked-by: Valerie Aurora Acked-by: Jeff Layton Cc: Theodore Ts'o Acked-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/page-flags.h | 2 +- mm/filemap.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index 5f38c460367e..4805fdec8d6e 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -198,7 +198,7 @@ static inline int __TestClearPage##uname(struct page *page) { return 0; } struct page; /* forward declaration */ TESTPAGEFLAG(Locked, locked) TESTSETFLAG(Locked, locked) -PAGEFLAG(Error, error) +PAGEFLAG(Error, error) TESTCLEARFLAG(Error, error) PAGEFLAG(Referenced, referenced) TESTCLEARFLAG(Referenced, referenced) PAGEFLAG(Dirty, dirty) TESTSCFLAG(Dirty, dirty) __CLEARPAGEFLAG(Dirty, dirty) PAGEFLAG(LRU, lru) __CLEARPAGEFLAG(LRU, lru) diff --git a/mm/filemap.c b/mm/filemap.c index 1a3dd5914726..b4ad8e36c81a 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -298,7 +298,7 @@ int filemap_fdatawait_range(struct address_space *mapping, loff_t start_byte, continue; wait_on_page_writeback(page); - if (PageError(page)) + if (TestClearPageError(page)) ret = -EIO; } pagevec_release(&pvec); -- cgit v1.2.3 From 110d74a921f4d272b47ef6104fcf937df808f4c8 Mon Sep 17 00:00:00 2001 From: Michel Lespinasse Date: Thu, 13 Jan 2011 15:46:11 -0800 Subject: mm: add FOLL_MLOCK follow_page flag. Move the code to mlock pages from __mlock_vma_pages_range() to follow_page(). This allows __mlock_vma_pages_range() to not have to break down work into 16-page batches. An additional motivation for doing this within the present patch series is that it'll make it easier for a later chagne to drop mmap_sem when blocking on disk (we'd like to be able to resume at the page that was read from disk instead of at the start of a 16-page batch). Signed-off-by: Michel Lespinasse Cc: Hugh Dickins Cc: Rik van Riel Cc: Peter Zijlstra Cc: Nick Piggin Cc: KOSAKI Motohiro Cc: Ingo Molnar Cc: "H. Peter Anvin" Cc: Thomas Gleixner Cc: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 1 + mm/memory.c | 22 ++++++++++++++++++ mm/mlock.c | 65 +++++------------------------------------------------- 3 files changed, 28 insertions(+), 60 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mm.h b/include/linux/mm.h index 721f451c3029..9ade803bcc1b 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1415,6 +1415,7 @@ struct page *follow_page(struct vm_area_struct *, unsigned long address, #define FOLL_GET 0x04 /* do get_page on page */ #define FOLL_DUMP 0x08 /* give error on hole if it would be zero */ #define FOLL_FORCE 0x10 /* get_user_pages read/write w/o permission */ +#define FOLL_MLOCK 0x40 /* mark page as mlocked */ typedef int (*pte_fn_t)(pte_t *pte, pgtable_t token, unsigned long addr, void *data); diff --git a/mm/memory.c b/mm/memory.c index b8f97b8575b7..15e1f19a3b10 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -1310,6 +1310,28 @@ struct page *follow_page(struct vm_area_struct *vma, unsigned long address, */ mark_page_accessed(page); } + if (flags & FOLL_MLOCK) { + /* + * The preliminary mapping check is mainly to avoid the + * pointless overhead of lock_page on the ZERO_PAGE + * which might bounce very badly if there is contention. + * + * If the page is already locked, we don't need to + * handle it now - vmscan will handle it later if and + * when it attempts to reclaim the page. + */ + if (page->mapping && trylock_page(page)) { + lru_add_drain(); /* push cached pages to LRU */ + /* + * Because we lock page here and migration is + * blocked by the pte's page reference, we need + * only check for file-cache page truncation. + */ + if (page->mapping) + mlock_vma_page(page); + unlock_page(page); + } + } unlock: pte_unmap_unlock(ptep, ptl); out: diff --git a/mm/mlock.c b/mm/mlock.c index 67b3dd8616dc..25cc9e88c540 100644 --- a/mm/mlock.c +++ b/mm/mlock.c @@ -159,10 +159,9 @@ static long __mlock_vma_pages_range(struct vm_area_struct *vma, { struct mm_struct *mm = vma->vm_mm; unsigned long addr = start; - struct page *pages[16]; /* 16 gives a reasonable batch */ int nr_pages = (end - start) / PAGE_SIZE; - int ret = 0; int gup_flags; + int ret; VM_BUG_ON(start & ~PAGE_MASK); VM_BUG_ON(end & ~PAGE_MASK); @@ -170,7 +169,7 @@ static long __mlock_vma_pages_range(struct vm_area_struct *vma, VM_BUG_ON(end > vma->vm_end); VM_BUG_ON(!rwsem_is_locked(&mm->mmap_sem)); - gup_flags = FOLL_TOUCH | FOLL_GET; + gup_flags = FOLL_TOUCH | FOLL_MLOCK; /* * We want to touch writable mappings with a write fault in order * to break COW, except for shared mappings because these don't COW @@ -185,63 +184,9 @@ static long __mlock_vma_pages_range(struct vm_area_struct *vma, nr_pages--; } - while (nr_pages > 0) { - int i; - - cond_resched(); - - /* - * get_user_pages makes pages present if we are - * setting mlock. and this extra reference count will - * disable migration of this page. However, page may - * still be truncated out from under us. - */ - ret = __get_user_pages(current, mm, addr, - min_t(int, nr_pages, ARRAY_SIZE(pages)), - gup_flags, pages, NULL); - /* - * This can happen for, e.g., VM_NONLINEAR regions before - * a page has been allocated and mapped at a given offset, - * or for addresses that map beyond end of a file. - * We'll mlock the pages if/when they get faulted in. - */ - if (ret < 0) - break; - - lru_add_drain(); /* push cached pages to LRU */ - - for (i = 0; i < ret; i++) { - struct page *page = pages[i]; - - if (page->mapping) { - /* - * That preliminary check is mainly to avoid - * the pointless overhead of lock_page on the - * ZERO_PAGE: which might bounce very badly if - * there is contention. However, we're still - * dirtying its cacheline with get/put_page: - * we'll add another __get_user_pages flag to - * avoid it if that case turns out to matter. - */ - lock_page(page); - /* - * Because we lock page here and migration is - * blocked by the elevated reference, we need - * only check for file-cache page truncation. - */ - if (page->mapping) - mlock_vma_page(page); - unlock_page(page); - } - put_page(page); /* ref from get_user_pages() */ - } - - addr += ret * PAGE_SIZE; - nr_pages -= ret; - ret = 0; - } - - return ret; /* 0 or negative error code */ + ret = __get_user_pages(current, mm, addr, nr_pages, gup_flags, + NULL, NULL); + return max(ret, 0); /* 0 or negative error code */ } /* -- cgit v1.2.3 From 088e54658f559a13c2b988086512d76fe9e8f846 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Thu, 13 Jan 2011 15:46:16 -0800 Subject: mm: remove likely() from mapping_unevictable() The mapping_unevictable() has a likely() around the mapping parameter. This mapping parameter comes from page_mapping() which has an unlikely() that the page will be set as PAGE_MAPPING_ANON, and if so, it will return NULL. One would think that this unlikely() means that the mapping returned by page_mapping() would not be NULL, but where page_mapping() is used just above mapping_unevictable(), that unlikely() is incorrect most of the time. This means that the "likely(mapping)" in mapping_unevictable() is incorrect most of the time. Running the annotated branch profiler on my main box which runs firefox, evolution, xchat and is part of my distcc farm, I had this: correct incorrect % Function File Line ------- --------- - -------- ---- ---- 12872836 1269443893 98 mapping_unevictable pagemap.h 51 35935762 1270265395 97 page_mapping mm.h 659 1306198001 143659 0 page_mapping mm.h 657 203131478 121586 0 page_mapping mm.h 657 5415491 1116 0 page_mapping mm.h 657 74899487 1116 0 page_mapping mm.h 657 203132845 224 0 page_mapping mm.h 659 5415464 27 0 page_mapping mm.h 659 13552 0 0 page_mapping mm.h 657 13552 0 0 page_mapping mm.h 659 242630 0 0 page_mapping mm.h 657 242630 0 0 page_mapping mm.h 659 74899487 0 0 page_mapping mm.h 659 The page_mapping() is a static inline, which is why it shows up multiple times. The mapping_unevictable() is also a static inline but seems to be used only once in my setup. The unlikely in page_mapping() was correct a total of 1909540379 times and incorrect 1270533123 times, with a 39% being incorrect. Perhaps this is enough to remove the unlikely from page_mapping() as well. Signed-off-by: Steven Rostedt Reviewed-by: KOSAKI Motohiro Acked-by: Nick Piggin Acked-by: Rik van Riel Cc: Lee Schermerhorn Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/pagemap.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h index 2d1ffe3cf1ee..9c66e994540f 100644 --- a/include/linux/pagemap.h +++ b/include/linux/pagemap.h @@ -48,7 +48,7 @@ static inline void mapping_clear_unevictable(struct address_space *mapping) static inline int mapping_unevictable(struct address_space *mapping) { - if (likely(mapping)) + if (mapping) return test_bit(AS_UNEVICTABLE, &mapping->flags); return !!mapping; } -- cgit v1.2.3 From e20e87795834f2f14cb53baf657b91d9c39f92c8 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Thu, 13 Jan 2011 15:46:17 -0800 Subject: mm: remove unlikely() from page_mapping() page_mapping() has a unlikely that the mapping has PAGE_MAPPING_ANON set. But running the annotated branch profiler on a normal desktop system doing vairous tasks (xchat, evolution, firefox, distcc), it is not really that unlikely that the mapping here will have the PAGE_MAPPING_ANON flag set: correct incorrect % Function File Line ------- --------- - -------- ---- ---- 35935762 1270265395 97 page_mapping mm.h 659 1306198001 143659 0 page_mapping mm.h 657 203131478 121586 0 page_mapping mm.h 657 5415491 1116 0 page_mapping mm.h 657 74899487 1116 0 page_mapping mm.h 657 203132845 224 0 page_mapping mm.h 659 5415464 27 0 page_mapping mm.h 659 13552 0 0 page_mapping mm.h 657 13552 0 0 page_mapping mm.h 659 242630 0 0 page_mapping mm.h 657 242630 0 0 page_mapping mm.h 659 74899487 0 0 page_mapping mm.h 659 The page_mapping() is a static inline, which is why it shows up multiple times. The unlikely in page_mapping() was correct a total of 1909540379 times and incorrect 1270533123 times, with a 39% being incorrect. With this much of an error, it's best to simply remove the unlikely and have the compiler and branch prediction figure this out. Signed-off-by: Steven Rostedt Cc: KOSAKI Motohiro Cc: Nick Piggin Cc: Rik van Riel Cc: Lee Schermerhorn Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/mm.h b/include/linux/mm.h index 9ade803bcc1b..57e11b2a18ba 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -657,7 +657,7 @@ static inline struct address_space *page_mapping(struct page *page) VM_BUG_ON(PageSlab(page)); if (unlikely(PageSwapCache(page))) mapping = &swapper_space; - else if (unlikely((unsigned long)mapping & PAGE_MAPPING_ANON)) + else if ((unsigned long)mapping & PAGE_MAPPING_ANON) mapping = NULL; return mapping; } -- cgit v1.2.3 From 9950474883e027e6e728cbcff25f7f2bf0c96530 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Thu, 13 Jan 2011 15:46:20 -0800 Subject: mm: kswapd: stop high-order balancing when any suitable zone is balanced Simon Kirby reported the following problem We're seeing cases on a number of servers where cache never fully grows to use all available memory. Sometimes we see servers with 4 GB of memory that never seem to have less than 1.5 GB free, even with a constantly-active VM. In some cases, these servers also swap out while this happens, even though they are constantly reading the working set into memory. We have been seeing this happening for a long time; I don't think it's anything recent, and it still happens on 2.6.36. After some debugging work by Simon, Dave Hansen and others, the prevaling theory became that kswapd is reclaiming order-3 pages requested by SLUB too aggressive about it. There are two apparent problems here. On the target machine, there is a small Normal zone in comparison to DMA32. As kswapd tries to balance all zones, it would continually try reclaiming for Normal even though DMA32 was balanced enough for callers. The second problem is that sleeping_prematurely() does not use the same logic as balance_pgdat() when deciding whether to sleep or not. This keeps kswapd artifically awake. A number of tests were run and the figures from previous postings will look very different for a few reasons. One, the old figures were forcing my network card to use GFP_ATOMIC in attempt to replicate Simon's problem. Second, I previous specified slub_min_order=3 again in an attempt to reproduce Simon's problem. In this posting, I'm depending on Simon to say whether his problem is fixed or not and these figures are to show the impact to the ordinary cases. Finally, the "vmscan" figures are taken from /proc/vmstat instead of the tracepoints. There is less information but recording is less disruptive. The first test of relevance was postmark with a process running in the background reading a large amount of anonymous memory in blocks. The objective was to vaguely simulate what was happening on Simon's machine and it's memory intensive enough to have kswapd awake. POSTMARK traceonly kanyzone Transactions per second: 156.00 ( 0.00%) 153.00 (-1.96%) Data megabytes read per second: 21.51 ( 0.00%) 21.52 ( 0.05%) Data megabytes written per second: 29.28 ( 0.00%) 29.11 (-0.58%) Files created alone per second: 250.00 ( 0.00%) 416.00 (39.90%) Files create/transact per second: 79.00 ( 0.00%) 76.00 (-3.95%) Files deleted alone per second: 520.00 ( 0.00%) 420.00 (-23.81%) Files delete/transact per second: 79.00 ( 0.00%) 76.00 (-3.95%) MMTests Statistics: duration User/Sys Time Running Test (seconds) 16.58 17.4 Total Elapsed Time (seconds) 218.48 222.47 VMstat Reclaim Statistics: vmscan Direct reclaims 0 4 Direct reclaim pages scanned 0 203 Direct reclaim pages reclaimed 0 184 Kswapd pages scanned 326631 322018 Kswapd pages reclaimed 312632 309784 Kswapd low wmark quickly 1 4 Kswapd high wmark quickly 122 475 Kswapd skip congestion_wait 1 0 Pages activated 700040 705317 Pages deactivated 212113 203922 Pages written 9875 6363 Total pages scanned 326631 322221 Total pages reclaimed 312632 309968 %age total pages scanned/reclaimed 95.71% 96.20% %age total pages scanned/written 3.02% 1.97% proc vmstat: Faults Major Faults 300 254 Minor Faults 645183 660284 Page ins 493588 486704 Page outs 4960088 4986704 Swap ins 1230 661 Swap outs 9869 6355 Performance is mildly affected because kswapd is no longer doing as much work and the background memory consumer process is getting in the way. Note that kswapd scanned and reclaimed fewer pages as it's less aggressive and overall fewer pages were scanned and reclaimed. Swap in/out is particularly reduced again reflecting kswapd throwing out fewer pages. The slight performance impact is unfortunate here but it looks like a direct result of kswapd being less aggressive. As the bug report is about too many pages being freed by kswapd, it may have to be accepted for now. The second test is a streaming IO benchmark that was previously used by Johannes to show regressions in page reclaim. MICRO traceonly kanyzone User/Sys Time Running Test (seconds) 29.29 28.87 Total Elapsed Time (seconds) 492.18 488.79 VMstat Reclaim Statistics: vmscan Direct reclaims 2128 1460 Direct reclaim pages scanned 2284822 1496067 Direct reclaim pages reclaimed 148919 110937 Kswapd pages scanned 15450014 16202876 Kswapd pages reclaimed 8503697 8537897 Kswapd low wmark quickly 3100 3397 Kswapd high wmark quickly 1860 7243 Kswapd skip congestion_wait 708 801 Pages activated 9635 9573 Pages deactivated 1432 1271 Pages written 223 1130 Total pages scanned 17734836 17698943 Total pages reclaimed 8652616 8648834 %age total pages scanned/reclaimed 48.79% 48.87% %age total pages scanned/written 0.00% 0.01% proc vmstat: Faults Major Faults 165 221 Minor Faults 9655785 9656506 Page ins 3880 7228 Page outs 37692940 37480076 Swap ins 0 69 Swap outs 19 15 Again fewer pages are scanned and reclaimed as expected and this time the test completed faster. Note that kswapd is hitting its watermarks faster (low and high wmark quickly) which I expect is due to kswapd reclaiming fewer pages. I also ran fs-mark, iozone and sysbench but there is nothing interesting to report in the figures. Performance is not significantly changed and the reclaim statistics look reasonable. Tgis patch: When the allocator enters its slow path, kswapd is woken up to balance the node. It continues working until all zones within the node are balanced. For order-0 allocations, this makes perfect sense but for higher orders it can have unintended side-effects. If the zone sizes are imbalanced, kswapd may reclaim heavily within a smaller zone discarding an excessive number of pages. The user-visible behaviour is that kswapd is awake and reclaiming even though plenty of pages are free from a suitable zone. This patch alters the "balance" logic for high-order reclaim allowing kswapd to stop if any suitable zone becomes balanced to reduce the number of pages it reclaims from other zones. kswapd still tries to ensure that order-0 watermarks for all zones are met before sleeping. Signed-off-by: Mel Gorman Reviewed-by: Minchan Kim Reviewed-by: KAMEZAWA Hiroyuki Reviewed-by: Eric B Munson Cc: Simon Kirby Cc: KOSAKI Motohiro Cc: Shaohua Li Cc: Dave Hansen Cc: Johannes Weiner Cc: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mmzone.h | 3 ++- mm/page_alloc.c | 8 +++--- mm/vmscan.c | 68 +++++++++++++++++++++++++++++++++++++++++++------- 3 files changed, 66 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 48906629335c..dad3612a7e39 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -639,6 +639,7 @@ typedef struct pglist_data { wait_queue_head_t kswapd_wait; struct task_struct *kswapd; int kswapd_max_order; + enum zone_type classzone_idx; } pg_data_t; #define node_present_pages(nid) (NODE_DATA(nid)->node_present_pages) @@ -654,7 +655,7 @@ typedef struct pglist_data { extern struct mutex zonelists_mutex; void build_all_zonelists(void *data); -void wakeup_kswapd(struct zone *zone, int order); +void wakeup_kswapd(struct zone *zone, int order, enum zone_type classzone_idx); bool zone_watermark_ok(struct zone *z, int order, unsigned long mark, int classzone_idx, int alloc_flags); bool zone_watermark_ok_safe(struct zone *z, int order, unsigned long mark, diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 0fd486467b4b..c31460918506 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1944,13 +1944,14 @@ __alloc_pages_high_priority(gfp_t gfp_mask, unsigned int order, static inline void wake_all_kswapd(unsigned int order, struct zonelist *zonelist, - enum zone_type high_zoneidx) + enum zone_type high_zoneidx, + enum zone_type classzone_idx) { struct zoneref *z; struct zone *zone; for_each_zone_zonelist(zone, z, zonelist, high_zoneidx) - wakeup_kswapd(zone, order); + wakeup_kswapd(zone, order, classzone_idx); } static inline int @@ -2028,7 +2029,8 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order, goto nopage; restart: - wake_all_kswapd(order, zonelist, high_zoneidx); + wake_all_kswapd(order, zonelist, high_zoneidx, + zone_idx(preferred_zone)); /* * OK, we're below the kswapd watermark and have kicked background diff --git a/mm/vmscan.c b/mm/vmscan.c index 7037cc8c60b6..3584067800e1 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -2246,11 +2246,14 @@ static int sleeping_prematurely(pg_data_t *pgdat, int order, long remaining) * interoperates with the page allocator fallback scheme to ensure that aging * of pages is balanced across the zones. */ -static unsigned long balance_pgdat(pg_data_t *pgdat, int order) +static unsigned long balance_pgdat(pg_data_t *pgdat, int order, + int classzone_idx) { int all_zones_ok; + int any_zone_ok; int priority; int i; + int end_zone = 0; /* Inclusive. 0 = ZONE_DMA */ unsigned long total_scanned; struct reclaim_state *reclaim_state = current->reclaim_state; struct scan_control sc = { @@ -2273,7 +2276,6 @@ loop_again: count_vm_event(PAGEOUTRUN); for (priority = DEF_PRIORITY; priority >= 0; priority--) { - int end_zone = 0; /* Inclusive. 0 = ZONE_DMA */ unsigned long lru_pages = 0; int has_under_min_watermark_zone = 0; @@ -2282,6 +2284,7 @@ loop_again: disable_swap_token(); all_zones_ok = 1; + any_zone_ok = 0; /* * Scan in the highmem->dma direction for the highest @@ -2400,10 +2403,12 @@ loop_again: * spectulatively avoid congestion waits */ zone_clear_flag(zone, ZONE_CONGESTED); + if (i <= classzone_idx) + any_zone_ok = 1; } } - if (all_zones_ok) + if (all_zones_ok || (order && any_zone_ok)) break; /* kswapd: all done */ /* * OK, kswapd is getting into trouble. Take a nap, then take @@ -2426,7 +2431,13 @@ loop_again: break; } out: - if (!all_zones_ok) { + + /* + * order-0: All zones must meet high watermark for a balanced node + * high-order: Any zone below pgdats classzone_idx must meet the high + * watermark for a balanced node + */ + if (!(all_zones_ok || (order && any_zone_ok))) { cond_resched(); try_to_freeze(); @@ -2451,6 +2462,36 @@ out: goto loop_again; } + /* + * If kswapd was reclaiming at a higher order, it has the option of + * sleeping without all zones being balanced. Before it does, it must + * ensure that the watermarks for order-0 on *all* zones are met and + * that the congestion flags are cleared. The congestion flag must + * be cleared as kswapd is the only mechanism that clears the flag + * and it is potentially going to sleep here. + */ + if (order) { + for (i = 0; i <= end_zone; i++) { + struct zone *zone = pgdat->node_zones + i; + + if (!populated_zone(zone)) + continue; + + if (zone->all_unreclaimable && priority != DEF_PRIORITY) + continue; + + /* Confirm the zone is balanced for order-0 */ + if (!zone_watermark_ok(zone, 0, + high_wmark_pages(zone), 0, 0)) { + order = sc.order = 0; + goto loop_again; + } + + /* If balanced, clear the congested flag */ + zone_clear_flag(zone, ZONE_CONGESTED); + } + } + return sc.nr_reclaimed; } @@ -2514,6 +2555,7 @@ static void kswapd_try_to_sleep(pg_data_t *pgdat, int order) static int kswapd(void *p) { unsigned long order; + int classzone_idx; pg_data_t *pgdat = (pg_data_t*)p; struct task_struct *tsk = current; @@ -2544,21 +2586,27 @@ static int kswapd(void *p) set_freezable(); order = 0; + classzone_idx = MAX_NR_ZONES - 1; for ( ; ; ) { unsigned long new_order; + int new_classzone_idx; int ret; new_order = pgdat->kswapd_max_order; + new_classzone_idx = pgdat->classzone_idx; pgdat->kswapd_max_order = 0; - if (order < new_order) { + pgdat->classzone_idx = MAX_NR_ZONES - 1; + if (order < new_order || classzone_idx > new_classzone_idx) { /* * Don't sleep if someone wants a larger 'order' - * allocation + * allocation or has tigher zone constraints */ order = new_order; + classzone_idx = new_classzone_idx; } else { kswapd_try_to_sleep(pgdat, order); order = pgdat->kswapd_max_order; + classzone_idx = pgdat->classzone_idx; } ret = try_to_freeze(); @@ -2571,7 +2619,7 @@ static int kswapd(void *p) */ if (!ret) { trace_mm_vmscan_kswapd_wake(pgdat->node_id, order); - balance_pgdat(pgdat, order); + balance_pgdat(pgdat, order, classzone_idx); } } return 0; @@ -2580,7 +2628,7 @@ static int kswapd(void *p) /* * A zone is low on free memory, so wake its kswapd task to service it. */ -void wakeup_kswapd(struct zone *zone, int order) +void wakeup_kswapd(struct zone *zone, int order, enum zone_type classzone_idx) { pg_data_t *pgdat; @@ -2590,8 +2638,10 @@ void wakeup_kswapd(struct zone *zone, int order) if (!cpuset_zone_allowed_hardwall(zone, GFP_KERNEL)) return; pgdat = zone->zone_pgdat; - if (pgdat->kswapd_max_order < order) + if (pgdat->kswapd_max_order < order) { pgdat->kswapd_max_order = order; + pgdat->classzone_idx = min(pgdat->classzone_idx, classzone_idx); + } if (!waitqueue_active(&pgdat->kswapd_wait)) return; if (zone_watermark_ok_safe(zone, order, low_wmark_pages(zone), 0, 0)) -- cgit v1.2.3 From e9da73d67729b58bba256123e2b4651e0d8a01ac Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:46:32 -0800 Subject: thp: compound_lock Add a new compound_lock() needed to serialize put_page against __split_huge_page_refcount(). Signed-off-by: Andrea Arcangeli Acked-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 34 ++++++++++++++++++++++++++++++++++ include/linux/page-flags.h | 12 +++++++++++- 2 files changed, 45 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/mm.h b/include/linux/mm.h index 57e11b2a18ba..3b1754ad8785 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -14,6 +14,7 @@ #include #include #include +#include struct mempolicy; struct anon_vma; @@ -305,6 +306,39 @@ static inline int is_vmalloc_or_module_addr(const void *x) } #endif +static inline void compound_lock(struct page *page) +{ +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + bit_spin_lock(PG_compound_lock, &page->flags); +#endif +} + +static inline void compound_unlock(struct page *page) +{ +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + bit_spin_unlock(PG_compound_lock, &page->flags); +#endif +} + +static inline unsigned long compound_lock_irqsave(struct page *page) +{ + unsigned long uninitialized_var(flags); +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + local_irq_save(flags); + compound_lock(page); +#endif + return flags; +} + +static inline void compound_unlock_irqrestore(struct page *page, + unsigned long flags) +{ +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + compound_unlock(page); + local_irq_restore(flags); +#endif +} + static inline struct page *compound_head(struct page *page) { if (unlikely(PageTail(page))) diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index 4805fdec8d6e..5a743cc87238 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -107,6 +107,9 @@ enum pageflags { #endif #ifdef CONFIG_MEMORY_FAILURE PG_hwpoison, /* hardware poisoned page. Don't touch */ +#endif +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + PG_compound_lock, #endif __NR_PAGEFLAGS, @@ -397,6 +400,12 @@ static inline void __ClearPageTail(struct page *page) #define __PG_MLOCKED 0 #endif +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +#define __PG_COMPOUND_LOCK (1 << PG_compound_lock) +#else +#define __PG_COMPOUND_LOCK 0 +#endif + /* * Flags checked when a page is freed. Pages being freed should not have * these flags set. It they are, there is a problem. @@ -406,7 +415,8 @@ static inline void __ClearPageTail(struct page *page) 1 << PG_private | 1 << PG_private_2 | \ 1 << PG_buddy | 1 << PG_writeback | 1 << PG_reserved | \ 1 << PG_slab | 1 << PG_swapcache | 1 << PG_active | \ - 1 << PG_unevictable | __PG_MLOCKED | __PG_HWPOISON) + 1 << PG_unevictable | __PG_MLOCKED | __PG_HWPOISON | \ + __PG_COMPOUND_LOCK) /* * Flags checked when a page is prepped for return by the page allocator. -- cgit v1.2.3 From 9180706344487700b40da9eca5dedd3d11cb33b4 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:46:32 -0800 Subject: thp: alter compound get_page/put_page Alter compound get_page/put_page to keep references on subpages too, in order to allow __split_huge_page_refcount to split an hugepage even while subpages have been pinned by one of the get_user_pages() variants. Signed-off-by: Andrea Arcangeli Acked-by: Rik van Riel Acked-by: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/mm/gup.c | 12 +++++++ arch/x86/mm/gup.c | 12 +++++++ include/linux/mm.h | 24 ++++++++++++-- mm/swap.c | 90 +++++++++++++++++++++++++++++++++++++++++++++++---- 4 files changed, 129 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/arch/powerpc/mm/gup.c b/arch/powerpc/mm/gup.c index d7efdbf640c7..fec13200868f 100644 --- a/arch/powerpc/mm/gup.c +++ b/arch/powerpc/mm/gup.c @@ -16,6 +16,16 @@ #ifdef __HAVE_ARCH_PTE_SPECIAL +static inline void get_huge_page_tail(struct page *page) +{ + /* + * __split_huge_page_refcount() cannot run + * from under us. + */ + VM_BUG_ON(atomic_read(&page->_count) < 0); + atomic_inc(&page->_count); +} + /* * The performance critical leaf functions are made noinline otherwise gcc * inlines everything into a single function which results in too much @@ -47,6 +57,8 @@ static noinline int gup_pte_range(pmd_t pmd, unsigned long addr, put_page(page); return 0; } + if (PageTail(page)) + get_huge_page_tail(page); pages[*nr] = page; (*nr)++; diff --git a/arch/x86/mm/gup.c b/arch/x86/mm/gup.c index 738e6593799d..06f56fcf9a77 100644 --- a/arch/x86/mm/gup.c +++ b/arch/x86/mm/gup.c @@ -105,6 +105,16 @@ static inline void get_head_page_multiple(struct page *page, int nr) atomic_add(nr, &page->_count); } +static inline void get_huge_page_tail(struct page *page) +{ + /* + * __split_huge_page_refcount() cannot run + * from under us. + */ + VM_BUG_ON(atomic_read(&page->_count) < 0); + atomic_inc(&page->_count); +} + static noinline int gup_huge_pmd(pmd_t pmd, unsigned long addr, unsigned long end, int write, struct page **pages, int *nr) { @@ -128,6 +138,8 @@ static noinline int gup_huge_pmd(pmd_t pmd, unsigned long addr, do { VM_BUG_ON(compound_head(page) != head); pages[*nr] = page; + if (PageTail(page)) + get_huge_page_tail(page); (*nr)++; page++; refs++; diff --git a/include/linux/mm.h b/include/linux/mm.h index 3b1754ad8785..a2718e1ed585 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -353,9 +353,29 @@ static inline int page_count(struct page *page) static inline void get_page(struct page *page) { - page = compound_head(page); - VM_BUG_ON(atomic_read(&page->_count) == 0); + /* + * Getting a normal page or the head of a compound page + * requires to already have an elevated page->_count. Only if + * we're getting a tail page, the elevated page->_count is + * required only in the head page, so for tail pages the + * bugcheck only verifies that the page->_count isn't + * negative. + */ + VM_BUG_ON(atomic_read(&page->_count) < !PageTail(page)); atomic_inc(&page->_count); + /* + * Getting a tail page will elevate both the head and tail + * page->_count(s). + */ + if (unlikely(PageTail(page))) { + /* + * This is safe only because + * __split_huge_page_refcount can't run under + * get_page(). + */ + VM_BUG_ON(atomic_read(&page->first_page->_count) <= 0); + atomic_inc(&page->first_page->_count); + } } static inline struct page *virt_to_head_page(const void *x) diff --git a/mm/swap.c b/mm/swap.c index 3f4854205b16..33f5292fe132 100644 --- a/mm/swap.c +++ b/mm/swap.c @@ -56,17 +56,93 @@ static void __page_cache_release(struct page *page) del_page_from_lru(zone, page); spin_unlock_irqrestore(&zone->lru_lock, flags); } +} + +static void __put_single_page(struct page *page) +{ + __page_cache_release(page); free_hot_cold_page(page, 0); } -static void put_compound_page(struct page *page) +static void __put_compound_page(struct page *page) { - page = compound_head(page); - if (put_page_testzero(page)) { - compound_page_dtor *dtor; + compound_page_dtor *dtor; + + __page_cache_release(page); + dtor = get_compound_page_dtor(page); + (*dtor)(page); +} - dtor = get_compound_page_dtor(page); - (*dtor)(page); +static void put_compound_page(struct page *page) +{ + if (unlikely(PageTail(page))) { + /* __split_huge_page_refcount can run under us */ + struct page *page_head = page->first_page; + smp_rmb(); + /* + * If PageTail is still set after smp_rmb() we can be sure + * that the page->first_page we read wasn't a dangling pointer. + * See __split_huge_page_refcount() smp_wmb(). + */ + if (likely(PageTail(page) && get_page_unless_zero(page_head))) { + unsigned long flags; + /* + * Verify that our page_head wasn't converted + * to a a regular page before we got a + * reference on it. + */ + if (unlikely(!PageHead(page_head))) { + /* PageHead is cleared after PageTail */ + smp_rmb(); + VM_BUG_ON(PageTail(page)); + goto out_put_head; + } + /* + * Only run compound_lock on a valid PageHead, + * after having it pinned with + * get_page_unless_zero() above. + */ + smp_mb(); + /* page_head wasn't a dangling pointer */ + flags = compound_lock_irqsave(page_head); + if (unlikely(!PageTail(page))) { + /* __split_huge_page_refcount run before us */ + compound_unlock_irqrestore(page_head, flags); + VM_BUG_ON(PageHead(page_head)); + out_put_head: + if (put_page_testzero(page_head)) + __put_single_page(page_head); + out_put_single: + if (put_page_testzero(page)) + __put_single_page(page); + return; + } + VM_BUG_ON(page_head != page->first_page); + /* + * We can release the refcount taken by + * get_page_unless_zero now that + * split_huge_page_refcount is blocked on the + * compound_lock. + */ + if (put_page_testzero(page_head)) + VM_BUG_ON(1); + /* __split_huge_page_refcount will wait now */ + VM_BUG_ON(atomic_read(&page->_count) <= 0); + atomic_dec(&page->_count); + VM_BUG_ON(atomic_read(&page_head->_count) <= 0); + compound_unlock_irqrestore(page_head, flags); + if (put_page_testzero(page_head)) + __put_compound_page(page_head); + } else { + /* page_head is a dangling pointer */ + VM_BUG_ON(PageTail(page)); + goto out_put_single; + } + } else if (put_page_testzero(page)) { + if (PageHead(page)) + __put_compound_page(page); + else + __put_single_page(page); } } @@ -75,7 +151,7 @@ void put_page(struct page *page) if (unlikely(PageCompound(page))) put_compound_page(page); else if (put_page_testzero(page)) - __page_cache_release(page); + __put_single_page(page); } EXPORT_SYMBOL(put_page); -- cgit v1.2.3 From 14fd403f2146f740942d78af4e0ee59396ad8eab Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:46:37 -0800 Subject: thp: export maybe_mkwrite huge_memory.c needs it too when it fallbacks in copying hugepages into regular fragmented pages if hugepage allocation fails during COW. Signed-off-by: Andrea Arcangeli Acked-by: Rik van Riel Acked-by: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 13 +++++++++++++ mm/memory.c | 13 ------------- 2 files changed, 13 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mm.h b/include/linux/mm.h index a2718e1ed585..6bef67d74adf 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -429,6 +429,19 @@ static inline void set_compound_order(struct page *page, unsigned long order) page[1].lru.prev = (void *)order; } +/* + * Do pte_mkwrite, but only if the vma says VM_WRITE. We do this when + * servicing faults for write access. In the normal case, do always want + * pte_mkwrite. But get_user_pages can cause write faults for mappings + * that do not have writing enabled, when used by access_process_vm. + */ +static inline pte_t maybe_mkwrite(pte_t pte, struct vm_area_struct *vma) +{ + if (likely(vma->vm_flags & VM_WRITE)) + pte = pte_mkwrite(pte); + return pte; +} + /* * Multiple processes may "see" the same page. E.g. for untouched * mappings of /dev/null, all processes see the same page full of diff --git a/mm/memory.c b/mm/memory.c index 1bbe9a22429c..bdf19366b705 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -2083,19 +2083,6 @@ static inline int pte_unmap_same(struct mm_struct *mm, pmd_t *pmd, return same; } -/* - * Do pte_mkwrite, but only if the vma says VM_WRITE. We do this when - * servicing faults for write access. In the normal case, do always want - * pte_mkwrite. But get_user_pages can cause write faults for mappings - * that do not have writing enabled, when used by access_process_vm. - */ -static inline pte_t maybe_mkwrite(pte_t pte, struct vm_area_struct *vma) -{ - if (likely(vma->vm_flags & VM_WRITE)) - pte = pte_mkwrite(pte); - return pte; -} - static inline void cow_user_page(struct page *dst, struct page *src, unsigned long va, struct vm_area_struct *vma) { /* -- cgit v1.2.3 From 8ac1f8320a0073f28cf9e0491af4cd98f504f92a Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:46:43 -0800 Subject: thp: pte alloc trans splitting pte alloc routines must wait for split_huge_page if the pmd is not present and not null (i.e. pmd_trans_splitting). The additional branches are optimized away at compile time by pmd_trans_splitting if the config option is off. However we must pass the vma down in order to know the anon_vma lock to wait for. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Andrea Arcangeli Acked-by: Rik van Riel Acked-by: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/mm/pgd.c | 2 +- arch/ia64/mm/hugetlbpage.c | 2 +- arch/sh/mm/hugetlbpage.c | 2 +- arch/sparc/mm/generic_32.c | 2 +- arch/sparc/mm/generic_64.c | 2 +- arch/sparc/mm/hugetlbpage.c | 2 +- arch/um/kernel/skas/mmu.c | 2 +- arch/x86/kernel/tboot.c | 2 +- include/linux/mm.h | 15 +++++++++------ mm/memory.c | 19 +++++++++++++------ mm/mremap.c | 8 +++++--- 11 files changed, 35 insertions(+), 23 deletions(-) (limited to 'include/linux') diff --git a/arch/arm/mm/pgd.c b/arch/arm/mm/pgd.c index 93292a18cf77..709244c66fa3 100644 --- a/arch/arm/mm/pgd.c +++ b/arch/arm/mm/pgd.c @@ -50,7 +50,7 @@ pgd_t *pgd_alloc(struct mm_struct *mm) if (!new_pmd) goto no_pmd; - new_pte = pte_alloc_map(mm, new_pmd, 0); + new_pte = pte_alloc_map(mm, NULL, new_pmd, 0); if (!new_pte) goto no_pte; diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c index 1841ee7e65f9..5ca674b74737 100644 --- a/arch/ia64/mm/hugetlbpage.c +++ b/arch/ia64/mm/hugetlbpage.c @@ -38,7 +38,7 @@ huge_pte_alloc(struct mm_struct *mm, unsigned long addr, unsigned long sz) if (pud) { pmd = pmd_alloc(mm, pud, taddr); if (pmd) - pte = pte_alloc_map(mm, pmd, taddr); + pte = pte_alloc_map(mm, NULL, pmd, taddr); } return pte; } diff --git a/arch/sh/mm/hugetlbpage.c b/arch/sh/mm/hugetlbpage.c index 9163db3e8d15..d7762349ea48 100644 --- a/arch/sh/mm/hugetlbpage.c +++ b/arch/sh/mm/hugetlbpage.c @@ -35,7 +35,7 @@ pte_t *huge_pte_alloc(struct mm_struct *mm, if (pud) { pmd = pmd_alloc(mm, pud, addr); if (pmd) - pte = pte_alloc_map(mm, pmd, addr); + pte = pte_alloc_map(mm, NULL, pmd, addr); } } diff --git a/arch/sparc/mm/generic_32.c b/arch/sparc/mm/generic_32.c index 5edcac184eaf..e6067b75f11c 100644 --- a/arch/sparc/mm/generic_32.c +++ b/arch/sparc/mm/generic_32.c @@ -50,7 +50,7 @@ static inline int io_remap_pmd_range(struct mm_struct *mm, pmd_t * pmd, unsigned end = PGDIR_SIZE; offset -= address; do { - pte_t * pte = pte_alloc_map(mm, pmd, address); + pte_t *pte = pte_alloc_map(mm, NULL, pmd, address); if (!pte) return -ENOMEM; io_remap_pte_range(mm, pte, address, end - address, address + offset, prot, space); diff --git a/arch/sparc/mm/generic_64.c b/arch/sparc/mm/generic_64.c index 04f2bf4cd571..3cb00dfd4bd6 100644 --- a/arch/sparc/mm/generic_64.c +++ b/arch/sparc/mm/generic_64.c @@ -92,7 +92,7 @@ static inline int io_remap_pmd_range(struct mm_struct *mm, pmd_t * pmd, unsigned end = PGDIR_SIZE; offset -= address; do { - pte_t * pte = pte_alloc_map(mm, pmd, address); + pte_t *pte = pte_alloc_map(mm, NULL, pmd, address); if (!pte) return -ENOMEM; io_remap_pte_range(mm, pte, address, end - address, address + offset, prot, space); diff --git a/arch/sparc/mm/hugetlbpage.c b/arch/sparc/mm/hugetlbpage.c index 5fdddf134caa..f4e97646ce23 100644 --- a/arch/sparc/mm/hugetlbpage.c +++ b/arch/sparc/mm/hugetlbpage.c @@ -214,7 +214,7 @@ pte_t *huge_pte_alloc(struct mm_struct *mm, if (pud) { pmd = pmd_alloc(mm, pud, addr); if (pmd) - pte = pte_alloc_map(mm, pmd, addr); + pte = pte_alloc_map(mm, NULL, pmd, addr); } return pte; } diff --git a/arch/um/kernel/skas/mmu.c b/arch/um/kernel/skas/mmu.c index 3d099f974785..1aee587e9c5d 100644 --- a/arch/um/kernel/skas/mmu.c +++ b/arch/um/kernel/skas/mmu.c @@ -31,7 +31,7 @@ static int init_stub_pte(struct mm_struct *mm, unsigned long proc, if (!pmd) goto out_pmd; - pte = pte_alloc_map(mm, pmd, proc); + pte = pte_alloc_map(mm, NULL, pmd, proc); if (!pte) goto out_pte; diff --git a/arch/x86/kernel/tboot.c b/arch/x86/kernel/tboot.c index c2f1b26141e2..998e972f3b1a 100644 --- a/arch/x86/kernel/tboot.c +++ b/arch/x86/kernel/tboot.c @@ -133,7 +133,7 @@ static int map_tboot_page(unsigned long vaddr, unsigned long pfn, pmd = pmd_alloc(&tboot_mm, pud, vaddr); if (!pmd) return -1; - pte = pte_alloc_map(&tboot_mm, pmd, vaddr); + pte = pte_alloc_map(&tboot_mm, NULL, pmd, vaddr); if (!pte) return -1; set_pte_at(&tboot_mm, vaddr, pte, pfn_pte(pfn, prot)); diff --git a/include/linux/mm.h b/include/linux/mm.h index 6bef67d74adf..14ddd98b063f 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1131,7 +1131,8 @@ static inline int __pmd_alloc(struct mm_struct *mm, pud_t *pud, int __pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long address); #endif -int __pte_alloc(struct mm_struct *mm, pmd_t *pmd, unsigned long address); +int __pte_alloc(struct mm_struct *mm, struct vm_area_struct *vma, + pmd_t *pmd, unsigned long address); int __pte_alloc_kernel(pmd_t *pmd, unsigned long address); /* @@ -1200,16 +1201,18 @@ static inline void pgtable_page_dtor(struct page *page) pte_unmap(pte); \ } while (0) -#define pte_alloc_map(mm, pmd, address) \ - ((unlikely(!pmd_present(*(pmd))) && __pte_alloc(mm, pmd, address))? \ - NULL: pte_offset_map(pmd, address)) +#define pte_alloc_map(mm, vma, pmd, address) \ + ((unlikely(pmd_none(*(pmd))) && __pte_alloc(mm, vma, \ + pmd, address))? \ + NULL: pte_offset_map(pmd, address)) #define pte_alloc_map_lock(mm, pmd, address, ptlp) \ - ((unlikely(!pmd_present(*(pmd))) && __pte_alloc(mm, pmd, address))? \ + ((unlikely(pmd_none(*(pmd))) && __pte_alloc(mm, NULL, \ + pmd, address))? \ NULL: pte_offset_map_lock(mm, pmd, address, ptlp)) #define pte_alloc_kernel(pmd, address) \ - ((unlikely(!pmd_present(*(pmd))) && __pte_alloc_kernel(pmd, address))? \ + ((unlikely(pmd_none(*(pmd))) && __pte_alloc_kernel(pmd, address))? \ NULL: pte_offset_kernel(pmd, address)) extern void free_area_init(unsigned long * zones_size); diff --git a/mm/memory.c b/mm/memory.c index bdf19366b705..567bca80ea53 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -394,9 +394,11 @@ void free_pgtables(struct mmu_gather *tlb, struct vm_area_struct *vma, } } -int __pte_alloc(struct mm_struct *mm, pmd_t *pmd, unsigned long address) +int __pte_alloc(struct mm_struct *mm, struct vm_area_struct *vma, + pmd_t *pmd, unsigned long address) { pgtable_t new = pte_alloc_one(mm, address); + int wait_split_huge_page; if (!new) return -ENOMEM; @@ -416,14 +418,18 @@ int __pte_alloc(struct mm_struct *mm, pmd_t *pmd, unsigned long address) smp_wmb(); /* Could be smp_wmb__xxx(before|after)_spin_lock */ spin_lock(&mm->page_table_lock); - if (!pmd_present(*pmd)) { /* Has another populated it ? */ + wait_split_huge_page = 0; + if (likely(pmd_none(*pmd))) { /* Has another populated it ? */ mm->nr_ptes++; pmd_populate(mm, pmd, new); new = NULL; - } + } else if (unlikely(pmd_trans_splitting(*pmd))) + wait_split_huge_page = 1; spin_unlock(&mm->page_table_lock); if (new) pte_free(mm, new); + if (wait_split_huge_page) + wait_split_huge_page(vma->anon_vma, pmd); return 0; } @@ -436,10 +442,11 @@ int __pte_alloc_kernel(pmd_t *pmd, unsigned long address) smp_wmb(); /* See comment in __pte_alloc */ spin_lock(&init_mm.page_table_lock); - if (!pmd_present(*pmd)) { /* Has another populated it ? */ + if (likely(pmd_none(*pmd))) { /* Has another populated it ? */ pmd_populate_kernel(&init_mm, pmd, new); new = NULL; - } + } else + VM_BUG_ON(pmd_trans_splitting(*pmd)); spin_unlock(&init_mm.page_table_lock); if (new) pte_free_kernel(&init_mm, new); @@ -3253,7 +3260,7 @@ int handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma, pmd = pmd_alloc(mm, pud, address); if (!pmd) return VM_FAULT_OOM; - pte = pte_alloc_map(mm, pmd, address); + pte = pte_alloc_map(mm, vma, pmd, address); if (!pte) return VM_FAULT_OOM; diff --git a/mm/mremap.c b/mm/mremap.c index 563fbdd6293a..b09eefaea0b8 100644 --- a/mm/mremap.c +++ b/mm/mremap.c @@ -47,7 +47,8 @@ static pmd_t *get_old_pmd(struct mm_struct *mm, unsigned long addr) return pmd; } -static pmd_t *alloc_new_pmd(struct mm_struct *mm, unsigned long addr) +static pmd_t *alloc_new_pmd(struct mm_struct *mm, struct vm_area_struct *vma, + unsigned long addr) { pgd_t *pgd; pud_t *pud; @@ -62,7 +63,8 @@ static pmd_t *alloc_new_pmd(struct mm_struct *mm, unsigned long addr) if (!pmd) return NULL; - if (!pmd_present(*pmd) && __pte_alloc(mm, pmd, addr)) + VM_BUG_ON(pmd_trans_huge(*pmd)); + if (pmd_none(*pmd) && __pte_alloc(mm, vma, pmd, addr)) return NULL; return pmd; @@ -147,7 +149,7 @@ unsigned long move_page_tables(struct vm_area_struct *vma, old_pmd = get_old_pmd(vma->vm_mm, old_addr); if (!old_pmd) continue; - new_pmd = alloc_new_pmd(vma->vm_mm, new_addr); + new_pmd = alloc_new_pmd(vma->vm_mm, vma, new_addr); if (!new_pmd) break; next = (new_addr + PMD_SIZE) & PMD_MASK; -- cgit v1.2.3 From 91a4ee2670e0ee2b0630a0eb24fb9a87ea3acd0a Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:46:44 -0800 Subject: thp: add pmd mmu_notifier helpers Add mmu notifier helpers to handle pmd huge operations. Signed-off-by: Andrea Arcangeli Acked-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mmu_notifier.h | 40 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 40 insertions(+) (limited to 'include/linux') diff --git a/include/linux/mmu_notifier.h b/include/linux/mmu_notifier.h index 43dcfbdc39de..cbfab1e9957d 100644 --- a/include/linux/mmu_notifier.h +++ b/include/linux/mmu_notifier.h @@ -243,6 +243,32 @@ static inline void mmu_notifier_mm_destroy(struct mm_struct *mm) __pte; \ }) +#define pmdp_clear_flush_notify(__vma, __address, __pmdp) \ +({ \ + pmd_t __pmd; \ + struct vm_area_struct *___vma = __vma; \ + unsigned long ___address = __address; \ + VM_BUG_ON(__address & ~HPAGE_PMD_MASK); \ + mmu_notifier_invalidate_range_start(___vma->vm_mm, ___address, \ + (__address)+HPAGE_PMD_SIZE);\ + __pmd = pmdp_clear_flush(___vma, ___address, __pmdp); \ + mmu_notifier_invalidate_range_end(___vma->vm_mm, ___address, \ + (__address)+HPAGE_PMD_SIZE); \ + __pmd; \ +}) + +#define pmdp_splitting_flush_notify(__vma, __address, __pmdp) \ +({ \ + struct vm_area_struct *___vma = __vma; \ + unsigned long ___address = __address; \ + VM_BUG_ON(__address & ~HPAGE_PMD_MASK); \ + mmu_notifier_invalidate_range_start(___vma->vm_mm, ___address, \ + (__address)+HPAGE_PMD_SIZE);\ + pmdp_splitting_flush(___vma, ___address, __pmdp); \ + mmu_notifier_invalidate_range_end(___vma->vm_mm, ___address, \ + (__address)+HPAGE_PMD_SIZE); \ +}) + #define ptep_clear_flush_young_notify(__vma, __address, __ptep) \ ({ \ int __young; \ @@ -254,6 +280,17 @@ static inline void mmu_notifier_mm_destroy(struct mm_struct *mm) __young; \ }) +#define pmdp_clear_flush_young_notify(__vma, __address, __pmdp) \ +({ \ + int __young; \ + struct vm_area_struct *___vma = __vma; \ + unsigned long ___address = __address; \ + __young = pmdp_clear_flush_young(___vma, ___address, __pmdp); \ + __young |= mmu_notifier_clear_flush_young(___vma->vm_mm, \ + ___address); \ + __young; \ +}) + #define set_pte_at_notify(__mm, __address, __ptep, __pte) \ ({ \ struct mm_struct *___mm = __mm; \ @@ -305,7 +342,10 @@ static inline void mmu_notifier_mm_destroy(struct mm_struct *mm) } #define ptep_clear_flush_young_notify ptep_clear_flush_young +#define pmdp_clear_flush_young_notify pmdp_clear_flush_young #define ptep_clear_flush_notify ptep_clear_flush +#define pmdp_clear_flush_notify pmdp_clear_flush +#define pmdp_splitting_flush_notify pmdp_splitting_flush #define set_pte_at_notify set_pte_at #endif /* CONFIG_MMU_NOTIFIER */ -- cgit v1.2.3 From 4e6af67e970a2ac287739a4c526c857b5bda27ec Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:46:44 -0800 Subject: thp: clear page compound split_huge_page must transform a compound page to a regular page and needs ClearPageCompound. Signed-off-by: Andrea Arcangeli Acked-by: Rik van Riel Reviewed-by: Christoph Lameter Acked-by: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/page-flags.h | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index 5a743cc87238..4835cae71047 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -347,7 +347,7 @@ static inline void set_page_writeback(struct page *page) * tests can be used in performance sensitive paths. PageCompound is * generally not used in hot code paths. */ -__PAGEFLAG(Head, head) +__PAGEFLAG(Head, head) CLEARPAGEFLAG(Head, head) __PAGEFLAG(Tail, tail) static inline int PageCompound(struct page *page) @@ -355,6 +355,13 @@ static inline int PageCompound(struct page *page) return page->flags & ((1L << PG_head) | (1L << PG_tail)); } +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +static inline void ClearPageCompound(struct page *page) +{ + BUG_ON(!PageHead(page)); + ClearPageHead(page); +} +#endif #else /* * Reduce page flag use as much as possible by overlapping @@ -392,6 +399,14 @@ static inline void __ClearPageTail(struct page *page) page->flags &= ~PG_head_tail_mask; } +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +static inline void ClearPageCompound(struct page *page) +{ + BUG_ON((page->flags & PG_head_tail_mask) != (1 << PG_compound)); + clear_bit(PG_compound, &page->flags); +} +#endif + #endif /* !PAGEFLAGS_EXTENDED */ #ifdef CONFIG_MMU -- cgit v1.2.3 From e7a00c45f29c0155007aa150bf231a70fa470365 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:46:45 -0800 Subject: thp: add pmd_huge_pte to mm_struct This increase the size of the mm struct a bit but it is needed to preallocate one pte for each hugepage so that split_huge_page will not require a fail path. Guarantee of success is a fundamental property of split_huge_page to avoid decrasing swapping reliability and to avoid adding -ENOMEM fail paths that would otherwise force the hugepage-unaware VM code to learn rolling back in the middle of its pte mangling operations (if something we need it to learn handling pmd_trans_huge natively rather being capable of rollback). When split_huge_page runs a pte is needed to succeed the split, to map the newly splitted regular pages with a regular pte. This way all existing VM code remains backwards compatible by just adding a split_huge_page* one liner. The memory waste of those preallocated ptes is negligible and so it is worth it. Signed-off-by: Andrea Arcangeli Acked-by: Rik van Riel Acked-by: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm_types.h | 3 +++ kernel/fork.c | 7 +++++++ 2 files changed, 10 insertions(+) (limited to 'include/linux') diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index bb7288a782fd..26bc4e2cd275 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -309,6 +309,9 @@ struct mm_struct { #endif #ifdef CONFIG_MMU_NOTIFIER struct mmu_notifier_mm *mmu_notifier_mm; +#endif +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + pgtable_t pmd_huge_pte; /* protected by page_table_lock */ #endif /* How many tasks sharing this mm are OOM_DISABLE */ atomic_t oom_disable_count; diff --git a/kernel/fork.c b/kernel/fork.c index 1499607e4da2..f78f50ba6cb2 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -529,6 +529,9 @@ void __mmdrop(struct mm_struct *mm) mm_free_pgd(mm); destroy_context(mm); mmu_notifier_mm_destroy(mm); +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + VM_BUG_ON(mm->pmd_huge_pte); +#endif free_mm(mm); } EXPORT_SYMBOL_GPL(__mmdrop); @@ -669,6 +672,10 @@ struct mm_struct *dup_mm(struct task_struct *tsk) mm->token_priority = 0; mm->last_interval = 0; +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + mm->pmd_huge_pte = NULL; +#endif + if (!mm_init(mm, tsk)) goto fail_nomem; -- cgit v1.2.3 From 47ad8475c000141eacb3ecda5e5ce4b43a9cd04d Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:46:47 -0800 Subject: thp: clear_copy_huge_page Move the copy/clear_huge_page functions to common code to share between hugetlb.c and huge_memory.c. Signed-off-by: Andrea Arcangeli Acked-by: Rik van Riel Acked-by: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 9 +++++++ mm/hugetlb.c | 70 +++-------------------------------------------------- mm/memory.c | 71 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 83 insertions(+), 67 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mm.h b/include/linux/mm.h index 14ddd98b063f..cc6ab1038f6f 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1589,5 +1589,14 @@ static inline int is_hwpoison_address(unsigned long addr) extern void dump_page(struct page *page); +#if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_HUGETLBFS) +extern void clear_huge_page(struct page *page, + unsigned long addr, + unsigned int pages_per_huge_page); +extern void copy_user_huge_page(struct page *dst, struct page *src, + unsigned long addr, struct vm_area_struct *vma, + unsigned int pages_per_huge_page); +#endif /* CONFIG_TRANSPARENT_HUGEPAGE || CONFIG_HUGETLBFS */ + #endif /* __KERNEL__ */ #endif /* _LINUX_MM_H */ diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 85855240933d..7bf223d6677b 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -394,71 +394,6 @@ static int vma_has_reserves(struct vm_area_struct *vma) return 0; } -static void clear_gigantic_page(struct page *page, - unsigned long addr, unsigned long sz) -{ - int i; - struct page *p = page; - - might_sleep(); - for (i = 0; i < sz/PAGE_SIZE; i++, p = mem_map_next(p, page, i)) { - cond_resched(); - clear_user_highpage(p, addr + i * PAGE_SIZE); - } -} -static void clear_huge_page(struct page *page, - unsigned long addr, unsigned long sz) -{ - int i; - - if (unlikely(sz/PAGE_SIZE > MAX_ORDER_NR_PAGES)) { - clear_gigantic_page(page, addr, sz); - return; - } - - might_sleep(); - for (i = 0; i < sz/PAGE_SIZE; i++) { - cond_resched(); - clear_user_highpage(page + i, addr + i * PAGE_SIZE); - } -} - -static void copy_user_gigantic_page(struct page *dst, struct page *src, - unsigned long addr, struct vm_area_struct *vma) -{ - int i; - struct hstate *h = hstate_vma(vma); - struct page *dst_base = dst; - struct page *src_base = src; - - for (i = 0; i < pages_per_huge_page(h); ) { - cond_resched(); - copy_user_highpage(dst, src, addr + i*PAGE_SIZE, vma); - - i++; - dst = mem_map_next(dst, dst_base, i); - src = mem_map_next(src, src_base, i); - } -} - -static void copy_user_huge_page(struct page *dst, struct page *src, - unsigned long addr, struct vm_area_struct *vma) -{ - int i; - struct hstate *h = hstate_vma(vma); - - if (unlikely(pages_per_huge_page(h) > MAX_ORDER_NR_PAGES)) { - copy_user_gigantic_page(dst, src, addr, vma); - return; - } - - might_sleep(); - for (i = 0; i < pages_per_huge_page(h); i++) { - cond_resched(); - copy_user_highpage(dst + i, src + i, addr + i*PAGE_SIZE, vma); - } -} - static void copy_gigantic_page(struct page *dst, struct page *src) { int i; @@ -2454,7 +2389,8 @@ retry_avoidcopy: return VM_FAULT_OOM; } - copy_user_huge_page(new_page, old_page, address, vma); + copy_user_huge_page(new_page, old_page, address, vma, + pages_per_huge_page(h)); __SetPageUptodate(new_page); /* @@ -2558,7 +2494,7 @@ retry: ret = -PTR_ERR(page); goto out; } - clear_huge_page(page, address, huge_page_size(h)); + clear_huge_page(page, address, pages_per_huge_page(h)); __SetPageUptodate(page); if (vma->vm_flags & VM_MAYSHARE) { diff --git a/mm/memory.c b/mm/memory.c index 567bca80ea53..60e1c68d8218 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -3645,3 +3645,74 @@ void might_fault(void) } EXPORT_SYMBOL(might_fault); #endif + +#if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_HUGETLBFS) +static void clear_gigantic_page(struct page *page, + unsigned long addr, + unsigned int pages_per_huge_page) +{ + int i; + struct page *p = page; + + might_sleep(); + for (i = 0; i < pages_per_huge_page; + i++, p = mem_map_next(p, page, i)) { + cond_resched(); + clear_user_highpage(p, addr + i * PAGE_SIZE); + } +} +void clear_huge_page(struct page *page, + unsigned long addr, unsigned int pages_per_huge_page) +{ + int i; + + if (unlikely(pages_per_huge_page > MAX_ORDER_NR_PAGES)) { + clear_gigantic_page(page, addr, pages_per_huge_page); + return; + } + + might_sleep(); + for (i = 0; i < pages_per_huge_page; i++) { + cond_resched(); + clear_user_highpage(page + i, addr + i * PAGE_SIZE); + } +} + +static void copy_user_gigantic_page(struct page *dst, struct page *src, + unsigned long addr, + struct vm_area_struct *vma, + unsigned int pages_per_huge_page) +{ + int i; + struct page *dst_base = dst; + struct page *src_base = src; + + for (i = 0; i < pages_per_huge_page; ) { + cond_resched(); + copy_user_highpage(dst, src, addr + i*PAGE_SIZE, vma); + + i++; + dst = mem_map_next(dst, dst_base, i); + src = mem_map_next(src, src_base, i); + } +} + +void copy_user_huge_page(struct page *dst, struct page *src, + unsigned long addr, struct vm_area_struct *vma, + unsigned int pages_per_huge_page) +{ + int i; + + if (unlikely(pages_per_huge_page > MAX_ORDER_NR_PAGES)) { + copy_user_gigantic_page(dst, src, addr, vma, + pages_per_huge_page); + return; + } + + might_sleep(); + for (i = 0; i < pages_per_huge_page; i++) { + cond_resched(); + copy_user_highpage(dst + i, src + i, addr + i*PAGE_SIZE, vma); + } +} +#endif /* CONFIG_TRANSPARENT_HUGEPAGE || CONFIG_HUGETLBFS */ -- cgit v1.2.3 From 936a5fe6e6148c0b3ea0d792b903847d9b9931a1 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:46:48 -0800 Subject: thp: kvm mmu transparent hugepage support This should work for both hugetlbfs and transparent hugepages. [akpm@linux-foundation.org: bring forward PageTransCompound() addition for bisectability] Signed-off-by: Andrea Arcangeli Cc: Avi Kivity Cc: Marcelo Tosatti Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/kvm/mmu.c | 91 ++++++++++++++++++++++++++++++++++++++-------- arch/x86/kvm/paging_tmpl.h | 9 ++++- include/linux/page-flags.h | 12 ++++++ virt/kvm/kvm_main.c | 32 +++++++++++++++- 4 files changed, 125 insertions(+), 19 deletions(-) (limited to 'include/linux') diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 9cafbb499813..47b2c3288b6b 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -554,14 +554,18 @@ static int host_mapping_level(struct kvm *kvm, gfn_t gfn) return ret; } -static int mapping_level(struct kvm_vcpu *vcpu, gfn_t large_gfn) +static bool mapping_level_dirty_bitmap(struct kvm_vcpu *vcpu, gfn_t large_gfn) { struct kvm_memory_slot *slot; - int host_level, level, max_level; - slot = gfn_to_memslot(vcpu->kvm, large_gfn); if (slot && slot->dirty_bitmap) - return PT_PAGE_TABLE_LEVEL; + return true; + return false; +} + +static int mapping_level(struct kvm_vcpu *vcpu, gfn_t large_gfn) +{ + int host_level, level, max_level; host_level = host_mapping_level(vcpu->kvm, large_gfn); @@ -2281,6 +2285,48 @@ static int kvm_handle_bad_page(struct kvm *kvm, gfn_t gfn, pfn_t pfn) return 1; } +static void transparent_hugepage_adjust(struct kvm_vcpu *vcpu, + gfn_t *gfnp, pfn_t *pfnp, int *levelp) +{ + pfn_t pfn = *pfnp; + gfn_t gfn = *gfnp; + int level = *levelp; + + /* + * Check if it's a transparent hugepage. If this would be an + * hugetlbfs page, level wouldn't be set to + * PT_PAGE_TABLE_LEVEL and there would be no adjustment done + * here. + */ + if (!is_error_pfn(pfn) && !kvm_is_mmio_pfn(pfn) && + level == PT_PAGE_TABLE_LEVEL && + PageTransCompound(pfn_to_page(pfn)) && + !has_wrprotected_page(vcpu->kvm, gfn, PT_DIRECTORY_LEVEL)) { + unsigned long mask; + /* + * mmu_notifier_retry was successful and we hold the + * mmu_lock here, so the pmd can't become splitting + * from under us, and in turn + * __split_huge_page_refcount() can't run from under + * us and we can safely transfer the refcount from + * PG_tail to PG_head as we switch the pfn to tail to + * head. + */ + *levelp = level = PT_DIRECTORY_LEVEL; + mask = KVM_PAGES_PER_HPAGE(level) - 1; + VM_BUG_ON((gfn & mask) != (pfn & mask)); + if (pfn & mask) { + gfn &= ~mask; + *gfnp = gfn; + kvm_release_pfn_clean(pfn); + pfn &= ~mask; + if (!get_page_unless_zero(pfn_to_page(pfn))) + BUG(); + *pfnp = pfn; + } + } +} + static bool try_async_pf(struct kvm_vcpu *vcpu, bool prefault, gfn_t gfn, gva_t gva, pfn_t *pfn, bool write, bool *writable); @@ -2289,20 +2335,25 @@ static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn, { int r; int level; + int force_pt_level; pfn_t pfn; unsigned long mmu_seq; bool map_writable; - level = mapping_level(vcpu, gfn); - - /* - * This path builds a PAE pagetable - so we can map 2mb pages at - * maximum. Therefore check if the level is larger than that. - */ - if (level > PT_DIRECTORY_LEVEL) - level = PT_DIRECTORY_LEVEL; + force_pt_level = mapping_level_dirty_bitmap(vcpu, gfn); + if (likely(!force_pt_level)) { + level = mapping_level(vcpu, gfn); + /* + * This path builds a PAE pagetable - so we can map + * 2mb pages at maximum. Therefore check if the level + * is larger than that. + */ + if (level > PT_DIRECTORY_LEVEL) + level = PT_DIRECTORY_LEVEL; - gfn &= ~(KVM_PAGES_PER_HPAGE(level) - 1); + gfn &= ~(KVM_PAGES_PER_HPAGE(level) - 1); + } else + level = PT_PAGE_TABLE_LEVEL; mmu_seq = vcpu->kvm->mmu_notifier_seq; smp_rmb(); @@ -2318,6 +2369,8 @@ static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn, if (mmu_notifier_retry(vcpu, mmu_seq)) goto out_unlock; kvm_mmu_free_some_pages(vcpu); + if (likely(!force_pt_level)) + transparent_hugepage_adjust(vcpu, &gfn, &pfn, &level); r = __direct_map(vcpu, v, write, map_writable, level, gfn, pfn, prefault); spin_unlock(&vcpu->kvm->mmu_lock); @@ -2655,6 +2708,7 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, u32 error_code, pfn_t pfn; int r; int level; + int force_pt_level; gfn_t gfn = gpa >> PAGE_SHIFT; unsigned long mmu_seq; int write = error_code & PFERR_WRITE_MASK; @@ -2667,9 +2721,12 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, u32 error_code, if (r) return r; - level = mapping_level(vcpu, gfn); - - gfn &= ~(KVM_PAGES_PER_HPAGE(level) - 1); + force_pt_level = mapping_level_dirty_bitmap(vcpu, gfn); + if (likely(!force_pt_level)) { + level = mapping_level(vcpu, gfn); + gfn &= ~(KVM_PAGES_PER_HPAGE(level) - 1); + } else + level = PT_PAGE_TABLE_LEVEL; mmu_seq = vcpu->kvm->mmu_notifier_seq; smp_rmb(); @@ -2684,6 +2741,8 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, u32 error_code, if (mmu_notifier_retry(vcpu, mmu_seq)) goto out_unlock; kvm_mmu_free_some_pages(vcpu); + if (likely(!force_pt_level)) + transparent_hugepage_adjust(vcpu, &gfn, &pfn, &level); r = __direct_map(vcpu, gpa, write, map_writable, level, gfn, pfn, prefault); spin_unlock(&vcpu->kvm->mmu_lock); diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h index 53210f1e94c2..6bccc24c4181 100644 --- a/arch/x86/kvm/paging_tmpl.h +++ b/arch/x86/kvm/paging_tmpl.h @@ -550,6 +550,7 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code, int r; pfn_t pfn; int level = PT_PAGE_TABLE_LEVEL; + int force_pt_level; unsigned long mmu_seq; bool map_writable; @@ -577,7 +578,11 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code, return 0; } - if (walker.level >= PT_DIRECTORY_LEVEL) { + if (walker.level >= PT_DIRECTORY_LEVEL) + force_pt_level = mapping_level_dirty_bitmap(vcpu, walker.gfn); + else + force_pt_level = 1; + if (!force_pt_level) { level = min(walker.level, mapping_level(vcpu, walker.gfn)); walker.gfn = walker.gfn & ~(KVM_PAGES_PER_HPAGE(level) - 1); } @@ -599,6 +604,8 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code, trace_kvm_mmu_audit(vcpu, AUDIT_PRE_PAGE_FAULT); kvm_mmu_free_some_pages(vcpu); + if (!force_pt_level) + transparent_hugepage_adjust(vcpu, &walker.gfn, &pfn, &level); sptep = FNAME(fetch)(vcpu, addr, &walker, user_fault, write_fault, level, &write_pt, pfn, map_writable, prefault); (void)sptep; diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index 4835cae71047..907f1605926b 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -409,6 +409,18 @@ static inline void ClearPageCompound(struct page *page) #endif /* !PAGEFLAGS_EXTENDED */ +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +static inline int PageTransCompound(struct page *page) +{ + return PageCompound(page); +} +#else +static inline int PageTransCompound(struct page *page) +{ + return 0; +} +#endif + #ifdef CONFIG_MMU #define __PG_MLOCKED (1 << PG_mlocked) #else diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 7f686251f711..85ab7db0d366 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -104,8 +104,36 @@ static pfn_t fault_pfn; inline int kvm_is_mmio_pfn(pfn_t pfn) { if (pfn_valid(pfn)) { - struct page *page = compound_head(pfn_to_page(pfn)); - return PageReserved(page); + struct page *head; + struct page *tail = pfn_to_page(pfn); + head = compound_head(tail); + if (head != tail) { + smp_rmb(); + /* + * head may be a dangling pointer. + * __split_huge_page_refcount clears PageTail + * before overwriting first_page, so if + * PageTail is still there it means the head + * pointer isn't dangling. + */ + if (PageTail(tail)) { + /* + * the "head" is not a dangling + * pointer but the hugepage may have + * been splitted from under us (and we + * may not hold a reference count on + * the head page so it can be reused + * before we run PageReferenced), so + * we've to recheck PageTail before + * returning what we just read. + */ + int reserved = PageReserved(head); + smp_rmb(); + if (PageTail(tail)) + return reserved; + } + } + return PageReserved(tail); } return true; -- cgit v1.2.3 From 32dba98e085f8b2b4345887df9abf5e0e93bfc12 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:46:49 -0800 Subject: thp: _GFP_NO_KSWAPD Transparent hugepage allocations must be allowed not to invoke kswapd or any other kind of indirect reclaim (especially when the defrag sysfs is control disabled). It's unacceptable to swap out anonymous pages (potentially anonymous transparent hugepages) in order to create new transparent hugepages. This is true for the MADV_HUGEPAGE areas too (swapping out a kvm virtual machine and so having it suffer an unbearable slowdown, so another one with guest physical memory marked MADV_HUGEPAGE can run 30% faster if it is running memory intensive workloads, makes no sense). If a transparent hugepage allocation fails the slowdown is minor and there is total fallback, so kswapd should never be asked to swapout memory to allow the high order allocation to succeed. Signed-off-by: Andrea Arcangeli Acked-by: Rik van Riel Acked-by: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/gfp.h | 5 ++++- mm/page_alloc.c | 3 ++- 2 files changed, 6 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/gfp.h b/include/linux/gfp.h index f54adfcbec9c..49d2356bb82d 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -34,6 +34,7 @@ struct vm_area_struct; #else #define ___GFP_NOTRACK 0 #endif +#define ___GFP_NO_KSWAPD 0x400000u /* * GFP bitmasks.. @@ -81,13 +82,15 @@ struct vm_area_struct; #define __GFP_RECLAIMABLE ((__force gfp_t)___GFP_RECLAIMABLE) /* Page is reclaimable */ #define __GFP_NOTRACK ((__force gfp_t)___GFP_NOTRACK) /* Don't track with kmemcheck */ +#define __GFP_NO_KSWAPD ((__force gfp_t)___GFP_NO_KSWAPD) + /* * This may seem redundant, but it's a way of annotating false positives vs. * allocations that simply cannot be supported (e.g. page tables). */ #define __GFP_NOTRACK_FALSE_POSITIVE (__GFP_NOTRACK) -#define __GFP_BITS_SHIFT 22 /* Room for 22 __GFP_FOO bits */ +#define __GFP_BITS_SHIFT 23 /* Room for 23 __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 */ diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 6e62d5f9d40b..bbd0423f2820 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2027,7 +2027,8 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order, goto nopage; restart: - wake_all_kswapd(order, zonelist, high_zoneidx, + if (!(gfp_mask & __GFP_NO_KSWAPD)) + wake_all_kswapd(order, zonelist, high_zoneidx, zone_idx(preferred_zone)); /* -- cgit v1.2.3 From 71e3aac0724ffe8918992d76acfe3aad7d8724a5 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:46:52 -0800 Subject: thp: transparent hugepage core Lately I've been working to make KVM use hugepages transparently without the usual restrictions of hugetlbfs. Some of the restrictions I'd like to see removed: 1) hugepages have to be swappable or the guest physical memory remains locked in RAM and can't be paged out to swap 2) if a hugepage allocation fails, regular pages should be allocated instead and mixed in the same vma without any failure and without userland noticing 3) if some task quits and more hugepages become available in the buddy, guest physical memory backed by regular pages should be relocated on hugepages automatically in regions under madvise(MADV_HUGEPAGE) (ideally event driven by waking up the kernel deamon if the order=HPAGE_PMD_SHIFT-PAGE_SHIFT list becomes not null) 4) avoidance of reservation and maximization of use of hugepages whenever possible. Reservation (needed to avoid runtime fatal faliures) may be ok for 1 machine with 1 database with 1 database cache with 1 database cache size known at boot time. It's definitely not feasible with a virtualization hypervisor usage like RHEV-H that runs an unknown number of virtual machines with an unknown size of each virtual machine with an unknown amount of pagecache that could be potentially useful in the host for guest not using O_DIRECT (aka cache=off). hugepages in the virtualization hypervisor (and also in the guest!) are much more important than in a regular host not using virtualization, becasue with NPT/EPT they decrease the tlb-miss cacheline accesses from 24 to 19 in case only the hypervisor uses transparent hugepages, and they decrease the tlb-miss cacheline accesses from 19 to 15 in case both the linux hypervisor and the linux guest both uses this patch (though the guest will limit the addition speedup to anonymous regions only for now...). Even more important is that the tlb miss handler is much slower on a NPT/EPT guest than for a regular shadow paging or no-virtualization scenario. So maximizing the amount of virtual memory cached by the TLB pays off significantly more with NPT/EPT than without (even if there would be no significant speedup in the tlb-miss runtime). The first (and more tedious) part of this work requires allowing the VM to handle anonymous hugepages mixed with regular pages transparently on regular anonymous vmas. This is what this patch tries to achieve in the least intrusive possible way. We want hugepages and hugetlb to be used in a way so that all applications can benefit without changes (as usual we leverage the KVM virtualization design: by improving the Linux VM at large, KVM gets the performance boost too). The most important design choice is: always fallback to 4k allocation if the hugepage allocation fails! This is the _very_ opposite of some large pagecache patches that failed with -EIO back then if a 64k (or similar) allocation failed... Second important decision (to reduce the impact of the feature on the existing pagetable handling code) is that at any time we can split an hugepage into 512 regular pages and it has to be done with an operation that can't fail. This way the reliability of the swapping isn't decreased (no need to allocate memory when we are short on memory to swap) and it's trivial to plug a split_huge_page* one-liner where needed without polluting the VM. Over time we can teach mprotect, mremap and friends to handle pmd_trans_huge natively without calling split_huge_page*. The fact it can't fail isn't just for swap: if split_huge_page would return -ENOMEM (instead of the current void) we'd need to rollback the mprotect from the middle of it (ideally including undoing the split_vma) which would be a big change and in the very wrong direction (it'd likely be simpler not to call split_huge_page at all and to teach mprotect and friends to handle hugepages instead of rolling them back from the middle). In short the very value of split_huge_page is that it can't fail. The collapsing and madvise(MADV_HUGEPAGE) part will remain separated and incremental and it'll just be an "harmless" addition later if this initial part is agreed upon. It also should be noted that locking-wise replacing regular pages with hugepages is going to be very easy if compared to what I'm doing below in split_huge_page, as it will only happen when page_count(page) matches page_mapcount(page) if we can take the PG_lock and mmap_sem in write mode. collapse_huge_page will be a "best effort" that (unlike split_huge_page) can fail at the minimal sign of trouble and we can try again later. collapse_huge_page will be similar to how KSM works and the madvise(MADV_HUGEPAGE) will work similar to madvise(MADV_MERGEABLE). The default I like is that transparent hugepages are used at page fault time. This can be changed with /sys/kernel/mm/transparent_hugepage/enabled. The control knob can be set to three values "always", "madvise", "never" which mean respectively that hugepages are always used, or only inside madvise(MADV_HUGEPAGE) regions, or never used. /sys/kernel/mm/transparent_hugepage/defrag instead controls if the hugepage allocation should defrag memory aggressively "always", only inside "madvise" regions, or "never". The pmd_trans_splitting/pmd_trans_huge locking is very solid. The put_page (from get_user_page users that can't use mmu notifier like O_DIRECT) that runs against a __split_huge_page_refcount instead was a pain to serialize in a way that would result always in a coherent page count for both tail and head. I think my locking solution with a compound_lock taken only after the page_first is valid and is still a PageHead should be safe but it surely needs review from SMP race point of view. In short there is no current existing way to serialize the O_DIRECT final put_page against split_huge_page_refcount so I had to invent a new one (O_DIRECT loses knowledge on the mapping status by the time gup_fast returns so...). And I didn't want to impact all gup/gup_fast users for now, maybe if we change the gup interface substantially we can avoid this locking, I admit I didn't think too much about it because changing the gup unpinning interface would be invasive. If we ignored O_DIRECT we could stick to the existing compound refcounting code, by simply adding a get_user_pages_fast_flags(foll_flags) where KVM (and any other mmu notifier user) would call it without FOLL_GET (and if FOLL_GET isn't set we'd just BUG_ON if nobody registered itself in the current task mmu notifier list yet). But O_DIRECT is fundamental for decent performance of virtualized I/O on fast storage so we can't avoid it to solve the race of put_page against split_huge_page_refcount to achieve a complete hugepage feature for KVM. Swap and oom works fine (well just like with regular pages ;). MMU notifier is handled transparently too, with the exception of the young bit on the pmd, that didn't have a range check but I think KVM will be fine because the whole point of hugepages is that EPT/NPT will also use a huge pmd when they notice gup returns pages with PageCompound set, so they won't care of a range and there's just the pmd young bit to check in that case. NOTE: in some cases if the L2 cache is small, this may slowdown and waste memory during COWs because 4M of memory are accessed in a single fault instead of 8k (the payoff is that after COW the program can run faster). So we might want to switch the copy_huge_page (and clear_huge_page too) to not temporal stores. I also extensively researched ways to avoid this cache trashing with a full prefault logic that would cow in 8k/16k/32k/64k up to 1M (I can send those patches that fully implemented prefault) but I concluded they're not worth it and they add an huge additional complexity and they remove all tlb benefits until the full hugepage has been faulted in, to save a little bit of memory and some cache during app startup, but they still don't improve substantially the cache-trashing during startup if the prefault happens in >4k chunks. One reason is that those 4k pte entries copied are still mapped on a perfectly cache-colored hugepage, so the trashing is the worst one can generate in those copies (cow of 4k page copies aren't so well colored so they trashes less, but again this results in software running faster after the page fault). Those prefault patches allowed things like a pte where post-cow pages were local 4k regular anon pages and the not-yet-cowed pte entries were pointing in the middle of some hugepage mapped read-only. If it doesn't payoff substantially with todays hardware it will payoff even less in the future with larger l2 caches, and the prefault logic would blot the VM a lot. If one is emebdded transparent_hugepage can be disabled during boot with sysfs or with the boot commandline parameter transparent_hugepage=0 (or transparent_hugepage=2 to restrict hugepages inside madvise regions) that will ensure not a single hugepage is allocated at boot time. It is simple enough to just disable transparent hugepage globally and let transparent hugepages be allocated selectively by applications in the MADV_HUGEPAGE region (both at page fault time, and if enabled with the collapse_huge_page too through the kernel daemon). This patch supports only hugepages mapped in the pmd, archs that have smaller hugepages will not fit in this patch alone. Also some archs like power have certain tlb limits that prevents mixing different page size in the same regions so they will not fit in this framework that requires "graceful fallback" to basic PAGE_SIZE in case of physical memory fragmentation. hugetlbfs remains a perfect fit for those because its software limits happen to match the hardware limits. hugetlbfs also remains a perfect fit for hugepage sizes like 1GByte that cannot be hoped to be found not fragmented after a certain system uptime and that would be very expensive to defragment with relocation, so requiring reservation. hugetlbfs is the "reservation way", the point of transparent hugepages is not to have any reservation at all and maximizing the use of cache and hugepages at all times automatically. Some performance result: vmx andrea # LD_PRELOAD=/usr/lib64/libhugetlbfs.so HUGETLB_MORECORE=yes HUGETLB_PATH=/mnt/huge/ ./largep ages3 memset page fault 1566023 memset tlb miss 453854 memset second tlb miss 453321 random access tlb miss 41635 random access second tlb miss 41658 vmx andrea # LD_PRELOAD=/usr/lib64/libhugetlbfs.so HUGETLB_MORECORE=yes HUGETLB_PATH=/mnt/huge/ ./largepages3 memset page fault 1566471 memset tlb miss 453375 memset second tlb miss 453320 random access tlb miss 41636 random access second tlb miss 41637 vmx andrea # ./largepages3 memset page fault 1566642 memset tlb miss 453417 memset second tlb miss 453313 random access tlb miss 41630 random access second tlb miss 41647 vmx andrea # ./largepages3 memset page fault 1566872 memset tlb miss 453418 memset second tlb miss 453315 random access tlb miss 41618 random access second tlb miss 41659 vmx andrea # echo 0 > /proc/sys/vm/transparent_hugepage vmx andrea # ./largepages3 memset page fault 2182476 memset tlb miss 460305 memset second tlb miss 460179 random access tlb miss 44483 random access second tlb miss 44186 vmx andrea # ./largepages3 memset page fault 2182791 memset tlb miss 460742 memset second tlb miss 459962 random access tlb miss 43981 random access second tlb miss 43988 ============ #include #include #include #include #define SIZE (3UL*1024*1024*1024) int main() { char *p = malloc(SIZE), *p2; struct timeval before, after; gettimeofday(&before, NULL); memset(p, 0, SIZE); gettimeofday(&after, NULL); printf("memset page fault %Lu\n", (after.tv_sec-before.tv_sec)*1000000UL + after.tv_usec-before.tv_usec); gettimeofday(&before, NULL); memset(p, 0, SIZE); gettimeofday(&after, NULL); printf("memset tlb miss %Lu\n", (after.tv_sec-before.tv_sec)*1000000UL + after.tv_usec-before.tv_usec); gettimeofday(&before, NULL); memset(p, 0, SIZE); gettimeofday(&after, NULL); printf("memset second tlb miss %Lu\n", (after.tv_sec-before.tv_sec)*1000000UL + after.tv_usec-before.tv_usec); gettimeofday(&before, NULL); for (p2 = p; p2 < p+SIZE; p2 += 4096) *p2 = 0; gettimeofday(&after, NULL); printf("random access tlb miss %Lu\n", (after.tv_sec-before.tv_sec)*1000000UL + after.tv_usec-before.tv_usec); gettimeofday(&before, NULL); for (p2 = p; p2 < p+SIZE; p2 += 4096) *p2 = 0; gettimeofday(&after, NULL); printf("random access second tlb miss %Lu\n", (after.tv_sec-before.tv_sec)*1000000UL + after.tv_usec-before.tv_usec); return 0; } ============ Signed-off-by: Andrea Arcangeli Acked-by: Rik van Riel Signed-off-by: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/include/asm/pgtable_64.h | 5 + include/linux/gfp.h | 3 + include/linux/huge_mm.h | 118 +++++ include/linux/mm.h | 4 + include/linux/mm_inline.h | 11 +- include/linux/page-flags.h | 21 + include/linux/rmap.h | 2 + include/linux/swap.h | 2 + mm/Makefile | 1 + mm/huge_memory.c | 901 ++++++++++++++++++++++++++++++++++++++ mm/internal.h | 4 + mm/memory.c | 84 +++- mm/rmap.c | 62 ++- mm/swap.c | 37 ++ 14 files changed, 1220 insertions(+), 35 deletions(-) create mode 100644 include/linux/huge_mm.h create mode 100644 mm/huge_memory.c (limited to 'include/linux') diff --git a/arch/x86/include/asm/pgtable_64.h b/arch/x86/include/asm/pgtable_64.h index 1fb61a74b2e1..b2df039a4119 100644 --- a/arch/x86/include/asm/pgtable_64.h +++ b/arch/x86/include/asm/pgtable_64.h @@ -286,6 +286,11 @@ static inline pmd_t pmd_mkwrite(pmd_t pmd) return pmd_set_flags(pmd, _PAGE_RW); } +static inline pmd_t pmd_mknotpresent(pmd_t pmd) +{ + return pmd_clear_flags(pmd, _PAGE_PRESENT); +} + #endif /* !__ASSEMBLY__ */ #endif /* _ASM_X86_PGTABLE_64_H */ diff --git a/include/linux/gfp.h b/include/linux/gfp.h index 49d2356bb82d..d95082cc6f4a 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -109,6 +109,9 @@ struct vm_area_struct; __GFP_HARDWALL | __GFP_HIGHMEM | \ __GFP_MOVABLE) #define GFP_IOFS (__GFP_IO | __GFP_FS) +#define GFP_TRANSHUGE (GFP_HIGHUSER_MOVABLE | __GFP_COMP | \ + __GFP_NOMEMALLOC | __GFP_NORETRY | __GFP_NOWARN | \ + __GFP_NO_KSWAPD) #ifdef CONFIG_NUMA #define GFP_THISNODE (__GFP_THISNODE | __GFP_NOWARN | __GFP_NORETRY) diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h new file mode 100644 index 000000000000..9301824c7491 --- /dev/null +++ b/include/linux/huge_mm.h @@ -0,0 +1,118 @@ +#ifndef _LINUX_HUGE_MM_H +#define _LINUX_HUGE_MM_H + +extern int do_huge_pmd_anonymous_page(struct mm_struct *mm, + struct vm_area_struct *vma, + unsigned long address, pmd_t *pmd, + unsigned int flags); +extern int copy_huge_pmd(struct mm_struct *dst_mm, struct mm_struct *src_mm, + pmd_t *dst_pmd, pmd_t *src_pmd, unsigned long addr, + struct vm_area_struct *vma); +extern int do_huge_pmd_wp_page(struct mm_struct *mm, struct vm_area_struct *vma, + unsigned long address, pmd_t *pmd, + pmd_t orig_pmd); +extern pgtable_t get_pmd_huge_pte(struct mm_struct *mm); +extern struct page *follow_trans_huge_pmd(struct mm_struct *mm, + unsigned long addr, + pmd_t *pmd, + unsigned int flags); +extern int zap_huge_pmd(struct mmu_gather *tlb, + struct vm_area_struct *vma, + pmd_t *pmd); + +enum transparent_hugepage_flag { + TRANSPARENT_HUGEPAGE_FLAG, + TRANSPARENT_HUGEPAGE_REQ_MADV_FLAG, + TRANSPARENT_HUGEPAGE_DEFRAG_FLAG, + TRANSPARENT_HUGEPAGE_DEFRAG_REQ_MADV_FLAG, +#ifdef CONFIG_DEBUG_VM + TRANSPARENT_HUGEPAGE_DEBUG_COW_FLAG, +#endif +}; + +enum page_check_address_pmd_flag { + PAGE_CHECK_ADDRESS_PMD_FLAG, + PAGE_CHECK_ADDRESS_PMD_NOTSPLITTING_FLAG, + PAGE_CHECK_ADDRESS_PMD_SPLITTING_FLAG, +}; +extern pmd_t *page_check_address_pmd(struct page *page, + struct mm_struct *mm, + unsigned long address, + enum page_check_address_pmd_flag flag); + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +#define HPAGE_PMD_SHIFT HPAGE_SHIFT +#define HPAGE_PMD_MASK HPAGE_MASK +#define HPAGE_PMD_SIZE HPAGE_SIZE + +#define transparent_hugepage_enabled(__vma) \ + (transparent_hugepage_flags & (1<vm_flags & VM_HUGEPAGE)) +#define transparent_hugepage_defrag(__vma) \ + ((transparent_hugepage_flags & \ + (1<vm_flags & VM_HUGEPAGE)) +#ifdef CONFIG_DEBUG_VM +#define transparent_hugepage_debug_cow() \ + (transparent_hugepage_flags & \ + (1<root->lock); \ + /* \ + * spin_unlock_wait() is just a loop in C and so the \ + * CPU can reorder anything around it. \ + */ \ + smp_mb(); \ + BUG_ON(pmd_trans_splitting(*____pmd) || \ + pmd_trans_huge(*____pmd)); \ + } while (0) +#define HPAGE_PMD_ORDER (HPAGE_PMD_SHIFT-PAGE_SHIFT) +#define HPAGE_PMD_NR (1< MAX_ORDER +#error "hugepages can't be allocated by the buddy allocator" +#endif +#else /* CONFIG_TRANSPARENT_HUGEPAGE */ +#define HPAGE_PMD_SHIFT ({ BUG(); 0; }) +#define HPAGE_PMD_MASK ({ BUG(); 0; }) +#define HPAGE_PMD_SIZE ({ BUG(); 0; }) + +#define transparent_hugepage_enabled(__vma) 0 + +#define transparent_hugepage_flags 0UL +static inline int split_huge_page(struct page *page) +{ + return 0; +} +#define split_huge_page_pmd(__mm, __pmd) \ + do { } while (0) +#define wait_split_huge_page(__anon_vma, __pmd) \ + do { } while (0) +#endif /* CONFIG_TRANSPARENT_HUGEPAGE */ + +#endif /* _LINUX_HUGE_MM_H */ diff --git a/include/linux/mm.h b/include/linux/mm.h index cc6ab1038f6f..78adec4ba9f4 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -111,6 +111,9 @@ extern unsigned int kobjsize(const void *objp); #define VM_SAO 0x20000000 /* Strong Access Ordering (powerpc) */ #define VM_PFN_AT_MMAP 0x40000000 /* PFNMAP vma that is fully mapped at mmap time */ #define VM_MERGEABLE 0x80000000 /* KSM may merge identical pages */ +#if BITS_PER_LONG > 32 +#define VM_HUGEPAGE 0x100000000UL /* MADV_HUGEPAGE marked this vma */ +#endif /* Bits set in the VMA until the stack is in its final location */ #define VM_STACK_INCOMPLETE_SETUP (VM_RAND_READ | VM_SEQ_READ) @@ -243,6 +246,7 @@ struct inode; * files which need it (119 of them) */ #include +#include /* * Methods to modify the page usage count. diff --git a/include/linux/mm_inline.h b/include/linux/mm_inline.h index 8835b877b8db..650f31eabdb1 100644 --- a/include/linux/mm_inline.h +++ b/include/linux/mm_inline.h @@ -20,13 +20,20 @@ static inline int page_is_file_cache(struct page *page) } static inline void -add_page_to_lru_list(struct zone *zone, struct page *page, enum lru_list l) +__add_page_to_lru_list(struct zone *zone, struct page *page, enum lru_list l, + struct list_head *head) { - list_add(&page->lru, &zone->lru[l].list); + list_add(&page->lru, head); __inc_zone_state(zone, NR_LRU_BASE + l); mem_cgroup_add_lru_list(page, l); } +static inline void +add_page_to_lru_list(struct zone *zone, struct page *page, enum lru_list l) +{ + __add_page_to_lru_list(zone, page, l, &zone->lru[l].list); +} + static inline void del_page_from_lru_list(struct zone *zone, struct page *page, enum lru_list l) { diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index 907f1605926b..4ca1241ef94e 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -410,11 +410,32 @@ static inline void ClearPageCompound(struct page *page) #endif /* !PAGEFLAGS_EXTENDED */ #ifdef CONFIG_TRANSPARENT_HUGEPAGE +/* + * PageHuge() only returns true for hugetlbfs pages, but not for + * normal or transparent huge pages. + * + * PageTransHuge() returns true for both transparent huge and + * hugetlbfs pages, but not normal pages. PageTransHuge() can only be + * called only in the core VM paths where hugetlbfs pages can't exist. + */ +static inline int PageTransHuge(struct page *page) +{ + VM_BUG_ON(PageTail(page)); + return PageHead(page); +} + static inline int PageTransCompound(struct page *page) { return PageCompound(page); } + #else + +static inline int PageTransHuge(struct page *page) +{ + return 0; +} + static inline int PageTransCompound(struct page *page) { return 0; diff --git a/include/linux/rmap.h b/include/linux/rmap.h index bb83c0da2071..e9fd04ca1e51 100644 --- a/include/linux/rmap.h +++ b/include/linux/rmap.h @@ -198,6 +198,8 @@ enum ttu_flags { }; #define TTU_ACTION(x) ((x) & TTU_ACTION_MASK) +bool is_vma_temporary_stack(struct vm_area_struct *vma); + int try_to_unmap(struct page *, enum ttu_flags flags); int try_to_unmap_one(struct page *, struct vm_area_struct *, unsigned long address, enum ttu_flags flags); diff --git a/include/linux/swap.h b/include/linux/swap.h index eba53e71d2cc..4d559325d919 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -208,6 +208,8 @@ extern unsigned int nr_free_pagecache_pages(void); /* linux/mm/swap.c */ extern void __lru_cache_add(struct page *, enum lru_list lru); extern void lru_cache_add_lru(struct page *, enum lru_list lru); +extern void lru_add_page_tail(struct zone* zone, + struct page *page, struct page *page_tail); extern void activate_page(struct page *); extern void mark_page_accessed(struct page *); extern void lru_add_drain(void); diff --git a/mm/Makefile b/mm/Makefile index 380772a9ccb8..2b1b575ae712 100644 --- a/mm/Makefile +++ b/mm/Makefile @@ -37,6 +37,7 @@ obj-$(CONFIG_MEMORY_HOTPLUG) += memory_hotplug.o obj-$(CONFIG_FS_XIP) += filemap_xip.o obj-$(CONFIG_MIGRATION) += migrate.o obj-$(CONFIG_QUICKLIST) += quicklist.o +obj-$(CONFIG_TRANSPARENT_HUGEPAGE) += huge_memory.o obj-$(CONFIG_CGROUP_MEM_RES_CTLR) += memcontrol.o page_cgroup.o obj-$(CONFIG_MEMORY_FAILURE) += memory-failure.o obj-$(CONFIG_HWPOISON_INJECT) += hwpoison-inject.o diff --git a/mm/huge_memory.c b/mm/huge_memory.c new file mode 100644 index 000000000000..0c1e8f939f7c --- /dev/null +++ b/mm/huge_memory.c @@ -0,0 +1,901 @@ +/* + * Copyright (C) 2009 Red Hat, Inc. + * + * This work is licensed under the terms of the GNU GPL, version 2. See + * the COPYING file in the top-level directory. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "internal.h" + +unsigned long transparent_hugepage_flags __read_mostly = + (1<page_table_lock); + + /* FIFO */ + pgtable = mm->pmd_huge_pte; + if (list_empty(&pgtable->lru)) + mm->pmd_huge_pte = NULL; + else { + mm->pmd_huge_pte = list_entry(pgtable->lru.next, + struct page, lru); + list_del(&pgtable->lru); + } + return pgtable; +} + +static int do_huge_pmd_wp_page_fallback(struct mm_struct *mm, + struct vm_area_struct *vma, + unsigned long address, + pmd_t *pmd, pmd_t orig_pmd, + struct page *page, + unsigned long haddr) +{ + pgtable_t pgtable; + pmd_t _pmd; + int ret = 0, i; + struct page **pages; + + pages = kmalloc(sizeof(struct page *) * HPAGE_PMD_NR, + GFP_KERNEL); + if (unlikely(!pages)) { + ret |= VM_FAULT_OOM; + goto out; + } + + for (i = 0; i < HPAGE_PMD_NR; i++) { + pages[i] = alloc_page_vma(GFP_HIGHUSER_MOVABLE, + vma, address); + if (unlikely(!pages[i])) { + while (--i >= 0) + put_page(pages[i]); + kfree(pages); + ret |= VM_FAULT_OOM; + goto out; + } + } + + for (i = 0; i < HPAGE_PMD_NR; i++) { + copy_user_highpage(pages[i], page + i, + haddr + PAGE_SHIFT*i, vma); + __SetPageUptodate(pages[i]); + cond_resched(); + } + + spin_lock(&mm->page_table_lock); + if (unlikely(!pmd_same(*pmd, orig_pmd))) + goto out_free_pages; + VM_BUG_ON(!PageHead(page)); + + pmdp_clear_flush_notify(vma, haddr, pmd); + /* leave pmd empty until pte is filled */ + + pgtable = get_pmd_huge_pte(mm); + pmd_populate(mm, &_pmd, pgtable); + + for (i = 0; i < HPAGE_PMD_NR; i++, haddr += PAGE_SIZE) { + pte_t *pte, entry; + entry = mk_pte(pages[i], vma->vm_page_prot); + entry = maybe_mkwrite(pte_mkdirty(entry), vma); + page_add_new_anon_rmap(pages[i], vma, haddr); + pte = pte_offset_map(&_pmd, haddr); + VM_BUG_ON(!pte_none(*pte)); + set_pte_at(mm, haddr, pte, entry); + pte_unmap(pte); + } + kfree(pages); + + mm->nr_ptes++; + smp_wmb(); /* make pte visible before pmd */ + pmd_populate(mm, pmd, pgtable); + page_remove_rmap(page); + spin_unlock(&mm->page_table_lock); + + ret |= VM_FAULT_WRITE; + put_page(page); + +out: + return ret; + +out_free_pages: + spin_unlock(&mm->page_table_lock); + for (i = 0; i < HPAGE_PMD_NR; i++) + put_page(pages[i]); + kfree(pages); + goto out; +} + +int do_huge_pmd_wp_page(struct mm_struct *mm, struct vm_area_struct *vma, + unsigned long address, pmd_t *pmd, pmd_t orig_pmd) +{ + int ret = 0; + struct page *page, *new_page; + unsigned long haddr; + + VM_BUG_ON(!vma->anon_vma); + spin_lock(&mm->page_table_lock); + if (unlikely(!pmd_same(*pmd, orig_pmd))) + goto out_unlock; + + page = pmd_page(orig_pmd); + VM_BUG_ON(!PageCompound(page) || !PageHead(page)); + haddr = address & HPAGE_PMD_MASK; + if (page_mapcount(page) == 1) { + pmd_t entry; + entry = pmd_mkyoung(orig_pmd); + entry = maybe_pmd_mkwrite(pmd_mkdirty(entry), vma); + if (pmdp_set_access_flags(vma, haddr, pmd, entry, 1)) + update_mmu_cache(vma, address, entry); + ret |= VM_FAULT_WRITE; + goto out_unlock; + } + get_page(page); + spin_unlock(&mm->page_table_lock); + + if (transparent_hugepage_enabled(vma) && + !transparent_hugepage_debug_cow()) + new_page = alloc_hugepage(transparent_hugepage_defrag(vma)); + else + new_page = NULL; + + if (unlikely(!new_page)) { + ret = do_huge_pmd_wp_page_fallback(mm, vma, address, + pmd, orig_pmd, page, haddr); + put_page(page); + goto out; + } + + copy_user_huge_page(new_page, page, haddr, vma, HPAGE_PMD_NR); + __SetPageUptodate(new_page); + + spin_lock(&mm->page_table_lock); + put_page(page); + if (unlikely(!pmd_same(*pmd, orig_pmd))) + put_page(new_page); + else { + pmd_t entry; + VM_BUG_ON(!PageHead(page)); + entry = mk_pmd(new_page, vma->vm_page_prot); + entry = maybe_pmd_mkwrite(pmd_mkdirty(entry), vma); + entry = pmd_mkhuge(entry); + pmdp_clear_flush_notify(vma, haddr, pmd); + page_add_new_anon_rmap(new_page, vma, haddr); + set_pmd_at(mm, haddr, pmd, entry); + update_mmu_cache(vma, address, entry); + page_remove_rmap(page); + put_page(page); + ret |= VM_FAULT_WRITE; + } +out_unlock: + spin_unlock(&mm->page_table_lock); +out: + return ret; +} + +struct page *follow_trans_huge_pmd(struct mm_struct *mm, + unsigned long addr, + pmd_t *pmd, + unsigned int flags) +{ + struct page *page = NULL; + + assert_spin_locked(&mm->page_table_lock); + + if (flags & FOLL_WRITE && !pmd_write(*pmd)) + goto out; + + page = pmd_page(*pmd); + VM_BUG_ON(!PageHead(page)); + if (flags & FOLL_TOUCH) { + pmd_t _pmd; + /* + * We should set the dirty bit only for FOLL_WRITE but + * for now the dirty bit in the pmd is meaningless. + * And if the dirty bit will become meaningful and + * we'll only set it with FOLL_WRITE, an atomic + * set_bit will be required on the pmd to set the + * young bit, instead of the current set_pmd_at. + */ + _pmd = pmd_mkyoung(pmd_mkdirty(*pmd)); + set_pmd_at(mm, addr & HPAGE_PMD_MASK, pmd, _pmd); + } + page += (addr & ~HPAGE_PMD_MASK) >> PAGE_SHIFT; + VM_BUG_ON(!PageCompound(page)); + if (flags & FOLL_GET) + get_page(page); + +out: + return page; +} + +int zap_huge_pmd(struct mmu_gather *tlb, struct vm_area_struct *vma, + pmd_t *pmd) +{ + int ret = 0; + + spin_lock(&tlb->mm->page_table_lock); + if (likely(pmd_trans_huge(*pmd))) { + if (unlikely(pmd_trans_splitting(*pmd))) { + spin_unlock(&tlb->mm->page_table_lock); + wait_split_huge_page(vma->anon_vma, + pmd); + } else { + struct page *page; + pgtable_t pgtable; + pgtable = get_pmd_huge_pte(tlb->mm); + page = pmd_page(*pmd); + pmd_clear(pmd); + page_remove_rmap(page); + VM_BUG_ON(page_mapcount(page) < 0); + add_mm_counter(tlb->mm, MM_ANONPAGES, -HPAGE_PMD_NR); + VM_BUG_ON(!PageHead(page)); + spin_unlock(&tlb->mm->page_table_lock); + tlb_remove_page(tlb, page); + pte_free(tlb->mm, pgtable); + ret = 1; + } + } else + spin_unlock(&tlb->mm->page_table_lock); + + return ret; +} + +pmd_t *page_check_address_pmd(struct page *page, + struct mm_struct *mm, + unsigned long address, + enum page_check_address_pmd_flag flag) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd, *ret = NULL; + + if (address & ~HPAGE_PMD_MASK) + goto out; + + pgd = pgd_offset(mm, address); + if (!pgd_present(*pgd)) + goto out; + + pud = pud_offset(pgd, address); + if (!pud_present(*pud)) + goto out; + + pmd = pmd_offset(pud, address); + if (pmd_none(*pmd)) + goto out; + if (pmd_page(*pmd) != page) + goto out; + VM_BUG_ON(flag == PAGE_CHECK_ADDRESS_PMD_NOTSPLITTING_FLAG && + pmd_trans_splitting(*pmd)); + if (pmd_trans_huge(*pmd)) { + VM_BUG_ON(flag == PAGE_CHECK_ADDRESS_PMD_SPLITTING_FLAG && + !pmd_trans_splitting(*pmd)); + ret = pmd; + } +out: + return ret; +} + +static int __split_huge_page_splitting(struct page *page, + struct vm_area_struct *vma, + unsigned long address) +{ + struct mm_struct *mm = vma->vm_mm; + pmd_t *pmd; + int ret = 0; + + spin_lock(&mm->page_table_lock); + pmd = page_check_address_pmd(page, mm, address, + PAGE_CHECK_ADDRESS_PMD_NOTSPLITTING_FLAG); + if (pmd) { + /* + * We can't temporarily set the pmd to null in order + * to split it, the pmd must remain marked huge at all + * times or the VM won't take the pmd_trans_huge paths + * and it won't wait on the anon_vma->root->lock to + * serialize against split_huge_page*. + */ + pmdp_splitting_flush_notify(vma, address, pmd); + ret = 1; + } + spin_unlock(&mm->page_table_lock); + + return ret; +} + +static void __split_huge_page_refcount(struct page *page) +{ + int i; + unsigned long head_index = page->index; + struct zone *zone = page_zone(page); + + /* prevent PageLRU to go away from under us, and freeze lru stats */ + spin_lock_irq(&zone->lru_lock); + compound_lock(page); + + for (i = 1; i < HPAGE_PMD_NR; i++) { + struct page *page_tail = page + i; + + /* tail_page->_count cannot change */ + atomic_sub(atomic_read(&page_tail->_count), &page->_count); + BUG_ON(page_count(page) <= 0); + atomic_add(page_mapcount(page) + 1, &page_tail->_count); + BUG_ON(atomic_read(&page_tail->_count) <= 0); + + /* after clearing PageTail the gup refcount can be released */ + smp_mb(); + + page_tail->flags &= ~PAGE_FLAGS_CHECK_AT_PREP; + page_tail->flags |= (page->flags & + ((1L << PG_referenced) | + (1L << PG_swapbacked) | + (1L << PG_mlocked) | + (1L << PG_uptodate))); + page_tail->flags |= (1L << PG_dirty); + + /* + * 1) clear PageTail before overwriting first_page + * 2) clear PageTail before clearing PageHead for VM_BUG_ON + */ + smp_wmb(); + + /* + * __split_huge_page_splitting() already set the + * splitting bit in all pmd that could map this + * hugepage, that will ensure no CPU can alter the + * mapcount on the head page. The mapcount is only + * accounted in the head page and it has to be + * transferred to all tail pages in the below code. So + * for this code to be safe, the split the mapcount + * can't change. But that doesn't mean userland can't + * keep changing and reading the page contents while + * we transfer the mapcount, so the pmd splitting + * status is achieved setting a reserved bit in the + * pmd, not by clearing the present bit. + */ + BUG_ON(page_mapcount(page_tail)); + page_tail->_mapcount = page->_mapcount; + + BUG_ON(page_tail->mapping); + page_tail->mapping = page->mapping; + + page_tail->index = ++head_index; + + BUG_ON(!PageAnon(page_tail)); + BUG_ON(!PageUptodate(page_tail)); + BUG_ON(!PageDirty(page_tail)); + BUG_ON(!PageSwapBacked(page_tail)); + + lru_add_page_tail(zone, page, page_tail); + } + + ClearPageCompound(page); + compound_unlock(page); + spin_unlock_irq(&zone->lru_lock); + + for (i = 1; i < HPAGE_PMD_NR; i++) { + struct page *page_tail = page + i; + BUG_ON(page_count(page_tail) <= 0); + /* + * Tail pages may be freed if there wasn't any mapping + * like if add_to_swap() is running on a lru page that + * had its mapping zapped. And freeing these pages + * requires taking the lru_lock so we do the put_page + * of the tail pages after the split is complete. + */ + put_page(page_tail); + } + + /* + * Only the head page (now become a regular page) is required + * to be pinned by the caller. + */ + BUG_ON(page_count(page) <= 0); +} + +static int __split_huge_page_map(struct page *page, + struct vm_area_struct *vma, + unsigned long address) +{ + struct mm_struct *mm = vma->vm_mm; + pmd_t *pmd, _pmd; + int ret = 0, i; + pgtable_t pgtable; + unsigned long haddr; + + spin_lock(&mm->page_table_lock); + pmd = page_check_address_pmd(page, mm, address, + PAGE_CHECK_ADDRESS_PMD_SPLITTING_FLAG); + if (pmd) { + pgtable = get_pmd_huge_pte(mm); + pmd_populate(mm, &_pmd, pgtable); + + for (i = 0, haddr = address; i < HPAGE_PMD_NR; + i++, haddr += PAGE_SIZE) { + pte_t *pte, entry; + BUG_ON(PageCompound(page+i)); + entry = mk_pte(page + i, vma->vm_page_prot); + entry = maybe_mkwrite(pte_mkdirty(entry), vma); + if (!pmd_write(*pmd)) + entry = pte_wrprotect(entry); + else + BUG_ON(page_mapcount(page) != 1); + if (!pmd_young(*pmd)) + entry = pte_mkold(entry); + pte = pte_offset_map(&_pmd, haddr); + BUG_ON(!pte_none(*pte)); + set_pte_at(mm, haddr, pte, entry); + pte_unmap(pte); + } + + mm->nr_ptes++; + smp_wmb(); /* make pte visible before pmd */ + /* + * Up to this point the pmd is present and huge and + * userland has the whole access to the hugepage + * during the split (which happens in place). If we + * overwrite the pmd with the not-huge version + * pointing to the pte here (which of course we could + * if all CPUs were bug free), userland could trigger + * a small page size TLB miss on the small sized TLB + * while the hugepage TLB entry is still established + * in the huge TLB. Some CPU doesn't like that. See + * http://support.amd.com/us/Processor_TechDocs/41322.pdf, + * Erratum 383 on page 93. Intel should be safe but is + * also warns that it's only safe if the permission + * and cache attributes of the two entries loaded in + * the two TLB is identical (which should be the case + * here). But it is generally safer to never allow + * small and huge TLB entries for the same virtual + * address to be loaded simultaneously. So instead of + * doing "pmd_populate(); flush_tlb_range();" we first + * mark the current pmd notpresent (atomically because + * here the pmd_trans_huge and pmd_trans_splitting + * must remain set at all times on the pmd until the + * split is complete for this pmd), then we flush the + * SMP TLB and finally we write the non-huge version + * of the pmd entry with pmd_populate. + */ + set_pmd_at(mm, address, pmd, pmd_mknotpresent(*pmd)); + flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE); + pmd_populate(mm, pmd, pgtable); + ret = 1; + } + spin_unlock(&mm->page_table_lock); + + return ret; +} + +/* must be called with anon_vma->root->lock hold */ +static void __split_huge_page(struct page *page, + struct anon_vma *anon_vma) +{ + int mapcount, mapcount2; + struct anon_vma_chain *avc; + + BUG_ON(!PageHead(page)); + BUG_ON(PageTail(page)); + + mapcount = 0; + list_for_each_entry(avc, &anon_vma->head, same_anon_vma) { + struct vm_area_struct *vma = avc->vma; + unsigned long addr = vma_address(page, vma); + BUG_ON(is_vma_temporary_stack(vma)); + if (addr == -EFAULT) + continue; + mapcount += __split_huge_page_splitting(page, vma, addr); + } + BUG_ON(mapcount != page_mapcount(page)); + + __split_huge_page_refcount(page); + + mapcount2 = 0; + list_for_each_entry(avc, &anon_vma->head, same_anon_vma) { + struct vm_area_struct *vma = avc->vma; + unsigned long addr = vma_address(page, vma); + BUG_ON(is_vma_temporary_stack(vma)); + if (addr == -EFAULT) + continue; + mapcount2 += __split_huge_page_map(page, vma, addr); + } + BUG_ON(mapcount != mapcount2); +} + +int split_huge_page(struct page *page) +{ + struct anon_vma *anon_vma; + int ret = 1; + + BUG_ON(!PageAnon(page)); + anon_vma = page_lock_anon_vma(page); + if (!anon_vma) + goto out; + ret = 0; + if (!PageCompound(page)) + goto out_unlock; + + BUG_ON(!PageSwapBacked(page)); + __split_huge_page(page, anon_vma); + + BUG_ON(PageCompound(page)); +out_unlock: + page_unlock_anon_vma(anon_vma); +out: + return ret; +} + +void __split_huge_page_pmd(struct mm_struct *mm, pmd_t *pmd) +{ + struct page *page; + + spin_lock(&mm->page_table_lock); + if (unlikely(!pmd_trans_huge(*pmd))) { + spin_unlock(&mm->page_table_lock); + return; + } + page = pmd_page(*pmd); + VM_BUG_ON(!page_count(page)); + get_page(page); + spin_unlock(&mm->page_table_lock); + + split_huge_page(page); + + put_page(page); + BUG_ON(pmd_trans_huge(*pmd)); +} diff --git a/mm/internal.h b/mm/internal.h index bd4f581f624a..69488205723d 100644 --- a/mm/internal.h +++ b/mm/internal.h @@ -134,6 +134,10 @@ static inline void mlock_migrate_page(struct page *newpage, struct page *page) } } +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +extern unsigned long vma_address(struct page *page, + struct vm_area_struct *vma); +#endif #else /* !CONFIG_MMU */ static inline int is_mlocked_vma(struct vm_area_struct *v, struct page *p) { diff --git a/mm/memory.c b/mm/memory.c index 60e1c68d8218..c50a195041ec 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -726,9 +726,9 @@ out_set_pte: return 0; } -static int copy_pte_range(struct mm_struct *dst_mm, struct mm_struct *src_mm, - pmd_t *dst_pmd, pmd_t *src_pmd, struct vm_area_struct *vma, - unsigned long addr, unsigned long end) +int copy_pte_range(struct mm_struct *dst_mm, struct mm_struct *src_mm, + pmd_t *dst_pmd, pmd_t *src_pmd, struct vm_area_struct *vma, + unsigned long addr, unsigned long end) { pte_t *orig_src_pte, *orig_dst_pte; pte_t *src_pte, *dst_pte; @@ -802,6 +802,16 @@ static inline int copy_pmd_range(struct mm_struct *dst_mm, struct mm_struct *src src_pmd = pmd_offset(src_pud, addr); do { next = pmd_addr_end(addr, end); + if (pmd_trans_huge(*src_pmd)) { + int err; + err = copy_huge_pmd(dst_mm, src_mm, + dst_pmd, src_pmd, addr, vma); + if (err == -ENOMEM) + return -ENOMEM; + if (!err) + continue; + /* fall through */ + } if (pmd_none_or_clear_bad(src_pmd)) continue; if (copy_pte_range(dst_mm, src_mm, dst_pmd, src_pmd, @@ -1004,6 +1014,15 @@ static inline unsigned long zap_pmd_range(struct mmu_gather *tlb, pmd = pmd_offset(pud, addr); do { next = pmd_addr_end(addr, end); + if (pmd_trans_huge(*pmd)) { + if (next-addr != HPAGE_PMD_SIZE) + split_huge_page_pmd(vma->vm_mm, pmd); + else if (zap_huge_pmd(tlb, vma, pmd)) { + (*zap_work)--; + continue; + } + /* fall through */ + } if (pmd_none_or_clear_bad(pmd)) { (*zap_work)--; continue; @@ -1280,11 +1299,27 @@ struct page *follow_page(struct vm_area_struct *vma, unsigned long address, pmd = pmd_offset(pud, address); if (pmd_none(*pmd)) goto no_page_table; - if (pmd_huge(*pmd)) { + if (pmd_huge(*pmd) && vma->vm_flags & VM_HUGETLB) { BUG_ON(flags & FOLL_GET); page = follow_huge_pmd(mm, address, pmd, flags & FOLL_WRITE); goto out; } + if (pmd_trans_huge(*pmd)) { + spin_lock(&mm->page_table_lock); + if (likely(pmd_trans_huge(*pmd))) { + if (unlikely(pmd_trans_splitting(*pmd))) { + spin_unlock(&mm->page_table_lock); + wait_split_huge_page(vma->anon_vma, pmd); + } else { + page = follow_trans_huge_pmd(mm, address, + pmd, flags); + spin_unlock(&mm->page_table_lock); + goto out; + } + } else + spin_unlock(&mm->page_table_lock); + /* fall through */ + } if (unlikely(pmd_bad(*pmd))) goto no_page_table; @@ -3179,9 +3214,9 @@ static int do_nonlinear_fault(struct mm_struct *mm, struct vm_area_struct *vma, * but allow concurrent faults), and pte mapped but not yet locked. * We return with mmap_sem still held, but pte unmapped and unlocked. */ -static inline int handle_pte_fault(struct mm_struct *mm, - struct vm_area_struct *vma, unsigned long address, - pte_t *pte, pmd_t *pmd, unsigned int flags) +int handle_pte_fault(struct mm_struct *mm, + struct vm_area_struct *vma, unsigned long address, + pte_t *pte, pmd_t *pmd, unsigned int flags) { pte_t entry; spinlock_t *ptl; @@ -3260,9 +3295,40 @@ int handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma, pmd = pmd_alloc(mm, pud, address); if (!pmd) return VM_FAULT_OOM; - pte = pte_alloc_map(mm, vma, pmd, address); - if (!pte) + if (pmd_none(*pmd) && transparent_hugepage_enabled(vma)) { + if (!vma->vm_ops) + return do_huge_pmd_anonymous_page(mm, vma, address, + pmd, flags); + } else { + pmd_t orig_pmd = *pmd; + barrier(); + if (pmd_trans_huge(orig_pmd)) { + if (flags & FAULT_FLAG_WRITE && + !pmd_write(orig_pmd) && + !pmd_trans_splitting(orig_pmd)) + return do_huge_pmd_wp_page(mm, vma, address, + pmd, orig_pmd); + return 0; + } + } + + /* + * Use __pte_alloc instead of pte_alloc_map, because we can't + * run pte_offset_map on the pmd, if an huge pmd could + * materialize from under us from a different thread. + */ + if (unlikely(__pte_alloc(mm, vma, pmd, address))) return VM_FAULT_OOM; + /* if an huge pmd materialized from under us just retry later */ + if (unlikely(pmd_trans_huge(*pmd))) + return 0; + /* + * A regular pmd is established and it can't morph into a huge pmd + * from under us anymore at this point because we hold the mmap_sem + * read mode and khugepaged takes it in write mode. So now it's + * safe to run pte_offset_map(). + */ + pte = pte_offset_map(pmd, address); return handle_pte_fault(mm, vma, address, pte, pmd, flags); } diff --git a/mm/rmap.c b/mm/rmap.c index a3197a8a295b..e41375a6b029 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -360,7 +360,7 @@ void page_unlock_anon_vma(struct anon_vma *anon_vma) * Returns virtual address or -EFAULT if page's index/offset is not * within the range mapped the @vma. */ -static inline unsigned long +inline unsigned long vma_address(struct page *page, struct vm_area_struct *vma) { pgoff_t pgoff = page->index << (PAGE_CACHE_SHIFT - PAGE_SHIFT); @@ -435,6 +435,8 @@ pte_t *__page_check_address(struct page *page, struct mm_struct *mm, pmd = pmd_offset(pud, address); if (!pmd_present(*pmd)) return NULL; + if (pmd_trans_huge(*pmd)) + return NULL; pte = pte_offset_map(pmd, address); /* Make a quick check before getting the lock */ @@ -489,35 +491,17 @@ int page_referenced_one(struct page *page, struct vm_area_struct *vma, unsigned long *vm_flags) { struct mm_struct *mm = vma->vm_mm; - pte_t *pte; - spinlock_t *ptl; int referenced = 0; - pte = page_check_address(page, mm, address, &ptl, 0); - if (!pte) - goto out; - /* * Don't want to elevate referenced for mlocked page that gets this far, * in order that it progresses to try_to_unmap and is moved to the * unevictable list. */ if (vma->vm_flags & VM_LOCKED) { - *mapcount = 1; /* break early from loop */ + *mapcount = 0; /* break early from loop */ *vm_flags |= VM_LOCKED; - goto out_unmap; - } - - if (ptep_clear_flush_young_notify(vma, address, pte)) { - /* - * Don't treat a reference through a sequentially read - * mapping as such. If the page has been used in - * another mapping, we will catch it; if this other - * mapping is already gone, the unmap path will have - * set PG_referenced or activated the page. - */ - if (likely(!VM_SequentialReadHint(vma))) - referenced++; + goto out; } /* Pretend the page is referenced if the task has the @@ -526,9 +510,39 @@ int page_referenced_one(struct page *page, struct vm_area_struct *vma, rwsem_is_locked(&mm->mmap_sem)) referenced++; -out_unmap: + if (unlikely(PageTransHuge(page))) { + pmd_t *pmd; + + spin_lock(&mm->page_table_lock); + pmd = page_check_address_pmd(page, mm, address, + PAGE_CHECK_ADDRESS_PMD_FLAG); + if (pmd && !pmd_trans_splitting(*pmd) && + pmdp_clear_flush_young_notify(vma, address, pmd)) + referenced++; + spin_unlock(&mm->page_table_lock); + } else { + pte_t *pte; + spinlock_t *ptl; + + pte = page_check_address(page, mm, address, &ptl, 0); + if (!pte) + goto out; + + if (ptep_clear_flush_young_notify(vma, address, pte)) { + /* + * Don't treat a reference through a sequentially read + * mapping as such. If the page has been used in + * another mapping, we will catch it; if this other + * mapping is already gone, the unmap path will have + * set PG_referenced or activated the page. + */ + if (likely(!VM_SequentialReadHint(vma))) + referenced++; + } + pte_unmap_unlock(pte, ptl); + } + (*mapcount)--; - pte_unmap_unlock(pte, ptl); if (referenced) *vm_flags |= vma->vm_flags; @@ -1202,7 +1216,7 @@ static int try_to_unmap_cluster(unsigned long cursor, unsigned int *mapcount, return ret; } -static bool is_vma_temporary_stack(struct vm_area_struct *vma) +bool is_vma_temporary_stack(struct vm_area_struct *vma) { int maybe_stack = vma->vm_flags & (VM_GROWSDOWN | VM_GROWSUP); diff --git a/mm/swap.c b/mm/swap.c index e0eeef940886..c02f93611a84 100644 --- a/mm/swap.c +++ b/mm/swap.c @@ -479,6 +479,43 @@ void __pagevec_release(struct pagevec *pvec) EXPORT_SYMBOL(__pagevec_release); +/* used by __split_huge_page_refcount() */ +void lru_add_page_tail(struct zone* zone, + struct page *page, struct page *page_tail) +{ + int active; + enum lru_list lru; + const int file = 0; + struct list_head *head; + + VM_BUG_ON(!PageHead(page)); + VM_BUG_ON(PageCompound(page_tail)); + VM_BUG_ON(PageLRU(page_tail)); + VM_BUG_ON(!spin_is_locked(&zone->lru_lock)); + + SetPageLRU(page_tail); + + if (page_evictable(page_tail, NULL)) { + if (PageActive(page)) { + SetPageActive(page_tail); + active = 1; + lru = LRU_ACTIVE_ANON; + } else { + active = 0; + lru = LRU_INACTIVE_ANON; + } + update_page_reclaim_stat(zone, page_tail, file, active); + if (likely(PageLRU(page))) + head = page->lru.prev; + else + head = &zone->lru[lru].list; + __add_page_to_lru_list(zone, page_tail, lru, head); + } else { + SetPageUnevictable(page_tail); + add_page_to_lru_list(zone, page_tail, LRU_UNEVICTABLE); + } +} + /* * Add the passed pages to the LRU, then drop the caller's refcount * on them. Reinitialises the caller's pagevec. -- cgit v1.2.3 From 0af4e98b6b095c74588af04872f83d333c958c32 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:46:55 -0800 Subject: thp: madvise(MADV_HUGEPAGE) Add madvise MADV_HUGEPAGE to mark regions that are important to be hugepage backed. Return -EINVAL if the vma is not of an anonymous type, or the feature isn't built into the kernel. Never silently return success. Signed-off-by: Andrea Arcangeli Acked-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/huge_mm.h | 6 ++++++ mm/huge_memory.c | 16 ++++++++++++++++ mm/madvise.c | 8 ++++++++ 3 files changed, 30 insertions(+) (limited to 'include/linux') diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h index 9301824c7491..d9ab70d776e2 100644 --- a/include/linux/huge_mm.h +++ b/include/linux/huge_mm.h @@ -97,6 +97,7 @@ extern void __split_huge_page_pmd(struct mm_struct *mm, pmd_t *pmd); #if HPAGE_PMD_ORDER > MAX_ORDER #error "hugepages can't be allocated by the buddy allocator" #endif +extern int hugepage_madvise(unsigned long *vm_flags); #else /* CONFIG_TRANSPARENT_HUGEPAGE */ #define HPAGE_PMD_SHIFT ({ BUG(); 0; }) #define HPAGE_PMD_MASK ({ BUG(); 0; }) @@ -113,6 +114,11 @@ static inline int split_huge_page(struct page *page) do { } while (0) #define wait_split_huge_page(__anon_vma, __pmd) \ do { } while (0) +static inline int hugepage_madvise(unsigned long *vm_flags) +{ + BUG(); + return 0; +} #endif /* CONFIG_TRANSPARENT_HUGEPAGE */ #endif /* _LINUX_HUGE_MM_H */ diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 763507932898..620891f4e54f 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -896,6 +896,22 @@ out: return ret; } +int hugepage_madvise(unsigned long *vm_flags) +{ + /* + * Be somewhat over-protective like KSM for now! + */ + if (*vm_flags & (VM_HUGEPAGE | VM_SHARED | VM_MAYSHARE | + VM_PFNMAP | VM_IO | VM_DONTEXPAND | + VM_RESERVED | VM_HUGETLB | VM_INSERTPAGE | + VM_MIXEDMAP | VM_SAO)) + return -EINVAL; + + *vm_flags |= VM_HUGEPAGE; + + return 0; +} + void __split_huge_page_pmd(struct mm_struct *mm, pmd_t *pmd) { struct page *page; diff --git a/mm/madvise.c b/mm/madvise.c index 319528b8db74..ecde40a401c1 100644 --- a/mm/madvise.c +++ b/mm/madvise.c @@ -71,6 +71,11 @@ static long madvise_behavior(struct vm_area_struct * vma, if (error) goto out; break; + case MADV_HUGEPAGE: + error = hugepage_madvise(&new_flags); + if (error) + goto out; + break; } if (new_flags == vma->vm_flags) { @@ -282,6 +287,9 @@ madvise_behavior_valid(int behavior) #ifdef CONFIG_KSM case MADV_MERGEABLE: case MADV_UNMERGEABLE: +#endif +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + case MADV_HUGEPAGE: #endif return 1; -- cgit v1.2.3 From 500d65d471018d9a13b0d51b7e141ed2a3555c1d Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:46:55 -0800 Subject: thp: pmd_trans_huge migrate bugcheck No pmd_trans_huge should ever materialize in migration ptes areas, because we split the hugepage before migration ptes are instantiated. Signed-off-by: Andrea Arcangeli Acked-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 1 + mm/memory.c | 5 +++++ mm/migrate.c | 7 ++++++- 3 files changed, 12 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/mm.h b/include/linux/mm.h index 78adec4ba9f4..2ec5138badab 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1490,6 +1490,7 @@ struct page *follow_page(struct vm_area_struct *, unsigned long address, #define FOLL_DUMP 0x08 /* give error on hole if it would be zero */ #define FOLL_FORCE 0x10 /* get_user_pages read/write w/o permission */ #define FOLL_MLOCK 0x40 /* mark page as mlocked */ +#define FOLL_SPLIT 0x80 /* don't return transhuge pages, split them */ typedef int (*pte_fn_t)(pte_t *pte, pgtable_t token, unsigned long addr, void *data); diff --git a/mm/memory.c b/mm/memory.c index c1a80e00458d..12ee1ea237f5 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -1305,6 +1305,10 @@ struct page *follow_page(struct vm_area_struct *vma, unsigned long address, goto out; } if (pmd_trans_huge(*pmd)) { + if (flags & FOLL_SPLIT) { + split_huge_page_pmd(mm, pmd); + goto split_fallthrough; + } spin_lock(&mm->page_table_lock); if (likely(pmd_trans_huge(*pmd))) { if (unlikely(pmd_trans_splitting(*pmd))) { @@ -1320,6 +1324,7 @@ struct page *follow_page(struct vm_area_struct *vma, unsigned long address, spin_unlock(&mm->page_table_lock); /* fall through */ } +split_fallthrough: if (unlikely(pmd_bad(*pmd))) goto no_page_table; diff --git a/mm/migrate.c b/mm/migrate.c index 690d0de993af..1a531b760b3b 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -113,6 +113,8 @@ static int remove_migration_pte(struct page *new, struct vm_area_struct *vma, goto out; pmd = pmd_offset(pud, addr); + if (pmd_trans_huge(*pmd)) + goto out; if (!pmd_present(*pmd)) goto out; @@ -632,6 +634,9 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private, /* page was freed from under us. So we are done. */ goto move_newpage; } + if (unlikely(PageTransHuge(page))) + if (unlikely(split_huge_page(page))) + goto move_newpage; /* prepare cgroup just returns 0 or -ENOMEM */ rc = -EAGAIN; @@ -1063,7 +1068,7 @@ static int do_move_page_to_node_array(struct mm_struct *mm, if (!vma || pp->addr < vma->vm_start || !vma_migratable(vma)) goto set_status; - page = follow_page(vma, pp->addr, FOLL_GET); + page = follow_page(vma, pp->addr, FOLL_GET|FOLL_SPLIT); err = PTR_ERR(page); if (IS_ERR(page)) -- cgit v1.2.3 From 79134171df238171daa4c024a42b77b401ccb00b Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:46:58 -0800 Subject: thp: transparent hugepage vmstat Add hugepage stat information to /proc/vmstat and /proc/meminfo. Signed-off-by: Andrea Arcangeli Acked-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/meminfo.c | 14 +++++++++++++- include/linux/mmzone.h | 1 + mm/huge_memory.c | 3 +++ mm/rmap.c | 20 ++++++++++++++++---- mm/vmstat.c | 1 + 5 files changed, 34 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c index a65239cfd97e..ed257d141568 100644 --- a/fs/proc/meminfo.c +++ b/fs/proc/meminfo.c @@ -100,6 +100,9 @@ static int meminfo_proc_show(struct seq_file *m, void *v) "VmallocChunk: %8lu kB\n" #ifdef CONFIG_MEMORY_FAILURE "HardwareCorrupted: %5lu kB\n" +#endif +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + "AnonHugePages: %8lu kB\n" #endif , K(i.totalram), @@ -128,7 +131,12 @@ static int meminfo_proc_show(struct seq_file *m, void *v) K(i.freeswap), K(global_page_state(NR_FILE_DIRTY)), K(global_page_state(NR_WRITEBACK)), - K(global_page_state(NR_ANON_PAGES)), + K(global_page_state(NR_ANON_PAGES) +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + + global_page_state(NR_ANON_TRANSPARENT_HUGEPAGES) * + HPAGE_PMD_NR +#endif + ), K(global_page_state(NR_FILE_MAPPED)), K(global_page_state(NR_SHMEM)), K(global_page_state(NR_SLAB_RECLAIMABLE) + @@ -150,6 +158,10 @@ static int meminfo_proc_show(struct seq_file *m, void *v) vmi.largest_chunk >> 10 #ifdef CONFIG_MEMORY_FAILURE ,atomic_long_read(&mce_bad_pages) << (PAGE_SHIFT - 10) +#endif +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + ,K(global_page_state(NR_ANON_TRANSPARENT_HUGEPAGES) * + HPAGE_PMD_NR) #endif ); diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index dad3612a7e39..02ecb0189b1d 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -114,6 +114,7 @@ enum zone_stat_item { NUMA_LOCAL, /* allocation from local node */ NUMA_OTHER, /* allocation from other node */ #endif + NR_ANON_TRANSPARENT_HUGEPAGES, NR_VM_ZONE_STAT_ITEMS }; /* diff --git a/mm/huge_memory.c b/mm/huge_memory.c index a313403b3c5e..7101112a5429 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -751,6 +751,9 @@ static void __split_huge_page_refcount(struct page *page) lru_add_page_tail(zone, page, page_tail); } + __dec_zone_page_state(page, NR_ANON_TRANSPARENT_HUGEPAGES); + __mod_zone_page_state(zone, NR_ANON_PAGES, HPAGE_PMD_NR); + ClearPageCompound(page); compound_unlock(page); spin_unlock_irq(&zone->lru_lock); diff --git a/mm/rmap.c b/mm/rmap.c index 92e14dcfe737..3825ae4bc32f 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -882,8 +882,13 @@ void do_page_add_anon_rmap(struct page *page, struct vm_area_struct *vma, unsigned long address, int exclusive) { int first = atomic_inc_and_test(&page->_mapcount); - if (first) - __inc_zone_page_state(page, NR_ANON_PAGES); + if (first) { + if (!PageTransHuge(page)) + __inc_zone_page_state(page, NR_ANON_PAGES); + else + __inc_zone_page_state(page, + NR_ANON_TRANSPARENT_HUGEPAGES); + } if (unlikely(PageKsm(page))) return; @@ -911,7 +916,10 @@ void page_add_new_anon_rmap(struct page *page, VM_BUG_ON(address < vma->vm_start || address >= vma->vm_end); SetPageSwapBacked(page); atomic_set(&page->_mapcount, 0); /* increment count (starts at -1) */ - __inc_zone_page_state(page, NR_ANON_PAGES); + if (!PageTransHuge(page)) + __inc_zone_page_state(page, NR_ANON_PAGES); + else + __inc_zone_page_state(page, NR_ANON_TRANSPARENT_HUGEPAGES); __page_set_anon_rmap(page, vma, address, 1); if (page_evictable(page, vma)) lru_cache_add_lru(page, LRU_ACTIVE_ANON); @@ -964,7 +972,11 @@ void page_remove_rmap(struct page *page) return; if (PageAnon(page)) { mem_cgroup_uncharge_page(page); - __dec_zone_page_state(page, NR_ANON_PAGES); + if (!PageTransHuge(page)) + __dec_zone_page_state(page, NR_ANON_PAGES); + else + __dec_zone_page_state(page, + NR_ANON_TRANSPARENT_HUGEPAGES); } else { __dec_zone_page_state(page, NR_FILE_MAPPED); mem_cgroup_update_file_mapped(page, -1); diff --git a/mm/vmstat.c b/mm/vmstat.c index 751a65e00aac..0c3b5048773e 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -880,6 +880,7 @@ static const char * const vmstat_text[] = { "numa_local", "numa_other", #endif + "nr_anon_transparent_hugepages", "nr_dirty_threshold", "nr_dirty_background_threshold", -- cgit v1.2.3 From ba76149f47d8c939efa0acc07a191237af900471 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:46:58 -0800 Subject: thp: khugepaged Add khugepaged to relocate fragmented pages into hugepages if new hugepages become available. (this is indipendent of the defrag logic that will have to make new hugepages available) The fundamental reason why khugepaged is unavoidable, is that some memory can be fragmented and not everything can be relocated. So when a virtual machine quits and releases gigabytes of hugepages, we want to use those freely available hugepages to create huge-pmd in the other virtual machines that may be running on fragmented memory, to maximize the CPU efficiency at all times. The scan is slow, it takes nearly zero cpu time, except when it copies data (in which case it means we definitely want to pay for that cpu time) so it seems a good tradeoff. In addition to the hugepages being released by other process releasing memory, we have the strong suspicion that the performance impact of potentially defragmenting hugepages during or before each page fault could lead to more performance inconsistency than allocating small pages at first and having them collapsed into large pages later... if they prove themselfs to be long lived mappings (khugepaged scan is slow so short lived mappings have low probability to run into khugepaged if compared to long lived mappings). Signed-off-by: Andrea Arcangeli Acked-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/huge_mm.h | 1 + include/linux/khugepaged.h | 66 +++ include/linux/sched.h | 1 + kernel/fork.c | 5 + mm/huge_memory.c | 1073 +++++++++++++++++++++++++++++++++++++++++++- 5 files changed, 1136 insertions(+), 10 deletions(-) create mode 100644 include/linux/khugepaged.h (limited to 'include/linux') diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h index d9ab70d776e2..43a694ef8904 100644 --- a/include/linux/huge_mm.h +++ b/include/linux/huge_mm.h @@ -25,6 +25,7 @@ enum transparent_hugepage_flag { TRANSPARENT_HUGEPAGE_REQ_MADV_FLAG, TRANSPARENT_HUGEPAGE_DEFRAG_FLAG, TRANSPARENT_HUGEPAGE_DEFRAG_REQ_MADV_FLAG, + TRANSPARENT_HUGEPAGE_DEFRAG_KHUGEPAGED_FLAG, #ifdef CONFIG_DEBUG_VM TRANSPARENT_HUGEPAGE_DEBUG_COW_FLAG, #endif diff --git a/include/linux/khugepaged.h b/include/linux/khugepaged.h new file mode 100644 index 000000000000..552f3184756c --- /dev/null +++ b/include/linux/khugepaged.h @@ -0,0 +1,66 @@ +#ifndef _LINUX_KHUGEPAGED_H +#define _LINUX_KHUGEPAGED_H + +#include /* MMF_VM_HUGEPAGE */ + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +extern int __khugepaged_enter(struct mm_struct *mm); +extern void __khugepaged_exit(struct mm_struct *mm); +extern int khugepaged_enter_vma_merge(struct vm_area_struct *vma); + +#define khugepaged_enabled() \ + (transparent_hugepage_flags & \ + ((1<flags)) + return __khugepaged_enter(mm); + return 0; +} + +static inline void khugepaged_exit(struct mm_struct *mm) +{ + if (test_bit(MMF_VM_HUGEPAGE, &mm->flags)) + __khugepaged_exit(mm); +} + +static inline int khugepaged_enter(struct vm_area_struct *vma) +{ + if (!test_bit(MMF_VM_HUGEPAGE, &vma->vm_mm->flags)) + if (khugepaged_always() || + (khugepaged_req_madv() && + vma->vm_flags & VM_HUGEPAGE)) + if (__khugepaged_enter(vma->vm_mm)) + return -ENOMEM; + return 0; +} +#else /* CONFIG_TRANSPARENT_HUGEPAGE */ +static inline int khugepaged_fork(struct mm_struct *mm, struct mm_struct *oldmm) +{ + return 0; +} +static inline void khugepaged_exit(struct mm_struct *mm) +{ +} +static inline int khugepaged_enter(struct vm_area_struct *vma) +{ + return 0; +} +static inline int khugepaged_enter_vma_merge(struct vm_area_struct *vma) +{ + return 0; +} +#endif /* CONFIG_TRANSPARENT_HUGEPAGE */ + +#endif /* _LINUX_KHUGEPAGED_H */ diff --git a/include/linux/sched.h b/include/linux/sched.h index f23b5bb6f52e..d747f948b34e 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -434,6 +434,7 @@ extern int get_dumpable(struct mm_struct *mm); #endif /* leave room for more dump flags */ #define MMF_VM_MERGEABLE 16 /* KSM may merge identical pages */ +#define MMF_VM_HUGEPAGE 17 /* set when VM_HUGEPAGE is set on vma */ #define MMF_INIT_MASK (MMF_DUMPABLE_MASK | MMF_DUMP_FILTER_MASK) diff --git a/kernel/fork.c b/kernel/fork.c index f78f50ba6cb2..25e429152ddc 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -66,6 +66,7 @@ #include #include #include +#include #include #include @@ -328,6 +329,9 @@ static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm) rb_parent = NULL; pprev = &mm->mmap; retval = ksm_fork(mm, oldmm); + if (retval) + goto out; + retval = khugepaged_fork(mm, oldmm); if (retval) goto out; @@ -546,6 +550,7 @@ void mmput(struct mm_struct *mm) if (atomic_dec_and_test(&mm->mm_users)) { exit_aio(mm); ksm_exit(mm); + khugepaged_exit(mm); /* must run before exit_mmap */ exit_mmap(mm); set_mm_exe_file(mm, NULL); if (!list_empty(&mm->mmlist)) { diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 7101112a5429..ae2bf08b1099 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -12,14 +12,111 @@ #include #include #include +#include +#include +#include #include #include #include "internal.h" +/* + * By default transparent hugepage support is enabled for all mappings + * and khugepaged scans all mappings. Defrag is only invoked by + * khugepaged hugepage allocations and by page faults inside + * MADV_HUGEPAGE regions to avoid the risk of slowing down short lived + * allocations. + */ unsigned long transparent_hugepage_flags __read_mostly = - (1< 0) { + int err = start_khugepaged(); + if (err) + ret = err; + } + + return ret; } static struct kobj_attribute enabled_attr = __ATTR(enabled, 0644, enabled_show, enabled_store); @@ -153,20 +260,212 @@ static struct attribute *hugepage_attr[] = { static struct attribute_group hugepage_attr_group = { .attrs = hugepage_attr, - .name = "transparent_hugepage", +}; + +static ssize_t scan_sleep_millisecs_show(struct kobject *kobj, + struct kobj_attribute *attr, + char *buf) +{ + return sprintf(buf, "%u\n", khugepaged_scan_sleep_millisecs); +} + +static ssize_t scan_sleep_millisecs_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t count) +{ + unsigned long msecs; + int err; + + err = strict_strtoul(buf, 10, &msecs); + if (err || msecs > UINT_MAX) + return -EINVAL; + + khugepaged_scan_sleep_millisecs = msecs; + wake_up_interruptible(&khugepaged_wait); + + return count; +} +static struct kobj_attribute scan_sleep_millisecs_attr = + __ATTR(scan_sleep_millisecs, 0644, scan_sleep_millisecs_show, + scan_sleep_millisecs_store); + +static ssize_t alloc_sleep_millisecs_show(struct kobject *kobj, + struct kobj_attribute *attr, + char *buf) +{ + return sprintf(buf, "%u\n", khugepaged_alloc_sleep_millisecs); +} + +static ssize_t alloc_sleep_millisecs_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t count) +{ + unsigned long msecs; + int err; + + err = strict_strtoul(buf, 10, &msecs); + if (err || msecs > UINT_MAX) + return -EINVAL; + + khugepaged_alloc_sleep_millisecs = msecs; + wake_up_interruptible(&khugepaged_wait); + + return count; +} +static struct kobj_attribute alloc_sleep_millisecs_attr = + __ATTR(alloc_sleep_millisecs, 0644, alloc_sleep_millisecs_show, + alloc_sleep_millisecs_store); + +static ssize_t pages_to_scan_show(struct kobject *kobj, + struct kobj_attribute *attr, + char *buf) +{ + return sprintf(buf, "%u\n", khugepaged_pages_to_scan); +} +static ssize_t pages_to_scan_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t count) +{ + int err; + unsigned long pages; + + err = strict_strtoul(buf, 10, &pages); + if (err || !pages || pages > UINT_MAX) + return -EINVAL; + + khugepaged_pages_to_scan = pages; + + return count; +} +static struct kobj_attribute pages_to_scan_attr = + __ATTR(pages_to_scan, 0644, pages_to_scan_show, + pages_to_scan_store); + +static ssize_t pages_collapsed_show(struct kobject *kobj, + struct kobj_attribute *attr, + char *buf) +{ + return sprintf(buf, "%u\n", khugepaged_pages_collapsed); +} +static struct kobj_attribute pages_collapsed_attr = + __ATTR_RO(pages_collapsed); + +static ssize_t full_scans_show(struct kobject *kobj, + struct kobj_attribute *attr, + char *buf) +{ + return sprintf(buf, "%u\n", khugepaged_full_scans); +} +static struct kobj_attribute full_scans_attr = + __ATTR_RO(full_scans); + +static ssize_t khugepaged_defrag_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + return single_flag_show(kobj, attr, buf, + TRANSPARENT_HUGEPAGE_DEFRAG_KHUGEPAGED_FLAG); +} +static ssize_t khugepaged_defrag_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t count) +{ + return single_flag_store(kobj, attr, buf, count, + TRANSPARENT_HUGEPAGE_DEFRAG_KHUGEPAGED_FLAG); +} +static struct kobj_attribute khugepaged_defrag_attr = + __ATTR(defrag, 0644, khugepaged_defrag_show, + khugepaged_defrag_store); + +/* + * max_ptes_none controls if khugepaged should collapse hugepages over + * any unmapped ptes in turn potentially increasing the memory + * footprint of the vmas. When max_ptes_none is 0 khugepaged will not + * reduce the available free memory in the system as it + * runs. Increasing max_ptes_none will instead potentially reduce the + * free memory in the system during the khugepaged scan. + */ +static ssize_t khugepaged_max_ptes_none_show(struct kobject *kobj, + struct kobj_attribute *attr, + char *buf) +{ + return sprintf(buf, "%u\n", khugepaged_max_ptes_none); +} +static ssize_t khugepaged_max_ptes_none_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t count) +{ + int err; + unsigned long max_ptes_none; + + err = strict_strtoul(buf, 10, &max_ptes_none); + if (err || max_ptes_none > HPAGE_PMD_NR-1) + return -EINVAL; + + khugepaged_max_ptes_none = max_ptes_none; + + return count; +} +static struct kobj_attribute khugepaged_max_ptes_none_attr = + __ATTR(max_ptes_none, 0644, khugepaged_max_ptes_none_show, + khugepaged_max_ptes_none_store); + +static struct attribute *khugepaged_attr[] = { + &khugepaged_defrag_attr.attr, + &khugepaged_max_ptes_none_attr.attr, + &pages_to_scan_attr.attr, + &pages_collapsed_attr.attr, + &full_scans_attr.attr, + &scan_sleep_millisecs_attr.attr, + &alloc_sleep_millisecs_attr.attr, + NULL, +}; + +static struct attribute_group khugepaged_attr_group = { + .attrs = khugepaged_attr, + .name = "khugepaged", }; #endif /* CONFIG_SYSFS */ static int __init hugepage_init(void) { -#ifdef CONFIG_SYSFS int err; +#ifdef CONFIG_SYSFS + static struct kobject *hugepage_kobj; - err = sysfs_create_group(mm_kobj, &hugepage_attr_group); - if (err) - printk(KERN_ERR "hugepage: register sysfs failed\n"); + err = -ENOMEM; + hugepage_kobj = kobject_create_and_add("transparent_hugepage", mm_kobj); + if (unlikely(!hugepage_kobj)) { + printk(KERN_ERR "hugepage: failed kobject create\n"); + goto out; + } + + err = sysfs_create_group(hugepage_kobj, &hugepage_attr_group); + if (err) { + printk(KERN_ERR "hugepage: failed register hugeage group\n"); + goto out; + } + + err = sysfs_create_group(hugepage_kobj, &khugepaged_attr_group); + if (err) { + printk(KERN_ERR "hugepage: failed register hugeage group\n"); + goto out; + } #endif - return 0; + + err = khugepaged_slab_init(); + if (err) + goto out; + + err = mm_slots_hash_init(); + if (err) { + khugepaged_slab_free(); + goto out; + } + + start_khugepaged(); + +out: + return err; } module_init(hugepage_init) @@ -285,6 +584,8 @@ int do_huge_pmd_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma, if (haddr >= vma->vm_start && haddr + HPAGE_PMD_SIZE <= vma->vm_end) { if (unlikely(anon_vma_prepare(vma))) return VM_FAULT_OOM; + if (unlikely(khugepaged_enter(vma))) + return VM_FAULT_OOM; page = alloc_hugepage(transparent_hugepage_defrag(vma)); if (unlikely(!page)) goto out; @@ -941,6 +1242,758 @@ int hugepage_madvise(unsigned long *vm_flags) return 0; } +static int __init khugepaged_slab_init(void) +{ + mm_slot_cache = kmem_cache_create("khugepaged_mm_slot", + sizeof(struct mm_slot), + __alignof__(struct mm_slot), 0, NULL); + if (!mm_slot_cache) + return -ENOMEM; + + return 0; +} + +static void __init khugepaged_slab_free(void) +{ + kmem_cache_destroy(mm_slot_cache); + mm_slot_cache = NULL; +} + +static inline struct mm_slot *alloc_mm_slot(void) +{ + if (!mm_slot_cache) /* initialization failed */ + return NULL; + return kmem_cache_zalloc(mm_slot_cache, GFP_KERNEL); +} + +static inline void free_mm_slot(struct mm_slot *mm_slot) +{ + kmem_cache_free(mm_slot_cache, mm_slot); +} + +static int __init mm_slots_hash_init(void) +{ + mm_slots_hash = kzalloc(MM_SLOTS_HASH_HEADS * sizeof(struct hlist_head), + GFP_KERNEL); + if (!mm_slots_hash) + return -ENOMEM; + return 0; +} + +#if 0 +static void __init mm_slots_hash_free(void) +{ + kfree(mm_slots_hash); + mm_slots_hash = NULL; +} +#endif + +static struct mm_slot *get_mm_slot(struct mm_struct *mm) +{ + struct mm_slot *mm_slot; + struct hlist_head *bucket; + struct hlist_node *node; + + bucket = &mm_slots_hash[((unsigned long)mm / sizeof(struct mm_struct)) + % MM_SLOTS_HASH_HEADS]; + hlist_for_each_entry(mm_slot, node, bucket, hash) { + if (mm == mm_slot->mm) + return mm_slot; + } + return NULL; +} + +static void insert_to_mm_slots_hash(struct mm_struct *mm, + struct mm_slot *mm_slot) +{ + struct hlist_head *bucket; + + bucket = &mm_slots_hash[((unsigned long)mm / sizeof(struct mm_struct)) + % MM_SLOTS_HASH_HEADS]; + mm_slot->mm = mm; + hlist_add_head(&mm_slot->hash, bucket); +} + +static inline int khugepaged_test_exit(struct mm_struct *mm) +{ + return atomic_read(&mm->mm_users) == 0; +} + +int __khugepaged_enter(struct mm_struct *mm) +{ + struct mm_slot *mm_slot; + int wakeup; + + mm_slot = alloc_mm_slot(); + if (!mm_slot) + return -ENOMEM; + + /* __khugepaged_exit() must not run from under us */ + VM_BUG_ON(khugepaged_test_exit(mm)); + if (unlikely(test_and_set_bit(MMF_VM_HUGEPAGE, &mm->flags))) { + free_mm_slot(mm_slot); + return 0; + } + + spin_lock(&khugepaged_mm_lock); + insert_to_mm_slots_hash(mm, mm_slot); + /* + * Insert just behind the scanning cursor, to let the area settle + * down a little. + */ + wakeup = list_empty(&khugepaged_scan.mm_head); + list_add_tail(&mm_slot->mm_node, &khugepaged_scan.mm_head); + spin_unlock(&khugepaged_mm_lock); + + atomic_inc(&mm->mm_count); + if (wakeup) + wake_up_interruptible(&khugepaged_wait); + + return 0; +} + +int khugepaged_enter_vma_merge(struct vm_area_struct *vma) +{ + unsigned long hstart, hend; + if (!vma->anon_vma) + /* + * Not yet faulted in so we will register later in the + * page fault if needed. + */ + return 0; + if (vma->vm_file || vma->vm_ops) + /* khugepaged not yet working on file or special mappings */ + return 0; + VM_BUG_ON(is_linear_pfn_mapping(vma) || is_pfn_mapping(vma)); + hstart = (vma->vm_start + ~HPAGE_PMD_MASK) & HPAGE_PMD_MASK; + hend = vma->vm_end & HPAGE_PMD_MASK; + if (hstart < hend) + return khugepaged_enter(vma); + return 0; +} + +void __khugepaged_exit(struct mm_struct *mm) +{ + struct mm_slot *mm_slot; + int free = 0; + + spin_lock(&khugepaged_mm_lock); + mm_slot = get_mm_slot(mm); + if (mm_slot && khugepaged_scan.mm_slot != mm_slot) { + hlist_del(&mm_slot->hash); + list_del(&mm_slot->mm_node); + free = 1; + } + + if (free) { + spin_unlock(&khugepaged_mm_lock); + clear_bit(MMF_VM_HUGEPAGE, &mm->flags); + free_mm_slot(mm_slot); + mmdrop(mm); + } else if (mm_slot) { + spin_unlock(&khugepaged_mm_lock); + /* + * This is required to serialize against + * khugepaged_test_exit() (which is guaranteed to run + * under mmap sem read mode). Stop here (after we + * return all pagetables will be destroyed) until + * khugepaged has finished working on the pagetables + * under the mmap_sem. + */ + down_write(&mm->mmap_sem); + up_write(&mm->mmap_sem); + } else + spin_unlock(&khugepaged_mm_lock); +} + +static void release_pte_page(struct page *page) +{ + /* 0 stands for page_is_file_cache(page) == false */ + dec_zone_page_state(page, NR_ISOLATED_ANON + 0); + unlock_page(page); + putback_lru_page(page); +} + +static void release_pte_pages(pte_t *pte, pte_t *_pte) +{ + while (--_pte >= pte) { + pte_t pteval = *_pte; + if (!pte_none(pteval)) + release_pte_page(pte_page(pteval)); + } +} + +static void release_all_pte_pages(pte_t *pte) +{ + release_pte_pages(pte, pte + HPAGE_PMD_NR); +} + +static int __collapse_huge_page_isolate(struct vm_area_struct *vma, + unsigned long address, + pte_t *pte) +{ + struct page *page; + pte_t *_pte; + int referenced = 0, isolated = 0, none = 0; + for (_pte = pte; _pte < pte+HPAGE_PMD_NR; + _pte++, address += PAGE_SIZE) { + pte_t pteval = *_pte; + if (pte_none(pteval)) { + if (++none <= khugepaged_max_ptes_none) + continue; + else { + release_pte_pages(pte, _pte); + goto out; + } + } + if (!pte_present(pteval) || !pte_write(pteval)) { + release_pte_pages(pte, _pte); + goto out; + } + page = vm_normal_page(vma, address, pteval); + if (unlikely(!page)) { + release_pte_pages(pte, _pte); + goto out; + } + VM_BUG_ON(PageCompound(page)); + BUG_ON(!PageAnon(page)); + VM_BUG_ON(!PageSwapBacked(page)); + + /* cannot use mapcount: can't collapse if there's a gup pin */ + if (page_count(page) != 1) { + release_pte_pages(pte, _pte); + goto out; + } + /* + * We can do it before isolate_lru_page because the + * page can't be freed from under us. NOTE: PG_lock + * is needed to serialize against split_huge_page + * when invoked from the VM. + */ + if (!trylock_page(page)) { + release_pte_pages(pte, _pte); + goto out; + } + /* + * Isolate the page to avoid collapsing an hugepage + * currently in use by the VM. + */ + if (isolate_lru_page(page)) { + unlock_page(page); + release_pte_pages(pte, _pte); + goto out; + } + /* 0 stands for page_is_file_cache(page) == false */ + inc_zone_page_state(page, NR_ISOLATED_ANON + 0); + VM_BUG_ON(!PageLocked(page)); + VM_BUG_ON(PageLRU(page)); + + /* If there is no mapped pte young don't collapse the page */ + if (pte_young(pteval)) + referenced = 1; + } + if (unlikely(!referenced)) + release_all_pte_pages(pte); + else + isolated = 1; +out: + return isolated; +} + +static void __collapse_huge_page_copy(pte_t *pte, struct page *page, + struct vm_area_struct *vma, + unsigned long address, + spinlock_t *ptl) +{ + pte_t *_pte; + for (_pte = pte; _pte < pte+HPAGE_PMD_NR; _pte++) { + pte_t pteval = *_pte; + struct page *src_page; + + if (pte_none(pteval)) { + clear_user_highpage(page, address); + add_mm_counter(vma->vm_mm, MM_ANONPAGES, 1); + } else { + src_page = pte_page(pteval); + copy_user_highpage(page, src_page, address, vma); + VM_BUG_ON(page_mapcount(src_page) != 1); + VM_BUG_ON(page_count(src_page) != 2); + release_pte_page(src_page); + /* + * ptl mostly unnecessary, but preempt has to + * be disabled to update the per-cpu stats + * inside page_remove_rmap(). + */ + spin_lock(ptl); + /* + * paravirt calls inside pte_clear here are + * superfluous. + */ + pte_clear(vma->vm_mm, address, _pte); + page_remove_rmap(src_page); + spin_unlock(ptl); + free_page_and_swap_cache(src_page); + } + + address += PAGE_SIZE; + page++; + } +} + +static void collapse_huge_page(struct mm_struct *mm, + unsigned long address, + struct page **hpage) +{ + struct vm_area_struct *vma; + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd, _pmd; + pte_t *pte; + pgtable_t pgtable; + struct page *new_page; + spinlock_t *ptl; + int isolated; + unsigned long hstart, hend; + + VM_BUG_ON(address & ~HPAGE_PMD_MASK); + VM_BUG_ON(!*hpage); + + /* + * Prevent all access to pagetables with the exception of + * gup_fast later hanlded by the ptep_clear_flush and the VM + * handled by the anon_vma lock + PG_lock. + */ + down_write(&mm->mmap_sem); + if (unlikely(khugepaged_test_exit(mm))) + goto out; + + vma = find_vma(mm, address); + hstart = (vma->vm_start + ~HPAGE_PMD_MASK) & HPAGE_PMD_MASK; + hend = vma->vm_end & HPAGE_PMD_MASK; + if (address < hstart || address + HPAGE_PMD_SIZE > hend) + goto out; + + if (!(vma->vm_flags & VM_HUGEPAGE) && !khugepaged_always()) + goto out; + + /* VM_PFNMAP vmas may have vm_ops null but vm_file set */ + if (!vma->anon_vma || vma->vm_ops || vma->vm_file) + goto out; + VM_BUG_ON(is_linear_pfn_mapping(vma) || is_pfn_mapping(vma)); + + pgd = pgd_offset(mm, address); + if (!pgd_present(*pgd)) + goto out; + + pud = pud_offset(pgd, address); + if (!pud_present(*pud)) + goto out; + + pmd = pmd_offset(pud, address); + /* pmd can't go away or become huge under us */ + if (!pmd_present(*pmd) || pmd_trans_huge(*pmd)) + goto out; + + new_page = *hpage; + if (unlikely(mem_cgroup_newpage_charge(new_page, mm, GFP_KERNEL))) + goto out; + + anon_vma_lock(vma->anon_vma); + + pte = pte_offset_map(pmd, address); + ptl = pte_lockptr(mm, pmd); + + spin_lock(&mm->page_table_lock); /* probably unnecessary */ + /* + * After this gup_fast can't run anymore. This also removes + * any huge TLB entry from the CPU so we won't allow + * huge and small TLB entries for the same virtual address + * to avoid the risk of CPU bugs in that area. + */ + _pmd = pmdp_clear_flush_notify(vma, address, pmd); + spin_unlock(&mm->page_table_lock); + + spin_lock(ptl); + isolated = __collapse_huge_page_isolate(vma, address, pte); + spin_unlock(ptl); + pte_unmap(pte); + + if (unlikely(!isolated)) { + spin_lock(&mm->page_table_lock); + BUG_ON(!pmd_none(*pmd)); + set_pmd_at(mm, address, pmd, _pmd); + spin_unlock(&mm->page_table_lock); + anon_vma_unlock(vma->anon_vma); + mem_cgroup_uncharge_page(new_page); + goto out; + } + + /* + * All pages are isolated and locked so anon_vma rmap + * can't run anymore. + */ + anon_vma_unlock(vma->anon_vma); + + __collapse_huge_page_copy(pte, new_page, vma, address, ptl); + __SetPageUptodate(new_page); + pgtable = pmd_pgtable(_pmd); + VM_BUG_ON(page_count(pgtable) != 1); + VM_BUG_ON(page_mapcount(pgtable) != 0); + + _pmd = mk_pmd(new_page, vma->vm_page_prot); + _pmd = maybe_pmd_mkwrite(pmd_mkdirty(_pmd), vma); + _pmd = pmd_mkhuge(_pmd); + + /* + * spin_lock() below is not the equivalent of smp_wmb(), so + * this is needed to avoid the copy_huge_page writes to become + * visible after the set_pmd_at() write. + */ + smp_wmb(); + + spin_lock(&mm->page_table_lock); + BUG_ON(!pmd_none(*pmd)); + page_add_new_anon_rmap(new_page, vma, address); + set_pmd_at(mm, address, pmd, _pmd); + update_mmu_cache(vma, address, entry); + prepare_pmd_huge_pte(pgtable, mm); + mm->nr_ptes--; + spin_unlock(&mm->page_table_lock); + + *hpage = NULL; + khugepaged_pages_collapsed++; +out: + up_write(&mm->mmap_sem); +} + +static int khugepaged_scan_pmd(struct mm_struct *mm, + struct vm_area_struct *vma, + unsigned long address, + struct page **hpage) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte, *_pte; + int ret = 0, referenced = 0, none = 0; + struct page *page; + unsigned long _address; + spinlock_t *ptl; + + VM_BUG_ON(address & ~HPAGE_PMD_MASK); + + pgd = pgd_offset(mm, address); + if (!pgd_present(*pgd)) + goto out; + + pud = pud_offset(pgd, address); + if (!pud_present(*pud)) + goto out; + + pmd = pmd_offset(pud, address); + if (!pmd_present(*pmd) || pmd_trans_huge(*pmd)) + goto out; + + pte = pte_offset_map_lock(mm, pmd, address, &ptl); + for (_address = address, _pte = pte; _pte < pte+HPAGE_PMD_NR; + _pte++, _address += PAGE_SIZE) { + pte_t pteval = *_pte; + if (pte_none(pteval)) { + if (++none <= khugepaged_max_ptes_none) + continue; + else + goto out_unmap; + } + if (!pte_present(pteval) || !pte_write(pteval)) + goto out_unmap; + page = vm_normal_page(vma, _address, pteval); + if (unlikely(!page)) + goto out_unmap; + VM_BUG_ON(PageCompound(page)); + if (!PageLRU(page) || PageLocked(page) || !PageAnon(page)) + goto out_unmap; + /* cannot use mapcount: can't collapse if there's a gup pin */ + if (page_count(page) != 1) + goto out_unmap; + if (pte_young(pteval)) + referenced = 1; + } + if (referenced) + ret = 1; +out_unmap: + pte_unmap_unlock(pte, ptl); + if (ret) { + up_read(&mm->mmap_sem); + collapse_huge_page(mm, address, hpage); + } +out: + return ret; +} + +static void collect_mm_slot(struct mm_slot *mm_slot) +{ + struct mm_struct *mm = mm_slot->mm; + + VM_BUG_ON(!spin_is_locked(&khugepaged_mm_lock)); + + if (khugepaged_test_exit(mm)) { + /* free mm_slot */ + hlist_del(&mm_slot->hash); + list_del(&mm_slot->mm_node); + + /* + * Not strictly needed because the mm exited already. + * + * clear_bit(MMF_VM_HUGEPAGE, &mm->flags); + */ + + /* khugepaged_mm_lock actually not necessary for the below */ + free_mm_slot(mm_slot); + mmdrop(mm); + } +} + +static unsigned int khugepaged_scan_mm_slot(unsigned int pages, + struct page **hpage) +{ + struct mm_slot *mm_slot; + struct mm_struct *mm; + struct vm_area_struct *vma; + int progress = 0; + + VM_BUG_ON(!pages); + VM_BUG_ON(!spin_is_locked(&khugepaged_mm_lock)); + + if (khugepaged_scan.mm_slot) + mm_slot = khugepaged_scan.mm_slot; + else { + mm_slot = list_entry(khugepaged_scan.mm_head.next, + struct mm_slot, mm_node); + khugepaged_scan.address = 0; + khugepaged_scan.mm_slot = mm_slot; + } + spin_unlock(&khugepaged_mm_lock); + + mm = mm_slot->mm; + down_read(&mm->mmap_sem); + if (unlikely(khugepaged_test_exit(mm))) + vma = NULL; + else + vma = find_vma(mm, khugepaged_scan.address); + + progress++; + for (; vma; vma = vma->vm_next) { + unsigned long hstart, hend; + + cond_resched(); + if (unlikely(khugepaged_test_exit(mm))) { + progress++; + break; + } + + if (!(vma->vm_flags & VM_HUGEPAGE) && + !khugepaged_always()) { + progress++; + continue; + } + + /* VM_PFNMAP vmas may have vm_ops null but vm_file set */ + if (!vma->anon_vma || vma->vm_ops || vma->vm_file) { + khugepaged_scan.address = vma->vm_end; + progress++; + continue; + } + VM_BUG_ON(is_linear_pfn_mapping(vma) || is_pfn_mapping(vma)); + + hstart = (vma->vm_start + ~HPAGE_PMD_MASK) & HPAGE_PMD_MASK; + hend = vma->vm_end & HPAGE_PMD_MASK; + if (hstart >= hend) { + progress++; + continue; + } + if (khugepaged_scan.address < hstart) + khugepaged_scan.address = hstart; + if (khugepaged_scan.address > hend) { + khugepaged_scan.address = hend + HPAGE_PMD_SIZE; + progress++; + continue; + } + BUG_ON(khugepaged_scan.address & ~HPAGE_PMD_MASK); + + while (khugepaged_scan.address < hend) { + int ret; + cond_resched(); + if (unlikely(khugepaged_test_exit(mm))) + goto breakouterloop; + + VM_BUG_ON(khugepaged_scan.address < hstart || + khugepaged_scan.address + HPAGE_PMD_SIZE > + hend); + ret = khugepaged_scan_pmd(mm, vma, + khugepaged_scan.address, + hpage); + /* move to next address */ + khugepaged_scan.address += HPAGE_PMD_SIZE; + progress += HPAGE_PMD_NR; + if (ret) + /* we released mmap_sem so break loop */ + goto breakouterloop_mmap_sem; + if (progress >= pages) + goto breakouterloop; + } + } +breakouterloop: + up_read(&mm->mmap_sem); /* exit_mmap will destroy ptes after this */ +breakouterloop_mmap_sem: + + spin_lock(&khugepaged_mm_lock); + BUG_ON(khugepaged_scan.mm_slot != mm_slot); + /* + * Release the current mm_slot if this mm is about to die, or + * if we scanned all vmas of this mm. + */ + if (khugepaged_test_exit(mm) || !vma) { + /* + * Make sure that if mm_users is reaching zero while + * khugepaged runs here, khugepaged_exit will find + * mm_slot not pointing to the exiting mm. + */ + if (mm_slot->mm_node.next != &khugepaged_scan.mm_head) { + khugepaged_scan.mm_slot = list_entry( + mm_slot->mm_node.next, + struct mm_slot, mm_node); + khugepaged_scan.address = 0; + } else { + khugepaged_scan.mm_slot = NULL; + khugepaged_full_scans++; + } + + collect_mm_slot(mm_slot); + } + + return progress; +} + +static int khugepaged_has_work(void) +{ + return !list_empty(&khugepaged_scan.mm_head) && + khugepaged_enabled(); +} + +static int khugepaged_wait_event(void) +{ + return !list_empty(&khugepaged_scan.mm_head) || + !khugepaged_enabled(); +} + +static void khugepaged_do_scan(struct page **hpage) +{ + unsigned int progress = 0, pass_through_head = 0; + unsigned int pages = khugepaged_pages_to_scan; + + barrier(); /* write khugepaged_pages_to_scan to local stack */ + + while (progress < pages) { + cond_resched(); + + if (!*hpage) { + *hpage = alloc_hugepage(khugepaged_defrag()); + if (unlikely(!*hpage)) + break; + } + + spin_lock(&khugepaged_mm_lock); + if (!khugepaged_scan.mm_slot) + pass_through_head++; + if (khugepaged_has_work() && + pass_through_head < 2) + progress += khugepaged_scan_mm_slot(pages - progress, + hpage); + else + progress = pages; + spin_unlock(&khugepaged_mm_lock); + } +} + +static struct page *khugepaged_alloc_hugepage(void) +{ + struct page *hpage; + + do { + hpage = alloc_hugepage(khugepaged_defrag()); + if (!hpage) { + DEFINE_WAIT(wait); + add_wait_queue(&khugepaged_wait, &wait); + schedule_timeout_interruptible( + msecs_to_jiffies( + khugepaged_alloc_sleep_millisecs)); + remove_wait_queue(&khugepaged_wait, &wait); + } + } while (unlikely(!hpage) && + likely(khugepaged_enabled())); + return hpage; +} + +static void khugepaged_loop(void) +{ + struct page *hpage; + + while (likely(khugepaged_enabled())) { + hpage = khugepaged_alloc_hugepage(); + if (unlikely(!hpage)) + break; + + khugepaged_do_scan(&hpage); + if (hpage) + put_page(hpage); + if (khugepaged_has_work()) { + DEFINE_WAIT(wait); + if (!khugepaged_scan_sleep_millisecs) + continue; + add_wait_queue(&khugepaged_wait, &wait); + schedule_timeout_interruptible( + msecs_to_jiffies( + khugepaged_scan_sleep_millisecs)); + remove_wait_queue(&khugepaged_wait, &wait); + } else if (khugepaged_enabled()) + wait_event_interruptible(khugepaged_wait, + khugepaged_wait_event()); + } +} + +static int khugepaged(void *none) +{ + struct mm_slot *mm_slot; + + set_user_nice(current, 19); + + /* serialize with start_khugepaged() */ + mutex_lock(&khugepaged_mutex); + + for (;;) { + mutex_unlock(&khugepaged_mutex); + BUG_ON(khugepaged_thread != current); + khugepaged_loop(); + BUG_ON(khugepaged_thread != current); + + mutex_lock(&khugepaged_mutex); + if (!khugepaged_enabled()) + break; + } + + spin_lock(&khugepaged_mm_lock); + mm_slot = khugepaged_scan.mm_slot; + khugepaged_scan.mm_slot = NULL; + if (mm_slot) + collect_mm_slot(mm_slot); + spin_unlock(&khugepaged_mm_lock); + + khugepaged_thread = NULL; + mutex_unlock(&khugepaged_mutex); + + return 0; +} + void __split_huge_page_pmd(struct mm_struct *mm, pmd_t *pmd) { struct page *page; -- cgit v1.2.3 From 5f24ce5fd34c3ca1b3d10d30da754732da64d5c0 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:47:00 -0800 Subject: thp: remove PG_buddy PG_buddy can be converted to _mapcount == -2. So the PG_compound_lock can be added to page->flags without overflowing (because of the sparse section bits increasing) with CONFIG_X86_PAE=y and CONFIG_X86_PAT=y. This also has to move the memory hotplug code from _mapcount to lru.next to avoid any risk of clashes. We can't use lru.next for PG_buddy removal, but memory hotplug can use lru.next even more easily than the mapcount instead. Signed-off-by: Andrea Arcangeli Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/page.c | 14 ++++++++------ include/linux/memory_hotplug.h | 14 +++++++++----- include/linux/mm.h | 21 +++++++++++++++++++++ include/linux/page-flags.h | 7 +------ mm/memory_hotplug.c | 14 ++++++++------ mm/page_alloc.c | 7 +++---- mm/sparse.c | 4 ++-- 7 files changed, 52 insertions(+), 29 deletions(-) (limited to 'include/linux') diff --git a/fs/proc/page.c b/fs/proc/page.c index b06c674624e6..6d8e6a9e93ab 100644 --- a/fs/proc/page.c +++ b/fs/proc/page.c @@ -116,15 +116,17 @@ u64 stable_page_flags(struct page *page) if (PageHuge(page)) u |= 1 << KPF_HUGE; - u |= kpf_copy_bit(k, KPF_LOCKED, PG_locked); - /* - * Caveats on high order pages: - * PG_buddy will only be set on the head page; SLUB/SLQB do the same - * for PG_slab; SLOB won't set PG_slab at all on compound pages. + * Caveats on high order pages: page->_count will only be set + * -1 on the head page; SLUB/SLQB do the same for PG_slab; + * SLOB won't set PG_slab at all on compound pages. */ + if (PageBuddy(page)) + u |= 1 << KPF_BUDDY; + + u |= kpf_copy_bit(k, KPF_LOCKED, PG_locked); + u |= kpf_copy_bit(k, KPF_SLAB, PG_slab); - u |= kpf_copy_bit(k, KPF_BUDDY, PG_buddy); u |= kpf_copy_bit(k, KPF_ERROR, PG_error); u |= kpf_copy_bit(k, KPF_DIRTY, PG_dirty); diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h index 31c237a00c48..24376fe7ee68 100644 --- a/include/linux/memory_hotplug.h +++ b/include/linux/memory_hotplug.h @@ -13,12 +13,16 @@ struct mem_section; #ifdef CONFIG_MEMORY_HOTPLUG /* - * Types for free bootmem. - * The normal smallest mapcount is -1. Here is smaller value than it. + * Types for free bootmem stored in page->lru.next. These have to be in + * some random range in unsigned long space for debugging purposes. */ -#define SECTION_INFO (-1 - 1) -#define MIX_SECTION_INFO (-1 - 2) -#define NODE_INFO (-1 - 3) +enum { + MEMORY_HOTPLUG_MIN_BOOTMEM_TYPE = 12, + SECTION_INFO = MEMORY_HOTPLUG_MIN_BOOTMEM_TYPE, + MIX_SECTION_INFO, + NODE_INFO, + MEMORY_HOTPLUG_MAX_BOOTMEM_TYPE = NODE_INFO, +}; /* * pgdat resizing functions diff --git a/include/linux/mm.h b/include/linux/mm.h index 2ec5138badab..7ab7d2b60041 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -397,6 +397,27 @@ static inline void init_page_count(struct page *page) atomic_set(&page->_count, 1); } +/* + * PageBuddy() indicate that the page is free and in the buddy system + * (see mm/page_alloc.c). + */ +static inline int PageBuddy(struct page *page) +{ + return atomic_read(&page->_mapcount) == -2; +} + +static inline void __SetPageBuddy(struct page *page) +{ + VM_BUG_ON(atomic_read(&page->_mapcount) != -1); + atomic_set(&page->_mapcount, -2); +} + +static inline void __ClearPageBuddy(struct page *page) +{ + VM_BUG_ON(!PageBuddy(page)); + atomic_set(&page->_mapcount, -1); +} + void put_page(struct page *page); void put_pages_list(struct list_head *pages); diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index 4ca1241ef94e..0db8037e2725 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -48,9 +48,6 @@ * struct page (these bits with information) are always mapped into kernel * address space... * - * PG_buddy is set to indicate that the page is free and in the buddy system - * (see mm/page_alloc.c). - * * PG_hwpoison indicates that a page got corrupted in hardware and contains * data with incorrect ECC bits that triggered a machine check. Accessing is * not safe since it may cause another machine check. Don't touch! @@ -96,7 +93,6 @@ enum pageflags { PG_swapcache, /* Swap page: swp_entry_t in private */ PG_mappedtodisk, /* Has blocks allocated on-disk */ PG_reclaim, /* To be reclaimed asap */ - PG_buddy, /* Page is free, on buddy lists */ PG_swapbacked, /* Page is backed by RAM/swap */ PG_unevictable, /* Page is "unevictable" */ #ifdef CONFIG_MMU @@ -233,7 +229,6 @@ PAGEFLAG(OwnerPriv1, owner_priv_1) TESTCLEARFLAG(OwnerPriv1, owner_priv_1) * risky: they bypass page accounting. */ TESTPAGEFLAG(Writeback, writeback) TESTSCFLAG(Writeback, writeback) -__PAGEFLAG(Buddy, buddy) PAGEFLAG(MappedToDisk, mappedtodisk) /* PG_readahead is only used for file reads; PG_reclaim is only for writes */ @@ -461,7 +456,7 @@ static inline int PageTransCompound(struct page *page) #define PAGE_FLAGS_CHECK_AT_FREE \ (1 << PG_lru | 1 << PG_locked | \ 1 << PG_private | 1 << PG_private_2 | \ - 1 << PG_buddy | 1 << PG_writeback | 1 << PG_reserved | \ + 1 << PG_writeback | 1 << PG_reserved | \ 1 << PG_slab | 1 << PG_swapcache | 1 << PG_active | \ 1 << PG_unevictable | __PG_MLOCKED | __PG_HWPOISON | \ __PG_COMPOUND_LOCK) diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index a2832c092509..e92f04749fcb 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -82,9 +82,10 @@ static void release_memory_resource(struct resource *res) #ifdef CONFIG_MEMORY_HOTPLUG_SPARSE #ifndef CONFIG_SPARSEMEM_VMEMMAP -static void get_page_bootmem(unsigned long info, struct page *page, int type) +static void get_page_bootmem(unsigned long info, struct page *page, + unsigned long type) { - atomic_set(&page->_mapcount, type); + page->lru.next = (struct list_head *) type; SetPagePrivate(page); set_page_private(page, info); atomic_inc(&page->_count); @@ -94,15 +95,16 @@ static void get_page_bootmem(unsigned long info, struct page *page, int type) * so use __ref to tell modpost not to generate a warning */ void __ref put_page_bootmem(struct page *page) { - int type; + unsigned long type; - type = atomic_read(&page->_mapcount); - BUG_ON(type >= -1); + type = (unsigned long) page->lru.next; + BUG_ON(type < MEMORY_HOTPLUG_MIN_BOOTMEM_TYPE || + type > MEMORY_HOTPLUG_MAX_BOOTMEM_TYPE); if (atomic_dec_return(&page->_count) == 1) { ClearPagePrivate(page); set_page_private(page, 0); - reset_page_mapcount(page); + INIT_LIST_HEAD(&page->lru); __free_pages_bootmem(page, 0); } diff --git a/mm/page_alloc.c b/mm/page_alloc.c index e7664b9f706c..9dfe49bceff4 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -449,8 +449,8 @@ __find_combined_index(unsigned long page_idx, unsigned int order) * (c) a page and its buddy have the same order && * (d) a page and its buddy are in the same zone. * - * For recording whether a page is in the buddy system, we use PG_buddy. - * Setting, clearing, and testing PG_buddy is serialized by zone->lock. + * For recording whether a page is in the buddy system, we set ->_mapcount -2. + * Setting, clearing, and testing _mapcount -2 is serialized by zone->lock. * * For recording page's order, we use page_private(page). */ @@ -483,7 +483,7 @@ static inline int page_is_buddy(struct page *page, struct page *buddy, * as necessary, plus some accounting needed to play nicely with other * parts of the VM system. * At each level, we keep a list of pages, which are heads of continuous - * free pages of length of (1 << order) and marked with PG_buddy. Page's + * free pages of length of (1 << order) and marked with _mapcount -2. Page's * order is recorded in page_private(page) field. * So when we are allocating or freeing one, we can derive the state of the * other. That is, if we allocate a small block, and both were @@ -5574,7 +5574,6 @@ static struct trace_print_flags pageflag_names[] = { {1UL << PG_swapcache, "swapcache" }, {1UL << PG_mappedtodisk, "mappedtodisk" }, {1UL << PG_reclaim, "reclaim" }, - {1UL << PG_buddy, "buddy" }, {1UL << PG_swapbacked, "swapbacked" }, {1UL << PG_unevictable, "unevictable" }, #ifdef CONFIG_MMU diff --git a/mm/sparse.c b/mm/sparse.c index 95ac219af379..93250207c5cf 100644 --- a/mm/sparse.c +++ b/mm/sparse.c @@ -671,10 +671,10 @@ static void __kfree_section_memmap(struct page *memmap, unsigned long nr_pages) static void free_map_bootmem(struct page *page, unsigned long nr_pages) { unsigned long maps_section_nr, removing_section_nr, i; - int magic; + unsigned long magic; for (i = 0; i < nr_pages; i++, page++) { - magic = atomic_read(&page->_mapcount); + magic = (unsigned long) page->lru.next; BUG_ON(magic == NODE_INFO); -- cgit v1.2.3 From f2d6bfe9ff0acec30b713614260e78b03d20e909 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Thu, 13 Jan 2011 15:47:01 -0800 Subject: thp: add x86 32bit support Add support for transparent hugepages to x86 32bit. Share the same VM_ bitflag for VM_MAPPED_COPY. mm/nommu.c will never support transparent hugepages. Signed-off-by: Johannes Weiner Signed-off-by: Andrea Arcangeli Reviewed-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/include/asm/pgtable-2level.h | 9 +++ arch/x86/include/asm/pgtable-3level.h | 23 +++++++ arch/x86/include/asm/pgtable.h | 117 ++++++++++++++++++++++++++++++++++ arch/x86/include/asm/pgtable_64.h | 109 ------------------------------- arch/x86/mm/pgtable.c | 4 +- include/linux/mm.h | 7 +- mm/Kconfig | 2 +- 7 files changed, 156 insertions(+), 115 deletions(-) (limited to 'include/linux') diff --git a/arch/x86/include/asm/pgtable-2level.h b/arch/x86/include/asm/pgtable-2level.h index 2334982b339e..98391db840c6 100644 --- a/arch/x86/include/asm/pgtable-2level.h +++ b/arch/x86/include/asm/pgtable-2level.h @@ -46,6 +46,15 @@ static inline pte_t native_ptep_get_and_clear(pte_t *xp) #define native_ptep_get_and_clear(xp) native_local_ptep_get_and_clear(xp) #endif +#ifdef CONFIG_SMP +static inline pmd_t native_pmdp_get_and_clear(pmd_t *xp) +{ + return __pmd(xchg((pmdval_t *)xp, 0)); +} +#else +#define native_pmdp_get_and_clear(xp) native_local_pmdp_get_and_clear(xp) +#endif + /* * Bits _PAGE_BIT_PRESENT, _PAGE_BIT_FILE and _PAGE_BIT_PROTNONE are taken, * split up the 29 bits of offset into this range: diff --git a/arch/x86/include/asm/pgtable-3level.h b/arch/x86/include/asm/pgtable-3level.h index 177b0165ea01..94b979d1b58d 100644 --- a/arch/x86/include/asm/pgtable-3level.h +++ b/arch/x86/include/asm/pgtable-3level.h @@ -104,6 +104,29 @@ static inline pte_t native_ptep_get_and_clear(pte_t *ptep) #define native_ptep_get_and_clear(xp) native_local_ptep_get_and_clear(xp) #endif +#ifdef CONFIG_SMP +union split_pmd { + struct { + u32 pmd_low; + u32 pmd_high; + }; + pmd_t pmd; +}; +static inline pmd_t native_pmdp_get_and_clear(pmd_t *pmdp) +{ + union split_pmd res, *orig = (union split_pmd *)pmdp; + + /* xchg acts as a barrier before setting of the high bits */ + res.pmd_low = xchg(&orig->pmd_low, 0); + res.pmd_high = orig->pmd_high; + orig->pmd_high = 0; + + return res.pmd; +} +#else +#define native_pmdp_get_and_clear(xp) native_local_pmdp_get_and_clear(xp) +#endif + /* * Bits 0, 6 and 7 are taken in the low part of the pte, * put the 32 bits of offset into the high part. diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h index 3278038e9706..001a3831567a 100644 --- a/arch/x86/include/asm/pgtable.h +++ b/arch/x86/include/asm/pgtable.h @@ -97,6 +97,11 @@ static inline int pte_young(pte_t pte) return pte_flags(pte) & _PAGE_ACCESSED; } +static inline int pmd_young(pmd_t pmd) +{ + return pmd_flags(pmd) & _PAGE_ACCESSED; +} + static inline int pte_write(pte_t pte) { return pte_flags(pte) & _PAGE_RW; @@ -145,6 +150,18 @@ static inline int pmd_large(pmd_t pte) (_PAGE_PSE | _PAGE_PRESENT); } +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +static inline int pmd_trans_splitting(pmd_t pmd) +{ + return pmd_val(pmd) & _PAGE_SPLITTING; +} + +static inline int pmd_trans_huge(pmd_t pmd) +{ + return pmd_val(pmd) & _PAGE_PSE; +} +#endif /* CONFIG_TRANSPARENT_HUGEPAGE */ + static inline pte_t pte_set_flags(pte_t pte, pteval_t set) { pteval_t v = native_pte_val(pte); @@ -219,6 +236,55 @@ static inline pte_t pte_mkspecial(pte_t pte) return pte_set_flags(pte, _PAGE_SPECIAL); } +static inline pmd_t pmd_set_flags(pmd_t pmd, pmdval_t set) +{ + pmdval_t v = native_pmd_val(pmd); + + return __pmd(v | set); +} + +static inline pmd_t pmd_clear_flags(pmd_t pmd, pmdval_t clear) +{ + pmdval_t v = native_pmd_val(pmd); + + return __pmd(v & ~clear); +} + +static inline pmd_t pmd_mkold(pmd_t pmd) +{ + return pmd_clear_flags(pmd, _PAGE_ACCESSED); +} + +static inline pmd_t pmd_wrprotect(pmd_t pmd) +{ + return pmd_clear_flags(pmd, _PAGE_RW); +} + +static inline pmd_t pmd_mkdirty(pmd_t pmd) +{ + return pmd_set_flags(pmd, _PAGE_DIRTY); +} + +static inline pmd_t pmd_mkhuge(pmd_t pmd) +{ + return pmd_set_flags(pmd, _PAGE_PSE); +} + +static inline pmd_t pmd_mkyoung(pmd_t pmd) +{ + return pmd_set_flags(pmd, _PAGE_ACCESSED); +} + +static inline pmd_t pmd_mkwrite(pmd_t pmd) +{ + return pmd_set_flags(pmd, _PAGE_RW); +} + +static inline pmd_t pmd_mknotpresent(pmd_t pmd) +{ + return pmd_clear_flags(pmd, _PAGE_PRESENT); +} + /* * Mask out unsupported bits in a present pgprot. Non-present pgprots * can use those bits for other purposes, so leave them be. @@ -527,6 +593,14 @@ static inline pte_t native_local_ptep_get_and_clear(pte_t *ptep) return res; } +static inline pmd_t native_local_pmdp_get_and_clear(pmd_t *pmdp) +{ + pmd_t res = *pmdp; + + native_pmd_clear(pmdp); + return res; +} + static inline void native_set_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep , pte_t pte) { @@ -616,6 +690,49 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, #define flush_tlb_fix_spurious_fault(vma, address) +#define mk_pmd(page, pgprot) pfn_pmd(page_to_pfn(page), (pgprot)) + +#define __HAVE_ARCH_PMDP_SET_ACCESS_FLAGS +extern int pmdp_set_access_flags(struct vm_area_struct *vma, + unsigned long address, pmd_t *pmdp, + pmd_t entry, int dirty); + +#define __HAVE_ARCH_PMDP_TEST_AND_CLEAR_YOUNG +extern int pmdp_test_and_clear_young(struct vm_area_struct *vma, + unsigned long addr, pmd_t *pmdp); + +#define __HAVE_ARCH_PMDP_CLEAR_YOUNG_FLUSH +extern int pmdp_clear_flush_young(struct vm_area_struct *vma, + unsigned long address, pmd_t *pmdp); + + +#define __HAVE_ARCH_PMDP_SPLITTING_FLUSH +extern void pmdp_splitting_flush(struct vm_area_struct *vma, + unsigned long addr, pmd_t *pmdp); + +#define __HAVE_ARCH_PMD_WRITE +static inline int pmd_write(pmd_t pmd) +{ + return pmd_flags(pmd) & _PAGE_RW; +} + +#define __HAVE_ARCH_PMDP_GET_AND_CLEAR +static inline pmd_t pmdp_get_and_clear(struct mm_struct *mm, unsigned long addr, + pmd_t *pmdp) +{ + pmd_t pmd = native_pmdp_get_and_clear(pmdp); + pmd_update(mm, addr, pmdp); + return pmd; +} + +#define __HAVE_ARCH_PMDP_SET_WRPROTECT +static inline void pmdp_set_wrprotect(struct mm_struct *mm, + unsigned long addr, pmd_t *pmdp) +{ + clear_bit(_PAGE_BIT_RW, (unsigned long *)pmdp); + pmd_update(mm, addr, pmdp); +} + /* * clone_pgd_range(pgd_t *dst, pgd_t *src, int count); * diff --git a/arch/x86/include/asm/pgtable_64.h b/arch/x86/include/asm/pgtable_64.h index b2df039a4119..975f709e09ae 100644 --- a/arch/x86/include/asm/pgtable_64.h +++ b/arch/x86/include/asm/pgtable_64.h @@ -182,115 +182,6 @@ extern void cleanup_highmap(void); #define __HAVE_ARCH_PTE_SAME -#ifdef CONFIG_TRANSPARENT_HUGEPAGE -static inline int pmd_trans_splitting(pmd_t pmd) -{ - return pmd_val(pmd) & _PAGE_SPLITTING; -} - -static inline int pmd_trans_huge(pmd_t pmd) -{ - return pmd_val(pmd) & _PAGE_PSE; -} -#endif /* CONFIG_TRANSPARENT_HUGEPAGE */ - -#define mk_pmd(page, pgprot) pfn_pmd(page_to_pfn(page), (pgprot)) - -#define __HAVE_ARCH_PMDP_SET_ACCESS_FLAGS -extern int pmdp_set_access_flags(struct vm_area_struct *vma, - unsigned long address, pmd_t *pmdp, - pmd_t entry, int dirty); - -#define __HAVE_ARCH_PMDP_TEST_AND_CLEAR_YOUNG -extern int pmdp_test_and_clear_young(struct vm_area_struct *vma, - unsigned long addr, pmd_t *pmdp); - -#define __HAVE_ARCH_PMDP_CLEAR_YOUNG_FLUSH -extern int pmdp_clear_flush_young(struct vm_area_struct *vma, - unsigned long address, pmd_t *pmdp); - - -#define __HAVE_ARCH_PMDP_SPLITTING_FLUSH -extern void pmdp_splitting_flush(struct vm_area_struct *vma, - unsigned long addr, pmd_t *pmdp); - -#define __HAVE_ARCH_PMD_WRITE -static inline int pmd_write(pmd_t pmd) -{ - return pmd_flags(pmd) & _PAGE_RW; -} - -#define __HAVE_ARCH_PMDP_GET_AND_CLEAR -static inline pmd_t pmdp_get_and_clear(struct mm_struct *mm, unsigned long addr, - pmd_t *pmdp) -{ - pmd_t pmd = native_pmdp_get_and_clear(pmdp); - pmd_update(mm, addr, pmdp); - return pmd; -} - -#define __HAVE_ARCH_PMDP_SET_WRPROTECT -static inline void pmdp_set_wrprotect(struct mm_struct *mm, - unsigned long addr, pmd_t *pmdp) -{ - clear_bit(_PAGE_BIT_RW, (unsigned long *)&pmdp->pmd); - pmd_update(mm, addr, pmdp); -} - -static inline int pmd_young(pmd_t pmd) -{ - return pmd_flags(pmd) & _PAGE_ACCESSED; -} - -static inline pmd_t pmd_set_flags(pmd_t pmd, pmdval_t set) -{ - pmdval_t v = native_pmd_val(pmd); - - return native_make_pmd(v | set); -} - -static inline pmd_t pmd_clear_flags(pmd_t pmd, pmdval_t clear) -{ - pmdval_t v = native_pmd_val(pmd); - - return native_make_pmd(v & ~clear); -} - -static inline pmd_t pmd_mkold(pmd_t pmd) -{ - return pmd_clear_flags(pmd, _PAGE_ACCESSED); -} - -static inline pmd_t pmd_wrprotect(pmd_t pmd) -{ - return pmd_clear_flags(pmd, _PAGE_RW); -} - -static inline pmd_t pmd_mkdirty(pmd_t pmd) -{ - return pmd_set_flags(pmd, _PAGE_DIRTY); -} - -static inline pmd_t pmd_mkhuge(pmd_t pmd) -{ - return pmd_set_flags(pmd, _PAGE_PSE); -} - -static inline pmd_t pmd_mkyoung(pmd_t pmd) -{ - return pmd_set_flags(pmd, _PAGE_ACCESSED); -} - -static inline pmd_t pmd_mkwrite(pmd_t pmd) -{ - return pmd_set_flags(pmd, _PAGE_RW); -} - -static inline pmd_t pmd_mknotpresent(pmd_t pmd) -{ - return pmd_clear_flags(pmd, _PAGE_PRESENT); -} - #endif /* !__ASSEMBLY__ */ #endif /* _ASM_X86_PGTABLE_64_H */ diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c index 65e92d58f942..500242d3c96d 100644 --- a/arch/x86/mm/pgtable.c +++ b/arch/x86/mm/pgtable.c @@ -362,7 +362,7 @@ int pmdp_test_and_clear_young(struct vm_area_struct *vma, if (pmd_young(*pmdp)) ret = test_and_clear_bit(_PAGE_BIT_ACCESSED, - (unsigned long *) &pmdp->pmd); + (unsigned long *)pmdp); if (ret) pmd_update(vma->vm_mm, addr, pmdp); @@ -404,7 +404,7 @@ void pmdp_splitting_flush(struct vm_area_struct *vma, int set; VM_BUG_ON(address & ~HPAGE_PMD_MASK); set = !test_and_set_bit(_PAGE_BIT_SPLITTING, - (unsigned long *)&pmdp->pmd); + (unsigned long *)pmdp); if (set) { pmd_update(vma->vm_mm, address, pmdp); /* need tlb flush only to serialize against gup-fast */ diff --git a/include/linux/mm.h b/include/linux/mm.h index 7ab7d2b60041..9c2695beab86 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -102,7 +102,11 @@ extern unsigned int kobjsize(const void *objp); #define VM_NORESERVE 0x00200000 /* should the VM suppress accounting */ #define VM_HUGETLB 0x00400000 /* Huge TLB Page VM */ #define VM_NONLINEAR 0x00800000 /* Is non-linear (remap_file_pages) */ +#ifndef CONFIG_TRANSPARENT_HUGEPAGE #define VM_MAPPED_COPY 0x01000000 /* T if mapped copy of data (nommu mmap) */ +#else +#define VM_HUGEPAGE 0x01000000 /* MADV_HUGEPAGE marked this vma */ +#endif #define VM_INSERTPAGE 0x02000000 /* The vma has had "vm_insert_page()" done on it */ #define VM_ALWAYSDUMP 0x04000000 /* Always include in core dumps */ @@ -111,9 +115,6 @@ extern unsigned int kobjsize(const void *objp); #define VM_SAO 0x20000000 /* Strong Access Ordering (powerpc) */ #define VM_PFN_AT_MMAP 0x40000000 /* PFNMAP vma that is fully mapped at mmap time */ #define VM_MERGEABLE 0x80000000 /* KSM may merge identical pages */ -#if BITS_PER_LONG > 32 -#define VM_HUGEPAGE 0x100000000UL /* MADV_HUGEPAGE marked this vma */ -#endif /* Bits set in the VMA until the stack is in its final location */ #define VM_STACK_INCOMPLETE_SETUP (VM_RAND_READ | VM_SEQ_READ) diff --git a/mm/Kconfig b/mm/Kconfig index 3982be2d721a..d774f77538ce 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -304,7 +304,7 @@ config NOMMU_INITIAL_TRIM_EXCESS config TRANSPARENT_HUGEPAGE bool "Transparent Hugepage Support" if EMBEDDED - depends on X86_64 && MMU + depends on X86 && MMU default y help Transparent Hugepages allows the kernel to use huge pages and -- cgit v1.2.3 From 0ca1634d4143c3579273ca53b993df19f5c98e92 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Thu, 13 Jan 2011 15:47:02 -0800 Subject: thp: mincore transparent hugepage support Handle transparent huge page pmd entries natively instead of splitting them into subpages. Signed-off-by: Johannes Weiner Signed-off-by: Andrea Arcangeli Reviewed-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/huge_mm.h | 3 +++ mm/huge_memory.c | 25 +++++++++++++++++++++++++ mm/mincore.c | 8 +++++++- 3 files changed, 35 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h index 43a694ef8904..25125fb6acf7 100644 --- a/include/linux/huge_mm.h +++ b/include/linux/huge_mm.h @@ -19,6 +19,9 @@ extern struct page *follow_trans_huge_pmd(struct mm_struct *mm, extern int zap_huge_pmd(struct mmu_gather *tlb, struct vm_area_struct *vma, pmd_t *pmd); +extern int mincore_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd, + unsigned long addr, unsigned long end, + unsigned char *vec); enum transparent_hugepage_flag { TRANSPARENT_HUGEPAGE_FLAG, diff --git a/mm/huge_memory.c b/mm/huge_memory.c index ae2bf08b1099..37e89a32a0b1 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -923,6 +923,31 @@ int zap_huge_pmd(struct mmu_gather *tlb, struct vm_area_struct *vma, return ret; } +int mincore_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd, + unsigned long addr, unsigned long end, + unsigned char *vec) +{ + int ret = 0; + + spin_lock(&vma->vm_mm->page_table_lock); + if (likely(pmd_trans_huge(*pmd))) { + ret = !pmd_trans_splitting(*pmd); + spin_unlock(&vma->vm_mm->page_table_lock); + if (unlikely(!ret)) + wait_split_huge_page(vma->anon_vma, pmd); + else { + /* + * All logical pages in the range are present + * if backed by a huge page. + */ + memset(vec, 1, (end - addr) >> PAGE_SHIFT); + } + } else + spin_unlock(&vma->vm_mm->page_table_lock); + + return ret; +} + pmd_t *page_check_address_pmd(struct page *page, struct mm_struct *mm, unsigned long address, diff --git a/mm/mincore.c b/mm/mincore.c index 9959bb41570e..a4e6b9d75c76 100644 --- a/mm/mincore.c +++ b/mm/mincore.c @@ -154,7 +154,13 @@ static void mincore_pmd_range(struct vm_area_struct *vma, pud_t *pud, pmd = pmd_offset(pud, addr); do { next = pmd_addr_end(addr, end); - split_huge_page_pmd(vma->vm_mm, pmd); + if (pmd_trans_huge(*pmd)) { + if (mincore_huge_pmd(vma, pmd, addr, next, vec)) { + vec += (next - addr) >> PAGE_SHIFT; + continue; + } + /* fall through */ + } if (pmd_none_or_clear_bad(pmd)) mincore_unmapped_range(vma, addr, next, vec); else -- cgit v1.2.3 From cd7548ab360c462118568eebb8c6da3bc303b02e Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Thu, 13 Jan 2011 15:47:04 -0800 Subject: thp: mprotect: transparent huge page support Natively handle huge pmds when changing page tables on behalf of mprotect(). I left out update_mmu_cache() because we do not need it on x86 anyway but more importantly the interface works on ptes, not pmds. Signed-off-by: Johannes Weiner Signed-off-by: Andrea Arcangeli Reviewed-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/huge_mm.h | 2 ++ mm/huge_memory.c | 27 +++++++++++++++++++++++++++ mm/mprotect.c | 8 +++++++- 3 files changed, 36 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h index 25125fb6acf7..c590b08c6fa6 100644 --- a/include/linux/huge_mm.h +++ b/include/linux/huge_mm.h @@ -22,6 +22,8 @@ extern int zap_huge_pmd(struct mmu_gather *tlb, extern int mincore_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd, unsigned long addr, unsigned long end, unsigned char *vec); +extern int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd, + unsigned long addr, pgprot_t newprot); enum transparent_hugepage_flag { TRANSPARENT_HUGEPAGE_FLAG, diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 37e89a32a0b1..7b55fe0e998b 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -948,6 +948,33 @@ int mincore_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd, return ret; } +int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd, + unsigned long addr, pgprot_t newprot) +{ + struct mm_struct *mm = vma->vm_mm; + int ret = 0; + + spin_lock(&mm->page_table_lock); + if (likely(pmd_trans_huge(*pmd))) { + if (unlikely(pmd_trans_splitting(*pmd))) { + spin_unlock(&mm->page_table_lock); + wait_split_huge_page(vma->anon_vma, pmd); + } else { + pmd_t entry; + + entry = pmdp_get_and_clear(mm, addr, pmd); + entry = pmd_modify(entry, newprot); + set_pmd_at(mm, addr, pmd, entry); + spin_unlock(&vma->vm_mm->page_table_lock); + flush_tlb_range(vma, addr, addr + HPAGE_PMD_SIZE); + ret = 1; + } + } else + spin_unlock(&vma->vm_mm->page_table_lock); + + return ret; +} + pmd_t *page_check_address_pmd(struct page *page, struct mm_struct *mm, unsigned long address, diff --git a/mm/mprotect.c b/mm/mprotect.c index 9402080d0d4a..5a688a2756be 100644 --- a/mm/mprotect.c +++ b/mm/mprotect.c @@ -88,7 +88,13 @@ static inline void change_pmd_range(struct vm_area_struct *vma, pud_t *pud, pmd = pmd_offset(pud, addr); do { next = pmd_addr_end(addr, end); - split_huge_page_pmd(vma->vm_mm, pmd); + if (pmd_trans_huge(*pmd)) { + if (next - addr != HPAGE_PMD_SIZE) + split_huge_page_pmd(vma->vm_mm, pmd); + else if (change_huge_pmd(vma, pmd, addr, newprot)) + continue; + /* fall through */ + } if (pmd_none_or_clear_bad(pmd)) continue; change_pte_range(vma->vm_mm, pmd, addr, next, newprot, -- cgit v1.2.3 From 0bbbc0b33d141f78a0d9218a54a47f50621220d3 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:47:05 -0800 Subject: thp: add numa awareness to hugepage allocations It's mostly a matter of replacing alloc_pages with alloc_pages_vma after introducing alloc_pages_vma. khugepaged needs special handling as the allocation has to happen inside collapse_huge_page where the vma is known and an error has to be returned to the outer loop to sleep alloc_sleep_millisecs in case of failure. But it retains the more efficient logic of handling allocation failures in khugepaged in case of CONFIG_NUMA=n. Signed-off-by: Andrea Arcangeli Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/gfp.h | 7 +++-- mm/huge_memory.c | 87 +++++++++++++++++++++++++++++++++++++++++++++-------- mm/mempolicy.c | 13 +++++--- 3 files changed, 87 insertions(+), 20 deletions(-) (limited to 'include/linux') diff --git a/include/linux/gfp.h b/include/linux/gfp.h index d95082cc6f4a..a3b148a91874 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -331,14 +331,17 @@ alloc_pages(gfp_t gfp_mask, unsigned int order) { return alloc_pages_current(gfp_mask, order); } -extern struct page *alloc_page_vma(gfp_t gfp_mask, +extern struct page *alloc_pages_vma(gfp_t gfp_mask, int order, struct vm_area_struct *vma, unsigned long addr); #else #define alloc_pages(gfp_mask, order) \ alloc_pages_node(numa_node_id(), gfp_mask, order) -#define alloc_page_vma(gfp_mask, vma, addr) alloc_pages(gfp_mask, 0) +#define alloc_pages_vma(gfp_mask, order, vma, addr) \ + alloc_pages(gfp_mask, order) #endif #define alloc_page(gfp_mask) alloc_pages(gfp_mask, 0) +#define alloc_page_vma(gfp_mask, vma, addr) \ + alloc_pages_vma(gfp_mask, 0, vma, addr) extern unsigned long __get_free_pages(gfp_t gfp_mask, unsigned int order); extern unsigned long get_zeroed_page(gfp_t gfp_mask); diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 0415a83afd66..f6559e7711bd 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -620,11 +620,26 @@ static int __do_huge_pmd_anonymous_page(struct mm_struct *mm, return ret; } +static inline gfp_t alloc_hugepage_gfpmask(int defrag) +{ + return GFP_TRANSHUGE & ~(defrag ? 0 : __GFP_WAIT); +} + +static inline struct page *alloc_hugepage_vma(int defrag, + struct vm_area_struct *vma, + unsigned long haddr) +{ + return alloc_pages_vma(alloc_hugepage_gfpmask(defrag), + HPAGE_PMD_ORDER, vma, haddr); +} + +#ifndef CONFIG_NUMA static inline struct page *alloc_hugepage(int defrag) { - return alloc_pages(GFP_TRANSHUGE & ~(defrag ? 0 : __GFP_WAIT), + return alloc_pages(alloc_hugepage_gfpmask(defrag), HPAGE_PMD_ORDER); } +#endif int do_huge_pmd_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, pmd_t *pmd, @@ -639,7 +654,8 @@ int do_huge_pmd_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma, return VM_FAULT_OOM; if (unlikely(khugepaged_enter(vma))) return VM_FAULT_OOM; - page = alloc_hugepage(transparent_hugepage_defrag(vma)); + page = alloc_hugepage_vma(transparent_hugepage_defrag(vma), + vma, haddr); if (unlikely(!page)) goto out; if (unlikely(mem_cgroup_newpage_charge(page, mm, GFP_KERNEL))) { @@ -862,7 +878,8 @@ int do_huge_pmd_wp_page(struct mm_struct *mm, struct vm_area_struct *vma, if (transparent_hugepage_enabled(vma) && !transparent_hugepage_debug_cow()) - new_page = alloc_hugepage(transparent_hugepage_defrag(vma)); + new_page = alloc_hugepage_vma(transparent_hugepage_defrag(vma), + vma, haddr); else new_page = NULL; @@ -1661,7 +1678,11 @@ static void collapse_huge_page(struct mm_struct *mm, unsigned long hstart, hend; VM_BUG_ON(address & ~HPAGE_PMD_MASK); +#ifndef CONFIG_NUMA VM_BUG_ON(!*hpage); +#else + VM_BUG_ON(*hpage); +#endif /* * Prevent all access to pagetables with the exception of @@ -1699,9 +1720,17 @@ static void collapse_huge_page(struct mm_struct *mm, if (!pmd_present(*pmd) || pmd_trans_huge(*pmd)) goto out; +#ifndef CONFIG_NUMA new_page = *hpage; - if (unlikely(mem_cgroup_newpage_charge(new_page, mm, GFP_KERNEL))) +#else + new_page = alloc_hugepage_vma(khugepaged_defrag(), vma, address); + if (unlikely(!new_page)) { + *hpage = ERR_PTR(-ENOMEM); goto out; + } +#endif + if (unlikely(mem_cgroup_newpage_charge(new_page, mm, GFP_KERNEL))) + goto out_put_page; anon_vma_lock(vma->anon_vma); @@ -1730,7 +1759,7 @@ static void collapse_huge_page(struct mm_struct *mm, spin_unlock(&mm->page_table_lock); anon_vma_unlock(vma->anon_vma); mem_cgroup_uncharge_page(new_page); - goto out; + goto out_put_page; } /* @@ -1765,10 +1794,19 @@ static void collapse_huge_page(struct mm_struct *mm, mm->nr_ptes--; spin_unlock(&mm->page_table_lock); +#ifndef CONFIG_NUMA *hpage = NULL; +#endif khugepaged_pages_collapsed++; out: up_write(&mm->mmap_sem); + return; + +out_put_page: +#ifdef CONFIG_NUMA + put_page(new_page); +#endif + goto out; } static int khugepaged_scan_pmd(struct mm_struct *mm, @@ -2001,11 +2039,16 @@ static void khugepaged_do_scan(struct page **hpage) while (progress < pages) { cond_resched(); +#ifndef CONFIG_NUMA if (!*hpage) { *hpage = alloc_hugepage(khugepaged_defrag()); if (unlikely(!*hpage)) break; } +#else + if (IS_ERR(*hpage)) + break; +#endif spin_lock(&khugepaged_mm_lock); if (!khugepaged_scan.mm_slot) @@ -2020,37 +2063,55 @@ static void khugepaged_do_scan(struct page **hpage) } } +static void khugepaged_alloc_sleep(void) +{ + DEFINE_WAIT(wait); + add_wait_queue(&khugepaged_wait, &wait); + schedule_timeout_interruptible( + msecs_to_jiffies( + khugepaged_alloc_sleep_millisecs)); + remove_wait_queue(&khugepaged_wait, &wait); +} + +#ifndef CONFIG_NUMA static struct page *khugepaged_alloc_hugepage(void) { struct page *hpage; do { hpage = alloc_hugepage(khugepaged_defrag()); - if (!hpage) { - DEFINE_WAIT(wait); - add_wait_queue(&khugepaged_wait, &wait); - schedule_timeout_interruptible( - msecs_to_jiffies( - khugepaged_alloc_sleep_millisecs)); - remove_wait_queue(&khugepaged_wait, &wait); - } + if (!hpage) + khugepaged_alloc_sleep(); } while (unlikely(!hpage) && likely(khugepaged_enabled())); return hpage; } +#endif static void khugepaged_loop(void) { struct page *hpage; +#ifdef CONFIG_NUMA + hpage = NULL; +#endif while (likely(khugepaged_enabled())) { +#ifndef CONFIG_NUMA hpage = khugepaged_alloc_hugepage(); if (unlikely(!hpage)) break; +#else + if (IS_ERR(hpage)) { + khugepaged_alloc_sleep(); + hpage = NULL; + } +#endif khugepaged_do_scan(&hpage); +#ifndef CONFIG_NUMA if (hpage) put_page(hpage); +#endif if (khugepaged_has_work()) { DEFINE_WAIT(wait); if (!khugepaged_scan_sleep_millisecs) diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 83b7df309fc4..368fc9d23610 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -1796,7 +1796,7 @@ static struct page *alloc_page_interleave(gfp_t gfp, unsigned order, } /** - * alloc_page_vma - Allocate a page for a VMA. + * alloc_pages_vma - Allocate a page for a VMA. * * @gfp: * %GFP_USER user allocation. @@ -1805,6 +1805,7 @@ static struct page *alloc_page_interleave(gfp_t gfp, unsigned order, * %GFP_FS allocation should not call back into a file system. * %GFP_ATOMIC don't sleep. * + * @order:Order of the GFP allocation. * @vma: Pointer to VMA or NULL if not available. * @addr: Virtual Address of the allocation. Must be inside the VMA. * @@ -1818,7 +1819,8 @@ static struct page *alloc_page_interleave(gfp_t gfp, unsigned order, * Should be called with the mm_sem of the vma hold. */ struct page * -alloc_page_vma(gfp_t gfp, struct vm_area_struct *vma, unsigned long addr) +alloc_pages_vma(gfp_t gfp, int order, struct vm_area_struct *vma, + unsigned long addr) { struct mempolicy *pol = get_vma_policy(current, vma, addr); struct zonelist *zl; @@ -1830,7 +1832,7 @@ alloc_page_vma(gfp_t gfp, struct vm_area_struct *vma, unsigned long addr) nid = interleave_nid(pol, vma, addr, PAGE_SHIFT); mpol_cond_put(pol); - page = alloc_page_interleave(gfp, 0, nid); + page = alloc_page_interleave(gfp, order, nid); put_mems_allowed(); return page; } @@ -1839,7 +1841,7 @@ alloc_page_vma(gfp_t gfp, struct vm_area_struct *vma, unsigned long addr) /* * slow path: ref counted shared policy */ - struct page *page = __alloc_pages_nodemask(gfp, 0, + struct page *page = __alloc_pages_nodemask(gfp, order, zl, policy_nodemask(gfp, pol)); __mpol_put(pol); put_mems_allowed(); @@ -1848,7 +1850,8 @@ alloc_page_vma(gfp_t gfp, struct vm_area_struct *vma, unsigned long addr) /* * fast path: default or task policy */ - page = __alloc_pages_nodemask(gfp, 0, zl, policy_nodemask(gfp, pol)); + page = __alloc_pages_nodemask(gfp, order, zl, + policy_nodemask(gfp, pol)); put_mems_allowed(); return page; } -- cgit v1.2.3 From 94fcc585fb85ad7b059c70872489b50044d401f3 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:47:08 -0800 Subject: thp: avoid breaking huge pmd invariants in case of vma_adjust failures An huge pmd can only be mapped if the corresponding 2M virtual range is fully contained in the vma. At times the VM calls split_vma twice, if the first split_vma succeeds and the second fail, the first split_vma remains in effect and it's not rolled back. For split_vma or vma_adjust to fail an allocation failure is needed so it's a very unlikely event (the out of memory killer would normally fire before any allocation failure is visible to kernel and userland and if an out of memory condition happens it's unlikely to happen exactly here). Nevertheless it's safer to ensure that no huge pmd can be left around if the vma is adjusted in a way that can't fit hugepages anymore at the new vm_start/vm_end address. Signed-off-by: Andrea Arcangeli Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/huge_mm.h | 19 ++++++++++++ mm/huge_memory.c | 80 +++++++++++++++++++++++++++++++++++++++++++++++-- mm/mmap.c | 2 ++ 3 files changed, 99 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h index c590b08c6fa6..827595228734 100644 --- a/include/linux/huge_mm.h +++ b/include/linux/huge_mm.h @@ -104,6 +104,19 @@ extern void __split_huge_page_pmd(struct mm_struct *mm, pmd_t *pmd); #error "hugepages can't be allocated by the buddy allocator" #endif extern int hugepage_madvise(unsigned long *vm_flags); +extern void __vma_adjust_trans_huge(struct vm_area_struct *vma, + unsigned long start, + unsigned long end, + long adjust_next); +static inline void vma_adjust_trans_huge(struct vm_area_struct *vma, + unsigned long start, + unsigned long end, + long adjust_next) +{ + if (!vma->anon_vma || vma->vm_ops || vma->vm_file) + return; + __vma_adjust_trans_huge(vma, start, end, adjust_next); +} #else /* CONFIG_TRANSPARENT_HUGEPAGE */ #define HPAGE_PMD_SHIFT ({ BUG(); 0; }) #define HPAGE_PMD_MASK ({ BUG(); 0; }) @@ -125,6 +138,12 @@ static inline int hugepage_madvise(unsigned long *vm_flags) BUG(); return 0; } +static inline void vma_adjust_trans_huge(struct vm_area_struct *vma, + unsigned long start, + unsigned long end, + long adjust_next) +{ +} #endif /* CONFIG_TRANSPARENT_HUGEPAGE */ #endif /* _LINUX_HUGE_MM_H */ diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 30c3cec82023..b6facc35e893 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1075,8 +1075,16 @@ pmd_t *page_check_address_pmd(struct page *page, goto out; if (pmd_page(*pmd) != page) goto out; - VM_BUG_ON(flag == PAGE_CHECK_ADDRESS_PMD_NOTSPLITTING_FLAG && - pmd_trans_splitting(*pmd)); + /* + * split_vma() may create temporary aliased mappings. There is + * no risk as long as all huge pmd are found and have their + * splitting bit set before __split_huge_page_refcount + * runs. Finding the same huge pmd more than once during the + * same rmap walk is not a problem. + */ + if (flag == PAGE_CHECK_ADDRESS_PMD_NOTSPLITTING_FLAG && + pmd_trans_splitting(*pmd)) + goto out; if (pmd_trans_huge(*pmd)) { VM_BUG_ON(flag == PAGE_CHECK_ADDRESS_PMD_SPLITTING_FLAG && !pmd_trans_splitting(*pmd)); @@ -2196,3 +2204,71 @@ void __split_huge_page_pmd(struct mm_struct *mm, pmd_t *pmd) put_page(page); BUG_ON(pmd_trans_huge(*pmd)); } + +static void split_huge_page_address(struct mm_struct *mm, + unsigned long address) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + + VM_BUG_ON(!(address & ~HPAGE_PMD_MASK)); + + pgd = pgd_offset(mm, address); + if (!pgd_present(*pgd)) + return; + + pud = pud_offset(pgd, address); + if (!pud_present(*pud)) + return; + + pmd = pmd_offset(pud, address); + if (!pmd_present(*pmd)) + return; + /* + * Caller holds the mmap_sem write mode, so a huge pmd cannot + * materialize from under us. + */ + split_huge_page_pmd(mm, pmd); +} + +void __vma_adjust_trans_huge(struct vm_area_struct *vma, + unsigned long start, + unsigned long end, + long adjust_next) +{ + /* + * If the new start address isn't hpage aligned and it could + * previously contain an hugepage: check if we need to split + * an huge pmd. + */ + if (start & ~HPAGE_PMD_MASK && + (start & HPAGE_PMD_MASK) >= vma->vm_start && + (start & HPAGE_PMD_MASK) + HPAGE_PMD_SIZE <= vma->vm_end) + split_huge_page_address(vma->vm_mm, start); + + /* + * If the new end address isn't hpage aligned and it could + * previously contain an hugepage: check if we need to split + * an huge pmd. + */ + if (end & ~HPAGE_PMD_MASK && + (end & HPAGE_PMD_MASK) >= vma->vm_start && + (end & HPAGE_PMD_MASK) + HPAGE_PMD_SIZE <= vma->vm_end) + split_huge_page_address(vma->vm_mm, end); + + /* + * If we're also updating the vma->vm_next->vm_start, if the new + * vm_next->vm_start isn't page aligned and it could previously + * contain an hugepage: check if we need to split an huge pmd. + */ + if (adjust_next > 0) { + struct vm_area_struct *next = vma->vm_next; + unsigned long nstart = next->vm_start; + nstart += adjust_next << PAGE_SHIFT; + if (nstart & ~HPAGE_PMD_MASK && + (nstart & HPAGE_PMD_MASK) >= next->vm_start && + (nstart & HPAGE_PMD_MASK) + HPAGE_PMD_SIZE <= next->vm_end) + split_huge_page_address(next->vm_mm, nstart); + } +} diff --git a/mm/mmap.c b/mm/mmap.c index 753f44d17047..73cc648873d6 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -589,6 +589,8 @@ again: remove_next = 1 + (end > next->vm_end); } } + vma_adjust_trans_huge(vma, start, end, adjust_next); + /* * When changing only vma->vm_end, we don't really need anon_vma * lock. This is a fairly rare case by itself, but the anon_vma -- cgit v1.2.3 From 8ee53820edfd1f3b6554c593f337148dd3d7fc91 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:47:10 -0800 Subject: thp: mmu_notifier_test_young For GRU and EPT, we need gup-fast to set referenced bit too (this is why it's correct to return 0 when shadow_access_mask is zero, it requires gup-fast to set the referenced bit). qemu-kvm access already sets the young bit in the pte if it isn't zero-copy, if it's zero copy or a shadow paging EPT minor fault we relay on gup-fast to signal the page is in use... We also need to check the young bits on the secondary pagetables for NPT and not nested shadow mmu as the data may never get accessed again by the primary pte. Without this closer accuracy, we'd have to remove the heuristic that avoids collapsing hugepages in hugepage virtual regions that have not even a single subpage in use. ->test_young is full backwards compatible with GRU and other usages that don't have young bits in pagetables set by the hardware and that should nuke the secondary mmu mappings when ->clear_flush_young runs just like EPT does. Removing the heuristic that checks the young bit in khugepaged/collapse_huge_page completely isn't so bad either probably but I thought it was worth it and this makes it reliable. Signed-off-by: Andrea Arcangeli Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/mmu.c | 34 ++++++++++++++++++++++++++++++++++ arch/x86/mm/gup.c | 3 +++ include/linux/mmu_notifier.h | 26 ++++++++++++++++++++++++++ mm/huge_memory.c | 6 ++++-- mm/mmu_notifier.c | 20 ++++++++++++++++++++ virt/kvm/kvm_main.c | 17 +++++++++++++++++ 7 files changed, 105 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index aa75f21a9fba..ffd7f8d29187 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -822,6 +822,7 @@ extern bool kvm_rebooting; #define KVM_ARCH_WANT_MMU_NOTIFIER int kvm_unmap_hva(struct kvm *kvm, unsigned long hva); int kvm_age_hva(struct kvm *kvm, unsigned long hva); +int kvm_test_age_hva(struct kvm *kvm, unsigned long hva); void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte); int cpuid_maxphyaddr(struct kvm_vcpu *vcpu); int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu); diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 47b2c3288b6b..f02b8edc3d44 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -945,6 +945,35 @@ static int kvm_age_rmapp(struct kvm *kvm, unsigned long *rmapp, return young; } +static int kvm_test_age_rmapp(struct kvm *kvm, unsigned long *rmapp, + unsigned long data) +{ + u64 *spte; + int young = 0; + + /* + * If there's no access bit in the secondary pte set by the + * hardware it's up to gup-fast/gup to set the access bit in + * the primary pte or in the page structure. + */ + if (!shadow_accessed_mask) + goto out; + + spte = rmap_next(kvm, rmapp, NULL); + while (spte) { + u64 _spte = *spte; + BUG_ON(!(_spte & PT_PRESENT_MASK)); + young = _spte & PT_ACCESSED_MASK; + if (young) { + young = 1; + break; + } + spte = rmap_next(kvm, rmapp, spte); + } +out: + return young; +} + #define RMAP_RECYCLE_THRESHOLD 1000 static void rmap_recycle(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn) @@ -965,6 +994,11 @@ int kvm_age_hva(struct kvm *kvm, unsigned long hva) return kvm_handle_hva(kvm, hva, 0, kvm_age_rmapp); } +int kvm_test_age_hva(struct kvm *kvm, unsigned long hva) +{ + return kvm_handle_hva(kvm, hva, 0, kvm_test_age_rmapp); +} + #ifdef MMU_DEBUG static int is_empty_shadow_page(u64 *spt) { diff --git a/arch/x86/mm/gup.c b/arch/x86/mm/gup.c index 269aa53932e0..dbe34b931374 100644 --- a/arch/x86/mm/gup.c +++ b/arch/x86/mm/gup.c @@ -8,6 +8,7 @@ #include #include #include +#include #include @@ -89,6 +90,7 @@ static noinline int gup_pte_range(pmd_t pmd, unsigned long addr, VM_BUG_ON(!pfn_valid(pte_pfn(pte))); page = pte_page(pte); get_page(page); + SetPageReferenced(page); pages[*nr] = page; (*nr)++; @@ -103,6 +105,7 @@ static inline void get_head_page_multiple(struct page *page, int nr) VM_BUG_ON(page != compound_head(page)); VM_BUG_ON(page_count(page) == 0); atomic_add(nr, &page->_count); + SetPageReferenced(page); } static inline void get_huge_page_tail(struct page *page) diff --git a/include/linux/mmu_notifier.h b/include/linux/mmu_notifier.h index cbfab1e9957d..cc2e7dfea9d7 100644 --- a/include/linux/mmu_notifier.h +++ b/include/linux/mmu_notifier.h @@ -61,6 +61,16 @@ struct mmu_notifier_ops { struct mm_struct *mm, unsigned long address); + /* + * test_young is called to check the young/accessed bitflag in + * the secondary pte. This is used to know if the page is + * frequently used without actually clearing the flag or tearing + * down the secondary mapping on the page. + */ + int (*test_young)(struct mmu_notifier *mn, + struct mm_struct *mm, + unsigned long address); + /* * change_pte is called in cases that pte mapping to page is changed: * for example, when ksm remaps pte to point to a new shared page. @@ -163,6 +173,8 @@ extern void __mmu_notifier_mm_destroy(struct mm_struct *mm); extern void __mmu_notifier_release(struct mm_struct *mm); extern int __mmu_notifier_clear_flush_young(struct mm_struct *mm, unsigned long address); +extern int __mmu_notifier_test_young(struct mm_struct *mm, + unsigned long address); extern void __mmu_notifier_change_pte(struct mm_struct *mm, unsigned long address, pte_t pte); extern void __mmu_notifier_invalidate_page(struct mm_struct *mm, @@ -186,6 +198,14 @@ static inline int mmu_notifier_clear_flush_young(struct mm_struct *mm, return 0; } +static inline int mmu_notifier_test_young(struct mm_struct *mm, + unsigned long address) +{ + if (mm_has_notifiers(mm)) + return __mmu_notifier_test_young(mm, address); + return 0; +} + static inline void mmu_notifier_change_pte(struct mm_struct *mm, unsigned long address, pte_t pte) { @@ -313,6 +333,12 @@ static inline int mmu_notifier_clear_flush_young(struct mm_struct *mm, return 0; } +static inline int mmu_notifier_test_young(struct mm_struct *mm, + unsigned long address) +{ + return 0; +} + static inline void mmu_notifier_change_pte(struct mm_struct *mm, unsigned long address, pte_t pte) { diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 915809b16edf..39d7df40c067 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1632,7 +1632,8 @@ static int __collapse_huge_page_isolate(struct vm_area_struct *vma, VM_BUG_ON(PageLRU(page)); /* If there is no mapped pte young don't collapse the page */ - if (pte_young(pteval)) + if (pte_young(pteval) || PageReferenced(page) || + mmu_notifier_test_young(vma->vm_mm, address)) referenced = 1; } if (unlikely(!referenced)) @@ -1892,7 +1893,8 @@ static int khugepaged_scan_pmd(struct mm_struct *mm, /* cannot use mapcount: can't collapse if there's a gup pin */ if (page_count(page) != 1) goto out_unmap; - if (pte_young(pteval)) + if (pte_young(pteval) || PageReferenced(page) || + mmu_notifier_test_young(vma->vm_mm, address)) referenced = 1; } if (referenced) diff --git a/mm/mmu_notifier.c b/mm/mmu_notifier.c index 438951d366f2..8d032de4088e 100644 --- a/mm/mmu_notifier.c +++ b/mm/mmu_notifier.c @@ -100,6 +100,26 @@ int __mmu_notifier_clear_flush_young(struct mm_struct *mm, return young; } +int __mmu_notifier_test_young(struct mm_struct *mm, + unsigned long address) +{ + struct mmu_notifier *mn; + struct hlist_node *n; + int young = 0; + + rcu_read_lock(); + hlist_for_each_entry_rcu(mn, n, &mm->mmu_notifier_mm->list, hlist) { + if (mn->ops->test_young) { + young = mn->ops->test_young(mn, mm, address); + if (young) + break; + } + } + rcu_read_unlock(); + + return young; +} + void __mmu_notifier_change_pte(struct mm_struct *mm, unsigned long address, pte_t pte) { diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 85ab7db0d366..4286d4766510 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -380,6 +380,22 @@ static int kvm_mmu_notifier_clear_flush_young(struct mmu_notifier *mn, return young; } +static int kvm_mmu_notifier_test_young(struct mmu_notifier *mn, + struct mm_struct *mm, + unsigned long address) +{ + struct kvm *kvm = mmu_notifier_to_kvm(mn); + int young, idx; + + idx = srcu_read_lock(&kvm->srcu); + spin_lock(&kvm->mmu_lock); + young = kvm_test_age_hva(kvm, address); + spin_unlock(&kvm->mmu_lock); + srcu_read_unlock(&kvm->srcu, idx); + + return young; +} + static void kvm_mmu_notifier_release(struct mmu_notifier *mn, struct mm_struct *mm) { @@ -396,6 +412,7 @@ static const struct mmu_notifier_ops kvm_mmu_notifier_ops = { .invalidate_range_start = kvm_mmu_notifier_invalidate_range_start, .invalidate_range_end = kvm_mmu_notifier_invalidate_range_end, .clear_flush_young = kvm_mmu_notifier_clear_flush_young, + .test_young = kvm_mmu_notifier_test_young, .change_pte = kvm_mmu_notifier_change_pte, .release = kvm_mmu_notifier_release, }; -- cgit v1.2.3 From 5a03b051ed87e72b959f32a86054e1142ac4cf55 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:47:11 -0800 Subject: thp: use compaction in kswapd for GFP_ATOMIC order > 0 This takes advantage of memory compaction to properly generate pages of order > 0 if regular page reclaim fails and priority level becomes more severe and we don't reach the proper watermarks. Signed-off-by: Andrea Arcangeli Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/compaction.h | 11 ++++++++--- mm/compaction.c | 31 ++++++++++++++++++++++++++----- mm/vmscan.c | 30 ++++++++++++++++++++---------- 3 files changed, 54 insertions(+), 18 deletions(-) (limited to 'include/linux') diff --git a/include/linux/compaction.h b/include/linux/compaction.h index 72cba4034785..dfa2ed4c0d26 100644 --- a/include/linux/compaction.h +++ b/include/linux/compaction.h @@ -11,6 +11,9 @@ /* The full zone was compacted */ #define COMPACT_COMPLETE 3 +#define COMPACT_MODE_DIRECT_RECLAIM 0 +#define COMPACT_MODE_KSWAPD 1 + #ifdef CONFIG_COMPACTION extern int sysctl_compact_memory; extern int sysctl_compaction_handler(struct ctl_table *table, int write, @@ -25,7 +28,8 @@ extern unsigned long try_to_compact_pages(struct zonelist *zonelist, bool sync); extern unsigned long compaction_suitable(struct zone *zone, int order); extern unsigned long compact_zone_order(struct zone *zone, int order, - gfp_t gfp_mask, bool sync); + gfp_t gfp_mask, bool sync, + int compact_mode); /* Do not skip compaction more than 64 times */ #define COMPACT_MAX_DEFER_SHIFT 6 @@ -70,9 +74,10 @@ static inline unsigned long compaction_suitable(struct zone *zone, int order) } static inline unsigned long compact_zone_order(struct zone *zone, int order, - gfp_t gfp_mask, bool sync) + gfp_t gfp_mask, bool sync, + int compact_mode) { - return 0; + return COMPACT_CONTINUE; } static inline void defer_compaction(struct zone *zone) diff --git a/mm/compaction.c b/mm/compaction.c index da25b5a2e476..60d52a7298d5 100644 --- a/mm/compaction.c +++ b/mm/compaction.c @@ -42,6 +42,8 @@ struct compact_control { unsigned int order; /* order a direct compactor needs */ int migratetype; /* MOVABLE, RECLAIMABLE etc */ struct zone *zone; + + int compact_mode; }; static unsigned long release_freepages(struct list_head *freelist) @@ -382,10 +384,10 @@ static void update_nr_listpages(struct compact_control *cc) } static int compact_finished(struct zone *zone, - struct compact_control *cc) + struct compact_control *cc) { unsigned int order; - unsigned long watermark = low_wmark_pages(zone) + (1 << cc->order); + unsigned long watermark; if (fatal_signal_pending(current)) return COMPACT_PARTIAL; @@ -395,12 +397,27 @@ static int compact_finished(struct zone *zone, return COMPACT_COMPLETE; /* Compaction run is not finished if the watermark is not met */ + if (cc->compact_mode != COMPACT_MODE_KSWAPD) + watermark = low_wmark_pages(zone); + else + watermark = high_wmark_pages(zone); + watermark += (1 << cc->order); + if (!zone_watermark_ok(zone, cc->order, watermark, 0, 0)) return COMPACT_CONTINUE; if (cc->order == -1) return COMPACT_CONTINUE; + /* + * Generating only one page of the right order is not enough + * for kswapd, we must continue until we're above the high + * watermark as a pool for high order GFP_ATOMIC allocations + * too. + */ + if (cc->compact_mode == COMPACT_MODE_KSWAPD) + return COMPACT_CONTINUE; + /* Direct compactor: Is a suitable page free? */ for (order = cc->order; order < MAX_ORDER; order++) { /* Job done if page is free of the right migratetype */ @@ -514,8 +531,9 @@ static int compact_zone(struct zone *zone, struct compact_control *cc) } unsigned long compact_zone_order(struct zone *zone, - int order, gfp_t gfp_mask, - bool sync) + int order, gfp_t gfp_mask, + bool sync, + int compact_mode) { struct compact_control cc = { .nr_freepages = 0, @@ -524,6 +542,7 @@ unsigned long compact_zone_order(struct zone *zone, .migratetype = allocflags_to_migratetype(gfp_mask), .zone = zone, .sync = sync, + .compact_mode = compact_mode, }; INIT_LIST_HEAD(&cc.freepages); INIT_LIST_HEAD(&cc.migratepages); @@ -569,7 +588,8 @@ unsigned long try_to_compact_pages(struct zonelist *zonelist, nodemask) { int status; - status = compact_zone_order(zone, order, gfp_mask, sync); + status = compact_zone_order(zone, order, gfp_mask, sync, + COMPACT_MODE_DIRECT_RECLAIM); rc = max(status, rc); /* If a normal allocation would succeed, stop compacting */ @@ -600,6 +620,7 @@ static int compact_node(int nid) .nr_freepages = 0, .nr_migratepages = 0, .order = -1, + .compact_mode = COMPACT_MODE_DIRECT_RECLAIM, }; zone = &pgdat->node_zones[zoneid]; diff --git a/mm/vmscan.c b/mm/vmscan.c index cfdef0bcc7ab..f5b762ae23a2 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -41,6 +41,7 @@ #include #include #include +#include #include #include @@ -2382,6 +2383,7 @@ loop_again: * cause too much scanning of the lower zones. */ for (i = 0; i <= end_zone; i++) { + int compaction; struct zone *zone = pgdat->node_zones + i; int nr_slab; @@ -2411,9 +2413,26 @@ loop_again: lru_pages); sc.nr_reclaimed += reclaim_state->reclaimed_slab; total_scanned += sc.nr_scanned; + + compaction = 0; + if (order && + zone_watermark_ok(zone, 0, + high_wmark_pages(zone), + end_zone, 0) && + !zone_watermark_ok(zone, order, + high_wmark_pages(zone), + end_zone, 0)) { + compact_zone_order(zone, + order, + sc.gfp_mask, false, + COMPACT_MODE_KSWAPD); + compaction = 1; + } + if (zone->all_unreclaimable) continue; - if (nr_slab == 0 && !zone_reclaimable(zone)) + if (!compaction && nr_slab == 0 && + !zone_reclaimable(zone)) zone->all_unreclaimable = 1; /* * If we've done a decent amount of scanning and @@ -2424,15 +2443,6 @@ loop_again: total_scanned > sc.nr_reclaimed + sc.nr_reclaimed / 2) sc.may_writepage = 1; - /* - * Compact the zone for higher orders to reduce - * latencies for higher-order allocations that - * would ordinarily call try_to_compact_pages() - */ - if (sc.order > PAGE_ALLOC_COSTLY_ORDER) - compact_zone_order(zone, sc.order, sc.gfp_mask, - false); - if (!zone_watermark_ok_safe(zone, order, high_wmark_pages(zone), end_zone, 0)) { all_zones_ok = 0; -- cgit v1.2.3 From 2c888cfbc1b45508a44763d85ba2e8ac43faff5f Mon Sep 17 00:00:00 2001 From: Rik van Riel Date: Thu, 13 Jan 2011 15:47:13 -0800 Subject: thp: fix anon memory statistics with transparent hugepages Count each transparent hugepage as HPAGE_PMD_NR pages in the LRU statistics, so the Active(anon) and Inactive(anon) statistics in /proc/meminfo are correct. Signed-off-by: Rik van Riel Signed-off-by: Andrea Arcangeli Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/huge_mm.h | 8 ++++++++ include/linux/mm_inline.h | 8 +++++--- mm/huge_memory.c | 10 ++++++++++ mm/memcontrol.c | 2 +- mm/vmscan.c | 11 ++++++----- 5 files changed, 30 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h index 827595228734..9b48c24df260 100644 --- a/include/linux/huge_mm.h +++ b/include/linux/huge_mm.h @@ -117,11 +117,19 @@ static inline void vma_adjust_trans_huge(struct vm_area_struct *vma, return; __vma_adjust_trans_huge(vma, start, end, adjust_next); } +static inline int hpage_nr_pages(struct page *page) +{ + if (unlikely(PageTransHuge(page))) + return HPAGE_PMD_NR; + return 1; +} #else /* CONFIG_TRANSPARENT_HUGEPAGE */ #define HPAGE_PMD_SHIFT ({ BUG(); 0; }) #define HPAGE_PMD_MASK ({ BUG(); 0; }) #define HPAGE_PMD_SIZE ({ BUG(); 0; }) +#define hpage_nr_pages(x) 1 + #define transparent_hugepage_enabled(__vma) 0 #define transparent_hugepage_flags 0UL diff --git a/include/linux/mm_inline.h b/include/linux/mm_inline.h index 650f31eabdb1..8f7d24712dc1 100644 --- a/include/linux/mm_inline.h +++ b/include/linux/mm_inline.h @@ -1,6 +1,8 @@ #ifndef LINUX_MM_INLINE_H #define LINUX_MM_INLINE_H +#include + /** * page_is_file_cache - should the page be on a file LRU or anon LRU? * @page: the page to test @@ -24,7 +26,7 @@ __add_page_to_lru_list(struct zone *zone, struct page *page, enum lru_list l, struct list_head *head) { list_add(&page->lru, head); - __inc_zone_state(zone, NR_LRU_BASE + l); + __mod_zone_page_state(zone, NR_LRU_BASE + l, hpage_nr_pages(page)); mem_cgroup_add_lru_list(page, l); } @@ -38,7 +40,7 @@ static inline void del_page_from_lru_list(struct zone *zone, struct page *page, enum lru_list l) { list_del(&page->lru); - __dec_zone_state(zone, NR_LRU_BASE + l); + __mod_zone_page_state(zone, NR_LRU_BASE + l, -hpage_nr_pages(page)); mem_cgroup_del_lru_list(page, l); } @@ -73,7 +75,7 @@ del_page_from_lru(struct zone *zone, struct page *page) l += LRU_ACTIVE; } } - __dec_zone_state(zone, NR_LRU_BASE + l); + __mod_zone_page_state(zone, NR_LRU_BASE + l, -hpage_nr_pages(page)); mem_cgroup_del_lru_list(page, l); } diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 892d8a17a7e5..f4f6041176a4 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1143,6 +1143,7 @@ static void __split_huge_page_refcount(struct page *page) int i; unsigned long head_index = page->index; struct zone *zone = page_zone(page); + int zonestat; /* prevent PageLRU to go away from under us, and freeze lru stats */ spin_lock_irq(&zone->lru_lock); @@ -1207,6 +1208,15 @@ static void __split_huge_page_refcount(struct page *page) __dec_zone_page_state(page, NR_ANON_TRANSPARENT_HUGEPAGES); __mod_zone_page_state(zone, NR_ANON_PAGES, HPAGE_PMD_NR); + /* + * A hugepage counts for HPAGE_PMD_NR pages on the LRU statistics, + * so adjust those appropriately if this page is on the LRU. + */ + if (PageLRU(page)) { + zonestat = NR_LRU_BASE + page_lru(page); + __mod_zone_page_state(zone, zonestat, -(HPAGE_PMD_NR-1)); + } + ClearPageCompound(page); compound_unlock(page); spin_unlock_irq(&zone->lru_lock); diff --git a/mm/memcontrol.c b/mm/memcontrol.c index a1bb59d4c9d0..f4ea3410fb4d 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1091,7 +1091,7 @@ unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan, case 0: list_move(&page->lru, dst); mem_cgroup_del_lru(page); - nr_taken++; + nr_taken += hpage_nr_pages(page); break; case -EBUSY: /* we don't affect global LRU but rotate in our LRU */ diff --git a/mm/vmscan.c b/mm/vmscan.c index f5b762ae23a2..0882014d2ce5 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1045,7 +1045,7 @@ static unsigned long isolate_lru_pages(unsigned long nr_to_scan, case 0: list_move(&page->lru, dst); mem_cgroup_del_lru(page); - nr_taken++; + nr_taken += hpage_nr_pages(page); break; case -EBUSY: @@ -1103,7 +1103,7 @@ static unsigned long isolate_lru_pages(unsigned long nr_to_scan, if (__isolate_lru_page(cursor_page, mode, file) == 0) { list_move(&cursor_page->lru, dst); mem_cgroup_del_lru(cursor_page); - nr_taken++; + nr_taken += hpage_nr_pages(page); nr_lumpy_taken++; if (PageDirty(cursor_page)) nr_lumpy_dirty++; @@ -1158,14 +1158,15 @@ static unsigned long clear_active_flags(struct list_head *page_list, struct page *page; list_for_each_entry(page, page_list, lru) { + int numpages = hpage_nr_pages(page); lru = page_lru_base_type(page); if (PageActive(page)) { lru += LRU_ACTIVE; ClearPageActive(page); - nr_active++; + nr_active += numpages; } if (count) - count[lru]++; + count[lru] += numpages; } return nr_active; @@ -1483,7 +1484,7 @@ static void move_active_pages_to_lru(struct zone *zone, list_move(&page->lru, &zone->lru[lru].list); mem_cgroup_add_lru_list(page, lru); - pgmoved++; + pgmoved += hpage_nr_pages(page); if (!pagevec_add(&pvec, page) || list_empty(list)) { spin_unlock_irq(&zone->lru_lock); -- cgit v1.2.3 From 37c2ac7872a9387542616f658d20ac25f5bdb32e Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:47:16 -0800 Subject: thp: compound_trans_order Read compound_trans_order safe. Noop for CONFIG_TRANSPARENT_HUGEPAGE=n. Signed-off-by: Andrea Arcangeli Cc: Daisuke Nishimura Cc: KAMEZAWA Hiroyuki Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 14 ++++++++++++++ mm/memcontrol.c | 12 ++++++------ mm/memory-failure.c | 12 ++++++------ 3 files changed, 26 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mm.h b/include/linux/mm.h index 9c2695beab86..ce97a2bb0b19 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -450,6 +450,20 @@ static inline int compound_order(struct page *page) return (unsigned long)page[1].lru.prev; } +static inline int compound_trans_order(struct page *page) +{ + int order; + unsigned long flags; + + if (!PageHead(page)) + return 0; + + flags = compound_lock_irqsave(page); + order = compound_order(page); + compound_unlock_irqrestore(page, flags); + return order; +} + static inline void set_compound_order(struct page *page, unsigned long order) { page[1].lru.prev = (void *)order; diff --git a/mm/memcontrol.c b/mm/memcontrol.c index f4ea3410fb4d..741206ffdace 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1027,10 +1027,6 @@ mem_cgroup_get_reclaim_stat_from_page(struct page *page) { struct page_cgroup *pc; struct mem_cgroup_per_zone *mz; - int page_size = PAGE_SIZE; - - if (PageTransHuge(page)) - page_size <<= compound_order(page); if (mem_cgroup_disabled()) return NULL; @@ -2286,8 +2282,10 @@ static int mem_cgroup_charge_common(struct page *page, struct mm_struct *mm, int ret; int page_size = PAGE_SIZE; - if (PageTransHuge(page)) + if (PageTransHuge(page)) { page_size <<= compound_order(page); + VM_BUG_ON(!PageTransHuge(page)); + } pc = lookup_page_cgroup(page); /* can happen at boot */ @@ -2558,8 +2556,10 @@ __mem_cgroup_uncharge_common(struct page *page, enum charge_type ctype) if (PageSwapCache(page)) return NULL; - if (PageTransHuge(page)) + if (PageTransHuge(page)) { page_size <<= compound_order(page); + VM_BUG_ON(!PageTransHuge(page)); + } count = page_size >> PAGE_SHIFT; /* diff --git a/mm/memory-failure.c b/mm/memory-failure.c index 1b43d0ffff65..548fbd70f026 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -203,7 +203,7 @@ static int kill_proc_ao(struct task_struct *t, unsigned long addr, int trapno, #ifdef __ARCH_SI_TRAPNO si.si_trapno = trapno; #endif - si.si_addr_lsb = compound_order(compound_head(page)) + PAGE_SHIFT; + si.si_addr_lsb = compound_trans_order(compound_head(page)) + PAGE_SHIFT; /* * Don't use force here, it's convenient if the signal * can be temporarily blocked. @@ -930,7 +930,7 @@ static int hwpoison_user_mappings(struct page *p, unsigned long pfn, static void set_page_hwpoison_huge_page(struct page *hpage) { int i; - int nr_pages = 1 << compound_order(hpage); + int nr_pages = 1 << compound_trans_order(hpage); for (i = 0; i < nr_pages; i++) SetPageHWPoison(hpage + i); } @@ -938,7 +938,7 @@ static void set_page_hwpoison_huge_page(struct page *hpage) static void clear_page_hwpoison_huge_page(struct page *hpage) { int i; - int nr_pages = 1 << compound_order(hpage); + int nr_pages = 1 << compound_trans_order(hpage); for (i = 0; i < nr_pages; i++) ClearPageHWPoison(hpage + i); } @@ -968,7 +968,7 @@ int __memory_failure(unsigned long pfn, int trapno, int flags) return 0; } - nr_pages = 1 << compound_order(hpage); + nr_pages = 1 << compound_trans_order(hpage); atomic_long_add(nr_pages, &mce_bad_pages); /* @@ -1166,7 +1166,7 @@ int unpoison_memory(unsigned long pfn) return 0; } - nr_pages = 1 << compound_order(page); + nr_pages = 1 << compound_trans_order(page); if (!get_page_unless_zero(page)) { /* @@ -1304,7 +1304,7 @@ static int soft_offline_huge_page(struct page *page, int flags) } done: if (!PageHWPoison(hpage)) - atomic_long_add(1 << compound_order(hpage), &mce_bad_pages); + atomic_long_add(1 << compound_trans_order(hpage), &mce_bad_pages); set_page_hwpoison_huge_page(hpage); dequeue_hwpoisoned_huge_page(hpage); /* keep elevated page count for bad page */ -- cgit v1.2.3 From a664b2d8555c659127bf8fe049a58449d394a707 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:47:17 -0800 Subject: thp: madvise(MADV_NOHUGEPAGE) Add madvise MADV_NOHUGEPAGE to mark regions that are not important to be hugepage backed. Return -EINVAL if the vma is not of an anonymous type, or the feature isn't built into the kernel. Never silently return success. Signed-off-by: Andrea Arcangeli Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/huge_mm.h | 14 ++++++++------ include/linux/khugepaged.h | 7 ++++--- include/linux/mm.h | 1 + mm/huge_memory.c | 41 ++++++++++++++++++++++++++++++----------- mm/madvise.c | 4 +++- 5 files changed, 46 insertions(+), 21 deletions(-) (limited to 'include/linux') diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h index 9b48c24df260..a8b7e42d19ec 100644 --- a/include/linux/huge_mm.h +++ b/include/linux/huge_mm.h @@ -52,10 +52,12 @@ extern pmd_t *page_check_address_pmd(struct page *page, #define HPAGE_PMD_SIZE HPAGE_SIZE #define transparent_hugepage_enabled(__vma) \ - (transparent_hugepage_flags & (1<vm_flags & VM_HUGEPAGE)) + ((transparent_hugepage_flags & \ + (1<vm_flags & VM_HUGEPAGE))) && \ + !((__vma)->vm_flags & VM_NOHUGEPAGE)) #define transparent_hugepage_defrag(__vma) \ ((transparent_hugepage_flags & \ (1< MAX_ORDER #error "hugepages can't be allocated by the buddy allocator" #endif -extern int hugepage_madvise(unsigned long *vm_flags); +extern int hugepage_madvise(unsigned long *vm_flags, int advice); extern void __vma_adjust_trans_huge(struct vm_area_struct *vma, unsigned long start, unsigned long end, @@ -141,7 +143,7 @@ static inline int split_huge_page(struct page *page) do { } while (0) #define wait_split_huge_page(__anon_vma, __pmd) \ do { } while (0) -static inline int hugepage_madvise(unsigned long *vm_flags) +static inline int hugepage_madvise(unsigned long *vm_flags, int advice) { BUG(); return 0; diff --git a/include/linux/khugepaged.h b/include/linux/khugepaged.h index 552f3184756c..6b394f0b5148 100644 --- a/include/linux/khugepaged.h +++ b/include/linux/khugepaged.h @@ -38,9 +38,10 @@ static inline void khugepaged_exit(struct mm_struct *mm) static inline int khugepaged_enter(struct vm_area_struct *vma) { if (!test_bit(MMF_VM_HUGEPAGE, &vma->vm_mm->flags)) - if (khugepaged_always() || - (khugepaged_req_madv() && - vma->vm_flags & VM_HUGEPAGE)) + if ((khugepaged_always() || + (khugepaged_req_madv() && + vma->vm_flags & VM_HUGEPAGE)) && + !(vma->vm_flags & VM_NOHUGEPAGE)) if (__khugepaged_enter(vma->vm_mm)) return -ENOMEM; return 0; diff --git a/include/linux/mm.h b/include/linux/mm.h index ce97a2bb0b19..956a35532f47 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -83,6 +83,7 @@ extern unsigned int kobjsize(const void *objp); #define VM_GROWSUP 0x00000200 #else #define VM_GROWSUP 0x00000000 +#define VM_NOHUGEPAGE 0x00000200 /* MADV_NOHUGEPAGE marked this vma */ #endif #define VM_PFNMAP 0x00000400 /* Page-ranges managed without "struct page", just pure PFN */ #define VM_DENYWRITE 0x00000800 /* ETXTBSY on write attempts.. */ diff --git a/mm/huge_memory.c b/mm/huge_memory.c index f4f6041176a4..fce667c0281d 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include "internal.h" @@ -1388,18 +1389,36 @@ out: return ret; } -int hugepage_madvise(unsigned long *vm_flags) +int hugepage_madvise(unsigned long *vm_flags, int advice) { - /* - * Be somewhat over-protective like KSM for now! - */ - if (*vm_flags & (VM_HUGEPAGE | VM_SHARED | VM_MAYSHARE | - VM_PFNMAP | VM_IO | VM_DONTEXPAND | - VM_RESERVED | VM_HUGETLB | VM_INSERTPAGE | - VM_MIXEDMAP | VM_SAO)) - return -EINVAL; - - *vm_flags |= VM_HUGEPAGE; + switch (advice) { + case MADV_HUGEPAGE: + /* + * Be somewhat over-protective like KSM for now! + */ + if (*vm_flags & (VM_HUGEPAGE | + VM_SHARED | VM_MAYSHARE | + VM_PFNMAP | VM_IO | VM_DONTEXPAND | + VM_RESERVED | VM_HUGETLB | VM_INSERTPAGE | + VM_MIXEDMAP | VM_SAO)) + return -EINVAL; + *vm_flags &= ~VM_NOHUGEPAGE; + *vm_flags |= VM_HUGEPAGE; + break; + case MADV_NOHUGEPAGE: + /* + * Be somewhat over-protective like KSM for now! + */ + if (*vm_flags & (VM_NOHUGEPAGE | + VM_SHARED | VM_MAYSHARE | + VM_PFNMAP | VM_IO | VM_DONTEXPAND | + VM_RESERVED | VM_HUGETLB | VM_INSERTPAGE | + VM_MIXEDMAP | VM_SAO)) + return -EINVAL; + *vm_flags &= ~VM_HUGEPAGE; + *vm_flags |= VM_NOHUGEPAGE; + break; + } return 0; } diff --git a/mm/madvise.c b/mm/madvise.c index ecde40a401c1..bbac126e03ed 100644 --- a/mm/madvise.c +++ b/mm/madvise.c @@ -72,7 +72,8 @@ static long madvise_behavior(struct vm_area_struct * vma, goto out; break; case MADV_HUGEPAGE: - error = hugepage_madvise(&new_flags); + case MADV_NOHUGEPAGE: + error = hugepage_madvise(&new_flags, behavior); if (error) goto out; break; @@ -290,6 +291,7 @@ madvise_behavior_valid(int behavior) #endif #ifdef CONFIG_TRANSPARENT_HUGEPAGE case MADV_HUGEPAGE: + case MADV_NOHUGEPAGE: #endif return 1; -- cgit v1.2.3 From 60ab3244ec85c44276c585a2a20d3750402e1cf4 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:47:18 -0800 Subject: thp: khugepaged: make khugepaged aware about madvise MADV_HUGEPAGE and MADV_NOHUGEPAGE were fully effective only if run after mmap and before touching the memory. While this is enough for most usages, it's little effort to make madvise more dynamic at runtime on an existing mapping by making khugepaged aware about madvise. MADV_HUGEPAGE: register in khugepaged immediately without waiting a page fault (that may not ever happen if all pages are already mapped and the "enabled" knob was set to madvise during the initial page faults). MADV_NOHUGEPAGE: skip vmas marked VM_NOHUGEPAGE in khugepaged to stop collapsing pages where not needed. [akpm@linux-foundation.org: tweak comment] Signed-off-by: Andrea Arcangeli Cc: Michael Kerrisk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/huge_mm.h | 6 ++++-- mm/huge_memory.c | 23 +++++++++++++++++++---- mm/madvise.c | 2 +- 3 files changed, 24 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h index a8b7e42d19ec..bddfba1d7b85 100644 --- a/include/linux/huge_mm.h +++ b/include/linux/huge_mm.h @@ -105,7 +105,8 @@ extern void __split_huge_page_pmd(struct mm_struct *mm, pmd_t *pmd); #if HPAGE_PMD_ORDER > MAX_ORDER #error "hugepages can't be allocated by the buddy allocator" #endif -extern int hugepage_madvise(unsigned long *vm_flags, int advice); +extern int hugepage_madvise(struct vm_area_struct *vma, + unsigned long *vm_flags, int advice); extern void __vma_adjust_trans_huge(struct vm_area_struct *vma, unsigned long start, unsigned long end, @@ -143,7 +144,8 @@ static inline int split_huge_page(struct page *page) do { } while (0) #define wait_split_huge_page(__anon_vma, __pmd) \ do { } while (0) -static inline int hugepage_madvise(unsigned long *vm_flags, int advice) +static inline int hugepage_madvise(struct vm_area_struct *vma, + unsigned long *vm_flags, int advice) { BUG(); return 0; diff --git a/mm/huge_memory.c b/mm/huge_memory.c index fce667c0281d..004c9c2aac78 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1389,7 +1389,8 @@ out: return ret; } -int hugepage_madvise(unsigned long *vm_flags, int advice) +int hugepage_madvise(struct vm_area_struct *vma, + unsigned long *vm_flags, int advice) { switch (advice) { case MADV_HUGEPAGE: @@ -1404,6 +1405,13 @@ int hugepage_madvise(unsigned long *vm_flags, int advice) return -EINVAL; *vm_flags &= ~VM_NOHUGEPAGE; *vm_flags |= VM_HUGEPAGE; + /* + * If the vma become good for khugepaged to scan, + * register it here without waiting a page fault that + * may not happen any time soon. + */ + if (unlikely(khugepaged_enter_vma_merge(vma))) + return -ENOMEM; break; case MADV_NOHUGEPAGE: /* @@ -1417,6 +1425,11 @@ int hugepage_madvise(unsigned long *vm_flags, int advice) return -EINVAL; *vm_flags &= ~VM_HUGEPAGE; *vm_flags |= VM_NOHUGEPAGE; + /* + * Setting VM_NOHUGEPAGE will prevent khugepaged from scanning + * this vma even if we leave the mm registered in khugepaged if + * it got registered before VM_NOHUGEPAGE was set. + */ break; } @@ -1784,7 +1797,8 @@ static void collapse_huge_page(struct mm_struct *mm, if (address < hstart || address + HPAGE_PMD_SIZE > hend) goto out; - if (!(vma->vm_flags & VM_HUGEPAGE) && !khugepaged_always()) + if ((!(vma->vm_flags & VM_HUGEPAGE) && !khugepaged_always()) || + (vma->vm_flags & VM_NOHUGEPAGE)) goto out; /* VM_PFNMAP vmas may have vm_ops null but vm_file set */ @@ -2007,8 +2021,9 @@ static unsigned int khugepaged_scan_mm_slot(unsigned int pages, break; } - if (!(vma->vm_flags & VM_HUGEPAGE) && - !khugepaged_always()) { + if ((!(vma->vm_flags & VM_HUGEPAGE) && + !khugepaged_always()) || + (vma->vm_flags & VM_NOHUGEPAGE)) { progress++; continue; } diff --git a/mm/madvise.c b/mm/madvise.c index bbac126e03ed..2221491ed503 100644 --- a/mm/madvise.c +++ b/mm/madvise.c @@ -73,7 +73,7 @@ static long madvise_behavior(struct vm_area_struct * vma, break; case MADV_HUGEPAGE: case MADV_NOHUGEPAGE: - error = hugepage_madvise(&new_flags, behavior); + error = hugepage_madvise(vma, &new_flags, behavior); if (error) goto out; break; -- cgit v1.2.3 From 22e5c47ee238abe636655c3862ed28d6eb084ad4 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 13 Jan 2011 15:47:20 -0800 Subject: thp: add compound_trans_head() helper Cleanup some code with common compound_trans_head helper. Signed-off-by: Andrea Arcangeli Cc: Hugh Dickins Cc: Johannes Weiner Cc: Marcelo Tosatti Cc: Avi Kivity Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/huge_mm.h | 18 ++++++++++++++++++ mm/ksm.c | 15 +++------------ virt/kvm/kvm_main.c | 38 ++++++++++++++------------------------ 3 files changed, 35 insertions(+), 36 deletions(-) (limited to 'include/linux') diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h index bddfba1d7b85..8e6c8c42bc3c 100644 --- a/include/linux/huge_mm.h +++ b/include/linux/huge_mm.h @@ -126,6 +126,23 @@ static inline int hpage_nr_pages(struct page *page) return HPAGE_PMD_NR; return 1; } +static inline struct page *compound_trans_head(struct page *page) +{ + if (PageTail(page)) { + struct page *head; + head = page->first_page; + smp_rmb(); + /* + * head may be a dangling pointer. + * __split_huge_page_refcount clears PageTail before + * overwriting first_page, so if PageTail is still + * there it means the head pointer isn't dangling. + */ + if (PageTail(page)) + return head; + } + return page; +} #else /* CONFIG_TRANSPARENT_HUGEPAGE */ #define HPAGE_PMD_SHIFT ({ BUG(); 0; }) #define HPAGE_PMD_MASK ({ BUG(); 0; }) @@ -144,6 +161,7 @@ static inline int split_huge_page(struct page *page) do { } while (0) #define wait_split_huge_page(__anon_vma, __pmd) \ do { } while (0) +#define compound_trans_head(page) compound_head(page) static inline int hugepage_madvise(struct vm_area_struct *vma, unsigned long *vm_flags, int advice) { diff --git a/mm/ksm.c b/mm/ksm.c index 4d5a681923bb..33781de0b6bf 100644 --- a/mm/ksm.c +++ b/mm/ksm.c @@ -415,20 +415,11 @@ out: static struct page *page_trans_compound_anon(struct page *page) { if (PageTransCompound(page)) { - struct page *head; - head = compound_head(page); + struct page *head = compound_trans_head(page); /* - * head may be a dangling pointer. - * __split_huge_page_refcount clears PageTail - * before overwriting first_page, so if - * PageTail is still there it means the head - * pointer isn't dangling. + * head may actually be splitted and freed from under + * us but it's ok here. */ - if (head != page) { - smp_rmb(); - if (!PageTransCompound(page)) - return NULL; - } if (PageAnon(head)) return head; } diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 4286d4766510..f29abeb6a912 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -104,34 +104,24 @@ static pfn_t fault_pfn; inline int kvm_is_mmio_pfn(pfn_t pfn) { if (pfn_valid(pfn)) { - struct page *head; + int reserved; struct page *tail = pfn_to_page(pfn); - head = compound_head(tail); + struct page *head = compound_trans_head(tail); + reserved = PageReserved(head); if (head != tail) { - smp_rmb(); /* - * head may be a dangling pointer. - * __split_huge_page_refcount clears PageTail - * before overwriting first_page, so if - * PageTail is still there it means the head - * pointer isn't dangling. + * "head" is not a dangling pointer + * (compound_trans_head takes care of that) + * but the hugepage may have been splitted + * from under us (and we may not hold a + * reference count on the head page so it can + * be reused before we run PageReferenced), so + * we've to check PageTail before returning + * what we just read. */ - if (PageTail(tail)) { - /* - * the "head" is not a dangling - * pointer but the hugepage may have - * been splitted from under us (and we - * may not hold a reference count on - * the head page so it can be reused - * before we run PageReferenced), so - * we've to recheck PageTail before - * returning what we just read. - */ - int reserved = PageReserved(head); - smp_rmb(); - if (PageTail(tail)) - return reserved; - } + smp_rmb(); + if (PageTail(tail)) + return reserved; } return PageReserved(tail); } -- cgit v1.2.3 From 29c1f677d424e8c5683a837fc4f03fc9f19201d7 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Thu, 13 Jan 2011 15:47:21 -0800 Subject: mm: migration: use rcu_dereference_protected when dereferencing the radix tree slot during file page migration migrate_pages() -> unmap_and_move() only calls rcu_read_lock() for anonymous pages, as introduced by git commit 989f89c57e6361e7d16fbd9572b5da7d313b073d ("fix rcu_read_lock() in page migraton"). The point of the RCU protection there is part of getting a stable reference to anon_vma and is only held for anon pages as file pages are locked which is sufficient protection against freeing. However, while a file page's mapping is being migrated, the radix tree is double checked to ensure it is the expected page. This uses radix_tree_deref_slot() -> rcu_dereference() without the RCU lock held triggering the following warning. [ 173.674290] =================================================== [ 173.676016] [ INFO: suspicious rcu_dereference_check() usage. ] [ 173.676016] --------------------------------------------------- [ 173.676016] include/linux/radix-tree.h:145 invoked rcu_dereference_check() without protection! [ 173.676016] [ 173.676016] other info that might help us debug this: [ 173.676016] [ 173.676016] [ 173.676016] rcu_scheduler_active = 1, debug_locks = 0 [ 173.676016] 1 lock held by hugeadm/2899: [ 173.676016] #0: (&(&inode->i_data.tree_lock)->rlock){..-.-.}, at: [] migrate_page_move_mapping+0x40/0x1ab [ 173.676016] [ 173.676016] stack backtrace: [ 173.676016] Pid: 2899, comm: hugeadm Not tainted 2.6.37-rc5-autobuild [ 173.676016] Call Trace: [ 173.676016] [] ? printk+0x14/0x1b [ 173.676016] [] lockdep_rcu_dereference+0x7d/0x86 [ 173.676016] [] migrate_page_move_mapping+0xca/0x1ab [ 173.676016] [] migrate_page+0x23/0x39 [ 173.676016] [] buffer_migrate_page+0x22/0x107 [ 173.676016] [] ? buffer_migrate_page+0x0/0x107 [ 173.676016] [] move_to_new_page+0x9a/0x1ae [ 173.676016] [] migrate_pages+0x1e7/0x2fa This patch introduces radix_tree_deref_slot_protected() which calls rcu_dereference_protected(). Users of it must pass in the mapping->tree_lock that is protecting this dereference. Holding the tree lock protects against parallel updaters of the radix tree meaning that rcu_dereference_protected is allowable. [akpm@linux-foundation.org: remove unneeded casts] Signed-off-by: Mel Gorman Cc: Minchan Kim Cc: KAMEZAWA Hiroyuki Cc: Milton Miller Cc: Nick Piggin Cc: Wu Fengguang Cc: [2.6.37.early] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/radix-tree.h | 16 ++++++++++++++++ mm/migrate.c | 4 ++-- 2 files changed, 18 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/radix-tree.h b/include/linux/radix-tree.h index ab2baa5c4884..23241c2fecce 100644 --- a/include/linux/radix-tree.h +++ b/include/linux/radix-tree.h @@ -145,6 +145,22 @@ static inline void *radix_tree_deref_slot(void **pslot) return rcu_dereference(*pslot); } +/** + * radix_tree_deref_slot_protected - dereference a slot without RCU lock but with tree lock held + * @pslot: pointer to slot, returned by radix_tree_lookup_slot + * Returns: item that was stored in that slot with any direct pointer flag + * removed. + * + * Similar to radix_tree_deref_slot but only used during migration when a pages + * mapping is being moved. The caller does not hold the RCU read lock but it + * must hold the tree lock to prevent parallel updates. + */ +static inline void *radix_tree_deref_slot_protected(void **pslot, + spinlock_t *treelock) +{ + return rcu_dereference_protected(*pslot, lockdep_is_held(treelock)); +} + /** * radix_tree_deref_retry - check radix_tree_deref_slot * @arg: pointer returned by radix_tree_deref_slot diff --git a/mm/migrate.c b/mm/migrate.c index 1a531b760b3b..89a6bc8cd307 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -248,7 +248,7 @@ static int migrate_page_move_mapping(struct address_space *mapping, expected_count = 2 + page_has_private(page); if (page_count(page) != expected_count || - (struct page *)radix_tree_deref_slot(pslot) != page) { + radix_tree_deref_slot_protected(pslot, &mapping->tree_lock) != page) { spin_unlock_irq(&mapping->tree_lock); return -EAGAIN; } @@ -320,7 +320,7 @@ int migrate_huge_page_move_mapping(struct address_space *mapping, expected_count = 2 + page_has_private(page); if (page_count(page) != expected_count || - (struct page *)radix_tree_deref_slot(pslot) != page) { + radix_tree_deref_slot_protected(pslot, &mapping->tree_lock) != page) { spin_unlock_irq(&mapping->tree_lock); return -EAGAIN; } -- cgit v1.2.3 From db16d5ec1f87f17511599bc77857dd1662b5a22f Mon Sep 17 00:00:00 2001 From: Greg Thelen Date: Thu, 13 Jan 2011 15:47:35 -0800 Subject: memcg: add page_cgroup flags for dirty page tracking This patchset provides the ability for each cgroup to have independent dirty page limits. Limiting dirty memory is like fixing the max amount of dirty (hard to reclaim) page cache used by a cgroup. So, in case of multiple cgroup writers, they will not be able to consume more than their designated share of dirty pages and will be forced to perform write-out if they cross that limit. The patches are based on a series proposed by Andrea Righi in Mar 2010. Overview: - Add page_cgroup flags to record when pages are dirty, in writeback, or nfs unstable. - Extend mem_cgroup to record the total number of pages in each of the interesting dirty states (dirty, writeback, unstable_nfs). - Add dirty parameters similar to the system-wide /proc/sys/vm/dirty_* limits to mem_cgroup. The mem_cgroup dirty parameters are accessible via cgroupfs control files. - Consider both system and per-memcg dirty limits in page writeback when deciding to queue background writeback or block for foreground writeback. Known shortcomings: - When a cgroup dirty limit is exceeded, then bdi writeback is employed to writeback dirty inodes. Bdi writeback considers inodes from any cgroup, not just inodes contributing dirty pages to the cgroup exceeding its limit. - When memory.use_hierarchy is set, then dirty limits are disabled. This is a implementation detail. An enhanced implementation is needed to check the chain of parents to ensure that no dirty limit is exceeded. Performance data: - A page fault microbenchmark workload was used to measure performance, which can be called in read or write mode: f = open(foo. $cpu) truncate(f, 4096) alarm(60) while (1) { p = mmap(f, 4096) if (write) *p = 1 else x = *p munmap(p) } - The workload was called for several points in the patch series in different modes: - s_read is a single threaded reader - s_write is a single threaded writer - p_read is a 16 thread reader, each operating on a different file - p_write is a 16 thread writer, each operating on a different file - Measurements were collected on a 16 core non-numa system using "perf stat --repeat 3". The -a option was used for parallel (p_*) runs. - All numbers are page fault rate (M/sec). Higher is better. - To compare the performance of a kernel without non-memcg compare the first and last rows, neither has memcg configured. The first row does not include any of these memcg patches. - To compare the performance of using memcg dirty limits, compare the baseline (2nd row titled "w/ memcg") with the the code and memcg enabled (2nd to last row titled "all patches"). root_cgroup child_cgroup s_read s_write p_read p_write s_read s_write p_read p_write mmotm w/o memcg 0.428 0.390 0.429 0.388 mmotm w/ memcg 0.411 0.378 0.391 0.362 0.412 0.377 0.385 0.363 all patches 0.384 0.360 0.370 0.348 0.381 0.363 0.368 0.347 all patches 0.431 0.402 0.427 0.395 w/o memcg This patch: Add additional flags to page_cgroup to track dirty pages within a mem_cgroup. Signed-off-by: KAMEZAWA Hiroyuki Signed-off-by: Andrea Righi Signed-off-by: Greg Thelen Acked-by: Daisuke Nishimura Cc: Balbir Singh Cc: Minchan Kim Cc: Wu Fengguang Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/page_cgroup.h | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) (limited to 'include/linux') diff --git a/include/linux/page_cgroup.h b/include/linux/page_cgroup.h index b02195dfc1b0..fdb5a92b5ac7 100644 --- a/include/linux/page_cgroup.h +++ b/include/linux/page_cgroup.h @@ -40,6 +40,9 @@ enum { PCG_USED, /* this object is in use. */ PCG_ACCT_LRU, /* page has been accounted for */ PCG_FILE_MAPPED, /* page is accounted as "mapped" */ + PCG_FILE_DIRTY, /* page is dirty */ + PCG_FILE_WRITEBACK, /* page is under writeback */ + PCG_FILE_UNSTABLE_NFS, /* page is NFS unstable */ PCG_MIGRATION, /* under page migration */ }; @@ -59,6 +62,10 @@ static inline void ClearPageCgroup##uname(struct page_cgroup *pc) \ static inline int TestClearPageCgroup##uname(struct page_cgroup *pc) \ { return test_and_clear_bit(PCG_##lname, &pc->flags); } +#define TESTSETPCGFLAG(uname, lname) \ +static inline int TestSetPageCgroup##uname(struct page_cgroup *pc) \ + { return test_and_set_bit(PCG_##lname, &pc->flags); } + /* Cache flag is set only once (at allocation) */ TESTPCGFLAG(Cache, CACHE) CLEARPCGFLAG(Cache, CACHE) @@ -78,6 +85,22 @@ SETPCGFLAG(FileMapped, FILE_MAPPED) CLEARPCGFLAG(FileMapped, FILE_MAPPED) TESTPCGFLAG(FileMapped, FILE_MAPPED) +SETPCGFLAG(FileDirty, FILE_DIRTY) +CLEARPCGFLAG(FileDirty, FILE_DIRTY) +TESTPCGFLAG(FileDirty, FILE_DIRTY) +TESTCLEARPCGFLAG(FileDirty, FILE_DIRTY) +TESTSETPCGFLAG(FileDirty, FILE_DIRTY) + +SETPCGFLAG(FileWriteback, FILE_WRITEBACK) +CLEARPCGFLAG(FileWriteback, FILE_WRITEBACK) +TESTPCGFLAG(FileWriteback, FILE_WRITEBACK) + +SETPCGFLAG(FileUnstableNFS, FILE_UNSTABLE_NFS) +CLEARPCGFLAG(FileUnstableNFS, FILE_UNSTABLE_NFS) +TESTPCGFLAG(FileUnstableNFS, FILE_UNSTABLE_NFS) +TESTCLEARPCGFLAG(FileUnstableNFS, FILE_UNSTABLE_NFS) +TESTSETPCGFLAG(FileUnstableNFS, FILE_UNSTABLE_NFS) + SETPCGFLAG(Migration, MIGRATION) CLEARPCGFLAG(Migration, MIGRATION) TESTPCGFLAG(Migration, MIGRATION) -- cgit v1.2.3 From 2a7106f2cb0768d00fe8c1eb42a754a7d8518f08 Mon Sep 17 00:00:00 2001 From: Greg Thelen Date: Thu, 13 Jan 2011 15:47:37 -0800 Subject: memcg: create extensible page stat update routines Replace usage of the mem_cgroup_update_file_mapped() memcg statistic update routine with two new routines: * mem_cgroup_inc_page_stat() * mem_cgroup_dec_page_stat() As before, only the file_mapped statistic is managed. However, these more general interfaces allow for new statistics to be more easily added. New statistics are added with memcg dirty page accounting. Signed-off-by: Greg Thelen Signed-off-by: Andrea Righi Acked-by: KAMEZAWA Hiroyuki Acked-by: Daisuke Nishimura Cc: Balbir Singh Cc: Minchan Kim Cc: Wu Fengguang Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memcontrol.h | 31 ++++++++++++++++++++++++++++--- mm/memcontrol.c | 16 +++++++--------- mm/rmap.c | 4 ++-- 3 files changed, 37 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 159a0762aeaf..067115ce6b3e 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -25,6 +25,11 @@ struct page_cgroup; struct page; struct mm_struct; +/* Stats that can be updated by kernel. */ +enum mem_cgroup_page_stat_item { + MEMCG_NR_FILE_MAPPED, /* # of pages charged as file rss */ +}; + extern unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan, struct list_head *dst, unsigned long *scanned, int order, @@ -121,7 +126,22 @@ static inline bool mem_cgroup_disabled(void) return false; } -void mem_cgroup_update_file_mapped(struct page *page, int val); +void mem_cgroup_update_page_stat(struct page *page, + enum mem_cgroup_page_stat_item idx, + int val); + +static inline void mem_cgroup_inc_page_stat(struct page *page, + enum mem_cgroup_page_stat_item idx) +{ + mem_cgroup_update_page_stat(page, idx, 1); +} + +static inline void mem_cgroup_dec_page_stat(struct page *page, + enum mem_cgroup_page_stat_item idx) +{ + mem_cgroup_update_page_stat(page, idx, -1); +} + unsigned long mem_cgroup_soft_limit_reclaim(struct zone *zone, int order, gfp_t gfp_mask); u64 mem_cgroup_get_limit(struct mem_cgroup *mem); @@ -293,8 +313,13 @@ mem_cgroup_print_oom_info(struct mem_cgroup *memcg, struct task_struct *p) { } -static inline void mem_cgroup_update_file_mapped(struct page *page, - int val) +static inline void mem_cgroup_inc_page_stat(struct page *page, + enum mem_cgroup_page_stat_item idx) +{ +} + +static inline void mem_cgroup_dec_page_stat(struct page *page, + enum mem_cgroup_page_stat_item idx) { } diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 741206ffdace..3d8a0c79dece 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1600,7 +1600,8 @@ bool mem_cgroup_handle_oom(struct mem_cgroup *mem, gfp_t mask) * possibility of race condition. If there is, we take a lock. */ -static void mem_cgroup_update_file_stat(struct page *page, int idx, int val) +void mem_cgroup_update_page_stat(struct page *page, + enum mem_cgroup_page_stat_item idx, int val) { struct mem_cgroup *mem; struct page_cgroup *pc = lookup_page_cgroup(page); @@ -1623,30 +1624,27 @@ static void mem_cgroup_update_file_stat(struct page *page, int idx, int val) goto out; } - this_cpu_add(mem->stat->count[idx], val); - switch (idx) { - case MEM_CGROUP_STAT_FILE_MAPPED: + case MEMCG_NR_FILE_MAPPED: if (val > 0) SetPageCgroupFileMapped(pc); else if (!page_mapped(page)) ClearPageCgroupFileMapped(pc); + idx = MEM_CGROUP_STAT_FILE_MAPPED; break; default: BUG(); } + this_cpu_add(mem->stat->count[idx], val); + out: if (unlikely(need_unlock)) unlock_page_cgroup(pc); rcu_read_unlock(); return; } - -void mem_cgroup_update_file_mapped(struct page *page, int val) -{ - mem_cgroup_update_file_stat(page, MEM_CGROUP_STAT_FILE_MAPPED, val); -} +EXPORT_SYMBOL(mem_cgroup_update_page_stat); /* * size of first charge trial. "32" comes from vmscan.c's magic value. diff --git a/mm/rmap.c b/mm/rmap.c index c30f33854f97..f21f4a1d6a1c 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -937,7 +937,7 @@ void page_add_file_rmap(struct page *page) { if (atomic_inc_and_test(&page->_mapcount)) { __inc_zone_page_state(page, NR_FILE_MAPPED); - mem_cgroup_update_file_mapped(page, 1); + mem_cgroup_inc_page_stat(page, MEMCG_NR_FILE_MAPPED); } } @@ -979,7 +979,7 @@ void page_remove_rmap(struct page *page) NR_ANON_TRANSPARENT_HUGEPAGES); } else { __dec_zone_page_state(page, NR_FILE_MAPPED); - mem_cgroup_update_file_mapped(page, -1); + mem_cgroup_dec_page_stat(page, MEMCG_NR_FILE_MAPPED); } /* * It would be tidy to reset the PageAnon mapping here, -- cgit v1.2.3 From dbd4ea78f002df283c95d9774837041735fa1bf9 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Thu, 13 Jan 2011 15:47:38 -0800 Subject: memcg: add lock to synchronize page accounting and migration Introduce a new bit spin lock, PCG_MOVE_LOCK, to synchronize the page accounting and migration code. This reworks the locking scheme of _update_stat() and _move_account() by adding new lock bit PCG_MOVE_LOCK, which is always taken under IRQ disable. 1. If pages are being migrated from a memcg, then updates to that memcg page statistics are protected by grabbing PCG_MOVE_LOCK using move_lock_page_cgroup(). In an upcoming commit, memcg dirty page accounting will be updating memcg page accounting (specifically: num writeback pages) from IRQ context (softirq). Avoid a deadlocking nested spin lock attempt by disabling irq on the local processor when grabbing the PCG_MOVE_LOCK. 2. lock for update_page_stat is used only for avoiding race with move_account(). So, IRQ awareness of lock_page_cgroup() itself is not a problem. The problem is between mem_cgroup_update_page_stat() and mem_cgroup_move_account_page(). Trade-off: * Changing lock_page_cgroup() to always disable IRQ (or local_bh) has some impacts on performance and I think it's bad to disable IRQ when it's not necessary. * adding a new lock makes move_account() slower. Score is here. Performance Impact: moving a 8G anon process. Before: real 0m0.792s user 0m0.000s sys 0m0.780s After: real 0m0.854s user 0m0.000s sys 0m0.842s This score is bad but planned patches for optimization can reduce this impact. Signed-off-by: KAMEZAWA Hiroyuki Signed-off-by: Greg Thelen Reviewed-by: Minchan Kim Acked-by: Daisuke Nishimura Cc: Andrea Righi Cc: Balbir Singh Cc: Wu Fengguang Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/page_cgroup.h | 31 ++++++++++++++++++++++++++++--- mm/memcontrol.c | 9 +++++++-- 2 files changed, 35 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/page_cgroup.h b/include/linux/page_cgroup.h index fdb5a92b5ac7..5b0c971d7cae 100644 --- a/include/linux/page_cgroup.h +++ b/include/linux/page_cgroup.h @@ -35,15 +35,18 @@ struct page_cgroup *lookup_page_cgroup(struct page *page); enum { /* flags for mem_cgroup */ - PCG_LOCK, /* page cgroup is locked */ + PCG_LOCK, /* Lock for pc->mem_cgroup and following bits. */ PCG_CACHE, /* charged as cache */ PCG_USED, /* this object is in use. */ - PCG_ACCT_LRU, /* page has been accounted for */ + PCG_MIGRATION, /* under page migration */ + /* flags for mem_cgroup and file and I/O status */ + PCG_MOVE_LOCK, /* For race between move_account v.s. following bits */ PCG_FILE_MAPPED, /* page is accounted as "mapped" */ PCG_FILE_DIRTY, /* page is dirty */ PCG_FILE_WRITEBACK, /* page is under writeback */ PCG_FILE_UNSTABLE_NFS, /* page is NFS unstable */ - PCG_MIGRATION, /* under page migration */ + /* No lock in page_cgroup */ + PCG_ACCT_LRU, /* page has been accounted for (under lru_lock) */ }; #define TESTPCGFLAG(uname, lname) \ @@ -117,6 +120,10 @@ static inline enum zone_type page_cgroup_zid(struct page_cgroup *pc) static inline void lock_page_cgroup(struct page_cgroup *pc) { + /* + * Don't take this lock in IRQ context. + * This lock is for pc->mem_cgroup, USED, CACHE, MIGRATION + */ bit_spin_lock(PCG_LOCK, &pc->flags); } @@ -130,6 +137,24 @@ static inline int page_is_cgroup_locked(struct page_cgroup *pc) return bit_spin_is_locked(PCG_LOCK, &pc->flags); } +static inline void move_lock_page_cgroup(struct page_cgroup *pc, + unsigned long *flags) +{ + /* + * We know updates to pc->flags of page cache's stats are from both of + * usual context or IRQ context. Disable IRQ to avoid deadlock. + */ + local_irq_save(*flags); + bit_spin_lock(PCG_MOVE_LOCK, &pc->flags); +} + +static inline void move_unlock_page_cgroup(struct page_cgroup *pc, + unsigned long *flags) +{ + bit_spin_unlock(PCG_MOVE_LOCK, &pc->flags); + local_irq_restore(*flags); +} + #else /* CONFIG_CGROUP_MEM_RES_CTLR */ struct page_cgroup; diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 3d8a0c79dece..d888956a2cfc 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1606,6 +1606,7 @@ void mem_cgroup_update_page_stat(struct page *page, struct mem_cgroup *mem; struct page_cgroup *pc = lookup_page_cgroup(page); bool need_unlock = false; + unsigned long uninitialized_var(flags); if (unlikely(!pc)) return; @@ -1617,7 +1618,7 @@ void mem_cgroup_update_page_stat(struct page *page, /* pc->mem_cgroup is unstable ? */ if (unlikely(mem_cgroup_stealed(mem))) { /* take a lock against to access pc->mem_cgroup */ - lock_page_cgroup(pc); + move_lock_page_cgroup(pc, &flags); need_unlock = true; mem = pc->mem_cgroup; if (!mem || !PageCgroupUsed(pc)) @@ -1640,7 +1641,7 @@ void mem_cgroup_update_page_stat(struct page *page, out: if (unlikely(need_unlock)) - unlock_page_cgroup(pc); + move_unlock_page_cgroup(pc, &flags); rcu_read_unlock(); return; } @@ -2211,9 +2212,13 @@ static int mem_cgroup_move_account(struct page_cgroup *pc, struct mem_cgroup *from, struct mem_cgroup *to, bool uncharge) { int ret = -EINVAL; + unsigned long flags; + lock_page_cgroup(pc); if (PageCgroupUsed(pc) && pc->mem_cgroup == from) { + move_lock_page_cgroup(pc, &flags); __mem_cgroup_move_account(pc, from, to, uncharge); + move_unlock_page_cgroup(pc, &flags); ret = 0; } unlock_page_cgroup(pc); -- cgit v1.2.3 From 50de1dd967d4ba3b8a90ebe7a4f5feca24191317 Mon Sep 17 00:00:00 2001 From: Daisuke Nishimura Date: Thu, 13 Jan 2011 15:47:43 -0800 Subject: memcg: fix memory migration of shmem swapcache In the current implementation mem_cgroup_end_migration() decides whether the page migration has succeeded or not by checking "oldpage->mapping". But if we are tring to migrate a shmem swapcache, the page->mapping of it is NULL from the begining, so the check would be invalid. As a result, mem_cgroup_end_migration() assumes the migration has succeeded even if it's not, so "newpage" would be freed while it's not uncharged. This patch fixes it by passing mem_cgroup_end_migration() the result of the page migration. Signed-off-by: Daisuke Nishimura Reviewed-by: Minchan Kim Acked-by: KAMEZAWA Hiroyuki Acked-by: Balbir Singh Cc: Minchan Kim Reviewed-by: Johannes Weiner Cc: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memcontrol.h | 5 ++--- mm/memcontrol.c | 5 ++--- mm/migrate.c | 2 +- 3 files changed, 5 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 067115ce6b3e..6a576f989437 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -98,7 +98,7 @@ extern int mem_cgroup_prepare_migration(struct page *page, struct page *newpage, struct mem_cgroup **ptr); extern void mem_cgroup_end_migration(struct mem_cgroup *mem, - struct page *oldpage, struct page *newpage); + struct page *oldpage, struct page *newpage, bool migration_ok); /* * For memory reclaim. @@ -251,8 +251,7 @@ mem_cgroup_prepare_migration(struct page *page, struct page *newpage, } static inline void mem_cgroup_end_migration(struct mem_cgroup *mem, - struct page *oldpage, - struct page *newpage) + struct page *oldpage, struct page *newpage, bool migration_ok) { } diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 6424ba0fce83..8ab841031436 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -2896,7 +2896,7 @@ int mem_cgroup_prepare_migration(struct page *page, /* remove redundant charge if migration failed*/ void mem_cgroup_end_migration(struct mem_cgroup *mem, - struct page *oldpage, struct page *newpage) + struct page *oldpage, struct page *newpage, bool migration_ok) { struct page *used, *unused; struct page_cgroup *pc; @@ -2905,8 +2905,7 @@ void mem_cgroup_end_migration(struct mem_cgroup *mem, return; /* blocks rmdir() */ cgroup_exclude_rmdir(&mem->css); - /* at migration success, oldpage->mapping is NULL. */ - if (oldpage->mapping) { + if (!migration_ok) { used = oldpage; unused = newpage; } else { diff --git a/mm/migrate.c b/mm/migrate.c index 5b7d1fd29621..46fe8cc13d67 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -768,7 +768,7 @@ skip_unmap: uncharge: if (!charge) - mem_cgroup_end_migration(mem, page, newpage); + mem_cgroup_end_migration(mem, page, newpage, rc == 0); unlock: unlock_page(page); -- cgit v1.2.3 From 9ce137eee4febaabca81143be07d4205d2bd52d4 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Tue, 11 Jan 2011 14:07:12 -0500 Subject: nfsd: don't support msnfs export option We've long had these pointless #ifdef MSNFS's sprinkled throughout the code--pointless because MSNFS is always defined (and we give no config option to make that easy to change). So we could just remove the ifdef's and compile the resulting code unconditionally. But as long as we're there: why not just rip out this code entirely? The only purpose is to implement the "msnfs" export option which turns on Windows-like behavior in some cases, and: - the export option isn't documented anywhere; - the userland utilities (which would need to be able to parse "msnfs" in an export file) don't support it; - I don't know how to maintain this, as I don't know what the proper behavior is; and - google shows no evidence that anyone has ever used this. Signed-off-by: J. Bruce Fields --- fs/nfsd/export.c | 4 ---- fs/nfsd/vfs.c | 41 ++--------------------------------------- include/linux/nfsd/export.h | 2 +- 3 files changed, 3 insertions(+), 44 deletions(-) (limited to 'include/linux') diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index c0fcb7ab7f6d..8b31e5f8795d 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -1,4 +1,3 @@ -#define MSNFS /* HACK HACK */ /* * NFS exporting and validation. * @@ -1444,9 +1443,6 @@ static struct flags { { NFSEXP_NOSUBTREECHECK, {"no_subtree_check", ""}}, { NFSEXP_NOAUTHNLM, {"insecure_locks", ""}}, { NFSEXP_V4ROOT, {"v4root", ""}}, -#ifdef MSNFS - { NFSEXP_MSNFS, {"msnfs", ""}}, -#endif { 0, {"", ""}} }; diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index b991125ce4a5..0a01e2fc5dda 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1,4 +1,3 @@ -#define MSNFS /* HACK HACK */ /* * File operations used by nfsd. Some of these have been ripped from * other parts of the kernel because they weren't exported, others @@ -875,15 +874,6 @@ static int nfsd_direct_splice_actor(struct pipe_inode_info *pipe, return __splice_from_pipe(pipe, sd, nfsd_splice_actor); } -static inline int svc_msnfs(struct svc_fh *ffhp) -{ -#ifdef MSNFS - return (ffhp->fh_export->ex_flags & NFSEXP_MSNFS); -#else - return 0; -#endif -} - static __be32 nfsd_vfs_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, loff_t offset, struct kvec *vec, int vlen, unsigned long *count) @@ -896,9 +886,6 @@ nfsd_vfs_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, err = nfserr_perm; inode = file->f_path.dentry->d_inode; - if (svc_msnfs(fhp) && !lock_may_read(inode, offset, *count)) - goto out; - if (file->f_op->splice_read && rqstp->rq_splice_ok) { struct splice_desc sd = { .len = 0, @@ -923,7 +910,6 @@ nfsd_vfs_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, fsnotify_access(file); } else err = nfserrno(host_err); -out: return err; } @@ -988,14 +974,6 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, int stable = *stablep; int use_wgather; -#ifdef MSNFS - err = nfserr_perm; - - if ((fhp->fh_export->ex_flags & NFSEXP_MSNFS) && - (!lock_may_write(file->f_path.dentry->d_inode, offset, *cnt))) - goto out; -#endif - dentry = file->f_path.dentry; inode = dentry->d_inode; exp = fhp->fh_export; @@ -1046,7 +1024,6 @@ out_nfserr: err = 0; else err = nfserrno(host_err); -out: return err; } @@ -1751,13 +1728,6 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, if (ndentry == trap) goto out_dput_new; - if (svc_msnfs(ffhp) && - ((atomic_read(&odentry->d_count) > 1) - || (atomic_read(&ndentry->d_count) > 1))) { - host_err = -EPERM; - goto out_dput_new; - } - host_err = -EXDEV; if (ffhp->fh_export->ex_path.mnt != tfhp->fh_export->ex_path.mnt) goto out_dput_new; @@ -1836,17 +1806,10 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, if (host_err) goto out_nfserr; - if (type != S_IFDIR) { /* It's UNLINK */ -#ifdef MSNFS - if ((fhp->fh_export->ex_flags & NFSEXP_MSNFS) && - (atomic_read(&rdentry->d_count) > 1)) { - host_err = -EPERM; - } else -#endif + if (type != S_IFDIR) host_err = vfs_unlink(dirp, rdentry); - } else { /* It's RMDIR */ + else host_err = vfs_rmdir(dirp, rdentry); - } dput(rdentry); diff --git a/include/linux/nfsd/export.h b/include/linux/nfsd/export.h index 8ae78a61eea4..bd316159278c 100644 --- a/include/linux/nfsd/export.h +++ b/include/linux/nfsd/export.h @@ -35,7 +35,7 @@ #define NFSEXP_NOHIDE 0x0200 #define NFSEXP_NOSUBTREECHECK 0x0400 #define NFSEXP_NOAUTHNLM 0x0800 /* Don't authenticate NLM requests - just trust */ -#define NFSEXP_MSNFS 0x1000 /* do silly things that MS clients expect */ +#define NFSEXP_MSNFS 0x1000 /* do silly things that MS clients expect; no longer supported */ #define NFSEXP_FSID 0x2000 #define NFSEXP_CROSSMOUNT 0x4000 #define NFSEXP_NOACL 0x8000 /* reserved for possible ACL related use */ -- cgit v1.2.3 From 2c6755988afc003a0332406a134fb6a1626f9b28 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Fri, 14 Jan 2011 02:36:43 +0000 Subject: fs: hlist UP debug fixup Po-Yu Chuang noticed that hlist_bl_set_first could crash on a UP system when LIST_BL_LOCKMASK is 0, because LIST_BL_BUG_ON(!((unsigned long)h->first & LIST_BL_LOCKMASK)); always evaulates to true. Fix the expression, and also avoid a dependency between bit spinlock implementation and list bl code (list code shouldn't know anything except that bit 0 is set when adding and removing elements). Eventually if a good use case comes up, we might use this list to store 1 or more arbitrary bits of data, so it really shouldn't be tied to locking either, but for now they are helpful for debugging. Signed-off-by: Nick Piggin --- include/linux/list_bl.h | 5 +++-- include/linux/rculist_bl.h | 3 ++- 2 files changed, 5 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/list_bl.h b/include/linux/list_bl.h index 9ee97e7f2be4..b2adbb4b2f73 100644 --- a/include/linux/list_bl.h +++ b/include/linux/list_bl.h @@ -16,7 +16,7 @@ * some fast and compact auxiliary data. */ -#if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) +#if defined(CONFIG_SMP) #define LIST_BL_LOCKMASK 1UL #else #define LIST_BL_LOCKMASK 0UL @@ -62,7 +62,8 @@ static inline void hlist_bl_set_first(struct hlist_bl_head *h, struct hlist_bl_node *n) { LIST_BL_BUG_ON((unsigned long)n & LIST_BL_LOCKMASK); - LIST_BL_BUG_ON(!((unsigned long)h->first & LIST_BL_LOCKMASK)); + LIST_BL_BUG_ON(((unsigned long)h->first & LIST_BL_LOCKMASK) != + LIST_BL_LOCKMASK); h->first = (struct hlist_bl_node *)((unsigned long)n | LIST_BL_LOCKMASK); } diff --git a/include/linux/rculist_bl.h b/include/linux/rculist_bl.h index b872b493724d..cf1244fbf3b6 100644 --- a/include/linux/rculist_bl.h +++ b/include/linux/rculist_bl.h @@ -11,7 +11,8 @@ static inline void hlist_bl_set_first_rcu(struct hlist_bl_head *h, struct hlist_bl_node *n) { LIST_BL_BUG_ON((unsigned long)n & LIST_BL_LOCKMASK); - LIST_BL_BUG_ON(!((unsigned long)h->first & LIST_BL_LOCKMASK)); + LIST_BL_BUG_ON(((unsigned long)h->first & LIST_BL_LOCKMASK) != + LIST_BL_LOCKMASK); rcu_assign_pointer(h->first, (struct hlist_bl_node *)((unsigned long)n | LIST_BL_LOCKMASK)); } -- cgit v1.2.3 From 1ac9ad1394fa542ac7ae0dc943ee3cda678799fa Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 12 Jan 2011 12:13:14 +0000 Subject: net: remove dev_txq_stats_fold() After recent changes, (percpu stats on vlan/tunnels...), we dont need anymore per struct netdev_queue tx_bytes/tx_packets/tx_dropped counters. Only remaining users are ixgbe, sch_teql, gianfar & macvlan : 1) ixgbe can be converted to use existing tx_ring counters. 2) macvlan incremented txq->tx_dropped, it can use the dev->stats.tx_dropped counter. 3) sch_teql : almost revert ab35cd4b8f42 (Use net_device internal stats) Now we have ndo_get_stats64(), use it, even for "unsigned long" fields (No need to bring back a struct net_device_stats) 4) gianfar adds a stats structure per tx queue to hold tx_bytes/tx_packets This removes a lockdep warning (and possible lockup) in rndis gadget, calling dev_get_stats() from hard IRQ context. Ref: http://www.spinics.net/lists/netdev/msg149202.html Reported-by: Neil Jones Signed-off-by: Eric Dumazet CC: Jarek Poplawski CC: Alexander Duyck CC: Jeff Kirsher CC: Sandeep Gopalpet CC: Michal Nazarewicz Signed-off-by: David S. Miller --- drivers/net/gianfar.c | 10 ++++------ drivers/net/gianfar.h | 10 ++++++++++ drivers/net/ixgbe/ixgbe_main.c | 23 ++++++++++++++++------- drivers/net/macvtap.c | 2 +- include/linux/netdevice.h | 5 ----- net/core/dev.c | 29 ----------------------------- net/sched/sch_teql.c | 26 +++++++++++++++++++++----- 7 files changed, 52 insertions(+), 53 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/gianfar.c b/drivers/net/gianfar.c index 45c4b7bfcf39..f1d4b450e797 100644 --- a/drivers/net/gianfar.c +++ b/drivers/net/gianfar.c @@ -433,7 +433,6 @@ static void gfar_init_mac(struct net_device *ndev) static struct net_device_stats *gfar_get_stats(struct net_device *dev) { struct gfar_private *priv = netdev_priv(dev); - struct netdev_queue *txq; unsigned long rx_packets = 0, rx_bytes = 0, rx_dropped = 0; unsigned long tx_packets = 0, tx_bytes = 0; int i = 0; @@ -449,9 +448,8 @@ static struct net_device_stats *gfar_get_stats(struct net_device *dev) dev->stats.rx_dropped = rx_dropped; for (i = 0; i < priv->num_tx_queues; i++) { - txq = netdev_get_tx_queue(dev, i); - tx_bytes += txq->tx_bytes; - tx_packets += txq->tx_packets; + tx_bytes += priv->tx_queue[i]->stats.tx_bytes; + tx_packets += priv->tx_queue[i]->stats.tx_packets; } dev->stats.tx_bytes = tx_bytes; @@ -2108,8 +2106,8 @@ static int gfar_start_xmit(struct sk_buff *skb, struct net_device *dev) } /* Update transmit stats */ - txq->tx_bytes += skb->len; - txq->tx_packets ++; + tx_queue->stats.tx_bytes += skb->len; + tx_queue->stats.tx_packets++; txbdp = txbdp_start = tx_queue->cur_tx; lstatus = txbdp->lstatus; diff --git a/drivers/net/gianfar.h b/drivers/net/gianfar.h index 68984eb88ae0..54de4135e932 100644 --- a/drivers/net/gianfar.h +++ b/drivers/net/gianfar.h @@ -907,12 +907,21 @@ enum { MQ_MG_MODE }; +/* + * Per TX queue stats + */ +struct tx_q_stats { + unsigned long tx_packets; + unsigned long tx_bytes; +}; + /** * struct gfar_priv_tx_q - per tx queue structure * @txlock: per queue tx spin lock * @tx_skbuff:skb pointers * @skb_curtx: to be used skb pointer * @skb_dirtytx:the last used skb pointer + * @stats: bytes/packets stats * @qindex: index of this queue * @dev: back pointer to the dev structure * @grp: back pointer to the group to which this queue belongs @@ -934,6 +943,7 @@ struct gfar_priv_tx_q { struct txbd8 *tx_bd_base; struct txbd8 *cur_tx; struct txbd8 *dirty_tx; + struct tx_q_stats stats; struct net_device *dev; struct gfar_priv_grp *grp; u16 skb_curtx; diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c index a060610a42db..602078b84892 100644 --- a/drivers/net/ixgbe/ixgbe_main.c +++ b/drivers/net/ixgbe/ixgbe_main.c @@ -6667,8 +6667,6 @@ netdev_tx_t ixgbe_xmit_frame_ring(struct sk_buff *skb, struct ixgbe_adapter *adapter, struct ixgbe_ring *tx_ring) { - struct net_device *netdev = tx_ring->netdev; - struct netdev_queue *txq; unsigned int first; unsigned int tx_flags = 0; u8 hdr_len = 0; @@ -6765,9 +6763,6 @@ netdev_tx_t ixgbe_xmit_frame_ring(struct sk_buff *skb, /* add the ATR filter if ATR is on */ if (test_bit(__IXGBE_TX_FDIR_INIT_DONE, &tx_ring->state)) ixgbe_atr(tx_ring, skb, tx_flags, protocol); - txq = netdev_get_tx_queue(netdev, tx_ring->queue_index); - txq->tx_bytes += skb->len; - txq->tx_packets++; ixgbe_tx_queue(tx_ring, tx_flags, count, skb->len, hdr_len); ixgbe_maybe_stop_tx(tx_ring, DESC_NEEDED); @@ -6925,8 +6920,6 @@ static struct rtnl_link_stats64 *ixgbe_get_stats64(struct net_device *netdev, struct ixgbe_adapter *adapter = netdev_priv(netdev); int i; - /* accurate rx/tx bytes/packets stats */ - dev_txq_stats_fold(netdev, stats); rcu_read_lock(); for (i = 0; i < adapter->num_rx_queues; i++) { struct ixgbe_ring *ring = ACCESS_ONCE(adapter->rx_ring[i]); @@ -6943,6 +6936,22 @@ static struct rtnl_link_stats64 *ixgbe_get_stats64(struct net_device *netdev, stats->rx_bytes += bytes; } } + + for (i = 0; i < adapter->num_tx_queues; i++) { + struct ixgbe_ring *ring = ACCESS_ONCE(adapter->tx_ring[i]); + u64 bytes, packets; + unsigned int start; + + if (ring) { + do { + start = u64_stats_fetch_begin_bh(&ring->syncp); + packets = ring->stats.packets; + bytes = ring->stats.bytes; + } while (u64_stats_fetch_retry_bh(&ring->syncp, start)); + stats->tx_packets += packets; + stats->tx_bytes += bytes; + } + } rcu_read_unlock(); /* following stats updated by ixgbe_watchdog_task() */ stats->multicast = netdev->stats.multicast; diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c index 21845affea13..5933621ac3ff 100644 --- a/drivers/net/macvtap.c +++ b/drivers/net/macvtap.c @@ -585,7 +585,7 @@ err: rcu_read_lock_bh(); vlan = rcu_dereference(q->vlan); if (vlan) - netdev_get_tx_queue(vlan->dev, 0)->tx_dropped++; + vlan->dev->stats.tx_dropped++; rcu_read_unlock_bh(); return err; diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index be4957cf6511..d971346b0340 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -520,9 +520,6 @@ struct netdev_queue { * please use this field instead of dev->trans_start */ unsigned long trans_start; - u64 tx_bytes; - u64 tx_packets; - u64 tx_dropped; } ____cacheline_aligned_in_smp; static inline int netdev_queue_numa_node_read(const struct netdev_queue *q) @@ -2265,8 +2262,6 @@ extern void dev_load(struct net *net, const char *name); extern void dev_mcast_init(void); extern struct rtnl_link_stats64 *dev_get_stats(struct net_device *dev, struct rtnl_link_stats64 *storage); -extern void dev_txq_stats_fold(const struct net_device *dev, - struct rtnl_link_stats64 *stats); extern int netdev_max_backlog; extern int netdev_tstamp_prequeue; diff --git a/net/core/dev.c b/net/core/dev.c index a3ef808b5e36..83507c265e48 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5523,34 +5523,6 @@ void netdev_run_todo(void) } } -/** - * dev_txq_stats_fold - fold tx_queues stats - * @dev: device to get statistics from - * @stats: struct rtnl_link_stats64 to hold results - */ -void dev_txq_stats_fold(const struct net_device *dev, - struct rtnl_link_stats64 *stats) -{ - u64 tx_bytes = 0, tx_packets = 0, tx_dropped = 0; - unsigned int i; - struct netdev_queue *txq; - - for (i = 0; i < dev->num_tx_queues; i++) { - txq = netdev_get_tx_queue(dev, i); - spin_lock_bh(&txq->_xmit_lock); - tx_bytes += txq->tx_bytes; - tx_packets += txq->tx_packets; - tx_dropped += txq->tx_dropped; - spin_unlock_bh(&txq->_xmit_lock); - } - if (tx_bytes || tx_packets || tx_dropped) { - stats->tx_bytes = tx_bytes; - stats->tx_packets = tx_packets; - stats->tx_dropped = tx_dropped; - } -} -EXPORT_SYMBOL(dev_txq_stats_fold); - /* Convert net_device_stats to rtnl_link_stats64. They have the same * fields in the same order, with only the type differing. */ @@ -5594,7 +5566,6 @@ struct rtnl_link_stats64 *dev_get_stats(struct net_device *dev, netdev_stats_to_stats64(storage, ops->ndo_get_stats(dev)); } else { netdev_stats_to_stats64(storage, &dev->stats); - dev_txq_stats_fold(dev, storage); } storage->rx_dropped += atomic_long_read(&dev->rx_dropped); return storage; diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c index af9360d1f6eb..84ce48eadff4 100644 --- a/net/sched/sch_teql.c +++ b/net/sched/sch_teql.c @@ -59,6 +59,10 @@ struct teql_master struct net_device *dev; struct Qdisc *slaves; struct list_head master_list; + unsigned long tx_bytes; + unsigned long tx_packets; + unsigned long tx_errors; + unsigned long tx_dropped; }; struct teql_sched_data @@ -274,7 +278,6 @@ static inline int teql_resolve(struct sk_buff *skb, static netdev_tx_t teql_master_xmit(struct sk_buff *skb, struct net_device *dev) { struct teql_master *master = netdev_priv(dev); - struct netdev_queue *txq = netdev_get_tx_queue(dev, 0); struct Qdisc *start, *q; int busy; int nores; @@ -314,8 +317,8 @@ restart: __netif_tx_unlock(slave_txq); master->slaves = NEXT_SLAVE(q); netif_wake_queue(dev); - txq->tx_packets++; - txq->tx_bytes += length; + master->tx_packets++; + master->tx_bytes += length; return NETDEV_TX_OK; } __netif_tx_unlock(slave_txq); @@ -342,10 +345,10 @@ restart: netif_stop_queue(dev); return NETDEV_TX_BUSY; } - dev->stats.tx_errors++; + master->tx_errors++; drop: - txq->tx_dropped++; + master->tx_dropped++; dev_kfree_skb(skb); return NETDEV_TX_OK; } @@ -398,6 +401,18 @@ static int teql_master_close(struct net_device *dev) return 0; } +static struct rtnl_link_stats64 *teql_master_stats64(struct net_device *dev, + struct rtnl_link_stats64 *stats) +{ + struct teql_master *m = netdev_priv(dev); + + stats->tx_packets = m->tx_packets; + stats->tx_bytes = m->tx_bytes; + stats->tx_errors = m->tx_errors; + stats->tx_dropped = m->tx_dropped; + return stats; +} + static int teql_master_mtu(struct net_device *dev, int new_mtu) { struct teql_master *m = netdev_priv(dev); @@ -422,6 +437,7 @@ static const struct net_device_ops teql_netdev_ops = { .ndo_open = teql_master_open, .ndo_stop = teql_master_close, .ndo_start_xmit = teql_master_xmit, + .ndo_get_stats64 = teql_master_stats64, .ndo_change_mtu = teql_master_mtu, }; -- cgit v1.2.3 From 51e7eed79c41180919ff94942895ba38467d9ad4 Mon Sep 17 00:00:00 2001 From: Tobias Klauser Date: Wed, 12 Jan 2011 22:14:56 +0000 Subject: etherdevice.h: Add is_unicast_ether_addr function From a check for !is_multicast_ether_addr it is not always obvious that we're checking for a unicast address. So add this helper function to make those code paths easier to read. Signed-off-by: Tobias Klauser Acked-by: Chris Metcalf Signed-off-by: David S. Miller --- include/linux/etherdevice.h | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'include/linux') diff --git a/include/linux/etherdevice.h b/include/linux/etherdevice.h index bec8b82889bf..ab68f785fd19 100644 --- a/include/linux/etherdevice.h +++ b/include/linux/etherdevice.h @@ -98,6 +98,17 @@ static inline int is_broadcast_ether_addr(const u8 *addr) return (addr[0] & addr[1] & addr[2] & addr[3] & addr[4] & addr[5]) == 0xff; } +/** + * is_unicast_ether_addr - Determine if the Ethernet address is unicast + * @addr: Pointer to a six-byte array containing the Ethernet address + * + * Return true if the address is a unicast address. + */ +static inline int is_unicast_ether_addr(const u8 *addr) +{ + return !is_multicast_ether_addr(addr); +} + /** * is_valid_ether_addr - Determine if the given Ethernet address is valid * @addr: Pointer to a six-byte array containing the Ethernet address -- cgit v1.2.3 From bba958783b1b4cb0a9420f4e11082467132a334c Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Fri, 14 Jan 2011 15:57:47 +0900 Subject: mmc: sh_mmcif: Convert to __raw_xxx() I/O accessors. When using the I/O accessors in raw mode from the boot stubs we don't want to bother with any of the complexity associated with readl/writel and friends. Furthermore, utilization within the context of the host driver itself is all performed on an ioremapped window, so using the __raw variants there doesn't pose any problem either. If and when barriers need to be added in the future, these will need to be explicitly written out, but this is so far not a concern for any of the affected CPUs in question. This fixes up the link error introduced by the ARM tree via its barrier refactoring: arch/arm/boot/compressed/mmcif-sh7372.o: In function `mmcif_loader': mmcif-sh7372.c:(.text+0x9e8): undefined reference to `outer_cache Following the change in: http://www.arm.linux.org.uk/developer/patches/viewpatch.php?id=6275/1 Reported-by: Simon Horman Signed-off-by: Paul Mundt --- include/linux/mmc/sh_mmcif.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mmc/sh_mmcif.h b/include/linux/mmc/sh_mmcif.h index bf173502d744..38d393092812 100644 --- a/include/linux/mmc/sh_mmcif.h +++ b/include/linux/mmc/sh_mmcif.h @@ -94,12 +94,12 @@ struct sh_mmcif_plat_data { static inline u32 sh_mmcif_readl(void __iomem *addr, int reg) { - return readl(addr + reg); + return __raw_readl(addr + reg); } static inline void sh_mmcif_writel(void __iomem *addr, int reg, u32 val) { - writel(val, addr + reg); + __raw_writel(val, addr + reg); } #define SH_MMCIF_BBS 512 /* boot block size */ -- cgit v1.2.3 From 412dc11d3fd01f96fdf4a8cbfbc5584a17dab7c8 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Wed, 24 Nov 2010 18:01:41 +0000 Subject: mfd: Add WM8326 support The WM8326 is a high performance variant of the WM832x series with no software visible differences. Signed-off-by: Mark Brown Signed-off-by: Samuel Ortiz --- drivers/mfd/wm831x-core.c | 7 +++++++ drivers/mfd/wm831x-i2c.c | 1 + drivers/mfd/wm831x-spi.c | 18 ++++++++++++++++++ include/linux/mfd/wm831x/core.h | 1 + 4 files changed, 27 insertions(+) (limited to 'include/linux') diff --git a/drivers/mfd/wm831x-core.c b/drivers/mfd/wm831x-core.c index d1b2d534b991..3fe9a58fe6c7 100644 --- a/drivers/mfd/wm831x-core.c +++ b/drivers/mfd/wm831x-core.c @@ -1541,6 +1541,12 @@ int wm831x_device_init(struct wm831x *wm831x, unsigned long id, int irq) dev_info(wm831x->dev, "WM8325 revision %c\n", 'A' + rev); break; + case WM8326: + parent = WM8326; + wm831x->num_gpio = 12; + dev_info(wm831x->dev, "WM8326 revision %c\n", 'A' + rev); + break; + default: dev_err(wm831x->dev, "Unknown WM831x device %04x\n", ret); ret = -EINVAL; @@ -1612,6 +1618,7 @@ int wm831x_device_init(struct wm831x *wm831x, unsigned long id, int irq) case WM8320: case WM8321: case WM8325: + case WM8326: ret = mfd_add_devices(wm831x->dev, -1, wm8320_devs, ARRAY_SIZE(wm8320_devs), NULL, wm831x->irq_base); diff --git a/drivers/mfd/wm831x-i2c.c b/drivers/mfd/wm831x-i2c.c index 156b19859e81..38be5201c783 100644 --- a/drivers/mfd/wm831x-i2c.c +++ b/drivers/mfd/wm831x-i2c.c @@ -108,6 +108,7 @@ static const struct i2c_device_id wm831x_i2c_id[] = { { "wm8320", WM8320 }, { "wm8321", WM8321 }, { "wm8325", WM8325 }, + { "wm8326", WM8326 }, { } }; MODULE_DEVICE_TABLE(i2c, wm831x_i2c_id); diff --git a/drivers/mfd/wm831x-spi.c b/drivers/mfd/wm831x-spi.c index 2789b151b0f9..0a8f772be88c 100644 --- a/drivers/mfd/wm831x-spi.c +++ b/drivers/mfd/wm831x-spi.c @@ -81,6 +81,8 @@ static int __devinit wm831x_spi_probe(struct spi_device *spi) type = WM8321; else if (strcmp(spi->modalias, "wm8325") == 0) type = WM8325; + else if (strcmp(spi->modalias, "wm8326") == 0) + type = WM8326; else { dev_err(&spi->dev, "Unknown device type\n"); return -EINVAL; @@ -184,6 +186,17 @@ static struct spi_driver wm8325_spi_driver = { .suspend = wm831x_spi_suspend, }; +static struct spi_driver wm8326_spi_driver = { + .driver = { + .name = "wm8326", + .bus = &spi_bus_type, + .owner = THIS_MODULE, + }, + .probe = wm831x_spi_probe, + .remove = __devexit_p(wm831x_spi_remove), + .suspend = wm831x_spi_suspend, +}; + static int __init wm831x_spi_init(void) { int ret; @@ -212,12 +225,17 @@ static int __init wm831x_spi_init(void) if (ret != 0) pr_err("Failed to register WM8325 SPI driver: %d\n", ret); + ret = spi_register_driver(&wm8326_spi_driver); + if (ret != 0) + pr_err("Failed to register WM8326 SPI driver: %d\n", ret); + return 0; } subsys_initcall(wm831x_spi_init); static void __exit wm831x_spi_exit(void) { + spi_unregister_driver(&wm8326_spi_driver); spi_unregister_driver(&wm8325_spi_driver); spi_unregister_driver(&wm8321_spi_driver); spi_unregister_driver(&wm8320_spi_driver); diff --git a/include/linux/mfd/wm831x/core.h b/include/linux/mfd/wm831x/core.h index a1239c48b41a..903280d21866 100644 --- a/include/linux/mfd/wm831x/core.h +++ b/include/linux/mfd/wm831x/core.h @@ -245,6 +245,7 @@ enum wm831x_parent { WM8320 = 0x8320, WM8321 = 0x8321, WM8325 = 0x8325, + WM8326 = 0x8326, }; struct wm831x { -- cgit v1.2.3 From 4c90aa94f6b3e33f57faaf19ef9819195dff61d3 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Fri, 26 Nov 2010 17:19:34 +0000 Subject: mfd: Provide pm_runtime_no_callbacks flag in cell data Allow MFD cells to have pm_runtime_no_callbacks() called on them during registration. This causes the runtime PM framework to ignore them, allowing use of runtime PM to suspend the device as a whole even if not all drivers for the MFD can usefully implement runtime PM. For example, RTCs are likely to run continuously regardless of the power state of the system. Signed-off-by: Mark Brown Signed-off-by: Samuel Ortiz --- drivers/mfd/mfd-core.c | 4 ++++ include/linux/mfd/core.h | 6 ++++++ 2 files changed, 10 insertions(+) (limited to 'include/linux') diff --git a/drivers/mfd/mfd-core.c b/drivers/mfd/mfd-core.c index ec99f681e773..d83ad0f141af 100644 --- a/drivers/mfd/mfd-core.c +++ b/drivers/mfd/mfd-core.c @@ -15,6 +15,7 @@ #include #include #include +#include #include static int mfd_add_device(struct device *parent, int id, @@ -82,6 +83,9 @@ static int mfd_add_device(struct device *parent, int id, if (ret) goto fail_res; + if (cell->pm_runtime_no_callbacks) + pm_runtime_no_callbacks(&pdev->dev); + kfree(res); return 0; diff --git a/include/linux/mfd/core.h b/include/linux/mfd/core.h index 5582ab3d3e48..835996e167e1 100644 --- a/include/linux/mfd/core.h +++ b/include/linux/mfd/core.h @@ -47,6 +47,12 @@ struct mfd_cell { /* don't check for resource conflicts */ bool ignore_resource_conflicts; + + /* + * Disable runtime PM callbacks for this subdevice - see + * pm_runtime_no_callbacks(). + */ + bool pm_runtime_no_callbacks; }; extern int mfd_add_devices(struct device *parent, int id, -- cgit v1.2.3 From e098aded79f24e2024139e82f778ff9db6dc142a Mon Sep 17 00:00:00 2001 From: Mattias Wallin Date: Thu, 2 Dec 2010 15:40:31 +0100 Subject: mfd: ab8500-core ioresources irq for subdrivers added This patch adds the ioresources used by subdrivers to retrieve their interrupt. Signed-off-by: Mattias Wallin Signed-off-by: Samuel Ortiz --- drivers/mfd/ab8500-core.c | 206 +++++++++++++++++++++++++++++++++++++++++++-- include/linux/mfd/ab8500.h | 4 +- 2 files changed, 201 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/drivers/mfd/ab8500-core.c b/drivers/mfd/ab8500-core.c index e464e94c4bc0..7f01a3a1d10f 100644 --- a/drivers/mfd/ab8500-core.c +++ b/drivers/mfd/ab8500-core.c @@ -397,12 +397,188 @@ static struct resource ab8500_poweronkey_db_resources[] = { }, }; +static struct resource ab8500_bm_resources[] = { + { + .name = "MAIN_EXT_CH_NOT_OK", + .start = AB8500_INT_MAIN_EXT_CH_NOT_OK, + .end = AB8500_INT_MAIN_EXT_CH_NOT_OK, + .flags = IORESOURCE_IRQ, + }, + { + .name = "BATT_OVV", + .start = AB8500_INT_BATT_OVV, + .end = AB8500_INT_BATT_OVV, + .flags = IORESOURCE_IRQ, + }, + { + .name = "MAIN_CH_UNPLUG_DET", + .start = AB8500_INT_MAIN_CH_UNPLUG_DET, + .end = AB8500_INT_MAIN_CH_UNPLUG_DET, + .flags = IORESOURCE_IRQ, + }, + { + .name = "MAIN_CHARGE_PLUG_DET", + .start = AB8500_INT_MAIN_CH_PLUG_DET, + .end = AB8500_INT_MAIN_CH_PLUG_DET, + .flags = IORESOURCE_IRQ, + }, + { + .name = "VBUS_DET_F", + .start = AB8500_INT_VBUS_DET_F, + .end = AB8500_INT_VBUS_DET_F, + .flags = IORESOURCE_IRQ, + }, + { + .name = "VBUS_DET_R", + .start = AB8500_INT_VBUS_DET_R, + .end = AB8500_INT_VBUS_DET_R, + .flags = IORESOURCE_IRQ, + }, + { + .name = "BAT_CTRL_INDB", + .start = AB8500_INT_BAT_CTRL_INDB, + .end = AB8500_INT_BAT_CTRL_INDB, + .flags = IORESOURCE_IRQ, + }, + { + .name = "CH_WD_EXP", + .start = AB8500_INT_CH_WD_EXP, + .end = AB8500_INT_CH_WD_EXP, + .flags = IORESOURCE_IRQ, + }, + { + .name = "VBUS_OVV", + .start = AB8500_INT_VBUS_OVV, + .end = AB8500_INT_VBUS_OVV, + .flags = IORESOURCE_IRQ, + }, + { + .name = "NCONV_ACCU", + .start = AB8500_INT_CCN_CONV_ACC, + .end = AB8500_INT_CCN_CONV_ACC, + .flags = IORESOURCE_IRQ, + }, + { + .name = "LOW_BAT_F", + .start = AB8500_INT_LOW_BAT_F, + .end = AB8500_INT_LOW_BAT_F, + .flags = IORESOURCE_IRQ, + }, + { + .name = "LOW_BAT_R", + .start = AB8500_INT_LOW_BAT_R, + .end = AB8500_INT_LOW_BAT_R, + .flags = IORESOURCE_IRQ, + }, + { + .name = "BTEMP_LOW", + .start = AB8500_INT_BTEMP_LOW, + .end = AB8500_INT_BTEMP_LOW, + .flags = IORESOURCE_IRQ, + }, + { + .name = "BTEMP_HIGH", + .start = AB8500_INT_BTEMP_HIGH, + .end = AB8500_INT_BTEMP_HIGH, + .flags = IORESOURCE_IRQ, + }, + { + .name = "USB_CHARGER_NOT_OKR", + .start = AB8500_INT_USB_CHARGER_NOT_OK, + .end = AB8500_INT_USB_CHARGER_NOT_OK, + .flags = IORESOURCE_IRQ, + }, + { + .name = "USB_CHARGE_DET_DONE", + .start = AB8500_INT_USB_CHG_DET_DONE, + .end = AB8500_INT_USB_CHG_DET_DONE, + .flags = IORESOURCE_IRQ, + }, + { + .name = "USB_CH_TH_PROT_R", + .start = AB8500_INT_USB_CH_TH_PROT_R, + .end = AB8500_INT_USB_CH_TH_PROT_R, + .flags = IORESOURCE_IRQ, + }, + { + .name = "MAIN_CH_TH_PROT_R", + .start = AB8500_INT_MAIN_CH_TH_PROT_R, + .end = AB8500_INT_MAIN_CH_TH_PROT_R, + .flags = IORESOURCE_IRQ, + }, + { + .name = "USB_CHARGER_NOT_OKF", + .start = AB8500_INT_USB_CHARGER_NOT_OKF, + .end = AB8500_INT_USB_CHARGER_NOT_OKF, + .flags = IORESOURCE_IRQ, + }, +}; + +static struct resource ab8500_debug_resources[] = { + { + .name = "IRQ_FIRST", + .start = AB8500_INT_MAIN_EXT_CH_NOT_OK, + .end = AB8500_INT_MAIN_EXT_CH_NOT_OK, + .flags = IORESOURCE_IRQ, + }, + { + .name = "IRQ_LAST", + .start = AB8500_INT_USB_CHARGER_NOT_OKF, + .end = AB8500_INT_USB_CHARGER_NOT_OKF, + .flags = IORESOURCE_IRQ, + }, +}; + +static struct resource ab8500_usb_resources[] = { + { + .name = "ID_WAKEUP_R", + .start = AB8500_INT_ID_WAKEUP_R, + .end = AB8500_INT_ID_WAKEUP_R, + .flags = IORESOURCE_IRQ, + }, + { + .name = "ID_WAKEUP_F", + .start = AB8500_INT_ID_WAKEUP_F, + .end = AB8500_INT_ID_WAKEUP_F, + .flags = IORESOURCE_IRQ, + }, + { + .name = "VBUS_DET_F", + .start = AB8500_INT_VBUS_DET_F, + .end = AB8500_INT_VBUS_DET_F, + .flags = IORESOURCE_IRQ, + }, + { + .name = "VBUS_DET_R", + .start = AB8500_INT_VBUS_DET_R, + .end = AB8500_INT_VBUS_DET_R, + .flags = IORESOURCE_IRQ, + }, +}; + +static struct resource ab8500_temp_resources[] = { + { + .name = "AB8500_TEMP_WARM", + .start = AB8500_INT_TEMP_WARM, + .end = AB8500_INT_TEMP_WARM, + .flags = IORESOURCE_IRQ, + }, +}; + static struct mfd_cell ab8500_devs[] = { #ifdef CONFIG_DEBUG_FS { .name = "ab8500-debug", + .num_resources = ARRAY_SIZE(ab8500_debug_resources), + .resources = ab8500_debug_resources, }, #endif + { + .name = "ab8500-sysctrl", + }, + { + .name = "ab8500-regulator", + }, { .name = "ab8500-gpadc", .num_resources = ARRAY_SIZE(ab8500_gpadc_resources), @@ -413,6 +589,22 @@ static struct mfd_cell ab8500_devs[] = { .num_resources = ARRAY_SIZE(ab8500_rtc_resources), .resources = ab8500_rtc_resources, }, + { + .name = "ab8500-bm", + .num_resources = ARRAY_SIZE(ab8500_bm_resources), + .resources = ab8500_bm_resources, + }, + { .name = "ab8500-codec", }, + { + .name = "ab8500-usb", + .num_resources = ARRAY_SIZE(ab8500_usb_resources), + .resources = ab8500_usb_resources, + }, + { + .name = "ab8500-poweron-key", + .num_resources = ARRAY_SIZE(ab8500_poweronkey_db_resources), + .resources = ab8500_poweronkey_db_resources, + }, { .name = "ab8500-pwm", .id = 1, @@ -425,14 +617,14 @@ static struct mfd_cell ab8500_devs[] = { .name = "ab8500-pwm", .id = 3, }, - { .name = "ab8500-charger", }, - { .name = "ab8500-audio", }, - { .name = "ab8500-usb", }, - { .name = "ab8500-regulator", }, + { .name = "ab8500-leds", }, { - .name = "ab8500-poweron-key", - .num_resources = ARRAY_SIZE(ab8500_poweronkey_db_resources), - .resources = ab8500_poweronkey_db_resources, + .name = "ab8500-denc", + }, + { + .name = "ab8500-temp", + .num_resources = ARRAY_SIZE(ab8500_temp_resources), + .resources = ab8500_temp_resources, }, }; diff --git a/include/linux/mfd/ab8500.h b/include/linux/mfd/ab8500.h index 85cf2c28fac6..1ad169655235 100644 --- a/include/linux/mfd/ab8500.h +++ b/include/linux/mfd/ab8500.h @@ -91,8 +91,8 @@ #define AB8500_INT_ID_DET_R4F 93 #define AB8500_INT_USB_CHG_DET_DONE 94 #define AB8500_INT_USB_CH_TH_PROT_F 96 -#define AB8500_INT_USB_CH_TH_PROP_R 97 -#define AB8500_INT_MAIN_CH_TH_PROP_F 98 +#define AB8500_INT_USB_CH_TH_PROT_R 97 +#define AB8500_INT_MAIN_CH_TH_PROT_F 98 #define AB8500_INT_MAIN_CH_TH_PROT_R 99 #define AB8500_INT_USB_CHARGER_NOT_OKF 103 -- cgit v1.2.3 From cdd137c9c86c201ddb7f42ec978d2da45e7b7a17 Mon Sep 17 00:00:00 2001 From: MyungJoo Ham Date: Thu, 23 Dec 2010 17:53:36 +0900 Subject: mfd: MAX8998/LP3974 hibernation support This patch makes the driver to save and restore register values for hibernation. Signed-off-by: MyungJoo Ham Signed-off-by: Kyungmin Park Signed-off-by: Samuel Ortiz --- drivers/mfd/max8998-irq.c | 7 +++ drivers/mfd/max8998.c | 108 ++++++++++++++++++++++++++++++++++++ include/linux/mfd/max8998-private.h | 2 + include/linux/mfd/max8998.h | 1 + 4 files changed, 118 insertions(+) (limited to 'include/linux') diff --git a/drivers/mfd/max8998-irq.c b/drivers/mfd/max8998-irq.c index c6b61fcc5808..3903e1fbb334 100644 --- a/drivers/mfd/max8998-irq.c +++ b/drivers/mfd/max8998-irq.c @@ -183,6 +183,13 @@ static irqreturn_t max8998_irq_thread(int irq, void *data) return IRQ_HANDLED; } +int max8998_irq_resume(struct max8998_dev *max8998) +{ + if (max8998->irq && max8998->irq_base) + max8998_irq_thread(max8998->irq_base, max8998); + return 0; +} + int max8998_irq_init(struct max8998_dev *max8998) { int i; diff --git a/drivers/mfd/max8998.c b/drivers/mfd/max8998.c index bb9977bebe78..5ce00ad5241a 100644 --- a/drivers/mfd/max8998.c +++ b/drivers/mfd/max8998.c @@ -25,6 +25,8 @@ #include #include #include +#include +#include #include #include #include @@ -135,6 +137,7 @@ static int max8998_i2c_probe(struct i2c_client *i2c, if (pdata) { max8998->ono = pdata->ono; max8998->irq_base = pdata->irq_base; + max8998->wakeup = pdata->wakeup; } mutex_init(&max8998->iolock); @@ -146,6 +149,8 @@ static int max8998_i2c_probe(struct i2c_client *i2c, ret = mfd_add_devices(max8998->dev, -1, max8998_devs, ARRAY_SIZE(max8998_devs), NULL, 0); + pm_runtime_set_active(max8998->dev); + if (ret < 0) goto err; @@ -178,10 +183,113 @@ static const struct i2c_device_id max8998_i2c_id[] = { }; MODULE_DEVICE_TABLE(i2c, max8998_i2c_id); +static int max8998_suspend(struct device *dev) +{ + struct i2c_client *i2c = container_of(dev, struct i2c_client, dev); + struct max8998_dev *max8998 = i2c_get_clientdata(i2c); + + if (max8998->wakeup) + set_irq_wake(max8998->irq, 1); + return 0; +} + +static int max8998_resume(struct device *dev) +{ + struct i2c_client *i2c = container_of(dev, struct i2c_client, dev); + struct max8998_dev *max8998 = i2c_get_clientdata(i2c); + + if (max8998->wakeup) + set_irq_wake(max8998->irq, 0); + /* + * In LP3974, if IRQ registers are not "read & clear" + * when it's set during sleep, the interrupt becomes + * disabled. + */ + return max8998_irq_resume(i2c_get_clientdata(i2c)); +} + +struct max8998_reg_dump { + u8 addr; + u8 val; +}; +#define SAVE_ITEM(x) { .addr = (x), .val = 0x0, } +struct max8998_reg_dump max8998_dump[] = { + SAVE_ITEM(MAX8998_REG_IRQM1), + SAVE_ITEM(MAX8998_REG_IRQM2), + SAVE_ITEM(MAX8998_REG_IRQM3), + SAVE_ITEM(MAX8998_REG_IRQM4), + SAVE_ITEM(MAX8998_REG_STATUSM1), + SAVE_ITEM(MAX8998_REG_STATUSM2), + SAVE_ITEM(MAX8998_REG_CHGR1), + SAVE_ITEM(MAX8998_REG_CHGR2), + SAVE_ITEM(MAX8998_REG_LDO_ACTIVE_DISCHARGE1), + SAVE_ITEM(MAX8998_REG_LDO_ACTIVE_DISCHARGE1), + SAVE_ITEM(MAX8998_REG_BUCK_ACTIVE_DISCHARGE3), + SAVE_ITEM(MAX8998_REG_ONOFF1), + SAVE_ITEM(MAX8998_REG_ONOFF2), + SAVE_ITEM(MAX8998_REG_ONOFF3), + SAVE_ITEM(MAX8998_REG_ONOFF4), + SAVE_ITEM(MAX8998_REG_BUCK1_VOLTAGE1), + SAVE_ITEM(MAX8998_REG_BUCK1_VOLTAGE2), + SAVE_ITEM(MAX8998_REG_BUCK1_VOLTAGE3), + SAVE_ITEM(MAX8998_REG_BUCK1_VOLTAGE4), + SAVE_ITEM(MAX8998_REG_BUCK2_VOLTAGE1), + SAVE_ITEM(MAX8998_REG_BUCK2_VOLTAGE2), + SAVE_ITEM(MAX8998_REG_LDO2_LDO3), + SAVE_ITEM(MAX8998_REG_LDO4), + SAVE_ITEM(MAX8998_REG_LDO5), + SAVE_ITEM(MAX8998_REG_LDO6), + SAVE_ITEM(MAX8998_REG_LDO7), + SAVE_ITEM(MAX8998_REG_LDO8_LDO9), + SAVE_ITEM(MAX8998_REG_LDO10_LDO11), + SAVE_ITEM(MAX8998_REG_LDO12), + SAVE_ITEM(MAX8998_REG_LDO13), + SAVE_ITEM(MAX8998_REG_LDO14), + SAVE_ITEM(MAX8998_REG_LDO15), + SAVE_ITEM(MAX8998_REG_LDO16), + SAVE_ITEM(MAX8998_REG_LDO17), + SAVE_ITEM(MAX8998_REG_BKCHR), + SAVE_ITEM(MAX8998_REG_LBCNFG1), + SAVE_ITEM(MAX8998_REG_LBCNFG2), +}; +/* Save registers before hibernation */ +static int max8998_freeze(struct device *dev) +{ + struct i2c_client *i2c = container_of(dev, struct i2c_client, dev); + int i; + + for (i = 0; i < ARRAY_SIZE(max8998_dump); i++) + max8998_read_reg(i2c, max8998_dump[i].addr, + &max8998_dump[i].val); + + return 0; +} + +/* Restore registers after hibernation */ +static int max8998_restore(struct device *dev) +{ + struct i2c_client *i2c = container_of(dev, struct i2c_client, dev); + int i; + + for (i = 0; i < ARRAY_SIZE(max8998_dump); i++) + max8998_write_reg(i2c, max8998_dump[i].addr, + max8998_dump[i].val); + + return 0; +} + +const struct dev_pm_ops max8998_pm = { + .suspend = max8998_suspend, + .resume = max8998_resume, + .freeze = max8998_freeze, + .restore = max8998_restore, +}; + static struct i2c_driver max8998_i2c_driver = { .driver = { .name = "max8998", .owner = THIS_MODULE, + .pm = &max8998_pm, }, .probe = max8998_i2c_probe, .remove = max8998_i2c_remove, diff --git a/include/linux/mfd/max8998-private.h b/include/linux/mfd/max8998-private.h index 7363dea6bbcd..effa5d3b96ae 100644 --- a/include/linux/mfd/max8998-private.h +++ b/include/linux/mfd/max8998-private.h @@ -159,10 +159,12 @@ struct max8998_dev { u8 irq_masks_cur[MAX8998_NUM_IRQ_REGS]; u8 irq_masks_cache[MAX8998_NUM_IRQ_REGS]; int type; + bool wakeup; }; int max8998_irq_init(struct max8998_dev *max8998); void max8998_irq_exit(struct max8998_dev *max8998); +int max8998_irq_resume(struct max8998_dev *max8998); extern int max8998_read_reg(struct i2c_client *i2c, u8 reg, u8 *dest); extern int max8998_bulk_read(struct i2c_client *i2c, u8 reg, int count, diff --git a/include/linux/mfd/max8998.h b/include/linux/mfd/max8998.h index f8c9f884aff2..686a744e63fb 100644 --- a/include/linux/mfd/max8998.h +++ b/include/linux/mfd/max8998.h @@ -88,6 +88,7 @@ struct max8998_platform_data { int buck1_set1; int buck1_set2; int buck2_set3; + bool wakeup; }; #endif /* __LINUX_MFD_MAX8998_H */ -- cgit v1.2.3 From 337ce5d1c5759644cea6c47220ce7e84f0398362 Mon Sep 17 00:00:00 2001 From: MyungJoo Ham Date: Tue, 4 Jan 2011 14:17:39 +0900 Subject: mfd: Support LP3974 RTC The first releases of LP3974 have a large delay in RTC registers, which requires 2 seconds of delay after writing to a rtc register (recommended by National Semiconductor's engineers) before reading it. If "rtc_delay" field of the platform data is true, the rtc driver assumes that such delays are required. Although we have not seen LP3974s without requiring such delays, we assume that such LP3974s will be released soon (or they have done so already) and they are supported by "lp3974" without setting "rtc_delay" at the platform data. This patch adds delays with msleep when writing values to RTC registers if the platform data has rtc_delay set. Signed-off-by: MyungJoo Ham Signed-off-by: Kyungmin Park Reviewed-by: Mark Brown Signed-off-by: Samuel Ortiz --- drivers/mfd/max8998.c | 26 +++++++++++++++++++--- drivers/regulator/max8998.c | 7 ++++++ drivers/rtc/rtc-max8998.c | 54 ++++++++++++++++++++++++++++++++++++++++----- include/linux/mfd/max8998.h | 4 ++++ 4 files changed, 83 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/drivers/mfd/max8998.c b/drivers/mfd/max8998.c index 5ce00ad5241a..bbfe86732602 100644 --- a/drivers/mfd/max8998.c +++ b/drivers/mfd/max8998.c @@ -42,6 +42,14 @@ static struct mfd_cell max8998_devs[] = { }, }; +static struct mfd_cell lp3974_devs[] = { + { + .name = "lp3974-pmic", + }, { + .name = "lp3974-rtc", + }, +}; + int max8998_read_reg(struct i2c_client *i2c, u8 reg, u8 *dest) { struct max8998_dev *max8998 = i2c_get_clientdata(i2c); @@ -146,11 +154,23 @@ static int max8998_i2c_probe(struct i2c_client *i2c, max8998_irq_init(max8998); - ret = mfd_add_devices(max8998->dev, -1, - max8998_devs, ARRAY_SIZE(max8998_devs), - NULL, 0); pm_runtime_set_active(max8998->dev); + switch (id->driver_data) { + case TYPE_LP3974: + ret = mfd_add_devices(max8998->dev, -1, + lp3974_devs, ARRAY_SIZE(lp3974_devs), + NULL, 0); + break; + case TYPE_MAX8998: + ret = mfd_add_devices(max8998->dev, -1, + max8998_devs, ARRAY_SIZE(max8998_devs), + NULL, 0); + break; + default: + ret = -EINVAL; + } + if (ret < 0) goto err; diff --git a/drivers/regulator/max8998.c b/drivers/regulator/max8998.c index 7568df6122ab..af52ebfc4927 100644 --- a/drivers/regulator/max8998.c +++ b/drivers/regulator/max8998.c @@ -835,6 +835,12 @@ static int __devexit max8998_pmic_remove(struct platform_device *pdev) return 0; } +static const struct platform_device_id max8998_pmic_id[] = { + { "max8998-pmic", TYPE_MAX8998 }, + { "lp3974-pmic", TYPE_LP3974 }, + { } +}; + static struct platform_driver max8998_pmic_driver = { .driver = { .name = "max8998-pmic", @@ -842,6 +848,7 @@ static struct platform_driver max8998_pmic_driver = { }, .probe = max8998_pmic_probe, .remove = __devexit_p(max8998_pmic_remove), + .id_table = max8998_pmic_id, }; static int __init max8998_pmic_init(void) diff --git a/drivers/rtc/rtc-max8998.c b/drivers/rtc/rtc-max8998.c index f22dee35f330..3f7bc6b9fefa 100644 --- a/drivers/rtc/rtc-max8998.c +++ b/drivers/rtc/rtc-max8998.c @@ -20,6 +20,7 @@ #include #include #include +#include #define MAX8998_RTC_SEC 0x00 #define MAX8998_RTC_MIN 0x01 @@ -73,6 +74,7 @@ struct max8998_rtc_info { struct i2c_client *rtc; struct rtc_device *rtc_dev; int irq; + bool lp3974_bug_workaround; }; static void max8998_data_to_tm(u8 *data, struct rtc_time *tm) @@ -124,10 +126,16 @@ static int max8998_rtc_set_time(struct device *dev, struct rtc_time *tm) { struct max8998_rtc_info *info = dev_get_drvdata(dev); u8 data[8]; + int ret; max8998_tm_to_data(tm, data); - return max8998_bulk_write(info->rtc, MAX8998_RTC_SEC, 8, data); + ret = max8998_bulk_write(info->rtc, MAX8998_RTC_SEC, 8, data); + + if (info->lp3974_bug_workaround) + msleep(2000); + + return ret; } static int max8998_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alrm) @@ -163,12 +171,29 @@ static int max8998_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alrm) static int max8998_rtc_stop_alarm(struct max8998_rtc_info *info) { - return max8998_write_reg(info->rtc, MAX8998_ALARM0_CONF, 0); + int ret = max8998_write_reg(info->rtc, MAX8998_ALARM0_CONF, 0); + + if (info->lp3974_bug_workaround) + msleep(2000); + + return ret; } static int max8998_rtc_start_alarm(struct max8998_rtc_info *info) { - return max8998_write_reg(info->rtc, MAX8998_ALARM0_CONF, 0x77); + int ret; + u8 alarm0_conf = 0x77; + + /* LP3974 with delay bug chips has rtc alarm bugs with "MONTH" field */ + if (info->lp3974_bug_workaround) + alarm0_conf = 0x57; + + ret = max8998_write_reg(info->rtc, MAX8998_ALARM0_CONF, alarm0_conf); + + if (info->lp3974_bug_workaround) + msleep(2000); + + return ret; } static int max8998_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alrm) @@ -187,10 +212,13 @@ static int max8998_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alrm) if (ret < 0) return ret; + if (info->lp3974_bug_workaround) + msleep(2000); + if (alrm->enabled) - return max8998_rtc_start_alarm(info); + ret = max8998_rtc_start_alarm(info); - return 0; + return ret; } static int max8998_rtc_alarm_irq_enable(struct device *dev, @@ -224,6 +252,7 @@ static const struct rtc_class_ops max8998_rtc_ops = { static int __devinit max8998_rtc_probe(struct platform_device *pdev) { struct max8998_dev *max8998 = dev_get_drvdata(pdev->dev.parent); + struct max8998_platform_data *pdata = dev_get_platdata(max8998->dev); struct max8998_rtc_info *info; int ret; @@ -249,10 +278,18 @@ static int __devinit max8998_rtc_probe(struct platform_device *pdev) ret = request_threaded_irq(info->irq, NULL, max8998_rtc_alarm_irq, 0, "rtc-alarm0", info); + if (ret < 0) dev_err(&pdev->dev, "Failed to request alarm IRQ: %d: %d\n", info->irq, ret); + dev_info(&pdev->dev, "RTC CHIP NAME: %s\n", pdev->id_entry->name); + if (pdata->rtc_delay) { + info->lp3974_bug_workaround = true; + dev_warn(&pdev->dev, "LP3974 with RTC REGERR option." + " RTC updates will be extremely slow.\n"); + } + return 0; out_rtc: @@ -273,6 +310,12 @@ static int __devexit max8998_rtc_remove(struct platform_device *pdev) return 0; } +static const struct platform_device_id max8998_rtc_id[] = { + { "max8998-rtc", TYPE_MAX8998 }, + { "lp3974-rtc", TYPE_LP3974 }, + { } +}; + static struct platform_driver max8998_rtc_driver = { .driver = { .name = "max8998-rtc", @@ -280,6 +323,7 @@ static struct platform_driver max8998_rtc_driver = { }, .probe = max8998_rtc_probe, .remove = __devexit_p(max8998_rtc_remove), + .id_table = max8998_rtc_id, }; static int __init max8998_rtc_init(void) diff --git a/include/linux/mfd/max8998.h b/include/linux/mfd/max8998.h index 686a744e63fb..c22b9a417c35 100644 --- a/include/linux/mfd/max8998.h +++ b/include/linux/mfd/max8998.h @@ -76,6 +76,9 @@ struct max8998_regulator_data { * @buck1_set1: BUCK1 gpio pin 1 to set output voltage * @buck1_set2: BUCK1 gpio pin 2 to set output voltage * @buck2_set3: BUCK2 gpio pin to set output voltage + * @wakeup: Allow to wake up from suspend + * @rtc_delay: LP3974 RTC chip bug that requires delay after a register + * write before reading it. */ struct max8998_platform_data { struct max8998_regulator_data *regulators; @@ -89,6 +92,7 @@ struct max8998_platform_data { int buck1_set2; int buck2_set3; bool wakeup; + bool rtc_delay; }; #endif /* __LINUX_MFD_MAX8998_H */ -- cgit v1.2.3 From 735a3d9efdc5aeebe201008e6655b235c7f02aeb Mon Sep 17 00:00:00 2001 From: MyungJoo Ham Date: Tue, 11 Jan 2011 12:20:05 +0100 Subject: regulator: Support MAX8998/LP3974 DVS-GPIO The previous driver did not support BUCK1-DVS3, BUCK1-DVS4, and BUCK2-DVS2 modes. This patch adds such modes and an option to block setting buck1/2 voltages out of the preset values. Signed-off-by: MyungJoo Ham Signed-off-by: Kyungmin Park Acked-by: Mark Brown Signed-off-by: Samuel Ortiz --- drivers/regulator/max8998.c | 87 ++++++++++++++++++++++++++++++++++----------- include/linux/mfd/max8998.h | 26 ++++++++++---- 2 files changed, 87 insertions(+), 26 deletions(-) (limited to 'include/linux') diff --git a/drivers/regulator/max8998.c b/drivers/regulator/max8998.c index af52ebfc4927..0ec49ca527a8 100644 --- a/drivers/regulator/max8998.c +++ b/drivers/regulator/max8998.c @@ -424,6 +424,9 @@ static int max8998_set_voltage_buck(struct regulator_dev *rdev, } } + if (pdata->buck_voltage_lock) + return -EINVAL; + /* no predefine regulator found */ max8998->buck1_idx = (buck1_last_val % 2) + 2; dev_dbg(max8998->dev, "max8998->buck1_idx:%d\n", @@ -451,18 +454,26 @@ buck1_exit: "BUCK2, i:%d buck2_vol1:%d, buck2_vol2:%d\n" , i, max8998->buck2_vol[0], max8998->buck2_vol[1]); if (gpio_is_valid(pdata->buck2_set3)) { - if (max8998->buck2_vol[0] == i) { - max8998->buck1_idx = 0; - buck2_gpio_set(pdata->buck2_set3, 0); - } else { - max8998->buck1_idx = 1; - ret = max8998_get_voltage_register(rdev, ®, - &shift, - &mask); - ret = max8998_write_reg(i2c, reg, i); - max8998->buck2_vol[1] = i; - buck2_gpio_set(pdata->buck2_set3, 1); + + /* check if requested voltage */ + /* value is already defined */ + for (j = 0; j < ARRAY_SIZE(max8998->buck2_vol); j++) { + if (max8998->buck2_vol[j] == i) { + max8998->buck2_idx = j; + buck2_gpio_set(pdata->buck2_set3, j); + goto buck2_exit; + } } + + if (pdata->buck_voltage_lock) + return -EINVAL; + + max8998_get_voltage_register(rdev, + ®, &shift, &mask); + ret = max8998_write_reg(i2c, reg, i); + max8998->buck2_vol[max8998->buck2_idx] = i; + buck2_gpio_set(pdata->buck2_set3, max8998->buck2_idx); +buck2_exit: dev_dbg(max8998->dev, "%s: SET3:%d\n", i2c->name, gpio_get_value(pdata->buck2_set3)); } else { @@ -707,6 +718,9 @@ static __devinit int max8998_pmic_probe(struct platform_device *pdev) platform_set_drvdata(pdev, max8998); i2c = max8998->iodev->i2c; + max8998->buck1_idx = pdata->buck1_default_idx; + max8998->buck2_idx = pdata->buck2_default_idx; + /* NOTE: */ /* For unused GPIO NOT marked as -1 (thereof equal to 0) WARN_ON */ /* will be displayed */ @@ -739,23 +753,46 @@ static __devinit int max8998_pmic_probe(struct platform_device *pdev) i = 0; while (buck12_voltage_map_desc.min + buck12_voltage_map_desc.step*i - != (pdata->buck1_max_voltage1 / 1000)) + < (pdata->buck1_voltage1 / 1000)) i++; - printk(KERN_ERR "i:%d, buck1_idx:%d\n", i, max8998->buck1_idx); max8998->buck1_vol[0] = i; ret = max8998_write_reg(i2c, MAX8998_REG_BUCK1_VOLTAGE1, i); + if (ret) + return ret; /* Set predefined value for BUCK1 register 2 */ i = 0; while (buck12_voltage_map_desc.min + buck12_voltage_map_desc.step*i - != (pdata->buck1_max_voltage2 / 1000)) + < (pdata->buck1_voltage2 / 1000)) i++; max8998->buck1_vol[1] = i; - printk(KERN_ERR "i:%d, buck1_idx:%d\n", i, max8998->buck1_idx); - ret = max8998_write_reg(i2c, MAX8998_REG_BUCK1_VOLTAGE2, i) - + ret; + ret = max8998_write_reg(i2c, MAX8998_REG_BUCK1_VOLTAGE2, i); + if (ret) + return ret; + + /* Set predefined value for BUCK1 register 3 */ + i = 0; + while (buck12_voltage_map_desc.min + + buck12_voltage_map_desc.step*i + < (pdata->buck1_voltage3 / 1000)) + i++; + + max8998->buck1_vol[2] = i; + ret = max8998_write_reg(i2c, MAX8998_REG_BUCK1_VOLTAGE3, i); + if (ret) + return ret; + + /* Set predefined value for BUCK1 register 4 */ + i = 0; + while (buck12_voltage_map_desc.min + + buck12_voltage_map_desc.step*i + < (pdata->buck1_voltage4 / 1000)) + i++; + + max8998->buck1_vol[3] = i; + ret = max8998_write_reg(i2c, MAX8998_REG_BUCK1_VOLTAGE4, i); if (ret) return ret; @@ -772,18 +809,28 @@ static __devinit int max8998_pmic_probe(struct platform_device *pdev) gpio_direction_output(pdata->buck2_set3, max8998->buck2_idx & 0x1); - /* BUCK2 - set preset default voltage value to buck2_vol[0] */ + /* BUCK2 register 1 */ i = 0; while (buck12_voltage_map_desc.min + buck12_voltage_map_desc.step*i - != (pdata->buck2_max_voltage / 1000)) + < (pdata->buck2_voltage1 / 1000)) i++; - printk(KERN_ERR "i:%d, buck2_idx:%d\n", i, max8998->buck2_idx); max8998->buck2_vol[0] = i; ret = max8998_write_reg(i2c, MAX8998_REG_BUCK2_VOLTAGE1, i); if (ret) return ret; + /* BUCK2 register 2 */ + i = 0; + while (buck12_voltage_map_desc.min + + buck12_voltage_map_desc.step*i + < (pdata->buck2_voltage2 / 1000)) + i++; + printk(KERN_ERR "i2:%d, buck2_idx:%d\n", i, max8998->buck2_idx); + max8998->buck2_vol[1] = i; + ret = max8998_write_reg(i2c, MAX8998_REG_BUCK2_VOLTAGE2, i); + if (ret) + return ret; } for (i = 0; i < pdata->num_regulators; i++) { diff --git a/include/linux/mfd/max8998.h b/include/linux/mfd/max8998.h index c22b9a417c35..61daa167b576 100644 --- a/include/linux/mfd/max8998.h +++ b/include/linux/mfd/max8998.h @@ -70,12 +70,20 @@ struct max8998_regulator_data { * @num_regulators: number of regultors used * @irq_base: base IRQ number for max8998, required for IRQs * @ono: power onoff IRQ number for max8998 - * @buck1_max_voltage1: BUCK1 maximum alowed voltage register 1 - * @buck1_max_voltage2: BUCK1 maximum alowed voltage register 2 - * @buck2_max_voltage: BUCK2 maximum alowed voltage + * @buck_voltage_lock: Do NOT change the values of the following six + * registers set by buck?_voltage?. The voltage of BUCK1/2 cannot + * be other than the preset values. + * @buck1_voltage1: BUCK1 DVS mode 1 voltage register + * @buck1_voltage2: BUCK1 DVS mode 2 voltage register + * @buck1_voltage3: BUCK1 DVS mode 3 voltage register + * @buck1_voltage4: BUCK1 DVS mode 4 voltage register + * @buck2_voltage1: BUCK2 DVS mode 1 voltage register + * @buck2_voltage2: BUCK2 DVS mode 2 voltage register * @buck1_set1: BUCK1 gpio pin 1 to set output voltage * @buck1_set2: BUCK1 gpio pin 2 to set output voltage + * @buck1_default_idx: Default for BUCK1 gpio pin 1, 2 * @buck2_set3: BUCK2 gpio pin to set output voltage + * @buck2_default_idx: Default for BUCK2 gpio pin. * @wakeup: Allow to wake up from suspend * @rtc_delay: LP3974 RTC chip bug that requires delay after a register * write before reading it. @@ -85,12 +93,18 @@ struct max8998_platform_data { int num_regulators; int irq_base; int ono; - int buck1_max_voltage1; - int buck1_max_voltage2; - int buck2_max_voltage; + bool buck_voltage_lock; + int buck1_voltage1; + int buck1_voltage2; + int buck1_voltage3; + int buck1_voltage4; + int buck2_voltage1; + int buck2_voltage2; int buck1_set1; int buck1_set2; + int buck1_default_idx; int buck2_set3; + int buck2_default_idx; bool wakeup; bool rtc_delay; }; -- cgit v1.2.3 From 92d50a4132977b932ed830fa58c05deeb5c524f0 Mon Sep 17 00:00:00 2001 From: Mattias Wallin Date: Tue, 7 Dec 2010 11:20:47 +0100 Subject: mfd: ab8500-core chip version cut 2.0 support This patch adds support for chip version 2.0 or cut 2.0. One new interrupt latch register - latch 12 - is introduced. Signed-off-by: Mattias Wallin Acked-by: Linus Walleij Signed-off-by: Samuel Ortiz --- drivers/mfd/ab8500-core.c | 36 +++++++++++++++++++++---------- include/linux/mfd/ab8500.h | 53 ++++++++++++++++++++++++++-------------------- 2 files changed, 55 insertions(+), 34 deletions(-) (limited to 'include/linux') diff --git a/drivers/mfd/ab8500-core.c b/drivers/mfd/ab8500-core.c index 86cc8874e6de..b6887014d687 100644 --- a/drivers/mfd/ab8500-core.c +++ b/drivers/mfd/ab8500-core.c @@ -52,6 +52,7 @@ #define AB8500_IT_LATCH8_REG 0x27 #define AB8500_IT_LATCH9_REG 0x28 #define AB8500_IT_LATCH10_REG 0x29 +#define AB8500_IT_LATCH12_REG 0x2B #define AB8500_IT_LATCH19_REG 0x32 #define AB8500_IT_LATCH20_REG 0x33 #define AB8500_IT_LATCH21_REG 0x34 @@ -98,7 +99,7 @@ * offset 0. */ static const int ab8500_irq_regoffset[AB8500_NUM_IRQ_REGS] = { - 0, 1, 2, 3, 4, 6, 7, 8, 9, 18, 19, 20, 21, + 0, 1, 2, 3, 4, 6, 7, 8, 9, 11, 18, 19, 20, 21, }; static int ab8500_get_chip_id(struct device *dev) @@ -252,6 +253,10 @@ static void ab8500_irq_sync_unlock(struct irq_data *data) if (new == old) continue; + /* Interrupt register 12 does'nt exist prior to version 0x20 */ + if (ab8500_irq_regoffset[i] == 11 && ab8500->chip_id < 0x20) + continue; + ab8500->oldmask[i] = new; reg = AB8500_IT_MASK1_REG + ab8500_irq_regoffset[i]; @@ -301,6 +306,10 @@ static irqreturn_t ab8500_irq(int irq, void *dev) int status; u8 value; + /* Interrupt register 12 does'nt exist prior to version 0x20 */ + if (regoffset == 11 && ab8500->chip_id < 0x20) + continue; + status = get_register_interruptible(ab8500, AB8500_INTERRUPT, AB8500_IT_LATCH1_REG + regoffset, &value); if (status < 0 || value == 0) @@ -554,6 +563,12 @@ static struct resource ab8500_usb_resources[] = { .end = AB8500_INT_VBUS_DET_R, .flags = IORESOURCE_IRQ, }, + { + .name = "USB_LINK_STATUS", + .start = AB8500_INT_USB_LINK_STATUS, + .end = AB8500_INT_USB_LINK_STATUS, + .flags = IORESOURCE_IRQ, + }, }; static struct resource ab8500_temp_resources[] = { @@ -670,8 +685,9 @@ int __devinit ab8500_init(struct ab8500 *ab8500) * 0x0 - Early Drop * 0x10 - Cut 1.0 * 0x11 - Cut 1.1 + * 0x20 - Cut 2.0 */ - if (value == 0x0 || value == 0x10 || value == 0x11) { + if (value == 0x0 || value == 0x10 || value == 0x11 || value == 0x20) { ab8500->revision = value; dev_info(ab8500->dev, "detected chip, revision: %#x\n", value); } else { @@ -684,18 +700,16 @@ int __devinit ab8500_init(struct ab8500 *ab8500) plat->init(ab8500); /* Clear and mask all interrupts */ - for (i = 0; i < 10; i++) { - get_register_interruptible(ab8500, AB8500_INTERRUPT, - AB8500_IT_LATCH1_REG + i, &value); - set_register_interruptible(ab8500, AB8500_INTERRUPT, - AB8500_IT_MASK1_REG + i, 0xff); - } + for (i = 0; i < AB8500_NUM_IRQ_REGS; i++) { + /* Interrupt register 12 does'nt exist prior to version 0x20 */ + if (ab8500_irq_regoffset[i] == 11 && ab8500->chip_id < 0x20) + continue; - for (i = 18; i < 24; i++) { get_register_interruptible(ab8500, AB8500_INTERRUPT, - AB8500_IT_LATCH1_REG + i, &value); + AB8500_IT_LATCH1_REG + ab8500_irq_regoffset[i], + &value); set_register_interruptible(ab8500, AB8500_INTERRUPT, - AB8500_IT_MASK1_REG + i, 0xff); + AB8500_IT_MASK1_REG + ab8500_irq_regoffset[i], 0xff); } ret = abx500_register_ops(ab8500->dev, &ab8500_ops); diff --git a/include/linux/mfd/ab8500.h b/include/linux/mfd/ab8500.h index 1ad169655235..37f56b7c4c15 100644 --- a/include/linux/mfd/ab8500.h +++ b/include/linux/mfd/ab8500.h @@ -74,30 +74,37 @@ #define AB8500_INT_ACC_DETECT_21DB_F 37 #define AB8500_INT_ACC_DETECT_21DB_R 38 #define AB8500_INT_GP_SW_ADC_CONV_END 39 -#define AB8500_INT_BTEMP_LOW 72 -#define AB8500_INT_BTEMP_LOW_MEDIUM 73 -#define AB8500_INT_BTEMP_MEDIUM_HIGH 74 -#define AB8500_INT_BTEMP_HIGH 75 -#define AB8500_INT_USB_CHARGER_NOT_OK 81 -#define AB8500_INT_ID_WAKEUP_R 82 -#define AB8500_INT_ID_DET_R1R 84 -#define AB8500_INT_ID_DET_R2R 85 -#define AB8500_INT_ID_DET_R3R 86 -#define AB8500_INT_ID_DET_R4R 87 -#define AB8500_INT_ID_WAKEUP_F 88 -#define AB8500_INT_ID_DET_R1F 90 -#define AB8500_INT_ID_DET_R2F 91 -#define AB8500_INT_ID_DET_R3F 92 -#define AB8500_INT_ID_DET_R4F 93 -#define AB8500_INT_USB_CHG_DET_DONE 94 -#define AB8500_INT_USB_CH_TH_PROT_F 96 -#define AB8500_INT_USB_CH_TH_PROT_R 97 -#define AB8500_INT_MAIN_CH_TH_PROT_F 98 -#define AB8500_INT_MAIN_CH_TH_PROT_R 99 -#define AB8500_INT_USB_CHARGER_NOT_OKF 103 +#define AB8500_INT_ADP_SOURCE_ERROR 72 +#define AB8500_INT_ADP_SINK_ERROR 73 +#define AB8500_INT_ADP_PROBE_PLUG 74 +#define AB8500_INT_ADP_PROBE_UNPLUG 75 +#define AB8500_INT_ADP_SENSE_OFF 76 +#define AB8500_INT_USB_PHY_POWER_ERR 78 +#define AB8500_INT_USB_LINK_STATUS 79 +#define AB8500_INT_BTEMP_LOW 80 +#define AB8500_INT_BTEMP_LOW_MEDIUM 81 +#define AB8500_INT_BTEMP_MEDIUM_HIGH 82 +#define AB8500_INT_BTEMP_HIGH 83 +#define AB8500_INT_USB_CHARGER_NOT_OK 89 +#define AB8500_INT_ID_WAKEUP_R 90 +#define AB8500_INT_ID_DET_R1R 92 +#define AB8500_INT_ID_DET_R2R 93 +#define AB8500_INT_ID_DET_R3R 94 +#define AB8500_INT_ID_DET_R4R 95 +#define AB8500_INT_ID_WAKEUP_F 96 +#define AB8500_INT_ID_DET_R1F 98 +#define AB8500_INT_ID_DET_R2F 99 +#define AB8500_INT_ID_DET_R3F 100 +#define AB8500_INT_ID_DET_R4F 101 +#define AB8500_INT_USB_CHG_DET_DONE 102 +#define AB8500_INT_USB_CH_TH_PROT_F 104 +#define AB8500_INT_USB_CH_TH_PROT_R 105 +#define AB8500_INT_MAIN_CH_TH_PROT_F 106 +#define AB8500_INT_MAIN_CH_TH_PROT_R 107 +#define AB8500_INT_USB_CHARGER_NOT_OKF 111 -#define AB8500_NR_IRQS 104 -#define AB8500_NUM_IRQ_REGS 13 +#define AB8500_NR_IRQS 112 +#define AB8500_NUM_IRQ_REGS 14 /** * struct ab8500 - ab8500 internal structure -- cgit v1.2.3 From 32385c7cf60a78375b63afc4f02001df84dfd1a0 Mon Sep 17 00:00:00 2001 From: Russell King Date: Fri, 14 Jan 2011 13:12:45 +0000 Subject: kernel: fix hlist_bl again __d_rehash is dereferencing an almost-NULL pointer on my ARM926. CONFIG_SMP=n and CONFIG_DEBUG_SPINLOCK=y. The faulting instruction is: strne r3, [r2, #4] and as can be seen from the register dump below, r2 is 0x00000001, hence the faulting 0x00000005 address. __d_rehash is essentially: spin_lock_bucket(b); entry->d_flags &= ~DCACHE_UNHASHED; hlist_bl_add_head_rcu(&entry->d_hash, &b->head); spin_unlock_bucket(b); which is: bit_spin_lock(0, (unsigned long *)&b->head.first); entry->d_flags &= ~DCACHE_UNHASHED; hlist_bl_add_head_rcu(&entry->d_hash, &b->head); __bit_spin_unlock(0, (unsigned long *)&b->head.first); bit_spin_lock(0, ptr) sets bit 0 of *ptr, in this case b->head.first if CONFIG_SMP or CONFIG_DEBUG_SPINLOCK is set: #if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) while (unlikely(test_and_set_bit_lock(bitnum, addr))) { while (test_bit(bitnum, addr)) { preempt_enable(); cpu_relax(); preempt_disable(); } } #endif So, b->head.first starts off NULL, and becomes a non-NULL (address 1). hlist_bl_add_head_rcu() does this: static inline void hlist_bl_add_head_rcu(struct hlist_bl_node *n, struct hlist_bl_head *h) { first = hlist_bl_first(h); n->next = first; if (first) first->pprev = &n->next; It is the store to first->pprev which is faulting. hlist_bl_first(): static inline struct hlist_bl_node *hlist_bl_first(struct hlist_bl_head *h) { return (struct hlist_bl_node *) ((unsigned long)h->first & ~LIST_BL_LOCKMASK); } but: #if defined(CONFIG_SMP) #define LIST_BL_LOCKMASK 1UL #else #define LIST_BL_LOCKMASK 0UL #endif So, we have one piece of code which sets bit 0 of addresses, and another bit of code which doesn't clear it before dereferencing the pointer if !CONFIG_SMP && CONFIG_DEBUG_SPINLOCK. With the patch below, I can again sucessfully boot the kernel on my Versatile PB/926 platform. Signed-off-by: Russell King --- include/linux/list_bl.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/list_bl.h b/include/linux/list_bl.h index b2adbb4b2f73..5bad17d1acde 100644 --- a/include/linux/list_bl.h +++ b/include/linux/list_bl.h @@ -16,7 +16,7 @@ * some fast and compact auxiliary data. */ -#if defined(CONFIG_SMP) +#if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK) #define LIST_BL_LOCKMASK 1UL #else #define LIST_BL_LOCKMASK 0UL -- cgit v1.2.3 From 359ab9f5b154cbd807a11e22792235f0f36b0cd5 Mon Sep 17 00:00:00 2001 From: MyungJoo Ham Date: Fri, 14 Jan 2011 14:46:11 +0900 Subject: power_supply: Add MAX17042 Fuel Gauge Driver The MAX17042 is a fuel gauge with an I2C interface for lithium-ion betteries. Unlike its predecessor MAX17040, MAX17042 uses 16bit registers. Besides, MAX17042 has much more features than MAX17040; e.g., a thermistor, current and current accumulation measurement, battery internal resistance estimate, average values of measurement, and others. This patch implements a driver for MAX17042. In this initial release, we have implemented the most basic features of a fuel gauge: measure the battery capacity and voltage. Signed-off-by: MyungJoo Ham Signed-off-by: Kyungmin Park Signed-off-by: Anton Vorontsov --- drivers/power/Kconfig | 10 ++ drivers/power/Makefile | 1 + drivers/power/max17042_battery.c | 239 +++++++++++++++++++++++++++++++++ include/linux/power/max17042_battery.h | 30 +++++ 4 files changed, 280 insertions(+) create mode 100644 drivers/power/max17042_battery.c create mode 100644 include/linux/power/max17042_battery.h (limited to 'include/linux') diff --git a/drivers/power/Kconfig b/drivers/power/Kconfig index 32165295cb1b..61bf5d724139 100644 --- a/drivers/power/Kconfig +++ b/drivers/power/Kconfig @@ -136,6 +136,16 @@ config BATTERY_MAX17040 in handheld and portable equipment. The MAX17040 is configured to operate with a single lithium cell +config BATTERY_MAX17042 + tristate "Maxim MAX17042/8997/8966 Fuel Gauge" + depends on I2C + help + MAX17042 is fuel-gauge systems for lithium-ion (Li+) batteries + in handheld and portable equipment. The MAX17042 is configured + to operate with a single lithium cell. MAX8997 and MAX8966 are + multi-function devices that include fuel gauages that are compatible + with MAX17042. + config BATTERY_Z2 tristate "Z2 battery driver" depends on I2C && MACH_ZIPIT2 diff --git a/drivers/power/Makefile b/drivers/power/Makefile index 545459f9f356..8385bfae8728 100644 --- a/drivers/power/Makefile +++ b/drivers/power/Makefile @@ -25,6 +25,7 @@ obj-$(CONFIG_BATTERY_BQ20Z75) += bq20z75.o obj-$(CONFIG_BATTERY_BQ27x00) += bq27x00_battery.o obj-$(CONFIG_BATTERY_DA9030) += da9030_battery.o obj-$(CONFIG_BATTERY_MAX17040) += max17040_battery.o +obj-$(CONFIG_BATTERY_MAX17042) += max17042_battery.o obj-$(CONFIG_BATTERY_Z2) += z2_battery.o obj-$(CONFIG_BATTERY_S3C_ADC) += s3c_adc_battery.o obj-$(CONFIG_CHARGER_PCF50633) += pcf50633-charger.o diff --git a/drivers/power/max17042_battery.c b/drivers/power/max17042_battery.c new file mode 100644 index 000000000000..c5c8805156cb --- /dev/null +++ b/drivers/power/max17042_battery.c @@ -0,0 +1,239 @@ +/* + * Fuel gauge driver for Maxim 17042 / 8966 / 8997 + * Note that Maxim 8966 and 8997 are mfd and this is its subdevice. + * + * Copyright (C) 2011 Samsung Electronics + * MyungJoo Ham + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * This driver is based on max17040_battery.c + */ + +#include +#include +#include +#include +#include +#include + +enum max17042_register { + MAX17042_STATUS = 0x00, + MAX17042_VALRT_Th = 0x01, + MAX17042_TALRT_Th = 0x02, + MAX17042_SALRT_Th = 0x03, + MAX17042_AtRate = 0x04, + MAX17042_RepCap = 0x05, + MAX17042_RepSOC = 0x06, + MAX17042_Age = 0x07, + MAX17042_TEMP = 0x08, + MAX17042_VCELL = 0x09, + MAX17042_Current = 0x0A, + MAX17042_AvgCurrent = 0x0B, + MAX17042_Qresidual = 0x0C, + MAX17042_SOC = 0x0D, + MAX17042_AvSOC = 0x0E, + MAX17042_RemCap = 0x0F, + MAX17402_FullCAP = 0x10, + MAX17042_TTE = 0x11, + MAX17042_V_empty = 0x12, + + MAX17042_RSLOW = 0x14, + + MAX17042_AvgTA = 0x16, + MAX17042_Cycles = 0x17, + MAX17042_DesignCap = 0x18, + MAX17042_AvgVCELL = 0x19, + MAX17042_MinMaxTemp = 0x1A, + MAX17042_MinMaxVolt = 0x1B, + MAX17042_MinMaxCurr = 0x1C, + MAX17042_CONFIG = 0x1D, + MAX17042_ICHGTerm = 0x1E, + MAX17042_AvCap = 0x1F, + MAX17042_ManName = 0x20, + MAX17042_DevName = 0x21, + MAX17042_DevChem = 0x22, + + MAX17042_TempNom = 0x24, + MAX17042_TempCold = 0x25, + MAX17042_TempHot = 0x26, + MAX17042_AIN = 0x27, + MAX17042_LearnCFG = 0x28, + MAX17042_SHFTCFG = 0x29, + MAX17042_RelaxCFG = 0x2A, + MAX17042_MiscCFG = 0x2B, + MAX17042_TGAIN = 0x2C, + MAx17042_TOFF = 0x2D, + MAX17042_CGAIN = 0x2E, + MAX17042_COFF = 0x2F, + + MAX17042_Q_empty = 0x33, + MAX17042_T_empty = 0x34, + + MAX17042_RCOMP0 = 0x38, + MAX17042_TempCo = 0x39, + MAX17042_Rx = 0x3A, + MAX17042_T_empty0 = 0x3B, + MAX17042_TaskPeriod = 0x3C, + MAX17042_FSTAT = 0x3D, + + MAX17042_SHDNTIMER = 0x3F, + + MAX17042_VFRemCap = 0x4A, + + MAX17042_QH = 0x4D, + MAX17042_QL = 0x4E, +}; + +struct max17042_chip { + struct i2c_client *client; + struct power_supply battery; + struct max17042_platform_data *pdata; +}; + +static int max17042_write_reg(struct i2c_client *client, u8 reg, u16 value) +{ + int ret = i2c_smbus_write_word_data(client, reg, value); + + if (ret < 0) + dev_err(&client->dev, "%s: err %d\n", __func__, ret); + + return ret; +} + +static int max17042_read_reg(struct i2c_client *client, u8 reg) +{ + int ret = i2c_smbus_read_word_data(client, reg); + + if (ret < 0) + dev_err(&client->dev, "%s: err %d\n", __func__, ret); + + return ret; +} + +static enum power_supply_property max17042_battery_props[] = { + POWER_SUPPLY_PROP_VOLTAGE_NOW, + POWER_SUPPLY_PROP_VOLTAGE_AVG, + POWER_SUPPLY_PROP_CAPACITY, +}; + +static int max17042_get_property(struct power_supply *psy, + enum power_supply_property psp, + union power_supply_propval *val) +{ + struct max17042_chip *chip = container_of(psy, + struct max17042_chip, battery); + + switch (psp) { + case POWER_SUPPLY_PROP_VOLTAGE_NOW: + val->intval = max17042_read_reg(chip->client, + MAX17042_VCELL) * 83; /* 1000 / 12 = 83 */ + break; + case POWER_SUPPLY_PROP_VOLTAGE_AVG: + val->intval = max17042_read_reg(chip->client, + MAX17042_AvgVCELL) * 83; + break; + case POWER_SUPPLY_PROP_CAPACITY: + val->intval = max17042_read_reg(chip->client, + MAX17042_SOC) / 256; + break; + default: + return -EINVAL; + } + return 0; +} + +static int __devinit max17042_probe(struct i2c_client *client, + const struct i2c_device_id *id) +{ + struct i2c_adapter *adapter = to_i2c_adapter(client->dev.parent); + struct max17042_chip *chip; + int ret; + + if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_WORD_DATA)) + return -EIO; + + chip = kzalloc(sizeof(*chip), GFP_KERNEL); + if (!chip) + return -ENOMEM; + + chip->client = client; + chip->pdata = client->dev.platform_data; + + i2c_set_clientdata(client, chip); + + chip->battery.name = "max17042_battery"; + chip->battery.type = POWER_SUPPLY_TYPE_BATTERY; + chip->battery.get_property = max17042_get_property; + chip->battery.properties = max17042_battery_props; + chip->battery.num_properties = ARRAY_SIZE(max17042_battery_props); + + ret = power_supply_register(&client->dev, &chip->battery); + if (ret) { + dev_err(&client->dev, "failed: power supply register\n"); + i2c_set_clientdata(client, NULL); + kfree(chip); + return ret; + } + + if (!chip->pdata->enable_current_sense) { + max17042_write_reg(client, MAX17042_CGAIN, 0x0000); + max17042_write_reg(client, MAX17042_MiscCFG, 0x0003); + max17042_write_reg(client, MAX17042_LearnCFG, 0x0007); + } + + return 0; +} + +static int __devexit max17042_remove(struct i2c_client *client) +{ + struct max17042_chip *chip = i2c_get_clientdata(client); + + power_supply_unregister(&chip->battery); + i2c_set_clientdata(client, NULL); + kfree(chip); + return 0; +} + +static const struct i2c_device_id max17042_id[] = { + { "max17042", 0 }, + { } +}; +MODULE_DEVICE_TABLE(i2c, max17042_id); + +static struct i2c_driver max17042_i2c_driver = { + .driver = { + .name = "max17042", + }, + .probe = max17042_probe, + .remove = __devexit_p(max17042_remove), + .id_table = max17042_id, +}; + +static int __init max17042_init(void) +{ + return i2c_add_driver(&max17042_i2c_driver); +} +module_init(max17042_init); + +static void __exit max17042_exit(void) +{ + i2c_del_driver(&max17042_i2c_driver); +} +module_exit(max17042_exit); + +MODULE_AUTHOR("MyungJoo Ham "); +MODULE_DESCRIPTION("MAX17042 Fuel Gauge"); +MODULE_LICENSE("GPL"); diff --git a/include/linux/power/max17042_battery.h b/include/linux/power/max17042_battery.h new file mode 100644 index 000000000000..7995deb8bfc1 --- /dev/null +++ b/include/linux/power/max17042_battery.h @@ -0,0 +1,30 @@ +/* + * Fuel gauge driver for Maxim 17042 / 8966 / 8997 + * Note that Maxim 8966 and 8997 are mfd and this is its subdevice. + * + * Copyright (C) 2011 Samsung Electronics + * MyungJoo Ham + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#ifndef __MAX17042_BATTERY_H_ +#define __MAX17042_BATTERY_H_ + +struct max17042_platform_data { + bool enable_current_sense; +}; + +#endif /* __MAX17042_BATTERY_H_ */ -- cgit v1.2.3 From 836cb711ad7960e52625b24195d6e70b79ab0816 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Fri, 14 Jan 2011 14:56:53 +0900 Subject: Revert update for dirty_ratio for memcg. The flags added by commit db16d5ec1f87f17511599bc77857dd1662b5a22f has no user now. We believe we'll use it soon but considering patch reviewing, the change itself should be folded into incoming set of "dirty ratio for memcg" patches. So, it's better to drop this change from current mainline tree. Signed-off-by: KAMEZAWA Hiroyuki Reviewed-by: Greg Thelen Signed-off-by: Linus Torvalds --- include/linux/page_cgroup.h | 23 ----------------------- 1 file changed, 23 deletions(-) (limited to 'include/linux') diff --git a/include/linux/page_cgroup.h b/include/linux/page_cgroup.h index 5b0c971d7cae..6d6cb7a57bb3 100644 --- a/include/linux/page_cgroup.h +++ b/include/linux/page_cgroup.h @@ -42,9 +42,6 @@ enum { /* flags for mem_cgroup and file and I/O status */ PCG_MOVE_LOCK, /* For race between move_account v.s. following bits */ PCG_FILE_MAPPED, /* page is accounted as "mapped" */ - PCG_FILE_DIRTY, /* page is dirty */ - PCG_FILE_WRITEBACK, /* page is under writeback */ - PCG_FILE_UNSTABLE_NFS, /* page is NFS unstable */ /* No lock in page_cgroup */ PCG_ACCT_LRU, /* page has been accounted for (under lru_lock) */ }; @@ -65,10 +62,6 @@ static inline void ClearPageCgroup##uname(struct page_cgroup *pc) \ static inline int TestClearPageCgroup##uname(struct page_cgroup *pc) \ { return test_and_clear_bit(PCG_##lname, &pc->flags); } -#define TESTSETPCGFLAG(uname, lname) \ -static inline int TestSetPageCgroup##uname(struct page_cgroup *pc) \ - { return test_and_set_bit(PCG_##lname, &pc->flags); } - /* Cache flag is set only once (at allocation) */ TESTPCGFLAG(Cache, CACHE) CLEARPCGFLAG(Cache, CACHE) @@ -88,22 +81,6 @@ SETPCGFLAG(FileMapped, FILE_MAPPED) CLEARPCGFLAG(FileMapped, FILE_MAPPED) TESTPCGFLAG(FileMapped, FILE_MAPPED) -SETPCGFLAG(FileDirty, FILE_DIRTY) -CLEARPCGFLAG(FileDirty, FILE_DIRTY) -TESTPCGFLAG(FileDirty, FILE_DIRTY) -TESTCLEARPCGFLAG(FileDirty, FILE_DIRTY) -TESTSETPCGFLAG(FileDirty, FILE_DIRTY) - -SETPCGFLAG(FileWriteback, FILE_WRITEBACK) -CLEARPCGFLAG(FileWriteback, FILE_WRITEBACK) -TESTPCGFLAG(FileWriteback, FILE_WRITEBACK) - -SETPCGFLAG(FileUnstableNFS, FILE_UNSTABLE_NFS) -CLEARPCGFLAG(FileUnstableNFS, FILE_UNSTABLE_NFS) -TESTPCGFLAG(FileUnstableNFS, FILE_UNSTABLE_NFS) -TESTCLEARPCGFLAG(FileUnstableNFS, FILE_UNSTABLE_NFS) -TESTSETPCGFLAG(FileUnstableNFS, FILE_UNSTABLE_NFS) - SETPCGFLAG(Migration, MIGRATION) CLEARPCGFLAG(Migration, MIGRATION) TESTPCGFLAG(Migration, MIGRATION) -- cgit v1.2.3 From 323b7fe8f8f6d5ac6214382cf30e8b3a80b265c9 Mon Sep 17 00:00:00 2001 From: Wolfram Sang Date: Fri, 14 Jan 2011 09:34:29 +0100 Subject: include/gpio.h: remove remaining __must_check-annotiations Commit 5f829e405ec4e96f711165a4a7b55c271d4363e2 (gpiolib: add missing functions to generic fallback) also introduced two. Signed-off-by: Wolfram Sang Cc: Greg KH Cc: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/gpio.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/gpio.h b/include/linux/gpio.h index 4b47ed96f131..32720baf70f1 100644 --- a/include/linux/gpio.h +++ b/include/linux/gpio.h @@ -35,13 +35,13 @@ static inline int gpio_request(unsigned gpio, const char *label) return -ENOSYS; } -static inline int __must_check gpio_request_one(unsigned gpio, +static inline int gpio_request_one(unsigned gpio, unsigned long flags, const char *label) { return -ENOSYS; } -static inline int __must_check gpio_request_array(struct gpio *array, size_t num) +static inline int gpio_request_array(struct gpio *array, size_t num) { return -ENOSYS; } -- cgit v1.2.3 From 415e12b2379239973feab91850b0dce985c6058a Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Fri, 7 Jan 2011 00:55:09 +0100 Subject: PCI/ACPI: Request _OSC control once for each root bridge (v3) Move the evaluation of acpi_pci_osc_control_set() (to request control of PCI Express native features) into acpi_pci_root_add() to avoid calling it many times for the same root complex with the same arguments. Additionally, check if all of the requisite _OSC support bits are set before calling acpi_pci_osc_control_set() for a given root complex. References: https://bugzilla.kernel.org/show_bug.cgi?id=20232 Reported-by: Ozan Caglayan Tested-by: Ozan Caglayan Signed-off-by: Rafael J. Wysocki Signed-off-by: Jesse Barnes --- drivers/acpi/apei/hest.c | 22 +++++++++------------- drivers/acpi/pci_root.c | 35 +++++++++++++++++++++++++++++++++++ drivers/pci/pci.h | 8 -------- drivers/pci/pcie/aer/aerdrv.c | 1 + drivers/pci/pcie/aer/aerdrv.h | 3 --- drivers/pci/pcie/portdrv.h | 3 --- drivers/pci/pcie/portdrv_acpi.c | 23 ++++------------------- include/acpi/apei.h | 6 ++++++ include/linux/pci-acpi.h | 6 ++++++ include/linux/pci.h | 11 +++++++++++ 10 files changed, 72 insertions(+), 46 deletions(-) (limited to 'include/linux') diff --git a/drivers/acpi/apei/hest.c b/drivers/acpi/apei/hest.c index daa7bc63f1d4..4ee58e72b730 100644 --- a/drivers/acpi/apei/hest.c +++ b/drivers/acpi/apei/hest.c @@ -195,24 +195,24 @@ static int __init setup_hest_disable(char *str) __setup("hest_disable", setup_hest_disable); -static int __init hest_init(void) +void __init acpi_hest_init(void) { acpi_status status; int rc = -ENODEV; unsigned int ghes_count = 0; if (acpi_disabled) - goto err; + return; if (hest_disable) { - pr_info(HEST_PFX "HEST tabling parsing is disabled.\n"); - goto err; + pr_info(HEST_PFX "Table parsing disabled.\n"); + return; } status = acpi_get_table(ACPI_SIG_HEST, 0, (struct acpi_table_header **)&hest_tab); if (status == AE_NOT_FOUND) { - pr_info(HEST_PFX "Table is not found!\n"); + pr_info(HEST_PFX "Table not found.\n"); goto err; } else if (ACPI_FAILURE(status)) { const char *msg = acpi_format_exception(status); @@ -226,15 +226,11 @@ static int __init hest_init(void) goto err; rc = hest_ghes_dev_register(ghes_count); - if (rc) - goto err; - - pr_info(HEST_PFX "HEST table parsing is initialized.\n"); + if (!rc) { + pr_info(HEST_PFX "Table parsing has been initialized.\n"); + return; + } - return 0; err: hest_disable = 1; - return rc; } - -subsys_initcall(hest_init); diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c index 96668ad09622..d9766797cd98 100644 --- a/drivers/acpi/pci_root.c +++ b/drivers/acpi/pci_root.c @@ -36,6 +36,7 @@ #include #include #include +#include #define PREFIX "ACPI: " @@ -47,6 +48,11 @@ static int acpi_pci_root_add(struct acpi_device *device); static int acpi_pci_root_remove(struct acpi_device *device, int type); static int acpi_pci_root_start(struct acpi_device *device); +#define ACPI_PCIE_REQ_SUPPORT (OSC_EXT_PCI_CONFIG_SUPPORT \ + | OSC_ACTIVE_STATE_PWR_SUPPORT \ + | OSC_CLOCK_PWR_CAPABILITY_SUPPORT \ + | OSC_MSI_SUPPORT) + static const struct acpi_device_id root_device_ids[] = { {"PNP0A03", 0}, {"", 0}, @@ -566,6 +572,33 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device) if (flags != base_flags) acpi_pci_osc_support(root, flags); + if (!pcie_ports_disabled + && (flags & ACPI_PCIE_REQ_SUPPORT) == ACPI_PCIE_REQ_SUPPORT) { + flags = OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL + | OSC_PCI_EXPRESS_NATIVE_HP_CONTROL + | OSC_PCI_EXPRESS_PME_CONTROL; + + if (pci_aer_available()) { + if (aer_acpi_firmware_first()) + dev_dbg(root->bus->bridge, + "PCIe errors handled by BIOS.\n"); + else + flags |= OSC_PCI_EXPRESS_AER_CONTROL; + } + + dev_info(root->bus->bridge, + "Requesting ACPI _OSC control (0x%02x)\n", flags); + + status = acpi_pci_osc_control_set(device->handle, &flags, + OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL); + if (ACPI_SUCCESS(status)) + dev_info(root->bus->bridge, + "ACPI _OSC control (0x%02x) granted\n", flags); + else + dev_dbg(root->bus->bridge, + "ACPI _OSC request failed (code %d)\n", status); + } + pci_acpi_add_bus_pm_notifier(device, root->bus); if (device->wakeup.flags.run_wake) device_set_run_wake(root->bus->bridge, true); @@ -603,6 +636,8 @@ static int __init acpi_pci_root_init(void) if (acpi_pci_disabled) return 0; + acpi_hest_init(); + pci_acpi_crs_quirks(); if (acpi_bus_register_driver(&acpi_pci_root_driver) < 0) return -ENODEV; diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 7d33f6673868..16ae9659346b 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -140,14 +140,6 @@ static inline void pci_no_msi(void) { } static inline void pci_msi_init_pci_dev(struct pci_dev *dev) { } #endif -#ifdef CONFIG_PCIEAER -void pci_no_aer(void); -bool pci_aer_available(void); -#else -static inline void pci_no_aer(void) { } -static inline bool pci_aer_available(void) { return false; } -#endif - static inline int pci_no_d1d2(struct pci_dev *dev) { unsigned int parent_dstates = 0; diff --git a/drivers/pci/pcie/aer/aerdrv.c b/drivers/pci/pcie/aer/aerdrv.c index 2b2b6508efde..58ad7917553c 100644 --- a/drivers/pci/pcie/aer/aerdrv.c +++ b/drivers/pci/pcie/aer/aerdrv.c @@ -17,6 +17,7 @@ #include #include +#include #include #include #include diff --git a/drivers/pci/pcie/aer/aerdrv.h b/drivers/pci/pcie/aer/aerdrv.h index 9656e3060412..80c11d131499 100644 --- a/drivers/pci/pcie/aer/aerdrv.h +++ b/drivers/pci/pcie/aer/aerdrv.h @@ -132,7 +132,6 @@ static inline int aer_osc_setup(struct pcie_device *pciedev) #ifdef CONFIG_ACPI_APEI extern int pcie_aer_get_firmware_first(struct pci_dev *pci_dev); -extern bool aer_acpi_firmware_first(void); #else static inline int pcie_aer_get_firmware_first(struct pci_dev *pci_dev) { @@ -140,8 +139,6 @@ static inline int pcie_aer_get_firmware_first(struct pci_dev *pci_dev) return pci_dev->__aer_firmware_first; return 0; } - -static inline bool aer_acpi_firmware_first(void) { return false; } #endif static inline void pcie_aer_force_firmware_first(struct pci_dev *pci_dev, diff --git a/drivers/pci/pcie/portdrv.h b/drivers/pci/pcie/portdrv.h index 8fcc03598b29..bd00a01aef14 100644 --- a/drivers/pci/pcie/portdrv.h +++ b/drivers/pci/pcie/portdrv.h @@ -20,9 +20,6 @@ #define get_descriptor_id(type, service) (((type - 4) << 4) | service) -extern bool pcie_ports_disabled; -extern bool pcie_ports_auto; - extern struct bus_type pcie_port_bus_type; extern int pcie_port_device_register(struct pci_dev *dev); #ifdef CONFIG_PM diff --git a/drivers/pci/pcie/portdrv_acpi.c b/drivers/pci/pcie/portdrv_acpi.c index 5982b6a63b89..a86b56e5f2f2 100644 --- a/drivers/pci/pcie/portdrv_acpi.c +++ b/drivers/pci/pcie/portdrv_acpi.c @@ -33,7 +33,7 @@ */ int pcie_port_acpi_setup(struct pci_dev *port, int *srv_mask) { - acpi_status status; + struct acpi_pci_root *root; acpi_handle handle; u32 flags; @@ -44,26 +44,11 @@ int pcie_port_acpi_setup(struct pci_dev *port, int *srv_mask) if (!handle) return -EINVAL; - flags = OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL - | OSC_PCI_EXPRESS_NATIVE_HP_CONTROL - | OSC_PCI_EXPRESS_PME_CONTROL; - - if (pci_aer_available()) { - if (aer_acpi_firmware_first()) - dev_dbg(&port->dev, "PCIe errors handled by BIOS.\n"); - else - flags |= OSC_PCI_EXPRESS_AER_CONTROL; - } - - status = acpi_pci_osc_control_set(handle, &flags, - OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL); - if (ACPI_FAILURE(status)) { - dev_dbg(&port->dev, "ACPI _OSC request failed (code %d)\n", - status); + root = acpi_pci_find_root(handle); + if (!root) return -ENODEV; - } - dev_info(&port->dev, "ACPI _OSC control granted for 0x%02x\n", flags); + flags = root->osc_control_set; *srv_mask = PCIE_PORT_SERVICE_VC; if (flags & OSC_PCI_EXPRESS_NATIVE_HP_CONTROL) diff --git a/include/acpi/apei.h b/include/acpi/apei.h index b3365025ff8d..c4dbb132d902 100644 --- a/include/acpi/apei.h +++ b/include/acpi/apei.h @@ -19,6 +19,12 @@ extern int hest_disable; extern int erst_disable; +#ifdef CONFIG_ACPI_APEI +void __init acpi_hest_init(void); +#else +static inline void acpi_hest_init(void) { return; } +#endif + typedef int (*apei_hest_func_t)(struct acpi_hest_header *hest_hdr, void *data); int apei_hest_parse(apei_hest_func_t func, void *data); diff --git a/include/linux/pci-acpi.h b/include/linux/pci-acpi.h index c8b6473c5f42..479d9bb88e11 100644 --- a/include/linux/pci-acpi.h +++ b/include/linux/pci-acpi.h @@ -40,4 +40,10 @@ static inline acpi_handle acpi_find_root_bridge_handle(struct pci_dev *pdev) { return NULL; } #endif +#ifdef CONFIG_ACPI_APEI +extern bool aer_acpi_firmware_first(void); +#else +static inline bool aer_acpi_firmware_first(void) { return false; } +#endif + #endif /* _PCI_ACPI_H_ */ diff --git a/include/linux/pci.h b/include/linux/pci.h index 63cbadce337e..12dd86a82a15 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -994,6 +994,9 @@ extern void pci_restore_msi_state(struct pci_dev *dev); extern int pci_msi_enabled(void); #endif +extern bool pcie_ports_disabled; +extern bool pcie_ports_auto; + #ifndef CONFIG_PCIEASPM static inline int pcie_aspm_enabled(void) { @@ -1003,6 +1006,14 @@ static inline int pcie_aspm_enabled(void) extern int pcie_aspm_enabled(void); #endif +#ifdef CONFIG_PCIEAER +void pci_no_aer(void); +bool pci_aer_available(void); +#else +static inline void pci_no_aer(void) { } +static inline bool pci_aer_available(void) { return false; } +#endif + #ifndef CONFIG_PCIE_ECRC static inline void pcie_set_ecrc_checking(struct pci_dev *dev) { -- cgit v1.2.3 From b6e335aeeb114dccb07eaa09e8b62ff9510cf745 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 29 Dec 2010 13:21:23 +0100 Subject: PCI/PM: Use pm_wakeup_event() directly for reporting wakeup events After recent changes related to wakeup events pm_wakeup_event() automatically checks if the given device is configured to signal wakeup, so pci_wakeup_event() may be a static inline function calling pm_wakeup_event() directly. Signed-off-by: Rafael J. Wysocki Signed-off-by: Jesse Barnes --- drivers/pci/pci.c | 16 ---------------- drivers/pci/pci.h | 6 ++++++ include/linux/pci.h | 1 - 3 files changed, 6 insertions(+), 17 deletions(-) (limited to 'include/linux') diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 6762dcae90ab..bf7ad2c09955 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -1297,22 +1297,6 @@ bool pci_check_pme_status(struct pci_dev *dev) return ret; } -/* - * Time to wait before the system can be put into a sleep state after reporting - * a wakeup event signaled by a PCI device. - */ -#define PCI_WAKEUP_COOLDOWN 100 - -/** - * pci_wakeup_event - Report a wakeup event related to a given PCI device. - * @dev: Device to report the wakeup event for. - */ -void pci_wakeup_event(struct pci_dev *dev) -{ - if (device_may_wakeup(&dev->dev)) - pm_wakeup_event(&dev->dev, PCI_WAKEUP_COOLDOWN); -} - /** * pci_pme_wakeup - Wake up a PCI device if its PME Status bit is set. * @dev: Device to handle. diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 16ae9659346b..f69d6e0fda75 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -74,6 +74,12 @@ extern void pci_pm_init(struct pci_dev *dev); extern void platform_pci_wakeup_init(struct pci_dev *dev); extern void pci_allocate_cap_save_buffers(struct pci_dev *dev); +static inline void pci_wakeup_event(struct pci_dev *dev) +{ + /* Wait 100 ms before the system can be put into a sleep state. */ + pm_wakeup_event(&dev->dev, 100); +} + static inline bool pci_is_bridge(struct pci_dev *pci_dev) { return !!(pci_dev->subordinate); diff --git a/include/linux/pci.h b/include/linux/pci.h index 12dd86a82a15..9e67dcd054c4 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -820,7 +820,6 @@ int pci_prepare_to_sleep(struct pci_dev *dev); int pci_back_from_sleep(struct pci_dev *dev); bool pci_dev_run_wake(struct pci_dev *dev); bool pci_check_pme_status(struct pci_dev *dev); -void pci_wakeup_event(struct pci_dev *dev); void pci_pme_wakeup_bus(struct pci_bus *bus); static inline int pci_enable_wake(struct pci_dev *dev, pci_power_t state, -- cgit v1.2.3 From 49731baa41df404c2c3f44555869ab387363af43 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 14 Jan 2011 18:43:57 +0100 Subject: block: restore multiple bd_link_disk_holder() support Commit e09b457b (block: simplify holder symlink handling) incorrectly assumed that there is only one link at maximum. dm may use multiple links and expects block layer to track reference count for each link, which is different from and unrelated to the exclusive device holder identified by @holder when the device is opened. Remove the single holder assumption and automatic removal of the link and revive the per-link reference count tracking. The code essentially behaves the same as before commit e09b457b sans the unnecessary kobject reference count dancing. While at it, note that this facility should not be used by anyone else than the current ones. Sysfs symlinks shouldn't be abused like this and the whole thing doesn't belong in the block layer at all. Signed-off-by: Tejun Heo Reported-by: Milan Broz Cc: Jun'ichi Nomura Cc: Neil Brown Cc: linux-raid@vger.kernel.org Cc: Kay Sievers Signed-off-by: Jens Axboe --- drivers/md/dm-table.c | 1 + drivers/md/md.c | 1 + fs/block_dev.c | 93 +++++++++++++++++++++++++++++++++++++++++---------- include/linux/fs.h | 8 ++++- 4 files changed, 84 insertions(+), 19 deletions(-) (limited to 'include/linux') diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index dffa0ac7c4f0..38e4eb1bb965 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -350,6 +350,7 @@ static void close_dev(struct dm_dev_internal *d, struct mapped_device *md) if (!d->dm_dev.bdev) return; + bd_unlink_disk_holder(d->dm_dev.bdev, dm_disk(md)); blkdev_put(d->dm_dev.bdev, d->dm_dev.mode | FMODE_EXCL); d->dm_dev.bdev = NULL; } diff --git a/drivers/md/md.c b/drivers/md/md.c index cf8594c5ea21..b76cfc89e1b5 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -1912,6 +1912,7 @@ static void unbind_rdev_from_array(mdk_rdev_t * rdev) MD_BUG(); return; } + bd_unlink_disk_holder(rdev->bdev, rdev->mddev->gendisk); list_del_rcu(&rdev->same_set); printk(KERN_INFO "md: unbind<%s>\n", bdevname(rdev->bdev,b)); rdev->mddev = NULL; diff --git a/fs/block_dev.c b/fs/block_dev.c index fe3f59c14a02..333a7bb4cb9c 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -432,6 +432,9 @@ static void init_once(void *foo) mutex_init(&bdev->bd_mutex); INIT_LIST_HEAD(&bdev->bd_inodes); INIT_LIST_HEAD(&bdev->bd_list); +#ifdef CONFIG_SYSFS + INIT_LIST_HEAD(&bdev->bd_holder_disks); +#endif inode_init_once(&ei->vfs_inode); /* Initialize mutex for freeze. */ mutex_init(&bdev->bd_fsfreeze_mutex); @@ -779,6 +782,23 @@ static struct block_device *bd_start_claiming(struct block_device *bdev, } #ifdef CONFIG_SYSFS +struct bd_holder_disk { + struct list_head list; + struct gendisk *disk; + int refcnt; +}; + +static struct bd_holder_disk *bd_find_holder_disk(struct block_device *bdev, + struct gendisk *disk) +{ + struct bd_holder_disk *holder; + + list_for_each_entry(holder, &bdev->bd_holder_disks, list) + if (holder->disk == disk) + return holder; + return NULL; +} + static int add_symlink(struct kobject *from, struct kobject *to) { return sysfs_create_link(from, to, kobject_name(to)); @@ -794,6 +814,8 @@ static void del_symlink(struct kobject *from, struct kobject *to) * @bdev: the claimed slave bdev * @disk: the holding disk * + * DON'T USE THIS UNLESS YOU'RE ALREADY USING IT. + * * This functions creates the following sysfs symlinks. * * - from "slaves" directory of the holder @disk to the claimed @bdev @@ -817,47 +839,83 @@ static void del_symlink(struct kobject *from, struct kobject *to) */ int bd_link_disk_holder(struct block_device *bdev, struct gendisk *disk) { + struct bd_holder_disk *holder; int ret = 0; mutex_lock(&bdev->bd_mutex); - WARN_ON_ONCE(!bdev->bd_holder || bdev->bd_holder_disk); + WARN_ON_ONCE(!bdev->bd_holder); /* FIXME: remove the following once add_disk() handles errors */ if (WARN_ON(!disk->slave_dir || !bdev->bd_part->holder_dir)) goto out_unlock; - ret = add_symlink(disk->slave_dir, &part_to_dev(bdev->bd_part)->kobj); - if (ret) + holder = bd_find_holder_disk(bdev, disk); + if (holder) { + holder->refcnt++; goto out_unlock; + } - ret = add_symlink(bdev->bd_part->holder_dir, &disk_to_dev(disk)->kobj); - if (ret) { - del_symlink(disk->slave_dir, &part_to_dev(bdev->bd_part)->kobj); + holder = kzalloc(sizeof(*holder), GFP_KERNEL); + if (!holder) { + ret = -ENOMEM; goto out_unlock; } - bdev->bd_holder_disk = disk; + INIT_LIST_HEAD(&holder->list); + holder->disk = disk; + holder->refcnt = 1; + + ret = add_symlink(disk->slave_dir, &part_to_dev(bdev->bd_part)->kobj); + if (ret) + goto out_free; + + ret = add_symlink(bdev->bd_part->holder_dir, &disk_to_dev(disk)->kobj); + if (ret) + goto out_del; + + list_add(&holder->list, &bdev->bd_holder_disks); + goto out_unlock; + +out_del: + del_symlink(disk->slave_dir, &part_to_dev(bdev->bd_part)->kobj); +out_free: + kfree(holder); out_unlock: mutex_unlock(&bdev->bd_mutex); return ret; } EXPORT_SYMBOL_GPL(bd_link_disk_holder); -static void bd_unlink_disk_holder(struct block_device *bdev) +/** + * bd_unlink_disk_holder - destroy symlinks created by bd_link_disk_holder() + * @bdev: the calimed slave bdev + * @disk: the holding disk + * + * DON'T USE THIS UNLESS YOU'RE ALREADY USING IT. + * + * CONTEXT: + * Might sleep. + */ +void bd_unlink_disk_holder(struct block_device *bdev, struct gendisk *disk) { - struct gendisk *disk = bdev->bd_holder_disk; + struct bd_holder_disk *holder; - bdev->bd_holder_disk = NULL; - if (!disk) - return; + mutex_lock(&bdev->bd_mutex); - del_symlink(disk->slave_dir, &part_to_dev(bdev->bd_part)->kobj); - del_symlink(bdev->bd_part->holder_dir, &disk_to_dev(disk)->kobj); + holder = bd_find_holder_disk(bdev, disk); + + if (!WARN_ON_ONCE(holder == NULL) && !--holder->refcnt) { + del_symlink(disk->slave_dir, &part_to_dev(bdev->bd_part)->kobj); + del_symlink(bdev->bd_part->holder_dir, + &disk_to_dev(disk)->kobj); + list_del_init(&holder->list); + kfree(holder); + } + + mutex_unlock(&bdev->bd_mutex); } -#else -static inline void bd_unlink_disk_holder(struct block_device *bdev) -{ } +EXPORT_SYMBOL_GPL(bd_unlink_disk_holder); #endif /** @@ -1380,7 +1438,6 @@ int blkdev_put(struct block_device *bdev, fmode_t mode) * unblock evpoll if it was a write holder. */ if (bdev_free) { - bd_unlink_disk_holder(bdev); if (bdev->bd_write_holder) { disk_unblock_events(bdev->bd_disk); bdev->bd_write_holder = false; diff --git a/include/linux/fs.h b/include/linux/fs.h index 3984f2358d1f..fb2190349cdf 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -666,7 +666,7 @@ struct block_device { int bd_holders; bool bd_write_holder; #ifdef CONFIG_SYSFS - struct gendisk * bd_holder_disk; /* for sysfs slave linkng */ + struct list_head bd_holder_disks; #endif struct block_device * bd_contains; unsigned bd_block_size; @@ -2058,12 +2058,18 @@ extern struct block_device *blkdev_get_by_dev(dev_t dev, fmode_t mode, extern int blkdev_put(struct block_device *bdev, fmode_t mode); #ifdef CONFIG_SYSFS extern int bd_link_disk_holder(struct block_device *bdev, struct gendisk *disk); +extern void bd_unlink_disk_holder(struct block_device *bdev, + struct gendisk *disk); #else static inline int bd_link_disk_holder(struct block_device *bdev, struct gendisk *disk) { return 0; } +static inline void bd_unlink_disk_holder(struct block_device *bdev, + struct gendisk *disk) +{ +} #endif #endif -- cgit v1.2.3 From ab0724ffee24409a7f81afb539b2ac29090bff3e Mon Sep 17 00:00:00 2001 From: Markus Trippelsdorf Date: Sat, 15 Jan 2011 00:07:22 +0100 Subject: PCI / ACPI: Fix build issue in pci_root.c for !CONFIG_PCIEPORTBUS The compilation of drivers/acpi/pci_root.c fails if CONFIG_PCIEPORTBUS is unset. Fix the problem. Signed-off-by: Rafael J. Wysocki Signed-off-by: Linus Torvalds --- include/linux/pci.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/linux') diff --git a/include/linux/pci.h b/include/linux/pci.h index 9e67dcd054c4..559d02897075 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -993,8 +993,13 @@ extern void pci_restore_msi_state(struct pci_dev *dev); extern int pci_msi_enabled(void); #endif +#ifdef CONFIG_PCIEPORTBUS extern bool pcie_ports_disabled; extern bool pcie_ports_auto; +#else +#define pcie_ports_disabled true +#define pcie_ports_auto false +#endif #ifndef CONFIG_PCIEASPM static inline int pcie_aspm_enabled(void) -- cgit v1.2.3 From 0029227f1bc30b6c809ae751f9e7af6cef900997 Mon Sep 17 00:00:00 2001 From: Andiry Xu Date: Mon, 27 Dec 2010 17:39:02 +0800 Subject: xHCI: synchronize irq in xhci_suspend() Synchronize the interrupts instead of free them in xhci_suspend(). This will prevent a double free when the host is suspended and then the card removed. Set the flag hcd->msix_enabled when using MSI-X, and check the flag in suspend_common(). MSI-X synchronization will be handled by xhci_suspend(), and MSI/INTx will be synchronized in suspend_common(). This patch should be queued for the 2.6.37 stable tree. Reported-by: Matthew Garrett Signed-off-by: Andiry Xu Signed-off-by: Sarah Sharp Cc: stable@kernel.org --- drivers/usb/core/hcd-pci.c | 7 ++++++- drivers/usb/host/xhci.c | 46 +++++++++++++++------------------------------- include/linux/usb/hcd.h | 1 + 3 files changed, 22 insertions(+), 32 deletions(-) (limited to 'include/linux') diff --git a/drivers/usb/core/hcd-pci.c b/drivers/usb/core/hcd-pci.c index b55d46070a25..f71e8e307e0f 100644 --- a/drivers/usb/core/hcd-pci.c +++ b/drivers/usb/core/hcd-pci.c @@ -405,7 +405,12 @@ static int suspend_common(struct device *dev, bool do_wakeup) return retval; } - synchronize_irq(pci_dev->irq); + /* If MSI-X is enabled, the driver will have synchronized all vectors + * in pci_suspend(). If MSI or legacy PCI is enabled, that will be + * synchronized here. + */ + if (!hcd->msix_enabled) + synchronize_irq(pci_dev->irq); /* Downstream ports from this root hub should already be quiesced, so * there will be no DMA activity. Now we can shut down the upstream diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c index 45e4a3108cc3..d48edfa5043a 100644 --- a/drivers/usb/host/xhci.c +++ b/drivers/usb/host/xhci.c @@ -226,7 +226,8 @@ static int xhci_setup_msi(struct xhci_hcd *xhci) static int xhci_setup_msix(struct xhci_hcd *xhci) { int i, ret = 0; - struct pci_dev *pdev = to_pci_dev(xhci_to_hcd(xhci)->self.controller); + struct usb_hcd *hcd = xhci_to_hcd(xhci); + struct pci_dev *pdev = to_pci_dev(hcd->self.controller); /* * calculate number of msi-x vectors supported. @@ -265,6 +266,7 @@ static int xhci_setup_msix(struct xhci_hcd *xhci) goto disable_msix; } + hcd->msix_enabled = 1; return ret; disable_msix: @@ -280,7 +282,8 @@ free_entries: /* Free any IRQs and disable MSI-X */ static void xhci_cleanup_msix(struct xhci_hcd *xhci) { - struct pci_dev *pdev = to_pci_dev(xhci_to_hcd(xhci)->self.controller); + struct usb_hcd *hcd = xhci_to_hcd(xhci); + struct pci_dev *pdev = to_pci_dev(hcd->self.controller); xhci_free_irq(xhci); @@ -292,6 +295,7 @@ static void xhci_cleanup_msix(struct xhci_hcd *xhci) pci_disable_msi(pdev); } + hcd->msix_enabled = 0; return; } @@ -647,6 +651,7 @@ int xhci_suspend(struct xhci_hcd *xhci) int rc = 0; struct usb_hcd *hcd = xhci_to_hcd(xhci); u32 command; + int i; spin_lock_irq(&xhci->lock); clear_bit(HCD_FLAG_HW_ACCESSIBLE, &hcd->flags); @@ -677,10 +682,15 @@ int xhci_suspend(struct xhci_hcd *xhci) spin_unlock_irq(&xhci->lock); return -ETIMEDOUT; } - /* step 5: remove core well power */ - xhci_cleanup_msix(xhci); spin_unlock_irq(&xhci->lock); + /* step 5: remove core well power */ + /* synchronize irq when using MSI-X */ + if (xhci->msix_entries) { + for (i = 0; i < xhci->msix_count; i++) + synchronize_irq(xhci->msix_entries[i].vector); + } + return rc; } @@ -694,7 +704,6 @@ int xhci_resume(struct xhci_hcd *xhci, bool hibernated) { u32 command, temp = 0; struct usb_hcd *hcd = xhci_to_hcd(xhci); - struct pci_dev *pdev = to_pci_dev(hcd->self.controller); int old_state, retval; old_state = hcd->state; @@ -729,9 +738,8 @@ int xhci_resume(struct xhci_hcd *xhci, bool hibernated) xhci_dbg(xhci, "Stop HCD\n"); xhci_halt(xhci); xhci_reset(xhci); - if (hibernated) - xhci_cleanup_msix(xhci); spin_unlock_irq(&xhci->lock); + xhci_cleanup_msix(xhci); #ifdef CONFIG_USB_XHCI_HCD_DEBUGGING /* Tell the event ring poll function not to reschedule */ @@ -765,30 +773,6 @@ int xhci_resume(struct xhci_hcd *xhci, bool hibernated) return retval; } - spin_unlock_irq(&xhci->lock); - /* Re-setup MSI-X */ - if (hcd->irq) - free_irq(hcd->irq, hcd); - hcd->irq = -1; - - retval = xhci_setup_msix(xhci); - if (retval) - /* fall back to msi*/ - retval = xhci_setup_msi(xhci); - - if (retval) { - /* fall back to legacy interrupt*/ - retval = request_irq(pdev->irq, &usb_hcd_irq, IRQF_SHARED, - hcd->irq_descr, hcd); - if (retval) { - xhci_err(xhci, "request interrupt %d failed\n", - pdev->irq); - return retval; - } - hcd->irq = pdev->irq; - } - - spin_lock_irq(&xhci->lock); /* step 4: set Run/Stop bit */ command = xhci_readl(xhci, &xhci->op_regs->command); command |= CMD_RUN; diff --git a/include/linux/usb/hcd.h b/include/linux/usb/hcd.h index dd6ee49a0844..a854fe89484e 100644 --- a/include/linux/usb/hcd.h +++ b/include/linux/usb/hcd.h @@ -112,6 +112,7 @@ struct usb_hcd { /* Flags that get set only during HCD registration or removal. */ unsigned rh_registered:1;/* is root hub registered? */ unsigned rh_pollable:1; /* may we poll the root hub? */ + unsigned msix_enabled:1; /* driver has MSI-X enabled? */ /* The next flag is a stopgap, to be removed when all the HCDs * support the new root-hub polling mechanism. */ -- cgit v1.2.3 From 98d530fe246b65fbd3cdeeeca319a80c46cb4793 Mon Sep 17 00:00:00 2001 From: Russell King - ARM Linux Date: Sat, 1 Jan 2011 23:00:23 +0000 Subject: Fix dmaengine_submit() return type desc->tx_submit's return type is dma_cookie_t, not int. Therefore, dmaengine_submit() should match this return type as it's just wrapping this detail. Signed-off-by: Russell King Signed-off-by: Dan Williams --- include/linux/dmaengine.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h index 9d8688b92d8b..830935b7c49c 100644 --- a/include/linux/dmaengine.h +++ b/include/linux/dmaengine.h @@ -532,7 +532,7 @@ static inline int dmaengine_resume(struct dma_chan *chan) return dmaengine_device_control(chan, DMA_RESUME, 0); } -static inline int dmaengine_submit(struct dma_async_tx_descriptor *desc) +static inline dma_cookie_t dmaengine_submit(struct dma_async_tx_descriptor *desc) { return desc->tx_submit(desc); } -- cgit v1.2.3 From 96a608a4bfd8468c21881b3f92024923886eb015 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Fri, 14 Jan 2011 17:51:11 -0800 Subject: ARM: PL08x: fix a warning drivers/dma/amba-pl08x.c: In function 'pl08x_start_txd': drivers/dma/amba-pl08x.c:205: warning: dereferencing 'void *' pointer We never dereference llis_va aside from assigning it to a struct pl08x_lli pointer or calculating the address of array element 0. Signed-off-by: Dan Williams --- include/linux/amba/pl08x.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/amba/pl08x.h b/include/linux/amba/pl08x.h index 933b4ed12be5..5b87b6aac3f8 100644 --- a/include/linux/amba/pl08x.h +++ b/include/linux/amba/pl08x.h @@ -118,7 +118,7 @@ struct pl08x_txd { dma_addr_t dst_addr; size_t len; dma_addr_t llis_bus; - void *llis_va; + struct pl08x_lli *llis_va; bool active; /* Default cctl value for LLIs */ u32 cctl; -- cgit v1.2.3 From 9875cf806403fae66b2410a3c2cc820d97731e04 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 14 Jan 2011 18:45:21 +0000 Subject: Add a dentry op to handle automounting rather than abusing follow_link() Add a dentry op (d_automount) to handle automounting directories rather than abusing the follow_link() inode operation. The operation is keyed off a new dentry flag (DCACHE_NEED_AUTOMOUNT). This also makes it easier to add an AT_ flag to suppress terminal segment automount during pathwalk and removes the need for the kludge code in the pathwalk algorithm to handle directories with follow_link() semantics. The ->d_automount() dentry operation: struct vfsmount *(*d_automount)(struct path *mountpoint); takes a pointer to the directory to be mounted upon, which is expected to provide sufficient data to determine what should be mounted. If successful, it should return the vfsmount struct it creates (which it should also have added to the namespace using do_add_mount() or similar). If there's a collision with another automount attempt, NULL should be returned. If the directory specified by the parameter should be used directly rather than being mounted upon, -EISDIR should be returned. In any other case, an error code should be returned. The ->d_automount() operation is called with no locks held and may sleep. At this point the pathwalk algorithm will be in ref-walk mode. Within fs/namei.c itself, a new pathwalk subroutine (follow_automount()) is added to handle mountpoints. It will return -EREMOTE if the automount flag was set, but no d_automount() op was supplied, -ELOOP if we've encountered too many symlinks or mountpoints, -EISDIR if the walk point should be used without mounting and 0 if successful. The path will be updated to point to the mounted filesystem if a successful automount took place. __follow_mount() is replaced by follow_managed() which is more generic (especially with the patch that adds ->d_manage()). This handles transits from directories during pathwalk, including automounting and skipping over mountpoints (and holding processes with the next patch). __follow_mount_rcu() will jump out of RCU-walk mode if it encounters an automount point with nothing mounted on it. follow_dotdot*() does not handle automounts as you don't want to trigger them whilst following "..". I've also extracted the mount/don't-mount logic from autofs4 and included it here. It makes the mount go ahead anyway if someone calls open() or creat(), tries to traverse the directory, tries to chdir/chroot/etc. into the directory, or sticks a '/' on the end of the pathname. If they do a stat(), however, they'll only trigger the automount if they didn't also say O_NOFOLLOW. I've also added an inode flag (S_AUTOMOUNT) so that filesystems can mark their inodes as automount points. This flag is automatically propagated to the dentry as DCACHE_NEED_AUTOMOUNT by __d_instantiate(). This saves NFS and could save AFS a private flag bit apiece, but is not strictly necessary. It would be preferable to do the propagation in d_set_d_op(), but that doesn't normally have access to the inode. [AV: fixed breakage in case if __follow_mount_rcu() fails and nameidata_drop_rcu() succeeds in RCU case of do_lookup(); we need to fall through to non-RCU case after that, rather than just returning with ungrabbed *path] Signed-off-by: David Howells Was-Acked-by: Ian Kent Signed-off-by: Al Viro --- Documentation/filesystems/Locking | 2 + Documentation/filesystems/vfs.txt | 14 +++ fs/dcache.c | 5 +- fs/namei.c | 226 ++++++++++++++++++++++++++++---------- include/linux/dcache.h | 7 +- include/linux/fs.h | 2 + 6 files changed, 198 insertions(+), 58 deletions(-) (limited to 'include/linux') diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking index 977d8919cc69..5f0c52a07386 100644 --- a/Documentation/filesystems/Locking +++ b/Documentation/filesystems/Locking @@ -19,6 +19,7 @@ prototypes: void (*d_release)(struct dentry *); void (*d_iput)(struct dentry *, struct inode *); char *(*d_dname)((struct dentry *dentry, char *buffer, int buflen); + struct vfsmount *(*d_automount)(struct path *path); locking rules: rename_lock ->d_lock may block rcu-walk @@ -29,6 +30,7 @@ d_delete: no yes no no d_release: no no yes no d_iput: no no yes no d_dname: no no no no +d_automount: no no yes no --------------------------- inode_operations --------------------------- prototypes: diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt index cae6d27c9f5b..726a4f6fa3c9 100644 --- a/Documentation/filesystems/vfs.txt +++ b/Documentation/filesystems/vfs.txt @@ -864,6 +864,7 @@ struct dentry_operations { void (*d_release)(struct dentry *); void (*d_iput)(struct dentry *, struct inode *); char *(*d_dname)(struct dentry *, char *, int); + struct vfsmount *(*d_automount)(struct path *); }; d_revalidate: called when the VFS needs to revalidate a dentry. This @@ -930,6 +931,19 @@ struct dentry_operations { at the end of the buffer, and returns a pointer to the first char. dynamic_dname() helper function is provided to take care of this. + d_automount: called when an automount dentry is to be traversed (optional). + This should create a new VFS mount record, mount it on the directory + and return the record to the caller. The caller is supplied with a + path parameter giving the automount directory to describe the automount + target and the parent VFS mount record to provide inheritable mount + parameters. NULL should be returned if someone else managed to make + the automount first. If the automount failed, then an error code + should be returned. + + This function is only used if DCACHE_NEED_AUTOMOUNT is set on the + dentry. This is set by __d_instantiate() if S_AUTOMOUNT is set on the + inode being added. + Example : static char *pipefs_dname(struct dentry *dent, char *buffer, int buflen) diff --git a/fs/dcache.c b/fs/dcache.c index 0c6d5c549d84..51f7bb6463af 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -1380,8 +1380,11 @@ EXPORT_SYMBOL(d_set_d_op); static void __d_instantiate(struct dentry *dentry, struct inode *inode) { spin_lock(&dentry->d_lock); - if (inode) + if (inode) { + if (unlikely(IS_AUTOMOUNT(inode))) + dentry->d_flags |= DCACHE_NEED_AUTOMOUNT; list_add(&dentry->d_alias, &inode->i_dentry); + } dentry->d_inode = inode; dentry_rcuwalk_barrier(dentry); spin_unlock(&dentry->d_lock); diff --git a/fs/namei.c b/fs/namei.c index 529e917ad2fc..16109da68bbf 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -896,51 +896,120 @@ int follow_up(struct path *path) } /* - * serialization is taken care of in namespace.c + * Perform an automount + * - return -EISDIR to tell follow_managed() to stop and return the path we + * were called with. */ -static void __follow_mount_rcu(struct nameidata *nd, struct path *path, - struct inode **inode) +static int follow_automount(struct path *path, unsigned flags, + bool *need_mntput) { - while (d_mountpoint(path->dentry)) { - struct vfsmount *mounted; - mounted = __lookup_mnt(path->mnt, path->dentry, 1); - if (!mounted) - return; - path->mnt = mounted; - path->dentry = mounted->mnt_root; - nd->seq = read_seqcount_begin(&path->dentry->d_seq); - *inode = path->dentry->d_inode; + struct vfsmount *mnt; + + if (!path->dentry->d_op || !path->dentry->d_op->d_automount) + return -EREMOTE; + + /* We want to mount if someone is trying to open/create a file of any + * type under the mountpoint, wants to traverse through the mountpoint + * or wants to open the mounted directory. + * + * We don't want to mount if someone's just doing a stat and they've + * set AT_SYMLINK_NOFOLLOW - unless they're stat'ing a directory and + * appended a '/' to the name. + */ + if (!(flags & LOOKUP_FOLLOW) && + !(flags & (LOOKUP_CONTINUE | LOOKUP_DIRECTORY | + LOOKUP_OPEN | LOOKUP_CREATE))) + return -EISDIR; + + current->total_link_count++; + if (current->total_link_count >= 40) + return -ELOOP; + + mnt = path->dentry->d_op->d_automount(path); + if (IS_ERR(mnt)) { + /* + * The filesystem is allowed to return -EISDIR here to indicate + * it doesn't want to automount. For instance, autofs would do + * this so that its userspace daemon can mount on this dentry. + * + * However, we can only permit this if it's a terminal point in + * the path being looked up; if it wasn't then the remainder of + * the path is inaccessible and we should say so. + */ + if (PTR_ERR(mnt) == -EISDIR && (flags & LOOKUP_CONTINUE)) + return -EREMOTE; + return PTR_ERR(mnt); } -} + if (!mnt) /* mount collision */ + return 0; -static int __follow_mount(struct path *path) -{ - int res = 0; - while (d_mountpoint(path->dentry)) { - struct vfsmount *mounted = lookup_mnt(path); - if (!mounted) - break; - dput(path->dentry); - if (res) - mntput(path->mnt); - path->mnt = mounted; - path->dentry = dget(mounted->mnt_root); - res = 1; + if (mnt->mnt_sb == path->mnt->mnt_sb && + mnt->mnt_root == path->dentry) { + mntput(mnt); + return -ELOOP; } - return res; + + dput(path->dentry); + if (*need_mntput) + mntput(path->mnt); + path->mnt = mnt; + path->dentry = dget(mnt->mnt_root); + *need_mntput = true; + return 0; } -static void follow_mount(struct path *path) +/* + * Handle a dentry that is managed in some way. + * - Flagged as mountpoint + * - Flagged as automount point + * + * This may only be called in refwalk mode. + * + * Serialization is taken care of in namespace.c + */ +static int follow_managed(struct path *path, unsigned flags) { - while (d_mountpoint(path->dentry)) { - struct vfsmount *mounted = lookup_mnt(path); - if (!mounted) - break; - dput(path->dentry); - mntput(path->mnt); - path->mnt = mounted; - path->dentry = dget(mounted->mnt_root); + unsigned managed; + bool need_mntput = false; + int ret; + + /* Given that we're not holding a lock here, we retain the value in a + * local variable for each dentry as we look at it so that we don't see + * the components of that value change under us */ + while (managed = ACCESS_ONCE(path->dentry->d_flags), + managed &= DCACHE_MANAGED_DENTRY, + unlikely(managed != 0)) { + /* Transit to a mounted filesystem. */ + if (managed & DCACHE_MOUNTED) { + struct vfsmount *mounted = lookup_mnt(path); + if (mounted) { + dput(path->dentry); + if (need_mntput) + mntput(path->mnt); + path->mnt = mounted; + path->dentry = dget(mounted->mnt_root); + need_mntput = true; + continue; + } + + /* Something is mounted on this dentry in another + * namespace and/or whatever was mounted there in this + * namespace got unmounted before we managed to get the + * vfsmount_lock */ + } + + /* Handle an automount point */ + if (managed & DCACHE_NEED_AUTOMOUNT) { + ret = follow_automount(path, flags, &need_mntput); + if (ret < 0) + return ret == -EISDIR ? 0 : ret; + continue; + } + + /* We didn't change the current path point */ + break; } + return 0; } int follow_down(struct path *path) @@ -958,13 +1027,37 @@ int follow_down(struct path *path) return 0; } +/* + * Skip to top of mountpoint pile in rcuwalk mode. We abort the rcu-walk if we + * meet an automount point and we're not walking to "..". True is returned to + * continue, false to abort. + */ +static bool __follow_mount_rcu(struct nameidata *nd, struct path *path, + struct inode **inode, bool reverse_transit) +{ + while (d_mountpoint(path->dentry)) { + struct vfsmount *mounted; + mounted = __lookup_mnt(path->mnt, path->dentry, 1); + if (!mounted) + break; + path->mnt = mounted; + path->dentry = mounted->mnt_root; + nd->seq = read_seqcount_begin(&path->dentry->d_seq); + *inode = path->dentry->d_inode; + } + + if (unlikely(path->dentry->d_flags & DCACHE_NEED_AUTOMOUNT)) + return reverse_transit; + return true; +} + static int follow_dotdot_rcu(struct nameidata *nd) { struct inode *inode = nd->inode; set_root_rcu(nd); - while(1) { + while (1) { if (nd->path.dentry == nd->root.dentry && nd->path.mnt == nd->root.mnt) { break; @@ -987,12 +1080,28 @@ static int follow_dotdot_rcu(struct nameidata *nd) nd->seq = read_seqcount_begin(&nd->path.dentry->d_seq); inode = nd->path.dentry->d_inode; } - __follow_mount_rcu(nd, &nd->path, &inode); + __follow_mount_rcu(nd, &nd->path, &inode, true); nd->inode = inode; return 0; } +/* + * Skip to top of mountpoint pile in refwalk mode for follow_dotdot() + */ +static void follow_mount(struct path *path) +{ + while (d_mountpoint(path->dentry)) { + struct vfsmount *mounted = lookup_mnt(path); + if (!mounted) + break; + dput(path->dentry); + mntput(path->mnt); + path->mnt = mounted; + path->dentry = dget(mounted->mnt_root); + } +} + static void follow_dotdot(struct nameidata *nd) { set_root(nd); @@ -1057,12 +1166,14 @@ static int do_lookup(struct nameidata *nd, struct qstr *name, struct vfsmount *mnt = nd->path.mnt; struct dentry *dentry, *parent = nd->path.dentry; struct inode *dir; + int err; + /* * See if the low-level filesystem might want * to use its own hash.. */ if (unlikely(parent->d_flags & DCACHE_OP_HASH)) { - int err = parent->d_op->d_hash(parent, nd->inode, name); + err = parent->d_op->d_hash(parent, nd->inode, name); if (err < 0) return err; } @@ -1092,20 +1203,25 @@ static int do_lookup(struct nameidata *nd, struct qstr *name, done2: path->mnt = mnt; path->dentry = dentry; - __follow_mount_rcu(nd, path, inode); - } else { - dentry = __d_lookup(parent, name); - if (!dentry) - goto need_lookup; + if (likely(__follow_mount_rcu(nd, path, inode, false))) + return 0; + if (nameidata_drop_rcu(nd)) + return -ECHILD; + /* fallthru */ + } + dentry = __d_lookup(parent, name); + if (!dentry) + goto need_lookup; found: - if (dentry->d_flags & DCACHE_OP_REVALIDATE) - goto need_revalidate; + if (dentry->d_flags & DCACHE_OP_REVALIDATE) + goto need_revalidate; done: - path->mnt = mnt; - path->dentry = dentry; - __follow_mount(path); - *inode = path->dentry->d_inode; - } + path->mnt = mnt; + path->dentry = dentry; + err = follow_managed(path, nd->flags); + if (unlikely(err < 0)) + return err; + *inode = path->dentry->d_inode; return 0; need_lookup: @@ -2203,11 +2319,9 @@ static struct file *do_last(struct nameidata *nd, struct path *path, if (open_flag & O_EXCL) goto exit_dput; - if (__follow_mount(path)) { - error = -ELOOP; - if (open_flag & O_NOFOLLOW) - goto exit_dput; - } + error = follow_managed(path, nd->flags); + if (error < 0) + goto exit_dput; error = -ENOENT; if (!path->dentry->d_inode) diff --git a/include/linux/dcache.h b/include/linux/dcache.h index 59fcd24b1468..ee6c26d142c3 100644 --- a/include/linux/dcache.h +++ b/include/linux/dcache.h @@ -167,6 +167,7 @@ struct dentry_operations { void (*d_release)(struct dentry *); void (*d_iput)(struct dentry *, struct inode *); char *(*d_dname)(struct dentry *, char *, int); + struct vfsmount *(*d_automount)(struct path *); } ____cacheline_aligned; /* @@ -205,13 +206,17 @@ struct dentry_operations { #define DCACHE_CANT_MOUNT 0x0100 #define DCACHE_GENOCIDE 0x0200 -#define DCACHE_MOUNTED 0x0400 /* is a mountpoint */ #define DCACHE_OP_HASH 0x1000 #define DCACHE_OP_COMPARE 0x2000 #define DCACHE_OP_REVALIDATE 0x4000 #define DCACHE_OP_DELETE 0x8000 +#define DCACHE_MOUNTED 0x10000 /* is a mountpoint */ +#define DCACHE_NEED_AUTOMOUNT 0x20000 /* handle automount on this dir */ +#define DCACHE_MANAGED_DENTRY \ + (DCACHE_MOUNTED|DCACHE_NEED_AUTOMOUNT) + extern seqlock_t rename_lock; static inline int dname_external(struct dentry *dentry) diff --git a/include/linux/fs.h b/include/linux/fs.h index 3984f2358d1f..4c00ed53be8f 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -242,6 +242,7 @@ struct inodes_stat_t { #define S_SWAPFILE 256 /* Do not truncate: swapon got its bmaps */ #define S_PRIVATE 512 /* Inode is fs-internal */ #define S_IMA 1024 /* Inode has an associated IMA struct */ +#define S_AUTOMOUNT 2048 /* Automount/referral quasi-directory */ /* * Note that nosuid etc flags are inode-specific: setting some file-system @@ -277,6 +278,7 @@ struct inodes_stat_t { #define IS_SWAPFILE(inode) ((inode)->i_flags & S_SWAPFILE) #define IS_PRIVATE(inode) ((inode)->i_flags & S_PRIVATE) #define IS_IMA(inode) ((inode)->i_flags & S_IMA) +#define IS_AUTOMOUNT(inode) ((inode)->i_flags & S_AUTOMOUNT) /* the read-only stuff doesn't really belong here, but any other place is probably as bad and I don't want to create yet another include file. */ -- cgit v1.2.3 From cc53ce53c86924bfe98a12ea20b7465038a08792 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 14 Jan 2011 18:45:26 +0000 Subject: Add a dentry op to allow processes to be held during pathwalk transit Add a dentry op (d_manage) to permit a filesystem to hold a process and make it sleep when it tries to transit away from one of that filesystem's directories during a pathwalk. The operation is keyed off a new dentry flag (DCACHE_MANAGE_TRANSIT). The filesystem is allowed to be selective about which processes it holds and which it permits to continue on or prohibits from transiting from each flagged directory. This will allow autofs to hold up client processes whilst letting its userspace daemon through to maintain the directory or the stuff behind it or mounted upon it. The ->d_manage() dentry operation: int (*d_manage)(struct path *path, bool mounting_here); takes a pointer to the directory about to be transited away from and a flag indicating whether the transit is undertaken by do_add_mount() or do_move_mount() skipping through a pile of filesystems mounted on a mountpoint. It should return 0 if successful and to let the process continue on its way; -EISDIR to prohibit the caller from skipping to overmounted filesystems or automounting, and to use this directory; or some other error code to return to the user. ->d_manage() is called with namespace_sem writelocked if mounting_here is true and no other locks held, so it may sleep. However, if mounting_here is true, it may not initiate or wait for a mount or unmount upon the parameter directory, even if the act is actually performed by userspace. Within fs/namei.c, follow_managed() is extended to check with d_manage() first on each managed directory, before transiting away from it or attempting to automount upon it. follow_down() is renamed follow_down_one() and should only be used where the filesystem deliberately intends to avoid management steps (e.g. autofs). A new follow_down() is added that incorporates the loop done by all other callers of follow_down() (do_add/move_mount(), autofs and NFSD; whilst AFS, NFS and CIFS do use it, their use is removed by converting them to use d_automount()). The new follow_down() calls d_manage() as appropriate. It also takes an extra parameter to indicate if it is being called from mount code (with namespace_sem writelocked) which it passes to d_manage(). follow_down() ignores automount points so that it can be used to mount on them. __follow_mount_rcu() is made to abort rcu-walk mode if it hits a directory with DCACHE_MANAGE_TRANSIT set on the basis that we're probably going to have to sleep. It would be possible to enter d_manage() in rcu-walk mode too, and have that determine whether to abort or not itself. That would allow the autofs daemon to continue on in rcu-walk mode. Note that DCACHE_MANAGE_TRANSIT on a directory should be cleared when it isn't required as every tranist from that directory will cause d_manage() to be invoked. It can always be set again when necessary. ========================== WHAT THIS MEANS FOR AUTOFS ========================== Autofs currently uses the lookup() inode op and the d_revalidate() dentry op to trigger the automounting of indirect mounts, and both of these can be called with i_mutex held. autofs knows that the i_mutex will be held by the caller in lookup(), and so can drop it before invoking the daemon - but this isn't so for d_revalidate(), since the lock is only held on _some_ of the code paths that call it. This means that autofs can't risk dropping i_mutex from its d_revalidate() function before it calls the daemon. The bug could manifest itself as, for example, a process that's trying to validate an automount dentry that gets made to wait because that dentry is expired and needs cleaning up: mkdir S ffffffff8014e05a 0 32580 24956 Call Trace: [] :autofs4:autofs4_wait+0x674/0x897 [] avc_has_perm+0x46/0x58 [] autoremove_wake_function+0x0/0x2e [] :autofs4:autofs4_expire_wait+0x41/0x6b [] :autofs4:autofs4_revalidate+0x91/0x149 [] __lookup_hash+0xa0/0x12f [] lookup_create+0x46/0x80 [] sys_mkdirat+0x56/0xe4 versus the automount daemon which wants to remove that dentry, but can't because the normal process is holding the i_mutex lock: automount D ffffffff8014e05a 0 32581 1 32561 Call Trace: [] __mutex_lock_slowpath+0x60/0x9b [] do_path_lookup+0x2ca/0x2f1 [] .text.lock.mutex+0xf/0x14 [] do_rmdir+0x77/0xde [] tracesys+0x71/0xe0 [] tracesys+0xd5/0xe0 which means that the system is deadlocked. This patch allows autofs to hold up normal processes whilst the daemon goes ahead and does things to the dentry tree behind the automouter point without risking a deadlock as almost no locks are held in d_manage() and none in d_automount(). Signed-off-by: David Howells Was-Acked-by: Ian Kent Signed-off-by: Al Viro --- Documentation/filesystems/Locking | 2 ++ Documentation/filesystems/vfs.txt | 21 +++++++++++- drivers/staging/autofs/dirhash.c | 5 ++- fs/afs/mntpt.c | 5 +-- fs/autofs4/autofs_i.h | 13 ------- fs/autofs4/dev-ioctl.c | 2 +- fs/autofs4/expire.c | 2 +- fs/autofs4/root.c | 11 +++--- fs/cifs/cifs_dfs_ref.c | 5 +-- fs/namei.c | 72 +++++++++++++++++++++++++++++++++++++-- fs/namespace.c | 14 ++++---- fs/nfs/namespace.c | 5 +-- fs/nfsd/vfs.c | 5 +-- include/linux/dcache.h | 11 ++++-- include/linux/namei.h | 3 +- 15 files changed, 126 insertions(+), 50 deletions(-) (limited to 'include/linux') diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking index 5f0c52a07386..cbf98b989b11 100644 --- a/Documentation/filesystems/Locking +++ b/Documentation/filesystems/Locking @@ -20,6 +20,7 @@ prototypes: void (*d_iput)(struct dentry *, struct inode *); char *(*d_dname)((struct dentry *dentry, char *buffer, int buflen); struct vfsmount *(*d_automount)(struct path *path); + int (*d_manage)(struct dentry *, bool); locking rules: rename_lock ->d_lock may block rcu-walk @@ -31,6 +32,7 @@ d_release: no no yes no d_iput: no no yes no d_dname: no no no no d_automount: no no yes no +d_manage: no no yes no --------------------------- inode_operations --------------------------- prototypes: diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt index 726a4f6fa3c9..4682586b147a 100644 --- a/Documentation/filesystems/vfs.txt +++ b/Documentation/filesystems/vfs.txt @@ -865,6 +865,7 @@ struct dentry_operations { void (*d_iput)(struct dentry *, struct inode *); char *(*d_dname)(struct dentry *, char *, int); struct vfsmount *(*d_automount)(struct path *); + int (*d_manage)(struct dentry *, bool); }; d_revalidate: called when the VFS needs to revalidate a dentry. This @@ -938,12 +939,30 @@ struct dentry_operations { target and the parent VFS mount record to provide inheritable mount parameters. NULL should be returned if someone else managed to make the automount first. If the automount failed, then an error code - should be returned. + should be returned. If -EISDIR is returned, then the directory will + be treated as an ordinary directory and returned to pathwalk to + continue walking. This function is only used if DCACHE_NEED_AUTOMOUNT is set on the dentry. This is set by __d_instantiate() if S_AUTOMOUNT is set on the inode being added. + d_manage: called to allow the filesystem to manage the transition from a + dentry (optional). This allows autofs, for example, to hold up clients + waiting to explore behind a 'mountpoint' whilst letting the daemon go + past and construct the subtree there. 0 should be returned to let the + calling process continue. -EISDIR can be returned to tell pathwalk to + use this directory as an ordinary directory and to ignore anything + mounted on it and not to check the automount flag. Any other error + code will abort pathwalk completely. + + If the 'mounting_here' parameter is true, then namespace_sem is being + held by the caller and the function should not initiate any mounts or + unmounts that it will then wait for. + + This function is only used if DCACHE_MANAGE_TRANSIT is set on the + dentry being transited from. + Example : static char *pipefs_dname(struct dentry *dent, char *buffer, int buflen) diff --git a/drivers/staging/autofs/dirhash.c b/drivers/staging/autofs/dirhash.c index d3f42c8325f7..a08bd7355035 100644 --- a/drivers/staging/autofs/dirhash.c +++ b/drivers/staging/autofs/dirhash.c @@ -88,14 +88,13 @@ struct autofs_dir_ent *autofs_expire(struct super_block *sb, } path.mnt = mnt; path_get(&path); - if (!follow_down(&path)) { + if (!follow_down_one(&path)) { path_put(&path); DPRINTK(("autofs: not expirable\ (not a mounted directory): %s\n", ent->name)); continue; } - while (d_mountpoint(path.dentry) && follow_down(&path)) - ; + follow_down(&path, false); // TODO: need to check error umount_ok = may_umount(path.mnt); path_put(&path); diff --git a/fs/afs/mntpt.c b/fs/afs/mntpt.c index e83c0336e7b5..f3e891d57a2c 100644 --- a/fs/afs/mntpt.c +++ b/fs/afs/mntpt.c @@ -273,10 +273,7 @@ static void *afs_mntpt_follow_link(struct dentry *dentry, struct nameidata *nd) break; case -EBUSY: /* someone else made a mount here whilst we were busy */ - while (d_mountpoint(nd->path.dentry) && - follow_down(&nd->path)) - ; - err = 0; + err = follow_down(&nd->path, false); default: mntput(newmnt); break; diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h index 0fffe1c24cec..eb67953452bb 100644 --- a/fs/autofs4/autofs_i.h +++ b/fs/autofs4/autofs_i.h @@ -229,19 +229,6 @@ int autofs4_wait(struct autofs_sb_info *,struct dentry *, enum autofs_notify); int autofs4_wait_release(struct autofs_sb_info *,autofs_wqt_t,int); void autofs4_catatonic_mode(struct autofs_sb_info *); -static inline int autofs4_follow_mount(struct path *path) -{ - int res = 0; - - while (d_mountpoint(path->dentry)) { - int followed = follow_down(path); - if (!followed) - break; - res = 1; - } - return res; -} - static inline u32 autofs4_get_dev(struct autofs_sb_info *sbi) { return new_encode_dev(sbi->sb->s_dev); diff --git a/fs/autofs4/dev-ioctl.c b/fs/autofs4/dev-ioctl.c index eff9a419469a..1442da4860e5 100644 --- a/fs/autofs4/dev-ioctl.c +++ b/fs/autofs4/dev-ioctl.c @@ -551,7 +551,7 @@ static int autofs_dev_ioctl_ismountpoint(struct file *fp, err = have_submounts(path.dentry); - if (follow_down(&path)) + if (follow_down_one(&path)) magic = path.mnt->mnt_sb->s_magic; } diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c index cc1d01365905..6a930b90d389 100644 --- a/fs/autofs4/expire.c +++ b/fs/autofs4/expire.c @@ -56,7 +56,7 @@ static int autofs4_mount_busy(struct vfsmount *mnt, struct dentry *dentry) path_get(&path); - if (!follow_down(&path)) + if (!follow_down_one(&path)) goto done; if (is_autofs4_dentry(path.dentry)) { diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index 651e4ef563b1..20225636a4e9 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -234,7 +234,7 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd) nd->flags); /* * For an expire of a covered direct or offset mount we need - * to break out of follow_down() at the autofs mount trigger + * to break out of follow_down_one() at the autofs mount trigger * (d_mounted--), so we can see the expiring flag, and manage * the blocking and following here until the expire is completed. */ @@ -243,7 +243,7 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd) if (ino->flags & AUTOFS_INF_EXPIRING) { spin_unlock(&sbi->fs_lock); /* Follow down to our covering mount. */ - if (!follow_down(&nd->path)) + if (!follow_down_one(&nd->path)) goto done; goto follow; } @@ -292,11 +292,10 @@ follow: * multi-mount with no root offset so we don't need * to follow it. */ - if (d_mountpoint(dentry)) { - if (!autofs4_follow_mount(&nd->path)) { - status = -ENOENT; + if (d_managed(dentry)) { + status = follow_down(&nd->path, false); + if (status < 0) goto out_error; - } } done: diff --git a/fs/cifs/cifs_dfs_ref.c b/fs/cifs/cifs_dfs_ref.c index c68a056f27fd..83479cf63f96 100644 --- a/fs/cifs/cifs_dfs_ref.c +++ b/fs/cifs/cifs_dfs_ref.c @@ -273,10 +273,7 @@ static int add_mount_helper(struct vfsmount *newmnt, struct nameidata *nd, break; case -EBUSY: /* someone else made a mount here whilst we were busy */ - while (d_mountpoint(nd->path.dentry) && - follow_down(&nd->path)) - ; - err = 0; + err = follow_down(&nd->path, false); default: mntput(newmnt); break; diff --git a/fs/namei.c b/fs/namei.c index 16109da68bbf..9d3033dc22e9 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -960,6 +960,7 @@ static int follow_automount(struct path *path, unsigned flags, /* * Handle a dentry that is managed in some way. + * - Flagged for transit management (autofs) * - Flagged as mountpoint * - Flagged as automount point * @@ -979,6 +980,16 @@ static int follow_managed(struct path *path, unsigned flags) while (managed = ACCESS_ONCE(path->dentry->d_flags), managed &= DCACHE_MANAGED_DENTRY, unlikely(managed != 0)) { + /* Allow the filesystem to manage the transit without i_mutex + * being held. */ + if (managed & DCACHE_MANAGE_TRANSIT) { + BUG_ON(!path->dentry->d_op); + BUG_ON(!path->dentry->d_op->d_manage); + ret = path->dentry->d_op->d_manage(path->dentry, false); + if (ret < 0) + return ret == -EISDIR ? 0 : ret; + } + /* Transit to a mounted filesystem. */ if (managed & DCACHE_MOUNTED) { struct vfsmount *mounted = lookup_mnt(path); @@ -1012,7 +1023,7 @@ static int follow_managed(struct path *path, unsigned flags) return 0; } -int follow_down(struct path *path) +int follow_down_one(struct path *path) { struct vfsmount *mounted; @@ -1029,14 +1040,19 @@ int follow_down(struct path *path) /* * Skip to top of mountpoint pile in rcuwalk mode. We abort the rcu-walk if we - * meet an automount point and we're not walking to "..". True is returned to + * meet a managed dentry and we're not walking to "..". True is returned to * continue, false to abort. */ static bool __follow_mount_rcu(struct nameidata *nd, struct path *path, struct inode **inode, bool reverse_transit) { + unsigned abort_mask = + reverse_transit ? 0 : DCACHE_MANAGE_TRANSIT; + while (d_mountpoint(path->dentry)) { struct vfsmount *mounted; + if (path->dentry->d_flags & abort_mask) + return true; mounted = __lookup_mnt(path->mnt, path->dentry, 1); if (!mounted) break; @@ -1086,6 +1102,57 @@ static int follow_dotdot_rcu(struct nameidata *nd) return 0; } +/* + * Follow down to the covering mount currently visible to userspace. At each + * point, the filesystem owning that dentry may be queried as to whether the + * caller is permitted to proceed or not. + * + * Care must be taken as namespace_sem may be held (indicated by mounting_here + * being true). + */ +int follow_down(struct path *path, bool mounting_here) +{ + unsigned managed; + int ret; + + while (managed = ACCESS_ONCE(path->dentry->d_flags), + unlikely(managed & DCACHE_MANAGED_DENTRY)) { + /* Allow the filesystem to manage the transit without i_mutex + * being held. + * + * We indicate to the filesystem if someone is trying to mount + * something here. This gives autofs the chance to deny anyone + * other than its daemon the right to mount on its + * superstructure. + * + * The filesystem may sleep at this point. + */ + if (managed & DCACHE_MANAGE_TRANSIT) { + BUG_ON(!path->dentry->d_op); + BUG_ON(!path->dentry->d_op->d_manage); + ret = path->dentry->d_op->d_manage(path->dentry, mounting_here); + if (ret < 0) + return ret == -EISDIR ? 0 : ret; + } + + /* Transit to a mounted filesystem. */ + if (managed & DCACHE_MOUNTED) { + struct vfsmount *mounted = lookup_mnt(path); + if (!mounted) + break; + dput(path->dentry); + mntput(path->mnt); + path->mnt = mounted; + path->dentry = dget(mounted->mnt_root); + continue; + } + + /* Don't handle automount points here */ + break; + } + return 0; +} + /* * Skip to top of mountpoint pile in refwalk mode for follow_dotdot() */ @@ -3530,6 +3597,7 @@ const struct inode_operations page_symlink_inode_operations = { }; EXPORT_SYMBOL(user_path_at); +EXPORT_SYMBOL(follow_down_one); EXPORT_SYMBOL(follow_down); EXPORT_SYMBOL(follow_up); EXPORT_SYMBOL(get_write_access); /* binfmt_aout */ diff --git a/fs/namespace.c b/fs/namespace.c index 3ddfd9046c44..d94ccd6ddafd 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1844,9 +1844,10 @@ static int do_move_mount(struct path *path, char *old_name) return err; down_write(&namespace_sem); - while (d_mountpoint(path->dentry) && - follow_down(path)) - ; + err = follow_down(path, true); + if (err < 0) + goto out; + err = -EINVAL; if (!check_mnt(path->mnt) || !check_mnt(old_path.mnt)) goto out; @@ -1940,9 +1941,10 @@ int do_add_mount(struct vfsmount *newmnt, struct path *path, down_write(&namespace_sem); /* Something was mounted here while we slept */ - while (d_mountpoint(path->dentry) && - follow_down(path)) - ; + err = follow_down(path, true); + if (err < 0) + goto unlock; + err = -EINVAL; if (!(mnt_flags & MNT_SHRINKABLE) && !check_mnt(path->mnt)) goto unlock; diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c index 74aaf3963c10..bfcb933e5755 100644 --- a/fs/nfs/namespace.c +++ b/fs/nfs/namespace.c @@ -176,10 +176,7 @@ out_err: path_put(&nd->path); goto out; out_follow: - while (d_mountpoint(nd->path.dentry) && - follow_down(&nd->path)) - ; - err = 0; + err = follow_down(&nd->path, false); goto out; } diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 230b79fbf005..0f79e33a65d7 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -88,8 +88,9 @@ nfsd_cross_mnt(struct svc_rqst *rqstp, struct dentry **dpp, .dentry = dget(dentry)}; int err = 0; - while (d_mountpoint(path.dentry) && follow_down(&path)) - ; + err = follow_down(&path, false); + if (err < 0) + goto out; exp2 = rqst_exp_get_by_name(rqstp, &path); if (IS_ERR(exp2)) { diff --git a/include/linux/dcache.h b/include/linux/dcache.h index ee6c26d142c3..1a87760d6532 100644 --- a/include/linux/dcache.h +++ b/include/linux/dcache.h @@ -168,6 +168,7 @@ struct dentry_operations { void (*d_iput)(struct dentry *, struct inode *); char *(*d_dname)(struct dentry *, char *, int); struct vfsmount *(*d_automount)(struct path *); + int (*d_manage)(struct dentry *, bool); } ____cacheline_aligned; /* @@ -214,8 +215,9 @@ struct dentry_operations { #define DCACHE_MOUNTED 0x10000 /* is a mountpoint */ #define DCACHE_NEED_AUTOMOUNT 0x20000 /* handle automount on this dir */ +#define DCACHE_MANAGE_TRANSIT 0x40000 /* manage transit from this dirent */ #define DCACHE_MANAGED_DENTRY \ - (DCACHE_MOUNTED|DCACHE_NEED_AUTOMOUNT) + (DCACHE_MOUNTED|DCACHE_NEED_AUTOMOUNT|DCACHE_MANAGE_TRANSIT) extern seqlock_t rename_lock; @@ -404,7 +406,12 @@ static inline void dont_mount(struct dentry *dentry) extern void dput(struct dentry *); -static inline int d_mountpoint(struct dentry *dentry) +static inline bool d_managed(struct dentry *dentry) +{ + return dentry->d_flags & DCACHE_MANAGED_DENTRY; +} + +static inline bool d_mountpoint(struct dentry *dentry) { return dentry->d_flags & DCACHE_MOUNTED; } diff --git a/include/linux/namei.h b/include/linux/namei.h index 18d06add0a40..8ef2c789c2a8 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -79,7 +79,8 @@ extern struct file *lookup_instantiate_filp(struct nameidata *nd, struct dentry extern struct dentry *lookup_one_len(const char *, struct dentry *, int); -extern int follow_down(struct path *); +extern int follow_down_one(struct path *); +extern int follow_down(struct path *, bool); extern int follow_up(struct path *); extern struct dentry *lock_rename(struct dentry *, struct dentry *); -- cgit v1.2.3 From 6f45b65672c8017d5e210e338bb5858a938ef445 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 14 Jan 2011 18:45:31 +0000 Subject: Add an AT_NO_AUTOMOUNT flag to suppress terminal automount Add an AT_NO_AUTOMOUNT flag to suppress terminal automounting of automount point directories. This can be used by fstatat() users to permit the gathering of attributes on an automount point and also prevent mass-automounting of a directory of automount points by ls. Signed-off-by: David Howells Acked-by: Ian Kent Signed-off-by: Al Viro --- fs/namei.c | 6 ++++++ fs/stat.c | 4 +++- include/linux/fcntl.h | 1 + include/linux/namei.h | 2 ++ 4 files changed, 12 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/fs/namei.c b/fs/namei.c index 9d3033dc22e9..dc50bfb2f5d6 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -908,6 +908,12 @@ static int follow_automount(struct path *path, unsigned flags, if (!path->dentry->d_op || !path->dentry->d_op->d_automount) return -EREMOTE; + /* We don't want to mount if someone supplied AT_NO_AUTOMOUNT + * and this is the terminal part of the path. + */ + if ((flags & LOOKUP_NO_AUTOMOUNT) && !(flags & LOOKUP_CONTINUE)) + return -EISDIR; /* we actually want to stop here */ + /* We want to mount if someone is trying to open/create a file of any * type under the mountpoint, wants to traverse through the mountpoint * or wants to open the mounted directory. diff --git a/fs/stat.c b/fs/stat.c index 12e90e213900..d5c61cf2b703 100644 --- a/fs/stat.c +++ b/fs/stat.c @@ -75,11 +75,13 @@ int vfs_fstatat(int dfd, const char __user *filename, struct kstat *stat, int error = -EINVAL; int lookup_flags = 0; - if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0) + if ((flag & ~(AT_SYMLINK_NOFOLLOW | AT_NO_AUTOMOUNT)) != 0) goto out; if (!(flag & AT_SYMLINK_NOFOLLOW)) lookup_flags |= LOOKUP_FOLLOW; + if (flag & AT_NO_AUTOMOUNT) + lookup_flags |= LOOKUP_NO_AUTOMOUNT; error = user_path_at(dfd, filename, lookup_flags, &path); if (error) diff --git a/include/linux/fcntl.h b/include/linux/fcntl.h index afc00af3229b..a562fa5fb4e3 100644 --- a/include/linux/fcntl.h +++ b/include/linux/fcntl.h @@ -45,6 +45,7 @@ #define AT_REMOVEDIR 0x200 /* Remove directory instead of unlinking file. */ #define AT_SYMLINK_FOLLOW 0x400 /* Follow symbolic links. */ +#define AT_NO_AUTOMOUNT 0x800 /* Suppress terminal automount traversal */ #ifdef __KERNEL__ diff --git a/include/linux/namei.h b/include/linux/namei.h index 8ef2c789c2a8..f276d4fa01fc 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -45,6 +45,7 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT, LAST_BIND}; * - ending slashes ok even for nonexistent files * - internal "there are more path components" flag * - dentry cache is untrusted; force a real lookup + * - suppress terminal automount */ #define LOOKUP_FOLLOW 0x0001 #define LOOKUP_DIRECTORY 0x0002 @@ -53,6 +54,7 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT, LAST_BIND}; #define LOOKUP_PARENT 0x0010 #define LOOKUP_REVAL 0x0020 #define LOOKUP_RCU 0x0040 +#define LOOKUP_NO_AUTOMOUNT 0x0080 /* * Intent data */ -- cgit v1.2.3 From 36d43a43761b004ad1879ac21471d8fc5f3157ec Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 14 Jan 2011 18:45:42 +0000 Subject: NFS: Use d_automount() rather than abusing follow_link() Make NFS use the new d_automount() dentry operation rather than abusing follow_link() on directories. Signed-off-by: David Howells Acked-by: Trond Myklebust Acked-by: Ian Kent Signed-off-by: Al Viro --- fs/nfs/dir.c | 4 ++- fs/nfs/inode.c | 4 +-- fs/nfs/internal.h | 1 + fs/nfs/namespace.c | 84 ++++++++++++++++++++++++-------------------------- include/linux/nfs_fs.h | 1 - 5 files changed, 46 insertions(+), 48 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index df8c03a02161..2c3eb33b904d 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -970,7 +970,7 @@ int nfs_lookup_verify_inode(struct inode *inode, struct nameidata *nd) { struct nfs_server *server = NFS_SERVER(inode); - if (test_bit(NFS_INO_MOUNTPOINT, &NFS_I(inode)->flags)) + if (IS_AUTOMOUNT(inode)) return 0; if (nd != NULL) { /* VFS wants an on-the-wire revalidation */ @@ -1173,6 +1173,7 @@ const struct dentry_operations nfs_dentry_operations = { .d_revalidate = nfs_lookup_revalidate, .d_delete = nfs_dentry_delete, .d_iput = nfs_dentry_iput, + .d_automount = nfs_d_automount, }; static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, struct nameidata *nd) @@ -1246,6 +1247,7 @@ const struct dentry_operations nfs4_dentry_operations = { .d_revalidate = nfs_open_revalidate, .d_delete = nfs_dentry_delete, .d_iput = nfs_dentry_iput, + .d_automount = nfs_d_automount, }; /* diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index ce00b704452c..d8512423ba72 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -300,7 +300,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr) else inode->i_op = &nfs_mountpoint_inode_operations; inode->i_fop = NULL; - set_bit(NFS_INO_MOUNTPOINT, &nfsi->flags); + inode->i_flags |= S_AUTOMOUNT; } } else if (S_ISLNK(inode->i_mode)) inode->i_op = &nfs_symlink_inode_operations; @@ -1208,7 +1208,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) /* Update the fsid? */ if (S_ISDIR(inode->i_mode) && (fattr->valid & NFS_ATTR_FATTR_FSID) && !nfs_fsid_equal(&server->fsid, &fattr->fsid) && - !test_bit(NFS_INO_MOUNTPOINT, &nfsi->flags)) + !IS_AUTOMOUNT(inode)) server->fsid = fattr->fsid; /* diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index bfa3a34af801..4644f04b4b46 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -252,6 +252,7 @@ extern char *nfs_path(const char *base, const struct dentry *droot, const struct dentry *dentry, char *buffer, ssize_t buflen); +extern struct vfsmount *nfs_d_automount(struct path *path); /* getroot.c */ extern struct dentry *nfs_get_root(struct super_block *, struct nfs_fh *); diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c index bfcb933e5755..f3fbb1bf3f18 100644 --- a/fs/nfs/namespace.c +++ b/fs/nfs/namespace.c @@ -97,9 +97,8 @@ Elong: } /* - * nfs_follow_mountpoint - handle crossing a mountpoint on the server - * @dentry - dentry of mountpoint - * @nd - nameidata info + * nfs_d_automount - Handle crossing a mountpoint on the server + * @path - The mountpoint * * When we encounter a mountpoint on the server, we want to set up * a mountpoint on the client too, to prevent inode numbers from @@ -109,84 +108,81 @@ Elong: * situation, and that different filesystems may want to use * different security flavours. */ -static void * nfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd) +struct vfsmount *nfs_d_automount(struct path *path) { struct vfsmount *mnt; - struct nfs_server *server = NFS_SERVER(dentry->d_inode); + struct nfs_server *server = NFS_SERVER(path->dentry->d_inode); struct dentry *parent; struct nfs_fh *fh = NULL; struct nfs_fattr *fattr = NULL; int err; - dprintk("--> nfs_follow_mountpoint()\n"); + dprintk("--> nfs_d_automount()\n"); - err = -ESTALE; - if (IS_ROOT(dentry)) - goto out_err; + mnt = ERR_PTR(-ESTALE); + if (IS_ROOT(path->dentry)) + goto out_nofree; - err = -ENOMEM; + mnt = ERR_PTR(-ENOMEM); fh = nfs_alloc_fhandle(); fattr = nfs_alloc_fattr(); if (fh == NULL || fattr == NULL) - goto out_err; + goto out; dprintk("%s: enter\n", __func__); - dput(nd->path.dentry); - nd->path.dentry = dget(dentry); - /* Look it up again */ - parent = dget_parent(nd->path.dentry); + /* Look it up again to get its attributes */ + parent = dget_parent(path->dentry); err = server->nfs_client->rpc_ops->lookup(parent->d_inode, - &nd->path.dentry->d_name, + &path->dentry->d_name, fh, fattr); dput(parent); - if (err != 0) - goto out_err; + if (err != 0) { + mnt = ERR_PTR(err); + goto out; + } if (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) - mnt = nfs_do_refmount(nd->path.mnt, nd->path.dentry); + mnt = nfs_do_refmount(path->mnt, path->dentry); else - mnt = nfs_do_submount(nd->path.mnt, nd->path.dentry, fh, - fattr); - err = PTR_ERR(mnt); + mnt = nfs_do_submount(path->mnt, path->dentry, fh, fattr); if (IS_ERR(mnt)) - goto out_err; + goto out; mntget(mnt); - err = do_add_mount(mnt, &nd->path, nd->path.mnt->mnt_flags|MNT_SHRINKABLE, + err = do_add_mount(mnt, path, path->mnt->mnt_flags | MNT_SHRINKABLE, &nfs_automount_list); - if (err < 0) { + switch (err) { + case 0: + dprintk("%s: done, success\n", __func__); + schedule_delayed_work(&nfs_automount_task, nfs_mountpoint_expiry_timeout); + break; + case -EBUSY: + /* someone else made a mount here whilst we were busy */ mntput(mnt); - if (err == -EBUSY) - goto out_follow; - goto out_err; + dprintk("%s: done, collision\n", __func__); + mnt = NULL; + break; + default: + mntput(mnt); + dprintk("%s: done, error %d\n", __func__, err); + mnt = ERR_PTR(err); + break; } - path_put(&nd->path); - nd->path.mnt = mnt; - nd->path.dentry = dget(mnt->mnt_root); - schedule_delayed_work(&nfs_automount_task, nfs_mountpoint_expiry_timeout); + out: nfs_free_fattr(fattr); nfs_free_fhandle(fh); - dprintk("%s: done, returned %d\n", __func__, err); - - dprintk("<-- nfs_follow_mountpoint() = %d\n", err); - return ERR_PTR(err); -out_err: - path_put(&nd->path); - goto out; -out_follow: - err = follow_down(&nd->path, false); - goto out; +out_nofree: + dprintk("<-- nfs_follow_mountpoint() = %p\n", mnt); + return mnt; } const struct inode_operations nfs_mountpoint_inode_operations = { - .follow_link = nfs_follow_mountpoint, .getattr = nfs_getattr, }; const struct inode_operations nfs_referral_inode_operations = { - .follow_link = nfs_follow_mountpoint, }; static void nfs_expire_automounts(struct work_struct *work) diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 0779bb8f95be..6023efa9f5d9 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -215,7 +215,6 @@ struct nfs_inode { #define NFS_INO_ADVISE_RDPLUS (0) /* advise readdirplus */ #define NFS_INO_STALE (1) /* possible stale inode */ #define NFS_INO_ACL_LRU_SET (2) /* Inode is on the LRU list */ -#define NFS_INO_MOUNTPOINT (3) /* inode is remote mountpoint */ #define NFS_INO_FLUSHING (4) /* inode is flushing out data */ #define NFS_INO_FSCACHE (5) /* inode can be cached by FS-Cache */ #define NFS_INO_FSCACHE_LOCK (6) /* FS-Cache cookie management lock */ -- cgit v1.2.3 From 1972580bb4edea3ed6fe273b2ca72f44f10f8c86 Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Fri, 14 Jan 2011 18:46:40 +0000 Subject: autofs4: Bump version Increase the autofs module sub-version so we can tell what kernel implementation is being used from user space debug logging. Signed-off-by: Ian Kent Signed-off-by: David Howells Signed-off-by: Al Viro --- include/linux/auto_fs4.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/auto_fs4.h b/include/linux/auto_fs4.h index 8b49ac48a5b7..e02982fa2953 100644 --- a/include/linux/auto_fs4.h +++ b/include/linux/auto_fs4.h @@ -24,7 +24,7 @@ #define AUTOFS_MIN_PROTO_VERSION 3 #define AUTOFS_MAX_PROTO_VERSION 5 -#define AUTOFS_PROTO_SUBVERSION 1 +#define AUTOFS_PROTO_SUBVERSION 2 /* Mask for expire behaviour */ #define AUTOFS_EXP_IMMEDIATE 1 -- cgit v1.2.3 From ab90911ff90cdab59b31c045c3f0ae480d14f29d Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 14 Jan 2011 18:46:51 +0000 Subject: Allow d_manage() to be used in RCU-walk mode Allow d_manage() to be called from pathwalk when it is in RCU-walk mode as well as when it is in Ref-walk mode. This permits __follow_mount_rcu() to call d_manage() directly. d_manage() needs a parameter to indicate that it is in RCU-walk mode as it isn't allowed to sleep if in that mode (but should return -ECHILD instead). autofs4_d_manage() can then be set to retain RCU-walk mode if the daemon accesses it and otherwise request dropping back to ref-walk mode. Signed-off-by: David Howells Signed-off-by: Al Viro --- Documentation/filesystems/Locking | 2 +- Documentation/filesystems/vfs.txt | 7 ++++++- fs/autofs4/root.c | 8 ++++++-- fs/namei.c | 15 ++++++++------- include/linux/dcache.h | 2 +- 5 files changed, 22 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking index cbf98b989b11..39707748ed2d 100644 --- a/Documentation/filesystems/Locking +++ b/Documentation/filesystems/Locking @@ -32,7 +32,7 @@ d_release: no no yes no d_iput: no no yes no d_dname: no no no no d_automount: no no yes no -d_manage: no no yes no +d_manage: no no yes (ref-walk) maybe --------------------------- inode_operations --------------------------- prototypes: diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt index 4682586b147a..3c4b2f1b64d0 100644 --- a/Documentation/filesystems/vfs.txt +++ b/Documentation/filesystems/vfs.txt @@ -865,7 +865,7 @@ struct dentry_operations { void (*d_iput)(struct dentry *, struct inode *); char *(*d_dname)(struct dentry *, char *, int); struct vfsmount *(*d_automount)(struct path *); - int (*d_manage)(struct dentry *, bool); + int (*d_manage)(struct dentry *, bool, bool); }; d_revalidate: called when the VFS needs to revalidate a dentry. This @@ -960,6 +960,11 @@ struct dentry_operations { held by the caller and the function should not initiate any mounts or unmounts that it will then wait for. + If the 'rcu_walk' parameter is true, then the caller is doing a + pathwalk in RCU-walk mode. Sleeping is not permitted in this mode, + and the caller can be asked to leave it and call again by returing + -ECHILD. + This function is only used if DCACHE_MANAGE_TRANSIT is set on the dentry being transited from. diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index 9194e274f849..dbd95512808c 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -36,7 +36,7 @@ static long autofs4_root_compat_ioctl(struct file *,unsigned int,unsigned long); static int autofs4_dir_open(struct inode *inode, struct file *file); static struct dentry *autofs4_lookup(struct inode *,struct dentry *, struct nameidata *); static struct vfsmount *autofs4_d_automount(struct path *); -static int autofs4_d_manage(struct dentry *, bool); +static int autofs4_d_manage(struct dentry *, bool, bool); const struct file_operations autofs4_root_operations = { .open = dcache_dir_open, @@ -450,7 +450,7 @@ done: return NULL; } -int autofs4_d_manage(struct dentry *dentry, bool mounting_here) +int autofs4_d_manage(struct dentry *dentry, bool mounting_here, bool rcu_walk) { struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb); @@ -464,6 +464,10 @@ int autofs4_d_manage(struct dentry *dentry, bool mounting_here) return 0; } + /* We need to sleep, so we need pathwalk to be in ref-mode */ + if (rcu_walk) + return -ECHILD; + /* Wait for pending expires */ do_expire_wait(dentry); diff --git a/fs/namei.c b/fs/namei.c index 373852012713..5c89695ae1e4 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -987,7 +987,8 @@ static int follow_managed(struct path *path, unsigned flags) if (managed & DCACHE_MANAGE_TRANSIT) { BUG_ON(!path->dentry->d_op); BUG_ON(!path->dentry->d_op->d_manage); - ret = path->dentry->d_op->d_manage(path->dentry, false); + ret = path->dentry->d_op->d_manage(path->dentry, + false, false); if (ret < 0) return ret == -EISDIR ? 0 : ret; } @@ -1048,13 +1049,12 @@ int follow_down_one(struct path *path) static bool __follow_mount_rcu(struct nameidata *nd, struct path *path, struct inode **inode, bool reverse_transit) { - unsigned abort_mask = - reverse_transit ? 0 : DCACHE_MANAGE_TRANSIT; - while (d_mountpoint(path->dentry)) { struct vfsmount *mounted; - if (path->dentry->d_flags & abort_mask) - return true; + if (unlikely(path->dentry->d_flags & DCACHE_MANAGE_TRANSIT) && + !reverse_transit && + path->dentry->d_op->d_manage(path->dentry, false, true) < 0) + return false; mounted = __lookup_mnt(path->mnt, path->dentry, 1); if (!mounted) break; @@ -1132,7 +1132,8 @@ int follow_down(struct path *path, bool mounting_here) if (managed & DCACHE_MANAGE_TRANSIT) { BUG_ON(!path->dentry->d_op); BUG_ON(!path->dentry->d_op->d_manage); - ret = path->dentry->d_op->d_manage(path->dentry, mounting_here); + ret = path->dentry->d_op->d_manage( + path->dentry, mounting_here, false); if (ret < 0) return ret == -EISDIR ? 0 : ret; } diff --git a/include/linux/dcache.h b/include/linux/dcache.h index 1a87760d6532..f958c19e3ca5 100644 --- a/include/linux/dcache.h +++ b/include/linux/dcache.h @@ -168,7 +168,7 @@ struct dentry_operations { void (*d_iput)(struct dentry *, struct inode *); char *(*d_dname)(struct dentry *, char *, int); struct vfsmount *(*d_automount)(struct path *); - int (*d_manage)(struct dentry *, bool); + int (*d_manage)(struct dentry *, bool, bool); } ____cacheline_aligned; /* -- cgit v1.2.3 From ea5b778a8b98c85a87d66bf844904f9c3802b869 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 14 Jan 2011 19:10:03 +0000 Subject: Unexport do_add_mount() and add in follow_automount(), not ->d_automount() Unexport do_add_mount() and make ->d_automount() return the vfsmount to be added rather than calling do_add_mount() itself. follow_automount() will then do the addition. This slightly complicates things as ->d_automount() normally wants to add the new vfsmount to an expiration list and start an expiration timer. The problem with that is that the vfsmount will be deleted if it has a refcount of 1 and the timer will not repeat if the expiration list is empty. To this end, we require the vfsmount to be returned from d_automount() with a refcount of (at least) 2. One of these refs will be dropped unconditionally. In addition, follow_automount() must get a 3rd ref around the call to do_add_mount() lest it eat a ref and return an error, leaving the mount we have open to being expired as we would otherwise have only 1 ref on it. d_automount() should also add the the vfsmount to the expiration list (by calling mnt_set_expiry()) and start the expiration timer before returning, if this mechanism is to be used. The vfsmount will be unlinked from the expiration list by follow_automount() if do_add_mount() fails. This patch also fixes the call to do_add_mount() for AFS to propagate the mount flags from the parent vfsmount. Signed-off-by: David Howells Signed-off-by: Al Viro --- Documentation/filesystems/vfs.txt | 23 ++++++++++++--------- fs/afs/mntpt.c | 25 ++++++----------------- fs/cifs/cifs_dfs_ref.c | 26 ++++++------------------ fs/internal.h | 2 ++ fs/namei.c | 42 ++++++++++++++++++++++++++++++++------- fs/namespace.c | 41 ++++++++++++++++++++++++++++++-------- fs/nfs/namespace.c | 24 ++++------------------ include/linux/mount.h | 7 +------ 8 files changed, 101 insertions(+), 89 deletions(-) (limited to 'include/linux') diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt index 3c4b2f1b64d0..94cf97b901d7 100644 --- a/Documentation/filesystems/vfs.txt +++ b/Documentation/filesystems/vfs.txt @@ -933,15 +933,20 @@ struct dentry_operations { dynamic_dname() helper function is provided to take care of this. d_automount: called when an automount dentry is to be traversed (optional). - This should create a new VFS mount record, mount it on the directory - and return the record to the caller. The caller is supplied with a - path parameter giving the automount directory to describe the automount - target and the parent VFS mount record to provide inheritable mount - parameters. NULL should be returned if someone else managed to make - the automount first. If the automount failed, then an error code - should be returned. If -EISDIR is returned, then the directory will - be treated as an ordinary directory and returned to pathwalk to - continue walking. + This should create a new VFS mount record and return the record to the + caller. The caller is supplied with a path parameter giving the + automount directory to describe the automount target and the parent + VFS mount record to provide inheritable mount parameters. NULL should + be returned if someone else managed to make the automount first. If + the vfsmount creation failed, then an error code should be returned. + If -EISDIR is returned, then the directory will be treated as an + ordinary directory and returned to pathwalk to continue walking. + + If a vfsmount is returned, the caller will attempt to mount it on the + mountpoint and will remove the vfsmount from its expiration list in + the case of failure. The vfsmount should be returned with 2 refs on + it to prevent automatic expiration - the caller will clean up the + additional ref. This function is only used if DCACHE_NEED_AUTOMOUNT is set on the dentry. This is set by __d_instantiate() if S_AUTOMOUNT is set on the diff --git a/fs/afs/mntpt.c b/fs/afs/mntpt.c index d23b2e344a78..aa59184151d0 100644 --- a/fs/afs/mntpt.c +++ b/fs/afs/mntpt.c @@ -241,7 +241,6 @@ error_no_devname: struct vfsmount *afs_d_automount(struct path *path) { struct vfsmount *newmnt; - int err; _enter("{%s,%s}", path->mnt->mnt_devname, path->dentry->d_name.name); @@ -249,24 +248,12 @@ struct vfsmount *afs_d_automount(struct path *path) if (IS_ERR(newmnt)) return newmnt; - mntget(newmnt); - err = do_add_mount(newmnt, path, MNT_SHRINKABLE, &afs_vfsmounts); - switch (err) { - case 0: - queue_delayed_work(afs_wq, &afs_mntpt_expiry_timer, - afs_mntpt_expiry_timeout * HZ); - _leave(" = %p {%s}", newmnt, newmnt->mnt_devname); - return newmnt; - case -EBUSY: - /* someone else made a mount here whilst we were busy */ - mntput(newmnt); - _leave(" = NULL [EBUSY]"); - return NULL; - default: - mntput(newmnt); - _leave(" = %d", err); - return ERR_PTR(err); - } + mntget(newmnt); /* prevent immediate expiration */ + mnt_set_expiry(newmnt, &afs_vfsmounts); + queue_delayed_work(afs_wq, &afs_mntpt_expiry_timer, + afs_mntpt_expiry_timeout * HZ); + _leave(" = %p {%s}", newmnt, newmnt->mnt_devname); + return newmnt; } /* diff --git a/fs/cifs/cifs_dfs_ref.c b/fs/cifs/cifs_dfs_ref.c index 0fc163808de3..7ed36536e754 100644 --- a/fs/cifs/cifs_dfs_ref.c +++ b/fs/cifs/cifs_dfs_ref.c @@ -351,7 +351,6 @@ free_xid: struct vfsmount *cifs_dfs_d_automount(struct path *path) { struct vfsmount *newmnt; - int err; cFYI(1, "in %s", __func__); @@ -361,25 +360,12 @@ struct vfsmount *cifs_dfs_d_automount(struct path *path) return newmnt; } - mntget(newmnt); - err = do_add_mount(newmnt, path, path->mnt->mnt_flags | MNT_SHRINKABLE, - &cifs_dfs_automount_list); - switch (err) { - case 0: - schedule_delayed_work(&cifs_dfs_automount_task, - cifs_dfs_mountpoint_expiry_timeout); - cFYI(1, "leaving %s [ok]" , __func__); - return newmnt; - case -EBUSY: - /* someone else made a mount here whilst we were busy */ - mntput(newmnt); - cFYI(1, "leaving %s [EBUSY]" , __func__); - return NULL; - default: - mntput(newmnt); - cFYI(1, "leaving %s [error %d]" , __func__, err); - return ERR_PTR(err); - } + mntget(newmnt); /* prevent immediate expiration */ + mnt_set_expiry(newmnt, &cifs_dfs_automount_list); + schedule_delayed_work(&cifs_dfs_automount_task, + cifs_dfs_mountpoint_expiry_timeout); + cFYI(1, "leaving %s [ok]" , __func__); + return newmnt; } const struct inode_operations cifs_dfs_referral_inode_operations = { diff --git a/fs/internal.h b/fs/internal.h index 9687c2ee2735..4931060fd089 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -70,6 +70,8 @@ extern void mnt_set_mountpoint(struct vfsmount *, struct dentry *, extern void release_mounts(struct list_head *); extern void umount_tree(struct vfsmount *, int, struct list_head *); extern struct vfsmount *copy_tree(struct vfsmount *, struct dentry *, int); +extern int do_add_mount(struct vfsmount *, struct path *, int); +extern void mnt_clear_expiry(struct vfsmount *); extern void __init mnt_init(void); diff --git a/fs/namei.c b/fs/namei.c index 5c89695ae1e4..c2e37727e3ab 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -900,6 +900,7 @@ static int follow_automount(struct path *path, unsigned flags, bool *need_mntput) { struct vfsmount *mnt; + int err; if (!path->dentry->d_op || !path->dentry->d_op->d_automount) return -EREMOTE; @@ -942,22 +943,49 @@ static int follow_automount(struct path *path, unsigned flags, return -EREMOTE; return PTR_ERR(mnt); } + if (!mnt) /* mount collision */ return 0; + /* The new mount record should have at least 2 refs to prevent it being + * expired before we get a chance to add it + */ + BUG_ON(mnt_get_count(mnt) < 2); + if (mnt->mnt_sb == path->mnt->mnt_sb && mnt->mnt_root == path->dentry) { + mnt_clear_expiry(mnt); + mntput(mnt); mntput(mnt); return -ELOOP; } - dput(path->dentry); - if (*need_mntput) - mntput(path->mnt); - path->mnt = mnt; - path->dentry = dget(mnt->mnt_root); - *need_mntput = true; - return 0; + /* We need to add the mountpoint to the parent. The filesystem may + * have placed it on an expiry list, and so we need to make sure it + * won't be expired under us if do_add_mount() fails (do_add_mount() + * will eat a reference unconditionally). + */ + mntget(mnt); + err = do_add_mount(mnt, path, path->mnt->mnt_flags | MNT_SHRINKABLE); + switch (err) { + case -EBUSY: + /* Someone else made a mount here whilst we were busy */ + err = 0; + default: + mnt_clear_expiry(mnt); + mntput(mnt); + mntput(mnt); + return err; + case 0: + mntput(mnt); + dput(path->dentry); + if (*need_mntput) + mntput(path->mnt); + path->mnt = mnt; + path->dentry = dget(mnt->mnt_root); + *need_mntput = true; + return 0; + } } /* diff --git a/fs/namespace.c b/fs/namespace.c index d94ccd6ddafd..bfcb701f9490 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1925,15 +1925,14 @@ static int do_new_mount(struct path *path, char *type, int flags, if (IS_ERR(mnt)) return PTR_ERR(mnt); - return do_add_mount(mnt, path, mnt_flags, NULL); + return do_add_mount(mnt, path, mnt_flags); } /* * add a mount into a namespace's mount tree - * - provide the option of adding the new mount to an expiration list + * - this unconditionally eats one of the caller's references to newmnt. */ -int do_add_mount(struct vfsmount *newmnt, struct path *path, - int mnt_flags, struct list_head *fslist) +int do_add_mount(struct vfsmount *newmnt, struct path *path, int mnt_flags) { int err; @@ -1963,9 +1962,6 @@ int do_add_mount(struct vfsmount *newmnt, struct path *path, if ((err = graft_tree(newmnt, path))) goto unlock; - if (fslist) /* add to the specified expiration list */ - list_add_tail(&newmnt->mnt_expire, fslist); - up_write(&namespace_sem); return 0; @@ -1975,7 +1971,36 @@ unlock: return err; } -EXPORT_SYMBOL_GPL(do_add_mount); +/** + * mnt_set_expiry - Put a mount on an expiration list + * @mnt: The mount to list. + * @expiry_list: The list to add the mount to. + */ +void mnt_set_expiry(struct vfsmount *mnt, struct list_head *expiry_list) +{ + down_write(&namespace_sem); + br_write_lock(vfsmount_lock); + + list_add_tail(&mnt->mnt_expire, expiry_list); + + br_write_unlock(vfsmount_lock); + up_write(&namespace_sem); +} +EXPORT_SYMBOL(mnt_set_expiry); + +/* + * Remove a vfsmount from any expiration list it may be on + */ +void mnt_clear_expiry(struct vfsmount *mnt) +{ + if (!list_empty(&mnt->mnt_expire)) { + down_write(&namespace_sem); + br_write_lock(vfsmount_lock); + list_del_init(&mnt->mnt_expire); + br_write_unlock(vfsmount_lock); + up_write(&namespace_sem); + } +} /* * process a list of expirable mountpoints with the intent of discarding any diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c index f3fbb1bf3f18..f32b8603dca8 100644 --- a/fs/nfs/namespace.c +++ b/fs/nfs/namespace.c @@ -149,26 +149,10 @@ struct vfsmount *nfs_d_automount(struct path *path) if (IS_ERR(mnt)) goto out; - mntget(mnt); - err = do_add_mount(mnt, path, path->mnt->mnt_flags | MNT_SHRINKABLE, - &nfs_automount_list); - switch (err) { - case 0: - dprintk("%s: done, success\n", __func__); - schedule_delayed_work(&nfs_automount_task, nfs_mountpoint_expiry_timeout); - break; - case -EBUSY: - /* someone else made a mount here whilst we were busy */ - mntput(mnt); - dprintk("%s: done, collision\n", __func__); - mnt = NULL; - break; - default: - mntput(mnt); - dprintk("%s: done, error %d\n", __func__, err); - mnt = ERR_PTR(err); - break; - } + dprintk("%s: done, success\n", __func__); + mntget(mnt); /* prevent immediate expiration */ + mnt_set_expiry(mnt, &nfs_automount_list); + schedule_delayed_work(&nfs_automount_task, nfs_mountpoint_expiry_timeout); out: nfs_free_fattr(fattr); diff --git a/include/linux/mount.h b/include/linux/mount.h index 1869ea24a739..af4765ea8fde 100644 --- a/include/linux/mount.h +++ b/include/linux/mount.h @@ -110,12 +110,7 @@ extern struct vfsmount *vfs_kern_mount(struct file_system_type *type, int flags, const char *name, void *data); -struct nameidata; - -struct path; -extern int do_add_mount(struct vfsmount *newmnt, struct path *path, - int mnt_flags, struct list_head *fslist); - +extern void mnt_set_expiry(struct vfsmount *mnt, struct list_head *expiry_list); extern void mark_mounts_for_expiry(struct list_head *mounts); extern dev_t name_to_dev_t(char *name); -- cgit v1.2.3 From 672c54466d24994eb9633f993862c89539504a42 Mon Sep 17 00:00:00 2001 From: Grant Likely Date: Thu, 13 Jan 2011 15:36:09 -0700 Subject: dt/flattree: Return virtual address from early_init_dt_alloc_memory_arch() The physical address is never used by the device tree code when allocating memory for unflattening. Change the architecture's alloc hook to return the virutal address instead. Signed-off-by: Grant Likely --- arch/microblaze/kernel/prom.c | 4 ++-- arch/mips/kernel/prom.c | 6 ++---- arch/powerpc/kernel/prom.c | 4 ++-- drivers/of/fdt.c | 8 +------- include/linux/of_fdt.h | 2 +- 5 files changed, 8 insertions(+), 16 deletions(-) (limited to 'include/linux') diff --git a/arch/microblaze/kernel/prom.c b/arch/microblaze/kernel/prom.c index c881393f07fd..bceaa5543e39 100644 --- a/arch/microblaze/kernel/prom.c +++ b/arch/microblaze/kernel/prom.c @@ -47,9 +47,9 @@ void __init early_init_dt_add_memory_arch(u64 base, u64 size) memblock_add(base, size); } -u64 __init early_init_dt_alloc_memory_arch(u64 size, u64 align) +void * __init early_init_dt_alloc_memory_arch(u64 size, u64 align) { - return memblock_alloc(size, align); + return __va(memblock_alloc(size, align)); } #ifdef CONFIG_EARLY_PRINTK diff --git a/arch/mips/kernel/prom.c b/arch/mips/kernel/prom.c index 9dbe58368953..a19811e98a41 100644 --- a/arch/mips/kernel/prom.c +++ b/arch/mips/kernel/prom.c @@ -45,11 +45,9 @@ void __init free_mem_mach(unsigned long addr, unsigned long size) return free_bootmem(addr, size); } -u64 __init early_init_dt_alloc_memory_arch(u64 size, u64 align) +void * __init early_init_dt_alloc_memory_arch(u64 size, u64 align) { - return virt_to_phys( - __alloc_bootmem(size, align, __pa(MAX_DMA_ADDRESS)) - ); + return __alloc_bootmem(size, align, __pa(MAX_DMA_ADDRESS)); } #ifdef CONFIG_BLK_DEV_INITRD diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c index 9e3132db718b..7185f0da7dc3 100644 --- a/arch/powerpc/kernel/prom.c +++ b/arch/powerpc/kernel/prom.c @@ -519,9 +519,9 @@ void __init early_init_dt_add_memory_arch(u64 base, u64 size) memblock_add(base, size); } -u64 __init early_init_dt_alloc_memory_arch(u64 size, u64 align) +void * __init early_init_dt_alloc_memory_arch(u64 size, u64 align) { - return memblock_alloc(size, align); + return __va(memblock_alloc(size, align)); } #ifdef CONFIG_BLK_DEV_INITRD diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c index c787c3d95c60..af824e7e0367 100644 --- a/drivers/of/fdt.c +++ b/drivers/of/fdt.c @@ -692,12 +692,6 @@ int __init early_init_dt_scan_chosen(unsigned long node, const char *uname, return 1; } -static void *__init early_device_tree_alloc(u64 size, u64 align) -{ - unsigned long mem = early_init_dt_alloc_memory_arch(size, align); - return __va(mem); -} - /** * unflatten_device_tree - create tree of device_nodes from flat blob * @@ -709,7 +703,7 @@ static void *__init early_device_tree_alloc(u64 size, u64 align) void __init unflatten_device_tree(void) { __unflatten_device_tree(initial_boot_params, &allnodes, - early_device_tree_alloc); + early_init_dt_alloc_memory_arch); /* Get pointer to OF "/chosen" node for use everywhere */ of_chosen = of_find_node_by_path("/chosen"); diff --git a/include/linux/of_fdt.h b/include/linux/of_fdt.h index 0ef22a1f129e..c84d900fbbb3 100644 --- a/include/linux/of_fdt.h +++ b/include/linux/of_fdt.h @@ -97,7 +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 u64 early_init_dt_alloc_memory_arch(u64 size, u64 align); +extern void * early_init_dt_alloc_memory_arch(u64 size, u64 align); extern u64 dt_mem_next_cell(int s, __be32 **cellp); /* -- cgit v1.2.3 From f03c65993b98eeb909a4012ce7833c5857d74755 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 14 Jan 2011 22:30:21 -0500 Subject: sanitize vfsmount refcounting changes Instead of splitting refcount between (per-cpu) mnt_count and (SMP-only) mnt_longrefs, make all references contribute to mnt_count again and keep track of how many are longterm ones. Accounting rules for longterm count: * 1 for each fs_struct.root.mnt * 1 for each fs_struct.pwd.mnt * 1 for having non-NULL ->mnt_ns * decrement to 0 happens only under vfsmount lock exclusive That allows nice common case for mntput() - since we can't drop the final reference until after mnt_longterm has reached 0 due to the rules above, mntput() can grab vfsmount lock shared and check mnt_longterm. If it turns out to be non-zero (which is the common case), we know that this is not the final mntput() and can just blindly decrement percpu mnt_count. Otherwise we grab vfsmount lock exclusive and do usual decrement-and-check of percpu mnt_count. For fs_struct.c we have mnt_make_longterm() and mnt_make_shortterm(); namespace.c uses the latter in places where we don't already hold vfsmount lock exclusive and opencodes a few remaining spots where we need to manipulate mnt_longterm. Note that we mostly revert the code outside of fs/namespace.c back to what we used to have; in particular, normal code doesn't need to care about two kinds of references, etc. And we get to keep the optimization Nick's variant had bought us... Signed-off-by: Al Viro --- drivers/mtd/mtdchar.c | 2 +- fs/anon_inodes.c | 2 +- fs/fs_struct.c | 35 ++++++++++----- fs/internal.h | 3 ++ fs/namei.c | 24 ----------- fs/namespace.c | 116 +++++++++++++++++++------------------------------- fs/pipe.c | 2 +- fs/super.c | 2 +- include/linux/mount.h | 4 +- include/linux/path.h | 2 - 10 files changed, 75 insertions(+), 117 deletions(-) (limited to 'include/linux') diff --git a/drivers/mtd/mtdchar.c b/drivers/mtd/mtdchar.c index ee4bb3330bdf..98240575a18d 100644 --- a/drivers/mtd/mtdchar.c +++ b/drivers/mtd/mtdchar.c @@ -1201,7 +1201,7 @@ err_unregister_chdev: static void __exit cleanup_mtdchar(void) { unregister_mtd_user(&mtdchar_notifier); - mntput_long(mtd_inode_mnt); + mntput(mtd_inode_mnt); unregister_filesystem(&mtd_inodefs_type); __unregister_chrdev(MTD_CHAR_MAJOR, 0, 1 << MINORBITS, "mtd"); } diff --git a/fs/anon_inodes.c b/fs/anon_inodes.c index cbe57f3c4d89..c5567cb78432 100644 --- a/fs/anon_inodes.c +++ b/fs/anon_inodes.c @@ -233,7 +233,7 @@ static int __init anon_inode_init(void) return 0; err_mntput: - mntput_long(anon_inode_mnt); + mntput(anon_inode_mnt); err_unregister_filesystem: unregister_filesystem(&anon_inode_fs_type); err_exit: diff --git a/fs/fs_struct.c b/fs/fs_struct.c index 68ca487bedb1..78b519c13536 100644 --- a/fs/fs_struct.c +++ b/fs/fs_struct.c @@ -4,6 +4,19 @@ #include #include #include +#include "internal.h" + +static inline void path_get_longterm(struct path *path) +{ + path_get(path); + mnt_make_longterm(path->mnt); +} + +static inline void path_put_longterm(struct path *path) +{ + mnt_make_shortterm(path->mnt); + path_put(path); +} /* * Replace the fs->{rootmnt,root} with {mnt,dentry}. Put the old values. @@ -17,11 +30,11 @@ void set_fs_root(struct fs_struct *fs, struct path *path) write_seqcount_begin(&fs->seq); old_root = fs->root; fs->root = *path; - path_get_long(path); + path_get_longterm(path); write_seqcount_end(&fs->seq); spin_unlock(&fs->lock); if (old_root.dentry) - path_put_long(&old_root); + path_put_longterm(&old_root); } /* @@ -36,12 +49,12 @@ void set_fs_pwd(struct fs_struct *fs, struct path *path) write_seqcount_begin(&fs->seq); old_pwd = fs->pwd; fs->pwd = *path; - path_get_long(path); + path_get_longterm(path); write_seqcount_end(&fs->seq); spin_unlock(&fs->lock); if (old_pwd.dentry) - path_put_long(&old_pwd); + path_put_longterm(&old_pwd); } void chroot_fs_refs(struct path *old_root, struct path *new_root) @@ -59,13 +72,13 @@ void chroot_fs_refs(struct path *old_root, struct path *new_root) write_seqcount_begin(&fs->seq); if (fs->root.dentry == old_root->dentry && fs->root.mnt == old_root->mnt) { - path_get_long(new_root); + path_get_longterm(new_root); fs->root = *new_root; count++; } if (fs->pwd.dentry == old_root->dentry && fs->pwd.mnt == old_root->mnt) { - path_get_long(new_root); + path_get_longterm(new_root); fs->pwd = *new_root; count++; } @@ -76,13 +89,13 @@ void chroot_fs_refs(struct path *old_root, struct path *new_root) } while_each_thread(g, p); read_unlock(&tasklist_lock); while (count--) - path_put_long(old_root); + path_put_longterm(old_root); } void free_fs_struct(struct fs_struct *fs) { - path_put_long(&fs->root); - path_put_long(&fs->pwd); + path_put_longterm(&fs->root); + path_put_longterm(&fs->pwd); kmem_cache_free(fs_cachep, fs); } @@ -118,9 +131,9 @@ struct fs_struct *copy_fs_struct(struct fs_struct *old) spin_lock(&old->lock); fs->root = old->root; - path_get_long(&fs->root); + path_get_longterm(&fs->root); fs->pwd = old->pwd; - path_get_long(&fs->pwd); + path_get_longterm(&fs->pwd); spin_unlock(&old->lock); } return fs; diff --git a/fs/internal.h b/fs/internal.h index 4931060fd089..12ccb86edef7 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -73,6 +73,9 @@ extern struct vfsmount *copy_tree(struct vfsmount *, struct dentry *, int); extern int do_add_mount(struct vfsmount *, struct path *, int); extern void mnt_clear_expiry(struct vfsmount *); +extern void mnt_make_longterm(struct vfsmount *); +extern void mnt_make_shortterm(struct vfsmount *); + extern void __init mnt_init(void); DECLARE_BRLOCK(vfsmount_lock); diff --git a/fs/namei.c b/fs/namei.c index c2e37727e3ab..8f7b41a14882 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -367,18 +367,6 @@ void path_get(struct path *path) } EXPORT_SYMBOL(path_get); -/** - * path_get_long - get a long reference to a path - * @path: path to get the reference to - * - * Given a path increment the reference count to the dentry and the vfsmount. - */ -void path_get_long(struct path *path) -{ - mntget_long(path->mnt); - dget(path->dentry); -} - /** * path_put - put a reference to a path * @path: path to put the reference to @@ -392,18 +380,6 @@ void path_put(struct path *path) } EXPORT_SYMBOL(path_put); -/** - * path_put_long - put a long reference to a path - * @path: path to put the reference to - * - * Given a path decrement the reference count to the dentry and the vfsmount. - */ -void path_put_long(struct path *path) -{ - dput(path->dentry); - mntput_long(path->mnt); -} - /** * nameidata_drop_rcu - drop this nameidata out of rcu-walk * @nd: nameidata pathwalk data to drop diff --git a/fs/namespace.c b/fs/namespace.c index d7fc05fac753..48809e21f270 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -183,7 +183,7 @@ static inline void mnt_dec_count(struct vfsmount *mnt) unsigned int mnt_get_count(struct vfsmount *mnt) { #ifdef CONFIG_SMP - unsigned int count = atomic_read(&mnt->mnt_longrefs); + unsigned int count = 0; int cpu; for_each_possible_cpu(cpu) { @@ -217,7 +217,7 @@ struct vfsmount *alloc_vfsmnt(const char *name) if (!mnt->mnt_pcp) goto out_free_devname; - atomic_set(&mnt->mnt_longrefs, 1); + this_cpu_add(mnt->mnt_pcp->mnt_count, 1); #else mnt->mnt_count = 1; mnt->mnt_writers = 0; @@ -624,8 +624,11 @@ static void commit_tree(struct vfsmount *mnt) BUG_ON(parent == mnt); list_add_tail(&head, &mnt->mnt_list); - list_for_each_entry(m, &head, mnt_list) + list_for_each_entry(m, &head, mnt_list) { m->mnt_ns = n; + atomic_inc(&m->mnt_longterm); + } + list_splice(&head, n->list.prev); list_add_tail(&mnt->mnt_hash, mount_hashtable + @@ -734,51 +737,30 @@ static inline void mntfree(struct vfsmount *mnt) deactivate_super(sb); } -#ifdef CONFIG_SMP -static inline void __mntput(struct vfsmount *mnt, int longrefs) +static void mntput_no_expire(struct vfsmount *mnt) { - if (!longrefs) { put_again: - br_read_lock(vfsmount_lock); - if (likely(atomic_read(&mnt->mnt_longrefs))) { - mnt_dec_count(mnt); - br_read_unlock(vfsmount_lock); - return; - } +#ifdef CONFIG_SMP + br_read_lock(vfsmount_lock); + if (likely(atomic_read(&mnt->mnt_longterm))) { + mnt_dec_count(mnt); br_read_unlock(vfsmount_lock); - } else { - BUG_ON(!atomic_read(&mnt->mnt_longrefs)); - if (atomic_add_unless(&mnt->mnt_longrefs, -1, 1)) - return; + return; } + br_read_unlock(vfsmount_lock); br_write_lock(vfsmount_lock); - if (!longrefs) - mnt_dec_count(mnt); - else - atomic_dec(&mnt->mnt_longrefs); + mnt_dec_count(mnt); if (mnt_get_count(mnt)) { br_write_unlock(vfsmount_lock); return; } - if (unlikely(mnt->mnt_pinned)) { - mnt_add_count(mnt, mnt->mnt_pinned + 1); - mnt->mnt_pinned = 0; - br_write_unlock(vfsmount_lock); - acct_auto_close_mnt(mnt); - goto put_again; - } - br_write_unlock(vfsmount_lock); - mntfree(mnt); -} #else -static inline void __mntput(struct vfsmount *mnt, int longrefs) -{ -put_again: mnt_dec_count(mnt); if (likely(mnt_get_count(mnt))) return; br_write_lock(vfsmount_lock); +#endif if (unlikely(mnt->mnt_pinned)) { mnt_add_count(mnt, mnt->mnt_pinned + 1); mnt->mnt_pinned = 0; @@ -789,12 +771,6 @@ put_again: br_write_unlock(vfsmount_lock); mntfree(mnt); } -#endif - -static void mntput_no_expire(struct vfsmount *mnt) -{ - __mntput(mnt, 0); -} void mntput(struct vfsmount *mnt) { @@ -802,7 +778,7 @@ void mntput(struct vfsmount *mnt) /* avoid cacheline pingpong, hope gcc doesn't get "smart" */ if (unlikely(mnt->mnt_expiry_mark)) mnt->mnt_expiry_mark = 0; - __mntput(mnt, 0); + mntput_no_expire(mnt); } } EXPORT_SYMBOL(mntput); @@ -815,33 +791,6 @@ struct vfsmount *mntget(struct vfsmount *mnt) } EXPORT_SYMBOL(mntget); -void mntput_long(struct vfsmount *mnt) -{ -#ifdef CONFIG_SMP - if (mnt) { - /* avoid cacheline pingpong, hope gcc doesn't get "smart" */ - if (unlikely(mnt->mnt_expiry_mark)) - mnt->mnt_expiry_mark = 0; - __mntput(mnt, 1); - } -#else - mntput(mnt); -#endif -} -EXPORT_SYMBOL(mntput_long); - -struct vfsmount *mntget_long(struct vfsmount *mnt) -{ -#ifdef CONFIG_SMP - if (mnt) - atomic_inc(&mnt->mnt_longrefs); - return mnt; -#else - return mntget(mnt); -#endif -} -EXPORT_SYMBOL(mntget_long); - void mnt_pin(struct vfsmount *mnt) { br_write_lock(vfsmount_lock); @@ -1216,7 +1165,7 @@ void release_mounts(struct list_head *head) dput(dentry); mntput(m); } - mntput_long(mnt); + mntput(mnt); } } @@ -1240,6 +1189,7 @@ void umount_tree(struct vfsmount *mnt, int propagate, struct list_head *kill) list_del_init(&p->mnt_list); __touch_mnt_namespace(p->mnt_ns); p->mnt_ns = NULL; + atomic_dec(&p->mnt_longterm); list_del_init(&p->mnt_child); if (p->mnt_parent != p) { p->mnt_parent->mnt_ghosts++; @@ -1969,7 +1919,7 @@ int do_add_mount(struct vfsmount *newmnt, struct path *path, int mnt_flags) unlock: up_write(&namespace_sem); - mntput_long(newmnt); + mntput(newmnt); return err; } @@ -2291,6 +2241,20 @@ static struct mnt_namespace *alloc_mnt_ns(void) return new_ns; } +void mnt_make_longterm(struct vfsmount *mnt) +{ + atomic_inc(&mnt->mnt_longterm); +} + +void mnt_make_shortterm(struct vfsmount *mnt) +{ + if (atomic_add_unless(&mnt->mnt_longterm, -1, 1)) + return; + br_write_lock(vfsmount_lock); + atomic_dec(&mnt->mnt_longterm); + br_write_unlock(vfsmount_lock); +} + /* * Allocate a new namespace structure and populate it with contents * copied from the namespace of the passed in task structure. @@ -2328,14 +2292,19 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, q = new_ns->root; while (p) { q->mnt_ns = new_ns; + atomic_inc(&q->mnt_longterm); if (fs) { if (p == fs->root.mnt) { + fs->root.mnt = mntget(q); + atomic_inc(&q->mnt_longterm); + mnt_make_shortterm(p); rootmnt = p; - fs->root.mnt = mntget_long(q); } if (p == fs->pwd.mnt) { + fs->pwd.mnt = mntget(q); + atomic_inc(&q->mnt_longterm); + mnt_make_shortterm(p); pwdmnt = p; - fs->pwd.mnt = mntget_long(q); } } p = next_mnt(p, mnt_ns->root); @@ -2344,9 +2313,9 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, up_write(&namespace_sem); if (rootmnt) - mntput_long(rootmnt); + mntput(rootmnt); if (pwdmnt) - mntput_long(pwdmnt); + mntput(pwdmnt); return new_ns; } @@ -2379,6 +2348,7 @@ struct mnt_namespace *create_mnt_ns(struct vfsmount *mnt) new_ns = alloc_mnt_ns(); if (!IS_ERR(new_ns)) { mnt->mnt_ns = new_ns; + atomic_inc(&mnt->mnt_longterm); new_ns->root = mnt; list_add(&new_ns->list, &new_ns->root->mnt_list); } diff --git a/fs/pipe.c b/fs/pipe.c index e2e95fb46a1e..89e9e19b1b2e 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -1292,7 +1292,7 @@ static int __init init_pipe_fs(void) static void __exit exit_pipe_fs(void) { unregister_filesystem(&pipe_fs_type); - mntput_long(pipe_mnt); + mntput(pipe_mnt); } fs_initcall(init_pipe_fs); diff --git a/fs/super.c b/fs/super.c index 4f6a3571a634..74e149efed81 100644 --- a/fs/super.c +++ b/fs/super.c @@ -1141,7 +1141,7 @@ static struct vfsmount *fs_set_subtype(struct vfsmount *mnt, const char *fstype) return mnt; err: - mntput_long(mnt); + mntput(mnt); return ERR_PTR(err); } diff --git a/include/linux/mount.h b/include/linux/mount.h index af4765ea8fde..604f122a2326 100644 --- a/include/linux/mount.h +++ b/include/linux/mount.h @@ -60,7 +60,7 @@ struct vfsmount { struct super_block *mnt_sb; /* pointer to superblock */ #ifdef CONFIG_SMP struct mnt_pcp __percpu *mnt_pcp; - atomic_t mnt_longrefs; + atomic_t mnt_longterm; /* how many of the refs are longterm */ #else int mnt_count; int mnt_writers; @@ -96,8 +96,6 @@ extern int mnt_clone_write(struct vfsmount *mnt); extern void mnt_drop_write(struct vfsmount *mnt); extern void mntput(struct vfsmount *mnt); extern struct vfsmount *mntget(struct vfsmount *mnt); -extern void mntput_long(struct vfsmount *mnt); -extern struct vfsmount *mntget_long(struct vfsmount *mnt); extern void mnt_pin(struct vfsmount *mnt); extern void mnt_unpin(struct vfsmount *mnt); extern int __mnt_is_readonly(struct vfsmount *mnt); diff --git a/include/linux/path.h b/include/linux/path.h index a581e8c06533..edc98dec6266 100644 --- a/include/linux/path.h +++ b/include/linux/path.h @@ -10,9 +10,7 @@ struct path { }; extern void path_get(struct path *); -extern void path_get_long(struct path *); extern void path_put(struct path *); -extern void path_put_long(struct path *); static inline int path_equal(const struct path *path1, const struct path *path2) { -- cgit v1.2.3 From fc8fe1e992ae0326a88edbe4d6793e840bbdd4ff Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sun, 16 Jan 2011 20:42:43 +0100 Subject: PCI / ACPI: Fix build of the AER driver for CONFIG_ACPI unset After commit 415e12b23792 ("PCI/ACPI: Request _OSC control once for each root bridge (v3)") include/linux/pci-acpi.h is included by drivers/pci/pcie/aer/aerdrv.c and if CONFIG_ACPI is unset, the bogus and unnecessary alternative definition of acpi_find_root_bridge_handle() causes a build error to occur. Remove the offending piece of garbage. Reported-and-tested-by: Stephen Rothwell Signed-off-by: Rafael J. Wysocki Signed-off-by: Linus Torvalds --- include/linux/pci-acpi.h | 3 --- 1 file changed, 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/pci-acpi.h b/include/linux/pci-acpi.h index 479d9bb88e11..44623500f419 100644 --- a/include/linux/pci-acpi.h +++ b/include/linux/pci-acpi.h @@ -35,9 +35,6 @@ static inline acpi_handle acpi_pci_get_bridge_handle(struct pci_bus *pbus) return acpi_get_pci_rootbridge_handle(pci_domain_nr(pbus), pbus->number); } -#else -static inline acpi_handle acpi_find_root_bridge_handle(struct pci_dev *pdev) -{ return NULL; } #endif #ifdef CONFIG_ACPI_APEI -- cgit v1.2.3 From 94ae85220a07d357d4937086c490854f63344de4 Mon Sep 17 00:00:00 2001 From: Russell King - ARM Linux Date: Sun, 16 Jan 2011 20:18:05 +0000 Subject: ARM: PL08x: cleanup comments Cleanup the formatting of comments, remove some which don't make sense anymore. Signed-off-by: Russell King [fix conflict with 96a608a4] Signed-off-by: Dan Williams --- drivers/dma/amba-pl08x.c | 146 ++++++++++++++++----------------------------- include/linux/amba/pl08x.h | 43 +++++++------ 2 files changed, 72 insertions(+), 117 deletions(-) (limited to 'include/linux') diff --git a/drivers/dma/amba-pl08x.c b/drivers/dma/amba-pl08x.c index bebc678ed4fc..297f48b0cba9 100644 --- a/drivers/dma/amba-pl08x.c +++ b/drivers/dma/amba-pl08x.c @@ -19,14 +19,14 @@ * this program; if not, write to the Free Software Foundation, Inc., 59 * Temple Place - Suite 330, Boston, MA 02111-1307, USA. * - * The full GNU General Public License is in this distribution in the - * file called COPYING. + * The full GNU General Public License is in this distribution in the file + * called COPYING. * * Documentation: ARM DDI 0196G == PL080 - * Documentation: ARM DDI 0218E == PL081 + * Documentation: ARM DDI 0218E == PL081 * - * PL080 & PL081 both have 16 sets of DMA signals that can be routed to - * any channel. + * PL080 & PL081 both have 16 sets of DMA signals that can be routed to any + * channel. * * The PL080 has 8 channels available for simultaneous use, and the PL081 * has only two channels. So on these DMA controllers the number of channels @@ -91,11 +91,9 @@ #define DRIVER_NAME "pl08xdmac" /** - * struct vendor_data - vendor-specific config parameters - * for PL08x derivatives + * struct vendor_data - vendor-specific config parameters for PL08x derivatives * @channels: the number of channels available in this variant - * @dualmaster: whether this version supports dual AHB masters - * or not. + * @dualmaster: whether this version supports dual AHB masters or not. */ struct vendor_data { u8 channels; @@ -241,10 +239,8 @@ static void pl08x_start_txd(struct pl08x_dma_chan *plchan, * * Disabling individual channels could lose data. * - * Disable the peripheral DMA after disabling the DMAC - * in order to allow the DMAC FIFO to drain, and - * hence allow the channel to show inactive - * + * Disable the peripheral DMA after disabling the DMAC in order to allow + * the DMAC FIFO to drain, and hence allow the channel to show inactive */ static void pl08x_pause_phy_chan(struct pl08x_phy_chan *ch) { @@ -367,6 +363,10 @@ static u32 pl08x_getbytes_chan(struct pl08x_dma_chan *plchan) /* * Allocate a physical channel for a virtual channel + * + * Try to locate a physical channel to be used for this transfer. If all + * are taken return NULL and the requester will have to cope by using + * some fallback PIO mode or retrying later. */ static struct pl08x_phy_chan * pl08x_get_phy_channel(struct pl08x_driver_data *pl08x, @@ -376,12 +376,6 @@ pl08x_get_phy_channel(struct pl08x_driver_data *pl08x, unsigned long flags; int i; - /* - * Try to locate a physical channel to be used for - * this transfer. If all are taken return NULL and - * the requester will have to cope by using some fallback - * PIO mode or retrying later. - */ for (i = 0; i < pl08x->vd->channels; i++) { ch = &pl08x->phy_chans[i]; @@ -495,9 +489,9 @@ struct pl08x_lli_build_data { }; /* - * Autoselect a master bus to use for the transfer - * this prefers the destination bus if both available - * if fixed address on one bus the other will be chosen + * Autoselect a master bus to use for the transfer this prefers the + * destination bus if both available if fixed address on one bus the + * other will be chosen */ static void pl08x_choose_master_bus(struct pl08x_lli_build_data *bd, struct pl08x_bus_data **mbus, struct pl08x_bus_data **sbus, u32 cctl) @@ -530,8 +524,7 @@ static void pl08x_choose_master_bus(struct pl08x_lli_build_data *bd, } /* - * Fills in one LLI for a certain transfer descriptor - * and advance the counter + * Fills in one LLI for a certain transfer descriptor and advance the counter */ static void pl08x_fill_lli_for_desc(struct pl08x_lli_build_data *bd, int num_llis, int len, u32 cctl) @@ -640,15 +633,11 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, * Choose bus to align to * - prefers destination bus if both available * - if fixed address on one bus chooses other - * - modifies cctl to choose an appropriate master */ pl08x_choose_master_bus(&bd, &mbus, &sbus, cctl); if (txd->len < mbus->buswidth) { - /* - * Less than a bus width available - * - send as single bytes - */ + /* Less than a bus width available - send as single bytes */ while (bd.remainder) { dev_vdbg(&pl08x->adev->dev, "%s single byte LLIs for a transfer of " @@ -659,10 +648,7 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, total_bytes++; } } else { - /* - * Make one byte LLIs until master bus is aligned - * - slave will then be aligned also - */ + /* Make one byte LLIs until master bus is aligned */ while ((mbus->addr) % (mbus->buswidth)) { dev_vdbg(&pl08x->adev->dev, "%s adjustment lli for less than bus width " @@ -674,7 +660,7 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, } /* - * Master now aligned + * Master now aligned * - if slave is not then we must set its width down */ if (sbus->addr % sbus->buswidth) { @@ -732,10 +718,8 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, if (lli_len == target_len) { /* - * Can send what we wanted - */ - /* - * Maintain alignment + * Can send what we wanted. + * Maintain alignment */ lli_len = (lli_len/mbus->buswidth) * mbus->buswidth; @@ -743,17 +727,14 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, } else { /* * So now we know how many bytes to transfer - * to get to the nearest boundary - * The next LLI will past the boundary - * - however we may be working to a boundary - * on the slave bus - * We need to ensure the master stays aligned + * to get to the nearest boundary. The next + * LLI will past the boundary. However, we + * may be working to a boundary on the slave + * bus. We need to ensure the master stays + * aligned, and that we are working in + * multiples of the bus widths. */ odd_bytes = lli_len % mbus->buswidth; - /* - * - and that we are working in multiples - * of the bus widths - */ lli_len -= odd_bytes; } @@ -793,8 +774,8 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, if (odd_bytes) { /* - * Creep past the boundary, - * maintaining master alignment + * Creep past the boundary, maintaining + * master alignment */ int j; for (j = 0; (j < mbus->buswidth) @@ -837,13 +818,9 @@ static int pl08x_fill_llis_for_desc(struct pl08x_driver_data *pl08x, } llis_va = txd->llis_va; - /* - * The final LLI terminates the LLI. - */ + /* The final LLI terminates the LLI. */ llis_va[num_llis - 1].lli = 0; - /* - * The final LLI element shall also fire an interrupt - */ + /* The final LLI element shall also fire an interrupt. */ llis_va[num_llis - 1].cctl |= PL080_CONTROL_TC_IRQ_EN; #ifdef VERBOSE_DEBUG @@ -891,7 +868,6 @@ static void pl08x_free_txd_list(struct pl08x_driver_data *pl08x, list_del(&txdi->node); pl08x_free_txd(pl08x, txdi); } - } } @@ -1020,10 +996,9 @@ static struct dma_async_tx_descriptor *pl08x_prep_dma_interrupt( } /* - * Code accessing dma_async_is_complete() in a tight loop - * may give problems - could schedule where indicated. - * If slaves are relying on interrupts to signal completion this - * function must not be called with interrupts disabled + * Code accessing dma_async_is_complete() in a tight loop may give problems. + * If slaves are relying on interrupts to signal completion this function + * must not be called with interrupts disabled. */ static enum dma_status pl08x_dma_tx_status(struct dma_chan *chan, @@ -1045,10 +1020,6 @@ pl08x_dma_tx_status(struct dma_chan *chan, return ret; } - /* - * schedule(); could be inserted here - */ - /* * This cookie not complete yet */ @@ -1273,11 +1244,10 @@ static int pl08x_prep_channel_resources(struct pl08x_dma_chan *plchan, } } else /* - * Else we're all set, paused and ready to roll, - * status will switch to PL08X_CHAN_RUNNING when - * we call issue_pending(). If there is something - * running on the channel already we don't change - * its state. + * Else we're all set, paused and ready to roll, status + * will switch to PL08X_CHAN_RUNNING when we call + * issue_pending(). If there is something running on the + * channel already we don't change its state. */ if (plchan->state == PL08X_CHAN_IDLE) plchan->state = PL08X_CHAN_PAUSED; @@ -1528,10 +1498,9 @@ bool pl08x_filter_id(struct dma_chan *chan, void *chan_id) /* * Just check that the device is there and active - * TODO: turn this bit on/off depending on the number of - * physical channels actually used, if it is zero... well - * shut it off. That will save some power. Cut the clock - * at the same time. + * TODO: turn this bit on/off depending on the number of physical channels + * actually used, if it is zero... well shut it off. That will save some + * power. Cut the clock at the same time. */ static void pl08x_ensure_on(struct pl08x_driver_data *pl08x) { @@ -1579,16 +1548,11 @@ static void pl08x_tasklet(unsigned long data) plchan->at = NULL; if (txd) { - /* - * Update last completed - */ + /* Update last completed */ plchan->lc = txd->tx.cookie; } - /* - * If a new descriptor is queued, set it up - * plchan->at is NULL here - */ + /* If a new descriptor is queued, set it up plchan->at is NULL here */ if (!list_empty(&plchan->pend_list)) { struct pl08x_txd *next; @@ -1615,11 +1579,10 @@ static void pl08x_tasklet(unsigned long data) plchan->state = PL08X_CHAN_IDLE; /* - * And NOW before anyone else can grab that free:d - * up physical channel, see if there is some memcpy - * pending that seriously needs to start because of - * being stacked up while we were choking the - * physical channels with data. + * And NOW before anyone else can grab that free:d up + * physical channel, see if there is some memcpy pending + * that seriously needs to start because of being stacked + * up while we were choking the physical channels with data. */ list_for_each_entry(waiting, &pl08x->memcpy.channels, chan.device_node) { @@ -1670,9 +1633,7 @@ static irqreturn_t pl08x_irq(int irq, void *dev) val = readl(pl08x->base + PL080_ERR_STATUS); if (val) { - /* - * An error interrupt (on one or more channels) - */ + /* An error interrupt (on one or more channels) */ dev_err(&pl08x->adev->dev, "%s error interrupt, register value 0x%08x\n", __func__, val); @@ -1696,9 +1657,7 @@ static irqreturn_t pl08x_irq(int irq, void *dev) mask |= (1 << i); } } - /* - * Clear only the terminal interrupts on channels we processed - */ + /* Clear only the terminal interrupts on channels we processed */ writel(mask, pl08x->base + PL080_TC_CLEAR); return mask ? IRQ_HANDLED : IRQ_NONE; @@ -1717,6 +1676,7 @@ static int pl08x_dma_init_virtual_channels(struct pl08x_driver_data *pl08x, int i; INIT_LIST_HEAD(&dmadev->channels); + /* * Register as many many memcpy as we have physical channels, * we won't always be able to use all but the code will have @@ -1950,9 +1910,7 @@ static int pl08x_probe(struct amba_device *adev, struct amba_id *id) /* Turn on the PL08x */ pl08x_ensure_on(pl08x); - /* - * Attach the interrupt handler - */ + /* Attach the interrupt handler */ writel(0x000000FF, pl08x->base + PL080_ERR_CLEAR); writel(0x000000FF, pl08x->base + PL080_TC_CLEAR); diff --git a/include/linux/amba/pl08x.h b/include/linux/amba/pl08x.h index 5b87b6aac3f8..3111385b8ca7 100644 --- a/include/linux/amba/pl08x.h +++ b/include/linux/amba/pl08x.h @@ -12,7 +12,6 @@ * * Please credit ARM.com * Documentation: ARM DDI 0196D - * */ #ifndef AMBA_PL08X_H @@ -55,8 +54,8 @@ enum { * @circular_buffer: whether the buffer passed in is circular and * shall simply be looped round round (like a record baby round * round round round) - * @single: the device connected to this channel will request single - * DMA transfers, not bursts. (Bursts are default.) + * @single: the device connected to this channel will request single DMA + * transfers, not bursts. (Bursts are default.) * @periph_buses: the device connected to this channel is accessible via * these buses (use PL08X_AHB1 | PL08X_AHB2). */ @@ -78,8 +77,7 @@ struct pl08x_channel_data { * @addr: current address * @maxwidth: the maximum width of a transfer on this bus * @buswidth: the width of this bus in bytes: 1, 2 or 4 - * @fill_bytes: bytes required to fill to the next bus memory - * boundary + * @fill_bytes: bytes required to fill to the next bus memory boundary */ struct pl08x_bus_data { dma_addr_t addr; @@ -92,10 +90,10 @@ struct pl08x_bus_data { * struct pl08x_phy_chan - holder for the physical channels * @id: physical index to this channel * @lock: a lock to use when altering an instance of this struct - * @signal: the physical signal (aka channel) serving this - * physical channel right now - * @serving: the virtual channel currently being served by this - * physical channel + * @signal: the physical signal (aka channel) serving this physical channel + * right now + * @serving: the virtual channel currently being served by this physical + * channel */ struct pl08x_phy_chan { unsigned int id; @@ -119,7 +117,6 @@ struct pl08x_txd { size_t len; dma_addr_t llis_bus; struct pl08x_lli *llis_va; - bool active; /* Default cctl value for LLIs */ u32 cctl; /* @@ -130,8 +127,8 @@ struct pl08x_txd { }; /** - * struct pl08x_dma_chan_state - holds the PL08x specific virtual - * channel states + * struct pl08x_dma_chan_state - holds the PL08x specific virtual channel + * states * @PL08X_CHAN_IDLE: the channel is idle * @PL08X_CHAN_RUNNING: the channel has allocated a physical transport * channel and is running a transfer on it @@ -152,7 +149,7 @@ enum pl08x_dma_chan_state { * @chan: wrappped abstract channel * @phychan: the physical channel utilized by this channel, if there is one * @phychan_hold: if non-zero, hold on to the physical channel even if we - * have no pending entries + * have no pending entries * @tasklet: tasklet scheduled by the IRQ to handle actual work etc * @name: name of channel * @cd: channel platform data @@ -166,8 +163,8 @@ enum pl08x_dma_chan_state { * @host: a pointer to the host (internal use) * @state: whether the channel is idle, paused, running etc * @slave: whether this channel is a device (slave) or for memcpy - * @waiting: a TX descriptor on this channel which is waiting for - * a physical channel to become available + * @waiting: a TX descriptor on this channel which is waiting for a physical + * channel to become available */ struct pl08x_dma_chan { struct dma_chan chan; @@ -189,16 +186,16 @@ struct pl08x_dma_chan { }; /** - * struct pl08x_platform_data - the platform configuration for the - * PL08x PrimeCells. + * struct pl08x_platform_data - the platform configuration for the PL08x + * PrimeCells. * @slave_channels: the channels defined for the different devices on the * platform, all inclusive, including multiplexed channels. The available - * physical channels will be multiplexed around these signals as they - * are requested, just enumerate all possible channels. - * @get_signal: request a physical signal to be used for a DMA - * transfer immediately: if there is some multiplexing or similar blocking - * the use of the channel the transfer can be denied by returning - * less than zero, else it returns the allocated signal number + * physical channels will be multiplexed around these signals as they are + * requested, just enumerate all possible channels. + * @get_signal: request a physical signal to be used for a DMA transfer + * immediately: if there is some multiplexing or similar blocking the use + * of the channel the transfer can be denied by returning less than zero, + * else it returns the allocated signal number * @put_signal: indicate to the platform that this physical signal is not * running any DMA transfer and multiplexing can be recycled * @lli_buses: buses which LLIs can be fetched from: PL08X_AHB1 | PL08X_AHB2 -- cgit v1.2.3 From 2fe17c1075836b66678ed2a305fd09b6773883aa Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 14 Jan 2011 13:07:43 +0100 Subject: fallocate should be a file operation Currently all filesystems except XFS implement fallocate asynchronously, while XFS forced a commit. Both of these are suboptimal - in case of O_SYNC I/O we really want our allocation on disk, especially for the !KEEP_SIZE case where we actually grow the file with user-visible zeroes. On the other hand always commiting the transaction is a bad idea for fast-path uses of fallocate like for example in recent Samba versions. Given that block allocation is a data plane operation anyway change it from an inode operation to a file operation so that we have the file structure available that lets us check for O_SYNC. This also includes moving the code around for a few of the filesystems, and remove the already unnedded S_ISDIR checks given that we only wire up fallocate for regular files. Signed-off-by: Christoph Hellwig Signed-off-by: Al Viro --- Documentation/filesystems/Locking | 3 +- fs/btrfs/file.c | 113 +++++++++++++++++ fs/btrfs/inode.c | 111 ---------------- fs/ext4/ext4.h | 2 +- fs/ext4/extents.c | 9 +- fs/ext4/file.c | 2 +- fs/gfs2/file.c | 258 ++++++++++++++++++++++++++++++++++++++ fs/gfs2/ops_inode.c | 258 -------------------------------------- fs/ocfs2/file.c | 8 +- fs/open.c | 4 +- fs/xfs/linux-2.6/xfs_file.c | 56 +++++++++ fs/xfs/linux-2.6/xfs_iops.c | 60 --------- include/linux/fs.h | 4 +- 13 files changed, 440 insertions(+), 448 deletions(-) (limited to 'include/linux') diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking index 651d5237c155..4471a416c274 100644 --- a/Documentation/filesystems/Locking +++ b/Documentation/filesystems/Locking @@ -60,7 +60,6 @@ ata *); ssize_t (*listxattr) (struct dentry *, char *, size_t); int (*removexattr) (struct dentry *, const char *); void (*truncate_range)(struct inode *, loff_t, loff_t); - long (*fallocate)(struct inode *inode, int mode, loff_t offset, loff_t len); int (*fiemap)(struct inode *, struct fiemap_extent_info *, u64 start, u64 len); locking rules: @@ -88,7 +87,6 @@ getxattr: no listxattr: no removexattr: yes truncate_range: yes -fallocate: no fiemap: no Additionally, ->rmdir(), ->unlink() and ->rename() have ->i_mutex on victim. @@ -437,6 +435,7 @@ prototypes: ssize_t (*splice_read)(struct file *, loff_t *, struct pipe_inode_info *, size_t, unsigned int); int (*setlease)(struct file *, long, struct file_lock **); + long (*fallocate)(struct file *, int, loff_t, loff_t); }; locking rules: diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 66836d85763b..a9e0a4eaf3d9 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include #include @@ -1237,6 +1238,117 @@ static int btrfs_file_mmap(struct file *filp, struct vm_area_struct *vma) return 0; } +static long btrfs_fallocate(struct file *file, int mode, + loff_t offset, loff_t len) +{ + struct inode *inode = file->f_path.dentry->d_inode; + struct extent_state *cached_state = NULL; + u64 cur_offset; + u64 last_byte; + u64 alloc_start; + u64 alloc_end; + u64 alloc_hint = 0; + u64 locked_end; + u64 mask = BTRFS_I(inode)->root->sectorsize - 1; + struct extent_map *em; + int ret; + + alloc_start = offset & ~mask; + alloc_end = (offset + len + mask) & ~mask; + + /* We only support the FALLOC_FL_KEEP_SIZE mode */ + if (mode & ~FALLOC_FL_KEEP_SIZE) + return -EOPNOTSUPP; + + /* + * wait for ordered IO before we have any locks. We'll loop again + * below with the locks held. + */ + btrfs_wait_ordered_range(inode, alloc_start, alloc_end - alloc_start); + + mutex_lock(&inode->i_mutex); + ret = inode_newsize_ok(inode, alloc_end); + if (ret) + goto out; + + if (alloc_start > inode->i_size) { + ret = btrfs_cont_expand(inode, alloc_start); + if (ret) + goto out; + } + + ret = btrfs_check_data_free_space(inode, alloc_end - alloc_start); + if (ret) + goto out; + + locked_end = alloc_end - 1; + while (1) { + struct btrfs_ordered_extent *ordered; + + /* the extent lock is ordered inside the running + * transaction + */ + lock_extent_bits(&BTRFS_I(inode)->io_tree, alloc_start, + locked_end, 0, &cached_state, GFP_NOFS); + ordered = btrfs_lookup_first_ordered_extent(inode, + alloc_end - 1); + if (ordered && + ordered->file_offset + ordered->len > alloc_start && + ordered->file_offset < alloc_end) { + btrfs_put_ordered_extent(ordered); + unlock_extent_cached(&BTRFS_I(inode)->io_tree, + alloc_start, locked_end, + &cached_state, GFP_NOFS); + /* + * we can't wait on the range with the transaction + * running or with the extent lock held + */ + btrfs_wait_ordered_range(inode, alloc_start, + alloc_end - alloc_start); + } else { + if (ordered) + btrfs_put_ordered_extent(ordered); + break; + } + } + + cur_offset = alloc_start; + while (1) { + em = btrfs_get_extent(inode, NULL, 0, cur_offset, + alloc_end - cur_offset, 0); + BUG_ON(IS_ERR(em) || !em); + last_byte = min(extent_map_end(em), alloc_end); + last_byte = (last_byte + mask) & ~mask; + if (em->block_start == EXTENT_MAP_HOLE || + (cur_offset >= inode->i_size && + !test_bit(EXTENT_FLAG_PREALLOC, &em->flags))) { + ret = btrfs_prealloc_file_range(inode, mode, cur_offset, + last_byte - cur_offset, + 1 << inode->i_blkbits, + offset + len, + &alloc_hint); + if (ret < 0) { + free_extent_map(em); + break; + } + } + free_extent_map(em); + + cur_offset = last_byte; + if (cur_offset >= alloc_end) { + ret = 0; + break; + } + } + unlock_extent_cached(&BTRFS_I(inode)->io_tree, alloc_start, locked_end, + &cached_state, GFP_NOFS); + + btrfs_free_reserved_data_space(inode, alloc_end - alloc_start); +out: + mutex_unlock(&inode->i_mutex); + return ret; +} + const struct file_operations btrfs_file_operations = { .llseek = generic_file_llseek, .read = do_sync_read, @@ -1248,6 +1360,7 @@ const struct file_operations btrfs_file_operations = { .open = generic_file_open, .release = btrfs_release_file, .fsync = btrfs_sync_file, + .fallocate = btrfs_fallocate, .unlocked_ioctl = btrfs_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = btrfs_ioctl, diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 64daf2acd0d5..902afbf50811 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -7098,116 +7098,6 @@ int btrfs_prealloc_file_range_trans(struct inode *inode, min_size, actual_len, alloc_hint, trans); } -static long btrfs_fallocate(struct inode *inode, int mode, - loff_t offset, loff_t len) -{ - struct extent_state *cached_state = NULL; - u64 cur_offset; - u64 last_byte; - u64 alloc_start; - u64 alloc_end; - u64 alloc_hint = 0; - u64 locked_end; - u64 mask = BTRFS_I(inode)->root->sectorsize - 1; - struct extent_map *em; - int ret; - - alloc_start = offset & ~mask; - alloc_end = (offset + len + mask) & ~mask; - - /* We only support the FALLOC_FL_KEEP_SIZE mode */ - if (mode & ~FALLOC_FL_KEEP_SIZE) - return -EOPNOTSUPP; - - /* - * wait for ordered IO before we have any locks. We'll loop again - * below with the locks held. - */ - btrfs_wait_ordered_range(inode, alloc_start, alloc_end - alloc_start); - - mutex_lock(&inode->i_mutex); - ret = inode_newsize_ok(inode, alloc_end); - if (ret) - goto out; - - if (alloc_start > inode->i_size) { - ret = btrfs_cont_expand(inode, alloc_start); - if (ret) - goto out; - } - - ret = btrfs_check_data_free_space(inode, alloc_end - alloc_start); - if (ret) - goto out; - - locked_end = alloc_end - 1; - while (1) { - struct btrfs_ordered_extent *ordered; - - /* the extent lock is ordered inside the running - * transaction - */ - lock_extent_bits(&BTRFS_I(inode)->io_tree, alloc_start, - locked_end, 0, &cached_state, GFP_NOFS); - ordered = btrfs_lookup_first_ordered_extent(inode, - alloc_end - 1); - if (ordered && - ordered->file_offset + ordered->len > alloc_start && - ordered->file_offset < alloc_end) { - btrfs_put_ordered_extent(ordered); - unlock_extent_cached(&BTRFS_I(inode)->io_tree, - alloc_start, locked_end, - &cached_state, GFP_NOFS); - /* - * we can't wait on the range with the transaction - * running or with the extent lock held - */ - btrfs_wait_ordered_range(inode, alloc_start, - alloc_end - alloc_start); - } else { - if (ordered) - btrfs_put_ordered_extent(ordered); - break; - } - } - - cur_offset = alloc_start; - while (1) { - em = btrfs_get_extent(inode, NULL, 0, cur_offset, - alloc_end - cur_offset, 0); - BUG_ON(IS_ERR(em) || !em); - last_byte = min(extent_map_end(em), alloc_end); - last_byte = (last_byte + mask) & ~mask; - if (em->block_start == EXTENT_MAP_HOLE || - (cur_offset >= inode->i_size && - !test_bit(EXTENT_FLAG_PREALLOC, &em->flags))) { - ret = btrfs_prealloc_file_range(inode, mode, cur_offset, - last_byte - cur_offset, - 1 << inode->i_blkbits, - offset + len, - &alloc_hint); - if (ret < 0) { - free_extent_map(em); - break; - } - } - free_extent_map(em); - - cur_offset = last_byte; - if (cur_offset >= alloc_end) { - ret = 0; - break; - } - } - unlock_extent_cached(&BTRFS_I(inode)->io_tree, alloc_start, locked_end, - &cached_state, GFP_NOFS); - - btrfs_free_reserved_data_space(inode, alloc_end - alloc_start); -out: - mutex_unlock(&inode->i_mutex); - return ret; -} - static int btrfs_set_page_dirty(struct page *page) { return __set_page_dirty_nobuffers(page); @@ -7310,7 +7200,6 @@ static const struct inode_operations btrfs_file_inode_operations = { .listxattr = btrfs_listxattr, .removexattr = btrfs_removexattr, .permission = btrfs_permission, - .fallocate = btrfs_fallocate, .fiemap = btrfs_fiemap, }; static const struct inode_operations btrfs_special_inode_operations = { diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 1de65f572033..0c8d97b56f34 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -2065,7 +2065,7 @@ extern int ext4_ext_map_blocks(handle_t *handle, struct inode *inode, extern void ext4_ext_truncate(struct inode *); extern void ext4_ext_init(struct super_block *); extern void ext4_ext_release(struct super_block *); -extern long ext4_fallocate(struct inode *inode, int mode, loff_t offset, +extern long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len); extern int ext4_convert_unwritten_extents(struct inode *inode, loff_t offset, ssize_t len); diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 4bdd160854eb..63a75810b7c3 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -3627,14 +3627,15 @@ static void ext4_falloc_update_inode(struct inode *inode, } /* - * preallocate space for a file. This implements ext4's fallocate inode + * preallocate space for a file. This implements ext4's fallocate file * operation, which gets called from sys_fallocate system call. * For block-mapped files, posix_fallocate should fall back to the method * of writing zeroes to the required new blocks (the same behavior which is * expected for file systems which do not support fallocate() system call). */ -long ext4_fallocate(struct inode *inode, int mode, loff_t offset, loff_t len) +long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len) { + struct inode *inode = file->f_path.dentry->d_inode; handle_t *handle; loff_t new_size; unsigned int max_blocks; @@ -3655,10 +3656,6 @@ long ext4_fallocate(struct inode *inode, int mode, loff_t offset, loff_t len) if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) return -EOPNOTSUPP; - /* preallocation to directories is currently not supported */ - if (S_ISDIR(inode->i_mode)) - return -ENODEV; - map.m_lblk = offset >> blkbits; /* * We can't just convert len to max_blocks because diff --git a/fs/ext4/file.c b/fs/ext4/file.c index bb003dc9ffff..2e8322c8aa88 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -210,6 +210,7 @@ const struct file_operations ext4_file_operations = { .fsync = ext4_sync_file, .splice_read = generic_file_splice_read, .splice_write = generic_file_splice_write, + .fallocate = ext4_fallocate, }; const struct inode_operations ext4_file_inode_operations = { @@ -223,7 +224,6 @@ const struct inode_operations ext4_file_inode_operations = { .removexattr = generic_removexattr, #endif .check_acl = ext4_check_acl, - .fallocate = ext4_fallocate, .fiemap = ext4_fiemap, }; diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c index fca6689e12e6..7cfdcb913363 100644 --- a/fs/gfs2/file.c +++ b/fs/gfs2/file.c @@ -19,6 +19,8 @@ #include #include #include +#include +#include #include #include #include @@ -610,6 +612,260 @@ static ssize_t gfs2_file_aio_write(struct kiocb *iocb, const struct iovec *iov, return generic_file_aio_write(iocb, iov, nr_segs, pos); } +static void empty_write_end(struct page *page, unsigned from, + unsigned to) +{ + struct gfs2_inode *ip = GFS2_I(page->mapping->host); + + page_zero_new_buffers(page, from, to); + flush_dcache_page(page); + mark_page_accessed(page); + + if (!gfs2_is_writeback(ip)) + gfs2_page_add_databufs(ip, page, from, to); + + block_commit_write(page, from, to); +} + +static int write_empty_blocks(struct page *page, unsigned from, unsigned to) +{ + unsigned start, end, next; + struct buffer_head *bh, *head; + int error; + + if (!page_has_buffers(page)) { + error = __block_write_begin(page, from, to - from, gfs2_block_map); + if (unlikely(error)) + return error; + + empty_write_end(page, from, to); + return 0; + } + + bh = head = page_buffers(page); + next = end = 0; + while (next < from) { + next += bh->b_size; + bh = bh->b_this_page; + } + start = next; + do { + next += bh->b_size; + if (buffer_mapped(bh)) { + if (end) { + error = __block_write_begin(page, start, end - start, + gfs2_block_map); + if (unlikely(error)) + return error; + empty_write_end(page, start, end); + end = 0; + } + start = next; + } + else + end = next; + bh = bh->b_this_page; + } while (next < to); + + if (end) { + error = __block_write_begin(page, start, end - start, gfs2_block_map); + if (unlikely(error)) + return error; + empty_write_end(page, start, end); + } + + return 0; +} + +static int fallocate_chunk(struct inode *inode, loff_t offset, loff_t len, + int mode) +{ + struct gfs2_inode *ip = GFS2_I(inode); + struct buffer_head *dibh; + int error; + u64 start = offset >> PAGE_CACHE_SHIFT; + unsigned int start_offset = offset & ~PAGE_CACHE_MASK; + u64 end = (offset + len - 1) >> PAGE_CACHE_SHIFT; + pgoff_t curr; + struct page *page; + unsigned int end_offset = (offset + len) & ~PAGE_CACHE_MASK; + unsigned int from, to; + + if (!end_offset) + end_offset = PAGE_CACHE_SIZE; + + error = gfs2_meta_inode_buffer(ip, &dibh); + if (unlikely(error)) + goto out; + + gfs2_trans_add_bh(ip->i_gl, dibh, 1); + + if (gfs2_is_stuffed(ip)) { + error = gfs2_unstuff_dinode(ip, NULL); + if (unlikely(error)) + goto out; + } + + curr = start; + offset = start << PAGE_CACHE_SHIFT; + from = start_offset; + to = PAGE_CACHE_SIZE; + while (curr <= end) { + page = grab_cache_page_write_begin(inode->i_mapping, curr, + AOP_FLAG_NOFS); + if (unlikely(!page)) { + error = -ENOMEM; + goto out; + } + + if (curr == end) + to = end_offset; + error = write_empty_blocks(page, from, to); + if (!error && offset + to > inode->i_size && + !(mode & FALLOC_FL_KEEP_SIZE)) { + i_size_write(inode, offset + to); + } + unlock_page(page); + page_cache_release(page); + if (error) + goto out; + curr++; + offset += PAGE_CACHE_SIZE; + from = 0; + } + + gfs2_dinode_out(ip, dibh->b_data); + mark_inode_dirty(inode); + + brelse(dibh); + +out: + return error; +} + +static void calc_max_reserv(struct gfs2_inode *ip, loff_t max, loff_t *len, + unsigned int *data_blocks, unsigned int *ind_blocks) +{ + const struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); + unsigned int max_blocks = ip->i_alloc->al_rgd->rd_free_clone; + unsigned int tmp, max_data = max_blocks - 3 * (sdp->sd_max_height - 1); + + for (tmp = max_data; tmp > sdp->sd_diptrs;) { + tmp = DIV_ROUND_UP(tmp, sdp->sd_inptrs); + max_data -= tmp; + } + /* This calculation isn't the exact reverse of gfs2_write_calc_reserve, + so it might end up with fewer data blocks */ + if (max_data <= *data_blocks) + return; + *data_blocks = max_data; + *ind_blocks = max_blocks - max_data; + *len = ((loff_t)max_data - 3) << sdp->sd_sb.sb_bsize_shift; + if (*len > max) { + *len = max; + gfs2_write_calc_reserv(ip, max, data_blocks, ind_blocks); + } +} + +static long gfs2_fallocate(struct file *file, int mode, loff_t offset, + loff_t len) +{ + struct inode *inode = file->f_path.dentry->d_inode; + struct gfs2_sbd *sdp = GFS2_SB(inode); + struct gfs2_inode *ip = GFS2_I(inode); + unsigned int data_blocks = 0, ind_blocks = 0, rblocks; + loff_t bytes, max_bytes; + struct gfs2_alloc *al; + int error; + loff_t next = (offset + len - 1) >> sdp->sd_sb.sb_bsize_shift; + next = (next + 1) << sdp->sd_sb.sb_bsize_shift; + + /* We only support the FALLOC_FL_KEEP_SIZE mode */ + if (mode & ~FALLOC_FL_KEEP_SIZE) + return -EOPNOTSUPP; + + offset = (offset >> sdp->sd_sb.sb_bsize_shift) << + sdp->sd_sb.sb_bsize_shift; + + len = next - offset; + bytes = sdp->sd_max_rg_data * sdp->sd_sb.sb_bsize / 2; + if (!bytes) + bytes = UINT_MAX; + + gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &ip->i_gh); + error = gfs2_glock_nq(&ip->i_gh); + if (unlikely(error)) + goto out_uninit; + + if (!gfs2_write_alloc_required(ip, offset, len)) + goto out_unlock; + + while (len > 0) { + if (len < bytes) + bytes = len; + al = gfs2_alloc_get(ip); + if (!al) { + error = -ENOMEM; + goto out_unlock; + } + + error = gfs2_quota_lock_check(ip); + if (error) + goto out_alloc_put; + +retry: + gfs2_write_calc_reserv(ip, bytes, &data_blocks, &ind_blocks); + + al->al_requested = data_blocks + ind_blocks; + error = gfs2_inplace_reserve(ip); + if (error) { + if (error == -ENOSPC && bytes > sdp->sd_sb.sb_bsize) { + bytes >>= 1; + goto retry; + } + goto out_qunlock; + } + max_bytes = bytes; + calc_max_reserv(ip, len, &max_bytes, &data_blocks, &ind_blocks); + al->al_requested = data_blocks + ind_blocks; + + rblocks = RES_DINODE + ind_blocks + RES_STATFS + RES_QUOTA + + RES_RG_HDR + gfs2_rg_blocks(al); + if (gfs2_is_jdata(ip)) + rblocks += data_blocks ? data_blocks : 1; + + error = gfs2_trans_begin(sdp, rblocks, + PAGE_CACHE_SIZE/sdp->sd_sb.sb_bsize); + if (error) + goto out_trans_fail; + + error = fallocate_chunk(inode, offset, max_bytes, mode); + gfs2_trans_end(sdp); + + if (error) + goto out_trans_fail; + + len -= max_bytes; + offset += max_bytes; + gfs2_inplace_release(ip); + gfs2_quota_unlock(ip); + gfs2_alloc_put(ip); + } + goto out_unlock; + +out_trans_fail: + gfs2_inplace_release(ip); +out_qunlock: + gfs2_quota_unlock(ip); +out_alloc_put: + gfs2_alloc_put(ip); +out_unlock: + gfs2_glock_dq(&ip->i_gh); +out_uninit: + gfs2_holder_uninit(&ip->i_gh); + return error; +} + #ifdef CONFIG_GFS2_FS_LOCKING_DLM /** @@ -765,6 +1021,7 @@ const struct file_operations gfs2_file_fops = { .splice_read = generic_file_splice_read, .splice_write = generic_file_splice_write, .setlease = gfs2_setlease, + .fallocate = gfs2_fallocate, }; const struct file_operations gfs2_dir_fops = { @@ -794,6 +1051,7 @@ const struct file_operations gfs2_file_fops_nolock = { .splice_read = generic_file_splice_read, .splice_write = generic_file_splice_write, .setlease = generic_setlease, + .fallocate = gfs2_fallocate, }; const struct file_operations gfs2_dir_fops_nolock = { diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c index c09528c07f3d..d8b26ac2e20b 100644 --- a/fs/gfs2/ops_inode.c +++ b/fs/gfs2/ops_inode.c @@ -18,8 +18,6 @@ #include #include #include -#include -#include #include #include "gfs2.h" @@ -1257,261 +1255,6 @@ static int gfs2_removexattr(struct dentry *dentry, const char *name) return ret; } -static void empty_write_end(struct page *page, unsigned from, - unsigned to) -{ - struct gfs2_inode *ip = GFS2_I(page->mapping->host); - - page_zero_new_buffers(page, from, to); - flush_dcache_page(page); - mark_page_accessed(page); - - if (!gfs2_is_writeback(ip)) - gfs2_page_add_databufs(ip, page, from, to); - - block_commit_write(page, from, to); -} - - -static int write_empty_blocks(struct page *page, unsigned from, unsigned to) -{ - unsigned start, end, next; - struct buffer_head *bh, *head; - int error; - - if (!page_has_buffers(page)) { - error = __block_write_begin(page, from, to - from, gfs2_block_map); - if (unlikely(error)) - return error; - - empty_write_end(page, from, to); - return 0; - } - - bh = head = page_buffers(page); - next = end = 0; - while (next < from) { - next += bh->b_size; - bh = bh->b_this_page; - } - start = next; - do { - next += bh->b_size; - if (buffer_mapped(bh)) { - if (end) { - error = __block_write_begin(page, start, end - start, - gfs2_block_map); - if (unlikely(error)) - return error; - empty_write_end(page, start, end); - end = 0; - } - start = next; - } - else - end = next; - bh = bh->b_this_page; - } while (next < to); - - if (end) { - error = __block_write_begin(page, start, end - start, gfs2_block_map); - if (unlikely(error)) - return error; - empty_write_end(page, start, end); - } - - return 0; -} - -static int fallocate_chunk(struct inode *inode, loff_t offset, loff_t len, - int mode) -{ - struct gfs2_inode *ip = GFS2_I(inode); - struct buffer_head *dibh; - int error; - u64 start = offset >> PAGE_CACHE_SHIFT; - unsigned int start_offset = offset & ~PAGE_CACHE_MASK; - u64 end = (offset + len - 1) >> PAGE_CACHE_SHIFT; - pgoff_t curr; - struct page *page; - unsigned int end_offset = (offset + len) & ~PAGE_CACHE_MASK; - unsigned int from, to; - - if (!end_offset) - end_offset = PAGE_CACHE_SIZE; - - error = gfs2_meta_inode_buffer(ip, &dibh); - if (unlikely(error)) - goto out; - - gfs2_trans_add_bh(ip->i_gl, dibh, 1); - - if (gfs2_is_stuffed(ip)) { - error = gfs2_unstuff_dinode(ip, NULL); - if (unlikely(error)) - goto out; - } - - curr = start; - offset = start << PAGE_CACHE_SHIFT; - from = start_offset; - to = PAGE_CACHE_SIZE; - while (curr <= end) { - page = grab_cache_page_write_begin(inode->i_mapping, curr, - AOP_FLAG_NOFS); - if (unlikely(!page)) { - error = -ENOMEM; - goto out; - } - - if (curr == end) - to = end_offset; - error = write_empty_blocks(page, from, to); - if (!error && offset + to > inode->i_size && - !(mode & FALLOC_FL_KEEP_SIZE)) { - i_size_write(inode, offset + to); - } - unlock_page(page); - page_cache_release(page); - if (error) - goto out; - curr++; - offset += PAGE_CACHE_SIZE; - from = 0; - } - - gfs2_dinode_out(ip, dibh->b_data); - mark_inode_dirty(inode); - - brelse(dibh); - -out: - return error; -} - -static void calc_max_reserv(struct gfs2_inode *ip, loff_t max, loff_t *len, - unsigned int *data_blocks, unsigned int *ind_blocks) -{ - const struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - unsigned int max_blocks = ip->i_alloc->al_rgd->rd_free_clone; - unsigned int tmp, max_data = max_blocks - 3 * (sdp->sd_max_height - 1); - - for (tmp = max_data; tmp > sdp->sd_diptrs;) { - tmp = DIV_ROUND_UP(tmp, sdp->sd_inptrs); - max_data -= tmp; - } - /* This calculation isn't the exact reverse of gfs2_write_calc_reserve, - so it might end up with fewer data blocks */ - if (max_data <= *data_blocks) - return; - *data_blocks = max_data; - *ind_blocks = max_blocks - max_data; - *len = ((loff_t)max_data - 3) << sdp->sd_sb.sb_bsize_shift; - if (*len > max) { - *len = max; - gfs2_write_calc_reserv(ip, max, data_blocks, ind_blocks); - } -} - -static long gfs2_fallocate(struct inode *inode, int mode, loff_t offset, - loff_t len) -{ - struct gfs2_sbd *sdp = GFS2_SB(inode); - struct gfs2_inode *ip = GFS2_I(inode); - unsigned int data_blocks = 0, ind_blocks = 0, rblocks; - loff_t bytes, max_bytes; - struct gfs2_alloc *al; - int error; - loff_t next = (offset + len - 1) >> sdp->sd_sb.sb_bsize_shift; - next = (next + 1) << sdp->sd_sb.sb_bsize_shift; - - /* We only support the FALLOC_FL_KEEP_SIZE mode */ - if (mode & ~FALLOC_FL_KEEP_SIZE) - return -EOPNOTSUPP; - - offset = (offset >> sdp->sd_sb.sb_bsize_shift) << - sdp->sd_sb.sb_bsize_shift; - - len = next - offset; - bytes = sdp->sd_max_rg_data * sdp->sd_sb.sb_bsize / 2; - if (!bytes) - bytes = UINT_MAX; - - gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &ip->i_gh); - error = gfs2_glock_nq(&ip->i_gh); - if (unlikely(error)) - goto out_uninit; - - if (!gfs2_write_alloc_required(ip, offset, len)) - goto out_unlock; - - while (len > 0) { - if (len < bytes) - bytes = len; - al = gfs2_alloc_get(ip); - if (!al) { - error = -ENOMEM; - goto out_unlock; - } - - error = gfs2_quota_lock_check(ip); - if (error) - goto out_alloc_put; - -retry: - gfs2_write_calc_reserv(ip, bytes, &data_blocks, &ind_blocks); - - al->al_requested = data_blocks + ind_blocks; - error = gfs2_inplace_reserve(ip); - if (error) { - if (error == -ENOSPC && bytes > sdp->sd_sb.sb_bsize) { - bytes >>= 1; - goto retry; - } - goto out_qunlock; - } - max_bytes = bytes; - calc_max_reserv(ip, len, &max_bytes, &data_blocks, &ind_blocks); - al->al_requested = data_blocks + ind_blocks; - - rblocks = RES_DINODE + ind_blocks + RES_STATFS + RES_QUOTA + - RES_RG_HDR + gfs2_rg_blocks(al); - if (gfs2_is_jdata(ip)) - rblocks += data_blocks ? data_blocks : 1; - - error = gfs2_trans_begin(sdp, rblocks, - PAGE_CACHE_SIZE/sdp->sd_sb.sb_bsize); - if (error) - goto out_trans_fail; - - error = fallocate_chunk(inode, offset, max_bytes, mode); - gfs2_trans_end(sdp); - - if (error) - goto out_trans_fail; - - len -= max_bytes; - offset += max_bytes; - gfs2_inplace_release(ip); - gfs2_quota_unlock(ip); - gfs2_alloc_put(ip); - } - goto out_unlock; - -out_trans_fail: - gfs2_inplace_release(ip); -out_qunlock: - gfs2_quota_unlock(ip); -out_alloc_put: - gfs2_alloc_put(ip); -out_unlock: - gfs2_glock_dq(&ip->i_gh); -out_uninit: - gfs2_holder_uninit(&ip->i_gh); - return error; -} - - static int gfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, u64 start, u64 len) { @@ -1562,7 +1305,6 @@ const struct inode_operations gfs2_file_iops = { .getxattr = gfs2_getxattr, .listxattr = gfs2_listxattr, .removexattr = gfs2_removexattr, - .fallocate = gfs2_fallocate, .fiemap = gfs2_fiemap, }; diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index cf254ce8c941..a6651956482e 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -1989,9 +1989,10 @@ int ocfs2_change_file_space(struct file *file, unsigned int cmd, return __ocfs2_change_file_space(file, inode, file->f_pos, cmd, sr, 0); } -static long ocfs2_fallocate(struct inode *inode, int mode, loff_t offset, +static long ocfs2_fallocate(struct file *file, int mode, loff_t offset, loff_t len) { + struct inode *inode = file->f_path.dentry->d_inode; struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); struct ocfs2_space_resv sr; int change_size = 1; @@ -2002,9 +2003,6 @@ static long ocfs2_fallocate(struct inode *inode, int mode, loff_t offset, if (!ocfs2_writes_unwritten_extents(osb)) return -EOPNOTSUPP; - if (S_ISDIR(inode->i_mode)) - return -ENODEV; - if (mode & FALLOC_FL_KEEP_SIZE) change_size = 0; @@ -2612,7 +2610,6 @@ const struct inode_operations ocfs2_file_iops = { .getxattr = generic_getxattr, .listxattr = ocfs2_listxattr, .removexattr = generic_removexattr, - .fallocate = ocfs2_fallocate, .fiemap = ocfs2_fiemap, }; @@ -2644,6 +2641,7 @@ const struct file_operations ocfs2_fops = { .flock = ocfs2_flock, .splice_read = ocfs2_file_splice_read, .splice_write = ocfs2_file_splice_write, + .fallocate = ocfs2_fallocate, }; const struct file_operations ocfs2_dops = { diff --git a/fs/open.c b/fs/open.c index 5b6ef7e2859e..e52389e1f05b 100644 --- a/fs/open.c +++ b/fs/open.c @@ -255,10 +255,10 @@ int do_fallocate(struct file *file, int mode, loff_t offset, loff_t len) if (((offset + len) > inode->i_sb->s_maxbytes) || ((offset + len) < 0)) return -EFBIG; - if (!inode->i_op->fallocate) + if (!file->f_op->fallocate) return -EOPNOTSUPP; - return inode->i_op->fallocate(inode, mode, offset, len); + return file->f_op->fallocate(file, mode, offset, len); } SYSCALL_DEFINE(fallocate)(int fd, int mode, loff_t offset, loff_t len) diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c index ef51eb43e137..a55c1b46b219 100644 --- a/fs/xfs/linux-2.6/xfs_file.c +++ b/fs/xfs/linux-2.6/xfs_file.c @@ -37,6 +37,7 @@ #include "xfs_trace.h" #include +#include static const struct vm_operations_struct xfs_file_vm_ops; @@ -882,6 +883,60 @@ out_unlock: return ret; } +STATIC long +xfs_file_fallocate( + struct file *file, + int mode, + loff_t offset, + loff_t len) +{ + struct inode *inode = file->f_path.dentry->d_inode; + long error; + loff_t new_size = 0; + xfs_flock64_t bf; + xfs_inode_t *ip = XFS_I(inode); + int cmd = XFS_IOC_RESVSP; + + if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE)) + return -EOPNOTSUPP; + + bf.l_whence = 0; + bf.l_start = offset; + bf.l_len = len; + + xfs_ilock(ip, XFS_IOLOCK_EXCL); + + if (mode & FALLOC_FL_PUNCH_HOLE) + cmd = XFS_IOC_UNRESVSP; + + /* check the new inode size is valid before allocating */ + if (!(mode & FALLOC_FL_KEEP_SIZE) && + offset + len > i_size_read(inode)) { + new_size = offset + len; + error = inode_newsize_ok(inode, new_size); + if (error) + goto out_unlock; + } + + error = -xfs_change_file_space(ip, cmd, &bf, 0, XFS_ATTR_NOLOCK); + if (error) + goto out_unlock; + + /* Change file size if needed */ + if (new_size) { + struct iattr iattr; + + iattr.ia_valid = ATTR_SIZE; + iattr.ia_size = new_size; + error = -xfs_setattr(ip, &iattr, XFS_ATTR_NOLOCK); + } + +out_unlock: + xfs_iunlock(ip, XFS_IOLOCK_EXCL); + return error; +} + + STATIC int xfs_file_open( struct inode *inode, @@ -1000,6 +1055,7 @@ const struct file_operations xfs_file_operations = { .open = xfs_file_open, .release = xfs_file_release, .fsync = xfs_file_fsync, + .fallocate = xfs_file_fallocate, }; const struct file_operations xfs_dir_file_operations = { diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c index a4ecc2188a09..bd5727852fd6 100644 --- a/fs/xfs/linux-2.6/xfs_iops.c +++ b/fs/xfs/linux-2.6/xfs_iops.c @@ -46,7 +46,6 @@ #include #include #include -#include #include #include @@ -505,64 +504,6 @@ xfs_vn_setattr( return -xfs_setattr(XFS_I(dentry->d_inode), iattr, 0); } -STATIC long -xfs_vn_fallocate( - struct inode *inode, - int mode, - loff_t offset, - loff_t len) -{ - long error; - loff_t new_size = 0; - xfs_flock64_t bf; - xfs_inode_t *ip = XFS_I(inode); - int cmd = XFS_IOC_RESVSP; - - if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE)) - return -EOPNOTSUPP; - - /* preallocation on directories not yet supported */ - error = -ENODEV; - if (S_ISDIR(inode->i_mode)) - goto out_error; - - bf.l_whence = 0; - bf.l_start = offset; - bf.l_len = len; - - xfs_ilock(ip, XFS_IOLOCK_EXCL); - - if (mode & FALLOC_FL_PUNCH_HOLE) - cmd = XFS_IOC_UNRESVSP; - - /* check the new inode size is valid before allocating */ - if (!(mode & FALLOC_FL_KEEP_SIZE) && - offset + len > i_size_read(inode)) { - new_size = offset + len; - error = inode_newsize_ok(inode, new_size); - if (error) - goto out_unlock; - } - - error = -xfs_change_file_space(ip, cmd, &bf, 0, XFS_ATTR_NOLOCK); - if (error) - goto out_unlock; - - /* Change file size if needed */ - if (new_size) { - struct iattr iattr; - - iattr.ia_valid = ATTR_SIZE; - iattr.ia_size = new_size; - error = -xfs_setattr(ip, &iattr, XFS_ATTR_NOLOCK); - } - -out_unlock: - xfs_iunlock(ip, XFS_IOLOCK_EXCL); -out_error: - return error; -} - #define XFS_FIEMAP_FLAGS (FIEMAP_FLAG_SYNC|FIEMAP_FLAG_XATTR) /* @@ -656,7 +597,6 @@ static const struct inode_operations xfs_inode_operations = { .getxattr = generic_getxattr, .removexattr = generic_removexattr, .listxattr = xfs_vn_listxattr, - .fallocate = xfs_vn_fallocate, .fiemap = xfs_vn_fiemap, }; diff --git a/include/linux/fs.h b/include/linux/fs.h index 177b4ddea418..09b5bd6a7c6b 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1552,6 +1552,8 @@ struct file_operations { ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, loff_t *, size_t, unsigned int); ssize_t (*splice_read)(struct file *, loff_t *, struct pipe_inode_info *, size_t, unsigned int); int (*setlease)(struct file *, long, struct file_lock **); + long (*fallocate)(struct file *file, int mode, loff_t offset, + loff_t len); }; #define IPERM_FLAG_RCU 0x0001 @@ -1582,8 +1584,6 @@ struct inode_operations { ssize_t (*listxattr) (struct dentry *, char *, size_t); int (*removexattr) (struct dentry *, const char *); void (*truncate_range)(struct inode *, loff_t, loff_t); - long (*fallocate)(struct inode *inode, int mode, loff_t offset, - loff_t len); int (*fiemap)(struct inode *, struct fiemap_extent_info *, u64 start, u64 len); } ____cacheline_aligned; -- cgit v1.2.3 From c2b3e74b78b24cb367289a75a2bd30e569e56e0e Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Mon, 13 Dec 2010 19:38:08 -0500 Subject: fs: Remove unlikely() from fput_light() In fput_light(), there's an unlikely(fput_needed), which running on my normal desktop doing firefox, xchat, evolution and part of my distcc farm, and running the annotate branch profiler shows that the unlikely is not very unlikely. correct incorrect % Function File Line ------- --------- - -------- ---- ---- 0 48 100 fput_light file.h 26 115828710 897415279 88 fput_light file.h 26 865271179 5286128445 85 fput_light file.h 26 19568539 8923664 31 fput_light file.h 26 12353677 3562279 22 fput_light file.h 26 267691 67062 20 fput_light file.h 26 15014853 348172 2 fput_light file.h 26 209258 205 0 fput_light file.h 26 1364164 0 0 fput_light file.h 26 Which gives 1032903812 times it was correct and 6203351846 times it was incorrect, or 85% incorrect. Cc: Andrew Morton Signed-off-by: Steven Rostedt Signed-off-by: Al Viro --- include/linux/file.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/file.h b/include/linux/file.h index b1e12970f617..e85baebf6279 100644 --- a/include/linux/file.h +++ b/include/linux/file.h @@ -23,7 +23,7 @@ extern struct file *alloc_file(struct path *, fmode_t mode, static inline void fput_light(struct file *file, int fput_needed) { - if (unlikely(fput_needed)) + if (fput_needed) fput(file); } -- cgit v1.2.3 From ecf5632dd189ab4c366cef853d6e5fe7adfe52e5 Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Sun, 16 Jan 2011 23:28:17 +0900 Subject: fs: fix address space warnings in ioctl_fiemap() The fi_extents_start field of struct fiemap_extent_info is a user pointer but was not marked as __user. This makes sparse emit following warnings: CHECK fs/ioctl.c fs/ioctl.c:114:26: warning: incorrect type in argument 1 (different address spaces) fs/ioctl.c:114:26: expected void [noderef] *dst fs/ioctl.c:114:26: got struct fiemap_extent *[assigned] dest fs/ioctl.c:202:14: warning: incorrect type in argument 1 (different address spaces) fs/ioctl.c:202:14: expected void const volatile [noderef] * fs/ioctl.c:202:14: got struct fiemap_extent *[assigned] fi_extents_start fs/ioctl.c:212:27: warning: incorrect type in argument 1 (different address spaces) fs/ioctl.c:212:27: expected void [noderef] *dst fs/ioctl.c:212:27: got char * Also add 'ufiemap' variable to eliminate unnecessary casts. Signed-off-by: Namhyung Kim Signed-off-by: Al Viro --- fs/ioctl.c | 10 +++++----- include/linux/fs.h | 4 ++-- 2 files changed, 7 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/fs/ioctl.c b/fs/ioctl.c index d6cc16476620..a59635e295fa 100644 --- a/fs/ioctl.c +++ b/fs/ioctl.c @@ -86,7 +86,7 @@ int fiemap_fill_next_extent(struct fiemap_extent_info *fieinfo, u64 logical, u64 phys, u64 len, u32 flags) { struct fiemap_extent extent; - struct fiemap_extent *dest = fieinfo->fi_extents_start; + struct fiemap_extent __user *dest = fieinfo->fi_extents_start; /* only count the extents */ if (fieinfo->fi_extents_max == 0) { @@ -173,6 +173,7 @@ static int fiemap_check_ranges(struct super_block *sb, static int ioctl_fiemap(struct file *filp, unsigned long arg) { struct fiemap fiemap; + struct fiemap __user *ufiemap = (struct fiemap __user *) arg; struct fiemap_extent_info fieinfo = { 0, }; struct inode *inode = filp->f_path.dentry->d_inode; struct super_block *sb = inode->i_sb; @@ -182,8 +183,7 @@ static int ioctl_fiemap(struct file *filp, unsigned long arg) if (!inode->i_op->fiemap) return -EOPNOTSUPP; - if (copy_from_user(&fiemap, (struct fiemap __user *)arg, - sizeof(struct fiemap))) + if (copy_from_user(&fiemap, ufiemap, sizeof(fiemap))) return -EFAULT; if (fiemap.fm_extent_count > FIEMAP_MAX_EXTENTS) @@ -196,7 +196,7 @@ static int ioctl_fiemap(struct file *filp, unsigned long arg) fieinfo.fi_flags = fiemap.fm_flags; fieinfo.fi_extents_max = fiemap.fm_extent_count; - fieinfo.fi_extents_start = (struct fiemap_extent *)(arg + sizeof(fiemap)); + fieinfo.fi_extents_start = ufiemap->fm_extents; if (fiemap.fm_extent_count != 0 && !access_ok(VERIFY_WRITE, fieinfo.fi_extents_start, @@ -209,7 +209,7 @@ static int ioctl_fiemap(struct file *filp, unsigned long arg) error = inode->i_op->fiemap(inode, &fieinfo, fiemap.fm_start, len); fiemap.fm_flags = fieinfo.fi_flags; fiemap.fm_mapped_extents = fieinfo.fi_extents_mapped; - if (copy_to_user((char *)arg, &fiemap, sizeof(fiemap))) + if (copy_to_user(ufiemap, &fiemap, sizeof(fiemap))) error = -EFAULT; return error; diff --git a/include/linux/fs.h b/include/linux/fs.h index 09b5bd6a7c6b..32b38cd829d3 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1483,8 +1483,8 @@ struct fiemap_extent_info { unsigned int fi_flags; /* Flags as passed from user */ unsigned int fi_extents_mapped; /* Number of mapped extents */ unsigned int fi_extents_max; /* Size of fiemap_extent array */ - struct fiemap_extent *fi_extents_start; /* Start of fiemap_extent - * array */ + struct fiemap_extent __user *fi_extents_start; /* Start of + fiemap_extent array */ }; int fiemap_fill_next_extent(struct fiemap_extent_info *info, u64 logical, u64 phys, u64 len, u32 flags); -- cgit v1.2.3 From 2a8652f4e0d11ee27b1d2870c600fd1300661a6e Mon Sep 17 00:00:00 2001 From: Roberto Sassu Date: Wed, 3 Nov 2010 11:11:15 +0100 Subject: ecryptfs: moved ECRYPTFS_SUPER_MAGIC definition to linux/magic.h The definition of ECRYPTFS_SUPER_MAGIC has been moved to the include file 'linux/magic.h' to become available to other kernel subsystems. Signed-off-by: Roberto Sassu Signed-off-by: Tyler Hicks --- fs/ecryptfs/ecryptfs_kernel.h | 1 - include/linux/magic.h | 1 + 2 files changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index 413a3c48f0bb..bc530a81e4ce 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -192,7 +192,6 @@ ecryptfs_get_key_payload_data(struct key *key) (((struct user_key_payload*)key->payload.data)->data); } -#define ECRYPTFS_SUPER_MAGIC 0xf15f #define ECRYPTFS_MAX_KEYSET_SIZE 1024 #define ECRYPTFS_MAX_CIPHER_NAME_SIZE 32 #define ECRYPTFS_MAX_NUM_ENC_KEYS 64 diff --git a/include/linux/magic.h b/include/linux/magic.h index ff690d05f129..62730ea2b56e 100644 --- a/include/linux/magic.h +++ b/include/linux/magic.h @@ -16,6 +16,7 @@ #define TMPFS_MAGIC 0x01021994 #define HUGETLBFS_MAGIC 0x958458f6 /* some random number */ #define SQUASHFS_MAGIC 0x73717368 +#define ECRYPTFS_SUPER_MAGIC 0xf15f #define EFS_SUPER_MAGIC 0x414A53 #define EXT2_SUPER_MAGIC 0xEF53 #define EXT3_SUPER_MAGIC 0xEF53 -- cgit v1.2.3 From d2763b4f44e16f44cc4156c9591e74df9dcd88be Mon Sep 17 00:00:00 2001 From: Naveen Kumar Gaddipati Date: Mon, 17 Jan 2011 20:40:58 -0800 Subject: Input: bu21013_ts - remove duplicate resolution parameters Remove duplicate display resolution parameters from platform data as one pair is quite enough. Signed-off-by: Naveen Kumar Gaddipati Signed-off-by: Dmitry Torokhov --- drivers/input/touchscreen/bu21013_ts.c | 4 ++-- include/linux/input/bu21013.h | 4 ---- 2 files changed, 2 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/drivers/input/touchscreen/bu21013_ts.c b/drivers/input/touchscreen/bu21013_ts.c index f7fa9ef4cd65..3c7e60b2f778 100644 --- a/drivers/input/touchscreen/bu21013_ts.c +++ b/drivers/input/touchscreen/bu21013_ts.c @@ -485,9 +485,9 @@ static int __devinit bu21013_probe(struct i2c_client *client, __set_bit(EV_ABS, in_dev->evbit); input_set_abs_params(in_dev, ABS_MT_POSITION_X, 0, - pdata->x_max_res, 0, 0); + pdata->touch_x_max, 0, 0); input_set_abs_params(in_dev, ABS_MT_POSITION_Y, 0, - pdata->y_max_res, 0, 0); + pdata->touch_y_max, 0, 0); input_set_drvdata(in_dev, bu21013_data); error = request_threaded_irq(pdata->irq, NULL, bu21013_gpio_irq, diff --git a/include/linux/input/bu21013.h b/include/linux/input/bu21013.h index e470d387dd49..05e03284b92a 100644 --- a/include/linux/input/bu21013.h +++ b/include/linux/input/bu21013.h @@ -12,8 +12,6 @@ * @cs_en: pointer to the cs enable function * @cs_dis: pointer to the cs disable function * @irq_read_val: pointer to read the pen irq value function - * @x_max_res: xmax resolution - * @y_max_res: ymax resolution * @touch_x_max: touch x max * @touch_y_max: touch y max * @cs_pin: chip select pin @@ -29,8 +27,6 @@ struct bu21013_platform_device { int (*cs_en)(int reset_pin); int (*cs_dis)(int reset_pin); int (*irq_read_val)(void); - int x_max_res; - int y_max_res; int touch_x_max; int touch_y_max; unsigned int cs_pin; -- cgit v1.2.3 From 2ce802f62ba32a7d95748ac92bf351f76affb6ff Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 20 Jan 2011 12:06:35 +0100 Subject: lockdep: Move early boot local IRQ enable/disable status to init/main.c During early boot, local IRQ is disabled until IRQ subsystem is properly initialized. During this time, no one should enable local IRQ and some operations which usually are not allowed with IRQ disabled, e.g. operations which might sleep or require communications with other processors, are allowed. lockdep tracked this with early_boot_irqs_off/on() callbacks. As other subsystems need this information too, move it to init/main.c and make it generally available. While at it, toggle the boolean to early_boot_irqs_disabled instead of enabled so that it can be initialized with %false and %true indicates the exceptional condition. Signed-off-by: Tejun Heo Acked-by: Peter Zijlstra Acked-by: Pekka Enberg Cc: Linus Torvalds LKML-Reference: <20110120110635.GB6036@htj.dyndns.org> Signed-off-by: Ingo Molnar --- arch/x86/xen/enlighten.c | 2 +- include/linux/kernel.h | 2 ++ include/linux/lockdep.h | 8 -------- init/main.c | 13 +++++++++++-- kernel/lockdep.c | 18 +----------------- kernel/trace/trace_irqsoff.c | 8 -------- 6 files changed, 15 insertions(+), 36 deletions(-) (limited to 'include/linux') diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 7e8d3bc80af6..50542efe45fb 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -1194,7 +1194,7 @@ asmlinkage void __init xen_start_kernel(void) per_cpu(xen_vcpu, 0) = &HYPERVISOR_shared_info->vcpu_info[0]; local_irq_disable(); - early_boot_irqs_off(); + early_boot_irqs_disabled = true; memblock_init(); diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 5a9d9059520b..d07d8057e440 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -243,6 +243,8 @@ extern int test_taint(unsigned flag); extern unsigned long get_taint(void); extern int root_mountflags; +extern bool early_boot_irqs_disabled; + /* Values used for system_state */ extern enum system_states { SYSTEM_BOOTING, diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index 71c09b26c759..f638fd78d106 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -436,16 +436,8 @@ do { \ #endif /* CONFIG_LOCKDEP */ #ifdef CONFIG_TRACE_IRQFLAGS -extern void early_boot_irqs_off(void); -extern void early_boot_irqs_on(void); extern void print_irqtrace_events(struct task_struct *curr); #else -static inline void early_boot_irqs_off(void) -{ -} -static inline void early_boot_irqs_on(void) -{ -} static inline void print_irqtrace_events(struct task_struct *curr) { } diff --git a/init/main.c b/init/main.c index 00799c1d4628..33c37c379e96 100644 --- a/init/main.c +++ b/init/main.c @@ -96,6 +96,15 @@ static inline void mark_rodata_ro(void) { } extern void tc_init(void); #endif +/* + * Debug helper: via this flag we know that we are in 'early bootup code' + * where only the boot processor is running with IRQ disabled. This means + * two things - IRQ must not be enabled before the flag is cleared and some + * operations which are not allowed with IRQ disabled are allowed while the + * flag is set. + */ +bool early_boot_irqs_disabled __read_mostly; + enum system_states system_state __read_mostly; EXPORT_SYMBOL(system_state); @@ -554,7 +563,7 @@ asmlinkage void __init start_kernel(void) cgroup_init_early(); local_irq_disable(); - early_boot_irqs_off(); + early_boot_irqs_disabled = true; /* * Interrupts are still disabled. Do necessary setups, then @@ -621,7 +630,7 @@ asmlinkage void __init start_kernel(void) if (!irqs_disabled()) printk(KERN_CRIT "start_kernel(): bug: interrupts were " "enabled early\n"); - early_boot_irqs_on(); + early_boot_irqs_disabled = false; local_irq_enable(); /* Interrupts are enabled now so all GFP allocations are safe. */ diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 42ba65dff7d9..0d2058da80f5 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -2291,22 +2291,6 @@ mark_held_locks(struct task_struct *curr, enum mark_type mark) return 1; } -/* - * Debugging helper: via this flag we know that we are in - * 'early bootup code', and will warn about any invalid irqs-on event: - */ -static int early_boot_irqs_enabled; - -void early_boot_irqs_off(void) -{ - early_boot_irqs_enabled = 0; -} - -void early_boot_irqs_on(void) -{ - early_boot_irqs_enabled = 1; -} - /* * Hardirqs will be enabled: */ @@ -2319,7 +2303,7 @@ void trace_hardirqs_on_caller(unsigned long ip) if (unlikely(!debug_locks || current->lockdep_recursion)) return; - if (DEBUG_LOCKS_WARN_ON(unlikely(!early_boot_irqs_enabled))) + if (DEBUG_LOCKS_WARN_ON(unlikely(early_boot_irqs_disabled))) return; if (unlikely(curr->hardirqs_enabled)) { diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c index 5cf8c602b880..92b6e1e12d98 100644 --- a/kernel/trace/trace_irqsoff.c +++ b/kernel/trace/trace_irqsoff.c @@ -453,14 +453,6 @@ void time_hardirqs_off(unsigned long a0, unsigned long a1) * Stubs: */ -void early_boot_irqs_off(void) -{ -} - -void early_boot_irqs_on(void) -{ -} - void trace_softirqs_on(unsigned long ip) { } -- cgit v1.2.3 From ca3e021417eed30ec2b64ce88eb0acf64aa9bc29 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Thu, 20 Jan 2011 14:44:24 -0800 Subject: memcg: fix USED bit handling at uncharge in THP Now, under THP: at charge: - PageCgroupUsed bit is set to all page_cgroup on a hugepage. ....set to 512 pages. at uncharge - PageCgroupUsed bit is unset on the head page. So, some pages will remain with "Used" bit. This patch fixes that Used bit is set only to the head page. Used bits for tail pages will be set at splitting if necessary. This patch adds this lock order: compound_lock() -> page_cgroup_move_lock(). [akpm@linux-foundation.org: fix warning] Signed-off-by: KAMEZAWA Hiroyuki Cc: Daisuke Nishimura Cc: Balbir Singh Cc: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memcontrol.h | 9 +++++ mm/huge_memory.c | 2 + mm/memcontrol.c | 91 ++++++++++++++++++++++++++-------------------- 3 files changed, 62 insertions(+), 40 deletions(-) (limited to 'include/linux') diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 6a576f989437..f512e189be5a 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -146,6 +146,10 @@ unsigned long mem_cgroup_soft_limit_reclaim(struct zone *zone, int order, gfp_t gfp_mask); u64 mem_cgroup_get_limit(struct mem_cgroup *mem); +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +void mem_cgroup_split_huge_fixup(struct page *head, struct page *tail); +#endif + #else /* CONFIG_CGROUP_MEM_RES_CTLR */ struct mem_cgroup; @@ -335,6 +339,11 @@ u64 mem_cgroup_get_limit(struct mem_cgroup *mem) return 0; } +static inline void mem_cgroup_split_huge_fixup(struct page *head, + struct page *tail) +{ +} + #endif /* CONFIG_CGROUP_MEM_CONT */ #endif /* _LINUX_MEMCONTROL_H */ diff --git a/mm/huge_memory.c b/mm/huge_memory.c index c4f634b3a48e..e187454d82f6 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1203,6 +1203,8 @@ static void __split_huge_page_refcount(struct page *page) BUG_ON(!PageDirty(page_tail)); BUG_ON(!PageSwapBacked(page_tail)); + mem_cgroup_split_huge_fixup(page, page_tail); + lru_add_page_tail(zone, page, page_tail); } diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 6d59a2bd520a..848b42195e5b 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1614,7 +1614,7 @@ void mem_cgroup_update_page_stat(struct page *page, if (unlikely(!mem || !PageCgroupUsed(pc))) goto out; /* pc->mem_cgroup is unstable ? */ - if (unlikely(mem_cgroup_stealed(mem))) { + if (unlikely(mem_cgroup_stealed(mem)) || PageTransHuge(page)) { /* take a lock against to access pc->mem_cgroup */ move_lock_page_cgroup(pc, &flags); need_unlock = true; @@ -2083,14 +2083,27 @@ struct mem_cgroup *try_get_mem_cgroup_from_page(struct page *page) return mem; } -/* - * commit a charge got by __mem_cgroup_try_charge() and makes page_cgroup to be - * USED state. If already USED, uncharge and return. - */ -static void ____mem_cgroup_commit_charge(struct mem_cgroup *mem, - struct page_cgroup *pc, - enum charge_type ctype) +static void __mem_cgroup_commit_charge(struct mem_cgroup *mem, + struct page_cgroup *pc, + enum charge_type ctype, + int page_size) { + int nr_pages = page_size >> PAGE_SHIFT; + + /* try_charge() can return NULL to *memcg, taking care of it. */ + if (!mem) + return; + + lock_page_cgroup(pc); + if (unlikely(PageCgroupUsed(pc))) { + unlock_page_cgroup(pc); + mem_cgroup_cancel_charge(mem, page_size); + return; + } + /* + * we don't need page_cgroup_lock about tail pages, becase they are not + * accessed by any other context at this point. + */ pc->mem_cgroup = mem; /* * We access a page_cgroup asynchronously without lock_page_cgroup(). @@ -2114,35 +2127,7 @@ static void ____mem_cgroup_commit_charge(struct mem_cgroup *mem, break; } - mem_cgroup_charge_statistics(mem, PageCgroupCache(pc), 1); -} - -static void __mem_cgroup_commit_charge(struct mem_cgroup *mem, - struct page_cgroup *pc, - enum charge_type ctype, - int page_size) -{ - int i; - int count = page_size >> PAGE_SHIFT; - - /* try_charge() can return NULL to *memcg, taking care of it. */ - if (!mem) - return; - - lock_page_cgroup(pc); - if (unlikely(PageCgroupUsed(pc))) { - unlock_page_cgroup(pc); - mem_cgroup_cancel_charge(mem, page_size); - return; - } - - /* - * we don't need page_cgroup_lock about tail pages, becase they are not - * accessed by any other context at this point. - */ - for (i = 0; i < count; i++) - ____mem_cgroup_commit_charge(mem, pc + i, ctype); - + mem_cgroup_charge_statistics(mem, PageCgroupCache(pc), nr_pages); unlock_page_cgroup(pc); /* * "charge_statistics" updated event counter. Then, check it. @@ -2152,6 +2137,34 @@ static void __mem_cgroup_commit_charge(struct mem_cgroup *mem, memcg_check_events(mem, pc->page); } +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + +#define PCGF_NOCOPY_AT_SPLIT ((1 << PCG_LOCK) | (1 << PCG_MOVE_LOCK) |\ + (1 << PCG_ACCT_LRU) | (1 << PCG_MIGRATION)) +/* + * Because tail pages are not marked as "used", set it. We're under + * zone->lru_lock, 'splitting on pmd' and compund_lock. + */ +void mem_cgroup_split_huge_fixup(struct page *head, struct page *tail) +{ + struct page_cgroup *head_pc = lookup_page_cgroup(head); + struct page_cgroup *tail_pc = lookup_page_cgroup(tail); + unsigned long flags; + + /* + * We have no races witch charge/uncharge but will have races with + * page state accounting. + */ + move_lock_page_cgroup(head_pc, &flags); + + tail_pc->mem_cgroup = head_pc->mem_cgroup; + smp_wmb(); /* see __commit_charge() */ + /* we don't need to copy all flags...*/ + tail_pc->flags = head_pc->flags & ~PCGF_NOCOPY_AT_SPLIT; + move_unlock_page_cgroup(head_pc, &flags); +} +#endif + /** * __mem_cgroup_move_account - move account of the page * @pc: page_cgroup of the page. @@ -2545,7 +2558,6 @@ direct_uncharge: static struct mem_cgroup * __mem_cgroup_uncharge_common(struct page *page, enum charge_type ctype) { - int i; int count; struct page_cgroup *pc; struct mem_cgroup *mem = NULL; @@ -2595,8 +2607,7 @@ __mem_cgroup_uncharge_common(struct page *page, enum charge_type ctype) break; } - for (i = 0; i < count; i++) - mem_cgroup_charge_statistics(mem, PageCgroupCache(pc), -1); + mem_cgroup_charge_statistics(mem, PageCgroupCache(pc), -count); ClearPageCgroupUsed(pc); /* -- cgit v1.2.3 From 2d6d9fd3a54a28c6f67f26eb6c74803307a1b11e Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 19 Jan 2011 22:27:14 +0100 Subject: ACPI: Introduce acpi_os_ioremap() Commit ca9b600be38c ("ACPI / PM: Make suspend_nvs_save() use acpi_os_map_memory()") attempted to prevent the code in osl.c and nvs.c from using different ioremap() variants by making the latter use acpi_os_map_memory() for mapping the NVS pages. However, that also requires acpi_os_unmap_memory() to be used for unmapping them, which causes synchronize_rcu() to be executed many times in a row unnecessarily and introduces substantial delays during resume on some systems. Instead of using acpi_os_map_memory() for mapping the NVS pages in nvs.c introduce acpi_os_ioremap() calling ioremap_cache() and make the code in both osl.c and nvs.c use it. Reported-by: Jeff Chua Signed-off-by: Rafael J. Wysocki Signed-off-by: Linus Torvalds --- drivers/acpi/nvs.c | 7 ++++--- drivers/acpi/osl.c | 12 +++++++----- include/linux/acpi.h | 3 --- include/linux/acpi_io.h | 16 ++++++++++++++++ 4 files changed, 27 insertions(+), 11 deletions(-) create mode 100644 include/linux/acpi_io.h (limited to 'include/linux') diff --git a/drivers/acpi/nvs.c b/drivers/acpi/nvs.c index 54b6ab8040a6..fa5a1df42b79 100644 --- a/drivers/acpi/nvs.c +++ b/drivers/acpi/nvs.c @@ -12,6 +12,7 @@ #include #include #include +#include #include /* @@ -80,7 +81,7 @@ void suspend_nvs_free(void) free_page((unsigned long)entry->data); entry->data = NULL; if (entry->kaddr) { - acpi_os_unmap_memory(entry->kaddr, entry->size); + iounmap(entry->kaddr); entry->kaddr = NULL; } } @@ -114,8 +115,8 @@ int suspend_nvs_save(void) list_for_each_entry(entry, &nvs_list, node) if (entry->data) { - entry->kaddr = acpi_os_map_memory(entry->phys_start, - entry->size); + entry->kaddr = acpi_os_ioremap(entry->phys_start, + entry->size); if (!entry->kaddr) { suspend_nvs_free(); return -ENOMEM; diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c index e2dd6de5d50c..b0931818cf98 100644 --- a/drivers/acpi/osl.c +++ b/drivers/acpi/osl.c @@ -38,6 +38,7 @@ #include #include #include +#include #include #include #include @@ -302,9 +303,10 @@ void __iomem *__init_refok acpi_os_map_memory(acpi_physical_address phys, acpi_size size) { struct acpi_ioremap *map, *tmp_map; - unsigned long flags, pg_sz; + unsigned long flags; void __iomem *virt; - phys_addr_t pg_off; + acpi_physical_address pg_off; + acpi_size pg_sz; if (phys > ULONG_MAX) { printk(KERN_ERR PREFIX "Cannot map memory that high\n"); @@ -320,7 +322,7 @@ acpi_os_map_memory(acpi_physical_address phys, acpi_size size) pg_off = round_down(phys, PAGE_SIZE); pg_sz = round_up(phys + size, PAGE_SIZE) - pg_off; - virt = ioremap_cache(pg_off, pg_sz); + virt = acpi_os_ioremap(pg_off, pg_sz); if (!virt) { kfree(map); return NULL; @@ -642,7 +644,7 @@ acpi_os_read_memory(acpi_physical_address phys_addr, u32 * value, u32 width) virt_addr = acpi_map_vaddr_lookup(phys_addr, size); rcu_read_unlock(); if (!virt_addr) { - virt_addr = ioremap_cache(phys_addr, size); + virt_addr = acpi_os_ioremap(phys_addr, size); unmap = 1; } if (!value) @@ -678,7 +680,7 @@ acpi_os_write_memory(acpi_physical_address phys_addr, u32 value, u32 width) virt_addr = acpi_map_vaddr_lookup(phys_addr, size); rcu_read_unlock(); if (!virt_addr) { - virt_addr = ioremap_cache(phys_addr, size); + virt_addr = acpi_os_ioremap(phys_addr, size); unmap = 1; } diff --git a/include/linux/acpi.h b/include/linux/acpi.h index eb176bb1b15b..a2e910e01293 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -306,9 +306,6 @@ extern acpi_status acpi_pci_osc_control_set(acpi_handle handle, u32 *mask, u32 req); extern void acpi_early_init(void); -int acpi_os_map_generic_address(struct acpi_generic_address *addr); -void acpi_os_unmap_generic_address(struct acpi_generic_address *addr); - #else /* !CONFIG_ACPI */ #define acpi_disabled 1 diff --git a/include/linux/acpi_io.h b/include/linux/acpi_io.h new file mode 100644 index 000000000000..7180013a4a3a --- /dev/null +++ b/include/linux/acpi_io.h @@ -0,0 +1,16 @@ +#ifndef _ACPI_IO_H_ +#define _ACPI_IO_H_ + +#include +#include + +static inline void __iomem *acpi_os_ioremap(acpi_physical_address phys, + acpi_size size) +{ + return ioremap_cache(phys, size); +} + +int acpi_os_map_generic_address(struct acpi_generic_address *addr); +void acpi_os_unmap_generic_address(struct acpi_generic_address *addr); + +#endif -- cgit v1.2.3 From 1c77ff22f539ceaa64ea43d6a26d867e84602cb7 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 19 Jan 2011 19:41:35 +0100 Subject: genirq: Remove __do_IRQ All architectures are finally converted. Remove the cruft. Signed-off-by: Thomas Gleixner Cc: Richard Henderson Cc: Mike Frysinger Cc: David Howells Cc: Tony Luck Cc: Greg Ungerer Cc: Michal Simek Acked-by: David Howells Cc: Kyle McMartin Acked-by: Benjamin Herrenschmidt Cc: Chen Liqin Cc: "David S. Miller" Cc: Chris Metcalf Cc: Jeff Dike --- Documentation/feature-removal-schedule.txt | 8 --- arch/alpha/Kconfig | 3 - arch/blackfin/Kconfig | 3 - arch/frv/Kconfig | 4 -- arch/ia64/Kconfig | 3 - arch/m68knommu/Kconfig | 4 -- arch/microblaze/Kconfig | 3 - arch/mips/Kconfig | 3 - arch/mn10300/Kconfig | 3 - arch/parisc/Kconfig | 4 -- arch/powerpc/Kconfig | 4 -- arch/score/Kconfig | 3 - arch/sparc/Kconfig | 4 -- arch/tile/Kconfig | 3 - arch/um/Kconfig.um | 3 - include/linux/irqdesc.h | 14 ---- kernel/irq/Kconfig | 3 - kernel/irq/handle.c | 111 ----------------------------- 18 files changed, 183 deletions(-) (limited to 'include/linux') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 8c594c45b6a1..b959659c5df4 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -357,14 +357,6 @@ Who: Dave Jones , Matthew Garrett ----------------------------- -What: __do_IRQ all in one fits nothing interrupt handler -When: 2.6.32 -Why: __do_IRQ was kept for easy migration to the type flow handlers. - More than two years of migration time is enough. -Who: Thomas Gleixner - ------------------------------ - What: fakephp and associated sysfs files in /sys/bus/pci/slots/ When: 2011 Why: In 2.6.27, the semantics of /sys/bus/pci/slots was redefined to diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig index fc95ee1bcf6f..943fe6930f77 100644 --- a/arch/alpha/Kconfig +++ b/arch/alpha/Kconfig @@ -68,9 +68,6 @@ config GENERIC_IOMAP bool default n -config GENERIC_HARDIRQS_NO__DO_IRQ - def_bool y - config GENERIC_HARDIRQS bool default y diff --git a/arch/blackfin/Kconfig b/arch/blackfin/Kconfig index 0a221d48152d..a37b2be23f18 100644 --- a/arch/blackfin/Kconfig +++ b/arch/blackfin/Kconfig @@ -50,9 +50,6 @@ config GENERIC_HARDIRQS config GENERIC_IRQ_PROBE def_bool y -config GENERIC_HARDIRQS_NO__DO_IRQ - def_bool y - config GENERIC_GPIO def_bool y diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig index f6bcb039cd6d..e504edeb3d84 100644 --- a/arch/frv/Kconfig +++ b/arch/frv/Kconfig @@ -33,10 +33,6 @@ config GENERIC_HARDIRQS bool default y -config GENERIC_HARDIRQS_NO__DO_IRQ - bool - default y - config TIME_LOW_RES bool default y diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index e0f5b6d7f849..be1faf991813 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -684,9 +684,6 @@ source "lib/Kconfig" config GENERIC_HARDIRQS def_bool y -config GENERIC_HARDIRQS_NO__DO_IRQ - def_bool y - config GENERIC_IRQ_PROBE bool default y diff --git a/arch/m68knommu/Kconfig b/arch/m68knommu/Kconfig index 704e7b92334c..7379cb0ce1af 100644 --- a/arch/m68knommu/Kconfig +++ b/arch/m68knommu/Kconfig @@ -52,10 +52,6 @@ config GENERIC_HARDIRQS bool default y -config GENERIC_HARDIRQS_NO__DO_IRQ - bool - default y - config GENERIC_CALIBRATE_DELAY bool default y diff --git a/arch/microblaze/Kconfig b/arch/microblaze/Kconfig index 5f5018a71a3d..5a6378493797 100644 --- a/arch/microblaze/Kconfig +++ b/arch/microblaze/Kconfig @@ -52,9 +52,6 @@ config GENERIC_TIME_VSYSCALL config GENERIC_CLOCKEVENTS def_bool y -config GENERIC_HARDIRQS_NO__DO_IRQ - def_bool y - config GENERIC_GPIO def_bool y diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 548e6cc3bc28..f5ecc0566bc2 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -793,9 +793,6 @@ config SCHED_OMIT_FRAME_POINTER bool default y -config GENERIC_HARDIRQS_NO__DO_IRQ - def_bool y - # # Select some configuration options automatically based on user selections. # diff --git a/arch/mn10300/Kconfig b/arch/mn10300/Kconfig index 8ed41cf2b08d..4638269152f6 100644 --- a/arch/mn10300/Kconfig +++ b/arch/mn10300/Kconfig @@ -34,9 +34,6 @@ config RWSEM_GENERIC_SPINLOCK config RWSEM_XCHGADD_ALGORITHM bool -config GENERIC_HARDIRQS_NO__DO_IRQ - def_bool y - config GENERIC_CALIBRATE_DELAY def_bool y diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig index 0888675c98dd..4b94ac4dc603 100644 --- a/arch/parisc/Kconfig +++ b/arch/parisc/Kconfig @@ -12,7 +12,6 @@ config PARISC select HAVE_IRQ_WORK select HAVE_PERF_EVENTS select GENERIC_ATOMIC64 if !64BIT - select GENERIC_HARDIRQS_NO__DO_IRQ help The PA-RISC microprocessor is designed by Hewlett-Packard and used in many of their workstations & servers (HP9000 700 and 800 series, @@ -79,9 +78,6 @@ config IRQ_PER_CPU bool default y -config GENERIC_HARDIRQS_NO__DO_IRQ - def_bool y - # unless you want to implement ACPI on PA-RISC ... ;-) config PM bool diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 959f38ccb9a7..e0b185dd718a 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -40,10 +40,6 @@ config GENERIC_HARDIRQS bool default y -config GENERIC_HARDIRQS_NO__DO_IRQ - bool - default y - config HAVE_SETUP_PER_CPU_AREA def_bool PPC64 diff --git a/arch/score/Kconfig b/arch/score/Kconfig index 4293fdcb5398..4f159acfbe33 100644 --- a/arch/score/Kconfig +++ b/arch/score/Kconfig @@ -53,9 +53,6 @@ config GENERIC_CLOCKEVENTS config SCHED_NO_NO_OMIT_FRAME_POINTER def_bool y -config GENERIC_HARDIRQS_NO__DO_IRQ - def_bool y - config GENERIC_SYSCALL_TABLE def_bool y diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig index 45d9c87d083a..989bb6415ea3 100644 --- a/arch/sparc/Kconfig +++ b/arch/sparc/Kconfig @@ -107,10 +107,6 @@ config NEED_PER_CPU_EMBED_FIRST_CHUNK config NEED_PER_CPU_PAGE_FIRST_CHUNK def_bool y if SPARC64 -config GENERIC_HARDIRQS_NO__DO_IRQ - bool - def_bool y if SPARC64 - config MMU bool default y diff --git a/arch/tile/Kconfig b/arch/tile/Kconfig index 4e8b82bca9e5..c16b98c2435d 100644 --- a/arch/tile/Kconfig +++ b/arch/tile/Kconfig @@ -10,9 +10,6 @@ config GENERIC_CSUM config GENERIC_HARDIRQS def_bool y -config GENERIC_HARDIRQS_NO__DO_IRQ - def_bool y - config GENERIC_IRQ_PROBE def_bool y diff --git a/arch/um/Kconfig.um b/arch/um/Kconfig.um index f8d1d0d47fe6..90a438acbfaf 100644 --- a/arch/um/Kconfig.um +++ b/arch/um/Kconfig.um @@ -120,9 +120,6 @@ config SMP If you don't know what to do, say N. -config GENERIC_HARDIRQS_NO__DO_IRQ - def_bool y - config NR_CPUS int "Maximum number of CPUs (2-32)" range 2 32 diff --git a/include/linux/irqdesc.h b/include/linux/irqdesc.h index 6a64c6fa81af..c1a95b7b58de 100644 --- a/include/linux/irqdesc.h +++ b/include/linux/irqdesc.h @@ -100,13 +100,6 @@ static inline struct irq_desc *move_irq_desc(struct irq_desc *desc, int node) #define get_irq_desc_data(desc) ((desc)->irq_data.handler_data) #define get_irq_desc_msi(desc) ((desc)->irq_data.msi_desc) -/* - * Monolithic do_IRQ implementation. - */ -#ifndef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ -extern unsigned int __do_IRQ(unsigned int irq); -#endif - /* * Architectures call this to let the generic IRQ layer * handle an interrupt. If the descriptor is attached to an @@ -115,14 +108,7 @@ extern unsigned int __do_IRQ(unsigned int irq); */ static inline void generic_handle_irq_desc(unsigned int irq, struct irq_desc *desc) { -#ifdef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ desc->handle_irq(irq, desc); -#else - if (likely(desc->handle_irq)) - desc->handle_irq(irq, desc); - else - __do_IRQ(irq); -#endif } static inline void generic_handle_irq(unsigned int irq) diff --git a/kernel/irq/Kconfig b/kernel/irq/Kconfig index 31d766bf5d2e..8e42fec7686d 100644 --- a/kernel/irq/Kconfig +++ b/kernel/irq/Kconfig @@ -9,9 +9,6 @@ menu "IRQ subsystem" config GENERIC_HARDIRQS def_bool y -config GENERIC_HARDIRQS_NO__DO_IRQ - def_bool y - # Select this to disable the deprecated stuff config GENERIC_HARDIRQS_NO_DEPRECATED def_bool n diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index e2347eb63306..3540a7190122 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -118,114 +118,3 @@ irqreturn_t handle_IRQ_event(unsigned int irq, struct irqaction *action) return retval; } - -#ifndef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ - -#ifdef CONFIG_ENABLE_WARN_DEPRECATED -# warning __do_IRQ is deprecated. Please convert to proper flow handlers -#endif - -/** - * __do_IRQ - original all in one highlevel IRQ handler - * @irq: the interrupt number - * - * __do_IRQ handles all normal device IRQ's (the special - * SMP cross-CPU interrupts have their own specific - * handlers). - * - * This is the original x86 implementation which is used for every - * interrupt type. - */ -unsigned int __do_IRQ(unsigned int irq) -{ - struct irq_desc *desc = irq_to_desc(irq); - struct irqaction *action; - unsigned int status; - - kstat_incr_irqs_this_cpu(irq, desc); - - if (CHECK_IRQ_PER_CPU(desc->status)) { - irqreturn_t action_ret; - - /* - * No locking required for CPU-local interrupts: - */ - if (desc->irq_data.chip->ack) - desc->irq_data.chip->ack(irq); - if (likely(!(desc->status & IRQ_DISABLED))) { - action_ret = handle_IRQ_event(irq, desc->action); - if (!noirqdebug) - note_interrupt(irq, desc, action_ret); - } - desc->irq_data.chip->end(irq); - return 1; - } - - raw_spin_lock(&desc->lock); - if (desc->irq_data.chip->ack) - desc->irq_data.chip->ack(irq); - /* - * REPLAY is when Linux resends an IRQ that was dropped earlier - * WAITING is used by probe to mark irqs that are being tested - */ - status = desc->status & ~(IRQ_REPLAY | IRQ_WAITING); - status |= IRQ_PENDING; /* we _want_ to handle it */ - - /* - * If the IRQ is disabled for whatever reason, we cannot - * use the action we have. - */ - action = NULL; - if (likely(!(status & (IRQ_DISABLED | IRQ_INPROGRESS)))) { - action = desc->action; - status &= ~IRQ_PENDING; /* we commit to handling */ - status |= IRQ_INPROGRESS; /* we are handling it */ - } - desc->status = status; - - /* - * If there is no IRQ handler or it was disabled, exit early. - * Since we set PENDING, if another processor is handling - * a different instance of this same irq, the other processor - * will take care of it. - */ - if (unlikely(!action)) - goto out; - - /* - * Edge triggered interrupts need to remember - * pending events. - * This applies to any hw interrupts that allow a second - * instance of the same irq to arrive while we are in do_IRQ - * or in the handler. But the code here only handles the _second_ - * instance of the irq, not the third or fourth. So it is mostly - * useful for irq hardware that does not mask cleanly in an - * SMP environment. - */ - for (;;) { - irqreturn_t action_ret; - - raw_spin_unlock(&desc->lock); - - action_ret = handle_IRQ_event(irq, action); - if (!noirqdebug) - note_interrupt(irq, desc, action_ret); - - raw_spin_lock(&desc->lock); - if (likely(!(desc->status & IRQ_PENDING))) - break; - desc->status &= ~IRQ_PENDING; - } - desc->status &= ~IRQ_INPROGRESS; - -out: - /* - * The ->end() handler has to deal with interrupts which got - * disabled while the handler was running. - */ - desc->irq_data.chip->end(irq); - raw_spin_unlock(&desc->lock); - - return 1; -} -#endif -- cgit v1.2.3 From 1daeddd5962acad1bea55e524fc0fadf32654a21 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Thu, 13 Jan 2011 09:30:49 -0800 Subject: rtc: Cleanup removed UIE emulation declaration rtc_dev_update_irq_enable_emul was removed in commit 042620a018afcfba1d678062b62e463b9e43a68d (UIE emulation is now handled via hrtimer), but the declaration was missed. This patch cleans it up. Signed-off-by: John Stultz LKML-Reference: <1294939849-20608-1-git-send-email-john.stultz@linaro.org> Signed-off-by: Thomas Gleixner --- include/linux/rtc.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rtc.h b/include/linux/rtc.h index 3c995b4d742c..ea760698e6fe 100644 --- a/include/linux/rtc.h +++ b/include/linux/rtc.h @@ -235,8 +235,6 @@ extern int rtc_irq_set_freq(struct rtc_device *rtc, struct rtc_task *task, int freq); extern int rtc_update_irq_enable(struct rtc_device *rtc, unsigned int enabled); extern int rtc_alarm_irq_enable(struct rtc_device *rtc, unsigned int enabled); -extern int rtc_dev_update_irq_enable_emul(struct rtc_device *rtc, - unsigned int enabled); void rtc_aie_update_irq(void *private); void rtc_uie_update_irq(void *private); -- cgit v1.2.3 From aa0be0f4659f91f31e45adc422b1788cb36ffddc Mon Sep 17 00:00:00 2001 From: John Stultz Date: Thu, 20 Jan 2011 15:26:12 -0800 Subject: RTC: Propagate error handling via rtc_timer_enqueue properly In cases where RTC hardware does not support alarms, the virtualized RTC interfaces did not have a way to propagate the error up to userland. This patch extends rtc_timer_enqueue so it catches errors from the hardware and returns them upwards to the virtualized interfaces. To simplify error handling, it also internalizes the management of the timer->enabled bit into rtc_timer_enqueue and rtc_timer_remove. Also makes rtc_timer_enqueue and rtc_timer_remove static. Reported-by: David Daney Reported-by: Andreas Schwab Reported-by: Geert Uytterhoeven Diagnosed-by: David Daney Tested-by: David Daney Signed-off-by: John Stultz LKML-Reference: <1295565973-14358-1-git-send-email-john.stultz@linaro.org> Signed-off-by: Thomas Gleixner --- drivers/rtc/interface.c | 49 ++++++++++++++++++++++++++++--------------------- include/linux/rtc.h | 2 -- 2 files changed, 28 insertions(+), 23 deletions(-) (limited to 'include/linux') diff --git a/drivers/rtc/interface.c b/drivers/rtc/interface.c index 90384b9f6b2c..f1ba2c696528 100644 --- a/drivers/rtc/interface.c +++ b/drivers/rtc/interface.c @@ -16,6 +16,9 @@ #include #include +static int rtc_timer_enqueue(struct rtc_device *rtc, struct rtc_timer *timer); +static void rtc_timer_remove(struct rtc_device *rtc, struct rtc_timer *timer); + static int __rtc_read_time(struct rtc_device *rtc, struct rtc_time *tm) { int err; @@ -175,16 +178,14 @@ int rtc_set_alarm(struct rtc_device *rtc, struct rtc_wkalrm *alarm) return err; if (rtc->aie_timer.enabled) { rtc_timer_remove(rtc, &rtc->aie_timer); - rtc->aie_timer.enabled = 0; } rtc->aie_timer.node.expires = rtc_tm_to_ktime(alarm->time); rtc->aie_timer.period = ktime_set(0, 0); if (alarm->enabled) { - rtc->aie_timer.enabled = 1; - rtc_timer_enqueue(rtc, &rtc->aie_timer); + err = rtc_timer_enqueue(rtc, &rtc->aie_timer); } mutex_unlock(&rtc->ops_lock); - return 0; + return err; } EXPORT_SYMBOL_GPL(rtc_set_alarm); @@ -195,15 +196,15 @@ int rtc_alarm_irq_enable(struct rtc_device *rtc, unsigned int enabled) return err; if (rtc->aie_timer.enabled != enabled) { - if (enabled) { - rtc->aie_timer.enabled = 1; - rtc_timer_enqueue(rtc, &rtc->aie_timer); - } else { + if (enabled) + err = rtc_timer_enqueue(rtc, &rtc->aie_timer); + else rtc_timer_remove(rtc, &rtc->aie_timer); - rtc->aie_timer.enabled = 0; - } } + if (err) + return err; + if (!rtc->ops) err = -ENODEV; else if (!rtc->ops->alarm_irq_enable) @@ -235,12 +236,9 @@ int rtc_update_irq_enable(struct rtc_device *rtc, unsigned int enabled) now = rtc_tm_to_ktime(tm); rtc->uie_rtctimer.node.expires = ktime_add(now, onesec); rtc->uie_rtctimer.period = ktime_set(1, 0); - rtc->uie_rtctimer.enabled = 1; - rtc_timer_enqueue(rtc, &rtc->uie_rtctimer); - } else { + err = rtc_timer_enqueue(rtc, &rtc->uie_rtctimer); + } else rtc_timer_remove(rtc, &rtc->uie_rtctimer); - rtc->uie_rtctimer.enabled = 0; - } out: mutex_unlock(&rtc->ops_lock); @@ -488,10 +486,13 @@ EXPORT_SYMBOL_GPL(rtc_irq_set_freq); * Enqueues a timer onto the rtc devices timerqueue and sets * the next alarm event appropriately. * + * Sets the enabled bit on the added timer. + * * Must hold ops_lock for proper serialization of timerqueue */ -void rtc_timer_enqueue(struct rtc_device *rtc, struct rtc_timer *timer) +static int rtc_timer_enqueue(struct rtc_device *rtc, struct rtc_timer *timer) { + timer->enabled = 1; timerqueue_add(&rtc->timerqueue, &timer->node); if (&timer->node == timerqueue_getnext(&rtc->timerqueue)) { struct rtc_wkalrm alarm; @@ -501,7 +502,13 @@ void rtc_timer_enqueue(struct rtc_device *rtc, struct rtc_timer *timer) err = __rtc_set_alarm(rtc, &alarm); if (err == -ETIME) schedule_work(&rtc->irqwork); + else if (err) { + timerqueue_del(&rtc->timerqueue, &timer->node); + timer->enabled = 0; + return err; + } } + return 0; } /** @@ -512,13 +519,15 @@ void rtc_timer_enqueue(struct rtc_device *rtc, struct rtc_timer *timer) * Removes a timer onto the rtc devices timerqueue and sets * the next alarm event appropriately. * + * Clears the enabled bit on the removed timer. + * * Must hold ops_lock for proper serialization of timerqueue */ -void rtc_timer_remove(struct rtc_device *rtc, struct rtc_timer *timer) +static void rtc_timer_remove(struct rtc_device *rtc, struct rtc_timer *timer) { struct timerqueue_node *next = timerqueue_getnext(&rtc->timerqueue); timerqueue_del(&rtc->timerqueue, &timer->node); - + timer->enabled = 0; if (next == &timer->node) { struct rtc_wkalrm alarm; int err; @@ -626,8 +635,7 @@ int rtc_timer_start(struct rtc_device *rtc, struct rtc_timer* timer, timer->node.expires = expires; timer->period = period; - timer->enabled = 1; - rtc_timer_enqueue(rtc, timer); + ret = rtc_timer_enqueue(rtc, timer); mutex_unlock(&rtc->ops_lock); return ret; @@ -645,7 +653,6 @@ int rtc_timer_cancel(struct rtc_device *rtc, struct rtc_timer* timer) mutex_lock(&rtc->ops_lock); if (timer->enabled) rtc_timer_remove(rtc, timer); - timer->enabled = 0; mutex_unlock(&rtc->ops_lock); return ret; } diff --git a/include/linux/rtc.h b/include/linux/rtc.h index ea760698e6fe..a0b639f8e805 100644 --- a/include/linux/rtc.h +++ b/include/linux/rtc.h @@ -244,8 +244,6 @@ int rtc_register(rtc_task_t *task); int rtc_unregister(rtc_task_t *task); int rtc_control(rtc_task_t *t, unsigned int cmd, unsigned long arg); -void rtc_timer_enqueue(struct rtc_device *rtc, struct rtc_timer *timer); -void rtc_timer_remove(struct rtc_device *rtc, struct rtc_timer *timer); void rtc_timer_init(struct rtc_timer *timer, void (*f)(void* p), void* data); int rtc_timer_start(struct rtc_device *rtc, struct rtc_timer* timer, ktime_t expires, ktime_t period); -- cgit v1.2.3 From 3dece370ecc7c6152b3fdd21c6de28179f6e643d Mon Sep 17 00:00:00 2001 From: Michal Simek Date: Fri, 21 Jan 2011 08:49:56 +0100 Subject: mm: System without MMU do not need pte_mkwrite The patch "thp: export maybe_mkwrite" (commit 14fd403f2146) breaks systems without MMU. Error log: CC arch/microblaze/mm/init.o In file included from include/linux/mman.h:14, from arch/microblaze/mm/consistent.c:24: include/linux/mm.h: In function 'maybe_mkwrite': include/linux/mm.h:482: error: implicit declaration of function 'pte_mkwrite' include/linux/mm.h:482: error: incompatible types in assignment Signed-off-by: Michal Simek CC: Andrea Arcangeli Reviewed-by: Rik van Riel CC: Andrew Morton CC: Ingo Molnar Signed-off-by: Linus Torvalds --- include/linux/mm.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/mm.h b/include/linux/mm.h index 956a35532f47..f6385fc17ad4 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -470,6 +470,7 @@ static inline void set_compound_order(struct page *page, unsigned long order) page[1].lru.prev = (void *)order; } +#ifdef CONFIG_MMU /* * Do pte_mkwrite, but only if the vma says VM_WRITE. We do this when * servicing faults for write access. In the normal case, do always want @@ -482,6 +483,7 @@ static inline pte_t maybe_mkwrite(pte_t pte, struct vm_area_struct *vma) pte = pte_mkwrite(pte); return pte; } +#endif /* * Multiple processes may "see" the same page. E.g. for untouched -- cgit v1.2.3 From d14fc1a74e846d7851f24fc9519fe87dc12a1231 Mon Sep 17 00:00:00 2001 From: Libor Pechacek Date: Fri, 14 Jan 2011 14:30:21 +0100 Subject: USB: serial: handle Data Carrier Detect changes Alan's commit 335f8514f200e63d689113d29cb7253a5c282967 introduced .carrier_raised function in several drivers. That also means tty_port_block_til_ready can now suspend the process trying to open the serial port when Carrier Detect is low and put it into tty_port.open_wait queue. We need to wake up the process when Carrier Detect goes high and trigger TTY hangup when CD goes low. Some of the devices do not report modem status line changes, or at least we don't understand the status message, so for those we remove .carrier_raised again. Signed-off-by: Libor Pechacek Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/serial/ch341.c | 10 ++++++++++ drivers/usb/serial/cp210x.c | 13 +------------ drivers/usb/serial/digi_acceleport.c | 10 ---------- drivers/usb/serial/generic.c | 20 ++++++++++++++++++++ drivers/usb/serial/keyspan_pda.c | 17 ----------------- drivers/usb/serial/pl2303.c | 11 +++++++++++ drivers/usb/serial/spcp8x5.c | 6 +++++- include/linux/usb/serial.h | 3 +++ 8 files changed, 50 insertions(+), 40 deletions(-) (limited to 'include/linux') diff --git a/drivers/usb/serial/ch341.c b/drivers/usb/serial/ch341.c index 63f7cc45bcac..7b8815ddf368 100644 --- a/drivers/usb/serial/ch341.c +++ b/drivers/usb/serial/ch341.c @@ -486,12 +486,22 @@ static void ch341_read_int_callback(struct urb *urb) if (actual_length >= 4) { struct ch341_private *priv = usb_get_serial_port_data(port); unsigned long flags; + u8 prev_line_status = priv->line_status; spin_lock_irqsave(&priv->lock, flags); priv->line_status = (~(data[2])) & CH341_BITS_MODEM_STAT; if ((data[1] & CH341_MULT_STAT)) priv->multi_status_change = 1; spin_unlock_irqrestore(&priv->lock, flags); + + if ((priv->line_status ^ prev_line_status) & CH341_BIT_DCD) { + struct tty_struct *tty = tty_port_tty_get(&port->port); + if (tty) + usb_serial_handle_dcd_change(port, tty, + priv->line_status & CH341_BIT_DCD); + tty_kref_put(tty); + } + wake_up_interruptible(&priv->delta_msr_wait); } diff --git a/drivers/usb/serial/cp210x.c b/drivers/usb/serial/cp210x.c index c3bea46e87fb..735ea03157ab 100644 --- a/drivers/usb/serial/cp210x.c +++ b/drivers/usb/serial/cp210x.c @@ -49,7 +49,6 @@ static int cp210x_tiocmset_port(struct usb_serial_port *port, struct file *, static void cp210x_break_ctl(struct tty_struct *, int); static int cp210x_startup(struct usb_serial *); static void cp210x_dtr_rts(struct usb_serial_port *p, int on); -static int cp210x_carrier_raised(struct usb_serial_port *p); static int debug; @@ -166,8 +165,7 @@ static struct usb_serial_driver cp210x_device = { .tiocmget = cp210x_tiocmget, .tiocmset = cp210x_tiocmset, .attach = cp210x_startup, - .dtr_rts = cp210x_dtr_rts, - .carrier_raised = cp210x_carrier_raised + .dtr_rts = cp210x_dtr_rts }; /* Config request types */ @@ -766,15 +764,6 @@ static int cp210x_tiocmget (struct tty_struct *tty, struct file *file) return result; } -static int cp210x_carrier_raised(struct usb_serial_port *p) -{ - unsigned int control; - cp210x_get_config(p, CP210X_GET_MDMSTS, &control, 1); - if (control & CONTROL_DCD) - return 1; - return 0; -} - static void cp210x_break_ctl (struct tty_struct *tty, int break_state) { struct usb_serial_port *port = tty->driver_data; diff --git a/drivers/usb/serial/digi_acceleport.c b/drivers/usb/serial/digi_acceleport.c index b92070c103cd..666e5a6edd82 100644 --- a/drivers/usb/serial/digi_acceleport.c +++ b/drivers/usb/serial/digi_acceleport.c @@ -455,7 +455,6 @@ static int digi_write_room(struct tty_struct *tty); static int digi_chars_in_buffer(struct tty_struct *tty); static int digi_open(struct tty_struct *tty, struct usb_serial_port *port); static void digi_close(struct usb_serial_port *port); -static int digi_carrier_raised(struct usb_serial_port *port); static void digi_dtr_rts(struct usb_serial_port *port, int on); static int digi_startup_device(struct usb_serial *serial); static int digi_startup(struct usb_serial *serial); @@ -511,7 +510,6 @@ static struct usb_serial_driver digi_acceleport_2_device = { .open = digi_open, .close = digi_close, .dtr_rts = digi_dtr_rts, - .carrier_raised = digi_carrier_raised, .write = digi_write, .write_room = digi_write_room, .write_bulk_callback = digi_write_bulk_callback, @@ -1339,14 +1337,6 @@ static void digi_dtr_rts(struct usb_serial_port *port, int on) digi_set_modem_signals(port, on * (TIOCM_DTR|TIOCM_RTS), 1); } -static int digi_carrier_raised(struct usb_serial_port *port) -{ - struct digi_port *priv = usb_get_serial_port_data(port); - if (priv->dp_modem_signals & TIOCM_CD) - return 1; - return 0; -} - static int digi_open(struct tty_struct *tty, struct usb_serial_port *port) { int ret; diff --git a/drivers/usb/serial/generic.c b/drivers/usb/serial/generic.c index e6833e216fc9..e4db5ad2bc55 100644 --- a/drivers/usb/serial/generic.c +++ b/drivers/usb/serial/generic.c @@ -479,6 +479,26 @@ int usb_serial_handle_break(struct usb_serial_port *port) } EXPORT_SYMBOL_GPL(usb_serial_handle_break); +/** + * usb_serial_handle_dcd_change - handle a change of carrier detect state + * @port: usb_serial_port structure for the open port + * @tty: tty_struct structure for the port + * @status: new carrier detect status, nonzero if active + */ +void usb_serial_handle_dcd_change(struct usb_serial_port *usb_port, + struct tty_struct *tty, unsigned int status) +{ + struct tty_port *port = &usb_port->port; + + dbg("%s - port %d, status %d", __func__, usb_port->number, status); + + if (status) + wake_up_interruptible(&port->open_wait); + else if (tty && !C_CLOCAL(tty)) + tty_hangup(tty); +} +EXPORT_SYMBOL_GPL(usb_serial_handle_dcd_change); + int usb_serial_generic_resume(struct usb_serial *serial) { struct usb_serial_port *port; diff --git a/drivers/usb/serial/keyspan_pda.c b/drivers/usb/serial/keyspan_pda.c index a10dd5676ccc..554a8693a463 100644 --- a/drivers/usb/serial/keyspan_pda.c +++ b/drivers/usb/serial/keyspan_pda.c @@ -679,22 +679,6 @@ static void keyspan_pda_dtr_rts(struct usb_serial_port *port, int on) } } -static int keyspan_pda_carrier_raised(struct usb_serial_port *port) -{ - struct usb_serial *serial = port->serial; - unsigned char modembits; - - /* If we can read the modem status and the DCD is low then - carrier is not raised yet */ - if (keyspan_pda_get_modem_info(serial, &modembits) >= 0) { - if (!(modembits & (1>>6))) - return 0; - } - /* Carrier raised, or we failed (eg disconnected) so - progress accordingly */ - return 1; -} - static int keyspan_pda_open(struct tty_struct *tty, struct usb_serial_port *port) @@ -881,7 +865,6 @@ static struct usb_serial_driver keyspan_pda_device = { .id_table = id_table_std, .num_ports = 1, .dtr_rts = keyspan_pda_dtr_rts, - .carrier_raised = keyspan_pda_carrier_raised, .open = keyspan_pda_open, .close = keyspan_pda_close, .write = keyspan_pda_write, diff --git a/drivers/usb/serial/pl2303.c b/drivers/usb/serial/pl2303.c index 45c40e6e9737..08c9181b8e48 100644 --- a/drivers/usb/serial/pl2303.c +++ b/drivers/usb/serial/pl2303.c @@ -678,9 +678,11 @@ static void pl2303_update_line_status(struct usb_serial_port *port, { struct pl2303_private *priv = usb_get_serial_port_data(port); + struct tty_struct *tty; unsigned long flags; u8 status_idx = UART_STATE; u8 length = UART_STATE + 1; + u8 prev_line_status; u16 idv, idp; idv = le16_to_cpu(port->serial->dev->descriptor.idVendor); @@ -702,11 +704,20 @@ static void pl2303_update_line_status(struct usb_serial_port *port, /* Save off the uart status for others to look at */ spin_lock_irqsave(&priv->lock, flags); + prev_line_status = priv->line_status; priv->line_status = data[status_idx]; spin_unlock_irqrestore(&priv->lock, flags); if (priv->line_status & UART_BREAK_ERROR) usb_serial_handle_break(port); wake_up_interruptible(&priv->delta_msr_wait); + + tty = tty_port_tty_get(&port->port); + if (!tty) + return; + if ((priv->line_status ^ prev_line_status) & UART_DCD) + usb_serial_handle_dcd_change(port, tty, + priv->line_status & UART_DCD); + tty_kref_put(tty); } static void pl2303_read_int_callback(struct urb *urb) diff --git a/drivers/usb/serial/spcp8x5.c b/drivers/usb/serial/spcp8x5.c index c7665d39d192..cbfb70bffdd0 100644 --- a/drivers/usb/serial/spcp8x5.c +++ b/drivers/usb/serial/spcp8x5.c @@ -133,7 +133,7 @@ struct spcp8x5_usb_ctrl_arg { /* how come ??? */ #define UART_STATE 0x08 -#define UART_STATE_TRANSIENT_MASK 0x74 +#define UART_STATE_TRANSIENT_MASK 0x75 #define UART_DCD 0x01 #define UART_DSR 0x02 #define UART_BREAK_ERROR 0x04 @@ -525,6 +525,10 @@ static void spcp8x5_process_read_urb(struct urb *urb) /* overrun is special, not associated with a char */ if (status & UART_OVERRUN_ERROR) tty_insert_flip_char(tty, 0, TTY_OVERRUN); + + if (status & UART_DCD) + usb_serial_handle_dcd_change(port, tty, + priv->line_status & MSR_STATUS_LINE_DCD); } tty_insert_flip_string_fixed_flag(tty, data, tty_flag, diff --git a/include/linux/usb/serial.h b/include/linux/usb/serial.h index 16d682f4f7c3..c9049139a7a5 100644 --- a/include/linux/usb/serial.h +++ b/include/linux/usb/serial.h @@ -347,6 +347,9 @@ extern int usb_serial_generic_prepare_write_buffer(struct usb_serial_port *port, extern int usb_serial_handle_sysrq_char(struct usb_serial_port *port, unsigned int ch); extern int usb_serial_handle_break(struct usb_serial_port *port); +extern void usb_serial_handle_dcd_change(struct usb_serial_port *usb_port, + struct tty_struct *tty, + unsigned int status); extern int usb_serial_bus_register(struct usb_serial_driver *device); -- cgit v1.2.3 From e94965ed5beb23c6fabf7ed31f625e66d7ff28de Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Wed, 15 Dec 2010 14:00:19 -0800 Subject: module: show version information for built-in modules in sysfs Currently only drivers that are built as modules have their versions shown in /sys/module//version, but this information might also be useful for built-in drivers as well. This especially important for drivers that do not define any parameters - such drivers, if built-in, are completely invisible from userspace. This patch changes MODULE_VERSION() macro so that in case when we are compiling built-in module, version information is stored in a separate section. Kernel then uses this data to create 'version' sysfs attribute in the same fashion it creates attributes for module parameters. Signed-off-by: Dmitry Torokhov Signed-off-by: Rusty Russell --- include/asm-generic/vmlinux.lds.h | 7 +++++ include/linux/module.h | 27 ++++++++++++++++ kernel/params.c | 65 ++++++++++++++++++++++++++++++++------- 3 files changed, 88 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index 68649336c4ad..6ebb81030d2d 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -364,6 +364,13 @@ VMLINUX_SYMBOL(__start___param) = .; \ *(__param) \ VMLINUX_SYMBOL(__stop___param) = .; \ + } \ + \ + /* Built-in module versions. */ \ + __modver : AT(ADDR(__modver) - LOAD_OFFSET) { \ + VMLINUX_SYMBOL(__start___modver) = .; \ + *(__modver) \ + VMLINUX_SYMBOL(__stop___modver) = .; \ . = ALIGN((align)); \ VMLINUX_SYMBOL(__end_rodata) = .; \ } \ diff --git a/include/linux/module.h b/include/linux/module.h index 8b17fd8c790d..50efcd3ae850 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -58,6 +58,12 @@ struct module_attribute { void (*free)(struct module *); }; +struct module_version_attribute { + struct module_attribute mattr; + const char *module_name; + const char *version; +}; + struct module_kobject { struct kobject kobj; @@ -161,7 +167,28 @@ extern struct module __this_module; Using this automatically adds a checksum of the .c files and the local headers in "srcversion". */ + +#ifdef MODULE #define MODULE_VERSION(_version) MODULE_INFO(version, _version) +#else +#define MODULE_VERSION(_version) \ + extern ssize_t __modver_version_show(struct module_attribute *, \ + struct module *, char *); \ + static struct module_version_attribute __modver_version_attr \ + __used \ + __attribute__ ((__section__ ("__modver"),aligned(sizeof(void *)))) \ + = { \ + .mattr = { \ + .attr = { \ + .name = "version", \ + .mode = S_IRUGO, \ + }, \ + .show = __modver_version_show, \ + }, \ + .module_name = KBUILD_MODNAME, \ + .version = _version, \ + } +#endif /* Optional firmware file (or files) needed by the module * format is simply firmware file name. Multiple firmware diff --git a/kernel/params.c b/kernel/params.c index 08107d181758..0da1411222b9 100644 --- a/kernel/params.c +++ b/kernel/params.c @@ -719,9 +719,7 @@ void destroy_params(const struct kernel_param *params, unsigned num) params[i].ops->free(params[i].arg); } -static void __init kernel_add_sysfs_param(const char *name, - struct kernel_param *kparam, - unsigned int name_skip) +static struct module_kobject * __init locate_module_kobject(const char *name) { struct module_kobject *mk; struct kobject *kobj; @@ -729,10 +727,7 @@ static void __init kernel_add_sysfs_param(const char *name, kobj = kset_find_obj(module_kset, name); if (kobj) { - /* We already have one. Remove params so we can add more. */ mk = to_module_kobject(kobj); - /* We need to remove it before adding parameters. */ - sysfs_remove_group(&mk->kobj, &mk->mp->grp); } else { mk = kzalloc(sizeof(struct module_kobject), GFP_KERNEL); BUG_ON(!mk); @@ -743,15 +738,36 @@ static void __init kernel_add_sysfs_param(const char *name, "%s", name); if (err) { kobject_put(&mk->kobj); - printk(KERN_ERR "Module '%s' failed add to sysfs, " - "error number %d\n", name, err); - printk(KERN_ERR "The system will be unstable now.\n"); - return; + printk(KERN_ERR + "Module '%s' failed add to sysfs, error number %d\n", + name, err); + printk(KERN_ERR + "The system will be unstable now.\n"); + return NULL; } - /* So that exit path is even. */ + + /* So that we hold reference in both cases. */ kobject_get(&mk->kobj); } + return mk; +} + +static void __init kernel_add_sysfs_param(const char *name, + struct kernel_param *kparam, + unsigned int name_skip) +{ + struct module_kobject *mk; + int err; + + mk = locate_module_kobject(name); + if (!mk) + return; + + /* We need to remove old parameters before adding more. */ + if (mk->mp) + sysfs_remove_group(&mk->kobj, &mk->mp->grp); + /* These should not fail at boot. */ err = add_sysfs_param(mk, kparam, kparam->name + name_skip); BUG_ON(err); @@ -796,6 +812,32 @@ static void __init param_sysfs_builtin(void) } } +ssize_t __modver_version_show(struct module_attribute *mattr, + struct module *mod, char *buf) +{ + struct module_version_attribute *vattr = + container_of(mattr, struct module_version_attribute, mattr); + + return sprintf(buf, "%s\n", vattr->version); +} + +extern struct module_version_attribute __start___modver[], __stop___modver[]; + +static void __init version_sysfs_builtin(void) +{ + const struct module_version_attribute *vattr; + struct module_kobject *mk; + int err; + + for (vattr = __start___modver; vattr < __stop___modver; vattr++) { + mk = locate_module_kobject(vattr->module_name); + if (mk) { + err = sysfs_create_file(&mk->kobj, &vattr->mattr.attr); + kobject_uevent(&mk->kobj, KOBJ_ADD); + kobject_put(&mk->kobj); + } + } +} /* module-related sysfs stuff */ @@ -875,6 +917,7 @@ static int __init param_sysfs_init(void) } module_sysfs_initialized = 1; + version_sysfs_builtin(); param_sysfs_builtin(); return 0; -- cgit v1.2.3 From 3b90a5b292321b2acac3921f77046ae195aef53f Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Mon, 24 Jan 2011 14:32:51 -0600 Subject: module: fix linker error for MODULE_VERSION when !MODULE and CONFIG_SYSFS=n lib/built-in.o:(__modver+0x8): undefined reference to `__modver_version_show' lib/built-in.o:(__modver+0x2c): undefined reference to `__modver_version_show' Simplest to just not emit anything: if they've disabled SYSFS they probably want the smallest kernel possible. Reported-by: Randy Dunlap Signed-off-by: Rusty Russell --- include/linux/module.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/module.h b/include/linux/module.h index 50efcd3ae850..e7c6385c6683 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -168,7 +168,7 @@ extern struct module __this_module; local headers in "srcversion". */ -#ifdef MODULE +#if defined(MODULE) || !defined(CONFIG_SYSFS) #define MODULE_VERSION(_version) MODULE_INFO(version, _version) #else #define MODULE_VERSION(_version) \ -- cgit v1.2.3 From b75be4204e7871869b2c268c00783703197aaa7d Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Wed, 5 Jan 2011 13:27:04 +0100 Subject: param: add null statement to compiled-in module params Add an unused struct declaration statement requiring a terminating semicolon to the compile-in case to provoke an error if __MODULE_INFO() is used without the terminating semicolon. Previously MODULE_ALIAS("foo") (no semicolon) compiled fine if MODULE was not selected. Cc: Dan Carpenter Signed-off-by: Linus Walleij Signed-off-by: Rusty Russell --- include/linux/moduleparam.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/moduleparam.h b/include/linux/moduleparam.h index 112adf8bd47d..07b41951e3fa 100644 --- a/include/linux/moduleparam.h +++ b/include/linux/moduleparam.h @@ -16,15 +16,17 @@ /* Chosen so that structs with an unsigned long line up. */ #define MAX_PARAM_PREFIX_LEN (64 - sizeof(unsigned long)) -#ifdef MODULE #define ___module_cat(a,b) __mod_ ## a ## b #define __module_cat(a,b) ___module_cat(a,b) +#ifdef MODULE #define __MODULE_INFO(tag, name, info) \ static const char __module_cat(name,__LINE__)[] \ __used __attribute__((section(".modinfo"), unused, aligned(1))) \ = __stringify(tag) "=" info #else /* !MODULE */ -#define __MODULE_INFO(tag, name, info) +/* This struct is here for syntactic coherency, it is not used */ +#define __MODULE_INFO(tag, name, info) \ + struct __module_cat(name,__LINE__) {} #endif #define __MODULE_PARM_TYPE(name, _type) \ __MODULE_INFO(parmtype, name##type, #name ":" _type) -- cgit v1.2.3 From 7ef88ad561457c0346355dfd1f53e503ddfde719 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Mon, 24 Jan 2011 14:45:10 -0600 Subject: BUILD_BUG_ON: make it handle more cases BUILD_BUG_ON used to use the optimizer to do code elimination or fail at link time; it was changed to first the size of a negative array (a nicer compile time error), then (in 8c87df457cb58fe75b9b893007917cf8095660a0) to a bitfield. This forced us to change some non-constant cases to MAYBE_BUILD_BUG_ON(); as Jan points out in that commit, it didn't work as intended anyway. bitfields: needs a literal constant at parse time, and can't be put under "if (__builtin_constant_p(x))" for example. negative array: can handle anything, but if the compiler can't tell it's a constant, silently has no effect. link time: breaks link if the compiler can't determine the value, but the linker output is not usually as informative as a compiler error. If we use the negative-array-size method *and* the link time trick, we get the ability to use BUILD_BUG_ON() under __builtin_constant_p() branches, and maximal ability for the compiler to detect errors at build time. We also document it thoroughly. Signed-off-by: Rusty Russell Cc: Jan Beulich Acked-by: Hollis Blanchard --- include/linux/kernel.h | 33 +++++++++++++++++++++++++++------ 1 file changed, 27 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index d07d8057e440..864712f3653d 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -575,12 +575,6 @@ struct sysinfo { char _f[20-2*sizeof(long)-sizeof(int)]; /* Padding: libc5 uses this.. */ }; -/* Force a compilation error if condition is true */ -#define BUILD_BUG_ON(condition) ((void)BUILD_BUG_ON_ZERO(condition)) - -/* Force a compilation error if condition is constant and true */ -#define MAYBE_BUILD_BUG_ON(cond) ((void)sizeof(char[1 - 2 * !!(cond)])) - /* Force a compilation error if a constant expression is not a power of 2 */ #define BUILD_BUG_ON_NOT_POWER_OF_2(n) \ BUILD_BUG_ON((n) == 0 || (((n) & ((n) - 1)) != 0)) @@ -592,6 +586,33 @@ struct sysinfo { #define BUILD_BUG_ON_ZERO(e) (sizeof(struct { int:-!!(e); })) #define BUILD_BUG_ON_NULL(e) ((void *)sizeof(struct { int:-!!(e); })) +/** + * BUILD_BUG_ON - break compile if a condition is true. + * @cond: the condition which the compiler should know is false. + * + * If you have some code which relies on certain constants being equal, or + * other compile-time-evaluated condition, you should use BUILD_BUG_ON to + * detect if someone changes it. + * + * The implementation uses gcc's reluctance to create a negative array, but + * gcc (as of 4.4) only emits that error for obvious cases (eg. not arguments + * to inline functions). So as a fallback we use the optimizer; if it can't + * prove the condition is false, it will cause a link error on the undefined + * "__build_bug_on_failed". This error message can be harder to track down + * though, hence the two different methods. + */ +#ifndef __OPTIMIZE__ +#define BUILD_BUG_ON(condition) ((void)sizeof(char[1 - 2*!!(condition)])) +#else +extern int __build_bug_on_failed; +#define BUILD_BUG_ON(condition) \ + do { \ + ((void)sizeof(char[1 - 2*!!(condition)])); \ + if (condition) __build_bug_on_failed = 1; \ + } while(0) +#endif +#define MAYBE_BUILD_BUG_ON(condition) BUILD_BUG_ON(condition) + /* Trap pasters of __FUNCTION__ at compile-time */ #define __FUNCTION__ (__func__) -- cgit v1.2.3 From 1765e3a4933ea0870fabd755feffc5473c4363ce Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Mon, 24 Jan 2011 14:45:10 -0600 Subject: Remove MAYBE_BUILD_BUG_ON Now BUILD_BUG_ON() can handle optimizable constants, we don't need MAYBE_BUILD_BUG_ON any more. Signed-off-by: Rusty Russell --- include/linux/gfp.h | 2 +- include/linux/kernel.h | 1 - include/linux/kmemcheck.h | 2 +- include/linux/virtio_config.h | 5 ++++- 4 files changed, 6 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/gfp.h b/include/linux/gfp.h index a3b148a91874..0b84c61607e8 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -249,7 +249,7 @@ static inline enum zone_type gfp_zone(gfp_t flags) ((1 << ZONES_SHIFT) - 1); if (__builtin_constant_p(bit)) - MAYBE_BUILD_BUG_ON((GFP_ZONE_BAD >> bit) & 1); + BUILD_BUG_ON((GFP_ZONE_BAD >> bit) & 1); else { #ifdef CONFIG_DEBUG_VM BUG_ON((GFP_ZONE_BAD >> bit) & 1); diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 864712f3653d..e2f4d6af2125 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -611,7 +611,6 @@ extern int __build_bug_on_failed; if (condition) __build_bug_on_failed = 1; \ } while(0) #endif -#define MAYBE_BUILD_BUG_ON(condition) BUILD_BUG_ON(condition) /* Trap pasters of __FUNCTION__ at compile-time */ #define __FUNCTION__ (__func__) diff --git a/include/linux/kmemcheck.h b/include/linux/kmemcheck.h index 08d7dc4ddf40..39f8453239f7 100644 --- a/include/linux/kmemcheck.h +++ b/include/linux/kmemcheck.h @@ -76,7 +76,7 @@ bool kmemcheck_is_obj_initialized(unsigned long addr, size_t size); \ _n = (long) &((ptr)->name##_end) \ - (long) &((ptr)->name##_begin); \ - MAYBE_BUILD_BUG_ON(_n < 0); \ + BUILD_BUG_ON(_n < 0); \ \ kmemcheck_mark_initialized(&((ptr)->name##_begin), _n); \ } while (0) diff --git a/include/linux/virtio_config.h b/include/linux/virtio_config.h index 0093dd7c1d6f..800617b4ddd5 100644 --- a/include/linux/virtio_config.h +++ b/include/linux/virtio_config.h @@ -109,7 +109,10 @@ static inline bool virtio_has_feature(const struct virtio_device *vdev, unsigned int fbit) { /* Did you forget to fix assumptions on max features? */ - MAYBE_BUILD_BUG_ON(fbit >= 32); + if (__builtin_constant_p(fbit)) + BUILD_BUG_ON(fbit >= 32); + else + BUG_ON(fbit >= 32); if (fbit < VIRTIO_TRANSPORT_F_START) virtio_check_driver_offered_feature(vdev, fbit); -- cgit v1.2.3 From 8c6a98b22b750c9eb52653ba643faa17db8d3881 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Mon, 24 Jan 2011 09:31:38 -0800 Subject: Input: sysrq - ensure sysrq_enabled and __sysrq_enabled are consistent Currently sysrq_enabled and __sysrq_enabled are initialised separately and inconsistently, leading to sysrq being actually enabled by reported as not enabled in sysfs. The first change to the sysfs configurable synchronises these two: static int __read_mostly sysrq_enabled = 1; static int __sysrq_enabled; Add a common define to carry the default for these preventing them becoming out of sync again. Default this to 1 to mirror previous behaviour. Signed-off-by: Andy Whitcroft Cc: stable@kernel.org Signed-off-by: Dmitry Torokhov --- drivers/char/sysrq.c | 2 +- include/linux/sysrq.h | 3 +++ kernel/sysctl.c | 3 ++- 3 files changed, 6 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/char/sysrq.c b/drivers/char/sysrq.c index c556ed9db13d..8e0dd254eb11 100644 --- a/drivers/char/sysrq.c +++ b/drivers/char/sysrq.c @@ -46,7 +46,7 @@ #include /* Whether we react on sysrq keys or just ignore them */ -static int __read_mostly sysrq_enabled = 1; +static int __read_mostly sysrq_enabled = SYSRQ_DEFAULT_ENABLE; static bool __read_mostly sysrq_always_enabled; static bool sysrq_on(void) diff --git a/include/linux/sysrq.h b/include/linux/sysrq.h index 387fa7d05c98..7faf933cced7 100644 --- a/include/linux/sysrq.h +++ b/include/linux/sysrq.h @@ -17,6 +17,9 @@ #include #include +/* Enable/disable SYSRQ support by default (0==no, 1==yes). */ +#define SYSRQ_DEFAULT_ENABLE 1 + /* Possible values of bitmask for enabling sysrq functions */ /* 0x0001 is reserved for enable everything */ #define SYSRQ_ENABLE_LOG 0x0002 diff --git a/kernel/sysctl.c b/kernel/sysctl.c index c33a1edb799f..3afce4dc9ba5 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -169,7 +169,8 @@ static int proc_taint(struct ctl_table *table, int write, #endif #ifdef CONFIG_MAGIC_SYSRQ -static int __sysrq_enabled; /* Note: sysrq code ises it's own private copy */ +/* Note: sysrq code uses it's own private copy */ +static int __sysrq_enabled = SYSRQ_DEFAULT_ENABLE; static int sysrq_sysctl_handler(ctl_table *table, int write, void __user *buffer, size_t *lenp, -- cgit v1.2.3 From 731f3f482ad3b2c58a1af2d0a9a634a82803706a Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Fri, 21 Jan 2011 03:05:28 +0000 Subject: NFS: nfsacl_{encode,decode} should return signed integer Clean up. The nfsacl_encode() and nfsacl_decode() functions return negative errno values, and each call site verifies that the returned value is not negative. Change the synopsis of both of these functions to reflect this usage. Document the synopsis and return values. Reported-by: Trond Myklebust Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- fs/nfs_common/nfsacl.c | 32 ++++++++++++++++++++++++++------ include/linux/nfsacl.h | 4 ++-- 2 files changed, 28 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs_common/nfsacl.c b/fs/nfs_common/nfsacl.c index fc1c52571c03..a3e78bd18679 100644 --- a/fs/nfs_common/nfsacl.c +++ b/fs/nfs_common/nfsacl.c @@ -72,9 +72,20 @@ xdr_nfsace_encode(struct xdr_array2_desc *desc, void *elem) return 0; } -unsigned int -nfsacl_encode(struct xdr_buf *buf, unsigned int base, struct inode *inode, - struct posix_acl *acl, int encode_entries, int typeflag) +/** + * nfsacl_encode - Encode an NFSv3 ACL + * + * @buf: destination xdr_buf to contain XDR encoded ACL + * @base: byte offset in xdr_buf where XDR'd ACL begins + * @inode: inode of file whose ACL this is + * @acl: posix_acl to encode + * @encode_entries: whether to encode ACEs as well + * @typeflag: ACL type: NFS_ACL_DEFAULT or zero + * + * Returns size of encoded ACL in bytes or a negative errno value. + */ +int nfsacl_encode(struct xdr_buf *buf, unsigned int base, struct inode *inode, + struct posix_acl *acl, int encode_entries, int typeflag) { int entries = (acl && acl->a_count) ? max_t(int, acl->a_count, 4) : 0; struct nfsacl_encode_desc nfsacl_desc = { @@ -224,9 +235,18 @@ posix_acl_from_nfsacl(struct posix_acl *acl) return 0; } -unsigned int -nfsacl_decode(struct xdr_buf *buf, unsigned int base, unsigned int *aclcnt, - struct posix_acl **pacl) +/** + * nfsacl_decode - Decode an NFSv3 ACL + * + * @buf: xdr_buf containing XDR'd ACL data to decode + * @base: byte offset in xdr_buf where XDR'd ACL begins + * @aclcnt: count of ACEs in decoded posix_acl + * @pacl: buffer in which to place decoded posix_acl + * + * Returns the length of the decoded ACL in bytes, or a negative errno value. + */ +int nfsacl_decode(struct xdr_buf *buf, unsigned int base, unsigned int *aclcnt, + struct posix_acl **pacl) { struct nfsacl_decode_desc nfsacl_desc = { .desc = { diff --git a/include/linux/nfsacl.h b/include/linux/nfsacl.h index f321b578edeb..fabcb1e5c460 100644 --- a/include/linux/nfsacl.h +++ b/include/linux/nfsacl.h @@ -51,10 +51,10 @@ nfsacl_size(struct posix_acl *acl_access, struct posix_acl *acl_default) return w; } -extern unsigned int +extern int nfsacl_encode(struct xdr_buf *buf, unsigned int base, struct inode *inode, struct posix_acl *acl, int encode_entries, int typeflag); -extern unsigned int +extern int nfsacl_decode(struct xdr_buf *buf, unsigned int base, unsigned int *aclcnt, struct posix_acl **pacl); -- cgit v1.2.3 From f61f6da0d53842e849bab7f69e1431bd3de1136d Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Fri, 21 Jan 2011 03:05:38 +0000 Subject: NFS: Prevent memory allocation failure in nfsacl_encode() nfsacl_encode() allocates memory in certain cases. This of course is not guaranteed to work. Since commit 9f06c719 "SUNRPC: New xdr_streams XDR encoder API", the kernel's XDR encoders can't return a result indicating possibly a failure, so a memory allocation failure in nfsacl_encode() has become fatal (ie, the XDR code Oopses) in some cases. However, the allocated memory is a tiny fixed amount, on the order of 40-50 bytes. We can easily use a stack-allocated buffer for this, with only a wee bit of nose-holding. Signed-off-by: Chuck Lever Signed-off-by: Trond Myklebust --- fs/nfs/nfs3acl.c | 4 ++-- fs/nfs_common/nfsacl.c | 22 +++++++++++++++------- fs/posix_acl.c | 17 +++++++++++++---- include/linux/posix_acl.h | 1 + 4 files changed, 31 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c index 9f88c5f4c7e2..274342771655 100644 --- a/fs/nfs/nfs3acl.c +++ b/fs/nfs/nfs3acl.c @@ -311,8 +311,8 @@ static int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, if (!nfs_server_capable(inode, NFS_CAP_ACLS)) goto out; - /* We are doing this here, because XDR marshalling can only - return -ENOMEM. */ + /* We are doing this here because XDR marshalling does not + * return any results, it BUGs. */ status = -ENOSPC; if (acl != NULL && acl->a_count > NFS_ACL_MAX_ENTRIES) goto out; diff --git a/fs/nfs_common/nfsacl.c b/fs/nfs_common/nfsacl.c index a3e78bd18679..84c27d69d421 100644 --- a/fs/nfs_common/nfsacl.c +++ b/fs/nfs_common/nfsacl.c @@ -42,6 +42,11 @@ struct nfsacl_encode_desc { gid_t gid; }; +struct nfsacl_simple_acl { + struct posix_acl acl; + struct posix_acl_entry ace[4]; +}; + static int xdr_nfsace_encode(struct xdr_array2_desc *desc, void *elem) { @@ -99,17 +104,22 @@ int nfsacl_encode(struct xdr_buf *buf, unsigned int base, struct inode *inode, .uid = inode->i_uid, .gid = inode->i_gid, }; + struct nfsacl_simple_acl aclbuf; int err; - struct posix_acl *acl2 = NULL; if (entries > NFS_ACL_MAX_ENTRIES || xdr_encode_word(buf, base, entries)) return -EINVAL; if (encode_entries && acl && acl->a_count == 3) { - /* Fake up an ACL_MASK entry. */ - acl2 = posix_acl_alloc(4, GFP_KERNEL); - if (!acl2) - return -ENOMEM; + struct posix_acl *acl2 = &aclbuf.acl; + + /* Avoid the use of posix_acl_alloc(). nfsacl_encode() is + * invoked in contexts where a memory allocation failure is + * fatal. Fortunately this fake ACL is small enough to + * construct on the stack. */ + memset(acl2, 0, sizeof(acl2)); + posix_acl_init(acl2, 4); + /* Insert entries in canonical order: other orders seem to confuse Solaris VxFS. */ acl2->a_entries[0] = acl->a_entries[0]; /* ACL_USER_OBJ */ @@ -120,8 +130,6 @@ int nfsacl_encode(struct xdr_buf *buf, unsigned int base, struct inode *inode, nfsacl_desc.acl = acl2; } err = xdr_encode_array2(buf, base + 4, &nfsacl_desc.desc); - if (acl2) - posix_acl_release(acl2); if (!err) err = 8 + nfsacl_desc.desc.elem_size * nfsacl_desc.desc.array_len; diff --git a/fs/posix_acl.c b/fs/posix_acl.c index 39df95a0ec25..b1cf6bf4b41d 100644 --- a/fs/posix_acl.c +++ b/fs/posix_acl.c @@ -22,6 +22,7 @@ #include +EXPORT_SYMBOL(posix_acl_init); EXPORT_SYMBOL(posix_acl_alloc); EXPORT_SYMBOL(posix_acl_clone); EXPORT_SYMBOL(posix_acl_valid); @@ -31,6 +32,16 @@ EXPORT_SYMBOL(posix_acl_create_masq); EXPORT_SYMBOL(posix_acl_chmod_masq); EXPORT_SYMBOL(posix_acl_permission); +/* + * Init a fresh posix_acl + */ +void +posix_acl_init(struct posix_acl *acl, int count) +{ + atomic_set(&acl->a_refcount, 1); + acl->a_count = count; +} + /* * Allocate a new ACL with the specified number of entries. */ @@ -40,10 +51,8 @@ posix_acl_alloc(int count, gfp_t flags) const size_t size = sizeof(struct posix_acl) + count * sizeof(struct posix_acl_entry); struct posix_acl *acl = kmalloc(size, flags); - if (acl) { - atomic_set(&acl->a_refcount, 1); - acl->a_count = count; - } + if (acl) + posix_acl_init(acl, count); return acl; } diff --git a/include/linux/posix_acl.h b/include/linux/posix_acl.h index d68283a898bb..54211c1cd926 100644 --- a/include/linux/posix_acl.h +++ b/include/linux/posix_acl.h @@ -71,6 +71,7 @@ posix_acl_release(struct posix_acl *acl) /* posix_acl.c */ +extern void posix_acl_init(struct posix_acl *, int); extern struct posix_acl *posix_acl_alloc(int, gfp_t); extern struct posix_acl *posix_acl_clone(const struct posix_acl *, gfp_t); extern int posix_acl_valid(const struct posix_acl *); -- cgit v1.2.3 From 778be232a207e79088ba70d832ac25dfea6fbf1a Mon Sep 17 00:00:00 2001 From: Andy Adamson Date: Tue, 25 Jan 2011 15:38:01 +0000 Subject: NFS do not find client in NFSv4 pg_authenticate The information required to find the nfs_client cooresponding to the incoming back channel request is contained in the NFS layer. Perform minimal checking in the RPC layer pg_authenticate method, and push more detailed checking into the NFS layer where the nfs_client can be found. Signed-off-by: Andy Adamson Signed-off-by: Trond Myklebust --- fs/nfs/callback.c | 109 +++++++++++----------------------------- fs/nfs/callback.h | 4 +- fs/nfs/callback_proc.c | 10 +--- fs/nfs/callback_xdr.c | 5 +- fs/nfs/client.c | 15 ++---- fs/nfs/internal.h | 3 +- fs/nfs/nfs4state.c | 6 --- include/linux/sunrpc/bc_xprt.h | 13 ----- include/linux/sunrpc/svc_xprt.h | 1 - net/sunrpc/svcsock.c | 4 +- 10 files changed, 42 insertions(+), 128 deletions(-) (limited to 'include/linux') diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c index 199016528fcb..e3d294269058 100644 --- a/fs/nfs/callback.c +++ b/fs/nfs/callback.c @@ -134,33 +134,6 @@ out_err: } #if defined(CONFIG_NFS_V4_1) -/* - * * CB_SEQUENCE operations will fail until the callback sessionid is set. - * */ -int nfs4_set_callback_sessionid(struct nfs_client *clp) -{ - struct svc_serv *serv = clp->cl_rpcclient->cl_xprt->bc_serv; - struct nfs4_sessionid *bc_sid; - - if (!serv->sv_bc_xprt) - return -EINVAL; - - /* on success freed in xprt_free */ - bc_sid = kmalloc(sizeof(struct nfs4_sessionid), GFP_KERNEL); - if (!bc_sid) - return -ENOMEM; - memcpy(bc_sid->data, &clp->cl_session->sess_id.data, - NFS4_MAX_SESSIONID_LEN); - spin_lock_bh(&serv->sv_cb_lock); - serv->sv_bc_xprt->xpt_bc_sid = bc_sid; - spin_unlock_bh(&serv->sv_cb_lock); - dprintk("%s set xpt_bc_sid=%u:%u:%u:%u for sv_bc_xprt %p\n", __func__, - ((u32 *)bc_sid->data)[0], ((u32 *)bc_sid->data)[1], - ((u32 *)bc_sid->data)[2], ((u32 *)bc_sid->data)[3], - serv->sv_bc_xprt); - return 0; -} - /* * The callback service for NFSv4.1 callbacks */ @@ -266,10 +239,6 @@ static inline void nfs_callback_bc_serv(u32 minorversion, struct rpc_xprt *xprt, struct nfs_callback_data *cb_info) { } -int nfs4_set_callback_sessionid(struct nfs_client *clp) -{ - return 0; -} #endif /* CONFIG_NFS_V4_1 */ /* @@ -359,78 +328,58 @@ void nfs_callback_down(int minorversion) mutex_unlock(&nfs_callback_mutex); } -static int check_gss_callback_principal(struct nfs_client *clp, - struct svc_rqst *rqstp) +/* Boolean check of RPC_AUTH_GSS principal */ +int +check_gss_callback_principal(struct nfs_client *clp, struct svc_rqst *rqstp) { struct rpc_clnt *r = clp->cl_rpcclient; char *p = svc_gss_principal(rqstp); + if (rqstp->rq_authop->flavour != RPC_AUTH_GSS) + return 1; + /* No RPC_AUTH_GSS on NFSv4.1 back channel yet */ if (clp->cl_minorversion != 0) - return SVC_DROP; + return 0; /* * It might just be a normal user principal, in which case * userspace won't bother to tell us the name at all. */ if (p == NULL) - return SVC_DENIED; + return 0; /* Expect a GSS_C_NT_HOSTBASED_NAME like "nfs@serverhostname" */ if (memcmp(p, "nfs@", 4) != 0) - return SVC_DENIED; + return 0; p += 4; if (strcmp(p, r->cl_server) != 0) - return SVC_DENIED; - return SVC_OK; + return 0; + return 1; } -/* pg_authenticate method helper */ -static struct nfs_client *nfs_cb_find_client(struct svc_rqst *rqstp) -{ - struct nfs4_sessionid *sessionid = bc_xprt_sid(rqstp); - int is_cb_compound = rqstp->rq_proc == CB_COMPOUND ? 1 : 0; - - dprintk("--> %s rq_proc %d\n", __func__, rqstp->rq_proc); - if (svc_is_backchannel(rqstp)) - /* Sessionid (usually) set after CB_NULL ping */ - return nfs4_find_client_sessionid(svc_addr(rqstp), sessionid, - is_cb_compound); - else - /* No callback identifier in pg_authenticate */ - return nfs4_find_client_no_ident(svc_addr(rqstp)); -} - -/* pg_authenticate method for nfsv4 callback threads. */ +/* + * pg_authenticate method for nfsv4 callback threads. + * + * The authflavor has been negotiated, so an incorrect flavor is a server + * bug. Drop packets with incorrect authflavor. + * + * All other checking done after NFS decoding where the nfs_client can be + * found in nfs4_callback_compound + */ static int nfs_callback_authenticate(struct svc_rqst *rqstp) { - struct nfs_client *clp; - RPC_IFDEBUG(char buf[RPC_MAX_ADDRBUFLEN]); - int ret = SVC_OK; - - /* Don't talk to strangers */ - clp = nfs_cb_find_client(rqstp); - if (clp == NULL) - return SVC_DROP; - - dprintk("%s: %s NFSv4 callback!\n", __func__, - svc_print_addr(rqstp, buf, sizeof(buf))); - switch (rqstp->rq_authop->flavour) { - case RPC_AUTH_NULL: - if (rqstp->rq_proc != CB_NULL) - ret = SVC_DENIED; - break; - case RPC_AUTH_UNIX: - break; - case RPC_AUTH_GSS: - ret = check_gss_callback_principal(clp, rqstp); - break; - default: - ret = SVC_DENIED; + case RPC_AUTH_NULL: + if (rqstp->rq_proc != CB_NULL) + return SVC_DROP; + break; + case RPC_AUTH_GSS: + /* No RPC_AUTH_GSS support yet in NFSv4.1 */ + if (svc_is_backchannel(rqstp)) + return SVC_DROP; } - nfs_put_client(clp); - return ret; + return SVC_OK; } /* diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h index d3b44f9bd747..46d93ce7311b 100644 --- a/fs/nfs/callback.h +++ b/fs/nfs/callback.h @@ -7,6 +7,7 @@ */ #ifndef __LINUX_FS_NFS_CALLBACK_H #define __LINUX_FS_NFS_CALLBACK_H +#include #define NFS4_CALLBACK 0x40000000 #define NFS4_CALLBACK_XDRSIZE 2048 @@ -37,7 +38,6 @@ enum nfs4_callback_opnum { struct cb_process_state { __be32 drc_status; struct nfs_client *clp; - struct nfs4_sessionid *svc_sid; /* v4.1 callback service sessionid */ }; struct cb_compound_hdr_arg { @@ -168,7 +168,7 @@ extern unsigned nfs4_callback_layoutrecall( extern void nfs4_check_drain_bc_complete(struct nfs4_session *ses); extern void nfs4_cb_take_slot(struct nfs_client *clp); #endif /* CONFIG_NFS_V4_1 */ - +extern int check_gss_callback_principal(struct nfs_client *, struct svc_rqst *); extern __be32 nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres *res, struct cb_process_state *cps); diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c index 4bb91cb2620d..829f406e91dd 100644 --- a/fs/nfs/callback_proc.c +++ b/fs/nfs/callback_proc.c @@ -373,17 +373,11 @@ __be32 nfs4_callback_sequence(struct cb_sequenceargs *args, { struct nfs_client *clp; int i; - __be32 status; + __be32 status = htonl(NFS4ERR_BADSESSION); cps->clp = NULL; - status = htonl(NFS4ERR_BADSESSION); - /* Incoming session must match the callback session */ - if (memcmp(&args->csa_sessionid, cps->svc_sid, NFS4_MAX_SESSIONID_LEN)) - goto out; - - clp = nfs4_find_client_sessionid(args->csa_addr, - &args->csa_sessionid, 1); + clp = nfs4_find_client_sessionid(args->csa_addr, &args->csa_sessionid); if (clp == NULL) goto out; diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c index 23112c263f81..14e0f9371d14 100644 --- a/fs/nfs/callback_xdr.c +++ b/fs/nfs/callback_xdr.c @@ -794,10 +794,9 @@ static __be32 nfs4_callback_compound(struct svc_rqst *rqstp, void *argp, void *r if (hdr_arg.minorversion == 0) { cps.clp = nfs4_find_client_ident(hdr_arg.cb_ident); - if (!cps.clp) + if (!cps.clp || !check_gss_callback_principal(cps.clp, rqstp)) return rpc_drop_reply; - } else - cps.svc_sid = bc_xprt_sid(rqstp); + } hdr_res.taglen = hdr_arg.taglen; hdr_res.tag = hdr_arg.tag; diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 192f2f860265..bd3ca32879e7 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -1206,16 +1206,11 @@ nfs4_find_client_ident(int cb_ident) * For CB_COMPOUND calls, find a client by IP address, protocol version, * minorversion, and sessionID * - * CREATE_SESSION triggers a CB_NULL ping from servers. The callback service - * sessionid can only be set after the CREATE_SESSION return, so a CB_NULL - * can arrive before the callback sessionid is set. For CB_NULL calls, - * find a client by IP address protocol version, and minorversion. - * * Returns NULL if no such client */ struct nfs_client * nfs4_find_client_sessionid(const struct sockaddr *addr, - struct nfs4_sessionid *sid, int is_cb_compound) + struct nfs4_sessionid *sid) { struct nfs_client *clp; @@ -1227,9 +1222,9 @@ nfs4_find_client_sessionid(const struct sockaddr *addr, if (!nfs4_has_session(clp)) continue; - /* Match sessionid unless cb_null call*/ - if (is_cb_compound && (memcmp(clp->cl_session->sess_id.data, - sid->data, NFS4_MAX_SESSIONID_LEN) != 0)) + /* Match sessionid*/ + if (memcmp(clp->cl_session->sess_id.data, + sid->data, NFS4_MAX_SESSIONID_LEN) != 0) continue; atomic_inc(&clp->cl_count); @@ -1244,7 +1239,7 @@ nfs4_find_client_sessionid(const struct sockaddr *addr, struct nfs_client * nfs4_find_client_sessionid(const struct sockaddr *addr, - struct nfs4_sessionid *sid, int is_cb_compound) + struct nfs4_sessionid *sid) { return NULL; } diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index 4644f04b4b46..cf9fdbdabc67 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -133,8 +133,7 @@ extern void nfs_put_client(struct nfs_client *); extern struct nfs_client *nfs4_find_client_no_ident(const struct sockaddr *); extern struct nfs_client *nfs4_find_client_ident(int); extern struct nfs_client * -nfs4_find_client_sessionid(const struct sockaddr *, struct nfs4_sessionid *, - int); +nfs4_find_client_sessionid(const struct sockaddr *, struct nfs4_sessionid *); extern struct nfs_server *nfs_create_server( const struct nfs_parsed_mount_data *, struct nfs_fh *); diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 2336d532cf66..e6742b57a04c 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -232,12 +232,6 @@ int nfs41_init_clientid(struct nfs_client *clp, struct rpc_cred *cred) status = nfs4_proc_create_session(clp); if (status != 0) goto out; - status = nfs4_set_callback_sessionid(clp); - if (status != 0) { - printk(KERN_WARNING "Sessionid not set. No callback service\n"); - nfs_callback_down(1); - status = 0; - } nfs41_setup_state_renewal(clp); nfs_mark_client_ready(clp, NFS_CS_READY); out: diff --git a/include/linux/sunrpc/bc_xprt.h b/include/linux/sunrpc/bc_xprt.h index c50b458b8a3f..082884295f80 100644 --- a/include/linux/sunrpc/bc_xprt.h +++ b/include/linux/sunrpc/bc_xprt.h @@ -47,14 +47,6 @@ static inline int svc_is_backchannel(const struct svc_rqst *rqstp) return 1; return 0; } -static inline struct nfs4_sessionid *bc_xprt_sid(struct svc_rqst *rqstp) -{ - if (svc_is_backchannel(rqstp)) - return (struct nfs4_sessionid *) - rqstp->rq_server->sv_bc_xprt->xpt_bc_sid; - return NULL; -} - #else /* CONFIG_NFS_V4_1 */ static inline int xprt_setup_backchannel(struct rpc_xprt *xprt, unsigned int min_reqs) @@ -67,11 +59,6 @@ static inline int svc_is_backchannel(const struct svc_rqst *rqstp) return 0; } -static inline struct nfs4_sessionid *bc_xprt_sid(struct svc_rqst *rqstp) -{ - return NULL; -} - static inline void xprt_free_bc_request(struct rpc_rqst *req) { } diff --git a/include/linux/sunrpc/svc_xprt.h b/include/linux/sunrpc/svc_xprt.h index 059877b4d85b..7ad9751a0d87 100644 --- a/include/linux/sunrpc/svc_xprt.h +++ b/include/linux/sunrpc/svc_xprt.h @@ -77,7 +77,6 @@ struct svc_xprt { size_t xpt_remotelen; /* length of address */ struct rpc_wait_queue xpt_bc_pending; /* backchannel wait queue */ struct list_head xpt_users; /* callbacks on free */ - void *xpt_bc_sid; /* back channel session ID */ struct net *xpt_net; struct rpc_xprt *xpt_bc_xprt; /* NFSv4.1 backchannel */ diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 7bd3bbba4710..d802e941d365 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -1609,9 +1609,7 @@ static struct svc_xprt *svc_bc_create_socket(struct svc_serv *serv, */ static void svc_bc_sock_free(struct svc_xprt *xprt) { - if (xprt) { - kfree(xprt->xpt_bc_sid); + if (xprt) kfree(container_of(xprt, struct svc_sock, sk_xprt)); - } } #endif /* CONFIG_NFS_V4_1 */ -- cgit v1.2.3 From ac751efa6a0d70f2c9daef5c7e3a92270f5c2dff Mon Sep 17 00:00:00 2001 From: Torben Hohn Date: Tue, 25 Jan 2011 15:07:35 -0800 Subject: console: rename acquire/release_console_sem() to console_lock/unlock() The -rt patches change the console_semaphore to console_mutex. As a result, a quite large chunk of the patches changes all acquire/release_console_sem() to acquire/release_console_mutex() This commit makes things use more neutral function names which dont make implications about the underlying lock. The only real change is the return value of console_trylock which is inverted from try_acquire_console_sem() This patch also paves the way to switching console_sem from a semaphore to a mutex. [akpm@linux-foundation.org: coding-style fixes] [akpm@linux-foundation.org: make console_trylock return 1 on success, per Geert] Signed-off-by: Torben Hohn Cc: Thomas Gleixner Cc: Greg KH Cc: Ingo Molnar Cc: Geert Uytterhoeven Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/mach-omap2/pm24xx.c | 4 +- arch/arm/mach-omap2/pm34xx.c | 4 +- arch/arm/mach-omap2/serial.c | 4 +- arch/parisc/kernel/pdc_cons.c | 4 +- drivers/char/bfin_jtag_comm.c | 8 +-- drivers/gpu/drm/nouveau/nouveau_drv.c | 8 +-- drivers/gpu/drm/radeon/radeon_device.c | 10 +-- drivers/staging/msm/msm_fb.c | 8 +-- drivers/staging/olpc_dcon/olpc_dcon.c | 10 +-- drivers/staging/sm7xx/smtcfb.c | 8 +-- drivers/tty/serial/sb1250-duart.c | 2 +- drivers/tty/tty_io.c | 4 +- drivers/tty/vt/selection.c | 4 +- drivers/tty/vt/vc_screen.c | 16 ++--- drivers/tty/vt/vt.c | 124 ++++++++++++++++----------------- drivers/tty/vt/vt_ioctl.c | 60 ++++++++-------- drivers/video/arkfb.c | 12 ++-- drivers/video/aty/aty128fb.c | 12 ++-- drivers/video/aty/atyfb_base.c | 10 +-- drivers/video/aty/radeon_pm.c | 10 +-- drivers/video/chipsfb.c | 8 +-- drivers/video/console/fbcon.c | 42 +++++------ drivers/video/da8xx-fb.c | 8 +-- drivers/video/fbmem.c | 12 ++-- drivers/video/fbsysfs.c | 20 +++--- drivers/video/geode/gxfb_core.c | 8 +-- drivers/video/geode/lxfb_core.c | 8 +-- drivers/video/i810/i810_main.c | 8 +-- drivers/video/jz4740_fb.c | 8 +-- drivers/video/mx3fb.c | 8 +-- drivers/video/nvidia/nvidia.c | 8 +-- drivers/video/ps3fb.c | 16 ++--- drivers/video/s3fb.c | 16 ++--- drivers/video/savage/savagefb_driver.c | 8 +-- drivers/video/sh_mobile_hdmi.c | 8 +-- drivers/video/sh_mobile_lcdcfb.c | 4 +- drivers/video/sm501fb.c | 8 +-- drivers/video/tmiofb.c | 10 +-- drivers/video/via/viafbdev.c | 8 +-- drivers/video/vt8623fb.c | 12 ++-- drivers/video/xen-fbfront.c | 4 +- fs/proc/consoles.c | 4 +- include/linux/console.h | 6 +- kernel/printk.c | 100 ++++++++++++++------------ 44 files changed, 336 insertions(+), 328 deletions(-) (limited to 'include/linux') diff --git a/arch/arm/mach-omap2/pm24xx.c b/arch/arm/mach-omap2/pm24xx.c index 9e5dc8ed51e9..97feb3ab6a69 100644 --- a/arch/arm/mach-omap2/pm24xx.c +++ b/arch/arm/mach-omap2/pm24xx.c @@ -134,7 +134,7 @@ static void omap2_enter_full_retention(void) /* Block console output in case it is on one of the OMAP UARTs */ if (!is_suspending()) - if (try_acquire_console_sem()) + if (!console_trylock()) goto no_sleep; omap_uart_prepare_idle(0); @@ -151,7 +151,7 @@ static void omap2_enter_full_retention(void) omap_uart_resume_idle(0); if (!is_suspending()) - release_console_sem(); + console_unlock(); no_sleep: if (omap2_pm_debug) { diff --git a/arch/arm/mach-omap2/pm34xx.c b/arch/arm/mach-omap2/pm34xx.c index 8cbbeade4b8a..a4aa1920a75c 100644 --- a/arch/arm/mach-omap2/pm34xx.c +++ b/arch/arm/mach-omap2/pm34xx.c @@ -398,7 +398,7 @@ void omap_sram_idle(void) if (!is_suspending()) if (per_next_state < PWRDM_POWER_ON || core_next_state < PWRDM_POWER_ON) - if (try_acquire_console_sem()) + if (!console_trylock()) goto console_still_active; /* PER */ @@ -481,7 +481,7 @@ void omap_sram_idle(void) } if (!is_suspending()) - release_console_sem(); + console_unlock(); console_still_active: /* Disable IO-PAD and IO-CHAIN wakeup */ diff --git a/arch/arm/mach-omap2/serial.c b/arch/arm/mach-omap2/serial.c index 302da7403a10..32e91a9c8b6b 100644 --- a/arch/arm/mach-omap2/serial.c +++ b/arch/arm/mach-omap2/serial.c @@ -812,7 +812,7 @@ void __init omap_serial_init_port(struct omap_board_data *bdata) oh->dev_attr = uart; - acquire_console_sem(); /* in case the earlycon is on the UART */ + console_lock(); /* in case the earlycon is on the UART */ /* * Because of early UART probing, UART did not get idled @@ -838,7 +838,7 @@ void __init omap_serial_init_port(struct omap_board_data *bdata) omap_uart_block_sleep(uart); uart->timeout = DEFAULT_TIMEOUT; - release_console_sem(); + console_unlock(); if ((cpu_is_omap34xx() && uart->padconf) || (uart->wk_en && uart->wk_mask)) { diff --git a/arch/parisc/kernel/pdc_cons.c b/arch/parisc/kernel/pdc_cons.c index 11bdd68e5762..fc770be465ff 100644 --- a/arch/parisc/kernel/pdc_cons.c +++ b/arch/parisc/kernel/pdc_cons.c @@ -169,11 +169,11 @@ static int __init pdc_console_tty_driver_init(void) struct console *tmp; - acquire_console_sem(); + console_lock(); for_each_console(tmp) if (tmp == &pdc_cons) break; - release_console_sem(); + console_unlock(); if (!tmp) { printk(KERN_INFO "PDC console driver not registered anymore, not creating %s\n", pdc_cons.name); diff --git a/drivers/char/bfin_jtag_comm.c b/drivers/char/bfin_jtag_comm.c index e397df3ad98e..16402445f2b2 100644 --- a/drivers/char/bfin_jtag_comm.c +++ b/drivers/char/bfin_jtag_comm.c @@ -183,16 +183,16 @@ bfin_jc_circ_write(const unsigned char *buf, int count) } #ifndef CONFIG_BFIN_JTAG_COMM_CONSOLE -# define acquire_console_sem() -# define release_console_sem() +# define console_lock() +# define console_unlock() #endif static int bfin_jc_write(struct tty_struct *tty, const unsigned char *buf, int count) { int i; - acquire_console_sem(); + console_lock(); i = bfin_jc_circ_write(buf, count); - release_console_sem(); + console_unlock(); wake_up_process(bfin_jc_kthread); return i; } diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.c b/drivers/gpu/drm/nouveau/nouveau_drv.c index 13bb672a16f4..f658a04eecf9 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drv.c +++ b/drivers/gpu/drm/nouveau/nouveau_drv.c @@ -234,9 +234,9 @@ nouveau_pci_suspend(struct pci_dev *pdev, pm_message_t pm_state) pci_set_power_state(pdev, PCI_D3hot); } - acquire_console_sem(); + console_lock(); nouveau_fbcon_set_suspend(dev, 1); - release_console_sem(); + console_unlock(); nouveau_fbcon_restore_accel(dev); return 0; @@ -359,9 +359,9 @@ nouveau_pci_resume(struct pci_dev *pdev) nv_crtc->lut.depth = 0; } - acquire_console_sem(); + console_lock(); nouveau_fbcon_set_suspend(dev, 0); - release_console_sem(); + console_unlock(); nouveau_fbcon_zfill_all(dev); diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c index 26091d602b84..0d478932b1a9 100644 --- a/drivers/gpu/drm/radeon/radeon_device.c +++ b/drivers/gpu/drm/radeon/radeon_device.c @@ -891,9 +891,9 @@ int radeon_suspend_kms(struct drm_device *dev, pm_message_t state) pci_disable_device(dev->pdev); pci_set_power_state(dev->pdev, PCI_D3hot); } - acquire_console_sem(); + console_lock(); radeon_fbdev_set_suspend(rdev, 1); - release_console_sem(); + console_unlock(); return 0; } @@ -905,11 +905,11 @@ int radeon_resume_kms(struct drm_device *dev) if (dev->switch_power_state == DRM_SWITCH_POWER_OFF) return 0; - acquire_console_sem(); + console_lock(); pci_set_power_state(dev->pdev, PCI_D0); pci_restore_state(dev->pdev); if (pci_enable_device(dev->pdev)) { - release_console_sem(); + console_unlock(); return -1; } pci_set_master(dev->pdev); @@ -920,7 +920,7 @@ int radeon_resume_kms(struct drm_device *dev) radeon_restore_bios_scratch_regs(rdev); radeon_fbdev_set_suspend(rdev, 0); - release_console_sem(); + console_unlock(); /* reset hpd state */ radeon_hpd_init(rdev); diff --git a/drivers/staging/msm/msm_fb.c b/drivers/staging/msm/msm_fb.c index 23fa049b51f2..a2f29d464051 100644 --- a/drivers/staging/msm/msm_fb.c +++ b/drivers/staging/msm/msm_fb.c @@ -347,7 +347,7 @@ static int msm_fb_suspend(struct platform_device *pdev, pm_message_t state) if ((!mfd) || (mfd->key != MFD_KEY)) return 0; - acquire_console_sem(); + console_lock(); fb_set_suspend(mfd->fbi, 1); ret = msm_fb_suspend_sub(mfd); @@ -358,7 +358,7 @@ static int msm_fb_suspend(struct platform_device *pdev, pm_message_t state) pdev->dev.power.power_state = state; } - release_console_sem(); + console_unlock(); return ret; } #else @@ -431,11 +431,11 @@ static int msm_fb_resume(struct platform_device *pdev) if ((!mfd) || (mfd->key != MFD_KEY)) return 0; - acquire_console_sem(); + console_lock(); ret = msm_fb_resume_sub(mfd); pdev->dev.power.power_state = PMSG_ON; fb_set_suspend(mfd->fbi, 1); - release_console_sem(); + console_unlock(); return ret; } diff --git a/drivers/staging/olpc_dcon/olpc_dcon.c b/drivers/staging/olpc_dcon/olpc_dcon.c index 9f26dc9408bb..56a283d1a74d 100644 --- a/drivers/staging/olpc_dcon/olpc_dcon.c +++ b/drivers/staging/olpc_dcon/olpc_dcon.c @@ -373,17 +373,17 @@ static void dcon_source_switch(struct work_struct *work) * * For now, we just hope.. */ - acquire_console_sem(); + console_lock(); ignore_fb_events = 1; if (fb_blank(fbinfo, FB_BLANK_UNBLANK)) { ignore_fb_events = 0; - release_console_sem(); + console_unlock(); printk(KERN_ERR "olpc-dcon: Failed to enter CPU mode\n"); dcon_pending = DCON_SOURCE_DCON; return; } ignore_fb_events = 0; - release_console_sem(); + console_unlock(); /* And turn off the DCON */ pdata->set_dconload(1); @@ -435,12 +435,12 @@ static void dcon_source_switch(struct work_struct *work) } } - acquire_console_sem(); + console_lock(); ignore_fb_events = 1; if (fb_blank(fbinfo, FB_BLANK_POWERDOWN)) printk(KERN_ERR "olpc-dcon: couldn't blank fb!\n"); ignore_fb_events = 0; - release_console_sem(); + console_unlock(); printk(KERN_INFO "olpc-dcon: The DCON has control\n"); break; diff --git a/drivers/staging/sm7xx/smtcfb.c b/drivers/staging/sm7xx/smtcfb.c index 0bc113c44d39..d007e4a12c14 100644 --- a/drivers/staging/sm7xx/smtcfb.c +++ b/drivers/staging/sm7xx/smtcfb.c @@ -1044,9 +1044,9 @@ static int __maybe_unused smtcfb_suspend(struct pci_dev *pdev, pm_message_t msg) /* when doing suspend, call fb apis and pci apis */ if (msg.event == PM_EVENT_SUSPEND) { - acquire_console_sem(); + console_lock(); fb_set_suspend(&sfb->fb, 1); - release_console_sem(); + console_unlock(); retv = pci_save_state(pdev); pci_disable_device(pdev); retv = pci_choose_state(pdev, msg); @@ -1105,9 +1105,9 @@ static int __maybe_unused smtcfb_resume(struct pci_dev *pdev) smtcfb_setmode(sfb); - acquire_console_sem(); + console_lock(); fb_set_suspend(&sfb->fb, 0); - release_console_sem(); + console_unlock(); return 0; } diff --git a/drivers/tty/serial/sb1250-duart.c b/drivers/tty/serial/sb1250-duart.c index a2f2b3254499..602d9845c52f 100644 --- a/drivers/tty/serial/sb1250-duart.c +++ b/drivers/tty/serial/sb1250-duart.c @@ -829,7 +829,7 @@ static void __init sbd_probe_duarts(void) #ifdef CONFIG_SERIAL_SB1250_DUART_CONSOLE /* * Serial console stuff. Very basic, polling driver for doing serial - * console output. The console_sem is held by the caller, so we + * console output. The console_lock is held by the caller, so we * shouldn't be interrupted for more console activity. */ static void sbd_console_putchar(struct uart_port *uport, int ch) diff --git a/drivers/tty/tty_io.c b/drivers/tty/tty_io.c index 464d09d97873..6158eae0f64a 100644 --- a/drivers/tty/tty_io.c +++ b/drivers/tty/tty_io.c @@ -3256,7 +3256,7 @@ static ssize_t show_cons_active(struct device *dev, struct console *c; ssize_t count = 0; - acquire_console_sem(); + console_lock(); for (c = console_drivers; c; c = c->next) { if (!c->device) continue; @@ -3271,7 +3271,7 @@ static ssize_t show_cons_active(struct device *dev, while (i--) count += sprintf(buf + count, "%s%d%c", cs[i]->name, cs[i]->index, i ? ' ':'\n'); - release_console_sem(); + console_unlock(); return count; } diff --git a/drivers/tty/vt/selection.c b/drivers/tty/vt/selection.c index ebae344ce910..c956ed6c83a3 100644 --- a/drivers/tty/vt/selection.c +++ b/drivers/tty/vt/selection.c @@ -316,9 +316,9 @@ int paste_selection(struct tty_struct *tty) /* always called with BTM from vt_ioctl */ WARN_ON(!tty_locked()); - acquire_console_sem(); + console_lock(); poke_blanked_console(); - release_console_sem(); + console_unlock(); ld = tty_ldisc_ref(tty); if (!ld) { diff --git a/drivers/tty/vt/vc_screen.c b/drivers/tty/vt/vc_screen.c index eab3a1ff99e4..a672ed192d33 100644 --- a/drivers/tty/vt/vc_screen.c +++ b/drivers/tty/vt/vc_screen.c @@ -202,7 +202,7 @@ vcs_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) /* Select the proper current console and verify * sanity of the situation under the console lock. */ - acquire_console_sem(); + console_lock(); attr = (currcons & 128); currcons = (currcons & 127); @@ -336,9 +336,9 @@ vcs_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) * the pagefault handling code may want to call printk(). */ - release_console_sem(); + console_unlock(); ret = copy_to_user(buf, con_buf_start, orig_count); - acquire_console_sem(); + console_lock(); if (ret) { read += (orig_count - ret); @@ -354,7 +354,7 @@ vcs_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) if (read) ret = read; unlock_out: - release_console_sem(); + console_unlock(); mutex_unlock(&con_buf_mtx); return ret; } @@ -379,7 +379,7 @@ vcs_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos) /* Select the proper current console and verify * sanity of the situation under the console lock. */ - acquire_console_sem(); + console_lock(); attr = (currcons & 128); currcons = (currcons & 127); @@ -414,9 +414,9 @@ vcs_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos) /* Temporarily drop the console lock so that we can read * in the write data from userspace safely. */ - release_console_sem(); + console_unlock(); ret = copy_from_user(con_buf, buf, this_round); - acquire_console_sem(); + console_lock(); if (ret) { this_round -= ret; @@ -542,7 +542,7 @@ vcs_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos) vcs_scr_updated(vc); unlock_out: - release_console_sem(); + console_unlock(); mutex_unlock(&con_buf_mtx); diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c index 76407eca9ab0..b230bd3f056f 100644 --- a/drivers/tty/vt/vt.c +++ b/drivers/tty/vt/vt.c @@ -1003,9 +1003,9 @@ static int vt_resize(struct tty_struct *tty, struct winsize *ws) struct vc_data *vc = tty->driver_data; int ret; - acquire_console_sem(); + console_lock(); ret = vc_do_resize(tty, vc, ws->ws_col, ws->ws_row); - release_console_sem(); + console_unlock(); return ret; } @@ -1271,7 +1271,7 @@ static void default_attr(struct vc_data *vc) vc->vc_color = vc->vc_def_color; } -/* console_sem is held */ +/* console_lock is held */ static void csi_m(struct vc_data *vc) { int i; @@ -1415,7 +1415,7 @@ int mouse_reporting(void) return vc_cons[fg_console].d->vc_report_mouse; } -/* console_sem is held */ +/* console_lock is held */ static void set_mode(struct vc_data *vc, int on_off) { int i; @@ -1485,7 +1485,7 @@ static void set_mode(struct vc_data *vc, int on_off) } } -/* console_sem is held */ +/* console_lock is held */ static void setterm_command(struct vc_data *vc) { switch(vc->vc_par[0]) { @@ -1545,7 +1545,7 @@ static void setterm_command(struct vc_data *vc) } } -/* console_sem is held */ +/* console_lock is held */ static void csi_at(struct vc_data *vc, unsigned int nr) { if (nr > vc->vc_cols - vc->vc_x) @@ -1555,7 +1555,7 @@ static void csi_at(struct vc_data *vc, unsigned int nr) insert_char(vc, nr); } -/* console_sem is held */ +/* console_lock is held */ static void csi_L(struct vc_data *vc, unsigned int nr) { if (nr > vc->vc_rows - vc->vc_y) @@ -1566,7 +1566,7 @@ static void csi_L(struct vc_data *vc, unsigned int nr) vc->vc_need_wrap = 0; } -/* console_sem is held */ +/* console_lock is held */ static void csi_P(struct vc_data *vc, unsigned int nr) { if (nr > vc->vc_cols - vc->vc_x) @@ -1576,7 +1576,7 @@ static void csi_P(struct vc_data *vc, unsigned int nr) delete_char(vc, nr); } -/* console_sem is held */ +/* console_lock is held */ static void csi_M(struct vc_data *vc, unsigned int nr) { if (nr > vc->vc_rows - vc->vc_y) @@ -1587,7 +1587,7 @@ static void csi_M(struct vc_data *vc, unsigned int nr) vc->vc_need_wrap = 0; } -/* console_sem is held (except via vc_init->reset_terminal */ +/* console_lock is held (except via vc_init->reset_terminal */ static void save_cur(struct vc_data *vc) { vc->vc_saved_x = vc->vc_x; @@ -1603,7 +1603,7 @@ static void save_cur(struct vc_data *vc) vc->vc_saved_G1 = vc->vc_G1_charset; } -/* console_sem is held */ +/* console_lock is held */ static void restore_cur(struct vc_data *vc) { gotoxy(vc, vc->vc_saved_x, vc->vc_saved_y); @@ -1625,7 +1625,7 @@ enum { ESnormal, ESesc, ESsquare, ESgetpars, ESgotpars, ESfunckey, EShash, ESsetG0, ESsetG1, ESpercent, ESignore, ESnonstd, ESpalette }; -/* console_sem is held (except via vc_init()) */ +/* console_lock is held (except via vc_init()) */ static void reset_terminal(struct vc_data *vc, int do_clear) { vc->vc_top = 0; @@ -1685,7 +1685,7 @@ static void reset_terminal(struct vc_data *vc, int do_clear) csi_J(vc, 2); } -/* console_sem is held */ +/* console_lock is held */ static void do_con_trol(struct tty_struct *tty, struct vc_data *vc, int c) { /* @@ -2119,7 +2119,7 @@ static int is_double_width(uint32_t ucs) return bisearch(ucs, double_width, ARRAY_SIZE(double_width) - 1); } -/* acquires console_sem */ +/* acquires console_lock */ static int do_con_write(struct tty_struct *tty, const unsigned char *buf, int count) { #ifdef VT_BUF_VRAM_ONLY @@ -2147,11 +2147,11 @@ static int do_con_write(struct tty_struct *tty, const unsigned char *buf, int co might_sleep(); - acquire_console_sem(); + console_lock(); vc = tty->driver_data; if (vc == NULL) { printk(KERN_ERR "vt: argh, driver_data is NULL !\n"); - release_console_sem(); + console_unlock(); return 0; } @@ -2159,7 +2159,7 @@ static int do_con_write(struct tty_struct *tty, const unsigned char *buf, int co if (!vc_cons_allocated(currcons)) { /* could this happen? */ printk_once("con_write: tty %d not allocated\n", currcons+1); - release_console_sem(); + console_unlock(); return 0; } @@ -2375,7 +2375,7 @@ rescan_last_byte: } FLUSH console_conditional_schedule(); - release_console_sem(); + console_unlock(); notify_update(vc); return n; #undef FLUSH @@ -2388,11 +2388,11 @@ rescan_last_byte: * us to do the switches asynchronously (needed when we want * to switch due to a keyboard interrupt). Synchronization * with other console code and prevention of re-entrancy is - * ensured with console_sem. + * ensured with console_lock. */ static void console_callback(struct work_struct *ignored) { - acquire_console_sem(); + console_lock(); if (want_console >= 0) { if (want_console != fg_console && @@ -2422,7 +2422,7 @@ static void console_callback(struct work_struct *ignored) } notify_update(vc_cons[fg_console].d); - release_console_sem(); + console_unlock(); } int set_console(int nr) @@ -2603,7 +2603,7 @@ static struct console vt_console_driver = { */ /* - * Generally a bit racy with respect to console_sem(). + * Generally a bit racy with respect to console_lock();. * * There are some functions which don't need it. * @@ -2629,17 +2629,17 @@ int tioclinux(struct tty_struct *tty, unsigned long arg) switch (type) { case TIOCL_SETSEL: - acquire_console_sem(); + console_lock(); ret = set_selection((struct tiocl_selection __user *)(p+1), tty); - release_console_sem(); + console_unlock(); break; case TIOCL_PASTESEL: ret = paste_selection(tty); break; case TIOCL_UNBLANKSCREEN: - acquire_console_sem(); + console_lock(); unblank_screen(); - release_console_sem(); + console_unlock(); break; case TIOCL_SELLOADLUT: ret = sel_loadlut(p); @@ -2688,10 +2688,10 @@ int tioclinux(struct tty_struct *tty, unsigned long arg) } break; case TIOCL_BLANKSCREEN: /* until explicitly unblanked, not only poked */ - acquire_console_sem(); + console_lock(); ignore_poke = 1; do_blank_screen(0); - release_console_sem(); + console_unlock(); break; case TIOCL_BLANKEDSCREEN: ret = console_blanked; @@ -2790,11 +2790,11 @@ static void con_flush_chars(struct tty_struct *tty) return; /* if we race with con_close(), vt may be null */ - acquire_console_sem(); + console_lock(); vc = tty->driver_data; if (vc) set_cursor(vc); - release_console_sem(); + console_unlock(); } /* @@ -2805,7 +2805,7 @@ static int con_open(struct tty_struct *tty, struct file *filp) unsigned int currcons = tty->index; int ret = 0; - acquire_console_sem(); + console_lock(); if (tty->driver_data == NULL) { ret = vc_allocate(currcons); if (ret == 0) { @@ -2813,7 +2813,7 @@ static int con_open(struct tty_struct *tty, struct file *filp) /* Still being freed */ if (vc->port.tty) { - release_console_sem(); + console_unlock(); return -ERESTARTSYS; } tty->driver_data = vc; @@ -2827,11 +2827,11 @@ static int con_open(struct tty_struct *tty, struct file *filp) tty->termios->c_iflag |= IUTF8; else tty->termios->c_iflag &= ~IUTF8; - release_console_sem(); + console_unlock(); return ret; } } - release_console_sem(); + console_unlock(); return ret; } @@ -2844,9 +2844,9 @@ static void con_shutdown(struct tty_struct *tty) { struct vc_data *vc = tty->driver_data; BUG_ON(vc == NULL); - acquire_console_sem(); + console_lock(); vc->port.tty = NULL; - release_console_sem(); + console_unlock(); tty_shutdown(tty); } @@ -2893,13 +2893,13 @@ static int __init con_init(void) struct vc_data *vc; unsigned int currcons = 0, i; - acquire_console_sem(); + console_lock(); if (conswitchp) display_desc = conswitchp->con_startup(); if (!display_desc) { fg_console = 0; - release_console_sem(); + console_unlock(); return 0; } @@ -2946,7 +2946,7 @@ static int __init con_init(void) printable = 1; printk("\n"); - release_console_sem(); + console_unlock(); #ifdef CONFIG_VT_CONSOLE register_console(&vt_console_driver); @@ -3037,7 +3037,7 @@ static int bind_con_driver(const struct consw *csw, int first, int last, if (!try_module_get(owner)) return -ENODEV; - acquire_console_sem(); + console_lock(); /* check if driver is registered */ for (i = 0; i < MAX_NR_CON_DRIVER; i++) { @@ -3122,7 +3122,7 @@ static int bind_con_driver(const struct consw *csw, int first, int last, retval = 0; err: - release_console_sem(); + console_unlock(); module_put(owner); return retval; }; @@ -3171,7 +3171,7 @@ int unbind_con_driver(const struct consw *csw, int first, int last, int deflt) if (!try_module_get(owner)) return -ENODEV; - acquire_console_sem(); + console_lock(); /* check if driver is registered and if it is unbindable */ for (i = 0; i < MAX_NR_CON_DRIVER; i++) { @@ -3185,7 +3185,7 @@ int unbind_con_driver(const struct consw *csw, int first, int last, int deflt) } if (retval) { - release_console_sem(); + console_unlock(); goto err; } @@ -3204,12 +3204,12 @@ int unbind_con_driver(const struct consw *csw, int first, int last, int deflt) } if (retval) { - release_console_sem(); + console_unlock(); goto err; } if (!con_is_bound(csw)) { - release_console_sem(); + console_unlock(); goto err; } @@ -3238,7 +3238,7 @@ int unbind_con_driver(const struct consw *csw, int first, int last, int deflt) if (!con_is_bound(csw)) con_driver->flag &= ~CON_DRIVER_FLAG_INIT; - release_console_sem(); + console_unlock(); /* ignore return value, binding should not fail */ bind_con_driver(defcsw, first, last, deflt); err: @@ -3538,7 +3538,7 @@ int register_con_driver(const struct consw *csw, int first, int last) if (!try_module_get(owner)) return -ENODEV; - acquire_console_sem(); + console_lock(); for (i = 0; i < MAX_NR_CON_DRIVER; i++) { con_driver = ®istered_con_driver[i]; @@ -3592,7 +3592,7 @@ int register_con_driver(const struct consw *csw, int first, int last) } err: - release_console_sem(); + console_unlock(); module_put(owner); return retval; } @@ -3613,7 +3613,7 @@ int unregister_con_driver(const struct consw *csw) { int i, retval = -ENODEV; - acquire_console_sem(); + console_lock(); /* cannot unregister a bound driver */ if (con_is_bound(csw)) @@ -3639,7 +3639,7 @@ int unregister_con_driver(const struct consw *csw) } } err: - release_console_sem(); + console_unlock(); return retval; } EXPORT_SYMBOL(unregister_con_driver); @@ -3934,9 +3934,9 @@ int con_set_cmap(unsigned char __user *arg) { int rc; - acquire_console_sem(); + console_lock(); rc = set_get_cmap (arg,1); - release_console_sem(); + console_unlock(); return rc; } @@ -3945,9 +3945,9 @@ int con_get_cmap(unsigned char __user *arg) { int rc; - acquire_console_sem(); + console_lock(); rc = set_get_cmap (arg,0); - release_console_sem(); + console_unlock(); return rc; } @@ -3994,12 +3994,12 @@ static int con_font_get(struct vc_data *vc, struct console_font_op *op) } else font.data = NULL; - acquire_console_sem(); + console_lock(); if (vc->vc_sw->con_font_get) rc = vc->vc_sw->con_font_get(vc, &font); else rc = -ENOSYS; - release_console_sem(); + console_unlock(); if (rc) goto out; @@ -4076,12 +4076,12 @@ static int con_font_set(struct vc_data *vc, struct console_font_op *op) font.data = memdup_user(op->data, size); if (IS_ERR(font.data)) return PTR_ERR(font.data); - acquire_console_sem(); + console_lock(); if (vc->vc_sw->con_font_set) rc = vc->vc_sw->con_font_set(vc, &font, op->flags); else rc = -ENOSYS; - release_console_sem(); + console_unlock(); kfree(font.data); return rc; } @@ -4103,12 +4103,12 @@ static int con_font_default(struct vc_data *vc, struct console_font_op *op) else name[MAX_FONT_NAME - 1] = 0; - acquire_console_sem(); + console_lock(); if (vc->vc_sw->con_font_default) rc = vc->vc_sw->con_font_default(vc, &font, s); else rc = -ENOSYS; - release_console_sem(); + console_unlock(); if (!rc) { op->width = font.width; op->height = font.height; @@ -4124,7 +4124,7 @@ static int con_font_copy(struct vc_data *vc, struct console_font_op *op) if (vc->vc_mode != KD_TEXT) return -EINVAL; - acquire_console_sem(); + console_lock(); if (!vc->vc_sw->con_font_copy) rc = -ENOSYS; else if (con < 0 || !vc_cons_allocated(con)) @@ -4133,7 +4133,7 @@ static int con_font_copy(struct vc_data *vc, struct console_font_op *op) rc = 0; else rc = vc->vc_sw->con_font_copy(vc, con); - release_console_sem(); + console_unlock(); return rc; } diff --git a/drivers/tty/vt/vt_ioctl.c b/drivers/tty/vt/vt_ioctl.c index 6b68a0fb4611..1235ebda6e1c 100644 --- a/drivers/tty/vt/vt_ioctl.c +++ b/drivers/tty/vt/vt_ioctl.c @@ -649,12 +649,12 @@ int vt_ioctl(struct tty_struct *tty, struct file * file, /* * explicitly blank/unblank the screen if switching modes */ - acquire_console_sem(); + console_lock(); if (arg == KD_TEXT) do_unblank_screen(1); else do_blank_screen(1); - release_console_sem(); + console_unlock(); break; case KDGETMODE: @@ -893,7 +893,7 @@ int vt_ioctl(struct tty_struct *tty, struct file * file, ret = -EINVAL; goto out; } - acquire_console_sem(); + console_lock(); vc->vt_mode = tmp; /* the frsig is ignored, so we set it to 0 */ vc->vt_mode.frsig = 0; @@ -901,7 +901,7 @@ int vt_ioctl(struct tty_struct *tty, struct file * file, vc->vt_pid = get_pid(task_pid(current)); /* no switch is required -- saw@shade.msu.ru */ vc->vt_newvt = -1; - release_console_sem(); + console_unlock(); break; } @@ -910,9 +910,9 @@ int vt_ioctl(struct tty_struct *tty, struct file * file, struct vt_mode tmp; int rc; - acquire_console_sem(); + console_lock(); memcpy(&tmp, &vc->vt_mode, sizeof(struct vt_mode)); - release_console_sem(); + console_unlock(); rc = copy_to_user(up, &tmp, sizeof(struct vt_mode)); if (rc) @@ -965,9 +965,9 @@ int vt_ioctl(struct tty_struct *tty, struct file * file, ret = -ENXIO; else { arg--; - acquire_console_sem(); + console_lock(); ret = vc_allocate(arg); - release_console_sem(); + console_unlock(); if (ret) break; set_console(arg); @@ -990,7 +990,7 @@ int vt_ioctl(struct tty_struct *tty, struct file * file, ret = -ENXIO; else { vsa.console--; - acquire_console_sem(); + console_lock(); ret = vc_allocate(vsa.console); if (ret == 0) { struct vc_data *nvc; @@ -1003,7 +1003,7 @@ int vt_ioctl(struct tty_struct *tty, struct file * file, put_pid(nvc->vt_pid); nvc->vt_pid = get_pid(task_pid(current)); } - release_console_sem(); + console_unlock(); if (ret) break; /* Commence switch and lock */ @@ -1044,7 +1044,7 @@ int vt_ioctl(struct tty_struct *tty, struct file * file, /* * Switching-from response */ - acquire_console_sem(); + console_lock(); if (vc->vt_newvt >= 0) { if (arg == 0) /* @@ -1063,7 +1063,7 @@ int vt_ioctl(struct tty_struct *tty, struct file * file, vc->vt_newvt = -1; ret = vc_allocate(newvt); if (ret) { - release_console_sem(); + console_unlock(); break; } /* @@ -1083,7 +1083,7 @@ int vt_ioctl(struct tty_struct *tty, struct file * file, if (arg != VT_ACKACQ) ret = -EINVAL; } - release_console_sem(); + console_unlock(); break; /* @@ -1096,20 +1096,20 @@ int vt_ioctl(struct tty_struct *tty, struct file * file, } if (arg == 0) { /* deallocate all unused consoles, but leave 0 */ - acquire_console_sem(); + console_lock(); for (i=1; iv_cols)) ret = -EFAULT; else { - acquire_console_sem(); + console_lock(); for (i = 0; i < MAX_NR_CONSOLES; i++) { vc = vc_cons[i].d; @@ -1135,7 +1135,7 @@ int vt_ioctl(struct tty_struct *tty, struct file * file, vc_resize(vc_cons[i].d, cc, ll); } } - release_console_sem(); + console_unlock(); } break; } @@ -1187,14 +1187,14 @@ int vt_ioctl(struct tty_struct *tty, struct file * file, for (i = 0; i < MAX_NR_CONSOLES; i++) { if (!vc_cons[i].d) continue; - acquire_console_sem(); + console_lock(); if (vlin) vc_cons[i].d->vc_scan_lines = vlin; if (clin) vc_cons[i].d->vc_font.height = clin; vc_cons[i].d->vc_resize_user = 1; vc_resize(vc_cons[i].d, cc, ll); - release_console_sem(); + console_unlock(); } break; } @@ -1367,7 +1367,7 @@ void vc_SAK(struct work_struct *work) struct vc_data *vc; struct tty_struct *tty; - acquire_console_sem(); + console_lock(); vc = vc_con->d; if (vc) { tty = vc->port.tty; @@ -1379,7 +1379,7 @@ void vc_SAK(struct work_struct *work) __do_SAK(tty); reset_vc(vc); } - release_console_sem(); + console_unlock(); } #ifdef CONFIG_COMPAT @@ -1737,10 +1737,10 @@ int vt_move_to_console(unsigned int vt, int alloc) { int prev; - acquire_console_sem(); + console_lock(); /* Graphics mode - up to X */ if (disable_vt_switch) { - release_console_sem(); + console_unlock(); return 0; } prev = fg_console; @@ -1748,7 +1748,7 @@ int vt_move_to_console(unsigned int vt, int alloc) if (alloc && vc_allocate(vt)) { /* we can't have a free VC for now. Too bad, * we don't want to mess the screen for now. */ - release_console_sem(); + console_unlock(); return -ENOSPC; } @@ -1758,10 +1758,10 @@ int vt_move_to_console(unsigned int vt, int alloc) * Let the calling function know so it can decide * what to do. */ - release_console_sem(); + console_unlock(); return -EIO; } - release_console_sem(); + console_unlock(); tty_lock(); if (vt_waitactive(vt + 1)) { pr_debug("Suspend: Can't switch VCs."); @@ -1781,8 +1781,8 @@ int vt_move_to_console(unsigned int vt, int alloc) */ void pm_set_vt_switch(int do_switch) { - acquire_console_sem(); + console_lock(); disable_vt_switch = !do_switch; - release_console_sem(); + console_unlock(); } EXPORT_SYMBOL(pm_set_vt_switch); diff --git a/drivers/video/arkfb.c b/drivers/video/arkfb.c index d583bea608fd..391ac939f011 100644 --- a/drivers/video/arkfb.c +++ b/drivers/video/arkfb.c @@ -23,7 +23,7 @@ #include #include #include -#include /* Why should fb driver call console functions? because acquire_console_sem() */ +#include /* Why should fb driver call console functions? because console_lock() */ #include