From 287e6611ab1eac76c2c5ebf6e345e04c80ca9c61 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 11 Feb 2016 14:16:27 +0100 Subject: libata: fix HDIO_GET_32BIT ioctl As reported by Soohoon Lee, the HDIO_GET_32BIT ioctl does not work correctly in compat mode with libata. I have investigated the issue further and found multiple problems that all appeared with the same commit that originally introduced HDIO_GET_32BIT handling in libata back in linux-2.6.8 and presumably also linux-2.4, as the code uses "copy_to_user(arg, &val, 1)" to copy a 'long' variable containing either 0 or 1 to user space. The problems with this are: * On big-endian machines, this will always write a zero because it stores the wrong byte into user space. * In compat mode, the upper three bytes of the variable are updated by the compat_hdio_ioctl() function, but they now contain uninitialized stack data. * The hdparm tool calling this ioctl uses a 'static long' variable to store the result. This means at least the upper bytes are initialized to zero, but calling another ioctl like HDIO_GET_MULTCOUNT would fill them with data that remains stale when the low byte is overwritten. Fortunately libata doesn't implement any of the affected ioctl commands, so this would only happen when we query both an IDE and an ATA device in the same command such as "hdparm -N -c /dev/hda /dev/sda" * The libata code for unknown reasons started using ATA_IOC_GET_IO32 and ATA_IOC_SET_IO32 as aliases for HDIO_GET_32BIT and HDIO_SET_32BIT, while the ioctl commands that were added later use the normal HDIO_* names. This is harmless but rather confusing. This addresses all four issues by changing the code to use put_user() on an 'unsigned long' variable in HDIO_GET_32BIT, like the IDE subsystem does, and by clarifying the names of the ioctl commands. Signed-off-by: Arnd Bergmann Reported-by: Soohoon Lee Tested-by: Soohoon Lee Cc: stable@vger.kernel.org Signed-off-by: Tejun Heo --- include/linux/ata.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/ata.h b/include/linux/ata.h index d2992bfa1706..c1a2f345cbe6 100644 --- a/include/linux/ata.h +++ b/include/linux/ata.h @@ -487,8 +487,8 @@ enum ata_tf_protocols { }; enum ata_ioctls { - ATA_IOC_GET_IO32 = 0x309, - ATA_IOC_SET_IO32 = 0x324, + ATA_IOC_GET_IO32 = 0x309, /* HDIO_GET_32BIT */ + ATA_IOC_SET_IO32 = 0x324, /* HDIO_SET_32BIT */ }; /* core structures */ -- cgit v1.2.3 From 91d31b9f8e7662726f273fc32b25f4099d78de4a Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Mon, 15 Feb 2016 05:22:18 +0000 Subject: regmap: add regmap_update_bits_base() Current regmap has many similar update functions like below, but the difference is very few. regmap_update_bits() regmap_update_bits_async() regmap_update_bits_check() regmap_update_bits_check_async() Furthermore, we can add *force* write option too in the future. This patch adds new regmap_update_bits_base() which is feature merged function. Above functions can be merged into it by macro. Signed-off-by: Kuninori Morimoto Signed-off-by: Mark Brown --- drivers/base/regmap/regmap.c | 40 ++++++++++++++++++++++++++++++++++++++++ include/linux/regmap.h | 11 +++++++++++ 2 files changed, 51 insertions(+) (limited to 'include/linux') diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index ee54e841de4a..4e35b2f41304 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -2647,6 +2647,46 @@ static int _regmap_update_bits(struct regmap *map, unsigned int reg, return ret; } +/** + * regmap_update_bits_base: + * Perform a read/modify/write cycle on the + * register map with change, async, force option + * + * @map: Register map to update + * @reg: Register to update + * @mask: Bitmask to change + * @val: New value for bitmask + * @change: Boolean indicating if a write was done + * @async: Boolean indicating asynchronously + * @force: Boolean indicating use force update + * + * if async was true, + * With most buses the read must be done synchronously so this is most + * useful for devices with a cache which do not need to interact with + * the hardware to determine the current register value. + * + * Returns zero for success, a negative number on error. + */ +int regmap_update_bits_base(struct regmap *map, unsigned int reg, + unsigned int mask, unsigned int val, + bool *change, bool async, bool force) +{ + int ret; + + map->lock(map->lock_arg); + + map->async = async; + + ret = _regmap_update_bits(map, reg, mask, val, change, force); + + map->async = false; + + map->unlock(map->lock_arg); + + return ret; +} +EXPORT_SYMBOL_GPL(regmap_update_bits_base); + /** * regmap_update_bits: Perform a read/modify/write cycle on the register map * diff --git a/include/linux/regmap.h b/include/linux/regmap.h index 18394343f489..28e50a3d2872 100644 --- a/include/linux/regmap.h +++ b/include/linux/regmap.h @@ -691,6 +691,9 @@ int regmap_raw_read(struct regmap *map, unsigned int reg, void *val, size_t val_len); int regmap_bulk_read(struct regmap *map, unsigned int reg, void *val, size_t val_count); +int regmap_update_bits_base(struct regmap *map, unsigned int reg, + unsigned int mask, unsigned int val, + bool *change, bool async, bool force); int regmap_update_bits(struct regmap *map, unsigned int reg, unsigned int mask, unsigned int val); int regmap_write_bits(struct regmap *map, unsigned int reg, @@ -937,6 +940,14 @@ static inline int regmap_bulk_read(struct regmap *map, unsigned int reg, return -EINVAL; } +static inline int regmap_update_bits_base(struct regmap *map, unsigned int reg, + unsigned int mask, unsigned int val, + bool *change, bool async, bool force) +{ + WARN_ONCE(1, "regmap API is disabled"); + return -EINVAL; +} + static inline int regmap_update_bits(struct regmap *map, unsigned int reg, unsigned int mask, unsigned int val) { -- cgit v1.2.3 From ca7a94464b5457a8dc5add19f6fc3bea59d6193f Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Mon, 15 Feb 2016 05:22:42 +0000 Subject: regmap: merge regmap_update_bits() into macro Current regmap has many similar update functions like below, but the difference is very few. regmap_update_bits() regmap_update_bits_async() regmap_update_bits_check() regmap_update_bits_check_async() Furthermore, we can add *force* write option too in the future. This patch merges regmap_update_bits() into macro by using regmap_update_bits_base(). Signed-off-by: Kuninori Morimoto Signed-off-by: Mark Brown --- drivers/base/regmap/regmap.c | 23 ----------------------- include/linux/regmap.h | 12 +++--------- 2 files changed, 3 insertions(+), 32 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index 4e35b2f41304..281898a97e8f 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -2687,29 +2687,6 @@ int regmap_update_bits_base(struct regmap *map, unsigned int reg, } EXPORT_SYMBOL_GPL(regmap_update_bits_base); -/** - * regmap_update_bits: Perform a read/modify/write cycle on the register map - * - * @map: Register map to update - * @reg: Register to update - * @mask: Bitmask to change - * @val: New value for bitmask - * - * Returns zero for success, a negative number on error. - */ -int regmap_update_bits(struct regmap *map, unsigned int reg, - unsigned int mask, unsigned int val) -{ - int ret; - - map->lock(map->lock_arg); - ret = _regmap_update_bits(map, reg, mask, val, NULL, false); - map->unlock(map->lock_arg); - - return ret; -} -EXPORT_SYMBOL_GPL(regmap_update_bits); - /** * regmap_write_bits: Perform a read/modify/write cycle on the register map * diff --git a/include/linux/regmap.h b/include/linux/regmap.h index 28e50a3d2872..500b36cbc7aa 100644 --- a/include/linux/regmap.h +++ b/include/linux/regmap.h @@ -65,6 +65,9 @@ struct reg_sequence { unsigned int delay_us; }; +#define regmap_update_bits(map, reg, mask, val) \ + regmap_update_bits_base(map, reg, mask, val, NULL, false, false) + #ifdef CONFIG_REGMAP enum regmap_endian { @@ -694,8 +697,6 @@ int regmap_bulk_read(struct regmap *map, unsigned int reg, void *val, int regmap_update_bits_base(struct regmap *map, unsigned int reg, unsigned int mask, unsigned int val, bool *change, bool async, bool force); -int regmap_update_bits(struct regmap *map, unsigned int reg, - unsigned int mask, unsigned int val); int regmap_write_bits(struct regmap *map, unsigned int reg, unsigned int mask, unsigned int val); int regmap_update_bits_async(struct regmap *map, unsigned int reg, @@ -948,13 +949,6 @@ static inline int regmap_update_bits_base(struct regmap *map, unsigned int reg, return -EINVAL; } -static inline int regmap_update_bits(struct regmap *map, unsigned int reg, - unsigned int mask, unsigned int val) -{ - WARN_ONCE(1, "regmap API is disabled"); - return -EINVAL; -} - static inline int regmap_write_bits(struct regmap *map, unsigned int reg, unsigned int mask, unsigned int val) { -- cgit v1.2.3 From 30ed9cb7a49b499ebc6061e4ff38e88cb4857cad Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Mon, 15 Feb 2016 05:23:01 +0000 Subject: regmap: merge regmap_update_bits_async() into macro Current regmap has many similar update functions like below, but the difference is very few. regmap_update_bits() regmap_update_bits_async() regmap_update_bits_check() regmap_update_bits_check_async() Furthermore, we can add *force* write option too in the future. This patch merges regmap_update_bits_async() into macro by using regmap_update_bits_base(). Signed-off-by: Kuninori Morimoto Signed-off-by: Mark Brown --- drivers/base/regmap/regmap.c | 34 ---------------------------------- include/linux/regmap.h | 12 ++---------- 2 files changed, 2 insertions(+), 44 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index 281898a97e8f..c2255f604c03 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -2710,40 +2710,6 @@ int regmap_write_bits(struct regmap *map, unsigned int reg, } EXPORT_SYMBOL_GPL(regmap_write_bits); -/** - * regmap_update_bits_async: Perform a read/modify/write cycle on the register - * map asynchronously - * - * @map: Register map to update - * @reg: Register to update - * @mask: Bitmask to change - * @val: New value for bitmask - * - * With most buses the read must be done synchronously so this is most - * useful for devices with a cache which do not need to interact with - * the hardware to determine the current register value. - * - * Returns zero for success, a negative number on error. - */ -int regmap_update_bits_async(struct regmap *map, unsigned int reg, - unsigned int mask, unsigned int val) -{ - int ret; - - map->lock(map->lock_arg); - - map->async = true; - - ret = _regmap_update_bits(map, reg, mask, val, NULL, false); - - map->async = false; - - map->unlock(map->lock_arg); - - return ret; -} -EXPORT_SYMBOL_GPL(regmap_update_bits_async); - /** * regmap_update_bits_check: Perform a read/modify/write cycle on the * register map and report if updated diff --git a/include/linux/regmap.h b/include/linux/regmap.h index 500b36cbc7aa..90c8b0e99f9d 100644 --- a/include/linux/regmap.h +++ b/include/linux/regmap.h @@ -67,6 +67,8 @@ struct reg_sequence { #define regmap_update_bits(map, reg, mask, val) \ regmap_update_bits_base(map, reg, mask, val, NULL, false, false) +#define regmap_update_bits_async(map, reg, mask, val)\ + regmap_update_bits_base(map, reg, mask, val, NULL, true, false) #ifdef CONFIG_REGMAP @@ -699,8 +701,6 @@ int regmap_update_bits_base(struct regmap *map, unsigned int reg, bool *change, bool async, bool force); int regmap_write_bits(struct regmap *map, unsigned int reg, unsigned int mask, unsigned int val); -int regmap_update_bits_async(struct regmap *map, unsigned int reg, - unsigned int mask, unsigned int val); int regmap_update_bits_check(struct regmap *map, unsigned int reg, unsigned int mask, unsigned int val, bool *change); @@ -956,14 +956,6 @@ static inline int regmap_write_bits(struct regmap *map, unsigned int reg, return -EINVAL; } -static inline int regmap_update_bits_async(struct regmap *map, - unsigned int reg, - unsigned int mask, unsigned int val) -{ - WARN_ONCE(1, "regmap API is disabled"); - return -EINVAL; -} - static inline int regmap_update_bits_check(struct regmap *map, unsigned int reg, unsigned int mask, unsigned int val, -- cgit v1.2.3 From 98c2dc48694a47109fff430a216fc13a9b45a4a1 Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Mon, 15 Feb 2016 05:23:17 +0000 Subject: regmap: merge regmap_update_bits_check() into macro Current regmap has many similar update functions like below, but the difference is very few. regmap_update_bits() regmap_update_bits_async() regmap_update_bits_check() regmap_update_bits_check_async() Furthermore, we can add *force* write option too in the future. This patch merges regmap_update_bits_check() into macro by using regmap_update_bits_base(). Signed-off-by: Kuninori Morimoto Signed-off-by: Mark Brown --- drivers/base/regmap/regmap.c | 25 ------------------------- include/linux/regmap.h | 14 ++------------ 2 files changed, 2 insertions(+), 37 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index c2255f604c03..ce24e9688b05 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -2710,31 +2710,6 @@ int regmap_write_bits(struct regmap *map, unsigned int reg, } EXPORT_SYMBOL_GPL(regmap_write_bits); -/** - * regmap_update_bits_check: Perform a read/modify/write cycle on the - * register map and report if updated - * - * @map: Register map to update - * @reg: Register to update - * @mask: Bitmask to change - * @val: New value for bitmask - * @change: Boolean indicating if a write was done - * - * Returns zero for success, a negative number on error. - */ -int regmap_update_bits_check(struct regmap *map, unsigned int reg, - unsigned int mask, unsigned int val, - bool *change) -{ - int ret; - - map->lock(map->lock_arg); - ret = _regmap_update_bits(map, reg, mask, val, change, false); - map->unlock(map->lock_arg); - return ret; -} -EXPORT_SYMBOL_GPL(regmap_update_bits_check); - /** * regmap_update_bits_check_async: Perform a read/modify/write cycle on the * register map asynchronously and report if diff --git a/include/linux/regmap.h b/include/linux/regmap.h index 90c8b0e99f9d..dd227dd5e5f8 100644 --- a/include/linux/regmap.h +++ b/include/linux/regmap.h @@ -69,6 +69,8 @@ struct reg_sequence { regmap_update_bits_base(map, reg, mask, val, NULL, false, false) #define regmap_update_bits_async(map, reg, mask, val)\ regmap_update_bits_base(map, reg, mask, val, NULL, true, false) +#define regmap_update_bits_check(map, reg, mask, val, change)\ + regmap_update_bits_base(map, reg, mask, val, change, false, false) #ifdef CONFIG_REGMAP @@ -701,9 +703,6 @@ int regmap_update_bits_base(struct regmap *map, unsigned int reg, bool *change, bool async, bool force); int regmap_write_bits(struct regmap *map, unsigned int reg, unsigned int mask, unsigned int val); -int regmap_update_bits_check(struct regmap *map, unsigned int reg, - unsigned int mask, unsigned int val, - bool *change); int regmap_update_bits_check_async(struct regmap *map, unsigned int reg, unsigned int mask, unsigned int val, bool *change); @@ -956,15 +955,6 @@ static inline int regmap_write_bits(struct regmap *map, unsigned int reg, return -EINVAL; } -static inline int regmap_update_bits_check(struct regmap *map, - unsigned int reg, - unsigned int mask, unsigned int val, - bool *change) -{ - WARN_ONCE(1, "regmap API is disabled"); - return -EINVAL; -} - static inline int regmap_update_bits_check_async(struct regmap *map, unsigned int reg, unsigned int mask, -- cgit v1.2.3 From 89d8d4b833b0b29e0e95bd0cd51e80f5ee7a6b0a Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Mon, 15 Feb 2016 05:23:37 +0000 Subject: regmap: merge regmap_update_bits_check_async() into macro Current regmap has many similar update functions like below, but the difference is very few. regmap_update_bits() regmap_update_bits_async() regmap_update_bits_check() regmap_update_bits_check_async() Furthermore, we can add *force* write option too in the future. This patch merges regmap_update_bits_check_async() into macro by using regmap_update_bits_base(). Signed-off-by: Kuninori Morimoto Signed-off-by: Mark Brown --- drivers/base/regmap/regmap.c | 37 ------------------------------------- include/linux/regmap.h | 15 ++------------- 2 files changed, 2 insertions(+), 50 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index ce24e9688b05..015135a656b7 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -2710,43 +2710,6 @@ int regmap_write_bits(struct regmap *map, unsigned int reg, } EXPORT_SYMBOL_GPL(regmap_write_bits); -/** - * regmap_update_bits_check_async: Perform a read/modify/write cycle on the - * register map asynchronously and report if - * updated - * - * @map: Register map to update - * @reg: Register to update - * @mask: Bitmask to change - * @val: New value for bitmask - * @change: Boolean indicating if a write was done - * - * With most buses the read must be done synchronously so this is most - * useful for devices with a cache which do not need to interact with - * the hardware to determine the current register value. - * - * Returns zero for success, a negative number on error. - */ -int regmap_update_bits_check_async(struct regmap *map, unsigned int reg, - unsigned int mask, unsigned int val, - bool *change) -{ - int ret; - - map->lock(map->lock_arg); - - map->async = true; - - ret = _regmap_update_bits(map, reg, mask, val, change, false); - - map->async = false; - - map->unlock(map->lock_arg); - - return ret; -} -EXPORT_SYMBOL_GPL(regmap_update_bits_check_async); - void regmap_async_complete_cb(struct regmap_async *async, int ret) { struct regmap *map = async->map; diff --git a/include/linux/regmap.h b/include/linux/regmap.h index dd227dd5e5f8..4d8e1edb4407 100644 --- a/include/linux/regmap.h +++ b/include/linux/regmap.h @@ -71,6 +71,8 @@ struct reg_sequence { regmap_update_bits_base(map, reg, mask, val, NULL, true, false) #define regmap_update_bits_check(map, reg, mask, val, change)\ regmap_update_bits_base(map, reg, mask, val, change, false, false) +#define regmap_update_bits_check_async(map, reg, mask, val, change)\ + regmap_update_bits_base(map, reg, mask, val, change, true, false) #ifdef CONFIG_REGMAP @@ -703,9 +705,6 @@ int regmap_update_bits_base(struct regmap *map, unsigned int reg, bool *change, bool async, bool force); int regmap_write_bits(struct regmap *map, unsigned int reg, unsigned int mask, unsigned int val); -int regmap_update_bits_check_async(struct regmap *map, unsigned int reg, - unsigned int mask, unsigned int val, - bool *change); int regmap_get_val_bytes(struct regmap *map); int regmap_get_max_register(struct regmap *map); int regmap_get_reg_stride(struct regmap *map); @@ -955,16 +954,6 @@ static inline int regmap_write_bits(struct regmap *map, unsigned int reg, return -EINVAL; } -static inline int regmap_update_bits_check_async(struct regmap *map, - unsigned int reg, - unsigned int mask, - unsigned int val, - bool *change) -{ - WARN_ONCE(1, "regmap API is disabled"); - return -EINVAL; -} - static inline int regmap_get_val_bytes(struct regmap *map) { WARN_ONCE(1, "regmap API is disabled"); -- cgit v1.2.3 From 28972eaa34f384eef5e33f36e00d8fa21ca44375 Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Mon, 15 Feb 2016 05:23:55 +0000 Subject: regmap: add regmap_field_update_bits_base() This patch adds new regmap_field_update_bits_base() which is using regmap_update_bits_base(). Current regmap_field_xxx() can be merged into it by macro. Signed-off-by: Kuninori Morimoto Signed-off-by: Mark Brown --- drivers/base/regmap/regmap.c | 27 +++++++++++++++++++++++++++ include/linux/regmap.h | 12 +++++++++++- 2 files changed, 38 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index 015135a656b7..e534105f47f6 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -1689,6 +1689,33 @@ int regmap_raw_write(struct regmap *map, unsigned int reg, } EXPORT_SYMBOL_GPL(regmap_raw_write); +/** + * regmap_field_update_bits_base(): + * Perform a read/modify/write cycle on the register field + * with change, async, force option + * + * @field: Register field to write to + * @mask: Bitmask to change + * @val: Value to be written + * @change: Boolean indicating if a write was done + * @async: Boolean indicating asynchronously + * @force: Boolean indicating use force update + * + * A value of zero will be returned on success, a negative errno will + * be returned in error cases. + */ +int regmap_field_update_bits_base(struct regmap_field *field, + unsigned int mask, unsigned int val, + bool *change, bool async, bool force) +{ + mask = (mask << field->shift) & field->mask; + + return regmap_update_bits_base(field->regmap, field->reg, + mask, val << field->shift, + change, async, force); +} +EXPORT_SYMBOL_GPL(regmap_field_update_bits_base); + /** * regmap_field_write(): Write a value to a single register field * diff --git a/include/linux/regmap.h b/include/linux/regmap.h index 4d8e1edb4407..23bf7657e485 100644 --- a/include/linux/regmap.h +++ b/include/linux/regmap.h @@ -773,9 +773,11 @@ void devm_regmap_field_free(struct device *dev, struct regmap_field *field); int regmap_field_read(struct regmap_field *field, unsigned int *val); int regmap_field_write(struct regmap_field *field, unsigned int val); +int regmap_field_update_bits_base(struct regmap_field *field, + unsigned int mask, unsigned int val, + bool *change, bool async, bool force); int regmap_field_update_bits(struct regmap_field *field, unsigned int mask, unsigned int val); - int regmap_fields_write(struct regmap_field *field, unsigned int id, unsigned int val); int regmap_fields_force_write(struct regmap_field *field, unsigned int id, @@ -954,6 +956,14 @@ static inline int regmap_write_bits(struct regmap *map, unsigned int reg, return -EINVAL; } +static inline int regmap_field_update_bits_base(struct regmap_field *field, + unsigned int mask, unsigned int val, + bool *change, bool async, bool force) +{ + WARN_ONCE(1, "regmap API is disabled"); + return -EINVAL; +} + static inline int regmap_get_val_bytes(struct regmap *map) { WARN_ONCE(1, "regmap API is disabled"); -- cgit v1.2.3 From 3674124b35894631f8f4d33ab041e713078bfd4b Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Mon, 15 Feb 2016 05:24:15 +0000 Subject: regmap: merge regmap_field_write() into macro This patch merges regmap_field_write() into macro by using regmap_field_update_bits_base(). Signed-off-by: Kuninori Morimoto Signed-off-by: Mark Brown --- drivers/base/regmap/regmap.c | 16 ---------------- include/linux/regmap.h | 4 +++- 2 files changed, 3 insertions(+), 17 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index e534105f47f6..228dce237658 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -1716,22 +1716,6 @@ int regmap_field_update_bits_base(struct regmap_field *field, } EXPORT_SYMBOL_GPL(regmap_field_update_bits_base); -/** - * regmap_field_write(): Write a value to a single register field - * - * @field: Register field to write to - * @val: Value to be written - * - * A value of zero will be returned on success, a negative errno will - * be returned in error cases. - */ -int regmap_field_write(struct regmap_field *field, unsigned int val) -{ - return regmap_update_bits(field->regmap, field->reg, - field->mask, val << field->shift); -} -EXPORT_SYMBOL_GPL(regmap_field_write); - /** * regmap_field_update_bits(): Perform a read/modify/write cycle * on the register field diff --git a/include/linux/regmap.h b/include/linux/regmap.h index 23bf7657e485..13e9ebdea1c7 100644 --- a/include/linux/regmap.h +++ b/include/linux/regmap.h @@ -74,6 +74,9 @@ struct reg_sequence { #define regmap_update_bits_check_async(map, reg, mask, val, change)\ regmap_update_bits_base(map, reg, mask, val, change, true, false) +#define regmap_field_write(field, val) \ + regmap_field_update_bits_base(field, ~0, val, NULL, false, false) + #ifdef CONFIG_REGMAP enum regmap_endian { @@ -772,7 +775,6 @@ struct regmap_field *devm_regmap_field_alloc(struct device *dev, void devm_regmap_field_free(struct device *dev, struct regmap_field *field); int regmap_field_read(struct regmap_field *field, unsigned int *val); -int regmap_field_write(struct regmap_field *field, unsigned int val); int regmap_field_update_bits_base(struct regmap_field *field, unsigned int mask, unsigned int val, bool *change, bool async, bool force); -- cgit v1.2.3 From 721ed64dda3774c619874866ca4f9a38ae6750af Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Mon, 15 Feb 2016 05:24:33 +0000 Subject: regmap: merge regmap_field_update_bits() into macro This patch merges regmap_field_update_bits() into macro by using regmap_field_update_bits_base(). Signed-off-by: Kuninori Morimoto Signed-off-by: Mark Brown --- drivers/base/regmap/regmap.c | 20 -------------------- include/linux/regmap.h | 4 ++-- 2 files changed, 2 insertions(+), 22 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index 228dce237658..606c9b53526b 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -1716,26 +1716,6 @@ int regmap_field_update_bits_base(struct regmap_field *field, } EXPORT_SYMBOL_GPL(regmap_field_update_bits_base); -/** - * regmap_field_update_bits(): Perform a read/modify/write cycle - * on the register field - * - * @field: Register field to write to - * @mask: Bitmask to change - * @val: Value to be written - * - * A value of zero will be returned on success, a negative errno will - * be returned in error cases. - */ -int regmap_field_update_bits(struct regmap_field *field, unsigned int mask, unsigned int val) -{ - mask = (mask << field->shift) & field->mask; - - return regmap_update_bits(field->regmap, field->reg, - mask, val << field->shift); -} -EXPORT_SYMBOL_GPL(regmap_field_update_bits); - /** * regmap_fields_write(): Write a value to a single register field with port ID * diff --git a/include/linux/regmap.h b/include/linux/regmap.h index 13e9ebdea1c7..e525beeaa2c6 100644 --- a/include/linux/regmap.h +++ b/include/linux/regmap.h @@ -76,6 +76,8 @@ struct reg_sequence { #define regmap_field_write(field, val) \ regmap_field_update_bits_base(field, ~0, val, NULL, false, false) +#define regmap_field_update_bits(field, mask, val)\ + regmap_field_update_bits_base(field, mask, val, NULL, false, false) #ifdef CONFIG_REGMAP @@ -778,8 +780,6 @@ int regmap_field_read(struct regmap_field *field, unsigned int *val); int regmap_field_update_bits_base(struct regmap_field *field, unsigned int mask, unsigned int val, bool *change, bool async, bool force); -int regmap_field_update_bits(struct regmap_field *field, - unsigned int mask, unsigned int val); int regmap_fields_write(struct regmap_field *field, unsigned int id, unsigned int val); int regmap_fields_force_write(struct regmap_field *field, unsigned int id, -- cgit v1.2.3 From e126edec184ea3049cc1f8b652c6eeb06aa65fda Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Mon, 15 Feb 2016 05:24:51 +0000 Subject: regmap: add regmap_fields_update_bits_base() This patch adds new regmap_fields_update_bits_base() which is using regmap_update_bits_base(). Current regmap_fields_xxx() can be merged into it by macro. Signed-off-by: Kuninori Morimoto Signed-off-by: Mark Brown --- drivers/base/regmap/regmap.c | 32 ++++++++++++++++++++++++++++++++ include/linux/regmap.h | 12 ++++++++++++ 2 files changed, 44 insertions(+) (limited to 'include/linux') diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index 606c9b53526b..0c7773fadd48 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -1716,6 +1716,38 @@ int regmap_field_update_bits_base(struct regmap_field *field, } EXPORT_SYMBOL_GPL(regmap_field_update_bits_base); +/** + * regmap_fields_update_bits_base(): + * Perform a read/modify/write cycle on the register field + * with change, async, force option + * + * @field: Register field to write to + * @id: port ID + * @mask: Bitmask to change + * @val: Value to be written + * @change: Boolean indicating if a write was done + * @async: Boolean indicating asynchronously + * @force: Boolean indicating use force update + * + * A value of zero will be returned on success, a negative errno will + * be returned in error cases. + */ +int regmap_fields_update_bits_base(struct regmap_field *field, unsigned int id, + unsigned int mask, unsigned int val, + bool *change, bool async, bool force) +{ + if (id >= field->id_size) + return -EINVAL; + + mask = (mask << field->shift) & field->mask; + + return regmap_update_bits_base(field->regmap, + field->reg + (field->id_offset * id), + mask, val << field->shift, + change, async, force); +} +EXPORT_SYMBOL_GPL(regmap_fields_update_bits_base); + /** * regmap_fields_write(): Write a value to a single register field with port ID * diff --git a/include/linux/regmap.h b/include/linux/regmap.h index e525beeaa2c6..2735a3df7eab 100644 --- a/include/linux/regmap.h +++ b/include/linux/regmap.h @@ -788,6 +788,9 @@ int regmap_fields_read(struct regmap_field *field, unsigned int id, unsigned int *val); int regmap_fields_update_bits(struct regmap_field *field, unsigned int id, unsigned int mask, unsigned int val); +int regmap_fields_update_bits_base(struct regmap_field *field, unsigned int id, + unsigned int mask, unsigned int val, + bool *change, bool async, bool force); /** * Description of an IRQ for the generic regmap irq_chip. @@ -966,6 +969,15 @@ static inline int regmap_field_update_bits_base(struct regmap_field *field, return -EINVAL; } +static inline int regmap_fields_update_bits_base(struct regmap_field *field, + unsigned int id, + unsigned int mask, unsigned int val, + bool *change, bool async, bool force) +{ + WARN_ONCE(1, "regmap API is disabled"); + return -EINVAL; +} + static inline int regmap_get_val_bytes(struct regmap *map) { WARN_ONCE(1, "regmap API is disabled"); -- cgit v1.2.3 From bbf2c46f46e23a496337e143cd012c013c6c7910 Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Mon, 15 Feb 2016 05:25:15 +0000 Subject: regmap: merge regmap_fields_write() into macro This patch merges regmap_fields_write() into macro by using regmap_fields_update_bits_base(). Signed-off-by: Kuninori Morimoto Signed-off-by: Mark Brown --- drivers/base/regmap/regmap.c | 22 ---------------------- include/linux/regmap.h | 5 +++-- 2 files changed, 3 insertions(+), 24 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index 0c7773fadd48..4b14745249ba 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -1748,28 +1748,6 @@ int regmap_fields_update_bits_base(struct regmap_field *field, unsigned int id, } EXPORT_SYMBOL_GPL(regmap_fields_update_bits_base); -/** - * regmap_fields_write(): Write a value to a single register field with port ID - * - * @field: Register field to write to - * @id: port ID - * @val: Value to be written - * - * A value of zero will be returned on success, a negative errno will - * be returned in error cases. - */ -int regmap_fields_write(struct regmap_field *field, unsigned int id, - unsigned int val) -{ - if (id >= field->id_size) - return -EINVAL; - - return regmap_update_bits(field->regmap, - field->reg + (field->id_offset * id), - field->mask, val << field->shift); -} -EXPORT_SYMBOL_GPL(regmap_fields_write); - int regmap_fields_force_write(struct regmap_field *field, unsigned int id, unsigned int val) { diff --git a/include/linux/regmap.h b/include/linux/regmap.h index 2735a3df7eab..5f438a4df5e6 100644 --- a/include/linux/regmap.h +++ b/include/linux/regmap.h @@ -79,6 +79,9 @@ struct reg_sequence { #define regmap_field_update_bits(field, mask, val)\ regmap_field_update_bits_base(field, mask, val, NULL, false, false) +#define regmap_fields_write(field, id, val) \ + regmap_fields_update_bits_base(field, id, ~0, val, NULL, false, false) + #ifdef CONFIG_REGMAP enum regmap_endian { @@ -780,8 +783,6 @@ int regmap_field_read(struct regmap_field *field, unsigned int *val); int regmap_field_update_bits_base(struct regmap_field *field, unsigned int mask, unsigned int val, bool *change, bool async, bool force); -int regmap_fields_write(struct regmap_field *field, unsigned int id, - unsigned int val); int regmap_fields_force_write(struct regmap_field *field, unsigned int id, unsigned int val); int regmap_fields_read(struct regmap_field *field, unsigned int id, -- cgit v1.2.3 From 48138609135fc9c363f034596e14bff5dbf9f33f Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Mon, 15 Feb 2016 05:25:32 +0000 Subject: regmap: merge regmap_fields_update_bits() into macro This patch merges regmap_fields_update_bits() into macro by using regmap_field_update_bits_base(). Signed-off-by: Kuninori Morimoto Signed-off-by: Mark Brown --- drivers/base/regmap/regmap.c | 26 -------------------------- include/linux/regmap.h | 4 ++-- 2 files changed, 2 insertions(+), 28 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index 4b14745249ba..79d7f51019d7 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -1760,32 +1760,6 @@ int regmap_fields_force_write(struct regmap_field *field, unsigned int id, } EXPORT_SYMBOL_GPL(regmap_fields_force_write); -/** - * regmap_fields_update_bits(): Perform a read/modify/write cycle - * on the register field - * - * @field: Register field to write to - * @id: port ID - * @mask: Bitmask to change - * @val: Value to be written - * - * A value of zero will be returned on success, a negative errno will - * be returned in error cases. - */ -int regmap_fields_update_bits(struct regmap_field *field, unsigned int id, - unsigned int mask, unsigned int val) -{ - if (id >= field->id_size) - return -EINVAL; - - mask = (mask << field->shift) & field->mask; - - return regmap_update_bits(field->regmap, - field->reg + (field->id_offset * id), - mask, val << field->shift); -} -EXPORT_SYMBOL_GPL(regmap_fields_update_bits); - /* * regmap_bulk_write(): Write multiple registers to the device * diff --git a/include/linux/regmap.h b/include/linux/regmap.h index 5f438a4df5e6..7d3d498fd3e8 100644 --- a/include/linux/regmap.h +++ b/include/linux/regmap.h @@ -81,6 +81,8 @@ struct reg_sequence { #define regmap_fields_write(field, id, val) \ regmap_fields_update_bits_base(field, id, ~0, val, NULL, false, false) +#define regmap_fields_update_bits(field, id, mask, val)\ + regmap_fields_update_bits_base(field, id, mask, val, NULL, false, false) #ifdef CONFIG_REGMAP @@ -787,8 +789,6 @@ int regmap_fields_force_write(struct regmap_field *field, unsigned int id, unsigned int val); int regmap_fields_read(struct regmap_field *field, unsigned int id, unsigned int *val); -int regmap_fields_update_bits(struct regmap_field *field, unsigned int id, - unsigned int mask, unsigned int val); int regmap_fields_update_bits_base(struct regmap_field *field, unsigned int id, unsigned int mask, unsigned int val, bool *change, bool async, bool force); -- cgit v1.2.3 From d6dde63e90cf0d34ae61e885a5eb254c59eb3358 Mon Sep 17 00:00:00 2001 From: Andreas Irestål Date: Tue, 16 Feb 2016 13:56:42 +0100 Subject: ASoC: adau17x1: Correct typos in file headers MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Signed-off-by: Andreas Irestål Signed-off-by: Mark Brown --- include/linux/platform_data/adau17x1.h | 2 +- sound/soc/codecs/adau1761-i2c.c | 2 +- sound/soc/codecs/adau1761-spi.c | 2 +- sound/soc/codecs/adau1761.c | 2 +- sound/soc/codecs/adau1781.c | 2 +- 5 files changed, 5 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/platform_data/adau17x1.h b/include/linux/platform_data/adau17x1.h index a81766cae230..9db1b905df24 100644 --- a/include/linux/platform_data/adau17x1.h +++ b/include/linux/platform_data/adau17x1.h @@ -1,5 +1,5 @@ /* - * Driver for ADAU1761/ADAU1461/ADAU1761/ADAU1961/ADAU1781/ADAU1781 codecs + * Driver for ADAU1361/ADAU1461/ADAU1761/ADAU1961/ADAU1381/ADAU1781 codecs * * Copyright 2011-2014 Analog Devices Inc. * Author: Lars-Peter Clausen diff --git a/sound/soc/codecs/adau1761-i2c.c b/sound/soc/codecs/adau1761-i2c.c index 348ccb17d3cc..a1b12671c307 100644 --- a/sound/soc/codecs/adau1761-i2c.c +++ b/sound/soc/codecs/adau1761-i2c.c @@ -1,5 +1,5 @@ /* - * Driver for ADAU1761/ADAU1461/ADAU1761/ADAU1961 codec + * Driver for ADAU1361/ADAU1461/ADAU1761/ADAU1961 codec * * Copyright 2014 Analog Devices Inc. * Author: Lars-Peter Clausen diff --git a/sound/soc/codecs/adau1761-spi.c b/sound/soc/codecs/adau1761-spi.c index 8bc1fbd25fcc..188334928c23 100644 --- a/sound/soc/codecs/adau1761-spi.c +++ b/sound/soc/codecs/adau1761-spi.c @@ -1,5 +1,5 @@ /* - * Driver for ADAU1761/ADAU1461/ADAU1761/ADAU1961 codec + * Driver for ADAU1361/ADAU1461/ADAU1761/ADAU1961 codec * * Copyright 2014 Analog Devices Inc. * Author: Lars-Peter Clausen diff --git a/sound/soc/codecs/adau1761.c b/sound/soc/codecs/adau1761.c index e7136b1956a3..b95d29dbd13d 100644 --- a/sound/soc/codecs/adau1761.c +++ b/sound/soc/codecs/adau1761.c @@ -1,5 +1,5 @@ /* - * Driver for ADAU1761/ADAU1461/ADAU1761/ADAU1961 codec + * Driver for ADAU1361/ADAU1461/ADAU1761/ADAU1961 codec * * Copyright 2011-2013 Analog Devices Inc. * Author: Lars-Peter Clausen diff --git a/sound/soc/codecs/adau1781.c b/sound/soc/codecs/adau1781.c index fde9068550a6..bc1bb56dae63 100644 --- a/sound/soc/codecs/adau1781.c +++ b/sound/soc/codecs/adau1781.c @@ -1,5 +1,5 @@ /* - * Driver for ADAU1781/ADAU1781 codec + * Driver for ADAU1381/ADAU1781 codec * * Copyright 2011-2013 Analog Devices Inc. * Author: Lars-Peter Clausen -- cgit v1.2.3 From 4ee34ea3a12396f35b26d90a094c75db95080baa Mon Sep 17 00:00:00 2001 From: Harvey Hunt Date: Wed, 24 Feb 2016 15:16:43 +0000 Subject: libata: Align ata_device's id on a cacheline The id buffer in ata_device is a DMA target, but it isn't explicitly cacheline aligned. Due to this, adjacent fields can be overwritten with stale data from memory on non coherent architectures. As a result, the kernel is sometimes unable to communicate with an ATA device. Fix this by ensuring that the id buffer is cacheline aligned. This issue is similar to that fixed by Commit 84bda12af31f ("libata: align ap->sector_buf"). Signed-off-by: Harvey Hunt Cc: linux-kernel@vger.kernel.org Cc: # 2.6.18 Signed-off-by: Tejun Heo --- include/linux/libata.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/libata.h b/include/linux/libata.h index bec2abbd7ab2..2c4ebef79d0c 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -720,7 +720,7 @@ struct ata_device { union { u16 id[ATA_ID_WORDS]; /* IDENTIFY xxx DEVICE data */ u32 gscr[SATA_PMP_GSCR_DWORDS]; /* PMP GSCR block */ - }; + } ____cacheline_aligned; /* DEVSLP Timing Variables from Identify Device Data Log */ u8 devslp_timing[ATA_LOG_DEVSLP_SIZE]; -- cgit v1.2.3 From 489061bba6c655a2f98d39be17df118c0fd09d57 Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Mon, 15 Feb 2016 05:25:54 +0000 Subject: regmap: add regmap_field_force_xxx() macros Signed-off-by: Kuninori Morimoto Signed-off-by: Mark Brown --- include/linux/regmap.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/linux') diff --git a/include/linux/regmap.h b/include/linux/regmap.h index 7d3d498fd3e8..d36ea89adc50 100644 --- a/include/linux/regmap.h +++ b/include/linux/regmap.h @@ -76,8 +76,12 @@ struct reg_sequence { #define regmap_field_write(field, val) \ regmap_field_update_bits_base(field, ~0, val, NULL, false, false) +#define regmap_field_force_write(field, val) \ + regmap_field_update_bits_base(field, ~0, val, NULL, false, true) #define regmap_field_update_bits(field, mask, val)\ regmap_field_update_bits_base(field, mask, val, NULL, false, false) +#define regmap_field_force_update_bits(field, mask, val) \ + regmap_field_update_bits_base(field, mask, val, NULL, false, true) #define regmap_fields_write(field, id, val) \ regmap_fields_update_bits_base(field, id, ~0, val, NULL, false, false) -- cgit v1.2.3 From e6ef243fa4660f3206137bd5f3e69b13a9b7c28a Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Mon, 15 Feb 2016 05:26:14 +0000 Subject: regmap: add regmap_fields_force_xxx() macros Signed-off-by: Kuninori Morimoto Signed-off-by: Mark Brown --- drivers/base/regmap/regmap.c | 12 ------------ include/linux/regmap.h | 6 ++++-- 2 files changed, 4 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index 79d7f51019d7..c7d4a636778d 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -1748,18 +1748,6 @@ int regmap_fields_update_bits_base(struct regmap_field *field, unsigned int id, } EXPORT_SYMBOL_GPL(regmap_fields_update_bits_base); -int regmap_fields_force_write(struct regmap_field *field, unsigned int id, - unsigned int val) -{ - if (id >= field->id_size) - return -EINVAL; - - return regmap_write_bits(field->regmap, - field->reg + (field->id_offset * id), - field->mask, val << field->shift); -} -EXPORT_SYMBOL_GPL(regmap_fields_force_write); - /* * regmap_bulk_write(): Write multiple registers to the device * diff --git a/include/linux/regmap.h b/include/linux/regmap.h index d36ea89adc50..e0960b3ff290 100644 --- a/include/linux/regmap.h +++ b/include/linux/regmap.h @@ -85,8 +85,12 @@ struct reg_sequence { #define regmap_fields_write(field, id, val) \ regmap_fields_update_bits_base(field, id, ~0, val, NULL, false, false) +#define regmap_fields_force_write(field, id, val) \ + regmap_fields_update_bits_base(field, id, ~0, val, NULL, false, true) #define regmap_fields_update_bits(field, id, mask, val)\ regmap_fields_update_bits_base(field, id, mask, val, NULL, false, false) +#define regmap_fields_force_update_bits(field, id, mask, val) \ + regmap_fields_update_bits_base(field, id, mask, val, NULL, false, true) #ifdef CONFIG_REGMAP @@ -789,8 +793,6 @@ int regmap_field_read(struct regmap_field *field, unsigned int *val); int regmap_field_update_bits_base(struct regmap_field *field, unsigned int mask, unsigned int val, bool *change, bool async, bool force); -int regmap_fields_force_write(struct regmap_field *field, unsigned int id, - unsigned int val); int regmap_fields_read(struct regmap_field *field, unsigned int id, unsigned int *val); int regmap_fields_update_bits_base(struct regmap_field *field, unsigned int id, -- cgit v1.2.3 From a528aca7f359f4b0b1d72ae406097e491a5ba9ea Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 29 Feb 2016 12:12:46 -0500 Subject: use ->d_seq to get coherency between ->d_inode and ->d_flags Games with ordering and barriers are way too brittle. Just bump ->d_seq before and after updating ->d_inode and ->d_flags type bits, so that verifying ->d_seq would guarantee they are coherent. Cc: stable@vger.kernel.org # v3.13+ Signed-off-by: Al Viro --- fs/dcache.c | 20 +++++--------------- include/linux/dcache.h | 4 +--- 2 files changed, 6 insertions(+), 18 deletions(-) (limited to 'include/linux') diff --git a/fs/dcache.c b/fs/dcache.c index 92d5140de851..2398f9f94337 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -269,9 +269,6 @@ static inline int dname_external(const struct dentry *dentry) return dentry->d_name.name != dentry->d_iname; } -/* - * Make sure other CPUs see the inode attached before the type is set. - */ static inline void __d_set_inode_and_type(struct dentry *dentry, struct inode *inode, unsigned type_flags) @@ -279,28 +276,18 @@ static inline void __d_set_inode_and_type(struct dentry *dentry, unsigned flags; dentry->d_inode = inode; - smp_wmb(); flags = READ_ONCE(dentry->d_flags); flags &= ~(DCACHE_ENTRY_TYPE | DCACHE_FALLTHRU); flags |= type_flags; WRITE_ONCE(dentry->d_flags, flags); } -/* - * Ideally, we want to make sure that other CPUs see the flags cleared before - * the inode is detached, but this is really a violation of RCU principles - * since the ordering suggests we should always set inode before flags. - * - * We should instead replace or discard the entire dentry - but that sucks - * performancewise on mass deletion/rename. - */ static inline void __d_clear_type_and_inode(struct dentry *dentry) { unsigned flags = READ_ONCE(dentry->d_flags); flags &= ~(DCACHE_ENTRY_TYPE | DCACHE_FALLTHRU); WRITE_ONCE(dentry->d_flags, flags); - smp_wmb(); dentry->d_inode = NULL; } @@ -370,9 +357,11 @@ static void dentry_unlink_inode(struct dentry * dentry) __releases(dentry->d_inode->i_lock) { struct inode *inode = dentry->d_inode; + + raw_write_seqcount_begin(&dentry->d_seq); __d_clear_type_and_inode(dentry); hlist_del_init(&dentry->d_u.d_alias); - dentry_rcuwalk_invalidate(dentry); + raw_write_seqcount_end(&dentry->d_seq); spin_unlock(&dentry->d_lock); spin_unlock(&inode->i_lock); if (!inode->i_nlink) @@ -1758,8 +1747,9 @@ static void __d_instantiate(struct dentry *dentry, struct inode *inode) spin_lock(&dentry->d_lock); if (inode) hlist_add_head(&dentry->d_u.d_alias, &inode->i_dentry); + raw_write_seqcount_begin(&dentry->d_seq); __d_set_inode_and_type(dentry, inode, add_flags); - dentry_rcuwalk_invalidate(dentry); + raw_write_seqcount_end(&dentry->d_seq); spin_unlock(&dentry->d_lock); fsnotify_d_instantiate(dentry, inode); } diff --git a/include/linux/dcache.h b/include/linux/dcache.h index 7781ce110503..c4b5f4b3f8f8 100644 --- a/include/linux/dcache.h +++ b/include/linux/dcache.h @@ -409,9 +409,7 @@ static inline bool d_mountpoint(const struct dentry *dentry) */ static inline unsigned __d_entry_type(const struct dentry *dentry) { - unsigned type = READ_ONCE(dentry->d_flags); - smp_rmb(); - return type & DCACHE_ENTRY_TYPE; + return dentry->d_flags & DCACHE_ENTRY_TYPE; } static inline bool d_is_miss(const struct dentry *dentry) -- cgit v1.2.3 From 7bcd79ac50d9d83350a835bdb91c04ac9e098412 Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Fri, 26 Feb 2016 23:40:50 +0800 Subject: block: bio: introduce helpers to get the 1st and last bvec The bio passed to bio_will_gap() may be fast cloned from upper layer(dm, md, bcache, fs, ...), or from bio splitting in block core. Unfortunately bio_will_gap() just figures out the last bvec via 'bi_io_vec[prev->bi_vcnt - 1]' directly, and this way is obviously wrong. This patch introduces two helpers for getting the first and last bvec of one bio for fixing the issue. Cc: stable@vger.kernel.org Reported-by: Sagi Grimberg Reviewed-by: Sagi Grimberg Reviewed-by: Christoph Hellwig Signed-off-by: Ming Lei Signed-off-by: Jens Axboe --- include/linux/bio.h | 37 +++++++++++++++++++++++++++++++++++++ 1 file changed, 37 insertions(+) (limited to 'include/linux') diff --git a/include/linux/bio.h b/include/linux/bio.h index 5349e6816cbb..cb6888824108 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -310,6 +310,43 @@ static inline void bio_clear_flag(struct bio *bio, unsigned int bit) bio->bi_flags &= ~(1U << bit); } +static inline void bio_get_first_bvec(struct bio *bio, struct bio_vec *bv) +{ + *bv = bio_iovec(bio); +} + +static inline void bio_get_last_bvec(struct bio *bio, struct bio_vec *bv) +{ + struct bvec_iter iter = bio->bi_iter; + int idx; + + if (!bio_flagged(bio, BIO_CLONED)) { + *bv = bio->bi_io_vec[bio->bi_vcnt - 1]; + return; + } + + if (unlikely(!bio_multiple_segments(bio))) { + *bv = bio_iovec(bio); + return; + } + + bio_advance_iter(bio, &iter, iter.bi_size); + + if (!iter.bi_bvec_done) + idx = iter.bi_idx - 1; + else /* in the middle of bvec */ + idx = iter.bi_idx; + + *bv = bio->bi_io_vec[idx]; + + /* + * iter.bi_bvec_done records actual length of the last bvec + * if this bio ends in the middle of one io vector + */ + if (iter.bi_bvec_done) + bv->bv_len = iter.bi_bvec_done; +} + enum bip_flags { BIP_BLOCK_INTEGRITY = 1 << 0, /* block layer owns integrity data */ BIP_MAPPED_INTEGRITY = 1 << 1, /* ref tag has been remapped */ -- cgit v1.2.3 From e0af29171aa8912e1ca95023b75ef336cd70d661 Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Fri, 26 Feb 2016 23:40:51 +0800 Subject: block: check virt boundary in bio_will_gap() In the following patch, the way for figuring out the last bvec will be changed with a bit cost introduced, so return immediately if the queue doesn't have virt boundary limit. Actually most of devices have not this limit. Reviewed-by: Sagi Grimberg Reviewed-by: Christoph Hellwig Signed-off-by: Ming Lei Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 4571ef1a12a9..cd06a41581b3 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -1372,6 +1372,13 @@ static inline void put_dev_sector(Sector p) page_cache_release(p.v); } +static inline bool __bvec_gap_to_prev(struct request_queue *q, + struct bio_vec *bprv, unsigned int offset) +{ + return offset || + ((bprv->bv_offset + bprv->bv_len) & queue_virt_boundary(q)); +} + /* * Check if adding a bio_vec after bprv with offset would create a gap in * the SG list. Most drivers don't care about this, but some do. @@ -1381,18 +1388,17 @@ static inline bool bvec_gap_to_prev(struct request_queue *q, { if (!queue_virt_boundary(q)) return false; - return offset || - ((bprv->bv_offset + bprv->bv_len) & queue_virt_boundary(q)); + return __bvec_gap_to_prev(q, bprv, offset); } static inline bool bio_will_gap(struct request_queue *q, struct bio *prev, struct bio *next) { - if (!bio_has_data(prev)) + if (!bio_has_data(prev) || !queue_virt_boundary(q)) return false; - return bvec_gap_to_prev(q, &prev->bi_io_vec[prev->bi_vcnt - 1], - next->bi_io_vec[0].bv_offset); + return __bvec_gap_to_prev(q, &prev->bi_io_vec[prev->bi_vcnt - 1], + next->bi_io_vec[0].bv_offset); } static inline bool req_gap_back_merge(struct request *req, struct bio *bio) -- cgit v1.2.3 From 25e71a99f10e444cd00bb2ebccb11e1c9fb672b1 Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Fri, 26 Feb 2016 23:40:52 +0800 Subject: block: get the 1st and last bvec via helpers This patch applies the two introduced helpers to figure out the 1st and last bvec, and fixes the original way after bio splitting. Cc: stable@vger.kernel.org Reported-by: Sagi Grimberg Reviewed-by: Sagi Grimberg Reviewed-by: Christoph Hellwig Signed-off-by: Ming Lei Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index cd06a41581b3..d7f6bca707ef 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -1394,11 +1394,16 @@ static inline bool bvec_gap_to_prev(struct request_queue *q, static inline bool bio_will_gap(struct request_queue *q, struct bio *prev, struct bio *next) { - if (!bio_has_data(prev) || !queue_virt_boundary(q)) - return false; + if (bio_has_data(prev) && queue_virt_boundary(q)) { + struct bio_vec pb, nb; + + bio_get_last_bvec(prev, &pb); + bio_get_first_bvec(next, &nb); - return __bvec_gap_to_prev(q, &prev->bi_io_vec[prev->bi_vcnt - 1], - next->bi_io_vec[0].bv_offset); + return __bvec_gap_to_prev(q, &pb, nb.bv_offset); + } + + return false; } static inline bool req_gap_back_merge(struct request *req, struct bio *bio) -- cgit v1.2.3 From a1a0e23e49037c23ea84bc8cc146a03584d13577 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 29 Feb 2016 18:28:53 -0500 Subject: writeback: flush inode cgroup wb switches instead of pinning super_block If cgroup writeback is in use, inodes can be scheduled for asynchronous wb switching. Before 5ff8eaac1636 ("writeback: keep superblock pinned during cgroup writeback association switches"), this could race with umount leading to super_block being destroyed while inodes are pinned for wb switching. 5ff8eaac1636 fixed it by bumping s_active while wb switches are in flight; however, this allowed in-flight wb switches to make umounts asynchronous when the userland expected synchronosity - e.g. fsck immediately following umount may fail because the device is still busy. This patch removes the problematic super_block pinning and instead makes generic_shutdown_super() flush in-flight wb switches. wb switches are now executed on a dedicated isw_wq so that they can be flushed and isw_nr_in_flight keeps track of the number of in-flight wb switches so that flushing can be avoided in most cases. v2: Move cgroup_writeback_umount() further below and add MS_ACTIVE check in inode_switch_wbs() as Jan an Al suggested. Signed-off-by: Tejun Heo Reported-by: Tahsin Erdogan Cc: Jan Kara Cc: Al Viro Link: http://lkml.kernel.org/g/CAAeU0aNCq7LGODvVGRU-oU_o-6enii5ey0p1c26D1ZzYwkDc5A@mail.gmail.com Fixes: 5ff8eaac1636 ("writeback: keep superblock pinned during cgroup writeback association switches") Cc: stable@vger.kernel.org #v4.5 Reviewed-by: Jan Kara Tested-by: Tahsin Erdogan Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 54 +++++++++++++++++++++++++++++++++++------------ fs/super.c | 1 + include/linux/writeback.h | 5 +++++ 3 files changed, 47 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 1f76d8950a57..5c46ed9f3e14 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -223,6 +223,9 @@ static void wb_wait_for_completion(struct backing_dev_info *bdi, #define WB_FRN_HIST_MAX_SLOTS (WB_FRN_HIST_THR_SLOTS / 2 + 1) /* one round can affect upto 5 slots */ +static atomic_t isw_nr_in_flight = ATOMIC_INIT(0); +static struct workqueue_struct *isw_wq; + void __inode_attach_wb(struct inode *inode, struct page *page) { struct backing_dev_info *bdi = inode_to_bdi(inode); @@ -317,7 +320,6 @@ static void inode_switch_wbs_work_fn(struct work_struct *work) struct inode_switch_wbs_context *isw = container_of(work, struct inode_switch_wbs_context, work); struct inode *inode = isw->inode; - struct super_block *sb = inode->i_sb; struct address_space *mapping = inode->i_mapping; struct bdi_writeback *old_wb = inode->i_wb; struct bdi_writeback *new_wb = isw->new_wb; @@ -424,8 +426,9 @@ skip_switch: wb_put(new_wb); iput(inode); - deactivate_super(sb); kfree(isw); + + atomic_dec(&isw_nr_in_flight); } static void inode_switch_wbs_rcu_fn(struct rcu_head *rcu_head) @@ -435,7 +438,7 @@ static void inode_switch_wbs_rcu_fn(struct rcu_head *rcu_head) /* needs to grab bh-unsafe locks, bounce to work item */ INIT_WORK(&isw->work, inode_switch_wbs_work_fn); - schedule_work(&isw->work); + queue_work(isw_wq, &isw->work); } /** @@ -471,20 +474,20 @@ static void inode_switch_wbs(struct inode *inode, int new_wb_id) /* while holding I_WB_SWITCH, no one else can update the association */ spin_lock(&inode->i_lock); - - if (inode->i_state & (I_WB_SWITCH | I_FREEING) || - inode_to_wb(inode) == isw->new_wb) - goto out_unlock; - - if (!atomic_inc_not_zero(&inode->i_sb->s_active)) - goto out_unlock; - + if (!(inode->i_sb->s_flags & MS_ACTIVE) || + inode->i_state & (I_WB_SWITCH | I_FREEING) || + inode_to_wb(inode) == isw->new_wb) { + spin_unlock(&inode->i_lock); + goto out_free; + } inode->i_state |= I_WB_SWITCH; spin_unlock(&inode->i_lock); ihold(inode); isw->inode = inode; + atomic_inc(&isw_nr_in_flight); + /* * In addition to synchronizing among switchers, I_WB_SWITCH tells * the RCU protected stat update paths to grab the mapping's @@ -494,8 +497,6 @@ static void inode_switch_wbs(struct inode *inode, int new_wb_id) call_rcu(&isw->rcu_head, inode_switch_wbs_rcu_fn); return; -out_unlock: - spin_unlock(&inode->i_lock); out_free: if (isw->new_wb) wb_put(isw->new_wb); @@ -847,6 +848,33 @@ restart: wb_put(last_wb); } +/** + * cgroup_writeback_umount - flush inode wb switches for umount + * + * This function is called when a super_block is about to be destroyed and + * flushes in-flight inode wb switches. An inode wb switch goes through + * RCU and then workqueue, so the two need to be flushed in order to ensure + * that all previously scheduled switches are finished. As wb switches are + * rare occurrences and synchronize_rcu() can take a while, perform + * flushing iff wb switches are in flight. + */ +void cgroup_writeback_umount(void) +{ + if (atomic_read(&isw_nr_in_flight)) { + synchronize_rcu(); + flush_workqueue(isw_wq); + } +} + +static int __init cgroup_writeback_init(void) +{ + isw_wq = alloc_workqueue("inode_switch_wbs", 0, 0); + if (!isw_wq) + return -ENOMEM; + return 0; +} +fs_initcall(cgroup_writeback_init); + #else /* CONFIG_CGROUP_WRITEBACK */ static struct bdi_writeback * diff --git a/fs/super.c b/fs/super.c index 1182af8fd5ff..74914b1bae70 100644 --- a/fs/super.c +++ b/fs/super.c @@ -415,6 +415,7 @@ void generic_shutdown_super(struct super_block *sb) sb->s_flags &= ~MS_ACTIVE; fsnotify_unmount_inodes(sb); + cgroup_writeback_umount(); evict_inodes(sb); diff --git a/include/linux/writeback.h b/include/linux/writeback.h index b333c945e571..d0b5ca5d4e08 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -198,6 +198,7 @@ void wbc_attach_and_unlock_inode(struct writeback_control *wbc, void wbc_detach_inode(struct writeback_control *wbc); void wbc_account_io(struct writeback_control *wbc, struct page *page, size_t bytes); +void cgroup_writeback_umount(void); /** * inode_attach_wb - associate an inode with its wb @@ -301,6 +302,10 @@ static inline void wbc_account_io(struct writeback_control *wbc, { } +static inline void cgroup_writeback_umount(void) +{ +} + #endif /* CONFIG_CGROUP_WRITEBACK */ /* -- cgit v1.2.3 From f21018427cb007a0894c36ad702990ab639cbbb4 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 3 Mar 2016 14:43:45 -0700 Subject: block: fix blk_rq_get_max_sectors for driver private requests Driver private request types should not get the artifical cap for the FS requests. This is important to use the full device capabilities for internal command or NVMe pass through commands. Signed-off-by: Christoph Hellwig Reported-by: Jeff Lien Tested-by: Jeff Lien Reviewed-by: Keith Busch Updated by me to use an explicit check for the one command type that does support extended checking, instead of relying on the ordering of the enum command values - as suggested by Keith. Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index d7f6bca707ef..413c84fbc4ed 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -895,7 +895,7 @@ static inline unsigned int blk_rq_get_max_sectors(struct request *rq) { struct request_queue *q = rq->q; - if (unlikely(rq->cmd_type == REQ_TYPE_BLOCK_PC)) + if (unlikely(rq->cmd_type != REQ_TYPE_FS)) return q->limits.max_hw_sectors; if (!q->limits.chunk_sectors || (rq->cmd_flags & REQ_DISCARD)) -- cgit v1.2.3 From e57cbaf0eb006eaa207395f3bfd7ce52c1b5539c Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Red Hat)" Date: Thu, 3 Mar 2016 17:18:20 -0500 Subject: tracing: Do not have 'comm' filter override event 'comm' field Commit 9f61668073a8d "tracing: Allow triggers to filter for CPU ids and process names" added a 'comm' filter that will filter events based on the current tasks struct 'comm'. But this now hides the ability to filter events that have a 'comm' field too. For example, sched_migrate_task trace event. That has a 'comm' field of the task to be migrated. echo 'comm == "bash"' > events/sched_migrate_task/filter will now filter all sched_migrate_task events for tasks named "bash" that migrates other tasks (in interrupt context), instead of seeing when "bash" itself gets migrated. This fix requires a couple of changes. 1) Change the look up order for filter predicates to look at the events fields before looking at the generic filters. 2) Instead of basing the filter function off of the "comm" name, have the generic "comm" filter have its own filter_type (FILTER_COMM). Test against the type instead of the name to assign the filter function. 3) Add a new "COMM" filter that works just like "comm" but will filter based on the current task, even if the trace event contains a "comm" field. Do the same for "cpu" field, adding a FILTER_CPU and a filter "CPU". Cc: stable@vger.kernel.org # v4.3+ Fixes: 9f61668073a8d "tracing: Allow triggers to filter for CPU ids and process names" Reported-by: Matt Fleming Signed-off-by: Steven Rostedt --- include/linux/trace_events.h | 2 ++ kernel/trace/trace_events.c | 14 ++++++++------ kernel/trace/trace_events_filter.c | 13 +++++++------ 3 files changed, 17 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/trace_events.h b/include/linux/trace_events.h index 429fdfc3baf5..925730bc9fc1 100644 --- a/include/linux/trace_events.h +++ b/include/linux/trace_events.h @@ -568,6 +568,8 @@ enum { FILTER_DYN_STRING, FILTER_PTR_STRING, FILTER_TRACE_FN, + FILTER_COMM, + FILTER_CPU, }; extern int trace_event_raw_init(struct trace_event_call *call); diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index ab09829d3b97..05ddc0820771 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -97,16 +97,16 @@ trace_find_event_field(struct trace_event_call *call, char *name) struct ftrace_event_field *field; struct list_head *head; - field = __find_event_field(&ftrace_generic_fields, name); + head = trace_get_fields(call); + field = __find_event_field(head, name); if (field) return field; - field = __find_event_field(&ftrace_common_fields, name); + field = __find_event_field(&ftrace_generic_fields, name); if (field) return field; - head = trace_get_fields(call); - return __find_event_field(head, name); + return __find_event_field(&ftrace_common_fields, name); } static int __trace_define_field(struct list_head *head, const char *type, @@ -171,8 +171,10 @@ static int trace_define_generic_fields(void) { int ret; - __generic_field(int, cpu, FILTER_OTHER); - __generic_field(char *, comm, FILTER_PTR_STRING); + __generic_field(int, CPU, FILTER_CPU); + __generic_field(int, cpu, FILTER_CPU); + __generic_field(char *, COMM, FILTER_COMM); + __generic_field(char *, comm, FILTER_COMM); return ret; } diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index f93a219b18da..6816302542b2 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -1043,13 +1043,14 @@ static int init_pred(struct filter_parse_state *ps, return -EINVAL; } - if (is_string_field(field)) { + if (field->filter_type == FILTER_COMM) { + filter_build_regex(pred); + fn = filter_pred_comm; + pred->regex.field_len = TASK_COMM_LEN; + } else if (is_string_field(field)) { filter_build_regex(pred); - if (!strcmp(field->name, "comm")) { - fn = filter_pred_comm; - pred->regex.field_len = TASK_COMM_LEN; - } else if (field->filter_type == FILTER_STATIC_STRING) { + if (field->filter_type == FILTER_STATIC_STRING) { fn = filter_pred_string; pred->regex.field_len = field->size; } else if (field->filter_type == FILTER_DYN_STRING) @@ -1072,7 +1073,7 @@ static int init_pred(struct filter_parse_state *ps, } pred->val = val; - if (!strcmp(field->name, "cpu")) + if (field->filter_type == FILTER_CPU) fn = filter_pred_cpu; else fn = select_comparison_fn(pred->op, field->size, -- cgit v1.2.3 From 5ea5c5e0a7f70b256417d3b6e36bd9851504babd Mon Sep 17 00:00:00 2001 From: "Yan, Zheng" Date: Sun, 14 Feb 2016 18:06:41 +0800 Subject: ceph: initial CEPH_FEATURE_FS_FILE_LAYOUT_V2 support Add support for the format change of MClientReply/MclientCaps. Also add code that denies access to inodes with pool_ns layouts. Signed-off-by: Yan, Zheng Reviewed-by: Sage Weil --- fs/ceph/addr.c | 4 ++++ fs/ceph/caps.c | 27 ++++++++++++++++++++++++--- fs/ceph/inode.c | 2 ++ fs/ceph/mds_client.c | 16 ++++++++++++++++ fs/ceph/mds_client.h | 1 + fs/ceph/super.h | 1 + include/linux/ceph/ceph_features.h | 1 + 7 files changed, 49 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index c22213789090..19adeb0ef82a 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -1756,6 +1756,10 @@ int ceph_pool_perm_check(struct ceph_inode_info *ci, int need) u32 pool; int ret, flags; + /* does not support pool namespace yet */ + if (ci->i_pool_ns_len) + return -EIO; + if (ceph_test_mount_opt(ceph_inode_to_client(&ci->vfs_inode), NOPOOLPERM)) return 0; diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index cdbf8cf3d52c..6fe0ad26a7df 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -2753,7 +2753,8 @@ static void handle_cap_grant(struct ceph_mds_client *mdsc, void *inline_data, int inline_len, struct ceph_buffer *xattr_buf, struct ceph_mds_session *session, - struct ceph_cap *cap, int issued) + struct ceph_cap *cap, int issued, + u32 pool_ns_len) __releases(ci->i_ceph_lock) __releases(mdsc->snap_rwsem) { @@ -2873,6 +2874,8 @@ static void handle_cap_grant(struct ceph_mds_client *mdsc, if (newcaps & (CEPH_CAP_ANY_FILE_RD | CEPH_CAP_ANY_FILE_WR)) { /* file layout may have changed */ ci->i_layout = grant->layout; + ci->i_pool_ns_len = pool_ns_len; + /* size/truncate_seq? */ queue_trunc = ceph_fill_file_size(inode, issued, le32_to_cpu(grant->truncate_seq), @@ -3411,6 +3414,7 @@ void ceph_handle_caps(struct ceph_mds_session *session, u32 inline_len = 0; void *snaptrace; size_t snaptrace_len; + u32 pool_ns_len = 0; void *p, *end; dout("handle_caps from mds%d\n", mds); @@ -3463,6 +3467,21 @@ void ceph_handle_caps(struct ceph_mds_session *session, p += inline_len; } + if (le16_to_cpu(msg->hdr.version) >= 8) { + u64 flush_tid; + u32 caller_uid, caller_gid; + u32 osd_epoch_barrier; + /* version >= 5 */ + ceph_decode_32_safe(&p, end, osd_epoch_barrier, bad); + /* version >= 6 */ + ceph_decode_64_safe(&p, end, flush_tid, bad); + /* version >= 7 */ + ceph_decode_32_safe(&p, end, caller_uid, bad); + ceph_decode_32_safe(&p, end, caller_gid, bad); + /* version >= 8 */ + ceph_decode_32_safe(&p, end, pool_ns_len, bad); + } + /* lookup ino */ inode = ceph_find_inode(sb, vino); ci = ceph_inode(inode); @@ -3518,7 +3537,8 @@ void ceph_handle_caps(struct ceph_mds_session *session, &cap, &issued); handle_cap_grant(mdsc, inode, h, inline_version, inline_data, inline_len, - msg->middle, session, cap, issued); + msg->middle, session, cap, issued, + pool_ns_len); if (realm) ceph_put_snap_realm(mdsc, realm); goto done_unlocked; @@ -3542,7 +3562,8 @@ void ceph_handle_caps(struct ceph_mds_session *session, issued |= __ceph_caps_dirty(ci); handle_cap_grant(mdsc, inode, h, inline_version, inline_data, inline_len, - msg->middle, session, cap, issued); + msg->middle, session, cap, issued, + pool_ns_len); goto done_unlocked; case CEPH_CAP_OP_FLUSH_ACK: diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index fb4ba2e4e2a5..5849b88bbed3 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -396,6 +396,7 @@ struct inode *ceph_alloc_inode(struct super_block *sb) ci->i_symlink = NULL; memset(&ci->i_dir_layout, 0, sizeof(ci->i_dir_layout)); + ci->i_pool_ns_len = 0; ci->i_fragtree = RB_ROOT; mutex_init(&ci->i_fragtree_mutex); @@ -756,6 +757,7 @@ static int fill_inode(struct inode *inode, struct page *locked_page, if (ci->i_layout.fl_pg_pool != info->layout.fl_pg_pool) ci->i_ceph_flags &= ~CEPH_I_POOL_PERM; ci->i_layout = info->layout; + ci->i_pool_ns_len = iinfo->pool_ns_len; queue_trunc = ceph_fill_file_size(inode, issued, le32_to_cpu(info->truncate_seq), diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index e7b130a637f9..911d64d865f1 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -100,6 +100,14 @@ static int parse_reply_info_in(void **p, void *end, } else info->inline_version = CEPH_INLINE_NONE; + if (features & CEPH_FEATURE_FS_FILE_LAYOUT_V2) { + ceph_decode_32_safe(p, end, info->pool_ns_len, bad); + ceph_decode_need(p, end, info->pool_ns_len, bad); + *p += info->pool_ns_len; + } else { + info->pool_ns_len = 0; + } + return 0; bad: return err; @@ -2298,6 +2306,14 @@ int ceph_mdsc_do_request(struct ceph_mds_client *mdsc, ceph_get_cap_refs(ceph_inode(req->r_old_dentry_dir), CEPH_CAP_PIN); + /* deny access to directories with pool_ns layouts */ + if (req->r_inode && S_ISDIR(req->r_inode->i_mode) && + ceph_inode(req->r_inode)->i_pool_ns_len) + return -EIO; + if (req->r_locked_dir && + ceph_inode(req->r_locked_dir)->i_pool_ns_len) + return -EIO; + /* issue */ mutex_lock(&mdsc->mutex); __register_request(mdsc, req, dir); diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h index ccf11ef0ca87..37712ccffcc6 100644 --- a/fs/ceph/mds_client.h +++ b/fs/ceph/mds_client.h @@ -44,6 +44,7 @@ struct ceph_mds_reply_info_in { u64 inline_version; u32 inline_len; char *inline_data; + u32 pool_ns_len; }; /* diff --git a/fs/ceph/super.h b/fs/ceph/super.h index 75b7d125ce66..9c458eb52245 100644 --- a/fs/ceph/super.h +++ b/fs/ceph/super.h @@ -287,6 +287,7 @@ struct ceph_inode_info { struct ceph_dir_layout i_dir_layout; struct ceph_file_layout i_layout; + size_t i_pool_ns_len; char *i_symlink; /* for dirs */ diff --git a/include/linux/ceph/ceph_features.h b/include/linux/ceph/ceph_features.h index c1ef6f14e7be..15151f3c4120 100644 --- a/include/linux/ceph/ceph_features.h +++ b/include/linux/ceph/ceph_features.h @@ -75,6 +75,7 @@ #define CEPH_FEATURE_CRUSH_TUNABLES5 (1ULL<<58) /* chooseleaf stable mode */ // duplicated since it was introduced at the same time as CEPH_FEATURE_CRUSH_TUNABLES5 #define CEPH_FEATURE_NEW_OSDOPREPLY_ENCODING (1ULL<<58) /* New, v7 encoding */ +#define CEPH_FEATURE_FS_FILE_LAYOUT_V2 (1ULL<<58) /* file_layout_t */ /* * The introduction of CEPH_FEATURE_OSD_SNAPMAPPER caused the feature -- cgit v1.2.3