From c393eaa85349941badf3ce5f087400dbaf3bbe02 Mon Sep 17 00:00:00 2001 From: Haifeng Xu Date: Fri, 26 Jul 2024 11:05:25 +0800 Subject: fs: don't flush in-flight wb switches for superblocks without cgroup writeback When deactivating any type of superblock, it had to wait for the in-flight wb switches to be completed. wb switches are executed in inode_switch_wbs_work_fn() which needs to acquire the wb_switch_rwsem and races against sync_inodes_sb(). If there are too much dirty data in the superblock, the waiting time may increase significantly. For superblocks without cgroup writeback such as tmpfs, they have nothing to do with the wb swithes, so the flushing can be avoided. Signed-off-by: Haifeng Xu Link: https://lore.kernel.org/r/20240726030525.180330-1-haifeng.xu@shopee.com Reviewed-by: Jan Kara Suggested-by: Jan Kara Signed-off-by: Christian Brauner --- include/linux/writeback.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/writeback.h b/include/linux/writeback.h index 1a54676d843a..56b85841ae4c 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -217,7 +217,7 @@ void wbc_account_cgroup_owner(struct writeback_control *wbc, struct page *page, size_t bytes); int cgroup_writeback_by_id(u64 bdi_id, int memcg_id, enum wb_reason reason, struct wb_completion *done); -void cgroup_writeback_umount(void); +void cgroup_writeback_umount(struct super_block *sb); bool cleanup_offline_cgwb(struct bdi_writeback *wb); /** @@ -324,7 +324,7 @@ static inline void wbc_account_cgroup_owner(struct writeback_control *wbc, { } -static inline void cgroup_writeback_umount(void) +static inline void cgroup_writeback_umount(struct super_block *sb) { } -- cgit v1.2.3 From c01a5d89e5c8abe638107be2a4ea9e4c7fcdd7f6 Mon Sep 17 00:00:00 2001 From: Wang Long Date: Fri, 2 Aug 2024 17:19:01 +0800 Subject: percpu-rwsem: remove the unused parameter 'read' In the function percpu_rwsem_release, the parameter `read` is unused, so remove it. Signed-off-by: Wang Long Link: https://lore.kernel.org/r/20240802091901.2546797-1-w@laoqinren.net Reviewed-by: Jan Kara Signed-off-by: Christian Brauner --- include/linux/fs.h | 2 +- include/linux/percpu-rwsem.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index fb0426f349fc..c0286d479c9d 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1683,7 +1683,7 @@ static inline bool __sb_start_write_trylock(struct super_block *sb, int level) #define __sb_writers_acquired(sb, lev) \ percpu_rwsem_acquire(&(sb)->s_writers.rw_sem[(lev)-1], 1, _THIS_IP_) #define __sb_writers_release(sb, lev) \ - percpu_rwsem_release(&(sb)->s_writers.rw_sem[(lev)-1], 1, _THIS_IP_) + percpu_rwsem_release(&(sb)->s_writers.rw_sem[(lev)-1], _THIS_IP_) /** * __sb_write_started - check if sb freeze level is held diff --git a/include/linux/percpu-rwsem.h b/include/linux/percpu-rwsem.h index 36b942b67b7d..c012df33a9f0 100644 --- a/include/linux/percpu-rwsem.h +++ b/include/linux/percpu-rwsem.h @@ -145,7 +145,7 @@ extern void percpu_free_rwsem(struct percpu_rw_semaphore *); #define percpu_rwsem_assert_held(sem) lockdep_assert_held(sem) static inline void percpu_rwsem_release(struct percpu_rw_semaphore *sem, - bool read, unsigned long ip) + unsigned long ip) { lock_release(&sem->dep_map, ip); } -- cgit v1.2.3 From 087adb4f0f91ee330446a70af899e6a996e5cc13 Mon Sep 17 00:00:00 2001 From: Mateusz Guzik Date: Tue, 6 Aug 2024 19:28:46 +0200 Subject: vfs: dodge smp_mb in break_lease and break_deleg in the common case These inlines show up in the fast path (e.g., in do_dentry_open()) and induce said full barrier regarding i_flctx access when in most cases the pointer is NULL. The pointer can be safely checked before issuing the barrier, dodging it in most cases as a result. It is plausible the consume fence would be sufficient, but I don't want to go audit all callers regarding what they before calling here. Signed-off-by: Mateusz Guzik Link: https://lore.kernel.org/r/20240806172846.886570-1-mjguzik@gmail.com Reviewed-by: Jeff Layton Signed-off-by: Christian Brauner --- include/linux/filelock.h | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/filelock.h b/include/linux/filelock.h index daee999d05f3..bb44224c6676 100644 --- a/include/linux/filelock.h +++ b/include/linux/filelock.h @@ -420,28 +420,38 @@ static inline int locks_lock_file_wait(struct file *filp, struct file_lock *fl) #ifdef CONFIG_FILE_LOCKING static inline int break_lease(struct inode *inode, unsigned int mode) { + struct file_lock_context *flctx; + /* * Since this check is lockless, we must ensure that any refcounts * taken are done before checking i_flctx->flc_lease. Otherwise, we * could end up racing with tasks trying to set a new lease on this * file. */ + flctx = READ_ONCE(inode->i_flctx); + if (!flctx) + return 0; smp_mb(); - if (inode->i_flctx && !list_empty_careful(&inode->i_flctx->flc_lease)) + if (!list_empty_careful(&flctx->flc_lease)) return __break_lease(inode, mode, FL_LEASE); return 0; } static inline int break_deleg(struct inode *inode, unsigned int mode) { + struct file_lock_context *flctx; + /* * Since this check is lockless, we must ensure that any refcounts * taken are done before checking i_flctx->flc_lease. Otherwise, we * could end up racing with tasks trying to set a new lease on this * file. */ + flctx = READ_ONCE(inode->i_flctx); + if (!flctx) + return 0; smp_mb(); - if (inode->i_flctx && !list_empty_careful(&inode->i_flctx->flc_lease)) + if (!list_empty_careful(&flctx->flc_lease)) return __break_lease(inode, mode, FL_DELEG); return 0; } -- cgit v1.2.3 From 8447d848e1dc6d42d1dcd00f133d7715fc732c47 Mon Sep 17 00:00:00 2001 From: Mateusz Guzik Date: Sat, 10 Aug 2024 08:47:53 +0200 Subject: vfs: only read fops once in fops_get/put In do_dentry_open() the usage is: f->f_op = fops_get(inode->i_fop); In generated asm the compiler emits 2 reads from inode->i_fop instead of just one. This popped up due to false-sharing where loads from that offset end up bouncing a cacheline during parallel open. While this is going to be fixed, the spurious load does not need to be there. This makes do_dentry_open() go down from 1177 to 1154 bytes. fops_put() is patched to maintain some consistency. No functional changes. Reviewed-by: Josef Bacik Signed-off-by: Mateusz Guzik Link: https://lore.kernel.org/r/20240810064753.1211441-1-mjguzik@gmail.com Signed-off-by: Christian Brauner --- include/linux/fs.h | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index c0286d479c9d..696c620f90e0 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2553,10 +2553,17 @@ struct super_block *sget(struct file_system_type *type, struct super_block *sget_dev(struct fs_context *fc, dev_t dev); /* Alas, no aliases. Too much hassle with bringing module.h everywhere */ -#define fops_get(fops) \ - (((fops) && try_module_get((fops)->owner) ? (fops) : NULL)) -#define fops_put(fops) \ - do { if (fops) module_put((fops)->owner); } while(0) +#define fops_get(fops) ({ \ + const struct file_operations *_fops = (fops); \ + (((_fops) && try_module_get((_fops)->owner) ? (_fops) : NULL)); \ +}) + +#define fops_put(fops) ({ \ + const struct file_operations *_fops = (fops); \ + if (_fops) \ + module_put((_fops)->owner); \ +}) + /* * This one is to be used *ONLY* from ->open() instances. * fops must be non-NULL, pinned down *and* module dependencies -- cgit v1.2.3 From 641bb4394f405cba498b100b44541ffc0aed5be1 Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Fri, 9 Aug 2024 12:38:56 +0200 Subject: fs: move FMODE_UNSIGNED_OFFSET to fop_flags This is another flag that is statically set and doesn't need to use up an FMODE_* bit. Move it to ->fop_flags and free up another FMODE_* bit. (1) mem_open() used from proc_mem_operations (2) adi_open() used from adi_fops (3) drm_open_helper(): (3.1) accel_open() used from DRM_ACCEL_FOPS (3.2) drm_open() used from (3.2.1) amdgpu_driver_kms_fops (3.2.2) psb_gem_fops (3.2.3) i915_driver_fops (3.2.4) nouveau_driver_fops (3.2.5) panthor_drm_driver_fops (3.2.6) radeon_driver_kms_fops (3.2.7) tegra_drm_fops (3.2.8) vmwgfx_driver_fops (3.2.9) xe_driver_fops (3.2.10) DRM_GEM_FOPS (3.2.11) DEFINE_DRM_GEM_DMA_FOPS (4) struct memdev sets fmode flags based on type of device opened. For devices using struct mem_fops unsigned offset is used. Mark all these file operations as FOP_UNSIGNED_OFFSET and add asserts into the open helper to ensure that the flag is always set. Link: https://lore.kernel.org/r/20240809-work-fop_unsigned-v1-1-658e054d893e@kernel.org Reviewed-by: Jeff Layton Reviewed-by: Jan Kara Signed-off-by: Christian Brauner --- include/linux/fs.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 696c620f90e0..0145bda465ff 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -146,8 +146,7 @@ typedef int (dio_iodone_t)(struct kiocb *iocb, loff_t offset, /* Expect random access pattern */ #define FMODE_RANDOM ((__force fmode_t)(1 << 12)) -/* File is huge (eg. /dev/mem): treat loff_t as unsigned */ -#define FMODE_UNSIGNED_OFFSET ((__force fmode_t)(1 << 13)) +/* FMODE_* bit 13 */ /* File is opened with O_PATH; almost nothing can be done with it */ #define FMODE_PATH ((__force fmode_t)(1 << 14)) @@ -2073,6 +2072,8 @@ struct file_operations { #define FOP_DIO_PARALLEL_WRITE ((__force fop_flags_t)(1 << 3)) /* Contains huge pages */ #define FOP_HUGE_PAGES ((__force fop_flags_t)(1 << 4)) +/* Treat loff_t as unsigned (e.g., /dev/mem) */ +#define FOP_UNSIGNED_OFFSET ((__force fop_flags_t)(1 << 5)) /* Wrap a directory iterator that needs exclusive inode access */ int wrap_directory_iterator(struct file *, struct dir_context *, -- cgit v1.2.3 From 1c48d441468c425e878d81c5c3e7ff8a9a594cc0 Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Fri, 16 Aug 2024 16:35:52 +0200 Subject: inode: remove __I_DIO_WAKEUP Afaict, we can just rely on inode->i_dio_count for waiting instead of this awkward indirection through __I_DIO_WAKEUP. This survives LTP dio and xfstests dio tests. Link: https://lore.kernel.org/r/20240816-vfs-misc-dio-v1-1-80fe21a2c710@kernel.org Reviewed-by: Josef Bacik Reviewed-by: Jeff Layton Signed-off-by: Christian Brauner --- include/linux/fs.h | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 0145bda465ff..8f8d274929d7 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2373,8 +2373,6 @@ static inline void kiocb_clone(struct kiocb *kiocb, struct kiocb *kiocb_src, * * I_REFERENCED Marks the inode as recently references on the LRU list. * - * I_DIO_WAKEUP Never set. Only used as a key for wait_on_bit(). - * * I_WB_SWITCH Cgroup bdi_writeback switching in progress. Used to * synchronize competing switching instances and to tell * wb stat updates to grab the i_pages lock. See @@ -2409,8 +2407,6 @@ static inline void kiocb_clone(struct kiocb *kiocb, struct kiocb *kiocb_src, #define __I_SYNC 7 #define I_SYNC (1 << __I_SYNC) #define I_REFERENCED (1 << 8) -#define __I_DIO_WAKEUP 9 -#define I_DIO_WAKEUP (1 << __I_DIO_WAKEUP) #define I_LINKABLE (1 << 10) #define I_DIRTY_TIME (1 << 11) #define I_WB_SWITCH (1 << 13) @@ -3227,7 +3223,9 @@ static inline ssize_t blockdev_direct_IO(struct kiocb *iocb, } #endif +bool inode_dio_finished(const struct inode *inode); void inode_dio_wait(struct inode *inode); +void inode_dio_wait_interruptible(struct inode *inode); /** * inode_dio_begin - signal start of a direct I/O requests @@ -3251,7 +3249,7 @@ static inline void inode_dio_begin(struct inode *inode) static inline void inode_dio_end(struct inode *inode) { if (atomic_dec_and_test(&inode->i_dio_count)) - wake_up_bit(&inode->i_state, __I_DIO_WAKEUP); + wake_up_var(&inode->i_dio_count); } extern void inode_set_flags(struct inode *inode, unsigned int flags, -- cgit v1.2.3 From 029c3f27fe84c5fd29d49a99cc5176433bf12209 Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Thu, 22 Aug 2024 11:28:38 +0200 Subject: fs: remove unused path_put_init() This helper has been unused for a while now. Link: https://lore.kernel.org/r/20240822-bewuchs-werktag-46672b3c0606@brauner Reviewed-by: Jan Kara Signed-off-by: Christian Brauner --- include/linux/path.h | 6 ------ 1 file changed, 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/path.h b/include/linux/path.h index ca073e70decd..7ea389dc764b 100644 --- a/include/linux/path.h +++ b/include/linux/path.h @@ -18,12 +18,6 @@ static inline int path_equal(const struct path *path1, const struct path *path2) return path1->mnt == path2->mnt && path1->dentry == path2->dentry; } -static inline void path_put_init(struct path *path) -{ - path_put(path); - *path = (struct path) { }; -} - /* * Cleanup macro for use with __free(path_put). Avoids dereference and * copying @path unlike DEFINE_FREE(). path_put() will handle the empty -- cgit v1.2.3 From 71ff58ce3428b2471efae5b00594e892b285c97c Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Thu, 22 Aug 2024 11:30:58 +0200 Subject: fs: s/__u32/u32/ for s_fsnotify_mask The underscore variants are for uapi whereas the non-underscore variants are for in-kernel consumers. Link: https://lore.kernel.org/r/20240822-anwerben-nutzung-1cd6c82a565f@brauner Reviewed-by: Jan Kara Signed-off-by: Christian Brauner --- include/linux/fs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 8f8d274929d7..cda1f2545ea0 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1266,7 +1266,7 @@ struct super_block { time64_t s_time_min; time64_t s_time_max; #ifdef CONFIG_FSNOTIFY - __u32 s_fsnotify_mask; + u32 s_fsnotify_mask; struct fsnotify_sb_info *s_fsnotify_info; #endif -- cgit v1.2.3 From da18ecbf0fb6fb73e6f9273e7aa15610f148ce17 Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Fri, 23 Aug 2024 14:47:35 +0200 Subject: fs: add i_state helpers The i_state member is an unsigned long so that it can be used with the wait bit infrastructure which expects unsigned long. This wastes 4 bytes which we're unlikely to ever use. Switch to using the var event wait mechanism using the address of the bit. Thanks to Linus for the address idea. Link: https://lore.kernel.org/r/20240823-work-i_state-v3-1-5cd5fd207a57@kernel.org Reviewed-by: Josef Bacik Reviewed-by: Jan Kara Signed-off-by: Christian Brauner --- include/linux/fs.h | 15 +++++++++++++++ 1 file changed, 15 insertions(+) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index cda1f2545ea0..d35de348d2ec 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -744,6 +744,21 @@ struct inode { void *i_private; /* fs or device private pointer */ } __randomize_layout; +/* + * Get bit address from inode->i_state to use with wait_var_event() + * infrastructre. + */ +#define inode_state_wait_address(inode, bit) ((char *)&(inode)->i_state + (bit)) + +struct wait_queue_head *inode_bit_waitqueue(struct wait_bit_queue_entry *wqe, + struct inode *inode, u32 bit); + +static inline void inode_wake_up_bit(struct inode *inode, u32 bit) +{ + /* Caller is responsible for correct memory barriers. */ + wake_up_var(inode_state_wait_address(inode, bit)); +} + struct timespec64 timestamp_truncate(struct timespec64 t, struct inode *inode); static inline unsigned int i_blocksize(const struct inode *node) -- cgit v1.2.3 From 2ed634c96ed1d6e4ee0497be176f9980866e81cb Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Fri, 23 Aug 2024 14:47:36 +0200 Subject: fs: reorder i_state bits so that we can use the first bits to derive unique addresses from i_state. Link: https://lore.kernel.org/r/20240823-work-i_state-v3-2-5cd5fd207a57@kernel.org Reviewed-by: Josef Bacik Reviewed-by: Jeff Layton Reviewed-by: Jan Kara Signed-off-by: Christian Brauner --- include/linux/fs.h | 38 +++++++++++++++++++++----------------- 1 file changed, 21 insertions(+), 17 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index d35de348d2ec..a868c9823c10 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2410,28 +2410,32 @@ static inline void kiocb_clone(struct kiocb *kiocb, struct kiocb *kiocb_src, * i_count. * * Q: What is the difference between I_WILL_FREE and I_FREEING? + * + * __I_{SYNC,NEW,LRU_ISOLATING} are used to derive unique addresses to wait + * upon. There's one free address left. */ -#define I_DIRTY_SYNC (1 << 0) -#define I_DIRTY_DATASYNC (1 << 1) -#define I_DIRTY_PAGES (1 << 2) -#define __I_NEW 3 +#define __I_NEW 0 #define I_NEW (1 << __I_NEW) -#define I_WILL_FREE (1 << 4) -#define I_FREEING (1 << 5) -#define I_CLEAR (1 << 6) -#define __I_SYNC 7 +#define __I_SYNC 1 #define I_SYNC (1 << __I_SYNC) -#define I_REFERENCED (1 << 8) +#define __I_LRU_ISOLATING 2 +#define I_LRU_ISOLATING (1 << __I_LRU_ISOLATING) + +#define I_DIRTY_SYNC (1 << 3) +#define I_DIRTY_DATASYNC (1 << 4) +#define I_DIRTY_PAGES (1 << 5) +#define I_WILL_FREE (1 << 6) +#define I_FREEING (1 << 7) +#define I_CLEAR (1 << 8) +#define I_REFERENCED (1 << 9) #define I_LINKABLE (1 << 10) #define I_DIRTY_TIME (1 << 11) -#define I_WB_SWITCH (1 << 13) -#define I_OVL_INUSE (1 << 14) -#define I_CREATING (1 << 15) -#define I_DONTCACHE (1 << 16) -#define I_SYNC_QUEUED (1 << 17) -#define I_PINNING_NETFS_WB (1 << 18) -#define __I_LRU_ISOLATING 19 -#define I_LRU_ISOLATING (1 << __I_LRU_ISOLATING) +#define I_WB_SWITCH (1 << 12) +#define I_OVL_INUSE (1 << 13) +#define I_CREATING (1 << 14) +#define I_DONTCACHE (1 << 15) +#define I_SYNC_QUEUED (1 << 16) +#define I_PINNING_NETFS_WB (1 << 17) #define I_DIRTY_INODE (I_DIRTY_SYNC | I_DIRTY_DATASYNC) #define I_DIRTY (I_DIRTY_INODE | I_DIRTY_PAGES) -- cgit v1.2.3 From 0fe340a98b584a31b07c664dc5ff0c923730581e Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Fri, 23 Aug 2024 14:47:38 +0200 Subject: inode: port __I_NEW to var event Port the __I_NEW mechanism to use the new var event mechanism. Link: https://lore.kernel.org/r/20240823-work-i_state-v3-4-5cd5fd207a57@kernel.org Reviewed-by: Josef Bacik Reviewed-by: Jan Kara Signed-off-by: Christian Brauner --- include/linux/writeback.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/writeback.h b/include/linux/writeback.h index 56b85841ae4c..8f651bb0a1a5 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -200,7 +200,8 @@ void inode_io_list_del(struct inode *inode); /* writeback.h requires fs.h; it, too, is not included from here. */ static inline void wait_on_inode(struct inode *inode) { - wait_on_bit(&inode->i_state, __I_NEW, TASK_UNINTERRUPTIBLE); + wait_var_event(inode_state_wait_address(inode, __I_NEW), + !(READ_ONCE(inode->i_state) & I_NEW)); } #ifdef CONFIG_CGROUP_WRITEBACK -- cgit v1.2.3 From 2b111edbe0a9c441605be5cfb73001dc98ec686f Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Fri, 23 Aug 2024 14:47:40 +0200 Subject: inode: make i_state a u32 Now that we use the wait var event mechanism make i_state a u32 and free up 4 bytes. This means we currently have two 4 byte holes in struct inode which we can pack. Link: https://lore.kernel.org/r/20240823-work-i_state-v3-6-5cd5fd207a57@kernel.org Reviewed-by: Josef Bacik Reviewed-by: Jan Kara Signed-off-by: Christian Brauner --- include/linux/fs.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index a868c9823c10..a1ef8c65d828 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -681,7 +681,8 @@ struct inode { #endif /* Misc */ - unsigned long i_state; + u32 i_state; + /* 32-bit hole */ struct rw_semaphore i_rwsem; unsigned long dirtied_when; /* jiffies of first dirtying */ -- cgit v1.2.3 From 5c40e050e6ac0218af7c520095729d440cc87e6b Mon Sep 17 00:00:00 2001 From: Michal Hocko Date: Thu, 29 Aug 2024 15:06:40 +0200 Subject: fs: drop GFP_NOFAIL mode from alloc_page_buffers There is only one called of alloc_page_buffers and it doesn't require __GFP_NOFAIL so drop this allocation mode. Signed-off-by: Michal Hocko Link: https://lore.kernel.org/r/20240829130640.1397970-1-mhocko@kernel.org Acked-by: Song Liu Reviewed-by: Jan Kara Signed-off-by: Christian Brauner --- include/linux/buffer_head.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h index 14acf1bbe0ce..7e903457967a 100644 --- a/include/linux/buffer_head.h +++ b/include/linux/buffer_head.h @@ -199,8 +199,7 @@ void folio_set_bh(struct buffer_head *bh, struct folio *folio, unsigned long offset); struct buffer_head *folio_alloc_buffers(struct folio *folio, unsigned long size, gfp_t gfp); -struct buffer_head *alloc_page_buffers(struct page *page, unsigned long size, - bool retry); +struct buffer_head *alloc_page_buffers(struct page *page, unsigned long size); struct buffer_head *create_empty_buffers(struct folio *folio, unsigned long blocksize, unsigned long b_state); void end_buffer_read_sync(struct buffer_head *bh, int uptodate); -- cgit v1.2.3 From 4356d575ef0f39a3e8e0ce0c40d84ce900ac3b61 Mon Sep 17 00:00:00 2001 From: Aleksa Sarai Date: Wed, 28 Aug 2024 20:19:43 +1000 Subject: fhandle: expose u64 mount id to name_to_handle_at(2) Now that we provide a unique 64-bit mount ID interface in statx(2), we can now provide a race-free way for name_to_handle_at(2) to provide a file handle and corresponding mount without needing to worry about racing with /proc/mountinfo parsing or having to open a file just to do statx(2). While this is not necessary if you are using AT_EMPTY_PATH and don't care about an extra statx(2) call, users that pass full paths into name_to_handle_at(2) need to know which mount the file handle comes from (to make sure they don't try to open_by_handle_at a file handle from a different filesystem) and switching to AT_EMPTY_PATH would require allocating a file for every name_to_handle_at(2) call, turning err = name_to_handle_at(-EBADF, "/foo/bar/baz", &handle, &mntid, AT_HANDLE_MNT_ID_UNIQUE); into int fd = openat(-EBADF, "/foo/bar/baz", O_PATH | O_CLOEXEC); err1 = name_to_handle_at(fd, "", &handle, &unused_mntid, AT_EMPTY_PATH); err2 = statx(fd, "", AT_EMPTY_PATH, STATX_MNT_ID_UNIQUE, &statxbuf); mntid = statxbuf.stx_mnt_id; close(fd); Reviewed-by: Jeff Layton Signed-off-by: Aleksa Sarai Link: https://lore.kernel.org/r/20240828-exportfs-u64-mount-id-v3-2-10c2c4c16708@cyphar.com Reviewed-by: Jan Kara Reviewed-by: Josef Bacik Signed-off-by: Christian Brauner --- include/linux/syscalls.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 4bcf6754738d..5758104921e6 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -870,7 +870,7 @@ asmlinkage long sys_fanotify_mark(int fanotify_fd, unsigned int flags, #endif asmlinkage long sys_name_to_handle_at(int dfd, const char __user *name, struct file_handle __user *handle, - int __user *mnt_id, int flag); + void __user *mnt_id, int flag); asmlinkage long sys_open_by_handle_at(int mountdirfd, struct file_handle __user *handle, int flags); -- cgit v1.2.3 From 2077006d4725c82c6e9612cec3a6c140921b067f Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Tue, 10 Sep 2024 10:16:39 +0200 Subject: uidgid: make sure we fit into one cacheline When I expanded uidgid mappings I intended for a struct uid_gid_map to fit into a single cacheline on x86 as they tend to be pretty performance sensitive (idmapped mounts etc). But a 4 byte hole was added that brought it over 64 bytes. Fix that by adding the static extent array and the extent counter into a substruct. C's type punning for unions guarantees that we can access ->nr_extents even if the last written to member wasn't within the same object. This is also what we rely on in struct_group() and friends. This of course relies on non-strict aliasing which we don't do. 99) If the member used to read the contents of a union object is not the same as the member last used to store a value in the object, the appropriate part of the object representation of the value is reinterpreted as an object representation in the new type as described in 6.2.6 (a process sometimes called "type punning"). Link: https://www.open-std.org/jtc1/sc22/wg14/www/docs/n2310.pdf Link: https://lore.kernel.org/r/20240910-work-uid_gid_map-v1-1-e6bc761363ed@kernel.org Reviewed-by: Jeff Layton Signed-off-by: Christian Brauner --- include/linux/user_namespace.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/user_namespace.h b/include/linux/user_namespace.h index 6030a8235617..3625096d5f85 100644 --- a/include/linux/user_namespace.h +++ b/include/linux/user_namespace.h @@ -21,9 +21,11 @@ struct uid_gid_extent { }; struct uid_gid_map { /* 64 bytes -- 1 cache line */ - u32 nr_extents; union { - struct uid_gid_extent extent[UID_GID_MAP_MAX_BASE_EXTENTS]; + struct { + struct uid_gid_extent extent[UID_GID_MAP_MAX_BASE_EXTENTS]; + u32 nr_extents; + }; struct { struct uid_gid_extent *forward; struct uid_gid_extent *reverse; -- cgit v1.2.3