From 44cdc1d952e3f7aa9944c1bbf38fc23f49885017 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 27 Sep 2020 11:18:30 -0400 Subject: convert ->f_ep_links/->fllink to hlist we don't care about the order of elements there Signed-off-by: Al Viro --- include/linux/fs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux/fs.h') diff --git a/include/linux/fs.h b/include/linux/fs.h index 0bd126418bb6..b484ba0e474a 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -946,7 +946,7 @@ struct file { #ifdef CONFIG_EPOLL /* Used by fs/eventpoll.c to link all the hooks to this file */ - struct list_head f_ep_links; + struct hlist_head f_ep_links; struct list_head f_tfile_llink; #endif /* #ifdef CONFIG_EPOLL */ struct address_space *f_mapping; -- cgit From 319c15174757aaedacc89a6e55c965416f130e64 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 1 Oct 2020 20:45:51 -0400 Subject: epoll: take epitem list out of struct file Move the head of epitem list out of struct file; for epoll ones it's moved into struct eventpoll (->refs there), for non-epoll - into the new object (struct epitem_head). In place of ->f_ep_links we leave a pointer to the list head (->f_ep). ->f_ep is protected by ->f_lock and it's zeroed as soon as the list of epitems becomes empty (that can happen only in ep_remove() by now). The list of files for reverse path check is *not* going through struct file now - it's a single-linked list going through epitem_head instances. It's terminated by ERR_PTR(-1) (== EP_UNACTIVE_POINTER), so the elements of list can be distinguished by head->next != NULL. epitem_head instances are allocated at ep_insert() time (by attach_epitem()) and freed either by ep_remove() (if it empties the set of epitems *and* epitem_head does not belong to the reverse path check list) or by clear_tfile_check_list() when the list is emptied (if the set of epitems is empty by that point). Allocations are done from a separate slab - minimal kmalloc() size is too large on some architectures. As the result, we trim struct file _and_ get rid of the games with temporary file references. Locking and barriers are interesting (aren't they always); see unlist_file() and ep_remove() for details. The non-obvious part is that ep_remove() needs to decide if it will be the one to free the damn thing *before* actually storing NULL to head->epitems.first - that's what smp_load_acquire is for in there. unlist_file() lockless path is safe, since we hit it only if we observe NULL in head->epitems.first and whoever had done that store is guaranteed to have observed non-NULL in head->next. IOW, their last access had been the store of NULL into ->epitems.first and we can safely free the sucker. OTOH, we are under rcu_read_lock() and both epitem and epitem->file have their freeing RCU-delayed. So if we see non-NULL ->epitems.first, we can grab ->f_lock (all epitems in there share the same struct file) and safely recheck the emptiness of ->epitems; again, ->next is still non-NULL, so ep_remove() couldn't have freed head yet. ->f_lock serializes us wrt ep_remove(); the rest is trivial. Note that once head->epitems becomes NULL, nothing can get inserted into it - the only remaining reference to head after that point is from the reverse path check list. Signed-off-by: Al Viro --- include/linux/fs.h | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'include/linux/fs.h') diff --git a/include/linux/fs.h b/include/linux/fs.h index b484ba0e474a..993c540e3b77 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -923,7 +923,7 @@ struct file { const struct file_operations *f_op; /* - * Protects f_ep_links, f_flags. + * Protects f_ep, f_flags. * Must not be taken from IRQ context. */ spinlock_t f_lock; @@ -946,8 +946,7 @@ struct file { #ifdef CONFIG_EPOLL /* Used by fs/eventpoll.c to link all the hooks to this file */ - struct hlist_head f_ep_links; - struct list_head f_tfile_llink; + struct hlist_head *f_ep; #endif /* #ifdef CONFIG_EPOLL */ struct address_space *f_mapping; errseq_t f_wb_err; -- cgit From 8a3c84b649b033024d2349f96234b26cbd6083a6 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Tue, 10 Nov 2020 16:50:21 -0800 Subject: vfs: separate __sb_start_write into blocking and non-blocking helpers Break this function into two helpers so that it's obvious that the trylock versions return a value that must be checked, and the blocking versions don't require that. While we're at it, clean up the return type mismatch. Signed-off-by: Darrick J. Wong Reviewed-by: Jan Kara Reviewed-by: Christoph Hellwig --- include/linux/fs.h | 21 +++++++++++---------- 1 file changed, 11 insertions(+), 10 deletions(-) (limited to 'include/linux/fs.h') diff --git a/include/linux/fs.h b/include/linux/fs.h index 0bd126418bb6..305989afd49c 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1581,7 +1581,8 @@ extern struct timespec64 current_time(struct inode *inode); */ void __sb_end_write(struct super_block *sb, int level); -int __sb_start_write(struct super_block *sb, int level, bool wait); +void __sb_start_write(struct super_block *sb, int level); +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_) @@ -1645,12 +1646,12 @@ static inline void sb_end_intwrite(struct super_block *sb) */ static inline void sb_start_write(struct super_block *sb) { - __sb_start_write(sb, SB_FREEZE_WRITE, true); + __sb_start_write(sb, SB_FREEZE_WRITE); } -static inline int sb_start_write_trylock(struct super_block *sb) +static inline bool sb_start_write_trylock(struct super_block *sb) { - return __sb_start_write(sb, SB_FREEZE_WRITE, false); + return __sb_start_write_trylock(sb, SB_FREEZE_WRITE); } /** @@ -1674,7 +1675,7 @@ static inline int sb_start_write_trylock(struct super_block *sb) */ static inline void sb_start_pagefault(struct super_block *sb) { - __sb_start_write(sb, SB_FREEZE_PAGEFAULT, true); + __sb_start_write(sb, SB_FREEZE_PAGEFAULT); } /* @@ -1692,12 +1693,12 @@ static inline void sb_start_pagefault(struct super_block *sb) */ static inline void sb_start_intwrite(struct super_block *sb) { - __sb_start_write(sb, SB_FREEZE_FS, true); + __sb_start_write(sb, SB_FREEZE_FS); } -static inline int sb_start_intwrite_trylock(struct super_block *sb) +static inline bool sb_start_intwrite_trylock(struct super_block *sb) { - return __sb_start_write(sb, SB_FREEZE_FS, false); + return __sb_start_write_trylock(sb, SB_FREEZE_FS); } @@ -2756,14 +2757,14 @@ static inline void file_start_write(struct file *file) { if (!S_ISREG(file_inode(file)->i_mode)) return; - __sb_start_write(file_inode(file)->i_sb, SB_FREEZE_WRITE, true); + sb_start_write(file_inode(file)->i_sb); } static inline bool file_start_write_trylock(struct file *file) { if (!S_ISREG(file_inode(file)->i_mode)) return true; - return __sb_start_write(file_inode(file)->i_sb, SB_FREEZE_WRITE, false); + return sb_start_write_trylock(file_inode(file)->i_sb); } static inline void file_end_write(struct file *file) -- cgit From 9b8523423b23ee3dfd88e32f5b7207be56a4e782 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Tue, 10 Nov 2020 16:50:21 -0800 Subject: vfs: move __sb_{start,end}_write* to fs.h Now that we've straightened out the callers, move these three functions to fs.h since they're fairly trivial. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig Reviewed-by: Jan Kara --- include/linux/fs.h | 21 ++++++++++++++++++--- 1 file changed, 18 insertions(+), 3 deletions(-) (limited to 'include/linux/fs.h') diff --git a/include/linux/fs.h b/include/linux/fs.h index 305989afd49c..6dabd019cab0 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1580,9 +1580,24 @@ extern struct timespec64 current_time(struct inode *inode); * Snapshotting support. */ -void __sb_end_write(struct super_block *sb, int level); -void __sb_start_write(struct super_block *sb, int level); -bool __sb_start_write_trylock(struct super_block *sb, int level); +/* + * These are internal functions, please use sb_start_{write,pagefault,intwrite} + * instead. + */ +static inline void __sb_end_write(struct super_block *sb, int level) +{ + percpu_up_read(sb->s_writers.rw_sem + level-1); +} + +static inline void __sb_start_write(struct super_block *sb, int level) +{ + percpu_down_read(sb->s_writers.rw_sem + level - 1); +} + +static inline bool __sb_start_write_trylock(struct super_block *sb, int level) +{ + return percpu_down_read_trylock(sb->s_writers.rw_sem + level - 1); +} #define __sb_writers_acquired(sb, lev) \ percpu_rwsem_acquire(&(sb)->s_writers.rw_sem[(lev)-1], 1, _THIS_IP_) -- cgit From 60b498852bf219c0bf2b0864c69972840978ca43 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 16 Nov 2020 15:21:18 +0100 Subject: fs: remove get_super_thawed and get_super_exclusive_thawed Just open code the wait in the only caller of both functions. Signed-off-by: Christoph Hellwig Reviewed-by: Jan Kara Reviewed-by: Hannes Reinecke Signed-off-by: Jens Axboe --- include/linux/fs.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'include/linux/fs.h') diff --git a/include/linux/fs.h b/include/linux/fs.h index 8667d0cdc71e..a61df0dd4f19 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1409,7 +1409,7 @@ enum { struct sb_writers { int frozen; /* Is sb frozen? */ - wait_queue_head_t wait_unfrozen; /* for get_super_thawed() */ + wait_queue_head_t wait_unfrozen; /* wait for thaw */ struct percpu_rw_semaphore rw_sem[SB_FREEZE_LEVELS]; }; @@ -3132,8 +3132,6 @@ extern struct file_system_type *get_filesystem(struct file_system_type *fs); extern void put_filesystem(struct file_system_type *fs); extern struct file_system_type *get_fs_type(const char *name); extern struct super_block *get_super(struct block_device *); -extern struct super_block *get_super_thawed(struct block_device *); -extern struct super_block *get_super_exclusive_thawed(struct block_device *bdev); extern struct super_block *get_active_super(struct block_device *bdev); extern void drop_super(struct super_block *sb); extern void drop_super_exclusive(struct super_block *sb); -- cgit From 4e7b5671c6a883d94b5428e1a9c141bbd56cb2a6 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 23 Nov 2020 13:38:40 +0100 Subject: block: remove i_bdev Switch the block device lookup interfaces to directly work with a dev_t so that struct block_device references are only acquired by the blkdev_get variants (and the blk-cgroup special case). This means that we now don't need an extra reference in the inode and can generally simplify handling of struct block_device to keep the lookups contained in the core block layer code. Signed-off-by: Christoph Hellwig Reviewed-by: Jan Kara Reviewed-by: Hannes Reinecke Acked-by: Tejun Heo Acked-by: Coly Li [bcache] Signed-off-by: Jens Axboe --- include/linux/fs.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux/fs.h') diff --git a/include/linux/fs.h b/include/linux/fs.h index a61df0dd4f19..b0b358309657 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -696,7 +696,6 @@ struct inode { struct list_head i_devices; union { struct pipe_inode_info *i_pipe; - struct block_device *i_bdev; struct cdev *i_cdev; char *i_link; unsigned i_dir_seq; -- cgit From 608af703519a58f5a7da4273809211cac27edfb2 Mon Sep 17 00:00:00 2001 From: Daniel Rosenberg Date: Thu, 19 Nov 2020 06:09:02 +0000 Subject: libfs: Add generic function for setting dentry_ops This adds a function to set dentry operations at lookup time that will work for both encrypted filenames and casefolded filenames. A filesystem that supports both features simultaneously can use this function during lookup preparations to set up its dentry operations once fscrypt no longer does that itself. Currently the casefolding dentry operation are always set if the filesystem defines an encoding because the features is toggleable on empty directories. Unlike in the encryption case, the dentry operations used come from the parent. Since we don't know what set of functions we'll eventually need, and cannot change them later, we enable the casefolding operations if the filesystem supports them at all. By splitting out the various cases, we support as few dentry operations as we can get away with, maximizing compatibility with overlayfs, which will not function if a filesystem supports certain dentry_operations. Signed-off-by: Daniel Rosenberg Reviewed-by: Theodore Ts'o Reviewed-by: Eric Biggers Reviewed-by: Gabriel Krisman Bertazi Signed-off-by: Jaegeuk Kim --- include/linux/fs.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux/fs.h') diff --git a/include/linux/fs.h b/include/linux/fs.h index 21cc971fd960..4a25ab4dbd3e 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -3186,6 +3186,7 @@ extern int generic_ci_d_hash(const struct dentry *dentry, struct qstr *str); extern int generic_ci_d_compare(const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name); #endif +extern void generic_set_encrypted_ci_d_ops(struct dentry *dentry); #ifdef CONFIG_MIGRATION extern int buffer_migrate_page(struct address_space *, -- cgit From 88149082bb8ef31b289673669e080ec6a00c2e59 Mon Sep 17 00:00:00 2001 From: Hao Li Date: Tue, 8 Dec 2020 10:08:43 +0800 Subject: fs: Handle I_DONTCACHE in iput_final() instead of generic_drop_inode() If generic_drop_inode() returns true, it means iput_final() can evict this inode regardless of whether it is dirty or not. If we check I_DONTCACHE in generic_drop_inode(), any inode with this bit set will be evicted unconditionally. This is not the desired behavior because I_DONTCACHE only means the inode shouldn't be cached on the LRU list. As for whether we need to evict this inode, this is what generic_drop_inode() should do. This patch corrects the usage of I_DONTCACHE. This patch was proposed in [1]. [1]: https://lore.kernel.org/linux-fsdevel/20200831003407.GE12096@dread.disaster.area/ Fixes: dae2f8ed7992 ("fs: Lift XFS_IDONTCACHE to the VFS layer") Signed-off-by: Hao Li Reviewed-by: Dave Chinner Reviewed-by: Ira Weiny Signed-off-by: Al Viro --- include/linux/fs.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'include/linux/fs.h') diff --git a/include/linux/fs.h b/include/linux/fs.h index 8667d0cdc71e..8bde32cf9711 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2878,8 +2878,7 @@ extern int inode_needs_sync(struct inode *inode); extern int generic_delete_inode(struct inode *inode); static inline int generic_drop_inode(struct inode *inode) { - return !inode->i_nlink || inode_unhashed(inode) || - (inode->i_state & I_DONTCACHE); + return !inode->i_nlink || inode_unhashed(inode); } extern void d_mark_dontcache(struct inode *inode); -- cgit From 52650c8b466bac399aec213c61d74bfe6f7af1a4 Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Mon, 14 Dec 2020 19:05:48 -0800 Subject: mm/gup: remove the vma allocation from gup_longterm_locked() Long ago there wasn't a FOLL_LONGTERM flag so this DAX check was done by post-processing the VMA list. These days it is trivial to just check each VMA to see if it is DAX before processing it inside __get_user_pages() and return failure if a DAX VMA is encountered with FOLL_LONGTERM. Removing the allocation of the VMA list is a significant speed up for many call sites. Add an IS_ENABLED to vma_is_fsdax so that code generation is unchanged when DAX is compiled out. Remove the dummy version of __gup_longterm_locked() as !CONFIG_CMA already makes memalloc_nocma_save(), check_and_migrate_cma_pages(), and memalloc_nocma_restore() into a NOP. Link: https://lkml.kernel.org/r/0-v1-5551df3ed12e+b8-gup_dax_speedup_jgg@nvidia.com Signed-off-by: Jason Gunthorpe Reviewed-by: Ira Weiny Cc: Dan Williams Cc: John Hubbard Cc: Pavel Tatashin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux/fs.h') diff --git a/include/linux/fs.h b/include/linux/fs.h index 8667d0cdc71e..1fcc2b00582b 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -3230,7 +3230,7 @@ static inline bool vma_is_fsdax(struct vm_area_struct *vma) { struct inode *inode; - if (!vma->vm_file) + if (!IS_ENABLED(CONFIG_FS_DAX) || !vma->vm_file) return false; if (!vma_is_dax(vma)) return false; -- cgit