From 38f1755a3e59a3f88e33030f8e4ee0421de2f05a Mon Sep 17 00:00:00 2001 From: Min-Hua Chen Date: Fri, 12 May 2023 00:46:25 +0800 Subject: fs: use correct __poll_t type Fix the following sparse warnings by using __poll_t instead of unsigned type. fs/eventpoll.c:541:9: sparse: warning: restricted __poll_t degrades to integer fs/eventfd.c:67:17: sparse: warning: restricted __poll_t degrades to integer Signed-off-by: Min-Hua Chen Message-Id: <20230511164628.336586-1-minhuadotchen@gmail.com> Signed-off-by: Christian Brauner --- include/linux/eventfd.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/eventfd.h b/include/linux/eventfd.h index 36a486505b08..98d31cdaca40 100644 --- a/include/linux/eventfd.h +++ b/include/linux/eventfd.h @@ -40,7 +40,7 @@ struct file *eventfd_fget(int fd); struct eventfd_ctx *eventfd_ctx_fdget(int fd); struct eventfd_ctx *eventfd_ctx_fileget(struct file *file); __u64 eventfd_signal(struct eventfd_ctx *ctx, __u64 n); -__u64 eventfd_signal_mask(struct eventfd_ctx *ctx, __u64 n, unsigned mask); +__u64 eventfd_signal_mask(struct eventfd_ctx *ctx, __u64 n, __poll_t mask); int eventfd_ctx_remove_wait_queue(struct eventfd_ctx *ctx, wait_queue_entry_t *wait, __u64 *cnt); void eventfd_ctx_do_read(struct eventfd_ctx *ctx, __u64 *cnt); -- cgit v1.2.3 From 247c8d2f9837a3e29e3b6b7a4aa9c36c37659dd4 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 16 May 2023 21:56:12 +0200 Subject: fs: pipe: reveal missing function protoypes A couple of functions from fs/pipe.c are used both internally and for the watch queue code, but the declaration is only visible when the latter is enabled: fs/pipe.c:1254:5: error: no previous prototype for 'pipe_resize_ring' fs/pipe.c:758:15: error: no previous prototype for 'account_pipe_buffers' fs/pipe.c:764:6: error: no previous prototype for 'too_many_pipe_buffers_soft' fs/pipe.c:771:6: error: no previous prototype for 'too_many_pipe_buffers_hard' fs/pipe.c:777:6: error: no previous prototype for 'pipe_is_unprivileged_user' Make the visible unconditionally to avoid these warnings. Fixes: c73be61cede5 ("pipe: Add general notification queue support") Signed-off-by: Arnd Bergmann Message-Id: <20230516195629.551602-1-arnd@kernel.org> Signed-off-by: Christian Brauner --- include/linux/pipe_fs_i.h | 4 ---- 1 file changed, 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/pipe_fs_i.h b/include/linux/pipe_fs_i.h index d2c3f16cf6b1..02e0086b10f6 100644 --- a/include/linux/pipe_fs_i.h +++ b/include/linux/pipe_fs_i.h @@ -261,18 +261,14 @@ void generic_pipe_buf_release(struct pipe_inode_info *, struct pipe_buffer *); extern const struct pipe_buf_operations nosteal_pipe_buf_ops; -#ifdef CONFIG_WATCH_QUEUE unsigned long account_pipe_buffers(struct user_struct *user, unsigned long old, unsigned long new); bool too_many_pipe_buffers_soft(unsigned long user_bufs); bool too_many_pipe_buffers_hard(unsigned long user_bufs); bool pipe_is_unprivileged_user(void); -#endif /* for F_SETPIPE_SZ and F_GETPIPE_SZ */ -#ifdef CONFIG_WATCH_QUEUE int pipe_resize_ring(struct pipe_inode_info *pipe, unsigned int nr_slots); -#endif long pipe_fcntl(struct file *, unsigned int, unsigned long arg); struct pipe_inode_info *get_pipe_info(struct file *file, bool for_splice); -- cgit v1.2.3 From ef104443bffa004f631729dfc924f0b84abbd602 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 16 May 2023 21:57:29 +0200 Subject: procfs: consolidate arch_report_meminfo declaration The arch_report_meminfo() function is provided by four architectures, with a __weak fallback in procfs itself. On architectures that don't have a custom version, the __weak version causes a warning because of the missing prototype. Remove the architecture specific prototypes and instead add one in linux/proc_fs.h. Signed-off-by: Arnd Bergmann Acked-by: Dave Hansen # for arch/x86 Acked-by: Helge Deller # parisc Reviewed-by: Alexander Gordeev Message-Id: <20230516195834.551901-1-arnd@kernel.org> Signed-off-by: Christian Brauner --- arch/parisc/include/asm/pgtable.h | 3 --- arch/powerpc/include/asm/pgtable.h | 3 --- arch/s390/include/asm/pgtable.h | 3 --- arch/s390/mm/pageattr.c | 1 + arch/x86/include/asm/pgtable.h | 1 + arch/x86/include/asm/pgtable_types.h | 3 --- arch/x86/mm/pat/set_memory.c | 1 + include/linux/proc_fs.h | 2 ++ 8 files changed, 5 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/arch/parisc/include/asm/pgtable.h b/arch/parisc/include/asm/pgtable.h index e715df5385d6..5656395c95ee 100644 --- a/arch/parisc/include/asm/pgtable.h +++ b/arch/parisc/include/asm/pgtable.h @@ -472,9 +472,6 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, #define pte_same(A,B) (pte_val(A) == pte_val(B)) -struct seq_file; -extern void arch_report_meminfo(struct seq_file *m); - #endif /* !__ASSEMBLY__ */ diff --git a/arch/powerpc/include/asm/pgtable.h b/arch/powerpc/include/asm/pgtable.h index 9972626ddaf6..6a88bfdaa69b 100644 --- a/arch/powerpc/include/asm/pgtable.h +++ b/arch/powerpc/include/asm/pgtable.h @@ -165,9 +165,6 @@ static inline bool is_ioremap_addr(const void *x) return addr >= IOREMAP_BASE && addr < IOREMAP_END; } - -struct seq_file; -void arch_report_meminfo(struct seq_file *m); #endif /* CONFIG_PPC64 */ #endif /* __ASSEMBLY__ */ diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h index 6822a11c2c8a..c55f3c3365af 100644 --- a/arch/s390/include/asm/pgtable.h +++ b/arch/s390/include/asm/pgtable.h @@ -42,9 +42,6 @@ static inline void update_page_count(int level, long count) atomic_long_add(count, &direct_pages_count[level]); } -struct seq_file; -void arch_report_meminfo(struct seq_file *m); - /* * The S390 doesn't have any external MMU info: the kernel page * tables contain all the necessary information. diff --git a/arch/s390/mm/pageattr.c b/arch/s390/mm/pageattr.c index 5ba3bd8a7b12..ca5a418c58a8 100644 --- a/arch/s390/mm/pageattr.c +++ b/arch/s390/mm/pageattr.c @@ -4,6 +4,7 @@ * Author(s): Jan Glauber */ #include +#include #include #include #include diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h index 15ae4d6ba476..5700bb337987 100644 --- a/arch/x86/include/asm/pgtable.h +++ b/arch/x86/include/asm/pgtable.h @@ -27,6 +27,7 @@ extern pgd_t early_top_pgt[PTRS_PER_PGD]; bool __init __early_make_pgtable(unsigned long address, pmdval_t pmd); +struct seq_file; void ptdump_walk_pgd_level(struct seq_file *m, struct mm_struct *mm); void ptdump_walk_pgd_level_debugfs(struct seq_file *m, struct mm_struct *mm, bool user); diff --git a/arch/x86/include/asm/pgtable_types.h b/arch/x86/include/asm/pgtable_types.h index 447d4bee25c4..ba3e2554799a 100644 --- a/arch/x86/include/asm/pgtable_types.h +++ b/arch/x86/include/asm/pgtable_types.h @@ -513,9 +513,6 @@ extern void native_pagetable_init(void); #define native_pagetable_init paging_init #endif -struct seq_file; -extern void arch_report_meminfo(struct seq_file *m); - enum pg_level { PG_LEVEL_NONE, PG_LEVEL_4K, diff --git a/arch/x86/mm/pat/set_memory.c b/arch/x86/mm/pat/set_memory.c index 7159cf787613..d1515756e369 100644 --- a/arch/x86/mm/pat/set_memory.c +++ b/arch/x86/mm/pat/set_memory.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include #include diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h index 0260f5ea98fe..253f2676d93a 100644 --- a/include/linux/proc_fs.h +++ b/include/linux/proc_fs.h @@ -158,6 +158,8 @@ int proc_pid_arch_status(struct seq_file *m, struct pid_namespace *ns, struct pid *pid, struct task_struct *task); #endif /* CONFIG_PROC_PID_ARCH_STATUS */ +void arch_report_meminfo(struct seq_file *m); + #else /* CONFIG_PROC_FS */ static inline void proc_root_init(void) -- cgit v1.2.3 From d0e135408e196921da2c85ee424235382c9ed614 Mon Sep 17 00:00:00 2001 From: "Fabio M. De Francesco" Date: Fri, 2 Jun 2023 12:33:07 +0200 Subject: highmem: Rename put_and_unmap_page() to unmap_and_put_page() With commit 849ad04cf562a ("new helper: put_and_unmap_page()"), Al Viro introduced the put_and_unmap_page() to use in those many places where we have a common pattern consisting of calls to kunmap_local() + put_page(). Obviously, first we unmap and then we put pages. Instead, the original name of this helper seems to imply that we first put and then unmap. Therefore, rename the helper and change the only known upstreamed user (i.e., fs/sysv) before this helper enters common use and might become difficult to find all call sites and instead easy to break the builds. Cc: Al Viro Signed-off-by: Fabio M. De Francesco Reviewed-by: Eric Biggers Message-Id: <20230602103307.5637-1-fmdefrancesco@gmail.com> Signed-off-by: Christian Brauner --- fs/sysv/dir.c | 22 +++++++++++----------- fs/sysv/namei.c | 8 ++++---- include/linux/highmem.h | 2 +- 3 files changed, 16 insertions(+), 16 deletions(-) (limited to 'include/linux') diff --git a/fs/sysv/dir.c b/fs/sysv/dir.c index cdb3d632c63d..0140010aa0c3 100644 --- a/fs/sysv/dir.c +++ b/fs/sysv/dir.c @@ -52,7 +52,7 @@ static int sysv_handle_dirsync(struct inode *dir) } /* - * Calls to dir_get_page()/put_and_unmap_page() must be nested according to the + * Calls to dir_get_page()/unmap_and_put_page() must be nested according to the * rules documented in mm/highmem.rst. * * NOTE: sysv_find_entry() and sysv_dotdot() act as calls to dir_get_page() @@ -103,11 +103,11 @@ static int sysv_readdir(struct file *file, struct dir_context *ctx) if (!dir_emit(ctx, name, strnlen(name,SYSV_NAMELEN), fs16_to_cpu(SYSV_SB(sb), de->inode), DT_UNKNOWN)) { - put_and_unmap_page(page, kaddr); + unmap_and_put_page(page, kaddr); return 0; } } - put_and_unmap_page(page, kaddr); + unmap_and_put_page(page, kaddr); } return 0; } @@ -131,7 +131,7 @@ static inline int namecompare(int len, int maxlen, * itself (as a parameter - res_dir). It does NOT read the inode of the * entry - you'll have to do that yourself if you want to. * - * On Success put_and_unmap_page() should be called on *res_page. + * On Success unmap_and_put_page() should be called on *res_page. * * sysv_find_entry() acts as a call to dir_get_page() and must be treated * accordingly for nesting purposes. @@ -166,7 +166,7 @@ struct sysv_dir_entry *sysv_find_entry(struct dentry *dentry, struct page **res_ name, de->name)) goto found; } - put_and_unmap_page(page, kaddr); + unmap_and_put_page(page, kaddr); } if (++n >= npages) @@ -209,7 +209,7 @@ int sysv_add_link(struct dentry *dentry, struct inode *inode) goto out_page; de++; } - put_and_unmap_page(page, kaddr); + unmap_and_put_page(page, kaddr); } BUG(); return -EINVAL; @@ -228,7 +228,7 @@ got_it: mark_inode_dirty(dir); err = sysv_handle_dirsync(dir); out_page: - put_and_unmap_page(page, kaddr); + unmap_and_put_page(page, kaddr); return err; out_unlock: unlock_page(page); @@ -321,12 +321,12 @@ int sysv_empty_dir(struct inode * inode) if (de->name[1] != '.' || de->name[2]) goto not_empty; } - put_and_unmap_page(page, kaddr); + unmap_and_put_page(page, kaddr); } return 1; not_empty: - put_and_unmap_page(page, kaddr); + unmap_and_put_page(page, kaddr); return 0; } @@ -352,7 +352,7 @@ int sysv_set_link(struct sysv_dir_entry *de, struct page *page, } /* - * Calls to dir_get_page()/put_and_unmap_page() must be nested according to the + * Calls to dir_get_page()/unmap_and_put_page() must be nested according to the * rules documented in mm/highmem.rst. * * sysv_dotdot() acts as a call to dir_get_page() and must be treated @@ -376,7 +376,7 @@ ino_t sysv_inode_by_name(struct dentry *dentry) if (de) { res = fs16_to_cpu(SYSV_SB(dentry->d_sb), de->inode); - put_and_unmap_page(page, de); + unmap_and_put_page(page, de); } return res; } diff --git a/fs/sysv/namei.c b/fs/sysv/namei.c index 2b2dba4c4f56..fcf163fea3ad 100644 --- a/fs/sysv/namei.c +++ b/fs/sysv/namei.c @@ -164,7 +164,7 @@ static int sysv_unlink(struct inode * dir, struct dentry * dentry) inode->i_ctime = dir->i_ctime; inode_dec_link_count(inode); } - put_and_unmap_page(page, de); + unmap_and_put_page(page, de); return err; } @@ -227,7 +227,7 @@ static int sysv_rename(struct mnt_idmap *idmap, struct inode *old_dir, if (!new_de) goto out_dir; err = sysv_set_link(new_de, new_page, old_inode); - put_and_unmap_page(new_page, new_de); + unmap_and_put_page(new_page, new_de); if (err) goto out_dir; new_inode->i_ctime = current_time(new_inode); @@ -256,9 +256,9 @@ static int sysv_rename(struct mnt_idmap *idmap, struct inode *old_dir, out_dir: if (dir_de) - put_and_unmap_page(dir_page, dir_de); + unmap_and_put_page(dir_page, dir_de); out_old: - put_and_unmap_page(old_page, old_de); + unmap_and_put_page(old_page, old_de); out: return err; } diff --git a/include/linux/highmem.h b/include/linux/highmem.h index 4de1dbcd3ef6..68da30625a6c 100644 --- a/include/linux/highmem.h +++ b/include/linux/highmem.h @@ -507,7 +507,7 @@ static inline void folio_zero_range(struct folio *folio, zero_user_segments(&folio->page, start, start + length, 0, 0); } -static inline void put_and_unmap_page(struct page *page, void *addr) +static inline void unmap_and_put_page(struct page *page, void *addr) { kunmap_local(addr); put_page(page); -- cgit v1.2.3 From a7bc2e8ddf3c8e1f5bfeb401f7ee112956cea259 Mon Sep 17 00:00:00 2001 From: chenzhiyin Date: Thu, 1 Jun 2023 05:24:00 -0400 Subject: fs.h: Optimize file struct to prevent false sharing In the syscall test of UnixBench, performance regression occurred due to false sharing. The lock and atomic members, including file::f_lock, file::f_count and file::f_pos_lock are highly contended and frequently updated in the high-concurrency test scenarios. perf c2c indentified one affected read access, file::f_op. To prevent false sharing, the layout of file struct is changed as following (A) f_lock, f_count and f_pos_lock are put together to share the same cache line. (B) The read mostly members, including f_path, f_inode, f_op are put into a separate cache line. (C) f_mode is put together with f_count, since they are used frequently at the same time. Due to '__randomize_layout' attribute of file struct, the updated layout only can be effective when CONFIG_RANDSTRUCT_NONE is 'y'. The optimization has been validated in the syscall test of UnixBench. performance gain is 30~50%. Furthermore, to confirm the optimization effectiveness on the other codes path, the results of fsdisk, fsbuffer and fstime are also shown. Here are the detailed test results of unixbench. Command: numactl -C 3-18 ./Run -c 16 syscall fsbuffer fstime fsdisk Without Patch ------------------------------------------------------------------------ File Copy 1024 bufsize 2000 maxblocks 875052.1 KBps (30.0 s, 2 samples) File Copy 256 bufsize 500 maxblocks 235484.0 KBps (30.0 s, 2 samples) File Copy 4096 bufsize 8000 maxblocks 2815153.5 KBps (30.0 s, 2 samples) System Call Overhead 5772268.3 lps (10.0 s, 7 samples) System Benchmarks Partial Index BASELINE RESULT INDEX File Copy 1024 bufsize 2000 maxblocks 3960.0 875052.1 2209.7 File Copy 256 bufsize 500 maxblocks 1655.0 235484.0 1422.9 File Copy 4096 bufsize 8000 maxblocks 5800.0 2815153.5 4853.7 System Call Overhead 15000.0 5772268.3 3848.2 ======== System Benchmarks Index Score (Partial Only) 2768.3 With Patch ------------------------------------------------------------------------ File Copy 1024 bufsize 2000 maxblocks 1009977.2 KBps (30.0 s, 2 samples) File Copy 256 bufsize 500 maxblocks 264765.9 KBps (30.0 s, 2 samples) File Copy 4096 bufsize 8000 maxblocks 3052236.0 KBps (30.0 s, 2 samples) System Call Overhead 8237404.4 lps (10.0 s, 7 samples) System Benchmarks Partial Index BASELINE RESULT INDEX File Copy 1024 bufsize 2000 maxblocks 3960.0 1009977.2 2550.4 File Copy 256 bufsize 500 maxblocks 1655.0 264765.9 1599.8 File Copy 4096 bufsize 8000 maxblocks 5800.0 3052236.0 5262.5 System Call Overhead 15000.0 8237404.4 5491.6 ======== System Benchmarks Index Score (Partial Only) 3295.3 Signed-off-by: chenzhiyin Message-Id: <20230601092400.27162-1-zhiyin.chen@intel.com> Signed-off-by: Christian Brauner --- include/linux/fs.h | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 21a981680856..9c2671b285a4 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -956,29 +956,35 @@ static inline int ra_has_index(struct file_ra_state *ra, pgoff_t index) index < ra->start + ra->size); } +/* + * f_{lock,count,pos_lock} members can be highly contended and share + * the same cacheline. f_{lock,mode} are very frequently used together + * and so share the same cacheline as well. The read-mostly + * f_{path,inode,op} are kept on a separate cacheline. + */ struct file { union { struct llist_node f_llist; struct rcu_head f_rcuhead; unsigned int f_iocb_flags; }; - struct path f_path; - struct inode *f_inode; /* cached value */ - const struct file_operations *f_op; /* * Protects f_ep, f_flags. * Must not be taken from IRQ context. */ spinlock_t f_lock; - atomic_long_t f_count; - unsigned int f_flags; fmode_t f_mode; + atomic_long_t f_count; struct mutex f_pos_lock; loff_t f_pos; + unsigned int f_flags; struct fown_struct f_owner; const struct cred *f_cred; struct file_ra_state f_ra; + struct path f_path; + struct inode *f_inode; /* cached value */ + const struct file_operations *f_op; u64 f_version; #ifdef CONFIG_SECURITY -- cgit v1.2.3 From 943211c87427f25bd22e0e63849fb486bb5f87fa Mon Sep 17 00:00:00 2001 From: Siddh Raman Pant Date: Mon, 5 Jun 2023 20:06:16 +0530 Subject: watch_queue: prevent dangling pipe pointer NULL the dangling pipe reference while clearing watch_queue. If not done, a reference to a freed pipe remains in the watch_queue, as this function is called before freeing a pipe in free_pipe_info() (see line 834 of fs/pipe.c). The sole use of wqueue->defunct is for checking if the watch queue has been cleared, but wqueue->pipe is also NULLed while clearing. Thus, wqueue->defunct is superfluous, as wqueue->pipe can be checked for NULL. Hence, the former can be removed. Tested with keyutils testsuite. Cc: stable@vger.kernel.org # 6.1 Signed-off-by: Siddh Raman Pant Acked-by: David Howells Message-Id: <20230605143616.640517-1-code@siddh.me> Signed-off-by: Christian Brauner --- include/linux/watch_queue.h | 3 +-- kernel/watch_queue.c | 12 ++++++------ 2 files changed, 7 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/watch_queue.h b/include/linux/watch_queue.h index fc6bba20273b..45cd42f55d49 100644 --- a/include/linux/watch_queue.h +++ b/include/linux/watch_queue.h @@ -38,7 +38,7 @@ struct watch_filter { struct watch_queue { struct rcu_head rcu; struct watch_filter __rcu *filter; - struct pipe_inode_info *pipe; /* The pipe we're using as a buffer */ + struct pipe_inode_info *pipe; /* Pipe we use as a buffer, NULL if queue closed */ struct hlist_head watches; /* Contributory watches */ struct page **notes; /* Preallocated notifications */ unsigned long *notes_bitmap; /* Allocation bitmap for notes */ @@ -46,7 +46,6 @@ struct watch_queue { spinlock_t lock; unsigned int nr_notes; /* Number of notes */ unsigned int nr_pages; /* Number of pages in notes[] */ - bool defunct; /* T when queues closed */ }; /* diff --git a/kernel/watch_queue.c b/kernel/watch_queue.c index e91cb4c2833f..d0b6b390ee42 100644 --- a/kernel/watch_queue.c +++ b/kernel/watch_queue.c @@ -42,7 +42,7 @@ MODULE_AUTHOR("Red Hat, Inc."); static inline bool lock_wqueue(struct watch_queue *wqueue) { spin_lock_bh(&wqueue->lock); - if (unlikely(wqueue->defunct)) { + if (unlikely(!wqueue->pipe)) { spin_unlock_bh(&wqueue->lock); return false; } @@ -104,9 +104,6 @@ static bool post_one_notification(struct watch_queue *wqueue, unsigned int head, tail, mask, note, offset, len; bool done = false; - if (!pipe) - return false; - spin_lock_irq(&pipe->rd_wait.lock); mask = pipe->ring_size - 1; @@ -603,8 +600,11 @@ void watch_queue_clear(struct watch_queue *wqueue) rcu_read_lock(); spin_lock_bh(&wqueue->lock); - /* Prevent new notifications from being stored. */ - wqueue->defunct = true; + /* + * This pipe can be freed by callers like free_pipe_info(). + * Removing this reference also prevents new notifications. + */ + wqueue->pipe = NULL; while (!hlist_empty(&wqueue->watches)) { watch = hlist_entry(wqueue->watches.first, struct watch, queue_node); -- cgit v1.2.3 From b6334e2cd46f0dab1131c0272426182c79b51114 Mon Sep 17 00:00:00 2001 From: Mao Zhu Date: Sun, 11 Jun 2023 08:33:14 -0400 Subject: fs: Fix comment typo Delete duplicated word in comment. Signed-off-by: Mao Zhu Message-Id: <20230611123314.5282-1-dengshaomin@cdjrlc.com> 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 9c2671b285a4..40bef9bf8749 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2675,7 +2675,7 @@ extern void evict_inodes(struct super_block *sb); void dump_mapping(const struct address_space *); /* - * Userspace may rely on the the inode number being non-zero. For example, glibc + * Userspace may rely on the inode number being non-zero. For example, glibc * simply ignores files with zero i_ino in unlink() and other places. * * As an additional complication, if userspace was compiled with -- cgit v1.2.3 From 2d8c9dcf7158060fcec9f891c0292ffdb4397523 Mon Sep 17 00:00:00 2001 From: Wen Yang Date: Thu, 15 Jun 2023 02:40:28 +0800 Subject: eventfd: add a uapi header for eventfd userspace APIs Create a uapi header include/uapi/linux/eventfd.h, move the associated flags to the uapi header, and include it from linux/eventfd.h. Suggested-by: Christian Brauner Signed-off-by: Wen Yang Reviewed-by: Matthew Wilcox (Oracle) Cc: Alexander Viro Cc: Jens Axboe Cc: Christian Brauner Cc: Christoph Hellwig Cc: Dylan Yudaken Cc: David Woodhouse Cc: Matthew Wilcox Cc: Eric Biggers Cc: linux-fsdevel@vger.kernel.org Cc: linux-kernel@vger.kernel.org Message-Id: Signed-off-by: Christian Brauner --- include/linux/eventfd.h | 6 +----- include/uapi/linux/eventfd.h | 11 +++++++++++ 2 files changed, 12 insertions(+), 5 deletions(-) create mode 100644 include/uapi/linux/eventfd.h (limited to 'include/linux') diff --git a/include/linux/eventfd.h b/include/linux/eventfd.h index 98d31cdaca40..b9d83652c097 100644 --- a/include/linux/eventfd.h +++ b/include/linux/eventfd.h @@ -9,12 +9,12 @@ #ifndef _LINUX_EVENTFD_H #define _LINUX_EVENTFD_H -#include #include #include #include #include #include +#include /* * CAREFUL: Check include/uapi/asm-generic/fcntl.h when defining @@ -23,10 +23,6 @@ * from eventfd, in order to leave a free define-space for * shared O_* flags. */ -#define EFD_SEMAPHORE (1 << 0) -#define EFD_CLOEXEC O_CLOEXEC -#define EFD_NONBLOCK O_NONBLOCK - #define EFD_SHARED_FCNTL_FLAGS (O_CLOEXEC | O_NONBLOCK) #define EFD_FLAGS_SET (EFD_SHARED_FCNTL_FLAGS | EFD_SEMAPHORE) diff --git a/include/uapi/linux/eventfd.h b/include/uapi/linux/eventfd.h new file mode 100644 index 000000000000..2eb9ab6c32f3 --- /dev/null +++ b/include/uapi/linux/eventfd.h @@ -0,0 +1,11 @@ +/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ +#ifndef _UAPI_LINUX_EVENTFD_H +#define _UAPI_LINUX_EVENTFD_H + +#include + +#define EFD_SEMAPHORE (1 << 0) +#define EFD_CLOEXEC O_CLOEXEC +#define EFD_NONBLOCK O_NONBLOCK + +#endif /* _UAPI_LINUX_EVENTFD_H */ -- cgit v1.2.3 From d7439fb1f4338fffd0bc68bb62d78f7712725f26 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Tue, 20 Jun 2023 13:28:32 +0200 Subject: fs: Provide helpers for manipulating sb->s_readonly_remount Provide helpers to set and clear sb->s_readonly_remount including appropriate memory barriers. Also use this opportunity to document what the barriers pair with and why they are needed. Suggested-by: Dave Chinner Signed-off-by: Jan Kara Reviewed-by: Dave Chinner Message-Id: <20230620112832.5158-1-jack@suse.cz> Signed-off-by: Christian Brauner --- fs/internal.h | 41 +++++++++++++++++++++++++++++++++++++++++ fs/namespace.c | 25 ++++++++++++++++--------- fs/super.c | 17 ++++++----------- include/linux/fs.h | 2 +- 4 files changed, 64 insertions(+), 21 deletions(-) (limited to 'include/linux') diff --git a/fs/internal.h b/fs/internal.h index bd3b2810a36b..b916b84809f3 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -120,6 +120,47 @@ void put_super(struct super_block *sb); extern bool mount_capable(struct fs_context *); int sb_init_dio_done_wq(struct super_block *sb); +/* + * Prepare superblock for changing its read-only state (i.e., either remount + * read-write superblock read-only or vice versa). After this function returns + * mnt_is_readonly() will return true for any mount of the superblock if its + * caller is able to observe any changes done by the remount. This holds until + * sb_end_ro_state_change() is called. + */ +static inline void sb_start_ro_state_change(struct super_block *sb) +{ + WRITE_ONCE(sb->s_readonly_remount, 1); + /* + * For RO->RW transition, the barrier pairs with the barrier in + * mnt_is_readonly() making sure if mnt_is_readonly() sees SB_RDONLY + * cleared, it will see s_readonly_remount set. + * For RW->RO transition, the barrier pairs with the barrier in + * __mnt_want_write() before the mnt_is_readonly() check. The barrier + * makes sure if __mnt_want_write() sees MNT_WRITE_HOLD already + * cleared, it will see s_readonly_remount set. + */ + smp_wmb(); +} + +/* + * Ends section changing read-only state of the superblock. After this function + * returns if mnt_is_readonly() returns false, the caller will be able to + * observe all the changes remount did to the superblock. + */ +static inline void sb_end_ro_state_change(struct super_block *sb) +{ + /* + * This barrier provides release semantics that pairs with + * the smp_rmb() acquire semantics in mnt_is_readonly(). + * This barrier pair ensure that when mnt_is_readonly() sees + * 0 for sb->s_readonly_remount, it will also see all the + * preceding flag changes that were made during the RO state + * change. + */ + smp_wmb(); + WRITE_ONCE(sb->s_readonly_remount, 0); +} + /* * open.c */ diff --git a/fs/namespace.c b/fs/namespace.c index 54847db5b819..5ba1eca6f720 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -309,9 +309,16 @@ static unsigned int mnt_get_writers(struct mount *mnt) static int mnt_is_readonly(struct vfsmount *mnt) { - if (mnt->mnt_sb->s_readonly_remount) + if (READ_ONCE(mnt->mnt_sb->s_readonly_remount)) return 1; - /* Order wrt setting s_flags/s_readonly_remount in do_remount() */ + /* + * The barrier pairs with the barrier in sb_start_ro_state_change() + * making sure if we don't see s_readonly_remount set yet, we also will + * not see any superblock / mount flag changes done by remount. + * It also pairs with the barrier in sb_end_ro_state_change() + * assuring that if we see s_readonly_remount already cleared, we will + * see the values of superblock / mount flags updated by remount. + */ smp_rmb(); return __mnt_is_readonly(mnt); } @@ -364,9 +371,11 @@ int __mnt_want_write(struct vfsmount *m) } } /* - * After the slowpath clears MNT_WRITE_HOLD, mnt_is_readonly will - * be set to match its requirements. So we must not load that until - * MNT_WRITE_HOLD is cleared. + * The barrier pairs with the barrier sb_start_ro_state_change() making + * sure that if we see MNT_WRITE_HOLD cleared, we will also see + * s_readonly_remount set (or even SB_RDONLY / MNT_READONLY flags) in + * mnt_is_readonly() and bail in case we are racing with remount + * read-only. */ smp_rmb(); if (mnt_is_readonly(m)) { @@ -588,10 +597,8 @@ int sb_prepare_remount_readonly(struct super_block *sb) if (!err && atomic_long_read(&sb->s_remove_count)) err = -EBUSY; - if (!err) { - sb->s_readonly_remount = 1; - smp_wmb(); - } + if (!err) + sb_start_ro_state_change(sb); list_for_each_entry(mnt, &sb->s_mounts, mnt_instance) { if (mnt->mnt.mnt_flags & MNT_WRITE_HOLD) mnt->mnt.mnt_flags &= ~MNT_WRITE_HOLD; diff --git a/fs/super.c b/fs/super.c index 5bf056087acc..5ee5da1fd498 100644 --- a/fs/super.c +++ b/fs/super.c @@ -944,8 +944,7 @@ int reconfigure_super(struct fs_context *fc) */ if (remount_ro) { if (force) { - sb->s_readonly_remount = 1; - smp_wmb(); + sb_start_ro_state_change(sb); } else { retval = sb_prepare_remount_readonly(sb); if (retval) @@ -953,12 +952,10 @@ int reconfigure_super(struct fs_context *fc) } } else if (remount_rw) { /* - * We set s_readonly_remount here to protect filesystem's - * reconfigure code from writes from userspace until - * reconfigure finishes. + * Protect filesystem's reconfigure code from writes from + * userspace until reconfigure finishes. */ - sb->s_readonly_remount = 1; - smp_wmb(); + sb_start_ro_state_change(sb); } if (fc->ops->reconfigure) { @@ -974,9 +971,7 @@ int reconfigure_super(struct fs_context *fc) WRITE_ONCE(sb->s_flags, ((sb->s_flags & ~fc->sb_flags_mask) | (fc->sb_flags & fc->sb_flags_mask))); - /* Needs to be ordered wrt mnt_is_readonly() */ - smp_wmb(); - sb->s_readonly_remount = 0; + sb_end_ro_state_change(sb); /* * Some filesystems modify their metadata via some other path than the @@ -991,7 +986,7 @@ int reconfigure_super(struct fs_context *fc) return 0; cancel_readonly: - sb->s_readonly_remount = 0; + sb_end_ro_state_change(sb); return retval; } diff --git a/include/linux/fs.h b/include/linux/fs.h index 40bef9bf8749..4caac7fdc5d3 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1248,7 +1248,7 @@ struct super_block { */ atomic_long_t s_fsnotify_connectors; - /* Being remounted read-only */ + /* Read-only state of the superblock is being changed */ int s_readonly_remount; /* per-sb errseq_t for reporting writeback errors via syncfs */ -- cgit v1.2.3