From 344bac8f0d73fe970cd9f5b2f132906317d29e8b Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Sun, 15 Dec 2024 21:17:05 +0100 Subject: fs: kill MNT_ONRB Move mnt->mnt_node into the union with mnt->mnt_rcu and mnt->mnt_llist instead of keeping it with mnt->mnt_list. This allows us to use RB_CLEAR_NODE(&mnt->mnt_node) in umount_tree() as well as list_empty(&mnt->mnt_node). That in turn allows us to remove MNT_ONRB. This also fixes the bug reported in [1] where seemingly MNT_ONRB wasn't set in @mnt->mnt_flags even though the mount was present in the mount rbtree of the mount namespace. The root cause is the following race. When a btrfs subvolume is mounted a temporary mount is created: btrfs_get_tree_subvol() { mnt = fc_mount() // Register the newly allocated mount with sb->mounts: lock_mount_hash(); list_add_tail(&mnt->mnt_instance, &mnt->mnt.mnt_sb->s_mounts); unlock_mount_hash(); } and registered on sb->s_mounts. Later it is added to an anonymous mount namespace via mount_subvol(): -> mount_subvol() -> mount_subtree() -> alloc_mnt_ns() mnt_add_to_ns() vfs_path_lookup() put_mnt_ns() The mnt_add_to_ns() call raises MNT_ONRB in @mnt->mnt_flags. If someone concurrently does a ro remount: reconfigure_super() -> sb_prepare_remount_readonly() { list_for_each_entry(mnt, &sb->s_mounts, mnt_instance) { } all mounts registered in sb->s_mounts are visited and first MNT_WRITE_HOLD is raised, then MNT_READONLY is raised, and finally MNT_WRITE_HOLD is removed again. The flag modification for MNT_WRITE_HOLD/MNT_READONLY and MNT_ONRB race so MNT_ONRB might be lost. Fixes: 2eea9ce4310d ("mounts: keep list of mounts in an rbtree") Cc: # v6.8+ Link: https://lore.kernel.org/r/20241215-vfs-6-14-mount-work-v1-1-fd55922c4af8@kernel.org Link: https://lore.kernel.org/r/ec6784ed-8722-4695-980a-4400d4e7bd1a@gmx.com [1] Signed-off-by: Christian Brauner --- include/linux/mount.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mount.h b/include/linux/mount.h index c34c18b4e8f3..04213d8ef837 100644 --- a/include/linux/mount.h +++ b/include/linux/mount.h @@ -50,7 +50,7 @@ struct path; #define MNT_ATIME_MASK (MNT_NOATIME | MNT_NODIRATIME | MNT_RELATIME ) #define MNT_INTERNAL_FLAGS (MNT_SHARED | MNT_WRITE_HOLD | MNT_INTERNAL | \ - MNT_DOOMED | MNT_SYNC_UMOUNT | MNT_MARKED | MNT_ONRB) + MNT_DOOMED | MNT_SYNC_UMOUNT | MNT_MARKED) #define MNT_INTERNAL 0x4000 @@ -64,7 +64,6 @@ struct path; #define MNT_SYNC_UMOUNT 0x2000000 #define MNT_MARKED 0x4000000 #define MNT_UMOUNT 0x8000000 -#define MNT_ONRB 0x10000000 struct vfsmount { struct dentry *mnt_root; /* root of the mounted tree */ -- cgit v1.2.3 From cacd9ae4bf801ff4125d8961bb9a3ba955e51680 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 7 Jan 2025 17:27:17 +0100 Subject: poll_wait: add mb() to fix theoretical race between waitqueue_active() and .poll() As the comment above waitqueue_active() explains, it can only be used if both waker and waiter have mb()'s that pair with each other. However __pollwait() is broken in this respect. This is not pipe-specific, but let's look at pipe_poll() for example: poll_wait(...); // -> __pollwait() -> add_wait_queue() LOAD(pipe->head); LOAD(pipe->head); In theory these LOAD()'s can leak into the critical section inside add_wait_queue() and can happen before list_add(entry, wq_head), in this case pipe_poll() can race with wakeup_pipe_readers/writers which do smp_mb(); if (waitqueue_active(wq_head)) wake_up_interruptible(wq_head); There are more __pollwait()-like functions (grep init_poll_funcptr), and it seems that at least ep_ptable_queue_proc() has the same problem, so the patch adds smp_mb() into poll_wait(). Link: https://lore.kernel.org/all/20250102163320.GA17691@redhat.com/ Signed-off-by: Oleg Nesterov Link: https://lore.kernel.org/r/20250107162717.GA18922@redhat.com Signed-off-by: Christian Brauner --- include/linux/poll.h | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/poll.h b/include/linux/poll.h index d1ea4f3714a8..fc641b50f129 100644 --- a/include/linux/poll.h +++ b/include/linux/poll.h @@ -41,8 +41,16 @@ typedef struct poll_table_struct { static inline void poll_wait(struct file * filp, wait_queue_head_t * wait_address, poll_table *p) { - if (p && p->_qproc && wait_address) + if (p && p->_qproc && wait_address) { p->_qproc(filp, wait_address, p); + /* + * This memory barrier is paired in the wq_has_sleeper(). + * See the comment above prepare_to_wait(), we need to + * ensure that subsequent tests in this thread can't be + * reordered with __add_wait_queue() in _qproc() paths. + */ + smp_mb(); + } } /* -- cgit v1.2.3 From 10b02a2cfec2f106db4897ad87732db56d71e6fd Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 7 Jan 2025 17:27:24 +0100 Subject: poll_wait: kill the obsolete wait_address check This check is historical and no longer needed, wait_address is never NULL. These days we rely on the poll_table->_qproc check. NULL if select/poll is not going to sleep, or it already has a data to report, or all waiters have already been registered after the 1st iteration. However, poll_table *p can be NULL, see p9_fd_poll() for example, so we can't remove the "p != NULL" check. Link: https://lore.kernel.org/all/20250106180325.GF7233@redhat.com/ Signed-off-by: Oleg Nesterov Link: https://lore.kernel.org/r/20250107162724.GA18926@redhat.com Signed-off-by: Christian Brauner --- include/linux/poll.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/poll.h b/include/linux/poll.h index fc641b50f129..57b6d1ccd8bf 100644 --- a/include/linux/poll.h +++ b/include/linux/poll.h @@ -41,7 +41,7 @@ typedef struct poll_table_struct { static inline void poll_wait(struct file * filp, wait_queue_head_t * wait_address, poll_table *p) { - if (p && p->_qproc && wait_address) { + if (p && p->_qproc) { p->_qproc(filp, wait_address, p); /* * This memory barrier is paired in the wq_has_sleeper(). -- cgit v1.2.3 From f005bf18a57aadf3af1e85a0f0151cb3688ee606 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 7 Jan 2025 17:27:43 +0100 Subject: poll: kill poll_does_not_wait() It no longer has users. Signed-off-by: Oleg Nesterov Link: https://lore.kernel.org/r/20250107162743.GA18947@redhat.com Signed-off-by: Christian Brauner --- include/linux/poll.h | 16 +++------------- 1 file changed, 3 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/include/linux/poll.h b/include/linux/poll.h index 57b6d1ccd8bf..12bb18e8b978 100644 --- a/include/linux/poll.h +++ b/include/linux/poll.h @@ -25,14 +25,14 @@ struct poll_table_struct; -/* +/* * structures and helpers for f_op->poll implementations */ typedef void (*poll_queue_proc)(struct file *, wait_queue_head_t *, struct poll_table_struct *); /* - * Do not touch the structure directly, use the access functions - * poll_does_not_wait() and poll_requested_events() instead. + * Do not touch the structure directly, use the access function + * poll_requested_events() instead. */ typedef struct poll_table_struct { poll_queue_proc _qproc; @@ -53,16 +53,6 @@ static inline void poll_wait(struct file * filp, wait_queue_head_t * wait_addres } } -/* - * Return true if it is guaranteed that poll will not wait. This is the case - * if the poll() of another file descriptor in the set got an event, so there - * is no need for waiting. - */ -static inline bool poll_does_not_wait(const poll_table *p) -{ - return p == NULL || p->_qproc == NULL; -} - /* * Return the set of events that the application wants to poll for. * This is useful for drivers that need to know whether a DMA transfer has -- cgit v1.2.3