kernel/linux.git/fs/pipe.c, branch v5.15.209

fs/pipe: Fix lockdep false-positive in watchqueue pipe_write()

2024-04-10T14:19:42+00:00

[ Upstream commit 055ca83559912f2cfd91c9441427bac4caf3c74e ] When you try to splice between a normal pipe and a notification pipe, get_pipe_info(..., true) fails, so splice() falls back to treating the notification pipe like a normal pipe - so we end up in iter_file_splice_write(), which first locks the input pipe, then calls vfs_iter_write(), which locks the output pipe. Lockdep complains about that, because we're taking a pipe lock while already holding another pipe lock. I think this probably (?) can't actually lead to deadlocks, since you'd need another way to nest locking a normal pipe into locking a watch_queue pipe, but the lockdep annotations don't make that clear. Bail out earlier in pipe_write() for notification pipes, before taking the pipe lock. Reported-and-tested-by: Closes: https://syzkaller.appspot.com/bug?extid=011e4ea1da6692cf881c Fixes: c73be61cede5 ("pipe: Add general notification queue support") Signed-off-by: Jann Horn Link: https://lore.kernel.org/r/20231124150822.2121798-1-jannh@google.com Signed-off-by: Christian Brauner Signed-off-by: Sasha Levin

pipe: wakeup wr_wait after setting max_usage

2024-02-23T07:54:33+00:00

[ Upstream commit e95aada4cb93d42e25c30a0ef9eb2923d9711d4a ] Commit c73be61cede5 ("pipe: Add general notification queue support") a regression was introduced that would lock up resized pipes under certain conditions. See the reproducer in [1]. The commit resizing the pipe ring size was moved to a different function, doing that moved the wakeup for pipe->wr_wait before actually raising pipe->max_usage. If a pipe was full before the resize occured it would result in the wakeup never actually triggering pipe_write. Set @max_usage and @nr_accounted before waking writers if this isn't a watch queue. Link: https://bugzilla.kernel.org/show_bug.cgi?id=212295 [1] Link: https://lore.kernel.org/r/20231201-orchideen-modewelt-e009de4562c6@brauner Fixes: c73be61cede5 ("pipe: Add general notification queue support") Reviewed-by: David Howells Cc: Signed-off-by: Lukas Schauer [Christian Brauner : rewrite to account for watch queues] Signed-off-by: Christian Brauner Signed-off-by: Sasha Levin

fs/pipe: move check to pipe_has_watch_queue()

2024-02-23T07:54:33+00:00

[ Upstream commit b4bd6b4bac8edd61eb8f7b836969d12c0c6af165 ] This declutters the code by reducing the number of #ifdefs and makes the watch_queue checks simpler. This has no runtime effect; the machine code is identical. Signed-off-by: Max Kellermann Message-Id: <20230921075755.1378787-2-max.kellermann@ionos.com> Reviewed-by: David Howells Signed-off-by: Christian Brauner Stable-dep-of: e95aada4cb93 ("pipe: wakeup wr_wait after setting max_usage") Signed-off-by: Sasha Levin

pipe: Fix missing lock in pipe_resize_ring()

2022-06-06T06:43:37+00:00

commit 189b0ddc245139af81198d1a3637cac74f96e13a upstream. pipe_resize_ring() needs to take the pipe->rd_wait.lock spinlock to prevent post_one_notification() from trying to insert into the ring whilst the ring is being replaced. The occupancy check must be done after the lock is taken, and the lock must be taken after the new ring is allocated. The bug can lead to an oops looking something like: BUG: KASAN: use-after-free in post_one_notification.isra.0+0x62e/0x840 Read of size 4 at addr ffff88801cc72a70 by task poc/27196 ... Call Trace: post_one_notification.isra.0+0x62e/0x840 __post_watch_notification+0x3b7/0x650 key_create_or_update+0xb8b/0xd20 __do_sys_add_key+0x175/0x340 __x64_sys_add_key+0xbe/0x140 do_syscall_64+0x5c/0xc0 entry_SYSCALL_64_after_hwframe+0x44/0xae Reported by Selim Enes Karaduman @Enesdex working with Trend Micro Zero Day Initiative. Fixes: c73be61cede5 ("pipe: Add general notification queue support") Reported-by: zdi-disclosures@trendmicro.com # ZDI-CAN-17291 Signed-off-by: David Howells Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman

pipe: make poll_usage boolean and annotate its access

2022-06-06T06:43:37+00:00

commit f485922d8fe4e44f6d52a5bb95a603b7c65554bb upstream. Patch series "Fix data-races around epoll reported by KCSAN." This series suppresses a false positive KCSAN's message and fixes a real data-race. This patch (of 2): pipe_poll() runs locklessly and assigns 1 to poll_usage. Once poll_usage is set to 1, it never changes in other places. However, concurrent writes of a value trigger KCSAN, so let's make KCSAN happy. BUG: KCSAN: data-race in pipe_poll / pipe_poll write to 0xffff8880042f6678 of 4 bytes by task 174 on cpu 3: pipe_poll (fs/pipe.c:656) ep_item_poll.isra.0 (./include/linux/poll.h:88 fs/eventpoll.c:853) do_epoll_wait (fs/eventpoll.c:1692 fs/eventpoll.c:1806 fs/eventpoll.c:2234) __x64_sys_epoll_wait (fs/eventpoll.c:2246 fs/eventpoll.c:2241 fs/eventpoll.c:2241) do_syscall_64 (arch/x86/entry/common.c:50 arch/x86/entry/common.c:80) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:113) write to 0xffff8880042f6678 of 4 bytes by task 177 on cpu 1: pipe_poll (fs/pipe.c:656) ep_item_poll.isra.0 (./include/linux/poll.h:88 fs/eventpoll.c:853) do_epoll_wait (fs/eventpoll.c:1692 fs/eventpoll.c:1806 fs/eventpoll.c:2234) __x64_sys_epoll_wait (fs/eventpoll.c:2246 fs/eventpoll.c:2241 fs/eventpoll.c:2241) do_syscall_64 (arch/x86/entry/common.c:50 arch/x86/entry/common.c:80) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:113) Reported by Kernel Concurrency Sanitizer on: CPU: 1 PID: 177 Comm: epoll_race Not tainted 5.17.0-58927-gf443e374ae13 #6 Hardware name: Red Hat KVM, BIOS 1.11.0-2.amzn2 04/01/2014 Link: https://lkml.kernel.org/r/20220322002653.33865-1-kuniyu@amazon.co.jp Link: https://lkml.kernel.org/r/20220322002653.33865-2-kuniyu@amazon.co.jp Fixes: 3b844826b6c6 ("pipe: avoid unnecessary EPOLLET wakeups under normal loads") Signed-off-by: Kuniyuki Iwashima Cc: Alexander Duyck Cc: Al Viro Cc: Davidlohr Bueso Cc: Kuniyuki Iwashima Cc: "Soheil Hassas Yeganeh" Cc: "Sridhar Samudrala" Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman

watch_queue: Fix lack of barrier/sync/lock between post and read

2022-03-16T13:23:44+00:00

commit 2ed147f015af2b48f41c6f0b6746aa9ea85c19f3 upstream. There's nothing to synchronise post_one_notification() versus pipe_read(). Whilst posting is done under pipe->rd_wait.lock, the reader only takes pipe->mutex which cannot bar notification posting as that may need to be made from contexts that cannot sleep. Fix this by setting pipe->head with a barrier in post_one_notification() and reading pipe->head with a barrier in pipe_read(). If that's not sufficient, the rd_wait.lock will need to be taken, possibly in a ->confirm() op so that it only applies to notifications. The lock would, however, have to be dropped before copy_page_to_iter() is invoked. Fixes: c73be61cede5 ("pipe: Add general notification queue support") Reported-by: Jann Horn Signed-off-by: David Howells Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman

watch_queue, pipe: Free watchqueue state after clearing pipe ring

2022-03-16T13:23:44+00:00

commit db8facfc9fafacefe8a835416a6b77c838088f8b upstream. In free_pipe_info(), free the watchqueue state after clearing the pipe ring as each pipe ring descriptor has a release function, and in the case of a notification message, this is watch_queue_pipe_buf_release() which tries to mark the allocation bitmap that was previously released. Fix this by moving the put of the pipe's ref on the watch queue to after the ring has been cleared. We still need to call watch_queue_clear() before doing that to make sure that the pipe is disconnected from any notification sources first. Fixes: c73be61cede5 ("pipe: Add general notification queue support") Reported-by: Jann Horn Signed-off-by: David Howells Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman

Revert "mm/gup: remove try_get_page(), call try_get_compound_head() directly"

2021-09-07T18:03:45+00:00

This reverts commit 9857a17f206ff374aea78bccfb687f145368be2e. That commit was completely broken, and I should have caught on to it earlier. But happily, the kernel test robot noticed the breakage fairly quickly. The breakage is because "try_get_page()" is about avoiding the page reference count overflow case, but is otherwise the exact same as a plain "get_page()". In contrast, "try_get_compound_head()" is an entirely different beast, and uses __page_cache_add_speculative() because it's not just about the page reference count, but also about possibly racing with the underlying page going away. So all the commentary about how "try_get_page() has fallen a little behind in terms of maintenance, try_get_compound_head() handles speculative page references more thoroughly" was just completely wrong: yes, try_get_compound_head() handles speculative page references, but the point is that try_get_page() does not, and must not. So there's no lack of maintainance - there are fundamentally different semantics. A speculative page reference would be entirely wrong in "get_page()", and it's entirely wrong in "try_get_page()". It's not about speculation, it's purely about "uhhuh, you can't get this page because you've tried to increment the reference count too much already". The reason the kernel test robot noticed this bug was that it hit the VM_BUG_ON() in __page_cache_add_speculative(), which is all about verifying that the context of any speculative page access is correct. But since that isn't what try_get_page() is all about, the VM_BUG_ON() tests things that are not correct to test for try_get_page(). Reported-by: kernel test robot Cc: John Hubbard Cc: Christoph Hellwig Cc: Andrew Morton Signed-off-by: Linus Torvalds

mm/gup: remove try_get_page(), call try_get_compound_head() directly

2021-09-03T16:58:11+00:00

try_get_page() is very similar to try_get_compound_head(), and in fact try_get_page() has fallen a little behind in terms of maintenance: try_get_compound_head() handles speculative page references more thoroughly. There are only two try_get_page() callsites, so just call try_get_compound_head() directly from those, and remove try_get_page() entirely. Also, seeing as how this changes try_get_compound_head() into a non-static function, provide some kerneldoc documentation for it. Link: https://lkml.kernel.org/r/20210813044133.1536842-4-jhubbard@nvidia.com Signed-off-by: John Hubbard Reviewed-by: Christoph Hellwig Cc: Matthew Wilcox Cc: Christian Borntraeger Cc: Heiko Carstens Cc: Vasily Gorbik Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

pipe: do FASYNC notifications for every pipe IO, not just state changes

2021-08-25T17:27:16+00:00

It turns out that the SIGIO/FASYNC situation is almost exactly the same as the EPOLLET case was: user space really wants to be notified after every operation. Now, in a perfect world it should be sufficient to only notify user space on "state transitions" when the IO state changes (ie when a pipe goes from unreadable to readable, or from unwritable to writable). User space should then do as much as possible - fully emptying the buffer or what not - and we'll notify it again the next time the state changes. But as with EPOLLET, we have at least one case (stress-ng) where the kernel sent SIGIO due to the pipe being marked for asynchronous notification, but the user space signal handler then didn't actually necessarily read it all before returning (it read more than what was written, but since there could be multiple writes, it could leave data pending). The user space code then expected to get another SIGIO for subsequent writes - even though the pipe had been readable the whole time - and would only then read more. This is arguably a user space bug - and Colin King already fixed the stress-ng code in question - but the kernel regression rules are clear: it doesn't matter if kernel people think that user space did something silly and wrong. What matters is that it used to work. So if user space depends on specific historical kernel behavior, it's a regression when that behavior changes. It's on us: we were silly to have that non-optimal historical behavior, and our old kernel behavior was what user space was tested against. Because of how the FASYNC notification was tied to wakeup behavior, this was first broken by commits f467a6a66419 and 1b6b26ae7053 ("pipe: fix and clarify pipe read/write wakeup logic"), but at the time it seems nobody noticed. Probably because the stress-ng problem case ends up being timing-dependent too. It was then unwittingly fixed by commit 3a34b13a88ca ("pipe: make pipe writes always wake up readers") only to be broken again when by commit 3b844826b6c6 ("pipe: avoid unnecessary EPOLLET wakeups under normal loads"). And at that point the kernel test robot noticed the performance refression in the stress-ng.sigio.ops_per_sec case. So the "Fixes" tag below is somewhat ad hoc, but it matches when the issue was noticed. Fix it for good (knock wood) by simply making the kill_fasync() case separate from the wakeup case. FASYNC is quite rare, and we clearly shouldn't even try to use the "avoid unnecessary wakeups" logic for it. Link: https://lore.kernel.org/lkml/20210824151337.GC27667@xsang-OptiPlex-9020/ Fixes: 3b844826b6c6 ("pipe: avoid unnecessary EPOLLET wakeups under normal loads") Reported-by: kernel test robot Tested-by: Oliver Sang Cc: Eric Biederman Cc: Colin Ian King Signed-off-by: Linus Torvalds