]> git.ipfire.org Git - thirdparty/kernel/linux.git/commit
fuse-uring: fix moving cancelled entry to ent_in_userspace list
authorJoanne Koong <joannelkoong@gmail.com>
Mon, 8 Jun 2026 19:21:49 +0000 (12:21 -0700)
committerMiklos Szeredi <mszeredi@redhat.com>
Mon, 15 Jun 2026 12:06:14 +0000 (14:06 +0200)
commit198f45eeb9f78b2a2d6d8be95e4e43468eb2c6bc
treed326bd986f067aa951d90a519b5166d8efbf1dd4
parentc146284c4355e96550e50e8f6e694223d107b621
fuse-uring: fix moving cancelled entry to ent_in_userspace list

fuse_uring_cancel() moves entries that are available (these have no reqs
attached) to the ent_in_userspace list. ent_list_request_expired()
checks the first entry on ent_in_userspace and dereferences
ent->fuse_req unconditionally, which will crash on a cancelled entry
that was moved to this list.

Fix this by freeing the entry and dropping queue_refs directly in
fuse_uring_cancel(). This is safe because cancel is the cancel handler
itself - after io_uring_cmd_done(), no more cancels will be dispatched
for this command, and teardown serializes with cancel via queue->lock.

Since cancel now decrements queue_refs, fuse_uring_abort() must no
longer gate fuse_uring_abort_end_requests() on queue_refs > 0, as
cancelled entries may have already dropped queue_refs while requests are
still queued. Remove the gate so abort always flushes requests and stops
queues.

Reported-by: Heechan Kang <gganji11@naver.com>
Tested-by: Heechan Kang <gganji11@naver.com>
Reviewed-by: Bernd Schubert <bernd@bsbernd.com>
Fixes: 4fea593e625c ("fuse: optimize over-io-uring request expiration check")
Cc: stable@vger.kernel.org
Suggested-by: Jian Huang Li <ali@ddn.com>
Suggested-by: Horst Birthelmer <horst@birthelmer.de>
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
fs/fuse/dev_uring.c
fs/fuse/dev_uring_i.h