io_uring: annotate offset timeout races
authorPavel Begunkov <asml.silence@gmail.com>
Fri, 19 May 2023 14:21:16 +0000 (15:21 +0100)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Fri, 11 Aug 2023 10:08:24 +0000 (12:08 +0200)
commit88b1958fb57dd99406fd9f4ba4c2daf5e01168b8
tree037ab0026a12a2864a47b707328eb5e09ac37c90
parenta78a8bcdc26de5ef3a0ee27c9c6c512e54a6051c
io_uring: annotate offset timeout races

commit 5498bf28d8f2bd63a46ad40f4427518615fb793f upstream.

It's racy to read ->cached_cq_tail without taking proper measures
(usually grabbing ->completion_lock) as timeout requests with CQE
offsets do, however they have never had a good semantics for from
when they start counting. Annotate racy reads with data_race().

Reported-by: syzbot+cb265db2f3f3468ef436@syzkaller.appspotmail.com
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/4de3685e185832a92a572df2be2c735d2e21a83d.1684506056.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
io_uring/timeout.c