diff options
| author | Pavel Begunkov <asml.silence@gmail.com> | 2021-04-13 11:43:00 +0100 |
|---|---|---|
| committer | Jens Axboe <axboe@kernel.dk> | 2021-04-14 13:07:27 -0600 |
| commit | c7d95613c7d6e003969722a290397b8271bdad17 (patch) | |
| tree | 5eddfc6c45f59cea3351e2b0e3ffc3adb5cb2d1e /fs | |
| parent | c60eb049f4a19ddddcd3ee97a9c79ab8066a6a03 (diff) | |
io_uring: fix early sqd_list removal sqpoll hangs
[ 245.463317] INFO: task iou-sqp-1374:1377 blocked for more than 122 seconds.
[ 245.463334] task:iou-sqp-1374 state:D flags:0x00004000
[ 245.463345] Call Trace:
[ 245.463352] __schedule+0x36b/0x950
[ 245.463376] schedule+0x68/0xe0
[ 245.463385] __io_uring_cancel+0xfb/0x1a0
[ 245.463407] do_exit+0xc0/0xb40
[ 245.463423] io_sq_thread+0x49b/0x710
[ 245.463445] ret_from_fork+0x22/0x30
It happens when sqpoll forgot to run park_task_work and goes to exit,
then exiting user may remove ctx from sqd_list, and so corresponding
io_sq_thread() -> io_uring_cancel_sqpoll() won't be executed. Hopefully
it just stucks in do_exit() in this case.
Fixes: dbe1bdbb39db ("io_uring: handle signals for IO threads like a normal thread")
Reported-by: Joakim Hassila <joj@mac.com>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Diffstat (limited to 'fs')
| -rw-r--r-- | fs/io_uring.c | 7 |
1 files changed, 5 insertions, 2 deletions
diff --git a/fs/io_uring.c b/fs/io_uring.c index bd14327c8e7e..dff34975d86b 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -6754,6 +6754,9 @@ static int io_sq_thread(void *data) current->flags |= PF_NO_SETAFFINITY; mutex_lock(&sqd->lock); + /* a user may had exited before the thread started */ + io_run_task_work_head(&sqd->park_task_work); + while (!test_bit(IO_SQ_THREAD_SHOULD_STOP, &sqd->state)) { int ret; bool cap_entries, sqt_spin, needs_sched; @@ -6770,10 +6773,10 @@ static int io_sq_thread(void *data) } cond_resched(); mutex_lock(&sqd->lock); - if (did_sig) - break; io_run_task_work(); io_run_task_work_head(&sqd->park_task_work); + if (did_sig) + break; timeout = jiffies + sqd->sq_thread_idle; continue; } |