io_uring: improve ctx hang handling

If io_ring_exit_work() can't get it done in 5 minutes, something is
going very wrong, don't keep spinning at HZ / 20 rate, it doesn't help
and it may take much of CPU time if there is a lot of workers stuck as
such.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/9e2d1ca81d569f6bc628af1a42ff6663bff7ce9c.1628471125.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
This commit is contained in:
Pavel Begunkov 2021-08-09 13:04:17 +01:00 коммит произвёл Jens Axboe
Родитель d3fddf6ddd
Коммит 58d3be2c60
1 изменённых файлов: 6 добавлений и 2 удалений

Просмотреть файл

@ -8795,6 +8795,7 @@ static void io_ring_exit_work(struct work_struct *work)
{
struct io_ring_ctx *ctx = container_of(work, struct io_ring_ctx, exit_work);
unsigned long timeout = jiffies + HZ * 60 * 5;
unsigned long interval = HZ / 20;
struct io_tctx_exit exit;
struct io_tctx_node *node;
int ret;
@ -8819,8 +8820,11 @@ static void io_ring_exit_work(struct work_struct *work)
io_sq_thread_unpark(sqd);
}
WARN_ON_ONCE(time_after(jiffies, timeout));
} while (!wait_for_completion_timeout(&ctx->ref_comp, HZ/20));
if (WARN_ON_ONCE(time_after(jiffies, timeout))) {
/* there is little hope left, don't run it too often */
interval = HZ * 60;
}
} while (!wait_for_completion_timeout(&ctx->ref_comp, interval));
init_completion(&exit.completion);
init_task_work(&exit.task_work, io_tctx_exit_cb);