io_uring: don't submit sqes when ctx->refs is dying

When IORING_SETUP_SQPOLL is enabled, io_ring_ctx_wait_and_kill() will wait
for sq thread to idle by busy loop:

    while (ctx->sqo_thread && !wq_has_sleeper(&ctx->sqo_wait))
        cond_resched();

Above loop isn't very CPU friendly, it may introduce a short cpu burst on
the current cpu.

If ctx->refs is dying, we forbid sq_thread from submitting any further
SQEs. Instead they just get discarded when we exit.

Signed-off-by: Xiaoguang Wang <xiaoguang.wang@linux.alibaba.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

diff --git a/fs/io_uring.c b/fs/io_uring.c
index 50f0794..0b51f21 100644 (file)
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -6175,7 +6175,8 @@ static int io_sq_thread(void *data)
                }
 
                mutex_lock(&ctx->uring_lock);
-               ret = io_submit_sqes(ctx, to_submit, NULL, -1);
+               if (likely(!percpu_ref_is_dying(&ctx->refs)))
+                       ret = io_submit_sqes(ctx, to_submit, NULL, -1);
                mutex_unlock(&ctx->uring_lock);
                timeout = jiffies + ctx->sq_thread_idle;
        }
@@ -7465,16 +7466,6 @@ static void io_ring_ctx_wait_and_kill(struct io_ring_ctx *ctx)
        percpu_ref_kill(&ctx->refs);
        mutex_unlock(&ctx->uring_lock);
 
-       /*
-        * Wait for sq thread to idle, if we have one. It won't spin on new
-        * work after we've killed the ctx ref above. This is important to do
-        * before we cancel existing commands, as the thread could otherwise
-        * be queueing new work post that. If that's work we need to cancel,
-        * it could cause shutdown to hang.
-        */
-       while (ctx->sqo_thread && !wq_has_sleeper(&ctx->sqo_wait))
-               cond_resched();
-
        io_kill_timeouts(ctx);
        io_poll_remove_all(ctx);