io_uring: warn when ring exit takes too long

We use system_unbound_wq to run io_ring_exit_work(), so it's hard to
monitor whether removal hang or not. Add WARN_ONCE to catch hangs.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

diff --git a/fs/io_uring.c b/fs/io_uring.c
index 01a7fa4a4889e09822ae3419c66fbe8a103b6eb0..945e54690b8110b9196e647d288ea19a44c7f21a 100644 (file)
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -8560,6 +8560,7 @@ static void io_tctx_exit_cb(struct callback_head *cb)
 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;
        struct io_tctx_exit exit;
        struct io_tctx_node *node;
        int ret;
@@ -8572,10 +8573,14 @@ static void io_ring_exit_work(struct work_struct *work)
         */
        do {
                io_uring_try_cancel_requests(ctx, NULL, NULL);
+
+               WARN_ON_ONCE(time_after(jiffies, timeout));
        } while (!wait_for_completion_timeout(&ctx->ref_comp, HZ/20));
 
        mutex_lock(&ctx->uring_lock);
        while (!list_empty(&ctx->tctx_list)) {
+               WARN_ON_ONCE(time_after(jiffies, timeout));
+
                node = list_first_entry(&ctx->tctx_list, struct io_tctx_node,
                                        ctx_node);
                exit.ctx = ctx;