io_uring: don't do flush cancel under inflight_lock

We can't safely cancel under the inflight lock. If the work hasn't been
started yet, then io_wq_cancel_work() simply marks the work as cancelled
and invokes the work handler. But if the work completion needs to grab
the inflight lock because it's grabbing user files, then we'll deadlock
trying to finish the work as we already hold that lock.

Instead grab a reference to the request, if it isn't already zero. If
it's zero, then we know it's going through completion anyway, and we
can safely ignore it. If it's not zero, then we can drop the lock and
attempt to cancel from there.

This also fixes a missing finish_wait() at the end of
io_uring_cancel_files().

Signed-off-by: Jens Axboe <axboe@kernel.dk>

diff --git a/fs/io_uring.c b/fs/io_uring.c
index 912d2648f8db7ef0f8526bc8a9a945e9b7a68e82..dcb0602c9fd2b0eb3193579473eea6a1c28e75d0 100644 (file)
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -4257,33 +4257,34 @@ static void io_uring_cancel_files(struct io_ring_ctx *ctx,
 
        while (!list_empty_careful(&ctx->inflight_list)) {
                enum io_wq_cancel ret = IO_WQ_CANCEL_NOTFOUND;
+               struct io_kiocb *cancel_req = NULL;
 
                spin_lock_irq(&ctx->inflight_lock);
                list_for_each_entry(req, &ctx->inflight_list, inflight_entry) {
-                       if (req->work.files == files) {
-                               ret = io_wq_cancel_work(ctx->io_wq, &req->work);
-                               break;
-                       }
+                       if (req->work.files != files)
+                               continue;
+                       /* req is being completed, ignore */
+                       if (!refcount_inc_not_zero(&req->refs))
+                               continue;
+                       cancel_req = req;
+                       break;
                }
-               if (ret == IO_WQ_CANCEL_RUNNING)
+               if (cancel_req)
                        prepare_to_wait(&ctx->inflight_wait, &wait,
-                                       TASK_UNINTERRUPTIBLE);
-
+                                               TASK_UNINTERRUPTIBLE);
                spin_unlock_irq(&ctx->inflight_lock);
 
-               /*
-                * We need to keep going until we get NOTFOUND. We only cancel
-                * one work at the time.
-                *
-                * If we get CANCEL_RUNNING, then wait for a work to complete
-                * before continuing.
-                */
-               if (ret == IO_WQ_CANCEL_OK)
-                       continue;
-               else if (ret != IO_WQ_CANCEL_RUNNING)
+               if (cancel_req) {
+                       ret = io_wq_cancel_work(ctx->io_wq, &cancel_req->work);
+                       io_put_req(cancel_req);
+               }
+
+               /* We need to keep going until we don't find a matching req */
+               if (!cancel_req)
                        break;
                schedule();
        }
+       finish_wait(&ctx->inflight_wait, &wait);
 }
 
 static int io_uring_flush(struct file *file, void *data)