[PATCH AUTOSEL 5.7 258/274] io_uring: fix overflowed reqs cancellation

[Sasha Levin]

From: Pavel Begunkov <asml.sile...@gmail.com>

[ Upstream commit 7b53d59859bc932b37895d2d37388e7fa29af7a5 ]

Overflowed requests in io_uring_cancel_files() should be shed only of
inflight and overflowed refs. All other left references are owned by
someone else.

If refcount_sub_and_test() fails, it will go further and put put extra
ref, don't do that. Also, don't need to do io_wq_cancel_work()
for overflowed reqs, they will be let go shortly anyway.

Signed-off-by: Pavel Begunkov <asml.sile...@gmail.com>
Signed-off-by: Jens Axboe <ax...@kernel.dk>
Signed-off-by: Sasha Levin <sas...@kernel.org>
---
 fs/io_uring.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/fs/io_uring.c b/fs/io_uring.c
index dd90c3fcd4f5..4038ed0a5c39 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -7467,10 +7467,11 @@ static void io_uring_cancel_files(struct io_ring_ctx 
*ctx,
                                finish_wait(&ctx->inflight_wait, &wait);
                                continue;
                        }
+               } else {
+                       io_wq_cancel_work(ctx->io_wq, &cancel_req->work);
+                       io_put_req(cancel_req);
                }
 
-               io_wq_cancel_work(ctx->io_wq, &cancel_req->work);
-               io_put_req(cancel_req);
                schedule();
                finish_wait(&ctx->inflight_wait, &wait);
        }
-- 
2.25.1