io_uring: close a small race gap for files cancel

The purpose of io_uring_cancel_files() is to wait for all requests
matching ->files to go/be cancelled. We should first drop files of a
request in io_req_drop_files() and only then make it undiscoverable for
io_uring_cancel_files.

First drop, then delete from list. It's ok to leave req->id->files
dangling, because it's not dereferenced by cancellation code, only
compared against. It would potentially go to sleep and be awaken by
following in io_req_drop_files() wake_up().

Fixes: 0f2122045b946 ("io_uring: don't rely on weak ->files references")
Cc: <stable@vger.kernel.org> # 5.5+
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 8cf6f22..b749578 100644 (file)
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -6098,15 +6098,15 @@ static void io_req_drop_files(struct io_kiocb *req)
        struct io_uring_task *tctx = req->task->io_uring;
        unsigned long flags;
 
+       put_files_struct(req->work.identity->files);
+       put_nsproxy(req->work.identity->nsproxy);
        spin_lock_irqsave(&ctx->inflight_lock, flags);
        list_del(&req->inflight_entry);
-       if (atomic_read(&tctx->in_idle))
-               wake_up(&tctx->wait);
        spin_unlock_irqrestore(&ctx->inflight_lock, flags);
        req->flags &= ~REQ_F_INFLIGHT;
-       put_files_struct(req->work.identity->files);
-       put_nsproxy(req->work.identity->nsproxy);
        req->work.flags &= ~IO_WQ_WORK_FILES;
+       if (atomic_read(&tctx->in_idle))
+               wake_up(&tctx->wait);
 }
 
 static void __io_clean_op(struct io_kiocb *req)