| Message ID | 7225a4d1-a499-42e1-83a1-87223e83ba14@kernel.dk (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | io_uring/rw: return IOU_ISSUE_SKIP_COMPLETE for multishot retry | expand |
diff --git a/io_uring/rw.c b/io_uring/rw.c index 47e097ab5d7e..0585ebcc9773 100644 --- a/io_uring/rw.c +++ b/io_uring/rw.c @@ -947,6 +947,8 @@ int io_read_mshot(struct io_kiocb *req, unsigned int issue_flags) */ if (io_kbuf_recycle(req, issue_flags)) rw->len = 0; + if (issue_flags & IO_URING_F_MULTISHOT) + return IOU_ISSUE_SKIP_COMPLETE; return -EAGAIN; }
If read multishot is being invoked from the poll retry handler, then we should return IOU_ISSUE_SKIP_COMPLETE rather than -EAGAIN. If not, then a CQE will be posted with -EAGAIN rather than triggering the retry when the file is flagged as readable again. Cc: stable@vger.kernel.org Reported-by: Sargun Dhillon <sargun@meta.com> Fixes: fc68fcda04910 ("io_uring/rw: add support for IORING_OP_READ_MULTISHOT") Signed-off-by: Jens Axboe <axboe@kernel.dk> ---