Skip to contentSkip to conversation

io_uring: mark reissue requests with REQ_F_PARTIAL_IO

PATCH New Jens Axboe 1 participant 1 message Jun 20, 2022

Subsystems (2)
lore
Share & tools
lore

Jens Axboe

author maintainer
#0 lore raw

PATCH 1/1 io_uring: mark reissue requests with REQ_F_PARTIAL_IO

If we mark for reissue, we assume that the buffer will remain stable. Hence if are using a provided buffer, we need to ensure that we stick with it for the duration of that request.

This only affects block devices that use provided buffers, as those are the only ones that get marked with REQ_F_REISSUE.

---

fs/io_uring.c +2 -2
index d3ee4fc532fa..87c65a358678 100644
@@ -3437,7 +3437,7 @@ static bool __io_complete_rw_common(struct io_kiocb *req, long res)
 	if (unlikely(res != req->cqe.res)) {
 		if ((res == -EAGAIN || res == -EOPNOTSUPP) &&
 		    io_rw_should_reissue(req)) {
-			req->flags |= REQ_F_REISSUE;
+			req->flags |= REQ_F_REISSUE | REQ_F_PARTIAL_IO;
 			return true;
 		}
 		req_set_fail(req);
@@ -3487,7 +3487,7 @@ static void io_complete_rw_iopoll(struct kiocb *kiocb, long res)
 		kiocb_end_write(req);
 	if (unlikely(res != req->cqe.res)) {
 		if (res == -EAGAIN && io_rw_should_reissue(req)) {
-			req->flags |= REQ_F_REISSUE;
+			req->flags |= REQ_F_REISSUE | REQ_F_PARTIAL_IO;
 			return;
 		}
 		req->cqe.res = res;
Signature · 1 line
Jens Axboe