Skip to content

Commit b8be3ae

Browse files
axboegregkh
authored andcommitted
io_uring/net: always use current transfer count for buffer put
A previous fix corrected the retry condition for when to continue a current bundle, but it missed that the current (not the total) transfer count also applies to the buffer put. If not, then for incrementally consumed buffer rings repeated completions on the same request may end up over consuming. Reported-by: Roy Tang (ErgoniaTrading) <royonia@ergonia.io> Cc: stable@vger.kernel.org Fixes: 3a08988 ("io_uring/net: only retry recv bundle for a full transfer") Link: axboe/liburing#1423 Signed-off-by: Jens Axboe <axboe@kernel.dk> (cherry picked from commit 51a4598) Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
1 parent 725fcba commit b8be3ae

File tree

1 file changed

+1
-1
lines changed

1 file changed

+1
-1
lines changed

io_uring/net.c

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -854,7 +854,7 @@ static inline bool io_recv_finish(struct io_kiocb *req, int *ret,
854854
if (sr->flags & IORING_RECVSEND_BUNDLE) {
855855
size_t this_ret = *ret - sr->done_io;
856856

857-
cflags |= io_put_kbufs(req, *ret, io_bundle_nbufs(kmsg, this_ret),
857+
cflags |= io_put_kbufs(req, this_ret, io_bundle_nbufs(kmsg, this_ret),
858858
issue_flags);
859859
if (sr->retry)
860860
cflags = req->cqe.flags | (cflags & CQE_F_MASK);

0 commit comments

Comments
 (0)