diff mbox series

io_uring: do not recycle buffer in READV

Message ID 20220721131325.624788-1-dylany@fb.com (mailing list archive)
State New
Headers show
Series io_uring: do not recycle buffer in READV | expand

Commit Message

Dylan Yudaken July 21, 2022, 1:13 p.m. UTC
READV cannot recycle buffers as it would lose some of the data required to
reimport that buffer.

Reported-by: Ammar Faizi <ammarfaizi2@gnuweeb.org>
Fixes: b66e65f41426 ("io_uring: never call io_buffer_select() for a buffer re-select")
Signed-off-by: Dylan Yudaken <dylany@fb.com>
---

I think going forward we can probably re-enable recycling for READV, or perhaps stick
it in the opdef to make it a bit more general. However since it is late in the merge
cycle I thought the simplest approach is best.

Worth noting the initial discussed approach of stashing the data in `struct io_rw` would not
have worked as that struct is full already apparently.

Dylan

 fs/io_uring.c | 8 ++++++++
 1 file changed, 8 insertions(+)


base-commit: ff6992735ade75aae3e35d16b17da1008d753d28

Comments

Jens Axboe July 21, 2022, 2:31 p.m. UTC | #1
On Thu, 21 Jul 2022 06:13:25 -0700, Dylan Yudaken wrote:
> READV cannot recycle buffers as it would lose some of the data required to
> reimport that buffer.
> 
> 

Applied, thanks!

[1/1] io_uring: do not recycle buffer in READV
      commit: 934447a603b22d98f45a679115d8402e1efdd0f7

Best regards,
diff mbox series

Patch

diff --git a/fs/io_uring.c b/fs/io_uring.c
index a01ea49f3017..b0180679584f 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -1737,6 +1737,14 @@  static void io_kbuf_recycle(struct io_kiocb *req, unsigned issue_flags)
 	    (req->flags & REQ_F_PARTIAL_IO))
 		return;
 
+	/*
+	 * READV uses fields in `struct io_rw` (len/addr) to stash the selected
+	 * buffer data. However if that buffer is recycled the original request
+	 * data stored in addr is lost. Therefore forbid recycling for now.
+	 */
+	if (req->opcode == IORING_OP_READV)
+		return;
+
 	/*
 	 * We don't need to recycle for REQ_F_BUFFER_RING, we can just clear
 	 * the flag and hence ensure that bl->head doesn't get incremented.