Re: [PATCH net] virtio_net: fix return value check in receive_mergeable()

From: Jason Wang
Date: Mon Dec 04 2017 - 21:59:34 EST




On 2017å12æ04æ 19:25, wangyunjian wrote:

-----Original Message-----
From: Jason Wang [mailto:jasowang@xxxxxxxxxx]
Sent: Monday, December 04, 2017 3:10 PM
To: wangyunjian <wangyunjian@xxxxxxxxxx>; linux-
kernel@xxxxxxxxxxxxxxx
Cc: mst@xxxxxxxxxx; caihe <caihe@xxxxxxxxxx>
Subject: Re: [PATCH net] virtio_net: fix return value check in
receive_mergeable()



On 2017å12æ04æ 14:02, wangyunjian wrote:
From: Yunjian Wang <wangyunjian@xxxxxxxxxx>

The function virtqueue_get_buf_ctx() could return NULL, the return
value 'buf' need to be checked with NULL, not value 'ctx'.

Signed-off-by: Yunjian Wang <wangyunjian@xxxxxxxxxx>
---
drivers/net/virtio_net.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 19a985e..559b215 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -756,7 +756,7 @@ static struct sk_buff *receive_mergeable(struct
net_device *dev,
int num_skb_frags;

buf = virtqueue_get_buf_ctx(rq->vq, &len, &ctx);
- if (unlikely(!ctx)) {
+ if (unlikely(!buf)) {
pr_debug("%s: rx error: %d buffers out of %d
missing\n",
dev->name, num_buf,
virtio16_to_cpu(vi->vdev,
Hi:

The path looks good but I'm not sure this is a real fix since we add
data and ctx simultaneously in virtqueue_add_inbuf_ctx().

Any bad result you see without this patch?

Now no problem has been found yet, and I found it when reviewing the code.

Thanks

Then I think it's better to go net-next. Michael, any idea on this?

Thanks


Thanks