xen/netfront: tolerate frags with no data

At least old Xen net backends seem to send frags with no real data
sometimes. In case such a fragment happens to occur with the frag limit
already reached the frontend will BUG currently even if this situation
is easily recoverable.

Modify the BUG_ON() condition accordingly.

Tested-by: Dietmar Hahn <dietmar.hahn@ts.fujitsu.com>
Signed-off-by: Juergen Gross <jgross@suse.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit is contained in:
Juergen Gross 2018-12-18 16:06:19 +01:00 committed by David S. Miller
parent 8742beb50f
commit d81c5054a5

View File

@ -905,7 +905,7 @@ static RING_IDX xennet_fill_frags(struct netfront_queue *queue,
if (skb_shinfo(skb)->nr_frags == MAX_SKB_FRAGS) { if (skb_shinfo(skb)->nr_frags == MAX_SKB_FRAGS) {
unsigned int pull_to = NETFRONT_SKB_CB(skb)->pull_to; unsigned int pull_to = NETFRONT_SKB_CB(skb)->pull_to;
BUG_ON(pull_to <= skb_headlen(skb)); BUG_ON(pull_to < skb_headlen(skb));
__pskb_pull_tail(skb, pull_to - skb_headlen(skb)); __pskb_pull_tail(skb, pull_to - skb_headlen(skb));
} }
if (unlikely(skb_shinfo(skb)->nr_frags >= MAX_SKB_FRAGS)) { if (unlikely(skb_shinfo(skb)->nr_frags >= MAX_SKB_FRAGS)) {