Commit f9aceca3 authored by KY Srinivasan's avatar KY Srinivasan Committed by Greg Kroah-Hartman

hyperv: Fix a bug in netvsc_start_xmit()

[ Upstream commit dedb845d ]

After the packet is successfully sent, we should not touch the skb
as it may have been freed. This patch is based on the work done by
Long Li <longli@microsoft.com>.

In this version of the patch I have fixed issues pointed out by David.
David, please queue this up for stable.
Signed-off-by: default avatarK. Y. Srinivasan <kys@microsoft.com>
Tested-by: default avatarLong Li <longli@microsoft.com>
Tested-by: default avatarSitsofe Wheeler <sitsofe@yahoo.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
Signed-off-by: default avatarGreg Kroah-Hartman <gregkh@linuxfoundation.org>
parent 58345c2e
...@@ -138,6 +138,7 @@ static int netvsc_start_xmit(struct sk_buff *skb, struct net_device *net) ...@@ -138,6 +138,7 @@ static int netvsc_start_xmit(struct sk_buff *skb, struct net_device *net)
struct hv_netvsc_packet *packet; struct hv_netvsc_packet *packet;
int ret; int ret;
unsigned int i, num_pages, npg_data; unsigned int i, num_pages, npg_data;
u32 skb_length = skb->len;
/* Add multipages for skb->data and additional 2 for RNDIS */ /* Add multipages for skb->data and additional 2 for RNDIS */
npg_data = (((unsigned long)skb->data + skb_headlen(skb) - 1) npg_data = (((unsigned long)skb->data + skb_headlen(skb) - 1)
...@@ -208,7 +209,7 @@ static int netvsc_start_xmit(struct sk_buff *skb, struct net_device *net) ...@@ -208,7 +209,7 @@ static int netvsc_start_xmit(struct sk_buff *skb, struct net_device *net)
ret = rndis_filter_send(net_device_ctx->device_ctx, ret = rndis_filter_send(net_device_ctx->device_ctx,
packet); packet);
if (ret == 0) { if (ret == 0) {
net->stats.tx_bytes += skb->len; net->stats.tx_bytes += skb_length;
net->stats.tx_packets++; net->stats.tx_packets++;
} else { } else {
kfree(packet); kfree(packet);
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment