[PATCH 1/3] vhost: fix skb leak in handle_rx()

2017-12-01 Thread wexu
From: Wei Xu Matthew found a roughly 40% tcp throughput regression with commit c67df11f(vhost_net: try batch dequing from skb array) as discussed in the following thread: https://www.mail-archive.com/netdev@vger.kernel.org/msg187936.html Eventually we figured out that it was a skb leak in handle

[PATCH 2/3] tun: free skb in early errors

2017-12-01 Thread wexu
From: Wei Xu tun_recvmsg() supports accepting skb by msg_control after commit ac77cfd4258f ("tun: support receiving skb through msg_control"), the skb if presented should be freed no matter how far it can go along, otherwise it would be leaked. This patch fixes several missed cases. Signed-off-

[PATCH 3/3] tap: free skb if flags error

2017-12-01 Thread wexu
From: Wei Xu tap_recvmsg() supports accepting skb by msg_control after commit 3b4ba04acca8 ("tap: support receiving skb from msg_control"), the skb if presented should be freed within the function, otherwise it would be leaked. Signed-off-by: Wei Xu Reported-by: Matthew Rosato --- drivers/ne

[PATCH net,stable v4 0/3] vhost: fix a few skb leaks

2017-12-01 Thread wexu
From: Wei Xu Matthew found a roughly 40% tcp throughput regression with commit c67df11f(vhost_net: try batch dequing from skb array) as discussed in the following thread: https://www.mail-archive.com/netdev@vger.kernel.org/msg187936.html v4: - fix zero iov iterator count in tap/tap_do_read()(Jas

[PATCH net,stable v3] vhost: fix a few skb leaks

2017-11-30 Thread wexu
From: Wei Xu Matthew found a roughly 40% tcp throughput regression with commit c67df11f(vhost_net: try batch dequing from skb array) as discussed in the following thread: https://www.mail-archive.com/netdev@vger.kernel.org/msg187936.html This is v3. v3: - move freeing skb from vhost to tun/tap

[PATCH 2/3] tun: free skb in early errors

2017-11-30 Thread wexu
From: Wei Xu tun_recvmsg() supports accepting skb by msg_control after commit ac77cfd4258f ("tun: support receiving skb through msg_control"), the skb if presented should be freed within the function, otherwise it would be leaked. Signed-off-by: Wei Xu Reported-by: Matthew Rosato --- drivers/

[PATCH 1/3] vhost: fix skb leak in handle_rx()

2017-11-30 Thread wexu
From: Wei Xu Matthew found a roughly 40% tcp throughput regression with commit c67df11f(vhost_net: try batch dequing from skb array) as discussed in the following thread: https://www.mail-archive.com/netdev@vger.kernel.org/msg187936.html Eventually we figured out that it was a skb leak in handle

[PATCH 3/3] tap: free skb if flags error

2017-11-30 Thread wexu
From: Wei Xu tap_recvmsg() supports accepting skb by msg_control after commit 3b4ba04acca8 ("tap: support receiving skb from msg_control"), the skb if presented should be freed within the function, otherwise it would be leaked. Signed-off-by: Wei Xu Reported-by: Matthew Rosato --- drivers/net

[PATCH net,stable v2] vhost: fix skb leak in handle_rx()

2017-11-29 Thread wexu
From: Wei Xu Matthew found a roughly 40% tcp throughput regression with commit c67df11f(vhost_net: try batch dequing from skb array) as discussed in the following thread: https://www.mail-archive.com/netdev@vger.kernel.org/msg187936.html Eventually we figured out that it was a skb leak in handle

[PATCH net,stable] vhost: fix skb leak in handle_rx()

2017-11-28 Thread wexu
From: Wei Xu Matthew found a roughly 40% tcp throughput regression with commit c67df11f(vhost_net: try batch dequing from skb array) as discussed in the following thread: https://www.mail-archive.com/netdev@vger.kernel.org/msg187936.html Eventually we figured out that it was a skb leak in handle