[PATCH v2] net: tcp6: fix double call of tcp_v6_fill_cb()

[PATCH v2] net: tcp6: fix double call of tcp_v6_fill_cb()

[Alexey Kodanev]

Regression from the following commit: 2dc49d1680.

tcp_v6_fill_cb() will be called twice if socket's state changes from
TCP_TIME_WAIT to TCP_LISTEN. That can result in control buffer data
corruption because in the second tcp_v6_fill_cb() call it's not copying
IP6CB(skb) anymore, but 'seq', 'end_seq', etc., so we can get weird and
unpredictable results. Performance loss of up to 1200% has been observed
in LTP/vxlan03 test.

This can be fixed by copying inet6_skb_parm to the beginning of 'cb'
only if xfrm6_policy_check() and tcp_v6_fill_cb() are going to be
called again.

Signed-off-by: Alexey Kodanev <alexey.kodanev@oracle.com>
---
 net/ipv6/tcp_ipv6.c |   11 +++++++++++
 1 files changed, 11 insertions(+), 0 deletions(-)

diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 5d46832..7d21a78 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1411,6 +1411,15 @@ static void tcp_v6_fill_cb(struct sk_buff *skb, const struct ipv6hdr *hdr,
 	TCP_SKB_CB(skb)->sacked = 0;
 }
 
+static void tcp_v6_restore_cb(struct sk_buff *skb)
+{
+	/* We need to move header back to the beginning if xfrm6_policy_check()
+	 * and tcp_v6_fill_cb() are going to be called again.
+	 */
+	memmove(IP6CB(skb), &TCP_SKB_CB(skb)->header.h6,
+		sizeof(struct inet6_skb_parm));
+}
+
 static int tcp_v6_rcv(struct sk_buff *skb)
 {
 	const struct tcphdr *th;
@@ -1543,6 +1552,7 @@ do_time_wait:
 			inet_twsk_deschedule(tw, &tcp_death_row);
 			inet_twsk_put(tw);
 			sk = sk2;
+			tcp_v6_restore_cb(skb);
 			goto process;
 		}
 		/* Fall through to ACK */
@@ -1551,6 +1561,7 @@ do_time_wait:
 		tcp_v6_timewait_ack(sk, skb);
 		break;
 	case TCP_TW_RST:
+		tcp_v6_restore_cb(skb);
 		goto no_tcp_socket;
 	case TCP_TW_SUCCESS:
 		;
-- 
1.7.1

Re: [PATCH v2] net: tcp6: fix double call of tcp_v6_fill_cb()

[Eric Dumazet]

On Thu, 2015-03-26 at 11:28 +0300, Alexey Kodanev wrote:
> Regression from the following commit: 2dc49d1680.
> 
> tcp_v6_fill_cb() will be called twice if socket's state changes from
> TCP_TIME_WAIT to TCP_LISTEN. That can result in control buffer data
> corruption because in the second tcp_v6_fill_cb() call it's not copying
> IP6CB(skb) anymore, but 'seq', 'end_seq', etc., so we can get weird and
> unpredictable results. Performance loss of up to 1200% has been observed
> in LTP/vxlan03 test.
> 
> This can be fixed by copying inet6_skb_parm to the beginning of 'cb'
> only if xfrm6_policy_check() and tcp_v6_fill_cb() are going to be
> called again.
> 
> Signed-off-by: Alexey Kodanev <alexey.kodanev@oracle.com>
> ---

Thanks Alexey !

Fixes: 2dc49d1680b53 ("tcp6: don't move IP6CB before xfrm6_policy_check()")
Acked-by: Eric Dumazet <edumazet@google.com>

Re: [PATCH v2] net: tcp6: fix double call of tcp_v6_fill_cb()

[Sergei Shtylyov]

Hello.

On 3/26/2015 11:28 AM, Alexey Kodanev wrote:

> Regression from the following commit: 2dc49d1680.

    Please also specify that commit's summary line in parens.

> tcp_v6_fill_cb() will be called twice if socket's state changes from
> TCP_TIME_WAIT to TCP_LISTEN. That can result in control buffer data
> corruption because in the second tcp_v6_fill_cb() call it's not copying
> IP6CB(skb) anymore, but 'seq', 'end_seq', etc., so we can get weird and
> unpredictable results. Performance loss of up to 1200% has been observed
> in LTP/vxlan03 test.

> This can be fixed by copying inet6_skb_parm to the beginning of 'cb'
> only if xfrm6_policy_check() and tcp_v6_fill_cb() are going to be
> called again.

> Signed-off-by: Alexey Kodanev <alexey.kodanev@oracle.com>

[...]

WBR, Sergei