Hi Keir:
[NET] gso: Fix rcv mss estimate
I noticed that with default TCP window sizes TSO + Xen would slow to a
crawl on certain machines. It turned out that there is a bug in the TCP
stack when it comes to receiving LRO (counter part of TSO for rx) packets.
The following fix has been applied upstream.
[TCP]: Fix rcv mss estimate for LRO
By passing a Linux-generated TSO packet straight back into Linux, Xen
becomes our first LRO user :) Unfortunately, there is at least one spot
in our stack that needs to be changed to cope with this.
The receive MSS estimate is computed from the raw packet size. This is
broken if the packet is GSO/LRO. Fortunately the real MSS can be found
in gso_size so we simply need to use that if it is non-zero.
Real LRO NICs should of course set the gso_size field in future.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Cheers,
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
--
diff -r 01a88ecec701 patches/linux-2.6.16.13/net-gso-5-rcv-mss.patch
--- /dev/null Thu Jan 01 00:00:00 1970 +0000
+++ b/patches/linux-2.6.16.13/net-gso-5-rcv-mss.patch Fri Sep 01 20:53:44 2006
+1000
@@ -0,0 +1,13 @@
+diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
+index 104af5d..1fa1536 100644
+--- a/net/ipv4/tcp_input.c
++++ b/net/ipv4/tcp_input.c
+@@ -127,7 +127,7 @@ static void tcp_measure_rcv_mss(struct s
+ /* skb->len may jitter because of SACKs, even if peer
+ * sends good full-sized frames.
+ */
+- len = skb->len;
++ len = skb_shinfo(skb)->gso_size ?: skb->len;
+ if (len >= icsk->icsk_ack.rcv_mss) {
+ icsk->icsk_ack.rcv_mss = len;
+ } else {
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel