| Message ID | 1467019279.6850.145.camel@edumazet-glaptop3.roam.corp.google.com |
|---|---|
| State | RFC, archived |
| Delegated to: | David Miller |
| Headers | show |
On Monday, June 06/27/16, 2016 at 14:51:19 +0530, Eric Dumazet wrote: > On Mon, 2016-06-27 at 14:29 +0530, Arjun V wrote: > > On Monday, June 06/27/16, 2016 at 13:36:22 +0530, Eric Dumazet wrote: > > > On Mon, 2016-06-27 at 12:47 +0530, Arjun V wrote: > > > > > > > Eric, > > > > > > > > Thanks for the quick patch. It didn't help much. We still see packets larger than 65536 with the patch. > > > > > > > > Below is the dump_stack() trace I am seeing for packets larger than 65536 in our xmit routine: > > > > > > What values do you get for skb->len ? > > > > > These are the skb->len values I am hitting in t4_eth_xmit() for around half hour run of netperf bidi(800 connections total) > > packet with skb->len 69570 detected > > packet with skb->len 66674 detected > > Ok please add : > > diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c > index 5c7ed147449c1b7ba029b12e033ad779a631460a..39bd2a2b8f4d931c514fe62a00541efef29157ad 100644 > --- a/net/ipv4/tcp.c > +++ b/net/ipv4/tcp.c > @@ -845,6 +845,7 @@ static unsigned int tcp_xmit_size_goal(struct sock *sk, u32 mss_now, > { > struct tcp_sock *tp = tcp_sk(sk); > u32 new_size_goal, size_goal; > + u32 res; > > if (!large_allowed || !sk_can_gso(sk)) > return mss_now; > @@ -862,7 +863,11 @@ static unsigned int tcp_xmit_size_goal(struct sock *sk, u32 mss_now, > size_goal = tp->gso_segs * mss_now; > } > > - return max(size_goal, mss_now); > + res = max(size_goal, mss_now); > + if (res >= 65536) > + pr_err_once("tcp_xmit_size_goal() -> %u size_goal %u, mss_now %u gso_max_size %u gso_segs %u\n", > + res, size_goal, mss_now, sk->sk_gso_max_size, tp->gso_segs); > + return res; > } > > static int tcp_send_mss(struct sock *sk, int *size_goal, int flags) > > Applied your patch. The above debug print is not getting invoked, when skb->len is greater than 65536.
On Mon, 2016-06-27 at 15:24 +0530, Arjun V wrote: > Applied your patch. > The above debug print is not getting invoked, when skb->len is greater than 65536. Interesting. It looks like we might have a bug in tcp_shift_skb_data() not respecting sk->sk_gso_max_size at all ... I will send a patch. Thanks.
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 5c7ed147449c1b7ba029b12e033ad779a631460a..39bd2a2b8f4d931c514fe62a00541efef29157ad 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -845,6 +845,7 @@ static unsigned int tcp_xmit_size_goal(struct sock *sk, u32 mss_now, { struct tcp_sock *tp = tcp_sk(sk); u32 new_size_goal, size_goal; + u32 res; if (!large_allowed || !sk_can_gso(sk)) return mss_now; @@ -862,7 +863,11 @@ static unsigned int tcp_xmit_size_goal(struct sock *sk, u32 mss_now, size_goal = tp->gso_segs * mss_now; } - return max(size_goal, mss_now); + res = max(size_goal, mss_now); + if (res >= 65536) + pr_err_once("tcp_xmit_size_goal() -> %u size_goal %u, mss_now %u gso_max_size %u gso_segs %u\n", + res, size_goal, mss_now, sk->sk_gso_max_size, tp->gso_segs); + return res; } static int tcp_send_mss(struct sock *sk, int *size_goal, int flags)