From mboxrd@z Thu Jan 1 00:00:00 1970 From: Li Yu Subject: Re: v3 for tcp friends? Date: Wed, 23 Jan 2013 11:14:26 +0800 Message-ID: <50FF5592.60008@gmail.com> References: <20120903.154833.1547153833820955116.davem@davemloft.net> <20120904.125841.2293649688957878987.davem@davemloft.net> <50FCEE64.80203@gmail.com> <50FCEEC5.9010404@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=GB2312 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: David Miller , netdev To: Bruce Curtis Return-path: Received: from mail-pb0-f54.google.com ([209.85.160.54]:63379 "EHLO mail-pb0-f54.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753328Ab3AWDPB (ORCPT ); Tue, 22 Jan 2013 22:15:01 -0500 Received: by mail-pb0-f54.google.com with SMTP id rr4so2522061pbb.13 for ; Tue, 22 Jan 2013 19:15:01 -0800 (PST) In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: =D3=DA 2013=C4=EA01=D4=C223=C8=D5 05:08, Bruce Curtis =D0=B4=B5=C0: > Thanks, Li >=20 > Started working on friends again, v4, more soon. >=20 >=20 :) I found another odd bug in TCP friends v3, the clients may hang at tcp_sendmsg() -> sk_stream_wait_memory() with or without my refcnt fix patch. Below shell script can reproduce this bug: #! /bin/sh -x sysctl -w net.ipv4.tcp_rmem=3D"4096 1073741824 1073741824" sysctl -w net.ipv4.tcp_wmem=3D"4096 1073741824 1073741824" sysctl -w net.ipv4.tcp_friends=3D1 msg=3D64K buf=3D256M pkill -9 netserver netserver netperf -t TCP_STREAM -l 1 -- -m ${msg} -M ${msg} -s ${buf} -S ${buf} -= 4 sysctl -w net.ipv4.tcp_friends=3D0 pkill -9 netserver netserver netperf -t TCP_STREAM -l 1 -- -m ${msg} -M ${msg} -s ${buf} -S ${buf} -= 4 ##################SCRIPT END################### netperf kernel stack is (by cat /proc/$netperf_pid/stack) [] sk_stream_wait_memory+0x2d9/0x2f0 [] tcp_sendmsg+0xf6c/0x1240 [] inet_sendmsg+0xf7/0x110 [] sock_sendmsg+0x7d/0xa0 [] sys_sendto+0x13d/0x190 [] system_call_fastpath+0x16/0x1b [] 0xffffffffffffffff netserver kernel stack is : [] sk_wait_data+0x8e/0xe0 [] tcp_recvmsg+0x5c3/0xbe0 [] inet_recvmsg+0xed/0x110 [] sock_recvmsg+0x84/0xb0 [] sys_recvfrom+0xee/0x170 [] system_call_fastpath+0x16/0x1b [] 0xffffffffffffffff And, "netstat -tnp" give us below results: Active Internet connections (w/o servers) Proto Recv-Q Send-Q Local Address Foreign Address State PID/Program name tcp 0 0 127.0.0.1:35451 127.0.0.1:12865 ESTABLISHED 2316/netperf tcp6 0 0 127.0.0.1:12865 127.0.0.1:35451 ESTABLISHED 2317/netserver (It seems that netperf hangs on the control connection of benchmark) I also try to fix this ... Thanks Yu > On Mon, Jan 21, 2013 at 12:55 AM, Li Yu > wrote: >=20 > 2013/1/21 Li Yu > >=20 > =D3=DA 2013=C4=EA01=D4=C221=C8=D5 15:29, Li Yu =D0=B4=B5=C0: >=20 > =D3=DA 2012=C4=EA09=D4=C205=C8=D5 00:58, David Miller =D0= =B4=B5=C0: >=20 > From: Bruce Curtis > > Date: Tue, 4 Sep 2012 08:10:23 -0700 >=20 > Will do, issues addressed, I'll get the patch out > later today or > tomorrow at the latest. >=20 >=20 > Thanks a lot Bruce. > -- > To unsubscribe from this list: send the line > "unsubscribe netdev" in > the body of a message to majordomo@vger.kernel.org > > More majordomo info at > http://vger.kernel.org/__majordomo-info.html > >=20 >=20 >=20 > Hi, Bruce, >=20 > I tested the TCP friends, found a bug here: >=20 > [ 106.541372] Pid: 1765, comm: client Not tainted > 3.7.0-rc1+ #25 > [ 106.543121] Call Trace: > [ 106.543950] [] > inet_sock_destruct+0x102/0x1f0 > [ 106.545687] [] __sk_free+0x1d/0x110 > [ 106.547209] [] sk_free+0x1c/0x20 > [ 106.548611] [] tcp_close+0x6c/0x3f0 > [ 106.549863] [] inet_release+0xda/0x= f0 > [ 106.551134] [] ? inet_release+0x20/= 0xf0 > [ 106.552419] [] ? mutex_unlock+0xe/0= x10 > [ 106.553658] [] sock_release+0x28/0x= a0 > [ 106.557366] [] sock_close+0x29/0x30 > [ 106.558831] [] __fput+0x122/0x210 > [ 106.560541] [] ____fput+0xe/0x10 > [ 106.562006] [] task_work_run+0x9e/0= xd0 > [ 106.563285] [] do_notify_resume+0x6= 1/0x70 > [ 106.564582] [] int_signal+0x12/0x17 >=20 >=20 > I also backported and tested it on stable kernel > 3.7.3/RHEL6 > kernel, this bug still exists. It seem that client may cl= ose > listening > sockets, may we need to add one reference count for liste= n > socket > before send its address to peer? >=20 >=20 > Sorry, I lost an important line of kernel log before above th= em: >=20 > [ 106.539367] IPv4: Attempt to release TCP socket in state 1= 0 > ffff880074abb5c0 >=20 > BTW: state 10 =3D TCP_LISTEN >=20 >=20 > It seem this patch works for me. >=20 > diff --git a/net/ipv4/inet_connection_sock.c > b/net/ipv4/inet_connection_sock.c > index 9641215..a625c02 100644 > --- a/net/ipv4/inet_connection_sock.c > +++ b/net/ipv4/inet_connection_sock.c > @@ -623,8 +623,11 @@ struct sock *inet_csk_clone(struct sock *sk, > const struct request_sock *req, > sock_hold(newsk); > was =3D xchg(&req->friend->sk_friend, ne= wsk); > /* If requester already connect()ed, may= be > sleeping */ > - if (was && !sock_flag(req->friend, SOCK_D= EAD)) > - sk->sk_state_change(req->friend); > + if (was) { > + if (!sock_flag(req->friend, SOCK_= DEAD)) > + =20 > sk->sk_state_change(req->friend); > + sock_put(was); > + } > } > newsk->sk_state =3D TCP_SYN_RECV; > diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c > index 5917485..7a63245 100644 > --- a/net/ipv4/tcp_output.c > +++ b/net/ipv4/tcp_output.c > @@ -2277,8 +2277,10 @@ struct sk_buff *tcp_make_synack(struct soc= k > *sk, struct dst_entry *dst, > memset(&opts, 0, sizeof(opts)); > /* Only try to make friends if enabled */ > - if (sysctl_tcp_friends) > + if (sysctl_tcp_friends) { > + sock_hold(sk); > skb->friend =3D sk; > + } > #ifdef CONFIG_SYN_COOKIES > if (unlikely(req->cookie_ts)) >=20 >=20 > And, our TCP friends v4? :) >=20 > Thanks >=20 > Yu >=20 >=20 >=20