From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vincent Pelletier Subject: Re: r8169: IO_PAGE_FAULT & netdev watchdog Date: Sat, 2 Jun 2012 15:42:28 +0200 Message-ID: <201206021542.28869.plr.vincent@gmail.com> References: <201205312331.06623.plr.vincent@gmail.com> <201206021108.53551.plr.vincent@gmail.com> <20120602105645.GA28769@electric-eye.fr.zoreil.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: netdev@vger.kernel.org To: Francois Romieu Return-path: Received: from mail-wg0-f44.google.com ([74.125.82.44]:45157 "EHLO mail-wg0-f44.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1759650Ab2FBNmf convert rfc822-to-8bit (ORCPT ); Sat, 2 Jun 2012 09:42:35 -0400 Received: by wgbdr13 with SMTP id dr13so2736197wgb.1 for ; Sat, 02 Jun 2012 06:42:33 -0700 (PDT) In-Reply-To: <20120602105645.GA28769@electric-eye.fr.zoreil.com> Sender: netdev-owner@vger.kernel.org List-ID: Le samedi 02 juin 2012 12:56:45, vous avez =E9crit : > And partly because the patch I sent included its content in the commi= t > message as well. :o/ I noticed the repetition after trying to apply on 3.4, and dropped one.= And=20 only then realised it was really already applied. > If the inlined patch makes a difference, you should see it with 3.4. It made a difference, when testing with netcat: without any change over= =20 vanilla 3.3.7, network trafic drops to 0 in a matter of seconds (up to = around=20 10s). With it, it stayed stable for 10 minutes, until I killed nc. I reproduced this with 3.4 as well (no patch =3D bug, patch =3D no prob= lem). In both version without patch, I got the watchdog warning 10 minutes af= ter=20 traffic drop - though without the IO_PAGE_FAULT message. I spent quite some time testing with nc in UDP mode first, and couldn't= =20 reproduce the issue (then I switched to TCP as said above). Does that m= ake any=20 sense ? I also noticed the significant lag at bootup when eth0 is brought up is= much=20 reduced on patched kernel. Does that makes sense ? =46WIW, the commands I used were based on: nc -l -p 5555 < /dev/zero > /dev/null With/without -u flag, and of course client-side equivalent command so t= he=20 connection was used full-duplex at maximum speed: 450Mb/s in TCP, 800+M= b/s in=20 UDP, each way. UDP was limited by CPU on one side (~450Mb/s upload from= =20 that box, 800Mb/s download, 100% cpu on it). Values are as reported by nload & htop. All tests were done in runlevel 2, with rsyslog manually started with i= ts init=20 script. > My life is a bit easier when you work somewhere in the main branch > (or in davem's -next but it is not relevant for regression fixes). I'm not sure: does 3.4 tarball from kernel.org qualify as "main branch"= ?=20 Otherwise, which git repos & branch should I use ? Regards, --=20 Vincent Pelletier