From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andy Furniss Subject: Re: WARNING: at net/ipv4/tcp_input.c Date: Sat, 03 May 2008 13:03:59 +0100 Message-ID: <481C54AF.1050109@andyfurniss.entadsl.com> References: <4815BCDF.6070704@andyfurniss.entadsl.com> <48162070.4030405@andyfurniss.entadsl.com> <481986FD.8070108@andyfurniss.entadsl.com> <48199342.9010406@andyfurniss.entadsl.com> Reply-To: lists@andyfurniss.entadsl.com Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Netdev To: =?ISO-8859-1?Q?Ilpo_J=E4rvinen?= Return-path: Received: from mx1.ukfsn.org ([77.75.108.10]:57090 "EHLO mail.ukfsn.org" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1756126AbYECMDy (ORCPT ); Sat, 3 May 2008 08:03:54 -0400 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: Ilpo J=E4rvinen wrote: > I suppose it flooded the log so that head of the log got lost already= =20 > before userspace could consume it... (I think it's ringbuffer or some= thing=20 > though I'm not sure), it might be wise to enlarge it (I think it's=20 > controlled through CONFIG_LOG_BUF_SHIFT). >=20 > Since I've not yet found any other path that could lead to it except = the=20 > yesterday's false-positive, getting the write queue state more accura= tely=20 > captured by the debug patch on the very first occassion might help. H= ere's=20 > less spammy version of the debug patch (I added one printout as well = to=20 > get more usefulness to transitional state reported). >=20 OK, increased CONFIG_LOG_BUF_SHIFT from 14 to 18 and reversed old /=20 applied new patch. Andy.