From mboxrd@z Thu Jan 1 00:00:00 1970 From: Patrick McHardy Subject: Re: [NET_SCHED 00/04]: External SFQ classifiers/flow classifier Date: Fri, 04 Apr 2008 12:45:44 +0200 Message-ID: <47F606D8.108@trash.net> References: <20080131175758.25151.20370.sendpatchset@localhost.localdomain> <47A4FB81.80700@fatooh.org> <47A74FEE.6080103@trash.net> <47EED247.50801@andyfurniss.entadsl.com> <47F22CF5.4080806@trash.net> <47F28758.1010401@andyfurniss.entadsl.com> <47F37139.9080303@andyfurniss.entadsl.com> <47F37E04.4010001@trash.net> <47F3B3C5.6000701@andyfurniss.entadsl.com> <47F6062E.9050804@andyfurniss.entadsl.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: 7bit Cc: Corey Hickey , Linux Netdev List To: lists@andyfurniss.entadsl.com Return-path: Received: from stinky.trash.net ([213.144.137.162]:33093 "EHLO stinky.trash.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752667AbYDDKpp (ORCPT ); Fri, 4 Apr 2008 06:45:45 -0400 In-Reply-To: <47F6062E.9050804@andyfurniss.entadsl.com> Sender: netdev-owner@vger.kernel.org List-ID: Andy Furniss wrote: > Andy Furniss wrote: > >> OK - FWIW I've been looking harder and have now seen the full range of >> +/- 32k so I assume it's just not getting initialised properly. > > Hmm I am not so sure now maybe it's a stats thing - I have seen flows > apparently stuck with -ve allot but behaving normally. > > I notice allot is short (16bits?) in sfq but see > > +struct tc_sfq_xstats > +{ > + __s32 allot; > +}; That doesn't matter, its promoted to a 32 bit value. But now that you mention it, the SFQ patch for iproute I submitted used a different type. Stephen has resynced it with the kernel in the mean time, but you might be using the broken version.