Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261306AbVDULbg (ORCPT ); Thu, 21 Apr 2005 07:31:36 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261302AbVDULbg (ORCPT ); Thu, 21 Apr 2005 07:31:36 -0400 Received: from cernmx06.cern.ch ([137.138.166.160]:23501 "EHLO cernmxlb.cern.ch") by vger.kernel.org with ESMTP id S261327AbVDULbE (ORCPT ); Thu, 21 Apr 2005 07:31:04 -0400 DomainKey-Signature: a=rsa-sha1; c=nofws; s=beta; d=cern.ch; q=dns; h=received:message-id:date:from:to:subject:mime-version:content-type:content-transfer-encoding; b=GvE7w/EkmBDc/uXFpJhXaODkQCCCxRYnzMRwbBDDewEv1GuF4YxxVki3ym0ctKYaGvQMDn9SN29EXk5krKflHJbYgf8baacPZHhic0S6mGsy7rObPNh0smAMLzJYS60z; Keywords: CERN SpamKiller Note: -51 Charset: west-latin X-Filter: CERNMX06 CERN MX v2.0 050413.1147 Release Message-ID: <42678EEA.6070109@cern.ch> Date: Thu, 21 Apr 2005 13:30:50 +0200 From: Andreas Hirstius User-Agent: Mozilla/5.0 (X11; U; Linux ia64; en-US; rv:1.7.7) Gecko/20050418 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Bartlomiej ZOLNIERKIEWICZ CC: Gelato technical , linux-kernel@vger.kernel.org Subject: Re: Serious performance degradation on a RAID with kernel 2.6.10-bk7 and later References: <42669357.9080604@cern.ch> <42668977.5060708@utah-nac.org> <42676501.8030805@cern.ch> <58cb370e05042102272ce70f2@mail.gmail.com> <42677587.8030707@cern.ch> In-Reply-To: <42677587.8030707@cern.ch> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 21 Apr 2005 11:30:50.0228 (UTC) FILETIME=[91D5DB40:01C54665] Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1012 Lines: 40 Hi, The fls() patch from David solves the problem :-)) Do you have an idea, when it will be in the mainline kernel?? Andreas Bartlomiej ZOLNIERKIEWICZ wrote: > > Hi! > >> A small update. >> >> Patching mm/filemap.c is not necessary in order to get the improved >> performance! >> It's sufficient to remove roundup_pow_of_two from |get_init_ra_size ... >> >> So a simple one-liner changes to picture dramatically. >> But why ?!?!? > > > roundup_pow_of_two() uses fls() and ia64 has buggy fls() implementation > [ seems that David fixed it but patch is not in the mainline yet]: > > http://www.mail-archive.com/linux-ia64@vger.kernel.org/msg01196.html > > That would also explain why you couldn't reproduce the problem on ia32 > Xeon machines. > > Bartlomiej > - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/