Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1760511AbXHJHGn (ORCPT ); Fri, 10 Aug 2007 03:06:43 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1756926AbXHJHGT (ORCPT ); Fri, 10 Aug 2007 03:06:19 -0400 Received: from smtp35.poczta.interia.pl ([80.48.65.35]:38334 "EHLO smtp4.poczta.interia.pl" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1756648AbXHJHGR (ORCPT ); Fri, 10 Aug 2007 03:06:17 -0400 Message-ID: <46BC0E5B.6030403@interia.pl> Date: Fri, 10 Aug 2007 09:06:03 +0200 From: =?ISO-8859-2?Q?Rafa=B3_Bilski?= User-Agent: Thunderbird 2.0.0.6 (X11/20070802) MIME-Version: 1.0 To: Dimitrios Apostolou Cc: linux-kernel@vger.kernel.org, Alan Cox , Andrew Morton Subject: Re: high system cpu load during intense disk i/o References: <200708031903.10063.jimis@gmx.net> <200708071750.51019.jimis@gmx.net> <46BA14A2.5030705@interia.pl> <200708091117.51687.jimis@gmx.net> In-Reply-To: <200708091117.51687.jimis@gmx.net> Content-Type: text/plain; charset=ISO-8859-2; format=flowed Content-Transfer-Encoding: 8bit X-EMID: 338daacc Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1341 Lines: 32 > So I would assume that delay_tsc() probably only makes the situation worse for > the libata tests, but the real problem is at __switch_to() and schedule(). Do > you agree with these assumptions? Yes. I agree that percentage of CPU time is unreasonably high for these functions. But not only for them. > Is there a way to for oprofile to report the time spent in function calls > depending on the call trace? I don't know any. > Thanks again for the help. I guess if that doesn't lead anywhere I'll just > start compiling older vanilla kernels and see when the problem dissapears. > But this needs a lot of time and I'm not sure for how long I'll be able to > not offer any service with those disks (I was supposed to use RAID 0 to > provide storage space with them, but with the current problems that wouldn't > be so wise). Probably this is the best and fastest way to diagnose problem. > Dimitris Regards Rafa? ---------------------------------------------------------------------- Ta?m Renaty i Andrzeja tu nie znajdziesz. Mamy co? lepszego ... >>>http://link.interia.pl/f1b41 - 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/