From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756785Ab1ISFvY (ORCPT ); Mon, 19 Sep 2011 01:51:24 -0400 Received: from TYO201.gate.nec.co.jp ([202.32.8.193]:62849 "EHLO tyo201.gate.nec.co.jp" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751010Ab1ISFvX (ORCPT ); Mon, 19 Sep 2011 01:51:23 -0400 Message-ID: <4E76D810.6010504@ce.jp.nec.com> Date: Mon, 19 Sep 2011 14:50:08 +0900 From: "Jun'ichi Nomura" User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.18) Gecko/20110621 Fedora/3.1.11-1.fc14 Thunderbird/3.1.11 MIME-Version: 1.0 To: Lukas Hejtmanek CC: Mike Snitzer , Kiyoshi Ueda , agk@redhat.com, linux-kernel@vger.kernel.org Subject: Re: request baset device mapper in Linux References: <20110720082640.GZ7561@ics.muni.cz> <4E280964.6070000@ct.jp.nec.com> <20110721132627.GY7561@ics.muni.cz> <4E291F28.2040609@ct.jp.nec.com> <20110722081903.GB7561@ics.muni.cz> <4E2A7806.1060601@ce.jp.nec.com> <20110724221619.GM16487@ics.muni.cz> <4E367256.8020206@ct.jp.nec.com> <20110908132756.GT31717@ics.muni.cz> <20110916140839.GE7428@ics.muni.cz> In-Reply-To: <20110916140839.GE7428@ics.muni.cz> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Lukas, On 09/16/11 23:08, Lukas Hejtmanek wrote: > On Thu, Sep 15, 2011 at 02:49:52PM -0400, Mike Snitzer wrote: >> Does "with your patch" mean against 3.0.3 or the original 2.6.32 kernel? >> >> If you apply the patch to the 3.0.3 kernel do you regain the performance? > > Unfortunately, the patch does not help with 3.0.3 kernel Thanks for the additional testing. So: 2.6.32.36 no-multipath: 2.8 GB/s multipath: 600 MB/s multipath w/ patch: 2.8-2.9 GB/s 3.0.3 no-multipath: ?? multipath: 2.5 GB/s multipath w/ patch: 2.5 GB/s(?) Have you tried 3.0.3 without multipath? > top looks like this: > 63 root 20 0 0 0 0 S 38 0.0 0:16.65 kswapd1 > 18 root 20 0 0 0 0 R 32 0.0 0:14.38 kworker/4:0 > 19 root 20 0 0 0 0 R 30 0.0 0:20.88 ksoftirqd/4 > 153 root 20 0 0 0 0 R 29 0.0 0:14.46 kworker/4:2 > 62 root 20 0 0 0 0 S 22 0.0 0:09.44 kswapd0 dm-multipath itself does not use workqueue for normal I/Os. So you might be hitting a different problem. Thanks, -- Jun'ichi Nomura, NEC Corporation