Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932927Ab1D1Otu (ORCPT ); Thu, 28 Apr 2011 10:49:50 -0400 Received: from cantor.suse.de ([195.135.220.2]:57988 "EHLO mx1.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754103Ab1D1Ott (ORCPT ); Thu, 28 Apr 2011 10:49:49 -0400 Subject: Re: [BUG] fatal hang untarring 90GB file, possibly writeback related. From: James Bottomley To: Chris Mason Cc: Colin Ian King , linux-fsdevel , linux-mm , linux-kernel , linux-ext4 In-Reply-To: <1303993705-sup-5213@think> References: <1303920553.2583.7.camel@mulgrave.site> <1303921583-sup-4021@think> <1303923000.2583.8.camel@mulgrave.site> <1303923177-sup-2603@think> <1303924902.2583.13.camel@mulgrave.site> <1303925374-sup-7968@think> <1303926637.2583.17.camel@mulgrave.site> <1303934716.2583.22.camel@mulgrave.site> <1303990590.2081.9.camel@lenovo> <1303993705-sup-5213@think> Content-Type: text/plain; charset="UTF-8" Date: Thu, 28 Apr 2011 09:49:43 -0500 Message-ID: <1304002183.2598.1.camel@mulgrave.site> Mime-Version: 1.0 X-Mailer: Evolution 2.32.1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1169 Lines: 28 On Thu, 2011-04-28 at 08:29 -0400, Chris Mason wrote: > Excerpts from Colin Ian King's message of 2011-04-28 07:36:30 -0400: > > One more data point to add, I've been looking at an identical issue when > > copying large amounts of data. I bisected this - and the lockups occur > > with commit > > 3e7d344970673c5334cf7b5bb27c8c0942b06126 - before that I don't see the > > issue. With this commit, my file copy test locks up after ~8-10 > > iterations, before this commit I can copy > 100 times and don't see the > > lockup. > > Well, that's really interesting. I tried with compaction on here and > couldn't trigger it, but this (very very lightly) tested patch might > help. > > It moves the writeout throttle before the goto restart, and also makes > sure we do at least one cond_resched before we loop. It seems to take longer, but with a PREEMPT kernel, kswapd eventually shoots up to 99% during the tar. James -- 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/