From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751572Ab1ITO4p (ORCPT ); Tue, 20 Sep 2011 10:56:45 -0400 Received: from gator34.hostgator.com ([184.173.197.218]:38797 "EHLO gator34.hostgator.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751437Ab1ITO4m (ORCPT ); Tue, 20 Sep 2011 10:56:42 -0400 X-Greylist: delayed 5808 seconds by postgrey-1.27 at vger.kernel.org; Tue, 20 Sep 2011 10:56:42 EDT Message-ID: <4E7892F1.1000107@gmail.com> Date: Tue, 20 Sep 2011 15:19:45 +0200 From: Slawomir Czarko-Wasiutycz User-Agent: Mozilla/5.0 (X11; Linux i686; rv:6.0.2) Gecko/20110906 Thunderbird/6.0.2 MIME-Version: 1.0 To: Andrea Arcangeli CC: Lin Ming , fa.linux.kernel@googlegroups.com, Linux Kernel Mailing List , Thomas Sattler , Johannes Weiner , Mel Gorman Subject: Re: iotop: khugepaged at 99.99% (2.6.38.3) References: <20110919175159.GL7800@redhat.com> In-Reply-To: <20110919175159.GL7800@redhat.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - gator34.hostgator.com X-AntiAbuse: Original Domain - vger.kernel.org X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12] X-AntiAbuse: Sender Address Domain - gmail.com X-BWhitelist: no X-Source: X-Source-Args: X-Source-Dir: X-Source-Sender: d51a4d067.access.telenet.be (gaia.garous.net) [81.164.208.103]:42878 X-Source-Auth: slawek@czarko.net X-Email-Count: 1 X-Source-Cap: cG9saWNhdDtwb2xpY2F0O2dhdG9yMzQuaG9zdGdhdG9yLmNvbQ== Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 09/19/2011 07:51 PM, Andrea Arcangeli wrote: > On Thu, Sep 15, 2011 at 02:43:32PM +0800, Lin Ming wrote: >> # cat /proc/`pgrep khugepaged`/io >> rchar: 0 >> wchar: 0 >> syscr: 0 >> syscw: 0 >> read_bytes: 0 >> write_bytes: 0 >> cancelled_write_bytes: 0 >> >> Andrea, >> >> From above output, all fields are zero. >> Does it mean that transparent huge page was not triggered/used at all? > Good idea to check it like that, yes that should confirm no > ->writepage was called by khugepaged through > compaction->migrate->writepage. > > It may have been used for migration, but the compaction code run by > khugepaged didn't trigger writes, or the write_bytes should have been >> 0. > With regard to Slawomir's problem, the kernel > kernel-PAE-2.6.40.4-5.fc15.i686 includes Mel's fix for the compaction > scan to stay in the right zone. I tried with previous kernel (2.6.40.3-0.fc15) and I haven't seen this problem when running the PC for 24 hours and building code in a loop inside the VM. With 2.6.40.4-5.fc15 I get it after a few hours. > > Slawomir could you run the command "cat /proc/`pgrep khugepaged`/io" > as root, so see if there's significant writeout going from khugepaged? For a long time there were just zeros appearing. After stalls started I got this: [root]# cat /proc/`pgrep khugepaged`/io rchar: 0 wchar: 0 syscr: 0 syscw: 0 read_bytes: 0 write_bytes: 24576 cancelled_write_bytes: 0 and after some time: [root]# cat /proc/`pgrep khugepaged`/io rchar: 0 wchar: 0 syscr: 0 syscw: 0 read_bytes: 0 write_bytes: 32768 cancelled_write_bytes: 0 > If there is you can try the patch in the below link. > > https://lkml.org/lkml/2011/7/26/103 > > But the fact it happens on top of VMplayer with a PAE guest, may also > be a variable to take into account, migrate does quite some pagetable > work. If VMplayer uses EPT/NTP (do you have EPT/NTP available as VT > feature in the host /proc/cpuinfo?) it's hard to see how that could be > related. I don't have EPT/NTP in /proc/cpuinfo. CPU is AMD Phenom(tm) II X6 1100T Processor Here are the flags from /proc/cpuinfo: flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc nonstop_tsc extd_apicid aperfmperf pni monitor cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt cpb npt lbrv svm_lock nrip_save pausefilter I'll build a kernel with the suggested patch and give it a try.