From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jaegeuk Kim Subject: Re: [LKP] [lkp] [f2fs] ec795418c4: fsmark.files_per_sec -36.3% regression Date: Thu, 4 Aug 2016 10:24:52 -0700 Message-ID: <20160804172452.GA12093@jaegeuk> References: <20160718020950.GB4986@yexl-desktop> <87twfmsndf.fsf@yhuang-mobile.sh.intel.com> <87bn18cvuu.fsf_-_@yhuang-mobile.sh.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Received: from sog-mx-3.v43.ch3.sourceforge.com ([172.29.43.193] helo=mx.sourceforge.net) by sfs-ml-3.v29.ch3.sourceforge.com with esmtp (Exim 4.76) (envelope-from ) id 1bVMOD-0000eq-Qw for linux-f2fs-devel@lists.sourceforge.net; Thu, 04 Aug 2016 17:25:01 +0000 Received: from mail.kernel.org ([198.145.29.136]) by sog-mx-3.v43.ch3.sourceforge.com with esmtps (TLSv1:AES256-SHA:256) (Exim 4.76) id 1bVMOC-0000ED-NX for linux-f2fs-devel@lists.sourceforge.net; Thu, 04 Aug 2016 17:25:01 +0000 Content-Disposition: inline In-Reply-To: <87bn18cvuu.fsf_-_@yhuang-mobile.sh.intel.com> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: linux-f2fs-devel-bounces@lists.sourceforge.net To: "Huang, Ying" Cc: linux-f2fs-devel@lists.sourceforge.net, lkp@01.org, kernel test robot , LKML Hi Huang, On Thu, Aug 04, 2016 at 10:00:41AM -0700, Huang, Ying wrote: > Hi, Jaegeuk, > > "Huang, Ying" writes: > > Hi, > > > > I checked the comparison result below and found this is a regression for > > fsmark.files_per_sec, not fsmark.app_overhead. > > > > Best Regards, > > Huang, Ying > > > > kernel test robot writes: > > > >> FYI, we noticed a -36.3% regression of fsmark.files_per_sec due to commit: > >> > >> commit ec795418c41850056feb956534edf059dc1155d4 ("f2fs: use percpu_rw_semaphore") > >> https://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs.git dev-test > > I found this has been merged by upstream. Do you have some plan to fix > it? Or you think the test itself has some problem? Sorry, too busy to take a look at this. The patch implements percpu_rw_semaphore which is intended to enhance FS scalability. Since I couldn't see any big regression in my test cases, could you check any debugging options which may give some overheads? Let me recheck this with whole my tests. Thanks, > > We have another 2 regressions > > - [lkp] [f2fs] 3bdad3c7ee: aim7.jobs-per-min -25.3% regression > - [lkp] [f2fs] b93f771286: aim7.jobs-per-min -81.2% regression > > they are merged by upstream now too. So same questions for them too. > > Best Regards, > Huang, Ying > > >> in testcase: fsmark > >> on test machine: 72 threads Haswell-EP with 128G memory > >> with following parameters: > > cpufreq_governor=performance/disk=1SSD/filesize=8K/fs=f2fs/iterations=8/nr_directories=16d/nr_files_per_directory=256fpd/nr_threads=4/sync_method=fsyncBeforeClose/test_size=72G > >> > >> > >> > >> Disclaimer: > >> Results have been estimated based on internal Intel analysis and are provided > >> for informational purposes only. Any difference in system hardware or software > >> design or configuration may affect actual performance. > >> ------------------------------------------------------------------------------ From mboxrd@z Thu Jan 1 00:00:00 1970 Content-Type: multipart/mixed; boundary="===============3523631828078492326==" MIME-Version: 1.0 From: Jaegeuk Kim To: lkp@lists.01.org Subject: Re: [f2fs] ec795418c4: fsmark.files_per_sec -36.3% regression Date: Thu, 04 Aug 2016 10:24:52 -0700 Message-ID: <20160804172452.GA12093@jaegeuk> In-Reply-To: <87bn18cvuu.fsf_-_@yhuang-mobile.sh.intel.com> List-Id: --===============3523631828078492326== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Hi Huang, On Thu, Aug 04, 2016 at 10:00:41AM -0700, Huang, Ying wrote: > Hi, Jaegeuk, > = > "Huang, Ying" writes: > > Hi, > > > > I checked the comparison result below and found this is a regression for > > fsmark.files_per_sec, not fsmark.app_overhead. > > > > Best Regards, > > Huang, Ying > > > > kernel test robot writes: > > > >> FYI, we noticed a -36.3% regression of fsmark.files_per_sec due to com= mit: > >> > >> commit ec795418c41850056feb956534edf059dc1155d4 ("f2fs: use percpu_rw_= semaphore") > >> https://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs.git dev-t= est > = > I found this has been merged by upstream. Do you have some plan to fix > it? Or you think the test itself has some problem? Sorry, too busy to take a look at this. The patch implements percpu_rw_semaphore which is intended to enhance FS scalability. Since I couldn't see any big regression in my test cases, coul= d you check any debugging options which may give some overheads? Let me recheck this with whole my tests. Thanks, > = > We have another 2 regressions > = > - [lkp] [f2fs] 3bdad3c7ee: aim7.jobs-per-min -25.3% regression > - [lkp] [f2fs] b93f771286: aim7.jobs-per-min -81.2% regression > = > they are merged by upstream now too. So same questions for them too. > = > Best Regards, > Huang, Ying > = > >> in testcase: fsmark > >> on test machine: 72 threads Haswell-EP with 128G memory > >> with following parameters: > > cpufreq_governor=3Dperformance/disk=3D1SSD/filesize=3D8K/fs=3Df2fs/iter= ations=3D8/nr_directories=3D16d/nr_files_per_directory=3D256fpd/nr_threads= =3D4/sync_method=3DfsyncBeforeClose/test_size=3D72G > >> > >> > >> > >> Disclaimer: > >> Results have been estimated based on internal Intel analysis and are p= rovided > >> for informational purposes only. Any difference in system hardware or = software > >> design or configuration may affect actual performance. > >> --===============3523631828078492326==-- From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S965168AbcHDRY4 (ORCPT ); Thu, 4 Aug 2016 13:24:56 -0400 Received: from mail.kernel.org ([198.145.29.136]:57470 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933845AbcHDRYz (ORCPT ); Thu, 4 Aug 2016 13:24:55 -0400 Date: Thu, 4 Aug 2016 10:24:52 -0700 From: Jaegeuk Kim To: "Huang, Ying" Cc: kernel test robot , lkp@01.org, LKML , linux-f2fs-devel@lists.sourceforge.net Subject: Re: [LKP] [lkp] [f2fs] ec795418c4: fsmark.files_per_sec -36.3% regression Message-ID: <20160804172452.GA12093@jaegeuk> References: <20160718020950.GB4986@yexl-desktop> <87twfmsndf.fsf@yhuang-mobile.sh.intel.com> <87bn18cvuu.fsf_-_@yhuang-mobile.sh.intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <87bn18cvuu.fsf_-_@yhuang-mobile.sh.intel.com> User-Agent: Mutt/1.6.0 (2016-04-01) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Huang, On Thu, Aug 04, 2016 at 10:00:41AM -0700, Huang, Ying wrote: > Hi, Jaegeuk, > > "Huang, Ying" writes: > > Hi, > > > > I checked the comparison result below and found this is a regression for > > fsmark.files_per_sec, not fsmark.app_overhead. > > > > Best Regards, > > Huang, Ying > > > > kernel test robot writes: > > > >> FYI, we noticed a -36.3% regression of fsmark.files_per_sec due to commit: > >> > >> commit ec795418c41850056feb956534edf059dc1155d4 ("f2fs: use percpu_rw_semaphore") > >> https://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs.git dev-test > > I found this has been merged by upstream. Do you have some plan to fix > it? Or you think the test itself has some problem? Sorry, too busy to take a look at this. The patch implements percpu_rw_semaphore which is intended to enhance FS scalability. Since I couldn't see any big regression in my test cases, could you check any debugging options which may give some overheads? Let me recheck this with whole my tests. Thanks, > > We have another 2 regressions > > - [lkp] [f2fs] 3bdad3c7ee: aim7.jobs-per-min -25.3% regression > - [lkp] [f2fs] b93f771286: aim7.jobs-per-min -81.2% regression > > they are merged by upstream now too. So same questions for them too. > > Best Regards, > Huang, Ying > > >> in testcase: fsmark > >> on test machine: 72 threads Haswell-EP with 128G memory > >> with following parameters: > > cpufreq_governor=performance/disk=1SSD/filesize=8K/fs=f2fs/iterations=8/nr_directories=16d/nr_files_per_directory=256fpd/nr_threads=4/sync_method=fsyncBeforeClose/test_size=72G > >> > >> > >> > >> Disclaimer: > >> Results have been estimated based on internal Intel analysis and are provided > >> for informational purposes only. Any difference in system hardware or software > >> design or configuration may affect actual performance. > >>