From mboxrd@z Thu Jan 1 00:00:00 1970 From: Edward Shishkin Subject: Re: Cleanup patches Date: Sun, 16 Nov 2014 10:50:06 +0100 Message-ID: <5468734E.5010807@gmail.com> References: <5c7c368b0711140634y569c8ce8ke4b2e813a2244ce2@mail.gmail.com> <5c7c368b0711141439u1cb89d82g8ee0928eabbae2a5@mail.gmail.com> <5467E20F.8010903@gmail.com> Mime-Version: 1.0 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=message-id:date:from:user-agent:mime-version:to:cc:subject :references:in-reply-to:content-type:content-transfer-encoding; bh=vq/jUIxbfxEZiiqKZ+J86pRofLB5XHWkGus3pdZzks0=; b=hEa+RBAqqmzyg5tQxfVNHJFpwOVVbfMEUKCSb9Vw6O+7FKTZYpKzOqEk2GAphmMpLD N5MPXLzKmFA6RDX7k43x7jgrFoMO4lKNv6oVXlftm0y90QLuLhtLmQsreMA6fdrOtxfm Ts046m1Mh1P2VucYqLzVO0l7KihxAvBRX6mDQYLj+MQA5fNiNSShXoD4SMRjnIi7TIuU PCasfpO06sYxhxOEIAjzTHS/8AnF2d4RRfHkez9Y1NAzX6JzEiIcEtDi411Dde2JiqjV sNtpqy8h6Iryca6weEweNeM+1MIBRaWQxWwayBePR2hhHNFK6zvubxowkUV1BCgwMWuz bMYg== In-Reply-To: Sender: reiserfs-devel-owner@vger.kernel.org List-ID: Content-Type: text/plain; charset="utf-8"; format="flowed" To: =?UTF-8?B?RHXFoWFuIMSMb2xpxIc=?= Cc: Ivan Shapovalov , reiserfs-devel On 11/16/2014 09:29 AM, Du=C5=A1an =C4=8Coli=C4=87 wrote: > On Sun, Nov 16, 2014 at 2:28 AM, Du=C5=A1an =C4=8Coli=C4=87 wrote: >> On Sun, Nov 16, 2014 at 12:30 AM, Du=C5=A1an =C4=8Coli=C4=87 wrote: >>> On Sun, Nov 16, 2014 at 12:30 AM, Edward Shishkin >>> wrote: >>>> On 11/16/2014 12:06 AM, Du=C5=A1an =C4=8Coli=C4=87 wrote: >>>>> These are results from bisecting of R4 slowdown with ccreg40 plug= in >>> Brainfart after hours of bisecting. I meant reg40 plugin. >>>> >>>> Hmm.. Are you sure? >>>> 2.6.10 doesn't know about ccreg... >>>> >>>> >>>> >>>>> with config and scripts I used, first double run is to see disper= sion >>>>> of consecutive tests: >>>>> >>>>> Reiser4 patch Real User System >>>>> 2.6.10-2 2m46.785s 0m3.214s 1m45.286s >>>>> 2.6.10-2 2m52.700s 0m3.198s 1m43.361s >>>>> 2.6.16-5 3m32.694s 0m2.900s 1m34.698s >>>>> 2.6.19-4 3m19.672s 0m3.120s 1m33.234s >>>>> 2.6.20 3m9.582s 0m3.164s 1m28.830s >>>>> 2.6.21 7m52.960s 0m2.992s 1m28.326s >>>>> 2.6.22-2 6m57.991s 0m2.852s 1m28.258s >>>>> >> Here are the more bisected results: >> >> Reiser4 patch Real User System >> 2.6.10-2 2m46.785s 0m3.214s 1m45.286s >> 2.6.10-2 2m52.700s 0m3.198s 1m43.361s >> 2.6.16-5 3m32.694s 0m2.900s 1m34.698s >> 2.6.19-4 3m19.672s 0m3.120s 1m33.234s >> 2.6.20 3m9.582s 0m3.164s 1m28.830s >> 2.6.21-rc1 3m13.643s 0m2.884s 1m30.482s >> 2.6.21-rc2 3m10.054s 0m2.904s 1m30.106s >> 2.6.21-rc3 7m39.374s 0m2.848s 1m29.570s >> 2.6.21-rc4 7m37.928s 0m2.904s 1m30.778s >> 2.6.21 7m52.960s 0m2.992s 1m28.326s >> 2.6.21 7m35.929s 0m3.044s 1m34.738s >> 2.6.22-2 6m57.991s 0m2.852s 1m28.258s >> > And some new relevant ones: > > 2.6.29 2m57.027s 0m3.096s 1m33.266s > 3.17.2 2m34. 1m20. > > 3.17 is cropped because I copied it by hand as network didn't work > with old userspace. > > Is this issue closed now? Thank you for your work! TBH, I thought that the performance issue introduced in 2.6.21-rc3 still takes place in modern kernels.. Edward. > > > Have a nice day > Dushan > > >> Obviously something happened between rc2 and rc3 >> >>>>> Have a nice day >>>>> >>>>> Dushan >>>>> >>>>> On Fri, Oct 24, 2014 at 11:17 PM, Du=C5=A1an =C4=8Coli=C4=87 wrote: >>>>>> OK after 7 years, can't believe how long ago it was, I decided t= o dig up >>>>>> some ancient hw and to try again to do this bisect. >>>>>> Questions: >>>>>> 1. Would it still be useful? >>>>>> 2. Any other tests that I can do on those old kernels that can b= e of >>>>>> value? >>>>>> 3. What would be best bisect range? Oldest and newest kernel? >>>>>> >>>>>> On Nov 14, 2007 11:39 PM, "Edward Shishkin" >>>>>> wrote: >>>>>>> On 11/15/07, Dushan Tcholich wrote: >>>>>>>> ... >>>>>>>>>> Well. There is a problem: starting from some point, performa= nce of >>>>>>>>>> reiser4 is substantially dropped for unknown reasons. As I r= emember, >>>>>>>>>> there were a lot of complaints about it. Also I have made a = brief >>>>>>>>>> test not so long ago (copy of linux source tree located in r= amfs to >>>>>>>>>> reiser4 partition): yeah, it is 3 times worse then it ought = to be, >>>>>>>>>> "vmstat 2" reports low bo-activity (something like 10000 blo= cks/s, >>>>>>>>>> instead of usual ~30000). >>>>>>>>>> >>>>>>>>>> It would be nice to find a changeset which kills performance= =2E >>>>>>>>>> >>>>>>>>>> Would you please look at this? It is non-trivial task, so ev= ery >>>>>>>>>> result would be ok (say, to know the first kernel in -mm ser= ies >>>>>>>>>> with slow reiser4). >>>>>>>>>> >>>>>>>>>> Hints: >>>>>>>>>> >>>>>>>>>> 1. The problem is in (default) unix-file plugin (nobody main= tained >>>>>>>>>> this for a long of time), so compression should be disa= bled. >>>>>>>>>> 2. I guess it should be something like bisecting. >>>>>>>>>> 3. I think that the problem appeared in ~2.6.17-mmXX kernel = when >>>>>>>>>> vs sent vfs patches with batch_write methods, and then= Andrew >>>>>>>>>> Morton evicted them because of some problems. However, >>>>>>>>>> I might be wrong here! >>>>>>>>>> >>>>>>>>> If you could give me an easy way to benchmark (guide me by ha= nd), I >>>>>>>>> could try to bisect it. >>>>>>>>> It would be easier if -mm could be bisected using git. >>>>>>>>> But my / is on r4+cc, so I don't know how could I do it? Mayb= e on >>>>>>>>> other machine? >>>>>>> Yes! Compression announced 15 March 2007, and it may happen >>>>>>> that some kernel you will need to boot are not support your "/"= =2E >>>>>>> So for bisecting you need the following: >>>>>>> 1. a machine with >=3D 512M RAM and "/" formatted with some fs >>>>>>> supported by old kernels. >>>>>>> 2. a spare partition. >>>>>>> 3. enable ramfs (it seems it is enabled by default in most dist= ros). >>>>>>> 4. put a tarball-to-copy in some working directory (I had >>>>>>> linux-2.6.9.tar.gz) >>>>>>> >>>>>>> Note, that some old -mm kernels are not compilable/bootable >>>>>>> (if so, pull the "hotfixes" patches from akpm's directory on ke= rnel.org) >>>>>>> >>>>>>> 5. edit the attached patches in accordance with your configurat= ions >>>>>>> 6. build and boot the testing kernel with reiser4 debug disable= d >>>>>>> (I think no needs to boot in single mode, or discard kde,= etc..) >>>>>>> 7. run "vmstat 2" >>>>>>> 8. run ./prepare_copy.sh && ./ncopy on another console >>>>>>> >>>>>>> I have the following: >>>>>>> real 6m27.970s >>>>>>> user 0m2.116s >>>>>>> sys 1m4.972s >>>>>>> >>>>>>> God, it is fairly bad results: On my machine real time should b= e >>>>>>> something like 2m20...... >>>> -- To unsubscribe from this list: send the line "unsubscribe reiserfs-deve= l" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html