Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758466AbXETNjE (ORCPT ); Sun, 20 May 2007 09:39:04 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1756012AbXETNiy (ORCPT ); Sun, 20 May 2007 09:38:54 -0400 Received: from neon.samage.net ([85.17.153.66]:34399 "EHLO neon.samage.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755980AbXETNix (ORCPT ); Sun, 20 May 2007 09:38:53 -0400 Message-ID: <4028.81.207.0.53.1179668330.squirrel@secure.samage.net> In-Reply-To: <465020B8.1010605@gmail.com> References: <20070519053532.GA32522@havoc.gtf.org> <1441.81.207.0.53.1179591630.squirrel@secure.samage.net> <464F7440.10900@garzik.org> <2741.81.207.0.53.1179616929.squirrel@secure.samage.net> <465020B8.1010605@gmail.com> Date: Sun, 20 May 2007 15:38:50 +0200 (CEST) Subject: Re: [PATCH] sata_sil: Greatly improve DMA support From: "Indan Zupancic" To: "Tejun Heo" Cc: "Jeff Garzik" , linux-ide@vger.kernel.org, "LKML" , "Alan Cox" User-Agent: SquirrelMail/1.4.8 MIME-Version: 1.0 Content-Type: text/plain;charset=UTF-8 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-Spam-Score: -1.8 X-Scan-Signature: b76837ecbba1170f708de966edddc2d0 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 3978 Lines: 84 On Sun, May 20, 2007 12:19, Tejun Heo wrote: > Indan Zupancic wrote: >> On Sun, May 20, 2007 00:03, Jeff Garzik wrote: >>> Indan Zupancic wrote: >>>> This patch seems to work with my SiI 3512, though I don't notice any >>>> difference, neither a speedup, nor a slowdown. Hdparm gives the same >>>> speeds (-tT), and cp -a'ing kernel sources is abysmal slow in both cases, >>>> (need to look into that one) so I didn't really test it that well. >>> >>> It won't result in much of a speedup, except in situations where IOMMU >>> or other situation that causes you to run into the 64k boundary being an >>> issue -- generally only on huge transfers. >>> >>> A good measure is to dd(1) to/from the block device, rather than using a >>> filesystem. As has been shown on LKML, the filesystem can really slow >>> things down in some cases. >> >> I didn't really expect a speedup, it's more that I've no regression to report. >> >> I could benchmark the patch more thoroughly, but right now I'm more worried >> about the crawling cp I just discovered. Talking about filesystems slowing down >> things... >> >> Test: >> >> $ cp -a linux-2.6/ /tmp/ >> >> done on the same ext3 partition. linux-2.6 contains source and git repo only, >> I'm compiling stuff with O=../obj. >> >> $ vmstat 10 >> procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu---- >> r b swpd free buff cache si so bi bo in cs us sy id wa >> 0 1 0 4168 3316 195700 0 0 739 494 530 393 15 3 66 16 >> 0 3 4 4120 2040 198196 0 0 14677 14111 1247 435 0 17 0 83 >> 0 1 4 3588 1444 199696 0 0 8892 9472 1362 438 0 12 0 88 >> 1 0 4 3772 4228 196012 0 0 764 454 1161 345 0 4 0 96 >> 0 1 4 3548 6156 193088 0 0 793 851 1158 340 0 4 0 96 >> 0 1 4 3852 7608 189096 0 0 798 523 1160 474 1 4 0 95 >> 1 1 4 3612 8684 186048 0 0 1244 864 1178 430 2 5 0 93 >> 0 1 4 90660 9308 96396 0 0 853 906 1244 578 7 6 0 87 >> 0 1 4 72280 9816 112368 0 0 830 854 1278 429 12 5 0 83 >> 1 0 4 52488 10296 130560 0 0 935 861 1178 418 1 6 0 94 >> 0 1 4 30500 10788 149776 0 0 977 858 1178 371 0 6 0 94 >> 0 1 4 9792 11244 167856 0 0 918 1394 1182 350 1 5 0 94 >> 0 1 4 4016 11216 172504 0 0 1017 858 1181 382 1 6 0 94 >> 0 1 4 3660 11484 171484 0 0 966 861 1182 410 1 6 0 94 >> >> It never finished, as I had no patience to copy about 900 Mb with this rate. >> >> As it's a git tree, I suppose it's heavily fragmented, but this is still rather >> pathetic. Should I blame cp, or is something else wrong? Any ideas how >> to figure this one out would be appreciated. Sorry for the off-topicness. > > Do things improve if you change the io scheduler to deadline? > > # echo deadline > /sys/block/sda/queue/scheduler I also tried noop, anticipatory, and now deadline, but it doesn't matter. > Also worth looking at is the following bug entry. > > http://bugzilla.kernel.org/show_bug.cgi?id=7372 > > There seems to be weird interaction among the scheduler / VM / IO. The > exact cause is still not verified. :-( I know, I posted a bugreport too, but for starvation with the anticipatory scheduler. Anyway, that bug seems unrelated to my case, as they have system unresponsiveness or other nastiness, while I only have a crawling cp, which I blame on weaknesses within ext3, a badly designed cp program and very fragmented filesystem. I just need to verify that, somehow. I'll try with older kernels later. Greetings, Indan - 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/