From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755794Ab1GMPmn (ORCPT ); Wed, 13 Jul 2011 11:42:43 -0400 Received: from adelie.canonical.com ([91.189.90.139]:33294 "EHLO adelie.canonical.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755452Ab1GMPml (ORCPT ); Wed, 13 Jul 2011 11:42:41 -0400 Date: Wed, 13 Jul 2011 10:42:32 -0500 (CDT) From: Manoj Iyer X-X-Sender: manjo@lazy To: Chris Ball cc: Arnd Bergmann , Manoj Iyer , linux-kernel@vger.kernel.org, jbarnes@virtuousgeek.org, matsumur@nts.ricoh.co.jp, linux-pci@vger.kernel.org, linux-mmc@vger.kernel.org Subject: Re: [PATCH] mmc: Added quirks for Ricoh 1180:e823 lower base clock frequency In-Reply-To: Message-ID: References: <1310419715-13254-1-git-send-email-manoj.iyer@canonical.com> <201107122000.50309.arnd@arndb.de> User-Agent: Alpine 2.02 (DEB 1266 2009-07-14) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Chris/Arnd, Here is a series of test I did with the patched kernel. == cold boot insert SD card == u@u:~/flash/flashbench$ sudo ./flashbench -O --erasesize=$[4 * 1024 * 1024] --blocksize=$[256 * 1024] /dev/mmcblk0 --open-au-nr=2 4MiB 4.96M/s 2MiB 6.3M/s 1MiB 6.23M/s 512KiB 6.23M/s 256KiB 6.26M/s == rerun the same test 2nd time === u@u:~/flash/flashbench$ sudo ./flashbench -O --erasesize=$[4 * 1024 * 1024] --blocksize=$[256 * 1024] /dev/mmcblk0 --open-au-nr=2 4MiB 6.28M/s 2MiB 6.29M/s 1MiB 6.29M/s 512KiB 6.29M/s 256KiB 6.26M/s == remove and reinsert the SD card == u@u:~/flash/flashbench$ sudo ./flashbench -O --erasesize=$[4 * 1024 * 1024] --blocksize=$[256 * 1024] /dev/mmcblk0 --open-au-nr=2 4MiB 6.42M/s 2MiB 6.28M/s 1MiB 6.22M/s 512KiB 6.23M/s 256KiB 6.26M/s == remove and reinsert the SD card == u@u:~/flash/flashbench$ sudo ./flashbench -O --erasesize=$[4 * 1024 * 1024] --blocksize=$[256 * 1024] /dev/mmcblk0 --open-au-nr=2 4MiB 6.42M/s 2MiB 6.3M/s 1MiB 6.22M/s 512KiB 6.25M/s 256KiB 6.24M/s == reboot, reinsert the SD card == u@u:~/flash/flashbench$ sudo ./flashbench -O --erasesize=$[4 * 1024 * 1024] --blocksize=$[256 * 1024] /dev/mmcblk0 --open-au-nr=2 [sudo] password for u: 4MiB 6.45M/s 2MiB 6.33M/s 1MiB 6.25M/s 512KiB 6.29M/s 256KiB 6.3M/s u@u:~/flash/flashbench$ == rerun the test again == u@u:~/flash/flashbench$ sudo ./flashbench -O --erasesize=$[4 * 1024 * 1024] --blocksize=$[256 * 1024] /dev/mmcblk0 --open-au-nr=2 4MiB 6.27M/s 2MiB 6.27M/s 1MiB 6.28M/s 512KiB 6.27M/s 256KiB 6.24M/s u@u:~/flash/flashbench$ == poweroff, Poweron, insert SD card == u@u:~/flash/flashbench$ sudo ./flashbench -O --erasesize=$[4 * 1024 * 1024] --blocksize=$[256 * 1024] /dev/mmcblk0 --open-au-nr=2 [sudo] password for u: 4MiB 6.45M/s 2MiB 6.33M/s 1MiB 6.24M/s 512KiB 6.29M/s 256KiB 6.32M/s u@u:~/flash/flashbench$ == rerun the test again == u@u:~/flash/flashbench$ sudo ./flashbench -O --erasesize=$[4 * 1024 * 1024] --blocksize=$[256 * 1024] /dev/mmcblk0 --open-au-nr=2 4MiB 6.29M/s 2MiB 6.28M/s 1MiB 6.29M/s 512KiB 6.28M/s 256KiB 6.25M/s u@u:~/flash/flashbench$ On Wed, 13 Jul 2011, Chris Ball wrote: > Hi Arnd, > > On Tue, Jul 12 2011, Arnd Bergmann wrote: >> I would very much expect that to be nonreproducible. The first row >> in each test is the result of a single write() system call and does >> not get averaged out. More importantly the time for each write >> depends a lot of the state of the card before the write. >> >> For instance when you do a lot of random writes to a card, optionally >> take it out and put it into a different machine, and then do a large >> linear write, that linear write will be very slow because the >> card has to garbage collect all the random writes that were done >> earlier. After a few writes (usually one is enough), it gets back >> to the full performance. > > That makes sense. Do you think this explains Manoj getting a slower > first file copy speed (757ms vs. 480ms) after applying his patch? > (Manoj, perhaps you could retry your test without GC being needed?) > > What would we expect lowering the SD base clock frequency from 200MHz > to 50MHz to do to performance theoretically? > > Thanks, > > - Chris. > -- > Chris Ball > One Laptop Per Child > > -- ==================== Manoj Iyer Ubuntu/Canonical Hardware Enablement ====================