From mboxrd@z Thu Jan 1 00:00:00 1970 From: OGAWA Hirofumi Subject: Re: [PATCH 0/2] MMC discard support (was [PATCH 0/7] Discard requests, v2) Date: Sun, 24 Aug 2008 22:39:02 +0900 Message-ID: <87fxour0qx.fsf@devron.myhome.or.jp> References: <1218299181.26926.88.camel@pmac.infradead.org> <20080816190858.4d150ea1@mjolnir.drzeus.cx> <20080822092448.GC20055@kernel.dk> <20080822131314.23ba745a@mjolnir.drzeus.cx> <20080822112010.GJ20055@kernel.dk> <87abf5ayvw.fsf@devron.myhome.or.jp> <20080823010255.1b4ad88c@mjolnir.drzeus.cx> <87vdxs7ga8.fsf@devron.myhome.or.jp> <20080824132335.15adc7a6@mjolnir.drzeus.cx> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Jens Axboe , linux-fsdevel@vger.kernel.org, David Woodhouse To: Pierre Ossman Return-path: Received: from mail.parknet.ad.jp ([210.171.162.6]:50494 "EHLO mail.officemail.jp" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753479AbYHXNjJ (ORCPT ); Sun, 24 Aug 2008 09:39:09 -0400 In-Reply-To: <20080824132335.15adc7a6@mjolnir.drzeus.cx> (Pierre Ossman's message of "Sun, 24 Aug 2008 13:23:35 +0200") Sender: linux-fsdevel-owner@vger.kernel.org List-ID: Pierre Ossman writes: > On Sat, 23 Aug 2008 08:59:11 +0900 > OGAWA Hirofumi wrote: > >> Pierre Ossman writes: >> > >> > I have tried those two and others, and I cannot see any noticeable >> > difference in performance. Please perform your own tests though as I >> > might have overlooked something. >> >> I see, sounds like erase commands of sd/mmc was ignored. > > If only. :) Yes :) > The device does take some time to perform the command, and the data is > indeed gone after the command has completed. So there is no doubt that > the sector gets erased. > > It might simply be that so few (no?) SD/MMC readers actually implement > these commands, so the card vendors haven't bothered implementing any > decent handling of them. I see. Umm... Or. The pre-erased card has no improvement. So, the card may be doing wear-leveling (ERASE_BLK_EN=1 may imply?), and write bandwidth is slow enough, it might hide the slowness of erase well... -- OGAWA Hirofumi