From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arnd Bergmann Subject: Re: [comments] MMC: Reliable write support. Date: Tue, 29 Mar 2011 09:01:31 +0200 Message-ID: <201103290901.31680.arnd@arndb.de> References: <1301001751-30785-1-git-send-email-andreiw@motorola.com> <201103251614.29893.arnd@arndb.de> Mime-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Return-path: Received: from moutng.kundenserver.de ([212.227.17.9]:53746 "EHLO moutng.kundenserver.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752418Ab1C2HBi (ORCPT ); Tue, 29 Mar 2011 03:01:38 -0400 In-Reply-To: Sender: linux-mmc-owner@vger.kernel.org List-Id: linux-mmc@vger.kernel.org To: Andrei Warkentin Cc: linux-mmc@vger.kernel.org On Tuesday 29 March 2011, Andrei Warkentin wrote: > On Fri, Mar 25, 2011 at 10:14 AM, Arnd Bergmann wrote: > > I confirmed with two MMC vendors that there is no "flush". Once the > DAT transfer completes, the data is stored on non-volatile storage as > soon as the busy status is cleared. > > Reliable writes are still "more reliable" because if the DAT transfer > is interrupted (power or reset through CMD0/CMD15 or hw pin for eMMC), > you have predictable flash contents. So it makes sense to map REQ_FUA > to it (and REQ_META, I would guess). Yes, sounds good. So I guess on MLC flash, a reliable write will go to a flash page that does not have data in any of its paired pages. Arnd