From mboxrd@z Thu Jan 1 00:00:00 1970 From: Chris Mason Subject: Re: [patch 5/8] mm: write_cache_pages integrity fix Date: Thu, 09 Oct 2008 10:12:55 -0400 Message-ID: <1223561575.14090.14.camel@think.oraclecorp.com> References: <20081009155039.139856823@suse.de> <20081009174822.621353840@suse.de> <1223556765.14090.2.camel@think.oraclecorp.com> <20081009132711.GB9941@wotan.suse.de> <1223559358.14090.11.camel@think.oraclecorp.com> <20081009135538.GC9941@wotan.suse.de> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Cc: Andrew Morton , Mikulas Patocka , linux-mm@kvack.org, linux-fsdevel@vger.kernel.org To: Nick Piggin Return-path: Received: from agminet01.oracle.com ([141.146.126.228]:54827 "EHLO agminet01.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752612AbYJIONS (ORCPT ); Thu, 9 Oct 2008 10:13:18 -0400 In-Reply-To: <20081009135538.GC9941@wotan.suse.de> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Thu, 2008-10-09 at 15:55 +0200, Nick Piggin wrote: > On Thu, Oct 09, 2008 at 09:35:58AM -0400, Chris Mason wrote: > > On Thu, 2008-10-09 at 15:27 +0200, Nick Piggin wrote: > > > > I don't think do_sync_mapping_range is broken as is. It simply splits > > the operations into different parts. The caller can request that we > > wait for pending IO first. > > It is. Not because of it's whacky API, but because it uses WB_SYNC_NONE. > > > > WB_SYNC_NONE none just means don't wait for IO in flight, and there are > > valid uses for it that will slow down if you switch them all to > > WB_SYNC_ALL. > > To write_cache_pages it means that, but further down the chain (eg. > block_write_full_page) it also means not to wait on other stuff. > > It has broadly meant "don't worry about data integirty" for a long time > AFAIKS. Sadly it has broadly meant different things to different people ;) You're right, block_write_full_page is broken. -chris