From: Theodore Tso Subject: Re: [PATCH 0/4] (RESEND) ext3[34] barrier changes Date: Mon, 19 May 2008 10:46:54 -0400 Message-ID: <20080519144654.GC15035@mit.edu> References: <482DDA56.6000301@redhat.com> <20080516130545.845a3be9.akpm@linux-foundation.org> <87ej7zcrqv.fsf@basil.nowhere.org> <200805190926.41970.chris.mason@oracle.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Andi Kleen , Andrew Morton , Eric Sandeen , linux-ext4@vger.kernel.org, linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org, jamie@shareable.org To: Chris Mason Return-path: Received: from BISCAYNE-ONE-STATION.MIT.EDU ([18.7.7.80]:40495 "EHLO biscayne-one-station.mit.edu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756727AbYESOtt (ORCPT ); Mon, 19 May 2008 10:49:49 -0400 Content-Disposition: inline In-Reply-To: <200805190926.41970.chris.mason@oracle.com> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Mon, May 19, 2008 at 09:26:41AM -0400, Chris Mason wrote: > > > > Didn't ext4 have some new checksum trick to avoid them? > > I didn't think checksumming avoided barriers completely. Just the barrier > before the commit block, not the barrier after. Funny thing, I was looking in this over the weekend. It currently avoids barriers entirely if journal_async_commit is enabled (which is not the default); if enabled, it effectively forces barrier=0. This is IMHO a bug. I'm working on a patch where "barrier=1" will use a barrier before and after, and "barrier=1,journal_async_commit" will use a barrier after. - Ted