From: "Martin K. Petersen" Subject: Re: high iowait problem(Bug 12309 on bugzilla.kernel.org) Date: Wed, 14 Apr 2010 10:25:08 -0400 Message-ID: References: <20100331124834.192500@gmx.com> <20100413135416.b537eee8.akpm@linux-foundation.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: In-Reply-To: <20100413135416.b537eee8.akpm@linux-foundation.org> (Andrew Morton's message of "Tue, 13 Apr 2010 13:54:16 -0700") Sender: linux-kernel-owner@vger.kernel.org To: Andrew Morton Cc: Frank Ren , mingo , "a.p.zijlstra" , "jens.axboe" , linux-kernel@vger.kernel.org, stable@kernel.org, dm-devel@redhat.com, Alasdair G Kergon , "Martin K. Petersen" List-Id: dm-devel.ids >>>>> "Andrew" == Andrew Morton writes: Andrew> I'm seeing just one device-mapper change whcih went in over that Andrew> timeframe: Andrew> : commit e0f5cfa7c18b411634e73923841eccd3a4c0ce7f Andrew> : Author: Martin K. Petersen Andrew> : Date: Mon Jan 11 03:21:50 2010 -0500 Andrew> : Andrew> : DM: Fix device mapper topology stacking The topology changes do not modify anything in our runtime I/O path. They only come into play during filesystem or logical volume creation on top of devices that support reporting alignment/physical block size. And you need recent dm and filesystem utilities for that to work. -- Martin K. Petersen Oracle Linux Engineering