Return-Path: Received: from mx2.netapp.com ([216.240.18.37]:49788 "EHLO mx2.netapp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754670Ab1BOPDo convert rfc822-to-8bit (ORCPT ); Tue, 15 Feb 2011 10:03:44 -0500 Subject: Re: [PATCH 10/16] pnfs: wave 3: coelesce across layout stripes From: Trond Myklebust To: "William A. (Andy) Adamson" Cc: linux-nfs@vger.kernel.org, Fred Isaman , Andy Adamon , Dean Hildebrand , Fred Isaman , Benny Halevy , Boaz Harrosh , Oleg Drokin , Tao Guo In-Reply-To: References: <1297711116-3139-1-git-send-email-andros@netapp.com> <1297711116-3139-11-git-send-email-andros@netapp.com> <1297726920.23841.51.camel@heimdal.trondhjem.org> Content-Type: text/plain; charset="UTF-8" Date: Tue, 15 Feb 2011 10:03:40 -0500 Message-ID: <1297782220.10103.13.camel@heimdal.trondhjem.org> Sender: linux-nfs-owner@vger.kernel.org List-ID: MIME-Version: 1.0 On Tue, 2011-02-15 at 09:43 -0500, William A. (Andy) Adamson wrote: > On Mon, Feb 14, 2011 at 6:42 PM, Trond Myklebust > wrote: > > On Mon, 2011-02-14 at 14:18 -0500, andros@netapp.com wrote: > >> From: Fred Isaman > >> > >> Add a pg_test layout driver hook which is used to avoid coelescing I/O across > >> layout stripes. > > > > Doesn't this belong before [PATCH 09/16] pnfs: wave 3: shift > > pnfs_update_layout locations? > > The pg_test uses the pg_lseg declared in [PATCH 09/16] pnfs: wave 3: > shift pnfs_update_layout locations, which is why the patches are > ordered this way. What prevents you from moving the pg_lseg declaration into this patch, and just relying on the initialisation being NULL? The current ordering means that applying 9/16 without 10/16 gives rise to broken stripe sizes. -- Trond Myklebust Linux NFS client maintainer NetApp Trond.Myklebust@netapp.com www.netapp.com