From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753518Ab1IHXPe (ORCPT ); Thu, 8 Sep 2011 19:15:34 -0400 Received: from acsinet15.oracle.com ([141.146.126.227]:64386 "EHLO acsinet15.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751483Ab1IHXP2 convert rfc822-to-8bit (ORCPT ); Thu, 8 Sep 2011 19:15:28 -0400 MIME-Version: 1.0 Message-ID: <52b5aee3-f424-42ae-830f-d1cf64fa49ef@default> Date: Thu, 8 Sep 2011 08:50:11 -0700 (PDT) From: Dan Magenheimer To: Andrew Morton Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org, jeremy@goop.org, hughd@google.com, ngupta@vflare.org, Konrad Wilk , JBeulich@novell.com, Kurt Hackel , npiggin@kernel.dk, riel@redhat.com, hannes@cmpxchg.org, matthew@wil.cx, Chris Mason , sjenning@linux.vnet.ibm.com, kamezawa.hiroyu@jp.fujitsu.com, jackdachef@gmail.com, cyclonusj@gmail.com, levinsasha928@gmail.com Subject: RE: [PATCH V8 3/4] mm: frontswap: add swap hooks and extend try_to_unuse References: <20110829164929.GA27216@ca-server1.us.oracle.com 20110907162703.7f8116b9.akpm@linux-foundation.org> In-Reply-To: <20110907162703.7f8116b9.akpm@linux-foundation.org> X-Priority: 3 X-Mailer: Oracle Beehive Extensions for Outlook 2.0.1.4.1.0 (410211) [OL 12.0.6557.5001] Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8BIT X-Source-IP: rtcsinet21.oracle.com [66.248.204.29] X-CT-RefId: str=0001.0A020207.4E694C8F.0157,ss=1,re=0.000,fgs=0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org > From: Andrew Morton [mailto:akpm@linux-foundation.org] > Sent: Wednesday, September 07, 2011 5:27 PM > To: Dan Magenheimer > Subject: Re: [PATCH V8 3/4] mm: frontswap: add swap hooks and extend try_to_unuse > > On Mon, 29 Aug 2011 09:49:29 -0700 > Dan Magenheimer wrote: > > > -static int try_to_unuse(unsigned int type) > > +int try_to_unuse(unsigned int type, bool frontswap, > > Are patches 2 and 3 in the wrong order? No, they've applied in that order and built after each patch properly for well over a year. At a minimum, frontswap.h must be created before patch 3of4, though I suppose the introduction of frontswap.c could be after patch 3of4... Note that frontswap.c (which calls try_to_unuse()) is non-functional (and isn't even built) until after patch 4of4 is applied. There is enough interdependency between the four parts that perhaps it should all be a single commit. I split it up for reviewer's convenience but apparently different reviewers use different review processes than I anticipated. :-} Bottom line though: yes, bisecting at any point in the patchset does work properly. Dan