From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from bombadil.infradead.org ([198.137.202.9]:59183 "EHLO bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750707AbcHAICY (ORCPT ); Mon, 1 Aug 2016 04:02:24 -0400 Date: Mon, 1 Aug 2016 01:02:23 -0700 From: Christoph Hellwig To: "Darrick J. Wong" Cc: david@fromorbit.com, linux-fsdevel@vger.kernel.org, vishal.l.verma@intel.com, bfoster@redhat.com, xfs@oss.sgi.com Subject: Re: [PATCH 11/47] xfs: move deferred operations into a separate file Message-ID: <20160801080223.GB30547@infradead.org> References: <146907695530.25461.3225785294902719773.stgit@birch.djwong.org> <146907703710.25461.16650495404061662831.stgit@birch.djwong.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <146907703710.25461.16650495404061662831.stgit@birch.djwong.org> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: I looked over this again and I really don't see the use case of merging it. Yes, the freed extent, rmap and reflink code is fairly similar, but there is all kinds of subtile differences that we need to paper over using methods and flags. I think we're better off not trying to share this code and have a separate, but easily understandable implementation for each btree. At least for the traditional traditional freed extent case the new code also is a lot less optimal than the previous version. From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from relay.sgi.com (relay3.corp.sgi.com [198.149.34.15]) by oss.sgi.com (Postfix) with ESMTP id 1EC877CFD for ; Mon, 1 Aug 2016 03:02:34 -0500 (CDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by relay3.corp.sgi.com (Postfix) with ESMTP id A1FD7AC002 for ; Mon, 1 Aug 2016 01:02:30 -0700 (PDT) Received: from bombadil.infradead.org ([198.137.202.9]) by cuda.sgi.com with ESMTP id qb8LzLZHVbngJ7xQ (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128 verify=NO) for ; Mon, 01 Aug 2016 01:02:29 -0700 (PDT) Date: Mon, 1 Aug 2016 01:02:23 -0700 From: Christoph Hellwig Subject: Re: [PATCH 11/47] xfs: move deferred operations into a separate file Message-ID: <20160801080223.GB30547@infradead.org> References: <146907695530.25461.3225785294902719773.stgit@birch.djwong.org> <146907703710.25461.16650495404061662831.stgit@birch.djwong.org> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <146907703710.25461.16650495404061662831.stgit@birch.djwong.org> List-Id: XFS Filesystem from SGI List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Errors-To: xfs-bounces@oss.sgi.com Sender: xfs-bounces@oss.sgi.com To: "Darrick J. Wong" Cc: linux-fsdevel@vger.kernel.org, vishal.l.verma@intel.com, bfoster@redhat.com, xfs@oss.sgi.com I looked over this again and I really don't see the use case of merging it. Yes, the freed extent, rmap and reflink code is fairly similar, but there is all kinds of subtile differences that we need to paper over using methods and flags. I think we're better off not trying to share this code and have a separate, but easily understandable implementation for each btree. At least for the traditional traditional freed extent case the new code also is a lot less optimal than the previous version. _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs