From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from verein.lst.de ([213.95.11.211]:33390 "EHLO newverein.lst.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755263AbdIHHdZ (ORCPT ); Fri, 8 Sep 2017 03:33:25 -0400 Date: Fri, 8 Sep 2017 09:33:24 +0200 From: Christoph Hellwig Subject: Re: [PATCH 07/17] xfs: use the proper state defines in xfs_bmap_del_extent_real Message-ID: <20170908073324.GA25065@lst.de> References: <20170903072956.3175-1-hch@lst.de> <20170903072956.3175-8-hch@lst.de> <20170907154708.GG61662@bfoster.bfoster> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20170907154708.GG61662@bfoster.bfoster> Sender: linux-xfs-owner@vger.kernel.org List-ID: List-Id: xfs To: Brian Foster Cc: Christoph Hellwig , linux-xfs@vger.kernel.org On Thu, Sep 07, 2017 at 11:47:09AM -0400, Brian Foster wrote: > The code looks correct because the states ultimately map to the correct > logic, but I think this is actually a misuse of the bmap states. The > contig states are used to define left and right extent contiguity. Here > (and in del_extent_[delay|cow]()) we use them to track extent > replacement alignment, which should be represented by the left and right > filling states (as is done for delalloc/unwritten conversion helpers). > Perhaps we should fix that all up? Sure, will do when reposting.