From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from userp1040.oracle.com ([156.151.31.81]:45429 "EHLO userp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755310AbcI2TqL (ORCPT ); Thu, 29 Sep 2016 15:46:11 -0400 Date: Thu, 29 Sep 2016 12:45:36 -0700 From: "Darrick J. Wong" Subject: Re: [PATCH 45/63] xfs: unshare a range of blocks via fallocate Message-ID: <20160929194536.GB14092@birch.djwong.org> References: <147503120985.30303.14151302091684456858.stgit@birch.djwong.org> <147503150585.30303.6967077248029323904.stgit@birch.djwong.org> <20160929170759.GJ4546@infradead.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20160929170759.GJ4546@infradead.org> Sender: linux-xfs-owner@vger.kernel.org List-ID: List-Id: xfs To: Christoph Hellwig Cc: david@fromorbit.com, linux-xfs@vger.kernel.org, Christoph Hellwig On Thu, Sep 29, 2016 at 10:07:59AM -0700, Christoph Hellwig wrote: > On Tue, Sep 27, 2016 at 07:58:25PM -0700, Darrick J. Wong wrote: > > Now that we have an fallocate flag to unshare a range of blocks, make > > XFS actually implement it. > > There is [no] unshared flag in here. Drat! I totally forgot to add the unshare flag back in! We even had a whole email thread about this months ago. :( :( Ok, I'll put in a flag. > Instead it performs an unshare without passing any flag, which will break > existing users of fallocate like the NFS server. > > v2: NFS doesn't pass around struct file pointers, which means that our > > unshare functions all crash when filp == NULL. We don't need filp > > anyway, so remove all the parts where we pass filp around. > > NFSD passes a file pointer just fine. Note that it really should > matter, but this comment confused me. Yeah, it confused me as well, so I'll just go back to your original changelog message. --D