From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mx1.redhat.com ([209.132.183.28]:60406 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753333AbdLFI2N (ORCPT ); Wed, 6 Dec 2017 03:28:13 -0500 Date: Wed, 6 Dec 2017 16:28:10 +0800 From: Eryu Guan Subject: Re: [PATCH v7] generic: initial fiemap range query test Message-ID: <20171206082810.GS2749@eguan.usersys.redhat.com> References: <1512057927-15022-1-git-send-email-nborisov@suse.com> <20171205082406.GK2749@eguan.usersys.redhat.com> <20171205165127.GD19219@magnolia> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20171205165127.GD19219@magnolia> Sender: fstests-owner@vger.kernel.org To: "Darrick J. Wong" Cc: Nikolay Borisov , linux-xfs@vger.kernel.org, fstests@vger.kernel.org, sandeen@sandeen.net List-ID: On Tue, Dec 05, 2017 at 08:51:27AM -0800, Darrick J. Wong wrote: > On Tue, Dec 05, 2017 at 04:24:06PM +0800, Eryu Guan wrote: > > On Thu, Nov 30, 2017 at 06:05:27PM +0200, Nikolay Borisov wrote: > > > Fiemap gained support for passing in optional offset len > > > which denote the range requested, so this patch adds > > > testcases for this functionality. Aditionally, a special "ranged" > > > argument is added to the require_xfs_io_command which checks > > > for the presence of fiemap range support. > > > > > > Signed-off-by: Nikolay Borisov > > > --- > > > > > > For the time being this test is expected to fail on XFS. > > > > > > V7: > > > * Adjusted for 64k block size filesystem > > > * Only use/require a test device and not a scratch device > > > > v7 looks good to me, test passes on ppc64 hosts too, thanks! > > > > But it seems that it's not decided yet whether xfs should trim the > > ranged fiemap query request, I'd like to see the expected behavior > > decided first. (BTW, v7 passes with latest RHEL7 kernel too, seems > > extents are not trimmed there.) > > The iomap callers need to tell iomap_apply if they want exactly the > range they asked for and no more (most of the write paths) or if they > can handle outside ranges (fiemap), and then we can turn on > XFS_BMAPI_ENTIRE reporting in xfs_iomap_begin. That means.. I can merge the test as-is? Thanks, Eryu