From mboxrd@z Thu Jan 1 00:00:00 1970 Content-Type: multipart/mixed; boundary="===============9218985994971525584==" MIME-Version: 1.0 From: Jan Kara To: lkp@lists.01.org Subject: Re: [fs] 36e2c7421f: xfstests.generic.095.fail Date: Mon, 12 Sep 2022 13:06:46 +0200 Message-ID: <20220912110646.bn4q2vh5jxii7zrx@quack3> In-Reply-To: <20220909144406.GA10618@lst.de> List-Id: --===============9218985994971525584== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable On Fri 09-09-22 16:44:06, Christoph Hellwig wrote: > On Thu, Sep 08, 2022 at 03:01:57PM +0800, kernel test robot wrote: > > Hi Christoph, > > = > > According to the commit message of below patch, we understand that it > > removes splice support for some device drivers, but we don't have enough > > knowledge to connect the code change with testcase result. We send this > > report for your reference, please kindly check if it is an expected > > result. Thanks. > = > This might be as simple wiring up iter_file_splice_write, but I don't > really have a good way to test udf, so adding the maintainer. Indeed. Thanks for notifying me. Just adding: .splice_write =3D iter_file_splice_write, to udf_file_operations fixes the problem for me and generic/095 passes fine. I've queued the fix to my tree. Honza -- = Jan Kara SUSE Labs, CR --===============9218985994971525584==--