From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
        id S1753178AbdHQSko convert rfc822-to-8bit (ORCPT <rfc822;w@1wt.eu>);
        Thu, 17 Aug 2017 14:40:44 -0400
Received: from mx2.suse.de ([195.135.220.15]:36357 "EHLO mx1.suse.de"
        rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP
        id S1752270AbdHQSkm (ORCPT <rfc822;linux-kernel@vger.kernel.org>);
        Thu, 17 Aug 2017 14:40:42 -0400
From: Luis Henriques <lhenriques@suse.com>
To: Ilya Dryomov <idryomov@gmail.com>
Cc: "Yan\, Zheng" <zyan@redhat.com>, Sage Weil <sage@redhat.com>,
        Ceph Development <ceph-devel@vger.kernel.org>,
        "linux-kernel\@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] ceph: check negative offsets on ceph_llseek()
References: <20170728105640.16268-1-lhenriques@suse.com>
        <87pobuz4s2.fsf@hermes>
        <CAOi1vP9RYiFuscyJqetRO3p68Bj14To5whwM6EByvtjChCk0_g@mail.gmail.com>
Date: Thu, 17 Aug 2017 19:40:39 +0100
In-Reply-To: <CAOi1vP9RYiFuscyJqetRO3p68Bj14To5whwM6EByvtjChCk0_g@mail.gmail.com>
        (Ilya Dryomov's message of "Thu, 17 Aug 2017 19:57:01 +0200")
Message-ID: <87lgmixck8.fsf@hermes>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 8BIT
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

Ilya Dryomov <idryomov@gmail.com> writes:

> On Thu, Aug 17, 2017 at 3:45 PM, Luis Henriques <lhenriques@suse.com> wrote:
>> Luis Henriques <lhenriques@suse.com> writes:
>>
>>> When a user requests SEEK_HOLE or SEEK_DATA with a negative offset
>>> ceph_llseek should return -ENXIO.  Currently -EINVAL is being returned for
>>> SEEK_DATA and 0 for SEEK_HOLE.
>>
>> Ping
>>
>> This patch should make xfstest generic/448 happy.
>
> It should or it does? ;)
>

Heh, it *does* make generic/448 happy.

>
> I saw generic/448 failures on ext4 with Darrick's recent test
>change,
> haven't tried ceph yet.

The generic/448 test currently fails with the error codes described in
the commit log.  I didn't found it useful to include in the commit
log, but here's the full test output:

File system supports the default behavior.
File system does not support unwritten extents.
File system magic#: 0xc36400
Allocation size: 4194304
18. Test file with negative SEEK_{HOLE,DATA} offsets
18.01 SEEK_HOLE expected -1 or -1, got 0.                         FAIL
18.02 SEEK_DATA expected -1 with errno -6, got -22.               FAIL

seek sanity check failed!


Cheers,
-- 
Luís