From: Anand Avati <anand.avati-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Subject: Re: regressions due to 64-bit ext4 directory cookies
Date: Thu, 28 Mar 2013 15:14:50 -0700
Message-ID: <CAFboF2xkvXx9YFYxBXupwg=s=3MaeQYm2KK2m8MFtEBPsxwQ7Q@mail.gmail.com>
References: <20130213224720.GE5938@thunk.org>
	<CAFboF2z1akN_edrY_fT915xfehfHGioA2M=PSHv0Fp3rD-5v5A@mail.gmail.com>
	<20130213230511.GW14195@fieldses.org>
	<20130213234430.GF5938@thunk.org>
	<CAFboF2zS+YAa0uUxMFUAbqgPh3Kb4xZu40WUjLyGn8qPoP+Oyw@mail.gmail.com>
	<5151BD5F.30607@itwm.fraunhofer.de> <5151C33E.2070008@redhat.com>
	<20130328140744.GA4989@thunk.org>
	<20130328175205.GD16651@lenny.home.zabbo.net>
	<CAFboF2ztc06G00z8ga35NrxgnT2YgBiDECgU_9kvVA_Go1_Bww@mail.gmail.com>
	<20130328183153.GG7080@fieldses.org>
	<CAFboF2w49Lc0vM0SerbJfL9_RuSHgEU+y_Yk7F4pLxeiqu+KRg@mail.gmail.com>
	<51549D74.1060703@redhat.com>
Mime-Version: 1.0
Content-Type: multipart/mixed; boundary="===============1707669704838522986=="
Cc: Eric Sandeen <sandeen-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>, linux-nfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	Theodore Ts'o <tytso-3s7WtUTddSA@public.gmane.org>, Zach Brown <zab-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
	Bernd Schubert <bernd.schubert-mPn0NPGs4xGatNDF+KUbs4QuADTiUCJX@public.gmane.org>,
	linux-ext4-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, gluster-devel-qX2TKyscuCcdnm+yROfE0A@public.gmane.org
To: Jeff Darcy <jdarcy-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
Return-path: <gluster-devel-bounces+gcfgd-gluster-devel=m.gmane.org-qX2TKyscuCcdnm+yROfE0A@public.gmane.org>
In-Reply-To: <51549D74.1060703-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
List-Unsubscribe: <https://lists.nongnu.org/mailman/options/gluster-devel>,
	<mailto:gluster-devel-request-qX2TKyscuCcdnm+yROfE0A@public.gmane.org?subject=unsubscribe>
List-Archive: <http://lists.nongnu.org/archive/html/gluster-devel>
List-Post: <mailto:gluster-devel-qX2TKyscuCcdnm+yROfE0A@public.gmane.org>
List-Help: <mailto:gluster-devel-request-qX2TKyscuCcdnm+yROfE0A@public.gmane.org?subject=help>
List-Subscribe: <https://lists.nongnu.org/mailman/listinfo/gluster-devel>,
	<mailto:gluster-devel-request-qX2TKyscuCcdnm+yROfE0A@public.gmane.org?subject=subscribe>
Errors-To: gluster-devel-bounces+gcfgd-gluster-devel=m.gmane.org-qX2TKyscuCcdnm+yROfE0A@public.gmane.org
Sender: gluster-devel-bounces+gcfgd-gluster-devel=m.gmane.org-qX2TKyscuCcdnm+yROfE0A@public.gmane.org
List-Id: linux-ext4.vger.kernel.org

--===============1707669704838522986==
Content-Type: multipart/alternative; boundary=001a11c1d64cba986a04d9037b05

--001a11c1d64cba986a04d9037b05
Content-Type: text/plain; charset=ISO-8859-1

On Thu, Mar 28, 2013 at 12:43 PM, Jeff Darcy <jdarcy-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> wrote:

> On 03/28/2013 02:49 PM, Anand Avati wrote:
> > Yes, it should, based on the theory of how ext4 was generating the
> > 63bits. But Jeff's test finds that the experiment is not matching the
> > theory.
>
> FWIW, I was able to re-run my test in between stuff related to That
> Other Problem.  What seems to be happening is that we read correctly
> until just after d_off 0x4000000000000000, then we suddenly wrap around
> - not to the very first d_off we saw, but to a pretty early one (e.g.
> 0x0041b6340689a32e).  This is all on a single brick, BTW, so it's pretty
> easy to line up the back-end and front-end d_off values which match
> perfectly up to this point.
>
> I haven't had a chance to ponder what this all means and debug it
> further.  Hopefully I'll be able to do so soon, but I figured I'd
> mention it in case something about those numbers rang a bell.
>

Of course, the unit tests (with artificial offsets) were done with brick
count >= 2. You have tested with DHT subvol count=1, which was not tested,
and sure enough, the code isn't handling it well. Just verified with the
unit tests that brick count = 1 condition fails to return the same d_off.

Posting a fixed version. Thanks for the catch!

Avati

--001a11c1d64cba986a04d9037b05
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<br><br><div class=3D"gmail_quote">On Thu, Mar 28, 2013 at 12:43 PM, Jeff D=
arcy <span dir=3D"ltr">&lt;<a href=3D"mailto:jdarcy-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org" target=3D"_=
blank">jdarcy-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org</a>&gt;</span> wrote:<br><blockquote class=3D"gmai=
l_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left=
:1ex">
<div class=3D"im">On 03/28/2013 02:49 PM, Anand Avati wrote:<br>
&gt; Yes, it should, based on the theory of how ext4 was generating the<br>
&gt; 63bits. But Jeff&#39;s test finds that the experiment is not matching =
the<br>
&gt; theory.<br>
<br>
</div>FWIW, I was able to re-run my test in between stuff related to That<b=
r>
Other Problem. =A0What seems to be happening is that we read correctly<br>
until just after d_off 0x4000000000000000, then we suddenly wrap around<br>
- not to the very first d_off we saw, but to a pretty early one (e.g.<br>
0x0041b6340689a32e). =A0This is all on a single brick, BTW, so it&#39;s pre=
tty<br>
easy to line up the back-end and front-end d_off values which match<br>
perfectly up to this point.<br>
<br>
I haven&#39;t had a chance to ponder what this all means and debug it<br>
further. =A0Hopefully I&#39;ll be able to do so soon, but I figured I&#39;d=
<br>
mention it in case something about those numbers rang a bell.<br></blockquo=
te><div><br></div><div>Of course, the unit tests (with artificial offsets) =
were done with brick count &gt;=3D 2. You have tested with DHT subvol count=
=3D1, which was not tested, and sure enough, the code isn&#39;t handling it=
 well. Just verified with the unit tests that brick count =3D 1 condition f=
ails to return the same d_off.</div>
<div><br></div><div>Posting a fixed version. Thanks for the catch!</div><di=
v><br></div><div>Avati</div><div><br></div></div>

--001a11c1d64cba986a04d9037b05--


--===============1707669704838522986==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
Gluster-devel mailing list
Gluster-devel-qX2TKyscuCcdnm+yROfE0A@public.gmane.org
https://lists.nongnu.org/mailman/listinfo/gluster-devel

--===============1707669704838522986==--