From mboxrd@z Thu Jan  1 00:00:00 1970
From: Robin Holt <holt@sgi.com>
Subject: Re: gianfar.c null pointer deref in gfar_start_xmit().
Date: Wed, 10 Aug 2011 04:32:46 -0500
Message-ID: <20110810093246.GL4926@sgi.com>
References: <20110803024438.GH4926@sgi.com>
 <20110809065407.GF3709@sgi.com>
 <1312873813.2531.54.camel@edumazet-laptop>
 <20110809130615.GT4926@sgi.com>
 <1312956061.2547.74.camel@edumazet-laptop>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: QUOTED-PRINTABLE
Cc: Robin Holt <holt@sgi.com>,
	Sandeep Gopalpet <Sandeep.Kumar@freescale.com>,
	"David S. Miller" <davem@davemloft.net>, netdev@vger.kernel.org
To: Eric Dumazet <eric.dumazet@gmail.com>
Return-path: <netdev-owner@vger.kernel.org>
Received: from relay3.sgi.com ([192.48.152.1]:43444 "EHLO relay.sgi.com"
	rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP
	id S1752633Ab1HJJcs (ORCPT <rfc822;netdev@vger.kernel.org>);
	Wed, 10 Aug 2011 05:32:48 -0400
Content-Disposition: inline
In-Reply-To: <1312956061.2547.74.camel@edumazet-laptop>
Sender: netdev-owner@vger.kernel.org
List-ID: <netdev.vger.kernel.org>

On Wed, Aug 10, 2011 at 08:01:01AM +0200, Eric Dumazet wrote:
> Le mardi 09 ao=FBt 2011 =E0 08:06 -0500, Robin Holt a =E9crit :
> > On Tue, Aug 09, 2011 at 09:10:13AM +0200, Eric Dumazet wrote:
> > > Le mardi 09 ao=FBt 2011 =E0 01:54 -0500, Robin Holt a =E9crit :
> > > > On Tue, Aug 02, 2011 at 09:44:38PM -0500, Robin Holt wrote:
> > > > >=20
> > > > > While using the v3.0 kernel on a Freescale P1010RDB with 3 mi=
nor patches
> > > > > (None which affect gianfar.c), I get a NULL pointer deref at:
> > > > >=20
> > > > > static int gfar_start_xmit(struct sk_buff *skb, struct net_de=
vice *dev)
> > > > > {
> > > > > ...
> > > > > 	regs =3D tx_queue->grp->regs;
> > > > >=20
> > > > > I put a BUG_ON(tx_queue->grp) just before this line and it di=
d trip.
> > > > > I have not looked at this any more than that.
> > > > >=20
> > > > > Any suggestions would be welcome.   To reproduce, all I need =
to do is
> > > > > a few sequences of pings.
> > > >=20
> > > > I was able to reproduce this with the net-next-2.6 kernel as we=
ll.
> > > >=20
> > >=20
> > > This driver incorrectly assumes a non dense txqueue array is poss=
ible
> > > for a netdev, but its not true.
> > >=20
> > > In the meantime, you could force it to use one tx_queue only.
> > >=20
> > > tx_queues =3D (u32 *)of_get_property(np, "fsl,num_tx_queues", NUL=
L);
> > > num_tx_qs =3D tx_queues ? *tx_queues : 1;
> >=20
> > I fixed up the .dts file and now it works.  I have not tested the R=
GMII
> > interface yet so I do not know if that supports multiple queues.  I=
 assume
> > I could find it in the documentation.  What might I be looking for?
> >=20
>=20
> What exact change did you in dts ?

I changed arch/powerpc/boot/dts/p1010si.dtsi and modified all lines tha=
t
had fsl,num_rx_queues and fsl,num_tx_queues then recompiled and rebuilt
my multi-image boot file.

It looks like my boot loader might be missing something.  The p1010
reference manual has a section 15.7.2 Multigroup Mode Initialization.
I assume that sections initialize before using multigroup mode steps ar=
e
not getting done in the boot loader.  Does that possibly sound like the
right section of the manual?

>=20
> Assuming your p1010rdb is single core powerpc, I fail to see why 8
> queues should be enabled on NIC. It only increases number of interrup=
ts
> if multiple flows are in use, and memory footprint.

Thanks,
Robin