On Wednesday 13 January 2010 10:54, Oliver Hartkopp wrote:
> Matthias Fuchs wrote:
> 
> >>>>> Any ideas where to dig?
> >>>> Please check in can_send() and can_rcv() if there are the same pointer 
> >>>> values
> >>>> in skb->sk .
> >>> Good advice. I added some printks. On a 2.6.28 kernel skb->sk is 
> >>> identical in 
> >>> can_send and can_rcv. Loopback is working fine.
> >>>
> >>> On 2.6.18 skb->sk is NULL in can_rcv.
> > 
> > I did some further testing. Doing the same tests on a Debian Etch system
> > were successful (Debian kernel 2.6.18-6-486). The problem seems to be with
> > the Centos/RHEL kernel 2.6.18-164.6.1.el5. This kernel contains a huge 
> > amount
> > of Redhat patches. I will see if any of them touch net/core/dev.c. Grr :-(
> 
> Maybe a diff with a vanilla 2.6.18 will help - and check especially for
> skb_orphan() calls ;-)

I dig a little bit deeper into the Centos/RHEL world. The RL kernel comes with
about 430 patches for the net code!!! And even more for the rest of the kernel.
Couldn't believe that number.

I took the net/core/dev.c and diffed it against a vanilla 2.6.18:

see http://www.pastebin.org/75442

There is a new skb_orphan in chunk "-1828,10 +1951,13 @@".


This is the redhat patch that caused it:

From: Herbert Xu <[email protected]>
Date: Wed, 6 May 2009 11:37:16 +0800
Subject: [net] tun: add packet accounting
Message-id: [email protected]
O-Subject: Re: [RHEL5.4 PATCH] tun: Add packet accounting
Bugzilla: 495863
RH-Acked-by: Neil Horman <[email protected]>
RH-Acked-by: Michael S. Tsirkin <[email protected]>

Hi:

RHEL5.4 BZ #495863

We need to add packet accounting to the tun driver so that users
such as virtio-net gets congestion feedback which is necessary to
prevent packet loss for protocols lacking congestion conctrol (such
as UDP) when used in a guest.

This is a backport of the following upstream patches:

commit 33dccbb050bbe35b88ca8cf1228dcf3e4d4b3554
Author: Herbert Xu <[email protected]>
Date:   Thu Feb 5 21:25:32 2009 -0800

commit 4cc7f68d65558f683c702d4fe3a5aac4c5227b97
Author: Herbert Xu <[email protected]>
Date:   Wed Feb 4 16:55:54 2009 -0800

commit 9a279bcbe347496799711155ed41a89bc40f79c5
Author: Herbert Xu <[email protected]>
Date:   Wed Feb 4 16:55:27 2009 -0800


I took a glace an the very last mentioned backported patch 9a279bcb.
Hmm this is one of our old frieds :-) 

See: http://lists.berlios.de/pipermail/socketcan-users/2009-June/000959.html

So to come to a conclusion: Good Socket-CAN, bad RH.

You really find 2.6.30 backports in RHEL 2.6.18 kernels.

So I think we have to revert he above patch in the RHEL kernel.

Matthias
_______________________________________________
Socketcan-core mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/socketcan-core

Reply via email to