linux-mips
[Top] [All Lists]

Re: Kernel oops on XXS1500 in au1000eth.c

To: Pete Popov <ppopov@mvista.com>
Subject: Re: Kernel oops on XXS1500 in au1000eth.c
From: Goswin von Brederlow <brederlo@informatik.uni-tuebingen.de>
Date: 13 Jan 2004 19:37:08 +0100
Cc: Goswin von Brederlow <brederlo@informatik.uni-tuebingen.de>, Linux MIPS mailing list <linux-mips@linux-mips.org>
In-reply-to: <1074018143.21864.13.camel@zeus.mvista.com>
Original-recipient: rfc822;linux-mips@linux-mips.org
References: <87lloblo27.fsf@mrvn.homelinux.org> <1074018143.21864.13.camel@zeus.mvista.com>
Sender: linux-mips-bounce@linux-mips.org
User-agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.4 (Reasonable Discussion)
Pete Popov <ppopov@mvista.com> writes:

> On Tue, 2004-01-13 at 10:07, Goswin von Brederlow wrote:
> > Hi,
> > 
> > when compiling a kernel for my XXS1500 I allways ended up with a
> > kernel oops in the network driver (au1000eth.c).
> > 
> > Finaly I checked the actual kernel source the running kernel was build
> > from and found "CONFIG_BCM5222_DUAL_PHY" was set. Setting that solves
> > the oops.
> > 
> > Maybe that could be caught in some way and prevented.
> 
> Well, the kernel shouldn't be crashing but as far as the BCM dual phy
> thing -- I'm not sure how to detect it at run time without knowing ahead
> of time that we've got one.  I admittedly haven't spent too much time
> thinking about this problem but I didn't see an easy way to handle it.
> 
> Pete

Maybe some

if (mach == MACH_XXS1500) ...

construct? The MACH_XXS1500 must be good for something.

Hopefully MyCable will change the name when they design a new board
with a different network thing.

MfG
        Goswin

<Prev in Thread] Current Thread [Next in Thread>