Re: forcedeth - WOL [SOLVED]

[Andrew Morton <akpm@osdl.org>]

On Thu, 28 Sep 2006 03:01:33 +0200
Björn Steinbrink <B.Steinbrink@gmx.de> wrote:

> On 2006.09.28 02:40:53 +0200, Björn Steinbrink wrote:
> > On 2006.09.28 02:04:48 +0200, Björn Steinbrink wrote:
> > > 
> > > Hi Andrew,
> > > 
> > > 
> > > On 2006.09.27 16:57:04 -0700, Andrew Morton wrote:
> > > > On Wed, 27 Sep 2006 22:38:06 +0200
> > > > Martin Filip <bugtraq@smoula.net> wrote:
> > > > 
> > > > > Hi,
> > > > > 
> > > > > Bj__rn Steinbrink p____e v St 27. 09. 2006 v 20:38 +0200:
> > > > > 
> > > > > > Did you check that WOL was enabled? I need to re-activate it after each
> > > > > > boot (I guess that's normal, not sure though).
> > > > > > The output of "ethtool eth0" should show:
> > > > > > 
> > > > > >         Supports Wake-on: g
> > > > > >         Wake-on: g
> > > > > > 
> > > > > Yes, of course :)
> > > > > 
> > > > > > Also, I remember a bugzilla entry in which it was said that the MAC was
> > > > > > somehow reversed by the driver. I that is still the case (I can't find
> > > > > > the bugzilla entry right now), you might just reverse the MAC address in
> > > > > > your WOL packet to workaround the bug.
> > > > > 
> > > > > Hey! this is really crazy :) but it works! To bo honest - I really do
> > > > > not know what crazy bug could cause problems like this. I thought it's
> > > > > NIC thing to manage all the work about WOL. I thought OS only sets NIC
> > > > > into "WOL mode".
> > > > > 
> > > > > But seeing this - one packet for windows and one magic packet for linux
> > > > > driver - I really do not get it.
> > > > > 
> > > > 
> > > > Are you saying that byte-reversing the MAC address make WOL work correctly?
> > > > 
> > > > What tool do you use to send the packet, and how is it being invoked?
> 
> As I like replying to myself, I of course intentionally forgot to answer
> this question... *sigh*
> 
> I used etherwake this time, but tried with a self-made tool (which works
> with my Broadcom NIC) when I discovered the bug report.
> 
> etherwake 01:23:45:67:89:ab -- works when I shutdown during POST
> etherwake ab:89:67:45:23:01 -- works when WOL was enabled with ethtool

OK, I use etherwake too.

> > > > Do we know if this reversal *always* happens with this driver, or only
> > > > sometimes?
> 
> I only tried 2.6.18 twice this time, but when I wrote my own tool to do
> it, I had probably 20-30 power on -> ethtool -> poweroff cycles before I
> decided to look into Bugzilla. As it looked like being fixed already and
> I did use the nForce NIC for testing only, I didn't spend any further
> time on it back then.

What I'm angling towards is: "is this just a driver bug"?