RE: Cpufreq for opteron

[paul.devriendt@amd.com]

> -----Original Message-----
> From: Pavel Machek [mailto:pavel@suse.cz]
> Sent: Monday, August 25, 2003 8:51 AM
> To: Devriendt, Paul
> Cc: davej@redhat.com; linux-kernel@vger.kernel.org; aj@suse.de;
> Langsdorf, Mark; Brunner, Richard
> Subject: Re: Cpufreq for opteron
> 
> 
> Hi!
> 
> > > 4) given good hardware and debugged driver, will any of those
> > > BUG_ON()s ever trigger?
> > 
> > Only if there are BIOS problems. 
> 
> In such case, I believe best idea is to leave them in as BUG_ON(). On
> broken BIOS, it will kill machine cleanly, and hopefully bios is going
> to be fixed.
> 
> If broken BIOS is seen in retail, we'll need to solve this other way.
> 
> Does this seem okay to you?

My concerns with the BUG_ON() approach are :
  1. Ease of me debugging the problem, as some of the state data I would
     want to see is global, so it might not be in a backtrace.
  2. Taking the machine down when exestuation could continue.

You have more kernel experience than I do, so I am willing to accept your
advice. I am ok with it.

Thanks. Paul.

> 								Pavel
> -- 
> When do you have a heart between your knees?
> [Johanka's followup: and *two* hearts?]
>