Re: [PATCH] Scheduling problems wrt SMP

[Linus Torvalds]

On Fri, 18 Dec 1998, David S. Miller wrote:
> 
> I can prove this.  The performance counters on the Ultra, unless
> broken, support my claims completely.

Uhh.. The performance counters may not be broken, but the Ultra SMP code
_was_ broken. You resheduled all the time, didn't you?

So of _course_ the heuristics made a big difference. My point is that on
the x86 we weren't that stupid in the first place because we used "hlt" 
to avoid rescheduling all the time. So your performance counter data is
irrelevant as far as I can tell.

                Linus

-
Linux SMP list: FIRST see FAQ at http://www.irisa.fr/prive/mentre/smp-faq/
To Unsubscribe: send "unsubscribe linux-smp" to [EMAIL PROTECTED]