Re: [patch] kstat change to see how much Linux SMP really scale well
Andi Kleen (ak@muc.de)
Sat, 13 Mar 1999 19:01:38 +0100
On Sat, Mar 13, 1999 at 06:08:15PM +0100, Ingo Molnar wrote:
>
> On Sat, 13 Mar 1999, Andi Kleen wrote:
>
> > > But I see a clever and more fun way to avoid losing ticks in both the irq
> > > handler and in the smp timer interrupt, and it's to always calc the delta
> > > between rdtsc inside the smp timer irq. I can go into that.
> >
> > Exactly what I had in mind. Actually I'm surprised that Linux doesn't do
> > that already @)
>
> this is a non-issue, really. It simply does not happen. I've profiled a
> slow and old dual-P100 for hours under heavy load and i never got a lost
> SMP timer interrupt.
I think it depends on the drivers. I assume your old P100
didn't have a polling SCSI controller that disabled local interrupts with
the iorequest spinlock for ages.
-Andi
--
This is like TV. I don't like TV.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/