Re: [PATCH 18/18] 2.6.17.9 perfmon2 patch for review: new x86_64 files

From: Stephane Eranian
Date: Wed Aug 23 2006 - 06:47:31 EST


Andi,

On Wed, Aug 23, 2006 at 12:19:44PM +0200, Andi Kleen wrote:
> > +config X86_64_PERFMON_AMD64
> > + tristate "Support 64-bit mode AMD64 hardware performance counters"
> > + depends on PERFMON
> > + default m
>
> No default m please. If someone just presses return in make oldconfig
> with a new kernel they don't want all kinds of new random optional drivers.
>
> I think I would prefer to call it _K8, because in theory new AMD CPUs
> might have difference performance counters.
>
I have a second thought on this. AMD has architected the performance counters.
Their specification is not part of a model specific documentation but
part of the AMD64 architecure. If they were to radically change the
way performance counters are defined, then the processor would violate the
architecture. They can certainely extened the architecture, e.g., have more
than 4 counters. Those new counters could be model-specific, in which case
I would create a model-specific PMU description for it. If they want
to extened the capability for all new processors, I think a cleaner
way would be to update the architecture and then we could update the
descsription we have.

What I don't not quite understand with the K7, K8 terminology is the
relation/dependencies with the AMD64 architecture specification.

--
-Stephane
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/