Stephane and Vince,

On 09.05.08 23:06:16, stephane eranian wrote:
> My main issue which this patch is that I wonder if the K7
> support does qualify as a AMD64 PMU which is how this
> patch would implement it.
> 
> Shouldn't we rather introduce an AMD_K7 PMU model
> for libpfm. We could reuse most of the code but it would
> logically be separate.
> 
> Opinions?

Unfortunately this became a little bit confusing. Although performance
monitoring is documented mainly in the AMD64 documentation and the
specifications of K8 and familiy 10h, performance counters has been
introduced with the K7. The PMUs of K7 and K8 can be treated as
compatible with the exception of newly introduced events. From that
point of view I tend to use the same PMU model as for K8. Actually we
could change from AMD64 to AMD, but this could confuse people even
more.

So, just to use performance monitoring on K7 as on K8, it's better to
have also the same PMU model.

-Robert

-- 
Advanced Micro Devices, Inc.
Operating System Research Center
email: [EMAIL PROTECTED]


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft 
Defy all challenges. Microsoft(R) Visual Studio 2008. 
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
perfmon2-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/perfmon2-devel

Reply via email to