lkml.org 
[lkml]   [2009]   [Jul]   [1]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH -tip] perf_counter: Add Generalized Hardware FPU support for AMD

* Paul Mackerras <paulus@samba.org> wrote:

> Ingo Molnar writes:
>
> > 19583739 vec-adds ( 2.01x scaled)
> > 20856051 vec-muls ( 2.01x scaled)
> > 20856051 vec-divs ( 2.01x scaled)
> > 25100224054 vec-idle-cycles ( 1.99x scaled)
> > 12540131 vec-busy-cycles ( 1.99x scaled)
> > 42286702 vec-ops ( 2.01x scaled)
> >
> > Paulus: would this categorization fit PowerPC too?
>
> Conceptually that looks nice, but unfortunately we don't have
> events that correspond to that categorization on any PowerPC with
> vector hardware (VMX/Altivec). POWER6 seems to have the most
> vector events, and they are mostly divided up along the lines of
> simple / complex / permute / load / store operations, and whether
> they are integer or floating-point operations.

Here's what we have on x86:

20177177044 vec-adds (scaled from 66.63%)
34101687027 vec-muls (scaled from 66.64%)
3984060862 vec-divs (scaled from 66.71%)
26349684710 vec-idle-cycles (scaled from 66.65%)
9052001905 vec-stall-cycles (scaled from 66.66%)
76440734242 vec-ops (scaled from 66.71%)

Could at least the idle/busy/stall/total generic stats be filled in
on powerpc, with a reasonable enough approximation? Those
utilization metrics are the most important ones when one tries to
figure out how well utilized the vector units are.

Ingo


\
 
 \ /
  Last update: 2009-07-01 15:15    [W:0.137 / U:0.040 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site