Re: benchmarking - it's now all(-1,0,1,5,6)% faster

Nicholas Clark <nick@unfortu.net> wrote:
:So I'm confused. It looks like some bits of perl are incredibly sensitive to
:cache alignment, or something similar. And as a consequence, perlbench is
:reliably reporting wildly varying timings because of this, and because it
:only tries a few, very specific things. Does this mean that it's still useful?

I think I remember seeing a profiler that emulates the x86 instruction set,
and so can give theoretically exact timings. Does this ring a bell for
anyone? I don't know if the emulation extended to details such as RAM
and cache sizes ...

Hugo

• benchmarking - it's now all(-1,0,1,5,6)% faster by Nicholas Clark
• Re: benchmarking - it's now all(-1,0,1,5,6)% faster by Michael G Schwern
• Re: benchmarking - it's now all(-1,0,1,5,6)% faster by hv
• Re: benchmarking - it's now all(-1,0,1,5,6)% faster by Peter Nimmervoll
• Re: benchmarking - it's now all(-1,0,1,5,6)% faster by Leopold Toetsch
• Re: benchmarking - it's now all(-1,0,1,5,6)% faster by Leopold Toetsch
• Re: benchmarking - it's now all(-1,0,1,5,6)% faster by Unknown Sender
• Re: benchmarking - it's now all(-1,0,1,5,6)% faster by A . Bergman
• Re: benchmarking - it's now all(-1,0,1,5,6)% faster by Nicholas Clark
• Re: benchmarking - it's now all(-1,0,1,5,6)% faster by hv
• Re: benchmarking - it's now all(-1,0,1,5,6)% faster by andreas.koenig
• Re: benchmarking - it's now all(-1,0,1,5,6)% faster by Nicholas Clark
• Re: benchmarking - it's now all(-1,0,1,5,6)% faster by andreas.koenig
• Re: benchmarking - it's now all(-1,0,1,5,6)% faster by Leopold Toetsch