Re: performance measurement and caches

Hans Boehm (boehm@parc.xerox.com) wrote:
> [Benchmark speed varied by a factor of two, apparently depending on
> memory location and cache behavior.]


Mary Fernandez wrote:
> I observed similar effects of procedure placement on cache
> performance on both MIPS and Intel 486. For a small set of large
> Modula-3 programs, procedure placement alone effected runtime by
> upto 15%; on the Intel up to 10%. Smaller effects than those you
> observed, but still confounding if you're trying to measure
> performance differences in that neighborhood. Just injecting NOPs
> at procedure boundaries perturbs alignment enough to produce
> measurable (>10%) differences in elapsed time.


I believe this is mostly due to the way 486, Pentium and PPro handles
code prefetching, both 486 and PPro really likes to have the top of
busy loops aligned near the beginning of a cache line, i.e. you can
get this effect in single-tasking mode, with no system traffic at all.


The Pentium relaxed the branch target requirement to 32-bit boundaries.


--
-Terje Mathisen (include std disclaimer) <Terje.Mathisen@hda.hydro.com>
--

Return to the comp.compilers page.
Search the comp.compilers archives again.