Re: [Gcl-devel] Boyer benchmark results

[Raymond Toy]

>>>>> "Camm" == Camm Maguire <address@hidden> writes:

    Camm> Paul has separately written speculating on cache.  The only logic I
    Camm> can conceive of being behind these numbers is if 1) gcl's assembler
    Camm> output via gcc is faster in cache then that output by cmucl, 2) cmucl
    Camm> either makes use of a superior copying gc algorithm for the cons
    Camm> storage (improving locality), or makes use of explicit prefetch
    Camm> instructions for much better memory access times than gcl, at least
    Camm> for cons storage, and 3) clisp also does something like 2), but not as

CMUCL certainly has a copying GC.  Don't know if it is superior or
not.  AFAIK, CMUCL doesn't do any kind of explicit prefetch---the
compiler is not smart enough to do that by itself.

Ray