> But didn't that 50% include those that went through
> indirections (which aren't
> caught by bitwise semitagging) ?
Yes, but the number of indirection enters is usually quite low (I have
ticky files pre-generated if you'd like to take a look).
Which reminds me: we can already eliminate the enter done by an
indirection by having several different types of indirection (9, to be
exact). I keep meaning to try this out, should only be a couple of
hours work.
> Of course, the best way to answer this question is to get
> some numbers :-)
Absolutely :)
> If I don't use the "-fspec-eval" flag on the spec branch,
> then it behaves in
> essentially the same way as the HEAD. Virtually all my
> changes only take affect when "-fspec-eval" is used.
>
> I can thus see how bitwise semi-tagging affects a
> non-speculative branch
> without having to patch the HEAD first :-))
> If it does turn out to be a big win, then I'll patch it in.
That'd be great. I use cachegrind to do most of my performance
monitoring these days - install cachegrind and run nofib with 'make
EXTRA_RUNTEST_OPTS=-cachegrind'. If you don't have much time, just run
the spectral suite and specify 'mode=fast' (the real suite doesn't
support fast yet). The nofib-analyse program knows how to interpret
cachegrind numbers.
Cheers,
Simon
_______________________________________________
Cvs-ghc mailing list
[EMAIL PROTECTED]
http://www.haskell.org/mailman/listinfo/cvs-ghc
