gcc -O3 -S vm/interp.c
This should produce a interp.s file.
Try also modifying vm/interp.c with the registers optimizations and
tell me how it goes for the benchmarks. If you have time for it of
course.
I tried the register optimizations and the results are interesting, if
inconclusive at the moment. There are some slight gains in all the
benchmark tests, but fannkuch is twice as fast! I will use this build of
the VM to try other tests.
Do you call 2 times speed increase inconclusive ? :)
I've updated the CVS with more registers optimizations for PPC. Could
you compile+test with it and send me the interp.s one more time ? I'm
not familiar with PPC assembly but looks like things can be further
optimized at some important places.
Nicolas
--
Neko : One VM to run them all
(http://nekovm.org)