Paolo Bonzini wrote:
I am agree, benchmarking is evil. My favorite benchmark is gcc
because it is my tool and I work on it. Compilation of gcc from
Spec2000 is >12% slower with df including last Paolo's patches
(which are a really big improvement).
Interesting enough, 176.gcc has a pretty bad result for runtime too in
SPEC2000:
176.gcc 1100 67.8 1622 1100 72.9 1509*
176.gcc 1100 68.1 1614* 1100 72.9 1509
176.gcc 1100 68.8 1600 1100 72.6 1515
By your reasoning, there is some probability that fixing this one
performance regression brings dataflow branch into shape for merging
into mainline... (even though so much has changed since 2.7.2 which is
used in SPEC2000).
Usually I check code size fo x86/x86_64. If it is bigger then something
is going wrong (it might be not true for itanium). It means more insns
(and probably performance degradation) and more compile time to process
insns (but i think for df generation more insns is only part of
slowdown). For df gcc code size is 0.5% bigger.
I'd like to use spec2006 gcc (as I remeber it is gcc 3.4) but running
spec20006 takes so long time for now. Still gcc-2.7 is good test for
big functions, switches etc. It is still a characteristic of modern gcc
versions.
