RE: optimization question

[yuasa@katura.tutics.tut.ac.jp (Taiichi Yuasa)]

>> Can anybody explain why above is the case. Is this just a
>> freak accident, or does it mean that the akcl compiler is so
>> smart, or does it mean that the akcl compiler is not so
>> smart, or ...

To answer your question, I need more information about your test program.
If the numbers in the "gc before/after" rows show the gc counter (i.e.,
the number of gc invocation so far), then gc must be the reason.
If this is the case, my guess is that your macro definitions may have done
something different from the original function definitions explicitly or
implicitly.

-- Taiichi