C++ compile time (again, take 2)

[Diego Novillo]

On Mon, 2003-06-16 at 11:03, law@redhat.com wrote:

> Out of curiosity, what did you have in mind to speed up CCP?
> 
Marking statements with a "don't simulate again" bit when we find that
they're either not interesting for CCP (produce no value) or they've
been found to produce a VARYING value in a previous pass.  This avoids
repeated hash table lookups which were popping up fairly high in gprof
reports.

I'm still testing the patch so I don't know if there are any bugs in it
yet.


Diego.