https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109695
--- Comment #38 from Aldy Hernandez <aldyh at gcc dot gnu.org> ---
(In reply to Andrew Macleod from comment #37)
> (In reply to CVS Commits from comment #36)
>
> > For the curious, a particular hot spot for IPA in this area was:
> >
> > ipcp_vr_lattice::meet_with_1 (const value_range *other_vr)
> > {
> > ...
> > ...
> > value_range save (m_vr);
> > m_vr.union_ (*other_vr);
> > return m_vr != save;
> > }
> >
> > The problem isn't the resizing (since we do that at most once) but the
> > fact that for some functions with lots of callers we end up a huge
> > range that gets copied and compared for every meet operation. Maybe
> > the IPA algorithm could be adjusted somehow??.
>
> isn't the the same thing as
>
> return m_vr.union_ (*other_vr);
>
> which should be much faster without all the copying... ?
Indeed. That is in the committed version of the patch. I just forgot to
update the message.
The reason I didn't originally do as you suggested was that there was a bug in
the union code that returned TRUE for a union that hadn't changed anything.
That has also been fixed in trunk.
