https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107591
--- Comment #9 from Andrew Macleod <amacleod at redhat dot com> ---
(In reply to Jakub Jelinek from comment #8)
> (In reply to Aldy Hernandez from comment #6)
> > FYI, the range operators have a relation field, so it should be able to tell
> > you that both operands are equal with VREL_EQ (?). You could use that to
> > optimize things.
>
> You're right, at least for irange trio.op1_op2 () in the mult fold_range is
> VREL_EQ and we could just go from that.
> For frange perhaps it isn't as perfect because of signed zeros, so in
> float
> foo (float x, float y)
> {
> if (x == y)
> return x * y;
> return 42.0f;
> }
> trio.op1_op2 () could be VREL_EQ even if one is signed and the other
> unsigned zero, and in that case -0.0 would need to be in the range too. But
> maybe that is good enough for now.
you could also test whether op1_range contains + and/or - 0, as well as
op2_range. VREL_EQ is a symbolic equality.. the ranges can still be distinct
and individually testable to see if you have a +0 and -0.. I guess you could
also test for equality of the ranges.. op1_range == op2_range
