https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64503
Martin Liška <marxin at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |marxin at gcc dot gnu.org
--- Comment #3 from Martin Liška <marxin at gcc dot gnu.org> ---
Hello.
I've been testing fix where I replaced exp2 with scalbln, as Jakub suggested.
On the other hand, -inf looks a bit suspicious and is created here:
ipa-inline.c:946, where
if (growth <= 0) {
badness = (sreal) (-SREAL_MIN_SIG + growth) << (SREAL_MAX_EXP / 256);
as one can see, it's (-1073741824+growth) << (536870911/256) which is
practically always -inf?
I am curious if it's desired calculation Honza?
Thanks,
Martin