https://gcc.gnu.org/bugzilla/show_bug.cgi?id=126313
Drea Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Summary|[16/17 Regression] Wrong |[16/17 Regression] Wrong
|code at -O2 on |code at -O2 on
|x86_64-pc-linux-gnu since |x86_64-pc-linux-gnu
|r14-868 |
Assignee|unassigned at gcc dot gnu.org |pinskia at gcc dot
gnu.org
Status|NEW |ASSIGNED
--- Comment #4 from Drea Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Jakub Jelinek from comment #3)
> #c1 started with r14-868-gb06cfb62229f17eca59fa4aabf853d7e17e2327b
> Ditto for #c0.
Yes that is what I would have expected. Since it exposes the issue from
ifcombine.
I am messing around with ifcombine right now so I can fix it.
I already have a class in phiopt which I have been thinking about expanding to
work on bb wise rather than one statement to save/restore the ranges on an ssa
name which i think can be used here too.
Won't be until mid next week though.
