https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121478
Andrew Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Known to fail| |13.1.0, 14.3.0, 15.1.0
Status|UNCONFIRMED |NEW
Last reconfirmed| |2025-08-09
Summary|[ICE] internal compiler |ICE in fold_convert_loc due
|error: in fold_convert_loc, |to null pointer type
|at fold-const.cc:2670 |
Ever confirmed|0 |1
--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Confirmed.
The front-end exposes nullptr_type to the middle-end for comparisons.
if (nullp_27(D) == 0)
this should always optimized to just true.
But I am not 100% sure if this is a front-end issue or a gimple one.
`-O2 -funswitch-loops` is enough to reproduce it. I can't figure out what goes
wrong with unswitching the loop dealing with that if statement too.
