https://gcc.gnu.org/bugzilla/show_bug.cgi?id=86122
Jakub Jelinek <jakub at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |jakub at gcc dot gnu.org
--- Comment #2 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Started with r249447. I'd go for:
--- match.pd.jj1 2018-05-25 14:34:42.268381974 +0200
+++ match.pd 2018-06-14 15:53:20.538712014 +0200
@@ -1771,10 +1771,11 @@ DEFINE_INT_AND_FLOAT_ROUND_FN (RINT)
(neg_inner_op @0 { wide_int_to_tree (type, wi::to_wide (cst)); })
/* Last resort, use some unsigned type. */
(with { tree utype = unsigned_type_for (type); }
- (view_convert (inner_op
- (view_convert:utype @0)
- (view_convert:utype
- { drop_tree_overflow (cst); })))))))))))))
+ (if (utype)
+ (view_convert (inner_op
+ (view_convert:utype @0)
+ (view_convert:utype
+ { drop_tree_overflow (cst); }))))))))))))))
/* (CST1 - A) +- CST2 -> CST3 - A */
(for outer_op (plus minus)
so that it is easily backportable and then we can discuss if we want
unsigned_type_for to support complex integer types or not.
