[Bug tree-optimization/68956] [6 regression] Vectorizer miscompilation of 416.gamess

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=68956

--- Comment #4 from Richard Biener <rguenth at gcc dot gnu.org> ---
(In reply to Ilya Enkovich from comment #2)
> Caused by a misprint.  Here is a patch:
> 
> --- a/gcc/tree-vect-stmts.c
> +++ b/gcc/tree-vect-stmts.c
> @@ -1304,8 +1304,8 @@ vect_init_vector (gimple *stmt, tree val, tree type,
> gimple_stmt_iterator *gsi)
>              all zeros or all ones value before building a vector.  */
>           if (VECTOR_BOOLEAN_TYPE_P (type))
>             {
> -             tree true_val = build_zero_cst (TREE_TYPE (type));
> -             tree false_val = build_all_ones_cst (TREE_TYPE (type));
> +             tree true_val = build_all_ones_cst (TREE_TYPE (type));
> +             tree false_val = build_zero_cst (TREE_TYPE (type));
> 
>               if (CONSTANT_CLASS_P (val))
>                 val = integer_zerop (val) ? false_val : true_val;

Doh.  The above is ok.