So the problem is vector generic decided to do comparisons in signed-boolean:32
types but the rest of the middle-end was not ready for that. Since we are
building
the comparison which will feed into a cond_expr here, using boolean_type_node is
better and also correct. The rest of the compiler thinks the ranges for
comparison is always [0,1] too.
Note this code does not currently lowers bigger vector sizes into smaller
vector sizes so using boolean_type_node here is better.
OK? bootstrapped and tested on x86_64-linux-gnu with no regressions.
gcc/ChangeLog:
PR middle-end/110554
* tree-vect-generic.cc (expand_vector_condition): For comparisons,
just build using boolean_type_node instead of the cond_type.
For non-comparisons/non-scalar-bitmask, build a ` != 0` gimple
that will feed into the COND_EXPR.
---
gcc/tree-vect-generic.cc | 8 ++++++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/gcc/tree-vect-generic.cc b/gcc/tree-vect-generic.cc
index df04a0db68d..a7e6cb87a5e 100644
--- a/gcc/tree-vect-generic.cc
+++ b/gcc/tree-vect-generic.cc
@@ -1121,7 +1121,7 @@ expand_vector_condition (gimple_stmt_iterator *gsi,
bitmap dce_ssa_names)
comp_width, comp_index);
tree aa2 = tree_vec_extract (gsi, comp_inner_type, a2,
comp_width, comp_index);
- aa = gimplify_build2 (gsi, code, cond_type, aa1, aa2);
+ aa = gimplify_build2 (gsi, code, boolean_type_node, aa1, aa2);
}
else if (a_is_scalar_bitmask)
{
@@ -1132,7 +1132,11 @@ expand_vector_condition (gimple_stmt_iterator *gsi,
bitmap dce_ssa_names)
build_zero_cst (TREE_TYPE (a)));
}
else
- aa = tree_vec_extract (gsi, cond_type, a, comp_width, comp_index);
+ {
+ result = tree_vec_extract (gsi, cond_type, a, comp_width, comp_index);
+ aa = gimplify_build2 (gsi, NE_EXPR, boolean_type_node, result,
+ build_zero_cst (cond_type));
+ }
result = gimplify_build3 (gsi, COND_EXPR, inner_type, aa, bb, cc);
if (!CONSTANT_CLASS_P (result))
constant_p = false;
--
2.31.1
