While vectorizable_store was already checking alignment requirement
of the stores and fall back to elementwise accesses if not honored
the vectorizable_load path wasn't doing this. After the previous
change to disregard alignment checking for VMAT_STRIDED_SLP in
get_group_load_store_type this now tripped on power.
Bootstrapped and tested on x86_64-unknown-linux-gnu, pushed.
PR tree-optimization/117720
* tree-vect-stmts.cc (vectorizable_load): For VMAT_STRIDED_SLP
verify the choosen load type is OK with regard to alignment.
---
gcc/tree-vect-stmts.cc | 16 +++++++++++++---
1 file changed, 13 insertions(+), 3 deletions(-)
diff --git a/gcc/tree-vect-stmts.cc b/gcc/tree-vect-stmts.cc
index 8700d1787b4..271c6da2a25 100644
--- a/gcc/tree-vect-stmts.cc
+++ b/gcc/tree-vect-stmts.cc
@@ -10650,9 +10650,19 @@ vectorizable_load (vec_info *vinfo,
of it. */
if (n == const_nunits)
{
- nloads = 1;
- lnel = const_nunits;
- ltype = vectype;
+ int mis_align = dr_misalignment (first_dr_info, vectype);
+ dr_alignment_support dr_align
+ = vect_supportable_dr_alignment (vinfo, dr_info, vectype,
+ mis_align);
+ if (dr_align == dr_aligned
+ || dr_align == dr_unaligned_supported)
+ {
+ nloads = 1;
+ lnel = const_nunits;
+ ltype = vectype;
+ alignment_support_scheme = dr_align;
+ misalignment = mis_align;
+ }
}
/* Else use the biggest vector we can load the group without
accessing excess elements. */
--
2.43.0
