When we classify an SLP access as VMAT_ELEMENTWISE we still consider
overrun - the reset of it is later overwritten. The following fixes
this, resolving a few RISC-V FAILs with --param vect-force-slp=1.
Bootstrap and regtest running on x86_64-unknown-linux-gnu.
* tree-vect-stmts.cc (get_group_load_store_type): For
VMAT_ELEMENTWISE there's no overrun.
---
gcc/tree-vect-stmts.cc | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/gcc/tree-vect-stmts.cc b/gcc/tree-vect-stmts.cc
index b09c016949f..66eda7d5ace 100644
--- a/gcc/tree-vect-stmts.cc
+++ b/gcc/tree-vect-stmts.cc
@@ -2093,7 +2093,6 @@ get_group_load_store_type (vec_info *vinfo, stmt_vec_info
stmt_info,
if (SLP_TREE_LANES (slp_node) == 1)
{
*memory_access_type = VMAT_ELEMENTWISE;
- overrun_p = false;
if (dump_enabled_p ())
dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
"single-element interleaving not supported
"
@@ -2110,7 +2109,8 @@ get_group_load_store_type (vec_info *vinfo, stmt_vec_info
stmt_info,
}
}
- overrun_p = loop_vinfo && gap != 0;
+ overrun_p = (loop_vinfo && gap != 0
+ && *memory_access_type != VMAT_ELEMENTWISE);
if (overrun_p && vls_type != VLS_LOAD)
{
dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
--
2.43.0
