It just clutters the dump files and takes up compile-time.
Bootstrapped and tested on x86_64-unknown-linux-gnu, pushed.
* tree-vect-slp.cc (vect_build_slp_tree_2): Disable SLP
reassociation for single-lane.
---
gcc/tree-vect-slp.cc | 2 ++
1 file changed, 2 insertions(+)
diff --git a/gcc/tree-vect-slp.cc b/gcc/tree-vect-slp.cc
index 2aedba61a4e..5a65a99d61e 100644
--- a/gcc/tree-vect-slp.cc
+++ b/gcc/tree-vect-slp.cc
@@ -2099,6 +2099,8 @@ vect_build_slp_tree_2 (vec_info *vinfo, slp_tree node,
for matching as we can succeed by means of builds from scalars
and have no good way to "cost" one build against another. */
else if (is_a <loop_vec_info> (vinfo)
+ /* Do not bother for single-lane SLP. */
+ && group_size > 1
/* ??? We don't handle !vect_internal_def defs below. */
&& STMT_VINFO_DEF_TYPE (stmt_info) == vect_internal_def
/* ??? Do not associate a reduction, this will wreck REDUC_IDX
--
2.43.0
