[PATCH] tree-optimization/97043 - fix latent wrong-code with SLP vectorization

Richard Biener rguenther@suse.de
Mon Sep 14 09:30:05 GMT 2020


When the unrolling decision comes late and would have prevented
eliding a SLP load permutation we can end up generating aligned
loads when the load is in fact unaligned.  Most of the time
alignment analysis figures out the load is in fact unaligned
but that cannot be relied upon.

The following removes the SLP load permutation eliding based on
the still premature vectorization factor.

This is only necessary on branches since on trunk this optimization
is delayed upon a point where the vectorization factor is final.

Bootstrap & regtest running on x86_64-unknown-linux-gnu.

2020-09-14  Richard Biener  <rguenther@suse.de>

	PR tree-optimization/97043
	* tree-vect-slp.c (vect_analyze_slp_instance): Do not
	elide a load permutation if the current vectorization
	factor is one.
---
 gcc/tree-vect-slp.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/gcc/tree-vect-slp.c b/gcc/tree-vect-slp.c
index f6331eeea86..3fdf56f9335 100644
--- a/gcc/tree-vect-slp.c
+++ b/gcc/tree-vect-slp.c
@@ -2309,9 +2309,8 @@ vect_analyze_slp_instance (vec_info *vinfo,
 		  /* The load requires permutation when unrolling exposes
 		     a gap either because the group is larger than the SLP
 		     group-size or because there is a gap between the groups.  */
-		  && (known_eq (unrolling_factor, 1U)
-		      || (group_size == DR_GROUP_SIZE (first_stmt_info)
-			  && DR_GROUP_GAP (first_stmt_info) == 0)))
+		  && group_size == DR_GROUP_SIZE (first_stmt_info)
+		  && DR_GROUP_GAP (first_stmt_info) == 0)
 		{
 		  SLP_TREE_LOAD_PERMUTATION (load_node).release ();
 		  continue;
-- 
2.26.2


More information about the Gcc-patches mailing list