diff mbox series

testsuite/115304 - properly guard gcc.dg/vect/slp-gap-1.c

Message ID 20240603124540.4F91F3832E66@sourceware.org
State New
Headers show
Series testsuite/115304 - properly guard gcc.dg/vect/slp-gap-1.c | expand

Commit Message

Richard Biener June 3, 2024, 12:45 p.m. UTC 
Testing on sparc shows we need vect_unpack and vect_perm.  This
isn't enough to resolve the GCN fail which ends up using interleaving.

Tested on x86_64-linux and sparc-solaris.

	PR testsuite/115304
	* gcc.dg/vect/slp-gap-1.c: Require vect_unpack and vect_perm.
---
 gcc/testsuite/gcc.dg/vect/slp-gap-1.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
diff mbox series

Patch

diff --git a/gcc/testsuite/gcc.dg/vect/slp-gap-1.c b/gcc/testsuite/gcc.dg/vect/slp-gap-1.c
index 36463ca22c5..9856da7a7f4 100644
--- a/gcc/testsuite/gcc.dg/vect/slp-gap-1.c
+++ b/gcc/testsuite/gcc.dg/vect/slp-gap-1.c
@@ -15,4 +15,4 @@  void pixel_sub_wxh(int16_t * __restrict diff, uint8_t *pix1, uint8_t *pix2) {
 /* We can vectorize this without peeling for gaps and thus without epilogue,
    but the only thing we can reliably scan is the zero-padding trick for the
    partial loads.  */
-/* { dg-final { scan-tree-dump-times "\{_\[0-9\]\+, 0" 6 "vect" { target vect64 } } } */
+/* { dg-final { scan-tree-dump-times "\{_\[0-9\]\+, 0" 6 "vect" { target { vect64 && { vect_unpack && vect_perm } } } } } */