In the Z backend we still set min-vect-loop-bound to 2 to work around
corner cases where awkward epilogue code gets generated in the
vectorizer. This has a particular bad impact when vectorizing loops
with a low iteration count. Due to this we do not vectorize the loop
in gen-vect-11/32 - what actually is a pity.
The patch sets min-vect-loop-bound back to the default value of 0 in
order to enable vectorization.
2019-11-05 Andreas Krebbel <kreb...@linux.ibm.com>
* gcc.dg/tree-ssa/gen-vect-11.c: Add --param min-vect-loop-bound=0
for IBM Z.
* gcc.dg/tree-ssa/gen-vect-11.c: Likewise.
---
gcc/testsuite/gcc.dg/tree-ssa/gen-vect-11.c | 6 +++++-
gcc/testsuite/gcc.dg/tree-ssa/gen-vect-32.c | 4 ++++
2 files changed, 9 insertions(+), 1 deletion(-)
diff --git a/gcc/testsuite/gcc.dg/tree-ssa/gen-vect-11.c
b/gcc/testsuite/gcc.dg/tree-ssa/gen-vect-11.c
index 650e73a5ee8..dd1c0ac3eba 100644
--- a/gcc/testsuite/gcc.dg/tree-ssa/gen-vect-11.c
+++ b/gcc/testsuite/gcc.dg/tree-ssa/gen-vect-11.c
@@ -1,6 +1,10 @@
/* { dg-do run { target vect_cmdline_needed } } */
/* { dg-options "-O2 -ftree-vectorize -fwrapv -fdump-tree-vect-details
-fvect-cost-model=dynamic" } */
-/* { dg-options "-O2 -ftree-vectorize -fwrapv -fdump-tree-vect-details
-fvect-cost-model=dynamic -mno-sse" { target { i?86-*-* x86_64-*-* } } } */
+/* { dg-additional-options "-mno-sse" { target { i?86-*-* x86_64-*-* } } } */
+/* The IBM Z backend sets the min-vect-loop-bound param to 2 to avoid
+ awkward epilogue code generation in some cases. This line needs to
+ be removed after finding an alternate way to fix this. */
+/* { dg-additional-options "--param min-vect-loop-bound=0" { target {
s390*-*-* } } } */
#include <stdlib.h>
diff --git a/gcc/testsuite/gcc.dg/tree-ssa/gen-vect-32.c
b/gcc/testsuite/gcc.dg/tree-ssa/gen-vect-32.c
index c4bee19b75a..378dd0b831c 100644
--- a/gcc/testsuite/gcc.dg/tree-ssa/gen-vect-32.c
+++ b/gcc/testsuite/gcc.dg/tree-ssa/gen-vect-32.c
@@ -1,6 +1,10 @@
/* { dg-do run { target vect_cmdline_needed } } */
/* { dg-options "-O2 -fno-tree-loop-distribute-patterns -ftree-vectorize
-fdump-tree-vect-details -fno-vect-cost-model" } */
/* { dg-additional-options "-mno-sse" { target { i?86-*-* x86_64-*-* } } } */
+/* The IBM Z backend sets the min-vect-loop-bound param to 2 to avoid
+ awkward epilogue code generation in some cases. This line needs to
+ be removed after finding an alternate way to fix this. */
+/* { dg-additional-options "--param min-vect-loop-bound=0" { target {
s390*-*-* } } } */
#include <stdlib.h>
--
2.23.0
