diff options
author | Jakub Jelinek <jakub@redhat.com> | 2023-04-22 20:16:08 +0200 |
---|---|---|
committer | Jakub Jelinek <jakub@redhat.com> | 2023-04-22 20:16:08 +0200 |
commit | 7a0cbaf7f802df209840d78740ffc749dadd1ce3 (patch) | |
tree | 51ac009f81e8bf7912d73aa3720b4f6abb7696ec | |
parent | 195270d28a534cc1f08478c6e0136f4fc13d247a (diff) | |
download | gcc-7a0cbaf7f802df209840d78740ffc749dadd1ce3.zip gcc-7a0cbaf7f802df209840d78740ffc749dadd1ce3.tar.gz gcc-7a0cbaf7f802df209840d78740ffc749dadd1ce3.tar.bz2 |
testsuite: Fix up pr109011-*.c tests for powerpc [PR109572]
As reported, pr109011-{4,5}.c tests fail on powerpc.
I thought they should have the same counts as the corresponding -{2,3}.c
tests, the only difference is that -{2,3}.c are int while -{4,5}.c are
long long. But there are 2 issues. One is that in the foo
function the vectorization costs comparison triggered in, while in -{2,3}.c
we use vectorization factor 4 and it was found beneficial, when using
long long it was just vf 2 and the scalar cost of doing
p[i] = __builtin_ctzll (q[i]) twice looked smaller than the vectorizated
statements. I could disable the cost model, but instead chose to add
some further arithmetics to those functions to make it beneficial even
with vf 2.
After that change, pr109011-4.c still failed; I was expecting 4 .CTZ calls
there on power9, 3 vectorized and one in scalar code, but for some reason
the scalar one didn't trigger. As I really want to count just the
vectorized calls, I've added the vect prefix on the variables to ensure
I'm only counting vectorized calls and decreased the 4 counts to 3.
2023-04-22 Jakub Jelinek <jakub@redhat.com>
PR testsuite/109572
* gcc.dg/vect/pr109011-1.c: In scan-tree-dump-times regexps match also
vect prefix to make sure we only count vectorized calls.
* gcc.dg/vect/pr109011-2.c: Likewise. On powerpc* expect just count 3
rather than 4.
* gcc.dg/vect/pr109011-3.c: In scan-tree-dump-times regexps match also
vect prefix to make sure we only count vectorized calls.
* gcc.dg/vect/pr109011-4.c: Likewise. On powerpc* expect just count 3
rather than 4.
(foo): Add 2 further arithmetic ops to the loop to make it appear
worthwhile for vectorization heuristics on powerpc.
* gcc.dg/vect/pr109011-5.c: In scan-tree-dump-times regexps match also
vect prefix to make sure we only count vectorized calls.
(foo): Add 2 further arithmetic ops to the loop to make it appear
worthwhile for vectorization heuristics on powerpc.
-rw-r--r-- | gcc/testsuite/gcc.dg/vect/pr109011-1.c | 8 | ||||
-rw-r--r-- | gcc/testsuite/gcc.dg/vect/pr109011-2.c | 8 | ||||
-rw-r--r-- | gcc/testsuite/gcc.dg/vect/pr109011-3.c | 4 | ||||
-rw-r--r-- | gcc/testsuite/gcc.dg/vect/pr109011-4.c | 10 | ||||
-rw-r--r-- | gcc/testsuite/gcc.dg/vect/pr109011-5.c | 6 |
5 files changed, 18 insertions, 18 deletions
diff --git a/gcc/testsuite/gcc.dg/vect/pr109011-1.c b/gcc/testsuite/gcc.dg/vect/pr109011-1.c index 16a5826..9bb8ee1 100644 --- a/gcc/testsuite/gcc.dg/vect/pr109011-1.c +++ b/gcc/testsuite/gcc.dg/vect/pr109011-1.c @@ -14,8 +14,8 @@ foo (long long *p, long long *q) p[i] = __builtin_popcountll (q[i]); } -/* { dg-final { scan-tree-dump-times " = \.POPCOUNT \\\(" 1 "optimized" { target { { i?86-*-* x86_64-*-* } && avx512vpopcntdq } } } } */ -/* { dg-final { scan-tree-dump-times " = \.POPCOUNT \\\(" 1 "optimized" { target { powerpc_p8vector_ok || s390_vx } } } } */ +/* { dg-final { scan-tree-dump-times " = \.POPCOUNT \\\(vect" 1 "optimized" { target { { i?86-*-* x86_64-*-* } && avx512vpopcntdq } } } } */ +/* { dg-final { scan-tree-dump-times " = \.POPCOUNT \\\(vect" 1 "optimized" { target { powerpc_p8vector_ok || s390_vx } } } } */ void bar (long long *p, long long *q) @@ -25,5 +25,5 @@ bar (long long *p, long long *q) p[i] = __builtin_clzll (q[i]); } -/* { dg-final { scan-tree-dump-times " = \.CLZ \\\(" 1 "optimized" { target { { i?86-*-* x86_64-*-* } && avx512cd } } } } */ -/* { dg-final { scan-tree-dump-times " = \.CLZ \\\(" 1 "optimized" { target { powerpc_p8vector_ok || s390_vx } } } } */ +/* { dg-final { scan-tree-dump-times " = \.CLZ \\\(vect" 1 "optimized" { target { { i?86-*-* x86_64-*-* } && avx512cd } } } } */ +/* { dg-final { scan-tree-dump-times " = \.CLZ \\\(vect" 1 "optimized" { target { powerpc_p8vector_ok || s390_vx } } } } */ diff --git a/gcc/testsuite/gcc.dg/vect/pr109011-2.c b/gcc/testsuite/gcc.dg/vect/pr109011-2.c index 191af89..6d18eb2 100644 --- a/gcc/testsuite/gcc.dg/vect/pr109011-2.c +++ b/gcc/testsuite/gcc.dg/vect/pr109011-2.c @@ -29,7 +29,7 @@ baz (int *p, int *q) p[i] = __builtin_ffs (q[i]); } -/* { dg-final { scan-tree-dump-times " = \.CLZ \\\(" 3 "optimized" { target { { { { i?86-*-* x86_64-*-* } && avx512cd } && lzcnt } && bmi } } } } */ -/* { dg-final { scan-tree-dump-times " = \.CTZ \\\(" 4 "optimized" { target powerpc_p9vector_ok } } } */ -/* { dg-final { scan-tree-dump-times " = \.CTZ \\\(" 2 "optimized" { target s390_vx } } } */ -/* { dg-final { scan-tree-dump-times " = \.POPCOUNT \\\(" 1 "optimized" { target s390_vx } } } */ +/* { dg-final { scan-tree-dump-times " = \.CLZ \\\(vect" 3 "optimized" { target { { { { i?86-*-* x86_64-*-* } && avx512cd } && lzcnt } && bmi } } } } */ +/* { dg-final { scan-tree-dump-times " = \.CTZ \\\(vect" 3 "optimized" { target powerpc_p9vector_ok } } } */ +/* { dg-final { scan-tree-dump-times " = \.CTZ \\\(vect" 2 "optimized" { target s390_vx } } } */ +/* { dg-final { scan-tree-dump-times " = \.POPCOUNT \\\(vect" 1 "optimized" { target s390_vx } } } */ diff --git a/gcc/testsuite/gcc.dg/vect/pr109011-3.c b/gcc/testsuite/gcc.dg/vect/pr109011-3.c index 2e631fc..0c6c826 100644 --- a/gcc/testsuite/gcc.dg/vect/pr109011-3.c +++ b/gcc/testsuite/gcc.dg/vect/pr109011-3.c @@ -28,5 +28,5 @@ baz (int *p, int *q) p[i] = __builtin_ffs (q[i]); } -/* { dg-final { scan-tree-dump-times " = \.POPCOUNT \\\(" 3 "optimized" { target { { { { i?86-*-* x86_64-*-* } && avx512vpopcntdq } && lzcnt } && bmi } } } } */ -/* { dg-final { scan-tree-dump-times " = \.CLZ \\\(" 3 "optimized" { target powerpc_p8vector_ok } } } */ +/* { dg-final { scan-tree-dump-times " = \.POPCOUNT \\\(vect" 3 "optimized" { target { { { { i?86-*-* x86_64-*-* } && avx512vpopcntdq } && lzcnt } && bmi } } } } */ +/* { dg-final { scan-tree-dump-times " = \.CLZ \\\(vect" 3 "optimized" { target powerpc_p8vector_ok } } } */ diff --git a/gcc/testsuite/gcc.dg/vect/pr109011-4.c b/gcc/testsuite/gcc.dg/vect/pr109011-4.c index ce1ee02..7bb3377 100644 --- a/gcc/testsuite/gcc.dg/vect/pr109011-4.c +++ b/gcc/testsuite/gcc.dg/vect/pr109011-4.c @@ -10,7 +10,7 @@ foo (long long *p, long long *q) { #pragma omp simd for (int i = 0; i < 2048; ++i) - p[i] = __builtin_ctzll (q[i]); + p[i] = 2 * q[i] + __builtin_ctzll (q[i]); } void @@ -29,7 +29,7 @@ baz (long long *p, long long *q) p[i] = __builtin_ffsll (q[i]); } -/* { dg-final { scan-tree-dump-times " = \.CLZ \\\(" 3 "optimized" { target { { { { i?86-*-* x86_64-*-* } && avx512cd } && lzcnt } && bmi } } } } */ -/* { dg-final { scan-tree-dump-times " = \.CTZ \\\(" 4 "optimized" { target powerpc_p9vector_ok } } } */ -/* { dg-final { scan-tree-dump-times " = \.CTZ \\\(" 2 "optimized" { target s390_vx } } } */ -/* { dg-final { scan-tree-dump-times " = \.POPCOUNT \\\(" 1 "optimized" { target s390_vx } } } */ +/* { dg-final { scan-tree-dump-times " = \.CLZ \\\(vect" 3 "optimized" { target { { { { i?86-*-* x86_64-*-* } && avx512cd } && lzcnt } && bmi } } } } */ +/* { dg-final { scan-tree-dump-times " = \.CTZ \\\(vect" 3 "optimized" { target powerpc_p9vector_ok } } } */ +/* { dg-final { scan-tree-dump-times " = \.CTZ \\\(vect" 2 "optimized" { target s390_vx } } } */ +/* { dg-final { scan-tree-dump-times " = \.POPCOUNT \\\(vect" 1 "optimized" { target s390_vx } } } */ diff --git a/gcc/testsuite/gcc.dg/vect/pr109011-5.c b/gcc/testsuite/gcc.dg/vect/pr109011-5.c index 51168ef..61fac0b 100644 --- a/gcc/testsuite/gcc.dg/vect/pr109011-5.c +++ b/gcc/testsuite/gcc.dg/vect/pr109011-5.c @@ -9,7 +9,7 @@ foo (long long *p, long long *q) { #pragma omp simd for (int i = 0; i < 2048; ++i) - p[i] = __builtin_ctzll (q[i]); + p[i] = 2 * q[i] + __builtin_ctzll (q[i]); } void @@ -28,5 +28,5 @@ baz (long long *p, long long *q) p[i] = __builtin_ffsll (q[i]); } -/* { dg-final { scan-tree-dump-times " = \.POPCOUNT \\\(" 3 "optimized" { target { { { { i?86-*-* x86_64-*-* } && avx512vpopcntdq } && lzcnt } && bmi } } } } */ -/* { dg-final { scan-tree-dump-times " = \.CLZ \\\(" 3 "optimized" { target powerpc_p8vector_ok } } } */ +/* { dg-final { scan-tree-dump-times " = \.POPCOUNT \\\(vect" 3 "optimized" { target { { { { i?86-*-* x86_64-*-* } && avx512vpopcntdq } && lzcnt } && bmi } } } } */ +/* { dg-final { scan-tree-dump-times " = \.CLZ \\\(vect" 3 "optimized" { target powerpc_p8vector_ok } } } */ |