From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by sourceware.org (Postfix, from userid 2153) id DB84F3858C83; Sat, 22 Apr 2023 18:16:40 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org DB84F3858C83 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org; s=default; t=1682187400; bh=8GN7suPLST/thg6Q5QH/4k4J8K9seT3sqW4dI3G2/0k=; h=From:To:Subject:Date:From; b=AObTw2uthI86bhunNaf+QTsNT/Ks1THetSiUjl7pCsWwcsH2kSxvyaeXcksEsyZyL YHynqgH9GPKNhsL4Dgm4pcNsxkAuRUPV6xkqIGgvBoCV6s3dDDbQ5U71NEwG8WdT2q IvRqvXbvymLDHYsw+ERvdoRbaR43sI9MaO8iKAY4= MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="utf-8" From: Jakub Jelinek To: gcc-cvs@gcc.gnu.org Subject: [gcc r14-166] testsuite: Fix up pr109011-*.c tests for powerpc [PR109572] X-Act-Checkin: gcc X-Git-Author: Jakub Jelinek X-Git-Refname: refs/heads/master X-Git-Oldrev: 195270d28a534cc1f08478c6e0136f4fc13d247a X-Git-Newrev: 7a0cbaf7f802df209840d78740ffc749dadd1ce3 Message-Id: <20230422181640.DB84F3858C83@sourceware.org> Date: Sat, 22 Apr 2023 18:16:40 +0000 (GMT) List-Id: https://gcc.gnu.org/g:7a0cbaf7f802df209840d78740ffc749dadd1ce3 commit r14-166-g7a0cbaf7f802df209840d78740ffc749dadd1ce3 Author: Jakub Jelinek Date: Sat Apr 22 20:16:08 2023 +0200 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 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. Diff: --- gcc/testsuite/gcc.dg/vect/pr109011-1.c | 8 ++++---- gcc/testsuite/gcc.dg/vect/pr109011-2.c | 8 ++++---- gcc/testsuite/gcc.dg/vect/pr109011-3.c | 4 ++-- gcc/testsuite/gcc.dg/vect/pr109011-4.c | 10 +++++----- 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 16a58267dc8..9bb8ee10a37 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 191af8945e5..6d18eb28bcf 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 2e631fc1506..0c6c8261790 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 ce1ee02516e..7bb337794b9 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 51168ef8708..61fac0bb549 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 } } } */