Remove vperm2f* and vperm2i builtins. Same effect can be achieved with builtin_shufflevector.
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@150064 91177308-0d34-0410-b5e6-96231b3b80d8
diff --git a/test/CodeGen/avx-shuffle-builtins.c b/test/CodeGen/avx-shuffle-builtins.c
index 6b2b1b1..18300bf 100644
--- a/test/CodeGen/avx-shuffle-builtins.c
+++ b/test/CodeGen/avx-shuffle-builtins.c
@@ -38,3 +38,21 @@
// CHECK: shufflevector{{.*}}<i32 3, i32 2, i32 1, i32 0, i32 7, i32 6, i32 5, i32 4>
return _mm256_permute_ps(a, 0x1b);
}
+
+__m256d test_mm256_permute2f128_pd(__m256d a, __m256d b) {
+ // Check if the mask is correct
+ // CHECK: shufflevector{{.*}}<i32 2, i32 3, i32 6, i32 7>
+ return _mm256_permute2f128_pd(a, b, 0x31);
+}
+
+__m256 test_mm256_permute2f128_ps(__m256 a, __m256 b) {
+ // Check if the mask is correct
+ // CHECK: shufflevector{{.*}}<i32 12, i32 13, i32 14, i32 15, i32 4, i32 5, i32 6, i32 7>
+ return _mm256_permute2f128_ps(a, b, 0x13);
+}
+
+__m256i test_mm256_permute2f128_si256(__m256i a, __m256i b) {
+ // Check if the mask is correct
+ // CHECK: shufflevector{{.*}}<i32 0, i32 1, i32 2, i32 3, i32 8, i32 9, i32 10, i32 11>
+ return _mm256_permute2f128_si256(a, b, 0x20);
+}
diff --git a/test/CodeGen/avx2-builtins.c b/test/CodeGen/avx2-builtins.c
index ba0338e..a5ddc8e 100644
--- a/test/CodeGen/avx2-builtins.c
+++ b/test/CodeGen/avx2-builtins.c
@@ -677,8 +677,8 @@
}
__m256i test_mm256_permute2x128_si256(__m256i a, __m256i b) {
- // CHECK: @llvm.x86.avx2.vperm2i128
- return _mm256_permute2x128_si256(a, b, 94);
+ // CHECK: shufflevector{{.*}}<i32 2, i32 3, i32 6, i32 7>
+ return _mm256_permute2x128_si256(a, b, 0x31);
}
__m128i test_mm256_extracti128_si256(__m256i a) {
diff --git a/test/CodeGen/builtins-x86.c b/test/CodeGen/builtins-x86.c
index acb5554..30138d6 100644
--- a/test/CodeGen/builtins-x86.c
+++ b/test/CodeGen/builtins-x86.c
@@ -414,9 +414,6 @@
tmp_V4i = __builtin_ia32_cvttpd2dq256(tmp_V4d);
tmp_V4i = __builtin_ia32_cvtpd2dq256(tmp_V4d);
tmp_V8i = __builtin_ia32_cvttps2dq256(tmp_V8f);
- tmp_V4d = __builtin_ia32_vperm2f128_pd256(tmp_V4d, tmp_V4d, 0x7);
- tmp_V8f = __builtin_ia32_vperm2f128_ps256(tmp_V8f, tmp_V8f, 0x7);
- tmp_V8i = __builtin_ia32_vperm2f128_si256(tmp_V8i, tmp_V8i, 0x7);
tmp_V4d = __builtin_ia32_vinsertf128_pd256(tmp_V4d, tmp_V2d, 0x7);
tmp_V8f = __builtin_ia32_vinsertf128_ps256(tmp_V8f, tmp_V4f, 0x7);
tmp_V8i = __builtin_ia32_vinsertf128_si256(tmp_V8i, tmp_V4i, 0x7);