summaryrefslogtreecommitdiffstats
path: root/llvm/test/CodeGen/X86/vector-shuffle-256-v4.ll
diff options
context:
space:
mode:
Diffstat (limited to 'llvm/test/CodeGen/X86/vector-shuffle-256-v4.ll')
-rw-r--r--llvm/test/CodeGen/X86/vector-shuffle-256-v4.ll152
1 files changed, 152 insertions, 0 deletions
diff --git a/llvm/test/CodeGen/X86/vector-shuffle-256-v4.ll b/llvm/test/CodeGen/X86/vector-shuffle-256-v4.ll
index 0bd1bd9d611..18b9970f070 100644
--- a/llvm/test/CodeGen/X86/vector-shuffle-256-v4.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-256-v4.ll
@@ -300,6 +300,82 @@ define <4 x double> @shuffle_v4f64_0167(<4 x double> %a, <4 x double> %b) {
ret <4 x double> %shuffle
}
+define <4 x double> @shuffle_v4f64_1054(<4 x double> %a, <4 x double> %b) {
+; AVX1-LABEL: shuffle_v4f64_1054:
+; AVX1: # BB#0:
+; AVX1-NEXT: vpermilpd {{.*#+}} xmm0 = xmm0[1,0]
+; AVX1-NEXT: vpermilpd {{.*#+}} xmm1 = xmm1[1,0]
+; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: shuffle_v4f64_1054:
+; AVX2: # BB#0:
+; AVX2-NEXT: vpermpd {{.*#+}} ymm1 = ymm1[0,1,1,0]
+; AVX2-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,0,2,2]
+; AVX2-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3]
+; AVX2-NEXT: retq
+ %shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 1, i32 0, i32 5, i32 4>
+ ret <4 x double> %shuffle
+}
+
+define <4 x double> @shuffle_v4f64_3254(<4 x double> %a, <4 x double> %b) {
+; AVX1-LABEL: shuffle_v4f64_3254:
+; AVX1: # BB#0:
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT: vpermilpd {{.*#+}} xmm0 = xmm0[1,0]
+; AVX1-NEXT: vpermilpd {{.*#+}} xmm1 = xmm1[1,0]
+; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: shuffle_v4f64_3254:
+; AVX2: # BB#0:
+; AVX2-NEXT: vpermpd {{.*#+}} ymm1 = ymm1[0,1,1,0]
+; AVX2-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[3,2,2,3]
+; AVX2-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3]
+; AVX2-NEXT: retq
+ %shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 3, i32 2, i32 5, i32 4>
+ ret <4 x double> %shuffle
+}
+
+define <4 x double> @shuffle_v4f64_3276(<4 x double> %a, <4 x double> %b) {
+; AVX1-LABEL: shuffle_v4f64_3276:
+; AVX1: # BB#0:
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT: vpermilpd {{.*#+}} xmm0 = xmm0[1,0]
+; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm1
+; AVX1-NEXT: vpermilpd {{.*#+}} xmm1 = xmm1[1,0]
+; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: shuffle_v4f64_3276:
+; AVX2: # BB#0:
+; AVX2-NEXT: vpermilpd {{.*#+}} ymm1 = ymm1[0,0,3,2]
+; AVX2-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[3,2,2,3]
+; AVX2-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3]
+; AVX2-NEXT: retq
+ %shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 3, i32 2, i32 7, i32 6>
+ ret <4 x double> %shuffle
+}
+
+define <4 x double> @shuffle_v4f64_1076(<4 x double> %a, <4 x double> %b) {
+; AVX1-LABEL: shuffle_v4f64_1076:
+; AVX1: # BB#0:
+; AVX1-NEXT: vpermilpd {{.*#+}} xmm0 = xmm0[1,0]
+; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm1
+; AVX1-NEXT: vpermilpd {{.*#+}} xmm1 = xmm1[1,0]
+; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: shuffle_v4f64_1076:
+; AVX2: # BB#0:
+; AVX2-NEXT: vpermilpd {{.*#+}} ymm1 = ymm1[0,0,3,2]
+; AVX2-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,0,2,2]
+; AVX2-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3]
+; AVX2-NEXT: retq
+ %shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 1, i32 0, i32 7, i32 6>
+ ret <4 x double> %shuffle
+}
+
define <4 x i64> @shuffle_v4i64_0000(<4 x i64> %a, <4 x i64> %b) {
; AVX1-LABEL: shuffle_v4i64_0000:
; AVX1: # BB#0:
@@ -608,6 +684,82 @@ define <4 x i64> @shuffle_v4i64_1251(<4 x i64> %a, <4 x i64> %b) {
ret <4 x i64> %shuffle
}
+define <4 x i64> @shuffle_v4i64_1054(<4 x i64> %a, <4 x i64> %b) {
+; AVX1-LABEL: shuffle_v4i64_1054:
+; AVX1: # BB#0:
+; AVX1-NEXT: vpermilpd {{.*#+}} xmm0 = xmm0[1,0]
+; AVX1-NEXT: vpermilpd {{.*#+}} xmm1 = xmm1[1,0]
+; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: shuffle_v4i64_1054:
+; AVX2: # BB#0:
+; AVX2-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,1,1,0]
+; AVX2-NEXT: vpshufd {{.*#+}} ymm0 = ymm0[2,3,0,1,6,7,4,5]
+; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
+; AVX2-NEXT: retq
+ %shuffle = shufflevector <4 x i64> %a, <4 x i64> %b, <4 x i32> <i32 1, i32 0, i32 5, i32 4>
+ ret <4 x i64> %shuffle
+}
+
+define <4 x i64> @shuffle_v4i64_3254(<4 x i64> %a, <4 x i64> %b) {
+; AVX1-LABEL: shuffle_v4i64_3254:
+; AVX1: # BB#0:
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT: vpermilpd {{.*#+}} xmm0 = xmm0[1,0]
+; AVX1-NEXT: vpermilpd {{.*#+}} xmm1 = xmm1[1,0]
+; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: shuffle_v4i64_3254:
+; AVX2: # BB#0:
+; AVX2-NEXT: vpermq {{.*#+}} ymm1 = ymm1[0,1,1,0]
+; AVX2-NEXT: vpermq {{.*#+}} ymm0 = ymm0[3,2,2,3]
+; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
+; AVX2-NEXT: retq
+ %shuffle = shufflevector <4 x i64> %a, <4 x i64> %b, <4 x i32> <i32 3, i32 2, i32 5, i32 4>
+ ret <4 x i64> %shuffle
+}
+
+define <4 x i64> @shuffle_v4i64_3276(<4 x i64> %a, <4 x i64> %b) {
+; AVX1-LABEL: shuffle_v4i64_3276:
+; AVX1: # BB#0:
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT: vpermilpd {{.*#+}} xmm0 = xmm0[1,0]
+; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm1
+; AVX1-NEXT: vpermilpd {{.*#+}} xmm1 = xmm1[1,0]
+; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: shuffle_v4i64_3276:
+; AVX2: # BB#0:
+; AVX2-NEXT: vpshufd {{.*#+}} ymm1 = ymm1[2,3,0,1,6,7,4,5]
+; AVX2-NEXT: vpermq {{.*#+}} ymm0 = ymm0[3,2,2,3]
+; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
+; AVX2-NEXT: retq
+ %shuffle = shufflevector <4 x i64> %a, <4 x i64> %b, <4 x i32> <i32 3, i32 2, i32 7, i32 6>
+ ret <4 x i64> %shuffle
+}
+
+define <4 x i64> @shuffle_v4i64_1076(<4 x i64> %a, <4 x i64> %b) {
+; AVX1-LABEL: shuffle_v4i64_1076:
+; AVX1: # BB#0:
+; AVX1-NEXT: vpermilpd {{.*#+}} xmm0 = xmm0[1,0]
+; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm1
+; AVX1-NEXT: vpermilpd {{.*#+}} xmm1 = xmm1[1,0]
+; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: shuffle_v4i64_1076:
+; AVX2: # BB#0:
+; AVX2-NEXT: vpshufd {{.*#+}} ymm1 = ymm1[2,3,0,1,6,7,4,5]
+; AVX2-NEXT: vpshufd {{.*#+}} ymm0 = ymm0[2,3,0,1,6,7,4,5]
+; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3],ymm1[4,5,6,7]
+; AVX2-NEXT: retq
+ %shuffle = shufflevector <4 x i64> %a, <4 x i64> %b, <4 x i32> <i32 1, i32 0, i32 7, i32 6>
+ ret <4 x i64> %shuffle
+}
+
define <4 x i64> @stress_test1(<4 x i64> %a, <4 x i64> %b) {
; AVX1-LABEL: stress_test1:
; AVX1: # BB#0:
OpenPOWER on IntegriCloud