[X86][SSE] Add more tests showing failure to lower shuffles as bit rotations

diff --git a/llvm/test/CodeGen/X86/vector-shuffle-128-v16.ll b/llvm/test/CodeGen/X86/vector-shuffle-128-v16.ll
index c813409..2049c1f 100644 (file)
--- a/llvm/test/CodeGen/X86/vector-shuffle-128-v16.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-128-v16.ll
@@ -1773,32 +1773,64 @@ define <16 x i8> @shuffle_v16i8_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_01_02_03_04_05_06(
   ret <16 x i8> %shuffle
 }
 
+define <16 x i8> @shuffle_v16i8_03_00_01_02_07_04_05_06_11_08_09_10_15_12_13_14(<16 x i8> %a) {
+; SSE2-LABEL: shuffle_v16i8_03_00_01_02_07_04_05_06_11_08_09_10_15_12_13_14:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    pxor %xmm1, %xmm1
+; SSE2-NEXT:    movdqa %xmm0, %xmm2
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm2 = xmm2[8],xmm1[8],xmm2[9],xmm1[9],xmm2[10],xmm1[10],xmm2[11],xmm1[11],xmm2[12],xmm1[12],xmm2[13],xmm1[13],xmm2[14],xmm1[14],xmm2[15],xmm1[15]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm2 = xmm2[3,0,1,2,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,7,4,5,6]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[3,0,1,2,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,4,5,6]
+; SSE2-NEXT:    packuswb %xmm2, %xmm0
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: shuffle_v16i8_03_00_01_02_07_04_05_06_11_08_09_10_15_12_13_14:
+; SSSE3:       # %bb.0:
+; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[3,0,1,2,7,4,5,6,11,8,9,10,15,12,13,14]
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: shuffle_v16i8_03_00_01_02_07_04_05_06_11_08_09_10_15_12_13_14:
+; SSE41:       # %bb.0:
+; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[3,0,1,2,7,4,5,6,11,8,9,10,15,12,13,14]
+; SSE41-NEXT:    retq
+;
+; AVX-LABEL: shuffle_v16i8_03_00_01_02_07_04_05_06_11_08_09_10_15_12_13_14:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[3,0,1,2,7,4,5,6,11,8,9,10,15,12,13,14]
+; AVX-NEXT:    retq
+  %shuffle = shufflevector <16 x i8> %a, <16 x i8> undef, <16 x i32> <i32 3, i32 0, i32 1, i32 2, i32 7, i32 4, i32 5, i32 6, i32 11, i32 8, i32 9, i32 10, i32 15, i32 12, i32 13, i32 14>
+  ret <16 x i8> %shuffle
+}
+
 ; PR44379
-define <16 x i8> @shuffle_v16i8_02_03_04_05_06_07_00_01_10_11_12_013_14_15_08_09(<16 x i8> %a) {
-; SSE-LABEL: shuffle_v16i8_02_03_04_05_06_07_00_01_10_11_12_013_14_15_08_09:
+define <16 x i8> @shuffle_v16i8_02_03_04_05_06_07_00_01_10_11_12_13_14_15_08_09(<16 x i8> %a) {
+; SSE-LABEL: shuffle_v16i8_02_03_04_05_06_07_00_01_10_11_12_13_14_15_08_09:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,2,3,0,4,5,6,7]
 ; SSE-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,6,7,4]
 ; SSE-NEXT:    retq
 ;
-; AVX1-LABEL: shuffle_v16i8_02_03_04_05_06_07_00_01_10_11_12_013_14_15_08_09:
+; AVX1-LABEL: shuffle_v16i8_02_03_04_05_06_07_00_01_10_11_12_13_14_15_08_09:
 ; AVX1:       # %bb.0:
 ; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[1,2,3,0,4,5,6,7]
 ; AVX1-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,6,7,4]
 ; AVX1-NEXT:    retq
 ;
-; AVX2-SLOW-LABEL: shuffle_v16i8_02_03_04_05_06_07_00_01_10_11_12_013_14_15_08_09:
+; AVX2-SLOW-LABEL: shuffle_v16i8_02_03_04_05_06_07_00_01_10_11_12_13_14_15_08_09:
 ; AVX2-SLOW:       # %bb.0:
 ; AVX2-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[1,2,3,0,4,5,6,7]
 ; AVX2-SLOW-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,6,7,4]
 ; AVX2-SLOW-NEXT:    retq
 ;
-; AVX2-FAST-LABEL: shuffle_v16i8_02_03_04_05_06_07_00_01_10_11_12_013_14_15_08_09:
+; AVX2-FAST-LABEL: shuffle_v16i8_02_03_04_05_06_07_00_01_10_11_12_13_14_15_08_09:
 ; AVX2-FAST:       # %bb.0:
 ; AVX2-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[2,3,4,5,6,7,0,1,10,11,12,13,14,15,8,9]
 ; AVX2-FAST-NEXT:    retq
 ;
-; AVX512VL-LABEL: shuffle_v16i8_02_03_04_05_06_07_00_01_10_11_12_013_14_15_08_09:
+; AVX512VL-LABEL: shuffle_v16i8_02_03_04_05_06_07_00_01_10_11_12_13_14_15_08_09:
 ; AVX512VL:       # %bb.0:
 ; AVX512VL-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[2,3,4,5,6,7,0,1,10,11,12,13,14,15,8,9]
 ; AVX512VL-NEXT:    retq

diff --git a/llvm/test/CodeGen/X86/vector-shuffle-128-v8.ll b/llvm/test/CodeGen/X86/vector-shuffle-128-v8.ll
index 95669b3..28c19e3 100644 (file)
--- a/llvm/test/CodeGen/X86/vector-shuffle-128-v8.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-128-v8.ll
@@ -2294,6 +2294,82 @@ define <8 x i16> @shuffle_v8i16_467uu675(<8 x i16> %a) {
   ret <8 x i16> %shuffle
 }
 
+define <8 x i16> @shuffle_v8i16_10325476(<8 x i16> %a) {
+; SSE-LABEL: shuffle_v8i16_10325476:
+; SSE:       # %bb.0:
+; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,0,3,2,4,5,6,7]
+; SSE-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,4,7,6]
+; SSE-NEXT:    retq
+;
+; AVX1-LABEL: shuffle_v8i16_10325476:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[1,0,3,2,4,5,6,7]
+; AVX1-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,4,7,6]
+; AVX1-NEXT:    retq
+;
+; AVX2-SLOW-LABEL: shuffle_v8i16_10325476:
+; AVX2-SLOW:       # %bb.0:
+; AVX2-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[1,0,3,2,4,5,6,7]
+; AVX2-SLOW-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,4,7,6]
+; AVX2-SLOW-NEXT:    retq
+;
+; AVX2-FAST-LABEL: shuffle_v8i16_10325476:
+; AVX2-FAST:       # %bb.0:
+; AVX2-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[2,3,0,1,6,7,4,5,10,11,8,9,14,15,12,13]
+; AVX2-FAST-NEXT:    retq
+;
+; AVX512VL-SLOW-LABEL: shuffle_v8i16_10325476:
+; AVX512VL-SLOW:       # %bb.0:
+; AVX512VL-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[1,0,3,2,4,5,6,7]
+; AVX512VL-SLOW-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,4,7,6]
+; AVX512VL-SLOW-NEXT:    retq
+;
+; AVX512VL-FAST-LABEL: shuffle_v8i16_10325476:
+; AVX512VL-FAST:       # %bb.0:
+; AVX512VL-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[2,3,0,1,6,7,4,5,10,11,8,9,14,15,12,13]
+; AVX512VL-FAST-NEXT:    retq
+  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32> <i32 1, i32 0, i32 3, i32 2, i32 5, i32 4, i32 7, i32 6>
+  ret <8 x i16> %shuffle
+}
+
+define <8 x i16> @shuffle_v8i16_12305674(<8 x i16> %a) {
+; SSE-LABEL: shuffle_v8i16_12305674:
+; SSE:       # %bb.0:
+; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,2,3,0,4,5,6,7]
+; SSE-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,6,7,4]
+; SSE-NEXT:    retq
+;
+; AVX1-LABEL: shuffle_v8i16_12305674:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[1,2,3,0,4,5,6,7]
+; AVX1-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,6,7,4]
+; AVX1-NEXT:    retq
+;
+; AVX2-SLOW-LABEL: shuffle_v8i16_12305674:
+; AVX2-SLOW:       # %bb.0:
+; AVX2-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[1,2,3,0,4,5,6,7]
+; AVX2-SLOW-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,6,7,4]
+; AVX2-SLOW-NEXT:    retq
+;
+; AVX2-FAST-LABEL: shuffle_v8i16_12305674:
+; AVX2-FAST:       # %bb.0:
+; AVX2-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[2,3,4,5,6,7,0,1,10,11,12,13,14,15,8,9]
+; AVX2-FAST-NEXT:    retq
+;
+; AVX512VL-SLOW-LABEL: shuffle_v8i16_12305674:
+; AVX512VL-SLOW:       # %bb.0:
+; AVX512VL-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[1,2,3,0,4,5,6,7]
+; AVX512VL-SLOW-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,6,7,4]
+; AVX512VL-SLOW-NEXT:    retq
+;
+; AVX512VL-FAST-LABEL: shuffle_v8i16_12305674:
+; AVX512VL-FAST:       # %bb.0:
+; AVX512VL-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[2,3,4,5,6,7,0,1,10,11,12,13,14,15,8,9]
+; AVX512VL-FAST-NEXT:    retq
+  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32> <i32 1, i32 2, i32 3, i32 0, i32 5, i32 6, i32 7, i32 4>
+  ret <8 x i16> %shuffle
+}
+
 ;
 ; Shuffle to logical bit shifts
 ;

diff --git a/llvm/test/CodeGen/X86/vector-shuffle-256-v16.ll b/llvm/test/CodeGen/X86/vector-shuffle-256-v16.ll
index 001288f..087f38d 100644 (file)
--- a/llvm/test/CodeGen/X86/vector-shuffle-256-v16.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-256-v16.ll
@@ -4508,6 +4508,78 @@ define <16 x i16> @shuffle_v16i16_00_01_02_04_07_06_05_04_08_09_10_11_15_14_13_1
   ret <16 x i16> %3
 }
 
+define <16 x i16> @shuffle_v16i16_01_00_03_02_05_04_07_06_09_08_11_10_13_12_15_14(<16 x i16> %a) {
+; AVX1-LABEL: shuffle_v16i16_01_00_03_02_05_04_07_06_09_08_11_10_13_12_15_14:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vpshuflw {{.*#+}} xmm1 = xmm0[1,0,3,2,4,5,6,7]
+; AVX1-NEXT:    vpshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,5,4,7,6]
+; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[1,0,3,2,4,5,6,7]
+; AVX1-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,4,7,6]
+; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-SLOW-LABEL: shuffle_v16i16_01_00_03_02_05_04_07_06_09_08_11_10_13_12_15_14:
+; AVX2-SLOW:       # %bb.0:
+; AVX2-SLOW-NEXT:    vpshuflw {{.*#+}} ymm0 = ymm0[1,0,3,2,4,5,6,7,9,8,11,10,12,13,14,15]
+; AVX2-SLOW-NEXT:    vpshufhw {{.*#+}} ymm0 = ymm0[0,1,2,3,5,4,7,6,8,9,10,11,13,12,15,14]
+; AVX2-SLOW-NEXT:    retq
+;
+; AVX2-FAST-LABEL: shuffle_v16i16_01_00_03_02_05_04_07_06_09_08_11_10_13_12_15_14:
+; AVX2-FAST:       # %bb.0:
+; AVX2-FAST-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[2,3,0,1,6,7,4,5,10,11,8,9,14,15,12,13,18,19,16,17,22,23,20,21,26,27,24,25,30,31,28,29]
+; AVX2-FAST-NEXT:    retq
+;
+; AVX512VL-SLOW-LABEL: shuffle_v16i16_01_00_03_02_05_04_07_06_09_08_11_10_13_12_15_14:
+; AVX512VL-SLOW:       # %bb.0:
+; AVX512VL-SLOW-NEXT:    vpshuflw {{.*#+}} ymm0 = ymm0[1,0,3,2,4,5,6,7,9,8,11,10,12,13,14,15]
+; AVX512VL-SLOW-NEXT:    vpshufhw {{.*#+}} ymm0 = ymm0[0,1,2,3,5,4,7,6,8,9,10,11,13,12,15,14]
+; AVX512VL-SLOW-NEXT:    retq
+;
+; AVX512VL-FAST-LABEL: shuffle_v16i16_01_00_03_02_05_04_07_06_09_08_11_10_13_12_15_14:
+; AVX512VL-FAST:       # %bb.0:
+; AVX512VL-FAST-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[2,3,0,1,6,7,4,5,10,11,8,9,14,15,12,13,18,19,16,17,22,23,20,21,26,27,24,25,30,31,28,29]
+; AVX512VL-FAST-NEXT:    retq
+  %shuffle = shufflevector <16 x i16> %a, <16 x i16> undef, <16 x i32> <i32 1, i32 0, i32 3, i32 2, i32 5, i32 4, i32 7, i32 6, i32 9, i32 8, i32 11, i32 10, i32 13, i32 12, i32 15, i32 14>
+  ret <16 x i16> %shuffle
+}
+
+define <16 x i16> @shuffle_v16i16_03_00_01_02_07_04_05_06_11_08_09_10_15_12_13_14(<16 x i16> %a) {
+; AVX1-LABEL: shuffle_v16i16_03_00_01_02_07_04_05_06_11_08_09_10_15_12_13_14:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vpshuflw {{.*#+}} xmm1 = xmm0[3,0,1,2,4,5,6,7]
+; AVX1-NEXT:    vpshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,7,4,5,6]
+; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[3,0,1,2,4,5,6,7]
+; AVX1-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,4,5,6]
+; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-SLOW-LABEL: shuffle_v16i16_03_00_01_02_07_04_05_06_11_08_09_10_15_12_13_14:
+; AVX2-SLOW:       # %bb.0:
+; AVX2-SLOW-NEXT:    vpshuflw {{.*#+}} ymm0 = ymm0[3,0,1,2,4,5,6,7,11,8,9,10,12,13,14,15]
+; AVX2-SLOW-NEXT:    vpshufhw {{.*#+}} ymm0 = ymm0[0,1,2,3,7,4,5,6,8,9,10,11,15,12,13,14]
+; AVX2-SLOW-NEXT:    retq
+;
+; AVX2-FAST-LABEL: shuffle_v16i16_03_00_01_02_07_04_05_06_11_08_09_10_15_12_13_14:
+; AVX2-FAST:       # %bb.0:
+; AVX2-FAST-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[6,7,0,1,2,3,4,5,14,15,8,9,10,11,12,13,22,23,16,17,18,19,20,21,30,31,24,25,26,27,28,29]
+; AVX2-FAST-NEXT:    retq
+;
+; AVX512VL-SLOW-LABEL: shuffle_v16i16_03_00_01_02_07_04_05_06_11_08_09_10_15_12_13_14:
+; AVX512VL-SLOW:       # %bb.0:
+; AVX512VL-SLOW-NEXT:    vpshuflw {{.*#+}} ymm0 = ymm0[3,0,1,2,4,5,6,7,11,8,9,10,12,13,14,15]
+; AVX512VL-SLOW-NEXT:    vpshufhw {{.*#+}} ymm0 = ymm0[0,1,2,3,7,4,5,6,8,9,10,11,15,12,13,14]
+; AVX512VL-SLOW-NEXT:    retq
+;
+; AVX512VL-FAST-LABEL: shuffle_v16i16_03_00_01_02_07_04_05_06_11_08_09_10_15_12_13_14:
+; AVX512VL-FAST:       # %bb.0:
+; AVX512VL-FAST-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[6,7,0,1,2,3,4,5,14,15,8,9,10,11,12,13,22,23,16,17,18,19,20,21,30,31,24,25,26,27,28,29]
+; AVX512VL-FAST-NEXT:    retq
+  %shuffle = shufflevector <16 x i16> %a, <16 x i16> undef, <16 x i32> <i32 3, i32 0, i32 1, i32 2, i32 7, i32 4, i32 5, i32 6, i32 11, i32 8, i32 9, i32 10, i32 15, i32 12, i32 13, i32 14>
+  ret <16 x i16> %shuffle
+}
+
 define <16 x i16> @insert_v16i16_0elt_into_zero_vector(i16* %ptr) {
 ; ALL-LABEL: insert_v16i16_0elt_into_zero_vector:
 ; ALL:       # %bb.0:

diff --git a/llvm/test/CodeGen/X86/vector-shuffle-256-v32.ll b/llvm/test/CodeGen/X86/vector-shuffle-256-v32.ll
index 6b04bad..5bd9e2d 100644 (file)
--- a/llvm/test/CodeGen/X86/vector-shuffle-256-v32.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-256-v32.ll
@@ -3058,6 +3058,24 @@ define <32 x i8> @shuffle_v32i8_31_32_33_34_35_36_37_38_39_40_41_42_43_44_45_46_
   ret <32 x i8> %shuffle
 }
 
+define <32 x i8> @shuffle_v32i8_03_00_01_02_07_04_05_06_11_08_09_10_15_12_13_14_19_16_17_18_23_20_21_22_27_24_25_26_31_28_29_30(<32 x i8> %a) {
+; AVX1-LABEL: shuffle_v32i8_03_00_01_02_07_04_05_06_11_08_09_10_15_12_13_14_19_16_17_18_23_20_21_22_27_24_25_26_31_28_29_30:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
+; AVX1-NEXT:    vmovdqa {{.*#+}} xmm2 = [3,0,1,2,7,4,5,6,11,8,9,10,15,12,13,14]
+; AVX1-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
+; AVX1-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
+; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2OR512VL-LABEL: shuffle_v32i8_03_00_01_02_07_04_05_06_11_08_09_10_15_12_13_14_19_16_17_18_23_20_21_22_27_24_25_26_31_28_29_30:
+; AVX2OR512VL:       # %bb.0:
+; AVX2OR512VL-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[3,0,1,2,7,4,5,6,11,8,9,10,15,12,13,14,19,16,17,18,23,20,21,22,27,24,25,26,31,28,29,30]
+; AVX2OR512VL-NEXT:    retq
+  %shuffle = shufflevector <32 x i8> %a, <32 x i8> undef, <32 x i32> <i32 3, i32 0, i32 1, i32 2, i32 7, i32 4, i32 5, i32 6, i32 11, i32 8, i32 9, i32 10, i32 15, i32 12, i32 13, i32 14, i32 19, i32 16, i32 17, i32 18, i32 23, i32 20, i32 21, i32 22, i32 27, i32 24, i32 25, i32 26, i32 31, i32 28, i32 29, i32 30>
+  ret <32 x i8> %shuffle
+}
+
 ; PR44379
 define <32 x i8> @shuffle_v32i8_02_03_04_05_06_07_00_01_10_11_12_13_14_15_08_09_18_19_20_21_22_23_16_17_26_27_28_29_30_31_24_25(<32 x i8> %a) {
 ; AVX1-LABEL: shuffle_v32i8_02_03_04_05_06_07_00_01_10_11_12_13_14_15_08_09_18_19_20_21_22_23_16_17_26_27_28_29_30_31_24_25:

diff --git a/llvm/test/CodeGen/X86/vector-shuffle-512-v32.ll b/llvm/test/CodeGen/X86/vector-shuffle-512-v32.ll
index c350e7f..99742b6 100644 (file)
--- a/llvm/test/CodeGen/X86/vector-shuffle-512-v32.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-512-v32.ll
@@ -197,6 +197,44 @@ define <32 x i16> @shuffle_v32i16_1_1_0_0_5_5_4_4_9_9_11_11_13_13_12_12_17_17_19
   ret <32 x i16> %c
 }
 
+define <32 x i16> @shuffle_v32i16_01_00_03_02_05_04_07_06_09_08_11_10_13_12_15_14_17_16_19_18_21_20_23_22_25_24_27_26_29_28_31_30(<32 x i16> %a) {
+; KNL-LABEL: shuffle_v32i16_01_00_03_02_05_04_07_06_09_08_11_10_13_12_15_14_17_16_19_18_21_20_23_22_25_24_27_26_29_28_31_30:
+; KNL:       ## %bb.0:
+; KNL-NEXT:    vextracti64x4 $1, %zmm0, %ymm1
+; KNL-NEXT:    vpshuflw {{.*#+}} ymm1 = ymm1[1,0,3,2,4,5,6,7,9,8,11,10,12,13,14,15]
+; KNL-NEXT:    vpshufhw {{.*#+}} ymm1 = ymm1[0,1,2,3,5,4,7,6,8,9,10,11,13,12,15,14]
+; KNL-NEXT:    vpshuflw {{.*#+}} ymm0 = ymm0[1,0,3,2,4,5,6,7,9,8,11,10,12,13,14,15]
+; KNL-NEXT:    vpshufhw {{.*#+}} ymm0 = ymm0[0,1,2,3,5,4,7,6,8,9,10,11,13,12,15,14]
+; KNL-NEXT:    vinserti64x4 $1, %ymm1, %zmm0, %zmm0
+; KNL-NEXT:    retq
+;
+; SKX-LABEL: shuffle_v32i16_01_00_03_02_05_04_07_06_09_08_11_10_13_12_15_14_17_16_19_18_21_20_23_22_25_24_27_26_29_28_31_30:
+; SKX:       ## %bb.0:
+; SKX-NEXT:    vpshufb {{.*#+}} zmm0 = zmm0[2,3,0,1,6,7,4,5,10,11,8,9,14,15,12,13,18,19,16,17,22,23,20,21,26,27,24,25,30,31,28,29,34,35,32,33,38,39,36,37,42,43,40,41,46,47,44,45,50,51,48,49,54,55,52,53,58,59,56,57,62,63,60,61]
+; SKX-NEXT:    retq
+  %shuffle = shufflevector <32 x i16> %a, <32 x i16> undef, <32 x i32> <i32 1, i32 0, i32 3, i32 2, i32 5, i32 4, i32 7, i32 6, i32 9, i32 8, i32 11, i32 10, i32 13, i32 12, i32 15, i32 14, i32 17, i32 16, i32 19, i32 18, i32 21, i32 20, i32 23, i32 22, i32 25, i32 24, i32 27, i32 26, i32 29, i32 28, i32 31, i32 30>
+  ret <32 x i16> %shuffle
+}
+
+define <32 x i16> @shuffle_v32i16_03_00_01_02_07_04_05_06_11_08_09_10_15_12_13_14_19_16_17_18_23_20_21_22_27_24_25_26_31_28_29_30(<32 x i16> %a) {
+; KNL-LABEL: shuffle_v32i16_03_00_01_02_07_04_05_06_11_08_09_10_15_12_13_14_19_16_17_18_23_20_21_22_27_24_25_26_31_28_29_30:
+; KNL:       ## %bb.0:
+; KNL-NEXT:    vextracti64x4 $1, %zmm0, %ymm1
+; KNL-NEXT:    vpshuflw {{.*#+}} ymm1 = ymm1[3,0,1,2,4,5,6,7,11,8,9,10,12,13,14,15]
+; KNL-NEXT:    vpshufhw {{.*#+}} ymm1 = ymm1[0,1,2,3,7,4,5,6,8,9,10,11,15,12,13,14]
+; KNL-NEXT:    vpshuflw {{.*#+}} ymm0 = ymm0[3,0,1,2,4,5,6,7,11,8,9,10,12,13,14,15]
+; KNL-NEXT:    vpshufhw {{.*#+}} ymm0 = ymm0[0,1,2,3,7,4,5,6,8,9,10,11,15,12,13,14]
+; KNL-NEXT:    vinserti64x4 $1, %ymm1, %zmm0, %zmm0
+; KNL-NEXT:    retq
+;
+; SKX-LABEL: shuffle_v32i16_03_00_01_02_07_04_05_06_11_08_09_10_15_12_13_14_19_16_17_18_23_20_21_22_27_24_25_26_31_28_29_30:
+; SKX:       ## %bb.0:
+; SKX-NEXT:    vpshufb {{.*#+}} zmm0 = zmm0[6,7,0,1,2,3,4,5,14,15,8,9,10,11,12,13,22,23,16,17,18,19,20,21,30,31,24,25,26,27,28,29,38,39,32,33,34,35,36,37,46,47,40,41,42,43,44,45,54,55,48,49,50,51,52,53,62,63,56,57,58,59,60,61]
+; SKX-NEXT:    retq
+  %shuffle = shufflevector <32 x i16> %a, <32 x i16> undef, <32 x i32> <i32 3, i32 0, i32 1, i32 2, i32 7, i32 4, i32 5, i32 6, i32 11, i32 8, i32 9, i32 10, i32 15, i32 12, i32 13, i32 14, i32 19, i32 16, i32 17, i32 18, i32 23, i32 20, i32 21, i32 22, i32 27, i32 24, i32 25, i32 26, i32 31, i32 28, i32 29, i32 30>
+  ret <32 x i16> %shuffle
+}
+
 define <32 x i16> @shuffle_v32i16_0zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz(<32 x i16> %a) {
 ; KNL-LABEL: shuffle_v32i16_0zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz:
 ; KNL:       ## %bb.0: