CodeGen peephole: fold redundant phys reg copies

[oota-llvm.git] / test / CodeGen / X86 / vector-shuffle-128-v16.ll

diff --git a/test/CodeGen/X86/vector-shuffle-128-v16.ll b/test/CodeGen/X86/vector-shuffle-128-v16.ll
index 25c377f05b6de36ad39a424da6f1d155a62947e1..13a9543ddd90cb9fa1d828763069ca8d51c92ba7 100644 (file)
--- a/test/CodeGen/X86/vector-shuffle-128-v16.ll
+++ b/test/CodeGen/X86/vector-shuffle-128-v16.ll
@@ -1,8 +1,9 @@
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -x86-experimental-vector-shuffle-legality | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE2
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -mattr=+ssse3 -x86-experimental-vector-shuffle-legality | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSSE3
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -mattr=+sse4.1 -x86-experimental-vector-shuffle-legality | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE41
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -mattr=+avx -x86-experimental-vector-shuffle-legality | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX1
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -mattr=+avx2 -x86-experimental-vector-shuffle-legality | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX2
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=x86-64 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE2
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -mattr=+ssse3 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSSE3
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -mattr=+sse4.1 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE41
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -mattr=+avx | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX1
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -mattr=+avx2 | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX2
 
 target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
 target triple = "x86_64-unknown-unknown"
@@ -249,21 +250,14 @@ define <16 x i8> @shuffle_v16i8_08_24_09_25_10_26_11_27_12_28_13_29_14_30_15_31(
 define <16 x i8> @shuffle_v16i8_16_00_16_01_16_02_16_03_16_04_16_05_16_06_16_07(<16 x i8> %a, <16 x i8> %b) {
 ; SSE2-LABEL: shuffle_v16i8_16_00_16_01_16_02_16_03_16_04_16_05_16_06_16_07:
 ; SSE2:       # BB#0:
-; SSE2-NEXT:    pxor %xmm2, %xmm2
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm2[0],xmm0[1],xmm2[1],xmm0[2],xmm2[2],xmm0[3],xmm2[3],xmm0[4],xmm2[4],xmm0[5],xmm2[5],xmm0[6],xmm2[6],xmm0[7],xmm2[7]
-; SSE2-NEXT:    movdqa %xmm0, %xmm3
-; SSE2-NEXT:    punpckhwd {{.*#+}} xmm3 = xmm3[4,4,5,5,6,6,7,7]
-; SSE2-NEXT:    movdqa {{.*#+}} xmm2 = [65535,0,65535,0,65535,0,65535,0]
-; SSE2-NEXT:    movdqa %xmm2, %xmm4
-; SSE2-NEXT:    pandn %xmm3, %xmm4
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm1
-; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[0,0,0,0]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
+; SSE2-NEXT:    movdqa {{.*#+}} xmm2 = [255,0,255,0,255,0,255,0,255,0,255,0,255,0,255,0]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[0,1,0,3]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm1 = xmm1[0,0,0,0,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,4,4,4,4]
 ; SSE2-NEXT:    pand %xmm2, %xmm1
-; SSE2-NEXT:    por %xmm1, %xmm4
-; SSE2-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3]
 ; SSE2-NEXT:    pandn %xmm0, %xmm2
 ; SSE2-NEXT:    por %xmm1, %xmm2
-; SSE2-NEXT:    packuswb %xmm4, %xmm2
 ; SSE2-NEXT:    movdqa %xmm2, %xmm0
 ; SSE2-NEXT:    retq
 ;
@@ -346,26 +340,20 @@ define <16 x i8> @shuffle_v16i8_03_02_01_00_07_06_05_04_19_18_17_16_23_22_21_20(
 ;
 ; SSSE3-LABEL: shuffle_v16i8_03_02_01_00_07_06_05_04_19_18_17_16_23_22_21_20:
 ; SSSE3:       # BB#0:
-; SSSE3-NEXT:    movdqa {{.*#+}} xmm2 = <3,2,1,0,7,6,5,4,u,u,u,u,u,u,u,u>
-; SSSE3-NEXT:    pshufb %xmm2, %xmm1
-; SSSE3-NEXT:    pshufb %xmm2, %xmm0
-; SSSE3-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; SSSE3-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]
+; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[6,4,2,0,14,12,10,8,7,5,3,1,15,13,11,9]
 ; SSSE3-NEXT:    retq
 ;
 ; SSE41-LABEL: shuffle_v16i8_03_02_01_00_07_06_05_04_19_18_17_16_23_22_21_20:
 ; SSE41:       # BB#0:
-; SSE41-NEXT:    movdqa {{.*#+}} xmm2 = <3,2,1,0,7,6,5,4,u,u,u,u,u,u,u,u>
-; SSE41-NEXT:    pshufb %xmm2, %xmm1
-; SSE41-NEXT:    pshufb %xmm2, %xmm0
-; SSE41-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; SSE41-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]
+; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[6,4,2,0,14,12,10,8,7,5,3,1,15,13,11,9]
 ; SSE41-NEXT:    retq
 ;
 ; AVX-LABEL: shuffle_v16i8_03_02_01_00_07_06_05_04_19_18_17_16_23_22_21_20:
 ; AVX:       # BB#0:
-; AVX-NEXT:    vmovdqa {{.*#+}} xmm2 = <3,2,1,0,7,6,5,4,u,u,u,u,u,u,u,u>
-; AVX-NEXT:    vpshufb %xmm2, %xmm1, %xmm1
-; AVX-NEXT:    vpshufb %xmm2, %xmm0, %xmm0
-; AVX-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX-NEXT:    vpunpcklbw {{.*#+}} 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]
+; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[6,4,2,0,14,12,10,8,7,5,3,1,15,13,11,9]
 ; AVX-NEXT:    retq
   %shuffle = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 3, i32 2, i32 1, i32 0, i32 7, i32 6, i32 5, i32 4, i32 19, i32 18, i32 17, i32 16, i32 23, i32 22, i32 21, i32 20>
   ret <16 x i8> %shuffle
@@ -374,21 +362,22 @@ define <16 x i8> @shuffle_v16i8_03_02_01_00_07_06_05_04_19_18_17_16_23_22_21_20(
 define <16 x i8> @shuffle_v16i8_03_02_01_00_31_30_29_28_11_10_09_08_23_22_21_20(<16 x i8> %a, <16 x i8> %b) {
 ; SSE2-LABEL: shuffle_v16i8_03_02_01_00_31_30_29_28_11_10_09_08_23_22_21_20:
 ; SSE2:       # BB#0:
-; SSE2-NEXT:    pxor %xmm2, %xmm2
-; SSE2-NEXT:    movdqa %xmm1, %xmm3
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm3 = xmm3[0],xmm2[0],xmm3[1],xmm2[1],xmm3[2],xmm2[2],xmm3[3],xmm2[3],xmm3[4],xmm2[4],xmm3[5],xmm2[5],xmm3[6],xmm2[6],xmm3[7],xmm2[7]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm3 = xmm3[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    movdqa %xmm0, %xmm4
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm2[8],xmm4[9],xmm2[9],xmm4[10],xmm2[10],xmm4[11],xmm2[11],xmm4[12],xmm2[12],xmm4[13],xmm2[13],xmm4[14],xmm2[14],xmm4[15],xmm2[15]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm4 = xmm4[3,2,1,0,4,5,6,7]
-; SSE2-NEXT:    movsd {{.*#+}} xmm3 = xmm4[0],xmm3[1]
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm1 = xmm1[8],xmm2[8],xmm1[9],xmm2[9],xmm1[10],xmm2[10],xmm1[11],xmm2[11],xmm1[12],xmm2[12],xmm1[13],xmm2[13],xmm1[14],xmm2[14],xmm1[15],xmm2[15]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm2[0],xmm0[1],xmm2[1],xmm0[2],xmm2[2],xmm0[3],xmm2[3],xmm0[4],xmm2[4],xmm0[5],xmm2[5],xmm0[6],xmm2[6],xmm0[7],xmm2[7]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[1,3,2,3]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
+; SSE2-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
+; 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 {{.*#+}} xmm3 = xmm2[3,2,1,0,4,5,6,7]
+; 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:    pshufd {{.*#+}} xmm1 = xmm0[2,3,2,3]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm1 = xmm1[3,2,1,0,4,5,6,7]
+; SSE2-NEXT:    punpcklqdq {{.*#+}} xmm3 = xmm3[0],xmm1[0]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[3,2,1,0,4,5,6,7]
-; SSE2-NEXT:    movsd {{.*#+}} xmm1 = xmm0[0],xmm1[1]
-; SSE2-NEXT:    packuswb %xmm3, %xmm1
-; SSE2-NEXT:    movdqa %xmm1, %xmm0
+; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm2[2,3,2,3]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm1 = xmm1[3,2,1,0,4,5,6,7]
+; SSE2-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; SSE2-NEXT:    packuswb %xmm3, %xmm0
 ; SSE2-NEXT:    retq
 ;
 ; SSSE3-LABEL: shuffle_v16i8_03_02_01_00_31_30_29_28_11_10_09_08_23_22_21_20:
@@ -481,6 +470,20 @@ define <16 x i8> @shuffle_v16i8_00_01_02_19_04_05_06_23_08_09_10_27_12_13_14_31(
   ret <16 x i8> %shuffle
 }
 
+define <16 x i8> @shuffle_v16i8_00_01_02_zz_04_05_06_zz_08_09_10_zz_12_13_14_zz(<16 x i8> %a) {
+; SSE-LABEL: shuffle_v16i8_00_01_02_zz_04_05_06_zz_08_09_10_zz_12_13_14_zz:
+; SSE:       # BB#0:
+; SSE-NEXT:    andps {{.*}}(%rip), %xmm0
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: shuffle_v16i8_00_01_02_zz_04_05_06_zz_08_09_10_zz_12_13_14_zz:
+; AVX:       # BB#0:
+; AVX-NEXT:    vandps {{.*}}(%rip), %xmm0, %xmm0
+; AVX-NEXT:    retq
+  %shuffle = shufflevector <16 x i8> %a, <16 x i8> zeroinitializer, <16 x i32> <i32 0, i32 1, i32 2, i32 19, i32 4, i32 5, i32 6, i32 23, i32 8, i32 9, i32 10, i32 27, i32 12, i32 13, i32 14, i32 31>
+  ret <16 x i8> %shuffle
+}
+
 define <16 x i8> @shuffle_v16i8_00_01_02_03_20_05_06_23_08_09_10_11_28_13_14_31(<16 x i8> %a, <16 x i8> %b) {
 ; SSE2-LABEL: shuffle_v16i8_00_01_02_03_20_05_06_23_08_09_10_11_28_13_14_31:
 ; SSE2:       # BB#0:
@@ -623,11 +626,8 @@ define <16 x i8> @PR20540(<8 x i8> %a) {
 ; SSE2-LABEL: PR20540:
 ; SSE2:       # BB#0:
 ; SSE2-NEXT:    pand {{.*}}(%rip), %xmm0
-; SSE2-NEXT:    pxor %xmm1, %xmm1
-; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[0,1,0,3]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm1 = xmm1[0,0,0,0,4,5,6,7]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,4,4,4,4]
-; SSE2-NEXT:    packuswb %xmm1, %xmm0
+; SSE2-NEXT:    packuswb %xmm0, %xmm0
+; SSE2-NEXT:    movq {{.*#+}} xmm0 = xmm0[0],zero
 ; SSE2-NEXT:    retq
 ;
 ; SSSE3-LABEL: PR20540:
@@ -649,28 +649,16 @@ define <16 x i8> @PR20540(<8 x i8> %a) {
 }
 
 define <16 x i8> @shuffle_v16i8_16_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz(i8 %i) {
-; SSE2-LABEL: shuffle_v16i8_16_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz:
-; SSE2:       # BB#0:
-; SSE2-NEXT:    movzbl %dil, %eax
-; SSE2-NEXT:    movd %eax, %xmm0
-; SSE2-NEXT:    retq
-;
-; SSSE3-LABEL: shuffle_v16i8_16_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz:
-; SSSE3:       # BB#0:
-; SSSE3-NEXT:    movd %edi, %xmm0
-; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
-; SSSE3-NEXT:    retq
-;
-; SSE41-LABEL: shuffle_v16i8_16_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz:
-; SSE41:       # BB#0:
-; SSE41-NEXT:    movd %edi, %xmm0
-; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
-; SSE41-NEXT:    retq
+; SSE-LABEL: shuffle_v16i8_16_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz:
+; SSE:       # BB#0:
+; SSE-NEXT:    movzbl %dil, %eax
+; SSE-NEXT:    movd %eax, %xmm0
+; SSE-NEXT:    retq
 ;
 ; AVX-LABEL: shuffle_v16i8_16_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz:
 ; AVX:       # BB#0:
-; AVX-NEXT:    vmovd %edi, %xmm0
-; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
+; AVX-NEXT:    movzbl %dil, %eax
+; AVX-NEXT:    vmovd %eax, %xmm0
 ; AVX-NEXT:    retq
   %a = insertelement <16 x i8> undef, i8 %i, i32 0
   %shuffle = shufflevector <16 x i8> zeroinitializer, <16 x i8> %a, <16 x i32> <i32 16, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
@@ -680,27 +668,28 @@ define <16 x i8> @shuffle_v16i8_16_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz(
 define <16 x i8> @shuffle_v16i8_zz_zz_zz_zz_zz_16_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz(i8 %i) {
 ; SSE2-LABEL: shuffle_v16i8_zz_zz_zz_zz_zz_16_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz:
 ; SSE2:       # BB#0:
-; SSE2-NEXT:    movzbl %dil, %eax
-; SSE2-NEXT:    movd %eax, %xmm0
-; SSE2-NEXT:    pslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5,6,7,8,9,10]
+; SSE2-NEXT:    shll $8, %edi
+; SSE2-NEXT:    pxor %xmm0, %xmm0
+; SSE2-NEXT:    pinsrw $2, %edi, %xmm0
 ; SSE2-NEXT:    retq
 ;
 ; SSSE3-LABEL: shuffle_v16i8_zz_zz_zz_zz_zz_16_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz:
 ; SSSE3:       # BB#0:
-; SSSE3-NEXT:    movd %edi, %xmm0
-; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = zero,zero,zero,zero,zero,xmm0[0],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
+; SSSE3-NEXT:    shll $8, %edi
+; SSSE3-NEXT:    pxor %xmm0, %xmm0
+; SSSE3-NEXT:    pinsrw $2, %edi, %xmm0
 ; SSSE3-NEXT:    retq
 ;
 ; SSE41-LABEL: shuffle_v16i8_zz_zz_zz_zz_zz_16_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz:
 ; SSE41:       # BB#0:
-; SSE41-NEXT:    movd %edi, %xmm0
-; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = zero,zero,zero,zero,zero,xmm0[0],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
+; SSE41-NEXT:    pxor %xmm0, %xmm0
+; SSE41-NEXT:    pinsrb $5, %edi, %xmm0
 ; SSE41-NEXT:    retq
 ;
 ; AVX-LABEL: shuffle_v16i8_zz_zz_zz_zz_zz_16_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz:
 ; AVX:       # BB#0:
-; AVX-NEXT:    vmovd %edi, %xmm0
-; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = zero,zero,zero,zero,zero,xmm0[0],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
+; AVX-NEXT:    vpxor %xmm0, %xmm0, %xmm0
+; AVX-NEXT:    vpinsrb $5, %edi, %xmm0, %xmm0
 ; AVX-NEXT:    retq
   %a = insertelement <16 x i8> undef, i8 %i, i32 0
   %shuffle = shufflevector <16 x i8> zeroinitializer, <16 x i8> %a, <16 x i32> <i32 0, i32 0, i32 0, i32 0, i32 0, i32 16, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>
@@ -708,16 +697,30 @@ define <16 x i8> @shuffle_v16i8_zz_zz_zz_zz_zz_16_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz(
 }
 
 define <16 x i8> @shuffle_v16i8_zz_uu_uu_zz_uu_uu_zz_zz_zz_zz_zz_zz_zz_zz_zz_16(i8 %i) {
-; SSE-LABEL: shuffle_v16i8_zz_uu_uu_zz_uu_uu_zz_zz_zz_zz_zz_zz_zz_zz_zz_16:
-; SSE:       # BB#0:
-; SSE-NEXT:    movd %edi, %xmm0
-; SSE-NEXT:    pslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0]
-; SSE-NEXT:    retq
+; SSE2-LABEL: shuffle_v16i8_zz_uu_uu_zz_uu_uu_zz_zz_zz_zz_zz_zz_zz_zz_zz_16:
+; SSE2:       # BB#0:
+; SSE2-NEXT:    shll $8, %edi
+; SSE2-NEXT:    pxor %xmm0, %xmm0
+; SSE2-NEXT:    pinsrw $7, %edi, %xmm0
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: shuffle_v16i8_zz_uu_uu_zz_uu_uu_zz_zz_zz_zz_zz_zz_zz_zz_zz_16:
+; SSSE3:       # BB#0:
+; SSSE3-NEXT:    shll $8, %edi
+; SSSE3-NEXT:    pxor %xmm0, %xmm0
+; SSSE3-NEXT:    pinsrw $7, %edi, %xmm0
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: shuffle_v16i8_zz_uu_uu_zz_uu_uu_zz_zz_zz_zz_zz_zz_zz_zz_zz_16:
+; SSE41:       # BB#0:
+; SSE41-NEXT:    pxor %xmm0, %xmm0
+; SSE41-NEXT:    pinsrb $15, %edi, %xmm0
+; SSE41-NEXT:    retq
 ;
 ; AVX-LABEL: shuffle_v16i8_zz_uu_uu_zz_uu_uu_zz_zz_zz_zz_zz_zz_zz_zz_zz_16:
 ; AVX:       # BB#0:
-; AVX-NEXT:    vmovd %edi, %xmm0
-; AVX-NEXT:    vpslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0]
+; AVX-NEXT:    vpxor %xmm0, %xmm0, %xmm0
+; AVX-NEXT:    vpinsrb $15, %edi, %xmm0, %xmm0
 ; AVX-NEXT:    retq
   %a = insertelement <16 x i8> undef, i8 %i, i32 0
   %shuffle = shufflevector <16 x i8> zeroinitializer, <16 x i8> %a, <16 x i32> <i32 0, i32 undef, i32 undef, i32 3, i32 undef, i32 undef, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 16>
@@ -728,29 +731,27 @@ define <16 x i8> @shuffle_v16i8_zz_zz_19_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz(
 ; SSE2-LABEL: shuffle_v16i8_zz_zz_19_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz:
 ; SSE2:       # BB#0:
 ; SSE2-NEXT:    movzbl %dil, %eax
-; SSE2-NEXT:    movd %eax, %xmm0
-; SSE2-NEXT:    pslldq {{.*#+}} xmm0 = zero,zero,xmm0[0,1,2,3,4,5,6,7,8,9,10,11,12,13]
+; SSE2-NEXT:    pxor %xmm0, %xmm0
+; SSE2-NEXT:    pinsrw $1, %eax, %xmm0
 ; SSE2-NEXT:    retq
 ;
 ; SSSE3-LABEL: shuffle_v16i8_zz_zz_19_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz:
 ; SSSE3:       # BB#0:
-; SSSE3-NEXT:    movd %edi, %xmm0
-; SSSE3-NEXT:    pslld $24, %xmm0
-; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = zero,zero,xmm0[3],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
+; SSSE3-NEXT:    movzbl %dil, %eax
+; SSSE3-NEXT:    pxor %xmm0, %xmm0
+; SSSE3-NEXT:    pinsrw $1, %eax, %xmm0
 ; SSSE3-NEXT:    retq
 ;
 ; SSE41-LABEL: shuffle_v16i8_zz_zz_19_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz:
 ; SSE41:       # BB#0:
-; SSE41-NEXT:    movd %edi, %xmm0
-; SSE41-NEXT:    pslld $24, %xmm0
-; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = zero,zero,xmm0[3],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
+; SSE41-NEXT:    pxor %xmm0, %xmm0
+; SSE41-NEXT:    pinsrb $2, %edi, %xmm0
 ; SSE41-NEXT:    retq
 ;
 ; AVX-LABEL: shuffle_v16i8_zz_zz_19_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz_zz:
 ; AVX:       # BB#0:
-; AVX-NEXT:    vmovd %edi, %xmm0
-; AVX-NEXT:    vpslld $24, %xmm0, %xmm0
-; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = zero,zero,xmm0[3],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
+; AVX-NEXT:    vpxor %xmm0, %xmm0, %xmm0
+; AVX-NEXT:    vpinsrb $2, %edi, %xmm0, %xmm0
 ; AVX-NEXT:    retq
   %a = insertelement <16 x i8> undef, i8 %i, i32 3
   %shuffle = shufflevector <16 x i8> zeroinitializer, <16 x i8> %a, <16 x i32> <i32 0, i32 1, i32 19, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
@@ -1109,40 +1110,33 @@ define <16 x i8> @shuffle_v16i8_uu_10_02_07_22_14_07_02_18_03_01_14_18_09_11_00(
 ; SSE2:       # BB#0: # %entry
 ; SSE2-NEXT:    pxor %xmm2, %xmm2
 ; SSE2-NEXT:    movdqa %xmm0, %xmm3
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm3 = xmm3[0],xmm2[0],xmm3[1],xmm2[1],xmm3[2],xmm2[2],xmm3[3],xmm2[3],xmm3[4],xmm2[4],xmm3[5],xmm2[5],xmm3[6],xmm2[6],xmm3[7],xmm2[7]
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm0 = xmm0[8],xmm2[8],xmm0[9],xmm2[9],xmm0[10],xmm2[10],xmm0[11],xmm2[11],xmm0[12],xmm2[12],xmm0[13],xmm2[13],xmm0[14],xmm2[14],xmm0[15],xmm2[15]
-; SSE2-NEXT:    pshufd {{.*#+}} xmm2 = xmm0[2,3,0,1]
-; SSE2-NEXT:    pshufd {{.*#+}} xmm4 = xmm0[3,1,2,3]
-; SSE2-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm3[0],xmm0[1],xmm3[1],xmm0[2],xmm3[2],xmm0[3],xmm3[3]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[2,1,2,3,4,5,6,7]
-; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,3,2,3]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,0,2,1,4,5,6,7]
-; SSE2-NEXT:    pshufd {{.*#+}} xmm5 = xmm3[3,1,2,3]
-; SSE2-NEXT:    punpcklwd {{.*#+}} xmm3 = xmm3[0],xmm2[0],xmm3[1],xmm2[1],xmm3[2],xmm2[2],xmm3[3],xmm2[3]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm2 = xmm3[0,1,2,3,6,5,6,7]
-; SSE2-NEXT:    pshufd {{.*#+}} xmm2 = xmm2[2,1,2,3]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm2 = xmm2[0,0,2,1,4,5,6,7]
-; SSE2-NEXT:    punpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm0[0]
-; SSE2-NEXT:    movdqa {{.*#+}} xmm3 = [0,65535,65535,65535,0,65535,65535,65535]
-; SSE2-NEXT:    pand %xmm3, %xmm2
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm1
-; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm1[0,1,0,3]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,1,2,3,4,5,6,7]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,5,6,7]
-; SSE2-NEXT:    pandn %xmm0, %xmm3
-; SSE2-NEXT:    por %xmm2, %xmm3
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm5[1,1,2,3,4,5,6,7]
-; SSE2-NEXT:    punpcklwd {{.*#+}} xmm4 = xmm4[0],xmm0[0],xmm4[1],xmm0[1],xmm4[2],xmm0[2],xmm4[3],xmm0[3]
-; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm4[0,2,2,3]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[2,1,0,3,4,5,6,7]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm2 = xmm5[0,2,2,1,4,5,6,7]
-; SSE2-NEXT:    punpcklwd {{.*#+}} xmm2 = xmm2[0],xmm0[0],xmm2[1],xmm0[1],xmm2[2],xmm0[2],xmm2[3],xmm0[3]
-; SSE2-NEXT:    movdqa {{.*#+}} xmm0 = [65535,65535,65535,65535,0,65535,65535,65535]
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm3 = xmm3[8],xmm2[8],xmm3[9],xmm2[9],xmm3[10],xmm2[10],xmm3[11],xmm2[11],xmm3[12],xmm2[12],xmm3[13],xmm2[13],xmm3[14],xmm2[14],xmm3[15],xmm2[15]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm4 = xmm3[0,3,0,1]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm4 = xmm4[0,1,2,2,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,4,5,7,7]
+; SSE2-NEXT:    movdqa {{.*#+}} xmm5 = [65535,65535,65535,0,65535,0,0,65535]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm2[0],xmm0[1],xmm2[1],xmm0[2],xmm2[2],xmm0[3],xmm2[3],xmm0[4],xmm2[4],xmm0[5],xmm2[5],xmm0[6],xmm2[6],xmm0[7],xmm2[7]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm2 = xmm0[0,1,0,3]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm2 = xmm2[0,3,1,3,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,4,5,6,4]
+; SSE2-NEXT:    pand %xmm5, %xmm2
+; SSE2-NEXT:    pandn %xmm4, %xmm5
+; SSE2-NEXT:    por %xmm2, %xmm5
+; SSE2-NEXT:    psrlq $16, %xmm3
+; SSE2-NEXT:    pshufd {{.*#+}} xmm2 = xmm3[0,2,2,3]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[3,1,1,3]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,1,2,1,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,7,4]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[1,3,2,3]
+; SSE2-NEXT:    punpckldq {{.*#+}} xmm2 = xmm2[0],xmm0[0],xmm2[1],xmm0[1]
+; SSE2-NEXT:    packuswb %xmm5, %xmm2
+; SSE2-NEXT:    movdqa {{.*#+}} xmm0 = [255,255,255,255,0,255,255,255,0,255,255,255,0,255,255,255]
 ; SSE2-NEXT:    pand %xmm0, %xmm2
-; SSE2-NEXT:    pslldq {{.*#+}} xmm1 = zero,zero,xmm1[0,1,2,3,4,5,6,7,8,9,10,11,12,13]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[0,1,0,3]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm1 = xmm1[0,1,3,3,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,5,5,5,7]
 ; SSE2-NEXT:    pandn %xmm1, %xmm0
 ; SSE2-NEXT:    por %xmm2, %xmm0
-; SSE2-NEXT:    packuswb %xmm3, %xmm0
 ; SSE2-NEXT:    retq
 ;
 ; SSSE3-LABEL: shuffle_v16i8_uu_10_02_07_22_14_07_02_18_03_01_14_18_09_11_00:
@@ -1358,3 +1352,283 @@ define <16 x i8> @shuffle_v16i8_uu_02_03_zz_uu_06_07_zz_uu_10_11_zz_uu_14_15_zz(
   %shuffle = shufflevector <16 x i8> %a, <16 x i8> zeroinitializer, <16 x i32> <i32 undef, i32 2, i32 3, i32 16, i32 undef, i32 6, i32 7, i32 16, i32 undef, i32 10, i32 11, i32 16, i32 undef, i32 14, i32 15, i32 16>
   ret <16 x i8> %shuffle
 }
+
+define <16 x i8> @shuffle_v16i8_bitcast_unpack(<16 x i8> %a, <16 x i8> %b) {
+; SSE-LABEL: shuffle_v16i8_bitcast_unpack:
+; SSE:       # BB#0:
+; SSE-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]
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: shuffle_v16i8_bitcast_unpack:
+; AVX:       # BB#0:
+; AVX-NEXT:    vpunpcklbw {{.*#+}} 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]
+; AVX-NEXT:    retq
+  %shuffle8  = shufflevector <16 x i8> %a, <16 x i8> %b, <16 x i32> <i32 7, i32 23, i32 6, i32 22, i32 5, i32 21, i32 4, i32 20, i32 3, i32 19, i32 2, i32 18, i32 1, i32 17, i32 0, i32 16>
+  %bitcast32 = bitcast <16 x i8> %shuffle8 to <4 x float>
+  %shuffle32 = shufflevector <4 x float> %bitcast32, <4 x float> undef, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
+  %bitcast16 = bitcast <4 x float> %shuffle32 to <8 x i16>
+  %shuffle16 = shufflevector <8 x i16> %bitcast16, <8 x i16> undef, <8 x i32> <i32 1, i32 0, i32 3, i32 2, i32 5, i32 4, i32 7, i32 6>
+  %bitcast8  = bitcast <8 x i16> %shuffle16 to <16 x i8>
+  ret <16 x i8> %bitcast8
+}
+
+define <16 x i8> @insert_dup_mem_v16i8_i32(i32* %ptr) {
+; SSE2-LABEL: insert_dup_mem_v16i8_i32:
+; SSE2:       # BB#0:
+; SSE2-NEXT:    movd {{.*#+}} xmm0 = mem[0],zero,zero,zero
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,1,0,3]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,4,4,4]
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: insert_dup_mem_v16i8_i32:
+; SSSE3:       # BB#0:
+; SSSE3-NEXT:    movd {{.*#+}} xmm0 = mem[0],zero,zero,zero
+; SSSE3-NEXT:    pxor %xmm1, %xmm1
+; SSSE3-NEXT:    pshufb %xmm1, %xmm0
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: insert_dup_mem_v16i8_i32:
+; SSE41:       # BB#0:
+; SSE41-NEXT:    movd {{.*#+}} xmm0 = mem[0],zero,zero,zero
+; SSE41-NEXT:    pxor %xmm1, %xmm1
+; SSE41-NEXT:    pshufb %xmm1, %xmm0
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: insert_dup_mem_v16i8_i32:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    vmovd {{.*#+}} xmm0 = mem[0],zero,zero,zero
+; AVX1-NEXT:    vpxor %xmm1, %xmm1, %xmm1
+; AVX1-NEXT:    vpshufb %xmm1, %xmm0, %xmm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: insert_dup_mem_v16i8_i32:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vpbroadcastb (%rdi), %xmm0
+; AVX2-NEXT:    retq
+  %tmp = load i32, i32* %ptr, align 4
+  %tmp1 = insertelement <4 x i32> zeroinitializer, i32 %tmp, i32 0
+  %tmp2 = bitcast <4 x i32> %tmp1 to <16 x i8>
+  %tmp3 = shufflevector <16 x i8> %tmp2, <16 x i8> undef, <16 x i32> zeroinitializer
+  ret <16 x i8> %tmp3
+}
+
+define <16 x i8> @insert_dup_mem_v16i8_sext_i8(i8* %ptr) {
+; SSE2-LABEL: insert_dup_mem_v16i8_sext_i8:
+; SSE2:       # BB#0:
+; SSE2-NEXT:    movsbl (%rdi), %eax
+; SSE2-NEXT:    movd %eax, %xmm0
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,1,0,3]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,4,4,4]
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: insert_dup_mem_v16i8_sext_i8:
+; SSSE3:       # BB#0:
+; SSSE3-NEXT:    movsbl (%rdi), %eax
+; SSSE3-NEXT:    movd %eax, %xmm0
+; SSSE3-NEXT:    pxor %xmm1, %xmm1
+; SSSE3-NEXT:    pshufb %xmm1, %xmm0
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: insert_dup_mem_v16i8_sext_i8:
+; SSE41:       # BB#0:
+; SSE41-NEXT:    movsbl (%rdi), %eax
+; SSE41-NEXT:    movd %eax, %xmm0
+; SSE41-NEXT:    pxor %xmm1, %xmm1
+; SSE41-NEXT:    pshufb %xmm1, %xmm0
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: insert_dup_mem_v16i8_sext_i8:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    movsbl (%rdi), %eax
+; AVX1-NEXT:    vmovd %eax, %xmm0
+; AVX1-NEXT:    vpxor %xmm1, %xmm1, %xmm1
+; AVX1-NEXT:    vpshufb %xmm1, %xmm0, %xmm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: insert_dup_mem_v16i8_sext_i8:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vpbroadcastb (%rdi), %xmm0
+; AVX2-NEXT:    retq
+  %tmp = load i8, i8* %ptr, align 1
+  %tmp1 = sext i8 %tmp to i32
+  %tmp2 = insertelement <4 x i32> zeroinitializer, i32 %tmp1, i32 0
+  %tmp3 = bitcast <4 x i32> %tmp2 to <16 x i8>
+  %tmp4 = shufflevector <16 x i8> %tmp3, <16 x i8> undef, <16 x i32> zeroinitializer
+  ret <16 x i8> %tmp4
+}
+
+define <16 x i8> @insert_dup_elt1_mem_v16i8_i32(i32* %ptr) {
+; SSE2-LABEL: insert_dup_elt1_mem_v16i8_i32:
+; SSE2:       # BB#0:
+; SSE2-NEXT:    movd {{.*#+}} xmm0 = mem[0],zero,zero,zero
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,1,0,3]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,1,1,1,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,5,5,5]
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: insert_dup_elt1_mem_v16i8_i32:
+; SSSE3:       # BB#0:
+; SSSE3-NEXT:    movd {{.*#+}} xmm0 = mem[0],zero,zero,zero
+; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: insert_dup_elt1_mem_v16i8_i32:
+; SSE41:       # BB#0:
+; SSE41-NEXT:    movd {{.*#+}} xmm0 = mem[0],zero,zero,zero
+; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: insert_dup_elt1_mem_v16i8_i32:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    vmovd {{.*#+}} xmm0 = mem[0],zero,zero,zero
+; AVX1-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: insert_dup_elt1_mem_v16i8_i32:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vpbroadcastb 1(%rdi), %xmm0
+; AVX2-NEXT:    retq
+  %tmp = load i32, i32* %ptr, align 4
+  %tmp1 = insertelement <4 x i32> zeroinitializer, i32 %tmp, i32 0
+  %tmp2 = bitcast <4 x i32> %tmp1 to <16 x i8>
+  %tmp3 = shufflevector <16 x i8> %tmp2, <16 x i8> undef, <16 x i32> <i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1>
+  ret <16 x i8> %tmp3
+}
+
+define <16 x i8> @insert_dup_elt2_mem_v16i8_i32(i32* %ptr) {
+; SSE2-LABEL: insert_dup_elt2_mem_v16i8_i32:
+; SSE2:       # BB#0:
+; SSE2-NEXT:    movd {{.*#+}} xmm0 = mem[0],zero,zero,zero
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,1,2,1]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[2,2,2,2,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,6,6,6]
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: insert_dup_elt2_mem_v16i8_i32:
+; SSSE3:       # BB#0:
+; SSSE3-NEXT:    movd {{.*#+}} xmm0 = mem[0],zero,zero,zero
+; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2]
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: insert_dup_elt2_mem_v16i8_i32:
+; SSE41:       # BB#0:
+; SSE41-NEXT:    movd {{.*#+}} xmm0 = mem[0],zero,zero,zero
+; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2]
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: insert_dup_elt2_mem_v16i8_i32:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    vmovd {{.*#+}} xmm0 = mem[0],zero,zero,zero
+; AVX1-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2]
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: insert_dup_elt2_mem_v16i8_i32:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vpbroadcastb 2(%rdi), %xmm0
+; AVX2-NEXT:    retq
+  %tmp = load i32, i32* %ptr, align 4
+  %tmp1 = insertelement <4 x i32> zeroinitializer, i32 %tmp, i32 0
+  %tmp2 = bitcast <4 x i32> %tmp1 to <16 x i8>
+  %tmp3 = shufflevector <16 x i8> %tmp2, <16 x i8> undef, <16 x i32> <i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2>
+  ret <16 x i8> %tmp3
+}
+
+define <16 x i8> @insert_dup_elt1_mem_v16i8_sext_i8(i8* %ptr) {
+; SSE2-LABEL: insert_dup_elt1_mem_v16i8_sext_i8:
+; SSE2:       # BB#0:
+; SSE2-NEXT:    movsbl (%rdi), %eax
+; SSE2-NEXT:    movd %eax, %xmm0
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,1,0,3]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,1,1,1,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,5,5,5]
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: insert_dup_elt1_mem_v16i8_sext_i8:
+; SSSE3:       # BB#0:
+; SSSE3-NEXT:    movsbl (%rdi), %eax
+; SSSE3-NEXT:    movd %eax, %xmm0
+; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: insert_dup_elt1_mem_v16i8_sext_i8:
+; SSE41:       # BB#0:
+; SSE41-NEXT:    movsbl (%rdi), %eax
+; SSE41-NEXT:    movd %eax, %xmm0
+; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: insert_dup_elt1_mem_v16i8_sext_i8:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    movsbl (%rdi), %eax
+; AVX1-NEXT:    vmovd %eax, %xmm0
+; AVX1-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: insert_dup_elt1_mem_v16i8_sext_i8:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    movsbl (%rdi), %eax
+; AVX2-NEXT:    shrl $8, %eax
+; AVX2-NEXT:    vmovd %eax, %xmm0
+; AVX2-NEXT:    vpbroadcastb %xmm0, %xmm0
+; AVX2-NEXT:    retq
+  %tmp = load i8, i8* %ptr, align 1
+  %tmp1 = sext i8 %tmp to i32
+  %tmp2 = insertelement <4 x i32> zeroinitializer, i32 %tmp1, i32 0
+  %tmp3 = bitcast <4 x i32> %tmp2 to <16 x i8>
+  %tmp4 = shufflevector <16 x i8> %tmp3, <16 x i8> undef, <16 x i32> <i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1>
+  ret <16 x i8> %tmp4
+}
+
+define <16 x i8> @insert_dup_elt2_mem_v16i8_sext_i8(i8* %ptr) {
+; SSE2-LABEL: insert_dup_elt2_mem_v16i8_sext_i8:
+; SSE2:       # BB#0:
+; SSE2-NEXT:    movsbl (%rdi), %eax
+; SSE2-NEXT:    movd %eax, %xmm0
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,1,2,1]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[2,2,2,2,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,6,6,6]
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: insert_dup_elt2_mem_v16i8_sext_i8:
+; SSSE3:       # BB#0:
+; SSSE3-NEXT:    movsbl (%rdi), %eax
+; SSSE3-NEXT:    movd %eax, %xmm0
+; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2]
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: insert_dup_elt2_mem_v16i8_sext_i8:
+; SSE41:       # BB#0:
+; SSE41-NEXT:    movsbl (%rdi), %eax
+; SSE41-NEXT:    movd %eax, %xmm0
+; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2]
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: insert_dup_elt2_mem_v16i8_sext_i8:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    movsbl (%rdi), %eax
+; AVX1-NEXT:    vmovd %eax, %xmm0
+; AVX1-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2]
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: insert_dup_elt2_mem_v16i8_sext_i8:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    movsbl (%rdi), %eax
+; AVX2-NEXT:    shrl $16, %eax
+; AVX2-NEXT:    vmovd %eax, %xmm0
+; AVX2-NEXT:    vpbroadcastb %xmm0, %xmm0
+; AVX2-NEXT:    retq
+  %tmp = load i8, i8* %ptr, align 1
+  %tmp1 = sext i8 %tmp to i32
+  %tmp2 = insertelement <4 x i32> zeroinitializer, i32 %tmp1, i32 0
+  %tmp3 = bitcast <4 x i32> %tmp2 to <16 x i8>
+  %tmp4 = shufflevector <16 x i8> %tmp3, <16 x i8> undef, <16 x i32> <i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2>
+  ret <16 x i8> %tmp4
+}