; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=x86_64-unknown-unknown | FileCheck %s --check-prefixes=SSE,SSE2
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+ssse3 | FileCheck %s --check-prefixes=SSE,SSSE3
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse4.1 | FileCheck %s --check-prefixes=SSE,SSE41
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefixes=AVX,AVX1
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefixes=AVX,AVX2OR512VL,AVX2,AVX2-SLOW
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2,+fast-variable-crosslane-shuffle,+fast-variable-perlane-shuffle | FileCheck %s --check-prefixes=AVX,AVX2OR512VL,AVX2,AVX2-FAST
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2,+fast-variable-perlane-shuffle | FileCheck %s --check-prefixes=AVX,AVX2OR512VL,AVX2,AVX2-FAST
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512vl,+avx512bw | FileCheck %s --check-prefixes=AVX,AVX2OR512VL,AVX512VL,AVX512VL-SLOW
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512vl,+avx512bw,+fast-variable-crosslane-shuffle,+fast-variable-perlane-shuffle | FileCheck %s --check-prefixes=AVX,AVX2OR512VL,AVX512VL,AVX512VL-FAST
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512vl,+avx512bw,+fast-variable-perlane-shuffle | FileCheck %s --check-prefixes=AVX,AVX2OR512VL,AVX512VL,AVX512VL-FAST
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+xop,+avx | FileCheck %s --check-prefixes=AVX,XOP,XOPAVX1
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+xop,+avx2 | FileCheck %s --check-prefixes=AVX,XOP,XOPAVX2

define <8 x i16> @shuffle_v8i16_01012323(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_01012323:
; SSE:       # %bb.0:
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,0,1,1]
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_01012323:
; AVX:       # %bb.0:
; AVX-NEXT:    vpermilps {{.*#+}} xmm0 = xmm0[0,0,1,1]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 1, i32 0, i32 1, i32 2, i32 3, i32 2, i32 3>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_67452301(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_67452301:
; SSE:       # %bb.0:
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[3,2,1,0]
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_67452301:
; AVX:       # %bb.0:
; AVX-NEXT:    vpermilps {{.*#+}} xmm0 = xmm0[3,2,1,0]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 6, i32 7, i32 4, i32 5, i32 2, i32 3, i32 0, i32 1>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_456789AB(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_456789AB:
; SSE2:       # %bb.0:
; SSE2-NEXT:    shufps {{.*#+}} xmm0 = xmm0[2,3],xmm1[0,1]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_456789AB:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    palignr {{.*#+}} xmm1 = xmm0[8,9,10,11,12,13,14,15],xmm1[0,1,2,3,4,5,6,7]
; SSSE3-NEXT:    movdqa %xmm1, %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_456789AB:
; SSE41:       # %bb.0:
; SSE41-NEXT:    palignr {{.*#+}} xmm1 = xmm0[8,9,10,11,12,13,14,15],xmm1[0,1,2,3,4,5,6,7]
; SSE41-NEXT:    movdqa %xmm1, %xmm0
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_456789AB:
; AVX:       # %bb.0:
; AVX-NEXT:    vpalignr {{.*#+}} xmm0 = xmm0[8,9,10,11,12,13,14,15],xmm1[0,1,2,3,4,5,6,7]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_00000000(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_00000000:
; SSE:       # %bb.0:
; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; SSE-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_00000000:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; AVX1-NEXT:    retq
;
; AVX2OR512VL-LABEL: shuffle_v8i16_00000000:
; AVX2OR512VL:       # %bb.0:
; AVX2OR512VL-NEXT:    vpbroadcastw %xmm0, %xmm0
; AVX2OR512VL-NEXT:    retq
;
; XOPAVX1-LABEL: shuffle_v8i16_00000000:
; XOPAVX1:       # %bb.0:
; XOPAVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
; XOPAVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; XOPAVX1-NEXT:    retq
;
; XOPAVX2-LABEL: shuffle_v8i16_00000000:
; XOPAVX2:       # %bb.0:
; XOPAVX2-NEXT:    vpbroadcastw %xmm0, %xmm0
; XOPAVX2-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_00004444(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_00004444:
; SSE:       # %bb.0:
; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
; SSE-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,4,4,4]
; SSE-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_00004444:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
; AVX1-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,4,4,4]
; AVX1-NEXT:    retq
;
; AVX2-SLOW-LABEL: shuffle_v8i16_00004444:
; AVX2-SLOW:       # %bb.0:
; AVX2-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
; AVX2-SLOW-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,4,4,4]
; AVX2-SLOW-NEXT:    retq
;
; AVX2-FAST-LABEL: shuffle_v8i16_00004444:
; AVX2-FAST:       # %bb.0:
; AVX2-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,0,1,0,1,0,1,8,9,8,9,8,9,8,9]
; AVX2-FAST-NEXT:    retq
;
; AVX512VL-SLOW-LABEL: shuffle_v8i16_00004444:
; AVX512VL-SLOW:       # %bb.0:
; AVX512VL-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
; AVX512VL-SLOW-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,4,4,4]
; AVX512VL-SLOW-NEXT:    retq
;
; AVX512VL-FAST-LABEL: shuffle_v8i16_00004444:
; AVX512VL-FAST:       # %bb.0:
; AVX512VL-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,0,1,0,1,0,1,8,9,8,9,8,9,8,9]
; AVX512VL-FAST-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_00004444:
; XOP:       # %bb.0:
; XOP-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
; XOP-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,4,4,4]
; XOP-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 0, i32 0, i32 0, i32 4, i32 4, i32 4, i32 4>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_u0u1u2u3(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_u0u1u2u3:
; SSE:       # %bb.0:
; SSE-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3]
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_u0u1u2u3:
; AVX:       # %bb.0:
; AVX-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 undef, i32 0, i32 undef, i32 1, i32 undef, i32 2, i32 undef, i32 3>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_u4u5u6u7(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_u4u5u6u7:
; SSE:       # %bb.0:
; SSE-NEXT:    punpckhwd {{.*#+}} xmm0 = xmm0[4,4,5,5,6,6,7,7]
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_u4u5u6u7:
; AVX:       # %bb.0:
; AVX-NEXT:    vpunpckhwd {{.*#+}} xmm0 = xmm0[4,4,5,5,6,6,7,7]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 undef, i32 4, i32 undef, i32 5, i32 undef, i32 6, i32 undef, i32 7>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_31206745(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_31206745:
; SSE:       # %bb.0:
; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[3,1,2,0,4,5,6,7]
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,1,3,2]
; SSE-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_31206745:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[3,1,2,0,4,5,6,7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,1,3,2]
; AVX1-NEXT:    retq
;
; AVX2-SLOW-LABEL: shuffle_v8i16_31206745:
; AVX2-SLOW:       # %bb.0:
; AVX2-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[3,1,2,0,4,5,6,7]
; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,1,3,2]
; AVX2-SLOW-NEXT:    retq
;
; AVX2-FAST-LABEL: shuffle_v8i16_31206745:
; AVX2-FAST:       # %bb.0:
; AVX2-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[6,7,2,3,4,5,0,1,12,13,14,15,8,9,10,11]
; AVX2-FAST-NEXT:    retq
;
; AVX512VL-SLOW-LABEL: shuffle_v8i16_31206745:
; AVX512VL-SLOW:       # %bb.0:
; AVX512VL-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[3,1,2,0,4,5,6,7]
; AVX512VL-SLOW-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,1,3,2]
; AVX512VL-SLOW-NEXT:    retq
;
; AVX512VL-FAST-LABEL: shuffle_v8i16_31206745:
; AVX512VL-FAST:       # %bb.0:
; AVX512VL-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[6,7,2,3,4,5,0,1,12,13,14,15,8,9,10,11]
; AVX512VL-FAST-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_31206745:
; XOP:       # %bb.0:
; XOP-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[3,1,2,0,4,5,6,7]
; XOP-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,1,3,2]
; XOP-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 3, i32 1, i32 2, i32 0, i32 6, i32 7, i32 4, i32 5>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_44440000(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_44440000:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,1,0,3]
; 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: shuffle_v8i16_44440000:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[8,9,8,9,8,9,8,9,0,1,0,1,0,1,0,1]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_44440000:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[8,9,8,9,8,9,8,9,0,1,0,1,0,1,0,1]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_44440000:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[8,9,8,9,8,9,8,9,0,1,0,1,0,1,0,1]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 4, i32 4, i32 4, i32 4, i32 0, i32 0, i32 0, i32 0>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_23016745(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_23016745:
; SSE:       # %bb.0:
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[1,0,3,2]
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_23016745:
; AVX:       # %bb.0:
; AVX-NEXT:    vpermilps {{.*#+}} xmm0 = xmm0[1,0,3,2]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 2, i32 3, i32 0, i32 1, i32 6, i32 7, i32 4, i32 5>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_23026745(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_23026745:
; SSE:       # %bb.0:
; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[2,3,0,2,4,5,6,7]
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,1,3,2]
; SSE-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_23026745:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[2,3,0,2,4,5,6,7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,1,3,2]
; AVX1-NEXT:    retq
;
; AVX2-SLOW-LABEL: shuffle_v8i16_23026745:
; AVX2-SLOW:       # %bb.0:
; AVX2-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[2,3,0,2,4,5,6,7]
; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,1,3,2]
; AVX2-SLOW-NEXT:    retq
;
; AVX2-FAST-LABEL: shuffle_v8i16_23026745:
; AVX2-FAST:       # %bb.0:
; AVX2-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[4,5,6,7,0,1,4,5,12,13,14,15,8,9,10,11]
; AVX2-FAST-NEXT:    retq
;
; AVX512VL-SLOW-LABEL: shuffle_v8i16_23026745:
; AVX512VL-SLOW:       # %bb.0:
; AVX512VL-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[2,3,0,2,4,5,6,7]
; AVX512VL-SLOW-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,1,3,2]
; AVX512VL-SLOW-NEXT:    retq
;
; AVX512VL-FAST-LABEL: shuffle_v8i16_23026745:
; AVX512VL-FAST:       # %bb.0:
; AVX512VL-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[4,5,6,7,0,1,4,5,12,13,14,15,8,9,10,11]
; AVX512VL-FAST-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_23026745:
; XOP:       # %bb.0:
; XOP-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[2,3,0,2,4,5,6,7]
; XOP-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,1,3,2]
; XOP-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 2, i32 3, i32 0, i32 2, i32 6, i32 7, i32 4, i32 5>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_23016747(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_23016747:
; SSE:       # %bb.0:
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[1,0,2,3]
; SSE-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,7,4,7]
; SSE-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_23016747:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[1,0,2,3]
; AVX1-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,7,4,7]
; AVX1-NEXT:    retq
;
; AVX2-SLOW-LABEL: shuffle_v8i16_23016747:
; AVX2-SLOW:       # %bb.0:
; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[1,0,2,3]
; AVX2-SLOW-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,7,4,7]
; AVX2-SLOW-NEXT:    retq
;
; AVX2-FAST-LABEL: shuffle_v8i16_23016747:
; AVX2-FAST:       # %bb.0:
; AVX2-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[4,5,6,7,0,1,2,3,12,13,14,15,8,9,14,15]
; AVX2-FAST-NEXT:    retq
;
; AVX512VL-SLOW-LABEL: shuffle_v8i16_23016747:
; AVX512VL-SLOW:       # %bb.0:
; AVX512VL-SLOW-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[1,0,2,3]
; AVX512VL-SLOW-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,7,4,7]
; AVX512VL-SLOW-NEXT:    retq
;
; AVX512VL-FAST-LABEL: shuffle_v8i16_23016747:
; AVX512VL-FAST:       # %bb.0:
; AVX512VL-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[4,5,6,7,0,1,2,3,12,13,14,15,8,9,14,15]
; AVX512VL-FAST-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_23016747:
; XOP:       # %bb.0:
; XOP-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[1,0,2,3]
; XOP-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,7,4,7]
; XOP-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 2, i32 3, i32 0, i32 1, i32 6, i32 7, i32 4, i32 7>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_75643120(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_75643120:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[3,1,2,0,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,5,6,4]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_75643120:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[14,15,10,11,12,13,8,9,6,7,2,3,4,5,0,1]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_75643120:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[14,15,10,11,12,13,8,9,6,7,2,3,4,5,0,1]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_75643120:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[14,15,10,11,12,13,8,9,6,7,2,3,4,5,0,1]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 7, i32 5, i32 6, i32 4, i32 3, i32 1, i32 2, i32 0>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_10545410(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_10545410:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,0]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,0,3,2,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,4,7,6]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_10545410:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[2,3,0,1,10,11,8,9,10,11,8,9,2,3,0,1]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_10545410:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[2,3,0,1,10,11,8,9,10,11,8,9,2,3,0,1]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_10545410:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[2,3,0,1,10,11,8,9,10,11,8,9,2,3,0,1]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 1, i32 0, i32 5, i32 4, i32 5, i32 4, i32 1, i32 0>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_54105410(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_54105410:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,0]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[3,2,1,0,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,4,7,6]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_54105410:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[10,11,8,9,2,3,0,1,10,11,8,9,2,3,0,1]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_54105410:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[10,11,8,9,2,3,0,1,10,11,8,9,2,3,0,1]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_54105410:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[10,11,8,9,2,3,0,1,10,11,8,9,2,3,0,1]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 5, i32 4, i32 1, i32 0, i32 5, i32 4, i32 1, i32 0>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_54101054(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_54101054:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,0]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[3,2,1,0,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,6,5,4]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_54101054:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[10,11,8,9,2,3,0,1,2,3,0,1,10,11,8,9]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_54101054:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[10,11,8,9,2,3,0,1,2,3,0,1,10,11,8,9]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_54101054:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[10,11,8,9,2,3,0,1,2,3,0,1,10,11,8,9]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 5, i32 4, i32 1, i32 0, i32 1, i32 0, i32 5, i32 4>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_04400440(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_04400440:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,0]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,2,0,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,4,4,6]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_04400440:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,8,9,0,1,0,1,8,9,8,9,0,1]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_04400440:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,8,9,0,1,0,1,8,9,8,9,0,1]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_04400440:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,8,9,0,1,0,1,8,9,8,9,0,1]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 4, i32 4, i32 0, i32 0, i32 4, i32 4, i32 0>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_40044004(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_40044004:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,0]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[2,0,0,2,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,6,6,4]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_40044004:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[8,9,0,1,0,1,8,9,8,9,0,1,0,1,8,9]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_40044004:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[8,9,0,1,0,1,8,9,8,9,0,1,0,1,8,9]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_40044004:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[8,9,0,1,0,1,8,9,8,9,0,1,0,1,8,9]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 4, i32 0, i32 0, i32 4, i32 4, i32 0, i32 0, i32 4>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_26405173(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_26405173:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,1,3,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,5,6,4]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,3,2,1]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,2,3,0,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,6,4,7]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_26405173:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[4,5,12,13,8,9,0,1,10,11,2,3,14,15,6,7]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_26405173:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[4,5,12,13,8,9,0,1,10,11,2,3,14,15,6,7]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_26405173:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[4,5,12,13,8,9,0,1,10,11,2,3,14,15,6,7]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 2, i32 6, i32 4, i32 0, i32 5, i32 1, i32 7, i32 3>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_20645173(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_20645173:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,1,3,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,5,6,4]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,3,2,1]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,0,2,3,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,6,4,7]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_20645173:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[4,5,0,1,12,13,8,9,10,11,2,3,14,15,6,7]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_20645173:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[4,5,0,1,12,13,8,9,10,11,2,3,14,15,6,7]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_20645173:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[4,5,0,1,12,13,8,9,10,11,2,3,14,15,6,7]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 2, i32 0, i32 6, i32 4, i32 5, i32 1, i32 7, i32 3>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_26401375(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_26401375:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,1,3,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,5,6,4]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,3,1,2]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,2,3,0,4,5,6,7]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_26401375:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[4,5,12,13,8,9,0,1,2,3,6,7,14,15,10,11]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_26401375:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[4,5,12,13,8,9,0,1,2,3,6,7,14,15,10,11]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_26401375:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[4,5,12,13,8,9,0,1,2,3,6,7,14,15,10,11]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 2, i32 6, i32 4, i32 0, i32 1, i32 3, i32 7, i32 5>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_66751643(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_66751643:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[3,1,2,3,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,6,5,7]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,3,2,0]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,1,3,2,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,5,4,6]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_66751643:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[12,13,12,13,14,15,10,11,2,3,12,13,8,9,6,7]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_66751643:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[12,13,12,13,14,15,10,11,2,3,12,13,8,9,6,7]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_66751643:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[12,13,12,13,14,15,10,11,2,3,12,13,8,9,6,7]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 6, i32 6, i32 7, i32 5, i32 1, i32 6, i32 4, i32 3>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_60514754(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_60514754:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,5,4,7]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[2,0,3,1,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,7,5,6]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_60514754:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[12,13,0,1,10,11,2,3,8,9,14,15,10,11,8,9]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_60514754:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[12,13,0,1,10,11,2,3,8,9,14,15,10,11,8,9]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_60514754:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[12,13,0,1,10,11,2,3,8,9,14,15,10,11,8,9]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> undef, <8 x i32> <i32 6, i32 0, i32 5, i32 1, i32 4, i32 7, i32 5, i32 4>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_00444444(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_00444444:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,0,2,2,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,4,4,4]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_00444444:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,0,1,8,9,8,9,8,9,8,9,8,9,8,9]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_00444444:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,0,1,8,9,8,9,8,9,8,9,8,9,8,9]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_00444444:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,0,1,8,9,8,9,8,9,8,9,8,9,8,9]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 0, i32 4, i32 4, i32 4, i32 4, i32 4, i32 4>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_44004444(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_44004444:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[2,2,0,0,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,4,4,4]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_44004444:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[8,9,8,9,0,1,0,1,8,9,8,9,8,9,8,9]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_44004444:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[8,9,8,9,0,1,0,1,8,9,8,9,8,9,8,9]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_44004444:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[8,9,8,9,0,1,0,1,8,9,8,9,8,9,8,9]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 4, i32 4, i32 0, i32 0, i32 4, i32 4, i32 4, i32 4>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_04404444(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_04404444:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,2,0,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,4,4,4]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_04404444:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,8,9,0,1,8,9,8,9,8,9,8,9]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_04404444:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,8,9,0,1,8,9,8,9,8,9,8,9]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_04404444:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,8,9,0,1,8,9,8,9,8,9,8,9]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 4, i32 4, i32 0, i32 4, i32 4, i32 4, i32 4>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_04400000(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_04400000:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,0,3]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,2,0,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,4,4,4]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_04400000:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,8,9,0,1,0,1,0,1,0,1,0,1]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_04400000:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,8,9,0,1,0,1,0,1,0,1,0,1]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_04400000:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,8,9,0,1,0,1,0,1,0,1,0,1]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 4, i32 4, i32 0, i32 0, i32 0, i32 0, i32 0>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_04404567(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_04404567:
; SSE:       # %bb.0:
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,2,0,4,5,6,7]
; SSE-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_04404567:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,2,2,0,4,5,6,7]
; AVX1-NEXT:    retq
;
; AVX2-SLOW-LABEL: shuffle_v8i16_04404567:
; AVX2-SLOW:       # %bb.0:
; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
; AVX2-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,2,2,0,4,5,6,7]
; AVX2-SLOW-NEXT:    retq
;
; AVX2-FAST-LABEL: shuffle_v8i16_04404567:
; AVX2-FAST:       # %bb.0:
; AVX2-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,8,9,0,1,8,9,10,11,12,13,14,15]
; AVX2-FAST-NEXT:    retq
;
; AVX512VL-SLOW-LABEL: shuffle_v8i16_04404567:
; AVX512VL-SLOW:       # %bb.0:
; AVX512VL-SLOW-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
; AVX512VL-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,2,2,0,4,5,6,7]
; AVX512VL-SLOW-NEXT:    retq
;
; AVX512VL-FAST-LABEL: shuffle_v8i16_04404567:
; AVX512VL-FAST:       # %bb.0:
; AVX512VL-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,8,9,0,1,8,9,10,11,12,13,14,15]
; AVX512VL-FAST-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_04404567:
; XOP:       # %bb.0:
; XOP-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
; XOP-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,2,2,0,4,5,6,7]
; XOP-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 4, i32 4, i32 0, i32 4, i32 5, i32 6, i32 7>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_0X444444(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_0X444444:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,1,2,2,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,4,4,4]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_0X444444:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,8,9,8,9,8,9,8,9,8,9,8,9]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_0X444444:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,8,9,8,9,8,9,8,9,8,9,8,9]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_0X444444:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,8,9,8,9,8,9,8,9,8,9,8,9]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 undef, i32 4, i32 4, i32 4, i32 4, i32 4, i32 4>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_44X04444(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_44X04444:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[2,2,2,0,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,4,4,4]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_44X04444:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[8,9,8,9,8,9,0,1,8,9,8,9,8,9,8,9]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_44X04444:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[8,9,8,9,8,9,0,1,8,9,8,9,8,9,8,9]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_44X04444:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[8,9,8,9,8,9,0,1,8,9,8,9,8,9,8,9]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 4, i32 4, i32 undef, i32 0, i32 4, i32 4, i32 4, i32 4>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_X4404444(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_X4404444:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,2,0,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,4,4,4]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_X4404444:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,8,9,0,1,8,9,8,9,8,9,8,9]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_X4404444:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,8,9,0,1,8,9,8,9,8,9,8,9]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_X4404444:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,8,9,0,1,8,9,8,9,8,9,8,9]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 undef, i32 4, i32 4, i32 0, i32 4, i32 4, i32 4, i32 4>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_0127XXXX(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_0127XXXX:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,1,3]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,7,6,7]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_0127XXXX:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,14,15,4,5,14,15,12,13,14,15]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_0127XXXX:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,14,15,4,5,14,15,12,13,14,15]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_0127XXXX:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,14,15,4,5,14,15,12,13,14,15]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 1, i32 2, i32 7, i32 undef, i32 undef, i32 undef, i32 undef>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_XXXX4563(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_XXXX4563:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[3,1,2,0]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,3,2,3,4,5,6,7]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,1,2,0]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_XXXX4563:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[12,13,6,7,4,5,6,7,8,9,10,11,12,13,6,7]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_XXXX4563:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[12,13,6,7,4,5,6,7,8,9,10,11,12,13,6,7]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_XXXX4563:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[12,13,6,7,4,5,6,7,8,9,10,11,12,13,6,7]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 undef, i32 undef, i32 undef, i32 undef, i32 4, i32 5, i32 6, i32 3>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_4563XXXX(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_4563XXXX:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[3,1,2,0]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,3,2,3,4,5,6,7]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,0,2,3]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_4563XXXX:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[8,9,10,11,12,13,6,7,8,9,10,11,0,1,2,3]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_4563XXXX:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[8,9,10,11,12,13,6,7,8,9,10,11,0,1,2,3]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_4563XXXX:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[8,9,10,11,12,13,6,7,8,9,10,11,0,1,2,3]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 4, i32 5, i32 6, i32 3, i32 undef, i32 undef, i32 undef, i32 undef>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_01274563(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_01274563:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,1,3]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,5,4,7]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,3,1,2]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_01274563:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,14,15,8,9,10,11,12,13,6,7]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_01274563:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,14,15,8,9,10,11,12,13,6,7]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_01274563:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,14,15,8,9,10,11,12,13,6,7]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 1, i32 2, i32 7, i32 4, i32 5, i32 6, i32 3>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_45630127(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_45630127:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[3,1,2,0]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,3,2,1,4,5,6,7]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,0,3,1]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_45630127:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[8,9,10,11,12,13,6,7,0,1,2,3,4,5,14,15]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_45630127:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[8,9,10,11,12,13,6,7,0,1,2,3,4,5,14,15]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_45630127:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[8,9,10,11,12,13,6,7,0,1,2,3,4,5,14,15]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 4, i32 5, i32 6, i32 3, i32 0, i32 1, i32 2, i32 7>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_37102735(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_37102735:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,6,5,7]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,1,3]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,5,6,4]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[3,2,1,0,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,4,5,6]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_37102735:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[6,7,14,15,2,3,0,1,4,5,14,15,6,7,10,11]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_37102735:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[6,7,14,15,2,3,0,1,4,5,14,15,6,7,10,11]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_37102735:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[6,7,14,15,2,3,0,1,4,5,14,15,6,7,10,11]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 3, i32 7, i32 1, i32 0, i32 2, i32 7, i32 3, i32 5>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_08192a3b(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_08192a3b:
; SSE:       # %bb.0:
; SSE-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_08192a3b:
; AVX:       # %bb.0:
; AVX-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 8, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_0c1d2e3f(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_0c1d2e3f:
; SSE:       # %bb.0:
; SSE-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[2,3,2,3]
; SSE-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_0c1d2e3f:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[2,3,2,3]
; AVX-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 12, i32 1, i32 13, i32 2, i32 14, i32 3, i32 15>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_4c5d6e7f(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_4c5d6e7f:
; SSE:       # %bb.0:
; SSE-NEXT:    punpckhwd {{.*#+}} xmm0 = xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_4c5d6e7f:
; AVX:       # %bb.0:
; AVX-NEXT:    vpunpckhwd {{.*#+}} xmm0 = xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 4, i32 12, i32 5, i32 13, i32 6, i32 14, i32 7, i32 15>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_48596a7b(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_48596a7b:
; SSE:       # %bb.0:
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,3,2,3]
; SSE-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_48596a7b:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,3,2,3]
; AVX-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 4, i32 8, i32 5, i32 9, i32 6, i32 10, i32 7, i32 11>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_08196e7f(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_08196e7f:
; SSE:       # %bb.0:
; SSE-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[0,3,2,3]
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,3,2,3]
; SSE-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; SSE-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_08196e7f:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[0,3,2,3]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,3,2,3]
; AVX1-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; AVX1-NEXT:    retq
;
; AVX2OR512VL-LABEL: shuffle_v8i16_08196e7f:
; AVX2OR512VL:       # %bb.0:
; AVX2OR512VL-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[0,3,2,3]
; AVX2OR512VL-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,3,2,3]
; AVX2OR512VL-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; AVX2OR512VL-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_08196e7f:
; XOP:       # %bb.0:
; XOP-NEXT:    vpperm {{.*#+}} xmm0 = xmm0[0,1],xmm1[0,1],xmm0[2,3],xmm1[2,3],xmm0[12,13],xmm1[12,13],xmm0[14,15],xmm1[14,15]
; XOP-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 8, i32 1, i32 9, i32 6, i32 14, i32 7, i32 15>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_0c1d6879(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_0c1d6879:
; SSE:       # %bb.0:
; SSE-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[2,0,2,3]
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,3,2,3]
; SSE-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; SSE-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_0c1d6879:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[2,0,2,3]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,3,2,3]
; AVX1-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; AVX1-NEXT:    retq
;
; AVX2OR512VL-LABEL: shuffle_v8i16_0c1d6879:
; AVX2OR512VL:       # %bb.0:
; AVX2OR512VL-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[2,0,2,3]
; AVX2OR512VL-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,3,2,3]
; AVX2OR512VL-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; AVX2OR512VL-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_0c1d6879:
; XOP:       # %bb.0:
; XOP-NEXT:    vpperm {{.*#+}} xmm0 = xmm0[0,1],xmm1[8,9],xmm0[2,3],xmm1[10,11],xmm0[12,13],xmm1[0,1],xmm0[14,15],xmm1[2,3]
; XOP-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 12, i32 1, i32 13, i32 6, i32 8, i32 7, i32 9>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_109832ba(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_109832ba:
; SSE:       # %bb.0:
; SSE-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[2,0,3,1,4,5,6,7]
; SSE-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,4,7,5]
; SSE-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_109832ba:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[2,0,3,1,4,5,6,7]
; AVX1-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,4,7,5]
; AVX1-NEXT:    retq
;
; AVX2-SLOW-LABEL: shuffle_v8i16_109832ba:
; AVX2-SLOW:       # %bb.0:
; AVX2-SLOW-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; AVX2-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[2,0,3,1,4,5,6,7]
; AVX2-SLOW-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,4,7,5]
; AVX2-SLOW-NEXT:    retq
;
; AVX2-FAST-LABEL: shuffle_v8i16_109832ba:
; AVX2-FAST:       # %bb.0:
; AVX2-FAST-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; AVX2-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[4,5,0,1,6,7,2,3,12,13,8,9,14,15,10,11]
; AVX2-FAST-NEXT:    retq
;
; AVX512VL-SLOW-LABEL: shuffle_v8i16_109832ba:
; AVX512VL-SLOW:       # %bb.0:
; AVX512VL-SLOW-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; AVX512VL-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[2,0,3,1,4,5,6,7]
; AVX512VL-SLOW-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,4,7,5]
; AVX512VL-SLOW-NEXT:    retq
;
; AVX512VL-FAST-LABEL: shuffle_v8i16_109832ba:
; AVX512VL-FAST:       # %bb.0:
; AVX512VL-FAST-NEXT:    vmovdqa {{.*#+}} xmm2 = [1,0,9,8,3,2,11,10]
; AVX512VL-FAST-NEXT:    vpermt2w %xmm1, %xmm2, %xmm0
; AVX512VL-FAST-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_109832ba:
; XOP:       # %bb.0:
; XOP-NEXT:    vpperm {{.*#+}} xmm0 = xmm0[2,3,0,1],xmm1[2,3,0,1],xmm0[6,7,4,5],xmm1[6,7,4,5]
; XOP-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 1, i32 0, i32 9, i32 8, i32 3, i32 2, i32 11, i32 10>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_8091a2b3(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_8091a2b3:
; SSE:       # %bb.0:
; SSE-NEXT:    punpcklwd {{.*#+}} xmm1 = xmm1[0],xmm0[0],xmm1[1],xmm0[1],xmm1[2],xmm0[2],xmm1[3],xmm0[3]
; SSE-NEXT:    movdqa %xmm1, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_8091a2b3:
; AVX:       # %bb.0:
; AVX-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm1[0],xmm0[0],xmm1[1],xmm0[1],xmm1[2],xmm0[2],xmm1[3],xmm0[3]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 8, i32 0, i32 9, i32 1, i32 10, i32 2, i32 11, i32 3>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_c4d5e6f7(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_c4d5e6f7:
; SSE:       # %bb.0:
; SSE-NEXT:    punpckhwd {{.*#+}} xmm1 = xmm1[4],xmm0[4],xmm1[5],xmm0[5],xmm1[6],xmm0[6],xmm1[7],xmm0[7]
; SSE-NEXT:    movdqa %xmm1, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_c4d5e6f7:
; AVX:       # %bb.0:
; AVX-NEXT:    vpunpckhwd {{.*#+}} xmm0 = xmm1[4],xmm0[4],xmm1[5],xmm0[5],xmm1[6],xmm0[6],xmm1[7],xmm0[7]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 12, i32 4, i32 13, i32 5, i32 14, i32 6, i32 15, i32 7>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_0213cedf(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_0213cedf:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshuflw {{.*#+}} xmm2 = xmm0[0,2,1,3,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm1[0,1,2,3,4,6,5,7]
; SSE2-NEXT:    movsd {{.*#+}} xmm0 = xmm2[0],xmm0[1]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_0213cedf:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshuflw {{.*#+}} xmm2 = xmm0[0,2,1,3,4,5,6,7]
; SSSE3-NEXT:    pshufhw {{.*#+}} xmm0 = xmm1[0,1,2,3,4,6,5,7]
; SSSE3-NEXT:    movsd {{.*#+}} xmm0 = xmm2[0],xmm0[1]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_0213cedf:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,4,6,5,7]
; SSE41-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,1,3,4,5,6,7]
; SSE41-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm1[4,5,6,7]
; SSE41-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_0213cedf:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,4,6,5,7]
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,2,1,3,4,5,6,7]
; AVX1-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm1[4,5,6,7]
; AVX1-NEXT:    retq
;
; AVX2-SLOW-LABEL: shuffle_v8i16_0213cedf:
; AVX2-SLOW:       # %bb.0:
; AVX2-SLOW-NEXT:    vpshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,4,6,5,7]
; AVX2-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,2,1,3,4,5,6,7]
; AVX2-SLOW-NEXT:    vpblendd {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
; AVX2-SLOW-NEXT:    retq
;
; AVX2-FAST-LABEL: shuffle_v8i16_0213cedf:
; AVX2-FAST:       # %bb.0:
; AVX2-FAST-NEXT:    vpshufb {{.*#+}} xmm1 = xmm1[8,9,12,13,10,11,14,15,u,u,u,u,u,u,u,u]
; AVX2-FAST-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,2,1,3,4,5,6,7]
; AVX2-FAST-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
; AVX2-FAST-NEXT:    retq
;
; AVX512VL-SLOW-LABEL: shuffle_v8i16_0213cedf:
; AVX512VL-SLOW:       # %bb.0:
; AVX512VL-SLOW-NEXT:    vpshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,4,6,5,7]
; AVX512VL-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,2,1,3,4,5,6,7]
; AVX512VL-SLOW-NEXT:    vpblendd {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
; AVX512VL-SLOW-NEXT:    retq
;
; AVX512VL-FAST-LABEL: shuffle_v8i16_0213cedf:
; AVX512VL-FAST:       # %bb.0:
; AVX512VL-FAST-NEXT:    vmovdqa {{.*#+}} xmm2 = [0,2,1,3,12,14,13,15]
; AVX512VL-FAST-NEXT:    vpermt2w %xmm1, %xmm2, %xmm0
; AVX512VL-FAST-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_0213cedf:
; XOP:       # %bb.0:
; XOP-NEXT:    vpperm {{.*#+}} xmm0 = xmm0[0,1,4,5,2,3,6,7],xmm1[8,9,12,13,10,11,14,15]
; XOP-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 2, i32 1, i32 3, i32 12, i32 14, i32 13, i32 15>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_443aXXXX(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_443aXXXX:
; SSE2:       # %bb.0:
; SSE2-NEXT:    movdqa {{.*#+}} xmm2 = [65535,65535,0,65535,65535,65535,65535,65535]
; SSE2-NEXT:    pand %xmm2, %xmm0
; SSE2-NEXT:    pandn %xmm1, %xmm2
; SSE2-NEXT:    por %xmm0, %xmm2
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm2[2,1,2,3]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,0,3,2,4,5,6,7]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_443aXXXX:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,xmm1[4,5,u,u,u,u,u,u,u,u]
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[8,9,8,9,6,7],zero,zero,xmm0[u,u,u,u,u,u,u,u]
; SSSE3-NEXT:    por %xmm1, %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_443aXXXX:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0,1],xmm1[2],xmm0[3,4,5,6,7]
; SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,1,2,3]
; SSE41-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,0,3,2,4,5,6,7]
; SSE41-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_443aXXXX:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm1[2],xmm0[3,4,5,6,7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,1,2,3]
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,3,2,4,5,6,7]
; AVX1-NEXT:    retq
;
; AVX2-SLOW-LABEL: shuffle_v8i16_443aXXXX:
; AVX2-SLOW:       # %bb.0:
; AVX2-SLOW-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm1[2],xmm0[3,4,5,6,7]
; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,1,2,3]
; AVX2-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,3,2,4,5,6,7]
; AVX2-SLOW-NEXT:    retq
;
; AVX2-FAST-LABEL: shuffle_v8i16_443aXXXX:
; AVX2-FAST:       # %bb.0:
; AVX2-FAST-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm1[2],xmm0[3,4,5,6,7]
; AVX2-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[8,9,8,9,6,7,4,5,8,9,10,11,12,13,14,15]
; AVX2-FAST-NEXT:    retq
;
; AVX512VL-SLOW-LABEL: shuffle_v8i16_443aXXXX:
; AVX512VL-SLOW:       # %bb.0:
; AVX512VL-SLOW-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm1[2],xmm0[3,4,5,6,7]
; AVX512VL-SLOW-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,1,2,3]
; AVX512VL-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,3,2,4,5,6,7]
; AVX512VL-SLOW-NEXT:    retq
;
; AVX512VL-FAST-LABEL: shuffle_v8i16_443aXXXX:
; AVX512VL-FAST:       # %bb.0:
; AVX512VL-FAST-NEXT:    vmovdqa {{.*#+}} xmm2 = [4,4,3,10,4,5,6,7]
; AVX512VL-FAST-NEXT:    vpermt2w %xmm1, %xmm2, %xmm0
; AVX512VL-FAST-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_443aXXXX:
; XOP:       # %bb.0:
; XOP-NEXT:    vpperm {{.*#+}} xmm0 = xmm0[8,9,8,9,6,7],xmm1[4,5],xmm0[8,9,10,11,12,13,14,15]
; XOP-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 4, i32 4, i32 3, i32 10, i32 undef, i32 undef, i32 undef, i32 undef>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_032dXXXX(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_032dXXXX:
; SSE2:       # %bb.0:
; SSE2-NEXT:    shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
; SSE2-NEXT:    shufps {{.*#+}} xmm0 = xmm0[0,1,2,0]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,5,6,7]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,1,2,3]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,3,2,1,4,5,6,7]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_032dXXXX:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,xmm1[10,11,u,u,u,u,u,u,u,u]
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,6,7,4,5],zero,zero,xmm0[u,u,u,u,u,u,u,u]
; SSSE3-NEXT:    por %xmm1, %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_032dXXXX:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm1[4,5,6,7]
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,6,7,4,5,10,11,0,1,10,11,0,1,2,3]
; SSE41-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_032dXXXX:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm1[4,5,6,7]
; AVX1-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,6,7,4,5,10,11,0,1,10,11,0,1,2,3]
; AVX1-NEXT:    retq
;
; AVX2-LABEL: shuffle_v8i16_032dXXXX:
; AVX2:       # %bb.0:
; AVX2-NEXT:    vpblendd {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
; AVX2-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,6,7,4,5,10,11,0,1,10,11,0,1,2,3]
; AVX2-NEXT:    retq
;
; AVX512VL-LABEL: shuffle_v8i16_032dXXXX:
; AVX512VL:       # %bb.0:
; AVX512VL-NEXT:    vmovdqa {{.*#+}} xmm2 = [0,3,2,13,0,13,0,1]
; AVX512VL-NEXT:    vpermt2w %xmm1, %xmm2, %xmm0
; AVX512VL-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_032dXXXX:
; XOP:       # %bb.0:
; XOP-NEXT:    vpperm {{.*#+}} xmm0 = xmm0[0,1,6,7,4,5],xmm1[10,11],xmm0[0,1],xmm1[10,11],xmm0[0,1,2,3]
; XOP-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 3, i32 2, i32 13, i32 undef, i32 undef, i32 undef, i32 undef>
  ret <8 x i16> %shuffle
}
define <8 x i16> @shuffle_v8i16_XXXdXXXX(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_XXXdXXXX:
; SSE:       # %bb.0:
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm1[2,2,3,3]
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_XXXdXXXX:
; AVX:       # %bb.0:
; AVX-NEXT:    vpermilps {{.*#+}} xmm0 = xmm1[2,2,3,3]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 undef, i32 undef, i32 undef, i32 13, i32 undef, i32 undef, i32 undef, i32 undef>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_012dXXXX(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_012dXXXX:
; SSE2:       # %bb.0:
; SSE2-NEXT:    movdqa {{.*#+}} xmm2 = [65535,65535,65535,0,65535,65535,65535,65535]
; SSE2-NEXT:    pand %xmm2, %xmm0
; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[2,2,3,3]
; SSE2-NEXT:    pandn %xmm1, %xmm2
; SSE2-NEXT:    por %xmm2, %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_012dXXXX:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,xmm1[10,11,u,u,u,u,u,u,u,u]
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5],zero,zero,xmm0[u,u,u,u,u,u,u,u]
; SSSE3-NEXT:    por %xmm1, %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_012dXXXX:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[2,2,3,3]
; SSE41-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0,1,2],xmm1[3],xmm0[4,5,6,7]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_012dXXXX:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[2,2,3,3]
; AVX-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0,1,2],xmm1[3],xmm0[4,5,6,7]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 1, i32 2, i32 13, i32 undef, i32 undef, i32 undef, i32 undef>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_XXXXcde3(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_XXXXcde3:
; SSE2:       # %bb.0:
; SSE2-NEXT:    movdqa {{.*#+}} xmm2 = [65535,65535,65535,65535,65535,65535,65535,0]
; SSE2-NEXT:    pand %xmm2, %xmm1
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,1,0,1]
; SSE2-NEXT:    pandn %xmm0, %xmm2
; SSE2-NEXT:    por %xmm1, %xmm2
; SSE2-NEXT:    movdqa %xmm2, %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_XXXXcde3:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[u,u,u,u,u,u,u,u],zero,zero,zero,zero,zero,zero,xmm0[6,7]
; SSSE3-NEXT:    pshufb {{.*#+}} xmm1 = xmm1[u,u,u,u,u,u,u,u,8,9,10,11,12,13],zero,zero
; SSSE3-NEXT:    por %xmm1, %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_XXXXcde3:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,1,0,1]
; SSE41-NEXT:    pblendw {{.*#+}} xmm0 = xmm1[0,1,2,3,4,5,6],xmm0[7]
; SSE41-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_XXXXcde3:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,1,0,1]
; AVX1-NEXT:    vpblendw {{.*#+}} xmm0 = xmm1[0,1,2,3,4,5,6],xmm0[7]
; AVX1-NEXT:    retq
;
; AVX2OR512VL-LABEL: shuffle_v8i16_XXXXcde3:
; AVX2OR512VL:       # %bb.0:
; AVX2OR512VL-NEXT:    vpbroadcastq %xmm0, %xmm0
; AVX2OR512VL-NEXT:    vpblendw {{.*#+}} xmm0 = xmm1[0,1,2,3,4,5,6],xmm0[7]
; AVX2OR512VL-NEXT:    retq
;
; XOPAVX1-LABEL: shuffle_v8i16_XXXXcde3:
; XOPAVX1:       # %bb.0:
; XOPAVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,1,0,1]
; XOPAVX1-NEXT:    vpblendw {{.*#+}} xmm0 = xmm1[0,1,2,3,4,5,6],xmm0[7]
; XOPAVX1-NEXT:    retq
;
; XOPAVX2-LABEL: shuffle_v8i16_XXXXcde3:
; XOPAVX2:       # %bb.0:
; XOPAVX2-NEXT:    vpbroadcastq %xmm0, %xmm0
; XOPAVX2-NEXT:    vpblendw {{.*#+}} xmm0 = xmm1[0,1,2,3,4,5,6],xmm0[7]
; XOPAVX2-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 undef, i32 undef, i32 undef, i32 undef, i32 12, i32 13, i32 14, i32 3>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_cde3XXXX(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_cde3XXXX:
; SSE2:       # %bb.0:
; SSE2-NEXT:    movdqa {{.*#+}} xmm2 = [65535,65535,65535,0,65535,65535,65535,65535]
; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[2,3,2,3]
; SSE2-NEXT:    pand %xmm2, %xmm1
; SSE2-NEXT:    pandn %xmm0, %xmm2
; SSE2-NEXT:    por %xmm1, %xmm2
; SSE2-NEXT:    movdqa %xmm2, %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_cde3XXXX:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,xmm0[6,7,u,u,u,u,u,u,u,u]
; SSSE3-NEXT:    pshufb {{.*#+}} xmm1 = xmm1[8,9,10,11,12,13],zero,zero,xmm1[u,u,u,u,u,u,u,u]
; SSSE3-NEXT:    por %xmm1, %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_cde3XXXX:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[2,3,2,3]
; SSE41-NEXT:    pblendw {{.*#+}} xmm0 = xmm1[0,1,2],xmm0[3],xmm1[4,5,6,7]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_cde3XXXX:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[2,3,2,3]
; AVX-NEXT:    vpblendw {{.*#+}} xmm0 = xmm1[0,1,2],xmm0[3],xmm1[4,5,6,7]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 12, i32 13, i32 14, i32 3, i32 undef, i32 undef, i32 undef, i32 undef>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_012dcde3(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_012dcde3:
; SSE2:       # %bb.0:
; SSE2-NEXT:    shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
; SSE2-NEXT:    shufps {{.*#+}} xmm0 = xmm0[0,3,2,1]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[3,1,2,0,4,5,6,7]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[3,1,2,0]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,3,2,1,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,5,7]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,3,2,1]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,3,0,2,4,5,6,7]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_012dcde3:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,xmm1[10,11,8,9,10,11,12,13],zero,zero
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5],zero,zero,zero,zero,zero,zero,zero,zero,xmm0[6,7]
; SSSE3-NEXT:    por %xmm1, %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_012dcde3:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm1[4,5,6,7]
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,10,11,8,9,10,11,12,13,6,7]
; SSE41-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_012dcde3:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm1[4,5,6,7]
; AVX1-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,10,11,8,9,10,11,12,13,6,7]
; AVX1-NEXT:    retq
;
; AVX2-LABEL: shuffle_v8i16_012dcde3:
; AVX2:       # %bb.0:
; AVX2-NEXT:    vpblendd {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
; AVX2-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,10,11,8,9,10,11,12,13,6,7]
; AVX2-NEXT:    retq
;
; AVX512VL-LABEL: shuffle_v8i16_012dcde3:
; AVX512VL:       # %bb.0:
; AVX512VL-NEXT:    vmovdqa {{.*#+}} xmm2 = [0,1,2,13,12,13,14,3]
; AVX512VL-NEXT:    vpermt2w %xmm1, %xmm2, %xmm0
; AVX512VL-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_012dcde3:
; XOP:       # %bb.0:
; XOP-NEXT:    vpperm {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5],xmm1[10,11,8,9,10,11,12,13],xmm0[6,7]
; XOP-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 1, i32 2, i32 13, i32 12, i32 13, i32 14, i32 3>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_0923cde7(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_0923cde7:
; SSE2:       # %bb.0:
; SSE2-NEXT:    movaps {{.*#+}} xmm2 = [65535,0,65535,65535,0,0,0,65535]
; SSE2-NEXT:    andps %xmm2, %xmm0
; SSE2-NEXT:    andnps %xmm1, %xmm2
; SSE2-NEXT:    orps %xmm2, %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_0923cde7:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    movaps {{.*#+}} xmm2 = [65535,0,65535,65535,0,0,0,65535]
; SSSE3-NEXT:    andps %xmm2, %xmm0
; SSSE3-NEXT:    andnps %xmm1, %xmm2
; SSSE3-NEXT:    orps %xmm2, %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_0923cde7:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1],xmm0[2,3],xmm1[4,5,6],xmm0[7]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_0923cde7:
; AVX:       # %bb.0:
; AVX-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1],xmm0[2,3],xmm1[4,5,6],xmm0[7]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 9, i32 2, i32 3, i32 12, i32 13, i32 14, i32 7>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_XXX1X579(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_XXX1X579:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm2 = xmm1[0,0,0,0]
; SSE2-NEXT:    movdqa {{.*#+}} xmm1 = [65535,65535,65535,65535,65535,65535,65535,0]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,1,1,1,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,7,7]
; SSE2-NEXT:    pand %xmm1, %xmm0
; SSE2-NEXT:    pandn %xmm2, %xmm1
; SSE2-NEXT:    por %xmm0, %xmm1
; SSE2-NEXT:    movdqa %xmm1, %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_XXX1X579:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm1 = xmm1[u,u,u,u,u,u],zero,zero,xmm1[u,u],zero,zero,zero,zero,xmm1[2,3]
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[u,u,u,u,u,u,2,3,u,u,10,11,14,15],zero,zero
; SSSE3-NEXT:    por %xmm1, %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_XXX1X579:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[0,0,0,0]
; SSE41-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,1,1,1,4,5,6,7]
; SSE41-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,7,7]
; SSE41-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6],xmm1[7]
; SSE41-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_XXX1X579:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[0,0,0,0]
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[1,1,1,1,4,5,6,7]
; AVX1-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,7,7]
; AVX1-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6],xmm1[7]
; AVX1-NEXT:    retq
;
; AVX2-SLOW-LABEL: shuffle_v8i16_XXX1X579:
; AVX2-SLOW:       # %bb.0:
; AVX2-SLOW-NEXT:    vpbroadcastd %xmm1, %xmm1
; AVX2-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[1,1,1,1,4,5,6,7]
; AVX2-SLOW-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,7,7]
; AVX2-SLOW-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6],xmm1[7]
; AVX2-SLOW-NEXT:    retq
;
; AVX2-FAST-LABEL: shuffle_v8i16_XXX1X579:
; AVX2-FAST:       # %bb.0:
; AVX2-FAST-NEXT:    vpbroadcastd %xmm1, %xmm1
; AVX2-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[2,3,2,3,2,3,2,3,8,9,10,11,14,15,u,u]
; AVX2-FAST-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6],xmm1[7]
; AVX2-FAST-NEXT:    retq
;
; AVX512VL-SLOW-LABEL: shuffle_v8i16_XXX1X579:
; AVX512VL-SLOW:       # %bb.0:
; AVX512VL-SLOW-NEXT:    vpbroadcastd %xmm1, %xmm1
; AVX512VL-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[1,1,1,1,4,5,6,7]
; AVX512VL-SLOW-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,7,7]
; AVX512VL-SLOW-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,5,6],xmm1[7]
; AVX512VL-SLOW-NEXT:    retq
;
; AVX512VL-FAST-LABEL: shuffle_v8i16_XXX1X579:
; AVX512VL-FAST:       # %bb.0:
; AVX512VL-FAST-NEXT:    vmovdqa {{.*#+}} xmm2 = [1,1,1,1,4,5,7,9]
; AVX512VL-FAST-NEXT:    vpermt2w %xmm1, %xmm2, %xmm0
; AVX512VL-FAST-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_XXX1X579:
; XOP:       # %bb.0:
; XOP-NEXT:    vpperm {{.*#+}} xmm0 = xmm0[2,3,2,3,2,3,2,3,8,9,10,11,14,15],xmm1[2,3]
; XOP-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 undef, i32 undef, i32 undef, i32 1, i32 undef, i32 5, i32 7, i32 9>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_XX4X8acX(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_XX4X8acX:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshuflw {{.*#+}} xmm1 = xmm1[0,2,2,3,4,5,6,7]
; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[0,1,2,0]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,6,7,4,7]
; SSE2-NEXT:    shufps {{.*#+}} xmm0 = xmm0[2,2],xmm1[2,3]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_XX4X8acX:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[u,u,u,u,8,9,u,u],zero,zero,zero,zero,zero,zero,xmm0[u,u]
; SSSE3-NEXT:    pshufb {{.*#+}} xmm1 = xmm1[u,u,u,u],zero,zero,xmm1[u,u,0,1,4,5,8,9,u,u]
; SSSE3-NEXT:    por %xmm1, %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_XX4X8acX:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm1 = xmm1[u,u,u,u,u,u,u,u,0,1,4,5,8,9,4,5]
; SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,2,3,3]
; SSE41-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm1[4,5,6,7]
; SSE41-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_XX4X8acX:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpshufb {{.*#+}} xmm1 = xmm1[u,u,u,u,u,u,u,u,0,1,4,5,8,9,4,5]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,2,3,3]
; AVX1-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm1[4,5,6,7]
; AVX1-NEXT:    retq
;
; AVX2-LABEL: shuffle_v8i16_XX4X8acX:
; AVX2:       # %bb.0:
; AVX2-NEXT:    vpshufb {{.*#+}} xmm1 = xmm1[u,u,u,u,u,u,u,u,0,1,4,5,8,9,4,5]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,2,3,3]
; AVX2-NEXT:    vpblendd {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
; AVX2-NEXT:    retq
;
; AVX512VL-LABEL: shuffle_v8i16_XX4X8acX:
; AVX512VL:       # %bb.0:
; AVX512VL-NEXT:    vmovdqa {{.*#+}} xmm2 = [4,5,4,5,8,10,12,10]
; AVX512VL-NEXT:    vpermt2w %xmm1, %xmm2, %xmm0
; AVX512VL-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_XX4X8acX:
; XOP:       # %bb.0:
; XOP-NEXT:    vpperm {{.*#+}} xmm0 = xmm0[8,9,10,11,8,9,10,11],xmm1[0,1,4,5,8,9,4,5]
; XOP-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 undef, i32 undef, i32 4, i32 undef, i32 8, i32 10, i32 12, i32 undef>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_8zzzzzzz(i16 %i) {
; SSE-LABEL: shuffle_v8i16_8zzzzzzz:
; SSE:       # %bb.0:
; SSE-NEXT:    movzwl %di, %eax
; SSE-NEXT:    movd %eax, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_8zzzzzzz:
; AVX:       # %bb.0:
; AVX-NEXT:    movzwl %di, %eax
; AVX-NEXT:    vmovd %eax, %xmm0
; AVX-NEXT:    retq
  %a = insertelement <8 x i16> undef, i16 %i, i32 0
  %shuffle = shufflevector <8 x i16> zeroinitializer, <8 x i16> %a, <8 x i32> <i32 8, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_z8zzzzzz(i16 %i) {
; SSE-LABEL: shuffle_v8i16_z8zzzzzz:
; SSE:       # %bb.0:
; SSE-NEXT:    pxor %xmm0, %xmm0
; SSE-NEXT:    pinsrw $1, %edi, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_z8zzzzzz:
; AVX:       # %bb.0:
; AVX-NEXT:    vpxor %xmm0, %xmm0, %xmm0
; AVX-NEXT:    vpinsrw $1, %edi, %xmm0, %xmm0
; AVX-NEXT:    retq
  %a = insertelement <8 x i16> undef, i16 %i, i32 0
  %shuffle = shufflevector <8 x i16> zeroinitializer, <8 x i16> %a, <8 x i32> <i32 2, i32 8, i32 3, i32 7, i32 6, i32 5, i32 4, i32 3>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_zzzzz8zz(i16 %i) {
; SSE-LABEL: shuffle_v8i16_zzzzz8zz:
; SSE:       # %bb.0:
; SSE-NEXT:    pxor %xmm0, %xmm0
; SSE-NEXT:    pinsrw $5, %edi, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_zzzzz8zz:
; AVX:       # %bb.0:
; AVX-NEXT:    vpxor %xmm0, %xmm0, %xmm0
; AVX-NEXT:    vpinsrw $5, %edi, %xmm0, %xmm0
; AVX-NEXT:    retq
  %a = insertelement <8 x i16> undef, i16 %i, i32 0
  %shuffle = shufflevector <8 x i16> zeroinitializer, <8 x i16> %a, <8 x i32> <i32 0, i32 0, i32 0, i32 0, i32 0, i32 8, i32 0, i32 0>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_zuuzuuz8(i16 %i) {
; SSE-LABEL: shuffle_v8i16_zuuzuuz8:
; SSE:       # %bb.0:
; SSE-NEXT:    pxor %xmm0, %xmm0
; SSE-NEXT:    pinsrw $7, %edi, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_zuuzuuz8:
; AVX:       # %bb.0:
; AVX-NEXT:    vpxor %xmm0, %xmm0, %xmm0
; AVX-NEXT:    vpinsrw $7, %edi, %xmm0, %xmm0
; AVX-NEXT:    retq
  %a = insertelement <8 x i16> undef, i16 %i, i32 0
  %shuffle = shufflevector <8 x i16> zeroinitializer, <8 x i16> %a, <8 x i32> <i32 0, i32 undef, i32 undef, i32 3, i32 undef, i32 undef, i32 6, i32 8>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_zzBzzzzz(i16 %i) {
; SSE-LABEL: shuffle_v8i16_zzBzzzzz:
; SSE:       # %bb.0:
; SSE-NEXT:    pxor %xmm0, %xmm0
; SSE-NEXT:    pinsrw $2, %edi, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_zzBzzzzz:
; AVX:       # %bb.0:
; AVX-NEXT:    vpxor %xmm0, %xmm0, %xmm0
; AVX-NEXT:    vpinsrw $2, %edi, %xmm0, %xmm0
; AVX-NEXT:    retq
  %a = insertelement <8 x i16> undef, i16 %i, i32 3
  %shuffle = shufflevector <8 x i16> zeroinitializer, <8 x i16> %a, <8 x i32> <i32 0, i32 1, i32 11, i32 3, i32 4, i32 5, i32 6, i32 7>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_def01234(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_def01234:
; SSE2:       # %bb.0:
; SSE2-NEXT:    psrldq {{.*#+}} xmm1 = xmm1[10,11,12,13,14,15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; SSE2-NEXT:    pslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5,6,7,8,9]
; SSE2-NEXT:    por %xmm1, %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_def01234:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    palignr {{.*#+}} xmm0 = xmm1[10,11,12,13,14,15],xmm0[0,1,2,3,4,5,6,7,8,9]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_def01234:
; SSE41:       # %bb.0:
; SSE41-NEXT:    palignr {{.*#+}} xmm0 = xmm1[10,11,12,13,14,15],xmm0[0,1,2,3,4,5,6,7,8,9]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_def01234:
; AVX:       # %bb.0:
; AVX-NEXT:    vpalignr {{.*#+}} xmm0 = xmm1[10,11,12,13,14,15],xmm0[0,1,2,3,4,5,6,7,8,9]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 13, i32 14, i32 15, i32 0, i32 1, i32 2, i32 3, i32 4>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_ueuu123u(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_ueuu123u:
; SSE2:       # %bb.0:
; SSE2-NEXT:    psrldq {{.*#+}} xmm1 = xmm1[10,11,12,13,14,15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; SSE2-NEXT:    pslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5,6,7,8,9]
; SSE2-NEXT:    por %xmm1, %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_ueuu123u:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    palignr {{.*#+}} xmm0 = xmm1[10,11,12,13,14,15],xmm0[0,1,2,3,4,5,6,7,8,9]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_ueuu123u:
; SSE41:       # %bb.0:
; SSE41-NEXT:    palignr {{.*#+}} xmm0 = xmm1[10,11,12,13,14,15],xmm0[0,1,2,3,4,5,6,7,8,9]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_ueuu123u:
; AVX:       # %bb.0:
; AVX-NEXT:    vpalignr {{.*#+}} xmm0 = xmm1[10,11,12,13,14,15],xmm0[0,1,2,3,4,5,6,7,8,9]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 undef, i32 14, i32 undef, i32 undef, i32 1, i32 2, i32 3, i32 undef>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_56701234(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_56701234:
; SSE2:       # %bb.0:
; SSE2-NEXT:    movdqa %xmm0, %xmm1
; SSE2-NEXT:    psrldq {{.*#+}} xmm1 = xmm1[10,11,12,13,14,15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; SSE2-NEXT:    pslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5,6,7,8,9]
; SSE2-NEXT:    por %xmm1, %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_56701234:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    palignr {{.*#+}} xmm0 = xmm0[10,11,12,13,14,15,0,1,2,3,4,5,6,7,8,9]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_56701234:
; SSE41:       # %bb.0:
; SSE41-NEXT:    palignr {{.*#+}} xmm0 = xmm0[10,11,12,13,14,15,0,1,2,3,4,5,6,7,8,9]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_56701234:
; AVX:       # %bb.0:
; AVX-NEXT:    vpalignr {{.*#+}} xmm0 = xmm0[10,11,12,13,14,15,0,1,2,3,4,5,6,7,8,9]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3, i32 4>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_u6uu123u(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_u6uu123u:
; SSE2:       # %bb.0:
; SSE2-NEXT:    movdqa %xmm0, %xmm1
; SSE2-NEXT:    psrldq {{.*#+}} xmm1 = xmm1[10,11,12,13,14,15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; SSE2-NEXT:    pslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5,6,7,8,9]
; SSE2-NEXT:    por %xmm1, %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_u6uu123u:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    palignr {{.*#+}} xmm0 = xmm0[10,11,12,13,14,15,0,1,2,3,4,5,6,7,8,9]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_u6uu123u:
; SSE41:       # %bb.0:
; SSE41-NEXT:    palignr {{.*#+}} xmm0 = xmm0[10,11,12,13,14,15,0,1,2,3,4,5,6,7,8,9]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_u6uu123u:
; AVX:       # %bb.0:
; AVX-NEXT:    vpalignr {{.*#+}} xmm0 = xmm0[10,11,12,13,14,15,0,1,2,3,4,5,6,7,8,9]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 undef, i32 6, i32 undef, i32 undef, i32 1, i32 2, i32 3, i32 undef>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_uuuu123u(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_uuuu123u:
; SSE:       # %bb.0:
; SSE-NEXT:    pslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5,6,7,8,9]
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_uuuu123u:
; AVX:       # %bb.0:
; AVX-NEXT:    vpslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5,6,7,8,9]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 undef, i32 undef, i32 undef, i32 undef, i32 1, i32 2, i32 3, i32 undef>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_bcdef012(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_bcdef012:
; SSE2:       # %bb.0:
; SSE2-NEXT:    psrldq {{.*#+}} xmm1 = xmm1[6,7,8,9,10,11,12,13,14,15],zero,zero,zero,zero,zero,zero
; SSE2-NEXT:    pslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5]
; SSE2-NEXT:    por %xmm1, %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_bcdef012:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    palignr {{.*#+}} xmm0 = xmm1[6,7,8,9,10,11,12,13,14,15],xmm0[0,1,2,3,4,5]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_bcdef012:
; SSE41:       # %bb.0:
; SSE41-NEXT:    palignr {{.*#+}} xmm0 = xmm1[6,7,8,9,10,11,12,13,14,15],xmm0[0,1,2,3,4,5]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_bcdef012:
; AVX:       # %bb.0:
; AVX-NEXT:    vpalignr {{.*#+}} xmm0 = xmm1[6,7,8,9,10,11,12,13,14,15],xmm0[0,1,2,3,4,5]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 11, i32 12, i32 13, i32 14, i32 15, i32 0, i32 1, i32 2>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_ucdeuu1u(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_ucdeuu1u:
; SSE2:       # %bb.0:
; SSE2-NEXT:    psrldq {{.*#+}} xmm1 = xmm1[6,7,8,9,10,11,12,13,14,15],zero,zero,zero,zero,zero,zero
; SSE2-NEXT:    pslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5]
; SSE2-NEXT:    por %xmm1, %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_ucdeuu1u:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    palignr {{.*#+}} xmm0 = xmm1[6,7,8,9,10,11,12,13,14,15],xmm0[0,1,2,3,4,5]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_ucdeuu1u:
; SSE41:       # %bb.0:
; SSE41-NEXT:    palignr {{.*#+}} xmm0 = xmm1[6,7,8,9,10,11,12,13,14,15],xmm0[0,1,2,3,4,5]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_ucdeuu1u:
; AVX:       # %bb.0:
; AVX-NEXT:    vpalignr {{.*#+}} xmm0 = xmm1[6,7,8,9,10,11,12,13,14,15],xmm0[0,1,2,3,4,5]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 undef, i32 12, i32 13, i32 14, i32 undef, i32 undef, i32 1, i32 undef>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_34567012(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_34567012:
; SSE2:       # %bb.0:
; SSE2-NEXT:    movdqa %xmm0, %xmm1
; SSE2-NEXT:    psrldq {{.*#+}} xmm1 = xmm1[6,7,8,9,10,11,12,13,14,15],zero,zero,zero,zero,zero,zero
; SSE2-NEXT:    pslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5]
; SSE2-NEXT:    por %xmm1, %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_34567012:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    palignr {{.*#+}} xmm0 = xmm0[6,7,8,9,10,11,12,13,14,15,0,1,2,3,4,5]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_34567012:
; SSE41:       # %bb.0:
; SSE41-NEXT:    palignr {{.*#+}} xmm0 = xmm0[6,7,8,9,10,11,12,13,14,15,0,1,2,3,4,5]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_34567012:
; AVX:       # %bb.0:
; AVX-NEXT:    vpalignr {{.*#+}} xmm0 = xmm0[6,7,8,9,10,11,12,13,14,15,0,1,2,3,4,5]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 3, i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_u456uu1u(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_u456uu1u:
; SSE2:       # %bb.0:
; SSE2-NEXT:    movdqa %xmm0, %xmm1
; SSE2-NEXT:    psrldq {{.*#+}} xmm1 = xmm1[6,7,8,9,10,11,12,13,14,15],zero,zero,zero,zero,zero,zero
; SSE2-NEXT:    pslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5]
; SSE2-NEXT:    por %xmm1, %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_u456uu1u:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    palignr {{.*#+}} xmm0 = xmm0[6,7,8,9,10,11,12,13,14,15,0,1,2,3,4,5]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_u456uu1u:
; SSE41:       # %bb.0:
; SSE41-NEXT:    palignr {{.*#+}} xmm0 = xmm0[6,7,8,9,10,11,12,13,14,15,0,1,2,3,4,5]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_u456uu1u:
; AVX:       # %bb.0:
; AVX-NEXT:    vpalignr {{.*#+}} xmm0 = xmm0[6,7,8,9,10,11,12,13,14,15,0,1,2,3,4,5]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 undef, i32 4, i32 5, i32 6, i32 undef, i32 undef, i32 1, i32 undef>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_u456uuuu(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_u456uuuu:
; SSE:       # %bb.0:
; SSE-NEXT:    psrldq {{.*#+}} xmm0 = xmm0[6,7,8,9,10,11,12,13,14,15],zero,zero,zero,zero,zero,zero
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_u456uuuu:
; AVX:       # %bb.0:
; AVX-NEXT:    vpsrldq {{.*#+}} xmm0 = xmm0[6,7,8,9,10,11,12,13,14,15],zero,zero,zero,zero,zero,zero
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 undef, i32 4, i32 5, i32 6, i32 undef, i32 undef, i32 undef, i32 undef>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_3456789a(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_3456789a:
; SSE2:       # %bb.0:
; SSE2-NEXT:    psrldq {{.*#+}} xmm0 = xmm0[6,7,8,9,10,11,12,13,14,15],zero,zero,zero,zero,zero,zero
; SSE2-NEXT:    pslldq {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm1[0,1,2,3,4,5]
; SSE2-NEXT:    por %xmm1, %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_3456789a:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    palignr {{.*#+}} xmm1 = xmm0[6,7,8,9,10,11,12,13,14,15],xmm1[0,1,2,3,4,5]
; SSSE3-NEXT:    movdqa %xmm1, %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_3456789a:
; SSE41:       # %bb.0:
; SSE41-NEXT:    palignr {{.*#+}} xmm1 = xmm0[6,7,8,9,10,11,12,13,14,15],xmm1[0,1,2,3,4,5]
; SSE41-NEXT:    movdqa %xmm1, %xmm0
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_3456789a:
; AVX:       # %bb.0:
; AVX-NEXT:    vpalignr {{.*#+}} xmm0 = xmm0[6,7,8,9,10,11,12,13,14,15],xmm1[0,1,2,3,4,5]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_u456uu9u(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_u456uu9u:
; SSE2:       # %bb.0:
; SSE2-NEXT:    psrldq {{.*#+}} xmm0 = xmm0[6,7,8,9,10,11,12,13,14,15],zero,zero,zero,zero,zero,zero
; SSE2-NEXT:    pslldq {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm1[0,1,2,3,4,5]
; SSE2-NEXT:    por %xmm1, %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_u456uu9u:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    palignr {{.*#+}} xmm1 = xmm0[6,7,8,9,10,11,12,13,14,15],xmm1[0,1,2,3,4,5]
; SSSE3-NEXT:    movdqa %xmm1, %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_u456uu9u:
; SSE41:       # %bb.0:
; SSE41-NEXT:    palignr {{.*#+}} xmm1 = xmm0[6,7,8,9,10,11,12,13,14,15],xmm1[0,1,2,3,4,5]
; SSE41-NEXT:    movdqa %xmm1, %xmm0
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_u456uu9u:
; AVX:       # %bb.0:
; AVX-NEXT:    vpalignr {{.*#+}} xmm0 = xmm0[6,7,8,9,10,11,12,13,14,15],xmm1[0,1,2,3,4,5]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 undef, i32 4, i32 5, i32 6, i32 undef, i32 undef, i32 9, i32 undef>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_56789abc(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_56789abc:
; SSE2:       # %bb.0:
; SSE2-NEXT:    psrldq {{.*#+}} xmm0 = xmm0[10,11,12,13,14,15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; SSE2-NEXT:    pslldq {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,xmm1[0,1,2,3,4,5,6,7,8,9]
; SSE2-NEXT:    por %xmm1, %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_56789abc:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    palignr {{.*#+}} xmm1 = xmm0[10,11,12,13,14,15],xmm1[0,1,2,3,4,5,6,7,8,9]
; SSSE3-NEXT:    movdqa %xmm1, %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_56789abc:
; SSE41:       # %bb.0:
; SSE41-NEXT:    palignr {{.*#+}} xmm1 = xmm0[10,11,12,13,14,15],xmm1[0,1,2,3,4,5,6,7,8,9]
; SSE41-NEXT:    movdqa %xmm1, %xmm0
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_56789abc:
; AVX:       # %bb.0:
; AVX-NEXT:    vpalignr {{.*#+}} xmm0 = xmm0[10,11,12,13,14,15],xmm1[0,1,2,3,4,5,6,7,8,9]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_u6uu9abu(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_u6uu9abu:
; SSE2:       # %bb.0:
; SSE2-NEXT:    psrldq {{.*#+}} xmm0 = xmm0[10,11,12,13,14,15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; SSE2-NEXT:    pslldq {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,xmm1[0,1,2,3,4,5,6,7,8,9]
; SSE2-NEXT:    por %xmm1, %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_u6uu9abu:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    palignr {{.*#+}} xmm1 = xmm0[10,11,12,13,14,15],xmm1[0,1,2,3,4,5,6,7,8,9]
; SSSE3-NEXT:    movdqa %xmm1, %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_u6uu9abu:
; SSE41:       # %bb.0:
; SSE41-NEXT:    palignr {{.*#+}} xmm1 = xmm0[10,11,12,13,14,15],xmm1[0,1,2,3,4,5,6,7,8,9]
; SSE41-NEXT:    movdqa %xmm1, %xmm0
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_u6uu9abu:
; AVX:       # %bb.0:
; AVX-NEXT:    vpalignr {{.*#+}} xmm0 = xmm0[10,11,12,13,14,15],xmm1[0,1,2,3,4,5,6,7,8,9]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 undef, i32 6, i32 undef, i32 undef, i32 9, i32 10, i32 11, i32 undef>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_0uuu1uuu(<8 x i16> %a) {
; SSE2-LABEL: shuffle_v8i16_0uuu1uuu:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,5,5,5]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_0uuu1uuu:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; SSSE3-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,5,5,5]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_0uuu1uuu:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pmovzxwq {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_0uuu1uuu:
; AVX:       # %bb.0:
; AVX-NEXT:    vpmovzxwq {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32> <i32 0, i32 undef, i32 undef, i32 undef, i32 1, i32 undef, i32 undef, i32 undef>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_0zzz1zzz(<8 x i16> %a) {
; SSE2-LABEL: shuffle_v8i16_0zzz1zzz:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pxor %xmm1, %xmm1
; SSE2-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; SSE2-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_0zzz1zzz:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pxor %xmm1, %xmm1
; SSSE3-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; SSSE3-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_0zzz1zzz:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pmovzxwq {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_0zzz1zzz:
; AVX:       # %bb.0:
; AVX-NEXT:    vpmovzxwq {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32> <i32 0, i32 9, i32 10, i32 11, i32 1, i32 13, i32 14, i32 15>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_0u1u2u3u(<8 x i16> %a) {
; SSE2-LABEL: shuffle_v8i16_0u1u2u3u:
; SSE2:       # %bb.0:
; SSE2-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_0u1u2u3u:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_0u1u2u3u:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pmovzxwd {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_0u1u2u3u:
; AVX:       # %bb.0:
; AVX-NEXT:    vpmovzxwd {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32> <i32 0, i32 undef, i32 1, i32 undef, i32 2, i32 undef, i32 3, i32 undef>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_0z1z2z3z(<8 x i16> %a) {
; SSE2-LABEL: shuffle_v8i16_0z1z2z3z:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pxor %xmm1, %xmm1
; SSE2-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_0z1z2z3z:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pxor %xmm1, %xmm1
; SSSE3-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_0z1z2z3z:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pmovzxwd {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_0z1z2z3z:
; AVX:       # %bb.0:
; AVX-NEXT:    vpmovzxwd {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32> <i32 0, i32 9, i32 1, i32 11, i32 2, i32 13, i32 3, i32 15>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_01100110(<8 x i16> %a) {
; SSE-LABEL: shuffle_v8i16_01100110:
; SSE:       # %bb.0:
; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,1,1,0,4,5,6,7]
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,1,0,1]
; SSE-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_01100110:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,1,1,0,4,5,6,7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,1,0,1]
; AVX1-NEXT:    retq
;
; AVX2-SLOW-LABEL: shuffle_v8i16_01100110:
; AVX2-SLOW:       # %bb.0:
; AVX2-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,1,1,0,4,5,6,7]
; AVX2-SLOW-NEXT:    vpbroadcastq %xmm0, %xmm0
; AVX2-SLOW-NEXT:    retq
;
; AVX2-FAST-LABEL: shuffle_v8i16_01100110:
; AVX2-FAST:       # %bb.0:
; AVX2-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,2,3,0,1,0,1,2,3,2,3,0,1]
; AVX2-FAST-NEXT:    retq
;
; AVX512VL-SLOW-LABEL: shuffle_v8i16_01100110:
; AVX512VL-SLOW:       # %bb.0:
; AVX512VL-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,1,1,0,4,5,6,7]
; AVX512VL-SLOW-NEXT:    vpbroadcastq %xmm0, %xmm0
; AVX512VL-SLOW-NEXT:    retq
;
; AVX512VL-FAST-LABEL: shuffle_v8i16_01100110:
; AVX512VL-FAST:       # %bb.0:
; AVX512VL-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,2,3,0,1,0,1,2,3,2,3,0,1]
; AVX512VL-FAST-NEXT:    retq
;
; XOPAVX1-LABEL: shuffle_v8i16_01100110:
; XOPAVX1:       # %bb.0:
; XOPAVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,1,1,0,4,5,6,7]
; XOPAVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,1,0,1]
; XOPAVX1-NEXT:    retq
;
; XOPAVX2-LABEL: shuffle_v8i16_01100110:
; XOPAVX2:       # %bb.0:
; XOPAVX2-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,1,1,0,4,5,6,7]
; XOPAVX2-NEXT:    vpbroadcastq %xmm0, %xmm0
; XOPAVX2-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32> <i32 0, i32 1, i32 1, i32 0, i32 0, i32 1, i32 1, i32 0>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_01u0u110(<8 x i16> %a) {
; SSE-LABEL: shuffle_v8i16_01u0u110:
; SSE:       # %bb.0:
; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,1,1,0,4,5,6,7]
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,1,0,1]
; SSE-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_01u0u110:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,1,1,0,4,5,6,7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,1,0,1]
; AVX1-NEXT:    retq
;
; AVX2-SLOW-LABEL: shuffle_v8i16_01u0u110:
; AVX2-SLOW:       # %bb.0:
; AVX2-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,1,1,0,4,5,6,7]
; AVX2-SLOW-NEXT:    vpbroadcastq %xmm0, %xmm0
; AVX2-SLOW-NEXT:    retq
;
; AVX2-FAST-LABEL: shuffle_v8i16_01u0u110:
; AVX2-FAST:       # %bb.0:
; AVX2-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,2,3,0,1,0,1,2,3,2,3,0,1]
; AVX2-FAST-NEXT:    retq
;
; AVX512VL-SLOW-LABEL: shuffle_v8i16_01u0u110:
; AVX512VL-SLOW:       # %bb.0:
; AVX512VL-SLOW-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,1,1,0,4,5,6,7]
; AVX512VL-SLOW-NEXT:    vpbroadcastq %xmm0, %xmm0
; AVX512VL-SLOW-NEXT:    retq
;
; AVX512VL-FAST-LABEL: shuffle_v8i16_01u0u110:
; AVX512VL-FAST:       # %bb.0:
; AVX512VL-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,2,3,2,3,0,1,0,1,2,3,2,3,0,1]
; AVX512VL-FAST-NEXT:    retq
;
; XOPAVX1-LABEL: shuffle_v8i16_01u0u110:
; XOPAVX1:       # %bb.0:
; XOPAVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,1,1,0,4,5,6,7]
; XOPAVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,1,0,1]
; XOPAVX1-NEXT:    retq
;
; XOPAVX2-LABEL: shuffle_v8i16_01u0u110:
; XOPAVX2:       # %bb.0:
; XOPAVX2-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,1,1,0,4,5,6,7]
; XOPAVX2-NEXT:    vpbroadcastq %xmm0, %xmm0
; XOPAVX2-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32> <i32 0, i32 1, i32 undef, i32 0, i32 undef, i32 1, i32 1, i32 0>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_467uu675(<8 x i16> %a) {
; SSE-LABEL: shuffle_v8i16_467uu675:
; SSE:       # %bb.0:
; SSE-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,6,7,5]
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,3,2,3]
; SSE-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_467uu675:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,6,7,5]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,3,2,3]
; AVX1-NEXT:    retq
;
; AVX2-SLOW-LABEL: shuffle_v8i16_467uu675:
; AVX2-SLOW:       # %bb.0:
; AVX2-SLOW-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,6,7,5]
; AVX2-SLOW-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,3,2,3]
; AVX2-SLOW-NEXT:    retq
;
; AVX2-FAST-LABEL: shuffle_v8i16_467uu675:
; AVX2-FAST:       # %bb.0:
; AVX2-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[8,9,12,13,14,15,10,11,8,9,12,13,14,15,10,11]
; AVX2-FAST-NEXT:    retq
;
; AVX512VL-SLOW-LABEL: shuffle_v8i16_467uu675:
; AVX512VL-SLOW:       # %bb.0:
; AVX512VL-SLOW-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,6,7,5]
; AVX512VL-SLOW-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,3,2,3]
; AVX512VL-SLOW-NEXT:    retq
;
; AVX512VL-FAST-LABEL: shuffle_v8i16_467uu675:
; AVX512VL-FAST:       # %bb.0:
; AVX512VL-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[8,9,12,13,14,15,10,11,8,9,12,13,14,15,10,11]
; AVX512VL-FAST-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_467uu675:
; XOP:       # %bb.0:
; XOP-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,6,7,5]
; XOP-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,3,2,3]
; XOP-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32> <i32 4, i32 6, i32 7, i32 undef, i32 undef, i32 6, i32 7, i32 5>
  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-LABEL: shuffle_v8i16_10325476:
; AVX512VL:       # %bb.0:
; AVX512VL-NEXT:    vprold $16, %xmm0, %xmm0
; AVX512VL-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_10325476:
; XOP:       # %bb.0:
; XOP-NEXT:    vprotd $16, %xmm0, %xmm0
; XOP-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-LABEL: shuffle_v8i16_12305674:
; AVX512VL:       # %bb.0:
; AVX512VL-NEXT:    vprolq $48, %xmm0, %xmm0
; AVX512VL-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_12305674:
; XOP:       # %bb.0:
; XOP-NEXT:    vprotq $48, %xmm0, %xmm0
; XOP-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
}

define <8 x i16> @shuffle_v8i16_02460246(<8 x i16> %a) {
; SSE2-LABEL: shuffle_v8i16_02460246:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,2,0,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,6,6,4]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,3,2,1]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,1,3,2,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,6,4,5]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_02460246:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,4,5,8,9,12,13,0,1,4,5,8,9,12,13]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_02460246:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,4,5,8,9,12,13,0,1,4,5,8,9,12,13]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_02460246:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,4,5,8,9,12,13,0,1,4,5,8,9,12,13]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> undef, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 0, i32 2, i32 4, i32 6>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_04040404(<8 x i16> %a) {
; SSE2-LABEL: shuffle_v8i16_04040404:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,0]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,0,2,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,6,4,6,4]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_04040404:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,0,1,8,9,0,1,8,9,0,1,8,9]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_04040404:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,0,1,8,9,0,1,8,9,0,1,8,9]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_04040404:
; AVX:       # %bb.0:
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,0,1,8,9,0,1,8,9,0,1,8,9]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> undef, <8 x i32> <i32 0, i32 4, i32 0, i32 4, i32 0, i32 4, i32 0, i32 4>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_02468ACE(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_02468ACE:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshuflw {{.*#+}} xmm1 = xmm1[0,2,2,3,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,4,6,6,7]
; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[0,2,2,3]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,2,3,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,6,6,7]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
; SSE2-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_02468ACE:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    movdqa {{.*#+}} xmm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15]
; SSSE3-NEXT:    pshufb %xmm2, %xmm1
; SSSE3-NEXT:    pshufb %xmm2, %xmm0
; SSSE3-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_02468ACE:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pxor %xmm2, %xmm2
; SSE41-NEXT:    pblendw {{.*#+}} xmm1 = xmm1[0],xmm2[1],xmm1[2],xmm2[3],xmm1[4],xmm2[5],xmm1[6],xmm2[7]
; SSE41-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0],xmm2[1],xmm0[2],xmm2[3],xmm0[4],xmm2[5],xmm0[6],xmm2[7]
; SSE41-NEXT:    packusdw %xmm1, %xmm0
; SSE41-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_02468ACE:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpxor %xmm2, %xmm2, %xmm2
; AVX1-NEXT:    vpblendw {{.*#+}} xmm1 = xmm1[0],xmm2[1],xmm1[2],xmm2[3],xmm1[4],xmm2[5],xmm1[6],xmm2[7]
; AVX1-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0],xmm2[1],xmm0[2],xmm2[3],xmm0[4],xmm2[5],xmm0[6],xmm2[7]
; AVX1-NEXT:    vpackusdw %xmm1, %xmm0, %xmm0
; AVX1-NEXT:    retq
;
; AVX2-LABEL: shuffle_v8i16_02468ACE:
; AVX2:       # %bb.0:
; AVX2-NEXT:    vpxor %xmm2, %xmm2, %xmm2
; AVX2-NEXT:    vpblendw {{.*#+}} xmm1 = xmm1[0],xmm2[1],xmm1[2],xmm2[3],xmm1[4],xmm2[5],xmm1[6],xmm2[7]
; AVX2-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0],xmm2[1],xmm0[2],xmm2[3],xmm0[4],xmm2[5],xmm0[6],xmm2[7]
; AVX2-NEXT:    vpackusdw %xmm1, %xmm0, %xmm0
; AVX2-NEXT:    retq
;
; AVX512VL-LABEL: shuffle_v8i16_02468ACE:
; AVX512VL:       # %bb.0:
; AVX512VL-NEXT:    # kill: def $xmm0 killed $xmm0 def $ymm0
; AVX512VL-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
; AVX512VL-NEXT:    vpmovdw %ymm0, %xmm0
; AVX512VL-NEXT:    vzeroupper
; AVX512VL-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_02468ACE:
; XOP:       # %bb.0:
; XOP-NEXT:    vpperm {{.*#+}} xmm0 = xmm0[0,1,4,5,8,9,12,13],xmm1[0,1,4,5,8,9,12,13]
; XOP-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_048C048C(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_048C048C:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[0,2,2,3]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm1 = xmm1[0,2,0,2,4,5,6,7]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,0,2,4,5,6,7]
; SSE2-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_048C048C:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[0,2,2,3]
; SSSE3-NEXT:    pshuflw {{.*#+}} xmm1 = xmm1[0,2,0,2,4,5,6,7]
; SSSE3-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
; SSSE3-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,0,2,4,5,6,7]
; SSSE3-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_048C048C:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pxor %xmm2, %xmm2
; SSE41-NEXT:    pblendw {{.*#+}} xmm1 = xmm1[0],xmm2[1,2,3],xmm1[4],xmm2[5,6,7]
; SSE41-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0],xmm2[1,2,3],xmm0[4],xmm2[5,6,7]
; SSE41-NEXT:    packusdw %xmm1, %xmm0
; SSE41-NEXT:    packusdw %xmm0, %xmm0
; SSE41-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_048C048C:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpxor %xmm2, %xmm2, %xmm2
; AVX1-NEXT:    vpblendw {{.*#+}} xmm1 = xmm1[0],xmm2[1,2,3],xmm1[4],xmm2[5,6,7]
; AVX1-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0],xmm2[1,2,3],xmm0[4],xmm2[5,6,7]
; AVX1-NEXT:    vpackusdw %xmm1, %xmm0, %xmm0
; AVX1-NEXT:    vpackusdw %xmm0, %xmm0, %xmm0
; AVX1-NEXT:    retq
;
; AVX2-LABEL: shuffle_v8i16_048C048C:
; AVX2:       # %bb.0:
; AVX2-NEXT:    vpxor %xmm2, %xmm2, %xmm2
; AVX2-NEXT:    vpblendw {{.*#+}} xmm1 = xmm1[0],xmm2[1,2,3],xmm1[4],xmm2[5,6,7]
; AVX2-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0],xmm2[1,2,3],xmm0[4],xmm2[5,6,7]
; AVX2-NEXT:    vpackusdw %xmm1, %xmm0, %xmm0
; AVX2-NEXT:    vpackusdw %xmm0, %xmm0, %xmm0
; AVX2-NEXT:    retq
;
; AVX512VL-LABEL: shuffle_v8i16_048C048C:
; AVX512VL:       # %bb.0:
; AVX512VL-NEXT:    vpbroadcastq {{.*#+}} xmm2 = [3377734080528384,3377734080528384]
; AVX512VL-NEXT:    vpermt2w %xmm1, %xmm2, %xmm0
; AVX512VL-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_048C048C:
; XOP:       # %bb.0:
; XOP-NEXT:    vpperm {{.*#+}} xmm0 = xmm0[0,1,8,9],xmm1[0,1,8,9],xmm0[0,1,8,9],xmm1[0,1,8,9]
; XOP-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 4, i32 8, i32 12, i32 0, i32 4, i32 8, i32 12>
  ret <8 x i16> %shuffle
}

;
; Shuffle to logical bit shifts
;
define <8 x i16> @shuffle_v8i16_z0z2z4z6(<8 x i16> %a) {
; SSE-LABEL: shuffle_v8i16_z0z2z4z6:
; SSE:       # %bb.0:
; SSE-NEXT:    pslld $16, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_z0z2z4z6:
; AVX:       # %bb.0:
; AVX-NEXT:    vpslld $16, %xmm0, %xmm0
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32><i32 8, i32 0, i32 8, i32 2, i32 8, i32 4, i32 8, i32 6>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_zzz0zzz4(<8 x i16> %a) {
; SSE-LABEL: shuffle_v8i16_zzz0zzz4:
; SSE:       # %bb.0:
; SSE-NEXT:    psllq $48, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_zzz0zzz4:
; AVX:       # %bb.0:
; AVX-NEXT:    vpsllq $48, %xmm0, %xmm0
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32><i32 8, i32 8, i32 8, i32 0, i32 8, i32 8, i32 8, i32 4>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_zz01zX4X(<8 x i16> %a) {
; SSE-LABEL: shuffle_v8i16_zz01zX4X:
; SSE:       # %bb.0:
; SSE-NEXT:    psllq $32, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_zz01zX4X:
; AVX:       # %bb.0:
; AVX-NEXT:    vpsllq $32, %xmm0, %xmm0
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32><i32 8, i32 8, i32 0, i32 1, i32 8, i32 undef, i32 4, i32 undef>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_z0X2z456(<8 x i16> %a) {
; SSE-LABEL: shuffle_v8i16_z0X2z456:
; SSE:       # %bb.0:
; SSE-NEXT:    psllq $16, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_z0X2z456:
; AVX:       # %bb.0:
; AVX-NEXT:    vpsllq $16, %xmm0, %xmm0
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32><i32 8, i32 0, i32 undef, i32 2, i32 8, i32 4, i32 5, i32 6>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_1z3zXz7z(<8 x i16> %a) {
; SSE-LABEL: shuffle_v8i16_1z3zXz7z:
; SSE:       # %bb.0:
; SSE-NEXT:    psrld $16, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_1z3zXz7z:
; AVX:       # %bb.0:
; AVX-NEXT:    vpsrld $16, %xmm0, %xmm0
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32><i32 1, i32 8, i32 3, i32 8, i32 undef, i32 8, i32 7, i32 8>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_1X3z567z(<8 x i16> %a) {
; SSE-LABEL: shuffle_v8i16_1X3z567z:
; SSE:       # %bb.0:
; SSE-NEXT:    psrlq $16, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_1X3z567z:
; AVX:       # %bb.0:
; AVX-NEXT:    vpsrlq $16, %xmm0, %xmm0
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32><i32 1, i32 undef, i32 3, i32 8, i32 5, i32 6, i32 7, i32 8>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_23zz67zz(<8 x i16> %a) {
; SSE-LABEL: shuffle_v8i16_23zz67zz:
; SSE:       # %bb.0:
; SSE-NEXT:    psrlq $32, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_23zz67zz:
; AVX:       # %bb.0:
; AVX-NEXT:    vpsrlq $32, %xmm0, %xmm0
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32><i32 2, i32 3, i32 8, i32 8, i32 6, i32 7, i32 8, i32 8>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_3zXXXzzz(<8 x i16> %a) {
; SSE-LABEL: shuffle_v8i16_3zXXXzzz:
; SSE:       # %bb.0:
; SSE-NEXT:    psrlq $48, %xmm0
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_3zXXXzzz:
; AVX:       # %bb.0:
; AVX-NEXT:    vpsrlq $48, %xmm0, %xmm0
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32><i32 3, i32 8, i32 undef, i32 undef, i32 undef, i32 8, i32 8, i32 8>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_01u3zzuz(<8 x i16> %a) {
; SSE-LABEL: shuffle_v8i16_01u3zzuz:
; SSE:       # %bb.0:
; SSE-NEXT:    movq {{.*#+}} xmm0 = xmm0[0],zero
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_01u3zzuz:
; AVX:       # %bb.0:
; AVX-NEXT:    vmovq {{.*#+}} xmm0 = xmm0[0],zero
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32> <i32 0, i32 1, i32 undef, i32 3, i32 8, i32 8, i32 undef, i32 8>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_0z234567(<8 x i16> %a) {
; SSE2-LABEL: shuffle_v8i16_0z234567:
; SSE2:       # %bb.0:
; SSE2-NEXT:    andps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_0z234567:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    andps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_0z234567:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pxor %xmm1, %xmm1
; SSE41-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1],xmm0[2,3,4,5,6,7]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_0z234567:
; AVX:       # %bb.0:
; AVX-NEXT:    vpxor %xmm1, %xmm1, %xmm1
; AVX-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1],xmm0[2,3,4,5,6,7]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32> <i32 0, i32 9, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_0zzzz5z7(<8 x i16> %a) {
; SSE2-LABEL: shuffle_v8i16_0zzzz5z7:
; SSE2:       # %bb.0:
; SSE2-NEXT:    andps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_0zzzz5z7:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    andps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_0zzzz5z7:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pxor %xmm1, %xmm1
; SSE41-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3,4],xmm0[5],xmm1[6],xmm0[7]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_0zzzz5z7:
; AVX:       # %bb.0:
; AVX-NEXT:    vpxor %xmm1, %xmm1, %xmm1
; AVX-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3,4],xmm0[5],xmm1[6],xmm0[7]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32> <i32 0, i32 8, i32 8, i32 8, i32 8, i32 5, i32 8, i32 7>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_0123456z(<8 x i16> %a) {
; SSE2-LABEL: shuffle_v8i16_0123456z:
; SSE2:       # %bb.0:
; SSE2-NEXT:    andps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_0123456z:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    andps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_0123456z:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pxor %xmm1, %xmm1
; SSE41-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1],xmm0[2,3,4,5,6],xmm1[7]
; SSE41-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_0123456z:
; AVX:       # %bb.0:
; AVX-NEXT:    vpxor %xmm1, %xmm1, %xmm1
; AVX-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1],xmm0[2,3,4,5,6],xmm1[7]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32> <i32 0, i32 9, i32 2, i32 3, i32 4, i32 5, i32 6, i32 15>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_fu3ucc5u(<8 x i16> %a, <8 x i16> %b) {
; SSE-LABEL: shuffle_v8i16_fu3ucc5u:
; SSE:       # %bb.0:
; SSE-NEXT:    pslldq {{.*#+}} xmm0 = zero,zero,xmm0[0,1,2,3,4,5,6,7,8,9,10,11,12,13]
; SSE-NEXT:    pshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,7,5,4,4]
; SSE-NEXT:    punpckhdq {{.*#+}} xmm1 = xmm1[2],xmm0[2],xmm1[3],xmm0[3]
; SSE-NEXT:    movdqa %xmm1, %xmm0
; SSE-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_fu3ucc5u:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpslldq {{.*#+}} xmm0 = zero,zero,xmm0[0,1,2,3,4,5,6,7,8,9,10,11,12,13]
; AVX1-NEXT:    vpshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,7,5,4,4]
; AVX1-NEXT:    vpunpckhdq {{.*#+}} xmm0 = xmm1[2],xmm0[2],xmm1[3],xmm0[3]
; AVX1-NEXT:    retq
;
; AVX2OR512VL-LABEL: shuffle_v8i16_fu3ucc5u:
; AVX2OR512VL:       # %bb.0:
; AVX2OR512VL-NEXT:    vpslldq {{.*#+}} xmm0 = zero,zero,xmm0[0,1,2,3,4,5,6,7,8,9,10,11,12,13]
; AVX2OR512VL-NEXT:    vpshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,7,5,4,4]
; AVX2OR512VL-NEXT:    vpunpckhdq {{.*#+}} xmm0 = xmm1[2],xmm0[2],xmm1[3],xmm0[3]
; AVX2OR512VL-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_fu3ucc5u:
; XOP:       # %bb.0:
; XOP-NEXT:    vpperm {{.*#+}} xmm0 = xmm1[14,15,10,11],xmm0[6,7,8,9],xmm1[8,9,8,9],xmm0[10,11,12,13]
; XOP-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 15, i32 undef, i32 3, i32 undef, i32 12, i32 12, i32 5, i32 undef>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_8012345u(<8 x i16> %a) {
; SSE-LABEL: shuffle_v8i16_8012345u:
; SSE:       # %bb.0:
; SSE-NEXT:    pslldq {{.*#+}} xmm0 = zero,zero,xmm0[0,1,2,3,4,5,6,7,8,9,10,11,12,13]
; SSE-NEXT:    retq
;
; AVX-LABEL: shuffle_v8i16_8012345u:
; AVX:       # %bb.0:
; AVX-NEXT:    vpslldq {{.*#+}} xmm0 = zero,zero,xmm0[0,1,2,3,4,5,6,7,8,9,10,11,12,13]
; AVX-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> zeroinitializer, <8 x i32> <i32 8, i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 undef>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_02468ace(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_02468ace:
; SSE2:       # %bb.0:
; SSE2-NEXT:    pshuflw {{.*#+}} xmm1 = xmm1[0,2,2,3,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,4,6,6,7]
; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[0,2,2,3]
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,2,3,4,5,6,7]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,6,6,7]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
; SSE2-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_02468ace:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    movdqa {{.*#+}} xmm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15]
; SSSE3-NEXT:    pshufb %xmm2, %xmm1
; SSSE3-NEXT:    pshufb %xmm2, %xmm0
; SSSE3-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_02468ace:
; SSE41:       # %bb.0:
; SSE41-NEXT:    pxor %xmm2, %xmm2
; SSE41-NEXT:    pblendw {{.*#+}} xmm1 = xmm1[0],xmm2[1],xmm1[2],xmm2[3],xmm1[4],xmm2[5],xmm1[6],xmm2[7]
; SSE41-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0],xmm2[1],xmm0[2],xmm2[3],xmm0[4],xmm2[5],xmm0[6],xmm2[7]
; SSE41-NEXT:    packusdw %xmm1, %xmm0
; SSE41-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_02468ace:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpxor %xmm2, %xmm2, %xmm2
; AVX1-NEXT:    vpblendw {{.*#+}} xmm1 = xmm1[0],xmm2[1],xmm1[2],xmm2[3],xmm1[4],xmm2[5],xmm1[6],xmm2[7]
; AVX1-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0],xmm2[1],xmm0[2],xmm2[3],xmm0[4],xmm2[5],xmm0[6],xmm2[7]
; AVX1-NEXT:    vpackusdw %xmm1, %xmm0, %xmm0
; AVX1-NEXT:    retq
;
; AVX2-LABEL: shuffle_v8i16_02468ace:
; AVX2:       # %bb.0:
; AVX2-NEXT:    vpxor %xmm2, %xmm2, %xmm2
; AVX2-NEXT:    vpblendw {{.*#+}} xmm1 = xmm1[0],xmm2[1],xmm1[2],xmm2[3],xmm1[4],xmm2[5],xmm1[6],xmm2[7]
; AVX2-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0],xmm2[1],xmm0[2],xmm2[3],xmm0[4],xmm2[5],xmm0[6],xmm2[7]
; AVX2-NEXT:    vpackusdw %xmm1, %xmm0, %xmm0
; AVX2-NEXT:    retq
;
; AVX512VL-LABEL: shuffle_v8i16_02468ace:
; AVX512VL:       # %bb.0:
; AVX512VL-NEXT:    # kill: def $xmm0 killed $xmm0 def $ymm0
; AVX512VL-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
; AVX512VL-NEXT:    vpmovdw %ymm0, %xmm0
; AVX512VL-NEXT:    vzeroupper
; AVX512VL-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_02468ace:
; XOP:       # %bb.0:
; XOP-NEXT:    vpperm {{.*#+}} xmm0 = xmm0[0,1,4,5,8,9,12,13],xmm1[0,1,4,5,8,9,12,13]
; XOP-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14>
  ret <8 x i16> %shuffle
}

define <8 x i16> @shuffle_v8i16_13579bdf(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: shuffle_v8i16_13579bdf:
; SSE2:       # %bb.0:
; SSE2-NEXT:    psrad $16, %xmm1
; SSE2-NEXT:    psrad $16, %xmm0
; SSE2-NEXT:    packssdw %xmm1, %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: shuffle_v8i16_13579bdf:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    psrad $16, %xmm1
; SSSE3-NEXT:    psrad $16, %xmm0
; SSSE3-NEXT:    packssdw %xmm1, %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: shuffle_v8i16_13579bdf:
; SSE41:       # %bb.0:
; SSE41-NEXT:    psrld $16, %xmm1
; SSE41-NEXT:    psrld $16, %xmm0
; SSE41-NEXT:    packusdw %xmm1, %xmm0
; SSE41-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_13579bdf:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpsrld $16, %xmm1, %xmm1
; AVX1-NEXT:    vpsrld $16, %xmm0, %xmm0
; AVX1-NEXT:    vpackusdw %xmm1, %xmm0, %xmm0
; AVX1-NEXT:    retq
;
; AVX2OR512VL-LABEL: shuffle_v8i16_13579bdf:
; AVX2OR512VL:       # %bb.0:
; AVX2OR512VL-NEXT:    vpsrld $16, %xmm1, %xmm1
; AVX2OR512VL-NEXT:    vpsrld $16, %xmm0, %xmm0
; AVX2OR512VL-NEXT:    vpackusdw %xmm1, %xmm0, %xmm0
; AVX2OR512VL-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_13579bdf:
; XOP:       # %bb.0:
; XOP-NEXT:    vpperm {{.*#+}} xmm0 = xmm0[2,3,6,7,10,11,14,15],xmm1[2,3,6,7,10,11,14,15]
; XOP-NEXT:    retq
  %shuffle = shufflevector <8 x i16> %a, <8 x i16> %b, <8 x i32> <i32 1, i32 3, i32 5, i32 7, i32 9, i32 11, i32 13, i32 15>
  ret <8 x i16> %shuffle
}

; PR40306
define <8 x i16> @shuffle_v8i16_9zzzuuuu(<8 x i16> %x) {
; SSE-LABEL: shuffle_v8i16_9zzzuuuu:
; SSE:       # %bb.0:
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; SSE-NEXT:    psrldq {{.*#+}} xmm0 = xmm0[14,15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; SSE-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_9zzzuuuu:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; AVX1-NEXT:    vpsrldq {{.*#+}} xmm0 = xmm0[14,15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; AVX1-NEXT:    retq
;
; AVX2-SLOW-LABEL: shuffle_v8i16_9zzzuuuu:
; AVX2-SLOW:       # %bb.0:
; AVX2-SLOW-NEXT:    vpbroadcastd %xmm0, %xmm0
; AVX2-SLOW-NEXT:    vpsrldq {{.*#+}} xmm0 = xmm0[14,15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; AVX2-SLOW-NEXT:    retq
;
; AVX2-FAST-LABEL: shuffle_v8i16_9zzzuuuu:
; AVX2-FAST:       # %bb.0:
; AVX2-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[2,3],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; AVX2-FAST-NEXT:    retq
;
; AVX512VL-SLOW-LABEL: shuffle_v8i16_9zzzuuuu:
; AVX512VL-SLOW:       # %bb.0:
; AVX512VL-SLOW-NEXT:    vbroadcastss %xmm0, %xmm0
; AVX512VL-SLOW-NEXT:    vpsrldq {{.*#+}} xmm0 = xmm0[14,15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; AVX512VL-SLOW-NEXT:    retq
;
; AVX512VL-FAST-LABEL: shuffle_v8i16_9zzzuuuu:
; AVX512VL-FAST:       # %bb.0:
; AVX512VL-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[2,3],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; AVX512VL-FAST-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_9zzzuuuu:
; XOP:       # %bb.0:
; XOP-NEXT:    extrq {{.*#+}} xmm0 = xmm0[2,3],zero,zero,zero,zero,zero,zero,xmm0[u,u,u,u,u,u,u,u]
; XOP-NEXT:    retq
  %r = shufflevector <8 x i16> zeroinitializer, <8 x i16> %x, <8 x i32> <i32 9, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef>
  ret <8 x i16> %r
}

; PR40318
define <8 x i16> @shuffle_v8i16_2zzzuuuu(<8 x i16> %x) {
; SSE-LABEL: shuffle_v8i16_2zzzuuuu:
; SSE:       # %bb.0:
; SSE-NEXT:    pslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5]
; SSE-NEXT:    psrldq {{.*#+}} xmm0 = xmm0[14,15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; SSE-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_2zzzuuuu:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5]
; AVX1-NEXT:    vpsrldq {{.*#+}} xmm0 = xmm0[14,15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; AVX1-NEXT:    retq
;
; AVX2-SLOW-LABEL: shuffle_v8i16_2zzzuuuu:
; AVX2-SLOW:       # %bb.0:
; AVX2-SLOW-NEXT:    vpslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5]
; AVX2-SLOW-NEXT:    vpsrldq {{.*#+}} xmm0 = xmm0[14,15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; AVX2-SLOW-NEXT:    retq
;
; AVX2-FAST-LABEL: shuffle_v8i16_2zzzuuuu:
; AVX2-FAST:       # %bb.0:
; AVX2-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[4,5],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; AVX2-FAST-NEXT:    retq
;
; AVX512VL-SLOW-LABEL: shuffle_v8i16_2zzzuuuu:
; AVX512VL-SLOW:       # %bb.0:
; AVX512VL-SLOW-NEXT:    vpslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5]
; AVX512VL-SLOW-NEXT:    vpsrldq {{.*#+}} xmm0 = xmm0[14,15],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; AVX512VL-SLOW-NEXT:    retq
;
; AVX512VL-FAST-LABEL: shuffle_v8i16_2zzzuuuu:
; AVX512VL-FAST:       # %bb.0:
; AVX512VL-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[4,5],zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero
; AVX512VL-FAST-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_2zzzuuuu:
; XOP:       # %bb.0:
; XOP-NEXT:    extrq {{.*#+}} xmm0 = xmm0[4,5],zero,zero,zero,zero,zero,zero,xmm0[u,u,u,u,u,u,u,u]
; XOP-NEXT:    retq
  %r = shufflevector <8 x i16> %x, <8 x i16> zeroinitializer, <8 x i32> <i32 2, i32 9, i32 10, i32 11, i32 undef, i32 undef, i32 undef, i32 undef>
  ret <8 x i16> %r
}

define <8 x i16> @shuffle_v8i16_3uu6zzzz(<8 x i16> %x) {
; SSE-LABEL: shuffle_v8i16_3uu6zzzz:
; SSE:       # %bb.0:
; SSE-NEXT:    pslldq {{.*#+}} xmm0 = zero,zero,xmm0[0,1,2,3,4,5,6,7,8,9,10,11,12,13]
; SSE-NEXT:    psrldq {{.*#+}} xmm0 = xmm0[8,9,10,11,12,13,14,15],zero,zero,zero,zero,zero,zero,zero,zero
; SSE-NEXT:    retq
;
; AVX1-LABEL: shuffle_v8i16_3uu6zzzz:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vpslldq {{.*#+}} xmm0 = zero,zero,xmm0[0,1,2,3,4,5,6,7,8,9,10,11,12,13]
; AVX1-NEXT:    vpsrldq {{.*#+}} xmm0 = xmm0[8,9,10,11,12,13,14,15],zero,zero,zero,zero,zero,zero,zero,zero
; AVX1-NEXT:    retq
;
; AVX2-SLOW-LABEL: shuffle_v8i16_3uu6zzzz:
; AVX2-SLOW:       # %bb.0:
; AVX2-SLOW-NEXT:    vpslldq {{.*#+}} xmm0 = zero,zero,xmm0[0,1,2,3,4,5,6,7,8,9,10,11,12,13]
; AVX2-SLOW-NEXT:    vpsrldq {{.*#+}} xmm0 = xmm0[8,9,10,11,12,13,14,15],zero,zero,zero,zero,zero,zero,zero,zero
; AVX2-SLOW-NEXT:    retq
;
; AVX2-FAST-LABEL: shuffle_v8i16_3uu6zzzz:
; AVX2-FAST:       # %bb.0:
; AVX2-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[6,7,8,9,10,11,12,13],zero,zero,zero,zero,zero,zero,zero,zero
; AVX2-FAST-NEXT:    retq
;
; AVX512VL-SLOW-LABEL: shuffle_v8i16_3uu6zzzz:
; AVX512VL-SLOW:       # %bb.0:
; AVX512VL-SLOW-NEXT:    vpslldq {{.*#+}} xmm0 = zero,zero,xmm0[0,1,2,3,4,5,6,7,8,9,10,11,12,13]
; AVX512VL-SLOW-NEXT:    vpsrldq {{.*#+}} xmm0 = xmm0[8,9,10,11,12,13,14,15],zero,zero,zero,zero,zero,zero,zero,zero
; AVX512VL-SLOW-NEXT:    retq
;
; AVX512VL-FAST-LABEL: shuffle_v8i16_3uu6zzzz:
; AVX512VL-FAST:       # %bb.0:
; AVX512VL-FAST-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[6,7,8,9,10,11,12,13],zero,zero,zero,zero,zero,zero,zero,zero
; AVX512VL-FAST-NEXT:    retq
;
; XOP-LABEL: shuffle_v8i16_3uu6zzzz:
; XOP:       # %bb.0:
; XOP-NEXT:    vpslldq {{.*#+}} xmm0 = zero,zero,xmm0[0,1,2,3,4,5,6,7,8,9,10,11,12,13]
; XOP-NEXT:    vpsrldq {{.*#+}} xmm0 = xmm0[8,9,10,11,12,13,14,15],zero,zero,zero,zero,zero,zero,zero,zero
; XOP-NEXT:    retq
  %r = shufflevector <8 x i16> %x, <8 x i16> zeroinitializer, <8 x i32> <i32 3, i32 undef, i32 undef, i32 6, i32 8, i32 8, i32 8, i32 8>
  ret <8 x i16> %r
}

define <8 x i16> @mask_v8i16_012345ef(<8 x i16> %a, <8 x i16> %b) {
; SSE2-LABEL: mask_v8i16_012345ef:
; SSE2:       # %bb.0:
; SSE2-NEXT:    shufps {{.*#+}} xmm0 = xmm0[3,0],xmm1[2,0]
; SSE2-NEXT:    shufps {{.*#+}} xmm1 = xmm1[0,1],xmm0[2,0]
; SSE2-NEXT:    movaps %xmm1, %xmm0
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: mask_v8i16_012345ef:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    shufps {{.*#+}} xmm0 = xmm0[3,0],xmm1[2,0]
; SSSE3-NEXT:    shufps {{.*#+}} xmm1 = xmm1[0,1],xmm0[2,0]
; SSSE3-NEXT:    movaps %xmm1, %xmm0
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: mask_v8i16_012345ef:
; SSE41:       # %bb.0:
; SSE41-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1,2],xmm0[3]
; SSE41-NEXT:    retq
;
; AVX-LABEL: mask_v8i16_012345ef:
; AVX:       # %bb.0:
; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm1[0,1,2],xmm0[3]
; AVX-NEXT:    retq
  %1 = bitcast <8 x i16> %a to <2 x i64>
  %2 = bitcast <8 x i16> %b to <2 x i64>
  %3 = and <2 x i64> %1, <i64 0, i64 -4294967296>
  %4 = and <2 x i64> %2, <i64 -1, i64 4294967295>
  %5 = or <2 x i64> %4, %3
  %6 = bitcast <2 x i64> %5 to <8 x i16>
  ret <8 x i16> %6
}

define <8 x i16> @insert_dup_mem_v8i16_i32(ptr %ptr) {
; SSE-LABEL: insert_dup_mem_v8i16_i32:
; SSE:       # %bb.0:
; SSE-NEXT:    movd {{.*#+}} xmm0 = mem[0],zero,zero,zero
; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; SSE-NEXT:    retq
;
; AVX1-LABEL: insert_dup_mem_v8i16_i32:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vmovd {{.*#+}} xmm0 = mem[0],zero,zero,zero
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; AVX1-NEXT:    retq
;
; AVX2OR512VL-LABEL: insert_dup_mem_v8i16_i32:
; AVX2OR512VL:       # %bb.0:
; AVX2OR512VL-NEXT:    vpbroadcastw (%rdi), %xmm0
; AVX2OR512VL-NEXT:    retq
;
; XOPAVX1-LABEL: insert_dup_mem_v8i16_i32:
; XOPAVX1:       # %bb.0:
; XOPAVX1-NEXT:    vmovd {{.*#+}} xmm0 = mem[0],zero,zero,zero
; XOPAVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
; XOPAVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; XOPAVX1-NEXT:    retq
;
; XOPAVX2-LABEL: insert_dup_mem_v8i16_i32:
; XOPAVX2:       # %bb.0:
; XOPAVX2-NEXT:    vpbroadcastw (%rdi), %xmm0
; XOPAVX2-NEXT:    retq
  %tmp = load i32, ptr %ptr, align 4
  %tmp1 = insertelement <4 x i32> zeroinitializer, i32 %tmp, i32 0
  %tmp2 = bitcast <4 x i32> %tmp1 to <8 x i16>
  %tmp3 = shufflevector <8 x i16> %tmp2, <8 x i16> undef, <8 x i32> zeroinitializer
  ret <8 x i16> %tmp3
}

define <8 x i16> @insert_dup_mem_v8i16_sext_i16(ptr %ptr) {
; SSE-LABEL: insert_dup_mem_v8i16_sext_i16:
; SSE:       # %bb.0:
; SSE-NEXT:    movzwl (%rdi), %eax
; SSE-NEXT:    movd %eax, %xmm0
; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; SSE-NEXT:    retq
;
; AVX1-LABEL: insert_dup_mem_v8i16_sext_i16:
; AVX1:       # %bb.0:
; AVX1-NEXT:    movzwl (%rdi), %eax
; AVX1-NEXT:    vmovd %eax, %xmm0
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; AVX1-NEXT:    retq
;
; AVX2OR512VL-LABEL: insert_dup_mem_v8i16_sext_i16:
; AVX2OR512VL:       # %bb.0:
; AVX2OR512VL-NEXT:    vpbroadcastw (%rdi), %xmm0
; AVX2OR512VL-NEXT:    retq
;
; XOPAVX1-LABEL: insert_dup_mem_v8i16_sext_i16:
; XOPAVX1:       # %bb.0:
; XOPAVX1-NEXT:    movzwl (%rdi), %eax
; XOPAVX1-NEXT:    vmovd %eax, %xmm0
; XOPAVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
; XOPAVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; XOPAVX1-NEXT:    retq
;
; XOPAVX2-LABEL: insert_dup_mem_v8i16_sext_i16:
; XOPAVX2:       # %bb.0:
; XOPAVX2-NEXT:    vpbroadcastw (%rdi), %xmm0
; XOPAVX2-NEXT:    retq
  %tmp = load i16, ptr %ptr, align 2
  %tmp1 = sext i16 %tmp to i32
  %tmp2 = insertelement <4 x i32> zeroinitializer, i32 %tmp1, i32 0
  %tmp3 = bitcast <4 x i32> %tmp2 to <8 x i16>
  %tmp4 = shufflevector <8 x i16> %tmp3, <8 x i16> undef, <8 x i32> zeroinitializer
  ret <8 x i16> %tmp4
}

define <8 x i16> @insert_dup_elt1_mem_v8i16_i32(ptr %ptr) {
; SSE-LABEL: insert_dup_elt1_mem_v8i16_i32:
; SSE:       # %bb.0:
; SSE-NEXT:    movd {{.*#+}} xmm0 = mem[0],zero,zero,zero
; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,1,1,1,4,5,6,7]
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; SSE-NEXT:    retq
;
; AVX1-LABEL: insert_dup_elt1_mem_v8i16_i32:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vmovd {{.*#+}} xmm0 = mem[0],zero,zero,zero
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[1,1,1,1,4,5,6,7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; AVX1-NEXT:    retq
;
; AVX2OR512VL-LABEL: insert_dup_elt1_mem_v8i16_i32:
; AVX2OR512VL:       # %bb.0:
; AVX2OR512VL-NEXT:    vpbroadcastw 2(%rdi), %xmm0
; AVX2OR512VL-NEXT:    retq
;
; XOPAVX1-LABEL: insert_dup_elt1_mem_v8i16_i32:
; XOPAVX1:       # %bb.0:
; XOPAVX1-NEXT:    vmovd {{.*#+}} xmm0 = mem[0],zero,zero,zero
; XOPAVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[1,1,1,1,4,5,6,7]
; XOPAVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; XOPAVX1-NEXT:    retq
;
; XOPAVX2-LABEL: insert_dup_elt1_mem_v8i16_i32:
; XOPAVX2:       # %bb.0:
; XOPAVX2-NEXT:    vpbroadcastw 2(%rdi), %xmm0
; XOPAVX2-NEXT:    retq
  %tmp = load i32, ptr %ptr, align 4
  %tmp1 = insertelement <4 x i32> zeroinitializer, i32 %tmp, i32 0
  %tmp2 = bitcast <4 x i32> %tmp1 to <8 x i16>
  %tmp3 = shufflevector <8 x i16> %tmp2, <8 x i16> undef, <8 x i32> <i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1>
  ret <8 x i16> %tmp3
}

define <8 x i16> @insert_dup_elt3_mem_v8i16_i32(ptr %ptr) {
; SSE2-LABEL: insert_dup_elt3_mem_v8i16_i32:
; SSE2:       # %bb.0:
; SSE2-NEXT:    movd {{.*#+}} xmm0 = mem[0],zero,zero,zero
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,1,1,1,4,5,6,7]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: insert_dup_elt3_mem_v8i16_i32:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    movd {{.*#+}} xmm0 = mem[0],zero,zero,zero
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[2,3,2,3,2,3,2,3,2,3,2,3,2,3,2,3]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: insert_dup_elt3_mem_v8i16_i32:
; SSE41:       # %bb.0:
; SSE41-NEXT:    movd {{.*#+}} xmm0 = mem[0],zero,zero,zero
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[2,3,2,3,2,3,2,3,2,3,2,3,2,3,2,3]
; SSE41-NEXT:    retq
;
; AVX1-LABEL: insert_dup_elt3_mem_v8i16_i32:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vbroadcastss (%rdi), %xmm0
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[3,3,3,3,4,5,6,7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; AVX1-NEXT:    retq
;
; AVX2OR512VL-LABEL: insert_dup_elt3_mem_v8i16_i32:
; AVX2OR512VL:       # %bb.0:
; AVX2OR512VL-NEXT:    vpbroadcastw 2(%rdi), %xmm0
; AVX2OR512VL-NEXT:    retq
;
; XOPAVX1-LABEL: insert_dup_elt3_mem_v8i16_i32:
; XOPAVX1:       # %bb.0:
; XOPAVX1-NEXT:    vbroadcastss (%rdi), %xmm0
; XOPAVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[3,3,3,3,4,5,6,7]
; XOPAVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; XOPAVX1-NEXT:    retq
;
; XOPAVX2-LABEL: insert_dup_elt3_mem_v8i16_i32:
; XOPAVX2:       # %bb.0:
; XOPAVX2-NEXT:    vpbroadcastw 2(%rdi), %xmm0
; XOPAVX2-NEXT:    retq
  %tmp = load i32, ptr %ptr, align 4
  %tmp1 = insertelement <4 x i32> zeroinitializer, i32 %tmp, i32 1
  %tmp2 = bitcast <4 x i32> %tmp1 to <8 x i16>
  %tmp3 = shufflevector <8 x i16> %tmp2, <8 x i16> undef, <8 x i32> <i32 3, i32 3, i32 3, i32 3, i32 3, i32 3, i32 3, i32 3>
  ret <8 x i16> %tmp3
}

define <8 x i16> @insert_dup_elt1_mem_v8i16_sext_i16(ptr %ptr) {
; SSE-LABEL: insert_dup_elt1_mem_v8i16_sext_i16:
; SSE:       # %bb.0:
; SSE-NEXT:    movswl (%rdi), %eax
; SSE-NEXT:    movd %eax, %xmm0
; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,1,1,1,4,5,6,7]
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; SSE-NEXT:    retq
;
; AVX1-LABEL: insert_dup_elt1_mem_v8i16_sext_i16:
; AVX1:       # %bb.0:
; AVX1-NEXT:    movswl (%rdi), %eax
; AVX1-NEXT:    vmovd %eax, %xmm0
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[1,1,1,1,4,5,6,7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; AVX1-NEXT:    retq
;
; AVX2-LABEL: insert_dup_elt1_mem_v8i16_sext_i16:
; AVX2:       # %bb.0:
; AVX2-NEXT:    movswl (%rdi), %eax
; AVX2-NEXT:    shrl $16, %eax
; AVX2-NEXT:    vmovd %eax, %xmm0
; AVX2-NEXT:    vpbroadcastw %xmm0, %xmm0
; AVX2-NEXT:    retq
;
; AVX512VL-LABEL: insert_dup_elt1_mem_v8i16_sext_i16:
; AVX512VL:       # %bb.0:
; AVX512VL-NEXT:    movswl (%rdi), %eax
; AVX512VL-NEXT:    shrl $16, %eax
; AVX512VL-NEXT:    vpbroadcastw %eax, %xmm0
; AVX512VL-NEXT:    retq
;
; XOPAVX1-LABEL: insert_dup_elt1_mem_v8i16_sext_i16:
; XOPAVX1:       # %bb.0:
; XOPAVX1-NEXT:    movswl (%rdi), %eax
; XOPAVX1-NEXT:    vmovd %eax, %xmm0
; XOPAVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[1,1,1,1,4,5,6,7]
; XOPAVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; XOPAVX1-NEXT:    retq
;
; XOPAVX2-LABEL: insert_dup_elt1_mem_v8i16_sext_i16:
; XOPAVX2:       # %bb.0:
; XOPAVX2-NEXT:    movswl (%rdi), %eax
; XOPAVX2-NEXT:    shrl $16, %eax
; XOPAVX2-NEXT:    vmovd %eax, %xmm0
; XOPAVX2-NEXT:    vpbroadcastw %xmm0, %xmm0
; XOPAVX2-NEXT:    retq
  %tmp = load i16, ptr %ptr, align 2
  %tmp1 = sext i16 %tmp to i32
  %tmp2 = insertelement <4 x i32> zeroinitializer, i32 %tmp1, i32 0
  %tmp3 = bitcast <4 x i32> %tmp2 to <8 x i16>
  %tmp4 = shufflevector <8 x i16> %tmp3, <8 x i16> undef, <8 x i32> <i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1>
  ret <8 x i16> %tmp4
}

define <8 x i16> @insert_dup_elt3_mem_v8i16_sext_i16(ptr %ptr) {
; SSE2-LABEL: insert_dup_elt3_mem_v8i16_sext_i16:
; SSE2:       # %bb.0:
; SSE2-NEXT:    movswl (%rdi), %eax
; SSE2-NEXT:    movd %eax, %xmm0
; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,1,1,1,4,5,6,7]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: insert_dup_elt3_mem_v8i16_sext_i16:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    movswl (%rdi), %eax
; SSSE3-NEXT:    movd %eax, %xmm0
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[2,3,2,3,2,3,2,3,2,3,2,3,2,3,2,3]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: insert_dup_elt3_mem_v8i16_sext_i16:
; SSE41:       # %bb.0:
; SSE41-NEXT:    movswl (%rdi), %eax
; SSE41-NEXT:    movd %eax, %xmm0
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[2,3,2,3,2,3,2,3,2,3,2,3,2,3,2,3]
; SSE41-NEXT:    retq
;
; AVX1-LABEL: insert_dup_elt3_mem_v8i16_sext_i16:
; AVX1:       # %bb.0:
; AVX1-NEXT:    movswl (%rdi), %eax
; AVX1-NEXT:    vmovd %eax, %xmm0
; AVX1-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[2,3,2,3,2,3,2,3,2,3,2,3,2,3,2,3]
; AVX1-NEXT:    retq
;
; AVX2-LABEL: insert_dup_elt3_mem_v8i16_sext_i16:
; AVX2:       # %bb.0:
; AVX2-NEXT:    movswl (%rdi), %eax
; AVX2-NEXT:    shrl $16, %eax
; AVX2-NEXT:    vmovd %eax, %xmm0
; AVX2-NEXT:    vpbroadcastw %xmm0, %xmm0
; AVX2-NEXT:    retq
;
; AVX512VL-LABEL: insert_dup_elt3_mem_v8i16_sext_i16:
; AVX512VL:       # %bb.0:
; AVX512VL-NEXT:    movswl (%rdi), %eax
; AVX512VL-NEXT:    shrl $16, %eax
; AVX512VL-NEXT:    vpbroadcastw %eax, %xmm0
; AVX512VL-NEXT:    retq
;
; XOPAVX1-LABEL: insert_dup_elt3_mem_v8i16_sext_i16:
; XOPAVX1:       # %bb.0:
; XOPAVX1-NEXT:    movswl (%rdi), %eax
; XOPAVX1-NEXT:    vmovd %eax, %xmm0
; XOPAVX1-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[2,3,2,3,2,3,2,3,2,3,2,3,2,3,2,3]
; XOPAVX1-NEXT:    retq
;
; XOPAVX2-LABEL: insert_dup_elt3_mem_v8i16_sext_i16:
; XOPAVX2:       # %bb.0:
; XOPAVX2-NEXT:    movswl (%rdi), %eax
; XOPAVX2-NEXT:    shrl $16, %eax
; XOPAVX2-NEXT:    vmovd %eax, %xmm0
; XOPAVX2-NEXT:    vpbroadcastw %xmm0, %xmm0
; XOPAVX2-NEXT:    retq
  %tmp = load i16, ptr %ptr, align 2
  %tmp1 = sext i16 %tmp to i32
  %tmp2 = insertelement <4 x i32> zeroinitializer, i32 %tmp1, i32 1
  %tmp3 = bitcast <4 x i32> %tmp2 to <8 x i16>
  %tmp4 = shufflevector <8 x i16> %tmp3, <8 x i16> undef, <8 x i32> <i32 3, i32 3, i32 3, i32 3, i32 3, i32 3, i32 3, i32 3>
  ret <8 x i16> %tmp4
}

define <8 x i16> @insert_dup_mem_v8i16_i64(ptr %ptr) {
; SSE-LABEL: insert_dup_mem_v8i16_i64:
; SSE:       # %bb.0:
; SSE-NEXT:    movq {{.*#+}} xmm0 = mem[0],zero
; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; SSE-NEXT:    retq
;
; AVX1-LABEL: insert_dup_mem_v8i16_i64:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vmovq {{.*#+}} xmm0 = mem[0],zero
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; AVX1-NEXT:    retq
;
; AVX2OR512VL-LABEL: insert_dup_mem_v8i16_i64:
; AVX2OR512VL:       # %bb.0:
; AVX2OR512VL-NEXT:    vpbroadcastw (%rdi), %xmm0
; AVX2OR512VL-NEXT:    retq
;
; XOPAVX1-LABEL: insert_dup_mem_v8i16_i64:
; XOPAVX1:       # %bb.0:
; XOPAVX1-NEXT:    vmovq {{.*#+}} xmm0 = mem[0],zero
; XOPAVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
; XOPAVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; XOPAVX1-NEXT:    retq
;
; XOPAVX2-LABEL: insert_dup_mem_v8i16_i64:
; XOPAVX2:       # %bb.0:
; XOPAVX2-NEXT:    vpbroadcastw (%rdi), %xmm0
; XOPAVX2-NEXT:    retq
  %tmp = load i64, ptr %ptr, align 4
  %tmp1 = insertelement <2 x i64> zeroinitializer, i64 %tmp, i32 0
  %tmp2 = bitcast <2 x i64> %tmp1 to <8 x i16>
  %tmp3 = shufflevector <8 x i16> %tmp2, <8 x i16> undef, <8 x i32> zeroinitializer
  ret <8 x i16> %tmp3
}

define <8 x i16> @insert_dup_elt1_mem_v8i16_i64(ptr %ptr) {
; SSE-LABEL: insert_dup_elt1_mem_v8i16_i64:
; SSE:       # %bb.0:
; SSE-NEXT:    movq {{.*#+}} xmm0 = mem[0],zero
; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,1,1,1,4,5,6,7]
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; SSE-NEXT:    retq
;
; AVX1-LABEL: insert_dup_elt1_mem_v8i16_i64:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vmovq {{.*#+}} xmm0 = mem[0],zero
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[1,1,1,1,4,5,6,7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; AVX1-NEXT:    retq
;
; AVX2OR512VL-LABEL: insert_dup_elt1_mem_v8i16_i64:
; AVX2OR512VL:       # %bb.0:
; AVX2OR512VL-NEXT:    vpbroadcastw 2(%rdi), %xmm0
; AVX2OR512VL-NEXT:    retq
;
; XOPAVX1-LABEL: insert_dup_elt1_mem_v8i16_i64:
; XOPAVX1:       # %bb.0:
; XOPAVX1-NEXT:    vmovq {{.*#+}} xmm0 = mem[0],zero
; XOPAVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[1,1,1,1,4,5,6,7]
; XOPAVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; XOPAVX1-NEXT:    retq
;
; XOPAVX2-LABEL: insert_dup_elt1_mem_v8i16_i64:
; XOPAVX2:       # %bb.0:
; XOPAVX2-NEXT:    vpbroadcastw 2(%rdi), %xmm0
; XOPAVX2-NEXT:    retq
  %tmp = load i64, ptr %ptr, align 4
  %tmp1 = insertelement <2 x i64> zeroinitializer, i64 %tmp, i32 0
  %tmp2 = bitcast <2 x i64> %tmp1 to <8 x i16>
  %tmp3 = shufflevector <8 x i16> %tmp2, <8 x i16> undef, <8 x i32> <i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1>
  ret <8 x i16> %tmp3
}

define <8 x i16> @insert_dup_elt3_mem_v8i16_i64(ptr %ptr) {
; SSE-LABEL: insert_dup_elt3_mem_v8i16_i64:
; SSE:       # %bb.0:
; SSE-NEXT:    movq {{.*#+}} xmm0 = mem[0],zero
; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[3,3,3,3,4,5,6,7]
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; SSE-NEXT:    retq
;
; AVX1-LABEL: insert_dup_elt3_mem_v8i16_i64:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vmovq {{.*#+}} xmm0 = mem[0],zero
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[3,3,3,3,4,5,6,7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; AVX1-NEXT:    retq
;
; AVX2OR512VL-LABEL: insert_dup_elt3_mem_v8i16_i64:
; AVX2OR512VL:       # %bb.0:
; AVX2OR512VL-NEXT:    vpbroadcastw 6(%rdi), %xmm0
; AVX2OR512VL-NEXT:    retq
;
; XOPAVX1-LABEL: insert_dup_elt3_mem_v8i16_i64:
; XOPAVX1:       # %bb.0:
; XOPAVX1-NEXT:    vmovq {{.*#+}} xmm0 = mem[0],zero
; XOPAVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[3,3,3,3,4,5,6,7]
; XOPAVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; XOPAVX1-NEXT:    retq
;
; XOPAVX2-LABEL: insert_dup_elt3_mem_v8i16_i64:
; XOPAVX2:       # %bb.0:
; XOPAVX2-NEXT:    vpbroadcastw 6(%rdi), %xmm0
; XOPAVX2-NEXT:    retq
  %tmp = load i64, ptr %ptr, align 4
  %tmp1 = insertelement <2 x i64> zeroinitializer, i64 %tmp, i32 0
  %tmp2 = bitcast <2 x i64> %tmp1 to <8 x i16>
  %tmp3 = shufflevector <8 x i16> %tmp2, <8 x i16> undef, <8 x i32> <i32 3, i32 3, i32 3, i32 3, i32 3, i32 3, i32 3, i32 3>
  ret <8 x i16> %tmp3
}

define <8 x i16> @insert_dup_elt7_mem_v8i16_i64(ptr %ptr) {
; SSE2-LABEL: insert_dup_elt7_mem_v8i16_i64:
; SSE2:       # %bb.0:
; SSE2-NEXT:    movq {{.*#+}} xmm0 = mem[0],zero
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,1,0,1]
; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,7,7,7]
; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,2,2,2]
; SSE2-NEXT:    retq
;
; SSSE3-LABEL: insert_dup_elt7_mem_v8i16_i64:
; SSSE3:       # %bb.0:
; SSSE3-NEXT:    movq {{.*#+}} xmm0 = mem[0],zero
; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[6,7,6,7,6,7,6,7,6,7,6,7,6,7,6,7]
; SSSE3-NEXT:    retq
;
; SSE41-LABEL: insert_dup_elt7_mem_v8i16_i64:
; SSE41:       # %bb.0:
; SSE41-NEXT:    movq {{.*#+}} xmm0 = mem[0],zero
; SSE41-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[6,7,6,7,6,7,6,7,6,7,6,7,6,7,6,7]
; SSE41-NEXT:    retq
;
; AVX1-LABEL: insert_dup_elt7_mem_v8i16_i64:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vmovddup {{.*#+}} xmm0 = mem[0,0]
; AVX1-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,7,7,7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,2,2,2]
; AVX1-NEXT:    retq
;
; AVX2OR512VL-LABEL: insert_dup_elt7_mem_v8i16_i64:
; AVX2OR512VL:       # %bb.0:
; AVX2OR512VL-NEXT:    vpbroadcastw 6(%rdi), %xmm0
; AVX2OR512VL-NEXT:    retq
;
; XOPAVX1-LABEL: insert_dup_elt7_mem_v8i16_i64:
; XOPAVX1:       # %bb.0:
; XOPAVX1-NEXT:    vmovddup {{.*#+}} xmm0 = mem[0,0]
; XOPAVX1-NEXT:    vpshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,7,7,7]
; XOPAVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,2,2,2]
; XOPAVX1-NEXT:    retq
;
; XOPAVX2-LABEL: insert_dup_elt7_mem_v8i16_i64:
; XOPAVX2:       # %bb.0:
; XOPAVX2-NEXT:    vpbroadcastw 6(%rdi), %xmm0
; XOPAVX2-NEXT:    retq
  %tmp = load i64, ptr %ptr, align 4
  %tmp1 = insertelement <2 x i64> zeroinitializer, i64 %tmp, i32 1
  %tmp2 = bitcast <2 x i64> %tmp1 to <8 x i16>
  %tmp3 = shufflevector <8 x i16> %tmp2, <8 x i16> undef, <8 x i32> <i32 7, i32 7, i32 7, i32 7, i32 7, i32 7, i32 7, i32 7>
  ret <8 x i16> %tmp3
}

define <8 x i16> @insert_dup_mem_v8i16_sext_i16_i64(ptr %ptr) {
; SSE-LABEL: insert_dup_mem_v8i16_sext_i16_i64:
; SSE:       # %bb.0:
; SSE-NEXT:    movzwl (%rdi), %eax
; SSE-NEXT:    movd %eax, %xmm0
; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; SSE-NEXT:    retq
;
; AVX1-LABEL: insert_dup_mem_v8i16_sext_i16_i64:
; AVX1:       # %bb.0:
; AVX1-NEXT:    movzwl (%rdi), %eax
; AVX1-NEXT:    vmovd %eax, %xmm0
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; AVX1-NEXT:    retq
;
; AVX2OR512VL-LABEL: insert_dup_mem_v8i16_sext_i16_i64:
; AVX2OR512VL:       # %bb.0:
; AVX2OR512VL-NEXT:    vpbroadcastw (%rdi), %xmm0
; AVX2OR512VL-NEXT:    retq
;
; XOPAVX1-LABEL: insert_dup_mem_v8i16_sext_i16_i64:
; XOPAVX1:       # %bb.0:
; XOPAVX1-NEXT:    movzwl (%rdi), %eax
; XOPAVX1-NEXT:    vmovd %eax, %xmm0
; XOPAVX1-NEXT:    vpshuflw {{.*#+}} xmm0 = xmm0[0,0,0,0,4,5,6,7]
; XOPAVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; XOPAVX1-NEXT:    retq
;
; XOPAVX2-LABEL: insert_dup_mem_v8i16_sext_i16_i64:
; XOPAVX2:       # %bb.0:
; XOPAVX2-NEXT:    vpbroadcastw (%rdi), %xmm0
; XOPAVX2-NEXT:    retq
  %tmp = load i16, ptr %ptr, align 2
  %tmp1 = sext i16 %tmp to i64
  %tmp2 = insertelement <2 x i64> zeroinitializer, i64 %tmp1, i32 0
  %tmp3 = bitcast <2 x i64> %tmp2 to <8 x i16>
  %tmp4 = shufflevector <8 x i16> %tmp3, <8 x i16> undef, <8 x i32> zeroinitializer
  ret <8 x i16> %tmp4
}