; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=x86_64-- -mattr=+sse2 | FileCheck %s --check-prefixes=SSE
; RUN: llc < %s -mtriple=x86_64-- -mattr=+avx  | FileCheck %s --check-prefixes=AVX,AVX1
; RUN: llc < %s -mtriple=x86_64-- -mattr=+avx2 | FileCheck %s --check-prefixes=AVX,AVX2,AVX2-SLOW
; RUN: llc < %s -mtriple=x86_64-- -mattr=+avx2,+fast-variable-crosslane-shuffle,+fast-variable-perlane-shuffle | FileCheck %s --check-prefixes=AVX,AVX2,AVX2-FAST-ALL
; RUN: llc < %s -mtriple=x86_64-- -mattr=+avx2,+fast-variable-perlane-shuffle | FileCheck %s --check-prefixes=AVX,AVX2,AVX2-FAST-PERLANE
; RUN: llc < %s -mtriple=x86_64-- -mattr=+avx512bw,+avx512vl | FileCheck %s --check-prefixes=AVX512

; These patterns are produced by LoopVectorizer for interleaved stores.

define void @vf2(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vecptr2, ptr %in.vecptr3, ptr %out.vec) nounwind {
; SSE-LABEL: vf2:
; SSE:       # %bb.0:
; SSE-NEXT:    movdqa (%rdi), %xmm0
; SSE-NEXT:    movdqa (%rdx), %xmm1
; SSE-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],mem[0],xmm0[1],mem[1]
; SSE-NEXT:    punpckldq {{.*#+}} xmm1 = xmm1[0],mem[0],xmm1[1],mem[1]
; SSE-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,1,3]
; SSE-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[0,2,1,3,4,5,6,7]
; SSE-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,4,6,5,7]
; SSE-NEXT:    movdqa %xmm0, (%r8)
; SSE-NEXT:    retq
;
; AVX-LABEL: vf2:
; AVX:       # %bb.0:
; AVX-NEXT:    vmovdqa (%rdi), %xmm0
; AVX-NEXT:    vmovdqa (%rdx), %xmm1
; AVX-NEXT:    vpunpckldq {{.*#+}} xmm0 = xmm0[0],mem[0],xmm0[1],mem[1]
; AVX-NEXT:    vpunpckldq {{.*#+}} xmm1 = xmm1[0],mem[0],xmm1[1],mem[1]
; AVX-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; AVX-NEXT:    vpshufb {{.*#+}} xmm0 = xmm0[0,1,8,9,2,3,10,11,4,5,12,13,6,7,14,15]
; AVX-NEXT:    vmovdqa %xmm0, (%r8)
; AVX-NEXT:    retq
;
; AVX512-LABEL: vf2:
; AVX512:       # %bb.0:
; AVX512-NEXT:    vmovdqa (%rdi), %xmm0
; AVX512-NEXT:    vmovdqa (%rdx), %xmm1
; AVX512-NEXT:    vpunpckldq {{.*#+}} xmm0 = xmm0[0],mem[0],xmm0[1],mem[1]
; AVX512-NEXT:    vpunpckldq {{.*#+}} xmm1 = xmm1[0],mem[0],xmm1[1],mem[1]
; AVX512-NEXT:    vmovdqa {{.*#+}} xmm2 = [0,2,8,10,1,3,9,11]
; AVX512-NEXT:    vpermi2w %xmm1, %xmm0, %xmm2
; AVX512-NEXT:    vmovdqa %xmm2, (%r8)
; AVX512-NEXT:    retq
  %in.vec0 = load <2 x i16>, ptr %in.vecptr0, align 32
  %in.vec1 = load <2 x i16>, ptr %in.vecptr1, align 32
  %in.vec2 = load <2 x i16>, ptr %in.vecptr2, align 32
  %in.vec3 = load <2 x i16>, ptr %in.vecptr3, align 32

  %concat01 = shufflevector <2 x i16> %in.vec0, <2 x i16> %in.vec1, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
  %concat23 = shufflevector <2 x i16> %in.vec2, <2 x i16> %in.vec3, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
  %concat0123 = shufflevector <4 x i16> %concat01, <4 x i16> %concat23, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
  %interleaved.vec = shufflevector <8 x i16> %concat0123, <8 x i16> poison, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 1, i32 3, i32 5, i32 7>

  store <8 x i16> %interleaved.vec, ptr %out.vec, align 32

  ret void
}

define void @vf4(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vecptr2, ptr %in.vecptr3, ptr %out.vec) nounwind {
; SSE-LABEL: vf4:
; SSE:       # %bb.0:
; SSE-NEXT:    movq {{.*#+}} xmm0 = mem[0],zero
; SSE-NEXT:    movq {{.*#+}} xmm1 = mem[0],zero
; SSE-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; SSE-NEXT:    movq {{.*#+}} xmm1 = mem[0],zero
; SSE-NEXT:    movq {{.*#+}} xmm2 = mem[0],zero
; SSE-NEXT:    punpcklwd {{.*#+}} xmm1 = xmm1[0],xmm2[0],xmm1[1],xmm2[1],xmm1[2],xmm2[2],xmm1[3],xmm2[3]
; SSE-NEXT:    movdqa %xmm0, %xmm2
; SSE-NEXT:    punpckldq {{.*#+}} xmm2 = xmm2[0],xmm1[0],xmm2[1],xmm1[1]
; SSE-NEXT:    punpckhdq {{.*#+}} xmm0 = xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; SSE-NEXT:    movdqa %xmm0, 16(%r8)
; SSE-NEXT:    movdqa %xmm2, (%r8)
; SSE-NEXT:    retq
;
; AVX1-LABEL: vf4:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vmovq {{.*#+}} xmm0 = mem[0],zero
; AVX1-NEXT:    vmovq {{.*#+}} xmm1 = mem[0],zero
; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm1[0],xmm0[0]
; AVX1-NEXT:    vmovq {{.*#+}} xmm3 = mem[0],zero
; AVX1-NEXT:    vmovq {{.*#+}} xmm4 = mem[0],zero
; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm5 = xmm4[0],xmm3[0]
; AVX1-NEXT:    vpunpcklwd {{.*#+}} xmm3 = xmm4[0],xmm3[0],xmm4[1],xmm3[1],xmm4[2],xmm3[2],xmm4[3],xmm3[3]
; AVX1-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm1[0],xmm0[0],xmm1[1],xmm0[1],xmm1[2],xmm0[2],xmm1[3],xmm0[3]
; AVX1-NEXT:    vpunpckldq {{.*#+}} xmm0 = xmm0[0],xmm3[0],xmm0[1],xmm3[1]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm1 = xmm5[3,1,2,3]
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm1 = xmm1[2,0,3,1,4,5,6,7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[3,1,2,3]
; AVX1-NEXT:    vpshuflw {{.*#+}} xmm2 = xmm2[2,0,3,1,4,5,6,7]
; AVX1-NEXT:    vpunpckldq {{.*#+}} xmm1 = xmm2[0],xmm1[0],xmm2[1],xmm1[1]
; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
; AVX1-NEXT:    vmovaps %ymm0, (%r8)
; AVX1-NEXT:    vzeroupper
; AVX1-NEXT:    retq
;
; AVX2-SLOW-LABEL: vf4:
; AVX2-SLOW:       # %bb.0:
; AVX2-SLOW-NEXT:    vmovq {{.*#+}} xmm0 = mem[0],zero
; AVX2-SLOW-NEXT:    vmovq {{.*#+}} xmm1 = mem[0],zero
; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm1[0],xmm0[0]
; AVX2-SLOW-NEXT:    vmovq {{.*#+}} xmm1 = mem[0],zero
; AVX2-SLOW-NEXT:    vmovq {{.*#+}} xmm2 = mem[0],zero
; AVX2-SLOW-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm2[0],xmm1[0]
; AVX2-SLOW-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
; AVX2-SLOW-NEXT:    vpshufb {{.*#+}} ymm1 = ymm0[0,1,8,9,u,u,u,u,2,3,10,11,u,u,u,u,u,u,u,u,20,21,28,29,u,u,u,u,22,23,30,31]
; AVX2-SLOW-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[2,3,0,1]
; AVX2-SLOW-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[u,u,u,u,0,1,8,9,u,u,u,u,2,3,10,11,20,21,28,29,u,u,u,u,22,23,30,31,u,u,u,u]
; AVX2-SLOW-NEXT:    vpblendd {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2],ymm0[3,4],ymm1[5],ymm0[6],ymm1[7]
; AVX2-SLOW-NEXT:    vmovdqa %ymm0, (%r8)
; AVX2-SLOW-NEXT:    vzeroupper
; AVX2-SLOW-NEXT:    retq
;
; AVX2-FAST-ALL-LABEL: vf4:
; AVX2-FAST-ALL:       # %bb.0:
; AVX2-FAST-ALL-NEXT:    vmovq {{.*#+}} xmm0 = mem[0],zero
; AVX2-FAST-ALL-NEXT:    vmovq {{.*#+}} xmm1 = mem[0],zero
; AVX2-FAST-ALL-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm1[0],xmm0[0]
; AVX2-FAST-ALL-NEXT:    vmovq {{.*#+}} xmm1 = mem[0],zero
; AVX2-FAST-ALL-NEXT:    vmovq {{.*#+}} xmm2 = mem[0],zero
; AVX2-FAST-ALL-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm2[0],xmm1[0]
; AVX2-FAST-ALL-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
; AVX2-FAST-ALL-NEXT:    vmovdqa {{.*#+}} ymm1 = [0,2,4,6,1,3,5,7]
; AVX2-FAST-ALL-NEXT:    vpermd %ymm0, %ymm1, %ymm0
; AVX2-FAST-ALL-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[0,1,4,5,8,9,12,13,2,3,6,7,10,11,14,15,16,17,20,21,24,25,28,29,18,19,22,23,26,27,30,31]
; AVX2-FAST-ALL-NEXT:    vmovdqa %ymm0, (%r8)
; AVX2-FAST-ALL-NEXT:    vzeroupper
; AVX2-FAST-ALL-NEXT:    retq
;
; AVX2-FAST-PERLANE-LABEL: vf4:
; AVX2-FAST-PERLANE:       # %bb.0:
; AVX2-FAST-PERLANE-NEXT:    vmovq {{.*#+}} xmm0 = mem[0],zero
; AVX2-FAST-PERLANE-NEXT:    vmovq {{.*#+}} xmm1 = mem[0],zero
; AVX2-FAST-PERLANE-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm1[0],xmm0[0]
; AVX2-FAST-PERLANE-NEXT:    vmovq {{.*#+}} xmm1 = mem[0],zero
; AVX2-FAST-PERLANE-NEXT:    vmovq {{.*#+}} xmm2 = mem[0],zero
; AVX2-FAST-PERLANE-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm2[0],xmm1[0]
; AVX2-FAST-PERLANE-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
; AVX2-FAST-PERLANE-NEXT:    vpshufb {{.*#+}} ymm1 = ymm0[0,1,8,9,u,u,u,u,2,3,10,11,u,u,u,u,u,u,u,u,20,21,28,29,u,u,u,u,22,23,30,31]
; AVX2-FAST-PERLANE-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[2,3,0,1]
; AVX2-FAST-PERLANE-NEXT:    vpshufb {{.*#+}} ymm0 = ymm0[u,u,u,u,0,1,8,9,u,u,u,u,2,3,10,11,20,21,28,29,u,u,u,u,22,23,30,31,u,u,u,u]
; AVX2-FAST-PERLANE-NEXT:    vpblendd {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2],ymm0[3,4],ymm1[5],ymm0[6],ymm1[7]
; AVX2-FAST-PERLANE-NEXT:    vmovdqa %ymm0, (%r8)
; AVX2-FAST-PERLANE-NEXT:    vzeroupper
; AVX2-FAST-PERLANE-NEXT:    retq
;
; AVX512-LABEL: vf4:
; AVX512:       # %bb.0:
; AVX512-NEXT:    vmovq {{.*#+}} xmm0 = mem[0],zero
; AVX512-NEXT:    vmovq {{.*#+}} xmm1 = mem[0],zero
; AVX512-NEXT:    vpunpcklqdq {{.*#+}} xmm0 = xmm1[0],xmm0[0]
; AVX512-NEXT:    vmovq {{.*#+}} xmm1 = mem[0],zero
; AVX512-NEXT:    vmovq {{.*#+}} xmm2 = mem[0],zero
; AVX512-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm2[0],xmm1[0]
; AVX512-NEXT:    vinserti128 $1, %xmm1, %ymm0, %ymm0
; AVX512-NEXT:    vmovdqa {{.*#+}} ymm1 = [0,4,8,12,1,5,9,13,2,6,10,14,3,7,11,15]
; AVX512-NEXT:    vpermw %ymm0, %ymm1, %ymm0
; AVX512-NEXT:    vmovdqa %ymm0, (%r8)
; AVX512-NEXT:    vzeroupper
; AVX512-NEXT:    retq
  %in.vec0 = load <4 x i16>, ptr %in.vecptr0, align 32
  %in.vec1 = load <4 x i16>, ptr %in.vecptr1, align 32
  %in.vec2 = load <4 x i16>, ptr %in.vecptr2, align 32
  %in.vec3 = load <4 x i16>, ptr %in.vecptr3, align 32

  %concat01 = shufflevector <4 x i16> %in.vec0, <4 x i16> %in.vec1, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
  %concat23 = shufflevector <4 x i16> %in.vec2, <4 x i16> %in.vec3, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
  %concat0123 = shufflevector <8 x i16> %concat01, <8 x i16> %concat23, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
  %interleaved.vec = shufflevector <16 x i16> %concat0123, <16 x i16> poison, <16 x i32> <i32 0, i32 4, i32 8, i32 12, i32 1, i32 5, i32 9, i32 13, i32 2, i32 6, i32 10, i32 14, i32 3, i32 7, i32 11, i32 15>

  store <16 x i16> %interleaved.vec, ptr %out.vec, align 32

  ret void
}

define void @vf8(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vecptr2, ptr %in.vecptr3, ptr %out.vec) nounwind {
; SSE-LABEL: vf8:
; SSE:       # %bb.0:
; SSE-NEXT:    movdqa (%rdi), %xmm0
; SSE-NEXT:    movdqa (%rsi), %xmm1
; SSE-NEXT:    movdqa (%rdx), %xmm2
; SSE-NEXT:    movdqa (%rcx), %xmm3
; SSE-NEXT:    movdqa %xmm2, %xmm4
; SSE-NEXT:    punpcklwd {{.*#+}} xmm4 = xmm4[0],xmm3[0],xmm4[1],xmm3[1],xmm4[2],xmm3[2],xmm4[3],xmm3[3]
; SSE-NEXT:    movdqa %xmm0, %xmm5
; SSE-NEXT:    punpcklwd {{.*#+}} xmm5 = xmm5[0],xmm1[0],xmm5[1],xmm1[1],xmm5[2],xmm1[2],xmm5[3],xmm1[3]
; SSE-NEXT:    movdqa %xmm5, %xmm6
; SSE-NEXT:    punpckldq {{.*#+}} xmm6 = xmm6[0],xmm4[0],xmm6[1],xmm4[1]
; SSE-NEXT:    punpckhdq {{.*#+}} xmm5 = xmm5[2],xmm4[2],xmm5[3],xmm4[3]
; SSE-NEXT:    punpckhwd {{.*#+}} xmm2 = xmm2[4],xmm3[4],xmm2[5],xmm3[5],xmm2[6],xmm3[6],xmm2[7],xmm3[7]
; SSE-NEXT:    punpckhwd {{.*#+}} xmm0 = xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
; SSE-NEXT:    movdqa %xmm0, %xmm1
; SSE-NEXT:    punpckldq {{.*#+}} xmm1 = xmm1[0],xmm2[0],xmm1[1],xmm2[1]
; SSE-NEXT:    punpckhdq {{.*#+}} xmm0 = xmm0[2],xmm2[2],xmm0[3],xmm2[3]
; SSE-NEXT:    movdqa %xmm0, 48(%r8)
; SSE-NEXT:    movdqa %xmm1, 32(%r8)
; SSE-NEXT:    movdqa %xmm5, 16(%r8)
; SSE-NEXT:    movdqa %xmm6, (%r8)
; SSE-NEXT:    retq
;
; AVX1-LABEL: vf8:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vmovdqa (%rdi), %xmm0
; AVX1-NEXT:    vmovdqa (%rsi), %xmm1
; AVX1-NEXT:    vmovdqa (%rdx), %xmm2
; AVX1-NEXT:    vmovdqa (%rcx), %xmm3
; AVX1-NEXT:    vpunpcklwd {{.*#+}} xmm4 = xmm2[0],xmm3[0],xmm2[1],xmm3[1],xmm2[2],xmm3[2],xmm2[3],xmm3[3]
; AVX1-NEXT:    vpunpcklwd {{.*#+}} xmm5 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
; AVX1-NEXT:    vpunpckhdq {{.*#+}} xmm6 = xmm5[2],xmm4[2],xmm5[3],xmm4[3]
; AVX1-NEXT:    vpunpckldq {{.*#+}} xmm4 = xmm5[0],xmm4[0],xmm5[1],xmm4[1]
; AVX1-NEXT:    vinsertf128 $1, %xmm6, %ymm4, %ymm4
; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm2 = xmm2[4],xmm3[4],xmm2[5],xmm3[5],xmm2[6],xmm3[6],xmm2[7],xmm3[7]
; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm0 = xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
; AVX1-NEXT:    vpunpckhdq {{.*#+}} xmm1 = xmm0[2],xmm2[2],xmm0[3],xmm2[3]
; AVX1-NEXT:    vpunpckldq {{.*#+}} xmm0 = xmm0[0],xmm2[0],xmm0[1],xmm2[1]
; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
; AVX1-NEXT:    vmovaps %ymm0, 32(%r8)
; AVX1-NEXT:    vmovaps %ymm4, (%r8)
; AVX1-NEXT:    vzeroupper
; AVX1-NEXT:    retq
;
; AVX2-LABEL: vf8:
; AVX2:       # %bb.0:
; AVX2-NEXT:    vmovdqa (%rdi), %xmm0
; AVX2-NEXT:    vmovdqa (%rdx), %xmm1
; AVX2-NEXT:    vinserti128 $1, (%rsi), %ymm0, %ymm0
; AVX2-NEXT:    vinserti128 $1, (%rcx), %ymm1, %ymm1
; AVX2-NEXT:    vpermq {{.*#+}} ymm2 = ymm1[0,2,0,2]
; AVX2-NEXT:    vmovdqa {{.*#+}} ymm3 = <u,u,u,u,0,1,8,9,u,u,u,u,2,3,10,11,u,u,u,u,4,5,12,13,u,u,u,u,6,7,14,15>
; AVX2-NEXT:    vpshufb %ymm3, %ymm2, %ymm2
; AVX2-NEXT:    vpermq {{.*#+}} ymm4 = ymm0[0,2,0,2]
; AVX2-NEXT:    vmovdqa {{.*#+}} ymm5 = <0,1,8,9,u,u,u,u,2,3,10,11,u,u,u,u,4,5,12,13,u,u,u,u,6,7,14,15,u,u,u,u>
; AVX2-NEXT:    vpshufb %ymm5, %ymm4, %ymm4
; AVX2-NEXT:    vpblendd {{.*#+}} ymm2 = ymm4[0],ymm2[1],ymm4[2],ymm2[3],ymm4[4],ymm2[5],ymm4[6],ymm2[7]
; AVX2-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[1,3,1,3]
; AVX2-NEXT:    vpshufb %ymm3, %ymm1, %ymm1
; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[1,3,1,3]
; AVX2-NEXT:    vpshufb %ymm5, %ymm0, %ymm0
; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2],ymm1[3],ymm0[4],ymm1[5],ymm0[6],ymm1[7]
; AVX2-NEXT:    vmovdqa %ymm0, 32(%r8)
; AVX2-NEXT:    vmovdqa %ymm2, (%r8)
; AVX2-NEXT:    vzeroupper
; AVX2-NEXT:    retq
;
; AVX512-LABEL: vf8:
; AVX512:       # %bb.0:
; AVX512-NEXT:    vmovdqa (%rdi), %xmm0
; AVX512-NEXT:    vmovdqa (%rdx), %xmm1
; AVX512-NEXT:    vinserti128 $1, (%rcx), %ymm1, %ymm1
; AVX512-NEXT:    vinserti128 $1, (%rsi), %ymm0, %ymm0
; AVX512-NEXT:    vinserti64x4 $1, %ymm1, %zmm0, %zmm0
; AVX512-NEXT:    vmovdqa64 {{.*#+}} zmm1 = [0,8,16,24,1,9,17,25,2,10,18,26,3,11,19,27,4,12,20,28,5,13,21,29,6,14,22,30,7,15,23,31]
; AVX512-NEXT:    vpermw %zmm0, %zmm1, %zmm0
; AVX512-NEXT:    vmovdqu64 %zmm0, (%r8)
; AVX512-NEXT:    vzeroupper
; AVX512-NEXT:    retq
  %in.vec0 = load <8 x i16>, ptr %in.vecptr0, align 32
  %in.vec1 = load <8 x i16>, ptr %in.vecptr1, align 32
  %in.vec2 = load <8 x i16>, ptr %in.vecptr2, align 32
  %in.vec3 = load <8 x i16>, ptr %in.vecptr3, align 32

  %concat01 = shufflevector <8 x i16> %in.vec0, <8 x i16> %in.vec1, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
  %concat23 = shufflevector <8 x i16> %in.vec2, <8 x i16> %in.vec3, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
  %concat0123 = shufflevector <16 x i16> %concat01, <16 x i16> %concat23, <32 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31>
  %interleaved.vec = shufflevector <32 x i16> %concat0123, <32 x i16> poison, <32 x i32> <i32 0, i32 8, i32 16, i32 24, i32 1, i32 9, i32 17, i32 25, i32 2, i32 10, i32 18, i32 26, i32 3, i32 11, i32 19, i32 27, i32 4, i32 12, i32 20, i32 28, i32 5, i32 13, i32 21, i32 29, i32 6, i32 14, i32 22, i32 30, i32 7, i32 15, i32 23, i32 31>

  store <32 x i16> %interleaved.vec, ptr %out.vec, align 32

  ret void
}

define void @vf16(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vecptr2, ptr %in.vecptr3, ptr %out.vec) nounwind {
; SSE-LABEL: vf16:
; SSE:       # %bb.0:
; SSE-NEXT:    movdqa (%rdi), %xmm0
; SSE-NEXT:    movdqa 16(%rdi), %xmm1
; SSE-NEXT:    movdqa (%rsi), %xmm5
; SSE-NEXT:    movdqa 16(%rsi), %xmm8
; SSE-NEXT:    movdqa (%rdx), %xmm3
; SSE-NEXT:    movdqa 16(%rdx), %xmm4
; SSE-NEXT:    movdqa (%rcx), %xmm6
; SSE-NEXT:    movdqa 16(%rcx), %xmm9
; SSE-NEXT:    movdqa %xmm3, %xmm7
; SSE-NEXT:    punpcklwd {{.*#+}} xmm7 = xmm7[0],xmm6[0],xmm7[1],xmm6[1],xmm7[2],xmm6[2],xmm7[3],xmm6[3]
; SSE-NEXT:    movdqa %xmm0, %xmm2
; SSE-NEXT:    punpcklwd {{.*#+}} xmm2 = xmm2[0],xmm5[0],xmm2[1],xmm5[1],xmm2[2],xmm5[2],xmm2[3],xmm5[3]
; SSE-NEXT:    movdqa %xmm2, %xmm10
; SSE-NEXT:    punpckhdq {{.*#+}} xmm10 = xmm10[2],xmm7[2],xmm10[3],xmm7[3]
; SSE-NEXT:    punpckldq {{.*#+}} xmm2 = xmm2[0],xmm7[0],xmm2[1],xmm7[1]
; SSE-NEXT:    punpckhwd {{.*#+}} xmm3 = xmm3[4],xmm6[4],xmm3[5],xmm6[5],xmm3[6],xmm6[6],xmm3[7],xmm6[7]
; SSE-NEXT:    punpckhwd {{.*#+}} xmm0 = xmm0[4],xmm5[4],xmm0[5],xmm5[5],xmm0[6],xmm5[6],xmm0[7],xmm5[7]
; SSE-NEXT:    movdqa %xmm0, %xmm5
; SSE-NEXT:    punpckhdq {{.*#+}} xmm5 = xmm5[2],xmm3[2],xmm5[3],xmm3[3]
; SSE-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm3[0],xmm0[1],xmm3[1]
; SSE-NEXT:    movdqa %xmm4, %xmm3
; SSE-NEXT:    punpcklwd {{.*#+}} xmm3 = xmm3[0],xmm9[0],xmm3[1],xmm9[1],xmm3[2],xmm9[2],xmm3[3],xmm9[3]
; SSE-NEXT:    movdqa %xmm1, %xmm6
; SSE-NEXT:    punpcklwd {{.*#+}} xmm6 = xmm6[0],xmm8[0],xmm6[1],xmm8[1],xmm6[2],xmm8[2],xmm6[3],xmm8[3]
; SSE-NEXT:    movdqa %xmm6, %xmm7
; SSE-NEXT:    punpckhdq {{.*#+}} xmm7 = xmm7[2],xmm3[2],xmm7[3],xmm3[3]
; SSE-NEXT:    punpckldq {{.*#+}} xmm6 = xmm6[0],xmm3[0],xmm6[1],xmm3[1]
; SSE-NEXT:    punpckhwd {{.*#+}} xmm4 = xmm4[4],xmm9[4],xmm4[5],xmm9[5],xmm4[6],xmm9[6],xmm4[7],xmm9[7]
; SSE-NEXT:    punpckhwd {{.*#+}} xmm1 = xmm1[4],xmm8[4],xmm1[5],xmm8[5],xmm1[6],xmm8[6],xmm1[7],xmm8[7]
; SSE-NEXT:    movdqa %xmm1, %xmm3
; SSE-NEXT:    punpckhdq {{.*#+}} xmm3 = xmm3[2],xmm4[2],xmm3[3],xmm4[3]
; SSE-NEXT:    punpckldq {{.*#+}} xmm1 = xmm1[0],xmm4[0],xmm1[1],xmm4[1]
; SSE-NEXT:    movdqa %xmm1, 96(%r8)
; SSE-NEXT:    movdqa %xmm3, 112(%r8)
; SSE-NEXT:    movdqa %xmm6, 64(%r8)
; SSE-NEXT:    movdqa %xmm7, 80(%r8)
; SSE-NEXT:    movdqa %xmm0, 32(%r8)
; SSE-NEXT:    movdqa %xmm5, 48(%r8)
; SSE-NEXT:    movdqa %xmm2, (%r8)
; SSE-NEXT:    movdqa %xmm10, 16(%r8)
; SSE-NEXT:    retq
;
; AVX1-LABEL: vf16:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vmovdqa (%rcx), %xmm8
; AVX1-NEXT:    vmovdqa 16(%rcx), %xmm5
; AVX1-NEXT:    vmovdqa (%rdx), %xmm9
; AVX1-NEXT:    vmovdqa 16(%rdx), %xmm6
; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm2 = xmm9[4],xmm8[4],xmm9[5],xmm8[5],xmm9[6],xmm8[6],xmm9[7],xmm8[7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm3 = xmm2[0,0,1,1]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm3, %ymm3
; AVX1-NEXT:    vmovdqa (%rsi), %xmm2
; AVX1-NEXT:    vmovdqa 16(%rsi), %xmm7
; AVX1-NEXT:    vmovdqa (%rdi), %xmm4
; AVX1-NEXT:    vmovdqa 16(%rdi), %xmm0
; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm1 = xmm4[4],xmm2[4],xmm4[5],xmm2[5],xmm4[6],xmm2[6],xmm4[7],xmm2[7]
; AVX1-NEXT:    vpmovzxdq {{.*#+}} xmm10 = xmm1[0],zero,xmm1[1],zero
; AVX1-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm10, %ymm1
; AVX1-NEXT:    vblendps {{.*#+}} ymm11 = ymm1[0],ymm3[1],ymm1[2],ymm3[3],ymm1[4],ymm3[5],ymm1[6],ymm3[7]
; AVX1-NEXT:    vpunpcklwd {{.*#+}} xmm1 = xmm6[0],xmm5[0],xmm6[1],xmm5[1],xmm6[2],xmm5[2],xmm6[3],xmm5[3]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm10 = xmm1[0,0,1,1]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm10, %ymm1
; AVX1-NEXT:    vpunpcklwd {{.*#+}} xmm3 = xmm0[0],xmm7[0],xmm0[1],xmm7[1],xmm0[2],xmm7[2],xmm0[3],xmm7[3]
; AVX1-NEXT:    vpmovzxdq {{.*#+}} xmm10 = xmm3[0],zero,xmm3[1],zero
; AVX1-NEXT:    vpshufd {{.*#+}} xmm3 = xmm3[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm3, %ymm10, %ymm3
; AVX1-NEXT:    vblendps {{.*#+}} ymm1 = ymm3[0],ymm1[1],ymm3[2],ymm1[3],ymm3[4],ymm1[5],ymm3[6],ymm1[7]
; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm3 = xmm6[4],xmm5[4],xmm6[5],xmm5[5],xmm6[6],xmm5[6],xmm6[7],xmm5[7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm5 = xmm3[0,0,1,1]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm3 = xmm3[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm3, %ymm5, %ymm3
; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm0 = xmm0[4],xmm7[4],xmm0[5],xmm7[5],xmm0[6],xmm7[6],xmm0[7],xmm7[7]
; AVX1-NEXT:    vpmovzxdq {{.*#+}} xmm5 = xmm0[0],zero,xmm0[1],zero
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm5, %ymm0
; AVX1-NEXT:    vblendps {{.*#+}} ymm0 = ymm0[0],ymm3[1],ymm0[2],ymm3[3],ymm0[4],ymm3[5],ymm0[6],ymm3[7]
; AVX1-NEXT:    vpunpcklwd {{.*#+}} xmm3 = xmm9[0],xmm8[0],xmm9[1],xmm8[1],xmm9[2],xmm8[2],xmm9[3],xmm8[3]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm5 = xmm3[0,0,1,1]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm3 = xmm3[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm3, %ymm5, %ymm3
; AVX1-NEXT:    vpunpcklwd {{.*#+}} xmm2 = xmm4[0],xmm2[0],xmm4[1],xmm2[1],xmm4[2],xmm2[2],xmm4[3],xmm2[3]
; AVX1-NEXT:    vpmovzxdq {{.*#+}} xmm4 = xmm2[0],zero,xmm2[1],zero
; AVX1-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm4, %ymm2
; AVX1-NEXT:    vblendps {{.*#+}} ymm2 = ymm2[0],ymm3[1],ymm2[2],ymm3[3],ymm2[4],ymm3[5],ymm2[6],ymm3[7]
; AVX1-NEXT:    vmovaps %ymm2, (%r8)
; AVX1-NEXT:    vmovaps %ymm0, 96(%r8)
; AVX1-NEXT:    vmovaps %ymm1, 64(%r8)
; AVX1-NEXT:    vmovaps %ymm11, 32(%r8)
; AVX1-NEXT:    vzeroupper
; AVX1-NEXT:    retq
;
; AVX2-LABEL: vf16:
; AVX2:       # %bb.0:
; AVX2-NEXT:    vmovdqa (%rcx), %xmm8
; AVX2-NEXT:    vmovdqa 16(%rcx), %xmm5
; AVX2-NEXT:    vmovdqa (%rdx), %xmm9
; AVX2-NEXT:    vmovdqa 16(%rdx), %xmm6
; AVX2-NEXT:    vpunpckhwd {{.*#+}} xmm2 = xmm9[4],xmm8[4],xmm9[5],xmm8[5],xmm9[6],xmm8[6],xmm9[7],xmm8[7]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm3 = xmm2[0,0,1,1]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm2, %ymm3, %ymm3
; AVX2-NEXT:    vmovdqa (%rsi), %xmm2
; AVX2-NEXT:    vmovdqa 16(%rsi), %xmm7
; AVX2-NEXT:    vmovdqa (%rdi), %xmm4
; AVX2-NEXT:    vmovdqa 16(%rdi), %xmm0
; AVX2-NEXT:    vpunpckhwd {{.*#+}} xmm1 = xmm4[4],xmm2[4],xmm4[5],xmm2[5],xmm4[6],xmm2[6],xmm4[7],xmm2[7]
; AVX2-NEXT:    vpmovzxdq {{.*#+}} xmm10 = xmm1[0],zero,xmm1[1],zero
; AVX2-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm10, %ymm1
; AVX2-NEXT:    vpblendd {{.*#+}} ymm11 = ymm1[0],ymm3[1],ymm1[2],ymm3[3],ymm1[4],ymm3[5],ymm1[6],ymm3[7]
; AVX2-NEXT:    vpunpcklwd {{.*#+}} xmm1 = xmm6[0],xmm5[0],xmm6[1],xmm5[1],xmm6[2],xmm5[2],xmm6[3],xmm5[3]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm10 = xmm1[0,0,1,1]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm10, %ymm1
; AVX2-NEXT:    vpunpcklwd {{.*#+}} xmm3 = xmm0[0],xmm7[0],xmm0[1],xmm7[1],xmm0[2],xmm7[2],xmm0[3],xmm7[3]
; AVX2-NEXT:    vpmovzxdq {{.*#+}} xmm10 = xmm3[0],zero,xmm3[1],zero
; AVX2-NEXT:    vpshufd {{.*#+}} xmm3 = xmm3[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm3, %ymm10, %ymm3
; AVX2-NEXT:    vpblendd {{.*#+}} ymm1 = ymm3[0],ymm1[1],ymm3[2],ymm1[3],ymm3[4],ymm1[5],ymm3[6],ymm1[7]
; AVX2-NEXT:    vpunpckhwd {{.*#+}} xmm3 = xmm6[4],xmm5[4],xmm6[5],xmm5[5],xmm6[6],xmm5[6],xmm6[7],xmm5[7]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm5 = xmm3[0,0,1,1]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm3 = xmm3[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm3, %ymm5, %ymm3
; AVX2-NEXT:    vpunpckhwd {{.*#+}} xmm0 = xmm0[4],xmm7[4],xmm0[5],xmm7[5],xmm0[6],xmm7[6],xmm0[7],xmm7[7]
; AVX2-NEXT:    vpmovzxdq {{.*#+}} xmm5 = xmm0[0],zero,xmm0[1],zero
; AVX2-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm0, %ymm5, %ymm0
; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0],ymm3[1],ymm0[2],ymm3[3],ymm0[4],ymm3[5],ymm0[6],ymm3[7]
; AVX2-NEXT:    vpunpcklwd {{.*#+}} xmm3 = xmm9[0],xmm8[0],xmm9[1],xmm8[1],xmm9[2],xmm8[2],xmm9[3],xmm8[3]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm5 = xmm3[0,0,1,1]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm3 = xmm3[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm3, %ymm5, %ymm3
; AVX2-NEXT:    vpunpcklwd {{.*#+}} xmm2 = xmm4[0],xmm2[0],xmm4[1],xmm2[1],xmm4[2],xmm2[2],xmm4[3],xmm2[3]
; AVX2-NEXT:    vpmovzxdq {{.*#+}} xmm4 = xmm2[0],zero,xmm2[1],zero
; AVX2-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm2, %ymm4, %ymm2
; AVX2-NEXT:    vpblendd {{.*#+}} ymm2 = ymm2[0],ymm3[1],ymm2[2],ymm3[3],ymm2[4],ymm3[5],ymm2[6],ymm3[7]
; AVX2-NEXT:    vmovdqa %ymm2, (%r8)
; AVX2-NEXT:    vmovdqa %ymm0, 96(%r8)
; AVX2-NEXT:    vmovdqa %ymm1, 64(%r8)
; AVX2-NEXT:    vmovdqa %ymm11, 32(%r8)
; AVX2-NEXT:    vzeroupper
; AVX2-NEXT:    retq
;
; AVX512-LABEL: vf16:
; AVX512:       # %bb.0:
; AVX512-NEXT:    vmovdqa (%rdi), %ymm0
; AVX512-NEXT:    vmovdqa (%rdx), %ymm1
; AVX512-NEXT:    vinserti64x4 $1, (%rsi), %zmm0, %zmm0
; AVX512-NEXT:    vinserti64x4 $1, (%rcx), %zmm1, %zmm1
; AVX512-NEXT:    vmovdqa64 {{.*#+}} zmm2 = [0,16,32,48,1,17,33,49,2,18,34,50,3,19,35,51,4,20,36,52,5,21,37,53,6,22,38,54,7,23,39,55]
; AVX512-NEXT:    vpermi2w %zmm1, %zmm0, %zmm2
; AVX512-NEXT:    vmovdqa64 {{.*#+}} zmm3 = [8,24,40,56,9,25,41,57,10,26,42,58,11,27,43,59,12,28,44,60,13,29,45,61,14,30,46,62,15,31,47,63]
; AVX512-NEXT:    vpermi2w %zmm1, %zmm0, %zmm3
; AVX512-NEXT:    vmovdqu64 %zmm3, 64(%r8)
; AVX512-NEXT:    vmovdqu64 %zmm2, (%r8)
; AVX512-NEXT:    vzeroupper
; AVX512-NEXT:    retq
  %in.vec0 = load <16 x i16>, ptr %in.vecptr0, align 32
  %in.vec1 = load <16 x i16>, ptr %in.vecptr1, align 32
  %in.vec2 = load <16 x i16>, ptr %in.vecptr2, align 32
  %in.vec3 = load <16 x i16>, ptr %in.vecptr3, align 32

  %concat01 = shufflevector <16 x i16> %in.vec0, <16 x i16> %in.vec1, <32 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31>
  %concat23 = shufflevector <16 x i16> %in.vec2, <16 x i16> %in.vec3, <32 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31>
  %concat0123 = shufflevector <32 x i16> %concat01, <32 x i16> %concat23, <64 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31, i32 32, i32 33, i32 34, i32 35, i32 36, i32 37, i32 38, i32 39, i32 40, i32 41, i32 42, i32 43, i32 44, i32 45, i32 46, i32 47, i32 48, i32 49, i32 50, i32 51, i32 52, i32 53, i32 54, i32 55, i32 56, i32 57, i32 58, i32 59, i32 60, i32 61, i32 62, i32 63>
  %interleaved.vec = shufflevector <64 x i16> %concat0123, <64 x i16> poison, <64 x i32> <i32 0, i32 16, i32 32, i32 48, i32 1, i32 17, i32 33, i32 49, i32 2, i32 18, i32 34, i32 50, i32 3, i32 19, i32 35, i32 51, i32 4, i32 20, i32 36, i32 52, i32 5, i32 21, i32 37, i32 53, i32 6, i32 22, i32 38, i32 54, i32 7, i32 23, i32 39, i32 55, i32 8, i32 24, i32 40, i32 56, i32 9, i32 25, i32 41, i32 57, i32 10, i32 26, i32 42, i32 58, i32 11, i32 27, i32 43, i32 59, i32 12, i32 28, i32 44, i32 60, i32 13, i32 29, i32 45, i32 61, i32 14, i32 30, i32 46, i32 62, i32 15, i32 31, i32 47, i32 63>

  store <64 x i16> %interleaved.vec, ptr %out.vec, align 32

  ret void
}

define void @vf32(ptr %in.vecptr0, ptr %in.vecptr1, ptr %in.vecptr2, ptr %in.vecptr3, ptr %out.vec) nounwind {
; SSE-LABEL: vf32:
; SSE:       # %bb.0:
; SSE-NEXT:    movdqa (%rdi), %xmm10
; SSE-NEXT:    movdqa 16(%rdi), %xmm13
; SSE-NEXT:    movdqa 32(%rdi), %xmm8
; SSE-NEXT:    movdqa 48(%rdi), %xmm4
; SSE-NEXT:    movdqa (%rsi), %xmm3
; SSE-NEXT:    movdqa 16(%rsi), %xmm1
; SSE-NEXT:    movdqa 32(%rsi), %xmm9
; SSE-NEXT:    movdqa (%rdx), %xmm0
; SSE-NEXT:    movdqa 16(%rdx), %xmm5
; SSE-NEXT:    movdqa 32(%rdx), %xmm6
; SSE-NEXT:    movdqa (%rcx), %xmm11
; SSE-NEXT:    movdqa 16(%rcx), %xmm14
; SSE-NEXT:    movdqa 32(%rcx), %xmm12
; SSE-NEXT:    movdqa %xmm0, %xmm7
; SSE-NEXT:    punpcklwd {{.*#+}} xmm7 = xmm7[0],xmm11[0],xmm7[1],xmm11[1],xmm7[2],xmm11[2],xmm7[3],xmm11[3]
; SSE-NEXT:    movdqa %xmm10, %xmm15
; SSE-NEXT:    punpcklwd {{.*#+}} xmm15 = xmm15[0],xmm3[0],xmm15[1],xmm3[1],xmm15[2],xmm3[2],xmm15[3],xmm3[3]
; SSE-NEXT:    movdqa %xmm15, %xmm2
; SSE-NEXT:    punpckhdq {{.*#+}} xmm2 = xmm2[2],xmm7[2],xmm2[3],xmm7[3]
; SSE-NEXT:    movdqa %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; SSE-NEXT:    punpckldq {{.*#+}} xmm15 = xmm15[0],xmm7[0],xmm15[1],xmm7[1]
; SSE-NEXT:    punpckhwd {{.*#+}} xmm0 = xmm0[4],xmm11[4],xmm0[5],xmm11[5],xmm0[6],xmm11[6],xmm0[7],xmm11[7]
; SSE-NEXT:    punpckhwd {{.*#+}} xmm10 = xmm10[4],xmm3[4],xmm10[5],xmm3[5],xmm10[6],xmm3[6],xmm10[7],xmm3[7]
; SSE-NEXT:    movdqa %xmm10, %xmm2
; SSE-NEXT:    punpckhdq {{.*#+}} xmm2 = xmm2[2],xmm0[2],xmm2[3],xmm0[3]
; SSE-NEXT:    movdqa %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; SSE-NEXT:    punpckldq {{.*#+}} xmm10 = xmm10[0],xmm0[0],xmm10[1],xmm0[1]
; SSE-NEXT:    movdqa %xmm5, %xmm0
; SSE-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm14[0],xmm0[1],xmm14[1],xmm0[2],xmm14[2],xmm0[3],xmm14[3]
; SSE-NEXT:    movdqa %xmm13, %xmm7
; SSE-NEXT:    punpcklwd {{.*#+}} xmm7 = xmm7[0],xmm1[0],xmm7[1],xmm1[1],xmm7[2],xmm1[2],xmm7[3],xmm1[3]
; SSE-NEXT:    movdqa %xmm7, %xmm2
; SSE-NEXT:    punpckhdq {{.*#+}} xmm2 = xmm2[2],xmm0[2],xmm2[3],xmm0[3]
; SSE-NEXT:    movdqa %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; SSE-NEXT:    punpckldq {{.*#+}} xmm7 = xmm7[0],xmm0[0],xmm7[1],xmm0[1]
; SSE-NEXT:    punpckhwd {{.*#+}} xmm5 = xmm5[4],xmm14[4],xmm5[5],xmm14[5],xmm5[6],xmm14[6],xmm5[7],xmm14[7]
; SSE-NEXT:    punpckhwd {{.*#+}} xmm13 = xmm13[4],xmm1[4],xmm13[5],xmm1[5],xmm13[6],xmm1[6],xmm13[7],xmm1[7]
; SSE-NEXT:    movdqa %xmm13, %xmm11
; SSE-NEXT:    punpckhdq {{.*#+}} xmm11 = xmm11[2],xmm5[2],xmm11[3],xmm5[3]
; SSE-NEXT:    punpckldq {{.*#+}} xmm13 = xmm13[0],xmm5[0],xmm13[1],xmm5[1]
; SSE-NEXT:    movdqa %xmm6, %xmm0
; SSE-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm12[0],xmm0[1],xmm12[1],xmm0[2],xmm12[2],xmm0[3],xmm12[3]
; SSE-NEXT:    movdqa %xmm8, %xmm5
; SSE-NEXT:    punpcklwd {{.*#+}} xmm5 = xmm5[0],xmm9[0],xmm5[1],xmm9[1],xmm5[2],xmm9[2],xmm5[3],xmm9[3]
; SSE-NEXT:    movdqa %xmm5, %xmm14
; SSE-NEXT:    punpckhdq {{.*#+}} xmm14 = xmm14[2],xmm0[2],xmm14[3],xmm0[3]
; SSE-NEXT:    punpckldq {{.*#+}} xmm5 = xmm5[0],xmm0[0],xmm5[1],xmm0[1]
; SSE-NEXT:    movdqa 48(%rdx), %xmm0
; SSE-NEXT:    punpckhwd {{.*#+}} xmm6 = xmm6[4],xmm12[4],xmm6[5],xmm12[5],xmm6[6],xmm12[6],xmm6[7],xmm12[7]
; SSE-NEXT:    movdqa 48(%rcx), %xmm12
; SSE-NEXT:    punpckhwd {{.*#+}} xmm8 = xmm8[4],xmm9[4],xmm8[5],xmm9[5],xmm8[6],xmm9[6],xmm8[7],xmm9[7]
; SSE-NEXT:    movdqa %xmm8, %xmm9
; SSE-NEXT:    punpckhdq {{.*#+}} xmm9 = xmm9[2],xmm6[2],xmm9[3],xmm6[3]
; SSE-NEXT:    punpckldq {{.*#+}} xmm8 = xmm8[0],xmm6[0],xmm8[1],xmm6[1]
; SSE-NEXT:    movdqa %xmm0, %xmm6
; SSE-NEXT:    punpcklwd {{.*#+}} xmm6 = xmm6[0],xmm12[0],xmm6[1],xmm12[1],xmm6[2],xmm12[2],xmm6[3],xmm12[3]
; SSE-NEXT:    movdqa 48(%rsi), %xmm2
; SSE-NEXT:    movdqa %xmm4, %xmm3
; SSE-NEXT:    punpcklwd {{.*#+}} xmm3 = xmm3[0],xmm2[0],xmm3[1],xmm2[1],xmm3[2],xmm2[2],xmm3[3],xmm2[3]
; SSE-NEXT:    movdqa %xmm3, %xmm1
; SSE-NEXT:    punpckhdq {{.*#+}} xmm1 = xmm1[2],xmm6[2],xmm1[3],xmm6[3]
; SSE-NEXT:    punpckldq {{.*#+}} xmm3 = xmm3[0],xmm6[0],xmm3[1],xmm6[1]
; SSE-NEXT:    punpckhwd {{.*#+}} xmm0 = xmm0[4],xmm12[4],xmm0[5],xmm12[5],xmm0[6],xmm12[6],xmm0[7],xmm12[7]
; SSE-NEXT:    punpckhwd {{.*#+}} xmm4 = xmm4[4],xmm2[4],xmm4[5],xmm2[5],xmm4[6],xmm2[6],xmm4[7],xmm2[7]
; SSE-NEXT:    movdqa %xmm4, %xmm2
; SSE-NEXT:    punpckhdq {{.*#+}} xmm2 = xmm2[2],xmm0[2],xmm2[3],xmm0[3]
; SSE-NEXT:    punpckldq {{.*#+}} xmm4 = xmm4[0],xmm0[0],xmm4[1],xmm0[1]
; SSE-NEXT:    movdqa %xmm4, 224(%r8)
; SSE-NEXT:    movdqa %xmm2, 240(%r8)
; SSE-NEXT:    movdqa %xmm3, 192(%r8)
; SSE-NEXT:    movdqa %xmm1, 208(%r8)
; SSE-NEXT:    movdqa %xmm8, 160(%r8)
; SSE-NEXT:    movdqa %xmm9, 176(%r8)
; SSE-NEXT:    movdqa %xmm5, 128(%r8)
; SSE-NEXT:    movdqa %xmm14, 144(%r8)
; SSE-NEXT:    movdqa %xmm13, 96(%r8)
; SSE-NEXT:    movdqa %xmm11, 112(%r8)
; SSE-NEXT:    movdqa %xmm7, 64(%r8)
; SSE-NEXT:    movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
; SSE-NEXT:    movaps %xmm0, 80(%r8)
; SSE-NEXT:    movdqa %xmm10, 32(%r8)
; SSE-NEXT:    movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
; SSE-NEXT:    movaps %xmm0, 48(%r8)
; SSE-NEXT:    movdqa %xmm15, (%r8)
; SSE-NEXT:    movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
; SSE-NEXT:    movaps %xmm0, 16(%r8)
; SSE-NEXT:    retq
;
; AVX1-LABEL: vf32:
; AVX1:       # %bb.0:
; AVX1-NEXT:    vmovdqa (%rcx), %xmm12
; AVX1-NEXT:    vmovdqa 16(%rcx), %xmm15
; AVX1-NEXT:    vmovdqa 32(%rcx), %xmm3
; AVX1-NEXT:    vmovdqa 48(%rcx), %xmm11
; AVX1-NEXT:    vmovdqa (%rdx), %xmm13
; AVX1-NEXT:    vmovdqa 16(%rdx), %xmm6
; AVX1-NEXT:    vmovdqa 32(%rdx), %xmm7
; AVX1-NEXT:    vmovdqa 48(%rdx), %xmm1
; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm2 = xmm13[4],xmm12[4],xmm13[5],xmm12[5],xmm13[6],xmm12[6],xmm13[7],xmm12[7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm5 = xmm2[0,0,1,1]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm5, %ymm8
; AVX1-NEXT:    vmovdqa (%rsi), %xmm14
; AVX1-NEXT:    vmovdqa 48(%rsi), %xmm2
; AVX1-NEXT:    vmovdqa (%rdi), %xmm5
; AVX1-NEXT:    vmovdqa 48(%rdi), %xmm4
; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm0 = xmm5[4],xmm14[4],xmm5[5],xmm14[5],xmm5[6],xmm14[6],xmm5[7],xmm14[7]
; AVX1-NEXT:    vpmovzxdq {{.*#+}} xmm9 = xmm0[0],zero,xmm0[1],zero
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm9, %ymm0
; AVX1-NEXT:    vblendps {{.*#+}} ymm0 = ymm0[0],ymm8[1],ymm0[2],ymm8[3],ymm0[4],ymm8[5],ymm0[6],ymm8[7]
; AVX1-NEXT:    vmovups %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX1-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm1[0],xmm11[0],xmm1[1],xmm11[1],xmm1[2],xmm11[2],xmm1[3],xmm11[3]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm8 = xmm0[0,0,1,1]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm8, %ymm8
; AVX1-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm4[0],xmm2[0],xmm4[1],xmm2[1],xmm4[2],xmm2[2],xmm4[3],xmm2[3]
; AVX1-NEXT:    vpmovzxdq {{.*#+}} xmm10 = xmm0[0],zero,xmm0[1],zero
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm10, %ymm0
; AVX1-NEXT:    vmovdqa 32(%rsi), %xmm10
; AVX1-NEXT:    vblendps {{.*#+}} ymm9 = ymm0[0],ymm8[1],ymm0[2],ymm8[3],ymm0[4],ymm8[5],ymm0[6],ymm8[7]
; AVX1-NEXT:    vmovdqa 32(%rdi), %xmm0
; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm1 = xmm1[4],xmm11[4],xmm1[5],xmm11[5],xmm1[6],xmm11[6],xmm1[7],xmm11[7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm8 = xmm1[0,0,1,1]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm8, %ymm1
; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm2 = xmm4[4],xmm2[4],xmm4[5],xmm2[5],xmm4[6],xmm2[6],xmm4[7],xmm2[7]
; AVX1-NEXT:    vpmovzxdq {{.*#+}} xmm4 = xmm2[0],zero,xmm2[1],zero
; AVX1-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm4, %ymm2
; AVX1-NEXT:    vblendps {{.*#+}} ymm8 = ymm2[0],ymm1[1],ymm2[2],ymm1[3],ymm2[4],ymm1[5],ymm2[6],ymm1[7]
; AVX1-NEXT:    vpunpcklwd {{.*#+}} xmm1 = xmm7[0],xmm3[0],xmm7[1],xmm3[1],xmm7[2],xmm3[2],xmm7[3],xmm3[3]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm2 = xmm1[0,0,1,1]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm2, %ymm1
; AVX1-NEXT:    vpunpcklwd {{.*#+}} xmm2 = xmm0[0],xmm10[0],xmm0[1],xmm10[1],xmm0[2],xmm10[2],xmm0[3],xmm10[3]
; AVX1-NEXT:    vpmovzxdq {{.*#+}} xmm4 = xmm2[0],zero,xmm2[1],zero
; AVX1-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm4, %ymm2
; AVX1-NEXT:    vmovdqa 16(%rsi), %xmm4
; AVX1-NEXT:    vblendps {{.*#+}} ymm11 = ymm2[0],ymm1[1],ymm2[2],ymm1[3],ymm2[4],ymm1[5],ymm2[6],ymm1[7]
; AVX1-NEXT:    vmovdqa 16(%rdi), %xmm1
; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm2 = xmm7[4],xmm3[4],xmm7[5],xmm3[5],xmm7[6],xmm3[6],xmm7[7],xmm3[7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm3 = xmm2[0,0,1,1]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm3, %ymm2
; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm0 = xmm0[4],xmm10[4],xmm0[5],xmm10[5],xmm0[6],xmm10[6],xmm0[7],xmm10[7]
; AVX1-NEXT:    vpmovzxdq {{.*#+}} xmm3 = xmm0[0],zero,xmm0[1],zero
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm3, %ymm0
; AVX1-NEXT:    vblendps {{.*#+}} ymm3 = ymm0[0],ymm2[1],ymm0[2],ymm2[3],ymm0[4],ymm2[5],ymm0[6],ymm2[7]
; AVX1-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm6[0],xmm15[0],xmm6[1],xmm15[1],xmm6[2],xmm15[2],xmm6[3],xmm15[3]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm2 = xmm0[0,0,1,1]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm2, %ymm0
; AVX1-NEXT:    vpunpcklwd {{.*#+}} xmm2 = xmm1[0],xmm4[0],xmm1[1],xmm4[1],xmm1[2],xmm4[2],xmm1[3],xmm4[3]
; AVX1-NEXT:    vpmovzxdq {{.*#+}} xmm7 = xmm2[0],zero,xmm2[1],zero
; AVX1-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm7, %ymm2
; AVX1-NEXT:    vblendps {{.*#+}} ymm0 = ymm2[0],ymm0[1],ymm2[2],ymm0[3],ymm2[4],ymm0[5],ymm2[6],ymm0[7]
; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm2 = xmm6[4],xmm15[4],xmm6[5],xmm15[5],xmm6[6],xmm15[6],xmm6[7],xmm15[7]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm6 = xmm2[0,0,1,1]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm6, %ymm2
; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm1 = xmm1[4],xmm4[4],xmm1[5],xmm4[5],xmm1[6],xmm4[6],xmm1[7],xmm4[7]
; AVX1-NEXT:    vpmovzxdq {{.*#+}} xmm4 = xmm1[0],zero,xmm1[1],zero
; AVX1-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm4, %ymm1
; AVX1-NEXT:    vblendps {{.*#+}} ymm1 = ymm1[0],ymm2[1],ymm1[2],ymm2[3],ymm1[4],ymm2[5],ymm1[6],ymm2[7]
; AVX1-NEXT:    vpunpcklwd {{.*#+}} xmm2 = xmm13[0],xmm12[0],xmm13[1],xmm12[1],xmm13[2],xmm12[2],xmm13[3],xmm12[3]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm4 = xmm2[0,0,1,1]
; AVX1-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm4, %ymm2
; AVX1-NEXT:    vpunpcklwd {{.*#+}} xmm4 = xmm5[0],xmm14[0],xmm5[1],xmm14[1],xmm5[2],xmm14[2],xmm5[3],xmm14[3]
; AVX1-NEXT:    vpmovzxdq {{.*#+}} xmm5 = xmm4[0],zero,xmm4[1],zero
; AVX1-NEXT:    vpshufd {{.*#+}} xmm4 = xmm4[2,2,3,3]
; AVX1-NEXT:    vinsertf128 $1, %xmm4, %ymm5, %ymm4
; AVX1-NEXT:    vblendps {{.*#+}} ymm2 = ymm4[0],ymm2[1],ymm4[2],ymm2[3],ymm4[4],ymm2[5],ymm4[6],ymm2[7]
; AVX1-NEXT:    vmovaps %ymm2, (%r8)
; AVX1-NEXT:    vmovaps %ymm1, 96(%r8)
; AVX1-NEXT:    vmovaps %ymm0, 64(%r8)
; AVX1-NEXT:    vmovaps %ymm3, 160(%r8)
; AVX1-NEXT:    vmovaps %ymm11, 128(%r8)
; AVX1-NEXT:    vmovaps %ymm8, 224(%r8)
; AVX1-NEXT:    vmovaps %ymm9, 192(%r8)
; AVX1-NEXT:    vmovups {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX1-NEXT:    vmovaps %ymm0, 32(%r8)
; AVX1-NEXT:    vzeroupper
; AVX1-NEXT:    retq
;
; AVX2-LABEL: vf32:
; AVX2:       # %bb.0:
; AVX2-NEXT:    vmovdqa (%rcx), %xmm12
; AVX2-NEXT:    vmovdqa 16(%rcx), %xmm15
; AVX2-NEXT:    vmovdqa 32(%rcx), %xmm3
; AVX2-NEXT:    vmovdqa 48(%rcx), %xmm11
; AVX2-NEXT:    vmovdqa (%rdx), %xmm13
; AVX2-NEXT:    vmovdqa 16(%rdx), %xmm6
; AVX2-NEXT:    vmovdqa 32(%rdx), %xmm7
; AVX2-NEXT:    vmovdqa 48(%rdx), %xmm1
; AVX2-NEXT:    vpunpckhwd {{.*#+}} xmm2 = xmm13[4],xmm12[4],xmm13[5],xmm12[5],xmm13[6],xmm12[6],xmm13[7],xmm12[7]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm5 = xmm2[0,0,1,1]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm2, %ymm5, %ymm8
; AVX2-NEXT:    vmovdqa (%rsi), %xmm14
; AVX2-NEXT:    vmovdqa 48(%rsi), %xmm2
; AVX2-NEXT:    vmovdqa (%rdi), %xmm5
; AVX2-NEXT:    vmovdqa 48(%rdi), %xmm4
; AVX2-NEXT:    vpunpckhwd {{.*#+}} xmm0 = xmm5[4],xmm14[4],xmm5[5],xmm14[5],xmm5[6],xmm14[6],xmm5[7],xmm14[7]
; AVX2-NEXT:    vpmovzxdq {{.*#+}} xmm9 = xmm0[0],zero,xmm0[1],zero
; AVX2-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm0, %ymm9, %ymm0
; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0],ymm8[1],ymm0[2],ymm8[3],ymm0[4],ymm8[5],ymm0[6],ymm8[7]
; AVX2-NEXT:    vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; AVX2-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm1[0],xmm11[0],xmm1[1],xmm11[1],xmm1[2],xmm11[2],xmm1[3],xmm11[3]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm8 = xmm0[0,0,1,1]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm0, %ymm8, %ymm8
; AVX2-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm4[0],xmm2[0],xmm4[1],xmm2[1],xmm4[2],xmm2[2],xmm4[3],xmm2[3]
; AVX2-NEXT:    vpmovzxdq {{.*#+}} xmm10 = xmm0[0],zero,xmm0[1],zero
; AVX2-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm0, %ymm10, %ymm0
; AVX2-NEXT:    vmovdqa 32(%rsi), %xmm10
; AVX2-NEXT:    vpblendd {{.*#+}} ymm9 = ymm0[0],ymm8[1],ymm0[2],ymm8[3],ymm0[4],ymm8[5],ymm0[6],ymm8[7]
; AVX2-NEXT:    vmovdqa 32(%rdi), %xmm0
; AVX2-NEXT:    vpunpckhwd {{.*#+}} xmm1 = xmm1[4],xmm11[4],xmm1[5],xmm11[5],xmm1[6],xmm11[6],xmm1[7],xmm11[7]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm8 = xmm1[0,0,1,1]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm8, %ymm1
; AVX2-NEXT:    vpunpckhwd {{.*#+}} xmm2 = xmm4[4],xmm2[4],xmm4[5],xmm2[5],xmm4[6],xmm2[6],xmm4[7],xmm2[7]
; AVX2-NEXT:    vpmovzxdq {{.*#+}} xmm4 = xmm2[0],zero,xmm2[1],zero
; AVX2-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm2, %ymm4, %ymm2
; AVX2-NEXT:    vpblendd {{.*#+}} ymm8 = ymm2[0],ymm1[1],ymm2[2],ymm1[3],ymm2[4],ymm1[5],ymm2[6],ymm1[7]
; AVX2-NEXT:    vpunpcklwd {{.*#+}} xmm1 = xmm7[0],xmm3[0],xmm7[1],xmm3[1],xmm7[2],xmm3[2],xmm7[3],xmm3[3]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm2 = xmm1[0,0,1,1]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm2, %ymm1
; AVX2-NEXT:    vpunpcklwd {{.*#+}} xmm2 = xmm0[0],xmm10[0],xmm0[1],xmm10[1],xmm0[2],xmm10[2],xmm0[3],xmm10[3]
; AVX2-NEXT:    vpmovzxdq {{.*#+}} xmm4 = xmm2[0],zero,xmm2[1],zero
; AVX2-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm2, %ymm4, %ymm2
; AVX2-NEXT:    vmovdqa 16(%rsi), %xmm4
; AVX2-NEXT:    vpblendd {{.*#+}} ymm11 = ymm2[0],ymm1[1],ymm2[2],ymm1[3],ymm2[4],ymm1[5],ymm2[6],ymm1[7]
; AVX2-NEXT:    vmovdqa 16(%rdi), %xmm1
; AVX2-NEXT:    vpunpckhwd {{.*#+}} xmm2 = xmm7[4],xmm3[4],xmm7[5],xmm3[5],xmm7[6],xmm3[6],xmm7[7],xmm3[7]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm3 = xmm2[0,0,1,1]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm2, %ymm3, %ymm2
; AVX2-NEXT:    vpunpckhwd {{.*#+}} xmm0 = xmm0[4],xmm10[4],xmm0[5],xmm10[5],xmm0[6],xmm10[6],xmm0[7],xmm10[7]
; AVX2-NEXT:    vpmovzxdq {{.*#+}} xmm3 = xmm0[0],zero,xmm0[1],zero
; AVX2-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm0, %ymm3, %ymm0
; AVX2-NEXT:    vpblendd {{.*#+}} ymm3 = ymm0[0],ymm2[1],ymm0[2],ymm2[3],ymm0[4],ymm2[5],ymm0[6],ymm2[7]
; AVX2-NEXT:    vpunpcklwd {{.*#+}} xmm0 = xmm6[0],xmm15[0],xmm6[1],xmm15[1],xmm6[2],xmm15[2],xmm6[3],xmm15[3]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm2 = xmm0[0,0,1,1]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm0, %ymm2, %ymm0
; AVX2-NEXT:    vpunpcklwd {{.*#+}} xmm2 = xmm1[0],xmm4[0],xmm1[1],xmm4[1],xmm1[2],xmm4[2],xmm1[3],xmm4[3]
; AVX2-NEXT:    vpmovzxdq {{.*#+}} xmm7 = xmm2[0],zero,xmm2[1],zero
; AVX2-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm2, %ymm7, %ymm2
; AVX2-NEXT:    vpblendd {{.*#+}} ymm0 = ymm2[0],ymm0[1],ymm2[2],ymm0[3],ymm2[4],ymm0[5],ymm2[6],ymm0[7]
; AVX2-NEXT:    vpunpckhwd {{.*#+}} xmm2 = xmm6[4],xmm15[4],xmm6[5],xmm15[5],xmm6[6],xmm15[6],xmm6[7],xmm15[7]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm6 = xmm2[0,0,1,1]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm2, %ymm6, %ymm2
; AVX2-NEXT:    vpunpckhwd {{.*#+}} xmm1 = xmm1[4],xmm4[4],xmm1[5],xmm4[5],xmm1[6],xmm4[6],xmm1[7],xmm4[7]
; AVX2-NEXT:    vpmovzxdq {{.*#+}} xmm4 = xmm1[0],zero,xmm1[1],zero
; AVX2-NEXT:    vpshufd {{.*#+}} xmm1 = xmm1[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm4, %ymm1
; AVX2-NEXT:    vpblendd {{.*#+}} ymm1 = ymm1[0],ymm2[1],ymm1[2],ymm2[3],ymm1[4],ymm2[5],ymm1[6],ymm2[7]
; AVX2-NEXT:    vpunpcklwd {{.*#+}} xmm2 = xmm13[0],xmm12[0],xmm13[1],xmm12[1],xmm13[2],xmm12[2],xmm13[3],xmm12[3]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm4 = xmm2[0,0,1,1]
; AVX2-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm2, %ymm4, %ymm2
; AVX2-NEXT:    vpunpcklwd {{.*#+}} xmm4 = xmm5[0],xmm14[0],xmm5[1],xmm14[1],xmm5[2],xmm14[2],xmm5[3],xmm14[3]
; AVX2-NEXT:    vpmovzxdq {{.*#+}} xmm5 = xmm4[0],zero,xmm4[1],zero
; AVX2-NEXT:    vpshufd {{.*#+}} xmm4 = xmm4[2,2,3,3]
; AVX2-NEXT:    vinserti128 $1, %xmm4, %ymm5, %ymm4
; AVX2-NEXT:    vpblendd {{.*#+}} ymm2 = ymm4[0],ymm2[1],ymm4[2],ymm2[3],ymm4[4],ymm2[5],ymm4[6],ymm2[7]
; AVX2-NEXT:    vmovdqa %ymm2, (%r8)
; AVX2-NEXT:    vmovdqa %ymm1, 96(%r8)
; AVX2-NEXT:    vmovdqa %ymm0, 64(%r8)
; AVX2-NEXT:    vmovdqa %ymm3, 160(%r8)
; AVX2-NEXT:    vmovdqa %ymm11, 128(%r8)
; AVX2-NEXT:    vmovdqa %ymm8, 224(%r8)
; AVX2-NEXT:    vmovdqa %ymm9, 192(%r8)
; AVX2-NEXT:    vmovups {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
; AVX2-NEXT:    vmovaps %ymm0, 32(%r8)
; AVX2-NEXT:    vzeroupper
; AVX2-NEXT:    retq
;
; AVX512-LABEL: vf32:
; AVX512:       # %bb.0:
; AVX512-NEXT:    vmovdqu64 (%rdi), %zmm0
; AVX512-NEXT:    vmovdqu64 (%rsi), %zmm1
; AVX512-NEXT:    vmovdqu64 (%rdx), %zmm2
; AVX512-NEXT:    vmovdqu64 (%rcx), %zmm3
; AVX512-NEXT:    vmovdqa64 {{.*#+}} zmm4 = <u,u,0,32,u,u,1,33,u,u,2,34,u,u,3,35,u,u,4,36,u,u,5,37,u,u,6,38,u,u,7,39>
; AVX512-NEXT:    vpermi2w %zmm3, %zmm2, %zmm4
; AVX512-NEXT:    vmovdqa64 {{.*#+}} zmm5 = <0,32,u,u,1,33,u,u,2,34,u,u,3,35,u,u,4,36,u,u,5,37,u,u,6,38,u,u,7,39,u,u>
; AVX512-NEXT:    vpermi2w %zmm1, %zmm0, %zmm5
; AVX512-NEXT:    movw $-21846, %ax # imm = 0xAAAA
; AVX512-NEXT:    kmovd %eax, %k1
; AVX512-NEXT:    vmovdqa32 %zmm4, %zmm5 {%k1}
; AVX512-NEXT:    vmovdqa64 {{.*#+}} zmm4 = <u,u,8,40,u,u,9,41,u,u,10,42,u,u,11,43,u,u,12,44,u,u,13,45,u,u,14,46,u,u,15,47>
; AVX512-NEXT:    vpermi2w %zmm3, %zmm2, %zmm4
; AVX512-NEXT:    vmovdqa64 {{.*#+}} zmm6 = <8,40,u,u,9,41,u,u,10,42,u,u,11,43,u,u,12,44,u,u,13,45,u,u,14,46,u,u,15,47,u,u>
; AVX512-NEXT:    vpermi2w %zmm1, %zmm0, %zmm6
; AVX512-NEXT:    vmovdqa32 %zmm4, %zmm6 {%k1}
; AVX512-NEXT:    vmovdqa64 {{.*#+}} zmm4 = <u,u,16,48,u,u,17,49,u,u,18,50,u,u,19,51,u,u,20,52,u,u,21,53,u,u,22,54,u,u,23,55>
; AVX512-NEXT:    vpermi2w %zmm3, %zmm2, %zmm4
; AVX512-NEXT:    vmovdqa64 {{.*#+}} zmm7 = <16,48,u,u,17,49,u,u,18,50,u,u,19,51,u,u,20,52,u,u,21,53,u,u,22,54,u,u,23,55,u,u>
; AVX512-NEXT:    vpermi2w %zmm1, %zmm0, %zmm7
; AVX512-NEXT:    vmovdqa32 %zmm4, %zmm7 {%k1}
; AVX512-NEXT:    vmovdqa64 {{.*#+}} zmm4 = <u,u,24,56,u,u,25,57,u,u,26,58,u,u,27,59,u,u,28,60,u,u,29,61,u,u,30,62,u,u,31,63>
; AVX512-NEXT:    vpermi2w %zmm3, %zmm2, %zmm4
; AVX512-NEXT:    vmovdqa64 {{.*#+}} zmm2 = <24,56,u,u,25,57,u,u,26,58,u,u,27,59,u,u,28,60,u,u,29,61,u,u,30,62,u,u,31,63,u,u>
; AVX512-NEXT:    vpermi2w %zmm1, %zmm0, %zmm2
; AVX512-NEXT:    vmovdqa32 %zmm4, %zmm2 {%k1}
; AVX512-NEXT:    vmovdqu64 %zmm2, 192(%r8)
; AVX512-NEXT:    vmovdqu64 %zmm7, 128(%r8)
; AVX512-NEXT:    vmovdqu64 %zmm6, 64(%r8)
; AVX512-NEXT:    vmovdqu64 %zmm5, (%r8)
; AVX512-NEXT:    vzeroupper
; AVX512-NEXT:    retq
  %in.vec0 = load <32 x i16>, ptr %in.vecptr0, align 32
  %in.vec1 = load <32 x i16>, ptr %in.vecptr1, align 32
  %in.vec2 = load <32 x i16>, ptr %in.vecptr2, align 32
  %in.vec3 = load <32 x i16>, ptr %in.vecptr3, align 32

  %concat01 = shufflevector <32 x i16> %in.vec0, <32 x i16> %in.vec1, <64 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31, i32 32, i32 33, i32 34, i32 35, i32 36, i32 37, i32 38, i32 39, i32 40, i32 41, i32 42, i32 43, i32 44, i32 45, i32 46, i32 47, i32 48, i32 49, i32 50, i32 51, i32 52, i32 53, i32 54, i32 55, i32 56, i32 57, i32 58, i32 59, i32 60, i32 61, i32 62, i32 63>
  %concat23 = shufflevector <32 x i16> %in.vec2, <32 x i16> %in.vec3, <64 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31, i32 32, i32 33, i32 34, i32 35, i32 36, i32 37, i32 38, i32 39, i32 40, i32 41, i32 42, i32 43, i32 44, i32 45, i32 46, i32 47, i32 48, i32 49, i32 50, i32 51, i32 52, i32 53, i32 54, i32 55, i32 56, i32 57, i32 58, i32 59, i32 60, i32 61, i32 62, i32 63>
  %concat0123 = shufflevector <64 x i16> %concat01, <64 x i16> %concat23, <128 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31, i32 32, i32 33, i32 34, i32 35, i32 36, i32 37, i32 38, i32 39, i32 40, i32 41, i32 42, i32 43, i32 44, i32 45, i32 46, i32 47, i32 48, i32 49, i32 50, i32 51, i32 52, i32 53, i32 54, i32 55, i32 56, i32 57, i32 58, i32 59, i32 60, i32 61, i32 62, i32 63, i32 64, i32 65, i32 66, i32 67, i32 68, i32 69, i32 70, i32 71, i32 72, i32 73, i32 74, i32 75, i32 76, i32 77, i32 78, i32 79, i32 80, i32 81, i32 82, i32 83, i32 84, i32 85, i32 86, i32 87, i32 88, i32 89, i32 90, i32 91, i32 92, i32 93, i32 94, i32 95, i32 96, i32 97, i32 98, i32 99, i32 100, i32 101, i32 102, i32 103, i32 104, i32 105, i32 106, i32 107, i32 108, i32 109, i32 110, i32 111, i32 112, i32 113, i32 114, i32 115, i32 116, i32 117, i32 118, i32 119, i32 120, i32 121, i32 122, i32 123, i32 124, i32 125, i32 126, i32 127>
  %interleaved.vec = shufflevector <128 x i16> %concat0123, <128 x i16> poison, <128 x i32> <i32 0, i32 32, i32 64, i32 96, i32 1, i32 33, i32 65, i32 97, i32 2, i32 34, i32 66, i32 98, i32 3, i32 35, i32 67, i32 99, i32 4, i32 36, i32 68, i32 100, i32 5, i32 37, i32 69, i32 101, i32 6, i32 38, i32 70, i32 102, i32 7, i32 39, i32 71, i32 103, i32 8, i32 40, i32 72, i32 104, i32 9, i32 41, i32 73, i32 105, i32 10, i32 42, i32 74, i32 106, i32 11, i32 43, i32 75, i32 107, i32 12, i32 44, i32 76, i32 108, i32 13, i32 45, i32 77, i32 109, i32 14, i32 46, i32 78, i32 110, i32 15, i32 47, i32 79, i32 111, i32 16, i32 48, i32 80, i32 112, i32 17, i32 49, i32 81, i32 113, i32 18, i32 50, i32 82, i32 114, i32 19, i32 51, i32 83, i32 115, i32 20, i32 52, i32 84, i32 116, i32 21, i32 53, i32 85, i32 117, i32 22, i32 54, i32 86, i32 118, i32 23, i32 55, i32 87, i32 119, i32 24, i32 56, i32 88, i32 120, i32 25, i32 57, i32 89, i32 121, i32 26, i32 58, i32 90, i32 122, i32 27, i32 59, i32 91, i32 123, i32 28, i32 60, i32 92, i32 124, i32 29, i32 61, i32 93, i32 125, i32 30, i32 62, i32 94, i32 126, i32 31, i32 63, i32 95, i32 127>

  store <128 x i16> %interleaved.vec, ptr %out.vec, align 32

  ret void
}