; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+sse4.1 | FileCheck %s --check-prefix=SSE
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx | FileCheck %s --check-prefix=AVX
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx2 | FileCheck %s --check-prefix=AVX
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx512f | FileCheck %s --check-prefix=AVX

; Combine tests involving SSE41 target shuffles (BLEND,INSERTPS,MOVZX)

declare <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8>, <16 x i8>)

define <16 x i8> @combine_vpshufb_as_movzx(<16 x i8> %a0) {
; SSE-LABEL: combine_vpshufb_as_movzx:
; SSE:       # %bb.0:
; SSE-NEXT:    pmovzxdq {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero
; SSE-NEXT:    retq
;
; AVX-LABEL: combine_vpshufb_as_movzx:
; AVX:       # %bb.0:
; AVX-NEXT:    vpmovzxdq {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero
; AVX-NEXT:    retq
  %res0 = call <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8> %a0, <16 x i8> <i8 0, i8 1, i8 2, i8 3, i8 -1, i8 -1, i8 -1, i8 -1, i8 undef, i8 undef, i8 undef, i8 undef, i8 -1, i8 -1, i8 -1, i8 -1>)
  ret <16 x i8> %res0
}

define <16 x i8> @PR50049(<48 x i8>* %p1, <48 x i8>* %p2) {
; SSE-LABEL: PR50049:
; SSE:       # %bb.0:
; SSE-NEXT:    movdqa (%rdi), %xmm2
; SSE-NEXT:    movdqa 16(%rdi), %xmm0
; SSE-NEXT:    movdqa 32(%rdi), %xmm1
; SSE-NEXT:    movdqa (%rsi), %xmm4
; SSE-NEXT:    movdqa 16(%rsi), %xmm5
; SSE-NEXT:    movdqa 32(%rsi), %xmm3
; SSE-NEXT:    movdqa {{.*#+}} xmm6 = <128,128,128,128,128,128,2,5,8,11,14,u,u,u,u,u>
; SSE-NEXT:    pshufb %xmm6, %xmm0
; SSE-NEXT:    movdqa {{.*#+}} xmm7 = <0,3,6,9,12,15,128,128,128,128,128,u,u,u,u,u>
; SSE-NEXT:    pshufb %xmm7, %xmm2
; SSE-NEXT:    por %xmm0, %xmm2
; SSE-NEXT:    pshufb %xmm6, %xmm5
; SSE-NEXT:    pshufb %xmm7, %xmm4
; SSE-NEXT:    por %xmm5, %xmm4
; SSE-NEXT:    pmovzxbw {{.*#+}} xmm5 = xmm4[0],zero,xmm4[1],zero,xmm4[2],zero,xmm4[3],zero,xmm4[4],zero,xmm4[5],zero,xmm4[6],zero,xmm4[7],zero
; SSE-NEXT:    pmovzxbw {{.*#+}} xmm0 = xmm2[0],zero,xmm2[1],zero,xmm2[2],zero,xmm2[3],zero,xmm2[4],zero,xmm2[5],zero,xmm2[6],zero,xmm2[7],zero
; SSE-NEXT:    pmullw %xmm5, %xmm0
; SSE-NEXT:    movdqa {{.*#+}} xmm5 = [255,255,255,255,255,255,255,255]
; SSE-NEXT:    pand %xmm5, %xmm0
; SSE-NEXT:    movdqa {{.*#+}} xmm6 = <8,u,9,u,10,u,128,u,128,u,128,u,128,u,128,u>
; SSE-NEXT:    pshufb %xmm6, %xmm4
; SSE-NEXT:    movdqa {{.*#+}} xmm7 = <128,u,128,u,128,u,1,u,4,u,7,u,10,u,13,u>
; SSE-NEXT:    pshufb %xmm7, %xmm3
; SSE-NEXT:    por %xmm4, %xmm3
; SSE-NEXT:    pshufb %xmm6, %xmm2
; SSE-NEXT:    pshufb %xmm7, %xmm1
; SSE-NEXT:    por %xmm2, %xmm1
; SSE-NEXT:    pmullw %xmm3, %xmm1
; SSE-NEXT:    pand %xmm5, %xmm1
; SSE-NEXT:    packuswb %xmm1, %xmm0
; SSE-NEXT:    retq
  %x1 = load <48 x i8>, <48 x i8>* %p1, align 16
  %x2 = load <48 x i8>, <48 x i8>* %p2, align 16
  %s1 = shufflevector <48 x i8> %x1, <48 x i8> poison, <16 x i32> <i32 0, i32 3, i32 6, i32 9, i32 12, i32 15, i32 18, i32 21, i32 24, i32 27, i32 30, i32 33, i32 36, i32 39, i32 42, i32 45>
  %s2 = shufflevector <48 x i8> %x2, <48 x i8> poison, <16 x i32> <i32 0, i32 3, i32 6, i32 9, i32 12, i32 15, i32 18, i32 21, i32 24, i32 27, i32 30, i32 33, i32 36, i32 39, i32 42, i32 45>
  %r = mul <16 x i8> %s1, %s2
  ret <16 x i8> %r
}