llvm/llvm/test/CodeGen/X86/stack-folding-int-avxvnniint16.ll

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -O3 -disable-peephole -verify-machineinstrs -mtriple=x86_64-unknown-unknown --show-mc-encoding -mattr=+avxvnniint16 | FileCheck %s

declare <4 x i32> @llvm.x86.avx2.vpdpwsud.128(<4 x i32> %A, <4 x i32> %B, <4 x i32> %C)
declare <8 x i32> @llvm.x86.avx2.vpdpwsud.256(<8 x i32> %A, <8 x i32> %B, <8 x i32> %C)
declare <4 x i32> @llvm.x86.avx2.vpdpwsuds.128(<4 x i32> %A, <4 x i32> %B, <4 x i32> %C)
declare <8 x i32> @llvm.x86.avx2.vpdpwsuds.256(<8 x i32> %A, <8 x i32> %B, <8 x i32> %C)
declare <4 x i32> @llvm.x86.avx2.vpdpwusd.128(<4 x i32> %A, <4 x i32> %B, <4 x i32> %C)
declare <8 x i32> @llvm.x86.avx2.vpdpwusd.256(<8 x i32> %A, <8 x i32> %B, <8 x i32> %C)
declare <4 x i32> @llvm.x86.avx2.vpdpwusds.128(<4 x i32> %A, <4 x i32> %B, <4 x i32> %C)
declare <8 x i32> @llvm.x86.avx2.vpdpwusds.256(<8 x i32> %A, <8 x i32> %B, <8 x i32> %C)
declare <4 x i32> @llvm.x86.avx2.vpdpwuud.128(<4 x i32> %A, <4 x i32> %B, <4 x i32> %C)
declare <8 x i32> @llvm.x86.avx2.vpdpwuud.256(<8 x i32> %A, <8 x i32> %B, <8 x i32> %C)
declare <4 x i32> @llvm.x86.avx2.vpdpwuuds.128(<4 x i32> %A, <4 x i32> %B, <4 x i32> %C)
declare <8 x i32> @llvm.x86.avx2.vpdpwuuds.256(<8 x i32> %A, <8 x i32> %B, <8 x i32> %C)

define <4 x i32> @test_int_x86_avx2_vpdpwsud_128(<4 x i32> %A, <4 x i32> %B, <4 x i32> %C) {
; CHECK-LABEL: test_int_x86_avx2_vpdpwsud_128:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovaps %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; CHECK-NEXT:    # encoding: [0xc5,0xf8,0x29,0x54,0x24,0xe8]
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop # encoding: [0x90]
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    vpdpwsud {{[-0-9]+}}(%r{{[sb]}}p), %xmm1, %xmm0 # 16-byte Folded Reload
; CHECK-NEXT:    # encoding: [0xc4,0xe2,0x72,0xd2,0x44,0x24,0xe8]
; CHECK-NEXT:    retq # encoding: [0xc3]
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{xmm16},~{xmm17},~{xmm18},~{xmm19},~{xmm20},~{xmm21},~{xmm22},~{xmm23},~{xmm24},~{xmm25},~{xmm26},~{xmm27},~{xmm28},~{xmm29},~{xmm30},~{xmm31},~{flags}"()
  %ret = call <4 x i32> @llvm.x86.avx2.vpdpwsud.128(<4 x i32> %A, <4 x i32> %B, <4 x i32> %C)
  ret <4 x i32> %ret
}

define <8 x i32> @test_int_x86_avx2_vpdpwsud_256(<8 x i32> %A, <8 x i32> %B, <8 x i32> %C) {
; CHECK-LABEL: test_int_x86_avx2_vpdpwsud_256:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; CHECK-NEXT:    # encoding: [0xc5,0xfc,0x11,0x54,0x24,0xd8]
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop # encoding: [0x90]
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    vpdpwsud {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm0 # 32-byte Folded Reload
; CHECK-NEXT:    # encoding: [0xc4,0xe2,0x76,0xd2,0x44,0x24,0xd8]
; CHECK-NEXT:    retq # encoding: [0xc3]
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{xmm16},~{xmm17},~{xmm18},~{xmm19},~{xmm20},~{xmm21},~{xmm22},~{xmm23},~{xmm24},~{xmm25},~{xmm26},~{xmm27},~{xmm28},~{xmm29},~{xmm30},~{xmm31},~{flags}"()
  %ret = call <8 x i32> @llvm.x86.avx2.vpdpwsud.256(<8 x i32> %A, <8 x i32> %B, <8 x i32> %C)
  ret <8 x i32> %ret
}

define <4 x i32> @test_int_x86_avx2_vpdpwsuds_128(<4 x i32> %A, <4 x i32> %B, <4 x i32> %C) {
; CHECK-LABEL: test_int_x86_avx2_vpdpwsuds_128:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovaps %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; CHECK-NEXT:    # encoding: [0xc5,0xf8,0x29,0x54,0x24,0xe8]
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop # encoding: [0x90]
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    vpdpwsuds {{[-0-9]+}}(%r{{[sb]}}p), %xmm1, %xmm0 # 16-byte Folded Reload
; CHECK-NEXT:    # encoding: [0xc4,0xe2,0x72,0xd3,0x44,0x24,0xe8]
; CHECK-NEXT:    retq # encoding: [0xc3]
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{xmm16},~{xmm17},~{xmm18},~{xmm19},~{xmm20},~{xmm21},~{xmm22},~{xmm23},~{xmm24},~{xmm25},~{xmm26},~{xmm27},~{xmm28},~{xmm29},~{xmm30},~{xmm31},~{flags}"()
  %ret = call <4 x i32> @llvm.x86.avx2.vpdpwsuds.128(<4 x i32> %A, <4 x i32> %B, <4 x i32> %C)
  ret <4 x i32> %ret
}

define <8 x i32> @test_int_x86_avx2_vpdpwsuds_256(<8 x i32> %A, <8 x i32> %B, <8 x i32> %C) {
; CHECK-LABEL: test_int_x86_avx2_vpdpwsuds_256:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; CHECK-NEXT:    # encoding: [0xc5,0xfc,0x11,0x54,0x24,0xd8]
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop # encoding: [0x90]
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    vpdpwsuds {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm0 # 32-byte Folded Reload
; CHECK-NEXT:    # encoding: [0xc4,0xe2,0x76,0xd3,0x44,0x24,0xd8]
; CHECK-NEXT:    retq # encoding: [0xc3]
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{xmm16},~{xmm17},~{xmm18},~{xmm19},~{xmm20},~{xmm21},~{xmm22},~{xmm23},~{xmm24},~{xmm25},~{xmm26},~{xmm27},~{xmm28},~{xmm29},~{xmm30},~{xmm31},~{flags}"()
  %ret = call <8 x i32> @llvm.x86.avx2.vpdpwsuds.256(<8 x i32> %A, <8 x i32> %B, <8 x i32> %C)
  ret <8 x i32> %ret
}

define <4 x i32> @test_int_x86_avx2_vpdpwusd_128(<4 x i32> %A, <4 x i32> %B, <4 x i32> %C) {
; CHECK-LABEL: test_int_x86_avx2_vpdpwusd_128:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovaps %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; CHECK-NEXT:    # encoding: [0xc5,0xf8,0x29,0x54,0x24,0xe8]
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop # encoding: [0x90]
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    vpdpwusd {{[-0-9]+}}(%r{{[sb]}}p), %xmm1, %xmm0 # 16-byte Folded Reload
; CHECK-NEXT:    # encoding: [0xc4,0xe2,0x71,0xd2,0x44,0x24,0xe8]
; CHECK-NEXT:    retq # encoding: [0xc3]
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{xmm16},~{xmm17},~{xmm18},~{xmm19},~{xmm20},~{xmm21},~{xmm22},~{xmm23},~{xmm24},~{xmm25},~{xmm26},~{xmm27},~{xmm28},~{xmm29},~{xmm30},~{xmm31},~{flags}"()
  %ret = call <4 x i32> @llvm.x86.avx2.vpdpwusd.128(<4 x i32> %A, <4 x i32> %B, <4 x i32> %C)
  ret <4 x i32> %ret
}

define <8 x i32> @test_int_x86_avx2_vpdpwusd_256(<8 x i32> %A, <8 x i32> %B, <8 x i32> %C) {
; CHECK-LABEL: test_int_x86_avx2_vpdpwusd_256:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; CHECK-NEXT:    # encoding: [0xc5,0xfc,0x11,0x54,0x24,0xd8]
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop # encoding: [0x90]
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    vpdpwusd {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm0 # 32-byte Folded Reload
; CHECK-NEXT:    # encoding: [0xc4,0xe2,0x75,0xd2,0x44,0x24,0xd8]
; CHECK-NEXT:    retq # encoding: [0xc3]
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{xmm16},~{xmm17},~{xmm18},~{xmm19},~{xmm20},~{xmm21},~{xmm22},~{xmm23},~{xmm24},~{xmm25},~{xmm26},~{xmm27},~{xmm28},~{xmm29},~{xmm30},~{xmm31},~{flags}"()
  %ret = call <8 x i32> @llvm.x86.avx2.vpdpwusd.256(<8 x i32> %A, <8 x i32> %B, <8 x i32> %C)
  ret <8 x i32> %ret
}

define <4 x i32> @test_int_x86_avx2_vpdpwusds_128(<4 x i32> %A, <4 x i32> %B, <4 x i32> %C) {
; CHECK-LABEL: test_int_x86_avx2_vpdpwusds_128:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovaps %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; CHECK-NEXT:    # encoding: [0xc5,0xf8,0x29,0x54,0x24,0xe8]
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop # encoding: [0x90]
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    vpdpwusds {{[-0-9]+}}(%r{{[sb]}}p), %xmm1, %xmm0 # 16-byte Folded Reload
; CHECK-NEXT:    # encoding: [0xc4,0xe2,0x71,0xd3,0x44,0x24,0xe8]
; CHECK-NEXT:    retq # encoding: [0xc3]
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{xmm16},~{xmm17},~{xmm18},~{xmm19},~{xmm20},~{xmm21},~{xmm22},~{xmm23},~{xmm24},~{xmm25},~{xmm26},~{xmm27},~{xmm28},~{xmm29},~{xmm30},~{xmm31},~{flags}"()
  %ret = call <4 x i32> @llvm.x86.avx2.vpdpwusds.128(<4 x i32> %A, <4 x i32> %B, <4 x i32> %C)
  ret <4 x i32> %ret
}

define <8 x i32> @test_int_x86_avx2_vpdpwusds_256(<8 x i32> %A, <8 x i32> %B, <8 x i32> %C) {
; CHECK-LABEL: test_int_x86_avx2_vpdpwusds_256:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; CHECK-NEXT:    # encoding: [0xc5,0xfc,0x11,0x54,0x24,0xd8]
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop # encoding: [0x90]
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    vpdpwusds {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm0 # 32-byte Folded Reload
; CHECK-NEXT:    # encoding: [0xc4,0xe2,0x75,0xd3,0x44,0x24,0xd8]
; CHECK-NEXT:    retq # encoding: [0xc3]
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{xmm16},~{xmm17},~{xmm18},~{xmm19},~{xmm20},~{xmm21},~{xmm22},~{xmm23},~{xmm24},~{xmm25},~{xmm26},~{xmm27},~{xmm28},~{xmm29},~{xmm30},~{xmm31},~{flags}"()
  %ret = call <8 x i32> @llvm.x86.avx2.vpdpwusds.256(<8 x i32> %A, <8 x i32> %B, <8 x i32> %C)
  ret <8 x i32> %ret
}

define <4 x i32> @test_int_x86_avx2_vpdpwuud_128(<4 x i32> %A, <4 x i32> %B, <4 x i32> %C) {
; CHECK-LABEL: test_int_x86_avx2_vpdpwuud_128:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovaps %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; CHECK-NEXT:    # encoding: [0xc5,0xf8,0x29,0x54,0x24,0xe8]
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop # encoding: [0x90]
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    vpdpwuud {{[-0-9]+}}(%r{{[sb]}}p), %xmm1, %xmm0 # 16-byte Folded Reload
; CHECK-NEXT:    # encoding: [0xc4,0xe2,0x70,0xd2,0x44,0x24,0xe8]
; CHECK-NEXT:    retq # encoding: [0xc3]
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{xmm16},~{xmm17},~{xmm18},~{xmm19},~{xmm20},~{xmm21},~{xmm22},~{xmm23},~{xmm24},~{xmm25},~{xmm26},~{xmm27},~{xmm28},~{xmm29},~{xmm30},~{xmm31},~{flags}"()
  %ret = call <4 x i32> @llvm.x86.avx2.vpdpwuud.128(<4 x i32> %A, <4 x i32> %B, <4 x i32> %C)
  ret <4 x i32> %ret
}

define <4 x i32> @test_int_x86_avx2_vpdpwuud_128_commuted(<4 x i32> %A, <4 x i32> %B, <4 x i32> %C) {
; CHECK-LABEL: test_int_x86_avx2_vpdpwuud_128_commuted:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovaps %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; CHECK-NEXT:    # encoding: [0xc5,0xf8,0x29,0x54,0x24,0xe8]
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop # encoding: [0x90]
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    vpdpwuud {{[-0-9]+}}(%r{{[sb]}}p), %xmm1, %xmm0 # 16-byte Folded Reload
; CHECK-NEXT:    # encoding: [0xc4,0xe2,0x70,0xd2,0x44,0x24,0xe8]
; CHECK-NEXT:    retq # encoding: [0xc3]
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{xmm16},~{xmm17},~{xmm18},~{xmm19},~{xmm20},~{xmm21},~{xmm22},~{xmm23},~{xmm24},~{xmm25},~{xmm26},~{xmm27},~{xmm28},~{xmm29},~{xmm30},~{xmm31},~{flags}"()
  %ret = call <4 x i32> @llvm.x86.avx2.vpdpwuud.128(<4 x i32> %A, <4 x i32> %C, <4 x i32> %B)
  ret <4 x i32> %ret
}

define <8 x i32> @test_int_x86_avx2_vpdpwuud_256(<8 x i32> %A, <8 x i32> %B, <8 x i32> %C) {
; CHECK-LABEL: test_int_x86_avx2_vpdpwuud_256:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; CHECK-NEXT:    # encoding: [0xc5,0xfc,0x11,0x54,0x24,0xd8]
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop # encoding: [0x90]
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    vpdpwuud {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm0 # 32-byte Folded Reload
; CHECK-NEXT:    # encoding: [0xc4,0xe2,0x74,0xd2,0x44,0x24,0xd8]
; CHECK-NEXT:    retq # encoding: [0xc3]
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{xmm16},~{xmm17},~{xmm18},~{xmm19},~{xmm20},~{xmm21},~{xmm22},~{xmm23},~{xmm24},~{xmm25},~{xmm26},~{xmm27},~{xmm28},~{xmm29},~{xmm30},~{xmm31},~{flags}"()
  %ret = call <8 x i32> @llvm.x86.avx2.vpdpwuud.256(<8 x i32> %A, <8 x i32> %B, <8 x i32> %C)
  ret <8 x i32> %ret
}

define <8 x i32> @test_int_x86_avx2_vpdpwuud_256_commuted(<8 x i32> %A, <8 x i32> %B, <8 x i32> %C) {
; CHECK-LABEL: test_int_x86_avx2_vpdpwuud_256_commuted:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; CHECK-NEXT:    # encoding: [0xc5,0xfc,0x11,0x54,0x24,0xd8]
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop # encoding: [0x90]
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    vpdpwuud {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm0 # 32-byte Folded Reload
; CHECK-NEXT:    # encoding: [0xc4,0xe2,0x74,0xd2,0x44,0x24,0xd8]
; CHECK-NEXT:    retq # encoding: [0xc3]
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{xmm16},~{xmm17},~{xmm18},~{xmm19},~{xmm20},~{xmm21},~{xmm22},~{xmm23},~{xmm24},~{xmm25},~{xmm26},~{xmm27},~{xmm28},~{xmm29},~{xmm30},~{xmm31},~{flags}"()
  %ret = call <8 x i32> @llvm.x86.avx2.vpdpwuud.256(<8 x i32> %A, <8 x i32> %C, <8 x i32> %B)
  ret <8 x i32> %ret
}

define <4 x i32> @test_int_x86_avx2_vpdpwuuds_128(<4 x i32> %A, <4 x i32> %B, <4 x i32> %C) {
; CHECK-LABEL: test_int_x86_avx2_vpdpwuuds_128:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovaps %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; CHECK-NEXT:    # encoding: [0xc5,0xf8,0x29,0x54,0x24,0xe8]
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop # encoding: [0x90]
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    vpdpwuuds {{[-0-9]+}}(%r{{[sb]}}p), %xmm1, %xmm0 # 16-byte Folded Reload
; CHECK-NEXT:    # encoding: [0xc4,0xe2,0x70,0xd3,0x44,0x24,0xe8]
; CHECK-NEXT:    retq # encoding: [0xc3]
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{xmm16},~{xmm17},~{xmm18},~{xmm19},~{xmm20},~{xmm21},~{xmm22},~{xmm23},~{xmm24},~{xmm25},~{xmm26},~{xmm27},~{xmm28},~{xmm29},~{xmm30},~{xmm31},~{flags}"()
  %ret = call <4 x i32> @llvm.x86.avx2.vpdpwuuds.128(<4 x i32> %A, <4 x i32> %B, <4 x i32> %C)
  ret <4 x i32> %ret
}

define <4 x i32> @test_int_x86_avx2_vpdpwuuds_128_commuted(<4 x i32> %A, <4 x i32> %B, <4 x i32> %C) {
; CHECK-LABEL: test_int_x86_avx2_vpdpwuuds_128_commuted:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovaps %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; CHECK-NEXT:    # encoding: [0xc5,0xf8,0x29,0x54,0x24,0xe8]
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop # encoding: [0x90]
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    vpdpwuuds {{[-0-9]+}}(%r{{[sb]}}p), %xmm1, %xmm0 # 16-byte Folded Reload
; CHECK-NEXT:    # encoding: [0xc4,0xe2,0x70,0xd3,0x44,0x24,0xe8]
; CHECK-NEXT:    retq # encoding: [0xc3]
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{xmm16},~{xmm17},~{xmm18},~{xmm19},~{xmm20},~{xmm21},~{xmm22},~{xmm23},~{xmm24},~{xmm25},~{xmm26},~{xmm27},~{xmm28},~{xmm29},~{xmm30},~{xmm31},~{flags}"()
  %ret = call <4 x i32> @llvm.x86.avx2.vpdpwuuds.128(<4 x i32> %A, <4 x i32> %C, <4 x i32> %B)
  ret <4 x i32> %ret
}

define <8 x i32> @test_int_x86_avx2_vpdpwuuds_256(<8 x i32> %A, <8 x i32> %B, <8 x i32> %C) {
; CHECK-LABEL: test_int_x86_avx2_vpdpwuuds_256:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; CHECK-NEXT:    # encoding: [0xc5,0xfc,0x11,0x54,0x24,0xd8]
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop # encoding: [0x90]
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    vpdpwuuds {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm0 # 32-byte Folded Reload
; CHECK-NEXT:    # encoding: [0xc4,0xe2,0x74,0xd3,0x44,0x24,0xd8]
; CHECK-NEXT:    retq # encoding: [0xc3]
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{xmm16},~{xmm17},~{xmm18},~{xmm19},~{xmm20},~{xmm21},~{xmm22},~{xmm23},~{xmm24},~{xmm25},~{xmm26},~{xmm27},~{xmm28},~{xmm29},~{xmm30},~{xmm31},~{flags}"()
  %ret = call <8 x i32> @llvm.x86.avx2.vpdpwuuds.256(<8 x i32> %A, <8 x i32> %B, <8 x i32> %C)
  ret <8 x i32> %ret
}

define <8 x i32> @test_int_x86_avx2_vpdpwuuds_256_commuted(<8 x i32> %A, <8 x i32> %B, <8 x i32> %C) {
; CHECK-LABEL: test_int_x86_avx2_vpdpwuuds_256_commuted:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; CHECK-NEXT:    # encoding: [0xc5,0xfc,0x11,0x54,0x24,0xd8]
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop # encoding: [0x90]
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    vpdpwuuds {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm0 # 32-byte Folded Reload
; CHECK-NEXT:    # encoding: [0xc4,0xe2,0x74,0xd3,0x44,0x24,0xd8]
; CHECK-NEXT:    retq # encoding: [0xc3]
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{xmm16},~{xmm17},~{xmm18},~{xmm19},~{xmm20},~{xmm21},~{xmm22},~{xmm23},~{xmm24},~{xmm25},~{xmm26},~{xmm27},~{xmm28},~{xmm29},~{xmm30},~{xmm31},~{flags}"()
  %ret = call <8 x i32> @llvm.x86.avx2.vpdpwuuds.256(<8 x i32> %A, <8 x i32> %C, <8 x i32> %B)
  ret <8 x i32> %ret
}