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

target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
target triple = "x86_64-unknown-unknown"

declare <4 x i32> @llvm.x86.avx512.vpdpwssd.128(<4 x i32>, <4 x i32>, <4 x i32>)
declare <8 x i32> @llvm.x86.avx512.vpdpwssd.256(<8 x i32>, <8 x i32>, <8 x i32>)
declare <4 x i32> @llvm.x86.avx512.vpdpwssds.128(<4 x i32>, <4 x i32>, <4 x i32>)
declare <8 x i32> @llvm.x86.avx512.vpdpwssds.256(<8 x i32>, <8 x i32>, <8 x i32>)
declare <4 x i32> @llvm.x86.avx512.vpdpbusd.128(<4 x i32>, <4 x i32>, <4 x i32>)
declare <8 x i32> @llvm.x86.avx512.vpdpbusd.256(<8 x i32>, <8 x i32>, <8 x i32>)
declare <4 x i32> @llvm.x86.avx512.vpdpbusds.128(<4 x i32>, <4 x i32>, <4 x i32>)
declare <8 x i32> @llvm.x86.avx512.vpdpbusds.256(<8 x i32>, <8 x i32>, <8 x i32>)

define <4 x i32> @stack_fold_vpdpwssd(<4 x i32> %a0, <4 x i32> %a1, <4 x i32> %a2) {
; CHECK-LABEL: stack_fold_vpdpwssd:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovaps %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    {vex} vpdpwssd {{[-0-9]+}}(%r{{[sb]}}p), %xmm1, %xmm0 # 16-byte Folded Reload
; CHECK-NEXT:    retq
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
  %2 = call <4 x i32> @llvm.x86.avx512.vpdpwssd.128(<4 x i32> %a0, <4 x i32> %a1, <4 x i32> %a2)
  ret <4 x i32> %2
}

define <4 x i32> @stack_fold_vpdpwssd_commuted(<4 x i32> %a0, <4 x i32> %a1, <4 x i32> %a2) {
; CHECK-LABEL: stack_fold_vpdpwssd_commuted:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovaps %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    {vex} vpdpwssd {{[-0-9]+}}(%r{{[sb]}}p), %xmm1, %xmm0 # 16-byte Folded Reload
; CHECK-NEXT:    retq
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
  %2 = call <4 x i32> @llvm.x86.avx512.vpdpwssd.128(<4 x i32> %a0, <4 x i32> %a2, <4 x i32> %a1)
  ret <4 x i32> %2
}

define <8 x i32> @stack_fold_vpdpwssd_256(<8 x i32> %a0, <8 x i32> %a1, <8 x i32> %a2) {
; CHECK-LABEL: stack_fold_vpdpwssd_256:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    {vex} vpdpwssd {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm0 # 32-byte Folded Reload
; CHECK-NEXT:    retq
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
  %2 = call <8 x i32> @llvm.x86.avx512.vpdpwssd.256(<8 x i32> %a0, <8 x i32> %a1, <8 x i32> %a2)
  ret <8 x i32> %2
}

define <8 x i32> @stack_fold_vpdpwssd_256_commuted(<8 x i32> %a0, <8 x i32> %a1, <8 x i32> %a2) {
; CHECK-LABEL: stack_fold_vpdpwssd_256_commuted:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    {vex} vpdpwssd {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm0 # 32-byte Folded Reload
; CHECK-NEXT:    retq
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
  %2 = call <8 x i32> @llvm.x86.avx512.vpdpwssd.256(<8 x i32> %a0, <8 x i32> %a2, <8 x i32> %a1)
  ret <8 x i32> %2
}

define <4 x i32> @stack_fold_vpdpwssds(<4 x i32> %a0, <4 x i32> %a1, <4 x i32> %a2) {
; CHECK-LABEL: stack_fold_vpdpwssds:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovaps %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    {vex} vpdpwssds {{[-0-9]+}}(%r{{[sb]}}p), %xmm1, %xmm0 # 16-byte Folded Reload
; CHECK-NEXT:    retq
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
  %2 = call <4 x i32> @llvm.x86.avx512.vpdpwssds.128(<4 x i32> %a0, <4 x i32> %a1, <4 x i32> %a2)
  ret <4 x i32> %2
}

define <4 x i32> @stack_fold_vpdpwssds_commuted(<4 x i32> %a0, <4 x i32> %a1, <4 x i32> %a2) {
; CHECK-LABEL: stack_fold_vpdpwssds_commuted:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovaps %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    {vex} vpdpwssds {{[-0-9]+}}(%r{{[sb]}}p), %xmm1, %xmm0 # 16-byte Folded Reload
; CHECK-NEXT:    retq
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
  %2 = call <4 x i32> @llvm.x86.avx512.vpdpwssds.128(<4 x i32> %a0, <4 x i32> %a2, <4 x i32> %a1)
  ret <4 x i32> %2
}

define <8 x i32> @stack_fold_vpdpwssds_256(<8 x i32> %a0, <8 x i32> %a1, <8 x i32> %a2) {
; CHECK-LABEL: stack_fold_vpdpwssds_256:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    {vex} vpdpwssds {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm0 # 32-byte Folded Reload
; CHECK-NEXT:    retq
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
  %2 = call <8 x i32> @llvm.x86.avx512.vpdpwssds.256(<8 x i32> %a0, <8 x i32> %a1, <8 x i32> %a2)
  ret <8 x i32> %2
}

define <8 x i32> @stack_fold_vpdpwssds_256_commuted(<8 x i32> %a0, <8 x i32> %a1, <8 x i32> %a2) {
; CHECK-LABEL: stack_fold_vpdpwssds_256_commuted:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    {vex} vpdpwssds {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm0 # 32-byte Folded Reload
; CHECK-NEXT:    retq
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
  %2 = call <8 x i32> @llvm.x86.avx512.vpdpwssds.256(<8 x i32> %a0, <8 x i32> %a2, <8 x i32> %a1)
  ret <8 x i32> %2
}

define <4 x i32> @stack_fold_vpdpbusd(<4 x i32> %a0, <4 x i32> %a1, <4 x i32> %a2) {
; CHECK-LABEL: stack_fold_vpdpbusd:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovaps %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    {vex} vpdpbusd {{[-0-9]+}}(%r{{[sb]}}p), %xmm1, %xmm0 # 16-byte Folded Reload
; CHECK-NEXT:    retq
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
  %2 = call <4 x i32> @llvm.x86.avx512.vpdpbusd.128(<4 x i32> %a0, <4 x i32> %a1, <4 x i32> %a2)
  ret <4 x i32> %2
}

define <4 x i32> @stack_fold_vpdpbusd_commuted(<4 x i32> %a0, <4 x i32> %a1, <4 x i32> %a2) {
; CHECK-LABEL: stack_fold_vpdpbusd_commuted:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovaps %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm2 # 16-byte Reload
; CHECK-NEXT:    {vex} vpdpbusd %xmm1, %xmm2, %xmm0
; CHECK-NEXT:    retq
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
  %2 = call <4 x i32> @llvm.x86.avx512.vpdpbusd.128(<4 x i32> %a0, <4 x i32> %a2, <4 x i32> %a1)
  ret <4 x i32> %2
}

define <8 x i32> @stack_fold_vpdpbusd_256(<8 x i32> %a0, <8 x i32> %a1, <8 x i32> %a2) {
; CHECK-LABEL: stack_fold_vpdpbusd_256:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    {vex} vpdpbusd {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm0 # 32-byte Folded Reload
; CHECK-NEXT:    retq
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
  %2 = call <8 x i32> @llvm.x86.avx512.vpdpbusd.256(<8 x i32> %a0, <8 x i32> %a1, <8 x i32> %a2)
  ret <8 x i32> %2
}

define <8 x i32> @stack_fold_vpdpbusd_256_commuted(<8 x i32> %a0, <8 x i32> %a1, <8 x i32> %a2) {
; CHECK-LABEL: stack_fold_vpdpbusd_256_commuted:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm2 # 32-byte Reload
; CHECK-NEXT:    {vex} vpdpbusd %ymm1, %ymm2, %ymm0
; CHECK-NEXT:    retq
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
  %2 = call <8 x i32> @llvm.x86.avx512.vpdpbusd.256(<8 x i32> %a0, <8 x i32> %a2, <8 x i32> %a1)
  ret <8 x i32> %2
}

define <4 x i32> @stack_fold_vpdpbusds(<4 x i32> %a0, <4 x i32> %a1, <4 x i32> %a2) {
; CHECK-LABEL: stack_fold_vpdpbusds:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovaps %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    {vex} vpdpbusds {{[-0-9]+}}(%r{{[sb]}}p), %xmm1, %xmm0 # 16-byte Folded Reload
; CHECK-NEXT:    retq
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
  %2 = call <4 x i32> @llvm.x86.avx512.vpdpbusds.128(<4 x i32> %a0, <4 x i32> %a1, <4 x i32> %a2)
  ret <4 x i32> %2
}

define <4 x i32> @stack_fold_vpdpbusds_commuted(<4 x i32> %a0, <4 x i32> %a1, <4 x i32> %a2) {
; CHECK-LABEL: stack_fold_vpdpbusds_commuted:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovaps %xmm2, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    vmovdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm2 # 16-byte Reload
; CHECK-NEXT:    {vex} vpdpbusds %xmm1, %xmm2, %xmm0
; CHECK-NEXT:    retq
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
  %2 = call <4 x i32> @llvm.x86.avx512.vpdpbusds.128(<4 x i32> %a0, <4 x i32> %a2, <4 x i32> %a1)
  ret <4 x i32> %2
}

define <8 x i32> @stack_fold_vpdpbusds_256(<8 x i32> %a0, <8 x i32> %a1, <8 x i32> %a2) {
; CHECK-LABEL: stack_fold_vpdpbusds_256:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    {vex} vpdpbusds {{[-0-9]+}}(%r{{[sb]}}p), %ymm1, %ymm0 # 32-byte Folded Reload
; CHECK-NEXT:    retq
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
  %2 = call <8 x i32> @llvm.x86.avx512.vpdpbusds.256(<8 x i32> %a0, <8 x i32> %a1, <8 x i32> %a2)
  ret <8 x i32> %2
}

define <8 x i32> @stack_fold_vpdpbusds_256_commuted(<8 x i32> %a0, <8 x i32> %a1, <8 x i32> %a2) {
; CHECK-LABEL: stack_fold_vpdpbusds_256_commuted:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmovups %ymm2, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; CHECK-NEXT:    #APP
; CHECK-NEXT:    nop
; CHECK-NEXT:    #NO_APP
; CHECK-NEXT:    vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm2 # 32-byte Reload
; CHECK-NEXT:    {vex} vpdpbusds %ymm1, %ymm2, %ymm0
; CHECK-NEXT:    retq
  %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
  %2 = call <8 x i32> @llvm.x86.avx512.vpdpbusds.256(<8 x i32> %a0, <8 x i32> %a2, <8 x i32> %a1)
  ret <8 x i32> %2
}