; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=amdgcn -mcpu=gfx700 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX7 %s
; RUN: llc -mtriple=amdgcn -mcpu=gfx803 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX8 %s
; RUN: llc -mtriple=amdgcn -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX9 %s
; RUN: llc -mtriple=amdgcn -mcpu=gfx906 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX9-DL %s
; RUN: llc -mtriple=amdgcn -mcpu=gfx1011 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX10-DL %s
; RUN: llc -mtriple=amdgcn -mcpu=gfx1012 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX10-DL %s

define amdgpu_kernel void @udot8_acc32(<8 x i4> addrspace(1)* %src1,
; GFX7-LABEL: udot8_acc32:
; GFX7:       ; %bb.0: ; %entry
; GFX7-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x9
; GFX7-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0xd
; GFX7-NEXT:    s_mov_b32 s3, 0xf000
; GFX7-NEXT:    s_mov_b32 s2, -1
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    s_load_dword s6, s[6:7], 0x0
; GFX7-NEXT:    s_load_dword s4, s[4:5], 0x0
; GFX7-NEXT:    s_load_dword s20, s[0:1], 0x0
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    s_lshr_b32 s7, s6, 28
; GFX7-NEXT:    s_bfe_u32 s14, s6, 0x40018
; GFX7-NEXT:    s_bfe_u32 s15, s6, 0x40014
; GFX7-NEXT:    s_bfe_u32 s16, s6, 0x40010
; GFX7-NEXT:    s_bfe_u32 s17, s6, 0x4000c
; GFX7-NEXT:    s_bfe_u32 s18, s6, 0x40008
; GFX7-NEXT:    s_bfe_u32 s19, s6, 0x40004
; GFX7-NEXT:    s_and_b32 s6, s6, 15
; GFX7-NEXT:    s_lshr_b32 s5, s4, 28
; GFX7-NEXT:    s_bfe_u32 s8, s4, 0x40018
; GFX7-NEXT:    s_bfe_u32 s9, s4, 0x40014
; GFX7-NEXT:    s_bfe_u32 s10, s4, 0x40010
; GFX7-NEXT:    s_bfe_u32 s11, s4, 0x4000c
; GFX7-NEXT:    s_bfe_u32 s12, s4, 0x40008
; GFX7-NEXT:    s_bfe_u32 s13, s4, 0x40004
; GFX7-NEXT:    s_and_b32 s4, s4, 15
; GFX7-NEXT:    v_mov_b32_e32 v0, s6
; GFX7-NEXT:    v_mov_b32_e32 v1, s20
; GFX7-NEXT:    v_mad_u32_u24 v0, s4, v0, v1
; GFX7-NEXT:    v_mov_b32_e32 v1, s19
; GFX7-NEXT:    v_mad_u32_u24 v0, s13, v1, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s18
; GFX7-NEXT:    v_mad_u32_u24 v0, s12, v1, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s17
; GFX7-NEXT:    v_mad_u32_u24 v0, s11, v1, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s16
; GFX7-NEXT:    v_mad_u32_u24 v0, s10, v1, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s15
; GFX7-NEXT:    v_mad_u32_u24 v0, s9, v1, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s14
; GFX7-NEXT:    v_mad_u32_u24 v0, s8, v1, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s7
; GFX7-NEXT:    v_mad_u32_u24 v0, s5, v1, v0
; GFX7-NEXT:    buffer_store_dword v0, off, s[0:3], 0
; GFX7-NEXT:    s_endpgm
;
; GFX8-LABEL: udot8_acc32:
; GFX8:       ; %bb.0: ; %entry
; GFX8-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX8-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    s_load_dword s6, s[6:7], 0x0
; GFX8-NEXT:    s_load_dword s2, s[4:5], 0x0
; GFX8-NEXT:    s_load_dword s18, s[0:1], 0x0
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    s_lshr_b32 s7, s6, 28
; GFX8-NEXT:    s_bfe_u32 s12, s6, 0x40018
; GFX8-NEXT:    s_bfe_u32 s13, s6, 0x40014
; GFX8-NEXT:    s_bfe_u32 s14, s6, 0x40010
; GFX8-NEXT:    s_bfe_u32 s15, s6, 0x4000c
; GFX8-NEXT:    s_bfe_u32 s16, s6, 0x40008
; GFX8-NEXT:    s_bfe_u32 s17, s6, 0x40004
; GFX8-NEXT:    s_and_b32 s6, s6, 15
; GFX8-NEXT:    s_lshr_b32 s3, s2, 28
; GFX8-NEXT:    s_bfe_u32 s4, s2, 0x40018
; GFX8-NEXT:    s_bfe_u32 s5, s2, 0x40014
; GFX8-NEXT:    s_bfe_u32 s8, s2, 0x40010
; GFX8-NEXT:    s_bfe_u32 s9, s2, 0x4000c
; GFX8-NEXT:    s_bfe_u32 s10, s2, 0x40008
; GFX8-NEXT:    s_bfe_u32 s11, s2, 0x40004
; GFX8-NEXT:    s_and_b32 s2, s2, 15
; GFX8-NEXT:    v_mov_b32_e32 v0, s6
; GFX8-NEXT:    v_mov_b32_e32 v1, s18
; GFX8-NEXT:    v_mad_u32_u24 v0, s2, v0, v1
; GFX8-NEXT:    v_mov_b32_e32 v1, s17
; GFX8-NEXT:    v_mad_u32_u24 v0, s11, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v1, s16
; GFX8-NEXT:    v_mad_u32_u24 v0, s10, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v1, s15
; GFX8-NEXT:    v_mad_u32_u24 v0, s9, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v1, s14
; GFX8-NEXT:    v_mad_u32_u24 v0, s8, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v1, s13
; GFX8-NEXT:    v_mad_u32_u24 v0, s5, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v1, s12
; GFX8-NEXT:    v_mad_u32_u24 v0, s4, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v1, s7
; GFX8-NEXT:    v_mad_u32_u24 v2, s3, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v0, s0
; GFX8-NEXT:    v_mov_b32_e32 v1, s1
; GFX8-NEXT:    flat_store_dword v[0:1], v2
; GFX8-NEXT:    s_endpgm
;
; GFX9-LABEL: udot8_acc32:
; GFX9:       ; %bb.0: ; %entry
; GFX9-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    s_load_dword s6, s[6:7], 0x0
; GFX9-NEXT:    s_load_dword s2, s[4:5], 0x0
; GFX9-NEXT:    s_load_dword s18, s[0:1], 0x0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    s_lshr_b32 s7, s6, 28
; GFX9-NEXT:    s_bfe_u32 s12, s6, 0x40018
; GFX9-NEXT:    s_bfe_u32 s13, s6, 0x40014
; GFX9-NEXT:    s_bfe_u32 s14, s6, 0x40010
; GFX9-NEXT:    s_bfe_u32 s15, s6, 0x4000c
; GFX9-NEXT:    s_bfe_u32 s16, s6, 0x40008
; GFX9-NEXT:    s_bfe_u32 s17, s6, 0x40004
; GFX9-NEXT:    s_and_b32 s6, s6, 15
; GFX9-NEXT:    s_lshr_b32 s3, s2, 28
; GFX9-NEXT:    s_bfe_u32 s4, s2, 0x40018
; GFX9-NEXT:    s_bfe_u32 s5, s2, 0x40014
; GFX9-NEXT:    s_bfe_u32 s8, s2, 0x40010
; GFX9-NEXT:    s_bfe_u32 s9, s2, 0x4000c
; GFX9-NEXT:    s_bfe_u32 s10, s2, 0x40008
; GFX9-NEXT:    s_bfe_u32 s11, s2, 0x40004
; GFX9-NEXT:    s_and_b32 s2, s2, 15
; GFX9-NEXT:    v_mov_b32_e32 v0, s6
; GFX9-NEXT:    v_mov_b32_e32 v1, s18
; GFX9-NEXT:    v_mad_u32_u24 v0, s2, v0, v1
; GFX9-NEXT:    v_mov_b32_e32 v1, s17
; GFX9-NEXT:    v_mad_u32_u24 v0, s11, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, s16
; GFX9-NEXT:    v_mad_u32_u24 v0, s10, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, s15
; GFX9-NEXT:    v_mad_u32_u24 v0, s9, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, s14
; GFX9-NEXT:    v_mad_u32_u24 v0, s8, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, s13
; GFX9-NEXT:    v_mad_u32_u24 v0, s5, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, s12
; GFX9-NEXT:    v_mad_u32_u24 v0, s4, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, s7
; GFX9-NEXT:    v_mad_u32_u24 v2, s3, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-NEXT:    global_store_dword v[0:1], v2, off
; GFX9-NEXT:    s_endpgm
;
; GFX9-DL-LABEL: udot8_acc32:
; GFX9-DL:       ; %bb.0: ; %entry
; GFX9-DL-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-DL-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    s_load_dword s2, s[6:7], 0x0
; GFX9-DL-NEXT:    s_load_dword s3, s[0:1], 0x0
; GFX9-DL-NEXT:    s_load_dword s4, s[4:5], 0x0
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    v_mov_b32_e32 v0, s2
; GFX9-DL-NEXT:    v_mov_b32_e32 v1, s3
; GFX9-DL-NEXT:    v_dot8_u32_u4 v2, s4, v0, v1
; GFX9-DL-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-DL-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-DL-NEXT:    global_store_dword v[0:1], v2, off
; GFX9-DL-NEXT:    s_endpgm
;
; GFX10-DL-LABEL: udot8_acc32:
; GFX10-DL:       ; %bb.0: ; %entry
; GFX10-DL-NEXT:    s_clause 0x1
; GFX10-DL-NEXT:    s_load_dwordx2 s[4:5], s[0:1], 0x34
; GFX10-DL-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
; GFX10-DL-NEXT:    ; implicit-def: $vcc_hi
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    s_load_dword s6, s[4:5], 0x0
; GFX10-DL-NEXT:    s_load_dword s0, s[0:1], 0x0
; GFX10-DL-NEXT:    s_load_dword s1, s[2:3], 0x0
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    v_mov_b32_e32 v0, s6
; GFX10-DL-NEXT:    v_dot8_u32_u4 v2, s0, s1, v0
; GFX10-DL-NEXT:    v_mov_b32_e32 v0, s4
; GFX10-DL-NEXT:    v_mov_b32_e32 v1, s5
; GFX10-DL-NEXT:    global_store_dword v[0:1], v2, off
; GFX10-DL-NEXT:    s_endpgm
                                       <8 x i4> addrspace(1)* %src2,
                                       i32 addrspace(1)* nocapture %dst) {
entry:
  %vec1 = load <8 x i4>, <8 x i4> addrspace(1)* %src1
  %vec2 = load <8 x i4>, <8 x i4> addrspace(1)* %src2

  %v1e0 = extractelement <8 x i4> %vec1, i64 0
  %cv1e0 = zext i4 %v1e0 to i32
  %v2e0 = extractelement <8 x i4> %vec2, i64 0
  %cv2e0 = zext i4 %v2e0 to i32
  %mul0 = mul nuw nsw i32 %cv1e0, %cv2e0

  %v1e1 = extractelement <8 x i4> %vec1, i64 1
  %cv1e1 = zext i4 %v1e1 to i32
  %v2e1 = extractelement <8 x i4> %vec2, i64 1
  %cv2e1 = zext i4 %v2e1 to i32
  %mul1 = mul nuw nsw i32 %cv1e1, %cv2e1

  %v1e2 = extractelement <8 x i4> %vec1, i64 2
  %cv1e2 = zext i4 %v1e2 to i32
  %v2e2 = extractelement <8 x i4> %vec2, i64 2
  %cv2e2 = zext i4 %v2e2 to i32
  %mul2 = mul nuw nsw i32 %cv1e2, %cv2e2

  %v1e3 = extractelement <8 x i4> %vec1, i64 3
  %cv1e3 = zext i4 %v1e3 to i32
  %v2e3 = extractelement <8 x i4> %vec2, i64 3
  %cv2e3 = zext i4 %v2e3 to i32
  %mul3 = mul nuw nsw i32 %cv1e3, %cv2e3

  %v1e4 = extractelement <8 x i4> %vec1, i64 4
  %cv1e4 = zext i4 %v1e4 to i32
  %v2e4 = extractelement <8 x i4> %vec2, i64 4
  %cv2e4 = zext i4 %v2e4 to i32
  %mul4 = mul nuw nsw i32 %cv1e4, %cv2e4

  %v1e5 = extractelement <8 x i4> %vec1, i64 5
  %cv1e5 = zext i4 %v1e5 to i32
  %v2e5 = extractelement <8 x i4> %vec2, i64 5
  %cv2e5 = zext i4 %v2e5 to i32
  %mul5 = mul nuw nsw i32 %cv1e5, %cv2e5

  %v1e6 = extractelement <8 x i4> %vec1, i64 6
  %cv1e6 = zext i4 %v1e6 to i32
  %v2e6 = extractelement <8 x i4> %vec2, i64 6
  %cv2e6 = zext i4 %v2e6 to i32
  %mul6 = mul nuw nsw i32 %cv1e6, %cv2e6

  %v1e7 = extractelement <8 x i4> %vec1, i64 7
  %cv1e7 = zext i4 %v1e7 to i32
  %v2e7 = extractelement <8 x i4> %vec2, i64 7
  %cv2e7 = zext i4 %v2e7 to i32
  %mul7 = mul nuw nsw i32 %cv1e7, %cv2e7

  %acc = load i32, i32 addrspace(1)* %dst, align 4
  %add1 = add i32 %mul0, %acc
  %add2 = add i32 %add1, %mul1
  %add3 = add i32 %add2, %mul2
  %add4 = add i32 %add3, %mul3
  %add5 = add i32 %add4, %mul4
  %add6 = add i32 %add5, %mul5
  %add7 = add i32 %add6, %mul6
  %add8 = add i32 %add7, %mul7

  store i32 %add8, i32 addrspace(1)* %dst, align 4
  ret void
}

; TODO: Remove the unnecessary instruction(that is zero-extending the
; 2nd MAD) to have the pattern-recognizer to kick in.
define amdgpu_kernel void @udot8_acc16(<8 x i4> addrspace(1)* %src1,
; GFX7-LABEL: udot8_acc16:
; GFX7:       ; %bb.0: ; %entry
; GFX7-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x9
; GFX7-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0xd
; GFX7-NEXT:    s_mov_b32 s3, 0xf000
; GFX7-NEXT:    s_mov_b32 s2, -1
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    buffer_load_ushort v0, off, s[0:3], 0
; GFX7-NEXT:    s_load_dword s4, s[4:5], 0x0
; GFX7-NEXT:    s_load_dword s5, s[6:7], 0x0
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    s_lshr_b32 s6, s4, 28
; GFX7-NEXT:    s_bfe_u32 s14, s5, 0x40018
; GFX7-NEXT:    s_bfe_u32 s15, s5, 0x40014
; GFX7-NEXT:    s_bfe_u32 s16, s5, 0x40010
; GFX7-NEXT:    s_bfe_u32 s17, s5, 0x4000c
; GFX7-NEXT:    s_bfe_u32 s18, s5, 0x40008
; GFX7-NEXT:    s_bfe_u32 s19, s5, 0x40004
; GFX7-NEXT:    s_lshr_b32 s13, s5, 28
; GFX7-NEXT:    s_and_b32 s5, s5, 15
; GFX7-NEXT:    s_bfe_u32 s7, s4, 0x40018
; GFX7-NEXT:    s_bfe_u32 s8, s4, 0x40014
; GFX7-NEXT:    s_bfe_u32 s9, s4, 0x40010
; GFX7-NEXT:    s_bfe_u32 s10, s4, 0x4000c
; GFX7-NEXT:    s_bfe_u32 s11, s4, 0x40008
; GFX7-NEXT:    s_bfe_u32 s12, s4, 0x40004
; GFX7-NEXT:    s_and_b32 s4, s4, 15
; GFX7-NEXT:    v_mov_b32_e32 v1, s5
; GFX7-NEXT:    v_mov_b32_e32 v2, s19
; GFX7-NEXT:    v_mov_b32_e32 v3, s18
; GFX7-NEXT:    v_mov_b32_e32 v4, s17
; GFX7-NEXT:    v_mov_b32_e32 v5, s16
; GFX7-NEXT:    v_mov_b32_e32 v6, s15
; GFX7-NEXT:    v_mov_b32_e32 v7, s14
; GFX7-NEXT:    s_waitcnt vmcnt(0)
; GFX7-NEXT:    v_mad_u32_u24 v0, s4, v1, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s12, v2, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s11, v3, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s10, v4, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s9, v5, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s8, v6, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s7, v7, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s13
; GFX7-NEXT:    v_mad_u32_u24 v0, s6, v1, v0
; GFX7-NEXT:    buffer_store_short v0, off, s[0:3], 0
; GFX7-NEXT:    s_endpgm
;
; GFX8-LABEL: udot8_acc16:
; GFX8:       ; %bb.0: ; %entry
; GFX8-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX8-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    v_mov_b32_e32 v0, s0
; GFX8-NEXT:    v_mov_b32_e32 v1, s1
; GFX8-NEXT:    flat_load_ushort v2, v[0:1]
; GFX8-NEXT:    s_load_dword s0, s[4:5], 0x0
; GFX8-NEXT:    s_load_dword s1, s[6:7], 0x0
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    s_lshr_b32 s2, s0, 28
; GFX8-NEXT:    s_bfe_u32 s10, s1, 0x40018
; GFX8-NEXT:    s_bfe_u32 s11, s1, 0x40014
; GFX8-NEXT:    s_bfe_u32 s12, s1, 0x40010
; GFX8-NEXT:    s_bfe_u32 s13, s1, 0x4000c
; GFX8-NEXT:    s_bfe_u32 s14, s1, 0x40008
; GFX8-NEXT:    s_bfe_u32 s15, s1, 0x40004
; GFX8-NEXT:    s_lshr_b32 s9, s1, 28
; GFX8-NEXT:    s_and_b32 s1, s1, 15
; GFX8-NEXT:    s_bfe_u32 s3, s0, 0x40018
; GFX8-NEXT:    s_bfe_u32 s4, s0, 0x40014
; GFX8-NEXT:    s_bfe_u32 s5, s0, 0x40010
; GFX8-NEXT:    s_bfe_u32 s6, s0, 0x4000c
; GFX8-NEXT:    s_bfe_u32 s7, s0, 0x40008
; GFX8-NEXT:    s_bfe_u32 s8, s0, 0x40004
; GFX8-NEXT:    s_and_b32 s0, s0, 15
; GFX8-NEXT:    v_mov_b32_e32 v3, s1
; GFX8-NEXT:    v_mov_b32_e32 v4, s15
; GFX8-NEXT:    v_mov_b32_e32 v5, s14
; GFX8-NEXT:    v_mov_b32_e32 v6, s13
; GFX8-NEXT:    v_mov_b32_e32 v7, s12
; GFX8-NEXT:    v_mov_b32_e32 v8, s11
; GFX8-NEXT:    v_mov_b32_e32 v9, s10
; GFX8-NEXT:    s_waitcnt vmcnt(0)
; GFX8-NEXT:    v_mad_u32_u24 v2, s0, v3, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s8, v4, v2
; GFX8-NEXT:    v_and_b32_e32 v2, 0xffff, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s7, v5, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s6, v6, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s5, v7, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s4, v8, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s3, v9, v2
; GFX8-NEXT:    v_mov_b32_e32 v3, s9
; GFX8-NEXT:    v_mad_u32_u24 v2, s2, v3, v2
; GFX8-NEXT:    flat_store_short v[0:1], v2
; GFX8-NEXT:    s_endpgm
;
; GFX9-LABEL: udot8_acc16:
; GFX9:       ; %bb.0: ; %entry
; GFX9-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-NEXT:    global_load_ushort v2, v[0:1], off
; GFX9-NEXT:    s_load_dword s0, s[4:5], 0x0
; GFX9-NEXT:    s_load_dword s1, s[6:7], 0x0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    s_lshr_b32 s2, s0, 28
; GFX9-NEXT:    s_bfe_u32 s10, s1, 0x40018
; GFX9-NEXT:    s_bfe_u32 s11, s1, 0x40014
; GFX9-NEXT:    s_bfe_u32 s12, s1, 0x40010
; GFX9-NEXT:    s_bfe_u32 s13, s1, 0x4000c
; GFX9-NEXT:    s_bfe_u32 s14, s1, 0x40008
; GFX9-NEXT:    s_bfe_u32 s15, s1, 0x40004
; GFX9-NEXT:    s_lshr_b32 s9, s1, 28
; GFX9-NEXT:    s_and_b32 s1, s1, 15
; GFX9-NEXT:    s_bfe_u32 s3, s0, 0x40018
; GFX9-NEXT:    s_bfe_u32 s4, s0, 0x40014
; GFX9-NEXT:    s_bfe_u32 s5, s0, 0x40010
; GFX9-NEXT:    s_bfe_u32 s6, s0, 0x4000c
; GFX9-NEXT:    s_bfe_u32 s7, s0, 0x40008
; GFX9-NEXT:    s_bfe_u32 s8, s0, 0x40004
; GFX9-NEXT:    s_and_b32 s0, s0, 15
; GFX9-NEXT:    v_mov_b32_e32 v3, s1
; GFX9-NEXT:    v_mov_b32_e32 v4, s15
; GFX9-NEXT:    v_mov_b32_e32 v5, s14
; GFX9-NEXT:    v_mov_b32_e32 v6, s13
; GFX9-NEXT:    v_mov_b32_e32 v7, s12
; GFX9-NEXT:    v_mov_b32_e32 v8, s11
; GFX9-NEXT:    v_mov_b32_e32 v9, s10
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    v_mad_u32_u24 v2, s0, v3, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s8, v4, v2
; GFX9-NEXT:    v_and_b32_e32 v2, 0xffff, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s7, v5, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s6, v6, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s5, v7, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s4, v8, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s3, v9, v2
; GFX9-NEXT:    v_mov_b32_e32 v3, s9
; GFX9-NEXT:    v_mad_u32_u24 v2, s2, v3, v2
; GFX9-NEXT:    global_store_short v[0:1], v2, off
; GFX9-NEXT:    s_endpgm
;
; GFX9-DL-LABEL: udot8_acc16:
; GFX9-DL:       ; %bb.0: ; %entry
; GFX9-DL-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-DL-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-DL-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-DL-NEXT:    global_load_ushort v2, v[0:1], off
; GFX9-DL-NEXT:    s_load_dword s0, s[4:5], 0x0
; GFX9-DL-NEXT:    s_load_dword s1, s[6:7], 0x0
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    s_lshr_b32 s2, s0, 28
; GFX9-DL-NEXT:    s_bfe_u32 s10, s1, 0x40018
; GFX9-DL-NEXT:    s_bfe_u32 s11, s1, 0x40014
; GFX9-DL-NEXT:    s_bfe_u32 s12, s1, 0x40010
; GFX9-DL-NEXT:    s_bfe_u32 s13, s1, 0x4000c
; GFX9-DL-NEXT:    s_bfe_u32 s14, s1, 0x40008
; GFX9-DL-NEXT:    s_bfe_u32 s15, s1, 0x40004
; GFX9-DL-NEXT:    s_lshr_b32 s9, s1, 28
; GFX9-DL-NEXT:    s_and_b32 s1, s1, 15
; GFX9-DL-NEXT:    s_bfe_u32 s3, s0, 0x40018
; GFX9-DL-NEXT:    s_bfe_u32 s4, s0, 0x40014
; GFX9-DL-NEXT:    s_bfe_u32 s5, s0, 0x40010
; GFX9-DL-NEXT:    s_bfe_u32 s6, s0, 0x4000c
; GFX9-DL-NEXT:    s_bfe_u32 s7, s0, 0x40008
; GFX9-DL-NEXT:    s_bfe_u32 s8, s0, 0x40004
; GFX9-DL-NEXT:    s_and_b32 s0, s0, 15
; GFX9-DL-NEXT:    v_mov_b32_e32 v3, s1
; GFX9-DL-NEXT:    v_mov_b32_e32 v4, s15
; GFX9-DL-NEXT:    v_mov_b32_e32 v5, s14
; GFX9-DL-NEXT:    v_mov_b32_e32 v6, s13
; GFX9-DL-NEXT:    v_mov_b32_e32 v7, s12
; GFX9-DL-NEXT:    v_mov_b32_e32 v8, s11
; GFX9-DL-NEXT:    v_mov_b32_e32 v9, s10
; GFX9-DL-NEXT:    s_waitcnt vmcnt(0)
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s0, v3, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s8, v4, v2
; GFX9-DL-NEXT:    v_and_b32_e32 v2, 0xffff, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s7, v5, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s6, v6, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s5, v7, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s4, v8, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s3, v9, v2
; GFX9-DL-NEXT:    v_mov_b32_e32 v3, s9
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s2, v3, v2
; GFX9-DL-NEXT:    global_store_short v[0:1], v2, off
; GFX9-DL-NEXT:    s_endpgm
;
; GFX10-DL-LABEL: udot8_acc16:
; GFX10-DL:       ; %bb.0: ; %entry
; GFX10-DL-NEXT:    s_load_dwordx2 s[2:3], s[0:1], 0x34
; GFX10-DL-NEXT:    ; implicit-def: $vcc_hi
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    v_mov_b32_e32 v0, s2
; GFX10-DL-NEXT:    v_mov_b32_e32 v1, s3
; GFX10-DL-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
; GFX10-DL-NEXT:    global_load_ushort v2, v[0:1], off
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    s_load_dword s0, s[0:1], 0x0
; GFX10-DL-NEXT:    s_load_dword s1, s[2:3], 0x0
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    s_and_b32 s2, s0, 15
; GFX10-DL-NEXT:    s_and_b32 s3, s1, 15
; GFX10-DL-NEXT:    s_waitcnt vmcnt(0)
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40004
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40004
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40008
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40008
; GFX10-DL-NEXT:    v_and_b32_e32 v2, 0xffff, v2
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x4000c
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x4000c
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40010
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40010
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40014
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40014
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40018
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40018
; GFX10-DL-NEXT:    s_lshr_b32 s0, s0, 28
; GFX10-DL-NEXT:    s_lshr_b32 s1, s1, 28
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s0, s1, v2
; GFX10-DL-NEXT:    global_store_short v[0:1], v2, off
; GFX10-DL-NEXT:    s_endpgm
                                       <8 x i4> addrspace(1)* %src2,
                                       i16 addrspace(1)* nocapture %dst) {
entry:
  %vec1 = load <8 x i4>, <8 x i4> addrspace(1)* %src1
  %vec2 = load <8 x i4>, <8 x i4> addrspace(1)* %src2

  %v1e0 = extractelement <8 x i4> %vec1, i64 0
  %cv1e0 = zext i4 %v1e0 to i16
  %v2e0 = extractelement <8 x i4> %vec2, i64 0
  %cv2e0 = zext i4 %v2e0 to i16
  %mul0 = mul nuw nsw i16 %cv1e0, %cv2e0

  %v1e1 = extractelement <8 x i4> %vec1, i64 1
  %cv1e1 = zext i4 %v1e1 to i16
  %v2e1 = extractelement <8 x i4> %vec2, i64 1
  %cv2e1 = zext i4 %v2e1 to i16
  %mul1 = mul nuw nsw i16 %cv1e1, %cv2e1

  %v1e2 = extractelement <8 x i4> %vec1, i64 2
  %cv1e2 = zext i4 %v1e2 to i16
  %v2e2 = extractelement <8 x i4> %vec2, i64 2
  %cv2e2 = zext i4 %v2e2 to i16
  %mul2 = mul nuw nsw i16 %cv1e2, %cv2e2

  %v1e3 = extractelement <8 x i4> %vec1, i64 3
  %cv1e3 = zext i4 %v1e3 to i16
  %v2e3 = extractelement <8 x i4> %vec2, i64 3
  %cv2e3 = zext i4 %v2e3 to i16
  %mul3 = mul nuw nsw i16 %cv1e3, %cv2e3

  %v1e4 = extractelement <8 x i4> %vec1, i64 4
  %cv1e4 = zext i4 %v1e4 to i16
  %v2e4 = extractelement <8 x i4> %vec2, i64 4
  %cv2e4 = zext i4 %v2e4 to i16
  %mul4 = mul nuw nsw i16 %cv1e4, %cv2e4

  %v1e5 = extractelement <8 x i4> %vec1, i64 5
  %cv1e5 = zext i4 %v1e5 to i16
  %v2e5 = extractelement <8 x i4> %vec2, i64 5
  %cv2e5 = zext i4 %v2e5 to i16
  %mul5 = mul nuw nsw i16 %cv1e5, %cv2e5

  %v1e6 = extractelement <8 x i4> %vec1, i64 6
  %cv1e6 = zext i4 %v1e6 to i16
  %v2e6 = extractelement <8 x i4> %vec2, i64 6
  %cv2e6 = zext i4 %v2e6 to i16
  %mul6 = mul nuw nsw i16 %cv1e6, %cv2e6

  %v1e7 = extractelement <8 x i4> %vec1, i64 7
  %cv1e7 = zext i4 %v1e7 to i16
  %v2e7 = extractelement <8 x i4> %vec2, i64 7
  %cv2e7 = zext i4 %v2e7 to i16
  %mul7 = mul nuw nsw i16 %cv1e7, %cv2e7

  %acc = load i16, i16 addrspace(1)* %dst, align 4
  %add1 = add i16 %mul0, %acc
  %add2 = add i16 %add1, %mul1
  %add3 = add i16 %add2, %mul2
  %add4 = add i16 %add3, %mul3
  %add5 = add i16 %add4, %mul4
  %add6 = add i16 %add5, %mul5
  %add7 = add i16 %add6, %mul6
  %add8 = add i16 %add7, %mul7

  store i16 %add8, i16 addrspace(1)* %dst, align 4
  ret void
}

; TODO: Remove the unnecessary instruction(that is zero-extending the
; 2nd MAD) to have the pattern-recognizer to kick in.
define amdgpu_kernel void @udot8_acc8(<8 x i4> addrspace(1)* %src1,
; GFX7-LABEL: udot8_acc8:
; GFX7:       ; %bb.0: ; %entry
; GFX7-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x9
; GFX7-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0xd
; GFX7-NEXT:    s_mov_b32 s3, 0xf000
; GFX7-NEXT:    s_mov_b32 s2, -1
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    buffer_load_ubyte v0, off, s[0:3], 0
; GFX7-NEXT:    s_load_dword s4, s[4:5], 0x0
; GFX7-NEXT:    s_load_dword s5, s[6:7], 0x0
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    s_lshr_b32 s6, s4, 28
; GFX7-NEXT:    s_bfe_u32 s14, s5, 0x40018
; GFX7-NEXT:    s_bfe_u32 s15, s5, 0x40014
; GFX7-NEXT:    s_bfe_u32 s16, s5, 0x40010
; GFX7-NEXT:    s_bfe_u32 s17, s5, 0x4000c
; GFX7-NEXT:    s_bfe_u32 s18, s5, 0x40008
; GFX7-NEXT:    s_bfe_u32 s19, s5, 0x40004
; GFX7-NEXT:    s_lshr_b32 s13, s5, 28
; GFX7-NEXT:    s_and_b32 s5, s5, 15
; GFX7-NEXT:    s_bfe_u32 s7, s4, 0x40018
; GFX7-NEXT:    s_bfe_u32 s8, s4, 0x40014
; GFX7-NEXT:    s_bfe_u32 s9, s4, 0x40010
; GFX7-NEXT:    s_bfe_u32 s10, s4, 0x4000c
; GFX7-NEXT:    s_bfe_u32 s11, s4, 0x40008
; GFX7-NEXT:    s_bfe_u32 s12, s4, 0x40004
; GFX7-NEXT:    s_and_b32 s4, s4, 15
; GFX7-NEXT:    v_mov_b32_e32 v1, s5
; GFX7-NEXT:    v_mov_b32_e32 v2, s19
; GFX7-NEXT:    v_mov_b32_e32 v3, s18
; GFX7-NEXT:    v_mov_b32_e32 v4, s17
; GFX7-NEXT:    v_mov_b32_e32 v5, s16
; GFX7-NEXT:    v_mov_b32_e32 v6, s15
; GFX7-NEXT:    v_mov_b32_e32 v7, s14
; GFX7-NEXT:    s_waitcnt vmcnt(0)
; GFX7-NEXT:    v_mad_u32_u24 v0, s4, v1, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s12, v2, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s11, v3, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s10, v4, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s9, v5, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s8, v6, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s7, v7, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s13
; GFX7-NEXT:    v_mad_u32_u24 v0, s6, v1, v0
; GFX7-NEXT:    buffer_store_byte v0, off, s[0:3], 0
; GFX7-NEXT:    s_endpgm
;
; GFX8-LABEL: udot8_acc8:
; GFX8:       ; %bb.0: ; %entry
; GFX8-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX8-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    v_mov_b32_e32 v0, s0
; GFX8-NEXT:    v_mov_b32_e32 v1, s1
; GFX8-NEXT:    flat_load_ubyte v2, v[0:1]
; GFX8-NEXT:    s_load_dword s0, s[4:5], 0x0
; GFX8-NEXT:    s_load_dword s1, s[6:7], 0x0
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    s_lshr_b32 s2, s0, 28
; GFX8-NEXT:    s_bfe_u32 s10, s1, 0x40018
; GFX8-NEXT:    s_bfe_u32 s11, s1, 0x40014
; GFX8-NEXT:    s_bfe_u32 s12, s1, 0x40010
; GFX8-NEXT:    s_bfe_u32 s13, s1, 0x4000c
; GFX8-NEXT:    s_bfe_u32 s14, s1, 0x40008
; GFX8-NEXT:    s_bfe_u32 s15, s1, 0x40004
; GFX8-NEXT:    s_lshr_b32 s9, s1, 28
; GFX8-NEXT:    s_and_b32 s1, s1, 15
; GFX8-NEXT:    s_bfe_u32 s3, s0, 0x40018
; GFX8-NEXT:    s_bfe_u32 s4, s0, 0x40014
; GFX8-NEXT:    s_bfe_u32 s5, s0, 0x40010
; GFX8-NEXT:    s_bfe_u32 s6, s0, 0x4000c
; GFX8-NEXT:    s_bfe_u32 s7, s0, 0x40008
; GFX8-NEXT:    s_bfe_u32 s8, s0, 0x40004
; GFX8-NEXT:    s_and_b32 s0, s0, 15
; GFX8-NEXT:    v_mov_b32_e32 v3, s1
; GFX8-NEXT:    v_mov_b32_e32 v4, s15
; GFX8-NEXT:    v_mov_b32_e32 v5, s14
; GFX8-NEXT:    v_mov_b32_e32 v6, s13
; GFX8-NEXT:    v_mov_b32_e32 v7, s12
; GFX8-NEXT:    v_mov_b32_e32 v8, s11
; GFX8-NEXT:    v_mov_b32_e32 v9, s10
; GFX8-NEXT:    s_waitcnt vmcnt(0)
; GFX8-NEXT:    v_mad_u32_u24 v2, s0, v3, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s8, v4, v2
; GFX8-NEXT:    v_and_b32_e32 v2, 0xff, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s7, v5, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s6, v6, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s5, v7, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s4, v8, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s3, v9, v2
; GFX8-NEXT:    v_mov_b32_e32 v3, s9
; GFX8-NEXT:    v_mad_u32_u24 v2, s2, v3, v2
; GFX8-NEXT:    flat_store_byte v[0:1], v2
; GFX8-NEXT:    s_endpgm
;
; GFX9-LABEL: udot8_acc8:
; GFX9:       ; %bb.0: ; %entry
; GFX9-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-NEXT:    global_load_ubyte v2, v[0:1], off
; GFX9-NEXT:    s_load_dword s0, s[4:5], 0x0
; GFX9-NEXT:    s_load_dword s1, s[6:7], 0x0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    s_lshr_b32 s2, s0, 28
; GFX9-NEXT:    s_bfe_u32 s10, s1, 0x40018
; GFX9-NEXT:    s_bfe_u32 s11, s1, 0x40014
; GFX9-NEXT:    s_bfe_u32 s12, s1, 0x40010
; GFX9-NEXT:    s_bfe_u32 s13, s1, 0x4000c
; GFX9-NEXT:    s_bfe_u32 s14, s1, 0x40008
; GFX9-NEXT:    s_bfe_u32 s15, s1, 0x40004
; GFX9-NEXT:    s_lshr_b32 s9, s1, 28
; GFX9-NEXT:    s_and_b32 s1, s1, 15
; GFX9-NEXT:    s_bfe_u32 s3, s0, 0x40018
; GFX9-NEXT:    s_bfe_u32 s4, s0, 0x40014
; GFX9-NEXT:    s_bfe_u32 s5, s0, 0x40010
; GFX9-NEXT:    s_bfe_u32 s6, s0, 0x4000c
; GFX9-NEXT:    s_bfe_u32 s7, s0, 0x40008
; GFX9-NEXT:    s_bfe_u32 s8, s0, 0x40004
; GFX9-NEXT:    s_and_b32 s0, s0, 15
; GFX9-NEXT:    v_mov_b32_e32 v3, s1
; GFX9-NEXT:    v_mov_b32_e32 v4, s15
; GFX9-NEXT:    v_mov_b32_e32 v5, s14
; GFX9-NEXT:    v_mov_b32_e32 v6, s13
; GFX9-NEXT:    v_mov_b32_e32 v7, s12
; GFX9-NEXT:    v_mov_b32_e32 v8, s11
; GFX9-NEXT:    v_mov_b32_e32 v9, s10
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    v_mad_u32_u24 v2, s0, v3, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s8, v4, v2
; GFX9-NEXT:    v_and_b32_e32 v2, 0xff, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s7, v5, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s6, v6, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s5, v7, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s4, v8, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s3, v9, v2
; GFX9-NEXT:    v_mov_b32_e32 v3, s9
; GFX9-NEXT:    v_mad_u32_u24 v2, s2, v3, v2
; GFX9-NEXT:    global_store_byte v[0:1], v2, off
; GFX9-NEXT:    s_endpgm
;
; GFX9-DL-LABEL: udot8_acc8:
; GFX9-DL:       ; %bb.0: ; %entry
; GFX9-DL-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-DL-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-DL-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-DL-NEXT:    global_load_ubyte v2, v[0:1], off
; GFX9-DL-NEXT:    s_load_dword s0, s[4:5], 0x0
; GFX9-DL-NEXT:    s_load_dword s1, s[6:7], 0x0
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    s_lshr_b32 s2, s0, 28
; GFX9-DL-NEXT:    s_bfe_u32 s10, s1, 0x40018
; GFX9-DL-NEXT:    s_bfe_u32 s11, s1, 0x40014
; GFX9-DL-NEXT:    s_bfe_u32 s12, s1, 0x40010
; GFX9-DL-NEXT:    s_bfe_u32 s13, s1, 0x4000c
; GFX9-DL-NEXT:    s_bfe_u32 s14, s1, 0x40008
; GFX9-DL-NEXT:    s_bfe_u32 s15, s1, 0x40004
; GFX9-DL-NEXT:    s_lshr_b32 s9, s1, 28
; GFX9-DL-NEXT:    s_and_b32 s1, s1, 15
; GFX9-DL-NEXT:    s_bfe_u32 s3, s0, 0x40018
; GFX9-DL-NEXT:    s_bfe_u32 s4, s0, 0x40014
; GFX9-DL-NEXT:    s_bfe_u32 s5, s0, 0x40010
; GFX9-DL-NEXT:    s_bfe_u32 s6, s0, 0x4000c
; GFX9-DL-NEXT:    s_bfe_u32 s7, s0, 0x40008
; GFX9-DL-NEXT:    s_bfe_u32 s8, s0, 0x40004
; GFX9-DL-NEXT:    s_and_b32 s0, s0, 15
; GFX9-DL-NEXT:    v_mov_b32_e32 v3, s1
; GFX9-DL-NEXT:    v_mov_b32_e32 v4, s15
; GFX9-DL-NEXT:    v_mov_b32_e32 v5, s14
; GFX9-DL-NEXT:    v_mov_b32_e32 v6, s13
; GFX9-DL-NEXT:    v_mov_b32_e32 v7, s12
; GFX9-DL-NEXT:    v_mov_b32_e32 v8, s11
; GFX9-DL-NEXT:    v_mov_b32_e32 v9, s10
; GFX9-DL-NEXT:    s_waitcnt vmcnt(0)
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s0, v3, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s8, v4, v2
; GFX9-DL-NEXT:    v_and_b32_e32 v2, 0xff, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s7, v5, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s6, v6, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s5, v7, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s4, v8, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s3, v9, v2
; GFX9-DL-NEXT:    v_mov_b32_e32 v3, s9
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s2, v3, v2
; GFX9-DL-NEXT:    global_store_byte v[0:1], v2, off
; GFX9-DL-NEXT:    s_endpgm
;
; GFX10-DL-LABEL: udot8_acc8:
; GFX10-DL:       ; %bb.0: ; %entry
; GFX10-DL-NEXT:    s_load_dwordx2 s[2:3], s[0:1], 0x34
; GFX10-DL-NEXT:    ; implicit-def: $vcc_hi
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    v_mov_b32_e32 v0, s2
; GFX10-DL-NEXT:    v_mov_b32_e32 v1, s3
; GFX10-DL-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
; GFX10-DL-NEXT:    global_load_ubyte v2, v[0:1], off
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    s_load_dword s0, s[0:1], 0x0
; GFX10-DL-NEXT:    s_load_dword s1, s[2:3], 0x0
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    s_and_b32 s2, s0, 15
; GFX10-DL-NEXT:    s_and_b32 s3, s1, 15
; GFX10-DL-NEXT:    s_waitcnt vmcnt(0)
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40004
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40004
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40008
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40008
; GFX10-DL-NEXT:    v_and_b32_e32 v2, 0xff, v2
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x4000c
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x4000c
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40010
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40010
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40014
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40014
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40018
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40018
; GFX10-DL-NEXT:    s_lshr_b32 s0, s0, 28
; GFX10-DL-NEXT:    s_lshr_b32 s1, s1, 28
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s0, s1, v2
; GFX10-DL-NEXT:    global_store_byte v[0:1], v2, off
; GFX10-DL-NEXT:    s_endpgm
                                      <8 x i4> addrspace(1)* %src2,
                                      i8 addrspace(1)* nocapture %dst) {
entry:
  %vec1 = load <8 x i4>, <8 x i4> addrspace(1)* %src1
  %vec2 = load <8 x i4>, <8 x i4> addrspace(1)* %src2

  %v1e0 = extractelement <8 x i4> %vec1, i64 0
  %cv1e0 = zext i4 %v1e0 to i8
  %v2e0 = extractelement <8 x i4> %vec2, i64 0
  %cv2e0 = zext i4 %v2e0 to i8
  %mul0 = mul nuw nsw i8 %cv1e0, %cv2e0

  %v1e1 = extractelement <8 x i4> %vec1, i64 1
  %cv1e1 = zext i4 %v1e1 to i8
  %v2e1 = extractelement <8 x i4> %vec2, i64 1
  %cv2e1 = zext i4 %v2e1 to i8
  %mul1 = mul nuw nsw i8 %cv1e1, %cv2e1

  %v1e2 = extractelement <8 x i4> %vec1, i64 2
  %cv1e2 = zext i4 %v1e2 to i8
  %v2e2 = extractelement <8 x i4> %vec2, i64 2
  %cv2e2 = zext i4 %v2e2 to i8
  %mul2 = mul nuw nsw i8 %cv1e2, %cv2e2

  %v1e3 = extractelement <8 x i4> %vec1, i64 3
  %cv1e3 = zext i4 %v1e3 to i8
  %v2e3 = extractelement <8 x i4> %vec2, i64 3
  %cv2e3 = zext i4 %v2e3 to i8
  %mul3 = mul nuw nsw i8 %cv1e3, %cv2e3

  %v1e4 = extractelement <8 x i4> %vec1, i64 4
  %cv1e4 = zext i4 %v1e4 to i8
  %v2e4 = extractelement <8 x i4> %vec2, i64 4
  %cv2e4 = zext i4 %v2e4 to i8
  %mul4 = mul nuw nsw i8 %cv1e4, %cv2e4

  %v1e5 = extractelement <8 x i4> %vec1, i64 5
  %cv1e5 = zext i4 %v1e5 to i8
  %v2e5 = extractelement <8 x i4> %vec2, i64 5
  %cv2e5 = zext i4 %v2e5 to i8
  %mul5 = mul nuw nsw i8 %cv1e5, %cv2e5

  %v1e6 = extractelement <8 x i4> %vec1, i64 6
  %cv1e6 = zext i4 %v1e6 to i8
  %v2e6 = extractelement <8 x i4> %vec2, i64 6
  %cv2e6 = zext i4 %v2e6 to i8
  %mul6 = mul nuw nsw i8 %cv1e6, %cv2e6

  %v1e7 = extractelement <8 x i4> %vec1, i64 7
  %cv1e7 = zext i4 %v1e7 to i8
  %v2e7 = extractelement <8 x i4> %vec2, i64 7
  %cv2e7 = zext i4 %v2e7 to i8
  %mul7 = mul nuw nsw i8 %cv1e7, %cv2e7

  %acc = load i8, i8 addrspace(1)* %dst, align 4
  %add1 = add i8 %mul0, %acc
  %add2 = add i8 %add1, %mul1
  %add3 = add i8 %add2, %mul2
  %add4 = add i8 %add3, %mul3
  %add5 = add i8 %add4, %mul4
  %add6 = add i8 %add5, %mul5
  %add7 = add i8 %add6, %mul6
  %add8 = add i8 %add7, %mul7

  store i8 %add8, i8 addrspace(1)* %dst, align 4
  ret void
}

; TODO: Remove the two unnecessary instructions(and+add after 2nd MAD)
; to have the pattern-recognizer to kick in.
define amdgpu_kernel void @udot8_acc4(<8 x i4> addrspace(1)* %src1,
; GFX7-LABEL: udot8_acc4:
; GFX7:       ; %bb.0: ; %entry
; GFX7-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x9
; GFX7-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0xd
; GFX7-NEXT:    s_mov_b32 s3, 0xf000
; GFX7-NEXT:    s_mov_b32 s2, -1
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    buffer_load_ubyte v0, off, s[0:3], 0
; GFX7-NEXT:    s_load_dword s4, s[4:5], 0x0
; GFX7-NEXT:    s_load_dword s5, s[6:7], 0x0
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    s_lshr_b32 s6, s4, 28
; GFX7-NEXT:    s_bfe_u32 s14, s5, 0x40018
; GFX7-NEXT:    s_bfe_u32 s15, s5, 0x40014
; GFX7-NEXT:    s_bfe_u32 s16, s5, 0x40010
; GFX7-NEXT:    s_bfe_u32 s17, s5, 0x4000c
; GFX7-NEXT:    s_bfe_u32 s18, s5, 0x40008
; GFX7-NEXT:    s_bfe_u32 s19, s5, 0x40004
; GFX7-NEXT:    s_lshr_b32 s13, s5, 28
; GFX7-NEXT:    s_and_b32 s5, s5, 15
; GFX7-NEXT:    s_bfe_u32 s7, s4, 0x40018
; GFX7-NEXT:    s_bfe_u32 s8, s4, 0x40014
; GFX7-NEXT:    s_bfe_u32 s9, s4, 0x40010
; GFX7-NEXT:    s_bfe_u32 s10, s4, 0x4000c
; GFX7-NEXT:    s_bfe_u32 s11, s4, 0x40008
; GFX7-NEXT:    s_bfe_u32 s12, s4, 0x40004
; GFX7-NEXT:    s_and_b32 s4, s4, 15
; GFX7-NEXT:    v_mov_b32_e32 v1, s5
; GFX7-NEXT:    v_mov_b32_e32 v2, s19
; GFX7-NEXT:    v_mov_b32_e32 v3, s18
; GFX7-NEXT:    v_mov_b32_e32 v4, s17
; GFX7-NEXT:    v_mov_b32_e32 v5, s16
; GFX7-NEXT:    v_mov_b32_e32 v6, s15
; GFX7-NEXT:    v_mov_b32_e32 v7, s14
; GFX7-NEXT:    s_waitcnt vmcnt(0)
; GFX7-NEXT:    v_mad_u32_u24 v0, s4, v1, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s12, v2, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s11, v3, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s10, v4, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s9, v5, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s8, v6, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s7, v7, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s13
; GFX7-NEXT:    v_mad_u32_u24 v0, s6, v1, v0
; GFX7-NEXT:    v_and_b32_e32 v0, 15, v0
; GFX7-NEXT:    buffer_store_byte v0, off, s[0:3], 0
; GFX7-NEXT:    s_endpgm
;
; GFX8-LABEL: udot8_acc4:
; GFX8:       ; %bb.0: ; %entry
; GFX8-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX8-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    v_mov_b32_e32 v0, s0
; GFX8-NEXT:    v_mov_b32_e32 v1, s1
; GFX8-NEXT:    flat_load_ubyte v2, v[0:1]
; GFX8-NEXT:    s_load_dword s0, s[4:5], 0x0
; GFX8-NEXT:    s_load_dword s1, s[6:7], 0x0
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    s_and_b32 s8, s0, 15
; GFX8-NEXT:    s_and_b32 s15, s1, 15
; GFX8-NEXT:    s_bfe_u32 s14, s1, 0x40004
; GFX8-NEXT:    v_mov_b32_e32 v4, s15
; GFX8-NEXT:    s_bfe_u32 s10, s1, 0x40018
; GFX8-NEXT:    s_bfe_u32 s11, s1, 0x40014
; GFX8-NEXT:    s_bfe_u32 s12, s1, 0x40010
; GFX8-NEXT:    s_bfe_u32 s13, s1, 0x40008
; GFX8-NEXT:    s_lshr_b32 s9, s1, 28
; GFX8-NEXT:    s_bfe_u32 s1, s1, 0x4000c
; GFX8-NEXT:    s_bfe_u32 s7, s0, 0x40004
; GFX8-NEXT:    v_mov_b32_e32 v5, s14
; GFX8-NEXT:    s_lshr_b32 s2, s0, 28
; GFX8-NEXT:    s_bfe_u32 s3, s0, 0x40018
; GFX8-NEXT:    s_bfe_u32 s4, s0, 0x40014
; GFX8-NEXT:    s_bfe_u32 s5, s0, 0x40010
; GFX8-NEXT:    s_bfe_u32 s6, s0, 0x40008
; GFX8-NEXT:    s_bfe_u32 s0, s0, 0x4000c
; GFX8-NEXT:    v_mov_b32_e32 v3, s1
; GFX8-NEXT:    v_mov_b32_e32 v6, s13
; GFX8-NEXT:    v_mul_u32_u24_e32 v3, s0, v3
; GFX8-NEXT:    v_and_b32_e32 v3, 15, v3
; GFX8-NEXT:    v_mov_b32_e32 v7, s12
; GFX8-NEXT:    v_mov_b32_e32 v8, s11
; GFX8-NEXT:    v_mov_b32_e32 v9, s10
; GFX8-NEXT:    s_waitcnt vmcnt(0)
; GFX8-NEXT:    v_mad_u32_u24 v2, s8, v4, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s7, v5, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s6, v6, v2
; GFX8-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX8-NEXT:    v_add_u32_e32 v2, vcc, v3, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s5, v7, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s4, v8, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s3, v9, v2
; GFX8-NEXT:    v_mov_b32_e32 v3, s9
; GFX8-NEXT:    v_mad_u32_u24 v2, s2, v3, v2
; GFX8-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX8-NEXT:    flat_store_byte v[0:1], v2
; GFX8-NEXT:    s_endpgm
;
; GFX9-LABEL: udot8_acc4:
; GFX9:       ; %bb.0: ; %entry
; GFX9-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-NEXT:    global_load_ubyte v2, v[0:1], off
; GFX9-NEXT:    s_load_dword s0, s[4:5], 0x0
; GFX9-NEXT:    s_load_dword s1, s[6:7], 0x0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    s_and_b32 s8, s0, 15
; GFX9-NEXT:    s_and_b32 s15, s1, 15
; GFX9-NEXT:    s_bfe_u32 s14, s1, 0x40004
; GFX9-NEXT:    v_mov_b32_e32 v4, s15
; GFX9-NEXT:    s_bfe_u32 s10, s1, 0x40018
; GFX9-NEXT:    s_bfe_u32 s11, s1, 0x40014
; GFX9-NEXT:    s_bfe_u32 s12, s1, 0x40010
; GFX9-NEXT:    s_bfe_u32 s13, s1, 0x40008
; GFX9-NEXT:    s_lshr_b32 s9, s1, 28
; GFX9-NEXT:    s_bfe_u32 s1, s1, 0x4000c
; GFX9-NEXT:    s_bfe_u32 s7, s0, 0x40004
; GFX9-NEXT:    v_mov_b32_e32 v5, s14
; GFX9-NEXT:    s_lshr_b32 s2, s0, 28
; GFX9-NEXT:    s_bfe_u32 s3, s0, 0x40018
; GFX9-NEXT:    s_bfe_u32 s4, s0, 0x40014
; GFX9-NEXT:    s_bfe_u32 s5, s0, 0x40010
; GFX9-NEXT:    s_bfe_u32 s6, s0, 0x40008
; GFX9-NEXT:    s_bfe_u32 s0, s0, 0x4000c
; GFX9-NEXT:    v_mov_b32_e32 v3, s1
; GFX9-NEXT:    v_mov_b32_e32 v6, s13
; GFX9-NEXT:    v_mul_u32_u24_e32 v3, s0, v3
; GFX9-NEXT:    v_and_b32_e32 v3, 15, v3
; GFX9-NEXT:    v_mov_b32_e32 v7, s12
; GFX9-NEXT:    v_mov_b32_e32 v8, s11
; GFX9-NEXT:    v_mov_b32_e32 v9, s10
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    v_mad_u32_u24 v2, s8, v4, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s7, v5, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s6, v6, v2
; GFX9-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX9-NEXT:    v_add_u32_e32 v2, v2, v3
; GFX9-NEXT:    v_mad_u32_u24 v2, s5, v7, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s4, v8, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s3, v9, v2
; GFX9-NEXT:    v_mov_b32_e32 v3, s9
; GFX9-NEXT:    v_mad_u32_u24 v2, s2, v3, v2
; GFX9-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX9-NEXT:    global_store_byte v[0:1], v2, off
; GFX9-NEXT:    s_endpgm
;
; GFX9-DL-LABEL: udot8_acc4:
; GFX9-DL:       ; %bb.0: ; %entry
; GFX9-DL-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-DL-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-DL-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-DL-NEXT:    global_load_ubyte v2, v[0:1], off
; GFX9-DL-NEXT:    s_load_dword s0, s[4:5], 0x0
; GFX9-DL-NEXT:    s_load_dword s1, s[6:7], 0x0
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    s_and_b32 s8, s0, 15
; GFX9-DL-NEXT:    s_and_b32 s15, s1, 15
; GFX9-DL-NEXT:    s_bfe_u32 s14, s1, 0x40004
; GFX9-DL-NEXT:    v_mov_b32_e32 v4, s15
; GFX9-DL-NEXT:    s_bfe_u32 s10, s1, 0x40018
; GFX9-DL-NEXT:    s_bfe_u32 s11, s1, 0x40014
; GFX9-DL-NEXT:    s_bfe_u32 s12, s1, 0x40010
; GFX9-DL-NEXT:    s_bfe_u32 s13, s1, 0x40008
; GFX9-DL-NEXT:    s_lshr_b32 s9, s1, 28
; GFX9-DL-NEXT:    s_bfe_u32 s1, s1, 0x4000c
; GFX9-DL-NEXT:    s_bfe_u32 s7, s0, 0x40004
; GFX9-DL-NEXT:    v_mov_b32_e32 v5, s14
; GFX9-DL-NEXT:    s_lshr_b32 s2, s0, 28
; GFX9-DL-NEXT:    s_bfe_u32 s3, s0, 0x40018
; GFX9-DL-NEXT:    s_bfe_u32 s4, s0, 0x40014
; GFX9-DL-NEXT:    s_bfe_u32 s5, s0, 0x40010
; GFX9-DL-NEXT:    s_bfe_u32 s6, s0, 0x40008
; GFX9-DL-NEXT:    s_bfe_u32 s0, s0, 0x4000c
; GFX9-DL-NEXT:    v_mov_b32_e32 v3, s1
; GFX9-DL-NEXT:    v_mov_b32_e32 v6, s13
; GFX9-DL-NEXT:    v_mul_u32_u24_e32 v3, s0, v3
; GFX9-DL-NEXT:    v_and_b32_e32 v3, 15, v3
; GFX9-DL-NEXT:    v_mov_b32_e32 v7, s12
; GFX9-DL-NEXT:    v_mov_b32_e32 v8, s11
; GFX9-DL-NEXT:    v_mov_b32_e32 v9, s10
; GFX9-DL-NEXT:    s_waitcnt vmcnt(0)
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s8, v4, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s7, v5, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s6, v6, v2
; GFX9-DL-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX9-DL-NEXT:    v_add_u32_e32 v2, v2, v3
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s5, v7, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s4, v8, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s3, v9, v2
; GFX9-DL-NEXT:    v_mov_b32_e32 v3, s9
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s2, v3, v2
; GFX9-DL-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX9-DL-NEXT:    global_store_byte v[0:1], v2, off
; GFX9-DL-NEXT:    s_endpgm
;
; GFX10-DL-LABEL: udot8_acc4:
; GFX10-DL:       ; %bb.0: ; %entry
; GFX10-DL-NEXT:    s_load_dwordx2 s[2:3], s[0:1], 0x34
; GFX10-DL-NEXT:    ; implicit-def: $vcc_hi
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    v_mov_b32_e32 v0, s2
; GFX10-DL-NEXT:    v_mov_b32_e32 v1, s3
; GFX10-DL-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
; GFX10-DL-NEXT:    global_load_ubyte v2, v[0:1], off
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    s_load_dword s0, s[0:1], 0x0
; GFX10-DL-NEXT:    s_load_dword s1, s[2:3], 0x0
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    s_and_b32 s2, s0, 15
; GFX10-DL-NEXT:    s_and_b32 s3, s1, 15
; GFX10-DL-NEXT:    s_bfe_u32 s4, s1, 0x40008
; GFX10-DL-NEXT:    s_bfe_u32 s5, s1, 0x4000c
; GFX10-DL-NEXT:    s_waitcnt vmcnt(0)
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40004
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40004
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40008
; GFX10-DL-NEXT:    s_bfe_u32 s3, s0, 0x4000c
; GFX10-DL-NEXT:    v_mul_u32_u24_e64 v3, s3, s5
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s4, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40010
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40010
; GFX10-DL-NEXT:    v_and_b32_e32 v3, 15, v3
; GFX10-DL-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX10-DL-NEXT:    v_add_nc_u32_e32 v2, v2, v3
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40014
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40014
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40018
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40018
; GFX10-DL-NEXT:    s_lshr_b32 s0, s0, 28
; GFX10-DL-NEXT:    s_lshr_b32 s1, s1, 28
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s0, s1, v2
; GFX10-DL-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX10-DL-NEXT:    global_store_byte v[0:1], v2, off
; GFX10-DL-NEXT:    s_endpgm
                                      <8 x i4> addrspace(1)* %src2,
                                      i4 addrspace(1)* nocapture %dst) {
entry:
  %vec1 = load <8 x i4>, <8 x i4> addrspace(1)* %src1
  %vec2 = load <8 x i4>, <8 x i4> addrspace(1)* %src2

  %v1e0 = extractelement <8 x i4> %vec1, i64 0
  %v2e0 = extractelement <8 x i4> %vec2, i64 0
  %mul0 = mul nuw nsw i4 %v1e0, %v2e0

  %v1e1 = extractelement <8 x i4> %vec1, i64 1
  %v2e1 = extractelement <8 x i4> %vec2, i64 1
  %mul1 = mul nuw nsw i4 %v1e1, %v2e1

  %v1e2 = extractelement <8 x i4> %vec1, i64 2
  %v2e2 = extractelement <8 x i4> %vec2, i64 2
  %mul2 = mul nuw nsw i4 %v1e2, %v2e2

  %v1e3 = extractelement <8 x i4> %vec1, i64 3
  %v2e3 = extractelement <8 x i4> %vec2, i64 3
  %mul3 = mul nuw nsw i4 %v1e3, %v2e3

  %v1e4 = extractelement <8 x i4> %vec1, i64 4
  %v2e4 = extractelement <8 x i4> %vec2, i64 4
  %mul4 = mul nuw nsw i4 %v1e4, %v2e4

  %v1e5 = extractelement <8 x i4> %vec1, i64 5
  %v2e5 = extractelement <8 x i4> %vec2, i64 5
  %mul5 = mul nuw nsw i4 %v1e5, %v2e5

  %v1e6 = extractelement <8 x i4> %vec1, i64 6
  %v2e6 = extractelement <8 x i4> %vec2, i64 6
  %mul6 = mul nuw nsw i4 %v1e6, %v2e6

  %v1e7 = extractelement <8 x i4> %vec1, i64 7
  %v2e7 = extractelement <8 x i4> %vec2, i64 7
  %mul7 = mul nuw nsw i4 %v1e7, %v2e7

  %acc = load i4, i4 addrspace(1)* %dst, align 4
  %add1 = add i4 %mul0, %acc
  %add2 = add i4 %add1, %mul1
  %add3 = add i4 %add2, %mul2
  %add4 = add i4 %add3, %mul3
  %add5 = add i4 %add4, %mul4
  %add6 = add i4 %add5, %mul5
  %add7 = add i4 %add6, %mul6
  %add8 = add i4 %add7, %mul7

  store i4 %add8, i4 addrspace(1)* %dst, align 4
  ret void
}

; TODO: Currently, permutation of udot8 is turned off due to a huge increase
; in the compile time.
define amdgpu_kernel void @udot8_CommutationInsideMAD(<8 x i4> addrspace(1)* %src1,
; GFX7-LABEL: udot8_CommutationInsideMAD:
; GFX7:       ; %bb.0: ; %entry
; GFX7-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x9
; GFX7-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0xd
; GFX7-NEXT:    s_mov_b32 s3, 0xf000
; GFX7-NEXT:    s_mov_b32 s2, -1
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    buffer_load_ubyte v0, off, s[0:3], 0
; GFX7-NEXT:    s_load_dword s4, s[4:5], 0x0
; GFX7-NEXT:    s_load_dword s5, s[6:7], 0x0
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    s_lshr_b32 s6, s4, 28
; GFX7-NEXT:    s_bfe_u32 s14, s5, 0x40018
; GFX7-NEXT:    s_bfe_u32 s15, s5, 0x40014
; GFX7-NEXT:    s_bfe_u32 s16, s5, 0x40010
; GFX7-NEXT:    s_bfe_u32 s17, s5, 0x4000c
; GFX7-NEXT:    s_bfe_u32 s18, s5, 0x40008
; GFX7-NEXT:    s_bfe_u32 s19, s5, 0x40004
; GFX7-NEXT:    s_lshr_b32 s13, s5, 28
; GFX7-NEXT:    s_and_b32 s5, s5, 15
; GFX7-NEXT:    s_bfe_u32 s7, s4, 0x40018
; GFX7-NEXT:    s_bfe_u32 s8, s4, 0x40014
; GFX7-NEXT:    s_bfe_u32 s9, s4, 0x40010
; GFX7-NEXT:    s_bfe_u32 s10, s4, 0x4000c
; GFX7-NEXT:    s_bfe_u32 s11, s4, 0x40008
; GFX7-NEXT:    s_bfe_u32 s12, s4, 0x40004
; GFX7-NEXT:    s_and_b32 s4, s4, 15
; GFX7-NEXT:    v_mov_b32_e32 v1, s5
; GFX7-NEXT:    v_mov_b32_e32 v2, s19
; GFX7-NEXT:    v_mov_b32_e32 v3, s18
; GFX7-NEXT:    v_mov_b32_e32 v4, s17
; GFX7-NEXT:    v_mov_b32_e32 v5, s16
; GFX7-NEXT:    v_mov_b32_e32 v6, s15
; GFX7-NEXT:    v_mov_b32_e32 v7, s14
; GFX7-NEXT:    s_waitcnt vmcnt(0)
; GFX7-NEXT:    v_mad_u32_u24 v0, s4, v1, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s12, v2, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s11, v3, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s10, v4, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s9, v5, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s8, v6, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s7, v7, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s13
; GFX7-NEXT:    v_mad_u32_u24 v0, s6, v1, v0
; GFX7-NEXT:    v_and_b32_e32 v0, 15, v0
; GFX7-NEXT:    buffer_store_byte v0, off, s[0:3], 0
; GFX7-NEXT:    s_endpgm
;
; GFX8-LABEL: udot8_CommutationInsideMAD:
; GFX8:       ; %bb.0: ; %entry
; GFX8-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX8-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    v_mov_b32_e32 v0, s0
; GFX8-NEXT:    v_mov_b32_e32 v1, s1
; GFX8-NEXT:    flat_load_ubyte v2, v[0:1]
; GFX8-NEXT:    s_load_dword s0, s[4:5], 0x0
; GFX8-NEXT:    s_load_dword s1, s[6:7], 0x0
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    s_and_b32 s8, s0, 15
; GFX8-NEXT:    s_and_b32 s15, s1, 15
; GFX8-NEXT:    s_bfe_u32 s14, s1, 0x40004
; GFX8-NEXT:    v_mov_b32_e32 v4, s15
; GFX8-NEXT:    s_bfe_u32 s10, s1, 0x40018
; GFX8-NEXT:    s_bfe_u32 s11, s1, 0x40014
; GFX8-NEXT:    s_bfe_u32 s12, s1, 0x40010
; GFX8-NEXT:    s_bfe_u32 s13, s1, 0x40008
; GFX8-NEXT:    s_lshr_b32 s9, s1, 28
; GFX8-NEXT:    s_bfe_u32 s1, s1, 0x4000c
; GFX8-NEXT:    s_bfe_u32 s7, s0, 0x40004
; GFX8-NEXT:    v_mov_b32_e32 v5, s14
; GFX8-NEXT:    s_lshr_b32 s2, s0, 28
; GFX8-NEXT:    s_bfe_u32 s3, s0, 0x40018
; GFX8-NEXT:    s_bfe_u32 s4, s0, 0x40014
; GFX8-NEXT:    s_bfe_u32 s5, s0, 0x40010
; GFX8-NEXT:    s_bfe_u32 s6, s0, 0x40008
; GFX8-NEXT:    s_bfe_u32 s0, s0, 0x4000c
; GFX8-NEXT:    v_mov_b32_e32 v3, s1
; GFX8-NEXT:    v_mov_b32_e32 v6, s13
; GFX8-NEXT:    v_mul_u32_u24_e32 v3, s0, v3
; GFX8-NEXT:    v_and_b32_e32 v3, 15, v3
; GFX8-NEXT:    v_mov_b32_e32 v7, s12
; GFX8-NEXT:    v_mov_b32_e32 v8, s11
; GFX8-NEXT:    v_mov_b32_e32 v9, s10
; GFX8-NEXT:    s_waitcnt vmcnt(0)
; GFX8-NEXT:    v_mad_u32_u24 v2, s8, v4, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s7, v5, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s6, v6, v2
; GFX8-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX8-NEXT:    v_add_u32_e32 v2, vcc, v2, v3
; GFX8-NEXT:    v_mad_u32_u24 v2, s5, v7, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s4, v8, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s3, v9, v2
; GFX8-NEXT:    v_mov_b32_e32 v3, s9
; GFX8-NEXT:    v_mad_u32_u24 v2, s2, v3, v2
; GFX8-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX8-NEXT:    flat_store_byte v[0:1], v2
; GFX8-NEXT:    s_endpgm
;
; GFX9-LABEL: udot8_CommutationInsideMAD:
; GFX9:       ; %bb.0: ; %entry
; GFX9-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-NEXT:    global_load_ubyte v2, v[0:1], off
; GFX9-NEXT:    s_load_dword s0, s[4:5], 0x0
; GFX9-NEXT:    s_load_dword s1, s[6:7], 0x0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    s_and_b32 s8, s0, 15
; GFX9-NEXT:    s_and_b32 s15, s1, 15
; GFX9-NEXT:    s_bfe_u32 s14, s1, 0x40004
; GFX9-NEXT:    v_mov_b32_e32 v4, s15
; GFX9-NEXT:    s_bfe_u32 s10, s1, 0x40018
; GFX9-NEXT:    s_bfe_u32 s11, s1, 0x40014
; GFX9-NEXT:    s_bfe_u32 s12, s1, 0x40010
; GFX9-NEXT:    s_bfe_u32 s13, s1, 0x40008
; GFX9-NEXT:    s_lshr_b32 s9, s1, 28
; GFX9-NEXT:    s_bfe_u32 s1, s1, 0x4000c
; GFX9-NEXT:    s_bfe_u32 s7, s0, 0x40004
; GFX9-NEXT:    v_mov_b32_e32 v5, s14
; GFX9-NEXT:    s_lshr_b32 s2, s0, 28
; GFX9-NEXT:    s_bfe_u32 s3, s0, 0x40018
; GFX9-NEXT:    s_bfe_u32 s4, s0, 0x40014
; GFX9-NEXT:    s_bfe_u32 s5, s0, 0x40010
; GFX9-NEXT:    s_bfe_u32 s6, s0, 0x40008
; GFX9-NEXT:    s_bfe_u32 s0, s0, 0x4000c
; GFX9-NEXT:    v_mov_b32_e32 v3, s1
; GFX9-NEXT:    v_mov_b32_e32 v6, s13
; GFX9-NEXT:    v_mul_u32_u24_e32 v3, s0, v3
; GFX9-NEXT:    v_and_b32_e32 v3, 15, v3
; GFX9-NEXT:    v_mov_b32_e32 v7, s12
; GFX9-NEXT:    v_mov_b32_e32 v8, s11
; GFX9-NEXT:    v_mov_b32_e32 v9, s10
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    v_mad_u32_u24 v2, s8, v4, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s7, v5, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s6, v6, v2
; GFX9-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX9-NEXT:    v_add_u32_e32 v2, v3, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s5, v7, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s4, v8, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s3, v9, v2
; GFX9-NEXT:    v_mov_b32_e32 v3, s9
; GFX9-NEXT:    v_mad_u32_u24 v2, s2, v3, v2
; GFX9-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX9-NEXT:    global_store_byte v[0:1], v2, off
; GFX9-NEXT:    s_endpgm
;
; GFX9-DL-LABEL: udot8_CommutationInsideMAD:
; GFX9-DL:       ; %bb.0: ; %entry
; GFX9-DL-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-DL-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-DL-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-DL-NEXT:    global_load_ubyte v2, v[0:1], off
; GFX9-DL-NEXT:    s_load_dword s0, s[4:5], 0x0
; GFX9-DL-NEXT:    s_load_dword s1, s[6:7], 0x0
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    s_and_b32 s8, s0, 15
; GFX9-DL-NEXT:    s_and_b32 s15, s1, 15
; GFX9-DL-NEXT:    s_bfe_u32 s14, s1, 0x40004
; GFX9-DL-NEXT:    v_mov_b32_e32 v4, s15
; GFX9-DL-NEXT:    s_bfe_u32 s10, s1, 0x40018
; GFX9-DL-NEXT:    s_bfe_u32 s11, s1, 0x40014
; GFX9-DL-NEXT:    s_bfe_u32 s12, s1, 0x40010
; GFX9-DL-NEXT:    s_bfe_u32 s13, s1, 0x40008
; GFX9-DL-NEXT:    s_lshr_b32 s9, s1, 28
; GFX9-DL-NEXT:    s_bfe_u32 s1, s1, 0x4000c
; GFX9-DL-NEXT:    s_bfe_u32 s7, s0, 0x40004
; GFX9-DL-NEXT:    v_mov_b32_e32 v5, s14
; GFX9-DL-NEXT:    s_lshr_b32 s2, s0, 28
; GFX9-DL-NEXT:    s_bfe_u32 s3, s0, 0x40018
; GFX9-DL-NEXT:    s_bfe_u32 s4, s0, 0x40014
; GFX9-DL-NEXT:    s_bfe_u32 s5, s0, 0x40010
; GFX9-DL-NEXT:    s_bfe_u32 s6, s0, 0x40008
; GFX9-DL-NEXT:    s_bfe_u32 s0, s0, 0x4000c
; GFX9-DL-NEXT:    v_mov_b32_e32 v3, s1
; GFX9-DL-NEXT:    v_mov_b32_e32 v6, s13
; GFX9-DL-NEXT:    v_mul_u32_u24_e32 v3, s0, v3
; GFX9-DL-NEXT:    v_and_b32_e32 v3, 15, v3
; GFX9-DL-NEXT:    v_mov_b32_e32 v7, s12
; GFX9-DL-NEXT:    v_mov_b32_e32 v8, s11
; GFX9-DL-NEXT:    v_mov_b32_e32 v9, s10
; GFX9-DL-NEXT:    s_waitcnt vmcnt(0)
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s8, v4, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s7, v5, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s6, v6, v2
; GFX9-DL-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX9-DL-NEXT:    v_add_u32_e32 v2, v3, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s5, v7, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s4, v8, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s3, v9, v2
; GFX9-DL-NEXT:    v_mov_b32_e32 v3, s9
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s2, v3, v2
; GFX9-DL-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX9-DL-NEXT:    global_store_byte v[0:1], v2, off
; GFX9-DL-NEXT:    s_endpgm
;
; GFX10-DL-LABEL: udot8_CommutationInsideMAD:
; GFX10-DL:       ; %bb.0: ; %entry
; GFX10-DL-NEXT:    s_load_dwordx2 s[2:3], s[0:1], 0x34
; GFX10-DL-NEXT:    ; implicit-def: $vcc_hi
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    v_mov_b32_e32 v0, s2
; GFX10-DL-NEXT:    v_mov_b32_e32 v1, s3
; GFX10-DL-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
; GFX10-DL-NEXT:    global_load_ubyte v2, v[0:1], off
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    s_load_dword s0, s[0:1], 0x0
; GFX10-DL-NEXT:    s_load_dword s1, s[2:3], 0x0
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    s_and_b32 s2, s0, 15
; GFX10-DL-NEXT:    s_and_b32 s3, s1, 15
; GFX10-DL-NEXT:    s_bfe_u32 s4, s0, 0x40008
; GFX10-DL-NEXT:    s_bfe_u32 s5, s1, 0x40008
; GFX10-DL-NEXT:    s_waitcnt vmcnt(0)
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40004
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40004
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x4000c
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x4000c
; GFX10-DL-NEXT:    v_mul_u32_u24_e64 v3, s2, s3
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s4, s5, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40010
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40010
; GFX10-DL-NEXT:    v_and_b32_e32 v3, 15, v3
; GFX10-DL-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX10-DL-NEXT:    v_add_nc_u32_e32 v2, v3, v2
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40014
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40014
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40018
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40018
; GFX10-DL-NEXT:    s_lshr_b32 s0, s0, 28
; GFX10-DL-NEXT:    s_lshr_b32 s1, s1, 28
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s0, s1, v2
; GFX10-DL-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX10-DL-NEXT:    global_store_byte v[0:1], v2, off
; GFX10-DL-NEXT:    s_endpgm
                                                      <8 x i4> addrspace(1)* %src2,
                                                      i4 addrspace(1)* nocapture %dst) {
entry:
  %vec1 = load <8 x i4>, <8 x i4> addrspace(1)* %src1
  %vec2 = load <8 x i4>, <8 x i4> addrspace(1)* %src2

  %v1e0 = extractelement <8 x i4> %vec1, i64 0
  %v2e0 = extractelement <8 x i4> %vec2, i64 0
  %mul0 = mul nuw nsw i4 %v1e0, %v2e0

  %v1e1 = extractelement <8 x i4> %vec1, i64 1
  %v2e1 = extractelement <8 x i4> %vec2, i64 1
  %mul1 = mul nuw nsw i4 %v1e1, %v2e1

  %v1e2 = extractelement <8 x i4> %vec1, i64 2
  %v2e2 = extractelement <8 x i4> %vec2, i64 2
  %mul2 = mul nuw nsw i4 %v1e2, %v2e2

  %v1e3 = extractelement <8 x i4> %vec1, i64 3
  %v2e3 = extractelement <8 x i4> %vec2, i64 3
  %mul3 = mul nuw nsw i4 %v1e3, %v2e3

  %v1e4 = extractelement <8 x i4> %vec1, i64 4
  %v2e4 = extractelement <8 x i4> %vec2, i64 4
  %mul4 = mul nuw nsw i4 %v1e4, %v2e4

  %v1e5 = extractelement <8 x i4> %vec1, i64 5
  %v2e5 = extractelement <8 x i4> %vec2, i64 5
  %mul5 = mul nuw nsw i4 %v1e5, %v2e5

  %v1e6 = extractelement <8 x i4> %vec1, i64 6
  %v2e6 = extractelement <8 x i4> %vec2, i64 6
  %mul6 = mul nuw nsw i4 %v1e6, %v2e6

  %v1e7 = extractelement <8 x i4> %vec1, i64 7
  %v2e7 = extractelement <8 x i4> %vec2, i64 7
  %mul7 = mul nuw nsw i4 %v1e7, %v2e7

  %acc = load i4, i4 addrspace(1)* %dst, align 4
  %add1 = add i4 %mul0, %acc
  %add2 = add i4 %mul1, %add1
  %add3 = add i4 %mul2, %add2
  %add4 = add i4 %mul3, %add3
  %add5 = add i4 %mul4, %add4
  %add6 = add i4 %mul5, %add5
  %add7 = add i4 %mul6, %add6
  %add8 = add i4 %mul7, %add7

  store i4 %add8, i4 addrspace(1)* %dst, align 4
  ret void
}

define amdgpu_kernel void @udot8_multiuses_mul1(<8 x i4> addrspace(1)* %src1,
; GFX7-LABEL: udot8_multiuses_mul1:
; GFX7:       ; %bb.0: ; %entry
; GFX7-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x9
; GFX7-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0xd
; GFX7-NEXT:    s_mov_b32 s3, 0xf000
; GFX7-NEXT:    s_mov_b32 s2, -1
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    s_load_dword s6, s[6:7], 0x0
; GFX7-NEXT:    s_load_dword s4, s[4:5], 0x0
; GFX7-NEXT:    s_load_dword s20, s[0:1], 0x0
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    s_bfe_u32 s19, s6, 0x40004
; GFX7-NEXT:    s_lshr_b32 s7, s6, 28
; GFX7-NEXT:    s_bfe_u32 s14, s6, 0x40018
; GFX7-NEXT:    s_bfe_u32 s15, s6, 0x40014
; GFX7-NEXT:    s_bfe_u32 s16, s6, 0x40010
; GFX7-NEXT:    s_bfe_u32 s17, s6, 0x4000c
; GFX7-NEXT:    s_bfe_u32 s18, s6, 0x40008
; GFX7-NEXT:    s_and_b32 s6, s6, 15
; GFX7-NEXT:    s_lshr_b32 s5, s4, 28
; GFX7-NEXT:    s_bfe_u32 s8, s4, 0x40018
; GFX7-NEXT:    s_bfe_u32 s9, s4, 0x40014
; GFX7-NEXT:    s_bfe_u32 s10, s4, 0x40010
; GFX7-NEXT:    s_bfe_u32 s11, s4, 0x4000c
; GFX7-NEXT:    s_bfe_u32 s12, s4, 0x40008
; GFX7-NEXT:    s_bfe_u32 s13, s4, 0x40004
; GFX7-NEXT:    s_and_b32 s4, s4, 15
; GFX7-NEXT:    v_mov_b32_e32 v0, s6
; GFX7-NEXT:    v_mov_b32_e32 v1, s20
; GFX7-NEXT:    v_mad_u32_u24 v1, s4, v0, v1
; GFX7-NEXT:    v_mov_b32_e32 v2, s19
; GFX7-NEXT:    v_mad_u32_u24 v0, s4, v0, v1
; GFX7-NEXT:    v_mad_u32_u24 v1, s13, v2, v1
; GFX7-NEXT:    v_mov_b32_e32 v2, s18
; GFX7-NEXT:    v_mad_u32_u24 v1, s12, v2, v1
; GFX7-NEXT:    v_mov_b32_e32 v2, s17
; GFX7-NEXT:    v_mad_u32_u24 v1, s11, v2, v1
; GFX7-NEXT:    v_mov_b32_e32 v2, s16
; GFX7-NEXT:    v_mad_u32_u24 v1, s10, v2, v1
; GFX7-NEXT:    v_mov_b32_e32 v2, s15
; GFX7-NEXT:    v_mad_u32_u24 v1, s9, v2, v1
; GFX7-NEXT:    v_mov_b32_e32 v2, s14
; GFX7-NEXT:    v_mad_u32_u24 v1, s8, v2, v1
; GFX7-NEXT:    v_mov_b32_e32 v2, s7
; GFX7-NEXT:    v_mad_u32_u24 v1, s5, v2, v1
; GFX7-NEXT:    v_add_i32_e32 v0, vcc, v1, v0
; GFX7-NEXT:    buffer_store_dword v0, off, s[0:3], 0
; GFX7-NEXT:    s_endpgm
;
; GFX8-LABEL: udot8_multiuses_mul1:
; GFX8:       ; %bb.0: ; %entry
; GFX8-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX8-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    s_load_dword s6, s[6:7], 0x0
; GFX8-NEXT:    s_load_dword s2, s[4:5], 0x0
; GFX8-NEXT:    s_load_dword s18, s[0:1], 0x0
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    s_bfe_u32 s17, s6, 0x40004
; GFX8-NEXT:    s_lshr_b32 s7, s6, 28
; GFX8-NEXT:    s_bfe_u32 s12, s6, 0x40018
; GFX8-NEXT:    s_bfe_u32 s13, s6, 0x40014
; GFX8-NEXT:    s_bfe_u32 s14, s6, 0x40010
; GFX8-NEXT:    s_bfe_u32 s15, s6, 0x4000c
; GFX8-NEXT:    s_bfe_u32 s16, s6, 0x40008
; GFX8-NEXT:    s_and_b32 s6, s6, 15
; GFX8-NEXT:    s_lshr_b32 s3, s2, 28
; GFX8-NEXT:    s_bfe_u32 s4, s2, 0x40018
; GFX8-NEXT:    s_bfe_u32 s5, s2, 0x40014
; GFX8-NEXT:    s_bfe_u32 s8, s2, 0x40010
; GFX8-NEXT:    s_bfe_u32 s9, s2, 0x4000c
; GFX8-NEXT:    s_bfe_u32 s10, s2, 0x40008
; GFX8-NEXT:    s_bfe_u32 s11, s2, 0x40004
; GFX8-NEXT:    s_and_b32 s2, s2, 15
; GFX8-NEXT:    v_mov_b32_e32 v0, s6
; GFX8-NEXT:    v_mov_b32_e32 v1, s18
; GFX8-NEXT:    v_mad_u32_u24 v1, s2, v0, v1
; GFX8-NEXT:    v_mov_b32_e32 v2, s17
; GFX8-NEXT:    v_mad_u32_u24 v0, s2, v0, v1
; GFX8-NEXT:    v_mad_u32_u24 v1, s11, v2, v1
; GFX8-NEXT:    v_mov_b32_e32 v2, s16
; GFX8-NEXT:    v_mad_u32_u24 v1, s10, v2, v1
; GFX8-NEXT:    v_mov_b32_e32 v2, s15
; GFX8-NEXT:    v_mad_u32_u24 v1, s9, v2, v1
; GFX8-NEXT:    v_mov_b32_e32 v2, s14
; GFX8-NEXT:    v_mad_u32_u24 v1, s8, v2, v1
; GFX8-NEXT:    v_mov_b32_e32 v2, s13
; GFX8-NEXT:    v_mad_u32_u24 v1, s5, v2, v1
; GFX8-NEXT:    v_mov_b32_e32 v2, s12
; GFX8-NEXT:    v_mad_u32_u24 v1, s4, v2, v1
; GFX8-NEXT:    v_mov_b32_e32 v2, s7
; GFX8-NEXT:    v_mad_u32_u24 v1, s3, v2, v1
; GFX8-NEXT:    v_add_u32_e32 v2, vcc, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v0, s0
; GFX8-NEXT:    v_mov_b32_e32 v1, s1
; GFX8-NEXT:    flat_store_dword v[0:1], v2
; GFX8-NEXT:    s_endpgm
;
; GFX9-LABEL: udot8_multiuses_mul1:
; GFX9:       ; %bb.0: ; %entry
; GFX9-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    s_load_dword s6, s[6:7], 0x0
; GFX9-NEXT:    s_load_dword s2, s[4:5], 0x0
; GFX9-NEXT:    s_load_dword s18, s[0:1], 0x0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    s_bfe_u32 s17, s6, 0x40004
; GFX9-NEXT:    s_lshr_b32 s7, s6, 28
; GFX9-NEXT:    s_bfe_u32 s12, s6, 0x40018
; GFX9-NEXT:    s_bfe_u32 s13, s6, 0x40014
; GFX9-NEXT:    s_bfe_u32 s14, s6, 0x40010
; GFX9-NEXT:    s_bfe_u32 s15, s6, 0x4000c
; GFX9-NEXT:    s_bfe_u32 s16, s6, 0x40008
; GFX9-NEXT:    s_and_b32 s6, s6, 15
; GFX9-NEXT:    s_lshr_b32 s3, s2, 28
; GFX9-NEXT:    s_bfe_u32 s4, s2, 0x40018
; GFX9-NEXT:    s_bfe_u32 s5, s2, 0x40014
; GFX9-NEXT:    s_bfe_u32 s8, s2, 0x40010
; GFX9-NEXT:    s_bfe_u32 s9, s2, 0x4000c
; GFX9-NEXT:    s_bfe_u32 s10, s2, 0x40008
; GFX9-NEXT:    s_bfe_u32 s11, s2, 0x40004
; GFX9-NEXT:    s_and_b32 s2, s2, 15
; GFX9-NEXT:    v_mov_b32_e32 v0, s6
; GFX9-NEXT:    v_mov_b32_e32 v1, s18
; GFX9-NEXT:    v_mad_u32_u24 v1, s2, v0, v1
; GFX9-NEXT:    v_mov_b32_e32 v2, s17
; GFX9-NEXT:    v_mad_u32_u24 v0, s2, v0, v1
; GFX9-NEXT:    v_mad_u32_u24 v1, s11, v2, v1
; GFX9-NEXT:    v_mov_b32_e32 v2, s16
; GFX9-NEXT:    v_mad_u32_u24 v1, s10, v2, v1
; GFX9-NEXT:    v_mov_b32_e32 v2, s15
; GFX9-NEXT:    v_mad_u32_u24 v1, s9, v2, v1
; GFX9-NEXT:    v_mov_b32_e32 v2, s14
; GFX9-NEXT:    v_mad_u32_u24 v1, s8, v2, v1
; GFX9-NEXT:    v_mov_b32_e32 v2, s13
; GFX9-NEXT:    v_mad_u32_u24 v1, s5, v2, v1
; GFX9-NEXT:    v_mov_b32_e32 v2, s12
; GFX9-NEXT:    v_mad_u32_u24 v1, s4, v2, v1
; GFX9-NEXT:    v_mov_b32_e32 v2, s7
; GFX9-NEXT:    v_mad_u32_u24 v1, s3, v2, v1
; GFX9-NEXT:    v_add_u32_e32 v2, v0, v1
; GFX9-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-NEXT:    global_store_dword v[0:1], v2, off
; GFX9-NEXT:    s_endpgm
;
; GFX9-DL-LABEL: udot8_multiuses_mul1:
; GFX9-DL:       ; %bb.0: ; %entry
; GFX9-DL-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-DL-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    s_load_dword s6, s[6:7], 0x0
; GFX9-DL-NEXT:    s_load_dword s2, s[4:5], 0x0
; GFX9-DL-NEXT:    s_load_dword s18, s[0:1], 0x0
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    s_bfe_u32 s17, s6, 0x40004
; GFX9-DL-NEXT:    s_lshr_b32 s7, s6, 28
; GFX9-DL-NEXT:    s_bfe_u32 s12, s6, 0x40018
; GFX9-DL-NEXT:    s_bfe_u32 s13, s6, 0x40014
; GFX9-DL-NEXT:    s_bfe_u32 s14, s6, 0x40010
; GFX9-DL-NEXT:    s_bfe_u32 s15, s6, 0x4000c
; GFX9-DL-NEXT:    s_bfe_u32 s16, s6, 0x40008
; GFX9-DL-NEXT:    s_and_b32 s6, s6, 15
; GFX9-DL-NEXT:    s_lshr_b32 s3, s2, 28
; GFX9-DL-NEXT:    s_bfe_u32 s4, s2, 0x40018
; GFX9-DL-NEXT:    s_bfe_u32 s5, s2, 0x40014
; GFX9-DL-NEXT:    s_bfe_u32 s8, s2, 0x40010
; GFX9-DL-NEXT:    s_bfe_u32 s9, s2, 0x4000c
; GFX9-DL-NEXT:    s_bfe_u32 s10, s2, 0x40008
; GFX9-DL-NEXT:    s_bfe_u32 s11, s2, 0x40004
; GFX9-DL-NEXT:    s_and_b32 s2, s2, 15
; GFX9-DL-NEXT:    v_mov_b32_e32 v0, s6
; GFX9-DL-NEXT:    v_mov_b32_e32 v1, s18
; GFX9-DL-NEXT:    v_mad_u32_u24 v1, s2, v0, v1
; GFX9-DL-NEXT:    v_mov_b32_e32 v2, s17
; GFX9-DL-NEXT:    v_mad_u32_u24 v0, s2, v0, v1
; GFX9-DL-NEXT:    v_mad_u32_u24 v1, s11, v2, v1
; GFX9-DL-NEXT:    v_mov_b32_e32 v2, s16
; GFX9-DL-NEXT:    v_mad_u32_u24 v1, s10, v2, v1
; GFX9-DL-NEXT:    v_mov_b32_e32 v2, s15
; GFX9-DL-NEXT:    v_mad_u32_u24 v1, s9, v2, v1
; GFX9-DL-NEXT:    v_mov_b32_e32 v2, s14
; GFX9-DL-NEXT:    v_mad_u32_u24 v1, s8, v2, v1
; GFX9-DL-NEXT:    v_mov_b32_e32 v2, s13
; GFX9-DL-NEXT:    v_mad_u32_u24 v1, s5, v2, v1
; GFX9-DL-NEXT:    v_mov_b32_e32 v2, s12
; GFX9-DL-NEXT:    v_mad_u32_u24 v1, s4, v2, v1
; GFX9-DL-NEXT:    v_mov_b32_e32 v2, s7
; GFX9-DL-NEXT:    v_mad_u32_u24 v1, s3, v2, v1
; GFX9-DL-NEXT:    v_add_u32_e32 v2, v0, v1
; GFX9-DL-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-DL-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-DL-NEXT:    global_store_dword v[0:1], v2, off
; GFX9-DL-NEXT:    s_endpgm
;
; GFX10-DL-LABEL: udot8_multiuses_mul1:
; GFX10-DL:       ; %bb.0: ; %entry
; GFX10-DL-NEXT:    s_clause 0x1
; GFX10-DL-NEXT:    s_load_dwordx2 s[4:5], s[0:1], 0x34
; GFX10-DL-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
; GFX10-DL-NEXT:    ; implicit-def: $vcc_hi
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    s_load_dword s6, s[4:5], 0x0
; GFX10-DL-NEXT:    s_load_dword s0, s[0:1], 0x0
; GFX10-DL-NEXT:    s_load_dword s1, s[2:3], 0x0
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    v_mov_b32_e32 v0, s6
; GFX10-DL-NEXT:    s_and_b32 s2, s0, 15
; GFX10-DL-NEXT:    s_and_b32 s3, s1, 15
; GFX10-DL-NEXT:    s_bfe_u32 s6, s0, 0x40004
; GFX10-DL-NEXT:    s_bfe_u32 s7, s1, 0x40004
; GFX10-DL-NEXT:    v_mad_u32_u24 v0, s2, s3, v0
; GFX10-DL-NEXT:    v_mad_u32_u24 v1, s6, s7, v0
; GFX10-DL-NEXT:    s_bfe_u32 s6, s0, 0x40008
; GFX10-DL-NEXT:    s_bfe_u32 s7, s1, 0x40008
; GFX10-DL-NEXT:    v_mad_u32_u24 v0, s2, s3, v0
; GFX10-DL-NEXT:    v_mad_u32_u24 v1, s6, s7, v1
; GFX10-DL-NEXT:    s_bfe_u32 s6, s0, 0x4000c
; GFX10-DL-NEXT:    s_bfe_u32 s7, s1, 0x4000c
; GFX10-DL-NEXT:    v_mad_u32_u24 v1, s6, s7, v1
; GFX10-DL-NEXT:    s_bfe_u32 s6, s0, 0x40010
; GFX10-DL-NEXT:    s_bfe_u32 s7, s1, 0x40010
; GFX10-DL-NEXT:    v_mad_u32_u24 v1, s6, s7, v1
; GFX10-DL-NEXT:    s_bfe_u32 s6, s0, 0x40014
; GFX10-DL-NEXT:    s_bfe_u32 s7, s1, 0x40014
; GFX10-DL-NEXT:    v_mad_u32_u24 v1, s6, s7, v1
; GFX10-DL-NEXT:    s_bfe_u32 s6, s0, 0x40018
; GFX10-DL-NEXT:    s_bfe_u32 s7, s1, 0x40018
; GFX10-DL-NEXT:    s_lshr_b32 s0, s0, 28
; GFX10-DL-NEXT:    s_lshr_b32 s1, s1, 28
; GFX10-DL-NEXT:    v_mad_u32_u24 v1, s6, s7, v1
; GFX10-DL-NEXT:    v_mad_u32_u24 v1, s0, s1, v1
; GFX10-DL-NEXT:    v_add_nc_u32_e32 v2, v0, v1
; GFX10-DL-NEXT:    v_mov_b32_e32 v0, s4
; GFX10-DL-NEXT:    v_mov_b32_e32 v1, s5
; GFX10-DL-NEXT:    global_store_dword v[0:1], v2, off
; GFX10-DL-NEXT:    s_endpgm
                                                <8 x i4> addrspace(1)* %src2,
                                                i32 addrspace(1)* nocapture %dst) {
entry:
  %vec1 = load <8 x i4>, <8 x i4> addrspace(1)* %src1
  %vec2 = load <8 x i4>, <8 x i4> addrspace(1)* %src2

  %v1e0 = extractelement <8 x i4> %vec1, i64 0
  %cv1e0 = zext i4 %v1e0 to i32
  %v2e0 = extractelement <8 x i4> %vec2, i64 0
  %cv2e0 = zext i4 %v2e0 to i32
  %mul0 = mul nuw nsw i32 %cv1e0, %cv2e0

  %v1e1 = extractelement <8 x i4> %vec1, i64 1
  %cv1e1 = zext i4 %v1e1 to i32
  %v2e1 = extractelement <8 x i4> %vec2, i64 1
  %cv2e1 = zext i4 %v2e1 to i32
  %mul1 = mul nuw nsw i32 %cv1e1, %cv2e1

  %v1e2 = extractelement <8 x i4> %vec1, i64 2
  %cv1e2 = zext i4 %v1e2 to i32
  %v2e2 = extractelement <8 x i4> %vec2, i64 2
  %cv2e2 = zext i4 %v2e2 to i32
  %mul2 = mul nuw nsw i32 %cv1e2, %cv2e2

  %v1e3 = extractelement <8 x i4> %vec1, i64 3
  %cv1e3 = zext i4 %v1e3 to i32
  %v2e3 = extractelement <8 x i4> %vec2, i64 3
  %cv2e3 = zext i4 %v2e3 to i32
  %mul3 = mul nuw nsw i32 %cv1e3, %cv2e3

  %v1e4 = extractelement <8 x i4> %vec1, i64 4
  %cv1e4 = zext i4 %v1e4 to i32
  %v2e4 = extractelement <8 x i4> %vec2, i64 4
  %cv2e4 = zext i4 %v2e4 to i32
  %mul4 = mul nuw nsw i32 %cv1e4, %cv2e4

  %v1e5 = extractelement <8 x i4> %vec1, i64 5
  %cv1e5 = zext i4 %v1e5 to i32
  %v2e5 = extractelement <8 x i4> %vec2, i64 5
  %cv2e5 = zext i4 %v2e5 to i32
  %mul5 = mul nuw nsw i32 %cv1e5, %cv2e5

  %v1e6 = extractelement <8 x i4> %vec1, i64 6
  %cv1e6 = zext i4 %v1e6 to i32
  %v2e6 = extractelement <8 x i4> %vec2, i64 6
  %cv2e6 = zext i4 %v2e6 to i32
  %mul6 = mul nuw nsw i32 %cv1e6, %cv2e6

  %v1e7 = extractelement <8 x i4> %vec1, i64 7
  %cv1e7 = zext i4 %v1e7 to i32
  %v2e7 = extractelement <8 x i4> %vec2, i64 7
  %cv2e7 = zext i4 %v2e7 to i32
  %mul7 = mul nuw nsw i32 %cv1e7, %cv2e7

  %acc = load i32, i32 addrspace(1)* %dst, align 4
  %add1 = add i32 %mul0, %acc
  %add = add i32  %mul0, %add1
  %add2 = add i32 %add1, %mul1
  %add3 = add i32 %add2, %mul2
  %add4 = add i32 %add3, %mul3
  %add5 = add i32 %add4, %mul4
  %add6 = add i32 %add5, %mul5
  %add7 = add i32 %add6, %mul6
  %add8 = add i32 %add7, %mul7

  %res = add i32 %add, %add8
  store i32 %res, i32 addrspace(1)* %dst, align 4
  ret void
}

define amdgpu_kernel void @udot8_acc32_vecMul(<8 x i4> addrspace(1)* %src1,
; GFX7-LABEL: udot8_acc32_vecMul:
; GFX7:       ; %bb.0: ; %entry
; GFX7-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x9
; GFX7-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0xd
; GFX7-NEXT:    s_mov_b32 s3, 0xf000
; GFX7-NEXT:    s_mov_b32 s2, -1
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    s_load_dword s6, s[6:7], 0x0
; GFX7-NEXT:    s_load_dword s4, s[4:5], 0x0
; GFX7-NEXT:    s_load_dword s20, s[0:1], 0x0
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    s_lshr_b32 s7, s6, 28
; GFX7-NEXT:    s_bfe_u32 s14, s6, 0x40018
; GFX7-NEXT:    s_bfe_u32 s15, s6, 0x40014
; GFX7-NEXT:    s_bfe_u32 s16, s6, 0x40010
; GFX7-NEXT:    s_bfe_u32 s17, s6, 0x4000c
; GFX7-NEXT:    s_bfe_u32 s18, s6, 0x40008
; GFX7-NEXT:    s_bfe_u32 s19, s6, 0x40004
; GFX7-NEXT:    s_and_b32 s6, s6, 15
; GFX7-NEXT:    s_lshr_b32 s5, s4, 28
; GFX7-NEXT:    s_bfe_u32 s8, s4, 0x40018
; GFX7-NEXT:    s_bfe_u32 s9, s4, 0x40014
; GFX7-NEXT:    s_bfe_u32 s10, s4, 0x40010
; GFX7-NEXT:    s_bfe_u32 s11, s4, 0x4000c
; GFX7-NEXT:    s_bfe_u32 s12, s4, 0x40008
; GFX7-NEXT:    s_bfe_u32 s13, s4, 0x40004
; GFX7-NEXT:    s_and_b32 s4, s4, 15
; GFX7-NEXT:    v_mov_b32_e32 v0, s6
; GFX7-NEXT:    v_mov_b32_e32 v1, s20
; GFX7-NEXT:    v_mad_u32_u24 v0, s4, v0, v1
; GFX7-NEXT:    v_mov_b32_e32 v1, s19
; GFX7-NEXT:    v_mad_u32_u24 v0, s13, v1, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s18
; GFX7-NEXT:    v_mad_u32_u24 v0, s12, v1, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s17
; GFX7-NEXT:    v_mad_u32_u24 v0, s11, v1, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s16
; GFX7-NEXT:    v_mad_u32_u24 v0, s10, v1, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s15
; GFX7-NEXT:    v_mad_u32_u24 v0, s9, v1, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s14
; GFX7-NEXT:    v_mad_u32_u24 v0, s8, v1, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s7
; GFX7-NEXT:    v_mad_u32_u24 v0, s5, v1, v0
; GFX7-NEXT:    buffer_store_dword v0, off, s[0:3], 0
; GFX7-NEXT:    s_endpgm
;
; GFX8-LABEL: udot8_acc32_vecMul:
; GFX8:       ; %bb.0: ; %entry
; GFX8-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX8-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    s_load_dword s6, s[6:7], 0x0
; GFX8-NEXT:    s_load_dword s2, s[4:5], 0x0
; GFX8-NEXT:    s_load_dword s18, s[0:1], 0x0
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    s_lshr_b32 s7, s6, 28
; GFX8-NEXT:    s_bfe_u32 s12, s6, 0x40018
; GFX8-NEXT:    s_bfe_u32 s13, s6, 0x40014
; GFX8-NEXT:    s_bfe_u32 s14, s6, 0x40010
; GFX8-NEXT:    s_bfe_u32 s15, s6, 0x4000c
; GFX8-NEXT:    s_bfe_u32 s16, s6, 0x40008
; GFX8-NEXT:    s_bfe_u32 s17, s6, 0x40004
; GFX8-NEXT:    s_and_b32 s6, s6, 15
; GFX8-NEXT:    s_lshr_b32 s3, s2, 28
; GFX8-NEXT:    s_bfe_u32 s4, s2, 0x40018
; GFX8-NEXT:    s_bfe_u32 s5, s2, 0x40014
; GFX8-NEXT:    s_bfe_u32 s8, s2, 0x40010
; GFX8-NEXT:    s_bfe_u32 s9, s2, 0x4000c
; GFX8-NEXT:    s_bfe_u32 s10, s2, 0x40008
; GFX8-NEXT:    s_bfe_u32 s11, s2, 0x40004
; GFX8-NEXT:    s_and_b32 s2, s2, 15
; GFX8-NEXT:    v_mov_b32_e32 v0, s6
; GFX8-NEXT:    v_mov_b32_e32 v1, s18
; GFX8-NEXT:    v_mad_u32_u24 v0, s2, v0, v1
; GFX8-NEXT:    v_mov_b32_e32 v1, s17
; GFX8-NEXT:    v_mad_u32_u24 v0, s11, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v1, s16
; GFX8-NEXT:    v_mad_u32_u24 v0, s10, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v1, s15
; GFX8-NEXT:    v_mad_u32_u24 v0, s9, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v1, s14
; GFX8-NEXT:    v_mad_u32_u24 v0, s8, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v1, s13
; GFX8-NEXT:    v_mad_u32_u24 v0, s5, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v1, s12
; GFX8-NEXT:    v_mad_u32_u24 v0, s4, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v1, s7
; GFX8-NEXT:    v_mad_u32_u24 v2, s3, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v0, s0
; GFX8-NEXT:    v_mov_b32_e32 v1, s1
; GFX8-NEXT:    flat_store_dword v[0:1], v2
; GFX8-NEXT:    s_endpgm
;
; GFX9-LABEL: udot8_acc32_vecMul:
; GFX9:       ; %bb.0: ; %entry
; GFX9-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    s_load_dword s6, s[6:7], 0x0
; GFX9-NEXT:    s_load_dword s2, s[4:5], 0x0
; GFX9-NEXT:    s_load_dword s18, s[0:1], 0x0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    s_lshr_b32 s7, s6, 28
; GFX9-NEXT:    s_bfe_u32 s12, s6, 0x40018
; GFX9-NEXT:    s_bfe_u32 s13, s6, 0x40014
; GFX9-NEXT:    s_bfe_u32 s14, s6, 0x40010
; GFX9-NEXT:    s_bfe_u32 s15, s6, 0x4000c
; GFX9-NEXT:    s_bfe_u32 s16, s6, 0x40008
; GFX9-NEXT:    s_bfe_u32 s17, s6, 0x40004
; GFX9-NEXT:    s_and_b32 s6, s6, 15
; GFX9-NEXT:    s_lshr_b32 s3, s2, 28
; GFX9-NEXT:    s_bfe_u32 s4, s2, 0x40018
; GFX9-NEXT:    s_bfe_u32 s5, s2, 0x40014
; GFX9-NEXT:    s_bfe_u32 s8, s2, 0x40010
; GFX9-NEXT:    s_bfe_u32 s9, s2, 0x4000c
; GFX9-NEXT:    s_bfe_u32 s10, s2, 0x40008
; GFX9-NEXT:    s_bfe_u32 s11, s2, 0x40004
; GFX9-NEXT:    s_and_b32 s2, s2, 15
; GFX9-NEXT:    v_mov_b32_e32 v0, s6
; GFX9-NEXT:    v_mov_b32_e32 v1, s18
; GFX9-NEXT:    v_mad_u32_u24 v0, s2, v0, v1
; GFX9-NEXT:    v_mov_b32_e32 v1, s17
; GFX9-NEXT:    v_mad_u32_u24 v0, s11, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, s16
; GFX9-NEXT:    v_mad_u32_u24 v0, s10, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, s15
; GFX9-NEXT:    v_mad_u32_u24 v0, s9, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, s14
; GFX9-NEXT:    v_mad_u32_u24 v0, s8, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, s13
; GFX9-NEXT:    v_mad_u32_u24 v0, s5, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, s12
; GFX9-NEXT:    v_mad_u32_u24 v0, s4, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, s7
; GFX9-NEXT:    v_mad_u32_u24 v2, s3, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-NEXT:    global_store_dword v[0:1], v2, off
; GFX9-NEXT:    s_endpgm
;
; GFX9-DL-LABEL: udot8_acc32_vecMul:
; GFX9-DL:       ; %bb.0: ; %entry
; GFX9-DL-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-DL-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    s_load_dword s2, s[6:7], 0x0
; GFX9-DL-NEXT:    s_load_dword s3, s[0:1], 0x0
; GFX9-DL-NEXT:    s_load_dword s4, s[4:5], 0x0
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    v_mov_b32_e32 v0, s2
; GFX9-DL-NEXT:    v_mov_b32_e32 v1, s3
; GFX9-DL-NEXT:    v_dot8_u32_u4 v2, s4, v0, v1
; GFX9-DL-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-DL-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-DL-NEXT:    global_store_dword v[0:1], v2, off
; GFX9-DL-NEXT:    s_endpgm
;
; GFX10-DL-LABEL: udot8_acc32_vecMul:
; GFX10-DL:       ; %bb.0: ; %entry
; GFX10-DL-NEXT:    s_clause 0x1
; GFX10-DL-NEXT:    s_load_dwordx2 s[4:5], s[0:1], 0x34
; GFX10-DL-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
; GFX10-DL-NEXT:    ; implicit-def: $vcc_hi
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    s_load_dword s6, s[4:5], 0x0
; GFX10-DL-NEXT:    s_load_dword s0, s[0:1], 0x0
; GFX10-DL-NEXT:    s_load_dword s1, s[2:3], 0x0
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    v_mov_b32_e32 v0, s6
; GFX10-DL-NEXT:    v_dot8_u32_u4 v2, s0, s1, v0
; GFX10-DL-NEXT:    v_mov_b32_e32 v0, s4
; GFX10-DL-NEXT:    v_mov_b32_e32 v1, s5
; GFX10-DL-NEXT:    global_store_dword v[0:1], v2, off
; GFX10-DL-NEXT:    s_endpgm
                                              <8 x i4> addrspace(1)* %src2,
                                              i32 addrspace(1)* nocapture %dst) {
entry:
  %vec1 = load <8 x i4>, <8 x i4> addrspace(1)* %src1
  %vec2 = load <8 x i4>, <8 x i4> addrspace(1)* %src2

  %cvec1 = zext <8 x i4> %vec1 to <8 x i32>
  %cvec2 = zext <8 x i4> %vec2 to <8 x i32>

  %mul = mul <8 x i32> %cvec1, %cvec2
  %mul0 = extractelement <8 x i32> %mul, i64 0
  %mul1 = extractelement <8 x i32> %mul, i64 1
  %mul2 = extractelement <8 x i32> %mul, i64 2
  %mul3 = extractelement <8 x i32> %mul, i64 3
  %mul4 = extractelement <8 x i32> %mul, i64 4
  %mul5 = extractelement <8 x i32> %mul, i64 5
  %mul6 = extractelement <8 x i32> %mul, i64 6
  %mul7 = extractelement <8 x i32> %mul, i64 7

  %acc = load i32, i32 addrspace(1)* %dst, align 4
  %add1 = add i32 %mul0, %acc
  %add2 = add i32 %add1, %mul1
  %add3 = add i32 %add2, %mul2
  %add4 = add i32 %add3, %mul3
  %add5 = add i32 %add4, %mul4
  %add6 = add i32 %add5, %mul5
  %add7 = add i32 %add6, %mul6
  %add8 = add i32 %add7, %mul7

  store i32 %add8, i32 addrspace(1)* %dst, align 4
  ret void
}

; TODO: Clean up the code(by default pk_mad_I16 should be generated), then
; support the pattern.
define amdgpu_kernel void @udot8_acc16_vecMul(<8 x i4> addrspace(1)* %src1,
; GFX7-LABEL: udot8_acc16_vecMul:
; GFX7:       ; %bb.0: ; %entry
; GFX7-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x9
; GFX7-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0xd
; GFX7-NEXT:    s_mov_b32 s3, 0xf000
; GFX7-NEXT:    s_mov_b32 s2, -1
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    buffer_load_ushort v0, off, s[0:3], 0
; GFX7-NEXT:    s_load_dword s4, s[4:5], 0x0
; GFX7-NEXT:    s_load_dword s5, s[6:7], 0x0
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    s_bfe_u32 s10, s4, 0x40004
; GFX7-NEXT:    s_bfe_u32 s17, s5, 0x40004
; GFX7-NEXT:    s_bfe_u32 s19, s5, 0x4000c
; GFX7-NEXT:    v_mov_b32_e32 v4, s17
; GFX7-NEXT:    s_bfe_u32 s14, s5, 0x40018
; GFX7-NEXT:    s_bfe_u32 s15, s5, 0x40014
; GFX7-NEXT:    s_bfe_u32 s16, s5, 0x40010
; GFX7-NEXT:    s_and_b32 s18, s5, 15
; GFX7-NEXT:    s_lshr_b32 s13, s5, 28
; GFX7-NEXT:    s_bfe_u32 s5, s5, 0x40008
; GFX7-NEXT:    s_bfe_u32 s12, s4, 0x4000c
; GFX7-NEXT:    v_mov_b32_e32 v2, s19
; GFX7-NEXT:    v_mul_u32_u24_e32 v2, s12, v2
; GFX7-NEXT:    v_mul_u32_u24_e32 v4, s10, v4
; GFX7-NEXT:    s_lshr_b32 s6, s4, 28
; GFX7-NEXT:    s_bfe_u32 s7, s4, 0x40018
; GFX7-NEXT:    s_bfe_u32 s8, s4, 0x40014
; GFX7-NEXT:    s_bfe_u32 s9, s4, 0x40010
; GFX7-NEXT:    s_and_b32 s11, s4, 15
; GFX7-NEXT:    v_mov_b32_e32 v3, s18
; GFX7-NEXT:    s_bfe_u32 s4, s4, 0x40008
; GFX7-NEXT:    v_mov_b32_e32 v1, s5
; GFX7-NEXT:    v_mul_u32_u24_e32 v1, s4, v1
; GFX7-NEXT:    v_lshlrev_b32_e32 v2, 16, v2
; GFX7-NEXT:    v_mul_u32_u24_e32 v3, s11, v3
; GFX7-NEXT:    v_lshlrev_b32_e32 v4, 16, v4
; GFX7-NEXT:    v_or_b32_e32 v1, v1, v2
; GFX7-NEXT:    v_or_b32_e32 v2, v3, v4
; GFX7-NEXT:    v_alignbit_b32 v3, v1, v2, 16
; GFX7-NEXT:    v_lshrrev_b32_e32 v4, 16, v1
; GFX7-NEXT:    v_mov_b32_e32 v5, s16
; GFX7-NEXT:    v_mov_b32_e32 v6, s15
; GFX7-NEXT:    v_mov_b32_e32 v7, s14
; GFX7-NEXT:    s_waitcnt vmcnt(0)
; GFX7-NEXT:    v_add_i32_e32 v0, vcc, v0, v2
; GFX7-NEXT:    v_add_i32_e32 v0, vcc, v3, v0
; GFX7-NEXT:    v_add_i32_e32 v0, vcc, v0, v1
; GFX7-NEXT:    v_add_i32_e32 v0, vcc, v4, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s9, v5, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s8, v6, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s7, v7, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s13
; GFX7-NEXT:    v_mad_u32_u24 v0, s6, v1, v0
; GFX7-NEXT:    buffer_store_short v0, off, s[0:3], 0
; GFX7-NEXT:    s_endpgm
;
; GFX8-LABEL: udot8_acc16_vecMul:
; GFX8:       ; %bb.0: ; %entry
; GFX8-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX8-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    v_mov_b32_e32 v0, s0
; GFX8-NEXT:    v_mov_b32_e32 v1, s1
; GFX8-NEXT:    flat_load_ushort v2, v[0:1]
; GFX8-NEXT:    s_load_dword s0, s[4:5], 0x0
; GFX8-NEXT:    s_load_dword s1, s[6:7], 0x0
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    s_lshr_b32 s2, s0, 28
; GFX8-NEXT:    s_bfe_u32 s10, s1, 0x40018
; GFX8-NEXT:    s_bfe_u32 s11, s1, 0x40014
; GFX8-NEXT:    s_bfe_u32 s12, s1, 0x40010
; GFX8-NEXT:    s_bfe_u32 s13, s1, 0x4000c
; GFX8-NEXT:    s_bfe_u32 s14, s1, 0x40008
; GFX8-NEXT:    s_bfe_u32 s15, s1, 0x40004
; GFX8-NEXT:    s_lshr_b32 s9, s1, 28
; GFX8-NEXT:    s_and_b32 s1, s1, 15
; GFX8-NEXT:    s_bfe_u32 s3, s0, 0x40018
; GFX8-NEXT:    s_bfe_u32 s4, s0, 0x40014
; GFX8-NEXT:    s_bfe_u32 s5, s0, 0x40010
; GFX8-NEXT:    s_bfe_u32 s6, s0, 0x4000c
; GFX8-NEXT:    s_bfe_u32 s7, s0, 0x40008
; GFX8-NEXT:    s_bfe_u32 s8, s0, 0x40004
; GFX8-NEXT:    s_and_b32 s0, s0, 15
; GFX8-NEXT:    v_mov_b32_e32 v3, s1
; GFX8-NEXT:    v_mov_b32_e32 v4, s15
; GFX8-NEXT:    v_mov_b32_e32 v5, s14
; GFX8-NEXT:    v_mov_b32_e32 v6, s13
; GFX8-NEXT:    v_mov_b32_e32 v7, s12
; GFX8-NEXT:    v_mov_b32_e32 v8, s11
; GFX8-NEXT:    v_mov_b32_e32 v9, s10
; GFX8-NEXT:    s_waitcnt vmcnt(0)
; GFX8-NEXT:    v_mad_u32_u24 v2, s0, v3, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s8, v4, v2
; GFX8-NEXT:    v_and_b32_e32 v2, 0xffff, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s7, v5, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s6, v6, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s5, v7, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s4, v8, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s3, v9, v2
; GFX8-NEXT:    v_mov_b32_e32 v3, s9
; GFX8-NEXT:    v_mad_u32_u24 v2, s2, v3, v2
; GFX8-NEXT:    flat_store_short v[0:1], v2
; GFX8-NEXT:    s_endpgm
;
; GFX9-LABEL: udot8_acc16_vecMul:
; GFX9:       ; %bb.0: ; %entry
; GFX9-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    s_load_dword s6, s[6:7], 0x0
; GFX9-NEXT:    s_load_dword s2, s[4:5], 0x0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    s_bfe_u32 s7, s6, 0x40018
; GFX9-NEXT:    s_lshr_b32 s12, s6, 28
; GFX9-NEXT:    s_pack_ll_b32_b16 s7, s7, s12
; GFX9-NEXT:    s_bfe_u32 s3, s2, 0x40018
; GFX9-NEXT:    s_lshr_b32 s4, s2, 28
; GFX9-NEXT:    s_bfe_u32 s13, s6, 0x40010
; GFX9-NEXT:    s_bfe_u32 s14, s6, 0x40014
; GFX9-NEXT:    s_pack_ll_b32_b16 s3, s3, s4
; GFX9-NEXT:    v_mov_b32_e32 v0, s7
; GFX9-NEXT:    v_pk_mul_lo_u16 v2, s3, v0
; GFX9-NEXT:    s_pack_ll_b32_b16 s3, s13, s14
; GFX9-NEXT:    s_bfe_u32 s5, s2, 0x40010
; GFX9-NEXT:    s_bfe_u32 s8, s2, 0x40014
; GFX9-NEXT:    s_bfe_u32 s15, s6, 0x40008
; GFX9-NEXT:    s_bfe_u32 s16, s6, 0x4000c
; GFX9-NEXT:    s_and_b32 s17, s6, 15
; GFX9-NEXT:    v_mov_b32_e32 v0, s3
; GFX9-NEXT:    s_pack_ll_b32_b16 s4, s5, s8
; GFX9-NEXT:    s_bfe_u32 s9, s2, 0x40008
; GFX9-NEXT:    s_bfe_u32 s10, s2, 0x4000c
; GFX9-NEXT:    s_bfe_u32 s6, s6, 0x40004
; GFX9-NEXT:    s_pack_ll_b32_b16 s3, s15, s16
; GFX9-NEXT:    v_pk_mul_lo_u16 v3, s4, v0
; GFX9-NEXT:    s_and_b32 s11, s2, 15
; GFX9-NEXT:    s_bfe_u32 s2, s2, 0x40004
; GFX9-NEXT:    v_mov_b32_e32 v0, s3
; GFX9-NEXT:    s_pack_ll_b32_b16 s4, s9, s10
; GFX9-NEXT:    s_pack_ll_b32_b16 s3, s17, s6
; GFX9-NEXT:    v_pk_mul_lo_u16 v4, s4, v0
; GFX9-NEXT:    s_pack_ll_b32_b16 s2, s11, s2
; GFX9-NEXT:    v_mov_b32_e32 v0, s3
; GFX9-NEXT:    v_pk_mul_lo_u16 v5, s2, v0
; GFX9-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-NEXT:    global_load_ushort v6, v[0:1], off
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    v_add_u32_e32 v6, v5, v6
; GFX9-NEXT:    v_add_u32_sdwa v5, v6, v5 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:WORD_1
; GFX9-NEXT:    v_add_u32_sdwa v5, v5, v4 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_0 src1_sel:BYTE_0
; GFX9-NEXT:    v_add_u32_sdwa v4, v5, v4 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:WORD_1
; GFX9-NEXT:    v_add_u32_e32 v4, v4, v3
; GFX9-NEXT:    v_add_u32_sdwa v3, v4, v3 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:WORD_1
; GFX9-NEXT:    v_add_u32_e32 v3, v3, v2
; GFX9-NEXT:    v_add_u32_sdwa v2, v3, v2 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:WORD_1
; GFX9-NEXT:    global_store_short v[0:1], v2, off
; GFX9-NEXT:    s_endpgm
;
; GFX9-DL-LABEL: udot8_acc16_vecMul:
; GFX9-DL:       ; %bb.0: ; %entry
; GFX9-DL-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-DL-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    s_load_dword s6, s[6:7], 0x0
; GFX9-DL-NEXT:    s_load_dword s2, s[4:5], 0x0
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    s_bfe_u32 s7, s6, 0x40018
; GFX9-DL-NEXT:    s_lshr_b32 s12, s6, 28
; GFX9-DL-NEXT:    s_pack_ll_b32_b16 s7, s7, s12
; GFX9-DL-NEXT:    s_bfe_u32 s3, s2, 0x40018
; GFX9-DL-NEXT:    s_lshr_b32 s4, s2, 28
; GFX9-DL-NEXT:    s_bfe_u32 s13, s6, 0x40010
; GFX9-DL-NEXT:    s_bfe_u32 s14, s6, 0x40014
; GFX9-DL-NEXT:    s_pack_ll_b32_b16 s3, s3, s4
; GFX9-DL-NEXT:    v_mov_b32_e32 v0, s7
; GFX9-DL-NEXT:    v_pk_mul_lo_u16 v2, s3, v0
; GFX9-DL-NEXT:    s_pack_ll_b32_b16 s3, s13, s14
; GFX9-DL-NEXT:    s_bfe_u32 s5, s2, 0x40010
; GFX9-DL-NEXT:    s_bfe_u32 s8, s2, 0x40014
; GFX9-DL-NEXT:    s_bfe_u32 s15, s6, 0x40008
; GFX9-DL-NEXT:    s_bfe_u32 s16, s6, 0x4000c
; GFX9-DL-NEXT:    s_and_b32 s17, s6, 15
; GFX9-DL-NEXT:    v_mov_b32_e32 v0, s3
; GFX9-DL-NEXT:    s_pack_ll_b32_b16 s4, s5, s8
; GFX9-DL-NEXT:    s_bfe_u32 s9, s2, 0x40008
; GFX9-DL-NEXT:    s_bfe_u32 s10, s2, 0x4000c
; GFX9-DL-NEXT:    s_bfe_u32 s6, s6, 0x40004
; GFX9-DL-NEXT:    s_pack_ll_b32_b16 s3, s15, s16
; GFX9-DL-NEXT:    v_pk_mul_lo_u16 v3, s4, v0
; GFX9-DL-NEXT:    s_and_b32 s11, s2, 15
; GFX9-DL-NEXT:    s_bfe_u32 s2, s2, 0x40004
; GFX9-DL-NEXT:    v_mov_b32_e32 v0, s3
; GFX9-DL-NEXT:    s_pack_ll_b32_b16 s4, s9, s10
; GFX9-DL-NEXT:    s_pack_ll_b32_b16 s3, s17, s6
; GFX9-DL-NEXT:    v_pk_mul_lo_u16 v4, s4, v0
; GFX9-DL-NEXT:    s_pack_ll_b32_b16 s2, s11, s2
; GFX9-DL-NEXT:    v_mov_b32_e32 v0, s3
; GFX9-DL-NEXT:    v_pk_mul_lo_u16 v5, s2, v0
; GFX9-DL-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-DL-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-DL-NEXT:    global_load_ushort v6, v[0:1], off
; GFX9-DL-NEXT:    s_waitcnt vmcnt(0)
; GFX9-DL-NEXT:    v_add_u32_e32 v6, v5, v6
; GFX9-DL-NEXT:    v_add_u32_sdwa v5, v6, v5 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:WORD_1
; GFX9-DL-NEXT:    v_add_u32_sdwa v5, v5, v4 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_0 src1_sel:BYTE_0
; GFX9-DL-NEXT:    v_add_u32_sdwa v4, v5, v4 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:WORD_1
; GFX9-DL-NEXT:    v_add_u32_e32 v4, v4, v3
; GFX9-DL-NEXT:    v_add_u32_sdwa v3, v4, v3 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:WORD_1
; GFX9-DL-NEXT:    v_add_u32_e32 v3, v3, v2
; GFX9-DL-NEXT:    v_add_u32_sdwa v2, v3, v2 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:WORD_1
; GFX9-DL-NEXT:    global_store_short v[0:1], v2, off
; GFX9-DL-NEXT:    s_endpgm
;
; GFX10-DL-LABEL: udot8_acc16_vecMul:
; GFX10-DL:       ; %bb.0: ; %entry
; GFX10-DL-NEXT:    s_load_dwordx2 s[2:3], s[0:1], 0x34
; GFX10-DL-NEXT:    ; implicit-def: $vcc_hi
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    v_mov_b32_e32 v0, s2
; GFX10-DL-NEXT:    v_mov_b32_e32 v1, s3
; GFX10-DL-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
; GFX10-DL-NEXT:    global_load_ushort v2, v[0:1], off
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    s_load_dword s0, s[0:1], 0x0
; GFX10-DL-NEXT:    s_load_dword s1, s[2:3], 0x0
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    s_and_b32 s2, s0, 15
; GFX10-DL-NEXT:    s_bfe_u32 s5, s0, 0x40004
; GFX10-DL-NEXT:    s_and_b32 s3, s1, 15
; GFX10-DL-NEXT:    s_bfe_u32 s4, s1, 0x40004
; GFX10-DL-NEXT:    s_pack_ll_b32_b16 s2, s2, s5
; GFX10-DL-NEXT:    s_pack_ll_b32_b16 s3, s3, s4
; GFX10-DL-NEXT:    s_bfe_u32 s4, s1, 0x40008
; GFX10-DL-NEXT:    v_pk_mul_lo_u16 v3, s2, s3
; GFX10-DL-NEXT:    s_bfe_u32 s5, s1, 0x4000c
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40008
; GFX10-DL-NEXT:    s_bfe_u32 s3, s0, 0x4000c
; GFX10-DL-NEXT:    s_pack_ll_b32_b16 s4, s4, s5
; GFX10-DL-NEXT:    s_pack_ll_b32_b16 s2, s2, s3
; GFX10-DL-NEXT:    s_bfe_u32 s3, s0, 0x40014
; GFX10-DL-NEXT:    v_pk_mul_lo_u16 v4, s2, s4
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40010
; GFX10-DL-NEXT:    s_bfe_u32 s4, s1, 0x40010
; GFX10-DL-NEXT:    s_bfe_u32 s5, s1, 0x40014
; GFX10-DL-NEXT:    s_pack_ll_b32_b16 s2, s2, s3
; GFX10-DL-NEXT:    s_pack_ll_b32_b16 s4, s4, s5
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40018
; GFX10-DL-NEXT:    s_lshr_b32 s1, s1, 28
; GFX10-DL-NEXT:    s_pack_ll_b32_b16 s1, s3, s1
; GFX10-DL-NEXT:    s_waitcnt vmcnt(0)
; GFX10-DL-NEXT:    v_add_nc_u32_e32 v2, v3, v2
; GFX10-DL-NEXT:    v_add_nc_u32_sdwa v2, v2, v3 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:WORD_1
; GFX10-DL-NEXT:    v_pk_mul_lo_u16 v3, s2, s4
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40018
; GFX10-DL-NEXT:    s_lshr_b32 s0, s0, 28
; GFX10-DL-NEXT:    v_add_nc_u32_sdwa v2, v2, v4 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_0 src1_sel:BYTE_0
; GFX10-DL-NEXT:    s_pack_ll_b32_b16 s0, s2, s0
; GFX10-DL-NEXT:    v_add_nc_u32_sdwa v2, v2, v4 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:WORD_1
; GFX10-DL-NEXT:    v_pk_mul_lo_u16 v4, s0, s1
; GFX10-DL-NEXT:    v_add_nc_u32_e32 v2, v2, v3
; GFX10-DL-NEXT:    v_add_nc_u32_sdwa v2, v2, v3 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:WORD_1
; GFX10-DL-NEXT:    v_add_nc_u32_e32 v2, v2, v4
; GFX10-DL-NEXT:    v_add_nc_u32_sdwa v2, v2, v4 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:WORD_1
; GFX10-DL-NEXT:    global_store_short v[0:1], v2, off
; GFX10-DL-NEXT:    s_endpgm
                                              <8 x i4> addrspace(1)* %src2,
                                              i16 addrspace(1)* nocapture %dst) {
entry:
  %vec1 = load <8 x i4>, <8 x i4> addrspace(1)* %src1
  %vec2 = load <8 x i4>, <8 x i4> addrspace(1)* %src2

  %cvec1 = zext <8 x i4> %vec1 to <8 x i16>
  %cvec2 = zext <8 x i4> %vec2 to <8 x i16>

  %mul = mul <8 x i16> %cvec1, %cvec2
  %mul0 = extractelement <8 x i16> %mul, i64 0
  %mul1 = extractelement <8 x i16> %mul, i64 1
  %mul2 = extractelement <8 x i16> %mul, i64 2
  %mul3 = extractelement <8 x i16> %mul, i64 3
  %mul4 = extractelement <8 x i16> %mul, i64 4
  %mul5 = extractelement <8 x i16> %mul, i64 5
  %mul6 = extractelement <8 x i16> %mul, i64 6
  %mul7 = extractelement <8 x i16> %mul, i64 7

  %acc = load i16, i16 addrspace(1)* %dst, align 4
  %add1 = add i16 %mul0, %acc
  %add2 = add i16 %add1, %mul1
  %add3 = add i16 %add2, %mul2
  %add4 = add i16 %add3, %mul3
  %add5 = add i16 %add4, %mul4
  %add6 = add i16 %add5, %mul5
  %add7 = add i16 %add6, %mul6
  %add8 = add i16 %add7, %mul7

  store i16 %add8, i16 addrspace(1)* %dst, align 4
  ret void
}

; TODO: Cleanup the code to generate MAD; pattern should be recognized then.
define amdgpu_kernel void @udot8_acc8_vecMul(<8 x i4> addrspace(1)* %src1,
; GFX7-LABEL: udot8_acc8_vecMul:
; GFX7:       ; %bb.0: ; %entry
; GFX7-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x9
; GFX7-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0xd
; GFX7-NEXT:    s_mov_b32 s3, 0xf000
; GFX7-NEXT:    s_mov_b32 s2, -1
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    buffer_load_ubyte v0, off, s[0:3], 0
; GFX7-NEXT:    s_load_dword s4, s[4:5], 0x0
; GFX7-NEXT:    s_load_dword s5, s[6:7], 0x0
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    s_bfe_u32 s6, s4, 0x4000c
; GFX7-NEXT:    s_bfe_u32 s13, s5, 0x4000c
; GFX7-NEXT:    s_bfe_u32 s15, s5, 0x40004
; GFX7-NEXT:    s_lshr_b32 s17, s5, 28
; GFX7-NEXT:    v_mov_b32_e32 v8, s13
; GFX7-NEXT:    s_bfe_u32 s14, s5, 0x40008
; GFX7-NEXT:    s_and_b32 s16, s5, 15
; GFX7-NEXT:    s_bfe_u32 s18, s5, 0x40018
; GFX7-NEXT:    s_bfe_u32 s19, s5, 0x40014
; GFX7-NEXT:    s_bfe_u32 s8, s4, 0x40004
; GFX7-NEXT:    v_mov_b32_e32 v6, s15
; GFX7-NEXT:    s_lshr_b32 s10, s4, 28
; GFX7-NEXT:    v_mov_b32_e32 v4, s17
; GFX7-NEXT:    v_mul_u32_u24_e32 v4, s10, v4
; GFX7-NEXT:    v_mul_u32_u24_e32 v6, s8, v6
; GFX7-NEXT:    v_mul_u32_u24_e32 v8, s6, v8
; GFX7-NEXT:    s_bfe_u32 s5, s5, 0x40010
; GFX7-NEXT:    s_bfe_u32 s7, s4, 0x40008
; GFX7-NEXT:    v_mov_b32_e32 v7, s14
; GFX7-NEXT:    s_and_b32 s9, s4, 15
; GFX7-NEXT:    v_mov_b32_e32 v5, s16
; GFX7-NEXT:    s_bfe_u32 s11, s4, 0x40018
; GFX7-NEXT:    v_mov_b32_e32 v3, s18
; GFX7-NEXT:    s_bfe_u32 s12, s4, 0x40014
; GFX7-NEXT:    v_mov_b32_e32 v2, s19
; GFX7-NEXT:    v_mul_u32_u24_e32 v2, s12, v2
; GFX7-NEXT:    s_bfe_u32 s4, s4, 0x40010
; GFX7-NEXT:    v_mov_b32_e32 v1, s5
; GFX7-NEXT:    v_mul_u32_u24_e32 v3, s11, v3
; GFX7-NEXT:    v_lshlrev_b32_e32 v4, 8, v4
; GFX7-NEXT:    v_mul_u32_u24_e32 v5, s9, v5
; GFX7-NEXT:    v_mul_u32_u24_e32 v7, s7, v7
; GFX7-NEXT:    v_lshlrev_b32_e32 v6, 8, v6
; GFX7-NEXT:    v_lshlrev_b32_e32 v8, 8, v8
; GFX7-NEXT:    v_or_b32_e32 v3, v3, v4
; GFX7-NEXT:    v_or_b32_e32 v4, v5, v6
; GFX7-NEXT:    v_or_b32_e32 v5, v7, v8
; GFX7-NEXT:    v_mul_u32_u24_e32 v9, s4, v1
; GFX7-NEXT:    v_lshlrev_b32_e32 v2, 8, v2
; GFX7-NEXT:    v_or_b32_e32 v2, v9, v2
; GFX7-NEXT:    v_lshlrev_b32_e32 v3, 16, v3
; GFX7-NEXT:    v_lshlrev_b32_e32 v5, 16, v5
; GFX7-NEXT:    v_or_b32_e32 v2, v2, v3
; GFX7-NEXT:    v_or_b32_e32 v3, v4, v5
; GFX7-NEXT:    v_alignbit_b32 v4, v2, v3, 8
; GFX7-NEXT:    v_alignbit_b32 v5, v2, v3, 16
; GFX7-NEXT:    v_lshrrev_b32_e32 v6, 24, v3
; GFX7-NEXT:    v_lshrrev_b32_e32 v7, 8, v2
; GFX7-NEXT:    v_lshrrev_b32_e32 v8, 16, v2
; GFX7-NEXT:    v_lshrrev_b32_e32 v2, 24, v2
; GFX7-NEXT:    s_waitcnt vmcnt(0)
; GFX7-NEXT:    v_add_i32_e32 v0, vcc, v0, v3
; GFX7-NEXT:    v_add_i32_e32 v0, vcc, v4, v0
; GFX7-NEXT:    v_add_i32_e32 v0, vcc, v5, v0
; GFX7-NEXT:    v_add_i32_e32 v0, vcc, v6, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s4, v1, v0
; GFX7-NEXT:    v_add_i32_e32 v0, vcc, v0, v7
; GFX7-NEXT:    v_add_i32_e32 v0, vcc, v0, v8
; GFX7-NEXT:    v_add_i32_e32 v0, vcc, v0, v2
; GFX7-NEXT:    buffer_store_byte v0, off, s[0:3], 0
; GFX7-NEXT:    s_endpgm
;
; GFX8-LABEL: udot8_acc8_vecMul:
; GFX8:       ; %bb.0: ; %entry
; GFX8-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX8-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    v_mov_b32_e32 v0, s0
; GFX8-NEXT:    v_mov_b32_e32 v1, s1
; GFX8-NEXT:    flat_load_ubyte v2, v[0:1]
; GFX8-NEXT:    s_load_dword s1, s[4:5], 0x0
; GFX8-NEXT:    s_load_dword s2, s[6:7], 0x0
; GFX8-NEXT:    s_mov_b32 s0, 0xffff
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    s_bfe_u32 s7, s1, 0x40004
; GFX8-NEXT:    s_bfe_u32 s9, s1, 0x4000c
; GFX8-NEXT:    s_bfe_u32 s14, s2, 0x40004
; GFX8-NEXT:    s_and_b32 s15, s2, 15
; GFX8-NEXT:    s_bfe_u32 s16, s2, 0x4000c
; GFX8-NEXT:    s_bfe_u32 s3, s1, 0x40014
; GFX8-NEXT:    s_lshr_b32 s5, s1, 28
; GFX8-NEXT:    s_bfe_u32 s10, s2, 0x40014
; GFX8-NEXT:    s_bfe_u32 s11, s2, 0x40010
; GFX8-NEXT:    s_lshr_b32 s12, s2, 28
; GFX8-NEXT:    s_bfe_u32 s13, s2, 0x40018
; GFX8-NEXT:    s_bfe_u32 s2, s2, 0x40008
; GFX8-NEXT:    s_and_b32 s8, s1, 15
; GFX8-NEXT:    v_mov_b32_e32 v4, s16
; GFX8-NEXT:    v_mov_b32_e32 v5, s9
; GFX8-NEXT:    v_mov_b32_e32 v6, s15
; GFX8-NEXT:    v_mov_b32_e32 v7, s14
; GFX8-NEXT:    v_mov_b32_e32 v8, s7
; GFX8-NEXT:    v_mul_u32_u24_sdwa v4, v5, v4 dst_sel:BYTE_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
; GFX8-NEXT:    v_mul_u32_u24_e32 v5, s8, v6
; GFX8-NEXT:    v_mul_u32_u24_sdwa v6, v8, v7 dst_sel:BYTE_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
; GFX8-NEXT:    s_bfe_u32 s4, s1, 0x40010
; GFX8-NEXT:    s_bfe_u32 s6, s1, 0x40018
; GFX8-NEXT:    v_mov_b32_e32 v9, s13
; GFX8-NEXT:    s_bfe_u32 s1, s1, 0x40008
; GFX8-NEXT:    v_mov_b32_e32 v3, s2
; GFX8-NEXT:    v_mov_b32_e32 v10, s12
; GFX8-NEXT:    v_mov_b32_e32 v11, s5
; GFX8-NEXT:    v_mov_b32_e32 v12, s11
; GFX8-NEXT:    v_mov_b32_e32 v13, s10
; GFX8-NEXT:    v_mov_b32_e32 v14, s3
; GFX8-NEXT:    v_mul_u32_u24_e32 v3, s1, v3
; GFX8-NEXT:    v_or_b32_e32 v5, v5, v6
; GFX8-NEXT:    v_mul_u32_u24_e32 v7, s6, v9
; GFX8-NEXT:    v_mul_u32_u24_sdwa v8, v11, v10 dst_sel:BYTE_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
; GFX8-NEXT:    v_mul_u32_u24_e32 v9, s4, v12
; GFX8-NEXT:    v_mul_u32_u24_sdwa v10, v14, v13 dst_sel:BYTE_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
; GFX8-NEXT:    v_and_b32_e32 v5, s0, v5
; GFX8-NEXT:    v_or_b32_sdwa v3, v3, v4 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
; GFX8-NEXT:    v_or_b32_e32 v9, v9, v10
; GFX8-NEXT:    v_or_b32_sdwa v7, v7, v8 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
; GFX8-NEXT:    v_and_b32_e32 v4, s0, v9
; GFX8-NEXT:    v_or_b32_e32 v3, v5, v3
; GFX8-NEXT:    v_or_b32_e32 v6, v4, v7
; GFX8-NEXT:    v_lshrrev_b32_e32 v7, 8, v3
; GFX8-NEXT:    v_lshrrev_b32_e32 v8, 8, v6
; GFX8-NEXT:    s_waitcnt vmcnt(0)
; GFX8-NEXT:    v_add_u32_e32 v2, vcc, v2, v5
; GFX8-NEXT:    v_add_u32_e32 v2, vcc, v7, v2
; GFX8-NEXT:    v_add_u32_sdwa v2, vcc, v3, v2 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:BYTE_2 src1_sel:BYTE_0
; GFX8-NEXT:    v_add_u32_sdwa v2, vcc, v3, v2 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:BYTE_3 src1_sel:DWORD
; GFX8-NEXT:    v_add_u32_e32 v2, vcc, v2, v4
; GFX8-NEXT:    v_add_u32_e32 v2, vcc, v8, v2
; GFX8-NEXT:    v_add_u32_sdwa v2, vcc, v6, v2 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:DWORD
; GFX8-NEXT:    v_add_u32_sdwa v2, vcc, v6, v2 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:BYTE_3 src1_sel:DWORD
; GFX8-NEXT:    flat_store_byte v[0:1], v2
; GFX8-NEXT:    s_endpgm
;
; GFX9-LABEL: udot8_acc8_vecMul:
; GFX9:       ; %bb.0: ; %entry
; GFX9-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-NEXT:    s_mov_b32 s2, 0xffff
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-NEXT:    global_load_ubyte v2, v[0:1], off
; GFX9-NEXT:    s_load_dword s0, s[4:5], 0x0
; GFX9-NEXT:    s_load_dword s1, s[6:7], 0x0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    s_bfe_u32 s3, s0, 0x40010
; GFX9-NEXT:    s_bfe_u32 s10, s1, 0x40010
; GFX9-NEXT:    s_bfe_u32 s11, s1, 0x40014
; GFX9-NEXT:    s_bfe_u32 s12, s1, 0x40018
; GFX9-NEXT:    s_lshr_b32 s13, s1, 28
; GFX9-NEXT:    s_and_b32 s14, s1, 15
; GFX9-NEXT:    s_bfe_u32 s15, s1, 0x40004
; GFX9-NEXT:    s_bfe_u32 s16, s1, 0x40008
; GFX9-NEXT:    v_mov_b32_e32 v3, s10
; GFX9-NEXT:    s_bfe_u32 s1, s1, 0x4000c
; GFX9-NEXT:    s_bfe_u32 s4, s0, 0x40014
; GFX9-NEXT:    v_mov_b32_e32 v4, s11
; GFX9-NEXT:    s_bfe_u32 s5, s0, 0x40018
; GFX9-NEXT:    v_mov_b32_e32 v5, s12
; GFX9-NEXT:    s_lshr_b32 s6, s0, 28
; GFX9-NEXT:    v_mov_b32_e32 v6, s13
; GFX9-NEXT:    s_and_b32 s7, s0, 15
; GFX9-NEXT:    v_mov_b32_e32 v7, s14
; GFX9-NEXT:    s_bfe_u32 s8, s0, 0x40004
; GFX9-NEXT:    v_mov_b32_e32 v8, s15
; GFX9-NEXT:    s_bfe_u32 s9, s0, 0x40008
; GFX9-NEXT:    v_mov_b32_e32 v9, s16
; GFX9-NEXT:    s_bfe_u32 s0, s0, 0x4000c
; GFX9-NEXT:    v_mov_b32_e32 v10, s1
; GFX9-NEXT:    v_mul_lo_u16_e32 v3, s3, v3
; GFX9-NEXT:    v_mul_lo_u16_sdwa v4, s4, v4 dst_sel:BYTE_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
; GFX9-NEXT:    v_mul_lo_u16_e32 v5, s5, v5
; GFX9-NEXT:    v_mul_lo_u16_sdwa v6, s6, v6 dst_sel:BYTE_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
; GFX9-NEXT:    v_mul_lo_u16_e32 v7, s7, v7
; GFX9-NEXT:    v_mul_lo_u16_sdwa v8, s8, v8 dst_sel:BYTE_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
; GFX9-NEXT:    v_or_b32_e32 v3, v3, v4
; GFX9-NEXT:    v_or_b32_sdwa v4, v5, v6 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
; GFX9-NEXT:    v_or_b32_e32 v5, v7, v8
; GFX9-NEXT:    v_mul_lo_u16_e32 v9, s9, v9
; GFX9-NEXT:    v_mul_lo_u16_sdwa v10, s0, v10 dst_sel:BYTE_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
; GFX9-NEXT:    v_and_b32_e32 v5, s2, v5
; GFX9-NEXT:    v_or_b32_sdwa v6, v9, v10 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
; GFX9-NEXT:    v_or_b32_e32 v6, v5, v6
; GFX9-NEXT:    v_lshrrev_b32_e32 v7, 8, v6
; GFX9-NEXT:    v_and_b32_e32 v3, s2, v3
; GFX9-NEXT:    v_or_b32_e32 v4, v3, v4
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    v_add_u32_e32 v2, v5, v2
; GFX9-NEXT:    v_add_u32_e32 v2, v2, v7
; GFX9-NEXT:    v_add_u32_sdwa v2, v2, v6 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:BYTE_0 src1_sel:BYTE_2
; GFX9-NEXT:    v_add_u32_sdwa v2, v2, v6 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:BYTE_3
; GFX9-NEXT:    v_add_u32_e32 v2, v2, v3
; GFX9-NEXT:    v_lshrrev_b32_e32 v3, 8, v4
; GFX9-NEXT:    v_add_u32_e32 v2, v2, v3
; GFX9-NEXT:    v_add_u32_sdwa v2, v2, v4 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:WORD_1
; GFX9-NEXT:    v_add_u32_sdwa v2, v2, v4 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:BYTE_3
; GFX9-NEXT:    global_store_byte v[0:1], v2, off
; GFX9-NEXT:    s_endpgm
;
; GFX9-DL-LABEL: udot8_acc8_vecMul:
; GFX9-DL:       ; %bb.0: ; %entry
; GFX9-DL-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-DL-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-DL-NEXT:    s_mov_b32 s2, 0xffff
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-DL-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-DL-NEXT:    global_load_ubyte v2, v[0:1], off
; GFX9-DL-NEXT:    s_load_dword s0, s[4:5], 0x0
; GFX9-DL-NEXT:    s_load_dword s1, s[6:7], 0x0
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    s_bfe_u32 s3, s0, 0x40010
; GFX9-DL-NEXT:    s_bfe_u32 s10, s1, 0x40010
; GFX9-DL-NEXT:    s_bfe_u32 s11, s1, 0x40014
; GFX9-DL-NEXT:    s_bfe_u32 s12, s1, 0x40018
; GFX9-DL-NEXT:    s_lshr_b32 s13, s1, 28
; GFX9-DL-NEXT:    s_and_b32 s14, s1, 15
; GFX9-DL-NEXT:    s_bfe_u32 s15, s1, 0x40004
; GFX9-DL-NEXT:    s_bfe_u32 s16, s1, 0x40008
; GFX9-DL-NEXT:    v_mov_b32_e32 v3, s10
; GFX9-DL-NEXT:    s_bfe_u32 s1, s1, 0x4000c
; GFX9-DL-NEXT:    s_bfe_u32 s4, s0, 0x40014
; GFX9-DL-NEXT:    v_mov_b32_e32 v4, s11
; GFX9-DL-NEXT:    s_bfe_u32 s5, s0, 0x40018
; GFX9-DL-NEXT:    v_mov_b32_e32 v5, s12
; GFX9-DL-NEXT:    s_lshr_b32 s6, s0, 28
; GFX9-DL-NEXT:    v_mov_b32_e32 v6, s13
; GFX9-DL-NEXT:    s_and_b32 s7, s0, 15
; GFX9-DL-NEXT:    v_mov_b32_e32 v7, s14
; GFX9-DL-NEXT:    s_bfe_u32 s8, s0, 0x40004
; GFX9-DL-NEXT:    v_mov_b32_e32 v8, s15
; GFX9-DL-NEXT:    s_bfe_u32 s9, s0, 0x40008
; GFX9-DL-NEXT:    v_mov_b32_e32 v9, s16
; GFX9-DL-NEXT:    s_bfe_u32 s0, s0, 0x4000c
; GFX9-DL-NEXT:    v_mov_b32_e32 v10, s1
; GFX9-DL-NEXT:    v_mul_lo_u16_e32 v3, s3, v3
; GFX9-DL-NEXT:    v_mul_lo_u16_sdwa v4, s4, v4 dst_sel:BYTE_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
; GFX9-DL-NEXT:    v_mul_lo_u16_e32 v5, s5, v5
; GFX9-DL-NEXT:    v_mul_lo_u16_sdwa v6, s6, v6 dst_sel:BYTE_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
; GFX9-DL-NEXT:    v_mul_lo_u16_e32 v7, s7, v7
; GFX9-DL-NEXT:    v_mul_lo_u16_sdwa v8, s8, v8 dst_sel:BYTE_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
; GFX9-DL-NEXT:    v_or_b32_e32 v3, v3, v4
; GFX9-DL-NEXT:    v_or_b32_sdwa v4, v5, v6 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
; GFX9-DL-NEXT:    v_or_b32_e32 v5, v7, v8
; GFX9-DL-NEXT:    v_mul_lo_u16_e32 v9, s9, v9
; GFX9-DL-NEXT:    v_mul_lo_u16_sdwa v10, s0, v10 dst_sel:BYTE_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
; GFX9-DL-NEXT:    v_and_b32_e32 v5, s2, v5
; GFX9-DL-NEXT:    v_or_b32_sdwa v6, v9, v10 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
; GFX9-DL-NEXT:    v_or_b32_e32 v6, v5, v6
; GFX9-DL-NEXT:    v_lshrrev_b32_e32 v7, 8, v6
; GFX9-DL-NEXT:    v_and_b32_e32 v3, s2, v3
; GFX9-DL-NEXT:    v_or_b32_e32 v4, v3, v4
; GFX9-DL-NEXT:    s_waitcnt vmcnt(0)
; GFX9-DL-NEXT:    v_add_u32_e32 v2, v5, v2
; GFX9-DL-NEXT:    v_add_u32_e32 v2, v2, v7
; GFX9-DL-NEXT:    v_add_u32_sdwa v2, v2, v6 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:BYTE_0 src1_sel:BYTE_2
; GFX9-DL-NEXT:    v_add_u32_sdwa v2, v2, v6 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:BYTE_3
; GFX9-DL-NEXT:    v_add_u32_e32 v2, v2, v3
; GFX9-DL-NEXT:    v_lshrrev_b32_e32 v3, 8, v4
; GFX9-DL-NEXT:    v_add_u32_e32 v2, v2, v3
; GFX9-DL-NEXT:    v_add_u32_sdwa v2, v2, v4 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:WORD_1
; GFX9-DL-NEXT:    v_add_u32_sdwa v2, v2, v4 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:BYTE_3
; GFX9-DL-NEXT:    global_store_byte v[0:1], v2, off
; GFX9-DL-NEXT:    s_endpgm
;
; GFX10-DL-LABEL: udot8_acc8_vecMul:
; GFX10-DL:       ; %bb.0: ; %entry
; GFX10-DL-NEXT:    s_load_dwordx2 s[2:3], s[0:1], 0x34
; GFX10-DL-NEXT:    ; implicit-def: $vcc_hi
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    v_mov_b32_e32 v0, s2
; GFX10-DL-NEXT:    v_mov_b32_e32 v1, s3
; GFX10-DL-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
; GFX10-DL-NEXT:    global_load_ubyte v2, v[0:1], off
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    s_load_dword s0, s[0:1], 0x0
; GFX10-DL-NEXT:    s_load_dword s1, s[2:3], 0x0
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    s_bfe_u32 s3, s0, 0x40004
; GFX10-DL-NEXT:    s_bfe_u32 s5, s1, 0x40004
; GFX10-DL-NEXT:    s_and_b32 s2, s0, 15
; GFX10-DL-NEXT:    v_mul_lo_u16_e64 v3, s3, s5
; GFX10-DL-NEXT:    s_and_b32 s3, s1, 15
; GFX10-DL-NEXT:    s_bfe_u32 s6, s0, 0x4000c
; GFX10-DL-NEXT:    s_bfe_u32 s5, s1, 0x4000c
; GFX10-DL-NEXT:    v_mul_lo_u16_e64 v4, s2, s3
; GFX10-DL-NEXT:    v_mul_lo_u16_e64 v5, s6, s5
; GFX10-DL-NEXT:    v_lshlrev_b16_e64 v3, 8, v3
; GFX10-DL-NEXT:    s_bfe_u32 s4, s0, 0x40008
; GFX10-DL-NEXT:    s_bfe_u32 s2, s1, 0x40008
; GFX10-DL-NEXT:    s_mov_b32 s3, 0xffff
; GFX10-DL-NEXT:    v_mul_lo_u16_e64 v6, s4, s2
; GFX10-DL-NEXT:    v_or_b32_e32 v3, v4, v3
; GFX10-DL-NEXT:    v_lshlrev_b16_e64 v5, 8, v5
; GFX10-DL-NEXT:    s_bfe_u32 s4, s0, 0x40014
; GFX10-DL-NEXT:    s_bfe_u32 s6, s1, 0x40014
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40010
; GFX10-DL-NEXT:    v_and_b32_e32 v3, s3, v3
; GFX10-DL-NEXT:    v_or_b32_sdwa v4, v6, v5 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
; GFX10-DL-NEXT:    v_mul_lo_u16_e64 v5, s4, s6
; GFX10-DL-NEXT:    s_bfe_u32 s5, s0, 0x40018
; GFX10-DL-NEXT:    s_bfe_u32 s7, s1, 0x40010
; GFX10-DL-NEXT:    s_lshr_b32 s0, s0, 28
; GFX10-DL-NEXT:    v_or_b32_e32 v4, v3, v4
; GFX10-DL-NEXT:    s_lshr_b32 s4, s1, 28
; GFX10-DL-NEXT:    v_mul_lo_u16_e64 v6, s2, s7
; GFX10-DL-NEXT:    v_mul_lo_u16_e64 v7, s0, s4
; GFX10-DL-NEXT:    v_lshlrev_b16_e64 v5, 8, v5
; GFX10-DL-NEXT:    v_lshrrev_b32_e32 v8, 8, v4
; GFX10-DL-NEXT:    s_bfe_u32 s0, s1, 0x40018
; GFX10-DL-NEXT:    v_mul_lo_u16_e64 v10, s5, s0
; GFX10-DL-NEXT:    v_or_b32_e32 v5, v6, v5
; GFX10-DL-NEXT:    v_lshlrev_b16_e64 v7, 8, v7
; GFX10-DL-NEXT:    v_and_b32_e32 v5, s3, v5
; GFX10-DL-NEXT:    s_waitcnt vmcnt(0)
; GFX10-DL-NEXT:    v_add_nc_u32_e32 v2, v3, v2
; GFX10-DL-NEXT:    v_or_b32_sdwa v3, v10, v7 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
; GFX10-DL-NEXT:    v_add_nc_u32_e32 v2, v2, v8
; GFX10-DL-NEXT:    v_or_b32_e32 v3, v5, v3
; GFX10-DL-NEXT:    v_add_nc_u32_sdwa v2, v2, v4 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:BYTE_0 src1_sel:BYTE_2
; GFX10-DL-NEXT:    v_add_nc_u32_sdwa v2, v2, v4 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:BYTE_3
; GFX10-DL-NEXT:    v_lshrrev_b32_e32 v4, 8, v3
; GFX10-DL-NEXT:    v_add_nc_u32_e32 v2, v2, v5
; GFX10-DL-NEXT:    v_add_nc_u32_e32 v2, v2, v4
; GFX10-DL-NEXT:    v_add_nc_u32_sdwa v2, v2, v3 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:WORD_1
; GFX10-DL-NEXT:    v_add_nc_u32_sdwa v2, v2, v3 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:BYTE_3
; GFX10-DL-NEXT:    global_store_byte v[0:1], v2, off
; GFX10-DL-NEXT:    s_endpgm
                                             <8 x i4> addrspace(1)* %src2,
                                             i8 addrspace(1)* nocapture %dst) {
entry:
  %vec1 = load <8 x i4>, <8 x i4> addrspace(1)* %src1
  %vec2 = load <8 x i4>, <8 x i4> addrspace(1)* %src2

  %cvec1 = zext <8 x i4> %vec1 to <8 x i8>
  %cvec2 = zext <8 x i4> %vec2 to <8 x i8>

  %mul = mul <8 x i8> %cvec1, %cvec2
  %mul0 = extractelement <8 x i8> %mul, i64 0
  %mul1 = extractelement <8 x i8> %mul, i64 1
  %mul2 = extractelement <8 x i8> %mul, i64 2
  %mul3 = extractelement <8 x i8> %mul, i64 3
  %mul4 = extractelement <8 x i8> %mul, i64 4
  %mul5 = extractelement <8 x i8> %mul, i64 5
  %mul6 = extractelement <8 x i8> %mul, i64 6
  %mul7 = extractelement <8 x i8> %mul, i64 7

  %acc = load i8, i8 addrspace(1)* %dst, align 4
  %add1 = add i8 %mul0, %acc
  %add2 = add i8 %add1, %mul1
  %add3 = add i8 %add2, %mul2
  %add4 = add i8 %add3, %mul3
  %add5 = add i8 %add4, %mul4
  %add6 = add i8 %add5, %mul5
  %add7 = add i8 %add6, %mul6
  %add8 = add i8 %add7, %mul7

  store i8 %add8, i8 addrspace(1)* %dst, align 4
  ret void
}

; TODO: Once the adictional "and+add" are removed, the pattern will be recognized.
define amdgpu_kernel void @udot8_acc4_vecMul(<8 x i4> addrspace(1)* %src1,
; GFX7-LABEL: udot8_acc4_vecMul:
; GFX7:       ; %bb.0: ; %entry
; GFX7-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x9
; GFX7-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0xd
; GFX7-NEXT:    s_mov_b32 s3, 0xf000
; GFX7-NEXT:    s_mov_b32 s2, -1
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    buffer_load_ubyte v0, off, s[0:3], 0
; GFX7-NEXT:    s_load_dword s4, s[4:5], 0x0
; GFX7-NEXT:    s_load_dword s5, s[6:7], 0x0
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    s_lshr_b32 s6, s4, 28
; GFX7-NEXT:    s_bfe_u32 s14, s5, 0x40018
; GFX7-NEXT:    s_bfe_u32 s15, s5, 0x40014
; GFX7-NEXT:    s_bfe_u32 s16, s5, 0x40010
; GFX7-NEXT:    s_bfe_u32 s17, s5, 0x4000c
; GFX7-NEXT:    s_bfe_u32 s18, s5, 0x40008
; GFX7-NEXT:    s_bfe_u32 s19, s5, 0x40004
; GFX7-NEXT:    s_lshr_b32 s13, s5, 28
; GFX7-NEXT:    s_and_b32 s5, s5, 15
; GFX7-NEXT:    s_bfe_u32 s7, s4, 0x40018
; GFX7-NEXT:    s_bfe_u32 s8, s4, 0x40014
; GFX7-NEXT:    s_bfe_u32 s9, s4, 0x40010
; GFX7-NEXT:    s_bfe_u32 s10, s4, 0x4000c
; GFX7-NEXT:    s_bfe_u32 s11, s4, 0x40008
; GFX7-NEXT:    s_bfe_u32 s12, s4, 0x40004
; GFX7-NEXT:    s_and_b32 s4, s4, 15
; GFX7-NEXT:    v_mov_b32_e32 v1, s5
; GFX7-NEXT:    v_mov_b32_e32 v2, s19
; GFX7-NEXT:    v_mov_b32_e32 v3, s18
; GFX7-NEXT:    v_mov_b32_e32 v4, s17
; GFX7-NEXT:    v_mov_b32_e32 v5, s16
; GFX7-NEXT:    v_mov_b32_e32 v6, s15
; GFX7-NEXT:    v_mov_b32_e32 v7, s14
; GFX7-NEXT:    s_waitcnt vmcnt(0)
; GFX7-NEXT:    v_mad_u32_u24 v0, s4, v1, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s12, v2, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s11, v3, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s10, v4, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s9, v5, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s8, v6, v0
; GFX7-NEXT:    v_mad_u32_u24 v0, s7, v7, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s13
; GFX7-NEXT:    v_mad_u32_u24 v0, s6, v1, v0
; GFX7-NEXT:    v_and_b32_e32 v0, 15, v0
; GFX7-NEXT:    buffer_store_byte v0, off, s[0:3], 0
; GFX7-NEXT:    s_endpgm
;
; GFX8-LABEL: udot8_acc4_vecMul:
; GFX8:       ; %bb.0: ; %entry
; GFX8-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX8-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    v_mov_b32_e32 v0, s0
; GFX8-NEXT:    v_mov_b32_e32 v1, s1
; GFX8-NEXT:    flat_load_ubyte v2, v[0:1]
; GFX8-NEXT:    s_load_dword s0, s[4:5], 0x0
; GFX8-NEXT:    s_load_dword s1, s[6:7], 0x0
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    s_and_b32 s8, s0, 15
; GFX8-NEXT:    s_and_b32 s15, s1, 15
; GFX8-NEXT:    s_bfe_u32 s14, s1, 0x40004
; GFX8-NEXT:    v_mov_b32_e32 v4, s15
; GFX8-NEXT:    s_bfe_u32 s10, s1, 0x40018
; GFX8-NEXT:    s_bfe_u32 s11, s1, 0x40014
; GFX8-NEXT:    s_bfe_u32 s12, s1, 0x40010
; GFX8-NEXT:    s_bfe_u32 s13, s1, 0x40008
; GFX8-NEXT:    s_lshr_b32 s9, s1, 28
; GFX8-NEXT:    s_bfe_u32 s1, s1, 0x4000c
; GFX8-NEXT:    s_bfe_u32 s7, s0, 0x40004
; GFX8-NEXT:    v_mov_b32_e32 v5, s14
; GFX8-NEXT:    s_lshr_b32 s2, s0, 28
; GFX8-NEXT:    s_bfe_u32 s3, s0, 0x40018
; GFX8-NEXT:    s_bfe_u32 s4, s0, 0x40014
; GFX8-NEXT:    s_bfe_u32 s5, s0, 0x40010
; GFX8-NEXT:    s_bfe_u32 s6, s0, 0x40008
; GFX8-NEXT:    s_bfe_u32 s0, s0, 0x4000c
; GFX8-NEXT:    v_mov_b32_e32 v3, s1
; GFX8-NEXT:    v_mov_b32_e32 v6, s13
; GFX8-NEXT:    v_mul_u32_u24_e32 v3, s0, v3
; GFX8-NEXT:    v_and_b32_e32 v3, 15, v3
; GFX8-NEXT:    v_mov_b32_e32 v7, s12
; GFX8-NEXT:    v_mov_b32_e32 v8, s11
; GFX8-NEXT:    v_mov_b32_e32 v9, s10
; GFX8-NEXT:    s_waitcnt vmcnt(0)
; GFX8-NEXT:    v_mad_u32_u24 v2, s8, v4, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s7, v5, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s6, v6, v2
; GFX8-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX8-NEXT:    v_add_u32_e32 v2, vcc, v3, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s5, v7, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s4, v8, v2
; GFX8-NEXT:    v_mad_u32_u24 v2, s3, v9, v2
; GFX8-NEXT:    v_mov_b32_e32 v3, s9
; GFX8-NEXT:    v_mad_u32_u24 v2, s2, v3, v2
; GFX8-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX8-NEXT:    flat_store_byte v[0:1], v2
; GFX8-NEXT:    s_endpgm
;
; GFX9-LABEL: udot8_acc4_vecMul:
; GFX9:       ; %bb.0: ; %entry
; GFX9-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-NEXT:    global_load_ubyte v2, v[0:1], off
; GFX9-NEXT:    s_load_dword s0, s[4:5], 0x0
; GFX9-NEXT:    s_load_dword s1, s[6:7], 0x0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    s_and_b32 s8, s0, 15
; GFX9-NEXT:    s_and_b32 s15, s1, 15
; GFX9-NEXT:    s_bfe_u32 s14, s1, 0x40004
; GFX9-NEXT:    v_mov_b32_e32 v4, s15
; GFX9-NEXT:    s_bfe_u32 s10, s1, 0x40018
; GFX9-NEXT:    s_bfe_u32 s11, s1, 0x40014
; GFX9-NEXT:    s_bfe_u32 s12, s1, 0x40010
; GFX9-NEXT:    s_bfe_u32 s13, s1, 0x40008
; GFX9-NEXT:    s_lshr_b32 s9, s1, 28
; GFX9-NEXT:    s_bfe_u32 s1, s1, 0x4000c
; GFX9-NEXT:    s_bfe_u32 s7, s0, 0x40004
; GFX9-NEXT:    v_mov_b32_e32 v5, s14
; GFX9-NEXT:    s_lshr_b32 s2, s0, 28
; GFX9-NEXT:    s_bfe_u32 s3, s0, 0x40018
; GFX9-NEXT:    s_bfe_u32 s4, s0, 0x40014
; GFX9-NEXT:    s_bfe_u32 s5, s0, 0x40010
; GFX9-NEXT:    s_bfe_u32 s6, s0, 0x40008
; GFX9-NEXT:    s_bfe_u32 s0, s0, 0x4000c
; GFX9-NEXT:    v_mov_b32_e32 v3, s1
; GFX9-NEXT:    v_mov_b32_e32 v6, s13
; GFX9-NEXT:    v_mul_u32_u24_e32 v3, s0, v3
; GFX9-NEXT:    v_and_b32_e32 v3, 15, v3
; GFX9-NEXT:    v_mov_b32_e32 v7, s12
; GFX9-NEXT:    v_mov_b32_e32 v8, s11
; GFX9-NEXT:    v_mov_b32_e32 v9, s10
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    v_mad_u32_u24 v2, s8, v4, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s7, v5, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s6, v6, v2
; GFX9-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX9-NEXT:    v_add_u32_e32 v2, v2, v3
; GFX9-NEXT:    v_mad_u32_u24 v2, s5, v7, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s4, v8, v2
; GFX9-NEXT:    v_mad_u32_u24 v2, s3, v9, v2
; GFX9-NEXT:    v_mov_b32_e32 v3, s9
; GFX9-NEXT:    v_mad_u32_u24 v2, s2, v3, v2
; GFX9-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX9-NEXT:    global_store_byte v[0:1], v2, off
; GFX9-NEXT:    s_endpgm
;
; GFX9-DL-LABEL: udot8_acc4_vecMul:
; GFX9-DL:       ; %bb.0: ; %entry
; GFX9-DL-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-DL-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-DL-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-DL-NEXT:    global_load_ubyte v2, v[0:1], off
; GFX9-DL-NEXT:    s_load_dword s0, s[4:5], 0x0
; GFX9-DL-NEXT:    s_load_dword s1, s[6:7], 0x0
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    s_and_b32 s8, s0, 15
; GFX9-DL-NEXT:    s_and_b32 s15, s1, 15
; GFX9-DL-NEXT:    s_bfe_u32 s14, s1, 0x40004
; GFX9-DL-NEXT:    v_mov_b32_e32 v4, s15
; GFX9-DL-NEXT:    s_bfe_u32 s10, s1, 0x40018
; GFX9-DL-NEXT:    s_bfe_u32 s11, s1, 0x40014
; GFX9-DL-NEXT:    s_bfe_u32 s12, s1, 0x40010
; GFX9-DL-NEXT:    s_bfe_u32 s13, s1, 0x40008
; GFX9-DL-NEXT:    s_lshr_b32 s9, s1, 28
; GFX9-DL-NEXT:    s_bfe_u32 s1, s1, 0x4000c
; GFX9-DL-NEXT:    s_bfe_u32 s7, s0, 0x40004
; GFX9-DL-NEXT:    v_mov_b32_e32 v5, s14
; GFX9-DL-NEXT:    s_lshr_b32 s2, s0, 28
; GFX9-DL-NEXT:    s_bfe_u32 s3, s0, 0x40018
; GFX9-DL-NEXT:    s_bfe_u32 s4, s0, 0x40014
; GFX9-DL-NEXT:    s_bfe_u32 s5, s0, 0x40010
; GFX9-DL-NEXT:    s_bfe_u32 s6, s0, 0x40008
; GFX9-DL-NEXT:    s_bfe_u32 s0, s0, 0x4000c
; GFX9-DL-NEXT:    v_mov_b32_e32 v3, s1
; GFX9-DL-NEXT:    v_mov_b32_e32 v6, s13
; GFX9-DL-NEXT:    v_mul_u32_u24_e32 v3, s0, v3
; GFX9-DL-NEXT:    v_and_b32_e32 v3, 15, v3
; GFX9-DL-NEXT:    v_mov_b32_e32 v7, s12
; GFX9-DL-NEXT:    v_mov_b32_e32 v8, s11
; GFX9-DL-NEXT:    v_mov_b32_e32 v9, s10
; GFX9-DL-NEXT:    s_waitcnt vmcnt(0)
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s8, v4, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s7, v5, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s6, v6, v2
; GFX9-DL-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX9-DL-NEXT:    v_add_u32_e32 v2, v2, v3
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s5, v7, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s4, v8, v2
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s3, v9, v2
; GFX9-DL-NEXT:    v_mov_b32_e32 v3, s9
; GFX9-DL-NEXT:    v_mad_u32_u24 v2, s2, v3, v2
; GFX9-DL-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX9-DL-NEXT:    global_store_byte v[0:1], v2, off
; GFX9-DL-NEXT:    s_endpgm
;
; GFX10-DL-LABEL: udot8_acc4_vecMul:
; GFX10-DL:       ; %bb.0: ; %entry
; GFX10-DL-NEXT:    s_load_dwordx2 s[2:3], s[0:1], 0x34
; GFX10-DL-NEXT:    ; implicit-def: $vcc_hi
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    v_mov_b32_e32 v0, s2
; GFX10-DL-NEXT:    v_mov_b32_e32 v1, s3
; GFX10-DL-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
; GFX10-DL-NEXT:    global_load_ubyte v2, v[0:1], off
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    s_load_dword s0, s[0:1], 0x0
; GFX10-DL-NEXT:    s_load_dword s1, s[2:3], 0x0
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    s_and_b32 s2, s0, 15
; GFX10-DL-NEXT:    s_and_b32 s3, s1, 15
; GFX10-DL-NEXT:    s_bfe_u32 s4, s1, 0x40008
; GFX10-DL-NEXT:    s_bfe_u32 s5, s1, 0x4000c
; GFX10-DL-NEXT:    s_waitcnt vmcnt(0)
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40004
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40004
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40008
; GFX10-DL-NEXT:    s_bfe_u32 s3, s0, 0x4000c
; GFX10-DL-NEXT:    v_mul_u32_u24_e64 v3, s3, s5
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s4, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40010
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40010
; GFX10-DL-NEXT:    v_and_b32_e32 v3, 15, v3
; GFX10-DL-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX10-DL-NEXT:    v_add_nc_u32_e32 v2, v2, v3
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40014
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40014
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    s_bfe_u32 s2, s0, 0x40018
; GFX10-DL-NEXT:    s_bfe_u32 s3, s1, 0x40018
; GFX10-DL-NEXT:    s_lshr_b32 s0, s0, 28
; GFX10-DL-NEXT:    s_lshr_b32 s1, s1, 28
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s2, s3, v2
; GFX10-DL-NEXT:    v_mad_u32_u24 v2, s0, s1, v2
; GFX10-DL-NEXT:    v_and_b32_e32 v2, 15, v2
; GFX10-DL-NEXT:    global_store_byte v[0:1], v2, off
; GFX10-DL-NEXT:    s_endpgm
                                             <8 x i4> addrspace(1)* %src2,
                                             i4 addrspace(1)* nocapture %dst) {
entry:
  %vec1 = load <8 x i4>, <8 x i4> addrspace(1)* %src1
  %vec2 = load <8 x i4>, <8 x i4> addrspace(1)* %src2

  %mul = mul <8 x i4> %vec1, %vec2
  %mul0 = extractelement <8 x i4> %mul, i64 0
  %mul1 = extractelement <8 x i4> %mul, i64 1
  %mul2 = extractelement <8 x i4> %mul, i64 2
  %mul3 = extractelement <8 x i4> %mul, i64 3
  %mul4 = extractelement <8 x i4> %mul, i64 4
  %mul5 = extractelement <8 x i4> %mul, i64 5
  %mul6 = extractelement <8 x i4> %mul, i64 6
  %mul7 = extractelement <8 x i4> %mul, i64 7

  %acc = load i4, i4 addrspace(1)* %dst, align 4
  %add1 = add i4 %mul0, %acc
  %add2 = add i4 %add1, %mul1
  %add3 = add i4 %add2, %mul2
  %add4 = add i4 %add3, %mul3
  %add5 = add i4 %add4, %mul4
  %add6 = add i4 %add5, %mul5
  %add7 = add i4 %add6, %mul6
  %add8 = add i4 %add7, %mul7

  store i4 %add8, i4 addrspace(1)* %dst, align 4
  ret void
}

define amdgpu_kernel void @udot8_variant1(i32 addrspace(1)* %v1addr,
; GFX7-LABEL: udot8_variant1:
; GFX7:       ; %bb.0: ; %entry
; GFX7-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x9
; GFX7-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0xd
; GFX7-NEXT:    s_mov_b32 s3, 0xf000
; GFX7-NEXT:    s_mov_b32 s2, -1
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    s_load_dword s4, s[4:5], 0x0
; GFX7-NEXT:    s_load_dword s5, s[6:7], 0x0
; GFX7-NEXT:    s_load_dword s20, s[0:1], 0x0
; GFX7-NEXT:    s_waitcnt lgkmcnt(0)
; GFX7-NEXT:    s_and_b32 s6, s4, 15
; GFX7-NEXT:    s_and_b32 s7, s5, 15
; GFX7-NEXT:    s_bfe_u32 s8, s4, 0x40004
; GFX7-NEXT:    s_bfe_u32 s10, s4, 0x40008
; GFX7-NEXT:    s_bfe_u32 s12, s4, 0x4000c
; GFX7-NEXT:    s_bfe_u32 s14, s4, 0x40010
; GFX7-NEXT:    s_bfe_u32 s16, s4, 0x40014
; GFX7-NEXT:    s_bfe_u32 s18, s4, 0x40018
; GFX7-NEXT:    s_lshr_b32 s4, s4, 28
; GFX7-NEXT:    v_mov_b32_e32 v0, s6
; GFX7-NEXT:    v_mov_b32_e32 v1, s20
; GFX7-NEXT:    v_mad_u32_u24 v0, s7, v0, v1
; GFX7-NEXT:    s_bfe_u32 s9, s5, 0x40004
; GFX7-NEXT:    s_bfe_u32 s11, s5, 0x40008
; GFX7-NEXT:    s_bfe_u32 s13, s5, 0x4000c
; GFX7-NEXT:    s_bfe_u32 s15, s5, 0x40010
; GFX7-NEXT:    s_bfe_u32 s17, s5, 0x40014
; GFX7-NEXT:    s_bfe_u32 s19, s5, 0x40018
; GFX7-NEXT:    s_lshr_b32 s5, s5, 28
; GFX7-NEXT:    v_mov_b32_e32 v1, s4
; GFX7-NEXT:    v_mad_u32_u24 v0, s5, v1, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s8
; GFX7-NEXT:    v_mad_u32_u24 v0, s9, v1, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s10
; GFX7-NEXT:    v_mad_u32_u24 v0, s11, v1, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s12
; GFX7-NEXT:    v_mad_u32_u24 v0, s13, v1, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s14
; GFX7-NEXT:    v_mad_u32_u24 v0, s15, v1, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s16
; GFX7-NEXT:    v_mad_u32_u24 v0, s17, v1, v0
; GFX7-NEXT:    v_mov_b32_e32 v1, s18
; GFX7-NEXT:    v_mad_u32_u24 v0, s19, v1, v0
; GFX7-NEXT:    buffer_store_dword v0, off, s[0:3], 0
; GFX7-NEXT:    s_endpgm
;
; GFX8-LABEL: udot8_variant1:
; GFX8:       ; %bb.0: ; %entry
; GFX8-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX8-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    s_load_dword s2, s[4:5], 0x0
; GFX8-NEXT:    s_load_dword s3, s[6:7], 0x0
; GFX8-NEXT:    s_load_dword s18, s[0:1], 0x0
; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
; GFX8-NEXT:    s_and_b32 s4, s2, 15
; GFX8-NEXT:    s_and_b32 s5, s3, 15
; GFX8-NEXT:    s_bfe_u32 s6, s2, 0x40004
; GFX8-NEXT:    s_bfe_u32 s8, s2, 0x40008
; GFX8-NEXT:    s_bfe_u32 s10, s2, 0x4000c
; GFX8-NEXT:    s_bfe_u32 s12, s2, 0x40010
; GFX8-NEXT:    s_bfe_u32 s14, s2, 0x40014
; GFX8-NEXT:    s_bfe_u32 s16, s2, 0x40018
; GFX8-NEXT:    s_lshr_b32 s2, s2, 28
; GFX8-NEXT:    v_mov_b32_e32 v0, s4
; GFX8-NEXT:    v_mov_b32_e32 v1, s18
; GFX8-NEXT:    v_mad_u32_u24 v0, s5, v0, v1
; GFX8-NEXT:    s_bfe_u32 s7, s3, 0x40004
; GFX8-NEXT:    s_bfe_u32 s9, s3, 0x40008
; GFX8-NEXT:    s_bfe_u32 s11, s3, 0x4000c
; GFX8-NEXT:    s_bfe_u32 s13, s3, 0x40010
; GFX8-NEXT:    s_bfe_u32 s15, s3, 0x40014
; GFX8-NEXT:    s_bfe_u32 s17, s3, 0x40018
; GFX8-NEXT:    s_lshr_b32 s3, s3, 28
; GFX8-NEXT:    v_mov_b32_e32 v1, s2
; GFX8-NEXT:    v_mad_u32_u24 v0, s3, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v1, s6
; GFX8-NEXT:    v_mad_u32_u24 v0, s7, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v1, s8
; GFX8-NEXT:    v_mad_u32_u24 v0, s9, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v1, s10
; GFX8-NEXT:    v_mad_u32_u24 v0, s11, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v1, s12
; GFX8-NEXT:    v_mad_u32_u24 v0, s13, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v1, s14
; GFX8-NEXT:    v_mad_u32_u24 v0, s15, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v1, s16
; GFX8-NEXT:    v_mad_u32_u24 v2, s17, v1, v0
; GFX8-NEXT:    v_mov_b32_e32 v0, s0
; GFX8-NEXT:    v_mov_b32_e32 v1, s1
; GFX8-NEXT:    flat_store_dword v[0:1], v2
; GFX8-NEXT:    s_endpgm
;
; GFX9-LABEL: udot8_variant1:
; GFX9:       ; %bb.0: ; %entry
; GFX9-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    s_load_dword s2, s[4:5], 0x0
; GFX9-NEXT:    s_load_dword s3, s[6:7], 0x0
; GFX9-NEXT:    s_load_dword s18, s[0:1], 0x0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    s_and_b32 s4, s2, 15
; GFX9-NEXT:    s_and_b32 s5, s3, 15
; GFX9-NEXT:    s_bfe_u32 s6, s2, 0x40004
; GFX9-NEXT:    s_bfe_u32 s8, s2, 0x40008
; GFX9-NEXT:    s_bfe_u32 s10, s2, 0x4000c
; GFX9-NEXT:    s_bfe_u32 s12, s2, 0x40010
; GFX9-NEXT:    s_bfe_u32 s14, s2, 0x40014
; GFX9-NEXT:    s_bfe_u32 s16, s2, 0x40018
; GFX9-NEXT:    s_lshr_b32 s2, s2, 28
; GFX9-NEXT:    v_mov_b32_e32 v0, s4
; GFX9-NEXT:    v_mov_b32_e32 v1, s18
; GFX9-NEXT:    v_mad_u32_u24 v0, s5, v0, v1
; GFX9-NEXT:    s_bfe_u32 s7, s3, 0x40004
; GFX9-NEXT:    s_bfe_u32 s9, s3, 0x40008
; GFX9-NEXT:    s_bfe_u32 s11, s3, 0x4000c
; GFX9-NEXT:    s_bfe_u32 s13, s3, 0x40010
; GFX9-NEXT:    s_bfe_u32 s15, s3, 0x40014
; GFX9-NEXT:    s_bfe_u32 s17, s3, 0x40018
; GFX9-NEXT:    s_lshr_b32 s3, s3, 28
; GFX9-NEXT:    v_mov_b32_e32 v1, s2
; GFX9-NEXT:    v_mad_u32_u24 v0, s3, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, s6
; GFX9-NEXT:    v_mad_u32_u24 v0, s7, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, s8
; GFX9-NEXT:    v_mad_u32_u24 v0, s9, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, s10
; GFX9-NEXT:    v_mad_u32_u24 v0, s11, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, s12
; GFX9-NEXT:    v_mad_u32_u24 v0, s13, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, s14
; GFX9-NEXT:    v_mad_u32_u24 v0, s15, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, s16
; GFX9-NEXT:    v_mad_u32_u24 v2, s17, v1, v0
; GFX9-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-NEXT:    global_store_dword v[0:1], v2, off
; GFX9-NEXT:    s_endpgm
;
; GFX9-DL-LABEL: udot8_variant1:
; GFX9-DL:       ; %bb.0: ; %entry
; GFX9-DL-NEXT:    s_load_dwordx4 s[4:7], s[0:1], 0x24
; GFX9-DL-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    s_load_dword s2, s[4:5], 0x0
; GFX9-DL-NEXT:    s_load_dword s3, s[0:1], 0x0
; GFX9-DL-NEXT:    s_load_dword s4, s[6:7], 0x0
; GFX9-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-DL-NEXT:    v_mov_b32_e32 v0, s2
; GFX9-DL-NEXT:    v_mov_b32_e32 v1, s3
; GFX9-DL-NEXT:    v_dot8_u32_u4 v2, s4, v0, v1
; GFX9-DL-NEXT:    v_mov_b32_e32 v0, s0
; GFX9-DL-NEXT:    v_mov_b32_e32 v1, s1
; GFX9-DL-NEXT:    global_store_dword v[0:1], v2, off
; GFX9-DL-NEXT:    s_endpgm
;
; GFX10-DL-LABEL: udot8_variant1:
; GFX10-DL:       ; %bb.0: ; %entry
; GFX10-DL-NEXT:    s_clause 0x1
; GFX10-DL-NEXT:    s_load_dwordx2 s[4:5], s[0:1], 0x34
; GFX10-DL-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
; GFX10-DL-NEXT:    ; implicit-def: $vcc_hi
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    s_load_dword s6, s[4:5], 0x0
; GFX10-DL-NEXT:    s_load_dword s0, s[0:1], 0x0
; GFX10-DL-NEXT:    s_load_dword s1, s[2:3], 0x0
; GFX10-DL-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-DL-NEXT:    v_mov_b32_e32 v0, s6
; GFX10-DL-NEXT:    v_dot8_u32_u4 v2, s1, s0, v0
; GFX10-DL-NEXT:    v_mov_b32_e32 v0, s4
; GFX10-DL-NEXT:    v_mov_b32_e32 v1, s5
; GFX10-DL-NEXT:    global_store_dword v[0:1], v2, off
; GFX10-DL-NEXT:    s_endpgm
                                          i32 addrspace(1)* %v2addr,
                                          i32 addrspace(1)* %dst) {
entry:
  %v1 = load i32, i32 addrspace(1)* %v1addr, align 4
  %v2 = load i32, i32 addrspace(1)* %v2addr, align 4
  %and = and i32 %v1, 15
  %and1 = and i32 %v2, 15
  %mul1 = mul nuw nsw i32 %and1, %and

  %shr = lshr i32 %v1, 4
  %and2 = and i32 %shr, 15
  %shr3 = lshr i32 %v2, 4
  %and4 = and i32 %shr3, 15
  %mul2 = mul nuw nsw i32 %and4, %and2

  %shr6 = lshr i32 %v1, 8
  %and7 = and i32 %shr6, 15
  %shr8 = lshr i32 %v2, 8
  %and9 = and i32 %shr8, 15
  %mul3 = mul nuw nsw i32 %and9, %and7

  %shr12 = lshr i32 %v1, 12
  %and13 = and i32 %shr12, 15
  %shr14 = lshr i32 %v2, 12
  %and15 = and i32 %shr14, 15
  %mul4 = mul nuw nsw i32 %and15, %and13

  %shr18 = lshr i32 %v1, 16
  %and19 = and i32 %shr18, 15
  %shr20 = lshr i32 %v2, 16
  %and21 = and i32 %shr20, 15
  %mul5 = mul nuw nsw i32 %and21, %and19

  %shr24 = lshr i32 %v1, 20
  %and25 = and i32 %shr24, 15
  %shr26 = lshr i32 %v2, 20
  %and27 = and i32 %shr26, 15
  %mul6 = mul nuw nsw i32 %and27, %and25

  %shr30 = lshr i32 %v1, 24
  %and31 = and i32 %shr30, 15
  %shr32 = lshr i32 %v2, 24
  %and33 = and i32 %shr32, 15
  %mul7 = mul nuw nsw i32 %and33, %and31

  %shr36 = lshr i32 %v1, 28
  %shr37 = lshr i32 %v2, 28
  %mul8 = mul nuw nsw i32 %shr37, %shr36
  %acc = load i32, i32 addrspace(1)* %dst, align 4

  %add1 = add i32 %mul1, %acc
  %add2 = add i32 %add1, %mul8
  %add3 = add i32 %add2, %mul2
  %add4 = add i32 %add3, %mul3
  %add5 = add i32 %add4, %mul4
  %add6 = add i32 %add5, %mul5
  %add7 = add i32 %add6, %mul6
  %add8 = add i32 %add7, %mul7
  store i32 %add8, i32 addrspace(1)* %dst, align 4
  ret void
}