; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck -check-prefixes=VI %s
; RUN: llc -mtriple=amdgcn -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX9 %s
; RUN: llc -mtriple=amdgcn -mcpu=gfx1010 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX10 %s
; RUN: llc -mtriple=amdgcn -mcpu=gfx1100 -mattr=+real-true16 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX11,GFX11-TRUE16 %s
; RUN: llc -mtriple=amdgcn -mcpu=gfx1100 -mattr=-real-true16 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX11,GFX11-FAKE16 %s

; FIXME: Need to handle non-uniform case for function below (load without gep).
; FIXME: VI or should be unnecessary
define amdgpu_kernel void @v_test_add_v2i16(ptr addrspace(1) %out, ptr addrspace(1) %in0, ptr addrspace(1) %in1) #1 {
; VI-LABEL: v_test_add_v2i16:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    s_load_dwordx2 s[4:5], s[4:5], 0x34
; VI-NEXT:    v_lshlrev_b32_e32 v2, 2, v0
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    v_mov_b32_e32 v1, s3
; VI-NEXT:    v_add_u32_e32 v0, vcc, s2, v2
; VI-NEXT:    v_addc_u32_e32 v1, vcc, 0, v1, vcc
; VI-NEXT:    v_mov_b32_e32 v3, s5
; VI-NEXT:    v_add_u32_e32 v2, vcc, s4, v2
; VI-NEXT:    v_addc_u32_e32 v3, vcc, 0, v3, vcc
; VI-NEXT:    flat_load_dword v4, v[0:1] glc
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    flat_load_dword v2, v[2:3] glc
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    v_mov_b32_e32 v0, s0
; VI-NEXT:    v_mov_b32_e32 v1, s1
; VI-NEXT:    v_add_u16_e32 v3, v4, v2
; VI-NEXT:    v_add_u16_sdwa v2, v4, v2 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
; VI-NEXT:    v_or_b32_e32 v2, v3, v2
; VI-NEXT:    flat_store_dword v[0:1], v2
; VI-NEXT:    s_endpgm
;
; GFX9-LABEL: v_test_add_v2i16:
; GFX9:       ; %bb.0:
; GFX9-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX9-NEXT:    s_load_dwordx2 s[6:7], s[4:5], 0x34
; GFX9-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    global_load_dword v1, v0, s[2:3] glc
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    global_load_dword v2, v0, s[6:7] glc
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    v_mov_b32_e32 v0, 0
; GFX9-NEXT:    v_pk_add_u16 v1, v1, v2
; GFX9-NEXT:    global_store_dword v0, v1, s[0:1]
; GFX9-NEXT:    s_endpgm
;
; GFX10-LABEL: v_test_add_v2i16:
; GFX10:       ; %bb.0:
; GFX10-NEXT:    s_clause 0x1
; GFX10-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX10-NEXT:    s_load_dwordx2 s[6:7], s[4:5], 0x34
; GFX10-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-NEXT:    global_load_dword v1, v0, s[2:3] glc dlc
; GFX10-NEXT:    s_waitcnt vmcnt(0)
; GFX10-NEXT:    global_load_dword v2, v0, s[6:7] glc dlc
; GFX10-NEXT:    s_waitcnt vmcnt(0)
; GFX10-NEXT:    v_mov_b32_e32 v0, 0
; GFX10-NEXT:    v_pk_add_u16 v1, v1, v2
; GFX10-NEXT:    global_store_dword v0, v1, s[0:1]
; GFX10-NEXT:    s_endpgm
;
; GFX11-LABEL: v_test_add_v2i16:
; GFX11:       ; %bb.0:
; GFX11-NEXT:    s_clause 0x1
; GFX11-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
; GFX11-NEXT:    s_load_b64 s[4:5], s[4:5], 0x34
; GFX11-NEXT:    v_and_b32_e32 v0, 0x3ff, v0
; GFX11-NEXT:    v_mov_b32_e32 v2, 0
; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_2)
; GFX11-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX11-NEXT:    s_waitcnt lgkmcnt(0)
; GFX11-NEXT:    global_load_b32 v1, v0, s[2:3] glc dlc
; GFX11-NEXT:    s_waitcnt vmcnt(0)
; GFX11-NEXT:    global_load_b32 v0, v0, s[4:5] glc dlc
; GFX11-NEXT:    s_waitcnt vmcnt(0)
; GFX11-NEXT:    v_pk_add_u16 v0, v1, v0
; GFX11-NEXT:    global_store_b32 v2, v0, s[0:1]
; GFX11-NEXT:    s_endpgm
  %tid = call i32 @llvm.amdgcn.workitem.id.x()
  %gep.out = getelementptr inbounds <2 x i16>, ptr addrspace(1) %out, i32 %tid
  %gep.in0 = getelementptr inbounds <2 x i16>, ptr addrspace(1) %in0, i32 %tid
  %gep.in1 = getelementptr inbounds <2 x i16>, ptr addrspace(1) %in1, i32 %tid
  %a = load volatile <2 x i16>, ptr addrspace(1) %gep.in0
  %b = load volatile <2 x i16>, ptr addrspace(1) %gep.in1
  %add = add <2 x i16> %a, %b
  store <2 x i16> %add, ptr addrspace(1) %out
  ret void
}

define amdgpu_kernel void @s_test_add_v2i16(ptr addrspace(1) %out, ptr addrspace(4) %in0, ptr addrspace(4) %in1) #1 {
; VI-LABEL: s_test_add_v2i16:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    s_load_dwordx2 s[4:5], s[4:5], 0x34
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    s_load_dword s2, s[2:3], 0x0
; VI-NEXT:    s_load_dword s3, s[4:5], 0x0
; VI-NEXT:    v_mov_b32_e32 v0, s0
; VI-NEXT:    v_mov_b32_e32 v1, s1
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    s_lshr_b32 s0, s2, 16
; VI-NEXT:    s_lshr_b32 s1, s3, 16
; VI-NEXT:    s_add_i32 s2, s2, s3
; VI-NEXT:    s_add_i32 s0, s0, s1
; VI-NEXT:    s_and_b32 s1, s2, 0xffff
; VI-NEXT:    s_lshl_b32 s0, s0, 16
; VI-NEXT:    s_or_b32 s0, s1, s0
; VI-NEXT:    v_mov_b32_e32 v2, s0
; VI-NEXT:    flat_store_dword v[0:1], v2
; VI-NEXT:    s_endpgm
;
; GFX9-LABEL: s_test_add_v2i16:
; GFX9:       ; %bb.0:
; GFX9-NEXT:    s_load_dwordx2 s[6:7], s[4:5], 0x34
; GFX9-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX9-NEXT:    v_mov_b32_e32 v0, 0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    s_load_dword s4, s[6:7], 0x0
; GFX9-NEXT:    s_load_dword s5, s[2:3], 0x0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    v_mov_b32_e32 v1, s4
; GFX9-NEXT:    v_pk_add_u16 v1, s5, v1
; GFX9-NEXT:    global_store_dword v0, v1, s[0:1]
; GFX9-NEXT:    s_endpgm
;
; GFX10-LABEL: s_test_add_v2i16:
; GFX10:       ; %bb.0:
; GFX10-NEXT:    s_clause 0x1
; GFX10-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX10-NEXT:    s_load_dwordx2 s[6:7], s[4:5], 0x34
; GFX10-NEXT:    v_mov_b32_e32 v0, 0
; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-NEXT:    s_load_dword s4, s[2:3], 0x0
; GFX10-NEXT:    s_load_dword s5, s[6:7], 0x0
; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-NEXT:    v_pk_add_u16 v1, s4, s5
; GFX10-NEXT:    global_store_dword v0, v1, s[0:1]
; GFX10-NEXT:    s_endpgm
;
; GFX11-LABEL: s_test_add_v2i16:
; GFX11:       ; %bb.0:
; GFX11-NEXT:    s_clause 0x1
; GFX11-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
; GFX11-NEXT:    s_load_b64 s[4:5], s[4:5], 0x34
; GFX11-NEXT:    v_mov_b32_e32 v0, 0
; GFX11-NEXT:    s_waitcnt lgkmcnt(0)
; GFX11-NEXT:    s_load_b32 s2, s[2:3], 0x0
; GFX11-NEXT:    s_load_b32 s3, s[4:5], 0x0
; GFX11-NEXT:    s_waitcnt lgkmcnt(0)
; GFX11-NEXT:    v_pk_add_u16 v1, s2, s3
; GFX11-NEXT:    global_store_b32 v0, v1, s[0:1]
; GFX11-NEXT:    s_endpgm
  %a = load <2 x i16>, ptr addrspace(4) %in0
  %b = load <2 x i16>, ptr addrspace(4) %in1
  %add = add <2 x i16> %a, %b
  store <2 x i16> %add, ptr addrspace(1) %out
  ret void
}

define amdgpu_kernel void @s_test_add_self_v2i16(ptr addrspace(1) %out, ptr addrspace(4) %in0) #1 {
; VI-LABEL: s_test_add_self_v2i16:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    s_load_dword s2, s[2:3], 0x0
; VI-NEXT:    v_mov_b32_e32 v0, s0
; VI-NEXT:    v_mov_b32_e32 v1, s1
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    s_lshr_b32 s0, s2, 16
; VI-NEXT:    s_and_b32 s1, s2, 0xffff
; VI-NEXT:    s_add_i32 s1, s1, s1
; VI-NEXT:    s_add_i32 s0, s0, s0
; VI-NEXT:    s_lshl_b32 s0, s0, 16
; VI-NEXT:    s_and_b32 s1, s1, 0xffff
; VI-NEXT:    s_or_b32 s0, s1, s0
; VI-NEXT:    v_mov_b32_e32 v2, s0
; VI-NEXT:    flat_store_dword v[0:1], v2
; VI-NEXT:    s_endpgm
;
; GFX9-LABEL: s_test_add_self_v2i16:
; GFX9:       ; %bb.0:
; GFX9-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX9-NEXT:    v_mov_b32_e32 v0, 0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    s_load_dword s2, s[2:3], 0x0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    v_pk_add_u16 v1, s2, s2
; GFX9-NEXT:    global_store_dword v0, v1, s[0:1]
; GFX9-NEXT:    s_endpgm
;
; GFX10-LABEL: s_test_add_self_v2i16:
; GFX10:       ; %bb.0:
; GFX10-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX10-NEXT:    v_mov_b32_e32 v0, 0
; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-NEXT:    s_load_dword s2, s[2:3], 0x0
; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-NEXT:    v_pk_add_u16 v1, s2, s2
; GFX10-NEXT:    global_store_dword v0, v1, s[0:1]
; GFX10-NEXT:    s_endpgm
;
; GFX11-LABEL: s_test_add_self_v2i16:
; GFX11:       ; %bb.0:
; GFX11-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
; GFX11-NEXT:    v_mov_b32_e32 v0, 0
; GFX11-NEXT:    s_waitcnt lgkmcnt(0)
; GFX11-NEXT:    s_load_b32 s2, s[2:3], 0x0
; GFX11-NEXT:    s_waitcnt lgkmcnt(0)
; GFX11-NEXT:    v_pk_add_u16 v1, s2, s2
; GFX11-NEXT:    global_store_b32 v0, v1, s[0:1]
; GFX11-NEXT:    s_endpgm
  %a = load <2 x i16>, ptr addrspace(4) %in0
  %add = add <2 x i16> %a, %a
  store <2 x i16> %add, ptr addrspace(1) %out
  ret void
}

; FIXME: VI should not scalarize arg access.
define amdgpu_kernel void @s_test_add_v2i16_kernarg(ptr addrspace(1) %out, <2 x i16> %a, <2 x i16> %b) #1 {
; VI-LABEL: s_test_add_v2i16_kernarg:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    s_lshr_b32 s4, s2, 16
; VI-NEXT:    s_lshr_b32 s5, s3, 16
; VI-NEXT:    s_add_i32 s2, s2, s3
; VI-NEXT:    s_add_i32 s4, s4, s5
; VI-NEXT:    s_and_b32 s2, s2, 0xffff
; VI-NEXT:    s_lshl_b32 s3, s4, 16
; VI-NEXT:    s_or_b32 s2, s2, s3
; VI-NEXT:    v_mov_b32_e32 v0, s0
; VI-NEXT:    v_mov_b32_e32 v1, s1
; VI-NEXT:    v_mov_b32_e32 v2, s2
; VI-NEXT:    flat_store_dword v[0:1], v2
; VI-NEXT:    s_endpgm
;
; GFX9-LABEL: s_test_add_v2i16_kernarg:
; GFX9:       ; %bb.0:
; GFX9-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX9-NEXT:    v_mov_b32_e32 v0, 0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    v_mov_b32_e32 v1, s3
; GFX9-NEXT:    v_pk_add_u16 v1, s2, v1
; GFX9-NEXT:    global_store_dword v0, v1, s[0:1]
; GFX9-NEXT:    s_endpgm
;
; GFX10-LABEL: s_test_add_v2i16_kernarg:
; GFX10:       ; %bb.0:
; GFX10-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX10-NEXT:    v_mov_b32_e32 v0, 0
; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-NEXT:    v_pk_add_u16 v1, s2, s3
; GFX10-NEXT:    global_store_dword v0, v1, s[0:1]
; GFX10-NEXT:    s_endpgm
;
; GFX11-LABEL: s_test_add_v2i16_kernarg:
; GFX11:       ; %bb.0:
; GFX11-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
; GFX11-NEXT:    v_mov_b32_e32 v0, 0
; GFX11-NEXT:    s_waitcnt lgkmcnt(0)
; GFX11-NEXT:    v_pk_add_u16 v1, s2, s3
; GFX11-NEXT:    global_store_b32 v0, v1, s[0:1]
; GFX11-NEXT:    s_endpgm
  %add = add <2 x i16> %a, %b
  store <2 x i16> %add, ptr addrspace(1) %out
  ret void
}

; FIXME: Eliminate or with sdwa
define amdgpu_kernel void @v_test_add_v2i16_constant(ptr addrspace(1) %out, ptr addrspace(1) %in0) #1 {
; VI-LABEL: v_test_add_v2i16_constant:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; VI-NEXT:    v_mov_b32_e32 v3, 0x1c8
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    v_mov_b32_e32 v1, s3
; VI-NEXT:    v_add_u32_e32 v0, vcc, s2, v0
; VI-NEXT:    v_addc_u32_e32 v1, vcc, 0, v1, vcc
; VI-NEXT:    flat_load_dword v2, v[0:1] glc
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    v_mov_b32_e32 v0, s0
; VI-NEXT:    v_mov_b32_e32 v1, s1
; VI-NEXT:    v_add_u16_e32 v4, 0x7b, v2
; VI-NEXT:    v_add_u16_sdwa v2, v2, v3 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:DWORD
; VI-NEXT:    v_or_b32_e32 v2, v4, v2
; VI-NEXT:    flat_store_dword v[0:1], v2
; VI-NEXT:    s_endpgm
;
; GFX9-LABEL: v_test_add_v2i16_constant:
; GFX9:       ; %bb.0:
; GFX9-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX9-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, 0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    global_load_dword v0, v0, s[2:3] glc
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    s_mov_b32 s2, 0x1c8007b
; GFX9-NEXT:    v_pk_add_u16 v0, v0, s2
; GFX9-NEXT:    global_store_dword v1, v0, s[0:1]
; GFX9-NEXT:    s_endpgm
;
; GFX10-LABEL: v_test_add_v2i16_constant:
; GFX10:       ; %bb.0:
; GFX10-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX10-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX10-NEXT:    v_mov_b32_e32 v1, 0
; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-NEXT:    global_load_dword v0, v0, s[2:3] glc dlc
; GFX10-NEXT:    s_waitcnt vmcnt(0)
; GFX10-NEXT:    v_pk_add_u16 v0, 0x1c8007b, v0
; GFX10-NEXT:    global_store_dword v1, v0, s[0:1]
; GFX10-NEXT:    s_endpgm
;
; GFX11-LABEL: v_test_add_v2i16_constant:
; GFX11:       ; %bb.0:
; GFX11-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
; GFX11-NEXT:    v_dual_mov_b32 v1, 0 :: v_dual_and_b32 v0, 0x3ff, v0
; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_1)
; GFX11-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX11-NEXT:    s_waitcnt lgkmcnt(0)
; GFX11-NEXT:    global_load_b32 v0, v0, s[2:3] glc dlc
; GFX11-NEXT:    s_waitcnt vmcnt(0)
; GFX11-NEXT:    v_pk_add_u16 v0, 0x1c8007b, v0
; GFX11-NEXT:    global_store_b32 v1, v0, s[0:1]
; GFX11-NEXT:    s_endpgm
  %tid = call i32 @llvm.amdgcn.workitem.id.x()
  %gep.out = getelementptr inbounds <2 x i16>, ptr addrspace(1) %out, i32 %tid
  %gep.in0 = getelementptr inbounds <2 x i16>, ptr addrspace(1) %in0, i32 %tid
  %a = load volatile <2 x i16>, ptr addrspace(1) %gep.in0
  %add = add <2 x i16> %a, <i16 123, i16 456>
  store <2 x i16> %add, ptr addrspace(1) %out
  ret void
}

; FIXME: Need to handle non-uniform case for function below (load without gep).
define amdgpu_kernel void @v_test_add_v2i16_neg_constant(ptr addrspace(1) %out, ptr addrspace(1) %in0) #1 {
; VI-LABEL: v_test_add_v2i16_neg_constant:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; VI-NEXT:    v_mov_b32_e32 v3, 0xfffffc21
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    v_mov_b32_e32 v1, s3
; VI-NEXT:    v_add_u32_e32 v0, vcc, s2, v0
; VI-NEXT:    v_addc_u32_e32 v1, vcc, 0, v1, vcc
; VI-NEXT:    flat_load_dword v2, v[0:1] glc
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    v_mov_b32_e32 v0, s0
; VI-NEXT:    v_mov_b32_e32 v1, s1
; VI-NEXT:    v_add_u16_e32 v4, 0xfcb3, v2
; VI-NEXT:    v_add_u16_sdwa v2, v2, v3 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:DWORD
; VI-NEXT:    v_or_b32_e32 v2, v4, v2
; VI-NEXT:    flat_store_dword v[0:1], v2
; VI-NEXT:    s_endpgm
;
; GFX9-LABEL: v_test_add_v2i16_neg_constant:
; GFX9:       ; %bb.0:
; GFX9-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX9-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, 0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    global_load_dword v0, v0, s[2:3] glc
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    s_mov_b32 s2, 0xfc21fcb3
; GFX9-NEXT:    v_pk_add_u16 v0, v0, s2
; GFX9-NEXT:    global_store_dword v1, v0, s[0:1]
; GFX9-NEXT:    s_endpgm
;
; GFX10-LABEL: v_test_add_v2i16_neg_constant:
; GFX10:       ; %bb.0:
; GFX10-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX10-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX10-NEXT:    v_mov_b32_e32 v1, 0
; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-NEXT:    global_load_dword v0, v0, s[2:3] glc dlc
; GFX10-NEXT:    s_waitcnt vmcnt(0)
; GFX10-NEXT:    v_pk_add_u16 v0, 0xfc21fcb3, v0
; GFX10-NEXT:    global_store_dword v1, v0, s[0:1]
; GFX10-NEXT:    s_endpgm
;
; GFX11-LABEL: v_test_add_v2i16_neg_constant:
; GFX11:       ; %bb.0:
; GFX11-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
; GFX11-NEXT:    v_dual_mov_b32 v1, 0 :: v_dual_and_b32 v0, 0x3ff, v0
; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_1)
; GFX11-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX11-NEXT:    s_waitcnt lgkmcnt(0)
; GFX11-NEXT:    global_load_b32 v0, v0, s[2:3] glc dlc
; GFX11-NEXT:    s_waitcnt vmcnt(0)
; GFX11-NEXT:    v_pk_add_u16 v0, 0xfc21fcb3, v0
; GFX11-NEXT:    global_store_b32 v1, v0, s[0:1]
; GFX11-NEXT:    s_endpgm
  %tid = call i32 @llvm.amdgcn.workitem.id.x()
  %gep.out = getelementptr inbounds <2 x i16>, ptr addrspace(1) %out, i32 %tid
  %gep.in0 = getelementptr inbounds <2 x i16>, ptr addrspace(1) %in0, i32 %tid
  %a = load volatile <2 x i16>, ptr addrspace(1) %gep.in0
  %add = add <2 x i16> %a, <i16 -845, i16 -991>
  store <2 x i16> %add, ptr addrspace(1) %out
  ret void
}

define amdgpu_kernel void @v_test_add_v2i16_inline_neg1(ptr addrspace(1) %out, ptr addrspace(1) %in0) #1 {
; VI-LABEL: v_test_add_v2i16_inline_neg1:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; VI-NEXT:    v_mov_b32_e32 v3, -1
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    v_mov_b32_e32 v1, s3
; VI-NEXT:    v_add_u32_e32 v0, vcc, s2, v0
; VI-NEXT:    v_addc_u32_e32 v1, vcc, 0, v1, vcc
; VI-NEXT:    flat_load_dword v2, v[0:1] glc
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    v_mov_b32_e32 v0, s0
; VI-NEXT:    v_mov_b32_e32 v1, s1
; VI-NEXT:    v_add_u16_e32 v4, -1, v2
; VI-NEXT:    v_add_u16_sdwa v2, v2, v3 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:DWORD
; VI-NEXT:    v_or_b32_e32 v2, v4, v2
; VI-NEXT:    flat_store_dword v[0:1], v2
; VI-NEXT:    s_endpgm
;
; GFX9-LABEL: v_test_add_v2i16_inline_neg1:
; GFX9:       ; %bb.0:
; GFX9-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX9-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, 0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    global_load_dword v0, v0, s[2:3] glc
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    v_pk_add_u16 v0, v0, -1
; GFX9-NEXT:    global_store_dword v1, v0, s[0:1]
; GFX9-NEXT:    s_endpgm
;
; GFX10-LABEL: v_test_add_v2i16_inline_neg1:
; GFX10:       ; %bb.0:
; GFX10-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX10-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX10-NEXT:    v_mov_b32_e32 v1, 0
; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-NEXT:    global_load_dword v0, v0, s[2:3] glc dlc
; GFX10-NEXT:    s_waitcnt vmcnt(0)
; GFX10-NEXT:    v_pk_add_u16 v0, v0, -1
; GFX10-NEXT:    global_store_dword v1, v0, s[0:1]
; GFX10-NEXT:    s_endpgm
;
; GFX11-LABEL: v_test_add_v2i16_inline_neg1:
; GFX11:       ; %bb.0:
; GFX11-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
; GFX11-NEXT:    v_dual_mov_b32 v1, 0 :: v_dual_and_b32 v0, 0x3ff, v0
; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_1)
; GFX11-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX11-NEXT:    s_waitcnt lgkmcnt(0)
; GFX11-NEXT:    global_load_b32 v0, v0, s[2:3] glc dlc
; GFX11-NEXT:    s_waitcnt vmcnt(0)
; GFX11-NEXT:    v_pk_add_u16 v0, v0, -1
; GFX11-NEXT:    global_store_b32 v1, v0, s[0:1]
; GFX11-NEXT:    s_endpgm
  %tid = call i32 @llvm.amdgcn.workitem.id.x()
  %gep.out = getelementptr inbounds <2 x i16>, ptr addrspace(1) %out, i32 %tid
  %gep.in0 = getelementptr inbounds <2 x i16>, ptr addrspace(1) %in0, i32 %tid
  %a = load volatile <2 x i16>, ptr addrspace(1) %gep.in0
  %add = add <2 x i16> %a, <i16 -1, i16 -1>
  store <2 x i16> %add, ptr addrspace(1) %out
  ret void
}

define amdgpu_kernel void @v_test_add_v2i16_inline_lo_zero_hi(ptr addrspace(1) %out, ptr addrspace(1) %in0) #1 {
; VI-LABEL: v_test_add_v2i16_inline_lo_zero_hi:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    v_mov_b32_e32 v1, s3
; VI-NEXT:    v_add_u32_e32 v0, vcc, s2, v0
; VI-NEXT:    v_addc_u32_e32 v1, vcc, 0, v1, vcc
; VI-NEXT:    flat_load_dword v2, v[0:1] glc
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    v_mov_b32_e32 v0, s0
; VI-NEXT:    v_mov_b32_e32 v1, s1
; VI-NEXT:    v_and_b32_e32 v3, 0xffff0000, v2
; VI-NEXT:    v_add_u16_e32 v2, 32, v2
; VI-NEXT:    v_or_b32_e32 v2, v2, v3
; VI-NEXT:    flat_store_dword v[0:1], v2
; VI-NEXT:    s_endpgm
;
; GFX9-LABEL: v_test_add_v2i16_inline_lo_zero_hi:
; GFX9:       ; %bb.0:
; GFX9-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX9-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, 0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    global_load_dword v0, v0, s[2:3] glc
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    v_pk_add_u16 v0, v0, 32
; GFX9-NEXT:    global_store_dword v1, v0, s[0:1]
; GFX9-NEXT:    s_endpgm
;
; GFX10-LABEL: v_test_add_v2i16_inline_lo_zero_hi:
; GFX10:       ; %bb.0:
; GFX10-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX10-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX10-NEXT:    v_mov_b32_e32 v1, 0
; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-NEXT:    global_load_dword v0, v0, s[2:3] glc dlc
; GFX10-NEXT:    s_waitcnt vmcnt(0)
; GFX10-NEXT:    v_pk_add_u16 v0, v0, 32
; GFX10-NEXT:    global_store_dword v1, v0, s[0:1]
; GFX10-NEXT:    s_endpgm
;
; GFX11-LABEL: v_test_add_v2i16_inline_lo_zero_hi:
; GFX11:       ; %bb.0:
; GFX11-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
; GFX11-NEXT:    v_dual_mov_b32 v1, 0 :: v_dual_and_b32 v0, 0x3ff, v0
; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_1)
; GFX11-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX11-NEXT:    s_waitcnt lgkmcnt(0)
; GFX11-NEXT:    global_load_b32 v0, v0, s[2:3] glc dlc
; GFX11-NEXT:    s_waitcnt vmcnt(0)
; GFX11-NEXT:    v_pk_add_u16 v0, v0, 32
; GFX11-NEXT:    global_store_b32 v1, v0, s[0:1]
; GFX11-NEXT:    s_endpgm
  %tid = call i32 @llvm.amdgcn.workitem.id.x()
  %gep.out = getelementptr inbounds <2 x i16>, ptr addrspace(1) %out, i32 %tid
  %gep.in0 = getelementptr inbounds <2 x i16>, ptr addrspace(1) %in0, i32 %tid
  %a = load volatile <2 x i16>, ptr addrspace(1) %gep.in0
  %add = add <2 x i16> %a, <i16 32, i16 0>
  store <2 x i16> %add, ptr addrspace(1) %out
  ret void
}

; The high element gives fp
define amdgpu_kernel void @v_test_add_v2i16_inline_fp_split(ptr addrspace(1) %out, ptr addrspace(1) %in0) #1 {
; VI-LABEL: v_test_add_v2i16_inline_fp_split:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; VI-NEXT:    v_mov_b32_e32 v3, 0x3f80
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    v_mov_b32_e32 v1, s3
; VI-NEXT:    v_add_u32_e32 v0, vcc, s2, v0
; VI-NEXT:    v_addc_u32_e32 v1, vcc, 0, v1, vcc
; VI-NEXT:    flat_load_dword v2, v[0:1] glc
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    v_mov_b32_e32 v0, s0
; VI-NEXT:    v_mov_b32_e32 v1, s1
; VI-NEXT:    v_add_u16_sdwa v3, v2, v3 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:DWORD
; VI-NEXT:    v_or_b32_sdwa v2, v2, v3 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_0 src1_sel:DWORD
; VI-NEXT:    flat_store_dword v[0:1], v2
; VI-NEXT:    s_endpgm
;
; GFX9-LABEL: v_test_add_v2i16_inline_fp_split:
; GFX9:       ; %bb.0:
; GFX9-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX9-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, 0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    global_load_dword v0, v0, s[2:3] glc
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    v_pk_add_u16 v0, v0, 1.0
; GFX9-NEXT:    global_store_dword v1, v0, s[0:1]
; GFX9-NEXT:    s_endpgm
;
; GFX10-LABEL: v_test_add_v2i16_inline_fp_split:
; GFX10:       ; %bb.0:
; GFX10-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX10-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX10-NEXT:    v_mov_b32_e32 v1, 0
; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-NEXT:    global_load_dword v0, v0, s[2:3] glc dlc
; GFX10-NEXT:    s_waitcnt vmcnt(0)
; GFX10-NEXT:    v_pk_add_u16 v0, v0, 1.0
; GFX10-NEXT:    global_store_dword v1, v0, s[0:1]
; GFX10-NEXT:    s_endpgm
;
; GFX11-LABEL: v_test_add_v2i16_inline_fp_split:
; GFX11:       ; %bb.0:
; GFX11-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
; GFX11-NEXT:    v_dual_mov_b32 v1, 0 :: v_dual_and_b32 v0, 0x3ff, v0
; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_1)
; GFX11-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX11-NEXT:    s_waitcnt lgkmcnt(0)
; GFX11-NEXT:    global_load_b32 v0, v0, s[2:3] glc dlc
; GFX11-NEXT:    s_waitcnt vmcnt(0)
; GFX11-NEXT:    v_pk_add_u16 v0, v0, 1.0
; GFX11-NEXT:    global_store_b32 v1, v0, s[0:1]
; GFX11-NEXT:    s_endpgm
  %tid = call i32 @llvm.amdgcn.workitem.id.x()
  %gep.out = getelementptr inbounds <2 x i16>, ptr addrspace(1) %out, i32 %tid
  %gep.in0 = getelementptr inbounds <2 x i16>, ptr addrspace(1) %in0, i32 %tid
  %a = load volatile <2 x i16>, ptr addrspace(1) %gep.in0
  %add = add <2 x i16> %a, <i16 0, i16 16256>
  store <2 x i16> %add, ptr addrspace(1) %out
  ret void
}

; FIXME: Need to handle non-uniform case for function below (load without gep).
define amdgpu_kernel void @v_test_add_v2i16_zext_to_v2i32(ptr addrspace(1) %out, ptr addrspace(1) %in0, ptr addrspace(1) %in1) #1 {
; VI-LABEL: v_test_add_v2i16_zext_to_v2i32:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    s_load_dwordx2 s[4:5], s[4:5], 0x34
; VI-NEXT:    v_lshlrev_b32_e32 v2, 2, v0
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    v_mov_b32_e32 v1, s3
; VI-NEXT:    v_add_u32_e32 v0, vcc, s2, v2
; VI-NEXT:    v_addc_u32_e32 v1, vcc, 0, v1, vcc
; VI-NEXT:    v_mov_b32_e32 v3, s5
; VI-NEXT:    v_add_u32_e32 v2, vcc, s4, v2
; VI-NEXT:    v_addc_u32_e32 v3, vcc, 0, v3, vcc
; VI-NEXT:    flat_load_dword v4, v[0:1] glc
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    flat_load_dword v3, v[2:3] glc
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    v_mov_b32_e32 v0, s0
; VI-NEXT:    v_mov_b32_e32 v1, s1
; VI-NEXT:    v_add_u16_e32 v2, v4, v3
; VI-NEXT:    v_add_u16_sdwa v3, v4, v3 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
; VI-NEXT:    flat_store_dwordx2 v[0:1], v[2:3]
; VI-NEXT:    s_endpgm
;
; GFX9-LABEL: v_test_add_v2i16_zext_to_v2i32:
; GFX9:       ; %bb.0:
; GFX9-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX9-NEXT:    s_load_dwordx2 s[6:7], s[4:5], 0x34
; GFX9-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX9-NEXT:    v_mov_b32_e32 v3, 0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    global_load_dword v1, v0, s[2:3] glc
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    global_load_dword v2, v0, s[6:7] glc
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    v_pk_add_u16 v0, v1, v2
; GFX9-NEXT:    v_lshrrev_b32_e32 v1, 16, v0
; GFX9-NEXT:    v_and_b32_e32 v0, 0xffff, v0
; GFX9-NEXT:    global_store_dwordx2 v3, v[0:1], s[0:1]
; GFX9-NEXT:    s_endpgm
;
; GFX10-LABEL: v_test_add_v2i16_zext_to_v2i32:
; GFX10:       ; %bb.0:
; GFX10-NEXT:    s_clause 0x1
; GFX10-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX10-NEXT:    s_load_dwordx2 s[6:7], s[4:5], 0x34
; GFX10-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-NEXT:    global_load_dword v1, v0, s[2:3] glc dlc
; GFX10-NEXT:    s_waitcnt vmcnt(0)
; GFX10-NEXT:    global_load_dword v2, v0, s[6:7] glc dlc
; GFX10-NEXT:    s_waitcnt vmcnt(0)
; GFX10-NEXT:    v_pk_add_u16 v0, v1, v2
; GFX10-NEXT:    v_mov_b32_e32 v2, 0
; GFX10-NEXT:    v_lshrrev_b32_e32 v1, 16, v0
; GFX10-NEXT:    v_and_b32_e32 v0, 0xffff, v0
; GFX10-NEXT:    global_store_dwordx2 v2, v[0:1], s[0:1]
; GFX10-NEXT:    s_endpgm
;
; GFX11-TRUE16-LABEL: v_test_add_v2i16_zext_to_v2i32:
; GFX11-TRUE16:       ; %bb.0:
; GFX11-TRUE16-NEXT:    s_clause 0x1
; GFX11-TRUE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
; GFX11-TRUE16-NEXT:    s_load_b64 s[4:5], s[4:5], 0x34
; GFX11-TRUE16-NEXT:    v_dual_mov_b32 v3, 0 :: v_dual_and_b32 v0, 0x3ff, v0
; GFX11-TRUE16-NEXT:    s_delay_alu instid0(VALU_DEP_1)
; GFX11-TRUE16-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX11-TRUE16-NEXT:    s_waitcnt lgkmcnt(0)
; GFX11-TRUE16-NEXT:    global_load_b32 v1, v0, s[2:3] glc dlc
; GFX11-TRUE16-NEXT:    s_waitcnt vmcnt(0)
; GFX11-TRUE16-NEXT:    global_load_b32 v0, v0, s[4:5] glc dlc
; GFX11-TRUE16-NEXT:    s_waitcnt vmcnt(0)
; GFX11-TRUE16-NEXT:    v_pk_add_u16 v2, v1, v0
; GFX11-TRUE16-NEXT:    v_mov_b16_e32 v1.h, 0
; GFX11-TRUE16-NEXT:    s_delay_alu instid0(VALU_DEP_2)
; GFX11-TRUE16-NEXT:    v_and_b32_e32 v0, 0xffff, v2
; GFX11-TRUE16-NEXT:    v_mov_b16_e32 v1.l, v2.h
; GFX11-TRUE16-NEXT:    global_store_b64 v3, v[0:1], s[0:1]
; GFX11-TRUE16-NEXT:    s_endpgm
;
; GFX11-FAKE16-LABEL: v_test_add_v2i16_zext_to_v2i32:
; GFX11-FAKE16:       ; %bb.0:
; GFX11-FAKE16-NEXT:    s_clause 0x1
; GFX11-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
; GFX11-FAKE16-NEXT:    s_load_b64 s[4:5], s[4:5], 0x34
; GFX11-FAKE16-NEXT:    v_and_b32_e32 v0, 0x3ff, v0
; GFX11-FAKE16-NEXT:    v_mov_b32_e32 v2, 0
; GFX11-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_2)
; GFX11-FAKE16-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX11-FAKE16-NEXT:    s_waitcnt lgkmcnt(0)
; GFX11-FAKE16-NEXT:    global_load_b32 v1, v0, s[2:3] glc dlc
; GFX11-FAKE16-NEXT:    s_waitcnt vmcnt(0)
; GFX11-FAKE16-NEXT:    global_load_b32 v0, v0, s[4:5] glc dlc
; GFX11-FAKE16-NEXT:    s_waitcnt vmcnt(0)
; GFX11-FAKE16-NEXT:    v_pk_add_u16 v0, v1, v0
; GFX11-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_1)
; GFX11-FAKE16-NEXT:    v_lshrrev_b32_e32 v1, 16, v0
; GFX11-FAKE16-NEXT:    v_and_b32_e32 v0, 0xffff, v0
; GFX11-FAKE16-NEXT:    global_store_b64 v2, v[0:1], s[0:1]
; GFX11-FAKE16-NEXT:    s_endpgm
  %tid = call i32 @llvm.amdgcn.workitem.id.x()
  %gep.out = getelementptr inbounds <2 x i32>, ptr addrspace(1) %out, i32 %tid
  %gep.in0 = getelementptr inbounds <2 x i16>, ptr addrspace(1) %in0, i32 %tid
  %gep.in1 = getelementptr inbounds <2 x i16>, ptr addrspace(1) %in1, i32 %tid
  %a = load volatile <2 x i16>, ptr addrspace(1) %gep.in0
  %b = load volatile <2 x i16>, ptr addrspace(1) %gep.in1
  %add = add <2 x i16> %a, %b
  %ext = zext <2 x i16> %add to <2 x i32>
  store <2 x i32> %ext, ptr addrspace(1) %out
  ret void
}

; FIXME: Need to handle non-uniform case for function below (load without gep).
define amdgpu_kernel void @v_test_add_v2i16_zext_to_v2i64(ptr addrspace(1) %out, ptr addrspace(1) %in0, ptr addrspace(1) %in1) #1 {
; VI-LABEL: v_test_add_v2i16_zext_to_v2i64:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    s_load_dwordx2 s[4:5], s[4:5], 0x34
; VI-NEXT:    v_lshlrev_b32_e32 v2, 2, v0
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    v_mov_b32_e32 v1, s3
; VI-NEXT:    v_add_u32_e32 v0, vcc, s2, v2
; VI-NEXT:    v_addc_u32_e32 v1, vcc, 0, v1, vcc
; VI-NEXT:    v_mov_b32_e32 v3, s5
; VI-NEXT:    v_add_u32_e32 v2, vcc, s4, v2
; VI-NEXT:    v_addc_u32_e32 v3, vcc, 0, v3, vcc
; VI-NEXT:    flat_load_dword v6, v[0:1] glc
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    flat_load_dword v2, v[2:3] glc
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    v_mov_b32_e32 v1, 0
; VI-NEXT:    v_mov_b32_e32 v4, s0
; VI-NEXT:    v_mov_b32_e32 v5, s1
; VI-NEXT:    v_mov_b32_e32 v3, v1
; VI-NEXT:    v_add_u16_e32 v0, v6, v2
; VI-NEXT:    v_add_u16_sdwa v2, v6, v2 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
; VI-NEXT:    flat_store_dwordx4 v[4:5], v[0:3]
; VI-NEXT:    s_endpgm
;
; GFX9-LABEL: v_test_add_v2i16_zext_to_v2i64:
; GFX9:       ; %bb.0:
; GFX9-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX9-NEXT:    s_load_dwordx2 s[6:7], s[4:5], 0x34
; GFX9-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX9-NEXT:    v_mov_b32_e32 v1, 0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    global_load_dword v2, v0, s[2:3] glc
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    global_load_dword v3, v0, s[6:7] glc
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    v_pk_add_u16 v0, v2, v3
; GFX9-NEXT:    v_alignbit_b32 v2, 0, v0, 16
; GFX9-NEXT:    v_and_b32_e32 v0, 0xffff, v0
; GFX9-NEXT:    v_mov_b32_e32 v3, v1
; GFX9-NEXT:    global_store_dwordx4 v1, v[0:3], s[0:1]
; GFX9-NEXT:    s_endpgm
;
; GFX10-LABEL: v_test_add_v2i16_zext_to_v2i64:
; GFX10:       ; %bb.0:
; GFX10-NEXT:    s_clause 0x1
; GFX10-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX10-NEXT:    s_load_dwordx2 s[6:7], s[4:5], 0x34
; GFX10-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-NEXT:    global_load_dword v1, v0, s[2:3] glc dlc
; GFX10-NEXT:    s_waitcnt vmcnt(0)
; GFX10-NEXT:    global_load_dword v2, v0, s[6:7] glc dlc
; GFX10-NEXT:    s_waitcnt vmcnt(0)
; GFX10-NEXT:    v_pk_add_u16 v0, v1, v2
; GFX10-NEXT:    v_mov_b32_e32 v1, 0
; GFX10-NEXT:    v_alignbit_b32 v2, 0, v0, 16
; GFX10-NEXT:    v_and_b32_e32 v0, 0xffff, v0
; GFX10-NEXT:    v_mov_b32_e32 v3, v1
; GFX10-NEXT:    global_store_dwordx4 v1, v[0:3], s[0:1]
; GFX10-NEXT:    s_endpgm
;
; GFX11-TRUE16-LABEL: v_test_add_v2i16_zext_to_v2i64:
; GFX11-TRUE16:       ; %bb.0:
; GFX11-TRUE16-NEXT:    s_clause 0x1
; GFX11-TRUE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
; GFX11-TRUE16-NEXT:    s_load_b64 s[4:5], s[4:5], 0x34
; GFX11-TRUE16-NEXT:    v_and_b32_e32 v0, 0x3ff, v0
; GFX11-TRUE16-NEXT:    s_delay_alu instid0(VALU_DEP_1)
; GFX11-TRUE16-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX11-TRUE16-NEXT:    s_waitcnt lgkmcnt(0)
; GFX11-TRUE16-NEXT:    global_load_b32 v1, v0, s[2:3] glc dlc
; GFX11-TRUE16-NEXT:    s_waitcnt vmcnt(0)
; GFX11-TRUE16-NEXT:    global_load_b32 v0, v0, s[4:5] glc dlc
; GFX11-TRUE16-NEXT:    s_waitcnt vmcnt(0)
; GFX11-TRUE16-NEXT:    v_pk_add_u16 v0, v1, v0
; GFX11-TRUE16-NEXT:    v_mov_b32_e32 v1, 0
; GFX11-TRUE16-NEXT:    s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
; GFX11-TRUE16-NEXT:    v_lshrrev_b32_e32 v2, 16, v0
; GFX11-TRUE16-NEXT:    v_dual_mov_b32 v3, v1 :: v_dual_and_b32 v0, 0xffff, v0
; GFX11-TRUE16-NEXT:    s_delay_alu instid0(VALU_DEP_2)
; GFX11-TRUE16-NEXT:    v_lshl_or_b32 v2, 0, 16, v2
; GFX11-TRUE16-NEXT:    global_store_b128 v1, v[0:3], s[0:1]
; GFX11-TRUE16-NEXT:    s_endpgm
;
; GFX11-FAKE16-LABEL: v_test_add_v2i16_zext_to_v2i64:
; GFX11-FAKE16:       ; %bb.0:
; GFX11-FAKE16-NEXT:    s_clause 0x1
; GFX11-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
; GFX11-FAKE16-NEXT:    s_load_b64 s[4:5], s[4:5], 0x34
; GFX11-FAKE16-NEXT:    v_and_b32_e32 v0, 0x3ff, v0
; GFX11-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_1)
; GFX11-FAKE16-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX11-FAKE16-NEXT:    s_waitcnt lgkmcnt(0)
; GFX11-FAKE16-NEXT:    global_load_b32 v1, v0, s[2:3] glc dlc
; GFX11-FAKE16-NEXT:    s_waitcnt vmcnt(0)
; GFX11-FAKE16-NEXT:    global_load_b32 v0, v0, s[4:5] glc dlc
; GFX11-FAKE16-NEXT:    s_waitcnt vmcnt(0)
; GFX11-FAKE16-NEXT:    v_pk_add_u16 v0, v1, v0
; GFX11-FAKE16-NEXT:    v_mov_b32_e32 v1, 0
; GFX11-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
; GFX11-FAKE16-NEXT:    v_alignbit_b32 v2, 0, v0, 16
; GFX11-FAKE16-NEXT:    v_dual_mov_b32 v3, v1 :: v_dual_and_b32 v0, 0xffff, v0
; GFX11-FAKE16-NEXT:    global_store_b128 v1, v[0:3], s[0:1]
; GFX11-FAKE16-NEXT:    s_endpgm
  %tid = call i32 @llvm.amdgcn.workitem.id.x()
  %gep.out = getelementptr inbounds <2 x i64>, ptr addrspace(1) %out, i32 %tid
  %gep.in0 = getelementptr inbounds <2 x i16>, ptr addrspace(1) %in0, i32 %tid
  %gep.in1 = getelementptr inbounds <2 x i16>, ptr addrspace(1) %in1, i32 %tid
  %a = load volatile <2 x i16>, ptr addrspace(1) %gep.in0
  %b = load volatile <2 x i16>, ptr addrspace(1) %gep.in1
  %add = add <2 x i16> %a, %b
  %ext = zext <2 x i16> %add to <2 x i64>
  store <2 x i64> %ext, ptr addrspace(1) %out
  ret void
}

; FIXME: Need to handle non-uniform case for function below (load without gep).
define amdgpu_kernel void @v_test_add_v2i16_sext_to_v2i32(ptr addrspace(1) %out, ptr addrspace(1) %in0, ptr addrspace(1) %in1) #1 {
; VI-LABEL: v_test_add_v2i16_sext_to_v2i32:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    s_load_dwordx2 s[4:5], s[4:5], 0x34
; VI-NEXT:    v_lshlrev_b32_e32 v2, 2, v0
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    v_mov_b32_e32 v1, s3
; VI-NEXT:    v_add_u32_e32 v0, vcc, s2, v2
; VI-NEXT:    v_addc_u32_e32 v1, vcc, 0, v1, vcc
; VI-NEXT:    v_mov_b32_e32 v3, s5
; VI-NEXT:    v_add_u32_e32 v2, vcc, s4, v2
; VI-NEXT:    v_addc_u32_e32 v3, vcc, 0, v3, vcc
; VI-NEXT:    flat_load_dword v4, v[0:1] glc
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    flat_load_dword v2, v[2:3] glc
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    v_mov_b32_e32 v0, s0
; VI-NEXT:    v_mov_b32_e32 v1, s1
; VI-NEXT:    v_add_u16_sdwa v3, v4, v2 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
; VI-NEXT:    v_add_u16_e32 v2, v4, v2
; VI-NEXT:    v_bfe_i32 v2, v2, 0, 16
; VI-NEXT:    v_bfe_i32 v3, v3, 0, 16
; VI-NEXT:    flat_store_dwordx2 v[0:1], v[2:3]
; VI-NEXT:    s_endpgm
;
; GFX9-LABEL: v_test_add_v2i16_sext_to_v2i32:
; GFX9:       ; %bb.0:
; GFX9-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX9-NEXT:    s_load_dwordx2 s[6:7], s[4:5], 0x34
; GFX9-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX9-NEXT:    v_mov_b32_e32 v3, 0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    global_load_dword v1, v0, s[2:3] glc
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    global_load_dword v2, v0, s[6:7] glc
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    v_pk_add_u16 v0, v1, v2
; GFX9-NEXT:    v_ashrrev_i32_e32 v1, 16, v0
; GFX9-NEXT:    v_bfe_i32 v0, v0, 0, 16
; GFX9-NEXT:    global_store_dwordx2 v3, v[0:1], s[0:1]
; GFX9-NEXT:    s_endpgm
;
; GFX10-LABEL: v_test_add_v2i16_sext_to_v2i32:
; GFX10:       ; %bb.0:
; GFX10-NEXT:    s_clause 0x1
; GFX10-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX10-NEXT:    s_load_dwordx2 s[6:7], s[4:5], 0x34
; GFX10-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-NEXT:    global_load_dword v1, v0, s[2:3] glc dlc
; GFX10-NEXT:    s_waitcnt vmcnt(0)
; GFX10-NEXT:    global_load_dword v2, v0, s[6:7] glc dlc
; GFX10-NEXT:    s_waitcnt vmcnt(0)
; GFX10-NEXT:    v_pk_add_u16 v0, v1, v2
; GFX10-NEXT:    v_mov_b32_e32 v2, 0
; GFX10-NEXT:    v_ashrrev_i32_e32 v1, 16, v0
; GFX10-NEXT:    v_bfe_i32 v0, v0, 0, 16
; GFX10-NEXT:    global_store_dwordx2 v2, v[0:1], s[0:1]
; GFX10-NEXT:    s_endpgm
;
; GFX11-LABEL: v_test_add_v2i16_sext_to_v2i32:
; GFX11:       ; %bb.0:
; GFX11-NEXT:    s_clause 0x1
; GFX11-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
; GFX11-NEXT:    s_load_b64 s[4:5], s[4:5], 0x34
; GFX11-NEXT:    v_and_b32_e32 v0, 0x3ff, v0
; GFX11-NEXT:    v_mov_b32_e32 v2, 0
; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_2)
; GFX11-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX11-NEXT:    s_waitcnt lgkmcnt(0)
; GFX11-NEXT:    global_load_b32 v1, v0, s[2:3] glc dlc
; GFX11-NEXT:    s_waitcnt vmcnt(0)
; GFX11-NEXT:    global_load_b32 v0, v0, s[4:5] glc dlc
; GFX11-NEXT:    s_waitcnt vmcnt(0)
; GFX11-NEXT:    v_pk_add_u16 v0, v1, v0
; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_1)
; GFX11-NEXT:    v_ashrrev_i32_e32 v1, 16, v0
; GFX11-NEXT:    v_bfe_i32 v0, v0, 0, 16
; GFX11-NEXT:    global_store_b64 v2, v[0:1], s[0:1]
; GFX11-NEXT:    s_endpgm
  %tid = call i32 @llvm.amdgcn.workitem.id.x()
  %gep.out = getelementptr inbounds <2 x i32>, ptr addrspace(1) %out, i32 %tid
  %gep.in0 = getelementptr inbounds <2 x i16>, ptr addrspace(1) %in0, i32 %tid
  %gep.in1 = getelementptr inbounds <2 x i16>, ptr addrspace(1) %in1, i32 %tid
  %a = load volatile <2 x i16>, ptr addrspace(1) %gep.in0
  %b = load volatile <2 x i16>, ptr addrspace(1) %gep.in1
  %add = add <2 x i16> %a, %b
  %ext = sext <2 x i16> %add to <2 x i32>
  store <2 x i32> %ext, ptr addrspace(1) %out
  ret void
}

; FIXME: Need to handle non-uniform case for function below (load without gep).
define amdgpu_kernel void @v_test_add_v2i16_sext_to_v2i64(ptr addrspace(1) %out, ptr addrspace(1) %in0, ptr addrspace(1) %in1) #1 {
; VI-LABEL: v_test_add_v2i16_sext_to_v2i64:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    s_load_dwordx2 s[4:5], s[4:5], 0x34
; VI-NEXT:    v_lshlrev_b32_e32 v2, 2, v0
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    v_mov_b32_e32 v1, s3
; VI-NEXT:    v_add_u32_e32 v0, vcc, s2, v2
; VI-NEXT:    v_addc_u32_e32 v1, vcc, 0, v1, vcc
; VI-NEXT:    v_mov_b32_e32 v3, s5
; VI-NEXT:    v_add_u32_e32 v2, vcc, s4, v2
; VI-NEXT:    v_addc_u32_e32 v3, vcc, 0, v3, vcc
; VI-NEXT:    flat_load_dword v0, v[0:1]
; VI-NEXT:    flat_load_dword v1, v[2:3]
; VI-NEXT:    v_mov_b32_e32 v4, s0
; VI-NEXT:    v_mov_b32_e32 v5, s1
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    v_add_u16_sdwa v2, v0, v1 dst_sel:DWORD dst_unused:UNUSED_PAD src0_sel:WORD_1 src1_sel:WORD_1
; VI-NEXT:    v_add_u16_e32 v0, v0, v1
; VI-NEXT:    v_bfe_i32 v0, v0, 0, 16
; VI-NEXT:    v_bfe_i32 v2, v2, 0, 16
; VI-NEXT:    v_ashrrev_i32_e32 v1, 31, v0
; VI-NEXT:    v_ashrrev_i32_e32 v3, 31, v2
; VI-NEXT:    flat_store_dwordx4 v[4:5], v[0:3]
; VI-NEXT:    s_endpgm
;
; GFX9-LABEL: v_test_add_v2i16_sext_to_v2i64:
; GFX9:       ; %bb.0:
; GFX9-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX9-NEXT:    s_load_dwordx2 s[6:7], s[4:5], 0x34
; GFX9-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX9-NEXT:    v_mov_b32_e32 v4, 0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    global_load_dword v1, v0, s[2:3]
; GFX9-NEXT:    global_load_dword v2, v0, s[6:7]
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    v_pk_add_u16 v1, v1, v2
; GFX9-NEXT:    v_lshrrev_b32_e32 v2, 16, v1
; GFX9-NEXT:    v_bfe_i32 v0, v1, 0, 16
; GFX9-NEXT:    v_bfe_i32 v2, v2, 0, 16
; GFX9-NEXT:    v_ashrrev_i32_e32 v1, 31, v0
; GFX9-NEXT:    v_ashrrev_i32_e32 v3, 31, v2
; GFX9-NEXT:    global_store_dwordx4 v4, v[0:3], s[0:1]
; GFX9-NEXT:    s_endpgm
;
; GFX10-LABEL: v_test_add_v2i16_sext_to_v2i64:
; GFX10:       ; %bb.0:
; GFX10-NEXT:    s_clause 0x1
; GFX10-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX10-NEXT:    s_load_dwordx2 s[6:7], s[4:5], 0x34
; GFX10-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX10-NEXT:    v_mov_b32_e32 v4, 0
; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-NEXT:    s_clause 0x1
; GFX10-NEXT:    global_load_dword v1, v0, s[2:3]
; GFX10-NEXT:    global_load_dword v2, v0, s[6:7]
; GFX10-NEXT:    s_waitcnt vmcnt(0)
; GFX10-NEXT:    v_pk_add_u16 v0, v1, v2
; GFX10-NEXT:    v_lshrrev_b32_e32 v1, 16, v0
; GFX10-NEXT:    v_bfe_i32 v0, v0, 0, 16
; GFX10-NEXT:    v_bfe_i32 v2, v1, 0, 16
; GFX10-NEXT:    v_ashrrev_i32_e32 v1, 31, v0
; GFX10-NEXT:    v_ashrrev_i32_e32 v3, 31, v2
; GFX10-NEXT:    global_store_dwordx4 v4, v[0:3], s[0:1]
; GFX10-NEXT:    s_endpgm
;
; GFX11-TRUE16-LABEL: v_test_add_v2i16_sext_to_v2i64:
; GFX11-TRUE16:       ; %bb.0:
; GFX11-TRUE16-NEXT:    s_clause 0x1
; GFX11-TRUE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
; GFX11-TRUE16-NEXT:    s_load_b64 s[4:5], s[4:5], 0x34
; GFX11-TRUE16-NEXT:    v_and_b32_e32 v0, 0x3ff, v0
; GFX11-TRUE16-NEXT:    v_mov_b32_e32 v4, 0
; GFX11-TRUE16-NEXT:    s_delay_alu instid0(VALU_DEP_2)
; GFX11-TRUE16-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX11-TRUE16-NEXT:    s_waitcnt lgkmcnt(0)
; GFX11-TRUE16-NEXT:    s_clause 0x1
; GFX11-TRUE16-NEXT:    global_load_b32 v1, v0, s[2:3]
; GFX11-TRUE16-NEXT:    global_load_b32 v0, v0, s[4:5]
; GFX11-TRUE16-NEXT:    s_waitcnt vmcnt(0)
; GFX11-TRUE16-NEXT:    v_pk_add_u16 v0, v1, v0
; GFX11-TRUE16-NEXT:    v_mov_b16_e32 v1.h, 0
; GFX11-TRUE16-NEXT:    s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_3)
; GFX11-TRUE16-NEXT:    v_mov_b16_e32 v1.l, v0.h
; GFX11-TRUE16-NEXT:    v_bfe_i32 v0, v0, 0, 16
; GFX11-TRUE16-NEXT:    s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
; GFX11-TRUE16-NEXT:    v_bfe_i32 v2, v1, 0, 16
; GFX11-TRUE16-NEXT:    v_ashrrev_i32_e32 v1, 31, v0
; GFX11-TRUE16-NEXT:    s_delay_alu instid0(VALU_DEP_2)
; GFX11-TRUE16-NEXT:    v_ashrrev_i32_e32 v3, 31, v2
; GFX11-TRUE16-NEXT:    global_store_b128 v4, v[0:3], s[0:1]
; GFX11-TRUE16-NEXT:    s_endpgm
;
; GFX11-FAKE16-LABEL: v_test_add_v2i16_sext_to_v2i64:
; GFX11-FAKE16:       ; %bb.0:
; GFX11-FAKE16-NEXT:    s_clause 0x1
; GFX11-FAKE16-NEXT:    s_load_b128 s[0:3], s[4:5], 0x24
; GFX11-FAKE16-NEXT:    s_load_b64 s[4:5], s[4:5], 0x34
; GFX11-FAKE16-NEXT:    v_and_b32_e32 v0, 0x3ff, v0
; GFX11-FAKE16-NEXT:    v_mov_b32_e32 v4, 0
; GFX11-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_2)
; GFX11-FAKE16-NEXT:    v_lshlrev_b32_e32 v0, 2, v0
; GFX11-FAKE16-NEXT:    s_waitcnt lgkmcnt(0)
; GFX11-FAKE16-NEXT:    s_clause 0x1
; GFX11-FAKE16-NEXT:    global_load_b32 v1, v0, s[2:3]
; GFX11-FAKE16-NEXT:    global_load_b32 v0, v0, s[4:5]
; GFX11-FAKE16-NEXT:    s_waitcnt vmcnt(0)
; GFX11-FAKE16-NEXT:    v_pk_add_u16 v0, v1, v0
; GFX11-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
; GFX11-FAKE16-NEXT:    v_lshrrev_b32_e32 v1, 16, v0
; GFX11-FAKE16-NEXT:    v_bfe_i32 v0, v0, 0, 16
; GFX11-FAKE16-NEXT:    v_bfe_i32 v2, v1, 0, 16
; GFX11-FAKE16-NEXT:    s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
; GFX11-FAKE16-NEXT:    v_ashrrev_i32_e32 v1, 31, v0
; GFX11-FAKE16-NEXT:    v_ashrrev_i32_e32 v3, 31, v2
; GFX11-FAKE16-NEXT:    global_store_b128 v4, v[0:3], s[0:1]
; GFX11-FAKE16-NEXT:    s_endpgm
  %tid = call i32 @llvm.amdgcn.workitem.id.x()
  %gep.out = getelementptr inbounds <2 x i64>, ptr addrspace(1) %out, i32 %tid
  %gep.in0 = getelementptr inbounds <2 x i16>, ptr addrspace(1) %in0, i32 %tid
  %gep.in1 = getelementptr inbounds <2 x i16>, ptr addrspace(1) %in1, i32 %tid
  %a = load <2 x i16>, ptr addrspace(1) %gep.in0
  %b = load <2 x i16>, ptr addrspace(1) %gep.in1
  %add = add <2 x i16> %a, %b
  %ext = sext <2 x i16> %add to <2 x i64>
  store <2 x i64> %ext, ptr addrspace(1) %out
  ret void
}

define <2 x i16> @add_inline_imm_neg1_0(<2 x i16> %x) {
; VI-LABEL: add_inline_imm_neg1_0:
; VI:       ; %bb.0:
; VI-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; VI-NEXT:    v_and_b32_e32 v1, 0xffff0000, v0
; VI-NEXT:    v_add_u16_e32 v0, -1, v0
; VI-NEXT:    v_or_b32_e32 v0, v0, v1
; VI-NEXT:    s_setpc_b64 s[30:31]
;
; GFX9-LABEL: add_inline_imm_neg1_0:
; GFX9:       ; %bb.0:
; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; GFX9-NEXT:    v_pk_sub_u16 v0, v0, 1
; GFX9-NEXT:    s_setpc_b64 s[30:31]
;
; GFX10-LABEL: add_inline_imm_neg1_0:
; GFX10:       ; %bb.0:
; GFX10-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; GFX10-NEXT:    v_pk_sub_u16 v0, v0, 1
; GFX10-NEXT:    s_setpc_b64 s[30:31]
;
; GFX11-LABEL: add_inline_imm_neg1_0:
; GFX11:       ; %bb.0:
; GFX11-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; GFX11-NEXT:    v_pk_sub_u16 v0, v0, 1
; GFX11-NEXT:    s_setpc_b64 s[30:31]
  %y = add <2 x i16> %x, <i16 -1, i16 0>
  ret <2 x i16> %y
}

define <2 x i16> @add_inline_imm_1_0(<2 x i16> %x) {
; VI-LABEL: add_inline_imm_1_0:
; VI:       ; %bb.0:
; VI-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; VI-NEXT:    v_and_b32_e32 v1, 0xffff0000, v0
; VI-NEXT:    v_add_u16_e32 v0, 1, v0
; VI-NEXT:    v_or_b32_e32 v0, v0, v1
; VI-NEXT:    s_setpc_b64 s[30:31]
;
; GFX9-LABEL: add_inline_imm_1_0:
; GFX9:       ; %bb.0:
; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; GFX9-NEXT:    v_pk_add_u16 v0, v0, 1
; GFX9-NEXT:    s_setpc_b64 s[30:31]
;
; GFX10-LABEL: add_inline_imm_1_0:
; GFX10:       ; %bb.0:
; GFX10-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; GFX10-NEXT:    v_pk_add_u16 v0, v0, 1
; GFX10-NEXT:    s_setpc_b64 s[30:31]
;
; GFX11-LABEL: add_inline_imm_1_0:
; GFX11:       ; %bb.0:
; GFX11-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; GFX11-NEXT:    v_pk_add_u16 v0, v0, 1
; GFX11-NEXT:    s_setpc_b64 s[30:31]
  %y = add <2 x i16> %x, <i16 1, i16 0>
  ret <2 x i16> %y
}

declare i32 @llvm.amdgcn.workitem.id.x() #0

attributes #0 = { nounwind readnone }
attributes #1 = { nounwind }