; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
; RUN: llc -mtriple=amdgcn -mcpu=gfx900 < %s | FileCheck -check-prefixes=GFX9 %s
; RUN: llc -mtriple=amdgcn -mcpu=gfx1010 < %s | FileCheck -check-prefixes=GFX10 %s
; RUN: llc -mtriple=amdgcn -mcpu=gfx1100 < %s | FileCheck -check-prefixes=GFX11 %s

define amdgpu_kernel void @sub_var_var_i1(ptr addrspace(1) %out, ptr addrspace(1) %in0, ptr addrspace(1) %in1) {
; GFX9-LABEL: sub_var_var_i1:
; 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_mov_b32_e32 v0, 0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    global_load_ubyte v1, v0, s[2:3] glc
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    global_load_ubyte v2, v0, s[6:7] glc
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    v_xor_b32_e32 v1, v1, v2
; GFX9-NEXT:    v_and_b32_e32 v1, 1, v1
; GFX9-NEXT:    global_store_byte v0, v1, s[0:1]
; GFX9-NEXT:    s_endpgm
;
; GFX10-LABEL: sub_var_var_i1:
; 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:    global_load_ubyte v1, v0, s[2:3] glc dlc
; GFX10-NEXT:    s_waitcnt vmcnt(0)
; GFX10-NEXT:    global_load_ubyte v2, v0, s[6:7] glc dlc
; GFX10-NEXT:    s_waitcnt vmcnt(0)
; GFX10-NEXT:    v_xor_b32_e32 v1, v1, v2
; GFX10-NEXT:    v_and_b32_e32 v1, 1, v1
; GFX10-NEXT:    global_store_byte v0, v1, s[0:1]
; GFX10-NEXT:    s_endpgm
;
; GFX11-LABEL: sub_var_var_i1:
; 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:    global_load_u8 v1, v0, s[2:3] glc dlc
; GFX11-NEXT:    s_waitcnt vmcnt(0)
; GFX11-NEXT:    global_load_u8 v2, v0, s[4:5] glc dlc
; GFX11-NEXT:    s_waitcnt vmcnt(0)
; GFX11-NEXT:    v_xor_b32_e32 v1, v1, v2
; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_1)
; GFX11-NEXT:    v_and_b32_e32 v1, 1, v1
; GFX11-NEXT:    global_store_b8 v0, v1, s[0:1]
; GFX11-NEXT:    s_endpgm
  %a = load volatile i1, ptr addrspace(1) %in0
  %b = load volatile i1, ptr addrspace(1) %in1
  %sub = sub i1 %a, %b
  store i1 %sub, ptr addrspace(1) %out
  ret void
}

define amdgpu_kernel void @sub_var_imm_i1(ptr addrspace(1) %out, ptr addrspace(1) %in) {
; GFX9-LABEL: sub_var_imm_i1:
; 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:    global_load_ubyte v1, v0, s[2:3] glc
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    v_and_b32_e32 v1, 1, v1
; GFX9-NEXT:    v_cmp_eq_u32_e32 vcc, 1, v1
; GFX9-NEXT:    s_xor_b64 s[2:3], vcc, -1
; GFX9-NEXT:    v_cndmask_b32_e64 v1, 0, 1, s[2:3]
; GFX9-NEXT:    global_store_byte v0, v1, s[0:1]
; GFX9-NEXT:    s_endpgm
;
; GFX10-LABEL: sub_var_imm_i1:
; 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:    global_load_ubyte v1, v0, s[2:3] glc dlc
; GFX10-NEXT:    s_waitcnt vmcnt(0)
; GFX10-NEXT:    v_and_b32_e32 v1, 1, v1
; GFX10-NEXT:    v_cmp_eq_u32_e32 vcc_lo, 1, v1
; GFX10-NEXT:    s_xor_b32 s2, vcc_lo, -1
; GFX10-NEXT:    v_cndmask_b32_e64 v1, 0, 1, s2
; GFX10-NEXT:    global_store_byte v0, v1, s[0:1]
; GFX10-NEXT:    s_endpgm
;
; GFX11-LABEL: sub_var_imm_i1:
; 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:    global_load_u8 v1, v0, s[2:3] glc dlc
; GFX11-NEXT:    s_waitcnt vmcnt(0)
; GFX11-NEXT:    v_and_b32_e32 v1, 1, v1
; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(SALU_CYCLE_1)
; GFX11-NEXT:    v_cmp_eq_u32_e32 vcc_lo, 1, v1
; GFX11-NEXT:    s_xor_b32 s2, vcc_lo, -1
; GFX11-NEXT:    v_cndmask_b32_e64 v1, 0, 1, s2
; GFX11-NEXT:    global_store_b8 v0, v1, s[0:1]
; GFX11-NEXT:    s_endpgm
  %a = load volatile i1, ptr addrspace(1) %in
  %sub = sub i1 %a, 1
  store i1 %sub, ptr addrspace(1) %out
  ret void
}

define amdgpu_kernel void @sub_i1_cf(ptr addrspace(1) %out, ptr addrspace(1) %a, ptr addrspace(1) %b) {
; GFX9-LABEL: sub_i1_cf:
; GFX9:       ; %bb.0: ; %entry
; GFX9-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; GFX9-NEXT:    s_load_dwordx2 s[8:9], s[4:5], 0x34
; GFX9-NEXT:    v_cmp_lt_u32_e32 vcc, 15, v0
; GFX9-NEXT:    ; implicit-def: $sgpr4_sgpr5
; GFX9-NEXT:    s_and_saveexec_b64 s[6:7], vcc
; GFX9-NEXT:    s_xor_b64 s[6:7], exec, s[6:7]
; GFX9-NEXT:    s_cbranch_execz .LBB2_2
; GFX9-NEXT:  ; %bb.1: ; %else
; GFX9-NEXT:    v_mov_b32_e32 v0, 0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    global_load_ubyte v0, v0, s[8:9] glc
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    v_and_b32_e32 v0, 1, v0
; GFX9-NEXT:    v_cmp_eq_u32_e64 s[4:5], 1, v0
; GFX9-NEXT:  .LBB2_2: ; %Flow
; GFX9-NEXT:    s_andn2_saveexec_b64 s[6:7], s[6:7]
; GFX9-NEXT:    s_cbranch_execz .LBB2_4
; GFX9-NEXT:  ; %bb.3: ; %if
; GFX9-NEXT:    v_mov_b32_e32 v0, 0
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    global_load_ubyte v0, v0, s[2:3] glc
; GFX9-NEXT:    s_waitcnt vmcnt(0)
; GFX9-NEXT:    s_andn2_b64 s[2:3], s[4:5], exec
; GFX9-NEXT:    v_and_b32_e32 v0, 1, v0
; GFX9-NEXT:    v_cmp_eq_u32_e32 vcc, 1, v0
; GFX9-NEXT:    s_and_b64 s[4:5], vcc, exec
; GFX9-NEXT:    s_or_b64 s[4:5], s[2:3], s[4:5]
; GFX9-NEXT:  .LBB2_4: ; %endif
; GFX9-NEXT:    s_or_b64 exec, exec, s[6:7]
; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
; GFX9-NEXT:    s_xor_b64 s[2:3], s[4:5], -1
; GFX9-NEXT:    v_mov_b32_e32 v0, 0
; GFX9-NEXT:    v_cndmask_b32_e64 v1, 0, 1, s[2:3]
; GFX9-NEXT:    global_store_byte v0, v1, s[0:1]
; GFX9-NEXT:    s_endpgm
;
; GFX10-LABEL: sub_i1_cf:
; GFX10:       ; %bb.0: ; %entry
; 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_cmp_lt_u32_e32 vcc_lo, 15, v0
; GFX10-NEXT:    ; implicit-def: $sgpr4
; GFX10-NEXT:    s_and_saveexec_b32 s5, vcc_lo
; GFX10-NEXT:    s_xor_b32 s5, exec_lo, s5
; GFX10-NEXT:    s_cbranch_execz .LBB2_2
; GFX10-NEXT:  ; %bb.1: ; %else
; GFX10-NEXT:    v_mov_b32_e32 v0, 0
; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-NEXT:    global_load_ubyte v0, v0, s[6:7] glc dlc
; GFX10-NEXT:    s_waitcnt vmcnt(0)
; GFX10-NEXT:    v_and_b32_e32 v0, 1, v0
; GFX10-NEXT:    v_cmp_eq_u32_e64 s4, 1, v0
; GFX10-NEXT:  .LBB2_2: ; %Flow
; GFX10-NEXT:    s_andn2_saveexec_b32 s5, s5
; GFX10-NEXT:    s_cbranch_execz .LBB2_4
; GFX10-NEXT:  ; %bb.3: ; %if
; GFX10-NEXT:    v_mov_b32_e32 v0, 0
; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-NEXT:    global_load_ubyte v0, v0, s[2:3] glc dlc
; GFX10-NEXT:    s_waitcnt vmcnt(0)
; GFX10-NEXT:    s_waitcnt_depctr 0xffe3
; GFX10-NEXT:    s_andn2_b32 s2, s4, exec_lo
; GFX10-NEXT:    v_and_b32_e32 v0, 1, v0
; GFX10-NEXT:    v_cmp_eq_u32_e32 vcc_lo, 1, v0
; GFX10-NEXT:    s_and_b32 s3, vcc_lo, exec_lo
; GFX10-NEXT:    s_or_b32 s4, s2, s3
; GFX10-NEXT:  .LBB2_4: ; %endif
; GFX10-NEXT:    s_or_b32 exec_lo, exec_lo, s5
; GFX10-NEXT:    s_waitcnt lgkmcnt(0)
; GFX10-NEXT:    s_xor_b32 s2, s4, -1
; GFX10-NEXT:    v_mov_b32_e32 v0, 0
; GFX10-NEXT:    v_cndmask_b32_e64 v1, 0, 1, s2
; GFX10-NEXT:    global_store_byte v0, v1, s[0:1]
; GFX10-NEXT:    s_endpgm
;
; GFX11-LABEL: sub_i1_cf:
; GFX11:       ; %bb.0: ; %entry
; 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:    s_mov_b32 s7, exec_lo
; GFX11-NEXT:    ; implicit-def: $sgpr6
; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_1)
; GFX11-NEXT:    v_cmpx_lt_u32_e32 15, v0
; GFX11-NEXT:    s_xor_b32 s7, exec_lo, s7
; GFX11-NEXT:    s_cbranch_execz .LBB2_2
; GFX11-NEXT:  ; %bb.1: ; %else
; GFX11-NEXT:    v_mov_b32_e32 v0, 0
; GFX11-NEXT:    s_waitcnt lgkmcnt(0)
; GFX11-NEXT:    global_load_u8 v0, v0, s[4:5] glc dlc
; GFX11-NEXT:    s_waitcnt vmcnt(0)
; GFX11-NEXT:    v_and_b32_e32 v0, 1, v0
; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_1)
; GFX11-NEXT:    v_cmp_eq_u32_e64 s6, 1, v0
; GFX11-NEXT:  .LBB2_2: ; %Flow
; GFX11-NEXT:    s_waitcnt lgkmcnt(0)
; GFX11-NEXT:    s_and_not1_saveexec_b32 s4, s7
; GFX11-NEXT:    s_cbranch_execz .LBB2_4
; GFX11-NEXT:  ; %bb.3: ; %if
; GFX11-NEXT:    v_mov_b32_e32 v0, 0
; GFX11-NEXT:    global_load_u8 v0, v0, s[2:3] glc dlc
; GFX11-NEXT:    s_waitcnt vmcnt(0)
; GFX11-NEXT:    s_and_not1_b32 s2, s6, exec_lo
; GFX11-NEXT:    v_and_b32_e32 v0, 1, v0
; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(SALU_CYCLE_1)
; GFX11-NEXT:    v_cmp_eq_u32_e32 vcc_lo, 1, v0
; GFX11-NEXT:    s_and_b32 s3, vcc_lo, exec_lo
; GFX11-NEXT:    s_or_b32 s6, s2, s3
; GFX11-NEXT:  .LBB2_4: ; %endif
; GFX11-NEXT:    s_or_b32 exec_lo, exec_lo, s4
; GFX11-NEXT:    s_xor_b32 s2, s6, -1
; GFX11-NEXT:    v_mov_b32_e32 v0, 0
; GFX11-NEXT:    v_cndmask_b32_e64 v1, 0, 1, s2
; GFX11-NEXT:    global_store_b8 v0, v1, s[0:1]
; GFX11-NEXT:    s_endpgm
entry:
  %tid = call i32 @llvm.amdgcn.workitem.id.x()
  %d_cmp = icmp ult i32 %tid, 16
  br i1 %d_cmp, label %if, label %else

if:
  %0 = load volatile i1, ptr addrspace(1) %a
  br label %endif

else:
  %1 = load volatile i1, ptr addrspace(1) %b
  br label %endif

endif:
  %2 = phi i1 [%0, %if], [%1, %else]
  %3 = sub i1 %2, -1
  store i1 %3, ptr addrspace(1) %out
  ret void
}

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