; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 6
; RUN: llc -mtriple=amdgcn -mcpu=gfx900 < %s | FileCheck %s

declare i32 @llvm.ctpop.i32(i32)
declare i64 @llvm.ctpop.i64(i64)
declare i32 @llvm.amdgcn.s.quadmask.i32(i32)
declare i64 @llvm.amdgcn.s.quadmask.i64(i64)

define amdgpu_ps i32 @shl32(i32 inreg %val0, i32 inreg %val1) {
; CHECK-LABEL: shl32:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_lshl_b32 s0, s0, s1
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = shl i32 %val0, %val1
  %cmp = icmp ne i32 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @shl64(i64 inreg %val0, i64 inreg %val1) {
; CHECK-LABEL: shl64:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_lshl_b64 s[0:1], s[0:1], s2
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = shl i64 %val0, %val1
  %cmp = icmp ne i64 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @lshr32(i32 inreg %val0, i32 inreg %val1) {
; CHECK-LABEL: lshr32:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_lshr_b32 s0, s0, s1
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = lshr i32 %val0, %val1
  %cmp = icmp ne i32 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @lshr64(i64 inreg %val0, i64 inreg %val1) {
; CHECK-LABEL: lshr64:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_lshr_b64 s[0:1], s[0:1], s2
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = lshr i64 %val0, %val1
  %cmp = icmp ne i64 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @ashr32(i32 inreg %val0, i32 inreg %val1) {
; CHECK-LABEL: ashr32:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_ashr_i32 s0, s0, s1
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = ashr i32 %val0, %val1
  %cmp = icmp ne i32 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @ashr64(i64 inreg %val0, i64 inreg %val1) {
; CHECK-LABEL: ashr64:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_ashr_i64 s[0:1], s[0:1], s2
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = ashr i64 %val0, %val1
  %cmp = icmp ne i64 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @abs32(i32 inreg %val0) {
; CHECK-LABEL: abs32:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_abs_i32 s0, s0
; CHECK-NEXT:    ;;#ASMSTART
; CHECK-NEXT:    ; use s0
; CHECK-NEXT:    ;;#ASMEND
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %neg = sub i32 0, %val0
  %cond = icmp sgt i32 %val0, %neg
  %result = select i1 %cond, i32 %val0, i32 %neg
  call void asm "; use $0", "s"(i32 %result)
  %cmp = icmp ne i32 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @absdiff32(i32 inreg %val0, i32 inreg %val1) {
; CHECK-LABEL: absdiff32:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_absdiff_i32 s0, s0, s1
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %diff = sub i32 %val0, %val1
  %result = call i32 @llvm.abs.i32(i32 %diff, i1 false)
  %cmp = icmp ne i32 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @and32(i32 inreg %val0, i32 inreg %val1) {
; CHECK-LABEL: and32:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_and_b32 s0, s0, s1
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = and i32 %val0, %val1
  %cmp = icmp ne i32 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @and64(i64 inreg %val0, i64 inreg %val1) {
; CHECK-LABEL: and64:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_and_b64 s[0:1], s[0:1], s[2:3]
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = and i64 %val0, %val1
  %cmp = icmp ne i64 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @and32_clear_one_bit(i32 inreg %val0) {
; CHECK-LABEL: and32_clear_one_bit:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_and_b32 s0, s0, 0x7fffffff
; CHECK-NEXT:    ;;#ASMSTART
; CHECK-NEXT:    ; use s0
; CHECK-NEXT:    ;;#ASMEND
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = and i32 %val0, 2147483647
  call void asm "; use $0", "s"(i32 %result)
  %cmp = icmp ne i32 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @and64_clear_one_bit(i64 inreg %val0) {
; CHECK-LABEL: and64_clear_one_bit:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_bitset0_b32 s0, 31
; CHECK-NEXT:    s_cmp_lg_u64 s[0:1], 0
; CHECK-NEXT:    ;;#ASMSTART
; CHECK-NEXT:    ; use s[0:1]
; CHECK-NEXT:    ;;#ASMEND
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = and i64 %val0, -2147483649
  call void asm "; use $0", "s"(i64 %result)
  %cmp = icmp ne i64 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @or32(i32 inreg %val0, i32 inreg %val1) {
; CHECK-LABEL: or32:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_or_b32 s0, s0, s1
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = or i32 %val0, %val1
  %cmp = icmp ne i32 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @or64(i64 inreg %val0, i64 inreg %val1) {
; CHECK-LABEL: or64:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_or_b64 s[0:1], s[0:1], s[2:3]
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = or i64 %val0, %val1
  %cmp = icmp ne i64 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @xor32(i32 inreg %val0, i32 inreg %val1) {
; CHECK-LABEL: xor32:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_xor_b32 s0, s0, s1
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = xor i32 %val0, %val1
  %cmp = icmp ne i32 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @xor64(i64 inreg %val0, i64 inreg %val1) {
; CHECK-LABEL: xor64:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_xor_b64 s[0:1], s[0:1], s[2:3]
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = xor i64 %val0, %val1
  %cmp = icmp ne i64 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @nand32(i32 inreg %val0, i32 inreg %val1) {
; CHECK-LABEL: nand32:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_nand_b32 s0, s0, s1
; CHECK-NEXT:    ;;#ASMSTART
; CHECK-NEXT:    ; use s0
; CHECK-NEXT:    ;;#ASMEND
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = and i32 %val0, %val1
  %result2 = xor i32 %result, -1
  call void asm "; use $0", "s"(i32 %result2)
  %cmp = icmp ne i32 %result2, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @nand64(i64 inreg %val0, i64 inreg %val1) {
; CHECK-LABEL: nand64:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_nand_b64 s[0:1], s[0:1], s[2:3]
; CHECK-NEXT:    ;;#ASMSTART
; CHECK-NEXT:    ; use s[0:1]
; CHECK-NEXT:    ;;#ASMEND
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = and i64 %val0, %val1
  %result2 = xor i64 %result, -1
  call void asm "; use $0", "s"(i64 %result2)
  %cmp = icmp ne i64 %result2, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @nor32(i32 inreg %val0, i32 inreg %val1) {
; CHECK-LABEL: nor32:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_nor_b32 s0, s0, s1
; CHECK-NEXT:    ;;#ASMSTART
; CHECK-NEXT:    ; use s0
; CHECK-NEXT:    ;;#ASMEND
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = or i32 %val0, %val1
  %result2 = xor i32 %result, -1
  call void asm "; use $0", "s"(i32 %result2)
  %cmp = icmp ne i32 %result2, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @nor64(i64 inreg %val0, i64 inreg %val1) {
; CHECK-LABEL: nor64:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_nor_b64 s[0:1], s[0:1], s[2:3]
; CHECK-NEXT:    ;;#ASMSTART
; CHECK-NEXT:    ; use s[0:1]
; CHECK-NEXT:    ;;#ASMEND
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = or i64 %val0, %val1
  %result2 = xor i64 %result, -1
  call void asm "; use $0", "s"(i64 %result2)
  %cmp = icmp ne i64 %result2, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @xnor32(i32 inreg %val0, i32 inreg %val1) {
; CHECK-LABEL: xnor32:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_xnor_b32 s0, s0, s1
; CHECK-NEXT:    ;;#ASMSTART
; CHECK-NEXT:    ; use s0
; CHECK-NEXT:    ;;#ASMEND
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = xor i32 %val0, %val1
  %result2 = xor i32 %result, -1
  call void asm "; use $0", "s"(i32 %result2)
  %cmp = icmp ne i32 %result2, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @xnor64(i64 inreg %val0, i64 inreg %val1) {
; CHECK-LABEL: xnor64:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_xnor_b64 s[0:1], s[0:1], s[2:3]
; CHECK-NEXT:    ;;#ASMSTART
; CHECK-NEXT:    ; use s[0:1]
; CHECK-NEXT:    ;;#ASMEND
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = xor i64 %val0, %val1
  %result2 = xor i64 %result, -1
  call void asm "; use $0", "s"(i64 %result2)
  %cmp = icmp ne i64 %result2, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @andn232(i32 inreg %val0, i32 inreg %val1) {
; CHECK-LABEL: andn232:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_andn2_b32 s0, s0, s1
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %nval1 = xor i32 %val1, -1
  %result = and i32 %val0, %nval1
  %cmp = icmp ne i32 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @nandn264(i64 inreg %val0, i64 inreg %val1) {
; CHECK-LABEL: nandn264:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_andn2_b64 s[0:1], s[0:1], s[2:3]
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %nval1 = xor i64 %val1, -1
  %result = and i64 %val0, %nval1
  %cmp = icmp ne i64 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @orn232(i32 inreg %val0, i32 inreg %val1) {
; CHECK-LABEL: orn232:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_orn2_b32 s0, s0, s1
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %nval1 = xor i32 %val1, -1
  %result = or i32 %val0, %nval1
  %cmp = icmp ne i32 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @orn264(i64 inreg %val0, i64 inreg %val1) {
; CHECK-LABEL: orn264:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_orn2_b64 s[0:1], s[0:1], s[2:3]
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %nval1 = xor i64 %val1, -1
  %result = or i64 %val0, %nval1
  %cmp = icmp ne i64 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @bfe_i32(i32 inreg %val0) {
; CHECK-LABEL: bfe_i32:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_bfe_i32 s0, s0, 0x80010
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %shl = shl i32 %val0, 8
  %result = ashr i32 %shl, 24
  %cmp = icmp ne i32 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @bfe_i64(i64 inreg %val0) {
; CHECK-LABEL: bfe_i64:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_bfe_i64 s[2:3], s[0:1], 0x80000
; CHECK-NEXT:    s_and_b32 s0, s0, 0xff
; CHECK-NEXT:    s_mov_b32 s1, 0
; CHECK-NEXT:    s_cmp_lg_u64 s[0:1], 0
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ;;#ASMSTART
; CHECK-NEXT:    ; use s[2:3]
; CHECK-NEXT:    ;;#ASMEND
; CHECK-NEXT:    ; return to shader part epilog
  %shl = shl i64 %val0, 56
  %result = ashr i64 %shl, 56
  call void asm "; use $0", "s"(i64 %result)
  %cmp = icmp ne i64 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @bfe_u32(i32 inreg %val0) {
; CHECK-LABEL: bfe_u32:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_bfe_u32 s0, s0, 0x80010
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %shl = shl i32 %val0, 8
  %result = lshr i32 %shl, 24
  %cmp = icmp ne i32 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @bfe_u64(i64 inreg %val0) {
; CHECK-LABEL: bfe_u64:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_and_b32 s0, s0, 0xff
; CHECK-NEXT:    s_mov_b32 s1, 0
; CHECK-NEXT:    s_cmp_lg_u64 s[0:1], 0
; CHECK-NEXT:    ;;#ASMSTART
; CHECK-NEXT:    ; use s[0:1]
; CHECK-NEXT:    ;;#ASMEND
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %shl = shl i64 %val0, 56
  %result = lshr i64 %shl, 56
  call void asm "; use $0", "s"(i64 %result)
  %cmp = icmp ne i64 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @bcnt032(i32 inreg %val0) {
; CHECK-LABEL: bcnt032:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_bcnt1_i32_b32 s0, s0
; CHECK-NEXT:    s_sub_i32 s0, 32, s0
; CHECK-NEXT:    s_cmp_lg_u32 s0, 0
; CHECK-NEXT:    ;;#ASMSTART
; CHECK-NEXT:    ; use s0
; CHECK-NEXT:    ;;#ASMEND
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = call i32 @llvm.ctpop.i32(i32 %val0) nounwind readnone
  %result2 = sub i32 32, %result
  call void asm "; use $0", "s"(i32 %result2)
  %cmp = icmp ne i32 %result2, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @bcnt064(i64 inreg %val0) {
; CHECK-LABEL: bcnt064:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_bcnt1_i32_b64 s0, s[0:1]
; CHECK-NEXT:    s_sub_u32 s0, 64, s0
; CHECK-NEXT:    s_subb_u32 s1, 0, 0
; CHECK-NEXT:    s_cmp_lg_u64 s[0:1], 0
; CHECK-NEXT:    ;;#ASMSTART
; CHECK-NEXT:    ; use s[0:1]
; CHECK-NEXT:    ;;#ASMEND
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = call i64 @llvm.ctpop.i64(i64 %val0) nounwind readnone
  %result2 = sub i64 64, %result
  call void asm "; use $0", "s"(i64 %result2)
  %cmp = icmp ne i64 %result2, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @bcnt132(i32 inreg %val0) {
; CHECK-LABEL: bcnt132:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_bcnt1_i32_b32 s0, s0
; CHECK-NEXT:    ;;#ASMSTART
; CHECK-NEXT:    ; use s0
; CHECK-NEXT:    ;;#ASMEND
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = call i32 @llvm.ctpop.i32(i32 %val0) nounwind readnone
  call void asm "; use $0", "s"(i32 %result)
  %cmp = icmp ne i32 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @bcnt164(i64 inreg %val0) {
; CHECK-LABEL: bcnt164:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_bcnt1_i32_b64 s0, s[0:1]
; CHECK-NEXT:    s_mov_b32 s1, 0
; CHECK-NEXT:    s_cmp_lg_u64 s[0:1], 0
; CHECK-NEXT:    ;;#ASMSTART
; CHECK-NEXT:    ; use s[0:1]
; CHECK-NEXT:    ;;#ASMEND
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = call i64 @llvm.ctpop.i64(i64 %val0) nounwind readnone
  call void asm "; use $0", "s"(i64 %result)
  %cmp = icmp ne i64 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @quadmask32(i32 inreg %val0) {
; CHECK-LABEL: quadmask32:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_quadmask_b32 s0, s0
; CHECK-NEXT:    ;;#ASMSTART
; CHECK-NEXT:    ; use s0
; CHECK-NEXT:    ;;#ASMEND
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = call i32 @llvm.amdgcn.s.quadmask.i32(i32 %val0) nounwind readnone
  call void asm "; use $0", "s"(i32 %result)
  %cmp = icmp ne i32 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @quadmask64(i64 inreg %val0) {
; CHECK-LABEL: quadmask64:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_quadmask_b64 s[0:1], s[0:1]
; CHECK-NEXT:    ;;#ASMSTART
; CHECK-NEXT:    ; use s[0:1]
; CHECK-NEXT:    ;;#ASMEND
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = call i64 @llvm.amdgcn.s.quadmask.i64(i64 %val0) nounwind readnone
  call void asm "; use $0", "s"(i64 %result)
  %cmp = icmp ne i64 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @not32(i32 inreg %val0) {
; CHECK-LABEL: not32:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_not_b32 s0, s0
; CHECK-NEXT:    ;;#ASMSTART
; CHECK-NEXT:    ; use s0
; CHECK-NEXT:    ;;#ASMEND
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = xor i32 %val0, -1
  call void asm "; use $0", "s"(i32 %result)
  %cmp = icmp ne i32 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}

define amdgpu_ps i32 @not64(i64 inreg %val0) {
; CHECK-LABEL: not64:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_not_b64 s[0:1], s[0:1]
; CHECK-NEXT:    ;;#ASMSTART
; CHECK-NEXT:    ; use s[0:1]
; CHECK-NEXT:    ;;#ASMEND
; CHECK-NEXT:    s_cselect_b64 s[0:1], -1, 0
; CHECK-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
; CHECK-NEXT:    ; return to shader part epilog
  %result = xor i64 %val0, -1
  call void asm "; use $0", "s"(i64 %result)
  %cmp = icmp ne i64 %result, 0
  %zext = zext i1 %cmp to i32
  ret i32 %zext
}


; --------------------------------------------------------------------------------
; Negative tests
; --------------------------------------------------------------------------------

@1 = extern_weak dso_local addrspace(4) constant i32

define amdgpu_ps i32 @si_pc_add_rel_offset_must_not_optimize() {
; CHECK-LABEL: si_pc_add_rel_offset_must_not_optimize:
; CHECK:       ; %bb.0:
; CHECK-NEXT:    s_getpc_b64 s[0:1]
; CHECK-NEXT:    s_add_u32 s0, s0, __unnamed_1@rel32@lo+4
; CHECK-NEXT:    s_addc_u32 s1, s1, __unnamed_1@rel32@hi+12
; CHECK-NEXT:    s_cmp_lg_u64 s[0:1], 0
; CHECK-NEXT:    s_cbranch_scc0 .LBB38_2
; CHECK-NEXT:  ; %bb.1: ; %endif
; CHECK-NEXT:    s_mov_b32 s0, 1
; CHECK-NEXT:    s_branch .LBB38_3
; CHECK-NEXT:  .LBB38_2: ; %if
; CHECK-NEXT:    s_mov_b32 s0, 0
; CHECK-NEXT:    s_branch .LBB38_3
; CHECK-NEXT:  .LBB38_3:
  %cmp = icmp ne ptr addrspace(4) @1, null
  br i1 %cmp, label %endif, label %if

if:
  ret i32 0

endif:
  ret i32 1
}