; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
; RUN: llc -mtriple=amdgcn -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefix=SI %s
; RUN: llc -mtriple=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefix=VI %s

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

define amdgpu_kernel void @sextload_i1_to_i32_trunc_cmp_eq_0(ptr addrspace(1) %out, ptr addrspace(1) %in) nounwind {
; SI-LABEL: sextload_i1_to_i32_trunc_cmp_eq_0:
; SI:       ; %bb.0:
; SI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x9
; SI-NEXT:    s_mov_b32 s7, 0xf000
; SI-NEXT:    s_mov_b32 s6, -1
; SI-NEXT:    s_mov_b32 s10, s6
; SI-NEXT:    s_mov_b32 s11, s7
; SI-NEXT:    s_waitcnt lgkmcnt(0)
; SI-NEXT:    s_mov_b32 s8, s2
; SI-NEXT:    s_mov_b32 s9, s3
; SI-NEXT:    buffer_load_ubyte v0, off, s[8:11], 0
; SI-NEXT:    s_mov_b32 s4, s0
; SI-NEXT:    s_mov_b32 s5, s1
; SI-NEXT:    s_waitcnt vmcnt(0)
; SI-NEXT:    v_and_b32_e32 v0, 1, v0
; SI-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
; SI-NEXT:    v_cndmask_b32_e64 v0, 0, 1, vcc
; SI-NEXT:    buffer_store_byte v0, off, s[4:7], 0
; SI-NEXT:    s_endpgm
;
; VI-LABEL: sextload_i1_to_i32_trunc_cmp_eq_0:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    s_mov_b32 s7, 0xf000
; VI-NEXT:    s_mov_b32 s6, -1
; VI-NEXT:    s_mov_b32 s10, s6
; VI-NEXT:    s_mov_b32 s11, s7
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    s_mov_b32 s8, s2
; VI-NEXT:    s_mov_b32 s9, s3
; VI-NEXT:    buffer_load_ubyte v0, off, s[8:11], 0
; VI-NEXT:    s_mov_b32 s4, s0
; VI-NEXT:    s_mov_b32 s5, s1
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    v_and_b32_e32 v0, 1, v0
; VI-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
; VI-NEXT:    v_cndmask_b32_e64 v0, 0, 1, vcc
; VI-NEXT:    buffer_store_byte v0, off, s[4:7], 0
; VI-NEXT:    s_endpgm
  %load = load i1, ptr addrspace(1) %in
  %ext = sext i1 %load to i32
  %cmp = icmp eq i32 %ext, 0
  store i1 %cmp, ptr addrspace(1) %out
  ret void
}

; FIXME: The negate should be inverting the compare.
define amdgpu_kernel void @zextload_i1_to_i32_trunc_cmp_eq_0(ptr addrspace(1) %out, ptr addrspace(1) %in) nounwind {
; SI-LABEL: zextload_i1_to_i32_trunc_cmp_eq_0:
; SI:       ; %bb.0:
; SI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x9
; SI-NEXT:    s_mov_b32 s7, 0xf000
; SI-NEXT:    s_mov_b32 s6, -1
; SI-NEXT:    s_mov_b32 s10, s6
; SI-NEXT:    s_mov_b32 s11, s7
; SI-NEXT:    s_waitcnt lgkmcnt(0)
; SI-NEXT:    s_mov_b32 s8, s2
; SI-NEXT:    s_mov_b32 s9, s3
; SI-NEXT:    buffer_load_ubyte v0, off, s[8:11], 0
; SI-NEXT:    s_mov_b32 s4, s0
; SI-NEXT:    s_mov_b32 s5, s1
; SI-NEXT:    s_waitcnt vmcnt(0)
; SI-NEXT:    v_and_b32_e32 v0, 1, v0
; SI-NEXT:    v_cmp_eq_u32_e32 vcc, 1, v0
; SI-NEXT:    s_xor_b64 s[0:1], vcc, -1
; SI-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; SI-NEXT:    buffer_store_byte v0, off, s[4:7], 0
; SI-NEXT:    s_endpgm
;
; VI-LABEL: zextload_i1_to_i32_trunc_cmp_eq_0:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    s_mov_b32 s7, 0xf000
; VI-NEXT:    s_mov_b32 s6, -1
; VI-NEXT:    s_mov_b32 s10, s6
; VI-NEXT:    s_mov_b32 s11, s7
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    s_mov_b32 s8, s2
; VI-NEXT:    s_mov_b32 s9, s3
; VI-NEXT:    buffer_load_ubyte v0, off, s[8:11], 0
; VI-NEXT:    s_mov_b32 s4, s0
; VI-NEXT:    s_mov_b32 s5, s1
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    v_and_b32_e32 v0, 1, v0
; VI-NEXT:    v_cmp_eq_u32_e32 vcc, 1, v0
; VI-NEXT:    s_xor_b64 s[0:1], vcc, -1
; VI-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; VI-NEXT:    buffer_store_byte v0, off, s[4:7], 0
; VI-NEXT:    s_endpgm
  %load = load i1, ptr addrspace(1) %in
  %ext = zext i1 %load to i32
  %cmp = icmp eq i32 %ext, 0
  store i1 %cmp, ptr addrspace(1) %out
  ret void
}

define amdgpu_kernel void @sextload_i1_to_i32_trunc_cmp_eq_1(ptr addrspace(1) %out, ptr addrspace(1) %in) nounwind {
; SI-LABEL: sextload_i1_to_i32_trunc_cmp_eq_1:
; SI:       ; %bb.0:
; SI-NEXT:    s_load_dwordx2 s[0:1], s[4:5], 0x9
; SI-NEXT:    s_mov_b32 s3, 0xf000
; SI-NEXT:    s_mov_b32 s2, -1
; SI-NEXT:    v_mov_b32_e32 v0, 0
; SI-NEXT:    s_waitcnt lgkmcnt(0)
; SI-NEXT:    buffer_store_byte v0, off, s[0:3], 0
; SI-NEXT:    s_endpgm
;
; VI-LABEL: sextload_i1_to_i32_trunc_cmp_eq_1:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx2 s[0:1], s[4:5], 0x24
; VI-NEXT:    s_mov_b32 s3, 0xf000
; VI-NEXT:    s_mov_b32 s2, -1
; VI-NEXT:    v_mov_b32_e32 v0, 0
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    buffer_store_byte v0, off, s[0:3], 0
; VI-NEXT:    s_endpgm
  %load = load i1, ptr addrspace(1) %in
  %ext = sext i1 %load to i32
  %cmp = icmp eq i32 %ext, 1
  store i1 %cmp, ptr addrspace(1) %out
  ret void
}

define amdgpu_kernel void @zextload_i1_to_i32_trunc_cmp_eq_1(ptr addrspace(1) %out, ptr addrspace(1) %in) nounwind {
; SI-LABEL: zextload_i1_to_i32_trunc_cmp_eq_1:
; SI:       ; %bb.0:
; SI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x9
; SI-NEXT:    s_mov_b32 s7, 0xf000
; SI-NEXT:    s_mov_b32 s6, -1
; SI-NEXT:    s_mov_b32 s10, s6
; SI-NEXT:    s_mov_b32 s11, s7
; SI-NEXT:    s_waitcnt lgkmcnt(0)
; SI-NEXT:    s_mov_b32 s8, s2
; SI-NEXT:    s_mov_b32 s9, s3
; SI-NEXT:    buffer_load_ubyte v0, off, s[8:11], 0
; SI-NEXT:    s_mov_b32 s4, s0
; SI-NEXT:    s_mov_b32 s5, s1
; SI-NEXT:    s_waitcnt vmcnt(0)
; SI-NEXT:    v_and_b32_e32 v0, 1, v0
; SI-NEXT:    buffer_store_byte v0, off, s[4:7], 0
; SI-NEXT:    s_endpgm
;
; VI-LABEL: zextload_i1_to_i32_trunc_cmp_eq_1:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    s_mov_b32 s7, 0xf000
; VI-NEXT:    s_mov_b32 s6, -1
; VI-NEXT:    s_mov_b32 s10, s6
; VI-NEXT:    s_mov_b32 s11, s7
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    s_mov_b32 s8, s2
; VI-NEXT:    s_mov_b32 s9, s3
; VI-NEXT:    buffer_load_ubyte v0, off, s[8:11], 0
; VI-NEXT:    s_mov_b32 s4, s0
; VI-NEXT:    s_mov_b32 s5, s1
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    v_and_b32_e32 v0, 1, v0
; VI-NEXT:    buffer_store_byte v0, off, s[4:7], 0
; VI-NEXT:    s_endpgm
  %load = load i1, ptr addrspace(1) %in
  %ext = zext i1 %load to i32
  %cmp = icmp eq i32 %ext, 1
  store i1 %cmp, ptr addrspace(1) %out
  ret void
}

define amdgpu_kernel void @sextload_i1_to_i32_trunc_cmp_eq_neg1(ptr addrspace(1) %out, ptr addrspace(1) %in) nounwind {
; SI-LABEL: sextload_i1_to_i32_trunc_cmp_eq_neg1:
; SI:       ; %bb.0:
; SI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x9
; SI-NEXT:    s_mov_b32 s7, 0xf000
; SI-NEXT:    s_mov_b32 s6, -1
; SI-NEXT:    s_mov_b32 s10, s6
; SI-NEXT:    s_mov_b32 s11, s7
; SI-NEXT:    s_waitcnt lgkmcnt(0)
; SI-NEXT:    s_mov_b32 s8, s2
; SI-NEXT:    s_mov_b32 s9, s3
; SI-NEXT:    buffer_load_ubyte v0, off, s[8:11], 0
; SI-NEXT:    s_mov_b32 s4, s0
; SI-NEXT:    s_mov_b32 s5, s1
; SI-NEXT:    s_waitcnt vmcnt(0)
; SI-NEXT:    v_and_b32_e32 v0, 1, v0
; SI-NEXT:    buffer_store_byte v0, off, s[4:7], 0
; SI-NEXT:    s_endpgm
;
; VI-LABEL: sextload_i1_to_i32_trunc_cmp_eq_neg1:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    s_mov_b32 s7, 0xf000
; VI-NEXT:    s_mov_b32 s6, -1
; VI-NEXT:    s_mov_b32 s10, s6
; VI-NEXT:    s_mov_b32 s11, s7
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    s_mov_b32 s8, s2
; VI-NEXT:    s_mov_b32 s9, s3
; VI-NEXT:    buffer_load_ubyte v0, off, s[8:11], 0
; VI-NEXT:    s_mov_b32 s4, s0
; VI-NEXT:    s_mov_b32 s5, s1
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    v_and_b32_e32 v0, 1, v0
; VI-NEXT:    buffer_store_byte v0, off, s[4:7], 0
; VI-NEXT:    s_endpgm
  %load = load i1, ptr addrspace(1) %in
  %ext = sext i1 %load to i32
  %cmp = icmp eq i32 %ext, -1
  store i1 %cmp, ptr addrspace(1) %out
  ret void
}

define amdgpu_kernel void @zextload_i1_to_i32_trunc_cmp_eq_neg1(ptr addrspace(1) %out, ptr addrspace(1) %in) nounwind {
; SI-LABEL: zextload_i1_to_i32_trunc_cmp_eq_neg1:
; SI:       ; %bb.0:
; SI-NEXT:    s_load_dwordx2 s[0:1], s[4:5], 0x9
; SI-NEXT:    s_mov_b32 s3, 0xf000
; SI-NEXT:    s_mov_b32 s2, -1
; SI-NEXT:    v_mov_b32_e32 v0, 0
; SI-NEXT:    s_waitcnt lgkmcnt(0)
; SI-NEXT:    buffer_store_byte v0, off, s[0:3], 0
; SI-NEXT:    s_endpgm
;
; VI-LABEL: zextload_i1_to_i32_trunc_cmp_eq_neg1:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx2 s[0:1], s[4:5], 0x24
; VI-NEXT:    s_mov_b32 s3, 0xf000
; VI-NEXT:    s_mov_b32 s2, -1
; VI-NEXT:    v_mov_b32_e32 v0, 0
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    buffer_store_byte v0, off, s[0:3], 0
; VI-NEXT:    s_endpgm
  %load = load i1, ptr addrspace(1) %in
  %ext = zext i1 %load to i32
  %cmp = icmp eq i32 %ext, -1
  store i1 %cmp, ptr addrspace(1) %out
  ret void
}


define amdgpu_kernel void @sextload_i1_to_i32_trunc_cmp_ne_0(ptr addrspace(1) %out, ptr addrspace(1) %in) nounwind {
; SI-LABEL: sextload_i1_to_i32_trunc_cmp_ne_0:
; SI:       ; %bb.0:
; SI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x9
; SI-NEXT:    s_mov_b32 s7, 0xf000
; SI-NEXT:    s_mov_b32 s6, -1
; SI-NEXT:    s_mov_b32 s10, s6
; SI-NEXT:    s_mov_b32 s11, s7
; SI-NEXT:    s_waitcnt lgkmcnt(0)
; SI-NEXT:    s_mov_b32 s8, s2
; SI-NEXT:    s_mov_b32 s9, s3
; SI-NEXT:    buffer_load_ubyte v0, off, s[8:11], 0
; SI-NEXT:    s_mov_b32 s4, s0
; SI-NEXT:    s_mov_b32 s5, s1
; SI-NEXT:    s_waitcnt vmcnt(0)
; SI-NEXT:    v_and_b32_e32 v0, 1, v0
; SI-NEXT:    buffer_store_byte v0, off, s[4:7], 0
; SI-NEXT:    s_endpgm
;
; VI-LABEL: sextload_i1_to_i32_trunc_cmp_ne_0:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    s_mov_b32 s7, 0xf000
; VI-NEXT:    s_mov_b32 s6, -1
; VI-NEXT:    s_mov_b32 s10, s6
; VI-NEXT:    s_mov_b32 s11, s7
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    s_mov_b32 s8, s2
; VI-NEXT:    s_mov_b32 s9, s3
; VI-NEXT:    buffer_load_ubyte v0, off, s[8:11], 0
; VI-NEXT:    s_mov_b32 s4, s0
; VI-NEXT:    s_mov_b32 s5, s1
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    v_and_b32_e32 v0, 1, v0
; VI-NEXT:    buffer_store_byte v0, off, s[4:7], 0
; VI-NEXT:    s_endpgm
  %load = load i1, ptr addrspace(1) %in
  %ext = sext i1 %load to i32
  %cmp = icmp ne i32 %ext, 0
  store i1 %cmp, ptr addrspace(1) %out
  ret void
}

define amdgpu_kernel void @zextload_i1_to_i32_trunc_cmp_ne_0(ptr addrspace(1) %out, ptr addrspace(1) %in) nounwind {
; SI-LABEL: zextload_i1_to_i32_trunc_cmp_ne_0:
; SI:       ; %bb.0:
; SI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x9
; SI-NEXT:    s_mov_b32 s7, 0xf000
; SI-NEXT:    s_mov_b32 s6, -1
; SI-NEXT:    s_mov_b32 s10, s6
; SI-NEXT:    s_mov_b32 s11, s7
; SI-NEXT:    s_waitcnt lgkmcnt(0)
; SI-NEXT:    s_mov_b32 s8, s2
; SI-NEXT:    s_mov_b32 s9, s3
; SI-NEXT:    buffer_load_ubyte v0, off, s[8:11], 0
; SI-NEXT:    s_mov_b32 s4, s0
; SI-NEXT:    s_mov_b32 s5, s1
; SI-NEXT:    s_waitcnt vmcnt(0)
; SI-NEXT:    v_and_b32_e32 v0, 1, v0
; SI-NEXT:    buffer_store_byte v0, off, s[4:7], 0
; SI-NEXT:    s_endpgm
;
; VI-LABEL: zextload_i1_to_i32_trunc_cmp_ne_0:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    s_mov_b32 s7, 0xf000
; VI-NEXT:    s_mov_b32 s6, -1
; VI-NEXT:    s_mov_b32 s10, s6
; VI-NEXT:    s_mov_b32 s11, s7
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    s_mov_b32 s8, s2
; VI-NEXT:    s_mov_b32 s9, s3
; VI-NEXT:    buffer_load_ubyte v0, off, s[8:11], 0
; VI-NEXT:    s_mov_b32 s4, s0
; VI-NEXT:    s_mov_b32 s5, s1
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    v_and_b32_e32 v0, 1, v0
; VI-NEXT:    buffer_store_byte v0, off, s[4:7], 0
; VI-NEXT:    s_endpgm
  %load = load i1, ptr addrspace(1) %in
  %ext = zext i1 %load to i32
  %cmp = icmp ne i32 %ext, 0
  store i1 %cmp, ptr addrspace(1) %out
  ret void
}

define amdgpu_kernel void @sextload_i1_to_i32_trunc_cmp_ne_1(ptr addrspace(1) %out, ptr addrspace(1) %in) nounwind {
; SI-LABEL: sextload_i1_to_i32_trunc_cmp_ne_1:
; SI:       ; %bb.0:
; SI-NEXT:    s_load_dwordx2 s[0:1], s[4:5], 0x9
; SI-NEXT:    s_mov_b32 s3, 0xf000
; SI-NEXT:    s_mov_b32 s2, -1
; SI-NEXT:    v_mov_b32_e32 v0, 1
; SI-NEXT:    s_waitcnt lgkmcnt(0)
; SI-NEXT:    buffer_store_byte v0, off, s[0:3], 0
; SI-NEXT:    s_endpgm
;
; VI-LABEL: sextload_i1_to_i32_trunc_cmp_ne_1:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx2 s[0:1], s[4:5], 0x24
; VI-NEXT:    s_mov_b32 s3, 0xf000
; VI-NEXT:    s_mov_b32 s2, -1
; VI-NEXT:    v_mov_b32_e32 v0, 1
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    buffer_store_byte v0, off, s[0:3], 0
; VI-NEXT:    s_endpgm
  %load = load i1, ptr addrspace(1) %in
  %ext = sext i1 %load to i32
  %cmp = icmp ne i32 %ext, 1
  store i1 %cmp, ptr addrspace(1) %out
  ret void
}

define amdgpu_kernel void @zextload_i1_to_i32_trunc_cmp_ne_1(ptr addrspace(1) %out, ptr addrspace(1) %in) nounwind {
; SI-LABEL: zextload_i1_to_i32_trunc_cmp_ne_1:
; SI:       ; %bb.0:
; SI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x9
; SI-NEXT:    s_mov_b32 s7, 0xf000
; SI-NEXT:    s_mov_b32 s6, -1
; SI-NEXT:    s_mov_b32 s10, s6
; SI-NEXT:    s_mov_b32 s11, s7
; SI-NEXT:    s_waitcnt lgkmcnt(0)
; SI-NEXT:    s_mov_b32 s8, s2
; SI-NEXT:    s_mov_b32 s9, s3
; SI-NEXT:    buffer_load_ubyte v0, off, s[8:11], 0
; SI-NEXT:    s_mov_b32 s4, s0
; SI-NEXT:    s_mov_b32 s5, s1
; SI-NEXT:    s_waitcnt vmcnt(0)
; SI-NEXT:    v_and_b32_e32 v0, 1, v0
; SI-NEXT:    v_cmp_eq_u32_e32 vcc, 1, v0
; SI-NEXT:    s_xor_b64 s[0:1], vcc, -1
; SI-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; SI-NEXT:    buffer_store_byte v0, off, s[4:7], 0
; SI-NEXT:    s_endpgm
;
; VI-LABEL: zextload_i1_to_i32_trunc_cmp_ne_1:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    s_mov_b32 s7, 0xf000
; VI-NEXT:    s_mov_b32 s6, -1
; VI-NEXT:    s_mov_b32 s10, s6
; VI-NEXT:    s_mov_b32 s11, s7
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    s_mov_b32 s8, s2
; VI-NEXT:    s_mov_b32 s9, s3
; VI-NEXT:    buffer_load_ubyte v0, off, s[8:11], 0
; VI-NEXT:    s_mov_b32 s4, s0
; VI-NEXT:    s_mov_b32 s5, s1
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    v_and_b32_e32 v0, 1, v0
; VI-NEXT:    v_cmp_eq_u32_e32 vcc, 1, v0
; VI-NEXT:    s_xor_b64 s[0:1], vcc, -1
; VI-NEXT:    v_cndmask_b32_e64 v0, 0, 1, s[0:1]
; VI-NEXT:    buffer_store_byte v0, off, s[4:7], 0
; VI-NEXT:    s_endpgm
  %load = load i1, ptr addrspace(1) %in
  %ext = zext i1 %load to i32
  %cmp = icmp ne i32 %ext, 1
  store i1 %cmp, ptr addrspace(1) %out
  ret void
}

; FIXME: This should be one compare.
define amdgpu_kernel void @sextload_i1_to_i32_trunc_cmp_ne_neg1(ptr addrspace(1) %out, ptr addrspace(1) %in) nounwind {
; SI-LABEL: sextload_i1_to_i32_trunc_cmp_ne_neg1:
; SI:       ; %bb.0:
; SI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x9
; SI-NEXT:    s_mov_b32 s7, 0xf000
; SI-NEXT:    s_mov_b32 s6, -1
; SI-NEXT:    s_mov_b32 s10, s6
; SI-NEXT:    s_mov_b32 s11, s7
; SI-NEXT:    s_waitcnt lgkmcnt(0)
; SI-NEXT:    s_mov_b32 s8, s2
; SI-NEXT:    s_mov_b32 s9, s3
; SI-NEXT:    buffer_load_ubyte v0, off, s[8:11], 0
; SI-NEXT:    s_mov_b32 s4, s0
; SI-NEXT:    s_mov_b32 s5, s1
; SI-NEXT:    s_waitcnt vmcnt(0)
; SI-NEXT:    v_and_b32_e32 v0, 1, v0
; SI-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
; SI-NEXT:    v_cndmask_b32_e64 v0, 0, 1, vcc
; SI-NEXT:    buffer_store_byte v0, off, s[4:7], 0
; SI-NEXT:    s_endpgm
;
; VI-LABEL: sextload_i1_to_i32_trunc_cmp_ne_neg1:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; VI-NEXT:    s_mov_b32 s7, 0xf000
; VI-NEXT:    s_mov_b32 s6, -1
; VI-NEXT:    s_mov_b32 s10, s6
; VI-NEXT:    s_mov_b32 s11, s7
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    s_mov_b32 s8, s2
; VI-NEXT:    s_mov_b32 s9, s3
; VI-NEXT:    buffer_load_ubyte v0, off, s[8:11], 0
; VI-NEXT:    s_mov_b32 s4, s0
; VI-NEXT:    s_mov_b32 s5, s1
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    v_and_b32_e32 v0, 1, v0
; VI-NEXT:    v_cmp_eq_u32_e32 vcc, 0, v0
; VI-NEXT:    v_cndmask_b32_e64 v0, 0, 1, vcc
; VI-NEXT:    buffer_store_byte v0, off, s[4:7], 0
; VI-NEXT:    s_endpgm
  %load = load i1, ptr addrspace(1) %in
  %ext = sext i1 %load to i32
  %cmp = icmp ne i32 %ext, -1
  store i1 %cmp, ptr addrspace(1) %out
  ret void
}

define amdgpu_kernel void @zextload_i1_to_i32_trunc_cmp_ne_neg1(ptr addrspace(1) %out, ptr addrspace(1) %in) nounwind {
; SI-LABEL: zextload_i1_to_i32_trunc_cmp_ne_neg1:
; SI:       ; %bb.0:
; SI-NEXT:    s_load_dwordx2 s[0:1], s[4:5], 0x9
; SI-NEXT:    s_mov_b32 s3, 0xf000
; SI-NEXT:    s_mov_b32 s2, -1
; SI-NEXT:    v_mov_b32_e32 v0, 1
; SI-NEXT:    s_waitcnt lgkmcnt(0)
; SI-NEXT:    buffer_store_byte v0, off, s[0:3], 0
; SI-NEXT:    s_endpgm
;
; VI-LABEL: zextload_i1_to_i32_trunc_cmp_ne_neg1:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx2 s[0:1], s[4:5], 0x24
; VI-NEXT:    s_mov_b32 s3, 0xf000
; VI-NEXT:    s_mov_b32 s2, -1
; VI-NEXT:    v_mov_b32_e32 v0, 1
; VI-NEXT:    s_waitcnt lgkmcnt(0)
; VI-NEXT:    buffer_store_byte v0, off, s[0:3], 0
; VI-NEXT:    s_endpgm
  %load = load i1, ptr addrspace(1) %in
  %ext = zext i1 %load to i32
  %cmp = icmp ne i32 %ext, -1
  store i1 %cmp, ptr addrspace(1) %out
  ret void
}

; FIXME: Need to handle non-uniform case for function below (load without gep).
define amdgpu_kernel void @masked_load_i1_to_i32_trunc_cmp_ne_neg1(ptr addrspace(1) %out, ptr addrspace(1) %in) nounwind {
; SI-LABEL: masked_load_i1_to_i32_trunc_cmp_ne_neg1:
; SI:       ; %bb.0:
; SI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x9
; SI-NEXT:    s_mov_b32 s7, 0xf000
; SI-NEXT:    v_mov_b32_e32 v1, 0
; SI-NEXT:    s_mov_b32 s10, 0
; SI-NEXT:    s_mov_b32 s11, s7
; SI-NEXT:    s_waitcnt lgkmcnt(0)
; SI-NEXT:    s_mov_b64 s[8:9], s[2:3]
; SI-NEXT:    buffer_load_sbyte v0, v[0:1], s[8:11], 0 addr64
; SI-NEXT:    s_mov_b32 s6, -1
; SI-NEXT:    s_mov_b32 s4, s0
; SI-NEXT:    s_mov_b32 s5, s1
; SI-NEXT:    s_waitcnt vmcnt(0)
; SI-NEXT:    v_cmp_ne_u32_e32 vcc, -1, v0
; SI-NEXT:    v_cndmask_b32_e64 v0, 0, 1, vcc
; SI-NEXT:    buffer_store_byte v0, off, s[4:7], 0
; SI-NEXT:    s_endpgm
;
; VI-LABEL: masked_load_i1_to_i32_trunc_cmp_ne_neg1:
; VI:       ; %bb.0:
; VI-NEXT:    s_load_dwordx4 s[0:3], s[4:5], 0x24
; 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_sbyte v0, v[0:1]
; VI-NEXT:    s_mov_b32 s3, 0xf000
; VI-NEXT:    s_mov_b32 s2, -1
; VI-NEXT:    s_waitcnt vmcnt(0)
; VI-NEXT:    v_cmp_ne_u32_e32 vcc, -1, v0
; VI-NEXT:    v_cndmask_b32_e64 v0, 0, 1, vcc
; VI-NEXT:    buffer_store_byte v0, off, s[0:3], 0
; VI-NEXT:    s_endpgm
  %tid.x = call i32 @llvm.amdgcn.workitem.id.x()
  %in.ptr = getelementptr i8, ptr addrspace(1) %in, i32 %tid.x
  %load = load i8, ptr addrspace(1) %in.ptr
  %masked = and i8 %load, 255
  %ext = sext i8 %masked to i32
  %cmp = icmp ne i32 %ext, -1
  store i1 %cmp, ptr addrspace(1) %out
  ret void
}