; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -passes=instcombine -S | FileCheck %s

target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
target triple = "x86_64-pc-win32"

%myStruct = type { float, [3 x float], [4 x float], i32 }

; make sure that we are not crashing when creating an illegal type
define void @func(%myStruct addrspace(1)* nocapture %p) nounwind {
; CHECK-LABEL: @func(
; CHECK-NEXT:    ret void
;
  %A = getelementptr inbounds %myStruct, %myStruct addrspace(1)* %p, i64 0
  %B = addrspacecast %myStruct addrspace(1)* %A to %myStruct*
  %C = getelementptr inbounds %myStruct, %myStruct* %B, i32 0, i32 1
  %D = getelementptr inbounds [3 x float], [3 x float]* %C, i32 0, i32 2
  %E = load float, float* %D, align 4
  %F = fsub float %E, undef
  ret void
}

@array = internal addrspace(3) global [256 x float] zeroinitializer, align 4
@scalar = internal addrspace(3) global float 0.000000e+00, align 4

define void @keep_necessary_addrspacecast(i64 %i, float** %out0, float** %out1) {
; CHECK-LABEL: @keep_necessary_addrspacecast(
; CHECK-NEXT:    [[T01:%.*]] = getelementptr [256 x float], [256 x float] addrspace(3)* @array, i64 0, i64 [[I:%.*]]
; CHECK-NEXT:    [[T0:%.*]] = addrspacecast float addrspace(3)* [[T01]] to float*
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr float, float addrspace(3)* @scalar, i64 [[I]]
; CHECK-NEXT:    [[T1:%.*]] = addrspacecast float addrspace(3)* [[TMP1]] to float*
; CHECK-NEXT:    store float* [[T0]], float** [[OUT0:%.*]], align 4
; CHECK-NEXT:    store float* [[T1]], float** [[OUT1:%.*]], align 4
; CHECK-NEXT:    ret void
;
  %t0 = getelementptr [256 x float], [256 x float]* addrspacecast ([256 x float] addrspace(3)* @array to [256 x float]*), i64 0, i64 %i
  %t1 = getelementptr [0 x float], [0 x float]* addrspacecast (float addrspace(3)* @scalar to [0 x float]*), i64 0, i64 %i
  store float* %t0, float** %out0, align 4
  store float* %t1, float** %out1, align 4
  ret void
}

declare void @escape_alloca(i16*)

; check that addrspacecast is stripped when trying to mark a GEP as inbounds
define { i8, i8 } @inbounds_after_addrspacecast() {
; CHECK-LABEL: @inbounds_after_addrspacecast(
; CHECK-NEXT:    [[T0:%.*]] = alloca i16, align 2
; CHECK-NEXT:    call void @escape_alloca(i16* nonnull [[T0]])
; CHECK-NEXT:    [[TMPCAST:%.*]] = bitcast i16* [[T0]] to [2 x i8]*
; CHECK-NEXT:    [[T1:%.*]] = addrspacecast [2 x i8]* [[TMPCAST]] to [2 x i8] addrspace(11)*
; CHECK-NEXT:    [[T2:%.*]] = getelementptr inbounds [2 x i8], [2 x i8] addrspace(11)* [[T1]], i64 0, i64 1
; CHECK-NEXT:    [[T3:%.*]] = load i8, i8 addrspace(11)* [[T2]], align 1
; CHECK-NEXT:    [[INSERT:%.*]] = insertvalue { i8, i8 } zeroinitializer, i8 [[T3]], 1
; CHECK-NEXT:    ret { i8, i8 } [[INSERT]]
;
  %t0 = alloca i16, align 2
  call void @escape_alloca(i16* %t0)
  %tmpcast = bitcast i16* %t0 to [2 x i8]*
  %t1 = addrspacecast [2 x i8]* %tmpcast to [2 x i8] addrspace(11)*
  %t2 = getelementptr [2 x i8], [2 x i8] addrspace(11)* %t1, i64 0, i64 1
  %t3 = load i8, i8 addrspace(11)* %t2, align 1
  %insert = insertvalue { i8, i8 } zeroinitializer, i8 %t3, 1
  ret { i8, i8 } %insert
}


declare spir_func <16 x i32> @my_extern_func()

; check that a bitcast is not generated when we need an addrspace cast
define void @bitcast_after_gep(<16 x i32>* %t0) {
; CHECK-LABEL: @bitcast_after_gep(
; CHECK-NEXT:    [[T4:%.*]] = addrspacecast <16 x i32>* [[T0:%.*]] to <16 x i32> addrspace(3)*
; CHECK-NEXT:    [[CALL:%.*]] = call spir_func <16 x i32> @my_extern_func()
; CHECK-NEXT:    store <16 x i32> [[CALL]], <16 x i32> addrspace(3)* [[T4]], align 64
; CHECK-NEXT:    ret void
;
  %t1 = bitcast <16 x i32>* %t0 to [16 x i32]*
  %t2 = addrspacecast [16 x i32]* %t1 to [16 x i32] addrspace(3)*
  %t3 = getelementptr inbounds [16 x i32], [16 x i32] addrspace(3)* %t2, i64 0, i64 0
  %t4 = bitcast i32 addrspace(3)* %t3 to <16 x i32> addrspace(3)*
  %call = call spir_func <16 x i32> @my_extern_func()
  store <16 x i32> %call, <16 x i32> addrspace(3)* %t4
  ret void
}