Fix APInt bit size in processDbgDeclares

[yaxunl]

Summary

processDbgDeclares assumes pointer size is the same for different addr spaces.
It uses pointer size for addr space 0 for all pointers, which causes assertion
in stripAndAccumulateInBoundsConstantOffsets for amdgcn---amdgiz since
pointer in addr space 5 has different size than in addr space 0.

This patch fixes that.

[rampitec]

LGTM

[arsenm]

lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp

1147 ↗

(On Diff #123037)

Is a vector of pointers allowed here? How could this never be a PointerType?

[yaxunl]

lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp

1147 ↗

(On Diff #123037)

There are lit tests in which Address is not pointer, e.g. CodeGen/X86/selectiondag-order.ll

define i64 @simulateWithDbgDeclare(<2 x i32> %a) local_unnamed_addr  {
entry:
  %rand = tail call i64 @lrand48() #3
  tail call void @llvm.dbg.declare(metadata i64 %rand, metadata !6, metadata !7), !dbg !8
  br label %body

[arsenm]

If this allows non-pointers, I’m very suspicious of using a pointer size for this of any address space. Something seems off about this. Should it use some maximum size int type?

[arsenm]

Should it just be the type size? There seems like no reason to use the low level pointer size query

[yaxunl]

In D40085#926766, @arsenm wrote:

Should it just be the type size? There seems like no reason to use the low level pointer size query

That sounds reasonable. Will do.

[yaxunl]

Revised by Matt's comments.

[arsenm]

LGTM

llvm/trunk/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp

for (const Instruction &I : BB) {
       assert(DI->getVariable() && "Missing variable");
       assert(DI->getDebugLoc() && "Missing location");
       const Value *Address = DI->getAddress();
       if (!Address)
         continue;
 
       // Look through casts and constant offset GEPs. These mostly come from
       // inalloca.
-      APInt Offset(DL.getPointerSizeInBits(0), 0);
+      APInt Offset(DL.getTypeSizeInBits(Address->getType()), 0);
       Address = Address->stripAndAccumulateInBoundsConstantOffsets(DL, Offset);
 
       // Check if the variable is a static alloca or a byval or inalloca
       // argument passed in memory. If it is not, then we will ignore this
       // intrinsic and handle this during isel like dbg.value.
       int FI = std::numeric_limits<int>::max();
       if (const auto *AI = dyn_cast<AllocaInst>(Address)) {
         auto SI = FuncInfo->StaticAllocaMap.find(AI);

llvm/trunk/test/CodeGen/AMDGPU/debugger-emit-prologue.ll

-; RUN: llc -O0 -mtriple=amdgcn--amdhsa -mcpu=fiji -mattr=+amdgpu-debugger-emit-prologue -verify-machineinstrs < %s | FileCheck %s
-; RUN: llc -O0 -mtriple=amdgcn--amdhsa -mcpu=fiji -verify-machineinstrs < %s | FileCheck %s --check-prefix=NOATTR
+; RUN: llc -O0 -mtriple=amdgcn--amdhsa-amdgiz -mcpu=fiji -mattr=+amdgpu-debugger-emit-prologue -verify-machineinstrs < %s | FileCheck %s
+; RUN: llc -O0 -mtriple=amdgcn--amdhsa-amdgiz -mcpu=fiji -verify-machineinstrs < %s | FileCheck %s --check-prefix=NOATTR
+target datalayout = "A5"
 
 ; CHECK: debug_wavefront_private_segment_offset_sgpr = [[SOFF:[0-9]+]]
 ; CHECK: debug_private_segment_buffer_sgpr = [[SREG:[0-9]+]]
 
 ; CHECK: v_mov_b32_e32 [[WGIDX:v[0-9]+]], s{{[0-9]+}}
 ; CHECK: buffer_store_dword [[WGIDX]], off, s[{{[0-9]+:[0-9]+}}], s[[SOFF]]
 ; CHECK: buffer_store_dword v0, off, s[{{[0-9]+:[0-9]+}}], s[[SOFF]] offset:16
 

 ; CHECK: DebuggerPrivateSegmentBufferSGPR: s[[SREG]]
 
 ; NOATTR-NOT: DebuggerWavefrontPrivateSegmentOffsetSGPR
 ; NOATTR-NOT: DebuggerPrivateSegmentBufferSGPR
 
 ; Function Attrs: nounwind
 define amdgpu_kernel void @test(i32 addrspace(1)* %A) #0 !dbg !12 {
 entry:
-  %A.addr = alloca i32 addrspace(1)*, align 4
-  store i32 addrspace(1)* %A, i32 addrspace(1)** %A.addr, align 4
-  call void @llvm.dbg.declare(metadata i32 addrspace(1)** %A.addr, metadata !17, metadata !18), !dbg !19
-  %0 = load i32 addrspace(1)*, i32 addrspace(1)** %A.addr, align 4, !dbg !20
+  %A.addr = alloca i32 addrspace(1)*, align 4, addrspace(5)
+  store i32 addrspace(1)* %A, i32 addrspace(1)* addrspace(5)* %A.addr, align 4
+  call void @llvm.dbg.declare(metadata i32 addrspace(1)* addrspace(5)* %A.addr, metadata !17, metadata !18), !dbg !19
+  %0 = load i32 addrspace(1)*, i32 addrspace(1)* addrspace(5)* %A.addr, align 4, !dbg !20
   %arrayidx = getelementptr inbounds i32, i32 addrspace(1)* %0, i32 0, !dbg !20
   store i32 1, i32 addrspace(1)* %arrayidx, align 4, !dbg !21
-  %1 = load i32 addrspace(1)*, i32 addrspace(1)** %A.addr, align 4, !dbg !22
+  %1 = load i32 addrspace(1)*, i32 addrspace(1)* addrspace(5)* %A.addr, align 4, !dbg !22
   %arrayidx1 = getelementptr inbounds i32, i32 addrspace(1)* %1, i32 1, !dbg !22
   store i32 2, i32 addrspace(1)* %arrayidx1, align 4, !dbg !23
-  %2 = load i32 addrspace(1)*, i32 addrspace(1)** %A.addr, align 4, !dbg !24
+  %2 = load i32 addrspace(1)*, i32 addrspace(1)* addrspace(5)* %A.addr, align 4, !dbg !24
   %arrayidx2 = getelementptr inbounds i32, i32 addrspace(1)* %2, i32 2, !dbg !24
   store i32 3, i32 addrspace(1)* %arrayidx2, align 4, !dbg !25
   ret void, !dbg !26
 }
 
 ; Function Attrs: nounwind readnone
 declare void @llvm.dbg.declare(metadata, metadata, metadata) #1
 
 attributes #0 = { nounwind "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="fiji" "unsafe-fp-math"="false" "use-soft-float"="false" }
 attributes #1 = { nounwind readnone }
 
 !llvm.dbg.cu = !{!0}
 !opencl.kernels = !{!3}
 !llvm.module.flags = !{!9, !10}
 !llvm.ident = !{!11}
 
 !0 = distinct !DICompileUnit(language: DW_LANG_C99, file: !1, producer: "clang version 3.9.0 (trunk 269772)", isOptimized: false, runtimeVersion: 0, emissionKind: FullDebug, enums: !2)
 !1 = !DIFile(filename: "test01.cl", directory: "/home/kzhuravl/Lightning/testing")
 !2 = !{}
 !3 = !{void (i32 addrspace(1)*)* @test, !4, !5, !6, !7, !8}
 !4 = !{!"kernel_arg_addr_space", i32 1}
 !5 = !{!"kernel_arg_access_qual", !"none"}
-!6 = !{!"kernel_arg_type", !"int*"}
-!7 = !{!"kernel_arg_base_type", !"int*"}
+!6 = !{!"kernel_arg_type", !"int addrspace(5)*"}
+!7 = !{!"kernel_arg_base_type", !"int addrspace(5)*"}
 !8 = !{!"kernel_arg_type_qual", !""}
 !9 = !{i32 2, !"Dwarf Version", i32 2}
 !10 = !{i32 2, !"Debug Info Version", i32 3}
 !11 = !{!"clang version 3.9.0 (trunk 269772)"}
 !12 = distinct !DISubprogram(name: "test", scope: !1, file: !1, line: 1, type: !13, isLocal: false, isDefinition: true, scopeLine: 1, flags: DIFlagPrototyped, isOptimized: false, unit: !0, variables: !2)
 !13 = !DISubroutineType(types: !14)
 !14 = !{null, !15}
 !15 = !DIDerivedType(tag: DW_TAG_pointer_type, baseType: !16, size: 64, align: 32)

llvm/trunk/test/CodeGen/AMDGPU/debugger-insert-nops.ll

-; RUN: llc -O0 -mtriple=amdgcn--amdhsa -mcpu=fiji -mattr=+amdgpu-debugger-insert-nops -verify-machineinstrs < %s | FileCheck %s --check-prefix=CHECK
-; RUN: llc -O0 -mtriple=amdgcn--amdhsa -mcpu=fiji -mattr=+amdgpu-debugger-insert-nops -verify-machineinstrs < %s | FileCheck %s --check-prefix=CHECKNOP
+; RUN: llc -O0 -mtriple=amdgcn--amdhsa-amdgiz -mcpu=fiji -mattr=+amdgpu-debugger-insert-nops -verify-machineinstrs < %s | FileCheck %s --check-prefix=CHECK
+; RUN: llc -O0 -mtriple=amdgcn--amdhsa-amdgiz -mcpu=fiji -mattr=+amdgpu-debugger-insert-nops -verify-machineinstrs < %s | FileCheck %s --check-prefix=CHECKNOP
+target datalayout = "A5"
 
 ; This test expects that we have one instance for each line in some order with "s_nop 0" instances after each.
 
 ; Check that each line appears at least once
 ; CHECK-DAG: test01.cl:2:3
 ; CHECK-DAG: test01.cl:3:3
 ; CHECK-DAG: test01.cl:4:3
 
 
 ; Check that each of each of the lines consists of the line output, followed by "s_nop 0"
 ; CHECKNOP: test01.cl:{{[234]}}:3
 ; CHECKNOP-NEXT: s_nop 0
 ; CHECKNOP: test01.cl:{{[234]}}:3
 ; CHECKNOP-NEXT: s_nop 0
 ; CHECKNOP: test01.cl:{{[234]}}:3
 ; CHECKNOP-NEXT: s_nop 0
 
 ; CHECK: test01.cl:5:{{[0-9]+}}
 ; CHECK-NEXT: s_nop 0
 ; CHECK-NEXT: s_endpgm
 
 ; Function Attrs: nounwind
 define amdgpu_kernel void @test(i32 addrspace(1)* %A) #0 !dbg !12 {
 entry:
-  %A.addr = alloca i32 addrspace(1)*, align 4
-  store i32 addrspace(1)* %A, i32 addrspace(1)** %A.addr, align 4
-  call void @llvm.dbg.declare(metadata i32 addrspace(1)** %A.addr, metadata !17, metadata !18), !dbg !19
-  %0 = load i32 addrspace(1)*, i32 addrspace(1)** %A.addr, align 4, !dbg !20
+  %A.addr = alloca i32 addrspace(1)*, align 4, addrspace(5)
+  store i32 addrspace(1)* %A, i32 addrspace(1)* addrspace(5)* %A.addr, align 4
+  call void @llvm.dbg.declare(metadata i32 addrspace(1)* addrspace(5)* %A.addr, metadata !17, metadata !18), !dbg !19
+  %0 = load i32 addrspace(1)*, i32 addrspace(1)* addrspace(5)* %A.addr, align 4, !dbg !20
   %arrayidx = getelementptr inbounds i32, i32 addrspace(1)* %0, i32 0, !dbg !20
   store i32 1, i32 addrspace(1)* %arrayidx, align 4, !dbg !20
-  %1 = load i32 addrspace(1)*, i32 addrspace(1)** %A.addr, align 4, !dbg !22
+  %1 = load i32 addrspace(1)*, i32 addrspace(1)* addrspace(5)* %A.addr, align 4, !dbg !22
   %arrayidx1 = getelementptr inbounds i32, i32 addrspace(1)* %1, i32 1, !dbg !22
   store i32 2, i32 addrspace(1)* %arrayidx1, align 4, !dbg !23
-  %2 = load i32 addrspace(1)*, i32 addrspace(1)** %A.addr, align 4, !dbg !24
+  %2 = load i32 addrspace(1)*, i32 addrspace(1)* addrspace(5)* %A.addr, align 4, !dbg !24
   %arrayidx2 = getelementptr inbounds i32, i32 addrspace(1)* %2, i32 2, !dbg !24
   store i32 3, i32 addrspace(1)* %arrayidx2, align 4, !dbg !25
   ret void, !dbg !26
 }
 
 ; Function Attrs: nounwind readnone
 declare void @llvm.dbg.declare(metadata, metadata, metadata) #1
 
 attributes #0 = { nounwind }
 attributes #1 = { nounwind readnone }
 
 !llvm.dbg.cu = !{!0}
 !opencl.kernels = !{!3}
 !llvm.module.flags = !{!9, !10}
 !llvm.ident = !{!11}
 
 !0 = distinct !DICompileUnit(language: DW_LANG_C99, file: !1, producer: "clang version 3.9.0 (trunk 268929)", isOptimized: false, runtimeVersion: 0, emissionKind: FullDebug, enums: !2)
 !1 = !DIFile(filename: "test01.cl", directory: "/home/kzhuravl/Lightning/testing")
 !2 = !{}
 !3 = !{void (i32 addrspace(1)*)* @test, !4, !5, !6, !7, !8}
 !4 = !{!"kernel_arg_addr_space", i32 1}
 !5 = !{!"kernel_arg_access_qual", !"none"}
-!6 = !{!"kernel_arg_type", !"int*"}
-!7 = !{!"kernel_arg_base_type", !"int*"}
+!6 = !{!"kernel_arg_type", !"int addrspace(5)*"}
+!7 = !{!"kernel_arg_base_type", !"int addrspace(5)*"}
 !8 = !{!"kernel_arg_type_qual", !""}
 !9 = !{i32 2, !"Dwarf Version", i32 2}
 !10 = !{i32 2, !"Debug Info Version", i32 3}
 !11 = !{!"clang version 3.9.0 (trunk 268929)"}
 !12 = distinct !DISubprogram(name: "test", scope: !1, file: !1, line: 1, type: !13, isLocal: false, isDefinition: true, scopeLine: 1, flags: DIFlagPrototyped, isOptimized: false, unit: !0, variables: !2)
 !13 = !DISubroutineType(types: !14)
 !14 = !{null, !15}
 !15 = !DIDerivedType(tag: DW_TAG_pointer_type, baseType: !16, size: 64, align: 32)

llvm/trunk/test/CodeGen/AMDGPU/debugger-reserve-regs.ll

-; RUN: llc -O0 -mtriple=amdgcn--amdhsa -mcpu=fiji -mattr=+amdgpu-debugger-reserve-regs -verify-machineinstrs < %s | FileCheck %s
-; RUN: llc -O0 -mtriple=amdgcn--amdhsa -mcpu=gfx901 -mattr=+amdgpu-debugger-reserve-regs -verify-machineinstrs < %s | FileCheck %s
+; RUN: llc -O0 -mtriple=amdgcn--amdhsa-amdgiz -mcpu=fiji -mattr=+amdgpu-debugger-reserve-regs -verify-machineinstrs < %s | FileCheck %s
+; RUN: llc -O0 -mtriple=amdgcn--amdhsa-amdgiz -mcpu=gfx901 -mattr=+amdgpu-debugger-reserve-regs -verify-machineinstrs < %s | FileCheck %s
+target datalayout = "A5"
 ; CHECK: reserved_vgpr_first = {{[0-9]+}}
 ; CHECK-NEXT: reserved_vgpr_count = 4
 ; CHECK: ReservedVGPRFirst: {{[0-9]+}}
 ; CHECK-NEXT: ReservedVGPRCount: 4
 
 ; Function Attrs: nounwind
 define amdgpu_kernel void @test(i32 addrspace(1)* %A) #0 !dbg !12 {
 entry:
-  %A.addr = alloca i32 addrspace(1)*, align 4
-  store i32 addrspace(1)* %A, i32 addrspace(1)** %A.addr, align 4
-  call void @llvm.dbg.declare(metadata i32 addrspace(1)** %A.addr, metadata !17, metadata !18), !dbg !19
-  %0 = load i32 addrspace(1)*, i32 addrspace(1)** %A.addr, align 4, !dbg !20
+  %A.addr = alloca i32 addrspace(1)*, align 4, addrspace(5)
+  store i32 addrspace(1)* %A, i32 addrspace(1)* addrspace(5)* %A.addr, align 4
+  call void @llvm.dbg.declare(metadata i32 addrspace(1)* addrspace(5)* %A.addr, metadata !17, metadata !18), !dbg !19
+  %0 = load i32 addrspace(1)*, i32 addrspace(1)* addrspace(5)* %A.addr, align 4, !dbg !20
   %arrayidx = getelementptr inbounds i32, i32 addrspace(1)* %0, i32 0, !dbg !20
   store i32 1, i32 addrspace(1)* %arrayidx, align 4, !dbg !21
-  %1 = load i32 addrspace(1)*, i32 addrspace(1)** %A.addr, align 4, !dbg !22
+  %1 = load i32 addrspace(1)*, i32 addrspace(1)* addrspace(5)* %A.addr, align 4, !dbg !22
   %arrayidx1 = getelementptr inbounds i32, i32 addrspace(1)* %1, i32 1, !dbg !22
   store i32 2, i32 addrspace(1)* %arrayidx1, align 4, !dbg !23
-  %2 = load i32 addrspace(1)*, i32 addrspace(1)** %A.addr, align 4, !dbg !24
+  %2 = load i32 addrspace(1)*, i32 addrspace(1)* addrspace(5)* %A.addr, align 4, !dbg !24
   %arrayidx2 = getelementptr inbounds i32, i32 addrspace(1)* %2, i32 2, !dbg !24
   store i32 3, i32 addrspace(1)* %arrayidx2, align 4, !dbg !25
   ret void, !dbg !26
 }
 
 ; Function Attrs: nounwind readnone
 declare void @llvm.dbg.declare(metadata, metadata, metadata) #1
 
 attributes #0 = { nounwind }
 attributes #1 = { nounwind readnone }
 
 !llvm.dbg.cu = !{!0}
 !opencl.kernels = !{!3}
 !llvm.module.flags = !{!9, !10}
 !llvm.ident = !{!11}
 
 !0 = distinct !DICompileUnit(language: DW_LANG_C99, file: !1, producer: "clang version 3.9.0 (trunk 268929)", isOptimized: false, runtimeVersion: 0, emissionKind: FullDebug, enums: !2)
 !1 = !DIFile(filename: "test01.cl", directory: "/home/kzhuravl/Lightning/testing")
 !2 = !{}
 !3 = !{void (i32 addrspace(1)*)* @test, !4, !5, !6, !7, !8}
 !4 = !{!"kernel_arg_addr_space", i32 1}
 !5 = !{!"kernel_arg_access_qual", !"none"}
-!6 = !{!"kernel_arg_type", !"int*"}
-!7 = !{!"kernel_arg_base_type", !"int*"}
+!6 = !{!"kernel_arg_type", !"int addrspace(5)*"}
+!7 = !{!"kernel_arg_base_type", !"int addrspace(5)*"}
 !8 = !{!"kernel_arg_type_qual", !""}
 !9 = !{i32 2, !"Dwarf Version", i32 2}
 !10 = !{i32 2, !"Debug Info Version", i32 3}
 !11 = !{!"clang version 3.9.0 (trunk 268929)"}
 !12 = distinct !DISubprogram(name: "test", scope: !1, file: !1, line: 1, type: !13, isLocal: false, isDefinition: true, scopeLine: 1, flags: DIFlagPrototyped, isOptimized: false, unit: !0, variables: !2)
 !13 = !DISubroutineType(types: !14)
 !14 = !{null, !15}
 !15 = !DIDerivedType(tag: DW_TAG_pointer_type, baseType: !16, size: 64, align: 32)

llvm/trunk/test/CodeGen/AMDGPU/hsa-metadata-kernel-debug-props.ll

-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx700 -filetype=obj -o - < %s | llvm-readobj -elf-output-style=GNU -notes | FileCheck --check-prefix=CHECK --check-prefix=GFX700 --check-prefix=NOTES %s
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx800 -filetype=obj -o - < %s | llvm-readobj -elf-output-style=GNU -notes | FileCheck --check-prefix=CHECK --check-prefix=GFX800 --check-prefix=NOTES %s
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -filetype=obj -o - < %s | llvm-readobj -elf-output-style=GNU -notes | FileCheck --check-prefix=CHECK --check-prefix=GFX900 --check-prefix=NOTES %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa-amdgiz -mcpu=gfx700 -filetype=obj -o - < %s | llvm-readobj -elf-output-style=GNU -notes | FileCheck --check-prefix=CHECK --check-prefix=GFX700 --check-prefix=NOTES %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa-amdgiz -mcpu=gfx800 -filetype=obj -o - < %s | llvm-readobj -elf-output-style=GNU -notes | FileCheck --check-prefix=CHECK --check-prefix=GFX800 --check-prefix=NOTES %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa-amdgiz -mcpu=gfx900 -filetype=obj -o - < %s | llvm-readobj -elf-output-style=GNU -notes | FileCheck --check-prefix=CHECK --check-prefix=GFX900 --check-prefix=NOTES %s
+target datalayout = "A5"
 
 declare void @llvm.dbg.declare(metadata, metadata, metadata)
 
 ; CHECK: ---
 ; CHECK:  Version: [ 1, 0 ]
 
 ; CHECK:  Kernels:
 ; CHECK:    - Name:       test
 ; CHECK:      SymbolName: 'test@kd'
 ; CHECK:      DebugProps:
 ; CHECK:        DebuggerABIVersion:                [ 1, 0 ]
 ; CHECK:        ReservedNumVGPRs:                  4
 ; GFX700:       ReservedFirstVGPR:                 8
 ; GFX800:       ReservedFirstVGPR:                 8
 ; GFX900:       ReservedFirstVGPR:                 11
 ; CHECK:        PrivateSegmentBufferSGPR:          0
 ; CHECK:        WavefrontPrivateSegmentOffsetSGPR: 11
 define amdgpu_kernel void @test(i32 addrspace(1)* %A) #0 !dbg !7 !kernel_arg_addr_space !12 !kernel_arg_access_qual !13 !kernel_arg_type !14 !kernel_arg_base_type !14 !kernel_arg_type_qual !15 {
 entry:
-  %A.addr = alloca i32 addrspace(1)*, align 4
-  store i32 addrspace(1)* %A, i32 addrspace(1)** %A.addr, align 4
-  call void @llvm.dbg.declare(metadata i32 addrspace(1)** %A.addr, metadata !16, metadata !17), !dbg !18
-  %0 = load i32 addrspace(1)*, i32 addrspace(1)** %A.addr, align 4, !dbg !19
+  %A.addr = alloca i32 addrspace(1)*, align 4, addrspace(5)
+  store i32 addrspace(1)* %A, i32 addrspace(1)* addrspace(5)* %A.addr, align 4
+  call void @llvm.dbg.declare(metadata i32 addrspace(1)* addrspace(5)* %A.addr, metadata !16, metadata !17), !dbg !18
+  %0 = load i32 addrspace(1)*, i32 addrspace(1)* addrspace(5)* %A.addr, align 4, !dbg !19
   %arrayidx = getelementptr inbounds i32, i32 addrspace(1)* %0, i64 0, !dbg !19
   store i32 777, i32 addrspace(1)* %arrayidx, align 4, !dbg !20
-  %1 = load i32 addrspace(1)*, i32 addrspace(1)** %A.addr, align 4, !dbg !21
+  %1 = load i32 addrspace(1)*, i32 addrspace(1)* addrspace(5)* %A.addr, align 4, !dbg !21
   %arrayidx1 = getelementptr inbounds i32, i32 addrspace(1)* %1, i64 1, !dbg !21
   store i32 888, i32 addrspace(1)* %arrayidx1, align 4, !dbg !22
-  %2 = load i32 addrspace(1)*, i32 addrspace(1)** %A.addr, align 4, !dbg !23
+  %2 = load i32 addrspace(1)*, i32 addrspace(1)* addrspace(5)* %A.addr, align 4, !dbg !23
   %arrayidx2 = getelementptr inbounds i32, i32 addrspace(1)* %2, i64 2, !dbg !23
   store i32 999, i32 addrspace(1)* %arrayidx2, align 4, !dbg !24
   ret void, !dbg !25
 }
 
 attributes #0 = { noinline nounwind "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="gfx800" "target-features"="+16-bit-insts,+amdgpu-debugger-emit-prologue,+amdgpu-debugger-insert-nops,+amdgpu-debugger-reserve-regs,+dpp,+fp64-fp16-denormals,+s-memrealtime,-fp32-denormals" "unsafe-fp-math"="false" "use-soft-float"="false" }
 
 !llvm.dbg.cu = !{!0}

 !6 = !{!"clang version 5.0.0"}
 !7 = distinct !DISubprogram(name: "test", scope: !1, file: !1, line: 1, type: !8, isLocal: false, isDefinition: true, scopeLine: 1, flags: DIFlagPrototyped, isOptimized: false, unit: !0, variables: !2)
 !8 = !DISubroutineType(types: !9)
 !9 = !{null, !10}
 !10 = !DIDerivedType(tag: DW_TAG_pointer_type, baseType: !11, size: 64)
 !11 = !DIBasicType(name: "int", size: 32, encoding: DW_ATE_signed)
 !12 = !{i32 1}
 !13 = !{!"none"}
-!14 = !{!"int*"}
+!14 = !{!"int addrspace(5)*"}
 !15 = !{!""}
 !16 = !DILocalVariable(name: "A", arg: 1, scope: !7, file: !1, line: 1, type: !10)
 !17 = !DIExpression(DW_OP_constu, 1, DW_OP_swap, DW_OP_xderef)
 !18 = !DILocation(line: 1, column: 30, scope: !7)
 !19 = !DILocation(line: 2, column: 3, scope: !7)
 !20 = !DILocation(line: 2, column: 8, scope: !7)
 !21 = !DILocation(line: 3, column: 3, scope: !7)
 !22 = !DILocation(line: 3, column: 8, scope: !7)
 !23 = !DILocation(line: 4, column: 3, scope: !7)
 !24 = !DILocation(line: 4, column: 8, scope: !7)
 !25 = !DILocation(line: 5, column: 1, scope: !7)