Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Support debug records (introduced in LLVM 19) #277

Open
RyanGlScott opened this issue Sep 25, 2024 · 0 comments
Open

Support debug records (introduced in LLVM 19) #277

RyanGlScott opened this issue Sep 25, 2024 · 0 comments
Labels
llvm/19.0

Comments

@RyanGlScott
Copy link
Contributor

LLVM 19 migrates away from the old style of recording debug information using llvm.dbg.* intrinsics to instead using debug records, as described in https://llvm.org/docs/RemoveDIsDebugInfo.html. This migration involved changes to bitcode that we must accommodate on the llvm-pretty-bc-parser side. The relevant LLVM commit is llvm/llvm-project@435d4c1, which introduces five new FUNC_CODEs:

  • FUNC_CODE_DEBUG_RECORD_LABEL (replacing llvm.dbg.label)
  • FUNC_CODE_DEBUG_RECORD_DECLARE (replacing llvm.dbg.declare)
  • FUNC_CODE_DEBUG_RECORD_ASSIGN (replacing llvm.dbg.assign)
  • FUNC_CODE_DEBUG_RECORD_VALUE_SIMPLE and FUNC_CODE_DEBUG_RECORD_VALUE (replacing llvm.dbg.value)

Here are examples of programs that use each of these codes, which llvm-pretty-bc-parser fails to parse:

FUNC_CODE_DEBUG_RECORD_LABEL

Minimal example:

int main(void) {
  goto end;
  end:
    return 0;
}

Compiling this with clang-19.1.0 -g -emit-llvm yields:

; ModuleID = 'test.c'
source_filename = "test.c"
target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128"
target triple = "x86_64-unknown-linux-gnu"

; Function Attrs: noinline nounwind optnone uwtable
define dso_local i32 @main() #0 !dbg !10 {
  %1 = alloca i32, align 4
  store i32 0, ptr %1, align 4
  br label %2, !dbg !14

2:                                                ; preds = %0
    #dbg_label(!15, !16)
  ret i32 0, !dbg !17
}

attributes #0 = { noinline nounwind optnone uwtable "frame-pointer"="all" "min-legal-vector-width"="0" "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="x86-64" "target-features"="+cmov,+cx8,+fxsr,+mmx,+sse,+sse2,+x87" "tune-cpu"="generic" }

!llvm.dbg.cu = !{!0}
!llvm.module.flags = !{!2, !3, !4, !5, !6, !7, !8}
!llvm.ident = !{!9}

!0 = distinct !DICompileUnit(language: DW_LANG_C11, file: !1, producer: "clang version 19.1.0 (/home/runner/work/llvm-project/llvm-project/clang a4bf6cd7cfb1a1421ba92bca9d017b49936c55e4)", isOptimized: false, runtimeVersion: 0, emissionKind: FullDebug, splitDebugInlining: false, nameTableKind: None)
!1 = !DIFile(filename: "test.c", directory: "/home/ryanscott/Documents/Hacking/Haskell/llvm-pretty-bc-parser", checksumkind: CSK_MD5, checksum: "ed19ff02bb0c1e0c6ad40fcff21451b3")
!2 = !{i32 7, !"Dwarf Version", i32 5}
!3 = !{i32 2, !"Debug Info Version", i32 3}
!4 = !{i32 1, !"wchar_size", i32 4}
!5 = !{i32 8, !"PIC Level", i32 2}
!6 = !{i32 7, !"PIE Level", i32 2}
!7 = !{i32 7, !"uwtable", i32 2}
!8 = !{i32 7, !"frame-pointer", i32 2}
!9 = !{!"clang version 19.1.0 (/home/runner/work/llvm-project/llvm-project/clang a4bf6cd7cfb1a1421ba92bca9d017b49936c55e4)"}
!10 = distinct !DISubprogram(name: "main", scope: !1, file: !1, line: 1, type: !11, scopeLine: 1, flags: DIFlagPrototyped, spFlags: DISPFlagDefinition, unit: !0)
!11 = !DISubroutineType(types: !12)
!12 = !{!13}
!13 = !DIBasicType(name: "int", size: 32, encoding: DW_ATE_signed)
!14 = !DILocation(line: 2, column: 3, scope: !10)
!15 = !DILabel(scope: !10, name: "end", file: !1, line: 3)
!16 = !DILocation(line: 3, column: 3, scope: !10)
!17 = !DILocation(line: 4, column: 5, scope: !10)
λ> parseBitCodeFromFile "test.bc"
Left (Error {errContext = ["@main","FUNCTION_BLOCK","FUNCTION_BLOCK_ID","value symbol table","MODULE_BLOCK","Bitstream"], errMessage = "instruction code 65 is unknown"})

FUNC_CODE_DEBUG_RECORD_DECLARE

Minimal example:

int main(void) {
  int x;
  return 0;
}

Compiling this with clang-19.1.0 -g -emit-llvm yields:

; ModuleID = 'test.c'
source_filename = "test.c"
target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128"
target triple = "x86_64-unknown-linux-gnu"

; Function Attrs: noinline nounwind optnone uwtable
define dso_local i32 @main() #0 !dbg !10 {
  %1 = alloca i32, align 4
  %2 = alloca i32, align 4
  store i32 0, ptr %1, align 4
    #dbg_declare(ptr %2, !15, !DIExpression(), !16)
  ret i32 0, !dbg !17
}

attributes #0 = { noinline nounwind optnone uwtable "frame-pointer"="all" "min-legal-vector-width"="0" "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="x86-64" "target-features"="+cmov,+cx8,+fxsr,+mmx,+sse,+sse2,+x87" "tune-cpu"="generic" }

!llvm.dbg.cu = !{!0}
!llvm.module.flags = !{!2, !3, !4, !5, !6, !7, !8}
!llvm.ident = !{!9}

!0 = distinct !DICompileUnit(language: DW_LANG_C11, file: !1, producer: "clang version 19.1.0 (/home/runner/work/llvm-project/llvm-project/clang a4bf6cd7cfb1a1421ba92bca9d017b49936c55e4)", isOptimized: false, runtimeVersion: 0, emissionKind: FullDebug, splitDebugInlining: false, nameTableKind: None)
!1 = !DIFile(filename: "test.c", directory: "/home/ryanscott/Documents/Hacking/Haskell/llvm-pretty-bc-parser", checksumkind: CSK_MD5, checksum: "9bb198fa990a9fe479335c87463915aa")
!2 = !{i32 7, !"Dwarf Version", i32 5}
!3 = !{i32 2, !"Debug Info Version", i32 3}
!4 = !{i32 1, !"wchar_size", i32 4}
!5 = !{i32 8, !"PIC Level", i32 2}
!6 = !{i32 7, !"PIE Level", i32 2}
!7 = !{i32 7, !"uwtable", i32 2}
!8 = !{i32 7, !"frame-pointer", i32 2}
!9 = !{!"clang version 19.1.0 (/home/runner/work/llvm-project/llvm-project/clang a4bf6cd7cfb1a1421ba92bca9d017b49936c55e4)"}
!10 = distinct !DISubprogram(name: "main", scope: !1, file: !1, line: 1, type: !11, scopeLine: 1, flags: DIFlagPrototyped, spFlags: DISPFlagDefinition, unit: !0, retainedNodes: !14)
!11 = !DISubroutineType(types: !12)
!12 = !{!13}
!13 = !DIBasicType(name: "int", size: 32, encoding: DW_ATE_signed)
!14 = !{}
!15 = !DILocalVariable(name: "x", scope: !10, file: !1, line: 2, type: !13)
!16 = !DILocation(line: 2, column: 7, scope: !10)
!17 = !DILocation(line: 3, column: 3, scope: !10)
λ> parseBitCodeFromFile "test.bc"
Left (Error {errContext = ["@main","FUNCTION_BLOCK","FUNCTION_BLOCK_ID","value symbol table","MODULE_BLOCK","Bitstream"], errMessage = "instruction code 62 is unknown"})

FUNC_CODE_DEBUG_RECORD_ASSIGN

Adapting the example from GaloisInc/crucible#1204:

__attribute__((noinline)) int foo(int x[2]) {
  return x[0];
}

int main(void) {
  int x[2] = { 0, 0 };
  return foo(x);
}

Compiling this with clang-19.1.0 -g -emit-llvm -O yields:

; ModuleID = 'test.c'
source_filename = "test.c"
target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128"
target triple = "x86_64-unknown-linux-gnu"

; Function Attrs: mustprogress nofree noinline norecurse nosync nounwind willreturn memory(argmem: read) uwtable
define dso_local i32 @foo(ptr nocapture noundef readonly %0) local_unnamed_addr #0 !dbg !10 {
    #dbg_value(ptr %0, !16, !DIExpression(), !17)
  %2 = load i32, ptr %0, align 4, !dbg !18, !tbaa !19
  ret i32 %2, !dbg !23
}

; Function Attrs: mustprogress nofree norecurse nosync nounwind willreturn memory(none) uwtable
define dso_local i32 @main() local_unnamed_addr #1 !dbg !24 {
  %1 = alloca [2 x i32], align 8, !DIAssignID !32
    #dbg_assign(i1 undef, !28, !DIExpression(), !32, ptr %1, !DIExpression(), !33)
  call void @llvm.lifetime.start.p0(i64 8, ptr nonnull %1) #3, !dbg !34
  store i64 0, ptr %1, align 8, !dbg !35, !DIAssignID !36
    #dbg_assign(i64 0, !28, !DIExpression(), !36, ptr %1, !DIExpression(), !33)
  %2 = call i32 @foo(ptr noundef nonnull %1), !dbg !37
  call void @llvm.lifetime.end.p0(i64 8, ptr nonnull %1) #3, !dbg !38
  ret i32 %2, !dbg !39
}

; Function Attrs: mustprogress nocallback nofree nosync nounwind willreturn memory(argmem: readwrite)
declare void @llvm.lifetime.start.p0(i64 immarg, ptr nocapture) #2

; Function Attrs: mustprogress nocallback nofree nosync nounwind willreturn memory(argmem: readwrite)
declare void @llvm.lifetime.end.p0(i64 immarg, ptr nocapture) #2

attributes #0 = { mustprogress nofree noinline norecurse nosync nounwind willreturn memory(argmem: read) uwtable "min-legal-vector-width"="0" "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="x86-64" "target-features"="+cmov,+cx8,+fxsr,+mmx,+sse,+sse2,+x87" "tune-cpu"="generic" }
attributes #1 = { mustprogress nofree norecurse nosync nounwind willreturn memory(none) uwtable "min-legal-vector-width"="0" "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="x86-64" "target-features"="+cmov,+cx8,+fxsr,+mmx,+sse,+sse2,+x87" "tune-cpu"="generic" }
attributes #2 = { mustprogress nocallback nofree nosync nounwind willreturn memory(argmem: readwrite) }
attributes #3 = { nounwind }

!llvm.dbg.cu = !{!0}
!llvm.module.flags = !{!2, !3, !4, !5, !6, !7, !8}
!llvm.ident = !{!9}

!0 = distinct !DICompileUnit(language: DW_LANG_C11, file: !1, producer: "clang version 19.1.0 (/home/runner/work/llvm-project/llvm-project/clang a4bf6cd7cfb1a1421ba92bca9d017b49936c55e4)", isOptimized: true, runtimeVersion: 0, emissionKind: FullDebug, splitDebugInlining: false, nameTableKind: None)
!1 = !DIFile(filename: "test.c", directory: "/home/ryanscott/Documents/Hacking/Haskell/llvm-pretty-bc-parser", checksumkind: CSK_MD5, checksum: "76d1b0749f7b8b079c2db3c06c693ee2")
!2 = !{i32 7, !"Dwarf Version", i32 5}
!3 = !{i32 2, !"Debug Info Version", i32 3}
!4 = !{i32 1, !"wchar_size", i32 4}
!5 = !{i32 8, !"PIC Level", i32 2}
!6 = !{i32 7, !"PIE Level", i32 2}
!7 = !{i32 7, !"uwtable", i32 2}
!8 = !{i32 7, !"debug-info-assignment-tracking", i1 true}
!9 = !{!"clang version 19.1.0 (/home/runner/work/llvm-project/llvm-project/clang a4bf6cd7cfb1a1421ba92bca9d017b49936c55e4)"}
!10 = distinct !DISubprogram(name: "foo", scope: !1, file: !1, line: 1, type: !11, scopeLine: 1, flags: DIFlagPrototyped | DIFlagAllCallsDescribed, spFlags: DISPFlagDefinition | DISPFlagOptimized, unit: !0, retainedNodes: !15)
!11 = !DISubroutineType(types: !12)
!12 = !{!13, !14}
!13 = !DIBasicType(name: "int", size: 32, encoding: DW_ATE_signed)
!14 = !DIDerivedType(tag: DW_TAG_pointer_type, baseType: !13, size: 64)
!15 = !{!16}
!16 = !DILocalVariable(name: "x", arg: 1, scope: !10, file: !1, line: 1, type: !14)
!17 = !DILocation(line: 0, scope: !10)
!18 = !DILocation(line: 2, column: 10, scope: !10)
!19 = !{!20, !20, i64 0}
!20 = !{!"int", !21, i64 0}
!21 = !{!"omnipotent char", !22, i64 0}
!22 = !{!"Simple C/C++ TBAA"}
!23 = !DILocation(line: 2, column: 3, scope: !10)
!24 = distinct !DISubprogram(name: "main", scope: !1, file: !1, line: 5, type: !25, scopeLine: 5, flags: DIFlagPrototyped | DIFlagAllCallsDescribed, spFlags: DISPFlagDefinition | DISPFlagOptimized, unit: !0, retainedNodes: !27)
!25 = !DISubroutineType(types: !26)
!26 = !{!13}
!27 = !{!28}
!28 = !DILocalVariable(name: "x", scope: !24, file: !1, line: 6, type: !29)
!29 = !DICompositeType(tag: DW_TAG_array_type, baseType: !13, size: 64, elements: !30)
!30 = !{!31}
!31 = !DISubrange(count: 2)
!32 = distinct !DIAssignID()
!33 = !DILocation(line: 0, scope: !24)
!34 = !DILocation(line: 6, column: 3, scope: !24)
!35 = !DILocation(line: 6, column: 7, scope: !24)
!36 = distinct !DIAssignID()
!37 = !DILocation(line: 7, column: 10, scope: !24)
!38 = !DILocation(line: 8, column: 1, scope: !24)
!39 = !DILocation(line: 7, column: 3, scope: !24)

FUNC_CODE_DEBUG_RECORD_VALUE_SIMPLE and FUNC_CODE_DEBUG_RECORD_VALUE

Minimal example of a program which compiles down to FUNC_CODE_DEBUG_RECORD_VALUE_SIMPLE:

int f(int x) {
  return x;
}

int main(void) {
  return 0;
}

Compiling this with clang-19.1.0 -g -emit-llvm -O yields:

; ModuleID = 'test.c'
source_filename = "test.c"
target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128"
target triple = "x86_64-unknown-linux-gnu"

; Function Attrs: mustprogress nofree norecurse nosync nounwind willreturn memory(none) uwtable
define dso_local noundef i32 @f(i32 noundef returned %0) local_unnamed_addr #0 !dbg !10 {
    #dbg_value(i32 %0, !15, !DIExpression(), !16)
  ret i32 %0, !dbg !17
}

; Function Attrs: mustprogress nofree norecurse nosync nounwind willreturn memory(none) uwtable
define dso_local noundef i32 @main() local_unnamed_addr #0 !dbg !18 {
  ret i32 0, !dbg !21
}

attributes #0 = { mustprogress nofree norecurse nosync nounwind willreturn memory(none) uwtable "min-legal-vector-width"="0" "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="x86-64" "target-features"="+cmov,+cx8,+fxsr,+mmx,+sse,+sse2,+x87" "tune-cpu"="generic" }

!llvm.dbg.cu = !{!0}
!llvm.module.flags = !{!2, !3, !4, !5, !6, !7, !8}
!llvm.ident = !{!9}

!0 = distinct !DICompileUnit(language: DW_LANG_C11, file: !1, producer: "clang version 19.1.0 (/home/runner/work/llvm-project/llvm-project/clang a4bf6cd7cfb1a1421ba92bca9d017b49936c55e4)", isOptimized: true, runtimeVersion: 0, emissionKind: FullDebug, splitDebugInlining: false, nameTableKind: None)
!1 = !DIFile(filename: "test.c", directory: "/home/ryanscott/Documents/Hacking/Haskell/llvm-pretty-bc-parser", checksumkind: CSK_MD5, checksum: "3dfad31b2954451c6a3a5da41ad424bf")
!2 = !{i32 7, !"Dwarf Version", i32 5}
!3 = !{i32 2, !"Debug Info Version", i32 3}
!4 = !{i32 1, !"wchar_size", i32 4}
!5 = !{i32 8, !"PIC Level", i32 2}
!6 = !{i32 7, !"PIE Level", i32 2}
!7 = !{i32 7, !"uwtable", i32 2}
!8 = !{i32 7, !"debug-info-assignment-tracking", i1 true}
!9 = !{!"clang version 19.1.0 (/home/runner/work/llvm-project/llvm-project/clang a4bf6cd7cfb1a1421ba92bca9d017b49936c55e4)"}
!10 = distinct !DISubprogram(name: "f", scope: !1, file: !1, line: 1, type: !11, scopeLine: 1, flags: DIFlagPrototyped | DIFlagAllCallsDescribed, spFlags: DISPFlagDefinition | DISPFlagOptimized, unit: !0, retainedNodes: !14)
!11 = !DISubroutineType(types: !12)
!12 = !{!13, !13}
!13 = !DIBasicType(name: "int", size: 32, encoding: DW_ATE_signed)
!14 = !{!15}
!15 = !DILocalVariable(name: "x", arg: 1, scope: !10, file: !1, line: 1, type: !13)
!16 = !DILocation(line: 0, scope: !10)
!17 = !DILocation(line: 2, column: 3, scope: !10)
!18 = distinct !DISubprogram(name: "main", scope: !1, file: !1, line: 5, type: !19, scopeLine: 5, flags: DIFlagPrototyped | DIFlagAllCallsDescribed, spFlags: DISPFlagDefinition | DISPFlagOptimized, unit: !0)
!19 = !DISubroutineType(types: !20)
!20 = !{!13}
!21 = !DILocation(line: 6, column: 3, scope: !18)
λ> parseBitCodeFromFile "test.bc"
Left (Error {errContext = ["@f","FUNCTION_BLOCK","FUNCTION_BLOCK_ID","value symbol table","MODULE_BLOCK","Bitstream"], errMessage = "instruction code 64 is unknown"})

FUNC_CODE_DEBUG_RECORD_VALUE (instruction code 61) is similar, although I'm not entirely sure how to make Clang emit such a function code.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
llvm/19.0
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@RyanGlScott