1 ================================
2 Source Level Debugging with LLVM
3 ================================
11 This document is the central repository for all information pertaining to debug
12 information in LLVM. It describes the :ref:`actual format that the LLVM debug
13 information takes <format>`, which is useful for those interested in creating
14 front-ends or dealing directly with the information. Further, this document
15 provides specific examples of what debug information for C/C++ looks like.
17 Philosophy behind LLVM debugging information
18 --------------------------------------------
20 The idea of the LLVM debugging information is to capture how the important
21 pieces of the source-language's Abstract Syntax Tree map onto LLVM code.
22 Several design aspects have shaped the solution that appears here. The
25 * Debugging information should have very little impact on the rest of the
26 compiler. No transformations, analyses, or code generators should need to
27 be modified because of debugging information.
29 * LLVM optimizations should interact in :ref:`well-defined and easily described
30 ways <intro_debugopt>` with the debugging information.
32 * Because LLVM is designed to support arbitrary programming languages,
33 LLVM-to-LLVM tools should not need to know anything about the semantics of
34 the source-level-language.
36 * Source-level languages are often **widely** different from one another.
37 LLVM should not put any restrictions of the flavor of the source-language,
38 and the debugging information should work with any language.
40 * With code generator support, it should be possible to use an LLVM compiler
41 to compile a program to native machine code and standard debugging
42 formats. This allows compatibility with traditional machine-code level
43 debuggers, like GDB or DBX.
45 The approach used by the LLVM implementation is to use a small set of
46 :ref:`intrinsic functions <format_common_intrinsics>` to define a mapping
47 between LLVM program objects and the source-level objects. The description of
48 the source-level program is maintained in LLVM metadata in an
49 :ref:`implementation-defined format <ccxx_frontend>` (the C/C++ front-end
50 currently uses working draft 7 of the `DWARF 3 standard
51 <http://www.eagercon.com/dwarf/dwarf3std.htm>`_).
53 When a program is being debugged, a debugger interacts with the user and turns
54 the stored debug information into source-language specific information. As
55 such, a debugger must be aware of the source-language, and is thus tied to a
56 specific language or family of languages.
58 Debug information consumers
59 ---------------------------
61 The role of debug information is to provide meta information normally stripped
62 away during the compilation process. This meta information provides an LLVM
63 user a relationship between generated code and the original program source
66 Currently, debug information is consumed by DwarfDebug to produce dwarf
67 information used by the gdb debugger. Other targets could use the same
68 information to produce stabs or other debug forms.
70 It would also be reasonable to use debug information to feed profiling tools
71 for analysis of generated code, or, tools for reconstructing the original
72 source from generated code.
74 TODO - expound a bit more.
78 Debugging optimized code
79 ------------------------
81 An extremely high priority of LLVM debugging information is to make it interact
82 well with optimizations and analysis. In particular, the LLVM debug
83 information provides the following guarantees:
85 * LLVM debug information **always provides information to accurately read
86 the source-level state of the program**, regardless of which LLVM
87 optimizations have been run, and without any modification to the
88 optimizations themselves. However, some optimizations may impact the
89 ability to modify the current state of the program with a debugger, such
90 as setting program variables, or calling functions that have been
93 * As desired, LLVM optimizations can be upgraded to be aware of the LLVM
94 debugging information, allowing them to update the debugging information
95 as they perform aggressive optimizations. This means that, with effort,
96 the LLVM optimizers could optimize debug code just as well as non-debug
99 * LLVM debug information does not prevent optimizations from
100 happening (for example inlining, basic block reordering/merging/cleanup,
101 tail duplication, etc).
103 * LLVM debug information is automatically optimized along with the rest of
104 the program, using existing facilities. For example, duplicate
105 information is automatically merged by the linker, and unused information
106 is automatically removed.
108 Basically, the debug information allows you to compile a program with
109 "``-O0 -g``" and get full debug information, allowing you to arbitrarily modify
110 the program as it executes from a debugger. Compiling a program with
111 "``-O3 -g``" gives you full debug information that is always available and
112 accurate for reading (e.g., you get accurate stack traces despite tail call
113 elimination and inlining), but you might lose the ability to modify the program
114 and call functions where were optimized out of the program, or inlined away
117 :ref:`LLVM test suite <test-suite-quickstart>` provides a framework to test
118 optimizer's handling of debugging information. It can be run like this:
122 % cd llvm/projects/test-suite/MultiSource/Benchmarks # or some other level
125 This will test impact of debugging information on optimization passes. If
126 debugging information influences optimization passes then it will be reported
127 as a failure. See :doc:`TestingGuide` for more information on LLVM test
128 infrastructure and how to run various tests.
132 Debugging information format
133 ============================
135 LLVM debugging information has been carefully designed to make it possible for
136 the optimizer to optimize the program and debugging information without
137 necessarily having to know anything about debugging information. In
138 particular, the use of metadata avoids duplicated debugging information from
139 the beginning, and the global dead code elimination pass automatically deletes
140 debugging information for a function if it decides to delete the function.
142 To do this, most of the debugging information (descriptors for types,
143 variables, functions, source files, etc) is inserted by the language front-end
144 in the form of LLVM metadata.
146 Debug information is designed to be agnostic about the target debugger and
147 debugging information representation (e.g. DWARF/Stabs/etc). It uses a generic
148 pass to decode the information that represents variables, types, functions,
149 namespaces, etc: this allows for arbitrary source-language semantics and
150 type-systems to be used, as long as there is a module written for the target
151 debugger to interpret the information.
153 To provide basic functionality, the LLVM debugger does have to make some
154 assumptions about the source-level language being debugged, though it keeps
155 these to a minimum. The only common features that the LLVM debugger assumes
156 exist are :ref:`source files <format_files>`, and :ref:`program objects
157 <format_global_variables>`. These abstract objects are used by a debugger to
158 form stack traces, show information about local variables, etc.
160 This section of the documentation first describes the representation aspects
161 common to any source-language. :ref:`ccxx_frontend` describes the data layout
162 conventions used by the C and C++ front-ends.
164 Debug information descriptors
165 -----------------------------
167 In consideration of the complexity and volume of debug information, LLVM
168 provides a specification for well formed debug descriptors.
170 Consumers of LLVM debug information expect the descriptors for program objects
171 to start in a canonical format, but the descriptors can include additional
172 information appended at the end that is source-language specific. All debugging
173 information objects start with a tag to indicate what type of object it is.
174 The source-language is allowed to define its own objects, by using unreserved
175 tag numbers. We recommend using with tags in the range 0x1000 through 0x2000
176 (there is a defined ``enum DW_TAG_user_base = 0x1000``.)
178 The fields of debug descriptors used internally by LLVM are restricted to only
179 the simple data types ``i32``, ``i1``, ``float``, ``double``, ``mdstring`` and
189 Most of the string and integer fields in descriptors are packed into a single,
190 null-separated ``mdstring``. The first field of the header is always an
191 ``i32`` containing the DWARF tag value identifying the content of the
194 For clarity of definition in this document, these header fields are described
195 below split inside an imaginary ``DIHeader`` construct. This is invalid
196 assembly syntax. In valid IR, these fields are stringified and concatenated,
197 separated by ``\00``.
199 The details of the various descriptors follow.
201 Compile unit descriptors
202 ^^^^^^^^^^^^^^^^^^^^^^^^
208 i32, ;; Tag = 17 (DW_TAG_compile_unit)
209 i32, ;; DWARF language identifier (ex. DW_LANG_C89)
210 mdstring, ;; Producer (ex. "4.0.1 LLVM (LLVM research group)")
211 i1, ;; True if this is optimized.
213 i32, ;; Runtime version
214 mdstring, ;; Split debug filename
215 i32 ;; Debug info emission kind (1 = Full Debug Info, 2 = Line Tables Only)
217 metadata, ;; Source directory (including trailing slash) & file pair
218 metadata, ;; List of enums types
219 metadata, ;; List of retained types
220 metadata, ;; List of subprograms
221 metadata, ;; List of global variables
222 metadata ;; List of imported entities
225 These descriptors contain a source language ID for the file (we use the DWARF
226 3.0 ID numbers, such as ``DW_LANG_C89``, ``DW_LANG_C_plus_plus``,
227 ``DW_LANG_Cobol74``, etc), a reference to a metadata node containing a pair of
228 strings for the source file name and the working directory, as well as an
229 identifier string for the compiler that produced it.
231 Compile unit descriptors provide the root context for objects declared in a
232 specific compilation unit. File descriptors are defined using this context.
233 These descriptors are collected by a named metadata ``!llvm.dbg.cu``. They
234 keep track of subprograms, global variables, type information, and imported
235 entities (declarations and namespaces).
246 i32 ;; Tag = 41 (DW_TAG_file_type)
248 metadata ;; Source directory (including trailing slash) & file pair
251 These descriptors contain information for a file. Global variables and top
252 level functions would be defined using this context. File descriptors also
253 provide context for source line correspondence.
255 Each input file is encoded as a separate file descriptor in LLVM debugging
258 .. _format_global_variables:
260 Global variable descriptors
261 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
267 i32, ;; Tag = 52 (DW_TAG_variable)
269 mdstring, ;; Display name (fully qualified C++ name)
270 mdstring, ;; MIPS linkage name (for C++)
271 i32, ;; Line number where defined
272 i1, ;; True if the global is local to compile unit (static)
273 i1 ;; True if the global is defined in the compile unit (not extern)
275 metadata, ;; Reference to context descriptor
276 metadata, ;; Reference to file where defined
277 metadata, ;; Reference to type descriptor
278 {}*, ;; Reference to the global variable
279 metadata, ;; The static member declaration, if any
282 These descriptors provide debug information about global variables. They
283 provide details such as name, type and where the variable is defined. All
284 global variables are collected inside the named metadata ``!llvm.dbg.cu``.
286 .. _format_subprograms:
288 Subprogram descriptors
289 ^^^^^^^^^^^^^^^^^^^^^^
295 i32, ;; Tag = 46 (DW_TAG_subprogram)
297 mdstring, ;; Display name (fully qualified C++ name)
298 mdstring, ;; MIPS linkage name (for C++)
299 i32, ;; Line number where defined
300 i1, ;; True if the global is local to compile unit (static)
301 i1, ;; True if the global is defined in the compile unit (not extern)
302 i32, ;; Virtuality, e.g. dwarf::DW_VIRTUALITY__virtual
303 i32, ;; Index into a virtual function
304 i32, ;; Flags - Artificial, Private, Protected, Explicit, Prototyped.
306 i32 ;; Line number where the scope of the subprogram begins
308 metadata, ;; Source directory (including trailing slash) & file pair
309 metadata, ;; Reference to context descriptor
310 metadata, ;; Reference to type descriptor
311 metadata, ;; indicates which base type contains the vtable pointer for the
313 {}*, ;; Reference to the LLVM function
314 metadata, ;; Lists function template parameters
315 metadata, ;; Function declaration descriptor
316 metadata ;; List of function variables
319 These descriptors provide debug information about functions, methods and
320 subprograms. They provide details such as name, return types and the source
321 location where the subprogram is defined.
330 i32, ;; Tag = 11 (DW_TAG_lexical_block)
332 i32, ;; Column number
333 i32 ;; Unique ID to identify blocks from a template function
335 metadata, ;; Source directory (including trailing slash) & file pair
336 metadata ;; Reference to context descriptor
339 This descriptor provides debug information about nested blocks within a
340 subprogram. The line number and column numbers are used to dinstinguish two
341 lexical blocks at same depth.
347 i32, ;; Tag = 11 (DW_TAG_lexical_block)
348 i32 ;; DWARF path discriminator value
350 metadata, ;; Source directory (including trailing slash) & file pair
351 metadata ;; Reference to the scope we're annotating with a file change
354 This descriptor provides a wrapper around a lexical scope to handle file
355 changes in the middle of a lexical block.
357 .. _format_basic_type:
359 Basic type descriptors
360 ^^^^^^^^^^^^^^^^^^^^^^
366 i32, ;; Tag = 36 (DW_TAG_base_type)
367 mdstring, ;; Name (may be "" for anonymous types)
368 i32, ;; Line number where defined (may be 0)
370 i64, ;; Alignment in bits
371 i64, ;; Offset in bits
373 i32 ;; DWARF type encoding
375 metadata, ;; Source directory (including trailing slash) & file pair (may be null)
376 metadata ;; Reference to context
379 These descriptors define primitive types used in the code. Example ``int``,
380 ``bool`` and ``float``. The context provides the scope of the type, which is
381 usually the top level. Since basic types are not usually user defined the
382 context and line number can be left as NULL and 0. The size, alignment and
383 offset are expressed in bits and can be 64 bit values. The alignment is used
384 to round the offset when embedded in a :ref:`composite type
385 <format_composite_type>` (example to keep float doubles on 64 bit boundaries).
386 The offset is the bit offset if embedded in a :ref:`composite type
387 <format_composite_type>`.
389 The type encoding provides the details of the type. The values are typically
390 one of the following:
398 DW_ATE_signed_char = 6
400 DW_ATE_unsigned_char = 8
402 .. _format_derived_type:
404 Derived type descriptors
405 ^^^^^^^^^^^^^^^^^^^^^^^^
411 i32, ;; Tag (see below)
412 mdstring, ;; Name (may be "" for anonymous types)
413 i32, ;; Line number where defined (may be 0)
415 i64, ;; Alignment in bits
416 i64, ;; Offset in bits
417 i32 ;; Flags to encode attributes, e.g. private
419 metadata, ;; Source directory (including trailing slash) & file pair (may be null)
420 metadata, ;; Reference to context
421 metadata, ;; Reference to type derived from
422 metadata ;; (optional) Objective C property node
425 These descriptors are used to define types derived from other types. The value
426 of the tag varies depending on the meaning. The following are possible tag
431 DW_TAG_formal_parameter = 5
433 DW_TAG_pointer_type = 15
434 DW_TAG_reference_type = 16
436 DW_TAG_ptr_to_member_type = 31
437 DW_TAG_const_type = 38
438 DW_TAG_volatile_type = 53
439 DW_TAG_restrict_type = 55
441 ``DW_TAG_member`` is used to define a member of a :ref:`composite type
442 <format_composite_type>` or :ref:`subprogram <format_subprograms>`. The type
443 of the member is the :ref:`derived type <format_derived_type>`.
444 ``DW_TAG_formal_parameter`` is used to define a member which is a formal
445 argument of a subprogram.
447 ``DW_TAG_typedef`` is used to provide a name for the derived type.
449 ``DW_TAG_pointer_type``, ``DW_TAG_reference_type``, ``DW_TAG_const_type``,
450 ``DW_TAG_volatile_type`` and ``DW_TAG_restrict_type`` are used to qualify the
451 :ref:`derived type <format_derived_type>`.
453 :ref:`Derived type <format_derived_type>` location can be determined from the
454 context and line number. The size, alignment and offset are expressed in bits
455 and can be 64 bit values. The alignment is used to round the offset when
456 embedded in a :ref:`composite type <format_composite_type>` (example to keep
457 float doubles on 64 bit boundaries.) The offset is the bit offset if embedded
458 in a :ref:`composite type <format_composite_type>`.
460 Note that the ``void *`` type is expressed as a type derived from NULL.
462 .. _format_composite_type:
464 Composite type descriptors
465 ^^^^^^^^^^^^^^^^^^^^^^^^^^
471 i32, ;; Tag (see below)
472 mdstring, ;; Name (may be "" for anonymous types)
473 i32, ;; Line number where defined (may be 0)
475 i64, ;; Alignment in bits
476 i64, ;; Offset in bits
478 i32 ;; Runtime languages
480 metadata, ;; Source directory (including trailing slash) & file pair (may be null)
481 metadata, ;; Reference to context
482 metadata, ;; Reference to type derived from
483 metadata, ;; Reference to array of member descriptors
484 metadata, ;; Base type containing the vtable pointer for this type
485 metadata, ;; Template parameters
486 mdstring ;; A unique identifier for type uniquing purpose (may be null)
489 These descriptors are used to define types that are composed of 0 or more
490 elements. The value of the tag varies depending on the meaning. The following
491 are possible tag values:
495 DW_TAG_array_type = 1
496 DW_TAG_enumeration_type = 4
497 DW_TAG_structure_type = 19
498 DW_TAG_union_type = 23
499 DW_TAG_subroutine_type = 21
500 DW_TAG_inheritance = 28
502 The vector flag indicates that an array type is a native packed vector.
504 The members of array types (tag = ``DW_TAG_array_type``) are
505 :ref:`subrange descriptors <format_subrange>`, each
506 representing the range of subscripts at that level of indexing.
508 The members of enumeration types (tag = ``DW_TAG_enumeration_type``) are
509 :ref:`enumerator descriptors <format_enumerator>`, each representing the
510 definition of enumeration value for the set. All enumeration type descriptors
511 are collected inside the named metadata ``!llvm.dbg.cu``.
513 The members of structure (tag = ``DW_TAG_structure_type``) or union (tag =
514 ``DW_TAG_union_type``) types are any one of the :ref:`basic
515 <format_basic_type>`, :ref:`derived <format_derived_type>` or :ref:`composite
516 <format_composite_type>` type descriptors, each representing a field member of
517 the structure or union.
519 For C++ classes (tag = ``DW_TAG_structure_type``), member descriptors provide
520 information about base classes, static members and member functions. If a
521 member is a :ref:`derived type descriptor <format_derived_type>` and has a tag
522 of ``DW_TAG_inheritance``, then the type represents a base class. If the member
523 of is a :ref:`global variable descriptor <format_global_variables>` then it
524 represents a static member. And, if the member is a :ref:`subprogram
525 descriptor <format_subprograms>` then it represents a member function. For
526 static members and member functions, ``getName()`` returns the members link or
527 the C++ mangled name. ``getDisplayName()`` the simplied version of the name.
529 The first member of subroutine (tag = ``DW_TAG_subroutine_type``) type elements
530 is the return type for the subroutine. The remaining elements are the formal
531 arguments to the subroutine.
533 :ref:`Composite type <format_composite_type>` location can be determined from
534 the context and line number. The size, alignment and offset are expressed in
535 bits and can be 64 bit values. The alignment is used to round the offset when
536 embedded in a :ref:`composite type <format_composite_type>` (as an example, to
537 keep float doubles on 64 bit boundaries). The offset is the bit offset if
538 embedded in a :ref:`composite type <format_composite_type>`.
549 i32, ;; Tag = 33 (DW_TAG_subrange_type)
555 These descriptors are used to define ranges of array subscripts for an array
556 :ref:`composite type <format_composite_type>`. The low value defines the lower
557 bounds typically zero for C/C++. The high value is the upper bounds. Values
558 are 64 bit. ``High - Low + 1`` is the size of the array. If ``Low > High``
559 the array bounds are not included in generated debugging information.
561 .. _format_enumerator:
563 Enumerator descriptors
564 ^^^^^^^^^^^^^^^^^^^^^^
570 i32, ;; Tag = 40 (DW_TAG_enumerator)
576 These descriptors are used to define members of an enumeration :ref:`composite
577 type <format_composite_type>`, it associates the name to the value.
586 i32, ;; Tag (see below)
588 i32, ;; 24 bit - Line number where defined
589 ;; 8 bit - Argument number. 1 indicates 1st argument.
593 metadata, ;; Reference to file where defined
594 metadata, ;; Reference to the type descriptor
595 metadata ;; (optional) Reference to inline location
598 These descriptors are used to define variables local to a sub program. The
599 value of the tag depends on the usage of the variable:
603 DW_TAG_auto_variable = 256
604 DW_TAG_arg_variable = 257
606 An auto variable is any variable declared in the body of the function. An
607 argument variable is any variable that appears as a formal argument to the
610 The context is either the subprogram or block where the variable is defined.
611 Name the source variable name. Context and line indicate where the variable
612 was defined. Type descriptor defines the declared type of the variable.
619 i32, ;; DW_TAG_expression
623 Complex expressions describe variable storage locations in terms of
624 prefix-notated DWARF expressions. Currently the only supported
625 operators are ``DW_OP_plus``, ``DW_OP_deref``, and ``DW_OP_piece``.
627 The ``DW_OP_piece`` operator is used for (typically larger aggregate)
628 variables that are fragmented across several locations. It takes two
629 i32 arguments, an offset and a size in bytes to describe which piece
630 of the variable is at this location.
633 .. _format_common_intrinsics:
635 Debugger intrinsic functions
636 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
638 LLVM uses several intrinsic functions (name prefixed with "``llvm.dbg``") to
639 provide debug information at various points in generated code.
646 void %llvm.dbg.declare(metadata, metadata)
648 This intrinsic provides information about a local element (e.g., variable).
649 The first argument is metadata holding the alloca for the variable. The second
650 argument is metadata containing a description of the variable.
657 void %llvm.dbg.value(metadata, i64, metadata)
659 This intrinsic provides information when a user source variable is set to a new
660 value. The first argument is the new value (wrapped as metadata). The second
661 argument is the offset in the user source variable where the new value is
662 written. The third argument is metadata containing a description of the user
665 Object lifetimes and scoping
666 ============================
668 In many languages, the local variables in functions can have their lifetimes or
669 scopes limited to a subset of a function. In the C family of languages, for
670 example, variables are only live (readable and writable) within the source
671 block that they are defined in. In functional languages, values are only
672 readable after they have been defined. Though this is a very obvious concept,
673 it is non-trivial to model in LLVM, because it has no notion of scoping in this
674 sense, and does not want to be tied to a language's scoping rules.
676 In order to handle this, the LLVM debug format uses the metadata attached to
677 llvm instructions to encode line number and scoping information. Consider the
678 following C fragment, for example:
692 Compiled to LLVM, this function would be represented like this:
696 define void @foo() #0 {
698 %X = alloca i32, align 4
699 %Y = alloca i32, align 4
700 %Z = alloca i32, align 4
701 call void @llvm.dbg.declare(metadata !{i32* %X}, metadata !10), !dbg !12
702 ; [debug line = 2:7] [debug variable = X]
703 store i32 21, i32* %X, align 4, !dbg !12
704 call void @llvm.dbg.declare(metadata !{i32* %Y}, metadata !13), !dbg !14
705 ; [debug line = 3:7] [debug variable = Y]
706 store i32 22, i32* %Y, align 4, !dbg !14
707 call void @llvm.dbg.declare(metadata !{i32* %Z}, metadata !15), !dbg !17
708 ; [debug line = 5:9] [debug variable = Z]
709 store i32 23, i32* %Z, align 4, !dbg !17
710 %0 = load i32* %X, align 4, !dbg !18
712 store i32 %0, i32* %Z, align 4, !dbg !18
713 %1 = load i32* %Y, align 4, !dbg !19
715 store i32 %1, i32* %X, align 4, !dbg !19
719 ; Function Attrs: nounwind readnone
720 declare void @llvm.dbg.declare(metadata, metadata) #1
722 attributes #0 = { nounwind ssp uwtable "less-precise-fpmad"="false"
723 "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf"
724 "no-infs-fp-math"="false" "no-nans-fp-math"="false"
725 "stack-protector-buffer-size"="8" "unsafe-fp-math"="false"
726 "use-soft-float"="false" }
727 attributes #1 = { nounwind readnone }
730 !llvm.module.flags = !{!8}
733 !0 = metadata !{i32 786449, metadata !1, i32 12,
734 metadata !"clang version 3.4 (trunk 193128) (llvm/trunk 193139)",
735 i1 false, metadata !"", i32 0, metadata !2, metadata !2, metadata !3,
736 metadata !2, metadata !2, metadata !""} ; [ DW_TAG_compile_unit ] \
737 [/private/tmp/foo.c] \
739 !1 = metadata !{metadata !"t.c", metadata !"/private/tmp"}
740 !2 = metadata !{i32 0}
741 !3 = metadata !{metadata !4}
742 !4 = metadata !{i32 786478, metadata !1, metadata !5, metadata !"foo",
743 metadata !"foo", metadata !"", i32 1, metadata !6,
744 i1 false, i1 true, i32 0, i32 0, null, i32 0, i1 false,
745 void ()* @foo, null, null, metadata !2, i32 1}
746 ; [ DW_TAG_subprogram ] [line 1] [def] [foo]
747 !5 = metadata !{i32 786473, metadata !1} ; [ DW_TAG_file_type ] \
749 !6 = metadata !{i32 786453, i32 0, null, metadata !"", i32 0, i64 0, i64 0,
750 i64 0, i32 0, null, metadata !7, i32 0, null, null, null}
751 ; [ DW_TAG_subroutine_type ] \
752 [line 0, size 0, align 0, offset 0] [from ]
753 !7 = metadata !{null}
754 !8 = metadata !{i32 2, metadata !"Dwarf Version", i32 2}
755 !9 = metadata !{metadata !"clang version 3.4 (trunk 193128) (llvm/trunk 193139)"}
756 !10 = metadata !{i32 786688, metadata !4, metadata !"X", metadata !5, i32 2,
757 metadata !11, i32 0, i32 0} ; [ DW_TAG_auto_variable ] [X] \
759 !11 = metadata !{i32 786468, null, null, metadata !"int", i32 0, i64 32,
760 i64 32, i64 0, i32 0, i32 5} ; [ DW_TAG_base_type ] [int] \
761 [line 0, size 32, align 32, offset 0, enc DW_ATE_signed]
762 !12 = metadata !{i32 2, i32 0, metadata !4, null}
763 !13 = metadata !{i32 786688, metadata !4, metadata !"Y", metadata !5, i32 3,
764 metadata !11, i32 0, i32 0} ; [ DW_TAG_auto_variable ] [Y] \
766 !14 = metadata !{i32 3, i32 0, metadata !4, null}
767 !15 = metadata !{i32 786688, metadata !16, metadata !"Z", metadata !5, i32 5,
768 metadata !11, i32 0, i32 0} ; [ DW_TAG_auto_variable ] [Z] \
770 !16 = metadata !{i32 786443, metadata !1, metadata !4, i32 4, i32 0, i32 0} \
771 ; [ DW_TAG_lexical_block ] [/private/tmp/t.c]
772 !17 = metadata !{i32 5, i32 0, metadata !16, null}
773 !18 = metadata !{i32 6, i32 0, metadata !16, null}
774 !19 = metadata !{i32 8, i32 0, metadata !4, null} ; [ DW_TAG_imported_declaration ]
775 !20 = metadata !{i32 9, i32 0, metadata !4, null}
777 This example illustrates a few important details about LLVM debugging
778 information. In particular, it shows how the ``llvm.dbg.declare`` intrinsic and
779 location information, which are attached to an instruction, are applied
780 together to allow a debugger to analyze the relationship between statements,
781 variable definitions, and the code used to implement the function.
785 call void @llvm.dbg.declare(metadata !{i32* %X}, metadata !10), !dbg !12
786 ; [debug line = 2:7] [debug variable = X]
788 The first intrinsic ``%llvm.dbg.declare`` encodes debugging information for the
789 variable ``X``. The metadata ``!dbg !12`` attached to the intrinsic provides
790 scope information for the variable ``X``.
794 !12 = metadata !{i32 2, i32 0, metadata !4, null}
795 !4 = metadata !{i32 786478, metadata !1, metadata !5, metadata !"foo",
796 metadata !"foo", metadata !"", i32 1, metadata !6,
797 i1 false, i1 true, i32 0, i32 0, null, i32 0, i1 false,
798 void ()* @foo, null, null, metadata !2, i32 1}
799 ; [ DW_TAG_subprogram ] [line 1] [def] [foo]
801 Here ``!12`` is metadata providing location information. It has four fields:
802 line number, column number, scope, and original scope. The original scope
803 represents inline location if this instruction is inlined inside a caller, and
804 is null otherwise. In this example, scope is encoded by ``!4``, a
805 :ref:`subprogram descriptor <format_subprograms>`. This way the location
806 information attached to the intrinsics indicates that the variable ``X`` is
807 declared at line number 2 at a function level scope in function ``foo``.
809 Now lets take another example.
813 call void @llvm.dbg.declare(metadata !{i32* %Z}, metadata !15), !dbg !17
814 ; [debug line = 5:9] [debug variable = Z]
816 The third intrinsic ``%llvm.dbg.declare`` encodes debugging information for
817 variable ``Z``. The metadata ``!dbg !17`` attached to the intrinsic provides
818 scope information for the variable ``Z``.
822 !16 = metadata !{i32 786443, metadata !1, metadata !4, i32 4, i32 0, i32 0} \
823 ; [ DW_TAG_lexical_block ] [/private/tmp/t.c]
824 !17 = metadata !{i32 5, i32 0, metadata !16, null}
826 Here ``!15`` indicates that ``Z`` is declared at line number 5 and
827 column number 0 inside of lexical scope ``!16``. The lexical scope itself
828 resides inside of subprogram ``!4`` described above.
830 The scope information attached with each instruction provides a straightforward
831 way to find instructions covered by a scope.
835 C/C++ front-end specific debug information
836 ==========================================
838 The C and C++ front-ends represent information about the program in a format
839 that is effectively identical to `DWARF 3.0
840 <http://www.eagercon.com/dwarf/dwarf3std.htm>`_ in terms of information
841 content. This allows code generators to trivially support native debuggers by
842 generating standard dwarf information, and contains enough information for
843 non-dwarf targets to translate it as needed.
845 This section describes the forms used to represent C and C++ programs. Other
846 languages could pattern themselves after this (which itself is tuned to
847 representing programs in the same way that DWARF 3 does), or they could choose
848 to provide completely different forms if they don't fit into the DWARF model.
849 As support for debugging information gets added to the various LLVM
850 source-language front-ends, the information used should be documented here.
852 The following sections provide examples of various C/C++ constructs and the
853 debug information that would best describe those constructs.
855 C/C++ source file information
856 -----------------------------
858 Given the source files ``MySource.cpp`` and ``MyHeader.h`` located in the
859 directory ``/Users/mine/sources``, the following code:
863 #include "MyHeader.h"
865 int main(int argc, char *argv[]) {
869 a C/C++ front-end would generate the following descriptors:
875 ;; Define the compile unit for the main source file "/Users/mine/sources/MySource.cpp".
879 metadata !1, ;; File/directory name
880 i32 4, ;; Language Id
881 metadata !"clang version 3.4 ",
882 i1 false, ;; Optimized compile unit
883 metadata !"", ;; Compiler flags
884 i32 0, ;; Runtime version
885 metadata !2, ;; Enumeration types
886 metadata !2, ;; Retained types
887 metadata !3, ;; Subprograms
888 metadata !2, ;; Global variables
889 metadata !2, ;; Imported entities (declarations and namespaces)
890 metadata !"" ;; Split debug filename
891 1, ;; Full debug info
895 ;; Define the file for the file "/Users/mine/sources/MySource.cpp".
898 metadata !"MySource.cpp",
899 metadata !"/Users/mine/sources"
907 ;; Define the file for the file "/Users/mine/sources/Myheader.h"
914 metadata !"./MyHeader.h",
915 metadata !"/Users/mine/sources",
920 ``llvm::Instruction`` provides easy access to metadata attached with an
921 instruction. One can extract line number information encoded in LLVM IR using
922 ``Instruction::getMetadata()`` and ``DILocation::getLineNumber()``.
926 if (MDNode *N = I->getMetadata("dbg")) { // Here I is an LLVM instruction
927 DILocation Loc(N); // DILocation is in DebugInfo.h
928 unsigned Line = Loc.getLineNumber();
929 StringRef File = Loc.getFilename();
930 StringRef Dir = Loc.getDirectory();
933 C/C++ global variable information
934 ---------------------------------
936 Given an integer global variable declared as follows:
942 a C/C++ front-end would generate the following descriptors:
947 ;; Define the global itself.
949 %MyGlobal = global int 100
952 ;; List of debug info of globals
956 ;; Define the compile unit.
961 metadata !"foo.cpp", ;; File
962 metadata !"/Volumes/Data/tmp", ;; Directory
963 metadata !"clang version 3.1 ", ;; Producer
964 i1 true, ;; Deprecated field
965 i1 false, ;; "isOptimized"?
966 metadata !"", ;; Flags
967 i32 0, ;; Runtime Version
968 metadata !1, ;; Enum Types
969 metadata !1, ;; Retained Types
970 metadata !1, ;; Subprograms
971 metadata !3, ;; Global Variables
972 metadata !1, ;; Imported entities
973 "", ;; Split debug filename
974 1, ;; Full debug info
975 } ; [ DW_TAG_compile_unit ]
977 ;; The Array of Global Variables
983 ;; Define the global variable itself.
989 metadata !"MyGlobal", ;; Name
990 metadata !"MyGlobal", ;; Display Name
991 metadata !"", ;; Linkage Name
995 i32 0, ;; IsLocalToUnit
996 i32 1, ;; IsDefinition
997 i32* @MyGlobal, ;; LLVM-IR Value
998 null ;; Static member declaration
999 } ; [ DW_TAG_variable ]
1005 metadata !"foo.cpp", ;; File
1006 metadata !"/Volumes/Data/tmp", ;; Directory
1010 metadata !5 ;; Unused
1011 } ; [ DW_TAG_file_type ]
1020 metadata !"int", ;; Name
1022 i64 32, ;; Size in Bits
1023 i64 32, ;; Align in Bits
1027 } ; [ DW_TAG_base_type ]
1029 C/C++ function information
1030 --------------------------
1032 Given a function declared as follows:
1036 int main(int argc, char *argv[]) {
1040 a C/C++ front-end would generate the following descriptors:
1042 .. code-block:: llvm
1045 ;; Define the anchor for subprograms.
1049 metadata !1, ;; File
1050 metadata !1, ;; Context
1051 metadata !"main", ;; Name
1052 metadata !"main", ;; Display name
1053 metadata !"main", ;; Linkage name
1054 i32 1, ;; Line number
1055 metadata !4, ;; Type
1056 i1 false, ;; Is local
1057 i1 true, ;; Is definition
1058 i32 0, ;; Virtuality attribute, e.g. pure virtual function
1059 i32 0, ;; Index into virtual table for C++ methods
1060 i32 0, ;; Type that holds virtual table.
1062 i1 false, ;; True if this function is optimized
1063 Function *, ;; Pointer to llvm::Function
1064 null, ;; Function template parameters
1065 null, ;; List of function variables (emitted when optimizing)
1066 1 ;; Line number of the opening '{' of the function
1069 ;; Define the subprogram itself.
1071 define i32 @main(i32 %argc, i8** %argv) {
1078 The following are the basic type descriptors for C/C++ core types:
1083 .. code-block:: llvm
1089 metadata !"bool", ;; Name
1090 i32 0, ;; Line number
1091 i64 8, ;; Size in Bits
1092 i64 8, ;; Align in Bits
1093 i64 0, ;; Offset in Bits
1101 .. code-block:: llvm
1107 metadata !"char", ;; Name
1108 i32 0, ;; Line number
1109 i64 8, ;; Size in Bits
1110 i64 8, ;; Align in Bits
1111 i64 0, ;; Offset in Bits
1119 .. code-block:: llvm
1125 metadata !"unsigned char",
1126 i32 0, ;; Line number
1127 i64 8, ;; Size in Bits
1128 i64 8, ;; Align in Bits
1129 i64 0, ;; Offset in Bits
1137 .. code-block:: llvm
1143 metadata !"short int",
1144 i32 0, ;; Line number
1145 i64 16, ;; Size in Bits
1146 i64 16, ;; Align in Bits
1147 i64 0, ;; Offset in Bits
1155 .. code-block:: llvm
1161 metadata !"short unsigned int",
1162 i32 0, ;; Line number
1163 i64 16, ;; Size in Bits
1164 i64 16, ;; Align in Bits
1165 i64 0, ;; Offset in Bits
1173 .. code-block:: llvm
1179 metadata !"int", ;; Name
1180 i32 0, ;; Line number
1181 i64 32, ;; Size in Bits
1182 i64 32, ;; Align in Bits
1183 i64 0, ;; Offset in Bits
1191 .. code-block:: llvm
1197 metadata !"unsigned int",
1198 i32 0, ;; Line number
1199 i64 32, ;; Size in Bits
1200 i64 32, ;; Align in Bits
1201 i64 0, ;; Offset in Bits
1209 .. code-block:: llvm
1215 metadata !"long long int",
1216 i32 0, ;; Line number
1217 i64 64, ;; Size in Bits
1218 i64 64, ;; Align in Bits
1219 i64 0, ;; Offset in Bits
1227 .. code-block:: llvm
1233 metadata !"long long unsigned int",
1234 i32 0, ;; Line number
1235 i64 64, ;; Size in Bits
1236 i64 64, ;; Align in Bits
1237 i64 0, ;; Offset in Bits
1245 .. code-block:: llvm
1252 i32 0, ;; Line number
1253 i64 32, ;; Size in Bits
1254 i64 32, ;; Align in Bits
1255 i64 0, ;; Offset in Bits
1263 .. code-block:: llvm
1269 metadata !"double",;; Name
1270 i32 0, ;; Line number
1271 i64 64, ;; Size in Bits
1272 i64 64, ;; Align in Bits
1273 i64 0, ;; Offset in Bits
1281 Given the following as an example of C/C++ derived type:
1285 typedef const int *IntPtr;
1287 a C/C++ front-end would generate the following descriptors:
1289 .. code-block:: llvm
1292 ;; Define the typedef "IntPtr".
1296 metadata !3, ;; File
1297 metadata !1, ;; Context
1298 metadata !"IntPtr", ;; Name
1299 i32 0, ;; Line number
1300 i64 0, ;; Size in bits
1301 i64 0, ;; Align in bits
1302 i64 0, ;; Offset in bits
1304 metadata !4 ;; Derived From type
1307 ;; Define the pointer type.
1313 metadata !"", ;; Name
1314 i32 0, ;; Line number
1315 i64 64, ;; Size in bits
1316 i64 64, ;; Align in bits
1317 i64 0, ;; Offset in bits
1319 metadata !5 ;; Derived From type
1322 ;; Define the const type.
1328 metadata !"", ;; Name
1329 i32 0, ;; Line number
1330 i64 0, ;; Size in bits
1331 i64 0, ;; Align in bits
1332 i64 0, ;; Offset in bits
1334 metadata !6 ;; Derived From type
1337 ;; Define the int type.
1343 metadata !"int", ;; Name
1344 i32 0, ;; Line number
1345 i64 32, ;; Size in bits
1346 i64 32, ;; Align in bits
1347 i64 0, ;; Offset in bits
1352 C/C++ struct/union types
1353 ------------------------
1355 Given the following as an example of C/C++ struct type:
1365 a C/C++ front-end would generate the following descriptors:
1367 .. code-block:: llvm
1370 ;; Define basic type for unsigned int.
1376 metadata !"unsigned int",
1377 i32 0, ;; Line number
1378 i64 32, ;; Size in Bits
1379 i64 32, ;; Align in Bits
1380 i64 0, ;; Offset in Bits
1385 ;; Define composite type for struct Color.
1389 metadata !1, ;; Compile unit
1391 metadata !"Color", ;; Name
1392 i32 1, ;; Line number
1393 i64 96, ;; Size in bits
1394 i64 32, ;; Align in bits
1395 i64 0, ;; Offset in bits
1397 null, ;; Derived From
1398 metadata !3, ;; Elements
1399 i32 0, ;; Runtime Language
1400 null, ;; Base type containing the vtable pointer for this type
1401 null ;; Template parameters
1405 ;; Define the Red field.
1409 metadata !1, ;; File
1410 metadata !1, ;; Context
1411 metadata !"Red", ;; Name
1412 i32 2, ;; Line number
1413 i64 32, ;; Size in bits
1414 i64 32, ;; Align in bits
1415 i64 0, ;; Offset in bits
1417 metadata !5 ;; Derived From type
1421 ;; Define the Green field.
1425 metadata !1, ;; File
1426 metadata !1, ;; Context
1427 metadata !"Green", ;; Name
1428 i32 3, ;; Line number
1429 i64 32, ;; Size in bits
1430 i64 32, ;; Align in bits
1431 i64 32, ;; Offset in bits
1433 metadata !5 ;; Derived From type
1437 ;; Define the Blue field.
1441 metadata !1, ;; File
1442 metadata !1, ;; Context
1443 metadata !"Blue", ;; Name
1444 i32 4, ;; Line number
1445 i64 32, ;; Size in bits
1446 i64 32, ;; Align in bits
1447 i64 64, ;; Offset in bits
1449 metadata !5 ;; Derived From type
1453 ;; Define the array of fields used by the composite type Color.
1455 !3 = metadata !{metadata !4, metadata !6, metadata !7}
1457 C/C++ enumeration types
1458 -----------------------
1460 Given the following as an example of C/C++ enumeration type:
1470 a C/C++ front-end would generate the following descriptors:
1472 .. code-block:: llvm
1475 ;; Define composite type for enum Trees
1479 metadata !1, ;; File
1480 metadata !1, ;; Context
1481 metadata !"Trees", ;; Name
1482 i32 1, ;; Line number
1483 i64 32, ;; Size in bits
1484 i64 32, ;; Align in bits
1485 i64 0, ;; Offset in bits
1487 null, ;; Derived From type
1488 metadata !3, ;; Elements
1489 i32 0 ;; Runtime language
1493 ;; Define the array of enumerators used by composite type Trees.
1495 !3 = metadata !{metadata !4, metadata !5, metadata !6}
1498 ;; Define Spruce enumerator.
1500 !4 = metadata !{i32 786472, metadata !"Spruce", i64 100}
1503 ;; Define Oak enumerator.
1505 !5 = metadata !{i32 786472, metadata !"Oak", i64 200}
1508 ;; Define Maple enumerator.
1510 !6 = metadata !{i32 786472, metadata !"Maple", i64 300}
1512 Debugging information format
1513 ============================
1515 Debugging Information Extension for Objective C Properties
1516 ----------------------------------------------------------
1521 Objective C provides a simpler way to declare and define accessor methods using
1522 declared properties. The language provides features to declare a property and
1523 to let compiler synthesize accessor methods.
1525 The debugger lets developer inspect Objective C interfaces and their instance
1526 variables and class variables. However, the debugger does not know anything
1527 about the properties defined in Objective C interfaces. The debugger consumes
1528 information generated by compiler in DWARF format. The format does not support
1529 encoding of Objective C properties. This proposal describes DWARF extensions to
1530 encode Objective C properties, which the debugger can use to let developers
1531 inspect Objective C properties.
1536 Objective C properties exist separately from class members. A property can be
1537 defined only by "setter" and "getter" selectors, and be calculated anew on each
1538 access. Or a property can just be a direct access to some declared ivar.
1539 Finally it can have an ivar "automatically synthesized" for it by the compiler,
1540 in which case the property can be referred to in user code directly using the
1541 standard C dereference syntax as well as through the property "dot" syntax, but
1542 there is no entry in the ``@interface`` declaration corresponding to this ivar.
1544 To facilitate debugging, these properties we will add a new DWARF TAG into the
1545 ``DW_TAG_structure_type`` definition for the class to hold the description of a
1546 given property, and a set of DWARF attributes that provide said description.
1547 The property tag will also contain the name and declared type of the property.
1549 If there is a related ivar, there will also be a DWARF property attribute placed
1550 in the ``DW_TAG_member`` DIE for that ivar referring back to the property TAG
1551 for that property. And in the case where the compiler synthesizes the ivar
1552 directly, the compiler is expected to generate a ``DW_TAG_member`` for that
1553 ivar (with the ``DW_AT_artificial`` set to 1), whose name will be the name used
1554 to access this ivar directly in code, and with the property attribute pointing
1555 back to the property it is backing.
1557 The following examples will serve as illustration for our discussion:
1559 .. code-block:: objc
1571 @synthesize p2 = n2;
1574 This produces the following DWARF (this is a "pseudo dwarfdump" output):
1576 .. code-block:: none
1578 0x00000100: TAG_structure_type [7] *
1579 AT_APPLE_runtime_class( 0x10 )
1581 AT_decl_file( "Objc_Property.m" )
1584 0x00000110 TAG_APPLE_property
1586 AT_type ( {0x00000150} ( int ) )
1588 0x00000120: TAG_APPLE_property
1590 AT_type ( {0x00000150} ( int ) )
1592 0x00000130: TAG_member [8]
1594 AT_APPLE_property ( {0x00000110} "p1" )
1595 AT_type( {0x00000150} ( int ) )
1596 AT_artificial ( 0x1 )
1598 0x00000140: TAG_member [8]
1600 AT_APPLE_property ( {0x00000120} "p2" )
1601 AT_type( {0x00000150} ( int ) )
1603 0x00000150: AT_type( ( int ) )
1605 Note, the current convention is that the name of the ivar for an
1606 auto-synthesized property is the name of the property from which it derives
1607 with an underscore prepended, as is shown in the example. But we actually
1608 don't need to know this convention, since we are given the name of the ivar
1611 Also, it is common practice in ObjC to have different property declarations in
1612 the @interface and @implementation - e.g. to provide a read-only property in
1613 the interface,and a read-write interface in the implementation. In that case,
1614 the compiler should emit whichever property declaration will be in force in the
1615 current translation unit.
1617 Developers can decorate a property with attributes which are encoded using
1618 ``DW_AT_APPLE_property_attribute``.
1620 .. code-block:: objc
1622 @property (readonly, nonatomic) int pr;
1624 .. code-block:: none
1626 TAG_APPLE_property [8]
1628 AT_type ( {0x00000147} (int) )
1629 AT_APPLE_property_attribute (DW_APPLE_PROPERTY_readonly, DW_APPLE_PROPERTY_nonatomic)
1631 The setter and getter method names are attached to the property using
1632 ``DW_AT_APPLE_property_setter`` and ``DW_AT_APPLE_property_getter`` attributes.
1634 .. code-block:: objc
1637 @property (setter=myOwnP3Setter:) int p3;
1638 -(void)myOwnP3Setter:(int)a;
1643 -(void)myOwnP3Setter:(int)a{ }
1646 The DWARF for this would be:
1648 .. code-block:: none
1650 0x000003bd: TAG_structure_type [7] *
1651 AT_APPLE_runtime_class( 0x10 )
1653 AT_decl_file( "Objc_Property.m" )
1656 0x000003cd TAG_APPLE_property
1658 AT_APPLE_property_setter ( "myOwnP3Setter:" )
1659 AT_type( {0x00000147} ( int ) )
1661 0x000003f3: TAG_member [8]
1663 AT_type ( {0x00000147} ( int ) )
1664 AT_APPLE_property ( {0x000003cd} )
1665 AT_artificial ( 0x1 )
1670 +-----------------------+--------+
1672 +=======================+========+
1673 | DW_TAG_APPLE_property | 0x4200 |
1674 +-----------------------+--------+
1676 New DWARF Attributes
1677 ^^^^^^^^^^^^^^^^^^^^
1679 +--------------------------------+--------+-----------+
1680 | Attribute | Value | Classes |
1681 +================================+========+===========+
1682 | DW_AT_APPLE_property | 0x3fed | Reference |
1683 +--------------------------------+--------+-----------+
1684 | DW_AT_APPLE_property_getter | 0x3fe9 | String |
1685 +--------------------------------+--------+-----------+
1686 | DW_AT_APPLE_property_setter | 0x3fea | String |
1687 +--------------------------------+--------+-----------+
1688 | DW_AT_APPLE_property_attribute | 0x3feb | Constant |
1689 +--------------------------------+--------+-----------+
1694 +--------------------------------+-------+
1696 +================================+=======+
1697 | DW_AT_APPLE_PROPERTY_readonly | 0x1 |
1698 +--------------------------------+-------+
1699 | DW_AT_APPLE_PROPERTY_readwrite | 0x2 |
1700 +--------------------------------+-------+
1701 | DW_AT_APPLE_PROPERTY_assign | 0x4 |
1702 +--------------------------------+-------+
1703 | DW_AT_APPLE_PROPERTY_retain | 0x8 |
1704 +--------------------------------+-------+
1705 | DW_AT_APPLE_PROPERTY_copy | 0x10 |
1706 +--------------------------------+-------+
1707 | DW_AT_APPLE_PROPERTY_nonatomic | 0x20 |
1708 +--------------------------------+-------+
1710 Name Accelerator Tables
1711 -----------------------
1716 The "``.debug_pubnames``" and "``.debug_pubtypes``" formats are not what a
1717 debugger needs. The "``pub``" in the section name indicates that the entries
1718 in the table are publicly visible names only. This means no static or hidden
1719 functions show up in the "``.debug_pubnames``". No static variables or private
1720 class variables are in the "``.debug_pubtypes``". Many compilers add different
1721 things to these tables, so we can't rely upon the contents between gcc, icc, or
1724 The typical query given by users tends not to match up with the contents of
1725 these tables. For example, the DWARF spec states that "In the case of the name
1726 of a function member or static data member of a C++ structure, class or union,
1727 the name presented in the "``.debug_pubnames``" section is not the simple name
1728 given by the ``DW_AT_name attribute`` of the referenced debugging information
1729 entry, but rather the fully qualified name of the data or function member."
1730 So the only names in these tables for complex C++ entries is a fully
1731 qualified name. Debugger users tend not to enter their search strings as
1732 "``a::b::c(int,const Foo&) const``", but rather as "``c``", "``b::c``" , or
1733 "``a::b::c``". So the name entered in the name table must be demangled in
1734 order to chop it up appropriately and additional names must be manually entered
1735 into the table to make it effective as a name lookup table for debuggers to
1738 All debuggers currently ignore the "``.debug_pubnames``" table as a result of
1739 its inconsistent and useless public-only name content making it a waste of
1740 space in the object file. These tables, when they are written to disk, are not
1741 sorted in any way, leaving every debugger to do its own parsing and sorting.
1742 These tables also include an inlined copy of the string values in the table
1743 itself making the tables much larger than they need to be on disk, especially
1744 for large C++ programs.
1746 Can't we just fix the sections by adding all of the names we need to this
1747 table? No, because that is not what the tables are defined to contain and we
1748 won't know the difference between the old bad tables and the new good tables.
1749 At best we could make our own renamed sections that contain all of the data we
1752 These tables are also insufficient for what a debugger like LLDB needs. LLDB
1753 uses clang for its expression parsing where LLDB acts as a PCH. LLDB is then
1754 often asked to look for type "``foo``" or namespace "``bar``", or list items in
1755 namespace "``baz``". Namespaces are not included in the pubnames or pubtypes
1756 tables. Since clang asks a lot of questions when it is parsing an expression,
1757 we need to be very fast when looking up names, as it happens a lot. Having new
1758 accelerator tables that are optimized for very quick lookups will benefit this
1759 type of debugging experience greatly.
1761 We would like to generate name lookup tables that can be mapped into memory
1762 from disk, and used as is, with little or no up-front parsing. We would also
1763 be able to control the exact content of these different tables so they contain
1764 exactly what we need. The Name Accelerator Tables were designed to fix these
1765 issues. In order to solve these issues we need to:
1767 * Have a format that can be mapped into memory from disk and used as is
1768 * Lookups should be very fast
1769 * Extensible table format so these tables can be made by many producers
1770 * Contain all of the names needed for typical lookups out of the box
1771 * Strict rules for the contents of tables
1773 Table size is important and the accelerator table format should allow the reuse
1774 of strings from common string tables so the strings for the names are not
1775 duplicated. We also want to make sure the table is ready to be used as-is by
1776 simply mapping the table into memory with minimal header parsing.
1778 The name lookups need to be fast and optimized for the kinds of lookups that
1779 debuggers tend to do. Optimally we would like to touch as few parts of the
1780 mapped table as possible when doing a name lookup and be able to quickly find
1781 the name entry we are looking for, or discover there are no matches. In the
1782 case of debuggers we optimized for lookups that fail most of the time.
1784 Each table that is defined should have strict rules on exactly what is in the
1785 accelerator tables and documented so clients can rely on the content.
1790 Standard Hash Tables
1791 """"""""""""""""""""
1793 Typical hash tables have a header, buckets, and each bucket points to the
1796 .. code-block:: none
1806 The BUCKETS are an array of offsets to DATA for each hash:
1808 .. code-block:: none
1811 | 0x00001000 | BUCKETS[0]
1812 | 0x00002000 | BUCKETS[1]
1813 | 0x00002200 | BUCKETS[2]
1814 | 0x000034f0 | BUCKETS[3]
1816 | 0xXXXXXXXX | BUCKETS[n_buckets]
1819 So for ``bucket[3]`` in the example above, we have an offset into the table
1820 0x000034f0 which points to a chain of entries for the bucket. Each bucket must
1821 contain a next pointer, full 32 bit hash value, the string itself, and the data
1822 for the current string value.
1824 .. code-block:: none
1827 0x000034f0: | 0x00003500 | next pointer
1828 | 0x12345678 | 32 bit hash
1829 | "erase" | string value
1830 | data[n] | HashData for this bucket
1832 0x00003500: | 0x00003550 | next pointer
1833 | 0x29273623 | 32 bit hash
1834 | "dump" | string value
1835 | data[n] | HashData for this bucket
1837 0x00003550: | 0x00000000 | next pointer
1838 | 0x82638293 | 32 bit hash
1839 | "main" | string value
1840 | data[n] | HashData for this bucket
1843 The problem with this layout for debuggers is that we need to optimize for the
1844 negative lookup case where the symbol we're searching for is not present. So
1845 if we were to lookup "``printf``" in the table above, we would make a 32 hash
1846 for "``printf``", it might match ``bucket[3]``. We would need to go to the
1847 offset 0x000034f0 and start looking to see if our 32 bit hash matches. To do
1848 so, we need to read the next pointer, then read the hash, compare it, and skip
1849 to the next bucket. Each time we are skipping many bytes in memory and
1850 touching new cache pages just to do the compare on the full 32 bit hash. All
1851 of these accesses then tell us that we didn't have a match.
1856 To solve the issues mentioned above we have structured the hash tables a bit
1857 differently: a header, buckets, an array of all unique 32 bit hash values,
1858 followed by an array of hash value data offsets, one for each hash value, then
1859 the data for all hash values:
1861 .. code-block:: none
1875 The ``BUCKETS`` in the name tables are an index into the ``HASHES`` array. By
1876 making all of the full 32 bit hash values contiguous in memory, we allow
1877 ourselves to efficiently check for a match while touching as little memory as
1878 possible. Most often checking the 32 bit hash values is as far as the lookup
1879 goes. If it does match, it usually is a match with no collisions. So for a
1880 table with "``n_buckets``" buckets, and "``n_hashes``" unique 32 bit hash
1881 values, we can clarify the contents of the ``BUCKETS``, ``HASHES`` and
1884 .. code-block:: none
1886 .-------------------------.
1887 | HEADER.magic | uint32_t
1888 | HEADER.version | uint16_t
1889 | HEADER.hash_function | uint16_t
1890 | HEADER.bucket_count | uint32_t
1891 | HEADER.hashes_count | uint32_t
1892 | HEADER.header_data_len | uint32_t
1893 | HEADER_DATA | HeaderData
1894 |-------------------------|
1895 | BUCKETS | uint32_t[n_buckets] // 32 bit hash indexes
1896 |-------------------------|
1897 | HASHES | uint32_t[n_hashes] // 32 bit hash values
1898 |-------------------------|
1899 | OFFSETS | uint32_t[n_hashes] // 32 bit offsets to hash value data
1900 |-------------------------|
1902 `-------------------------'
1904 So taking the exact same data from the standard hash example above we end up
1907 .. code-block:: none
1917 | ... | BUCKETS[n_buckets]
1919 | 0x........ | HASHES[0]
1920 | 0x........ | HASHES[1]
1921 | 0x........ | HASHES[2]
1922 | 0x........ | HASHES[3]
1923 | 0x........ | HASHES[4]
1924 | 0x........ | HASHES[5]
1925 | 0x12345678 | HASHES[6] hash for BUCKETS[3]
1926 | 0x29273623 | HASHES[7] hash for BUCKETS[3]
1927 | 0x82638293 | HASHES[8] hash for BUCKETS[3]
1928 | 0x........ | HASHES[9]
1929 | 0x........ | HASHES[10]
1930 | 0x........ | HASHES[11]
1931 | 0x........ | HASHES[12]
1932 | 0x........ | HASHES[13]
1933 | 0x........ | HASHES[n_hashes]
1935 | 0x........ | OFFSETS[0]
1936 | 0x........ | OFFSETS[1]
1937 | 0x........ | OFFSETS[2]
1938 | 0x........ | OFFSETS[3]
1939 | 0x........ | OFFSETS[4]
1940 | 0x........ | OFFSETS[5]
1941 | 0x000034f0 | OFFSETS[6] offset for BUCKETS[3]
1942 | 0x00003500 | OFFSETS[7] offset for BUCKETS[3]
1943 | 0x00003550 | OFFSETS[8] offset for BUCKETS[3]
1944 | 0x........ | OFFSETS[9]
1945 | 0x........ | OFFSETS[10]
1946 | 0x........ | OFFSETS[11]
1947 | 0x........ | OFFSETS[12]
1948 | 0x........ | OFFSETS[13]
1949 | 0x........ | OFFSETS[n_hashes]
1957 0x000034f0: | 0x00001203 | .debug_str ("erase")
1958 | 0x00000004 | A 32 bit array count - number of HashData with name "erase"
1959 | 0x........ | HashData[0]
1960 | 0x........ | HashData[1]
1961 | 0x........ | HashData[2]
1962 | 0x........ | HashData[3]
1963 | 0x00000000 | String offset into .debug_str (terminate data for hash)
1965 0x00003500: | 0x00001203 | String offset into .debug_str ("collision")
1966 | 0x00000002 | A 32 bit array count - number of HashData with name "collision"
1967 | 0x........ | HashData[0]
1968 | 0x........ | HashData[1]
1969 | 0x00001203 | String offset into .debug_str ("dump")
1970 | 0x00000003 | A 32 bit array count - number of HashData with name "dump"
1971 | 0x........ | HashData[0]
1972 | 0x........ | HashData[1]
1973 | 0x........ | HashData[2]
1974 | 0x00000000 | String offset into .debug_str (terminate data for hash)
1976 0x00003550: | 0x00001203 | String offset into .debug_str ("main")
1977 | 0x00000009 | A 32 bit array count - number of HashData with name "main"
1978 | 0x........ | HashData[0]
1979 | 0x........ | HashData[1]
1980 | 0x........ | HashData[2]
1981 | 0x........ | HashData[3]
1982 | 0x........ | HashData[4]
1983 | 0x........ | HashData[5]
1984 | 0x........ | HashData[6]
1985 | 0x........ | HashData[7]
1986 | 0x........ | HashData[8]
1987 | 0x00000000 | String offset into .debug_str (terminate data for hash)
1990 So we still have all of the same data, we just organize it more efficiently for
1991 debugger lookup. If we repeat the same "``printf``" lookup from above, we
1992 would hash "``printf``" and find it matches ``BUCKETS[3]`` by taking the 32 bit
1993 hash value and modulo it by ``n_buckets``. ``BUCKETS[3]`` contains "6" which
1994 is the index into the ``HASHES`` table. We would then compare any consecutive
1995 32 bit hashes values in the ``HASHES`` array as long as the hashes would be in
1996 ``BUCKETS[3]``. We do this by verifying that each subsequent hash value modulo
1997 ``n_buckets`` is still 3. In the case of a failed lookup we would access the
1998 memory for ``BUCKETS[3]``, and then compare a few consecutive 32 bit hashes
1999 before we know that we have no match. We don't end up marching through
2000 multiple words of memory and we really keep the number of processor data cache
2001 lines being accessed as small as possible.
2003 The string hash that is used for these lookup tables is the Daniel J.
2004 Bernstein hash which is also used in the ELF ``GNU_HASH`` sections. It is a
2005 very good hash for all kinds of names in programs with very few hash
2008 Empty buckets are designated by using an invalid hash index of ``UINT32_MAX``.
2013 These name hash tables are designed to be generic where specializations of the
2014 table get to define additional data that goes into the header ("``HeaderData``"),
2015 how the string value is stored ("``KeyType``") and the content of the data for each
2021 The header has a fixed part, and the specialized part. The exact format of the
2028 uint32_t magic; // 'HASH' magic value to allow endian detection
2029 uint16_t version; // Version number
2030 uint16_t hash_function; // The hash function enumeration that was used
2031 uint32_t bucket_count; // The number of buckets in this hash table
2032 uint32_t hashes_count; // The total number of unique hash values and hash data offsets in this table
2033 uint32_t header_data_len; // The bytes to skip to get to the hash indexes (buckets) for correct alignment
2034 // Specifically the length of the following HeaderData field - this does not
2035 // include the size of the preceding fields
2036 HeaderData header_data; // Implementation specific header data
2039 The header starts with a 32 bit "``magic``" value which must be ``'HASH'``
2040 encoded as an ASCII integer. This allows the detection of the start of the
2041 hash table and also allows the table's byte order to be determined so the table
2042 can be correctly extracted. The "``magic``" value is followed by a 16 bit
2043 ``version`` number which allows the table to be revised and modified in the
2044 future. The current version number is 1. ``hash_function`` is a ``uint16_t``
2045 enumeration that specifies which hash function was used to produce this table.
2046 The current values for the hash function enumerations include:
2050 enum HashFunctionType
2052 eHashFunctionDJB = 0u, // Daniel J Bernstein hash function
2055 ``bucket_count`` is a 32 bit unsigned integer that represents how many buckets
2056 are in the ``BUCKETS`` array. ``hashes_count`` is the number of unique 32 bit
2057 hash values that are in the ``HASHES`` array, and is the same number of offsets
2058 are contained in the ``OFFSETS`` array. ``header_data_len`` specifies the size
2059 in bytes of the ``HeaderData`` that is filled in by specialized versions of
2065 The header is followed by the buckets, hashes, offsets, and hash value data.
2071 uint32_t buckets[Header.bucket_count]; // An array of hash indexes into the "hashes[]" array below
2072 uint32_t hashes [Header.hashes_count]; // Every unique 32 bit hash for the entire table is in this table
2073 uint32_t offsets[Header.hashes_count]; // An offset that corresponds to each item in the "hashes[]" array above
2076 ``buckets`` is an array of 32 bit indexes into the ``hashes`` array. The
2077 ``hashes`` array contains all of the 32 bit hash values for all names in the
2078 hash table. Each hash in the ``hashes`` table has an offset in the ``offsets``
2079 array that points to the data for the hash value.
2081 This table setup makes it very easy to repurpose these tables to contain
2082 different data, while keeping the lookup mechanism the same for all tables.
2083 This layout also makes it possible to save the table to disk and map it in
2084 later and do very efficient name lookups with little or no parsing.
2086 DWARF lookup tables can be implemented in a variety of ways and can store a lot
2087 of information for each name. We want to make the DWARF tables extensible and
2088 able to store the data efficiently so we have used some of the DWARF features
2089 that enable efficient data storage to define exactly what kind of data we store
2092 The ``HeaderData`` contains a definition of the contents of each HashData chunk.
2093 We might want to store an offset to all of the debug information entries (DIEs)
2094 for each name. To keep things extensible, we create a list of items, or
2095 Atoms, that are contained in the data for each name. First comes the type of
2096 the data in each atom:
2103 eAtomTypeDIEOffset = 1u, // DIE offset, check form for encoding
2104 eAtomTypeCUOffset = 2u, // DIE offset of the compiler unit header that contains the item in question
2105 eAtomTypeTag = 3u, // DW_TAG_xxx value, should be encoded as DW_FORM_data1 (if no tags exceed 255) or DW_FORM_data2
2106 eAtomTypeNameFlags = 4u, // Flags from enum NameFlags
2107 eAtomTypeTypeFlags = 5u, // Flags from enum TypeFlags
2110 The enumeration values and their meanings are:
2112 .. code-block:: none
2114 eAtomTypeNULL - a termination atom that specifies the end of the atom list
2115 eAtomTypeDIEOffset - an offset into the .debug_info section for the DWARF DIE for this name
2116 eAtomTypeCUOffset - an offset into the .debug_info section for the CU that contains the DIE
2117 eAtomTypeDIETag - The DW_TAG_XXX enumeration value so you don't have to parse the DWARF to see what it is
2118 eAtomTypeNameFlags - Flags for functions and global variables (isFunction, isInlined, isExternal...)
2119 eAtomTypeTypeFlags - Flags for types (isCXXClass, isObjCClass, ...)
2121 Then we allow each atom type to define the atom type and how the data for each
2122 atom type data is encoded:
2128 uint16_t type; // AtomType enum value
2129 uint16_t form; // DWARF DW_FORM_XXX defines
2132 The ``form`` type above is from the DWARF specification and defines the exact
2133 encoding of the data for the Atom type. See the DWARF specification for the
2134 ``DW_FORM_`` definitions.
2140 uint32_t die_offset_base;
2141 uint32_t atom_count;
2142 Atoms atoms[atom_count0];
2145 ``HeaderData`` defines the base DIE offset that should be added to any atoms
2146 that are encoded using the ``DW_FORM_ref1``, ``DW_FORM_ref2``,
2147 ``DW_FORM_ref4``, ``DW_FORM_ref8`` or ``DW_FORM_ref_udata``. It also defines
2148 what is contained in each ``HashData`` object -- ``Atom.form`` tells us how large
2149 each field will be in the ``HashData`` and the ``Atom.type`` tells us how this data
2150 should be interpreted.
2152 For the current implementations of the "``.apple_names``" (all functions +
2153 globals), the "``.apple_types``" (names of all types that are defined), and
2154 the "``.apple_namespaces``" (all namespaces), we currently set the ``Atom``
2159 HeaderData.atom_count = 1;
2160 HeaderData.atoms[0].type = eAtomTypeDIEOffset;
2161 HeaderData.atoms[0].form = DW_FORM_data4;
2163 This defines the contents to be the DIE offset (eAtomTypeDIEOffset) that is
2164 encoded as a 32 bit value (DW_FORM_data4). This allows a single name to have
2165 multiple matching DIEs in a single file, which could come up with an inlined
2166 function for instance. Future tables could include more information about the
2167 DIE such as flags indicating if the DIE is a function, method, block,
2170 The KeyType for the DWARF table is a 32 bit string table offset into the
2171 ".debug_str" table. The ".debug_str" is the string table for the DWARF which
2172 may already contain copies of all of the strings. This helps make sure, with
2173 help from the compiler, that we reuse the strings between all of the DWARF
2174 sections and keeps the hash table size down. Another benefit to having the
2175 compiler generate all strings as DW_FORM_strp in the debug info, is that
2176 DWARF parsing can be made much faster.
2178 After a lookup is made, we get an offset into the hash data. The hash data
2179 needs to be able to deal with 32 bit hash collisions, so the chunk of data
2180 at the offset in the hash data consists of a triple:
2185 uint32_t hash_data_count
2186 HashData[hash_data_count]
2188 If "str_offset" is zero, then the bucket contents are done. 99.9% of the
2189 hash data chunks contain a single item (no 32 bit hash collision):
2191 .. code-block:: none
2194 | 0x00001023 | uint32_t KeyType (.debug_str[0x0001023] => "main")
2195 | 0x00000004 | uint32_t HashData count
2196 | 0x........ | uint32_t HashData[0] DIE offset
2197 | 0x........ | uint32_t HashData[1] DIE offset
2198 | 0x........ | uint32_t HashData[2] DIE offset
2199 | 0x........ | uint32_t HashData[3] DIE offset
2200 | 0x00000000 | uint32_t KeyType (end of hash chain)
2203 If there are collisions, you will have multiple valid string offsets:
2205 .. code-block:: none
2208 | 0x00001023 | uint32_t KeyType (.debug_str[0x0001023] => "main")
2209 | 0x00000004 | uint32_t HashData count
2210 | 0x........ | uint32_t HashData[0] DIE offset
2211 | 0x........ | uint32_t HashData[1] DIE offset
2212 | 0x........ | uint32_t HashData[2] DIE offset
2213 | 0x........ | uint32_t HashData[3] DIE offset
2214 | 0x00002023 | uint32_t KeyType (.debug_str[0x0002023] => "print")
2215 | 0x00000002 | uint32_t HashData count
2216 | 0x........ | uint32_t HashData[0] DIE offset
2217 | 0x........ | uint32_t HashData[1] DIE offset
2218 | 0x00000000 | uint32_t KeyType (end of hash chain)
2221 Current testing with real world C++ binaries has shown that there is around 1
2222 32 bit hash collision per 100,000 name entries.
2227 As we said, we want to strictly define exactly what is included in the
2228 different tables. For DWARF, we have 3 tables: "``.apple_names``",
2229 "``.apple_types``", and "``.apple_namespaces``".
2231 "``.apple_names``" sections should contain an entry for each DWARF DIE whose
2232 ``DW_TAG`` is a ``DW_TAG_label``, ``DW_TAG_inlined_subroutine``, or
2233 ``DW_TAG_subprogram`` that has address attributes: ``DW_AT_low_pc``,
2234 ``DW_AT_high_pc``, ``DW_AT_ranges`` or ``DW_AT_entry_pc``. It also contains
2235 ``DW_TAG_variable`` DIEs that have a ``DW_OP_addr`` in the location (global and
2236 static variables). All global and static variables should be included,
2237 including those scoped within functions and classes. For example using the
2249 Both of the static ``var`` variables would be included in the table. All
2250 functions should emit both their full names and their basenames. For C or C++,
2251 the full name is the mangled name (if available) which is usually in the
2252 ``DW_AT_MIPS_linkage_name`` attribute, and the ``DW_AT_name`` contains the
2253 function basename. If global or static variables have a mangled name in a
2254 ``DW_AT_MIPS_linkage_name`` attribute, this should be emitted along with the
2255 simple name found in the ``DW_AT_name`` attribute.
2257 "``.apple_types``" sections should contain an entry for each DWARF DIE whose
2262 * DW_TAG_enumeration_type
2263 * DW_TAG_pointer_type
2264 * DW_TAG_reference_type
2265 * DW_TAG_string_type
2266 * DW_TAG_structure_type
2267 * DW_TAG_subroutine_type
2270 * DW_TAG_ptr_to_member_type
2272 * DW_TAG_subrange_type
2278 * DW_TAG_packed_type
2279 * DW_TAG_volatile_type
2280 * DW_TAG_restrict_type
2281 * DW_TAG_interface_type
2282 * DW_TAG_unspecified_type
2283 * DW_TAG_shared_type
2285 Only entries with a ``DW_AT_name`` attribute are included, and the entry must
2286 not be a forward declaration (``DW_AT_declaration`` attribute with a non-zero
2287 value). For example, using the following code:
2297 We get a few type DIEs:
2299 .. code-block:: none
2301 0x00000067: TAG_base_type [5]
2302 AT_encoding( DW_ATE_signed )
2304 AT_byte_size( 0x04 )
2306 0x0000006e: TAG_pointer_type [6]
2307 AT_type( {0x00000067} ( int ) )
2308 AT_byte_size( 0x08 )
2310 The DW_TAG_pointer_type is not included because it does not have a ``DW_AT_name``.
2312 "``.apple_namespaces``" section should contain all ``DW_TAG_namespace`` DIEs.
2313 If we run into a namespace that has no name this is an anonymous namespace, and
2314 the name should be output as "``(anonymous namespace)``" (without the quotes).
2315 Why? This matches the output of the ``abi::cxa_demangle()`` that is in the
2316 standard C++ library that demangles mangled names.
2319 Language Extensions and File Format Changes
2320 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2322 Objective-C Extensions
2323 """"""""""""""""""""""
2325 "``.apple_objc``" section should contain all ``DW_TAG_subprogram`` DIEs for an
2326 Objective-C class. The name used in the hash table is the name of the
2327 Objective-C class itself. If the Objective-C class has a category, then an
2328 entry is made for both the class name without the category, and for the class
2329 name with the category. So if we have a DIE at offset 0x1234 with a name of
2330 method "``-[NSString(my_additions) stringWithSpecialString:]``", we would add
2331 an entry for "``NSString``" that points to DIE 0x1234, and an entry for
2332 "``NSString(my_additions)``" that points to 0x1234. This allows us to quickly
2333 track down all Objective-C methods for an Objective-C class when doing
2334 expressions. It is needed because of the dynamic nature of Objective-C where
2335 anyone can add methods to a class. The DWARF for Objective-C methods is also
2336 emitted differently from C++ classes where the methods are not usually
2337 contained in the class definition, they are scattered about across one or more
2338 compile units. Categories can also be defined in different shared libraries.
2339 So we need to be able to quickly find all of the methods and class functions
2340 given the Objective-C class name, or quickly find all methods and class
2341 functions for a class + category name. This table does not contain any
2342 selector names, it just maps Objective-C class names (or class names +
2343 category) to all of the methods and class functions. The selectors are added
2344 as function basenames in the "``.debug_names``" section.
2346 In the "``.apple_names``" section for Objective-C functions, the full name is
2347 the entire function name with the brackets ("``-[NSString
2348 stringWithCString:]``") and the basename is the selector only
2349 ("``stringWithCString:``").
2354 The sections names for the apple hash tables are for non-mach-o files. For
2355 mach-o files, the sections should be contained in the ``__DWARF`` segment with
2358 * "``.apple_names``" -> "``__apple_names``"
2359 * "``.apple_types``" -> "``__apple_types``"
2360 * "``.apple_namespaces``" -> "``__apple_namespac``" (16 character limit)
2361 * "``.apple_objc``" -> "``__apple_objc``"