\input texinfo @setfilename stabs.info @c @finalout @ifinfo @format START-INFO-DIR-ENTRY * Stabs:: The "stabs" debugging information format. END-INFO-DIR-ENTRY @end format @end ifinfo @ifinfo This document describes the stabs debugging symbol tables. Copyright 1992, 1993 Free Software Foundation, Inc. Contributed by Cygnus Support. Written by Julia Menapace. Permission is granted to make and distribute verbatim copies of this manual provided the copyright notice and this permission notice are preserved on all copies. @ignore Permission is granted to process this file through Tex and print the results, provided the printed document carries copying permission notice identical to this one except for the removal of this paragraph (this paragraph not being relevant to the printed manual). @end ignore Permission is granted to copy or distribute modified versions of this manual under the terms of the GPL (for which purpose this text may be regarded as a program in the language TeX). @end ifinfo @setchapternewpage odd @settitle STABS @titlepage @title The ``stabs'' debug format @author Julia Menapace, Jim Kingdon, David MacKenzie @author Cygnus Support @page @tex \def\$#1${{#1}} % Kluge: collect RCS revision info without $...$ \xdef\manvers{\$Revision$} % For use in headers, footers too {\parskip=0pt \hfill Cygnus Support\par \hfill \manvers\par \hfill \TeX{}info \texinfoversion\par } @end tex @vskip 0pt plus 1filll Copyright @copyright{} 1992, 1993 Free Software Foundation, Inc. Contributed by Cygnus Support. Permission is granted to make and distribute verbatim copies of this manual provided the copyright notice and this permission notice are preserved on all copies. @end titlepage @ifinfo @node Top @top The "stabs" representation of debugging information This document describes the stabs debugging format. @menu * Overview:: Overview of stabs * Program structure:: Encoding of the structure of the program * Constants:: Constants * Example:: A comprehensive example in C * Variables:: * Types:: Type definitions * Symbol Tables:: Symbol information in symbol tables * Cplusplus:: Appendixes: * Example2.c:: Source code for extended example * Example2.s:: Assembly code for extended example * Stab Types:: Symbol types in a.out files * Symbol Descriptors:: Table of Symbol Descriptors * Type Descriptors:: Table of Symbol Descriptors * Expanded reference:: Reference information by stab type * Questions:: Questions and anomolies * XCOFF-differences:: Differences between GNU stabs in a.out and GNU stabs in XCOFF * Sun-differences:: Differences between GNU stabs and Sun native stabs * Stabs-in-ELF:: Stabs in an ELF file. @end menu @end ifinfo @node Overview @chapter Overview of stabs @dfn{Stabs} refers to a format for information that describes a program to a debugger. This format was apparently invented by @c FIXME! <> at the University of California at Berkeley, for the @code{pdx} Pascal debugger; the format has spread widely since then. This document is one of the few published sources of documentation on stabs. It is believed to be comprehensive for stabs used by C. The lists of symbol descriptors (@pxref{Symbol Descriptors}) and type descriptors (@pxref{Type Descriptors}) are believed to be completely comprehensive. Stabs for COBOL-specific features and for variant records (used by Pascal and Modula-2) are poorly documented here. Other sources of information on stabs are @cite{Dbx and Dbxtool Interfaces}, 2nd edition, by Sun, 1988, and @cite{AIX Version 3.2 Files Reference}, Fourth Edition, September 1992, "dbx Stabstring Grammar" in the a.out section, page 2-31. This document is believed to incorporate the information from those two sources except where it explictly directs you to them for more information. @menu * Flow:: Overview of debugging information flow * Stabs Format:: Overview of stab format * String Field:: The @code{.stabs} @var{string} field * C example:: A simple example in C source * Assembly code:: The simple example at the assembly level @end menu @node Flow @section Overview of debugging information flow The GNU C compiler compiles C source in a @file{.c} file into assembly language in a @file{.s} file, which the assembler translates into a @file{.o} file, which the linker combines with other @file{.o} files and libraries to produce an executable file. With the @samp{-g} option, GCC puts in the @file{.s} file additional debugging information, which is slightly transformed by the assembler and linker, and carried through into the final executable. This debugging information describes features of the source file like line numbers, the types and scopes of variables, and function names, parameters, and scopes. For some object file formats, the debugging information is encapsulated in assembler directives known collectively as @dfn{stab} (symbol table) directives, which are interspersed with the generated code. Stabs are the native format for debugging information in the a.out and XCOFF object file formats. The GNU tools can also emit stabs in the COFF and ECOFF object file formats. The assembler adds the information from stabs to the symbol information it places by default in the symbol table and the string table of the @file{.o} file it is building. The linker consolidates the @file{.o} files into one executable file, with one symbol table and one string table. Debuggers use the symbol and string tables in the executable as a source of debugging information about the program. @node Stabs Format @section Overview of stab format There are three overall formats for stab assembler directives, differentiated by the first word of the stab. The name of the directive describes which combination of four possible data fields follows. It is either @code{.stabs} (string), @code{.stabn} (number), or @code{.stabd} (dot). IBM's XCOFF assembler uses @code{.stabx} (and some other directives such as @code{.file} and @code{.bi}) instead of @code{.stabs}, @code{.stabn} or @code{.stabd}. The overall format of each class of stab is: @example .stabs "@var{string}",@var{type},0,@var{desc},@var{value} .stabn @var{type},0,@var{desc},@var{value} .stabd @var{type},0,@var{desc} .stabx "@var{string}",@var{value},@var{type},@var{sdb-type} @end example @c what is the correct term for "current file location"? My AIX @c assembler manual calls it "the value of the current location counter". For @code{.stabn} and @code{.stabd}, there is no @var{string} (the @code{n_strx} field is zero; see @xref{Symbol Tables}). For @code{.stabd}, the @var{value} field is implicit and has the value of the current file location. For @code{.stabx}, the @var{sdb-type} field is unused for stabs and can always be set to 0. The number in the @var{type} field gives some basic information about which type of stab this is (or whether it @emph{is} a stab, as opposed to an ordinary symbol). Each valid type number defines a different stab type. Further, the stab type defines the exact interpretation of, and possible values for, any remaining @var{string}, @var{desc}, or @var{value} fields present in the stab. @xref{Stab Types}, for a list in numeric order of the valid type field values for stab directives. @node String Field @section The @code{.stabs} @var{string} field For @code{.stabs} the @var{string} field holds the meat of the debugging information. The generally unstructured nature of this field is what makes stabs extensible. For some stab types the string field contains only a name. For other stab types the contents can be a great deal more complex. The overall format is of the @var{string} field is: @example "@var{name}:@var{symbol-descriptor} @var{type-information}" @end example @var{name} is the name of the symbol represented by the stab. @var{name} can be omitted, which means the stab represents an unnamed object. For example, @samp{:t10=*2} defines type 10 as a pointer to type 2, but does not give the type a name. Omitting the @var{name} field is supported by AIX dbx and GDB after about version 4.8, but not other debuggers. GCC sometimes uses a single space as the name instead of omitting the name altogether; apparently that is supported by most debuggers. The @var{symbol_descriptor} following the @samp{:} is an alphabetic character that tells more specifically what kind of symbol the stab represents. If the @var{symbol_descriptor} is omitted, but type information follows, then the stab represents a local variable. For a list of symbol descriptors, see @ref{Symbol Descriptors}. The @samp{c} symbol descriptor is an exception in that it is not followed by type information. @xref{Constants}. @var{type-information} is either a @var{type_number}, or @samp{@var{type_number}=}. The @var{type_number} alone is a type reference, referring directly to a type that has already been defined. The @samp{@var{type_number}=} form is a type definition, where the number represents a new type which is about to be defined. The type definition may refer to other types by number, and those type numbers may be followed by @samp{=} and nested definitions. In a type definition, if the character that follows the equals sign is non-numeric then it is a @var{type_descriptor}, and tells what kind of type is about to be defined. Any other values following the @var{type_descriptor} vary, depending on the @var{type_descriptor}. If a number follows the @samp{=} then the number is a @var{type_reference}. For a full description of types, @ref{Types}. @xref{Type Descriptors}, for a list of @var{type_descriptor} values. There is an AIX extension for type attributes. Following the @samp{=} is any number of type attributes. Each one starts with @samp{@@} and ends with @samp{;}. Debuggers, including AIX's dbx and GDB 4.10, skip any type attributes they do not recognize. GDB 4.9 and other versions of dbx may not do this. Because of a conflict with C++ (@pxref{Cplusplus}), new attributes should not be defined which begin with a digit, @samp{(}, or @samp{-}; GDB may be unable to distinguish those from the C++ type descriptor @samp{@@}. The attributes are: @table @code @item a@var{boundary} @var{boundary} is an integer specifying the alignment. I assume it applies to all variables of this type. @item s@var{size} Size in bits of a variable of this type. @item p@var{integer} Pointer class (for checking). Not sure what this means, or how @var{integer} is interpreted. @item P Indicate this is a packed type, meaning that structure fields or array elements are placed more closely in memory, to save memory at the expense of speed. @end table All this can make the @var{string} field quite long. All versions of GDB, and some versions of dbx, can handle arbitrarily long strings. But many versions of dbx cretinously limit the strings to about 80 characters, so compilers which must work with such dbx's need to split the @code{.stabs} directive into several @code{.stabs} directives. Each stab duplicates exactly all but the @var{string} field. The @var{string} field of every stab except the last is marked as continued with a double-backslash at the end. Removing the backslashes and concatenating the @var{string} fields of each stab produces the original, long string. @node C example @section A simple example in C source To get the flavor of how stabs describe source information for a C program, let's look at the simple program: @example main() @{ printf("Hello world"); @} @end example When compiled with @samp{-g}, the program above yields the following @file{.s} file. Line numbers have been added to make it easier to refer to parts of the @file{.s} file in the description of the stabs that follows. @node Assembly code @section The simple example at the assembly level This simple ``hello world'' example demonstrates several of the stab types used to describe C language source files. @example 1 gcc2_compiled.: 2 .stabs "/cygint/s1/users/jcm/play/",100,0,0,Ltext0 3 .stabs "hello.c",100,0,0,Ltext0 4 .text 5 Ltext0: 6 .stabs "int:t1=r1;-2147483648;2147483647;",128,0,0,0 7 .stabs "char:t2=r2;0;127;",128,0,0,0 8 .stabs "long int:t3=r1;-2147483648;2147483647;",128,0,0,0 9 .stabs "unsigned int:t4=r1;0;-1;",128,0,0,0 10 .stabs "long unsigned int:t5=r1;0;-1;",128,0,0,0 11 .stabs "short int:t6=r1;-32768;32767;",128,0,0,0 12 .stabs "long long int:t7=r1;0;-1;",128,0,0,0 13 .stabs "short unsigned int:t8=r1;0;65535;",128,0,0,0 14 .stabs "long long unsigned int:t9=r1;0;-1;",128,0,0,0 15 .stabs "signed char:t10=r1;-128;127;",128,0,0,0 16 .stabs "unsigned char:t11=r1;0;255;",128,0,0,0 17 .stabs "float:t12=r1;4;0;",128,0,0,0 18 .stabs "double:t13=r1;8;0;",128,0,0,0 19 .stabs "long double:t14=r1;8;0;",128,0,0,0 20 .stabs "void:t15=15",128,0,0,0 21 .align 4 22 LC0: 23 .ascii "Hello, world!\12\0" 24 .align 4 25 .global _main 26 .proc 1 27 _main: 28 .stabn 68,0,4,LM1 29 LM1: 30 !#PROLOGUE# 0 31 save %sp,-136,%sp 32 !#PROLOGUE# 1 33 call ___main,0 34 nop 35 .stabn 68,0,5,LM2 36 LM2: 37 LBB2: 38 sethi %hi(LC0),%o1 39 or %o1,%lo(LC0),%o0 40 call _printf,0 41 nop 42 .stabn 68,0,6,LM3 43 LM3: 44 LBE2: 45 .stabn 68,0,6,LM4 46 LM4: 47 L1: 48 ret 49 restore 50 .stabs "main:F1",36,0,0,_main 51 .stabn 192,0,0,LBB2 52 .stabn 224,0,0,LBE2 @end example @node Program structure @chapter Encoding for the structure of the program @menu * Main Program:: Indicate what the main program is * Source Files:: The path and name of the source file * Include Files:: Names of include files * Line Numbers:: * Procedures:: * Block Structure:: @end menu @node Main Program @section Main Program Most languages allow the main program to have any name. The @code{N_MAIN} stab type is used for a stab telling the debugger what name is used in this program. Only the name is significant; it will be the name of a function which is the main program. Most C compilers do not use this stab; they expect the debugger to simply assume that the name is @samp{main}, but some C compilers emit an @code{N_MAIN} stab for the @samp{main} function. @node Source Files @section Paths and names of the source files Before any other stabs occur, there must be a stab specifying the source file. This information is contained in a symbol of stab type @code{N_SO}; the string contains the name of the file. The value of the symbol is the start address of portion of the text section corresponding to that file. With the Sun Solaris2 compiler, the @code{desc} field contains a source-language code. Some compilers (for example, GCC2 and SunOS4 @file{/bin/cc}) also include the directory in which the source was compiled, in a second @code{N_SO} symbol preceding the one containing the file name. This symbol can be distinguished by the fact that it ends in a slash. Code from the cfront C++ compiler can have additional @code{N_SO} symbols for nonexistent source files after the @code{N_SO} for the real source file; these are believed to contain no useful information. For example: @example .stabs "/cygint/s1/users/jcm/play/",100,0,0,Ltext0 # @r{100 is N_SO} .stabs "hello.c",100,0,0,Ltext0 .text Ltext0: @end example Instead of @code{N_SO} symbols, XCOFF uses a @code{.file} assembler directive which assembles to a standard COFF @code{.file} symbol; explaining this in detail is outside the scope of this document. @node Include Files @section Names of include files There are several different schemes for dealing with include files: the traditional @code{N_SOL} approach, Sun's @code{N_BINCL} approach, and the XCOFF @code{C_BINCL} approach (which despite the similar name has little in common with @code{N_BINCL}). An @code{N_SOL} symbol specifies which include file subsequent symbols refer to. The string field is the name of the file and the value is the text address corresponding to the start of the previous include file and the start of this one. To specify the main source file again, use an @code{N_SOL} symbol with the name of the main source file. A @code{N_BINCL} symbol specifies the start of an include file. In an object file, only the name is significant. The Sun linker puts data into some of the other fields. The end of the include file is marked by a @code{N_EINCL} symbol (which has no name field). In an ojbect file, there is no significant data in the @code{N_EINCL} symbol; the Sun linker puts data into some of the fields. @code{N_BINCL} and @code{N_EINCL} can be nested. If the linker detects that two source files have identical stabs with a @code{N_BINCL} and @code{N_EINCL} pair (as will generally be the case for a header file), then it only puts out the stabs once. Each additional occurance is replaced by an @code{N_EXCL} symbol. I believe the Sun (SunOS4, not sure about Solaris) linker is the only one which supports this feature. For the start of an include file in XCOFF, use the @file{.bi} assembler directive, which generates a @code{C_BINCL} symbol. A @file{.ei} directive, which generates a @code{C_EINCL} symbol, denotes the end of the include file. Both directives are followed by the name of the source file in quotes, which becomes the string for the symbol. The value of each symbol, produced automatically by the assembler and linker, is an offset into the executable which points to the beginning (inclusive, as you'd expect) and end (inclusive, as you would not expect) of the portion of the COFF linetable which corresponds to this include file. @code{C_BINCL} and @code{C_EINCL} do not nest. @node Line Numbers @section Line Numbers A @code{N_SLINE} symbol represents the start of a source line. The @var{desc} field contains the line number and the @var{value} field contains the code address for the start of that source line. On most machines the address is absolute; for Sun's stabs-in-ELF, it is relative to the function in which the @code{N_SLINE} symbol occurs. GNU documents @code{N_DSLINE} and @code{N_BSLINE} symbols for line numbers in the data or bss segments, respectively. They are identical to @code{N_SLINE} but are relocated differently by the linker. They were intended to be used to describe the source location of a variable declaration, but I believe that GCC2 actually puts the line number in the desc field of the stab for the variable itself. GDB has been ignoring these symbols (unless they contain a string field) at least since GDB 3.5. XCOFF uses COFF line numbers instead, which are outside the scope of this document, ammeliorated by adequate marking of include files (@pxref{Include Files}). For single source lines that generate discontiguous code, such as flow of control statements, there may be more than one line number entry for the same source line. In this case there is a line number entry at the start of each code range, each with the same line number. @node Procedures @section Procedures All of the following stabs use the @code{N_FUN} symbol type. A function is represented by an @samp{F} symbol descriptor for a global (extern) function, and @samp{f} for a static (local) function. The next @code{N_SLINE} symbol can be used to find the line number of the start of the function. The value field is the address of the start of the function (absolute for @code{a.out}; relative to the start of the file for Sun's stabs-in-ELF). The type information of the stab represents the return type of the function; thus @samp{foo:f5} means that foo is a function returning type 5. The type information of the stab is optionally followed by type information for each argument, with each argument preceded by @samp{;}. An argument type of 0 means that additional arguments are being passed, whose types and number may vary (@samp{...} in ANSI C). This extension is used by Sun's Solaris compiler. GDB has tolerated it (i.e., at least parsed the syntax, if not necessarily used the information) at least since version 4.8; I don't know whether all versions of dbx will tolerate it. The argument types given here are not redundant with the symbols for the arguments themselves (@pxref{Parameters}), they are the types of the arguments as they are passed, before any conversions might take place. For example, if a C function which is declared without a prototype takes a @code{float} argument, the value is passed as a @code{double} but then converted to a @code{float}. Debuggers need to use the types given in the arguments when printing values, but if calling the function they need to use the types given in the symbol defining the function. If the return type and types of arguments of a function which is defined in another source file are specified (i.e., a function prototype in ANSI C), traditionally compilers emit no stab; the only way for the debugger to find the information is if the source file where the function is defined was also compiled with debugging symbols. As an extension the Solaris compiler uses symbol descriptor @samp{P} followed by the return type of the function, followed by the arguments, each preceded by @samp{;}, as in a stab with symbol descriptor @samp{f} or @samp{F}. This use of symbol descriptor @samp{P} can be distinguished from its use for register parameters (@pxref{Parameters}) by the fact that it has symbol type @code{N_FUN}. The AIX documentation also defines symbol descriptor @samp{J} as an internal function. I assume this means a function nested within another function. It also says symbol descriptor @samp{m} is a module in Modula-2 or extended Pascal. Procedures (functions which do not return values) are represented as functions returning the @code{void} type in C. I don't see why this couldn't be used for all languages (inventing a @code{void} type for this purpose if necessary), but the AIX documentation defines @samp{I}, @samp{P}, and @samp{Q} for internal, global, and static procedures, respectively. These symbol descriptors are unusual in that they are not followed by type information. For any of the above symbol descriptors, after the symbol descriptor and the type information, there is optionally a comma, followed by the name of the procedure, followed by a comma, followed by a name specifying the scope. The first name is local to the scope specified, and seems to be redundant with the name of the symbol (before the @samp{:}). The name specifying the scope is the name of a procedure specifying that scope. This feature is used by GCC, and presumably Pascal, Modula-2, etc., compilers, for nested functions. If procedures are nested more than one level deep, only the immediately containing scope is specified, for example: @example int foo (int x) @{ int bar (int y) @{ int baz (int z) @{ return x + y + z; @} return baz (x + 2 * y); @} return x + bar (3 * x); @} @end example @noindent produces the stabs: @example .stabs "baz:f1,baz,bar",36,0,0,_baz.15 # @r{36 is N_FUN} .stabs "bar:f1,bar,foo",36,0,0,_bar.12 .stabs "foo:F1",36,0,0,_foo @end example The stab representing a procedure is located immediately following the code of the procedure. This stab is in turn directly followed by a group of other stabs describing elements of the procedure. These other stabs describe the procedure's parameters, its block local variables, and its block structure. Going back to our ``hello world'' example program, @example 48 ret 49 restore @end example @noindent The @code{.stabs} entry after this code fragment shows the @var{name} of the procedure (@code{main}); the type descriptor @var{desc} (@code{F}, for a global procedure); a reference to the predefined type @code{int} for the return type; and the starting @var{address} of the procedure. Here is an exploded summary (with whitespace introduced for clarity), followed by line 50 of our sample assembly output, which has this form: @example .stabs "@var{name}: @var{desc} @r{(global proc @samp{F})} @var{return_type_ref} @r{(int)} ",N_FUN, NIL, NIL, @var{address} @end example @example 50 .stabs "main:F1",36,0,0,_main @end example @node Block Structure @section Block Structure The program's block structure is represented by the @code{N_LBRAC} (left brace) and the @code{N_RBRAC} (right brace) stab types. The variables defined inside a block precede the @code{N_LBRAC} symbol for most compilers, including GCC. Other compilers, such as the Convex, Acorn RISC machine, and Sun @code{acc} compilers, put the variables after the @code{N_LBRAC} symbol. The values of the @code{N_LBRAC} and @code{N_RBRAC} symbols are the start and end addresses of the code of the block, respectively. For most machines, they are relative to the starting address of this source file. For the Gould NP1, they are absolute. For Sun's stabs-in-ELF, they are relative to the function in which they occur. The @code{N_LBRAC} and @code{N_RBRAC} stabs that describe the block scope of a procedure are located after the @code{N_FUN} stab that represents the procedure itself. Sun documents the @code{desc} field of @code{N_LBRAC} and @code{N_RBRAC} symbols as containing the nesting level of the block. However, dbx seems not to care, and GCC always sets @code{desc} to zero. @node Constants @chapter Constants The @samp{c} symbol descriptor indicates that this stab represents a constant. This symbol descriptor is an exception to the general rule that symbol descriptors are followed by type information. Instead, it is followed by @samp{=} and one of the following: @table @code @item b @var{value} Boolean constant. @var{value} is a numeric value; I assume it is 0 for false or 1 for true. @item c @var{value} Character constant. @var{value} is the numeric value of the constant. @item e @var{type-information} , @var{value} Constant whose value can be represented as integral. @var{type-information} is the type of the constant, as it would appear after a symbol descriptor (@pxref{String Field}). @var{value} is the numeric value of the constant. GDB 4.9 does not actually get the right value if @var{value} does not fit in a host @code{int}, but it does not do anything violent, and future debuggers could be extended to accept integers of any size (whether unsigned or not). This constant type is usually documented as being only for enumeration constants, but GDB has never imposed that restriction; I don't know about other debuggers. @item i @var{value} Integer constant. @var{value} is the numeric value. The type is some sort of generic integer type (for GDB, a host @code{int}); to specify the type explicitly, use @samp{e} instead. @item r @var{value} Real constant. @var{value} is the real value, which can be @samp{INF} (optionally preceded by a sign) for infinity, @samp{QNAN} for a quiet NaN (not-a-number), or @samp{SNAN} for a signalling NaN. If it is a normal number the format is that accepted by the C library function @code{atof}. @item s @var{string} String constant. @var{string} is a string enclosed in either @samp{'} (in which case @samp{'} characters within the string are represented as @samp{\'} or @samp{"} (in which case @samp{"} characters within the string are represented as @samp{\"}). @item S @var{type-information} , @var{elements} , @var{bits} , @var{pattern} Set constant. @var{type-information} is the type of the constant, as it would appear after a symbol descriptor (@pxref{String Field}). @var{elements} is the number of elements in the set (Does this means how many bits of @var{pattern} are actually used, which would be redundant with the type, or perhaps the number of bits set in @var{pattern}? I don't get it), @var{bits} is the number of bits in the constant (meaning it specifies the length of @var{pattern}, I think), and @var{pattern} is a hexadecimal representation of the set. AIX documentation refers to a limit of 32 bytes, but I see no reason why this limit should exist. This form could probably be used for arbitrary constants, not just sets; the only catch is that @var{pattern} should be understood to be target, not host, byte order and format. @end table The boolean, character, string, and set constants are not supported by GDB 4.9, but it will ignore them. GDB 4.8 and earlier gave an error message and refused to read symbols from the file containing the constants. This information is followed by @samp{;}. @node Example @chapter A Comprehensive Example in C To describe the other stab types, we'll examine a second program, @code{example2}, which builds on the first example to introduce the rest of the stab types, symbol descriptors, and type descriptors used in C. @xref{Example2.c} for the complete @file{.c} source, and @pxref{Example2.s} for the @file{.s} assembly code. This description includes parts of those files. @section Flow of control and nested scopes @table @strong @item Directive: @code{.stabn} @item Types: @code{N_SLINE}, @code{N_LBRAC}, @code{N_RBRAC} (cont.) @end table Consider the body of @code{main}, from @file{example2.c}. It shows more about how @code{N_SLINE}, @code{N_RBRAC}, and @code{N_LBRAC} stabs are used. @example 20 @{ 21 static float s_flap; 22 int times; 23 for (times=0; times < s_g_repeat; times++)@{ 24 int inner; 25 printf ("Hello world\n"); 26 @} 27 @}; @end example Here we have a single source line, the @code{for} line, that generates non-linear flow of control, and non-contiguous code. In this case, an @code{N_SLINE} stab with the same line number proceeds each block of non-contiguous code generated from the same source line. The example also shows nested scopes. The @code{N_LBRAC} and @code{N_LBRAC} stabs that describe block structure are nested in the same order as the corresponding code blocks, those of the for loop inside those for the body of main. @noindent This is the label for the @code{N_LBRAC} (left brace) stab marking the start of @code{main}. @example 57 LBB2: @end example @noindent In the first code range for C source line 23, the @code{for} loop initialize and test, @code{N_SLINE} (68) records the line number: @example .stabn N_SLINE, NIL, @var{line}, @var{address} 58 .stabn 68,0,23,LM2 59 LM2: 60 st %g0,[%fp-20] 61 L2: 62 sethi %hi(_s_g_repeat),%o0 63 ld [%fp-20],%o1 64 ld [%o0+%lo(_s_g_repeat)],%o0 65 cmp %o1,%o0 66 bge L3 67 nop @exdent label for the @code{N_LBRAC} (start block) marking the start of @code{for} loop 68 LBB3: 69 .stabn 68,0,25,LM3 70 LM3: 71 sethi %hi(LC0),%o1 72 or %o1,%lo(LC0),%o0 73 call _printf,0 74 nop 75 .stabn 68,0,26,LM4 76 LM4: @exdent label for the @code{N_RBRAC} (end block) stab marking the end of the @code{for} loop 77 LBE3: @end example @noindent Now we come to the second code range for source line 23, the @code{for} loop increment and return. Once again, @code{N_SLINE} (68) records the source line number: @example .stabn, N_SLINE, NIL, @var{line}, @var{address} 78 .stabn 68,0,23,LM5 79 LM5: 80 L4: 81 ld [%fp-20],%o0 82 add %o0,1,%o1 83 st %o1,[%fp-20] 84 b,a L2 85 L3: 86 .stabn 68,0,27,LM6 87 LM6: @exdent label for the @code{N_RBRAC} (end block) stab marking the end of the @code{for} loop 88 LBE2: 89 .stabn 68,0,27,LM7 90 LM7: 91 L1: 92 ret 93 restore 94 .stabs "main:F1",36,0,0,_main 95 .stabs "argc:p1",160,0,0,68 96 .stabs "argv:p20=*21=*2",160,0,0,72 97 .stabs "s_flap:V12",40,0,0,_s_flap.0 98 .stabs "times:1",128,0,0,-20 @end example @noindent Here is an illustration of stabs describing nested scopes. The scope nesting is reflected in the nested bracketing stabs (@code{N_LBRAC}, 192, appears here). @example .stabn N_LBRAC,NIL,NIL, @var{block-start-address} 99 .stabn 192,0,0,LBB2 ## begin proc label 100 .stabs "inner:1",128,0,0,-24 101 .stabn 192,0,0,LBB3 ## begin for label @end example @noindent @code{N_RBRAC} (224), ``right brace'' ends a lexical block (scope). @example .stabn N_RBRAC,NIL,NIL, @var{block-end-address} 102 .stabn 224,0,0,LBE3 ## end for label 103 .stabn 224,0,0,LBE2 ## end proc label @end example @node Variables @chapter Variables @menu * Stack Variables:: Variables allocated on the stack. * Global Variables:: Variables used by more than one source file. * Register variables:: Variables in registers. * Common Blocks:: Variables statically allocated together. * Statics:: Variables local to one source file. * Parameters:: Variables for arguments to functions. @end menu @node Stack Variables @section Automatic Variables Allocated on the Stack If a variable is declared whose scope is local to a function and whose lifetime is only as long as that function executes (C calls such variables automatic), they can be allocated in a register (@pxref{Register variables}) or on the stack. For variables allocated on the stack, each variable has a stab with the symbol descriptor omitted. Since type information should being with a digit, @samp{-}, or @samp{(}, only digits, @samp{-}, and @samp{(} are precluded from being used for symbol descriptors by this fact. However, the Acorn RISC machine (ARM) is said to get this wrong: it puts out a mere type definition here, without the preceding @code{@var{typenumber}=}. This is a bad idea; there is no guarantee that type descriptors are distinct from symbol descriptors. These stabs have the @code{N_LSYM} stab type. The value of the stab is the offset of the variable within the local variables. On most machines this is an offset from the frame pointer and is negative. The location of the stab specifies what block it is defined in; see @ref{Block Structure}. For example, the following C code @example int main () @{ int x; @} @end example produces the following stabs @example .stabs "main:F1",36,0,0,_main # @r{36 is N_FUN} .stabs "x:1",128,0,0,-12 # @r{128 is N_LSYM} .stabn 192,0,0,LBB2 # @r{192 is N_LBRAC} .stabn 224,0,0,LBE2 # @r{224 is N_RBRAC} @end example @xref{Procedures} for more information on the first stab, and @ref{Block Structure} for more information on the @code{N_LBRAC} and @code{N_RBRAC} stabs. @node Global Variables @section Global Variables A variable whose scope which is not specific to just one source file is represented by the @samp{G} symbol descriptor. These stabs use the @code{N_GSYM} stab type. The type information for the stab (@pxref{String Field}) gives the type of the variable. For example, the following source code: @example char g_foo = 'c'; @end example @noindent yields the following assembly code: @example .stabs "g_foo:G2",32,0,0,0 # @r{32 is N_GSYM} .global _g_foo .data _g_foo: .byte 99 @end example The address of the variable represented by the @code{N_GSYM} is not contained in the @code{N_GSYM} stab. The debugger gets this information from the external symbol for the global variable. In the example above, the @code{.global _g_foo} and @code{_g_foo:} lines tell the assembler to produce an external symbol. @node Register variables @section Register variables @c According to an old version of this manual, AIX uses C_RPSYM instead @c of C_RSYM. I am skeptical; this should be verified. Register variables have their own stab type, @code{N_RSYM}, and their own symbol descriptor, @samp{r}. The stab's value field contains the number of the register where the variable data will be stored. The value is the register number. AIX defines a separate symbol descriptor @samp{d} for floating point registers. This seems unnecessary; why not just just give floating point registers different register numbers? I have not verified whether the compiler actually uses @samp{d}. If the register is explicitly allocated to a global variable, but not initialized, as in @example register int g_bar asm ("%g5"); @end example the stab may be emitted at the end of the object file, with the other bss symbols. @node Common Blocks @section Common Blocks A common block is a statically allocated section of memory which can be referred to by several source files. It may contain several variables. I believe Fortran is the only language with this feature. A @code{N_BCOMM} stab begins a common block and an @code{N_ECOMM} stab ends it. The only thing which is significant about these two stabs is their name, which can be used to look up a normal (non-debugging) symbol which gives the address of the common block. Then each stab between the @code{N_BCOMM} and the @code{N_ECOMM} specifies a member of that common block; its value is the offset within the common block of that variable. The @code{N_ECOML} stab type is documented for this purpose, but Sun's Fortran compiler uses @code{N_GSYM} instead. The test case I looked at had a common block local to a function and it used the @samp{V} symbol descriptor; I assume one would use @samp{S} if not local to a function (that is, if a common block @emph{can} be anything other than local to a function). @node Statics @section Static Variables Initialized static variables are represented by the @samp{S} and @samp{V} symbol descriptors. @samp{S} means file scope static, and @samp{V} means procedure scope static. @c This is probably not worth mentioning; it is only true on the sparc @c for `double' variables which although declared const are actually in @c the data segment (the text segment can't guarantee 8 byte alignment). @c (although GCC @c 2.4.5 has a bug in that it uses @code{N_FUN}, so neither dbx nor GDB can @c find the variables) In a.out files, @code{N_STSYM} means the data segment, @code{N_FUN} means the text segment, and @code{N_LCSYM} means the bss segment. In XCOFF files, each symbol has a section number, so the stab type need not indicate the segment. In ECOFF files, the storage class is used to specify the section, so the stab type need not indicate the segment. @c In ELF files, it apparently is a big mess. See kludge in dbxread.c @c in GDB. FIXME: Investigate where this kludge comes from. @c @c This is the place to mention N_ROSYM; I'd rather do so once I can @c coherently explain how this stuff works for stabs-in-ELF. @c For example, the source lines @example static const int var_const = 5; static int var_init = 2; static int var_noinit; @end example @noindent yield the following stabs: @example .stabs "var_const:S1",36,0,0,_var_const # @r{36 is N_FUN} . . . .stabs "var_init:S1",38,0,0,_var_init # @r{38 is N_STSYM} . . . .stabs "var_noinit:S1",40,0,0,_var_noinit # @r{40 is N_LCSYM} @end example @node Parameters @section Parameters Parameters to a function are represented by a stab (or sometimes two, see below) for each parameter. The stabs are in the order in which the debugger should print the parameters (i.e., the order in which the parameters are declared in the source file). The exact form of the stab depends on how the parameter is being passed. Parameters passed on the stack use the symbol descriptor @samp{p}, and the @code{N_PSYM} symbol type. The value of the symbol is an offset used to locate the parameter on the stack; its exact meaning is machine-dependent but on most machines it is an offset from the frame pointer. If the parameter is passed in a register, then the traditional way to do this is to provide two symbols for each argument: @example .stabs "arg:p1" . . . ; N_PSYM .stabs "arg:r1" . . . ; N_RSYM @end example Debuggers are expected to use the second one to find the value, and the first one to know that it is an argument. Because this is kind of ugly, some compilers use symbol descriptor @samp{P} or @samp{R} to indicate an argument which is in a register. The symbol value is the register number. @samp{P} and @samp{R} mean the same thing, the difference is that @samp{P} is a GNU invention and @samp{R} is an IBM (XCOFF) invention. As of version 4.9, GDB should handle either one. Symbol type @code{C_RPSYM} is used with @samp{R} and @code{N_RSYM} is used with @samp{P}. According to the AIX documentation symbol descriptor @samp{D} is for a parameter passed in a floating point register. This seems unnecessary---why not just use @samp{R} with a register number which indicates that it's a floating point register? I haven't verified whether the system actually does what the documentation indicates. There is at least one case where GCC uses a @samp{p} and @samp{r} pair rather than @samp{P}; this is where the argument is passed in the argument list and then loaded into a register. On the sparc and hppa, for a @samp{P} symbol whose type is a structure or union, the register contains the address of the structure. On the sparc, this is also true of a @samp{p} and @samp{r} pair (using Sun cc) or a @samp{p} symbol. However, if a (small) structure is really in a register, @samp{r} is used. And, to top it all off, on the hppa it might be a structure which was passed on the stack and loaded into a register and for which there is a @samp{p} and @samp{r} pair! I believe that symbol descriptor @samp{i} is supposed to deal with this case, (it is said to mean "value parameter by reference, indirect access", I don't know the source for this information) but I don't know details or what compilers or debuggers use it, if any (not GDB or GCC). It is not clear to me whether this case needs to be dealt with differently than parameters passed by reference (see below). There is another case similar to an argument in a register, which is an argument which is actually stored as a local variable. Sometimes this happens when the argument was passed in a register and then the compiler stores it as a local variable. If possible, the compiler should claim that it's in a register, but this isn't always done. Some compilers use the pair of symbols approach described above (@samp{@var{arg}:p} followed by @samp{@var{arg}:}); this includes GCC1 (not GCC2) on the sparc when passing a small structure and GCC2 (sometimes) when the argument type is float and it is passed as a double and converted to float by the prologue (in the latter case the type of the @samp{@var{arg}:p} symbol is double and the type of the @samp{@var{arg}:} symbol is float). GCC, at least on the 960, uses a single @samp{p} symbol descriptor for an argument which is stored as a local variable but uses @code{N_LSYM} instead of @code{N_PSYM}. In this case the value of the symbol is an offset relative to the local variables for that function, not relative to the arguments (on some machines those are the same thing, but not on all). If the parameter is passed by reference (e.g., Pascal VAR parameters), then type symbol descriptor is @samp{v} if it is in the argument list, or @samp{a} if it in a register. Other than the fact that these contain the address of the parameter other than the parameter itself, they are identical to @samp{p} and @samp{R}, respectively. I believe @samp{a} is an AIX invention; @samp{v} is supported by all stabs-using systems as far as I know. @c Is this paragraph correct? It is based on piecing together patchy @c information and some guesswork Conformant arrays refer to a feature of Modula-2, and perhaps other languages, in which the size of an array parameter is not known to the called function until run-time. Such parameters have two stabs, a @samp{x} for the array itself, and a @samp{C}, which represents the size of the array. The value of the @samp{x} stab is the offset in the argument list where the address of the array is stored (it this right? it is a guess); the value of the @samp{C} stab is the offset in the argument list where the size of the array (in elements? in bytes?) is stored. The following are also said to go with @code{N_PSYM}: @example "name" -> "param_name:#type" -> pP (<>) -> pF Fortran function parameter -> X (function result variable) -> b (based variable) value -> offset from the argument pointer (positive). @end example As a simple example, the code @example main (argc, argv) int argc; char **argv; @{ @end example produces the stabs @example .stabs "main:F1",36,0,0,_main # @r{36 is N_FUN} .stabs "argc:p1",160,0,0,68 # @r{160 is N_PSYM} .stabs "argv:p20=*21=*2",160,0,0,72 @end example The type definition of argv is interesting because it contains several type definitions. Type 21 is pointer to type 2 (char) and argv (type 20) is pointer to type 21. @node Types @chapter Defining Types Now let's look at some variable definitions involving complex types. This involves understanding better how types are described. In the examples so far types have been described as references to previously defined types or defined in terms of subranges of or pointers to previously defined types. The section that follows discusses the other type descriptors that may follow the @samp{=} sign in a type definition. @menu * Builtin types:: Integers, floating point, void, etc. * Miscellaneous Types:: Pointers, sets, files, etc. * Cross-references:: Referring to a type not yet defined. * Subranges:: A type with a specific range. * Arrays:: An aggregate type of same-typed elements. * Strings:: Like an array but also has a length. * Enumerations:: Like an integer but the values have names. * Structures:: An aggregate type of different-typed elements. * Typedefs:: Giving a type a name. * Unions:: Different types sharing storage. * Function Types:: @end menu @node Builtin types @section Builtin types Certain types are built in (@code{int}, @code{short}, @code{void}, @code{float}, etc.); the debugger recognizes these types and knows how to handle them. Thus don't be surprised if some of the following ways of specifying builtin types do not specify everything that a debugger would need to know about the type---in some cases they merely specify enough information to distinguish the type from other types. The traditional way to define builtin types is convolunted, so new ways have been invented to describe them. Sun's @code{acc} uses special builtin type descriptors (@samp{b} and @samp{R}), and IBM uses negative type numbers. GDB can accept all three, as of version 4.8; dbx just accepts the traditional builtin types and perhaps one of the other two formats. The following sections describe each of these formats. @menu * Traditional Builtin Types:: Put on your seatbelts and prepare for kludgery * Builtin Type Descriptors:: Builtin types with special type descriptors * Negative Type Numbers:: Builtin types using negative type numbers @end menu @node Traditional Builtin Types @subsection Traditional Builtin types Often types are defined as subranges of themselves. If the array bounds can fit within an @code{int}, then they are given normally. For example: @example .stabs "int:t1=r1;-2147483648;2147483647;",128,0,0,0 # @r{128 is N_LSYM} .stabs "char:t2=r2;0;127;",128,0,0,0 @end example Builtin types can also be described as subranges of @code{int}: @example .stabs "unsigned short:t6=r1;0;65535;",128,0,0,0 @end example If the lower bound of a subrange is 0 and the upper bound is -1, it means that the type is an unsigned integral type whose bounds are too big to describe in an int. Traditionally this is only used for @code{unsigned int} and @code{unsigned long}. For larger types, GCC 2.4.5 puts out bounds in octal, with a leading 0. In this case a negative bound consists of a number which is a 1 bit followed by a bunch of 0 bits, and a positive bound is one in which a bunch of bits are 1. All known versions of dbx and GDB version 4 accept this, but GDB 3.5 refuses to read the whole file containing such symbols. So GCC 2.3.3 did not output the proper size for these types. @c FIXME: Update this for the 2.4.5 output, not 2.3.3 @example .stabs "unsigned int:t4=r1;0;-1;",128,0,0,0 .stabs "long long int:t7=r1;0;-1;",128,0,0,0 @end example If the lower bound of a subrange is 0 and the upper bound is negative, it means that it is an unsigned integral type whose size in bytes is the absolute value of the upper bound. I believe this is a Convex convention for @code{unsigned long long}. If the lower bound of a subrange is negative and the upper bound is 0, it means that the type is a signed integral type whose size in bytes is the absolute value of the lower bound. I believe this is a Convex convention for @code{long long}. To distinguish this from a legitimate subrange, the type should be a subrange of itself. I'm not sure whether this is the case for Convex. If the upper bound of a subrange is 0, it means that this is a floating point type, and the lower bound of the subrange indicates the number of bytes in the type: @example .stabs "float:t12=r1;4;0;",128,0,0,0 .stabs "double:t13=r1;8;0;",128,0,0,0 @end example However, GCC writes @code{long double} the same way it writes @code{double}, so there is no way to distinguish. @example .stabs "long double:t14=r1;8;0;",128,0,0,0 @end example Complex types are defined the same way as floating-point types; there is no way to distinguish a single-precision complex from a double-precision floating-point type. The C @code{void} type is defined as itself: @example .stabs "void:t15=15",128,0,0,0 @end example I'm not sure how a boolean type is represented. @node Builtin Type Descriptors @subsection Defining Builtin Types using Builtin Type Descriptors There are various type descriptors to define builtin types: @table @code @c FIXME: clean up description of width and offset, once we figure out @c what they mean @item b @var{signed} @var{char-flag} @var{width} ; @var{offset} ; @var{nbits} ; Define an integral type. @var{signed} is @samp{u} for unsigned or @samp{s} for signed. @var{char-flag} is @samp{c} which indicates this is a character type, or is omitted. I assume this is to distinguish an integral type from a character type of the same size, for example it might make sense to set it for the C type @code{wchar_t} so the debugger can print such variables differently (Solaris does not do this). Sun sets it on the C types @code{signed char} and @code{unsigned char} which arguably is wrong. @var{width} and @var{offset} appear to be for small objects stored in larger ones, for example a @code{short} in an @code{int} register. @var{width} is normally the number of bytes in the type. @var{offset} seems to always be zero. @var{nbits} is the number of bits in the type. Note that type descriptor @samp{b} used for builtin types conflicts with its use for Pascal space types (@pxref{Miscellaneous Types}); they can be distinguished because the character following the type descriptor will be a digit, @samp{(}, or @samp{-} for a Pascal space type, or @samp{u} or @samp{s} for a builtin type. @item w Documented by AIX to define a wide character type, but their compiler actually uses negative type numbers (@pxref{Negative Type Numbers}). @item R @var{fp_type} ; @var{bytes} ; Define a floating point type. @var{fp_type} has one of the following values: @table @code @item 1 (NF_SINGLE) IEEE 32-bit (single precision) floating point format. @item 2 (NF_DOUBLE) IEEE 64-bit (double precision) floating point format. @item 3 (NF_COMPLEX) @item 4 (NF_COMPLEX16) @item 5 (NF_COMPLEX32) @c "GDB source" really means @file{include/aout/stab_gnu.h}, but trying @c to put that here got an overfull hbox. These are for complex numbers. A comment in the GDB source describes them as Fortran complex, double complex, and complex*16, respectively, but what does that mean? (i.e., Single precision? Double precison?). @item 6 (NF_LDOUBLE) Long double. This should probably only be used for Sun format long double, and new codes should be used for other floating point formats (@code{NF_DOUBLE} can be used if a long double is really just an IEEE double, of course). @end table @var{bytes} is the number of bytes occupied by the type. This allows a debugger to perform some operations with the type even if it doesn't understand @var{fp_code}. @item g @var{type-information} ; @var{nbits} Documented by AIX to define a floating type, but their compiler actually uses negative type numbers (@pxref{Negative Type Numbers}). @item c @var{type-information} ; @var{nbits} Documented by AIX to define a complex type, but their compiler actually uses negative type numbers (@pxref{Negative Type Numbers}). @end table The C @code{void} type is defined as a signed integral type 0 bits long: @example .stabs "void:t19=bs0;0;0",128,0,0,0 @end example The Solaris compiler seems to omit the trailing semicolon in this case. Getting sloppy in this way is not a swift move because if a type is embedded in a more complex expression it is necessary to be able to tell where it ends. I'm not sure how a boolean type is represented. @node Negative Type Numbers @subsection Negative Type numbers Since the debugger knows about the builtin types anyway, the idea of negative type numbers is simply to give a special type number which indicates the built in type. There is no stab defining these types. I'm not sure whether anyone has tried to define what this means if @code{int} can be other than 32 bits (or other types can be other than their customary size). If @code{int} has exactly one size for each architecture, then it can be handled easily enough, but if the size of @code{int} can vary according the compiler options, then it gets hairy. The best way to do this would be to define separate negative type numbers for 16-bit @code{int} and 32-bit @code{int}; therefore I have indicated below the customary size (and other format information) for each type. The information below is currently correct because AIX on the RS6000 is the only system which uses these type numbers. If these type numbers start to get used on other systems, I suspect the correct thing to do is to define a new number in cases where a type does not have the size and format indicated below (or avoid negative type numbers in these cases). Also note that part of the definition of the negative type number is the name of the type. Types with identical size and format but different names have different negative type numbers. @table @code @item -1 @code{int}, 32 bit signed integral type. @item -2 @code{char}, 8 bit type holding a character. Both GDB and dbx on AIX treat this as signed. GCC uses this type whether @code{char} is signed or not, which seems like a bad idea. The AIX compiler (xlc) seems to avoid this type; it uses -5 instead for @code{char}. @item -3 @code{short}, 16 bit signed integral type. @item -4 @code{long}, 32 bit signed integral type. @item -5 @code{unsigned char}, 8 bit unsigned integral type. @item -6 @code{signed char}, 8 bit signed integral type. @item -7 @code{unsigned short}, 16 bit unsigned integral type. @item -8 @code{unsigned int}, 32 bit unsigned integral type. @item -9 @code{unsigned}, 32 bit unsigned integral type. @item -10 @code{unsigned long}, 32 bit unsigned integral type. @item -11 @code{void}, type indicating the lack of a value. @item -12 @code{float}, IEEE single precision. @item -13 @code{double}, IEEE double precision. @item -14 @code{long double}, IEEE double precision. The compiler claims the size will increase in a future release, and for binary compatibility you have to avoid using @code{long double}. I hope when they increase it they use a new negative type number. @item -15 @code{integer}. 32 bit signed integral type. @item -16 @code{boolean}. 32 bit type. How is the truth value encoded? Is it the least significant bit or is it a question of whether the whole value is zero or non-zero? @item -17 @code{short real}. IEEE single precision. @item -18 @code{real}. IEEE double precision. @item -19 @code{stringptr}. @xref{Strings}. @item -20 @code{character}, 8 bit unsigned character type. @item -21 @code{logical*1}, 8 bit type. This Fortran type has a split personality in that it is used for boolean variables, but can also be used for unsigned integers. 0 is false, 1 is true, and other values are non-boolean. @item -22 @code{logical*2}, 16 bit type. This Fortran type has a split personality in that it is used for boolean variables, but can also be used for unsigned integers. 0 is false, 1 is true, and other values are non-boolean. @item -23 @code{logical*4}, 32 bit type. This Fortran type has a split personality in that it is used for boolean variables, but can also be used for unsigned integers. 0 is false, 1 is true, and other values are non-boolean. @item -24 @code{logical}, 32 bit type. This Fortran type has a split personality in that it is used for boolean variables, but can also be used for unsigned integers. 0 is false, 1 is true, and other values are non-boolean. @item -25 @code{complex}. A complex type consisting of two IEEE single-precision floating point values. @item -26 @code{complex}. A complex type consisting of two IEEE double-precision floating point values. @item -27 @code{integer*1}, 8 bit signed integral type. @item -28 @code{integer*2}, 16 bit signed integral type. @item -29 @code{integer*4}, 32 bit signed integral type. @item -30 @code{wchar}. Wide character, 16 bits wide, unsigned (what format? Unicode?). @end table @node Miscellaneous Types @section Miscellaneous Types @table @code @item b @var{type-information} ; @var{bytes} Pascal space type. This is documented by IBM; what does it mean? Note that this use of the @samp{b} type descriptor can be distinguished from its use for builtin integral types (@pxref{Builtin Type Descriptors}) because the character following the type descriptor is always a digit, @samp{(}, or @samp{-}. @item B @var{type-information} A volatile-qualified version of @var{type-information}. This is a Sun extension. A volatile-qualified type means that references and stores to a variable of that type must not be optimized or cached; they must occur as the user specifies them. @item d @var{type-information} File of type @var{type-information}. As far as I know this is only used by Pascal. @item k @var{type-information} A const-qualified version of @var{type-information}. This is a Sun extension. A const-qualified type means that a variable of this type cannot be modified. @item M @var{type-information} ; @var{length} Multiple instance type. The type seems to composed of @var{length} repetitions of @var{type-information}, for example @code{character*3} is represented by @samp{M-2;3}, where @samp{-2} is a reference to a character type (@pxref{Negative Type Numbers}). I'm not sure how this differs from an array. This appears to be a Fortran feature. @var{length} is a bound, like those in range types; see @ref{Subranges}. @item S @var{type-information} Pascal set type. @var{type-information} must be a small type such as an enumeration or a subrange, and the type is a bitmask whose length is specified by the number of elements in @var{type-information}. @item * @var{type-information} Pointer to @var{type-information}. @end table @node Cross-references @section Cross-references to other types If a type is used before it is defined, one common way to deal with this is just to use a type reference to a type which has not yet been defined. The debugger is expected to be able to deal with this. Another way is with the @samp{x} type descriptor, which is followed by @samp{s} for a structure tag, @samp{u} for a union tag, or @samp{e} for a enumerator tag, followed by the name of the tag, followed by @samp{:}. for example the following C declarations: @example struct foo; struct foo *bar; @end example produce @example .stabs "bar:G16=*17=xsfoo:",32,0,0,0 @end example Not all debuggers support the @samp{x} type descriptor, so on some machines GCC does not use it. I believe that for the above example it would just emit a reference to type 17 and never define it, but I haven't verified that. Modula-2 imported types, at least on AIX, use the @samp{i} type descriptor, which is followed by the name of the module from which the type is imported, followed by @samp{:}, followed by the name of the type. There is then optionally a comma followed by type information for the type (This differs from merely naming the type (@pxref{Typedefs}) in that it identifies the module; I don't understand whether the name of the type given here is always just the same as the name we are giving it, or whether this type descriptor is used with a nameless stab (@pxref{String Field}), or what). The symbol ends with @samp{;}. @node Subranges @section Subrange types The @samp{r} type descriptor defines a type as a subrange of another type. It is followed by type information for the type which it is a subrange of, a semicolon, an integral lower bound, a semicolon, an integral upper bound, and a semicolon. The AIX documentation does not specify the trailing semicolon, in an effort to specify array indexes more cleanly, but a subrange which is not an array index has always included a trailing semicolon (@pxref{Arrays}). Instead of an integer, either bound can be one of the following: @table @code @item A @var{offset} The bound is passed by reference on the stack at offset @var{offset} from the argument list. @xref{Parameters}, for more information on such offsets. @item T @var{offset} The bound is passed by value on the stack at offset @var{offset} from the argument list. @item a @var{register-number} The bound is pased by reference in register number @var{register-number}. @item t @var{register-number} The bound is passed by value in register number @var{register-number}. @item J There is no bound. @end table Subranges are also used for builtin types; see @ref{Traditional Builtin Types}. @node Arrays @section Array types Arrays use the @samp{a} type descriptor. Following the type descriptor is the type of the index and the type of the array elements. If the index type is a range type, it will end in a semicolon; if it is not a range type (for example, if it is a type reference), there does not appear to be any way to tell where the types are separated. In an effort to clean up this mess, IBM documents the two types as being separated by a semicolon, and a range type as not ending in a semicolon (but this is not right for range types which are not array indexes, @pxref{Subranges}). I think probably the best solution is to specify that a semicolon ends a range type, and that the index type and element type of an array are separated by a semicolon, but that if the index type is a range type, the extra semicolon can be omitted. GDB (at least through version 4.9) doesn't support any kind of index type other than a range anyway; I'm not sure about dbx. It is well established, and widely used, that the type of the index, unlike most types found in the stabs, is merely a type definition, not type information (@pxref{String Field}) (that is, it need not start with @var{type-number}@code{=} if it is defining a new type). According to a comment in GDB, this is also true of the type of the array elements; it gives @samp{ar1;1;10;ar1;1;10;4} as a legitimate way to express a two dimensional array. According to AIX documentation, the element type must be type information. GDB accepts either. The type of the index is often a range type, expressed as the letter r and some parameters. It defines the size of the array. In the example below, the range @code{r1;0;2;} defines an index type which is a subrange of type 1 (integer), with a lower bound of 0 and an upper bound of 2. This defines the valid range of subscripts of a three-element C array. For example, the definition @example char char_vec[3] = @{'a','b','c'@}; @end example @noindent produces the output @example .stabs "char_vec:G19=ar1;0;2;2",32,0,0,0 .global _char_vec .align 4 _char_vec: .byte 97 .byte 98 .byte 99 @end example If an array is @dfn{packed}, it means that the elements are spaced more closely than normal, saving memory at the expense of speed. For example, an array of 3-byte objects might, if unpacked, have each element aligned on a 4-byte boundary, but if packed, have no padding. One way to specify that something is packed is with type attributes (@pxref{String Field}), in the case of arrays another is to use the @samp{P} type descriptor instead of @samp{a}. Other than specifying a packed array, @samp{P} is identical to @samp{a}. @c FIXME-what is it? A pointer? An open array is represented by the @samp{A} type descriptor followed by type information specifying the type of the array elements. @c FIXME: what is the format of this type? A pointer to a vector of pointers? An N-dimensional dynamic array is represented by @example D @var{dimensions} ; @var{type-information} @end example @c Does dimensions really have this meaning? The AIX documentation @c doesn't say. @var{dimensions} is the number of dimensions; @var{type-information} specifies the type of the array elements. @c FIXME: what is the format of this type? A pointer to some offsets in @c another array? A subarray of an N-dimensional array is represented by @example E @var{dimensions} ; @var{type-information} @end example @c Does dimensions really have this meaning? The AIX documentation @c doesn't say. @var{dimensions} is the number of dimensions; @var{type-information} specifies the type of the array elements. @node Strings @section Strings Some languages, like C or the original Pascal, do not have string types, they just have related things like arrays of characters. But most Pascals and various other languages have string types, which are indicated as follows: @table @code @item n @var{type-information} ; @var{bytes} @var{bytes} is the maximum length. I'm not sure what @var{type-information} is; I suspect that it means that this is a string of @var{type-information} (thus allowing a string of integers, a string of wide characters, etc., as well as a string of characters). Not sure what the format of this type is. This is an AIX feature. @item z @var{type-information} ; @var{bytes} Just like @samp{n} except that this is a gstring, not an ordinary string. I don't know the difference. @item N Pascal Stringptr. What is this? This is an AIX feature. @end table @node Enumerations @section Enumerations Enumerations are defined with the @samp{e} type descriptor. @c FIXME: Where does this information properly go? Perhaps it is @c redundant with something we already explain. The source line below declares an enumeration type. It is defined at file scope between the bodies of main and @code{s_proc} in @file{example2.c}. The type definition is located after the @code{N_RBRAC} that marks the end of the previous procedure's block scope, and before the @code{N_FUN} that marks the beginning of the next procedure's block scope. Therefore it does not describe a block local symbol, but a file local one. The source line: @example enum e_places @{first,second=3,last@}; @end example @noindent generates the following stab @example .stabs "e_places:T22=efirst:0,second:3,last:4,;",128,0,0,0 @end example The symbol descriptor (T) says that the stab describes a structure, enumeration, or union tag. The type descriptor e, following the 22= of the type definition narrows it down to an enumeration type. Following the e is a list of the elements of the enumeration. The format is @var{name:value,}. The list of elements ends with a @samp{;}. There is no standard way to specify the size of an enumeration type; it is determined by the architecture (normally all enumerations types are 32 bits). There should be a way to specify an enumeration type of another size; type attributes would be one way to do this @xref{Stabs Format}. @node Structures @section Structures @table @strong @item Directive: @code{.stabs} @item Type: @code{N_LSYM} or @code{C_DECL} @item Symbol Descriptor: @code{T} @item Type Descriptor: @code{s} @end table The following source code declares a structure tag and defines an instance of the structure in global scope. Then a typedef equates the structure tag with a new type. A seperate stab is generated for the structure tag, the structure typedef, and the structure instance. The stabs for the tag and the typedef are emited when the definitions are encountered. Since the structure elements are not initialized, the stab and code for the structure variable itself is located at the end of the program in .common. @example 6 struct s_tag @{ 7 int s_int; 8 float s_float; 9 char s_char_vec[8]; 10 struct s_tag* s_next; 11 @} g_an_s; 12 13 typedef struct s_tag s_typedef; @end example The structure tag is an @code{N_LSYM} stab type because, like the enum, the symbol is file scope. Like the enum, the symbol descriptor is @samp{T}, for enumeration, struct or tag type. The symbol descriptor @samp{s} following the @samp{16=} of the type definition narrows the symbol type to struct. Following the struct symbol descriptor is the number of bytes the struct occupies, followed by a description of each structure element. The structure element descriptions are of the form @var{name:type, bit offset from the start of the struct, number of bits in the element}. @example <128> N_LSYM - type definition .stabs "name:sym_desc(struct tag) Type_def(16)=type_desc(struct type) struct_bytes elem_name:type_ref(int),bit_offset,field_bits; elem_name:type_ref(float),bit_offset,field_bits; elem_name:type_def(17)=type_desc(array) index_type(range of int from 0 to 7); element_type(char),bit_offset,field_bits;;", N_LSYM,NIL,NIL,NIL 30 .stabs "s_tag:T16=s20s_int:1,0,32;s_float:12,32,32; s_char_vec:17=ar1;0;7;2,64,64;s_next:18=*16,128,32;;",128,0,0,0 @end example In this example, two of the structure elements are previously defined types. For these, the type following the name: part of the element description is a simple type reference. The other two structure elements are new types. In this case there is a type definition embedded after the name:. The type definition for the array element looks just like a type definition for a standalone array. The @code{s_next} field is a pointer to the same kind of structure that the field is an element of. So the definition of structure type 16 contains an type definition for an element which is a pointer to type 16. @node Typedefs @section Giving a Type a Name To give a type a name, use the @samp{t} symbol descriptor. The type specified by the type information (@pxref{String Field}) for the stab. For example, @example .stabs "s_typedef:t16",128,0,0,0 @end example specifies that @code{s_typedef} refers to type number 16. Such stabs have symbol type @code{N_LSYM} or @code{C_DECL}. If instead, you are specifying the tag name for a structure, union, or enumeration, use the @samp{T} symbol descriptor instead. I believe C is the only language with this feature. If the type is an opaque type (I believe this is a Modula-2 feature), AIX provides a type descriptor to specify it. The type descriptor is @samp{o} and is followed by a name. I don't know what the name means---is it always the same as the name of the type, or is this type descriptor used with a nameless stab (@pxref{String Field})? There optionally follows a comma followed by type information which defines the type of this type. If omitted, a semicolon is used in place of the comma and the type information, and the type is much like a generic pointer type---it has a known size but little else about it is specified. @node Unions @section Unions Next let's look at unions. In example2 this union type is declared locally to a procedure and an instance of the union is defined. @example 36 union u_tag @{ 37 int u_int; 38 float u_float; 39 char* u_char; 40 @} an_u; @end example This code generates a stab for the union tag and a stab for the union variable. Both use the @code{N_LSYM} stab type. Since the union variable is scoped locally to the procedure in which it is defined, its stab is located immediately preceding the @code{N_LBRAC} for the procedure's block start. The stab for the union tag, however is located preceding the code for the procedure in which it is defined. The stab type is @code{N_LSYM}. This would seem to imply that the union type is file scope, like the struct type @code{s_tag}. This is not true. The contents and position of the stab for @code{u_type} do not convey any infomation about its procedure local scope. @display <128> N_LSYM - type .stabs "name:sym_desc(union tag)type_def(22)=type_desc(union) byte_size(4) elem_name:type_ref(int),bit_offset(0),bit_size(32); elem_name:type_ref(float),bit_offset(0),bit_size(32); elem_name:type_ref(ptr to char),bit_offset(0),bit_size(32);;" N_LSYM, NIL, NIL, NIL @end display @smallexample 105 .stabs "u_tag:T23=u4u_int:1,0,32;u_float:12,0,32;u_char:21,0,32;;", 128,0,0,0 @end smallexample The symbol descriptor, @samp{T}, following the @samp{name:} means that the stab describes an enumeration, struct or union tag. The type descriptor @samp{u}, following the @samp{23=} of the type definition, narrows it down to a union type definition. Following the @samp{u} is the number of bytes in the union. After that is a list of union element descriptions. Their format is @var{name:type, bit offset into the union, number of bytes for the element;}. The stab for the union variable follows. @display <128> N_LSYM - local variable (with no symbol descriptor) .stabs "name:type_ref(u_tag)", N_LSYM, NIL, NIL, frame_ptr_offset @end display @example 130 .stabs "an_u:23",128,0,0,-20 @end example @node Function Types @section Function types There are various types for function variables. These types are not used in defining functions; see symbol descriptor @samp{f}; they are used for things like pointers to functions. The simple, traditional, type is type descriptor @samp{f} is followed by type information for the return type of the function, followed by a semicolon. This does not deal with functions the number and type of whose parameters are part of their type, as found in Modula-2 or ANSI C. AIX provides extensions to specify these, using the @samp{f}, @samp{F}, @samp{p}, and @samp{R} type descriptors. First comes the type descriptor. Then, if it is @samp{f} or @samp{F}, this is a function, and the type information for the return type of the function follows, followed by a comma. Then comes the number of parameters to the function and a semicolon. Then, for each parameter, there is the name of the parameter followed by a colon (this is only present for type descriptors @samp{R} and @samp{F} which represent Pascal function or procedure parameters), type information for the parameter, a comma, @samp{0} if passed by reference or @samp{1} if passed by value, and a semicolon. The type definition ends with a semicolon. For example, @example int (*g_pf)(); @end example @noindent generates the following code: @example .stabs "g_pf:G24=*25=f1",32,0,0,0 .common _g_pf,4,"bss" @end example The variable defines a new type, 24, which is a pointer to another new type, 25, which is defined as a function returning int. @node Symbol Tables @chapter Symbol information in symbol tables This chapter describes the format of symbol table entries and how stab assembler directives map to them. It also describes the transformations that the assembler and linker make on data from stabs. Each time the assembler encounters a stab in its input file it puts each field of the stab into corresponding fields in a symbol table entry of its output file. If the stab contains a string field, the symbol table entry for that stab points to a string table entry containing the string data from the stab. Assembler labels become relocatable addresses. Symbol table entries in a.out have the format: @c FIXME: should refer to external, not internal. @example struct internal_nlist @{ unsigned long n_strx; /* index into string table of name */ unsigned char n_type; /* type of symbol */ unsigned char n_other; /* misc info (usually empty) */ unsigned short n_desc; /* description field */ bfd_vma n_value; /* value of symbol */ @}; @end example For @code{.stabs} directives, the @code{n_strx} field holds the offset in bytes from the start of the string table to the string table entry containing the @var{string} field. For other classes of stabs (@code{.stabn} and @code{.stabd}) this field is zero. Symbol table entries with @code{n_type} fields containing a value greater or equal to 0x20 originated as stabs generated by the compiler (with one random exception). Those with n_type values less than 0x20 were placed in the symbol table of the executable by the assembler or the linker. The linker concatenates object files and does fixups of externally defined symbols. You can see the transformations made on stab data by the assembler and linker by examining the symbol table after each pass of the build, first the assemble and then the link. To do this, use @samp{nm -ap}. This dumps the symbol table, including debugging information, unsorted. For stab entries the columns are: @var{value}, @var{other}, @var{desc}, @var{type}, @var{string}. For assembler and linker symbols, the columns are: @var{value}, @var{type}, @var{string}. There are a few important things to notice about symbol tables. Where the @var{value} field of a stab contains a frame pointer offset, or a register number, that value is unchanged by the rest of the build. Where the @var{value} field of a stab contains an assembly language label, it is transformed by each build step. The assembler turns it into a relocatable address and the linker turns it into an absolute address. This source line defines a static variable at file scope: @example 3 static int s_g_repeat @end example @noindent The following stab describes the symbol: @example 26 .stabs "s_g_repeat:S1",38,0,0,_s_g_repeat @end example @noindent The assembler transforms the stab into this symbol table entry in the @file{.o} file. The location is expressed as a data segment offset. @example 21 00000084 - 00 0000 STSYM s_g_repeat:S1 @end example @noindent in the symbol table entry from the executable, the linker has made the relocatable address absolute. @example 22 0000e00c - 00 0000 STSYM s_g_repeat:S1 @end example Stabs for global variables do not contain location information. In this case the debugger finds location information in the assembler or linker symbol table entry describing the variable. The source line: @example 1 char g_foo = 'c'; @end example @noindent generates the stab: @example 21 .stabs "g_foo:G2",32,0,0,0 @end example The variable is represented by the following two symbol table entries in the object file. The first one originated as a stab. The second one is an external symbol. The upper case @samp{D} signifies that the @code{n_type} field of the symbol table contains 7, @code{N_DATA} with local linkage. The value field following the file's line number is empty for the stab entry. For the linker symbol it contains the relocatable address corresponding to the variable. @example 19 00000000 - 00 0000 GSYM g_foo:G2 20 00000080 D _g_foo @end example @noindent These entries as transformed by the linker. The linker symbol table entry now holds an absolute address. @example 21 00000000 - 00 0000 GSYM g_foo:G2 @dots{} 215 0000e008 D _g_foo @end example @node Cplusplus @chapter GNU C++ stabs @menu * Basic Cplusplus types:: * Simple classes:: * Class instance:: * Methods:: Method definition * Protections:: * Method Modifiers:: * Virtual Methods:: * Inheritence:: * Virtual Base Classes:: * Static Members:: @end menu Type descriptors added for C++ descriptions: @table @code @item # method type (@code{##} if minimal debug) @item @@ Member (class and variable) type. It is followed by type information for the offset basetype, a comma, and type information for the type of the field being pointed to. (FIXME: this is acknowledged to be gibberish. Can anyone say what really goes here?). Note that there is a conflict between this and type attributes (@pxref{String Field}); both use type descriptor @samp{@@}. Fortunately, the @samp{@@} type descriptor used in this C++ sense always will be followed by a digit, @samp{(}, or @samp{-}, and type attributes never start with those things. @end table @node Basic Cplusplus types @section Basic types for C++ << the examples that follow are based on a01.C >> C++ adds two more builtin types to the set defined for C. These are the unknown type and the vtable record type. The unknown type, type 16, is defined in terms of itself like the void type. The vtable record type, type 17, is defined as a structure type and then as a structure tag. The structure has four fields: delta, index, pfn, and delta2. pfn is the function pointer. << In boilerplate $vtbl_ptr_type, what are the fields delta, index, and delta2 used for? >> This basic type is present in all C++ programs even if there are no virtual methods defined. @display .stabs "struct_name:sym_desc(type)type_def(17)=type_desc(struct)struct_bytes(8) elem_name(delta):type_ref(short int),bit_offset(0),field_bits(16); elem_name(index):type_ref(short int),bit_offset(16),field_bits(16); elem_name(pfn):type_def(18)=type_desc(ptr to)type_ref(void), bit_offset(32),field_bits(32); elem_name(delta2):type_def(short int);bit_offset(32),field_bits(16);;" N_LSYM, NIL, NIL @end display @smallexample .stabs "$vtbl_ptr_type:t17=s8 delta:6,0,16;index:6,16,16;pfn:18=*15,32,32;delta2:6,32,16;;" ,128,0,0,0 @end smallexample @display .stabs "name:sym_dec(struct tag)type_ref($vtbl_ptr_type)",N_LSYM,NIL,NIL,NIL @end display @example .stabs "$vtbl_ptr_type:T17",128,0,0,0 @end example @node Simple classes @section Simple class definition The stabs describing C++ language features are an extension of the stabs describing C. Stabs representing C++ class types elaborate extensively on the stab format used to describe structure types in C. Stabs representing class type variables look just like stabs representing C language variables. Consider the following very simple class definition. @example class baseA @{ public: int Adat; int Ameth(int in, char other); @}; @end example The class @code{baseA} is represented by two stabs. The first stab describes the class as a structure type. The second stab describes a structure tag of the class type. Both stabs are of stab type @code{N_LSYM}. Since the stab is not located between an @code{N_FUN} and a @code{N_LBRAC} stab this indicates that the class is defined at file scope. If it were, then the @code{N_LSYM} would signify a local variable. A stab describing a C++ class type is similar in format to a stab describing a C struct, with each class member shown as a field in the structure. The part of the struct format describing fields is expanded to include extra information relevent to C++ class members. In addition, if the class has multiple base classes or virtual functions the struct format outside of the field parts is also augmented. In this simple example the field part of the C++ class stab representing member data looks just like the field part of a C struct stab. The section on protections describes how its format is sometimes extended for member data. The field part of a C++ class stab representing a member function differs substantially from the field part of a C struct stab. It still begins with @samp{name:} but then goes on to define a new type number for the member function, describe its return type, its argument types, its protection level, any qualifiers applied to the method definition, and whether the method is virtual or not. If the method is virtual then the method description goes on to give the vtable index of the method, and the type number of the first base class defining the method. When the field name is a method name it is followed by two colons rather than one. This is followed by a new type definition for the method. This is a number followed by an equal sign and the type descriptor @samp{#}, indicating a method type, and a second @samp{#}, indicating that this is the @dfn{minimal} type of method definition used by GCC2, not larger method definitions used by earlier versions of GCC. This is followed by a type reference showing the return type of the method and a semi-colon. The format of an overloaded operator method name differs from that of other methods. It is @samp{op$::@var{operator-name}.} where @var{operator-name} is the operator name such as @samp{+} or @samp{+=}. The name ends with a period, and any characters except the period can occur in the @var{operator-name} string. The next part of the method description represents the arguments to the method, preceeded by a colon and ending with a semi-colon. The types of the arguments are expressed in the same way argument types are expressed in C++ name mangling. In this example an @code{int} and a @code{char} map to @samp{ic}. This is followed by a number, a letter, and an asterisk or period, followed by another semicolon. The number indicates the protections that apply to the member function. Here the 2 means public. The letter encodes any qualifier applied to the method definition. In this case, @samp{A} means that it is a normal function definition. The dot shows that the method is not virtual. The sections that follow elaborate further on these fields and describe the additional information present for virtual methods. @display .stabs "class_name:sym_desc(type)type_def(20)=type_desc(struct)struct_bytes(4) field_name(Adat):type(int),bit_offset(0),field_bits(32); method_name(Ameth)::type_def(21)=type_desc(method)return_type(int); :arg_types(int char); protection(public)qualifier(normal)virtual(no);;" N_LSYM,NIL,NIL,NIL @end display @smallexample .stabs "baseA:t20=s4Adat:1,0,32;Ameth::21=##1;:ic;2A.;;",128,0,0,0 .stabs "class_name:sym_desc(struct tag)",N_LSYM,NIL,NIL,NIL .stabs "baseA:T20",128,0,0,0 @end smallexample @node Class instance @section Class instance As shown above, describing even a simple C++ class definition is accomplished by massively extending the stab format used in C to describe structure types. However, once the class is defined, C stabs with no modifications can be used to describe class instances. The following source: @example main () @{ baseA AbaseA; @} @end example @noindent yields the following stab describing the class instance. It looks no different from a standard C stab describing a local variable. @display .stabs "name:type_ref(baseA)", N_LSYM, NIL, NIL, frame_ptr_offset @end display @example .stabs "AbaseA:20",128,0,0,-20 @end example @node Methods @section Method defintion The class definition shown above declares Ameth. The C++ source below defines Ameth: @example int baseA::Ameth(int in, char other) @{ return in; @}; @end example This method definition yields three stabs following the code of the method. One stab describes the method itself and following two describe its parameters. Although there is only one formal argument all methods have an implicit argument which is the @code{this} pointer. The @code{this} pointer is a pointer to the object on which the method was called. Note that the method name is mangled to encode the class name and argument types. Name mangling is described in the @sc{arm} (@cite{The Annotated C++ Reference Manual}, by Ellis and Stroustrup, @sc{isbn} 0-201-51459-1); @file{gpcompare.texi} in Cygnus GCC distributions describes the differences between GNU mangling and @sc{arm} mangling. @c FIXME: Use @xref, especially if this is generally installed in the @c info tree. @c FIXME: This information should be in a net release, either of GCC or @c GDB. But gpcompare.texi doesn't seem to be in the FSF GCC. @example .stabs "name:symbol_desriptor(global function)return_type(int)", N_FUN, NIL, NIL, code_addr_of_method_start .stabs "Ameth__5baseAic:F1",36,0,0,_Ameth__5baseAic @end example Here is the stab for the @code{this} pointer implicit argument. The name of the @code{this} pointer is always @code{this}. Type 19, the @code{this} pointer is defined as a pointer to type 20, @code{baseA}, but a stab defining @code{baseA} has not yet been emited. Since the compiler knows it will be emited shortly, here it just outputs a cross reference to the undefined symbol, by prefixing the symbol name with @samp{xs}. @example .stabs "name:sym_desc(register param)type_def(19)= type_desc(ptr to)type_ref(baseA)= type_desc(cross-reference to)baseA:",N_RSYM,NIL,NIL,register_number .stabs "this:P19=*20=xsbaseA:",64,0,0,8 @end example The stab for the explicit integer argument looks just like a parameter to a C function. The last field of the stab is the offset from the argument pointer, which in most systems is the same as the frame pointer. @example .stabs "name:sym_desc(value parameter)type_ref(int)", N_PSYM,NIL,NIL,offset_from_arg_ptr .stabs "in:p1",160,0,0,72 @end example << The examples that follow are based on A1.C >> @node Protections @section Protections In the simple class definition shown above all member data and functions were publicly accessable. The example that follows contrasts public, protected and privately accessable fields and shows how these protections are encoded in C++ stabs. @c FIXME: What does "part of the string" mean? Protections for class member data are signified by two characters embedded in the stab defining the class type. These characters are located after the name: part of the string. @samp{/0} means private, @samp{/1} means protected, and @samp{/2} means public. If these characters are omited this means that the member is public. The following C++ source: @example class all_data @{ private: int priv_dat; protected: char prot_dat; public: float pub_dat; @}; @end example @noindent generates the following stab to describe the class type all_data. @display .stabs "class_name:sym_desc(type)type_def(19)=type_desc(struct)struct_bytes data_name:/protection(private)type_ref(int),bit_offset,num_bits; data_name:/protection(protected)type_ref(char),bit_offset,num_bits; data_name:(/num omited, private)type_ref(float),bit_offset,num_bits;;" N_LSYM,NIL,NIL,NIL @end display @smallexample .stabs "all_data:t19=s12 priv_dat:/01,0,32;prot_dat:/12,32,8;pub_dat:12,64,32;;",128,0,0,0 @end smallexample Protections for member functions are signified by one digit embeded in the field part of the stab describing the method. The digit is 0 if private, 1 if protected and 2 if public. Consider the C++ class definition below: @example class all_methods @{ private: int priv_meth(int in)@{return in;@}; protected: char protMeth(char in)@{return in;@}; public: float pubMeth(float in)@{return in;@}; @}; @end example It generates the following stab. The digit in question is to the left of an @samp{A} in each case. Notice also that in this case two symbol descriptors apply to the class name struct tag and struct type. @display .stabs "class_name:sym_desc(struct tag&type)type_def(21)= sym_desc(struct)struct_bytes(1) meth_name::type_def(22)=sym_desc(method)returning(int); :args(int);protection(private)modifier(normal)virtual(no); meth_name::type_def(23)=sym_desc(method)returning(char); :args(char);protection(protected)modifier(normal)virual(no); meth_name::type_def(24)=sym_desc(method)returning(float); :args(float);protection(public)modifier(normal)virtual(no);;", N_LSYM,NIL,NIL,NIL @end display @smallexample .stabs "all_methods:Tt21=s1priv_meth::22=##1;:i;0A.;protMeth::23=##2;:c;1A.; pubMeth::24=##12;:f;2A.;;",128,0,0,0 @end smallexample @node Method Modifiers @section Method Modifiers (const, volatile, const volatile) << based on a6.C >> In the class example described above all the methods have the normal modifier. This method modifier information is located just after the protection information for the method. This field has four possible character values. Normal methods use @samp{A}, const methods use @samp{B}, volatile methods use @samp{C}, and const volatile methods use @samp{D}. Consider the class definition below: @example class A @{ public: int ConstMeth (int arg) const @{ return arg; @}; char VolatileMeth (char arg) volatile @{ return arg; @}; float ConstVolMeth (float arg) const volatile @{return arg; @}; @}; @end example This class is described by the following stab: @display .stabs "class(A):sym_desc(struct)type_def(20)=type_desc(struct)struct_bytes(1) meth_name(ConstMeth)::type_def(21)sym_desc(method) returning(int);:arg(int);protection(public)modifier(const)virtual(no); meth_name(VolatileMeth)::type_def(22)=sym_desc(method) returning(char);:arg(char);protection(public)modifier(volatile)virt(no) meth_name(ConstVolMeth)::type_def(23)=sym_desc(method) returning(float);:arg(float);protection(public)modifer(const volatile) virtual(no);;", @dots{} @end display @example .stabs "A:T20=s1ConstMeth::21=##1;:i;2B.;VolatileMeth::22=##2;:c;2C.; ConstVolMeth::23=##12;:f;2D.;;",128,0,0,0 @end example @node Virtual Methods @section Virtual Methods << The following examples are based on a4.C >> The presence of virtual methods in a class definition adds additional data to the class description. The extra data is appended to the description of the virtual method and to the end of the class description. Consider the class definition below: @example class A @{ public: int Adat; virtual int A_virt (int arg) @{ return arg; @}; @}; @end example This results in the stab below describing class A. It defines a new type (20) which is an 8 byte structure. The first field of the class struct is @samp{Adat}, an integer, starting at structure offset 0 and occupying 32 bits. The second field in the class struct is not explicitly defined by the C++ class definition but is implied by the fact that the class contains a virtual method. This field is the vtable pointer. The name of the vtable pointer field starts with @samp{$vf} and continues with a type reference to the class it is part of. In this example the type reference for class A is 20 so the name of its vtable pointer field is @samp{$vf20}, followed by the usual colon. Next there is a type definition for the vtable pointer type (21). This is in turn defined as a pointer to another new type (22). Type 22 is the vtable itself, which is defined as an array, indexed by a range of integers between 0 and 1, and whose elements are of type 17. Type 17 was the vtable record type defined by the boilerplate C++ type definitions, as shown earlier. The bit offset of the vtable pointer field is 32. The number of bits in the field are not specified when the field is a vtable pointer. Next is the method definition for the virtual member function @code{A_virt}. Its description starts out using the same format as the non-virtual member functions described above, except instead of a dot after the @samp{A} there is an asterisk, indicating that the function is virtual. Since is is virtual some addition information is appended to the end of the method description. The first number represents the vtable index of the method. This is a 32 bit unsigned number with the high bit set, followed by a semi-colon. The second number is a type reference to the first base class in the inheritence hierarchy defining the virtual member function. In this case the class stab describes a base class so the virtual function is not overriding any other definition of the method. Therefore the reference is to the type number of the class that the stab is describing (20). This is followed by three semi-colons. One marks the end of the current sub-section, one marks the end of the method field, and the third marks the end of the struct definition. For classes containing virtual functions the very last section of the string part of the stab holds a type reference to the first base class. This is preceeded by @samp{~%} and followed by a final semi-colon. @display .stabs "class_name(A):type_def(20)=sym_desc(struct)struct_bytes(8) field_name(Adat):type_ref(int),bit_offset(0),field_bits(32); field_name(A virt func ptr):type_def(21)=type_desc(ptr to)type_def(22)= sym_desc(array)index_type_ref(range of int from 0 to 1); elem_type_ref(vtbl elem type), bit_offset(32); meth_name(A_virt)::typedef(23)=sym_desc(method)returning(int); :arg_type(int),protection(public)normal(yes)virtual(yes) vtable_index(1);class_first_defining(A);;;~%first_base(A);", N_LSYM,NIL,NIL,NIL @end display @c FIXME: bogus line break. @example .stabs "A:t20=s8Adat:1,0,32;$vf20:21=*22=ar1;0;1;17,32; A_virt::23=##1;:i;2A*-2147483647;20;;;~%20;",128,0,0,0 @end example @node Inheritence @section Inheritence Stabs describing C++ derived classes include additional sections that describe the inheritence hierarchy of the class. A derived class stab also encodes the number of base classes. For each base class it tells if the base class is virtual or not, and if the inheritence is private or public. It also gives the offset into the object of the portion of the object corresponding to each base class. This additional information is embeded in the class stab following the number of bytes in the struct. First the number of base classes appears bracketed by an exclamation point and a comma. Then for each base type there repeats a series: two digits, a number, a comma, another number, and a semi-colon. The first of the two digits is 1 if the base class is virtual and 0 if not. The second digit is 2 if the derivation is public and 0 if not. The number following the first two digits is the offset from the start of the object to the part of the object pertaining to the base class. After the comma, the second number is a type_descriptor for the base type. Finally a semi-colon ends the series, which repeats for each base class. The source below defines three base classes @code{A}, @code{B}, and @code{C} and the derived class @code{D}. @example class A @{ public: int Adat; virtual int A_virt (int arg) @{ return arg; @}; @}; class B @{ public: int B_dat; virtual int B_virt (int arg) @{return arg; @}; @}; class C @{ public: int Cdat; virtual int C_virt (int arg) @{return arg; @}; @}; class D : A, virtual B, public C @{ public: int Ddat; virtual int A_virt (int arg ) @{ return arg+1; @}; virtual int B_virt (int arg) @{ return arg+2; @}; virtual int C_virt (int arg) @{ return arg+3; @}; virtual int D_virt (int arg) @{ return arg; @}; @}; @end example Class stabs similar to the ones described earlier are generated for each base class. @c FIXME!!! the linebreaks in the following example probably make the @c examples literally unusable, but I don't know any other way to get @c them on the page. @c One solution would be to put some of the type definitions into @c separate stabs, even if that's not exactly what the compiler actually @c emits. @smallexample .stabs "A:T20=s8Adat:1,0,32;$vf20:21=*22=ar1;0;1;17,32; A_virt::23=##1;:i;2A*-2147483647;20;;;~%20;",128,0,0,0 .stabs "B:Tt25=s8Bdat:1,0,32;$vf25:21,32;B_virt::26=##1; :i;2A*-2147483647;25;;;~%25;",128,0,0,0 .stabs "C:Tt28=s8Cdat:1,0,32;$vf28:21,32;C_virt::29=##1; :i;2A*-2147483647;28;;;~%28;",128,0,0,0 @end smallexample In the stab describing derived class @code{D} below, the information about the derivation of this class is encoded as follows. @display .stabs "derived_class_name:symbol_descriptors(struct tag&type)= type_descriptor(struct)struct_bytes(32)!num_bases(3), base_virtual(no)inheritence_public(no)base_offset(0), base_class_type_ref(A); base_virtual(yes)inheritence_public(no)base_offset(NIL), base_class_type_ref(B); base_virtual(no)inheritence_public(yes)base_offset(64), base_class_type_ref(C); @dots{} @end display @c FIXME! fake linebreaks. @smallexample .stabs "D:Tt31=s32!3,000,20;100,25;0264,28;$vb25:24,128;Ddat: 1,160,32;A_virt::32=##1;:i;2A*-2147483647;20;;B_virt: :32:i;2A*-2147483647;25;;C_virt::32:i;2A*-2147483647; 28;;D_virt::32:i;2A*-2147483646;31;;;~%20;",128,0,0,0 @end smallexample @node Virtual Base Classes @section Virtual Base Classes A derived class object consists of a concatination in memory of the data areas defined by each base class, starting with the leftmost and ending with the rightmost in the list of base classes. The exception to this rule is for virtual inheritence. In the example above, class @code{D} inherits virtually from base class @code{B}. This means that an instance of a @code{D} object will not contain its own @code{B} part but merely a pointer to a @code{B} part, known as a virtual base pointer. In a derived class stab, the base offset part of the derivation information, described above, shows how the base class parts are ordered. The base offset for a virtual base class is always given as 0. Notice that the base offset for @code{B} is given as 0 even though @code{B} is not the first base class. The first base class @code{A} starts at offset 0. The field information part of the stab for class @code{D} describes the field which is the pointer to the virtual base class @code{B}. The vbase pointer name is @samp{$vb} followed by a type reference to the virtual base class. Since the type id for @code{B} in this example is 25, the vbase pointer name is @samp{$vb25}. @c FIXME!! fake linebreaks below @smallexample .stabs "D:Tt31=s32!3,000,20;100,25;0264,28;$vb25:24,128;Ddat:1, 160,32;A_virt::32=##1;:i;2A*-2147483647;20;;B_virt::32:i; 2A*-2147483647;25;;C_virt::32:i;2A*-2147483647;28;;D_virt: :32:i;2A*-2147483646;31;;;~%20;",128,0,0,0 @end smallexample Following the name and a semicolon is a type reference describing the type of the virtual base class pointer, in this case 24. Type 24 was defined earlier as the type of the @code{B} class @code{this} pointer. The @code{this} pointer for a class is a pointer to the class type. @example .stabs "this:P24=*25=xsB:",64,0,0,8 @end example Finally the field offset part of the vbase pointer field description shows that the vbase pointer is the first field in the @code{D} object, before any data fields defined by the class. The layout of a @code{D} class object is a follows, @code{Adat} at 0, the vtable pointer for @code{A} at 32, @code{Cdat} at 64, the vtable pointer for C at 96, the virtual base pointer for @code{B} at 128, and @code{Ddat} at 160. @node Static Members @section Static Members The data area for a class is a concatenation of the space used by the data members of the class. If the class has virtual methods, a vtable pointer follows the class data. The field offset part of each field description in the class stab shows this ordering. << How is this reflected in stabs? See Cygnus bug #677 for some info. >> @node Example2.c @appendix Source code for extended example @example 1 char g_foo = 'c'; 2 register int g_bar asm ("%g5"); 3 static int s_g_repeat = 2; 4 int (*g_pf)(); 5 6 struct s_tag @{ 7 int s_int; 8 float s_float; 9 char s_char_vec[8]; 10 struct s_tag* s_next; 11 @} g_an_s; 12 13 typedef struct s_tag s_typedef; 14 15 char char_vec[3] = @{'a','b','c'@}; 16 17 main (argc, argv) 18 int argc; 19 char* argv[]; 20 @{ 21 static float s_flap; 22 int times; 23 for (times=0; times < s_g_repeat; times++)@{ 24 int inner; 25 printf ("Hello world\n"); 26 @} 27 @}; 28 29 enum e_places @{first,second=3,last@}; 30 31 static s_proc (s_arg, s_ptr_arg, char_vec) 32 s_typedef s_arg; 33 s_typedef* s_ptr_arg; 34 char* char_vec; 35 @{ 36 union u_tag @{ 37 int u_int; 38 float u_float; 39 char* u_char; 40 @} an_u; 41 @} 42 43 @end example @node Example2.s @appendix Assembly code for extended example @example 1 gcc2_compiled.: 2 .stabs "/cygint/s1/users/jcm/play/",100,0,0,Ltext0 3 .stabs "example2.c",100,0,0,Ltext0 4 .text 5 Ltext0: 6 .stabs "int:t1=r1;-2147483648;2147483647;",128,0,0,0 7 .stabs "char:t2=r2;0;127;",128,0,0,0 8 .stabs "long int:t3=r1;-2147483648;2147483647;",128,0,0,0 9 .stabs "unsigned int:t4=r1;0;-1;",128,0,0,0 10 .stabs "long unsigned int:t5=r1;0;-1;",128,0,0,0 11 .stabs "short int:t6=r1;-32768;32767;",128,0,0,0 12 .stabs "long long int:t7=r1;0;-1;",128,0,0,0 13 .stabs "short unsigned int:t8=r1;0;65535;",128,0,0,0 14 .stabs "long long unsigned int:t9=r1;0;-1;",128,0,0,0 15 .stabs "signed char:t10=r1;-128;127;",128,0,0,0 16 .stabs "unsigned char:t11=r1;0;255;",128,0,0,0 17 .stabs "float:t12=r1;4;0;",128,0,0,0 18 .stabs "double:t13=r1;8;0;",128,0,0,0 19 .stabs "long double:t14=r1;8;0;",128,0,0,0 20 .stabs "void:t15=15",128,0,0,0 21 .stabs "g_foo:G2",32,0,0,0 22 .global _g_foo 23 .data 24 _g_foo: 25 .byte 99 26 .stabs "s_g_repeat:S1",38,0,0,_s_g_repeat 27 .align 4 28 _s_g_repeat: 29 .word 2 @c FIXME! fake linebreak in line 30 30 .stabs "s_tag:T16=s20s_int:1,0,32;s_float:12,32,32;s_char_vec: 17=ar1;0;7;2,64,64;s_next:18=*16,128,32;;",128,0,0,0 31 .stabs "s_typedef:t16",128,0,0,0 32 .stabs "char_vec:G19=ar1;0;2;2",32,0,0,0 33 .global _char_vec 34 .align 4 35 _char_vec: 36 .byte 97 37 .byte 98 38 .byte 99 39 .reserve _s_flap.0,4,"bss",4 40 .text 41 .align 4 42 LC0: 43 .ascii "Hello world\12\0" 44 .align 4 45 .global _main 46 .proc 1 47 _main: 48 .stabn 68,0,20,LM1 49 LM1: 50 !#PROLOGUE# 0 51 save %sp,-144,%sp 52 !#PROLOGUE# 1 53 st %i0,[%fp+68] 54 st %i1,[%fp+72] 55 call ___main,0 56 nop 57 LBB2: 58 .stabn 68,0,23,LM2 59 LM2: 60 st %g0,[%fp-20] 61 L2: 62 sethi %hi(_s_g_repeat),%o0 63 ld [%fp-20],%o1 64 ld [%o0+%lo(_s_g_repeat)],%o0 65 cmp %o1,%o0 66 bge L3 67 nop 68 LBB3: 69 .stabn 68,0,25,LM3 70 LM3: 71 sethi %hi(LC0),%o1 72 or %o1,%lo(LC0),%o0 73 call _printf,0 74 nop 75 .stabn 68,0,26,LM4 76 LM4: 77 LBE3: 78 .stabn 68,0,23,LM5 79 LM5: 80 L4: 81 ld [%fp-20],%o0 82 add %o0,1,%o1 83 st %o1,[%fp-20] 84 b,a L2 85 L3: 86 .stabn 68,0,27,LM6 87 LM6: 88 LBE2: 89 .stabn 68,0,27,LM7 90 LM7: 91 L1: 92 ret 93 restore 94 .stabs "main:F1",36,0,0,_main 95 .stabs "argc:p1",160,0,0,68 96 .stabs "argv:p20=*21=*2",160,0,0,72 97 .stabs "s_flap:V12",40,0,0,_s_flap.0 98 .stabs "times:1",128,0,0,-20 99 .stabn 192,0,0,LBB2 100 .stabs "inner:1",128,0,0,-24 101 .stabn 192,0,0,LBB3 102 .stabn 224,0,0,LBE3 103 .stabn 224,0,0,LBE2 104 .stabs "e_places:T22=efirst:0,second:3,last:4,;",128,0,0,0 @c FIXME: fake linebreak in line 105 105 .stabs "u_tag:T23=u4u_int:1,0,32;u_float:12,0,32;u_char:21,0,32;;", 128,0,0,0 106 .align 4 107 .proc 1 108 _s_proc: 109 .stabn 68,0,35,LM8 110 LM8: 111 !#PROLOGUE# 0 112 save %sp,-120,%sp 113 !#PROLOGUE# 1 114 mov %i0,%o0 115 st %i1,[%fp+72] 116 st %i2,[%fp+76] 117 LBB4: 118 .stabn 68,0,41,LM9 119 LM9: 120 LBE4: 121 .stabn 68,0,41,LM10 122 LM10: 123 L5: 124 ret 125 restore 126 .stabs "s_proc:f1",36,0,0,_s_proc 127 .stabs "s_arg:p16",160,0,0,0 128 .stabs "s_ptr_arg:p18",160,0,0,72 129 .stabs "char_vec:p21",160,0,0,76 130 .stabs "an_u:23",128,0,0,-20 131 .stabn 192,0,0,LBB4 132 .stabn 224,0,0,LBE4 133 .stabs "g_bar:r1",64,0,0,5 134 .stabs "g_pf:G24=*25=f1",32,0,0,0 135 .common _g_pf,4,"bss" 136 .stabs "g_an_s:G16",32,0,0,0 137 .common _g_an_s,20,"bss" @end example @node Stab Types @appendix Table of stab types The following are all the possible values for the stab type field, for @code{a.out} files, in numeric order. This does not apply to XCOFF. The symbolic names are defined in the file @file{include/aout/stabs.def}. @menu * Non-stab symbol types:: Types from 0 to 0x1f * Stab symbol types:: Types from 0x20 to 0xff @end menu @node Non-stab symbol types @appendixsec Non-stab symbol types The following types are used by the linker and assembler, not by stab directives. Since this document does not attempt to describe aspects of object file format other than the debugging format, no details are given. @c Try to get most of these to fit on a single line. @iftex @tableindent=1.5in @end iftex @table @code @item 0x0 N_UNDF Undefined symbol @item 0x2 N_ABS File scope absolute symbol @item 0x3 N_ABS | N_EXT External absolute symbol @item 0x4 N_TEXT File scope text symbol @item 0x5 N_TEXT | N_EXT External text symbol @item 0x6 N_DATA File scope data symbol @item 0x7 N_DATA | N_EXT External data symbol @item 0x8 N_BSS File scope BSS symbol @item 0x9 N_BSS | N_EXT External BSS symbol @item 0x0c N_FN_SEQ Same as @code{N_FN}, for Sequent compilers @item 0x0a N_INDR Symbol is indirected to another symbol @item 0x12 N_COMM Common---visible after shared library dynamic link @item 0x14 N_SETA Absolute set element @item 0x16 N_SETT Text segment set element @item 0x18 N_SETD Data segment set element @item 0x1a N_SETB BSS segment set element @item 0x1c N_SETV Pointer to set vector @item 0x1e N_WARNING Print a warning message during linking @item 0x1f N_FN File name of a @file{.o} file @end table @node Stab symbol types @appendixsec Stab symbol types The following symbol types indicate that this is a stab. This is the full list of stab numbers, including stab types that are used in languages other than C. @table @code @item 0x20 N_GSYM Global symbol; see @ref{N_GSYM}. @item 0x22 N_FNAME Function name (for BSD Fortran); see @ref{N_FNAME}. @item 0x24 N_FUN Function name (@pxref{Procedures}) or text segment variable (@pxref{Statics}). @item 0x26 N_STSYM Data segment file-scope variable; see @ref{Statics}. @item 0x28 N_LCSYM BSS segment file-scope variable; see @ref{Statics}. @item 0x2a N_MAIN Name of main routine; see @ref{Main Program}. @c FIXME: discuss this in the Statics node where we talk about @c the fact that the n_type indicates the section. @item 0x2c N_ROSYM Variable in @code{.rodata} section; see @ref{Statics}. @item 0x30 N_PC Global symbol (for Pascal); see @ref{N_PC}. @item 0x32 N_NSYMS Number of symbols (according to Ultrix V4.0); see @ref{N_NSYMS}. @item 0x34 N_NOMAP No DST map; see @ref{N_NOMAP}. @c FIXME: describe this solaris feature in the body of the text (see @c comments in include/aout/stab.def). @item 0x38 N_OBJ Object file (Solaris2). @c See include/aout/stab.def for (a little) more info. @item 0x3c N_OPT Debugger options (Solaris2). @item 0x40 N_RSYM Register variable; see @ref{N_RSYM}. @item 0x42 N_M2C Modula-2 compilation unit; see @ref{N_M2C}. @item 0x44 N_SLINE Line number in text segment; see @ref{Line Numbers}. @item 0x46 N_DSLINE Line number in data segment; see @ref{Line Numbers}. @item 0x48 N_BSLINE Line number in bss segment; see @ref{Line Numbers}. @item 0x48 N_BROWS Sun source code browser, path to @file{.cb} file; see @ref{N_BROWS}. @item 0x4a N_DEFD GNU Modula2 definition module dependency; see @ref{N_DEFD}. @item 0x4c N_FLINE Function start/body/end line numbers (Solaris2). @item 0x50 N_EHDECL GNU C++ exception variable; see @ref{N_EHDECL}. @item 0x50 N_MOD2 Modula2 info "for imc" (according to Ultrix V4.0); see @ref{N_MOD2}. @item 0x54 N_CATCH GNU C++ @code{catch} clause; see @ref{N_CATCH}. @item 0x60 N_SSYM Structure of union element; see @ref{N_SSYM}. @item 0x62 N_ENDM Last stab for module (Solaris2). @item 0x64 N_SO Path and name of source file; see @ref{Source Files}. @item 0x80 N_LSYM Stack variable (@pxref{Stack Variables}) or type (@pxref{Typedefs}). @item 0x82 N_BINCL Beginning of an include file (Sun only); see @ref{Include Files}. @item 0x84 N_SOL Name of include file; see @ref{Include Files}. @item 0xa0 N_PSYM Parameter variable; see @ref{Parameters}. @item 0xa2 N_EINCL End of an include file; see @ref{Include Files}. @item 0xa4 N_ENTRY Alternate entry point; see @ref{N_ENTRY}. @item 0xc0 N_LBRAC Beginning of a lexical block; see @ref{Block Structure}. @item 0xc2 N_EXCL Place holder for a deleted include file; see @ref{Include Files}. @item 0xc4 N_SCOPE Modula2 scope information (Sun linker); see @ref{N_SCOPE}. @item 0xe0 N_RBRAC End of a lexical block; see @ref{Block Structure}. @item 0xe2 N_BCOMM Begin named common block; see @ref{Common Blocks}. @item 0xe4 N_ECOMM End named common block; see @ref{Common Blocks}. @item 0xe8 N_ECOML Member of a common block; see @ref{Common Blocks}. @c FIXME: How does this really work? Move it to main body of document. @item 0xea N_WITH Pascal @code{with} statement: type,,0,0,offset (Solaris2). @item 0xf0 N_NBTEXT Gould non-base registers; see @ref{Gould}. @item 0xf2 N_NBDATA Gould non-base registers; see @ref{Gould}. @item 0xf4 N_NBBSS Gould non-base registers; see @ref{Gould}. @item 0xf6 N_NBSTS Gould non-base registers; see @ref{Gould}. @item 0xf8 N_NBLCS Gould non-base registers; see @ref{Gould}. @end table @c Restore the default table indent @iftex @tableindent=.8in @end iftex @node Symbol Descriptors @appendix Table of Symbol Descriptors These tell in the @code{.stabs} @var{string} field what kind of symbol the stab represents. They follow the colon which follows the symbol name. @xref{String Field}, for more information about their use. @c Please keep this alphabetical @table @code @c In TeX, this looks great, digit is in italics. But makeinfo insists @c on putting it in `', not realizing that @var should override @code. @c I don't know of any way to make makeinfo do the right thing. Seems @c like a makeinfo bug to me. @item @var{digit} @itemx ( @itemx - Variable on the stack; see @ref{Stack Variables}. @item a Parameter passed by reference in register; see @ref{Parameters}. @item c Constant; see @ref{Constants}. @item C Conformant array bound (Pascal, maybe other languages), @xref{Parameters}. Name of a caught exception (GNU C++). These can be distinguished because the latter uses N_CATCH and the former uses another symbol type. @item d Floating point register variable; see @ref{Register variables}. @item D Parameter in floating point register; see @ref{Parameters}. @item f File scope function; see @ref{Procedures}. @item F Global function; see @ref{Procedures}. @item G Global variable; see @ref{Global Variables}. @item i @xref{Parameters}. @item I Internal (nested) procedure; see @ref{Procedures}. @item J Internal (nested) function; see @ref{Procedures}. @item L Label name (documented by AIX, no further information known). @item m Module; see @ref{Procedures}. @item p Argument list parameter; see @ref{Parameters}. @item pP @xref{Parameters}. @item pF Fortran Function parameter; see @ref{Parameters}. @item P Unfortunately, three separate meanings have been independently invented for this symbol descriptor. At least the GNU and Sun uses can be distinguished by the symbol type. Global Procedure (AIX) (symbol type used unknown); see @ref{Procedures}. Register parameter (GNU) (symbol type N_PSYM); see @ref{Parameters}. Prototype of function referenced by this file (Sun acc) (symbol type N_FUN). @item Q Static Procedure; see @ref{Procedures}. @item R Register parameter @xref{Parameters}. @item r Register variable; see @ref{Register variables}. @item S File scope variable; see @ref{Statics}. @item t Type name; see @ref{Typedefs}. @item T enumeration, struct or union tag; see @ref{Typedefs}. @item v Parameter passed by reference; see @ref{Parameters}. @item V Procedure scope static variable; see @ref{Statics}. @item x Conformant array; see @ref{Parameters}. @item X Function return variable; see @ref{Parameters}. @end table @node Type Descriptors @appendix Table of Type Descriptors These tell in the @code{.stabs} @var{string} field what kind of type is being defined. They follow the type number and an equals sign. @xref{Overview}, for more information about their use. @table @code @item @var{digit} @itemx ( Type reference; see @ref{String Field}. @item - Reference to builtin type; see @ref{Negative Type Numbers}. @item # Method (C++); see @ref{Cplusplus}. @item * Pointer; see @ref{Miscellaneous Types}. @item & Reference (C++). @item @@ Type Attributes (AIX); see @ref{String Field}. Member (class and variable) type (GNU C++); see @ref{Cplusplus}. @item a Array; see @ref{Arrays}. @item A Open array; see @ref{Arrays}. @item b Pascal space type (AIX); see @ref{Miscellaneous Types}. Builtin integer type (Sun); see @ref{Builtin Type Descriptors}. @item B Volatile-qualified type; see @ref{Miscellaneous Types}. @item c Complex builtin type; see @ref{Builtin Type Descriptors}. @item C COBOL Picture type. See AIX documentation for details. @item d File type; see @ref{Miscellaneous Types}. @item D N-dimensional dynamic array; see @ref{Arrays}. @item e Enumeration type; see @ref{Enumerations}. @item E N-dimensional subarray; see @ref{Arrays}. @item f Function type; see @ref{Function Types}. @item F Pascal function parameter; see @ref{Function Types} @item g Builtin floating point type; see @ref{Builtin Type Descriptors}. @item G COBOL Group. See AIX documentation for details. @item i Imported type; see @ref{Cross-references}. @item k Const-qualified type; see @ref{Miscellaneous Types}. @item K COBOL File Descriptor. See AIX documentation for details. @item M Multiple instance type; see @ref{Miscellaneous Types}. @item n String type; see @ref{Strings}. @item N Stringptr; see @ref{Strings}. @item o Opaque type; see @ref{Typedefs}. @item p Procedure; see @ref{Function Types}. @item P Packed array; see @ref{Arrays}. @item r Range type; see @ref{Subranges}. @item R Builtin floating type; see @ref{Builtin Type Descriptors} (Sun). Pascal subroutine parameter; see @ref{Function Types} (AIX). Detecting this conflict is possible with careful parsing (hint: a Pascal subroutine parameter type will always contain a comma, and a builtin type descriptor never will). @item s Structure type; see @ref{Structures}. @item S Set type; see @ref{Miscellaneous Types}. @item u Union; see @ref{Unions}. @item v Variant record. This is a Pascal and Modula-2 feature which is like a union within a struct in C. See AIX documentation for details. @item w Wide character; see @ref{Builtin Type Descriptors}. @item x Cross-reference; see @ref{Cross-references}. @item z gstring; see @ref{Strings}. @end table @node Expanded reference @appendix Expanded reference by stab type @c FIXME: This appendix should go away, see N_PSYM or N_SO for an example. For a full list of stab types, and cross-references to where they are described, see @ref{Stab Types}. This appendix just duplicates certain information from the main body of this document; eventually the information will all be in one place. Format of an entry: The first line is the symbol type expressed in decimal, hexadecimal, and as a #define (see devo/include/aout/stab.def). The second line describes the language constructs the symbol type represents. The third line is the stab format with the significant stab fields named and the rest NIL. Subsequent lines expand upon the meaning and possible values for each significant stab field. # stands in for the type descriptor. Finally, any further information. @menu * N_GSYM:: Global variable * N_FNAME:: Function name (BSD Fortran) * N_PC:: Pascal global symbol * N_NSYMS:: Number of symbols * N_NOMAP:: No DST map * N_RSYM:: Register variable * N_M2C:: Modula-2 compilation unit * N_BROWS:: Path to .cb file for Sun source code browser * N_DEFD:: GNU Modula2 definition module dependency * N_EHDECL:: GNU C++ exception variable * N_MOD2:: Modula2 information "for imc" * N_CATCH:: GNU C++ "catch" clause * N_SSYM:: Structure or union element * N_ENTRY:: Alternate entry point * N_SCOPE:: Modula2 scope information (Sun only) * Gould:: non-base register symbols used on Gould systems * N_LENG:: Length of preceding entry @end menu @node N_GSYM @section 32 - 0x20 - N_GYSM @display Global variable. .stabs "name", N_GSYM, NIL, NIL, NIL @end display @example "name" -> "symbol_name:#type" # -> G @end example Only the @var{name} field is significant. The location of the variable is obtained from the corresponding external symbol. @node N_FNAME @section 34 - 0x22 - N_FNAME Function name (for BSD Fortran) @display .stabs "name", N_FNAME, NIL, NIL, NIL @end display @example "name" -> "function_name" @end example Only the "name" field is significant. The location of the symbol is obtained from the corresponding extern symbol. @node N_PC @section 48 - 0x30 - N_PC Global symbol (for Pascal) @display .stabs "name", N_PC, NIL, NIL, value @end display @example "name" -> "symbol_name" <> value -> supposedly the line number (stab.def is skeptical) @end example @display @file{stabdump.c} says: global pascal symbol: name,,0,subtype,line << subtype? >> @end display @node N_NSYMS @section 50 - 0x32 - N_NSYMS Number of symbols (according to Ultrix V4.0) @display 0, files,,funcs,lines (stab.def) @end display @node N_NOMAP @section 52 - 0x34 - N_NOMAP No DST map for symbol (according to Ultrix V4.0). I think this means a variable has been optimized out. @display name, ,0,type,ignored (stab.def) @end display @node N_RSYM @section 64 - 0x40 - N_RSYM register variable @display .stabs "name:type",N_RSYM,0,RegSize,RegNumber (Sun doc) @end display @node N_M2C @section 66 - 0x42 - N_M2C Modula-2 compilation unit @display .stabs "name", N_M2C, 0, desc, value @end display @example "name" -> "unit_name,unit_time_stamp[,code_time_stamp] desc -> unit_number value -> 0 (main unit) 1 (any other unit) @end example @node N_BROWS @section 72 - 0x48 - N_BROWS Sun source code browser, path to @file{.cb} file <> "path to associated .cb file" Note: type field value overlaps with N_BSLINE @node N_DEFD @section 74 - 0x4a - N_DEFD GNU Modula2 definition module dependency GNU Modula-2 definition module dependency. Value is the modification time of the definition file. Other is non-zero if it is imported with the GNU M2 keyword @code{%INITIALIZE}. Perhaps @code{N_M2C} can be used if there are enough empty fields? @node N_EHDECL @section 80 - 0x50 - N_EHDECL GNU C++ exception variable <> "name is variable name" Note: conflicts with N_MOD2. @node N_MOD2 @section 80 - 0x50 - N_MOD2 Modula2 info "for imc" (according to Ultrix V4.0) Note: conflicts with N_EHDECL <> @node N_CATCH @section 84 - 0x54 - N_CATCH GNU C++ @code{catch} clause GNU C++ @code{catch} clause. Value is its address. Desc is nonzero if this entry is immediately followed by a CAUGHT stab saying what exception was caught. Multiple CAUGHT stabs means that multiple exceptions can be caught here. If Desc is 0, it means all exceptions are caught here. @node N_SSYM @section 96 - 0x60 - N_SSYM Structure or union element Value is offset in the structure. <> @node N_ENTRY @section 164 - 0xa4 - N_ENTRY Alternate entry point. Value is its address. <> @node N_SCOPE @section 196 - 0xc4 - N_SCOPE Modula2 scope information (Sun linker) <> @node Gould @section Non-base registers on Gould systems These are used on Gould systems for non-base registers syms. However, the following values are not the values used by Gould; they are the values which GNU has been documenting for these values for a long time, without actually checking what Gould uses. I include these values only because perhaps some someone actually did something with the GNU information (I hope not, why GNU knowingly assigned wrong values to these in the header file is a complete mystery to me). @example 240 0xf0 N_NBTEXT ?? 242 0xf2 N_NBDATA ?? 244 0xf4 N_NBBSS ?? 246 0xf6 N_NBSTS ?? 248 0xf8 N_NBLCS ?? @end example @node N_LENG @section - 0xfe - N_LENG Second symbol entry containing a length-value for the preceding entry. The value is the length. @node Questions @appendix Questions and anomalies @itemize @bullet @item @c I think this is changed in GCC 2.4.5 to put the line number there. For GNU C stabs defining local and global variables (@code{N_LSYM} and @code{N_GSYM}), the @var{desc} field is supposed to contain the source line number on which the variable is defined. In reality the @var{desc} field is always 0. (This behavior is defined in @file{dbxout.c} and putting a line number in @var{desc} is controlled by @samp{#ifdef WINNING_GDB}, which defaults to false). GDB supposedly uses this information if you say @samp{list @var{var}}. In reality, @var{var} can be a variable defined in the program and GDB says @samp{function @var{var} not defined}. @item In GNU C stabs, there seems to be no way to differentiate tag types: structures, unions, and enums (symbol descriptor @samp{T}) and typedefs (symbol descriptor @samp{t}) defined at file scope from types defined locally to a procedure or other more local scope. They all use the @code{N_LSYM} stab type. Types defined at procedure scope are emited after the @code{N_RBRAC} of the preceding function and before the code of the procedure in which they are defined. This is exactly the same as types defined in the source file between the two procedure bodies. GDB overcompensates by placing all types in block #1, the block for symbols of file scope. This is true for default, @samp{-ansi} and @samp{-traditional} compiler options. (Bugs gcc/1063, gdb/1066.) @item What ends the procedure scope? Is it the proc block's @code{N_RBRAC} or the next @code{N_FUN}? (I believe its the first.) @item @c FIXME: This should go with the other stuff about global variables. Global variable stabs don't have location information. This comes from the external symbol for the same variable. The external symbol has a leading underbar on the _name of the variable and the stab does not. How do we know these two symbol table entries are talking about the same symbol when their names are different? (Answer: the debugger knows that external symbols have leading underbars). @c FIXME: This is absurdly vague; there all kinds of differences, some @c of which are the same between gnu & sun, and some of which aren't. @c In particular, I'm pretty sure GCC works with Sun dbx by default. @c @item @c Can GCC be configured to output stabs the way the Sun compiler @c does, so that their native debugging tools work? It doesn't by @c default. GDB reads either format of stab. (GCC or SunC). How about @c dbx? @end itemize @node XCOFF-differences @appendix Differences between GNU stabs in a.out and GNU stabs in XCOFF @c FIXME: Merge *all* these into the main body of the document. The AIX/RS6000 native object file format is XCOFF with stabs. This appendix only covers those differences which are not covered in the main body of this document. @itemize @bullet @item BSD a.out stab types correspond to AIX XCOFF storage classes. In general the mapping is @code{N_@var{stabtype}} becomes @code{C_@var{stabtype}}. Some stab types in a.out are not supported in XCOFF; most of these use @code{C_DECL}. @c FIXME: Get C_* types for the block, figure out whether it is always @c used (I suspect not), explain clearly, and move to node Statics. Exception: initialised static @code{N_STSYM} and un-initialized static @code{N_LCSYM} both map to the @code{C_STSYM} storage class. But the destinction is preserved because in XCOFF @code{N_STSYM} and @code{N_LCSYM} must be emited in a named static block. Begin the block with @samp{.bs s[RW] data_section_name} for @code{N_STSYM} or @samp{.bs s bss_section_name} for @code{N_LCSYM}. End the block with @samp{.es}. @c FIXME: I think they are trying to say something about whether the @c assembler defaults the value to the location counter. @item If the XCOFF stab is a @code{N_FUN} (@code{C_FUN}) then follow the string field with @samp{,.} instead of just @samp{,}. @end itemize I think that's it for @file{.s} file differences. They could stand to be better presented. This is just a list of what I have noticed so far. There are a @emph{lot} of differences in the information in the symbol tables of the executable and object files. Mapping of a.out stab types to XCOFF storage classes: @example stab type storage class ------------------------------- N_GSYM C_GSYM N_FNAME unknown N_FUN C_FUN N_STSYM C_STSYM N_LCSYM C_STSYM N_MAIN unkown N_PC unknown N_RSYM C_RSYM unknown C_RPSYM N_M2C unknown N_SLINE unknown N_DSLINE unknown N_BSLINE unknown N_BROWSE unchanged N_CATCH unknown N_SSYM unknown N_SO unknown N_LSYM C_LSYM various C_DECL N_BINCL unknown N_SOL unknown N_PSYM C_PSYM N_EINCL unknown N_ENTRY C_ENTRY N_LBRAC unknown N_EXCL unknown N_SCOPE unknown N_RBRAC unknown N_BCOMM C_BCOMM N_ECOMM C_ECOMM N_ECOML C_ECOML N_LENG unknown @end example @node Sun-differences @appendix Differences between GNU stabs and Sun native stabs @c FIXME: Merge all this stuff into the main body of the document. @itemize @bullet @item GNU C stabs define @emph{all} types, file or procedure scope, as @code{N_LSYM}. Sun doc talks about using @code{N_GSYM} too. @item Sun C stabs use type number pairs in the format (@var{a},@var{b}) where @var{a} is a number starting with 1 and incremented for each sub-source file in the compilation. @var{b} is a number starting with 1 and incremented for each new type defined in the compilation. GNU C stabs use the type number alone, with no source file number. @end itemize @node Stabs-in-ELF @appendix Using stabs with the ELF object file format The ELF object file format allows tools to create object files with custom sections containing any arbitrary data. To use stabs in ELF object files, the tools create two custom sections, a section named @code{.stab} which contains an array of fixed length structures, one struct per stab, and a section named @code{.stabstr} containing all the variable length strings that are referenced by stabs in the @code{.stab} section. The byte order of the stabs binary data matches the byte order of the ELF file itself, as determined from the @code{EI_DATA} field in the @code{e_ident} member of the ELF header. @c Is "source file" the right term for this concept? We don't mean that @c there is a separate one for include files (but "object file" or @c "object module" isn't quite right either; the output from ld -r is a @c single object file but contains many source files). The first stab in the @code{.stab} section for each source file is synthetic, generated entirely by the assembler, with no corresponding @code{.stab} directive as input to the assembler. This stab contains the following fields: @table @code @item n_strx Offset in the @code{.stabstr} section to the source filename. @item n_type @code{N_UNDF}. @item n_other Unused field, always zero. @item n_desc Count of upcoming symbols, i.e., the number of remaining stabs for this source file. @item n_value Size of the string table fragment associated with this source file, in bytes. @end table The @code{.stabstr} section always starts with a null byte (so that string offsets of zero reference a null string), followed by random length strings, each of which is null byte terminated. The ELF section header for the @code{.stab} section has its @code{sh_link} member set to the section number of the @code{.stabstr} section, and the @code{.stabstr} section has its ELF section header @code{sh_type} member set to @code{SHT_STRTAB} to mark it as a string table. Because the linker does not process the @code{.stab} section in any special way, none of the addresses in the @code{n_value} field of the stabs are relocated by the linker. Instead they are relative to the source file (or some entity smaller than a source file, like a function). To find the address of each section corresponding to a given source file, the (compiler? assembler?) puts out symbols giving the address of each section for a given source file. Since these are normal ELF symbols, the linker can relocate them correctly. They are named @code{Bbss.bss} for the bss section, @code{Ddata.data} for the data section, and @code{Drodata.rodata} for the rodata section. I haven't yet figured out how the debugger gets the address for the text section. @contents @bye